Rac1 Function Is Required for Src-induced Transformation

The proto-oncogene c-Src has been implicated in the development and progression of a number of human cancers including those of colon and breast. Accumulating evidence indicates that activated alleles of Src may induce cell transformation through Ras-ERK-dependent and -independent pathways. Here we show that Rac1 activity is strongly elevated in Src-transformed cells and that this small G protein is a critical component of the pathway connecting oncogenic Src with cell transformation. We further show that Vav2 and the ubiquitously expressed Rac1 guanine nucleotide exchange factor Tiam1 are phosphorylated in tyrosine residues in cells transfected with active and oncogenic Src. Moreover, phosphorylation of Tiam1 in cells treated with pervanadate, a potent inhibitor of tyrosine phosphatases, was partially inhibited by the Src inhibitor SU6656. Using truncated mutants of Tiam1, we demonstrate that multiple sites can be tyrosine-phosphorylated by Src. Furthermore, Tiam1 cooperated with Src to induce activation of Rac1 in vivo and the formation of membrane ruffles. Similarly, activation of JNK and the c-jun promoter by Src were also potently increased by Tiam1. Together, these results suggest that Vav2 and Tiam1 may act as downstream effectors of Src, thereby regulating Rac1-dependent pathways that participate in Src-induced cell transformation.

The cytoplasmic tyrosine kinase Src was initially identified as the protein product of the first discovered viral oncogene, v-src, which is responsible for the transforming ability of the Rous sarcoma virus (1)(2)(3). Its normal cellular counterpart, the product of the proto-oncogene c-src (4,5), was later found to play a key role in signal transduction processes by acting downstream from a large variety of cell surface receptors (reviewed in Ref. 6). c-Src was also found to be overexpressed and highly active in a number of distinct tumor types, including those affecting the colon, breast, pancreas, and central nervous system, where it may play an important role in tumor development and in the acquisition of the metastatic phenotype (reviewed in Ref. 7).
The molecular mechanisms underlying the transforming potential of v-Src have been extensively explored. In particular, expression of oncogenic and activated forms of Src results in tyrosine phosphorylation of numerous intracellular substrates and the consequent activation of multiple signaling pathways that ultimately control cell proliferation. For example, active Src can potently stimulate the Ras-ERK 1 pathway, primarily by the tyrosine phosphorylation of Shc by Src, followed by the recruitment to the membrane of the adapter protein Grb2 and the Ras-guanine nucleotide exchange factor Sos, thereby promoting Ras activation (8). However, several studies have indicated that Src can also utilize biochemical routes distinct from the Ras-ERK pathway to promote its biological responses (9 -11). Alternative pathways activated by Src may contribute to the activation of NF-B, STAT3, and E2F that can then stimulate the expression of several genes implicated in cell cycle control, such as cyclin D and c-Myc (12)(13)(14)(15)(16). Of interest, the latter has been shown to be dependent on the activation of a GTP-binding protein of the Rho family, Rac1, but not on the Ras-ERK pathway (10).
Rho GTPases, including RhoA, Rac1, and Cdc42, can control transcriptional events in the nucleus in addition to their effects on the organization of the cellular cytoskeleton (17)(18)(19). In particular, Rac1 has emerged as a key upstream activator of multiple signaling pathways regulating gene expression. For example, Rac1 regulates the transcriptional activity of c-Jun through the JNK pathway (19,20) and the serum-responsive element through the transcriptional activation of the serum response factor (17). As for other small GTPases, the functional activity of Rac1 is tightly regulated in vivo. Guanine nucleotide exchange factors (GEFs) activate these GTPases by promoting the exchange of their inactive GDP-bound forms to their active GTP-bound species. Most known GEFs for these proteins share a highly related structural domain of about 250 amino acids, termed the Dbl homology (DH) domain, which is adjacent to a pleckstrin homology (PH) domain (reviewed in Ref. 21). The biochemical connection between receptor and nonreceptor tyrosine kinases and the activation of Rac1 involves the activation of Rac GEFs and/or the inhibition of GTPase-activating proteins acting on Rac1. For example, we have shown that the tyrosine phosphorylation of the Rac1 exchange factor Vav1, which is predominantly expressed in hematopoietic cells, leads to the activation of the GDP/GTP exchange activity of Rac1 (22,23). Interestingly, the exchange activity of Vav2, a more widely expressed member of the Vav family, is also increased upon tyrosine phosphorylation (24), suggesting that this GEF may be involved in Rac1 activation by Src kinases in nonhematopoietic cells. However, whether Rac1 activation plays a role in cell transformation by activated alleles of Src and, if so, the precise nature of the Rac GEFs involved are both still poorly defined.
