G proteins activate ATP-sensitive K+ channels by antagonizing ATP-dependent gating

To determine whether G proteins activate cardiac ATP-sensitive K+ (KATP) channels by regulating intracellular ATP (ATPi)-dependent gating, currents were measured in inside-out patches. When ATPi closed KATP channels, activators of endogenous G proteins, GTP (plus adenosine or acetylcholine), GTP gamma S, or AlF-4 stimulated channels, an effect prevented by GDP beta S. In the absence of ATPi, G protein activators were ineffective. Intracellular nucleoside diphosphates restored KATP channel openings after the "rundown" of spontaneous activity. Only when ATPi suppressed nucleoside diphosphate-induced openings, GTP gamma S or AlF-4 enhanced KATP channel activity. Active forms of exogenous G protein subunits (G alpha i-1, G alpha i-2, or G alpha o) activated only KATP channels closed by ATPi. G proteins stimulate cardiac KATP channels apparently by antagonizing ATPi-dependent inhibitory gating. Regulation of ligand-dependent gating represents a distinct type of G protein modulation of ion channels.

Possible involvement of a pertussis toxin-sensitive GTP-binding protein in protein transport into nuclei isolated from rat liver

Nuclear protein transport was inhibited in permeabilized HeLa cells which had been prepared after culture with pertussis toxin, suggesting that the pertussis toxin-sensitive protein(s) might be involved in the nuclear protein transport. To investigate the mechanism whereby pertussis toxin inhibited the nuclear protein transport, an accumulation of proteins containing a nuclear localization signal sequence (NLS) into isolated rat liver nuclei was investigated. The NLS-containing protein accumulation required ATP and cytosolic proteins, and was temperature- and wheat germ agglutinin-sensitive as had been observed in permeabilized cells. Non-hydrolyzable GTP analogues, such as guanosine 5'-(gamma-thio)triphosphate and guanosine 5'-(beta, gamma-methylene)triphosphate, but not ATP analogues, inhibited the NLS-containing protein accumulation in the isolated nuclei. The NLS-containing protein accumulation was also inhibited by prior treatment of the nuclei with pertussis toxin plus NAD, and the effect of pertussis toxin was blocked when guanosine 5'-(gamma-thio)triphosphate was simultaneously added during the pretreatment with pertussis toxin. The inhibition induced by pertussis toxin and the blockage by guanosine 5'-(gamma-thio)triphosphate were well correlated to ADP-ribosylation of 40-kDa protein in nuclear fraction. These results suggested that the nuclear pertussis toxin-sensitive GTP-binding protein is involved in a pathway of nuclear protein transport.

Association of the beta gamma subunits of trimeric GTP-binding proteins with 90-kDa heat shock protein, hsp90

GTP-binding proteins (G proteins), predominantly located at the inner surface of the plasma membranes of mammalian cells, dissociate into their constituent alpha and beta gamma subunits upon stimulation of G protein-coupled receptors by agonists. In the present studies, cytoplasmic proteins which might have an affinity for the dissociated beta gamma subunits were investigated by means of beta gamma subunit-immobilized affinity-column (beta gamma-immobilized column) chromatography. When soluble fractions obtained from various materials including rat liver, bovine brain, and HL-60 cells were applied to a beta gamma-immobilized column, some proteins were specifically eluted from the column with high-salt and detergent-containing solutions. One of the beta gamma subunit-binding proteins, of which the molecular weight was approximately 93,000 on SDS-PAGE, appeared to be commonly present in all tissues tested. The 93-kDa beta gamma-binding protein was identified as 90-kDa heat shock protein, hsp90, based on the findings of its partial amino acid sequences and its immunoreactivity to a monoclonal anti-hsp90 antibody. The brain hsp90 inhibited beta gamma-supported pertussis toxin-catalyzed ADP-ribosylation of alpha subunits. The hsp90 was also capable of binding to beta gamma subunits which had been reconstituted into phospholipid vesicles. The binding of hsp90 to beta gamma subunits was inhibited by the addition of GDP-bound alpha subunits, but not by GTP gamma S-bound ones. These results suggested that hsp90 could associate functionally with free beta gamma subunits dissociated from trimeric G proteins in vitro.

Pertussis toxin-catalyzed ADP-ribosylation of GTP-binding proteins with digoxigenin-conjugated NAD. Identification of the proteins in plasma membranes and nuclei

ADP-ribose moiety containing digoxigenin was transferred by pertussis toxin (IAP) to the alpha subunit of Gi (Gi alpha) from digoxigenin-conjugated NAD (DIG-NAD) in a beta gamma subunit-dependent manner. ADP-ribosylation of Gi alpha with DIG-NAD plus IAP was inhibited by native NAD. These results indicate that non-radiolabeled DIG-NAD also serves as the substrate for IAP-catalyzed ADP-ribosylation of G proteins. Using DIG-NAD and fluorescein isothiocyanate-labeled anti-digoxigenin antibody, IAP-sensitive G protein(s) was found to be exist in nuclei as well as plasma membranes of rat liver and HeLa cells. Thus, DIG-NAD is useful to identify pertussis toxin-substrate G proteins.

Opposite coupling of prostaglandin E receptor EP3C with Gs and G(o). Stimulation of Gs and inhibition of G(o)

We recently identified four isoforms of bovine prostaglandin E receptor EP3 subtype, which are coupled to different signaling pathways; EP3A is coupled to inhibition of adenylate cyclase, while EP3B and EP3C are coupled to its stimulation and EP3D is coupled to phosphatidylinositol turnover, in addition to the adenylate cyclase system (Namba, T., Sugimoto, Y., Negishi, M., Irie, A., Ushikubi, F., Kakizuka, Ito, S., A., Ichikawa, A., and Narumiya, S. (1993) Nature 365, 166-170). We examined here the identity of coupled G proteins and their regulation by one of the isoforms, EP3C, in the membranes of EP3C cDNA-transfected Chinese hamster ovary cells. M&B 28767, an EP3 agonist, stimulated the GTPase activity in the pertussis toxin (PT)-treated cell membrane, but inhibited it in the cholera toxin (CT)-treated cell membrane, while the agonist neither stimulated nor inhibited it in the both PT- and CT-treated cell membrane. In the PT- and CT-treated cell membrane reconstituted with various G proteins, M&B 28767 inhibited the GTPase activity of G(o), but stimulated that of Gs. On the other hand, M&B 28767 did not affect the GTPase activity of Gi1, Gi2, or Gi3. M&B 28767 increased the apparent affinity of G(o) for GDP without any change in that for GTP, as assessed by displacement of [35S]GTP gamma S (guanosine 5'-O-(3-thiotriphosphate)) binding to G(o). In contrast, M&B 28767 increased the apparent affinity of Gs for GTP but decreased that for GDP. These results demonstrated that the EP3 receptor isoform is coupled to two different G proteins, and oppositely regulates their activities, inhibition of G(o), and stimulation of Gs.

