Effects of systemic glucocorticosteroids on peripheral neutrophil functions in asthmatic subjects: an ex vivo study

In 21 asthmatic subjects, several functions of isolated peripheral neutrophils (chemokinesis and chemotaxis toward 10% E. coli; superoxide anion generation after PMA; leukotriene B4 (LTB4) release from whole blood and isolated neutrophtls, before and after different stimuli) were evaluated during an acute exacerbation of asthma, and after 14 – 54 days of treatment with systemic glucocorticosteroids (GCS). During acute exacerbation, superoxide anion generation was higher in asthmatics than in eleven normal subjects (39.2 ± 14.1 vs. 25.2 ± 7.3 nmol, p < 0.05); there was a significant correlation between FEV1 (% of predicted) and neutrophil chemotaxis (r = −0.52, p = 0.04). After treatment, there was no significant change in all neutrophil functions, except for a decrease in neutrophil chemotaxis in subjects who showed an FEV1 increase > 20% after GCS treatment (from 131 ± 18 to 117 ± 21 μm, p = 0.005). Chemokinesis sicantly decreased in all subjects, and the changes significantly correlated with an arbitrary score of the total administered dose of GCS (r = 0.57, p < 0.05). These data suggest that neutrophil activation plays a minor role in asthma, and that treatment with GCS is not able to modify most functions of peripheral neutrophils in asthmatic subjects; chemotaxis seems to be related only to the severity of the asthma and it could reflect the improvement of the disease.


Introduction
Asthma has been defined as reversible airway obstruction, associated with nonspecific bronchial hyperresponsiveness, which is sustained by a chronic airway inflammation. Pathology of fatal asthma 2 and direct in vivo measurements of indices of airway inflammation by bronchoalveolar lavage (BAL) and bronchial biopsy 3'4 has confirmed the major role played by the recruitment of inflammatory cells in the airways of subjects with asthma of different severity.
While eosinophils are pivotal cells in asthma, the role of neutrophils in the pathophysiology of asthma is still uncertain. Neutrophil counts are increased in bronchial (BL) and bronchoalveolar lavage (BAL) of animals and humans after acute exposure to allergens, chemicals (like toluene diisocyanate, TDI) or oxidants (like ozone); [5][6][7] but in the stable phase, asthmatics show neutrophil counts in BAL or in bronchial biopsy that are not different from those found in normal 9 subjects.' Markers of activation of neutrophils have been inconsistently demonstrated in the (C) 1995 Rapid Science Publishers blood of asthmatic subjects. Peripheral neutrophils of asthmatic subjects seem to have higher releasability of mediators, 1-15 higher superoxide anion generation 1 and higher migratory activity 11 than in normal subjects. However, the differences between asthmatic and normal subjects are mild; the observed abnormalities are not strictly related to the severity of asthma and they do not change significantly with the treatment of asthma. Thus, the significance of these abnormalities in the pathophysiology of the disease has not been demonstrated. Since neutrophils contain many destructive proteases and are able to release arachidonic acid metabolites, such as the potent chemoattractant leukotriene B4 (LTB4), these cells can be considered to play a pathologic role in asthma.
