INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Ozone, the primary pollutant of photochemical smog, is a major health concern. Ozone inhalation produces airway inflammation and detrimental effects on lung function. In controlled human exposure studies, ozone induces airway inflammation characterized by increased neutrophil number, protein content, and proinflammatory cytokine concentrations in bronchoalveolar lavage fluid (BALF) (1). In addition, ozone has been shown to cause an increase in bronchoreactivity to methacholine in asthmatic and healthy individuals and to allergen in asthmatic individuals (5). High ambient ozone concentrations are associated with decrements in lung function and an overall increase in respiratory symptoms (6). Epidemiologic studies correlate high ambient ozone concentrations with increased hospital visits for respiratory illnesses, including asthma (7, 8).

One mechanism by which ozone exposure may cause lung injury is through the release of tachykinin neuropeptides from sensory nerves. Sensory nerve fibers in the airways are found beneath the epithelium, near smooth muscle cells and submucosal glands, and around arterial vessels (9). In the airways, release of tachykinins such as substance P from sensory nerves can stimulate mucus secretion, bronchoconstriction, plasma exudation, and immune cell responses (10). Inhaled irritants that can stimulate airway sensory nerve fibers to release tachykinins include capsaicin (11), cigarette smoke (12), sulfur dioxide (13), and ozone (14). Although tachykinins have been documented to be released from sensory nerves after ozone exposure (14), it is unclear whether these neuropeptides play a protective or contributing role in ozone-induced lung injury.

Acute exposures to ozone in animals have provided conflicting results regarding the role of sensory nerves in ozone-induced lung injury. Experiments using ozone-exposed guinea pigs depleted of tachykinin-containing sensory nerve fibers by capsaicin treatment suggested that sensory neurons contribute to ozone-induced airway inflammation. Capsaicin treatment attenuated ozone-induced airway hyperresponsiveness and neutrophilic inflammation (15) and decreased airway permeability as determined by Evans blue dye extravasation (16). However, results obtained from ozone exposure experiments in rats treated neonatally with capsaicin suggested that capsaicin-sensitive sensory fibers act to protect the lung from ozone damage, thereby reducing inflammation. Rats that were treated with capsaicin as neonates and exposed to ozone as adults demonstrated increased airway hyperreactivity (17) and bronchoalveolar lavage (BAL) neutrophilia (18) compared with ozone-exposed vehicle-treated rats.

In the present study we have used novel methods to study the role of sensory neurons in ozone-induced lung inflammation. We have examined the response to ozone in two types of genetically manipulated mice that exhibit altered innervation of the airways. We have used CCSP-NGF transgenic mice, which overexpress nerve growth factor (NGF) from the airway epithelial-specific Clara cell secretory protein (CCSP) promoter and exhibit a hyperinnervation of the airways by sympathetic and tachykinin-containing sensory nerve fibers (19). We have also used mice deficient in the low-affinity nerve growth factor receptor p75 gene (NGFR), which exhibit deficits in tachykinin-containing sensory nerve fibers (20). In addition, we treated mice with neurokinin receptor antagonists to block the action of the tachykinins released from sensory nerves during ozone exposure. Lastly we examined cytokine production in the different types of mice as a possible mechanism by which sensory neurons affect ozone-induced lung inflammation. Our results demonstrate that in contrast to previous studies in rats, tachykinins promote ozone-induced airway inflammation in mice.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animals

CCSP-NGF transgenic mice were described in greater detail previously (19). For this study, CCSP-NGF mice were backcrossed onto a C57BL/6 background for 6 to 8 generations. Mice carrying a mutation of the gene encoding the low-affinity NGF receptor p75^NGFR were obtained courtesy of Kuo-Fen Lee (Massachusetts Institute of Technology, Cambridge, MA). Mice heterozygous for the NGFR null mutation were backcrossed to C57BL/6 mice for 5 to 6 generations and then interbred to generate homozygous mutant mice on a C57BL/6 background. C57BL/6 mice were obtained from Jackson Laboratories (Bar Harbor, ME). All mice were housed in an AAALAC (Association for Assessment and Accreditation of Laboratory Animal Care) accredited facility according to National Institutes of Health guidelines. Mice were maintained on a 14-h light/10-h dark schedule and given food and water ad libitum.

