INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The atmosphere in urban areas is heavily polluted with nitrogen dioxide (NO₂) and suspended particulate matter (SPM) that are derived primarily from vehicle emissions (1, 2). The number of diesel-powered cars has been increasing in Europe and Japan. Diesel-powered vehicles emit more NO₂ and particulates than gasoline-powered cars do. As a result, the proportion of diesel exhaust particles (DEP) to SPM is very high in the urban atmosphere. Suspended particle matter generated outdoors consists of very fine particles and therefore can invade buildings easily (2).

In Japan, the prevalence of allergic rhinitis among schoolchildren is significantly higher in districts that are strongly polluted by NO₂ and SPM than in unpolluted districts (3). However, despite numerous epidemiologic attempts to show the relationship between NO₂ and allergic asthma, no experimental data have provided evidence that NO₂ induces allergic bronchial asthma (4). We recently reported that the intratracheal instillation of DEP in mice induces chronic airway inflammation characterized by the infiltration of eosinophils and lymphocytes, airway hyperresponsiveness, and mucus hypersecretion (7). We also developed a murine model of allergic asthma using the intratracheal instillation of DEP and ovalbumin (OVA). In this model, DEP enhanced airway inflammation, mucus hypersecretion and hyperresponsiveness as indicated by increases in OVA-specific immunoglobulin (Ig) G₁ production and interleukin (IL)-5 expression (8, 9). These observations suggest that DEP in conjunction with allergens may cause allergic asthma in humans. This mechanism may be clinically relevant, especially because levels of allergens are increased in modern buildings that are relatively airtight (10).

The effects of intratracheal instillation of DEP suspension in mice may differ from those of daily diesel exhaust (DE) inhalation. Although the effects of intranasal or intratracheal instillation of DEP suspensions on allergic reactions have been analyzed (7, 11, 12), the effects of DE inhalation on airway inflammation and hyperresponsiveness in mice have not yet been investigated. Therefore, we exposed mice to DE and sensitized them with OVA, then analyzed airway inflammation, airway responsiveness, the production of antigen-specific IgE and IgG₁, and the local expression of cytokines.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Materials

Acetylcholine (ACh), diethyl ether, thimerosal, phenylmethane sulphonyl fluoride, and Tween 20 were purchased from Nacarai Tesque (Kyoto, Japan). Rat anti-mouse IgE, horseradish peroxidase-conjugated streptavidin, bovine serum albumin (BSA), ethylenediamine tetraacetic acid (EDTA), 4-methyl-umbelliferyl--galactoside and OVA (grade V) were obtained from Sigma Chemical Co., Ltd. (St. Louis, MO). Biotinylated rabbit anti-mouse IgG₁ and -D-galactosidase-conjugated streptavidin were purchased from Zymed Laboratories (San Francisco, CA). Leupeptin and pepstatin were purchased from Peptide Institute (Osaka, Japan). Phosphate-buffered saline (PBS, pH 7.4) was obtained from Nissui Pharmaceutical Co., Ltd. (Tokyo, Japan). Diff-Quik staining solution was purchased from International Reagents Co., Ltd. (Kobe, Japan). Rat anti-mouse IgE monoclonal antibody was obtained from Yamasa Shoyu Co., Ltd. (Chiba, Japan). Schiff's reagent was purchased from Merck (Darmstadt, Germany). All other chemicals were of the highest grade available.

Animals

Male C3H/HeN mice (6 wk old) were obtained from Japan Clea Co. (Tokyo, Japan). They were fed a commercial stock diet CE2 (Japan Clea Co.) and water ad libitum. The animals were housed in a chamber maintained at 24 to 26° C with 55 to 75% humidity and a cycle of 14 h light to 10 h dark. The study adhered to the National Institutes of Health guidelines for the use of experimental animals.

