Bronchial Hyperreactivity after Inhalation of Trimellitic Anhydride Dust in Guinea Pigs after Intradermal Sensitization to the Free Hapten1 , 2
JAMES P. HAYES, ROB DANIEL, ROSEMARY D. TEE, PETER J. BARNES, ANTHONY J. NEWMAN TAYLOR, and K. FAN CHUNG Introduction
T rimellitic anhydride (TMA) is known to cause occupational asthma (1) and pulmonary hemorrhage with hemolytic anemia in humans (2). Inhalation of TMA dust may provoke immediate and late asthmatic responses in sensitized persons (3). Nonspecific bronchial hyperreactivity is associated with occupational asthma, and transient increases in bronchial reactivity may follow inhalation of specific occupational agents (4). An animal model of increased airway responsiveness caused by the acid anhydrides may provide an opportunity of examining the mechanisms underlying this characteristic feature of occupational asthma. Previous workers have demonstrated systemic and pulmonary immune responses in various animal species after tracheal instillation of free TMA (5). Botham and coworkers (6) demonstrated an immunologic response in guinea pigs sensitized to free TMA administered by the intradermal route, and they also observed an alteration in respiratory rate immediately after inhalation of TMA dust in sensitized animals. We have previously reported that intradermal injection of free TMA induces a specific IgG-I, and in some animals IgE response to TMA conjugated to guinea pig albumin (GPSA). Inhaled and intravenous challenge with TMA-GPSA provokes an acute change in bronchopulmonary response in sensitized guinea pigs (7). In this study, we examined the effect of inhaled TMA dust on airway reactivity and cellular infiltration in sensitized and in nonsensitized guinea pigs 8 h after exposure. Wealso examined the time course of this response by assessing the effect of TMA dust at 2 and at 24 h after exposure in sensitized guinea pigs. Methods Sensitization Male Dunkin-Hartley guinea pigs weighing
SUMMARY We have developed In the guinea pig, an animal model of bronchial hyperreactlvlty provoked by Inhalation of trimellltic anhydride (TMA) dust, a known cause of occupational asthma in humans, after intradermal sensitization to the free hapten. Male Dunkin-Hartley guinea pigs (n = 6) were Injected intradermally with 0.1 ml of 30% TMA In corn oil. Control animals (n = 7) were injected with 0.1 ml corn 011 alone. On Days 21 to 28 after sensitization, guinea pigs were challenged (nose only) to 12 mg/m 3 of inhalable TMA dust for 30 min. Bronchial reactivity was measured in sensitized animals and in control animals at 8 h after exposure to the dust. We also measured bronchial reactivity In sensitized exposed guinea pigs at 2 h (n 5) and at 24 h (n 5). Pulmonary inflation pressure (PIP)was used to assess bronchopulmonary response. Blood samples were taken for assessment of IgG-1 antibodies to TMA conjugated to guinea-pig albumin. The concentration of acetylcholine required to Induce a 100% increase in PIP was used to assess bronchial reactivity. The lungs were eviscerated for histologic examination. All guinea pigs injected Intradermally with TMA had high titers of specific IgG-1 antibodies to TMA conjugated to guinea-pig albumin. There was a significant increase in bronchial reactivity in sensitized guinea pigs 8 h after exposure to the TMA dust compared with that In the control animals. There was also a significant eosinophilic Inflammatory influx In the subeplthelium of the sensitized groups. Exposure to free TMA dust in sensitized guinea pigs induces an increase In bronchial reactivity accompanied by an eosinophilic inflammatory exudate. This model may be suitable for assessing the underlying mechanisms of occupational asthma caused by acid anhydrides and possibly to other low molecular weight chemicals.
=
=
AM REV RESPIR DIS 1992; 146:1311-1314
200 to 250 g (n = 16)were injected intradermally with 0.1ml of 30010 TMA (Aldrich, Gillingham, UK) suspended in corn oil. Control guinea pigs (n = 7) were injected with 0.1 ml of corn oil alone.
was read on the spectrophotometer at an optical density of 420 nm. The substitution ratio for TMA-GPSA was 21:1.
Preparation of Trimellitic Anhydride Conjugate
Plates werecoated with 5 ug/ml TMA-GPSA in coating buffer and incubated overnight at 4 0 C. The next day the plates were washed four times in PBS/1\veen. Serum was diluted to 1:50solution, and plates were coated with PBS/1\veen. Serum was added in doubling dilutions across the plate and incubated for 3 h at 27 0 C. The plates were again washed four times with PBS/Tween and rabbit an-
TMA was conjugated to guinea pig serum albumin (GPSA) as follows. Four 250-mg aliquots of GPSA (Sigma, Poole, UK) were mixed with 50mlof9% NaHC03 on ice, and 250 mg of ground TMA dust were added and allowed to stand for 10min. Each aliquot was stirred separately for 1 h. The aliquots were pooled and centrifuged for 10min at 600 x g. The supernatant was separated and dialyzed with 0.02 M NH 4HC03 for 48 h with six changes. This was further dialyzed with distilled water for 24 h and lyophilized. The degree of substitution was assessed by UV spectrophotometry. Briefly 1.0 mg TMA-GPSA were added to 1 ml of 0.1 M borate buffer at pH 9.3. A 0.03-M solution of 2,4,6trinitrobenzene-sulfonic acid (TNBS) (Aldrich) in borate buffer was prepared. Then, 25 III of TNBS solution were added at 1 ml of the TMA-GPSA solution and incubated for 20 min at room temperature. The sample
ELISA for Assessment of IgG-l Antibodies
(Received in originalform August 12, 1991 and in revised form June 24, 1992) 1 From the Departments of Thoracic Medicine and Occupational & Environmental Medicine,Royal Brompton National Heart & Lung Hospital, National Heart & Lung Institute, London, United Kingdom. 2 Correspondence and requests for reprints should be addressed to K. F. Chung, Department of Thoracic Medicine, National Heart & Lung Institute, Dovehouse St., London SW3 6LY UK.