In this study, we first examined the role of Rac1 activation in Src-induced transformation. We found that v-Src-transformed NIH 3T3 cells display remarkably high levels of GTP-bound Rac1 and that molecules interfering with the activation of Rac1 or the ability of Rac1 to induce downstream pathways potently inhibited the focus forming activity of v-Src. In search for the underlying mechanism leading to Rac1 activation, we found that endogenous Vav2, which is limitedly expressed in NIH 3T3 cells, is tyrosine-phosphorylated in v-Src transformed cells. In addition, we observed that Tiam1, a specific Rac1-GEF that is highly expressed in NIH 3T3 cells, is also tyrosinephosphorylated in v-Src-transformed cells. Furthermore, we found that active Src strongly potentiates Tiam1-induced Rac1 activation and the consequent stimulation of cellular responses controlled by this small GTPase. These findings suggest that Tiam1 is a novel downstream effector of Src and that Src may utilize Tiam1 and Vav2 to regulate the activity of Rac1, thereby initiating the activation of a variety of cytoplasmic and nuclear events that are necessary for cellular transformation.

MATERIALS AND METHODS
Cell Lines and Transfections-NIH 3T3 fibroblasts were maintained in Dulbecco's modified Eagle's medium (Sigma) supplemented with 10% (v/v) calf serum (Bio Whittaker). To establish NIH 3T3 cells stably expressing v-Src (NIH/v-Src), the cells were transfected by the calciumphosphate precipitation technique with pCEFL-v-Src. The cells were selected in culture medium containing G418 (750 g/ml) for 3 weeks. Human embryonic kidney (HEK) 293T cells were maintained in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum. Transient transfections in HEK 293T cells were performed using the LipofectAMINE Plus reagent (Invitrogen) according to the manufacturer's instructions.
DNA Constructs-The expression vectors pSM-SrcYF (constitutively active) and pSM-SrcYF-KM (dominant negative) were kindly provided by H. Varmus. The plasmids pCEFL-SrcYF and pCEFL-SrcYF-KM were generated by cloning SrcYF and SrcYF-KM in the mammalian expression vector pCEFL. The coding sequence for the Cdc42/Rac1 interactive binding (CRIB) domain of the p21-activated serine/threonine kinase 1 (PAK1; amino acids 67-150) was amplified by PCR and subcloned into pGEX-4T3 in frame with glutathione S-transferase (GST). Full-length Tiam1 was cloned by PCR from murine brain cDNA (Clontech) and subcloned in the vector pCEFL with a C-terminal HA tag. Expression vectors for an N-terminal truncated form of Tiam1 lacking the first 390 amino acids (Tiam1⌬N) and a C-terminal fragment of Tiam1 corresponding to the last 580 amino acids (TiamC580) were kindly provided by J. Collard. The expression vector (pcDNAIII derivatives) for proto-Vav2 was previously reported (24). The plasmid pCEFL EGFP-Vav2 wild type was generated by cloning Vav2 wild type in pCEFL in frame with the C-terminal EGFP coding sequence. Constitutively activated small G proteins RhoA, Rac1, and Cdc42 were generated by replacing glutamine for leucine (QL mutants) in a position analogous to codon 61 in Ras and subcloned as AU5 epitope-tagged forms in pCEFL as previously reported (23). Dominant negative forms of Rac1 (Rac1 N17) and Cdc42 (Cdc42 N17) were obtained by replacing threonine 17 for asparagine in their corresponding coding sequence (19). Expression vectors for HA-tagged JNK and a plasmid encoding a luciferase gene driven by a wild type murine c-jun promoter, pJLuc, have been previously described (18).
