- Email: [email protected]
Light and immunofluorescent study of the Arthus reaction in the rabbit lung
Light and immunofluorescent study of the Arthus reaction in the rabbit lung Donald C. Zavala, and Roger Oskvig,
M.D., Mitchell 1. Rhodes, M.D. Iowa City, Iowa
M.D.,
Hal B. Richerson,
M.D.,
A localized Arthus reaction was produced in the lwng of sensitized rabbits by delivery of antigen into a lower lobe broncbns rising a method of selective bronchial catheterization under &oroscopy. The rabbits were sensitized with bovine immwnoglobulin G (B-IgG) in incomplete Freund’s adjnvant (IFA) to prodwe precipitating antibody without classic delayed hypersensitivity. Pulmonary histopathology was stndiea at intervals following antigen challenge, rising light and immnno@orescent microscopy. Gross lesions peripheral to the lower lobe bronohm receiving antigen were fozlnd within 18 hr. Subsequent necrosis redted in a dense scar by 6 wk. Microscopically, early lesians were typified by localized bronchitis, bronehiolitis, alveolitis, and vasculitis with exuberant exzcdates containing predominantly polymorphonzlclear leukocytes. Extensive focal necrosis was present by 78 hr. Immnnofiuorescent studies revealed the presence of B-IgG, rabbit IgG, and complement (C3) in and around bronchi, bronchioles, alveoli, and vessels. No grannlomatozls lesions were foma, and proliferation of alveolar lining cells ~a.3 not detected in these studies. Thus, the lung can participate in an acute Arthur reaction following local antigen challenge in systemically sensitiaed animals. The pathology more closely resembles a necrotizing bacterial pneumonia than an interstitial or hypersensitivity pneumonitis under the conditions of this experimental system. Implications for human disease are speculative.
Various experimental models of hypersensitivity pneumonitis in rabbits have been developed to study the inflammatory reaction of the immune animal to antigen and its relation to antibodies. A number of investigators including 0pie.l Fried,z and Cannon and associates3have produced a localized pneumonitis, bronchitis, and vasculitis in sensitized rabbits by instillation of horse serum or egg albumin directly into the lung or intratracheally. The effects of rabbit antiratlung antibodies have been studied by Read,4 Nagaya, Schauble, and Sicker,” Hagadorn, Vasquez, and Kinney,6 and Willoughby and Dixon7 by intracheal or intravenous routes of challenge. Bensch, Dominquez, and Liebow reported on the absorption of antigenically intact protein molecules across the pulmonary air-tissue barrier into the circulation of dogs after instillation into the trachea. Recently, Richerson, Cheng, and Bauserman,s, lo produced an acute allergic alveolitis, peribronchiolitis, and vasculitis in sensitized rabbits using aerosolized ovalbumin. From the Pulmonary and Allergy-Immunology Divisions, Department of Medicine, of Iowa College of Medicine. Supported by Grant No. 5598 from the American Lung Association. Received for publication July 26, 1974. Accepted for publication Nov. 19, 1974. Reprint requests to: Dr. D. C. Zavala, Department of Internal Medicine, University Hospitals, Iowa City, Iowa 52242. F-01. 56, No.
6, pp. 450-463
University
of
Iowa
VOLUME NUMBER
56 6
Arthus
reaction
in rabbit
lung
451
The purpose of this report is to describe the pulmonary pathologic changes occurring in a type III (Arthus) immunologic reaction. The lesions were produced by intrabronchial challenge in rabbits sensitized with bovine-IgG (B-IgG) in incomplete Freund’s adjuvant (IFA). A unique feature of the experiment was the utilization of a nonsurgical rabbit model of selective bronchial catheterization. None of the previous animal models delivered the antigen directly into a selected peripheral segment of the lung without highly sophisticated equipment, inhalation anesthesia, aerosolization, tracheostomy, intratracheal instillation, or transthoracic puncture. MATERIALS
AND
METHODS
A total of 54 male and female New Zealand white rabbits, weighing 2.7 to 3.6 kg, were used. In the study, 30 rabbits were immunized by injection with 1 mg of B-IgG in 0.15 M sodium chloride emulsified in IFA volume of 0.2 ml, distributed among the 4 footpads. Three weeks after immunization a subcutaneous booster injection consisting of 1 mg of antigen emulsified in an equal part of IFA was given at 4 injection sites (each shoulder), for a total of 4 mg antigen per animal. Skin tests for sensitizing activity and blood sampling for antibody formation were carried out on the twenty-first day following antigen footpad injections and then repeated 1 wk after booster injections. The animals were skin-tested with 1 mg B-IgG in 0.1 ml saline injected intradermally into closely shaved flank skin. Precipitins were measured using Ouchterlony’s method of double immunodiffusion in agar gel.11 One to two weeks after the booster injection, the rabbits mere challenged by instillation of 2 mg of antigen in 0.1 ml saline solution into one of the segmental bronchi of either lower lobe. Selective bronchial catheterization was performed as described previously.12 Each rabbit was prepared by an intramuscular injection of 0.5 ml of Innovar-Vet and secured on a peg board. After spraying the oropharynx and larynx with 4% lidoeaine, a flexible steel guide mire was inserted under direct vision through the vocal cords, and a small plastic tube passed over the guide wire into the trachea. The wire was removed leaving the endotraeheal tube in position. Bronchial catheterization was carried out by inserting a tiny polyethylene radiopaque catheter through the tracheal tube and guiding it under fluoroscopic control to the desired location in the lungs where the antigen was deposited. Following challenge, 3 rabbits were sacrificed with intravenous pentobarbital (150 mg) at 6, 12, 24, 48, 72 hr and 1, 2, 3, 4, and 6 wk. The lungs were removed in toto, inflated to approximate in situ size with 10% buffered formalin delivered into the trachea, and placed in a jar of formalin. After fixation of lung tissue, traverse sections were cut from each lobe, including the lesion produced at the site of intrabronchial challenge, placed in fresh buffered formalin, and prepared for light microscopy. In addition, sections were taken from the heart, kidney, liver, and spleen. Paraffin-embedded tissue was cut in 5 p sections and then stained with hematoxylin and eosin. Sections of fresh lung tissue from challenge and control sites also were studied by fluorescent antibody technique, as described by Beutner and Nisengard.13 Tissues taken from 2 IFA rabbits at 6, 24, 48, 72 hr, and 1 wk postchallenge were unsnap” frozen in liquid nitrogen, sectioned on a cryostat, and prepared using fluorescein-conjugated antirabbit.IgG, antibovine*IgG, and antiC3, each diluted 1:30.* The sections were viewed on a Leitz Ortholux II fluorescence microscope using heat and suppression filters. Appropriate pictures were taken on Polaroid and color slide film. Twelve control animals were used consisting of 3 unimmunized rabbits challenged with B-IgG, 6 B-IgG immunized rabbits challenged with different protein antigens (ovalbumin and human serum albumin, 3 each), and 3 B-IgG immunized, unchallenged rabbits. *The anti-R-IgG had an antibody protein concentration of 16.8 mg/ml the anti-BeIgG had a protein concentration of 3.5 mg/ml (Miles anti-C3 had a protein concentration of 1.3 mg/ml (Meloy) .
