Molecular characterization of T cells across disease states in the central nervous system

Abstract

The local cytokine milieu shapes the nature and function of immune cells and by extension the overall course of tissue-specific immune responses. In the context of cancer and autoimmunity, two opposing immune responses, the local immune environment can lead to dysfunctional T cell states. Understanding the mechanisms that distinguish these two states is key to identifying unique pathways that could be induced in autoimmunity and relieved without causing autoimmunity in cancer. We sought to characterize unique and shared T cell states in the context of multiple sclerosis (MS) and glioblastoma (GBM) using a combination of immunophenotyping and functional approaches, including ultra low-input transcriptional profiling. First we use a series of unbiased approaches to identify functional differences between auto-reactive T cells derived from MS or healthy donors.

We found that MS-derived CCR6⁺ T cells produce pathogenic inflammatory cytokines IFN-[upsilon], IL-17, and GM-CSF, while healthy controls produce IL-10 in response to myelin antigen. We also identified a transcriptional signature that characterizes these cells from MS donors and highlights the role of CTLA-4 signaling in autoreactive T cells derived from healthy donors. Next, we sought to better understand the role of another co-inhibitory receptor, PD-1, specifically in the context of Treg and CD4⁺ T effector function in GBM, a central nervous system cancer. We identify unique signatures of dysfunction in both the CD4⁺ T cell and Treg compartment correlated with PD-1 expression. Finally, adverse development of autoimmunity in the context of treatment with blocking CTLA-4 antibodies suggests that studies directly comparing T cells in the context of autoimmunity and cancer can yield valuable insight into mechanisms of T cell dysfunction.

Through a transcriptional comparison of isolated T cell subsets from the spinal fluid of MS patients, tumors of GBM patients, and matched blood we identify common biological mechanisms of CNS T cells and identify unique signatures of CNS exhaustion. Taken together, our data suggests that the CNS may be enriched for pathogenic cells in the context of both MS and GBM.