Inhibition of Immune Responses by ITAM-Bearing Receptors

See allHide authors and affiliations

Science's STKE  31 Jan 2006:
Vol. 2006, Issue 320, pp. re1
DOI: 10.1126/stke.3202006re1


Cells of the immune system possess many multisubunit receptors that are composed of a ligand-binding subunit associated with distinct signaling adaptors containing one or more immunoreceptor tyrosine-based activation motifs (ITAMs). These receptors include the T cell receptor, the B cell receptor, and many Fc receptors, as well as families of activating receptors on myeloid and natural killer cells. Receptors that associate with ITAM-containing adaptors classically have been viewed as transducing activating signals involving phosphorylation of the tyrosines within the ITAM and recruitment of Syk family tyrosine kinases. Receptors associated with ITAM-containing adaptors in myeloid cells have also been implicated in inhibition of cellular activation. Here, we discuss these new negative roles for signaling by receptors that associate with ITAM-bearing adaptors in myeloid and other cell types within the immune system.


Many activating receptors found on hematopoietic cells are composed of multiple subunits, with the ligand-binding subunit distinct from the subunit that transduces the activating signal. These receptors include the T cell receptor (TCR), the B cell receptor (BCR), and Fc receptors (FcRs) for several different classes of immunoglobulins, as well as various activating receptors found on natural killer (NK) cells and myeloid cells (Fig. 1). The ligand-binding subunits of these receptors typically noncovalently associate with their signaling partners through interactions involving basic amino acids located in the transmembrane domain of the receptors with acidic amino acids located in the transmembrane domain of the signaling adaptor subunits. The transmembrane signaling adaptors in this family typically have a minimal extracellular domain and contain one or more cytoplasmic immunoreceptor tyrosine-based activation motifs (ITAMs), containing the consensus sequence YxxL/I-(x)6-8-YxxL/I (x denotes any amino acid), through which they propagate downstream signals (1). This family of ITAM-containing transmembrane signaling adaptors includes the CD3γ, δ, ε, and ζ subunits of the TCR; Igα and Igβ (also known as CD79a and CD79b) of the BCR; the γ chain of FcεRI (FcRγ), which associates with several FcRs, as well as other myeloid and NK receptors; and DAP12, which also associates with many NK and myeloid receptors.

Fig. 1.

Association of ITAM-bearing adaptors and cell surface receptors. The α and β chains of T cell receptor (TCR) associate with the CD3 family of ITAM-containing adaptors. CD3δ, γ, and ε each contain one cytoplasmic ITAM and have an extracellular Ig domain, whereas CD3ζ has three ITAMs and forms a disulfide-bonded homodimer. CD3δ, γ, and ε are found exclusively in T cells, whereas ζ is also found in NK cells, where it associates with several activating NK receptors. The B cell receptor (BCR) is composed of surface immunoglobulin (sIg) associated with a disulfide-bonded heterodimer of Igα and Igβ, each of which contains one ITAM and an extracellular Ig domain similar to CD3δ, γ, and ε. Igα and Igβ are exclusively expressed in B cells. The FcεRIγ (FcRγ) ITAM-containing adaptor is expressed in myeloid cells, B cells, and NK cells, as well as in some T cells. It has a minimal extracellular domain with one cytoplasmic ITAM and forms a disulfide-bonded homodimer (in some cells FcεRIγ forms a disulfide-bonded heterodimer with CD3ζ). FcεRIγ can associate with various receptors, including FcαRI (also known as CD89) depicted here. DAP12 has a structure and expression pattern similar to that of FcRγ. In myeloid cells and NK cells, DAP12 associates with various receptors, including TREM-2 depicted here.

Activating signals through ITAMs in transmembrane signaling adaptors have been well studied over the past two decades [for reviews, see (25)]. After receptor ligation, the tyrosines in the ITAM are typically phosphorylated by Src family kinases. These phosphorylated tyrosines then serve as docking sites for the recruitment of the Syk family tyrosine kinases Syk, ZAP-70, or both, which bind the doubly phosphorylated ITAM through tandem Src homology 2 (SH2) domains. Syk recruitment and activation leads to the activation of many downstream signaling pathways through phosphorylation of signaling and adaptor proteins. These pathways include phospholipase C–γ activation leading to calcium flux and Ras activation leading to extracellular signal–regulated kinase (ERK) activation (Fig. 2).