Ozone Exposure

Exposures were conducted in a 24-cubic-foot stainless steel horizontal laminar flow exposure chamber (Baker Company, Sanford, ME). During exposure mice were housed separately in individual compartments of stainless steel wire cages. Ozone was generated by passing oxygen through a small chamber containing an ultraviolet lamp (UVP, Inc., Upland, CA) and diluted to the desired concentration by mixing with filtered air. Ozone concentration in the chamber was monitored with a Dasibi 1003 ozone analyzer (Glendale, CA), and the desired concentration was maintained by adjusting the intensity of the ultraviolet lamp with a rheostat. For control air-exposed mice, an identical procedure was followed, but the ultraviolet light was not turned on. After a 3-h exposure, mice were returned to their cages and given food and water ad libitum.

Analysis of Airway Innervation

At 18 h after exposure to ozone, lungs were fixed by intratracheal instillation and embedded in paraffin. Tissue sections were stained with rabbit polyclonal antibodies to synaptophysin (Dako, Carpinteria, CA) diluted 1:100, and bound antibodies were detected with biotinylated goat anti-rabbit antibodies (Jackson ImmunoResearch, West Grove, PA) diluted 1:3,500. Sections were incubated with 500 ng/ml streptavidin-conjugated horseradish peroxidase (Jackson ImmunoResearch) followed by diaminobenzidine. Substance P was measured in lung homogenates of normal and NGFR knockout mice as previously described (19) using a commercially available enzyme immunoassay (Cayman Chemical, Ann Arbor, MI).

Neurokinin Receptor Antagonists

Approximately 30 min before ozone exposure, mice were treated with a neurokinin receptor antagonist, a combination of receptor antagonists, or vehicle. The NK1 receptor antagonist used in this study, RP 67580 (2-[1-imino-2-(2 methoxy phenyl) ethyl]-7,7 diphenyl-4-perhydroisoindolone (3aR, 7aR)), was generously supplied by Rhone-Poulenc Rorer (Vitry sur Seine, France). RP 67580 was dissolved at 45 mg/ml in 0.1 N methanesulfonic acid, diluted with water to 10 mg/ml and then diluted with 0.9% NaCl for administration intraperitoneally at 1 mg/kg. The NK2 and NK3 receptor antagonists used in this study were SR 48968 {(S)-N-methyl-N-[4-(-acetylamino-4-phenylpiperidino)-2-(3,4-dichloro-phenyl)-butyl] benzamide} and SR 142801 [(S)- (N)-(1-(3-(1-benzoyl-3-(3,4-dichlorophenyl) piperidin-3-yl) propyl)-4-phenylpiperidin-4-yl)-N-methylacetamide], respectively. Sanofi Recherche (Montpellier, France) generously supplied these reagents. SR 48968 and SR 142801 were dissolved in 0.01% Tween 80, diluted with 0.9% NaCl, and administered intraperitoneally at 1 mg/kg.

BAL

Mice were lavaged with 5 × 0.8 ml ice-cold lavage buffer (137 mM NaCl, 2.7 mM KCl, 4.3 mM Na₂HPO₄.7H₂O, 1.4 mM KH₂PO₄, 0.4 mM ethylenediaminetetraacetic acid [EDTA]). The first lavage sample was spun at 3,100 × g for 5 min, and the supernatant was frozen at 70° C for subsequent protein and cytokine assays. The cell pellet was resuspended and pooled with the remaining lavage samples. The total cell count was recorded, and 5 × 10⁴ cells were spun onto a slide and stained with Diff-Quik (DADE, Miami, FL). Cell differential was performed on 400 cells by an investigator who was unaware of the identity of the samples.

