Published ahead of print on February 5, 2003, doi:10.1164/rccm.200201-055OC
© 2003 American Thoracic Society
Neurokinins Modulate Hyperventilation-induced Bronchoconstriction in Canine Peripheral AirwaysDepartment of Environmental Health Sciences, School of Public Health, The Johns Hopkins University, Baltimore, Maryland Correspondence and requests for reprints should be addressed to Arthur N. Freed, National Heart, Lung, and Blood Institute, Two Rockledge Center, Suite 7186, 6701 Rockledge Drive, Bethesda, MD 20892–7924. E-Mail: freeda{at}nhlbi.nih.gov
This study was designed to test the hypotheses that (1) neurokinin (NK) receptor activity modulates hyperventilation-induced bronchoconstriction (HIB) in canine peripheral airways and (2) NK receptor activity is stimulated via hyperventilation-induced eicosanoid production and release. A bronchoscope was used in anesthetized dogs to record peripheral airway resistance (Rp); to test airway reactivity to NK A (NKA), substance P, and hypertonic saline; and to examine HIB before and after combined treatment with NK-1 (CP 99,994) and NK-2 (SR 48,968) receptor antagonists. Bronchoalveolar lavage fluid cells, prostaglandin D2, and cysteinyl leukotrienes from hyperventilated airways pretreated with either vehicle or NK antagonists were also measured. Pretreatment with NK-1 and NK-2 antagonists significantly attenuated HIB and the response to substance P, virtually abolished the response to NKA, and had little effect on the response to HS. Blockade of NK-1 and NK-2 receptors did not affect either the cell profiles or the mediator concentrations recovered in bronchoalveolar lavage fluid after hyperventilation. We conclude that NKs modulate the development of HIB and appear to do so via hyperventilation-induced eicosanoid production and release.
Key Words: animal model • exercise-induced asthma • neurokinins Hyperventilation of cool, dry air causes mild airway narrowing in normal subjects and severe airway obstruction in individuals with asthma (1). Although the latter case is commonly referred to as exercise-induced asthma (EIA), in general, these responses might better be lumped under the rubric of hyperventilation-induced bronchoconstriction (HIB). Penetration into the lung periphery of cool dry air is believed to result in airway hyperosmolarity (2), which in turn initiates the release of bronchoactive mediators from osmosensitive cells (3, 4) and ultimately results in transient airway obstruction. This scenario is supported by numerous studies using pharmacologic antagonists of either mediator production or receptor activity (5–7), and one study documenting increased levels of leukotrienes and prostaglandins in bronchoalveolar lavage fluid (BALF) recovered from subjects with asthma after isocapnic hyperpnea (8). The hyperventilation-induced production and release of leukotrienes and prostanoids have been carefully documented in rat, guinea pig, and canine models of HIB (reviewed in 9). In addition to eicosanoid mediators, the local release of neurokinins (NKs) from afferent C fibers (10) has been reported to initiate HIB in rodents via NK receptors (11, 12). However, it is unclear whether this phenomenon occurs in mammals other than rodents. Also in question is whether hyperventilation with dry gas stimulates NK release, which in turn initiates airway narrowing via the synthesis of cysteinyl leukotrienes (13), or whether leukotrienes modulate HIB via the secondary release of neuropeptides (14, 15). Recent data in guinea pigs showing that leukotriene antagonists inhibited hyperventilation-induced NK release in BALF and plasma support the latter scenario (15). The fact that NK receptor antagonists do not abolish HIB in guinea pigs (16) is consistent with the notion that other parallel pathways (e.g., eicosanoid metabolism) are likely to be stimulated while hyperventilating cold, dry air. NKs and their receptors modulate goblet cell secretion, bronchovascular leakage, inflammatory cell infiltration, airways obstruction, and bronchial hyperreactivity in animal and human subjects (17–24). Although NKs have been implicated as potential modulators of asthma, we know of only one study that attempted to examine the effect of NK antagonists on HIB in human subjects. In that study, NK-1 receptor antagonist FK-888 shortened the duration of exercise-induced bronchoconstriction in patients with asthma but failed to inhibit either the development or the magnitude of the airway obstruction (25). This may be due to the fact that the NK-2 receptor is the predominant receptor subtype producing airway constriction in human subjects (26, 27). Thus, in light of the prominent role played by NKs in rodent models of HIB, it is likely that these mediators play an important role in the modulation of HIB in other mammalian models and in human subjects. The purpose of this study was to test the hypotheses that (1) NK receptor activity modulates HIB in canine peripheral airways and (2) NK receptor activity is stimulated via hyperventilation-induced eicosanoid production and release.