1311
1312
tiguinea pig IgO-1 (ICN Biomedicals Ltd., High Wycombe Bucks, UK), diluted to 1:2,500, was added and incubated for 2 h at 27° C. The plates were again washed with PBS/1\veen. Antirabbit IgO-peroxidase (ICN Biomedical Ltd.) was added at 1:5,000dilution and incubated at 27° C for 2 h. Finally, the plates were washed in PBS/Tween and chromogen was added. The reaction was stopped after 6 min with 1M H 2S04 and read at an optical density of 492 nm. The titers of IgO-1 were calculated from the last dilution that was double the control sera.
Trimellitic Anhydride Dust Exposure On Days 21 to 28, guinea pigs were placed in a conical container with their nostrils protruding, fixed in an exposure chamber with an internal volume of 30 L, and exposed to 12 mg/rn" of inhalable dust for 30 min. The dust was generated by passing air (12L/min) through a Wright's dust feeder above the chamber. Dust levels were measured before and during exposure gravimetrically by extracting air samples using a dust sampler (AFC 123; CaselleLondon Ltd., Bedford, UK) onto fiberglassfilters (Whatman, Maidstone, UK) at 2 L/min for 5 min. Particle size was measured by a cascade impactor (Marple, Andersen Samples Ltd., Atlanta, GA). Measurement of Bronchopulmonary Response and Bronchial Reactivity After exposure to free TMA dust, sensitized guinea pigsweredivided into three groups and were assessed at 2 h (n = 5), at 8 h (n = 6), or at 24 h (n = 5). Control animals (n = 7) were assessed at 8 h after exposure to TMA dust. All guinea pigs were anesthetized intraperitoneally with 25 g/kg urethane (Sigma), tracheostomized, and ventilated at a tidal volume of 10ml/kg and a respiratory rate of 60 breaths/min on an animal ventilator (Harvard Apparatus Ltd., Edenbridge, Kent, UK). A cannula was inserted into the left carotid artery to monitor blood pressure. Pulmonary inflation pressure(PIP) was measured through a side-arm tube attached at the origin of the outlet tube by a pressure transducer (Sensym 744; Farnell, Leeds, UK) connected to a twochannel recorder (Multitracer 2; Lectromed, Letchworth, UK). Acetylcholine (Sigma) was administered in increasing 1/2-log concentrations via an ultrasonic nebulizer (DeVilbiss Co., Somerset, PA). Each dose of acetylcholine (initial dose, 10-5 M) was administered for 20 breaths. The concentration of acetylcholine required to induce a 100070 increase in PIP (PC I OO) was used as an index of bronchial reactivity. Blood samples (0.3 ml) were taken from all guinea pigs at this time for measurement of IgG-1 antibodies to TMA-GPSA. Histopathology At the end of the experiment, the guinea pigs were killed. The lungs were inflated with 4% formol saline, and the trachea was tied. The tissue was fixed in formaldehyde, sectioned (3-JJm sections), and stained with carbolchromotrope to demonstrate eosinophil granules. The slides were coded and read without
HAYES, DANIEL, TEE, BARNES, NEWMAN TAYLOR, AND CHUNG
5 10 4
10
.
3 10
~
2 10
~
10
~
>-
i
0 000 00000 000 0000 0
!:
0 0
10 ·2
8
~
...s2
J1.
B'
~
10.3
A
~
'0
.c 0
i8'
fA
&
10.4
A
10. 5
.. ••
•
A
..J
!i!)
2h
8h
sensitized
controls
controls
Fig. 1. IgG-1 antibody titers to trimellitic anhydride conjugated to guinea pig serum albumin (TMA-GPSA) in guinea pigs given intradermal injections with 0.1 mlof 30% TMA. No antibodies to TMA-GPSA were in control guinea pigs injected with 0.1 ml corn oil alone.
prior knowledge of which group of animals they came from. In order to quantify the number of eosinophils in the airways, a section of tissue from the right or left main bronchus (chosen randomly) and a similar section diametrically opposite were selected. Eosinophils were counted within a given area using a graticule (size56 mm': magnification 400) positioned from the epithelium extending into the subepithelial layers. At the same time, mononuclear cells and polymorphonuclear cells were also counted.
Statistical Analysis Data are expressed as mean and SEM. PC 100 values were analyzed as 10glo. A paired t test was used to compare bronchial reactivity and eosinophil counts in the sensitized group assessed at 8 h and the nonsensitized group. To determine the time course of the response, bronchial reactivity and eosinophil counts in the three sensitized groups were compared using analysis of variance. A p value < 0.05 was considered significant. Results
All guinea pigs given intradermal injections of TMA had high levels of specific antibodies to TMA-GPSA (figure 1). There was no significant change in blood pressure among the groups during the course of the experiment. There was no significant difference in baseline PIP measurements between the control group and the sensitized group at 8 h. There was a significant increase in PC lO O in the sensitized group measured at 8 h (0.26 mM, 1.4, geometric mean, [GSEM]) compared with the control group (3.6 mM, 2.2 GSEM; p < 0.02). There was no significant difference in bronchial reactivity in the sensitized guinea pigs at 2 h (1.4 mM, 2.1 GSEM) and at 24 h (0.62 mM, 3.7 GSEM) compared with the sensitizedgroup at 8 h (fig-
8h
24h
sensitized
Fig. 2. Effect of inhaled trimellitic anhydride (TMA) dust on bronchial reactivity, which was measured at 2 h (closed triangles), at 8 h (open triangles), and at 24 h (closed circles) in sensitized guinea pigs. Control guinea pigs (open circles) were measured at 8 h. The concentration of acetylcholine required to induce an increase of 100% in pulmonary inflation pressure (pC.oo) was used to measure bronchial reactivity. There was a significant decrease in acetylcholine Pc'oo in sensitized animals compared with that in control animals. (p < 0.02.) Horizontal bars indicate mean values.
ure 2), although the increase in airway responsiveness was maximal at 8 h. In the control guinea pigs eosinophils were occasionally found in the extrapulmonary airways as foci beneath surface epithelium and less often within the epithelium. They were not a feature of the adventitia. Similarly, in lung sections occasional eosinophils wereseen in and beneath the surface epithelium. In the sensitized guinea pigs there were significant increases in eosinophil numbers in both the extrapulmonary airway and the lung
ie
40
!
