Specific and saturable binding of pp60v-src to plasma membranes: evidence for a myristyl-src receptor

The molecular basis for membrane association of pp60v-src, the transforming protein of Rous sarcoma virus, was investigated in a cell-free system. Newly synthesized pp60v-src polypeptide, produced by in vitro translation of src mRNA, rapidly bound to plasma membranes. Binding was saturable and dependent on the presence of myristate at the amino terminus of pp60v-src. Prior treatment of membranes with heat or trypsin greatly decreased subsequent binding of pp60v-src. Membrane binding of pp60v-src was competed by a myristylated peptide containing the first 11 amino acids of the mature src sequence, but not by non-myristylated src peptide or other myristylated peptides. The specificity, saturability, and competitive nature of pp60v-src binding provide evidence for the existence of a src receptor in the plasma membrane.     

The ADP/ATP carrier is the 32-kilodalton receptor for an NH2-terminally myristylated src peptide but not for pp60src polypeptide.

Membrane binding of pp60src is initiated via its myristylated NH2 terminus. To identify a candidate pp60src docking protein or receptor in the membrane, a radiolabelled peptide corresponding to the pp60src NH2-terminal membrane binding domain was cross-linked to fibroblast membranes and found to specifically label a 32-kDa protein. This protein was purified by appending an affinity tag to the peptide probe so that the cross-linked complex could be isolated via affinity chromatography. Microsequencing indicated that the 32-kDa protein was the mitochondrial ADP/ATP carrier (AAC). This result was further confirmed by the ability of an antibody to the AAC to immunoprecipitate the cross-linked complex, by the ability of certain inhibitors of the AAC to block cross-linking, and by membrane fractionation to show that complex formation occurred essentially exclusively in the mitochondrial fraction. While the AAC bound the myristyl-src peptide in a specific manner both in vitro and in vivo, its localization to the inner membrane of the mitochondrion precludes its being a pp60src binding protein. An analysis of pp60v-src binding in vitro was consistent with this expectation. Thus, use of a myristyl-src peptide revealed an unexpected and previously unidentified binding capacity of the AAC, most likely related to the ability of long-chain fatty acyl coenzyme As to serve as AAC inhibitors. The amphipathic nature of the pp60src NH2 terminus suggests alternative strategies for uncovering pp60src membrane binding species.     

Identification of platelet membrane proteins that interact with amino-terminal peptides of pp60c-src.

Platelets contain exceptionally high levels of pp60c-src and, thus, provide a convenient system for investigating the physiological function of this protein-tyrosine kinase. We have employed chemical cross-linking of myristylated amino-terminal peptides of pp60c-src to platelet membranes in order to identify platelet membrane components that interact with pp60c-src to regulate or mediate its activity. We detected specific binding of radioiodinated peptides to platelet membrane proteins of 32, 50, 92, and 105 kDa. The 32-kDa protein may be related to the putative src receptor component recently identified in fibroblast membranes. The most reactive platelet protein, however, is the 50-kDa protein, which is either absent or nonreactive in fibroblast membranes. Binding of src peptides to this protein was saturable, and we estimate the presence of approximately 1 x 10(6) of the 50-kDa binding sites per platelet. The specificity of the peptide binding to the 50- and 32-kDa platelet proteins was analyzed by competition with different peptides. The binding sites displayed an absolute requirement for an N-myristoyl moiety and a strong preference for pp60c-src amino-terminal sequences. The identification of these src-interacting proteins may help to decipher the biochemical pathways in which platelet pp60c-src is involved.     

Reconstitution of the Rous sarcoma virus transforming protein pp60v-src into phospholipid vesicles.

