We have implemented a disease deconstruction pipeline using single cell technologies applied to inflamed tissues in humans to discover new cell types, cell states and pathways. We then use genetically modified mouse models to define mechanisms and therapeutic insights. By shifting in an iterative fashion between humans and mice, our studies unravel the most relevant immune abnormalities. Using single cell RNA-sequencing, CyTOF and flow sorting, we have identified novel T, B, macrophage and stromal cell subsets in RA synovial and systemic lupus erythematosus (SLE) nephritis kidney biopsies. Below are summaries of several key findings.
We carried out mass cytometry (CyTOF) for deep immunophenotyping and performed unbiased clustering using single cell RNA sequencing of synovial tissues in rheumatoid arthritis (RA). We identified a new population of T helper cells that we have named T peripheral helper (Tph) cells, to distinguish them from T follicular helper (Tfh) cells (Rao et al Nature 2017; 542:110-114). Tph cells are found in leukocyte aggregates and tertiary lymphoid structures in chronic inflammatory reactions in peripheral tissues where they drive autoantibody production. Tph cells express high levels of PD1 and ICOS, like Tfh cells, but instead of expressing CXCR5 and BCL6 and localizing in lymph node germinal centers, they express CCR2 and home to peripheral tissues where they secrete CXCL13, the ligand for CXCR5, and drive T-B clusters and provide help to B cells in autoimmune lesions.
We are determining if these pathological T cells drive autoantibody production across autoimmune diseases in general. Targeting Tph cells may provide a new and much more specific option in therapeutic development by targeting only the pathological T cells and avoiding gross blocking of all T cells like current therapeutics.
We are studying the role of stromal fibroblasts in autoimmune disease. In recent years, fibroblastic stromal cells have been found to play key roles in regulating immune responses in lymph nodes and tumors. We have found that in inflamed peripheral tissues like RA, stromal fibroblasts function as the main inflammatory cells that account for production of most of the IL-6 and chemokines that recruit and sustain chronic inflammation.
Pathologically expanded fibroblastic stromal cells in RA: Using key markers, we sorted fibroblast subsets from RA synovium and showed that fibroblast subsets exist that have different transcriptomic signatures and functions (Mizoguchi et al Nat. Commun. 2018; 9:789). Using a combination of CyTOF and single cell and sorted cell subset RNA sequencing, we found that a distinct population of CD90+HLADR+ fibroblasts are markedly expanded in RA.
This population is highly inflammatory and produces nearly all of the IL-6 and most of the chemokines secreted in the synovial tissue in RA. This population is expanded in a disease specific manner and correlates with disease activity. (Mizoguchi et al Nat. Commun. 2018; 9:789; Zhang et al Nat. Immunol 2019 In Press biorxiv http://dx.doi.org/10.1101/351130). We are now examining stromal populations across a series of autoimmune conditions including RA, SLE, and IBD to identify shared and unique stromal cell populations that mediate inflammation and tissue damage.
Current work focuses on the factors that drive fibroblast differentiation and pathologic function. For example, we found that fibroblasts in the synovium display a transcriptional gradient that corresponds to their positional identity within the tissue. Notch3 and Notch ligands drive a major signaling pathway relevant to the fibroblast transcriptional gradient. Graph based clustering and trajectory analysis of more than 25,000 single cells reveal the relationship between anatomically and transcriptionally distinct fibroblasts. Currently, no drugs that target fibroblasts have been approved as medical treatments, but we are identifying the pathways that regulate fibroblast activation and can be targeted therapeutically.
Previously, we found that mesenchymal cadherins, important adhesion molecules in tissue morphogenesis, including cadherin-11 are highly upregulated on activated fibroblasts in tissues and regulate their inflammatory behavior (Chang et al. PNAS 2011; 108:8402-8407). When fibroblast cadherin-11 is deleted or blocked, synovial lining formation is attenuated and the inflammatory response in mouse models of inflammatory arthritis resolves (Lee et al. Science 2007; 315:1006-1010).