..~
30
A
~
c
::3
I
.!! 20
i.0 c
0
Ii
a ::3
z
0
~
•
~ 9 8
• .....
t
10
'0
JE
t
t
•
8
2
24
hours
controls
sensitized
Fig. 3. Effect of inhaled trimellitic anhydride dust (12 mg/m3 for 30 min) on the number of eosinophils per unit airway tissue in sensitized guinea pigs at 2 h (closed triangles), at 8 h (open triangles), and at 24 h (closed circles). Control guinea pigs (open circles) were assessed 8 h after exposure. There was a significant increase in the number of eosinophils in airway tissue in sensitized animals compared with that in control animals. (p < 0.005.) Horizontal bars indicate mean values.
BRONCHIAL HYPERREACTIVITY
1313
CAUSED BY TRIMELLITIC ANHYDRIDE
B
Fig. 4. Photomicrograph of bronchial tissue stained with carbol chromotrope(magnification, 400) showing transverse sections of main bronchus from a guinea pig sensitized to trimellitic anhydride (A) and a control guinea pig (B). Longitudinal sections of bronchus from a sensitized guinea pig (C) and a control guinea pig (D) are also shown. The arrow indicates a typical eosinophil.
sections. In the extrapulmonary airways these occasionally extended to the deeper layers. In the lungs eosinophils were found both internal and external to the bronchial smooth muscle and between blocks of smooth muscle. Eosinophil counts were significantly greater in the sensitized group at 8 h (21.7 cells ± 1.8) compared with the control group (4.9 cells ± 1.8; p < 0.01). Occasional alveolar foci of hemorrhage were seen in both sensitized and nonsensitized animals. A small but significant increase in mononuclear cells was also seen in the sensitized group (5.0 ± 1.1, mean ± SEM) compared with the control group (2.4 ± 0.43; p < 0.05). The exact nature of these cells remains uncertain, as specific lymphocyte stains were not used. There was no significant difference in the eosinophil counts between the three sensitized groups (figure 3), but as with the changes in airway responsiveness, the eosinophil count in the subepithelium was maximal 8 h after exposure to TMA dust (figure 4). There was no significant correlation between the number of eo-
sinophils in the subepithelium and the level of bronchial reactivity in the three sensitized groups of guinea pigs. However, there was a significant correlation (p < 0.005, R = 0.73, R2 = 0.53) between the number of eosinophils in the subepithelium and airway responsiveness in the sensitized group and the nonsensitized group at 8 h after exposure to TMA dust (figure 5). Discussion
We have found that guinea pigs injected intradermally with TMA in corn oil developed evidence of a specific immunologic response, and subsequent inhalation of TMA dust provoked an increase in airway responsiveness to inhaled acetylcholine that was associated with subepithelial eosinophil and mononuclear cell infiltration. These responses did not occur in guinea pigs injected with corn oil alone. Wedid not see any increase in baseline PIP measurements in the sensitized groups compared with that in the control group. There was no significant difference in the bronchial reactivity mea-
surements and the degree of eosinophil or mononuclear cell infiltration found in the three groups of sensitized guinea pigs. However, both the bronchial reactivity and cellular infiltration were maximal 8 h after exposure to the TMA dust. Occasional alveolar foci of hemorrhage wereseen in both the control and the sen10 -6
!:
•
10 -5
0 0
U 0-
10 -4
III
.s
'0 ~ u
10 -3
su III
10 -2
~ ...J
10 -1
ClI
, 0
10
20
30
40
Number of eosinophlls per unit area (56mm2)
Fig. 5. The correlation between the number of eosinophils in the subepithelium and the concentration of acetylcholine required to induce a 100% increase in baseline pulmonary inflation pressure (PIP) in sensitized and in control guinea pigs assessed 8 h after inhalation of TMA dust. (p < 0.005, R == 0.73 R2 = 0.53.)
1314
HAYES, DANIEL, TEE, BARNES, NEWMAN TAYLOR, AND CHUNG
sitized groups. This may have been ar- in one study an association was found tefactual or a direct effect of TMA on between the number of eosinophils and the alveoli,but it would not have account- T lymphocytes in the airways of sensied for the changes seen in the sensitized tized guinea pigs. Most recently, Sanjar group in this study. Although we did not and coworkers (13) developed a model find a direct relationship between the lev- of bronchial hyperreactivity and eoel of bronchial reactivity and the degree sinophilia in bronchoalveolar lavage fluid of eosinophil or mononuclear cell in- in guinea pigs sensitized to ovalbumin. filtration in the three groups of sensitized In this study inhibition of the eosinophilguinea pigs, a strong correlation between ic response by dexamethasone did not rebronchial reactivity and eosinophil in- sult in inhibition of the increase in bronfiltration in the sensitized group assessed chial reactivity, suggesting that although 8 h after inhalation of TMA dust and there is an association between these two the control group (p < 0.005), suggest- events, a causal relationship may not ing an association (or relationship) be- exist. Acid anhydrides are known to cause tween airway reactivity and the inflammatory infiltrate. occupational asthma in humans. In the Severalstudies have examined different majority of cases it is associated with eimethods of sensitization with TMA in ther an IgE or IgG-4 antibody response. a variety of species. However, few stud- Specific inhalation tests in sensitized subies have investigated the effect of TMA jects may provoke both reduced airway on bronchopulmonary responses in pre- caliber and increased airway responsiveviously sensitized animals. Botham and ness (14). In occupational asthma the late coworkers (6), using a similar method of response has been associated with an insensitization, observed an acute change crease in nonspecific bronchial reactiviin respiratory rate in guinea pigs exposed ty, and an increase in bronchial reactivito free TMA dust. However, the exact na- ty at 3 h has been found to be predictive ture of this response