INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

SEVERAL STUDIES FROM OUR LABORATORY (5, 6, 29) have demonstrated that rats late in pregnancy and throughout lactation are more susceptible to acute ozone (O₃)-induced pulmonary inflammation and damage than virgin female rats, rats early in pregnancy, or postlactating rats. Data from one of these studies (29) suggested that a significant portion of the difference in pulmonary response between lactating and postlactating animals stems from the greater inhaled O₃ dose in lactating rats due to their metabolically driven higher ventilation. In addition to the difference in total inhaled O₃ dose, a significant difference in regional dose to the lower lung of lactating and postlactating rats must also occur because of the considerably larger tidal volume in lactating rats that drives O₃ deeper into the alveolar region (15). It seems probable that O₃ delivered deeply into the alveolar region may have a disproportionately greater effect on the induction of inflammatory changes. The earlier study by Weideman et al. (29) did not address this concept because accurate assessment of the regional dose from the available data was not possible. In the study presented here, the relative O₃ dose received by the lower lungs of pregnant, lactating, and prepregnant (virgin female) rats is estimated, and the possibility that the relative dose to this region can account for documented differences in O₃-induced inflammatory responses among these physiological states is explored.

Biological responses in the lung after O₃ exposure are not directly dependent on the concentration of O₃ in inhaled air but rather on the rate and duration of delivery of O₃ or O₃ metabolites to critical reactive sites in the respiratory tract tissues (i.e., O₃ dose to the tissue). Among groups of dissimilar animals (e.g., different physiological states) exposed to the same O₃ concentration, this critical parameter of dose may vary due to inherent differences in, for example, lung chemistry or ventilation. The lower lung was targeted for dose estimates because it is believed that O₃-induced lung inflammation is initiated primarily by reactions in the lower air spaces rather than in the upper airways. The amount of O₃ or active O₃ metabolite that reacts with specified components of the lower lung was measured by the methods of Hatch and colleagues (7, 8) and Santrock et al. (23) and was assumed to be proportional to the O₃ dose to the lower lung. The rationale for this method is that ¹⁸O derived from inhaled ¹⁸O₃ acts as a tracer for O₃ or O₃ metabolites that react with cellular or acellular components of interest (i.e., ¹⁸O₃-derived reaction products). For example, Hatch et al. (8) considered ¹⁸O incorporated into the bronchoalveolar lavage (BAL) fluid (BALF) cell pellet and the surfactant pellet to be estimates of the O₃ dose to the alveolar region of the lung because this is primarily where these targets are located. Experiments from the study presented here show that differences in O₃-induced pulmonary inflammation among pregnant, lactating, and virgin female rats were consistent with differences in O₃ surfactant dose measured as ¹⁸O incorporated into the surfactant pellet collected by BAL (i.e., ¹⁸O₃-derived reaction products incorporated into the surfactant pellet). Additional experiments comparing the antioxidant status of the lungs of pregnant, lactating, and virgin female rats were performed because the responses to O₃ are believed to be modulated by antioxidant factors in the lung (3). These experiments revealed that measurements of surfactant O₃ dose and inflammatory response among these physiological groups were inversely correlated with pulmonary ascorbic acid (AA) concentration.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Experimental animals. The rats used in this study were Sprague-Dawley supplied by Charles River Laboratories. Pregnant and lactating rats were mated by the breeder at 8-9 wk of age. The rats were received ~1 wk before exposure and were housed singly in solid-bottom cages in a room maintained at 20-23°C and 45-55% relative humidity with a 6 AM to 6 PM light cycle. Purina rodent chow and water were provided ad libitum. On the day of exposure to O₃ or air, the pregnant rats were at gestation day 17 (day of vaginal plug called day 1) and lactating rats were 13 days postpartum. At exposure, virgin female rats were between 10 and 14 wk, pregnant rats ~12 wk, and lactating rats 13-14 wk of age.

