Altered TCR Signaling from Geometrically Repatterned Immunological Synapses

See allHide authors and affiliations

Science  18 Nov 2005:
Vol. 310, Issue 5751, pp. 1191-1193
DOI: 10.1126/science.1119238


The immunological synapse is a specialized cell-cell junction that is defined by large-scale spatial patterns of receptors and signaling molecules yet remains largely enigmatic in terms of formation and function. We used supported bilayer membranes and nanometer-scale structures fabricated onto the underlying substrate to impose geometric constraints on immunological synapse formation. Analysis of the resulting alternatively patterned synapses revealed a causal relation between the radial position of T cell receptors (TCRs) and signaling activity, with prolonged signaling from TCR microclusters that had been mechanically trapped in the peripheral regions of the synapse. These results are consistent with a model of the synapse in which spatial translocation of TCRs represents a direct mechanism of signal regulation.

Antigen recognition by T cells requires interactions between TCRs and antigenic peptides bound to major histocompatibility complex molecules (pMHCs) on antigen-presenting cells (APCs) (1). A coordinated recognition process ensues in which surface proteins on the T cell, along with their respective ligands, become organized into micron-scale, spatially patterned motifs known as immunological synapses (ISs) (2, 3). TCRs and pMHCs occupy a region designated the central supramolecular activation cluster (c-SMAC). This is surrounded by a ring of interactions between leukocyte function-associated antigen–1 (LFA-1) on the T cell and intercellular adhesion molecule-1 (ICAM-1) on the APC, termed the peripheral supramolecular activation cluster (p-SMAC). Ultimately, an elaborate collage of adhesion, costimulatory, and signaling molecules, along with cytoskeletal attachments (4, 5) and lipid rafts (68), organizes to sustain signaling over the hours required for full T cell activation. Although previous observations have suggested that the spatial organization of the c-SMAC is directly involved both in increasing and extinguishing TCR signaling (9, 10), a direct causal relation between changes in the synaptic pattern and signaling remains to be established.

We have developed an experimental platform that enables direct manipulation of IS patterns in living T cells. A supported membrane, consisting of a continuous and fluid lipid bilayer coating a silica substrate (11), is used to create an artificial APC surface (12). Inclusion of glycosylphosphatidylinositol (GPI)–linked pMHC and ICAM-1 into the supported membrane is sufficient to enable IS formation between a T cell and the synthetic surface (13). This hybrid live cell–synthetic bilayer IS is illustrated schematically in Fig. 1. Fluidity is a characteristic property of supported bilayers and distinguishes them from solid and polymeric substrates. Movement within the bilayer, however, can be manipulated by fabricating geometrically defined patterns of solid-state structures on the substrate (Fig. 1) (14). We posited that such substrate-imposed constraints might be used to guide molecular motion in the supported bilayer and linked cell-surface receptors to generate alternatively patterned synapses.

Fig. 1.

Diagram of a hybrid live T cell–supported membrane junction. Receptors on the cell surface engage cognate ligands in the supported membrane and become subject to constraints on mobility imposed by physical barriers. The cytoskeleton is represented schematically to reflect the active source of central organization observed in our experiments.

Silica substrates displaying various configurations of chromium lines (100 nm wide and 5 nm high) were fabricated using electron-beam lithography (15). Supported proteolipid membranes were assembled on these substrates by vesicle fusion. As receptors on the T cell surface engage their respective ligands in the bilayer, they become subject to the geometrical restrictions on mobility imposed by the chromium lines. T cell blasts expressing the cloned AND TCR, specific for moth cytochrome c (MCC) 88-103 peptide bound to I-Ek (pMHC), were used in all IS formation experiments. In control experiments with nonactivating peptide, T cells failed to form IS patterns on substrates with chromium lines, confirming specificity of the system (fig. S1). Under conditions of specific antigen recognition, a series of IS patterns, formed under different geometries of constraint patterns, were compared with the unrestricted IS patterns (Fig. 2A). Arrays of parallel lines were found to restrict protein mobility in one dimension and skew the synaptic pattern from a circular to a rectangular shape in which a central band of TCR-pMHC is flanked by two bands of LFA-1–ICAM-1 (Fig. 2B). Multifocal ISs were observed to form in response to grid constraint patterns, which create an array of isolated membrane corrals (Fig. 2C). More elaborate constraint designs, such as a mosaic of concentric hexagonal barriers (Fig. 2D), were also used. A diverse collection of spatially mutated IS patterns were generated to investigate the effects of spatial constraints on synaptic signaling.

Fig. 2.

Synapse formation is altered by geometrical constraints of the substrate. T cells were incubated with fluorescently labeled anti-TCR H57 Fab (green) before being introduced to supported bilayers containing GPI-linked pMHC (unlabeled) and ICAM-1 (red). Chromium lines are visible in brightfield, although they are only 100 nm across, verified by electron microscopy. Images are at 10 min after cells were introduced. IS on unpatterned substrate (A), 2-μm parallel lines (B), 5-μm square grid (C), and concentric hexagonal barriers (spacing 1 μm) (D). TCR distribution (grayscale) on 1-μm square grid (E). Transport map of (E) formed by drawing arrows toward the TCR cluster within the enhanced grid (F).