Protein and Cytokine Analyses

The concentration of protein in the lavage samples was determined using a commercially available Bradford assay (Bio-Rad Laboratories, Hercules, CA) with bovine serum albumin as the standard. Mouse KC and interleukin 6 (IL-6) concentrations in the lavage samples were measured using Quantikine M (R&D Systems, Minneapolis, MN), a quantitative sandwich ELISA. Tumor necrosis factor- (TNF-) was measured by ELISA using reagents obtained commercially (Endogen, Woburn, MA).

Statistical Analysis

Data are expressed as mean ± SEM. Group means were compared using Student's t test or analysis of variance (ANOVA) followed by Fisher's protected least significant difference test. For neutrophil data, a square root transformation was performed before statistical analysis to achieve normally distributed data and equal variance among groups.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Altered Innervation in CCSP-NGF Transgenic and NGFR Knockout Mice

To demonstrate altered innervation in CCSP-NGF transgenic mice, lung sections were stained with antibodies against synaptophysin, a synaptic vesicle protein that serves as a neuronal marker (Figure 1). This analysis indicated that the airways in CCSP-NGF mice were surrounded by a dense network of neuronal fibers (Figures 1A and 1D). We previously showed that lungs from these mice exhibited increased substance P-immunoreactive fibers and a fivefold increase in substance P concentrations over nontransgenic mice (19). Histologically lungs from NGFR knockout mice (Figures 1C and 1F) appeared grossly similar to wild-type mice (Figures 1B and 1E). NGFR knockout mice were previously shown to have deficits in tachykinin-containing sensory nerve fibers and functional impairment in nociception in the skin, but the effect of the mutation in the lung was not examined (20). To quantitate tachykinin innervation in NGFR knockout mice, lung substance P concentrations were measured by enzyme immunoassay. The amount of substance P in the lungs of NGFR knockout mice was reduced approximately 50% compared with wild-type mice (Figure 2). These data demonstrate that the innervation of the lung is altered in CCSP-NGF transgenic and NGFR knockout mice.

View larger version (171K): [in this window] [in a new window]   Figure 1.   Lung histology in mice with altered airway innervation. Mice were exposed to 1.5 ppm ozone for 3 h or were sham-exposed to air. At 18 h after exposure, lungs were fixed, embedded in paraffin, and subjected to immunohistochemical analysis for synaptophysin as described in METHODS. A-C are air-exposed, D-F are ozone-exposed; A and D are CCSP-NGF transgenic mice, B and E are wild-type mice, and C and F are NGFR knockout mice. Arrows indicate nerve fibers underneath the epithelium that exhibited synaptophysin immunoreactivity. Arrowheads indicate inflammatory infiltrate present in the airways of ozone-exposed animals.

View larger version (16K): [in this window] [in a new window]   Figure 2.   Substance P levels in NGFR knockout and wild-type mice. Substance P was measured in lung homogenates by enzyme immunoassay as described (19). NGFR knockout mice (n = 5) had significantly lower concentrations of substance P than wild-type mice (n = 5). *p < 0.05.

Effect of Airway Innervation on Ozone-induced Lung Inflammation

CCSP-NGF transgenic, NGFR knockout, and wild-type mice were exposed to ozone to determine whether the genetic alterations in these mice affected the inflammatory response. Figure 1 shows lung histology in CCSP-NGF ( panel D), wild-type ( panel E ), and NGFR knockout mice ( panel F ) 18 h after exposure to 1.5 ppm ozone for 3 h. This time point was shown in pilot studies to correspond to the peak of neutrophil infiltration into the airways. Ozone-exposed mice exhibited isolated clusters of inflammatory infiltrates composed of neutrophils and macrophages in the airways. Ozone-induced inflammation appeared qualitatively similar in the three types of mice.

Cellular influx into the airways was quantitated by performing differential cell counts on cells isolated by lung lavage. The total number of cells recovered by lavage was not significantly different among the three types of mice 18 h after ozone exposure. CCSP-NGF mice demonstrated a significant increase in the number of neutrophils in BALF over wild-type mice, whereas NGFR knockout mice exhibited a significant decrease compared with wild-type mice (Figure 3). NGFR knockout mice had a significantly higher number of macrophages than wild-type and transgenic mice at this time point (Figure 3). NGFR knockout mice tended to have more lymphocytes than wild-type and CCSP-NGF mice, both in exposed and unexposed mice (Figure 3 and data not shown). However, the number of lymphocytes was not different between exposed and unexposed mice and in none of the groups did lymphocytes exceed 1% of total cells. No difference in the number of eosinophils was observed among the different types of mice (Figure 3) or between air- and ozone-exposed mice (data not shown).

