INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Leukocyte recruitment into the airways is a feature of bronchial asthma. Increased number of both activated T cells and eosinophils have been a consistent finding in bronchial biopsies (1), bronchoalveolar lavage fluid (BALF) (2, 3), and in peripheral blood (4) derived from asthmatic patients. Moreover, exposure of asthmatic patients to a relevant allergen further increases leukocyte recruitment into asthmatic airways through the upregulation of vascular adhesion molecules (5) and release of cytokines from mast cells (6) and activated Th-2-like lymphocytes (7). Eosinophils are considered to play a key role in leading to disruption of the bronchial epithelium, enhanced bronchial responsiveness, and airway obstruction (8, 9). Cytokines encoded in the interleukin-4 (IL-4) gene cluster on Chromosome 5, specifically IL-3, IL-5, and granulocyte-macrophage colony-stimulating factor (GM-CSF), play a key role in eosinophil recruitment, migration, and activation associated with allergen exposure (7, 10), although this does not preclude other cytokines and mediators from being involved.

Over the past decade, a new family of cytokine peptides, known as the chemokines, has been identified. These chemokines are 8- to 10-kD, basic heprin-binding polypeptides that have been subdivided into C-X-C, C-C, and C branches according to their amino-acid sequences (11, 12). The chemokines regulated on activation, normal T-cell expressed and probably secreted (RANTES), macrophage inflammatory peptide-1 (MIP-1), and macrophage chemotactic peptide (MCP-1) are members of the C-C branch and are thought to play an important role in the allergic inflammatory process. RANTES and MIP-1 are chemotactic for lymphocytes, monocytes, and eosinophils (13, 14), whereas the activities of MCP-1 are restricted to monocytes, lymphocytes (15, 16), and basophils (17). In vivo evidence for the involvement of these chemokines in leukocyte recruitment has been obtained from experimental animal studies. Injection of RANTES into dog skin results in the recruitment of monocytes and eosinophils at the injection site (18). Similarly, at 4 to 8 h after intratracheal administration of Schistosoma mansoni eggs into mouse airways, there occurs recruitment of eosinophils in parallel with MIP-1 release (19). Increased levels of RANTES and MIP-1 have been detected in the nasal secretions of atopic patients exposed to local allergen challenge (20), and MCP-1 has been reported to be increased in nasal washings of ragweed-sensitive subjects during the pollen allergen season (21). Recently, these three chemokines have been found to be present in increased levels in the bronchoalveolar lavage fluid (BALF) of patients with active asthma (22). Immunoreactivity for MCP-1 is also increased in the bronchial epithelium of asthmatic patients (23). More recently, we have reported that allergen challenge leads to the release of RANTES and MIP-1 into asthmatic airways, at 4 h after challenge (24, 25), whereas Sur and colleagues have reported RANTES release at 24 h after allergen challenge (26). In the present study we have investigated the presence of RANTES, MIP-1, and MCP-1 immunoreactivity in BALF derived from normal subjects and patients with mild allergic asthma. We further studied the effect of endobronchial (local) allergen challenge on the release of these chemokines, and their relationship with leukocyte recruitment in BALF at 4 and 24 h after asthmatic subjects were exposed to allergen. We also report the production of RANTES, MIP-1, and MCP-1 by BALF cells.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Seven normal subjects (six male and one female; median age: 24 yr; age range: 20 to 40 yr) and 25 patients with mild asthma (18 male and seven female; median age: 31; age range: 20 to 55 yr) were recruited for the study; their clinical characteristics are summarized in Tables 1 and 2. At the time of enrollment, all of the asthmatic subjects had stable pulmonary function, with a baseline FEV₁  70% of that predicted for their age and height. None of the asthmatic subjects had been treated with inhaled or oral corticosteroids, sodium cromoglycate, or theophylline for at least 6 wk prior to participation in the study. All of asthmatic subjects were atopic as defined by a > 3 mm skin-wheal response to one or more of five common allergens (Dermatophagoides pteronyssinus, mixed grass pollen, dog, feathers, and cat dander (Hollister Stier, Elkhart, IN). All of the asthmatic subjects were hyperresponsive to inhaled methacholine, with their geometric mean concentration of methacholine needed to evoke a 20% decline in FEV₁ (PC₂₀) value ranging from 0.1 to 2.9 mg/ml (Tables 1 and 2). The study was approved by the Southampton Hospitals and University Ethical Committee, and patients gave their written informed consent.

