Published ahead of print on May 18, 2006, doi:10.1165/rcmb.2005-0391OC
© 2006 American Thoracic Society DOI: 10.1165/rcmb.2005-0391OC Effects of Nitric Oxide Synthases in Chronic Allergic Airway Inflammation and RemodelingDepartments of Medicine and Pathology, School of Medicine, University of São Paulo, São Paulo, Brazil Correspondence and requests for reprints should be addressed to Iolanda F. L. C. Tibério, M.D., Ph.D., Departamento de Clínica Médica, Faculdade de Medicina da Universidade de São Paulo, Av. Dr. Arnaldo, 455Sala 1216, 01246-903, São Paulo, SP, Brazil. E-mail: iocalvo{at}uol.com.br
The precise role of each nitric oxide (NO) synthase (NOS) isoform in the pathobiology of asthma is not well established. Our objective was to investigate the contribution of constitutive NO synthase (cNOS) and inducible NOS (iNOS) isoforms to lung mechanics and inflammatory and remodeling responses in an experimental model of chronic allergic pulmonary inflammation. Guinea pigs were submitted to seven ovalbumin exposures with increasing doses (1 5 mg/ml) for 4 wk. The animals received either chronic L-NAME (N-nitro-L-arginine methyl ester, in drinking water) or 1400W (iNOS-specific inhibitor, intraperitoneal) treatments. At 72 h after the seventh inhalation of ovalbumin solution, animals were anesthetized, mechanically ventilated, exhaled NO was collected, and lung mechanical responses were evaluated before and after antigen challenge. Both L-NAME and 1400W treatments increased baseline resistance and decreased elastance of the respiratory system in nonsensitized animals. After challenge, L-NAME increased resistance of the respiratory system and collagen deposition on airways, and decreased peribronchial edema and mononuclear cell recruitment. Administration of 1400W reduced resistance of the respiratory system response, eosinophilic and mononuclear cell recruitment, and collagen and elastic fibers content in airways. L-NAME treatment reduced both iNOS- and neuronal NOS-positive eosinophils, and 1400W diminished only the number of eosinophils expressing iNOS. In this experimental model, inhibition of NOS-derived NO by L-NAME treatment amplifies bronchoconstriction and increases collagen deposition. However, blockage of only iNOS attenuates bronchoconstriction and inflammatory and remodeling processes.
Key Words: collagen deposition experimental asthma model nitric oxide
Airway inflammation in asthma is under the control of complex and incompletely understood mechanisms involving the release of a variety of neurotransmitters, inflammatory mediators, and other signaling molecules, including nitric oxide (NO) (1, 2). NO is involved in several biological processes, including host defense, immune regulation, platelet aggregation, neurotransmission, and inflammation (2). NO is synthesized from L-arginine by NO synthase (NOS). Both neuronal NOS (nNOS) and endothelial NOS (eNOS) are considered constitutive, and are involved in vasodilation and bronchodilation. The inducible NOS (iNOS) is stimulated by many proinflammatory cytokines, and is expressed in several types of inflammatory cells (2). Excessive NO production is detected in the air exhaled by individuals with asthma and in experimental asthma models, and has been correlated to the intensity of airway inflammation (16). However, the exact role of each NOS in this process is still a matter of controversy. There is recent evidence suggesting a role of NO in cellular proliferation, collagen deposition, and myofibroblast differentiation (710). However, there are few studies evaluating the effects of chronic and specific iNOS inhibition in experimental models of chronic allergic inflammation. Because airway remodeling is an important clinical feature of asthma, and influences the decrease in lung function observed in patients with asthma (11), studies that evaluate the role of NO and NOS in this process are necessary. Based on these observations, the present study evaluated, by means of a chronic allergic pulmonary inflammation model, the differences in the effects of iNOS- and cNOS-derived NO in lung mechanics, airway inflammation, and collagen and elastic fibers deposition in airway wall. To achieve these objectives, guinea pigs with chronic allergic pulmonary inflammation were chronically treated with either a false substrate of NOS (L-NAME) or a specific and highly selective iNOS inhibitor (1400W).
