Membrane Translocation of Protein Kinase Cθ during T Lymphocyte Activation Requires Phospholipase C-γ-generated Diacylglycerol

Protein kinase Cθ (PKCθ) is the only PKC isoform recruited to the immunological synapse after T cell receptor stimulation, suggesting that its activation mechanism differs from that of the other isoforms. Previous studies have suggested that this selective PKCθ recruitment may operate via a Vav-regulated, cytoskeletal-dependent mechanism, independent of the classical phospholipase C/diacylglycerol pathway. Here, we demonstrate that, together with tyrosine phosphorylation of PKCθ in the regulatory domain, binding of phospholipase C-dependent diacylglycerol is required for PKCθ recruitment to the T cell synapse. In addition, we demonstrate that diacylglycerol kinase α-dependent diacylglycerol phosphorylation provides the negative signal required for PKCθ inactivation, ensuring fine control of the T cell activation response.

Protein kinase C (PKC) is the only PKC isoform recruited to the immunological synapse after T cell receptor stimulation, suggesting that its activation mechanism differs from that of the other isoforms. Previous studies have suggested that this selective PKC recruitment may operate via a Vav-regulated, cytoskeletal-dependent mechanism, independent of the classical phospholipase C/diacylglycerol pathway.
Here, we demonstrate that, together with tyrosine phosphorylation of PKC in the regulatory domain, binding of phospholipase C-dependent diacylglycerol is required for PKC recruitment to the T cell synapse. In addition, we demonstrate that diacylglycerol kinase ␣-dependent diacylglycerol phosphorylation provides the negative signal required for PKC inactivation, ensuring fine control of the T cell activation response.
During an immune response, T lymphocyte interaction with an antigen-presenting cell (APC) 1 at the so-called immunological synapse (1) governs lymphocyte responsiveness. The molecular signaling complex formed at the cell-cell contact site triggers a cascade of biochemical events, culminating in cell proliferation and the execution of T cell effector functions (2).
An important consequence of T cell receptor (TCR) stimulation is phospholipase C-␥1 (PLC-␥1) activation. This enzyme is recruited to the membrane by adaptor proteins and is also activated directly by tyrosine phosphorylation (3,4). The activated enzyme hydrolyzes phosphatidylinositol 4,5-diphosphate to inositol 1,4,5-triphosphate, which stimulates Ca 2ϩ release from intracellular stores and to diacylglycerol (DAG). The importance of Ca 2ϩ and DAG in T cell activation was demonstrated several years ago, when it was shown that a combination of Ca 2ϩ ionophore and phorbol esters, which function as DAG analogues, could mimic TCR signals, leading to full T cell activation (5). The relevance of PLC-␥1 in TCR signals was further highlighted by the impaired TCR activation in PLC-␥1deficient cell lines (6). Similarly, mutations in the T cell adapter LAT, which prevents PLC-␥1 recruitment and activation (3), abrogate expression of several T cell activation-associated genes, including interleukin-2 (IL-2) (7). PLC-␥1-dependent Ca 2ϩ elevation activates the phosphatase calcineurin, leading to dephosphorylation of the transcription factor NF-AT (8). The exact role of PLC-␥-generated DAG nonetheless remains a matter of controversy. According to conventional models of TCR signaling, cell responses to increased DAG are due to protein kinase C (PKC) activation (9). The recent discovery of RasGRP, a Ras guanine exchange factor containing a DAG-binding C1like domain, has nevertheless established a direct link between TCR-mediated PLC-␥1 activation and the Ras-extracellular signal-regulated kinase pathway (10).
Is well established that DAG generation is responsible for membrane translocation and activation of classical and novel PKC family members (11,12). T lymphocytes express several classical and novel PKC isoforms, although only the novel isoform PKC is selectively recruited to the immunological synapse (13). Analysis of PKC knockout mice showed the importance of this isoform in regulating TCR-derived signals and demonstrated the requirement for PKC in activating the downstream elements AP-1, NF-B, and IL-2 in T cells (14). There is thus great interest in dissecting the mechanism underlying selective PKC recruitment to TCR contact sites during antigen stimulation. Recent experiments suggested that PKC recruitment to the cell membrane is mediated by Vav/ Rac activation through a PLC-␥-independent mechanism that involves cytoskeletal reorganization (15)(16)(17)(18)(19). Nonetheless, analysis of Vav1-deficient mice showed that during T cell activation, this exchange factor for Rac acts as an upstream regulator of PLC-␥1 (20). According to these results, PLC-mediated DAG generation would be a direct consequence of Vav-dependent cytoskeletal reorganization, making it difficult to assess the specific contribution of DAG generation among Vav-regulated signals.
