INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Asthma is a chronic inflammatory disease of the airways in which various resident and migrated cell-derived molecules play a role (1). Because reactive oxygen and related species, including nitric oxide (NO), have a potent proinflammatory action (2, 3), these molecules may be involved in the airway inflammatory process. In animal models, allergen- (4, 5) and ozone-induced (6) airway inflammation and airway hyperresponsiveness are largely modified by inhibitors of synthesis of reactive oxygen and related species or by scavengers of radical species, supporting this hypothesis. Further, in asthmatic airways, NO production is increased, possibly via inducible NO synthase (iNOS) (7), and steroid treatment reduces the NO generation (8), suggesting that NO may be partly responsible for the asthmatic airway inflammation.

Other types of reactive oxygen such as superoxide anion (O₂) may also be exaggerated in asthmatic airways via upregulation of xanthine oxidase (XO) in microvascular endothelial cells (5) and NADPH oxidase in the infiltrated eosinophils (9). NO rapidly reacts with O₂, which is released from inflammatory cells including eosinophils, and results in the formation of the highly proinflammatory molecule peroxynitrite (10). However, the role of peroxynitrite in the inflammatory process of the late allergic response (LAR) after allergen challenge, which most resembles asthmatic airway inflammation, has not yet been elucidated. The aim of this study was to examine the role of peroxynitrite in the microvascular hyperpermeability during the LAR in sensitized guinea pigs. We assessed the NO, O₂, and peroxynitrite production by measuring the NO concentration in the expired air, O₂ generating enzyme activity, and peroxynitrite-induced nitration product immunostaining, respectively. We quantified the airway microvascular permeability by means of Monastral blue dye trapping between the postcapillary endothelium. The functional role of the NO, O₂, and peroxynitrite on the microvascular permeability was assessed using each molecule's synthase inhibitor or scavenger. Further, we also quantified the eosinophil accumulation into the airways during the LAR and examined the role of NO, O₂, and peroxynitrite in the eosinophil response. We found that peroxynitrite formed by NO and O₂ is an important molecule for the microvascular hyperpermeability but not the eosinophil accumulation during the LAR.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animal Sensitization

Male Dunkin-Hartley guinea pigs (Funabashi Farm, Sendai, Japan) weighing 150 to 200 g were used. On Days 0 and 1, all animals were sensitized with a subcutaneous injection of 10 µg ovalbumin (OA) and 100 mg aluminum hydroxide. This sensitization procedure has been reported to develop an IgE-driven response (11). On Day 6, all animals were exposed to 2% OA in saline using an ultrasonic nebulizer (NE-U12; Omron, Tokyo, Japan; output 0.8 ml/min) for 3 min in a Plexiglas exposure chamber (24.5 × 40.5 × 15.0 cm) under spontaneous breathing. All the experiments performed in this study were conducted with the consent of the Ethics Committee for Use of Experimental Animals of the Tohoku University School of Medicine.

Allergen Challenge

On Day 21, the animals were pretreated intraperitoneally with pyriramine (10 mg/kg) to prevent the animals from dying in the early-phase airway response. Each NO synthase (NOS) inhibitor, XO inhibitor, peroxynitrite scavenger, or vehicle thereof was also injected intraperitoneally. Thirty minutes after the pretreatment, the animals were exposed to saline aerosol (control group) or 1% OA aerosol (challenged group) using the above-described nebulizer system for 1 min. Two hours after the inhalation, all animals were anesthetized intraperitoneally with urethane (2 g/kg). A water-filled esophageal catheter (outer diameter, 1.52 mm) was inserted to monitor the transpulmonary pressure (Ptp). Because Ptp reflects changes in both airway resistance and parenchymal compliance, we judged the LAR occurrence using this parameter. When the Ptp reached twice that of the baseline value in the challenged group, the animals were examined as described later. In the control group, each study was performed at about 5 to 6 h after saline inhalation, which is compatible with the time at which LAR occurs in OA-challenged animals.

