ENVIRONMENTAL RESEARCH54, 39--51 (1991)
Structural and Biochemical Effects in Lungs of Japanese Quail following a 1-Week Exposure to Ozone P . J. A . ROMBOUT,* J. A . M . A . D O R M A N S , t L . VAN BREE,* AND M . MARRA*
*Laboratory for Toxicology and ~Laboratory for Pathology, National Institute of Public Health and Environmental Protection, P.O. Box 1, Bilthoven, The Netherlands Received July 12, 1990 The effect of ozone inhalation on birds was investigated. Japanese quail were exposed continuously to 0, 0.3, 1.0, and 3.0 mg/m3 ozone (0, 0.15, 0.50, and 1.50 ppm, respectively) for 7 days. Pulmonary effects were determined by light and electron microscopy as well as by biochemistry. Focal areas of hemorrhages were noticed in the birds exposed to 1.0 mg/m 3 ozone. Additional effects after exposure to 1 mg/m3 included loss of cilia in trachea and bronchi, an inflammatory response, and necrosis of air capillary epithelial cells. Following exposure to 3 mg/m3 many atria of tertiary bronchi were completely obstructed by extensive hemorrhages, metaplasia of atrial wall cells, and hypertrophy of smooth muscle cells. Lung biochemistry data revealed that in the 3 mg/m3 group lactate dehydrogenase, glucose6-phosphate dehydrogenase, and glutathione reductase activities w e r e significantly increased. In the 0.3 and 1.0 mg/m3 exposure groups no effects on lung antioxidant enzymes were observed. In conclusion, Japanese quail appear to respond to ozone exposure in a different way than mammals. Since no signs of repair in air capillary epithelium after 7 days of continuous exposure were observed, the quail seems to lack the morphological and biochemical repair ability as is observed in mammals. Therefore, more research of the effects of ozone on birds seems to be necessary, both from a mechanistic and an ecological point of view. © 1991AcademicPress,Inc.
INTRODUCTION Ozone (03) is the major toxic air pollutant during the summer season. One-hour mean 03 concentrations of 200-400 ixg/m 3 (100-200 ppb) occur in industrialized areas. Eight- and twelve-hour mean 03 concentrations reach 80-90% and 70-80% of the one-hr mean maximum, respectively (Rombout et al., 1986a). Exposure of mammals to 03 concentrations of 160-400 p,g/m 3 for several hours per day during several days results in an array of effects. The most noticeable are lung function decrements and morphological alterations in the respiratory tract, and alterations in the activity of pulmonary antioxidant enzymes and the defense against respiratory infections (World Health Organization, 1987; Lippmann, 1989). Similar exposure conditions are phytotoxic for domestic crops and wild vegetation (Tonneijck, 1989). Furthermore, 03 is thought to be partly responsible for the devastating effects of air pollution on the decline of forests (Krause and Prinz, 1989). To our knowledge, data on the effects in birds following 03 exposure or exposure to air pollution in general are scarce. Quilligan et al. (1958) reported on the mortality in chickens after 03 exposure. The lack of more data appears to be somewhat peculiar since the avian species are an important order in the animal world. Adverse effects o f O 3 exposure on a specific species, or on birds in general, constitute a potential threat to the vitality of a species in heavily polluted areas. 39 0013-9351/91 $3.00 Copyright© 1991by AcademicPress, Inc. All rightsof reproductionin any formreserved,
40
R O M B O U T ET A L.
In addition, those effects may have a large impact on various ecosystems. In particular, birds flying at higher altitudes appear to be at risk since they are exposed to higher than ground level 03 concentrations, due to the positive concentration gradient (Wolff et al., 1987). The avian respiratory system differs markedly in its fundamental architecture from that of mammals. In short, the trachea branches via 2 primary bronchi into about 20 secondary bronchi. These secondary bronchi branch into many tertiary bronchi which form an open network of tubuli. The tertiary bronchus is lined by atria separated from each other by atrial septa. The atria narrow from the smaller infundibula into air capillaries which are intermingled with blood capillaries. The avian lungs may respond differently by nature and extent to oxidant gas exposure than lungs from mammalian species for several reasons. The gas exchange area per unit body weight is small compared to mammals (Scheid, 1979), since birds possess a higher surface density (surface area of the blood-gas barrier per unit volume of exchange tissue) (Maina and King, 1984). Furthermore, birds have a greater diffusing capacity than mammals (Maina and Settle, 1982) that accommodates the rather high metabolic rate during flight, which is presumably 7-15 times the basic metabolic rate (Lasiewski, 1972), resulting in a higher demand for oxygen. The consequent high respiration rate leads to an increased 03 flux to the tissue (in terms of ~g O3/sec/m 2 lung surface) when 03 polluted air is inhaled. A different response between birds and mammals might further be caused by differences in the ability to cope with reactive oxygen species, by differences in the antioxidant status, and by differences in the susceptibility of the very thin epithelium to 03. The unknown effectiveness of avian repair mechanisms in the respiratory tract constitutes yet another highly uncertain element in the qualitative risk assessment of 03 exposure of birds. The objective of the present study was to investigate the nature and the extent of the response of an avian species to various concentrations of 03. For this purpose Japanese quail were continuously exposed for 7 days to 0.3, 1.0, and 3.0 mg/m 3 03. The effects were characterized by morphology and activities of several antioxidant enzymes. MATERIALS AND METHODS
Animal Exposure After an acclimatization period of 1 week four groups of 8-week-old male Japanese quail (Coturnix coturnixjaponica) (110 -+ 10 g) were continuously exposed for 7 days to 0, 0.3, 1.0, or 3.0 mg/m 3 03 (0, 0.15, 0.50, or 1.5 ppm, respectively) with a maximal deviation of 10% in rectangular stainless steel and glass inhalation chambers (0.90 x 0.60 x 0.45 m) (Marra and Rombout, 1990). The quail were obtained from the CIVO-TNO Institute, Zeist, The Netherlands. The first three groups comprised eight animals, and the fourth group comprised six animals. Each group was housed in one chamber. The light cycle was 12-hr light/12-hr dark. Food (Turkey breeding food, Hope Farms, Woerden, The Netherlands) and water were available ad libitum. Ambient air was filtered through an active carbon filter and a high efficiency
OZONE TOXICITY IN AVIAN LUNG
41
particle filter. A flow of 6 m3/hr was maintained through the chambers. The temperature and relative humidity were conditioned at 22 +- I°C and 55 --- 5%, respectively. Permanganate- and Neomycin-impregnated animal cage boards (Upjohn, Kalamazoo, MI) prevented ammonia buildup from excreta. Food was renewed and the exposure chambers were cleaned daily. During this feeding and cleaning period (30 min), the O 3 exposures were interrupted. A negative pressure of -+ 100 Pa was maintained in the chambers. 03 was generated by irradiation of oxygen by ultraviolet light (254 nm). The 03/oxygen mixture was metered into the inlet stream of the exposure chambers by stainless steel massflow controllers (Model F-201, Hitec, Vorden, The Netherlands). The 03 concentration in each chamber was measured for 2 rain every 8 rain with a chemiluminescence O3-analyzer (Model 8002, Bendix Corp., Lewisburg, WV), which was calibrated daily by means of its internal 03 source. Before and after the experimental period, calibration was performed by gas-phase titration with an NBS-traceable mixture of nitric oxide in nitrogen. The consumption of nitric oxide and the production of nitrogen dioxide during this procedure were measured with a nitrogen oxide analyzer (Model PW9762, Philips, Eindhoven, The Netherlands). The exposures were controlled by a microcomputer equipped with the required interfaces (Minc-11, Digital Equipment Corp., Maynard, MA). The set points of the massflow controllers inthe oxygen/O3 streams were adjusted according to the deviations of the measured concentrations from the desired concentrations. Microscopy
Prior to autopsy the animals were anesthetized by pentobarbital sodium (60 mg/kg body wt ip). For histology the tracheae of three animals per group (two in the 3.0 mg/m 3 group) were cannulated. The lungs were fixed in situ for 1 hr at a pressure of 20 cm HzO with 2% glutaraldehyde in 0.1 M sodiumcacodylate buffer, pH 7.3,350 mOsm. (Maina and King, 1982). Hereafter the trachea and lungs were removed from the thorax in toto and were left immersed in fixative for at least 48 hr before processing. For light microscopical (LM) examination 5-t~m paraffin cross sections of trachea, lung, heart, liver, kidney, spleen, and bursa of Fabricius fixed by I0% buffered formalin were cut and routinely stained with hematoxylin and eosin. For scanning electron microscopy (SEM) lung sections about 4 mm thick were additionally fixed in OsO4, dehydrated in a graded acetone series, and dried in a critical point drying device (Polaron Instruments, Watford, UK). The dried samples were mounted on aluminium stubs and sputter coated with a thin layer of gold in a Polaron coating unit E 5000. The specimens were examined in a Philips PSEM 501 B scanning electron microscope (Eindhoven, The Netherlands). Additionally, for transmission electron microscopy (TEM) 10 small blocks of lung tissue from two control animals and two animals of the 1.0 mg/m 3 group were embedded in Epon. One-micrometer sections were cut and stained with toluidine blue. U1trathin sections were cut with an LKB ultramicrotome (Stockholm, Sweden), double-stained with uranyl acetate and lead citrate, and examined in a Philips EM 400 electron microscope.
