INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Epidemiologic studies have suggested that asthmatics are more sensitive than healthy persons to the effects of particulate matter (PM) air pollution. Levels of respirable PM₁₀ (less than or equal to 10 µm mass median aerodynamic diameter [MMAD]) below current air quality standards have been positively correlated with the risk of hospitalization because of asthma (1). PM₁₀ is also associated with a decline in peak expiratory flow rate and increased respiratory symptoms such as cough and wheezing in asthmatic children (4). The chemical and physical properties of PM responsible for aggravation of asthmatic symptoms and the biologic mechanisms involved have not been elucidated.

In contrast to coarse PM, which is derived primarily from breakdown and resuspension of crustal materials, the fine fraction of ambient PM (< 2.5 µm MMAD) is mainly generated from secondary transformation products of NO_x and SO₂ or the combustion of wood and fossil fuels (7). Significant quantities of water-soluble transition metals may be present in fine ambient PM (8), and these metals may affect human health. Metal-containing PM have been recovered from lungs (9) and bronchoalveolar lavage fluid (10) of human subjects. In addition, PM₁₀ and airborne iron were associated with respiratory symptoms and decreased pulmonary function in adults living near a steel factory (11). Recent toxicologic studies using residual oil fly ash (ROFA), a fugitive fine PM sample with a high content of bioavailable transition metals including V, Ni, and Fe, have demonstrated that these metals are associated with lung injury in both healthy animals and animal models of cardiopulmonary injury (12). We have recently demonstrated that metal composition of ROFA is critical in the development of airway hyperreactivity to acetylcholine challenge in healthy rats (17). Although the metal content of ROFA is significantly greater than the content of typical fine ambient PM, it has been used in these studies as a surrogate particle to investigate metal-related mechanisms of responses to PM in animal models. These studies implicate metals as toxicologically important constituents of particulate matter.

In the present study, pulmonary physiologic and inflammatory effects of exposure to ROFA were examined in a mouse model of allergic airways disease with features similar to human allergic asthma. Balb/c strain mice were selected because they are typically used in studies of allergic responses to ovalbumin and because they are high IgE responders (18, 19). We hypothesized that ROFA exposure interacts with allergen challenge in sensitized mice, resulting in greater than additive physiologic and inflammatory responses. Our objectives were to examine the time course of responses after challenge and exposure, and to examine mediators that may be relevant to the observed physiologic responses. The results from this study show that several physiologic and inflammatory responses to ROFA in allergic mice are greater than additive, indicating an interaction between these treatments. An early increase in T-helper lymphocyte class 2 (Th2) proallergic cytokines may spur the development of these responses.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animals

Female Balb/cJ mice, 7 wk old on arrival and weighing 18 to 21 g, were obtained from Jackson Laboratories (Bar Harbor, ME). Mice were provided with Prolab RMH-3000 meal (PMI Feeds, St. Louis, MO) and tap water ad libitum. Animals were maintained on a 12-h light/dark cycle at 22 ± 2° C and 50 ± 10% relative humidity in an AAALAC-approved facility, and held for a minimum of 4 d before treatment. Protocols used in this study were reviewed and approved by the EPA Institutional Animal Care and Use Committee, and were conducted using national guidelines for the care and protection of animals.

Allergen Sensitization and Challenge and Exposure to Residual Oil Fly Ash

Mice were sensitized with 10 mg ovalbumin (OVA; Grade V, Sigma Chemical, St. Louis, MO) in 0.3 ml aluminum hyroxide gel adjuvant (Alhydrogel; Accurate, Westbury, NY) injected intraperitoneally on two consecutive days. Control mice received Alhydrogel only. Two to 3 wk later, all mice were challenged with an aerosol of 1% ovalbumin in sterile saline for 1 h in a wire rack inside a 135 L chamber. The solution was nebulized using three jets of a six jet atomizer (Model 9306; TSI, Minneapolis, MN) at 8 L/min/jet, and house air was added to make up a total flow rate of 40 L/min (18 exchanges/h). The following data were collected over the course of these challenges (mean ± SD): temperature, 73 ± 2° C; relative humidity, 96 ± 3%; aerosol concentration, 158 ± 3 mg/m³. Mice that were sensitized and challenged with ovalbumin were defined as allergic, whereas mice that were only challenged with ovalbumin were defined as nonallergic. Allergic status was confirmed by analysis of serum IgE levels (see below).

