Signaling Network of the Btk Family Kinases

Yun Qiu1 and Hsing-Jien Kung*,2

1Department of Laboratory Medicine and Pathology and Cancer Center, 420 Delaware Street SE, University of Minnesota, Minneapolis, Minnesota, MN 55455, USA; 2Department of Biological Chemistry and Cancer Center, University of California at Davis, School of Medicine, 4645 2nd Avenue, Sacramento, California, CA 95817, USA

The Btk family kinases represent new members of non- immunode®ciency disease, X-linked agammaglobuline- receptor tyrosine kinases, which include Btk/Atk, Itk/ mia (XLA) (Vihinen et al., 1997a) and murine X-linked Emt/Tsk, Bmx/Etk, and Tec. They are characterized by immunode®ciency (Xid) (Vihinen et al., 1995). The having four structural modules: PH (pleckstrin homol- XLA phenotype includes the lack of mature circulating ogy) domain, SH3 (Src homology 3) domain, SH2 (Src B cells and immunoglobulins. Itk, expressed primarily homology 2) domain and kinase (Src homology 1) in a subset of T cells, plays an essential role in T-cell domain. Increasing evidence suggests that, like Src- development and proliferation (Liao and Littman, family kinases, Btk family kinases play central but 1995). It transmits signals emanating from the T-cell diverse modulatory roles in various cellular processes. antigen receptor and mast FceR (August et al., 1994; They participate in signal transduction in response to Gibson et al., 1996). Itk knock-out mice lack mature virtually all types of extracellular stimuli which are thymocytes and are defective in T-cell receptor transmitted by growth factor receptors, cytokine recep- signaling. Tec was originally discovered in hepatocarci- tors, G-protein coupled receptors, antigen-receptors and noma cells, but was subsequently found to be expressed integrins. They are regulated by many non-receptor abundantly in hematopoietic cells (Mano et al., 1993). tyrosine kinases such as Src, Jak, Syk and FAK family Tec responds to a variety of stimuli including IL3, IL6 kinases. In turn, they regulate many of major signaling and SCF (Mano et al., 1995; Matsuda et al., 1995; pathways including those of PI3K, PLCg and PKC. Both Tang et al., 1994). Bmx/Etk cDNA clones were genetic and biochemical approaches have been used to originally identi®ed in a bone marrow library (Tamag- dissect the signaling pathways and elucidate their roles in none et al., 1994) and subsequently in prostate cancer growth, dierentiation and apoptosis. An emerging new cells (Qiu et al., 1998b; Robinson et al., 1996). Etk/ role of this family of kinases is cytoskeletal reorganiza- Bmx is required for IL6-induced neuroendocrine tion and cell motility. The physiological importance of dierentiation of prostate cancer cells and transforma- these kinases was amply demonstrated by their link to tioan of ®broblasts and epithelial cells induced by Src the development of immunode®ciency diseases, due to (Qiu et al., 1998b; Tsai et al., 2000). It is also induced germ-line mutations. The present article attempts to by IL3 and GM-CSF in mast cell lines (Ekman et al., review the structure and functions of Btk family kinases 2000). Our recent ®ndings suggest that Etk/BMX may by summarizing our current knowledge on the interacting also be involved in regulation of cell motility (Chen et partners associated with the dierent modules of the al., submitted). Aside from and perhaps related to the kinases and the diverse signaling pathways in which they various physiological and pathological processes they are involved. Oncogene (2000) 19, 5651 ± 5661. are involved, Btk family of kinases appear to be major regulators of apoptosis (Uckun et al., 1996; Uckun, Keywords: PH domain; Btk; tyrosine kinase; apoptosis; 1998; Xue, 1999). This will be elaborated further in this motility review. The versatile roles of Btk kinases are re¯ected by their protein ± protein and lipid ± protein interaction Introduction modular structures. In addition to the N-terminal Pleckstrin homology (PH) domain and the C-terminal Btk family tyrosine kinases represent a newly emerging kinase domain (SH1), Src Homology domain 1, there group of non-receptor tyrosine kinases, which include are a Src Homology 3 (SH3) domain and a Src Btk/Atk (Tsukada et al., 1993; Vetrie et al., 1993), Tec Homology 2 (SH2) domain. Between their PH domains (Sato et al., 1994a), Itk/Emt/Tsk (Siliciano et al., 1992) and SH3 domains, there exist the Tec Homology (TH) and Bmx/Etk (Qiu et al., 1998b; Tamagnone et al., domains, which contain several SH3-binding, proline- 1994) (Figure 1). The de®ning feature of these kinases rich sequences (PXPPXP) shared by this kinase family, is the presence of a PH (pleckstrin homology) domain with the exception of Etk/Bmx. These protein modules at their N-terminus. Btk has a restricted expression in a recognize and bind to a variety of cellular factors to subset of B cells and in myeloid cells. It is involved in form signal complexes. In addition, the presence of signaling by B-cell antigen receptor, mast cell FceR, PXXP motif and the SH3 domain establishes an IL5 receptor and IL6 receptor (Aoki et al., 1994; de intramolecular interaction which folds the Btk family Weers et al., 1994; Kawakami et al., 1994; Matsuda et kinases in a `closed' form and subject the kinases to al., 1995; Sato et al., 1994a,b). Naturally occurring regulation by stimuli which activate molecules or mutations of Btk were initially identi®ed in human ligands that disrupt this interaction. This is important, as the Btk family kinases, unlike-Src family kinases, do not have a C-terminal tyrosine residue (Y530 in human *Correspondence: H-J Kung c-Src), whose phosphorylation and interaction with the Signaling network of the Btk family kinases Y Qiu and H-J Kung 5652 With the discovery of PI3K and other phosphoinosi- side kinases, the role of phospholipids as signaling molecules is now well established. Several groups investigated the binding ability to phosphoinositides by Btk PH domain (Fukuda et al., 1996; Kojima et al., 1997; Rameh et al., 1997; Salim et al., 1996). The results from these studies on the speci®city and anity to various phospholipids are somewhat contradictory possibly due to the dierent experimental approaches and conditions. However, Rameh et al. (1997) showed that several PH domains, including Btk PH domain, bind with relatively higher anity to the lipid products of PI3-kinase such as PtdIns-3,4,5-P3 in vitro than to Figure 1 Overview of Btk family kinases PtdIns-4-P and PtdIns-4,5-P2, which are constitutively produced in cells. This selectivity is diminished by the xid mutation (R28C) in Btk, indicating that phospho- internal SH2 domain control the kinase activity of the inositide binding to the PH domain is of physiological Src kinases. In this review, we will ®rst summarize the relevance. The interaction between the PH domains structure of these modules and the signaling molecules and phospholipids is thought to play a role in that have been identi®ed to associate with each protein membrane recruitment of these kinases in response to module of Btk family kinases, as well as their potential extracellular stimuli. This is consistent with the reports roles in signaling transduced by Btk family kinases. that several Btk kinases are activated by PI3-kinase This will be followed by a discussion and to some (August et al., 1997; Li et al., 1997a; Qiu et al., 1998b) extent, speculation on the signal pathways whereby the and that Btk becomes membrane localized when PTEN Btk family kinases are activated by upstream activators a lipid phosphatase which counteracts PI3K is and transmit signals to the downstream eectors. The defective (Satterthwaite et al., 2000; Shan et al., 2000). overall biological consequences of these signal pathways will be discussed. Due to the vast literature on Interaction with heterotrimeric G-protein subunits this subject, it is dicult to be inclusive for subject matters as well as citations. Readers are referred to Heterotrimeric G proteins play critical roles in signal several excellent reviews on the related subjects of Btk transduction initiated by hormones, neurotransmitters family kinases (Mano, 1999; Rawlings and Witte, 1995; and various biologically active molecules. The PH-TH Vihinen et al., 1997). domain of Btk has been shown to bind to bg,Gaq and Ga12 subunit of heterotrimeric G-proteins (Langhans- Rajasekaran et al., 1995; Bence et al., 1997; Touhara et al., 1994; Tsukada et al., 1994) and the association results The structure and interacting partners of Btk family in elevation of kinase activity (Jiang et al., 1998; kinases Langhans-Rajasekaran et al., 1995; Ma and Huang, 1998). In addition, Tec and Bmx/Etk are also regulated PH domain and TH domain by Ga12/13. Kinase-dead Bmx/Etk is shown to inhibit Btk family kinases are thus far the only tyrosine kinase Ga12- and Ga13-induced activation of serum response family, which contains a PH domain. The PH domain factor (SRF) (Mao et al., 1998). Furthermore, over- was originally identi®ed as an internal repeat in expression of Tec in NIH3T3 cells resulted in increases of pleckstrin and later found in a wide variety of signaling actin polymerization and formation of stress ®bers as molecules ranging from protein kinases, phospholi- well as translocation of RhoA to particulate fractions, pases, GTPases to adaptor proteins as well as suggesting a role of Btk kinases in regulation of Rho cytoskeleton proteins (Rebecchi and Scarlata, 1998). GTPases (Mao et al., 1998). These observations indicate The mutations in PH domain of Btk have been shown the important role of Btk family kinases in signal to associate with XLA suggesting the important role of transduction induced by G proteins. PH domain in regulation of Btk kinase activity (Rawlings et al., 1993; Thomas et al., 1993). Although Interaction with F-actin the exact function of the PH domain has yet to be elucidated, accumulating evidence suggests that PH The actin cytoskeleton plays an essential role in a domains are multifunctional signaling modules which variety of cellular processes including cell division, cell are able to bind to phospholipids as well as many shape, motility, chemotaxis, etc. Btk is shown to protein partners including heterotrimeric G-proteins, colocalize with actin ®bers upon stimulation of mast PKC isoforms, Stat 3, F-actin, Fas and FAK (see cell by high anity IgE receptor in a PH domain below). The interaction between PH domains with dependent manner (Yao et al., 1999). The F-actin (but various signaling molecules is likely transient and not G-actin) binding site has been mapped to the basic dynamic in the process of signal transduction. residues of a 10-amino-acid stretch of the N-terminal region of Btk. This region is highly conserved among all members of the Btk family, and F-actin binding is Interaction with phosphoinositides likely a common property for all members. The PH Phosphoinositides were recognized previously as pre- domain of Btk has been shown to promote actin cursors for second messengers involved in cell surface ®lament bundle formation in vitro (Yao et al., 1999). receptor-coupled signaling transduction pathways. Recent observations from our laboratory indicate that