Glucocorticosteroids (GCS) and other antiinflammatory drugs now represent the first choice in the treatment of asthma. The efficacy of inhaled or systemic GCS can be evaluated by the changes in pulmonary function tests and bronchial reactivity, but also by the imjrovement in the markers of airway inflammation. GCS can Mediators of Inflammation Vol 4 1995 251 modify several functions of the inflammatory cells by i.m or oral route, or equivalent doses of involved in asthma, and these effects can explain deflazacort (Flantadin, Lepetit, Italy). They were the efficacy of these drugs in asthma treatment, examined every 2 weeks until a significant Some in vitro studies have shown that GCS can improvement in asthma symptoms occurred. At stabilize lysosomes, inhibit chemotaxis and other the end of the treatment period, they repeated neutrophil functions, 7-2 but these effects occur pulmonary function tests and blood measureat very high concentrations of GCS, and their ments. clinical significance has not been proven. Few in In the same time, eleven normal subjects from vivo or ex vivo studies have evaluated the effect our Unit team were examined as regards the of physiologically important concentrations of spontaneous and 20 I.tM calcium ionophore GCS on neutrophil functions, showing effects on A23187-induced LTB4 release from whole blood, some 9-2 but not on other functions. 8'22 migratory activity and superoxide anion genera-The aim of our study was to assess whether tion of isolated peripheral neutrophils. treatment with systemic GCS in subjects with exacerbation of asthma was able to modify several functions of peripheral neutrophils Mthod potentially involved in the pathophysiology of asthma, such as chemotaxis, release of oxygen Pulmonary function tests: Expiratory flowradicals and LTB 4. Furthermore, the effect of volume curves were performed by means of a GCS treatment on these neutrophil functions was HP Pulmonary Desk System 47804/A; three related to the improvement in the airway acceptable manoeuvres were obtained each time, obstruction as assessed by spirometry, according to ATS criteria. (2) leucyl-phenylalanine (f-MLP, Sigma) or 300 btg/ml increase in FEV > 12% after acute administration zymosan (ZAS, Sigma). Control incubations were of inhaled betai-agonists or after a short term performed in the absence of stimuli. After incucourse of intensive treatment with bronchodilabation, serum was obtained by centrifugation at tors and GCS; 23 or (3) methacholine PD20FEV 1 500 x g for 10 min and divided into aliquots lower than 1 mg. The mean duration of asthma for subsequent eicosanoid assay. Leukotriene B4 was 11 4-9 years. At the time of the first (LTB4) concentration in the supernatant was examination, all subjects were under regular measured using radioimmunoassay (RIA)with a treatment with bronchodilators (inhaled beta2tritiated tracer (Amersham, UK); an LTB 4 stanagonists and/or oral theophylline) and inhaled dard was purchased from Upjohn (Kalamazoo, anti-inflammatory drugs (sodium cromoglycate or MI, USA); and the specific antiserum was a gift nedocromil sodium in eight subjects, beclofrom Dr Frank Carey and Dr Robert Forder, ICI methasone dipropionate 800-1500 lg daily in Pharmaceuticals, UK. The sensitivity of LTB4 RIA 13 subjects). The aetiology of asthma was allergic was 4.3 4-0.9 pg (coefficient of intra-assay variain eight subjects, occupational in two subjects, tion: 12%). and intrinsic in the remaining eleven subjects. The first examination was done when subjects Preparation of isolated neutrophils: For preparaexperienced a spontaneous exacerbation of tion of granulocytes, ACD-anticoagulated blood asthma, with increase in asthma symptoms (60ml) was centrifuged at 180 x g for 10 min; despite their current treatment. At this time, they the supernatant platelet-rich plasma was disperformed spirometry and collected a blood carded, and the remaining aliquot was mixed sample to measure total and differential white with a 3.5% solution of dextrane T500 (Pharmablood cells (WBC), migratory activity and other cia, Uppsala, Sweden) in saline solution, to a markers of activity of peripheral neutrophils, final concentration of 1.5%. After allowing differ-Systemic GCS were added to their regular treat-ential cell layering for 60 min at room temperament: 6-methyl-prednisolone (Urbason, Hoechst, ture, leukocyte-rich plasma was collected and Italy) 40 mg/day for 3 days and then 20 mg/day centrifuged at 270 x g for 10 min; after dischar-ging the supernatant, the pellet was resuspended trifuged at 1 500 x g for 5 min at 4C to remove in 7.5 ml phosphate-buffered saline without cells. The reduction of cytochrome C was meacalcium and magnesium (PBS, Sigma)containing sured by reading the adsorbance of the super-0.5% bovine serum albumin (BSA), carefully nant at 550 nm wavelength. The intra-assay layered onto 3.0 ml of Ficoll-Hypaque (Lymvariation was 4.9%. phoprep, Nycomed AS, Oslo, Norway) and centrifuged at 350 x g for 30 min. Mononuclear Release of LTB4 after different stimuli. A suspencells at the interface between PBS and Ficollsion of isolated granulocytes (concentration, Hypaque were removed (see below); lysis of red 7.5 x 106 cells/ml) was incubated for 30 min at blood cells was performed by adding 6 ml of 37C in the presence of either 20 btM calcium distilled water to the pellet for 15 s, and osmo-ionophore A 23187, or 0.2 mM N-formyl-methiolarity was re-equilibrated to normal levels with nyl-leucyl-phenylalanine, or 300 l.tg/ml zymosan. 2 ml of 3.5% NaC1 solution. After further cen-Control incubations were performed in the trifugation (270 x g for 5 min), the final pellet absence of stimuli. At the end of the incubation, was resuspended in 1 ml of PBS and cell count a supernatant was obtained by centrifugation at performed in a Thoma-Zeiss microscope 1 500 x g for 10 min and divided into aliquots chamber. Cell viability was evaluated by Trypan for subsequent eicosanoid assay. The LTB4 conblue exclusion and was always >95%. Neucentration in the supernatantwas measured using trophil purification, evaluated on a standard the same radioimmunoassay (RIA) as described May-Grunwald-Giemsa smear, was also > 959/o. previously. Purified neutrophils were used to perform measurements of migratory activity, generation of Statistical analysis: Data are reported as mean _+ superoxide anion, and release of LTB4 after dif-standard deviation (S.D.). Paired and unpaired ferent stimuli.
