ARTHRITIS & RHEUMATISM Vol. 50, No. 5, May 2004, pp 1430–1436 DOI 10.1002/art.20166 © 2004, American College of Rheumatology Secretion of Oncostatin M by Neutrophils in Rheumatoid Arthritis Andrew Cross,1 Steven W. Edwards,1 Roger C. Bucknall,2 and Robert J. Moots1 Objective. Neutrophils are known to express and release a large number of proinflammatory cytokines when they are stimulated by inflammatory stimuli. The objective of this study was to determine whether neutrophils express oncostatin M (OSM), a member of the interleukin-6 family of cytokines that has been implicated in the pathogenesis of inflammatory joint disease. Methods. Neutrophils were isolated from the blood of healthy volunteer donors and from the blood and synovial fluid of patients with rheumatoid arthritis (RA). OSM levels were measured in cell extracts and in culture supernatants by Western blotting. Total RNA was isolated from control and granulocyte–macrophage colony-stimulating factor (GM-CSF)–treated neutrophils, and OSM messenger RNA levels were quantified by hybridization of a radiolabeled probe. Results. GM-CSF stimulated a rapid and transient expression and release of OSM from blood neutrophils, which was more rapid than the expression and release from blood monocytes. A 28-kd protein was identified in cell extracts, but an additional 25-kd isoform was detected in culture supernatants. Synovial fluid neutrophils could not be stimulated to express OSM, but this cytokine was detected in cell-free supernatants at various levels. Conclusion. Blood neutrophils can be stimulated to express and rapidly release large quantities of OSM. We propose that this important cytokine is released from neutrophils as they infiltrate rheumatoid joints and, thus, contribute to the complex cytokine network that characterizes RA. Rheumatoid arthritis (RA) is a severe, chronic inflammatory condition characterized by a profound inflammatory immune response that leads to joint damage and destruction, as well as extraarticular disease with a significant impact on both morbidity and mortality (1,2). The pathologic processes underlying RA have not been fully elucidated. However, it is clear that there is a dysregulation of both the cellular immune system and the cytokine network (3–5). The success of anti– proinflammatory cytokine therapy directed at tumor necrosis factor ␣ (TNF␣) and, to a lesser extent, interleukin-1␤ (IL-1␤), has highlighted the importance of these cytokines in rheumatoid disease processes (6– 10). However, such therapies are not beneficial to all patients, which suggests that other processes are also important. Similarly, the most-studied cellular component of the immune response in RA is the T cell. However, development of a truly effective anti–T cell therapy has remained elusive, which suggests that other cellular components of the immune system play important pathologic roles in RA. Polymorphonuclear neutrophils are circulating phagocytes that participate in immune and inflammatory responses to both endogenous and exogenous stimuli. These cells are the most abundant of the immune system and represent the greatest proportion of cells that infiltrate the rheumatoid joint (11). However, they have been subjected to a disproportionately low number of investigations in RA. These cells are clearly present in vast numbers in synovial effusions in patients with RA (12,13). They are also focused in synovial tissue at sites of erosions and are capable of inducing profound inflammatory responses and damage (14,15). We and other investigators have shown that neutrophils are not merely short-lived, terminally differentiated cells, but rather, they are able to live for long periods of time Supported by Aintree Arthritis Trust UK and Wyeth Pharmaceuticals UK. 1 Andrew Cross, PhD, Steven W. Edwards, PhD, Robert J. Moots, MD, PhD: University of Liverpool, Liverpool, UK; 2Roger C. Bucknall, MD: Royal Liverpool University Hospital, Liverpool, UK. Address correspondence and reprint requests to Robert J. Moots, MD, PhD, University of Liverpool, Academic Rheumatology Unit, University Hospital Aintree, Longmoor Lane, Liverpool L9 7AL, UK. E-mail: firstname.lastname@example.org. Submitted for publication June 13, 2003; accepted in revised form January 10, 2004. 