|Home | About | Journals | Submit | Contact Us | Français|
Trimellitic anhydride (TMA) is a small molecular weight industrial compound that will cause asthma-like symptoms in humans. Some of these TMA-induced symptoms can be reproduced in the guinea pig. In the guinea pig model of TMA-induced asthma, intratracheal instillation of TMA coupled to guinea pig serum albumin causes an immediate bronchoconstriction and increase in airway microvascular leakage with concomitant decrease in circulating platelets and white blood cells and subsequent cellular infiltration of mononuclear cells, neutrophils and eosinophils into the bronchoalveolar lavage fluid. In addition, in the lung tissue an increase in eosinophil peroxidase activity (a measure of eosinophil numbers) occurs. The purpose of this study was to determine whether complement system activation was essential for any of these TMA-induced events. Guinea pigs pretreated with cobra venom factor (CVF) had significantly reduced amounts of complement component C3 in the lavage fluid 24 hours after TMA conjugated to guinea pig serum albumin challenge indicating that the CVF treatment was successful in depleting complement proteins. Pretreatment with CVF did not affect the immediate TMA-induced bronchoconstriction nor the TMA-induced microvascular leakage. In animals depleted of the complement system by pretreatment with CVF the TMA-induced increase in mononuclear cells, total white blood cells, red blood cells, and EPO activity in the bronchoalveolar lavage was significantly reduced. Thus, our results suggest that in the guinea pig, the complement system is an important source of mediators for cellular infiltration into the lung after exposure to this acid anhydride and that inhibiting complement activation may be useful in preventing the inflammatory cell infiltration in TMA-induced asthma.
TMA is a small molecular weight chemical that is used in the paint and plastics industry. Estimates indicate that approximately 20,000 workers are exposed annually to acid anhydrides, including TMA. It is a strong respiratory sensitizer when breathed in as a dust or when applied to the skin. Syndromes in sensitized workers exposed to TMA can include: 1) an immediate and delayed onset asthma characterized by coughing, wheezing and dyspnea; 2) a late respiratory systemic syndrome characterized by cough, myalgias, fever, chills, pulmonary infiltration and pulmonary hemorrhage; 3) a pulmonary disease–anemia; and 4) an irritant response. Evidence indicates that the first three responses are immunological in nature as serum antibodies against TMA can be demonstrated (Berstein et al., 1982).
Previous studies by others have demonstrated that actively sensitizing guinea pigs by intradermal injection of TMA in corn oil and later challenging them will result in TMA-induced asthma-like conditions similar to that seen in humans (Arakawa et al., 1993; Hayes et al., 1992a). Intradermal injection of TMA generates high titers of circulating IgG1 antibodies to TMA conjugated to guinea pig serum albumin, TMA-GPSA (Hayes et al., 1992b). Upon intratracheal challenge with TMA-GPSA, these sensitized guinea pigs respond with an immediate bronchoconstriction and microvascular leakage of proteins and a subsequent inflammatory cell infiltration into the airways.
Work done by Leach et al. (1987) in a rat model of TMA-induced lung injury showed increases in absolute and relative lung weights, external hemorrhagic lung foci, alveolar hemorrhage and pneumonitis after repeated inhalation of TMA dust. Deposition of IgG and the third component of complement (C3) in the lung and mediastinal lymph nodes was also demonstrated. These data in the rat suggest that complement system activation is occurring after TMA challenge and may be a critical factor in the tissue damage and/or other TMA-induced phenomenon.
Lung tissue damage in the form of hemorrhagic foci has also been documented in sensitized guinea pigs after exposure to TMA dust (Tao et al., 1991). The guinea pigs were either actively sensitized with TMA or passively sensitized with IgG2 or IgG1 antibodies against TMA. Lung tissue damage occurred in the guinea pigs regardless of the method of sensitization. Thus, Tao et al. (1991) speculated that immune complex-mediated complement system activation was responsible for the lung tissue damage.
