T-Cell Receptor Signalling: The Complete Pathway, Branch by Branch

The T-cell receptor is the most finely calibrated detector in immunology. It must discriminate an agonist peptide present at a few dozen copies from a self peptide present at tens of thousands, and it does so with a receptor that has no catalytic activity of its own. Every downstream event — the calcium flux, the Ras switch, the NF-κB translocation, the IL-2 burst — is built from borrowed kinases, adaptors with no enzymatic function, and phosphatases that decide how much signal survives.

Signal 1: the receptor, the co-receptors and the Lck set-point

The TCR/CD3 complex binds peptide–MHC through its αβ heterodimer and signals through ten ITAMs spread across the CD3γε, CD3δε and ζζ chains. None of those chains is an enzyme. Phosphorylation comes from Lck, tethered to the tail of CD4 or CD8 by a zinc-clasp motif. That tether is the point of the co-receptor: by binding the same MHC molecule the TCR is engaging, it delivers Lck to an engaged receptor rather than a bystander.

Lck activity is set by two opposing enzymes, and this is where much of the discrimination happens. Csk phosphorylates the inhibitory C-terminal tyrosine, locking Lck closed; CD45 removes it. Because CD45 also strips the activation-loop phosphate, its effect is dose-dependent — too little and Lck stays shut, too much and it is disarmed. Around 40% of Lck in a resting cell is already primed: the receptor is held at a threshold, not switched off. Segregating the bulky CD45 ectodomain out of close-contact zones turns that small imbalance into a decision, and Fyn handles the weak-agonist end of the range.

ZAP-70 and the LAT signalosome

Doubly phosphorylated ITAMs create a bidentate docking site for the tandem SH2 domains of ZAP-70. Docking alone is not enough: ZAP-70 arrives autoinhibited and must be licensed by Lck on Y315/Y319. That two-step requirement is a proofreading step, and it is why transient, low-affinity contacts fail to propagate. ZAP-70 deficiency in humans gives a selective CD8 lymphopenia with non-functional CD4 cells.

Active ZAP-70 has two substrates that matter, and neither is an enzyme. LAT is a palmitoylated adaptor whose phosphorylated tyrosines nucleate a condensate holding PLCγ1, Gads, Grb2·SOS and PI3K; Gads brings in SLP-76, which recruits Itk and Vav1. Mutating the one LAT tyrosine that binds PLCγ1 causes lymphoproliferation in mice rather than simple immunodeficiency, because the branches downstream are not equally weighted. Vav1 then drives Rac1, WASP and cofilin to build the actin sheet that holds the synapse together.

The fork at PLCγ1: calcium, Ras–ERK and PKCθ

Itk phosphorylates PLCγ1, which hydrolyses PIP₂ into diacylglycerol and IP₃; everything distal is a division of those two products. IP₃ empties the ER store, STIM1 senses the depletion and opens ORAI1, and the sustained Ca²⁺ plateau loads calmodulin to activate the phosphatase calcineurin. Calcineurin then dephosphorylates NFAT and exposes its nuclear localisation signal — the step ciclosporin and tacrolimus block.

DAG has two receivers. RasGRP1 loads Ras with GTP at the membrane; SOS, delivered by Grb2 on LAT, is a second GEF allosterically primed by Ras-GTP itself. RasGRP1 makes the first Ras-GTP, which turns SOS into a positive-feedback amplifier — the basis of the digital, all-or-none ERK response seen in single cells. Ras drives Raf → MEK1/2 → ERK1/2, and ERK phosphorylates Elk-1 and RSK to induce c-Fos, which pairs with JNK-driven c-Jun as AP-1.

The second receiver is PKCθ, the only PKC isoform stably held at the synapse centre. It phosphorylates CARMA1, which oligomerises with BCL10 and the paracaspase MALT1 to activate IKK, degrade IκBα and release NF-κB. The branches converge on the IL-2 promoter combinatorially: its composite NFAT–AP-1 site needs both partners at once, and NFAT arriving alone transcribes the anergy programme instead.

