Class I phosphoinositide 3-kinases (PI3Ks) regulate important cellular processes such as mitogenesis, apoptosis, and cytoskeletal functions. They include PI3K␣, -␤, and -␦ isoforms coupled to receptor tyrosine kinases and a PI3K␥ isoform activated by receptor-stimulated G proteins. This study examines the direct interaction of purified recombinant PI3K␥ catalytic subunit (p110␥) and G␤␥ complexes. When phosphatidylinositol was used as a substrate, G␤␥ stimulated p110␥ lipid kinase activity more than 60-fold (EC 50 , ϳ20 nM). Stimulation was inhibited by G␣ o -GDP or wortmannin in a concentration-dependent fashion. Stoichiometric binding of a monoclonal antibody to the putative pleckstrin homology domain of p110␥ did not affect G␤␥-mediated enzymatic stimulation, whereas incubation of G␤␥ with a synthetic peptide resembling a predicted G␤␥ effector domain of type 2 adenylyl cyclase selectively inhibited activation of p110␥. G␤␥ complexes bound to N-as well as C-terminal deletion mutants of p110␥. Correspondingly, these enzymatically inactive N-and C-terminal mutants inhibited G␤␥-mediated activation of wild type p110␥. Our data suggest that (i) p110␥ directly interacts with G␤␥, (ii) the pleckstrin homology domain is not the only region important for G␤␥-mediated activation of the lipid kinase, and (iii) G␤␥ binds to at least two contact sites of p110␥, one of which is close to or within the catalytic core of the enzyme.

We have previously reported that, in venous myocytes, Gbetagamma scavengers inhibit angiotensin AT1A receptor-induced stimulation of L-type Ca2+ channels (1). Here, we demonstrate that intracellular infusion of purified Gbetagamma complexes stimulates the L-type Ca2+ channel current in a concentration-dependent manner. Additional intracellular dialysis of GDP-bound inactive Galphao or of a peptide corresponding to the Gbetagamma binding region of the beta-adrenergic receptor kinase completely inhibited the Gbetagamma-induced stimulation of Ca2+ channel currents. The gating properties of the channel were not affected by intracellular application of Gbetagamma, suggesting that Gbetagamma increased the whole-cell calcium conductance. In addition, both the angiotensin AT1A receptor- and the Gbetagamma-induced stimulation of L-type Ca2+ channels were blocked by pretreatment of the cells with wortmannin, at nanomolar concentrations. Correspondingly, intracellular infusion of an enzymatically active purified recombinant Gbetagamma-sensitive phosphoinositide 3-kinase, PI3Kgamma, mimicked Gbetagamma-induced stimulation of Ca2+ channels. Both Gbetagamma- and PI3Kgamma-induced stimulations of Ca2+ channel currents were reduced by protein kinase C inhibitors suggesting that the Gbetagamma/PI3Kgamma-activated transduction pathway involves a protein kinase C. These results indicate for the first time that Gbetagamma dimers stimulate the vascular L-type Ca2+ channels through a Gbetagamma-sensitive PI3K.

Secretory vesicles store neurotransmitters that are released by exocytosis. Their membrane contains transporters responsible for transmitter loading that are driven by an electrochemical proton gradient across the vesicle membrane. We have now examined whether uptake of noradrenaline is regulated by heterotrimeric G proteins. In streptolysin O-permeabilized PC 12 cells, GTP-analogues and AlF 4 -inhibited noradrenaline uptake, an effect that was sensitive to treatment with pertussis toxin. Inhibition of uptake was prevented by Gαo-specific antibodies and mimicked by purified activated Gαo 2 . No effect was seen when Gαo 2 in its inactive GDP-bound form or purified activated Gαo 1 , Gαi 1 and Gαi 2 were tested. Down-regulation of uptake remained unchanged when exocytosis was inhibited by the light chain of tetanus toxin. Vesicular acidification was not affected whereas binding of [ 3 H]reserpine was reduced by GTPγS and Gαo 2 . These data suggest that the monoamine transporter rather than the vacuolar ATPase is affected. We conclude that catecholamine uptake is controlled by Gαo 2 , suggesting a novel function for heterotrimeric G proteins in the control of neurotransmitter storage.

We previously reported that, in the membranes of HL-60 cells during activation of G-proteins, a phosphate transfer reaction occurs which involves transient G-protein beta subunit (G beta) phosphorylation [Wieland, Nürnberg, Ulibarri, Kaldenberg-Stasch, Schultz and Jakobs (1993) J. Biol. Chem. 268, 18111-18118]. Here, the generality of this phenomenon is evaluated by studying membranes of various tissues obtained from different mammalian species. All membranes tested expressed at least G beta 1 and G beta 2 subunits. Cell membranes from bovine and porcine brain and liver, rat brain and human blood cells exhibited predominantly G beta 1 or both subtypes at roughly equal concentrations. In contrast, significantly more G beta 2 immunoreactivity was detected in membranes from human placenta. Bovine and porcine liver membranes exhibited weak, G beta-specific immunoreactive signals. Conversely, these membranes showed the highest levels of G beta phosphorylation after incubation with [gamma-32P]GTP or 35S-labelled guanosine 5'-[gamma-thio]triphosphate. Interestingly, G beta-specific phosphorylation of membranes from human erythrocytes and platelets was very weak. G beta phosphorylation was confirmed by immunoprecipitation with G beta-specific antibodies, and the target amino acid was identified as histidine. On SDS/PAGE, phosphorylated or thiophosphorylated G beta-proteins differed in their apparent molecular size from unmodified G beta-proteins. Moreover, phosphorylated G beta-proteins differed in a species-dependent fashion in their electrophoretic mobility. Solubilization of membrane proteins with detergent did not abolish G beta phosphorylation. In contrast, reconstituted purified Gi/Go proteins showed no G beta phosphorylation. From these experiments we conclude that: (i) G beta phosphorylation represents a general phenomenon occurring in the cells of various species to different degrees, (ii) phosphorylated G beta-proteins exhibit species-dependent diverse electrophoretic mobilities, and (iii) G beta phosphorylation requires a membrane-associated cofactor(s) which is lost during routine G-protein purification.

Synthesis and pharmacological properties of new potent direct activators of heterotrimeric G proteins are described. Compounds were synthesized from protected amino acids with alkylamines using coupling reagents (CDI, DCC, and EDC). Alkyl-substituted amino acid amides and their corresponding di- and triamines were subjected to structure-activity analysis. All compounds activated membrane-bound HL-60 GTPases in a pertussis toxin-sensitive fashion. This suggests a specific effect of compounds on the carboxy terminus of a defined subclass of heterotrimeric G proteins, i.e., members of the G alpha i subfamily. Elongation of the alkyl chain and increasing the number of amino groups enhanced the potency of compounds on HL-60 membrane-bound GTPase. N-(2,5-Diaminopentyl)dodecylamine (21) was selected to study its mode of action employing purified pertussis toxin-sensitive G proteins. It stimulated G alpha subunits by inducing the release of bound GDP. In contrast to receptors G beta gamma complexes were not required for 21-mediated activation of G alpha. Moderate isoform selectivity of its action was observed within a group of highly homologous members of the Gi subfamily with G alpha o1 being activated at lowest concentrations, whereas higher concentrations were necessary for the stimulation of G alpha i1 or transducin. We conclude that these compounds represent important tools for studying G protein-dependent cellular functions.