Residual oil fly ash (ROFA) was collected from the flue gas of a power plant burning a low sulfur (1%) no. 6 residual oil downstream from a 2.5-µm cutoff cyclone, and therefore represents a fugitive fine emission source PM sample that may be released into the environment (13, 17). Endotoxin was not detected in this sample of ROFA (17). One to 3 h after ovalbumin challenge, mice were anesthetized in a 2.7 L Plexiglas chamber with methoxyflurane vapor (Mallinckrodt, Mundelein, IL). Anesthetized mice were intratracheally instilled with sterile saline vehicle or ROFA in saline using a 100-µl syringe and round-tip needle (dose = 3 mg/kg, ~ 50 µl volume; ~ 60 µg ROFA instilled per mouse) as previously described (20). This dose does not represent exposure to ambient levels of PM, but is relevant for occupational exposures and useful for understanding potential mechanisms of responses to PM in allergic asthma. Saline was used as a vehicle control since the saline-soluble portion of ROFA particles (94% by weight) has been shown to mediate toxic effects in previous studies (13). After 1, 3, 8, or 15 d of recovery, airway responsiveness to intravenously administered methacholine (MCh) challenge was assessed in each mouse. After these measurements, serum samples were collected for IgE analysis, and bronchoalveolar lavage (BAL) was performed to collect and analyze cells, proteins, enzymes, and cytokines.

Airway Responsiveness Measurements

Mice were anesthetized intraperitoneally with urethane (1.5 g/kg) and tracheostomized using a custom-built cannula (0.89 mm I.D., 1.27 mm O.D.) with ports for inspiration, expiration, and monitoring of airflow and pressure, similar to a previously described design (21). Mice were ventilated with a constant flow of air (0.833 ml/s) regulated by a mass flow controller (FC-260; Tylan, Carson, CA). Small pneumatic solenoids (NVZ 110-6GZ-M5; SMC Pneumatics, Indianapolis, IN) controlled inspiration (180 ms; tidal volume = 0.15 ml) and allowed passive expiration (220 ms; f = 150 breaths/min). Spontaneous respiration was inhibited by intraperitoneal administration of succinylcholine chloride (15 mg/kg). Body temperature was maintained at 37° C using an isothermal heating pad (Braintree Scientific, Braintree, MA), and heart rate and waveform were monitored (SRA-200; Micro-Med, Louisville, KY). A single point pressure transducer (Ailtech, Deerfield, IL) was connected to one port of the tracheal cannula and a pneumotachograph (SCXL004DN; SenSym, Milpitas, CA) was connected to two ports closer to the point of insertion in the trachea to provide airway pressure and flow signals, respectively. A custom-written software program (Mouse Reactivity System; U.S. EPA, Research Triangle Park, NC) was used to control ventilation and calculate respiratory indices. The flow signal was integrated to obtain volume, and maximal airway opening pressure, total respiratory system compliance (CT), total respiratory system resistance (RT), and resistance at 70% tidal volume were calculated for each breath cycle. Compliance and RT were calculated using the constant flow inflation single compartment model (22). All values were averaged over 6-s intervals and stored in data files. After establishing a stable baseline over a period of at least 2 min, mice were challenged via the jugular vein with doubling doses of methacholine (MCh) in saline (0.5, 1, 2, 4 µg) using a single catheter and needle. The volumes (5, 10, 20, 40 µl in 1-s delivery) were administered every 2 min using a syringe pump (Model 200; KD Scientific, Boston, MA) for precise timing and delivery of MCh. To standardize lung volumes prior to the first dose and 1 min after each dose, the expiratory port was briefly occluded until airway pressure reached ~ 30 cm H₂O. This maneuver eliminated artifactual differences in baseline resistance and compliance caused by focal atelectasis, which could affect responses to subsequent doses of MCh. The maximal change in RT after each dose is expressed as percent increase over baseline immediately prior to each dose [(peak RT  baseline RT)/baseline RT], whereas the maximal change in CT is expressed as percent decrease from baseline prior to each dose [(baseline CT  minimum CT)/baseline CT]. Maximal changes in both RT and CT were typically observed 12 to 30 s after drug infusion.