Fig. 2.

ITAM signaling cascade in hematopoietic cells. After ligand binding to receptor, the tyrosines within the ITAM are phosphorylated by members of the Src family of tyrosine kinases. This allows the recruitment and activation of Syk (in myeloid cells, B cells, and NK cells), ZAP-70 (in NK cells and T cells), or both, which in turn leads to phosphorylation of the BLNK or SLP-76 family of adaptor proteins. This is a central event that results in the activation of multiple signaling cascades, including Tec family kinases (Tec, ITK, or BTK), PLC-γ leading to calcium flux and NF-AT translocation to the nucleus, DAG production leading to ERK activation, and PKC activation leading to activation of the transcription factor NF-κB. BLNK or SLP-76 also couple receptor ligation to the activation of the GTP/GDP exchange factor Vav, resulting in activation of c-Jun N-terminal kinase (JNK). Abbreviations: DAG, diacylglycerol; IP3, inositol 1,4,5-trisphosphate; NF-AT, nuclear factor of activated T cells; NF-κB, nuclear factor kappa B; PI3K, phosphatidylinositol 3-kinase; PKC, protein kinase C; PLC-γ, phospholipase C–γ.

Activating signals through ITAMs in immune cells are often counterbalanced by inhibitory signals through receptors that contain immunoreceptor tyrosine-based inhibitory motifs (ITIMs) in their cytoplasmic tails (6). ITIMs are defined by the sequence (I/V/L/S)xYxx(L/V) and bind to the protein tyrosine phosphatases SHP-1 or SHP-2 or the inositol lipid phosphatase SHIP (7). These phosphatases dephosphorylate signaling proteins or lipids, thereby dampening responses. The balance of activating signals through ITAMs and inhibitory signals through ITIMs is thought to determine the level of response for many immune receptor systems. In B cells, the ITIM-containing receptor CD22 regulates signaling by the BCR, Igα, and Igβ complex. In NK cells, activating receptors associated with DAP12, with FcRγ, CD3ζ, or with both FcRγ and CD3ζ are regulated by the major histocompatibility complex (MHC) class I–binding, ITIM-containing inhibitory receptors [Ly49 receptors in mice and killer cell immunoglobulin-like receptors (KIRs) in humans]. Particularly interesting are cases in which activating and inhibitory receptors on the same cell bind to the same ligand, such as FcγRIII (also known as CD16) (ITAM-associated) and FcγRII (also known as CD32) (ITIM-containing) on myeloid cells, where the response to ligand binding is determined by the relative abundance of the two receptors. In addition to ITIM-containing receptors, members of the SLAM subfamily of the CD2 family of receptors in some circumstances can also inhibit ITAM-mediated activation of NK cells. These receptors contain one or more immunoreceptor tyrosine-based switch motifs (ITSMs) in their cytoplasmic domains that associate with activating signaling adaptors (SAP) or inhibitory signaling adaptors (EAT-2, ERT, or both) to mediate opposing functions (810). Thus, numerous inhibitory receptors have evolved to regulate the activation of cells by ITAM-bearing receptor complexes.

Inhibitory ITAM Signaling

Although activation through ITAM-containing signaling adaptors has been documented for many receptors and cell types, two recent studies demonstrate that two of the "activating" adaptors can also mediate inhibitory signals in myeloid cells (11, 12). The DAP12 adaptor, through an undefined receptor, inhibited the response of macrophages to pathogen-initiated signals through Toll-like receptors (TLRs) (11). Similarly, FcRγ, the signaling chain for the IgA-specific Fc receptor (FcαRI), inhibited IgG-mediated phagocytosis in monocytes and IgE-mediated degranulation in mast cell transfectants (12). Inhibition by both DAP12 and FcRγ required the presence of the tyrosine residues within the ITAMs—the same tyrosines that are required for ITAM-mediated activating signals—and, therefore, was not due to other signaling motifs present in the adaptors. Additionally, DAP12 and FcRγ signals inhibited functional responses that these adaptors can elicit themselves when cross-linked—macrophage inflammatory cytokine production and mast cell degranulation, respectively. These two studies highlight additional mechanisms that myeloid cells use to regulate their functional responses.

