Heterotrimeric G proteins in vertebrates constitute a family molecular switches that transduce the activation of a populous group of cell-surface receptors to a group of diverse effector units. The receptors include the photopigments such as rhodopsin and prominent families such as the adrenergic, muscarinic acetylcholine, and chemokine receptors involved in regulating a broad spectrum of responses in humans. Signals from receptors are sensed by heterotrimeric G proteins and transduced to effectors such as adenylyl cyclases, phospholipases, and various ion channels. Physiological regulation of G protein-linked receptors allows for integration of signals that directly or indirectly effect the signaling from receptor→G protein→effector(s). Steroid hormones can regulate signaling via transcriptional control of the activities of the genes encoding members of G protein-linked pathways. Posttranscriptional mechanisms are under physiological control, altering the stability of preexisting mRNA and affording an additional level for regulation. Protein phosphorylation, protein prenylation, and proteolysis constitute major posttranslational mechanisms employed in the physiological regulation of G protein-linked signaling. Drawing upon mechanisms at all three levels, physiological regulation permits integration of demands placed on G protein-linked signaling.
More than a decade has passed since the highly cited review of G protein signaling by Dr. Alfred G. Gliman was published (188). Although the central elements of G protein signaling remain the cell-surface receptors coupled to G proteins, the family of nearly 20 heterotrimeric G proteins, and the ever-expanding, diverse groups of effector units [e.g., adenylyl cyclases (AC), phospholipases, various ion channels], detailing the physiological aspects of signaling through this pathways continues to stimulate and challenge us. A Medline search of the topic of G protein signaling including publications since the Gilman review reveals more than 10,000 citations relevant to the structure and function of members of the three cardinal elements of G protein signaling and how they relate to physiology as well as pathophysiology. Needless to say, the daunting challenge of covering adequately this volume of highly regarded scientific literature is surpassed only by the challenge of limiting the number of citations to 400–500 articles. Let us be explicit in stating at the outset that this review is not meant as a comprehensive analysis of all that has been published on G protein signaling in the last decade, but rather as a snapshot of a complex and large, work in progress. We shall try to highlight the nature of the basic elements constituting G protein-linked signaling and to construct a consensus for understanding how physiological regulation occurs using transcriptional, posttranscriptional, and posttranslational mechanisms well known to the molecular and cell biologist.
In an effort to anticipate criticism of our inability to include all the most relevant articles, we have created many figures and tables, richly annotated with references. Careful analysis of the citations will reveal the inclusion of many references as the most complete, and perhaps up-to-date review of the specific topic. We make no excuses that the systems adopted as “typical” for this review are highly familiar to the authors. It is true that many of the central themes described in the review could be expressed in a multitude of contexts. Because the superfamily of G protein-linked receptors will probably rise to somewhere between 500 and 1,000, few will agree with our forced selection of a given type or even subfamily of receptor for more detailed consideration of its physiological regulation. Perhaps more telling is that the challenging task of reviewing a broad field of literature does result in the emergence of “sound” from “noise,” “pinnacles” from “peaks,” and “timeless” from “timely” with regard to discoveries in this field. Wherever possible, these seminal discoveries and changes in how we understand or approach the study of physiological regulation of signaling via heterotrimeric G proteins are highlighted. The review is structured to illuminate the formidable tasks that lay ahead in exploitation of all that we have learned from studies of the structure and function of individual members of the troika that transduces cell membrane signals into a diverse downstream network of pathways involved in cell proliferation, differentiation, and apoptosis to a more unified understanding of the integration of these inputs.
II. ELEMENTS OF G PROTEIN-LINKED SIGNALING
A. G Protein-Linked Receptors
Evolution of our understanding of G protein-linked receptors (GPLR) increased rapidly after the first molecular cloning of a well-known member of the GPLR that bind hormones. In 1986, the primary sequences of the hamster (139) and turkey (626) β-adrenergic receptors were reported, providing long-awaited information that proved critical to understanding, classifying, and probing GPLR. Before 1986, what we knew about GPLR was garnered first from pharmacological studies, then radioligand binding studies, and later through heroic efforts in protein purification of these low-abundance membrane proteins (357,529). The identification of seven hydrophobic sequences similar in length and in organization to that of bacteriorhodopsin permitted modeling of GPLR and the molecular tools with which to test many details of emerging hypotheses (Fig.1). By the end of 1998, more than 300 members of the superfamily of hormone receptors linked to G proteins have been entered in the GenBank. If one also includes in the listing of GPLR the expanding subfamily of odorant receptors (69,97), it is likely that ∼1% of the mammalian genome encodes GPLR.
Several useful approaches have been reported to the classification of GPLR. Molecular cloning provides a wealth of information, best managed as a database with automated data collection, structured flatfiles, and data query mechanisms. One such database was reported within the National Center for Biotechnology Information, first available in 1994 (307). Such databases are extremely useful tools, providing means to search homologies, to segregate paralogues (variants of receptors created by gene duplication) and orthologues (same GPLR, but different species), and to discern family, subfamilies, and other relationships among the members of this superfamily of receptors.
Using 10 residues conserved among more than a hundred different GPLR and analysis of sequence alignment, BIN maps have been created that can discriminate among the families based on the length of consecutive segments, termed “partitions” (387). Patterns of relationships develop that are provocative. The key residues are the NH2-terminal Met termed “α,” the last nonconserved residue termed “ω,” and the following, intervening conserved residues (in transmembrane-spanning regions, see Fig. 1): Asn (TMSR1), Asp (TMSR2), Cys (top of TMR3), Arg (bottom of TMSR3), Try (TMSR4), Cys (bottom loop between TMSR4 and TMSR5), Pro (TMSR5), Pro (TMSR6), and Pro (TMSR7). Although only a proposed method of analysis, BIN mapping may prove valuable in assigning identities to orphan receptors (387).
Based simply on the chemical nature of the ligand, GPLR may be classified into families and subfamilies (537). A simple, nonexhaustive listing of the prominent members of families is provided in Table 1. Table 1 reveals the marked diversity of the ligands/stimulants toward which GPLR apparently have evolved. Visual excitation operates via GPLR signaling, relying on the capture of a photon by the 9-cis-retinal covalently coupled via Lys-296 to the opsin photopigment rhodopsin (7,86, 134, 455, 542). The physiology and pathophysiology of vision can be explored elsewhere (422). Gustatory signals (taste) and odorant signals (smell) likewise operate via GPLR (1, 60,365), the odorant family likely representing the most populous of all GPLR families with membership expected in the range of 1,000–2,000 receptors (22). A genetic analysis of mammalian olfaction reveals many intriguing aspects of this GPLR signaling-based sensory physiology (457). Small molecule receptors in the classification include the most well-characterized receptors for biogenic amines, such as dopamine, epinephrine, and serotonin. Included in this group are both receptors for ATP found ubiquitously in mammals (350,356) and for cAMP, a molecule that is central to the sensory biology of Dictyostelium discoidum, the slime mold (165, 254, 380). Thus GPLR mediate the actions of a chemically diverse family of ligands (including photons) and do so in a broad spectrum of organisms, from human to mold.
The adrenergic receptors typify the relationship that can exist among subfamilies in the GPLR superfamily (413). At least nine adrenergic receptor paralogues have been identified. In the Ahlquist era, a discrimination between α- and β-adrenergic agonists was discovered. Although pharmacology created both agonists and antagonists that collectively drove the subclassification further for both the α- and β-adrenergic ligands, few envisioned the revelations emerging from molecular cloning that provided no less than nine members to the adrenergic receptor gene subfamilies (535,536). The β-adrenergic receptor subfamily, initially composed of the β1- and β2-members, saw the number expand to three most recently, with the discovery of the β3-adrenergic receptor (152). The α-adrenergic subfamilies are composed of α1- and α2-adrenergic groups, with three gene products for each (α1 with A, B, and D; α2 with A, B, and C) (440). Molecular cloning has revealed the existence of other subfamilies, such as the muscarinic acetylcholine receptor with five members (343, 406, 609) and the dopamine receptor subfamily with at least seven members (6, 88, 124, 274,382).
The receptor subfamily for 5-hydroxytryptamine (5-HT or serotonin) provides an impressive display of diversity (Fig.2) (88, 94,103, 127, 218-220). There exists at least 13 members of the rat serotonin receptor subfamily, each the product of a separate gene (161). In some cases (5-HT1B/D, 5-HT2, and 5-HT3) serotonin receptors were identified pharmacologically, before analysis by molecular cloning (219). Many, however, were generated by molecular screening of DNA libraries under low-stringency hybridization conditions. Although the sequence homology chart provides information on the extent to which various members of the subfamily of serotonin receptors display homology at the level of protein sequence, they do not reveal information on the nature of the signal propagation. Combined expression and characterization of the cloned genes will enable a methodical approach to understanding the basis and need for the many members of this subfamily of GPLR.
The β2-adrenergic receptor provides an excellent model from which to discuss the general structure of GPLR (530, 531, 535). The primary sequence information deduced from the molecular cloning (139) revealed the presence of seven domains of the sequence, rich in hydrophobic residues and spanning 22–28 residues in length (Fig. 1). On the basis of the hydrodynamic and detergent solubility data, it was speculated that these hydrophobic domains of the receptor spanned the lipid bilayer of the membrane, organizing the receptor in a fashion that resembled the deduced structure of bacteriorhodopsin (164, 226,469). Subsequent biochemical analysis and topographical analysis by site-directed, antipeptide antibodies established the NH2-terminal domain and three segments intervening between the membrane-spanning domains as exofacial, whereas three additional intervening segments as well as the COOH-terminal tail of the receptor were localized to the cytoplasmic face of the lipid bilayer (Fig. 3) (17,589-591). Studies of N-glycosylation sites of the β2-adrenergic receptor confirmed the location of the NH2-terminal sequence as exofacial (181,182). In a similar manner, study of the protein phosphorylation of GPLR confirmed the COOH-terminal tail and the segment between transmembrane-spanning regions 3 and 4, 5 and 6 display sites for phosphorylation by a variety of protein kinases, including protein kinase (PK) A (222), PKC (456), and members of the GPLR kinases, termed GRK (292, 348). The NH2-terminal segments of GPLR vary in size from 7 to 595 amino acids and the COOH-terminal region from 12 to 359 amino acids. Loops vary in size from 12 to 359 amino acids. Nearly one-half of the known GPLR that act as receptors for hormones, autacoids, and neurotransmitters also possess a conserved protein motif, either Asp-Arg-Tyr or Glu-Arg-Trp located in the NH2-terminal portion of the intracellular loop between transmembrane-spanning regions 3 and 4. G protein-linked receptors for calcitonin, CRH, growth hormone-releasing hormone, glucagon, and parathyroid hormone lack this motif, while retaining all of the other features discussed above (536).
Although commonly displayed in the “serpentine” arrangement for the sake of simplicity, GPLR have been shown by a number of indirect but compelling data to be organized much like a basket composed of the transmembrane-spanning domains (187, 235,235, 529, 590). Photoaffinity labeling with high-affinity antagonist ligands have revealed the close proximity of transmembrane segments 6 and 7 with segments 1 and 2 (Fig. 4). These receptors for monoamines invariably possess an acidic side chain of Asp in transmembrane-spanning segment 3 necessary for formation of a salt bridge with the amino group of the ligand (531). A hydrophobic pocket critical to binding is created by aromatic residues of transmembrane-spanning segments 6 and 7. The hydroxyl groups of the ligand family are believed to be hydrogen bonded through Ser and Thr residues in transmembrane-spanning segments 4 and 5. At least for the β2-adrenergic receptor, these data and the overall similarity to bacteriorhodopsin (385) suggest the existence of a binding “pocket” in which ligands interact. For the photopigment rhodopsin, the retinal chromophore attached covalently to Lys-296 of transmembrane-spanning segment 7 appears to be buried ∼20 Å into the lipid bilayer. The hydrophobic, tryptic core of the β2-adrenergic receptor retains the ability to activate Gs in response to agonists (477). Fluorescence energy transfer measurements with the β2-adrenergic receptor, likewise, suggest that the binding of antagonists occurs within the dimensions of the lipid bilayer (98,167, 302, 338,531), although establishing the depth to which ligands penetrate the lipid bilayer will require more precise analysis.
Many GPLR bind protein ligands. The size and complexity of protein ligands for GPLR span from small peptides [thyrotropin-releasing hormone (TRH)] to large glycoproteins [luteinizing hormone (LH) and thyrotropin-stimulating hormone (TSH)] (619). Whereas the small molecule ligands may bind and activate their GPLR via a domain embedded in the bilayer, GPLR that bind protein ligands often display large NH2-terminal, exofacial domains (100, 150, 225,265, 381, 619). There is a positive correlation between the ligand size and the GPLR NH2 terminus (266). Even more intriguing is the thrombin receptor, a substrate for the thrombin protesers carrying an NH2-terminal sequence in which its ligand is embedded (8, 57, 123, 129,585). Cleavage of the exofacial domain of the receptors yields an activating peptide ligand, tethered to the receptor in its basal state (57, 585). Other members of this novel class of protease-activated receptors (PAR) have been cloned (57), providing an additional dimension to our understanding of ways in which GPLR can be activated. This mechanism of activation, irreversible by nature, poses some interesting possibilities about the manner in which thrombin receptors and other members of the PAR family are activated, desensitized, and downregulated (56, 57, 129,428, 559, 561).
By definition, all GPLR not only provide the discriminator activity for ligand binding but must propagate the activation to a G protein(s) via protein-protein interactions (187). First analyzed for the β1- and β2-adrenergic receptors, the NH2-terminal and COOH-terminal extremes of the segment intervening between transmembrane-spanning regions 5 and 6 appear to play a prominent role in G protein interaction. Proteolytic cleavage products of native receptor (433, 477, 615) and forms of the receptor with mutations in these regions (530, 531) provided model systems in which to ascertain important G protein contact sites for GPLR. Mutagenesis not only identifies regions critical to G protein coupling by GPLR, but also reveal some residues that alter the very nature of the G proteins to which a GPLR couples (614, 615). The early work in the β-adrenergic receptors has been since confirmed by similar studies in a wide variety of GPLR. Taken together, the data provide a compelling picture of receptor-G protein contact sites that transduce agonist binding and transformation to a receptor with high affinity for agonist into changes in G protein binding of and activation by GTP.
B. Heterotrimeric G Proteins
1. Family classification
G proteins are members of a superfamily of GTPases that are fundamentally conserved from bacteria to mammals and play diverse roles in many aspects of cell regulation (190). The family of receptor-coupled G proteins has a unique heterotrimeric composition, and there is structural and functional diversity among each of the three polypeptide components of a G protein heterotrimer (228, 533). In general, G proteins are classified by reference to their α-subunits, although the newly appreciated regulatory roles for the tightly associated βγ-dimers suggest that this nomenclature should be revised to account for the potentially complex variety of G protein heterotrimers that can be assembled from these distinct α-, β-, and γ-monomeric gene products (105, 107, 108,189).
2. G protein activation and deactivation
Agonist-liganded receptors associate with and promote activation of G proteins by stimulating release of GDP bound to the guanine nucleotide-binding site of the α-subunit. Release of GDP is followed by GTP binding, causing dissociation of the heterotrimer into derivative substrate α- and βγ-dimer. The intrinsic GTPase activity of the α-subunit determines the lifetime of this active (dissociated) state of the G protein. Hydrolysis of bound GTP to GDP allows the α-subunit to reassociate with the βγ-dimer, ready for another round of receptor-regulated activation (48). Both the α- and βγ- subunits can regulate G protein-coupled effectors in a selective manner that can be either independent, synergistic, or antagonistic (408). The GTPase activity of Gα and its association with βγ-subunits are both regulated by accessory proteins. The accessory proteins involved, “regulators of G protein signaling” have been afforded the acronym RGS proteins. The RGS proteins may bind to Gαi subunits, for example, accelerating the rate of GTP hydrolysis (30,32, 33, 135). The Gβγ subunits are regulated by the protein phosducin, which binds Gβγ tightly, preventing their interaction with Gα and effectors (39). Twenty years of biochemical and genetic studies of the heterotrimeric G proteins have recently culminated in determination of the three-dimensional structure of several G protein α-subunits in both their GDP- and GTP-liganded states and of a G protein αβγ-heterotrimer (30, 111,113, 298, 323, 324,384, 515, 516, 522,540, 557, 587). These new structural data are treated in a later segment.
Gα subunits are extrinsic membrane proteins. Lipid modification generally involving myristoylation or palmitoylation at the NH2 terminus is essential for membrane anchoring of these proteins (76, 84). Posttranslational features of Gα subunit function are detailed in section v.
Structural and functional classification of G protein oligomers has been defined by the α-subunits (see Table2). As might be expected of proteins that perform certain highly conserved functions (for example, association with activated hormone receptors, GTP binding, and hydrolysis as well as association with βγ-dimers), the primary sequence of all known Gα subunits contains ∼20% invariant conserved amino acids. Outside of these regions, the sequences of the G proteins diverge. Four families of these proteins, termed s, i, q, and 12/13, have been proposed based on amino acid sequence comparisons (228, 611).
The Gαs class contains Gαs and Gαolf, which are 88% identical (273). Both proteins activate AC and are substrates for ADP-ribosylation catalyzed by the A1 protomer of a toxin elaborated byVibrio cholera (i.e., cholera toxin). This posttranslational modification inhibits the intrinsic GTPase activity of the G proteins (273) (see sect. vC ).
The Gαi class contains Gαi-1, Gαi-2, Gαi-3, the retinal Gα, Gαt, two forms of the brain-specific Gα subunit Gαo-1 and Gαo-2, as well as Gαz. All members of this class (with the exception of Gαz) contain a conserved COOH-terminal cysteine residue that is the site of ADP-ribosylation catalyzed by a toxin elaborated by Bortadella pertussis (i.e., pertussis toxin). This irreversible, covalent modification uncouples the G protein from its activating receptor (see sect. vC ). Blockade of cellular responses to stimulation by pertussis toxin treatment has been an effective experimental procedure employed to implicate this class of Gα subunits in specific cellular signaling processes. The Gαt subunit activates retinal cGMP phosphodiesterase, the major effector in vertebrate phototransduction. Members of the Gαi and Gαo subfamilies are implicated in the regulation of ion channel activity and regulation of phospholipase (PL) C, whereas the function of Gαz is not known (85, 349).
The Gαq class contains five family members, Gα11, Gα14, Gα15, Gα16, and Gαq. These closely related proteins are substrates for neither cholera toxin- nor pertussis toxin-catalyzed ADP-ribosylation. The Gαq subunits are notable regulators of the β-class of phosphoinositide-specific PLC (PLC-β). Gαq and Gα11 are widely expressed in mammalian tissues. The expression other members of the Gαq class, in contrast, is restricted to stromal and epithelial cells as well as to cells of the hematopoietic lineage. These Gα subunits also activate PLC-β isoforms and may exhibit a preference for members of the PLC-β2 family that also display a similarly restricted pattern of expression (4, 10,200, 327, 328, 416,512, 513).
The final class of pertussis toxin- and cholera toxin-resistant Gα subunits contains two proteins, Gα12 and Gα13. These α-subunits are widely expressed (510). The functions of Gα12 and Gα13 have not been clearly defined. Overexpression of activated forms of these proteins transforms fibroblasts (312). Expression and activation of Gα12 and Gα13 occurs in differentiation of P19 embryonic stem cells in response to retinoic acid (264). Activation of Gα13 leads to selective activation of mitogen-activated protein kinases, especially jun NH2-terminal kinases (263). Other data implicate Gα12 and Gα13 in regulation of the Na+/Cl− antiporter activity (9, 414, 534).
4. β-Subunit/γ-subunit complexes
The realization that the Gβγ subunits play direct roles in regulation of effectors has focused attention on these tightly associated βγ-dimers (see Table 3). Five distinct members of the Gβ family have been identified (595). Each β-subunit displays 340 amino acid residues and a molecular mass of ∼35,000 Da. The linear sequence of these proteins consists of seven or eight tandem-repeats with a central conserved Trp-Asp sequence that has been termed a “WD-40” motif (178). The Gγ family are more divergent, displaying seven family members to date, at least six of which are mammalian isoforms. These γ-subunits vary in molecular masses from 7.3 to 8.5 kDa and are considerably more diverse in primary sequence than the Gβ subunits (30, 107, 253,294, 408, 491). The predicted COOH-terminal sequences of all Gγ subunits contain the sequence CysAAX, where A is any aliphatic amino acids. The Gγ subunits are modified by prenylation of the Cys residue by removal of the final three amino acids by proteolysis, and then by carboxymethylation of the newly generated COOH-terminal Cys. The retinal Gγ1subunit is modified by farnesylation, whereas the other mammalian Gγ subunits all are modified by geranylgeranylation (see sect.vB ). This acyl chain modification is necessary for association of the Gβγ dimer with the lipid bilayer. Gβ and Gγ subunits are tightly associated and cannot be dissociated except under strongly denaturing conditions. Acylation of Gγ is not required for assembly of the Gβγ dimer, although the mechanisms by which these proteins are assembled posttranslationally are not known. Acylation does appear to play critical roles both in membrane association of the Gβγ dimer, in the association of Gβγ with Gα, and in interactions with AC (175, 195,253, 279, 569).