Generation of Diesel Exhaust

A 4JB-1-type, light-duty (2,740 ml), four-cylinder diesel engine (Isuzu Automobile Co., Tokyo, Japan) was connected to an Eddy current dynamometer (ECDY; Meidensha, Tokyo, Japan). The engine was operated using standard diesel fuel at a speed of 1,500 rpm under a load of 10 torque (kg/m) (13). Diesel exhaust fumes were diluted with clean air to a constant particle concentration (3 mg DEP/m³). The concentrations of nitric oxide (NO), NO₂, sulfur dioxide (SO₂), and carbon dioxide (CO₂) in the diluted DE were 19.0 ± 2.4, 4.08 ± 0.29, 1.26 ± 0.10, and 3,100 ± 220 parts per million, respectively.

Study Protocol

Eighty C3H/HeN mice were divided into four groups: control animals (air-saline), DE-saline, air-OVA, and DE-OVA. The animals were exposed to clean air or diluted DE in separate chambers (2.2 m³) for 12 h/d (10:00 P.M. to 10:00 A.M.) for 5 wk. After the first 7 d, two of the groups were sensitized by intraperitoneal injection with 1 mg OVA dissolved in 0.5 ml of an aluminum hydroxide (alum) gel suspension (3 mg/ml) in saline solution. After an additional 4 wk, these mice (air-OVA and DE-OVA) were challenged with OVA by exposure for 15 min to an aerosol of 1% OVA in saline administered through an ultrasonic nebulizer (NE-U07; Omron Co., Tokyo, Japan) in a chamber of 2,000 cm³ volume. The two groups of nonsensitized mice (air-saline and DE-saline) were injected with saline instead of OVA-alum and challenged with a saline aerosol. All animals were killed 24 h after the challenge with saline or OVA.

Bronchoalveolar Lavage

After collecting blood from the animals, the tracheae were cannulated. The lungs were lavaged with 1.2 ml sterile saline at 37° C, instilled bilaterally using a syringe. The lavaged fluid was harvested by gentle aspiration. This procedure was repeated three times. For all treatment groups, an average of 90% of the instilled 3.6 ml was retrieved. The fluids from three lavages were combined, cooled to 4° C, and centrifuged at 300 × g for 10 min. Total cell counts were determined on fresh fluid specimens using a hemocytometer. Differential cell counts were assessed on cytologic preparations. The slides were prepared using a Cytospin (Tomy Seiko Co., Ltd., Tokyo, Japan) and stained with Diff-Quik. A total of 300 cells was counted under oil immersion microscopy.

Histologic Evaluation of Eosinophils and Goblet Cells in the Lungs

After exsanguination, the lungs were removed and fixed by intratracheal instillation with 10% neutral phosphate-buffered formalin at a pressure of 20 cm H₂O. After at least 72 h, slices (4-5 mm) of all pulmonary lobes were embedded in paraffin. Sections (3 µm) were prepared and stained with Diff-Quik to quantitate the number of infiltrating eosinophils. The length of the basement membrane of all airways in each specimen was measured by video micrometer (VM-30; Olympus, Tokyo, Japan). The eosinophils below the bronchial epithelium in each sample were counted using a micrometer (AX80; Olympus) under oil immersion. The results were expressed as the number of inflammatory cells per millimeter of basement membrane. To quantitate goblet cells, the sections were stained with periodic acid- Schiff. Stained goblet cells in the bronchial epithelium were counted using the micrometer. The results were expressed as the number of goblet cells per millimeter of basement membrane.

Measurement of Airway Responsiveness

Respiratory resistance (Rrs) was measured as described previously (14), with a minor modification. Briefly, the mice were anesthetized with pentobarbital sodium (50 mg/kg, given intraperitoneally), and a tracheostomy was performed using an 18-gauge cannula. The animals were then mechanically ventilated with a rodent respirator (Model 683; Harvard Apparatus, South Natick, MA) in a plethysmograph box with a pneumotachometer (BUXCO Electronics, Inc., Sharon, CT) at a constant tidal volume (0.3 ml) and a rate of 120 breaths/min. Spontaneous respiration was inhibited by pancuronium bromide (1 mg/kg, injected intramuscularly). The endotracheal pressure, flow, Rrs, and dynamic lung compliance were recorded continuously on a six-channel recorder (BUXCO Electronics). The 4-s average of Rrs and dynamic lung compliance was also recorded. An ACh challenge was performed by inhalation of an ACh solution (0.313 to 10 mg/ml) for 2 min. The solution was aerosolized using an ultrasonic nebulizer (NE-U07; Omron Co.).

