Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

In the last decades, both the incidence of allergic bronchial asthma (BA) and the level of air pollution caused by ozone have increased (1). The question of whether there exists a causal relationship between the two events still remains unsolved. In most Western industrialized areas, the average concentrations of ozone over a 1-yr span range from 40 to 80 µg/m³. Maximal levels in the summer season sustained over days or weeks can exceed 200 µg/m³ (2) and hourly means can surpass even 350 µg/m³ (3). The exposure of healthy individuals to ozone leads to: (1) irritative cough and substernal chest pain on inspiration; (2) decrements in forced vital capacity and forced expiratory volume at 1 s (FEV₁) values; and (3) neutrophilic inflammation of the airway mucosa accompanied by increased levels of inflammatory mediators and proteins in bronchoalveolar lavage (BAL) (4). Elevated levels of proinflammatory interleukins (IL-6, IL-8) and arachidonic acid metabolites (prostaglandin [PG] E₂, PGF₂, leukotriene [LT] B₄, and thromboxane B₂) have been reported (8, 10, 11). Further, the inflammation caused by ozone exposure was found in nasal epithelial cells as indicated by increased levels of IL-1, IL-6, tumor necrosis factor-, and intercellular adhesion molecule expression (12). The effects of acute and chronic ozone exposure on anatomy and cellular constituents of the lungs and the airways have been studied in detail in animal models (for review, see 13). For example, a simulated urban profile of ambient ozone for 78 wk in rats caused epithelial inflammation, interstitial fibrosis, and bronchiolar epithelial cell injury (14). Ozone has adverse effects on membranes. It increases their permeability for proteins and thus may enable allergens to penetrate into deeper tissue areas more easily.

BA is characterized by a unique type of airway inflammation accompanied by the influx of T-helper (Th)2 T cells, eosinophils, and the development of airway hyperresponsiveness (AHR). Ozone, when applied for short terms, has comparable effects: it also increases airway responsiveness (AR) and in asthmatics, an eosinophilic inflammation is observed (15) suggesting an augmented (Th)2 response. Data for chronic exposures, however, are limited. In humans, the effects of chronic ozone exposure can be delineated from epidemiologic studies. It is an important observation that increased ambient ozone levels have been associated with increased asthma attacks and morbidity (16). Animal models offer the potential advantage to assess the relationship between airway inflammation, AHR, and ozone exposure under controlled environmental conditions at the molecular level. To date, only limited data are available addressing this issue. In monkeys, ozone (2,000 µg/m³) enhanced the development of allergy to inhaled platinum (17). A long-term experiment over 4 wk with repeated exposures to ozone in concentrations between 260 and 480 µg/m³ (total ozone exposure, 384 h) enhanced the allergic response in an anaphylactic shock model in ovalbumin (OVA)-allergic mice (18). These data lead to the hypothesis that chronic ozone exposure may shift the immune system toward a Th2 response (19) and aggravates asthma (1, 20).

The present study was carried out to analyze the effects of intermittent ozone exposures continued over 4 wk (180 to 500 µg/m³, 3 times weekly, 4 h each) on the development of allergic immune responses after OVA-aerosol exposure. Typical indicators of allergic sensitization (immediate cutaneous hypersensitivity [ICHS] reaction, production of allergen-specific antibodies, development of increased AR) and the pattern of airway inflammation (BAL fluid [BALF] content of inflammatory cells, IL-4, IL-5, and interferon [IFN]-) were assessed in two mouse strains (BALB/c and C57BL/6) differing in propensity to develop allergic reactions. The data presented here indicate ozone has a potential priming effect on allergic immune responses.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals

Pathogen-free female BALB/c and C57BL/6 mice (Bundesinstitut für gesundheitlichen Verbraucherschutz und Veterinärmedizin, Berlin, Germany), aged 6 to 8 wk and weighing 18 to 22 g, were used. The animals were housed under standard conditions, fed an OVA-free diet, and supplied with water ad libitum. The breathing parameter values of spontaneously breathing BALB/c mice were determined under standard conditions at room air and temperature.

Exposure Chamber

A chamber 700 × 700 × 400 mm (height × width × depth) made of safety glass with an inlet on the top and an outlet an the bottom for ozone and air supply was used for the exposures. The gases were pumped with 600 liters/h through the chamber. Mice were exposed to room air or to different ozone concentrations (180, 250, and 500 µg/m³) for 4 h three times/wk over a period of 4 wk.