The complement system consists of approximately 20 proteins that circulate in the blood. This collection of proteins is part of the humoral immune system and without activation, the proteins have little or no biological activity. When the system is activated by antigen-antibody interactions, bacterial or parasite invasion, etc., many of the complement proteins acquire biological activity. Some are involved with clearance of foreign cells and particles or lysing cells, whereas some components have chemotactic activity toward granulocytes. Products of complement system activation having biological activity include C5a and C3a. Both of these compounds will promote smooth muscle contraction and induce increased vascular permeability. C5a is a 74 amino acid glycopeptide cleaved from the much larger protein, C5, during activation. It is one of the classical chemotactic agents along with leukotriene B4 and platelet-activating factor. C5a exhibits chemotactic activity toward monocytes, basophils, neutrophils and eosinophils (Glovsky et al., 1979). In rabbits, intratracheal instillation of C5 fragments (not necessarily only C5a) causes a diffuse neutrophilic infiltrate, which is maximal within 4 hr (Shaw et al., 1978). Similarly, exposure of rabbits to an aerosol of human C5a des arg leads to accumulation of granulocytes within intrapulmonary airways (Irvin et al., 1986). C3a is a 77 amino acid protein cleaved from the larger complement system protein C3. The appearance of C3a in the bronchoalveolar lavage has been used as an indication of complement system activation in the lung in asthma (van de Graaf et al., 1992) and in hypersensitivity pneumonitis (Yoshizawa et al., 1988).
TMA exposure in a sensitized guinea pig results in an immediate bronchoconstriction, increases in airway microvascular leakage and increased cellular infiltration into the lung. These TMA-induced allergic events in the lung have not been tested for complement system participation. Therefore, the purpose of this study was to determine if complement system activation is necessary for TMA-induced bronchoconstriction, increases in microvascular leakage and subsequent WBC infiltration into the lung tissue and airspace.
Guinea pigs (Charles River, Inc., Portage, MI or Sasco, Inc., Omaha, NE) weighing 200 to 250 g were sensitized on day zero by intradermal injection of 0.1 ml corn oil containing 0.30% TMA flakes (Aldrich Chemical Co., Milwaukee, WI). On day 20, guinea pigs received injections i.p. with 50 U/kg CVF or its vehicle at three times (-28, -24 and -20 hr) and once on day 21 at -4 hr before intratracheal instillation of GPSA or TMA-GPSA. Groups of guinea pigs receiving CVF pretreatment are designated as CVF TMA-GPSA in the figures and table. Groups of guinea pigs designated in the figures and table as GPSA or TMA-GPSA are pretreated with vehicle for CVF. A total of 80 guinea pigs were used in this study. TMA was conjugated to GPSA by stirring 1.5 g TMA flakes with 200 mg GPSA in 10 ml of 9% NaHCO3 for 1 hr. This preparation was dialyzed against H2O until no traces of Na-TMA were evident by spectrophotometrically scanning from 210 nm to 310 nm. This conjugate of TMA-GPSA was used in experiments measuring the bronchoconstriction and circulating cell changes. For experiments investigating cellular infiltration into the airspace, the conjugate of TMA-GPSA was freeze-dried and suspended at a final concentration of 100 mg protein/ml H2O, pH 7.2. Degree of substitution was determined by the 2, 4, 6-trinitrobenzene sulfonic acid method as described by Snyder and Sobocinski (1975) and was found to be 18 mol TMA per mol GPSA.