Study Design

The asthmatic patients were divided into four groups; five individuals formed part of more than one data group (27). Group A (n = 8) had a single bronchoscopy for bronchoalveolar lavage (BAL). Group B (n = 9) had allergen instilled into the middle lobe (ML) and saline into the right upper lobe (RUL), followed 4 h later by a second bronchoscopy for BAL of the ML and RUL. Group C (n = 8) underwent an identical procedure to that for Group B, but with the second bronchoscopy undertaken 24 h after allergen challenge and saline instillation. In order to determine the effects of bronchoscopy and BAL independently of the effects of allergen, Group D (n = 5) was treated exactly as Group B, with the exception that both the ML and RUL were exposed to saline, with BAL being performed in the ML 4 h later.

The allergen extract (mixed grass pollen or D. pteronyssinus) (Hollister Stier) used for local bronchial challenge was that which produced the largest wheal response on skin-prick testing. In each subject, a skin-wheal dose-response series was undertaken, using 10-fold dilutions of allergen, and the allergen concentration chosen for the segmental bronchial challenge was 10 times lower than the concentration that produced a 3-mm wheal response. The median concentration of allergen inducing a 3-mm wheal response was 10⁴ (range: 10³ to 10⁶) (Table 2).

Bronchoscopy and Local Challenge

Fiberoptic bronchoscopy was undertaken as previously described (28). All subjects received 2.5 mg salbutamol by nebulization 15 min prior to the procedure, and intravenous atropine 0.6 mg and midazolam 3 to 8 mg. Oxygen (100%) was administered via nasal prongs throughout the procedure, and oxygen saturation (Sa_O2) was monitored with a digital oximeter (Minolta, Middlesex, UK). Fiberoptic bronchoscopy was done with an Olympus IT-20 bronchoscope (Olympus Optical Co., Tokyo, Japan). The sham challenge was undertaken in the RUL with 20 ml of prewarmed saline, and the allergen challenge was performed in the medial segment with 20 ml of allergen solution. Either 4 or 24 h later, the second bronchoscopy was performed and BAL was undertaken with six 20-ml aliquots of pre-warmed 0.9% saline solution. On completion of the procedure, subjects were observed for 3 h, and additional nebulized salbutamol was given as necessary.

Processing of BAL Fluid

The BAL fluid (BALF) obtained from bronchoscopies was immediately filtered through a 0.9 mm-gauge sieve and centrifuged at 400 × g for 10 min at 4° C to separate cells from fluid. BALF thus processed was then aliquoted into plastic tubes and frozen at 70° C. The cell pellet was resuspended in 5 ml of RPMI-1640 (containing 5% human AB serum with 100 U/ml streptomycin), and the cells were counted in a hemocytometer. A 100 µl aliquot of cells (1 × 10⁶/ml) was subjected to cytocentrifugation (Cytospin, Shandon Southern, Runcorn, UK) and air dried to obtain differential cell counts after staining with May- Grunwald-Giemsa stain. Six hundred cells were counted per cytospin.

BAL Cell Culture

BAL cells obtained from asthmatic subjects undergoing endobronchial allergen challenge were washed in RPMI-1640, resuspended in culture medium (10⁶ cells/ml), and cultured at 37° C in an incubator containing 5% CO₂. The culture medium consisted of RPMI-1640 with 5% human AB serum, 1 mM 1-glutamine, 2 mM sodium pyruvate, 100 U/ml streptomycin, and 0.5 µg/ml fungizone (Gibco, Paisley, UK). After 24 h incubation, culture supernatant of BAL cells was collected and kept at 70° C for measurements of chemokines.

Measurements of Chemokines

Measurement of the chemokines RANTES, MIP-1, and MCP-1 in both BALF (10 × concentrated, using 3-kDA-cutoff Centricon filters; Amicon Inc., Beverly, MA) and BAL cell culture supernatant (diluted up to 5-fold to get the sample concentration onto the linear part of the enzyme-linked immunoassay [ELISA] standard curves) was done with ELISA kits according to the manufacturer's protocol (R&D Systems, Abingdon, UK). Study samples and standard dilutions of specific cytokine were assayed in duplicate. Each assay had demonstrated no measurable cross-reactivity to other cytokines as determined by ourselves and as stated by the manufacturer. The limits of detection for RANTES, MIP-1, and MCP-1 were 5 pg/ml, 10 pg/ml, and 15 pg/ ml, respectively. Concentrations of chemokines in BAL cell culture supernatant were corrected for their initial dilution, whereas concentrations of chemokines in BALF were given as measured in 10×-concentrated BALF.

