INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Ozone is a strong photochemical oxidant and the principal air pollutant in many urban areas during the summer months. It is a major health concern because of the potential toxic effects related to its oxidative properties. The respiratory system is the primary target of ozone toxicity. It is known that the ozone- induced pulmonary injury may be present at several levels of the respiratory tract, but the nature, degree, and location of the injury may be highly variable. Animal studies have indicated that the principal lesions resulting from inhalation of high ambient concentrations of ozone are located in the nasal cavity and the pulmonary centriacinar region, the junction between conducting airways and alveolar gas exchange regions (1).

These affected regions appear to be species-independent, whereas the epithelial necrosis and inflammatory response within this region appear to be species-dependent (7, 8). Some differences in response are observed in the centriacinar region of different species (such as rodents and monkeys) according to the anatomic features of the region. In species whose centriacinar region is composed of terminal bronchioles opening directly into alveolar ducts such as rats and mice, the acute response to ozone exposure is necrosis of epithelial cells of terminal conducting airways and the transition between air-conducting and ventilatory units (i.e., distal terminal bronchiole and proximal alveolar duct) (9). Associated with this epithelial necrosis is an infiltrate of acute inflammatory cells comprised of large numbers of neutrophils in the peribronchiolar connective tissues in the centriacinar region (2, 6).

In the rat, the transition from bronchiolar epithelium in terminal conducting airways to alveolar lining cells in the proximal gas exchange region is abrupt (9). Experimental studies with rats at concentrations near the low ambient range (0.1 parts per million [ppm]) suggest that the lungs of rats are relatively insensitive to these environmentally relevant concentrations of ozone.

In those species whose centriacinar regions of the lungs include extensive and well-developed respiratory bronchioles such as humans, monkeys, and carnivores, the acute lesion is focused in the respiratory bronchiolar region, where the transition is a minimum of three generations of airway that contain bronchiolar epithelium on one side of the airway lumen and alveolar gas exchange on the other side (9). Studies in monkeys have shown significant changes in the respiratory bronchiole after short- and long-term exposures to 0.15 ppm ozone. Further, at 0.25 ppm ozone, a 4- to 10-fold increase in response to ozone (primarily necrosis of bronchiolar epithelial cells) was observed in monkeys as compared with rats (13).

Even though isolated tracheobronchial airway cells (14) and tracheal explants (15) have been exposed to ozone in vitro, respiratory bronchioles are extremely difficult to isolate for explant culture and have not been exposed ex vivo to ozone. Thus, the purpose of the study was to evaluate ferrets in vivo in comparison to monkeys and rats, as appropriate animal models to evaluate ozone-induced epithelial injury and repair in the region of the central acinus. We selected the ferret for the most detailed analysis, because ferretslike monkeys and humanshave well-developed respiratory bronchioles (11, 12) and because the central acinus of ferrets has not been evaluated for ozone-induced epithelial injury and inflammation.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animals and Exposure

Sprague-Dawley rats (n = 12), approximately 10 wk of age, obtained from a specific pathogen-free colony (Banton & Kingmann, Freeman, CA) were immediately housed in stainless-steel chambers and supplied with filtered air upon arrival at our facility. Young adult male rhesus monkeys (approximately 4 yr of age) (n = 8) were used in this study. All monkeys were born at the California Regional Primate Research Center and their medical records were known and considered in their selection. Young adult male ferrets (approximately 18 mo of age) were obtained from an outbred colony from Animal Resources Services at the University of California, Davis (n = 16). All animals were given a comprehensive physical examination, including a chest radiograph and complete blood count, before exposure.

Monkeys, ferrets, and rats were transferred to exposure chambers for 1 wk acclimatization during which they breathed filtered air. All animals remained in the exposure chambers at all times before and during the exposure period. Ozone or filtered air exposures (equal numbers per group: rats 6:6, monkeys 4:4, and ferrets 8:8) were performed in stainless steel chambers as previously described (6) at the exposure facility of the California Regional Primate Research Center, using a Sander model 25 ozonizer (Germany). For all of the exposures the chamber concentration was set for 1.0 ppm ozone or 0.0 ppm (filtered air) for 8 h and the concentration was measured using a Dasibi 1003AH ozone analyzer (Dasibi Environmental Corp., Glendale, CA), that was calibrated against a Dasibi UV Photometer model 1008 PC, which in turn was calibrated against a National Bureau of Standards reference photometer located at the California Air Resources Board Quality Assurance Standards Laboratory, Sacramento, CA. Ozone concentrations were logged at 30-s intervals using a microcomputer-based data acquisition system, and the data points recorded over an entire 8-h exposure period were averaged to compute the mean ozone concentrations ± SD for each chamber during a given exposure. The gas was delivered into a mixing chamber where it was diluted to the appropriate exposure concentration using a mixture of 95% air and 5% carbon dioxide.

