INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The cysteinyl-leukotrienes and prostaglandins (collectively known as eicosanoids) are lipid mediators thought to have an important role in the pathophysiology of asthma (1, 2). To date this role has been largely investigated indirectly by blocking the effect of the mediator using receptor antagonists and synthesis inhibitors, although a closer understanding of the role of eicosanoids in asthma would be helped by a noninvasive method for assessing airway production. Eicosanoids have been measured in bronchoalveolar lavage fluid in humans (3, 4), but this is invasive and not suitable for serial assessments. Urinary leukotriene E₄ levels have proved useful to demonstrate cysteinyl-leukotriene release during allergen (5) and aspirin-induced asthma (6), but the technique may not be as sensitive as one that directly samples the airway.

Recent developments in the methodology of sputum induction and processing have resulted in a technique that is suitable for collecting sputum samples from most normal and asthmatic subjects not able to produce sputum spontaneously (7, 8). Concentrations of many molecular markers of eosinophilic inflammation in sputum are considerably greater than in BAL, are repeatable, and successfully discriminate normal from asthmatic subjects (8). Sputum analysis also allows molecular markers of inflammation to be related to cellular markers. The purpose of the current study is to investigate the use of induced sputum to assess airway eicosanoid production.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Healthy and asthmatic nonsmoking adults were recruited from the staff and clinic of Glenfield Hospital. The healthy subjects gave no history of respiratory diseases, had negative allergen skin prick tests, had a FEV₁/FVC > 80% and normal methacholine airway responsiveness (PC₂₀ > 16 mg/ml). Subjects with asthma gave a suggestive history and had objective evidence of variable airflow obstruction as indicated by one or more of the following: (1) methacholine airway hyperresponsiveness (PC₂₀ < 8 mg/ml); (2) a > 15% improvement in FEV₁ 10 min after 200 µg albuterol or; (3) a > 20% maximum morning to evening variability in peak expiratory flow (PEF) measured twice daily over 14 d. Subjects with asthma were subdivided into those with mild episodic asthma requiring treatment with as-required inhaled ₂-agonists only (n = 10) and those with persistent asthma requiring regular treatment with inhaled corticosteroids (n = 10) (Table 1). Six subjects (four receiving inhaled corticosteroids) were studied within 48 h of an exacerbation of asthma during which their PEF was < 50% of predicted or best recorded (whichever was less). The exacerbation was managed as suggested by the British Thoracic Society guidelines (9). All received oral prednisolone 30 mg daily and nebulized albuterol 2.5 mg four hourly and when required. Subjects gave full consent to participate after detailed verbal and written explanation of the study. The protocol was approved by the Glenfield Hospital ethics committee.

Measurements

Subjects attended on two occasions. On the first occasion spirometry and allergen skin prick tests to six common allergens were performed and the subjects were asked to record twice-daily PEF. On the second occasion a methacholine inhalation test was performed followed after recovery by sputum induction. Because of concerns about the safety of sputum induction in acute severe asthma, a spontaneous sample was obtained within 48 h of their admission (mean, 15 h). Within-subject repeatability was assessed in 10 subjects (three normal controls) who had a second sputum induction while clinically stable within 14 d of the first.

FEV₁ was measured using a dry bellows spirometer (Vitalograph, Buckingham, UK) as the best of successive readings within 100 ml, and methacholine challenge was performed using a tidal breathing method as previously described (10). Sputum was induced and processed using a well-established method (7, 8). Sputum was induced using 3, 4, and 5% saline inhaled in sequence for 5 min via an ultrasonic nebulizer (Medix, Basingstoke, UK; output, 0.9 ml/min; mass median diameter, 5.5 µm) 10 min after inhaled albuterol 200 µg. FEV₁ was measured after each inhalation, and subjects were asked to blow their noses and rinse their mouths with water before expectoration to minimize nasal contamination of the sample. Sputum was stored on ice and processed immediately after expectoration. Sputum free from salivary contamination was selected, weighed, and mixed with four times its volume of 0.1% dithiothreitol (DTT) maintained at 4° C. A subset of nine samples was split immediately after expectoration, and in one of the samples attempts were made to minimize ex vivo prostanoid and leukotriene production and breakdown by adding 10⁵ M indomethacin, 5 × 10⁵ M nordihydroguaiaretic acid, 3 × 10⁵ M L-serine-sodium tetraborate, and 2 × 10⁵ M L-cysteine to the 0.1% DTT. A further six samples were split and portions dispersed with and without DTT. After rocking on a bench rocker for 15 min the sample was diluted with four volumes of phosphate-buffered saline and rocked for a further 5 min. The suspension was filtered through a 48-µm nylon gauze and centrifuged at 2,000 rpm (790 × g) for 10 min. The supernatant was stored at 70° C for future analysis. Cytospins were air dried and stained with Romanowski stain (11) before 400 nonsquamous cells were counted by a blinded observer.

