INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Asthma is a chronic inflammatory disease associated with paroxysmal bronchospasm, bronchial hyperresponsiveness, and the presence of an eosinophilic infiltrate in the airways (1). The development of tissue eosinophilia necessitates processes of eosinophil hematopoietic generation, mobilization into the blood, endothelial adhesion, chemotaxis to inflammatory foci, and survival in the tissues (2, 3). Interleukin (IL)-3, IL-5, and granulocyte-macrophage colony-stimulating factor (GM-CSF) induce the differentiation of eosinophils from progenitors in the bone marrow, increase the pool of available eosinophils in the circulation, and promote local eosinophil survival (4). In contrast, the adhesion and locomotion of eosinophils are regulated by specific chemoattractants. These include complement factor C5a; the lipid mediators platelet-activating factor, leukotriene B₄, and 5-oxo-6,8,11,14-ecosatetraenoic acid (5-oxo-ETE); and the chemokines macrophage inflammatory protein 1 (MIP-1), monocyte chemotactic protein (MCP)-2, MCP-3, MCP-4, RANTES (regulation on activation, normal T cell expressed and secreted), and eotaxin (5, 6). While the helper T cell type 2 (Th2) cytokines IL-4 and IL-5 are produced predominantly by infiltrating T cells (7), a number of eosinophilopoietic cytokines and eosinophil-active chemokines can be produced by resident airway cells such as epithelial cells (8), endothelial cells (9), alveolar macrophages (10), fibroblasts (11), and smooth muscle cells (12, 13).

Eotaxin is a C-C () chemokine that was identified by the microsequencing of high-performance liquid chromatography (HPLC)-purified proteins from bronchoalveolar lavage (BAL) fluid of ovalbumin-sensitized and challenged guinea pigs (14). Through the selective expression of its specific receptor, CCR3, human eotaxin is a potent chemoattractant for eosinophils (15, 16), basophils (17), and Th2-like T lymphocytes (18), all of which are found in tissues undergoing allergic reactions (1, 7, 19). Intratracheal instillation of eotaxin in rodents is followed by a marked lung eosinophilia (20). Similarly, injection of eotaxin in the skin of animals is associated with the rapid accumulation of eosinophils (14, 16, 20). In a guinea pig model of allergic airways disease, local eotaxin generation parallels the entry of eosinophils into the lungs (23). Furthermore, compared with sensitized and challenged wild-type mice, eotaxin-deficient mice exhibit a marked decrease in allergen-induced eosinophil recruitment into the airways (24).

We have shown the increased expression of eotaxin in the lungs of subjects of asthma compared with normal controls (25). In vitro eosinophil chemotaxis by bronchoalveolar lavage fluid from subjects with asthma was partly inhibited with antibodies against eotaxin to a greater extent than with antibodies against RANTES or MCP-4. At sites of inflammation in asthma, allergic rhinitis, and chronic sinusitis, eotaxin immunoreactivity has been localized to various inflammatory cells including monocyte-macrophages, T cells, and eosinophils (25, 26). However, the observation that eotaxin is also expressed by epithelial cells (25, 26) suggests that structural cells may also contribute to tissue eosiniphilia through eotaxin generation in human allergic inflammation. Indeed, tumor necrosis factor  (TNF-) and IL-1cytokines that are increased in allergic inflammationhave been shown to act on bronchial and alveolar epithelial cell lines to upregulate eotaxin synthesis and release (27).

Airway myocytes are recognized as important target cells in asthma owing to their ability to contract in response to inflammatory cell products including histamine, eicosanoids, cytokines, and proteins released from eosinophils (28). Further, through growth and proliferation, airway smooth muscle (ASM) cells have been implicated in lung remodeling, a feature of asthma that can contribute to persistent airway narrowing and bronchial hyperresponsiveness (28). Work has suggested that the functions of ASM are not restricted to contractile and growth responses. In vitro, ASM has also been shown to express immunoglobulin receptors (29), HLA-DR (30), vascular cell adhesion molecule 1 (VCAM-1), intercellular adhesion molecule 1 (ICAM-1) (31), and a limited repertoire of cytokines including RANTES (12), IL-1 (32), IL-6, and IL-11 (33). These observations raise the possibility that ASM has the potential to regulate inflammatory responses.