42
ROMBOUT ET AL.
Biochemistry
The biochemical analyses for antioxidant enzymes were hampered by the inability to perfuse the respiratory system adequately. Therefore it was decided to bleed the lungs carefully. After anesthesia the birds were killed by cutting the abdominal aorta. The lungs of five birds per group (four in the 3.0 mg/m 3 group) were excised, trimmed free, blotted, weighed, and homogenized at 4°C in a glass-Teflon homogenizer in a sucrose-mannitol buffer (Mustafa et al., 1977). The homogenate was centrifuged at 105,000g for 60 min at 4°C. After measuring its volume, the supernatant was used immediately for biochemical analyses. The protein content (Lowry et al., 1951) and the activities of glucose-6-phosphate dehydrogenase (G6PDH) (Mustafa et al., 1977), glutathione peroxidase (GSH-Px) (Lawrence and Burk, 1976), glutathione reductase (GSH-Red) (Racker, 1955), and lactate dehydrogenase (LDH) (Van Bree et al., 1988) were measured in the supernatant. Enzyme activities are expressed in units per whole lung (1 unit equals 1 I~mole product formed/rain), units per gram lung, and units per gram cytosolic protein. Data are presented as means -+ standard error of the mean. The Student's t test (two-sided) was used for statistical analyses. RESU LTS In short, the avian respiratory system is illustrated here with LM, TEM, and SEM photographs of control quail. The secondary bronchi are lined by a pseudostratified columnar ciliated epithelium. The number of mucous glands decreases and the number of mucous goblet cells increases distally. The ciliated areas and the areas with mucous cells of the trachea orient clearly parallel with the axis of the trachea. The secondary bronchi (Fig. la) branch into many tertiary bronchi (parabronchi) which form an open network of tubuli. The tertiary bronchus is lined by atria separated from each other by atrial septa (Fig. 2a). The walls of the atria and septa are composed of high squamous or low cuboidal epithelial cells. These cells are characterized by microvilli at their apical surface and by lamellar bodies in their cytoplasm (Figs. 3 and 4). We suppose that small cytoplasmic strings remain partly covering the surface after the lamellar body has been opened at the luminal surface and surfactant has been released. A few muscle cells are located in the top of the atrial septa close to the lumen of the tertiary bronchi (Fig. 4). The atria narrow from the smaller infundibula into air capillaries which are intermingled with blood capillaries (Figs. 5a and 6A). The air capillaries are clad with nothing but epithelial cells with very thin cytoplasmic extensions, reminiscent of the mammalian type I pneumocyte (Fig. 7). Macrophages were not present in these air capillaries. No histological changes were observed in the control group. The animals of the 3.0 mg/m 3 O3 group developed laboured breathing with ongoing exposure. One animal of this group had to be autopsied after 5 days of exposure. At the end of the exposure period the relative lung weight of the animals of the 3.0 mg/m 3 03 group was significantly increased (Table 1). No statistically significant effects in all exposure groups were noted with respect to body weight
OZONE TOXICITY IN AVIAN LUNG
43
FIG. 1. (a) Epithelium of a secondary bronchus, where ciliated cells and mucous cells are intermingled. Control quail, SEM, 2000x. (b) Extensive loss and shortening of cilia is evident in the secondary bronchial epithelium. 1 mg/m 3 03, SEM, 2000x.
gain and weights of heart, liver, kidneys, spleen, and bursa of Fabricius. At macroscopical examination the lungs of the animals of the 3.0 mg/m 3 03 group were strongly hemorrhagic. Macroscopically, no alterations were observed in the other groups. Continuous exposure to 0.3 mg/m 3 03 for 7 days resulted in small focal hemorrhages in the capillary network near a tertiary bronchus in one of three animals. As no macroscopical changes were seen at autopsy this incidental finding has limited value and can be observed as an artifact. No other significant abnormalities were seen in lungs of the animals exposed to 0.3 mg/m 3 03. After exposure to 1.0 mg/m 3 03 for 7 days loss and shortening of cilia were observed in about 40% of the surface area of the trachea and the bronchi (Fig. lb). The epithelium o f the secondary bronchi was hypertrophic. To a large extent hemorrhages and inflammatory foci of lymphocytes, macrophages, and heterophylic granulocytes occurred diffusely in the capillary network (Fig. 8). Erythrocytes were present in the atria. With TEM no alterations were observed in the epithelium of the atrial walls. On the other hand many epithelial cells of the air capillaries were necrotic and desquamated, leaving a denuded basal lamina (Fig. 6B). This basal lamina was sometimes partly covered by cytoplasmic extensions of recruited macrophages. Focally, perivascular and interstitial edema were seen. In the animals exposed to 3.0 mg/m 3 03, the effects were clearly more severe. The cilia in the airways disappeared almost completely. Hypertrophy of the epithelium of the secondary bronchi was present and the lumina contained mucus
44
ROMBOUT ET AL.
FIG. 2. (a) Cross sections of tertiary bronchi, in-between air capillary network is present. The bronchi are circularly lined by atria separated from each other by thin septa. Control quail, LM, 80 ×. (b) The tertiary bronchi are closed by hypertrophied smooth muscle ceils. The air capillary network shows extensive hemorrhages. 3 mg/m 3 03 for 5 days, LM, 80×.
and erythrocytes. Hemorrhages, squamous metaplasia of the atrial epithelium with abundant mitoses, and a hypertrophy of the underlying smooth muscle cells resulted in complete closure of almost all atria (Figs. 2b and 9). Focally the atrial epithelium was necrotic, leaving smooth muscle cells exposed to 03. The capillary network of nearly the whole lung was hemorrhagic to such an extent that the lumina of tertiary bronchi were filled with massive conglomerations of erythrocytes. No histological differences were observed in the heart, liver; kidneys, spleen, and bursa of Fabricius between the animals of the control group and the group exposed to 3.0 mg/m 3 03. The results of the biochemical analysis are summarized in Table 1. Statistically significant pulmonary biochemical alterations were only observed in the 3.0 mg/ m 3 03 group. In this group G6PDH, GSH-Red, and L D H activities, expressed per whole lung tissue or per gram protein, were significantly increased. However, expressed per gram lung tissue the GSH-Px activity was decreased and the L D H activity was increased. DISCUSSION The trachea, secondary bronchi, and the entire parabronchial lung of Japanese quail are affected by a continuous 7-day exposure to 1.0 or 3.0 mg/m 3 03. The air-blood capillary network appears to be extremely vulnerable. Massive hemorrhages occur in the air-blood capillary network after exposure to 1.0 mg/m 3 03 for
OZONE TOXICITY IN AVIAN LUNG
45
FIG. 3. Longitudinal section of atrial septum. At both sides flat atrial wall cells are present with lamellar bodies in their cytoplasm. The cells are covered with microvilli. Control quail, TEM, 5400 ×. FIG. 4. Top of a longitudinally sectioned atrial septum. Smooth muscle cells are situated in the center. Thin cytoplasmic extensions of atrial wall cells form the epithelial lining. Control quail, TEM, 5400x.
