The contribution of reactive oxygen species (ROS), in particular hydroxyl radical (OH·), to joint inflammation was examined in rats developing adjutant arthritis (AA) by treatment with ROS scavengers dimethylthiourea (DMTU) and DMSO. Adjutant arthritis was induced in Sprague-Dawley (SD) rats by a single intradermal (i.d.) injection of Mycobacterium tuberculosis (MT) in oil on day O. By day 14, all rats exhibited arthritis in the hind limbs and the majority had involvement of the forelimbs. A marked inflammatory cell influx (75% neutrophils) was present in the synovial fluid. These cells, in vitro, spontaneously produced OH· (0.96% ± 0.28 OH· units/h per 10⁵ cells). In contrast, spontaneous OH· production by normal circulating leucocytes was absent (0.07 ± 0.03 OH· units/h per 10⁵ cells). Adjuvant-injected rats were treated with DMTU (500, 250 and 100 mg/kg), DMSO (330 and 165 mg/kg), or saline (disease control) once daily on days 8, 9, and 10 and twice daily on days 11, 12, and 13 postadjuvant injection. Both DMTU and DMSO significantly reduced the clinical evidence of arthritis (clinical scores: DMTU [500 mg/kg] = 0, P < 0.0001; DMSO [3.0 mL/kg] = 0.4 ± 0.3, P < 0.01, compared with disease control = 2.3 ± 0.3). Synovial fluid cell accumulation was also significantly reduced (DMTU [500 mg/kg] = 0.5 ± 0.1 x 10⁵ cells/four joints, P < 0.0001; DMSO [3.0 mL/kg] 2.75 ± 1.9 x 10⁵ cells/four joints, P < 0.01 compared with disease control = 11.76 ± 1.7 x 10⁵ cells/four/joints). Neither treatment inhibited cutaneous delayed type hypersensitivity (DTH) to the disease inducing antigen. Furthermore, DMTU (550 mg/kg) did not cause neutropenia nor inhibit peritoneal neutrophil accumulation in response to a chemotactic stimulus. This study demonstrates the attenuation of adjutant arthritis by ROS scavengers and suggests a pivotal role for ROS, particularly OH·, in the mediation of joint inflammation in this disease.

Considerable evidence implicates reactive oxygen species (ROS) as mediators of inflammation and tissue destruction in inflammatory disorders, such as rheumatoid arthritis (RA) and adjuvant arthritis (AA). ^{1, 2} Histological studies in AA have demonstrated a pronounced mononuclear inflammatory cell infiltrate in particular tissues. ³ In the synovial fluid of patients with RA and rats with AA the most prominent inflammatory cell is the neutrophil. ⁴ The products of these cells have direct access to the synovium and articular cartilage and may be important mediators of injury to these components of the joint.

In vitro studies have demonstrated that activated neutrophils produce a number of ROS including hydrogen peroxide (H₂O₂) and superoxide anion (O₂·¯) Both O₂·¯ and H₂O₂ can react with a number of biological substrates; however, the ability of neutrophils to use (O₂·¯ and H₂O₂ directly to mediate extracellular effects is limited in part by the poor reactivity of these radicals in aqueous solutions. ^{5, 6} Under physiological conditions, (O₂·¯ rapidly dissimulates to produce (O₂·^­ and H₂O₂. Although H₂O₂ is also a reactive oxidant, neutrophils consume most of this metabolite to produce other radicals. ⁶ In the Haber-Weiss reaction, a more injurious radical, the hydroxyl radical (OH·) is produced from O₂·¯ and H₂O₂ using a metal ion, such as iron, as catalyst. ⁵ In the presence of the enzyme myeloperoxidase (MPO), H₂0₂ is used to oxidize Cl¯ ions into hypochlorous acid (HOCl), another powerful oxidant. Hypochlorous acid can inactivate antiproteinases allowing for uncontrolled proteinase activity. ⁶