Guinea pigs sensitized to TMA were anesthetized with ketamine (30 mg/kg, i.m.) and xylazine (5 mg/kg, i.m.), tracheotomized and the left jugular vein cannulated for i.v. injections. Blood pressure was monitored via the femoral artery using a Statham PM23Db pressure transducer. Respiration was arrested with succinylcholine chloride (2.1 mg/kg, i.v.) and the animal was artificially respirated using a Harvard constant volume ventilator (frequency 50/min, volume 10 ml/kg). Body temperature was maintained at 37°C using a Harvard homeothermic blanket. Airflow was measured using a Fleisch 000 pneumotach connected to a Statham PM15ETC ± 25 mm H2O differential pressure transducer. Transpulmonary pressure was monitored by connecting one inlet of a Statham PM15ETC ± 07PSID differential pressure transducer to an 18-ga needle inserted through the fifth or sixth intercostal space (intrapleural pressure), and the other inlet to the end of the endotracheal tube (intratracheal pressure). The flow and transpulmonary pressure signals were fed into an on-line pulmonary mechanics computer (Model 6, Buxco Electronics, Inc., Sharon, CT) that calculated pulmonary resistance and dynamic lung compliance on a breath-to-breath basis (Amdur and Mead, 1958). For these calculations, the transpulmonary pressure and the volume are sampled at points of zero flow (i.e., the start and end of inspiration) and dynamic lung compliance is computed as the change in volume divided by the change in transpulmonary pressure. Also, transpulmonary pressure and tracheal airflow are sampled at isovolumetric levels on inspiration and expiration and the pulmonary resistance is computed as the change in pressure divided by the change in flow. The parameters were simultaneously recorded on a Grass Model 6B polygraph and digitized using a Buxco Model LS-12 data logger. Every 0.1 min the average of the breaths within that interval are recorded on an IBM XT microcomputer utilizing LOGIC software (Branch Technology, Dexter, MI). Animals were allowed to stabilize 15 to 20 min before experimental manipulations. Guinea pigs were given 200 μl (4 mg protein) TMA-GPSA by intratracheal instillation. Blood samples (0.5 ml) were taken into EDTA via a cannula in the carotid artery at -2, +2, +7 and +21 min from intratracheal instillation of TMA-GPSA. White blood cells and platelets were counted on a hemocytometer after dilution and lysis of red blood cells. Blood smears are stained with a modified Wrights' stain (Diff Quik, American Scientific Products, McGaw Park, IL) and at least 200 cells are counted and categorized as neutrophils, eosinophils or mononuclear cells as determined by their morphology.
Evans blue dye (30 mg/kg in 0.15 M NaCl) was injected i.v. 5 min before TMA-GPSA instillation. The chest was opened 21 min after TMA-GPSA instillation and an 18-ga needle was inserted through the left ventricle into the aorta and 50 ml of PBS was injected to flush dye from the bronchial circulation. The trachea and lungs were removed, weighed and the parenchyma was carefully scraped off. The trachea and main bronchi were separated from the remainder of the airway tree (intrapulmonary sections) and placed in formamide for 24 hr at 56°C. The amount of Evans blue dye in the tissue samples was determined by comparing the absorbance at 620 nm of the tissue samples with a standard curve of known amounts of Evans blue in formamide.
On day 21 sensitized guinea pigs were given the histamine antagonist, pyrilamine (6.1 mg/kg, i.p.). Thirty min later TMA-GPSA (4 mg protein in 40 μl H2O) was instilled intratracheally under ketamine and xylazine anesthesia. Twenty-four hr after instillation of TMA-GPSA, guinea pigs were given a lethal dose of pentobarbital and their lungs lavaged via a tracheal cannula with 30 ml of room temperature PBS in 5-ml increments. Five ml of PBS were instilled into the lungs via a tracheal cannula, recovered immediately and set into a siliconized glass tube on ice. Twenty-six to 28 ml of bronchoalveolar lavage fluid were recovered from the guinea pigs. After the lavage the tubes were centrifuged at 120 × g for 10 min. The BAL supernatant was collected and frozen for later analysis of protein content as described below. The BAL cell pellet was resuspended in 1.0 ml of PBS, pH 7.2 for determination of WBC numbers. One-half of the BAL cell suspension was centrifuged at 500 × g for 10 min. The supernatant was discarded and the cells were resuspended in one-half ml of 0.5% hexadecyltrimethylammonium bromide (Sigma Chemical Co., St Louis, MO) in 50 mM KH2PO4 buffer pH 6.0. This cell suspension was used for MPO analysis. For EPO analysis, 50 μl of the original BAL cell suspension was added to 375 μl PBS plus 25 μl 2% Triton in Tris buffer, pH 8.0. Both the samples for MPO and EPO were frozen, thawed and sonicated (Heat System-Ultrasonic Processor, Heat Systems Ultrasonics, Inc., Farmingdale, NY) three times. MPO and EPO activities were then measured spectrophotometrically (DU 50, Beckman Instruments, Inc., Irvine, CA) as described below. Total cell counts were done as well as differential cell counts on stained slides as described above from the half of the cell suspension in PBS. Total protein in the BAL supernatant was determined by a method described by Lowry et al. (1951). This procedure indirectly measures lung permeability because large proteins from the blood vessels are normally excluded from the lung lumen.