is unknown. Previ- of the late asthmatic response (15). Few ously, we reported an immediate increase studies have reported the pathologic in bronchopulmonary response to TMA- changes associated with the late asthmatGPSA both by inhalation or by intrave- ic response in patients with occupationnous injection in guinea pigs sensitized al asthma caused by low molecular to TMA by intradermal injection (7). To weight chemicals. Fabbri and coworkers our knowledge this is the first report on (16) found an increase in neutrophils in the effect of a low molecular weight bronchoalveolar lavage fluid in patients chemical known to cause occupational with a late asthmatic response after exasthma on airway responsiveness and air posure to toluene diisocyanate. Paggiaro way disease in previously sensitized and coworkers (17) have reported persistence of inflammatory cells in the subanimals. Sheppard and coworkers (8, 9) found mucosa and increased eosinophils in the an increase in airway responsiveness as- bronchoalveolar lavage fluid in patients sociated with an influx of polymor- with chronic asthma induced by toluene phonuclear leukocytes into the airway of diisocyanate. In our study we demonnonsensitized guinea pigs exposed to tol- strated two important features that may uene diisocyanate; however, this increase be associated with occupational asthma, in airway responsiveness also occurred in namely, increased airway responsiveness the absence of a cellular infiltrate, and and a predominantly eosinophilic init was mediated by release of tachykinins. flammatory infiltrate. In conclusion, we have developed a This is unlikely to be a valid model of hypersensitivity-induced asthma in hu- model of bronchial hyperresponsiveness mans. Biagnini and coworkers (10) sen- and eosinophil infiltration in the airways sitized cyanomologous monkeys to plati- after exposure to free TMA dust in num salts when inhaled concurrently guinea pigs previously sensitized to the with ozone. Subsequent inhalation of the free hapten. The provocation of increased platinum salt resulted in an increase in airway responsiveness and eosinophil inbronchial reactivity to methacholine in filtration of the airway epithelium in sensensitized animals. However, the patho- sitized animals by inhaled TMA dust closely parallels changes observed in oclogic findings were not reported. Our results are similar to guinea pig cupational asthma in humans. models of sensitization and subsequent Acknowledgment exposure by inhalation to ovalbumin and Ascaris (11, 12). These studies have The writers thank Dr. Peter Jeffrey, Senior reported an increase in inflammatory Lecturer, Department of Lung Pathology, for cells in the airways and bronchoalveolar his help and advice during the course of this lavage fluid of the sensitized animals, and study.
References 1. Fawcett JW, Newman TaylorAJ, Pepys J. Asthma due to inhaled chemical agents-epoxy resin systems containing phthalic anhydride, trimellitic anhydride and triethylene tetramine. Clin Allergy 1977; 7:1-14. 2. Zeiss CR, Patterson R, Pruzansky J J, Miller MM, Rosenberg M, Levitz D. Trimellitic-anhydride induced airway syndromes: clinical and immunologic syndromes. J Allergy Clin Immunol 1977; 60:96-103. 3. Howe W, Venables KM, Topping MD, et al. Tetrachlorophthalic anhydride asthma: evidence for specific 19Bantibody. J AllergyClin Immunol1983; 71:5-11. 4. Chang-YeungM, Lam S. Occupational asthma. Am Rev Respir Dis 1986; 133:686-703. 5. Sale SR, Patterson R, Zeiss CR, Fiore M, Harris KE, Yawn D. Immune responses of dogs and rabbits to intrabronchial trimellitic anhydride. Int Arch Allergy Appl Immunol 1982; 67:329-34. 6. Botham PA, Rattray NJ, Woodcock DR, Walsh ST, Hext PM. The induction of respiratory allergy in guinea pigs following intradermal injection of trimellitic anhydride: a comparison with the response to 2,4-dinitrochlorobenzene. Toxicol Lett 1989; 47:25-39. 7. Hayes JP, Daniel R, TeeRD, Barnes PJ, Chung KF, Newman Taylor AJ. Specific immunologic and bronchopulmonary responses to a protein conjugate of trimellitic anhydride in guinea pigs sensitised to the free hapten. Clin Exp Allergy 1992; 22: 694-700. 8. Gordon T, Sheppard D, McDonald DM, Distefano S, Scypinski L. Airway hyperresponsiveness and inflammation induced by toluene diisocyanate in guinea pigs. Am Rev Respir Dis 1985;132:1106-12. 9. Thompson JE, Scypinski LA, Gordon T, Sheppard D. Tachykinins mediate the acute increase in airway responsiveness caused by toluene diisocyanate in guinea pigs. Am Rev Respir Dis 1987; 136:43-9. 10. Biagnini RE, Moorman WJ, Lewis TR, Bernstein IL. Ozone enhancement of platinum asthma in a primate model. Am Rev Respir Dis 1986; 134:719-25. 11. Iljima H, Ishii M, Yamauchi K, et al. Bronchoalveolar lavage and histologic characterization of late asthmatic response in guinea pigs. Am Rev Respir Dis 1987; 136:922-9. 12. Frew AJ, Moqbel R, Azzawi M, et al. T lymphocytes and eosinophils in allergen-induced late phase asthmatic responses in the guinea pig. Am Rev Respir Dis 1990; 141:407-13. 13. Sanjar S, Aoki S, Kristersson A, Smith D, Morley J. Antigen challenge induces pulmonary airway eosinophil accumulation and airway hyperreactivity in sensitised guinea pigs: the effect of antiasthma drugs. Br J Pharmacol 1991; 99:679-86. 14. Venables K. Low molecular weight chemicals, hypersensitivity, and direct toxicity: the acid anhydrides. Br J Ind Med 1989; 46:222-32. 15. Durham SR, Graneek BJ, Hawkins R, Newman Taylor AJ. The temporal relationship between increases in airway responsiveness to histamine and late asthmatic reactions induced by occupational agents. J Allergy Clin Immunol1987; 79:398-406. 16. Fabbri LM, Boschetto P, Zocca E, et al. Bronchoalveolar neutrophilia during late asthmatic reactions induced by toluene diisocyanate. Am Rev Respir Dis 1987; 136:36-42. 17. Paggiaro P, Bacci E, Paoletti P, et al. Bronchoalveolar lavage and morphology of the airways after cessation of exposure in asthmatic subjects sensitized to toluene diisocyanate. Chest 1990; 98:536-42.