Protocol of Inhibition Study

NOS inhibition study. The animals were divided into three groups. Group 1: NOS inhibitor (N-nitro-L-arginine methyl ester [L-NAME], 100 mg/kg given intraperitoneally), treated, OA-challenged (n = 7). Group 2: inactive enantiomer (N-nitro-D-arginine methyl ester [D-NAME], 100 mg/kg given intraperitoneally) treated, saline-exposed (n = 6). Group 3: D-NAME treated, OA-challenged (n = 6).

XO inhibition study. The animals were divided into three groups. Group 1: XO inhibitor {4-amino-6-hydroxypyrazolo(3,4-d)pyrimidine [AHPP], 50 mg/kg given intraperitoneally} treated, OA-challenged (n = 8). Group 2: vehicle for AHPP (0.1 N NaOH, 1 ml/kg given intraperitoneally) treated, saline-exposed (n = 6). Group 3: NaOH treated, OA-challenged (n = 8).

Peroxynitrite scavenger study. The animals were divided into three groups. Group 1: peroxynitrite scavenger [2-phenyl-1,2-benzisoselenazole-3(2H)-one (ebselen), 30 mg/kg given intraperitoneally] treated, OA-challenged (n = 7). Group 2: vehicle for peroxynitrite scavenger (100% DMSO 1 ml/kg given intraperitoneally) treated, saline-exposed (n = 6). Group 3: DMSO treated, OA-challenged (n = 8).

The dose of each inhibitor or scavenger was chosen to cause maximal inhibition according to previous studies (12).

Quantification of Airway Microvascular Permeability

Monastral blue dye (particle size: 5 to 300 nm) was sonicated for 5 min and filtrated using a 5 µm Millipore filter just before use. One minute after the intravenous administration of Monastral blue dye (30 mg/ kg), the thorax was opened and the systemic and pulmonary circulation were perfused with 1% paraformaldehyde (PFA) in 50 mM phosphate-buffered saline (PBS) as previously reported (15). Then, the trachea was removed and tracheal whole mounts were prepared. The lower part of the trachea was immersed in 1% PFA for 2 h, washed in distilled water for 2 h and soaked in glycerol for 20 h at room temperature, followed by dehydration in 100% ethanol. Next, the tissue was immersed in toluene, then in 100% ethanol, and hydrated in distilled water. The hydrated trachea was opened longitudinally along the ventral midline and flattened between two glass slides held tightly by clips for 24 h in 100% ethanol, cleared in toluene for 15 min, and mounted on a glass slide. Three images of the membranous portion from each tracheal whole mount preparation were viewed with image-analyzing software (MacScope; Mitani Co., Fukui, Japan) using an Apple Macintosh computer connected to the microscope. The Monastral blue dye trapped between the endothelium was quantified as the area density (16), that is, the percentage of the blood vessels labeled with Monastral blue in the tracheal whole mount. Because the tissue must be dissected longitudinally when using Monastral blue technique, we quantified the microvascular permeability only in the trachea.

Quantification of Eosinophil Accumulation into the Airways

The trachea was immersed in 10% formalin for 3 d, and then embedded in paraffin and sectioned at a thickness of 4 µm. The obtained section was stained using Hansel's stain. Eosinophils were counted both in the epithelial layer and the submucosal area.

Measurement of NO Concentration in the Expired Gas

In another set of experiments, we measured the expired NO concentration. The animals were divided into three groups. Group 1: D-NAME treated, saline-exposed (n = 5). Group 2: D-NAME treated, OA-challenged (n = 5). Group 3: L-NAME treated, OA-challenged (n = 5). At the LAR, the trachea was cannulated and the lungs were ventilated artificially with a small animal constant-volume ventilator (Model SN-480-7; Shinano Seisakusho, Tokyo, Japan) at a frequency of 45 strokes/min with a tidal volume of 10 ml/kg body weight. During mechanical ventilation, NO free gas (NO concentration < 1 ppb) was supplied from the inspiratory port of the ventilator, and the expired gas from the animal was collected from the expiratory port to the vacuum sampling bag for 3 min. The collected gas was adequately mixed manually, and then the NO concentration of the expired gas was determined by a chemiluminescence NO analyzer (Sievers 280; Sievers Instruments Inc., Boulder, CO), as previously reported (17).