Oncogene Signaling network of the Btk family kinases Y Qiu and H-J Kung 5653 endogenous Etk/Bmx can also translocate to plasma Interaction with STATs membrane and colocalize with F-actin exclusively in membrane rues in ®bronectin-stimulated endothelial Saharinen et al. (1997) ®rst reported that Bmx/Etk is a cell HUVEC. These observations suggest the possibility potent activator of STAT1, 3, and 5 in Cos1 cells. The that Btk family kinases are able to translocate to the level of STATs activation induced by Bmx/Etk is actin cytoskeleton where Btk family kinases and their higher than that by Src family kinases. Recent reports potential downstream targets such as Rho family con®rm this data and extend to epithelial cells as well GTPases work coordinately to reorganize cytoskeleton as ®broblasts (Tsai et al., 2000). These authors showed in response to stimuli. that STAT1, 3 and 5 are tyrosine phosphorylated by Bmx/Etk and at least for STAT3, there is a direct association of the molecule with the PH domain of the Interaction with PKC isoforms Bmx/Etk. Importantly, this activation is independent PKC is a family of serine/threonine kinases that play of JAK's, a common activator of STATs and provides essential roles in a variety of biological processes, such an additional activation source for STATs. Further- as proliferation, dierentiation and development. more, Src is known to activate STAT3 and Src- Several PKC isoforms have been found to associate mediated cellular transformation requires STATs with Btk kinases and modulate their kinase activity activation (Bromberg et al., 1998; Cao et al., 1996; (Kawakami et al., 1995; Saharinen et al., 1997; Yao et Turkson et al., 1998); yet the detailed mechanisms al., 1994, 1997). The sequence requirement for the whereby Src activates STAT3 was unclear. Recent interaction between Btk and PKC was mapped to the studies implicate two pathways: ®rst, Src activates p38/ region compassing residues 28 to 45, which is over- MAPK which, in turn, induces serine phosphorylation lapped with phospholipid binding site, suggesting that on STAT3 and activates STAT3 (Turkson et al., 1999); phospholipid binding may compete with PKC binding and the second, Src activates Bmx/Etk by phosphor- and PKC may regulate membrane targeting of Btk ylating Bmx/Etk at tyrosine 566, which then interacts kinases. Btk has been shown to phosphorylate and with STAT3, tyrosine-phosphorylating STAT3 and activate PKC bI (Kawakami et al., 2000), which is activates STAT3 (Tsai et al., 2000). Since the sites of consistent with the observations of similar phenotypes phosphorylation by these two pathways are distinct, it between PKC b knockout mice and the Btk knockout is conceivable that these two pathways may synergize mice (Leitges et al., 1996). Btk and PKC bI may form with each other. In a separate study by Wen et al. a negative feedback loop in BCR-mediated signaling: (1999), a constitutively active Bmx/Etk is shown to be Btk phosphorylates and activates PKC bI and the able to activate STAT1, 3 and 5 with rapid kinetics, activated PKC bI, in turn, phosphorylates Btk and again suggesting that the activation of STATs by Bmx/ downregulates Btk activity. This negative feedback Etk is direct. STATs are transcriptional factors that loop may be involved in maintaining the dynamic modulate the transcription of Cyclin D, p21 and p27 homeostasis of Btk activity in the signaling network. and aect the growth and apoptosis behaviors of the Another novel PKC isoform, PKCd, is shown to cells. This could be one way how Btk family kinases inhibit STATs activation by Bmx/Etk, possibly regulate apoptosis. through PKC-mediated protein tyrosine phosphotase(s) such as SHP-1 (Saharinen et al., 1997). These data Interaction with FAK suggest that PKC isoforms may downregulate Btk family kinases through distinct mechanisms. Focal Adhesion Kinase (FAK) is one of the key mediators of integrin signaling events which control the cellular responses to extracellular matrix, such as Interaction with BP-135/TFII-I survival, migration and proliferation. Recent evidence BAP-135 was identi®ed as a 135 kD tyrosine phos- suggests that, in both endothelial and epithelial cells, phorylated protein associated with Btk in RAMOS Etk/Bmx is regulated by FAK in response to cells treated with pervanadate (Yang and Desiderio, extracellular matrix proteins. We also observed the 1997). BAP135 was later found to be identical to a colocalization of Etk and FAK at membrane rues ubiquitously expressed multifunctional transcription where the F-actins are enriched. These data are factor TFII-I (Grueneberg et al., 1997), which is consistent with the observations from others that Btk deleted in Williams-Beuren syndrome (Perez Jurado kinases are involved in integrin signaling and regula- et al., 1998). TFII-I is constitutively associated with tion of GTPases and actin bundle formation (Mao et Btk in resting B cells but dissociates from Btk upon al., 1998; Nore et al., 2000; Yao et al., 1999). The anti-Ig antibody stimulation, which is accompanied by interaction between Etk and FAK is shown to depend an increase of TFII-I in the nucleus. In the Xid B cells, on PH domain of Etk and FERM domain of FAK which carry a point mutation (R28C) in the PH (Chen et al., submitted). It is likely that this interaction domain of Btk, TFII-I no longer associates with Btk could be important for Etk activation by extracellular and increased level of TFII-I is accumulated in the matrix. The role of Etk in integrin signaling remains to nucleus. These data suggest that Btk may play a role in be investigated. regulation of nuclear export or import of TFII-I. BAP- 135/TFII-I is also a substrate of Btk in vivo and Interaction with Fas phosphorylation of TFII-I by Btk enhances its transcriptional activity (Novina et al., 1998). Since Fas is a member of tumor necrosis factor (TNF) TFII-I is a ubiquitous transcription factor present in all receptor family and one of the major regulators of cell types, it is possibly also regulated by other apoptosis in various cell types. Recently, it was shown members of Btk family kinases. that Btk constitutively associates with Fas in wild-type

