Bivalent anti-CD3 Fab dimers drive T cells into a highly efficient fratricide with potential clinical application

Abstract Non-covalent anti-CD3 Fab dimers (Bi-Fabs) bind to the TCR/CD3 complex and activate T cells. In the presence of proper co-stimulation, anti-CD3 Bi-Fabs promote T cell cycling and acquisition of effector functions on CD4 and CD8 T cells more robustly than other bivalent antibody species like F(ab′)2 and mAb with the same specificity. Surprisingly, the net effect of anti-CD3 Bi-Fab treatment on T cells in vitro is contraction of CD4 and CD8 T cells, via AICD. This is in contrast with the net T cell proliferation observed with soluble anti-CD3 mAbs that are stimulatory in the presence of proper co-stimulation. Bi-Fab binding to the TCR/CD3 complex, and direct cellular contact with Bi-Fab engaged T cells, are required to observe the net contraction of treated T cells. Using perforin-deficient T cells, and Fas-FasL blockade, we show this net contraction results from T cell apoptosis via Fas-FasL and CTL degranulation. Structural studies on Bi-Fabs suggest a rigid rod like structure. We hypothesize that the rigid rod allows for a potent crosslinking structure with both paratopes from the anti-CD3 Fabs capable to engage effector T cells on fratricide. In vivo, mice injected with anti-CD3 Bi-Fabs present depletion of T cells from blood, and absence of any adverse effects derived from T cell activation like cytokine storm. We have begun exploring the therapeutic potential of T cell depletion in mouse models of autoimmune disease with promising results.