Signaling through the T cell receptor (TCR) controls adaptive immune responses. Antigen binding to TCRαβ transmits signals through the plasma membrane to induce phosphorylation of the CD3 cytoplasmic tails by incompletely understood mechanisms. Here we show that cholesterol bound to the TCRβ transmembrane region keeps the TCR in a resting, inactive conformation that cannot be phosphorylated by active kinases. Only TCRs that spontaneously detached from cholesterol could switch to the active conformation (termed primed TCRs) and then be phosphorylated. Indeed, by modulating cholesterol binding genetically or enzymatically, we could switch the TCR between the resting and primed states. The active conformation was stabilized by binding to peptide-MHC, which thus controlled TCR signaling. These data are explained by a model of reciprocal allosteric regulation of TCR phosphorylation by cholesterol and ligand binding. Our results provide both a molecular mechanism and a conceptual framework for how lipid-receptor interactions regulate signal transduction.

No abstract

Ligand binding to the TCR causes a conformational change at the CD3 subunits to expose the CD3ε cytoplasmic proline-rich sequence (PRS). It was suggested that the PRS is important for TCR signaling and T cell activation. It has been shown that the purified, recombinant SH3.1 domain of the adaptor molecule noncatalytic region of tyrosine kinase (Nck) can bind to the exposed PRS of CD3ε, but the molecular mechanism of how full-length Nck binds to the TCR in cells has not been investigated so far. Using the in situ proximity ligation assay and copurifications, we show that the binding of Nck to the TCR requires partial phosphorylation of CD3ε, as it is based on two cooperating interactions. First, the SH3.1(Nck) domain has to bind to the nonphosphorylated and exposed PRS, that is, the first ITAM tyrosine has to be in the unphosphorylated state. Second, the SH2(Nck) domain has to bind to the second ITAM tyrosine in the phosphorylated state. Likewise, mutations of the SH3.1 and SH2 domains in Nck1 resulted in the loss of Nck1 binding to the TCR. Furthermore, expression of an SH3.1-mutated Nck impaired TCR signaling and T cell activation. Our data suggest that the exact pattern of CD3ε phosphorylation is critical for TCR functioning.

In the last decade an increasing number of plasma membrane (PM) proteins have been shown to be non-randomly distributed but instead forming submicron-sized oligomers called nanoclusters. Nanoclusters exist independently of the ligand-bound state of the receptors and their existence implies a high degree of lateral organisation of the PM and its proteins. The mechanisms that drive receptor nanoclustering are largely unknown. One well-defined example of a transmembrane receptor that forms nanoclusters is the T cell antigen receptor (TCR), a multisubunit protein complex whose nanoclustering influences its activity. Membrane lipids, namely cholesterol and sphingomyelin, have been shown to contribute to TCR nanoclustering. However, the identity of the membrane microdomain in which the TCR resides remains controversial. Using a GFP-labeled TCR we show here that the resting TCR localized in the disordered domain of giant PM vesicles (GPMVs) and PM spheres (PMSs) and that single and nanoclustered TCRs are found in the high-density fractions in sucrose gradients. Both findings are indicative of non-raft localization. We discuss possible mechanisms of TCR nanoclustering in T cells. This article is part of a Special Issue entitled: Nanoscale membrane organisation and signalling.