In identifying the pathways that activate stromal cell populations in autoimmune diseases, we defined signal specific gene expression modules in fibroblasts that regulate their secretion of IL-6 in chronically inflamed tissues. IL-6 is part of a larger program of inflammatory factors and transcription factor regulators that are co-expressed as an inflammatory module dependent upon an autocrine feedback loop mediated by cell surface IL-6 receptor family members, including IL-6R, oncostatin M receptor and LIF receptor (Nguyen et al. Immunity 2017; 46:220-232). We are now examining the expression of these inflammatory modules in RA and other autoimmune tissues that implicate fibroblasts as key regulators of peripheral tissue inflammation.
I am co-leading a 50 MM NIH-FNIH consortium to carry out large scale single cell analyses on RA and SLE. Patient samples are collected from multiple sites in the US and UK and sent for processing for CyTOF and single cell RNA sequencing carried out in the Single Cell Genomics Core which our lab operates. Data analysis and integration of cell and molecular data with histological and clinical data are performed in an interdisciplinary group of immunologists and computational biologists. New T, B, macrophage and stromal cell types are being identified in tissues from human autoimmune diseases that are not “found it textbooks” and do not correspond to know cell types and states. Thus, we are defining (redefining) pathological cell types and states in autoimmune lesions.
Our laboratory defined the system of CD1 restricted lipid antigen presentation to T cells (Beckman et al. Nature 1994; 372:691-694). CD1 molecules (CD1a, b, c and d) are MHC class I-like proteins that contain hydrophobic channels that bind the acyl chain tails of lipid antigens. This allows lipids to bind to CD1 antigen presenting molecules analogous to peptide binding to MHC. The genes encoding the CD1a, b, c and d molecules represent a distinct lineage of antigen presenting elements that open T cell recognition to the universe of lipid containing self and foreign antigens. These antigens include glycolipids (including sphingolipids, diacylglycerols), lipopeptides and fatty acids that are found in the cell walls of microbes or are self-lipids antigens in mammalian cells. CD1 restricted iNKT cells can secrete Th1, Th2 or Th17 cytokines and mediate host defense against microbial infections or immunopathology.
iNKT cells function as cellular adjuvants transactivating other leukocytes. iNKT cells are early sensors of danger and microbial infection. We have focused on how iNKT cells transactivate DCs which leads to DC instruction and maturation. The activated DC then presents self-lipid antigens that drive even stronger iNKT cell activation in a cycle. (Cohen et al. Cell Host Microbe 2011; 10:437-450) (Brigl et al. J Exp Med. 2011; 208:1163-1177). We are now studying how iNKT cells regulate DC and macrophage transactivation and production of IL-1b and IL-18 via inflammasome and non-inflammasome mechanisms.
Recently, we identified a distinct population of iNKT cells that is highly enriched and function as regulatory T cells in adipose tissue. They provide homeostatic regulation to prevent inflammation in adipose tissue. When inflammation in adipose tissue occurs, it leads to insulin resistance and type 2 diabetes. We found that adipose tissue iNKT cells regulate the regulators. Namely, iNKT cells in adipose tissue are the major source of IL-2 that is required for Treg expansion and function. When iNKT cells are deficient, Treg cell numbers are profoundly reduced in adipose tissue and adipose tissue inflammation, diabetes and obesity worsen.
The important role of innate T cells in adipose tissue is underscored by our finding that they also regulate thermogenesis to set body temperature. When iNKT cells are absent, body temperature and metabolic rate are reduced altering energy utilization. When stimulated, iNKT cells drive the uncoupling of oxidative phosphorylation from ATP generation resulting in heat production and browning of white fat with weight loss and improvement in glucose tolerance (Lynch et al. Nat Immunol. 2015; 16:85-95) (Lynch et al. Cell Metab. 2016; 24:510-519). We are now defining the mechanisms by which iNKT cells regulate inflammation and control thermogenesis in adipose tissue.