GK* and brain G beta gamma activate muscarinic K+ channel through the same mechanism

A pertussis toxin-sensitive G protein (GK) links muscarinic cholinergic and A1-purinergic receptors with an inwardly rectifying K+ (KACh) channel in cardiac atrial cell membranes. Although the beta gamma subunits of pertussis toxin-sensitive G proteins (G beta gamma) have been reported to fully activate this channel, it is not known whether exogenously applied G beta gamma interacts with the KACh channel through the same mechanism as the active subunit of endogenous GK (GK*). We examined in inside-out patches the relationship between the concentration of GTP and KACh channel activity with or without preactivation of the channels by either GTP gamma S (guanosine 5'-3-O-(thio)triphosphate) or G beta gamma purified from bovine brain. In the control, KACh channels were activated by intracellular GTP (with acetylcholine in the pipette) in a positive cooperative manner (Hill coefficient approximately 2.5). As the channels were preactivated by GTP gamma S to progressively higher levels, the GTP channel activity relationship shifted more to the left, but the Hill coefficients of the curves remained the same. The same changes were observed when KACh channels were preactivated with brain G beta gamma. These results indicate that endogenous GK* and exogenous G beta gamma share a common molecular mechanism to activate the KACh channel.

Induction of starfish oocyte maturation by the beta gamma subunit of starfish G protein and possible existence of the subsequent effector in cytoplasm

beta gamma subunits of G proteins were purified from starfish oocytes, and their role in the induction of oocyte maturation by 1-methyladenine was investigated. When injected into starfish oocytes, the purified beta gamma subunit of the starfish G protein induced germinal vesicle breakdown (GVBD) faster than that of bovine brain G protein. Injection of the starfish beta gamma into cytoplasm near the germinal vesicle (GV) induced GVBD earlier than when injected into the GV or the cytoplasm near the plasma membrane. Fluorescent-labeled beta gamma was retained in the injected area even after GVBD. Injected beta gamma also induced the formation of maturation-promoting factor as well as an increase of histone H1 kinase activity. These results suggest that beta gamma dissociates from alpha-subunit by the stimulation of 1-methyladenine and interacts with a cytoplasmic effector, which results in formation of active cdc2 kinase.

Alterations of guanine nucleotide-binding proteins in post-mortem human brain in alcoholics

Qualitative and quantitative alterations of G proteins in membrane preparations from parietal and temporal cortex regions in post-mortem brains obtained from alcoholics and controls matched with respect to age and post-mortem delay were investigated by Western-blotting with polyclonal antibodies against specific G protein subunits and functional photoaffinity GTP binding. Quantitative immunoblotting showed that only Gs alpha (52 kDa species) in temporal cortex was significantly decreased (30%, P < 0.05) in alcoholics compared with controls. Moreover, ethanol-stimulated photoaffinity GTP labeling of Gs alpha and Gi/o alpha was decreased in alcoholics in both cortex regions. These results suggest that disturbances of G protein-mediated signal transduction may be involved in the pathophysiology of alcoholics.

NAD glycohydrolase specifically induced by retinoic acid in human leukemic HL-60 cells. Identification of the NAD glycohydrolase as leukocyte cell surface antigen CD38

Human leukemic HL-60 cells are differentiated into granulocytic cells by retinoic acid, and this differentiation is preceded by the induction of an ecto-enzyme of NAD glycohydrolase (NADase). The NADase specifically induced by retinoic acid appeared to be encoded by human leukocyte cell surface antigen CD38 as follows. 1) There was an early expression of CD38 mRNA, together with the induction of the NADase activity, in the retinoic acid-treated HL-60 cells. 2) The time course of the expression of CD38 antigen on the cell surface was well correlated with that of the induction of NADase activity. 3) The NADase activity solubilized from the differentiated HL-60 cell membrane could be immunoprecipitated with an anti-CD38 monoclonal antibody. 4) Introduction of the CD38 cDNA into Escherichia coli cells resulted in the expression of an NADase, the activity of which was inhibited by dithiothreitol. The NADase activity in the differentiated cells was also inhibited by the reducing reagent. These results clearly indicated that the dithiothreitol-sensitive NADase activity induced by retinoic acid in HL-60 cells is attributed to the molecule of human leukocyte cell surface antigen CD38, which contains cysteine-rich cytoplasmic domain within its molecule.

Botulinum ADP-ribosyltransferase C3 induces elevation of the vitelline coat of ascidian eggs

Botulinum exoenzyme C3 ADP-ribosylates a 23 kDa protein of unfertilized eggs of the ascidian, Halocynthia roretzi. Microinjection of C3 into the eggs induced elevation of the egg vitelline coat. Co-injection of heparin or EGTA with C3 inhibited the inducing effect of C3. The vitelline coat of eggs which had been previously co-injected with heparin and C3 was elevated by addition of calcium ionophore, but not by insemination. C3 also induced an increased formation of inositol 1,4,5-triphosphate (IP3) in ascidian egg membranes. Thus the ADP-ribosylation of small GTP-binding protein by C3 seems to be responsible for elevation of the vitelline coat of ascidian eggs through IP3 formation and intracellular calcium mobilization.

Production of monoclonal antibodies that inhibit ADP-ribosylation of small GTP-binding proteins catalyzed by Clostridium botulinum ADP-ribosyltransferase C3

Four monoclonal antibodies that inhibited ADP-ribosylation of 23 kDa protein(s) of ascidian eggs catalyzed by Clostridium botulinum ADP-ribosyltransferase C3 were produced. They also inhibited C3-catalyzed ADP-ribosylation of the 24 kDa protein of rat liver cytosol. By the immunoprecipitation technique, it was found that they recognized small GTP-binding proteins of ascidian eggs and mammalian brains, but did not interact with the rat brain activator of the ADP-ribosyltransferase reaction. The antibody can also immunoprecipitate recombinant Rho A irrespective as to whether the Rho A is the GDP-bound form or the GTPrS-bound form. Thus the antibodies are novel and useful tools in analyzing the physiological roles of the Rho family of GTP-binding proteins.

[Involvement of G protein in receptor-effector coupling]

G proteins consist of three subunits, alpha, beta and gamma, and bind with GTP or GDP to mediate the transformation and amplification of the signals between receptor on the cell membrane and the intracellular effector system (enzymes or ion channels), which produces various types of messenger. Bacterial toxins such as cholera and pertussis are widely used for research of signal transduction, owing to their ability for ADP-ribosylation of some types of G proteins to modify their functions. Network of signal transduction involving G proteins expands in cells of various organs, and relationship between G protein and receptors, or effectors, has been revealed day by day. Recent information of them are reviewed here.

Activation by G protein beta gamma subunits of beta-adrenergic and muscarinic receptor kinase

We have shown previously that GTP-binding regulatory protein (G protein) beta gamma subunits stimulate the agonist- or light-dependent phosphorylation of muscarinic acetylcholine receptors (mAChRs) and rhodopsin by a protein kinase partially purified from porcine brain (mAChR kinase) but not the phosphorylation of rhodopsin by rhodopsin kinase (Haga, K., and Haga, T. (1992) J. Biol. Chem. 267, 2222-2227). We report here that the mAChR kinase phosphorylates beta-adrenergic receptors (beta-ARs) purified from bovine lung in an agonist-dependent manner, and the phosphorylation is also stimulated by G protein beta gamma subunits. We also report that recombinant beta-adrenergic receptor kinase 1 (beta-ARK1) expressed in COS-7 cells phosphorylates mAChRs (human m2 subtype) and rhodopsin in an agonist- or light-dependent manner, respectively, and that this phosphorylation is stimulated by G protein beta gamma subunits. By contrast, the beta gamma subunits do not stimulate the phosphorylation of mAChRs or rhodopsin by a beta-ARK1 mutant lacking a part of the carboxyl-terminal region which is present in beta-ARKs but not in rhodopsin kinase. These results indicate that the beta-ARK1 is the same as or very similar to the mAChR kinase but is distinguished from the rhodopsin kinase with respect to activation by the beta gamma subunits and that the extra carboxyl-terminal sequence in beta-ARKs is required for the stimulation by the beta gamma subunits.

