Multiple features of the p59fyn src homology 4 domain define a motif for immune-receptor tyrosine-based activation motif (ITAM) binding and for plasma membrane localization.

The src family tyrosine kinase p59fyn binds to a signaling motif contained in subunits of the TCR known as the immune-receptor tyrosine-based activation motif (ITAM). This is a specific property of p59fyn because two related src family kinases, p60src and p56lck, do not bind to ITAMs. In this study, we identify the residues of p59fyn that are required for binding to ITAMs. We previously demonstrated that the first 10 residues of p59fyn direct its association with the ITAM. Because this region of src family kinases also directs their fatty acylation and membrane association (Resh, M.D. 1993, Biochim. Biophys. Acta 1155:307-322; Resh, M.D. 1994. Cell. 76:411-413), we determined whether fatty acylation and membrane association of p59fyn correlates with its ability to bind ITAMs. Four residues (Gly2, Cys3, Lys7, and Lys9) were required for efficient binding of p59fyn to the TCR. Interestingly, the same four residues are present in p56lyn, the other src family tyrosine kinase known to bind to the ITAM, suggesting that this set of residues constitutes an ITAM recognition motif. These residues were also required for efficient fatty acylation (myristoylation at Gly2 and palmitoylation at Cys3), and plasma membrane targeting of p59fyn. Thus, the signals that direct p59fyn fatty acylation and plasma membrane targeting also direct its specific ability to bind to TCR proteins.

Structures of active conformations of Gi alpha 1 and the mechanism of GTP hydrolysis.

Mechanisms of guanosine triphosphate (GTP) hydrolysis by members of the G protein alpha subunit-p21ras superfamily of guanosine triphosphatases have been studied extensively but have not been well understood. High-resolution x-ray structures of the GTP gamma S and GDP.AlF4- complexes formed by the G protein Gi alpha 1 demonstrate specific roles in transition-state stabilization for two highly conserved residues. Glutamine204 (Gln61 in p21ras) stabilizes and orients the hydrolytic water in the trigonal-bipyramidal transition state. Arginine 178 stabilizes the negative charge at the equatorial oxygen atoms of the pentacoordinate phosphate intermediate. Conserved only in the G alpha family, this residue may account for the higher hydrolytic rate of G alpha proteins relative to those of the p21ras family members. The fold of Gi alpha 1 differs from that of the homologous Gt alpha subunit in the conformation of a helix-loop sequence located in the alpha-helical domain that is characteristic of these proteins; this site may participate in effector binding. The amino-terminal 33 residues are disordered in GTP gamma S-Gi alpha 1, suggesting a mechanism that may promote release of the beta gamma subunit complex when the alpha subunit is activated by GTP.

G-protein alpha subunit Gi(alpha)2 mediates erythropoietin signal transduction in human erythroid precursors.

Erythropoietin induces a dose-dependent increase in cytosolic calcium in human erythroblasts that is mediated by a voltage-independent Ca2+ channel. Inhibition of this response to erythropoietin by pertussis toxin suggests involvement of guanine nucleotide-binding regulatory proteins (G-proteins). The role of G-proteins in regulation of the erythropoietin-modulated Ca2+ channel was delineated here by microinjection of G-protein modulators or subunits into human erythroid precursors. This is the first report on the use of microinjection to study erythropoietin signal transduction in normal precursor cells. Fura-2 loaded day-10 burst-forming units-erythroid-derived erythroblasts were used for microinjection and free intracellular calcium concentration ([Ca(i)]) was measured with digital video imaging. BCECF (1,2',7'-bis(2-carboxyethyl)-5-(and -6-)-carboxyfluorescein) was included in microinjectate, and an increase in BCECF fluorescence was evidence of successful microinjection. Cells were microinjected with nonhydrolyzable analogues of GTP, GTPgammaS or GDPbetaS, which maintain the alpha subunit in an activated or inactivated state, respectively. [Ca(i)] increased significantly in a dose-dependent manner after microinjection of GTPgammaS. However, injection of GDPbetaS blocked the erythropoietin-induced calcium increase, providing direct evidence that activation of a G-protein is required. To delineate which G-protein subunits are involved, alpha or betagamma transducin subunits were purified and microinjected as a sink for betagamma or alpha subunits in the erythroblast, respectively. Transducin betagamma, but not alpha, subunits eliminated the calcium response to erythropoietin, demonstrating the primary role of the alpha subunit. Microinjected antibodies to Gi(alpha)2, but not Gi(alpha)1 or Gi(alpha)3, blocked the erythropoietin-stimulated [Ca(i)] rise, identifying Gi(alpha)2 as the subunit involved. This was confirmed by the ability of microinjected recombinant myristoylated Gi(alpha)2, but not Gi(alpha)1 or Gi(alpha)3 subunits, to reconstitute the response of pertussis toxin-treated erythroblasts to erythropoietin. These data directly demonstrate a physiologic function of G-proteins in hematopoietic cells and show that Gi(alpha)2 is required in erythropoietin modulation of [Ca(i)] via influx through calcium channels.

