Juvenile idiopathic arthritis (JIA) is a heterogeneous group of arthropathies of unknown etiology. Both genetic and environmental factors are believed to play a role in susceptibility to JIA. One way to improve the understanding of the etiopathogenesis of JIA is to define better biological phenotypes of JIA. A common feature of the different subtypes of JIA is a tumor-like expansion of the inflamed synovial tissue, which is infiltrated by inflammatory cells including macrophages, plasma cells and lymphocytes. The production of inflammatory cytokines by these cells is thought to be instrumental in the development and perpetuation of the inflammatory arthritis. Several cytokines secreted by activated macrophages and lymphocytes such as interleukin (IL)1, IL6, and tumor necrosis factor-α (TNFα) have been implicated in the perpetuation of inflammatory response in JIA 1. Other studies have implicated IL1β, IL6, IL8, soluble IL2-receptor (sIL2R), and TNFα in JIA 23456789.

CD40 is a member of the TNF-receptor family which is expressed on the surface of antigen presenting cells including B-cells, activated macrophages, dendritic cells and monocytes 10. Signaling through CD40 induces antigen presenting cells to express different immune accessory molecules important in cell-cell interactions 11. CD40 interacts with CD40 ligand (CD40L or CD154) which is predominantly expressed on activated CD4+ T-cells 12. CD154 is transiently expressed on activated T-cells and plays a crucial role in B-cell function 13. Activated T-cells expressing CD154 can also interact with CD40 on endothelial cells to induce production of inflammatory cytokines 14. The transient expression of CD154 allows T-cells to stimulate selected CD40-bearing cells to participate in the immune response. Abnormal expression of CD154 has been reported in patients with systemic lupus erythematosus (SLE) and rheumatoid arthritis (RA) 1516.

In addition to existing in a membrane bound form, CD154 can also be released as a soluble molecule (sCD154), which has been shown to have biological functions 17. Increased levels of sCD154 have been demonstrated in patients with SLE, 18, RA, 19, inflammatory bowel diseases (IBD) 20, systemic sclerosis 2122 and Sjogren's syndrome 23. To our knowledge the concentrations and possible role of sCD154 has not been investigated in JIA. Our objectives were to examine the levels of multiple cytokines including sCD154 in children with JIA and healthy controls, and to determine the association of sCD154 with other inflammatory cytokines reported to be elevated in JIA. We also sought to determine if sCD154 levels differed between the different subtypes of JIA.

Serum samples were obtained from 77 children (63% female) who fulfilled the International League of Associations for Rheumatology (ILAR) criteria for the diagnosis of JIA 24. There were 10 patients each with systemic JIA and rheumatoid factor (RF)-positive polyarticular JIA, 14 with RF-negative polyarticular JIA, 20 with persistent oligoarticular JIA, 11 with extended oligoarticular JIA and with enthesitis related arthritis (ERA), and 1 with undifferentiated arthritis. The mean age of onset of JIA in the patients was 7.5 years. A total of 81 pediatric controls (52% female) were also studied with a mean age of 12 years (range 7–16 years). Controls were children obtained from an Associated Regional and University Pathologists (ARUP) institute of clinical and experimental pathology children's normal values study. The majority of the cases and controls were of Northern European ancestry. Serum was separated from peripheral blood soon after being drawn from subjects and aliquots were stored at -80°C until the time of cytokine determination. Subjects were enrolled after providing informed consent, under protocols approved by the Institutional Review Board of the University of Utah.

Information about disease variables at the time of enrollment was available on 72 children with JIA. Active disease was defined as having arthritis in at least 1 joint. Seven children did not have active disease. Variables collected included number of affected joints, ESR, and medications used to treat JIA at the time of enrollment (table 1). At the time of collection, 71% were on a non-steroidal anti inflammatory agent, 46% were on a disease modifying anti-rheumatic drug, 25% were on methotrexate, 19% were on an anti-TNF agent, and 12% were on corticosteroids.

The serum levels of IL1β, IL5, IL6, IL8, IL13, sCD154, IFNγ, sIL2R, and TNFα were highly significantly elevated in cases compared to the controls (Table 2). Levels of IL2, IL4, IL10 and IL12 were not significantly different between cases and controls. The levels of IL2, IL4, IL5, and IFNγ were lower than 5 pg/ml in over 90% of subjects, while the levels of IL12 were lower than 5 pg/ml in 85% of subjects. The median levels of IL2, IL4, IL5, IL12, and IFNγ, were ~0 pg/ml among both cases and controls (Table 2). For this reason, these five cytokines were not included in subsequent analyses.

Utilizing a multiplex assay for cytokines and inflammatory markers, we characterized the profile of these biomarkers in a well defined cohort of JIA patients and controls. We observed striking elevations of a number of cytokines. Of special interest to us were the increased serum levels of sCD154 in children with different JIA subtypes. This elevation was most pronounced in children with systemic, RF-positive polyarticular and ERA subtypes, and modest among children with the oligoarticular subtypes. To our knowledge, ours is the first study to investigate sCD154 levels in JIA. Elevated levels of sCD154 have been found in other autoimmune disorders including SLE, RA, IBD, systemic sclerosis and Sjogren's syndrome 181920212223. Expression of CD40 on synovial monocytes, fibroblasts and dendritic cells in RA has been reported 2930. After stimulation in vitro, peripheral blood T-cells from patients with RA had increased and longer expression of CD154 compared to T-cells from controls 16. Blocking CD40 on synovial fibroblasts from RA patients with soluble anti-CD40 antibodies prior to co-culture with synovial mononuclear cells resulted in decreased TNFα levels 16. Moreover, blockade of CD154 with an antibody ameliorates collagen induced arthritis in a murine model of RA 31. These observations suggest a major role for CD154 in the perpetuation of inflammation in RA.

