1. Top of page
  2. Abstract
  6. Acknowledgements


To investigate the involvement of proinflammatory and destructive mediators in oncostatin M (OSM)–induced joint pathology, using gene-deficient mice.


An adenoviral vector expressing murine OSM was injected into the joints of naive wild-type mice and mice deficient for interleukin-1 (IL-1), IL-6, tumor necrosis factor α (TNFα), or inducible nitric oxide synthase (iNOS). Reverse transcription–polymerase chain reaction was used to study gene expression. Inflammation and cartilage proteoglycan (PG) depletion were assessed by histology. OSM and IL-1 levels in synovial fluid from patients with juvenile idiopathic arthritis (JIA) were measured by enzyme-linked immunosorbent assay.


Adenoviral expression of murine OSM led to joint inflammation, bone apposition, chondrophyte formation, articular cartilage PG depletion, and VDIPEN neoepitope expression in wild-type mice. A unique and consistent observation was the focal PG depletion and disorganization of the growth plate cartilage during the first week of inflammation. Synovial IL-1β, IL-6, TNFα, and iNOS gene expression was strongly induced. Of these factors, only deficiency in IL-1 markedly reduced inflammation and PG depletion and completely prevented growth plate damage. In addition, this is the first study in which OSM was detected in JIA synovial fluid. Most samples were also IL-1β positive.


IL-1, but not IL-6, TNFα, or iNOS, plays an important role in joint disease induced by intraarticular gene transfer of OSM in mice. The effect of OSM on murine connective tissue and the presence of OSM in human synovial fluid make involvement of OSM in human arthropathies very likely.

Oncostatin M (OSM) is a multifunctional cytokine that belongs to the interleukin-6 (IL-6) family (1). Elevated levels of OSM can be detected in the synovial fluid, but not in the serum, of patients with rheumatoid arthritis (RA) (2). Immunohistochemical analysis of RA synovial tissue showed that synovial macrophages are the source of OSM in the inflamed joint (3). A pathologic role for OSM in RA is suspected, because OSM by itself can induce joint inflammation in animals. Injecting recombinant human OSM into the joints of goats induced the influx of polymorphonuclear cells (PMNs), followed by cells of the macrophage/monocyte lineage (4). Joint inflammation was also induced by adenoviral expression of murine OSM in mice (5, 6). The enhanced expression of adhesion molecules such as E-selectin and P-selectin (7, 8), CXC chemokines (8), and the CC chemokine monocyte chemotactic protein 1 (9) by OSM could contribute to the influx of inflammatory cells. Furthermore, synovial fibroblasts displayed a transformed phenotype under the influence of OSM (5), suggesting involvement of OSM in pannus formation.

Besides chronic joint inflammation, RA is also characterized by destruction of articular cartilage and bone. Cartilage consists of a framework of collagen fibers in which proteoglycans (PGs) are entrapped. These PGs can retain water, which enables the cartilage to resist compressive forces. Proinflammatory cytokines such as IL-1 (10, 11) are involved in cartilage degradation. Results from experiments in recent years suggest a similarly important role for OSM in the cartilage degradation of RA. OSM was shown to induce collagen release from bovine cartilage in vitro (12). It also stimulated PG release and suppressed PG synthesis in porcine articular cartilage explants (13). Injecting OSM into the joints of goats (4) decreased the cartilage PG content. In humans, OSM concentrations in synovial fluid correlate positively with levels of cartilage degradation markers (14). OSM was also the first cytokine that, in combination with IL-1α, was demonstrated to induce collagen release from human cartilage (3).

The development of joint inflammation and cartilage damage in experimental arthritis can be greatly influenced by the expression of proinflammatory cytokines and other mediators. We previously demonstrated that blocking of IL-1 could prevent inhibition of PG synthesis in experimental arthritis (15, 16). The formation of nitric oxide (NO) was shown to be involved in IL-1–induced inhibition of PG synthesis in vitro (17), and PG loss was reduced in experimental arthritis in mice deficient for the inducible NO synthase (iNOS) gene (18). Studies entailing blocking of tumor necrosis factor α (TNFα) showed involvement of TNFα in the early phase of joint inflammation (19, 20), while studies of experimental arthritis in IL-6–deficient mice showed involvement of IL-6 in the chronicity of arthritis (21). In the present study, we investigated the involvement of these proinflammatory mediators in OSM-induced joint disease. We injected an adenoviral vector expressing murine OSM into the joints of mice deficient for IL-1, IL-6, TNFα, or iNOS and studied the effects of these gene deletions on OSM-induced joint pathology.

Ubiquitous transgenic overexpression of bovine OSM has been found to be lethal for newborn mice. One mouse survived and developed growth plate disorganization, with enhanced growth of the hind legs (22). Growth plate damage (23, 24) as well as localized growth abnormalities (25) are characteristic features of juvenile idiopathic arthritis (JIA). Therefore, we studied the effects of murine OSM gene transfer not only on development of inflammation and articular cartilage damage, but also on the growth plate. Furthermore, we studied expression of OSM in the synovial fluid of patients with JIA.


  1. Top of page
  2. Abstract
  6. Acknowledgements


For this study, male mice deficient for the following genes were used: IL-1α and IL-1β (26), IL-6 (27), TNFα (28), and iNOS (29). C57BL/6 and C57BL/6 × 129Sv mice were used as wild-type controls. Breeding colonies were kept at the Central Animal Facilities of the University of Nijmegen. Animals used in the experiments were between 11 and 13 weeks of age. All mice were housed in filter-top cages under specific pathogen–free conditions. A standard diet and water were provided ad libitum. After injection of the adenoviral vector, the mice were housed in isolators. Experiments were performed according to national and institutional regulations for animal use.

Adenoviral vectors and intraarticular injection.

The construction of adenoviral vectors expressing murine OSM (AdMuOSM) or murine IL-17 (AdmIL-17) has been described previously (30, 31). AdDL70-3, a vector without insert, was used as a control vector. For in vivo experiments, the virus was diluted in physiologic saline, and 2 × 106 plaque-forming units (PFU) in a total volume of 6 μl were injected into the knee joint cavity. Construction of NIH3T3 cells overexpressing human IL-1β will be described elsewhere (Joosten L: unpublished observations). A total of 2.5 × 104 cells were injected into the knee joint.

Histologic evaluation of knee joints.

