A new look at TCR signaling to NF-κB

Abstract

Antigen stimulation of TCR signaling to NF-κB is required for T cell proliferation and differentiation of effector cells. The TCR-to-NF-κB pathway is generally viewed as a linear sequence of events in which TCR engagement triggers a cytoplasmic cascade of protein-protein interactions and post-translational modifications, ultimately culminating in the nuclear translocation of NF-κB. However, recent findings suggest a more complex picture in which distinct signalosomes, previously unrecognized proteins, and newly identified regulatory mechanisms play key roles in signal transmission. In this review, we evaluate recent data and suggest areas of future emphasis in the study of this important pathway.

The current consensus model of TCR signaling to NF-κB

In the last decade, much progress has been made in defining molecular mechanisms by which the TCR activates the NF-κB transcription factor. Most of the key mediators in this cascade are now defined, and many key signal transmission mechanisms have been elucidated [1–4] (Figures 1 and 2). The general consensus understanding is that engagement of the TCR by an MHC-antigen complex initiates downstream CD3 ITAM phosphorylation by the Src family kinases, FYN and LCK. Phosphorylated CD3 activates the T cell specific tyrosine kinase, ZAP-70, which phosphorylates the adapter proteins LAT and SLP-76, causing SLP-76 to bind to VAV1. The VAV1-SLP76-ITK complex activates PLCγ1, generating IP3 and DAG, which ultimately trigger calcium release and PKC activation, respectively. Activation of a specific PKC isoform, PKCθ, connects the above described TCR proximal signaling events to distal events that ultimately lead to NF-κB activation. Importantly, PKCθ activation is also driven by engagement of the T cell costimulatory receptor CD28 by B7 ligands on antigen presenting cells. This molecular interaction activates PI3K, inducing recruitment of PDK1 and AKT to the plasma membrane. At the immune synapse (IS), PDK1 phosphorylates and activates PKCθ. PKCθ-mediated phosphorylation of CARMA1 triggers a conformational change, causing CARMA1 to bind to BCL10 and MALT1, forming the CBM complex. Through a mechanism that may involve TRAF6, both BCL10 and MALT1 become polyubiquitinated. The IKK complex is then recruited to the CBM complex via the IKKγ polyubiquitin binding motif. This association leads to polyubiquitination of IKKγ and phosphorylation of IKKβ by TAK1, activating IKKβ. IKKβ then phosphorylates IκBα, triggering its proteasomal degradation, enabling nuclear translocation of canonical NF-κB heterodimers comprised of p65 (RELA) and p50 proteins. Once in the nucleus, NF-κB governs the transcription of numerous genes involved in T cell survival, proliferation, and effector functions.

Figure 1. New developments in the TCR-to-NF-κB signaling pathway.

The TCR transmits signals through the LAT-SLP76 complex, and possibly through GLK to activate PKCθ. CD28 signals through PI3K and PDK1 to activate PKCθ. Activated PKCθ induces formation of the CARMA1, BCL10, MALT1 (CBM) complex. Newly identified proteins acting on CARMA1 and BCL10 are shown on the left. MALT1 enzymatic protein cleaving functions and caspase-8 recruitment are shown on the right. The CBM complex transmits activating signals to IKK through the ubiquitin ligases TRAF6, TRAF2 and/or MIB2. IKK phosphorylates IκBα leading to IκBα ubiquitination and degradation allowing NF-κB nuclear translocation and gene transcription.

Figure 2. Negative regulation of TCR-to-NF-κB signaling.

The TCR/CD28 activation signal, shown by green arrows, passes through multiple intermediate signaling proteins ultimately causing NF-κB activation. Negative modulation activities are depicted by red arrows. Yellow ovals indicate membrane proximal and cytosolic signalosomes described in detail in the text.

Recent data suggest that aspects of the consensus model for TCR signaling are overly simplistic, and that additional molecules play a role in the TCR-to-NF-κB cascade. Here, we summarize data suggesting that multiple signalosomes participate in TCR activation of NF-κB, and describe the negative regulatory mechanisms that control this pathway. We also discuss evidence for connections between control of NF-κB activation and other cellular processes, such as actin remodeling. Overall, the emerging picture is that the TCR-to-NF-κB signaling cascade is a crucial nexus which both governs and is regulated by a diverse network of T cell biological processes.

New developments in the TCR-to-NF-κB pathway

Deletion of the genes encoding PKCθ and CBM complex proteins results in impaired TCR-induced NF-κB activation. Recent work also identifies a number of additional molecules that regulate this pathway (Figure 1).

PKCθ

Phosphorylated PKCθ connects LAT and SLP76 with the CBM complex [4, 5]. The protein kinase PDK1 is considered essential for PKCθ activation as PDK1-deficient Jurkat and primary CD4 T cells show a defect in PKCθ phosphorylation and NF-κB activation [6, 7]. However, there is a lack of in vitro evidence that PDK1 directly phosphorylates PKCθ. Moreover, PDK1 activation is dependent on CD28 engagement, while PKCθ IS translocation and NF-κB activation can occur in a purely CD3-dependent manner, without participation of CD28 [6, 8–10]. These observations suggest that another kinase links the TCR-CD3 complex with PKCθ. Indeed, GLK, a SLP76-regulated kinase, was recently reported to directly phosphorylate PKCθ both in vitro and in primary T cells and T cell lines in response to TCR stimulation [11]. Additionally, GLK-deficient murine lymph node cells exhibit reduced PKCθ- and IKK-phosphorylation, correlating with reduced cytokine and antibody production. Collectively, these data suggest that PDK1 and GLK1 might function together to induce PKCθ phosphorylation and activate NF-κB (Figure 1). Alternatively, GLK and/or PDK1 may be utilized in an exclusive manner to phosphorylate PKCθ, depending on the activation and/or differentiation state of the T cells, the type of antigen-bearing stimulatory cell, etc. Elucidation of such details will require a careful comparison of the GLK and PDK1 knockout models under a variety of T cell activation paradigms.

