HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints/Permissions Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Bonnefoy-Bérard, N. Articles by Altman, A. Search for Related Content PubMed PubMed Citation Articles by Bonnefoy-Bérard, N. Articles by Altman, A.

Vav: Function and Regulation in Hematopoietic Cell Signaling

^a INSERM U80, Hôpital E. Herriot, Lyon, France;
^b The La Jolla Institute for Allergy and Immunology, La Jolla, California, USA

Key Words. Vav • Hematopoietic cell • Antigen receptor • Signal transduction • Ras • Protein tyrosine kinase • Guanine nucleotide exchange • Dbl-homology

Dr. Amnon Altman, La Jolla Institute for Allergy and Immunology, 11149 North Torrey Pines Road, La Jolla, CA 92037, USA.

	Abstract

Top Abstract Introduction Isolation and Expression of... Domain Structure of Vav Phosphorylation of Vav Role of Vav in... Transforming Activity of vav... The Interaction of Vav... Future Perspectives: The Vav... References

 
Vav, a 95 kDa proto-oncogene product expressed specifically in hematopoietic cells, was originally isolated as a transforming human oncogene. Vav contains an array of functional domains that are involved in interactions with other proteins and, possibly, with lipids. These include, among others, a putative guanine nucleotide exchange domain, a cysteine-rich region similar to the phorbol ester/diacylglycerol-binding domain of protein kinase C, a pleckstrin-homology domain, and Src-homology 2 and 3 (SH2 and SH3, respectively) domains. The presence of these domains, the transforming activity of the vav oncogene, and the rapid increase in tyrosine phosphorylation of Vav induced by triggering of diverse receptors indicate that it plays an important role in hematopoietic cell signaling pathways. Such a role is supported by recent studies using "knockout" mice and transiently transfected T cells, in which Vav deletion or overexpression, respectively, had marked effects on lymphocyte development or activation. The presence of a putative guanine nucleotide exchange domain, the prototype of which is found in the dbl oncogene product, implies that Vav functions as a guanine nucleotide exchange factor (GEF) for one (or more) members of the Ras-like family of small GTP-binding proteins. In support of such a role, Vav preparations were found in some (but not other) studies to mediate in vitro-specific GEF activity for Ras. Additional studies are required to identify the physiological regulators and targets of Vav, and its exact role in hematopoietic cell development and signaling.

	Introduction

Top Abstract Introduction Isolation and Expression of... Domain Structure of Vav Phosphorylation of Vav Role of Vav in... Transforming Activity of vav... The Interaction of Vav... Future Perspectives: The Vav... References

 
Protein tyrosine kinases (PTKs) play critical roles in intracellular transmission of signals initiated by diverse cell surface receptors, which lead to cellular activation, differentiation and/or growth [1,2]. Rapid PTK activation and the subsequent phosphorylation of proteins on tyrosine promote the formation of membrane-bound multimeric protein complexes [1,3]. Protein-protein interactions within such complexes are mediated by conserved functional domains, e.g., Src-homology 2 and 3 (SH2 and SH3, respectively) domains [3–5]. These signaling complexes amplify the initial signal and transmit it to a cascade of intracellular serine/threonine kinases [6]. Stimulation of these enzymes causes activation of multiple transcription factors and, hence, induction of cell type-specific gene expression.

Primary cells and established lines of hematopoietic lineage have been widely used as models for analyzing the molecular basis of PTK-mediated and other signal transduction pathways. These cells can be activated via multiple, structurally well-defined surface receptors such as antigen-specific receptors on T [7–9] and B [10] lymphocytes, or widely expressed cytokine receptors [11]. While many of the proteins involved in PTK-coupled signaling, e.g., Src-family PTKs, phospholipase C- (PLC) and phosphatidylinositol 3-kinase (PI3-K), are expressed ubiquitously and are linked to many different receptors, others are expressed in a more restricted manner and are, therefore, thought to play more specialized roles in signaling pathways unique to given cell types. In hematopoietic cells, examples of the latter include isoforms of the CD45 phosphotyrosine phosphatase [12], the p70^zap/p72^syk PTKs [13–15] and Vav. The latter was originally identified as a transforming human oncogene [16], and has received considerable recent attention [17,18] after it was discovered that Vav contains several conserved signaling motifs, and is rapidly phosphorylated on tyrosine following ligation of various receptors. New information on Vav is rapidly being accumulated. This review summarizes current knowledge in this area and presents perspectives on its potential function(s) in the hematopoietic system.

	Isolation and Expression of Vav

Top Abstract Introduction Isolation and Expression of... Domain Structure of Vav Phosphorylation of Vav Role of Vav in... Transforming Activity of vav... The Interaction of Vav... Future Perspectives: The Vav... References

 
vav was identified as a transforming oncogene during gene transfer studies that involved transfection of murine NIH 3T3 fibroblasts with DNA from human carcinomas and tumorigenicity assays in athymic nude mice [16]. Analysis of cDNA clones corresponding to the vav oncogene and subsequent cloning of the complete human and murine proto-vav cDNA [16,19,20] revealed that 5' proto-vav sequences were replaced during the initial transfection studies by sequences of the cotransfected pSV2neo selection vector. A correction was later introduced in the original sequence when a murine proto-vav cDNA was subsequently isolated [20]. The corrected cDNA sequence of proto-vav predicts proteins of 845 (murine) or 846 (human) amino acids, respectively, displaying 91% amino acid identity [19–21]. The gene encoding human proto-Vav maps to the p12p13.2 region of chromosome 19 [22], a region of karyotypic abnormalities in several hematopoietic malignancies.

Vav mRNA and protein are expressed in a highly specific manner in hematopoietic tissues and lines [16, 19–21]. In developing mouse embryos, vav expression was similarly restricted to hematopoietic tissues, with the exception of tooth buds which also expressed vav transcripts by in situ hybridization [23]. An antiserum raised against a synthetic Vav peptide (residues 576-589 of proto-Vav) reacts with a 95 kDa protein, corresponding to the proto-vav gene product, present in various hematopoietic cells, and detects an 85-88 kDa protein in fibroblasts transformed by onco-vav [19,21].

The expression of Vav appears to be developmentally regulated. Low transcript levels were first detected in 11- to 12-day-old mouse embryos and expression was significantly higher on day 14.5 [23]. In the developing or adult thymus, Vav was localized mainly in cortical areas, and immunocytochemical analysis confirmed heterogenous expression levels in isolated thymocytes. In contrast, spleen and lymph node cells displayed a more uniform expression. Vav was preferentially localized in the cytoplasm of adult thymocytes, spleen and bone marrow cells [23]. It remains to be determined whether this distribution is characteristic of other hematopoietic cells and whether cellular activation affects it. The findings that transfected Vav associates with the autophosphorylated epidermal growth factor (EGF) or platelet-derived growth factor (PDGF) receptors in fibroblasts [24,25] raise the possibility that physiological triggering of hematopoietic receptors, and the resulting activation of associated nonreceptor PTKs, may induce translocation of Vav to a membrane-bound signaling complex. In addition, some Vav is also found in the nucleus by immunofluorescence and cell fractionation studies [25,26], and Vav translocation to the nucleus was observed in a prolactin-stimulated T cell line [27].