Inhibition by DAP12 in macrophages

The inhibitory effect of DAP12 was revealed by examining macrophages from DAP12-deficient mice (11). These macrophages lack the ability to signal through many potential receptors that pair with DAP12 in myeloid cells, including MDL-1, TREM-1, TREM-2, TREM-3, TREM-5 (also known as CD300LB), PILRβ, MAIR-II (also known as CD300c), SIRPβ, CD200R2, CD200R3, and CD200R4 (13). When DAP12-deficient macrophages were activated through TLRs using different pathogen products or their mimetics, such as lipopolysaccharide (LPS) or CpG DNA, higher concentrations of the proinflammatory cytokines tumor necrosis factor (TNF), interleukin-6 (IL-6), and IL-12 p40 were produced than was produced by wild-type macrophages (11). This suggests that in wild-type macrophages, a DAP12 signal is required to limit TLR-induced cytokine production. This inhibitory effect of DAP12 was counterintuitive given that cross-linking DAP12-paired receptors, such as TREM-1 or MAIR-II, in macrophages has been shown to induce the secretion of these same cytokines (14, 15). The negative regulation of TLR signaling was seen in vivo, with the DAP12-deficient mice having increased TNF in the sera in response to LPS injection and increased susceptibility to endotoxic shock in a model system in which mice were treated with 𝒟-galactosamine and LPS (11). The DAP12-deficient mice also have enhanced responses to infection with the intracellular bacterium Listeria monocytogenes, consistent with increased proinflammatory cytokine production in vivo.

The direct mechanism for DAP12 inhibition of TLR responses is not clear, but it requires the ITAM tyrosines and involves the tyrosine kinase Syk, because the phenotype of Syk-deficient macrophages is identical to that of DAP12-deficient macrophages—increased inflammatory cytokine production in response to TLR-mediated activation. Downstream of TLR ligation, there is a selective effect of DAP12 deficiency on the mitogen-activated protein kinase (MAPK) ERK—after treatment with LPS, DAP12-deficient macrophages have increased ERK phosphorylation, a measure of ERK activation, in comparison with wild-type macrophages. Although the nuclear factor κB (NF-κB) transcription factor is well documented to regulate inflammatory cytokine production downstream of TLRs, ERK signals are also required for TLR-induced TNF production (16). These results suggest that a signal downstream of Syk negatively regulates the ERK pathway. Members of the MAPK phosphatase family are candidates for the mediators of this Syk-dependent inhibition of ERK activation. It is also possible that DAP12 recruits a phosphatase more proximal to the plasma membrane in a Syk-dependent manner, such as the tyrosine phosphatase SHP-1 or a related phosphatase (Fig. 3), which would be similar to the mechanism by which FcαRI is inhibited (discussed below). Additionally, DAP12 may induce specific negative regulators of TLR signaling, such as IRAK-M, though this is not a likely mechanism, because DAP12 was also shown to inhibit FcγRIII-mediated activation, which signals in a manner distinct from that of the TLRs (Fig. 3). Although induction of a specific negative regulator by the inhibitory DAP12 signal is appealing, it is also plausible that the inhibitory function of DAP12 is due to sequestration of a common signaling intermediate in the DAP12 and TLR pathways (Fig. 3). Though there are not many similarities within these signaling pathways, proteins in the ERK pathway are candidates. Although this study showed a striking effect of the presence of DAP12 on inflammatory cytokine production by macrophages, the receptor(s) that initiates the inhibitory DAP12 signal remains to be determined.

Fig. 3.

Possible mechanisms for DAP12-mediated inhibition of TLR responses. (A) In the absence of inhibition, a TLR signal through a critical signaling component X is required for full activation of macrophages. (B) An inhibitory signal induced by a DAP12-associated receptor causes the recruitment and sequestration of the critical signaling component X so that a lesser amount is available for signaling downstream of the TLR, resulting in reduced TLR-mediated activation. (C) An inhibitory signal induced by a DAP12-associated receptor causes the recruitment of a phosphatase (Y) that dephosphorylates the critical signaling component X, reducing the activation of X and, therefore, resulting in lower levels of cellular activation. (D) An inhibitory signal induced by a DAP12-associated receptor induces the production or activation of Z, a specific negative regulator of TLR signaling.