Dogs were anesthetized, intubated with a dual-portal endotracheal tube, and mechanically ventilated with room air. End-tidal CO2 was maintained at 4.5%, and heart rate, mean arterial pressure, and body temperature were recorded (28).
A bronchoscope was wedged in two contralateral sublobar segments in each dog (mean ± SE = 23 ± 1.2 kg, n = 6). A map of the branching pattern of the airways from the origin of the specific lobar bronchus to the wedge position was recorded to allow identical bronchoscope placement on subsequent days. Sublobar airways were ventilated (200 ml/min) via the bronchoscope with 5% CO2 in dry air. After recording baseline peripheral resistance (Rp) in airway 1 (28), reactivity ( The effect of NK-1 and NK-2 receptor blockade on the peripheral airway response to an aerosol hypertonic saline (HS) challenge (4,400 mmol/kg H2O, 60 seconds) (29) was examined in six dogs (22 ± 1.9 kg), three of which were used in the previous experiment. The protocol was similar to that described above in which HS challenge was done instead of either NK aerosol challenges or DAC. Recording Rp at only 0.5, 2, and 5 minutes after the second aerosol challenge shortened this protocol. Finally, two contralateral sublobar segments in each dog (22 ± 1.6 kg, n = 7) were wedged with a bronchoscope to obtain cell and mediator data. After recording baseline Rp in both sublobar segments, either NK-1 and NK-2 receptor antagonists or their vehicle was administered intravenously. DAC was done in each segment, and bronchoalveolar lavage was done immediately after DAC (0 minutes) in one segment and 5 minutes after the DAC in the other segment. The same dogs used in the HS challenge study (plus 1) were used for these experiments. BALF differential cell counts, prostaglandin D2, and leukotriene C4-D4-E4 (LTC4–E4) were analyzed as previously described (28). Rp data were analyzed using a Friedman repeated measures analysis of variance on ranks in conjunction with a Student–Newman–Keuls test for the comparison of treatment means. All other data were analyzed using a Wilcoxon signed rank test.
Effects of NK-1 and NK-2 Receptor Inhibition of Bronchoconstriction Induced by DAC, NKA, SP, and HS DAC increased Rp by approximately 110% before and approximately 140% after treatment with the NK-1/NK-2 vehicle (Figures 1A, 1C, and 1D) . In contrast, DAC increased Rp by approximately 140% before and by only approximately 60% after treatment with the NK-1/NK-2 antagonists (Figures 1B–1D). Dry air–induced changes in Rp were significantly attenuated (p < 0.05) after the DAC. This was evident when comparing HIB before and after treatment with NK-1/NK-2 antagonists (Figure 1B) and when comparing HIB after drug or vehicle treatment done 1 week apart (Figure 1D). Dry air–induced changes in Rp were similar over time (Figure 1C). It is important to note that HIB was previously shown to be repeatable over the time course of this study and that the magnitude of HIB normally tends to increase in response to consecutive challenges (30–32).
NKA increased Rp by approximately 130% before and approximately 165% after treatment with the NK-1/NK-2 vehicle (Figures 2A, 2C, and 2D) . In contrast, NKA increased Rp by approximately 100% before and by only approximately 20% after treatment with the NK-1/NK-2 antagonists (Figures 2B–D). NKA-induced changes in Rp were significantly attenuated (p < 0.05) at 0.5 and 2 minutes after the challenge. This was evident when comparing NKA-induced bronchoconstriction before and after treatment with NK-1/NK-2 antagonists (Figure 2B) and when comparing NKA-induced bronchoconstriction after drug or vehicle treatment done 1 week apart (Figure 2D). Except for the 2-minute timepoint, NKA-induced changes in Rp were similar over time (Figure 2C, compare pretreatment values in Figure 2A with those in Figure 2B).
SP increased Rp by approximately 80% before and after treatment with the NK-1/NK-2 vehicle (Figures 3A, 3C, and 3D) . In contrast, SP increased Rp by approximately 60% before and by only approximately 24% after treatment with the NK-1/NK-2 antagonists (Figures 3B–3D). SP-induced changes in Rp were significantly attenuated (p < 0.05) at 2 and 10 minutes after the challenge. This was evident only when comparing SP-induced bronchoconstriction before and after treatment with NK-1/NK-2 antagonists (Figure 3B). Our inability to detect significant differences when comparing SP-induced bronchoconstriction after drug and vehicle treatment done 1 week apart (Figure 3D) probably reflects the differences in baseline Rp recorded at those times (Figure 3D, compare post-treatment values in Figure 3A with those in Figure 3B).