Oncogene Signaling network of the Btk family kinases Y Qiu and H-J Kung 5654 DT-40 lymphoma B-cells and human B-cell precursor phosphatidylinositol 3-kinase, Crk, and 14-3-3 proteins leukemia cell line NALM-6-UMI and functions as a through dierent protein-interacting modules, leading to negative regulator of Fas-mediated apoptosis (Vassilev the formation of multi-component signaling complexes. et al., 1999). The interaction between Btk and Fas is Several groups have identi®ed p120 c-Cbl as a ligand for independent on Fas ligation and involved in both the the SH3 domain of Btk kinases in vitro (Bunnell et al., PH domain and the kinase domain of Btk. This 1996; Cory et al., 1995), however, the in vivo relevance of interaction is shown to disrupt the association of these ®ndings remains to be established. The interaction FADD with Fas, which may provide a mechanism by between Itk SH3 domain and Cbl seems to be transient which Btk prevents apoptosis in these cells. and is only detected within 30 min upon stimulation in Jurkat cells, while Src SH3 domain binds to Cbl constitutively. These data suggest that these two kinases Interaction with PTPD1 may bind to distinct proline-rich sequences in Cbl and PTPD1 is a tyrosine phosphatase containing an ezrin the accessibility of these sites may be modulated by domain. It was originally identi®ed as a substrate for modi®cations of the surrounding residues, such as Src (Moller et al., 1994). PTPD1 was found to phosphorylation or ubiquitination. associate with the PH domain of Etk/Bmx (Jui et al., 2001). This interaction leads to activation of Etk/Bmx Interaction with WASP with concomitant tyrosine phosphorylation of PTPD1. How and whether the phosphorylation of PTPD1 WASP (Wiskott-Aldrich Syndrome Protein) whose aects its phosphatase activity is unclear, although defect is associated with the disease aecting multiple Etk/Bmx does not seem to be de-phosphorylated. Thus hematopoietic lineages. The B and T lymphocytes the regulatory eect of PTPD1 on Etk/Bmx is not derived from WAS patients fail to proliferate and through its enzyme activity, but rather on PTPD1's migrate in response to stimuli and have altered ability to bind and disrupt the intramolecular folding cytoskeletal architecture. The functional defects of the of Etk/Bmx. That this interaction is functional is lymphocytes of human WAS are similar to the shown by the observation that the PTPD1 induces the phenotype of the B cells of Xid mice and the T cells activation of STAT3 by Etk/Bmx. of Itk knock-out mice, suggesting a common signaling pathway involving WASP and Btk kinases. SH3 domains of Btk kinases have been shown to bind to SH3 domain WASP in vitro (Bunnell et al., 1996; Cory et al., 1996); SH3 domains are protein modules that recognize the physiological relevance of this interaction will need proline-rich motifs present in many proteins. Muta- to be further investigated. tions in the SH3 domains of several proto-oncogenes such as c-Src and c-Abl increase their transforming Interaction with Syk/ZAP-70 potential, suggesting SH3 domains may down-regulate their kinase activity. Mutation of Y233 or deletion of Syk/ZAP-70 family kinases contain two SH2 domains SH3 domain increases the transforming potential of and are expressed in the cells of immune system Btk mutant (E41K) without aecting its kinase activity (Turner et al., 2000). Loss of Syk expression results in directly, suggesting that the SH3 domain regulates Btk profound hematopoietic defects (Cheng et al., 1995; function through binding to regulatory proteins or Turner et al., 1995). Along with Src family kinases and internal folding (Afar et al., 1996). Btk family kinases, Syk and ZAP-70 have been shown to interact with ITAMs (Immunoreceptor tyrosine- based activation motifs) located in the cytoplasmic Interaction with CD28 domains of immune receptors. Syk preferentially binds T cell activation requires both interaction of the TCR to phosphorylated SH3 domains of Btk and Itk via its with antigen bound to MHC molecules and costimula- SH2 domain (Morrogh et al., 1999). This interaction is tory signals initiated by adhesion molecules such as thought to be important for PKCBI activation CD28. Tec SH3 domain is shown to bind to the proline- (Kawakami et al., 2000). rich motifs in CD28 upon CD28 ligation in T cells and The major autophosphorylation site of Btk (Y233) is involved in transcriptional regulation of cytokines and located in its SH3 domain (Park et al., 1996). It is phosphorylation of p62 Dok in response to CD28 interesting that phosphorylation of Y233 alters the activation (Yang et al., 1999; Yang and Olive, 1999). A binding speci®city for its protein partners (Morrogh et kinase-dead Tec can suppress interleukin-2 expression al., 1999; Park et al., 1996). The Btk SH3 domain binds while a similar Itk mutant does not. These data indicate to c-Cbl constitutively regardless of the phosphoryla- the speci®c interaction between Tec SH3 domain and tion status of Y233. However, phosphorylation of CD28 and distinct roles of Tec and Itk in T cell Y233 signi®cantly diminishes the binding of WASP. activation. On the other hand, Syk preferentially interacts with phosphorylated SH3 domain (Morrogh et al., 1999). These observations indicate the important roles of the Interaction with c-Cbl SH3 domains in regulation of Btk kinases. Cbl is a 120-kD proto-oncogene product which contains multiple proline-rich stretches and several potential Interaction with Sab phosphotyrosine-containing motifs. Cbl is rapidly tyrosine-phosphorylated in response to a variety of stimuli Sab, a negative regulator of Btk activity, was identi®ed and becomes associated with a number of intracellular by using a Far Western cloning strategy. Sab is signaling molecules such as protein tyrosine kinases, speci®cally associated with Btk SH3 domain but not

