Portal de Periódicos da CAPES

Dok-R plays a pivotal role in angiopoietin-1-dependent cell migration through recruitment and activation of Pak

2001; Springer Nature; Volume: 20; Issue: 21 Linguagem: Inglês

10.1093/emboj/20.21.5919

1460-2075

Zubin Master, Nina Jones, Jennifer M. Tran, J. V. Jones, Robert S. Kerbel, Daniel Dumont,

Galectins and Cancer Biology

Article1 November 2001free access Dok-R plays a pivotal role in angiopoietin-1-dependent cell migration through recruitment and activation of Pak Zubin Master Zubin Master Division of Molecular and Cellular Biology, Sunnybrook and Women's College Health Sciences Centre, 2075 Bayview Avenue, Research Building, S-wing, Room 227, Toronto, Ontario, M4N 3M5 Canada Department of Medical Biophysics, University of Toronto, Ontario, M5G 2M9 Canada Search for more papers by this author Nina Jones Nina Jones Samuel Lunenfeld Research Institute, Mount Sinai Hospital, 600 University Avenue, Toronto, Ontario, M5G 1X5 Canada Search for more papers by this author Jennifer Tran Jennifer Tran Division of Molecular and Cellular Biology, Sunnybrook and Women's College Health Sciences Centre, 2075 Bayview Avenue, Research Building, S-wing, Room 227, Toronto, Ontario, M4N 3M5 Canada Department of Laboratory Medicine and Pathobiology, University of Toronto, Ontario, M5G 1L5 Canada Search for more papers by this author Jamie Jones Jamie Jones Division of Molecular and Cellular Biology, Sunnybrook and Women's College Health Sciences Centre, 2075 Bayview Avenue, Research Building, S-wing, Room 227, Toronto, Ontario, M4N 3M5 Canada Search for more papers by this author Robert S. Kerbel Robert S. Kerbel Division of Molecular and Cellular Biology, Sunnybrook and Women's College Health Sciences Centre, 2075 Bayview Avenue, Research Building, S-wing, Room 227, Toronto, Ontario, M4N 3M5 Canada Department of Medical Biophysics, University of Toronto, Ontario, M5G 2M9 Canada Department of Laboratory Medicine and Pathobiology, University of Toronto, Ontario, M5G 1L5 Canada Search for more papers by this author Daniel J. Dumont Corresponding Author Daniel J. Dumont Division of Molecular and Cellular Biology, Sunnybrook and Women's College Health Sciences Centre, 2075 Bayview Avenue, Research Building, S-wing, Room 227, Toronto, Ontario, M4N 3M5 Canada Department of Medical Biophysics, University of Toronto, Ontario, M5G 2M9 Canada Toronto-Sunnybrook Regional Cancer Center, Toronto, Ontario, Canada, M4N 3M5 Search for more papers by this author Zubin Master Zubin Master Division of Molecular and Cellular Biology, Sunnybrook and Women's College Health Sciences Centre, 2075 Bayview Avenue, Research Building, S-wing, Room 227, Toronto, Ontario, M4N 3M5 Canada Department of Medical Biophysics, University of Toronto, Ontario, M5G 2M9 Canada Search for more papers by this author Nina Jones Nina Jones Samuel Lunenfeld Research Institute, Mount Sinai Hospital, 600 University Avenue, Toronto, Ontario, M5G 1X5 Canada Search for more papers by this author Jennifer Tran Jennifer Tran Division of Molecular and Cellular Biology, Sunnybrook and Women's College Health Sciences Centre, 2075 Bayview Avenue, Research Building, S-wing, Room 227, Toronto, Ontario, M4N 3M5 Canada Department of Laboratory Medicine and Pathobiology, University of Toronto, Ontario, M5G 1L5 Canada Search for more papers by this author Jamie Jones Jamie Jones Division of Molecular and Cellular Biology, Sunnybrook and Women's College Health Sciences Centre, 2075 Bayview Avenue, Research Building, S-wing, Room 227, Toronto, Ontario, M4N 3M5 Canada Search for more papers by this author Robert S. Kerbel Robert S. Kerbel Division of Molecular and Cellular Biology, Sunnybrook and Women's College Health Sciences Centre, 2075 Bayview Avenue, Research Building, S-wing, Room 227, Toronto, Ontario, M4N 3M5 Canada Department of Medical Biophysics, University of Toronto, Ontario, M5G 2M9 Canada Department of Laboratory Medicine and Pathobiology, University of Toronto, Ontario, M5G 1L5 Canada Search for more papers by this author Daniel J. Dumont Corresponding Author Daniel J. Dumont Division of Molecular and Cellular Biology, Sunnybrook and Women's College Health Sciences Centre, 2075 Bayview Avenue, Research Building, S-wing, Room 