Blood Retrieval and Analysis

The mice were anesthetized with diethyl ether. The chest and abdominal walls were opened, and blood was retrieved by cardiac puncture. Plasma was prepared and frozen at 80° C until the assay for OVA-specific IgE and IgG₁.

Ovalbumin-specific IgE Determination

The titer of OVA-specific IgE antibody was measured by IgE-capture ELISA (15). Briefly, microplate wells (Dynatech, Chantilly, VA) were coated with a monoclonal rat anti-mouse IgE antibody for 3 h at 37° C, followed by incubation for 1 h at 37° C in PBS containing 1% BSA and 0.01% thimerosal. Plasma samples were diluted cumulatively with 1% BSA-PBS. After they were washed with PBS containing 0.05% Tween 20 (PBST), diluted plasma samples were added to the wells and incubated overnight at 4° C. After another wash with PBST, biotinylated OVA was added to each well and incubated for 1 h at room temperature. Wells were washed and incubated with -D-galactosidase-conjugated streptavidin for 1 h at room temperature. After the final washing, wells were incubated with 4-methylumbelliferyl--galactoside as the enzyme substrate for 2 h at 37° C. The enzyme reaction was stopped with 0.1 M glycine-NaOH buffer solution (pH 10.3). The fluorescence intensity was read using a microplate fluorescence reader (Fluoroskan Flow Laboratories, Costa Mesa, CA). Each plate included a previously screened plasma standard that contained a high anti-OVA antibody titer. The resulting titers were expressed as a percentage of the standard titer. Cut-off values for antibody-positive plasma were set at twice the mean fluorescence of preimmune plasma.

Ovalbumin-specific IgG₁ Determination

Ovalbumin-specific IgG₁ was measured by solid-phase antigen ELISA. Microplate wells were coated with OVA overnight at 4° C. The wells then were incubated for 1 h at room temperature with 1% BSA-PBS containing 0.01% thimerosal. Plasma samples were diluted with 1% BSA-PBS cumulatively. After washing with PBST, diluted plasma samples were added to the wells and incubated for 1 h at room temperature. After washing again with PBST, the wells were incubated for 1 h at room temperature with biotinylated rabbit anti-mouse IgG₁ antibody, washed, and incubated with horseradish peroxidase-conjugated streptavidin for 1 h at room temperature. The wells then were washed and incubated in the dark with o-phenylenediamine and H₂O₂ for 30 min at room temperature. The enzyme reaction was stopped with 4 N H₂SO₄. The absorbance at 490 nm was determined using a microplate reader (Model 3550; Bio-Rad Laboratories, Hercules, CA). Each plate included a plasma standard that contained a high titer of anti-OVA IgG₁ antibodies. All results were expressed as a percentage of the standard titer. The cut-off value for antibody-positive plasma was set at twice the mean absorbance of preimmune plasma.

Quantitation of Cytokine Levels in Lung-tissue Supernatants

The lungs were removed after exsanguination. They were quickly frozen in liquid nitrogen and stored at 80° C until the assay was performed. Each lung was homogenized in 10 mM potassium phosphate buffer (pH 7.4) containing 0.1 mM EDTA, 0.1 mM phenylmethane sulphonyl fluoride, 1 µM pepstatin, and 2 µM leupeptin. The homogenate then was centrifuged at 105,000 g for 1 h at 4° C. The supernatant was stored at 80° C.

