The role of kinins and cytokines in rheumatoid arthritis.

Abstract

Introduction: Rheumatoid arthritis (RA) is a systemic inflammatory disease characterized by inflammatory synovitis. The histopathological features include synovial hyperplasia, an inflammatory cell infiltration, angiogenesis and an inflammatory exudate into the synovial joint with progression to bone and joint destruction. While the exact aetiology of RA is unknown, a number of inflammatory cells and mediators have been implicated in the pathogenesis. Kinins are vasoactive peptides that have the capacity to induce the cardinal features of inflammation and considerable evidence exists for a role for the kallikrein-kinin cascade in inflammatory arthritis. The proinflarnmatory cytokines are also important mediators in rheumatoid arthritis and there is evidence for a functional relationship between the kallikrein-kinin and cytokine cascades in rheumatoid arthritis. Methods: Following approval from the Ethics Committee of the University of Natal, synovial tissue samples were obtained at arthroscopy from patients with RA and at autopsy (for controls). The tissue samples were processed for light microscopy and immunostained by the immunoperoxidase method to detect tissue kallikrein and the kinin B 1 and B2 receptors. The intensity of the immunostaining was quantified by image analysis. Blood and synovial fluid samples were obtained from patients with RA and blood from age and sex matched healthy volunteers. The RA patients were assessed clinically to determine the degree of disease activity and the presence or concomitant diseases. Disease activity was determined by the duration of morning stiffness, the twenty eight tender and swollen joint counts, pain on a visual analogue scale, patient's and physician's global assessment of disease activity (Likert scale), a local activity index, the modified Health Assessment Questionnaire (HAQ), disease activity score (DAS) and the erythrocyte sedimentation rate (ESR) and C reactive protein (CRP). In the synovial fluid (SF) samples, the functional activity of tissue kallikrein (TK) was demonstrated using an amidolytic assay and the total amount of TK was measured in a sandwich enzyme linked immunosorbent assay (ELISA). The Pearson's correlation coefficient was used to correlate the TK levels with measures of disease activity. Further, basal and generated kinins were measured in the SF by competitive ELISA, and the levels correlated with measures of disease activity. In the cytokine study, interleukin lP (IL IP) and tumour necrosis factor p (TNFP) were measured in the synovial fluid samples by ELISA, and the relationship between the cytokine levels and disease activity as well as TK, basal and generated kinins determined. Neutrophils were isolated from the blood and synovial fluid samples from the rheumatoid arthritis patients and from the blood samples from healthy volunteers. The circulating and synovial fluid neutrophils were immunostained to detect tissue kallkrein, the kinin moiety in the kininogen molecule and kinin BI and B2 receptors, and the immunofluorescence visualized by confocal microscopy. The images were digitally analysed using the Analysis 2.1 Pro system. The Kruskal Wallis and one-way ANOVA tests were used to compare the mean intensity of imrnunostaining in the control neutrophils with that present on the circulating and SF neutrophils harvested from RA patients. The intensity of labeling for these antigens was correlated with measures of disease activity. Results: 1. Synovial tissue samples: Labeling for tissue kallikrein was observed in the synovial lining and endothelial cells in control and rheumatoid tissue. There was a significant increase in the intensity of TK labeling in the endothelial cells of the rheumatoid tissue (p < 0.05). The kinin Bl and B2 receptor were visualized in the synovial lining cells, endothelial cells and the subintimal fibroblasts and macrophages in the control and rheumatoid synovial samples, with a significant increase in B 1 receptor labeling in the synovial lining cells in rhewnatoid synovial tissue (p < 0.01). 2. Tissue kallikrein activity and the total TK concentration was measured in the synovial fluid obtained from 20 patients with RA. There was no direct correlation between the between the enzymic and antigenic tissue kallikrein. There w as a significant negative correlation between the enzymic TK and the twenty eight swollen joint count (r = -0.464; p <0.05). 3. There was a significant negative correlation between the basal kinin and generated kinin levels (r= -0.454; p < 0.05). In addition, there was a negative correlation between the basal kinin levels and the C RP (r = -0.537; p < 0.05) and the disease activity score (r = -0.458; p < 0.05). In contrast, there was a positive correlation between the generated kinin levels and the twenty-eight tender and swollen joint counts (r = 0.536; p < 0.05 and r = 0.509; p < 0.05 respectively), the ESR (r = 0.598; p < 0.01), CRP (r = 0.725; p < 0.01) and the disease activity score (r = 0.676; p < 0.01). There was a significant correlation between the SF levels of IL IP and pain (r = 0.462; p < 0.05), physician's global assessment of disease activity (r = 0.549; p < 0.05), 28 tender joint count (r = 0.4 72; p < 0.05) and CRP (r = 0.530; p < 0.05). Although there appeared to be a correlation between the IL 1 p and disease activity score, this was not significant (r = 0.412; p = 0.07). In addition, the levels of synovial fluid lNF a correlated with the 28 tender joint (r = 0.458; p < 0.05) count and CRP (r = 0.653; p < 0.01). 4. There appeared to be a trend towards a negative correlation between the SF amidase TK levels and IL 1 p, however this was not significant. While there was no direct relationship between the SF levels of IL 1 P and the generated kinins, there was a positive correlation between low to moderate levels of IL 1 p and the generated kinins (r = 0.51, p < 0.05). In contrast, there was a negative correlation with higher levels of IL Ip (r = -0.5, p < 0.05). 