Background: The association between cigarette smoking and emphysema is strong, and occupational exposures to various dusts have been linked with higher incidences of certain types of emphysema. Epidemiology has associated air pollution exposures with increased morbidity and mortality among people with pre-existing chronic obstructive pulmonary disease (COPD). Because emphysema frequently coexists with other components of COPD (e.g., bronchitis) and is not easily distinguished on a clinical scale, we have little knowledge of pollution-induced risks for morbidity or mortality among people having COPD consisting predominantly of emphysema. Animal models may help determine the exacerbating effects of pollutants on pre-existing emphysema.

Degradation of elastin due to unchecked intrapulmonary elastastinolytic protease activity is the primary hypothesis of emphysema pathogenesis. Smoking causes pulmonary inflammation which leads to increased intrapulmonary release of proteases (such as neutrophil elastase and metalloproteinases), myeloperoxidase (MPO), and oxidants. Proteases damage the lung extracellular matrix. MPO and oxidants damage pulmonary parenchymal cells and the extracellular matrix, they inactivate intrapulmonary protease inhibitors, and they activate latent metalloproteinases having elastinolytic activity. Inflammation is amplified by release of cytokines from damaged parenchymal cells and from chemotactic peptides released from proteolysis. Cigarette smoke (CS) also contains oxidants that may directly inactivate protease inhibitors and may activate metalloproteinases and the complement system.

Ozone (O3) is an oxidant gas pollutant that can cause pulmonary fibrosis and changes in epithelial populations at high exposures. It does not cause emphysema in experimental animals, however, it does induce a cascade of free radicals, lipid hydroperoxides, aldehydes, hydrogen peroxide, and other reactive intermediates that lead to cell damage. Ozone also causes pulmonary inflammation that peaks after a few days but persists to a much lesser degree for months in exposed rodents. These oxidative and inflammatory effects may potentiate the emphysema-inducing effects of CS.

Hypothesis and Specific Aims: Our hypothesis is that chronic, repeated exposure to O3 will amplify cigarette smoke-induced emphysema in mice. The aims of the project are to determine; 1) whether O3 interacts with CS to augment emphysema in mice; and 2) the mechanisms contributing to the changes in response.

Methods: Female B6C3F1 mice are being exposed to filtered air (FA), low O3 (0.08 ppm, "LO"), high O3 (0.30 ppm, "HO"), CS (250 mg total particulate material/m3), LO + CS ("LOCS"), and HO + CS ("HOCS"). The CS exposures are conducted for 6 hours/day, 5 days/week. The O3 exposures are conducted for 8 hours/night, 5 nights/week. The CS concentration is one that previously produced emphysema in female B6C3F1 mice and represents a daily deposited particle dose approximately equivalent to that of a 3 pack/day smoker. The low O3 concentration is at the current primary NAAQS and the high O3 concentration greatly exceeds the NAAQS and may simulate peak concentrations in heavily polluted areas.

The actual chamber O3 concentrations are set higher than the target concentrations listed above to adjust for altitude (1,700 m) and yield partial pressures equivalent to 0.08 and 0.3 ppm at sea level. The actual exposure concentrations are: LO = 0.10 ppm; HO = 0.37 ppm. At a smoke particle concentration of 250 µg/m3, the gas concentrations are as follows: CO = 245 ppm, CO2 = 0.06%; HC = 126 ppm; NOx = 11.3 ppm. Impactor data collected to date indicate a mass median aerodynamic diameter of 0.33 µm, with a geometric SD of 1.5.

Mice were killed after 4 days/nights (~1 week), 4 weeks, and 15 weeks of exposure. The final termination is scheduled at the end of 32 weeks of exposure, near the end of June 2000. Bronchoalveolar lavage fluid (BALF) is analyzed for glutathione, proteases, antiproteases, total inflammatory cells, differential cell counts, production of superoxide from stimulated and non-stimulated adherent macrophages, and pro-inflammatory cytokines. Elastin degradation products are measured in BALF from the 15- and 32-week termination times. Lung total elastin will be measured at the final termination time. Lungs are fixed for histopathology (1 and 4 weeks), histochemistry (collagen and elastin), and/or histomorphometry (quantification of emphysema -15 and 32 weeks).

