Background/Purpose: Anti-SSA antibody positive (SSA+ [Ro52 or Ro60]) and negative (SSA-) Sjögren disease (SjD) have differing clinical phenotypes and prognostic features; however, the pathogenesis driving these two endotypes remains unclear. We define a novel pathogenic SjD cell, fibroblasts (fbs), and shed new light on traditional pathogenic processes (Ro52/60) using integrated spatial sequencing techniques. Our objective is to define the salivary gland (SG) fibroblasts (fb) cellular neighborhoods and interactions driving pathology in SSA+ compared to SSA- SjD with a focus on Ro52/60

Methods: We performed spatial analyses with Xenium and CosMx of human SG tissue derived from 2 SSA+ and 2 SSA- patients fulfilling 2016 ACR/EULAR criteria for SjD. After data integration, we labeled cell types and proceeded to neighborhood analyses using graph attention neural networks. We performed immunofluorescence for Ro52 and Ro60 in SSA+ and SSA- SjD SG (n=9).

Results: Our data integration and analyses using Xenium and CosMX yielded nine major cell types. Next, we performed neighborhood analyses, comparing SSA+ and SSA- SjD neighborhoods and found SSA+ fbs interact most with ductal epithelial and T cells. In contrast, SSA- fbs interact with ductal epithelial cells alone. We compared neighborhoods that predominate in SSA+ or SSA- SjD SGs and that had similar cell subset composition. For example, neighborhood clusters 1 and 7 show plasma cells and seromucous acini neighboring with fbs; yet cluster 1 is found nearly exclusively in SSA+ samples and cluster 7 is found in SSA- SjD SGs, respectively. In the SSA+ predominant cluster 1, fb transcripts include JAK1, IRF4, and STAT6, whereas the SSA- cluster 7 fbs have important transcripts that include CXCL12/14 and IFITM3. Analyses of top differential transcripts show SSA+ fbs expressed CSF1R and CD40, whereas SSA- fbs express CCL28, IL10, and IFI6. We used seqFISH to orthogonally identify 12 cell subsets in SSA+ and SSA- SjD SGs. We labeled our glands with Ro52 and Ro60. Ro52 expression is lower in SSA- compared to SSA+ SjD cell subsets except in T-cells, ki67 high B-cells, and follicular dendritic cells. In contrast, Ro60 expression is lower in SSA- compared to SSA+ SjD cell subsets except for basal and luminal ductal cells. Both Ro52 and Ro60 are most differentially expressed by serostatus in myoepithelial cells. SSA+ non-lymphocytes express more Ro52 than lymphocytes. This pattern is reversed in SSA- SGs. Akin to Ro52, SSA+ non-lymphocytes express more Ro60 than lymphocytes. In contrast to Ro52, SSA- non-lymphocytes express more Ro60 than lymphocytes.

Conclusion: By integrating multiple spatial profiling techniques, we identify unique fb subsets driving cellular interactions in SSA+ and SSA- SjD. Ro52/Ro60 have long been hypothesized to be central to SjD pathogenesis. For the first time, we determine cell-specific Ro52/60 expression in SSA+ compared to SSA- SGs. Ro52/Ro60 expression differs by cell subsets in SSA+ and SSA- SjD, with myoepithelial cells showing the most divergent Ro52/Ro60 expression by serostatus, a potential nidus for autoantibody formation. These findings provide new pathogenic insight into SjD fb by serostatus and gain new perspectives on Ro52 and Ro60 in a cell-specific manner.

« Back to ACR Convergence 2025

ACR Meeting Abstracts - https://acrabstracts.org/abstract/integration-of-multiple-spatial-transcriptomics-reveals-novel-insights-into-sjogren-disease-salivary-gland-fibroblast-by-peripheral-serostatus/