Hydroxyl radical production by neutrophils is evidenced in a number of studies using various detection systems including the decomposition of DMSO to methane,⁷ spin resonance/spin trapping system⁸ and by hydroxylation of salicylic acid. ⁹ These detection systems are indirect measures of OH· concentration and rely on oxidation of indicator compounds to more stable products. Evidence for an important role of OH· in neutrophil mediated tissue injury has been demonstrated in models of mucosal injury in rat stomach,¹⁰ pulmonary⁷ and renal injury in rats,¹¹ and acute glomerulonephritis in rabbits. ⁹ These studies involved the use of various ROS scavengers including dimethylthiourea (DMTU), desferrioxamine(DFX), and superoxide dismutase (SOD). ^7,9 - 11. In the current study the effect of treatment with highly diffusable ROS scavengers, DMSO and DMTU in joint injury in AA was investigated.

Induction of adjuvant arthritis

All studies were performed in outbred male Sprague-Dawley (SD) rats (170-200 g). Adjuvant was prepared by suspending 200 mg. of Mycobacterium tuberculosis (MT: H37 RA: Difco Laboratories, Detroit, MI, USA) in 5 ml absolute ethanol and 5 ml diethyl ether. This was mixed for 30 min on a rotating wheel, centrifuged 5 min at 800 g and the supernatant discarded. The resultant pellet was dried at 37°C for 3 h. The delipidated MT was ground with a mortar and pestle. Squalane (saturated hydrocarbon: Sigma Chemical Co., St Louis, MO, USA) was gradually added, during grinding to give a final concentration of 10 mg/ml. This suspension was sonicated (Sonifier Cell Disruptor B-30: Consolidated Ultrasonics, Melbourne, Vic., Aust.) for 30 s immediately prior to each usage. On day 0, test animals were injected intradermally (i.d.) at the base of the tail with 0.15 ml of adjuvant. Control animals were injected similarly at the base of the tail with 0.15 ml squalane.

Treatment protocols

DMTL treatment    1.3-Dimethyl-2-thiourea (Aldrich Chem. Co. Inc., WI, USA) was dissolved in distilled water at final concentrations of 250, 125, and 50 mg/ml⁷ Treated animals received i.p. injections of 500, 250, or 100 mg/kg once daily on 8, 9, and 10 and twice daily on days 11, 12 and 13 postadjuvant injection.

DMSO treatment    Dimethyl sulfoxide (Ajax Chemicals, Sydney, NSW, Australia) at doses of 330 or 165 mg/kg was administered i.p. at the same treatment intervals as DMTU.

Experimental Design

Unless otherwise indicated, all data refer to day 14 after initiation of AA. The groups studied are summarized in Table 1.

Assessment of adjuvant arthritis

Development of arthritis was studied on days 3, 7, 11, and 14 after adjuvant injection and the following parameters assessed.

Clinical disease score    The severity of clinical disease was rated 0-3 as follows: no swelling or limitation in movement = 0; distinct swelling of the distal hindlimb joints but no limitation of movement = 1; moderate swelling of hindlimb and forelimb joints and minor restriction of mobility = 2; and marked swelling in all limbs, involvement of the tip of the tail and significant limitation in mobility = 3.

Hindpaw swelling by volume replacement    Plastic tubes were filled to the rim with water containing 1% detergent (pyroneg; Diversey, Seven Hills, NSW, Australia) and weighed. Under light ether anaesthesia, each hindpaw was then immersed vertically up to the level of the line joining the medial and lateral malleolus of the ankle joint. The tubes were again weighed and the weight difference recorded as volume displaced. Changes in hindpaw volume are expressed as a percentage of initial hindpaw volume (i.e. day 0).

Synovial fluid leucocyte count    This was assessed on day 14 only. Joints were exposed by removal of the overlying skin and the joint cavities flushed with 3Ýml per joint of HBSS using a 26 gauge needle and syringe. Lavage fluid from both knee joints and ankle joints was collected and pooled. Cells were then washed in HBSS and counted in a modified Neubeaur haemocytometer (Improved Neubauer, Weber, UK).