Inhibition by the FcαRI-FcRγ complex in macrophages and RBL cells

In their study of inhibition through the FcRγ ITAM-containing adaptor, Pasquier et al. (12) investigated the mechanism for the long-recognized phenomenon by which monomeric IgA, such as is found in serum, can inhibit IgG-mediated activation of myeloid cells, which in turn affects a plethora of functional consequences, including phagocytosis, oxidative burst, and cytokine production (17). Consistent with the in vitro findings, patients deficient in IgA have an increased incidence of autoimmune disease and allergic disorders (18). On human blood myeloid cells, IgA binds FcαRI (also known as CD89), which associates with the FcRγ ITAM-containing adaptor (17). Pasquier et al. first identified FcαRI as a receptor that can inhibit IgG-mediated phagocytosis in human monocytes (12). A FcαRI-FcRγ chimera (composed of the extracellular domain of FcαRI, the transmembrane domain of FcαRI with a R209L mutation, and the intracellular domain of FcRγ) transfected into the rat mast cell line RBL was then used to identify the requirements for FcαRI-mediated inhibitory signaling (Fig. 4). The RBL transfectants were activated by cross-linking the IgE receptor (FcεRI), using degranulation as a readout. The degree of cross-linking of the FcαRI-FcRγ chimera was critical for inhibition—Fab fragments of an antibody to FcαRI caused inhibition, whereas Fab fragments of the same antibody with additional cross-linking induced activation of the RBL cells transfected with the FcαRI-FcRγ chimera. This was also seen with natural ligands—the low-valency interaction of monomeric IgA with FcαRI caused inhibition, whereas multivalent IgA complexes induced activation (Fig. 4). Of note, Fab fragments of antibodies recognizing other receptors (FcεRI and FcγRIIb) that signal through the FcRγ ITAM-bearing adaptor were not able to transduce inhibitory signals. The authors used this finding to support the idea that the receptor complex composed of FcαRI and FcRγ subunits is unique in its ability to initiate ITAM-dependent inhibitory signals. Alternatively, the affinity of the antibodies used for cross-linking FcεRI and FcγRIIb may not be appropriate to induce the inhibitory signals.

Fig. 4.

Inhibition of FcεRI signal transduction by an FcαRI-FcRγ chimera. In the report by Pasquier et al. (12), an FcαRI-FcRγ chimera was introduced into the RBL mast cell line expressing an endogenous FcεRI (composed of α, β, and γ subunits). The authors used two experimental models. (A) The RBL cells were pretreated either with monomeric IgA (blue) (low avidity) or with monomeric IgA plus a secondary cross-linking F(ab)′2 (green) (high avidity), followed by activation with antigen to induce degranulation through FcεRI using anti-DNP IgE and DNP (purple). Treatment with the low-avidity monomeric IgA resulted in the inhibition of IgE-mediated degranulation. (B) The RBL cells were pretreated with a Fab fragment of a monoclonal antibody that recognizes FcαRI (red) (low avidity) or a Fab fragment plus cross-linking F(ab)′2 (green) (high avidity). Treatment with the low-avidity Fab alone resulted in the inhibition of IgE-mediated degranulation.