Ozone Generator

Ozone was produced by a ComEDHF1 generator (Anseros, Tüb-ingen, Germany) using high voltage up to 10 kV. The ozone concentration was continuously monitored by an Ozomat MP6080 photometric analyzer (Anseros). The signal was fed into a proportional integral differential (PID) controller, Peripheral PID (Anseros), that analyzed the temporal changes in ozone concentration and regulated the generator to stable ozone concentrations.

Generation of Aerosols

Aerosols were generated with a common jet nebulizer (Pari-Boy; Pari-Werke, Starnberg, Germany). The generator (Pari-Master) produced a pressure of 1.4 to 1.5 bar and an airflow of 20 liters/min. The median mass diameter was 3.6 µm and the mass portion with a size below 5 µm was 65%. Methacholine (MCh) was delivered with the Pari-Boy. Each concentration (0 to 30 mg/ml in phosphate-buffered saline [PBS]) was aerosolized for 5 min. OVA aerosol was generated from an OVA solution (10 mg/ml in PBS, Grade V; Sigma, Deisenhofen, Germany) and applied for 20 min each day.

Sensitization Protocol

Mice were sensitized to OVA via the airways by aerosol administration five times/wk (Figure 1). The exposures were carried out over a period of 4 wk as described earlier.

View larger version (14K): [in this window] [in a new window]   Figure 1.   Schematic drawing of the sensitization protocol. Animals were exposed either to ozone (180 to 500 µg/m3) three times/wk for 4 h or to OVA-aerosol five times/wk for 20 min or to both agents.

Measurement of Serum Immunoglobulins

Total immunoglobulin (Ig) E and anti-OVA IgE, IgG₁, and IgG_2a antibody concentrations were measured by enzyme-linked immunosorbent assay (ELISA) as previously described (21, 22). Antimouse IgE, IgG₁, and IgG_2a monoclonal antibodies were obtained from Pharmingen, Hamburg, Germany.

Assessment of Leukocyte Distribution in BALF

At 24 h after the last aerosol exposure, the animals were killed, the tracheas were cannulated, and BAL was performed by two repeated lavages with 0.8 ml ice-cold PBS each. BALFs of each animal were pooled and the recovered volume and the total cell number were determined. Mean recovery volume was 1.4 ± 0.2 ml and no significant difference was detected between the study groups. Cytospins were prepared for each sample by centrifugation of 50 µl BALF (100 × g, 5 min). After fixation, cytospins were stained with Diff-Quik (Baxter Dade, Dudingen, Switzerland). Cells were classified as either macrophages, neutrophils, eosinophils, or lymphocytes, using standard morphologic criteria. Cell-free lavage fluids were stored at 20°C until analysis.

Measurement of Cytokines in BALFs

The cytokines IL-4, IL-5, and IFN- were measured by ELISA as previously described (21). Antibodies were obtained from Pharmingen. Sensitivities (in pg/ml) were 10 for IL-4, 30 for IL-5, 10 for IL-2, and 100 for IFN-.

AR

AR was assessed by head-out body plethysmography as described previously (23). Briefly, four mice were placed in four body plethysmographs attached to an exposure chamber (Crown Glass, Somerville, NJ). The airflow was measured with a PTM 378/1.2 pneumotachograph (Hugo Sachs Electronics, March-Hugstetten, Germany) and an 8-T2 differential pressure transducer (Gaeltec, Dunvegan, UK). The airflow was analyzed in response to various concentrations of MCh (0 to 30 mg/ml, 1 min) delivered by a jet nebulizer (Pari-Boy; Pari-Werke). The concentration of MCh causing a 50% reduction in midexpiratory airflow (MCh₅₀) was determined.

Assessment of ICHS

Intracutaneous skin testing was performed as previously described (24). Briefly, abdominal skin was shaved, and 50 µl of test solution was injected intradermally. To improve visibility of the intradermal reaction, 100 µl of a 0.5% Evans blue solution was injected intravenously 15 min before skin testing. The following skin-test solutions were used: PBS as a negative control; Compound 48/80 (5 mg/ml; Sigma) as a positive control; and OVA at 0.5, 5, 50, and 100 µg/ml (Sigma). After 15 min of development, the blue flair reactions were assessed as described (24). The skin tests were classified according to their intensity into three classes for each concentration applied. Because four concentrations were injected, the maximal score for each mouse was 12 points when the animal reacted at all concentrations in the highest possible degree.