JAMES P. HAYES, ROB DANIEL, ROSEMARY D. TEE, PETER J. BARNES, ANTHONY J. NEWMAN TAYLOR, and K. FAN CHUNG Introduction
T rimellitic anhydride (TMA) is known to cause occupational asthma (1) and pulmonary hemorrhage with hemolytic anemia in humans (2). Inhalation of TMA dust may provoke immediate and late asthmatic responses in sensitized persons (3). Nonspecific bronchial hyperreactivity is associated with occupational asthma, and transient increases in bronchial reactivity may follow inhalation of specific occupational agents (4). An animal model of increased airway responsiveness caused by the acid anhydrides may provide an opportunity of examining the mechanisms underlying this characteristic feature of occupational asthma. Previous workers have demonstrated systemic and pulmonary immune responses in various animal species after tracheal instillation of free TMA (5). Botham and coworkers (6) demonstrated an immunologic response in guinea pigs sensitized to free TMA administered by the intradermal route, and they also observed an alteration in respiratory rate immediately after inhalation of TMA dust in sensitized animals. We have previously reported that intradermal injection of free TMA induces a specific IgG-I, and in some animals IgE response to TMA conjugated to guinea pig albumin (GPSA). Inhaled and intravenous challenge with TMA-GPSA provokes an acute change in bronchopulmonary response in sensitized guinea pigs (7). In this study, we examined the effect of inhaled TMA dust on airway reactivity and cellular infiltration in sensitized and in nonsensitized guinea pigs 8 h after exposure. Wealso examined the time course of this response by assessing the effect of TMA dust at 2 and at 24 h after exposure in sensitized guinea pigs. Methods Sensitization Male Dunkin-Hartley guinea pigs weighing
SUMMARY We have developed In the guinea pig, an animal model of bronchial hyperreactlvlty provoked by Inhalation of trimellltic anhydride (TMA) dust, a known cause of occupational asthma in humans, after intradermal sensitization to the free hapten. Male Dunkin-Hartley guinea pigs (n = 6) were Injected intradermally with 0.1 ml of 30% TMA In corn oil. Control animals (n = 7) were injected with 0.1 ml corn 011 alone. On Days 21 to 28 after sensitization, guinea pigs were challenged (nose only) to 12 mg/m 3 of inhalable TMA dust for 30 min. Bronchial reactivity was measured in sensitized animals and in control animals at 8 h after exposure to the dust. We also measured bronchial reactivity In sensitized exposed guinea pigs at 2 h (n 5) and at 24 h (n 5). Pulmonary inflation pressure (PIP)was used to assess bronchopulmonary response. Blood samples were taken for assessment of IgG-1 antibodies to TMA conjugated to guinea-pig albumin. The concentration of acetylcholine required to Induce a 100% increase in PIP was used to assess bronchial reactivity. The lungs were eviscerated for histologic examination. All guinea pigs injected Intradermally with TMA had high titers of specific IgG-1 antibodies to TMA conjugated to guinea-pig albumin. There was a significant increase in bronchial reactivity in sensitized guinea pigs 8 h after exposure to the TMA dust compared with that In the control animals. There was also a significant eosinophilic Inflammatory influx In the subeplthelium of the sensitized groups. Exposure to free TMA dust in sensitized guinea pigs induces an increase In bronchial reactivity accompanied by an eosinophilic inflammatory exudate. This model may be suitable for assessing the underlying mechanisms of occupational asthma caused by acid anhydrides and possibly to other low molecular weight chemicals.
=
=
AM REV RESPIR DIS 1992; 146:1311-1314
200 to 250 g (n = 16)were injected intradermally with 0.1ml of 30010 TMA (Aldrich, Gillingham, UK) suspended in corn oil. Control guinea pigs (n = 7) were injected with 0.1 ml of corn oil alone.
was read on the spectrophotometer at an optical density of 420 nm. The substitution ratio for TMA-GPSA was 21:1.
Preparation of Trimellitic Anhydride Conjugate
Plates werecoated with 5 ug/ml TMA-GPSA in coating buffer and incubated overnight at 4 0 C. The next day the plates were washed four times in PBS/1\veen. Serum was diluted to 1:50solution, and plates were coated with PBS/1\veen. Serum was added in doubling dilutions across the plate and incubated for 3 h at 27 0 C. The plates were again washed four times with PBS/Tween and rabbit an-
TMA was conjugated to guinea pig serum albumin (GPSA) as follows. Four 250-mg aliquots of GPSA (Sigma, Poole, UK) were mixed with 50mlof9% NaHC03 on ice, and 250 mg of ground TMA dust were added and allowed to stand for 10min. Each aliquot was stirred separately for 1 h. The aliquots were pooled and centrifuged for 10min at 600 x g. The supernatant was separated and dialyzed with 0.02 M NH 4HC03 for 48 h with six changes. This was further dialyzed with distilled water for 24 h and lyophilized. The degree of substitution was assessed by UV spectrophotometry. Briefly 1.0 mg TMA-GPSA were added to 1 ml of 0.1 M borate buffer at pH 9.3. A 0.03-M solution of 2,4,6trinitrobenzene-sulfonic acid (TNBS) (Aldrich) in borate buffer was prepared. Then, 25 III of TNBS solution were added at 1 ml of the TMA-GPSA solution and incubated for 20 min at room temperature. The sample
ELISA for Assessment of IgG-l Antibodies
(Received in originalform August 12, 1991 and in revised form June 24, 1992) 1 From the Departments of Thoracic Medicine and Occupational & Environmental Medicine,Royal Brompton National Heart & Lung Hospital, National Heart & Lung Institute, London, United Kingdom. 2 Correspondence and requests for reprints should be addressed to K. F. Chung, Department of Thoracic Medicine, National Heart & Lung Institute, Dovehouse St., London SW3 6LY UK.
