Nafamostat Mesilate, a Potent Serine Protease Inhibitor, Inhibits Airway

Home , Nafamostat

Nafamostat Mesilate, a Potent Serine Protease Inhibitor, Inhibits Airway Eosinophilic Inflammation and Airway Epithelial Remodeling in a Murine Model of Allergic Asthma

Masayuki Ishizaki1, Hiroyuki Tanaka1,*, Daisuke Kajiwara1, Tatsuyuki Toyohara1, Keiko Wakahara1, Naoki Inagaki1, and Hiroichi Nagai1 1Laboratory of Pharmacology, Department of Bioactive Molecules, Gifu Pharmaceutical University, 5-6-1 Mitahora-higashi, Gifu 502-8585, Japan

Received June 24, 2008; Accepted September 29, 2008

Abstract. To clarify the involvement of serine proteases in the development of allergic airway inflammation, we investigated the effect of nafamostat mesilate, a serine protease inhibitor, in a murine model of allergic asthma. Mice were sensitized to ovalbumin (OA) with alum and then exposed to 1% OA for 30 min, three times every 4th day. Nafamostat mesilate was administered orally for 10 days during the allergen challenge. In sensitized mice, repeated allergen challenge induced an increase in tryptase proteolytic activity in bronchoalveolar lavage fluid (BALF). In addition, marked increases in the numbers of inflammatory cells, levels of T helper type 2 (Th2) cytokines and eotaxin in BALF, numbers of goblet cells in the epithelium, and level of OA- specific IgE in serum were observed after repetitive allergen inhalation. Treatment with nafamostat mesilate significantly inhibited not only increased proteolytic activities, but also increases in the numbers of eosinophils and lymphocytes in the BALF. Nafamostat mesilate also dose-dependently inhibited increases in the levels of interleukin-13 and eotaxin in BALF and goblet cell hyperplasia. These findings suggest that increased serine protease activity in the airways is involved in the development of antigen-induced allergic eosinophilic inflammation and epithelial remodeling in bronchial asthma.

Keywords: eosinophil, interleukin (IL)-13, serine protease, tryptase

Introduction effector cells, particularly in the early phase of the allergic reaction and the immediate asthmatic response. Bronchial asthma is a chronic inflammatory disease Through antigen cross-linking to high affinity IgE of the airways, characterized by airway eosinophilic receptors on the cell surface, mast cells discharge a inflammation, various airflow limitations, hyperrespon- group of mediators, including histamine and tryptase (3). siveness to a variety of stimuli, and airway wall It has also been reported that mast cells have a potential remodeling such as goblet cell hyperplasia in the role in cellular infiltrates and airway inflammation by epithelium, subepithelial fibrosis, and smooth muscle producing tryptase, lipid mediators such as leukotrienes hyperplasia/hypertrophy (1). Airway inflammation is and prostanoids, and cytokines (3). associated with the infiltration of eosinophils, lympho- Serine protease tryptase, localized in mast cell cytes, and mast cells into the airways; and increased granules, constitutes one of the most abundant mast cell numbers of these cells have been found in the broncho- proteins and is released in airways during allergen alveolar lavage fluid (BALF) and bronchial biopsies challenge in atopic subjects (4). It is also present in from asthmatic individuals (2). Mast cells are the main increased concentrations in induced sputum samples obtained from asthmatic patients (5). Tryptase has potent *Corresponding author. [email protected] biologic activities that may contribute to the inflam- Published online in J-STAGE on November 14, 2008 (in advance) matory response in bronchial asthma. Tryptase promotes doi: 10.1254/jphs.08162FP human mast cell degranulation (6), induces eosinophil