View larger version (20K): [in this window] [in a new window]   Figure 3.   BAL cell profile in ozone-exposed mice. BAL was performed 18 h after exposure to 1.5 ppm ozone, and cell differentials were determined. CCSP-NGF mice (n = 8) exhibited increased neutrophilic inflammation compared with wild-type mice (n = 17) whereas NGFR knockout mice (n = 9) demonstrated decreased neutrophilic inflammation compared with wild-type mice. *p < 0.05; **p < 0.01.

To determine whether the observed difference in neutrophil influx represented a change in the peak neutrophil response or an alteration in the kinetics of the response, mice were also lavaged 4 and 48 h after exposure to 1.5 ppm ozone. In air-exposed mice and in ozone-treated mice 4 and 18 h after exposure, the total number of cells recovered by lavage was not significantly different among the three types of mice. At 48 h after ozone exposure, CCSP-NGF transgenic mice had a significantly higher number of total cells (27.3 ± 2.1 × 10⁴) than wild-type (19.6 ± 2.8 × 10⁴; p < 0.05) or NGFR knockout mice (18.4 ± 1.8 × 10⁴; p < 0.01). For all the types of mice, the highest numbers of neutrophils were recovered at 18 h and were lower at 4 and 48 h after ozone exposure (Figure 4). The numbers of neutrophils recovered in BAL of CCSP-NGF and NGFR knockout mice were not different from wild-type C57BL/6 mice in air-exposed mice or at the the 4-h time point; however, CCSP-NGF mice had more neutrophils than NGFR knockout mice at the 48-h time point. These results indicate that mice with altered airway innervation exhibit similar kinetics of neutrophilic inflammation, but an altered peak response. In addition, no significant differences in neutrophil number were observed among air-exposed NGFR knockout, C57BL/6, and CCSP-NGF mice (Figure 4). These data indicate that manipulation of airway innervation can alter the magnitude of ozone-induced neutrophilic inflammation and that tachykinin-containing sensory neurons surrounding the airways are involved in producing or amplifying the effects of ozone.

View larger version (22K): [in this window] [in a new window]   Figure 4.   Time course of neutrophilic inflammation in genetically altered mice. NGFR knockout mice, wild-type C57BL/6 mice, and CCSP-NGF transgenic mice were exposed to air or 1.5 ppm ozone. Inflammation in ozone-exposed mice was assessed by lavage 4, 18, and 48 h after exposure. The number of neutrophils in lavage fluid at the different times is shown. n = 4 per group for air exposed, n = 8 per group for 4 h, n = 8 to 17 per group for 18 h, and n = 8 per group for 48 h. *p < 0.05; **p < 0.01.

To determine the extent of plasma leakage into the airways of the genetically altered mice, protein concentration was measured in BALF of NGFR knockout, C57BL/6, and CCSP-NGF mice 4, 18, and 48 h after exposure to 1.5 ppm ozone. As shown in Figure 5, the concentration of protein in BAL increased in a time-dependent manner after exposure but no significant differences were observed among the different types of mice. In addition, no significant differences were present among the air-exposed animals.

View larger version (25K): [in this window] [in a new window]   Figure 5.   Protein concentration in BALF. NGFR knockout mice, C57BL/6 mice and CCSP-NGF transgenic mice were exposed to air or 1.5 ppm ozone. Protein concentration in lavage fluid was assessed at 4, 18, and 48 h after exposure to ozone. No significant differences in BAL protein content were found among the different types of mice.