HS increased Rp by approximately 65% before and approximately 59% after treatment with the NK-1/NK-2 vehicle (Figures 4A, 4C, and 4D) . In contrast, HS increased Rp by approximately 80% before and by approximately 45% after treatment with the NK-1/NK-2 antagonists (Figures 4B–4D). However, the tendency for HS-induced bronchoconstriction to decrease was not significant (p > 0.05) (Figures 4B and 4D).
NS increased Rp by approximately 22% before and approximately 16% after treatment with the NK-1/NK-2 vehicle (Figures 5A, 5C, and 5D) . NS increased Rp by approximately 31% before and by approximately 18% after treatment with the NK-1/NK-2 antagonists (Figures 5B–5D). Post-treatment changes in Rp were significantly reduced (p < 0.05) at 0.5 minutes after the challenge in vehicle treated airways only (Figure 5A).
Changes in heart rate and body temperature during vehicle control and NK-1/NK-2 experiments were similar. A significant decrease in mean arterial pressure was observed during NK-1/NK-2 experiments when compared with vehicle control experiments (Table 1) .
Effects of NK-1 and NK-2 Receptor Inhibition on Bronchoalveolar Lavage Cells and Mediators Recovered During HIB Total cells per milliliter of BALF recovered at 0 (p = 0.938; Figure 6A) and 5 minutes (p = 0.578; Figure 6B) after DAC from vehicle-treated and NK-1/NK-2-treated airways were similar. Cell viability was also similar across treatments (99.6–100%). Drug treatment did not significantly affect any cells of interest that were recovered in BALF after DAC (Figure 6).
Concentrations of prostaglandin D2 (p = 0.219) and leukotriene C4–E4 (p = 0.375) in BALF samples recovered at 0 and 5 minutes after DAC, respectively, from vehicle- and NK-1/NK-2–treated airways did not differ significantly (Figure 7) .
This study provides the first pharmacologic evidence supporting a role for NK receptors in the development of HIB in canine peripheral airways. We show that specific simultaneous antagonism of NK-1 and NK-2 receptors reduced HIB by approximately 50%. Similar levels of inhibition were achieved using a 5-lipoxygenase antagonist (32) in dogs and either a 5-lipoxygenase (33, 34) or a leukotriene D4 receptor (35, 36) antagonist in subjects with asthma. These observations are consistent with the hypothesis that NKs and leukotrienes work in concert to produce hyperventilation-induced airway obstruction (11, 12). We used aerosols of NKA and SP to test the efficacy of simultaneous NK-1 and NK-2 receptor antagonism in our canine model of EIA. We found that NK-1 and NK-2 receptor antagonism essentially abolished NKA-induced bronchoconstriction, which was reduced on average by approximately 75%. The residual increase in Rp was not significant when compared with baseline tone (Figures 2B and 2D). Sherwood and colleagues (37) previously reported that aerosolized NKA caused NK-2 receptor-mediated bronchoconstriction in dogs. Related work revealed that the NK-2 receptor is the predominant receptor subtype producing airway constriction in canine airways (38). In vitro experiments with human airway tissue suggest a similar finding (26, 27), although NK1 receptors were recently reported to contract small and medium sized isolated human bronchi. Small airway contraction appears to be mediated via prostanoid release (39), whereas medium airway narrowing appears to result directly via smooth muscle receptor activation and inositol phosphate release (40). To our knowledge, the bronchoactive effects of aerosolized SP in dog have not been described, although intravenous bolus injections of SP causes dose-dependent bronchoconstriction in this species (41). Bronchoconstriction in response to aerosolized SP was smaller and more variable than that elicited by NKA (Figure 3). SP-induced bronchoconstriction was reduced by only approximately 45% and reflects primarily a reduction in the peak response (Figure 3B). Unlike NKA-induced bronchoconstriction, the residual increase in Rp was significant when compared with baseline tone (Figures 3B and 3D). Although this suggests that NK-1 receptors were at best partially blocked, the estimated plasma concentration of CP 99,994 used in this study (approximately 12.5 µg/ml, based on delivered dose and blood volume calculated for a 20-kg dog) greatly exceeded that which blocks NK-1 receptors in dogs (50 ng/ml) (38). Thus, it is possible that SP acts via a novel NK receptor-independent pathway similar to that described by Grant and colleagues (42) or through nonspecific peptide-membrane phospholipid interactions as suggested by Kroegel and colleagues (43). Although these issues preclude an assessment of the specific role of NK-1 receptors in the development of HIB in our canine model, these