As with inhibition through DAP12 (11), suppression through the FcαRI-FcRγ chimera required the tyrosines in the ITAM (12). Under the conditions that resulted in inhibition, the amount of tyrosine phosphorylation of the ITAMs was less than when the cells were activated by extensive cross-linking of the FcαRI-FcRγ chimera. Although Syk was associated with the FcαRI-FcRγ chimera under both inhibitory and activating conditions, there was slightly less Syk associated with the receptors during inhibition. More striking was the association of the protein tyrosine phosphatase SHP-1 with the receptor, which was observed only under inhibitory conditions. This suggested that the mechanism for inhibition was the dephosphorylation of signaling components required for activation through the high-affinity IgE receptor by IgE cross-linking. This was indeed the case—Syk, LAT (linker of activated T cells), and ERK phosphorylation downstream of FcεRI cross-linking were all decreased when the RBL cells were stimulated under inhibitory conditions through the FcαRI-FcRγ chimera. This suggests cross talk between different myeloid receptors expressed in a cell, in that recruitment of SHP-1 to the FcαRI-FcRγ chimera inhibited signaling downstream of FcεRI. Although other studies have shown the recruitment and activation of the tyrosine phosphatases SHP-1 and SHP-2 and the inositol phosphatase SHIP to the ITAMs in the FcγRIIa receptors and to the ITAMs in the high-affinity IgE receptor complexes in myeloid cells (1921), those studies have pointed to a role for these phosphatases in controlling positive signaling through the same receptor, not a heterologous receptor. Furthermore, ERK activation was required for SHP-1 association with the FcαRI-FcRγ chimeric receptor in the transfected RBL cells—almost no SHP-1 was coimmunoprecipitated with the FcαRI-FcRγ chimera in the presence of an inhibitor of MEK1, the upstream kinase responsible for activating ERK (12).

Together, the studies by Hamerman et al. (11) and Pasquier et al. (12) make a compelling argument for ITAMs as dual signaling switches—in some circumstances activating and in other circumstances inhibitory toward heterologous receptors. The study of FcαRI-FcRγ chimera signaling clearly shows that this switch is controlled by the avidity of the ligand cross-linking the receptor (12). Low-avidity interactions resulted in inhibition of signaling of heterologous receptors (such as inhibition of the high-affinity IgE receptor by low-avidity ligands of the FcαRI-FcRγ chimera), whereas high-avidity interactions resulted in activation through the ligand-bound receptor itself (Fig. 4). A similar model was suggested for the mechanism by which DAP12 mediated inhibition of TLR signaling (11); however, confirmation awaits identification of the ligand(s) and DAP12-associated receptors that mediate this inhibitory signal. Thus far, a high-affinity and high-avidity interaction—antibody against the receptor plus cross-linking—has been the stimulus for all reports of activation of myeloid cells through DAP12-associated receptors and stimulation of cytokine production. Therefore, we hypothesize the existence of a low-affinity or low-avidity natural ligand for a DAP12-associated macrophage receptor that would result in the inhibitory signal. Furthermore, the ligand must be expressed on macrophages themselves because pure macrophage cultures showed the DAP12-dependent inhibition. Suppression of macrophage inflammatory responses by DAP12 may contribute to the quiescent state of macrophages in an uninfected host or in the presence of low-level infection. An increase in the affinity or abundance of a ligand for a DAP12-associated receptor during an infection might allow macrophages to switch from an inhibitory DAP12 signal to an activating signal that synergizes with or augments the TLR-induced inflammatory response.

Inhibition of FcεRI signals in mast cells

An early description of inhibition by ITAM-containing receptors was reported by Torigoe et al. (22) in a study examining activation of RBL mast cells through FcεRI with IgE and antigen. FcεRI is composed of three chains: the ligand-binding α chain and the ITAM-bearing β and γ subunits. RBL cells were coated with two IgE molecules against two antigenically distinct haptens that bind with different affinities to their respective IgE. The cells were then stimulated with a mixture of the two haptens, with the low-affinity hapten used at ~40 times the concentration of the high-affinity hapten. Robust FcεRI phosphorylation was found when cells were treated with either hapten alone or with a mixture of the two haptens. In contrast, phosphorylation of the downstream signaling components Syk, Pyk2, and ERK was decreased when the cells were exposed to a mixture of the low- and high-affinity haptens, to a level much lower than that seen with the low-affinity hapten alone. Interestingly, the further downstream the signaling component from the receptor, the greater the decrease in phosphorylation; that is, the decrease in phosphorylation of the proximal signaling component Syk was much smaller than the decrease in phosphorylation of ERK, which is further downstream from the receptor. This excess of low-affinity hapten was also able to inhibit mast cell degranulation, the functional outcome downstream of FcεRI cross-linking. With the use of an immunoprecipitation technique that differentiated between the FcεRI bound to high-affinity or low-affinity ligand, the amount of phosphorylation of the receptor ITAMs bound to the high-affinity ligand was determined to be decreased by about two-thirds when an excess of low-affinity hapten ligand was also bound to FcεRI on these cells. The authors hypothesized that the inhibitory effects were due to sequestration of Lyn, the Src family kinase responsible for ITAM phosphorylation, by the high numbers of IgE receptors cross-linked with the low-affinity ligand. However, it is equally plausible that phosphatase recruitment to the ITAMs in inhibitory conditions caused the reduced activation of downstream signaling components, similar to that shown by Pasquier et al. for FcαRI (12).