1311
1312
tiguinea pig IgO-1 (ICN Biomedicals Ltd., High Wycombe Bucks, UK), diluted to 1:2,500, was added and incubated for 2 h at 27° C. The plates were again washed with PBS/1\veen. Antirabbit IgO-peroxidase (ICN Biomedical Ltd.) was added at 1:5,000dilution and incubated at 27° C for 2 h. Finally, the plates were washed in PBS/Tween and chromogen was added. The reaction was stopped after 6 min with 1M H 2S04 and read at an optical density of 492 nm. The titers of IgO-1 were calculated from the last dilution that was double the control sera.
Trimellitic Anhydride Dust Exposure On Days 21 to 28, guinea pigs were placed in a conical container with their nostrils protruding, fixed in an exposure chamber with an internal volume of 30 L, and exposed to 12 mg/rn" of inhalable dust for 30 min. The dust was generated by passing air (12L/min) through a Wright's dust feeder above the chamber. Dust levels were measured before and during exposure gravimetrically by extracting air samples using a dust sampler (AFC 123; CaselleLondon Ltd., Bedford, UK) onto fiberglassfilters (Whatman, Maidstone, UK) at 2 L/min for 5 min. Particle size was measured by a cascade impactor (Marple, Andersen Samples Ltd., Atlanta, GA). Measurement of Bronchopulmonary Response and Bronchial Reactivity After exposure to free TMA dust, sensitized guinea pigsweredivided into three groups and were assessed at 2 h (n = 5), at 8 h (n = 6), or at 24 h (n = 5). Control animals (n = 7) were assessed at 8 h after exposure to TMA dust. All guinea pigs were anesthetized intraperitoneally with 25 g/kg urethane (Sigma), tracheostomized, and ventilated at a tidal volume of 10ml/kg and a respiratory rate of 60 breaths/min on an animal ventilator (Harvard Apparatus Ltd., Edenbridge, Kent, UK). A cannula was inserted into the left carotid artery to monitor blood pressure. Pulmonary inflation pressure(PIP) was measured through a side-arm tube attached at the origin of the outlet tube by a pressure transducer (Sensym 744; Farnell, Leeds, UK) connected to a twochannel recorder (Multitracer 2; Lectromed, Letchworth, UK). Acetylcholine (Sigma) was administered in increasing 1/2-log concentrations via an ultrasonic nebulizer (DeVilbiss Co., Somerset, PA). Each dose of acetylcholine (initial dose, 10-5 M) was administered for 20 breaths. The concentration of acetylcholine required to induce a 100070 increase in PIP (PC I OO) was used as an index of bronchial reactivity. Blood samples (0.3 ml) were taken from all guinea pigs at this time for measurement of IgG-1 antibodies to TMA-GPSA. Histopathology At the end of the experiment, the guinea pigs were killed. The lungs were inflated with 4% formol saline, and the trachea was tied. The tissue was fixed in formaldehyde, sectioned (3-JJm sections), and stained with carbolchromotrope to demonstrate eosinophil granules. The slides were coded and read without
HAYES, DANIEL, TEE, BARNES, NEWMAN TAYLOR, AND CHUNG
5 10 4
10
.
3 10
~
2 10
~
10
~
>-
i
0 000 00000 000 0000 0
!:
0 0
10 ·2
8
~
...s2
J1.
B'
~
10.3
A
~
'0
.c 0
i8'
fA
&
10.4
A
10. 5
.. ••
•
A
..J
!i!)
2h
8h
sensitized
controls
controls
Fig. 1. IgG-1 antibody titers to trimellitic anhydride conjugated to guinea pig serum albumin (TMA-GPSA) in guinea pigs given intradermal injections with 0.1 mlof 30% TMA. No antibodies to TMA-GPSA were in control guinea pigs injected with 0.1 ml corn oil alone.
prior knowledge of which group of animals they came from. In order to quantify the number of eosinophils in the airways, a section of tissue from the right or left main bronchus (chosen randomly) and a similar section diametrically opposite were selected. Eosinophils were counted within a given area using a graticule (size56 mm': magnification 400) positioned from the epithelium extending into the subepithelial layers. At the same time, mononuclear cells and polymorphonuclear cells were also counted.
Statistical Analysis Data are expressed as mean and SEM. PC 100 values were analyzed as 10glo. A paired t test was used to compare bronchial reactivity and eosinophil counts in the sensitized group assessed at 8 h and the nonsensitized group. To determine the time course of the response, bronchial reactivity and eosinophil counts in the three sensitized groups were compared using analysis of variance. A p value < 0.05 was considered significant. Results
All guinea pigs given intradermal injections of TMA had high levels of specific antibodies to TMA-GPSA (figure 1). There was no significant change in blood pressure among the groups during the course of the experiment. There was no significant difference in baseline PIP measurements between the control group and the sensitized group at 8 h. There was a significant increase in PC lO O in the sensitized group measured at 8 h (0.26 mM, 1.4, geometric mean, [GSEM]) compared with the control group (3.6 mM, 2.2 GSEM; p < 0.02). There was no significant difference in bronchial reactivity in the sensitized guinea pigs at 2 h (1.4 mM, 2.1 GSEM) and at 24 h (0.62 mM, 3.7 GSEM) compared with the sensitizedgroup at 8 h (fig-
8h
24h
sensitized
Fig. 2. Effect of inhaled trimellitic anhydride (TMA) dust on bronchial reactivity, which was measured at 2 h (closed triangles), at 8 h (open triangles), and at 24 h (closed circles) in sensitized guinea pigs. Control guinea pigs (open circles) were measured at 8 h. The concentration of acetylcholine required to induce an increase of 100% in pulmonary inflation pressure (pC.oo) was used to measure bronchial reactivity. There was a significant decrease in acetylcholine Pc'oo in sensitized animals compared with that in control animals. (p < 0.02.) Horizontal bars indicate mean values.
ure 2), although the increase in airway responsiveness was maximal at 8 h. In the control guinea pigs eosinophils were occasionally found in the extrapulmonary airways as foci beneath surface epithelium and less often within the epithelium. They were not a feature of the adventitia. Similarly, in lung sections occasional eosinophils wereseen in and beneath the surface epithelium. In the sensitized guinea pigs there were significant increases in eosinophil numbers in both the extrapulmonary airway and the lung
ie
40
!