Oncogene Signaling network of the Btk family kinases Y Qiu and H-J Kung 5655 with Src family kinases. Association between Sab and (Qiu et al., 1998b). Phosphorylation of a highly Btk results in the inhibition of Btk activity and conserved tyrosine residue (Y551 of Btk and Y566 of dramatically reduces both early and late BCR-mediated Etk) by the Src family kinases has been shown to be events including calcium mobilization, IP3 production required for activation of these kinases in vivo and programmed cell death (Yamadori et al., 1999). (Rawlings et al., 1996; Tsai et al., 2000). Several groups used the yeast two-hybrid system to search for potential substrates and regulatory proteins by using SH2 domain kinase-inactive catalytic domains of Btk kinases as SH2 domains recognize phosphotyrosine containing baits. peptides and proteins. Mutations in the SH2 domain of Btk are associated with impaired B cell function and Interaction with BRDG1 may cause the XLA immunode®ciency in humans, possibly due to disruption of phosphotyrosine binding BRDG1 (BCR downstream signaling 1), a docking sites (Vihinen et al., 1999). Several proteins have been protein containing a PH domain and proline-rich shown to interact with the SH2 domains of Btk kinases motifs, was identi®ed as a binding protein for the and to modulate their functions. The SH2 domain of kinase domain of Tec by yeast two-hybrid system Btk is shown to be required for the activation of (Ohya et al., 1999). It was shown that BRDG1 is an in PLCg2 in DT40 cells (Takata and Kurosaki, 1996), vivo substrate of Tec but not Btk or Etk/Bmx, although Btk has not been shown to bind to PLCg indicating it is a speci®c substrate for Tec. Co- directly. expression of BRDG1 enhances Tec activity. The TH domain of Tec is required for this regulation. The TH domain of Itk has been shown to interact with its own Interaction with SLPs/BLNK SH3 domain and keep the kinase in a `close' SLP (SH2 domain-containing leukocyte protein) adap- conformation (Andreotti et al., 1997). Such an tors SLP-65/BLNK/BASH and SLP-76 are tyrosine interaction may be also present in Tec. Thus, BRDG1 phosphorylated by Syk or ZAP-70 in activated B and may activate Tec by disrupting the intramolecular T lymphocytes. Tyrosine phosphorylation of SLP interaction. BRDG1 is speci®c for regulation of Tec. It adapters allows the SH2-mediated recruitment of PLCg is likely that other BRDG1-like molecules may regulate and Btk kinases and brings PLCg to the vicinity of Btk other members of Btk family kinases. kinases and leads to phosphorylation and activation of phospholipase C by these kinases (Su et al., 1999). Interaction with Grb10/GrbIR Grb10/GrbIR, an adaptor molecule containing a PH Interaction with Dok-1 domain and an SH2 domain, was shown to interact Dok-1 or p62Dok is a 62 kd protein ®rst identi®ed as a with kinase domain of Tec by yeast two-hybrid molecule associated with rasGAP (Ellis et al., 1990) screening (Mano et al., 1998). Grb10 is known to bind and as such may be involved in the regulation of ras receptor tyrosine kinases in a ligand-dependent man- signaling. This protein was recently shown to be ner, but not to be phosphorylated on tyrosine residues. associated with Tec (but not Btk or Itk) and is a However, Grb10 becomes profoundly tyrosine-phos- substrate of Tec (Yoshida et al., 2000). Indeed, the phorylated by Tec but not by Syk, Jak2 or insulin tyrosine phosphorylation of Dok-1 by Tec, enables receptor in co-transfection in 293 cells. Co-expression Dok-1 to bind the SH2 domain of Tec, which in turn of Grb10 suppresses the cytokine-driven and Tec- hyperphosphorylates Dok-1. The authors speculated driven activation of the c-fos promoter. It is possible that in addition to its function in regulating ras, the that tyrosine phosphorylated Grb10 may recruit some multiple phosphotyrosine residues on Dok-1 might SH2 domain containing inhibitory proteins such as allow Tec to transmit other signals. phosphotase SHP-1 to the vicinity of kinase domain of Tec which, in turn, dephosphorylates Tec and downregulates its kinase activity. Thus, Grb10 may serve as Interaction with vav a negative feedback regulator in Tec mediated Vav is a guanine nucleotide exchange factor for rac, signaling. cdc42 and rho family of G proteins, involved in cell migration, spreading and membrane ruing. By yeast Interaction with PI3K two-hybrid system, it was found that Tec could complex with vav through its SH2 domain, and was In a yeast two-hybrid screen, Tec was found to likely the kinase that phosphorylates vav upon gp130 associate with the SH2 domain of p85b subunit of stimulation (Takahashi-Tezuka et al., 1997). Machide PI3K as well as that of the p55PIK subunit et al. (1995) reported the interaction between vav and (Takahashi-Tezuka et al., 1997). These interactions Tec TH-domain, suggesting that there may be multiple depend on the autokinase activity of Tec and contacts between these two molecules. This interaction presumably the phosphotyrosine residue(s) on the Tec would be one way how Btk family kinases exerts its molecule. As such, it is understandable that the kinase eect on Rho and cytoskeletal reorganization. domain is required for this interaction in yeast cells where the endogenous tyrosine kinase activity is undetectable. The contact site between Tec and PI3K Kinase domain subunit may be somewhere else on the molecule, such The kinase or catalytic domains of Btk family kinases as the SH2 or SH3 region. The implication of this are highly conserved, sharing about 70% similarity ®nding is that PI3Kb, which is an upstream activator

Oncogene Signaling network of the Btk family kinases Y Qiu and H-J Kung 5656 of Btk family of kinases, can also be a downstream IL6 via PI3K (Qiu et al., 1998b) and presumably Tyk2. substrate, and Btk can activate PI3K pathway. Critical It is interesting that in prostate epithelial cell line to this hypothesis is the demonstration of increased LNCaP, IL6 also engages erbB2 and erbB3, which is a activity of PI3K after association with Tec, which has potent activator of PI3K (Qiu et al., 1998a). This yet to be provided. perhaps explains why in this cell line, IL6 treatment induces a robust activation of Bmx/Etk. The activation of Btk and Bmx/Etk by IL3 and GM-CSF in mast cells have been demonstrated and both involve the activa- The functions and signal transductions of the tion by PI3K (Ekman et al., 2000). Likewise, the Btk family kinases activation of Itk and Tec by IL2 also involves PI3K (August et al., 1997). It thus seems that PI3K is a Upstream activators common activator of BTK family of kinases. As the Btk family kinases contain several protein ± The activation of Btk family of kinases by G-protein protein interaction modules and interact with a multi- coupled receptor has been identi®ed primarily in cases tude of signal molecules, it is not surprising, that they involving Gaq and the related Ga12/13. In both cases, participate in multiple signal pathways and integrate evidence for direct activation of Btk family kinases by signals emanating from virtually all types of cell these proteins was provided by in vitro experiments. surface receptors, including antigen receptors, cytokine Direct association of Gaq and Btk has also been receptors, growth factor receptors and G-protein shown. As described above, the interaction takes place coupled receptors. at the TH domain, which likely serves to unfold the The Btk family kinases were originally recognized molecule. There is also a report describing that Gbg for their requirement in the signaling by T-cell associates directly with Btk (Langhans-Rajasekaran et receptor. B-cell receptor and mast cell receptor (FceRI) al., 1995). The recent work of Gu et al., (2000) showed (Aoki et al., 1994; Gibson et al., 1996; Hata et al., that Gas and Gai protein directly activates Src-family 1998; Heyeck et al., 1997; Kawakami et al., 1994; Liao proteins, the possibility exists that G proteins other and Littman, 1995; Rawlings et al., 1993; Thomas et than Gaq and Ga12 may also activate Btk family al., 1993; Tsukada et al., 1993; Vetrie et al., 1993). kinases. These receptors themselves are not tyrosine kinases, As most of the Btk family Ga12 kinases are but invariably activate Src family kinases as well as expressed in hematopoietic cells, considerably less is Syk family kinases. The Btk family of kinases is found known about their activation by growth factor to be directly phosphorylated and activated by Src receptors as compared to cytokine receptors. Limited (Rawlings et al., 1996; Tsai et al., 2000). Syk, which do data however suggest that growth factors can also not directly participate in the phosphorylation of Btk, induce the activation of this family of kinases. For apparently plays a crucial role in Btk's activation, since instance, SCF (ligand for c-kit receptor) is capable of in Syk- cells, the Btk activity is drastically reduced, activating Tec (Tang et al., 1994) and EGF, TGFa despite the presence of normal active Lyn kinase, a (ligand for EGF receptor), HRG (ligand for erbB3 and Src-family kinase (Kawakami et al., 2000). As erbB4) (Qiu unpublished work) and VEGF (ligand for described above, Btk activation also requires PI3K VGEFR) (Alitalo personal communication) are shown signal. The activation of PI3K is thought to be to activate Etk/Bmx. This is perhaps not surprising, as provided by Lyn kinase which directly associates with all these receptors are capable of activating PI3K, and the p85 subunit of PI3K and at least in vitro it activates some of them also activate Src-like kinases. While all the activity of PI3K (Pleiman et al., 1994; Yamanashi these growth factor receptors have intrinsic tyrosine et al., 1992). This hypothesis has recently been kinase activity, their activation of Btk family kinases challenged by at least two reports which showed that may depend on the availability of Src-like kinases, Syk, but not Lyn is required for PI3K activation, using which may explain why the activation is cell type DT-40 chicken B-lymphoma cell lines carrying null dependent. alleles of Syk, Btk or Lyn as a model (Beitz et al., In summary, our current understanding on activa- 1999; Craxton et al., 1999). Irrespective of the details, tion of Btk family kinases ®ts in a two-step model participation of Src family kinase, Syk family kinase (Figure 2). In the ®rst step, it requires a signal to and PI3K in concert to activate BTK family kinases, disrupt the intramolecular interaction between the SH3 upon antigen stimulation, has been solidly documen- domain and proline-rich region of these kinases. This ted. could be achieved by binding of the PH domains of Among the cytokines that activate Btk family these kinases to phosphoinositides, bg subunits of G kinases are IL2, IL3, IL6, GCSF and erythropoietin proteins or the FERM domain of FAK. These and the list is likely to grow. Although some of these interactions will lead to conformational change of the receptors engage the Src family kinases, a common PH domain and allow these kinases to translocate to signal transducer of these receptors are Jak family of cytoplasmic membrane where the Src family kinases kinases, Jak1, Jak2 or Tyk2 (Ihle and Kerr, 1995). IL6 are located. Subsequently, the highly conserved tyro- was shown to activate Btk and Tec in B-lymphocytes sine residue (equivalent to Src Y416) in the catalytical and hepatocytes respectively. The activated IL6 domain is transphosphorylated by Src kinases. This receptors form stable complexes with Btk and Tec via tyrosine phosphorylation induces the autokinase activ- Jak family kinases. It was further shown that this ity, which then phosphorylates a tyrosine residue in the association leads to direct phosphorylation of Btk by SH3 domain to further prevent the intramolecular Jak1 (Takahashi-Tezuka et al., 1997). By contrast, in folding due to the interaction between the proline-rich prostate epithelial cells, Etk/Bmx is not directly motif in the TH domain and the SH3 domain. The associated with gp130, but it is strongly activated by above activation model has gained support by the