Eicosanoid Assay

Two normal subjects had insufficient sputum supernatant for prostanoid assays. PGD₂, PGE₂, PGF₂, and TXB₂ were measured by modified stable isotope dilution assays that used gas chromatography-negative ion chemical ionization-mass spectroscopy (GC-NICI-MS) as previously described (12). Deuterium-labeled internal standards of PGD₂, PGE₂, PGF₂, and TXB₂ were added and the samples were extracted twice with ethyl acetate after acidification to pH 3. The extract was converted to pentafluorobenzyl ester by treatment with a mixture of 12.5% pentafluorobenzyl bromide in acetonitrile and diisopropylethylamine and subjected to TLC plates using the solvent system ethyl acetate/heptane (80:20, vol/vol). The methoxime derivitives of PGD₂, PGE₂, and TXB₂ were made by treatment with 0.5% methoxamine hydrochloride in pyridine followed by formation of the trimethylsilyl derivatives by treatment with N,O-bis (trimethylsilyl) trifluoroacetamine in pyridine. PGF₂ and 9, 11-PGF₂ (not found in samples), PGE₂, TXB₂, and PGD₂ were quantified by measuring the ratio of the intensity of ion m/z 569/573, m/z 524/528, m/z 614/617, and 524/530/ 531/532, respectively. Analysis was performed using a Nermag R10-10C mass spectrometer (Nermag, Fairfield, NJ) operating in the negative ion mode coupled to a Varian-Vista 6000 gas chromatograph (Varian-Vista, Sunnyvale, CA) using a 15 m DB 1701 or 5 m SPB-1 fused silica capillary column (J&W Scientific, Inc., Folsom, CA; Supelco, Inc., Bellefonte, PA). LTC₄/D₄/E₄ were measured by enzyme immunoassay employing a cysteinyl-leukotriene polyclonal antiserum (Cayman Chemical, Ann Arbor, MI) after prior purification on C18 columns (Altech, Los Altos, CA). Samples were spiked with 4,000 cpm of tritiated LTC₄ standard (Dupont NEN Research Products, Boston, MA) before purification and analysis. The recovery was 80 to 85%. The intraassay and interassay coefficient of variability of the cysteinyl-leukotriene assay was 5 to 10% and 10 to 15%, respectively, across the range of concentrations measured.

Analysis

Sputum differential eosinophil counts and eicosanoid concentrations (corrected for the sputum dilution and expressed as ng/ml sputum) were log-transformed, described as geometric mean (log SEM) and compared between normal subjects and subjects with asthma by unpaired t test and between normal subjects and the different categories of asthma by one-way analysis of variance. Log transformation of the LTC₄/D₄/E₄ concentrations did not achieve a normal distribution, and these data were described as median and compared between groups using the Kruskal-Wallis test. Log sputum eosinophil counts and log eicosanoid concentrations were correlated using Pearson's correlation coefficient. Within-subject repeatability of sputum eicosanoid concentration was expressed as a 95% range for repeat measures, as described by Chinn (13).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Cysteinyl-leukotriene and prostaglandin concentrations were not significantly different in sputum treated with biosynthesis and breakdown inhibitors (Table 2). Median sputum LTC₄/ D₄/E₄ concentrations were 4.1 and 2.9 ng/ml in sputum dispersed with and without DTT (p = 0.1). Median sputum LTC₄/D₄/E₄ concentration was significantly greater in subjects with asthma (median, 9.4 ng/ml) than in normal control subjects (6.4 ng/ml; p < 0.02) and were significantly greater in subjects with asthma treated with inhaled corticosteroids (median, 11.4 ng/ml; p < 0.05) or studied within 48 h of an acute severe exacerbation of asthma (13 ng/ml; p < 0.02) (Table 1 and Figure 1). Between-category differences in concentrations of TXB₂, PGD₂, and PGF₂ were similar, although there was more within-category variability and the differences were not significantly different (Table 1 and Figure 2). Concentrations of PGE₂ did not differ significantly between categories (Table 1 and Figure 2).