(Miles Laboratories) Laboratories), and
; the
452
Zavala
FIG. 1. intact Arrow
denotes
J. ALLERGY
et al.
lung removed site of lesion
from the IFA-sensitized which resembles an
rabbit infarct.
72 hr after
CLIN. IMMUNOL. DECEMBER 1975
antigen
challenge.
RESULTS Skin tests
Skin reactivity was used for the selection of experimental animals. Intracutaneous injections of 1 mg B-IgG produced equally strong skin reactions in 30 IFA-immunized rabbits when carried out 3 wk after the footpad injections and repeated 1 wk after the booster injections. Additional animals with poor or equivocal results were excluded from the study. Strongly positive reactions were present at 6 hr and were progressive to 48 hr, at which time the lesions consisted of dark hemorrhagic centers measuring 0.5 to 0.8 cm, surrounded by edematous, erythematous halos, 3.0 to 3.5 cm in diameter. The lesions were unchanged or slightly decreased at 72 hr after which the halos gradually subsided but the centers became necrotic. Unsensitized rabbits also were skintested with the same dose of antigen (1 mg B-IgG) and showed no skin irritation. Precipitins
Precipitins were demonstrated in the immune sera (postimmunization and postbooster) of all experimental animals used in this study to a dilution of 1:4 against a concentration of 0.1 mg/ml of antigen in the center well. Pathology
Intrabronchial challenge of the sensitized rabbits produced the following pathologic changes. At autopsy, gross findings were apparent after 12 hr, with the initial lesions appearing around one of the lower lobe basal bronchi
VOLUME NUMBER
56 6
Arthus
FIG. 2. Histopathology intense inflammatory A large
necrotic
of rabbit (PMNs) and area
is located
lung round nearby,
72
hr cell
not
postchallenge. infiltration seen
in these
reaction
involving
A vasculitis arteries
views.
(x200.)
in
rabbit
is
present (A) and
lung
with veins
453
an (8).
454
Zovala
et
al.
FIG. 3. Histopathology of rabbit lung 7 hr postchallenge. An acute with thickened septa, cellular infiltrates, and exudate in aveolar magnification in B shows intro-alveolar macrophages, red blood cells. (x500.) In C, the arrow points to a bizarre, multinucleated alveolar space. (x500.)
J. ALLERGY
CLIN. IMMUNOL. DECEMBER 1975
alveolitis is shown in A spaces. (x200.) Higher ceils, PMNs, and round giant cell located in the
VOLUME NUMBER
Arthus
56 6
FIG.
3C.
For
legend
see opposite
reaction
in
rabbit
lung
455
page.
and its branches that received the antigen. Within 72 hr the lesion often resembled an infarct (Fig. 1). Usually a necrotic center developed that excavated by the third to fourth week postchallenge and then collapsed and formed a dense scar after the sixth week. Microscopically, the initial early reaction occurred in and around the bronchi, followed by spread into the surrounding lung. The first changes were seen at 6 hr postchallenge and consisted of a regional peribronchitis, perihronchiolitis, alveolitis, and vasculitis. The early cellular infiltrates were made up of the following: (1) focal collections of PMNs’ in the lumen and walls of bronchi and bronchioles, (2) a mixture of PMNs and lymphocytes around bronchi, bronchioles, and pulmonary vessels. (3) PMNs and red blood cells filling dilated alveolar capillaries, and (4) focal intra-alveolar exudation of PMNs and red blood cells. By 72 hr there was extensive focal necrosis and ulceration of the walls of the bronchi and bronchioles with extensive intraluminal exudation and occasional rupture. A predominantly round cell infiltrate surrounded the smaller airwayqand there was hypcrplasia of lymphoid tissue around many of the larger bronchi. A moderate to intense vasculitis, involving arteries and veins beyond the lesion. There were rxt,ensire perivascular (Fig. 2), did not extend infiltrates consisting of PMNs, lymphocytes with some large mononuclear cells, and plasma cells. An alvcolitis was clearly evident with thickened septa, heavy cellular infiltrates, and scattered fluid-exudate in the alveolar spaces (Fig. 3, “Polymorphonuelear leukocytes
that,
in the
rabbits,
have
cytoplasmic,
cosinophilic
granules.