Here we used several approaches to evaluate the exact contribution of PLC-␥-dependent DAG generation in the control of TCR-mediated PKC recruitment and activation. We used pharmacological inhibitors, PKC mutants, and expression of constitutive active or transdominant negative forms of DGK␣, which acts as a negative modulator of DAG levels, to show that tyrosine phosphorylation of and DAG binding to PKC are indispensable for selective PKC recruitment to the immunological synapse in T lymphocytes.
Recombinant Plasmids-The plasmid encoding EGFP-PKC under a pEF promoter was generated by excising PKC from the PKC-EGFP construct (Clontech, Palo Alto, CA) with XhoI and SacII (this end was blunted). The 2.1-kb fragment encoding the PKC cDNA was subcloned in the pEGFP-Bos vector (pEGFP-PKC) at the EGFP C terminus. To generate mutant constructs, Tyr 90 was replaced by Phe or Asp using the QuikChange TM site-directed mutagenesis kit from Stratagene and appropriate oligonucleotides. The kinase-dead and membrane-targeted versions of DGK␣ fused to EGFP have been described (21).
Cell Culture and Transfection-The JHM1-2.2 cell line was generated by stable transfection of the human muscarinic subtype 1 receptor in the Jurkat leukemic human T cell line (22). Jurkat and JHM1-2.2 cells were maintained in RPMI 1640 medium containing 10% heatinactivated fetal calf serum, 2 mM L-glutamine, 1 mM sodium pyruvate, 10 mM Hepes, and 100 g/ml each penicillin and streptomycin (complete medium). Cells in exponential growth were electroporated with 20 g of DNA constructs using a Gene Pulser (270 V, 975 microfarads; Bio-Rad). Cells were immediately transferred to 30 ml of complete medium and assayed after 24 h.
Blood Collection and Lymphocyte Purification-Blood was drawn from healthy volunteers. Buffy coat (40 ml) was overlaid on a Ficoll- Hypaque gradient (1:2) and centrifuged, and the lymphocyte-containing interface band was collected and washed with serum-free RPMI 1640 medium. The cell pellet was suspended in RPMI 1640 with 10% fetal calf serum, and cells were counted in a Neubauer hemocytometer.
Anti-CD3/CD28 Stimulation and Immunofluorescence-Cells were harvested 24 -36 h after electroporation. For anti-CD3/CD28 stimulation, slides were precoated with a 1:1 mixture of anti-CD3/CD28 antibody (final concentration 10 g/ml each) in 50 mM Tris-HCl buffer, pH 8. Lymphocytes were plated onto slides and examined by confocal microscopy after 15 min. For pharmacological inhibition, cells were pretreated (30 min) with inhibitors before plating onto anti-CD3/CD28coated slides. For stimulation with antibody-coated microspheres, antibodies were adsorbed to microspheres by mixing 0.5 g of antibody in phosphate-buffered saline with 0.5 ϫ 10 6 microspheres (final volume 1 ml) and incubating (1.5 h, room temperature) with continuous mixing; 1.5 ml of 1% bovine serum albumin in PBS was added, and mixing continued (30 min). Microspheres were washed three times with phosphate-buffered saline and resuspended in phosphate-buffered saline. For stimulation, 10 6 transfected cells were mixed with antibody-coated beads at a 2:1 cell/bead ratio and plated on chambered slides. A time series of images was captured by confocal microscopy.
Analysis of Cell Surface CD69 or CD25 Expression-Cells were plated in anti-CD3/CD28-coated 6-well plates. Cell surface CD69 or CD25 expression was analyzed 20 h later using phycoerythrin-conjugated anti-human CD69 or phycoerythrin-conjugated anti-human CD25 monoclonal antibody. Immunofluorescence intensity of cells was determined by flow cytometry (EPICS-XL; Beckman Coulter). For inhibitor treatment, cells were incubated with vehicle or inhibitors for 30 min and then plated as above. For analysis of CD69 expression in transfected cells, cells were seeded 24 h after transfection onto anti-CD3/CD28-coated plates, and CD69 expression was analyzed after 6 or 20 h, gating for green fluorescent protein-positive and -negative cells.