Serum IgE Assay

After airway responsiveness measurements, blood was obtained from the descending aorta for IgE analysis. Serum samples were stored at 80° C. Serum IgE concentrations were determined with an enzyme-linked immunosorbent assay (ELISA) as described previously (23). Briefly, 96-well plates were coated overnight with 100 µl monoclonal rat antimouse IgE (2.5 µg/ml; Southern BioTech), washed with TRIS-buffered saline (TBS) containing 0.5% Tween 20, blocked for an hour with 200 µl TBS containing 1% bovine serum albumin (Sigma), and washed again. Serial 4× dilutions of samples and purified mouse IgE standard (800-1.56 ng/ml; Sigma) in duplicate were applied for an hour, washed, and a biotinylated secondary antibody (sheep anti-mouse IgE; The Binding Site, Birmingham, UK) was applied. After 1 h, streptavidin alkaline phosphatase (Jackson Labs, Westgrove, PA) and substrate, p-nitrophenylphosphate disodium (Sigma), were added. Plates were read at 405 nm on a Thermomax plate reader (Molecular Devices Corp., Menlo Park, CA). Concentrations of IgE were calculated using a 4-parameter equation fit of the standard curve.

BAL Fluid Cells, Proteins, and Cytokines

Mice were lavaged with two aliquots of Ca²⁺- and Mg²⁺-free Hanks' balanced salt solution (28 ml/kg; HBSS). Approximately 90% of the total injected volume was consistently recovered. The lavage fluid was placed on ice and centrifuged at 360 × g for 12 min at 4° C. Supernatants were collected for biochemical and cytokine analysis, and cells were resuspended in 1.0 ml HBSS and counted (Coulter, Hialeah, FL). Slides of BAL fluid cells were made (Cytospin 3; Shandon, Pittsburgh, PA) and stained with Leuko-Stat (Fisher Scientific, Fair Lawn, NJ), and at least 500 cells per sample were differentiated. Assays for total protein, albumin, lactate dehydrogenase (LDH) and N-acetyl--D-glucosamidase (NAG) were modified for use on the Cobas Fara II centrifugal spectrophotometer (Hoffman-La Roche) as previously described (17).

Lavage fluid supernatants for cytokine analyses were mixed with fetal bovine serum (10%) to prevent nonspecific binding to vessel walls and stored at 80° C. Interleukin (IL)-4 and IL-5 protein concentrations in BAL fluid supernatant were determined using an ELISA with matched antibody pairs according to the manufacturer's suggested protocol (PharMingen, San Diego, CA). For IL-4 detection, rat IgG₁ clone 11B11 (capture) and biotinylated rat IgG1 clone BVD6-24G2 (detection) mAb were used. For IL-5 detection, rat IgG1 clone TRFK5 (capture) and biotinylated rat IgG2a clone TRFK4 (detection) mAb were used. Recombinant mouse cytokines (PharMingen) were used as standards. Concentrations were calculated using a quadratic equation fit of the standard curve.

Statistical Analysis

Deviations from Gaussian distribution were tested using the Kolmogorov-Smirnov test (GraphPad Prism, San Diego, CA). Pulmonary function and lavage fluid biochemical data were normally distributed and were analyzed without transformation. BAL fluid cellular data were not normally distributed and a square root transformation of the data was performed before analysis. Logarithm transformations of cytokine and IgE data were performed to normalize these data. In order to test for particular sensitivity of the ovalbumin allergic group to ROFA, all data were analyzed using two-way analysis of variance (GraphPad Prism) to test for interaction between row and column factors (allergic status and ROFA exposure) and effects cause by individual factors. Statistically significant interactions were referred to as greater than additive or synergistic. If an effect of interaction between the two factors was significant, effects caused by individual factors were not displayed since these are difficult to interpret (GraphPad Prism). Differences were considered significant if p (Type I error) < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Bronchoalveolar Lavage Fluid Proteins and Enzymes

Exposure to residual oil fly ash (ROFA) resulted in significant increases in biochemical markers of lung injury in BAL fluid supernatant (Table 1). Highly significant increases (p < 0.001) in protein, albumin, lactate dehydrogenase (LDH), and N-acetyl--D-glucosamidase (NAG) were found 1 and 3 d after treatment, which were attributable to ROFA exposure. Significant increases in these parameters associated with exposure to ROFA were also found 8 d after treatment, but there were no ROFA-induced effects by 15 d after exposure. Allergen sensitization and challenge was associated with a highly significant increase in protein 1 d after challenge and significant increases in all biochemical parameters 8 d after challenge. There were no significant interactions between ROFA exposure and allergen sensitization and challenge with respect to biochemical parameters at any time after treatment.

