Molecular features of the viral and cellular Src kinases involved in interactions with the GTPase-activating protein.

GTPase-activating protein (GAP) enhances the rate of GTP hydrolysis by cellular Ras proteins and is implicated in mitogenic signal transduction. GAP is phosphorylated on tyrosine in cells transformed by Rous sarcoma virus and serves as an in vitro substrate of the viral Src (v-Src) kinase. Our previous studies showed that GAP complexes stably with normal cellular Src (c-Src), although its association with v-Src is less stable. To further investigate the molecular basis for interactions between GAP and the Src kinases, we examined GAP association with and phosphorylation by a series of c-Src and v-Src mutants. Analysis of GAP association with c-Src/v-Src chimeric proteins demonstrates that GAP associates stably with Src proteins possessing low kinase activity and poorly with activated Src kinases, especially those that lack the carboxy-terminal segment of c-Src containing the regulatory amino acid Tyr-527. Phosphorylated Tyr-527 is a major determinant of c-Src association with GAP, as demonstrated by c-Src point mutants in which Tyr-527 is changed to Phe. While the isolated amino-terminal half of the c-Src protein is insufficient for stable GAP association, analysis of point substitutions of highly conserved amino acid residues in the c-Src SH2 region indicate that this region also influences Src-GAP complex formation. Therefore, our results suggest that both Tyr-527 phosphorylation and the SH2 region contribute to stable association of c-Src with GAP. Analysis of in vivo phosphorylation of GAP by v-Src mutants containing deletions encompassing the SH2, SH3, and unique regions suggests that the kinase domain of v-Src contains sufficient substrate specificity for GAP phosphorylation. Even though tyrosine phosphorylation of GAP correlates to certain extent with the transforming ability of various c-Src and v-Src mutants, our data suggest that other GAP-associated proteins may also have roles in Src-mediated oncogenic transformation. These findings provide additional evidence for the specificity of Src interactions with GAP and support the hypothesis that these interactions contribute to the biological functions of the Scr kinases.     

Reconstitution of interactions between the Src tyrosine kinases and Ras GTPase-activating protein using a baculovirus expression system.

Ras GTPase-activating protein (GAP) has been implicated in mitogenic signal transduction downstream of oncogenic and receptor tyrosine kinases. Previous studies have suggested that GAP is phosphorylated by oncogenic viral Src (v-Src) and that GAP is associated with a complex containing normal cellular Src (c-Src) in vertebrate fibroblasts. To investigate molecular interactions between the Src kinases and GAP, we developed an in vitro system for reconstituting Src-GAP complexes. For this purpose, we constructed recombinant baculovirus vectors that direct expression of Rous sarcoma virus v-Src, chicken c-Src, and bovine GAP in infected Sf9 insect cells. In vitro reconstitution experiments using baculovirus-expressed proteins demonstrate that both v-Src and c-Src associate in complexes with GAP. In addition, in vitro and in vivo phosphorylation analyses indicate that GAP serves as a substrate for both the v-Src and c-Src tyrosine kinases. To determine which structural features of GAP are involved in interactions with the Src kinases, we constructed recombinant baculoviruses that encode deletion mutants of bovine GAP. Deletion of the GAP amino-terminal portion containing Src homology 2 regions, which are highly conserved structural motifs postulated to mediate interactions among proteins, diminishes GAP phosphorylation and association with Src. This reconstitution system should facilitate further studies of molecular interactions between the Src kinases and GAP.     

RACK1 inhibits the serum- and anchorage-independent growth of v-Src transformed cells

Cancer cells are capable of serum- and anchorage-independent growth, and focus formation on monolayers of normal cells. Previously, we showed that RACK1 inhibits c-Src kinase activity and NIH3T3 cell growth. Here, we show that RACK1 partially inhibits v-Src kinase activity, and the serum- and anchorage-independent growth of v-Src transformed cells, but has no effect on focus formation. RACK1-overexpressing v-Src cells show disassembly of podosomes, which are actin-rich structures that are distinctive to fully transformed cells. Together, our results demonstrate that RACK1 overexpression in v-Src cells partially reverses the transformed phenotype of the cells. Our results identify an endogenous inhibitor of the oncogenic Src tyrosine kinase and of cell transformation.     

v-Src's hold over actin and cell adhesions

The oncoprotein v-Src and its cellular homologue (c-Src) are tyrosine kinases that modulate the actin cytoskeleton and cell adhesions. Through the concerted action of their protein-interaction and kinase domains, they are targeted to cell matrix integrin adhesions or cadherin-dependent junctions between epithelial cells, where they phosphorylate substrates that induce adhesion turnover and actin re-modelling. Recent experiments have defined some of the key targets and effector pathways that mediate the pleiotropic oncogenic effects of v-Src.     