Bronchoalveolar Lavage (BAL), Necropsy, and Lung Fixation

After 8 h of ozone exposure and 1 h of filtered air, monkeys were anesthetized with ketamine (30 to 35 mg/kg) injected subcutaneously, and killed using sodium pentobarbital overdose. Similarly, rats and ferrets were killed using sodium pentobarbital overdose. In ferrets, a catheter was placed in the trachea and 30 ml in four equal volumes (approximately 7 ml) of phosphate-buffered saline (PBS) was used to lavage the lung by a syringe. In rats, a catheter was placed in the trachea and 10 ml of PBS was used to lavage the lung by a syringe. In monkeys, the chest was opened, the left cranial lobe was cannulated, and 20 ml of PBS was used to lavage the lung by a syringe. Lavage fluid was placed on ice and processed immediately. The amount of lavage fluid recovered was recorded, specimens centrifuged at 300 × g for 5 min, the supernatant decanted and placed on ice, and the cell pellet resuspended in 5 to 10 ml of Hanks' balanced salt solution without calcium and magnesium. A total nucleated cell count was done using a cell counter (Coulter Electronics, Inc., Hialeah, FL) after erythrocyte lysis by Zab-oglobin (Coulter Electronics, Inc.). A cytocentrifuge (Shandon Southern Instruments Inc., Sewickley, PA) was used to prepare slides (2,000 rpm, 20 s) that were stained with a modified Wright's stain (LeukoStat; Fisher Scientific, Pittsburgh, PA) for differential cell counts. Three hundred cells per animal were counted by light microscopy (×1,000) and the proportion of macrophages, neutrophils, eosinophils, and lymphocytes was determined. Values were expressed as cells/ml from lavage of the lobe. Lavage fluid was centrifuged (259 × g, 5 min), and total protein was determined using the method of Bradford (16) and the BioradProtein assay (Bio-Rad, Richmond, CA). Right cranial lung lobes were fixed with 10% zinc- formalin (Anatech Ltd., Battle Creek, MI) by infusion through a tracheal cannula at 30 cm H₂O pressure for a minimum of 1 h.

Labeling and Observation of Injured Cells

Left cranial lung lobes from ferrets, monkeys, and left lungs of rats were instilled with up to 60 ml of 12 µM ethidium homodimer-1 (Molecular Probes Inc., Eugene, OR) in Waymouths media at room temperature for 15 min, lavaged with PBS, and instillation fixed for 30 min in 1% glutaraldehyde/1% paraformaldehyde (cacodylate buffer, 440 mOsm, pH 7.4). Subsequent to airway dissection, samples were mounted on coverslips using glue, and stained with a second DNA-binding fluorochrome YoPro-1 (Molecular Probes Inc., Eugene, OR). The samples were then immersed in a small Petri dish for observation on a Bio-Rad 600 Confocal Microscope (Hercules, CA) with water-immersion lenses. Two first-generation respiratory bronchioles (monkeys and ferrets) and terminal bronchioles (rats) were digitally captured. This method marks necrotic epithelial cells red (ethidium-positive) in a green epithelial background (17).

Selection of Airway Tissue for Morphology

We used a dissecting microscope with dual viewing capability, and a cool fiber optics illuminator to sample two acini from the right cranial lung lobe of ferrets, monkeys, and rats. The main airway was dissected along the axial pathway, beginning at the level of the lobar bronchus. This plane allows the majority of small side branches along the axial pathway to be available for sampling (18).

Airway Processing and Sectioning

Ferret, monkey, and rat airways were dehydrated in a graded series of ethanols and embedded in paraffin for light microscopy according to standard procedures. Several 5-µm sections were cut sagittally from the embedded tissue along one axial pathway. The tissues were stained with hematoxylin-eosin and evaluated by light microscopy.

Statistical Analysis

Differences between groups were analyzed using analysis of variance (ANOVA) and Fisher's least significant difference test when there were more than two comparisons (SYSTAT 5 for the Macintosh, Version 5.2; SYSTAT, Inc., Evanston, IL). All data are expressed as mean ± SEM unless otherwise stated. Statistical significance is accepted for p < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

BAL Cells and Protein

There was constant recovery of BAL fluid (BALF) from filtered air-exposed and ozone-exposed ferret, monkey, and rat lungs. We recovered 81 ± 8% for ferrets, 85 ± 7% for monkeys and 90 ± 5% for rats of the intratracheally instilled PBS used for the BAL. The numbers of neutrophils (PMN), macrophages (Mac), and lymphocytes (LYM) recovered per milliliter of BAL are shown in Figure 1. Neutrophils were significantly increased in ozone-exposed as compared with filtered air-exposed ferrets, monkeys, and rats. There were 3- to 4-fold more neutrophils in monkeys and ferrets per milliliter of BAL than in rats. There were no significant differences in the number of macrophages or lymphocytes between the two exposure groups. Eosinophil and mast cell numbers were so small that they were not reported. Similar to the increase in neutrophils, lavaged protein was significantly increased in ozone-exposed as compared with filtered air-exposed ferrets, monkeys, and rats (Figure 1).