	Domain Structure of Vav

Top Abstract Introduction Isolation and Expression of... Domain Structure of Vav Phosphorylation of Vav Role of Vav in... Transforming Activity of vav... The Interaction of Vav... Future Perspectives: The Vav... References

 
Early analysis of the putative sequence of the vav oncogene and proto-oncogene products identified several structural motifs potentially mediating interaction of Vav with cellular components [16,19–21]. As our understanding of protein-protein interactions in diverse signal transduction pathways has recently increased, additional functional motifs were identified in Vav and interactions of this protein with a number of ligands have been reported. Although much remains to be studied, it is clear that Vav contains a host of conserved motifs that enable it to interact, directly or indirectly, with other proteins, lipids and possibly, nucleic acids. The abundance of such motifs suggests that Vav could be coupled to distinct signaling pathways, thereby possibly mediating crosstalk among such pathways at several levels. The following description of these structural domains corresponds to their historic order of discovery. A schematic view of the domain structure of Vav is shown in Figure 1. Table 1 summarizes reported or suspected associations of cellular components with Vav domains.

View larger version (19K): [in this window] [in a new window]   Figure 1. Schematic structure of proto-Vav. The domains are approximately up to scale and are discussed in detail in the text. Several key features of the protein are indicated: the N-terminal deletion that converts Vav to a transforming oncogene [16, 19, 21]; Tyr-174 (within a Y*EDL motif) which is phosphorylated by p56lck; a conserved LLLQEL motif in the GEF domain which, in the case of Dbl (LLLKEL), is essential for transforming and exchange activities [68,110]; a potential substrate site for cAMP-dependent protein kinase (RRGDS*Y), Cys-528 which is required for transforming [21] and diglyceride-stimulated exchange [39] activities; and a potential SH3-binding site (IPPPPG). Domain/sites

A region downstream of the leucine zipper-like domain, corresponding to residues 132-176 in proto-Vav, is rich in acidic residues (49% glutamic or aspartic acid) [16,21]. This region is largely responsible for the low isoelectric point (6.08) of the predicted protein sequence. Although similar acidic regions are known to mediate interactions with other proteins, the exact function of this domain in Vav is unknown.

The Cysteine-Rich Domain
Residues 516-563 of proto-Vav constitute a cysteine-rich, zinc finger-like region [16,19,21]. This domain (Fig. 1) is essential for the transforming activity of the vav oncogene since mutations of conserved cysteine or histidine residues in it abolished transformation of NIH 3T3 fibroblasts by vav [21]. Although the homology between Vav and bona fide zinc finger domains was questioned [30], a recent study did demonstrate binding of zinc to recombinant Vav [32]. This motif, His-X₁₂-Cys-X₂-Cys-X₁₃-Cys-X₂-Cys-X₄-His-X₂-Cys-X₆-Cys [20, 21, 30], is highly homologous to the phorbol ester [33]- and zinc [34]-binding domains of protein kinase C (PKC) enzymes. A similar domain is also found in several other proteins [20], i.e., diacylglycerol (DAG) kinase, n-chimerin [35], the C. elegans-derived protein, Unc-13 [36], the Raf-1 serine/threonine kinase and, interestingly, two other putative guanine nucleotide exchange factors (GEFs), i.e., Fgd1 [37] and Lfc [38]. Like PKC enzymes [33] (with the exception of PKC), the cysteine-rich domains of n-chimerin [35] and Unc-13 [36)] also bind phorbol ester/DAG, and the enzymatic activities of PKC and n-chimerin are stimulated by phorbol ester.

We found recently that vav-transfected COS cells bound two- to three-fold higher levels of radiolabeled phorbol ester than did control vector-transfected cells [39]. Increased binding was not seen in cells expressing Vav in which Cys-528 had been mutated. The homologous cysteine residue in PKC is essential for phorbol ester binding [33]. In contrast, attempts to demonstrate in vitro phorbol ester binding to lysates of Vav-expressing insect cells or to an E. coli-derived fusion protein of the cysteine-rich Vav domain proved negative [32]. This discrepancy could reflect the dependence of phorbol ester binding to Vav on additional, yet-to-be-defined cellular cofactor(s) missing in the in vitro binding assay. It is also possible that Vav may bind more efficiently other lipid(s), e.g., ceramides, the second messenger products of the sphingomyelin signaling pathway which are structurally related to DAG [40]. In support of this notion, Vav can be activated by synthetic ceramides [39]. PKC, the only PKC isoform that does not bind phorbol ester [32], was also recently found to become activated by ceramide [41].

SH2/SH3 Domains
SH2 and SH3 domains, which are frequently present together, represent modular regions of ~100 and ~60 amino acids, respectively, that participate in protein-protein interactions during signal transduction. While SH2 domains recognize phosphotyrosine (PTyr) in the context of short flanking amino acid sequences, SH3 domains bind proline-rich sequences [3–5]. The C-terminal region of Vav, beginning at residue ~612 (proto-Vav) constitutes two SH3 domains separated by a single SH2 domain (Fig. 1) [3, 24, 25]. This arrangement is shared by Grb2, an adaptor protein that couples several PTK receptors to Ras activation via its association with the ubiquitous Ras GEF, Sos [5]. This similarity suggests that the SH3-SH2-SH3 domains of Vav may have a similar adaptor function. Nevertheless, both the SH2 and SH3 domains of Grb2 and Vav appear to bind distinct, though possibly partially overlapping, sets of proteins (see below). A short proline-rich sequence found just upstream of the proximal SH3 domain of Vav could potentially bind SH3 domains.

The Vav SH2 domain associates with ligand-activated (autophosphorylated) EGF and PDGF receptors in vav-transfected fibroblasts [24, 25]. Mutations in four highly conserved amino acids within the SH2 domain, which are essential for SH2-PTyr interactions, abolished the association of Vav with the activated EGF receptor. However, only two of these mutations reduced its transforming activity [42], suggesting a role for the SH2 domain in the transforming activity of Vav. Since ligation of many hematopoietic receptors causes activation of nonreceptor PTKs, it is likely that Vav associates in an analogous manner, via its SH2 domain, with such PTK(s) following activation. Indeed, a glutathione S-transferase (GST) Vav-SH2 fusion protein was recently reported to bind the T cell-derived p70^zap PTK in an activation (i.e., tyrosine phosphorylation)-dependent manner and, similarly, Vav immunoprecipitates from activated Jurkat cells contained immunoreactive p70^zap, estimated at ~5% of the total p70^zap present in the cells [43]. More prominent, unidentified 74 and 80 kDa proteins also bound to Vav-SH2 in vitro [43]. Vav also associated in vivo and in vitro, via its SH2 domain, with several members of the Jak family [44] of PTKs, and this association was augmented by GM-CSF stimulation [45]. In activated B cells, Vav is transiently associated with a single unidentified 70 kDa PTyr-containing protein, Vap-1 [46]. Vap-1 may be related to a highly tyrosine-phosphorylated 70-75 kDa protein(s) in activated T or B cell lysates which we found to associate with a GST-Vav-SH2 fusion protein in vitro. Recent screening of a phosphopeptide library identified a PTyr-containing sequence, PTyr-(Met/Leu/Glu)-Glu-Pro, as a Vav-SH2-binding consensus motif [47]. The presence of such a motif, Tyr-Glu-Glu-Pro, in the cytoplasmic domain of CD19 [48, 49] may mediate the association of Vav with CD19 following crosslinking of the latter in B lymphocytes [50].