View larger version (13K): [in this window] [in a new window]   Figure 1.   Sputum cysteinyl-leukotriene concentration in each category.

View larger version (15K): [in this window] [in a new window]   Figure 2.   Sputum prostaglandin concentration (ng/ml) in each category.

There was no correlation between the log sputum concentration of LTC₄/D₄/E₄, TXB₂, and PGD₂ and the log sputum eosinophil count. However, the log sputum eosinophil count negatively correlated (r = 0.48; p < 0.01) (Figure 3) with the log sputum PGE₂ concentration and positively correlated (r = 0.45; p < 0.02) with the log sputum PGF₂ concentration in subjects with asthma. The sputum supernatant PGE₂ concentration did not correlate with the sputum macrophage differential cell count (r = 0.07). The 95% ranges for repeat measures were 1.9-, 3.8-, 3.9-, 4.5-, and 5.9-fold for PGE₂, LTC₄/D₄/ E₄, PGD₂, TXB₂ and PGF₂, respectively.

View larger version (10K): [in this window] [in a new window]   Figure 3.   Relationship between sputum differential eosinophil count and sputum supernatant PGE2 concentration in subjects with asthma.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

We have shown that induced sputum cysteinyl-leukotriene concentration was significantly greater in subjects with asthma than in normal control subjects, but there were no significant differences in concentrations of other eicosanoids. Sputum cysteinyl-leukotriene concentrations were greater in subjects with more persistent and severe asthma, suggesting that they might be more functionally important in these groups. Most of these subjects were treated with inhaled corticosteroids, supporting the theory that corticosteroids do not directly reduce cysteinyl-leukotriene production (14). We found considerable within-subject variability of most sputum eicosanoid concentrations, which could reflect variable airway production of these mediators or problems with the methodology. The fact that sputum PGE₂ could be measured repeatably is more in keeping with the former.

We have addressed a number of potential problems with using induced sputum to assess airway eicosanoid production by performing appropriate control experiments. Ex vivo production, or breakdown of, seems unlikely since immediate incubation of sputum with leukotriene and prostaglandin biosynthesis and breakdown inhibitors did not affect sputum eicosanoid concentrations. Similarly, important interference in the cysteinyl-leukotriene assay by DTT is unlikely since concentrations were not significantly different in sputum treated with and that treated without DTT. We cannot exclude the possibility that the cysteinyl-leukotriene concentration in sputum from subjects with asthma might be increased by the effect of hypertonic saline on mast cells and other mediator-producing cells in the airway. However, sputum concentrations of the mast cell product PGD₂ were not elevated and subjects were pretreated with albuterol before sputum induction, which would be expected to reduce mediator release. Furthermore, concentrations of cysteinyl-leukotrienes were greatest in spontaneously produced sputum from subjects with acute severe asthma. It is possible that our inclusion of spontaneous sputum has biased our comparisons and correlations, but we consider this unlikely since inflammatory cell counts and the concentration of most inflammatory mediators have been shown to be similar in induced and spontaneous sputum samples (15). A further concern was variable dilution of sputum eicosanoids by contaminating saliva. It is difficult to design control spiking experiments because of the different physical properties of sputum and contaminating fluid, so we attempted to minimize this by selecting sputum from surrounding fluid. Our median squamous cell contamination was less than 5%, suggesting that any effect of salivary contamination was small.

Earlier investigators measured slow-reacting substance- type activity in sputum from asthmatics using a bioassay or by high pressure liquid chromatography with mixed results (16, 17). More recently cysteinyl-leukotrienes have been shown to be present in increased concentrations in bronchial wash and bronchoalveolar lavage (BAL) samples from subjects with stable asthma (3), after allergen challenge in atopic asthma (14, 18) and after aspirin challenge in aspirin-sensitive asthma (12). Our findings using newer techniques to obtain and process sputum support these earlier findings. The relatively noninvasive nature of sputum induction and the fact that cysteinyl-leukotriene concentrations are present in considerably greater concentration in induced sputum than in BAL suggest that this technique has a number of advantages over bronchoscopic methods.

