The mechanism of the anti-inflammatory activity of glucocorticosteroids.

Abstract

Activation of membrane phospholipases by a variety of stimuli is thought to play a key role in the production of arachidonic acid metabolites. By inhibiting phospholipase A 2 (PLA2), glucocorticosteroids (GC) prevent biosynthesis of the proinflammatory eicosanoids such as prostaglandins, hydroxy acids and leukotrienes [1]. The effect of the GC depends upon receptor occupancy which is coupled to translational and transcriptional events within the target cell and the production of a 'second messenger' protein(s) with anti-phospholipase activity [2]. Proteins with anti-phospholipase activity have been isolated from rat peritoneal leucocytes [2, 3] and steroid-stimulated rabbit polymorphonuclear leucocytes (PMNL) and termed macrocortin and lipomodulin respectively. It is generally accepted that this sequence of events is the mechanism by which GC exert their anti-inflammatory effect. While investigating the effects of GC and BW 755C, on inflammatory reactions in rabbit dermis, observations led us to question the anti-inflammatory mechanism of GC in this model. Inflammatory reactions were initiated by injection of reagents into the dermis of the shaved backs of female NZW rabbits. Plasma exudation (PE) was monitored by the extravasation of 125I rabbit serum albumin [4] and PMNL accumulation by an immunoradiometric assay of a novel PMNL specific marker protein [5]. Reversed passive Arthus reactions (RPAR, figure legend), arachidonic acid (100/tg), and LTB 4 together with PGE 2 (200 ng each), caused concentration and timedependent increases in PE and PMNL accumulation into the dermis. Plasma exudation and PMNL accumulation were temporally related and maximal at 3 hours. No PE occurred when any of these agents were injected into the dermis of rabbits made neutropaenic by treatment with nitrogen mustard (1.75 mg/kg l ) suggesting that the PE is PMNL dependent. In contrast, bradykinin (BK) induced maximal PE within 10 minutes of injection in both control and neutropaenic rabbits demonstrating its inflammatory effect is rapid and independent of PMNL. Dexamethasone (0.05-5.0 mg/kg -~) dosed 3 hours prior to initiating inflammatory reactions caused a doserelated inhibition of PE. Maximal inhibition was 40% for arachidonic acid and LTB 4 + PGE z and 60% for BK and RPAR with EDs0 achieved between 0.05 and 0.1 mg/kg 1 for each reaction. Therefore, dexamethasone is equally efficacious against PMNL dependent and independent inflammatory reactions in rabbit skin suggesting a common site(s) of action, perhaps the microvascular endothelium. In addition to inhibiting PE, paramethasone reduced the number of PMNL which accumulated into inflammatory lesions induced by LTB 4 + PGE 2 and a RPAR as shown in the figure. It is unlikely that the anti-inflammatory efficacy of GC in this model is solely the result of reduced eicosanoid biosynthesis via inhibition of PLA z since GC also inhibit inflammation induced by the pro-inflammatory eicosanoids LTB 4 + PGE 2 and arachidonic acid. It is tempting to speculate that the anti-inflammatory effect of GC against the RPAR and BK induced inflammation also occur by an alternative mechanism. It has been suggested that mixed inhibitors, compounds which inhibit both cyclo-oxygenase and lipoxygenase activity thus reducing pro-inflammatory eicosanoid biosynthesis, will have 'steroid-like' anti-inflammatory effects [6]. However, in marked contrast to GC the mixed inhibitor BW755C, when dosed orally at 50 (see figure) or 100 mg/kg 1, had no effect on PE or PMNL accumulation into RPAR or LTB 4 + PGE 2 inflammatory sites. Although its inactivity against inflammation induced by LTB 4 + PGEz is predictable, and in contrast to GC, failure to inhibit a steroid-sensitive RPAR implies (1), that the mixed inhibitor BW755C does not have 'steroid-like' anti-inflammatory effects and (2), that the efficacy of GC against immune complex mediated inflammation is not due to inhibition of