BAL and nasal lavage. BAL was performed as previously described (5). Briefly, the rats were exsanguinated via the abdominal aorta after an overdose of intraperitoneal pentobarbital sodium. The trachea was cannulated with a blunted 15-gauge needle, and the chest cavity was opened. The lungs were inflated to total capacity with calcium- and magnesium-free phosphate-buffered saline (PBS; pH 7.2) that flowed passively from a reservoir maintained at a height of 25 cm of buffer. The PBS remained in the lungs for 30 s after total inflation (i.e., 30-s dwell time) and was then briskly withdrawn manually with a syringe attached to the tracheal cannula via a three-way adapter. This BALF (BALF 1) was stored on ice, and the procedure was repeated five times. The BALF was centrifuged at 4°C at 450 g (low-speed centrifugation), and the supernatant and cells were processed and saved for later analytic procedures.

After BAL, the nasal passages of some rats were lavaged retrogradely via a blunted, 1.8-cm-long, 15-gauge stainless steel needle that was inserted into the trachea just distal to the larynx, advanced anteriorly through the nasopharyngeal orifice, and secured with suture thread. The jaw of the rat was removed so that pressure could be applied on both sides of the catheter to prevent leakage into the oral cavity during lavage. A syringe containing 1.4 ml of PBS was attached to the needle hub, the contents were slowly forced through the nasal passages, and the nasal lavage (NL) fluid (NLF) was collected as it dripped from the nares.

Animal exposures. The rats were restrained in tube holders and acutely exposed to either 0.5, 0.8, or 1.1 parts/million (ppm) O₃ or charcoal-filtered air for 1-4 h beginning at approximately 9 AM in a stainless steel "nose-only" apparatus (CH Technologies, Westwood, NJ). O₃ was generated by passing a mixture of 1% O₂ (¹⁶O or ¹⁸O, 99% purity) in argon through an electric arc generator at a precisely controlled flow rate in the range of 30-100 ml/min. The efficiency of conversion of O₂ to O₃ in this system was ~4%. The O₃ output was mixed with a stream of charcoal-filtered air (~15 l/min), and the O₃ concentration generated was maintained within ±2.0% of the target by controlling the flow of O₃ from the arc generator. The O₃ concentration was monitored continuously with a Dasibi 1003-PC O₃ analyzer (Dasibi Environmental, Glendale, CA), which was calibrated quarterly with a Monitor Laboratories (San Diego, CA) calibrator. The background O₃ concentration in the charcoal-filtered air was 0.00-0.02 ppm.

Separate groups of animals of each status were exposed to identical concentrations of either ¹⁸O₃ (tracer isotope) or ¹⁶O₃ (common isotope) and killed at different times after exposure to accurately assess both dose and inflammatory responses. This was necessary because the optimum time for the collection of material for measurement of ¹⁸O₃-derived reaction products (dose) is immediately after exposure because of the relatively short half-life of these reaction products (~6 h; Ref. 23), whereas the optimum time for detection of inflammatory changes is generally the day after exposure. Exposure to ¹⁸O₃ for estimation of dose was 3 h in duration, and exposure to ¹⁶O₃ for measurement of inflammatory changes was 4 h in duration. Because ¹⁸O₃-derived reaction products were used as a measure of relative dose (as opposed to absolute dose), these estimates of relative dose after 3 h of exposure were valid for rats exposed to O₃ for 4 h. All three physiological groups were exposed to 0.5 and 0.8 ppm in both ¹⁸O₃ and ¹⁶O₃ exposure regimens; however, only the virgin female group was exposed to 1.1 ppm. Virgin female rats were exposed to the higher concentration of O₃ to produce surfactant doses and inflammatory responses that were approximately comparable to those of the pregnant and lactating groups exposed to 0.8 ppm O₃.