An artificial membrane system was developed to study the molecular basis for interaction of pp60v-src, the Rous sarcoma virus transforming protein, with lipid bilayers. pp60v-src was extracted from cell membranes by detergent solubilization and reincorporated into phospholipid vesicles. Reconstituted pp60v-src retained tyrosine kinase activity and was integrally associated with the liposome through a 10-kilodalton (kDa) amino-terminal domain. The same 10-kDa domain was shown to anchor pp60v-src to the plasma membrane of transformed cells. Reconstitution experiments performed with nonmyristylated pp60v-src proteins revealed that these polypeptides did not interact with phospholipid vesicles. In contrast, myristylated, soluble pp60v-src molecules (including a highly purified pp60v-src preparation) could be reconstituted into liposomes, but their interaction with the liposomal bilayer was not mediated by the 10-kDa amino-terminal domain. When membrane proteins were included during reconstitution of purified pp60v-src, binding through the 10-kDa anchor was restored. A model is presented to accommodate the different types of interactions of pp60v-src with liposomes; the model postulates the existence of an additional membrane component that anchors the pp60v-src polypeptide to the phospholipid bilayer.     

The six amino-terminal amino acids of p60src are sufficient to cause myristylation of p21v-ras.

We have used oligonucleotide-directed mutagenesis to replace the N-terminal amino acids of p21v-ras with residues which mimic the amino terminus of p60v-src. p21v-ras protein possessing only the first five amino acids of p60src was not myristylated, while substitution of residue 6 (serine) produced a protein p21(GSSKS) which incorporated [3H]myristic acid that was stable to hydroxylamine, sensitive to inhibitors of protein synthesis, and found in both the normally nonacylated precursor and mature forms of p21(GSSKS). This defines the minimum framework of the p60v-src myristylation signal (glycine 2 and serine 6) and identifies serine 6 as a crucial part of that signal for myristylation of a protein in vivo.     

Processing of p60v-src to its myristylated membrane-bound form.

p60src of wild-type Rous sarcoma virus is myristylated at its N-terminal glycine residue. We have shown previously that this myristylation is necessary for p60src membrane association and for cell transformation by using src mutants with alterations within the N-terminal 30 kilodaltons of p60src. In this study we analyzed the process of p60src myristylation in wild type- and mutant-infected cells. All myristylated src proteins examined lack the initiator methionine, but two mutant src proteins lacking the initiator methionine are not myristylated, indicating that removal of the initiator methionine and myristylation are not obligatorily coupled. Analysis of the kinetics of myristylation and the association of p60src with cellular proteins p50 and p90 indicated that myristylation occurs before p60src becomes membrane associated and that transient association with p50 and p90 occurs regardless of myristylation. Myristylation is required for stable association of p60src with the plasma membrane but is not sufficient for membrane association. A mutant with an src deletion of amino acids 169 through 264 has an src protein that is myristylated but not membrane bound, remaining stably associated with p50 and p90. This mutant is transformation defective. Several N-terminal deletion mutants possessing tyrosine kinase activity have myristylated and membrane-bound src proteins but are not fully active in cell transformation, suggesting that additional N-terminal functional domains exist.     

Identification of a membrane-binding domain within the amino-terminal region of human immunodeficiency virus type 1 Gag protein which interacts with acidic phospholipids.

Retroviral Gag proteins are targeted to the plasma membrane, where they play the central role in virion formation. Several studies have suggested that the membrane-binding signal is contained within the amino-terminal matrix sequence; however, the precise location has never been determined for the Gag protein of any retrovirus. In this report, we show that the first 31 residues of human immunodeficiency virus type 1 Gag protein can function independently as a membrane-targeting domain when fused to heterologous proteins. A bipartite membrane-targeting motif was identified, consisting of the myristylated N-terminal 14 amino acids and a highly basic region that binds acidic phospholipids. Replacement of the N-terminal membrane-targeting domain of pp60v-src with that of human immunodeficiency virus type 1 Gag elicits efficient membrane binding and a transforming phenotype. Removal of myristate or the basic region results in decreased membrane binding of Gag-Src chimeras in vitro and impaired virion formation by Pr55gag in vivo. We propose that the N-terminal Gag sequence functions as a targeting signal to direct interaction with acidic phospholipids on the cytoplasmic leaflet of the plasma membrane.     