Measurement of XO and Xanthine Dehydrogenase (XD) Enzyme Activity

According to the method of Akaike and coworkers (18), XO and XD enzyme activity was measured. The animals were divided into three groups. Group 1: D-NAME treated, saline-exposed (n = 8). Group 2: D-NAME treated, OA-challenged (n = 10). Group 3: L-NAME treated, OA-challenged (n = 8). After the anesthesia with urethane, the animals were exsanguinated by cutting the left ventricle. Immediately after, the animals were perfused with ice cold PBS and then with inhibitor cocktail (ice-cold PBS containing 2 mM EDTA, 2 mM p-amidino PMSF, 10 mM dithiothreitol, and 0.5 µg/ml leupeptin) via the ascending aorta and pulmonary artery. The trachea and lung were removed and divided into airway and parenchyma. Next, tissues were homogenized in 50 mM potassium phosphate buffer (pH, 7.6) containing the inhibitor cocktail at 4° C. The homogenates were centrifuged at 3,000 rpm for 10 min and the supernatants were recentrifuged at 100,000 × g for 1 h at 4° C. The supernatants were filtered with an 0.45-µm Millipore filter. In order to remove the low molecular-weight compounds (e.g., xanthine and hypoxanthine), they were dialyzed for 5 h against 10 L of PBS at 4° C with cellulose tubing (Seamless Cellulose Tubing size 8/32; Sankou Pure Chemical Industries, Tokyo, Japan) before determination of the XO enzyme activity. All samples were assayed for their XO activity using pterin as the substrate in a spectrofluorometer (Model 650-40; Hitachi Ltd., Tokyo, Japan) with excitation at 345 nm and emission at 390 nm. The volume of the assay mixture was 1.0 ml in PBS, which consisted of 9 µM pterin and 50 µl of the sample solution. Reactions proceeded for 1 h at 37° C. To measure both XO and XD activity, the above reactions were carried out in the presence of 9 µM methylene blue. To confirm the specificity of the activity, 20 µM alloprinol were added to the sample and the reaction was carried out. The activity was determined as the formation of isoxanthopterin. We employed the airway/parenchyma ratio as an index of XO plus XD activity, because the inflammatory site during the LAR is the airways and not the parenchyma.

Nitrotyrosine Immunostaining Study

The trachea was perfused with 1% PFA, immersed in 4% PFA fixative solution for 12 h at 4° C, and further immersed for 24 h at 4° C in 0.1 M phosphate buffer containing 15% sucrose. The tissues were then sectioned at a thickness of 6 µm with a cryostat. All sections were mounted on chrom-alum gelatin-coated glass slides. Endogenous peroxidase activity was reduced by incubation in 3% hydrogen peroxide in 100% methanol for 5 min at room temperature. After washing in PBS, sections were incubated with primary antibody (polyclonal nitrotyrosine rabbit IgG, 1:100 dilution at a final concentration is 3.92 µg/ml) (Upstate Biotechnology, Lake Placid, NY) or coincubated with excessive nitrotyrosine (as negative control) for 12 h at 4° C (19). In order to reduce nonspecific binding of the antibody, tissues were preincubated with 4% skim milk in PBS containing 0.3% Triton-X for 30 min and then incubated with 10% inactivated normal goat serum for 30 min at room temperature. The immunoreactions were visualized by the indirect immunoperoxidase method using ENVISION polymer reagent, which is antirabbit IgG from goat conjugated with peroxidase labeled dextran (DAKO Japan Ltd., Kyoto, Japan) for 1 h at room temperature. The diaminobenzidine reaction was performed, followed by counterstaining with hematoxylin. The tissue sections were photographed with Fuji Neopan F-film (Iso 32; Fuji Photo Film Co., Tokyo, Japan).