Activated Src tyrosine kinase phosphorylates Tyr-457 of bovine GTPase-activating protein (GAP) in vitro and the corresponding residue of rat GAP in vivo.

GTPase-activating protein (GAP) is a key regulator of the cellular Ras protein, which is implicated in oncogenic signal transduction pathways downstream of the viral Src (v-Src) kinase. Previous studies demonstrated that v-Src induces tyrosine phosphorylation of GAP, suggesting that GAP may provide a biochemical link between v-Src and Ras signaling pathways. To determine the precise residues in GAP phosphorylated by Src kinases, we used a baculovirus/insect cell expression system for investigating in vitro phosphorylation of GAP. Phosphopeptide mapping analysis revealed that v-Src and normal cellular Src (c-Src) phosphorylate tyrosine residues in bovine GAP at one major site and one minor site in vitro. Significantly, the major site of GAP phosphorylation in vitro is also the major site of in vivo tyrosine phosphorylation of GAP in rat fibroblasts transformed by v-Src. Analyses of GAP deletion mutants and TrpE-GAP fusion proteins established that Tyr-457 of bovine GAP (and the corresponding residue of rat and human GAP) is the major site of tyrosine phosphorylation. Our results demonstrate that the v-Src kinase induces phosphorylation of the same tyrosine residue of GAP in vitro and in vivo, suggesting that GAP is a direct substrate of activated Src kinases in vivo. Because epidermal growth factor receptor phosphorylates the equivalent tyrosine residue in human GAP (Tyr-460), these findings are consistent with the hypothesis that specific phosphorylation of GAP at this site may have a physiologically important role in regulating mitogenic Ras signaling pathways.     

Localization of the viral and cellular Src kinases to perinuclear vesicles in fibroblasts.

Previous studies have shown that the viral oncogene product, v-Src, is localized to a perinuclear structure. Here, we demonstrate by immunofluorescence analysis that the perinuclear structure corresponds to a local concentration of vesicles. Overexpressed normal cellular Src, c-Src, and a temperature-sensitive mutant of v-Src also are associated with these perinuclear vesicles. This perinuclear localization is observed in a variety of cell types using several different antibodies to Src, and it is independent of the fixation method. Immunoelectron ultracryomicroscopic analysis of rat fibroblast cells transformed by v-Src demonstrates an association of this protein with the limiting membranes of vesicles concentrated in the perinuclear region. These vesicles appear at the electron microscopic level to be multivesicular bodies on the basis of their size (0.3-1.0 microns in diameter), large electron-transparent lumens, and electron-dense vesicular inclusions. Morphometric analysis indicates that approximately 20% of the total cell v-Src protein is associated with these structures. This subpopulation of v-Src may have been recovered from the plasma membrane via the endocytotic pathway in a manner analogous to endocytosis of the epidermal growth factor receptor. Localization of the Src tyrosine kinases to these perinuclear endocytotic vesicles may be necessary for oncogenic transformation by v-Src and for normal functions of c-Src.     

The SH3 domain of Src can downregulate its kinase activity in the absence of the SH2 domain-pY527 interaction

The contact between the SH2 domain and the C-terminal tail of c-Src inhibits its kinase activity via a complex network of interactions, including the SH3 domain. We examined the role of the SH3 domain in v-Src, where the C-terminal tail is mutated and unbound. We used the v-Src variants Prague C (PRC) and Schmidt-Ruppin A (SRA), which are of low and high kinase activities, respectively, to measure phosphorylation in vitro by immunoprecipitated kinases produced in Saccharomyces cerevisiae. Swapping the regulatory domains between SRA and PRC revealed that N117D, I96T, and V124L mutations in the n-src- and RT-loops of the SH3 domain of PRC are responsible for the low kinase activity of PRC. Moreover, introducing D117N, R95W, T96I, and L124V into activated c-Src(Y527F) caused a 2.5-fold increase in its activity. The mutations in the CD linker KP249,250DG and L255A, which were shown to activate c-Src, had no effect on the activity of the "SH2-activated" Src kinases. Together our data suggest that in the "SH2-activated" forms of Src, the SH3 domain continues to influence the kinase activity via the direct contacts of the n-src- and RT-loops with the kinase N-terminal lobe.     