View larger version (27K): [in this window] [in a new window]   Figure 1.   The number of cells recovered by BAL per ml, in thousands, is shown for filtered air (FA)-exposed and ozone (O3)-exposed ferrets (A), monkeys (B), and rats (C ) on the left axis. On the right axis, the lavaged protein in mg/ml (hatched bars) is shown. Bars represent the mean ± SEM. Mac = mononuclear phagocytes; PMN = neutrophils; LYM = lymphocytes. Ferrets n = 8, monkeys n = 4, and rats n = 6 per group. *Significantly greater (p < 0.05) than filtered air-exposed animals.

Histopathology

Lungs from ferrets, monkeys, and rats exposed to filtered air and ozone were examined at all airway levels by light microscopy. Ferrets, monkeys, and rats exposed to filtered air showed no abnormal morphologic changes. The principal alteration detected by light microscopy in first-generation respiratory bronchioles of ozone-exposed ferrets and monkeys was moderate-to-marked degeneration and necrosis of the bronchiolar epithelium (Figures 2 and 3). There was occasional desquamation and sloughing of epithelial lining cells and subsequent denudation of the basement membrane. In addition, there were aggregates and clusters of neutrophils and macrophages in the alveolar outpockets of the respiratory bronchioles as well as intramurally in the terminal bronchioles. There was mild to moderate swelling of the subjacent interstitium. In contrast, rats showed little degeneration and necrosis of the bronchiolar epithelium. We observed no desquamation or sloughing of the epithelium, but aggregates of neutrophils along with mild swelling of the subjacent interstitium were present (Figure 4).

View larger version (68K): [in this window] [in a new window]   Figure 2.   Representative light micrographs of respiratory bronchioles in filtered air-exposed (A) and ozone-exposed (B) ferrets. Histopathologic changes reveal a marked influx of neutrophils and cellular debris in respiratory bronchioles of ozone-exposed ferrets (inset). Bars = 50 µm.

View larger version (66K): [in this window] [in a new window]   Figure 3.   Representative light micrographs of respiratory bronchioles in filtered air-exposed (A) and ozone-exposed (B) monkeys. Histopathologic changes reveal a marked influx of neutrophils and epithelial sloughing in respiratory bronchioles in ozone-exposed monkeys (inset). Bars = 50 µm.

View larger version (73K): [in this window] [in a new window]   Figure 4.   Representative light micrographs of terminal bronchioles in filtered air-exposed (A) and ozone-exposed (B) rats. Histopathologic changes reveal a mild influx of neutrophils into the interstitium of terminal bronchioles in ozone-exposed rats (inset). Bars = 50 µm.

Laser Confocal Microscopy of Injured Epithelial Cells

Laser confocal microscopic images of injured bronchiolar epithelial cells in the centriacinar region after acute ozone exposure were used as a direct interspecies comparison among ferrets, monkeys, and rats (Figure 5). The degree of injury of bronchiolar epithelial cells after 8-h ozone exposure was very similar in ferrets and monkeys. Rats showed a greatly reduced number of injured bronchiolar epithelial cells compared with ferrets and monkeys. Lesions were most pronounced in the respiratory bronchiolar epithelium in ferrets and monkeys, and were markedly less in terminal bronchioles and proximal alveolar ducts in rats.

View larger version (87K): [in this window] [in a new window]   Figure 5.   Laser confocal microscope images of injured epithelial cells in red (increased permeability to the DNA-binding fluorochrome ethidium homodimer-1 in living tissue) and viable cells in green identified after fixation using a second DNA-binding fluorochrome Yo-Pro-1 in representative bronchioles from ferrets (A), rhesus monkeys (B), and rats (C ) after ozone exposure. Note the similar degree of injury in ferrets and monkeys, but the milder injury in rats. No injured epithelial cells were observed in filtered air controls. Bar = 20 µm.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This study yielded two novel observations in the evaluation of acute ozone-induced injury to the centriacinar region of ferret, monkey, and rat lungs: ozone exposure (1) induces severe centriacinar epithelial necrosis and inflammation in ferrets and monkeys, and (2) induces moderate centriacinar epithelial necrosis and inflammation in rats. Consequently, ozone inhalation in ferrets results in similar airway epithelial injury compared with monkeys, and therefore represents a better model of human ozone exposure than rodents.

