INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The eicosanoids are a family of lipid mediators derived from arachadonic acid and consist of the cysteinyl leukotrienes (cys LTs) and prostaglandins (PG). They are formed by the actions of 5-lipoxygenase and cyclooxygenase respectively. While the cys LTs, LTC₄, LTD₄, and LTE₄ cause potent bronchoconstriction, mucosal edema, and increased mucus secretion within the airways, the biological functions of the prostanoids are more diverse. PGD₂ and PGF₂ provoke potent bronchoconstriction whereas PGE₂ and prostacyclin protect against bronchoconstriction. PGE₂ has also been shown to have anti-inflammatory effects in vitro and in vivo in human subjects.

To date, study of the function of these mediators in asthma has involved work using in vitro models (1, 2), studies on the effects of agents that antagonize eicosanoid synthesis or receptor interaction (3), and studies involving models of provoked asthma (6). In vivo cys LT generation has been monitored indirectly by measuring LTE₄ concentrations in the urine (7, 9, 10) or directly by sampling airway secretions at baseline values (11) and after allergen challenge (8, 12, 13).

In more recent years, sputum induction has been validated as a safe, well tolerated, reproducible, and relatively noninvasive technique for investigating airway inflammation in asthma (14, 15). Sputum is dispersed using the reducing agent dithiothretiol (DTT), which disrupts disulfide bonds within the mucus phase of the sputum allowing dispersion of the cells and measurement of mediators in the supernatants (15). In keeping with the findings in bronchoalveolar lavage fluid (BALF) (11), the concentrations of cys LTs in sputum have been shown to be significantly higher in asthmatics than normal control subjects and to be elevated after acute exacerbations of asthma (16). The concentrations of prostanoids including PGD₂, PGE₂, and PGF₂ were not significantly different in the induced sputum of asthmatics compared with normal control subjects (16).

We wished to further examine whether the noninvasive technique of sputum induction could detect changes in these inflammatory mediators after allergen inhalation in sensitized atopic individuals. Although the potent antagonists at the cys LT receptor can be shown to attenuate both the early and late asthmatic reaction (EAR and LAR) to allergen challenge (3, 4), suggesting that these mediators play a role in both phases of allergen-induced asthmatic reactions, the results of studies monitoring cys LT generation during the LAR have been inconclusive. Release of cys LTs and prostaglandins during the EAR has been confirmed both in urine and in bronchoalveolar lavage (BAL) immediately after allergen challenge (5, 6, 17). However, increased urinary LTE₄ during the LAR has been detected by some workers (5, 6) but not by others (10, 18, 19). Similarly, although certain groups (12, 20) have shown a significant increase in cys LTs in BALF associated with the LAR, this has not been confirmed by all workers (13). We were therefore keen to establish whether increased cys LT generation could be detected at 24 h after allergen challenge in atopic asthmatic volunteers with a confirmed LAR. We chose to sample airway secretions directly using sputum induction to provide information on concentrations of the eicosanoid mediators and inflammatory cell numbers simultaneously.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Fourteen atopic asthmatics (9 female, 5 male) were identified during the screening stage of a larger interventional bronchoscopy study. They were selected on the basis of a demonstrated decrease in forced expiratory volume in one second (FEV₁) of  15% during the LAR to inhalational allergen challenge. The inclusion criteria included a history of mild asthma precipitated by an aeroallergen, a positive skin prick test to at least one allergen extract, an FEV₁ greater than 70% predicted, the ability to produce a sputum sample, and an LAR in response to allergen inhalation. All patients were nonsmokers. No patient had had an upper respiratory tract infection in the 6 wk preceding the trial or had used inhaled or oral corticosteroids or other treatment, apart from inhaled short-acting ₂-agonists for at least 12 wk before entering the trial. All patients were studied out of their allergen season. The study was approved by the East London and City Health Authority Ethics Committee and each patient gave written informed consent.