1430 SECRETION OF OSM BY NEUTROPHILS IN RA under conditions of inflammation and can up-regulate a variety of receptors, including class II major histocompatibility complex molecules, and a range of cytokines (16–18). A clear role for neutrophils in both the initiation and development of experimental arthritis in mice has recently been demonstrated (19). Oncostatin M (OSM) is a pleiotropic cytokine of the IL-6 family (20), but its role in inflammation is currently ambiguous. For example, OSM has been shown to have antiinflammatory effects by regulating tissue inhibitor of metalloproteinases 1 and antiproteases, inhibiting IL-1–induced IL-8 production by lung fibroblasts (21–24), and stimulating hepatocytes to secrete acute-phase proteins (25). Furthermore, anti-OSM antibodies have been shown to ameliorate experimental arthritis in mice (26). Conversely, OSM has been shown to exert proinflammatory effects in other situations. For example, it induces adhesion and chemotaxis in neutrophils and induces chemokine production by endothelial cells and synovial fibroblasts (27,28). Enhanced expression of OSM induces inflammation and arthritis in mice, and it is found in elevated concentrations in rheumatoid synovial fluid. Oncostatin synergizes with IL-1 to promote cartilage degradation, and it induces granulocyte colony-stimulating factor and granulocyte–macrophage colony-stimulating factor (GM-CSF) production by endothelial cells (29). Clearly, all of these properties may be of potential relevance in RA. In this study, we observed enhanced production of OSM in neutrophils stimulated by GM-CSF. We describe the dynamics of OSM production and secretion by neutrophils from the peripheral blood of healthy volunteer donors and in matched peripheral blood and synovial fluid samples from patients with RA. MATERIALS AND METHODS Materials. Neutrophil isolation medium was obtained from Cardinal Associates (Santa Fe, NM), Ficoll-Paque from Amersham Pharmacia (Uppsala, Sweden), RPMI 1640 medium from Gibco BRL (Paisley, UK), and fetal calf serum from Sigma (Poole, UK). The cytokines used were GM-CSF (Glaxo, Greenford, UK) and TNF␣ (Calbiochem, Nottingham, UK). Anti-OSM antibody (catalog no. RDI-ONCSabrP) and recombinant OSM (catalog no. RDI-310) were purchased from Research Diagnostics (Flanders, NJ); anti-OSM antibody obtained from Abcam (catalog no. ab9633; Abcam, Cambridge, UK) was used to confirm the results. A BD Atlas Nylon Array (catalog no. 7744-1, human cytokine/receptor) was used for gene expression analysis and was purchased from BD Clontech (Basingstoke, UK) Cell isolation and culture. This study was approved by the Institutional Review Board of the South Sefton Research 1431 Ethics Committee. Peripheral blood cells were prepared from heparinized venous blood obtained from healthy donors, and patients with RA (fulfilling the American College of Rheumatology [formerly, the American Rheumatism Association] 1987 revised criteria ), and patients with other inflammatory joint arthropathies. Cells were separated into neutrophil and mononuclear cell fractions by using neutrophil isolation medium (as described in the manufacturer’s instructions) (31). Contaminating erythrocytes were removed by hypotonic lysis. Neutrophils from the synovial fluid of RA patients were isolated soon after aspiration, by use of Ficoll-Paque in a 1-step centrifugation method (32). Neutrophils were routinely examined for purity and viability using trypan blue exclusion (⬎97% and ⬎98%, respectively) immediately after isolation. Purity was confirmed using morphologic analysis of cytospin preparations and CD15 and/or CD16 expression. Purified neutrophils were resuspended in 1640 RPMI, supplemented with 10% fetal calf serum at 5 ⫻ 106 cells/ml, and cultured at 37°C in a humidified chamber containing 5% CO2. Cytokines were added as indicated. U937 cells were cultured and maintained in RPMI 1640 medium supplemented with 10% fetal calf