Somatostatin induces release of the alpha subunits of pertussis toxin-sensitive G proteins in native membranes and in intact GH4C1 rat pituitary cells

Incubation of GH4C1 rat pituitary cell membranes with the poorly hydrolyzable GTP analogue, GTP gamma S, produces a decrease in the pertussis toxin-catalyzed ADP-ribosylation of 40-kDa protein in the membrane pellet and the release of an alpha-like substrate from the membrane into the supernatant fraction; these effects do not occur with the inactive GDP analogue, GDP beta S. The resolved supernatant fraction from GTP gamma S-stimulated membranes is significantly activated to pertussis toxin-catalyzed [32P]ADP-ribosylation by the addition of purified beta gamma complex. Immunoblot analysis identifies the released pertussis toxin substrate as alpha subunits of Gi2, Gi3, and G(o) in the resolved supernatant. The physiological agonist, somatostatin, also stimulates the release of Gi2 and G(o) alpha subunits but not Gi3 from GH4C1 cell membranes in the presence of a low concentration of GTP gamma S (20 nM). The effects of somatostatin are inhibited by pretreatment of GH4C1 cells with pertussis toxin. Furthermore, the addition of somatostatin to intact GH4C1 cells decreases the level of Gi2 alpha subunits in the crude membrane whereas immunoblot analysis of the 274,000 x g supernatant (cytosolic fraction) clearly shows the presence of Gi2 alpha subunits. These data indicate that pertussis toxin-sensitive G proteins in GH4C1 cells dissociate into alpha subunits and beta gamma complex with the release of the alpha subunits from the membranes upon somatostatin activation.

Distribution of guanine nucleotide-binding protein in the brain of the reeler mutant mouse

The localization of a GTP-binding protein (G(o)) in the cerebellar and cerebral cortex and hippocampus of the normal and reeler mutant mouse was immunohistochemically examined using affinity-purified antibody raised against the alpha subunit of G(o). Although the general distribution pattern of G(o)-immunoreactive products in the brain of the normal mouse, i.e., abundant in the neuropil but absent from neuronal cell bodies, is also seen in the reeler brain, some differences are present, as described below. Strong G(o)-immunoreactive products are found in the molecular layer of the cerebellar cortex of the normal mouse. In the reeler cerebellum, in addition to the strong G(o)-immunoreactivity of the thin molecular layer, moderate G(o)-immunoreactivities are also found in the granular cell layer and the central cerebellar mass. G(o)-immunoreactive products are distributed throughout all layers of the cerebral cortex of the normal and reeler mouse. However, layer I of the normal cerebral cortex is more strongly stained with this antibody than the underlying layers, whereas the upper third of the reeler cerebral cortex is more strongly stained than the lower two-thirds. In the hippocampus of the normal mouse, G(o)-immunoreactive products are localized in the neuropil of the stratum oriens, stratum radiatum and stratum lacunosum-moleculare, but absent from the cell bodies of the pyramidal cells and their apical dendritic shafts. Such a distribution pattern of G(o)-immunoreactive products is also seen in the hippocampus of the reeler mouse, except that G(o)-immunonegative pyramidal cells split into 2 or 3 laminae.(ABSTRACT TRUNCATED AT 250 WORDS)

A protein kinase C inhibitor, staurosporine, activates phospholipase D via a pertussis toxin-sensitive GTP-binding protein in rabbit peritoneal neutrophils

In rabbit peritoneal neutrophils prelabeled with [3H] lyso platelet-activating factor, a protein kinase C inhibitor, staurosporine (> 1 microM), increased [3H]phosphatidylethanol ([3H]PEt) level in the presence of ethanol in a concentration- and time-dependent manner, providing evidence for staurosporine activation of phospholipase D (PLD). The staurosporine activation of the enzyme absolutely required both extracellular calcium and cytochalasin B, and was almost completely inhibited by pretreatment of the cells with pertussis toxin (IAP). In a reconstituted system where the purified Gi1 had been incorporated into phospholipid vesicles, staurosporine activated GTPase activity of Gi1 in a concentration-dependent fashion, with a maximal 4-5-fold effect. ADP-ribosylation by IAP of Gi1 in vesicles significantly suppressed the staurosporine activation. As with the GTPase activity of Gi1, GTPase activities of other purified IAP-sensitive G proteins, such as Gi2 and G(o), were significantly stimulated by staurosporine, but the cholera toxin substrate Gs was appreciably less sensitive to the staurosporine stimulation. The staurosporine activation of GTPase was also observed in rabbit neutrophil membranes from control cells, but not in membranes from IAP-treated neutrophils. From these results, we conclude that the staurosporine activation of PLD in rabbit neutrophils is attributed to the direct activation of an IAP-sensitive G protein in a similar manner to receptors occupied by agonists. By contrast, staurosporine failed to activate phosphoinositide-specific phospholipase C (PI-PLC) under the conditions in which it activated PLD, indicating that there exists a PLD activation pathway independent of PI-PLC. Furthermore, it was found that N-acetyl-beta-glucosaminidase release from the granules of intact neutrophils was evoked by staurosporine to almost the same extent as by fMLP (100 nM), but O2- generation was not affected. These results suggest a possibility that PLD pathway plays an important role in enzyme release, but is not sufficient for O2- generation, in rabbit peritoneal neutrophils.

The primary structure of the alpha subunit of a starfish guanosine-nucleotide-binding regulatory protein involved in 1-methyladenine-induced oocyte maturation

Starfish-oocyte maturation induced by 1-methyladenine (MeAde) was inhibited by microinjection of pertussis toxin (PTX). The inhibition appeared to result from PTX-catalyzed ADP-ribosylation of a 39-kDa guanosine-nucleotide-binding regulatory protein (G protein) in the oocyte. These results strongly support the hypothesis that the MeAde-induced signals operate via a membrane receptor and are carried by the PTX-sensitive G protein. When PTX-injected oocytes were treated with dithiothreitol, 85% of them reinitiated meiosis, suggesting that dithiothreitol did not act on the MeAde receptor. We constructed a cDNA library from the immature ovary of starfish, Asterina pectinifera, and screened it with the cDNA of the alpha subunit of an inhibitory rat G protein (Gi-2). A positive cDNA clone contained an open reading frame of 1062 bases which had 74% identity with the rat Gi-2 cDNA. The deduced amino acid sequence was 85% and 89% identical to rat Gi-2 and rat Gi-1, respectively. The alpha subunit of the G protein purified from cortices of starfish oocytes was digested by trypsin and the resulting four peptides were microsequenced. Comparison of these amino acid sequences with the predicted one indicated that the isolated cDNA clone encoded the alpha subunit of the PTX-sensitive G protein in oocytes. The C-terminal sequence, KNNLKDCGLF, was identical to that of Gi, suggesting that the cysteine residue is the site of ADP-ribosylation.