Inhibition by DAP12 in IPC and microglia

Interferon (IFN) α/β–producing cells (IPCs) respond to specific TLR stimuli, such as CpG DNA, resulting in the production of large quantities of type I IFNs (23). In mice and humans, DAP12-associated receptors present on IPCs inhibited TLR-induced cytokine responses (24, 25). Fuchs et al. reported that the DAP12-associated receptor NKp44 is present on human IPCs and that cross-linking this receptor with a monoclonal antibody inhibited the secretion of IFN-α induced by treatment with CpG DNA (24). Although this NKp44-induced inhibition may be mediated by ITAM-dependent DAP12 signaling, it is also possible that NKp44 itself delivers this negative signal because the cytoplasmic domain of NKp44 contains a putative ITIM-like sequence (26). Although the tyrosine in this putative ITIM can be phosphorylated, it has not been shown to associate with any phosphatase or to mediate inhibition in NK cells. We propose that the DAP12 ITAM itself is responsible for the inhibition. Siglec-H, a DAP12-associated receptor expressed specifically in mouse IPCs, also inhibits CpG-induced IFN-α/β production, and DAP12-deficient IPCs produce higher concentrations of type I IFNs in vitro and in vivo than do wild-type IPCs (25). This is similar to the effect of DAP12 deficiency on TLR-induced cytokine production by mouse macrophages (11). Interestingly, inhibition through NKp44 and Siglec-H was demonstrated using plate-bound antibody to cross-link these receptors (24, 25). This contrasts with the outcome of cross-linking other DAP12-associated receptors with antibodies, which results in activation rather than inhibition of the receptor-bearing cells. Negative signaling through another DAP12-associated receptor, TREM-2, has been described by Takahashi et al. in myeloid lineage microglial cells (27). Although the main focus of this study was on the positive role of TREM-2 in the phagocytosis of apoptotic cells by microglia, TREM-2 was also shown to modestly inhibit induction of TNF and iNOS (inducible nitric oxide synthase) mRNA. The authors suggested that this mechanism prevents an inflammatory response by microglia after the phagocytosis of apoptotic cells.

Inhibition by FcRγ in macrophages

Mosser and colleagues reported that signaling through FcRs for IgG on mouse macrophages inhibited TLR4-dependent LPS signaling (2830). The original observation was that IgG-coated sheep red blood cells selectively inhibited macrophage LPS-induced IL-12 production, but not TNF production (29). Through the analysis of gene-disrupted macrophages, this was shown to be mediated by a receptor that pairs with FcRγ, but not FcγRIII, and not by the ITIM-containing FcγRII (30). The authors concluded that it must be mediated by FcγRI, at that time the only other receptor for IgG known to be expressed on macrophages; however, it is equally possible that it is caused by the newly identified FcγRIV (31). Additionally, the mechanism for this suppression of IL-12 was found to be increased IL-10 secretion by IgG-treated macrophages (28, 30). This is, therefore, distinct from the DAP12-mediated suppression of LPS responses discussed above (11), because DAP12 signals inhibited, not increased, IL-10 production. DAP12 also inhibited TNF production, as well as IL-12 production.