Red blood cells were measured in the BAL as an indication of the amount of lung damage incurred 24 hr after TMA challenge. The BAL cell suspension was freeze-thawed and centrifuged at 500 × g for 10 min to remove debris. The optical density of the supernatant was measured at 412 nm as an indication of the hemoglobin content. Ward (1979) used this method to estimate hemorrhage in lung tissue. Lysis of washed guinea pig red blood cell suspensions in water has verified that the O.D.412 is proportional to the number of RBC between 0.1 to 1 × 107 RBC/ml.
Myeloperoxidase was extracted and assayed as described for heart and skin samples (Bradley et al., 1982; Ormrod et al., 1987). The lung was removed and the right and left caudal lobes and each of one-half of the trachea plus main bronchi were suspended in 3 ml 0.5% hexadecyltrimethylammonium bromide in 50 mM potassium phosphate buffer, pH 6.0 and then homogenized for 1 min on ice using a Janke and Kunkel Ultra-Turrax T25 tissue processor (Janke and Kunkel GmbH & Co., Staufen, Germany). Those lung samples destined for MPO assay were then freeze-thawed three times, sonicated for 15 sec, centrifuged at 12,500 × g for 30 min at 4°C and the supernatant frozen until assayed.
For the MPO assay, the hydrolysis of 0.15 mM hydrogen peroxide was monitored by the change in O.D. at 460 nm of 0.5 mM o-dianisidine hydrochloride (Sigma Chemical Co., St. Louis, MO) in 50 mM potassium phosphate buffer, pH 6.0, containing 5 mM aminotriazole in a 1 ml reaction volume at 25°C. Units of MPO activity were calculated as outlined in the Worthington Enzyme Manual (1972) and the activity expressed as units MPO/g lung tissue. In the absence of hydrogen peroxide there was no change in the O.D. at 460 nm indicating the activity being measured was a peroxide-dependent event. The activity was completely inhibited by 0.1 mM sodium azide but not inhibited by 5 mM aminotriazole indicating that the activity being measured was due to MPO rather than eosinophil peroxidase or catalase. We have measured MPO activity from isolated guinea pig neutrophils and found that MPO activity increased proportionally from 105 to 2 × 107 cells with a correlation coefficient (R2) of 0.704 and that 2 × 107 cells contain approximately 2 units MPO which is in close agreement with other studies (Bradley et al., 1982; Ormrod et al., 1987). MPO measurements did not correlate with the numbers of eosinophils in BAL (R2 of 0.053).
Eosinophil peroxidase activity measurements have been described by Strathet al.(1985) andChenget al.(1993). Lung tissue was homogenized as described above for myeloperoxidase activity. The homogenate was then centrifuged at 400 × g for 30 min and the pellet resuspended in 4 ml of the same buffer. These samples were sonicated and freeze-thawed three times. EPO activity in the pellet was assayed by monitoring O.D. changes at 490 nm using the substrate solution 0.1 mM o-phenylenediamine dihydrochloride (Sigma) in 0.05 M Tris-HCl, pH 8.0, containing Triton X-100 and 1 mM hydrogen peroxide. We have verified that aminotriazole inhibits this peroxidase activity. We have also determined that EPO activity in the lavage increases proportionally with numbers of eosinophils in the BAL (R2 of 0.793) but not with the numbers of neutrophils (R2 of 0.096). TMA-GPSA treatment had no effect on EPO activity/106 eosinophils in the BAL.