The epithelial necrosis observed by laser confocal microscopy in ferrets after 8 h of ozone exposure was comparable to that in monkeys, but not rats and thus supports our initial hypothesis that species with respiratory bronchioles are particularly susceptible to ozone-induced injury. Interspecies variability in response to inhaled ozone depends to a significant degree on an inherent responsiveness to oxidant injury of those tissues and their anatomic and structural differences. It appears that the regions in the lung that are affected by acute ozone exposure are species-independent, whereas the injury and tissue response are species-dependent. It has been established from a variety of studies that the central acinus (junction between conducting airways and the alveolar gas exchange region) receives the greatest dose of ozone for humans and monkeys, as well as for rats, guinea pigs, and rabbits (7, 8, 19). Therefore this region in the lung seems to be critical in evaluation of ozone- induced epithelial damage. These models also indicate that for a given exposure concentration, the degree of injury of the inhaled ozone to the respiratory acinus is approximately twice as high in humans and monkeys as compared with rats (20).

This is probably dependent on major differences in the structure of the airways at the central acinar region among humans, monkeys, rats, guinea pigs, and rabbits. Humans, like monkeys, have extensive respiratory bronchioles comprised of bronchiolar epithelium and gas exchange epithelium (10). Respiratory bronchioles are partially alveolarized airways interposed between terminal bronchioles and alveolar ducts. The sensitivity of respiratory bronchioles to inhaled toxicants such as cigarette smoke, coal dust, and ozone is well described and recognized in species such as primates and humans (5, 21, 22). In humans, respiratory bronchioles are developed to at least two or three generations. Proximal generations of respiratory bronchioles are lined by nonciliated bronchiolar epithelial cells, ciliated cells, and type 1 epithelial cells. Distal generations are lined by nonciliated cuboidal bronchiolar and type 1 and 2 epithelial cells. Only a small number of laboratory animals are suitable models for studies inducing ozone injury to the region involving the respiratory bronchioles. Most commonly used rodents and lagomorphs lack well-developed respiratory bronchioles as compared with those in humans (9). In contrast, ferretslike monkeys and humanshave well-developed respiratory bronchioles with at least two to three generations (11). Ferrets also have large numbers of tracheal and submucosal glands with a similar secretory profile to humans (11, 23). Because of the anatomic and physiologic similarities to primates and humans, the ferret has the potential to be an important model for evaluation of airborne pollutants (12, 24, 25).

In addition to morphologic evaluation of ozone-induced lung injury, we used BAL as a quantitative parameter for the total inflammatory response of the entire airway tree and pulmonary parenchyma. However, the precise anatomic location of the inflammatory exudate along the airway tree recovered by BAL is uncertain (26). BAL revealed significantly higher numbers of neutrophils in ozone-exposed ferrets compared with filtered air-exposed ferrets. There were also 3- to 4-fold more neutrophils in monkeys and ferrets per milliliter of BAL than in rats. This pronounced neutrophilic response in ozone-exposed ferrets and monkeys is in agreement with the histopathologic changes of greater tissue injury, epithelial necrosis, and an acute inflammatory infiltrate in the centriacinar region.

Another parameter to evaluate ozone-induced toxicity is BAL total protein, which serves as a sensitive indicator of cellular membrane integrity loss. It has been used to quantitatively assess ozone-induced toxicity in a variety of animal and human studies (26, 27). Lavaged protein is considered to be a sensitive indicator of increased epithelial permeability, and is detectable at early stages, even before cellular breakdown occurs (28). Ferrets showed approximately a 1.5-fold increase in lavaged protein as compared with rats.

Studies in humans (29) found a similar inflammatory response to the findings in rhesus monkeys (5). Healthy, nonsmoking volunteers acutely exposed to ozone had their BAL performed and cells and supernatant were evaluated for various indicators of inflammation. There was an 8.2-fold increase in the number of polymorphonuclear leukocytes (PMNs) and a marked increase in the amount of inflammatory mediators such as prostaglandin E₂ (PGE₂) in BAL after ozone exposure. Ozone exposure in monkeys showed comparable increases in neutrophils and PGE₂ in BAL to humans (5). The monkey model of ozone exposure, as well as the results of this study in ferrets, accurately reflect the direction and general magnitude of inflammatory changes observed in BAL after human voluntary ozone exposure.

We conclude from our results that acute ozone exposure in ferrets induces severe airway inflammation and epithelial necrosis in the centriacinar region, particularly in the respiratory bronchioles. We propose that the ferret represents a suitable model of human ozone exposure, for the lesion is very similar compared with that observed in monkeys. Although monkeys have proven to be the species with the most similar response to humans (5, 29), they are limited for experimentation by their availability and cost. Ferrets could be used as a model to investigate the pathogenetic mechanisms involved in ozone, as well as other air pollutants, in airway injury in humans.