7 days, indicating damage to the endothelium. This hemorrhagic response to 03 exposure is potentially life threatening at the 3.0 mg/m 3 03 level since the gas exchange capacity is jeopardized by the obstruction of the airflow in many atria. The small diameter of the air capillaries (3-10 Ixm) facilitates the complete closure by exudate and blood cells. This may lead to a higher airflow to other still open airways possibly resulting in a higher 03 dose to these airways and the surrounding gas exchange tissue, ultimately resulting in total lung failure with ongoing exposure. Since the quail were at rest, the effects registrated at 3.0 mg/m 3 03 were less than would be expected when the birds are very active. The avian respiratory system differs fundamentally from the mammalian lung. In birds, no tree-like branching pattern of the airways terminating in alveoli exists as in mammals; rather, the system is composed of two separate components, the lung itself and the air sacs, which allow the volume changes required for tidal ventilation (Scheid, 1979). The avian lung is composed of an interconnecting network of three generations of bronchi with numerous air capillaries. The air sacs can be considered as bellows which provide a unidirectional ventilatory flow to the otherwise relatively rigid parabronchial lung. The gas exchange is very efficient due to the unidirectional ventilatory flow, the crosscurrent flow system, and the relative continuous high oxygen concentration. The air capillaries are lined by epithelial cells with very large surfaces. Evans et al. (1985) concluded from rat studies that tolerance to 03 is absent in those
46
ROMBOUT ET AL.
FIG. 5. (a) View from a tertiary bronchus into atria and deeper lying infundibula. Septa (arrowheads) separate the various atria from each other. Control quail, SEM, 260 ×. (b) Two tertiary bronchi with atria, closed by hemorrhages and hypertrophied smooth muscle cells. Also the capillary network in between shows hemorrhages in contrast with the open capillary network in a control quail (Fig. 5a). 3 mg/m 3 03, SEM, 100×.
pneumocytes whose surface area, exposed to 03, is too large for the biochemical mechanism within the cell to protect it. Tolerance to 03 is reestablished by a decreased surface area of type I cells. The air capillary epithelial cells of Japanese quail, the equivalent of the mammalian type I pneumocyte, have a large surface area to volume ratio and appear to have a low metabolic activity. In a histochemical study of many enzymes in different regions of the chicken lung, the exchange area, mainly consisting of endothelial and air capillary epithelial cells, shows very weak enzyme activities (e.g., LDH and G6PDH) in contrast to strong activities in the bronchi and atrial epithelium (Tyler and Pearse, 1966). In mammalian species necrosis of the epithelial type I cell after 03 exposure is followed by a repair phase and proliferation of the interposed type II cell. The onset of injury and consecutive repair can be observed from Day 2 onward (Plopper et al., 1979; Evans et al., 1985). The air capillary epithelial cell appears to be the only cell type covering the avian air capillaries (MacDonald, 1970; Maina, 1987) as also observed in this study. Mitoses following damage of air capillary epithelium appear to be absent. After 7 days of exposure a denuded basal lamina is still observed, indicating the absence of repair processes. More research is needed to elucidate these processes. The atrial epithelial cells are often compared with the mammalian type II pneumocyte, because of the morphological similarities of microvillous structures and lamellar bodies. But there are some striking differences. While type II cells are interposed between type I cells in the gas exchange region, the atrial epithe-
OZONE TOXICITY IN AVIAN LUNG
47
FIG. 6. (A) Air capillaries (a) are alternately situated between erythrocyte-containing capillaries (b). Control quail, TEM, 3200 ×. (B) All air capillaries are devoid of epithelial cells (arrowheads). In the left upper corner a pyknotic epithelial cell is present, continuous in space with the interstitium, filled with lymphocytes and macrophages. I mg/m 3 03, TEM, 3500x.
lial cells are strictly located in the air conducting area. After exposure to 3 mg/m 3 03, but not after exposure to 1 mg/m 3 03, there is a proliferation of atrial epithelial cells, but this appears to have no effect on reepithelization of the air capillary spaces. Remarkably, the activities of LDH, G6PDH, and GSH-Px in control quail lung (this study) and control rat lung (Rombout et al., 1986b; Rietjens et al., 1985) are very comparable. Mammalian species exposed to oxidant gases respond by increased activities of antioxidant enzymes, probably originating from the result of shifts in lung cell populations and in intracellular biochemical changes. No biochemical reaction could be observed in the quail lung after continuous exposure to 0.3 or 1.0 mg/m 3 03, in contrast to increased enzyme activities in rats after 7 days of exposure to 0.3-0.8 mg/m 3 03 (Mustafa and Tierney, 1978). The GSH-Px activity, expressed in units per gram of lung, was significantly decreased in the 3.0 mg/m 3 group by an increase in lung weight. The elevated enzyme activities, as observed following exposure to 3 mg/m 3 03, are difficult to explain and are possibly due to the massive influx of erythrocytes and inflammatory cells as well as to the cellular changes in the atrial region (Tyler and Pearse, 1966). Assuming that an increase of antioxidant enzyme activities can be regarded as an adaptive response following oxidant exposure, the lack of effect up to 1.0 mg/m 3 03 exposure, as observed in this study, supposes the limited ability of the quail lung to respond adequately in this respect, thereby rendering this tissue more susceptible compared to mammalian lung.
48
ROMBOUT ET AL.
FIG. 7. Detail of the air-blood barrier (a-b). A thin air capillary epithelial cell with a nucleus is separated from a thin endothelial cell by a basal lamina. Control quail, TEM, 8600x.
TABLE 1 EFFECTS OF CONTINUOUS OZONE EXPOSURE FOR 7 DAYS ON BODY WEIGHT, LUNG WEIGHT, AND LUNG ANTIOXIDANT ENZYMES OF JAPANESE QUAILS mg/m 3 0 3 Body weight (g) Relative lung weight (mg/g bw) Protein (rag/lung) Protein (mg/g lung) G6PDH units/lung units/g lung units/g protein GSH-Px units/lung units/g lung units/g protein GSH-Red units/lung units/g lung units/g protein LDH units/lung units/g lung units/g protein
0 118 0.75 67,1 72.8
-+ 20 ~ -+ 0.09 -+ 11.2 -+ 8.0
0.3 114 0.84 75.5 76.5
_+ 6 -+ 0.11 -+ 12.1 -+ 4.0
1.0 109 0.87 59.9 73.2
-+ 24 _+ 0.20 _+ 12.7 -+ 8.6
3.0 114 1.20 73.8 53.6
-+ 8 + 0.14"** _+ 8.6 _+ 7.1
1.00 _+ 0.29 1.12 _+ 0.41 15.8 _+ 6.6
0.94 -+ 0.54 1.00 -+ 0.71 13.4 + 10.1
0.75 _+ 0.10 0.95 _+ 0.24 13.0 _+ 2.9
3.44 -+ 2.16" 2.47 -+ 1.58 44.9 -+ 23.4*
10.5 -+ 2.7 11.4 _+ 2.1 156 _+ 26
10.3 -+ 2.1 10.4 -+ 1.3 136 -+ 16
8.3 -+ 2.5 10.0 - 1.9 137 -+ 18
10.8 _+ 2.4 7.7 -+ 1.8' 145 -+ 16
1.27 _+ 0.24 1.38 +_ 0.25 19.1 + 3.5
1.38 -+ 0.18 1.41 +- 0.12 18.5 +- 2.1
1.10 -+ 0.22 1.34 -+ 0.14 18.6 -+ 3.1
2.61 _+ 0.42*** 1.89 _+ 0.40 35.3 _+ 3.6***
50.5 -+ 6.8 55.8 -+ 11.7 775 -+ 189
53.1 -+ 7.5 54.4 _+ 8.0 715 + 134
38.9 _+ 12.2 46.9 _+ 7.0 644 _+ 91
122 -+ 20*** 88 + 18.2" 1650 -+ 13"**
a Each value represents mean + SE of five animals (n = 4 for the 3 mg/m3 03 group). * P < 0.05; **P < 0.01; ***P < 0.001.