5. Structures of heterotrimeric Gα and Gβ/γ subunits
Structures of Gαt and Gαi have been solved in their GTP-, GDP- and AlF4-liganded states (30, 111, 113, 298,323, 516). A mutant of Gαi-1 in which Gly-203 is substituted with Ala has been solved in its GDP and phosphate-bound state. The structures of Gβγ complexes with Gαi-1 and phosducin have been described (454), and a very recent study reports the structure of the heterotrimeric catalytic core of AC complexed with Gαs (558) (Fig. 6). These structures have provided invaluable insight into the mechanisms by which G proteins interact with and are oriented relative to cell membranes and their cognate receptors, as well as the processes of receptor-catalyzed guanine nucleotide exchange, GTP hydrolysis, and effector activation.
Gα subunits are composed of essentially two distinct domains, a Ras-like GTPase domain and a predominantly helical domain that is unique to the Gα subunits. The bound guanine nucleotide is held at the interface of these domains. Three switch regions within the GTPase domain (switches I, II, and III) change conformation in response to the guanine nucleotide-liganded state of the Gα subunit, and significantly, all three switch regions form significant contacts with Gβγ and effectors. In the GTP-bound state, the switch regions are held in place by contacts to the terminal γ-phosphate of the nucleotide, whereas these regions appear to be less ordered (more flexible) in crystals of the GDP-liganded G proteins (Fig.5).
Structures of two Gβγ dimers have been determined. Gβ1γ1 (the retinal Gβγ species) has been crystallized alone and in combination with either a Gα subunit or the regulator phosducin (515). A second Gβγ dimer, Gβ1γ2, has been solved in the Gα-bound state (587). The Gβ structure is dominated by a β-propeller, which is a structural motif found in a number of different proteins (522) (Fig.6). This motif is composed of seven repeats of four-stranded α-sheets that are arranged around a small central hole. This prominent symmetry arises from the internally repeated WD-40 motif found in the Gβ subunits and a functionally diverse group of proteins. Gγ subunits contain a membrane-anchoring COOH-terminal farnesyl group. Gγ associates tightly with Gβ through a coiled-coil structure. The presence of a region of sequence enriched in basic amino acids on the side of the β-propeller abutting this coiled-coil region suggests a possible role in an electrostatic interaction with membrane lipids. It is noteworthy that the Gβ1γ1 and Gβ1γ2 structures differ in the relative orientation of the coiled-coil, which may reflect differences in interactions either with receptors or with effectors, or perhaps with both.
The structures of the individual components of a G protein heterotrimer have been augmented by descriptions of the structures of two Gαβγ complexes. The major interaction between Gα and the βγ-complex involves the NH2 terminus. These structures indicate that association with Gβγ produces a significant change in conformation of the switch I and II regions of the G protein (Fig. 6). The structure of Gβγ is not appreciably changed by association with Gα. This important observation suggests that receptor-promoted guanine nucleotide exchange alters the conformation of the switch II region of Gα, disrupting interactions that are critical for Gβγ binding (587). The structure of guanosine 5′-O-(3-thiotriphosphate) (GTPγS)-liganded Gαs in complex the catalytic domains of AC reveals a unique interaction between the switch II region of the G protein and its effector protein (Fig. 7) (558). A number of excellent reviews discuss these structures and the implications of the information they reveal for understanding G protein function (215, 522).
C. G Protein-Linked Effectors
G protein α- and βγ-subunits regulate the activities of a structurally diverse group of effector molecules. These include enzymes engaged in the synthesis and degradation of intracellular second messengers, as well as ion-selective channels (see Table4). In some cases, direct regulation of these effectors by G protein subunits has been demonstrated unequivocally by in vitro reconstitution, whereas in other instances, a timely example being the mitogen-activated protein kinases (MAP kinase) cascade, G protein regulation of the effectors may be indirect. We consider these classes of G protein effectors further below. It is also noteworthy that members of at least two classes of G protein effectors, the PLC-β enzymes and the cGMP-phosphodiesterases, function as GTPase-activating proteins (RGS-like functions) for their G protein regulators.
Adenylyl cyclase catalyzes the formation of cAMP from the substrate Mg2+-ATP. These enzymes are expressed in a wide range of species from bacteria, yeasts, and slime molds to mammals. In general, AC are membrane-bound proteins, although certain of the bacterial and possibly mammalian enzymes are cytosolic. The yeast and bacterial AC are peripheral membrane proteins, whereas theDictyostelium enzyme is an integral membrane protein with a single transmembrane spanning domain (79). The G protein-regulated AC isoforms identified in higher eukaryotes share a common motif (159, 511). These enzymes are large, single polypeptides (molecular masses in the range of 120 kDa). In general, mammalian AC appear to be proteins embedded in the plasma membrane. The cloning of the first of these enzymes revealed a deduced topology that comprises an apparently duplicated structure consisting of a short NH2 terminus, two membrane-spanning regions of six α-helices which link ∼40-kDa cytoplasmic domains (termed domains C1 and C2). The helical portions of the transmembrane domains are linked by short extramembranous regions. This topological organization is unique among enzymes but common to several ATP-dependent transport proteins including the P-glycoprotein and cystic fibrosis transmembrane conductance regulator (18, 313). Whether or not these transmembrane AC isoforms function as membrane transport proteins remains a provocative question.
Complementary DNA encoding nine distinct mammalian AC isoforms have been cloned and expressed (95, 159,176, 258, 549). These proteins are designated types I—IX with evidence for existence of alternatively spliced transcripts of unknown significance, further adding to the complexity. Homologs have been identified both inDictyostelium and Drosophila. The C1 and C2 domains of the mammalian AC are well conserved (50–90% identity). It is also noteworthy that the C1 and C2 domains are homologous to each other and to the catalytic domains of a number of membrane-bound guanylyl cyclases (603). Catalytic activity requires both the C1 and C2 domains, and point mutations in either of these two domains can inhibit catalytic activity. Soluble guanylyl cyclase isoforms function as heterodimers, and this also appears to be the case for the AC. Expression of either the NH2- and COOH-terminal halves of the molecule or, remarkably, of the individual C1 and C2 domains devoid of their membrane anchors reconstitutes AC activity. These soluble AC are activated both by Gαs and by the plant diterpene forskolin (547). Although neither the C1 nor C2 domains display sequences homologous to those involved in binding nucleotides, it is clear that the enzyme must interact with its substrates as well as with so-called “P-site” nucleotide regulators (272). That binding sites for these substrates and regulators are formed through the interaction between the two domains seems obvious based on such considerations. The recently described structure of a C2 domain homodimer revealed that forskolin and nucleotide substrates bind at the C1/C2 domain interface (558).
Where examined, the mammalian AC isoforms are generally widely expressed, although types I and II appear to be primarily neuronal, and type III is restricted to olfactory epithelium (258). All forms of AC identified in mammalian systems are stimulated by GTP-liganded Gαs. The enzymes differ in their susceptibilities to regulation by βγ-subunits, by members of the Gαi class, by Ca2+/calmodulin, and by PKC. βγ-Subunits are effective inhibitors of the type I enzyme but stimulate activity of the type II and type IV enzymes in a manner that is highly conditional on costimulation by Gαs. It is noteworthy that as with other βγ-subunit-dependent phenomena, stimulation of type II AC requires considerably higher concentrations of βγ-subunits than activating concentrations of α-subunits (548, 551, 552). Thus abundant Gi heterotrimers are likely to be the physiologically important source of βγ-subunits for this mode of regulation of AC.
Although the Gi family was initially identified as the G proteins responsible for inhibition of AC activity, mechanisms proposed for the inhibitory mode of regulation have been the focus of intense debate. A failure to observe inhibition of adenylyl cyclase activity by isolated Gαi led to the proposition that sequestration of Gαs by βγ-subunits might be the mechanism underpinning the inhibitory response. More recent investigations reveal that all three isoforms of Gαi are equally effective inhibitors of the types V and VI enzymes. Type I AC is selectively inhibited by Gαo, whereas the types I and V enzymes can be inhibited by Gαz (550, 551,554).
Intracellular Ca2+ is an important regulator of AC activity. The types I, VIII, and to a lesser extent III enzymes are potently activated by Ca2+/calmodulin, whereas the other known isoenzymes are rather insensitive to this activator (598). Studies using intact cells have implicated a number of phosphorylation-dependent mechanisms in control of AC activity. Of these, control by cAMP-dependent PKA and by PKC has been most widely studied. Although both protein kinases can phosphorylate certain AC isoforms in vitro, the effects on adenylyl cyclase activity observed are modest in comparison with those observed in intact cells. It seems likely that PKA and PKC control AC activity by an indirect mechanism(s) (604).
The distinct patterns of regulation of the AC isoforms suggest that these enzymes may function as integrators of G protein-mediated signaling pathways. Regulatory input from Gq-coupled receptors can control AC activity by Ca2+ or PKC-dependent processes. Gi can directly inhibit the types I, V, and VI enzymes, whereas Gi-derived βγ-subunits can activate the type II enzyme in a conditional manner. All of the AC isoforms described to date continue to share the ability to be stimulated by GTP-loaded Gαs.
3. cGMP phosphodiesterases
The cGMP phosphodiesterase (PDE) plays a central role in visual excitation in vertebrate rod photoreceptor cells. Absorption of a photon by the photopigment rhodopsin leads to activation of the PDE which, in turn, catalyzes the hydrolysis of cGMP, causing closure of cGMP-gated cation channels and in hyperpolarization of the cell membrane (214). The cGMP PDE exists in both membrane-bound and soluble forms. Irrespective of localization, the protein is a heterotetramer composed of three distinct polypeptides Pα, Pβ, and Pγ with molecular masses of 88, 84, and 14 kDa, respectively, found in a molar ratio of 1:1:2. The Pα and Pβ polypeptides both contain catalytic sites, whereas the Pγ subunit is an inhibitor of their enzymatic activities (52). The retinal Gt protein mediates regulation of the PDE by rhodopsin. GTP-loaded Gαt binds the Pγ subunit, releasing inhibition of the Pα and Pβ catalytic subunits. Pγ is a GTPase-activating protein (GAP) for Gαt, increasing its rate of GTP hydrolysis (12). Guanosine 5′-triphosphate hydrolysis releases the Pγ subunit to reassociate with and to inhibit the catalytic subunits of the enzyme, completing the activation/deactivation cycle.
4. Inositol lipid-specific PLC
Three families of PLC enzymes classified as -β, -γ, and -δ have been identified in mammalian systems. There are multiple isoenzymes within each class (464). The PLC enzymes share some common structural features including two regions of conserved sequence, termed “X” and “Y” domains, that contain residues important for substrate binding and catalysis. Outside of these sequences, the proteins diverge extensively. The four PLC-β isoenzymes are targets for activation by Gα- and Gβγ subunits. Members of the Gq family of G protein α-subunits couple receptors to activation of the PLC-β enzymes in a pertussis toxin-insensitive manner (512, 555,556). G protein regulation of the PLC-β enzymes has been studied using both reconsititution assays with purified proteins and exogenously provided substrates as well as by transient transfection of COS-7 cells with vectors for the expression of the PLC enzymes and G protein subunits. Experiments using transient expression of Gα subunits and PLC-β isoenzymes in COS-7 cells suggest that there are differences in susceptibility of the individual PLC-β enzymes to activation by the various Gq family members (10, 416, 617,618). However, experiments with purified PLC-β1, -β2, and -β3 show little difference in activation by purified Gαq, Gα11, and Gα16 (229,311, 513).
In many systems, receptor regulation of inositol lipid hydrolysis is sensitive to pertussis toxin (216). The observation that G protein βγ-subunits activate the PLC-β enzymes both in vitro and in transient transfection assays provides a biochemical explanation for this finding (54, 55, 78,137, 247). βγ-Subunits can activate all of the PLC-β isoenzymes. Phospholipase C-β2 and PLC-β3 appear more sensitive to stimulation by βγ-subunits than PLC-β1 and PLC-β4(430, 463). In the case of PLC-β3, Gαq and βγ-subunits are approximately equally effective activators on a molar basis. For other members of the PLC-β family, the αq-subunits are considerably more potent (50- to 100-fold) than Gβγ subunits.
The PLC-β enzymes appear to function as GAP for their G protein regulators. The Gαq subunits used in these activation experiments were liganded with nonhydrolyzable guanine nucleotide analogs (35). Under these conditions with GTPase-resistant GTP analogs, it is possible that the apparent potency of α-subunits is truely an overestimate. In general, such reconstitution studies employ Gβγ subunits purified from bovine brain. Since a large number of βγ-dimers can be assembled from individual members of the β- and γ-subunit multigene families, it is possible that individual Gβγ dimers not tested may prove to be more effective PLC activators. A limited number of studies have been performed to test this possibility (53, 253). With the exception of Gβ1γ2, Gβγ dimers of various combinations tested appear equipotent and effective as PLC activators. In fact, the reduced potency of Gβγ subunits (compared with α-subunits) as activators of the PLC-β enzymes may provide a degree of selectivity in receptor regulation of the PLC-β enzymes. Accordingly, only activation of abundant G protein heterotrimers (for example, members of the pertussis toxin-sensitive Goand Gi families) would produce sufficient βγ-subunits for activation for PLC-β by this manner.
One unique characteristic of the Gq family of heterotrimeric G proteins is that, when purified and reconstituted with appropriate receptors which promote the guanine nucleotide exchange step in the G protein activation cycle, their intrinsic steady-state GTPase activities are much lower than those of members of the other heterotrimeric G protein families (229,311, 426). This finding was paradoxical, since direct measurement of PLC-mediated increases in intracellular Ca2+ revealed that upon addition of an agonist, the Gq/PLC-β system was activated extremely rapidly (216). Ross and co-workers (34) studied regulation of PLC-β1 in a reconstituted system containing purified M1 muscarinic cholinergic receptors and a purified mixture of Gαq and Gα11. In this system, receptor-promoted binding of GTPγS to the G protein and PLC-catalyzed phosphatidylinositol 4,5-bisphosphate (PIP2) hydrolysis are tightly coupled. When the nonhydrolyzable GTPγS was replaced by hydrolyzable GTP, PLC activation was much reduced (35). This apparent uncoupling between G protein and effector in the presence of GTP suggests that PLC-β1 is a GAP for its Gαq/11 activator (35). Promotion of GTP hydrolysis accelerates the deactivation rate of the Gαq/11 subunits and, in turn, accelerates the deactivation of PLC-β1. The GAP activity of PLC-β1 enables the inositol lipid signaling system to respond rapidly to receptor deactivation. In this way, inositol signaling is regulated not only by the rate of receptor-catalyzed GDP/GTP exchange of the α-subunit, but also by acceleration of the rate of GTP hydrolysis accelerated by the effector. More detailed studies of the time courses of PIP2 hydrolysis by PLC-β1 in this reconstitution system suggest that, in the presence of saturating agonist, receptor-Gqcomplexes can remain stable over multiple GTPase cycles (436).
All PLC enzymes contain two highly conserved X and Y domains. The three-dimensional structure of a portion of PLC-δ containing the X and Y domains has been reported recently (153). The other regions of the PLC enzymes display divergent sequences, focusing attention on the unique regions of the PLC-β enzymes as possible sites of G protein association. Although the precise structural nature of the interaction between the PLC-β enzymes and G protein α- and βγ-subunits is not known, several lines of evidence support the idea that these regions which comprise the NH2 terminus, the inter X-Y region, as well as the COOH terminus play important roles in the PLC-β G protein coupling.
A number of independent experiments suggest that the COOH terminus of the PLC-β enzymes contains sites for interaction with G protein α-subunits. Rhee and co-workers (431) reported that truncation of PLC-β1 and PLC-β2 immediately after the Y domain produces catalytically active proteins that can be activated by G protein βγ-subunits but not by Gα subunits. Simon and colleagues (616) found that overexpression of the COOH terminus of PLC-β1 in COS-7 cells blocked activation of cotransfected PLC-β1 by muscarinic cholinergic receptors and G protein α-subunits. Two short peptides corresponding to a portion of the PLC-β1 amino acid sequence displayed the activity to block activation of PLC-β1 by Gαq in vitro, whereas a peptide of identical amino acid composition but different sequence did not (616). Ross and co-workers (34, 35) found that several recombinantly expressed fragments of the COOH terminus of PLC-β1 functioned as GAP for purified Gαqwhen reconstituted with purified muscarinic cholinergic receptors. The PLC-β “tail” with the most effective GAP activity did not correspond to the region identified as a site of Gαqinteraction with PLC-β1 using the synthetic peptide approach described above (616). It is possible that the G protein PLC-β interactions that provide the basis for stimulation of GTPase activity and activation of PLC catalytic activity involve different parts of the PLC-β COOH terminus. Interestingly, some of the PLC-β COOH-terminal tails inhibited basal PLC-β1 activity in a manner that could be overcome by Gαq. Perhaps PLC-β1 can form oligomers, and this self-association may have some importance for the mechanism by which G protein α-subunits activate the enzymes (436).
The site of interaction between the PLC-β enzymes and G protein βγ-subunits is less well defined. Because removal of the COOH terminus of PLC-β1 and PLC-β2 does not diminish the capacity of βγ-subunits to activate the enzymes, it is reasonable to presume that the remaining portion of the enzymes contain the βγ-subunit interaction site. The X and Y domains are common to all PLC enzymes. Therefore, the most likely sites for βγ-subunit interaction are the NH2 terminus and the inter X-Y region. There exists experimental evidence suggesting that both of these regions of the proteins are important for activation by βγ-subunits. The PLC-β and -δ isoforms contain a pleckstrin homology (PH) domain at their NH2 terminus. PH domains have been found in a diverse group of proteins including guanine nucleotide exchange factors for small G proteins, as well as for some protein kinases (334). It is clear that a subset of PH domains mediate selective interactions of proteins with inositol lipids and phosphates (334). The PLC-δ PH domain has been studied intensely. Both intact PLC-δ and the isolated PLC-δ PH domain bind PIP2 and inositol 1,4,5-trisphosphate (IP3). Binding of PIP2 to the PLC-δ PH domain appears to anchor the enzyme to the lipid bilayer, allowing it to function in a “scooting” mode of catalysis (177). The PLC-β1 and PLC-β2 PH domains do not appear to serve an analogous function, since these enzymes bind to membranes with high affinity in a manner that does not depend on PIP2. The structure of a complex between the PLC-δ PH domain and IP3 identifies amino acid residues that participate in formation of ionic and hydrogen bonds with the inositide headgroup. The PH domains of the PLC-β isoforms have substitutions of key residues that interact with IP3 in the PH domains of the δ-isoform, which presumably accounts for their inability to bind to PIP2 (480).
Studies with various other proteins suggest that some PH domains may mediate protein interactions with Gβγ subunits. The β-adrenergic receptor kinase (β-ARK) is recruited to the plasma membrane and activated by Gβγ subunits (330). β-Adrenergic receptor kinase has an NH2-terminal PH domain, and several lines of evidence suggest that amino acid residues in the COOH-terminal portion of this motif mediate interactions with βγ-subunits (29). As discussed above, the finding that removal of the COOH terminus of the PLC-β proteins does not impair activation by βγ-subunits would be consistent with a role for the NH2 terminus in this process. Removal of the NH2 terminus of PLC-β2 produced a catalytically inactive protein, and there has been no direct investigation of a role for the NH2-terminal PH domain of the PLC-β enzymes in activation by Gβγ subunits. Other work implicates the highly charged residues of the inter X-Y region in regulation of PLC-β2 by βγ-subunits. Peptide fragments from this region of PLC-β2 expressed as fusion proteins bind to purified βγ-subunits in vitro, and overexpression of this region of PLC-β2 in COS-7 cells blocks activation of cotransfected PLC-β2 (315,616).
5. Phosphoinositide 3-kinases
Phosphoinositide 3-kinases (PI 3-kinases) are a family of ATP-dependent enzymes that catalyze the phosphorylation of the D-3 OH group of the three “conventional”myo-inositol-containing phospholipids, phosphatidylinositol (PI), phosphatidylinositol 4-phosphate, and PIP2 to form phosphatidylinositol 3-phosphate, phosphatidylinositol 3,4-bisphosphate, and phosphatidylinositol 3,4,5-trisphosphate, respectively. Members of this multigene family have been identified in plants, yeast, Drosophila, and mammals (578). Biochemical and genetic evidence suggests that PI 3-kinases control processes of receptor-regulated signal transduction, protein trafficking, and mitogenesis (578). Much of the biochemical evidence implicating PI 3-kinase in processes of cell regulation comes from pharmacological studies with several inhibitory compounds. The fungal metabolite wortmannin is the most widely employed example of this group. These inhibitor agents are potent, producing effects at nanomolar concentrations, but their degree of selectivity for PI 3-kinase activity is still open to question.