serum and 1 mM L-glutamine. Cells were stimulated with phorbol myristate acetate (80 ng/ml), and both the supernatant and cells were collected and processed as described below. RNA isolation and analysis. Total RNA was extracted from 3 ⫻ 107 isolated control and GM-CSF–treated (45 minutes) peripheral blood neutrophils using TRIzol reagent (Gibco BRL), as described in the manufacturer’s instructions. RNA was then further purified using an RNeasy mini kit (Qiagen, Crawley, UK). Probe preparation utilized a polymerase chain reaction complementary DNA (cDNA) synthesis kit (BD Clontech), which was then labeled with 32P-dATP (50 Ci) for 30 minutes at 50°C. Labeled cDNA was purified from unincorporated 32P-labeled nucleotides and small (⬍0.1-kb) cDNA fragments using spin-column chromatography. Hybridization of cDNA probes. The cDNA prepared above were then hybridized to BD Atlas Nylon Array filter arrays, exactly as described in the manufacturer’s instructions. Briefly, 2 identical Atlas array membranes were prehybridized with prewarmed hybridization solution (ExpressHyb; BD Clontech) containing denatured salmon testes DNA (0.1 mg/ml) for 30 minutes at 68°C. Labeled probe was denatured using a denaturing solution and, after neutralization, was added to the prehybridized membrane. Hybridization was performed overnight at 68°C, with continuous agitation for 18 hours. The membranes were then washed consecutively in prewarmed wash solution 1 (2⫻ saline– sodium citrate [SSC] and sodium dodecyl sulfate [SDS; 1% weight/volume]) and wash solution 2 (0.1⫻ SSC and SDS [0.5% w/v]) for 30 minutes each at 68°C, with continuous agitation. The membranes were then removed from the hybridization containers, blotted dry, sealed in plastic wrap, and analyzed. A Storm 840 PhosphorImager (Molecular Dynamics, Chesham, UK) was used for membrane analysis. Captured images were analyzed and hybridization signals quantified using ImageQuant version 5.2 software (Molecular Dynamics). Background signals were subtracted from each membrane, and signals were normalized, using levels of housekeeping genes as standards. Results were interpreted with the AtlasInfo database (BD Clontech). 1432 CROSS ET AL Statistical analysis. Data sets were analyzed using Student’s t-test. Figure 1. Oncostatin M (OSM) expression by healthy control blood neutrophils in response to granulocyte–macrophage colonystimulating factor (GM-CSF). A and B, Results of 1 of 3 experiments in which mRNA from control and GM-CSF–treated neutrophils (50 units/ml for 45 minutes) was hybridized to an Atlas macroarray filter. Only that portion of the filter with the OSM target (spotted in duplicate) is shown. The hybridization signal in B was 150–200 times greater than that in A, based on quantitation by PhosphorImager analysis. C, Healthy control blood neutrophils incubated for 2 hours at 37°C in the absence (0) and presence of 50 units/ml of GM-CSF. At the indicated times, samples were obtained and centrifuged, and OSM levels in the cell pellets and culture supernatants were detected by Western blotting. A Western blot of actin in cell pellets was performed to confirm equality of loading (bottom). Western analysis. In experiments in which both the supernatant and pellet were analyzed, cells were cultured at 2 ⫻ 107/ml. At each time point, 200 l was removed and centrifuged and 150 l of supernatant was removed and added to 30 l of boiling 5⫻ sample buffer with phenylmethylsulfonyl fluoride (PMSF; 1 mM). Cell pellets were then washed in phosphate buffered saline, centrifuged, and then lysed with 150 l of ice-cold lysis buffer (20 mM Tris, pH 7.5, 1 mM EDTA, 150 mM NaCl containing 1% v/v Igepal, 10 mM NaF, 10 g/ml of aprotinin, 10 g/ml of leupeptin, and 10 g/ml of pepstatin A) before the addition of boiling 5⫻ sample buffer and 1 mM PMSF. All samples were stored at ⫺80°C until analyzed. SDS–polyacrylamide gel electrophoresis (using 12% gels) was used to separate protein extracts, and OSM detection was performed using the primary antibody and an enhanced chemiluminescence detection system. Densitometry on carefully exposed blots (to