Nafamostat mesylate, a serine protease inhibitor, demonstrates novel antimicrobial properties and effectiveness in Chlamydia-induced arthritis

INTRODUCTION: Effective treatment of reactive arthritis would ideally achieve both control of inflammation and eradication of persisting arthritogenic pathogens. We use a model of experimental Chlamydia trachomatis-induced arthritis (CtIA) to evaluate the effectiveness of nafamostat mesilate (NM), a serine protease inhibitor with complement-modifying effects and anticoagulant properties. To date clinical use of NM has largely been in Asia and has been primarily confined to inflammatory states such as pancreatitis. METHODS: In vitro studies examined inhibition of Chlamydia proliferation using fibroblast cell lines as targets and phase contrast microscopy. In vivo studies used an established protocol, experimental CtIA, induced in Lewis rats by injection of synoviocyte-packaged C. trachomatis. NM was dissolved in water and administered by daily intraperitoneal injection at a dose of 10 mg/kg beginning the day prior to the administration of Chlamydia. Readouts in vivo included (i) joint swelling, (ii) histopathology scoring of severity of arthritis, (iii) host clearance of the pathogen (by ELISA, the IDEIA PCE Chlamydia). RESULTS: NM exerted a dose-dependent inhibition of chlamydial proliferation in vitro. Without NM, the mean number of inclusion bodies (IB) per well was 17,886 (± 1415). At 5 μg/mL NM, there were 8,490 (± 756) IB, at 25 μg/mL NM there were 35 IB and at 50 μg/mL NM no IB was observed. Chlamydial antigens in each well along the concentration gradient were assayed by ELISA, demonstrating that at 25 μg/mL NM inhibition of Chlamydia was almost complete. In the experimental arthritis model, joint swelling was significantly reduced with NM treatment: average joint width for the NM-treated animals was 8.55 mm (s.d. ± 0.6578, n = 10) versus 11.18 mm (s.d. ± 0.5672, n = 10) in controls (P < 0.001). Histopathology scoring indicated that NM resulted in a marked attenuation of the inflammatory infiltration and joint damage: mean pathology score in NM-treated animals was 10.9 (± 2.45, n = 11) versus 15.9 (± 1.45, n = 10) in controls (P < 0.0001). With respect to persistence of Chlamydia within the synovial tissues, NM treatment was accompanied by a reduction in the microbial load in the joint: mean optical density (O.D.) for ELISA with NM treatment was 0.05 (± 0.02, n = 4) versus 0.18 (± 0.05, n = 4) in controls (P < 0.001). CONCLUSIONS: NM is a protease inhibitor not previously recognized to possess antimicrobial properties. The present study demonstrates for the first time that NM exerts significant impact on C. trachomatis-induced arthritis and suggests that such approaches may prove clinically useful in chronic reactive arthritis.