Pharmac. Ther. Vol. 50, pp. 271-283, 1991 Printed in Grcat Britain.All rights reserved
0163-7258/91 $0.00+ 0.50 © 1991 PergamonPressplc
Specialist Subject Editor: C. W. TAYLOR
THE REGULATION OF ADENYLYL CYCLASE BY RECEPTOR-OPERATED G PROTEINS ALEXANDER LEVITZKI a n d ALLAN BAR-SINAI Department of Biological Chemistry, Hebrew University of Jerusalem, 91904 Jerusalem, Israel Abstract--The receptor regulated adenylyl cyclase system is a multiprotein complex which is a member of the family of the receptor-effector systems whose signal is transduced by heterotrimeric GTP-binding proteins. The system consists of stimulatoryand inhibitory receptors (Rsand Ri), stimulatoryand inhibitory G proteins (Gs and Gi) and the adenylyl cyclase enzyme (C). While quite specific in situ, receptors (stimulatory or inhibitory) from one source can activate the appropriate G protein from other cell types or species which in turn can act on C from other sources. Studies with chimeric proteins have shown that the various specificities (stimulatory or inhibitory) can be mapped to defined domains in both receptors and G proteins. The mechanism by which the heterotrimeric G proteins couple to the stimulatory and inhibitory signals is discussed in detail. Specifically, the data supporting collision coupling vs the shuttle mechanism is reviewed, as well as the role of fl~,subunits in both the stimulatory and inhibitory signals.
CONTENTS 1. Introduction 2. The Functional Units of Hormonally Regulated Adenylyl Cyclase 3. Universality of the System 4. Stimulation of Adenylyl Cyclase 4.1. General features 4.2. Catalytic role of the receptor and 'collision coupling' 4.3. G s 'shuttle' mechanisms 4.4. Conformational transitions 5. The (Gs-Gi) Subunit Dissociation Mechanism for Hormonal Activation and Inhibition 5.1. General considerations 5.2. fl), Subunits are tightly associated with activated cyclase 5.3. Kinetic considerations 6. Mechanism of Gi Action References
1. I N T R O D U C T I O N Historically, the hormone-activated adenylyl cyclase was the first receptor-controlled intracellular messenger system discovered. Today it remains one of the best studied systems, one in which all of the essential protein components are known (Gilman, 1987; Levitzki, 1988; Birnbaumer et aL, 1990), although other less essential control elements may yet be discovered. The adenylyl cyclase system is part of a family of receptor-effector systems which depend on heterotrimeric G proteins to control transduction of the signal from the receptor to the effector (Taylor, 1990). The receptors in this family all possess very similar putative three dimensional structures, spanning the lipid bilayer seven times. The effectors of this family, however, are much less well known, with only adenylyl cyclase having been recently cloned (Krupinski et ai., 1989). The beterotdmeric G protiens are, in turn, members of the GTPase supeffamily that includes many of the most important proteins involved in signal transduction
271 271 272 273 273 274 274 275 275 275 277 278 279 281
and control of cellular processes (Bourne et al., 1990, 1991). We shall try, in this review, to summarize our understanding of the mechanism of signal transduction through G proteins to adenylyl cyclase. Although we believe that many of the mechanisms involved in signal transduction through the adenylyl cyclase system can be generalized throughout the family of G protein controlled receptors, care should always be used in applying mechanisms from one receptor-effector system to another. 2. THE F U N C T I O N A L UNITS O F HORMONALLY REGULATED A D E N Y L Y L CYCLASE Receptor-regulated adenylyl cyclase, is a multiprotein system composed of 5 functional units (Fig. 1). (a) A stimulatory receptor R~ which binds the stimulatory hormone or the neurotransmitter. We shall discuss in detail the fl-adrenoceptor which is a 271
272
A. LEVITZKIand A. BAR-SINAI
GTp ATP
,N c A M P + PPi