Oncogene Signaling network of the Btk family kinases Y Qiu and H-J Kung 5657

Figure 2 A two-step model of the activation of Btk family kinases

following observations. First, in a cell type that lacks and eventual MAPK/JNK activation (Jiang et al., PTEN, a lipid phosphatase which counteracts PI3K, 1998; Kawakami et al., 1997, 1998; Schaeer et al., Itk is constitutively localized on the plasma membrane 1999). The interaction with vav can potentially activate and hypersensitive to T cell receptor activation (Shan the rac/cdc42/Rho pathways which also lead to JNK et al., 2000). Second, the attachment of a myristylation and p38MAPK activation, and SRF activation (Mao et site or the extracellular and transmembrane domain of al., 1998). The involvement of Tec in the activation of cytokine receptor to Tec lead to constitutive activation PI3K and Akt has also been reported (Craxton et al., and membrane localization, in the absence of PI3K 1999). Activation of AKT leads to the activation of activation (Li et al., 1997a; Yoshida et al., 2000). transcriptional factor Forkhead and NFkB (Brunet et Third, the importance of unfolding the PH domain al., 1999; Kane et al., 1999). Thus, the Btk family during activation is underscored by the ®nding that kinases have potentials to activate all the major signal deletion of the PH domain leads to constitutive pathways. The transcriptional factors that can poten- activation of Etk/Bmx in either prostate or lung tially be activated range from AP-1, NFkB to Fork- epithelial cells (Qiu et al., 1998b; Wen et al., 1999). head and SRF. In addition, STAT3 is shown to This kinase, which is not localized to membrane and directly associate with Etk/Bmx and Btk (Tsai et al., may not be as fully functional as the activated wild- 2000; Uckun, 1998; Wen et al., 1999). In the case of type, apparently can perform certain functions such as Etk/Bmx, this association leads to phosphorylation and the activation of STATs and the consequential activation of STAT3. Similarly, STAT1 and 5 are also activation of Cyclin D1, which does not require activated by Etk/Bmx, with the concomitant activation membrane localization (Tsai et al., 2000; Wen et al., of the target gene Cyclin D. Based on our current 1999). If disruption of the intramolecular folding is a understanding of how STATs are activated, one key step in the activation, molecules that have high presumes that Etk/Bmx phosphorylates STATs in the anity toward the PH domain or the SH3 domain are cytoplasm, resulting in the translocation of STATs expected to activate Btk family kinases. This is indeed dimer into the nucleus where they act as transcriptional the case for Gaq. PTPD1 and FAK. It is suggested factors. Interestingly, in a similar manner, a basic that intramolecular folding begins with the interaction transcriptional factor TFII-I is activated by Btk between the proline-rich motif in the TH domain and (Novina et al., 1999). the SH3 domain (Andreotti et al., 1997). In this regard, Aside from the various transcriptional activations Etk/Bmx would be an exception as it does not contain in the nucleus to reprogram the gene expression, the extensive proline-rich motif and thus, if intramolecular signals of Btk family of kinases also have profound folding exists for this molecule, there must be other eects on cytoskeletal reorganization and cell moti- types of interactions. lity. As described above, the Btk family kinases have the potential to activate rac, cdc42 and rho A. Furthermore, through the conserved basic residues at Downstream signal the N-terminus, the Btk family kinases bind Given the multi-modular structure of the Btk family speci®cally to F-actin (Yao et al., 1999) and are kinases and the multitude of signaling molecules that postulated to participate in actin-bundle formation. can interact with them, it is to be expected that these Treatment of cells with thrombin, which activates kinases impact multiple signal pathways and generate Ga12 and Btk, leads translocation of Rho and pleiotropic eects. Figure 3 summarizes some of the induces actin reorganization. Recent work further signals emanating from Btk family of kinases. The showed that Etk/Bmx, in response to ®bronectin and most well-studied downstream signal is perhaps the integrin, is relocated to membrane ruing together PLCg2 and PKCbI, with concomitant induction of with FAK (Chen et al., submitted). In addition, in sustained calcium in¯ux (Scharenberg and Kinet, 1998) DT-40 cells lacking Btk, the actin pattern as well as

