INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Nitric oxide (NO) is a molecular gas that can be detected in the exhaled gas of animals and humans (1, 2). The expired gas of individuals with asthma contains higher levels of NO (fraction of expired NO, FE_NO) than that of normal individuals; these levels rise after bronchoprovocation and fall after treatment with glucocorticoids (3). Additionally, eosinophil numbers have been reported to positively correlate with FE_NO in a group of subjects with stable mild-to-moderate asthma (11). In this regard, several investigative groups have hypothesized that FE_NO correlates with airway inflammation (3, 6, 12). However, the precise mechanisms responsible for the increased FE_NO documented in asthma are unknown.

Leukotrienes are proinflammatory molecules that are detected during asthma exacerbations and are released during induced asthmatic bronchospasm in models associated with increased NO including allergen inhalation and cold air hyperventilation (13). Inhalation of cysteinyl leukotrienes has been associated with increased airway eosinophils (16, 17). Recently, a relationship between leukotrienes and FE_NO in naturally occurring asthma has been suggested by reports demonstrating that administration of a cysteinyl leukotriene antagonist reduces FE_NO and prevents increases in FE_NO during reduction of inhaled corticosteroid dose (18, 19). One of these studies further noted that leukotriene antagonism also blocks increases in serum eosinophilic cationic protein associated with tapering of inhaled steroids (19).

In view of the data reviewed above, we hypothesized that leukotrienes may promote the increased FE_NO encountered in asthma and that if so, this effect could be mediated through leukotriene-induced airway eosinophils. To test these hypotheses, we examined FE_NO in seven subjects with mild, atopic asthma who developed increased sputum eosinophils following leukotriene E₄ (LTE₄) inhalation and compared these values with those after methacholine inhalation.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Adult nonsmokers with asthma were enrolled. All subjects had a history of asthma (20), positive skin tests to common aeroallergens, and a baseline FEV₁  65% predicted with either a 12% improvement in FEV₁ after inhalation of a -agonist or a methacholine provocative concentration causing a 20% fall in FEV₁ (PC₂₀) of  8 mg/ml. Subjects had been free of upper respiratory infection for at least 6 wk and were using no asthma medications except for short acting -agonists, which were withheld for at least 8 h before all testing. No systemic or inhaled corticosteroids had been used within 8 wk of enrollment. The study protocol was approved by the Human Research Committee at the Brigham and Women's Hospital. All subjects gave their written informed consent.

Protocol

The study protocol, as outlined in Figure 1, entailed four sessions, all of which began before 9:00 A.M. At the first visit a screening examination, baseline spirometry, and baseline sputum induction were performed. Female subjects provided a urine specimen for pregnancy testing. Subjects returned to the laboratory within 48 h for the second visit and provided exhaled gas for measurement of FE_NO. Following bronchoprovocation with either LTE₄ or methacholine, exhaled gas for FE_NO was collected every 30 min for 4 h. At the end of this 4-h observation period, sputum was induced. Following a washout period of at least 7 d, subjects returned to the laboratory for visits 3 and 4, which were identical to visits 1 and 2, respectively, except that bronchoprovocation with the alternate agonist (methacholine or LTE₄) was performed at visit 4. Study subjects were identified as those who demonstrated an increase in sputum eosinophils following LTE₄ inhalation.

View larger version (12K): [in this window] [in a new window]   Figure 1.   Schematic diagram of the study protocol.

Spirometry

Spirometry was performed according to published guidelines (21) with a Collins Survey III spirometer (Warren E. Collins, Braintree, MA). Baseline forced expiratory maneuvers were repeated until three consecutive maneuvers with  5% difference between the highest and lowest FEV₁ were achieved, and the highest FEV₁ from a technically adequate maneuver was recorded.

Sputum Induction

Sputum induction and processing were performed according to established methodology (22). In brief, subjects inhaled 360 µg of albuterol and then performed three peak-flow maneuvers prior to sputum induction, using a Mini-Wright peak-flow meter (Armstrong Medical, Lincolnshire, IL). The best peak expiratory flow rate (PEFR) was recorded as the subjects' baseline. Sputum was induced by inhalation of hypertonic saline (NaCl, 3%). Hypertonic aerosols were generated with a DeVilbiss Ultraneb 2000 ultrasonic nebulizer (Divilbiss Health Care, Somerset, PA) at maximal output (measured to be 4 ml/min). Subjects wearing noseclips inhaled the aerosol through a 90-cm segment of ventilator circuit tubing. At 2-min intervals, subjects were instructed to clear their mouths of saliva, take two large inhalations from the aerosol tubing, and expectorate sputum into a cup. After sputum expectoration, PEFR was recorded. Subjects inhaled hypertonic saline for 12 min or until PEFR fell by 20%, whichever came first. No subjects terminated the saline inhalation prior to 12 min due to a reduction in PEFR.

