Detection of tumor necrosis factor α and receptors in the serum and synovial fluid of patients with rheumatoid arthritis and osteoarthritis.

全文

(1)109. Detection of Tumor Necrosis Factor- and Receptors in the Serum and Synovial Fluid of Patients With Rheumatoid Arthritis and Osteoarthritis 1. 2. 2. Chung-Ming Huang, Chi-Lan Chen , Tsai-Hsiu Lin , Hsiu-Jain Wang 2. Division of Immunology and Rheumatology, Department of Internal Medicine; Department of Clinical Laboratory, 1. China Medical College Hospital, Taichung, and Foo Yin Institute of Technology, Kaohsiung, Taiwan, R.O.C.. Tumor necrosis factor (TNF) induces the production of two forms of soluble receptors (p55 and p75) that are present in human serum at concentrations that increase in inflammatory rheumatic disease. The purpose of this study was to evaluate the usefulness of TNF-. and soluble. TNF receptor (sTNFR) for distinguishing between rheumatoid arthritis (RA), osteoarthritis (OA) and normal individuals. Serum and synovial fluid (SF) from patients with RA and OA, and serum from normal control subjects were analyzed for p55, p75 and TNFimmunosorbent assays (ELISA). TNF-. by enzyme-linked. was detectable in 20% of RA sera. However, it could not be. measured in OA or control sera. Levels of TNF-. were significantly higher in SF of RA group. when compared with the OA group. Although p55 sTNFR concentrations in serum showed a significant increase in RA than in the controls, there was no difference between patients with OA and the controls. Measurement of p55 sTNFR in SF showed a significant difference between the RA and OA groups. Furthermore, p75 sTNFR in serum was markedly higher in RA and OA groups than in the controls. p75 in SF was significantly different between RA and OA patients. There was no significant correlation between p55, p75, TNF-. and disease activity, including. clinical and laboratory parameters (erythrocyte sedimentation rate and C-reactive protein levels). In conclusion, our study demonstrates that TNF- and sTNFR are up-regulated in patients with RA and produced locally in the joints. In addition, serum and SF levels of sTNFR and TNFdo not seem to be related to disease activity. ( Mid Taiwan J Med 1999;4:109-15 ). Key words disease activity, rheumatoid arthritis, soluble tumor necrosis factor receptor, tumor necrosis factor-. inducible cytokine produced primarily by. INTRODUCTION. Tumor necrosis factor (TNF)-. is an. monocytes and macrophages. Many biological activities of TNF-. Received January 14, 1999.. Revised March 5, 1999.. Accepted March 16, 1999. Address reprint requests to. Chung-Ming Huang, Division of. Immunology and Rheumatology, Department of Internal Medicine, China Medical College Hospital, No 2, Yuh-Der Road, Taichung, Taiwan, R.O.C.. appear to be involved in. the pathogenesis of chronic inflammation and arthritis. TNF-. can induce resorption of. cartilage and bone [1], endothelial adherence and activation of granulocytes [2], and class I.