Properties of 1-methyladenine receptors in starfish oocyte membranes: involvement of pertussis toxin-sensitive GTP-binding protein in the receptor-mediated signal transduction

In response to a meiosis-inducing hormone, 1-methyladenine (1-MA), starfish oocytes undergo reinitiation of meiosis with germinal vesicle breakdown. The 1-MA-initiated signal is, however, inhibited by prior microinjection of pertussis toxin into the oocytes, suggesting that a guanine nucleotide-binding protein (G protein) serving as the substrate of pertussis toxin is involved in the 1-MA receptor-mediated signal. We thus investigated properties of 1-MA receptors by means of binding of the radiolabeled ligand to the oocyte membranes. There were apparently two forms of 1-MA receptors with high and low affinities in the membranes. The high-affinity form was converted into the low-affinity one in the presence of a non-hydrolyzable analogue of GTP. A 39-kDa protein, which had been identified as the alpha-subunit of the major substrate G protein for pertussis toxin, was also ADP-ribosylated by cholera toxin only when 1-MA was added to the membranes. The ADP-ribosylated 39-kDa alpha-subunit could be immunoprecipitated with antibodies raised against the carboxy-terminal site of mammalian inhibitory G-alpha. These results indicate that 1-MA receptors are functionally coupled with the 39-kDa pertussis toxin-substrate G protein in starfish oocyte membranes.

Activation of nucleoside diphosphate kinase by mastoparan, a peptide isolated from wasp venom

We have previously reported that GDP-bound alpha beta gamma-trimeric GTP-binding (G) proteins can be converted into the active GTP-bound form with nucleoside diphosphate (NDP) kinase and ATP, although its exact activation mechanism still remains to be resolved. In the present study, we investigated whether NDP kinase activity was modified by mastoparan, a wasp venom peptide that is known to activate G proteins as an agonist-receptor complex. The activity of NDP kinase measured by the formation of GTP from ATP and GDP was markedly stimulated, when the kinase was incubated with mastoparan. The concentration of mastoparan required for the activation was much lower than that observed for the peptide-induced activation of G proteins under similar assay conditions. There was also an increase in the phosphorylated intermediate of NDP kinase as well as the catalytic activity upon its incubation with mastoparan. These results suggest that mastoparan not only activates G proteins directly via guanine nucleotide exchange reaction but also stimulates NDP kinase activity.

On the mechanism of G protein beta gamma subunit activation of the muscarinic K+ channel in guinea pig atrial cell membrane. Comparison with the ATP-sensitive K+ channel

The mechanism of G protein beta gamma subunit (G beta gamma)-induced activation of the muscarinic K+ channel (KACh) in the guinea pig atrial cell membrane was examined using the inside-out patch clamp technique. G beta gamma and GTP-gamma S-bound alpha subunits (G alpha *'s) of pertussis toxin (PT)-sensitive G proteins were purified from bovine brain. Either in the presence or absence of Mg2+, G beta gamma activated the KACh channel in a concentration-dependent fashion. 10 nM G beta gamma almost fully activated the channel in 132 of 134 patches (98.5%). The G beta gamma-induced maximal channel activity was equivalent to or sometimes larger than the GTP-gamma S-induced one. Half-maximal activation occurred at approximately 6 nM G beta gamma. Detergent (CHAPS) and boiled G beta gamma preparation could not activate the KACh channel. G beta gamma suspended by Lubrol PX instead of CHAPS also activated the channel. Even when G beta gamma was pretreated in Mg(2+)-free EDTA internal solution containing GDP analogues (24-48 h) to inactivate possibly contaminating G i alpha *'s, the G beta gamma activated the channel. Furthermore, G beta gamma preincubated with excessive GDP-bound G o alpha did not activate the channel. These results indicate that G beta gamma itself, but neither the detergent CHAPS nor contaminating G i alpha *, activates the KACh channel. Three different kinds of G i alpha * at 10 pM-10 nM could weakly activate the KACh channel. However, they were effective only in 40 of 124 patches (32.2%) and their maximal channel activation was approximately 20% of that induced by GTP-gamma S or G beta gamma. Thus, G i alpha * activation of the KACh channel may not be significant. On the other hand, G i alpha *'s effectively activated the ATP-sensitive K+ channel (KATP) in the ventricular cell membrane when the KATP channel was maintained phosphorylated by the internal solution containing 100 microM Mg.ATP. G beta gamma inhibited adenosine or mACh receptor-mediated, intracellular GTP-induced activation of the KATP channel. G i alpha *'s also activated the phosphorylated KATP channel in the atrial cell membrane, but did not affect the background KACh channel. G beta gamma subsequently applied to the same patch caused prominent KACh channel activation. The above results may indicate two distinct regulatory systems of cardiac K+ channels by PT-sensitive G proteins: G i alpha activation of the KATP channel and G beta gamma activation of the KACh channel.

Direct activation of purified Go-type GTP binding protein by tricyclic antidepressants

A growing body of evidence suggests that tricyclic antidepressant agents (TCAs) interact with GTP binding proteins (G proteins). We have investigated if TCAs directly alter the function of the purified Go protein which is specifically expressed in neuronal tissue. Several TCAs markedly enhanced the GTPase activity of Go protein in a pertussis toxin-susceptible manner, whereas MAO-inhibitor and anxiolytic agent did not. This enhancing effect of TCAs on Go function may be due to an increase in the GDP-GTP exchange reaction occurring on Go. Thus, it is very likely that TCAs can modify various signal transduction by directly interacting with G proteins in brain cells.

Molecular heterogeneity of the beta gamma-subunits of GTP-binding proteins in bovine brain membranes

The guanine nucleotide-binding proteins (G proteins) are heterotrimers composed of alpha-, beta-, and gamma-subunits, and each of the constituent subunits has been reported to exhibit a molecular heterogeneity. The beta- and gamma-subunits form a functional unit that does not separate under physiological conditions and interact with various alpha-subunits that appear to mainly regulate specific effectors. We thus purified the beta gamma-complex of G proteins from bovine brain membranes and found that there were chromatographically multiple forms of beta gamma-subunits which could be reassociated with various alpha-subunits. The major findings observed with the purified proteins were summarized as follows. (a) The constituent beta gamma-subunits in the brain membrane G proteins appeared to be divided into two groups in their elution profiles from a hydrophobic column. (b) Each of the two groups contained at least five different components of beta gamma-subunits upon analyzing by a high-resolution, anion-exchange column. (c) Distribution of the heterogeneous beta gamma-subunits was not identical among various trimeric G proteins such as Gi, G0, and Gs. (d) The heterogeneous beta gamma-components were able to interact with a specific alpha-subunit resulting in the alpha beta gamma-trimer that served as the substrate of pertussis toxin-catalyzed ADP-ribosylation. (e) However, the apparent abilities of some beta gamma-subunits to support the toxin-induced modification were significantly different in a special comparison between the two beta gamma-groups that were eluted from the hydrophobic column. These results indicated that there were multiple forms of beta gamma-subunits associating with the specific alpha-subunit of a trimeric G protein and that some of those had different affinities for various alpha-subunits in terms of their tight associations. A possible role of the heterogeneity in beta gamma-subunits is also discussed in terms of G protein-mediated signal transductions.