227, Toronto, Ontario, M4N 3M5 Canada Department of Medical Biophysics, University of Toronto, Ontario, M5G 2M9 Canada Toronto-Sunnybrook Regional Cancer Center, Toronto, Ontario, Canada, M4N 3M5 Search for more papers by this author Author Information Zubin Master1,2, Nina Jones3, Jennifer Tran1,4, Jamie Jones1, Robert S. Kerbel1,2,4 and Daniel J. Dumont 1,2,5 1Division of Molecular and Cellular Biology, Sunnybrook and Women's College Health Sciences Centre, 2075 Bayview Avenue, Research Building, S-wing, Room 227, Toronto, Ontario, M4N 3M5 Canada 2Department of Medical Biophysics, University of Toronto, Ontario, M5G 2M9 Canada 3Samuel Lunenfeld Research Institute, Mount Sinai Hospital, 600 University Avenue, Toronto, Ontario, M5G 1X5 Canada 4Department of Laboratory Medicine and Pathobiology, University of Toronto, Ontario, M5G 1L5 Canada 5Toronto-Sunnybrook Regional Cancer Center, Toronto, Ontario, Canada, M4N 3M5 *Corresponding author. Division of Molecular and Cellular Biology, Sunnybrook and Women's College Health Sciences Centre, 2075 Bayview Avenue, Research Building, S-wing, Room 227, Toronto, Ontario, Canada M4N 3M5, E-mail: [email protected] The EMBO Journal (2001)20:5919-5928https://doi.org/10.1093/emboj/20.21.5919 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Tek/Tie-2 is an endothelial cell (EC)-specific receptor tyrosine kinase that plays a critical role in angiogenesis via its regulation by the angiopoietin family of growth factor ligands. Angiopoietin-1 (Ang1) can promote EC migration; however, the signaling mechanisms underlying this process remain elusive. Here we demonstrate that Dok-R/Dok-2 can associate with Tek in ECs following Ang1 stimulation, resulting in tyrosine phosphorylation of Dok-R and the subsequent recruitment of Nck and the p21-activating kinase (Pak/Pak1) to the activated receptor. Ang1-mediated migration is increased upon Dok-R overexpression and this requires a functional Nck binding site on Dok-R. Localization of this Dok-R–Nck–Pak complex to the activated Tek receptor at the cellular membrane is coincident with activation of Pak kinase. The ability of Dok-R to bind Nck is required for maximal activation of Pak and overexpression of Pak results in increased Ang1-mediated cell motility. Our study outlines a novel signaling pathway underlying Ang1-driven cell migration that involves Dok-R and its recruitment of Nck and the subsequent activation of Pak. Introduction Angiogenesis, the process by which new blood vessels arise from the pre-existing vasculature, occurs during embryonic vascular development as well as in normal physiological and pathological growth in adults. In the adult, sprouting angiogenesis is the most common type of vascular growth and it is characterized by the activation of vascular endothelial cells (ECs), which are induced by various angiogenic stimuli to undergo several distinct cellular processes such as proliferation, migration and cytoskeletal reorganization. Collectively, these events are crucial for proper establishment of a vascular network (Klagsbrun and Moses, 1999). The organization and remodeling of blood vessels is a tightly regulated process controlled in part by paracrine signals, many of which are initiated by binding of growth factor ligands to their cognate transmembrane receptor tyrosine kinases (RTKs) expressed on the surface of ECs. Ligand binding leads to RTK activation and autophosphorylation. The vascular endothelial growth factor (VEGF) receptors (VEGFRs) and the TIE receptors are two such subfamilies of RTKs that are expressed almost exclusively on ECs (Mustonen and Alitalo, 1995). Much of our understanding of the importance of these RTKs in angiogenesis has stemmed from genetic experiments, which have clearly demonstrated a role for signaling pathways mediated through these receptors in discrete phases of mouse vascular development. While the ligand VEGF and its receptors, VEGFR-1 and VEGFR-2, are required for the establishment of a primitive vascular network, the angiopoietins and the TIE receptors, Tie/Tie-1 and Tek/Tie-2, play a subsequent role in the remodeling and stabilization of the vasculature once it has been specified and patterned (reviewed in Jones et al., 2001). The angiopoietins are a new family of growth factor ligands