Assessment of Cysteinyl-LTs  C₄, D₄, and E₄ in BALF

A commercially available ELISA (Amersham Pharmacia, Freiburg, Germany) was used to measure cysteinyl-LTs C₄, D₄, and E₄. The assays were performed as recommended by the manufacturer.

Statistical Analysis

Results are presented as mean values ± standard error of the mean (SEM), unless otherwise stated. One-way analysis of variance was used to determine the levels of difference between animal groups.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Ozone Induces Th2-Like Responses in BALB/c Mice

BALB/c mice are a genetically predisposed "Th2-high responder" strain. Chronic ozone exposures (over 4 wk, 3 times weekly, 4 h each d) stimulated total IgE production in a dose-dependent manner (Table 1). The analysis of BALF obtained 24 h after the last exposure showed increased levels in the production of IL-4 and IL-5. IFN-, a prototypic Th1 cytokine, remained unchanged. Assessment of the cellular infiltrate revealed significant increases at 500 µg/m³ ozone (Table 2). Most pronounced increases were observed among eosinophils (P < 0.05), lymphocytes (P < 0.05), and neutrophils (P < 0.01). This protocol did not alter the level of AR (11.3 ± 0.5 versus 11.2 ± 0.5 mg/ml MCh; Figure 2) as assessed by head-out body plethysmography and expressed as MCh₅₀.

View larger version (16K): [in this window] [in a new window]   Figure 2.   AR measured by head-out body plethysmography 48 h after last ozone exposure and 24 h after last OVA exposure in BALB/c mice. Open bars, ozone exposure; filled bars, OVA and ozone exposure. Ordinate, MCh50. Abscissa, ozone concentration (µg/m3). Means ± SEM; n = 8 mice per group. Significant difference indicated as **P < 0.01.

Ozone Enhances an Allergic Phenotype in BALB/c Mice

To analyze the effect of ozone on the development of allergic sensitization, BALB/c mice were exposed to ozone and OVA-aerosol. OVA-aerosol exposure alone induced weak allergic immune responses to OVA as indicated by weak positive skin-test responses (ICHS, 2.8 ± 2.3 U, maximal response = 12 U per mouse; Figure 3A) and induction of low levels of allergen-specific (OVA) IgG₁ and IgG_2a antibody production (4.2 LU/ml) (Figures 3B and 3C), as well as moderate increases in IL-5 levels in BALFs (from 43 ± 9 to 176 ± 34 pg/ml; Table 3). The airways were infiltrated by few numbers of lymphocytes, neutrophils, and eosinophils (Table 2). Further, an insignificant trend toward increased AR was apparent (11.3 ± 0.5 versus 9.4 ± 2.4 mg/ml MCh). In contrast, the combination of ozone and OVA-aerosol exposure resulted in augmented immune responses to OVA. At 500 µg/m³ of ozone, the ICHS score to OVA was increased more than 3-fold as compared with OVA-aerosol exposure alone (73 versus 23% of maximal response; Figure 3A). An increase was detected in anti-OVA IgG₁ antibody titers (4.2 versus 7.6 LU/ml; Figure 3B). Synergistic effects were also observed on the recruitment of lymphocytes (2.4 ± 0.35 versus 7.9 ± 0.75 cells/ml BALF × 10³) and eosinophils (1.7 ± 0.57 versus 9.1 ± 0.75 cells/ml BALF × 10³) into the airways (Table 2). These effects were present only at an ozone concentration of 500 µg/m³. In contrast, the numbers of neutrophils and macrophages remained unchanged. In terms of IL-4 and IL-5 production, no additive effect was detected. An increased AR, as indicated by a higher sensitivity to MCh challenge (11.3 versus 7.2 mg/ml MCh), was detected in mice sensitized to OVA and simultaneously exposed to ozone (500 µg/m³; Figure 2).

