Transmembrane adaptor proteins (TRAPs) are a group of integral membrane proteins that recruit and assemble signaling molecules proximal to the antigen receptors (Horejsi et al. 2004). These molecules are characterized by short extracellular domains followed by a transmembrane domain and long cytoplasmatic tails containing mainly tyrosine-based interaction motifs. The phosphorylation status of these motifs changes after immunoreceptor ligation and Zap-70 activation, thereby regulating their ability to bind SH2-domain-containing effector proteins. So far, seven different members have been identified that can be further divided into two groups based on their targeting to lipid rafts in the plasma membrane. Lipid raft-associated TRAPs include LAT (linker for activation of T cell), PAG (protein associated with glycosphingolipid-enriched microdomains, also called Csk-binding protein Cbp), NTAL (non-T-cell activation linker) and LIME (Lck-interacting membrane protein), all of which have a juxtamembrane CxxC motif that when palmitoylated targets TRAPs to membrane microdomains. Although this posttranslational modification seems to be essential for LAT function (further discussed below), the significance of dual acylation for targeting of the other TRAPs remains unclear. The non-lipid raft-associated TRAPs encompass LAX (linker for activation of X cells), SIT (SHP2-interacting TRAP) and TRIM (TCR-interacting molecule).
Together TRAPs provide an arsenal of targeting sites that can connect immu-noreceptor signaling to downstream effectors. Several members contain multiple motifs for recruitment of Grb2 or GADS through YxN motifs. This suggests that the different TRAPs in general are associated with the fine-tuning and signal assembly related to specific receptor systems and that there is some level of redundancy between members. Gene targeting studies of different TRAPs support this notion to some extent, but they also suggest that defects in signal organization due to lack of specific TRAPs can have severe long-term effects, including autoimmu-nity. It is important to point out that TRAPs contribute in a unique way in the fine-tuning of different signals, including negative regulation of immune receptor activation. Interestingly, recent results suggest that there also can be interplay by two functionally opposing TRAPS in the fine-tuning of TCR signaling. This added complexity together with the different membrane distribution and binding abilities provides the immune cell with a set of focal points for spatiotemporal signal regulation. A detailed description of all TRAPS and their functions can be found elsewhere (Horejsi et al. 2004; Simeoni et al. 2005). Here we will focus on two functionally opposing scaffolding complexes involving TRAPs.
LAT has an essential role in TCR signaling and provides a hallmark example of the complex signal integration needed for T-cell activation (Fig. 1) (Weber et al. 1998; Zhang et al. 1998a). Gene targeting of Lat in mice leads to a complete arrest early in T-cell development (DN3 stage) due to defect signaling downstream of the pre-TCR (Zhang et al. 1999b). Cell line defects in LAT display normal Z-chain phosphorylation and recruitment of Zap-70, but show defects in PLC-y activation, calcium flux and activation of Ras and Erk (Finco et al. 1998; Zhang et al. 1999a). The signaling defects in these cell lines can be rescued by wild-type LAT, but not by a mutant with a disrupted CxxC motif, suggesting that lipid raft association is essential for normal function (Zhang et al. 1998b). The cytoplasmic part of LAT contains a total of nine tyrosine-based interaction motifs where the four distal sites (Y132LVV, Y171VNV, Y191VNV and Y226ENL) seem to be the primary targets for phosphorylation by Zap-70 or Syk and functionally most important. In agreement with this, mice with knock-in mutation in all four of these distal tyrosines (LATY-4F) have similar phenotype as the LAT-/- mice (Sommers et al. 2001). Tyrosine 132 binds specifically the SH2 domain of PLCy, whereas the three distal tyrosines bind the SH2 domains of Grb2 or the Grb2-related adaptor protein GADS (Zhang et al. 2000). Binding of Grb2 leads to recruitment of the Ras guanine
