RNA Identification of PRIME Cells Predicting Rheumatoid Arthritis Flares

Abstract

Background: Rheumatoid arthritis, like many inflammatory diseases, is characterized by episodes of quiescence and exacerbation (flares). The molecular events leading to flares are unknown.

Methods: We established a clinical and technical protocol for repeated home collection of blood in patients with rheumatoid arthritis to allow for longitudinal RNA sequencing (RNA-seq). Specimens were obtained from 364 time points during eight flares over a period of 4 years in our index patient, as well as from 235 time points during flares in three additional patients. We identified transcripts that were differentially expressed before flares and compared these with data from synovial single-cell RNA-seq. Flow cytometry and sorted-blood-cell RNA-seq in additional patients were used to validate the findings.

Results: Consistent changes were observed in blood transcriptional profiles 1 to 2 weeks before a rheumatoid arthritis flare. B-cell activation was followed by expansion of circulating CD45-CD31-PDPN+ preinflammatory mesenchymal, or PRIME, cells in the blood from patients with rheumatoid arthritis; these cells shared features of inflammatory synovial fibroblasts. Levels of circulating PRIME cells decreased during flares in all 4 patients, and flow cytometry and sorted-cell RNA-seq confirmed the presence of PRIME cells in 19 additional patients with rheumatoid arthritis.

Conclusions: Longitudinal genomic analysis of rheumatoid arthritis flares revealed PRIME cells in the blood during the period before a flare and suggested a model in which these cells become activated by B cells in the weeks before a flare and subsequently migrate out of the blood into the synovium. (Funded by the National Institutes of Health and others.).

Copyright © 2020 Massachusetts Medical Society.

Figures

Figure 1.. Study Overview and Validation of In-Home Assessments of Disease Activity and Gene Expression.

Panel A shows an overview of the collection of clinical data and specimens over time. The Routine Assessment of Patient Index Data 3 (RAPID3) questionnaire and the Disease Activity Score 28 (DAS28) were used to assess disease activity weekly at home and monthly in the clinic, respectively. Green indicates the time leading up to a flare, and purple the time during a flare. Panel B shows a scatterplot (with locally weighted smoothing) of the relationship between the change in scores on the RAPID3 questionnaire and the DAS28 in the index patient. DAS28 scores range from 2 to 10, with higher scores indicating more disease activity. RAPID3 scores range from 0 to 30, with higher scores indicating more severe disease. The blue line represents the point estimates, and the shaded area represents the 95% confidence interval. Panel C shows neutrophil, lymphocyte, and monocyte counts in paired clinical complete blood counts conducted with blood drawn by venipuncture as compared with the CIBERSORTx-inferred blood counts from RNA sequencing (RNA-seq) data obtained with the use of fingerstick blood specimens (38 paired specimens). The shaded areas represent the 95% confidence intervals.

Figure 2.. Clinical and Transcriptional Characteristics of Rheumatoid Arthritis Flares in the Index Patient.

Panel A shows disease activity (measured with the RAPID3 questionnaire; 356 scores included in the analysis) over the course of 4 years in the index patient. Panel B shows a volcano plot of differential gene expression during flares (46 specimens) and during baseline (33 specimens), with significance (−log10 false discovery rate [FDR]) plotted against the log2 relative expression (flare:baseline ratio). Gray points indicate genes with no significant difference in expression between flares and baseline (with FDR >0.1), red indicates genes with significantly increased expression during a flare (FDR <0.1 and log2 expression ratio >0), and blue indicates genes with significantly decreased expression during a flare (FDR <0.1 and log2 expression ratio <0). Panels C and D show pathways enriched among genes with significantly increased (Panel C) or decreased (Panel D) expression during a flare relative to baseline. The dashed line represents the threshold for significance (FDR <0.05, or −log10 FDR >1.3). NLS denotes nuclear localization signal.

Figure 3.. Transcriptional Characteristics of Immune Activation before Symptom Onset in Rheumatoid Arthritis Flares.

Panel A shows RAPID3 disease-activity scores over time (measured in days). The box represents disease activity from day −56 to day 28 relative to the start of a flare (day 0). Panel B shows hierarchical clustering of z scores of 2791 significantly differentially expressed genes over time. Significant clusters are labeled by color. Antecedent cluster 2 (AC2) and AC3 refer to clusters of genes with expression that changed before flares. Panel C shows a detailed representation of the cluster 1, AC2, and AC3 genes depicted in Panel B over the time to a flare. Panel D shows the mean standardized cluster gene expression over the time to a flare. Light gray lines represent expression of individual genes in the cluster. The dashed horizontal line represents the mean baseline gene expression (weeks −8 to −4). Panel E shows pathways enriched in cluster 1, AC2, and AC3. The dashed vertical line represents the threshold for significance (FDR 10 FDR >1.3).

Figure 4.. Expression of AC3 Genes in PRIME Cells.

Panel A shows synovial cell subtype marker genes in clusters identified in blood. Enrichment scores of 200 single-cell RNA-seq (scRNA-seq) marker genes from 18 synovial subset cell types are shown. The dashed vertical line represents the threshold for significance (FDR 10 FDR >1.3). ABC denotes age-associated B cell, CTL cytotoxic T lymphocyte, IFN interferon, Tph/fh T peripheral helper/follicular helper, Treg T regulatory. Panel B shows the means of the standardized expression of genes that are common to synovial sublining fibroblasts (CD34+DKK+HLA−DRA+ fibroblasts) and AC3 in blood over the time to a flare (the dashed vertical line represents the start of a flare); I bars represent 95% confidence intervals. For standardization, the mean expression level for each gene was calculated across flares per week and then normalized across weeks. Panel C shows a Venn diagram of the numbers of AC3 genes that decreased during flares in the index patient (Patient 1) and the three additional patients (Patients 2, 3, and 4). Shading is used to highlight overlap between the index patient and the other patients. Panel D shows the log2 relative expression ratio for AC3 genes in PRIME cells (flow-sorted CD45−CD31−PDPN+ cells) as compared with hematopoietic cells (flow-sorted CD45+) and in stained peripheral blood mononuclear cells (not flow sorted) as compared with hematopoietic cells (flow-sorted CD45+) as a technical control for the stress of flow sorting. The box-and-whisker plots indicate the median, interquartile range, and 1.5 times the interquartile range; the P value is from a Mann–Whitney U test.

Figure 5.. Model of Blood and Synovial Gene-Expression Changes before and during Rheumatoid Arthritis Flares.

Inflammatory signals activate naive B cells (AC2), which in turn activate PRIME cells (AC3); these cells harbor the signature of synovial sublining fibroblast genes. According to this model, PRIME cells demarginate and are increased in blood before a flare and then decrease just after symptom onset; these cells or their progeny are increased in inflammatory synovium in patients with rheumatoid arthritis, where they contribute to and may be sufficient to cause joint inflammation.