that bind specifically to the Tek receptor. To date, four angiopoietins (Ang1–4) bind Tek and behave as either agonists (Ang1 and Ang4) or context-dependent antagonists (Ang2 and Ang3) of Tek kinase activity (Davis et al., 1996; Maisonpierre et al., 1997; Valenzuela et al., 1999). Mice engineered to lack Ang1 or Tek or to overexpress the Tek antagonistic ligand, Ang2, die in utero due to vascular hemorrhaging, defects in cardiac development, capillary sprouting and the proper patterning of vessels (Dumont et al., 1994; Sato et al., 1995; Suri et al., 1996; Maisonpierre et al., 1997). Such genetic studies of Ang1/Tek signaling have been further complemented by in vitro experiments, which have demonstrated that Ang1 can induce EC sprout formation, motility and can stabilize tubule structures formed in collagen matrices (Koblizek et al., 1998; Witzenbichler et al., 1998; Fujikawa et al., 1999; Hayes et al., 1999; Jones et al., 1999). These reports have clearly established the necessity of Tek signaling for proper formation and maturation of the vasculature during embryogenesis. Although these findings suggest a role for Ang1 signaling in EC migration and angiogenic sprout formation, the molecular mechanisms by which these processes occur have yet to be defined. Ang1 binding to Tek results in tyrosine phosphorylation of the receptor and thus provides high affinity binding sites for downstream adapter and docking proteins that are subsequently recruited to the receptor via their Src homology 2 (SH2) or phosphotyrosine binding (PTB) domains. Two such proteins that are recruited to the Tek receptor in a phosphotyrosine-dependent manner include the regulatory subunit of phosphatidylinositol (PI) 3-kinase, p85, and the docking protein Dok-R (Jones and Dumont, 1998; Kontos et al., 1998; Jones et al., 1999). Recruitment of p85 to tyrosine phosphorylated Tek has been shown to cause activation of PI3-kinase, which regulates both EC survival and motility (Kontos et al., 1998; Witzenbichler et al., 1998; Fujikawa et al., 1999; Jones et al., 1999; Papapetropoulos et al., 2000). Interestingly, inhibition of PI3-kinase activity by pharmacological agents only partially decreases Ang1-mediated cell migration, suggesting that an alternative PI3-kinase-independent pathway may regulate motility downstream of the Tek receptor (Jones et al., 1999). Dok-R (also known as Dok-2 and FRIP) is a member of a new class of docking proteins, termed the DOK family, that includes p62 (Dok-1), Dok-L (Dok-3) and possibly IRS-3 (Carpino et al., 1997; Lavan et al., 1997; Yamanashi and Baltimore, 1997; Nelms et al., 1998; Cong et al., 1999; Lemay et al., 2000). Structurally, the DOK family members are characterized by an N-terminal pleckstrin homology (PH) domain followed by a central PTB domain and a proline- and tyrosine-rich C-terminal tail. Dok-R is recruited to activated Tek via its PTB domain, which results in its subsequent tyrosine phosphorylation, thereby establishing binding sites for the small GTPase-activating protein for Ras, p120RasGAP (RasGAP) and the adapter protein Nck (Jones and Dumont, 1998). In addition, Dok-R can constitutively interact with the adapter protein Crk (Jones and Dumont, 1998). Both RasGAP and Nck play a role in cell adhesion, motility and actin polymerization (McGlade et al., 1993; Ren et al., 1993; Noguchi et al., 1999; Chen et al., 2000; Kulkarni et al., 2000; Howe, 2001), which implies that Dok-R may serve a scaffolding role for proteins that modulate migration. Recently, a role for Dok in cellular migration had been proposed by Noguchi et al. (1999), as overexpression of Dok potentiates cellular motility in response to insulin in an Nck-dependent fashion. Furthermore, Nck associates with a host of effector proteins involved in cytoskeletal rearrangement and motility such as the serine/threonine kinase Pak (Bokoch et al., 1996; Kiosses et al., 1999; Sells et al., 1999). Interestingly, VEGF-mediated assembly of focal adhesion complexes and migration has been shown to require the interaction of Nck with Pak and subsequent Pak activation (Stoletov et al., 2001). An equivalent Nck binding site to that on Dok has also been identified on Dok-R, suggesting a potential role for Dok-R in cell migration mediated through