Antibody for the ELISA was an IgG fraction of goat anti-guinea pig C3 from Cappel Research Products, Durham, NC. Alkaline phosphatase was conjugated to this antibody using glutaraldehyde by standard methods (Hornbeck, 1991). Wells of polystyrene plates were coated with a 50 μl volume of IgG fraction of goat anti-guinea pig C3 in PBS and 0.05% NaN3 for 16 hr at room temperature. The remaining nonspecific binding sites were blocked by incubation with 50 μl/well of a blocking buffer containing 0.1% bovine serum albumin in 0.17 M H3BO3, 0.12 M NaCl, 0.05% Tween 20, 0.05% NaN3 and 1 mM EDTA for 30 min at 37°C. The BAL diluted in PBS was added and incubated for 2 hr at 37°C. The plates were again washed three times and 50 μl of blocking buffer containing BSA was added for 10 min. After washing, 50 μl of alkaline phosphatase-labeled goat anti-guinea pig IgG antibody to C3 were added to the wells and incubated for 2 hr at 37°C. After washing three times, 50 μl of the alkaline phosphatase substrate p-nitrophenyl phosphate (1 mg/ml, Sigma) in 2-amino-2-methyl-1, 3 propanediol, pH 10.0, was added. After 60 min the optical density at 405 nm was measured on an ELISA reader. Guinea pig C3 concentrations in the BAL were calculated from a calibration curve obtained by using a guinea pig C3 standard (Diamedix Corp, Miami, FL). This ELISA showed no activity to the carboxy terminal 20 amino acids of guinea pig C3a or to guinea pig C5.
Guinea pig serum albumin, bovine serum albumin, Evans blue, formamide, H3BO3, NaCl, pyrilamine, Triton X-100, hydrogen peroxide, sodium azide, Tween 20, p-nitrophenyl phosphate, 2-amino-2-methy1-1, 3 propanediol and aminotriazole were purchased from Sigma. Evans blue dye was purchased from Fischer Scientific Co., Fair Lawn, NJ. Ketamine was purchased from Aveco, Inc., Fort Dodge, IO. Xylazine was purchased from Mobay Corp., Shawnee, KS. Cobra venom factor (Diamedix Corp., Miami, FL) is a protein isolated from the venom of cobras (Naja naja kaouthia). When injected into animals it will cause a marked decrease in C3 levels. The potency of CVF is expressed in units based on the reciprocal of the highest dilution of the preparation that when incubated at 30°C for 60 min with an equal volume of guinea pig serum (complement source) diluted 1:10, will lower the C3 titer by at least 95%. The vehicle for CVF was 0.15 M NaCl, 0.005 M sodium phosphate, 0.00015 M calcium chloride, 0.005 M magnesium chloride and 0.5% gelatin at pH 7.5.
In experiments to determine the effect of complement system depletion by CVF pretreatment on the infiltration of cells into the lungs an analysis of variance was used to test if significant differences existed between and among treatments. Dunnett's t test was used to compare vehicle and CVF TMA-GPSA treatments with TMA-GPSA treatment. TMA-GPSA treatment was compared to CVF TMA-GPSA treatment by Mann-Whitney U test where appropriate. For determining the effects of complement system depletion on the time course of TMA-induced percent decrease in compliance and percent increase in resistance and changes in circulating blood cell counts, the two-tailed t test used was Satterthwaite's approximation which does not assume equal variances (Snedecor and Cochran, 1980). Level of significance for all tests is P ≤ .05 and is indicated in the figures by an asterisk or a dagger.
The purpose of this study was to test our hypothesis that the complement system was necessary for TMA-induced bronchoconstriction, circulating cell changes and microvascular leakage into the airway tree. To test our hypothesis we pretreated guinea pigs with CVF to deplete the alternative pathway of the complement system. Actively sensitized guinea pigs were challenged by intratracheal instillation of TMA-GPSA and the percent increase in pulmonary resistance and percent decrease in compliance were measured as described in “Methods. ” Control animals were sensitized to TMA and challenged with GPSA. As seen in figure 1 pretreatment with CVF to deplete the complement system had no effect on the TMA-induced bronchoconstriction over the monitoring period of 20 min. Maximum percent increase in pulmonary resistance for normal and complement depleted guinea pigs was 145.4 ± 59.7 and 123.8 ± 35.8, respectively. Percent decrease in dynamic lung compliance was -78.1 ± 5.8 and -76.3 ± 6.6 for the two groups. TMA-GPSA challenge also caused a significant and transient decrease in the number of WBC and an immediate and sustained decrease in the numbers of circulating platelets (fig. 2). Depletion of the complement system prior to intratracheal instillation of TMA-GPSA had no effect on the transient decrease in the number of circulating WBC. However, CVF pretreatment significantly attenuated the TMA-induced decrease in numbers of circulating platelets at +7 and +21 min after instillation. In guinea pigs not depleted of the complement system, numbers of circulating platelets remained low at the 21-min time point showing that they did not return to the circulation after the bronchoconstriction had attenuated.