Serum IgE

Total serum IgE was greatly increased by ovalbumin sensitization and challenge in comparison with nonsensitized groups (Figure 1). A trend toward increasing values with time after challenge was observed in both OVA and OVA + ROFA groups, and, accordingly, the difference between allergen-sensitized and nonsensitized groups was greater with time. Exposure to ROFA did not significantly increase IgE concentrations, nor was there any interaction with ovalbumin sensitization and challenge at any time.

View larger version (21K): [in this window] [in a new window]   Figure 1.   Total serum IgE levels in mice 1, 3, 8, and 15 d after treatment. Mice were injected intraperitoneally with aluminum hydroxide adjuvant only, challenged with ovalbumin aerosol, and intratracheally instilled with saline vehicle (Control), injected with adjuvant only, challenged, and instilled with residual oil fly ash (ROFA; 3 mg/kg), sensitized and challenged with ovalbumin and instilled with saline (OVA), or sensitized and challenged with ovalbumin and instilled with residual oil fly ash (OVA + ROFA). Values shown are means ± SE (n = 4 to 8 mice per group). Two-way analysis of variance (ANOVA) was used to determine significant differences associated with (a) exposure to ROFA (b) ovalbumin sensitization and challenge, or (c) an interaction between these factors (a and c not significant in this analysis). **Significantly greater (p < 0.001) IgE levels in ovalbumin-sensitized and challenged mice than in nonsensitized mice.

BAL Fluid Cytokines

BAL fluid IL-4 concentrations were synergistically increased by combined ROFA exposure and OVA sensitization and challenge 1 and 3 d after treatment (Figure 2). The level of IL-4 in OVA + ROFA mice was 8-fold greater than that in OVA mice 1 d postexposure and 3-fold greater 3 d postexposure. The concentration of IL-5 was significantly increased by both ROFA exposure and allergen sensitization and challenge 1 d after treatment, but no significant interaction was observed. By 3 d postexposure, a significant interaction was observed between the treatments; the level of IL-5 in OVA + ROFA mice was 3-fold greater when compared with OVA-treated mice. Significantly increased levels of IL-4 and IL-5 were observed in allergen-sensitized and challenged mice as long as 8 d after challenge.

View larger version (26K): [in this window] [in a new window]   Figure 2.   Cytokine concentrations in BAL fluid of mice 1, 3, 8, and 15 d after treatment (groups described in Figure 1). Values shown are means ± SE (n = 4 to 8 mice per group). Significant differences are shown (no asterisk, p < 0.05; *p < 0.01; **p < 0.001) because of exposure to ROFA (a), because of ovalbumin sensitization and challenge (b), or because of an interaction between these factors (c).

BAL Fluid Cells

A small but significant increase in lavage fluid macrophage numbers was found 3 d after exposure to ROFA, whereas allergen sensitization and challenge caused a significant increase in macrophages 8 d after challenge (Figure 3). ROFA exposure induced a 25-fold increase in neutrophil numbers 1 d after exposure (p < 0.001). Neutrophil numbers thereafter declined, but the effect of ROFA exposure was still highly significant 3 d after exposure. Both ROFA exposure and ovalbumin sensitization and challenge induced significant increases in neutrophil numbers 8 d after exposure, but there were no significant interactions of ROFA and OVA on neutrophil numbers at any time after treatment. Numbers of lymphocytes were significantly increased only by allergen sensitization and challenge, and peaked 8 d after challenge. Essentially no eosinophils were recovered from nonallergic groups. Eosinophils in the OVA group increased from 3% of total cells 1 d after challenge to 24% 3 d after challenge and 71% 8 d after challenge, and then declined to 20% of total cells 15 d after challenge (Figure 3). A highly significant effect of allergen sensitization and challenge on eosinophil numbers was observed at all time points (not shown where interaction was observed). ROFA exposure and allergen sensitization and challenge interacted synergistically to increase BAL fluid eosinophil numbers 1 and 3 d after exposure (Figure 3). The number of eosinophils in the OVA + ROFA group was 8.3 times the sum of eosinophils in the ROFA group and the OVA group 1 d after exposure and 3.5 times the sum 3 d after exposure. Eosinophil numbers peaked 8 d after challenge and exposure, which was a highly significant effect of allergen sensitization and challenge only; by this time there were no longer any significant interactions between ROFA exposure and OVA sensitization.