Drugs

Ovalbumin (Class V), Monastral blue B suspension, paraformaldehyde, Triton-X, diaminobenzidine, dithiothreitol, leupeptin, pterin, EDTA, L-NAME, D-NAME, methylene blue, and ebselen were obtained from Sigma Chemical Co. (St. Louis, MO). Hydrogen peroxide, NaN₃, and p-amidino PMSF were purchased from Wako Pure Chemical Industries (Osaka, Japan), Hansel's stain from Torii Pharmaceutical (Tokyo, Japan), and AHPP, alloprinol, and isoxanthopterin from Aldrich Chemical Co. Inc. (Milwaukee, WI).

Statistical Analysis

Data are expressed as mean ± SEM. Multiple comparisons of mean data of dye extravasation, quantification of eosinophils, expired NO concentration, and XO plus XD enzyme activity among the groups were performed by one-way analysis of variance (ANOVA) followed by Scheffe's test as a post hoc test. Probability values of < 0.05 were considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

All animals showed rapid breathing and cyanosis during and immediately after OA inhalation. About 45 min later, the condition of the animals returned to the level of the preinhalation challenge; 4.5 to 6 h after the inhalation, the Ptp values rose to twice that of the baseline values in OA challenged but not in saline exposed animals.

NO and LAR

OA challenge caused remarkable Monastral blue dye extravasation in D-NAME treated animals compared with those of the control group during the LAR (Figure 1). L-NAME pretreatment almost completely inhibited the dye extravasation in the challenged group. The area density of the dye in D-NAME treated, saline-exposed, D-NAME treated, OA-challenged, and L-NAME treated, OA-challenged groups were 0.13 ± 0.07, 5.5 ± 1.2, and 1.5 ± 0.54%, respectively (Figure 2). A representative view of eosinophil accumulation from the control and challenged group in the LAR is shown in Figure 3. OA inhalation caused significant eosinophil accumulation both in the epithelium (p < 0.01) and the submucosa (p < 0.01) in the LAR compared with control animal (Figure 4). L-NAME pretreatment partially but significantly reduced the eosinophil accumulation both in the epithelium (p < 0.05) and the submucosa (p < 0.05) (Figure 4). In D-NAME pretreated animals, OA inhalation also caused a significant increase in the expired NO concentration (22.8 ± 2.3 ppb, p < 0.05) compared with those of control (D-NAME treated, saline-exposed) animals (17.4 ± 1.6 ppb) during the LAR (Figure 5). L-NAME pretreatment almost completely abolished the NO concentration in the expired air (2.4 ± 0.49 ppb, p < 0.01 compared with both D-NAME pretreated, saline-exposed and D-NAME pretreated, OA-challenged animals) (Figure 5).

View larger version (169K): [in this window] [in a new window]   View larger version (168K): [in this window] [in a new window]   View larger version (172K): [in this window] [in a new window]   Figure 1.   Representative view of Monastral-blue-labeled airway microvessels in D-NAME treated, saline-exposed (A), in D-NAME treated, ovalbumin (OA)-challenged (B), and in L-NAME treated, OA-challenged animal (C ). Scale bar = 300 µm.

View larger version (13K): [in this window] [in a new window]   Figure 2.   Effect of NO synthase (NOS) inhibitor (L-NAME) and inactive enantiomer D-NAME on airway microvascular permeability during the late allergic response (LAR). All values are mean ± SEM of six to seven animals. **p < 0.01 compared with the value of D-NAME treated, saline-exposed group, p < 0.05 compared with the value of D-NAME treated, OA-challenged group.

