de Brum-Fernandes AJ, de Falco V, da Rocha FAC, Jancar S. Paradoxical effect of dexamethasone administration to rabbits with antigen-induced arthritis, ...
In flammopharmacology. 1992;1:289-294. Copyright C) Kluwer Academic Publishers by - Printed in the Netherlands
PARADOXICAL EFFECT OF DEXAMETHASONE ADMINISTRATION TO RABBITS WITH ANTIGEN-INDUCED ARTHRITIS AJ. DE BRUM-FERNANDES I*, V. DE FALCO 1, F.~. CASTRO DA ROCHA 1 and S. JANCAR z aFaculty of Medicine, University of S~o Paulo, 2Department of Immunology, Institute of Biomedical Sciences, University of Silo Paulo, Brazil *Correspondence
ABSTRACT de Brum-Fernandes AJ, de Falco V, da Rocha FAC, Jancar S. Paradoxical effect of dexamethasone administration to rabbits with antigen-induced arthritis, lnfiammopharmacology. 1992;1:289-294. Systemic dexamethasone phosphate administration to rabbits 2 h before induction of arthritis by antigen promoted a decrease in cellular influx, vascular permeability and eicosanoid concentrations in the synovial fluids. When administered 24 h before the articular challenge a significant increase in all parameters assessed was observed. A third group of animals receiving an injection of dexamethasone at 24 h and at 2 h before arthritis induction gave results similar to the group receiving one dose 2 h before the intra-articular challenge. The results suggest that a blockade of the hypothalamicpituitary-adrenal axis could be responsible for the findings in animals treated 24 h before induction of arthritis, and that endogenous glucocorticoids can be important for the down-regulation of the articular inflammatory reaction.
Keywords: arthritis, dexamethasone, eicosanoids
INTRODUCTION Glucocorticoids influence many physiological and pathological processes involved in inflammation, the inhibition of the synthesis of eicosanoids being one of their most remarkable actions. This is probably due to their induction of the synthesis of lipocortin, a peptide that blocks the action of phospholipase A 2 and the release of arachidonic acid from membrane phospholipids [1,2]. However, glucocorticoids interfere with other metabolic and endocrinological systems making their use as anti-inflammatory drugs troublesome because of the various and sometimes serious side-effects arising with long-term therapy. The discontinuation of glucocorticoid therapy in patients with arthritis usually leads to a sharp worsening in articular symptoms, even when the dose is slowly reduced. A better understanding of the mechanisms involved in glucocorticoid actions in articular inflammatory processes could contribute to better therapeutic management of such patients. Antigen-induced arthritis (AIA) in rabbiis can provide a useful tool for the study of articular inflammation therapy. Besides its pathological and physiopathological resemblances to human arthritis [3-9], this experimental model shares common inflammatory mediators with human pathology, namely, the presence of prostaglandins [113] and leukotrienes [11] in the synovial fluid. In the present study, the
290
de Brum-Fernandeset al.
effects of dexamethasone administration on various inflammatory parameters in rabbits with AIA were investigated, and an apparent paradoxical effect was observed when administered 24 h before induction of arthritis.
MATERIALS AND METHODS
Animals New Zealand white rabbits of either sex, aged three to four months, were obtained from the animal facilities of the Faculty of Medicine, University of S~o Paulo. They were caged individually and fed appropriate rabbit chow and had free access to water.
Induction of arthritis Arthritis was induced according to a modification [4] of a previously described method [7]. Briefly, animals were first immunized with bovine serum albumin (BSA, Sigma Co.) 5 mg in 2.0 mL of Freund's Complete Adjuvant (Difco Co.). Three weeks later their backs were shaved and each animal received 5 intradermal injections of 0.2 mL BSA (0.1% in sterile apyrogenic 0.9% NaC1, saline). Animals that developed an intense cutaneous Arthus reaction, characterized by central necrosis, were submitted, one week later, to an injection of the antigen in the left knee (0.4 mL of BSA 0.2% in sterile apyrogenic saline) and an equal volume of sterile apyrogenic saline in the contralaterai joint. Animals which did not respond appropriately to the cutaneous challenge after five consecutive weeks were not used. Four hours after the induction of arthritis, the animals were killed by an intravenous injection of 2.0 mL KCI (20 mg/mL). The knees were dissected and the joints washed with 2 mL of saline containing EDTA (1 mg/mL). Synovial washings were collected and placed in ice.