View larger version (30K): [in this window] [in a new window]   Figure 3.   Inflammatory cell numbers in BAL fluid of mice 1, 3, 8, and 15 d after treatment (groups described in Figure 1). Values shown are means + SE (n = 4 to 8 mice per group). Significant differences are shown (no asterisk, p < 0.05; *p < 0.01; **p < 0.001) because of exposure to ROFA (a), because of ovalbumin sensitization and challenge (b), or because of an interaction between these factors (c).

Airway Responsiveness to MCh Challenge

Significant effects of treatments on baseline total respiratory system resistance were observed only on Day 1, when resistance in ROFA-exposed groups was increased 11% in comparison with saline-instilled groups (p = 0.03), and on Day 3, when resistance in OVA-allergic groups was increased 15% in comparison with nonallergic groups (p = 0.02; data not shown). ROFA exposure induced hyperreactivity to methacholine infusion on Day 1, indicated by significantly greater resistance at the lower three doses of drug (Figure 4) and a shift to the left of the dose-response curves. Ovalbumin sensitization and challenge caused a significant increase in resistance at the highest dose of MCh on Day 1, indicating an increase in maximal contractile responsiveness. The OVA-induced increase in maximal responsiveness was also observed Days 3 and 8 after challenge, but it was not quite significant 15 d after challenge (p = 0.08). A significant interaction between OVA and ROFA was observed 8 d after exposure at the lowest dose of MCh, indicating a greater than additive increase in sensitivity induced by these two factors. At this dose, the increase in total respiratory system resistance in the OVA + ROFA group was approximately twice that observed in the other three groups.

View larger version (28K): [in this window] [in a new window]   Figure 4.   Airway responsiveness to intravenously administered methacholine (MCh) challenge, expressed as percent increase in total respiratory system resistance (RT) over baseline, in anesthetized mechanically ventilated Balb/cJ mice 1, 3, 8, and 15 d after treatment (groups described in Figure 1). Results shown are means and SE (n = 4 to 8 mice per group). At individual doses of MCh, two-way analysis of variance (ANOVA) was used to determine significant differences (no asterisk, p < 0.05; *p < 0.01) associated with exposure to ROFA (a), ovalbumin sensitization and challenge (b), or an interaction between these factors (c).

Baseline respiratory system compliance was not affected by exposure to ROFA or allergen challenge 1, 3, or 15 d after exposure. However, on Day 8 a significant interaction between ROFA exposure and OVA sensitization and challenge resulted in lower baseline compliance prior to the two highest doses of MCh (data not shown). ROFA exposure caused a significant reduction in compliance 1 d after exposure in response to lower doses of MCh (Figure 5). A greater than additive decrease in compliance was observed at the highest dose of MCh on Day 1 because of interaction of ROFA and OVA effects. On Day 3, a highly significant reduction in compliance was observed at the highest dose of MCh because of OVA sensitization and challenge. By Day 8, compliance was further reduced after combined exposure to ROFA and OVA, which was more pronounced than the concurrent increase in total respiratory system resistance. Significant interactions between ROFA and OVA effects were observed on Day 8 at all doses of MCh except dose 3 (p = 0.07). If interaction was not considered, the decrease in compliance caused by OVA sensitization and challenge was highly significant (p < 0.001) for all doses of MCh (Figure 2; not shown where interaction was significant). By Day 15, a significant interaction between OVA and ROFA factors was still observed at one dose of MCh, and a significant effect of OVA sensitization and challenge was observed at two doses. These results show that effects of OVA and ROFA on compliance were longer-lasting compared with effects of these treatments on respiratory system resistance.