Tachykinin Receptor Antagonists

To examine the role of tachykinins released from sensory neurons surrounding the airways in modulating ozone-induced lung inflammation, mice were treated with neurokinin receptor antagonists. In our initial studies, we treated wild-type C57BL/6 mice with neurokinin receptor antagonists and assessed the effects 18 h after ozone exposure. C57BL/6 mice were treated with the NK1 receptor antagonist RP 67580 (21), the NK2 receptor antagonist SR 48968 (22), the combination of RP 67580 and SR 48968, or vehicle before exposure to 1.5 ppm ozone. As shown in Figure 6, mice receiving RP 67580 (RP) had a significantly lower number of neutrophils in BALF as compared with vehicle-treated mice (Veh), whereas mice treated with SR 48968 (SR) were not significantly different from vehicle-treated control mice. In addition, mice receiving both RP 67580 and SR 48968 (RP + SR) had a significantly lower number of neutrophils compared with treatment with SR 48968 alone, confirming the inhibitory effect of NK1 receptor blockade. Overall, treatment with the NK2 receptor antagonist SR 48968 resulted in a trend toward more neutrophils (p = 0.07), which may explain why combined treatment with RP 67580 and SR 48968 did not result in a significant reduction compared with vehicle-treated mice. Treatment with the NK3 receptor antagonist SR 142801 had no effect on ozone-induced neutrophilic inflammation (not shown). These data show that blocking the actions of tachykinins at the NK1 receptor reduces the magnitude of the neutrophilic infiltrate into the airways of ozone-exposed mice and suggest that tachykinins promote ozone-induced neutrophilic inflammation in wild-type C57BL/6 mice.

View larger version (13K): [in this window] [in a new window]   Figure 6.   Effect of neurokinin receptor antagonists on ozone-induced neutrophilic inflammation in wild-type C57BL/6 mice. RP 67580 (RP), SR 48968 (SR), a combination of the two antagonists (RP + SR), or vehicle (Veh) were administered to groups of 16 mice before exposure to 1.5 ppm ozone. Inflammation was assessed 18 h later by lung lavage. The number of neutrophils recovered for each group is shown. Treatment with the NK1 receptor antagonist significantly inhibited neutrophilic inflammation compared with vehicle-treated mice. *p < 0.05; **p < 0.01.

CCSP-NGF transgenic mice were treated with neurokinin receptor antagonists and the effects on ozone-induced neutrophilic inflammation were assessed. As shown in Figure 7, mice receiving RP 67580 had a significantly lower number of neutrophils in BALF 18 h after exposure compared with vehicle-treated ozone-exposed mice. As with wild-type mice, treatment of CCSP-NGF mice with the combination of RP 67580 and SR 48968 did not result in a significant reduction in neutrophils compared with vehicle-treated mice. These data suggest that tachykinins acting through NK1 receptors are involved in producing the increased neutrophilic inflammation observed in CCSP-NGF mice.

View larger version (12K): [in this window] [in a new window]   Figure 7.   Effect of neurokinin receptor antagonists on ozone-induced neutrophilic inflammation in CCSP-NGF mice. RP 67580 (RP), a combination of RP 67580 and SR 48968 (RP + SR), or vehicle (Veh) were administered to groups of 16 to 17 mice before exposure to 1.5 ppm ozone. Neutrophil influx into the airways was assessed 18 h later by lung lavage. Treatment with the NK1 receptor antagonist RP 67580 significantly inhibited neutrophilic inflammation in CCSP-NGF mice. *p < 0.05.

