Immunoreceptor tyrosine-based activation motifs (ITAMs) are crucial for TCR signal transduction. Each ITAM consists of a conserved YxxL/I-X6-8-YxxL/I motif, where the tyrosine residues play a central role in signal initiation (14–16). These tyrosines are phosphorylated upon TCR engagement, predominantly by LCK, creating docking sites for signaling molecules such as ZAP70 (Table 1). This phosphorylation is essential, as substituting tyrosine (Y) with phenylalanine (F), a residue that cannot be phosphorylated, renders the ITAM non-functional, effectively abolishing downstream signaling (17). Furthermore, ITAM phosphorylation must be reversible to allow dynamic regulation of TCR signaling. This has been demonstrated using phosphomimetic variants, where glutamic acid substitutions mimic the negative charge of phosphorylated tyrosines, impairing T-cell activation and highlighting the importance of dynamic phosphorylation (18).

The TCR-CD3 complex contains ten ITAMs, with one ITAM on the cytoplasmic tails of CD3ϵ, CD3δ, and CD3γ, and three ITAMs on each ζ chain (14–16). ITAMs are not unique to the TCR; various receptors on hematopoietic cells, including the B-cell receptor (BCR), several Fc receptors, and certain NK-cell receptors signal through cytoplasmic ITAMs (14, 19). Interestingly, receptors that bind abundant or polyvalent antigens, such as the BCR, typically contain two to four ITAMs (20). In contrast, the αβ TCR, which demonstrates extremely high antigen sensitivity by showing transient calcium signaling to even a single agonist pMHC ligand (21, 22), contains a much larger number of ITAMs. This might suggest that the presence of multiple ITAMs in the TCR-CD3 complex contribute to signal amplification. Indeed, studies in murine models have shown that reducing the number of ITAMs below seven per TCR-CD3 complex impairs TCR function during central tolerance in the thymus, leading to autoimmune disorders (23). However, recent findings suggest that modified TCR-CD3 complexes with only four functional ITAMs are more responsive to weak antigens than their wild-type (WT) counterparts with ten ITAMs, reflecting the nuanced roles of ITAMs in modulating signal strength (24).

Moreover, the exact amino acid sequences of each ITAM within the TCR-CD3 complex differ (Figure 1), resulting in different phosphorylation efficiency by kinases such as LCK and in varied binding affinities to downstream signaling molecules. This diversity renders ITAMs functionally non-equivalent, with distinct qualitative contributions to TCR signaling (23, 25). Evidence for this functional diversity comes from studies in mice engineered to have a single ITAM sequence across all ten positions. These mice exhibit impaired thymocyte development, failing to transition effectively from double-negative (DN) to double-positive (DP) stages, due to reduced TCR expression and signaling (25).

Interestingly, ITAMs are not limited to signal activation; some also play regulatory roles. For instance, monophosphorylation of the CD3ϵ ITAM (ϵITAM) recruits the C-terminal Src kinase (CSK) (1), while monophosphorylated ITAMs of CD3δ (δITAM) and ζ (ζITAM) recruit the SH2-containing protein tyrosine phosphatase-1 (SHP1) (3, 24) (Table 1). Both are negative regulators of TCR signaling, highlighting the dual role of ITAM phosphorylation in balancing activation with feedback inhibition to regulate T-cell activation. Although our understanding of the qualitative and quantitative roles of individual ITAMs within the TCR-CD3 complex remains incomplete, current data emphasize the importance of ITAM multiplicity and diversity during T-cell development, signaling and tolerance (Figure 2).

The unique properties of individual ITAMs are likely influenced by their interaction with other motifs within the CD3 cytoplasmic tails. However, the lack of structural data limits our understanding of these interactions. Cryo-electron microscopy (Cryo-EM) studies of the TCR-CD3 complex have yet to resolve the cytoplasmic tails, due to their high flexibility. This flexibility may facilitate a range of molecular interactions but could also constrain or modulate specific signaling events. Further research into the three-dimensional organization and dynamics of these tails is essential to elucidate how their configurations shape TCR signaling.

Basic-rich stretch (BRS) motifs are arginine and lysine-enriched amino-acid sequences that mediate electrostatic interactions with nearby negatively charged molecules, such as acidic lipids (26). These motifs are present in several transmembrane proteins, including FcϵRIγ, CD43, CD44, CD28, 4-1BB (CD137) and ICAM1/2 (27), as well as cytosolic proteins (28, 29). Within the TCR-CD3 complex, BRS motifs are found in the cytoplasmic tails of CD3ϵ and ζ (Figure 1).

The BRS of CD3ϵ (ϵBRS) is located in the juxtamembrane region of its cytoplasmic tail (30, 31) (Figure 1). This motif binds to different acidic phospholipids found in the plasma membrane and various cellular organelles, anchoring CD3ϵ to the membrane (31). The ζ chain, in contrast, contains three BRS motifs: two between ζITAM1 and ζITAM2, and one between ζITAM2 and ζITAM3 (ζBRS1-3) (32).

Landmark studies propose that BRS motifs sequester and protect ITAMs from phosphorylation in the absence of TCR-ligand engagement by binding to phospholipids in the inner leaflet of the plasma membrane (33–36). However, contrary to this protective role, mutations that disrupt the ability of either the εBRS or ζBRSs to interact with the plasma membrane unexpectedly lead to reduced TCR signaling (31, 37). This apparent paradox highlights the complexity of BRS functions. Mice lacking a functional ϵBRS exhibit impaired DN3-DN4 thymocyte transition and defective positive selection, underscoring the importance of the ϵBRS in TCR signaling during T-cell development (38) (Figure 2).

Beyond lipid interactions, the ϵBRS has also been reported to interact with signaling proteins, such as GRK2, CAST, LCK and p85 (1, 30, 39, 40) (Table 1). However, no protein-protein interactions involving the ζBRSs have been identified to date.

The mechanisms underlying BRS functions remain incompletely understood. Whether their role in TCR signaling depends primarily on lipid binding, protein interactions, or a combination of both is still unclear. Loss-of-function mutations in the ϵBRS, for example, reduce TCR surface levels in both developing and peripheral T cells and impair TCR localization to the immunological synapse (31). These findings suggest a role for the ϵBRS in synapse formation, endosomal recycling, and TCR degradation. However, the precise molecular mechanisms governing these processes require further investigation.