355 356 M Ishizaki et al and neutrophil migration (7), enhances histamine-induced Materials and Methods airway contraction (8, 9), and stimulates microvascular leakage (7) and proliferation of lung fibroblasts and Animals airway smooth muscle cells (10, 11). Seven-week-old male BALB/c mice were purchased Tryptase activity causes proteolytic activation by from Japan SLC (Shizuoka). The animals were housed specific cleavage of the N-terminus of protease-activated in plastic cages in an air-conditioned room at 22 ± 1°C receptors (PARs) (12). So far, four PARs have been with a relative humidity of 60 ± 5% and had free cloned and characterized: PAR-1 – PAR-4. Trypsin and access to food and water. Experiments were undertaken mast cell tryptase are known to be the endogenous following the guidelines for the care and use of experi- stimuli for PAR-2. PAR-2 is expressed by airway mental animals of the Japanese Association for epithelial cells, fibroblasts, myocytes, smooth muscle Laboratory Animals Science in 1987. cells, and nerve fibers of the respiratory tract (13). Therefore, the tryptase–PAR-2 interaction is thought to Reagents be involved in the onset, perpetuation and aggravation of Nafamostat mesilate was kindly provided by Torii airway allergic disease, especially bronchial asthma (14). Pharmaceutical Co., Ltd. (Tokyo). The following Recently, Erba et al. (15) reported that gabexate drugs and chemicals were purchased commercially: mesilate is a highly potent and specific inhibitor of S-2288 (H-D-Ile-Pro-Arg p-nitroanilide; Chromogenix- human tryptase. This compound is a synthetic low Instrumentation Laboratory SpA, Milan, Italy), molecular weight inhibitor of trypsin-like serine pro- ovalbumin (OA; Seikagaku Kogyo, Tokyo), prednisolone teases (16 – 18), but shows more than 100-fold higher 21-acetate (Sigma, St. Louis, MO, USA), bovine serum affinity for human tryptase than for the other proteases albumin (BSA, Seikagaku Kogyo), Türk solution such as thrombin and trypsin. Gabexate is used clinically (Wako Pure Chemical Industries Ltd., Osaka), sodium for the treatment of acute pancreatitis, disseminated pentobarbitone (Abbott Laboratory, Chicago, IL, USA), intravascular coagulation (DIC), and as an anticoagulant disodium ethylenediaminetetraacetic acid (EDTA-2Na; for extracorporeal circulation with hemodialysis in Nacalai Tesque Inc., Kyoto), Diff-Quik solution Japan (19). Nafamostat mesilate (6-amidino-2-naphthyl (International Reagent Corp. Ltd., Kobe), monoclonal rat p-guanidinobenzoate dimethane sulfonate), which is anti-mouse IgE antibody (LO-ME-3; Serotec Co., Ltd., another structurally related serine protease inhibitor, Oxford, UK), and peroxidase-conjugated streptavidin has been recently reported to be far more potent than (Dako, Glostrup, Denmark). gabexate (20). The affinity of nafamostat mesilate for human tryptase is 157- and 8814-fold higher than Sensitization and antigen challenge those for trypsin and thrombin, respectively (21). Thus, Experiments were performed as reported previously nafamostat mesilate appears to be useful for research (23). Mice were actively sensitized by intraperitoneal into the pathophysiological roles of tryptase, especially injections of 50 μg OA with 1 mg alum on day 0 and 12. allergic airway inflammation, using animal models. Starting on day 22, they were exposed to OA (1% w/v Recently, Chen et al. demonstrated that nafamostat diluted in sterile physiological saline) for 30 min, three mesilate inhibited antigen-induced airway eosinophilia, times every 4th day. Negative control animals were IgE production, and interleukin (IL)-4 and pro- exposed to saline in a similar manner. Nafamostat inflammatory cytokines such as IL-6 and tumor necrosis mesilate was dissolved in pure water. Prednisolone was factor (TNF)-α in BALF, but it augmented the produc- suspended in 0.5% carboxymethylcellulose. Nafamostat tion of IL-12, which is critical for the induction of Th1 mesilate (30, 100, or 300 mg/kg) and prednisolone responses, in BALF (22). However, it is still unclear (5 mg/kg) were administered orally for 10 days, from whether nafamostat mesilate can modulate antigen- days 21 to 30. Bronchoalveolar lavage (BAL) was induced T helper type 2/1 (Th2/Th1) imbalance and performed 24 h after the final antigen challenge. airway remodeling and whether it can attenuate these phenotypes through the inhibition of serine protease BAL activation. To evaluate airway inflammation, we examined the Therefore, in the present study, we investigated the accumulation of inflammatory cells in BALF. Experi- therapeutic potential of the potent tryptase inhibitor ments were performed according to previously described nafamostat mesilate on allergic airway inflammation and methods (24). Animals were killed with an intraperito- airway remodeling in a murine model of allergic asthma. neal injection of sodium pentobarbitone (100 mg/kg). The trachea was cannulated and the left bronchi were tied for histological examination. Then, the