Oncogene Signaling network of the Btk family kinases Y Qiu and H-J Kung 5658

Figure 3 The signals transmitted by Btk family kinases

the ability of actin to aggregate dier from that in (Uckun, 1998), while overexpression of Etk/Bmx the wild-type cells (Nore et al., 2000). These protects prostate cancer cells from ionization and observations suggest that Btk family kinases are Thapsigargen induced cell death (Xue et al., 1999). involved in cytoskeletal reorganization and cell These are some of the examples in support of an movement, and their aberrant activation may be anti-apoptotic role of Btk family kinases. The related to metastasis of cancer cells. This is consistent protective role of Btk family kinases could be due with the observations that B-cells lacking Btk to their ability to activate Akt and NF-kB (Bajpai et (derived from Xid mice, targeted knock-out mice or al., 2000; Kitaura et al., 2000; Petro et al., 2000). On patients of Brutons' Syndrome) fail to migrate, so are the other hand, it was reported that DT-40 lacking the cells derived from Wiskott-Aldrich Syndrome, of both alleles of Btk became resistant to UV induced which the target gene is WASP, a cellular factor apoptosis (Uckun et al., 1996). The failure to interacting with Btk (Cory et al., 1996). establish stable Btk clones was attributed to its potential to induce apoptosis via the activation of p38MAPK (Islam et al., 2000). Moreover, over- Biological consequences expression Etk/Bmx in mast cells and Btk in B-cells The overall consequences of the various signals counteract respectively the anti-apoptotic eect of transmitted by the Btk family kinases are varied GM-CSF (Ekman et al., 2000) and IL6 (Uckun, including growth, dierentiation and apoptosis (for a 1998). How can BTK family kinases be both pro- general discussion, see Hsueh and Scheuermann, and anti-apoptosis? This question actually is not 2000). The involvement of Btk in growth, dierentia- dicult to answer, considering the multitude of tion and development has long been recognized in signals the Btk family kinases transmits, some pro- Xid and XLA cells, as the de®ciency of Btk leads to apoptotic (e.g., p38MAPK) and other anti-apoptotic the failure of B-cells to proliferate upon antigen (e.g., Akt and NF-kB). Depending on the balance of stimulation and subsequent maturation. The ability of these signals, the overall consequence can be either Btk to complement a weakly oncogenic Src in the pro- or anti-apoptotic. Another plausible mechanism transformation of ®broblasts and the observation that is that the kinase, through post-translational mod- Etk/Bmx is the key eector of Src in the transforma- i®cation, may exist in two forms, one pro-apoptotic tion of hepatocytes and ®broblasts suggest their and one anti-apoptotic. For instance, it was found participation in anchorage-independent growth (Li et that Etk/Bmx has a caspase-3 cleavage site at the al., 1997b; Tsai et al., 2000). The role of Etk/Bmx in SH3 domain, and can be cleaved into a protein the neuroendocrine dierentiation of prostate cancer which contains only the SH2 and kinase domain (CY cells has also been documented using a dominant- Huang and YM Wu personal communication). It is negative mutant of Etk/Bmx (Qiu et al., 1998b). conceivable that the wild-type functions to protect There is also signi®cant evidence implicating the Btk the cells from apoptosis by activating Akt, whereas family kinases in the apoptosis pathway, paradoxi- the truncated form may be engaged in apoptosis cally both anti-apoptotic and pro-apoptotic. The B- pathway. In either case, accumulating evidence cells lacking Btk are prone to apoptosis, suggesting suggests that the Btk family kinases are intimately that Btk serves a protective role. Overexpression of involved in the interaction with the cellular apoptotic Btk was shown to inhibit Fas mediated apoptosis machinery and might serve as an apoptosis switch.

Oncogene Signaling network of the Btk family kinases Y Qiu and H-J Kung 5659 Concluding remarks complexes and to de®ne precisely the biochemical nature of these complexes. In parallel, a genetic Non-receptor tyrosine kinases can be grouped into ®ve approach needs to be taken to identify functionally major families: Src, Jak, Abl, Syk, Fak and Btk (see relevant components. Numerous studies have already other this issue). Increasing evidence suggests that these taken advantage of cells devoid of the genes encoding kinases serve multiple roles in diversifying and Btk family of kinases and their interacting partners. amplifying the signals emanating from receptors Deletion of these kinases from the genome often causes located on the cell surface. They do so by crosstalk severe cellular and organismal defects. The work from with the receptors as well as non-receptor tyrosine Witte and his colleagues elegantly illustrates this kinases. The Btk family kinases represent a prime approach (Satterthwaite et al., 2000; Satterthwaite example. By virtue of having at least four protein ± and Witte, 2000). These authors cleverly used a strain protein interaction domains, these kinases are endowed of mouse that have a low dose of the Btk to cross with with an exceptional ability to interact and to form mice that carry deletions in other signal molecules signal complexes. We know, based on the data suspected to be important in the Btk signaling (e.g., reviewed here, that the Btk family kinases transmit lyn7/7,Gaq7/7, etc). These mice are powerful tools to signals from all known classes of cell surface receptors assess the importance of the genetic interactions in Btk including those of growth factors, cytokines, antigens, signaling in the in vivo setting. While these data are not and G-protein coupled, and interact with all known always consistent with the data generated in vitro, families of non-receptor tyrosine kinases except Abl. pointing to the complexity of the in vivo context, they They also transduce signals stemmed from cell are valuable and necessary to provide further directions attachment (®bronectin/integrins) and oxidative stress. for investigations. Only by taking both genetic and As such, the Btk family kinases may play signi®cant biochemical approaches can one achieve a thorough roles in modulating cellular responses and behavior. It understanding of the complex signaling network of Btk therefore comes as no surprise that Btk family kinases family of kinases. are involved in multiple cellular processes including cell proliferation, apoptosis, dierentiation and migration. As illustrated in this review, there are numerous Acknowledgments cellular factors that can potentially interact with this The authors wish to thank Drs Hsiu-Ming Shih, Chi-Ying family of kinases. Yet, in a given cell and in a given Huang, Yi-Mi Wu and Li-Fen Lee for providing stimulat- cellular compartment, there must be some rules ing discussion and data before publication. The work is governing the speci®city of the association. The supported by NIH, ACS and AHA grants to Y Qiu, and challenge in the future is to identify the signal NIH, PCRP and DOD grants to H-J Kung.

References

AfarDE,ParkH,HowellBW,RawlingsDJ,CooperJand Cory GO, MacCarthy-Morrogh L, Banin S, Gout I, Brickell Witte ON. (1996). Mol. Cell. Biol., 16, 3465 ± 3471. PM, Levinsky RJ, Kinnon C and Lovering RC. (1996). J. Andreotti AH, Bunnell SC, Feng S, Berg LJ and Schreiber Immunol., 157, 3791 ± 3795. SL. (1997). Nature, 385, 93 ± 97. Craxton A, Jiang A, Kurosaki T and Clark EA. (1999). J. Aoki Y, Isselbacher KJ and Pillai S. (1994). Proc. Natl. Acad. Biol. Chem., 274, 30644 ± 30650. Sci. USA, 91, 10606 ± 10609. de Weers M, Brouns GS, Hinshelwood S, Kinnon C, August A, Gibson S, Kawakami Y, Kawakami T, Mills GB Schuurman RK, Hendriks RW and Borst J. (1994). J. and Dupont B. (1994). Proc. Natl. Acad. Sci. USA, 91, Biol. Chem., 269, 23857 ± 23860. 9347 ± 9351. Ekman N, Arighi E, Rajantie I, Saharinen P, Ristimaki A, August A, Sadra A, Dupont B and Hanafusa H. (1997). Silvennoinen O and Alitalo K. (2000). Oncogene, 19, Proc. Natl. Acad. Sci. USA, 94, 11227 ± 11232. 4151 ± 4158. Bajpai UD, Zhang K, Teutsch M, Sen R and Wortis HH. Ellis C, Moran M, McCormick F and Pawson T. (1990). (2000). J. Exp. Med., 191, 1735 ± 1744. Nature, 343, 377 ± 381. Beitz LO, Fruman DA, Kurosaki T, Cantley LC and Fukuda M, Kojima T, Kabayama H and Mikoshiba K. Scharenberg AM. (1999). J. Biol. Chem., 274, 32662 ± (1996). J. Biol. Chem., 271, 30303 ± 30306. 32666. Gibson S, August A, Kawakami Y, Kawakami T, Dupont B Bence K, Ma W, Kozasa T and Huang XY. (1997). Nature, and Mills GB. (1996). J. Immunol., 156, 2716 ± 2722. 389, 296 ± 299. Grueneberg DA, Henry RW, Brauer A, Novina CD, Bromberg JF, Fan Z, Brown C, Mendelsohn J and Darnell Jr Cheriyath V, Roy AL and Gilman M. (1997). Genes JE. (1998). Cell Growth Dier., 9, 505 ± 512. Dev., 11, 2482 ± 2493. BrunetA,BonniA,ZigmondMJ,LinMZ,JuoP,HuLS, Gu C, Ma YC, Benjamin J, Littman D, Chao MV and Huang Anderson MJ, Arden KC, Blenis J and Greenberg ME. XY. (2000). J. Biol. Chem., 275, 20726 ± 20733. (1999). Cell, 96, 857 ± 868. Hata D, Kawakami Y, Inagaki N, Lantz CS, Kitamura T, Bunnell SC, Henry PA, Kolluri R, Kirchhausen T, Rickles Khan WN, Maeda-Yamamoto M, Miura T, Han W, RJ and Berg LJ. (1996). J. Biol. Chem., 271, 25646 ± 25656. Hartman SE, Yao L, Nagai H, Goldfeld AE, Alt FW, Cao X, Tay A, Guy GR and Tan YH. (1996). Mol. Cell. Galli SJ, Witte ON and Kawakami T. (1998). J. Exp. Biol., 16, 1595 ± 1603. Med., 187, 1235 ± 1247. Cheng AM, Rowley B, Pao W, Hayday A, Bolen JB and HeyeckSD,WilcoxHM,BunnellSCandBergLJ.(1997).J. Pawson T. (1995). Nature, 378, 303 ± 306. Biol. Chem., 272, 25401 ± 25408. Cory GO, Lovering RC, Hinshelwood S, MacCarthy- Hsueh RC and Scheuermann RH. (2000). Adv. Immunol., 75, Morrogh L, Levinsky RJ and Kinnon C. (1995). J. Exp. 283 ± 316. Med., 182, 611 ± 615. Ihle JN and Kerr IM. (1995). Trends Genet., 11, 69 ± 74.