Knee joints were dissected, fixed in formalin, decalcified, dehydrated, and embedded in paraffin. Standard 7-μm frontal sections were prepared. Sections were stained with Safranin O and counterstained with fast green for assessing cartilage damage. Histopathologic findings were scored on 5 semi-serial sections of the joint. Scoring was performed in a blinded manner by 2 independent observers. Cartilage depletion was scored from 0 (normal Safranin O staining; no depletion) to 3 (complete loss of Safranin O staining; complete depletion). Joint inflammation was also scored on a 0–3-point scale.

NIMP-R14 staining.

The influx of PMNs was assessed by staining knee joint sections for the presence of the NIMP-R14 epitope (32), which is present mainly on neutrophils. Sections were deparaffinized, treated with 0.1% trypsin in 0.1% CaCl2, pH 7.8 and preincubated for 15 minutes with 20% normal rabbit serum before incubation for 1 hour with anti–NIMP-R14 antibodies (a kind gift from Dr. M. Strath, London, UK). After incubation with a peroxidase-labeled rabbit anti-rat secondary antibody in 5% normal mouse serum/phosphate buffered saline (PBS) for 30 minutes, the sections were incubated with diaminobenzidine (1 mg/ml in 50 mM Tris HCl, pH 7.6, 0.001% H2O2) for 10 minutes. Sections were counterstained with hematoxylin for 30 seconds. Normal rat immunoglobulin was used as a negative control.

Image analysis of newly formed bone.

The area of newly formed bone was measured using the QWin image analysis system (Leica, Cambridge, UK). Images of Safranin O–stained sections were captured using a JVC 3-CCD color video camera and displayed on a computer monitor. For each joint, 4 measurements of the length of the original cortical bone (marked by a precipitation line in the staining) and the area of newly formed bone on the femur were performed in a standardized manner. The amount of newly formed bone is expressed as μm2 of new bone/10 μm of cortical bone.

Isolation of synovial RNA and semiquantitative reverse transcription–polymerase chain reaction (RT-PCR).

Synovial messenger RNA (mRNA) was isolated and quantitated as described previously (33). The patellae (with surrounding synovium) were isolated from the knee joints, and 2 pieces of tissue adjacent to the patella were punched out with a 3-mm biopsy punch (Stiefel, Wächters Bach, Germany). The tissue was immediately frozen in liquid nitrogen. Tissue samples were homogenized in a freeze mill, thawed in 1 ml of TRIzol reagent, and further processed according to the manufacturer's protocol. All reagents for RNA isolation and RT-PCR were obtained from Life Technologies (Breda, The Netherlands). Isolated RNA was treated with DNase I before being reverse transcribed into complementary DNA (cDNA) with Moloney murine leukemia virus reverse transcriptase.

After increasing numbers of PCR cycles, samples were obtained and run on an agarose gel. The cycle number at which the PCR product was first detected on the gel was obtained as a measure for the amount of specific mRNA originally present in the isolated synovial RNA. PCR for GAPDH was performed to verify that equal amounts of cDNA were used. Primers for IL-1β, TNFα, GAPDH (34), and IL-6 (6) were used as described previously. The iNOS primers used (at 55°C, 1 mM MgCl2) were as follows: forward CCC-TAA-GAG-TCA-CCA-AAA-TGG, reverse CTA-CAG-TTC-CGA-GCG-TCA-AA. The OSM primers used (at 55°C, 1 mM MgCl2) were as follows: forward CTT-GGA-GCC-CTA-TAT-CCG-CC, reverse GTG-TGG-AGC-CAT-CGT-CCC-ATT-C. Primers were designed using Oligo 4.0 and Primer software (Molecular Biology Insights, Cascade, CO).

Ex vivo PG synthesis.

PG synthesis was assessed by 35S-sulfate incorporation in patellar cartilage. Patellae from knee joints injected with adenoviral vectors and from the uninjected contralateral knee joints were dissected, with a minimum amount of surrounding synovium, under sterile conditions. The ex vivo synthesis assays were performed with RPMI supplemented with 100 units/ml penicillin, 100 μg/ml streptomycin, 1 mmole/liter pyruvate, and 5% fetal calf serum (Life Technologies). The patellae were placed separately in 200 μl of medium containing 4 μCi 35S-sulfate and incubated for 3 hours at 37°C in a humidified atmosphere of 5% CO2. After labeling, the patellae were washed with physiologic saline and fixed overnight in 4% formalin. Fixed patellae were decalcified in 5% formic acid for 4 hours, dissected, and dissolved in 0.25 ml Lumasolve (Omnilabo, Breda, The Netherlands). After addition of 1 ml of Lipoluma (Omnilabo), the 35S-sulfate content of each patella was measured by liquid scintillation counting in a TriLux 1450 MicroBeta (Perkin-Elmer Wallac, Turku, Finland). Data for joints injected with adenoviral vectors are presented as a percentage of the normal chondrocyte PG synthesis in the contralateral, uninjected knee joint.

VDIPEN neoepitope staining.

Irreversible PG damage was assessed by immunohistochemistry for the VDIPEN neoepitope. Joint sections were deparaffinized, rehydrated, and digested for 1 hour at 37°C in 0.25 units/ml of chondroitinase ABC in 0.1M Tris HCl, pH 8.0 (Sigma, Zwijndrecht, The Netherlands) to remove chondroitin sulfate from the PGs. Sections were treated with 1% H2O2 in methanol for 20 minutes, followed by treatment with 0.1% Triton X-100 in PBS for 5 minutes. After incubation with 1.5% normal goat serum for 20 minutes, sections were incubated overnight at 4°C with affinity-purified rabbit anti-VDIPEN IgG (a kind gift from Dr. I. Singer, Rahway, NJ) or with normal rabbit IgG. The next day, the sections were incubated with biotinylated goat anti-rabbit IgG followed by labeling with avidin–peroxidase (Elite kit; Vector, Burlingame, CA). Peroxidase development was performed using nickel enhancement to increase sensitivity. Sections were counterstained with 2% orange G for 5 minutes.

Synovial fluid from patients with JIA.

Synovial fluid was collected from children with JIA who attended the University Medical Center Nijmegen or the St. Maartens Clinic for intraarticular glucocorticosteroid injection. The samples were collected in accordance with local ethics legislation and were made available anonymously to the authors after informed consent was received from the parents. Diagnosis according to the revised International League of Associations for Rheumatology criteria (35) is described in Table 1. The synovial fluid was centrifuged for 10 minutes at 3,000 revolutions per minute, aliquoted, and stored at −70°C.