Reagents-Human recombinant PDGF-BB (Intergen) was used at a final concentration of 25 ng/ml. The SU6656 and PP2 (Calbiochem) inhibitors were used at 2 M. Specific antibodies to HA and GFP were purchased from Covance. Polyclonal antibody against the C terminus (C-16) of Tiam1 was from Santa Cruz (sc-872). Polyclonal antibody against a unique acidic region of Vav2 was from Bustelo's lab (25).
Determination of Protein Tyrosine Phosphorylation-For prediction of potential sites for tyrosine phosphorylation by members of the Src family, amino acid sequences were analyzed using the Scansite (scansite.mit.edu) and the NetPhos 2.0 prediction servers (www.cbs.dtu.dk/ services/NetPhos). For experimental determination of tyrosine phosphorylation, the proteins were immunoprecipitated using specific antibodies, and the immunoprecipitates were subjected to Western blot analysis against phosphotyrosines using a combination of monoclonal antibodies from Upstate Biotechnology, Inc. (4G10) and Santa Cruz (PY-99). The proteins were visualized by enhanced chemiluminescence detection (Amersham Biosciences) using goat anti-mouse IgGs coupled to horseradish peroxidase as the secondary antibody (ICN).
In Vivo Rac1 Guanine Nucleotide Exchange Assay-In vivo Rac1 activity was assessed by a modified method described elsewhere (26,27). Briefly, the cells were lysed with ice-cold buffer containing 10 mM Tris, 100 mM NaCl, 1% Triton X-100, 0.5 mM EDTA, 40 mM ␤-glycerophosphate, 10 mM MgCl 2 , 1 mM Na 3 VO 4 , 10 g/ml aprotinin, 10 g/ml leupeptin, and 1 mM phenylmethylsulfonyl fluoride. The lysates were incubated for 30 min with a purified, bacterially expressed GST fusion protein containing the CRIB domain of PAK1, previously bound to glutathione-Sepharose beads (Amersham Biosciences), followed by three washes with lysis buffer. The GTP-bound forms of Rac1 associated with GST-CRIB were quantified by Western blot analysis using a monoclonal antibody against Rac1 (BD Transduction Laboratories).
Focus Forming Assays-NIH 3T3 cells were transfected by the calcium phosphate precipitation technique with different expression plasmids together with 1 g of pcDNAIII-␤-gal, a plasmid expressing the enzyme ␤-galactosidase, adjusting the total amount of plasmid DNA with empty vector. The day after transfection, the cells were washed in medium supplemented with 5% calf serum and then maintained in the same medium for 2-3 weeks. The foci were stained and scored as described (18). Duplicate plates were fixed with 1ϫ PBS solution containing 2% (v/v) formaldehyde and 0.2% (v/v) glutaraldehyde and stained at 37°C for ␤-galactosidase activity with a 1ϫ PBS solution containing 2 mM MgCl 2 , 5 mM K 3 Fe(CN) 6 , 5 mM K 4 Fe(CN) 6 , and 0.1% 5-bromo-4-chloro-3-indolyl-␤-D-galactopyranoside (X-gal) to evaluate the transfection efficiency.
JNK Kinase Assay-HEK 293T cells were seeded at 70 -80% confluence and transfected with Polyfect (Qiagen) with the expression vector for HA-tagged JNK1 alone or in combination with other plasmids. After transfection, the cells were cultured for 18 h and incubated in serumfree medium for 2 h. The cells were washed with cold PBS and lysed at 4°C in JNK lysis buffer containing 25 mM HEPES, pH 7.5, 0.3 M NaCl, 1.5 mM MgCl 2 , 0.2 mM EDTA, 0.5 mM dithiothreitol, 20 mM ␤-glycerophosphate, 1 mM vanadate, 1% Triton X-100, 1 mM phenylmethylsulfonyl fluoride, 20 g/ml aprotinin, and 20 g/ml leupeptin. HA-tagged JNK1 was immunoprecipitated from cleared lysates at 4°C for 2 h with anti-HA monoclonal antibody (Babco). Immunocomplexes were recovered with protein G-Sepharose (Amersham Biosciences). The beads were washed three times with ice-cold lysis buffer, once with 100 mM Tris, pH 7.5, 0.5 M LiCl, and once with kinase reaction buffer containing 12.5 mM MOPS, pH 7.5, 12.5 mM ␤-glycerophosphate, 7.5 mM MgCl 2 , 0.5 mM EGTA, 0.5 mM sodium fluoride, and 0.5 mM vanadate. The samples were resuspended in 30 l of kinase reaction buffer containing 10 Ci of [␥-32 P]ATP/reaction and 20 M of unlabeled ATP. 1 g of purified, bacterially expressed GST-c-Jun was used as substrate for the JNK in vitro kinase assay. After 20 min at 30°C, the reactions were terminated by the addition of 10 l of 5ϫ Laemmli buffer. The samples were analyzed by SDS gel electrophoresis on 12% acrylamide gels, and autoradiographies were performed with the aid of an intensifying screen.