Table 1

Phenotypic Characterization of Synovial Leucocytes

Freshly collected synovial leucocytes were cytospun onto glass slides using a centrifuge (Shandon Cytospin II, Shandon Instruments, Runcorn, UK) at 500 g for 5 min. Cytospin slides were stained using mAb to leucocyte subsets and an indirect immunoperoxidase technique. Briefly, cells were pre-incubated with 10% normal rat serum (NRtS) and 5% BSA (Pentex BSA Fraction V; Miles Inc., Kankakee, IL, USA) in PBS for 20 min at room temperature. Cells were then incubated with mAb directed against rat Ia antigen (MRC OX-6; donated by A.F. Williams, MRC Cellular Immonology Unit, University of Oxford, Oxford, UK), T cell receptor (R73: Serotec, Kidlington, Oxford, UK), and a monocyte/macrophage marker (ED1: Serotec) diluted 1/10, 1/100 and 1/1000, respectively, for 1 hour at room temperature. Endogenous peroxidase activity was quenched by incubation in 1.0% H₂O₂ (H₂0₂; Ajax Chemicals) in methanol for 20 min. at room temperature. Bound mAb were labeled with rabbit antimouse IgG-horseradish peroxidase complex (Dakopatts, Glostrup, Denmark) diluted 1/50 in 5% NRtS in 1% BSA/PBS for 30 min at room temperature. Cells were treated with 0.6 mg/ml diaminobenzidine tetra-hydrochloride (0.06% DAB; Sigma) and 0.03% H₂0₂ in 0.05 mmol/L Tris-HCl pH 7.6 and counterstained with haematoxylin (Harris' Haematoxylin; BDH Chemicals, Kilsyth, Vic., Australia). Cryostat cut frozen tissue sections of spleen and cytospin preparations of peritoneal macrophages and neutrophils served as positive controls for the leucocyte markers. Staining using normal mouse glubulin as the primary antibody was performed on cytospin preparations as a negative control. For Leishmann staining, cells were incubated with 1 ml of Leishmann stain (BDH Chemicals) for 2 min followed by 2 ml Leishmann buffer (BDH Chemicals) for 12 min then washed thoroughly with water. Neutrophils were identified by their typical nuclear morphology. One hundred cells were counted on each cytospin preparation to determine the percentage of cells positive for each marker.

DMTU Effect On Circulating Neutrophil Number

Blood (20 L) was collected and diluted in 1.98 ml of Unopette reservoir diluent (Becton-Dickinson, Ritherford, NJ, USA) to lyse erythrocytes. Cells were counted using a haemoncytometer (Improved Neubauer; Weber, UK) and neutrophils were identified by their typical nuclear morphology.

DMTU Effect On Peritoneal Neutrophil Accumulation

Rats (n = 5) were treated with 500 mg/kg of DMTU using the above protocol. Control rats (n = 5) were untreated. After 6 days, peritoneal neutrophil accumulation was induced by a 5 ml i.p. injection of thioglycollate at a concentration of 38.5 mg/ml in distilled water (medium brewer modified; BBL Microbiology System, Cockeysville, MD, USA) 24 hours after final DMTU treatment. Neutrophls were harvested 6 hours later by peritoneal lavage using 30 ml of HBSS and the cell yield determined by counting a haemocytometer.

Assessment of Cutaneous Delayed Type Hypersensitivity

Cutaneous delayed type hypersensitivity (DTH) response to MT was assessed in treated and control rats developing AA. ¹² A small portion of the hair on the back of animals was shaved, and 50 L of MT (0.05 mg) in oil or oil alone was injected i.d. 72 h prior to killing. After the rats were killed on day 14, skin thickness was measured using calibrated skin fold calipers.

Statistics

Data are expressed as mean ± s.e.m. Statistical significance was assessed by Student's t-test for unpaired data.