The ELISAs for IL-5, granulocyte/macrophage colony-stimulating factor (GM-CSF), IL-2, interferon gamma (IFN-), IL-10, and tumor necrosis factor alpha (TNF-) were performed using matching antibody pairs (Endogen, Cambridge, MA). The ELISA for IL-4 was performed using matching antibody pairs (Amersham, Buckinghamshire, England) according to the manufacturer's instructions. The second antibodies were conjugated to horseradish peroxidase. Subtractive readings at 550 nm and 450 nm were converted to pg/ml using values obtained from standard curves generated with varying concentrations of recombinant IL-5, IL-4, GM-CSF, IL-2, IFN-, IL-10 and TNF-. The detection limits of the assays were < 5 pg/ml, < 5 pg/ml, < 5 pg/ ml, < 3 pg/ml, < 15 pg/ml, < 12 pg/ml, and < 10 pg/ml, respectively.

Statistical Analysis

Data are expressed as means ± SEM. Differences in the numbers of infiltrating inflammatory cells and goblet cells, airway hyperresponsiveness, levels of cytokine protein, and immunoglobulin titers among groups were determined using analysis of variance (Statview; Abacus Concepts, Inc., Berkeley, CA). If differences among groups were significant (p < 0.05), the Fisher protected least-significant-difference test was used to distinguish between pairs of groups. A level of p < 0.05 was accepted as significant.

The correlation coefficients among the number of inflammatory cells (total cells, macrophages, eosinophils, and neutrophils in bronchoalveolar lavage [BAL] fluid), the levels of OVA-specific antibodies (IgE and IgG₁), and the local levels of cytokines (IL-5, GM-CSF, IL-2, IL-4, IFN-, IL-10 and TNF-) were calculated for each mouse (n = 40).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Effects of Diesel Exhaust Exposure on the Cell Composition of Bronchoalveolar Lavage Fluid after Ovalbumin Challenge

Total cells and differential cell counts in the BAL fluid collected from individual mice were determined (Figure 1). The total number of cells as well as the number of macrophages and neutrophils in the BAL fluid recovered from DE-exposed groups (DE-saline and DE-OVA) were significantly higher than in BAL fluid recovered from the unexposed groups (air-saline and air-OVA). However, no significant differences in the numbers of macrophages and neutrophils were observed between the DE-saline and DE-OVA groups. Eosinophils were observed only in OVA-sensitized groups (air-OVA and DE-OVA). The number of eosinophils was 13 times higher in the DE-OVA than in the air-OVA group (p < 0.0001). Thus, although the ratio of eosinophils to total cells was very small, DE inhalation accompanied by OVA sensitization significantly increased the number of eosinophils in the BAL fluid.

View larger version (34K): [in this window] [in a new window]   Figure 1.   Cellular profile of bronchoalveolar lavage fluid. Mice were exposed to clean air or DE for 5 wk, with or without OVA sensitization. Bronchoalveolar lavage was performed 24 h after the OVA challenge. Total cell counts were determined on fresh BAL fluid, and differential cell counts were assessed using Diff-Quik staining. Results are expressed as means ± SEM of 10 mice per group. *p < 0.05, compared with air-saline group; p < 0.05, compared with DE-saline group; p < 0.05, compared with air-OVA group (Fisher protected least-significant-difference test).

Effects of Diesel Exhaust Exposure on Eosinophilic Infiltration and Goblet Cell Hyperplasia after Ovalbumin Challenge

The number of eosinophils and goblet cells per length of basement membrane of the bronchial epithelium and body weight are summarized in Table 1. After DE and/or OVA treatment, no significant differences in body weight were observed among the four groups. Ovalbumin sensitization alone (air-OVA) caused significant eosinophilic infiltration below the bronchial epithelium, which was markedly enhanced by the combined DE and OVA treatment (DE-OVA). Thus, the number of eosinophils in the DE-OVA group was 80 and 4 times higher than in the air-saline and air-OVA groups, respectively (p < 0.0001). Only a few eosinophils were present in the nonsensitized groups (air-saline and DE-saline).