Preparation of samples for estimation of dose. Animals acutely exposed to ¹⁸O₃ were killed immediately (5-10 min) after exposure, and the cell pellets from the low-speed centrifugation of BALFs 1-6 were combined in a 1.5-ml microfuge tube and stored at 70°C. The supernatants from the low-speed centrifugation of BALFs 1-3 were combined and centrifuged at 27,000 g (high-speed centrifugation) for 30 min. The resulting supernatant was discarded, and the pelleted material that contained most of the extracellular surface-active lipid-protein complex (22) was stored at 70°C. Although this procedure does not give a quantitative recovery of surfactant, it is a frequently employed and accepted method for its collection. Therefore, this high-speed pellet is hereafter referred to as surfactant or surfactant pellet. The surfactant and cell pellet samples were lyophilized, transferred into tared 5 × 3.5-mm-diameter silver capsules, and weighed on a microbalance. The samples were secured by crimping the capsules and were then stored at 70°C. All samples were analyzed within 4 mo for ¹⁸O concentration as described in ¹⁸O analysis.

Preparation of samples for measurement of inflammatory response. Groups of animals acutely exposed to ¹⁶O₃ (common isotope) were killed 20 ± 1 h after exposure. Aliquots of BALF 1 supernatant from the low-speed centrifugation were frozen at 70°C for later protein analysis (25). The cell pellets from all six lavages (BALFs 1-6) were pooled and enumerated, and viability was determined by trypan blue exclusion at a magnification of ×400. The viability of all groups was >90%, and there were no differences among groups. Slides of cells from the BALF pellets were prepared with a cytospin centrifuge, fixed with methanol, and stained with a Hemacolor stain set (EM Diagnostic Systems, Gibbstown, NJ) for differentiation of 500 inflammatory cells/rat at a magnification of ×1,000.

Preparation of samples for antioxidant analysis. Other groups of animals were acutely exposed to air or 0.8 ppm ¹⁶O₃ and killed immediately after exposure for BAL and NL. Aliquots of BALF 1 and NLF supernatants from the low-speed centrifugations were stabilized in 2.5% perchloric acid and stored at 70°C. Both stabilized supernatants were later analyzed for AA and uric acid (UA).

Lung tissue from additional groups of naive rats was prepared for subsequent determination of several antioxidant species. Approximately 0.2 and 0.3 g wet weight of lung tissue perfused via the pulmonary artery and devoid of major bronchi were homogenized with a Kinematica tissue homogenizer (model PT 10-35, Brinkmann Instruments, Westbury, NY) in 3.0 ml of 3.0% perchloric acid and 3.0 ml of 1.15% KCl-50 mM Tris buffer, pH 7.6, respectively. The homogenates were centrifuged at 20,000 g for 30 min, and the supernatants were stored frozen at 70°C pending analysis for AA, UA, and total glutathione (perchloric acid supernatant) and for glutathione peroxidase, glutathione reductase, superoxide dismutase, glucose-6-phosphate dehydrogenase (G6PDH), and catalase (Tris buffer supernatant). A third piece of perfused lung tissue (~0.5 g wet wt) was stored at 70°C for subsequent vitamin E analysis.

Antioxidant analyses. Aliquots used for enzymatic activity determinations were clarified after being thawed by centrifugation at 12,000 g for 20 min. Glutathione peroxidase activity was determined from the consumption of NADPH in the presence of tert-butyl hydroperoxide and glutathione reductase activity from the reduction of dithio(bis)nitrobenzoic acid (9). G6PDH activity was determined by the method of Lohr and Waller (13), superoxide dismutase activity by inhibition of the reduction of pyrogallol (14), and catalase by the method of Wheeler et al. (30). All enzymatic analyses were performed with a COBAS FARA autoanalyzer. Perchloric acid supernatants of tissue homogenates and BALFs were clarified by centrifugation at 20,000 g for 20 min and assayed for AA and UA by HPLC with amperometric detection (11). The perchloric acid tissue homogenate supernatant was also analyzed for total glutathione (oxidized plus reduced) with a cycling assay with the COBAS FARA autoanalyzer (1). Tissue vitamin E concentration was determined by HPLC analysis of the concentrated heptane extract of a 80% ethanol tissue homogenate (27).