Reporter Gene Assay-The plasmid encoding a luciferase gene driven by a wild type murine c-jun promoter (pJLuc) was kindly provided by R. Prywes. The cells were transfected with different expression plasmids together with 0.1 g of the pJLuc and 0.01 g of pRL-null (a plasmid expressing the luciferase from Renilla reniformis) as an internal control. In all cases, the total amount of plasmid DNA was adjusted with pcDNAIII-␤-gal. Firefly and Renilla luciferase activities present in cellular lysates were assayed using a dual luciferase reporter system (Promega), and light emission was quantitated using the Monolight 2010 luminometer as specified by the manufacturer (Analytical Luminescence Laboratory).
Fluorescence Microscopy-Porcine aortic endothelial (PAE) cells were grown in 24-well plates on coverslips and transfected using Fu-GENE 6 (Roche Applied Science) with the pCEFL-EGFP plasmid along with additional expression vectors. The Cells were serum-starved for 8 h and washed twice with PBS. The cells were then fixed with 2% paraformaldehyde (in 1ϫ PBS) for 20 min and permeabilized with 0.5% Triton X-100 (in 1ϫ PBS) for 10 min. The coverslips were blocked with 0.5% bovine serum albumin and incubated for 1 h with phalloidin conjugated to Texas Red. The coverslips were mounted and visualized using Axioplan2 microscope (Zeiss). The digital images were captured using Spotcam.

Rac1 Is Constitutively Active in v-Src Transformed Cells-
To examine the contribution of Rac1 in v-Src-mediated cell transformation, we first explored the status of Rac1 activity in v-Src-transformed cells. To this end, we established NIH 3T3 lines stably expressing v-Src (NIH/v-Src). As a control, these cells exhibited a substantially increased level of phosphotyrosine containing proteins, whose extent of phosphorylation was strongly reduced by the highly selective Src inhibitor SU6656 (28) (Fig. 1A). For the analysis of Rac1 activity in vivo, we used a GST fusion protein containing the CRIB domain of the PAK1, prebound to glutathione-Sepharose beads, to bind and affinity precipitate the activated, GTP-bound forms of Rac1. Using this assay, we observed remarkably high levels of Rac1-GTP in NIH/v-Src cells when compared with those in parental, nontransformed NIH 3T3 cells (Fig. 1B). Total Rac1 expression levels were comparable between these two cell populations, and although normal NIH 3T3 cells exhibited low levels of active Rac1 under basal conditions, the amount of active Rac1 in these cells was rapidly and transiently increased upon stimulation with PDGF, which served as a control (Fig.  1C). Thus, stable expression of v-Src results in a remarkable and persistent increase in the levels of active Rac1, comparable with that achievable by acute exposure of NIH 3T3 cells to polypeptide growth factors.
