Arthritis Research & Therapy (Mar 2022)

Single-cell transcriptomics of popliteal lymphatic vessels and peripheral veins reveals altered lymphatic muscle and immune cell populations in the TNF-Tg arthritis model

  • H. Mark Kenney,
  • Chia-Lung Wu,
  • Alayna E. Loiselle,
  • Lianping Xing,
  • Christopher T. Ritchlin,
  • Edward M. Schwarz

DOI
https://doi.org/10.1186/s13075-022-02730-z
Journal volume & issue
Vol. 24, no. 1
pp. 1 – 20

Abstract

Read online

Abstract Background Lymphatic dysfunction exists in tumor necrosis factor transgenic (TNF-Tg) mice and rheumatoid arthritis (RA) patients. While joint-draining TNF-Tg popliteal lymphatic vessels (PLVs) have deficits in contractility during end-stage arthritis, the nature of lymphatic muscle cells (LMCs) and their TNF-altered transcriptome remain unknown. Thus, we performed single-cell RNA-sequencing (scRNAseq) on TNF-Tg LMCs in PLVs efferent to inflamed joints versus wild-type (WT) controls. Methods Single-cell suspensions of PLVs were sorted for smooth muscle cells (SMCs), which was validated by Cspg4-Cre;tdTomato reporter gene expression. Single-cell RNA-seq was performed on a 10x Genomics platform and analyzed using the Seurat R package. Uniform Manifold Approximation and Projections (UMAPs) and Ingenuity Pathway Analysis software were used to assess cell clusters and functional genomics in WT vs. TNF-Tg populations. Results Fluorescent imaging of Cspg4-Cre;tdTomato vessels demonstrated dim PLVs and strong reporter gene expression in the adjacent superficial saphenous vein, which was corroborated by flow cytometry of LMCs and vascular smooth muscle cells (VSMCs) from these vessels. Due to their unique morphology, these populations could also be readily detected by scatter analysis of cells from non-fluorescent mice. Bioinformatics analysis of flow sorted WT and TNF-Tg cells identified 20 unique cell clusters that together were 22.4% LMCs, 15.0% VSMCs, and 62.6% non-muscle cells of 8879 total cells. LMCs and M2-macrophages were decreased, while inflammatory monocytes were increased in TNF-Tg lower limb vasculature. SMC populations were defined by Cald1, Tpm1, and Pdgfrb expression and were enriched in myofibroblast-like gene expression. TNF-Tg LMCs exhibited enhanced functional genomics associated with cell death, phagocyte recruitment, and joint inflammation. Among the most prominent TNF-induced genes in SMCs were Mmp3, Cxcl12, and Ccl19, and the most downregulated genes were Zbtb16, Galnt15, and Apod. Conclusions Single-cell RNA-seq can be used to investigate functional genomics of lower limb vasculature in mice. Our findings confirm the inflammatory transcriptome of TNF-Tg vessels and altered gene expression in SMC populations. This study further supports a potential role of mesenchymal stromal cells in inflammatory-erosive arthritis pathogenesis, and warrants future studies to define the effects of this TNF-altered transcriptome on PLV function and joint homeostasis.

Keywords