avoid film saturation) was performed with Image 1.44 VDM software (National Institutes of Health, Bethesda, MD; online at: http://rsb.info.nih.gov/nih-image/). Ponceau S–stained actin on membranes confirmed equivalence of loading. RESULTS OSM expression by peripheral blood neutrophils from healthy control subjects. Neutrophils isolated from the blood of healthy volunteers expressed negligible levels of messenger RNA (mRNA) for OSM, as detected by hybridization to a DNA filter array (Figure 1A). However, in the same preparation of neutrophils treated for 45 minutes with GM-CSF, far higher levels of mRNA for OSM were detected (Figure 1B). Quantitation of hybridization signals obtained from control and GM-CSF–treated neutrophils indicated a 150–200-fold increase in counts following GM-CSF treatment. Neutrophils isolated from the blood of healthy controls exhibited low, but detectable, levels of OSM protein, as measured by Western blotting (Figure 1C). Cellular levels of this protein increased rapidly after GM-CSF treatment, peaking at 90–120 minutes after stimulation and declining thereafter (Figure 2). Thus, by 5 hours after stimulation with GM-CSF, cellular levels of OSM were approaching control, unstimulated levels. Control blood neutrophils secreted extremely low levels of OSM that were below the level of detection of the assay. However, following GM-CSF treatment, OSM was detectable in neutrophil supernatants and reached maximal levels between 90 and 120 minutes after stim- Figure 2. Kinetics of OSM expression and secretion by healthy control blood neutrophils in response to GM-CSF. Neutrophils from the blood of healthy controls were incubated as described in Figure 1. At the indicated times after addition of GM-CSF, OSM levels in cell pellets (solid bars) and culture supernatants (open bars) were determined by Western blotting. Values are the mean and SD (n ⫽ 3 samples). Relative levels of OSM are expressed as the percentage of a positive control (phorbol myristate acetate–treated U937 cells). See Figure 1 for definitions. SECRETION OF OSM BY NEUTROPHILS IN RA Figure 3. Kinetics of OSM expression and secretion by healthy control peripheral blood mononuclear cells (PBMCs) in response to GM-CSF. PBMCs from healthy controls were incubated as described in Figure 1. At the indicated times after addition of GM-CSF, OSM levels in cell pellets (solid bars) and culture supernatants (open bars) were determined by Western blotting. Values are the mean and SD (n ⫽ 3 samples). Relative levels of OSM are expressed as the percentage of a positive control (phorbol myristate acetate–treated U937 cells). See Figure 1 for other definitions. ulation (Figures 1C and 2). After this time, secreted levels also declined. Thus, GM-CSF treatment resulted in a rapid activation of both the expression and the secretion of OSM from neutrophils. Both expression and secretion were transient, falling to baseline (unstimulated) levels by ⬃6–8 hours. It is curious to note, that while only a single band of OSM was detected intracellularly (at 28 kd), 2 proteins were detected in culture supernatants, at 28 kd and 25 kd. We then tested whether another proinflammatory cytokine, TNF␣, was also capable of stimulating OSM expression and secretion. TNF␣ was shown to stimulate increased OSM expression by neutrophils, but very little secretion of this cytokine was detected (data not shown). OSM expression by peripheral blood mononuclear cells (PBMCs) from healthy controls. GM-CSF could also stimulate OSM expression in PBMCs, but peak expression was detected by 2–3 hours after stimulation (Figure 3), compared with the very rapid expression (60–90 minutes) observed following stimulation of neutrophils. Similarly, secretion of OSM was observed after GM-CSF treatment, but again, this was slower than that observed after stimulation of neutrophils, with maximal levels detected extracellularly by 4–5 hours after stimulation. Similarly, TNF␣ could stimulate OSM 1433 expression by mononuclear cells, but again, very little secretion was observed (data not shown). Thus, GM-CSF stimulated OSM expression and