Comparative efficacy of a secretory phospholipase A₂ inhibitor with conventional anti-inflammatory agents in a rat model of antigen-induced arthritis

Rheumatoid arthritis (RA) is an immune-based chronic inflammatory synovitis presenting with pain, stiffness and swelling of the affected joints. RA results in secondary bone and cartilage destruction causing joint deformity. Current therapies include conventional non-steroidal anti-inflammatory agents (NSAIDs), corticosteroids such as prednisolone, disease-modifying anti-rheumatic-drugs, such as methotrexate or leflunomide, and biological therapies such as the inhibitors of tumour necrosis factor alpha (TNFα), etanercept, adulimumab and infliximab [1]. No single agent is completely effective at treating disease pathology and is devoid of side effects; consequently, a safe and effective treatment for RA remains elusive. In the mid-1980's, phospholipase A,₂ (PLA₂) enzymes were found to be highly expressed in the synovial fluid of RA patients [2]. PLA₂ forms a group of enzymes that metabolise phosphoglycerides to release lipid mediators such as lysophospholipids and arachidonic acid. These metabolites can be converted into the pro-inflammatory platelet activating factor (PAF) and eicosanoids (prostaglandins, thromboxanes, and leukotrienes), respectively [3]. As opposed to cytosolic PLA₂ enzymes which have physiological functions within virtually all cells [4], secretory PLA₂ (sPLA₂) enzymes are known to be active during inflammation, and thus have been an attractive target for anti-inflammatory drug development [3]. sPLA₂ enzymes also have agonistic activity at the M-type receptor, through which they can promote inflammation via degranulation of mast cells, cytokine release or secretion of elastase, an activator of the complement cascade extrinsic pathway [5‐8]. sPLA₂ enzyme concentrations have been found to be elevated in the synovial fluid of patients with RA [2, 9]. Correlations have also been found between serum levels of sPLA₂ and clinical markers of disease such as the number of active and effused joints, erythrocyte sedimentation rate, Lansbury index, elevated platelet count, and low hemoglobin in RA patients [10, 11]. Arthritic joints have also been shown to have high expression of sPLA₂ group IIa within the synovial lining, while sPLA₂ IIa expression in healthy joints is virtually absent [12]. Furthermore, intra-articular injections of human recombinant sPLA₂ caused acute inflammatory arthritic-like symptoms in rats [13] and rabbits [14], although transgenic mice over-expressing human sPLA₂ did not spontaneously develop arthritis [15, 16].

Researchers from Eli Lilly performed a phase I clinical trial using an inhibitor of sPLA₂ group IIa (LY315920) given intravenously to patients with active RA, which provided significant improvement in swollen and tender joints after three days [17]. Following this, a larger scale Phase II trial was conducted to evaluate the oral efficacy of LY333013, a methyl ester prodrug of LY315920. The results from this trial indicated that although there were significant dose-response related improvements after one week of treatment, there was no significant effect following four and eight weeks of treatment [17]. Potential explanations for this failure include the lack of sufficient inhibitor concentration in the synovial fluid to inhibit local joint sPLA₂, and that all patients were already receiving disease-modifying anti-arthritic drug therapy throughout the trial [17, 18]. Therefore, there is still a need to establish whether there may be a pathogenic role of sPLA₂ enzymes in RA.

We have previously reported that a synthetic small molecule inhibitor of group IIa sPLA₂ (sPLA₂I; 5-(4-benzyloxyphenyl)-4S-(7-phenylheptanoylamino)-pentanoic acid) is orally active and has therapeutic efficacy in rat models of intestinal ischemia-reperfusion injury [19] and inflammatory bowel disease [20]. There has also been evidence of efficacy with this compound in a small, preliminary investigation in adjuvant-induced arthritic rats [21]. To evaluate this finding, the present study reports a full investigation of the potential of this agent to prevent and reverse signs of inflammatory disease in the rat antigen-induced arthritis model. In addition, we compared the in vivo activity of this sPLA₂I to the conventional anti-arthritic agents, infliximab, leflunomide and prednisolone. In this rodent model of RA, we found that the sPLA₂I reduced all measured markers of arthritis pathology and was more effective than conventional anti-arthritic treatments.