Airway microvascular leakage was measured 21 min after TMA-GPSA challenge by extravasation of Evans blue dye into the trachea and bronchi or in the remainder of the airway tree. We found that the amount of Evans blue was not altered between guinea pigs depleted of their complement system by pretreatment with CVF (66.8 ± 13.3 ng/mg wet weight of trachea plus main bronchi) and those guinea pigs pretreated with the vehicle for CVF (72.8 ± 11.5 ng/mg wet weight of trachea plus main bronchi). However, there was significantly greater amounts of Evans blue per mg wet tissue weight in the trachea and bronchi when compared to the remainder of the airway (25.0 ± 3.1 ng/mg and 21.8 ± 2.7 ng/mg wet weight of intrapulmonary section in CVF-treated and vehicle-treated animals, respectively). Evans blue extravasation into the lung tissue was measured only at 21 min after instillation of TMA. In our previous studies untreated guinea pig lungs had less than 10 ng Evans blue dye/mg wet weight of lung tissue (Regal et al., 1993a).
In another group of guinea pigs we measured TMA-induced cellular infiltration 24 hr after instillation of GPSA or TMA-GPSA. Total WBC in the lavage increased significantly 24 hr after instillation of TMA-GPSA when compared to GPSA treated animals (fig. 3). Pretreating guinea pigs with CVF to deplete their complement system resulted in a significant decrease in the numbers of total WBC in the lavage when compared to animals instilled with TMA-GPSA but not depleted of their complement system. Differential counts of the WBC recovered from the BAL are shown in figure 4. A significant increase in numbers of mononuclear cells, neutrophils and eosinophils in the BAL 24 hr after TMA-GPSA instillation was measured when compared to animals instilled with GPSA. CVF pretreatment significantly reduced the TMA-induced increase in mononuclear cells in the BAL.
Total MPO and EPO activities also increased in BAL 24 hr after instillation of TMA-GPSA when compared to animals instilled with GPSA (fig. 5). Depletion of the complement system by pretreatment with CVF had no effect on the TMA-induced increase in MPO in the lavage 24 hr after instillation, but did have a significant effect on the EPO in the lavage. The data in figure 5 suggest that the complement system is important in TMA-induced eosinophil infiltration but not neutrophil infiltration into the airspace.
TMA-GPSA instillation caused an increase in EPO activity in the lung tissue but not the trachea (table 1). The instillation of TMA-GPSA did not affect the MPO activity in either the lung or the trachea. CVF pretreatment did not significantly affect the TMA-induced increase in EPO activity in the lung.
Total protein in the BAL was measured as an indication of lung microvascular permeability. Protein recovered in the BAL 24 hr after challenge was significantly increased due to the instillation of TMA-GPSA (fig. 6). CVF pretreatment did not significantly affect the amount of protein in the BAL 24 hr after TMA-GPSA instillation. Red blood cells also increased in the BAL 24 hr after instilling TMA-GPSA (fig. 7). Pretreating guinea pigs with CVF to deplete the complement system proteins resulted in a significant reduction in the TMA-induced RBC accumulation in the BAL.