To evaluate the hypersecretion of mucus in the airways, the lung specimens were stained with periodic acid-Schiff. The number of goblet cells increased significantly in the DE-OVA group, reaching levels 36 times and 6 times higher than in the air-saline and air-OVA groups, respectively (p < 0.0001). The lungs of nonsensitized animals (air-saline and DE-saline) contained only a few goblet cells.

Respiratory Resistance after Diesel Exhaust Exposure and Ovalbumin Challenge

To compare the effects of DE and OVA on airway hyperresponsiveness, we determined Rrs in animals after cumulative ACh inhalation. Although baseline Rrs in the air-OVA group increased, the airway responsiveness to ACh was significantly greater in the DE-OVA group than in the other three groups. To further evaluate the effects of DE exposure on Rrs, we calculated the provocative ACh concentration that caused a 50% increase in Rrs (PC₁₅₀), as shown in Table 2. The decrease in PC₁₅₀ represents an increase in airway responsiveness to ACh, but not an increase in baseline Rrs. In both the DE-OVA and air-OVA groups the PC₁₅₀ values were significantly lower than in the air-saline group (p < 0.005). The PC₁₅₀ value in the DE-OVA group (2.1 mg ACh/ml) was significantly lower than in the other three groups (p < 0.001). These data suggest that DE exposure accompanied by OVA sensitization enhanced airway hyperresponsiveness.

Ovalbumin-specific IgE and IgG₁ Levels in Serum

To examine whether DE affects antigen-specific immunoglobulin production the levels of OVA-specific IgE and IgG₁ in the serum were analyzed 1 d after the challenge with aerosolized OVA (Figure 2). These analyses found that OVA sensitization resulted in elevated titers of OVA-specific IgE, which were enhanced further by DE exposure. Thus, the IgE titer in the DE-OVA group was 2.2 times higher than that in the air-OVA group (p < 0.025). In both the air-saline and DE-saline groups, the titer of OVA-specific IgE was below the detection limit.

View larger version (23K): [in this window] [in a new window]   Figure 2.   Ovalbumin-specific IgE and IgG1 antibody levels in the serum. Mice were exposed to clean air or DE for 5 wk with or without OVA sensitization. Plasma samples were retrieved 24 h after the OVA challenge. Ovalbumin-specific immunoglobulins were analyzed using ELISA. Results are expressed as means ± SEM of 18 mice per group. N.D. = below the detection limit *p < 0.05, compared with air-saline group; p < 0.05, compared with DE- saline group; p < 0.05, compared with air-OVA group (Fisher protected least-significant-difference test).

Similarly, OVA sensitization significantly elevated the levels of OVA-specific IgG₁. Exposure to DE further enhanced these levels; IgG₁ titers in the DE-OVA group were 1.6 times higher than in the air-OVA group (p < 0.025). The IgG₁ titers in both the air-saline and DE-saline groups were below the detection limits. These results indicate that DE had an adjuvant effect on the production of both IgG₁ and IgE.

Local Cytokine Expression

To determine which cytokines were involved in the DE-mediated enhancement of antigen-induced airway inflammation and hyperresponsiveness, we quantitated protein levels of IL-5, IL-4, GM-CSF, IL-2, IFN-, IL-10 and TNF- in lung-tissue supernatants from the four groups (Table 3).

Ovalbumin sensitization alone led to enhancement of IL-5, IL-4, GM-CSF, and IL-2 levels as compared with the air-saline group. Levels of IL-10 and TNF- were reduced. Diesel exhaust alone caused reductions in the concentration of IL-10 and TNF-, compared with the air-saline. The combination of DE and OVA produced higher levels of IL-5 and IL-2 than air-saline. The enhanced IL-5 expression induced by OVA sensitization was increased further by the combined DE and OVA treatment. Specifically, the IL-5 levels in the DE-OVA group were 1.6 times higher than those in the air-OVA group (p < 0.0001). On the other hand, the enhanced GM-CSF and IL-4 expressions caused by OVA sensitization were significantly reduced by DE inhalation (p < 0.05). Diesel exhaust inhalation did not influence expression of IL-2 induced by OVA sensitization.