Nck. We wanted to examine whether Dok-R could potentially modulate an alternative, or PI3-kinase-independent, pathway for Ang1-mediated migration downstream of the Tek receptor (Jones et al., 1999). We demonstrate that Ang1 stimulation of ECs causes the phosphotyrosine-dependent binding of Dok-R to Tek in vivo and the subsequent tyrosine phosphorylation of Dok-R. Once phosphorylated, Dok-R is able to recruit Nck in ECs, which leads to tyrosine phosphorylation of Nck. To investigate the role of Dok-R in Ang1-mediated migration, we engineered several point mutations in Dok-R, which abrogate its ability to associate with either Nck or RasGAP. Overexpression of wild-type Dok-R potentiates Ang1-mediated cell motility, which is dependent upon phosphorylation of tyrosine (Y) 351 of Dok-R since mutation of Y351 to phenylalanine abolishes the ability of Dok-R to interact with Nck (Jones and Dumont, 1999). We further demonstrate that the Nck binding protein, Pak, is recruited to Tek at the cell membrane and that this recruitment results in an Ang1-dependent Pak kinase activation in ECs. Furthermore, we demonstrate that Pak kinase activity is dependent on phosphorylation of Y351 on Dok-R. Moreover, we substantiate a role for Pak in Ang1-driven migration by demonstrating that its overexpression potentiates Ang1-mediated motility and requires phosphorylation of Dok-R on Y351. Our results demonstrate a novel pathway underlying Ang1/Tek-mediated migration, which requires Dok-R phosphorylation and recruitment of an Nck–Pak complex at the membrane, which leads to Pak activation in ECs. Results Angiopoietin-1 stimulation induces Dok-R tyrosine phosphorylation and migration of ECs SVR ECs express relatively high levels of Tek and Dok-R (Jones and Dumont, 1998 and data not shown). To assess whether Dok-R could associate with Tek and become tyrosine phosphorylated in response to Ang1 stimulation of ECs, we stimulated SVRs with conditioned medium containing Ang1 (Jones et al., 1999; Teichert-Kuliszewska et al., 2001). Briefly, conditioned medium collected from HEK 293T cells stably expressing a form of Ang1 fused to the Myc and His epitope tags, herein referred to as Ang1-MH, was used as stimulant while conditioned medium harvested from the untransfected parental cell line, referred to as Mock, was used as the control stimulant. Treatment of SVR ECs with Ang1-MH stimulated the tyrosine phosphorylation of Tek and its co-immunoprecipitation with Dok-R (Figure 1A), which is the first documentation that this interaction occurs in ECs. Furthermore, Ang1-MH stimulation of SVRs also resulted in the tyrosine phosphorylation of Nck and its co-immunoprecipitation with phosphorylated Dok-R (Figure 1B). The increased amount of phosphorylated Nck found in Dok-R immunoprecipitates relative to Nck immunoprecipitates suggests that most of the Nck associated with Dok-R is tyrosine phosphorylated. Figure 1.Ang1-MH stimulation induces EC migration and tyrosine phosphorylation and association of a Tek/Dok-R–Nck complex. (A) DMEM alone (−) or Ang1-MH (+)-conditioned medium was used to stimulate SVR ECs for 10 min in the absence of phosphatase inhibitors. Proteins from Tek immunoprecipitates were separated by SDS–PAGE and immunoblotted with antibodies recognizing phosphotyrosine (pY), Tek and Dok-R, which revealed that Ang1-MH can stimulate the tyrosine phosphorylation of Tek and the co-immunoprecipitation of tyrosine phosphorylated Dok-R. Both lysates contained equal concentrations of Tek receptor. (B) Ang1-MH (+) stimulation of SVR ECs results in the tyrosine phosphorylation of Dok-R and co-immunoprecipitation of tyrosine phosphorylated Tek and Nck when compared with DMEM-stimulated lysates (−). Immunoprecipitations of Dok-R (first two lanes, anti-Dok-R blot) and Nck (last two lanes, anti-Nck blot) with their respective antibodies revealed that equal amounts of protein were immunoprecipitated from both stimulated (+) and unstimulated (−) lysates. Immunoprecipitation of tyrosine phosphorylated Nck from Ang1-MH-stimulated lysates could co-immunoprecipitate more tyrosine phosphorylated Dok-R than unstimulated (−) lysates (last two lanes). (C) SVR ECs were seeded in the top chamber of a modified Boyden chamber while chemoattractants (Mock, Ang1-MH or 50 ng/ml