In asthmatic lungs, eosinophilia can be observed within and around ASM (34). However, the possibility that human ASM cells are a source of eotaxin has not been investigated. In this study, we show that airway myocytes express eotaxin in vitro and in vivo in subjects with asthma. ASM production of eotaxin was upregulated by TNF- and IL-1 stimulation, and eotaxin accounted for a significant proportion of the eosinophil chemoattractant activity produced by ASM cells.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Cell Culture

Human ASM cells from two sources were used. Bronchial/tracheal smooth muscle cells (B/TSMCs) were purchased from Clonetics (San Diego, CA). These cells stained positively for -smooth muscle actin and negatively for factor VIII, CD45, and CD3. B/TSMCs were grown as recommended by the supplier in their optimal medium (SmGM-2; Clonetics) containing 5% fetal bovine serum (FBS) at 37° C in a humidified incubator with 5% CO₂. The second source of ASM cells was human tracheas, which were obtained from lung transplant donors in accordance with procedures approved by the University of Pennsylvania Committee on Studies Involving Human Beings. Tracheal smooth muscle cells (TSMCs) were isolated and purified from the trachealis muscle as described by Panettieri and coworkers (33, 35). TSMCs were grown at 37°C with 5% CO₂ in Ham's F12 medium supplemented with 10% FBS, penicillin (10³ U/ml), streptomycin (1 mg/ml), NaOH (12 mM), CaCl₂ (1.7 mM), L-glutamine (2 mM), and 25 mM HEPES. These cells retain smooth muscle-specific actin expression and have the requisite receptor/second-messenger systems necessary to support both contractile and relaxant responses (35). B/TSMCs and TSMCs grow with the hill-and-valley appearance characteristic of smooth muscle in culture, and are elongated and spindle shaped with a central nucleus.

Cell Stimulation

Confluent B/TSMCs and TSMCs in passages 3-9 were growth arrested by FBS deprivation for 48 h (31). After serum deprivation, the cells were stimulated in fresh, serum-free medium containing TNF- and/or IL-I (R&D Systems, Minneapolis, MN) in a concentration- and time-dependent manner.

Semiquantitative RT-PCR and Southern Analysis

Total cellular RNA was extracted from culture flasks using the Trizol isolation reagent (Life Technologies, Gaithersburg, MD). Reverse transcription was performed by using 2 µg of total RNA in a first-strand cDNA synthesis reaction with the Moloney murine leukemia virus reverse transcriptase (RT; Life Technologies). Comparison was achieved by amplifying the constitutively expressed -actin gene (a housekeeping gene) and eotaxin under subsaturating conditions in parallel tubes as previously described (25, 36). -Actin was used as the standard to control for variations in RNA isolation, cDNA synthesis, and polymerase chain reaction (PCR) performance. A sample of cDNA was subjected to sequential cycles of amplification (20, 25, 30, 35, and 40 cycles). Samples were amplified at 94° C for 1 min, 60° C for 2 min, and 72° C for 3 min. A 322-bp fragment was generated using specific primers for human eotaxin (25). The optical density obtained for each amplified fragment was plotted against the number of cycles. The amounts of PCR-generated bands increase logarithmically up to a certain number of cycles, reaching a plateau thereafter. Under these conditions, it was established when PCR were in the exponential (quantifiable) phase. The quantification was achieved by scanning the band intensities obtained on ethidium bromide-stained agarose gels with an Instant Imager system 2000 (Pharmacia Biotech, Piscataway, NJ). Bands were transferred to nylon membranes and Southern analysis was performed with an internal primer for eotaxin (25) to verify the specificity of the PCR product. Enhanced chemiluminescence detection (Boehringer GmbH, Manheim, Germany) was used for Southern blots according to the instructions of the manufacturer.