The function of the two SH3 domains of Vav is unknown. The C-terminal SH3 domain of Vav, but not of Grb2, was recently found to bind in vitro directly and to associate in intact cells with heterogenous ribonucleoprotein K, a 65 kDa protein localized in the cytosol and nuclear extracts of several cell types [26, 31]. This protein contains proline-rich sequences which mediate the interaction, and an immobilized synthetic peptide corresponding to one of these sequences bound Vav from T cell lysates [26]. The functional implications of this interaction are unknown. In addition, a fusion protein consisting of the tandem SH3-SH2-SH3 domains of Vav (but not the individual domains) was found to bind in vitro another unidentified, 45 kDa poly(rC)-specific RNA-binding protein [31].

Direct association between Vav and Grb2 was recently documented by several approaches, including a yeast two-hybrid screen [51, 52]. In one instance, Vav-Grb2 association in intact cells was observed only when the two proteins were overexpressed [52], suggesting a relatively low stoichiometry. This may explain our failure to demonstrate this association by coimmunoprecipitation in Jurkat cells. The Vav-Grb2 interaction was mediated by an unusual protein-protein binding mode, i.e., dimerization of specific SH3 domains of each of the two partners. This finding could explain the reported binding of Vav from resting or activated T cells to a recombinant Grb2 protein [53]. Based on the recent description of a physical association between Vav and the regulatory (p85) subunit of PI3-K following CD19 crosslinking in B cells, and the failure to detect tyrosine phosphorylation of either Vav or p85 under these conditions [50], it was suggested that the SH3 domain(s) of Vav may bind to proline-rich sequences in p85; however, other mechanisms of association, including indirect association via some intermediate protein, cannot be excluded.

Guanine Nucleotide Exchange Domain
Considerable progress in our understanding of the function and regulation of small Ras-like GTP-binding (G) proteins was made with the recent discovery of the GEF proteins. These enzymes activate small G proteins by catalyzing the exchange of bound GDP for GTP, a critical step in the stimulation of downstream effector cascades [54^–56]. Exchange proteins isolated to date fall within two subfamilies: a group of GEFs that are specific for Ras proteins, and another group specific for the Rho family of small G proteins which are involved in organization of the cytoskeleton. Cdc25 and Cdc24 from S. cerevisiae represent the prototypic GEFs for these two subfamilies, respectively. Each group of GEFs is characterized by several conserved sequence regions [55].

Soon after the discovery of Vav, it was noted that a stretch of ~180 amino acids, beginning with residue ~198 of proto-Vav, displays a ~25% homology with domains previously identified in Cdc24 and in two mammalian proteins, Bcr and Dbl [20, 57]. This domain was subsequently termed the Dbl-homology (DH) domain. Genetic and molecular cloning techniques led to the recent isolation of cDNA encoding additional yeast (Scd1; [58]) or mammalian members of this DH family, i.e., Ect2 [59], Tim [60], Lbc [61], Abr [62], Tiam-1 [63], Ost [64], FGD1 [37], Dbs [65] and Lfc [38]. The domain structures of these proteins are shown in Figure 2. Recombinant Ect2 binds to several members of the Rho family [59] but, like Tim, FGD1, Lfc, and Dbs, has not been found to date to display detectable GEF activity [59]. The active members of the family (with the target proteins for which they display GEF activity in parentheses) are Cdc24 (Cdc42) [66], Dbl (Cdc42, RhoA) [67^–70], Lbc (Rho) [70], Tiam-1 (Rac1, Cdc42>RhoA) [71], Bcr and Abr (Cdc42>Rho>Rac) [72], and Ost (RhoA, Cdc42) [64]. The GEF activity of these proteins usually correlates with their ability to associate with the nucleotide-free form of their respective targets. Ost, however, represents an exception in that, despite its GEF activity for RhoA and Cdc42, it does not bind these small GTPases but, rather, binds GTP-Rac [64], suggesting that it is a Rac1 effector which, in turn, activates RhoA and Cdc42. Two Ras-specific GEFs, Ras-GRF [73^–75] and Sos1 [63], are unique in that in addition to a distal Ras-specific active exchange domain, they display an inactive DH domain near their N-terminus [73].

View larger version (33K): [in this window] [in a new window]   Figure 2. Structure and biological activity of Dbl-family members. The proteins are drawn to scale with their tandem DH and PH domains aligned. The bar at the upper left corresponds to 100 amino acid residues. The size and domain structure of mammalian Vav2 [125] and yeast Scd1 [58] (with undetermined GEF activities) are very similar to Vav and Cdc24, respectively. mSos1 and rasGRF, which display an inactive DH domain and, in addition, an active Ras-specific GEF domain, are shown for comparison. Other domains shown are the acidic (A) domain of Vav, the cysteine-rich (C) domains of Vav, Fgd1 and Lfc, the SH2 (2) or SH3 (3) domains of Vav and Dbs, the GAP domains of Bcr and Abr, and the serine/threonine kinase domain of Bcr. ND: not detected (or not done).

	Phosphorylation of Vav

Top Abstract Introduction Isolation and Expression of... Domain Structure of Vav Phosphorylation of Vav Role of Vav in... Transforming Activity of vav... The Interaction of Vav... Future Perspectives: The Vav... References

 
Vav is constitutively phosphorylated on tyrosine and serine residues and, at trace amounts, on threonine in resting T cells, B cells, macrophages or transfected NIH 3T3 cells [17, 24, 25, 46]. Triggering of diverse receptors leads to a rapid increase in the tyrosine phosphorylation of Vav. These include antigen receptors on T or B lymphocytes [24, 25, 46, 81, 82]; the FcRI receptor on basophilic leukemia cells [25]; type I, II or III receptors for the Fc portion of IgG [83, 84] or the lipopolysaccharide receptor [85] on monocytes/macrophages; accessory receptors, i.e., CD28 on T cells [86, 87] and CD19 on B cells [50]; receptors for the cytokines interleukin 2 (IL-2) [88], IL-3 [45, 51] and IL-5 [89], interferon [90], erythropoietin [51, 91] and GM-CSF [45, 51, 92]; and, finally, several PTK receptors such as c-kit [45, 93], Flk2/Flt3 [94, 95], and the EGF [24, 25, 42], PDGF [24] or insulin [96] receptors in hematopoietic stem cells or in vav-transfected fibroblasts. Ligand binding to the prolactin receptor, on the other hand, induced increased phosphorylation on serine/threonine, but not on tyrosine, residues [27]. In a comparative analysis of the tyrosine phosphorylation of Vav induced in intact T cells by anti-CD3 or anti-CD28 antibodies, or by cells transfected with the natural CD28 ligand, B7-1, the latter was the most potent inducer of tyrosine phosphorylation [86]. Furthermore, B7-1-induced phosphorylation of Vav was more prolonged. Precise mapping of the tyrosine phosphorylation sites in Vav has not been reported.