456
Zavala
et al.
FIG. 4. Histopathology of rabbit lung 72 hr postchallenge. margin of the lesion is indicated by arrows. The challenged just out of the photograph. (x50.)
J. ALLERGY
CLIN. IMMUNOL. DECEMBER 1975
The line of demarcation bronchus is located
at the inferiorly
A). The intra-alveolar cells consisted of macrophages, red blood cells, PMNs, a few round cells, and an occasional multinucleated giant cell (Fig. 3, B and C) . A leading line of demarcation was observed at the outer margins of the lesions (Fig. 4)) possibly representing antigen-antibody interaction. The contralateral lung was normal except for filling of a moderate number of alveolar capillaries with eosinophilic granulocytes. By 1 wk closely grouped necrotic areas of variable sizes developed at the challenge sites (Fig. 5), which usually were evacuated by the third to fourth weeks. The resulting cavity then collapsed and, after the sixth week, formed a dense scar (Fig. 6). At no time did the animals appear ill. Microscopic studies of the heart, liver, spleen, and kidneys were normal on all animals. Halving the challenging dose of antigen to 1 mg B-IgG in 12 IFA-immunized rabbits resulted in lung lesions that were less extensive and intense. had completely clean lungs on light microscopy All of the control mimals except for an occasional small focus of intra-alveolar macrophages, a rare vessel surrounded with a few small round cells, and minimal 1,ymphoid collections in peribronchial and peribronchiolar locations. Immunofluorescent microscopy on control animals demonstrated a rare fluorescent phagocytic cell with antisera to both IgG and C3. The specificity of each conjugate was confirmed by immunoelectrophoresis. lmmunoglobulins deposited in animal tissue were identified by prior inhibition using unlabeled antirabbit IgG, antirabbit C3 antibody, and
VOLUME NUMBER
56 6
Arthus
FIG. 5. Histopathology Note proximity of the
of rabbit lung 1 wk postchallenge. lesion to a bronchiole. (x50.)
FIG. 6. Histopathology resulting in retraction
of rabbit lung 6 wk of the pleura. (x30.)
postchallenge.
reaction
A large
Arrow
in rabbit
necrotic
points
area
lung
457
is evident.
to a healed
scar,
458
FIG. seen
Zavala
et
J. ALLERGY
al.
7. lmmunoftuorescent in
the
bronchial
(A)
microscopy and alveolar
of
rabbit
(B) air
lung spaces.
at 6 hr (x100.)
postchallenge.
CLIN. IMMUNOL. DECEMBER 1975
Free
B-IgG
is
VOLUME NUMBER
56 6
FIG. 8. lmmunofluorescent seen in bronchi (A) and B-IgG. A x100; 8, x540.)
Arthus
reaction
in rabbit
microscopy of rabbit lung at 48 hr postchallenge. alveoli (B). Arrow points to an alveolar phagocytic
lung
459
8-IgG again cell containing
is
460
Zavala
J. ALLERGY
et al.
FIG. 9. lmmunofluorescent in peribronchial and vessel. (xl 00.)
microscopy perivascular
of rabbit lung at 6 hr postchallenge. areas. 6 indicates bronchus and V
CLIN. IMMUNOL. DECEMBER 1975
R-IgG indicates
is seen blood
antibovine IgG. Subsequently, incubation was again carried out with fluoresceinlabeled specific immunoglobulins. Immunofluorescent microscopy on sensitized rabbits demonstrated initiation of a localized Arthus reaction 6 hr after intrabronchial challenge with bovineIgG. Specimens at 6 and 48 hr showed localized B-IgG antigen, rabbit-IgG (R-IgG), and complement (C3). B-IgG was found in bronchial and alveolar air spaces (Fig. 7) and in phagocytic cells located in and around the airways and blood vessels (Fig. 8). R-IgG was present along with B-IgG (Figs. 9 and lo), but there was no R-IgG and B-IgG in bronchial, bronchiolar, or alveolar lining cells. At 48 hr, most of the complement (C3) was no longer free but appeared as brightly fluorescing granules in phagocytic cells, located in and around blood vessels and airways. Some free granular C3 was present in the alveolar spaces. At 1 wk, B-IgG and complement were no longer detectable, although phagocytic cells containing RIgG granules were still found scattered in lung parenchyma with some peribronchiolar and perivascular localization (Fig. 11). All of the various stages of this reaction were localized entirely at the challenge site. DISCUSSION
We have described the morphologic sensitized rabbits following intrabronchial
changes occurring in the lungs of challenge by selective catheterization.
VOLUME NUMBER
Arthus
56 6
FIG. 10. lmmunofluorescent present in the bronchi but
microscopy notably absent
reaction
of rabbit lung at 48 hr in the lining cells. [xl 00.)
in rabbit
postchallenge.
lung
R-IgG
461
is
The animals were immunized with B-IgG, emulsified in IFA. Circulating precipitins and positive skin tests were demonstrated in all animals. The experimental lesions began in 6 hr with regional peribronchiolit~is, alveolitis, and vasculitis and then rapidly progressed to focal necrosis in 72 hr and central necrosis in 1 wk. There was extensive intraluminal exudation in the bronchi, a mixed perivascular infiltration of PM& and lymphocytes, and an alveolitis with septal widening and cellular infiltrates. The necrotic core evacuated in 3 to 4 wk, leaving a cavity that soon collapsed and formed a residual scar. The rabbits were extremely tolerant of their localized immune lung injury and at no time did they appear ill. An apparent pulmonary Arthus reaction was shown in the sensit.ized rabbits by immunofluorescent microscopy. Antigen and complement were found in bronchial and alveolar air spaces and in alveolar macrophages. Bronchial, bronchiolar, and alveolar lining cells took no active part in this reaction. Apparently there was no proliferation of alveolar type II pneumocytes, as observed by light microscopy. In this study of experimental immune lung injury, a nonsurgical rabbit model using bronchial catheterization provided the advantage of delivering the antigen into a selected pulmonary segment. The remainder of the lung served as its own control. Using a similar technique of selective bronchial catheterization, Richerson
462
Zavala
J. ALLERGY
et al.