The TCR contains a Proline rich sequence (PRS) within the CD3ϵ cytoplasmic tail (Figure 1), which recruits the adaptor protein NCK through its first Src-homology 3 (SH3.1) domain, facilitating efficient TCR signaling (41) (Table 1). This interaction also stabilizes the association between LCK and the TCR, further enhancing signal propagation (42). Additionally, the PRS serves as a binding site for other signaling molecules, including NUMB (43) and EPS8 (44), underscoring its multifunctional role in TCR signaling (Table 1).

Studies in PRS-knock-in mice (PxxP to AxxA) revealed significant impairments in T-cell development. Thymocytes exhibited a partial arrest at the DN3-DN4 transition, reduced efficiency in both positive and negative selection at the DP stage, and defects in maturation into CD4 single positive (SP) and CD8 SP thymocytes. All these observations support a role for the PRS at check points where effective pre-TCR and TCR signaling is required (45, 46) (Figure 2). Interesting, immature thymocytes in these mice displayed increased TCR surface levels, likely as a compensatory mechanism to the reduced signaling output per TCR molecule.

Furthermore, it was demonstrated that PRS mutations in CD3ϵ selectively affect in vitro T-cell responses to weak but not strong antigenic peptides (47). This finding aligns with studies showing that the absence of NCK impairs T-cell activation under weak antigenic stimulation (48). Together, these results highlight the PRS as a key regulator of TCR signaling, particularly in modulating responses to antigens of low strength.

A unique feature of the TCR is the receptor kinase (RK) motif, located within the ITAM of the CD3ϵ cytoplasmic tail (Figure 1). This motif, defined by the sequence RKxQRxxY, directly recruits the kinase LCK by interacting with its SH3 domain (2) (Table 1). Structural studies revealed that the RK motif binds to the RT loop and n-SRC loop of LCK’s SH3 domain. Notably, most RT loop residues are not conserved in related kinases, providing a potential explanation for the unique role of LCK in T-cell development and activation (49, 50).

The specificity of the SH3(LCK)-RK(CD3ϵ) interaction is underscored by the rarity of similar motifs. To date, only two similar motifs have been identified: one in the adaptor protein SKAP1 (RKxx(Y)xxY) (51) and another in Candida albicans (52). These findings highlight the exceptional role of the RK motif in mediating TCR-specific recruitment of LCK. Functional studies demonstrate that mutations of the RK motif reduce TCR signaling and T-cell activation and negatively affect thymocyte development (2) (Figure 2). These results establish the RK motif as a critical regulator of TCR signaling and highlight its essential role in coordinating proximal T-cell activation.

The cytoplasmic tail of CD3ε contains an endoplasmic reticulum (ER) retention motif (NQRRI) that plays a crucial role in regulating the assembly and surface levels of the TCR (53). This motif ensures that only fully assembled TCR-CD3 complexes are transported to the plasma membrane. Mutations in the ER retention sequence allow the surface expression of incomplete αβγϵ and αβδϵ complexes that lack the ζ chains, highlighting its critical quality-control function (54).

CD3γ holds a membrane proximal di-leucine motif involved in TCR downregulation after phosphorylation of the serine residue five amino acids prior by protein kinase C (PKC) (55, 56). This motif has been shown to be important for controlling T-cell homeostasis and the response to virus infections (57, 58). CD3δ also contains a di-leucine motif, which is however missing the precedent serine residue and is therefore currently being considered nonfunctional.

The diversity of motifs within the CD3 subunits highlights their specialized contributions to the TCR function. CD3ϵ stands out as being particularly enriched in signaling and regulatory motifs – including the BRS, PRS, RK, ITAM, and ER retention motif – suggesting a central role in both assembly and signal integration. In contrast, ζ, CD3γ and CD3δ seem to exhibit more singular functions: signal amplification (ζ), downregulating the TCR through the di-leucine motif (CD3γ) or by recruiting the negative regulator SHP1 (CD3δ and ζ). The reasons behind this asymmetry in motif distribution remain an open question. It is possible that evolutionary pressures have tailored CD3ϵ to act as a multifunctional hub, while ζ, CD3γ and CD3δ play more ancillary roles. Future studies focusing on the distinct contributions of these subunits may reveal new layers of complexity in TCR regulation.

In the absence of ligand binding, the αβ TCR-CD3 complex exists in a compact, resting conformation (Figure 3). Cryo-EM studies using nanodiscs aiming to mimic native lipid environments reveal that the TCRαβ ectodomains are tightly associated with the CD3 ectodomains and the transmembrane and juxtamembrane regions remain compacted, preserving the TCR in a closed conformation (64). This finding aligns with independent molecular dynamics simulations (65). It however contrasts with cryo-EM studies conducted in detergent environments, where this compact conformation was not observed (66–69), plausibly because detergents disrupt crucial lipid-TCR interactions. These interactions are essential for maintaining the receptor’s membrane dynamics and are insufficiently mimicked by detergents.

The importance of lipid interactions for TCR function is supported by both computational and experimental evidence. In silico molecular dynamics simulations suggest that TCR-lipid interactions stabilize the receptor’s resting state (70), while experimental studies using biophotonic approaches demonstrate that dephosphorylation of phosphoinositides in the plasma membrane releases the CD3ϵ cytoplasmic tail, enhancing its accessibility to LCK and facilitating signaling (71). Further evidence for the important roles of lipids for TCR functions was suggested by a study showing that cholesterol plays a key role in stabilizing the TCR’s resting state by interacting with the TCRβ transmembrane (TM) region (59, 72). This interaction is key in suppressing phosphorylation of the ζITAMs (73). In line with this, recent structural studies have identified two cholesterol molecules embedded within the TM regions of the TCR-CD3 complex (64, 67), further underscoring the importance of lipid interactions for the TCR.

Additional lipid interactions help guard the resting TCR state. For example, the BRS motifs in the CD3ϵ and ζ cytoplasmic tails bind the inner leaflet of the plasma membrane, shielding the ITAMs from premature phosphorylation and regulating their availability for downstream signaling (34, 35, 70, 71). These interactions provide a compelling regulatory mechanism, ensuring the TCR remains inactive in resting T cells. Interestingly, while some basal phosphorylation of the ζ chain can be detected in non-stimulated cells, the εITAMs appear to be fully protected from phosphorylation (35, 74). This differential “tonic” phosphorylation may be explained by the fact that the CD3ϵ cytoplasmic domain has a higher affinity for lipids than the ζ cytoplasmic domain (27). Thereby, specific tyrosines in ζ could become transiently accessible to kinases in a subset of resting TCRs (35). This basal phosphorylation of ζ is largely mediated by LCK (75).