This study is the first investigation of sPLA₂ IIa inhibition in the antigen-induced arthritis model of RA. We have previously demonstrated the usefulness of this model in establishing the efficacy of other experimental compounds and conventional anti-inflammatory drugs [23]. In the present study we have compared the efficacy of sPLA₂ group IIa enzyme inhibition, using an orally-active and specific small molecule sPLA₂I, with currently used anti-arthritic drugs in reducing antigen-induced arthritic pathology. We demonstrate that inhibition of sPLA₂ IIa alleviates the clinical signs and pathological changes associated with RA, with a greater reliability than some conventional anti-rheumatoid therapies.

sPLA₂ IIa is a secretory enzyme that converts arachidonic acid (AA)-containing phospholipids to free AA, and has been shown to be highly expressed in affected joint tissues in patients with RA [12]. sPLA₂ IIa forms the apex of an autacoids cascade in the synovium of arthritic joints. The heparin binding domain of sPLA₂ localises to lipid rafts, bringing the enzyme into close proximity to downstream mediators of this cascade, such as cycloxygenase and lipoxygenase [29]. Additionally, sPLA₂ IIa is a ligand for the M-type receptor located on inflammatory cells [5]. Signalling through the M-type receptor in mast cells results in degranulation [30]; in neutrophils it mediates an increase in cPLA₂ [29]; and in monocytes induces exocytosis of cytokines, including TNFα [31]. Current RA therapies, such as the TNFα inhibitor, infliximab, or the NSAID, ibuprofen, target the mediators downstream of sPLA₂ IIa. Specific inhibition of sPLA₂ IIa may, therefore, be a valid target to develop novel disease modifying anti-rheumatic drugs (DMARDs) which are more efficacious than existing therapies, given high concentrations of sPLA₂ IIa in arthritic joints [23]. This study confirms this hypothesis demonstrating that an orally-active sPLA₂I in a rat model of RA provides significant benefits over inhibition of downstream mediators of inflammation currently used as standard therapies in the treatment of RA.

In this study, we used a potent and orally active inhibitor of group IIa sPLA2 enzymes (sPLA₂I) [20, 21]. Oral administration of this drug to rats, prior to the induction of arthritis and daily throughout the trial, was found to be effective at reducing joint swelling and gait score when administered at both 1 and 5 mg/kg/day. However, sPLA₂I at the lower 1 mg/kg dose failed to reduce the disease progression as demonstrated by histopathology, when compared to untreated controls. Therefore, doses of 5 and 10 mg/kg/day were used to examine efficacy of reversing established arthritic damage. It is likely that the effect of sPLA₂I demonstrated in the prevention trials is due to action on the effector, rather than induction-phase of the immune response as rats were pre-sensitised to the antigen at -21 and -14 days. However, the design of this study did not allow us to discriminate between the action of the drug on both phases. In the reversal therapeutic trial, sPLA₂I was compared to conventional arthritis treatments infliximab, leflunomide and prednisolone. Rats were treated from Day 2, as this is near the maximal response in knee swelling and gait scores as seen in our first experimental trial. A separate group of rats euthanased at Day 2 showed a significant degree of histopathological damage (data not shown) validating the choice to initiate treatment at this time point.

Both 5 and 10 mg/kg/day sPLA₂I significantly reduced both gait score and joint swelling over the course of the study in the reversal trial. Of the conventional treatments, although all were able to demonstrate a significant benefit at certain individual time points, only infliximab reduced inflammation (knee width) and prednisolone reduced pain (gait score) with overall statistical significance. Additionally, treatment with infliximab or prednisolone showed no significant reduction in histopathology score, but both leflunomide and sPLA₂I (10 mg/kg/day) were effective in reducing the joint histopathology to a significant degree. Overall, sPLA₂I alleviated all aspects of RA pathology with a greater reliability than any of the conventional therapeutics in this study.