Oncogene Signaling network of the Btk family kinases Y Qiu and H-J Kung 5660 Islam TC, Branden LJ, Kohn DB, Islam KB and Edvard Ohya K, Kajigaya S, Kitanaka A, Yoshida K, Miyazato A, Smith CI. (2000). J. Gene Med., 2, 204 ± 209. Yamashita Y, Yamanaka T, Ikeda U, Shimada K, Ozawa Jiang Y, Ma W, Wan Y, Kozasa T, Hattori S and Huang XY. K and Mano H. (1999). Proc. Natl. Acad. Sci. USA, 96, (1998). Nature, 395, 808 ± 813. 11976 ± 11981. Jui HY, Tseng RJ, Wen X, Fang HI, Huang LM, Chen KY, Park H, Wahl MI, Afar DE, Turck CW, Rawlings DJ, Tam Kung HJ, Ann DK and Shih HM. (2001). J. Biol. Chem., C, Scharenberg AM, Kinet JP and Witte ON. (1996). in press. Immunity, 4, 515 ± 525. Kane LP, Shapiro VS, Stokoe D and Weiss A. (1999). Curr. Perez Jurado LA, Wang YK, Peoples R, Coloma A, Cruces J Biol., 9, 601 ± 604. and Francke U. (1998). Hum. Mol. Genet., 7, 325 ± 334. Kawakami Y, Hartman SE, Holland PM, Cooper JA and Petro JB, Rahman SM, Ballard DW and Khan WN. (2000). Kawakami T. (1998). J. Immunol., 161, 1795 ± 1802. J. Exp. Med., 191, 1745 ± 1754. Kawakami Y, Kitaura J, Hartman SE, Lowell CA, Pleiman CM, Hertz WM and Cambier JC. (1994). Science, Siraganian RP and Kawakami T. (2000). Proc. Natl. 263, 1609 ± 1612. Acad. Sci. USA, 97, 7423 ± 7428. Qiu Y, Ravi L and Kung HJ. (1998a). Nature, 393, 83 ± 85. Kawakami Y, Miura T, Bissonnette R, Hata D, Khan WN, Qiu Y, Robinson D, Pretlow TG and Kung HJ. (1998b). Kitamura T, Maeda-Yamamoto M, Hartman SE, Yao L, Proc. Natl. Acad. Sci. USA, 95, 3644 ± 3649. Alt FW and Kawakami T. (1997). Proc. Natl. Acad. Sci. Rameh LE, Arvidsson A, Carraway III KL, Couvillon AD, USA, 94, 3938 ± 3942. Rathbun G, Crompton A, VanRenterghem B, Czech MP, Kawakami Y, Yao L, Miura T, Tsukada S, Witte ON and Ravichandran KS, Burako SJ, Wang DS, Chen CS and Kawakami T. (1994). Mol. Cell. Biol., 14, 5108 ± 5113. Cantley LC. (1997). J. Biol. Chem., 272, 22059 ± 22066. Kawakami Y, Yao L, Tashiro M, Gibson S, Mills GB and Rawlings DJ, Saran DC, Tsukada S, Largaespada DA, Kawakami T. (1995). J. Immunol., 155, 3556 ± 3562. Grimaldi JC, Cohen L, Mohr RN, Bazan JF, Howard M, Kitaura J, Asai K, Maeda-Yamamoto M, Kawakami Y, Copeland NG, et al. (1993). Science, 261, 358 ± 361. Kikkawa U and Kawakami T. (2000). J. Exp. Med., 192, Rawlings DJ, Scharenberg AM, Park H, Wahl MI, Lin S, 729 ± 740. Kato RM, Fluckiger AC, Witte ON and Kinet JP. (1996). Kojima T, Fukuda M, Watanabe Y, Hamazato F and Science, 271, 822 ± 825. Mikoshiba K. (1997). Biochem. Biophys. Res. Commun., Rawlings DJ and Witte ON. (1995). Semin. Immunol., 7, 236, 333 ± 339. 237 ± 246. Langhans-Rajasekaran SA, Wan Y and Huang XY. (1995). Rebecchi MJ and Scarlata S. (1998). Annu. Rev. Biophys. Proc. Natl. Acad. Sci. USA, 92, 8601 ± 8605. Biomol. Struct., 27, 503 ± 528. Leitges M, Schmedt C, Guinamard R, Davoust J, Schaal S, Robinson D, He F, Pretlow T and Kung H-J. (1996). Proc. Stabel S and Tarakhovsky A. (1996). Science, 273, 788 ± Natl. Acad. Sci. USA, 93, 5958 ± 5962. 791. Saharinen P, Ekman N, Sarvas K, Parker P, Alitalo K and Li T, Rawlings DJ, Park H, Kato RM, Witte ON and Silvennoinen O. (1997). Blood, 90, 4341 ± 4353. Satterthwaite AB. (1997a). Oncogene, 15, 1375 ± 1383. Salim K, Bottomley MJ, Querfurth E, Zvelebil MJ, Gout I, Li Z, Wahl MI, Eguinoa A, Stephens LR, Hawkins PT and ScaifeR,MargolisRL,GiggR,SmithCI,DriscollPC, Witte ON. (1997b). Proc. Natl. Acad. Sci. USA, 94, Water®eld MD and Panayotou G. (1996). EMBO J., 15, 13820 ± 13825. 6241 ± 6250. Liao XC and Littman DR. (1995). Immunity, 3, 757 ± 769. Sato K, Mano H, Ariyama T, Inazawa J, Yazaki Y and Hirai Ma YC and Huang XY. (1998). Proc. Natl. Acad. Sci. USA, H. (1994a). Leukemia, 8, 1663 ± 1672. 95, 12197 ± 12201. Sato S, Katagiri T, Takaki S, Kikuchi Y, Hitoshi Y, Machide M, Mano H and Todokoro K. (1995). Oncogene, Yonehara S, Tsukada S, Kitamura D, Watanabe T, Witte 11, 619 ± 625. O, et al. (1994b). J. Exp. Med., 180, 2101 ± 2111. Mano H. (1999). Int. J. Hematol., 69, 6±12. Satterthwaite AB, Willis F, Kanchanastit P, Fruman D, ManoH,ManoK,TangB,KoehlerM,YiT,GilbertDJ, Cantley LC, Helgason CD, Humphries RK, Lowell CA, Jenkins NA, Copeland NG and Ihle JN. (1993). Oncogene, Simon M, Leitges M, Tarakhovsky A, Tedder TF, Lesche 8, 417 ± 424. R, Wu H and Witte ON. (2000). Proc. Natl. Acad. Sci. Mano H, Ohya K, Miyazato A, Yamashita Y, Ogawa W, USA, 97, 6687 ± 6692. Inazawa J, Ikeda U, Shimada K, Hatake K, Kasuga M, Satterthwaite AB and Witte ON. (2000). Immunol. Rev., 175, Ozawa K and Kajigaya S. (1998). Genes Cells, 3, 431 ± 441. 120 ± 127. Mano H, Yamashita Y, Sato K, Yazaki Y and Hirai H. Schaeer EM, Debnath J, Yap G, McVicar D, Liao XC, (1995). Blood, 85, 343 ± 350. Littman DR, Sher A, Varmus HE, Lenardo MJ and Mao J, Xie W, Yuan H, Simon MI, Mano H and Wu D. Schwartzberg PL. (1999). Science, 284, 638 ± 641. (1998). EMBO J., 17, 5638 ± 5646. Scharenberg AM and Kinet JP. (1998). Cell, 94, 5±8. Matsuda T, Takahashi-Tezuka M, Fukada T, Okuyama Y, Shan X, Czar MJ, Bunnell SC, Liu P, Liu Y, Schwartzberg FujitaniY,TsukadaS,ManoH,HiraiH,WitteONand PL and Wange RL. (2000). Mol. Cell. Biol., 20, 6945 ± Hirano T. (1995). Blood, 85, 627 ± 633. 6957. Moller NP, Moller KB, Lammers R, Kharitonenkov A, Siliciano JD, Morrow TA and Desiderio SV. (1992). Proc. Sures I and Ullrich A. (1994). Proc. Natl. Acad. Sci. USA, Natl. Acad. Sci. USA, 89, 11194 ± 11198. 91, 7477 ± 7481. Su YW, Zhang Y, Schweikert J, Koretzky GA, Reth M and Morrogh LM, Hinshelwood S, Costello P, Cory GO and Wienands J. (1999). Eur. J. Immunol., 29, 3702 ± 3711. Kinnon C. (1999). Eur. J. Immunol., 29, 2269 ± 2279. Takahashi-Tezuka M, Hibi M, Fujitani Y, Fukada T, Nore BF, Vargas L, Mohamed AJ, Branden LJ, Backesjo Yamaguchi T and Hirano T. (1997). Oncogene, 14, CM, Islam TC, Mattsson PT, Hultenby K, Christensson B 2273 ± 2282. and Smith CI. (2000). Eur. J. Immunol., 30, 145 ± 154. Takata M and Kurosaki T. (1996). J. Exp. Med., 184, 31 ± 40. Novina CD, Cheriyath V and Roy AL. (1998). J. Biol. Tamagnone L, Lahtinen I, Mustonen T, Virtaneva K, Chem., 273, 33443 ± 33448. Francis F, Muscatelli F, Alitalo R, Smith CI, Larsson C Novina CD, Kumar S, Bajpai U, Cheriyath V, Zhang K, and Alitalo K. (1994). Oncogene, 9, 3683 ± 3688. Pillai S, Wortis HH and Roy AL. (1999). Mol. Cell. Biol., Tang B, Mano H, Yi T and Ihle JN. (1994). Mol. Cell. Biol., 19, 5014 ± 5024. 14, 8432 ± 8437.