Evaluation of vascular permeability in synovial membranes Thirty minutes before the intra-articular injection the animals received an intravenous injection of Evans Blue dye (20 mg/kg). Sephadex G-200 chromatography of rabbit serum collected 4 h after the dye injected showed that at least 95% of the dye was eluted with the last protein fraction, identified as albumin by immunoelectrophoresis (data not shown). The synovial washings were centrifuged at 150 g at 4"C for 15 rain and the concentration of Evans Blue dye estimated colorimetrically at 630 nm.
Cytology of synovial fluids Total cell counts on the synovial fluid were performed using a modified Neubauer chamber. Smears of the cell pellets obtained after centrifugation of the syaovial washings were stained by Leishman's method and differential counts made.
ParadoxicalEffectof Dexamethasonein AIA
291
Determination of eicosanoids in the synovial washings All procedures involving synovial washings were performed in ice or at 4~ After centrifugation and determination of Evans Blue dye concentration, synovial washings were stored at -70~ until required for the extraction of eicosanoids. This was done by chromatography in SEP-PAK C18 columns (Waters Associates, Milford, MA, USA): fixed amounts of the washings were acidified by addition of 3.0 mL of szllne-HCl pH 3.3, these were then centrifuged at 680 g for 20 rain at 4~ to remove precipitable proteins. Supernatants were passed through SEP-PAK columns, then washed sequentially with 5.0 mL of distilled water and 5.0 mL of absolute methanol. Eluates in methanol were dried under nitrogen and resuspended in phosphatebuffered saline (PBS) for the radioimmunoassay. Recovery of eicosanoids after the extraction procedure was verified by adding tritiated eicosanoids to syuovial washings. Recoveries were 77.1% for prostaglandin E2, 71.2% for leukotriene C4 and 85.2% for leukotriene B4. Eicosanoid concentrations were determined by radioimmunoassay using commercially available kits (Leukotriene C4/D4/E4/F 4 3I-1 RIA KIT with Bio-Mag Magnetic Separation; Prostaglandin E 2 3H RIA KIT with Bio-Mag Magnetic Separation; and Leukotriene B4 3H RIA KIT with Bio-Mag Magnetic Separation, Advanced Magnetics Inc, MA, USA). Results were corrected for the dilutions/concentrations introduced by the extraction procedures.
Pharmacological manipulations Five animals received dexamethasone phosphate (Decadron Injet~ivel, Merck, Sharp and Dohme, Brazil), 0.15 mg/kg i.v. 2 h before the induction of arthritis. Four animals received the same dose 24 h before the articular challenge, and four others the same dose on two occasions: 24 h and 2 h before the induction of arthritis. As a control, five animals received saline i.v. 2 h before the articular challenge.
RESULTS Antigen (BSA) injection in the left knees of previously immunized, non-treated rabbits induced a significant accumulation of cells at 4 h following the induction of arthritis, compared with the contralateral saline-injected joints (Table 1). Although both polymorphonuclear and mononuclear cell numbers were increased, polymorphonuclear ceils predominated. Evans Blue dye concentrations in the antigen-injected knees were also higher than in the control knees, demonstrating an increase in the vascular permeability in the inflamed joints. In addition, eicosanoid levels in the synovial washings were higher in the antigen-injected than in the control knees. Dexamethasone administration 2 h before the induction of arthritis significantly inhibited the increase in vascular permeability, as assessed by the Evans Blue dye method (Figure 1). Although the number of mononuclear cells was not sionificantly different from the control, dexamethasone reduced the number of polymorphonuclear cells in the synovial washings. A significant reduction in the concentration of all eicosanoids assayed was also detected. In contrast, when the same dose of dexamethasone was administered 24 h before the intra-articular injection, a significant
292
de Brum-Fernandeset ai.
TABLE I Cell number, vascular permeability and eicosanoid concentrations in synovial washings of non-treated rabbits with antigen-induced arthritis Intra-articular injection Saline
BSA
PMN (cells/ram 3) Mononuclear (cells/ram 3)
103 -+ 153 8 _+ 7
Evans Blue (/zg/mL) PGE 2 (ng/mL) Peptido-LTs (ng/mL) LTB4 (ng/mL)
7.4 0.21 1.54 0.41
_+ 0.34 -+ 0.12 -+ 0.89 _+ 0.11
23847_+ 7757*** 1051 __. 438*** 89 ___24*** 4.09 __. 1.81" 4.94 -+ 1.34"* 0.63 -+ 0.09*
*p