secretion from both neutrophils and mononuclear cells, but the kinetics of activation and secretion were much faster for neutrophils. OSM expression by neutrophils from healthy controls and patients with RA. Cellular levels of OSM were not significantly different in neutrophils isolated from healthy controls compared with those from patients with RA, as analyzed immediately after purification (Figure 4). Similarly, neutrophils from the blood of patients with RA could be stimulated by GM-CSF to express and secrete OSM, and the levels observed following simulation were not significantly different from those obtained with healthy control neutrophils (data not shown). In the absence of GM-CSF, cellular levels of OSM remained fairly constant during culture of blood neutrophils from controls or patients (Figure 5). However, cellular levels of OSM were rapidly enhanced by GM-CSF treatment. Figure 4. Oncostatin M (OSM) expression by neutrophils isolated from the blood of healthy controls and patients with rheumatoid arthritis (RA). Neutrophils were isolated from the blood of healthy controls and patients with RA, and cellular OSM levels were measured by Western blotting immediately after isolation. Neutrophils were also incubated at 37°C for 1 hour with 50 units/ml of granulocyte– macrophage colony-stimulating factor (GM-CSF) prior to measurement of OSM levels and densitometry. Values are the mean and SD (n ⫽ 6 GM-CSF–treated samples, n ⫽ 6 control samples, and n ⫽ 9 RA samples). Similar effects of GM-CSF on RA blood neutrophils were observed. 1434 Figure 5. OSM expression by blood neutrophils from patients with RA. Neutrophils were isolated from the blood of patients with RA and then incubated in the absence or presence of GM-CSF for up to 3 hours at 37°C. At the indicated times, samples were removed for measurement of cellular OSM levels by Western blotting. Values are the mean and SD (n ⫽ 3 samples). Levels of OSM in healthy controls, as measured by densitometry, were designated as 100%. See Figure 4 for definitions. Neutrophils isolated from the synovial fluid of patients with RA were shown to express OSM. Immediately after isolation, cellular levels were slightly decreased compared with levels in paired blood neutrophils, but this did not reach statistical significance (P ⫽ 0.07; n ⫽ 8). However, GM-CSF could not enhance the expression of this cytokine (Figure 6). Indeed, in the absence of GM-CSF (and in contrast to blood neutrophils), expression of OSM rapidly declined and was virtually undetectable by 3 hours of culture ex vivo (Figures 6 and 7). In contrast, GM-CSF slightly delayed the rate at which OSM levels per cell decreased, but no clear up-regulation of OSM was seen after GM-CSF treatment, as was observed in blood cells from controls or patients. OSM levels in RA synovial fluid. Cell-free synovial fluid samples derived from patients with RA were Figure 6. OSM expression by synovial fluid neutrophils from patients with RA. Neutrophils were isolated from the synovial fluid of patients with RA and then incubated in the absence or presence of GM-CSF for up to 3 hours at 37°C. At the indicated times, samples were removed for measurement of cellular OSM levels by Western blotting. See Figure 4 for definitions. CROSS ET AL Figure 7. OSM expression by synovial fluid neutrophils from patients with RA. Neutrophils were isolated from the synovial fluid of patients with RA and then incubated in the absence or presence of GM-CSF for up to 3 hours at 37°C. At the indicated times, samples were removed for measurement of cellular OSM levels by Western blotting. Values are the mean and SD (n ⫽ 3 samples). Levels of OSM in healthy controls, as measured by densitometry, were designated as 100%. See Figure 4 for definitions. tested for levels of OSM by Western blotting, using a protein extract from U937 cells as a positive control. OSM was detectable in almost all samples examined, but there was wide variation in the levels. Furthermore, the relative levels of the 2 isoforms of extracellular OSM (at 25 kd and 28 kd) varied widely. For example, some samples contained