ELISA

Supernatants from serum-deprived B/TSMCs (cytokine stimulated or unstimulated) were collected from culture flasks, centrifuged at 1,200 rpm for 7 min at 4° C to remove cellular debris, and stored at 80° C until use. Eotaxin release in supernatants was assayed by a sandwich enzyme-linked imunosorbent assay (ELISA) as described previously (25, 27). Each well of a high-binding efficiency 96-well ELISA plate was coated with a mouse anti-human eotaxin monoclonal antibody (2A12; see References 25 and 27) at 400 ng in phosphate-buffered saline (PBS) for 4 h at room temperature. All incubations were carried out in a humidified atmosphere. Residual binding sites were blocked with 3% bovine serum albumin (BSA, 200 µl/well; Sigma Chemical Co., St. Louis, MO) in PBS with 0.02% azide and incubated overnight at room temperature. After washing with PBS, standard eotaxin solution or B/ TSMC supernatants (50 µl/well in PBS containing 3% BSA) were added in duplicate to the coated wells, incubated for 2 h at room temperature, and washed again with PBS. Fifty microliters of rabbit anti-eotaxin affinity-purified polyclonal antibody (650 µg/ml) was subsequently added at 1:1,000 in 3% BSA-PBS. After a 2-h incubation at room temperature the plates were washed three times with PBS, and 50 µl of horseradish peroxidase-linked goat anti-rabbit IgG (Kirkegaard & Perry, Gaithersburg, MD) diluted 1:1,000 in PBS containing 3% BSA was added to each well and incubated for 90 min at room temperature. After washing, the binding was visualized with 3,3',5,5'-tetramethylbenzidine substrate according to the instruction of the manufacturer (Kirkegaard & Perry). The reaction was stopped after 15 min by adding 100 µl of 1 M H₃PO₄ per well, and absorbance at 650 nm was measured. Under these conditions, this assay is sensitive to 31 pg/ml.

PBMC and Eosinophil Purification from Peripheral Blood

Peripheral blood mononuclear cells (PBMCs) and granulocytes were purified from peripheral blood of nonatopic volunteers by Ficoll-Paque (Pharmacia Biotech) density centrifugation. After removing the mononuclear cell fraction, granulocytes were obtained by dextran sedimentation. Human eosinophils were further purified by negative selection with anti-CD16- and anti-CD3-coated immunomagnetic microbeads, using a magnetic cell sorting system (Miltenyi Biotec, Bergisch-Gladbach, Germany) at 4° C. The degree of purity of eosinophil populations, estimated after staining with Giemsa, was between 92 and 100%. Freshly isolated PBMCs and eosinophils were resuspended at concentrations of 2 × 10⁶ cells/ml in RPMI 1640 supplemented with penicillin (100 U/ml), streptomycin (100 µg/ml), 1 mM L-glutamine, and 1 mM sodium pyruvate. Cells were incubated for 1 h at 37° C in a humidified atmosphere of 5% CO₂ (25). Cells were washed, resuspended, counted, and used for the chemotaxis assays.

Chemotaxis Assay

Experiments were performed with a 48-well microchemotaxis chamber (NeuroProbe, Cabin John, MD) and carried out as previously described (15, 25). Migration of human PBMCs and eosinophils in response to recombinant eotaxin or B/TSMC supernatant was performed on a polycarbonate filter (5-µm pore size) and a polyvinylpyrrolidone-free polycarbonate filter (3-µm pore size) was used for PBMCs. Eosinophils (2 × 10⁶ cells/ml) and mononuclear cells (2 × 10⁶ cells/ml) were loaded into the chambers and incubated at 37° C, 5% CO₂ (60 min for eosinophils and 90 min for mononuclear cells). The filters were subsequently fixed and stained with a RAL kit (Labonord, France). Only cells morphologically identified as eosinophils or mononuclear cells were counted by microscopy in five high-power fields (magnification ×400) as previously described (14, 25).

For neutralization experiments, B/TSMC supernatants were preincubated with anti-eotaxin, anti-RANTES, or anti-eotaxin plus anti-RANTES (at 1:100 dilution) at 37° C for 1 h as described (25). The specificity of these antibodies has been shown previously (25, 27). Normal rabbit serum was used as a negative control.

Immunocytochemistry

B/TSMCs grown on eight-well glass slides (Naige Nunc, Naperville, IL) were serum deprived and stimulated with cytokines or incubated with medium alone. Slides were fixed with acetone-methanol (70:30), air dried, and stored at 20° C until use. Immunoreactive eotaxin was detected as described previously (25). Briefly, after treatment with Tris-buffered saline (TBS) containing 10% goat serum and 0.5% BSA for 30 min, slides were incubated with anti-eotaxin antiserum at a final dilution of 1:200 for 90 min at 37° C, followed by incubation for 1 h at 37° C with fluorescein isothiocyanate (FITC)-labeled swine anti-rabbit IgG (5 µg/ml). After each incubation with antibody, slides were extensively washed with TBS. Nuclei of cells were stained for 2 min with Hoechst 33258 dye (bisbenzimide; Sigma Chemical Co.). Normal rabbit serum (NRS) was used at the same dilution as a negative control. Slides were visualized with a Zeiss Axiophot fluorescence microscope (Carl Zeiss [Oberkochen], Ltd., Welwyn Garden City, UK).