Transformation Induced by v-Src Is Inhibited by the Block-ade of the Rac1 Pathway-We next explored the contribution of Rac1 to the ability of v-Src to transform NIH 3T3 cells. In this regard, we noticed that v-Src-induced foci are morphologically distinct from the spread out foci induced by an activated form of an upstream kinase of ERK1/2, MEK1 EE, which are typical of other oncogenes stimulating the ERK pathway, such as ras and raf (29, 30) (Fig. 2). Indeed, v-Src-induced foci are highly compact and well delineated, resembling more closely the foci induced by members of the Rho GEF family and activated forms of Rho GTPases (30). When v-Src was cotransfected with expression vectors for either a dominant negative form of Rac1 (Rac1 N17), which interferes with Rac1 activation (31), or a GST fusion protein containing the CRIB domain of PAK (PAK-N), that prevents Rac1 function (32), the number of foci induced by v-Src were significantly reduced. In contrast, the number of foci induced by MEK1 EE was not affected by these Rac1interfering constructs. Together, these results suggest that Rac1 is an integral component of the transforming pathway utilized by v-Src.
Vav2 and Tiam1 Are Phosphorylated on Tyrosine Residues by Active Src and Endogenous Tyrosine Kinases-On the basis of the above observations, we next set out to further define the mechanism underlying Rac1 activation by Src. Because the ubiquitously expressed Vav2 can be regulated by tyrosine kinases (24), we confirmed that endogenous Vav2 was tyrosinephosphorylated in NIH/v-Src cells (Fig. 3A). However, the expression levels of this GEF for Rho GTPases are relatively low, and large amounts of cellular proteins need to be immunoprecipitated to detect Vav2 tyrosine phosphorylation. In addition, we have recently observed that dominant negative forms of Vav2 inhibit only partially the effects mediated by Src (10).  1. Rac1 is activated in v-Srctransformed NIH 3T3 cells. A, NIH 3T3 cells and NIH 3T3 cells stably transformed with v-Src (NIH/v-Src) were incubated in serum-free medium with vehicle (Me 2 SO) or SU6656 (2 M) for 3 h. The lysates were subjected to Western blot (WB) analysis with anti-phosphotyrosine (PY) antibody. B, NIH 3T3 and NIH/v-Src cells were serum-deprived for 3 h, and the cell lysates were incubated with GST-PAK-N for 30 min to affinity precipitate (AP) active Rac1. PAK-bound Rac1 and total Rac1 in the corresponding total lysates (TL) were analyzed by WB with a monoclonal antibody against Rac1. C, NIH 3T3 cells were incubated with serum-free medium for 3 h and treated with PDGF for the indicated times. The levels of active Rac1 and total Rac1 were analyzed as in B.
Thus, other Rac1 GEFs may also contribute to the activation of Rac1 by Src in addition to Vav2.
Among the several members of the growing superfamily of GEFs for Rho proteins, Tiam1 has been shown to have high specificity for Rac1 and to be widely expressed. Of interest, a detailed examination of the amino acid sequence of Tiam1 revealed the existence of potential consensus sites for phosphorylation by members of the Src family (scansite.mit.edu). Thus, we set out to determine whether Tiam1 was a Src effector leading to Rac1 activation. As a first step, we asked whether Src could induce Tiam1 phosphorylation on tyrosine residues. As shown in Fig. 3A, endogenous Tiam1 was clearly tyrosinephosphorylated in NIH/v-Src cells, as judged by its immunodetection by anti-phosphotyrosine-specific antibodies in Tiam1 immunoprecipitates. We next investigated whether an HAtagged Tiam1 can be phosphorylated in cells transiently transfected with an activated form of Src. As shown in Fig. 3B, immunoprecipitated Tiam1 was only slightly phosphorylated by overexpression of c-Src, whereas an active Src (SrcYF), in which Tyr 527 was mutated to Phe, induced a much stronger effect. Both c-Src and SrcYF also effectively tyrosine-phosphorylated ectopically expressed Vav2. To confirm that the kinase activity of Src was required for the tyrosine phosphorylation of Tiam1 and Vav2 by Src, we used an inactive form of SrcYF with an additional mutation that renders the kinase inactive (SrcYF-KM). Neither Tiam1 nor Vav2 were phosphorylated when cotransfected with SrcYF-KM. Together, these results indicate that Src can phosphorylate Tiam1 on tyrosine residues and that this effect is dependent on the kinase activity of Src.