A GTP-binding protein in rat liver nuclei serving as the specific substrate of pertussis toxin-catalyzed ADP-ribosylation

The ADP-ribosyl moiety of NAD was transferred to a 40-kDa protein when rat liver nuclei were incubated with pertussis toxin. The 40-kDa substrate in the nuclei displayed unique properties as follows, some of which were apparently distinct from those observed with the toxin-substrate GTP-binding protein (Gi) in the liver plasma membranes. 1) The nuclear 40-kDa protein was recognized with antibodies reacting with the alpha-subunits (alpha i-1 and alpha i-2) of Gi, but not with anti-Go-alpha-subunit antibody. 2) The nuclear protein had a higher mobility than alpha-subunit of the plasma membrane-bound Gi upon electrophoresis with a urea/sodium dodecyl sulfate-containing polyacrylamide gel. 3) The nuclear protein was not extracted from the nuclei with 1% Triton X-100, whereas Gi was easily solubilized from the plasma membranes. 4) There was a beta gamma-subunit-like activity in the nuclei, which was assayed by an ability to support pertussis toxin-catalyzed ADP-ribosylation of a purified alpha-subunit of Gi. Moreover, a 36-kDa protein in the nuclei was recognized with antibody raised against purified beta-subunits of Gi. 5) Pertussis toxin-induced ADP-ribosylation of the nuclear protein was selectively inhibited by the addition of a nonhydrolyzable GTP analogue, and its inhibitory action was competitively blocked by the simultaneous addition of GDP or its analogues, as had been observed with plasma membrane-bound Gi. It thus appeared that a novel form of alpha beta gamma-trimeric GTP-binding protein serving as the substrate of pertussis toxin was present in rat liver nuclei. In order to examine a possible role of the nuclear GTP-binding protein, rats were injected with carbon tetrachloride, a necrosis inducer of hepatocytes. There was a marked increase in the nuclear substrate activity from 3-6 days after the injection, without a significant change in the activity of Gi in the plasma membranes. The time course of the increase corresponded with a recovering stage from the hepatocyte necrosis. These results suggested that the nuclear GTP-binding protein found in the present study might be involved at some stages in the hepatocyte growth.

Modification of the function of pertussis toxin substrate GTP-binding protein by cholera toxin-catalyzed ADP-ribosylation

The alpha-subunit of Gi-2, in addition to that of Gs (GTP-binding proteins involved in adenylate cyclase inhibition and stimulation, respectively) was ADP-ribosylated by cholera toxin in HL-60 cell membranes when a chemotactic receptor was stimulated by formyl-Met-Leu-Phe (fMLP), and the sites modified by cholera and pertussis toxins on the alpha-subunit of Gi-2 were different (Iiri, T., Tohkin, M., Morishima, N., Ohoka, Y., Ui, M., and Katada, T. (1989) J. Biol. Chem. 264, 21394-21400). In order to investigate how the functions of Gi-2 were modified by cholera toxin, the ADP-ribosylated and unmodified proteins were purified from HL-60 cell membranes that had been incubated in the presence and absence of cholera toxin, respectively. The modified Gi-2 displayed unique properties as follows. 1) The ADP-ribosylated alpha-subunit had a more acidic pI than the unmodified one, leading to a partial resolution of the modified Gir2 trimer from the unmodified protein by an anion column chromatography. 2) When the purified proteins were incubated with [gamma-32P]GTP, the radioactivity was more greatly retained in the modified Gi-2 than in the unmodified protein. 3) The actual catalytic rate (kcat) of GTP hydrolysis was, indeed, markedly inhibited by cholera toxin-induced modification. 4) There was an increase in the apparent affinity of Gi-2 for GDP by cholera toxin-induced modification. 5) The modified Gi-2 exhibited a low substrate activity for pertussis toxin-catalyzed ADP-ribosylation. 6) A high-affinity fMLP binding to HL-60 cell membranes was more effectively reconstituted with the ADP-ribosylated Gi-2 than with the unmodified protein. These results suggested that the agonist-fMLP receptor complex was effectively coupled with the ADP-ribosylated Gi-2, resulting in the GTP-bound form, and that the hydrolysis of GTP on the modified alpha-subunit was selectively attenuated. Thus, cholera toxin ADP-ribosylated Gi-2 appeared to be not only a less sensitive pertussis toxin substrate but also an efficient signal transducer between receptors and effectors.

Functional modification by cholera-toxin-catalyzed ADP-ribosylation of a guanine-nucleotide-binding regulatory protein serving as the substrate of pertussis toxin

The alpha subunits of Gi (Gi alpha) and Gs (guanine-nucleotide-binding proteins involved in adenylate cyclase inhibition and stimulation, respectively) was ADP-ribosylated by cholera toxin in differentiated HL-60 cell membranes upon stimulation of chemotactic receptors by fMLF (fM, N-formylmethionine). The ADP-ribosylation site of Gi alpha modified by cholera toxin appeared to be different from that modified by pertussis toxin [Iiri, T., Tohkin, M., Morishima, N., Ohoka, Y., Ui, M. & Katada, T. (1989) J. Biol. Chem. 264, 21,394-21,400]. This allowed us to investigate how the two types of ADP-ribosylation influence the function of the signal-coupling protein. The major findings observed in HL-60 cell membranes, where the same Gi alpha molecule was ADP-ribosylated by treatment of the membranes with either toxin, are summarized as follows. (a) More fMLF bound with a high affinity to cholera-toxin-treated membranes than to the control membranes. The high-affinity binding was, however, not observed in pertussis-toxin-treated membranes. (b) Although fMLF stimulated guanine nucleotide binding and GTPase activity in control membranes, stimulation was almost completely abolished in pertussis-toxin-treated membranes. In contrast, fMLF-dependent stimulation of GTPase activity, but not that of guanine nucleotide binding was attenuated in cholera-toxin-treated membranes. (c) Gi alpha, once modified by cholera toxin, still served as a substrate of pertussis-toxin-catalyzed ADP-ribosylation; however, the ADP-ribosylation rate of modified Gi was much lower than that of intact Gi. These results suggested that Gi ADP-ribosylated by cholera toxin was effectively capable of coupling with fMLF receptors, resulting in formation of high-affinity fMLF receptors, and that hydrolysis of GTP bound to the alpha subunit was selectively impaired by its ADP-ribosylation by cholera toxin. Thus, unlike the ADP-ribosylation of Gi by pertussis toxin, cholera-toxin-induced modification would be of great advantage to the interaction of Gi with receptors and effectors that are regulated by the signal-coupling protein. This type of modification might also be a candidate for unidentified G proteins which were less sensitive to pertussis toxin and appeared to be involved in some signal-transduction systems.

Purification and characterization of a GTP-binding protein serving as pertussis toxin substrate in starfish oocytes

In response to a meiosis-inducing hormone, 1-methyladenine (1-MA), starfish oocytes undergo reinitiation of meiosis with germinal vesicle breakdown. The 1-MA-initiated signal is, however, inhibited by prior microinjection of pertussis toxin into the oocytes (Shilling, F., Chiba, K., Hoshi, M., Kishimoto, T., and Jaffe, L.A. (1989) Dev. Biol. 133, 605-608), suggesting that a pertussis-toxin-sensitive guanine-nucleotide-binding protein (G protein) is involved in the 1-MA-induced signal transduction. Based on these findings, we purified a G protein serving as the substrate of pertussis toxin from the plasma membranes of starfish oocytes. The purified G protein had an alpha beta gamma-trimeric structure consisting of 39-kDa alpha, 37-kDa beta, and 8-kDa gamma subunits. The 39-kDa alpha subunit contained a site for ADP-ribosylation catalyzed by pertussis toxin. The alpha subunit was also recognized by antibodies specific for a common GTP-binding site of many mammalian alpha subunits or a carboxy-terminal ADP-ribosylation site of mammalian inhibitory G-alpha. An antibody raised against mammalian 36-/35-kDa beta subunits strongly reacted with the 37-kDa beta subunit of starfish G protein. The purified starfish G protein had a GTP-binding activity with a high affinity and displayed a low GTPase activity. The activity of the G protein serving as the substrate for pertussis-toxin-catalyzed ADP-ribosylation was inhibited by its association with a non-hydrolyzable GTP analogue. Thus, the starfish G protein appeared to be similar to mammalian G proteins at least in terms of its structure and properties of nucleotide binding and the pertussis toxin substrate. A possible role of the starfish G protein is also discussed in the signal transduction between 1-MA receptors and reinitiation of meiosis with germinal vesicle breakdown.