Subjects attended the laboratory on four separate days between 8:00 and 10:00 A.M. At the first screening visit a full history, physical examination, skin prick tests, and lung function tests were performed. Skin prick tests were to extracts of house dust mite, grass, cat, dog, and diluent control (Soluprick; ALK, Hørsholm, Denmark). A weal of at least 3 mm greater than the negative control was considered positive. One week later the patient attended for baseline sputum induction. The following day the patient attended for an inhalational allergen challenge and 24 h later sputum induction was repeated.

Inhalational Allergen Challenge

Patients were challenged with extracts of the allergen that they were most sensitive to (Aquagen; ALK). The inhalational allergen challenge was performed using a standardized protocol (21) using a breath-activated dosimeter (Mefar, Brescia, Italy). Increasing concentrations of allergen were inhaled under identical conditions (delivering 50, 125, 250, 500, 1,250, 2,500, 5,000 then 12,500 breath units as required) until a decrease in FEV₁ greater than or equal to 20% was recorded. FEV₁ was then monitored at 5-min intervals for 45 min, 15-minute intervals for 2 h, then at 30-min intervals for 12 h. Changes observed between 0-2 h postchallenge were recorded as the EAR, the LAR was defined as changes recorded between 2 to 12 h.

Sputum Induction and Processing

Sputum induction and processing were performed using an accepted protocol (14, 15). Volunteers inhaled 200 µg salbutamol 15 min before sputum induction with a mist of 3%, 4%, then 5% hypertonic saline generated via a low-output ultrasonic nebulizer (Sonix 2000; Medix Ltd, Harlow, UK) for 7 min each. After each inhalation patients were asked to blow their nose, rinse their mouth out with water, then attempt to expectorate sputum. All expectorated sputum samples were collected on ice while further inhalations took place. After each inhalation the FEV₁ was recorded before proceeding. Induction would have been abandoned should any patient have developed either a decrease in FEV₁ > 20%, a troublesome cough, or shortness of breath.

Plugs of sputum were selected and dispersed using 4 volumes of cooled 0.1% DTT (Sigma Pharmaceuticals, Dorset, UK) freshly diluted in Dulbecco's phosphate-buffered saline (D-PBS) (Sigma). After 15 min an equal volume of cooled D-PBS was added and the sample filtered through a 48-µm nylon gauze and centrifuged at 2,000 revolutions per minute for 10 min. The supernatants were stored immediately at 80° C while the cell pellet was resuspended in D-PBS for the cell viability and total cell counts to be estimated by trypan blue staining (Sigma) and cytospins were prepared. Differential cell counts were performed on cytocentrifuge preparations stained with the Romanowsky stain (22) by two observers blind to clinical details. In all cases 400 nucleated nonsquamous cells were counted and results are expressed as percentage of total nonsquamous cells.

Eicosanoid Assay

Sputum supernatants were frozen immediately and stored at 80° C until analysis. Eicosanoid concentrations were measured in the supernatants, corrected for sputum dilution during processing, and expressed as ng/ml of original sputum sample.

Cysteinyl-leukotrienes. Total LTC₄, LTD₄, and LTE₄ were measured in the sputum supernatants, by enzyme immunoassay, employing a cysteinyl-leukotriene polyclonal serum (Cayman Chemical, Ann Abor, 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 rate was 80 to 85%. Samples were corrected for recovery. The sensitivity of the ELISA was 7 pg/ml (the lowest point on the standard curve). The intra- and interassay coefficient of variability of the cysteinyl leukotrienes was 5 to 10% and 10 to 15% respectively across the range of concentrations measured.

Prostanoids. PGD₂ and PGE₂ were measured by modified stable isotope dilution assays that used gas chromatography-negative ion chemical ionization-mass spectroscopy (GC-NICI-MS) as previously described (23). Samples were spiked with 100 to 150 pg of deuterium-labeled internal standards of PGD₂ and PGE₂ before purification, derivation, and analysis. The samples were extracted twice with ethyl acetate after acidification to pH 3. The extract was converted to a pentafluorobenzyl ester by treatment with a mixture of 12.5% pentaluorobenzyl bromide in acetonitrile and diisopropylethylamine and subjected to thin layer chromatography (TLC) plates using the solvent system ethyl acetate/heptane (80:20, vol/vol). The methoxime derivatives of PGD₂ and PGE₂ 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. PGD₂ and PGE₂ were quantified by measuring the ratio of the intensity of mass-to-charge (m/z) 524/530/ 531/532 and m/z 524/528 respectively. Analysis was performed using a mass spectrometer (Nermag R10-10C; Fairfield, NJ) operating in the negative ion mode coupled to a gas chromatograph (Varian Vista 6000; Sunnyvale, CA) using a 15-m DB1701 fused silica capillary column (J and W Scientific, Folsom, CA) or a 5-m SPB-1 (Supelco Inc., Bellefonte, PA) nonpolar methylsilicone column.