In Situ Hybridization and Immunocytochemistry on Lung Sections of Individuals with Asthma

To determine whether ASM has the capacity to produce eotaxin in vivo, in situ hybridization (37) and immunocytochemistry (38) were performed on sections of the major airways from six subjects with asthma using an eotaxin cRNA probe and anti-eotaxin polyclonal antibodies (25, 26), respectively. A separate set of sections was also stained with monoclonal antibody for eotaxin (27) and irrelevant mouse monoclonal antibody of the same isotype as negative control. The airway sections were obtained from 11 subjects (6 with asthma and 5 nonasthmatic) who have been described previously in detail (34). The subjects were part of the St. Paul's Hospital Lung Study (Vancouver, Canada), for which ethical approval was obtained. A clinical diagnosis of asthma was made on the basis of an evaluation of patient medical files by a respiratory physician. The clinical criteria used to establish this diagnosis included prior physician diagnosis and treatment for asthma, documented evidence of variable airflow obstruction greater than 15%, and bronchial hyperresponsiveness. Immediately after resection, lung specimens were prepared for in situ hybridization and immunocytochemistry as described previously (34).

Statistical Analysis

Statistical significance was determined using a Student's t test. p Values < 0.05 were considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Constitutive and Cytokine-Induced Eotaxin mRNA Expression

In initial experiments, total RNA purified from TSMCs and B/ TSMCs was reverse transcribed and the cDNA amplified by PCR with specific primers for eotaxin. Gel electrophoresis revealed bands that corresponded to the predicted length of the eotaxin cDNA product (322 bp). The specificity of the PCR was confirmed by Southern analysis using an internal primer for eotaxin (Figure 1). Eotaxin mRNA expression was further studied in TSMCs cultured in the presence of TNF-, IL-1, or medium alone by semiquantitative RT-PCR using the constitutively expressed -actin as a standard (Figure 2). Expression of mRNA for eotaxin was found in cells cultured in medium alone and was increased after 24 h of stimulation with TNF- or IL-1 by 11-fold and sevenfold, respectively. Constitutive expression of eotaxin mRNA was also observed in B/ TSMCs (Figure 3). After 24 h of TNF- or IL-1 stimulation, this was increased by threefold and twofold, respectively. In B/ TSMCs, a combination of TNF- and IL-1 caused a greater increase in eotaxin mRNA expression compared with either cytokine alone.

View larger version (78K): [in this window] [in a new window]   Figure 1.   RT-PCR and Southern blotting demonstrating eotaxin mRNA expression in B/TSMCs. RNA purified from B/TSMCs was examined for the presence of eotaxin transcripts by RT-PCR (A). For positive controls, RNA from A549 cells, Hep-2 cells, and PBMCs was used. Putative eotaxin bands (322 bp) were transferred to nylon membranes and hybridized with a digoxigenin-labeled internal oligonucleotide specific for eotaxin (B) as described in METHODS.

View larger version (35K): [in this window] [in a new window]   Figure 2.   Expression of eotaxin mRNA in cytokine-stimulated and unstimulated TSMCs by semiquantitative RT-PCR. (A) RT-PCR was performed using total RNA purified from confluent TSMCs cultured in serum-free medium alone or in the presence of 25 ng/ml IL-1 or TNF- for 24 h. Eotaxin-specific signals of 322 bp were obtained. (B) RT-PCR for -actin. (C ) Below each blot a quantitative comparison is made at each dose between the eotoxin-specific signal and the -actin-specific signal of unstimulated TSMCs.