RESULTS
PKC is the only PKC isoform recruited to the immunological synapse after TCR stimulation, suggesting that its activation mechanism differs from that of the other isoforms. To assess the changes in subcellular PKC location following TCR stimulation in intact T cells, we generated an EGFP-PKC construct. When transiently transfected Jurkat cells were stimulated with anti-CD3/CD28-coated polystyrene beads, this construct translocated to the cell-bead contact zone (Fig. 1A), as described for the endogenous PKC. To analyze the specificity of the PKC recruitment mechanism in response to TCR engagement, protein translocation was evaluated in JHM1-2.2, a Jurkat subcell line stably transfected with the human muscarinic type I receptor. Carbachol stimulation of these cells induces DAG-dependent activation signals through PLC-␤ activation (22). As in the parental Jurkat line, PKC translocated to the membrane when these cells were stimulated with anti-CD3/CD28-coated beads (not shown) or with immobilized anti-CD3/CD28 antibody (Fig. 1B, upper panel). Under these conditions, some perinuclear localization was also seen. Cell treatment with the Src-like kinase-specific inhibitor PP2 (23) prevented EGFP-PKC translocation after TCR activation. Translocation was also observed after carbachol stimulation, but, in this case, PP2 treatment did not prevent membrane localization of the EGFP-PKC construct (Fig. 1B, lower panel).
CD69 is a lymphocyte activation marker; its expression is DAG-mediated and correlates with PKC activation (17,24). We thus analyzed PKC activation by measuring CD69 expression. Cell surface CD69 expression increased in response to both TCR and human muscarinic type I receptor stimulation (Fig. 1C). As for EGFP-PKC membrane translocation, PP2 treatment of cells prevented CD69 expression in response to anti-CD3/CD28 stimulation but did not affect CD69 induction after carbachol stimulation. This validates the use of CD69 expression as an indicator of PKC translocation.
To assess the contribution of DAG generation to PKC translocation, EGFP-PKC translocation was determined in intact Jurkat cells following treatment with various pharmacological inhibitors known to interfere with DAG-based signals. Treatment with the PLC-␥ inhibitor U73122 prevented EGFP-PKC translocation to the membrane, indicating that PLC-␥-dependent DAG generation is essential for membrane localization (Fig. 2A). The addition of calphostin C, which competes with PKCs for the DAG binding site, also impeded EGFP-PKC membrane translocation after TCR engagement ( Fig. 2A). Finally, treatment with the PKC inhibitor bisindolylmaleimide I hydrochloride (25) did not affect PKC translocation ( Fig. 2A), indicating that activity of other PKC isoforms is not involved in PKC translocation. Although translocation to the plasma membrane is blocked by U73122 and calphostin C, EGFP-PKC appears to remain associated to internal membranes ( Fig. 2A).
In response to TCR triggering, PKC is tyrosine-phosphorylated and can be detected in anti-Tyr(P) immunopellets (26). Inhibition of PLC-␥ activation did not alter the amount of PKC immunoprecipitated by anti-Tyr(P) after TCR stimulation (Fig.  2B). PLC-␥ inhibition thus blocks PKC translocation, without affecting Lck-mediated PKC Tyr phosphorylation or its recruitment to Tyr-phosphorylated protein complexes. To corre- late PKC membrane translocation with its biological activity, we examined CD69 expression after PLC-␥ inhibition. Following U73122 treatment, CD69 expression was severely impaired, to an extent similar to that seen after PP2 inhibition (Fig. 2C).
To demonstrate that DAG-dependent PKC activation is a physiological mechanism in the lymphocyte immune response and is not cell line-specific, similar experiments were performed using peripheral blood lymphocytes (Fig. 3A). Both blockade of PLC-␥ activation and treatment with the PKCspecific inhibitor rottlerin (25) block CD69 up-regulation. This supports the finding that PLC-␥-mediated DAG production also controls CD69 expression in human lymphocytes via PKC activation. PKC regulates lymphocyte activation through expression of the transcription factors NF-B and AP-1 and subsequent induction of the IL-2 and CD25 genes (14,24,27,28). Specific inhibition of PLC-␥ or PKC diminished surface CD25 expression and impaired lymphocyte progression to later activation stages (Fig. 3B).