Statistical Analysis

Clinical characteristics are expressed as mean ± SEM; all sputum data are expressed as the median with the minimum to maximum range shown. Statistical analyses were performed with the aid of a commercially available statistical package (Minitab for Windows, Minitab Release 9.2; Minitab Inc., State College, PA). Statistical comparisons of before and after the allergen inhalation were performed by using two-tailed Wilcoxon's signed-rank tests. Eicosaniod concentrations, eosinophil percentages, and absolute numbers of eosinophils × 10⁶/ml were normally distributed after log transformation, so relationships between log sputum eosinophil counts and log eicosaniod concentrations have been examined using Pearson's correlation coefficient. Values of p less than 0.05 were considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The 14 patients studied had a mean decrease in FEV₁ of 26.1% (± 1.5) during the EAR and a mean decrease in FEV₁ of 33.0% (± 3.5) during the LAR. They produced adequate sputum for differential cell counts and analysis of cys LTs before and after allergen challenge. Their clinical characteristics are shown in Table 1. There was insufficient sputum supernatant for GC-NICI-MS from Patients 9 and 14. 

The sputum characteristics, differential cell counts, and eicosaniod concentrations before and 24 h after allergen challenge are shown in Table 2. There was a significant correlation between the cys LT concentration at baseline and eosinophil numbers, whether eosinophils are expressed as a percentage of the differential cell count (Figure 1) or as eosinophils × 10⁶ per ml of sputum (r = 0.83, p < 0.001). There were no correlations between eosinophil numbers and PGD₂ or PGE₂ concentrations at baseline values. However, the concentration of PGE₂ correlated to the percentage of macrophages at baseline (Figure 1).

View larger version (13K): [in this window] [in a new window]   Figure 1.   The relationship between the percentages of eosinophils and immunoreactive cys LTs (LTC4/LTD4/LTE4) ng/ml sputum and the percentages of macrophages and PGE2 ng/ml sputum in baseline sputum samples.

Allergen inhalation produced a significant increase in sputum eosinophils from a median of 1.9% at baseline to 16.4% at 24 h postchallenge (p = 0.001, Figure 2). This was associated with a small but significant increase in the percentage of sputum lymphocytes from 0.9% to 1.3% (p = 0.011) and a reciprocal decrease in the median macrophage percentage (p = 0.025). There was also a significant increase in the concentration of cys LTs in sputum 24 h after allergen inhalation (p = 0.002, Figure 2). There was a significant correlation between cys LT concentration in ng/ml sputum and eosinophil numbers in sputum samples collected at 24 h postchallenge, whether eosinophils are expressed as a percentage of the differential cell count (r = 0.58, p < 0.05) or as eosinophils × 10⁶ per ml of sputum (r = 0.61, p < 0.001). The percentage increase in eosinophils correlated with the percentage increase in cys LTs (calculated as the postchallenge/baseline × 100) as shown in Figure 3.

View larger version (21K): [in this window] [in a new window]   Figure 2.   Effect of allergen inhalation on induced sputum eosinophil percentages and immunoreactive cys LTs (LTC4/LTD4/LTE4) ng/ml sputum. The individual and median values are shown.

View larger version (15K): [in this window] [in a new window]   Figure 3.   Relationship between the percentage increase in sputum eosinophil percentage and the percentage increase in immunoreactive cys LTs (LTC4/LTD4/LTE4) per ng/ml sputum at 24 h after allergen challenge.