OZONE TOXICITY IN AVIAN LUNG
49
Fro. 8. Hemorrhages and influxes of inflammatory cells in the interstitium and air capillary space nearby atria. 1 mg/m 3 03, LM, 200x.
FIG. 9. Atrial wall epithelium shows metaplasia (arrowheads). The underlying capillary network is filled with inflammatory cells and erythrocytes. 3 mg/m 3 03, LM, 200x.
50
ROMBOUT ET AL.
Brown et al. (1982) described the resistance of chickens to acute pulmonary injury by O2, paraquat, and silica. Pulmonary lavage failed to reveal an influx of inflammatory cells, and no pulmonary edema was detected. However, in our experiments inflammatory cells were present both interstitially and inside the atrial lumina and air capillaries. Figures 10-12 in the article of Brown et al., however, reveal that the atrial lumina display an influx of inflammatory cells. To what extent a comparison is possible between a pulmonary lavage in chicken and in mammals is not clear. In conclusion, continuous exposure to an ambient concentration of 03 induced damage in the gas exchange area of quail lungs. Together with the apparent lack of adequate morphological and biochemical repair capacity, it gives rise to serious concern about potential health effects on an important order of animals in our environment. In particular birds with a high metabolic rate that are flying at rather high altitudes for a considerable part of the day might be at risk. More research concerning the vulnerability of birds to air pollution and the mechanism of damage and repair of their respiratory tract is inevitable to predict the effect of air pollution on various ecosystems. ACKNOWLEDGMENT The authors are grateful for the expert technical assistance of G. Riool-Nesselaar, J. Bos, and A. J. F. Boere.
REFERENCES Brown, Ch., Pratt, Ph. C., and Lynn, W. S. (1982). Characterization of pulmonary cellular influx differentials to known toxic agents between species. Inflammation 6, 327-341. Evans, M. J., Dekker, N. P., Cabral-Anderson, L. J., and Shami, S. G. (1985). Morphological basis of tolerance to ozone. Exp. Mol. Pathol. 42, 366-376. Krause, G. H. M., and Prinz, B. (1989). Current knowledge of ozone on vegetation forest effects and emerging issues. In "Atmospheric Ozone Research and Its Policy Implications" (T. Schneider, S. D. Lee, G. J. R. Wolters, and L. D. Grant, Eds.), pp. 45-55. Elsevier, Amsterdam. Lasiewski, R. C. (1972). Respiratory function in birds. In "Avian Biology" (D. S. Farner and J. R. King, Eds.), Vol. II, pp. 287-335. Academic Press, New York. Lawrence, R. A., and Burk, R. F. (1976). Glutathione peroxidase activity in selenium deficient rat liver. Biochem. Biophys. Res. Commun. 71, 952-958. Lippmann, M. (1989). Health effects of ozone: A critical review. J. Air Pollut. Control Assoc. 39, 672-695. Lowry, O. H., Rosenbrough, N. J., Farr, A. L., and Randall, R. J. (1951). Protein measurement with the Folin-phenol reagent. J. Biol. Chem. 193, 265-275. MacDonald, J. W. (1970). Observations on the histology of the lung of Gallus domesticus. Brit. Vet. J. 126, 89-93. Maina, J. N. (1987). Morphometrics of the avian lung. 4. The structural design of the charadriiform lung. Respir. Physiol. 68, 9%120. Maina, J. N., and King, A. S. (1982). Morphometrics of the avian lung. 2. The wild mallard (Arias Platyrhynchis) and graylag goose (Anser anser). Respir. Physiol. 50, 299-310. Maina, J. N., and King, A. S. (1984). Correlation between structure and function in the design of the bat lung: A morphometric study. J. Exp. Biol. 111, 43-61. Maina, J. N., and Settle, G. (1982). Allometric comparison of some morphometric parameters of avian and mammalian lungs. J. Physiol. (London) 330. Marra, M., and Rombout, P. J. A. (1990). Design and performance of an inhalation chamber for exposing laboratory animals to oxidant air pollutants. Inhalation Toxicol. 2, 187-204.
OZONE TOXICITY IN AVIAN LUNG
51
Mustafa, M. G., Hacker, M. D., Hussain, M. Z., and Lee, S. D. (1977). Biochemical effects of environmental oxidant pollutants in animal lungs. In "Biochemical Effects of Environmental Pollutants" (S. D. Lee, Ed.), pp. 59-96. Ann Arbor Science, Ann Arbor, MI. Mustafa, M. G., and Tierney, D. F. (1978). Biochemical and metabolic changes in the lung with oxygen, ozone and nitrogen dioxide toxicity. Amer. Rev. Respir. Dis. 118, 1061-1090. Plopper, C. G., Chow, C. K., Dungworth, D. L., and Tyler, W. S. (1979). Pulmonary alterations in rats exposed to 0.2 ppm and 0.1 ppm ozone: A correlated morphological and biochemical study. Arch. Environ. Health 34, 390-395. Quilligan, J. J., Boche, R. D., Falk, H., and Kotin, P. (1958). The toxicity of ozone for young chicks. A.M.A. Arch. Ind. Health 18, 16-22. Racker, E. (1955). Glutathione reductase (liver and yeast). In "Methods in Enzymology" (S. P. Colowick and N. O. Kaplan, Eds.), Vol. 2, pp. 722-725. Academic Press, New York. Rietjens, I. M. C. M., Bree, L. van, Marra, M., Poelen, M. C. M., Rombout, P. J. A., and Alink, G. M. (1985). Glutathione pathway enzyme activities and the ozone sensitivity of lung cell populations derived from ozone exposed rats. Toxicology 37, 205-214. Rombout, P. J. A., Lioy, P. J., and Goldstein, B. D. (1986a). Rationale for an eight hour ozone standard. J. Air Pollut. Control Assoc. 36, 913-917. Rombout, P. J. A.; Van Bree, L., Heisterkamp, S. H., and Marra, M. (1986b). Towards an exposureeffect model for short term ozone exposure of rats. In "7th World Clean Air Congress, Sydney 1986" (H. F. Hartman, Ed.), pp. 203-211. Holmes, Melbourne. Scheid, P. (1979). Mechanism of gas exchange in bird lungs. Rev. Physiol. Biochem. Pharmacol. 86, 137-186. Tonneijck, A. E. G. (1989). Evaluation of ozone effects on vegetation in the Netherlands. In "Atmospheric Ozone Research and Its Policy Implications" (T. Schneider, S. D. Lee, G. J. R. Wolters and L. D. Grant, Eds.), pp. 251-260. Elsevier Science Publ., Amsterdam. Tyler, W. S., and Pearse, A. G. E. (1966). Functional and analytical histochemistry of the chicken lung lobule with particular reference to surfactant. Poult. Sci. 45, 501-511. Van Bree, L., Haagsman, H. P., Van Golde, L. M. G., and Rombout, P. J. A. (1988). Phosphatidylcholine synthesis in isolated type II pneumocytes from ozone-exposed rats. Arch Toxieol. 61, 224-228. Wolff, G. T., Lioy, P. J., and Taylor, R. S. (1987). The diurnal variations of ozone at different altitudes on a rural mountain in the eastern United States. J. Air PoUut. Control Assoc. 37, 45--48. World Health Organization. (1987). "Air Quality Guidelines for Europe: Ozone and Other Photochemical Oxidants," pp. 315-326. Copenhagen.