CARMA1

CARMA1, a critical target of PKCθ phosphorylation, resides in lymphocytes in an inactive state. Extensive CARMA1 mutagenesis data suggest that this inactive state is maintained by intramolecular interactions that prevent the CARMA1 caspase recruitment domain (CARD) from interacting with the CARD of Bcl10 [12, 13]. PKCθ phosphorylates human CARMA1 at three serine residues, S552, S645 and S637 (S564, S657, S649 in the mouse) [1]. Phosphorylation at S552 and S645 is critical for the CARMA1 conformational change that enables binding to BCL10-MALT1, leading to further signal transmission [14, 15] (CBM complex post-translational modifications are reviewed in [1]). In contrast, S649 phosphorylation suppresses CARMA1-mediated NF-κB and JNK activation [16]. While PKCθ induces CARMA1 S649 phosphorylation in vitro and PKC inhibitors block the phosphorylation event in vivo, there is to date little evidence that PKCθ directly phosphorylates CARMA1 at S649 in living cells. Also, the phosphorylation events of opposing function do not occur simultaneously; the activating phosphorylation at S552/S645 is early and transient, whereas the inhibitory phosphorylation at S649 is delayed, but sustained for a longer duration. Considering both the absence of evidence of direct PKCθ phosphorylation of CARMA1 S649 and the rapid PKCθ activation kinetics post-TCR activation, it remains a strong possibility that the delayed CARMA1 S649 phosphorylation is mediated by an unidentified PKCθ-dependent kinase that is activated and/or interacts with CARMA1 in a delayed manner (Figure 2).

Apart from PKCθ, three additional protein kinases, HPK-1, CK1α and CaMKII are capable of phosphorylating CARMA1 and regulating CBM complex activity (Figure 1). In Jurkat T cells, siRNA knockdown of HPK-1 reduced IKK enzymatic activity, NF-κB nuclear translocation, and T cell survival [17]. HPK-1 over-expression produced the opposite effect [18]. Demonstration of TCR-regulated interaction between HPK1 and CARMA1 explained the ability of HPK-1 to stimulate NF-κB activation. These authors also provided evidence that HPK-1 phosphorylates CARMA1 at S551, a site distinct from the residues modified by PKCθ [18]. However, the overall role of HPK-1 in lymphocyte function remains a hotly debated issue, as mouse knockout data demonstrate that HPK-1 suppresses lymphocyte activation, due to its inhibitory effect on the proximal signaling protein SLP76 [19, 20] (see Negative regulation of TCR signaling to NF-κB, below).

CK1α, another kinase that acts via modification of CARMA1, was identified by mass spectrometry as a CARMA1 binding partner. CK1α induces TCR-mediated NF-κB activation in Jurkat and IL-2 secretion in primary human T cells [21], probably by recruiting activated IKKβ to the CBM complex. Although CK1α phosphorylates CARMA1 at S608, limiting NF-κB activation, the positive CK1α function supercedes its negative role. [21]. Thus CK1α enzymatic activity opposes its IKK recruiting function, indicating that CK1α is also a bi-functional regulator of the TCR-to-NF-κB pathway.

Accumulating data suggest that the TCR-to-NF-κB pathway is also regulated by the protein kinase, CaMKII. TCR stimulation results in a rapid increase in T cell cytosolic calcium, which binds CaM (Calmodulin), triggering activation of CaMKII and the phosphatase, calcineurin (see also BCL10 section, below) (Figure 1). Both enzymes influence the TCR-to-NF-κB pathway. In Jurkat, following APC+antigen or CD3+CD28 stimulation [22, 23], CaMKII is recruited to the IS. CaMKII inhibition by siRNA knockdown (γ or δ isoform) or by the pharmacological inhibitor, KN-93, reduces NF-κB activation [23]. However, the mechanism of CaMKII action is unclear. In a cell-free system, CaMKII can phosphorylate CARMA1 at S109 [23] and BCL10 at S138 [24] or at S48 and T91 [22]. Studies in Jurkat showed CARMA1 S109 phosphorylation assists in CARMA-BCL10 binding [23], and BCL10 T91 phosphorylation promotes K63-polyubiquitination of BCL10 and signaling to IKK [22]. In contrast, there is evidence that BCL10 phosphorylation at S138 might have a negative effect on NF-κB activation [24]. Considering the above data, the mechanism by which CaMKII connects calcium-CaM signals to NF-κB remains largely unclear. Future studies employing genetic deletion or RNAi silencing of CaMKII in primary T cells will be required to better clarify the role of CaMKII in TCR activation of NF-κB.

Interestingly, CARMA1 phosphorylation is necessary but not sufficient for CBM complex formation. An additional essential step involves interaction with the integrin receptor regulatory protein ADAP (Figure 1). ADAP-deficient T cells show impaired TCR- and CD28-dependent proliferation and reduced cytokine secretion. ADAP binds both CARMA1, facilitating CBM complex assembly [25], and TAK1, promoting IKK activation [26], and these actions might account for ADAP’s effect on NF-κB activation and cytokine secretion. The molecular mechanism by which ADAP enhances CBM complex assembly remains unclear. The ADAP-CARMA1 interaction is also critical for CDK2 and Cyclin E expression [27]. As cyclins regulate progression through cell cycle, this observation suggests a molecular basis for the impaired T cell proliferation observed in ADAP deficiency.

BCL10

As a result of the above modifications and interactions of CARMA1, the constitutively associated BCL10-MALT1 complex associates with CARMA1, forming the CBM complex [3, 4]. Immediately following TCR stimulation, BCL10 is phosphorylated and ubiquitinated. BCL10 post-translational modification is complex, with many reported sites of modification. The regulation and significance of many modifications remain poorly understood (see [1, 5] for review). To date, evidence suggests that IKKβ is one of the principal kinases responsible for BCL10 phosphorylation (Figure 2) (see above for discussion of BCL10 phosphorylation by CaMKII). Initially, IKKβ-phosphorylation of BCL10 stabilizes the CBM complex, but subsequent IKKβ phosphorylation of BCL10 triggers BCL10 dissociation from MALT1 [28] and/or BCL10 degradation [29]. Thus BCL10 phosphorylation may be an important step for terminating TCR signals to NF-κB. Interestingly, calcineurin, a calcium dependent phosphatase, has been reported to dephosphorylate BCL10 [30, 31], stabilizing the CBM complex and enhancing NF-κB activation (Figure 1). Calcineurin, primarily known for its role in TCR-induced activation of the NFAT transcription factor, is also involved in lysosome-dependent degradation of PLCγ1 and PKCθ upon T cell re-stimulation, resulting in T cell anergy [32]. As PKCθ is essential for NF-κB activation, these seemingly conflicting roles of calcineurin in TCR activation of NF-κB require further examination.