Myristoylation-regulated direct interaction between calcium-bound calmodulin and N-terminal region of pp60v-src

pp60v-src tyrosine protein kinase was suggested to interact with Ca2+-bound calmodulin (Ca2+/CaM) through the N-terminal region based on its structural similarities to CAP-23/NAP-22, a myristoylated neuron-specific protein, whose myristoyl group is essential for interaction with Ca2+/CaM; (1) the N terminus of pp60v-src is myristoylated like CAP-23/NAP-22; (2) both lysine residues are required for the myristoylation-dependent interaction and serine residues that are thought to regulate the interaction through the phosphorylations located in the N-terminal region of pp60v-src. To verify this possibility, we investigated the direct interaction between pp60v-src and Ca2+/CaM using a myristoylated peptide corresponding to the N-terminal region of pp60v-src. The binding assay indicated that only the myristoylated peptide binds to Ca2+/CaM, and the non-myristoylated peptide is not able to bind to Ca2+/CaM. Analyses of the binding kinetics revealed two independent reactions with the dissociation constants (KD) of 2.07 x 10(-9)M (KD1) and 3.93 x 10(-6)M (KD2), respectively. Two serine residues near the myristoyl moiety of the peptide (Ser2, Ser11) were phosphorylated by protein kinase C in vitro, and the phosphorylation drastically reduced the interaction. NMR experiments indicated that two molecules of the myristoylated peptide were bound around the hydrophobic clefts of a Ca2+/CaM molecule. The small-angle X-ray scattering analyses showed that the size of the peptide-Ca2+/CaM complex is 2-3A smaller than that of the known Ca2+/CaM-target molecule complexes. These results demonstrate clearly the direct interaction between pp60v-src and Ca2+/CaM in a novel manner different from that of known Ca2+/CaM, the target molecules, interactions.     

In vitro synthesis of pp60v-src: myristylation in a cell-free system.

Covalent attachment of myristic acid to pp60v-src, the transforming protein of Rous sarcoma virus, was studied in a cell-free system. Using a synthetic peptide containing the first 11 amino acids of the mature pp60v-src polypeptide sequence as a substrate, we probed lysates from a variety of cells and tissues for N-myristyl transferase (NMT) activity. Nearly every eucaryotic cell type tested contained NMT, including avian, mammalian, insect, and plant cells. Since NMT activity was detected in rabbit reticulocyte lysates, we took advantage of the translational capability of these lysates to determine the precise point during translation at which myristate is attached to pp60v-src. src mRNA, transcribed from cloned v-src DNA, was translated in reticulocyte lysates which had been depleted of endogenous myristate. Addition of [3H]myristate to lysates 10 min after the start of synchronized translation resulted in a dramatic decrease in the incorporation of radiolabeled myristate into pp60v-src polypeptide chains. These results imply that although myristate can be attached posttranslationally to synthetic peptide substrates, myristylation in vivo is apparently a very early cotranslational event which occurs before the first 100 amino acids of the nascent polypeptide chain are polymerized.     

Characterization of pp60src phosphorylation in vitro in Rous sarcoma virus-transformed cell membranes.

Phosphorylation of the src gene product pp60v-src was studied in plasma membrane fractions prepared from Rous sarcoma virus-transformed vole cells. Upon addition of [gamma-32P]ATP to isolated membrane vesicles, phosphate was incorporated into a 60,000-dalton polypeptide identified as pp60v-src. In the presence of vanadate, pp60v-src phosphorylation was stimulated ca. 30-fold. At low concentrations of ATP (1 microM), this reaction occurred almost exclusively on the carboxy-terminal 26,000-dalton region of pp60v-src. However, at higher ATP concentrations (100 microM), additional sites of phosphorylation were evident in the amino-terminal 34,000-dalton region. Kinetic analyses, performed under conditions in which ATP hydrolysis was minimal, revealed that the phosphorylation reaction at the carboxy terminus exhibited a higher Vmax and a lower Km for ATP than those occurring at the amino terminus. In addition, the amino-terminal region of pp60v-src was more rapidly dephosphorylated than the carboxy-terminal region. These results indicate that interaction of pp60v-src with the plasma membrane may limit the extent of amino-terminal phosphorylation by lowering the rate of the reaction and the affinity for the substrate while increasing its susceptibility to phosphoprotein phosphatases. We suggest that the use of transformed-cell membrane preparations provides a model system for studying the possible regulatory roles of phosphorylation and dephosphorylation on pp60v-src function.     