¹⁸O analysis. Determination of the ¹⁸O concentration in lyophilized surfactant and cell pellet samples was accomplished by the analytic method described by Hatch et al. (8). In brief, this method consists of mass-spectrophotometric determination of the ratio of ¹⁸O to ¹⁶O in the sample after conversion of O₂ to CO₂. The raw data thus determined (¹⁸O-to-¹⁶O ratios) were corrected for background tissue ¹⁸O that results from the presence of ~0.2% of atmospheric O₂ in the ¹⁸O isomeric form. This background ratio of ¹⁸O to ¹⁶O is identical in all tissues and can be determined by analysis of plasma from air-breathing rats. The background ¹⁸O-to-¹⁶O ratio established from rat plasma was subtracted from each surfactant and cell pellet sample ¹⁸O-to-¹⁶O ratio obtained from rats exposed to ¹⁸O₃. The resulting corrected ratios represented the fractional enrichment of ¹⁸O due to ¹⁸O₃ exposure. The corrected sample ratios were then converted to "excess micrograms of ¹⁸O per gram of dry weight" (8).

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Comparison of pulmonary responses of virgin female, pregnant, and lactating rats to acute O₃ exposure. Data on inflammation and damage from rats exposed to 0.5, 0.8, or 1.1 (virgin female rats only) ppm O₃ for 4 h are shown in Figs. 1 (BALF PMN) and 2 (BALF protein). Both responses were greater in pregnant and lactating rats than in virgin female rats, and the absolute differences increased as the O₃ concentration increased. Two-factor ANOVA showed that there were significant differences with respect to physiological group, O₃ concentration, and the interaction of these two factors for both BALF protein and PMN responses. Subsequent single-factor ANOVA followed by post hoc testing indicated significant differences among the physiological groups at 0.5 and 0.8 ppm O₃ concentrations as shown in Figs. 1 and 2. This pattern of response is similar to those previously observed after 6-h exposures (5, 29).

View larger version (23K): [in this window] [in a new window]   Fig. 1.   Pulmonary polymorphonuclear leukocyte (PMN) response of virgin female, pregnant, and lactating rats 20 h after exposure to 0.5, 0.8, or 1.1 parts/million (ppm) ozone (O3; 16O3) for 4 h. In all groups, the predominant inflammatory cell type was the macrophage. Bronchoalveolar lavage fluid (BALF) PMN percentage of inflammatory cells for each O3-exposed rat was adjusted for background PMNs by subtraction of mean BALF PMN value obtained from air-exposed control rats. Each data point is mean ± SE of 5-7 animals. Mean BALF PMN values for air-exposed virgin female, pregnant, and lactating rats were 0.5, 0.9, and 3.3%, respectively. Significantly different (P < 0.05 by ANOVA with Tukey-Kramer post hoc test) from: a the other 2 groups at the same concentration; b virgin female group at the same concentration.

View larger version (22K): [in this window] [in a new window]   Fig. 2.   Pulmonary protein response (protein concentration in BALF 1) of virgin female, pregnant, and lactating rats 20 h after exposure to 0.5, 0.8, or 1.1 ppm 16O3 for 4 h. BALF 1 protein concentration of each O3-exposed rat was adjusted for background protein by subtraction of mean BALF 1 protein concentration obtained from air-exposed control rats. Each data point is mean ± SE of 5-7 animals. Mean BALF 1 protein concentrations for air-exposed virgin female, pregnant, and lactating rats were 89, 114, and 114 µg/ml, respectively. a Significantly different from virgin female and pregnant groups at the same concentration, P < 0.05 by ANOVA with Tukey-Kramer post hoc test.

Assessment of relative O₃ dose in the lower lungs of virgin female, pregnant, and lactating rats. The relative dose of O₃ (O₃ itself or reactive metabolites of O₃) delivered to reactive sites in the lower lung was estimated by measuring the concentration of ¹⁸O in lavageable surfactant and inflammatory cells immediately after exposure to ¹⁸O₃. To explore the validity of this estimation, the linearity of incorporation of ¹⁸O was investigated in virgin female rats exposed to 0.8 ppm ¹⁸O₃ for periods of 1-4 h (Fig. 3). Note that incorporation of ¹⁸O was reasonably linear in both surfactant and lavaged cells, although the rate was considerably greater in the surfactant.