All guinea pigs received humane care in compliance with National Institutes of Health guidelines, and all experiments described in this study were approved by the institutional review board of University of São Paulo (São Paulo, Brazil).
Induction of Chronic Allergic Pulmonary Inflammation
Chronic L-NAME Treatment Chronic L-NAME treatment (60 mg/kg/animal/d) was performed as previously described (6). Briefly, guinea pigs received L-NAME dissolved in drinking water ad libitum (OVA-L or NS-L groups; see EXPERIMENTAL GROUPS below), beginning 24 h after the fourth inhalation of either OVA or normal saline. Control animals received only sterile drinking water.
1400W Treatment
Experimental Groups
Concentration of Exhaled NO
Pulmonary Mechanics Evaluation
After baseline measurements of exhaled NO, Ptr, and Still under anesthesia, and after the post-challenge measurements of pulmonary mechanics, the anterior chest wall was opened and lungs were washed with heparinized saline solution (1:40). Guinea pigs were then exsanguinated, a positive end-expiratory pressure of 5 cm H2O was applied to the respiratory system, the airways were occluded at the end of expiration, and the lungs were removed en bloc.
Morphometric Studies Peribronchial edema was quantified in randomly selected, transversely sectioned noncartilaginous airways. The number of points of the integrating eyepiece falling on areas of edema was counted in three areas of each airway wall and termed "edema index" (EI; µm2). Mononuclear (MN) cells and eosinophils present in the airway wall were counted in three randomly selected areas of the same airways used to measure peribronchial edema, and were expressed as cells/unit area (104 µm2) (46, 12).
Evaluation of iNOS and nNOS Expression
Airway Remodeling
Statistical Analysis
Inhalation Time No differences were observed among the groups during inhalations 1 to 4. The saline-exposed animals stayed in contact with antigen for 900 s in all inhalations. At the fifth, sixth, and seventh inhalations, OVA-exposed animals showed lower values of inhalation time (seventh inhalation [mean ± SEM: 300.2 ± 38.0 s]) compared with normal saline-exposed animals (900.0 ± 0 s) (P < 0.001). Neither L-NAME nor 1400W treatments affected the time during which the animals could stay in contact with the antigen before the onset of respiratory symptoms (seventh inhalation [OVA-L: 343.0 ± 112.3 s; OVA-W: 311.1 ± 142.1 s).
Concentration of Exhaled NO
Respiratory Mechanics Evaluation Figures 3A and 3B show baseline (pre-OVA challenge) values of Rrs and Ers, respectively. There were no significant differences in baseline values of Rrs or Ers between the OVA and the NS groups. Both L-NAME and 1400W treatments increased baseline values of Rrs in both OVA and saline-exposed guinea pigs (P < 0.05). In addition, 1400W treatment resulted in baseline values of Rrs greater than those resulting from L-NAME treatment (P < 0.001). In contrast, both L-NAME and 1400W treatments reduced baseline values of Ers, and these differences were statistically significant only in the saline-exposed animals (P < 0.05) (Figure 3B). We observed no significant differences in Ers values when L-NAME and 1400W treatments were compared.
After OVA challenge or normal saline inhalation, we evaluated the percentage of maximal increase of both Rrs (%Rrs) (Figure 3C) and Ers (%Ers) (Figure 3D). The group exposed to OVA showed greater values of %Rrs and %Ers compared with saline-exposed animals (P < 0.001). L-NAME treatment increased %Rrs in OVA-exposed guinea pigs, (P < 0.01). In contrast, 1400W treatment resulted in lower maximal response of Rrs induced by OVA challenge (P < 0.01). OVA-exposed guinea pigs that were treated with L-NAME (OVA-L group) showed greater values of %Ers compared with that of the OVA-W group (P < 0.05).