FIG. 1. Components involved in the dual regulation of adenylyl cyclase. The catalyst C, which is a transmembrane protein spanning the membrane several times, responds to the stimulatory G protein G, and to the inhibitory G protein G i, both of which are localized to the inner leaflet of the membrane. It is not yet clear whether G i interacts directly with C or not. The stimulatory signal is initiated by the stimulatory receptor 1~ and the inhibitory signal by the inhibitory receptor R~, both of which span the membrane a putative seven times. Cholera toxin catalyzed ADP-ribosylation of G, inhibits the GTPase activity of (3, whereas pertussis toxin catalyzed ADP-ribosylation of G i inhibits GDP to GTP exchange o n G i . In both cases the transmission of the message to C is modified. Inhibition of the G, GTPase locks the G,C complex in its active state. Inhibition of the GDP to GTP exchange on G~ locks G~ in its inactive conformation thus nullifying inhibitory action on C. representative member of this family of receptors (Levitzki, 1988; Lefkowitz and Caron, 1988). This receptor, like all known receptors which interact with G proteins, is a transmembrane glycoprotein with 7 putative membrane-spanning sequences. The fladrenoceptor is the only member of this large family of receptors which has been purified and successfully reconstituted with other pure components of the adenylyl cyclase system (Feder et al., 1986a,b; May et al., 1985). (b) The G s protein which binds GTP and activates the catalytic unit, adenylyl cyclase (C). Gs is composed of 3 subunits: as, which possesses the GTP-binding site and is the target for cholera toxin catalyzed ADP-ribosylation by N A D +, and the fl and subunits, which are tightly associated with each other (Cassey and Gilman, 1988 for review). The heterotrimeric Gs protein is localized in the inner leaflet of the plasma membrane and is much less hydrophobic than either the fl-adrenoceptor or the catalytic unit of adenylyl cyclase. (c) The catalytic unit (C) is a hydrophobic protein with a multitude of transmembrane spanning domains carrying its catalytic site at the cytoplasmic side of the membrane (Krupinski et al., 1989). (d) Inhibitory receptors, P~ which bind inhibitory neurotransmitters or inhibitory hormones and which, like fl-adrenoceptors, span the membrane 7 times. (e) The inhibitory GTP-binding protein Gi. This protein, like G~, is a heterotrimer composed of three subunits: 0~, fl and 7, where the fl~ complex is highly similar or identical to the one found in G, (Cassey and Gilman, 1988 for review). The subunit ~'i is homologous to g~ but is a substrate for ADP-ribosylation catalyzed by the toxin of Bordetella pertussis (pertussis toxin), but not by cholera toxin (Gilman, 1987). In addition, 0~i is myristoylated anchoring it in the membrane, while ~,, undergoes no such post-translational modifi-
cation (Jones et al., 1990; Mumby et al., 1990). The homology is, as expected, high in the sequences involved in the binding of GTP, but lower at the carboxy terminal regions which participate in the G protein-receptor interactions. It is also likely that the other regions of divergence in the sequence are involved in the interactions with the effector and with the inhibitory receptor. Since cts and ~i are likely to interact at different domains of the catalyst C (Section 5), it is to be expected that the effector domain of ct will be different for each G protein. Recently, studies using point mutations, deletion studies and ~--~i chimeras have shed much light on the function of the various domains of the ~t subunit. This work has recently been extensively reviewed (Bourne et al., 1991; see also Osawa et al., 1990). Briefly, however, the extreme C-terminal end is thought to participate in receptor contact, the Nterminal end may act as a function attenuator, and the effector region seems to map to the C-terminal portion of the protein. However, the assignment of specific functions to the various domains must be thought of as putative for the present. So far it has been generally accepted that it is the ~-subunit which transmits the signal to the effector. Recently, however, it was pointed out that there may be differences between the fly subunits of different G proteins (Cassey and Gilman, 1988) and therefore, each class of receptors may recognize a specific subset of ctfly structures where the heterotrimeric combination of the subunits generates the specificity to both the receptor and the effector. When more is known about the diversity of fl~s and their distribution in different G proteins, roles in the specificity of the signal produced by the G protein may become clearer. The multitude of genes coding for fl and ~ subunits (Birnbaumer et al., 1990; Gautam et al., 1990) suggests that the specificity of the G protein may be determined by the ~fly heterotrimer rather than by the ~t-subunit alone. As far as Gs and G i a r e concerned, however, it seems that the fly subunits are identical and functionally interchangeable (for reviews see Gilman, 1987; Birnbaumer et al., 1990).