right air Effect of Tryptase Inhibitor on Asthma Model 357 lumen was washed four times with 0.5 ml calcium- Co., Ltd., Saitama) was used as a standard. and magnesium-free PBS containing 0.1% BSA and Optical densities of the enzymatic reactions were read 0.05 mM EDTA-2Na. This procedure was repeated three using an automatic ELISA plate reader (Multiscan MS times (total volume: 1.3 ml, recovery >85%). BALF ver 8; Lab-Systems Oy, Helsinki, Finland) at 450 nm from each animal was pooled in a plastic tube, cooled on (reference 690 nm) and analyzed using Deltasoft 3 ice, and centrifuged (150 × g) at 4°C for 10 min. Cell (Biometallics, Inc., Princeton, NJ, USA). The detection pellets were resuspended in the same buffer (0.5 ml). limit was 1 ng/ml. BALF was stained with Türk solution and the number of nucleated cells was counted in a Burker chamber. A Histological study differential count was made on a smear prepared with a The left lungs were distended with 10% buffered cytocentrifuge (Cytospin II; Shandon, Cheshire, UK) formalin via the trachea (10 cmH2O) for 30 min and then and stained with Diff-Quik solution (based on standard excised and immersed in fresh fixative for 24 h. Tissues morphologic criteria) of at least 300 cells (magnification were sliced and embedded in paraffin, and 6 μm sections ×500). The BALF supernatant was stored at −30°C for were stained with hematoxylin & eosin for general subsequent determination of cytokine production. morphology and periodic acid-Schiff (15) to identify goblet cells in the epithelium. Tryptase activity in BALF Goblet cell hyperplasia was examined with a method To evaluate tryptase activity in BALF, p-nitroaniline previously described by Padrid et al. (25) with a slight release from the chromogenic tryptase substrate S-2288 modification (24) using a Leica image analysis system was monitored at 405 nm according to previously (Leica, Cambridge, UK). Briefly, two to four specimens described methods (24). Ten microliters of 6 mM of the PAS-stained histological preparations of the left substrate solution and 70 μl of 57 mM Tris-HCl (pH 8.3) lobe, in which the total length of the epithelial basement were placed in a flat-bottom, 96-well microtiter plate, membrane of the bronchioles were 1.0 to 2.5 mm, were and then 40 μl of BALF was added to initiate the selected and the pathological changes were evaluated reaction. Immediately after the addition of BALF, according to a modified five-point scoring system using optical density was monitored and the microplate was a Leica microscope (×20 objectives). The preparations, incubated at 37°C. Tryptase activity was represented as in which the maximum internal diameter of the the difference from optical density at zero time. bronchioles was two times as large as the minimum internal diameter or exceeded two times, were not used Cytokine and eotaxin levels in BALF for analysis. The hyperplasia of the goblet cells in the The amounts of cytokines and a chemokine, eotaxin, epithelial lining was expressed by a score according to in BALF supernatant were measured using an enzyme- the percentage of the goblet cells in the epithelial cells. linked immunosorbent assay (ELISA for IL-4, IL-5, and To minimize sampling errors, a five-pointing scoring interferon (IFN)-γ, from Endogen Inc., Woburn, MA, system (grade 0 to 4) (24) was adopted: grade 0 (no USA; ELISA for IL-13 and eotaxin, from R&D Systems goblet cells); grade 1, <25%; grade 2, 25% – 50%; Inc., Minneapolis, MN, USA). The detection-limit of grade 3, 50% – 75%; grade 4, ≥75%. The mean scores of each kit was 5 pg/ml for IL-4 and IL-5, 10 pg/ml for the total epithelial cells in two to four preparations of IFN-γ, 1.5 pg/ml for IL-13, and 3 pg/ml for eotaxin. one mouse were counted. The mean scores of hyperplasia of the goblet cells were calculated in seven or Measurement of serum antigen-specific IgE eight animals. Immediately prior to the first and final antigen challenges, blood was collected and sera were obtained Statistical analyses by centrifugation and stored at −80°C. Antigen-specific Values are presented as the mean ± S.E.M. Statistical IgE in serum was measured using an ELISA as pre- significance between two groups was determined by the viously described (23, 24). Briefly, serum OA-specific two-tailed Student’s t-test or the Mann-Whitney’s U-test IgE was measured by coating flat-bottom, 96-well after the variances of the data were evaluated with the microtiter plates (Nunc Immuno-Plate 96-F; Roskilde, F-test. In multiple comparisons, Dunnett’s test was Denmark) with a monoclonal rat anti-mouse IgE used after the analysis of variances with Bartlett’s test. antibody at a concentration of 5 μg/ml. After blocking P-values less than 0.05 were considered to be significant. with 1% BSA, serum dilutions were incubated for 1 h followed by biotinylated-OA and peroxidase-conjugated streptavidin. Sequentially diluted monoclonal anti-OA IgE (donated by Dr. M. Kiniwa, Taiho Pharmaceutical 358 M Ishizaki et al