Peribronchial EI
Inflammatory Cell Infiltrates Figures 4B and 4C show the numbers of MN cells (B) and eosinophils (C) in the wall of distal (noncartilaginous) airways of the six experimental groups studied. There were no significant differences among the normal saline-exposed groups (NS, NS-L, and NS-W). The number of MN cells and eosinophils in OVA-exposed guinea pigs was greater than that observed in saline-exposed guinea pigs (P < 0.001). In OVA-exposed animals, both L-NAME and 1400W treatments (P < 0.05) reduced the number of MN cells in airways. However, only 1400W treatment attenuated the eosinophilic infiltration around airways (P < 0.001 compared with OVA and OVA-L groups). OVA-exposed guinea pigs showed an increase in both nNOS-positive (Figure 4D) and iNOS-positive (Figure 4E) eosinophils around airways (P < 0.01) compared with normal saline-exposed animals. L-NAME treatment attenuated these responses (P < 0.05), and 1400W reduced only iNOS-positive eosinophils in sensitized animals (P < 0.001) compared with animals that received vehicle. Figure 5 shows representative photomicrographs of the distal airway walls of guinea pigs stained through immunohistochemistry for nNOS (A and B) and iNOS (C and D). In the photomicrographs of airway walls of guinea pigs exposed to aerosolized OVA, we noted a large number of iNOS-positive (D) and nNOS-positive (B) inflammatory cells; mainly eosinophils. In normal saline-exposed animals, we observed low numbers of inflammatory cells stained for nNOS (A) and iNOS (C) around distal airway walls.
Characterization of Airway Remodeling Figures 6B, 6E, 6H, and 6L show the collagenous fibers in the control (B) and OVA-exposed animals (E, H, and L) stained with Picrossirius and observed using polarized light. In saline-exposed guinea pigs, there was a weak, red-orange birefringence of peribronchial interstitium in tissue sections (Figure 6B), coincident with the maintenance of the peribronchiolar extracellular matrix (ECM) architecture (Figure 6A). In contrast, in OVA-exposed animals, there was a diffuse increase of birefringence (Figure 6E) in peribronchiolar ECM seen in H&E preparations (Figure 6D). Figure 6H shows that chronic L-NAME treatment increased collagen deposition around the distal airway wall without interfering with elastic content (Figure 6I). Figures 6L and 6M show that chronic 1400W treatment decreased both collagen and elastic fiber deposition around the distal airway wall, with a reduction in inflammatory cell infiltrate (Figure 6J).
Quantification of Airway Remodeling Measurements of airway remodeling are shown in Figure 7. Airway wall area and the collagen and elastic airway content were similar in all saline-exposed animals. Airway wall area was significantly higher in OVA-exposed animals (Figure 7A) compared with the saline-exposed groups (P < 0.05). Neither L-NAME nor 1400W treatments affected airway wall area. OVA-exposed animals also showed greater values of percent collagen compared with saline-exposed animals (P < 0.05; Figure 7B). L-NAME-treatment resulted in an increase in collagen deposition in airway wall, and treatment with 1400W resulted in a decrease in this response (P < 0.05 compared with OVA group).
We observed that OVA-exposed animals had an increase in the content of elastic airway fibers compared with the saline-exposed groups (P < 0.05; Figure 7C). Only 1400W treatment reduced this response (P < 0.05 compared with OVA and OVA-L groups).