Three structurally and functionally related families of PI 3-kinases, termed types I, II, and III, have been described (578). These proteins share a common catalytic core with homology to sequences found in an extended family of “lipid-like protein kinases” which contains an ATP-binding site and amino acid residues essential for catalysis. The type I enzymes are related to the Streptomyces cerevisiae Vps34 protein and function as PI-specific PI 3-kinases. The type II family of PI 3-kinases contain a calcium-dependent lipid binding (CalB) domain and appear to preferentially phosphorylate either PI or phosphatidylinositol 4-phosphate. The type I PI 3-kinases were the first members of this multigene family to be described, and these proteins have been studied in most detail. These catalytic subunits all share a COOH-terminal lipid kinase domain with variable NH2-terminal domains that contain sites for interaction with regulatory subunits. To date, three members of this class have been described (578). P110α and -β associate with the SH2 domain-containing p55 or p85 adapter proteins, which appears to mediate activation of these PI 3-kinase catalysts by receptor and nonreceptor tyrosine kinases. p110γ does not associate with p85 (142,579, 620).
Several mammalian cells contain a PI 3-kinase activity that is sensitive to activation by the βγ-subunits of heterotrimeric G proteins. Recombinantly expressed p110γ can be activated by GTP-liganded Gαi subunits and βγ-subunits of heterotrimeric G proteins in vitro, but the effects of the G protein subunits in p110δ are considerably smaller than those reported for βγ-subunit responsive PI 3-kinases activities (224,260, 527). Recent work has identified a 101-kDa adapter protein that associates with p110γ and confers sensitivity to stimulation by βγ-subunits (526). Other mechanisms for G protein stimulation of PI 3-kinase also may exist. In neutrophils and U 937 cells, the Gβγ subunit-responsive PI 3-kinase does not appear to be associated with a p85 adapter, whereas a p85-associated PI 3-kinase activity from human platelets can be activated by Gβγ subunits (83). In platelets, a 46-kDa protein that cross-reacts with an antibody to the p85 PI 3-kinase subunit and a 100-kDa wortmannin-binding protein may be responsible for Gβγ subunit-stimulated PI 3-kinase activity (24, 83).
Phospholipase A2 catalyzes the rate-limiting step in the synthesis of prostanoids, the release of arachidonic acid from plasma membrane lipids (130). Two broad classes of these enzymes exist: low-molecular-weight secretory forms and high-molecular-weight cytosolic enzymes. Regulation of both classes is complex, and there is evidence that both forms of the enzyme play important roles in agonist-stimulated arachidonate metabolism (130). Evidence that Gβγ subunits activate a retinal PLA2 activity has been presented (262). The identity of the Gβγ-subunit regulated PLA2 isoform involved and the physiological importance of this regulation have not been established.
7. Ion channels
G proteins play important roles in the gating and modulation of ion channels that are selective for K+, Ca2+, and Na+ (105). In some cases, direct regulation of ion channels by G proteins appears likely, although the technical limitations inherent in measuring ion channel activities makes it difficult to exclude indirect, particularly second messenger-dependent mechanisms. In this regard, the regulation of cardiac K+ channels by Gαi and Gβγ subunits remains the only case in which a direct interaction of the G protein likely underlies regulation of ion channel activity. Receptor type- and cell type-specific inhibition or activation of certain Ca2+-, Na+-, and Cl−-selective ion channels by G protein subunits has also been reported, but the mechanism(s) responsible for this regulation is not known (64, 105).
a) atrial k+ channels. Neuroendocrine control of cardiac function, notably by vagal innervation or by circulating catecholamines, is an important mechanism for control of the rate and force of contraction of the heart. Acetylcholine binds to M2 muscarinic cholinergic receptors to activate an inwardly rectifying K+ channel (IK.ACh), through a pertussis toxin-sensitive mechanism (65). Hyperpolarization contributes to a slowing of the rate of contraction of the heart. A number of lines of evidence suggested that this ion channel is regulated by a G protein-dependent mechanism. Notable are the following observations, in cell-attached patch studies, acetylcholine has to be present in the patch pipette to activate the channel, and in perfused cell-attached experiments as well as in work using excised membrane patches, hydrolysis-resistant GTP analogs activate these channels (105). The gene encoding the channel responsible for IK.ACh has been cloned and termed IGIRK1(106). Excised membrane patches have been used to investigate the roles of G protein subunits in regulation of IK.ACh. In initial studies of this type, addition of either Gβγ subunits or Gαi subunits was found to produce substantial (up to 500-fold) increases in channel activity (294,610). Activation by the Gα and Gβγ subunits was not additive, and more recent work suggests that Gβγ subunits preferentially activate IK.ACh, whereas Gα subunits stimulate activity of another type of K+ channel, IK.ATP (37).
Further work reveals an additional level of complexity for the regulation of IK.ACh. The IK.ACh channel can be activated by a wide variety of G protein-coupled receptors in cardiac atria. An additional mechanism for activation of these channels by receptors may involve arachidonic acid and its metabolites 12-hydroperoxyeicosatetraenoic acid, 8-hydroxyeicosatetraenoic acid, and leukotrienes (LT) including LTD4 and LTC4(104). It is possible that some subset of the reported effects of Gβγ subunits on the IK.ACh activity may be mediated indirectly by these lipid messengers, reflecting perhaps generation by Gβγ subunit-stimulated PLA2 activity. Reconstitution of purified components in defined membrane systems will be required to resolve these issues (104).
b) ca2+ channels. Functional classification of Ca2+ channels has identified L-, N-, T-, and P-type currents. G proteins regulate several Ca2+ channels. Both the L- and N-type Ca2+channels of neurons which are both high-threshold, slowly inactivating currents and can be distinguished by sensitivity to ω-conotoxin and the L-type dihydropyridine-sensitive cardiac Ca2+ channel are regulated by G proteins (104,625). Although genes for several of these channels have been cloned, it is not yet clear how the polypeptides encoded relate to the measured currents. What is clear is that these channels are multimers composed of a common α1-subunit that forms the conducting pore. The α-subunit is variably associated with α2-, β-, γ-, and δ-subunits that presumably serve regulatory roles but whose precise functions are incompletely understood (625). Inhibition of voltage-dependent Ca2+ channels by G protein-coupled receptors was first reported in chick dorsal root ganglion preparations and subsequently has been studied in a number of cell lines (364). A variety of receptors cause inhibition of L-type Ca2+channels in neuroglioma-blastoma NG108–15 cells. Receptors for Leu-enkephalin, somatostatin, and norepinephrine activate channels through a pertussis toxin-sensitive pathway, whereas those receptors for bradykinin do not. Expression of pertussis toxin-resistant Gαo and Gαi mutants in these cells enables the cells to overcome inhibition of receptor-mediated ion channel caused by treatment with pertussis toxin. These observations suggest a role for Gαo in coupling the opioid and catecholamine receptors to the inhibition of Ca2+ channels (553). Studies using rat GH3 pituitary cells provide evidence for an involvement of both Gα and Gβγ subunits in receptor-dependent inhibition of Ca2+ channel activity. For the GH3 cell studies, antisense oligonucleotides were employed successfully as a strategy to ablate the expression of specific G protein subunits (230). By these approaches, evidence was provided that Gαo-1 and Gαo-2 mediate muscarinic and stomatostatin-dependent inhibition of the Ca2+ current. Subsequent work extended these observations to define involvement of Gβ1 and Gβ3 in these receptor-selective processes (297, 299-301). Although the data obtained are intriguing, this work does not address the specificity or indeed the mechanisms by which G proteins regulate Ca2+channel activity. Additionally, the approach cannot reveal what parts of the G protein activation cycle or presumed G protein effector interactions have been disrupted by these manipulations.
Gαs may play a role in the activation of a cardiac L-type Ca2+ channel. Reconstitution of Gαs with membrane patches restored Ca2+channel activity when the channel was inactivated by prolonged depolarization (i.e., rundown). To investigate the mechanism(s) involved in this phenomenon further, Ca2+channel-containing preparations were reconstituted into artificial lipid bilayers. In this system, Gαs increased Ca2+ channel activity 13- to 25-fold (248). These types of experiment, however, cannot distinguish membrane-delimited second messenger-dependent activation mechanisms from direct modulation. It is noteworthy that other studies suggest a major mechanism for Ca2+ channel modulation via activation of a cAMP-dependent protein kinase pathway in intact cardiac myocytes (62).
c) na+ channels. β-Adrenergic stimulation of cardiac cells produces a rapid inhibition of the fast Na+ current, and studies using whole cell recordings as well as single-channel patch measurements suggest a direct role for Gαs in the inhibition of this current (492). This is the only report of this phenomenon, yet other studies provide evidence for second messenger-dependent pathways in the control of this channel. Activation of an amiloride-sensitive Na+ channel in a renal epithelial cell line by a pertussis toxin-sensitive mechanism has been described. Reconstitution studies implicated roles for Gαi-3 in the activation of the Na+ channel. The existence of an indirect rather than direct mechanism cannot be excluded, and PLA2 and arachidonic acid metabolites have been implicated as mediators of Na+ channel activation in this system (341).
d) cl− channels. G protein-dependent activation of Cl− channels has been reported in cardiac and epithelial cells. Molecular characterization of Cl− channels is not as advanced as that of the action channels described above. There is evidence that the cyctic fibrosis transmembrane conductance regulator (CFTR) channel in airway epithelial cells can be inhibited by pertussis toxin-sensitive G proteins, in a fashion that permits stimulation of the channel by a cAMP-dependent mechanism. An unidentified cardiac Cl−channel has been reported to be gated directly by Gαs, although a role for second messenger-dependent pathways cannot be discounted (156).
8. Tyrosine kinases and the MAP kinases cascade
Activation of G protein-coupled receptors results in stimulation of the MAP kinase pathway (490). This includes receptors for a variety of peptide and nonpeptide ligands. These receptors couple to MAP kinase activation via pertussis toxin-sensitive (presumably Gi- or Go-mediated pathways) and pertussis toxin-insensitive (presumably Gq or Gs-mediated pathways) (351). Despite considerable investigation, the precise mechanisms involved in coupling G protein activation to stimulation of the MAP kinase cascade are not known. Although analogies with the mating factor receptor-initated protein kinase cascade in budding yeast points to a central role for the Gβγ dimer in this process (141), direct regulation of kinase cascade components by G protein subunits has not been demonstrated in mammalian systems.
What is clear is that G protein-coupled receptors use multiple pathways that ultimately converge upon MAP kinase activation. The pathways coupling receptor and nonreceptor tyrosine kinases to MAP kinase activation are known in detail. Tyrosine phosphorylation creates binding sites for recruitment of the Grb2 SOS adapter/exchange factor complex which, in turn, catalyzes GTP-dependent activation of the ras low-molecular-weight GTP-binding protein. Activated ras recruits and activates the Raf protein kinase that in turn initiates the activation of MAP kinase kinase (MEK) and MAP kinase molecules, ultimately resulting in activation of nuclear gene transcription. Although there may be roles for additional signaling molecules including PI 3-kinase and the PLD product phosphatidic acid in ras-mediated raf activation, the aforementioned series of interactions clearly comprise the primary mechanism involved (354). Activation of PKC has been shown to result in MAP kinase activation in many systems. The mechanism involved may involve direct phosphorylation of the raf protein kinase. Stimulation of PLC-β with consequent diacylglycerol generation and Ca2+mobilization may therefore play a role in MAP kinase activation by both Gq and Gi (in this case by βγ-subunit-mediated activation) family G proteins (223). Other data suggest that there are PKC-independent mechanisms for G protein-coupled receptor-mediated MAP kinase activation. One mechanism employed by receptors for a variety of agonists including lysophosphatidic acid (LPA) and bombesin may involve phosphorylation of the shc adapter protein by an as yet unidentified tyrosine kinase resulting in SOS recruitment and ras activation (574-576). Work from several laboratories has begun to define roles for PH-domain-containing proteins as effectors for βγ-subunits, and in this regard, Bruton’s tyrosine kinase is an attractive effector (482). Recombinantly expressed PH-domain-containing proteins including the β-ARK-1 PH domain exert dominant negative effects on G protein receptor-mediated MAP kinase activation, which also suggests an effector role for βγ-subunits in this pathway (252,256, 304).
More recent studies have presented evidence that several distinct nonreceptor tyrosine kinases couple GPLR and the MAP kinase regulatory network. The high rate of recombination in avian B (DT40) cells afforded the ability to examine what role, if any, the Src-related tyrosine kinases Lyn and Syk played in the signaling via G proteins (Gq and Gi) coupled to muscarinic acetylcholine receptors (588). Interestingly, deficiency of Lyn was found to block signaling to the level of MAP kinase via Gq, but not via Gi. Loss of Syk eliminated muscarinic signaling via both Gq and Gi. These were the first insights into possible cross talk at the level of heterotrimeric G proteins.
The mechanism(s) by which G proteins may activate protein tyrosine kinases and thereby elements of the MAP kinase regulatory network have been under intense investigation. Direct stimulation of Bruton’s tyrosine kinase by the α-subunit of Gq was established via reconstitution of purified G protein subunits with Btk. Only the α-subunit of Gq and not those for Gi-1, Go, and Gz activated Btk in the presence of GTP (25). The Gβγ subunits examined failed to activate Btk and the elimination of Btk block signaling from G protein-coupled receptors and Gq to MAP kinase, p38 (25).
More recently, Btk has been shown to be an effector for both Gq as well as G12 proteins (268). Additionally, in this same study, G12 protein is shown to stimulate the activity of a Ras-GTPase-activating protein (Gap1) both in vitro and in vivo. A PH domain and Btk motif appear to constitute necessary binding motifs for G12 binding to both Btk and Gap1. These data herald a new dimension in our understanding of G protein signaling in which the artificially distinct linear pathways of GPLR and tyrosine kinases signaling are replaced by an interactive network rich with cross talk between these two prominent modalities of cell signaling.
D. Regulators of G Protein Signaling
As discussed in greater detail in sectioniiC , interaction with certain effectors can stimulate the GTPase activity of their cognate Gα subunit activators. The best-characterized examples of RGS activity are the phospholipase C-β enzymes that stimulate GTPase activity of Gαq in vitro and the Pγ subunit of cGMP phosphodiesterase which stimulates GTP hydrolysis by Gαt. It seems likely that other G protein effectors also may function as GTPase activators for their activating Gα subunits (AC and ion channels being likely candidates).
2. RGS proteins
Derivatives of genetic studies in yeast and nematodes, a large family of mammalian proteins that themselves appear to have no effector functions, have been shown to act as selective GTPase activators and thereby RGS proteins (32, 135,136). These regulators of G protein signaling act on a range of Gα subunits. The S. cerevisiae gene productsst2 was identified using genetic approaches as a negative regulator of the G protein-dependent pheromone signaling pathway. Inactivating mutations in this gene produce a supersensitivity to pheromone phenotype (140, 296). Further genetic analysis indicated that sst2 was acting at the level of a G protein, although a biochemical description of the mechanisms of action of sst2p was lacking. A nematode gene EGL-10 that functioned as a negative regulator of the egg-laying response also was identified. The protein encoded by the EGL-10 gene displayed a pronounced sequence similarity to sst2p(306).
The realization that multiple homologs of sst2 andEGL-10 are present in higher eukaryotes led to a rapid identification of a family of mammalian RGS proteins (32). Biochemical analyses in vitro using recombinant proteins as well as studies in intact cells reveal that these RGS proteins are selective GTPase activators for certain heterotrimeric Gα subunits. GAIP is a human RGS protein that has been shown to interact with Gαi-3. GAIP is expressed in many tissues (33). A numerical suffix has been adopted to assign names for the remaining mammalian RGS proteins. RGS homologs including GOS8 and BL34/1R20 exhibit more restricted expression patterns (192). Some RGS proteins, such as RGS1 (BL34/1R20), have expression restricted to a single cell type (B lymphocytes). RGS2 (GOS8) is found only in human blood lymphocytes. RGS3 was cloned from a B-lymphocyte cDNA library using an oligonucleotide probe corresponding to a sequence conserved between RGS1 and -2 (231). RGS4 was discovered through an expression cloning strategy designed to identify rat brain cDNA that could stimulate recovery from pheromone-induced growth arrest when expressed in yeast (444). A number of studies indicate that RGS1, RGS2, RGS3, and RGS4 all regulate G protein signaling in mammalian cells. Transient expression of RGS1, for example, blunts MAP kinase activation in response to platelet-activating factor (PAF), whereas expression of each of the four RGS proteins attenuates MAP kinase activation by interleukin-8 (192).
The results from a growing number of reconstitution studies indicate that RGS proteins act as GAP for Gα subunits. On the basis of single-turnover kinetic studies, either RGS4 or GAIP produce a 40-fold increase in the rate of GTP hydrolysis by Gα subtypes. This work led to speculation that RGS proteins may accelerate GTP hydrolysis by stabilizing G protein α-subunits their active conformation (227). The recent determination of the crystal structure of an RGS2 Gαi-1 complex reveals that the core domain of RGS2 does in fact bind to the various switch regions of the G protein α-subunit. Unlike GAP for small GTPases, however, RGS2 does not contribute catalytic residues that interact directly with the substrate. The RGS2 Gαi-1 structure reveals also that the binding site for RGS4 on Gαi would be expected to preclude interactions of the G protein with its effectors (557).
3. Receptor-activity modifying proteins
Receptor-activity modifying proteins, termed RAMP, have been isolated and cloned that regulate the transport and ligand specificity of the calcitonin receptor-like receptor (CRLR) (372). Composed of at least three members, the RAMP are single-transmembrane proteins with a molecular mass of ∼14 kDa. The tissue expression of mRNA of RAMP differs among RAMP1, -2, and -3 but appear broad in humans. The RAMP are essential for the transport of the CRLR to the plasma membrane. If RAMP1 presents the CRLR to the plasma membrane, the ligand specificity of the mature glycoprotein receptor is that of calcitonin gene-related peptide. If RAMP2 presents the receptor, it is core-glycosylated and displays a ligand specificity of the adrenomedullin receptor. The extent to which RAMP contribute physiologically to the ability of tissue/cells to respond to different peptides remains to be established. Likewise, the family of GPLR sensitive to RAMP regulation remains unknown.
III. TRANSCRIPTIONAL REGULATION OF SIGNALING PATHWAYS
A. Physiological Perspective
Very early in the study of physiological regulation now known to be G protein mediated was it observed that hormones whose actions typically require hours or days to be manifested were in some way influencing the ability of other agents to act acutely. Often referred to as “permissive” in nature, this level of regulation was obvious in the action of many hormones, such as thyroid hormone, glucocorticoids, and sex steroids, that act on nuclear receptors (373). The overproduction or lack of such permissive hormones often had profound effects on the ability of other hormones such as catecholamines, neurotransmitters, and drugs to act acutely at target sites. The constellation of pathophysiologies that accompany overproduction/underproduction of glucocorticoids (Cushing’s disease/Addison’s disease) or thyroid hormones (hyperthyroidism/hypothyroidism), for example, are well known to the physiologist and include a large dose of permissive effects exerted on G protein-mediated signaling. To highlight the basis for the regulation of G protein-mediated signaling by permissive hormones that act on nuclear receptors, a few prominent examples that have been explored in greater detail are described. The mechanisms supporting the physiological regulation of the examples displayed are likely candidates for the mechanisms that are involved in permissive effects of steroids, retinoids, and thyroid hormones on other G protein-mediated pathways.
B. Transcriptional Activation
For steroids, retinoids, thyroid hormones, and other agents that operate through nuclear receptors (373), transcriptional regulation is the prominent mechanism of action controlling physiological output of this class of hormones (117,360). Responses mediated by catecholamine receptors, especially the β-adrenergic receptors, are exquisitely sensitive to the effects of hormones that act on nuclear receptors (373). In some respects, the expansion of our knowledge of nuclear receptor function and gene expression has simplified the task of explaining, for example, how hyperthyroidism results in increased cardiovascular tone and enhanced signaling by a variety of hormones, including the catecholamines.
1. Thyroid hormones
Hyperthyroidism leads to amplified responses to catecholamine stimulation, and the effects appear to be manifest at the most proximal point of the signaling paradigm for GPLR, the receptor itself (13, 14). Thyroid hormone administration in vivo or in vitro mimics hyperstimulation of adrenergic pathways, and the changes in the abundance of β1- and β2-adrenergic receptors mirror these changes. Administration of thyroid hormones leads to an overproduction of the receptors, manifest in the time frame of hours to days typical of agents acting through nuclear receptors. Despite the fact that the pathophysiology of thyroid hormone overproduction on adrenergic signaling has been known for decades, the precise linkage between thyroid hormone and the expression of the β1-adrenergic gene has been established only quite recently.