3. UNIVERSALITY OF THE SYSTEM One of the key observations in the 1970s was that stimulatory receptors from one cell type can hybridize with a heterologous adenylyl cyclase system when transplanted into the appropriate cell type. For example, fl-adrenoceptors from turkey erythrocytes and glucagon receptors from rat liver can be transferred into Friend erythroleukemia cells and activate its adenylyl cyclase (Orly and Schramm, 1976; Schramm et al., 1977; Schramm, 1979). Kinetic experiments also supported the hypothesis that two different receptor types (namely A2 adenosine receptors and flm-adrenoceptors) in the same cell couple to a single pool of adenylyl cyclase and therefore of G, proteins, both in turkey erythrocytes (Sevilla et al., 1977; Tolkovsky and Levitzki, 1978b) and rat brain (Braun and Levitzki, 1979). These experiments suggested already in the late 1970s, that receptors
273
Regulation of adenylyl cyclase which differ markedly in their pharmacological specificity possess common structural domains which participate in the interaction between the stimulatory receptor and G,. Recent cloning and sequencing studies on f12- and flradrenoceptors support this hypothesis. It was found (Lefkowitz and Caron, 1988) that certain sequences are involved with ligand binding while others are involved in interactions with the G protein. This feature seems to be generally true for receptors which interact with G proteins. Chimeras of f12- and ~te-adrenoceptors have been constructed, such that the pharmacological specificity is that of the cyclase inhibitory ~2-adrenoceptor whereas the biochemical activity is of the fl-type. Thus an ~t?adrenergic ligand (clonidine), which is normally associated with adenylyl cyclase inhibition, activates the enzyme when it interacts with the ~2-fl2 adrenoceptor chimera where the G~/fl-adrenoceptor interface is preserved (Kobilka et al., 1988). Recently, other studies involving M I muscarinic-fl-adrenergic chimeras (Wong et al., 1990) and ill-t2 adrenergic chimeras (Marullo et al., 1990) have shown that this principle can be widely applied. These experiments constitute an elegant proof for the concept of structural separation between the ligand binding domain and the G-protein interaction domain on the receptor. Most strikingly, the mechanism of interaction between these two domains within the protein seems to be independent of the specificity of the domains. G~ seems to be universal also in respect to C since it can couple to a variety of C units from different sources. For example, a functional fl-adrenoceptordependent adenylyl cyclase complex can be reconstituted in $49 cyc- membranes using G~ from various sources (e.g. wild type $49 lymphoma cells, rabbit liver, turkey erythrocytes and human erythrocytes) (for review, see Levitzki, 1986; Ross and Gilman, 1980). We have shown that the catalytic unit of adenylyl cyclase from two different species, bovine brain and rabbit myocardium couples equipotently with G~ from either turkey erythrocytes or rabbit liver (Feder et al., 1986a). These results suggest that C is also a highly conserved molecule. Studies using monoclonal antibodies against C reveal that bovine brain C and rabbit myocardial C are similar, but not identical (Mollner and Pfeuffer, 1988). Cloning work from Reed and coworkers (Krupinski et al., 1989; Bakalyar and Reed, 1990) on adenylyl cyclase reveals that there are a few classes of Grresponsive cyclases all similar in overall structure, but quite diverse in amino acid sequence. A few variants of Gi are also known from cloning studies, but the functional differences are not yet understood (Birnbaumer et al., 1990). It is not clear at this point how many Gi type proteins are truly involved with the inhibition of adenylyl cyclase and whether they participate in other signal transduction pathways. G~_ 2, however, is considered the most likely candidate for mediating the inhibitory tonus on adenylyl cyclase (Simonds et ai., 1989; Senogles et al., 1990). The universality of the components has allowed more flexibility in reconstitution experiments in which components from various species have been coreconstituted to produce functional interactions (Feder et al., 1986a,b; May et al., 1985).