View larger version (13K): [in this window] [in a new window]   Figure 3.   Enhanced sensitization of BALB/c by ozone. Nonaerosolized (NIL), or with OVA-aerosol-sensitized (OVA), or with OVA aerosol-sensitized and ozone-exposed mice were analyzed 48 h after the last ozone exposure and 24 h after the last OVA exposure. (A) ICHS reaction. OVA (0.5, 5, 50, and 100 µg/ml) was injected intradermally into BALB/c mice. Maximal score was 12 points per animal. Ordinate, ICHS score. Significant differences indicated as *P < 0.05; ***P < 0.01; n = 8 mice per group. (B) Serum concentration of anti- OVA IgG1 (LU/ml BAL); abscissa, Nil, untreated mice, ozone concentration (µg/ m3). Means ± SEM; n = 8-12 mice per group. Significant difference indicated as *P < 0.05. (C) Serum concentration anti-OVA IgG2a (LU/ml BAL); abscissa, Nil, untreated mice, ozone concentration (µg/ m3). Means ± SEM; n = 8-12 mice per group. Differences were not significant.

It has been previously demonstrated that LTs are able to mediate acute bronchoconstriction. Therefore the total content of cysteinyl-LTs (LTC₄, LTD₄, and LTE₄) in BALF was analyzed by ELISA. Neither OVA nor ozone alone changed the cysteinyl-LT level in BALF. However, the combined exposure to ozone induced a dose-dependent increase in cysteinyl-LTs parallel to the increase in AHR (Figure 4; P < 0.05 versus treatment without ozone).

View larger version (20K): [in this window] [in a new window]   Figure 4.   Content of cysteinyl-LTs C4/D4/E4 in BALF measured by ELISA 48 h after last ozone exposure and 24 h after last OVA exposure in BALB/c mice. Open bars, ozone exposure; filled bars, OVA and ozone exposure. Ordinate, cysteinyl-LTs (pg/ml). Abscissa, ozone concentration (µg/m3). Means ± SEM; n = 8 mice per group. Significant difference indicated as *P < 0.05.

Induction of an Allergic Phenotype in C57BL/6 Mice

To assess whether the effects of ozone were restricted to "Th2-high responder" mice, "Th2-low responder" C57BL/6 animals were treated according to the previously described protocols. On the basis of the results in BALB/c mice, C57BL/6 animals were exposed to 500 µg/m³ ozone. This dose had no effect on IL-4, IL-5, or IFN- levels in BALFs (data not shown). More than 95% of cells in BALF were marcophages, and no additional influx of eosinophils, lymphocytes, or neutrophils was detected (data not shown). In contrast to BALB/c, C57BL/6 mice showed lower baseline total IgE levels, which were not significantly elevated in ozone-exposed mice (27 ± 3 versus 35 ± 6 ng/ml, mean ± SEM, before and after ozone, respectively).

In C57BL/6 mice, OVA-aerosol exposure triggered the production of high anti-OVA IgG_2a (390 LU/ml) and low anti-OVA IgG₁ antibody titers (1.0 LU/ml; Figures 5B and 5C). Positive skin-test reactions to OVA were not detectable (Figure 5A). When ozone and OVA were applied simultaneously, the IgG response pattern was shifted toward higher anti-OVA IgG₁ (3.0 LU/ml) and lower anti-OVA IgG_2a (140 LU/ml) antibody production. As a consequence, positive ICHS reactions (40% of maximal response) were observed only in OVA mice simultaneously exposed to ozone. In terms of airway inflammation as determined by IL-4, IL-5, and IFN- levels and cell differentiation in BALF, no additional effects were observed after ozone exposure. The AR was not significantly affected by either of the treatments.