On the mechanism of basal and agonist-induced activation of the G protein-gated muscarinic K+ channel in atrial myocytes of guinea pig heart

Using the patch clamp technique, we examined the agonist-free, basal interaction between the muscarinic acetylcholine (m-ACh) receptor and the G protein (GK)-gated muscarinic K+ channel (IK.ACh), and the modification of this interaction by ACh binding to the receptor in single atrial myocytes of guinea pig heart. In the whole cell clamp mode, guanosine-5'-O-(3-thiotriphosphate) (GTP-gamma S) gradually increased the IK.ACh current in the absence of agonists (e.g., acetylcholine). This increase was inhibited in cells that were pretreated with islet-activating protein (IAP, pertussis toxin) or N-ethylmaleimide (NEM). In inside-out patches, even in the absence of agonists, intracellular GTP caused openings of IK.ACh in a concentration-dependent manner in approximately 80% of the patches. Channel activation by GTP in the absence of agonist was much less than that caused by GTP-gamma S. The agonist-independent, GTP-induced activation of IK.ACh was inhibited by the A promoter of IAP (with nicotinamide adenine dinucleotide) or NEM. As the ACh concentration was increased, the GTP-induced maximal open probability of IK.ACh was increased and the GTP concentration for the half-maximal activation of IK.ACh was decreased. Intracellular GDP inhibited the GTP-induced openings of IK.ACh in a concentration-dependent fashion. The half-inhibition of IK.ACh openings occurred at a much lower concentration of GDP in the absence of agonists than in the presence of ACh. From these results, we concluded (a) that the interaction between the m-ACh receptor and GK is essential for basal stimulation of IK.ACh, and (b) that ACh binding to the receptor accelerates the turnover of GK and increases GK's affinity to GTP analogues over GDP.

Amino acid sequence determination of the novel forms of Go alpha purified from bovine brain membranes

Previously we have reported that there are at least four different forms of Go alpha in bovine brain membranes which can be distinguished by their elution profiles from Mono Q column and their immunological reactivities. The four alpha-subunits are referred to as alpha o1, alpha o2, alpha o3 and alpha o4 in their elution orders from the column. Partial amino acid sequences of the purified alpha o1 and alpha o2 were determined and compared with the predicted sequences of two classes of Go alpha cDNAs, termed Go alpha-1 and Go alpha-2. There were at least two unique fragments corresponding with the predicted amino acid sequence of the Go alpha-2 cDNA but different from that of the Go alpha-1 cDNA upon tryptic digestion of alpha o1- or alpha o2-subunit. The alpha o3- and alpha o4-subunits, but not alpha o1-and alpha o2-subunits, were recognized by an antibody raised against a unique amino acid sequence predicted from Go alpha-1 cDNA. These results suggest that alpha o1,2 subunits and alpha o3,4 subunits are encoded by Go alpha-2 cDNA and Go alpha-1 cDNA, respectively.

Direct activation of GTP-binding proteins by venom peptides that contain cationic clusters within their alpha-helical structures

Direct interactions of venom peptides that contained a cysteine-stabilized alpha-helical motif within their internal molecules with alpha beta gamma-trimeric GTP-binding proteins (G proteins) were studied in reconstituted phospholipid vesicles. Mast cell-degranulating (MCD) peptide stimulated the steady-state rate of GTP hydrolysis catalyzed by the reconstituted G proteins. Synthetic D-MCD peptide, the optical isomer of MCD peptide, was also effective in the activation of G proteins as L-MCD peptide. The stimulations by L- and D-peptides were both abolished in G proteins that had been ADP-ribosylated by pertussis toxin. Charybdotoxin also stimulated, though slightly, the GTPase activity of G proteins. Such a stimulation was, however, not observed upon the incubation of G proteins with other venom peptides such as apamin, sarafotoxin and endothelin. Thus, in comparison of the amino acid sequences of their venom peptides, the extent of the activation of G proteins appeared to be correlated with the number of basic amino acid residues around the alpha-helix. These results suggest that cationic clusters at one side of the alpha-helical surface are more important in the direct activation of G proteins than a specific, alpha-helical structure.

Characterization of cytosolic pertussis toxin-sensitive GTP-binding protein in mastocytoma P-815 cells

We have characterized a soluble pertussis toxin (PT)-sensitive GTP-binding protein (G-protein) present in mouse mastocytoma P-815 cells. 65% of total ADP-ribosylation of PT substrate having a molecular mass of 40 kDa on SDS-polyacrylamide gel electrophoresis in cell homogenate was detected in the supernatant after centrifugation at 100,000 x g for 90 min. [32P]ADP-ribosylation of cytosolic PT substrate was significantly enhanced on the addition of exogenous beta gamma complex. The molecular mass of the cytosolic PT substrate was estimated to be about 80 kDa on an Ultrogel AcA 44 column, but the beta gamma complex was not detected in the cytosol by using the anti-beta gamma complex antibody. Furthermore, the cytosolic PT substrate was found to have some unique properties: [35S]GTP gamma S binding was not inhibited by GDP and [32P]ADP-ribosylation was not affected by GTP gamma S treatment. Only after the cytosolic PT substrate had been mixed with exogenous beta gamma complex, did it copurify with exogenous beta gamma complex by several column chromatographies including an Octyl-Sepharose CL-4B column. The PT substrate was identified as Gi2 alpha by Western blot analysis and peptide mapping with S. aureus V8 protease. These results suggest that Gi2 alpha without beta gamma complex exists with an apparent molecular mass of about 80 kDa in the cytosolic fraction of P-815 cells.

Identification of a botulinum C3-like enzyme in bovine brain that catalyzes ADP-ribosylation of GTP-binding proteins

A novel enzyme activity was found in bovine brain cytosol that transfers the ADP-ribosyl moiety of NAD to proteins with Mr values of 22,000 and 25,000. The substrates were the same GTP-binding proteins serving as the substrate of an ADP-ribosyltransferase C3 which was produced by a type C strain of Clostridium botulinum. The brain enzyme was partially purified from the cytosol and had a molecular mass of approximately 20,000 on a gel filtration column. The brain endogenous enzyme displayed unique properties similar to those observed with botulinum C3 enzyme. The enzyme activity was markedly stimulated by a protein factor that had been initially found in the cytosol as an activator for botulinum C3-catalyzed ADP-ribosylation (Ohtsuka, T., Nagata, K., Iiri, T., Nozawa, Y., Ueno, K., Ui, M., and Katada, T. (1989) J. Biol. Chem. 264, 15000-15005). The activity of the brain enzyme was also affected by certain types of detergents or phospholipids. The substrate of the brain enzyme was specific for GTP-binding proteins serving as the substrate of botulinum C3 enzyme; the alpha-subunits of trimeric GTP-binding proteins which served as the substrate of cholera or pertussis toxin were not ADP-ribosylated by the endogenous enzyme. Thus, this is the first report showing an endogenous enzyme in mammalian cells that catalyzes ADP-ribosylation of small molecular weight GTP-binding proteins.