Table 1. OSM and IL-1β in synovial fluid of patients with JIA*
PatientAge, years/sexJIA subtypeOSM, pg/mlIL-1β, pg/ml
  • *

    All samples were obtained from knee joints, except sample 10, which was obtained from an ankle joint. In patients 2, 4, and 6, the arthritic leg was longer than the unaffected leg. Bony overgrowth of the knee occurred in patients 1, 6, and 10. Oncostatin M (OSM) and interleukin-1β (IL-1β) were determined by enzyme-linked immunosorbent assay, performed twice, in duplicate. Values are the mean ± SD. JIA = juvenile idiopathic arthritis; ND = not detectable; RF = rheumatoid factor; NM = not measured.

16/FOligoarthritis18.4 ± 4.15.2 ± 0.8
310.2/FRF− polyarthritis12.7 ± 0.55.1 ± 1.0
47.6/FOligoarthritis17.5 ± 1.4ND
5 (left knee)9.1/FOligoarthritis10.1 ± 0.55.4 ± 0.8
5 (right knee)9.1/FOligoarthritis13.7 ± 0.4ND
715.4/FRF− polyarthritisND11.3 ± 2.7
813/FRF− polyarthritis9.6 ± 1.62.2 ± 0.5
95.7/FSystemic23.1 ± 0.24.2 ± 1.1
106.2/FOligoarthritis27.4 ± 0.956.7 ± 5.2
1111.3/FOligoarthritis4.1 ± 0.5ND
124.4/FOligoarthritis16.3 ± 2.511.7 ± 2.7

Cytokine measurement in synovial fluid.

The amount of OSM and IL-1β in synovial fluid was determined by enzyme-linked immunosorbent assay (ELISA), using the Quantikine human OSM immunoassay or the Quantikine human IL-1β immunoassay (R&D Systems, Minneapolis, MN), according to the manufacturer's protocol.

Statistical analysis.

The rank sum test was used for statistical comparison between groups. P values less than 0.05 were considered significant.


  1. Top of page
  2. Abstract
  6. Acknowledgements

Joint pathology after OSM gene transfer.

In wild-type mice, intraarticular injection of AdMuOSM induced joint inflammation (Figures 1A and B) that was characterized by the influx of PMNs (Figure 1C) and mononuclear cells as well as by synovial hyperplasia. The AdMuOSM-induced inflammation led to cartilage PG loss in the patella and femur, as demonstrated by loss of red staining in the Safranin O–stained knee joint sections (Figures 1A and B). Both inflammation and cartilage PG loss lasted for at least 4 weeks. The periosteum became activated (Figure 1A), and apposition of new bone occurred at this site (Figure 1B). No additional apposition or remodeling of the new bone occurred after day 14. Chondrophytes, abnormal cartilaginous masses that can develop on the articular surface of bone, were formed on the patella and femur (Figure 1E) and increased in size until at least week 4. In contrast, injection of 2 × 106 PFU of the control vector AdDL70-3 did not induce joint inflammation, PG loss, bone apposition, or chondrophyte formation (Figures 1D and F).

thumbnail image

Figure 1. Joint pathology induced by the adenoviral murine oncostatin M vector (AdMuOSM) in wild-type mice. Inflammation and cartilage proteoglycan (PG) depletion on A, day 7 and B, day 14 after injection of AdMuOSM into the knee joint of wild-type mice. PG depletion is shown by reduced red staining of the upper cartilage layer of the patella and femur. In B, arrows indicate periosteal bone apposition. C, NIMPR-14 staining of polymorphonuclear cells in the inflamed synovium on day 7 after AdMuOSM injection. D, Normal joint histology on day 14 after injection of the control vector AdDL70-3. Similar histology was observed on day 7 after injection (results not shown). E, Chondrophyte formation (arrow) at the femoral head on day 14 after AdMuOSM injection. F, No chondrophyte formation is observed at the femoral head on day 14 after AdDL70-3 injection. F = femur; P = patella; S = inflamed synovium. (Safranin O stained [except in D]; original magnification × 50 in A, B, and D;× 200 in C, E, and F.)

Download figure to PowerPoint

Growth plate damage after OSM gene transfer.

A unique feature of OSM gene transfer that has not been previously reported is that the AdMuOSM vector also caused damage to the growth plate cartilage. In all wild-type mice studied, we observed PG depletion and loss of matrix integrity in the growth plates (Figure 2A) adjacent to the periosteum. The PG content in the growth plates recovered after the first week of inflammation, but the matrix integrity and the normal arrangement of chondrocytes were not restored. This growth plate damage was not observed after injection of the control vector AdDL70-3 (Figure 2B). It also was not observed after local gene transfer of IL-1 or IL-17 (Figures 2C and D), although both induced joint inflammation and articular cartilage PG depletion (Joosten LAB, Koenders MI: unpublished observations). These observations exclude the possibility that the damage occurred as a general consequence of local proinflammatory cytokine overexpression.

thumbnail image

Figure 2. Growth plate damage induced by the adenoviral murine oncostatin M vector (AdMuOSM) in wild-type mice. A, Proteoglycan depletion and disorganization of the growth plate on day 7 after injection of AdMuOSM. B, Normal growth plate on day 7 after injection of AdDL70-3. Note also the absence of inflammation after AdDL70-3 injection. C, Normal growth plate after injection of 2.5 × 104 NIH3T3 cells expressing human interleukin-1β. D, Normal growth plate after injection of the adenoviral murine interleukin-17 vector. (Safranin O stained; original magnification × 200.)

Download figure to PowerPoint

Cytokine dependence of OSM-induced joint pathology.