Analysis of Correlation Coefficients Among the Examined Factors

To determine which cytokines were involved in DE-mediated enhancement of airway inflammation, immunoglobulin production and its expression, the correlation coefficients among the levels of local cytokines, the levels of OVA-specific antibodies, and the number of inflammatory cells were calculated for each mouse. These analyses showed that the amounts of IgE, IgG₁, and IL-2 were significantly correlated with each other (Figure 3). The amounts of IL-2 and the numbers of eosinophils also were moderately correlated with the IL-5 levels. The correlation coefficients between IL-2 and IgE, IL-2 and IgG₁, IL-2 and IL-5, and IL-5 and eosinophils were 0.648, 0.657, 0.762, and 0.663, respectively (p < 0.0001, n = 40). Although IL-10 levels were correlated with TNF- levels (r = 0.621, p < 0.0001, n = 40), they were not correlated with the other factors. Finally, no correlations existed among the amounts of GM-CSF, IL-4, or INF- and the other factors examined.

View larger version (22K): [in this window] [in a new window]   Figure 3.   Correlations among examined pathological factors  Air-saline (open circle), DE-saline (closed triangle), air-OVA (open square), and DE-OVA (closed square). The correlations among the numbers of eosinophils in BAL fluid (BALF), OVA-specific antibodies (IgE and IgG1), and local cytokine expression (IL-5 and IL-2) were calculated for each mouse (n = 40).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

To elucidate the involvement of DE in allergic airway inflammation and hyperresponsiveness, we exposed mice to DE for 5 wk accompanied by sensitization to OVA. Our results indicated that exposure to DE enhanced airway inflammation and was associated with a marked infiltration of eosinophils. The recruitment of eosinophils was accompanied by the hyperplasia of goblet cells in the bronchial epithelium. Exposure to DE also increased airway hyperresponsiveness, production of OVA-specific IgE and IgG₁, and the expression of IL-5 in the murine lungs induced by antigen sensitization. In contrast, DE inhalation alone induced infiltration of macrophages and neutrophils in lungs but not eosinophilic inflammation, hyperplasia of goblet cells, and airway hyperresponsiveness.

To our knowledge, this is the first report to show that the inhalation of DE enhances allergic airway inflammation, hyperresponsiveness, and goblet cell hyperplasia in mice. Previous experiments have used DEP suspensions instead of DE fumes. Although the intratracheal or intranasal administration of DEP is useful for estimating the effects of DEP alone (7, 11, 12), it does not adequately represent the exposure caused by the daily inhalation of DE fumes. Our inhalation experiments, therefore, are more suitable for evaluating DE's effects on human health. However, the exposed DEP concentration in the present study (3 mg/m³) is higher than that in an urban atmosphere (< 200 µg/m³). Therefore, we have to investigate whether a much lower concentration of DE enhances these allergic reactions.

We previously showed that the intratracheal administration of DEP with OVA challenge to C3H/HeN mice induced airway inflammation, hyperresponsiveness, and mucus hypersecretion (9). The enhanced infiltration of eosinophils was associated with an increased production of antigen-specific IgG₁, but not with IgE. These pathogenic features can be explained by the effects of DEP administration on the expression of IL-5 and IL-2 in murine lung tissue (8, 9).

Either NO₂ or SO₂ alone induces airway hyperreactivity, but not IgE production and airway inflammation, in animals and humans (6). The effects of DE exposure described in this study were similar to those of intratracheal instillation of DEP. Therefore, it is reasonable to infer that particulates in DE, rather than the fairly high concentrations of gaseous components, mainly enhanced allergic reactions in the present study. However, some of the effects of DE inhalation and intratracheal DEP administration differed. For example, the DE inhalation-induced hyperplasia of goblet cells and increases in cytokine expression were less than those observed after the intratracheal instillation of DEP. The OVA-specific IgE production was significantly enhanced after DE inhalation, but not after DEP administration, even though the same protocol for sensitization was used for both approaches. Finally, the levels of OVA-specific IgG₁ were higher after DE inhalation than after DEP administration (9). These effects may be due to the gaseous components in DE and/or the route and amounts of particles administered; the amount of DEP inhaled in this study was calculated to be about 200 µg/wk/mouse, which significantly exceeded the exposure achieved with the intratracheal instillation of DEP (25 µg/wk/mouse) (9).