It has been proposed that PKC Tyr 90 is the phosphorylatable residue in response to TCR activation (26). To assess the specific contribution of Tyr phosphorylation in TCR-elicited PKC translocation, two EGFP-PKC mutants were prepared. In one, Tyr 90 was replaced by Phe, preventing protein phosphorylation; in the other, Tyr 90 was replaced with Asp, which mimics the negative charge of phosphorylated Tyr, simulating phosphorylation at this residue. When analyzed by confocal microscopy, the EGFP-PKC(Y90F) protein did not translocate to the membrane after TCR stimulation (Fig. 4A).
The EGFP-PKC(Y90D) mutant was found in cytosol in unstimulated cells, and translocation was observed only after TCR engagement (Fig. 4B). Furthermore, PP2 inhibition of the Lck pathway or U73122 inhibition of DAG generation also blocked EGFP-PKC(Y90D) translocation following stimulation.
Earlier studies ruled out DAG requirement for PKC translocation, based on the observation that this protein is found at the membrane after TCR stimulation in a PLC-␥1-deficient Jurkat subcell line (19). As these authors showed by fractionation analysis, we observed that PKC translocated to the plasma membrane when immobilized anti-CD3/CD28 was used as a stimulus (Fig. 5). Western blot analysis of the two PLC-␥ isoforms known to be present in T lymphocytes showed that, whereas PLC-␥1 was absent in these cells, PLC-␥2 was present at high levels. Analysis after TCR stimulation demonstrated that, although to a lower extent than in parental Jurkat cells, Ca 2ϩ levels were increased in these PLC-␥1-deficient cells. We confirmed that this Ca 2ϩ increase was PLC-␥-dependent, since it was blocked by U73122 treatment (Fig. 5). TCR-induced Ca 2ϩ release was not blocked by the physiologically inactive analog U73343 (not shown).
In T lymphocytes, clearance of PLC-generated DAG is regulated by DGK␣, which converts this lipid to PA (21). To correlate PKC activation with DAG clearance, we determined CD69 expression in Jurkat cells transiently transfected with a catalytically inactive DGK␣ form (DGK␣kd) known to induce higher, sustained DAG levels in response to TCR activation (29). In basal conditions, CD69 expression was up-regulated in transfected compared with untransfected cells (Fig. 6A). After TCR stimulation, CD69 expression was markedly elevated in cells expressing DGK␣kd, correlating with the increased DAG generation in these cells. Pharmacological inhibition of PLC-␥ and Lck returned CD69 expression to basal levels, indicating that the CD69 increase in DGK␣kd-expressing cells was due to attenuation of endogenous DGK activity and not to further regulation. We next measured CD69 expression in Jurkat cells transiently transfected with a constitutively active DGK␣ construct (Myr-EGFP-DGK␣). Myr-EGFP-DGK␣ locates at the plasma membrane, attenuating receptor-regulated responses by phosphorylating PLC-␥-generated DAG (21). Accordingly, in Myr-EGFP-DGK␣-transfected cells, cell surface CD69 expression was down-modulated compared with untransfected cells (Fig. 6B).
To confirm the role of DGK␣ as a negative modulator of PKC relocalization, we studied TCR-induced endogenous PKC translocation in Myr-EGFP-DGK-transfected Jurkat cells. Endogenous PKC translocation to the membrane was detected by confocal microscopy following TCR triggering or after the addition of the DAG analog PDBu to the cells (Fig.  6C). In TCR-stimulated Myr-EGFP-DGK-expressing cells, PKC translocation from cytosol to the membrane was impaired (Fig. 6D). Myr-EGFP-DGK expression did not affect PDBu-induced membrane translocation of PKC, showing that only TCR-induced signals were affected. DISCUSSION PKC translocation to the T cell synapse, which is formed when a major histocompatibility complex-bound peptide in APC contacts specific T cells, is crucial for subsequent T cell activation (13). Following TCR/CD28 costimulation, PKC activates IKK␤, leading to NF-B activation; this effect is T cell-specific, since PKC does not activate NF-B efficiently in nonlymphoid cells (28). In mature primary T cells, the NF-B cascade is the major and physiologically most important target of PKC in the TCR/CD28 costimulatory pathway leading to IL-2 production.