RT-PCR revealed synovial expression of OSM gene in the knee joint injected with AdMuOSM, but not in that injected with AdDL70-3 (Figure 3A). On day 3, the first OSM PCR product was detected 2 cycles after detection of the first GAPDH PCR product. On day 7, it was detected a mean (±SD) of 5 ± 1 cycles after detection of the first GAPDH PCR product, indicating a decrease in OSM gene expression. OSM can induce expression of IL-6 (36) and can enhance the effects of other proinflammatory cytokines such as IL-1 and TNFα (3). Semiquantitative RT-PCR analysis showed up-regulated expression of mRNA for IL-1β, IL-6, and TNFα associated with the AdMuOSM-induced inflammation (Figure 3B). Because these cytokines can have great influence on joint inflammation, chondrocyte metabolism, and cartilage damage, we determined their role in the AdMuOSM-induced joint disease by injecting 2 × 106 PFU AdMuOSM into the knee joints of mice deficient for IL-1, IL-6, or TNFα. Mice deficient for the iNOS gene also received the injection. This factor (iNOS) plays a role in the suppression of cartilage PG synthesis and was also up-regulated at the mRNA level (Figure 3B).

thumbnail image

Figure 3. Oncostatin M (OSM) and cytokine gene expression during adenoviral murine OSM (AdMuOSM)–induced inflammation. A, OSM gene expression in synovium 3 days after injection of AdMuOSM, as detected by semiquantitative reverse transcription–polymerase chain reaction (RT-PCR). Results represent the linear part of the PCR reaction (26 PCR cycles for OSM and 22 cycles for GAPDH). B, Enhanced gene expression for interleukin-1β (IL-1β), IL-6, tumor necrosis factor α (TNFα), and inducible nitric oxide synthase (iNOS) on day 3 after injection of the adenoviral vector. Gene expression was compared between synovia from AdMuOSM-injected and contralateral AdDL70-3–injected knee joints in 4 mice, as described in Patients and Methods. The control vector in the contralateral knee joint served as a within-animal control. Gene expression was examined in 2 groups of 2 pooled synovia. Equal amounts of complementary DNA were used, as assessed by PCR for GAPDH. Semiquantitative RT-PCR was performed at least twice.

Download figure to PowerPoint

IL-1 was observed to play an important role during the first week of AdMuOSM-induced joint pathology. Histologic scoring showed a reduced synovial infiltrate on day 7 after injection of AdMuOSM in IL-1α/β–deficient mice (Table 2). The influence of IL-1 on joint inflammation decreased after day 7, and by day 14, inflammation in wild-type and IL-1α/β–deficient mice did not differ (Table 2). Deficiency for TNFα, IL-6, and iNOS did not affect AdMuOSM-induced joint inflammation. Inflammation in the TNFα-deficient mice tended to be reduced on day 7 after injection, but this difference versus wild-type mice did not reach statistical significance (P = 0.05).

Table 2. AdMuOSM-induced joint pathology in wild-type and cytokine-deficient mice*
MiceInflammationPG depletionRelative PG synthesis, %, day 4Chondrophyte incidence, day 14§Bone apposition, μm2 of new bone/ 10 μm of cortical bone, day 14Growth plate damage incidence, day 7§
Day 7Day 14Day 7Day 14
  • *

    Data shown are for 1 representative experiment with 5–9 mice per group. Except where indicated otherwise, values are the mean ± SD. IL-1α/β = interleukin-1α/β; TNFα = tumor necrosis factor α; iNOS = inducible nitric oxide synthase.

  • Scored on a 0–3-point scale. Data for proteoglycan (PG) depletion are for patellar cartilage; similar results (not shown) were obtained for femoral cartilage.

  • Ex vivo patellar PG synthesis was compared with synthesis in the patella of the uninjected contralateral knee joint. Values >100% indicate increased PG synthesis in the patella from the adenoviral murine oncostatin M (AdMuOSM)–injected knee joint. Injection of AdDL70-3 induced only a slight increase of PG synthesis (mean ± SD 111 ± 14.9%). Six to 8 patellae per group were used, and PG synthesis was measured twice.

  • §

    Percentage of mice.

  • P < 0.05 versus wild-type mice, by rank sum test.

  • #

    P < 0.005 versus wild-type mice, by rank sum test.

Wild-type1.7 ± 0.60.9 ± 0.41.8 ± 0.61.9 ± 0.6142 ± 34.888338 ± 47100
IL-1α/β−/−1.0 ± 0.20.7 ± 0.40.5 ± 0.3#2.0 ± 0.7167 ± 10.3100317 ± 410
TNFα−/−1.0 ± 0.51.0 ± 0.61.4 ± 0.51.5 ± 0.4NM100374 ± 74100
IL-6−/−1.5 ± 0.51.0 ± 0.62.3 ± 0.52.4 ± 0.4NM100312 ± 76100
iNOS−/−1.8 ± 0.71.1 ± 0.41.6 ± 1.12.2 ± 1.2NM80355 ± 62100

Cartilage PG depletion was also significantly reduced in the IL-1α/β–deficient mice during the first week of inflammation (Table 2). Ex vivo PG synthesis was increased after injection of AdMuOSM into wild-type mice (Table 2). This suggests that the observed PG depletion in wild-type mice is caused by enhanced PG breakdown. The fact that the ex vivo PG synthesis increased further in IL-1α/β–deficient mice suggests that endogenous IL-1 can at least partly counteract OSM-induced stimulation of PG synthesis. Deficiency for TNFα, IL-6, and iNOS did not affect the AdMuOSM-induced PG loss on days 7 and 14 after injection of the vector.

By day 14, PG loss in the IL-1α/β–deficient mice had developed to the same extent as that in the wild-type mice (Table 2), with marked expression of the matrix metalloproteinase (MMP)–generated VDIPEN neoepitope (Figures 4A and C). In contrast, the AdDL70-3 control vector did not lead to generation of the VDIPEN neoepitope (Figure 4B).

thumbnail image

Figure 4. VDIPEN neoepitope staining in cartilage after injection of AdMuOSM. A, Positive VDIPEN staining in the cartilage of a wild-type mouse 14 days after AdMuOSM injection. Staining is detected in the matrix surrounding the chondrocytes at the surface and in the deeper cartilage layers. B, Negative VDIPEN staining in a wild-type mouse 14 days after injection with AdDL70-3. C, Positive VDIPEN staining in an interleukin-1α/β–deficient mouse 14 days after injection of AdMuOSM. D, Negative staining in a wild-type mouse 14 days after injection of AdMuOSM, when preimmune serum was used. (Original magnification × 760.)

Download figure to PowerPoint

Cytokine dependence of OSM-induced growth plate damage.

The endogenous role of IL-1 was even more prominent in the AdMuOSM-induced growth plate damage. No loss of PGs or disruption of the matrix integrity was found in the growth plates of AdMuOSM-treated IL-1α/β–deficient mice (Table 2). In contrast to the IL-1α/β–deficient mice, growth plate damage did develop in all mice deficient for TNFα, IL-6, or iNOS (Table 2). Similar to the situation in wild-type mice, the PG content of the affected growth plate was restored on day 14 of inflammation in mice deficient for TNFα, IL-6, and iNOS.