Requirement of phosphatidylinositol-3 kinase modification for its association with p60src.

When purified p60v-src was mixed with lysates of chicken embryo fibroblasts and immunoprecipitated with anti-Src antibody, phosphatidylinositol (PI)-3 kinase activity was found to be present in the Src protein immunoprecipitates. The level of bound PI-3 kinase activity was 5 to 10 times higher in lysates obtained from cells transformed by the src, fps, or yes oncogene than in lysates of uninfected cells. This increase in associated PI-3 kinase activity appears to be due to increased binding of this enzyme to p60v-src. This change most likely resulted from tyrosine phosphorylation of PI-3 kinase or an associated protein, since the PI-3 kinase activity that can bind to p60v-src was depleted by antiphosphotyrosine antibody. Binding of PI-3 kinase did not require either p60src protein kinase activity or autophosphorylation of p60v-src tyrosine residues. Furthermore, binding was markedly decreased by deletions in the N-terminal SH2 region but unchanged by deletion of the C-terminal half of p60v-src containing the catalytic domain. Taking these data together, it appears that PI-3 kinase or its associated protein is phosphorylated on tyrosine and that the phosphorylated form can bind to the N-terminal half of p60v-src, which contains the SH2 domain.     

Highly specific antibody to Rous sarcoma virus src gene product recognizes a novel population of pp60v-src and pp60c-src molecules

Antiserum to the Rous sarcoma virus (RSV)-transforming protein, pp60v-src, was produced in rabbits immunized with p60 expressed in Escherichia coli. alpha p60 serum immunoprecipitated quantitatively more pp60v-src than did tumor-bearing rabbit (TBR) sera. When RSV-transformed cell lysates were preadsorbed with TBR serum, the remaining lysate contained additional pp60v-src, which was recognized only by reimmunoprecipitation with alpha p60 serum and not by TBR serum. In subcellular fractions of RSV-infected chicken embryo fibroblasts (RSV-CEFs) and field vole cells probed with TBR serum, the majority of the pp60v-src was associated with the plasma membrane-enriched P100 fraction. However, alpha p60 serum revealed equal distribution of pp60v-src and its kinase activity between the P1 (nuclear) and P100 fractions. The same results were obtained for pp60c-src in uninfected CEFs. On discontinuous sucrose gradients nearly 50% of the P1-pp60v-src sedimented with nuclei, in fractions where no plasma membrane was detected. Indirect immunofluorescence microscopy of RSV-CEFs with alpha p60 serum revealed a distinct pattern of perinuclear fluorescence, in addition to staining at the cell periphery. Thus the use of a highly specific antibody reveals that enzymatically active pp60v-src and pp60c-src molecules are present in other intracellular structures, probably juxtareticular nuclear membranes, in addition to the plasma membrane in normal, uninfected, and wild-type RSV-infected cells.     

The myristylation signal of p60v-src functionally complements the N-terminal fps-specific region of P130gag-fps.

The P130gag-fps protein-tyrosine kinase of Fujinami sarcoma virus contains an N-terminal fps-specific domain (Nfps) that is important for oncogenicity. The N-terminal 14 amino acids of p60v-src, which direct myristylation and membrane association, can replace the gag-Nfps sequences of P130gag-fps (residues 1 to 635), producing a highly transforming src-fps polypeptide. Conversely, gag-Nfps can restore modest transforming activity to a nonmyristylated v-src polypeptide. These results emphasize the modular construction of protein-tyrosine kinases and indicate that Nfps, possibly in conjunction with gag, functions in the subcellular localization of P130gag-fps.     