View larger version (34K): [in this window] [in a new window]   Figure 3.   Eotaxin mRNA expression in cytokine-stimulated and unstimulated B/TSMCs by semiquantitative RT-PCR. (A) Total RNA purified from confluent B/TSMCs cultured in serum-free medium alone or in the presence of 1 ng/ml IL-1, 100 ng/ml TNF-, or 100 ng/ml TNF- plus 1 ng/ml IL-1 for 24 h was subjected to RT-PCR using specific primers for eotaxin. (B) RT-PCR for -actin. (C ) Densitometric analysis of semiquantitative RT-PCR for eotaxin mRNA in cytokine-stimulated and unstimulated B/TSMCs. Below each blot a quantitative comparison is made at each dose between the eotaxin-specific signal and the -actin-specific signal of unstimulated B/TSMCs.

Time Course of Cytokine-Induced Eotaxin mRNA Accumulation

Treatment of B/TSMCs with TNF- (100 ng/ml) induced an increase in eotaxin mRNA expression at 6 h, reaching a maximal response at 24 h (Figure 4A). Levels of eotaxin mRNA remained elevated for the duration of the experiment (up to 48 h). Stimulation with IL-1 (1 ng/ml) also resulted in an increase in eotaxin mRNA by 6 h; this reached a maximum at 12 h and was still elevated at 24 and 48 h (Figure 4B).

View larger version (58K): [in this window] [in a new window]   Figure 4.   Kinetics of TNF-- and IL-1-stimulated accumulation of eotaxin mRNA in B/TSMCs by semiquantitative RT-PCR. Confluent serum-deprived B/TSMCs were stimulated with TNF- (100 ng/ml) or IL-1 (1 ng/ml) for 6, 12, 24, and 48 h. Eotaxin mRNA expression was examined at each time point after TNF- (A) and IL-1 (B) stimulation by semiquantitative RT-PCR. Below each blot a quantitative comparison is made at each dose between the eotaxin-specific signal and the -actin-specific signal of unstimulated B/TSMCs.

Dose-Response Effect of TNF- and IL-1 on Eotaxin mRNA Expression

The addition of increasing doses of TNF- (1, 10, 25, 50, and 100 ng/ml) to the medium 24 h before harvesting was associated with increasing eotaxin mRNA expression up to stimulation with 25 ng/ml TNF-, after which eotaxin mRNA levels declined modestly but were still higher than in cells incubated with medium alone (Figure 5A). IL-1 also had a dose-dependent effect on eotaxin mRNA induction. At 24 h, the maximal eotaxin mRNA response was observed with 25 ng/ml IL-1, and all IL-1 concentrations resulted in increased eotaxin mRNA compared with cells cultured in medium alone (Figure 5B).

View larger version (74K): [in this window] [in a new window]   Figure 5.   Dose-response effect of TNF- and IL-1 on the expression of eotaxin mRNA in B/TSMCs by semiquantitative RT-PCR. The effect of increasing concentrations of TNF- (A) and IL-1 (B) on eotaxin mRNA expression in B/TSMCs was determined by semiquantitative RT-PCR. Total RNA was extracted 24 h after the addition of the cytokines to the culture medium. Below each blot a quantitative comparison is made at each dose between the eotaxin-specific signal and the -actin-specific signal of unstimulated B/TSMCs.

Examination of Eotaxin Expression in B/TSMCs by Immunocytochemistry

Immunocytochemistry of B/TSMCs with anti-eotaxin antiserum confirmed the expression and localization of eotaxin immunoreactivity in B/TSMCs cultured in medium alone and in those stimulated with IL-1 or TNF- (Figure 6 and data that is not shown). The expression of eotaxin mRNA assessed by the intensity of the signal was increased after TNF- or IL-1 stimulation. No positive signal was observed when normal rabbit serum was used instead of the primary antibody.

View larger version (77K): [in this window] [in a new window]   Figure 6.   Expression of eotaxin in ASM in vitro and in vivo. (Top) Immunofluorescent staining of eotaxin in unstimulated B/TSMCs. B/ TSMCs grown on eight-well glass slides were fixed in acetone-methanol. Slides were incubated with rabbit anti-eotaxin antiserum followed by swine anti-rabbit IgG conjugated to FITC. (Middle) Cross-section of an intermediate airway of a subject with asthma showing eotaxin immunoreactivity in smooth muscle cells (large arrows) and epithelium (small arrow). Cryostat sections were cut from frozen lung tissue from asthmatic subjects. Slides were incubated with anti-eotaxin polyclonal antibodies, the appropriate secondary antibodies, and a tertiary layer of streptavidin-peroxidase. Sections were developed with diaminobenzidine and positive cells stained darker than surrounding cells. (Bottom) High-power magnification of transverse section of a large airway in a subject with asthma, demonstrating eotaxin immunoreactivity in bundles of smooth muscle (arrows) by the streptavidin-peroxidase method of immunostaining.