The identity of the PTK(s) that phosphorylate Vav under physiological conditions is unknown. However, several findings implicate p56^lck, at least in mature T cells. These include the deficient antigen receptor-induced tyrosine phosphorylation of Vav [97] in a mutant Jurkat T cell leukemia line lacking this kinase [98], the coimmunoprecipitation of Vav and p56^lck [97], the ability of recombinant p56^lck to phosphorylate Vav in vitro [81] and, finally, the in vitro binding of tyrosine-phosphorylated Vav from activated T cells to recombinant GST fusion proteins expressing the SH2 domains of p56^lck or p59^fyn [Bonnefoy-Bérard et al., in preparation]. Tyr-174 in proto-Vav, which is embedded within the acidic domain (Fig. 1), lies in a motif, Tyr-Glu-Asp-Leu, predicted to interact with the SH2 domains of Src-family PTKs [48]. Indeed, using short GST fusion proteins expressing the relevant region of Vav, we found that Tyr-174 is a substrate for recombinant p56^lck in vitro (Bonnefoy-Bérard et al., in preparation). The functional significance of phosphorylation at this residue is unknown.

Vav may also be a substrate for Syk-family PTKs since recombinant p72^syk phosphorylated in vitro-translated Vav (Gulbins et al., unpublished observations). This is supported by several findings: first, Vav was found to associate, via its SH2 domain, with another phosphorylated member of the Syk family, p70^zap [43]; second, the Tyr-Glu-Asp-Leu motif in Vav was an efficient substrate for purified spleen-derived p72^syk [99] or baculovirus-derived p70^zap Vav in vitro (N. Isakov, personal communication). Overall, these findings suggest that activated p70^zap/p72^syk associates with the SH2 domain of Vav and then phosphorylates its Tyr-174 which, in turn, creates a binding site for the SH2 domain of an Src-family PTK, e.g., p56^lck. However, despite the above-documented association of Vav with Src- or Syk-family PTKs and its phosphorylation on tyrosine, attempts to detect PTK activity in Vav immunoprecipitates have been largely unsuccessful [17, 43, 46]. This may reflect the unstable and/or transient nature of the interaction between Vav and its PTK(s). In addition, Vav, or some other protein(s) in the complex, may negatively regulate the enzymatic activity of an associated PTK.

The association of CD28 with another PTK, Itk/Emt, and the ability of CD28 crosslinking to induce rapid activation and tyrosine phosphorylation of this kinase [87], coupled with the somewhat slower course of Vav tyrosine phosphorylation (relative to Itk/Emt) in response to CD28 ligation [86, 87], raise the possibility that Itk/Emt may also be involved, directly or indirectly, in regulating Vav. Similarly, the ability of ligated cytokine receptors, which are coupled to the Jak family of PTKs [44], to induce tyrosine phosphorylation of Vav [45, 51, 88, 89, 91^–93], and the association between Jak kinases and Vav-SH2 [45], suggest some role for these kinases. However, more work is needed in order to definitely identify the "Vav PTK(s)".

Phosphorylation of Vav on tyrosine could modulate its function, either directly (see below) or by inducing its association with other proteins via the SH2 domains of the latter. Coimmunoprecipitation experiments combined with anti-PTyr immunoblotting do not reveal unambiguously whether observed associations with Vav are direct or mediated by some intermediate(s) and, in the case of a direct association, whether this reflects binding of Vav-SH2 to other tyrosine-phosphorylated proteins versus binding of PTyr residues in Vav to SH2-containing proteins (or both). Discrimination between these possibilities requires in vitro binding studies using purified proteins and their SH2 domains or analysis of mutated proteins. Furthermore, in activated T cells (and probably in other hematopoietic cells), Vav apparently comigrates with other tyrosine-phosphorylated proteins [86]. Probing of Jurkat cell lysates with SH2-and/or SH3-containing recombinant fusion proteins revealed tyrosine phosphorylation-dependent association of Vav with c-Crk, Nck, Shc, the N-terminal SH2 domain of GAP, and the C-terminal SH2 domains of PLC1 or the p85 subunit of PI3-K [53]. The latter association may account for the reported interaction between Vav and p85 in B cells following CD19 crosslinking [50], although PTyr could not be detected by immunoblotting in the p85-associated fraction of Vav.

Although earlier studies suggested that the constitutive serine/threonine phosphorylation of Vav is not increased following cellular activation [46], we found recently a modest but reproducible increase in the serine phosphorylation of Vav upon T cell activation (Bonnefoy-Bérard et al., in preparation). Furthermore, Vav coprecipitated with a serine kinase activity that phosphorylated, among others, Vav itself in immune complex kinase assays. The association of Vav with a serine/threonine kinase activity was recently confirmed independently [43]. The nature of the associated kinase, and the potential functional implications of this phosphorylation are unknown.

	Role of Vav in Hematopoiesis, Lymphocyte Development and Activation

Top Abstract Introduction Isolation and Expression of... Domain Structure of Vav Phosphorylation of Vav Role of Vav in... Transforming Activity of vav... The Interaction of Vav... Future Perspectives: The Vav... References

 
Given the highly specific expression of Vav in hematopoietic tissues [16, 19^–21], and the correlation between the onset of its expresssion and hematopoietic activity in the fetal liver [23], Vav is likely to play a role in hematopoiesis. This notion is supported by the finding that vav expression is first detected in in vitro-differentiating murine embryonic stem (ES) cells just prior to and/or during the early onset of hematopoietic differentiation [100^–102]. Stable expression of vav antisense RNA in ES clones inhibited their spontaneous or cytokine-induced differentiation into erythroid and nonerythroid hematopoietic cells, without interfering with the development of embryoid bodies [100]. The degree of suppression of hematopoiesis correlated with the level of expression of vav antisense RNA in individual clones, and paralleled a marked reduction in the induction of several markers that characterize the differentiation of ES cells into erythroid or myeloid lineage cells [101]. These observations suggest that Vav plays a critical role in the development of committed hematopoietic cells from multipotent stem cells. However, an opposite conclusion can be drawn from two other recent studies with ES clones in which both vav alleles have been inactivated by homologous recombination [102, 103]. These cells retained the ability to differentiate into erythroid, myeloid and mast cell progenitors. The reasons for this discrepancy are not clear, but may reflect the use of different strategies for inactivating (or reducing) vav expression.