Sputum Processing

The expectorated sputum was weighed and diluted 1:1 wt/vol with 0.1% dithiothreitol (Sputolysin; Behring Diagnostics, Somerville, NJ). The sample was gently vortexed and placed in a shaking water bath at 37° C for 15 min. A 1-ml volume of the suspension was removed for cell count and differential count and the remainder was centrifuged at 1,000 × g for 10 min at room temperature. The resulting supernatant was aliquoted and stored at 70° C.

Total cell counts were determined with a hemocytometer (Hausser Scientific, Horsham, PA), and the sample was diluted with normal saline to a final concentration of 1.6 × 10⁵ cells/ml for use in the preparation of cytospin slides. Cytocentrifuge slides containing 250 µl/slide were spun for 6 min at 800 rpm (Shandon 3, Pittsburgh, PA).

Differential counts of eosinophils, lymphocytes, neutrophils, macrophages, epithelial cells, and squamous cells were determined on LeukoStat (Fisher Scientific) stained by a single investigator (O.B.) who had no knowledge of the source of the specimen at the time of counting. For each sample, 300 nonsquamous cells were counted, and the percentage of each type of cell was expressed as the percentage of the total nonsquamous cells. In a multicenter asthma therapeutics trial using this technique, the median intrasubject standard deviation of eosinophil counts on repeated measures is 1% (25).

Plethysmography

On the day of each bronchial provocation procedure, subjects performed plethysmography (Warren E. Collins, Braintree, MA) to determine baseline specific airway conductance (sGaw) using established techniques (17, 26). The mean value from a set of three maneuvers was recorded as the value for that set. Subjects were required to perform plethysmography until there was  10% difference between the mean sGaw of two consecutive sets. Once a stable baseline had been established, the mean value of a single set of three maneuvers during the bronchial provocation procedures was recorded as the value for that set.

Methacholine Challenge

On the day of methacholine inhalation the subjects established a stable baseline sGaw as described above. Methacholine challenge was performed according to published techniques with minimal modifications as described below (17, 26). While wearing noseclips, subjects inhaled five breaths of saline from a DeVilbiss nebulizer (model 646) attached to a 20 psi compressed air source and a calibrated dosimeter (S+M Instruments, Doylestown, PA). The dosimeter provided an output of 6 µl/breath. Subjects held their breath for 10 s after each inhalation. Airway caliber was assessed by plethysmography 3 min after the last inhalation. If the mean sGaw was > 90% of the presaline value, subjects went on to inhale five breaths of methacholine (0.078 mg/ml, Methapharm, Brantford, ON, Canada). If sGaw was > 65% of the postsaline value 3 min after methacholine inhalation, subjects went on to inhale doubling concentrations of methacholine until the sGaw had fallen to  65% of the postsaline value.

Leukotriene Challenge

Leukotriene challenge was performed according to published techniques with minimal modifications as described below (17, 26). Stock solutions of LTE₄ (Biomol; Plymouth Meeting, PA) containing 333 µg/ml LTE₄ in methanol/ammonium acetate 70:30, pH 5.6, were stored at 80° C under argon. High-performance liquid chromatography (HPLC) was performed every 8 wk to confirm purity of the agonist. On the day of leukotriene inhalation challenge, subjects performed plethysmography to establish a stable baseline sGaw as described above. While wearing noseclips they then inhaled eight breaths of saline from a DeVilbiss as described above. Subjects held their breath for 5 s after each inhalation. Airway caliber was assessed by plethysmography 2 min after the last inhalation. If the mean sGaw was > 90% of the presaline value, subjects went on to inhale eight breaths of LTE₄ (0.1 µg/ml). sGaw was assessed at 5, 10, and 15 min after inhalation, with the mean sGaw at 15 min recorded as the value for that concentration. If sGaw was > 65% of the postsaline value, subjects went on to inhale 1/3-log increasing concentrations of LTE₄ until the sGaw had fallen to  65% of the postsaline value.

NO Collection and Analysis

Exhaled gas for NO determination was collected before plethysmography (at baseline), at threshold, and every 30 min thereafter. The technical aspects of the offline collection were those specified by the ATS (27). In brief, subjects provided exhaled gas under pressure-regulated flow-restricted conditions, using a T-piece with one-way valves on the inspiratory and expiratory limbs. The inspiratory limb was attached to a source of NO-free air; the expiratory limb included adapters for a pressure manometer (Mercury Medical, Clearwater, FL) and the Mylar bag connector. After three to five tidal breaths through the apparatus, the subject inhaled to total lung capacity. During this inspiration a Mylar balloon was attached to the expiratory limb of the valve apparatus. Without a breath hold, the subject exhaled to residual volume (RV) maintaining a mouth pressure of 10 mm Hg. At this pressure, the flow through the valve apparatus and Mylar bag was 377 ml/s. As the subject approached RV and was no longer able to maintain mouth pressure at 10 mm Hg, he or she was coached to terminate the exhalation and inhale. During this second inhalation the Mylar bag was sealed and removed from the expiratory limb of the apparatus. The NO concentration in the Mylar bag was measured using a chemiluminescence analyzer (model 280; Sievers, Boulder, CO) by an individual without knowledge of the identity of the specimen.

