Computational characterization of T cells in inflammatory diseases

Abstract

T cells play critical roles in adaptive immunity. The diversity of T cell receptor (TCR) sequences enables T cells to recognize antigens derived from various pathogens in a specific way. Antigen stimulation can induce the formation of memory T cells, which maintain long-term immunity against previously encountered pathogens. Tissue resident memory T cells (TRMs) reside in peripheral organs without recirculating through the bloodstream. They can rapidly respond to local alarming signals and mediate inflammatory reactions. Especially the recent advent of single cell RNA sequencing (scRNA-seq) and its computational analysis enables the unbiased characterization of complex cell types from different tissues. The combination of scRNA-seq with cell surface protein measurement (CITE-seq) and TCR sequencing (TCR-seq) results in deep insights into T cell identities, clonality, and functions. While recent research characterized many different T cell types and their potential roles in immunity, a deep understanding of the role of T cell subtypes, especially TRMs, in different organs and inflammatory diseases remains elusive. In this thesis, we characterized T cell subtypes and interpreted their functions in distinct organs and diseases using single cell sequencing and computational analysis. In the first project, we profiled T cells from blood and kidney biopsies of both controls and patients with ANCA-associated glomerulonephritis using scRNA-seq combined with CITE-seq. We demonstrated that T helper 17 (Th17) TRMs are present in control and diseased human kidneys. Experimental mouse models showed that bacteria-induced Th17 TRMs exacerbates autoimmune kidney disease. In the second project, we performed scRNA-seq on memory T cells sorted from liver samples from subjects with non-alcoholic fatty liver disease (NAFLD) or non-alcoholic steatohepatitis (NASH). We identified multi-cytokine producing TRMs that are enriched in NASH and contribute to the progression of fibrosis. We also demonstrated the correlation between Bacteroides in the small intestine and hepatic multi-cytokine producing TRMs in NASH patients. In the third project, we profiled immune cell populations in the bronchoalveolar fluid and peripheral blood specimens obtained from patients with severe COVID-19 or bacterial pneumonia. Using simultaneous scRNA-seq, CITE-seq and TCR-seq, we found that a TRM-like Th17 cell population expressing GM-CSF and IL-17A was expanded in bronchoalveolar fluid from COVID-19 patients. Interactome analysis showed that these Th17 cells interact with macrophages and cytotoxic T cells. We also found that high levels of serum GM-CSF and IL-17A correlated with severe COVID-19. Overall, this thesis provides a systematic characterization of T cells in different inflammatory diseases covering different organs. The computational analysis addressed important aspects of the pathogenic role of TRMs in renal autoimmunity, chronic liver inflammation and the current COVID-19 pandemic. Our results suggest a potential link between bacterial/viral infection and autoimmune-like inflammatory responses. We also provide insights that might open novel avenues for therapeutic intervention for inflammatory diseases.