(2) TNF and sTNFR in RA and OA. 110. major histocompatibility complex expression,. of patients with RA, OA and gout [19].. as well as stimulation of fibroblast growth via. However, SF levels were not tested in their. platelet-derived growth factor [3], synovial cells. study. In our study, we were interested in the. prostaglandin E2 and collagenase production [4]. The beneficial effect of an anti-TNFantibody in an animal model of type II collagen-induced arthritis supports the. production of sTNFR in the synovial joints at and sTNFR sites of enhanced TNF-. hypothesis that TNF-. is pivotal in the. pathogenesis of rheumatoid arthritis (RA) [5,6]. One mechanism involved in the regulation. production in RA. We determined the concentrations of TNF- and sTNFR in serum and SF of patients with RA and OA to investigate whether these could be used clinically to distinguish patients from different forms of arthritis and to research for the. of TNF function is the inducible proteolytic. relationship between TNF-. cleavage of cell soluble TNF receptors. clinical disease activity of RA.. , sTNFR and. (sTNFR), which results in down-regulation of the membrane receptors and formation of soluble forms of the receptors. These soluble forms of the receptors can block TNF. MATERIALS AND METHODS. Patients. function, by competing for TNF [7-10]. Studies. Thirty patients with rheumatic disease,. of serum levels of sTNFR have indicated that. who visited the Division of Immunology and. cleavage of the TNF receptors takes place. Rheumatology at China Medical College. constantly in vivo even in normal individuals,. Hospital, were studied. Among those patients,. and is enhanced in some disease states.. 15 patients (median age 61 years, range 41-73). Recently, soluble binding proteins identical to. had a definite RA according to the 1987. the extracellular portions of the two sTNFR. Revised American College of Rheumatology. were characterized [11-13]. These receptors are. Criteria [20] and 15 patients (median age 59. commonly known as the 75 kDa or type A. years, range 45-70) had clinically and. and 55 kDa or type B sTNFR. sTNFR are. radiologically diagnosed OA predominantly. differentially expressed on macrophages,. involving the knee joints. An age-matched. granulocytes, activated T and B lymphocytes,. control group included 15 healthy volunteers. and various other cell lines [14]. They have. with a median age of 57 years (range 39 - 70. been shown to increase after administration of. years). The majority of patients with RA and. TNF [15]. Production of TNF in disease. OA were treated with nonsteroidal anti-. condition is therefore likely to result in. inflammatory drugs (NSAIDs). In addition,. increased serum concentrations of sTNFR.. several of the patients with RA were treated. Both p55 and p75 sTNFR levels were. with low dose steroids (n = 12) and/or disease-. significantly higher in RA patients as. modifying anti-rheumatic drug (DMARD), i.e.,. compared with osteoarthritis (OA) patients. methotrexate (n = 5), sulfasalazine (n = 6),. from previous reports [16-18]. Differences are. hydroxychloroquine (n = 8), D-penicillamine. apparent in the serum, but are most markedly. (n = 1), or azathioprine (n = 1). No patient had. in. been injected with steroids in the same joint in. the. synovial. fluid. (SF). samples.. Furthermore, several studies have reported a. the three months previous to the study.. correlation between an elevation in sTNFR in. Clinical Evaluation. serum or SF and disease activity in some patients [16,17]. However, other. reports. disprove against such an association [18]. In Taiwan, Lee et al. have demonstrated the concentration of TNF-. and sTNFR in serum. To evaluate the relationship between TNF- , sTNFR and clinical disease activity of RA, we used a summation of scores based on six features [duration of morning stiffness for more than 1 hour, warmth, joint tenderness,.