Status: Exposures began in April 1999. Problems occurred in maintaining a high O3 concentration in the exposure chamber containing the HOCS group, probably because of scrubbing of the O3 by CS-associated compounds that were deposited on chamber components and animals' bodies. Therefore, a new group of animals was ordered and the exposures were re-started in June 1999. Scrubbing of O3 by CS contaminants continued to be a problem; this was remedied by daily chamber washing and frequent changing of animal baskets. Problems with the O3 monitors also necessitated the re-starting of certain groups.

Mice from Blocks A and B (FA, LO, HO, CS, LOCS, and HOCS exposure groups) and Block C (FA, HO, CS, and HOCS) were killed after 4 days/nights (~1 week), 4 weeks, and 15 weeks of exposure. Mice from Block B were also killed after 32 weeks of exposure (end of January 2000). The BALF from these animals has been analyzed. Histopathologic evaluation of lungs from the 1-week time points has been performed.

A collaboration with Dr. Gary Hatch at EPA/NHEERL was established. Lung tissue from this study is being analyzed in Dr. Hatch's laboratory for the status of a variety of antioxidants (e.g., a-tocopherol, ascorbate) and antioxidant enzymes (e.g., superoxide dismutase, glutathione reductase).

Preliminary Results - Histopathology: After 4 days/nights of exposure, there were no significant changes in the lungs of mice exposed to FA or LO. The main changes in lungs of mice exposed to HO included minimal, diffuse centriacinar alveolitis and terminal bronchiolitis. The inflammation was characterized by increased interstitial and luminal macrophages and neutrophils. Macrophages predominated over neutrophils. In scattered foci, slight alveolar epithelial hyperplasia was present. The terminal bronchiolitis (slight to minimal) was characterized by increased neutrophils in the walls of terminal bronchioles and increased mitotic figures in the bronchiolar epithelium. The main findings in the lungs of mice exposed to CS included a minimal, multifocal alveolitis, a minimal multifocal to diffuse alveolar macrophage hyperplasia, and a minimal bronchiolitis. The alveolitis was more scattered and not restricted to centriacinar foci. The bronchiolitis differed from the terminal bronchiolitis seen in HO-exposed animals, because it involved bronchioles proximal to the terminal bronchioles and was not directly associated with centriacinar alveolitis. The changes in mice exposed to LOCS were similar to those in mice exposed to CS alone. The changes in the lungs of mice exposed to HOCS consisted of minimal multifocal to diffuse alveolar macrophage hyperplasia, minimal bronchiolitis, minimal to mild centriacinar alveolitis, and terminal bronchiolitis. The lesions appeared much like the addition of the CS lesions and the HO lesions, except the multifocal alveolitis was no longer present. Also, the centriacinar alveolitis exhibited more interstitial neutrophils and fewer luminal neutrophils than the centriacinar alveolitis in HO-exposed animals' lungs. Also within the HOCS group, the amount of centriacinar damage in the form of interstitial thickening due to cellular infiltrates and dilation of the centriacinus appeared greater than in HO-exposed lungs.

A preliminary evaluation after 4 weeks of exposure indicated no significant changes in the lungs of mice exposed to FA or LO. In lungs of animals exposed to HO for 4 weeks, there was slightly less centriacinar inflammation than in lungs from animals exposed to HO for 1 week. At that time point, however, there was a slight, diffuse increase in macrophages and inflammation of small and terminal bronchioles. In mice exposed to CS, there was mild multifocal alveolitis and minimal bronchiolitis. Thus, the amount of alveolitis was increased after 4 weeks of CS exposure, i.e., more alveoli were affected and they contained more macrophages and neutrophils at 4 weeks than at 1 week of exposure. The lesions in mice exposed to LOCS were like those in mice exposed to CS alone. The lesions in mice exposed to HOCS were like those in mice exposed to CS or LOCS except the macrophages were larger and there was more debris. The debris appeared to be cytoplasmic fragments of macrophages and karyorrhectic debris of neutrophils. Also, macrophages that had phagocytosed neutrophils or neutrophil nuclear fragments were observed.