Fig. 1 Integrated scaffolding by LAT and Slp-76 couples TCR to downstream mediators. Phosphorylation of LAT by Zap-70 after TCR activation leads to recruitment of Grb2-SOS and GADS-Slp76 to the three C-terminal YxN motifs. Grb2-SOS is involved in activating the Ras-Erk pathway, while recruitment of GADS brings Slp-76 to the membrane, where it is phosphorylated by Zap-70 at critical tyrosine residues. This creates binding sites for the SH2 domains of Vav, Nck and the Tec kinase ITK. Vav and Nck regulate cytoskeletal reorganization, while ITK phosphor-ylates and activates PLC-y. The SH2 domain of PLC-y binds directly to phosphorylated Y132LVV in LAT, and after activation PLC-y will hydrolyze PIP2 to generate IP3 and DAG. IP3 induces release of cytoplasmic Ca2+ and activation of NFAT through calcineurin. DAG activates PKC isoforms and RasGRP to regulate the activity of NF-kB and AP1. Together, this complex scaffolding connects the activated TCR complex to the transcriptional machinery regulation of the immune response
Fig. 1 Integrated scaffolding by LAT and Slp-76 couples TCR to downstream mediators. Phosphorylation of LAT by Zap-70 after TCR activation leads to recruitment of Grb2-SOS and GADS-Slp76 to the three C-terminal YxN motifs. Grb2-SOS is involved in activating the Ras-Erk pathway, while recruitment of GADS brings Slp-76 to the membrane, where it is phosphorylated by Zap-70 at critical tyrosine residues. This creates binding sites for the SH2 domains of Vav, Nck and the Tec kinase ITK. Vav and Nck regulate cytoskeletal reorganization, while ITK phosphor-ylates and activates PLC-y. The SH2 domain of PLC-y binds directly to phosphorylated Y132LVV in LAT, and after activation PLC-y will hydrolyze PIP2 to generate IP3 and DAG. IP3 induces release of cytoplasmic Ca2+ and activation of NFAT through calcineurin. DAG activates PKC isoforms and RasGRP to regulate the activity of NF-kB and AP1. Together, this complex scaffolding connects the activated TCR complex to the transcriptional machinery regulation of the immune response exchange factor SOS, thereby contributing to the activation of the Ras-MAPK pathway. The SH2 domain of PLCy is specifically recruited to Y132 and places the enzyme in close proximity of its substrate PtdIns(4,5)P2, leading to an increase in cytoplasmic calcium by IP3 and activation of both PKC isozymes and RasGRP by DAG. Results suggest that the recruitment of RasGRP by DAG is the preferred pathway for activation of Erk downstream of the TCR (Dower et al. 2000; Layer et al. 2003; Priatel et al. 2002). The activation of PLCy is dependent on the GADS-mediated recruitment of the multifunctional adaptor Slp-76 to LAT, since both LAT-deficient and Slp-76-deficient cells are unable to activate PLCy (Finco et al. 1998; Yablonski et al. 1998). Recruitment of GADS to LAT brings Slp-76 to the membrane, where it is phosphorylated by Zap-70 at critical tyrosine residues in the N-terminal region (Singer et al. 2004). This creates binding sites for the SH2 domains of Vav, Nck and the Tec-kinase ITK. PLCy is then phosphorylated by the recruited ITK and activated. Like LAT, Slp-76 can bind PLCy, directly creating an interconnected signaling module of LAT, GADS, Slp-76, ITK and PLCy that controls the generation of the second messengers IP3 and DAG downstream of immunoreceptors (Koretzky et al. 2006). Additional regulation is introduced since both PLCy and ITK contain PH domains for binding to phosphatidylinositols in the membrane (Schwartzberg et al. 2005). The Slp-76-Vav-Nck interactions are important in regulating processes leading to reorganization of the actin cytoskeleton and generation of the immunological synapse (further discussed below).