Statistics

Data for exhaled NO during the challenge protocols are presented as percentage of baseline values unless otherwise indicated and were analyzed by two-way analysis of variance (SigmaStat; Jandel Scientific, San Rafael, CA). Data for sputum eosinophils pre- and postchallenge and airway responses were analyzed with Student's paired t test. Relationships between continuous variables were examined by linear regression.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Sixteen subjects completed screening for the study. Two additional subjects enrolled but did not complete both challenges, and three other subjects were not able to produce any sputum; these five individuals were not considered further. The group of screened subjects included nine females and seven males whose ages ranged from 21 to 55 yr. Baseline FEV₁ ranged from 64 to 109% pred, and baseline FE_NO ranged from 4.3 to 29.5 ppb.

Seven study subjects were identified by sputum eosinophil deferential counts, which increased by one or more standard deviation for repeated measures (1% of the total count [25]) after LTE₄ inhalation. This group included four males and three females whose ages ranged from 26 to 51 yr. Baseline FEV₁ ranged from 75 to 93% pred, and baseline FE_NO ranged from 9.1 to 29.5 ppb. Baseline demographic and lung function data of these subjects are presented in Table 1.

Airway Responses in the Study Subjects

The airway responses to bronchoprovocation are detailed in Table 2. Bronchoprovocation with LTE₄ produced a decrease of approximately 62% in sGaw, which was higher than that induced by methacholine (61.7 ± 5.2% following LTE₄ versus 41.7 ± 2.1% following methacholine, p < 0.01)

Sputum Differential Counts following Bronchoprovocation

In the seven study subjects who were identified by a total increase in eosinophils of  1% following LTE₄ inhalation, eosinophils rose from 4.01 ± 0.89% pre-LTE₄ to 8.33 ± 1.52% post-LTE₄. However, this rise in eosinophils was specific to LTE₄ because in these subjects, eosinophils fell following methacholine challenge (Figure 2). The mean change in sputum eosinophils from baseline after LTE₄ inhalation was larger than that after methacholine inhalation (+4.31 ± 1.25% versus 1.14 ± 0.93%). As compared to prechallenge values, there were no significant differences in the differential counts of other cell types after LTE₄ inhalation (Table 3). Methacholine inhalation produced small, but significant increases in the percentage of lymphocytes and neutrophils in the induced sputum (0.2 + 0.1 prechallenge versus 0.5 ± 0.2 postchallenge and 49.5 ± 6.8 prechallenge versus 62.0 ± 7.1 postchallenge, respectively, p < 0.05). In these seven subjects, the individual change in sputum eosinophils was not correlated with baseline NO, LTE₄ PC₃₅, or the degree of bronchoconstriction induced by LTE₄ (data not shown).

View larger version (14K): [in this window] [in a new window]   Figure 2.   Effect of methacholine and LTE4 challenge on induced sputum eosinophils for the study subjects (n = 7). In these subjects, eosinophils increased from 4.01 ± 0.89% before LTE4 challenge to 8.33 ± 1.52% afterward and declined from 6.21 ± 1.85% before methacholine inhalation to 5.07 ± 1.80% afterward.

NO Responses following Bronchoprovocation

LTE₄ inhalation in these seven subjects produced no significant change in FE_NO (one-way ANOVA, p > 0.05, Figure 3). FE_NO ranged from a maximum of 117.7 ± 11.0% of baseline at time 0 (challenge threshold) to a minimum of 74.7 ± 9.0% of baseline at 210 min postchallenge. Similarly, methacholine inhalation produced no significant change in FE_NO (one-way ANOVA, p > 0.05); FE_NO ranged from a maximum of 133.1 ± 9.4% of baseline at challenge threshold to a minimum of 87.1 ± 3.3% of baseline at 30 min after challenge. FE_NO following LTE₄ inhalation was not greater than FE_NO following methacholine inhalation (two-way ANOVA, p = 0.38). In these subjects, there was no significant relationship between percent fall in sGaw induced by LTE₄ inhalation and the change in FE_NO from baseline at 4 h after challenge.

