The ability of T cells to engage functionally distinct ligands throughout development and peripheral surveillance has been attributed to TCR cross-reactivity. Yet, it is still unclear to what extent TCRs really are cross-reactive, or how cross-reactivity impacts signaling. These questions are important to understanding the mechanisms for, and consequences of, TCR ligand discrimination in normal and pathogenic immunity. Here we wish to apply a powerful methodology, peptide-MHC yeast display coupled with deep sequencing, to explore the interplay between TCR cross-reactivity, structure, and function by asking: 1-How cross-reactive are TCRs? Our peptide-MHC libraries enable, for the first time, an experimental determination of TCR/pMHC cross-reactivity that can gauge the extent of cross-reactivity between different TCRs to a common pMHC, autoimmune TCRs, and TCRs isolated from different types of T cells. 2-Do TCR/pMHC cross-reactivity and binding topology play instructive roles in TCR signaling? Most 'alternative'agonist peptides characterized in the literature are simply cognate ligands with conservative substitutions that are recognized in similar structural modes by a TCR, leading to the common assumption that TCR signaling is structurally indiscriminate. In contrast, we found that a non-homologous peptide discovered by pMHC yeast display was recognized with high affinity by a TCR in an unusual docking geometry that failed to activate intracellular signaling. This was the first demonstration that the docking mode of the TCR/pMHC complex is an important variable in TCR signaling. We wish to identify peptides that are recognized by several TCRs in diverse structural modes, and ask how the biophysical parameters of TCR cross-reactivity modulate downstream signaling in mouse and human systems where state-of-the-art tools exist to interrogate mechanism. 3-Do TCR/pMHC cross-reactivity and docking geometry influence human autoimmune disease? TCR cross-reactivity, molecular mimicry, and several aberrant structural features of self-antigen recognition have been proposed to play roles in development of autoimmune disease. Indeed, several MBP-reactive TCRs derived from patients with Multiple Sclerosis assume a very unusual binding mode with MBP presented by HLA-DR15. We wish to test these hypotheses by probing the cross-reactive properties of MBP-reactive human TCRs, and identifying peptides that are recognized in different binding modes than the MBP autoantigen. 4-What is the molecular basis of non- signaling TCR/pMHC interactions in natural human immunity? Unexpectedly, we recently found that a high percentage of T cell clones isolated from human blood using HIV-DR4 tetramers do not signal in response to these peptide antigens, apparently engaged in 'non-productive', but specific, TCR/pMHC interactions in vivo. We wish to determine the structural basis for how TCR binding is decoupled from signaling by these peptides, and isolate cognate endogenous agonist peptides. Collectively, we propose to both interrogate and manipulate basic aspects of TCR cross-reactivity in a manner that is directly germane to human translational immunology.
T cells provide critical defense against infectious pathogens and cancers but, conversely, they can also react against self antigens in autoimmune diseases. We are utilizing a variety of interfacial approaches from protein engineering, biophysics and cell biology to visualize the structural logic by which T cell receptors translate the recognition of self and foreign peptide-Major Histocompatibility Complex proteins into distinct signals, and to ask if we can manipulate, or tune this signal through engineering of the peptide ligands. Our hope is that this approach can be leveraged for the betterment of human health through identification of new ligands and the design of immunotherapeutic drugs.
|Birnbaum, Michael E; Mendoza, Juan L; Sethi, Dhruv K et al. (2014) Deconstructing the peptide-MHC specificity of T cell recognition. Cell 157:1073-87|