(3) Chung-Ming Huang, et al.. 111. Table 1 . p55 sTNFR (pg/ml) in serum and synovial fluid from patients with RA, OA and control subjects Serum Group. No.. RA OA Control. 15 15 15. *p < 0.05;. Mean 2036.5 1303.8 1100.0. SD. Synovial fluid Range. 864.0* 630.2 366.8*. 939 494 400. Mean 7310.8 5535.0. 3280 2380 1600. SD 2678.9 2731.2. Range 2910 2750. 10000 10000. p < 0.05. RA= rheumatoid arthritis; OA= osteoarthritis.. Table 2. p75 sTNFR (pg/ml) in serum and synovial fluid from patients with RA, OA and control subjects Serum Group. No.. RA OA Control. 15 15 15. *p < 0.005;. p < 0.01;. Mean 4041.2 3149.1 2250.0. SD 1026* 1603.1 988.5*. Synovial fluid Range 2200 1000 500. Mean 6814.0 5382.5. 5000 5000 3500. SD 2885.7 2701.5. Range 2320 2180. 10000 10000. p < 0.05. RA= rheumatoid arthritis; OA= osteoarthritis.. stored at -20 0 C until being assayed for cytokines. Assays for TNF-. and sTNFR. sTNFR was measured by an enzymelinked immunosorbent assay (ELISA) (R&D Systems, Minneapolis, MN, USA) [22]. TNFwas also measured by the ELISA method, (Endogen, Woburn, MA, USA). All samples were tested in duplicate. The interassay variations were 6.1% and 4.1% for the p55 and p75 assays, respectively, and < 10% for the assay. The lower limits of detection for. TNFFig. 1. TNF-. level in synovial fluid (SF) from patients. p55 and p75 were 30 pg/ml and 10 pg/ml,. with rheumatoid arthritis (RA) and osteoarthritis (OA) (p <. respectively. The sensitivity of the TNF-. 0.01).. ELISA was < 5 pg/ml. Statistical Analysis. effusion, synovial thickening and levels of erythrocyte sedimentation rate (ESR), Creactive protein (CRP)]. The joint was globally defined as having “high inflammatory activity” if 4-6 parameters were present, and “low inflammatory activity” if < 2 were present [21]. Sample Preparation SF and peripheral venous blood samples. For paired samples, differences were tested by Mann-Whitney U test. Spearman coefficient was used to assess a correlation between cytokine (TNF- , p55 and p75 sTNFR) and disease activity parameters (ESR and CRP). Results were expressed as mean standard deviation (SD). A p value of less than 0.05 was considered significant.. were collected at the same time. All SF samples were aspirated from the knee joint and treated with sheep hyaluronidase (2000 U/ml). SF and blood samples were spun at 2000 rpm for 10 minutes. Then the SF supernatant and plasma were collected and. RESULTS. TNF-. in Serum and SF. TNF-. was detectable in only three (20%). of 15 serum samples from the RA patients and.

(4) TNF and sTNFR in RA and OA. 112. none of those from the OA group and the. cytokines play key roles in inflammatory. control subjects. Thirteen (86%) of the 15 SF. processes, and particular interest has been. samples from RA patients, and eight (53%) of. focused on elucidating the pathways regulated. 15 SF samples from the OA patients had. by these mediators and their corresponding. in SF of RA. antagonists and endogenous binding proteins. patients was significantly higher when. [23,24]. Measuring cytokines and their soluble. 