Next, we took advantage of the observation that reactive oxygen species, such as H 2 O 2 , activate c-Src (33) to examine whether endogenous tyrosine kinases of the Src family were able to phosphorylate Tiam1. As shown in Fig. 4A, endogenous Tiam1 was tyrosine-phosphorylated upon H 2 O 2 treatment. In the presence of H 2 O 2 , orthovanadate is oxidized to pervanadate, which increases protein tyrosine phosphorylation because of its inhibitory effect on protein-tyrosine phosphatases (34,35). Interestingly, the combined use of low concentrations of H 2 O 2 and vanadate, which had no effect when added alone, induced a strong phosphorylation of Tiam1. As an approach to determine whether this phosphorylation was mediated, at least in part, by members of the Src family, the cells were preincubated with the Src inhibitors PP2 and SU6656, the latter being a newly characterized compound with a high specificity toward Src kinases (28). Both compounds partially inhibited the tyrosine phosphorylation of Tiam1 induced by pervanadate (Fig.  4B). Collectively, these results suggest that in addition to v-Src, endogenous tyrosine kinases of the Src family can phosphorylate the Rac1 GEF Tiam1.
Tiam1 exhibits a number of structural domains, an N-terminal PH domain, a Ras-binding domain, a PDZ domain, and a C-terminal DH/PH domain (reviewed in Ref. 21). To delineate the regions of Tiam1 that are phosphorylated by Src, we made use of various truncated mutants of Tiam1 and examined them for tyrosine phosphorylation in HEK 293T cells transfected with active Src. The expressed constructs included the first 750 amino acids and the N-terminal PH domain (TiamN750); a C-terminal fragment containing the rest of the molecule including the Ras-binding domain, PDZ, and DH/PH domain (TiamC841); and a construct that consisted of the last 580 amino acids, including the DH/PH domain (TiamC580). As shown in Fig. 5, the three fragments were all effectively phosphorylated in tyrosine residues by Src. These data indicated that Tiam1 contains multiple phosphor-acceptor sites that are susceptible to Src-induced tyrosine phosphorylation.
Src Potentiates Tiam1-induced Rac1 Activation and Promotes Rac1-dependent Signaling-We next asked whether Src is able to regulate Tiam1 function. First, we determined whether Src and Tiam1 could cooperate to stimulate the accumulation of activated, GTP-bound Rac1 in vivo. For these experiments, we expressed in HEK 293T cells limited amounts of cDNA for SrcYF and Tiam1, which induced only a slight activation of Rac1 when expressed alone. When Tiam1 was cotransfected with SrcYF, we detected a strong potentiation of Rac1 activation (Fig. 6). As expected, SrcYF also strongly potentiated the activation of Rac1 by Vav2.
Of interest, the stimulatory effect of active Src on the in vivo GEF activity of Tiam1 was also reflected in the ability of SrcYF to promote morphological and transcriptional events regulated by Rac1. As an approach to study the effects of Src on cell morphology, we used PAE cells, in which Rho GTPases promote characteristic morphological changes by controlling a variety of actin-based cytostructures (36). Indeed, as shown in Fig. 7, in these cells activated forms of RhoA (RhoAQL) induce the for- mation of stress fibers, whereas activated forms of Rac1 (Rac1QL) and Cdc42 (Cdc42QL) induce membrane ruffling and the formation of fillopodia, respectively. Expression of a limited amount of full-length Tiam1 in PAE cells failed to change the cellular morphology when compared with control cells transfected only with GFP, whereas limited amounts of SrcYF alone induced cell elongation and the formation of several membrane processes. Interestingly, when Tiam1 was coexpressed with SrcYF, ϳ40% of transfected PAE cells, as judged by GFP expression, exhibited lamellipodia and/or membrane ruffles, resembling the phenotype of the cells transfected with RacQL or Tiam1⌬N, an N-terminally truncated, active form of Tiam1 (37).