View larger version (16K): [in this window] [in a new window]   Figure 3.   Effect of methacholine and LTE4 challenge on exhaled NO for the study subjects (n = 7). Baseline nitric oxide determinations were made prior to inhalation of agonist whereas determinations made at time = 0 occurred at the nadir response to the inhaled agonists. LTE4 inhalation produced no significant change in FENO (one-way ANOVA, p > 0.05). Similarly, methacholine inhalation produced no significant change in FENO (one-way ANOVA, p > 0.05). FENO following LTE4 inhalation was similar to FENO.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Although a leukotriene receptor antagonist has been shown to block spontaneous increases in FE_NO when inhaled corticosteroids are withdrawn (19), it is unclear whether leukotrienes directly mediate an increase in exhaled NO. In addition, the relationship between acute airway eosinophilia and FE_NO is not known. We measured FE_NO in subjects with increased eosinophils in induced sputum following LTE₄ inhalation, and compared these values with measurements made after methacholine inhalation. We found that in subjects with asthma, bronchial provocation with LTE₄ to a degree that produces increases in sputum eosinophils does not induce elevations in FE_NO.

Our data clearly demonstrate that inhalation of the cysteinyl leukotriene LTE₄ at doses that are adequate to produce airway narrowing and sputum eosinophilia does not acutely increase FE_NO in subjects with asthma. Although it is possible that the doses of LTE₄ administered in this study were insufficient to stimulate NO production and that an increase in FE_NO would have been seen with higher doses of LTE₄ or a greater degree of LTE₄-induced airway narrowing, several lines of evidence make this possibility unlikely. First, it has been demonstrated that bronchoprovocation with allergen resulting in moderate degrees of airway narrowing (similar to that induced in the present study) causes increases in FE_NO (10). Second, we know that in our subjects the doses of LTE₄ administered were sufficient to provoke sputum eosinophilia, yet we found no increase in FE_NO after LTE₄ inhalation. Lastly, a subanalysis demonstrated no relationship between LTE₄ PC₂₀ or the percentage decrease in sGaw and change in FE_NO.

It is unlikely that LTE₄-induced NO production conceivably occurs over a longer time period than we measured since FE_NO rises quickly after other methods of bronchoprovocation. Previously reported data in animals and humans have demonstrated that other inflammatory stimuliallergen exposure and cold air hyperventilationinduce increases in FE_NO within 15 min (10, 28).

As these stimuli are associated with leukotriene release, these observations taken together with our findings, suggest that although the cysteinyl leukotrienes may be a component of the inflammatory environment of the asthmatic airway, LTE₄ alone is not a direct mediator of NO formation. Perhaps other effects of leukotrienes that were not measured in this study, such as changes in vascular permeability, are responsible for modulating FE_NO.

We initially hypothesized that because airway eosinophil numbers correlate with FE_NO, eosinophils could directly promote the formation of NO as human eosinophils are known to express both NOS1 and NOS2 (29, 30). Despite this pathophysiologically plausible mechanism, we found that the induction of sputum eosinophils by LTE₄ was not associated with increased FE_NO; our seven subjects with increased sputum eosinophils 4 h after LTE₄ inhalation had no increase in FE_NO as compared with the prechallenge baseline. These data may appear to conflict with the conclusions of Jatakanon and colleagues, who demonstrated a positive correlation between exhaled NO and sputum eosinophils in subjects with stable asthma and who hypothesized a causal relationship between FE_NO and airway eosinophils mediated by NO-induced enhancement of Th2-like lymphocyte proliferation (11). However, these authors studied subjects with stable asthma and naturally occurring sputum eosinophils, both of which likely result from activation of various inflammatory pathways involving multiple mediators, whereas we studied subjects with asthma following exogenous bronchoprovocation by a single mediator. Our findings suggest that eosinophils acutely recruited to the airway by LTE₄ may be insufficiently activated to participate in the pathobiology of NO production. Alternatively, it is possible that eosinophils do not play a direct role in increased exhaled NO. Our results, in concert with those reported by Obata and colleagues, who found no increases in exhaled NO in subjects with asthma who developed increased sputum eosinophils following plicatic acid inhalation, support this conclusion (31).

Recent reports indicate that administration of a cysteinyl leukotriene antagonist reduces FE_NO within 48 h in children with asthma and attenuates the increase in FE_NO and serum eosinophil cationic protein seen that otherwise follows a 50% reduction of inhaled corticosteroid doses in subjects with moderate asthma (18, 19). In view of these data, our finding that LTE₄ inhalation does not acutely increase FE_NO suggests that leukotrienes may be a necessary, but not sufficient, mediator of increased NO production in asthma.

In summary, neither LTE₄ inhalation nor the acute increase in sputum eosinophils induced by LTE₄ inhalation results in increased FE_NO in asthma. Leukotrienes alone do not appear sufficient to induce NO production. Further, if eosinophils are involved in enhancing NO formation at all, their mobilization by leukotrienes is an insufficient stimulus to induce NO production.