View larger version (21K): [in this window] [in a new window]   Fig. 3.   Incorporation of 18O into surfactant and cell pellets immediately after exposure of virgin female rats to 0.8 ppm 18O3 for 1-4 h. Each data point is mean ± SE of 4-6 rats.

View larger version (23K): [in this window] [in a new window]   Fig. 4.   Relative surfactant O3 dose as a function of O3 concentration in virgin female, pregnant, and lactating rats exposed for 3 h to 18O3. Each data point is mean ± SE of 4-6 rats.

View larger version (24K): [in this window] [in a new window]   Fig. 5.   Relative cell pellet O3 dose as a function of O3 concentration in virgin female, pregnant, and lactating rats exposed for 3 h to 18O3. Each data point is mean ± SE of 4-6 rats.

Relationship between relative O₃ dose to the lower lung and pulmonary responses in virgin female, pregnant, and lactating rats. To establish the relationship between PMN inflammation and relative O₃ dose in the lower lung, mean values of BALF PMNs for each physiological group at each concentration (Fig. 1) were regressed against their respective mean O₃ doses estimated as surfactant-incorporated ¹⁸O (also termed "relative surfactant O₃ dose"). The result, illustrated in Fig. 6, indicates very good correlation between PMN inflammation and relative surfactant O₃ dose regardless of the physiological status. Analysis of the PMN values from individual animals (not mean values) showed that the slopes of the linear regressions of virgin female, pregnant, and lactating rats were not significantly different from the common regression slope derived from all data when the relative dose was used as the regressor (P > 0.05; F = 2.94; Fig. 6), whereas the slopes were significantly different when concentration was used as the regressor (P < 0.05; F = 8.90; Fig. 1). Data of BALF protein were analyzed similarly and are illustrated in Fig. 7. Although the F ratio improved from 44.51 to 11.26 when relative dose (Fig. 7) was used as the regressor instead of concentration (Fig. 2), individual regression slopes of the physiological groups were still significantly different from the common regression slope (P < 0.05), probably due mainly to the exaggerated BALF protein response in the lactating group after 0.8 ppm O₃. It is thought that the protein response in this group was exaggerated because of localized regions of breakdown of the blood-alveolar barrier resulting in pockets of air spaces flooded with plasma proteins (see DISCUSSION).

View larger version (24K): [in this window] [in a new window]   Fig. 6.   Pulmonary PMN response of virgin female, pregnant, and lactating rats 20 h after exposure to 0.5, 0.8, or 1.1 ppm 16O3 for 4 h regressed against mean relative surfactant O3 dose determined in separate groups of rats immediately after exposure to 18O3. PMN response data are the same as illustrated in Fig. 1, and relative surfactant O3 dose data are the same as shown in Fig. 4.

View larger version (22K): [in this window] [in a new window]   Fig. 7.   Pulmonary protein response of virgin female, pregnant, and lactating rats 20 h after exposure to 0.5, 0.8, or 1.1 ppm 16O3 for 4 h regressed against mean relative surfactant O3 dose determined in separate groups of rats immediately after exposure to 18O3. High-dose lactating group mean data point was not used in fitting the regression line. Protein response data are the same as illustrated in Fig. 2, and relative surfactant O3 dose data are the same as shown in Fig. 4.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