Cumulative evidence support that the αβ TCR-CD3 complex undergoes conformational changes upon binding to activating pMHC (64, 76–78). This structural transition results in the exposure of critical motifs within the CD3 cytoplasmic tails, allowing the simultaneous recruitment of multiple signaling proteins (2, 37, 41, 59, 62, 72, 79, 80) (Figure 3). It has been proposed that mechanical forces generated during the TCR-pMHC binding induce the formation of so-called catch bonds that stabilize the interaction between the TCR and agonistic pMHC, prolonging signaling (81, 82).

Although, the key role of LCK in TCR phosphorylation is broadly accepted, the mechanism of LCK recruitment to the ligand-engaged but unphosphorylated TCR remains an active area of study. As LCK is the only SFK member associated with CD4 and CD8 co-receptors, the intuitive model proposes that LCK is recruited to TCR-pMHC class I and TCR-pMHC class II via CD8 and CD4, respectively (83). However, T-cell signaling can be induced in a co-receptor-independent manner, suggesting the existence of additional mechanisms of LCK docking at the TCR-CD3 complex.

Indeed, it has been shown that the RK motif within CD3ϵ recruits LCK via its SH3 domain to the ligand-engaged TCR (2). The ϵBRS motif further enhances this process by binding LCK’s unique domain, increasing the local concentration of LCK at the TCR (1). With each TCR-CD3 complex containing two CD3ϵ chains, the recruitment of two LCK molecules may promote trans-phosphorylation, further accelerating TCR activation. However, experimental evidence supporting this hypothesis is still lacking. Not all ITAM tyrosines are equally phosphorylated, and phosphorylation of individual tyrosines has distinct functional consequences (1). For example, phosphorylation of the first tyrosine in the ϵITAM stabilizes the LCK-CD3ϵ interaction, while phosphorylation at the second tyrosine disrupts it (18). The PRS motif within CD3ϵ also contributes to LCK recruitment by binding the adaptor protein NCK, which in turn stabilizes LCK localization at the TCR (41, 42, 47, 84, 85). NCK localization at the TCR is further stabilized when the NCK SH2 domain binds to the phosphorylated second tyrosine of CD3ϵ (86), while phosphorylation of the first CD3ϵ tyrosine disrupts NCK binding (44, 86). In addition, LCK interacts with phosphotyrosine motifs in the TCR-CD3-ZAP70 complex via its SH2 domain to further amplify the signaling (87–89). The importance of the LCK SH2 domain is highlighted by the observation that LCK with mutated SH2 does not trigger TCR signaling in Jurkat cells (90).

Together, these findings suggest that multiprotein interactions, which are highly dynamic and precisely regulated, control LCK concentration nearby the TCR upon ligand binding.

Meanwhile, large membrane proteins such as the phosphatases CD45 and CD148 are excluded from the close-contact zones of TCR-pMHC contact. This steric exclusion prevents premature dephosphorylation, ensuring efficient propagation of the TCR signal (61, 91–94).

Dual phosphorylation of the ITAMs by LCK creates high-affinity docking sites for the SH2 domains of ZAP70 (19, 95–97) (Figure 3). ZAP70’s binding follows a hierarchical pattern, with the highest affinity for the membrane-proximal ζITAM1, δITAM and γITAM (15), followed by the additional ITAMs of ζ (ζITAM2/3) and the lowest affinity for ϵITAM (97–99). Upon binding to dually phosphorylated ITAMs, ZAP70 undergoes a conformational change that repositions its SH2 domains and relieves autoinhibitory interactions within its linker region (100). This structural rearrangement primes ZAP70 for phosphorylation by LCK, fully activating ZAP70’s kinase activity (96, 97).

Once activated, ZAP70 phosphorylates the key scaffolding proteins LAT and SLP76 (101), creating platforms for recruiting additional signaling molecules necessary for downstream signal propagation (6, 102, 103). The efficiency of signal propagation depends on the stability of the TCR-pMHC interaction: a half-life long enough to sustain ITAM phosphorylation by ZAP70 is crucial for engaging downstream pathways, ultimately leading to T-cell activation, proliferation, and differentiation (104, 105).

These proximal signaling events do not occur in isolation, but rather within the specialized microenvironment of the immunological synapse. Upon TCR engagement, local calcium (Ca²⁺) concentrations surge due to the co-localization of TCRs and CRAC channels. Ca²⁺ can directly bind to lipid phosphate groups, neutralizing their charge and disrupting ionic interactions with the ϵBRS and ζBRSs. This disruption further exposes ITAM tyrosines for phosphorylation and enhances signal amplification (106). This Ca²⁺-dependent mechanism may allow bystander TCRs, not directly engaged with antigen, to become activated as well. Although monomeric pMHC alone might be sufficient to initiate proximal TCR signaling events (21, 107–109), TCR clustering within the synapse significantly enhances signal transduction by concentrating ITAMs, kinases, and other signaling molecules in close proximity (110, 111), ensuring effective downstream signal propagation.

LCK activity is highly regulated by maintaining a balance between its active and auto-inhibited states. Two key tyrosine residues control this balance: the activating tyrosine in the catalytic domain (Y394 in humans), which must be phosphorylated for full kinase activity, and the inhibitory tyrosine in the C-terminal domain (Y505 in humans). Phosphorylation of Y505 facilitates an intramolecular bond with the SH2 domain, locking LCK in a closed, inactive conformation (116, 117).

The equilibrium between these active and inactive states is highly dynamic and regulated by a network of phosphatases and kinases, including the tyrosine kinase CSK, which phosphorylates Y505 (118), the membrane-bound protein tyrosine phosphatases CD45 and CD148 (75, 119, 120), as well as cytosolic phosphatases such as PTPN6 (SHP1) (121), PTPN22 (122), and the PEST domain-enriched tyrosine phosphatase (PEP) (123, 124). Furthermore, it has been recently proposed that LCK binds to the homodimeric receptor CD146, a cell-adhesion molecule, which leads to LCK’s trans-autophosphorylation and activation (125). Together, these regulatory mechanisms ensure that LCK activity, and consequently TCR phosphorylation, remains tightly controlled in the absence of ligand engagement, preventing unwanted activation.