Fig. 2. Effect of nafamostat mesilate on the proteolytic activity of tryptase in BALF obtained from mice after repeated antigen Fig. 1. Tryptase-like proteolytic activity in BALF obtained from challenge. Results were represented as the means ± S.E.M. of seven sensitized BALB/c mice 24 h after the final antigen challenge. to nine animals. The wavelength for optical density determination Optical density was increased with proteolysis of S-2288, a was 405 nm. OA, ovalbumin-inhaled; Pred, prednisolone; Sal, saline- chromogenic substrate sensitive to tryptase. Values represent the † inhaled. *P<0.05 (vs OA group); P<0.05 (vs OA group). means ± S.E.M. of seven or eight animals. The wavelength for optical density determination was 405 nm. OA, ovalbumin-inhaled; Sal, saline-inhaled. *P<0.05, **P<0.01 (vs Sal group).

Results

Tryptase activity in BALF To investigate whether repeated allergen challenge increases tryptase-like protease activity, we examined the proteolytic activity in BALF from antigen-exposed mice compared with those of saline-exposed mice. As shown in Fig. 1, tryptase substrate S-2288 was not degraded in saline-exposed mice. In contrast, an increase in the degradation of S-2288 was observed in BALF of OA-exposed mice, and the optical density was significantly and linearly increased from 10 to 60 min. These results indicate that final antigen challenge induced tryptase-like protease activity in BALF in the repeatedly challenged mice.

Effect of nafamostat mesilate on tryptase activity in BALF Nafamostat mesilate has a potent inhibitory effect against human tryptase (20). To clarify whether nafamostat mesilate has an inhibitory effect on murine tryptase, tryptase activities in BALF were examined with S-2288 as described above (Fig. 2). The drug was administered for 10 days during allergen challenge. Doses of nafamostat mesilate were determined in a Fig. 3. Effect of nafamostat mesilate on antigen-induced leukocyte previous report (26) because the biological half-life of accumulation in BALF obtained from sensitized BALB/c mice 24 h this compound is short (23.1 min) and it is particularly after the final antigen challenge. Values represent the means ± S.E.M. unstable in oral administration (27). In addition, we did of seven to nine mice in each group. OA, ovalbumin-inhaled; Pred, prednisolone; Sal, saline-inhaled. *P<0.05, **P<0.01, ***P<0.001 not observe decreases in body weight by nafamostat (vs OA group); †P<0.05, ††P<0.01 (vs OA group). mesilate even at a dose of 300 mg/kg. Nafamostat mesilate strongly inhibited the increased tryptase activity and normalized it at a dose of 300 mg/kg, although Effect of nafamostat mesilate on airway inflammation prednisolone had no effect. These findings indicate that To investigate the role of tryptase-like protease in nafamostat mesilate has a potent inhibitory effect on antigen-induced inflammatory infiltrates in the airways, murine tryptase-like protease activity. we examined the effect of nafamostat mesilate in this Effect of Tryptase Inhibitor on Asthma Model 359