An area of intense recent investigation involves elucidation of the mechanism and purpose of BCL10 ubiquitination. Reports indicate that BCL10 is modified by K63-polyubiquitination [33, 34] or both K48- and K63-polyubiquitination with K63- preceding K48-polyubiquitination [35]. Mutagenesis data suggest that both K48- and K63-polyubiquitin chains are conjugated exclusively to lysines 31 and 63 of Bcl10 [35]. K63-polyubiquitination of Bcl10 is triggered by interaction with CARMA1 [12, 13]. The identity of the BCL10 ubiquitin ligase is highly controversial, with NEDD4 and Itch [36], cIAP2 [37], βTrCP [29], and TRAF6 [35] all implicated as E3 ligases targeting BCL10. These uncertainties aside, the emerging consensus is that BCL10 ubiquitination serves two important purposes; initial transmission of the NF-κB activation signal [33, 35], followed by BCL10 degradation and signal termination [29, 33, 35, 36] (see Negative regulation of TCR signaling to NF-κB, below). BCL10 ubiquitination is important for signal transmission, as ubiquitin chains create binding sites for IKKγ [35] and p62 [33] (Figure 2), two molecules critical for NF-κB activation. Interestingly, the requirement for p62 and certain other signaling molecules seems to depend on the stage of T cell differentiation (see Box 1).

Box 1. Differences in TCR-to-NF-κB signaling between naïve and effector/memory T cells.

Compared to naïve T cells, effector and memory cells demonstrate a more rapid response to antigen stimulation. This accelerated response of effector/memory T cells may partly reflect differentiation-dependent changes in NF-κB signaling mechanisms. One proposed mechanism is that crucial signaling proteins in differentiated T cells are constitutively phosphorylated, allowing more rapid activation. However, investigations have failed to reveal differences in phosphorylation states of ZAP70 and PLC-γ between naïve and memory cells [111]. A second possible mechanism is that signaling proteins in effector/memory T cells may be constitutively associated with lipid rafts, cholesterol rich regions in the cell membrane that are sites of signaling molecule aggregation. Studies have indeed demonstrated increased interaction of signaling proteins with lipid rafts in memory CD4 and CD8 T cells [112, 113], but the functional consequences have not been established.

A third potential mechanism is that key signaling proteins may be more highly expressed in effector/memory T cells. Single cell experimental studies and computer modeling suggest a correlation between increased signaling protein expression and T cell functional capacity [114]. Multiple studies have shown higher total ZAP70 protein expression in CD4 effector and memory T cells, compared to naïve T cells [115, 116]. Furthermore, data show that IFN-γ production occurs selectively from effector T cells expressing the highest levels of ZAP70, and that IFN-γ expression can be reduced by silencing ZAP70 using siRNA. Another study demonstrated higher levels of the adaptor protein BCL10 in both effector CD4 and memory CD8 T cells, compared to naïve T cells [33], although the functional importance of this difference remains unexplored. Importantly, elevations of signaling protein levels do not universally accompany T cell differentiation. For example, relative to naïve T cells, SLP-76 levels are increased in effector T cells but decreased in memory T cells [117].

A final possible mechanism is that differentiated T cells might utilize signaling molecules that are distinct from those involved in naïve cell signaling. Two studies demonstrated that the scaffolding protein p62 is required for TCR-to-NF-κB signaling in effector T cells, but not in naïve T cells [33, 118]. Consistent with these findings, p62 expression is dramatically increased concomitant with TCR stimulation and effector differentiation [119]. These studies indicate that differentiation-dependent increases in expression of certain intermediate signaling proteins and utilization of distinct signaling proteins may enhance TCR signaling in effector/memory T cells, vs. naïve T cells.

Apart from changes in signal transduction, alterations in transcription factor utilization and chromatin modification contribute to the rapid responses of memory T cells to antigen stimulation. The increase in IFN-γ production by memory CD4 T cells compared to naïve cells is associated with a switch from T-bet- to NF-κB-dependent transcription. Additionally, memory cells exhibit increased expression and promoter binding of the NF-κB subunit p50 [120]. Also, the kinetics of gene transcription may be accelerated in memory T cells by histone acetylation of the promoter regions of specific genes, including perforin and Eomes, facilitating transcription factor binding [121]. Thus, epigenetic modifications and differential transcription factor usage may contribute to the rapid transcriptional response of memory T cells.

Another emerging area of interest regarding the biology of BCL10 involves the relationship between TCR-dependent post-translational modification of Bcl10 and regulation of the T cell actin cytoskeletion [38]. Although current evidence suggests that actin cytoskeletal dynamics do not directly influence NF-κB activation, actin polymerization has a profound influence on the immunological synapse and TCR microclusters (see Cell membrane and cytosolic complexes in TCR signaling to NF-κB, below), indirectly linking regulation of the actin cytoskeleton to TCR signaling to downstream transcription factors [39, 40]. Also, there are multiple points of intersection between TCR-to-NF-κB signaling and regulation of the actin cytoskeleton, suggesting an as-yet poorly understood mechanistic coupling of the regulation of these pathways. TCR stimulation alters the T cell actin cytoskeleton, enabling T cell spreading and conjugation with APCs. In Jurkat cells, successful actin remodeling post-TCR stimulation requires BCL10 S138 phosphorylation [38] and IKKβ-Homer3 association at the IS [41] (Figure 1). Interestingly, BCL10 S138 phosphorylation is also required for macrophage Fc receptor-induced actin polymerization, leading to phagosome formation [42]. Phosphorylation of S138 is unrelated to CARMA1-BCL10-MALT1 mediated NF-κB activation in both T cells and macrophages [38, 42]. Similarly, the IKKβ-Homer3 mediated actin regulation in T cells is also NF-κB independent [41]. The identity of the precise mechanisms that link phospho-BCL10 and Homer3 to actin remains to be determined.

MALT1

MALT1, the third member of the CBM complex, plays a more nebulous role in TCR activation of NF-κB. Initial studies suggested MALT1 functions as a scaffolding protein that participates in NF-κB activation by connecting the CBM complex with TRAF6. Later studies showed that MALT1 contributes less to TCR activation of NF-κB than BCL10 or PKCθ in primary mouse T cells [43] and that silencing MALT1 expression in Jurkat reduces but does not completely block IκBα phosphorylation [44]. Additionally, in B cells, BCL10 is essential for RELA activation, whereas MALT1 is dispensable [45]. Thus, there is an emerging understanding of MALT1 as a protein which modifies antigen receptor-mediated signaling to NF-κB, without being strictly required for NF-κB activation.