However, it is unclear how TCR engagement would mechanistically instruct LCK’s catalytic activity beyond the recruitment of active LCK molecules close to the TCR. A substantial proportion of LCK is constitutively active in resting T cells (126–128). Some of this active LCK is located near the TCR, even in the receptor’s resting state, poised for immediate engagement when the TCR transitions to its prime state (Figure 3) (2). Whether TCR triggering increases the pool of active LCK and what the mechanism and kinetics would be is still a subject of debate (126, 129–131). A straightforward explanation, pending experimental confirmation, is that this pre-activated LCK directly phosphorylates the ligand-bound TCR without requiring further activation of additional LCK molecules. Such a mechanism would enable the rapid initiation of TCR signaling.

LCK exists either as a free pool or in association with the co-receptors CD4 or CD8 (5, 132). As LCK is the dominant SFK in T-cell activation, and the only family member capable of interacting with the co-receptor CD4 and CD8, an intuitive model of TCR signaling postulated that CD4 or CD8 deliver LCK to the TCR-pMHC complex, serving as the primary trigger of TCR signaling (133).

However, this paradigm has evolved with evidence showing TCR signaling can occur independently of co-receptor engagement. Experiments using anti-CD3 antibodies, CARs, or MHC class I mutants with a 10-fold weaker CD8-affinity than WT (134), demonstrate that TCR signaling does not strictly require co-receptor interaction. Additionally, unconventional T cells, which recognize non-MHC antigens, naturally bypass co-receptor involvement. Collectively, these findings reveal that co-receptor aggregation of the TCR is not an essential mechanism for TCR signal initiation (Figure 4). While co-receptor-bound LCK is not strictly required for TCR triggering, it plays a significant role in enhancing the sensitivity of TCR signaling, especially in response to antigens with suboptimal affinity (135–138). Recent insights into the physiological roles of CD4-LCK and CD8-LCK were addressed in a landmark study using knock-in mice expressing a LCK CA mutant (C20A and C23A), which does not interact with CD4 or CD8 (137). Although these mice showed reduced positive selection, their phenotype was relatively mild especially in terms of cytotoxic responses to viral infections and tumors.

In sharp contrast, LCK-deficient or kinase-dead LCK models exhibit severe impairments (49, 137, 139–141). This suggests that free LCK can largely compensate for the loss of co-receptor-associated LCK in vivo (137).

LCK associates with CD4 through a zinc clasp formed by cysteine-rich motifs in the CD4 cytoplasmic tail and the N-terminal part of LCK (5, 132, 142). A large proportion of CD4 molecules are coupled with LCK in developing DP thymocytes (136), with a modest increase in CD4-LCK coupling observed during CD4⁺ T-cell maturation (143). This is likely due to the higher affinity of LCK for CD4 compared to CD8 (132) and the expression of a truncated CD8 isoform unable to bind LCK (144), leading to preferential sequestration of LCK to CD4 in the DP stage (145). Estimates from co-immunoprecipitation studies suggest that between 36% and 80% of CD4 is bound to LCK in primary murine CD4⁺ T cells from lymph nodes (143, 146). Given the relatively weak co-receptor-LCK interaction and possible artifacts upon cells lysis with detergents, methods that do not require cell lysis, such as in situ proximity ligation assays (PLA) or quantitative single-molecule microscopy may provide more accurate occupancy measurements.

Genetic disruption of the CD4-LCK interaction reduces CD4⁺ SP thymocyte numbers and impairs helper T-cell functions. However, some CD4⁺ T-cell functions can be rescued by kinase-dead CD4-bound LCK, suggesting a kinase-independent role for CD4-LCK in stabilizing CD4 localization at the plasma membrane (137). Similarly, chimeras formed by CD4 fused to kinase-deleted LCK were shown to be more efficient at supporting full activation of a CD4-dependent T-cell hybridoma than the full-length LCK chimeric fusion protein (147). This kinase-independent stabilization of CD4 by LCK is essential for regulating the surface levels and trafficking of CD4 in primary T cells, as previously demonstrated in transgenic non-lymphoid cell lines (148, 149). This suggests that CD4 has LCK-independent or at least kinase-independent roles in T-cell development and signaling.

While CD4-bound LCK is not strictly required for TCR signaling, it enhances signal transduction (136, 150), fine-tunes antigen sensitivity (137), and stabilizes CD4 at the cell surface (137, 148, 149). This underscores the modulatory role of co-receptor-associated LCK in antigen sensitivity and signal amplification.

LCK associates with CD8 in a manner similar to CD4, though CD8 and CD4 differ structurally and functionally. CD8 exists in two isoforms, CD8α and CD8β, which form homo- or heterodimers. Conventional CD8⁺ αβ T cells express the CD8αβ heterodimer, while γδ T cells and NK cells primarily express the CD8αα homodimer (151). LCK associates with the CD8α subunit (152). Despite the potential for CD8αα to bind two LCK molecules, CD8β enhances the interaction between LCK and CD8α, raising questions about how much LCK associates with CD8αα homodimers (153).

In DP thymocytes, CD8-bound LCK is less abundant compared to CD4-bound LCK (136). However, during T-cell maturation, the proportion of CD8-bound LCK increases, correlating with enhanced homeostatic responses to self-antigens in mature CD8⁺ T cells (136, 143).

Although CD8-LCK is essential for fine-tuning cytotoxic T-cell responses to weak antigens in a kinase-dependent manner, CD8-bound LCK appears largely dispensable for the in vivo development of cytotoxic T cells and anti-viral/anti-tumor responses (137). However, supraphysiological CD8-LCK stoichiometry, in T cells expressing a chimeric CD8.4 co-receptor, consisting of the extracellular CD8α domain and the intracellular LCK-binding CD4 part, has relatively strong effects on enhancing the positive selection of very weakly self-reactive thymocytes (154), decreasing antigen affinity threshold for negative selection (136), and inducing antigen-independent memory-like T cells (also known as virtual memory T cells) (155). This suggests that the relatively minor role of the CD8-LCK interaction in vivo might be, at least partially, caused by the relatively low CD8-LCK stoichiometry, which makes free LCK dominant even in WT CD8⁺ T cells.

