Myung-Chul Kim1, Umasankar De2, Nicholas Borcherding3, Joon Paek4, Nicole Winn5, Indraneel Bhattacharyya6, Qing Yu7, Ryan Kolb8, Theodore Drashansky9, Akaluck Thatayatikom10, Weizhou Zhang8 and Seunghee Cha11, 1Jeju National University, Jeju-si, South Korea, 2University of Florida, Gainesville, FL, 3Washington University School of Medicine in St. Louis, Washington, DC, 4Department of Oral & Maxillofacial Diagnostic Sciences, Gainesville, FL, 5Division of Oral Medicine, Department of Oral and Maxillofacial Diagnostic Sciences, University of Florida College of Dentistry, Gainesville, FL, 6Department of Oral & Maxillofacial Diagnostic Sciences, University of Florida College of Dentistry, Gainesville, FL, 7The Forsyth Institute, Cambridge, MA, 8Department of Pathology & Immunology, University of Florida, Gainesville, FL, 9Cellularity, Inc, Florham Park, NJ, 10AdventHealth for Children, Orlando, FL, Orlando, FL, 11Department of Oral & Maxillofacial Diagnostic Sciences, University of Florida College of Dentistry and Center for Orphaned Autoimmune Disorders (COAD), Gainesville, FL
Background/Purpose: This study was performed to characterize key immune cell subsets and their interactions in patients with childhood Sjögren's disease (cSjD) for the first time in the field. Unlike Sjögren's disease (SjD), diagnostic criteria, treatment protocols, and natural history are unavailable in cSjD. In the current study, we aimed to address the following questions: 1) What distinguishes cSjD patients from healthy individuals?, 2) What distinguishes cSjD from symptomatic non-cSjD?, 3) Is cSjD a distinct disease entity from adult SjD?, 4) What involves biopsy positivity in non-cSjD?, and 5) Most importantly, what are the transcriptome and interactome profiles contributing to recurrent parotitis (RP), a chief complaint that prompts the most clinical visits from young children with cSjD?
Methods: Five groups of four PBMC samples from the University of Florida (UF) pediatric cohort were profiled by single-cell RNA-seq (scRNA-seq). The groups include: cSjD, non-cSjD, biopsy-positive non-cSjD (Bx), biopsy-positive non-cSjD with RP (BxRP), and healthy controls (HC) (Figure 1). cSjD patients were diagnosed based on the 2016 ACR/EULAR criteria for primary SjD. The CellChat R package identified cell-to-cell interaction strength and frequency. Flow cytometry for immune subsets and regulatory T cell (Treg) functional assays for Treg suppression were also performed.
Results: Prominent effector memory CD4+ T cells (Tem) noted in cSjD were characterized by genes involving inflammation and cytolysis. A unique cluster of proinflammatory CD14+ monocytes enriched with type I and type II IFN-related genes was highly specific to cSjD compared with the published scRNA-seq dataset from adult SjD (Figure 2). Surprisingly, in vitro Treg function in cSjD was intact and distinct from reduced Treg functionality in SjD. The X-C-motif-chemokine-ligand-1 (XCL1)+ NK subset was expanded significantly, exhibiting functional enhancement in NK-mediated cytotoxicity and transmigration. Subjects with RP had more M1-like CD14+ monocytes, Tem, and XCL1+ NK cells. Importantly, effector CD8+ T and XCL1+ NK cells were analyzed to be activated by dendritic cells or memory B cells, based on our interactome analyses (Figure 3). Significant upregulation of CCL3 was noted in NK cells from patients in the RP group, compared to those not in the RP group.
Conclusion: Our study, based on scRNA-seq analysis of PBMC from our UF cohort, revealed that: 1) Enhanced immune cell interactions and inflammatory monocyte clusters enriched with IFN-related genes are the key features in cSjD. 2) M2-polarized monocytes distinguish non-cSjD from cSjD. 3) Tregs in cSjD are functionally competent, whereas the regulatory function of Tregs in adult SjD are suppressed. 4) Biopsy-positive non-cSjD contains proinflammatory M1-polarized monocytes. 5) Patients with RP present inflammatory immune subsets and enhanced interactions similar to cSjD. A subset of more activated NK cell subset might also contribute to RP through preferential recruitment of effector types of T cells, including MAIT-like CD8+ T cells, by CCL3-producing NK cells.
Overall study scheme. (A) Single cell RNA sequencing applied to peripheral blood mononuclear cells from 5 groups of interest is present. (B) Study design, groups, and analyses that will be covered by this study.
Identification and characterization of proinflammatory myeloid subpopulations by scRNA-seq. (A) Six myeloid subsets are presented on the tSNE plot. (B) Among the top 20 DEGs defining each subset, representative genes are presented on the heatmap with Cluster 4 characteristically distinguished from others by the distinct upregulation of IFN-related genes. (C) Hex density enrichment plot reveals enrichment pattern of proinflammatory and inflammatory responses and type I and II IFN signatures in Cluster 4 of each group. Red and blue numbers on each quadrant illustrate upward and downward trends, respectively, compared to HC.
Molecular and interactome characterization of immune subsets involved in recurrent parotitis. (A) Hex density enrichment plot revealing enrichment pattern of "M1 macrophage" and "leukocyte-mediated immunity" in the classic monocyte subsets across the groups. RP exhibits a more inflammatory phenotype than those without, as evidenced by the RP-preferential shift toward the inflammatory signatures. The numbers in red and blue on each quadrant illustrate upward and downward trends, respectively, compared to HC. (B) Dot plot presenting RP-related upregulation of CCL3 and TRDC genes in NK subsets. Statistical significance was obtained by comparing two groups of interest with a non-parametric Wilcoxon rank-sum t-test. * p < 0.05, ** p < 0.01, and *** p< 0.001. (C) Identification of significant ligand and receptor pairs between immune subsets of interest, including effector CD8+ T and XCL1+ NK subsets. These subsets have shown increased communication probability with B and myeloid subsets via MHC class I-related genes in cSjD RP, compared to the groups without RP.
M. Kim: None; U. De: None; N. Borcherding: Omniscope, 2; J. Paek: None; N. Winn: None; I. Bhattacharyya: None; Q. Yu: None; R. Kolb: None; T. Drashansky: Celularity, Inc., 3, 10, 11; A. Thatayatikom: None; W. Zhang: None; S. Cha: None.
.jpg)
.jpg)
.jpg)