VEGF) were placed in the bottom well. Stimulation with Ang1-MH and VEGF resulted in a 5-fold increase in migration over Mock-stimulated cells. Data points are represented as the number of migrated cells per field. All experiments were performed in triplicate and differences between Mock-stimulated and ligand-stimulated migration values were statistically significant (P <0.05). Download figure Download PowerPoint We have shown previously that Ang1-MH functions as a potent chemoattractant for Py4-1 and human umbilical vein ECs (Jones et al., 1999). Thus, we set out to determine whether SVRs would also respond in a similar fashion. Using a modified Boyden chamber motility assay, SVR cells migrated as robustly with Ang1-MH treatment as that seen with VEGF when compared with Mock-treated cells (Figure 1B). This result was not anticipated since most other EC types tested generally display a 2-fold increase in Ang1-MH-mediated migration as compared with a 5-fold increase seen in VEGF-induced cell migration (Jones et al., 1999). This enhanced migration of SVRs may reflect the relatively high levels of Tek expression seen in SVRs (data not shown). SVR ECs respond to Ang1-MH stimulation by activating Tek, resulting in tyrosine phosphorylation of Dok-R and Nck and thus, serve as a model to further investigate the role of Dok-R and Nck in Ang1-mediated EC migration. Phosphorylation of Y351 on Dok-R is required for interaction with Nck We have shown that Ang1-MH stimulation results in the phosphorylation of Dok-R and its association with Nck. The SH2 domains of Nck and RasGAP are known to interact with phosphorylated tyrosine within the context YxxP (Songyang et al., 1993). We have previously generated point mutations in Dok-R that alter tyrosine residues that mediate binding to Nck (Y351F) and RasGAP (Y276,304F–DM) to phenylalanine (Jones and Dumont, 1999; see Figure 2A). To determine whether these point mutations in Dok-R affected its recruitment to activated Tek, we introduced Tek and either Dok-R, or one of its mutants, into HEK 293T cells. As reported previously, overexpression of Tek in HEK 293T cells results in activation of the receptor in an Ang1-independent context (Jones and Dumont, 1998; Jones et al., 1999). Immunoprecipitation of Tek results in its tyrosine phosphorylation and co-immunoprecipitation of Dok-R and its mutants (Figure 2B). Dok-R binding to activated Tek also results in tyrosine phosphorylation of Dok-R and mutant forms. Furthermore, co-immunoprecipitation experiments also revealed that Dok-R and Dok-RDM could interact with Nck, but that this binding was abolished in Dok-RY351F-transfected cells (Figure 2B). These results demonstrate that the differently mutagenized Dok-R polypeptides are efficiently recruited to Tek and that Y351 on Dok-R is required for downstream recruitment of Nck. Figure 2.Dok-R recruits Nck through tyrosine phosphorylation of Y351. (A) The positions of three tyrosine residues (276, 304 and 351) in Dok-R found within the context YxxP are indicated. Double mutations in Y276 and Y304 to phenylalanine (F) abolish Dok-R binding to RasGAP (Dok-RDM) while mutation of Y351 to F abrogates binding to Nck (Dok-RY351F). (B) Overexpression of Tek with Dok-R, or its mutants, in HEK 293T cells resulted in the tyrosine phosphorylation of Dok-R, Dok-RDM and Dok-RY351F and their co-immunoprecipitation with tyrosine phosphorylated Tek. Nck could co-immunoprecipitate with Dok-R or Dok-RDM, but this interaction was abrogated in Dok-RY351F-transfected cells. Equal amounts of Tek and Dok-R/mutant were observed in transfected cell lysates (CL). (C) Lysates prepared from Tek-Dok-R and Tek-Dok-RY351F cell lines stimulated with Mock or Ang1-MH for 1 h in the absence of phosphatase inhibitors were immunoprecipitated with anti-Tek antibodies. Ang1-MH can stimulate the tyrosine phosphorylation of Tek in stable cell lines and the co-immunoprecipitation of tyrosine phosphorylated Dok-R or its mutants. Mock stimulation of these cells resulted in basal levels of Tek tyrosine phosphorylation. Immunoprecipitation of Tek also resulted in the co-immunoprecipitation of Nck in Ang1-MH-stimulated lysates of Tek-Dok-R cells, but not Tek-Dok-RY351F mutant cells. Immunoblotting with anti-Tek antibodies revealed equal