4. STIMULATION OF ADENYLYL CYCLASE 4.1. GENERALF~.nPY.S In the native state both hormone binding to R~ and GTP binding to G, at the Gt,-subunit are required to induce activation of C to the cAMP-producing state. The active state of the system decays concomitantly with the hydrolysis of GTP to GDP and Pi at the G, regulatory site. Replenishment of G, with GTP and the continued presence of hormone at the receptor allow the system to regain its active cAMP-producing state. This type of 'on-off" cycle accounts well for the properties of hormone-dependent adenylyl cyclase (see Levitzki, 1988, for review) and actually for all processes of activation of effectors through G proteins. The hallmarks of adenylyl cyclase activation by receptors and G proteins can be summarized as follows: (1) The hormone-dependent adenylyl cyclases possess a slow (k~t ~_ 15 min -l) hormone-dependent GTPase activity. This activity has been demonstrated in fl-adrenoceptor-dependent cyclase (Cassel and Selinger, 1976), in glucagon-dependent cyclase, pancreoenzymin-dependent cyclase, and in prostaglandin El-dependent cyclase (Levitzki, 1982 for review) as well as in fl-adrenoceptor G, reconstituted systems (Lefkowitz and Carom 1988; Gilman, 1987). (2) Nonhydrolyzable GTP analogs such as GPPNHP and GTP?S, generate a constitutively active enzyme due to the complete blockade of the 'off" GTPase reaction (Gilman, 1987, for review). (3) Inhibition of the GTPase reaction by cholera toxin catalyzed ADP-ribosylation of ~q activates the cyclase due to the slow-down of the 'off" GTPase reaction (see Gilman, 1987; Levitzki, 1982, 1986, for reviews). (4) From independent measurements of the rate of enzyme activation by hormone and guanyl nucleotide (the 'on' reaction) and the decay of the cAMPproducing state to its basal state (the 'off" reaction), one can compute the fraction of the total pool of cyclase which is in the active state in the presence of GTP (Fig. 2). These measurements (Arad and
°,,
0163-7258/91 $0.00+ 0.50 © 1991 PergamonPressplc
Specialist Subject Editor: C. W. TAYLOR
THE REGULATION OF ADENYLYL CYCLASE BY RECEPTOR-OPERATED G PROTEINS ALEXANDER LEVITZKI a n d ALLAN BAR-SINAI Department of Biological Chemistry, Hebrew University of Jerusalem, 91904 Jerusalem, Israel Abstract--The receptor regulated adenylyl cyclase system is a multiprotein complex which is a member of the family of the receptor-effector systems whose signal is transduced by heterotrimeric GTP-binding proteins. The system consists of stimulatoryand inhibitory receptors (Rsand Ri), stimulatoryand inhibitory G proteins (Gs and Gi) and the adenylyl cyclase enzyme (C). While quite specific in situ, receptors (stimulatory or inhibitory) from one source can activate the appropriate G protein from other cell types or species which in turn can act on C from other sources. Studies with chimeric proteins have shown that the various specificities (stimulatory or inhibitory) can be mapped to defined domains in both receptors and G proteins. The mechanism by which the heterotrimeric G proteins couple to the stimulatory and inhibitory signals is discussed in detail. Specifically, the data supporting collision coupling vs the shuttle mechanism is reviewed, as well as the role of fl~,subunits in both the stimulatory and inhibitory signals.