In addition to its ability to regulate the cytoskeleton, Rac1 is also involved in signaling pathways leading to nuclear events through the activation of JNK that subsequently leads to the stimulation of c-jun transcription (19). Thus, we tested the effects of the combination of Tiam1 and active Src on the activity of JNK in HEK 293T cells. For these experiments, the cells were cotransfected with a HA-tagged form of JNK1, and HA immunoprecipitates were subjected to an in vitro JNK assay using GST-c-Jun as a substrate. As shown in Fig. 8A, when expressed in limited amounts, Tiam1 and SrcYF alone induced a minimal activation of JNK, but coexpression of Tiam1 with SrcYF induced a robust stimulation of the activity of JNK, which was similar to that induced by Tiam1⌬N. This potentiation of JNK activation by SrcYF together with Tiam1 was abolished by a dominant negative form of Rac1 (Rac1 N17), whereas a dominant negative form of Cdc42 (Cdc42 N17) had only a limited effect (less than 20% of inhibition) (data not shown). This suggests that the activating effect on JNK exerted by SrcYF and Tiam1 is specifically mediated by Rac1.
Finally, to test whether the cooperation of Tiam1 and SrcYF could also have effects on gene expression regulation, we performed a gene reporter assay using NIH 3T3 cells transfected with the c-jun promoter followed by a luciferase gene. Consistent with the activation of the JNK pathway, the combination of both Tiam1 and SrcYF induced a strong potentiation of the activation of the c-jun promoter (Fig. 8B). Together, our results indicate that SrcYF can cooperate with Tiam1 to induce the activation of Rac1 and the consequent stimulation of downstream signaling pathways regulated by this small GTPase. DISCUSSION Although activated Src can stimulate the Ras-ERK pathway (8), cells transformed by v-src often exhibit only a limited increase in the levels of active Ras and ERK1/2 (9,11), which suggests that this potent oncogene may utilize signaling routes in addition to or other than the ERK pathway to exert its remarkable biological activity. Indeed, Src can stimulate c-myc expression in a Ras-ERK-independent manner (10), a process that is required for Src-induced transformation (38). Other members of the mitogen-activated protein kinase superfamily, such as the c-Jun N-terminal kinase (JNK) and p38, have been also reported to be highly activated in v-Src-transformed NIH 3T3 fibroblasts and to play a major role in the phosphorylation and activation of STAT3, a transcription factor that is also required for Src-transformation (9). In addition, induction of cyclin D expression by Src through a network of kinases including ERK, JNK, and p38 has been shown to be required for G 1 phase cell cycle progression in Src transformed cells (15). This body of evidence clearly shows that oncogenic and active Src induce cell transformation through the activation of a complex array of signaling pathways.
Of interest, many of the biochemical routes stimulated by Src overlap with those regulated by Rac1. In fact, this small GTPase of the Rho family is required for the ability of activated Src to stimulate c-myc expression (10). In line with this observation, we found that cells transformed by v-src exhibit remarkable increased levels of GTP-bound Rac1 in vivo, which is aligned with the recently reported increase in Rac-GTP levels in neuroretina cells expressing a temperature-sensitive mutant of v-Src (11). These observations may help explain why v-srcinduced foci resemble more closely the punctuated morphology of foci induced by constitutively activated Rho GTPases and active guanine-nucleotide exchange factors rather than the more diffuse morphology displayed typically by Ras and the active forms of Raf and MEK (29,30). Furthermore, inhibition of Rac1 function dramatically diminished the ability of v-src to transform NIH 3T3 cells, collectively supporting that this GTPase is an integral component of the transforming pathway elicited by v-src.
Several GEFs for small GTPases have been found to be regulated by tyrosine phosphorylation, including the three members of the Vav family, proto-Dbl, Ras-GEF, and the RGScontaining GEFs for Rho PDZ-RhoGEF and LARG (22-24, 39 -41). Among them, Vav1 is the best example of a GEF that is tightly regulated by tyrosine phosphorylation (22,23). However, Vav1 is predominantly expressed in hematopoietic cells (42). Vav2, a protein structurally related to Vav1, exhibits enhanced GEF activity toward Rac1, RhoA, RhoG, and Cdc42 (24,43,44) and can also be regulated by tyrosine phosphorylation (24). Because Vav2 is ubiquitously expressed, this GEF represents a likely candidate to mediate the activation of Rho GTPases by Src (42). However, we have recently observed that the ability of Src to stimulate c-myc expression by a Rac1-dependent pathway in NIH 3T3 cells can be only partially explained by the activation of Vav2 (10). Indeed, expression of Vav2 in NIH 3T3 cells is limited, and we and others have shown that tyrosine phosphorylation of endogenous Vav2 is detectable in NIH 3T3 cells exposed to PDGF treatment upon immunoprecipitation of large amounts of cellular proteins (10,25).