Albumin Measurements in BALF

Measurements of albumin in BALF samples (nonconcentrated BALF) were undertaken with rocket immunoelectrophoresis (29). Sensitivity of the assay was 1 µg/ml, and the coefficient of variation (cv) of repeat measurements was < 5%.

Statistical Analyses

Statistical analyses of RANTES, MIP-1, and MCP-1 in BALF and BAL cell supernatant was performed with Wilcoxon's paired rank test. Spearman's rank correlation coefficient was used to investigate any relationship between cell counts and RANTES, MIP-1, and MCP-1 concentrations. A value of p < 0.05 was regarded as statistically significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Steady-state Asthma

BALF and cell population. BALF was obtained from all subjects taking part in the study. There was no significant difference in the volume of lavage fluid recovered from the normal and the asthmatic subjects (59.3 ± 4.3 ml versus 60.8 ± 5.1 ml, p > 0.05). Differential cell counts showed that BALF from the asthmatic subjects contained more inflammatory cells; however, only eosinophil numbers reached statistical significance (p < 0.05) (Table 3). Baseline measurement of albumin in BALF failed to reveal any significant difference between the normal controls and the subjects with mild asthma (30.5 µg/ml, range: 15 to 70 µg/ml; versus 47.2 µg/ml, range: 15 to 95 µg/ml, respectively).

Chemokine concentrations in BALF. Measurement of the chemokines RANTES, MIP-1, and MCP-1 in BALF with ELISA showed that these chemokines were present in both normal controls and stable asthmatic patients. No difference between the two groups was found in the baseline levels of the three chemokines (Figure 1). Because the recovery of BALF can vary from one subject to another, levels of RANTES, MIP-1, and MCP-1 were also analyzed in relation to concentration of albumin (data not shown). The comparative results obtained did not differ from those that were expressed per milliliter of BALF. There was no correlation between the level of any of the chemokines and the total number of leukocytes or leukocyte subsets present in BALF (data not shown).

View larger version (9K): [in this window] [in a new window]   Figure 1.   Levels of RANTES (A), MIP-1 (B), and MCP-1 (C ) immunoreactivity in BALF (10× concentrated) of normal and asthmatic subjects. Horizontal lines represent median values.

Allergen-induced Asthma

BALF and cell population. There was no significant difference in the volume of fluid recovered from saline- or allergen-exposed bronchial segments (54.3 ± 3.3 ml versus 45.5 ± 4 ml, respectively). Eosinophil numbers were significantly increased in BALF from the allergen-challenged as compared with the saline-challenged segments. At 4 h after allergen challenge (Group B), there was a 5-fold increase in eosinophils (p < 0.05), whereas at 24 h (Group C) the eosinophil count was 19-fold greater than at the saline-exposed site (p < 0.02). No significant difference was observed in the numbers of macrophages, lymphocytes, or neutrophils at the two sites (Table 4). As compared with baseline lavage (Group D), there was a significant increase in the number of neutrophils at the saline-challenge site, p < 0.005 (Table 4). Levels of albumin in BALF from the saline and allergen lavages were not significantly different at either 4 h (72.5 µg/ml, range: 31 to 124 µg/ml; versus 79.0 µg/ml, range: 33 to 225 µg/ml, respectively) or 24 h (59 µg/ ml, range: 21 to 136 µg/ml; versus 84 µg/ml, range: 25 to 177 µg/ ml, respectively).

Chemokine concentrations in BALF. As compared with those at the saline site, the levels of RANTES, MIP-1, and MCP-1 were greatly increased at the allergen-challenge site 4 h after challenge (median: 11.5 pg/ml versus 175 pg/ml, p = 0.02; 88 pg/ml versus 258 pg/ml, p < 0.05; and 450 pg/ml versus 900 pg/ml, p < 0.05, respectively (Figure 2). At the 24-h time point, the three chemokines were still detectable at both the allergen- and saline-exposed sites, although the levels were substantially lower than those measured at 4 h, there being no statistically significant differences between the concentrations of the three chemokines at the two sites at this later time point. Concentrations of the three chemokines in BALF obtained 4 h after challenge were significantly increased over those observed at 24 h, p  0.02 (Figure 2). The double saline challenge also induced the release of MIP-1 and MCP-1, but not of RANTES, into the BALF (Figure 3). However, levels of these chemokines induced by saline were substantially lower than those observed in the allergen-challenged segments. As compared with the double saline challenge, allergen challenge induced 2- to 3-fold greater levels of MIP-1 and MCP-1 in BALF (Figures 2 and 3).