The chromium barriers also enabled us to provide insight into basic mechanisms of IS formation. For example, a 1-μm grid caused fragmentation of the IS into more than 100 microsynaptic TCR clusters that were stable for more than 30 min (Fig. 2E) despite the rapid TCR-pMHC off rate (∼0.06 s–1) (16). Because TCR motion can only be constrained by the grid through engagement with pMHC, the stability of corralled TCR microclusters indicates that the TCRs in each microcluster move collectively as a multimeric unit. Otherwise, individual TCRs would percolate over the barriers during disengagements from pMHC, and the stable trapping of microclusters would not be observed. The position of each TCR-pMHC microcluster within its corral revealed the direction of transport and could be used to compile a transport map of the IS (Fig. 2F). The microclusters on grids were generally “pulled” to the corner of the corral nearest the center of the IS, and images could be quantified to reveal the high degree of centralized TCR organization in frustrated synapses (fig. S2). Typically, one TCR-pMHC cluster is observed per corral for the 1-, 2-, and 5-μm square grids that were studied, suggesting that TCR clustering occurred only after pMHC engagement. Thus, if TCR were substantially preclustered, one would expect a stochastic distribution of microclusters within the corrals rather than the even distributions we observed on the 1-μm and 2-μm grids. Collectively, this set of observations supports a three-step process by which the mature IS is formed: (i) TCR engagement of pMHC, (ii) TCR-pMHC assembly into microclusters, and (iii) directed transport of microclusters to form the c-SMAC.

Using cytoplasmic distribution of phosphorylated tyrosine (pY) residues associated with TCR clusters, we next measured signaling activity specific to each TCR cluster within constrained synapse motifs (17) (see also fig. S3). At early time points, pY patterns were similar in both native and repatterned synapses (Fig. 3, A and B). However, at 5 min, TCR clusters in the natively pattered IS were observed only in the c-SMAC region and had very low pY levels (Fig. 3C). In contrast, TCR clusters that had been stably restrained to the periphery of the contact area by the substrate grids retained high specific pY levels (Fig. 3D). This effect was restricted to the periphery, because TCR clusters trapped in more central regions of spatially modified synapses lost their pY signal in a time frame similar to those observed in native synapses. The duration of TCR-pY signaling thus correlated with radial position of the TCR rather than with cluster size. Overall, the extent of specific pY associated with TCR clusters above the local background was significantly greater in the IS that had been spatially constrained by the grid (Fig. 3E).

Fig. 3.

TCR-specific phosphotyrosine (pY) signaling in native and repatterned synapses. T cells, which had been incubated with fluorescently labeled anti-TCR H57 Fab, were allowed to interact with pMHC-ICAM membranes for either 2 or 5 min before being fixed and stained for pY. (A) Synapse on unpatterned membrane at 2 min. TCR clusters are distributed, and relatively enhanced pY staining colocalizes with each cluster. The diffuse ring of pY staining in the periphery is likely associated with cortical actin. (B) Synapse on a 2-μm chromium grid at 2 min. (C) Synapse on unpatterned membrane at 5 min. (D) Synapse on a 2-μm chromium grid at 5 min. (E) Statistical results for % TCR colocalization with pY. Black, cells off pattern; gray, cells on 2-μm grids. Results are from three independent experiments at 2 min (a minimum of 9 cells per experiment both on and off patterns; total 31 on, 51 off) and four independent experiments at 5 minutes (a minimum of 7 cells per experiment on and off patterns; total 39 on, 53 off). Data from the 1-min time point (not shown) had extremely high standard deviation because cell population was not well synchronized. (F) Intracellular calcium is elevated in cells on grids. T cells were loaded with the ratiometric calcium-sensitive dye fura-2 and allowed to interact with pMHC-ICAM membranes. Fura-2 fluorescence emission ratio was integrated from 5 min to 20 min in cells on and off 2-μm grids (five independent experiments; total 49 on, 57 off).

Another key measure of signaling activity is the flux of intracellular Ca2+ induced by TCR antigen recognition, which integrates the outputs of all TCR signaling events in the IS (18). The integrated Ca2+ response was significantly higher in cells with spatially constrained IS as compared with those with native synapses (Fig. 3F). Thus, mechanical trapping of TCR in the IS periphery augments early TCR-associated pY levels, as well as the elevation of cytoplasmic Ca2+.

These experiments provide insight into how signaling is extinguished in individual TCR clusters in the IS, which may be attributed to temporal or spatial processes such as recruitment of inhibitors (19) or changes in the actin cytoskeleton that feed back on signaling (20). The hybrid live cell–supported membrane platform made it possible to physically impede receptor translocation to prevent c-SMAC formation, allowing us to resolve that radial location represents a critical parameter in the IS. In physiological terms, it is possible that some APCs may use their own cytoskeletons to restrict transport of pMHC or costimulatory molecules in a related manner. Impeding TCR cluster translocation to the c-SMAC might thus represent a means of augmenting T cell activation (21, 22). Potentially, the ability to induce spatial modifications in model cell-cell interfaces could be useful in exploring spatial organization of membrane domains and proteins on the cell surface, receptor signaling activity, and cytoskeletal regulation processes.

Supporting Online Material

Materials and Methods

SOM Text

Figs. S1 to S4

References and Notes

View Abstract

Navigate This Article