Moreover, the protease activity of MALT1 is associated with regulation of diverse cellular functions, most of which are distinct from NF-κB signal transduction. MALT1 cleaves target proteins after arginine residues preceded by a serine [44, 46]. To date, four MALT1 substrates, BCL10, A20, CYLD and RELB, have been identified (Figure 1). MALT1 cleavage of BCL10 and CYLD regulates T cell adhesion [44] and JNK activation [47] respectively, with no detected effect on NF-κB signal transduction. Another target of MALT1 cleavage is A20, a known inhibitor of NF-κB activation [46]. However, MALT1 cleaves only a small fraction of the total cellular A20 pool [46], and blockade of MALT1 protease activity fails to limit IKK activation [48]. Thus, the significance of A20 cleavage as a mechanism regulating TCR signaling to NF-κB remains uncertain. Among known MALT1 substrates, RELB is most compelling as an NF-κB signal mediator that is cleaved to regulate NF-κB activation. MALT1-dependent RELB cleavage and concomitant degradation leads to increased RELA and c-REL DNA binding [49, 50]. How RELB degradation affects RELA or c-REL activity is not well understood, but it might reflect competition between different NF-κB heterodimers for DNA binding sites, or inhibition of canonical activation by gene products of the non-canonical (i.e., RELB-dependent) pathway.

Caspase-8

The protease caspase-8 is well-known as a mediator of apoptosis. Impaired lymphocyte activation in caspase-8-deficient mice also suggested a role of caspase-8 in the TCR-to-NF-κB pathway [51], a finding now confirmed and extended by several additional studies (Figure 1). Caspase-8 recruits IKK to activated Bcl10-MALT1 by a TRAF6-dependent mechanism [51, 52], and MALT1 activates caspase-8, causing caspase-8-mediated cleavage of the NF-κB inducer c-FLIP_L [53] (reviewed in [54]). Additionally, caspase-8 proteolytic activity is required for CD3/CD28-dependent NF-κB activation, as T cells treated with the caspase inhibitor Z-VAD-FMK or expressing the caspase-8 catalytically inactive mutant, C360S, fail to induce NF-κB [51]. Interestingly, in HEK293 cells, neither Z-VAD-FMK nor caspase-8 C360S blocks caspase-8-induced NF-κB activation [55], suggesting that the mechanistic link between caspase-8 and NF-κB activation might be cell-type specific. Thus, additional mechanistic data are required to yield a clear understanding of how caspase-8 participates in this signaling pathway.

IKK complex

A key step in TCR activation of NF-κB is CBM complex-dependent K63-polyubiquitination of IKKγ. TRAF6 has been proposed as the ubiquitin ligase that performs this key function [1, 3, 54] (Figure 1). However, TRAF6-deficient T cells do not show impaired NF-κB activation [56], suggesting either that TRAF6 does not play such a role, or that there are redundant ubiquitin ligases that compensate for loss of TRAF6. For example, TRAF2, an ubiquitin ligase recruited by the caspase-8-FLIP complex is also capable of polyubiquitinating IKKγ [53]. Another possible candidate is the ubiquitin ligase MIB2, which binds to BCL10 and promotes IKKγ ubiquitination and NF-κB activation in transfection-overexpression experiments [57] (Figure 1). Thus, it is possible that several ubiquitin ligases contribute to K63-polyubiquitination of IKKγ. It is also possible that there is a key role for modification of IKKγ by M1 (linear head-to-tail linked) polyubiquitin chains. M1 polyubiquitination plays a key role in TNF receptor activation of NF-κB [58], but involvement in TCR-mediated activation of IKK is undetermined. Resolution of these lingering mechanistic questions will require compelling genetic and biochemical evidence using primary T cells to definitively demonstrate which ligase(s) and which ubiquitin modifications are essential.

Activation of IKK requires a combination of IKKγ ubquitination and IKKβ phosphorylation. The latter process is mediated by the protein kinase TAK1 [59] (Figure 1). TAK1 activation seems to be dependent on PKCθ, but not on the CBM complex members CARMA1 and BCL10 [60]. The adapter molecule ADAP is required for TAK1 recruitment to PKCθ [26]. However, the precise mechanism of PKCθ-mediated TAK1 activation is not well-defined. The IKKβ phosphorylation events can be countered by the PP4R1-PP4c phosphatase complex, limiting IKK activation and T cell function [61] (Figure 2).

The ‘All-or-None’ NF-κB response

At a single cell level, signaling pathways may be broadly classified as digital or analog. Whereas analog signaling is a graded response that is proportional to stimulus intensity, digital signaling produces an all-or-none response, irrespective of intensity of the activation stimulus [62]. In T cells, increasing TCR stimulation strength leads to an increasing percentage of cells with IκBα phosphorylation and RELA activation, without altering per cell level of the two proteins [63]. This observation suggests a digital signaling mechanism similar to findings regarding TNFα-induced NF-κB signaling [64], and TCR induced ERK phosphorylation [65, 66] or NFATc2 activation [67]. In contrast, bypassing the TCR via stimulation with PMA+ionomycin leads to a graded (analog) NF-κB response in T cells [63, 67], indicating that signal digitization occurs at an early TCR-proximal step. The molecular switch enabling digitization of TCR signals remains to be determined.

The above studies demonstrate that, rather than the early understanding of TCR activation of NF-κB as a simple linear cascade (PKCθ→CBM complex→IKK), there is a complex web of interconnected signaling molecules that link the TCR to NF-κB. In particular, there is accumulating evidence of multiple signaling inputs converging on key proteins, particularly CARMA1 and BCL10. Additionally, the mechanistic understanding of the crucial functions of certain mediators in transmitting the activating signal to NF-κB, e.g., MALT1, is incomplete. Also, the precise roles and relative importance of kinases recently implicated as regulators of this pathway, including PDK1, GLK, HPK-1, and CK1α remain to be well-defined. Cross-talk between the TCR-to-NF-κB pathway and other cellular processes, such as regulation of the actin cytoskeleton and T cell-APC interactions, is an emerging area of interest for which there is currently very limited mechanistic understanding.

Cell surface and cytosolic complexes in TCR signaling to NF-κB

Cell surface TCR microclusters and the immunological synapse

The concept of the immunological synapse developed in the late nineties when imaging revealed micrometer sized clusters, composed of surface receptors and signaling proteins, at the T cell-APC intercellular contact site [68, 69]. Initially, the IS was proposed as the site at which intracellular signaling from the TCR is initiated and sustained. Later studies showed that initiation of TCR signaling and tyrosine kinase activation can be detected within seconds following stimulation and before IS formation, in peripheral TCR-rich regions termed “microclusters” [70, 71]. Current data suggest that the earliest signaling events occur in these peripheral microclusters, which move centripetally and eventually fuse to form the mature IS.