FIG. 11. lmmunofluorescent remains. Arrows denote and normal lung. (x100.)
microscopy margin of the
of rabbit lung at 1 wk postchallenge. reaction between fluorescent rabbit
CLIN. IMMUNOL. DECEMBER 1975
R-IgG alone immunoglobulin
and co-worke&* studied the histopathology of classic delayed hypersensitivity in the lung using purified protein derivative (PPD) . Although significant differences in tissue response are noted between their work and this study, there is a remarkable similarity in the absence of proliferation of type II alveolar lining cells (granular pneumocytes). Apparently the lung’s response in the Arthus reaction and delayed hypersensitivity differs from its response to other forms of injury (e.g., various drugs, oxygen, radiation), which are typified by proliferating, atypical alveolar pneumocytes. Any comparison with diseases of human lungs is speculative. Nevertheless, under the conditions of this experimental model the resulting tissue reaction pneumonia with a localized vasculitis rather more closely resembled a necrotizing than any of the hypersensitivity lung diseases or several types of necrotizing vasculitis which include Wegener’s granulomatosis, the eosinophilic vasculitis described by Churg and Straus.ql” giant cell arteritis, periarteritis nodosa, and systemic vasculitis.16 The resulting cavity healed like a lung abscess. No other nonchallenged areas of the organs were involved in the reaction, including same lung. Renal involvement, common in human vasculitis, was absent in our animal study. Also, pulmonary granulomas were totally absent, such as those seen in fungal and mycobacterial infections, sarcoidosis, talc pneumoconiosis, Wegener’s granulomatosis, and hypersensitivity pneumonitis. The occasional
VOLUME NUMBER
56 6
Arthus
reaction
in rabbit
lung
463
bizarre, multinucleated giant cell seen in the alveolar spaces resembles those reported in giant cell interstitial pneumonia (GIP) ; however, in GIP these cells are more numerous and are accompanied by varying numbers of desquamated alveolar pneumocytes.17~ I8 REFERENCES 1 Opie, E. L. : Inflammatory reaction of the immune animal to antigen (Arthus phenomena) and its relation to antibodies, J. Immunol. 9: 231, 1924. 2 Fried, B. M.: Allergic lobar pneumonia-experimental study, J. Exp. Med. 57: 111, 1933. 3 Cannon, P. R., Walsh, T. E., and Marshall, C. E.: Acute local anaphylaetic inflammation in the lungs, Am. J. Pathol. 17: 777, 1941. 4 Read, J. J.: The pathological changes produced by anti-lung serum, J. Pathol. 76: 403, 1958. sensitivity pneumonitis, 5 Nagaya, H., Schauble, M. K., and Sieker, H. 0.: Experimental Am. Rev. Respir. Dis. 91: 375, 1965. 6 Hagadorn, J. E., Vasquez, J. J., and Kinney, T. R.: Immunopathologic studies of an experimental model resembling Goodpasture’s syndrome, Am. J. Pathol. 57: 17, 1969. 7 Willoughby, W. F., and Dixon, F. J.: Experimental hemorrhagic pneumonitis produced by heterologous anti-lung antibody, J. Immunol. 104: 28, 1970. 8 Bensch, K. G., Dominguez, E., and Liebow, A. A.: Absorption of intact protein molecules across the pulmonary air-tissue barrier, Science 157: 1204, 196’7. 9 Richerson, H. B., Cheng, F. H. F., and Bauserman, S. C.: Acute experimental hypersensitivity pneumonit.is in rabbits, Am. Rev. Respir. Dis. 164: 568, 1971. 10 Richerson, H. B.: Varieties of acute immunologic damage to the rabbit lung, Ann. N. Y. Acad. Sci. 221: 340, 1974. 11 Ouchterlony, 0. : Diffusion-in-gel-methods for immunological analysis. II, Progress in allergy, Base1 and New York, 1962, S. Karger, vol. 6, p. 30. bronchial catheterization for the study of 12 Zavala, D. C., and Rhodes, M. L.: Selective experimental lung damage in the rabbit, Proc. Sot. Exp. Biol. Med. 144: 509, 1973. immunofluorescence in clinical immuno13 Beutner, E. H., and Nisengard, R. J.: Defined pathology, Protocol, Department of Microbiology, School of Medicine, State University of New York at Buffalo, 1971. 14 Richerson, H. B., Anuras; J., Rhodes, M. L., and Zavala, D. C.: Comparative histopathology of classic delayed hypersensitivity in skin and lung of rabbits, J, ALLERGY CLIN. IMMUNOL. 53: 87, 1974. (Abst.) 15 Churg, J., and Strauss, L.: Allergic granulomatosis, allergic angiitis, and periarteritis nodosa, Am. J. Pathol. 27: 277, 1951. 16 Claman, H. N., Karr, R,. M., Kohler, P. F., et al: Systemic (allergic?) vasculitis, J.
ALLERGY CLIN. IMMUNOL. 54: 54,1974. 17 Reddy, (GIP), 18 Liebow, Smith,
J. A., Gorelick, D. F., and Christianson, C. S.: Giant cell interstitial pneumonia Chest 58: 319, 1970. A. A.: New concepts and entities in pulmonary disease, in Liebow, A. A., and D. D., editors: The lung, Baltimore, 1968, The Williams & Wilkins Co., p. 332.