The role of NO in patients with asthma and in animal models of allergic airway inflammation, as well as the relative contribution of each of the NOS isoforms for the production of this molecule, remains incompletely understood. In the present study, we intended to clarify the importance of cNOS and iNOS isoforms in an experimental model of chronic allergic airway inflammation. In this regard, we used L-NAME, a nonselective false substrate that inhibits NO produced by all NOS isoforms, and 1400W, an iNOS inhibitor. Because 1400W is specific and highly selective for iNOS inhibition, the differences observed among the experimental groups is probably related to the absence of cNOS-derived NO in the experimental group that received L-NAME treatment. Previous studies had demonstrated that 1400W is one of the most selective inhibitors of iNOS. In addition, 1400W presents an in vivo selectivity at least 100-fold greater than other NOS inhibitors, such as aminoguanidine (13, 16). Other iNOS inhibitors are, at most, 30-fold more potent against iNOS than against eNOS, whereas 1400W is over 5,000- and 200-fold more potent against iNOS than it is against eNOS and nNOS, respectively (13, 16). It is important to emphasize that the differences in the dose schedules of 1400W and L-NAME used in the present study were related to their pharmacokinetics, as previously determined (13, 17). Because high doses of 1400W (50 mg/kg, intraperitoneally) can cause some toxic effects (13, 16) related to the nervous and cardiovascular systems, including induction of convulsions, we used a lower dose (2 mg/kg, intraperitoneally, for 4 d). That choice of dose was based on a study that found a significant reduction in LPS-induced vascular leakage using doses of 1400W within the range of 15 mg/kg (13). Although the effects of these treatments may be influenced by continuous or pulsatile NOS inhibition, their selectivity was demonstrated by analyzing the results of exhaled NO and immunohistochemical studies detecting eosinophils positive for nNOS and iNOS (Figures 2 and 4). Exhaled NO has been considered a useful marker of airway inflammation in human asthma (1, 2) and in animal models of allergic inflammation (36). In the present study, OVA-exposed animals presented an increase in exhaled NO compared with saline-exposed animals. In experimental models of pulmonary inflammation, the concentrations of NO detected in the exhaled air have been predominantly used to evaluate the efficacy of tested treatments. Noteworthy is the observation that the values of exhaled NO observed in saline-exposed animals (20 ppb) are similar to the values found in other studies with guinea pigs (3). Interestingly, all animals receiving L-NAME presented a significant reduction in the concentrations of exhaled NO. In contrast, 1400W reduced the concentration of exhaled NO only in OVA-exposed guinea pigs. Evaluating the number of eosinophils positive for NOS, we observed that L-NAME treatment reduced the number of eosinophils positive for both NOS enzymes; however, 1400W reduced only the iNOS-positive eosinophils, without modifying the number of cells positive for nNOS. Despite the fact that these results confirmed the effectiveness of both treatments in exhaled NO reduction, they also suggest that 1400W treatment had high selectivity to iNOS isoforms that are present mainly in inflammatory situations. We have previously studied this model of chronic allergic airway inflammation in guinea pigs (46, 12). Guinea pigs chronically exposed to OVA presented hyperresponsiveness to methacholine, an intense bronchoconstriction after antigen challenge, and an increase in the number of eosinophils and CD4+ lymphocytes in both bronchoalveolar lavage and lung tissues. In fact, in this experimental model, eosinophils comprise 7075% of inflammatory cells in bronchoalveolar lavage fluid (12). Recently, we have also demonstrated that, in this experimental model, there is an increase in the number of MN cells and eosinophils positive for iNOS around airways of OVA-exposed guinea pigs (5). In addition, we observed an increase in the content of collagen fibers around airway wall (6). In the present study, we observed that NO produced by iNOS increases both collagen and elastic fiber content in the airway wall. Additionally, it contributes to bronchoconstriction and amplifies inflammatory responses. The inhalation time is the time that the animals stayed in contact with antigen without respiratory distress symptoms, and reflects an acute response to antigen exposure. After the fifth inhalation, OVA-exposed animals stayed in contact with antigen for a lower length of time compared with saline-exposed animals (46, 12), and neither L-NAME nor 1400W treatments influenced the inhalation time in the seventh inhalation compared with vehicle-treated animals. These results suggest that NO did not interfere with acute responses to antigen exposure. Although NO has been considered one of the mediators that influence airway smooth muscle responses (2, 18), other factors are involved in acute responses to antigen contact, including activation of different pathways that control airway smooth muscle contraction and mast