Receptor-induced PKC translocation to membranes is a complex process that involves protein phosphorylation, proteinlipid, and protein-protein interactions (30,31). Here we used several biochemical and genetic approaches to show that TCRmediated PKC translocation to the cell membrane requires an integrated two-step signal involving PKC Tyr phosphorylation as well as DAG generation at the plasma membrane.
The use of JHM1-2.2, a Jurkat-derived T lymphoid cell line stably transfected with the human muscarinic type I receptor, demonstrates that, through activation of distinct PLC isoforms, both TCR and G protein-coupled receptors can induce PKC translocation. Nonetheless, experiments using the Lck inhibitor PP2 indicate that Src tyrosine kinase-imposed regulation is restricted to TCR-induced PKC translocation, which implies receptor-dependent selective pathways.
Activation of the Src tyrosine kinases p56Lck and p59Fyn is a very early event in the TCR-triggered signaling cascade (2). Cell treatment with PP2 thus affects the phosphorylation/activation of several molecules including different types of adapters and signaling mediators, among them PLC␥-1. When DAG FIG. 5. PKC translocation in PLC␥-1-deficient Jurkat cells. A, PLC␥-1-deficient Jurkat cells were transfected, and PKC translocation was determined by confocal microscopy after anti-CD3/CD28 stimulation (15 min). B, to analyze PLC-␥ isoform expression, Jurkat wild type, PLC␥-1-deficient, and PLC-␥1-reconstituted cells were lysed, and PLC␥-1 and -2 were determined by Western blot. The regulatory subunit of phosphatidylinositol 3-kinase, p85, was used as loading control. C, to measure Ca 2ϩ generation after TCR triggering, Jurkat wild type and PLC␥-1-deficient cells were analyzed by flow cytometry for Ca 2ϩ release after stimulation with anti-CD3/CD28 antibodies (10 g/ml; 50 s). Where indicated, cells were pretreated with U73122 (1 M; 30 min) before stimulation. generation pathways are disrupted, PKC translocation is blocked, even when PKC can be precipitated with anti-Tyr(P) antibody. This lack of membrane association is reflected by an abrogated CD69 response, indicating that PKC translocation requires an integrated signaling mechanism in which Tyr phosphorylation and DAG generation are essential for subcellular relocation of the protein.
The experiments in peripheral blood lymphocytes demonstrate that PKC function is similar in human and in murine lymphocytes. Blockade of PKC activity results in a reduction in IL-2R␣ chain, preventing correct assembly of the heterotrimeric high affinity IL-2R. PKC-deficient mice show a severe defect in TCR/CD28-induced AP-1 and NF-B activation (14). The IL-2 gene promoter has consensus binding sites for these transcription factors; thus, neither IL-2 nor the IL-2R␣ chain (CD25) is efficiently induced in PKCϪ/Ϫ mice, resulting in diminished T cell proliferative responses (14). Accordingly, high IL-2 concentrations can restore lymphocyte activation and proliferation in PKCϪ/Ϫ mice (14,32).
Previous studies suggested that phosphorylation on Tyr 90 in PKC was a direct consequence of TCR stimulation (26). Here we confirm the importance of phosphorylation at this residue, showing that substitution of Tyr 90 by Phe prevents anti-CD3/ CD28-induced PKC translocation to the plasma membrane. Tyr 90 is not located within a consensus binding sequence for Src homology 2 domains (33), suggesting that Tyr 90 phosphorylation may induce a conformational change. This could allow PKC binding to DAG by exposing the C1 domains. Tyr 90 mutation would thus prevent PKC translocation by rendering PKC(Y90F) unable to interact with DAG.
Studies with the Asp 90 mutant, which mimics Tyr 90 phosphorylation, indicate that phosphorylation of this residue, albeit necessary, is not sufficient for TCR-induced PKC translocation and that an accessory signal is required. The disruption of anti-CD3/CD28-induced PKC translocation following pharmacological inhibition of PLC-␥ suggests that PLC-␥-mediated DAG generation constitutes this additional membrane targeting signal.