LCK delivered to the TCR by CD4 or CD8 during thymic selection was proposed to impose MHC restriction on the developing αβ TCR repertoire (156). Supporting this theory, TCRs in mice lacking both, co-receptors and MHC (quad-deficient mice), develop non-pMHC-specificities resembling those of antibodies (145, 157) and are enriched for CD1d-restricted iNKT receptors (137). However, the co-receptor-LCK interaction is not strictly required for MHC restriction, as TCR transgenic monoclonal MHC class I- or II-restricted thymocytes develop normally in the LCK CA mice (137). Given the prominent role of CD8 and CD4 co-receptors in the lineage commitment, these observations reveal LCK-independent roles of the co-receptors in T-cell triggering, probably by stabilizing the pMHC-engaged TCR (158, 159). Although a role of CD4 in TCR-pMHCII stabilization has been strongly questioned based on its low affinity to isolated pMHCII complexes (150), recent evidence shows that the CD4-pMHCII affinity increases substantially when the pMHC is engaged by a cognate TCR (160).

The role of free LCK versus co-receptor-bound LCK in TCR signaling has been a subject of significant interest. Current evidence suggests that free LCK is more efficient than CD8-bound LCK in initiating ITAM phosphorylation (161) and that CD8-bound LCK is recruited only later, stabilizing the whole TCR-pMHCI-CD8-LCK complex (158, 161). This model is supported by studies showing that free LCK moves faster and is more associated with the TCR at the steady-state in thymocytes than co-receptor-bound LCK (128). These findings indicate that free LCK may be particularly critical for triggering the ligand-engaged TCR in contexts where co-receptors do not interact with the ligand, such as during antibody stimulation, in CAR T cells, or in unconventional T cells that recognize non-MHC antigens. Mathematical modeling further emphasizes the role of co-receptor-bound LCK in signal amplification rather than signal initiation. Recruitment of CD4-LCK enhances TCR phosphorylation threefold compared to free LCK alone. However, when recruited to a pre-phosphorylated TCR, this effect is amplified 30- to 40-fold, additionally suggesting that co-receptors may act to boost signaling after the initial trigger has been established (162).

As mentioned above, co-receptor-bound LCK is crucial for signaling triggered by low-affinity ligands. These ligands bind shortly to the TCR and may require additional mechanisms to recruit sufficient LCK to complete all phosphorylation steps before the ligand dissociates. The kinetic proofreading model highlights the need for all proximal signaling steps to occur within the limited duration of the TCR-pMHC interaction to enable downstream signaling (104, 105, 135, 163–165). The cooperative engagement of free and co-receptor-bound LCK may reduce the time required for LCK to access and phosphorylate the ITAMs. This is supported by studies in compound heterozygous mice expressing the LCK CA mutant, which is unable to bind CD8 (free LCK only) and a kinase-dead LCK mutant bound to CD8. These mice exhibit weaker responses to suboptimal antigens than homozygous LCK WT or even LCK CA mice (137), suggesting that the kinase activity of CD8-LCK facilitates TCR signaling when the duration of the TCR-pMHC bond is the limiting factor.

During T-cell development, CD4 and CD8 co-receptors are highly expressed in DP thymocytes, ensuring preferential selection of TCRs recognizing MHC antigens and enabling discrimination between self and non-self MHC molecules. These high co-receptor levels reduce the availability of free LCK for TCRs that recognize non-MHC ligands, preventing the positive selection of DP thymocytes bearing such TCRs (166) and lowering the relative abundance of positively selected iNKT cells (137). In contrast to DP cells, SP thymocytes and mature T cells express only one co-receptor that matches the MHC-restriction of their TCR. In these cells, it is still not clear whether and how the roles of free LCK and co-receptor-bound LCK differ. Studies show that LCK-deficient primary T cells expressing only the LCK CA mutant, which cannot bind co-receptors, exhibit weaker ζ-chain phosphorylation than cells expressing WT LCK, suggesting that both free and co-receptor-bound LCK pools are critical for robust TCR-CD3 signal transduction (161).

Despite substantial advances, key aspects of free versus co-receptor-bound LCK remain unresolved. Open questions include the precise stoichiometry of CD8- and CD4-bound LCK specially in mature conventional T cells (136, 143, 167), potential differences in the kinase activities of free versus co-receptor-bound LCK (128, 168), and whether these pools perform distinct mechanistic roles during TCR signal initiation (137, 161). The balance between free and co-receptor-bound LCK reflects a finely tuned mechanism that ensures both sensitivity and efficiency in TCR signaling. Relative contributions of these LCK pools may depend on factors such as LCK and co-receptor expression levels, the proportion of LCK coupled to the co-receptors, ligand affinity, and the presence of phosphatases. Understanding this balance is critical for comprehending TCR signaling dynamics and could provide insights into therapeutic strategies for modulate immune responses in cancer immunotherapy and other immunomodulatory contexts.

Similar to αβ TCRs, γδ TCRs assemble with the CD3 signaling complex. In humans, this includes CD3ϵγ, CD3ϵδ, and ζζ dimers, while in mice, the complex features two CD3ϵγ dimers and ζζ (176). Structural and functional differences between human and mouse γδ T cells complicate direct translation of findings between species.

Key structural differences between αβ and γδ TCRs lie in their constant regions and connecting peptides, which influence receptor assembly, surface orientation, and charge distribution (177, 178). For example, the extracellular domains of γδ TCRs exhibit greater conformational flexibility compared to αβ TCRs (179, 180). This increased flexibility may enable γδ T cells to recognize a wider range of ligands with varying sizes and structures. A recent cryo-EM study uncovered variability in γδ TCR assembly based on the Vγ chain usage (179) (Figure 5). For instance, the Vγ9Vδ2 TCR appears monomeric and exhibits substantial conformational flexibility in its γδ TCR ectodomains and connecting peptides, resembling the dynamics of Fabs (fragment antigen-binding) regions and hinge linkers of membrane-bound Igs. FLIM-FRET analysis further supports this, showing that other γδ TCRs, such as Vγ4Vδ1, Vγ8Vδ3, and Vγ3Vδ1, also exist as monomers (179). In contrast, the Vγ5Vδ1 TCR forms a dimeric structure, which is crucial for its activation, and FLIM-FRET analysis indicates that Vγ2Vδ1 TCRs also assemble as dimers (179) (Table 2). These structural differences underscore the unique assembly mechanisms and flexibility of γδ TCRs compared to the more rigid αβ TCR complexes (Figure 5). While cryo-EM studies have provided valuable insights into γδ TCR assembly (179, 180), future studies employing native-like nanodiscs instead of detergent environments could further illuminate the diverse conformations of γδ TCRs. Although these structural variations suggest that the mechanisms of TCR activation may not be fully conserved between αβ and γδ TCRs, some parallels might exist. For instance, cholesterol has been shown to restrain signaling in both Vγ9Vδ2 and Vγ5Vδ1 TCRs, possibly by altering the conformation of the ζ chains (179). This suggests that certain regulatory processes may be shared between αβ and γδ TCRs, by conservation of key residues across TCR subtypes and species. The following sections will further explore both shared and unique mechanisms of activation between αβ TCRs and γδ TCRs.