Student's t-tests were performed when appropriate. Regression analysis was used to correlate Migratory activity. Chemokinesis and chemotaxis pulmonary function tests with indices of neutroof granulocytes were measured by a modified phil activity in the blood, a level of sigm4ficance Boyden chamber technique. A 0.7 ml aliquot of a lower than 5% was considered significant.
cell suspension adjusted to 1 x 106/ml was put into the upper compartment of a Boyden chamber, using a 13 mm diameter and 3 lm Rult pore size filter (Millipore Co, Bedford, USA). Dulbecco solution or 109/o E. coli culture super-At the time of asthma exacerbation, all subjects natant was placed in the lower compartment of were currently symptomatic with a moderate to the chamber to study chemokinesis (random severe airway obstruction (mean FEVx, migration) or chemotaxis, respectively. Incuba-1.55 +_ 0.48 1, 51.1 __+ 15.1% of the predicted tion was performed for 2 h at 37C in a CO2 value). Haematology showed high eosinophil incubator. Filters were fixed on a microscope count (>400 cells/ll) in nine of 21 subjects slide and stained with haematoxylin-eosin.
(43%). The evaluation of the peripheral neutro-Migration of granulocytes was measured accordphil functions in the asthmatic subjects showed ing to the 'leading-front' technique on ten ran-no significant difference with respect to normal domly selected low power (x 40) fields. All subjects as regards the spontaneous and 20 tM samples were processed in triplicate. The intra-calcium ionophore A23187-induced LTB 4 producassay and inter-assay variations were 7% and 13%, In order to subtract any O2-dependent change in kinesis showed a trend in increasing in asthmatic adsorbance, the same assay was also carried out subjects in comparison with normal subjects in the presence of 10 btl of 3 mg/ml superoxide (91.6 _+ 15.6 lm vs. 86.4 _+ 6.8 lm, p 0.07).
dismutase. The reaction was stopped by placing In asthmatic subjects there was a mild sigthe tubes in ice, and the suspension was cen-nificant inverse relationship between FEV (in % Mediators of Inflammation Vol 4 1995 253 of the predicted value) and neutrophil chemotaxis (r -0.52, p 0.04).
At the end of the treatment with systemic GCS, when symptoms of exacerbation were recovered in all subjects, mean FEV and FVC significantly increased with respect to the pre-treatment values ( Table 1). Haematology showed a significant increase in total WBC and neutrophil count, and a significant decrease in eosinophil count. A significant reduction of neutrophil chemokinesis was observed after treatment. No change was observed in chemotaxis and superoxide anion generation from isolated peripheral neutrophils ( Table 1).
The spontaneous and induced release of LTB 4 from whole blood and from isolated peripheral neutrophils was not significantly reduced after treatment with GCS (  An arbitrary score of the total dose of GCS administered during all periods of treatment (daily dose of 6-methyl-prednisolone or equivalent dose of deflazacort x number of days of treatment) was computed for each patient. Responders had a similar score to non-responders. This score was not significantly related to changes in the haematology or indices of neutrophil functions; there was only a significant relationship between this score and the decrease after treatment in chemokinesis from isolated peripheral neutrophils (r 0.57, p < 0.05).
There was no difference in the changes of indices of neutrophil activity between subjects who received regular beta.-agonists in addition to systemic GCS treatment and subjects who did not, nor between subjects who were previously treated with inhaled GCS and those treated with cromones.