The family of Rac1 GEFs has extended dramatically in the last years (reviewed in Ref. 21). Among them, Tiam1 is one of the GEFs exhibiting the highest specificity for Rac1. Tiam1 is widely expressed, and detectable expression levels of Tiam1 are found in cell lysates from many cell types tested, including HEK 293T, NIH 3T3, HaCat, and primary cultures of glial cells (results not shown). Thus, our results showing that v-Src and active c-Src can phosphorylate and potentiate Tiam1-induced Rac1 activity may explain the activation of Rac1 by Src in several tissues and cell lines. In addition, the fact that oncogenic Src phosphorylates Tiam1 suggests that tyrosine phosphorylation may be a novel mechanism of regulation of this GEF. Thus far, Tiam1 has been shown to be regulated by phosphorylation in threonine residues by protein kinase C and Ca 2ϩ /calmodulin-dependent protein kinase II upon agonist stimulation (45,46). Similarly, Tiam1 has also been shown to be phosphorylated in breast cancer cells upon stimulation with heregulin-␤1, although in this study the nature of the phosphorylated residues was not addressed (47).
These observations raise the possibility that cell surface receptors that stimulate cellular Src can use Tiam1 to promote the activity of Rac. However, the nature of the activating mechanism(s) acting upstream from Src can determine the choice of the Src downstream targets. For example, phosphorylation of Vav2 by Src is a critical component of the pathway by which PGDF receptors activate Rac1 (10), but unlike Vav2, Tiam1 is not tyrosine-phosphorylated in response to PDGF in NIH 3T3 cells (results not shown). Conversely, Vav2 is not tyrosinephosphorylated upon integrin activation, despite the ability of integrin receptors to potently activate Src family tyrosine kinases (44). Whether integrins instead activate Tiam1 through Src is under investigation. Similarly, further work may be required to elucidate which of the many intracellular or extracellular initiated pathways that converge to stimulate the activity of cellular Src kinases (8) may utilize Tiam1 to initiate the activation of Rac1 and its downstream signaling routes.
On the other hand, Tiam1 was identified in a search of genes implicated in the process of invasion and metastasis (48) and harbors oncogenic activity (37). It is interesting that Tiam1 has been shown to interact with a number of regulatory molecules that may be important for the metastatic and tumorigenic effects of this GEF. For instance, Ras can interact with Tiam1 through the Ras-binding domain, thereby increasing its Rac1 exchange ability, a process that is required for Ras-induced tumorigenesis (49). It has also been suggested that the interaction of Tiam1 with ankyrins plays a pivotal role in regulating Rac1 signaling and cytoskeletal functions required for the oncogenic and metastatic effects of Tiam1 in breast cancer progression (50). In this context, our results showing that Tiam1 can be regulated by v-Src may be relevant for the ability oncogenic Src to stimulate Rac1-dependent pathways that lead to cell transformation.
The emerging picture from this study is that Vav2 and Tiam1, the latter a widely expressed Rac1 guanine nucleotide exchange factor, are tyrosine-phosphorylated in Src-transformed cells and that activated Src potentiates the stimulation of Rac1 by Tiam1, thereby activating signaling pathways reg-ulated by this small GTPase. Thus, Tiam1 represents a novel downstream effector of Src, which may be relevant for the ability of Src to stimulate signaling pathways through Rac1. In turn, how tyrosine phosphorylation of Tiam1 contributes to the activation of its GEF activity in vivo is at the present unknown. Tyrosine phosphorylation of Tiam1 may control the activity of its regulatory proteins or, alternatively, may activate the intrinsic catalytic exchange activity of this GEF. We can envision that future studies exploring the mechanism by which Src-dependent tyrosine phosphorylation of Tiam1 affects its activity will provide further insights into how oncogenic alleles of Src activate Rac1, thereby promoting cell transformation.