In αβ T cells, the binding of pMHC induces the stabilization of TCR conformational states exposing the signaling motifs within the CD3 complex. Mice with mutations in the CD3ϵ stalk region (ϵC80G), which prevents the exposure of the PRS motif upon αβ TCR engagement (181), demonstrate that this transition is essential for αβ T-cell development. Whether this mutation solely prevents PRS exposure or also impairs the switch of the αβ TCR to its active state is to date unknown. Interestingly, this mutation differentially affects γδ T-cell subsets, suggesting that γδ TCRs may not represent a single, uniform receptor, but rather a collection of receptors, each with distinct activation mechanisms (Table 2). For instance, Vγ1⁺, Vγ7⁺ and Vγ5⁺ (non-dentritic epidermal T cells (DETC)) γδ T cells develop normally in ϵC80G mice, while Vγ4⁺ and Vγ5⁺ (DETC) subsets are reduced (181). Additional evidence for this diversity among γδ TCRs comes from a study showing that the human Vγ9Vδ2 TCR, expressed in a Jurkat cell line, does not expose the PRS upon stimulation with its endogenous ligand (182), which was also true for the T22-specific murine G8 (Vγ4⁺) and KN6 γδ TCR (182, 183). Together, these findings highlighting the heterogeneous nature of γδ TCRs and their distinct activation mechanisms.

γδ TCR signaling is also initiated by the phosphorylation of CD3 ITAMs, likely mediated by SFKs. Inhibition of CSK, a negative regulator of SFKs, enhances ERK phosphorylation in γδ T cells, underscoring the importance of SFKs in γδ TCR signaling (184). Unlike αβ TCR signaling, which predominantly relies on LCK, evidence suggests that different γδ T cell subsets may rely on distinct SFKs.

γδ T cells express significantly lower levels of LCK compared to αβ T cells (185). Moreover, γδ T cells are present in LCK-deficient mice, which lack αβ T cells, indicating that LCK is not essential for all γδ T-cell subsets (186). Supporting this, patients with loss-of-function mutations in LCK exhibit a drastic reduction in αβ T cells, yet a relative increase in γδ T cells, suggesting that certain γδ T-cell populations are LCK-independent (187, 188). Notably, in one patient, Vδ1⁺ γδ T cells accumulated, whereas Vδ2⁺ γδ T cells were reduced, indicating that Vδ2⁺, but not Vδ1⁺, γδ T cells may depend more on LCK (188).

In mice, IL-17-producing Vγ6⁺ and Vγ4⁺ γδ T cells lack LCK expression and can function without it, consistent with earlier findings that LCK is dispensable for the development of these subsets (189–191). In contrast, Vγ3⁺ T-cell development is severely impaired in Lck^-/-, Lck/Fyn^-/- and Cd45^-/- mice (49, 192) and murine CD27⁺ γδ thymocytes, which are primed to differentiate into IFNγ-producing γδ T cells, exhibit high LCK expression (190). Global gene expression analyses reveal that the B cell-specific lymphoid kinase (BLK) is expressed in murine γδ T cells but not in αβ T cells, mediating the development of Vγ6⁺ IL-17-producing γδ T cells (193, 194), indicating that other SFKs beyond LCK may play a role in γδ TCR signaling.

γδ TCR signaling strength has been linked to distinct effector fates in mice. Strong γδ TCR signaling promotes the differentiation of IFNγ-producing subsets, predominantly Vγ1⁺, Vγ5⁺, and Vγ7⁺ γδ T cells (195, 196), which likely depend on LCK. In contrast, weaker γδ TCR signaling favors the development of IL-17-producing subsets (majority of Vγ4⁺ and Vγ6⁺ γδ T cells) (195, 196), which may rely on alternative SFKs like BLK.

How SFKs are recruited to the γδ TCR presents an intriguing question, especially since most γδ T cells do not express the co-receptors CD4 or CD8 and recognize their ligands without the participation of CD4 or CD8. This suggests that alternative mechanisms, potentially involving CD3 cytoplasmic motifs such as the BRS or RK motif, facilitate the SFK recruitment. If these motifs are functional in γδ T-cell subsets needs to be addressed in the future. While ZAP70 is essential for αβ T-cell development, some γδ T cells persist in ZAP70-deficient mice, suggesting that ZAP70 is not universally required for all γδ subsets (197). Consistent with this, studies using Zap70^-/-, Sykb^-/-, and Zap70/Sykb^-/- double KO mice show that SYK is required for the generation of neonatal Vγ1⁺ and Vγ4⁺ γδ T cells, while Vγ5⁺ cells rely on either ZAP70 or SYK, and Vγ6⁺ cells depend on both (198). Given that SYK activation is less dependent than ZAP70 on SFKs (199, 200), it is likely that LCK-dependent γδ T-cell subsets preferentially use ZAP70, whereas SYK-dependent subsets do not rely on LCK. However, the γδ T cells existing in the different models do not fully support this (oversimplifying) idea (Table 2) and further investigation is needed.

It has been proposed that γδ TCR signal initiation by nonclassical MHC class 1b antigens requires the exclusion of phosphatases (CD45) from the vicinity of the TCR, as consequence of cell-cell close contact zones (183). This adds to the idea that tipping the equilibrium between kinases and phosphatases in the vicinity of the engaged-TCR aids to activation. Whether this is the case for all γδ TCRs and all γδ T-cell subsets needs to be elucidated.