The v-Src and c-Src tyrosine kinases immunoprecipitated from Rous sarcoma virus-transformed cells display different peptide substrate specificities

In the cells transformed by Rous sarcoma virus (RSV), two Src proteins are expressed: the ubiquitous tyrosine kinase c-Src and the v-Src, the product of the transforming gene of the virus. Using three synthetic peptide substrates widely used for testing Src kinase activity, we show that they are phosphorylated with different efficiencies by the v-Src and c-Src tyrosine kinases immunoprecipitated from the tumor cell line H19. The v-Src displays higher efficiency (Vmax/Km ratio) toward all three peptides used, but the Vmax of v-Src is much lower than Vmax of c-Src with two peptides out of three. This difference in substrate specificity, if ignored, may cause misestimation of the amounts of active c-Src and v-Src in RSV-transformed cells. On the other hand, the different peptide substrate specificities may also reflect different protein substrate specificities of the v-Src and c-Src kinases in vivo.     

The v-Src SH3 domain facilitates a cell adhesion-independent association with focal adhesion kinase

Integrins facilitate cell attachment to the extracellular matrix, and these interactions generate cell survival, proliferation, and motility signals. Integrin signals are relayed in part by focal adhesion kinase (FAK) activation and the formation of a transient signaling complex initiated by Src homology 2 (SH2)-dependent binding of Src family protein-tyrosine kinases to the FAK Tyr-397 autophosphorylation site. Here we show that in viral Src (v-Src)-transformed NIH3T3 fibroblasts, an adhesion-independent FAK-Src signaling complex occurs. Co-expression studies in human 293T cells showed that v-Src could associate with and phosphorylate a Phe-397 FAK mutant at Tyr-925 promoting Grb2 binding to FAK in suspended cells. In vitro, glutathione S-transferase fusion proteins of the v-Src SH3 but not c-Src SH3 domain bound to FAK in lysates of NIH3T3 fibroblasts. The v-Src SH3-binding sites were mapped to known proline-X-X-proline (PXXP) SH3-binding motifs in the FAK N- (residues 371-377) and C-terminal domains (residues 712-718 and 871-882) by in vitro pull-down assays, and these sites are composed of a PXXPXXPhi (where Phi is a hydrophobic residue) v-Src SH3 binding consensus. Sequence comparisons show that residues in the RT loop region of the c-Src and v-Src SH3 domains differ. Substitution of c-Src RT loop residues (Arg-97 and Thr-98) for those found in the v-Src SH3 domain (Trp-97 and Ile-98) enhanced the binding of distinct NIH3T3 cellular proteins to a glutathione S-transferase fusion protein of the c-Src (Trp-97 + Ile-98) SH3 domain. FAK was identified as a c-Src (Trp-97 + Ile-98) SH3 domain target in fibroblasts, and co-expression studies in 293T cells showed that full-length c-Src (Trp-97 + Ile-98) could associate in vivo with Phe-397 FAK in an SH2-independent manner. These studies establish a functional role for the v-Src SH3 domain in stabilizing an adhesion-independent signaling complex with FAK.     

Regulation of c-Src activity by the expression of wild-type v-Src and its kinase-dead double Y416F-K295N mutant

Active, wild-type v-Src and its kinase-dead double Y416F-K295N mutant were expressed in hamster fibroblasts. Expression of the active v-Src induced activation of endogenous c-Src and increased general protein-tyrosine phosphorylation in the infected cells. Expression of the kinase-dead mutant induced hypophosphorylation of Tyr416 of the endogenous c-Src. The inactivation of c-Src was reversible, as confirmed by in vitro kinase activity of c-Src immunoprecipitated from the kinase-dead v-Src-expressing cells. Both activation and inactivation of c-Src may be explained by direct interaction of the v-Src and c-Src that may either facilitate transphosphorylation of the regulatory Tyr416 in the activation loop, or prevent it by formation of transient dead-end complexes of the Y416F-K295N mutant with c-Src. The interaction was also indicated by co-localization of v- and c-Src proteins in immunofluorescent images of the infected cells. These results suggest that dimerization of Src plays an important role in the regulation of Src tyrosine kinase activity.     