5. Immunoreactive TK, kinin moiety and the Bl and B2 receptors were visualized on the circulating neutrophils from the healthy volunteers and the circulating and SF neutrophils from the RA patients. There was no statistically significant difference in the mean intensity of TK labeling in the circulating neutrophils from healthy volunteers (n=8) and the circulating and synovial fluid neutrophils of the RA patients n=8). However, when the intensity of labeling of the SF neutrophils (n=80) from the RA patients was compared to the circulating neutrophils (n=80) of healthy volunteers, there was a significant loss of TK labeling in the SF neutrophils of the RA patients (1-Way ANOVA, p < 0.01). In the RA patients, there was a loss of the kinin moiety from both the SF and circulating neutrophils compared to controls (Kruskal-Wallis: p < 0.05 and < 0.01 respectively). Although there was a clear increase in the intensity of labeling of the kinin BI receptor on the SF neutrophils from RA patients (n=8), when compared to the circulating neutrophils from healthy volunteers (n=8), the mean values did not reach significance (Kruskal Wallis; p > 0.05). However, a significant increase in Bl receptor labeling was observed on both the circulating and SF neutrophils of the RA patients (n=80) when compared to circulating neutrophils of the healthy volunteers (n=80) (1 Way ANOVA, p < 0.01 and < 0.05 respectively). In addition, there was a positive correlation between the immunoreactivity for the BI kinin receptor on the circulating neutrophils from the RA patients and the local activity index (r = 0.783; p < 0.05). Although there was a clear increase of the kinin B2 receptor on the circulating and SF neutrophils from the RA patients compared to circulating neutrophils from healthy volunteers, this was only significant for the circulating neutrophils from the RA patients (Kruskal-Wallis, p = 0.05). There was no correlation between the intensity of labeling of TK, the kinin moiety and the B2 receptor and measures of disease activity. Discussion and conclusions 1. This study provides the first evidence for the localization of TK and the kinin receptors in control and rheumatoid synovial tissue using antibodies specific for each protein and standard immunolabelling techniques. Synovial fibroblasts, macrophages and endothelial cells through the release of enzymes and cytokines have the ability to mediate the inflammatory changes and cartilage and bone destruction in RA. The presence of TK and the kinin receptors in these cells therefore provides evidence for a pathogenetic role for the kallikrein-kinin cascade in RA. 2. In addition, in RA there is an exudation of fluid into the joint space. Tissue kallikrein has been previously reported in the synovial fluid obtained from RA patients, however the correlation of TK levels and disease activity has not been previously studied. The negative correlation between enzymic TK and the twenty-eight swollen joint count, an indicator of disease activity suggests that there is a consumption of TK in inflammation, presumably due to increased kininogenase activity. Similarly, the kinin generating capacity of synovial fluid obtained from RA patients has been previously reported, however, this is the first study demonstrating a link between kinin generating capacity and validated markers of disease activity. The kinin generating capacity is a complex and dynamic cascade involving the bioregulation of all the components of the kallikrein-kinin system and is therefore more likely to accurately define the role of kinins in inflammatory arthritis than are individual components of the kallikrein-kinin cascade. Measurement of tissue kallikrein and basal kinins is affected by the presence of natural inhibitors and their short half-life in biological fluids. In addition to the synovial fluid study a decrease in the urinary TK activity and an increase in urine kinin generated kinins was demonstrated, suggesting that there is a systemic activation of the kallikrein kinin cascade in RA. 3. Although there is evidence for an interaction between the kallikrein-kinin and cytokine cascades in inflammation, in this study a direct correlation between the levels of interleukin 1 J3 and tumour necrosis factor J3 in the synovial fluid and TK and generated kinin levels was not found. This may be due to the wide variations in the levels of cytokines, the presence of inhibitors and anti-inflammatory cytokines, or a complex and dynamic relationship between the two cascades. However, the correlation of both generated kinins and interleukin l J3 and tumour necrosis factor J3 with disease activity provides circumstantial evidence for a synergistic role for these mediators in inflammatory arthritis. 4. In the neutrophil study, loss of immunoreactive tissue kallkrein and kinin moiety from the neutrophils obtained from RA patients was demonstrated. This finding supports the hypothesis that kinins are released from the neutrophils by the enzymatic action of tissue kallkrein and suggests that the kallkrein-kinin system is activated both locally and systemically in patients with RA. Further, there was upregulation of the both the kinin B 1 and B2 receptors on the neutrophils from the RA patients. While the B2 receptor is thought mediate most of the actions of kinins, the correlation of the intensity of B 1 receptors and the local activity index implies that the B 1 receptor may be important in inflammation. 5. These findings provide convincing evidence for the role of the kallikrein-kinin cascade in the pathogenesis of inflammation in RA. Further development of kinin receptor antagonists may provide a novel therapeutic modality.