Table 1. Effect of nafamostat mesilate on increase in cytokines and eotaxin in BALF obtained from sensitized BALB/c mice 24 h after the final antigen challenge Group Treatment IL-4 (pg/ml) IL-5 (pg/ml) IL-13 (pg/ml) IFN-γ (pg/ml) Eotaxin (pg/ml)

Sal vehicle 18.4 ± 2.2 N.D. 1.5 ± 1.0*** 305.6 ± 43.1 N.D. OA vehicle 23.5 ± 4.0 63.9 ± 15.5 143.1 ± 31.0 191.7 ± 41.3 19.1 ± 3.5 Nafamostat 30 mg/kg 26.1 ± 6.7 69.7 ± 9.2 138.7 ± 14.6 177.6 ± 27.9 16.4 ± 1.1 100 mg/kg 18.1 ± 5.0 55.2 ± 5.9 72.9 ± 12.3 178.7 ± 25.0 13.3 ± 1.9 300 mg/kg 16.3 ± 3.2 42.0 ± 4.4 54.5 ± 8.7† 178.4 ± 28.7 11.3 ± 1.7† Pred 5 mg/kg 11.7 ± 3.9 25.6 ± 4.3* 57.8 ± 8.2** 267.5 ± 21.2 13.0 ± 2.9 Values represent the means ± S.E.M. of seven to nine mice in each group. N.D., not detected; OA, sensitized and ovalbumin-inhaled; Pred, prednisolone; Sal, sensitized and saline-inhaled. *P<0.05, **P<0.01, ***P<0.001 (vs OA group); †P<0.05 (vs OA group). model. As shown in Fig. 3, repeated antigen challenge Effect of nafamostat mesilate on goblet cell hyperplasia induced increases in the numbers of total leukocytes, To evaluate the effect of nafamostat mesilate on macrophages, neutrophils, eosinophils, and lymphocytes antigen-induced goblet cells hyperplasia in the airway in BALF compared with those in saline-exposed animals. epithelium, which is a cardinal feature of bronchial In contrast, the oral administration of nafamostat asthma, lung sections were stained with PAS for mesilate at doses of 100 and 300 mg/kg significantly detection (Fig. 4), and then the goblet cell hyperplasia and dose-dependently inhibited the increases in the was quantitatively estimated in terms of grade as numbers of total leukocytes, eosinophils, and lympho- described in Materials and Methods (Fig. 4e). As shown cytes in BALF after antigen inhalation. The inhibitory in Fig. 4a, histological analyses of lungs from saline- potency of nafamostat mesilate at a dose of 300 mg/kg exposed mice showed normal lung histology. In was the same as that of prednisolone at a dose of contrast, the number of goblet cells in the epithelium 5mg/kg. was dramatically increased after repeated antigen challenge (Fig. 4b), and the goblet cells in the epithelium Effect of nafamostat mesilate on cytokine and chemokine showed hypertrophic features. Antigen-induced goblet production in BALF cell hyperplasia was dose-dependently and significantly To investigate the inhibitory mechanism of nafamostat reduced by treatment with nafamostat mesilate at doses mesilate on antigen-induced airway eosinophilia, we of 100 and 300 mg/kg (Fig. 4c), with efficacies similar next examined the effect on Th1/Th2 cytokines and to that of prednisolone at a dose of 5 mg/kg (Fig. 4d). chemokine production in BALF. Repeated antigen provocation induced significant increases in IL-5 and Effect of nafamostat mesilate on antigen-specific IgE in IL-13 production and a tendency to decrease IFN-γ serum production in BALF compared to those of saline-inhaled All sensitized animals produced detectable levels of mice (Table 1). IL-4 production in BALF tended to the antigen-specific IgE antibody in serum before increase after repeated challenge in sensitized mice. exposure at day 21, and there were no significant CCL-11, eotaxin-1, is a critical chemokine for the differences in the levels among groups (data not migration of eosinophils into inflamed tissues via shown). Repeated aeroantigen inhalation caused marked binding to the CCR3 receptor. Therefore, we measured increases in the antigen-specific IgE antibody (Table 2). the level in BALF. As expected, a significant increase in Nafamostat mesilate did not affect the increases in the level of eotaxin in BALF was observed after the antigen specific IgE levels, whereas prednisolone antigen challenge, whereas the chemokine was not showed a tendency to inhibit it. detected in BALF of saline-inhaled mice (Table 1). In contrast, treatment with nafamostat mesilate signifi- Discussion cantly inhibited the increases in IL-13 and eotaxin production in BALF in a dose-dependent manner. In In the present study, we investigated the therapeutic addition, the drug showed the tendency to inhibit IL-4 potential of the potent tryptase inhibitor nafamostat and IL-5 production in BALF, although nafamostat mesilate on allergen-induced airway eosinophilic mesilate did not reverse the decrease in IFN-γ produc- inflammation and epithelial remodeling in a murine tion after antigen inhalation. The inhibitory potency of model of allergic asthma. Repeated allergen challenge nafamostat mesilate at a dose of 300 mg/kg was almost caused an increase in tryptase proteolytic activity in the equivalent to that of prednisolone (Table 1). BALF of sensitized mice. Treatment with nafamostat 360 M Ishizaki et al

