Membrane-mediated cooperativity of antigen receptors and CD2 adhesion receptors in CAR-T cell activation

Abstract

Chimeric antigen receptor (CAR)-T cells exhibit a major defect in antigen recognition, often requiring 100-1000-fold higher antigen densities for activation than the native T cell receptor (TCR). Recent work has traced this defect to the inefficient use of CD2/CD58 adhesion by CAR-T cells and demonstrated that matching the extracellular dimensions of CD2/CD58 and CAR/antigen complexes can dramatically enhance antigen sensitivity. Remarkably, modest nanometre-scale changes in CD2/CD58 dimensions produce orders-of-magnitude changes in antigen sensitivity, while having little effect on T cell activation at high antigen densities. Here, we present a general model for the length-dependent cooperativity of antigen receptors and CD2 adhesion receptors in CAR-T cell activation, combining statistical-physical theory with coarse-grained simulations. The model predicts that the antigen concentration required for activation increases sharply with the length mismatch between antigen-receptor and adhesion-receptor complexes until a critical mismatch is reached, at which the complexes segregate. Beyond this critical mismatch when complexes segregate, antigen sensitivity becomes largely independent of complex length, consistent with experimental observations. For small to moderate mismatches, the antigen concentration required for activation exhibits a Gaussian dependence on length mismatch with a characteristic width of only a few nanometres, explaining how nanometre-scale changes in CD2 length can alter antigen sensitivity by orders of magnitude. Fitting the model to measurements of CD2-length-dependent CAR-T cell activation further enables estimation of characteristic molecular length scales and concentrations of antigen-receptor and adhesion-receptor complexes. These results provide a quantitative physical framework connecting nanometre-scale organization of the cell-cell interface to CAR-T cell antigen sensitivity and highlight molecular dimensions as a design parameter for T cell-based immunotherapies.