amounts of Tek protein expression in all lysates. Download figure Download PowerPoint In order to easily manipulate Dok-R signaling pathways, we utilized a 293Tek cell line that responds to Ang1-MH stimulation in a manner virtually identical to SVR and other ECs (Jones et al., 1999). We generated 293Tek cell lines stably expressing vector alone (Tek-Vec cells), Dok-R (Tek-Dok-R cells) or mutant forms of Dok-R unable to bind either Nck (Tek-Dok-RY351F cells) or RasGAP (Tek-Dok-RDM cells). Stimulation of these cell lines with Ang1-MH resulted in the tyrosine phosphorylation of wild-type Dok-R and its mutant forms, and their co-immunoprecipitation with activated Tek (Figure 2C and data not shown). Notably, a phosphoprotein similar to the size of Nck co-immunoprecipitates with Tek only from Ang1-MH-stimulated Tek-Dok-R lysates, but not from Tek-Dok-RY351F lysates (Figure 2C). These results illustrate that Ang1-MH can promote the tyrosine phosphorylation and association with the activated Tek receptor of Dok-R and its mutants. Phosphorylation of Y351 on Dok-R is required for Ang1-mediated migration Since Ang1-MH stimulation of ECs leads to the recruitment of Nck to Dok-R and both Nck and RasGAP have been shown to influence cellular migration, we thus set out to determine whether Dok-R can mediate Ang1-dependent cell migration. We chose several independent clones of Tek-Dok-R, Tek-Dok-RDM and Tek-Dok-RY351F stable cell lines that had comparable expression levels of Dok-R proteins (Figure 3A, inset). Ang1-MH stimulation of 293Tek cells stably expressing vector alone (Tek-Vec) produced a significant 3-fold increase in cell motility over stimulation with Mock (Figure 3A). Overexpression of Dok-R in Tek-Dok-R cells potentiated Ang1-mediated cell motility by almost 2-fold over Tek-Vec cells. This potentiation was not abrogated in Tek-Dok-RDM cells, whereas the Tek-Dok-RY351F cells exhibited a significant reduction in Ang1-mediated migration (Figure 3A). The presence of endogenous hDok-R in 293Tek cells (data not shown) may account for the ability of Dok-RY351F to interfere with normal hDok-R function in these cells. Importantly, we also tested whether basic fibroblast growth factor (bFGF) could affect cell migration and found that all of the 293Tek stable lines responded similarly to bFGF-stimulated chemotaxis (Figure 3A), which is consistent with our inability to detect Dok-R tyrosine phosphorylation following bFGF stimulation (data not shown). The ability of these cells to undergo similar levels of bFGF-dependent migration serves to illustrate that the lack of Ang1-dependent motility observed in Tek-Dok-RY351F cells is not due to a generalized inhibition of cell migration by the Dok-RY351F protein. Taken together, these results demonstrate that Dok-R mediates Ang1-dependent cell migration and that Y351 on Dok-R is required for this activity. Figure 3.Nck recruitment to Dok-R is required for Ang1-mediated migration. (A) Lysates prepared from Tek-Dok-R or mutant cell lines were subjected to immunoblot analysis using anti-HA antibodies that recognized transfected Dok-R/mutant (inset). Equal Dok-R-expressing clones of Tek-Vec, Tek-Dok-R, Tek-Dok-RDM and Tek-Dok-RY351F stable cell lines (inset) were placed in a migration assay as described previously using Mock, Ang1-MH or 30 ng/ml bFGF in Mock media as chemoattractants. Ang1-MH stimulation of Tek-Vec cells resulted in a 3-fold increase in migration over Mock, and expression of Dok-R or Dok-RDM potentiated this Ang1-dependent motility. However, Tek-Dok-RY351F cells displayed a dramatic reduction in migration. Stimulation of cells with bFGF resulted in relatively equal migration levels. All experiments were performed in triplicate using three different clones (data not shown). Migration differences between Mock and Ang1-MH stimulations of all cell lines, and of Ang1-MH stimulations of Tek-Dok-RY351F compared with Tek-Vec/Tek-Dok-R/Tek-Dok-RDM were all found to be statistically significant (P <0.05). (B) Tek-Dok-R and Tek-Dok-RY351F stable cell lines seeded on coverslips were stimulated with Mock- or Ang1-MH-conditioned media for 4 h and then immunostained using antibodies against actin. Mock-stimulated cells displayed a diffused dispersion of actin at the cell membrane with