Thyroid hormone treatment of cells was shown very early to stimulate the expression of β1-adrenergic receptors in cardiac muscle cells in culture. The 5′-flanking regions for the human, macaque, rat, mouse, and ovine β1-adrenergic receptor genes have been cloned (355, 494). Analogous to the β2-adrenergic receptor, the proximal promoter regions of these genes are rich in GC sequences and lack a well-defined TATA box. The transcriptional start site (TSS) for rat and human β1-adrenergic receptor genes is found at approximately −253 relative to the ATG (494). The machinery involved in regulating basal expression and hormonal responsiveness of the β1-adrenergic receptor gene is still obscure. The transcriptional activity of the full-length promoter of the β1-adrenergic receptor gene extending from −3311 to −1 is ∼10% of a truncated promoter extending from 484 and −1. Treating ventricular myocytes with thyroid hormone was shown to increase the expression of receptor mRNA within 1–2 h of challenge (13). The ability of thyroid hormones to elevate receptor mRNA levels correlated well with their ability to increase the relative rates of transcription of the receptor gene in nuclear run-on assays within only minutes of addition of thyroid hormone to cells in culture. This gene activation and increase in mRNA accumulation was sensitive to inhibition by actinomycin D. The transcriptional regulation of genes by thyroid hormones has been the focus of years of detailed analysis. It is clear that regulation of the genes encoding elements of G protein-linked signaling by thyroid hormones offers little mechanistically to that already known. More importantly, current research has provided molecular descriptions of prominent examples of physiological regulation by thyroid hormones expressed in G protein-linked pathways, much like the basis for enhanced adrenergic tone in hyperthyroidism and blunted cardiovascular responses in hypothyroid states in humans. Detailed information of these molecular details was remarkably obscure a decade ago.
Thyroid hormones (3,3′,5-triiodothyronine, T3) play an important role in the regulation of the hormone-sensitive AC system in the heart. Both α- and β-adrenergic receptor action in the heart is regulated by thyroid status (13, 376). Perinatal administration of propylthiouracil to yield hypothyroidism provoked a decline, whereas T3 administration stimulated an increase, in the expression of cardiac α1- and α2-adrenergic receptors (376). The transcription of the β1-adrenergic receptor gene is increased in response to T3 in the heart in vivo and in neonatal rat ventricular myocytes in primary culture (13). Altered thyroid status also influences the β1-adrenergic receptor signaling pathway in the heart by regulating the abundance and coupling of G protein α-subunits (14). These effects form a basis for the marked influence of T3 on β-adrenergic catecholamine action in the heart, which is blunted in hypothyroidism and markedly potentiated in hyperthyroidism.
Progressive truncations in the promoter region of the rat β1-adrenergic receptor gene revealed that the 5′-untranslated region (5′-UTR) can be truncated to −394 without significant loss of activity. Foreshortening of the 5′-UTR to −330, in contrast, abrogated the activity of the promoter (15). Therefore, the minimal promoter of the β1-adrenergic receptor gene appears to be contained within 60 bp of sequence between −389 and −330 of the 5′-UTR. The short promoter enables high levels of expression in all cells, even those that do not normally express β1-adrenergic receptor. Regulatory regions of the β1-adrenergic receptor promoter were localized between two regions that lie 5′ and 3′ to the transcriptional start site. The 3′-regulatory region suppresses expression of the gene and could be localized to the nucleotides between −125 to −100, relative to the ATG. A series of 4-bp mutations introduced across this region in the context of the −3311 to −100 permitted detailed analysis of the β1-adrenergic receptor promoter (Fig.8). Mutations of the sequences between −121 and −118 (domain D) restored nearly full promoter activity, whereas the mutations of domains A, B, and C had no effect on basal activity. The 5′-regulatory region was localized to the sequences between −2870 and −2740 (15). These between −2870 and −2740 in tandem participate with those between −125 to −100 are responsible for low levels of expression of the β1-adrenergic receptor in specific cells and tissues.
The T3-responsive element (TRE) in the β1-adrenergic receptor promoter has been identified between domains A and C in the −125 to −100 region by site-directed mutagenesis in tandem with transient transfections into ventricular myocytes (15). The TRE generally are composed of hexametric half-sites oriented as direct repeats of an AGGTCA motif spaced by 4 bp, palindromes separated by 0–1 bp or inverted palindromes spaced by 6 bp (572). The β1-adrenergic receptor TRE is organized as a direct repeat of the motif that is separated by 5 bp instead of the usual 4 bp (Fig. 8). This organization results in a TRE that functionally is weaker than comparable 4-bp spaced TRE. The β1-adrenergic receptor TRE bound the nuclear thyroid hormone a andb as monomers and even better as heterodimers with the retinoid X receptor (15, 432). These results suggest that in vivo, nuclear thyroid hormone receptors bind preferentially to this site as a heterodimeric complex containing retinoid X receptor.
Thyroid hormone effects on G protein expression have been reported in a few tissues (15, 453). For adipocytes, alterations in thyroid states in vivo have been shown to result in alterations in catecholamine responsiveness (592). Hypothyroidism, for example, elevates the expression of Gαi-2 and Gαo, while increasing the levels of Gβγ subunits to a greater extent (453). These changes in the relative abundance of Gαi-2 may well provide the basis for the impaired lipolytic response of the adipocytes from hypothyroid animals, while not altering the expression of the β-adrenergic receptors in these cells. In neonatal rat ventricular myocytes, thyroid hormone increases the expression of Gαsby four- to fivefold, while suppressing the expression of Gαi-2 (14). These thryoid hormone-induced changes in the expression of G protein subunits and β-adrenergic receptors provide a biochemical explanation for the enhanced cardiovascular response noted for in hyperthyroidism and a generalized model for influences of thyroid hormones on G protein signaling.
Equally prominent as a regulator of G protein-linked signaling are the glucocorticoids. Glucocorticoids administered in vivo lead to changes in physiology mediated by GPLR, including adrenergic receptors (202, 321), muscarinic receptors (151, 243), and those for serotonin (507). G protein expression is also sensitive to administration of glucocorticoids (290, 383,471) or deficiency (470) in response to adrenalectomy. The β-adrenergic receptors provide an ideal model for the study of the permissive effects of glucocorticoids on G protein signaling elements. Naturally occurring and synthetic glucocorticoids, corticosterone and dexamethasone, respectively, exert a profound effect on catecholamine responses. Glucocorticoids potentiate and adrenal steroid insufficiency blunts many of these G protein-mediated pathways. The physiological basis for the actions of glucocorticoids are treated exhaustively elsewhere (487). The potentiation of β-adrenergic action by glucocorticoids reflects their effects on β-adrenergic receptor expression. A variety of cells in culture display a sharp increase in β-adrenergic receptor expression at 12–48 h after challenge with dexamethasone (117,196, 202, 359). Many features of glucocorticoid regulation of β-adrenergic receptors observed in acutely prepared white and brown fat cells in vivo are also observed with some fidelity in various model cell lines in culture (321).
Early studies of glucocorticoid action on receptor mRNA levels provided the first insight, i.e., a threefold increase in β-adrenergic receptor mRNA within 2 h of challenge with dexamethasone (118, 202, 592). The ability of actinomycin D to block the response, the relatively stable half-life of the receptor mRNA in challenged cells, and the character of the response argued for transcriptional activation of the receptor gene. Nuclear run-on assays of transcriptional activation revealed a sevenfold increase in the relative rate of transcription for the β-adrenergic receptor gene in nuclei isolated from steroid-treated as compared with naive cells. Although not fully detailed, glucocorticoid-response elements (GRE) have been identified in the genes of many GPLR and elements, including the β1- and β2-adrenergic receptors. For the mammalian β2-adrenergic receptor, a GRE located at position −1031 in the 5′-UTR appears to be obligate for glucocorticoid-induced activation of the gene (201,359). Disruption of this GRE abolishes glucocorticoid-induced elevation of receptor mRNA. Similar studies have revealed that α2-adrenergic receptor expression can be upregulated in response to glucocorticoids (213). Studies of α2-receptor expression in several pancreatic β-cell lines in culture reveal that glucocorticoids, but not progestins and sex steroids, increase expression of the receptor as well as the downstream signaling mediated by these receptors (213). It is likely that many other members of the GPLR superfamily are selectively upregulated by steroid hormones. Only few of the members of the GPLR superfamily have been studied to date.
Regulation of G protein subunits by glucocorticoids has been studied in a variety of systems. Adrenalectomy leads to a loss of catecholamine-stimulated cAMP accumulation and lipolysis (470). In fat cells, adenalectomy leads to a loss in the amount of Gαs and Gβ/γ subunit expression (470), providing an explanation for the loss of lipolytic responsiveness. Dexamethasone treatment of adrenalectomized animals leads to rectification of the catecholamine responsiveness of the fat cells derived from the animals. Changes in the levels of expression of G protein subunits in adrenalectomized animals reflect changes in subunit mRNA and suggest that glucocorticoids activate genes encoding Gαs and Gβ/γ subunits. In the liver, studies have clearly shown that glucocorticoid administration in 2- to 3-day-old neonatal animals leads to increased expression of Gαs, Gαi-2, and Gβ/γ subunits (290,289). The increase in Gαs appears to be responsible for the increased AC activity observed in the neonatal liver of animals challenged with glucocorticoids. Treating osteosarcoma UMR-106–01 cells with dexamethasone in culture has been shown to increase both the expression of Gαq/11 as well as PLC-β activity (383). At least one study reports the ability of glucocorticoids to increase AC signaling, in the absence of changes in Gαs or Gαi, suggesting regulation of the expression of the AC catalyst itself (378). The promoter regions of genes encoding G protein subunits (237), AC, and PLC-β isoforms have not been characterized in sufficient detail to allow description of the GRE that may be responsible for glucocorticoid-regulated expression.
The regulation of GPLR by retinoids has not been studied extensively. Retinoic acid is an important morphogen in early mammalian development and undoubtably regulates signaling events mediated by G proteins. Retinoic acid increased transcription of β1-adrenergic receptor gene in mouse F9 teratocarcinoma stem cells (173). Expression of β1- but not β2-adrenergic receptors increases severalfold in response to retinoid acid treatment. It has not been established if the increase in β1-adrenergic receptor expression is a direct result of retinoic acid or a reflection of the morphogen stimulating the stem cells to differentiate to primitive endoderm. Analysis of the promoter sequence of the β1-adrenergic receptor gene has been informative. For the gene encoding the β1-adrenergic receptor, sequences between −125 and −100 contain two direct repeats of the AGGT (A/G) motifs that are separated by 5 bp. This organization is more efficient as a retinoic acid responsive element than as a TRE (572).
Recent studies have focused on whether or not the region between −125 and −100 confers retinoic acid responsiveness to the β1-adrenergic receptor gene. Transient transfection experiments have been conducted in F9 cells with chimeric genes containing the β1-adrenergic receptor promoter driving the luciferase reporter gene. Because the effect of retinoic acid on β1-adrenergic receptor expression occurs in the primitive endodermal F9 phenotype, F9 cells must be exposed first to 100 nM retinoic acid (or buffer as a control) for 2 days before transfection. Cells are then transiently transfected and reexposed to either 100 nM retinoic acid buffer for an additional 48 h. Transcription from the promoter containing the −125 to −100 sequence was found to increase threefold in F9 cells in response to retinoic acid. The induction of gene expression by retinoic acid was dependent on the repeats A and C, indicating that the same nucleotides were involved in mediating the thyroid hormone as well as the retinoic acid responses (15, 432).
4. Sex steroids
Estradiol and progestins appear to have little generalized effects on G protein signaling, although increases in expression of β-adrenergic receptor expression have been noted in lymphocytes of estrogenized female patients (378, 400). Testosterone, in contrast, is associated with changes in adrenergic signaling. In adipose tissue, kidney, and prostate, increases in β2-adrenergic receptor expression accompany administration of testosterone in vivo (117). Studies performed in the rat ventral prostate reveal depression in the expression of β2-adrenergic receptors within 4 days of orchidectomy (120). Testosterone administration to castrated animals provoked activation of the receptor gene, elevation of receptor mRNA, and a 3.5-fold increase in β2-adrenergic receptors within 8 h (120). These data suggest that in the context of the well-known target tissue for sex steroid action, changes in the expression of G protein signaling elements can be observed after administration of a steroid. The physiological significance of these changes in signaling elements remains obscure. Genes for other GPLR display transcriptional activation by sex steroids. The gene for α1β-adrenergic receptor shows estrogen-dependent activation in the hypothalamus-preoptic area of the brain (281), whereas in the cortex, estrogen appears to reduce α1β-receptor mRNA levels (281). The α2D-adrenergic receptors in the hypothalamus-preoptic region, in contrast, display a sharp reduction in response to estradiol treatment in vivo (282). Estrogen regulation of D2 dopamine receptors in the brain has been suggested to involve changes in splicing, rather than a solely transcriptional control mechanism (197).
Although confined largely to studies on white adipose tissue in rodents, levels of sex steroids have been shown to influence the expression of G proteins. Ovarectomized animals show a decline in mRNA and expression of Gαs, but neither Gαi-1nor Gαi-2 in epididymal fat (131,132). Estradiol replacement rectified these changes in G protein expression. Similarly, studies in castrated rodents reveal a sharp decline in Gαs, Gαi-1 and Gαi-2 levels, but only in epididymal fat tissue (138, 321). In this case, the decline in Gαs, but not Gαi-1 and Gαi-2, was fully rectified by treatment with testosterone (138). Because all of the isoforms of AC known to date are activated by Gαs, these changes in Gαs expression stimulated by estrogenic and androgenic hormones may play an important role in setting the tone of the cAMP-PKA-CREB sensing pathways in these and perhaps other target sites. Detailed analysis of steroid responsive elements in the promoter regions of genes constituting the receptors, G proteins, and effectors in these signaling devices has not been reported.
Activation of genes encoding elements of G protein signaling is not restricted to hormones such as glucocorticoids, thyroid hormones, and retinoids but includes a unique version of a positive, or feed-forward loop of regulation. In this type of regulation, the output of a signaling pathway would be expected to amplify the expression of the elements of the pathway from which it is produced. One prominent example of a feed-forward loop is the regulation of the gene encoding the β2-adrenergic receptor by the product of the effector AC, i.e., cAMP. Short-term challenge of cells in culture with a β-adrenergic agonist leads to a three- to fivefold increase in the level of receptor mRNA (115). Receptor mRNA stability is relatively constant at these early times, and nuclear run-on experiments established transcriptional activation of the β2-adrenergic receptor gene. Analysis of the 5′-UTR of the receptor gene identified a palindrome with the motif TGACGTA which can constitute a cAMP response element (CRE) capable of binding its specific binding protein termed, CREB. A reporter gene fused to the 5′-UTR of the human β2-adrenergic receptor gene provided a direct test of the hypothesis that a functional CRE was present (114). In tandem with DNA footprinting and gel-shift analysis, the presence of the CRE was established, providing the molecular basis for the elevation of the receptor mRNA that can occur rapidly upon agonist stimulation.
The presence of a functional CRE in the β-adrenergic receptor gene presents a paradox to the physiological responses and biochemical changes most often associated with stimulation of G protein signaling, namely, attenuation of the signal throughput. Countless reports have documented and studied the well-known desensitization and downregulation of GPLR that occur for β-adrenergic receptors and virtually all other members of this superfamily. If a CRE is commonly found in the 5′-UTR of these genes, why does agonist stimulation provoke a general decline in receptor expression, i.e., agonist-induced downregulation? Agonist-induced destablization of receptor mRNA has been discovered and described biochemically in recent years (see sect. iv). Perhaps the transcriptional activation of a CRE is not sufficient to compensate for a posttranscriptional destabilization of the receptor mRNA. Evidence in support of this proposal is provided by analysis of a similar relationship, when cells are simultaneously challenged with a glucocorticoid (upregulation) as well as an agonist (downregulation). Depending on the timing and extent of the opposing stimuli, either receptor upregulation by glucocorticoid can dominate or agonist-induced downregulation can dominant (208). Thus the regulation appears to be dynamic, capable of pushing the levels of receptor mRNA to achieve either increased synthesis (glucocorticoid) or decreased synthesis (agonist). The CRE may provide a rectifying element to the agonist-induced downregulation of receptor synthesis by ensuring increased synthesis and continued availability of receptor during short periods of sustained agonist stimulation.
It is well known that G protein signaling provides “cross-regulation” among the various pathways. Activation of any one pathway generally leads to attenuation of the pathway as well as potentiation of an opposing pathway. For example, activation of the stimulatory AC pathway attenuates expression of the receptors and perhaps Gαs, while stimulating a fourfold increase in the mRNA and product of the opposing Gαi-2-mediated pathway (207). One could envision the existence of CRE in the genes encoding elements of the inhibitory AC pathway that oppose the stimulatory response. Detailed analysis of the promoter and 5′-flanking regions of the genes encoding Gαi family members would provide a direct test of this intriguing possibility.
C. Transcriptional Repression
Gene regulation most often is discussed in terms of activation, as exemplified by the strong transcriptional activation observed for many G protein-signaling elements in response to thyroid hormones, sex steroids, and glucocorticoids. Although far less common that activation, repression of transcription does occur in the case of a least one well-known member of the GPLR superfamily, the β1-adrenergic receptor. Glucocorticoids stimulate a robust expression of β2-adrenergic receptors, while decreasing the steady-state levels of expression of β1-adrenergic receptors (196,291). The glucocorticoid-induced decline in β1-adrenergic receptors has been explored in some detail in the rat C6 glioma cell line, a cell line which expresses both β1- and β2-adrenergic receptors. Radioligand binding and immunoblotting demonstrate a frank increase in the β2- and a sharp decline in the β1-adrenergic receptor expression (291). Levels of β1-adrenergic receptor mRNA decline within 2 h of challenge of the cells with dexamethasone. Nuclear run-on analysis of the regulation revealed that the relative rate of transcription of the β1-adrenergic receptor gene was reduced in nuclei prepared from cells treated with the glucocorticoid (291). Study of glucocorticoid regulation of β1- and β2-adrenergic receptor expression in mouse 3T3-L1 embryonic fibroblasts provide additional support for the data obtained with C6 cells (196). Recent studies using a reported gene construct and the 5′-flanking promoter region of the β1-adrenergic receptor gene reveal a glucocorticoid-responsive domain capable of repressing gene transcription located between nucleotides −950 and −926 relative to the translational ATG in the rat β1-adrenergic receptor gene (16). The physiological output of the a glucocorticoid-induced upregulation of β2-adrenergic receptors and downregulation of β1-adrenergic receptors would likely alter the pharmacological character of the target tissues or cells to be predominantly β2-adrenergic.
Other members of the GPLR superfamily display downregulation in response to glucocorticoids. The 5-HT1A receptors of the hippocampus display a marked increase in expression in adrenalectomized animals. Glucocorticoid treatment rectifies this increase in 5-HT1A receptor expression (637). Nuclear run-on assays show an increase in the relative rate of transcription in nuclei from the adrenalectomized animal tissue (637). Treating frontocortical astrocytes with dexamethasone downregulates the expression of the 5-HT7receptors. Taken together, these data suggest that the genes for some 5-HT receptors are transcriptionally repressed by glucocorticoids. The downstream consequences of these changes in serotonin receptors in the central nervous system are not known. Muscarinic receptor expression in the canine airway (151) and guinea pig lung (486) display a similar response to glucocorticoids, i.e., a downregulation. The decrease in M2 and M3receptors in the canine airway may provide an explanation for some of the well-known benefits of glucocorticoids for pulmonary conditions, in addition to the increase in β-adrenergic tone described in greater detail above.
In astroglial cells in culture, study of the role of cAMP in the expression of α2A-adrenergic receptors has revealed evidence for a transcriptional repression of the α2A-receptor gene. Treating the cultures directly with cAMP analogs or the diterpene forskolin stimulated a profound (90%) decrease in the mRNA levels of the α2A-receptor (462). The mRNA half-life was unaffected by elevation of intracellular cAMP levels, suggesting that transcriptional repression was the mechanism by which cAMP was exerting its negative influence on expression of the α2A-adrenergic receptors. Study of the effects of activating PKC by phorbol esters in these same cells suggests that like cAMP, activation of PKC provokes a marked transcriptional repression of the α2A-adenergic receptor gene (462).
D. Transcriptional Basis for Physiological Regulation
On the basis of a large volume of published work since 1987 highlighted in this section, transcriptional control emerges as an important element of physiological regulation. The examples of hyper- and hypothyroidism, of Cushings syndrome and adrenal insufficiency, and of alterations in G protein signaling that accompany puberty and menopause provide ample evidence that the basic operation of G protein-linked signaling responds to changes in agents that alter transcription of the gene that encode the receptors, G proteins, and effectors that compose the basic pathways. The extreme examples that enable us to identify and characterize the underlying biochemical/molecular basis for the physiological regulation should not distract us from the obvious conclusion that in the “normal” physiological state these agents contribute to establishing the “gain” in these macromolecular signaling devices. Equally important, pharmacological intervention in a myriad of pathophysiological conditions that require the use of endocrine agents, hormones, and drugs have the potential of altering the signaling pathways described above. As our knowledge of G protein signaling in the central nervous system expands, the molecular basis for the psychological dimensions of endocrine agents may become much more lucid in terms of transcriptional regulation of the GPLR, heterotrimeric G proteins, and effectors that function in neurotransmission and sensory perception.