Oncogene Signaling network of the Btk family kinases Y Qiu and H-J Kung 5661 Thomas JD, Sideras P, Smith CI, Vorechovsky I, Chapman Vihinen M, Kwan SP, Lester T, Ochs HD, Resnick I, Valiaho V and Paul WE. (1993). Science, 261, 355 ± 358. J, Conley ME and Smith CI. (1999). Hum. Mutat., 13, Touhara K, Inglese J, Pitcher JA, Shaw G and Lefkowitz RJ. 280 ± 285. (1994). J. Biol. Chem., 269, 10217 ± 10220. VihinenM,MattssonPTandSmithCIE.(1997b).Front. Tsai YT, Su YH, Fang SS, Huang TN, Qiu Y, Jou YS, Shih Biosci., 2, 27 ± 42. HM, Kung HJ and Chen RH. (2000). Mol. Cell. Biol., 20, Wen X, Lin HH, Shih HM, Kung HJ and Ann DK. (1999). J. 2043 ± 2054. Biol. Chem., 274, 38204 ± 38210. Tsukada S, Saran DC, Rawlings DJ, Parolini O, Allen RC, Xue LY, Qiu Y, He J, Kung HJ and Oleinick NL. (1999). Klisak I, Sparkes RS, Kubagawa H, Mohandas T, Quan S, Oncogene, 18, 3391 ± 3398. et al. (1993). Cell, 72, 279 ± 290. Yamadori T, Baba Y, Matsushita M, Hashimoto S, Tsukada S, Simon MI, Witte ON and Katz A. (1994). Proc. Kurosaki M, Kurosaki T, Kishimoto T and Tsukada S. Natl. Acad. Sci. USA, 91, 11256 ± 11260. (1999). Proc. Natl. Acad. Sci. USA, 96, 6341 ± 6346. Turkson J, Bowman T, Adnane J, Zhang Y, Djeu JY, Yamanashi Y, Fukui Y, Wongsasant B, Kinoshita Y, Sekharam M, Frank DA, Holzman LB, Wu J, Sebti S and Ichimori Y, Toyoshima K and Yamamoto T. (1992). Jove R. (1999). Mol. Cell. Biol., 19, 7519 ± 7528. Proc. Natl. Acad. Sci. USA, 89, 1118 ± 1122. Turkson J, Bowman T, Garcia R, Caldenhoven E, De Groot Yang W and Desiderio S. (1997). Proc. Natl. Acad. Sci. USA, RP and Jove R. (1998). Mol. Cell. Biol., 18, 2545 ± 2552. 94, 604 ± 609. Turner M, Mee PJ, Costello PS, Williams O, Price AA, Yang WC, Ghiotto M, Barbarat B and Olive D. (1999). J. Duddy LP, Furlong MT, Geahlen RL and Tybulewicz VL. Biol. Chem., 94, 604 ± 609. (1995). Nature, 378, 298 ± 302. Yang WC and Olive D. (1999). Eur. J. Immunol., 29, 1842 ± Turner M, Schweighoer E, Colucci F, Di Santo JP and 1849. Tybulewicz VL. (2000). Immunol. Today, 21, 148 ± 154. Yao L, Janmey P, Frigeri LG, Han W, Fujita J, Kawakami Uckun FM. (1998). Biochem. Pharmacol., 56, 683 ± 691. Y, Apgar JR and Kawakami T. (1999). J. Biol. Chem., 274, Uckun FM, Waddick KG, Mahajan S, Jun X, Takata M, 19752 ± 19761. Bolen J and Kurosaki T. (1996). Science, 273, 1096 ± 1100. Yao L, Kawakami Y and Kawakami T. (1994). Proc. Natl. Vassilev A, Ozer Z, Navara C, Mahajan S and Uckun FM. Acad.Sci.USA,91, 9175 ± 9179. (1999). J. Biol. Chem., 274, 1646 ± 1656. YaoL,SuzukiH,OzawaK,DengJ,LehelC,FukamachiH, Vetri D, Vorechovsky I, Sideras P, Holland J, Davies A, Anderson WB, Kawakami Y and Kawakami T. (1997). J. Flinter F, Hammarstrom L, Kinnon C, Levinsky R, Biol. Chem., 272, 13033 ± 13039. Bobrow M, et al. (1993). Nature, 361, 226 ± 233. Yoshida K, Yamashita Y, Miyazato A, Ohya K, Kitanaka Vihinen M, Belohradsky BH, Haire RN, Holinski-Feder E, A, Ikeda U, Shimada K, Yamanaka T, Ozawa K and Kwan SP, Lappalainen I, Lehvaslaiho H, Lester T, Meindl Mano H. (2000). J. Biol. Chem., 275, 24945 ± 24952. A, Ochs HD, Ollila J, Vorechovsky I, Weiss M and Smith CI. (1997a). Nucleic Acids Res., 25, 166 ± 171. Vihinen M, Cooper MD, de Saint Basile G, Fischer A, Good RA, Hendriks RW, Kinnon C, Kwan SP, Litman GW, Notarangelo LD, et al. (1995). Immunol. Today, 16, 460 ± 465.

Oncogene