Inhibition by Igα in B cells

Inhibitory ITAM signaling has also been suggested for B cells. The BCR includes a heterodimer composed of Igα and Igβ, which are ITAM-containing adaptors. Many studies have investigated the role of these ITAM adaptors in B cell development and function. Kraus et al. examined B cells from mice that were lacking the cytoplasmic region of Igα (32). Although these mice had previously been shown to have defective B cell maturation and, therefore, reduced numbers of mature splenic B cells, this study examined the response of immature B cells in the bone marrow expressing a monoclonal BCR to a soluble autoantigen. Unexpectedly, they found that the immature B cells lacking the cytoplasmic tail of Igα had an increased response to autoantigen, resulting in the loss of BCR specificity for the autoantigen through endogenous light-chain rearrangement. This could also be seen by a heightened calcium flux to BCR cross-linking in the mutant immature B cells. Because the Igα cytoplasmic domain contains tyrosine residues outside the ITAM and the developmental stage of the immature B cells assayed in the wild-type and the Igα cytoplasmic truncation mice may have been different, the authors generated a specific ITAM-mutant mouse in which the tyrosines of the Igα ITAM had been mutated to phenylalanine and, therefore, could not be phosphorylated and propagate a signal (33). These mice had a less severe developmental defect than the Igα cytoplasmic truncation mice, which allowed the analysis of mature splenic B cells. The mature splenic B cells in the Igα ITAM-mutant mice had an increased calcium flux in response to BCR cross-linking, similar to the immature bone marrow B cells from the mice with an Igα cytoplasmic truncation. The splenic B cells containing the Igα ITAM-mutant had twice the surface abundance of BCR in comparison with wild-type B cells, as well as slightly (17%) higher levels of CD19 and slightly (19%) lower levels of CD22 than wild-type B cells. Because CD19 and CD22 are thought to help set the activation threshold of B cells, the authors proposed that the lack of an activating signal from Igα in the B cells led them to adapt by increasing their surface BCRs and decreasing their threshold for activation, resulting in hyperresponsiveness. Several studies have also implicated the Igα cytoplasmic domain in BCR internalization after cross-linking (34, 35); therefore, it is also possible that the increased BCR signal was due to lack of BCR internalization after signaling in the Igα-mutant B cells, leading to a sustained and heightened response. An alternative explanation for the hyperresponsiveness of the Igα-mutant B cells is that Igα can, in some circumstances, transduce inhibitory signals through its ITAM, similarly to DAP12 and FcRγ, and when this inhibition is lacking, heightened responses result.

Inhibition of BCR signaling in B cells has also been reported for the LMP2A protein of Epstein-Barr virus (EBV). LMP2A has 12 transmembrane-spanning regions with both the N terminus and C terminus located within the cell. It has minimal extracellular domains and contains an ITAM in its N-terminal cytoplasmic domain (36). In vitro, the LMP2A ITAM blocks BCR-induced signaling in EBV-transformed B cells, possibly by sequestering the Src family kinase Lyn (37, 38). This has been proposed as a mechanism for maintaining viral latency by preventing normal B cell activation (39). In contrast, in transgenic mice, LMP2A provides the tonic signal required for B cell survival, which is normally provided by the BCR. This potentially could allow the survival of EBV-infected B cells in infected humans (40). Whether the inhibitory signal or the survival signal provided by the LMP2A ITAM is the predominant in vivo function of LMP2A during EBV infection remains to be determined.

Signal strength and T cell development

Signals initiated by high- and low-affinity ligands for the TCR and CD3 complex on T cells have also been shown to transduce different signals, resulting in distinct functional outcomes both for developing T cells in the thymus and for mature T cells in the secondary lymphoid organs. During development, T cells make several lineage choices that may be mediated by the strength of signal through the ITAM-bearing CD3 subunits of the TCR complex. The first of these is the decision by early thymocyte progenitors to adopt the αβ or γδ T cell lineage, marked by the expression of the appropriate TCR. Strong signals through the TCR, as are usually given by γδ TCRs, promoted a γδ T cell fate, whereas weak signals, typical of the pre-TCR (TCRβ + pre-Tα), promoted an αβ T cell fate (41, 42). In both studies, the authors were able to change the fate of the T cells by experimentally altering the signal strength, particularly through the ERK pathway. This is reminiscent of another lineage choice made by developing T cells—that of CD4 versus CD8 lineage—which typically matches the responsiveness of the TCR for MHC class II and MHC class I, respectively. Strong αβ TCR signals resulting in higher, sustained ERK activation promote CD4 lineage commitment, whereas weak αβ TCR signals resulting in lower, transient ERK activation promote CD8 lineage commitment (43, 44). ERK activation has also been implicated in signaling for positive and negative selection of CD4+CD8+ thymocytes (45).