The importance of the LAT-PLCy interaction in immune homeostasis is dramatically illustrated in knock-in mice where the PLCy binding site is mutated to phenylalanine (Aguado et al. 2002; Sommers et al. 2002). T cells from these mice display similar defects in signaling, as seen in cell lines with the same mutation characterized by reduced phosphorylation of PLCy and an inability to induce intra-cellular calcium release after TCR stimulation. This results in a severe defect in T-cell development and supports the notion that the LAT-PLCy interaction is necessary for pre-TCR signaling. Surprisingly, these mice develop a polyclonal lymphoproliferative disease and autoimmunity involving constitutively active CD4+ cells in the periphery, a high amount of T-helper-2 (TH2) cytokines secreted and induced B-cell maturation. Interestingly, the activation of Erk was relatively normal in these mice, suggesting additional mechanisms for signal regulation of the Ras-Erk pathway. Another (and very likely) explanation for the drastic lymphopro-liferative disease is the defect in the development of CD4+CD25+Foxp3+ regulatory T cells (Treg) seen in these mice (Koonpaew et al. 2006). Considering the essential role of Treg in immune tolerance (Sakaguchi et al. 2006a), the absence of these cells would clearly be problematic and contribute to the phenotype. A clear defect in immune development is also seen in knock-in mice where the three distal tyrosines in LAT are mutated (Nunez-Cruz et al. 2003). These LAT-Y3F mice also have an early developmental block, but retain a significant amount of y§T cells with an activated phenotype that produce large amounts of TH2 cytokines. Jointly, these models suggest that both binding of PLCy and Grb2/GADS complexes to LAT contribute to the balancing of signals needed for correct development, while at the same time selecting T cells with an appropriate antigen repertoire. Signal integration at the level of LAT can also involve proteins complexes with inhibitory functions. The adaptor protein Gab2 is recruited to LAT via its association with the SH3 domain of Grb2 or GADS. Gab2 has a negative effect on TCR activation, probably by recruiting the protein tyrosine phosphatase (PTP) SHP-2 to LAT (Yamasaki et al. 2003). A recent report also describes the clustering of a Grb2-Dok-2-SHIP regulatory complex to LAT, which when disrupted by the use of siRNA increases proximal signaling and IL-2 production (Dong et al. 2006). These results fit well with results from Dok-1/Dok-2 double deficient mice describing enhanced TCR-induced cytokine production and proliferation (Yasuda et al. 2007). Targeting of the 5'-phosphoinositol-phosphatase SHIP1 to LAT in the membrane will localize this phosphatase in the proximity of its lipid substrate and thereby enable modulation of the recruitment and activation of PH-domain-containing signaling proteins. Together, these examples add to the complexity of signal integration at the level of
LAT and may indicate that the multiple Grb2-binding sites seen in some of the TRAPs will recruit functionally opposing protein complexes. Dissecting the spatiotemporal control of such LAT scaffolding will be necessary to fully understand the multitude of proximal signal integration downstream of ITAM-containing immunoreceptors. Furthermore, this information will also be relevant to the analysis of the complex phenotypes seen in different LAT mouse models.
Reversible palmitoylation through the opposing activity of acyltransferases and thioesterases is important in the functional regulation of several proteins (Resh 2006). In immune cell signaling, lipid modification, including palmitoylation, is central for correct membrane targeting of key signaling molecules, such as Src kinases and Ras. Although the significance of LAT targeting to lipid rafts is still controversial, it seems clear that the dual palmitoylation of LAT contributes to the correct tethering in the plasma membrane (Zhang et al. 1998b). In agreement with this, the dynamic regulation of LAT palmitoylation has recently been suggested to be an important mechanism in inducing non-responsiveness in T cells and anergy (Hundt et al. 2006). Considering that T-cell anergy is not caused by a permanent failure in TCR-induced signaling pathways, but represents a temporal and reversible defect in activation due to specific uncoupling of downstream signal pathways, a dynamic modulation of LAT function would provide an attractive model for anergy regulation.
Together this rather extensive analysis of the molecular mechanism associated with LAT-mediated scaffolding in T cells has revealed that site-specific recruitment and specific protein assembly are critical in fine-tuning proximal TCR signals.
1.1.2 NTAL and PAG: Transmembrane Adaptors with Regulatory Functions
In the absence of an appropriate activating stimulus, the TCR-signaling machinery needs to be kept in an inactive state. Regulation of signals controlling the initiation of lymphocyte activation is central in preventing the cell from launching an attack on normal cells. Equally important is the release from inhibition when activation is necessary. This also includes mechanisms that are able to terminate an immune response after successful clearance of the pathogen. It is now evident that dynamic scaffolding of inhibitory complexes at the plasma membrane is functionally important for immune homeostasis. Adaptor molecules that negatively regulate T-cell activation include NTAL, PAG, SIT and LAX, in addition to cytoplasmatic adaptors such as Cbl (casitas B-lineage lymphoma) and SAP (SLAM-associated protein) (Horejsi et al. 2004; Latour and Veillette 2004; Mueller 2004). The scaffolding mediated by the two TRAPs NTAL and PAG plays different roles in immune control. NTAL is expressed in B cells, mast cells and NK cells, but not in naïve T cells. Five of the eight tyrosine phosphorylation motifs in the cytoplasmatic region are YxN motifs that become tyrosine phosphorylated after BCR or Fc-receptor stimulation and associate with Grb2, Gab1 and Cbl (Brdicka et al. 2002; Janssen et al. 2003). Due to its lack of expression in naïve T cells and its similarity with LAT, NTAL was originally thought to be a LAT-like linker downstream of the B