This study confirms that greater air space PMN and protein levels are induced in pregnant and lactating rats than in virgin female rats by acute exposure to the same concentrations of O₃. In addition, the present study extends the investigation of responses among these physiological states to include the relationship between pulmonary inflammation and an estimate of relative O₃ dose in the lower lung. This estimate consists of surfactant-incorporated ¹⁸O reaction products derived from inhaled ¹⁸O₃. The concentration of ¹⁸O in surfactant is thought to be proportional to the O₃ molecules that penetrate into the air spaces of the lower lung and escape scavenging by antioxidant molecules in the ELF. Although the chemical identity of the ¹⁸O₃-derived reaction products is not identified in this analysis, it is probable that incorporated ¹⁸O reflects chemically and biologically active O₃ metabolites. The validity of surfactant ¹⁸O₃-derived reaction products as an estimate of O₃ dose to the lower lung is supported by the linear incorporation of ¹⁸O with respect to time of exposure to ¹⁸O₃ (Fig. 3). In addition, the greater rate of incorporation of ¹⁸O into the surfactant of lactating rats compared with the surfactant of virgin female rats (Fig. 4) is consistent with the higher tidal volumes in lactating rats (28, 29).

Data from three previous studies (8, 10, 24) have shown that there is a direct correlation between ¹⁸O₃-derived reaction products in the BAL high-speed pellet (surfactant) and indicators of toxicity and inflammation. This positive correlation is confirmed by the results of this study that, in addition, show that the relationship between surfactant dose and PMN inflammation is linear among three physiologically diverse groups of rats exposed to several O₃ concentrations (Fig. 6). Thus the data illustrated in Fig. 6 demonstrate that the surfactant O₃ dose (¹⁸O₃-derived reaction products in surfactant) is a reliable measure of the effectiveness of O₃ in causing pulmonary PMN inflammation regardless of physiological status. This relationship implies that the O₃ and/or metabolites of O₃ that react with surfactant are either the same as or proportional to the chemical species responsible for initiating inflammation. It can also be inferred from Fig. 6 that the difference in PMN response to acute O₃ exposure among these physiological states (Fig. 1) can probably be explained by factors that determine the surfactant O₃ dose in the lower lung, such as ventilation and the concentration of key antioxidants, not by differences in the inherent sensitivity of pulmonary tissues to O₃ or its proximal toxicants.

Although not as sensitive to O₃ exposure as BALF PMNs, the BALF protein response to acute O₃ exposure is a reasonably good linear function of surfactant O₃ dose (Fig. 7) provided that the exaggerated response of lactating rats after exposure to 0.8 ppm is excluded. The basis for this exclusion is the suspected localized breakdown of the blood-air barrier in the high-concentration lactating group. Although the majority of airway protein detected in all other groups probably consisted of plasma proteins that transuded a relatively intact epithelial-capillary barrier, in the 0.8 ppm O₃-exposed lactating group much of the airway protein is believed to have leaked from ruptured blood vessels through a badly disrupted alveolar epithelium. This is suggested by the relatively large number of red blood cells (RBCs) in the BALF of this group (mean ~20 × 10⁶) compared with the few RBCs in the BALF of all other groups (mean ~1.4 × 10⁶). Such leakage of blood would greatly increase the BALF protein concentration but would not significantly affect the percentage of BALF PMNs because of the low concentration of PMNs in blood relative to other types of leukocytes.

In theory, lavageable inflammatory cells (largely macrophages) present in the RTLF of the normal lung should function similarly to surfactant molecules as reactive sites for O₃ and its metabolites. In actuality, however, the concentration of ¹⁸O₃-derived reaction products in the BALF pellet of inflammatory cells was not as good an indicator of O₃ dose as was surfactant ¹⁸O (compare Figs. 4 and 5). There are two probable reasons for the unreliability of cell pellet ¹⁸O₃-derived reaction products as an estimate of dose. First, the rate of incorporation of ¹⁸O into the cell pellet is slow relative to its incorporation into surfactant (see Fig. 3). Thus cell pellet-associated ¹⁸O is not responsive to relatively small increments in O₃ dose. Second, exposure to high O₃ concentrations causes some sloughing of epithelial cells, producing cellular fragments or debris of unknown ¹⁸O concentration that contaminate the inflammatory cells in the BALF cell pellet.