ELISA

To confirm the release of eotaxin from smooth muscle, we measured the levels of eotaxin in B/TSMC supernatant after incubation of cells with medium alone, TNF- (1, 10, 25, 50, and 100 ng/ml), or IL-1 (1, 10, 25, 50, and 100 ng/ml). Eotaxin was detected in supernatants from B/TSMCs in the absence of cytokine stimulation. Stimulation for 24 h with TNF- and IL-1 resulted in a marked increase in eotaxin release at all concentrations tested (Figure 7).

View larger version (16K): [in this window] [in a new window]   Figure 7.   Release of eotaxin protein by B/TSMCs detected by ELISA. B/TSMCs were stimulated for 24 h with increasing concentrations of TNF- and IL-1. Supernatants were recovered and eotaxin protein was measured by ELISA. (A) TNF- stimulation; (B) IL-1 stimulation. Each point represents the mean ± SD of four measurements. Measurements were made in duplicate on two supernatants from separate cultures for each point. *p < 0.05; **p < 0.01; ***p < 0.001.

Contribution of Eotaxin to Eosinophil and PBMC Chemotaxis Mediated by B/TSMC Supernatant

As a functional assay of B/TSMC-derived eotaxin, supernatants from cultures stimulated for 24 h with TNF- or IL-1 were examined for their chemotactic activity on eosinophils. Recombinant eotaxin increased the chemotaxis of eosinophils but not PBMCs (not shown). In contrast, B/TSMC supernatants induced the chemotaxis of both eosinophils and PBMCs at all concentrations tested (1, 1:2, 1:4, 1:8, 1:16, 1:32, 1:64; Figure 8). Anti-eotaxin antibodies had a marked inhibitory effect on the chemotactic activity of B/TSMC supernatant for eosinophils but not for PBMCs (Figure 8, Table 1).

View larger version (24K): [in this window] [in a new window]   Figure 8.   Cellular migration induced by B/TSMC supernatant in filter assays and inhibition of chemotactic activity with polyclonal anti-eotaxin antibody. Eosinophils (Eos) or PBMCs were incubated in a Boyden chamber with serial dilutions of supernatant (St) recovered from B/TSMCs stimulated for 24 h with TNF- (100 ng/ ml) (solid bars). To determine the contribution of eotaxin to eosinophil and PBMC chemotaxis, supernatants were preincubated with polyclonal anti-eotaxin antibodies (hatched bars) or normal rabbit serum. (A) Chemotaxis of human eosinophils; (B) chemotaxis of human PBMCs. The results of a representative experiment of three experiments are shown. Each point represents the mean number of cells ± SD in five fields (magnification ×400) per well in one chemotaxis chamber (25). Similar results were obtained using supernatants from B/TSMCs stimulated with 100 ng/ml IL-1 (Table 1). Control NRS had no effect on chemotaxis (not shown). *p < 0.001; **p < 0.01.

We compared the contribution of eotaxin with that of RANTES in eosinophil chemotaxis mediated by supernatant from B/TSMCs stimulated with TNF- or IL-1. The addition of antibodies to either chemokine alone resulted in significant decreases in eosinophil chemotaxis (Table 1); however, the difference between anti-eotaxin and anti-RANTES treatment was not significant in IL-1- or TNF--stimulated cultures. Anti-eotaxin plus anti-RANTES caused an inhibition of eosinophil chemotaxis that exceeded 70% compared with the eosinophil chemoattractant activity of supernatants without the antibodies. This inhibition was significant compared with the inhibitory effect of either antibody alone.