Signal 2: CD28 and the PI3K–AKT–mTOR arm

CD28 engagement by CD80/CD86 does not start a separate pathway so much as lower the threshold and extend the duration of the one already running. Its YMNM motif recruits PI3K, generating PIP₃ that brings PDK1 and AKT to the membrane; PDK1 supplies T308 and mTORC2 S473, while PTEN opposes the arm by returning PIP₃ to PIP₂.

AKT output is largely a matter of what it switches off. It excludes FOXO1 from the nucleus, removing the programme that maintains quiescence and naive homing, and inactivates GSK-3β — the kinase that re-phosphorylates NFAT to expel it, so inhibition prolongs NFAT residence. AKT also stabilises c-Myc and sustains Bcl-xL. mTOR ties this to physiology: mTORC1 licenses aerobic glycolysis and biases differentiation toward effector fates, which is why rapamycin suppresses effectors while expanding memory and Treg populations.

The brakes: CTLA-4, PD-1, Cbl-b and DGKα

CTLA-4 and CD28 compete for the same ligands, and CTLA-4 binds roughly twenty-fold more tightly. Its dominant mechanism is not signalling but trans-endocytosis: it strips CD80 and CD86 off the antigen-presenting cell and degrades them, removing signal 2 for every T cell nearby. That is why CTLA-4 knockout mice die of lymphoproliferation within weeks while PD-1 knockouts develop only late, organ-restricted autoimmunity. PD-1 works intracellularly instead: its ITSM recruits SHP-2, with help from SHP-1, which dephosphorylates CD28 preferentially over the TCR itself.

Two intrinsic brakes complete the picture, and both enforce anergy. Cbl-b is an E3 ligase that ubiquitinates PI3K and Vav1; CD28 signalling degrades it, so its removal is how costimulation is recorded, and Cbl-b-deficient T cells activate without CD28 at all. DGKα converts DAG to phosphatidic acid, terminating the RasGRP and PKCθ branches at source; it is upregulated in anergic and exhausted cells, and inhibiting it restores Ras–ERK signalling.

Key targets and matching reagents

Studying TCR signalling in vivo

The proximal half of this pathway is intracellular and best measured biochemically, but the synapse and its checkpoints are surface events, and therefore directly addressable with functional-grade antibodies.

1. Engaging or depleting the T-cell compartment

Anti-CD3 (145-2C11) is the workhorse: plate-bound it drives polyclonal activation, and given systemically it induces transient depletion and receptor modulation. Anti-CD4 (GK1.5) and anti-CD8a (Ly-2) deplete the respective subsets and are the standard way to assign a phenotype to a lineage. Because αCD3 is a potent mitogen, low endotoxin matters more here than almost anywhere: LPS contamination drives cytokine release indistinguishable from an on-target response. Run isotype-matched functional-grade controls throughout.

2. Manipulating costimulation and the checkpoints

Anti-CD28 supplies signal 2 alongside αCD3 and separates anergy from deletion. Anti-CTLA-4 (9D9) and anti-PD-1 (RMP1-14) reproduce the clinical mechanisms in mouse tumour models, and their non-overlapping kinetics — CTLA-4 at priming, PD-1 in the effector phase — make sequencing studies informative rather than merely additive. Read them out with downstream nodes: NFAT, NF-κB p65 and c-Myc report whether proximal signalling was restored.

3. Quantifying the intracellular cascade

Lck, Fyn, ZAP-70, Itk and Csk cover the proximal set-point; LAT, SLP-76, Gads and Vav1 the signalosome; Ras, MEK1, ERK2, JNK1 and AP-1/c-Jun the MAPK branch; PI3K, PDK1, AKT1, PTEN, mTOR, FOXO1 and GSK-3β the metabolic arm. For NF-κB use BCL10, MALT1, IKKβ, IκBα and p65, with Bcl-xL, Cbl-b and DGKα for survival and anergy, and Rac1, WASP and cofilin-1 for the cytoskeleton.

Featured products for this pathway

Frequently asked questions

For research use only. Not for use in diagnostic or therapeutic procedures.