SNAP-25 palmitoylation and plasma membrane targeting require a functional secretory pathway.

Synaptosomal-associated protein of 25 kDa (SNAP-25) is a palmitoylated membrane protein essential for neurotransmitter release from synaptic terminals. We used neuronal cell lines to study the biosynthesis and posttranslational processing of SNAP-25 to investigate how palmitoylation contributes to the subcellular localization of the protein. SNAP-25 was synthesized as a soluble protein that underwent palmitoylation approximately 20 min after synthesis. Palmitoylation of the protein coincided with its stable membrane association. Treatment of cells with brefeldin A or other disrupters of transport inhibited palmitoylation of newly synthesized SNAP-25 and abolished membrane association. These results demonstrate that the processing of SNAP-25 and its targeting to the plasma membrane depend on an intact transport mechanism along the exocytic pathway. The kinetics of SNAP-25 palmitoylation and membrane association and the sensitivity of these parameters to brefeldin A suggest a novel trafficking pathway for targeting proteins to the plasma membrane. In vitro, SNAP-25 stably associated with membranes was not released from the membrane after chemical deacylation. We propose that palmitoylation of SNAP-25 is required for initial membrane targeting of the protein but that other interactions can maintain membrane association in the absence of fatty acylation.

Dual lipid modification motifs in G(alpha) and G(gamma) subunits are required for full activity of the pheromone response pathway in Saccharomyces cerevisiae.

To establish the biological function of thioacylation (palmitoylation), we have studied the heterotrimeric guanine nucleotide-binding protein (G protein) subunits of the pheromone response pathway of Saccharomyces cerevisiae. The yeast G protein gamma subunit (Ste18p) is unusual among G(gamma) subunits because it is farnesylated at cysteine 107 and has the potential to be thioacylated at cysteine 106. Substitution of either cysteine results in a strong signaling defect. In this study, we found that Ste18p is thioacylated at cysteine 106, which depended on prenylation of cysteine 107. Ste18p was targeted to the plasma membrane even in the absence of prenylation or thioacylation. However, G protein activation released prenylation- or thioacylation-defective Ste18p into the cytoplasm. Hence, lipid modifications of the G(gamma) subunit are dispensable for G protein activation by receptor, but they are required to maintain the plasma membrane association of G(betagamma) after receptor-stimulated release from G(alpha). The G protein alpha subunit (Gpa1p) is tandemly modified at its N terminus with amide- and thioester-linked fatty acids. Here we show that Gpa1p was thioacylated in vivo with a mixture of radioactive myristate and palmitate. Mutation of the thioacylation site in Gpa1p resulted in yeast cells that displayed partial activation of the pathway in the absence of pheromone. Thus, dual lipidation motifs on Gpa1p and Ste18p are required for a fully functional pheromone response pathway.

TPCK and quercetin act synergistically with vanadate to increase protein-tyrosine phosphorylation in avian cells.

We have found four compounds that act synergistically with the phosphotyrosine phosphatase inhibitor sodium orthovanadate (Na3VO4) to greatly increase the extent of protein-tyrosine phosphorylation in both uninfected chick embryo fibroblasts (CEF) and their Rous sarcoma virus-transformed counterparts (RSV-CEF). These four inhibitors fall into two categories: the chymotrypsin-specific protease inhibitors, tosyl-phenylalanine-chloromethyl ketone (TPCK) and N-carbobenzoxy-1-phenylalanine-chloromethyl ketone (ZPCK); and certain bioflavonoids which can competitively inhibit ATP-binding, quercetin and phloretin.

Signalling functions of protein palmitoylation.

Covalent lipid modifications anchor numerous signalling proteins to the cytoplasmic face of the plasma membrane. These modifications mediate protein-membrane and protein-protein interactions and are often essential for function. Protein palmitoylation, due to its reversible nature, may be particularly important for modulating protein function during cycles of activation and deactivation. Despite intense investigation, the exact functions of protein palmitoylation are not well understood. However, it is clear that palmitoylation can affect a protein's affinity for membranes, subcellular localization, and interactions with other proteins. In this review, recent advances in understanding the functions and mechanisms of protein palmitoylation are discussed, with particular emphasis on how this lipid affects the biochemistry and cell biology of signalling proteins.

Differential effects of acyl-CoA binding protein on enzymatic and non-enzymatic thioacylation of protein and peptide substrates.