The inflammation in synovial tissue of children with JIA is characterized by the presence of antigen presenting cells and activated T-cells, and is indistinguishable from the pathology observed in patients with RA 32. Our results suggest that sCD154 might also play a role in the pathogenesis of most of the JIA subtypes. It is conceivable that the activated CD4 T-cells in JIA synovium may be the source of sCD154, although it is possible that other cells, such as activated platelets release sCD154. Cell-bound as well as sCD154 have been shown to activate endothelial cells in vitro, and increase production of leukemia inhibitory factor, IL6 and granulocyte-macrophage colony stimulating factors 14. We speculate that sCD154 could possibly interact with cells bearing CD40, possibly enhancing antigen presentation, and production of other inflammatory mediators in JIA.

Our correlation analyses showed significant association between sCD154 and the pro-inflammatory markers IL6, IL1 β, IL8 and TNFα. It has been shown that sCD154 has the potential to activate endothelial cells in vitro, and increase production of IL6 14. Elevated levels of IL6 in JIA have been reported by several authors 2345673334. Indeed, levels of IL6 were strikingly higher among all JIA subtypes compared to controls in our study. IL6 was also significant in the logistic regression analysis suggesting that it likely plays a major role in concert with other cytokines in the pathogenesis of JIA. The increased levels of the pro-inflammatory cytokines IL1β and IL8 observed in our cohort have also been reported in different forms of JIA as well as RA 345835. In agreement with earlier findings, we observed that levels of TNFα were elevated in all subtypes except those with persistent oligoarticular JIA, in which fewer joint involvement and less systemic inflammation is observed 68. Together these observations support a primary or secondary role for the inflammatory cytokines IL6, IL1β, IL8 and TNFα in different JIA subtypes.

The elevated levels of sIL2R and IL10 in the different subtypes of JIA are in agreement with previous studies 269 although with some subtle differences which may be attributable to the differences in the size of the cohorts. For example, although we observed increased levels of sIL2R in a JIA subtype equivalent to ERA in a study by Mangge et al 8, these concentrations were not significantly different from controls. Nevertheless, sIL2R levels tend to be highest in systemic or polyarticular JIA suggesting that it maybe a biomarker of JIA. Our logistic regression analysis suggests that IL10 might have a protective effect on disease, as might be anticipated by its inhibitory activity.

Some cytokines were barely detected in the serum in a majority of subjects detected in our study. For example, undetectable levels of IL2, and IFNγ have been reported by other investigators 3436. On the other hand, some of the results described here have not previously been reported by others. For instance some authors have reported finding no significant elevations in the levels of IL1β, IL8, or TNFα 24833. Differences in the results reported in this study and others could reflect heterogeneity of the different JIA cohorts. Other differences could be due to methodological issues, such as test specimen (serum versus plasma) and assay methodology. It is likely that some cytokines are more elevated in the synovial fluid and not in the serum, especially in children with persistently oligoarticular JIA.

Our study has several strengths. Our cohort of ~80 JIA patients and 80 pediatric controls is larger than most series. Our patients have been classified using the ILAR criteria, which results in more homogeneous subtypes compared to earlier studies. To our knowledge sCD154 has not been measured in JIA patients before. We also measured several other cytokines previously implicated in JIA. The multiplex assay used small volumes of serum to assay thirteen cytokines. This method has also been successfully used by de Jager et al33, but sCD154 and sIL2R were not among the cytokines included in their study. They also found that IL6 differed between controls and patients. We have used an objective partially automated multianalyte assay which we have found to reliably measure cytokines and other analytes in a variety of disorders including immune deficiency 26, coronary artery disease 37 and acute rheumatic fever 38. We believe the differences in cytokine profiles, as well as in the inflammatory marker sCD154, suggest that these might play a role in the pathogenesis of at least some subtypes of JIA.

The past decade has seen the emergence of biological therapeutic agents directed against several mediators of inflammation. Our findings of elevated TNFα is in agreement with the response seen with anti-TNF therapy in children with JIA. The elevated levels of IL1β and IL6 support the roles for biological agents targeting these cytokines as well. Finally, sCD154 could also be a potential target of biological therapy, although some studies in individuals with SLE raise the concern for pro-thrombotic effects, possibly due to effects of these agents on platelets and/or the endothelium 39. If the importance of the CD40–CD154 pathway in JIA is confirmed, and safety concerns could be addressed, therapy directed at inhibiting this pathway deserves further exploration in JIA.

We have demonstrated that sCD154 levels are elevated in sera from children with JIA. Furthermore, sCD154 levels correlated with several other pro-inflammatory cytokines, which are also elevated in children with JIA. The potential implications of sCD154 as a biomarker for treatment and monitoring patients is of particular appeal since sCD154 is a molecule associated with T cell activation and may play a role in the pathogenesis of JIA.

The authors declare that they have no competing interests.

SP conceived of the study, participated in the design and coordination of study, performed the statistical analysis, and wrote the manuscript, TBM Carried out the Luminex analyses, and helped to draft the manuscript, AET Participated in the design of the study, helped to coordinate the study, and to draft the manuscript, AW Participated in the design of the study, sample acquisition, and drafting of the manuscript, BC Enrolled subjects, involved in acquisition of samples and assisted with data collection, ASZ Participated in the design of the study, and helped to draft the manuscript, BMcN Participated in the design of the study, and helped to draft the manuscript, JFB Participated in the design of the study, and helped to draft the manuscript, HRH Senior author Participated in the design of the study, directed the Luminex analyses, and helped to draft the manuscript. All authors participated in the writing, read and approved the final manuscript.