IV. POSTTRANSCRIPTIONAL REGULATION OF SIGNALING PATHWAYS
1. Regulatable mRNA
The steady-state expression of proteins, including the elements of G protein-linked pathways, reflects to a large degree the activation and/or suppression of the individual genes encoding these elements. Regulation of gene expression is not the only mechanism for controlling the level of mRNA and protein. Over the past decade, regulation of steady-state levels of mRNA have been shown to include an important posttranscriptional component (261, 337, 475,476, 613). Although first revealed by studies of mRNA with remarkably short half-lives (58,59, 110, 217, 438,474, 496), posttranscriptional regulation of mRNA is not confined to this small subset of cases but applies to a much broader group of mRNA, including those that encode elements of G protein-signaling devices.
The intracellular levels of highly regulated mRNA (e.g., mRNA of granulocyte/macrophage colony-stimulating factor, tumor necrosis factor-α, and the oncogenes c-fos and c-myc) are markedly influenced by the rate of degradation (261,476). Regulation of mRNA stability and turnover has been shown to be multifaceted, reflecting not only various cytosolic and nuclear-associated factors and polyadenylation, but also cognate sequences of the 3′-UTR of mRNA, such as the tandem repeats of AUUUA pentamers (41, 61, 90-92,362, 438, 508,573), AUUUUA hexamers (561), and nonamers, such as UUAUUUA (U/A) (U/A) (322) and UUAUUUAUU (638).
Ribonucleic acid binding proteins implicated in regulating mRNA stability and turnover can be segregated and discussed in terms of several classes. The heterogeneous, nuclear ribonucleoprotein particles display roles in several steps of mRNA maturation including the packaging, translocation, and splicing of heterogenous nuclear RNA (143, 541, 613). The splicing and further processing of pre-mRNA, which possess introns, a 5′-cap and 3′-poly (A)+ tract involves yet another class of RNA-binding proteins, termed “sNRP”, the small nuclear RNA-binding proteins (308, 525). Messenger RNA binding proteins predominantly cytosolic in nature include the 72,000-kDa poly(A)-binding protein that binds to long (∼25 nucleotides/protein) stretches of poly(A)+ and stabilizes the RNA to 3′- to 5′-nuclease activity (34). A class of smaller (ranging from 30,000 to 40,000 kDa) cytosolic mRNA-binding proteins have been identified that display recognition of AU-rich domains in the 3′-UTR of mRNA (41,61, 362, 404, 446,573, 573, 634). Despite the fact that several of these RNA-binding proteins have been purified to near homogeneity, the precise role(s) that these smaller RNA-binding proteins play in regulating mRNA stability and turnover has yet to be fully elucidated.
2. GPLR mRNA
Virtually all members of the large superfamily of GPLR display attenuation and downregulation in the face of a chronic challenge with agonist. Presumed to reflect a common need to rectify the sensitivity of a chronically stimulated pathway, agonist-induced downregulation of the elements in a pathway serves to restore balance among integrated pathways. For GPLR, e.g., the β2-adrenergic receptor, agonist-induced downregulation of GPLR provides an explanation for long-term attenuation to chronic stimuli characteristically observed for members of this receptor superfamily (16,16, 117, 119, 201,203, 221, 446,590). β2-Adrenergic receptors provide a useful model in which to discuss this basic paradigm in signaling. Steady-state levels of the β2-adrenergic receptor and of its mRNA decline sharply after a challenge with agonist (201, 206, 208). The basis for the decline in receptor mRNA induced by agonist is not transcriptional suppression but rather posttranscriptional destabilization of receptor mRNA (208).
B. Destabilization of mRNA
1. Agonist-induced destabilization of GPLR
The ability to measure accurately the absolute level of “rare” mRNA for low-abundance membrane proteins provided a major advance to understanding the mechanisms by which their steady-state levels were being regulated physiologically. First reported by DNA-excess solution hybridization, the levels of β2-adrenergic receptor mRNA were established to be ∼0.5 pg/μg total cellular RNA or ∼1.0 attomole (19–18 mol) of receptor mRNA/μg total cellular RNA (201, 202). As noted in sectionivA, transcription of the genes encoding GPLR, G proteins, and effectors were found to display complex regulation. On the basis of the literature, it was quite logical to speculate that the reduction in receptor mRNA that accompanied chronic stimulation of agonist was transcriptional suppression. For the β2-adrenergic receptor, reduced transcription seemed the likely candidate for the agonist-induced decline in receptor mRNA (Fig. 9). Depending on the cell type examined, the extent and time course of downregulation of receptor mRNA was quite variable. For those GPLR whose mRNA levels have been studied under conditions of agonist-induced decline of receptor expression, the loss of receptor mRNA is 50–70%, and the time courses range from 2 to 12 h for the full decline (201,243, 560).
The hamster β2-adrenergic receptor gene displays two consensus sequences for CRE (114-117). The linkage between agonist activation of AC and elevation of intracellular cAMP levels fostered the novel hypothesis for the β2-adrenergic receptor gene that the CRE were repressing rather than activating transcription (201). Analysis of the rate of transcription using nuclear run-on assays with nuclei prepared from control as compared with agonist-treated cells revealed no inhibition (208). Furthermore, poisoning of transcription with actinomycin D had no effect on the extent or time course of agonist-induced decline in receptor mRNA. The critical observation to our understanding of agonist-induced downregulation of receptor and receptor mRNA was provided through the analysis of the mRNA half-life. The half-life for the mRNA for various GPLR was established by use of actinomycin D to block transcription and analysis of mRNA levels using either DNA-excess solution hybridization or use of riboprobes. Comparison of the half-lives for β2-adrenergic receptor mRNA obtained from control cells versus cells challenged with β2-adrenergic agonist for 12 h showed a decline from a half-life of 12–14 h to a half-life of ∼6 h (201, 206,208). Other cell types displayed the same agonist-induced decline in the half-life of β2-adrenergic receptor mRNA, providing the first demonstration of a posttranscriptional mechanism for physiological regulation of a member of the superfamily of GPLR.
2. cAMP-induced destabilization of GPLR
β-Adrenergic agonist or treatment of cells with analogs of cAMP displayed the ability to destabilize pools of existing receptor mRNA, although agonist-induced destabilization was more profound than that of cAMP alone (206). Further insight into the mechanism of posttranscriptional regulation of receptor mRNA was provided by analysis of the role of agonist versus cAMP in variants of the S49 mouse lymphoma cells, variants with specific lesions in the stimulatory AC pathway. The UNC, cyc−, and H21A variants possess lesions proximal to AC and display downregulation of β2-adrenergic receptor mRNA if challenged with the diterpene activator of AC forskolin (188). Forskolin activates AC directly and increases intracellular cAMP accumulation. Agonist treatment of UNC and cyc−, however, failed to stimulate agonist-induced downregulation, reflecting the loss of receptor-Gαs signaling in these variants (206). The H21A variant of Gαs which couples to receptor, but not to AC, displayed agonist-induced downregulation of β2-adrenergic receptor mRNA, but to a reduced extent. Based on the results obtained with these variants, downregulation of receptor mRNA appears to be induced by elevation of intracellular cAMP and also by some mechanism in which activation of AC by agonist is not obligate. The kin− variant of S49 mouse lymphoma are devoid of catalytic subunit of PKA (188). The kin− cells respond to either agonist or forskolin by increasing intracellular cAMP levels, reflecting the downstream lesion in this variant. Interestingly, the kin− cells did not display downregulation of receptor mRNA in response to either agonist or forskolin (208). Thus agonist-induced downregulation of receptor mRNA, although not requiring activation of AC, requires PKA activity for expression. Because expression of CREB itself requires active PKA, it is not possible to determine if the loss of agonist-induced downregulation of receptor mRNA is due to the loss of PKA activity, CREB activity, or both. The nature of other factors, other than cAMP, in the mediation of posttranscriptional downregulation of receptor mRNA remain obscure.
C. β-ARB Protein, a GPLR-Specific mRNA Binding Protein
1. β-ARB protein and β-adrenergic receptor mRNA
Discovery of a posttranscriptional mechanism for physiological regulation of G protein signaling elements, such as the β2-adrenergic receptor, focused attention on the turnover of receptor mRNA in response to agonist stimulation as well as to the protein(s) involved in the process. Studies of RNA binding proteins have been facilitated by the use of ultraviolet-induced cross-linking of radiolabeled RNA species to binding proteins, followed by digestion of the nucleic acid and SDS-polyacrylamide gel electrophoretic separation and identification of the binding protein(s) (613). Through the use of photo-cross-linking of labeled mRNA molecules to proteins in cellular extracts, a 35,000-kDa protein was identified with properties consistent with those expected for an RNA-binding protein selective for mRNA of receptors that display agonist-induced downregulation of their messages and protein expression (446). Identification of proteins specifically involved in the binding of RNA was enabled by the ability to create uniformly radiolabeled, capped, polyadenylated, mRNA for a variety of GPLR, including those that display and those that do not display agonist-induced destabilization of receptor mRNA. Adaptation of this strategy to the most thoroughly studied GPLR, the β2-adrenergic receptor, identified this highly specific 35-kDa protein in cell extracts (446).
The 35,000-kDa β-adrenergic receptor mRNA-binding protein, termed β-ARB protein (446), demonstrates several RNA binding properties similar to A+U-rich element (ARE) binding proteins reported by others (61, 362, 446,446, 573, 573,635). This 35-kDa cytosolic βARB RNA-binding protein displays the following properties: binds selectively to mRNA of the β1- and β2-adrenergic receptors (243, 446) and to other GPLR mRNA displaying AU-rich domains (559), fails to bind both rat and human β3-mRNA (559), and is induced by agonist treatment, its levels varying inversely with the level of receptor mRNA (446). Although sharing some properties with the mRNA binding protein AUF1, β-ARB protein has been shown to be distinct and likely a member of a family of mRNA binding proteins that includes both AUF1 and β-ARB protein (439). The cognate sequences of β2-adrenergic receptor mRNA important for binding to β-ARB protein has been established in vitro via competition studies with 3′-UTR of highly regulated mRNA and RNA variants with specific mutations in the cognate domains, as well as via radiolabeling of these 3′-UTR followed by ultraviolet-catalyzed cross-linking to cytosolic preparations containing β-ARB protein (243) (Fig. 10).
Recently, analysis of the agonist regulation of human β2-adrenergic receptor mRNA stability has been reported (128). By mutagenesis and deletion analysis, the 3′-UTR was identified as the cognate domain required for destabilization, confirming the earlier studies performed on the hamster β2-adrenergic receptor mRNA and thrombin receptor mRNA (243, 559, 560). In the human β2-adrenergic receptor mRNA, the region of the 3′-UTR at positions 329–337 was identified as the cognate sequence regulating stability, both in response to agonist or cAMP elevation (128).
A hormonally regulated LH/human chorinoic gonadotrophin (hCG) receptor mRNA binding protein has been identified (287), which displays properties much like the β-ARB protein (128,446). This LH/hCG receptor mRNA binding protein binds the LH/hCG receptor mRNA specifically, increases during agonist-induced receptor downregulation, and has a restricted domain of expression (287). Unlike the properties of the β-ARB protein, the LH/hCG receptor mRNA binding protein binds to the open reading frame sequences and displays a molecular mass of ∼50 kDa (287). Much further study of this putativetrans-acting RNA binding protein, termed LRBP-1 (287), and of the cognate sequence(s) of the LH/hCG receptor mRNA will be required before more detailed comparisons with the β-ARB protein can be performed.
2. β-ARB protein and mRNA of other GPLR
The predictive value of the presence of the cognate sequence to identify receptors displaying posttranscriptional regulation has been explored. The GenBank was scanned for sequences that would harbor the cognate domain for β-ARB protein, and a number of GPLR were identified, including the protease-activated GPLR for thrombin (559). With the use of thrombin receptor as a model system, the hypothesis that presence of the cognate domain would dictate agonist and/or cAMP-induced destablization of receptor mRNA was tested. The thrombin receptor, whose mRNA harbors several AU-rich sequence in its 3′-UTR region, was found to bind β-ARB protein specifically and to display both agonist-induced as well as cAMP-induced destabilization of receptor mRNA and its mRNA (559) (Fig. 10). These observations provide compelling evidence that the presence of the cognate sequence for binding of β-ARB protein in an mRNA targets the mRNA for posttranscriptional regulation (destabilization) in response to both agonist and cAMP. Agonist-induced destablilization of M1 muscarinic acetylcholine receptor mRNA has been demonstrated. As found for the β2-adrenergic receptor mRNA, the destabilization sequence is in the 3′-UTR. More extensive study of a broader spectrum of members of the superfamily of GPLR will be required before the degree of universality of this process (and protein) can be established.
3. β-ARB protein and mRNA regulation in vivo
The sequences established for binding of β-ARB protein to hamster β2-adrenergic receptor mRNA in vitro provided strong, but circumstantial, evidence for the involvement of both the cognate sequences and β-ARB protein in the posttranscriptional regulation of receptor mRNA. These hypotheses were tested in vivo to ascertain if the disruption of the cognate binding sites on the target mRNA would influence either basal, steady-state levels of mRNA or the ability of agonists to induce the destabilization of preexisting mRNA (561). Site-directed mutagenesis in tandem with stable expression of receptor mRNA with mutated 3′-UTR in Chinese hamster ovary cells provided the tools with which to answer this provocative question. Although both pentamer and hexamer core ARE of receptor mRNA bind avidly to the β-ARB protein, only the 20-nucleotide, A+U-rich sequence consisting of an AUUUUA hexamer flanked by U-rich regions was shown to be obligate for binding of β-ARB protein and regulation of β2-adrenergic receptor mRNA stability in vivo (561). Receptor mRNA in which the hexamer-containing ARE was disrupted in the AUUUUA or in the 3′-flanking, U-rich region displayed increased half-life but more importantly had lost the capacity to be downregulated in response to either agonist or elevation of intracellular cAMP (Fig.11). Thus the existence of elements within the 20-nucleotide ARE studied in detail for the β2-adrenergic receptor and thrombin receptor appear to predict aspects of the physiological regulation of other GPLR RNA harboring such motifs.
4. Posttranscriptional regulation of G proteins and effectors
Studies on a variety of GPLR suggest that at least some major subset of this superfamily is subject to physiological regulation by posttranscriptional mechanisms. One cannot generalize to the broad spectrum of receptors, since there are examples of agonist-induced downregulation of receptor and receptor mRNA that are not consistent with the posttranscriptional mechanism. The β3-adrenergic receptor, for example, displays agonist-induced downregulation of receptor and receptor mRNA, but does so without a change in the apparent stability of its mRNA (559). In this case, agonist fails to influence mRNA stability, but rather acts at the level of transcription of the β2-adrenergic receptor gene to exert a repression. Thus β1-, β2-, but not β3-adrenergic receptors display a posttranscriptional mechanism to attenuate chronic stimulation of the receptor. Similarities in signaling patterns or even high homology at the protein level have no predictive value when looking for examples of posttranscriptional regulation. Studies of the cross-regulation operating among G protein-linked pathways provide a number of instances in which attenuation of signaling results from a reduction in the expression of key elements in a pathway. Whether or not the mechanisms responsible for the attenuation and reduction in expression include posttranscriptional, destabilization of preexisting mRNA will require more detailed studies at the level of G proteins as well as G protein-sensitive effectors.
It must be highlighted that the studies of β-ARB protein have not revealed the precise nature of the molecule itself, its protein sequence, gene structure, or function. The β-ARB protein may be a nuclease catalyzing the degradation of preexisting mRNA that harbor the appropriate cognate sequence or may be an RNA-binding protein devoid of nuclease activity whose function is to target the association of the mRNA with a nuclease. Although speculation, one can envision that β-ARB protein may represent a member of a larger array of RNA-binding proteins with differing requirements for recognition of mRNA (439, 559, 561). Answers to these provocative and important questions will follow quickly upon the successful molecular cloning of the gene encoding this interesting molecule.
D. Physiological Implications
There exists compelling evidence to support the notion that important features of physiological regulation operate via posttranscriptional mechanisms to alter the levels of mRNA for key elements in signaling pathways (128, 243,287, 446, 559-561). Attenuation of signaling in the face of a chronic challenge of agonist is a both paradigm in biology as well as a major problem in therapeutics. Treatment of pulmonary diseases, such as asthma, include the use of β-adrenergic agonists to open airways. Because this condition, not unlike many others, is chronic, the therapeutic intervention is by necessity chronic also. This dynamic of agonist treatment followed by therapeutic relief set the stage for attenuation, which only amplifies the use of the agonist and exacerbation of the desensitization. In the terms of β-adrenergic receptors in the lung and airways, chronic management of asthma exemplifies the mechanics of agonist-induced downregulation of receptor and receptor mRNA. Many other hormones, neurotransmitters, and drugs that exert their actions via G proteins can be expected to operate in a similar manner.
As mentioned in section iiiB2 , glucocorticoids operate via gene activation, increasing mRNA levels of GPLR, such as the β-adrenergic receptors, that harbor a GRE in the promoter regions of their genes. Studied in the hamster DDT1-MF2 vas deferens smooth muscle cells, the interplay between transcriptional activation of the β-adrenergic receptor genes by glucocorticoids and the ability of agonists to destabilize preexisting receptor mRNA is as remarkable as it is revealing. Neither gene activation nor posttranscriptional destabilization of mRNA dominates to the exclusion of the other; thus both contribute to the control of receptor mRNA (201, 202, 205,208). Rather, the opposing actions of glucocorticoids and agonists display a dynamic relationship in which both processes remain operational and capable of rectifying major swings in intracellular mRNA in response to activation of the opposing mechanism (Fig.12). The clinical experience confirms the fundamental data gleaned from the model cell systems studied in vitro. Chronic use of β-adrenergic agonists provokes desensitization that, if unchecked, limits the therapeutic use of the agonists. Once attenuation and receptor downregulation have occurred, glucocorticoids offer a rational approach to “rescue” the patient from the downward spiral and to restore the β-adrenergic sensitivity of the lungs and airways. Both the physiological and pharmacological description of this interaction between steroids and β-adrenergic agonists provide a useful example of physiological regulation of G protein signaling. To what extent, if any, regulation of G proteins and effectors may contribute to the overall process remains to be established.
V. POSTTRANSLATIONAL REGULATION OF SIGNALING PATHWAYS
A. Protein Phosphorylation
Persistent signaling of G protein-coupled devices provokes an attenuation of the biological response, a universal paradigm in biology often termed “desensitization.” Classic examples of desensitization include attenuation in visual perception, odorant recognition, bacterial chemotaxis, and adaptation of humans to various drugs, neurotransmitters, and hormones used in therapy. Posttranslational modifications to protein elements in G protein signaling are generally responsible for desensitization, especially in the most rapid aspects of the process. Although the repertoire of posttranslational modifications for proteins is significant, protein phosphorylation has emerged as the overarching mechanism supporting desensitization. Understanding the role of protein phosphorylation in regulatory biology has been a major effort of modern biology and requires knowledge not only of potential sites of a specific protein for phosphorylation, but also knowledge of the protein kinases and phosphatases with the responsibility to execute the necessary covalent changes that control function and/or organization of the signaling elements. Progress in the identification of protein kinases and phosphatases that target GPLR, their G protein partners, and effectors has been impressive during the past decade, and many, in-depth, and timely reviews are highlighted for the interested reader seeking further detail into this fascinating mode of regulation.
G protein-linked receptors are well-known substrates for protein phosphorylation by both serine/threonine protein kinases such as PKA and PKC (169, 222) as well as by tyrosine kinases (361, 560), discussed in detail below. The persistent activation of a GPLR, such as the β-adrenergic receptor, provokes in attenuation of a response, e.g., cAMP generation, in what on the surface appears to be a simple negative-feedback loop. The character of the desensitization and the kinases responsible for the diminished functional capacity of the receptor are not uniform and can be dissected into two types, homologous and heterologous. Discriminating between the two forms of desensitization is informative, providing insights into the sites of receptor phosphorylation, the protein kinases that catalyze the phosphorylation, and the temporal aspects of the process.
a) heterologous and homologous desensitization. The ability of multiple agonists operating through discrete receptors both to activate a common effector and to attenuate the ability of other receptor agonists to act typifies the heterologous form of desensitization (102,240). A multitude of receptors activate the Gsprotein and thereby AC (188). Production of cAMP common to all of the stimuli provokes activation of PKA and resultant phosphorylation of many GPLR (222). Thus a single agonist can attenuate the signaling capacity of a larger number of GPLR, even receptors not actually engaged in the primary agonist-induced response. Protein kinase A, a serine/threonine protein kinase, phosphorylates receptors on canonical sites and renders the receptor molecules less capable of activating AC (222). As discussed below, effectors as well as GPLR may be substrates for PKA (257). Temporal considerations are simple, since AC are rapidly activated by a number of hormones. The dose-response relationship between receptor occupancy and activation of PKA is nonlinear, and full activation of the kinase is obtained at subsaturation of the receptor binding sites by agonist. On the basis of these considerations, heterologous desensitization would appear to operate at very low levels of agonist, amplifying the activation of receptor by agonist binding to phosphorylation, potentially, of the full complement of the receptor as well as of those other GPLR with canonical PKA phosphorylation sites (344). Numerous studies have implicated PKC also in the regulation of GPLR (49, 271, 632). Signaling of GPLR has been shown to be attenuated (222) as well as activated by changes in PKC activity (504,505). Using both selective inhibitors of PKA and PKC as well as antisense RNA to suppress both of these serine/threonine kinases, recent studies have revealed the basis for many paradoxical findings, i.e., cell type-specific patterns of desensitization (504). The role and contribution of various protein kinases in a number of widely used cell lines in culture to desensitization are best addressed when clones with null backgrounds for each kinase can be created (632). Serine-262 appeared to be a site of PKA action, serine-261 or serine-262 for PKC action and coincident phosphorylation by PKC and PKA yielded additive desensitization for hamster β2-adrenergic receptors (632). Cell type-specific patterns of kinase action can influence not only the temporal features of desensitization, but also the magnitude of the attenuation to a chronic stimuli exerted through phosphorylation of those G protein signaling elements with motifs for multiple protein kinases.