cell receptor. However, deletion of the gene encoding NTAL (LAB in mouse) did not significantly affect B-cell development or BCR-mediated signaling (Wang et al. 2005; Zhu et al. 2004). Although a recent report suggests a regulatory role in the internalization of BCR (Mutch et al. 2007), the significance of NTAL in B cell function remains unclear. Interestingly, NTAL can partly rescue the phenotype of LAT-deficient cells and the development of peripheral T cells in LAT -/- mice. However, NTAL does not bind PLC-y or recruit Slp proteins, and transgenic mice expressing NTAL on a LAT-deficient background develop a severe lymphoprolif-erative syndrome almost identical to the LATY136F mice (Janssen et al. 2004; Koonpaew et al. 2004). So far, these results suggest that the Grb2-binding sites in NTAL can compensate for the loss of LAT except for the critical recruitment and activation of PLCy. An important development in the understanding of NTAL in immune cells was reported recently (Zhu et al. 2006). It turns out that the expression of NTAL is upregulated during T-cell activation and that aged mice (>6 months) deficient in NTAL develop an autoimmune syndrome characterized by enlarged spleens and production of autoantibodies. T cells from these mice are hyperactive, and TCR stimulation results in increased phosphorylation of LAT, PLCy, Erk and Akt. In comparison, T cells from NTAL transgenic mice display reduced activation of the same molecules, further supporting a negative effect on T-cell activation. Interestingly, the data suggest that there is a functional connection between NTAL and LAT, since the targeting of LAT to lipid rafts is somehow related to NTAL expression. In the absence of NTAL, more LAT was distributed to rafts leading to increased LAT phosphorylation and calcium mobilization, whereas overexpression of NTAL had the opposite effect. Together these data support a model where NTAL deficiency leads to enhanced TCR signaling in the periphery and gradual accumulation of autoreactive T cells that result in the development of an autoimmune syndrome. Thus, upregulation of NTAL after TCR activation provides a negative feedback loop important for T-cell homeostasis.
The cloning and initial characterization of PAG led to some interesting models of proximal control of TCR signaling. Cbp/PAG is ubiquitously expressed and was originally observed as an 80-kDa phosphoprotein present in lipid rafts and as a component of a Fyn-associated complex in T cells (Brdicka et al. 2000; Kawabuchi et al. 2000; Marie-Cardine et al. 1999). Like other TRAPs, it is a type-III transmembrane protein with a short extracellular domain, a transmembrane region followed by a palmitoylation motif and a cytoplasmatic part containing multiple sites for tyrosine phosphorylation. These sites serve as substrates for the Src kinase Fyn as Cbp/PAG phosphorylation in Fyn -/- cells is almost absent (Yasuda et al. 2002). Analyzed in vitro, PAG phosphorylation creates binding sites for the SH2 domains of several signaling molecules, but in vivo only the C-terminal Src kinase (Csk) and Fyn itself are reproducibly found to interact with PAG (Brdicka et al. 2000; Kawabuchi et al. 2000). Csk plays an important role in negative regulation of Src kinases by phosphorylation of a C-terminal inhibitory tyrosine residue that induces an intramolecular interaction between this site and the SH2 domain (Bergman et al. 1992; Chow et al. 1993; Okada et al. 1991). Since Src kinases are membrane targeted, this attenuation requires the membrane recruitment of the cytosolic Csk. In agreement with this, constitutive targeting of Csk to the plasma membrane leads to almost complete block of TCR signaling (Cloutier et al. 1995). The inhibitory action by Csk is counteracted by the tyrosine phosphatase CD45 through dephosphorylation of the C-terminal site (for review, see Thomas and Brown, 1999; Veillette et al. 2002). Site-directed mutagenesis of Cbp/PAG identified Tyr317 (Tyr314 in mouse) in an YSSV motif as the site for Csk binding (Brdicka et al. 2000; Kawabuchi et al. 2000; Thomas and Brown 1999). In a naive T cell, PAG is kept in a hyperphosphorylated state, probably due to its association with Fyn, and thereby constitutively targets Csk to lipid rafts. Interestingly, the binding to Cbp/PAG significantly increases Csk activity that can be further modulated by PKA-mediated phosphorylation of Csk (Vang et al. 2003). This could provide a system for tonic inhibition of Src kinases that is localized in proximity to the Cbp/PAG-associated Csk. Upon TCR stimulation, PAG is rapidly dephos-phorylated by a mechanism that appears to involve CD45 activity, and Csk is subsequently removed from lipid rafts (Brdicka et al. 2000; Davidson et al. 2003; Torgersen et al. 2001). This dissociation of Csk results in less negative control of Src kinases, allows for phosphorylation of specific substrates including ITAM sequences and is a necessary permissive event to allow T-cell activation to proceed.