M.D., Mitchell 1. Rhodes, M.D. Iowa City, Iowa
M.D.,
Hal B. Richerson,
M.D.,
A localized Arthus reaction was produced in the lwng of sensitized rabbits by delivery of antigen into a lower lobe broncbns rising a method of selective bronchial catheterization under &oroscopy. The rabbits were sensitized with bovine immwnoglobulin G (B-IgG) in incomplete Freund’s adjnvant (IFA) to prodwe precipitating antibody without classic delayed hypersensitivity. Pulmonary histopathology was stndiea at intervals following antigen challenge, rising light and immnno@orescent microscopy. Gross lesions peripheral to the lower lobe bronohm receiving antigen were fozlnd within 18 hr. Subsequent necrosis redted in a dense scar by 6 wk. Microscopically, early lesians were typified by localized bronchitis, bronehiolitis, alveolitis, and vasculitis with exuberant exzcdates containing predominantly polymorphonzlclear leukocytes. Extensive focal necrosis was present by 78 hr. Immnnofiuorescent studies revealed the presence of B-IgG, rabbit IgG, and complement (C3) in and around bronchi, bronchioles, alveoli, and vessels. No grannlomatozls lesions were foma, and proliferation of alveolar lining cells ~a.3 not detected in these studies. Thus, the lung can participate in an acute Arthur reaction following local antigen challenge in systemically sensitiaed animals. The pathology more closely resembles a necrotizing bacterial pneumonia than an interstitial or hypersensitivity pneumonitis under the conditions of this experimental system. Implications for human disease are speculative.
Various experimental models of hypersensitivity pneumonitis in rabbits have been developed to study the inflammatory reaction of the immune animal to antigen and its relation to antibodies. A number of investigators including 0pie.l Fried,z and Cannon and associates3have produced a localized pneumonitis, bronchitis, and vasculitis in sensitized rabbits by instillation of horse serum or egg albumin directly into the lung or intratracheally. The effects of rabbit antiratlung antibodies have been studied by Read,4 Nagaya, Schauble, and Sicker,” Hagadorn, Vasquez, and Kinney,6 and Willoughby and Dixon7 by intracheal or intravenous routes of challenge. Bensch, Dominquez, and Liebow reported on the absorption of antigenically intact protein molecules across the pulmonary air-tissue barrier into the circulation of dogs after instillation into the trachea. Recently, Richerson, Cheng, and Bauserman,s, lo produced an acute allergic alveolitis, peribronchiolitis, and vasculitis in sensitized rabbits using aerosolized ovalbumin. From the Pulmonary and Allergy-Immunology Divisions, Department of Medicine, of Iowa College of Medicine. Supported by Grant No. 5598 from the American Lung Association. Received for publication July 26, 1974. Accepted for publication Nov. 19, 1974. Reprint requests to: Dr. D. C. Zavala, Department of Internal Medicine, University Hospitals, Iowa City, Iowa 52242. F-01. 56, No.
6, pp. 450-463
University
of
Iowa
VOLUME NUMBER
56 6
Arthus
reaction
in rabbit
lung
451
The purpose of this report is to describe the pulmonary pathologic changes occurring in a type III (Arthus) immunologic reaction. The lesions were produced by intrabronchial challenge in rabbits sensitized with bovine-IgG (B-IgG) in incomplete Freund’s adjuvant (IFA). A unique feature of the experiment was the utilization of a nonsurgical rabbit model of selective bronchial catheterization. None of the previous animal models delivered the antigen directly into a selected peripheral segment of the lung without highly sophisticated equipment, inhalation anesthesia, aerosolization, tracheostomy, intratracheal instillation, or transthoracic puncture. MATERIALS
AND
METHODS
A total of 54 male and female New Zealand white rabbits, weighing 2.7 to 3.6 kg, were used. In the study, 30 rabbits were immunized by injection with 1 mg of B-IgG in 0.15 M sodium chloride emulsified in IFA volume of 0.2 ml, distributed among the 4 footpads. Three weeks after immunization a subcutaneous booster injection consisting of 1 mg of antigen emulsified in an equal part of IFA was given at 4 injection sites (each shoulder), for a total of 4 mg antigen per animal. Skin tests for sensitizing activity and blood sampling for antibody formation were carried out on the twenty-first day following antigen footpad injections and then repeated 1 wk after booster injections. The animals were skin-tested with 1 mg B-IgG in 0.1 ml saline injected intradermally into closely shaved flank skin. Precipitins were measured using Ouchterlony’s method of double immunodiffusion in agar gel.11 One to two weeks after the booster injection, the rabbits mere challenged by instillation of 2 mg of antigen in 0.1 ml saline solution into one of the segmental bronchi of either lower lobe. Selective bronchial catheterization was performed as described previously.12 Each rabbit was prepared by an intramuscular injection of 0.5 ml of Innovar-Vet and secured on a peg board. After spraying the oropharynx and larynx with 4% lidoeaine, a flexible steel guide mire was inserted under direct vision through the vocal cords, and a small plastic tube passed over the guide wire into the trachea. The wire was removed leaving the endotraeheal tube in position. Bronchial catheterization was carried out by inserting a tiny polyethylene radiopaque catheter through the tracheal tube and guiding it under fluoroscopic control to the desired location in the lungs where the antigen was deposited. Following challenge, 3 rabbits were sacrificed with intravenous pentobarbital (150 mg) at 6, 12, 24, 48, 72 hr and 1, 2, 3, 4, and 6 wk. The lungs were removed in toto, inflated to approximate in situ size with 10% buffered formalin delivered into the trachea, and placed in a jar of formalin. After fixation of lung tissue, traverse sections were cut from each lobe, including the lesion produced at the site of intrabronchial challenge, placed in fresh buffered formalin, and prepared for light microscopy. In addition, sections were taken from the heart, kidney, liver, and spleen. Paraffin-embedded tissue was cut in 5 p sections and then stained with hematoxylin and eosin. Sections of fresh lung tissue from challenge and control sites also were studied by fluorescent antibody technique, as described by Beutner and Nisengard.13 Tissues taken from 2 IFA rabbits at 6, 24, 48, 72 hr, and 1 wk postchallenge were unsnap” frozen in liquid nitrogen, sectioned on a cryostat, and prepared using fluorescein-conjugated antirabbit.IgG, antibovine*IgG, and antiC3, each diluted 1:30.* The sections were viewed on a Leitz Ortholux II fluorescence microscope using heat and suppression filters. Appropriate pictures were taken on Polaroid and color slide film. Twelve control animals were used consisting of 3 unimmunized rabbits challenged with B-IgG, 6 B-IgG immunized rabbits challenged with different protein antigens (ovalbumin and human serum albumin, 3 each), and 3 B-IgG immunized, unchallenged rabbits. *The anti-R-IgG had an antibody protein concentration of 16.8 mg/ml the anti-BeIgG had a protein concentration of 3.5 mg/ml (Miles anti-C3 had a protein concentration of 1.3 mg/ml (Meloy) .