236.9. receptors in biological fluids may not only be. . TNF-. measurable TNF-. compared with SF of OA patients (270.2 pg/ml vs 61.6. 57.9 pg/ml, p < 0.01) (Fig. 1).. helpful in investigating pathogenic mechanisms but may also have a potential value for. p55 sTNFR in Serum and SF As shown in Table 1, serum p55 sTNFR. evaluation of disease activity and prognosis. TNF-. levels in patients with RA, OA and control 864 pg/ml, 1303.8. subjects were 2036.5 pg/ml, and 1100. 630.2. 366.8 pg/ml, respectively.. The serum level of p55 sTNFR in RA patients was significantly higher than those in the control subjects (p < 0.05). The p55 sTNFR levels in SF between the RA and OA groups were significantly different (7310.8. 2678.9 pg/ml vs 5535. 2731.2 pg/ml, p. has been demonstrated to inhibit. the proteoglycan synthesis in cartilage and increase bone erosion. In this study, TNFwas found in the SF of 86% of RA patients and 53% of OA patients. In addition, levels of TNF-. were significantly higher in SF of RA. group when compared with OA group. On the other hand, TNF-. was detectable the sera. of 20% of the RA patients, but none of the OA . Our results are. < 0.05).. patients contained TNF-. p75 sTNFR in Serum and SF. lower than those obtained by Saxne et al. [25]. No significant difference in serum p75 sTNFR was found between patients with RA (4041.2. 1026 pg/ml) and OA (3149.1. 1603.1. pg/ml, Table 2). On the contrary, the serum sTNFR levels in RA and OA patients were significantly higher than in the controls (2250 988.5 pg/ml, p < 0.005 and p < 0.01, respectively). The p75 sTNFR level in SF was significantly higher in RA patients (6814 2885.7 pg/ml) than in OA patients (5382.5 2701.5 pg/ml, p < 0.05).. and Lee et al. [19], they have demonstrated that 50% to 79% RA patients had detectable TNFlevels in sera. The reason for the discrepancy is uncertain, possible explanations include differences in patient groups or methods of assay. In our study, the serum concentrations of p75 sTNFR in RA and OA patients were higher than those in control subjects. The serum levels of p55 sTNFR in patients with RA were higher than those in the control groups. However, the SF concentrations of p55 and p75. Correlation Between Cytokine and Disease. sTNFR in RA patients were significantly. Activity. higher than those in OA patients. These results. The correlation between TNF-. , p55 and. p75 sTNFR levels and laboratory variables as well as with clinical variables were determined. Serum and SF levels of TNF- , p55 and p75 sTNFR were not significantly correlated with the clinical disease activity and laboratory parameters (data not shown).. were in agreement as reported by Steiner et al. [16] and Cope et al. [17]. This suggests that increased levels of sTNFR generally occur in inflamed joints (in RA, and even in OA patients), which may be due to increased shedding of the receptor from the cell surface in the joints. It is not clear which factors (cytokines), apart from TNFthe release of TNF-. DISCUSSION. It has been well established that some. itself, regulate. receptor [26].. As the report of Roux-Lombard et al. [18], our results did not show any correlation.