Activation of a receptor followed by attenuation of the function of the occupied receptor only, typifies homologous desensitization (102, 249). Although sharing with heterologous desensitization a primary role for protein phosphorylation, homologous desensitization of GPLR provides novel capabilities, whereby protein kinases target only the occupied receptors (240) rather than all receptors, occupied or vacant, as does PKA. The discovery of a family of protein kinases that recognize only the agonist-bound form of the receptor revealed the basis for homologous desensitization (249,441). The GRK are serine/threonine kinases with substrate specificity for receptors that are agonist bound and activated (81, 209, 293, 318,419). Six GRK, products of separate genes, have been identified (26, 318, 348,419, 423). Rhodopsin kinase (GRK1), the first to be identified, has a restricted domain of expression, namely, the photoreceptor cells (27, 424). The β-adrenergic receptor kinase (β-ARK) or GRK2, in contrast, is expressed widely, and its successful cloning spawned the identification of an additional four members of the GRK family, each of which displays expression in a number of tissues (249). The GRK can be assayed with peptide substrates, but as perhaps expected, display much higher affinity for the full-length, liganded receptors (28, 168, 319,420). The receptor specificity of GRK1 assayed in vitro is largely confined to rhodopsin and to a very limited extent the β-adrenergic receptors (27). GRK4 prefers β-adrenergic receptors and is most abundant in the testis (169). Reconstitution studies performed in vitro with GRK and either purified or recombinant GPLR provide some insight into the potential of GRK to discriminate among the various substrates (168,169). Putative sites for GRK phosphorylation for β2-adrenergic receptors identified in in vitro studies (168) have been challenged by mutagenesis studies in vivo (495). Sites other than those identified for GRK2 and GRK5 in vitro appear to be responsible for in vivo GRK-dependent desensitization.
Detailed analysis of the receptor specificity of GRK in vivo, however, is a more formidable task and remains to be established. Functionally, the GRK can be dissected into an NH2-terminal region believed to target GRK to specific classes of activated receptors, a COOH-terminal region necessary for targeting of the GRK themselves to intracellular targets, and a centrally located, conserved catalytic domain of ∼270 amino acid residues (563). GRK1 is fanesylated (251), GRK4 and -6 are palmitoylated (448), whereas GRK2 and -3 are not lipid modified but display two protein motifs implicated in targeting proteins to the membrane, a Gβγ binding domain and a PH, PIP2-binding domain (125).
A second major distinction of the GRK is the functional dependence of the biology of their phosphorylation on an additional class of interacting proteins, the arrestins (162). Unlike the PKA and PKC, which phosphorylate and attenuate the function of their receptor substrates, phosphorylation of receptors by GRK in the absence of arrestins fails to decrease substantially the functional integrity of the receptor or photoreceptor. These accessory, inhibitory proteins that bind to the GRK-phosphorylated receptors and enable the functional expression of the desensitization were first identified and characterized in the visual system (424). The discovery of the role of arrestin in mediating desensitization in vision was followed by a search for an analogous molecule to mediate GRK-dependent phosphorylation of GPLR, such as the β-adrenergic receptors and muscarinic acetylcholine receptors. Although initially used to identify a role in extravisual sites, arrestin was later shown to display little affinity for GPLR, other than the photopigment rhodopsin. β-Arrestins were discovered soon thereafter and found in wide distribution (42, 314, 331,528). The members of the arrestin family (arrestin, β-arrestin, and arrestin 3) all bind to properly phosphorylated GPLR stoichiometrically for expression of their inhibitory role in signaling. Overexpression of arrestins typically results in more profound desensitization, suggesting that the amount of arrestin may be limiting with respect to availability of properly phosphorylated receptors. Attenuation of GPLR signaling after overexpression of β-arrestin or arrestin 3, however, must be viewed with considerable caution. More recently, a role of arrestins in mediating intracellular localization of GPLR involving clathrin has been discovered (163, 193), which could quite apart from its role in desensitization exerted at the G protein level alter the cellular localization and/or endocytosis of GPLR (565). This novel aspect of arrestin biology is discussed below in terms of a broader understanding of GPLR trafficking, sequestration, and degradation.
As is true for most membrane-localized signaling elements with extracellular domains for sensing stimuli, GPLR require extensive posttranslational processing to both provide, cycle, and remove (degrade) receptors from the plasma membrane. It is therefore not remarkable that a wealth of reports have accumulated showing members of the GPLR subject to internalization, sequestration, and degradation. The more pressing question, properly set against the background presented above, is whether or not GPLR display these basic aspects of receptor biology, but rather is internalization, sequestration, and/or degradation of receptors regulated by agonists in such a way as to contribute meaningfully to desensitization (590,591)? Proponents for the role of internalization/sequestration in agonist-induced desensitization would cite numerous studies that document the ability of agonist challenge to alter each or all of these parameters. Opponents, in contrast, would argue that all membrane proteins are subject to these processes, and it is not remarkable that activation of signaling may influence the overall process. The most compelling arguments against the tenet is that chemical (245) and thermal means of arresting internalization that are well known in cell biology fail to block agonist-induced desensitization (583). Mutation of specific phosphorylation sites and truncations yield equivocal support for either position, since the mutations themselves may alter protein folding or protein-protein interactions that appear to result in an alteration in desensitization. Even an ostensibly fool-proof approach of staining fixed, nonpermeabilized cells with antibodies to extracellular sequences or GPLR fails to provide results of unequivocal interpretation (591). Some studies display little loss of surface immunoreactivity during desensitization (584), whereas others show significant loss of cell-surface receptors and relocalization of immunoreactivity to intracellular compartments not enriched with plasma membrane markers (593). One set of observations seem universal, i.e., the magnitude of GPLR found to internalize or to sequester is substantially less than the amount predicted by the magnitude of the desensitization achieved. It is rare for cells to sequester more than 50% of GPLR, even under conditions in which attenuation is much more profound.
Some rectification of the apparent paradoxes in the field emerges from a more complete understanding of the complexities that underlie receptor interactions with numerous other proteins and classes of proteins. The GRK have been shown to bind and phosphorylate tubulin (82) as well as GPLR adding another dimension to possible roles in cytoskeletal action. Arrestins have been shown to mediate the GPLR interaction with clathrin-coated vesicles (193), primary elements in the trafficking of proteins within cells for many purposes (421, 543). Overexpression of the nonvisual arrestins attenuates signaling through β1-adrenergic receptors, much like overexpression of GRK2 (170). Although overexpression of β-arrestin and arrestin 3 can drive GPLR into clathrin-coated pits, at normal levels of expression arrestin may (193) or may not act as a primary adaptor between a properly phosphorylated GPLR and clathrin (193). Analysis of mutant forms of M2acetylcholine receptors provide additional evidence that desensitization and sequestration of receptor are phosphorylation-dependent but distinct pathways (240). Other recent studies of the μ-opioid receptor clearly demonstrate that phosphorylation is not an obligatory event in desensitization of AC regulation by this GPLR (72). The visual arrestins do not appear to display this same adaptor function with respect to clathrin. It is worth noting that the fate of the GPLR as well as their ligands is important to consider, since signaling may well include a phase beyond the cell-surface interlude that occurs intracellularly, reflecting downstream roles in signaling for either the receptor, or the ligand, or both (331).
Quantitative descriptions of receptor endocytosis and recycling phenomenon provide invaluable tools for further analysis and modeling. Details of the arrestin- and clathrin-independent mechanisms of receptor sequestration remain obscure and crucial to our understanding of receptor function and trafficking in response to chronic activation by agonists. Much of our knowledge of GPLR trafficking has been gleaned from studies of a rather small number of GPLR, prominent among them β-adrenergic receptors, and new paradigms may arise as our information on broader groups of GPLR is acquired.
b) counterregulation of gplr via tyrosine kinases. Despite an extensive literature on the interaction between signaling of GPLR and growth factor receptors with intrinsic tyrosine kinase activity, only recently have details emerged about this important regulatory physiology (204, 285). The ability of insulin to counterregulate catecholamine action has been known for decades, and facets of insulin action operating at many postreceptor levels have been documented. Analysis of GPLR signaling from the agonist activation of β-adrenergic receptors to the distal stimulation of cAMP generation revealed a new target for study, the GPLR itself. Treating cells with insulin abrogates catecholamine signaling via β-adrenergic receptors but does not alter the ability of Gs to activate AC or the functional status of Gs itself (204). The discovery that β-adrenergic receptors display increased phosphorylation in cells treated briefly with insulin was quite unexpected and raised an important question of whether or not GPLR are phosphorylated on tyrosine residues, since agonist-induced desensitization proceeds by the action of serine/threonine-specific enzymes PKA, PKC, and GRK.
Stimulating cells with insulin [or insulin-like growth factor I (IGF-I)] results in a functional uncoupling of the β-adrenergic receptor from Gs and in a sharp increase in phosphotyrosine content (204). Because activation of a growth factor receptor results in the tyrosine phosphorylation of a GPLR, it was imperative to discern if the signaling was propagated through a cascade of kinases, as is often the case in growth factor signaling, or a more direct tete-á-tete interaction. Reconstitution of purified insulin receptors and recombinant β-adrenergic receptors demonstrated conclusively in vitro that a GPLR could function as a substrate for a tyrosine kinase receptor (20). Phosphorylation of the β-adrenergic receptor by the insulin receptor was insulin dependent (half-maximal at ∼20 μU/ml), rapid, and yields a stoichiometry of 1–2 mol phosphate/mol GPLR. Counterregulation at the most proximal point imaginable, receptor to receptor, was verified in vivo by mapping studies in which the sites phosphorylated by insulin were revealed to be Y350, Y354, and Y364 (283), in the COOH-terminal intracellular tail of the β-adrenergic receptor (531). These same residues were found to be phosphorylated in the defined, reconstitution studies performed in vitro, establishing that the interaction was receptor to receptor (283), placing the β-adrenergic receptor in the same category as other insulin receptor substrates (IRSI and IRSII). Furthermore, mutation of these tyrosyl residues to phenylalanine altered substantially the ability of insulin to counterregulate catecholamine action (284).
Because IGF-I also counterregulates catecholamine action, it was not unexpected that it also stimulated the phosphorylation of β-adrenergic receptors, much like insulin. Unexpected was the determination that IGF-I preferentially phosphorylates Y132 and Y141 residues in the first intracellular loop between transmembrane spanning domains I and II of this heptihelical receptor (284). Notably, these data were obtained by use of matrix-assisted, laser-desorption ionization (MALDI) mass spectrometry of unlabeled receptor phosphopeptides; the first application of MALDI mass spectrometry to this general area (284).
Further examination of the motifs surrounding the sites of the best peptide substrates for in vitro phosphorylation, peptide L339, harbors residue Y350 that lies in a sequence motif (Tyr-Gly Asn-Gly) with similarity to motifs known to interact with Caenorhabditis elegans sem5 src-homology 2 (SH2) domains when phosphorylated (20, 361). An Asn residue at position +2 from the tyrosine residue with residues +1 and +3 containing aliphatic side chains is the general character of these sites. Analysis of the phosphorylation sites for IGF-I likewise revealed a potential binding site for Shc (285, 361). Provocative are these observations because they foster the possibility that GPLR, once properly phosphorylated by a tyrosine kinase, can make use of adaptor molecules with the capability of directing interactions with a myriad of protein targets. Other examples of GPLR acting as substrates of tyrosine kinases have been reported for the bradykinin, angiotensin II, and cholecystokinin receptors of the GPLR superfamily (361). The analysis of tyrosine phosphorylation in low-abundance GPLR remains a formidable task, and the broader appreciation for tyrosine phosphorylation and regulation of GPLR will likely suffer until this obstacle has been circumvented.
Are potential Shc or Grb2 binding sites of GPLR functional? For the β-adrenergic receptor, insulin-dependent Grb2 association occurs (286). Grb2 binds the β2-adrenergic receptor stoichiometrically, based on the availability of the tyrosine-350 residue and its phosphorylation by the insulin receptor. Insulin has been shown not only to uncouple the β-adrenergic receptor from Gs but to sequester the β-adrenergic receptor in much the same manner as does agonist-induced desensitization. Tantalizing results suggest that both PI 3-kinase and the GTPase dynamin associate with the β-adrenergic receptor via Grb2. Grb2 itself associates only with the membranes of insulin-treated cells and in so doing leads to a Grb2-dependent shift in the agonist affinity of the β-adrenergic receptor, much like GTP does (506). The Grb2 responses were dependent on the integrity of the SH2 binding domains of Grb2 and were sensitive to competition by peptides known to interrupt SH2 interactions. Thus Grb2 binds the β-adrenergic receptor in an insulin-stimulated fashion, uncouples from Gs, and appears to bind two downstream elements involved in cellular trafficking of membrane-bound molecules. Much further work will be required to detail the signals propagated through protein-protein interactions involving Grb2, Shc, and other adaptor molecules with which tyrosine-phosphorylated GPLR may function.
In the last decade, a vast literature on the nature of the protein kinases that employ GPLR as substrates has emerged. Knowledge about the role of protein phosphatases in the resensitization of GPLR, in contrast, has lagged, as is a common observation in the general field of regulation of proteins via phosphorylation. The role of dephosphorylation in receptor resensitization seems obvious. In general, GPLR recover from agonist-induced desensitization through some process other than expression of receptors de novo. Several protein phosphatases have been implicated in resensitization (169), which then leads to the questions of where in the cell does the dephosphorylation occur, by what phosphatase(s), and how is the resensitized receptor introduced again into the signaling process (331)? Models suggesting that receptors are endocytosed and then dephosphorylated by phosphatases enriched in an endosomal compartment that is often acidified ignore the organization of the heptihelical receptors that would appear to maintain the intracellular sites of phosphorylation of the receptor to the exterior and not the interior of the endosomes. In addition, there is no reason to assume that receptors cannot be resensitized in situ at the membrane, which would offer some real advantages and economies to cells that must repeatedly respond to agonist stimulation. In support of this possibility is a growing literature on the microdomains of caveoli and the role of both anchoring proteins and scaffolding proteins in organizing G protein-linked signaling systems (158,386, 407, 437). Although not a likely model for all GPLR, the visual system in Drosophilahas been dissected genetically with great precision. The existence of a family of protein phosphatases involved in the regulation of GPLR signaling cascades has been postulated based on analysis of retinal degeneration C (RGDC) mutants of Drosophila(581). Central roles of protein-protein binding motifs in the organization and function of rhodopsin signaling have emerged (566, 581). For other GPLR, only recently has a role for A-kinase associated proteins (AKAP) as scaffolds for signaling been discovered. The 250-kDa AKAP termed gravin organizes the β2-adrenergic receptor with PKA, PKC, and protein phosphatase 2B. The loss of gravin disrupts the signaling process, particularly the recovery of sensitivity (503).
c) physiological implications of gplr phosphorylation. Marked progress has been made in the study of phosphorylation of GPLR in the past decade: sites of phosphorylation have been identified; two new classes of protein kinases, the GRK and tyrosine kinases, regulating GPLR have been detailed; roles for arrestins and adaptor molecules such as Grb2 in receptor trafficking have been discovered; and the complexity of multisite phosphorylation of GPLR is only now being appreciated. The profound role of phosphorylation, protein kinases, and phosphatases in normal GPLR signaling highlights the possibility of their roles in signaling of GPLR in pathophysiological states. β-Adrenergic receptor signaling functions prominently in myocardial regulation, and recent studies in transgenic mice suggest that important features of cardiac pathophysiology result from alterations in the phosphorylation of these receptors (310, 467). Enhanced cardiac function is obtained either by overexpression of β-adrenergic receptors (38) or by expression of an inhibitor of GRK (β-ARK1) (305). More profound are the consequences of β-ARK-1 deficiency in mice, lethal during late gestation (466). In fact, the virtual absence of endogenous GRK2/NARK1 activity in the GRK2 −/− embryos and its early lethality highlights the role of GRK ion cardiac development (259). Novel observations abound from the development of mice in which specific elements of the signaling pathway are either deleted genetically or intentionally overexpressed. The extent to which bona fide models of congestive heart failure and other cardiac pathophysiologies can be mimicked in transgenic mice through application of this powerful technology to the study of GPLR signaling is not obvious and awaits more critical review of existing models as well as generation of new lesions of interest.
2. G proteins as phosphoproteins
Evidence for both serine/threonine and tyrosine phosphorylation of several Gα subunits has been presented. Several of these studies link phosphorylation to changes in Gα subunit function that may underlie adaptive responses to stimulation or provide mechanisms for cross talk among different G protein-mediated effector pathways or between G protein-mediated pathways and signaling systems initiated by receptor tyrosine kinases.
a) phosphorylation of gαi. Protein kinase C-dependent phosphorylation of Gαi-2has been suggested to play a role in regulating the inhibitory actions of this G protein on AC activity (74). Protein kinase C activation results in an attenuated capacity of δ-opiate agonists and epinephrine acting through α2-adrenergic receptors to inhibit AC activity. Protein phosphatase inhibitors and phorbol esters both elicited an inhibitory AC response, and studies using broken-cell preparations suggest that an attenuation of G protein-dependent inhibition of AC was responsible (532). Gαi-2 was shown to be phosphorylated in unstimulated cells, and activation of PKC produced an increase in phosphorylation of this G protein α-subunit. Studies using PKC activators and inhibitors demonstrate that Gαi-2phosphorylation correlated with attenuation of AC inhibition (532).
Similar findings have been reported in studies using rat hepatocytes (74, 75). Phorbol esters, vasopressin, and angiotensin II elicited increases in the phosphorylation of Gαi-2 in primary hepatocytes from lean Zucker rats. Phosphorylation of Gαi-2 appears to involve a loss of Gi function. The ability of nonhydrolyzable guanine nucleotides to inhibit forskolin-stimulated AC activity was reduced in membranes prepared from treated hepatocytes. Hepatocytes from obese Zucker rats appeared to be resistant to both agonist-induced phosphorylation of Gαi-2 and to guanine nucleotide-dependent inhibition of AC. The basal level of Gαi-2 phosphorylation in hepatocytes from obese Zucker rats was higher than in hepatocytes from lean animals (73). Activation of PKC doubled the level of phosphorylation of Gαi-2 in the hepatocytes from lean animals but had little effect on the phosphorylation state of Gαi-2 in hepatocytes from obese animals. Incubation of hepatocytes from lean animals with ligands that lead to the phosphorylation of Gαi-2 abolished the ability of low concentrations of the hydrolysis-resistant GTP analog Gpp[NH]p to inhibit AC expressed in isolated membranes. Treatment of hepatocyte plasma membranes from lean, but not obese, Zucker rats with pure PKC led to Gαi-2 phosphorylation and inhibition of AC, and this increased phosphorylation of Gαi-2 could be reversed by alkaline phosphatase treatment.
Further studies suggested that a variety of agonists including glucagon, TH-glucagon ([1-N-α-trinitrophenylhistidine, 12-homoarginine]glucagon), arginine vasopressin, angiotensin II, the phorbol ester 12-O-tetradecanoylphorbol 13-acetate, and 8-bromo-cAMP all promote phosphorylation of Gαi-2, whereas no increased phosphorylation of either Gαi-3 or Gαs is observed (74). Phosphoamino acid analysis indicates serine as the major site of phosphorylation of Gαi-2 in response to PKC activators. In vitro phosphorylation by PKC has been demonstrated, but the sites of phosphorylation have not been defined.
b) phosphorylation of gα s by pp60c-src. A number of lines of evidence suggest that cross talk exists between the cellular signal transduction pathways involving tyrosine phosphorylation catalyzed by members of the pp60c-src kinase family and G proteins. pp60c-src has been reported to catalyze tyrosine phosphorylation of Gαs in vitro with the GDP-liganded form of the G protein monomer appear to be the preferred substrate. Phosphorylation produced a modest increase in affinity of Gαs for nonhydrolyzable guanine nucleotides (562).
c) phosphorylation of gαz. Phosphorylation of Gαz has been observed in human platelets (80). Phosphorylation is elicited by agents that activate PKC including phorbol 12-myristate 13-acetate, thrombin, and the thromboxane A2 analog U-46619. Studies using purified components in vitro demonstrate that PKC catalyzes the rapid and nearly stoichiometric phosphorylation of recombinant Gαz, with the modification occurring preferentially for the GDP-bound form of the subunit. Phosphoamino acid analysis and partial digestion suggested that phosphorylation of both recombinant Gαz and platelet Gαz occurs at a serine residue near the NH2terminus. Studies using an antiserum that only recognizes the unphosphorylated form of the α-subunit suggests that PKC-mediated phosphorylation of Gαz achieves high stoichiometry in intact platelets (349).
d) phosphorylation of gαq/11. In several systems, tyrosine kinase inhibitors block activation of inositol lipid hydrolysis and Ca2+ signaling by G protein-coupled receptors. Receptor activation was reported to induce phosphorylation of Gαq/11 on a tyrosine residue (Tyr-356) of the Gαq/11 subunit (571). This tyrosine phosphorylation event was shown to be essential for Gαq/11 activation. Reconstitution studies indicated that tyrosine phosphorylation of Gαq/11 promoted interaction with receptors. These provocative studies suggest a permissive role for tyrosine phosphorylation in coupling receptors to Gαq/11activation (379). The protein kinase responsible for phosphorylation Gαq/11 and the mechanism of this effect or receptor coupling remains to be established.
e) phosphorylation of gαt. The GDP-bound α-subunit of transducin (but not the GTP-bound form) has been reported to be phosphorylated on tyrosine residues by the insulin receptor kinase and on serine residues by PKC. The Gαtβγ heterotrimer was found, in contrast, to be poorly phosphorylated by the insulin receptor kinase and is not phosphorylated by PKC. Holotransducin or any of its subunits were not substrates for phosphorylation by cAMP-dependent protein kinase. The physiological significance of these modifications has not been detailed (636).