View larger version (27K): [in this window] [in a new window]   Figure 5.   Airway responsiveness to intravenously administered methacholine (MCh) challenge, expressed as percent decrease in total respiratory system compliance (CT) from baseline, in anesthetized ventilated Balb/cJ mice 1, 3, 8, and 15 d after treatment (groups described in Figure 1). Results shown are means and SE (n = 4 to 8 mice per group). Significant differences are shown (no asterisk, p < 0.05; *p < 0.01; **p < 0.001) because of ROFA exposure (a), because of ovalbumin sensitization and challenge (b), or because of an interaction between these factors (c).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The results of this study showed that physiologic and inflammatory responses to ovalbumin challenge in sensitized mice were enhanced after exposure to ROFA, an emission source PM with a high content of bioavailable transition metals. Interaction between the effects of ROFA and allergen was demonstrated by greater than additive increases in lavage fluid IL-4 and IL-5, eosinophil numbers, and airway responsiveness to methacholine challenge. The time course of these responses suggests that an early synergistic increase in Th2 cytokines, especially IL-4, contributes to increased allergic inflammation, decreased pulmonary function and airway hyperresponsiveness, which are manifest at later time points.

Exposure to ROFA resulted in a decline in lung function in mice 1 d after exposure, but not at later time points. At this time, baseline resistance and sensitivity to methacholine were increased in ROFA-exposed mice compared with nonexposed mice. Previous studies have shown that ROFA produces significant airway hyperreactivity in normal rats 4 d after exposure (17). This finding is likely to be related to a higher dose (8.3 mg/kg) than that used in the present study (3 mg/kg) as well as species differences in responsiveness to ROFA. Mice sensitized and challenged with ovalbumin (OVA) antigen exhibit features similar to human asthma, including airway hyperreactivity to cholinergic challenge and pulmonary inflammation characterized by increases in eosinophil numbers (18, 19, 24). In contrast to previous studies using OVA, a mild sensitization/challenge protocol was used in order to more readily detect interaction between OVA and ROFA treatments. Mice were challenged only once with OVA 2 wk after sensitization, resulting in an influx of eosinophils in BAL fluid comprising just 3% of total cells 1 d after challenge. Previous studies used multiple challenge days (24) or examined effects later than 1 d after challenge (18), resulting in greater numbers of lavaged eosinophils. In contrast to the effects observed with ROFA, OVA-induced hyperresponsiveness to MCh was observed at least 15 d after exposure. Similarly, increased numbers of pulmonary eosinophils and lymphocytes were observed at least 15 d after exposure. These results are consistent with another murine model in which airway hyperresponsiveness was shown to be dependent on CD4⁺ T-lymphocytes, and correlated with numbers of pulmonary eosinophils, which may be effector cells of the physiologic response (25). The combination of exposure to ROFA and challenge with OVA in sensitized mice resulted in greater than additive increases in eosinophilia and allergic cytokines IL-4 and IL-5, occurring at 1 and 3 d after exposure, and airway responsiveness, occurring primarily 8 and 15 d after exposure. The effects of combined exposure were most significant with respect to compliance 8 d after exposure, which was coincidental with maximal numbers of eosinophils. Both baseline compliance and MCh-induced reduction in compliance were significantly reduced, indicating that combined exposure resulted in loss of elasticity in the distal lung. The decrease in respiratory system compliance may be due to alterations in smooth muscle tone, surfactant function, or possibly connective tissue elements of the lung parenchyma. The timing of these effects indicates that ROFA increases early proallergic inflammatory responses which render the lung susceptible to nonspecific airway hyperresponsiveness at later time points.

Recent studies have shown that intratracheal instillation of diesel exhaust particles (DEP) can enhance OVA-induced allergic inflammatory responses in mice. In contrast to ROFA, which contains a high proportion of water-soluble sulfate and metals, especially V, Ni, and Fe, and a small carbon core (13, 17), DEP are primarily composed of elemental and organic carbon species, and have very little sulfate and metals (26). Intratracheal instillation of OVA and DEP (700 µg) over a 6-wk period resulted in significant increases in serum OVA-specific IgG₁, airway mucus-producing goblet cells, BAL fluid neutrophils, eosinophils, and IL-5 (27), and respiratory resistance in response to acetylcholine administration (28) in comparison with mice treated with OVA or DEP alone. These studies suggest that DEP can enhance responses to allergens in susceptible persons, who might be at greater risk living in urban or industrial areas with high levels of diesel exhaust pollution. Although two-way analysis of variance was not used to test for interaction in these studies, a greater than additive increase in respiratory system resistance was apparent in mice exposed to both OVA and DEP (28). Strain differences in responses to DEP are apparent as ICR (28) and C3H/He (29) mice, but not Balb/c mice (29) demonstrated airway hyperresponsiveness in response to OVA and DEP 1 d after the final exposure. In the present study, ROFA increased airway hyperresponsiveness in allergic Balb/cJ mice, whereas a higher dose of DEP had no significant effect in allergic Balb/c mice (29), suggesting that ROFA is more potent than DEP in this regard.