The success of targeted sPLA₂ IIa inhibition with the sPLA₂I used in this study could be attributed to the actions of sPLA₂ IIa upstream of many of the targets of the conventional therapies. For instance, the binding of sPLA₂ IIa to the M-type receptors has been shown to cause TNFα, IL6 and IL12 release from monocytes [32]; in this way sPLA₂I has clinical similarities to those of infliximab, a TNFα inhibitor. Additionally, sPLA₂ inhibition prevents mast cell degranulation by inhibiting sPLA₂ IIa interaction with the M-type receptor [6] whereas leflunomide induces mast cell apoptosis [33]. Mast cells have been postulated as the link between the antigen-antibody complex triggered inflammation and sustained, chronic RA, as mast cell deficient mice are resistant to the development of RA in an arthritogeneic serum model [29]. This mechanism could explain the success of early leflunomide intervention in clinical trials, and the significant reduction in histopathology score, by both sPLA₂I and leflunomide, in this study [34, 35]. We have previously demonstrated that another commonly used anti-arthritic agent, ibuprofen, was unable to reduce the degree of histological damage in the same model, despite providing a therapeutic effect for joint swelling and gait scoring [23]. This is in contrast to leflunomide, which was able to reduce histopathology without providing an overall significant benefit to gait score and joint swelling after Day 8. Inhibition of sPLA₂ IIa, however, caused significant reductions in gait score, joint swelling and histopathology. This is evidence that each of these facets of RA is mediated by separate, but communicating, mechanisms. sPLA₂ inhibition reduces prostaglandin synthesis by COX, through reducing the concentration of free AA, whilst ibuprofen is a direct inhibitor of COX. This inhibition of prostaglandin synthesis, which both reduces inflammation and inhibits pain, may account for the significant difference seen in gait scores/joint swelling with both these drugs. Inhibition of sPLA₂ IIa also reduces mast cell degranulation and neutrophil infiltration by preventing binding to M-type receptors [33]. This attenuation of immune cell function may mimic the mast cell-specific repression of leflunomide, which causes cell cycle arrest and mast cell apoptosis, in alleviating joint histopathology [33]. The multiple actions of sPLA₂ IIa in arthritis, and its absence in the synovium of healthy joints, are likely the reason for its success in this study when compared to conventional therapeutics. In addition, conventional therapeutics, with the exception of prednisolone, target single downstream mediators of sPLA₂ IIa actions.

Specific inhibition of sPLA₂ enzymes has an advantage over conventional RA therapeutics, in that it targets multiple possible pathways of RA pathogenesis (Figure 7), without affecting normally occurring biological processes. There have been no demonstrated side effects of sPLA₂ IIa inhibition in animal models of disease and the phase II clinical trial of an sPLA₂I in humans provided evidence of some liver toxicity only at the highest doses (1,000 mg/day) [17]. Conversely, all conventional therapeutics in this study failed to show significant benefit to every pathology measurement of the antigen-induced arthritis model. In addition, many of the conventional therapeutics have off-target effects that mitigate the benefits provided. Leflunomide and infliximab have been shown to inhibit osteogenic cell proliferation [36]; because of this, both treatments pose additional risks for post-menopausal women. Leflunomide and prednisolone actively suppress the immune system; leflunomide induces the apoptosis of several types of immune cells [33, 37, 38], and prednisolone induces cell cycle arrest, thereby inhibiting cellular proliferation. This study also confirms a catabolic effect of daily prednisolone administration as has been previously shown [24] and, although prednisolone can act as an inhibitor of sPLA₂ IIa transcription [39], the use of a low-dose steroid in a chronic condition can potentiate morbidity for the patient [40‐42].

Previously, the efficacy of sPLA₂ IIa inhibition has been tested in phase II clinical trials as an adjunct to DMARD therapy without success [17]. The administration of DMARDS in conjunction with the sPLA₂I may have masked any benefits provided by sPLA₂I. Here we show, for the first time, that sPLA₂ IIa inhibition has the potential to be a useful monotherapy for the treatment of RA, and may be a more effective chronic therapy than the conventional RA therapeutics.

We have previously shown the sPLA₂I used in this study to be an orally-active, highly selective drug for the treatment of intestinal ischemia reperfusion injuries [19] and inflammatory bowel disease [20] in rat models, and now demonstrate its efficacy in a model of RA. Inhibition of sPLA₂ IIa using a chemically different sPLA₂ enzyme inhibitor, has previously been trialed in human RA. In this clinical trial, the sPLA₂ inhibitor was administered as an adjunct therapy to DMARD and glucocorticoid therapy, which may have masked any benefits to patients [17]. By directly comparing sPLA₂ inhibition to conventional therapies in a rodent model of antigen-induced arthritis, we have provided a rationale and evidence for the use of sPLA₂I as a replacement for DMARD/glucocorticoid therapy in future clinical trials.