3. Effectors as phosphoproteins
Studies with intact cells using pharmacological activators and inhibitors of phosphorylation clearly demonstrate an important role for protein phosphorylation in modulation of G protein effector activities. Phosphorylation of G proteins as well as GPLR likely accounts for much of these phenomena. Phosphorylation of G protein-coupled effectors has only been demonstrated in a surprisingly small number of cases.
Phospholipase C-β enzymes have been reported to be substrates for phosphorylation by both PKC and PKA. The macrophage PAF receptor stimulates PLC activity. Addition of dibutyryl cAMP to these cells resulted in an attenuation of receptor-regulated PLC activity and phosphorylation of PLC-β3 (267,342). Phorbol 12-myristate 13-acetate (PMA) also stimulated phosphorylation of PLC-β3 in these cells. Studies using PAF receptor mutants suggest that phosphorylation of PLC-β3 by PKC is involved in inhibition of PLC-β activation. Phosphorylation by PKA also has been linked to regulation of PLC-β2. In this case, phosphorylation of the enzyme by PKA was demonstrated both in vitro and in vivo. Phosphorylation of PLC-β2 apparently attenuated activation by βγ-subunits (342). Studies using intact cells implicate a number of phosphorylation-dependent mechanisms in the control of AC activity. Of the mechanisms suspected, control by cAMP-dependent PKA and by PKC has been most widely studied. Although both PKA and PKC can phosphorylate certain AC isoforms in vitro, the effects of phosphorylation on AC activity observed are modest in comparison with observations made in intact cells. It seems likely that these protein kinases control AC activity by an indirect mechanism(s) (546). Adenylyl cyclase II activity is increased by phorbol ester treatment of cells. Mutagenesis studies identify threonine-1057 as a site of PKC-mediated phosphorylation of AC. Although phosphorylation of the threonine-1057 had little effect on basal AC II activity, it provoked a marked reduction in the ability of the enzyme to be stimulated by βγ-subunits (550,552). Recombinantly expressed type II AC activity is increased by PMA treatment of Sf9 cells. Treatment with PMA was found to result in an increased maximum velocity without affecting the Michaelis constant of AC for ATP. In this system, PMA treatment does not alter sensitivity to Gαs but enhances stimulation at all concentrations of activated Gαs tested. Studies using epitope-tagged type II AC confirm the PMA-dependent phosphorylation. Phosphorylation of type II AC by PKC therefore appears to play a role in sensitization of the enzyme to G protein activation.
Evidence for inhibition of AC by protein phosphorylation also has been presented. Adenylyl cyclase type II isoenzyme appears to be the major target. Inhibition of this AC isoform is antagonized by treatment with inhibitors of Ca2+/calmodulin-dependent protein kinase and by expression of constitutively activated calmodulin kinase II. Increases in intracellular Ca2+ levels produce phosphorylation of AC II. Site-directed mutagenesis of a calmodulin kinase II consensus site (Ser-1076 to Ala-1076) in AC III reduces greatly Ca2+-stimulated phosphorylation and inhibition of AC III in vivo. These data suggest that Ca2+ inhibition of AC III in vivo is due to direct phosphorylation of the enzyme by calmodulin kinase II (101).
B. Lipid Modifications of Proteins
1. Palmitoylation of GPLR
Covalent lipid attachment to protein is a ubiquitous posttranslational modification (148, 180,392, 601). Lipid modifications to proteins include prenylation, myristolyation, and palmitoylation. Prenylation consists of farnesylation or genanylgeranylation of cysteinyl residues of target proteins, whereas myristoylation occurs on NH2-terminal glycinyl residues of a target protein. Palmitoylation occurs by thioesterification of cysteinyl residues of target proteins and is further differentiated from either prenylation or myristoylation in being a reversible posttranslational modification, much like protein phosphorylation.
Addition of a 16-carbon fatty acid to a protein provides a hydrophobic domain or anchor potentially of interaction with the lipid bilayer and perhaps hydrophobic domains of other proteins. For the heptihelical GPLR that are intergral membrane proteins with seven, hydrophobic transmembrane-spanning domains, palmitoylation seems unlikely to contribute meaningfully to localization of GPLR as it does for G protein subunits that are membrane associated, but not integral membrane proteins. Various members of the GPLR superfamily lack the potential site for palmitoylation, usually confined to the COOH-terminal regions of the receptor arising from the seventh transmembrane-spanning segment (392). Mutation of the expected sites of palmitoylation confirms through loss of metabolic labeling the palmitoylation of GPLR but produces little effect on the ability of the receptor to properly fold and transit to/from the plasma membrane (389). Alterations in the phosphorylation and desensitization of the β-adrenergic receptor mutant lacking the site for palmitoylation have been reported (50,388, 389). Rigorous biochemical demonstration of stoichiometric palmitoylation of GPLR has only been provided for rhodopsin, more than a decade ago. Metabolic labeling studies do detect labeled palmitate associated with several GPLR (392). The more rigorous determinations of the true stoichiometry of fatty acid per mole of receptor is difficult to assess in very low-abundance membrane proteins like GPLR and remains to be established for most members of this family.
A most intriguing role for palmitoylation has evolved from careful study of the β-adrenergic receptor (51). The β-adrenergic receptor displays radiolabeling when isolated from cells metabolically labeled with radiolabeled palmitate. This labeling of the receptors is lost in cells expressing mutant receptor, lacking a cysteinlyl residues found at ∼340 residues from the NH2terminus (50). Palmitoylation of this COOH-terminal site of most GPLR would create a short “fourth” intracellular loop that is attached. However, unlike the seven hydrophobic segments in all GPLR, this fourth segment would not traverse the lipid bilayer (392). The reversibility of palmitoylation takes on a greater potential significance for GPLR biology in general and β-adrenergic receptor biology in particular by virtue of its dynamic regulation by agonists. In addition, the absence of the palmitoylation site, or perhaps the absence of a cysteinyl residue at that location, leads to enhanced basal phosphorylation of adjacent canonical sites by PKA (388, 389). The nonpalmitoylated β-adrenergic receptor appears hyperphosphorylated and desensitized, although the functional significance of this observation in vivo remains to be established. Although speculation, palmitoylation may provide a hydrophobic domain to drive protein-protein interactions with other elements of GPLR signaling, rather than protein-lipid bilayer interactions. Much further study will be required before the physiological significance of dynamic palmitoylation of GPLR can be adequately addressed.
2. Modification of heterotrimeric Gα and Gβγ subunits by fatty acids
As discussed above, both Gα and Gγ subunits are modified by covalent attachment of fatty acids. Myristoylation of G protein α-subunits and prenylation of γ-subunits are stable modifications. Mutational studies indicate that these modifications have profound effects on the localization and function of these G protein subunits. Palmitoylation of α-subunits is dynamic in intact cells and thus has the potential to be a regulated posttranslational modification (84). As discussed below, receptor activation of Gs increases palmitate turnover on the α-subunit, presumably by stimulating deacylation (84). The enzymes responsible for these modifications have not been isolated, but the potential importance of these enzymes as regulators of reversible acylation and dynamic control of Gα subunit activity is receiving increased attention.
a) nh 2 -terminal myristoylation/palmitoylation of gαsubunits. Gα subunits are dually modified at their NH2 terminus by addition of prenyl and myristoyl groups (339, 340, 399). As discussed below, these modifications are required for membrane association and for the consequent signaling functions of these proteins. Because palmitoylation appears to be a regulated event, several studies have focused on examining the functional consequenses of this posttranslational modification.
The most convincing demonstration of a functional role for palmitoylation of a Gα subunit is derived from studies of Gαz. Palmitoylation of Gαz provokes a dramatic inhibition of its response to the GTPase-accelerating activity of Gz GAP (568). Gz GAP is a selective member of the regulators of G protein signaling (RGS) protein family GAP. This palmitoylation is accompanied by a decrease in the affinity of Gz GAP for the GTP-bound form of Gαz and a decrease in the maximal rate of GTP hydrolysis. A similar decrease is observed in response to the GAP activity of Gα-interacting protein (GAIP). The palmitoylation of Gαi-1 provides a further example as palmitoylation inhibited its response to RGS4. Reversible palmitoylation therefore appears to be a mechanism by which G protein signaling responses may be prolonged through the suppression of RGS actions (568).
Gα11 is dually palmitoylated, and the role of this modification in G protein function has been examined by mutagenesis. The wild-type Gα11 is membrane associated, whereas a nonpalmitoylated mutant is present only in the soluble fraction. Monopalmitoylated forms of Gα11 are found distributed between soluble and membrane fractions. The degree of stimulation produced by activation of a coexpressed TRH receptor with either the wild-type or the mutant forms of Gα11 correlates well with the extent to which these G protein α-subunits associate with membranes, i.e., greater activation is noted with those G protein α-subunits that have been palmitoylated.
Functional interactions between myristoylation and palmitoylation have been examined using Gαi-1, which is dually acylated at its NH2 terminus with myristate and palmitate. Inhibition of protein myristoylation by treatment with 2-hydroxymyristate prevents neither the incorporation of palmitate nor the membrane association of this protein expressed in COS cells. A palmitoylation defective Gαi-1 mutant was found primarily in the cytoplasmic fraction. These data suggest that myristoylation is not an absolute requirement for palmitoylation of Gαi-1 and that palmitoylation plays a key role in membrane association (246). The role of palmitoylation in myristoylation, βγ-interactions, and membrane attachment also involves the NH2 terminus of the Gαi-1 subunit. Coexpression studies indicated that turnover of palmitate is slower on Gαi-1 subunits coexpressed with β- and γ-subunits than it is for the Gαi-1 subunit expressed alone (601). A nonpalmitoylated mutant that was predominantly soluble was capable of forming a heterotrimer with coexpressed βγ-subunits, restoring its membrane localization (602). Similar findings have been reported for Gαo (612). Perhaps changes in βγ-subunit association could modulate palmitoylation during signaling.
The enzymes catalyzing palmitoylation and of Gα subunit have not been identified. A palmitoyl-CoA:protein S-palmitoyltransferase activity that acylates G protein α-subunits in vitro has been identified in brain. Palmitoyltransferase activity is membrane associated and requires detergent for solubilization. G protein heterotrimers are the preferred substrate for this enzyme, and both myristoylated and nonmyristoylated Gα subunits are recognized as substrates. In rat liver, palmitoyltransferase activity is high in the plasma membrane-enriched fractions and present at low or undetectable levels in Golgi, endoplasmic reticulum, and mitochondria (147). This subcellular localization suggests that the palmitoyltransferase plays a role in regulated cycles of G protein acylation and deacylation. Palmitoyl thioesterases that catalyze the deacylation of low-molecular-weight GTP-binding proteins have been described, but their roles in control of heterotrimeric G protein function remain to be established (77, 517).
b) isoprenylation of heterotrimeric g proteinγ-subunits. Heterotrimeric Gγ subunits are processed by prenylation of a cysteine residue at a site initially four amino acids from the COOH terminus, endoproteolytic truncation of the three terminal amino acids, and methylation of the now COOH-terminal prenylcysteine residue (84). Prenylation of Gα is not required for association with Gβ, although only Gβγ dimers containing prenylated Gγ are competent to interact with Gα subunits (253). It is noteworthy that Gγ1 is farnesylated, whereas all other known Gγ subunits are geranylgeranylated. Isoprenylation of Gγ is required for G protein-effector interactions. Reconstitution studies reveal, for example, that only isoprenylated Gγ subunits can stimulate PLC-β2 activity (53). Gβγ dimers containing the Gγ1 subunit are generally less effective in assays of effector stimulation or in receptor reconstitution studies. Experiments in which the farnesyl and geranylgeranyl groups are switched between Gγ subunits suggest an additional role for this covalent modification, i.e., in modulating interactions of G protein heterotrimers with adenosine receptors and perhaps other GPLR (627).
Adenosine 5′-diphosphate-ribosyltransferases catalyze the transfer of the ADP-ribose moiety of NAD to specific amino acid residues of their target proteins (395). The earliest example of ADP-ribosylation, the ability of diphtheria toxin andPseudomonas aeruginosa exotoxin A to mono-ADP-ribosylate a modified histidinyl residue of eukaryotic elongation factor 2, results in a blockade of protein synthesis and subsequent cell death. For G protein-linked signaling, prominent are the ADP-ribosyltransferase activities of cholera toxin and pertussis toxin, whose mechanisms of action target the α-subunits of heterotrimeric G proteins (570). The details of cholera toxin-catalyzed ADP-ribosylation of a specific argininyl residue and activation (impaired GTPase) of Gαs, AC, and cAMP generation are well described (189). Likewise, for pertussis toxin which ADP-ribosylates the COOH-terminal cysteinyl residue of Gαi, Gαo, and Gαt blocking the ability of GPLR from transducing activation to the G protein, much information on this novel form of posttranslational modification is known (570). Over the past decade mono-ADP-ribosyltransferases have been identified and their genes cloned from animal tissues (395). Some appear to be glycosylphosphatidylinositol-linked exoenzymes (418).
The activity of ADP-ribosyltransferases is regulated by members of a family of factors, collectively termed ADP-ribosylation factors (ARF) (278). The ARF stimulate the ability of cholera toxin to ADP-ribosylate Gαs but have been revealed as critical elements to vesicular trafficking in cells and to activation of the membrane-associated phospholipase D isozymes. For the interested readers, more-detailed reviews of these emerging areas can be obtained in a number of excellent reviews (44,395).
VI. G PROTEIN-LINKED SIGNALING AND HUMAN DISEASES
Investigating the basis of pathophysiologies in humans can be approached from several vantage points. Experimental mutagenesis performed in the laboratory often reveals critical elements controlling gene expression or the nature of the gene product’s structure that by their very nature might yield alterations in the complex biochemical and physiological pathways governing normal health and development. Likewise, advances in human genetics and the application of molecular genetics to inherited diseases have illuminated invaluable insights into these same elements. To facilitate the analysis of human diseases, the Online Mendelian Inheritance of Man (OMIM) was created as an outgrowth of the detailed cataloging by Victor McKusick (Johns Hopkins University). The outline of human diseases derivative of mutations in G protein signaling reflects the access to this wonderful, researcher-oriented database (http://www3.ncbi.nlm.nih.gov/Omim) maintained by the National Center for Biotechnology Information. Readers are encouraged to explore their interests in the leads presented below, making use of this timely resource.
A. GPLR-Based Diseases
1. Gain-of-function mutations in GPLR
Although one can speculate that mutations affecting the function of critical signaling elements will be identified which provoke a pathophysiology, the search to identify the basis of human diseases and the search to understand the functional domains of important signaling molecules by mutagenesis converge relatively few times, but produce important advances in our knowledge of humans (520). Naturally occurring mutations in receptors for LH (500) and TSH (434) that provoke profound endocrinopathies were identified just before results of detailed mutagenesis of the adrenergic family of GPLR (332) which revealed the ability of specific mutations to constitutively activate receptors in the absence of ligand. By virtue of this constitutive activation, these gain-of-function mutant receptors cause chronic downstream activation of signaling pathways in the absence of the upstream, regulated events (333). Because only a small portion of the complement of any receptor requires activation to yield a full range of physiological responses, germ-line heterozygous gain-of-function mutations would be predicted to be dominant.
Based on current knowledge, one might anticipate that some gain-of-function mutations in GPLR might not result in clinical manifestations. Mechanisms like agonist-induced desensitization and downregulation of receptors, in theory, could compensate for or rectify the activity provoked by some mutations that constitutively activate a GPLR. Accordingly, a GPLR that operates via Gαs may constitutively activate AC and PKA, providing a negative feedback operating at two distinct levels, either at PKA that can phosphorylate and desensitize the receptor or at the level of cAMP-driven, downregulation of receptor expression (203,205, 206, 394,445). Certain protein kinases, like those of the GRK family (210, 211, 330,347), that require the ligand-bound form of the receptor may or may not be able to recognize a mutated form a GPLR that is constitutively activated. Details of how the known gain-of-function mutations apparently bypass the normal regulatory controls that operate physiologically remain obscure.
To the extent that the downstream signaling of a receptor has been elucidated, pathophysiologies derivative of constitutive activation of GPLR are predictable. The activating mutations of the LH receptor provoke familial male precocious puberty (FMPP), in an autosomal dominant fashion (499, 500). Female carriers of the LH receptor mutation, however, do not display the clinical manifestations of precocious puberty, reflecting perhaps a greater influence of LH action alone on the acquisition of secondary sexual characteristics in the male. The most common site of mutations in patients with FMPP are in the sixth transmembrane domain of the LH receptor, with the Asp-578 residue subject to mutation to glycine, tyrosine, or other residues (326, 500). The third and sixth transmembrane domains of GPLR have been shown to be critical to agonist binding and activation of GPLR members, such as the adrenergic receptors (332, 333,460, 483). The Asp578Gly mutations, in particular, provoke robust cAMP accumulation in transfected cells in vitro and is associated with the earliest onset of FMPP, within the first year of age (325).
The TSH receptor is the site of gain-of-function germline mutations that have been shown capable of provoking either hereditary toxic thyroid hyperplasia or toxic thyroid adenoma (149,434). Additional GPLR in which mutations lead to gain-of-function include the photopigment rhodopsin, in which congenital night blindness can result (451,509, 520), and the receptor for PTH and PTH-related peptide that is responsible for Jansen-type metaphyseal chondrodysplasia (489). Activating mutations of rhodopsin capable of causing congenital night blindness and retinitis pigmentosum have been detailed (452). The opsin photopigments, such as rhodopsin, present a special case in which mutations that alter the proper folding of the heptihelical molecule and retention of the pigment in the endoplasmic reticulum provoke a dominant-negative influence culminating in degeneration of photoreceptors and autosomal-dominant retinitis pigmentosa (146).
Congenital stationary night blindness might appear superficially as an activating mutation of rhodopsin but was shown to be the result of various null mutations in the rhodopsin kinase gene whose action is to dampen the activation of the bleached photopigment (623), a likely cause of some forms of Oguchi disease. The molecular genetics of inherited variation in color vision are available eleswhere (405). Two gain-of-function mutations in the parathyroid hormone (PTH)/PTH-related peptide (PTHrP)receptor yield Jansen’s metaphyseal chondrodysplasia (276, 277,488, 489). The His223Arg mutation occurs in the first intracellular loop of this GPLR (488). A second mutation occurring in the sixth transmembrane-spanning domain, Thr410Pro, also constitutively activates the PTH/PTHrP receptor in an agonist-independent fashion (488). Interestingly, introducing these same mutations into the highly conserved histidine and threonine residues of several other receptors has no obvious activating effect (488). It is likely that many other disorders derivative of mutations in GPLR that produce gain-of-function will be identified.
2. Loss-of-function mutations in GPLR
Loss-of-function mutations of GPLR have been identified in ACTH, follicle-stimulating hormone, growth hormone-releasing hormone (GHRH), LH, and TSH receptors as well as the photopigments. Detailed tabulation of the mutations of these GPLR has been compiled (518, 520). The mutations can yield loss of receptor mRNA synthesis or protein translation, inappropriate targeting of the synthesized receptor, or loss-of-function of the receptor, although the mutant receptor may be properly expressed and targeted. Based on the generalized structure of GPLR and their functional domains, known loss-of-function by the molecule might reflect alteration of agonist binding capacity, of agonist activation of the receptor, or of signal transduction from the receptor to the G protein. The bulk of these receptor mutations produce well-characterized, inherited diseases and are transmitted as autosomal recessives, requiring either homozygosity for a defect or perhaps compound heterozygous defects for clinical manifestation.