..~
30
A
~
c
::3
I
.!! 20
i.0 c
0
Ii
a ::3
z
0
~
•
~ 9 8
• .....
t
10
'0
JE
t
t
•
8
2
24
hours
controls
sensitized
Fig. 3. Effect of inhaled trimellitic anhydride dust (12 mg/m3 for 30 min) on the number of eosinophils per unit airway tissue in sensitized guinea pigs at 2 h (closed triangles), at 8 h (open triangles), and at 24 h (closed circles). Control guinea pigs (open circles) were assessed 8 h after exposure. There was a significant increase in the number of eosinophils in airway tissue in sensitized animals compared with that in control animals. (p < 0.005.) Horizontal bars indicate mean values.
BRONCHIAL HYPERREACTIVITY
1313
CAUSED BY TRIMELLITIC ANHYDRIDE
B
Fig. 4. Photomicrograph of bronchial tissue stained with carbol chromotrope(magnification, 400) showing transverse sections of main bronchus from a guinea pig sensitized to trimellitic anhydride (A) and a control guinea pig (B). Longitudinal sections of bronchus from a sensitized guinea pig (C) and a control guinea pig (D) are also shown. The arrow indicates a typical eosinophil.
sections. In the extrapulmonary airways these occasionally extended to the deeper layers. In the lungs eosinophils were found both internal and external to the bronchial smooth muscle and between blocks of smooth muscle. Eosinophil counts were significantly greater in the sensitized group at 8 h (21.7 cells ± 1.8) compared with the control group (4.9 cells ± 1.8; p < 0.01). Occasional alveolar foci of hemorrhage were seen in both sensitized and nonsensitized animals. A small but significant increase in mononuclear cells was also seen in the sensitized group (5.0 ± 1.1, mean ± SEM) compared with the control group (2.4 ± 0.43; p < 0.05). The exact nature of these cells remains uncertain, as specific lymphocyte stains were not used. There was no significant difference in the eosinophil counts between the three sensitized groups (figure 3), but as with the changes in airway responsiveness, the eosinophil count in the subepithelium was maximal 8 h after exposure to TMA dust (figure 4). There was no significant correlation between the number of eo-
sinophils in the subepithelium and the level of bronchial reactivity in the three sensitized groups of guinea pigs. However, there was a significant correlation (p < 0.005, R = 0.73, R2 = 0.53) between the number of eosinophils in the subepithelium and airway responsiveness in the sensitized group and the nonsensitized group at 8 h after exposure to TMA dust (figure 5). Discussion
We have found that guinea pigs injected intradermally with TMA in corn oil developed evidence of a specific immunologic response, and subsequent inhalation of TMA dust provoked an increase in airway responsiveness to inhaled acetylcholine that was associated with subepithelial eosinophil and mononuclear cell infiltration. These responses did not occur in guinea pigs injected with corn oil alone. Wedid not see any increase in baseline PIP measurements in the sensitized groups compared with that in the control group. There was no significant difference in the bronchial reactivity mea-
surements and the degree of eosinophil or mononuclear cell infiltration found in the three groups of sensitized guinea pigs. However, both the bronchial reactivity and cellular infiltration were maximal 8 h after exposure to the TMA dust. Occasional alveolar foci of hemorrhage wereseen in both the control and the sen10 -6
!:
•
10 -5
0 0
U 0-
10 -4
III
.s
'0 ~ u
10 -3
su III
10 -2
~ ...J
10 -1
ClI
, 0
10
20
30
40
Number of eosinophlls per unit area (56mm2)
Fig. 5. The correlation between the number of eosinophils in the subepithelium and the concentration of acetylcholine required to induce a 100% increase in baseline pulmonary inflation pressure (PIP) in sensitized and in control guinea pigs assessed 8 h after inhalation of TMA dust. (p < 0.005, R == 0.73 R2 = 0.53.)
1314
HAYES, DANIEL, TEE, BARNES, NEWMAN TAYLOR, AND CHUNG
sitized groups. This may have been ar- in one study an association was found tefactual or a direct effect of TMA on between the number of eosinophils and the alveoli,but it would not have account- T lymphocytes in the airways of sensied for the changes seen in the sensitized tized guinea pigs. Most recently, Sanjar group in this study. Although we did not and coworkers (13) developed a model find a direct relationship between the lev- of bronchial hyperreactivity and eoel of bronchial reactivity and the degree sinophilia in bronchoalveolar lavage fluid of eosinophil or mononuclear cell in- in guinea pigs sensitized to ovalbumin. filtration in the three groups of sensitized In this study inhibition of the eosinophilguinea pigs, a strong correlation between ic response by dexamethasone did not rebronchial reactivity and eosinophil in- sult in inhibition of the increase in bronfiltration in the sensitized group assessed chial reactivity, suggesting that although 8 h after inhalation of TMA dust and there is an association between these two the control group (p < 0.005), suggest- events, a causal relationship may not ing an association (or relationship) be- exist. Acid anhydrides are known to cause tween airway reactivity and the inflammatory infiltrate. occupational asthma in humans. In the Severalstudies have examined different majority of cases it is associated with eimethods of sensitization with TMA in ther an IgE or IgG-4 antibody response. a variety of species. However, few stud- Specific inhalation tests in sensitized subies have investigated the effect of TMA jects may provoke both reduced airway on bronchopulmonary responses in pre- caliber and increased airway responsiveviously sensitized animals. Botham and ness (14). In occupational asthma the late coworkers (6), using a similar method of response has been associated with an insensitization, observed an acute change crease in nonspecific bronchial reactiviin respiratory rate in guinea pigs exposed ty, and an increase in bronchial reactivito free TMA dust. However, the exact na- ty at 3 h has been found to be predictive ture of this response