The present study showed a significant increase in serum total IgE in OVA-sensitized and challenged mice, but ROFA did not further increase total IgE, indicating that IgE does not play a major role in the enhancement of late inflammatory and physiologic responses by ROFA in allergic mice. Studies with intratracheally instilled DEP and OVA indicate that IgG, rather than IgE antibody production, is more closely associated with allergic inflammation and hyperresponsiveness, at least in ICR and C3H/He mice (27, 29). Allergen-induced hyperresponsiveness and pulmonary eosinophilia have been demonstrated in the presence of normal levels of allergen-specific IgG in IgE-deficient mice (30), and passive transfer of IgG1 is also capable of mediating these responses (31). These studies suggest a possible role for antigen-specific IgG in the responses to ROFA and OVA.

Changes in lavage fluid protein, LDH, and NAG represent significant increases in lung epithelial permeability, cytotoxicity, and lysosomal enzyme release, respectively. Increases in these biochemical indices of lung injury were associated with both ROFA exposure and allergen sensitization and challenge, but the effect of combined treatments was not greater than additive. Because these mediators did not correlate with airway hyperresponsiveness in OVA + ROFA mice, it is unlikely that they contribute significantly to physiologic responses associated with particle exposure in allergic mice.

The IL-4 concentration in BAL fluid was greatest in OVA + ROFA versus OVA-only mice 1 d after exposure, suggesting that this cytokine, a central mediator of the Th2 pathway promoting allergic reactions, may be upregulated by ROFA in allergic mice. ROFA has also been shown to slightly increase VCAM-1 adhesion molecule expression in normal rats (32). This molecule is upregulated by IL-4 and may be important for eosinophil adhesion and recruitment to lung airways. IL-4 is also critical for mucus cell hypertrophy and hypersecretion, another essential feature of allergic asthma (33, 34). In contrast to the results reported here with ROFA, enhancement of allergic responses by DEP was not associated with any increase in lung IL-4 levels in comparison with unexposed allergic mice (27, 29). The difference in IL-4 levels might be attributable to a longer time period after the first challenge with OVA (27), in which case an early increase in IL-4 after the first challenge in DEP-exposed mice could have subsided to levels observed in unexposed allergic mice. However, this appears unlikely since IL-4 levels in DEP-exposed allergic mice were either not different or marginally reduced compared with levels in unexposed allergic mice 24 h after a single challenge (29). Interleukin-5 is an essential mediator for eosinophil recruitment, activation, and survival (35). In this study, IL-5 levels were elevated in OVA + ROFA mice compared with OVA-only mice at 1 and 3 d postexposure, and a significant interaction between ROFA and allergen was found 3 d after exposure. Both IL-4 and IL-5 may contribute to the observed differences in eosinophil numbers and airway hyperresponsiveness between ROFA-exposed allergic mice and unexposed allergic mice.

In summary, the combination of exposure to ROFA and allergen challenge in sensitized mice resulted in greater than additive physiologic and inflammatory responses, which were associated with early increases in concentrations of Th2 cytokines in BAL fluid. These results complement other recent mechanistic studies that have correlated health effects with metal content of PM in normal animals and animal models of cardiopulmonary disease (12), and contribute to the understanding of mechanisms responsible for health effects of PM in allergic asthmatics. These data are consistent with epidemiologic studies which suggest that persons with preexisting respiratory disease, especially allergic asthma, may be at greater risk of adverse physiologic effects of PM (1). Future studies should examine whether the effects observed here with intratracheal instillation of animals may be replicated using relevant inhalation exposure protocols.