Numerous loss-of-function mutations of the V2 vasopressin receptor have been described (425, 473). Mutations include frame-shift, missense, and nonsense mutations, not confined to any one domain of the V2 vasopressin receptor. This array of mutations produces a rather uniform clinical entity, nephrogenic diabetes insipidus (NDI) (36). Although vasopressin levels are elevated, patients with NDI are unable to concentrate urine. The V2 receptor gene maps to Xq28. One of two familial types of NDI is inherited in an X-linked manner and involves mutations in this gene (375,425, 473, 577), whereas the autosomal recessive disorder involves a mutation of the aquaporin-2 gene. Because males are hemizygous for the V2 receptor gene, a single mutation that affects receptor function will yield clinical NDI in males, whereas most female carriers will not manifest the clinical features of NDI. Males with X-linked NDI derivative of a mutation in the V2 vasopressin receptor gene likewise lack the clotting factor response, although those actions of vasopressin mediated by the V1a and V1breceptors, such as the hypertensive response, are unaffected. The occurrence of more than 70 distinct loss-of-function mutations that occur in the V2 receptor gene has no obvious explanation. The V2 gene is clearly a stand out compared with the prevalence of loss-of-function mutations in other GPLR.
Loss-of-function mutations have been identified in the calcium-sensing receptor that can precipitate familial benign hypercalcemia (heterozygous) or neonatal severe hyperparathyroidism (homozygous) (66, 67, 443). Mutations in the ACTH receptor that reduce function yield familial glucocorticoid deficiency (109, 564). Familial growth hormone deficiency can result from loss-of-function mutations in the GHRH receptor detailed elsewhere (518), including a missense mutation in the extracellular domain that disrupts the ability of the receptor to signal (191,368). The central role of cell signaling via G proteins in endocrine regulation provides a basis for the preponderance of endocrine diseases involving loss-of-function mutations of GPLR in humans. Aganglionic megacolon (also known as Hirschsprung disease) has been shown to result from a variety of mutations in the endothelinB receptor (450). Loss-of-function mutations in the thromboxane A2 receptor, especially the Arg-60 to Leu substitution, results in congenital bleeding (234). A Gly-40 to Ser mutation of the glucagon receptor reduces receptor affinity for glucagon by threefold and shows some linkage to type II diabetes mellitus (212) as well as a possible role in essential hypertension (87). Needless to say, loss-of-function mutations are not confined to any one subset or family of the GPLR superfamily.
A long list has accumulated of clinical conditions associated with possible alterations in GPLR that may not involve mutations of the receptor themselves, but rather alterations in the molecules with which they interact. For adrenergic receptors, as an example, the list includes chronic activation by agonist (and pheochromocytoma), antagonist withdrawal, asthma, congestive heart failure, hypertension, and myocardial ischemia. Much further analysis of these and other important disease states will be required to elucidate the precise mechanisms by which GPLR-mediated signaling is being influenced.
B. Heterotrimeric G Protein-Based Diseases
1. Gain-of-function alterations in G proteins
Elaboration of toxin by the infectious microbe Vibrio cholera is associated with increased transluminal movement of water and electrolytes characteristic of the uncontrollable, often fatal diarrhea accompanying infection with cholera. The realization that the pathophysiology of cholera intoxication resulted from a covalent modification (ADP-ribosylation) of the Gαs, which in turn activated the molecule (188), heightened awareness that G protein-controlled pathways critical to physiological regulation could be the targets not only of bacterial toxins but also heritable or spontaneous mutations with the same influence. As the nature of heterotrimeric G protein biology was better understood, so was the unique role of the intrinsic GTPase in determining the activation state of the α-subunit (188).
Sporadic pituitary somatotroph tumors that secrete growth hormone as well as thyroid hormone hypersecreting-thyroid adenomas were found to reflect mutations of the Gαs molecule (353). On the basis of study of the residues of G protein α-subunits important in the GTPase function, it was perhaps predictable that the residues mutated in these endocrine tumors involved several sites well known to affect GTPase activity. Arginine-201 and Gln-227 are common sites for mutations of this nature, and mutations of both of these sites reduce intrinsic GTPase activity, responsible for turning off the GTP-activation cycle. The reduction in GTPase activity creates a gain-of-function mutant that is constitutively activated in the absence of the usual stimulation via agonist-activated GPLR. Because elevated cAMP levels are mitogenic for many cell types (188), including those of several endocrine tissues, it is no wonder that these mutations produce not only hypersecretion by cAMP-sensitive pathways but also hypertrophy and/or hyperplasia.
Gain-of-function mutations in G protein α-subunits have been engineered into transgenic mice in an effort to evaluate the role of these subunits in signaling in vivo (93). Targeted, inducible expression of Gαi-2, harboring a Glu205Leu activating mutation, in liver, skeletal muscle, and adipose tissue of transgenic mice revealed an important role of Gαi-2 in regulation of insulin action (93). Mice harboring the Glu205Leu Gαi-2 expression transgene display a reduced fasting glucose level, increased tolerance to glucose loading, and biochemical manifestations suggesting that the activated G protein produces an insulinomimetic effect (93, 198). These mice provide another example in which signaling via G proteins cross over to signaling in the realm of growth factor receptors with intrinsic tyrosine kinase activity, such as insulin and IGF-I receptors. Expression of activated forms of other heterotrimeric G proteins in transgenic mice has not been reported. Expression of constitutively activated Gαq and Gα13 in Chinese hamster ovary cells and COS cells in vitro, for example, triggers apoptosis (3) and provides and opportunity for follow-up studies in transgenic mice.
In humans, activating mutations of Gαi-2, much like the Glu205Leu, have been identified in neoplasms of the ovary as well as the cortex of the adrenal (352, 353). A Gly38Asp germline mutation of the α-subunit of transducin (Gt) has been identified in an autosomal dominant form of stationary night blindness (144). The mutation in a G protein α-subunit whose expression is confined almost exclusively to the retinal rod appears analogous to the activating mutation observed in p21ras and constitutively activates the phototransduction pathway causing night blindness, often referred to as “Nougaret disease” for the French 17th century butcher traced as the origin (145).
A most intriguing example of gain-of-function mutations in G protein α-subunits causing human disease is the case of McCune-Albright syndrome (607, 608). McCune-Albright syndrome is a sporadic disease typified by precocious puberty, monoostotic or polyostotic fibrous dysplasia, cafe au lait pigmentation, and several endocrinopathies. Hyperthyroidism, Cushing’s syndrome, hyperparathyroidism, acromegaly, and hepatomegaly are frequently observed in patients with this syndrome. The hyperactivity of the endocrine tissues appears to result from mutation of Arg-201 in Gα which generates gain of function (608). The expression of activated Gαs is not uniform in McCune-Albright syndrome, but rather is a complex pattern reflecting the occurrence of a mutation in the multicell, developing embryo generating a mosaic of expression that reflects the fate map of the cell in which the mutation arises (607,608). The patients with McCune-Albright syndrome are considered to be true chimeras, the severity of the disease manifest largely by the spectrum of organs and tissues in which activated Gαs occurs. Other studies suggest that when severe, the McCune-Albright syndrome may be the cause, although not specified as such, of early death in childhood, especially when the activated G protein is expressed in tissues such as the liver, heart, and gastrointestinal tract, all nonclassic targets of McCune-Albright syndrome (501). Thus activating mutations of Gαs can provoke a range of pathologies, from thyroid adenomas and fibrous dysplasia to the constellation of alterations observed in McCune-Albright syndrome (521).
Targeted expression of constitutively activated mutant forms of Gαs in transgenic mice may provide a useful model for study of the role of Gαs in mammalian development. Overexpression of wild-type Gαs in hearts of transgenic mice has been accomplished using the rat α-myosin heavy chain promoter (179). Although mRNA levels for Gαs increased nearly 40-fold, Gαsexpression in the heart increased less than 3-fold, and the cardiac phenotype was little effected (179). The physiological mechanism(s) by which the relative levels of G protein subunit expression are regulated remains to be established and will be essential to our understanding of G protein signaling in human diseases.
2. Loss-of-function alterations in G proteins
Much like toxins of V. cholera have emerged which target Gαs and activate it through covalent modification and reduction of intrinsic GTPase activity, toxins elaborated byBordetella pertussis offer an example of using the same covalent modification to disable signaling by G protein α-subunits of the Gαi-2 family (188, 570). Targeting a cysteinyl residue in the COOH-terminal five residues of these Gα subunits for ADP-ribosylation, pertussis toxin effectively blocks the ability of the G protein to transduce signals to effectors, such as AC. Unlike the activating mutation of Gαs catalyzed by cholera toxin that both activates and dissociates the α-subunit from the β/γ-complex, the ADP-ribosylation of Gαi-2 catalyzed by pertussis toxin stabilizes the α-subunit in the heterotrimeric, inactive state (570).
Albright and co-workers (2, 520) described a disorder in which resistance to PTH was evident. Patients with this disorder showed a subnormal cAMP response in their urine after challenge with PTH, suggesting a lesion proximal to the production of the cyclic nucleotide. This disorder, termed pseudohypoparathyroidism (type 1a), was later found to be associated with resistance to other hormones that activate cAMP synthesis. Biochemical determinations revealed a loss-of-function coinciding with a reduction in the expression of the Gαs protein that activates AC (335). Pseudohypoparathyroidism is associated with other phenotypic features such as obesity, short stature, mental retardation, and bone anomalies, termed Albright hereditary osteodystrophy (519). The colocalization of the Gαs gene and the locus for pseudohypoparathyroidism 1a to the long arm of chromosome 20 provides further genetic evidence to support mutation of Gαs as the basis for the disease. The mutations of Gαs causing pseudohypoparathyroidism in humans are varied but provoke a reduced level of expression of this G protein subunit (606). Creation of viable (−/−) knock-out mice of the Gαs gene has not been reported, although tried repeatedly. Homozygous loss of Gαs is lethal in the embryo, and complex patterns of tissue-specific imprinting have been observed for the heterozygotes with the null allelle (631). These data and the phenotypic changes observed in patients with a heterozygous, germline deficiency of Gαs suggest an obligate and important role for Gαs in normal mammalian development.
Study of the physiological outcome of the deficiency of a G protein subunit in vivo that may be necessary for in utero development poses a formidable problem, addressable by the use of an inducible, tissue-specific approach in which a transgene which is silent in utero is activated at birth and expresses RNA antisense to a defined region of the targeted subunit. Using the phosphoenolpyruvate carboxykinase (PEPCK) gene promoter to drive high-level expression of antisense RNA, it has been possible to generate mice with loss-of-function status for several specific G protein subunits, including Gαi-2 (397) and Gαq (172). Mice with Gαi-2deficiency in liver, adipose tissue, and skeletal muscle yield a severely runted phenotype that does not affect longevity or fecundity (397). Knock-out (−/−) mice deficient for Gαi-2 were reported in which a runted phenotype with ulcerative colitis and adenocarcinoma of the colon was variably present (478, 479).
More detailed analysis of the mice harboring the transgene and Gαi-2 deficiency revealed insulin resistance (398), presenting much like patients with non-insulin-dependent diabetes mellitus (NIDDM). The frank insulin resistance was observed at several well-known insulin responses including glucose tolerance, glucose transport, and GLUT4 transporter recruitment to glycogen synthase activation in skeletal muscle and antilipolytic effects of insulin in fat tissue (398). As noted above, the overexpression of the constitutively activated Gαi-2 provoked the opposite phenotype, enhanced glucose tolerance (93). Thus one can anticipate the occurrence of NIDDM that may reflect mutations in Gαi-2 function.
This same approach permitted the creation of transgenic mice in which Gαq was deficient in skeletal muscle, liver, and adipose tissue (172). The two most important phenotypic characteristics of the mice harboring the inducible vector for RNA antisense to Gαq was their relative obesity compared with control littermates and the absence of PLC activation by a variety of hormones acting at these target tissues. Interestingly, the Gαq-deficient mice displayed fat cells that showed little ability to breakdown stored triglyceride in response to a variety of agents, including catecholamines. Analysis of lipolysis in fat cells prepared acutely from the control littermates with inhibitors of PKC revealed the existence of a non-cAMP-dependent pathway of lipolysis susceptible to inhibition by protein kinase inhibitors, absent in the Gαq-deficient mice (172), and long sought after in the field of lipid metabolism.
With the use of homologous recombination, mice deficient in Gαq were created. Platelets from the Gαq-deficient mice displayed defective platelet activation (417). The role of Gαq in these systems cannot be replaced by Gαi or by β/γ-subunits from other heterotrimeric G proteins, suggesting that the effects reflect deficiency in Gαq itself and not some indirect effect of its absence. Overexpression of Gαq in transgenic mouse heart, in contrast, has been shown to lead to failure of cardiac contraction (126).
Loss-of-function mutants have been sought for other G protein subunits, such as Gα11 and Gα13, both of which failed to produce viable transgenic mice (415). The absence of Gα13 appeared to impair the ability of endothelial cells to organize vascular beds and systems, provoking death in utero. No human disease stated has been linked to loss-of-function mutations of either Gα11 or Gα13.
C. Newly Emerging Therapies for G Protein Signaling Defects
The development of an mammalian expression vector with the qualities of conditional expression and tissue specificity combined with antisense RNA technology provides a novel platform for the parallel development of new therapeutic approaches to disease states in which a mutant gene product is responsible for the pathology (172, 397, 398). McCune-Albright syndrome, discussed in sectionviB1 , is but one example of the occurrence of an activating mutation of a G protein that results in endocrinopathies and aberrations in early childhood development.
In combination, antisense RNA technology and conditional tissue-specific expression vectors offer a new approach (358) to treating severe endocrine and nonendocrine manifestations of diseases, such as McCune-Albright syndrome. Theoretically, the mosaic character of McCune-Albright syndrome provides the opportunity to address tissue-specific manifestations that may be life-threatening, such as hepatobiliary disease or cardiac disease. Tissue-specific elements can be assembled to provide expression limited to target sites, as shown for tissue-specific expression in liver, adipose, and skeletal muscle using the pPCK-AS vector and in heart using the α-myosin heavy chain promoter. Further engineering and/or discovery of tissue-specific expression elements that complement the therapeutic targets will be a formidable task. Tissue-specific expression of transgenes has succeeded in a number of tissues. Designing expression vectors to accommodate the tissues in which antisense RNA must be expressed to ablate the products of mutant genes would appear to pose no insurmountable problems.
Conditional expression provides an additional level of safety for gene therapy. Useful in avoiding lethality of “knock-outs” targeting gene products necessary for development in utero, the PEPCK promoter demonstrated the feasibility of the approach. A variety of other opportunities for conditional expression can be envisioned based on the two-hybrid system in yeast used to probe protein-protein interactions. The DNA binding and activation domains of a specific promoter could be constructed to hybridize upon binding of a small organic molecule, already approved for use in humans by the Food and Drug Administration. Expression of RNA antisense to the mRNA of the mutant G protein in McCune-Albright syndrome, for example, could be conditional based on the administration of a small molecule drug, providing an additional safeguard to that provided through the use of tissue-specific elements for expression. The selection of the proper vector for gene therapy is beyond the scope of this short review, but it should be noted that Moloney murine leukemia virus-based vectors have been employed successfully to express transgenes with PEPCK gene elements in mice. Although speculative, the use of conditional, tissue-specific vectors to express RNA antisense to mRNA of mutant proteins would appear to offer a novel dimension to gene therapy, readily adaptable to a host of human disease states, including McCune-Albright syndrome and a variety of GPLR-based diseases outlined above.
VII. CLOSING COMMENTARY
In the decade since the Gilman review, we have enjoyed an extraordinary expansion in our knowledge concerning signal propagation via heterotrimeric G proteins. To provide a fruitful discussion of the physiological regulation of this signaling requires a current inventory of members of the families and “superfamilies” of the fundamental elements of the system, i.e., GPLR, heterotrimeric G proteins, and their effectors. The growth in each portfolio has been impressive, and the number of new partners of protein-protein interactions discovered for each of the basic elements is equally impressive. The RAMP molecules, let alone scaffolding, cytoskeletal, and adaptor molecules identified as partners for GPLR, could not have been envisioned but just a few years ago. The RGS molecules, their target G proteins, and the ways in which they regulate signaling both biochemically and physiologically remain a high profile target for research. Effectors for G proteins now include tyrosine kinases, such as Btk, while both G proteins and GPLR have been shown themselves to be targets for tyrosine kinases. G proteins have established roles in more complex biological processes, including oncogenesis, differentiation, development, and insulin action, but to name a few examples. Disruption of these complex pathways by mutation often yield well-known human pathologies, such as McCune-Albright syndrome. The full understanding of these human pathologies will require a broader appreciation of the physiological, biochemical, and molecular features of signaling as well as pathophysiologies derivative of alterations in the signaling process.
Our original goals guiding the evolution of this review over the past 18 mo recognized two real limitations, our inability to include all relevant primary citations and our need to ensure the development of the broader work at the expense of more in-depth discussions of each topic. The solution to the vexing aspects of these two forced decisions was to include many current, high-quality reviews as primary sources for the reader to delve further into the primary data. To accusations that this compromised the review, we rightfully plead nolo contendere and justifiably will suffer some just, yet unavoidable criticism. In truth, it is a sense of incompleteness that punishes us more severely, as we admit to the desire to describe more fully the rapid advances in cell biology that have permitted us to approach signaling by exploring the cellular locale of the G protein-coupled devices that transduce the signals. Not only can we detect receptor sequestration as made visible by epifluorescence and confocal microscopy, but also we can examine the dynamics of changes in cellular localization as an output for understanding how changes in localization and/or composition of these signaling devices contributes to the biology of the system. Much more study of this dimension of the problem will be required before we can appreciate adequately the basis for the rich diversity that exists among the key elements of G protein signaling. Review of the cell biology of G protein signaling devices would be an equally ambitious topic, intentionally avoided herein.
The task of enhancing the level of description of G protein signaling devices in situ is most formidable. Even the limitations of the very best in confocal microscopy seem intolerable to those seeking greater levels of definition. Impressive advances in large-scale expression and purification of signaling elements, good fortune, and intense labor aimed at obtaining usable crystals for X-ray analysis have yielded great dividends to the structural biologist and cell biologist alike. The cocrystallization of G protein heterotrimers alone, with RGS, and effectors in tandem with X-ray analysis is providing structural information on the elements whose localization we long to capture dynamically at lower resolution in living cells. These two approaches to study G protein elements share a continuum that will become populated with more information as advances in “soft” structural biology merge with intense efforts to describe large macromolecular targets such as G protein signaling devices by confocal and epifluorescence microscopy.
As we delve further into how the localization, stoichiometry, protein-protein interactions, and structural features of G protein signaling contribute to the ultimate read-out, i.e., physiological regulation, new opportunities for understanding pathophysiologies in humans will emerge. Whether through delineation of mutations that alter expression, structure and/or function of components of G protein signaling devices, our task will be to then provide a broader analysis relating these alterations to the physiological read-out. Among the opportunities created is the identification of new targets for therapeutic intervention. Perhaps it will be new, small organic molecules targeted at RGS or RAMP that will be used to alter signaling throughput in conditions where hyper- or hypostimulation of the pathways is the basis for human pathophysiologies. Antisense oligodeoxynucleotides, newly created inverse agonists, and development of new drug molecules that can alter important, posttranslational modifications of components of G protein-linking signaling devices already enjoy major research investments in both the public and private sectors. The investment market for this portifolio likely will be bullish for some time.
The final integration of G protein signaling emerges as physiological regulation. Once molecular biology completes the identification of the members of the GPLR, G protein subunits, effectors, and interacting proteins, the task at the cellular level will be to explore how regulation of transcriptional activation/suppression, posttranscriptional regulation, and posttranslational modification of these proteins contribute to expression, localization, structure, function, and regulation. Transgenic mouse technology will enable binary questions of loss-of-function and gain-of-function to be posed in the context of a multicellular, complex model. In time, the convergence of these gains in understanding in tandem with their application to the study of human disease will culminate in elucidating the fullest dimensions of physiological regulation of G protein signaling. The pace and intensity of the effort suggest that much will be gained well before another decade has passed.
We express our appreciation to the many investigators that provided preprints and advance copies of papers submitted for publication throughout the execution of this review. In particular, the advice and help of Drs. Suleiman W. Bahouth (Dept. of Pharmacology, Univ. of Tennessee Health Sciences Center, Memphis, TN), Hsien-yu Wang (Dept. of Physiology & Biophysics, SUNY, Stony Brook, NY), and Kevin R. Lynch (Dept. of Pharmacology, Univ. of Virginia Health Sciences Center, Charlottesville, VA) were invaluable in this project. Finally, we thank Roxanne Brockner for the patient and competent management of the references and informatics required in this effort.
We acknowledge the late Dr. Martin Rodbell, a scientist who contributed so much to our molecular understanding of signal transduction and to the meaning of creative thinking.
The research in the author’s laboratories has been supported generously by The National Institutes of Health and the American Cancer Society.
Address for reprint requests and other correspondence: C.C. Malbon, Dept. of Molecular Pharmacology, Diabetes and Metabolic Research Center, University Medical Center, SUNY/Stony Brook, Stony Brook, NY 11794-8651.
- Copyright © 1999 The American Physiological Society