(Miles Laboratories) Laboratories), and
; the
452
Zavala
FIG. 1. intact Arrow
denotes
J. ALLERGY
et al.
lung removed site of lesion
from the IFA-sensitized which resembles an
rabbit infarct.
72 hr after
CLIN. IMMUNOL. DECEMBER 1975
antigen
challenge.
RESULTS Skin tests
Skin reactivity was used for the selection of experimental animals. Intracutaneous injections of 1 mg B-IgG produced equally strong skin reactions in 30 IFA-immunized rabbits when carried out 3 wk after the footpad injections and repeated 1 wk after the booster injections. Additional animals with poor or equivocal results were excluded from the study. Strongly positive reactions were present at 6 hr and were progressive to 48 hr, at which time the lesions consisted of dark hemorrhagic centers measuring 0.5 to 0.8 cm, surrounded by edematous, erythematous halos, 3.0 to 3.5 cm in diameter. The lesions were unchanged or slightly decreased at 72 hr after which the halos gradually subsided but the centers became necrotic. Unsensitized rabbits also were skintested with the same dose of antigen (1 mg B-IgG) and showed no skin irritation. Precipitins
Precipitins were demonstrated in the immune sera (postimmunization and postbooster) of all experimental animals used in this study to a dilution of 1:4 against a concentration of 0.1 mg/ml of antigen in the center well. Pathology
Intrabronchial challenge of the sensitized rabbits produced the following pathologic changes. At autopsy, gross findings were apparent after 12 hr, with the initial lesions appearing around one of the lower lobe basal bronchi
VOLUME NUMBER
56 6
Arthus
FIG. 2. Histopathology intense inflammatory A large
necrotic
of rabbit (PMNs) and area
is located
lung round nearby,
72
hr cell
not
postchallenge. infiltration seen
in these
reaction
involving
A vasculitis arteries
views.
(x200.)
in
rabbit
is
present (A) and
lung
with veins
453
an (8).
454
Zovala
et
al.
FIG. 3. Histopathology of rabbit lung 7 hr postchallenge. An acute with thickened septa, cellular infiltrates, and exudate in aveolar magnification in B shows intro-alveolar macrophages, red blood cells. (x500.) In C, the arrow points to a bizarre, multinucleated alveolar space. (x500.)
J. ALLERGY
CLIN. IMMUNOL. DECEMBER 1975
alveolitis is shown in A spaces. (x200.) Higher ceils, PMNs, and round giant cell located in the
VOLUME NUMBER
Arthus
56 6
FIG.
3C.
For
legend
see opposite
reaction
in
rabbit
lung
455
page.
and its branches that received the antigen. Within 72 hr the lesion often resembled an infarct (Fig. 1). Usually a necrotic center developed that excavated by the third to fourth week postchallenge and then collapsed and formed a dense scar after the sixth week. Microscopically, the initial early reaction occurred in and around the bronchi, followed by spread into the surrounding lung. The first changes were seen at 6 hr postchallenge and consisted of a regional peribronchitis, perihronchiolitis, alveolitis, and vasculitis. The early cellular infiltrates were made up of the following: (1) focal collections of PMNs’ in the lumen and walls of bronchi and bronchioles, (2) a mixture of PMNs and lymphocytes around bronchi, bronchioles, and pulmonary vessels. (3) PMNs and red blood cells filling dilated alveolar capillaries, and (4) focal intra-alveolar exudation of PMNs and red blood cells. By 72 hr there was extensive focal necrosis and ulceration of the walls of the bronchi and bronchioles with extensive intraluminal exudation and occasional rupture. A predominantly round cell infiltrate surrounded the smaller airwayqand there was hypcrplasia of lymphoid tissue around many of the larger bronchi. A moderate to intense vasculitis, involving arteries and veins beyond the lesion. There were rxt,ensire perivascular (Fig. 2), did not extend infiltrates consisting of PMNs, lymphocytes with some large mononuclear cells, and plasma cells. An alvcolitis was clearly evident with thickened septa, heavy cellular infiltrates, and scattered fluid-exudate in the alveolar spaces (Fig. 3, “Polymorphonuelear leukocytes
that,
in the
rabbits,
have
cytoplasmic,
cosinophilic
granules.
456
Zavala
et al.