The induction of antibody production after DEP administration has been reported previously. For example, intranasal instillation of DEP and antigen resulted in increased levels of antigen-specific IgE in murine (11) and human sera (12). Furthermore, DEP and the polyaromatic-hydrocarbon fraction derived from DEP enhanced IgE production from purified human B cells in vitro (16, 17). In contrast, the intratracheal instillation of DEP and OVA enhances IgG₁, but not IgE, production in mice (8, 9). Although these results seem to be discrepant, the type of immunoglobulin produced may depend on the administration route of DEP. Although we cannot indicate whether DE inhalation actually affects sensitization or immunoglobulin production, DE inhalation enhanced both IgE and IgG₁ productions, and these may induce allergic reactions. Diesel exhaust inhalation enhances infiltration of macrophages, i.e., antigen-presenting cells; it can also enhance OVA sensitization and allergic reactions.

The mechanisms underlying allergic asthma remain unknown because it involves numerous factors, including antibody production, chemical mediators, cytokine expression, and activation of different cells. The airway inflammation of asthma is unique in that the airway wall is infiltrated by Th2 cells (18), eosinophils, and mast cells (19, 20). Each of these cells is thought to contribute to the physiologic changes that characterize asthma (21). In particular, allergen-specific IgE is believed to play a central role in the hypersensitivity reactions via mast cells. Such IgE-mediated reactions are followed by chronic inflammation leading to increased airway hyperresponsiveness. However, evidence suggests the existence of alternative or additional pathways of hypersensitivity reactions. First, immediate hypersensitivity and airway hyperresponsiveness were induced by the administration of OVA-specific IgE or IgG₁, but not by IgG_2a or IgG₃ (22). Second, allergen-specific IgA, IgE, and IgG₁ antibodies reportedly contribute to antigen-specific eosinophil degranulation via FcR, FcRI, FcRII and FcRII on the surface of the eosinophil (23, 24). Together with antigen, IgG₁ acts as a strong agonist for eosinophil degranulation in vitro (25, 26). These observations suggest that both IgG and IgE can contribute to allergic airway inflammation and airway hyperresponsiveness via eosinophils.

On the other hand, the bronchial hyperreactivity and eosinophilic inflammation caused by exposure to the allergen occur even in mice that are deficient in IgE, mast cells, or B cells (27). Aeroallergen-induced eosinophilic inflammation, airway damage, and airway hyperreactivity in mice can occur independently of IL-4 and allergen-specific immunoglobulin (31). In these instances of eosinophilic pathological changes, IL-5 plays a central role even in the absence of immunoglobulin production (30, 31).

In the present study, the levels of IgG₁, IgE, IL-5, and IL-2 were significantly correlated with each other and increased markedly after DE inhalation and antigen challenge. The high correlation among these parameters suggests that IL-2 and IL-5 play important roles in enhancing airway inflammation and hyperresponsiveness. The elevated IL-2 and IL-5 levels observed in this study may promote the production of OVA-specific immunoglobulin (32) and eosinophil activation (35), thereby causing airway inflammation with epithelial damage and airway hyperresponsiveness.

Although our results demonstrate that DE inhalation enhances IL-5 expression, we cannot conclude whether the mechanisms underlying hypersensitivity and epithelial damage by DE and OVA are involved with immunoglobulin production and/or eosinophils. To clarify the mechanisms underlying air-pollutant-induced allergic asthma, further studies need to examine the activation and degranulation of eosinophils and epithelial damage in more detail.