approximately equal levels of both isoforms (e.g., sample 6), while others contained primarily the 28-kd isoform (e.g., sample 11) or primarily the 25-kd isoform (e.g., sample 42) (Figure 8). The relative levels of OSM detected in a range of Figure 8. Relative oncostatin M (OSM) levels in synovial fluid from patients with rheumatoid arthritis (RA). Cell-free synovial fluid was isolated from different patients with RA (indicated by the numbers across the top), and 10 l of each sample (diluted in 200 l of sample buffer) was analyzed by Western blotting to determine relative OSM levels. Positive represents a cell extract obtained after phorbol myristate acetate treatment of U937 cells. SECRETION OF OSM BY NEUTROPHILS IN RA Figure 9. Relative oncostatin M (OSM) levels in synovial fluid from patients with rheumatoid arthritis (RA). Cell-free synovial fluid was isolated from different patients with RA (indicated by the numbers across the bottom), and 10 l of each sample (diluted in 200 l of sample buffer) was analyzed by Western blotting to determine relative OSM levels. Data are shown after quantitation by densitometry. Positive represents a cell extract obtained after phorbol myristate acetate treatment of U937 cells. rheumatoid synovial fluid samples are shown in Figure 9. In some of the samples, OSM levels were just above the level of detection, whereas in others, OSM levels exceeded those in the positive control. DISCUSSION Human neutrophils secrete a wide range of cytokines, and these secreted molecules have the capacity to direct the progress of an inflammatory reaction by influencing the activity of immune cells and tissues. In common with other cytokines known to be expressed by neutrophils, oncostatin M is not secreted in significant amounts by resting blood neutrophils, but its expression and secretion are rapidly activated by proinflammatory signals such as GM-CSF, which is known to be secreted by synovial cells in RA (33). This activation of expression is extremely rapid, with maximal cellular levels occurring within 1–2 hours of stimulation and maximal secreted levels occurring by 2 hours of stimulation. This contrasts with the rather slower kinetics of expression and secretion observed following addition of GM-CSF to PBMCs. Neutrophils have typically been reported to activate cytokine expression more rapidly than do mononuclear cells (17). As with other cytokines, we show that on a cell basis, neutrophils secrete ⬃20% of the levels of OSM as mononuclear phagocytes. However, since neutrophil numbers during inflammation can be far higher than mononuclear cell numbers, neutrophil-derived OSM expression is likely to be of extreme importance in disease pathogenesis. 1435 Neutrophils have previously been reported to express OSM, but, to our knowledge, there are no reports of 2 isoforms of this cytokine in the literature. While we detected only a 28-kd form in cell extracts (the reported molecular mass of the cytokine ), we consistently detected an additional form at 25 kd in neutrophil supernatants. These isoforms were detected when the Western blots were developed using 2 separate anti-OSM antibodies. Curiously, we also detected both isoforms in the synovial fluid of patients with RA, but there was great variation between donors in both the levels of OSM in different fluids and the relative levels of the 2 isoforms detected. For example, in some samples, both isoforms were detected in approximately equal amounts, while in others, either the 28-kd or the 25-kd isoform predominated. The molecular identity of the 25-kd isoform is unknown, but it may represent a processed form of the cytokine, perhaps as a result of partial proteolytic cleavage. It would also be extremely interesting to determine whether the 2 isoforms have different stabilities or different biologic functions. It will then be possible to determine whether these 2 molecules play different roles in RA, and hence, the biologic significance of the variation in the presence of these 2 molecules in different synovial fluid samples could be assessed. Somewhat to our surprise, we could not