FIG. 4. Histopathology of rabbit lung 72 hr postchallenge. margin of the lesion is indicated by arrows. The challenged just out of the photograph. (x50.)
J. ALLERGY
CLIN. IMMUNOL. DECEMBER 1975
The line of demarcation bronchus is located
at the inferiorly
A). The intra-alveolar cells consisted of macrophages, red blood cells, PMNs, a few round cells, and an occasional multinucleated giant cell (Fig. 3, B and C) . A leading line of demarcation was observed at the outer margins of the lesions (Fig. 4)) possibly representing antigen-antibody interaction. The contralateral lung was normal except for filling of a moderate number of alveolar capillaries with eosinophilic granulocytes. By 1 wk closely grouped necrotic areas of variable sizes developed at the challenge sites (Fig. 5), which usually were evacuated by the third to fourth weeks. The resulting cavity then collapsed and, after the sixth week, formed a dense scar (Fig. 6). At no time did the animals appear ill. Microscopic studies of the heart, liver, spleen, and kidneys were normal on all animals. Halving the challenging dose of antigen to 1 mg B-IgG in 12 IFA-immunized rabbits resulted in lung lesions that were less extensive and intense. had completely clean lungs on light microscopy All of the control mimals except for an occasional small focus of intra-alveolar macrophages, a rare vessel surrounded with a few small round cells, and minimal 1,ymphoid collections in peribronchial and peribronchiolar locations. Immunofluorescent microscopy on control animals demonstrated a rare fluorescent phagocytic cell with antisera to both IgG and C3. The specificity of each conjugate was confirmed by immunoelectrophoresis. lmmunoglobulins deposited in animal tissue were identified by prior inhibition using unlabeled antirabbit IgG, antirabbit C3 antibody, and
VOLUME NUMBER
56 6
Arthus
FIG. 5. Histopathology Note proximity of the
of rabbit lung 1 wk postchallenge. lesion to a bronchiole. (x50.)
FIG. 6. Histopathology resulting in retraction
of rabbit lung 6 wk of the pleura. (x30.)
postchallenge.
reaction
A large
Arrow
in rabbit
necrotic
points
area
lung
457
is evident.
to a healed
scar,
458
FIG. seen
Zavala
et
J. ALLERGY
al.
7. lmmunoftuorescent in
the
bronchial
(A)
microscopy and alveolar
of
rabbit
(B) air
lung spaces.
at 6 hr (x100.)
postchallenge.
CLIN. IMMUNOL. DECEMBER 1975
Free
B-IgG
is
VOLUME NUMBER
56 6
FIG. 8. lmmunofluorescent seen in bronchi (A) and B-IgG. A x100; 8, x540.)
Arthus
reaction
in rabbit
microscopy of rabbit lung at 48 hr postchallenge. alveoli (B). Arrow points to an alveolar phagocytic
lung
459
8-IgG again cell containing
is
460
Zavala
J. ALLERGY
et al.
FIG. 9. lmmunofluorescent in peribronchial and vessel. (xl 00.)
microscopy perivascular
of rabbit lung at 6 hr postchallenge. areas. 6 indicates bronchus and V
CLIN. IMMUNOL. DECEMBER 1975
R-IgG indicates
is seen blood
antibovine IgG. Subsequently, incubation was again carried out with fluoresceinlabeled specific immunoglobulins. Immunofluorescent microscopy on sensitized rabbits demonstrated initiation of a localized Arthus reaction 6 hr after intrabronchial challenge with bovineIgG. Specimens at 6 and 48 hr showed localized B-IgG antigen, rabbit-IgG (R-IgG), and complement (C3). B-IgG was found in bronchial and alveolar air spaces (Fig. 7) and in phagocytic cells located in and around the airways and blood vessels (Fig. 8). R-IgG was present along with B-IgG (Figs. 9 and lo), but there was no R-IgG and B-IgG in bronchial, bronchiolar, or alveolar lining cells. At 48 hr, most of the complement (C3) was no longer free but appeared as brightly fluorescing granules in phagocytic cells, located in and around blood vessels and airways. Some free granular C3 was present in the alveolar spaces. At 1 wk, B-IgG and complement were no longer detectable, although phagocytic cells containing RIgG granules were still found scattered in lung parenchyma with some peribronchiolar and perivascular localization (Fig. 11). All of the various stages of this reaction were localized entirely at the challenge site. DISCUSSION
We have described the morphologic sensitized rabbits following intrabronchial
changes occurring in the lungs of challenge by selective catheterization.
VOLUME NUMBER
Arthus
56 6
FIG. 10. lmmunofluorescent present in the bronchi but
microscopy notably absent
reaction
of rabbit lung at 48 hr in the lining cells. [xl 00.)
in rabbit
postchallenge.
lung
R-IgG
461
is
The animals were immunized with B-IgG, emulsified in IFA. Circulating precipitins and positive skin tests were demonstrated in all animals. The experimental lesions began in 6 hr with regional peribronchiolit~is, alveolitis, and vasculitis and then rapidly progressed to focal necrosis in 72 hr and central necrosis in 1 wk. There was extensive intraluminal exudation in the bronchi, a mixed perivascular infiltration of PM& and lymphocytes, and an alveolitis with septal widening and cellular infiltrates. The necrotic core evacuated in 3 to 4 wk, leaving a cavity that soon collapsed and formed a residual scar. The rabbits were extremely tolerant of their localized immune lung injury and at no time did they appear ill. An apparent pulmonary Arthus reaction was shown in the sensit.ized rabbits by immunofluorescent microscopy. Antigen and complement were found in bronchial and alveolar air spaces and in alveolar macrophages. Bronchial, bronchiolar, and alveolar lining cells took no active part in this reaction. Apparently there was no proliferation of alveolar type II pneumocytes, as observed by light microscopy. In this study of experimental immune lung injury, a nonsurgical rabbit model using bronchial catheterization provided the advantage of delivering the antigen into a selected pulmonary segment. The remainder of the lung served as its own control. Using a similar technique of selective bronchial catheterization, Richerson
462
Zavala
J. ALLERGY
et al.