Substrate-specific modulation of Src-mediated phosphorylation of Ras and caseins by sphingosines and other substrate modulators.

It is important for the understanding of protein kinase action to differentiate between regulation at the enzyme and at the substrate levels. For example, the inhibitors dinitrophenol-tyrosine and tyrphostins act at the enzyme level to inhibit phosphorylation of all substrates by c-Src and v-Src kinases. In contrast, polylysine acts at the substrate level to stimulate Src-mediated phosphorylation of beta-casein but to inhibit phosphorylation of alpha-casein. Here we demonstrate novel enzyme-specific and substrate-specific modulations of Src kinase activity of potential physiological significance. At the enzyme level, we observed that c-Src kinase preferentially phosphorylates alpha-casein, while the v-Src kinase prefers beta-casein. At the substrate level we observed substrate-specific modulation by physiological factors including sphingosine, sphingosine derivatives and the ganglioside GM3. Galactosyl-sphingosine (psychosine) was more effective in stimulating phosphorylation of beta-casein and poly(E1A1Y1) than sphingosine. Glucosyl- and lactosyl-sphingosine were ineffective. Rat was extensively phosphorylated by c-Src in the presence of polylysine, and to a lesser extent in the sphingosine and galactosyl-sphingosine. These unexpected differences point out another potential mechanism for regulation of c-Src and v-Src kinase activities and may help to explain some of the pleotyptic manifestations of protein tyrosine kinase actions.     

Tyrosine phosphorylation of focal adhesion kinase at sites in the catalytic domain regulates kinase activity: a role for Src family kinases.

Focal adhesion kinase (FAK) is a widely expressed nonreceptor protein-tyrosine kinase implicated in integrin-mediated signal transduction pathways and in the process of oncogenic transformation by v-Src. Elevation of FAK's phosphotyrosine content, following both cell adhesion to extracellular matrix substrata and cell transformation by Rous sarcoma virus, correlates directly with an increased kinase activity. To help elucidate the role of FAK phosphorylation in signal transduction events, we used a tryptic phosphopeptide mapping approach to identify tyrosine sites of phosphorylation responsive to both cell adhesion and Src transformation. We have identified four tyrosines, 397, 407, 576, and 577, which are phosphorylated in mouse BALB/3T3 fibroblasts in an adhesion-dependent manner. Tyrosine 397 has been previously recognized as the major site of FAK autophosphorylation. Phosphorylation of tyrosines 407, 576, and 577, which are previously unrecognized sites, is significantly elevated in the presence of c-Src in vitro and v-Src in vivo. Tyrosines 576 and 577 lie within catalytic subdomain VIII--a region recognized as a target for phosphorylation-mediated regulation of protein kinase activity. We found that maximal kinase activity of FAK immune complexes requires phosphorylation of both tyrosines 576 and 577. Our results indicate that phosphorylation of FAK by Src (or other Src family kinases) is an important step in the formation of an active signaling complex.     

Phospholipase D prevents apoptosis in v-Src-transformed rat fibroblasts and MDA-MB-231 breast cancer cells

Phospholipase D (PLD) activity is elevated in response to mitogenic and oncogenic signals. PLD also cooperates with overexpressed tyrosine kinases to transform rat fibroblasts. 3Y1 rat fibroblasts overexpressing the tyrosine kinase c-Src undergo apoptosis in response to serum withdrawal. We report here that elevated expression of either PLD1 or PLD2 in these cells prevents apoptosis induced by serum withdrawal. 3Y1 cells transformed by the activated tyrosine kinase v-Src have elevated PLD activity and are resistant to apoptosis induced by serum withdrawal. However, if PLD activity is blocked, the v-Src-transformed cells underwent apoptosis. MDA-MB-231 cells are a human breast cancer cell line with substantially elevated levels of PLD activity. Inhibiting PLD activity in these cells similarly rendered them sensitive to the apoptotic insult of serum withdrawal. These data indicate that elevated PLD activity generates a survival signal(s) allowing cells to overcome default apoptosis programs.     