detect significantly elevated levels of OSM in neutrophils isolated from the blood of RA patients. Indeed, neutrophils isolated from the blood of these patients behaved in a manner similar to that of healthy control neutrophils in terms of their responsiveness to GM-CSF. Synovial fluid neutrophils behaved quite differently, however. OSM was clearly detected in freshly isolated synovial fluid neutrophils, but in contrast to blood cells, the levels of this cytokine rapidly declined as they were cultured in vitro in the absence of GM-CSF. Thus, after 3 hours of incubation in culture, cellular levels of this cytokine in synovial fluid neutrophils were virtually undetectable. Furthermore, the addition of GM-CSF did not stimulate OSM expression in these synovial fluid neutrophils, but rather, partially slowed the rate of disappearance of this cytokine. From these observations we conclude that the synovial fluid neutrophils have already been triggered to express and secrete OSM within the joints, and so, are desensitized to further stimulation. Indeed, this scenario may explain the presence of OSM within these joint fluids, although it must be stressed that cells other than neutrophils that are present within diseased joints can also contribute to the presence of this cytokine. Thus, based upon the observations in this study, we conclude that neutrophils are triggered to secrete OSM very soon after 1436 CROSS ET AL entering the diseased joint, possibly when in transit through the pannus, where they will encounter GM-CSF in significant amounts, and that the release of OSM contributes to the pathologic disease processes. As the role of OSM in disease pathogenesis remains enigmatic, it is uncertain as yet if this neutrophil-dependent cytokine release is harmful or beneficial to the disease process. REFERENCES 1. Bresnihan B. Pathogenesis of joint damage in rheumatoid arthritis. J Rheumatol 1999;26:717–9. 2. Grossman JM, Brahn E. Rheumatoid arthritis: current clinical and research directions. J Womens Health 1997;6:627–38. 3. Beaulieu AD, McColl SR. Differential expression of two major cytokines produced by neutrophils, interleukin-8 and the interleukin-1 receptor antagonist, in neutrophils isolated from the synovial fluid and peripheral blood of patients with rheumatoid arthritis. Arthritis Rheum 1994;37:855–9. 4. Farahat MN, Yanni G, Poston R, Panayi GS. Cytokine expression in synovial membranes of patients with rheumatoid arthritis and osteoarthritis. Ann Rheum Dis 1993;52:870–5. 5. Ridderstad A, Abedi-Valugerdi M, Moller E. Cytokines in rheumatoid arthritis. Ann Med 1991;23:219–23. 6. Abramson SB, Amin A. Blocking the effects of IL-1 in rheumatoid arthritis protects bone and cartilage. Rheumatology (Oxford) 2002;41:972–80. 7. Catrina AI, Lampa J, Ernestam S, af Klint E, Bratt J, Klareskog L, et al. Anti-tumour necrosis factor (TNF)-␣ therapy (etanercept) down-regulates serum matrix metalloproteinase (MMP)-3 and MMP-1 in rheumatoid arthritis. Rheumatology (Oxford) 2002;41: 484–9. 8. Feige U, Hu YL, Gasser J, Campagnuolo G, Munyakazi L, Bolon B. Anti-interleukin-1 and anti-tumor necrosis factor-␣ synergistically inhibit adjuvant arthritis in Lewis rats. Cell Mol Life Sci 2000;57:1457–70. 9. Geborek P, Crnkic M, Petersson IF, Saxne T. Etanercept, infliximab, and leflunomide in established rheumatoid arthritis: clinical experience using a structured follow up programme in southern Sweden. Ann Rheum Dis 2002;61:793–8. 10. Maini RN, Taylor PC. Anti-cytokine therapy for rheumatoid arthritis. Annu Rev Med 2000;51:207–29. 11. Edwards SW, Hallett MB. Seeing the wood for the trees: the forgotten role of neutrophils in rheumatoid arthritis. Immunol Today 1997;18:320–4. 12. Chatham WW, Swaim R, Frohsin H Jr, Heck LW, Miller EJ, Blackburn WD Jr. Degradation of human articular cartilage by neutrophils in synovial fluid. Arthritis Rheum 1993;36:51–8. 13. Kitsis E, Weissmann G. The role of the neutrophil in rheumatoid arthritis. Clin Orthop 1991;265:63–72. 14. Kowanko IC, Ferrante A, Clemente G, Youssef PP, Smith M. Tumor necrosis factor priming of peripheral blood neutrophils from rheumatoid arthritis patients. J Clin Immunol 1996;16: 216–21. 