is unknown. Previ- of the late asthmatic response (15). Few ously, we reported an immediate increase studies have reported the pathologic in bronchopulmonary response to TMA- changes associated with the late asthmatGPSA both by inhalation or by intrave- ic response in patients with occupationnous injection in guinea pigs sensitized al asthma caused by low molecular to TMA by intradermal injection (7). To weight chemicals. Fabbri and coworkers our knowledge this is the first report on (16) found an increase in neutrophils in the effect of a low molecular weight bronchoalveolar lavage fluid in patients chemical known to cause occupational with a late asthmatic response after exasthma on airway responsiveness and air posure to toluene diisocyanate. Paggiaro way disease in previously sensitized and coworkers (17) have reported persistence of inflammatory cells in the subanimals. Sheppard and coworkers (8, 9) found mucosa and increased eosinophils in the an increase in airway responsiveness as- bronchoalveolar lavage fluid in patients sociated with an influx of polymor- with chronic asthma induced by toluene phonuclear leukocytes into the airway of diisocyanate. In our study we demonnonsensitized guinea pigs exposed to tol- strated two important features that may uene diisocyanate; however, this increase be associated with occupational asthma, in airway responsiveness also occurred in namely, increased airway responsiveness the absence of a cellular infiltrate, and and a predominantly eosinophilic init was mediated by release of tachykinins. flammatory infiltrate. In conclusion, we have developed a This is unlikely to be a valid model of hypersensitivity-induced asthma in hu- model of bronchial hyperresponsiveness mans. Biagnini and coworkers (10) sen- and eosinophil infiltration in the airways sitized cyanomologous monkeys to plati- after exposure to free TMA dust in num salts when inhaled concurrently guinea pigs previously sensitized to the with ozone. Subsequent inhalation of the free hapten. The provocation of increased platinum salt resulted in an increase in airway responsiveness and eosinophil inbronchial reactivity to methacholine in filtration of the airway epithelium in sensensitized animals. However, the patho- sitized animals by inhaled TMA dust closely parallels changes observed in oclogic findings were not reported. Our results are similar to guinea pig cupational asthma in humans. models of sensitization and subsequent Acknowledgment exposure by inhalation to ovalbumin and Ascaris (11, 12). These studies have The writers thank Dr. Peter Jeffrey, Senior reported an increase in inflammatory Lecturer, Department of Lung Pathology, for cells in the airways and bronchoalveolar his help and advice during the course of this lavage fluid of the sensitized animals, and study.
References 1. Fawcett JW, Newman TaylorAJ, Pepys J. Asthma due to inhaled chemical agents-epoxy resin systems containing phthalic anhydride, trimellitic anhydride and triethylene tetramine. Clin Allergy 1977; 7:1-14. 2. Zeiss CR, Patterson R, Pruzansky J J, Miller MM, Rosenberg M, Levitz D. Trimellitic-anhydride induced airway syndromes: clinical and immunologic syndromes. J Allergy Clin Immunol 1977; 60:96-103. 3. Howe W, Venables KM, Topping MD, et al. Tetrachlorophthalic anhydride asthma: evidence for specific 19Bantibody. J AllergyClin Immunol1983; 71:5-11. 4. Chang-YeungM, Lam S. Occupational asthma. Am Rev Respir Dis 1986; 133:686-703. 5. Sale SR, Patterson R, Zeiss CR, Fiore M, Harris KE, Yawn D. Immune responses of dogs and rabbits to intrabronchial trimellitic anhydride. Int Arch Allergy Appl Immunol 1982; 67:329-34. 6. Botham PA, Rattray NJ, Woodcock DR, Walsh ST, Hext PM. The induction of respiratory allergy in guinea pigs following intradermal injection of trimellitic anhydride: a comparison with the response to 2,4-dinitrochlorobenzene. Toxicol Lett 1989; 47:25-39. 7. Hayes JP, Daniel R, TeeRD, Barnes PJ, Chung KF, Newman Taylor AJ. Specific immunologic and bronchopulmonary responses to a protein conjugate of trimellitic anhydride in guinea pigs sensitised to the free hapten. Clin Exp Allergy 1992; 22: 694-700. 8. Gordon T, Sheppard D, McDonald DM, Distefano S, Scypinski L. Airway hyperresponsiveness and inflammation induced by toluene diisocyanate in guinea pigs. Am Rev Respir Dis 1985;132:1106-12. 9. Thompson JE, Scypinski LA, Gordon T, Sheppard D. Tachykinins mediate the acute increase in airway responsiveness caused by toluene diisocyanate in guinea pigs. Am Rev Respir Dis 1987; 136:43-9. 10. Biagnini RE, Moorman WJ, Lewis TR, Bernstein IL. Ozone enhancement of platinum asthma in a primate model. Am Rev Respir Dis 1986; 134:719-25. 11. Iljima H, Ishii M, Yamauchi K, et al. Bronchoalveolar lavage and histologic characterization of late asthmatic response in guinea pigs. Am Rev Respir Dis 1987; 136:922-9. 12. Frew AJ, Moqbel R, Azzawi M, et al. T lymphocytes and eosinophils in allergen-induced late phase asthmatic responses in the guinea pig. Am Rev Respir Dis 1990; 141:407-13. 13. Sanjar S, Aoki S, Kristersson A, Smith D, Morley J. Antigen challenge induces pulmonary airway eosinophil accumulation and airway hyperreactivity in sensitised guinea pigs: the effect of antiasthma drugs. Br J Pharmacol 1991; 99:679-86. 14. Venables K. Low molecular weight chemicals, hypersensitivity, and direct toxicity: the acid anhydrides. Br J Ind Med 1989; 46:222-32. 15. Durham SR, Graneek BJ, Hawkins R, Newman Taylor AJ. The temporal relationship between increases in airway responsiveness to histamine and late asthmatic reactions induced by occupational agents. J Allergy Clin Immunol1987; 79:398-406. 16. Fabbri LM, Boschetto P, Zocca E, et al. Bronchoalveolar neutrophilia during late asthmatic reactions induced by toluene diisocyanate. Am Rev Respir Dis 1987; 136:36-42. 17. Paggiaro P, Bacci E, Paoletti P, et al. Bronchoalveolar lavage and morphology of the airways after cessation of exposure in asthmatic subjects sensitized to toluene diisocyanate. Chest 1990; 98:536-42.