Although initial attempts to generate viable vav-"knockout" mice by homologous gene targeting in the germ line have failed for reasons which are not clear [102], more recent experiments led to the successful isolation of such mutant mice (L. J. Tybulewicz, personal communication). Thus, in contrast to the former conclusion [102], Vav knockout is not lethal. Nevertheless, the original failure led to an alternative approach, i.e., complementation of RAG-2-deficient blastocysts [104] by reconstitution with vav^–/– ES cells which has recently been achieved by several groups [103, 105, 106]. These mice displayed a severe reduction in the number of thymic double-negative (CD4^–8^–), double-positive and single-positive thymocytes and mature peripheral T cells, and absence of the mature CD4⁺ T cell receptor for antigen (TCR)^high subset of thymocytes. Similarly, total numbers of peripheral B cells were reduced and peritoneal CD5⁺ B cells were absent. T or B cell proliferation induced by anti-antigen receptor antibodies was severely compromised although antigen-nonspecific stimulation was normal. As a result, anti-TCR/CD3-induced IL-2 production was deficient. The antigen receptor-stimulated increase in intracellular Ca²⁺ concentration was reported to be intact in thymocytes or splenic T cells of the vav^–/– in one study [106], but was defective in double-positive thymocytes and peripheral B cells in another [105]. Activation defects also included nearly absent CD5, CD69 (in thymocytes) or CD25 (in splenic T cells) upregulation in response to anti-TCR/CD3 antibodies [105]. This phenotype is consistent with a primary role of Vav in T cell maturation, general lymphocyte expansion and antigen receptor-dependent activation, but not with CD4/CD8 commitment per se. Further analysis of these mice will be instrumental in deciphering the function of Vav in hematopoietic cells.

An important role for Vav in antigen receptor-dependent T cell activation was also implied from another recent study in which the effect of vav cDNA expression vectors on the activation of a transiently cotransfected IL-2 promoter:reporter element was evaluated in Jurkat T cells [107]. Surprisingly, only the vav proto-oncogene, but not the N-terminal truncated oncogenic version, spontanously activated and, in addition, enhanced the anti-TCR antibody-induced activation of the IL-2 promoter or its isolated NFIL-2A or NFAT elements. This effect was sensitive to the calcineurin-blocking drug, FK506, or to a specific PTK inhibitor, and required expression of a functional TCR/CD3, p56^lck and CD45. Nevertheless, Vav overexpression did not have an effect on Ca²⁺ mobilization or the overall tyrosine phosphorylation profile [107]. A similar effect of Vav on NFAT was recently found in another study [108].

	Transforming Activity of vav and Related Genes

Top Abstract Introduction Isolation and Expression of... Domain Structure of Vav Phosphorylation of Vav Role of Vav in... Transforming Activity of vav... The Interaction of Vav... Future Perspectives: The Vav... References

 
As noted earlier, vav was identified as a transforming gene, a result of deleting the N-terminus of the corresponding proto-oncogene product [16, 19, 21]. Addition of the missing upstream sequence suppresses the oncogenic activity [21]. This region may, therefore, negatively regulate the biological activity of Vav. Whether this is achieved via an autoregulatory inhibitory effect or by association with another protein(s) is unknown. The transforming activity of vav was demonstrated only in fibroblasts. It is not known whether abnormalities in its expression or structure are associated with human hematopoietic malignancies. However, the lack of activity of the vav oncogene in IL-2 promoter activation assays [107] would suggest that transformation following ectopic onco-Vav expression in fibroblasts does not reflect its true biological functions in hematopoietic cells.

Studies relying on a strategy similar to the one used for the isolation of vav, i.e., transfection of NIH 3T3 fibroblasts with DNA from mostly human tumors [59^–61, 64, 109] or a murine hematopoietic cell line [38, 65], or an attempt to isolate bcr-homologous genes [62], led independently to the isolation of eight other transforming genes whose products, Dbl [109], Ect2 [59], Tim [60], Lbc [61], Abr [62], Ost [64], Lfc [38] and Dbs [65], share little with vav (or with one another) other than tandem DH-PH domains (Fig. 2). Other approaches aimed at isolating genes involved in tumor cell invasion [63] or in fasciogenital dysplasia [37] also led to the identification of proteins displaying tandem DH/PH domains, i.e., Tiam-1 and FGD1, respectively. The biological activity of these genes is usually unmasked by their rearrangement during transfection experiments, or by their chromosomal translocation in the case of fasciogenital dysplasia. This, in turn, can lead to truncation of their protein products. However, in those instances where the truncated regions have been defined, they do not share any obvious homology and, in the case of Tiam-1, truncations of either N- or C-terminal segments, which are structurally unrelated, confer invasive properties.

These observations suggest that structural changes, or ectopic (or over-) expression of these otherwise unrelated proteins can cause transformation which is mediated by their DH and/or PH domains. This is supported by the findings that mutations or deletions of these two domains in Dbl [68, 110], Ect2 [59], Ost [64], Lfc [38], Dbs [65] or Vav (A. Munshi et al., unpublished observations) abolish their transforming and/or guanine nucleotide exchange activity. It is tempting to speculate that the PH domain serves to localize these proteins to the plasma membrane where they can encounter their small G protein targets and utilize the DH domain to catalyze guanine nucleotide exchange on them. This is also supported by the recent demonstration that targeting to the plasma membrane via an added isoprenylation signal can compensate for the loss of transforming activity of the PH domain-deleted Lfc protein [38]. However, these domains may not be sufficient for transformation. For example, overexpression of C-terminus-truncated Tiam-1 protein (which lacks the DH/PH domains) also induces invasiveness [63]. On the other hand, mutations in the SH2 [42] or cysteine-rich [21] domains of Vav (but not deletion of the cysteine-rich Lfc domain [38]) can abolish the transforming activity, indicating that domains other than the DH/PH domains contribute to the transforming activity of Vav.

Analysis of NIH 3T3 fibroblasts transformed by vav, dbl, ect2, tim, lbc, and ost yielded several clues regarding the potential function of, and mechanism of transformation by, the corresponding gene products. First, cells transformed by these oncogenes form similar foci of rounded or cuboidal, highly piled-up and nonrefractile cells, and display giant multinucleated cells, a result of syncitia formation [16, 59^–61, 64, 109, 111, 112]. This morphology is distinct from that of ras-transformed cells but similar, on the other hand, to that of cells transformed by an oncogenic version of rhoA, including the presence of actin stress fibers and focal adhesions [112]. Tiam-1-transfected T lymphoma cells with invasive properties also form giant multinucleated cells [63]. The demonstrated role of RhoA in regulating cell motility and avidity of the LFA-1 adhesion receptor, and the relationship of these two proteins to invasiveness of tumor cells, suggest that the transforming activity of these DH/PH-containing proteins is mediated, at least in part, via their action on small G proteins, such as RhoA, that regulate organization of the cytoskeleton and, by extension, the motility and invasive properties of cells [63]. Second, vav or dbl-transformed cells display elevated constitutive activity of Erk2, a mitogen-activated protein (MAP) kinase [111^–113]. Furthermore, a cotransfected dominant-negative Erk2 construct reduced the transforming activity of dbl in a focus-forming assay [112]. The two oncogenes did not stimulate, however, transcription driven by a Ras-response element [112]. Since Erk activation is Ras-dependent, it is difficult to reconcile these results. However, more recent data suggest that dbl is a much more effective activator of the c-Jun N-terminal and p38 subfamilies of MAP kinases than of Erk kinases, and that this activation is Rac-dependent but Ras-independent [114]. With the current rapid progress in characterizing regulators and effectors of the Rho family [115, 116], a clearer view of the mechanisms that couple distinct members of the Dbl family to MAP kinase signaling cascades is likely to emerge soon.