View larger version (17K): [in this window] [in a new window]   Figure 2.   Concentrations of RANTES (A), MIP-1 (B), and MCP-1 (C ) immunoreactivity in BALF 10× concentrated) obtained 4 and 24 h after saline and allergen challenge. Horizontal lines represent median values.

View larger version (11K): [in this window] [in a new window]   Figure 3.   Concentrations of RANTES (A), MIP-1 (B), and MCP-1 (C ) immunoreactivity in BALF (10× concentrated) obtained 4 h after double saline challenge. Horizontal lines represent median values.

There was a significant correlation between the concentrations of RANTES and the numbers of eosinophils in BALF at 4 h after allergen exposure (r = 0.68, p = 0.04), but not after saline exposure (r = 0.21, p = 0.60). In contrast, at 24 h after challenge, no correlations were found between RANTES levels and eosinophil numbers at the saline- (r = 0.07, p = 0.86) or allergen-challenged (r = 0.08, p = 0.91) sites. There were no correlations between MIP-1 and eosinophil numbers at the saline- or allergen-challenge sites at either 4 h (r = 0.06, p = 0.06, versus r = 0.03, p = 0.93) or 24 h (r = 0.07, p = 0.86, versus r = 0.81, p = 0.01). Similarly, BALF concentrations of RANTES, MIP-1, and MCP-1 did not correlate with either the number of lymphocytes or of macrophages obtained from the saline- and allergen-challenged sites at the two time points (data not shown).

Release of Chemokines by BAL Cells

To investigate whether BAL cells spontaneously release RANTES, MIP-1, and MCP-1, concentrations of these chemokines were measured in the 24 h culture supernatant of BAL cells obtained 4 h after challenge. BAL cells derived from the allergen-challenge site released more RANTES and MCP-1 than did BAL cells from the saline-challenge site (median: 660.0 pg/ml versus 105.0 pg/ml, and 5.7 ng/ml versus 2.6 ng/ml; respectively, p < 0.05 (Figure 4). BAL cells derived from the allergen-challenge site also released greater levels of MIP- than did cells from the saline-challenge site (5.5 ng/ml versus 4.4 ng/ml), but this difference failed to reach statistical significance.

View larger version (12K): [in this window] [in a new window]   Figure 4.   Concentrations of RANTES (A), MIP-1 (B), and MCP-1 (C ) immunoreactivity in the 24 h culture supernatant of BAL cells obtained 4 h after saline and allergen challenge. Horizontal lines represent median values. Concentrations of chemokines were corrected for their initial 5-fold dilution.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Bronchial asthma is a chronic inflammatory disease that is characterized by infiltration of eosinophils and lymphocytes into the airway tissue. The present study has demonstrated that the chemokines RANTES, MIP-1, and MCP-1 are present in the epithelial lining fluid of both normal and mildly asthmatic subjects, with no difference in the concentrations of the three chemokines in the two populations. Of importance is that endobronchial allergen challenge of sensitized asthmatic airways induced the release of these chemokines in BALF at concentrations that were greatly increased at 4 h. However, by 24 h, levels of the chemokines had almost returned to normal values. In demonstrating ongoing release of RANTES and MCP-1 by BAL cells from the allergen-exposed site after 24 h of culture, we suggest that these chemokines may recruit leukocytes into asthmatic airways.

To study leukocytes infiltrating asthmatic airways, we performed endobronchial allergen challenge in patients with mild asthma. This allowed us to demonstrate that eosinophil recruitment is already evident at 4 h after allergen exposure, and that the numbers of eosinophils increase further at 24 h. In concurrence with our previous study, we found a nonspecific recruitment of neutrophils, which was predominantly related to the bronchoscopic procedure (27). Allergen challenge did not induce a significant increase in the number of lymphocytes or macrophages.