Fig. 4. Effect of nafamostat mesilate on antigen-induced goblet cell hyperplasia in sensitized BALB/c mice 24 h after the final antigen challenge (×200). a – d: Lung sections stained with periodic acid-Schiff. Scale bar = 100 μm. a: Saline-exposed mice, b: OA-exposed mice, c: nafamostat mesilate (300 mg/kg)-administered mice, and d: prednisolone (5 mg/kg)-administrated mice. e: Goblet cell hyperplasia was scored by 5 grades (0 to 4). Values represent the means ± S.E.M. of seven to nine animals in each group. OA, ovalbumin-inhaled; Pred, prednisolone; Sal, saline-inhaled. ***P<0.001 (vs OA group); ††P<0.01, †††P<0.001 (vs OA group).

Table 2. Effect of nafamostat mesilate on increases in serum anti- atopic diseases. gen-specific IgE after repeated antigen challenge To date, the role of PAR-2 in allergic airway inflam- Group Treatment Specific IgE (ng/ml) mation has been investigated using PAR-2 gene– Sal vehicle 166.3 ± 12.0** deficient mice, transgenic mice, or PAR-2 agonists. However, it is controversial as to whether or not the OA vehicle 349.1 ± 38.4 PAR-2 activation is involved in antigen-induced airway Nafamostat 30 mg/kg 322.2 ± 29.8 eosinophilic inflammation, and the participation of 100 mg/kg 252.7 ± 22.1 PAR2 activation in the pathogenesis of allergen-induced ± 300 mg/kg 333.5 20.3 airway remodeling has not been investigated. For ± Pred 5 mg/kg 259.8 14.9 instance, Schmidlin et al. reported that both airway Values represent the means ± S.E.M. of seven to nine mice in each eosinophilia and airway hyperresponsiveness to metha- group. OA, sensitized and ovalbumin-inhaled; Pred, prednisolone; choline were suppressed by a deficiency in the PAR-2 Sal, sensitized and saline-inhaled. **P<0.01 (vs OA group). gene, whereas the overexpression of PAR-2 augmented eosinophilic inflammation (28). In agreement with their mesilate during antigen inhalation dose-dependently and findings, Ebeling et al. recently reported that admin- significantly inhibited the increased tryptase proteolytic istration of a PAR-2–activating peptide via the nasal activity in BALF. In addition, increases in the numbers cavity enhanced antigen-induced airway inflammation of eosinophils and lymphocytes in the airway lumen, and hyperresponsiveness to methacholine. In addition, the levels of IL-13 and eotaxin in BALF, and goblet cell they demonstrated that PAR-2 activation could also hyperplasia in the epithelium were all significantly exaggerate established airway eosinophilic inflammation inhibited by the tryptase inhibitor in a dose-dependent (29). Moreover, Takizawa et al. clearly showed that the manner. These findings indicate that inhibition of serine activation of PAR-2 was involved in antigen-induced protease activity in the airways, for example, using airway eosinophilia via the increased production of nafamostat mesilate, is a viable therapeutic approach for eotaxin in the airways using PAR-2–deficient mice (30). the treatment of allergic eosinophilic inflammation in These findings clearly demonstrated that PAR-2 activa- Effect of Tryptase Inhibitor on Asthma Model 361 tion in the airways may have detrimental effects in increases in Th2 cytokine levels in BALF after allergen asthma; however, it has not been fully elucidated challenge in sensitized mice. Because Th2 cytokines whether or not the enzymatic activities of endogenous are involved in allergic inflammation through the PAR-2 ligands are increased after antigen challenge and differentiation of naïve CD4+ T cells into Th2 cells and how tryptase is involved in eosinophilic inflammation. eosinophils, cytokine/chemokine production at the local In contrast to these observations, it is reported that site and the activation of inflammatory leukocytes and activation of PAR-2 inhibited lipopolysaccharide- resident cells such as epithelial cells, fibroblasts, and induced airway neutrophilia in mice (31). De smooth muscle cells, the reduction of Th2 cytokine Campo et al. also clearly demonstrated that stimulation production by this inhibitor may affect the recruitment of PAR-2 inhibited antigen-induced airway eosinophilia of inflammatory infiltrates into the airways and the and hyperresponsiveness to methacholine in allergic prolongation of eosinophil survival. In contrast to Th2 mice through cyclooxygenase-2–dependent production cytokines, nafamostat mesilate did not affect the reduc- of prostaglandin E2. These observations indicate that tion in IFN-γ, a major Th1 cytokine, by repeated antigen airway PAR-2 activates critical anti-inflammatory challenge. These data suggest that nafamostat mesilate pathways in allergic airway (32). can inhibit antigen-induced airway eosinophilia by Therefore, prior to evaluating the effect of nafamostat inhibiting Th2 cytokine production and the chemokine mesilate, we first examined whether tryptase-like production responsible for chemoattraction of eosino- activity was increased after repeated antigen challenge phils. in our model. As a result, repeated antigen challenge Another question to be addressed is whether increased caused increases in tryptase-like activity in BALF tryptase-like activity is involved in the development of compared with saline-exposed mice, suggesting that airway remodeling in the epithelium. Mucus hyper- mast cells in the airways produce tryptase after antigen secretion from hyperplastic goblet cells causes airway challenge in this model. In contrast, but as expected, mucous plugging, especially in peripheral airways. nafamostat mesilate clearly inhibited the increase in Mucous plugging is reported to be a major contributing tryptase-like activity in a dose-dependent manner, factor in mortality rates associated with severe acute whereas prednisolone did not. Furthermore, we investi- asthma (34, 35). Therefore, the inhibition of goblet cell gated whether repeated antigen challenge caused hyperplasia is a therapeutic approach for the treatment of increases in the number of mast cells around airways bronchial asthma. Mucous hypersecretion is associated and examined the effect of nafamostat mesilate on it. with airway inflammation in asthma. A variety of The number of mast cells prior to the final antigen inflammatory mediators have been shown to stimulate challenge tended to increase, but was not significantly mucus secretion, including histamine, prostaglandins, different from those of saline-exposed animals and the leukotrienes, platelet activating factor, and eosinophil inhibitors had no effect on this (unpublished data). These cationic protein (36, 37). Recently, the cytokines IL-4, data demonstrate that nafamostat mesilate has the IL-9, and IL-13 have been shown to be involved in capacity to inhibit tryptase activity in this model. mucus secretion (38 – 40). In contrast, there is no The main question to be addressed is whether or not evidence that tryptase directly induces goblet cell hyper- increased tryptase-like activity induced by repeated plasia in vitro (41) or in vivo. In keeping with recent antigen challenge is involved in the development of findings, the present observation suggests that airway eosinophilic