Purification and characterization of a new GTP-binding protein of Mr 24,000 in bovine brain membranes

A GTP-binding protein with an Mr of 24,000 was purified from a cholate extract of bovine brain membranes in addition to the previously reported alpha beta gamma-trimeric GTP-binding proteins (G proteins). Partial amino acid sequence analysis of the purified 24-kDa protein revealed that it was not identical to any of the low Mr GTP-binding proteins already reported, but similar to the rac-gene products serving as the substrate of an ADP-ribosyltransferase (C3) purified from the culture medium of Clostridium botulinum type C. However, the 24-kDa protein was not ADP-ribosylated by the botulinum C3 enzyme. The 24-kDa protein was purified as a nucleotide-free form and characterized by the following unique properties distinct from those of alpha beta gamma-trimeric G proteins. (1) Mg2+ was essentially required for nucleotide binding to the 24-kDa protein; there was a progressive increase in its binding affinity for nucleotides as the concentration of the divalent cation was increased. (2) Nucleotides previously bound to the 24-kDa protein were rapidly dissociated from the protein in Mg(2+)-free medium, in accord with the fact that the protein was indeed purified as a nucleotide-free form with Mg(2+)-free solutions. (3) The 24-kDa protein apparently exhibited much lower GTPase activity than do alpha beta gamma-trimeric G proteins because the product GDP was released from the 24-kDa protein in exchange for the substrate GTP only at a very low rate. Based on these findings, a possible role of the 24-kDa protein in cellular signalling is discussed in comparison with well characterized alpha beta gamma-trimeric G proteins.

Interaction of guanine-nucleotide-binding regulatory proteins with chemotactic peptide receptors in differentiated human leukemic HL-60 cells

Human leukemic HL-60 cells were differentiated into neutrophil-like cells by treatment with dimethylsulfoxide (Me2SO) or N6,O2'-dibutyryladenosine 3',5'-phosphate (Bt2cAMP), and membrane fractions were prepared from the differentiated cells. Receptors for fMLF (fM,N-formylmethionine) and guanine-nucleotide-binding regulatory proteins (G proteins) serving as the substrate for pertussis toxin (islet-activating protein; IAP) were extracted from cell membranes then reconstituted into phospholipid vesicles. The binding of fMLF to the reconstituted vesicles (or the membranes) was determined with 10 nM [3H] fMLF. In both cases, high-affinity binding to vesicle preparations from the Me2SO- and Bt2cAMP-induced cells was abolished following treatment with IAP, suggesting that fMLF receptors were functionally coupled to IAP-sensitive G proteins in each of the two vesicle types. However, the high-affinity fMLF binding was much higher in vesicle preparations originating from Bt2cAMP-induced cells than in those from Me2SO-induced cells, although the amount of IAP-substrate G protein reconstituted into the each phospholipid vesicles preparation was not significantly different from the other. The G proteins of the two differentiated cells were both identified as inhibitory forms (Gi-2) based on their electrophoretic mobilities and immunoblot analyses. When purified Gi-2 from rat brain was reconstituted into the two IAP-treated vesicles, high-affinity fMLF binding was restored in a similar manner in both. IAP-substrate G proteins partially purified from the two differentiated HL-60 cells were also effective in restoring high-affinity fMLF binding to the IAP-treated vesicles. However, a significant difference was observed that the reconstituted binding was higher with the G-protein-rich fraction from Bt2cAMP-induced cells than with that from Me2SO-induced cells, with each of the two IAP-treated vesicle types. These results suggest that the different high-affinity binding of fMLF observed in the two differentiated HL-60 cells are due to a difference in the property of endogenous G proteins rather than fMLF receptors, though the two G proteins are indistinguishable from each other in terms of the subtype of G protein, Gi-2.

Direct interactions of mastoparan and compound 48/80 with GTP-binding proteins

The effects of mastoparan and compound 48/80 on the activities of alpha beta gamma-trimeric GTP-binding proteins (G proteins) were studied with purified Go and Gi-1 which had been reconstituted into phospholipid vesicles. Pertussis toxin-catalyzed ADP-ribosylation of Go or Gi-1 was inhibited by mastoparan or compound 48/80, suggesting that the G proteins were dissociated into their constituent alpha- and beta gamma-subunits in the presence of these compounds. The steady-state rate of GTP hydrolysis catalyzed by Go or Gi-1 was stimulated by the two compounds. Both the stimulations were due to increases in the rate of the GDP-GTP exchange reaction occurring on the G proteins. However, the modes stimulation of the GTPase activity depended on the type of G protein used, and the stimulations caused by the two compounds were differently affected by pertussis toxin-catalyzed ADP-ribosylation of G proteins. Moreover, the mastoparan-induced stimulation of the GTPase activity was partially inhibited by compound 48/80. Thus, the two histamine secretagogues mastoparan and compound 48/80 appear to activate G proteins differently, though they interact with the signal-transducing proteins, at least partly, at a common binding site.

Conversion of GDP into GTP by nucleoside diphosphate kinase on the GTP-binding proteins

A direct interaction of alpha beta gamma trimeric GTP binding proteins (G proteins; G0 and Gs) with nucleoside diphosphate kinase (NDP kinase) was investigated with homogeneously purified proteins. There was a progressive release of 32Pi from [gamma-32P]ATP when GDP-bound G0 was incubated together with NDP kinase. The Pi release induced by the interaction of G0 with NDP kinase was not accompanied by the dissociation of GDP bound to the alpha-subunit of G0. This was a sharp contrast to G protein-catalyzed GTP hydrolysis observed with GTP as the substrate; the dissociation of bound GDP was essentially required for the following binding of the substrate, GTP, to be hydrolyzed. A kinetic analysis displayed different properties for the substrate of NDP kinase between free GDP and G protein-bound GDP. NDP kinase-dependent phosphorylation of GDP on G0 was indeed demonstrated with adenosine 5'-(3-O-thio)triphosphate as the phosphate donor; there was a formation of guanosine 5'-(3-O-thio)triphosphate-bound G0 from the ATP analogue. Moreover, purified Gs was readily ADP-ribosylated by cholera toxin in the presence of NDP kinase, ATP, and an ADP-ribosylation factor, also suggesting that the nucleotide form on Gs was certainly GTP. These results indicate that NDP kinase can transfer the gamma-phosphate of ATP directly to GDP bound to G proteins and that this phosphorylation results in the activation of the signal-coupling proteins. A possible role of the new activation mechanism of G proteins is discussed in comparison with the previously characterized GDP-GTP exchange pathway by the agonist-receptor complex.

Identification of sites for alkylation by N-ethylmaleimide and pertussis toxin-catalyzed ADP-ribosylation on GTP-binding proteins

An alpha beta gamma-trimeric GTP-binding protein (Go) serving as the substrate of pertussis toxin-(IAP) catalyzed ADP-ribosylation was purified from rat brain membranes. The constituent alpha-subunit (alpha o) was alkylated with N-ethylmaleimide (NEM), and the functionally important sulfhydryl groups were investigated. There were at least two cysteine residues highly reactive to NEM on the GDP-bound form of alpha o. These alkylations resulted in loss of its ability to be ADP-ribosylated by IAP and to associate with beta gamma, but leaving the GTP-binding site of alpha o intact. The reacted cysteine residues were identified by the sequencing of tryptic fragments of alpha o. One of the alkylation sites was Cys-351, which was four amino acid residues away from the carboxyl-terminus of the molecule. The Cys-351 was proven to be also a site for IAP-catalyzed ADP-ribosylation. Possible roles of cysteine residues on the alpha-subunit of Go are discussed in the functions of the signal transducing protein.