View larger version (138K): [in this window] [in a new window]   View larger version (145K): [in this window] [in a new window]   Figure 3.   Eosinophil accumulation within the airway wall during the LAR. In D-NAME treated, saline-exposed animals, few eosinophils were seen (A). In contrast, much eosinophil accumulation both in the epithelium and the submucosa was observed in D-NAME treated, OA-challenged animals (B). Scale bar = 20 µm.

View larger version (15K): [in this window] [in a new window]   Figure 4.   Quantification of eosinophil infiltration in the epithelium (A) and submucosa (B) during the LAR. In the D-NAME treated, OA-exposed group (n = 6), the number of eosinophils significantly increased both in the epithelial layer and the submucosal area compared with the D-NAME treated, saline-exposed group (n = 6). In the L-NAME treated, OA-exposed group (n = 7), the number of eosinophils was partially but significantly decreased both in the epithelial layer and the submucosal area compared with the D-NAME treated, OA-exposed group. All values are mean ± SEM. **p < 0.01 compared with the values of D-NAME treated, saline-exposed group, p < 0.05 compared with the values of D-NAME treated, OA-challenged group.

View larger version (16K): [in this window] [in a new window]   Figure 5.   Expired NO concentration from guinea pig during the LAR. In the D-NAME treated, OA-challenged group (n = 5), the NO level in expired gas was significantly higher than that of the D-NAME treated, saline-exposed group (n = 5, p < 0.05). In the L-NAME treated, OA-challenged group (n = 5), the NO was almost abolished. All values are mean ± SEM.

O₂ and LAR

During the LAR, XO inhibitor AHPP pretreatment significantly inhibited the area density of Monastral blue dye (2.9 ± 0.69%, p < 0.05) compared with those of the vehicle for AHPP (NaOH) pretreated group (7.4 ± 1.3%) in the OA-challenged animals (Figure 6). In contrast to the NOS inhibitor study, AHPP pretreatment did not have a significant inhibitory effect on eosinophil accumulation in the airways in the LAR after OA challenge in the epithelium (38.3 ± 7.70/mm for vehicle treated and 34.8 ± 9.37/mm for AHPP treated) or in the submucosal area (1,150 ± 190/mm² for vehicle treated and 732 ± 158/mm² for AHPP treated). The XO plus XD activity in the airways during the LAR is shown in Figure 7. In D-NAME pretreated, OA-challenged animals, airway XO plus XD activity (expressed as airway/parenchyma ratio) was significantly elevated during the LAR (5.18 ± 1.1, p < 0.01) compared with those of the saline-exposed group (1.15 ± 0.36). L-NAME pretreatment did not affect the XO plus XD activity airway/parenchyma ratio in the OA-challenged animals in the LAR (Figure 7).

View larger version (16K): [in this window] [in a new window]   Figure 6.   Effect of xanthine oxidase (XO) inhibitor (AHPP) on airway microvascular hyperpermeability during the LAR. In the vehicle for AHPP (NaOH) treated, OA-challenged group (n = 8), vascular permeability was significantly increased compared with the vehicle treated, saline-exposed group (n = 6). Pretreatment with AHPP (n = 8) significantly reduced the response during the LAR. All values are mean ± SEM. *p < 0.05; **p < 0.01 compared with the value of vehicle treated, saline-exposed group, p < 0.05 compared with the value of vehicle treated, OA-challenged group.

View larger version (18K): [in this window] [in a new window]   Figure 7.   XO plus xanthine dehydrogenase (XD) enzyme activity of guinea pig lung during the LAR. Data of a D-NAME treated, saline-exposed group (n = 8), a D-NAME treated, OA-challenged group (n = 10), and an L-NAME treated, OA-exposed group (n = 8) are presented. All values are mean ± SEM. **p < 0.01 compared with the value of D-NAME treated, saline-exposed group.