inflammation associated with Th2 nafamostat mesilate inhibited goblet cell hyperplasia cytokine production in the airways. In the present study, indirectly via the inhibition of IL-13 production in the we examined the effect of nafamostat mesilate on airways, rather than directly via the inhibition of tryptase antigen-induced inflammatory infiltrates in the airways, activity. Th2/Th1 cytokine and eotaxin production, and epithelial In this study, there is a discrepancy between the doses changes in a mouse model of allergic asthma. This that inhibited tryptase activity and cytokine productions inhibitor significantly suppressed airway eosinophilic in BALF and those that inhibited airway eosinophilia inflammation, IL-13 and eotaxin production, and goblet and epithelial remodeling. It is assumed that nafamostat cell hyperplasia in the epithelium. In allergic inflam- mesilate can attenuate the asthma-like phenotypes not matory responses, many cytokines and chemokines are only through inhibiting tryptase activity in the airways, released from various cell types. In bronchial asthma, but by suppressing the other mechanisms such as CD4+ T cells produce and secrete a large amount of Th2 complement inhibition and/or coagulation, although cytokines such as IL-4, IL-5, and IL-13; and these the Ki value for tryptase is 1000 times less than that cytokines promote allergic airway inflammation (33). for C1r, C1s, thrombin, or factor Xa (21). Nafamostat In the present study, nafamostat mesilate inhibited the mesilate was originally developed as an inhibitor of 362 M Ishizaki et al complements (42). Certainly, in a rat asthma model, 1991;87:893–910. nafamostat mesilate was used as a complement inhibitor 3 Metcalfe DD, Baram D, Mekori YA. Mast cells. Physiol Rev. and reduced airway inflammation (43). In addition, it 1997;77:1033–1079. has been also reported that the deficiency of C3 or CR3 4 Wenzel SE, Fowler AA 3rd, Schwartz LB. Activation of pulmonary mast cells by bronchoalveolar allergen challenge. reduced Th2 responses with allergen challenge (44). In vivo release of histamine and tryptase in atopic subjects with In contrast to complements, previous investigations and without asthma. Am Rev Respir Dis. 1988;137:1002–1008. have demonstrated that dysregulated coagulation and 5 Fahy JV, Liu J, Wong H, Boushey HA. Cellular and biochemical fibrinolysis contribute to the pathogenesis of asthma (45, analysis of induced sputum from asthmatic and from healthy 46) and that thrombin activity in BALF was increased subjects. Am Rev Respir Dis. 1993;147:1126–1131. by segmental antigen challenge in asthmatics (47). 6 He S, Gaca MD, Walls AF. A role for tryptase in the activation Furthermore, Shinagawa et al. clearly demonstrated that of human mast cells: Modulation of histamine release by tryptase and inhibitors of tryptase. J Pharmacol Exp Ther. factor Xa was involved in allergen-induced mucus 1998;286:289–297. production in the airways probably through augmenta- 7 He S, Walls AF. Human mast cell tryptase: A stimulus of tion of amphiregulin production (48). microvascular leakage and mast cell activation. Eur J Pharmacol. Here, we evaluated the effect of the potent tryptase 1997;328:89–97. inhibitor nafamostat mesilate on allergic airway inflam- 8 Molinari JF, Scuri M, Moore WR, Clark J, Tanaka R, Abraham mation and airway remodeling in a mouse model of WM. Inhaled tryptase causes bronchoconstriction in sheep via allergic asthma. Nafamostat mesilate significantly and histamine release. Am J Respir Crit Care Med. 1996;154:649– 653. dose-dependently suppressed antigen-induced increases 9 Johnson PR, Ammit AJ, Carlin SM, Armour CL, Caughey GH, in tryptase-like activity in the airways and inhibited Black JL. Mast cell tryptase potentiates histamine-induced airway eosinophilic inflammation, Th2-related cytokine contraction in human sensitized bronchus. Eur Respir J. and chemokine production in BALF, and goblet cell 1997;10:38–43. hyperplasia in the epithelium. Recently, we also 10 Frungieri MB, Weidinger S, Meineke V, Kohn FM, Mayerhofer examined the role of PAR-2 on allergic airway inflam- A. Proliferative action of mast-cell tryptase is mediated by par2, mation and airway remodeling using PAR-2 gene cox2, prostaglandins, and ppargamma: Possible relevance to knockout mice, which were backcrossed into BALB/c human fibrotic disorders. Proc Natl Acad Sci U S A. 2002;99: 15072–15077. mice at least more than 5 times. We found that defi- 11 Brown JK, Jones CA, Rooney LA, Caughey GH, Hall IP. ciency of the PAR-2 gene clearly attenuated antigen- Tryptase’s potent mitogenic effects in human airway smooth induced airway eosinophilia and Th2 cytokine produc- muscle cells are via nonproteolytic actions. Am J Physiol Lung tion in BALF compared with those of wild-type Cell Mol Physiol. 2002;282:L197–L206. BALB/c mice (Y. Nagira, Y. Ezaki, and H. Tanaka; 12 Ossovskaya VS, Bunnett NW. Protease-activated receptors: unpublished data). These findings demonstrate the Contribution to physiology and disease. Physiol Rev. 2004;84: important role of tryptase in the pathogenesis of allergic 579–621. 13 Lan RS, Stewart GA, Henry PJ. Role of protease-activated airway inflammation and the potential benefit of tryptase receptors in airway function: A target for therapeutic inter- inhibitors such as nafamostat mesilate for treating vention? Pharmacol Ther. 2002;95:239–257. airway eosinophilic inflammation in asthma therapy. 14 Reed CE, Kita H. The role of protease activation of inflammation in allergic respiratory diseases. J Allergy Clin Immunol. Acknowledgments 2004;114:997–1008; quiz 1009. 15 Erba F, Fiorucci L, Pascarella S, Menegatti E, Ascenzi P, Ascoli This study was supported in part by a donation from F. Selective inhibition of human mast cell tryptase by gabexate mesylate, an antiproteinase drug. Biochem Pharmacol. 2001;61: the late Professor Emeritus Akihide Koda, for which we 271–276. are very grateful. The authors would like to thank Drs. 16 Aramoto H, Saito H, Shigematsu H, Muto T. [Synthetic protease Yoshinori Itoh and Satoshi Tanaka for their helpful inhibitors in the treatment of disseminated intravascular coagula- comments during the present study and Mr. Daniel tion]. Nippon Rinsho. 1993;51:93–98. (text in Japanese with Mrozek for skilful assistance in preparation of this English abstract) manuscript. 17 Ramjee MK, Henderson IM, McLoughlin SB, Padova A. The kinetic and structural characterization of the reaction of nafamostat with bovine pancreatic trypsin. Thromb Res. 2000; References 98:559–569. 18 Menegatti E, Bolognesi M, Scalia S, Bortolotti F, Guarneri M, 1 Bousquet J, Jeffery PK, Busse WW, Johnson M, Vignola AM. Ascenzi P. Gabexate mesylate inhibition of serine proteases: Asthma. From bronchoconstriction to airways inflammation and Thermodynamic and computer-graphics analysis. J Pharm Sci. remodeling. Am J Respir Crit Care Med. 2000;161:1720–1745. 1986;75:1171–1174. 2 Kay AB. Asthma and inflammation. J Allergy Clin Immunol. 19 Yoshikawa T, Murakami M, Furukawa Y, Kato H, Takemura S, Effect of Tryptase Inhibitor on Asthma Model 363