Increased activity of c-Src and Csk in fibroblasts transformed by v-src oncogene.

When c-Src and v-Src were immunoprecipitated together from hamster fibroblasts transformed by Rous sarcoma virus containing v-src oncogene, the total Src activity was almost threefold higher compared to c-Src activity in the control cells. The activity of v-Src immunoprecipitated separately, however, accounting for only 40% of the total Src activity, indicating that c-Src is activated upon transformation. An increased activity of Csk was also found in RSV-transformed cells. It decreased upon serum stimulation in parallel with an increase in Src kinase activity. In nontransformed cells, serum stimulation induced an enhanced Csk activity, but no changes in c-Src activity were observed. This may suggest that Csk may have more functions in hamster fibroblasts, in addition to its inhibitory effect on c-Src.     

Activation of c-Src in cells bearing v-Crk and its suppression by Csk.

The protein product of the CT10 virus, p47gag-crk (v-Crk), which contains Src homology region 2 (SH2) and 3 (SH3) domains but lacks a kinase domain, is believed to cause an increase in cellular protein tyrosine phosphorylation. A candidate tyrosine kinase, Csk (C-terminal Src kinase), has been implicated in c-Src Tyr-527 phosphorylation, which negatively regulates the protein tyrosine kinase of pp60c-src (c-Src). To investigate how c-Src kinase activity is regulated in vivo, we first looked at whether v-Crk can activate c-Src kinase. We found that cooverexpression of v-Crk and c-Src caused elevation of c-Src kinase activity, resulting in an increase of tyrosine phosphorylation of cellular proteins and morphological transformation of rat 3Y1 fibroblasts. v-Crk and c-Src complexes were not detected, although v-Crk bound to a variety of tyrosine-phosphorylated proteins in cells overexpressing v-Crk and c-Src. Overexpression of Csk in these transformed cells caused reversion to normal phenotypes and also reduced the level of c-Src kinase activity. However, Csk did not cause reversion of cells transformed by v-Src or c-Src527F, in which Tyr-527 was changed to Phe. These results strongly suggest that Csk acts on Tyr-527 of c-Src and suppresses c-Src kinase activity in vivo. Because Csk can suppress transformation by cooverexpression of v-Crk and c-Src, we suggest that v-Crk causes activation of c-Src in vivo by altering the phosphorylation state of Tyr-527.     

v-Src rescues actin-based cytoskeletal architecture and cell motility and induces enhanced anchorage independence during oncogenic transformation of focal adhesion kinase-null fibroblasts

The ability of the focal adhesion kinase (FAK) to integrate signals from extracellular matrix and growth factor receptors requires the integrity of Tyr397, a major autophosphorylation site that mediates the Src homology 2-dependent binding of Src family kinases. However, the precise roles played by FAK in specific Src-induced pathways, especially as they relate to oncogenic transformation, remain unclear. Here, we investigate the role of FAK in v-Src-induced oncogenic transformation by transducing temperature-sensitive v-Src (ts72v-Src) into p53-null FAK+/+ or FAK-/- mouse embryo fibroblasts (MEF). At the permissive temperature (PT), ts72v-Src induced abundant tyrosine phosphorylation, morphological transformation and cytoskeletal rearrangement in FAK-/- MEF, including the restoration of cell polarity, typical focal adhesion complexes, and longitudinal F-actin stress fibers. v-Src rescued the haptotactic, linear directional, and invasive motility defects of FAK-/- cells to levels found in FAK+/+ or FAK+/+-[ts72v-Src] cells, and, in the case of monolayer wound healing motility, there was an enhancement. Src activation failed to increase the high basal tyrosine phosphorylation of the Crk-associated substrate, CAS, found in FAK-/- MEF, indicating that CAS phosphorylation alone is insufficient to induce motility in the absence of FAK- or v-Src-induced cytoskeletal remodeling. Compared with FAK+/+[ts72v-Src] controls, FAK-/-[ts72v-Src] clones exhibited 7-10-fold higher anchorage-independent proliferation that could not be attributed to variations in either v-Src protein level or stability. Re-expression of FAK diminished the colony-forming activities of FAK-/-[ts72v-Src] without altering ts72v-Src expression levels, suggesting that FAK attenuates Src-induced anchorage independence. Our data also indicate that the enhanced Pyk2 level found in FAK-/- MEF plays no role in v-Src-induced anchorage independence. Overall, our data indicate that FAK, although dispensable, attenuates v-Src-induced oncogenic transformation by modulating distinct signaling and cytoskeletal pathways.     

