0127: Single-cell Profiling of Dermal Endothelial Cells Reveals Potential Cell-Cell Interactions in Patients with APS and a History of Cardiac Valve Disease

Background/Purpose: Some of the rarer manifestations of antiphospholipid syndrome (APS), such as cardiac valve disease, remain poorly understood at the molecular level. A previous study profiling skin biopsies from a few patients with marked livedo racemosa preliminarily identified the endothelial cell (EC) Hippo-YAP1-CCN2 axis as a key player in APS vasculopathy. Here, we expanded dermal single-cell profiling to include 30 individuals with primary APS, with the goal of identifying unique tissue signatures associated with some of the more poorly understood features of APS, including cardiac valve disease and nephropathy. In this abstract, we focused on the ligand-receptor pairs predicted to be driving cellular communication in the dermis of patients with a history of cardiac valve disease.

Methods: Single-cell RNA sequencing was used to characterize upper-thigh skin biopsies (epidermis removed) from 30 patients with primary APS (including a subset of eight patients with confirmed valve disease) and 14 healthy controls. Cell-cell interaction analysis was performed using CellPhoneDB to identify enriched (i.e., highly correlated) ligand-receptor pairs between ECs and other potentially interacting cell types. Differential expression and pathway enrichment were evaluated using g:Profiler. Associations with a history of valve disease were tested via logistic regression.

Results: Thirteen cell types were detected within the dermis, with endothelial cells representing 18% of the population in primary APS patients, compared with 16% in controls (p< 0.001). In patients with valve disease (n=8), versus those without (n=22), endothelial cells showed a markedly increased potential for interactions with myeloid cells, smooth muscle cells, and fibroblasts, amongst others (Figure 1). For myeloid cells, highlighted interactions between EC ligands and myeloid cell receptors included CCL23-CCR1, ICAM2-CD209, JAG2-NOTCH1, and DLL1-NOTCH1, with the potential to support inflammatory signaling, leukocyte adhesion, and immune activation. Featured EC interactions with smooth muscle included ANGPT2-TIE2, EDN1-EDNRA, JAG2-NOTCH3, and EFNB1-EPHB6, suggesting increased vascular constriction and remodeling. Notable ligand-receptor interactions between endothelial cells and fibroblasts included PDGFB-PDGFRA, CCL2-CCR2, FN1-ITGA5, and WNT2B-SFRP2, emphasizing fibroblast activation and extracellular matrix (ECM) transformation. Finally, pathway analysis underscored the potential for more leukocyte adhesion, angiogenesis, and ECM remodeling in the tissues of APS patients with a history of valve pathology.

Conclusion: Our findings reveal that endothelial cells of patients with a history of valve disease have significant potential—through their production of chemokines, cell adhesion molecules, and ECM-interacting molecules—to engage in proinflammatory and profibrotic interactions with myeloid cells, smooth muscle cells, and fibroblasts. The identified interactions and pathways provide a roadmap for future research into the valvular pathology that causes significant morbidity for some patients living with APS.