Both enzymatic and autocatalytic mechanisms have been proposed to account for protein thioacylation (commonly known as palmitoylation). Acyl-CoA binding proteins (ACBP) strongly suppress non-enzymatic thioacylation of cysteinyl-containing peptides by long-chain acyl-CoAs. At physiological concentrations of ACBP, acyl-CoAs, and membrane lipids, the rate of spontaneous acylation is expected to be too slow to contribute significantly to thioacylation of signaling proteins in mammalian cells (Leventis et al., Biochemistry 36 (1997) 5546-5553). Here we characterized the effects of ACBP on enzymatic thioacylation. A protein S-acyltransferase activity previously characterized using G-protein alpha-subunits as a substrate (Dunphy et al., J. Biol. Chem., 271 (1996) 7154-7159), was capable of thioacylating short lipid-modified cysteinyl-containing peptides. The minimum requirements for substrate recognition were a free cysteine thiol adjacent to a hydrophobic lipid anchor, either myristate or farnesyl isoprenoid. PAT activity displayed specificity for the acyl donor, efficiently utilizing long-chain acyl-CoAs, but not free fatty acid or S-palmitoyl-N-acetylcysteamine. ACBP only modestly inhibited enzymatic thioacylation of a myristoylated peptide or G-protein alpha-subunits under conditions where non-enzymatic thioacylation was reduced to background. Thus, protein S-acyltransferase remains active in the presence of physiological concentrations of ACBP and acyl-CoA in vitro and is likely to represent the predominant mechanism of thioacylation in vivo.

Crystallization and preliminary crystallographic studies of Gi alpha 1 and mutants of Gi alpha 1 in the GTP and GDP-bound states.

Several different crystal forms of Gi alpha 1 have been grown and analyzed. Crystals of native protein containing bound GTP gamma S belong to space group P3(1)2(1) or P3(2)2(1) with cell dimensions a,b = 80.6 A and c = 106.3 A and diffract to a resolution of 1.9 A using synchrotron radiation. Crystals of native protein containing bound GDP belong to space group I4 with cell dimensions a,b = 121.3 A, and c = 67.7 A and diffract to 3.0 A. Data sets from crystals grown using mutant proteins have also been obtained and characterized.

Immunological characterization of proteins detected by phosphotyrosine antibodies in cells transformed by Rous sarcoma virus.

Phosphotyrosine antibodies were used to identify tyrosine-phosphorylated proteins in Rous sarcoma virus (RSV)-transformed chicken embryo fibroblasts. A large number of tyrosine phosphoproteins were detected. A similar set of proteins was observed in RSV-transformed murine cells. An 85,000-dalton protein, however, was present in transformed avian cells but missing in transformed murine cells. Neither the 85,000-dalton protein nor any of the other tyrosine phosphoproteins appeared to be viral structural proteins. Use of RSV mutants encoding partially deleted src gene products enabled us to identify a 60,000-dalton cellular tyrosine phosphoprotein that comigrated with wild-type pp60v-src. With the exception of calpactin I, the major tyrosine phosphoproteins detected in immunoblots appeared to be different from several previously characterized substrates of pp60v-src with similar molecular masses (ezrin, vinculin, and the fibronectin receptor).

Nonmyristoylated p60v-src fails to phosphorylate proteins of 115-120 kDa in chicken embryo fibroblasts.

We have used anti-phosphotyrosine antibodies to identify a large number of tyrosine phosphoproteins in Rous sarcoma virus (RSV)-transformed chicken embryo fibroblasts. Most of these proteins were found in the 100,000 X g supernatant when cells were separated into soluble and particulate fractions; however, one group of proteins, of 115-120 kDa, was found in the particulate fraction. The phosphotyrosine content of the 115- to 120-kDa proteins was greatly reduced in chicken embryo fibroblasts infected with mutants of RSV (NY314 and SD10) encoding nonmyristoylated forms of the viral src gene product that do not associate with cellular membranes. Since RSV mutants NY314 and SD10 do not transform cells, phosphorylation of this group of 115- to 120-kDa membrane proteins may be related to the process of transformation.

Selectivity of the beta-adrenergic receptor among Gs, Gi's, and Go: assay using recombinant alpha subunits in reconstituted phospholipid vesicles.

The selective regulation of Gs (long and short forms), Gi's (1, 2, and 3), and Go by the beta-adrenergic receptor was assessed quantitatively after coreconstitution of purified receptor, purified G-protein beta gamma subunits, and individual recombinant G-protein alpha subunits that were expressed in and purified from Escherichia coli. Receptor and beta gamma subunits were incorporated into phospholipid vesicles, and the alpha subunits bound to the vesicles stoichiometrically with respect to beta gamma. Efficient regulation of alpha subunit by receptor required the presence of beta gamma. Regulation of G proteins was measured according to the stimulation of the initial rate of GTP gamma S binding, steady-state GTPase activity, and equilibrium GDP/GDP exchange. The assays yielded qualitatively similar results. GDP/GDP exchange was a first-order reaction for each subunit. The rate constant increased linearly with the concentration of agonist-liganded receptor, and the dependence of the rate constant on receptor concentration was a reproducible measurement of the efficiency with which receptor regulated each G protein. Reconstituted alpha s (long or short form) was stimulated by receptor to approximately the extent described previously for natural Gs. Both alpha i,1 and alpha i,3 were regulated with 25-33% of that efficiency. Stimulation of alpha o and alpha i,2 was weak, and stimulation of alpha o was barely detectable over its high basal exchange rate. Reduction of the receptor with dithiothreitol increased the exchange rates for all G proteins but did not alter the relative selectivity of the receptor.