Peroxynitrite and LAR

The peroxynitrite scavenger ebselen almost completely inhibited airway Monastral blue dye extravasation in the LAR after OA challenge; the area density of Monastral blue dye was 7.4 ± 1.3% in vehicle for ebselen pretreated and 1.4 ± 1.0% in ebselen pretreated animals (p < 0.05) (Figure 8). As in the AHPP study, airway eosinophil accumulation in the LAR was not significantly affected by ebselen pretreatment in the epithelium (47.2 ± 6.00/mm for vehicle treated and 37.8 ± 4.83/ mm for ebselen treated) or in the submucosal area (1,440 ± 264/mm² for vehicle treated and 1,380 ± 307/mm² for ebselen treated). The generation of nitrotyrosine in the airways was investigated by immunohistochemical study with antinitrotyrosine rabbit polyclonal antibody. As shown in Figure 9, strong staining with nitrotyrosine antibody in infiltrated polynuclear cells and microvascular endothelium was evident in D-NAME treated, OA-challenged animals, but not in saline-exposed animals during the LAR. L-NAME pretreatment diminished the positive staining in both polynuclear cells and microvascular endothelium during the LAR after OA challenge (Figure 9). In contrast, AHPP pretreatment abolished the nitrotyrosine staining only in the microvascular endothelium but not in polynuclear cells (Figures 10A and 10B). As in the L-NAME study, ebselen pretreatment completely inhibited the nitrotyrosine formation in both infiltrated cells and microvascular endothelium (Figures 10C and 10D).

View larger version (14K): [in this window] [in a new window]   Figure 8.   Effect of peroxynitrite scavenger (ebselen) on airway microvascular permeability during the LAR. All values are mean ± SEM of six to eight animals. **p < 0.01 compared with the value of vehicle for ebselen (DMSO) treated, saline-exposed group, p < 0.05 compared with the value of vehicle treated, OA-challenged animals.

View larger version (149K): [in this window] [in a new window]   View larger version (130K): [in this window] [in a new window]   View larger version (143K): [in this window] [in a new window]   View larger version (151K): [in this window] [in a new window]   Figure 9.   Immunohistochemical localization of nitrotyrosine in the D-NAME treated, saline-exposed (A), D-NAME treated, OA-challenged (B) and L-NAME treated, OA-challenged animal (C ). Negative control (primary antibody was coincubated with excessive nitrotyrosine) is shown in Panel D. Nitrotyrosine positive cells, infiltrated polynuclear cells, and microvascular endothelium (arrowhead ) are identified as brown in D-NAME treated, OA-challenged but not in those of the other interventions. Scale bar = 20 µm.

View larger version (125K): [in this window] [in a new window]   View larger version (151K): [in this window] [in a new window]   View larger version (135K): [in this window] [in a new window]   View larger version (145K): [in this window] [in a new window]   Figure 10.   Immunohistochemical localization of nitrotyrosine. Data of vehicle for AHPP (NaOH) treated, OA-challenged (A), AHPP treated, OA-challenged (B), vehicle for ebselen (DMSO) treated, OA-challenged (C ) and ebselen treated, OA-challenged animal (D). In each vehicle treated, OA-challenged group, strong staining was observed in both infiltrated cells and microvascular endothelium (arrowhead ). Ebselen pretreatment abolished the staining in both infiltrated cells and microvascular endothelium. In contrast, AHPP diminished the staining in the endothelium but not in the infiltrated cells. Scale bar = 20 µm.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animal LAR models are useful tools for investigating asthmatic airway inflammation. In the inflammatory process and subsequent airway narrowing, microvascular permeability plays a key role (1). However, reports concerning the vascular response in the LAR have been scanty. In the present study, we have shown that airway microvascular hyperpermeability occurs in the LAR. Further, inhibition of NO, O₂, and peroxynitrite has a potent inhibitory effect on the permeability.