receptors were reported to play little if any role in canine bronchoconstriction (38). Despite this fact, it is possible that incomplete NK-1 receptor blockade contributed to the trend toward increased leukotrienes seen in Figure 7 and may reflect a complex interaction between NK-1 and NK-2 receptor subtypes. Simultaneous oral administration of CP 99,994 and SR 48,968 was reported to reduce the biologic activity (i.e., plasma level) of SR 48,968 in dogs (38). However, a comparison of the combined NK-1/NK-2 treatment used in this study with some preliminary trials using SR 48,968 alone did not reveal any noticeable differences in the efficacy of the NK-2 antagonist when delivered intravenously. This may be due to differences in dose and in the route of administration of these drugs. The use of NK-1/NK-2 antagonists significantly decreased mean arterial pressure. This decrease in mean arterial pressure could theoretically contribute to the inhibitory effect of NK-1/NK-2 treatment on HIB via the reduction of hyperventilation-induced bronchovascular hyperemia and edema formation. However, previous work showed that HIB in canine peripheral airways was not affected by changes in bronchial arterial blood pressure and that bronchovascular leakage did not contribute to the development of HIB (44). Until now, there were no published data to implicate a role for excitatory neuropeptides in our canine model of EIA. Hyperventilation with dry air causes mucosal injury in canine (45) and human subjects with asthma (8). Thus, bronchoalveolar lavage cell profiles were examined primarily to determine whether NK-1/NK-2 antagonists inhibited epithelial cell shedding (i.e., mucosal injury), which typically occurs during hyperventilation with dry air (28, 45). NK-1 and NK-2 receptor antagonists were reported to reduce the accumulation of ciliated epithelial cells in a guinea pig model of allergic asthma (21). However, NK-1/NK-2 receptor blockade does not attenuate hyperventilation-induced mucosal injury in our canine model (Figure 6). The fact that NK-1/NK-2 antagonists inhibit HIB in dogs (Figure 1) is consistent with the hypothesis that local injury of the bronchial mucosa initiates the release of NKs from afferent C fibers. NK release is in turn believed to stimulate airway narrowing via the release of cysteinyl leukotrienes (13, 14). Although NK-induced eicosanoid release has been reported in lung (46–48), leukotrienes and prostanoids are themselves capable of stimulating neuropeptide release (49, 50), suggesting that both NK- and eicosanoid-induced pathways may modulate HIB. Thus, it is possible that hyperventilation-induced mast cell degranulation (44, 51) initiates the release of neuropeptides from sensory nerve fibers, resulting in bronchoconstriction. This concept fits with the observation that treatment with cylcooxygenase and 5-lipoxygenase inhibitors (52) failed to provide significantly greater protection against HIB than either one alone (32, 53), and suggests that leukotrienes and prostanoids may be equipotent in stimulating neuropeptide release. Lai and Lee (15) using a guinea pig model of HIB provided compelling evidence in support of the hypothesis that leukotrienes modulate HIB via the secondary release of neuropeptides. They reported that leukotriene synthesis and receptor antagonists inhibited hyperventilation-induced SP and leukotriene release in BALF. They also showed that treatment with SR 48,968 did not affect hyperpnea-induced increases of leukotrienes in BALF. Our canine data (Figures 1 and 7) and data from guinea pigs (14, 15) suggest that hyperventilation with dry air initiates a cascade of events that starts with eicosanoid metabolism and release and results in downstream NK activity. In contrast to this potential mechanism, Yang and colleagues (13) reported that NKs mediate hyperpnea-induced bronchoconstriction by triggering cysteinyl leukotriene synthesis in guinea pig airways. However, the use of bile to monitor leukotriene production in their study does not allow us to exclude the synthesis of leukotrienes from organs other than the lung. This problem is illustrated by the fact that leukotriene levels increased in plasma but not BALF samples recovered within minutes after hyperpnea from guinea pigs treated with an NK-2 receptor antagonist (15). Even so, we cannot discount the possibility that both pathways are simultaneously activated and contribute to the development of HIB. The fact that neither NK nor eicosanoid antagonists abolish HIB in either guinea pigs (14, 16) or dogs (Figure 1) (32, 53) is consistent with the notion that other pathways are likely to be simultaneously stimulated while hyperventilating cold dry air. Finally, hyperventilation-induced