Understanding the interplay between the different SFKs and γδ TCR downstream signaling pathways will be essential for fully elucidating the molecular mechanisms governing γδ T-cell function and their potential therapeutic applications.

The role of γδ T cells in tumor surveillance is well-established. Studies in mice have shown that the absence of γδ TCRs increases susceptibility to certain cancers (201). In humans, the presence of γδ T cells is the most favorable survival predictor across 25 malignancies and most solid tumors (202). Their presence at tumor sites correlates with improved clinical outcomes in colorectal, gastric, and head and neck cancers (203–205). However, the role of human γδ T cells in cancer appears to be influenced by their subset-specific characteristics. Vδ1⁺ T cells, predominantly found in tissues, form diverse subsets by pairing with various Vγ chains. While some Vδ1⁺ T cells recognize glycolipids presented by CD1c and CD1d molecules (206, 207), the specific TCR antigens for most Vδ1⁺ T cells remain unknown. These cells exhibit potent anti-tumor activity in colorectal cancer, multiple myeloma, and chronic lymphocytic leukemia (208–210). However, they can also exert immunosuppressive effects (211–213). In contrast, the human Vδ2⁺ subset exclusively pairs with the Vγ9 chain to form Vγ9Vδ2 T cells, the predominant γδ T-cell population in peripheral blood. These cells recognize phosphoantigens derived from pathogens or tumor cells (214) and have been extensively studied in clinical settings due to their strong anti-tumor activity (215–220). The roles of human Vδ3⁺ and Vδ5⁺ γδ T cells in cancer remain less characterized (171, 221–224).

Emerging therapeutic strategies aim to leverage the unique properties of γδ T cells. These include the development of bispecific cell engagers, adoptive transfer of expanded Vδ1⁺ or Vδ2⁺ T-cell populations, and genetic engineering approaches such as CAR γδ T cells or αβ T cells engineered to express specific γδ TCRs (225). Recent clinical trials exploring γδ T cells in cancer immunotherapy have shown promising, but still modest, results emphasizing their potential as a tool in cancer treatment (171, 213, 226).

To fully harness the therapeutic potential of γδ T cells, a deeper understanding of the molecular mechanisms underlying γδ TCR signaling is essential. In particular, the proximal signaling steps leading to γδ T-cell activation require further investigation. The following sections will examine how insights into proximal αβ TCR signaling events have driven the development of novel therapeutic approaches (4), providing a framework to inspire innovations in γδ T cell-based therapies.

CARs are engineered receptors that redirect T cells to recognize and kill tumor cells. A CAR typically consists of three main components: an extracellular binding domain, generally a single-chain variable fragment (scFv) derived from an antibody, that recognizes the target antigen; a TM region anchoring the receptor in the membrane; and intracellular signaling domains responsible for activating the T cell. In the following section, we focus on different intracellular signaling domains, without delving into other structural advances such as in the binding domain or TM region of CAR designs.

Early CAR designs were based on the ζ chain of the TCR, known for its strong signaling capacity. These first-generation CARs successfully drove T-cell proliferation, cytokine production, and cytotoxicity in vitro and in vivo (227, 228). However, their limited persistence in vivo hindered their therapeutic potential. To address this, second-generation CARs incorporated co-stimulatory domains originating from co-stimulatory receptors such as 4-1BB (also known as CD137 and TNFSFR9) or CD28, alongside the ζ chain (henceforth referred to as BBζ and 28ζ CARs, respectively; Figure 6). This innovation significantly improved CAR T-cell persistence and tumor-eradicating capability (229). Notably, CD28 contains two basic-rich stretches (BRSs) that interact with the plasma membrane and LCK, whereas 4-1BB has only one juxtamembrane BRS. It was shown that the higher signaling intensity of 28ζ CARs, compared to BBζ CARs, results from constitutive LCK association with CD28 (230). However, recent findings also revealed that 28ζ CAR signaling can occur independent of LCK, relying instead on the kinase FYN, whereas both TCR and BBζ CAR signaling are LCK-dependent (231).

Most CAR designs retain the ζ chain, assuming its multiple ITAMs would amplify signaling. However, it has become evident that reducing ITAM numbers may actually improve therapeutic outcomes. CARs with a single functional ζITAM (1XX CARs) demonstrated superior anti-tumor efficacy compared to those with two or three ITAMs in a CD19⁺ tumor in vivo model (232). These 1XX ITAM CARs promote a favorable memory T-cell phenotype with enhanced persistence (232), supporting the notion that reducing signaling potency can reduce activation-induced cell death (AICD) and exhaustion, particularly in cases of high tumor antigen expression. These promising findings have already inspired clinical trials using the 1XX CAR format (233, 234).

Recent advances in CAR design have focused on incorporating signaling domains beyond the ζ chain ITAMs, particularly leveraging the unique functionalities of CD3 subunits such as CD3ϵ, CD3δ, and CD3γ (Figure 6). These efforts aim to enhance CAR T-cell efficacy, persistence, and safety.

For instance, BBεζ CARs, which combine a 4-1BB co-stimulatory domain with CD3ϵ and ζ chains, exhibit superior anti-tumor activity in leukemia models compared to BBζ CARs (2, 8) by recruiting LCK through the RK motif (2). Similarly, the BRS motif of CD3ε modulates interactions with LCK and p85, reducing cytokine production while increasing T-cell persistence for a ε28ζ CAR (CD3ε, CD28, and ζ chain) relative to traditional 28ζ CARs (1). CARs incorporating the CD3ϵ cytoplasmic tail together with a CD28 co-stimulatory domain (28ϵCAR) showed increased efficacy in a solid tumor models while minimizing inflammatory cytokine release compared to a 28ζ CAR (235). Additionally, CD3ϵ-containing CARs regulate LCK activity through the recruitment of the kinase CSK via the mono-phosphorylated ITAM of CD3ϵ, providing negative feedback control (1). Mutations to the BRS, RK, or ITAM motifs in CD3ε-based CARs impaired CAR T-cell function in vitro, underscoring their critical role in optimizing CAR signaling (3).