CONTENTS 1. Introduction 2. The Functional Units of Hormonally Regulated Adenylyl Cyclase 3. Universality of the System 4. Stimulation of Adenylyl Cyclase 4.1. General features 4.2. Catalytic role of the receptor and 'collision coupling' 4.3. G s 'shuttle' mechanisms 4.4. Conformational transitions 5. The (Gs-Gi) Subunit Dissociation Mechanism for Hormonal Activation and Inhibition 5.1. General considerations 5.2. fl), Subunits are tightly associated with activated cyclase 5.3. Kinetic considerations 6. Mechanism of Gi Action References
1. I N T R O D U C T I O N Historically, the hormone-activated adenylyl cyclase was the first receptor-controlled intracellular messenger system discovered. Today it remains one of the best studied systems, one in which all of the essential protein components are known (Gilman, 1987; Levitzki, 1988; Birnbaumer et aL, 1990), although other less essential control elements may yet be discovered. The adenylyl cyclase system is part of a family of receptor-effector systems which depend on heterotrimeric G proteins to control transduction of the signal from the receptor to the effector (Taylor, 1990). The receptors in this family all possess very similar putative three dimensional structures, spanning the lipid bilayer seven times. The effectors of this family, however, are much less well known, with only adenylyl cyclase having been recently cloned (Krupinski et ai., 1989). The beterotdmeric G protiens are, in turn, members of the GTPase supeffamily that includes many of the most important proteins involved in signal transduction
271 271 272 273 273 274 274 275 275 275 277 278 279 281
and control of cellular processes (Bourne et al., 1990, 1991). We shall try, in this review, to summarize our understanding of the mechanism of signal transduction through G proteins to adenylyl cyclase. Although we believe that many of the mechanisms involved in signal transduction through the adenylyl cyclase system can be generalized throughout the family of G protein controlled receptors, care should always be used in applying mechanisms from one receptor-effector system to another. 2. THE F U N C T I O N A L UNITS O F HORMONALLY REGULATED A D E N Y L Y L CYCLASE Receptor-regulated adenylyl cyclase, is a multiprotein system composed of 5 functional units (Fig. 1). (a) A stimulatory receptor R~ which binds the stimulatory hormone or the neurotransmitter. We shall discuss in detail the fl-adrenoceptor which is a 271
272
A. LEVITZKIand A. BAR-SINAI
GTp ATP
,N c A M P + PPi
FIG. 1. Components involved in the dual regulation of adenylyl cyclase. The catalyst C, which is a transmembrane protein spanning the membrane several times, responds to the stimulatory G protein G, and to the inhibitory G protein G i, both of which are localized to the inner leaflet of the membrane. It is not yet clear whether G i interacts directly with C or not. The stimulatory signal is initiated by the stimulatory receptor 1~ and the inhibitory signal by the inhibitory receptor R~, both of which span the membrane a putative seven times. Cholera toxin catalyzed ADP-ribosylation of G, inhibits the GTPase activity of (3, whereas pertussis toxin catalyzed ADP-ribosylation of G i inhibits GDP to GTP exchange o n G i . In both cases the transmission of the message to C is modified. Inhibition of the G, GTPase locks the G,C complex in its active state. Inhibition of the GDP to GTP exchange on G~ locks G~ in its inactive conformation thus nullifying inhibitory action on C. representative member of this family of receptors (Levitzki, 1988; Lefkowitz and Caron, 1988). This receptor, like all known receptors which interact with G proteins, is a transmembrane glycoprotein with 7 putative membrane-spanning sequences. The fladrenoceptor is the only member of this large family of receptors which has been purified and successfully reconstituted with other pure components of the adenylyl cyclase system (Feder et al., 1986a,b; May et al., 1985). (b) The G s protein which binds GTP and activates the catalytic unit, adenylyl cyclase (C). Gs is composed of 3 subunits: as, which possesses the GTP-binding site and is the target for cholera toxin catalyzed ADP-ribosylation by N A D +, and the fl and subunits, which are tightly associated with each other (Cassey and Gilman, 1988 for review). The heterotrimeric Gs protein is localized in the inner leaflet of the plasma membrane and is much less hydrophobic than either the fl-adrenoceptor or the catalytic unit of adenylyl cyclase. (c) The catalytic unit (C) is a hydrophobic protein with a multitude of transmembrane spanning domains