Differential Accumulation and Activation of Monocyte and Dendritic Cell Subsets in Inflamed Synovial Fluid Discriminates Between Juvenile Idiopathic Arthritis and Septic Arthritis

Abstract

Despite their distinct etiology, several lines of evidence suggest that innate immunity plays a pivotal role in both juvenile idiopathic arthritis (JIA) and septic arthritis (SA) pathophysiology. Indeed, monocytes and dendritic cells (DC) are involved in the first line of defense against pathogens and play a critical role in initiating and orchestrating the immune response. The aim of this study was to compare the number and phenotype of monocytes and DCs in peripheral blood (PB) and synovial fluid (SF) from patients with JIA and SA to identify specific cell subsets and activation markers associated with pathophysiological mechanisms and that could be used as biomarkers to discriminate both diseases. The proportion of intermediate and non-classical monocytes in the SF and PB, respectively, were significantly higher in JIA than in SA patients. In contrast the proportion of classical monocytes and their absolute numbers were higher in the SF from SA compared with JIA patients. Higher expression of CD64 on non-classical monocyte was observed in PB from SA compared with JIA patients. In SF, higher expression of CD64 on classical and intermediate monocyte as well as higher CD163 expression on intermediate monocytes was observed in SA compared with JIA patients. Moreover, whereas the number of conventional (cDC), plasmacytoid (pDC) and inflammatory (infDC) DCs was comparable between groups in PB, the number of CD141+ cDCs and CD123+ pDCs in the SF was significantly higher in JIA than in SA patients. CD14+ infDCs represented the major DC subset in the SF of both groups with potent activation assessed by high expression of HLA-DR and CD86 and significant up-regulation of HLA-DR expression in SA compared with JIA patients. Finally, higher activation of SF DC subsets was monitored in SA compared with JIA with significant up-regulation of CD86 and PDL2 expression on several DC subsets. Our results show the differential accumulation and activation of innate immune cells between septic and inflammatory arthritis. They strongly indicate that the relative high numbers of CD141+ cDC and CD123+ pDCs in SF are specific for JIA while the over-activation of DC and monocyte subsets is specific for SA.

Keywords: CD123 pDC; CD141 cDC; dendritic cells; juvenile idiopathic arthritis; monocytes; multiparametric flow cytometry; septic arthritis.

Copyright © 2020 Cren, Nziza, Carbasse, Mahe, Dufourcq-Lopez, Delpont, Chevassus, Khalil, Mura, Duroux-Richard, Apparailly, Jeziorski and Louis-Plence.

Figures

Figure 1

Representative staining of monocyte subsets. Representative dot plots of the classical (CD14++CD16−, red), intermediate (CD14++CD16+, blue) and non-classical (CD14+CD16++, green) monocytes and representative histograms for the various markers (CCR2, CD163, CD64, HLA-DR, and SLAN) on each subset. (A) in peripheral blood (B) in synovial fluid of patients with JIA (left) or SA (right). Black histograms correspond to negative control MFI.

Figure 2

Representative staining of DC subsets. Representative dot plot of the CD1c+ cDC (1, red), CD14+ InfDCs (2, blue), CD141+ cDC (3, orange) and CD123+ pDC(4, green) and representative histograms for the various markers (PDL2, CD86, HLA-DR) on each subset (A) in peripheral blood and (B) in synovial fluid of patients with JIA (left) or SA (right). Black histograms correspond to negative control MFI.

Figure 3

Stratification of patients with JIA and SA based on monocyte proportion and all cell counts. (A) Heatmap of monocyte proportion and all cell counts in peripheral blood (PB) and synovial fluid (SF) by multi-parametric flow cytometry. In the color scale, yellow indicates relative higher and blue relative lower cell numbers/proportion. Gray data are undetermined. The dendrogram shows the results of the non-supervised hierarchical clustering analysis; in the heatmap: rows, cell subsets; and columns, individual patients with JIA (gray, n = 15) and SA (black, n = 14). (B) Heatmap showing markers that were significantly differentially expressed between patients with SA and JIA. The dendrogram shows the results of the non-supervised hierarchical clustering analysis using an FDR cut-off of 5%. (C) Direct comparison of the 5 differentially expressed cells counts in synovial fluid (SF) and non-classical monocyte proportion in peripheral blood (PB) from JIA (white, n = 15) and SA (gray, n = 14) patients. Data are represented for all patients and horizontal bars are the mean ± SEM. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001 (two-sided t or Mann–Whitney test, according to data distribution after FDR adjustment).

Figure 4

Stratification of patients with JIA and SA based on DC and monocyte activation markers. (A) Heatmap showing the mean fluorescence intensities (MFI, rows) of activation markers that were differentially expressed in JIA (gray, n = 15) and SA (black, n = 14) patients (column). The dendrogram shows the results of the non-supervised hierarchical clustering analysis using an FDR cut-off of 5%. Yellow indicates relative higher expression, blue relative lower expression of the markers and gray undetermined data. Direct comparison of the expression of (B) CD64 and (C) CD163 in monocyte subsets and (D) CD86, (E) PDL2, and (F) HLA-DR in DC subsets in blood (PB) and synovial fluid (SF) samples of patients with JIA (white, n = 15) and SA (gray, n = 14); Data are represented for all patients and horizontal bars are mean ± SEM. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.001 (two-sided t or Mann–Whitney test, according to data distribution after FDR adjustment).

Figure 5

Stratification of patients with JIA and SA based on all parameters. (A) Heatmap showing the markers (markers, rows) that were differentially represented in JIA (gray, n = 15) and SA (black, n = 14) patients (column). The dendrogram shows the results of the non-supervised hierarchical clustering analysis using an FDR cut-off of 5%. Yellow indicates relative higher expression, blue relative lower expression of the markers. (B) Receiver operating curves (ROC) for markers with AUC values > 0.94.