Expanding on this, CARs incorporating signaling domains from CD3δ or CD3γ in the context of the 41BB co-stimulatory domain show enhanced anti-tumor efficacy, reduced tonic signaling, and lower cytokine secretion compared to BBζ CARs (3). CD3δ, for instance, recruits SHP1 through monophosphorylation of the second ITAM tyrosine, providing regulatory control and promoting stem-like T-cell properties and CAR functionality (3). Furthermore, mutations in the di-leucine motifs of δCARs and γCARs boost surface expression, T-cell cytotoxicity and activation, highlighting the role of this motif in regulating CAR functionality (3).

These findings underscore the critical importance of understanding proximal TCR signaling to guide the rational design of CARs. While current FDA-approved CAR therapies rely on second-generation ζ chain constructs, emerging evidence suggests that alternative designs incorporating CD3ϵ, δ, or γ signaling domains may provide significant advantages, including enhanced efficacy, reduced tonic signaling, and improved T-cell persistence (Figure 6). The next frontier in CAR T-cell optimization could lie in harnessing the unique properties of these underexplored CD3 subunits. Clinical trials will be essential to determine whether these novel CAR constructs can surpass the performance of current therapies. Moreover, the strategic combination of different CD3 signaling domains (1, 2) represents a promising avenue to further refine CAR designs, potentially unlocking more durable and effective cancer immunotherapies.

Alternative CAR formats have been developed to bypass the earliest TCR signaling steps, ITAM phosphorylation, by directly integrating downstream signaling proteins (236, 237). These herein termed bypassCARs are coupled directly to molecules such as LCK, FYN, the kinase domain (KD) of ZAP70 (ZAP70_KD), LAT, SLP76, or PLCγ1, without including any co-stimulatory domains (238). Among these designs, the ZAP70_KD-bypassCAR (Figure 6) demonstrated superior tumor control in a solid-tumor mouse model compared to a conventional BBζ CAR (238). An additional study demonstrated that adding the CD28 signaling domain before the ZAP70_KD (Figure 6) resulted in a construct capable of inducing tumor remission comparable to a BBζ CAR and more durable than a 28ζ CAR in a Nalm6 tumor mouse model (239).

These findings highlight the potential of bypassCARs to overcome limitations in traditional designs, but further studies are needed to optimize their efficacy, evaluate possible undesired tonic signaling and thereby, premature exhaustion, and evaluate their clinical relevance across diverse tumor models.

TCR-like chimeric receptors represent an exciting frontier in engineered T-cell therapies, leveraging the entire TCR-CD3 complex for signal transduction instead of isolated CD3 chains. Ideally, these receptors function independently of MHC.

One such approach, TCR-fusion constructs (TRuCs), involves fusing a scFv to the extracellular domain of CD3ϵ. TRuCs have shown superior efficacy in both hematopoietic and solid tumor models compared to second-generation BBζ and 28ζ CARs (240, 241) and are currently being investigated in clinical trials with promising interim results (242).

In another strategy, antibody-TCRs (AbTCRs), a full, anti-tumor Fab-fragment is fused to the constant domains of the TCRγ and TCRδ chains, creating new chimeric receptors that outperformed BBζ and 28ζ CARs in controlling B-cell tumors in vivo. These AbTCR-T cells also exhibited reduced cytokine production and exhaustion marker levels compared to 28ζ CAR T cells (243). Using the constant domains of TCRγ and TCRδ prevents undesired pairing with endogenous TCRα and TCRβ chains, reducing the risk of generating off-target specificities that could lead to autoimmunity. However, how these AbTCRs are coupled to proximal signaling pathways has yet to be elucidated, therefore, the exact mechanisms underlying their superior performance remain unclear.

Synthetic TCR and Antigen Receptors (STARs) (244) and HLA-independent TCRs (HITs) (245) replace the variable regions of TCRα and TCRβ with antibody-derived variable regions. In preclinical mouse models, STAR and HIT T cells provided superior tumor control compared to 28ζ CARs, especially in tumors with low antigen expression (245). STAR and HIT T cells were also more effective at infiltrating solid tumors, less prone to exhaustion, and persisted longer than BBζ and 28ζ CARs. Interestingly, STAR and HIT T cells produced similar or even higher amounts of cytokines than BBζ and 28ζ CARs upon tumor recognition, likely due to their enhanced sensitivity (243, 245, 246). This can eventually be a drawback for these receptors, as increased cytokine production could pose the risk for severe side effects such as cytokine release syndrome (CRS) or immune effector cell-associated neurotoxicity syndrome (ICANS) (247).

TCR-like chimeric receptors might represent a major advance in T cell-based therapies by leveraging the complete TCR complex to enhance proximal signaling and anti-tumor efficacy. Constructs such as TRuCs, AbTCRs, STARs, and HITs have demonstrated superior performance in preclinical models, particularly in targeting tumors with low antigen expression and addressing limitations of second-generation ζCARs. Their ability to persist longer, resist exhaustion, and infiltrate solid tumors underscores their potential for treating refractory cancers.

Despite these promising preclinical data, clinical experience with most of these next-generation designs remains scarce: only a small number of early-phase trials currently evaluate 1XX CARs (NCT05757700), CD3ϵ-based CARs (NCT06373081), or TCR-like receptors such as TRuCs (NCT03907852), and STARs (NCT06321289, NCT03953599, NCT05548088, NCT05518357, NCT04508842) so their behavior in real-world settings is still largely unknown. In the clinic, mesothelin-targeting TRuC-T cells (NCT03907852 (242),) have shown objective responses in heavily pretreated solid tumors but also highlighted on-target/off-tumor and cytokine-mediated toxicities, illustrating both the promise and risks of engaging the full TCR-CD3 complex. Early CD19-targeting 1XX CAR T and STAR-T trials will similarly test whether tuning ζ ITAM content or harnessing complete TCR signaling can improve persistence and efficacy without exacerbating CRS, ICANS or antigen escape, but mature data are not yet available. Preclinical and first-in-human results therefore suggest that some constructs may mitigate challenges of conventional CAR T cells, yet broad implementation may be limited by manufacturing and regulatory complexity and by unresolved questions about long-term persistence, exhaustion and safety in heterogeneous tumors. Systematic head-to-head clinical evaluation of ζ-based CARs, CD3ϵ/δ/γ-containing CARs, bypassCARs and TCR-like receptors will be essential to define which designs offer the most favorable balance between efficacy, safety and feasibility, advancing the next generation of cancer immunotherapies.