OSM in synovial fluid of patients with JIA.

The OSM-induced joint disease in the mice resembled the arthritic changes that are observed in human arthropathies such as RA and JIA. For this reason, we were interested in determining whether OSM could be detected in JIA synovial fluid, as was already demonstrated for RA synovial fluid (2, 3). Indeed, we could measure OSM in 77% of the JIA synovial fluid samples that were examined by ELISA (Table 1). Furthermore, 70% of the OSM-positive samples were also positive for IL-1β (Table 1).


  1. Top of page
  2. Abstract
  6. Acknowledgements

OSM is produced in the inflamed joints of patients with RA (3), and results of several in vitro experiments suggest that OSM could play an important role in cartilage damage in RA. OSM induced collagen release from bovine cartilage explants (12) and, in combination with IL-1, from human cartilage explants (3). Furthermore, OSM stimulated PG release and suppressed PG synthesis in porcine articular cartilage (13). In the present study, we used an adenoviral vector expressing murine OSM to investigate in vivo the influence of proinflammatory mediators, which are important for RA, on OSM-induced joint pathology.

The AdMuOSM vector induced inflammation, cartilage PG depletion, periosteal bone apposition, and chondrophyte formation in the joints of naive mice. Semiquantitative RT-PCR analysis showed increased expression of mRNA for IL-6, TNFα, IL-1β, and iNOS in the AdMuOSM-injected knee joint. A relationship with OSM-induced pathology was investigated in mice deficient for these proinflammatory factors.

OSM is a strong inducer of IL-6 gene expression. We had previously observed that AdMuOSM-induced inflammation was not inhibited by IL-6 deficiency (6). The present results demonstrate that cartilage PG depletion is also not affected in these mice. In contrast to the important role of IL-6 in experimental arthritis (21, 37), the present results do not indicate such a role for IL-6 in AdMuOSM-induced joint disease. A positive correlation between TNFα and OSM concentrations has been demonstrated in synovial fluid obtained from patients with RA (14). However, whether there is a direct relationship between these cytokines was, until now, not clear. Results of the present study show that cartilage PG depletion induced by AdMuOSM was not affected by TNFα deficiency. Although inflammation in TNFα-deficient mice tended to be reduced on day 7, by day 14 inflammation in these mice did not differ from that in wild-type mice. This is consistent with reports showing that TNFα is important during disease onset, but that TNFα deficiency or inhibition did not prevent development of severe arthritis in experimental models (38, 39). Taken together, our results do not suggest an important role for TNFα in OSM-induced joint pathology.

Nitric oxide is produced in the inflamed joints of patients with RA (40) and can contribute to IL-1–induced inhibition of PG synthesis (16, 41). We previously demonstrated that cartilage PG depletion, but not joint inflammation, is significantly reduced in iNOS-deficient mice with zymosan-induced arthritis (18). In the present experiments, PG depletion did not differ between iNOS-deficient and wild-type mice, indicating that NO formation is not essential for AdMuOSM-induced cartilage damage.

Joint inflammation and cartilage PG depletion in IL-1α/β–deficient mice were significantly reduced on day 7. This shows an important role for IL-1 in AdMuOSM-induced joint pathology. IL-1 has been shown to be a key factor in the development of experimental arthritis (10, 42). Both inhibition of PG synthesis and stimulation of PG breakdown can induce PG depletion in arthritis, and IL-1 can be involved in both processes (11, 43). In our experiments, ex vivo PG synthesis in patellae from AdMuOSM-injected joints was not inhibited, but rather was increased. This excess in PG synthesis, however, could not prevent articular cartilage PG loss. Breakdown of PGs would, therefore, be the main cause of the observed PG loss in wild-type mice. In the IL-1α/β–deficient mice, ex vivo PG synthesis was even further increased, and this could (at least in part) contribute to the reduced PG loss in these mice. The increased PG synthesis in the IL-1α/β–deficient mice furthermore indicates that in wild-type mice, IL-1 will partly inhibit the elevation of PG synthesis. We previously observed that blocking of IL-1 in experimental arthritis completely prevented inhibition of PG synthesis but did not influence inflammation-induced PG breakdown (15).

Bell et al (4) reported that coinjecting human OSM with recombinant human IL-1 receptor antagonist (IL-1Ra) into the joints of goats could not attenuate OSM-induced cartilage PG depletion. In a previous study, we observed that prolonged high concentrations of IL-1Ra were necessary to prevent IL-1–induced inhibition of PG synthesis in antigen-induced arthritis. These concentrations could be achieved with mini–osmotic pumps but not by bolus injection of IL-1Ra (15). The negative results described by Bell et al could therefore be attributable to poor pharmacokinetics of IL-1Ra in the joint. In the present study, we used IL-1α/β–deficient mice to circumvent these problems and observed clear involvement of IL-1 in the OSM-induced PG loss that occurred during the first week of inflammation.

We have previously shown that repeated injections of IL-1 induce inflammation and cartilage damage in the murine knee joint (11). In that study, the polymorphonuclear cell was the predominant cell type in the inflammatory infiltrate. A role for PMNs in cartilage damage has been shown in vitro (44) and in vivo (45). Recently, OSM was shown to selectively recruit PMNs in an in vitro flow chamber assay (46). Using NIMP-R14 staining, we could detect PMNs in the inflamed synovium of both wild-type and IL-1α/β–deficient mice (results not shown), suggesting that IL-1 is not necessary for OSM-induced PMN influx. This, however, does not exclude a relationship between IL-1 and PMNs in the observed PG depletion. Activation of PMNs might differ between wild-type and IL-1α/β–deficient mice; this requires further investigation.

During AdMuOSM-induced inflammation, irreversible damage to the PG network occurred, as demonstrated by the presence of the MMP-induced VDIPEN neoepitope. Expression of VDIPEN was shown to correlate with severe cartilage damage in murine arthritis (47). The VDIPEN neoepitope was also detected in cartilage from IL-1α/β–deficient mice, indicating that irreversible PG damage can occur independent of IL-1. Future research is needed to identify the enzyme that is responsible for this irreversible damage and its relationship to OSM.