Recent data also cast doubt on the model that the IS represents the primary site of sustained signaling to downstream mediators. Specifically, TCR signaling to NF-κB can occur in the absence of an IS [72, 73]. Also, as detailed below, accumulating data point towards the existence of cytosolic signaling-clusters/signalosomes that may sustain and regulate TCR signal transduction. Although the IS might not be involved in signal initiation or maintenance, the IS has certain other critical functions. Experimental studies and computer modeling revealed that at low antigen-MHC concentrations, the IS amplifies TCR activation by grouping together antigens, TCR and signaling proteins. This allows the TCR to rapidly engage the small number of stimulatory ligands, resulting in efficient signal transduction [74]. Also, the IS may serve as a site of TCR internalization and degradation, serving to limit signal strength and/or duration [75]. Interestingly, ubiquitination of as-yet undefined c-SMAC proteins leads to interaction with the ubiquitin-binding protein, TSG101, a component of the endosomal sorting complex required for transport (ESCRT). This interaction is required for c-SMAC formation and organization, as well as for termination of signaling by microclusters and TCR degradation [76]. These observations suggest that TCR signaling and signal termination by TCR degradation are intimately connected.

Cytosolic T cell signalosomes

Emerging data suggest that several distinct cellular sites may together coordinate TCR signaling to NF-κB. Immediately following TCR engagment, the integral membrane protein LAT and the cytosolic protein SLP76 are recruited to the region immediately below the TCR microclusters [3]. ZAP70-mediated phosphorylation of LAT and SLP76 enables these adaptor proteins to bind downstream mediators, transmitting the activation signal. Data suggest that following initial recruitment, LAT and SLP76 dissociate from TCR microclusters in endosomes [77] (Figure 2). SLP76 remians on the the outer cytosolic side of the endosome, and its attachment to endosomes is mediated indirectly through LAT. Endosome-associated LAT and SLP76 remain phosphorylated, indicating that these adaptors are capable of binding effector molecules and continuing the signaling cascade. LAT-SLP76 endosomes might represent sites of continous signaling to downstream mediators, similar to the signaling complexes on intracellular endosomal membranes observed downstream of the EGFR [78, 79]. LAT-SLP76 endosomes might also have a role in termination of TCR signaling, as inhibition of microcluster movement results in enhanced TCR signaling [80, 81].

There is also evidence for cytosolic signalosomes containing BCL10 and MALT1 that specifically sustain NF-κB-activation. Upon PKCθ activation, CARMA1 is recruited to the plasma membrane, apparently undergoing a conformational change that opens access to binding sites for BCL10 and MALT1, leading to activation of IKK [3, 4] (Figure 2). But the mechanism that links the membrane bound CBM complex to IKK activation is not well defined. One possibility is that BCL10-MALT1 clusters interact transiently with membrane bound CARMA1, followed by redistribution to a cytosolic site at which these proteins transmit the signal to IκBα-NF-κB. Biochemical analysis of Jurkat cells revealed the existence of two complexes [34], an early membrane-bound CBM complex (Figure 2) and a late BCL10-MALT-IKK complex, and this late complex was shown to inducibly interact with IκBα. Additionally, in a cell free system, purified recombinant BCL10 and MALT1 proteins along with TRAF6, UBC13, and TAK1 were sufficient for IKK activation, demonstrating that the BCL10-MALT1 complex can activate IKK in a manner that is independent of concurrent physical association with CARMA1 [48] (i.e., although CARMA1 is required for transmitting the activating signal to BCL10, CARMA1 does not need to remain associated with BCL10-MALT1). Importantly, IKK activation was mediated by a small fraction of the total BCL10 and MALT1 pool that existed as high molecular weight oligomers, providing biochemical evidence of BCL10-MALT1 signalosomes. However, it is also important to note that there are as yet no published biochemical data showing that similar signaling-competent oligomers of BCL10-MALT1 exist in intact cells or cellular lysates.

Imaging studies in primary mouse T cells and D10 T cells demonstrate the presence of TCR-induced cytosolic BCL10-MALT1 clusters, termed POLKADOTS [9, 82], at locations that are distinct from membrane bound PKCθ, a marker for the IS. Furthermore, these BCL10-MALT1 clusters are also enriched in the downstream signaling protein TRAF6 [82]. Formation of the POLKADOTS clusters is dependent both on expression of the autophagy adaptor protein p62 and K63-polyubiquitination of Bcl10 [33]. Interaction between Bcl10-Malt1 and p62 and the resultant signal transmission to IKK is driven by BCL10 K63-polyubiquitination (Figure 2). The formation of the proposed “POLKADOTS signalosome” is strongly correlated with IKKα/β phosphorylation [33] and RelA nuclear translocation [63], supporting a direct role in NF-κB activation.

The above studies provide compelling evidence that TCR signaling to NF-κB involves several organized signalosomes that concentrate groups of interacting signaling proteins. Interestingly, these signalosomes are not universally required for TCR signaling to NF-κB. Specifically, NF-κB activation can occur in the absence of the IS, and the POLKADOTS signalosome apparently plays no role in naïve T cells (see Box 1). Although early data suggest that such signalosomes represent cellular platforms for coordinating and precisely regulating positive and negative signals to NF-κB, much further work is needed to fully define the components and functions of these intriguing macromolecular structures. Areas for future focus include: determining how TSG101 and the ESCRT complex interact with microclusters and c-SMAC constitutents to regulate signal transmission by the TCR; better establishing the role of endosomal LAT-SLP76 in activation of downstream transcription factors; establishing the mechanistic connection between the IS and POLKADOTS signalosome; and defining in molecular detail how mechanistic requirements for NF-κB activation change with T cell differentiation (see Box 1).

Negative regulation of TCR signaling to NF-κB

TCR activation of NF-κB is critical for T cell proliferation and differentiation. However, unrestricted and persistent NF-κB activation can lead to development of autoimmune diseases and neoplasms [83], cellular senescence [84], or apoptosis [85]. In order to strike a balance between productive T cell activation and deleterious consequences of excessive NF-κB activation, TCR signaling has to be precisely regulated. After T cell activation, negative regulation starts at the level of cell surface receptors (e.g., by TCR endocytosis and degradation) and continues at multiple steps of cytoplasmic and nuclear signaling. In this review, we focus on cytoplasmic mechanisms that limit TCR-to-NF-κB signaling. (Cell surface receptor and nuclear regulation mechanisms are reviewed in [2, 3] and [86], respectively). Cytosolic negative regulators and mechanisms of action are listed in Table 1 and illustrated in Figure 2.