Action of brefeldin A blocked by activation of a pertussis-toxin-sensitive G protein.

In many mammalian cells brefeldin A interferes with mechanisms that keep the Golgi appartus separate from the endoplasmic reticulum. The earliest effect of brefeldin A is release of the coat protein beta-COP from the Golgi. This release is blocked by pretreatment with GTP-gamma S or AlF4- (ref. 12). The AlF4- ion activates heterotrimeric G proteins but not proteins of the ras superfamily, suggesting that a heterotrimeric G protein might control membrane transfer from the endoplasmic reticulum to the Golgi. We report here that mastoparan, a peptide that activates heterotrimeric G proteins, promotes binding of beta-COP to Golgi membranes in vitro and antagonizes the effect of brefeldin A on beta-COP in perforated cells and on isolated Golgi membranes. This inhibition is greatly diminished if cells are pretreated with pertussis toxin before perforation. Thus, a heterotrimeric G protein of the Gi/Go subfamily regulates association of coat components with Golgi membranes.

Enrichment of G-protein palmitoyltransferase activity in low density membranes: in vitro reconstitution of Galphai to these domains requires palmitoyltransferase activity.

Many signaling proteins are targeted to low density, sphingomyelin- and cholesterol-enriched membranes, also called lipid rafts. These domains organize receptor-mediated signaling events at the plasma membrane. Fatty acylation is one mechanism for targeting proteins to rafts. It was therefore of interest to determine if protein palmitoyltransferase activity is also present in these domains. In this study, protein palmitoyltransferase activity, assayed using G-protein alpha subunits as a substrate, was found to be highly enriched in low density membranes derived from cells that express caveolin as well as those that do not. Depletion of cellular cholesterol with the drug methyl-beta-cyclodextrin resulted in inhibition of palmitoyltransferase activity and a redistribution of the remaining activity to membranes of higher density. This effect was reversed by adding cholesterol to cyclodextrin-treated cells. When reconstituted into cell membranes, the population of purified recombinant G(alphai) that was palmitoylated was highly enriched in the low density membrane fractions, whereas the bulk unmodified G(alphai)-protein was largely excluded. This effect required palmitoyltransferase activity and was abolished if the palmitoylated cysteine was mutated. Thus, palmitoyltransferase facilitates the enrichment of fatty acylated signaling molecules in plasma membrane subdomains.

Interactions of the bovine brain A1-adenosine receptor with recombinant G protein alpha-subunits. Selectivity for rGi alpha-3.

The ability of the bovine brain A1-adenosine receptor to discriminate between different G protein subtypes was tested using G protein alpha-subunits synthesized in Escherichia coli (rG alpha-subunits). When combined with a 3-fold molar excess of beta gamma-subunit purified from bovine brain and used at high concentrations, all three subtypes of rGi alpha (rGi alpha-1, rGi alpha-2, and rGi alpha-3) and rGo alpha were capable of reconstituting guanine nucleotide-sensitive high-affinity binding of the agonist radioligand (-)-N6-3-[125I] (iodo-4-hydroxyphenylisopropyl) adenosine ([125I]HPIA) to the purified A1-adenosine receptor (Kd approximately 1.2 nM). Titration of the A1-adenosine receptor with increasing amounts of rG alpha revealed a approximately 10-fold higher affinity for rGi alpha-3 compared with rGi alpha-1, rGi alpha-2, and rGo alpha. This selectivity was also observed in the absence of beta gamma. Other alpha-subunits (rGs alpha-s, rGs alpha-L, rGs alpha PT, and rGz alpha) did not promote [125I]HPIA binding to the purified receptor. In N-ethylmaleimide-treated bovine brain membranes, rGi alpha-3 was the only rG alpha-subunit capable of reconstituting high-affinity agonist binding. Similarly, rGi alpha-3 competed potently with rGo alpha for activation by the agonist-liganded A1-adenosine receptor, whereas a approximately 50-fold molar excess of rGo alpha was required to quench the receptor-mediated release of [alpha-32P]GDP from rGi alpha-3. Hence, in spite of the extensive homology between alpha-subunits belonging to the Gi/Go group, the A1-adenosine receptor appears to discriminate between the subtypes. This specificity is likely to govern transmembrane signaling pathways in vivo.