Expanded fibroblast subset drives pathology in rheumatoid arthritis

The role of fibroblasts in end organ fibrosis is well established, but insights into their roles in chronic
inflammatory diseases in peripheral tissues like rheumatoid arthritis (RA) is still emerging. We identified a
highly expanded inflammatory subpopulation of fibroblasts in the sublining region of RA synovial tissue. It
accounts for >50% of all fibroblasts in the synovium in RA, but it is a rare population in osteoarthritis (OA). The
expanded population is distinguished by high expression of CD90 (Thy1, a sublining marker) and HLA-DR, and
the production of IL-6 and many chemokines. We hypothesize that these CD90+DR+IL-6+ fibroblasts are key in
driving inflammation directly by secreting inflammatory factors and indirectly by recruiting and activating
leukocytes to maintain chronic inflammation. When analyzing single cell RNA-seq data from the RA/SLE
Accelerating Medicines Partnership (AMP) Consortium, we found that markers of lining and sublining
fibroblasts in synovium were not absolute – but instead represented a gradient in gene expression in trajectory
analysis. We found that this transcriptional gradient corresponds to an anatomic spatial gradient in the
synovium emanating from blood vessels. Our data suggest that Notch signaling is a dominant driver of the
gradient starting with fibroblasts around blood vessels and extending to sublining fibroblasts that express
Notch3 receptors and Jagged (Jag)1 Notch ligands.
Here, we wish to determine if Notch 3 signaling specifically on fibroblasts drives the spatial pattering
and the differentiation of sublining fibroblasts. To accomplish this, in Aim 1 we use mixed cell organoids with
endothelial tubules and fibroblasts to compare spatial pattering and differentiation of Notch3 deficient
compared to control fibroblasts. In Aim 2, we determine the location of the CD90+DR+ inflammatory cytokine
producing fibroblasts and Notch3+ fibroblasts in the synovium and determine which fibroblast population(s)
most significantly associate with leukocytes (T cells, B cells and macrophages). In Aim 3 we activate synovial
fibroblast lines with inflammatory cytokines that are found in RA, in the presence or absence of Notch ligands.
We use flow cytometry, RNA-seq, LDA, and trajectory analyses to compare fibroblast cell states induced in
vitro with those found in the synovium in RA. Then, we extend the Notch gradient concept from fibroblast
differentiation to how fibroblast-derived Notch ligands activate attached T cell in organoids. Finally, in Aim 4,
we determine if targeted, conditional disruption of Notch signaling in fibroblasts or targeted conditional deletion
of Notch ligands in fibroblasts prevents inflammatory arthritis in mouse models. Together, these studies will
advance our knowledge of how fibroblasts differentiate in RA to become drivers of inflammation and pathology
in chronically inflamed synovial tissues, and how they might be targeted therapeutically in murine models.

Grant Number: 5R01AR063709-10
NIH Institute/Center: NIH
Principal Investigator: Michael Brenner