Allergen inhalation did not produce a significant change in the sputum concentration of PGD₂ or PGE₂ at 24 h postchallenge (Table 2). There was no correlation between the magnitude of either the EAR or LAR and these mediators.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Using the relatively noninvasive technique of sputum induction, we have shown that in mild atopic asthmatics baseline eosinophil numbers correlated closely with the concentration of cys LTs in induced sputum, whereas baseline macrophages were related to baseline PGE₂ concentrations (Figure 1). In these atopic asthmatics who demonstrated a LAR, allergen inhalation was associated with an increase in both induced sputum eosinophil numbers and sputum cys LT concentrations (Figure 2) at 24 h postchallenge. The increase in cys LTs was related to the increase in sputum eosinophils (Figure 3). There were no changes in the concentrations of PGD₂ or PGE₂.

To our knowledge, this it the first time that the noninvasive technique of sputum induction has been used to directly demonstrate evidence of allergen-induced cys LT production within the airways. We believe that this study provides clear evidence for increased generation of cys LTs within the airways after the LAR. Previous workers have exploited the fact that 4 to 6% of all cys LTs synthesized within the body appear within the urine as LTE₄ (24, 25). By careful collection of frequent urine samples some workers have shown biphasic cys LT generation after allergen challenge during the EAR (5, 17) and LAR (5, 6), whereas others have been unable to detect cys LT generation during the LAR (10, 18, 19). Using fiber optic bronchoscopy, direct evidence of airway release of eicosanoids has been gathered immediately after endobronchial allergen challenge (8, 26) and after inhalational allergen challenge (12). Endobronchial allergen challenge was associated with an early phase increase in PGD₂, thromboxane B₂ and LTC₄ (8) detected in BAL collected at 5 min after challenge. Measurements of cys LTs in BAL after inhalational allergen challenge in asthmatics who developed a LAR have been controversial, either showing a small but significant increase in immunoreactive LTC₄ detected in BAL at 6 h postchallenge (12, 20) or being unable to detect increased LTC₄ generation at 12 h postchallenge (13). Using the noninvasive technique of sputum induction, we have demonstrated that cys LT generation occurs within the airways after the LAR and may be detected up to 24 h after allergen inhalation. This is consistent with the finding that potent cys LT receptor antagonists significantly attenuate the LAR (3, 4).

Cys LTs are synthesized by eosinophils, mast cells, and basophils directly and by a variety of other cells including endothelial cells and platelets by transcellular metabolism of the unstable intermediate, LTA₄. The close correlation between baseline cys LTs concentrations and eosinophil numbers suggests that in steady-state mild atopic asthma eosinophils are a principal source of cys LTs within the airways. There was no correlation between eosinophil numbers and the prostanoid mediators PGD₂ or PGE₂. The correlation between baseline macrophage counts and PGE₂ suggests that these cells may be the major source of PGE₂ in the sputum of asthmatics at baseline. This is consistent with the finding that PGE₂ and thromboxane are the major lipid mediators released by macrophages (27).

We believe that sputum induction may offer several advantages over bronchoscopic techniques in studies designed to monitor mediator release. First, the fact that this technique is largely noninvasive would allow repeated examinations to occur over a more detailed time course in the same individual than is possible using bronchoscopy. Second, the fact that the concentration of mediators is significantly higher in sputum than in BAL, with baseline concentrations of cys LTs in asthmatics in sputum at 3.5 ng/ml versus 0.36 pg/ml (11) or 0.32 pg/ml (26) in BAL, suggests that it may be a more sensitive tool to detect subtle changes in mediator expression. This may be because the dilution factor of lower airway secretions is low and fixed between samples in induced sputum, particularly when selected sputum plugs are processed, whereas considerable and variable dilution of lower airway secretions may occur during sampling for BAL. The concentrations of cys LTs and prostanoids have recently been measured in induced sputum supernatants and been shown to be unaffected by treatment with DTT (16). In addition, cys LTs and prostanoid concentrations appear to be stable in sputum supernatants stored at 80° C even in the absence of eicosaniod biosynthesis and breakdown inhibitors (16), suggesting that supernatants collected during routine sputum processing are suitable for examination. The disadvantage of sputum induction is that for some patients production of an adequate volume of sputum is difficult. Although we present results from 14 patients here, a further seven patients with a LAR were identified who were unable to produce enough sputum for analysis of the sputum supernatant either before or after allergen challenge. This represents a success rate of 75% of sputum induction attempts, which is consistent with those found by other groups (14).