Cytokine Analyses

One potential mechanism by which neurokinin receptor signaling may modulate an inflammatory response is through control of cytokine gene expression. The recruitment of leukocytes is a hallmark of the inflammatory response. KC, a member of the CXC chemokine family, mediates neutrophil chemotaxis and activation (23). TNF- is a proinflammatory cytokine and a major regulator of chemokine gene expression (24). IL-6 possesses proinflammatory properties and is detected at increased concentrations in BALF after acute ozone exposures in humans (3). The concentrations of KC, TNF-, and IL-6 were measured in the BALF of NGFR knockout, C57BL/6, and CCSP-NGF mice exposed to 1.5 ppm ozone or air (Figure 8). Cytokine concentrations were measured in BALF collected 4 and 18 h after exposure. As shown in Figure 8A, the concentration of KC 4 h after ozone exposure was significantly higher in CCSP-NGF mice compared with both C57BL/6 and NGFR knockout mice. At 18 h postexposure, the level of KC in BALF was significantly higher in C57BL/6 than in NGFR knockout mice; however, no significant differences were apparent between the CCSP-NGF mice and C57BL/6 mice (Figure 8B). The concentration of TNF- was also significantly higher in CCSP-NGF mice compared with C57BL/6 mice 4 h after exposure (Figure 8C). At 18 h after ozone exposure, the concentrations of TNF- were increased in all groups of mice; however, no significant differences were found among the different types of mice (Figure 8D). IL-6 concentrations were higher at 4 h than at 18 h after exposure to ozone, but no significant differences in IL-6 concentrations were found among CCSP-NGF, C57BL/6, and NGFR knockout mice at any time point (Figures 8E and 8F). Concentrations of the cytokines in air-exposed mice were at or below the level of detection of these assays (not shown). These data indicate that airway innervation can modulate proinflammatory gene expression.

View larger version (22K): [in this window] [in a new window]   Figure 8.   Lavage fluid cytokine concentrations of ozone-exposed mice. The cytokines KC (A and B), TNF- (C and D), and IL-6 (E and F ) were measured in BALF by ELISA at 4 h (A, C, and E ) and 18 h (B, D, and F ) after exposure to 1.5 ppm ozone. KC was significantly higher in CCSP-NGF transgenic mice compared with wild-type and NGFR knockout mice at 4 h and lower in NGFR knockout mice compared with wild-type mice at 18 h. Compared with C57BL/6 mice, TNF- was also significantly higher in CCSP-NGF mice at 4 h after exposure. No significant differences in the concentration of IL-6 were observed among the three types of mice at 4 or 18 h after ozone exposure. *p < 0.05; **p < 0.01.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

We have used genetic manipulation of airway innervation in mice and treatment with neurokinin antagonists to examine the role of airway sensory nerves in ozone-induced lung inflammation. Genetically altered mice with increased innervation of the airways by tachykinin-containing sensory nerve fibers exhibited an approximately 2-fold increase in the peak number of neutrophils in BALF after ozone exposure. Mice with decreased sensory innervation exhibited an approximately 50% reduction in the number of neutrophils at the same time. Experiments in which mice were treated with neurokinin receptor antagonists confirmed that tachykinin signaling promotes ozone-induced lung inflammation. Treatment of wild-type mice with tachykinin receptor inhibitors resulted in a maximal 50% inhibition of ozone-induced neutrophilic inflammation. Treatment of CCSP-NGF transgenic mice resulted in a maximal 30% inhibition of neutrophil influx compared with vehicle-treated transgenic mice. The fact that antagonist-treated CCSP-NGF mice exhibited more severe inflammation than wild-type mice on an absolute basis suggested that higher doses of neurokinin receptor antagonists may be required because of the increased capacity of CCSP-NGF mice to release tachykinins. Alternatively, other mechanisms besides tachykinin signaling, e.g., direct effects of NGF (25), may contribute to the inflammation observed in CCSP-NGF mice. Our results indicate that airway innervation can modulate the inflammatory response to ozone and suggest that airway sensory nerves are involved in promoting or enhancing ozone-induced neutrophilic inflammation in mice.

Previous studies using either capsaicin to deplete tachykinins or neurokinin receptor antagonists to block tachykinin signaling have indicated that sensory nerves enhance ozone-induced lung injury in guinea pigs. Capsaicin pretreatment reduced ozone-induced inflammation, vascular permeability, and airway hyperreactivity (15, 16, 26). Guinea pigs treated with FK-244, an antagonist for NK1 and NK2 receptors, had significantly fewer neutrophils in BALF compared with vehicle-treated animals (27). In conjunction with these results, our data indicate that release of tachykinins from airway sensory nerves triggers common proinflammatory pathways in both mice and guinea pigs.