(5) Chung-Ming Huang, et al.. 113. between disease activity and the levels of p55 and p75 sTNFR or TNF-. in RA patients. This. lack of correlation with established disease activity variables may indicate that, rather. biologic characterization of a specific tumor necrosis factor-. inhibitor. J Biol Chem 1989;264:11966-73.. 9. Engelmann H, Novick D, Wallach D. Tumor necrosis factor binding protein purified from human urine: evidence for immunological cross reactivity with cell. than disease activity, sTNFR may more likely. surface tumor necrosis factor receptors. J Biol Chem. reflect the disease process of RA.. 1990;265:1531-6.. In summary, we observed higher levels of TNF-. and sTNFR in SF, but not in serum, in. patients with RA than OA. This may suggest that TNF-. and sTNFR are up-regulated in RA. patients and produced locally in the joints. Moreover, serum and SF levels of TNF-. and. sTNFR do not seem to be related to disease activity.. 10. Porteu F, Nathan C. Shedding of tumor necrosis factor receptors by activated human neutrophil. J Exp Med 1990;172:599-607. 11. Seckinger P, Isaaz S, Dayer JM. A human inhibitor of tumor necrosis factor alpha. J Exp Med 1988;167:15116. 12. Engelmann H, Novick D, Wallach D. Two tumor necrosis factor binding proteins purified from human urine. J Biol Chem 1990;265:1531-6. 13. Kohon T, Brewer MT, Baker SL, et al. A sector tumor necrosis factor receptor gene product can shed naturally occurring tumor necrosis factor inhibitor.. ACKNOWLEDGMENT. Proc Natl Acad Sci USA 1990;87:8331-5.. This study was supported by a grant (DMR-87-040) from the China Medical College Hospital.. 14. Aggarwal BB, Essalu TE, Hass PE. Characterization of receptors for human tumor necrosis factor and their regulation by. interferon. Nature 1985;318:665-8.. 15. Lantz M, Malik S, Slevin ML, et al. Infusion of tumor necrosis factor causes an increase in circulation TNFbinding protein in humans. Cytokines 1992;2:402-6.. REFERENCES 1.. 16. Steiner G, Studnicka-Benke A, Witzmann G, et al. stimulates. Soluble receptors for tumor necrosis factor and. resorption and inhibits synthesis of proteoglycan in. Saklatvala J. Tumor necrosis factor-. interleukin-2 in serum and synovial fluid of patients. cartilage. Nature 1986;322:547-9.. with rheumatoid arthritis, reactive arthritis and. 2. Shalaby MR, Aggarwal BB, Rinderknecht E, et al.. 3.. osteoarthritis. J Rheumatol 1995;22:406-12.. Activation of human polymorphonuclear neutrophil. 17. Cope AP, Aderka D, Doherty M, et al. Increased levels. function by interferon gamma and tumor necrosis. of soluble tumor necrosis factor receptors in the. factor. J Immunol 1985;135:2069-73.. sera and synovial fluid of patients with rheumatic. Vilcek J, Palombella VJ, Henriksen-DeStefano D, et al. Fibroblast growth enhancing activity of tumor. disease. Arthritis Rheum 1992;35:1160-9. 18. Roux-Lombard P, Punzi L, Hasler F, et al. Soluble. necrosis factor and its relationship to other polypep-. tumor necrosis factor receptors in human inflam-. tide growth factors. J Exp Med 1986;163:632-43.. matory synovial fluids. Arthritis Rheum 1993;36:485-9.. 4. Dayer JM, Beutler B, Cerami A. Cachectin/tumor necrosis. factor. stimulates. collagenase. and. prostaglandin E2 production by human synovial cells and dermal fibroblasts. J Exp Med 1985;162:21638.. factor receptor in serum of patients with arthritis. J Formos Med Assoc 1997;96:573-8. 20.Arnett FC, Edworthy SM, Bloch DA, et al. The American Rheumatism Association 1987 revised. 5. Brennan FM, Chantry D, Jackson A, et al. Inhibitory effect of TNF-. 19. Lee CS, Chen KH, Wang PC. Soluble tumor necrosis. antibodies on synovial cell. interleukin-1 production in rheumatoid arthritis. Lancet 1989;II:244-7. 6. Haworth C, Brennan FM, Chantry D, et al.. criteria for the classification of rheumatoid arthritis. Arthritis Rheum 1988:3:315-24. 21. Doherty M, Richards N, Hornby J, et al. Relation between synovial fluid C3 degradation products and local joint inflammation in rheumatoid arthritis,. Expression of granulocyte-macrophage colony. osteoarthritis, and crystal associated arthropathy.. stimulating factor (GM-CSF) in rheumatoid arthritis: Eur J regulation by tumor necrosis factor-. Ann Rheum Dis 1988;47:190-7.. Immunol 1991;21:2575-9.. 22. Adolf GR, Apfler I. A monoclonal antibody-based enzyme immunoassay for quantitation of human. 7. Olsson I, Lantz M, Nilsson E, et al. Isolation and. tumor necrosis factor binding protein I, a soluble. characterization of a tumor necrosis binding protein. fragment of 60 kD TNF receptor in biological fluids.. from urine. Eur J Haematol 1989;42:270-5. 8. Seckingger P, Isaaz S, Dayer JM. Purification and. J Immunol Methods 1991;143:127-36. 23. Holt I, Cooper RG, Denton J, et al. Cytokine inter-.

(6) TNF and sTNFR in RA and OA. 114 relationships and their association with disease. factor-. activity in arthritis. Br J Rheumatol 1992;33:1145-8.. serum. Arthritis Rheum 1988;31:1041-5.. in rheumatoid arthritis synovial fluid and. 24. Rubin LA. The soluble interleukin-2 receptor in. 26. Taylor DJ. Cytokine combinations increase p75. rheumatic disease. [Editorial] Arthritis Rheum. tumor necrosis factor receptor binding and stimulate. 1990;33:1145-8.. receptor shedding in rheumatoid synovial fibroblast.. 25. Saxne T, Palladino MA, Heinegard JD, et al. Detection of tumor necrosis factor-. but not tumor necrosis. Arthritis Rheum 1994;37:232-5..

(7) 115. 1. 2. 2 2. p55. p55. p75. 1. p75. 20% p55 p55. p75 p55. p75. ESR. CRP. 1999;4:109-15. 1/14/1999 3/16/1999. 3/5/1999.

(8)