TABLE 1.

Cytosolic proteins involved in negative regulation of TCR-to-NF-κB signaling

Protein mediator	Proposed mechanism	Gene deletion or silencing phenotype	Ref
SHP-1 (phosphatase)	Dephosphorylates LCK, ZAP70, SLP-76	me/me “moth-eaten” mouse TCR-induced thymocyte hyper-proliferation, increased IL-2 production Increased CD8 effector cell expansion with viral infection	[87–90, 110]
STS-1, STS-2 (phosphatase)	Dephosphorylates ZAP-70	Sts1−/− Sts2−/− double knockout T cell hyperproliferation Increased IL-2, -4, -5, -10, IFN-γ secretion Increased susceptibility in murine multiple scelerosis model	[91]
CSK (protein kinase)	LCK inhibitory phosphorylation	siRNA silencing: Increased TCR-induced IL-2 secretion	[94]
CBL-B, C-CBL (E3 ubiquitin ligase)	C-CBL: LAT trafficking to endosomes	Cbl-b−/−: increased lymphocyte proliferation and secretion of IL-2 and antibody c-Cbl−/−: lymphoid hyperplasia	[77, 107, 108]
HPK-1 (protein kinase)	SLP-76 inhibitory phosphorylation increases 14-3-3 binding	Map4k1−/−: T cell hyperproliferation, increased T cell cytokines and B cell antibody production Increased susceptibility to autoimmune encephalitis	[19, 20]
PP2A (phosphatase)	Dephosphorylates CARMA1 and IKK	siRNA silencing: Increased IL-2 and IFN-γ secretion TAX mediated PP2A inhibition: Constitutive IKK activation	[97, 98]
CK1α (protein kinase) *	Phosphorylates CARMA1	CK1α inactive kinase mutant: Increased NF-κB activity CK1α silencing: reduced NF-κB activity, T cell IL-2 secretion and proliferation (contradictory results suggest CK1α is a bi- functional regulator. Dominant effect is positive regulation)	[21]
PKCθ (protein kinase) * (direct or indirect effect)	Late CARMA1 phosphorylation at S649	Mutation of S649 phosphorylation site increases IKK and NF-κB activation in Jurkat cells (PKCθ knockout blocks NF-κB activation. Dominant effect is positive regulation)	[16]
GAKIN (kinesin 13B)	Competes with BCL10 for CARMA1 binding	shRNA silencing: Increased IKK activation and IL-2 secretion	[99]
CRADD (adaptor protein)	Competes with CARMA1 to bind BCL10	Cradd−/−: Increased T cell CARMA1-BCL10 binding, RELA activation and cytokine secretion	[100]
autophagy pathway, proteasomes	BCL10 degradation	Autophagy-deficient T cells: Increased IL-2 secretion, CD25 expression; decreased survival Inhibition of BCL10 phosphorylation: Increased secretion of IL-2 and TNFα	[28, 29, 33, 36, 109]
A20 (deubiquitinase & ubiquitin ligase)	Removal of K63- polyubiquitin from MALT1	Tnfaip3−/−: Multi-organ inflammation, leading to death. B cell conditional knock out: Lupus-like disease, increased antibody secretion	[101, 102]
CYLD (deubiquitinase)	Removal of K63- polyubiquitin from TAK1	Cyld−/−: Increased IKKβ phosphorylation. Hyperresponsive T and B cells. Spontaneous colonic infiltration	[104, 105]
PP4R1-PP4c (phosphatase)	IKKα/β dephosphorylation	PP4R1-silenced T cells demonstrate enhanced IKK phosphorylation and NF-κB activation	[61]

^*Bi-functional NF-κB regulator with stimulatory effect overriding negative function

TCR-proximal regulatory mechanisms

Successful transmission of the activation signal from the TCR to NF-κB requires dynamic regulation at multiple intermediate steps. TCR engagement results in the phosphorylation and/or ubiquitination of many intermediates, and these post-translational modifications activate or repress signal transmission. The TCR-proximal signaling events involving ZAP70, LCK, LAT and SLP-76 are dynamically regulated by multiple protein phosphatases and kinases (Figure 2). Data suggest that SHP-1, a phosphatase well known for its ability to negatively regulate T cell signaling, dephosphorylates LCK [87], ZAP70 [88] and SLP-76 [89]. Mice with T cell-specific deletion of SHP-1 exhibit increased expansion of CD8 effector (but not memory) T cells in response to viral infection [90]. In addition to SHP-1, ZAP70 can also be de-phosphorylated by Sts-1 and Sts-2, limiting TCR stimulation. Deletion of both Sts-1 and Sts-2 causes T cell hyper-responsiveness, with augmented cytokine secretion in response to antigen stimulation [91]. In stimulated T cells, Sts-1 and -2 deletion results in accumulation of hyper-phosphorylated and ubiquitinated proteins [92]. As deletion of either Sts-1 or Sts-2, alone, does not produce any detectable phenotype, there is likely wide overlap in the function of these related molecules [93].

LCK activity is controlled by two opposing enzymes: CSK, a protein kinase, and CD45, a protein phosphatase. Data show that CSK-mediated inhibitory phosphorylation of LCK diminishes TCR activation (Figure 2), whereas CSK silencing amplifies TCR-induced IL-2 production [94]. CD45 removes this inhibitory phosphate group from LCK, allowing LCK to participate in TCR signaling [95]. SLP-76 activity is also limited by phosphorylation. After the initial transient ZAP70-mediated activating phosphorylation of SLP-76 (at Y112, Y128 and Y145; reviewed in [96]), HPK-1 adds a phosphate group to SLP-76 at S376 (Figure 2), causing increased binding to the inhibitory protein 14-3-3 [19, 20]. T cells lacking HPK-1 demonstrate CD3 induced hyper-proliferation and increased cytokine secretion, although it is unclear whether this effect is mediated by the effect of HPK-1 on NF-κB or on other T cell transcription factors (e.g., NFAT and AP-1).