Kondo M. Effects of fut-175, a new synthetic protease inhibitor 33 Robinson DS, Hamid Q, Ying S, Tsicopoulos A, Barkans J, on endotoxin-induced disseminated intravascular coagulation in Bentley AM, et al. Predominant th2-like bronchoalveolar t- rats. Haemostasis. 1983;13:374–378. lymphocyte population in atopic asthma. N Engl J Med. 1992; 20 Sendo T, Sumimura T, Itoh Y, Goromaru T, Aki K, Yano T, 326:298–304. et al. Involvement of proteinase-activated receptor-2 in mast cell 34 Aikawa T, Shimura S, Sasaki H, Ebina M, Takishima T. Marked tryptase-induced barrier dysfunction in bovine aortic endothelial goblet cell hyperplasia with mucus accumulation in the airways cells. Cell Signal. 2003;15:773–781. of patients who died of severe acute asthma attack. Chest. 21 Mori S, Itoh Y, Shinohata R, Sendo T, Oishi R, Nishibori M. 1992;101:916–921. Nafamostat mesilate is an extremely potent inhibitor of human 35 Saetta M, Di Stefano A, Rosina C, Thiene G, Fabbri LM. tryptase. J Pharmacol Sci. 2003;92:420–423. Quantitative structural analysis of peripheral airways and 22 Chen CL, Wang SD, Zeng ZY, Lin KJ, Kao ST, Tani T, et al. arteries in sudden fatal asthma. Am Rev Respir Dis. 1991;143: Serine protease inhibitors nafamostat mesilate and gabexate 138–143. mesilate attenuate allergen-induced airway inflammation and 36 Lundgren JD, Shelhamer JH. Pathogenesis of airway mucus eosinophilia in a murine model of asthma. J Allergy Clin hypersecretion. J Allergy Clin Immunol. 1990;85:399–417. Immunol. 2006;118:105–112. 37 Nadel JA. Role of enzymes from inflammatory cells on airway 23 Nagai H, Maeda Y, Tanaka H. The effect of anti-il-4 monoclonal submucosal gland secretion. Respiration. 1991;58 Suppl 1:3–5. antibody, rapamycin and interferon-gamma on airway hyper- 38 Lee JJ, McGarry MP, Farmer SC, Denzler KL, Larson KA, reactivity to acetylcholine in mice. Clin Exp Allergy. 1997;27: Carrigan PE, et al. Interleukin-5 expression in the lung 218–224. epithelium of transgenic mice leads to pulmonary changes 24 Tanaka H, Masuda T, Tokuoka S, Komai M, Nagao K, pathognomonic of asthma. J Exp Med. 1997;185:2143–2156. Takahashi Y, et al. The effect of allergen-induced airway 39 Temann UA, Geba GP, Rankin JA, Flavell RA. Expression of inflammation on airway remodeling in a murine model of interleukin 9 in the lungs of transgenic mice causes airway allergic asthma. Inflamm Res. 2001;50:616–624. inflammation, mast cell hyperplasia, and bronchial hyper- 25 Padrid P, Snook S, Finucane T, Shiue P, Cozzi P, Solway J, et al. responsiveness. J Exp Med. 1998;188:1307–1320. Persistent airway hyperresponsiveness and histologic alterations 40 Temann UA, Prasad B, Gallup MW, Basbaum C, Ho SB, Flavell after chronic antigen challenge in cats. Am J Respir Crit Care RA, et al. A novel role for murine il-4 in vivo: Induction of Med. 1995;151:184–193. muc5ac gene expression and mucin hypersecretion. Am J Respir 26 Ino Y, Sato T, Koshiyama Y, Suzuki K, Oda M, Iwaki M. Cell Mol Biol. 1997;16:471–478. Effects of fut-175, a novel synthetic protease inhibitor, on the 41 Chokki M, Yamamura S, Eguchi H, Masegi T, Horiuchi H, development of adjuvant arthritis in rats and some biological Tanabe H, et al. Human airway trypsin-like protease increases reactions dependent on complement activation. Gen Pharmacol. mucin gene expression in airway epithelial cells. Am J Respir 1987;18:513–516. Cell Mol Biol. 2004;30:470–478. 27 Tsukagoshi S. [pharmacokinetics studies of nafamostat mesilate 42 Fujii S, Hitomi Y. New synthetic inhibitors of c1r, c1 esterase, (fut), a synthetic protease inhibitor, which has been used for the thrombin, plasmin, kallikrein and trypsin. Biochim Biophys treatments of dic and acute pancreatitis, and as an anticoagulant Acta. 1981;661:342–345. in extracorporeal circulation]. Gan To Kagaku Ryoho. 2000;27: 43 Abe M, Shibata K, Akatsu H, Shimizu N, Sakata N, Katsuragi T, 767–774. (text in Japanese) et al. Contribution of anaphylatoxin c5a to late airway responses 28 Schmidlin F, Amadesi S, Dabbagh K, Lewis DE, Knott P, after repeated exposure of antigen to allergic rats. J Immunol. Bunnett NW, et al. Protease-activated receptor 2 mediates 2001;167:4651–4660. eosinophil infiltration and hyperreactivity in allergic inflam- 44 Drouin SM, Corry DB, Kildsgaard J, Wetsel RA. Cutting edge: mation of the airway. J Immunol. 2002;169:5315–5321. The absence of c3 demonstrates a role for complement in th2 29 Ebeling C, Forsythe P, Ng J, Gordon JR, Hollenberg M, effector functions in a murine model of pulmonary allergy. J Vliagoftis H. Proteinase-activated receptor 2 activation in the Immunol. 2001;167:4141–4145. airways enhances antigen-mediated airway inflammation and 45 Oh CK, Ariue B, Alban RF, Shaw B, Cho SH. Pai-1 promotes airway hyperresponsiveness through different pathways. J extracellular matrix deposition in the airways of a murine asthma Allergy Clin Immunol. 2005;115:623–630. model. Biochem Biophys Res Commun. 2002;294:1155–1160. 30 Takizawa T, Tamiya M, Hara T, Matsumoto J, Saito N, Kanke 46 Yuda H, Adachi Y, Taguchi O, Gabazza EC, Hataji O, Fujimoto T, et al. Abrogation of bronchial eosinophilic inflammation H, et al. Activated protein c inhibits bronchial hyperresponsive- and attenuated eotaxin content in protease-activated receptor 2- ness and th2 cytokine expression in mice. Blood. 2004;103: deficient mice. J Pharmacol Sci. 2005;98:99–102. 2196–2204. 31 Moffatt JD, Lever R, Page CP. Effects of inhaled thrombin 47 Terada M, Kelly EA, Jarjour NN. Increased thrombin activity receptor agonists in mice. Br J Pharmacol. 2004;143:269–275. after allergen challenge: A potential link to airway remodeling? 32 De Campo BA, Henry PJ. Stimulation of protease-activated Am J Respir Crit Care Med. 2004;169:373–377. receptor-2 inhibits airway eosinophilia, hyperresponsiveness and 48 Shinagawa K, Martin JA, Ploplis VA, Castellino FJ. Coagulation bronchoconstriction in a murine model of allergic inflammation. factor xa modulates airway remodeling in a murine model of Br J Pharmacol. 2005;144:1100–1108. asthma. Am J Respir Crit Care Med. 2007;175:136–143.