In contrast, results obtained from ozone exposure experiments using rats treated with capsaicin or neurokinin receptor antagonists suggest that capsaicin-sensitive sensory fibers act to protect the lung from ozone damage thereby reducing inflammation. Capsaicin treatment of neonatal rats resulted in increased interstitial inflammation and epithelial injury after ozone exposure (28). Similarly, ozone-exposed capsaicin-treated rats demonstrated airway hyperreactivity (17) and increased BAL neutrophilia (18) compared with vehicle-treated animals exposed to ozone. Rats administered a combination of NK1 and NK2 receptor antagonists exhibited increased neutrophilic inflammation when analyzed 4 h after ozone exposure (18). These results show that there can be species-specific modulation of ozone-induced lung injury by airway sensory nerves, with tachykinins promoting inflammation in mice and guinea pigs and inhibiting inflammation in rats. Although previous studies in other species employed capsaicin to destroy sensory nerves and we used genetic manipulations to alter the innervation of the airways, the effects of tachykinins on ozone-induced inflammation were confirmed by the use of neurokinin receptor antagonists in these studies. These data argue for real species differences in the control of ozone-induced inflammation by tachykinins as opposed to differences in the methodology used to manipulate neurokinin receptor signaling. We identified NK1 as the receptor subtype promoting ozone-induced inflammation in mice. In the guinea-pig and rat studies, ozone-induced inflammation was altered by combined blockade of NK1 and NK2 receptors, but the relative contribution of signaling through the individual receptors was not determined (18, 27).

Our results indicated that RP 67580, which specifically binds to NK1 receptors, was the most effective antagonist in inhibiting ozone-induced lung inflammation. These observations implicate substance P signaling through NK1 receptors as the most likely mechanism of the observed effects on inflammation. This concept is in agreement with previous studies evaluating the role of substance P and the NK1 receptor in the regulation of inflammation. Substance P stimulates adhesion molecule gene expression, releases neutrophil chemotactic activity, and enhances neutrophil migration in vitro (29). In the rat trachea, the NK1 receptor antagonist CP 96345 reduced leukocyte adhesion and plasma leakage induced by substance P, capsaicin, and hypertonic saline challenge (31). Treatment of C57BL/6 mice with CP 96345 attenuated inflammatory cell recruitment to the lungs induced by intratracheal challenge with sheep erythrocytes in antigen-primed animals (32). NK1 receptor knockout mice were protected from lung immune complex injury (33) and demonstrated reduced pancreatitis-associated lung injury as characterized by neutrophil influx and microvascular permeability (34). These data imply that the activation of the NK1 receptor intensifies the cellular immune response in the lungs, suggesting a physiologically significant role for substance P in the regulation of inflammatory responses.

One potential mechanism by which neurokinin receptor signaling may modulate an inflammatory response is through control of cytokine gene expression. NK1 receptor activation has been shown to upregulate in vivo the expression of proinflammatory cytokines such as IL-1, IL-6, and TNF- (35). We examined the expression of KC, TNF-, and IL-6, three proinflammatory cytokines whose expression is upregulated after ozone exposure (4, 38, 39). Of these three molecules, KC expression was most influenced by altered airway innervation. The concentration of KC was significantly higher in the lavage fluid of CCSP-NGF mice than in wild-type or NGFR knockout mice 4 h after ozone exposure (Figure 8A). KC is a chemokine that attracts neutrophils to sites of inflammation and so would be expected to be upregulated before the peak neutrophil influx, as has been found previously (40). KC concentrations were lower in NGFR knockout mice than in wild-type mice, but this difference was significant only at 18 h after ozone exposure (Figure 8B). For TNF- and IL-6, differences in cytokine concentrations among the three types of mice did not correlate well with differences in neutrophilic inflammation (Figures 8C to 8F). Therefore the modulation of neutrophilic inflammation by tachykinins may be mediated by other cytokines that we did not examine or may occur by alternative mechanisms independent of cytokine production. Our studies document a role for sensory nerves in promoting ozone-induced neutrophilic inflammation in mice and provide a model system for dissecting molecular mechanisms by which the nervous system modulates inflammatory processes after exposure to oxidant pollutants.