Periosteal bone apposition and chondrophyte formation were induced in both wild-type and gene-deficient mice. Periosteal bone apposition can occur in the short tubular bones of the phalanges, metacarpals, and metatarsals, and also in the long bones during JIA (48, 49). To our knowledge, little is known about the significance of periosteal bone apposition in JIA. Bone apposition was not induced by overexpression of IL-1 or IL-17 (data not shown), which excludes the possibility that bone apposition was a general consequence of inflammation. We previously had observed that OSM could enhance in vitro the bone morphogenetic protein 2–induced differentiation of C2C12 cells toward the osteoblastic lineage (6). This suggests that OSM could play a positive regulatory role during bone formation by enhancing the activity of bone-forming factors. Chondrophyte and osteophyte formation is common in osteoarthritis (OA). Osteophytes can also develop in RA and JIA with secondary OA, but this happens less frequently.

In general, adenovirally mediated gene transfer to the joint results in a transient transgene expression (50, 51) lasting from 1 to 2 weeks. We observed that most of the changes in the murine knee joint had already developed during the first week, when OSM gene expression was demonstrated. The involvement of IL-1 provides circumstantial evidence for a relationship between transgene expression and the observed joint pathology. This was evident on day 7 but not on day 14. After day 7, OSM-induced inflammation subsided, and the growth plate PG content returned to normal levels. During the first week, periosteal activation, leading to bone apposition on day 14, also took place. Thereafter, the process of new bone formation did not proceed. Surprisingly, articular cartilage PG depletion continued after day 7. This is probably not directly related to OSM activity but could be a result of irreversible cartilage damage, delayed repair mechanisms, or morphologic changes (e.g., chondrophyte formation), which could influence cartilage integrity.

A unique finding associated with injection of AdMuOSM is that the growth plate became damaged. This was not observed with vectors expressing either IL-1 or IL-17, although both induced articular cartilage damage. Such growth plate damage has not been previously observed in experimental arthritis in mice of the same age. This process was demonstrated to be dependent on endogenous IL-1. In growth plates, there is a balance between cartilage matrix degradation, proliferation, matrix formation, and hypertrophy. Expression of IL-1 mRNA has been detected in the growth plate of developing bones in mice (52). In vitro results of studies using growth plate chondrocytes from the rat suggested that IL-1 induces resting growth cells to acquire a phenotype of growth zone cells in an autocrine manner (53). Furthermore, IL-1 could play a role in the bone and cartilage resorption processes that occur in the growth plate during the formation of new bone. OSM could either enhance or modify the autocrine effects of IL-1 on growth plate chondrocytes, thereby leading to growth plate PG loss, disorganization, and finally growth abnormalities.

Growth plate changes in patients with JIA have been reported. Magnetic resonance imaging studies have shown epiphyseal cartilage loss in the knees of patients with JIA (23), and in unilateral juvenile arthritis the femoral epiphysis of the arthritic side was observed to be enlarged (24). Although IL-6 is found in elevated concentrations in serum and synovial fluid in JIA (54), the presence of OSM has, as far as we know, not been investigated in these patients. Using ELISA techniques, we detected OSM in synovial fluid of most of the examined children, and we could also detect IL-1β in most of our OSM-positive samples. Our experiments in the cytokine-deficient mice indicated that OSM, in the presence of IL-1, could cause serious risks to the integrity of growth plate cartilage. It is possible that the combination of these cytokines is similarly involved in growth plate damage in JIA.

Both increased growth and growth retardation occur frequently in JIA (25, 55). Most of our synovial fluid samples were obtained from patients with oligoarthritis who had involvement of the knee joint. A study by Simon et al (56) showed a relationship between age at disease onset, involvement of the knee, and localized growth abnormalities in oligoarthritis (formerly called monoarticular and pauciarticular RA). In patients in whom JIA began before age 9 years, the involved side was the longer one. Disease onset after this age led to rapid premature closure of the growth plate and shortening of the involved side. Among the positive samples in our study, the highest concentrations of OSM were found in those obtained from the younger children, which could implicate a role for OSM in increased growth of the involved side. This is further supported by the finding that a mouse transgenic for bovine OSM had enlarged hind limbs (22). We recently began collaborations in order to increase the number of synovial fluid samples available for study from patients with the different forms of JIA. We hope that this will also enable us to further characterize OSM expression during the time course of the disease.

In conclusion, our results demonstrate an important role for endogenous IL-1 in AdMuOSM-induced joint pathology, but no involvement of TNFα, IL-6, or iNOS. The induction of growth plate damage in mice adds a newly recognized pathologic consequence of OSM expression that would be particularly relevant in JIA. The AdMuOSM vector provides a useful tool to further investigate this process in more detail. Our results in the cytokine-deficient mice and the detection of OSM in synovial fluid of patients with JIA suggest that the proinflammatory and cartilage-damaging effects of OSM are relevant in human arthropathies such as RA and JIA.


  1. Top of page
  2. Abstract
  6. Acknowledgements

We thank Astrid Holthuysen for the VDIPEN staining and Dr. P. Schwarzenberger (New Orleans, LA) for the AdmIL-17 vector.