no membrane protrusions. However, cells stimulated with Ang1-MH displayed localized regions of actin and Tek-Dok-R cells displayed a greater frequency and extent of membrane projections (arrowheads) than did Tek-Dok-RY351F cells. Download figure Download PowerPoint Migrating cells form membrane extensions of localized actin at a focal point of the cell during directional chemotaxis, which facilitates movement by extending the cell body prior to contraction (Sheetz et al., 1998). To further confirm that Dok-R plays a role in Ang1-mediated migration using another biological assay, we chose to examine actin localization in Tek-Dok-R and Tek-Dok-RY351F cells. Mock-treated Tek-Dok-R and Tek-Dok-RY351F cells displayed non-localized actin that was dispersed throughout the cell membrane (Figure 3B). However, Ang1-MH stimulation resulted in the redistribution of actin at focal points on the cell membrane. Interestingly, the Tek-Dok-R cells displayed a greater number and more extensive membrane protrusions as compared with Tek-Dok-RY351F cells, suggesting that recruitment of Nck may be required for Dok-R to mediate cellular morphological changes in response to Ang1/Tek signaling (Figure 3B). These findings implicate Dok-R and Nck in rearrangements of the actin cytoskeleton to facilitate Ang1-mediated motility. Pak colocalizes with Tek through a Dok-R–Nck interaction Nck has previously been shown to constitutively associate with Pak whose activation has been implicated in cytoskeletal rearrangements leading to cell migration downstream of many growth factor receptors (Bokoch et al., 1996; Lu et al., 1997; Stoletov et al., 2001). To establish whether Pak could potentially contribute to Ang1-mediated cell migration upon recruitment to Dok-R, we first examined whether Pak could colocalize with Tek at the cell membrane. HEK 293T cells grown on coverslips were transfected with Tek and either Dok-R or Dok-RY351F under conditions where Tek is activated. Indirect immunofluorescence with antibodies directed against Tek, Dok-R and Pak was used to detect the intracellular distribution of these proteins in vivo. As anticipated, both Tek (red) and the different forms of Dok-R (green) were colocalized at the cell membrane as indicated by the merged image (Figure 4 and data not shown). To demonstrate that Tek and Dok-R colocalization requires an active Tek kinase, Dok-R and its mutants were co-transfected with a kinase-inactive Tek (TekK853A). A requirement for an active Tek kinase was demonstrated by a dramatic lack of Dok-R localization at the membrane as Dok-R (green) was found predominantly in the cytoplasm, which was evident upon merging the two images, Tek (orange) (Figure 4, inset). Somewhat similar to the pattern of Tek and Dok-R colocalization, an enrichment of Pak was localized at the membrane (white arrowheads) in Tek + Dok-R-transfected cells, which was completely absent in Tek + Dok-RY351F cells (Figure 4). It is important to note that not every cell was transfected here (asterisk) and since our transfected Dok-RY351F must first displace endogenous hDok-R, this may explain the small amount of Pak localization seen in some Tek + Dok-RY351F cells (merge). Figure 4.Pak colocalizes with Tek through a Dok-R–Nck interaction. Antibodies recognizing Tek, Dok-R and Pak were used to detect these proteins in HEK 293T cells overexpressing Tek with either Dok-R or Dok-RY351F by confocal microscopy. Cellular distributions of Tek (red) were found at the cell membrane and were colocalized with Dok-R/Dok-RY351F (green) as seen by the merged image (yellow). The colocalization seen between Tek and Dok-R, or its mutants, specifically required an active Tek kinase, since cells co-transfected with TekK853A and Dok-R display diffuse staining throughout the cytoplasm rather than at the membrane (inset). Similarly, in cells overexpressing activated Tek and Dok-R, Pak (green) was enriched at the cell membrane (white arrows), which colocalized with Tek, as indicated by the merged image. However, cells expressing Tek and Dok-RY351F did not have this enriched localization of Pak at the membrane and no colocalization with Tek can be seen (merge). Download figure Download PowerPoint Activated Tek promotes the formation of a Dok-R, Nck and Pak ternary com