RANTES, MIP-1, and MCP-1 are potent leukocyte chemoattractants in vitro, and they are thought to be involved in the allergic inflammatory process. In the present study we have shown that these three cytokines are constitutively present in the airways of both normal subjects and patients with steady-state asthma, suggesting that they may regulate cell trafficking under physiologic conditions. Our findings fail to support Alam and colleagues' report that these chemokines are increased in BALF from patients with stable asthma (22). Since our patients were to undergo segmental allergen challenge, they were deliberately chosen to have milder disease as compared with those taking part in Alam and colleagues' study. We have previously demonstrated that endobronchial allergen challenge leads to the release of RANTES and MIP-1 at 4 h after challenge, and have shown that these two chemokines have biologic activity on eosinophils and lymphocytes, respectively (24, 25). The present study extends these observations by demonstrating that allergen challenge also induces the release of MCP-1 into asthmatic airways. Of particular interest in this study was the finding that levels of these chemokines were high at the 4-h time point and had returned to baseline values at 24 h after allergen exposure. Our findings do not support those in a report from Sur and associates indicating that allergen challenge induces RANTES release at 24 h after challenge (26).

In a previous study we demonstrated that endobronchial challenge with both allergen and saline causes release of the C-X-C chemokine IL-8 into asthmatic airways (27). However, in this previous study, we were unable to show any difference in IL-8 levels in BALF with the two exposures, suggesting that instrumentation and saline washing of the airways alone is sufficient to generate IL-8, probably from the epithelium. The present study extends this observation by showing that saline challenge alone induced the release of small amounts of MIP-1 and MCP-1, but not of RANTES, at 4 h after challenge; however, in contrast to the case with IL-8, allergen exposure results in a further increase in lavage levels of the C-C chemokines.

To investigate whether RANTES, MIP-1 and MCP-1 are involved in leukocyte recruitment, we sought correlations between these chemokines and cells infiltrating asthmatic airways. A significant correlation was found at the allergen-challenge site between the concentration of RANTES in BALF and the number of eosinophils recovered from the exposed segment at 4 h but not at 24 h. We have recently purified RANTES from BALF of six asthmatic subjects exposed to segmental allergen challenge, and have demonstrated its biologic activity on eosinophils (24). In this last study, levels of RANTES were also found to correlate with eosinophil numbers in BALF. These observations are consistent with the release of RANTES being involved in eosinophil recruitment during the early phase of the allergen-induced late asthmatic response (4 h after challenge). Chemoattractants other than RANTES may dominate eosinophil recruitment at 24 h after challenge. In a recent study it was suggested that IL-5 is the major eosinophil attractant associated with eosinophil recruitment at this time point (26). However, the possibility cannot be excluded that RANTES could have bound proteoglycans on the extracellular matrix (30), in which case this cytokine could recruit eosinophils by a hapotactic mechanism.

We have not been able to show a correlation between the high level of MIP-1 after allergen challenge and eosinophil numbers. MIP-1 has been reported to be an eosinophil chemoattractant in vitro, but as compared with RANTES it is relatively weak (31). RANTES, MIP-1, and MCP-1 exert chemotactic effects on monocytes when studied in vitro; however, on allergen challenge, there was no increase in the number of these cells in BALF at either 4 h or 24 hours after challenge. Similarly, the three chemokines have also been shown to exert chemotactic effects on basophils and on both CD4⁺ and CD8⁺ T lymphocytes. Although this matter was not specifically examined in the present study, increased numbers of activated T cells and basophils are known to occur from 4 to 24 h after allergen challenge of asthmatic airways (5, 32).

In the present study we did not investigate the cellular source of the chemokines detected in BALF. However, a number of cells, including alveolar macrophages, neutrophils, endothelial cells, and bronchial epithelial cells are known sources of these cytokines (12), and as such could contribute to airway chemokine release. Interestingly, free cells lavaged from the airways at 4 h after allergen as compared with saline exposure generated appreciably more RANTES and MCP-1 but not MIP-1 after ex vivo culture for 24 h. The finding that BAL cells from the allergen- and saline-challenge sites release similar amounts of MIP-1 suggests that the bronchoscopic procedure itself recruits and activates cells into asthmatic airways, and that these cells subsequently release this cytokine into the BAL cell culture supernatant.

Other chemokines that may play an important role in allergic inflammation include eotaxin, MCP-3, and MCP-4. However, there are no reagents commercially available for measuring these cytokines in biologic fluids. The development of sensitive assays for detecting these novel chemokines may allow their role in bronchial asthma to be defined.

In summary, the present study has demonstrated that RANTES, MIP-1, and MCP-1 are constitutively present in the BALF of both normal and asthmatic subjects. Endobronchial allergen challenge of asthmatic airways leads to release of additional amounts of these chemokines. Moreover, BAL cells derived from the allergen-challenge site spontaneously release RANTES and MCP-1. These findings suggest that the C-C chemokines contribute in the regulation of cell trafficking into asthmatic airways, especially after exposure to allergen, such as occurs during the late asthmatic response.