Animal model of grain worker's lung.

We examined the light microscopic changes in the lungs of rabbits exposed to grain dust for variable periods of time, to determine whether an animal model of grain worker's lung could be developed. Experimental animals were exposed to grain dust at a concentration of 20 mg/m3 for 7 hr/day, 5 days/week, for up to 6 months. The lungs of these rabbits demonstrated a granulomatous interstitial pneumonitis associated with exudation of mononuclear cells into the alveoli and conducting airways. These changes appeared within 5 days of the onset of exposure and reached a peak at 3 weeks but were sustained through the longest exposure interval. No abnormalities were observed in the lungs of control rabbits. These results show three points of consistency with those obtained in epidemiologic studies of grain elevator workers. First, the rapid appearance of the experimental changes suggests that the mechanism of tissue injury may not be immunologic. Second, the occurrence of the histopathologic alterations in the interstitium, alveoli, and airways corresponds with the combined restrictive and obstructive ventilatory defect described in the human epidemiologic studies. Third, the absence of lung fibrosis in rabbits exposed to dust for 6 months suggests that the pneumonitis is reversible. Thus this experimental model shows promise of helping to clarify the nature and mechanism of the adverse pulmonary effects of grain dust.


Introduction
A number of epidemiologic studies have shown that employment as a grain elevator worker is associated with a variable increase in respiratory symptoms and an adverse effect on pulmonary function (1)(2)(3)(4)(5). The pulmonary function changes are both acute and chronic and show exposure-response relationships with the level of respirable dust (6). The respiratory symptoms and pulmonary function changes develop within the initial 3 months of employment (7) and improve over a 2-month period of lay-off (8).
There have been occasional reports of grain workers who were allergic to grain dust (9). However, allergy or hypersensitivity would not appear to explain the collective abnormalities detected in epidemiologic studies (10,11). On the other hand, there have been two reports of a generalized increase in serum a-1-antitrypsin levels in grain elevator workers (11,12). This observation, together with the apparent absence of an immunologic mechanism, has been interpreted as suggesting that the pulmonary abnormalities that occur in this occupational group may be due to a nonimmunologic inflammatory process (11).
Grain elevator dust and grainAdust extracts have been shown both to activate complement (13,14) and to form a nonimmunologic precipitate with human IgG (14). These properties suggest the basis for a possible nonimmunologic inflammatory process which may be activated when grain dust is inhaled into the lung. An animal model would provide the potential to characterize further the nature and mechanism ofthe adverse pulmonary effects of grain dust. Mice exposed to 1.2-5.4 g/m3 of grain dust for 8-24 hr/day, 5 days/week for up to 4 months have been reported to develop an increased number of macrophages in their airways (15). However, rats exposed to 60 mg/m3 of grain dust for 2 hr/day for up to 2 months, demonstrated a granulomatous interstitial pneumonitis (16). The concentration of grain dust used in these studies would be considered extreme by usual industry standards (6,10,17).
The following report describes an animal model of grain worker's lung in rabbits exposed to grain dust under conditions that more closely approximate those in the contemporary grain industry. Light microscopic examination of the lungs of exposed rabbits shows an inflammatory process in both the airways, alveoli, and interstitium that is well developed after only 5 days, reaches a peak at 3 weeks, and is still present at 6 months. The results obtained in this model show a number of features which parallel those reported in epidemiologic studies of grain elevator workers.

Rabbits
New Zealand White rabbits weighing 2 kg were obtained in a single batch from one supplier. The controls were housed continuously in the animal quarters, while the experimentals were moved into a separate room on the days when they were exposed to grain dust.

Dust Exposure Chambers
Two exposure chambers were used which consisted of a one cubic meter working area supplied with grain dust from a fluidized bed. The chamber has been described in detail previously (18). The dust delivery rate was adjusted to maintain a dust level of about 20 mg/m3 in the chambers.

Grain Dust
A single batch of grain dust was used in these studies. This was obtained from the air exhaust collectors of two large grain elevators in Thunder Bay, Canada (Saskatchewan Wheat Pool 4 and Pbol 7). These elevators process several species of grain, often simultaneously; they handle wheat predominantly and to a lesser extent oats, rye, and barley. The stock dust was thoroughly mixed and stored at 4°C. Prior to use, the largest dust particles were removed on a 65 mesh sieve.
The level of grain dust in each chamber was measured using a Bendix 44 pump and a Millipore filter cassette holding a 37-mm polytetrafluoroethylene filter (Millipore, Flouropore, 1.0-pRm pore size). The filters were desiccated and weighed in a standard rmanner before and after use (6). The chambers were monitorek1 during most days when the rabbits were exposed, and a time weighted average was obtained covering the full 7-hr period. The airborne dust in the chambers was further characterized on several occasions using an Andersen cascade impactor (19).

Experimental Design
A total of 47 rabbits were employed, of which 23 were controls and 24 experimentals. Both groups were subdivided into five subgroups containing four or more animals each. Animals in excess of four per subgroup were included to allow for premature deaths. The experimental animals were exposed to 20 mg/m3 of grain dust for 7 hr/day, 5 days/week, for a duration of 1, 3, 6, 12, or 24 weeks. Exposures were staggered so that all rabbits were approximately the same age at the time of sacrifice. Four control animals were sacrificed on each occasion that a subgroup of experimental rabbits reached the end of their exposure period. This study was approved by the Animal Care Committee of the University of Toronto.