cell degranulation (19). We observed that both L-NAME and 1400W treatments increased the resistance of respiratory system in saline-exposed animals. Because the role of iNOS is more pronounced in inflammatory situations, few studies have evaluated the effects of iNOS inhibition on physiologic situations. We have previously shown that there is a basal expression of iNOS in resident cells around airways in guinea pigs not exposed to an inflammatory stimulus (5). In addition, Guo and colleagues (20) showed that iNOS is continuously produced by airway epithelium in normal humans. Interestingly, the effects of 1400W treatment in the increase of Rrs were more pronounced than those observed with L-NAME treatment, regardless of chronic OVA exposure. These data are in agreement with previous studies that suggested the presence of a cNOS bronchodilation effect (2, 9, 18). In addition, we suggest that NO produced by iNOS can also contribute to the control of airway smooth muscle responses. In contrast to the increase in Rrs observed with NOS inhibition, both L-NAME and 1400W treatments decreased baseline values of Ers in saline-exposed guinea pigs, suggesting an effect of NO on small airways and/or distal airspaces. However, it was previously shown that the effects of NO are more pronounced in proximal airways because there is a decrease in nitrergic nerves toward distal airways (18). In this context, Lilly and colleagues (21) showed that vasointestinal peptide injection in isolated guinea pig trachea has the capacity to relax whole lungs in part by stimulating the generation of NO due to nNOS activation. One possible explanation for the decrease in Ers induced by NOS inhibition is an indirect interference of NO with surfactant A by peroxynitrite production, modifying the alveolar tension, as previously suggested by others (22). Hickman-Davis and colleagues (23) demonstrated that surfactant protein A may help regulate NO production. In addition, Que and colleagues (24) demonstrated that mice with genetic deletion of S-nitrosoglutathione reductase do not present bronchial hyperresponsiveness, and this effect could be related to an endogenous effect of S-nitrosothiols controlled by S-nitrosoglutathione reductase. Discordance between airway mechanics and airway inflammation may reflect different NOS products: S-nitrosoglutathione (GSNO) versus NO (iNOS) (24). In the present study, we also observed these divergences, as L-NAME partially reduced inflammation but induced bronchoconstriction. However, the evaluation of iNOS effects by 1400W treatment showed a similar effect on the attenuation of both inflammatory and mechanical parameters. Many studies have focused on the role of NO in the modulation of airway smooth muscle contraction in different models of experimental pulmonary allergic inflammation (5, 6, 2527). It has been previously shown that NO that is mainly derived from constitutive isoforms of NOS attenuates bronchoconstriction induced by allergen in sensitized experimental animals (2, 9, 25). In contrast, others have observed that nNOS-derived NO could contribute to airway constriction (28). Our results suggest a protective effect of NO derived from cNOS, as the increase in Rrs induced by antigen challenge was significantly greater in guinea pigs treated with L-NAME, but not in the experimental group that received 1400W (Figure 3). In addition, the iNOS isoform contributed to airway constriction, mainly in the proximal airways, because the group of guinea pigs that received 1400W presented an increase in Rrs induced by OVA challenge lower than that seen in the group that did not receive NOS antagonists. It has been previously suggested that iNOS could contribute to antigen-induced airway constriction. Several mechanisms reported in the literature tried to explain how NO could interfere in airway tone (2) and the ability of NO to control airway tone could be related to both GMPc-dependent and independent mechanisms (2931). We demonstrate that only L-NAME treatment was involved in the attenuation of peribronchial edema. The effects of NO in peribronchial edema could be an effect of NO inhibition reducing airway inflammation. In this context, it was expected that 1400W treatment attenuated the peribronchial edema formation since this treatment reduced airway inflammation more than L-NAME treatment. However, the effects of L-NAME in edema formation could be attributed to other effects of NO, including vasodilation and modulation of capillary leakage (32). There have been several studies showing that NO plays a role as a vasodilator, and that this function is probably derived from NO produced by constitutive isoforms (2, 6, 9). The precise role of NO in airway inflammation remains unclear, and depends on experimental protocol and the inhibitors used. We observed that L-NAME treatment diminished only MN cells in the airway wall, and specific iNOS inhibition with 1400W reduced both eosinophilic and MN cell recruitment, confirming previous reports suggesting a proinflammatory iNOS effect (25, 33). Although we have recently observed that acute treatment with L-NAME inhibits eosinophilic recruitment in guinea pigs with chronically allergic