Download files
Abstract
T cell activation is an essential part of the immune response and requires segregation of signalling proteins into specialised membrane domains to transmit and sustain T-cell receptor (TCR) signalling. Engaged TCRs initiate signalling through an extended protein network around the adaptor protein Linker for Activation of T cells (LAT). TCR-LAT networks are critical for membrane segregation, sustained signalling and actin cytoskeleton remodeling although the driving forces and interplay of components are not well understood.
LAT exists in clusters at the cell membrane before and after T cell signalling. LAT clusters have been associated with ordered membrane domains (lipid rafts) and the actin cytoskeleton, raising questions on the nature and influence these processes may have on LAT signalling during T cell activation.
In this study it is demonstrated that the activation of T cells increased membrane order at the T cell activation site, observed using the polarity-sensitive membrane dye Laurdan. The increase in membrane order was independent of LAT cross-linking at a distal site, achieved by specific biotinylation of LAT and adsorption onto streptavidin-beads. LAT clustering induced in this manner also increased local membrane condensation at the bead and was independent of TCR triggering. This suggests that both TCR triggering and LAT cross-linking contribute to membrane re-organisation and the formation of raft domains.
High resolution single molecule imaging of LAT and phosphorylated LAT in the plasma membrane of T cells and HeLa cells was conducted using photo-activatable localisation microscopy (PALM) and direct stochastic optical reconstruction microscopy (dSTORM). Local point pattern analyses were used to quantitatively map LAT clustering.
In HeLa cells, the clustering behaviour of the non-signalling LAT(YAllF) mutant was found to be similar to that of wildtype LAT. The palmitoylation-deficient LAT(C26/29S) mutant formed larger clusters containing more LAT molecules. Conditions of reduced lipid raft abundance (using 7-ketocholesterol) or disruption of the actin cytoskeleton (using Latrunculin B) were also found to increase LAT clustering, and the effect was greatest when both lipid rafts and actin were disrupted, suggesting that both raft association and cortical actin contribute to LAT clustering.
In T cells, LAT clustering was characterised in both quiescent cells and cells activated on antibody-coated glass surfaces. In activated T cells, more LAT molecules were found at TCR activation sites, even when biotinylated LAT was cross-linked to streptavidin-coated beads outside of the activation zone. This suggests that LAT was recruited from an intracellular pool. Cluster analysis revealed that TCR-recruited LAT clusters had an intermediate size. In activated T cells, two distinct LAT clusters co-existed which either did or did not contain phosphorylated LAT. Clusters that were not phosphorylated may have existed prior to T cell activation.
LAT clusters in live cells were observed to grow quickly from foci near or at the cell membrane resembling vesicles docking to the plasma membrane. These clusters remained in situ for up to a minute before disappearing and were also observed occurring within discrete sites at the membrane.
Recruitment of the non-signalling LAT(YAllF) mutant was as efficient as WT LAT whereas the palmitoylation-deficient LAT(C26/29S) mutant was neither recruited nor phosphorylated. This suggests that recruitment precedes, and is independent of LAT phosphorylation.
Together these observations imply that LAT clustering is not solely a consequence of the lateral accumulation of surface-membrane molecules at sites of TCR activation. Instead, the data lead towards a new model of LAT recruitment via sub-synaptic vesicles, adding a new dimension to the assembly and signalling of LAT during early TCR activation.