Protein kinase C phosphorylates pp60src at a novel site

The transforming protein of Rous sarcoma virus (pp60v-src) and its normal cellular homolog (pp60c-src) are demonstrated to be phosphorylated at serine 12 in vivo under certain conditions. We propose that protein kinase C is responsible for this modification based on the following evidence. First, the tumor promoters, 12-O-tetradecanoylphorbol-13-acetate and teleocidin, and synthetic diacylglycerol, known activators of protein kinase C in vivo, cause nearly complete phosphorylation of pp60src at serine 12. Second, among five purified serine/threonine-specific protein kinases tested, only protein kinase C phosphorylates pp60c-src and pp60v-src in vitro at serine 12. Third, purified protein kinase C phosphorylates a synthetic peptide corresponding to the N-terminal 20 amino acids of pp60c-src at serine 12. The physiological significance of this novel phosphorylation is discussed.     

Transformation-specific tyrosine phosphorylation of a novel cellular protein in chicken cells expressing oncogenic variants of the avian cellular src gene.

We used myristylated and nonmyristylated c-src-based variants and phosphotyrosine-specific antibodies to reevaluate the role of tyrosine phosphorylation in cellular transformation by pp60src. Prior methods used to detect tyrosine-phosphorylated proteins failed to discriminate predicted differences in tyrosine phosphorylation which are clearly observed with phosphotyrosine-specific antibodies and Western blotting (immunoblotting). Here we report the observation of a 120,000-Mr protein whose phosphorylation on tyrosine correlates with the induction of morphological transformation. p120 was not observed in cells overexpressing the regulated, nononcogenic pp60c-src, whereas phosphorylation of p120 was greatly enhanced in cells expressing activated, oncogenic pp60527F. Furthermore, phosphorylation of p120 was not induced by expression of the activated but nonmyristylated src variant pp602A/527F, which is transformation defective. p120 partitioned preferentially with cellular membranes, consistent with the observation that transforming src proteins are membrane associated. Although a number of additional putative substrates were identified and partially characterized with respect to intracellular localization, tyrosine phosphorylation of these proteins was not tightly linked to transformation.     

Intracellular localization and processing of pp60v-src proteins expressed by two distinct temperature-sensitive mutants of Rous sarcoma virus.

The transforming protein of Rous sarcoma virus, pp60v-src, is known to be a tyrosine protein kinase, but the mechanism of cell transformation remains unclear. In further investigating pp60v-src structure and function, we have analyzed two temperature-sensitive (ts) Rous sarcoma virus src gene mutants, tsLA29 and tsLA32. The mutations in tsLA29 and tsLA32 map in the carboxy-terminal region and the amino-terminal half of pp60v-src, respectively, and encode mutant proteins with either temperature-labile (tsLA29) or -stable (tsLA32) kinase activities. Here we examined the intracellular processing and localization of these pp60v-src mutants and extended our characterization of transformation parameters expressed by cells infected by the Rous sarcoma virus variants. No obvious defects in functional integrity of the tsLA32 pp60v-src could yet be demonstrated, whereas the tsLA29 pp60v-src was perturbed not only in kinase activity, but also in aspects of protein processing and localization. Analysis of transformation parameters expressed by infected cells demonstrated the complete temperature lability of both mutants.     

The src protein contains multiple domains for specific attachment to membranes.