  1. Top of page
  2. Abstract
  6. Acknowledgements
  • 1
    Rose TM, Bruce AG. Oncostatin M is a member of a cytokine family that includes leukemia-inhibitory factor, granulocyte colony-stimulating factor, and interleukin 6. Proc Natl Acad Sci U S A 1991; 88: 86415.
  • 2
    Okamoto H, Yamamura M, Morita Y, Harada S, Makino H, Ota Z. The synovial expression and serum levels of interleukin-6, interleukin-11, leukemia inhibitory factor, and oncostatin M in rheumatoid arthritis. Arthritis Rheum 1997; 40: 10961105.
  • 3
    Cawston TE, Curry VA, Summers CA, Clark IM, Riley GP, Life PF, et al. The role of oncostatin M in animal and human connective tissue collagen turnover and its localization within the rheumatoid joint. Arthritis Rheum 1998; 41: 176071.
  • 4
    Bell MC, Carroll GJ, Chapman HM, Mills JN, Hui W. Oncostatin M induces leukocyte infiltration and cartilage proteoglycan degradation in vivo in goat joints. Arthritis Rheum 1999; 42: 254351.
  • 5
    Langdon C, Kerr C, Hassen M, Hara T, Arsenault AL, Richards CD. Murine oncostatin M stimulates mouse synovial fibroblasts in vitro and induces inflammation and destruction in mouse joints in vivo. Am J Pathol 2000; 157: 118796.
  • 6
    De Hooge AS, van de Loo FA, Bennink MB, de Jong DS, Arntz OJ, Lubberts E, et al. Adenoviral transfer of murine oncostatin M elicits periosteal bone apposition in knee joints of mice, despite synovial inflammation and up-regulated expression of interleukin-6 and receptor activator of nuclear factor-kappaB ligand. Am J Pathol 2002; 160: 173343.
  • 7
    Yao L, Pan J, Setiadi H, Patel KD, McEver RP. Interleukin 4 or oncostatin M induces a prolonged increase in P-selectin mRNA and protein in human endothelial cells. J Exp Med 1996; 184: 8192.
  • 8
    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: 15868.
  • 9
    Langdon C, Leith J, Smith F, Richards CD. Oncostatin M stimulates monocyte chemoattractant protein-1– and interleukin-1–induced matrix metalloproteinase 1 production by human synovial fibroblasts in vitro. Arthritis Rheum 1997; 40: 213946.
  • 10
    Pettipher ER, Higgs GA, Henderson B. Interleukin 1 induces leukocyte infiltration and cartilage proteoglycan degradation in the synovial joint. Proc Natl Acad Sci U S A 1986; 83: 874953.
  • 11
    Van de Loo AA, van den Berg WB. Effects of murine recombinant interleukin 1 on synovial joints in mice: measurement of patellar cartilage metabolism and joint inflammation. Ann Rheum Dis 1990; 49: 23845.
  • 12
    Cawston TE, Ellis AJ, Humm G, Lean E, Ward D, Curry V. Interleukin-1 and oncostatin M in combination promote the release of collagen fragments from bovine nasal cartilage in culture. Biochem Biophys Res Commun 1995; 215: 37785.
  • 13
    Hui W, Bell M, Carroll G. Oncostatin M (OSM) stimulates resorption and inhibits synthesis of proteoglycan in porcine articular cartilage explants. Cytokine 1996; 8: 495500.
  • 14
    Manicourt DH, Poilvache P, van Egeren A, Devogelaer JP, Lenz ME, Thonar EJ. Synovial fluid levels of tumor necrosis factor α and oncostatin M correlate with levels of markers of the degradation of crosslinked collagen and cartilage aggrecan in rheumatoid arthritis but not in osteoarthritis. Arthritis Rheum 2000; 43: 2818.
  • 15
    Van de Loo FA, Joosten LA, van Lent PL, Arntz OJ, van den Berg WB. Role of interleukin-1, tumor necrosis factor α, and interleukin-6 in cartilage proteoglycan metabolism and destruction: effect of in situ blocking in murine antigen- and zymosan-induced arthritis. Arthritis Rheum 1995; 38: 16472.
  • 16
    Van de Loo FA, Arntz OJ, Otterness IG, van den Berg WB. Protection against cartilage proteoglycan synthesis inhibition by antiinterleukin 1 antibodies in experimental arthritis. J Rheumatol 1992; 19: 34856.
  • 17
    Taskiran D, Stefanovic-Racic M, Georgescu H, Evans C. Nitric oxide mediates suppression of cartilage proteoglycan synthesis by interleukin-1. Biochem Biophys Res Commun 1994; 200: 1428.
  • 18
    Van de Loo FA, Arntz OJ, van Enckevort FH, van Lent PL, van den Berg WB. Reduced cartilage proteoglycan loss during zymosan-induced gonarthritis in NOS2-deficient mice and in anti–interleukin-1–treated wild-type mice with unabated joint inflammation. Arthritis Rheum 1998; 41: 63446.
  • 19
    Joosten LA, Helsen MM, van de Loo FA, van den Berg WB. Anticytokine treatment of established type II collagen–induced arthritis in DBA/1 mice: a comparative study using anti-TNFα, anti–IL-1α/β, and IL-1Ra. Arthritis Rheum 1996; 39: 797809.
  • 20
    Joosten LA, Helsen MM, Saxne T, van de Loo FA, Heinegard D, van den Berg WB. IL-1 alpha beta blockade prevents cartilage and bone destruction in murine type II collagen-induced arthritis, whereas TNF-alpha blockade only ameliorates joint inflammation. J Immunol 1999; 163: 504955.
  • 21
    De Hooge AS, van de Loo FA, Arntz OJ, van den Berg WB. Involvement of IL-6, apart from its role in immunity, in mediating a chronic response during experimental arthritis. Am J Pathol 2000; 157: 208191.
  • 22
    Malik N, Haugen HS, Modrell B, Shoyab M, Clegg CH. Developmental abnormalities in mice transgenic for bovine oncostatin M. Mol Cell Biol 1995; 15: 234958.
  • 23
    Senac MO Jr, Deutsch D, Bernstein BH, Stanley P, Crues JV III, Stoller DW, et al. MR imaging in juvenile rheumatoid arthritis. AJR Am J Roentgenol 1988; 150: 8738.
  • 24
    Sundberg J, Brattstrom M. Juvenile rheumatoid gonarthritis. II. Disturbance of ossification and growth. Acta Rheumatol Scand 1965; 11: 27990.
  • 25
    White PH. Growth abnormalities in children with juvenile rheumatoid arthritis. Clin Orthop 1990; Oct: 4650.
  • 26
    Yamada H, Mizumo S, Horai R, Iwakura Y, Sugawara I. Protective role of interleukin-1 in mycobacterial infection in IL-1 alpha/beta double-knockout mice. Lab Invest 2000; 80: 75967.
  • 27
    Kopf M, Baumann H, Freer G, Freudenberg M, Lamers M, Kishimoto T, et al. Impaired immune and acute-phase responses in interleukin-6-deficient mice. Nature 1994; 368: 33942.
  • 28