Structural and Biochemical Effects in Lungs of Japanese Quail following a 1-Week Exposure to Ozone P . J. A . ROMBOUT,* J. A . M . A . D O R M A N S , t L . VAN BREE,* AND M . MARRA*
*Laboratory for Toxicology and ~Laboratory for Pathology, National Institute of Public Health and Environmental Protection, P.O. Box 1, Bilthoven, The Netherlands Received July 12, 1990 The effect of ozone inhalation on birds was investigated. Japanese quail were exposed continuously to 0, 0.3, 1.0, and 3.0 mg/m3 ozone (0, 0.15, 0.50, and 1.50 ppm, respectively) for 7 days. Pulmonary effects were determined by light and electron microscopy as well as by biochemistry. Focal areas of hemorrhages were noticed in the birds exposed to 1.0 mg/m 3 ozone. Additional effects after exposure to 1 mg/m3 included loss of cilia in trachea and bronchi, an inflammatory response, and necrosis of air capillary epithelial cells. Following exposure to 3 mg/m3 many atria of tertiary bronchi were completely obstructed by extensive hemorrhages, metaplasia of atrial wall cells, and hypertrophy of smooth muscle cells. Lung biochemistry data revealed that in the 3 mg/m3 group lactate dehydrogenase, glucose6-phosphate dehydrogenase, and glutathione reductase activities w e r e significantly increased. In the 0.3 and 1.0 mg/m3 exposure groups no effects on lung antioxidant enzymes were observed. In conclusion, Japanese quail appear to respond to ozone exposure in a different way than mammals. Since no signs of repair in air capillary epithelium after 7 days of continuous exposure were observed, the quail seems to lack the morphological and biochemical repair ability as is observed in mammals. Therefore, more research of the effects of ozone on birds seems to be necessary, both from a mechanistic and an ecological point of view. © 1991AcademicPress,Inc.
INTRODUCTION Ozone (03) is the major toxic air pollutant during the summer season. One-hour mean 03 concentrations of 200-400 ixg/m 3 (100-200 ppb) occur in industrialized areas. Eight- and twelve-hour mean 03 concentrations reach 80-90% and 70-80% of the one-hr mean maximum, respectively (Rombout et al., 1986a). Exposure of mammals to 03 concentrations of 160-400 p,g/m 3 for several hours per day during several days results in an array of effects. The most noticeable are lung function decrements and morphological alterations in the respiratory tract, and alterations in the activity of pulmonary antioxidant enzymes and the defense against respiratory infections (World Health Organization, 1987; Lippmann, 1989). Similar exposure conditions are phytotoxic for domestic crops and wild vegetation (Tonneijck, 1989). Furthermore, 03 is thought to be partly responsible for the devastating effects of air pollution on the decline of forests (Krause and Prinz, 1989). To our knowledge, data on the effects in birds following 03 exposure or exposure to air pollution in general are scarce. Quilligan et al. (1958) reported on the mortality in chickens after 03 exposure. The lack of more data appears to be somewhat peculiar since the avian species are an important order in the animal world. Adverse effects o f O 3 exposure on a specific species, or on birds in general, constitute a potential threat to the vitality of a species in heavily polluted areas. 39 0013-9351/91 $3.00 Copyright© 1991by AcademicPress, Inc. All rightsof reproductionin any formreserved,
40
R O M B O U T ET A L.
In addition, those effects may have a large impact on various ecosystems. In particular, birds flying at higher altitudes appear to be at risk since they are exposed to higher than ground level 03 concentrations, due to the positive concentration gradient (Wolff et al., 1987). The avian respiratory system differs markedly in its fundamental architecture from that of mammals. In short, the trachea branches via 2 primary bronchi into about 20 secondary bronchi. These secondary bronchi branch into many tertiary bronchi which form an open network of tubuli. The tertiary bronchus is lined by atria separated from each other by atrial septa. The atria narrow from the smaller infundibula into air capillaries which are intermingled with blood capillaries. The avian lungs may respond differently by nature and extent to oxidant gas exposure than lungs from mammalian species for several reasons. The gas exchange area per unit body weight is small compared to mammals (Scheid, 1979), since birds possess a higher surface density (surface area of the blood-gas barrier per unit volume of exchange tissue) (Maina and King, 1984). Furthermore, birds have a greater diffusing capacity than mammals (Maina and Settle, 1982) that accommodates the rather high metabolic rate during flight, which is presumably 7-15 times the basic metabolic rate (Lasiewski, 1972), resulting in a higher demand for oxygen. The consequent high respiration rate leads to an increased 03 flux to the tissue (in terms of ~g O3/sec/m 2 lung surface) when 03 polluted air is inhaled. A different response between birds and mammals might further be caused by differences in the ability to cope with reactive oxygen species, by differences in the antioxidant status, and by differences in the susceptibility of the very thin epithelium to 03. The unknown effectiveness of avian repair mechanisms in the respiratory tract constitutes yet another highly uncertain element in the qualitative risk assessment of 03 exposure of birds. The objective of the present study was to investigate the nature and the extent of the response of an avian species to various concentrations of 03. For this purpose Japanese quail were continuously exposed for 7 days to 0.3, 1.0, and 3.0 mg/m 3 03. The effects were characterized by morphology and activities of several antioxidant enzymes. MATERIALS AND METHODS
Animal Exposure After an acclimatization period of 1 week four groups of 8-week-old male Japanese quail (Coturnix coturnixjaponica) (110 -+ 10 g) were continuously exposed for 7 days to 0, 0.3, 1.0, or 3.0 mg/m 3 03 (0, 0.15, 0.50, or 1.5 ppm, respectively) with a maximal deviation of 10% in rectangular stainless steel and glass inhalation chambers (0.90 x 0.60 x 0.45 m) (Marra and Rombout, 1990). The quail were obtained from the CIVO-TNO Institute, Zeist, The Netherlands. The first three groups comprised eight animals, and the fourth group comprised six animals. Each group was housed in one chamber. The light cycle was 12-hr light/12-hr dark. Food (Turkey breeding food, Hope Farms, Woerden, The Netherlands) and water were available ad libitum. Ambient air was filtered through an active carbon filter and a high efficiency
OZONE TOXICITY IN AVIAN LUNG
41
particle filter. A flow of 6 m3/hr was maintained through the chambers. The temperature and relative humidity were conditioned at 22 +- I°C and 55 --- 5%, respectively. Permanganate- and Neomycin-impregnated animal cage boards (Upjohn, Kalamazoo, MI) prevented ammonia buildup from excreta. Food was renewed and the exposure chambers were cleaned daily. During this feeding and cleaning period (30 min), the O 3 exposures were interrupted. A negative pressure of -+ 100 Pa was maintained in the chambers. 03 was generated by irradiation of oxygen by ultraviolet light (254 nm). The 03/oxygen mixture was metered into the inlet stream of the exposure chambers by stainless steel massflow controllers (Model F-201, Hitec, Vorden, The Netherlands). The 03 concentration in each chamber was measured for 2 rain every 8 rain with a chemiluminescence O3-analyzer (Model 8002, Bendix Corp., Lewisburg, WV), which was calibrated daily by means of its internal 03 source. Before and after the experimental period, calibration was performed by gas-phase titration with an NBS-traceable mixture of nitric oxide in nitrogen. The consumption of nitric oxide and the production of nitrogen dioxide during this procedure were measured with a nitrogen oxide analyzer (Model PW9762, Philips, Eindhoven, The Netherlands). The exposures were controlled by a microcomputer equipped with the required interfaces (Minc-11, Digital Equipment Corp., Maynard, MA). The set points of the massflow controllers inthe oxygen/O3 streams were adjusted according to the deviations of the measured concentrations from the desired concentrations. Microscopy
Prior to autopsy the animals were anesthetized by pentobarbital sodium (60 mg/kg body wt ip). For histology the tracheae of three animals per group (two in the 3.0 mg/m 3 group) were cannulated. The lungs were fixed in situ for 1 hr at a pressure of 20 cm HzO with 2% glutaraldehyde in 0.1 M sodiumcacodylate buffer, pH 7.3,350 mOsm. (Maina and King, 1982). Hereafter the trachea and lungs were removed from the thorax in toto and were left immersed in fixative for at least 48 hr before processing. For light microscopical (LM) examination 5-t~m paraffin cross sections of trachea, lung, heart, liver, kidney, spleen, and bursa of Fabricius fixed by I0% buffered formalin were cut and routinely stained with hematoxylin and eosin. For scanning electron microscopy (SEM) lung sections about 4 mm thick were additionally fixed in OsO4, dehydrated in a graded acetone series, and dried in a critical point drying device (Polaron Instruments, Watford, UK). The dried samples were mounted on aluminium stubs and sputter coated with a thin layer of gold in a Polaron coating unit E 5000. The specimens were examined in a Philips PSEM 501 B scanning electron microscope (Eindhoven, The Netherlands). Additionally, for transmission electron microscopy (TEM) 10 small blocks of lung tissue from two control animals and two animals of the 1.0 mg/m 3 group were embedded in Epon. One-micrometer sections were cut and stained with toluidine blue. U1trathin sections were cut with an LKB ultramicrotome (Stockholm, Sweden), double-stained with uranyl acetate and lead citrate, and examined in a Philips EM 400 electron microscope.