	The Interaction of Vav with Ras

Top Abstract Introduction Isolation and Expression of... Domain Structure of Vav Phosphorylation of Vav Role of Vav in... Transforming Activity of vav... The Interaction of Vav... Future Perspectives: The Vav... References

 
The identification of a DH domain in Vav [20, 57] led to the speculation that it might function as a GEF for some member(s) of the Rho family of small G proteins [18, 55]. However, Vav immunoprecipitates, in vitro-translated Vav, or partially purified Vav from transiently transfected COS cells displayed in vitro exchange activity for Ras [39, 81, 82, 113, 117], but not for RhoA, Rac1, Rac2 or Ral [68, 113]. The proto-oncogene product possessed low exchange activity, but the activity could be stimulated in intact cells or in vitro by two apparently independent pathways. Stimulation of T or B cells through their antigen receptors stimulated exchange activity via a PTK-dependent pathway since it paralleled the increase in the tyrosine phosphorylation of Vav, and could be blocked by pretreating the cells with a PTK inhibitor [81, 82]. Furthermore, Vav was phosphorylated and enzymatically activated by recombinant p56^lck in vitro [81]. Although an SH3-mediated association between Grb2 (which constitutively binds Sos) and Vav was recently documented [51, 52], it is not likely to fully account for the Vav-associated exchange activity for two reasons: first, a Vav fragment generated by in vitro translation in a wheat germ lysate system was active [81] and, second, since optimal association of Grb2 with Sos requires both SH3 domains of the former, it is highly unlikely that the same Grb2 molecule could bind simultaneously Vav and Sos with a physiologically significant stoichiometry.

The second pathway could be induced by treating Vav with phorbol esters or synthetic diglycerides, which did not induce an increase in its tyrosine phosphorylation. This activation was PKC-independent since it was not prevented by a specific PKC inhibitor [39, 113]. Genetic analysis of Vav mutants indicated that the cysteine-rich domain of Vav and, more specifically, Cys-528, were essential for the phorbol ester-mediated activation and for the increased binding of labeled phorbol ester to vav-transfected cells [39]. Since Ras and MAP kinases can be activated via PTK-independent signaling pathways, e.g., by trimeric G proteins [6], stimulation of the exchange activity of Vav by a diglyceride-mediated pathway may allow receptors that are not coupled to PTKs to stimulate Ras (and/or some other small G protein?), thereby triggering additional downstream signaling events. Conversely, PTK-mediated Vav activation may contribute to Ras-dependent signaling events initiated by antigen or cytokine receptors, both of which have been shown to be coupled to Ras via their associated nonreceptor PTKs [9^–11, 118, 119].

The expression of a phorbol ester-stimulated Ras GEF in hematopoietic cells may account, at least in part, for the observed Ras activation by phorbol ester in T [120] and B [121] cells, in contrast to fibroblasts, in which the same treatment does not activate Ras [118]. In T cells, this effect was attributed to the proximal action of PKC which indirectly activates Ras via inhibition of GAP [120]. However, the ability of phorbol ester to directly stimulate a Ras GEF would provide an additional mechanism to account for Ras activation by phorbol ester in T or B cells. The existence of a phorbol ester-activated, PKC-independent Ras activation pathway in lymphocytes is reinforced by the finding that phorbol myristate acetate-mediated Ras activation in B cells is apparently PKC-independent [121]. This notion is also supported by our finding that phorbol ester caused Ras activation in proto-Vav-expressing, but not in control, fibroblasts [113]. Another member of the Dbl family, Lfc [38], and the chimerin family of Rho-specific GAPs, also display a similar cysteine-rich domain, and chimerin is activated by phorbol ester [35], thereby representing another example of phorbol ester-responsive proteins that regulate the activity of Ras-related small G proteins. Interestingly, 2-chimerin expresses, like Vav, both a phorbol ester-responsive cysteine-rich domain and an SH2 domain [122]. The presence of these two domains in the same protein may provide a link between phorbol ester- and PTK-regulated responses.

In contrast to proto-Vav, the truncated oncogenic protein [113], as well as an in vitro-translated Vav fragment (residues 143-597) that lacks N- and C-terminal sequences [81], displayed constitutively elevated exchange activity for Ras that was not elevated further by phorbol ester treatment [113]. This correlated with increased exchange activity and the level of active (GTP-bound) Ras in cell lysates from onco-vav-transformed fibroblasts. The correlation between the transforming and constitutive exchange activities of onco-Vav suggests that activation of some small G protein(s) which is the physiological target of Vav contributes to Vav-mediated transformation.

A Vav-Ras interaction is supported by our findings that, first, Vav and Ras coimmunoprecipitate from T cells and, second, a recombinant GST-Ras fusion protein bound Vav from Jurkat or transfected COS cells, albeit with an apparently low stoichiometry (Fig. 3). This represents, most likely, direct binding since recombinant Vav was found in a recent independent study to bind Ras in vitro (although it failed to stimulate guanine nucleotide exchange); binding was dependent on the same residues of Ras that are essential for complex formation between it and two well-established Ras GEFs, i.e., Cdc25 or Sos ([123]; D. Broek and J. Han, personal communication). However, a recent study [111] reported lack of GEF activity in Vav preparations from several sources, and it was recently concluded that the Ras exchange activity documented in Vav immunoprecipitates from prolactin-stimulated cells was mediated by an associated protein rather than by Vav itself [27]. Potential reasons for these discrepancies are discussed below.