carrying its catalytic site at the cytoplasmic side of the membrane (Krupinski et al., 1989). (d) Inhibitory receptors, P~ which bind inhibitory neurotransmitters or inhibitory hormones and which, like fl-adrenoceptors, span the membrane 7 times. (e) The inhibitory GTP-binding protein Gi. This protein, like G~, is a heterotrimer composed of three subunits: 0~, fl and 7, where the fl~ complex is highly similar or identical to the one found in G, (Cassey and Gilman, 1988 for review). The subunit ~'i is homologous to g~ but is a substrate for ADP-ribosylation catalyzed by the toxin of Bordetella pertussis (pertussis toxin), but not by cholera toxin (Gilman, 1987). In addition, 0~i is myristoylated anchoring it in the membrane, while ~,, undergoes no such post-translational modifi-
cation (Jones et al., 1990; Mumby et al., 1990). The homology is, as expected, high in the sequences involved in the binding of GTP, but lower at the carboxy terminal regions which participate in the G protein-receptor interactions. It is also likely that the other regions of divergence in the sequence are involved in the interactions with the effector and with the inhibitory receptor. Since cts and ~i are likely to interact at different domains of the catalyst C (Section 5), it is to be expected that the effector domain of ct will be different for each G protein. Recently, studies using point mutations, deletion studies and ~--~i chimeras have shed much light on the function of the various domains of the ~t subunit. This work has recently been extensively reviewed (Bourne et al., 1991; see also Osawa et al., 1990). Briefly, however, the extreme C-terminal end is thought to participate in receptor contact, the Nterminal end may act as a function attenuator, and the effector region seems to map to the C-terminal portion of the protein. However, the assignment of specific functions to the various domains must be thought of as putative for the present. So far it has been generally accepted that it is the ~-subunit which transmits the signal to the effector. Recently, however, it was pointed out that there may be differences between the fly subunits of different G proteins (Cassey and Gilman, 1988) and therefore, each class of receptors may recognize a specific subset of ctfly structures where the heterotrimeric combination of the subunits generates the specificity to both the receptor and the effector. When more is known about the diversity of fl~s and their distribution in different G proteins, roles in the specificity of the signal produced by the G protein may become clearer. The multitude of genes coding for fl and ~ subunits (Birnbaumer et al., 1990; Gautam et al., 1990) suggests that the specificity of the G protein may be determined by the ~fly heterotrimer rather than by the ~t-subunit alone. As far as Gs and G i a r e concerned, however, it seems that the fly subunits are identical and functionally interchangeable (for reviews see Gilman, 1987; Birnbaumer et al., 1990).
3. UNIVERSALITY OF THE SYSTEM One of the key observations in the 1970s was that stimulatory receptors from one cell type can hybridize with a heterologous adenylyl cyclase system when transplanted into the appropriate cell type. For example, fl-adrenoceptors from turkey erythrocytes and glucagon receptors from rat liver can be transferred into Friend erythroleukemia cells and activate its adenylyl cyclase (Orly and Schramm, 1976; Schramm et al., 1977; Schramm, 1979). Kinetic experiments also supported the hypothesis that two different receptor types (namely A2 adenosine receptors and flm-adrenoceptors) in the same cell couple to a single pool of adenylyl cyclase and therefore of G, proteins, both in turkey erythrocytes (Sevilla et al., 1977; Tolkovsky and Levitzki, 1978b) and rat brain (Braun and Levitzki, 1979). These experiments suggested already in the late 1970s, that receptors
273
Regulation of adenylyl cyclase which differ markedly in their pharmacological specificity possess common structural domains which participate in the interaction between the stimulatory receptor and G,. Recent cloning and sequencing studies on f12- and flradrenoceptors support this hypothesis. It was found (Lefkowitz and Caron, 1988) that certain sequences are involved with ligand binding while others are involved in interactions with the G protein. This feature seems to be generally true for receptors which interact with G proteins. Chimeras of f12- and ~te-adrenoceptors have been constructed, such that the pharmacological specificity is that of the cyclase inhibitory ~2-adrenoceptor whereas the biochemical