Although differences in ligand affinity that affect the intensity and duration of ERK activation through the TCR-CD3 complex play an important role during many aspects of T cell development, there is as yet no evidence that different-affinity ligands cause these TCR complexes to differentially affect other (non-TCR) receptors in the responding T cell. However, low-affinity "antagonist" peptide-MHC ligands for a TCR can inhibit the response of a T cell that is exposed to high-affinity agonist peptide-MHC ligands for the same TCR presented on the same antigen-presenting cell (46). Furthermore, Dittel et al. showed that in a T cell clone expressing two distinct TCRs, presentation of an antagonist peptide for one TCR resulted in inhibition of functional responses induced by presentation of an agonist peptide to the other TCR (47). Stimulation with the antagonist peptide alone led to the deposition of SHP-1 on the unengaged TCR, suggesting a mechanism for TCR antagonism. Stefanova et al. showed that this ability of antagonist peptides to inhibit responses to agonist peptides was controlled by a negative-feedback loop involving ERK, SHP-1, and Lck, a Src family kinase that phosphorylates the CD3 ITAMs (48). In T cells presented with low-affinity antagonist peptides, SHP-1 associated with Lck and prevented complete CD3ζ ITAM phosphorylation and subsequent signaling for cellular activation. In T cells presented with high-affinity agonist peptides, rapid and robust ERK activation caused the direct serine phosphorylation of Lck, which interfered with SHP-1 binding and, therefore, allowed full activation of the TCR signaling cascade, including complete CD3ζ ITAM phosphorylation. Interestingly, inhibition of ERK activation with a MEK inhibitor after presentation of an agonist peptide reduced the CD3ζ phosphorylation to amounts seen with antagonist peptides. That antagonist peptides can reduce the response to agonist peptides when both are presented to the same T cell suggests that SHP-1 can desensitize other TCRs on the same cell. The inhibition mediated by the FcαRI-FcRγ complex in myeloid cells in which SHP-1 associated with the receptor itself after a weak signal may also be the result of an interaction between SHP-1 and a Src family kinase. In contrast to the role of ERK in removing SHP-1 from the TCR complex, Pasquier et al. showed that ERK activity is required for SHP-1 association with FcRγ (12). Therefore, these negative-feedback loops involving ERK and SHP-1 downstream of ITAM-containing adaptors may be distinct in myeloid cells and T cells.


It is now clear that ITAM-containing adaptors can mediate not only activating signals, but also inhibitory signals, propagated through tyrosine residues within the ITAMs, which help to set the activation threshold of the cell. Several distinct mechanisms may be responsible for negative regulation through these ITAM-bearing receptors, depending on the particular receptor or cell type. In addition, the amount of ligand or the affinity of ligand available to engage these ITAM-bearing receptor complexes may dictate whether the net response is to enhance or suppress cellular activation initiated by other signaling receptors. Inhibition mediated by ITAM-bearing receptors might be a result of (i) recruitment and activation of a phosphatase or other specific inhibitory signaling component; (ii) recruitment and sequestration of kinases—thereby depriving other receptors of these enzymes critical for activation; or (iii) induction of immune-suppressive cytokines that can dampen cellular responses initiated by other activating receptors. Because this is an emerging field of study, many questions remain to be answered regarding the mechanism of inhibition, both for the ligand interactions that lead to inhibition, as well as the signaling intermediates and pathways used. Unraveling the biochemical basis for negative ITAM signaling undoubtedly will provide new insights into the fine-tuning necessary to regulate immune responses to ensure protective immunity, yet avoid autoimmunity.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.
  30. 30.
  31. 31.
  32. 32.
  33. 33.
  34. 34.
  35. 35.
  36. 36.
  37. 37.
  38. 38.
  39. 39.
  40. 40.
  41. 41.
  42. 42.
  43. 43.
  44. 44.
  45. 45.
  46. 46.
  47. 47.
  48. 48.
  49. 49.
View Abstract

Stay Connected to Science Signaling

Navigate This Article