Pharmacological inhibition of PLC-␥ blocked PKC membrane translocation but did not alter its Tyr phosphorylation or its recruitment to Tyr-phosphorylated protein complexes. Under these conditions, PKC appears to be associated to internal membranes rather than being returned to cytosol. Other DAGregulated proteins were shown to translocate to internal localizations through protein-protein interactions; thus, ␤-COP, a Golgi protein, acts as a cellular receptor for PKC⑀ (34), whereas the ␤ 2 chimerin-interacting protein Tmp21 is found in the Golgi and endoplasmic reticulum (35). By inducing a conformational change in the absence of membrane DAG generation, Tyr phosphorylation might thus allow PKC association with cytoskeletal and/or internal membrane-associated proteins. PLC␥-dependent DAG generation would thus be necessary to ensure correct PKC association with the membrane after Tyr phosphorylation. It is important to note that if the absence of DAG caused PKC to remain bound to a cytoskeletal or particulate fraction, a fractionation assay would not distinguish this localization from association with the plasma membrane. This may clarify the previously reported PKC association to the particulate fraction in the absence of PLC␥-1 activation (19).
The DAG generation requirement for TCR-induced PKC membrane recruitment is further confirmed by the use of both constitutive active and transdominant negative DGK␣ forms. This lipid kinase phosphorylates DAG to PA, acting as a negative modulator of DAG-regulated signals (21). The mutants used here are known to modify existing DAG levels without affecting DAG generation and showed the negative role of DGK␣ in RasGRP-induced Ras activation (29). As previously demonstrated for RasGRP, we show that plasma membrane PKC localization is also negatively regulated by DGK␣, confirming the role of this lipid kinase as a modulator of the intensity and duration of TCR-derived signals.
It was recently reported that selective PKC recruitment to the TCR engagement site may operate via a phosphatidylinositol 3-kinase-dependent, PLC-␥/DAG-independent mechanism (19). These authors used PLC-␥1-deficient Jurkat cells to show PKC translocation to the membrane. A model was suggested in which phosphatidylinositol 3-kinase acting upstream of the Vav/Rac pathway would regulate PKC translocation via cytoskeletal rearrangement (15,17,19). The Vav/Rac pathway reorganizes the T cell actin cytoskeleton and facilitates TCR capping (36,37). PKC may bind to a cytoskeletal or scaffold protein associated with the immunological synapse. We nonetheless found high PLC-␥2 levels in these cells and found that the TCR-dependent Ca 2ϩ elevation was U73122-sensitive, indicating TCR-dependent activation of this PLC isoform. The data thus suggest that PLC-␥ isoforms may have an overlapping role in lymphocytes and that these PLC-␥1-deficient cells cannot be used to rule out a DAG requirement in PKC translocation. The phenotype analysis of VavϪ/Ϫ mice showed disruption of PLC-␥ activation, demonstrating that PLC-␥ is one of the main effectors of Vav activation (20). Together, these data suggest that Vav-induced, PKC-mediated lymphocyte activation requires PLC-␥ for PKC translocation. Experiments are currently under way to analyze the functional relationship between the Lck/PLC-␥ and phosphatidylinositol 3-kinase/Vav pathways in spatio-temporal regulation of PKC translocation.
DAG generation during T cell activation is a key step in Ras/extracellular signal-regulated kinase pathway initiation through RasGRP translocation (29). Here we report the essential role of DAG in PKC membrane localization. Altogether, these conclusions allow us to suggest that PLC-␥mediated DAG generation at the T cell synapse has an essential role in the polarized recruitment of signaling molecules during T cell activation. Whereas Tyr phosphorylation of adaptors and cytoskeletal reorganization ensure relocation of signaling proteins to the T cell synapse, the local increase in DAG levels at the membrane is an essential step to allow rapid recruitment of C1-containing proteins to the proximity of the TCR. A model can be envisaged (Fig. 7) in which Lck-triggered phosphorylation of several effectors, together with Vav/Rac-induced dynamic cytoskeletal reorganization, activates PLC-␥1 to increase local DAG levels. Simultaneously, Tyr phosphorylation of PKC promotes a conformational change, exposing its C1-DAG-binding domains that anchor the protein to the membrane. Fine control of PKC activation requires that the elevation in DAG levels be rapid and transient. This is regulated by DGK␣ translocation to the membrane, where the lipid kinase decreases membrane DAG levels by phosphorylating DAG to PA, dissociating PKC, and terminating the signal. DAG generation at the site of the immunological synapse allows the spatio-temporal protein relocalization that ensures signal transmission during a T cell response.