15. Mohr W, Pelster B, Wessinghage D. Polymorphonuclear granulocytes in rheumatic tissue destruction. VI. The occurrence of PMNs in menisci of patients with rheumatoid arthritis. Rheumatol Int 1984;5:39–44. 16. Cassatella MA, Gasperini S, Russo MP. Cytokine expression and release by neutrophils. Ann N Y Acad Sci 1997;832:233–42. 17. Quayle JA, Adams S, Bucknall RC, Edwards SW. Interleukin-1 expression by neutrophils in rheumatoid arthritis. Ann Rheum Dis 1995;54:930–3. 18. Cross A, Bucknall RC, Castella MA, Edwards SW, Moots RJ. Synovial fluid neutrophils transcribe and express class II major histocompatibility complex molecules in rheumatoid arthritis. Arthritis Rheum 2003;48:2796–806. 19. Wipke BT, Allen PM. Essential role of neutrophils in the initiation and progression of a murine model of rheumatoid arthritis. J Immunol 2001;167:1601–8. 20. Lahiri T, Laporte JD, Moore PE, Panettieri RA Jr, Shore SA. Interleukin-6 family cytokines: signaling and effects in human airway smooth muscle cells. Am J Physiol Lung Cell Mol Physiol 2001;280:L1225–32. 21. Langdon C, Leith J, Smith F, Richards CD. Oncostatin M stimulates monocyte chemoattractant protein-1– and interleukin1–induced matrix metalloproteinase-1 production by human synovial fibroblasts in vitro. Arthritis Rheum 1997;40:2139–46. 22. Richards CD, Kerr C, Tanaka M, Hara T, Miyajima A, Pennica D, et al. Regulation of tissue inhibitor of metalloproteinase-1 in fibroblasts and acute phase proteins in hepatocytes in vitro by mouse oncostatin M, cardiotrophin-1, and IL-6. J Immunol 1997; 159:2431–7. 23. Wahl AF, Wallace PM. Oncostatin M in the anti-inflammatory response. Ann Rheum Dis 2001;60:75–80. 24. Wallace PM, MacMaster JF, Rouleau KA, Brown TJ, Loy JK, Donaldson KL, et al. Regulation of inflammatory responses by oncostatin M. J Immunol 1999;162:5547–55. 25. Richards CD, Shoyab M, Brown TJ, Gauldie J. Selective regulation of metalloproteinase inhibitor (TIMP-1) by oncostatin M in fibroblasts in culture. J Immunol 1993;150:5596–603. 26. Plater-Zyberk C, Buckton J, Thompson S, Spaull J, Zanders E, Papworth J, et al. Amelioration of arthritis in two murine models using antibodies to oncostatin M. Arthritis Rheum 2001;44: 2697–702. 27. Kerfoot SM, Raharjo E, Ho M, Kaur J, Serirom S, McCafferty DM, et al. Exclusive neutrophil recruitment with oncostatin M in a human system. Am J Pathol 2001;159:1531–9. 28. Modur V, Feldhaus MJ, Weyrich AS, Jicha DL, Prescott SM, Zimmerman GA, et al. Oncostatin M is a proinflammatory mediator: in vivo effects correlate with endothelial cell expression of inflammatory cytokines and adhesion molecules. J Clin Invest 1997;100:158–68. 29. Catterall JB, Carrere S, Koshy PJ, Degnan BA, Shingleton WD, Brinckerhoff CE, et al. Synergistic induction of matrix metalloproteinase 1 by interleukin-1␣ and oncostatin M in human chondrocytes involves signal transducer and activator of transcription and activator protein 1 transcription factors via a novel mechanism. Arthritis Rheum 2001;44:2296–310. 30. Arnett FC, Edworthy SM, Bloch DA, McShane DJ, Fries JF, Cooper NS, et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 1988;31:315–24. 31. Fossati G, Moulding DA, Spiller DG, Moots RJ, White MR, Edwards SW. The mitochondrial network of human neutrophils: role in chemotaxis, phagocytosis, respiratory burst activation and commitment to apoptosis. J Immunol 2003;170:1964–72. 32. Edwards SW, Hughes V, Barlow J, Bucknall R. Immunological detection of myeloperoxidase in synovial fluid from patients with rheumatoid arthritis. Biochem J 1988;250:81–5. 33. Alvaro-Gracia JM, Zvaifler NJ, Brown CB, Kaushansky K, Firestein GS. Cytokines in chronic inflammatory arthritis. VI. Analysis of the synovial cells involved in granulocyte-macrophage colonystimulating factor production and gene expression in rheumatoid arthritis and its regulation by IL-1 and tumor necrosis factor-␣. J Immunol 1991;146:3365–71. 34. Bruce AG, Hoggatt IH, Rose TM. Oncostatin M is a differentiation factor for myeloid leukemia cells. J Immunol 1992;149: 1271–5.