Preliminary Results - BALF: Separately, CS alone caused a marked increase, and O3 caused a lesser, concentration-related increase, in BALF macrophages and neutrophils at 1, 4, and 15 weeks (Figure 9). Responses were greatest at 4 weeks. The effects of the combination LOCS exposure were not significantly different from the effect of CS alone. High O3 in combination with CS caused an increase in macrophages at 1 and 15 weeks and

Figure 9: Inflammatory cell responses to CS and /or O3. BALF was collected, and total recovered cells were counted (A). Differential cell counts were performed on Wright-Giemsa-stained cytocentrifuge slide preparations, and results for macrophages (B) and neutrophils (C) are depicted.

Figure 10. Effects of CS and/or O3 exposures on BALF glutathione concentrations (A), total protein (B), and alkaline phosphatase activity (C).

suppressed the usual number of macrophages associated with CS exposure alone at 4 weeks. At 1 and 4 weeks, the HOCS combination also significantly suppressed the neutrophilic inflammatory response that was normally associated with CS exposure alone. This suppression in neutrophil numbers was corroborated by the histologic findings (see above) where lungs of mice in the HOCS group after 1 week of exposure appeared to have fewer luminal neutrophils in foci of alveolitis than lungs from mice exposed to CS alone. After 4 weeks, the increased amounts of neutrophilic luminal debris associated with the HOCS-induced alveolitis also suggests that fewer neutrophils were recoverable by BAL.

Cigarette smoke caused an increase in total glutathione in the BALF at 1 week with a further increase at 4 weeks, and this partially resolved by 15 weeks (Fig. 10). Ozone caused a concentration-dependent increase in glutathione with a similar peak at 4 weeks. The effects of CS and O3 combined were nearly additive at the early time points. The increases in all exposure groups consisted nearly entirely of the reduced form of glutathione. Cigarette smoke and O3 both caused an increase in total protein, with the effects of O3 being concentration-dependent, but in contrast to the results for glutathione, the effects were not additive, and the greatest effects were observed at the earliest time point. Cigarette smoke and O3 also both caused increases in alkaline phosphatase activity in the lavage fluid, but the increases were not additive in the combined exposures. The effects on this enzyme indicated less resolution at 15 weeks than the results for protein or glutathione.

Cigarette smoke caused a rapid increase in gelatinolytic activity in BALF, as determined by zymography on SDS-polyacrylamide gels containing 1% gelatin. This activity was particularly associated with a 92 Kda protein (results not shown). The increased activity was sustained for at least 15 weeks. Low O3 had little effect, but HO caused a small increase in activity, particularly of a proteinase with molecular weight of approximately 60 Kda. The effect of combined HOCS at early times suppressed the CS-induced increases in the 92 Kda activity, but this suppression was no longer observed at 4 or 15 weeks.

Interim Conclusions: Interactions between CS and O3 are complex, time dependent, and may be additive or antagonistic depending on the parameter examined. The development of emphysema requires a lengthy period of CS exposure in this mouse model, as it does in humans. The early and complex interactions between CS and O3 may enhance, diminish, or have no effect on the development of emphysema. The summary conclusions must await completion of the morphometric evaluations for emphysema.

Schedule for Completion: Exposure of mice from Block C (FA, HO, CS, and HOCS groups) will reach the final evaluation time (32 weeks) in late June 2000. The lung sections from the 4-, 15-, and 32-week time points are ongoing histopathology and histomorphometry for emphysema at this time. Histochemistry for lung collagen and elastin, BALF cytokine analysis, BALF elastin degradation products, and total lung elastin content have yet to be performed. This work should be completed and a manuscript submitted for publication by the end of this year.