View larger version (43K): [in this window] [in a new window]   Figure 3. The association of Vav with Ras.A) Coimmunoprecipitation of Vav with Ras from intact cells. Lysates (1 x 107 cells/group) from resting or anti-CD3-stimulated (1 min) Jurkat cells were immunoprecipitated with an anti-Ras monoclonal antibody (lanes 1, 2) or with a control, isotype-matched antibody (lane 3). Immunoprecipitates were separated by SDS/7.5% PAGE, and transferred to a nitrocellulose membrane which was immunoblotted with a monoclonal anti-Vav antibody. As a positive control, a total cell lysate (1 x 106 cells) was similarly immunoblotted (CL, lane 4). The prominent ~50 kDa bands represent the heavy chain of the immunoprecipitating antibodies.B) Binding of Vav to Ras in vitro. Lysates were prepared from 1 x 107 resting (lanes 1, 3) or anti-CD3-stimulated (2 min; lanes 2, 4) Jurkat cells, or from COS-1 cells (5 x 106/group; lanes 5, 6) which were transfected 48 h earlier with proto (pVav)- or onco (oVav)-vav expression vectors. The lysates were incubated overnight at 4°C with 10 µg control GST protein (lanes 1, 2) or a GST-Ha-Ras fusion protein (lanes 3-6). Proteins were immobilized on glutathione beads which were washed with lysis buffer, and eluted with SDS sample buffer. The eluted proteins were separated by SDS/7.5% PAGE, transferred to a nitrocellulose membrane, and immunoblotted with a monoclonal anti-Vav antibody. The positions of the Vav proteins are indicated by arrowheads, and molecular weight markers are shown to the left of each panel.

	Future Perspectives: The Vav Dilemma

Top Abstract Introduction Isolation and Expression of... Domain Structure of Vav Phosphorylation of Vav Role of Vav in... Transforming Activity of vav... The Interaction of Vav... Future Perspectives: The Vav... References

 
In spite of a large body of information on Vav since its isolation [16], the exact function and regulation of this proto-oncogene product is still poorly understood. The highly specific expression of Vav in hematopoietic cells suggests that it plays a role in signaling pathways unique to these cells. Its increased tyrosine phosphorylation in response to stimulation with different ligands indicates that Vav is coupled to diverse hematopoietic receptors, i.e., antigen, accessory (e.g., CD28 in T cells), Fc and cytokine receptors. This is not unique to Vav since these receptors share other intracellular signal transducers such as Src or Syk family PTKs, the putative regulators of Vav. Moreover, ligation of many of these receptors leads to activation of Ras, which is one likely target for Vav.

A major question regarding the physiological role of Vav concerns the function of its putative guanine nucleotide exchange domain. Scrutiny of the relevant literature reveals a set of findings that support a role for Vav as a Ras-interacting protein, first and foremost, the demonstration by us [39, 81, 82, 113] that different Vav preparations possess exchange activity for Ras. As would be expected from these in vitro-based findings, Vav-transformed NIH 3T3 cells display increased GTP-Ras levels [113]. A small (~50%) but reproducible increase in GTP-Ras was also seen in another recent study; this effect was less pronounced, however, than the parallel ~four-fold increase seen in cells transformed by Ras-GRF, a bona fide Ras-specific GEF ([112]; R. Khosravi-Far and C. Der, personal communication). Other supporting evidence includes: a) the documented direct association between Vav and Ras and the finding that the same residues of Ras mediate the association of Ras with Sos/Cdc25 on one hand, and with Vav on the other; nevertheless, the affinity of the Vav-Ras interaction appears to be lower (D. Broek and J. Han, personal communication); and b) the inhibition of vav-mediated effects in intact cells by dominant-negative ras in cotransfection assays [107, 111, 112, 124].

Other data, however, are inconsistent with the function of Vav as an upstream regulator of Ras. Negative data include the failure of Vav preparations to cause guanine nucleotide exchange on Ras in vitro, or of vav-transformed cells to display an increase of active Ras, in one study [111]; the lack of activation of a ras-response element by oncogenic Vav (but not by Ras-GRF) in intact cells [112]; and the inability of farnesylation inhibitors (that block the function of Ras) to inhibit vav-mediated transformation [111]. In addition, transforming oncogenes of the dbl family, including vav, induce a very different morphology than transforming ras mutants in transfected fibroblasts. Oncogenic Vav and/or Dbl were found to activate the Erk2 [112] or JNK [114] kinases via a Ras-independent pathway. Finally, on theoretical grounds based on the structure of its exchange domain, Vav was grouped among proven or putative GEFs that mediate guanine nucleotide exchange on Rac/Rho small G proteins, rather than on Ras [55].

How can these seemingly contradictory findings be resolved and incorporated into a unifying scheme regarding the function of Vav? We would like to suggest a potential scenario for Vav action in hematopoietic cells which is based on three reasonable assumptions: first, Vav is a multifunctional protein that possesses effector activities in addition to guanine nucleotide exchange; this notion is supported by the presence of several conserved functional domains in Vav, and by the finding that, in addition to the exchange domain [68, 110], mutations in the Cys-rich [21] or SH2 [42] domains of Vav (which are not likely to directly affect the exchange activity) can also abolish its transforming activity. Thus, depending on experimental circumstances (e.g., cell type and the response being studied), some effects of Vav may appear to be more or less dependent on an intact function of Ras, or on its being a Ras GEF, than others.

Second, by comparison to Sos, Cdc25 or Ras-GRF, Vav may represent a relatively inefficient Ras activator, thereby accounting for difficulties in reproducibly documenting its exchange activity. The true physiologic substrate for Vav could be some other yet-to-be-identified small G protein(s), in which case the observed in vitro exchange activity towards Ras may represent a fortuitous cross-reactivity. The action of Vav on such target protein(s) in addition to, rather than on, Ras may explain the apparently Ras-independent effects of Vav and the distinct morphology that its oncogenic version induces in transformed cells. It is also possible that Vav does mediate, in fact, efficient exchange activity on some known small G protein(s) (including Ras), but optimal experimental conditions for demonstrating such activity remain unknown. This situation may not be unique to Vav but, rather, may represent a general characteristic of other Dbl-like proteins which, to date, have not been shown to possess GEF activity, i.e., Ect2, Tim, Dbs, Lfc, and FGD1 [37, 38, 59, 60, 65]. As discussed by others [55], the difficulty in demonstrating exchange activity of members of this family may reflect the fact that they require additional unknown factors to act on their substrates. These may include, among others, a critical cofactor, e.g., a lipid [79] or ß subunits of a trimeric G protein [77] that could bind to the PH domain, or an essential post-translational modification, both of which may occur only in mammalian cells. This may represent a similar situation to the biological activity of several regulators of small G proteins which depends on post-translational modification of the latter [55]. It is important to note that negative studies [59, 111] have used recombinant DH proteins derived from E. coli or baculovirus expression systems. Examples also exist of mammalian proteins that are biologically inactive when expressed in insect cells from a baculovirus vector.

Third, the large body of evidence which supports interactions between Vav and both Ras and Rho small G proteins could potentially be reconciled by implicating Vav as an important linker between Ras- and Rho-dependent signaling pathways in hematopoietic cells. It is clear that a cross-talk between these pathways occurs [115, 116]. Examples of members of the Dbl family that can function simultaneously as effectors and GEFs for distinct small GTPases (including Ras and Cdc42, respectively) exist in the form of yeast Cdc24 and Scd1 [58], and mammalian Ost [64].