In the present study, the expired NO concentration was elevated a small but significant amount in the LAR, suggesting exaggerated NO production in the airways at that time. In asthmatics, it has been reported that the allergen-induced late phase bronchospastic response is associated with an elevation of exhaled NO (20), which is compatible with our results. The mechanism of the elevated NO production in the LAR appears to be via iNOS induced by inflammatory cytokines such as TNF-, IL-1, and IFN- (2). In rats, it has been reported that allergen inhalation in sensitized animals caused an enhancement of the iNOS expression in both messenger RNA and protein in the LAR (21). Because, in the present study, the NOS inhibitor L-NAME significantly inhibited the OA- induced airway microvascular hyperpermeability during the LAR, the excessively produced NO seems to be involved in the phenomenon. The relatively small elevation in exhaled NO during the LAR may be due to the NO conversion to peroxynitrite via the reaction with O₂, which was also upregulated during the period.

In the present study, OA challenge enhanced the airway XO plus XD enzyme activity by about five times compared with the control group during the LAR, suggesting that O₂ production was also exaggerated at the time. XO has been reported to localize in vascular endothelium and connective tissues in airway (22). Among them, endothelial cells of the microvasculature are the most abundant source of the enzyme (23). In the present study, the XO inhibitor AHPP significantly reduced the OA-induced airway microvascular hyperpermeability during the LAR, indicating that O₂ as well as NO are key molecules of the response.

Because separate inhibition of both NO and O₂ production largely reduced but did not abolish the OA-induced microvascular hyperpermeability during the LAR, it is unlikely that each molecule alone has the ability to cause the microvascular hyperpermeability during this period. It has been reported that NO reacts with O₂ very rapidly (k = 6.7 × 10⁹ M¹S¹), resulting in peroxynitrite formation (24). Recently, it has been reported that the exogenous administration of peroxynitrite causes edema formation in rat skin (25). Therefore, peroxynitrite may be responsible for the OA-induced airway microvascular hyperpermeability during the LAR.

The above hypothesis is supported by the following two findings observed in the present study. First, peroxynitrite causes nitration of tyrosine residues and results in nitrotyrosine formation (19). In the present study, during the LAR, strong nitrotyrosine immunostaining was observed in the airway microvascular endothelium and infiltrated inflammatory cells (eosinophils), suggesting that peroxynitrite formation occurred during the period. A very recent study has shown that nitrotyrosine formation occurs via myeloperoxidase (MPO) pathways as well as by peroxynitrite (26). However, the source of MPO is mainly neutrophils, which were negligible in the present study. Therefore, the nitrotyrosine formation observed in the present study reveals peroxynitrite production. Second, peroxynitrite scavenger ebselen almost completely abolished the OA-induced airway microvascular hyperpermeability as well as the nitrotyrosine formation during the LAR. Compared with the high selectivity of L-NAME and AHPP for NOS (12) and XO (13), respectively, ebselen has an inhibitory effect on enzymes of inflammatory mediators as well as a peroxynitrite scavenging action. However, ebselen is more specific for peroxynitrite scavenging as the concentration that gives a 50% inhibition of the maximal response (IC₅₀) value is around 2.0 µM (27), which is lower than that for glutathione peroxidase activity (IC₅₀ = 10 µM) (28), 5-lipoxygenase inhibition (IC₅₀ = 27 µM) (29), and the inhibition of leukotriene B₄ formation (IC₅₀ = 4.0 µM) (29). In the present study, the estimated ebselen concentration was around 3.2 µM according to a previous report (25). This concentration seems sufficient to scavenge peroxynitrite, but it may have some effect on leukotriene B₄ formation. Because leukotriene B₄ has a potent neutrophil chemotactic action, this mechanism may affect the results observed in the present study. However, in the present study, neutrophil accumulation was not observed even without ebselen, suggesting that ebselen's effect on leukotriene B₄ inhibition is not important.