pulmonary inflammation (5, 6), we did not obtain the same results after chronic L-NAME treatment (6). Regarding this issue, Birrel and colleagues (34) showed that L-NAME, but not 1400W, acute treatments attenuated eosinophilic recruitment. These differences could be related to a compensatory increase in iNOS (35), no effects on eosinophilopoiesis (36), or differences in apoptosis modulation (37). Taylor and colleagues (37) suggested that NO could have different effects on inflammatory cell apoptosis (anti- and pro-apoptotic properties), depending on the concentration, flux, and source of NO. There is a very small number of previous studies comparing L-NAME and 1400W in models of airway or pulmonary inflammation, and none evaluating airway remodeling. McCluskie and colleagues (38) evaluated the effects of NO in an aerosolized LPS-driven animal model of airway inflammation. Their study involved the assessment of exhaled NO and the inflammatory response in mice treated with L-NAME (100 mg/kg) or 1400W (30 or 100 mg/kg) 2 h before and 1 after challenge. The treatments with both drugs resulted in a similar decrease in exhaled NO and airway neutrophilia. Iijima and colleagues (39) also evaluated the relationship between eosinophil recruitment and NO in sensitized mice using 1400W (1.0 mg/kg) and L-NAME (10 mg/kg) administered 0.5 h before and 8, 20, 32, and 44 h after OVA challenge. The authors demonstrated that both L-NAME and 1400W reduced the number of eosinophils in the bronchoalveolar lavage fluid. Airway remodeling, characterized by structural changes in the airway wall, has an important clinical significance by contributing to the irreversibility of lung function alterations observed in patients with asthma (11). Our study showed some important results concerning the airway remodeling process. First, NO derived from cNOS protected against airway collagen deposition without interfering with elastic fiber content. Second, the specific inhibition of iNOS by 1400W treatment decreased both collagen and elastic fiber content in the airway wall. The relevance of iNOS activation to the airway remodeling process was poorly defined, and other authors have been unable to observe the same result (10). These differences could be related to the use of genetically NOS-deficient mice. We suggest that conflicting results have been described using NOS knockout mice, and the maturation and development of the innate immune system may be important for the expression of the allergic response. The present study evaluated whether NO pathways are involved in airway remodeling, mainly considering the ECM component. Recently, some studies with different experimental protocols have suggested some possible mechanisms (7, 26, 40, 41). Hogaboam and colleagues (7) studied a model of nonfibrotic lung granuloma. They showed that L-NAME induces an increase in C-C chemokine receptors, CCR2 and CCR3, mRNA and reduces macrophage chemoattractant protein-1 and eotaxin in isolated lung fibroblast culture, increasing the collagen content. Previous reports suggested that arginase could interfere with myofibroblast differentiation, increasing collagen production. Meurs and colleagues (26) also suggested that allergen-induced deficiency of cNOS-derived NO enhances arginase activity. Finally, NO can have some effects in metalloproteinases, modifying collagen degradation (40). We found no other studies that evaluated the effects of iNOS-derived NO on elastic fiber deposition in asthma models. One possible explanation for the inhibition of collagen and elastic fiber content in airways may be related to the effects of NO on the modulation of inflammatory response, modifying cytokine or chemokine profiles and neutrophilic responses (8, 33). However, more studies are needed to clarify these unsolved questions. In conclusion, the nonspecific inhibition of NO production by L-NAME treatment in guinea pigs with allergen-induced chronic pulmonary inflammation potentiates the bronchoconstiction and the airway remodeling, particularly the airway collagen deposition. In contrast, blockage of iNOS-derived NO attenuates antigen-induced airway constriction, inflammatory, and remodeling processes, reducing both collagen and elastic fiber deposition.
The authors thank David I. Kasahara and Beatriz M. Saraiva-Romanholo for their assistance in L-NAME administration and exhaled NO measurements, respectively. They also thank Professor Brendan J. R. Whittle (William Harvey Research Institute, London, UK) for helping them to choose the exact dose of 1400W.
This work was supported by Fundação de Amparo à Pesquisa do Estado de São Paulo and Conselho Nacional de Desenvolvimento Científico and Tecnológico, Brazil. This study was presented, in part, at the International Meeting of the European Respiratory Society in Copenhagen, 2005. Originally Published in Press as DOI: 10.1165/rcmb.2005-0391OC on May 18, 2006 Conflict of Interest Statement: None of the authors has a financial relationship with a commercial entity that has an interest in the subject of this manuscript. Received in original form October 18, 2005 Accepted in final form May 11, 2006
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||