Possible interaction of alpha 1-adrenergic receptor with pertussis-toxin-sensitive guanine-nucleotide-binding regulatory proteins (G proteins) responsible for phospholipase C activation in rat liver plasma membranes

Islet-activating protein (IAP; pertussis toxin) was employed to test the hypothesis that IAP-sensitive GTP-binding regulatory proteins (G proteins) are coupled with alpha 1-adrenergic receptor in rat liver plasma membranes. The high-affinity state of the binding of alpha 2-adrenergic agonist, which is known to be coupled with IAP-sensitive G protein, was abolished in IAP-treated plasma membranes. IAP treatment of plasma membranes could also diminish the high-affinity state of the alpha 1-adrenergic receptor for the agonist. Restoration of the high-affinity state of the alpha 1-adrenergic receptor for the agonist occurred on reconstitution of the bovine brain IAP-sensitive G proteins. The alpha 1-adrenergic receptor agonist stimulated inositol triphosphate (InsP3) production from [3H]inositol-labeled liver plasma membranes in a concentration-dependent manner. IAP treatment also decreased alpha 1-adrenergic-agonist-induced InsP3 production but not completely. From these results, we concluded that there is a possibility that both IAP-sensitive and IAP-insensitive G proteins were involved in alpha 1-adrenergic-receptor-stimulated phospholipase C activation in rat liver plasma membranes.

Distinctive regulation of the functional linkage between the human cation-independent mannose 6-phosphate receptor and GTP-binding proteins by insulin-like growth factor II and mannose 6-phosphate

The rat insulin-like growth factor II (IGF-II) receptor develops transmembrane signaling functions by directly coupling to a guanine nucleotide-binding protein (G protein) having a 40-kDa alpha subunit, Gi-2, whereas recent studies have indicated that the IGF-II receptor is a molecule identical to the cation-independent mannose 6-phosphate receptor (CI-MPR), a receptor implicated in lysosomal enzyme sorting. In this study, by using vesicles reconstituted with the clonal human CI-MPR and G proteins, we indicated that the CI-MPR could stimulate guanosine 5'-O-(3-thiotriphosphate) (GTP gamma S) binding and GTPase activities of Gi proteins in response to IGF-II. The stimulatory effect of IGF-II on Gi-2 depended on the reconstituted amount of the CI-MPR; it could not be found in vesicles reconstituted with Gi-2 alone; and it was also observed on Gi-1 reconstituted with the CI-MPR in phospholipid vesicles. Of interest, such stimulatory effect was not reproduced by Man-6-P in CI-MPR vesicles reconstituted with either G protein. Furthermore, the affinity for Man-6-P-mediated beta-glucuronidase binding to several kinds of native cell membranes was not reduced by 100 microM GTP gamma S. Instead, however, Man-6-P dose-dependently inhibited IGF-II-induced Gi-2 activation with an IC50 of 6 microM in vesicles reconstituted with the CI-MPR and Gi-2. The action of 100 nM IGF-II was completely abolished by 1 mM Man-6-P. Such an inhibitory effect of Man-6-P was reproduced by 4000 times lower concentrations of beta-glucuronidase or similar concentrations of fructose 1-phosphate, but not by mannose or glucose 6-phosphate. These results indicate that the human CI-MPR has two distinct signaling functions that positively or negatively regulate the activity of Gi-2 in response to the binding of IGF-II or Man-6-P.

Molecular heterogeneity of the subclasses of islet-activating protein (pertussis toxin)-sensitive GTP-binding proteins in porcine thyroid tissue

From porcine thyroid cell membranes, we purified five GTP-binding proteins (G-proteins); Nos. 1 to 3 have 41-kDa alpha-subunits, and Nos. 4 and 5 have 40-kDa alpha-subunits. They were chromatographically (Mono Q) separable and served as specific substrates for islet-activating protein (pertussis toxin). G-proteins 1 and 2 were indistinguishable from porcine brain Gi1 with respect to three criteria, i.e., mobility on sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), pI of the ADP-ribosylated alpha-subunit, and immunoreactivity. G-protein 3 was identified as Gi3 by immunoreactivity. The SDS-PAGE and isoelectric focusing (IEF) analyses identified G-proteins 4 and 5 as being chromatographically heterogeneous subtypes of Gi2 in comparison with a pure porcine brain preparation. The IEF analysis also disclosed that each of the Gi1, Gi2, and Gi3 subspecies isolated in the present study has a minor component characterized by a slightly lower pI of its alpha-subunit. We conclude that porcine thyroid tissue contains at least Gi1, Gi2, and Gi3, and that each is made up of heterogeneous populations.

A simple structure encodes G protein-activating function of the IGF-II/mannose 6-phosphate receptor

The insulin-like growth factor-II/mannose 6-phosphate receptor (IGF-II/man6PR) can directly interact with and activate Gi-2, a GTP binding protein (G protein). We found that the segment of residues 2410-2423 in the human IGF-II/man6PR activates Gi-2 in a manner similar to G-coupled receptors. We observed a hierarchy of the segment action when tested on various G proteins, with an order of Gi-2 greater than Gi-1 approximately Gi-3 greater than Go. The segment had no effect on Gs or low molecular weight G proteins. The segment action depended on its primary structure and was potentiated when the segment was connected with a part of the receptor transmembrane region. Finally, the Gi-2-activating function of the human IGF-II/man6PR could be blocked by an antibody against the segment, indicating a critical role for this small region of the receptor.

Purification and characterization of five different alpha subunits of guanine-nucleotide-binding proteins in bovine brain membranes. Their physiological properties concerning the activities of adenylate cyclase and atrial muscarinic K+ channels

We have purified five different alpha subunits of guanine-nucleotide-binding proteins (G proteins) from bovine brain membranes as active forms bound to guanosine 5'-[gamma-thio]triphosphate (GTP[gamma S]). All the purified alpha subunits were interacted with beta gamma subunits and served as a substrate for pertussin-catalyzed ADP-ribosylation. Based on the findings of immunoblot analyses using specific antibodies raised against various alpha subunits of G proteins, three of them were identified as alpha i-1, alpha i-2 and alpha i-3, and the other two were classified into alpha o type. One of the alpha o-type proteins was the most abundant in the brain membranes (termed alpha o), and the other (alpha o2) appeared to differ from alpha o in its proteolytic digestion data. The physiological properties of these purified GTP[gamma S]-bound alpha subunits towards adenylate cyclase and atrial muscarinic K+ channels were studied. The nucleotide-bound forms of alpha i-1, alpha i-2, alpha i-3 and alpha o2 inhibited the adenylate cyclase activity of S49 cyc- membranes which had been reconstituted with GTP[gamma S]-treated Gs; this inhibition appeared to be mainly competitive with the activated Gs, alpha i-1 having the most potent inhibitory activity among them. GTP[gamma S]-bound alpha o, however, could not inhibit the Gs-stimulated activity at all. On the other hand, all the GTP[gamma S]-bound alpha subunits activated atrial muscarinic K+ channels, accompanied by a lag time, at picomolar concentrations. The beta gamma subunits resolved from G proteins also activated the K+ channels without a lag time at nanomolar concentration. The maximum activation by the beta gamma subunits appeared to be more potent than that by any of the alpha subunits. These results suggest that alpha and beta gamma subunits might activate the K+ channels by mechanisms different from each other.