The physiological significance of Csk targeting to Cbp/PAG has been seriously questioned by two independent reports describing the absence of any clear pheno-type in Cbp/PAG deficient mice (Dobenecker et al. 2005; Xu et al. 2005). Specifically, embryogenesis, T-cell development and T-cell functions appeared normal in the absence of Cbp/PAG. Taken together with the severe phenotype of Csk-deficient mice (Imamoto and Soriano 1993; Nada et al. 1993), this information clearly suggests that Cbp/PAG is not the only anchoring molecule for Csk and that some redundancy must exist. In fact, several Csk-binding adaptor proteins have been described including Dok-related adaptors, paxillin, LIME and SIT, arguing for coordinated action of several scaffolding proteins in the regulation of Csk (Lemay et al. 2000; Pfrepper et al. 2001; Sabe et al. 1994). Together this opens for a model where Csk targeting by distinct adaptors couple Csk modulation onto specific pathways on an individual basis and provide control modules for the Src kinases involved.
The non-covalent association of Fyn to Cbp/PAG has not been fully characterized. Dephosphorylation of Cbp/PAG after TCR stimulation of normal human T cells was originally reported not to affect the binding of Fyn to Cbp/PAG, suggesting a phosphotyrosine-independent association. However, recent results from mice demonstrate a similar release of Fyn from Cbp/PAG after TCR crosslinking as with Csk (Davidson et al. 2007). In fact, it is suggested that the release of Fyn precedes the dephosphorylation of Tyr314 and release of Csk. The reason for this discrepancy is unknown, but could potentially be influenced by Fyn SH3 domain binding to PAG (see below). This study also includes the analyses of Cbp/PAG transgenic mice arguing for a distinct role of Cbp/PAG-associated Fyn in the regulation of T-cell anergy, implying that Fyn can regulate other processes than Csk recruitment when associated with Cbp/PAG. Whether Cbp/PAG truly provides independent regulation of Fyn or Csk remains to be clarified, but studies like this might provide insight into the functional role of scaffolding to Cbp/PAG.
Structurally Cbp/PAG also has certain unique features like two proline-rich regions and a PDZ-domain binding motif in its C-terminus. We recently identified the first proline-rich region as a ligand for the SH3 domain of Fyn, suggesting that Fyn interacts with PAG through both the SH3 and SH2 domains (Solheim et al. 2008). This provides an example of coordinated, dual domain docking of Fyn as seen in certain other cases (Arold et al. 2001; Nakamoto et al. 1996). Furthermore, it opens for the possibility that Fyn associated with PAG is insensitive to Csk-mediated phosphorylation at the regulatory Y528 site, since the SH2 domain already is engaged and unable to form a closed conformation. The possible transient release of Fyn after T-cell activation described above suggests that the SH2 domain is most important in binding. However, it appears that the engagement of the SH3 domain is more significant in regulation of Fyn kinase activity. This would be in agreement with the general model of Src regulation where engagement of the SH3 domain causes a structural change that releases Fyn from an inhibitory intramolecular interaction between the SH3 domain and the SH2-kinase linker region (for review, see Hubbard and Till 2000; Sicheri and Kuriyan 1997).
The C-terminal VTRL sequence of PAG interacts with the PDZ-domain of the cytoplasmic adaptor EBP-50 [ezrin, radixin, moesin (ERM)-binding protein 50], which in turn can bind to ERM proteins and connect this protein complex to the actin cytoskeleton (Brdickova et al. 2001; Itoh et al. 2002). This opens the possibility that PAG is involved in the rearrangement of the cellular framework influencing formation of the immunological synapse. In this context it is interesting to note that reduction of PAG expression by siRNA seems to influence Src kinase activity related to cell adhesion and to spread in a similar way as in Csk-deficient cells (Shima et al. 2003).
Was this article helpful?
All Natural Immune Boosters Proven To Fight Infection, Disease And More. Discover A Natural, Safe Effective Way To Boost Your Immune System Using Ingredients From Your Kitchen Cupboard. The only common sense, no holds barred guide to hit the market today no gimmicks, no pills, just old fashioned common sense remedies to cure colds, influenza, viral infections and more.