Eotaxin mRNA and Immunoreactivity in ASM of Individuals with Asthma

Using immunocytochemistry, eotaxin immunoreactivity (Figure 6) was detected in ASM of all six subjects with asthma who were investigated. As demonstrated previously, eotaxin protein also localized to the airway epithelium and infiltrating cells in the submucosa. Similar results of eotaxin immunoreactivity were obtained with polyclonal and monoclonal antibodies. Weak eotaxin immunoreactivity was observed in the airway smooth muscle of lung sections of subjects without asthma (data not shown). Using in situ hybridization, eotaxin transcripts were also detected in ASM of patients with asthma (data not shown).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In this study, we show that cultured human airway myocytes constitutively express eotaxin mRNA and protein. TNF- and IL-1 stimulation enhanced the accumulation of eotaxin transcripts in a concentration-dependent manner and a combination of the two cytokines had a greater effect than either alone. Kinetics of cytokine-induced eotaxin mRNA upregulation were observed; maximal responses occurred at 12 h and 24 h after IL-1 and TNF- stimulation, respectively. In vivo, signals for eotaxin mRNA and immunoreactivity were demonstrated in smooth muscle around asthmatic airways. Although we did not show any eosinophil infiltration around the smooth muscle cells in this study, we have previously reported an increased number of eosinophils in the smooth muscle area of larger airways (34). Furthermore, the release of eotaxin protein from ASM was confirmed in culture supernatants, which had a strong chemoattractant activity for eosinophils and PBMCs. The chemotactic responses of eosinophils to ASM supernatant were found to be partly inhibited with antibodies directed against eotaxin or RANTES, and a combined blockade of both chemokines caused an inhibition of eosinophil chemotaxis that exceeded 70%. Collectively, our findings suggest, but do not prove, that human ASM may contribute to airway inflammation in asthma through the expression, production, and release of eotaxin. The results of this study complement and extend previous investigations localizing eotaxin to macrophages, T cells, eosinophils, and epithelial cells in bronchial biopsies and BAL from patients with asthma (25), and observations that eotaxin is expressed in ASM of the allergen-challenged mouse (39) and guinea pig lungs (40). Our findings further implicate airway myocytes as effector cells with the capacity to regulate inflammatory responses in asthmatic lungs.

A number of processes believed to be important in the pathogenesis of asthma have been ascribed to the activities of TNF- and IL-1 (41). These cytokines are found at increased levels in lung lavage fluid from patients with asthma and their spontaneous release is augmented in alveolar macrophages from adult patients with asthma and wheezy infants (42, 43). Although TNF- and IL-1 can cause the airway hyperresponsiveness and eosinophilia that characterize asthma, neither is a potent chemoattractant for eosiniphils (41). On the other hand, the contribution of IL-1 and TNF- to eosinophil recruitment has been demonstrated with the use of IL-1 receptor antagonist proteins and soluble TNF receptors in animal models of eosinophilic airway inflammation; inhibition of either activational pathway significantly decreased allergen- induced eosinophil migration into the airways (2, 41). Together, these observations suggest that effects of IL-1 and TNF- on eosinophil recruitment to the airways are indirect. Indeed, in addition to IL-4 and IL-13, TNF- and IL-1 induce VCAM-1 expression on vascular endothelial cells (44), a key step in the recruitment of eosinophils and mononuclear cells to the airways after allergen exposure.

It is germane to the observation that TNF- and IL-1 upregulate eotaxin production by ASM that mast cellswhich also comprise a source of TNF- and IL-1have been isolated from human ASM tissue (45). TNF- and IL-1 are therefore potentially available to myocytes in vivo, particularly after IgE-mediated triggering of mast cells by allergen. Interestingly, Hakonarson and coworkers (32) have demonstrated that exposure of ASM tissue or cultured cells to serum from individuals with asthma induces the elaboration of IL-1 by ASM. Although the factor(s) responsible for this activity have not been definitively identified, the ability of ASM to produce IL-1 suggests an autocrine mechanism by which eotaxin production could be increased. Within the range of concentrations that we found effective in the upregulation of eotaxin in ASM, TNF- and IL-1 have previously been shown to stimulate the production of eotaxin, RANTES, and MCP-4 by airway epithelial cells (8, 27). The combined effect that we observed between IL-1 and TNF- in the induction of eotaxin mRNA is consistent with results obtained for eotaxin expression in a bronchial epithelial cell line (BEAS-2B cells; see Reference 27).