Nitric oxide controls src kinase activity through a sulfhydryl group modification-mediated Tyr-527-independent and Tyr-416-linked mechanism.

c-Src kinase was activated when either murine NIH3T3 fibroblast cells or immunoprecipitated c-Src proteins were treated with nitric oxide generator, S-nitroso-N-acetyl penicillamine (SNAP) or sodium nitroprusside. Nitric oxide (NO) scavenger hemoglobin and N(2)O(3) scavenger homocysteine abolished the SNAP-mediated c-Src kinase activation. Phosphoamino acid analysis and peptide mapping of in vitro labeled phospho-c-Src proteins revealed that SNAP promoted the autophosphorylation at tyrosine, which preferentially took place at Tyr-416. Peptide mapping of in vivo labeled c-Src kinase excluded the involvement of phospho-Tyr-527 dephosphorylation in the SNAP-mediated activation mechanism. Correspondingly, protein-tyrosine phosphatase inhibitor Na(3)VO(4) did not abolish the SNAP-mediated activation of Src kinase, and the constitutively activated v-Src kinase was also further up-regulated in activity by SNAP. SNAP, however, failed to up-regulate the kinase activity of Phe-416 mutant v-Src. 2-Mercaptoethanol or dithiothreitol, which should disrupt N(2)O(3)-mediated S-nitrosylation and subsequent formation of the S-S bond, abolished the up-regulated catalytic activity, and the activity was regained after re-exposing the enzyme to SNAP. Exposure of Src kinase to SNAP promoted both autophosphorylation and S-S bond-mediated aggregation of the kinase molecules, demonstrating a linkage between the two events. These results suggest that the NO/N(2)O(3)-provoked S-nitrosylation/S-S bond formation destabilizes the Src structure for Tyr-416 autophosphorylation-associated activation bypassing the Tyr-527-linked regulation.     

Translocation of Src kinase to the cell periphery is mediated by the actin cytoskeleton under the control of the Rho family of small G proteins

We have isolated Swiss 3T3 subclones that are resistant to the mitogenic and morphological transforming effects of v-Src as a consequence of aberrant translocation of the oncoprotein under low serum conditions. In chicken embryo and NIH 3T3 fibroblasts under similar conditions, v-Src rapidly translocates from the perinuclear region to the focal adhesions upon activation of the tyrosine kinase, resulting in downstream activation of activator protein-1 and mitogen- activated protein kinase, which are required for the mitogenic and transforming activity of the oncoprotein. Since serum deprivation induces cytoskeletal disorganization in Swiss 3T3, we examined whether regulators of the cytoskeleton play a role in the translocation of v- Src, and also c-Src, in response to biological stimuli. Actin stress fibers and translocation of active v-Src to focal adhesions in quiescent Swiss 3T3 cells were restored by microinjection of activated Rho A and by serum. Double labeling with anti-Src and phalloidin demonstrated that v-Src localized along the reformed actin filaments in a pattern that would be consistent with trafficking in complexes along the stress fibers to focal adhesions. Furthermore, treatment with the actin-disrupting drug cytochalasin D, but not the microtubule- disrupting drug nocodazole, prevented v-Src translocation. In addition to v-Src, we observed that PDGF-induced, Rac-mediated membrane ruffling was accompanied by translocation of c-Src from the cytoplasm to the plasma membrane, an effect that was also blocked by cytochalasin D. Thus, we conclude that translocation of Src from its site of synthesis to its site of action at the cell membrane requires an intact cytoskeletal network and that the small G proteins of the Rho family may specify the peripheral localization in focal adhesions or along the membrane, mediated by their effects on the cytoskeleton.