Both NOS inhibition by L-NAME and peroxynitrite scavenging by ebselen diminished the nitrotyrosine formation in both microvascular endothelium and infiltrated inflammatory cells. In contrast, the XO inhibitor AHPP inhibited the nitrotyrosine formation in vascular endothelium but not in infiltrated inflammatory cells. This may suggest that the exaggeration of XO activity mainly occurs in the airway microvascular endothelium during the LAR. On the other hand, the O₂ production observed in the infiltrated cells (eosinophils) may be via NADPH oxidase in the cells (9). In the present study, the XO inhibitor AHPP inhibited about 61% of the OA-induced airway microvascular hyperpermeability, which is relatively less effective compared with NOS inhibitor (72.7% inhibition) and peroxynitrite scavenger (81.0% inhibition). This seems to be due to the fact that the O₂ from NADPH oxidase in activated eosinophils also reacts with NO, resulting in peroxynitrite formation.

In the present study, compared with the strong inhibitory effect of L-NAME, AHPP, and ebselen on OA-challenge- induced microvascular hyperpermeability during the LAR, eosinophil accumulation was weakly inhibited by L-NAME but not by AHPP or ebselen. Possible explanations for the discrepancy are as follows. First, there is evidence that airway microvascular hyperpermeability quantified by Monastral blue dye trapping between the endothelium occurs mainly in the smallest postcapillary venules, whereas leukocyte attachment, which is a key step of the cell infiltration into the airways, occurs mostly in the largest postcapillary venules (15). Thus, the sites where the vascular hyperpermeability and inflammatory cell infiltration occur are different. Second, airway microvascular hyperpermeability is mainly caused by endothelial contraction at the leaky site. On the other hand, inflammatory cell infiltration into the airways is largely controlled by chemical factors, including cytokines, adhesion molecules, and chemokines. Therefore, airway microvascular hyperpermeability and eosinophil infiltration into the tissues seem to occur, at least in part, independently. The above hypothesis is supported by a recent study that showed, using radiolabeled albumin, that the antibodies to both intercellular adhesion molecule-1 (ICAM-1) and ₄ integrin reduced eosinophil infiltration but not microvascular permeability (30).

The mechanisms by which the NOS inhibitor induced a slight but significant reduction in eosinophil accumulation into the airways during the LAR in the present study are unclear. NO itself has been reported to have an inhibitory effect on the expression of ICAM-1, vascular cell adhesion molecule-1 (VCAM-1) (31), and E-selectin (32), which are known to have a key role in the eosinophil infiltration into the tissues. Accordingly, NOS inhibitor pretreatment may enhance eosinophil accumulation into the airways. However, in an allergic mice model, endogenous NO inhibition by L-NAME pretreatment (33) and iNOS knockout (34) had a significant inhibitory effect on eosinophil infiltration into the airways, as also shown in the present study. Taken together, in allergic airway inflammation, NO may facilitate eosinophil infiltration into the airway via its potent actions of vasodilation and microvascular hyperpermeability. NOS inhibitor causes vasoconstriction via endogenous NO depletion, resulting in decreased local blood flow. This may be another explanation for the inhibitory effect of L-NAME on eosinophil accumulation.

In summary, allergen challenge caused airway microvascular hyperpermeability and eosinophil infiltration into the airways during the LAR. NO production and XO plus XD activity were enhanced at that time. Either NOS or XO inhibition largely inhibited the vascular response. NO reacts with O₂ very rapidly, resulting in peroxynitrite formation. Actually, peroxynitrite production assessed by nitrotyrosine immunostaining was exaggerated, and the peroxynitrite scavenger almost completely inhibited the microvascular hyperpermeability. Therefore, peroxynitrite may be a final molecule among the reactive oxygen and related species in airway microvascular hyperpermeability during the LAR. Because airway microvascular hyperpermeability causes macromolecule leakage into the airways, resulting in airway inflammation and subsequent airway hyperresponsiveness (35), peroxynitrite scavengers may be useful for asthma therapy.