42
ROMBOUT ET AL.
Biochemistry
The biochemical analyses for antioxidant enzymes were hampered by the inability to perfuse the respiratory system adequately. Therefore it was decided to bleed the lungs carefully. After anesthesia the birds were killed by cutting the abdominal aorta. The lungs of five birds per group (four in the 3.0 mg/m 3 group) were excised, trimmed free, blotted, weighed, and homogenized at 4°C in a glass-Teflon homogenizer in a sucrose-mannitol buffer (Mustafa et al., 1977). The homogenate was centrifuged at 105,000g for 60 min at 4°C. After measuring its volume, the supernatant was used immediately for biochemical analyses. The protein content (Lowry et al., 1951) and the activities of glucose-6-phosphate dehydrogenase (G6PDH) (Mustafa et al., 1977), glutathione peroxidase (GSH-Px) (Lawrence and Burk, 1976), glutathione reductase (GSH-Red) (Racker, 1955), and lactate dehydrogenase (LDH) (Van Bree et al., 1988) were measured in the supernatant. Enzyme activities are expressed in units per whole lung (1 unit equals 1 I~mole product formed/rain), units per gram lung, and units per gram cytosolic protein. Data are presented as means -+ standard error of the mean. The Student's t test (two-sided) was used for statistical analyses. RESU LTS In short, the avian respiratory system is illustrated here with LM, TEM, and SEM photographs of control quail. The secondary bronchi are lined by a pseudostratified columnar ciliated epithelium. The number of mucous glands decreases and the number of mucous goblet cells increases distally. The ciliated areas and the areas with mucous cells of the trachea orient clearly parallel with the axis of the trachea. The secondary bronchi (Fig. la) branch into many tertiary bronchi (parabronchi) which form an open network of tubuli. The tertiary bronchus is lined by atria separated from each other by atrial septa (Fig. 2a). The walls of the atria and septa are composed of high squamous or low cuboidal epithelial cells. These cells are characterized by microvilli at their apical surface and by lamellar bodies in their cytoplasm (Figs. 3 and 4). We suppose that small cytoplasmic strings remain partly covering the surface after the lamellar body has been opened at the luminal surface and surfactant has been released. A few muscle cells are located in the top of the atrial septa close to the lumen of the tertiary bronchi (Fig. 4). The atria narrow from the smaller infundibula into air capillaries which are intermingled with blood capillaries (Figs. 5a and 6A). The air capillaries are clad with nothing but epithelial cells with very thin cytoplasmic extensions, reminiscent of the mammalian type I pneumocyte (Fig. 7). Macrophages were not present in these air capillaries. No histological changes were observed in the control group. The animals of the 3.0 mg/m 3 O3 group developed laboured breathing with ongoing exposure. One animal of this group had to be autopsied after 5 days of exposure. At the end of the exposure period the relative lung weight of the animals of the 3.0 mg/m 3 03 group was significantly increased (Table 1). No statistically significant effects in all exposure groups were noted with respect to body weight
OZONE TOXICITY IN AVIAN LUNG
43
FIG. 1. (a) Epithelium of a secondary bronchus, where ciliated cells and mucous cells are intermingled. Control quail, SEM, 2000x. (b) Extensive loss and shortening of cilia is evident in the secondary bronchial epithelium. 1 mg/m 3 03, SEM, 2000x.
gain and weights of heart, liver, kidneys, spleen, and bursa of Fabricius. At macroscopical examination the lungs of the animals of the 3.0 mg/m 3 03 group were strongly hemorrhagic. Macroscopically, no alterations were observed in the other groups. Continuous exposure to 0.3 mg/m 3 03 for 7 days resulted in small focal hemorrhages in the capillary network near a tertiary bronchus in one of three animals. As no macroscopical changes were seen at autopsy this incidental finding has limited value and can be observed as an artifact. No other significant abnormalities were seen in lungs of the animals exposed to 0.3 mg/m 3 03. After exposure to 1.0 mg/m 3 03 for 7 days loss and shortening of cilia were observed in about 40% of the surface area of the trachea and the bronchi (Fig. lb). The epithelium o f the secondary bronchi was hypertrophic. To a large extent hemorrhages and inflammatory foci of lymphocytes, macrophages, and heterophylic granulocytes occurred diffusely in the capillary network (Fig. 8). Erythrocytes were present in the atria. With TEM no alterations were observed in the epithelium of the atrial walls. On the other hand many epithelial cells of the air capillaries were necrotic and desquamated, leaving a denuded basal lamina (Fig. 6B). This basal lamina was sometimes partly covered by cytoplasmic extensions of recruited macrophages. Focally, perivascular and interstitial edema were seen. In the animals exposed to 3.0 mg/m 3 03, the effects were clearly more severe. The cilia in the airways disappeared almost completely. Hypertrophy of the epithelium of the secondary bronchi was present and the lumina contained mucus
44
ROMBOUT ET AL.
FIG. 2. (a) Cross sections of tertiary bronchi, in-between air capillary network is present. The bronchi are circularly lined by atria separated from each other by thin septa. Control quail, LM, 80 ×. (b) The tertiary bronchi are closed by hypertrophied smooth muscle ceils. The air capillary network shows extensive hemorrhages. 3 mg/m 3 03 for 5 days, LM, 80×.
and erythrocytes. Hemorrhages, squamous metaplasia of the atrial epithelium with abundant mitoses, and a hypertrophy of the underlying smooth muscle cells resulted in complete closure of almost all atria (Figs. 2b and 9). Focally the atrial epithelium was necrotic, leaving smooth muscle cells exposed to 03. The capillary network of nearly the whole lung was hemorrhagic to such an extent that the lumina of tertiary bronchi were filled with massive conglomerations of erythrocytes. No histological differences were observed in the heart, liver; kidneys, spleen, and bursa of Fabricius between the animals of the control group and the group exposed to 3.0 mg/m 3 03. The results of the biochemical analysis are summarized in Table 1. Statistically significant pulmonary biochemical alterations were only observed in the 3.0 mg/ m 3 03 group. In this group G6PDH, GSH-Red, and L D H activities, expressed per whole lung tissue or per gram protein, were significantly increased. However, expressed per gram lung tissue the GSH-Px activity was decreased and the L D H activity was increased. DISCUSSION The trachea, secondary bronchi, and the entire parabronchial lung of Japanese quail are affected by a continuous 7-day exposure to 1.0 or 3.0 mg/m 3 03. The air-blood capillary network appears to be extremely vulnerable. Massive hemorrhages occur in the air-blood capillary network after exposure to 1.0 mg/m 3 03 for
OZONE TOXICITY IN AVIAN LUNG
45
FIG. 3. Longitudinal section of atrial septum. At both sides flat atrial wall cells are present with lamellar bodies in their cytoplasm. The cells are covered with microvilli. Control quail, TEM, 5400 ×. FIG. 4. Top of a longitudinally sectioned atrial septum. Smooth muscle cells are situated in the center. Thin cytoplasmic extensions of atrial wall cells form the epithelial lining. Control quail, TEM, 5400x.