    Pasparakis M, Alexopoulou L, Episkopou V, Kollias G. Immune and inflammatory responses in TNF alpha-deficient mice: a critical requirement for TNF alpha in the formation of primary B cell follicles, follicular dendritic cell networks and germinal centers, and in the maturation of the humoral immune response. J Exp Med 1996; 184: 13971411.
  • 29
    MacMicking JD, Nathan C, Hom G, Chartrain N, Fletcher DS, Trumbauer M, et al. Altered responses to bacterial infection and endotoxic shock in mice lacking inducible nitric oxide synthase. Cell 1995; 81: 64150.
  • 30
    Kerr C, Langdon C, Graham F, Gauldie J, Hara T, Richards CD. Adenovirus vector expressing mouse oncostatin M induces acute-phase proteins and TIMP-1 expression in vivo in mice. J Interferon Cytokine Res 1999; 19: 11951205.
  • 31
    Schwarzenberger P, La Russa V, Miller A, Ye P, Huang W, Zieske A, et al. IL-17 stimulates granulopoiesis in mice: use of an alternate, novel gene therapy-derived method for in vivo evaluation of cytokines. J Immunol 1998; 161: 63839.
  • 32
    McLaren DJ, Strath M, Smithers SR. Schistosoma mansoni: evidence that immunity in vaccinated and chronically infected CBA/Ca mice is sensitive to treatment with a monoclonal antibody that depletes cutaneous effector cells. Parasite Immunol 1987; 9: 66782.
  • 33
    Van Meurs JB, van Lent PL, Joosten LA, van der Kraan PM, van den Berg WB. Quantification of mRNA levels in joint capsule and articular cartilage of the murine knee joint by RT-PCR: kinetics of stromelysin and IL-1 mRNA levels during arthritis. Rheumatol Int 1997; 16: 197205.
  • 34
    Joosten LA, Lubberts E, Durez P, Helsen MM, Jacobs MJ, Goldman M, et al. Role of interleukin-4 and interleukin-10 in murine collagen-induced arthritis: protective effect of interleukin-4 and interleukin-10 treatment on cartilage destruction. Arthritis Rheum 1997; 40: 24960.
  • 35
    Petty RE, Southwood TR, Baum J, Bhettay E, Glass DN, Manners P, et al. Revision of the proposed classification criteria for juvenile idiopathic arthritis: Durban, 1997. J Rheumatol 1998; 25: 19914.
  • 36
    Brown TJ, Rowe JM, Liu JW, Shoyab M. Regulation of IL-6 expression by oncostatin M. J Immunol 1991; 147: 217580.
  • 37
    Alonzi T, Fattori E, Lazzaro D, Costa P, Probert L, Kollias G, et al. Interleukin 6 is required for the development of collagen-induced arthritis. J Exp Med 1998; 187: 4618.
  • 38
    Campbell IK, O'Donnell K, Lawlor KE, Wicks IP. Severe inflammatory arthritis and lymphadenopathy in the absence of TNF. J Clin Invest 2001; 107: 151927.
  • 39
    Van den Berg WB. Lessons from animal models of arthritis. Curr Rheumatol Rep 2002; 4: 2329.
  • 40
    Sakurai H, Kohsaka H, Liu MF, Higashiyama H, Hirata Y, Kanno K, et al. Nitric oxide production and inducible nitric oxide synthase expression in inflammatory arthritides. J Clin Invest 1995; 96: 235763.
  • 41
    Presle N, Cipolletta C, Jouzeau JY, Abid A, Netter P, Terlain B. Cartilage protection by nitric oxide synthase inhibitors after intraarticular injection of interleukin-1β in rats. Arthritis Rheum 1999; 42: 2094102.
  • 42
    Van den Berg WB, Joosten LA, Helsen M, van de Loo FA. Amelioration of established murine collagen-induced arthritis with anti-IL-1 treatment. Clin Exp Immunol 1994; 95: 23743.
  • 43
    Steinberg JJ, Hubbard JR, Sledge CB. Chondrocyte-mediated breakdown of cartilage. J Rheumatol 1987 May;14 Spec No: 558.
  • 44
    Moore AR, Iwamura H, Larbre JP, Scott DL, Willoughby DA. Cartilage degradation by polymorphonuclear leucocytes: in vitro assessment of the pathogenic mechanisms. Ann Rheum Dis 1993; 52: 2731.
  • 45
    Van Lent PL, van den Hoek AE, van den Bersselaar LA, van de Loo FA, Eykholt HE, Brouwer WF, et al. Early cartilage degradation in cationic immune complex arthritis in mice: relative role of interleukin 1, the polymorphonuclear cell (PMN) and PMN elastase. J Rheumatol 1994; 21: 3219.
  • 46
    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: 15319.
  • 47
    Van Meurs JB, van Lent PL, Holthuysen AE, Singer II, Bayne EK, van den Berg WB. Kinetics of aggrecanase- and metalloproteinase-induced neoepitopes in various stages of cartilage destruction in murine arthritis. Arthritis Rheum 1999; 42: 112839.
  • 48
    Resnick D, Sartoris D, Cone RO. Diagnostic tests and procedures in rheumatic disease. In: KelleyWN, HarrisEDJr, RuddyS, SledgeCB, editors. Textbook of rheumatology. Philadelphia: WB Saunders; 1989. p. 650708.
  • 49
    Martel W, Holt JF, Cassidy JT. Roentgenologic manifestations of juvenile rheumatoid arthritis. Am J Roentgenol 1962; 88: 40023.
  • 50
    Evans CH, Ghivizzani SC, Oligino TA, Robbins PD. Future of adenoviruses in the gene therapy of arthritis. Arthritis Res 2001; 3: 1426.
  • 51
    Lubberts E, Joosten LA, Oppers B, van den Bersselaar L, Coenen-De Roo CJ, Kolls JK, et al. IL-1-independent role of IL-17 in synovial inflammation and joint destruction during collagen-induced arthritis. J Immunol 2001; 167: 100413.
  • 52
    Takacs L, Kovacs EJ, Smith MR, Young HA, Durum SK. Detection of IL-1 alpha and IL-1 beta gene expression by in situ hybridization: tissue localization of IL-1 mRNA in the normal C57BL/6 mouse. J Immunol 1988; 141: 308195.
  • 53
    Horan J, Dean DD, Kieswetter K, Schwartz Z, Boyan BD. Evidence that interleukin-1, but not interleukin-6, affects costochondral chondrocyte proliferation, differentiation, and matrix synthesis through an autocrine pathway. J Bone Miner Res 1996; 11: 111929.
  • 54
    Lepore L, Pennesi M, Saletta S, Perticarari S, Presani G, Prodan M. Study of IL-2, IL-6, TNF alpha, IFN gamma and beta in the serum and synovial fluid of patients with juvenile chronic arthritis. Clin Exp Rheumatol 1994; 12: 5615.
  • 55
    Bernstein BH, Stobie D, Singsen BH, Koster-King K, Kornreich HK, Hanson V. Growth retardation in juvenile rheumatoid arthritis (JRA). Arthritis Rheum 1977; 20 Suppl 2: 2126.
  • 56
    Simon S, Whiffen J, Shapiro F. Leg-length discrepancies in monoarticular and pauciarticular juvenile rheumatoid arthritis. J Bone Joint Surg Am 1981; 63: 20915.