View larger version (10K): [in this window] [in a new window]   Figure 5.   Induction of sensitization to OVA by ozone in C57BL/6. Nonaerosolized mice (NIL), or with OVA-aerosol-sensitized (OVA) or with OVA-aerosol-sensitized and ozone (500 µg/m3)-exposed (OVA + ozone) mice were analyzed 48 h after the last ozone exposure and 24 h after the last OVA exposure. (A) ICHS reaction. OVA (0.5, 5, 50, and 100 µg/ml) was injected intradermally into BALB/c mice. Maximal score was 12 points per animal. Ordinate, ICHS score. Significant differences indicated as *P < 0.05; ***P < 0.01; n = 8 mice per group. (B) Serum concentration anti-OVA IgG1 (LU/ ml BAL); abscissa, Nil, untreated mice, ozone concentration (µg/m3). Means ± SEM; n = 8-12 mice per group. Significant difference indicated as *P < 0.05. (C) Serum concentration anti-OVA IgG2a (LU/ml BAL); abscissa, Nil, untreated mice, ozone concentration (µg/m3). Means ± SEM; n = 8-12 mice per group. Differences were not significant.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The results presented here reveal the induction of a Th2-like immune response in BALB/c mice by ozone, as indicated by a dose-dependent increase in the levels of total serum IgE, IL-4, and IL-5 levels as well as eosinophil and lymphocyte numbers in BALFs. These results support the hypothesis that ozone directs the immune response toward a Th2 pattern under certain conditions (19, 25). To assess the effect of ozone exposure on the development of allergic immune responses, BALB/c mice were exposed to OVA-aerosol. In contrast to protocols of intraperitoneal sensitization by OVA adsorbed to Al(OH)₃ (21, 26), the protocol of the present study induced only low levels of allergen-specific antibody production and minimal signs of airway inflammation that did not result in the induction of increased AR. When such OVA-sensitized mice were additionally exposed to ozone, the allergen-specific Th2 immune response was enhanced. In the IgE responder strain BALB/c, ozone augmented the allergic response to OVA and induced increased AR. In contrast, in the "IgE-low responder" strain C57BL/6, ozone did not affect OVA-induced airway inflammation or airway reactivity, but rather induced the development of positive skin-test responses to OVA, most likely due to a shift in the anti-OVA IgG₁/IgG_2a antibody profile.

Similar effects of ozone on the enhancement of allergic reaction have been described. Increased allergic responses due to ozone inhalations (260 to 480 µg/m³) were found in an anaphlylatic shock model of OVA-allergic mice (18). However, these authors did not investigate the underlying immunologic mechanisms. Biagini and colleagues observed increases in skin-test sensitivity in a monkey model of occupational allergy toward inhaled platinum after high ozone exposures (2,000 µg/m³, 6 h/d, 5 d/wk for 12 wk [17, 27]). The combined ozone and platinum exposure did not affect baseline pulmonary function, but increased the specific (platinum) and nonspecific MCh bronchial responsiveness.

Similarly, in the present study a combined ozone and OVA exposure increased AR in BALB/c mice. Only under these conditions was increased AR detected. This effect was small but significant. A hallmark of allergic reactions is the infiltration of eosinophils at the site of inflammation. This influx of eosinophils is probably due to increased local production of chemotactic and survival factors such as IL-5 as observed in BALFs. It is likely that the development of increased AR is dependent on the degree of inflammation, which was further augmented by the combined exposure to ozone and OVA and might thus surpass a certain trigger level for the induction of increased AR. Further, it is likely that elevated cysteinyl-LT levels contribute to the development of increased AR. Cysteinyl-LTs are potent mediators of bronchoconstriction, vascular and nonvascular smooth-muscle constriction, increased vascular permeability, and epithelial mucous secretion (28). It is widely considered that cysteinyl-LTs are important mediators in bronchial asthma (29). Increased LT levels have been found in human BALF after ozone exposure (30). Sources for LT production include eosinophils, mast cells, and macrophages. It is likely that in our model these cell types served as cellular sources. Under certain conditions, mast cells produced and released IL-4. IL-4 plays an important role to direct T-cell effector responses toward Th-2. In our model of enhanced sensitization to OVA by ozone, elevated levels of IL-4, Igs, and LTs, therefore, might serve as prerequisites for the development of increased AR. However, other mediatorse.g., the growth factor granulocyte macrophage colony-stimulating factor (31) and prostanoids (32, 33)are produced after ozone exposure by mast cells, bronchial epithelial cells, and alveolar macrophages, and might also affect AR.

In our study ozone was administered repetitively to closely simulate the natural situation of high ozone levels on sunny days during the summer season. The concentrations used lie within the range observed outdoors in summer (> 350 µg/m³, hourly mean [3]). The outcome of ozone effects, however, is heavily dependent on the activity status of the individual examined. In experiments performed with resting subjects involving 1 to 2 h of exposure to levels as high as 800 µg/m³, only minor effects were observed (34). In contrast, exposure to ozone concentrations ranging from 240 to 800 µg/m³ for 2 h while performing intermittent heavy exercise impaired lung functions, including FEV₁, in a dose-dependent manner (35).

Taken together, the data in this study demonstrate that ozone is a modulator of the T-cell effector response toward the Th2 phenotype. This effect on allergen-specific immune responses was observed in both IgE-high and IgE-low responder strains. The effective concentrations applied are close to ambient concentration. Ozone, therefore, may be a candidate contributing to the increase in prevalence and incidence of allergic diseases during the last decades.