The proteins encoded by the oncogene v-src and its cellular counterpart c-src (designated generically here as pp60src) are tightly associated with both plasma membranes and intracellular membranes. This association is due in part to the amino-terminal myristylation of pp60src, but several lines of evidence suggest that amino-terminal portions of the protein itself are also involved. We now report that pp60src contains at least three domains which, in conjunction with myristylation, are capable of mediating attachment to membranes and determining subcellular localization. We identified these domains by fusing various portions of pp60src to pyruvate kinase, which is normally a cytoplasmic protein. Amino acids 1 to 14 of pp60src are sufficient to mediate both myristylation and the attachment of pyruvate kinase to cytoplasmic granules. In contrast, amino acids 38 to 111 mediate association with the plasma membrane and perinuclear membranes, whereas amino acids 204 to 259 mediate association primarily with perinuclear membranes. We conclude that pp60src contains independent domains that target the protein to distinctive subcellular locations and thus may facilitate diverse biological functions of the protein.     

Inhibition of the tyrosine kinase activity of v-src, v-fgr, and v-yes gene products by a monoclonal antibody which binds both amino and carboxy peptide fragments of pp60v-src.

A monoclonal antibody, R2D2, raised to the src gene product of Rous sarcoma virus was found to inhibit the tyrosine protein kinase activity of pp60v-src in autophosphorylation reactions and in reactions involving exogenously added substrates, such as casein and histone. R2D2 also inhibited the enzymatic activity of two related viral transforming proteins, pp70gag-fgr and pp90gag-yes. The inhibitory ability of R2D2 was dependent upon immunoglobulin concentration and could be demonstrated in both immune complexes formed directly with R2D2 and preformed immune complexes to which R2D2 was added. Binding sites in both the amino-terminal 110 amino acid residues and the carboxy-terminal 240 amino acids of pp60v-src were identified for R2D2. These results indicate that at least part of the epitope recognized by R2D2 resides within a region of the src protein which is required for protein kinase activity. The localization of the R2D2 epitope to the amino- as well as to the carboxy-terminal portions of pp60v-src, together with results of studies analyzing the relative binding efficiencies of R2D2 to the intact protein and to V-8 proteolytic fragments of pp60v-src, are consistent with the view that the R2D2 epitope is conformational in nature and that it is assembled from residues contained within both N-terminal and C-terminal regions of the molecule.     

Aberrant protein phosphorylation at tyrosine is responsible for the growth-inhibitory action of pp60v-src expressed in the yeast Saccharomyces cerevisiae.

Expression of pp60v-src, the transforming protein of Rous sarcoma virus, arrests the growth of the yeast Saccharomyces cerevisiae. To determine the basis of this growth arrest, yeast strains were constructed that expressed either wild-type v-src or various mutant v-src genes under the control of the galactose-inducible, glucose repressible GAL1 promoter. When shifted to galactose medium, cells expressing wild-type v-src ceased growth immediately and lost viability, whereas cells expressing a catalytically inactive mutant (K295M) continued to grow normally, indicating that the kinase activity of pp60v-src is required for its growth inhibitory effect. Mutants of v-src altered in the SH2/SH3 domain (XD4, XD6, SPX1, and SHX13) and a mutant lacking a functional N-terminal myristoylation signal (MM4) caused only a partial inhibition of growth, indicating that complete growth inhibition requires either targeting of the active kinase or binding of the kinase to phosphorylated substrates, or both. Cells arrested by v-src expression displayed aberrant microtubule structures, alterations in DNA content and elevated p34CDC28 kinase activity. Immunoblotting with antiphosphotyrosine antibody showed that many yeast proteins, including the p34CDC28 kinase, became phosphorylated at tyrosine in cells expressing v-src. Both the growth inhibition and the tyrosine-specific protein phosphorylation observed following v-src expression were reversed by co-expression of a mammalian phosphotyrosine-specific phosphoprotein phosphatase (PTP1B). However a v-src mutant with a small insertion in the catalytic domain (SRX5) had the same lethal effect as wild-type v-src, yet induced only very low levels of protein-tyrosine phosphorylation. These results indicate that inappropriate phosphorylation at tyrosine is the primary cause of the lethal effect of pp60v-src expression but suggest that only a limited subset of the phosphorylated proteins are involved in this effect.