FIG. 11. lmmunofluorescent remains. Arrows denote and normal lung. (x100.)
microscopy margin of the
of rabbit lung at 1 wk postchallenge. reaction between fluorescent rabbit
CLIN. IMMUNOL. DECEMBER 1975
R-IgG alone immunoglobulin
and co-worke&* studied the histopathology of classic delayed hypersensitivity in the lung using purified protein derivative (PPD) . Although significant differences in tissue response are noted between their work and this study, there is a remarkable similarity in the absence of proliferation of type II alveolar lining cells (granular pneumocytes). Apparently the lung’s response in the Arthus reaction and delayed hypersensitivity differs from its response to other forms of injury (e.g., various drugs, oxygen, radiation), which are typified by proliferating, atypical alveolar pneumocytes. Any comparison with diseases of human lungs is speculative. Nevertheless, under the conditions of this experimental model the resulting tissue reaction pneumonia with a localized vasculitis rather more closely resembled a necrotizing than any of the hypersensitivity lung diseases or several types of necrotizing vasculitis which include Wegener’s granulomatosis, the eosinophilic vasculitis described by Churg and Straus.ql” giant cell arteritis, periarteritis nodosa, and systemic vasculitis.16 The resulting cavity healed like a lung abscess. No other nonchallenged areas of the organs were involved in the reaction, including same lung. Renal involvement, common in human vasculitis, was absent in our animal study. Also, pulmonary granulomas were totally absent, such as those seen in fungal and mycobacterial infections, sarcoidosis, talc pneumoconiosis, Wegener’s granulomatosis, and hypersensitivity pneumonitis. The occasional
VOLUME NUMBER
56 6
Arthus
reaction
in rabbit
lung
463
bizarre, multinucleated giant cell seen in the alveolar spaces resembles those reported in giant cell interstitial pneumonia (GIP) ; however, in GIP these cells are more numerous and are accompanied by varying numbers of desquamated alveolar pneumocytes.17~ I8 REFERENCES 1 Opie, E. L. : Inflammatory reaction of the immune animal to antigen (Arthus phenomena) and its relation to antibodies, J. Immunol. 9: 231, 1924. 2 Fried, B. M.: Allergic lobar pneumonia-experimental study, J. Exp. Med. 57: 111, 1933. 3 Cannon, P. R., Walsh, T. E., and Marshall, C. E.: Acute local anaphylaetic inflammation in the lungs, Am. J. Pathol. 17: 777, 1941. 4 Read, J. J.: The pathological changes produced by anti-lung serum, J. Pathol. 76: 403, 1958. sensitivity pneumonitis, 5 Nagaya, H., Schauble, M. K., and Sieker, H. 0.: Experimental Am. Rev. Respir. Dis. 91: 375, 1965. 6 Hagadorn, J. E., Vasquez, J. J., and Kinney, T. R.: Immunopathologic studies of an experimental model resembling Goodpasture’s syndrome, Am. J. Pathol. 57: 17, 1969. 7 Willoughby, W. F., and Dixon, F. J.: Experimental hemorrhagic pneumonitis produced by heterologous anti-lung antibody, J. Immunol. 104: 28, 1970. 8 Bensch, K. G., Dominguez, E., and Liebow, A. A.: Absorption of intact protein molecules across the pulmonary air-tissue barrier, Science 157: 1204, 196’7. 9 Richerson, H. B., Cheng, F. H. F., and Bauserman, S. C.: Acute experimental hypersensitivity pneumonit.is in rabbits, Am. Rev. Respir. Dis. 164: 568, 1971. 10 Richerson, H. B.: Varieties of acute immunologic damage to the rabbit lung, Ann. N. Y. Acad. Sci. 221: 340, 1974. 11 Ouchterlony, 0. : Diffusion-in-gel-methods for immunological analysis. II, Progress in allergy, Base1 and New York, 1962, S. Karger, vol. 6, p. 30. bronchial catheterization for the study of 12 Zavala, D. C., and Rhodes, M. L.: Selective experimental lung damage in the rabbit, Proc. Sot. Exp. Biol. Med. 144: 509, 1973. immunofluorescence in clinical immuno13 Beutner, E. H., and Nisengard, R. J.: Defined pathology, Protocol, Department of Microbiology, School of Medicine, State University of New York at Buffalo, 1971. 14 Richerson, H. B., Anuras; J., Rhodes, M. L., and Zavala, D. C.: Comparative histopathology of classic delayed hypersensitivity in skin and lung of rabbits, J, ALLERGY CLIN. IMMUNOL. 53: 87, 1974. (Abst.) 15 Churg, J., and Strauss, L.: Allergic granulomatosis, allergic angiitis, and periarteritis nodosa, Am. J. Pathol. 27: 277, 1951. 16 Claman, H. N., Karr, R,. M., Kohler, P. F., et al: Systemic (allergic?) vasculitis, J.
ALLERGY CLIN. IMMUNOL. 54: 54,1974. 17 Reddy, (GIP), 18 Liebow, Smith,
J. A., Gorelick, D. F., and Christianson, C. S.: Giant cell interstitial pneumonia Chest 58: 319, 1970. A. A.: New concepts and entities in pulmonary disease, in Liebow, A. A., and D. D., editors: The lung, Baltimore, 1968, The Williams & Wilkins Co., p. 332.