Histopathology
The rabbits were euthanized with an overdose of sodium nembutol. The trachea was ligated and a catheter tied in proximally that was connected to a water manometer. Air was injected into the trachea to maintain a pressure of 20 cm H20. The chest cavity was opened and catheters tied into the right atrium and aorta. The pulmonary artery was perfused free of blood with cold heparinized saline and then with cold universal fixative. The trachea, right atrium, and aorta were tied off, and the heart and lungs removed and left in universal fixative overnight, and then in 10% buffered formalin. Tissue sections were prepared for light microscopy from lungs of both the control and experimental rabbits. These were obtained from blocks which contained both central and peripheral portions of lung in a single section. The sections were stained using hematoxylin and eosin.

Chamber Conditions
The average monthly temperature in the chambers ranged from 16 to 24°C, with a daily minimum of 10 and maximum of 27 (Table 1). The average monthly total dust concentration was similar in both chambers and approximated 20 mg/m3. The lowest dust concentration on any day was 15.4 and the highest 29.3. The dust inside the chamber was analyzed using an Andersen cascade impactor on three occasions at monthly intervals. About 45% of the particles were greater than 9 ,um in size while the majority of the remainder were between 3.3 and 5.8 ,um and a small proportion were 1.1 ,m or smaller ( Table  2).

Rabbit Weights and Survival
The initial weights of the experimental and control rabbits were comparable (Table 3). Both groups gained to about the same degree and approximately doubled their weight over the experimental period. A total of three animals died spontaneously during the course of the experiment, all of which were controls.

Histopathology
Sections of lung from the control and experimental rabbits were examined by light microscopy. All experimental animals demonstrated histopathologic changes after 5 days of exposure to grain dust. The extent of the tissue involvement reached a peak at 3 weeks and was sustained through the longest exposure interval of 24 weeks. The changes consisted of an interstitial pneumonitis that extended into the walls of some blood vessels and airways (Fig. 1). The interstitial pneumonitis was patchy but widespread and was characterized by vascular congestion, edema, and a cellular infiltrate composed predominantly of mononuclear and giant cells often forming granulomata. Neutrophils and eosinophils also were evident, more so within the initial 3 to 6 weeks. In addition, there was extensive cellular exudation into the alveolar spaces and conducting airways. The cellular exudate present in the airways was comparable to the tissue infiltrate. The lungs of the control rabbits showed no abnormalities (Fig.   2).

Additional Studies
In additional studies, the histopathologic changes in the lungs have been examined in greater detail, including the use of scanning and transmission electron microscopy. The histopathologic changes seen in the lungs of the 5 2.1-3.3 11.4 8.8 9.9 exposed rabbits were characterized by an interstitial 6 1.  (21) and also to a hypersensitivity pneumonitis 8 0.4-0.7 (22), but the rapidity of onset seems to weigh strongly against the involvement of an immunological mechanism. There was no indication even after 5 days of exposure of Also, the pulmonary function of the control and experia predominantly eosinophilic response suggestive of a mental rabbits has been measured at the beginning and type 1 allergic reaction (23). Also, the acute inflammation end of the study period. Further, bronchoalveolar lavage and hemorrhagic necrosis of a type 3 reaction was not has been performed at the end of the experiment, and seen at any time point (24).

the cells obtained have been characterized both mor-
We and others have hypothesized that the adverse pulphologically and functionally by their phagocytic activity monary effects of inhaled grain dust may be due to the and migration out of capillary tubes. These results will known complement activating properties of this material be published separately. (11,25). Nevertheless, the histopathologic changes seen in our model did not resemble the acute changes described Discussion when a complement derived chemotactic factor has been instilled intratracheally in rabbits (26). This difference The daily and weekly exposure intervals utilized in the does not necessarily exclude the possible role of compleforegoing study were selected to represent those of terment in the response to grain dust, since a chronic, difminal grain elevator workers. The concentration of grain, fuse effect of low levels of complement activation may . Light microscopy at magnification of 320 x of lung section obtained from rabbit exposed to grain dust for 3 weeks. differ from that of a single instillation of a large bolus. The response to grain dust may be conditioned additionally by the simultaneous presence of particulates. However, our results cannot be entirely explained on this basis, since the effect of particulates alone tends to be manifested by macrophage exudation into the air spaces. and sometimes by the evolution of fibrosis (27). The morphology of the histopathologic lung changes in our exposed rabbits was indicative of an inflammatory response in both the pulmonary airways and the interstitium. This is consistent with our epidemiologic findings in grain elevator workers, which showed pulmonary function changes suggestive of both an obstructive and restrictive ventilatory defect (6).
The experimental pneumonitis observed in our rabbits was not associated with obvious fibrosis by light microscopy. However, examination of sections by polarized light or using the picrosirius stain showed a small increase in the amount of fibrous tissue in some alveolar septae. This slight increase in fibrous tissue would be consistent with our demonstration that the adverse effects of chronic grain dust exposure in humans are at least partially reversible after a period of absence from exposure (8).
The consistency of our findings in rabbits and humans chronically exposed to grain dust suggests that the experimental model has the potential to provide clinically meaningful information about the nature and mechanism of the adverse pulmonary effects of grain dust. This work was supported by grants from Physicians' Services Incorporated Foundation and the Ontario Thoracic Society