TCR-distal regulatory mechanisms

In the case of the CBM complex and other TCR-distal mediators of NF-κB activation, these proteins were identified relatively recently, and regulatory mechanisms are therefore only now being recognized. One protein newly identified as a CARMA1 regulator is the serine-threonine phosphatase, PP2A (Figure 2). As discussed above, PKCθ phosphorylation of CARMA1 at Ser645 is critical for assembly of the CBM complex [1]. PP2A removes the phosphate group from Ser645, destabilizing the CARMA1-BCL10 interaction and reducing NF-κB activation. Further, siRNA silencing of PP2A results in higher TCR-induced IL-2 and IFN-γ production [97]. However, it is also important to note that PP2A directly de-activates the IKK complex [98], making it unclear whether inhibition of NF-κB activation is primarily due to PP2A effects on CARMA1 or IKK.

Also discussed above, CARMA1 phosphorylation at S649 and S608 by PKCθ (directly or indirectly) [16] and CK1α [21], respectively, impairs NF-κB signaling. A phosphorylation-independent mechanism of CARMA1 regulation involves binding of CARMA1 to the kinesin motor protein GAKIN [99]. This interaction results in CARMA1 movement away from PKCθ at the IS and reduces CARMA1 interaction with BCL10 (Figure 2). Post-TCR stimulation CARMA1-BCL10 binding is also interrupted by CRADD, which competes with CARMA1 to bind BCL10 (Figure 2). Accordingly, CRADD-deficient murine T cells show increased CARMA1-BCL10 binding and RELA activation. CRADD deletion results in increased T cell cytokine production with concanavalin A stimulation, but does not yield any identified autoimmune phenotype [100].

Regulation of TCR signaling by ubiquitin modifying enzymes

The ubiquitin editing enzyme, A20, has been long known as an inhibitor of TNF-dependent NF-κB activation. A20 can both remove K63-polyubiquitin chains from and add K48-polyubiquitin chains to target proteins [101]. A20 has a high basal expression in T cells, and it can limit TCR-induced NF-κB activity by removing K63-polyubiquitin chains from the CBM complex protein, MALT1 [102] (Figure 2). Interestingly, MALT1 can cleave A20 in a TCR-dependent manner, and this proteolytic cleavage has been suggested to disrupt the ability of A20 to limit TCR activation of NF-κB [46]. A20 is also capable of cleaving K63-polyubiquitin chains from RIP1, TRAF6, and IKKγ, limiting TNF-, IL-1-, and LPS-dependent NF-κB activation [103]. However, there is so far no evidence that these specific mechanisms regulate the TCR-to-NF-κB pathway. As A20 activity is critical for down-regulating NF-κB activity in numerous cell types, A20 deficient mice (Tnfaip3^−/−) die shortly after birth due to multi-organ inflammation [101]. Although conditional inactivation of A20 in B cells results in autoantibody production and lupus-like disease [101], the in vivo role of A20 in T cells remains uncertain.

A second ubiquitin modifying enzyme, CYLD, has been reported as a modulator of JNK and NF-κB activation [101]. CYLD^−/− mice develop spontaneous autoimmune colitis. Biochemical studies of CYLD^−/− peripheral T cells and B cells showed constitutively phosphorylated IKKβ and activated NF-κB [104, 105]. This is explained by loss of CYLD mediated TAK1 K63-deubiquitination [106], which suppresses TAK1 activation in wild-type lymphocytes (Figure 2). Curiously, in CD3+CD28 stimulated Jurkat T cells, MALT1 cleaves CYLD increasing activation of JNK, but not NF-κB [47]. Further studies of A20^−/− and CYLD^−/− T cells will be required to determine the relative importance of these regulators of protein ubiquitination in TCR-to-NF-κB signaling, and to assess whether there is any redundancy in their activities.

Regulation of TCR signaling by molecular sequestration

Emerging data suggest a novel mechanism of TCR signal transduction regulation, whereby cytosolic mediators are selectively sequestered in cytosolic vesicular compartments, making them unavailable for signal transmission. The ubiquitin ligases of CBL family, CBL-B and C-CBL, negatively regulate T cell activation. Both CBL-B and C-CBL knock out results in higher NF-κB activation, lymphoid hyperplasia, and increased IL-2 secretion from T cells [107]. Reports indicate that CBL activity might depend on promotion of TCR trafficking to lysosomes [107] and/or on ubiquitination of the adaptor protein LAT, targeting LAT for endosomal sequestration [108] (Figure 2). A similar mechanism involving the distal TCR signaling adaptor BCL10 was also recently reported. Data suggest that BCL10 is selectively targeted for sequestration within autophagosomes [33], followed by lysosomal degradation [33, 36]. The proteasome also contributes to TCR-dependent BCL10 degradation via an unclear mechanism [29, 33]. The role of autophagy in limiting TCR activation of NF-κB is further supported by the observation that T cell-specific inactivation of the autophagy genes ATG7 [109] or ATG3 [33] resulted in increased IL-2 secretion and CD25 expression. BCL10 autophagy is highly selective, as the BCL10 binding partner MALT1 was neither detected inside autophagosomes nor degraded [33]. The process whereby MALT1 is separated from BCL10, protecting MALT1 from degradation might depend on autophagosome formation [33] and/or IKKβ-mediated BCL10 phosphorylation [28, 29]. Data showing that phosphorylation of Bcl10 by IKKβ causes BCL10 to dissociate from the CBM complex [28] suggests that IKKβ has a dual function, both positively and negatively regulating the TCR-to-NF-κB pathway.

Interestingly, the process of BCL10 autophagy occurs at the POLKADOTS signalosome, requiring the same molecular interactions that play a central role in TCR activation of NF-κB. Specifically, autophagy of BCL10 requires interaction between p62 and K63-polyubiquitinated BCL10 [33, 35, 36] (Figure 2). Thus, p62 is an additional bi-functional regulator of TCR signaling to NF-κB, controlling both assembly of the POLKADOTS signalosome and degradation of its key signaling component, BCL10. How NF-κB activation and BCL10 autophagy (a signal limiting mechanism) are coordinately regulated at the POLKADOTS signalosome will require further investigation.

The above studies illustrate that negative regulatory mechanisms act at multiple points in the TCR-to-NF-κB signaling cascade. Interestingly, diverse independent mechanisms including phosphorylation, de-phosphorylation, ubiquitination, de-ubiquitination, protein trafficking, and protein degradation contribute to limiting activation of TCR signaling. These data suggest that negative regulation of this pathway is critical for T cell function and/or survival, in line with emerging data that unrestrained NF-κB activation has a spectrum of possible deleterious outcomes.