7 days, indicating damage to the endothelium. This hemorrhagic response to 03 exposure is potentially life threatening at the 3.0 mg/m 3 03 level since the gas exchange capacity is jeopardized by the obstruction of the airflow in many atria. The small diameter of the air capillaries (3-10 Ixm) facilitates the complete closure by exudate and blood cells. This may lead to a higher airflow to other still open airways possibly resulting in a higher 03 dose to these airways and the surrounding gas exchange tissue, ultimately resulting in total lung failure with ongoing exposure. Since the quail were at rest, the effects registrated at 3.0 mg/m 3 03 were less than would be expected when the birds are very active. The avian respiratory system differs fundamentally from the mammalian lung. In birds, no tree-like branching pattern of the airways terminating in alveoli exists as in mammals; rather, the system is composed of two separate components, the lung itself and the air sacs, which allow the volume changes required for tidal ventilation (Scheid, 1979). The avian lung is composed of an interconnecting network of three generations of bronchi with numerous air capillaries. The air sacs can be considered as bellows which provide a unidirectional ventilatory flow to the otherwise relatively rigid parabronchial lung. The gas exchange is very efficient due to the unidirectional ventilatory flow, the crosscurrent flow system, and the relative continuous high oxygen concentration. The air capillaries are lined by epithelial cells with very large surfaces. Evans et al. (1985) concluded from rat studies that tolerance to 03 is absent in those
46
ROMBOUT ET AL.
FIG. 5. (a) View from a tertiary bronchus into atria and deeper lying infundibula. Septa (arrowheads) separate the various atria from each other. Control quail, SEM, 260 ×. (b) Two tertiary bronchi with atria, closed by hemorrhages and hypertrophied smooth muscle cells. Also the capillary network in between shows hemorrhages in contrast with the open capillary network in a control quail (Fig. 5a). 3 mg/m 3 03, SEM, 100×.
pneumocytes whose surface area, exposed to 03, is too large for the biochemical mechanism within the cell to protect it. Tolerance to 03 is reestablished by a decreased surface area of type I cells. The air capillary epithelial cells of Japanese quail, the equivalent of the mammalian type I pneumocyte, have a large surface area to volume ratio and appear to have a low metabolic activity. In a histochemical study of many enzymes in different regions of the chicken lung, the exchange area, mainly consisting of endothelial and air capillary epithelial cells, shows very weak enzyme activities (e.g., LDH and G6PDH) in contrast to strong activities in the bronchi and atrial epithelium (Tyler and Pearse, 1966). In mammalian species necrosis of the epithelial type I cell after 03 exposure is followed by a repair phase and proliferation of the interposed type II cell. The onset of injury and consecutive repair can be observed from Day 2 onward (Plopper et al., 1979; Evans et al., 1985). The air capillary epithelial cell appears to be the only cell type covering the avian air capillaries (MacDonald, 1970; Maina, 1987) as also observed in this study. Mitoses following damage of air capillary epithelium appear to be absent. After 7 days of exposure a denuded basal lamina is still observed, indicating the absence of repair processes. More research is needed to elucidate these processes. The atrial epithelial cells are often compared with the mammalian type II pneumocyte, because of the morphological similarities of microvillous structures and lamellar bodies. But there are some striking differences. While type II cells are interposed between type I cells in the gas exchange region, the atrial epithe-
OZONE TOXICITY IN AVIAN LUNG
47
FIG. 6. (A) Air capillaries (a) are alternately situated between erythrocyte-containing capillaries (b). Control quail, TEM, 3200 ×. (B) All air capillaries are devoid of epithelial cells (arrowheads). In the left upper corner a pyknotic epithelial cell is present, continuous in space with the interstitium, filled with lymphocytes and macrophages. I mg/m 3 03, TEM, 3500x.
lial cells are strictly located in the air conducting area. After exposure to 3 mg/m 3 03, but not after exposure to 1 mg/m 3 03, there is a proliferation of atrial epithelial cells, but this appears to have no effect on reepithelization of the air capillary spaces. Remarkably, the activities of LDH, G6PDH, and GSH-Px in control quail lung (this study) and control rat lung (Rombout et al., 1986b; Rietjens et al., 1985) are very comparable. Mammalian species exposed to oxidant gases respond by increased activities of antioxidant enzymes, probably originating from the result of shifts in lung cell populations and in intracellular biochemical changes. No biochemical reaction could be observed in the quail lung after continuous exposure to 0.3 or 1.0 mg/m 3 03, in contrast to increased enzyme activities in rats after 7 days of exposure to 0.3-0.8 mg/m 3 03 (Mustafa and Tierney, 1978). The GSH-Px activity, expressed in units per gram of lung, was significantly decreased in the 3.0 mg/m 3 group by an increase in lung weight. The elevated enzyme activities, as observed following exposure to 3 mg/m 3 03, are difficult to explain and are possibly due to the massive influx of erythrocytes and inflammatory cells as well as to the cellular changes in the atrial region (Tyler and Pearse, 1966). Assuming that an increase of antioxidant enzyme activities can be regarded as an adaptive response following oxidant exposure, the lack of effect up to 1.0 mg/m 3 03 exposure, as observed in this study, supposes the limited ability of the quail lung to respond adequately in this respect, thereby rendering this tissue more susceptible compared to mammalian lung.
48
ROMBOUT ET AL.
FIG. 7. Detail of the air-blood barrier (a-b). A thin air capillary epithelial cell with a nucleus is separated from a thin endothelial cell by a basal lamina. Control quail, TEM, 8600x.
TABLE 1 EFFECTS OF CONTINUOUS OZONE EXPOSURE FOR 7 DAYS ON BODY WEIGHT, LUNG WEIGHT, AND LUNG ANTIOXIDANT ENZYMES OF JAPANESE QUAILS mg/m 3 0 3 Body weight (g) Relative lung weight (mg/g bw) Protein (rag/lung) Protein (mg/g lung) G6PDH units/lung units/g lung units/g protein GSH-Px units/lung units/g lung units/g protein GSH-Red units/lung units/g lung units/g protein LDH units/lung units/g lung units/g protein
0 118 0.75 67,1 72.8
-+ 20 ~ -+ 0.09 -+ 11.2 -+ 8.0
0.3 114 0.84 75.5 76.5
_+ 6 -+ 0.11 -+ 12.1 -+ 4.0
1.0 109 0.87 59.9 73.2
-+ 24 _+ 0.20 _+ 12.7 -+ 8.6
3.0 114 1.20 73.8 53.6
-+ 8 + 0.14"** _+ 8.6 _+ 7.1
1.00 _+ 0.29 1.12 _+ 0.41 15.8 _+ 6.6
0.94 -+ 0.54 1.00 -+ 0.71 13.4 + 10.1
0.75 _+ 0.10 0.95 _+ 0.24 13.0 _+ 2.9
3.44 -+ 2.16" 2.47 -+ 1.58 44.9 -+ 23.4*
10.5 -+ 2.7 11.4 _+ 2.1 156 _+ 26
10.3 -+ 2.1 10.4 -+ 1.3 136 -+ 16
8.3 -+ 2.5 10.0 - 1.9 137 -+ 18
10.8 _+ 2.4 7.7 -+ 1.8' 145 -+ 16
1.27 _+ 0.24 1.38 +_ 0.25 19.1 + 3.5
1.38 -+ 0.18 1.41 +- 0.12 18.5 +- 2.1
1.10 -+ 0.22 1.34 -+ 0.14 18.6 -+ 3.1
2.61 _+ 0.42*** 1.89 _+ 0.40 35.3 _+ 3.6***
50.5 -+ 6.8 55.8 -+ 11.7 775 -+ 189
53.1 -+ 7.5 54.4 _+ 8.0 715 + 134
38.9 _+ 12.2 46.9 _+ 7.0 644 _+ 91
122 -+ 20*** 88 + 18.2" 1650 -+ 13"**
a Each value represents mean + SE of five animals (n = 4 for the 3 mg/m3 03 group). * P < 0.05; **P < 0.01; ***P < 0.001.