increases in airway surface fluid osmolality are believed to contribute to the development of HIB (54). If changes in airway surface fluid osmolality per se contribute to HIB and if these osmotic-induced changes are modulated by NK activity, then NKs should modulate HS-induced bronchoconstriction in a similar manner. We previously reported that DAC caused acute airway surface fluid hyperosmolality, acute mucosal injury, mediator release, transient bronchoconstriction, prolonged leukocyte infiltration, late airway obstruction, and delayed bronchial hyperreactivity in canine peripheral airways (45, 52, 54). In contrast to DAC, HS aerosol challenge produced only airway surface fluid hyperosmolality, mediator release, transient bronchoconstriction, and delayed bronchial hyperreactivity (29). In addition, although HS stimulated SP release and glandular mucous exocytosis in human subjects, it failed to elicit any vascular processes such as plasma exudation or mucosal edema (55). Data depicted in Figure 4 reveal that unlike HIB, NK-1/NK-2 receptors play little if any role in the development of HS-induced bronchoconstriction. Similar results were reported for human subjects treated with CP 99,994 (56). Although BALF cell profiles are not acutely affected (Figure 6), the observations noted previously here suggest that hyperventilation-induced mucosal injury initiates NK receptor activity, leukocyte infiltration, and airway obstruction independently of its accompanying hypertonic component. In summary, NK receptor activity modulates HIB in canine peripheral airways and does so in part via an eicosanoid-mediated pathway(s). In addition to significantly inhibiting HIB, NK-1 and NK-2 receptor antagonists partially reduced SP-induced bronchoconstriction, virtually abolished NKA-induced bronchoconstriction, and had little effect on HS-induced bronchoconstriction. The latter phenomenon further delineates the role of airway surface fluid hyperosmolality in hyperventilation-induced changes in peripheral airway function. NK-1 and NK-2 receptor antagonists did not inhibit hyperventilation-induced release of prostaglandin D2 and leukotriene C4-D4-E4 production (Figure 7), which typically occurs in canine peripheral airways (28). This lack of effect on eicosanoid mediator release may reflect inadequate NK1 receptor antagonism, but it is more likely that HIB is modulated via the release of NKs from sensory C fibers. Although presently there exists a dearth of data focusing on the participation of NKs in the development of HIB in human subjects, the prominent role played by NKs in rodent and canine models of EIA suggests that these mediators play an important role in the modulation of HIB in human subjects. Finally, the fact that this work focuses on peripheral airways and that NK2 receptor density tends to be greatest in the central airways of all species examined to date (57) suggest that our data may actually underestimate the modulatory role of sensory neural mechanisms in HIB.
Supported by National Institutes of Health National Heart, Lung, and Blood Institute, grants HL51930 and HL63186. This article was written by Arthur N. Freed in his private capacity. The views expressed in the article do not necessarily represent the views of the National Institutes of Health, Department of Health and Human Services, or the United States. Received in original form January 29, 2002; accepted in final form February 4, 2003
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
METHODS
Rp) to sequentially aerosolized (60 seconds, 200 ml/min, 5% CO2 in dry air) NK A (NKA; 200 µg/ml) and substance P (SP; 200 µg/ml) was recorded. Rp was recorded at 0.5, 2, 5, 10, and 15 minutes after each challenge and was allowed to recover between the SP and NKA challenges. During the time between the SP and NKA challenges, baseline Rp and the response to dry air challenge (DAC) were recorded in the contralateral sublobar segment (airway 2). DAC was done by increasing the flow rate of 5% CO2 in dry air from 200 to 2,000 ml/min for 2 minutes. After Rp stabilized in both sublobar segments (i.e., Rp remained virtually unchanged for three consecutive readings), either NK-1 (CP 99,994) and NK-2 (SR 48,968) receptor antagonists (1 mg/kg each + 1 ml polyethlyene glycol [PEG]200 + 2 ml 0.9% NaCl) or its vehicle was administered intravenously. Thirty minutes later, baseline Rp was recorded, and the SP and NKA challenges were repeated in airway 1. DAC was repeated as soon as Rp stabilized in airway 2. This protocol was repeated 1 week later in the same sublobar segments using NK-1 and NK-2 receptor antagonists if vehicle was used the previous week or vice versa. 
0.05 compared with baseline; 
p 
p 