Other data from this study showed that lung tissue and BALF AA concentrations are lower in pregnant and lactating rats than in virgin female rats (Tables 1 and 2). This observed difference, especially the apparent difference in AA concentration in the ELF, could be an important factor modulating the rate of O₃ penetration to surfactant molecules and other reactive sites in the lower lung. The RTLF (ELF plus fluid lining the upper airways) is the initial compartment that inhaled O₃ encounters. It is believed that O₃ is essentially completely removed from the air phase by reactive absorption in this fluid layer (18, 19). Although the specific reactions of O₃ within the RTLF are not known, it is thought that O₃ can initiate a cascade of potentially toxic reactions that includes reactions with proteins and sites of unsaturation in fatty acids (19-21). Alternatively, O₃ may be scavenged by certain antioxidant molecules, thus limiting entry into the more toxic pathways (21). We hypothesize that an important factor in modulating both the incorporation of ¹⁸O into surfactant molecules and the induction of inflammation is the rate of removal of O₃ by antioxidants in the RTLF. In rats, the most prominent antioxidant in the RTLF is AA (3). It has been demonstrated in model systems that AA is depleted by O₃ and also that AA can effectively remove O₃ from an airstream (2, 12, 26). Results from the experiments reported here (Table 2), as well as those from a previously reported study (5), show that AA is depleted in BALF 1 by acute O₃ exposure and thus are consistent with the model experiments mentioned above. Furthermore, the results of the present study are consistent with the concept that the BALF 1 AA concentration reflects the concentration of AA in the ELF and the underlying tissues. The reaction of inhaled O₃ with AA in the ELF upsets the equilibrium between the ELF and the underlying cells, ultimately depleting AA in the epithelial tissues and compromising their capacity for protection against oxidant damage. Because pregnant and lactating rats have lower concentrations of AA in their ELF and tissues than virgin female rats (Tables 1 and 2), presumably a larger amount of inhaled O₃ will enter into toxic reactive pathways in the former physiological states, leading to both greater inflammation and greater incorporation into surfactant molecules. Data from the present study are consistent with, but do not confirm, this scenario because there may be other variables (e.g., ventilation) among pregnant, lactating, and virgin female rats that can affect O₃-induced inflammation and incorporation of ¹⁸O₃-derived reaction products into surfactant.

NL data (Table 2) suggest that AA in the lining fluid of the nasal passages reacts with inhaled O₃ and thus may function to protect the lower lung from O₃ exposure. A study by others (4) showed that the nasal tissues (tissues anterior to the posterior pharynx) of tidal-breathing humans remove an average of 40% of inhaled O₃. Therefore, it follows that factors affecting the efficiency of removal of O₃ by the nasal passages have biological significance. One such factor may be the presence of antioxidants in the nasal tissues and lining fluid. The data from this study suggest that AA in the nasal passages of the rat reacts with and removes inhaled O₃, much as UA is believed to do in humans (16, 17). The NL data from this study also indicate that, unlike the airways and tissues of the lung, there is no difference in AA concentration in the nasal passage RTLF among naive pregnant, lactating, and virgin female rats. Therefore, the observed differences in inflammation and surfactant O₃ dose among these physiological states are unlikely to be caused by differential removal of O₃ in the nasal passages by AA.

There are data from other laboratories that suggest the importance of ELF AA concentration in protection against the effects of acute O₃ exposure. For example, in a study investigating the effect of dietary restriction on acute O₃ toxicity in rats, Kari et al. (10) reported relationships similar to those observed here. Their data showed an inverse relationship between BALF AA concentration and both pulmonary inflammation and incorporation of ¹⁸O₃-derived reaction products into lung surfactant and cells. In their study, BALF glutathione was also elevated in diet-restricted animals and, therefore, like AA might have influenced the mitigation of O₃ effects in these rats.

In conclusion, this study has demonstrated that quantitatively disparate inflammatory responses among prepregnant (virgin), pregnant, and lactating rats acutely exposed to O₃ can be explained by proportional O₃ doses in the lower lung as estimated by ¹⁸O₃-derived reaction products incorporated into the surfactant collected by BAL. AA concentration in the BALF and lung tissue correlated inversely with inflammation and surfactant O₃ dose in these groups of rats and may, in part, mediate the observed differences.