A number of studies have identified ASM as a source of cytokines and chemokines in vitro (12, 32, 33). In the context of eosinophilic inflammation in asthma, it is pertinent that ASM produces constitutively the eosinophilopoietic agent GM-CSF, and that this expression can be increased with IL-1 or TNF- (13). The induction of RANTES expression by TNF- in human ASM has been reported (12). In addition to attracting eosinophils, RANTES causes the migration of monocyte-macrophages and T lymphocytes in vitro (5). However, when injected in the skin of the rhesus monkey, RANTES is a less effective eosinophil chemoattractant than eotaxin (16). In the absence of cytokine stimulation, eotaxin mRNA expression was easily detected by RT-PCR. However, the baseline of eotaxin mRNA expression was different in the TNF- and IL-1 experiments. This most likely occurred because the experiments were done using different passages and on different dates. Using Northern analysis, previous studies have demonstrated the importance of TNF- and IL-1 in the rapid induction and mobilization of eotaxin mRNA expression (27). After cytokine stimulation, the time that we found eotaxin mRNA expression to be maximal (12 to 24 h) in ASM coincides with the time frame in which increased eosinophils begin appearing in the airways after allergen exposure. Maximal induction of RANTES in ASM was reported at 96 h after cytokine stimulation, suggesting a prominent role for RANTES in sustaining inflammatory cell recruitment over a more prolonged period (12).

We found that both eotaxin and RANTES represented a considerable proportion of ASM-derived eosinophil chemoattratant activity. Similar results were obtained using supernatants from TNF- and IL-1-stimulated ASM, suggesting that like TNF-, IL-1 can also induce RANTES production by ASM. Eosinophil chemotaxis was obtained with highly diluted culture supernatants of stimulated ASM. Several possibilities may account for this observation. First, previous studies of chemotactic response of human eosinophils have been carried out using recombinant, but not natural, eotaxin. Recombinant eotaxin likely has reduced activity due to differences in glycosylation. Second, chemotactic responses are difficult to compare accurately between studies because differences may be related to the source of eosinophils, and may also be highly dependent on small changes in ionic content and pH of the supernatant (46). Third, the elevated polyclonal antibody concentrations (1:10 dilution, data not shown) required for complete blockade of eotaxin activity may reflect the difference in affinity for the eotaxin between CCR-3 and the polyclonal antibody. The neutralizing activity of neutralizing antibodies is limited to the recognition of those few key residues within the region that play a critical role in chemokine binding or in the modulation of chemokine activity (47). The structural features of the eotaxin and anti-eotaxin antibody interaction must be determined before conclusions can be drawn. Interestingly, the blockade of both chemokines caused a decrease in ASM-derived eosinophil chemotaxis of more than 70%. This is in contrast to findings that the eosinophil chemotaxis elicited by asthmatic BAL was inhibited by 25-45% with anti-eotaxin and anti-RANTES antibodies (25). TNF- and IL-1-stimulated ASM appear, therefore, to reduce a more restricted profile of eosinophil chemoattractants than what is present in the lungs of subjects with asthma. The remainder of eosinophil chemotactic activity produced by ASM might be attributed to MIP-1 and IL-8, other eosinophil chemoattractant that have also been shown to be produced by airway myocytes (12, 48). The growing number of cytokines and chemokines that are potentially produced by ASM suggests that these cells may also express other eosinophil chemoattractants in addition to eotaxin, RANTES, MIP-1, and IL-8.

The receptor for eotaxin, CCR3, has been reported to be selectively expressed by human Th2 cells (18). Because CCR3⁺ T cells compose only ~ 1% of peripheral blood T cells, a low level of CCR3 expression and consequently the high contamination by negative cells might explain our observation that anti-eotaxin antibodies had no effect on PBMC chemotaxis induced by supernatant from stimulated ASM. Eosinophils and basophils are known to release a variety of substances that are directly active on ASM. The attraction of T cells to ASM expressing eotaxin may also be important in the pathogenesis of asthma. Activated T lymphocytes have been demonstrated to adhere to TNF--stimulated ASM via integrins and CD44 (31). Such interactions were shown to induced DNA synthesis in ASM, suggesting a role for T cells in airway remodeling. In addition, studies demonstrating the expression of MHC class II on ASM raise the possibility that airway myocytes could act as antigen-presenting cells in asthmatic airways (30). However, the effect of Th2 versus Th1 T lymphocyte interactions with ASM remain to be investigated.

In conclusion, the results of this study show that ASM has the capacity to produce eotaxin, and might thereby contribute to the recruitment of eosinophils, basophils, and Th2 cells in the airways of subjects with asthma.