These considerations should serve as useful guidelines for future studies aimed at identifying the true physiological functions of Vav (as well as other members of the DH family). For example, based on the demonstration that other GEFs form physical complexes with the nucleotide-free form of their substrates while effectors associate with the GTP-loaded forms of their respective GTPases, the Vav GEF domain may be used as a probe to isolate potential substrate(s) using the yeast two-hybrid system or by screening cDNA expression libraries, or to assess its binding to various "empty" or guanine nucleotide-loaded small G proteins in vitro. Second, as new ligands that bind to distinct functional domains of Vav (proteins, lipids, etc.) are identified, our understanding of the factors that regulate its activity will increase. In addition, functional analysis of a recently isolated Vav-related protein which is expressed ubiquitously [125] would be informative.

Finally, it should be pointed out that many conclusions regarding the function of Vav are based on studies in cells that do not normally express the protein, i.e., NIH 3T3 fibroblasts. In hematopoietic cells, Vav may interact with signal-transducing elements that are not expressed in other cell types, e.g., Syk family PTKs. As a result, it may display effector activities or be subject to regulatory effects that are unique to hematopoietic cells. This is, in fact, clearly demonstrated by the finding that it is the vav proto-oncogene, rather than the oncogene, which is active in T cells [107]. Valuable information about the true function of Vav would thus emerge either from "knockout" strategies designed to abolish its expression, or by stably overexpressing Vav in transfected hematopoietic cell lines or in transgenic animals. These types of studies, which will certainly be actively pursued in the near future, will likely shed light on the function and regulation of this important proto-oncogene product in the hematopoietic system.

	Acknowledgments

 
This work was supported in part by NIH grants GM50819 and CA35299. We thank our many colleagues who participated in the early phase of the work in our laboratory, in particular E. Gulbins, M. Coggeshall, G. Baier and C. Langlet, and A. Martinez for editorial assistance. [This is publication number 118 from the La Jolla Institute for Allergy and Immunology].

	References

Top Abstract Introduction Isolation and Expression of... Domain Structure of Vav Phosphorylation of Vav Role of Vav in... Transforming Activity of vav... The Interaction of Vav... Future Perspectives: The Vav... References

 

1. Cantley LC, Auger KR, Carpenter C et al. Oncogenes and signal transduction. Cell1991;64:281–302.[Medline]

2. Schlessinger J, Ullrich A. Growth factor signaling by receptor tyrosine kinases.Neuron1992;9:383–391.[Medline]

3. Koch CA, Anderson D, Moran MF et al. SH2 and SH3 domains: elements that control interactions of cytoplasmic signaling molecules.Science1991;252:668–674.[Abstract/Free Full Text]

4. Mayer BJ, Baltimore D. Signaling through SH2 and SH3 domains.Trends Cell Biol1993;3:8–13.[Medline]

5. Pawson T, Schlessinger J. SH2 and SH3 domains.Curr Biol1993;3:434–442.[Medline]

6. Johnson GL, Vaillancourt RR. Sequential protein kinase reactions controlling cell growth and differentiation.Curr Opin Cell Biol1994;6:230–238.[Medline]

7. Altman A, Mustelin T, Coggeshall KM. T lymphocyte activation: a biological model of signal transduction.CRC Critical Rev Immunol1990;10:347–391.

8. Mustelin T. T cell antigen receptor signaling: three families of tyrosine kinases and a phosphatase.Immunity1994;1:351–356.[Medline]

9. Weiss A, Littman DR. Signal transduction by lymphocyte antigen receptors.Cell1994;76:263–274.[Medline]

10. Pleiman CM, D'Ambrosio D, Cambier JC. The B-cell antigen receptor complex: structure and signal transduction.Immunol Today1994;15:393–399.[Medline]

11. Kishimoto T, Taga T, Akira S. Cytokine signal transduction.Cell1994;76:253–262.[Medline]

12. Thomas ML. The regulation of B- and T-lymphocyte activation by the transmembrane protein tyrosine phosphatase CD45.Curr Opin Cell Biol1994;6:247–252.[Medline]

13. Chan AC, Iwashima M, Turck CW et al. ZAP-70: a 70 kd protein-tyrosine kinase that associates with the TCR chain.Cell1992;71:649–662.[Medline]

14. Kolanus W, Romero C, Seed B. T cell activation by clustered tyrosine kinases.Cell1993;74:171–183.[Medline]

15. Taniguchi T, Kobayashi T, Kondo J et al. Molecular cloning of a porcine gene syk that encodes a 72-kDa protein-tyrosine kinase showing high susceptibility to proteolysis.J Biol Chem1991;266:15790–15796.[Abstract/Free Full Text]

16. Katzav S, Martin-Zanca D, Barbacid M. vav, a novel human oncogene derived from a locus ubiquitously expressed in hematopoietic cells.EMBO J1989;8:2283–2290.[Medline]

17. Hu P, Margolis B, Schlessinger J. Vav: a potential link between tyrosine kinases and Ras-like GTPases in hematopoietic cell signaling.Bioessays1993;15:179–183.[Medline]

18. Puil L, Pawson T. Vagaries of Vav.Curr Biol1992;2:275–277.[Medline]

19. Katzav S, Cleveland JL, Heslop HE et al. Loss of the amino-terminal helix-loop-helix domain of the vav proto-oncogene activates its transforming potential.Mol Cell Biol1991;11:1912–1920.[Abstract/Free Full Text]

20. Adams JM, Houston H, Allen J et al. The hematopoietically expressed vav proto-oncogene shares homology with the dbl GDP-GTP exchange factor, the bcr gene and a yeast gene (CDC24) involved in cytoskeletal organization.Oncogene1992;7:611–618.[Medline]

21. Coppola S, Bryant S, Koda T et al. Mechanism of activation of the vav protooncogene.Cell Growth Differ1991;2:95–105.[Abstract]

22. Martinerie C, Cannizzaro LA, Croce CM et al. The human VAV proto-oncogene maps to chromosome region 19p1219p13.2.Hum Genet1990;86:65–68.[Medline]

23. Bustelo XR, Rubin SD, Suen KL et al. Developmental expression of the vav protooncogene.Cell Growth Differ1993;4:297–308.[Abstract]

24. Bustelo XR, Ledbetter JA, Barbacid M. Product of vav proto-oncogene defines a new class of tyrosine protein kinase substrates.Nature1992;356:68–71.[Medline]

25. Margolis B, Hu P, Katzav S et al. Tyrosine phosphorylation of vav proto-oncogene product containing SH2 domain and transcription factor motifs.Nature1992;356:71–74.[Medline]

26. Hobert H, Jallal B, Schlessinger J et al. Novel signaling pathway suggested by SH3 domain-mediated p95^vav/heterogenous ribonucleoprotein K interaction.J Biol Chem1994;269:20225–20228.[Abstract/Free Full Text]

27. Clevenger CV, Ngo W, Sokol DL et al. Vav is necessary for prolactin-stimulated proliferation and is translocated into the nucleus of a T-cell line.J Biol Chem1995;270:13246–13253.[Abstract/Free Full Text]

28. Katzav S. vav: a molecule for all haemopoiesis? Br J Haematol1992;81:141–144.