activity is of the fl-type. Thus an ~t?adrenergic ligand (clonidine), which is normally associated with adenylyl cyclase inhibition, activates the enzyme when it interacts with the ~2-fl2 adrenoceptor chimera where the G~/fl-adrenoceptor interface is preserved (Kobilka et al., 1988). Recently, other studies involving M I muscarinic-fl-adrenergic chimeras (Wong et al., 1990) and ill-t2 adrenergic chimeras (Marullo et al., 1990) have shown that this principle can be widely applied. These experiments constitute an elegant proof for the concept of structural separation between the ligand binding domain and the G-protein interaction domain on the receptor. Most strikingly, the mechanism of interaction between these two domains within the protein seems to be independent of the specificity of the domains. G~ seems to be universal also in respect to C since it can couple to a variety of C units from different sources. For example, a functional fl-adrenoceptordependent adenylyl cyclase complex can be reconstituted in $49 cyc- membranes using G~ from various sources (e.g. wild type $49 lymphoma cells, rabbit liver, turkey erythrocytes and human erythrocytes) (for review, see Levitzki, 1986; Ross and Gilman, 1980). We have shown that the catalytic unit of adenylyl cyclase from two different species, bovine brain and rabbit myocardium couples equipotently with G~ from either turkey erythrocytes or rabbit liver (Feder et al., 1986a). These results suggest that C is also a highly conserved molecule. Studies using monoclonal antibodies against C reveal that bovine brain C and rabbit myocardial C are similar, but not identical (Mollner and Pfeuffer, 1988). Cloning work from Reed and coworkers (Krupinski et al., 1989; Bakalyar and Reed, 1990) on adenylyl cyclase reveals that there are a few classes of Grresponsive cyclases all similar in overall structure, but quite diverse in amino acid sequence. A few variants of Gi are also known from cloning studies, but the functional differences are not yet understood (Birnbaumer et al., 1990). It is not clear at this point how many Gi type proteins are truly involved with the inhibition of adenylyl cyclase and whether they participate in other signal transduction pathways. G~_ 2, however, is considered the most likely candidate for mediating the inhibitory tonus on adenylyl cyclase (Simonds et ai., 1989; Senogles et al., 1990). The universality of the components has allowed more flexibility in reconstitution experiments in which components from various species have been coreconstituted to produce functional interactions (Feder et al., 1986a,b; May et al., 1985).
4. STIMULATION OF ADENYLYL CYCLASE 4.1. GENERALF~.nPY.S In the native state both hormone binding to R~ and GTP binding to G, at the Gt,-subunit are required to induce activation of C to the cAMP-producing state. The active state of the system decays concomitantly with the hydrolysis of GTP to GDP and Pi at the G, regulatory site. Replenishment of G, with GTP and the continued presence of hormone at the receptor allow the system to regain its active cAMP-producing state. This type of 'on-off" cycle accounts well for the properties of hormone-dependent adenylyl cyclase (see Levitzki, 1988, for review) and actually for all processes of activation of effectors through G proteins. The hallmarks of adenylyl cyclase activation by receptors and G proteins can be summarized as follows: (1) The hormone-dependent adenylyl cyclases possess a slow (k~t ~_ 15 min -l) hormone-dependent GTPase activity. This activity has been demonstrated in fl-adrenoceptor-dependent cyclase (Cassel and Selinger, 1976), in glucagon-dependent cyclase, pancreoenzymin-dependent cyclase, and in prostaglandin El-dependent cyclase (Levitzki, 1982 for review) as well as in fl-adrenoceptor G, reconstituted systems (Lefkowitz and Carom 1988; Gilman, 1987). (2) Nonhydrolyzable GTP analogs such as GPPNHP and GTP?S, generate a constitutively active enzyme due to the complete blockade of the 'off" GTPase reaction (Gilman, 1987, for review). (3) Inhibition of the GTPase reaction by cholera toxin catalyzed ADP-ribosylation of ~q activates the cyclase due to the slow-down of the 'off" GTPase reaction (see Gilman, 1987; Levitzki, 1982, 1986, for reviews). (4) From independent measurements of the rate of enzyme activation by hormone and guanyl nucleotide (the 'on' reaction) and the decay of the cAMPproducing state to its basal state (the 'off" reaction), one can compute the fraction of the total pool of cyclase which is in the active state in the presence of GTP (Fig. 2). These measurements (Arad and
°,,