We selected the 24-h time point after allergen challenge for our initial study as our primary interest was to demonstrate changes in cys LT expression. Our hypothesis was that eosinophils contribute significantly to cys LT production. Studies of the inflammatory processes occurring during the LAR have shown increased eosinophil numbers in BAL (28) and bronchial biopsies (29) at 24 h, while analysis of induced sputum after allergen challenge has shown an increase in eosinophils and eosinophil granule proteins (14). It therefore seemed likely that increased cys LTs could be detected at 24 h. In studies examining the fate of inhaled radiolabeled LTC₄, it has been shown that cys LTs are cleared rapidly from the airways of asthmatics and that, even after allergen inhalation, up to 60% of an inhaled dose of LTC₄ is cleared from the lungs within 15 min (30). By 6 h the majority of such a dose is cleared from the body in the urine. The fact that such high concentrations of cys LTs were detectable in sputum at 24 h postchallenge suggests that there was ongoing synthesis of cys LTs even at 24 h postchallenge. Although there is a significant relationship between the increase in eosinophil numbers and the increase in cys LTs, the causal relationship between them is more difficult to establish. Whereas an increase in activated eosinophils may generate an increase in cys LTs, inhalation of LTD₄ (31) and LTE₄ (32) has been shown to cause eosinophil recruitment into the airways. Although we have indeed been able to demonstrate an increase in cys LT production within the airways at 24 h postchallenge, this time point examined changes that occurred after rather than during the LAR. We may well have been too late to detect either peak cys LT generation during the LAR or to detect cys LT synthesis by other cell types such as mast cells.

The lack of increase in concentrations of the prostaglandins PGD₂, and PGE₂ at 24 h after allergen challenge is striking and may well explain the relative lack of effect of nonsteroidal treatments on allergen-induced bronchoconstriction. However, it is likely that 24 h postchallenge is too late to detect the PGD₂ release from mast cells, as increased PGD₂ has been demonstrated in BAL during the EAR (8, 26) but not in the LAR at 12 h postchallenge (13). Further studies examining sputum induced at various time points over a tighter time course would be feasible given the noninvasive nature of sputum induction and would allow the changing profile of eicosaniod production after allergen challenge to be studied.

Several lines of work have suggested that PGE₂ is protective against the development of airway inflammation (33) including the recent demonstration that there was a negative correlation between PGE₂ concentrations in sputum and sputum eosinophil counts in asthmatics over a wide range of asthma severity (16). Inhalation of PGE₂ before allergen challenge has been shown to attenuate the bronchoconstriction (34) and airway inflammation associated with the LAR (35) in human asthmatics. We have not reproduced a significant negative correlation between PGE₂ concentrations and eosinophil counts in this study, although we have shown a similar negative trend in our study patients (r = 0.3, not significant [ns]). The explanation for this is that only very mild atopic asthmatics were recruited to this study, who represent a relatively homogenous group. Similarly, we have shown no protective effect of high concentrations of PGE₂ in baseline induced sputum in protecting against development of the LAR. Again, this is probably due to the selection bias of this study, in that only patients with a demonstrated late-phase reaction with a decrease in FEV₁ > 15% were examined in this study. It would be interesting to measure PGE₂ concentrations in a wider range of patients including those 25 to 30% of atopic asthmatics who demonstrate an isolated EAR after allergen inhalation.

In summary, we have shown that in mild atopic asthmatics baseline sputum cys LT concentrations correlated with baseline eosinophil numbers in sputum, that the LAR was associated with an increase in both eosinophil and cys LT concentrations, and that this increase in eosinophils correlated with the increase in cys LT levels. We suggest that the relatively noninvasive nature of sputum induction and the fact that it provides a sensitive mechanism to detect changes in airway eicosanoid concentrations makes it a useful method for examining eicosanoid metabolism within the airways.