Concluding remarks

Recent findings have significantly altered our views of TCR signaling to NF-κB. Certain mediators, such as PKCθ and CARMA1, are phosphorylated by multiple kinases, dispelling early models that postulated a simple linear cascade by which the TCR transmits activating signals to NF-κB. Studies have also revealed that cytosolic mediators in this pathway do not all coalesce in a single TCR-proximal signalosome at the IS. Rather, cytosolic signalosomes far removed from the TCR are also integral to NF-κB activation. Despite these new complexities, precise molecular mechanisms of signal transmission are gradually being revealed, with a number of essential post-translational modifications and their downstream effects now identified. For example, Bcl10, once thought to be a simple adaptor connecting CARMA1 to MALT1, is now known to undergo K63-polyubiquitination, triggering association with p62 and activation of IKK. Additionally, numerous mechanisms of negative regulation have been recognized, indicating that the TCR-to-NF-κB pathway is tightly modulated. Four kinases (PKCθ, HPK-1, CK1α, and IKKβ) and two ubiquitin-binding proteins (TSG101 and p62) play a dual role in signaling to NF-κB, with both activating and signal limiting activities. Precise control of NF-κB activation may guarantee production of exact levels of cell cycle proteins, thereby avoiding failed cell division and apoptosis. Emerging data also suggest that signaling to NF-κB is coupled to seemingly unrelated biological processes such as regulation of the actin cytoskeleton. Additionally, there may be considerable differences in mechanisms of TCR activation of NF-κB utilized by T cells at distinct stages of differentiation. Despite these recent advances in our understanding of this pathway, there remain many mechanistic details that are poorly understood. Important questions for the field are outlined in Box 2.

Box 2. Directions for future research.

• Elucidation of differences between naïve and effector T cells in the TCR-to-NF-κB pathway. In particular, it is unclear why effector/memory T cells require a cytosolic signalosome to successfully activate NF-κB (see Box 1), and why levels of signaling proteins differ between naïve and effector/memory cells.

• Defining how the CBM complex and the POLKADOTS signalosome are mechanistically connected to IKK activation. Related questions include how BCL10 is physically separated from MALT1, allowing selective BCL10 degradation, and whether MALT1 function changes after its signaling partner BCL10 is degraded.

• Revealing the mechanism and purpose of coupling NF-κB activation with other T cell signaling cascades. In particular, it is currently unclear why JNK activation and actin polymerization are closely networked to the TCR-to-NF-κB cascade.

• Exploring the role of microRNAs in TCR-to-NF-κB signaling. MicroRNAs are emerging as key regulators of T cell activation, with likely effects on signaling to NF-κB. Indeed, recent data show mir-146 and 155 affect mediators utilized by the TCR-to-NF-κB cascade and alter T cell behavior [122–125].

• Determining the role of atypical ubiquitination in TCR-to-NF-κB signaling. To date, most research in this pathway has focused on ‘typical’ K48- and K63-polyubiquitination, due to the existence of highly-specific tools that enable detailed study of these modifications. The six ‘atypical’ ubiquitins have demonstrated function in NF-κB activation [58] but there is so far little data regarding how these under-studied ubiquitin polymers modulate TCR signaling to NF-κB.

Highlights.

• TCR activation of NF-κB involves a complex web of interconnected signal transducers

• Many negative regulatory mechanisms cooperatively modulate the TCR-to-NF-κB pathway

• Both a plasma membrane and a cytosolic signalosome participate in TCR signaling to NF-κB

Box-Glossary

T cell receptor (TCR) complex

Heteromeric complex comprised of the T cell receptor α-β heterodimer and CD3 complex. TCR recognizes peptide antigens presented by major histocompatibility complex (MHC) proteins, while the CD3 chains are the signal transducing subunits

CD28

A cell surface receptor providing a “co-stimulatory” signal required in conjunction with TCR signaling for productive activation of naïve T cells

ITAM

Immunotyrosine activation motif. Conserved amino acid sequence containing tyrosine residues located in cytosolic regions of certain cell surface receptors, including CD3 chains of T cells. Phosphorylated tyrosines serve as binding site for downstream signaling proteins

IKK

IκB kinase. A trimeric protein complex composed of two catalytic subunits, IKKα and IKKβ, and a homodimer of the regulatory subunit, IKKγ/NEMO. The canonical NF-κB activation pathway converges on IKK activation, resulting in phosphorylation and subsequent degradation of IκB, the protein responsible for NF-κB cytosolic sequestration

NF-κB

Nuclear Factor kappa B. A five member family of transcription factors that exist in homo- or hetero-dimers. Members are RELA, RELB, cREL, p100/p52 and p105/p50. “NF-κB” is generally used to refer to the RELA-p50 activating heterodimer. In un-stimulated T cells, NF-κB is sequestered in the cytosol by its binding partner IκB. IκB is phosphorylated by IKK, triggering proteasomal degradation, freeing NF-κB to translocate to the nucleus and drive transcription of target genes

K48- vs. K63-polyubiquitination

Distinct polyubiquitin chain conjugations in which polymeric linkages are via lysine 48 or lysine 63, respectively. K48-polyubiquitinated proteins are generally degraded by the proteasome. The consequences of K63-polyubiquitination are more diverse, ranging from modification of protein function to targeting proteins for degradation by the autophagy-lysosomal pathway

Microclusters

Nanometer-sized clusters of ligand-engaged TCR that initiate signal transduction. The cytoplasmic face of microclusters is enriched in specific signaling molecules and phosphotyrosine

IS (immunological synapse)/SMAC (supramolecular activation cluster)

Microclusters fuse to form the micrometer sized IS between the APC and T cell. The IS is composed of concentric rings of signaling molecules: in general, the central SMAC (cSMAC) contains the TCR and signaling partners, and the peripheral SMAC (pSMAC) contains molecules primarily involved in T cell-APC adhesion

Signalosomes

An organized cluster of proteins that coordinately controls cell signal transmission and regulation

Naïve T cells

T cells that have never encountered antigen. Entry into cell cycle and concomitant effector/memory differentiation requires sustained stimulation from hours to days by engagement of TCR and CD28 molecules

Effector T cells

Short-lived differentiated T cells with immunological activities that can include cytokine production and cytolytic activity. Activation requirements are less stringent than for naïve T cells, requiring shorter TCR engagement (~ 1 hr) and no costimulation

Memory T cells

Long-lived differentiated T cells that persist in a resting state until re-encounter with specific antigen. Activating requirements are similar or identical to effector cells. Upon re-stimulation, memory cells proliferate and differentiate into secondary effector and memory populations. Memory cells may be directly derived from stimulated naïve cells (divergent development model), or surviving effector cells (linear development model) (Reviewed in [126, 127])