Background Adult-onset Still's disease (AOSD) and Behçet's disease (BD) are both systemic inflammatory diseases, the causes of which are largely unknown. They have been recently classified as autoinflammatory diseases, a group of diseases in which innate rather than acquired immunity plays important roles in their pathogenesis. As AOSD and BD are clinically distinct diseases, their cytokine networks should also be different.
Objectives In this study, we attempted to quantify the levels of multiple cytokines in the serum of patients by utilizing a beads-array technique and ELISA, and then compared the serum cytokine profiles of the two diseases by factor analysis. We then sought to clarify the hierarchical relationship between interleukin (IL-) 17A and interferon (IFN-) α using peripheral blood mononuclear cells (PBMCs).
Methods We quantified the serum levels of 10 cytokines (IFN-α, IFN-γ, IL-1β, IL-2, IL-4, IL-6, IL-10, IL-12p70, IL-17A and tumor necrosis factor α) in 16 AOSD patients and 28 BD patients using multiplex bead array assays and IL-18 along with ELISA. The data were then subjected to factor analysis. We next stimulated PBMCs from three healthy volunteers in vitro with class A CpG oligodeoxynucleotides (ODNs) in the presence or absence of IL-17A for 15 hours. We performed flowcytometric analysis to examine the expression of intracellular IFN-α in plasmacytoid dendritic cells (pDCs).
Results Two factors were extracted from the factor analysis using the data on 8 cytokines that were detectable in the serum of the patients. IL-17A and IFN-α, the levels of which showed a strong positive correlation in the serum of BD patients (Fig. A), were the main components of Factor 1, while Factor 2 consisted of IL-6, IL-10 and IL-18 (Fig. B). Patients were also plotted on a plane determined by Factors 1 and 2 according to each patient's factor scores. Those who were high in Factor 1 but low in Factor 2 were likely to be BD patients and vice versa, many of those who were high in Factor 2 but low in Factor 1 were AOSD patients. In terms of flowcytometric analysis, IL-17A alone did not induce IFN-α expression in pDCs, but it did substantially increase IFN-α-positive pDCs induced by CpG ODNs.
Conclusions The cytokines examined were clearly separated into distinct groups by the factor analysis. Similarly, the AOSD and BD patients could be separated, although roughly. High levels of serum IL-6, -10 and -18 suggest AOSD while high levels of IFN-α and IL-17A indicate BD. To establish patterns of correlation among cytokines, it is important to focus on the cytokine concentrations in each patient, rather than the average cytokine concentrations in each of the diseases. In terms of the hierarchical relationship between IFN-α and IL-17A, a previous report suggested that IFN-α blocks IL-17A production in PBMCs from BD patients. Thus, we examined the effect of IL-17A on IFN-α production. It should be noted that the real stimulus for IFN-α release in BD is unknown. In addition, cells other than pDCs may be involved in the production of IFN-α. Nevertheless, understanding the hierarchical relationship among cytokines should prove to be helpful in clarifying the pathogenesis of various inflammatory diseases.
Disclosure of Interest None declared