Discussion
We showed that treatment with systemic GCS, at doses able to induce a significant improvement in pulmonary function in asthmatic subjects, was able to cause minimal effects on some indices of activity of peripheral blood neutrophils. Chemotaxis towards E. coli endotoxin was the only marker of neutrophil activity which significantly reduced after GCS in subjects who Table 3. Mean values (-t-SD) of haematology, migratory activity and superoxide anion generation from isolated peripheral neutrophils in asthmatic subjects who responded or not to systemic GCS treatment with an FEV1 increase of 20% or more responded to GCS treatment with an increase in FEV1, but no change in LTB4 release and superoxide anion generation from neutrophils was observed. These data confirm and extend the results obtained by Gin and coworkers 19 on six asthmatic patients; in particular, the effect was obtained only in subjects clinically sensitive to GCS and not in subjects who did not show any significant change in FEV after treatment. Therefore, the decrease in neutrophil chemotaxis could be considered as a consequence of GCS treatment on airway inflammation, and not as a nonspecific effect of steroids directly on peripheral blood cells, because the same effect was not observed in subjects who were non-responders to GCS treatment, though they received similar doses of GCS. On the other hand, chemokinesis was affected by GCS in a dose-dependent manner in all subjects, independently from the degree of clinical response, suggesting that this neutrophil function can be directly affected by GCS and not by an effect on the mediators of the asthmatic inflammation. These results are in contrast with previous reports, suggesting a lack of effect of GCS on neutrophil locomotion in normal and asthmatic subjects, 18'25 but they agree with our previous results obtained in asthmatic subjects tested with high dose inhaled beclomethasone dipropionate for 1 month. 26 It is not surprising that some asthmatic patients did not show a significant improvement in FEV1 after GCS treatment. It is well known that some patients are resistant to the steroids, 27 and that the effect of GCS on FEV can require a different duration of treatment. Although all our patients showed at the diagnosis the typical functional abnormalities of asthma (reversibility and/ or bronchial hyperreactivity), it is possible that some of them had a concomitant chronic obstructive pulmonary disease of variable degree which could contribute to the airway obstruction.
Patients who were responders to GCS showed, in effect, higher eosinophil counts in the blood than non-responders. Furthermore, the posttreatment evaluation was performed after improvement in symptoms, which could not correspond to an improvement in pulmonary function. However, the duration of treatment and the total administered dose of GCS were equivalent in responders and non-responders, suggesting that the lack of clinical response was not due to an insufficient treatment in non-responders.
The lack of effect of GCS on LTB4 release from whole blood and isolated neutrophils after several stimuli confirm what has been previously reported after a short-term treatment with GCS. 22 LTB 4 release from neutrophils could be considered as a way to recruit new cells in the airways after an initiating stimulus; because neutrophils are usual inhabitants of the airways, z8 this function could be physiologically important. Although some methodological problems need to be considered (the dose and the characteristic of the different physiological and non-physiological stimuli to induce LTB4 release from isolated neutrophils), these results confirm a minor role of neutrophil activation in asthma. The higher release of LTB4 from whole blood after GCS is due to the higher number of peripheral neu-tr0Phils, induced by the well-known effect of GCS on leukocyte margination. The different increase in LTB 4 release in whole blood and in isolated neutrophils after the various stimuli was due to the different amount of neutrophils obtained in the different preparations, and to the presence of platelets in whole blood which can increase the availability of arachidonic acid as a substrate for LTB4 production from neutrophil lypoxygenase.
We must consider, however, that neutrophils of the airways, obtained by bronchial or broncho-alveolar lavage (BAL), could show a higher degree of activation, and a greater mod-Mediators of Inflammation Vol 4 1995 255 ulation in their functions by anti-inflammatory treatment, with respect to blood neutrophils. Considering that several studies showed a good correlation between activation of eosinophils derived from the blood and from BAL fluid, 29 the difference between blood and BAL neutrophils, although hypothetical, is not very probable.
As an additional finding, this study shows that most functions of peripheral neutrophils from asthmatic subjects during asthma exacerbations are not different from those measured in a small group of normal subjects, except for superoxide anion generation. Furthermore, chemotaxis toward E. coli endotoxin was significantly correlated with an index of asthma severity like FEV.
The partial disagreement between previous data 1-6 and our observation regarding the increased capability of peripheral neutrophils to generate arachidonic acid metabolites and superoxide anion, could be explained by the small group of normal subjects in our study, by some methodological differences and by the different severity of the disease.
In conclusion, our study shows that neutrophil activation plays a minor role in asthma, and that treatment with systemic GCS is not able to modify most neutrophil functions in asthmatic subjects. Chemotaxis only seems to be related to the severity of asthma and it could reflect the improvement of the disease after GCS treatment.
However, further studies on airway neutrophils are needed.