OZONE TOXICITY IN AVIAN LUNG
49
Fro. 8. Hemorrhages and influxes of inflammatory cells in the interstitium and air capillary space nearby atria. 1 mg/m 3 03, LM, 200x.
FIG. 9. Atrial wall epithelium shows metaplasia (arrowheads). The underlying capillary network is filled with inflammatory cells and erythrocytes. 3 mg/m 3 03, LM, 200x.
50
ROMBOUT ET AL.
Brown et al. (1982) described the resistance of chickens to acute pulmonary injury by O2, paraquat, and silica. Pulmonary lavage failed to reveal an influx of inflammatory cells, and no pulmonary edema was detected. However, in our experiments inflammatory cells were present both interstitially and inside the atrial lumina and air capillaries. Figures 10-12 in the article of Brown et al., however, reveal that the atrial lumina display an influx of inflammatory cells. To what extent a comparison is possible between a pulmonary lavage in chicken and in mammals is not clear. In conclusion, continuous exposure to an ambient concentration of 03 induced damage in the gas exchange area of quail lungs. Together with the apparent lack of adequate morphological and biochemical repair capacity, it gives rise to serious concern about potential health effects on an important order of animals in our environment. In particular birds with a high metabolic rate that are flying at rather high altitudes for a considerable part of the day might be at risk. More research concerning the vulnerability of birds to air pollution and the mechanism of damage and repair of their respiratory tract is inevitable to predict the effect of air pollution on various ecosystems. ACKNOWLEDGMENT The authors are grateful for the expert technical assistance of G. Riool-Nesselaar, J. Bos, and A. J. F. Boere.
REFERENCES Brown, Ch., Pratt, Ph. C., and Lynn, W. S. (1982). Characterization of pulmonary cellular influx differentials to known toxic agents between species. Inflammation 6, 327-341. Evans, M. J., Dekker, N. P., Cabral-Anderson, L. J., and Shami, S. G. (1985). Morphological basis of tolerance to ozone. Exp. Mol. Pathol. 42, 366-376. Krause, G. H. M., and Prinz, B. (1989). Current knowledge of ozone on vegetation forest effects and emerging issues. In "Atmospheric Ozone Research and Its Policy Implications" (T. Schneider, S. D. Lee, G. J. R. Wolters, and L. D. Grant, Eds.), pp. 45-55. Elsevier, Amsterdam. Lasiewski, R. C. (1972). Respiratory function in birds. In "Avian Biology" (D. S. Farner and J. R. King, Eds.), Vol. II, pp. 287-335. Academic Press, New York. Lawrence, R. A., and Burk, R. F. (1976). Glutathione peroxidase activity in selenium deficient rat liver. Biochem. Biophys. Res. Commun. 71, 952-958. Lippmann, M. (1989). Health effects of ozone: A critical review. J. Air Pollut. Control Assoc. 39, 672-695. Lowry, O. H., Rosenbrough, N. J., Farr, A. L., and Randall, R. J. (1951). Protein measurement with the Folin-phenol reagent. J. Biol. Chem. 193, 265-275. MacDonald, J. W. (1970). Observations on the histology of the lung of Gallus domesticus. Brit. Vet. J. 126, 89-93. Maina, J. N. (1987). Morphometrics of the avian lung. 4. The structural design of the charadriiform lung. Respir. Physiol. 68, 9%120. Maina, J. N., and King, A. S. (1982). Morphometrics of the avian lung. 2. The wild mallard (Arias Platyrhynchis) and graylag goose (Anser anser). Respir. Physiol. 50, 299-310. Maina, J. N., and King, A. S. (1984). Correlation between structure and function in the design of the bat lung: A morphometric study. J. Exp. Biol. 111, 43-61. Maina, J. N., and Settle, G. (1982). Allometric comparison of some morphometric parameters of avian and mammalian lungs. J. Physiol. (London) 330. Marra, M., and Rombout, P. J. A. (1990). Design and performance of an inhalation chamber for exposing laboratory animals to oxidant air pollutants. Inhalation Toxicol. 2, 187-204.
OZONE TOXICITY IN AVIAN LUNG
51
Mustafa, M. G., Hacker, M. D., Hussain, M. Z., and Lee, S. D. (1977). Biochemical effects of environmental oxidant pollutants in animal lungs. In "Biochemical Effects of Environmental Pollutants" (S. D. Lee, Ed.), pp. 59-96. Ann Arbor Science, Ann Arbor, MI. Mustafa, M. G., and Tierney, D. F. (1978). Biochemical and metabolic changes in the lung with oxygen, ozone and nitrogen dioxide toxicity. Amer. Rev. Respir. Dis. 118, 1061-1090. Plopper, C. G., Chow, C. K., Dungworth, D. L., and Tyler, W. S. (1979). Pulmonary alterations in rats exposed to 0.2 ppm and 0.1 ppm ozone: A correlated morphological and biochemical study. Arch. Environ. Health 34, 390-395. Quilligan, J. J., Boche, R. D., Falk, H., and Kotin, P. (1958). The toxicity of ozone for young chicks. A.M.A. Arch. Ind. Health 18, 16-22. Racker, E. (1955). Glutathione reductase (liver and yeast). In "Methods in Enzymology" (S. P. Colowick and N. O. Kaplan, Eds.), Vol. 2, pp. 722-725. Academic Press, New York. Rietjens, I. M. C. M., Bree, L. van, Marra, M., Poelen, M. C. M., Rombout, P. J. A., and Alink, G. M. (1985). Glutathione pathway enzyme activities and the ozone sensitivity of lung cell populations derived from ozone exposed rats. Toxicology 37, 205-214. Rombout, P. J. A., Lioy, P. J., and Goldstein, B. D. (1986a). Rationale for an eight hour ozone standard. J. Air Pollut. Control Assoc. 36, 913-917. Rombout, P. J. A.; Van Bree, L., Heisterkamp, S. H., and Marra, M. (1986b). Towards an exposureeffect model for short term ozone exposure of rats. In "7th World Clean Air Congress, Sydney 1986" (H. F. Hartman, Ed.), pp. 203-211. Holmes, Melbourne. Scheid, P. (1979). Mechanism of gas exchange in bird lungs. Rev. Physiol. Biochem. Pharmacol. 86, 137-186. Tonneijck, A. E. G. (1989). Evaluation of ozone effects on vegetation in the Netherlands. In "Atmospheric Ozone Research and Its Policy Implications" (T. Schneider, S. D. Lee, G. J. R. Wolters and L. D. Grant, Eds.), pp. 251-260. Elsevier Science Publ., Amsterdam. Tyler, W. S., and Pearse, A. G. E. (1966). Functional and analytical histochemistry of the chicken lung lobule with particular reference to surfactant. Poult. Sci. 45, 501-511. Van Bree, L., Haagsman, H. P., Van Golde, L. M. G., and Rombout, P. J. A. (1988). Phosphatidylcholine synthesis in isolated type II pneumocytes from ozone-exposed rats. Arch Toxieol. 61, 224-228. Wolff, G. T., Lioy, P. J., and Taylor, R. S. (1987). The diurnal variations of ozone at different altitudes on a rural mountain in the eastern United States. J. Air PoUut. Control Assoc. 37, 45--48. World Health Organization. (1987). "Air Quality Guidelines for Europe: Ozone and Other Photochemical Oxidants," pp. 315-326. Copenhagen.