LTE₄, the end product of LTC₄ and D₄ metabolism, can be measured in the urine, and there is increasing interest in the use of measures of urinary excretion of LTE₄ to assess airway cysteinyl-leukotriene production (5, 6, 19). Intervention studies show that urinary LTE₄ concentrations increase after allergen challenge in atopic asthma (5) and after aspirin challenge in subjects with aspirin-induced asthma (6). Urinary LTE₄ concentrations may also be increased in subjects with acute severe asthma (5, 20). Only 4 to 9% of an intravenous or inhaled dose of LTC₄ appears in the urine (19), suggesting that this technique might be relatively insensitive to minor differences or changes in airway cysteinyl-leukotriene concentrations. In support of this, urinary LTE₄ excretion is similar in subjects with stable asthma and in normal control subjects (5). Furthermore, exercise-induced asthma, which is effectively inhibited by treatment with leukotiene receptor antagonists, suggesting it is at least partly due to airway cysteinyl-leukotriene production (1, 21), is not always associated with significant changes in urinary LTE₄ excretion (21, 22). We have shown that induced sputum cysteinyl-leukotriene concentration are significantly greater in subjects with stable asthma, suggesting that measurement in sputum is more sensitive than measurement in urine.

Although the magnitude of the differences in sputum concentrations of PGD₂, PGF₂, and TXB₂ was similar to the differences in cysteinyl-leukotrienes, there was more within-category variability and the differences were not statistically significant. This is in keeping with the variable effect of nonsteroidal anti-inflammatory drugs on airway function and responsiveness in stable asthma (2). Our findings contrast with those of a number of bronchoscopy studies showing increased concentrations or proportions of broncoconstrictor prostaglandins in bronchial wash and BAL in stable asthma and after allergen challenge (4, 14, 23). Possible explanations for the difference include greater precision of bronchoscopic measurements, difference in subjects studied, or spurious elevation of airway prostaglandin production caused by trauma of the inflamed airway wall in asthma during bronchoscopy. Assessment of airway prostaglandin production using both techniques in the same subjects might help resolve this issue.

Our technique allows us to obtain simultaneous information on cellular markers of airway inflammation. We did not find a positive correlation between the sputum eosinophil count and the sputum cysteinyl-leukotriene or bronchoconstrictor prostaglandin concentration in our heterogeneous population of subjects with asthma, suggesting that additional cell types might be an important source or that the cell producing these eicosanoids is not proportionately represented in sputum. We did find a negative correlation between the sputum eosinophil count and the supernatant PGE₂ concentration. One interpretation of this relationship is that PGE₂ exerts an anti-inflammatory effect in the airway. The fact that PGE₂ and its analogues have a number of bronchoprotective and anti-inflammatory effects in vitro (24), and inhaled PGE₂ inhibits the late bronchoconstrictor (25) and inflammatory (26) response to allergen in atopic asthma, would support such a role. Another possible explanation for the relationship between the sputum eosinophil count and PGE₂ concentration is that active eosinophilic airway inflammation damages cells producing PGE₂ such as airway epithelial cells. However, sputum concentrations of PGF₂ (which is produced by similar cell types to PGE₂) (27) correlated positively with the sputum eosinophil count, arguing against this interpretation. Finally we have considered whether the reduced PGE₂ concentration in the sputum supernatant of samples containing a high proportion of eosinophils may be due to reduced numbers of macrophages since the macrophage is a potential source of PGE₂ (27). The lack of correlation between the sputum differential macrophage count and sputum supernatant PGE₂ concentration suggests that this is not the case.

In conclusion, we have shown that eicosanoids are present in high concentrations in induced sputum and that concentrations of cysteinyl-leukotrienes are present in significantly higher concentrations in subjects with asthma than in normal control subjects. We suggest that this technique is a potentially useful method to assess airway production of eicosanoids and to relate this to the underlying cellular inflammation. The noninvasive nature of sputum induction suggests that this technique might be particularly useful for serial assessment after intervention with drugs and/or bronchial challenge. Our estimates of within-subject repeatability of sputum eicosanoid concentrations will help in the planning of these studies.