CVF was used in this study to deplete guinea pigs of complement system proteins. In our laboratory, we verified that CVF treatment by itself did not alter the base line compliance and resistance, microvascular leakage or cellular composition of the bronchoalveolar lavage fluid (unpublished data). Guinea pigs were actively sensitized to ovalbumin, pretreated with CVF and exposed to an aerosol of normal saline solution. The BAL parameters in these animals 24 hr after saline aerosol were not significantly different from the control animals. These experiments demonstrate that pretreating guinea pigs with CVF using our protocol has no significant effects on the parameters measured in this study except to deplete the alternative pathway complement proteins. By depleting the guinea pigs of complement proteins, we were able to test whether activation of the complement system was required in any of the phenomena resulting from TMA-GPSA instillation. Our previously unpublished studies completed in our laboratory had demonstrated that a similar injection schedule of CVF resulted in more than 97% reduction in the total hemolytic complement activity in serum without significant changes in WBC populations occurring in the BAL or lung tissue. To insure that CVF treatment was also depleting complement proteins locally in the lung we used an ELISA method to measure C3 in the BAL. C3 is in measurable quantities in guinea pig lungs instilled with GPSA (fig. 8). C3 quantities in the recoverable lavage fluid are significantly increased in animals instilled with TMA-GPSA. This may reflect an increased pulmonary microvascular permeability as shown in the BAL protein data (fig. 6). C3 quantities in the lavage fluid from guinea pigs pretreated with CVF and instilled with TMA-GPSA are significantly reduced from that found in animals not pretreated with CVF and instilled with GPSA. These data confirm that pretreating animals with CVF will deplete the complement system, both in the circulation and locally in the lung.
The purpose of this study was to determine if the complement system is important in TMA-induced bronchoconstriction, circulating cell changes, airway microvascular leakage and the subsequent cellular infiltration into the lungs. We used a guinea pig model of TMA-induced asthma that has been thoroughly characterized by other researchers (Arakawa et al., 1993; Hayes et al., 1992a; Hayes et al., 1992b). Dermal exposure of the guinea pig to TMA results in TMA-induced asthma-like conditions similar to that seen in humans (Arakawa et al., 1993; Hayes et al., 1992a), including bronchoconstriction and cellular infiltration. We did not attempt to measure a late onset physiologic response in our model. Previous studies of Pauluhn and Eben (1991) using respiratory rate as an indicator had not demonstrated a late onset respiratory response in the guinea pig. In our study, TMA-GPSA instillation in sensitized guinea pigs results in an immediate bronchoconstriction, with a transient decrease in circulating WBC and sustained decrease in circulating platelets, an increase in microvascular leakage as measured by Evans blue dye and an increase in WBC infiltration into the lungs. The WBC infiltration is due to an increase in mononuclear cells, neutrophils and eosinophils. Thus, this model mimics many aspects of TMA-induced asthma in humans, including immediate bronchoconstriction and airway microvascular leakage followed by cellular infiltration.
High titers of circulating IgG1 antibodies to TMA-GPSA are present in this guinea pig model as measured by ELISA (Hayes et al., 1992b). TMA is a highly reactive chemical and is readily able to bind to nucleophilic groups such as lysine and cysteine on tissue or serum proteins (Botham et al., 1989). Similarly, in humans, Zeiss et al.(1992) have shown that exposure to TMA dust results in circulating antibodies specific to TMA conjugated to human serum albumin. Therefore, to elicit the asthmatic response in the sensitized guinea pig we used intratracheal challenge with the preformed TMA-protein conjugate. This exposure eliminates possible irritant effects due to free TMA.
Leach et al.(1987) demonstrated C3 deposition in lung tissue of sensitized rats after inhalation of TMA, indicating that the complement system had been activated. Leach et al.(1987) concluded that the concurrent lung damage was consistent with the “immune complex injury syndrome” as described by Ward (1979).Tao et al.(1991) demonstrated that TMA-induced pulmonary hemorrhage in the guinea pig is similar to that seen in the rat. TMA challenge of guinea pigs passively sensitized with the complement fixing antibody IgG2 to TMA results in pulmonary hemorrhage, prompting them to speculate that immune complexes and complement activation were also important in TMA-induced pulmonary hemorrhage. Both of these studies suggest that TMA-induced complement system activation occurs. In addition, products of complement system activation are known to be potent agents in causing bronchoconstriction, microvascular leakage and cell infiltration. Therefore, we proposed that the complement system was necessary for the TMA-induced bronchoconstriction, circulating cell changes, cellular infiltration into the lungs and microvascular leakage. The results of our studies suggest that after TMA-GPSA instillation the complement system is activated and is essential for changes in circulating platelets, changes in eosinophil and mononuclear cell infiltration into the airspace, as well as the hemorrhage as measured by increased RBC in the BAL.
CVF was used to deplete the complement system in guinea pigs. Adequate doses of CVF given i.p. to a guinea pig will activate the complement system and subsequently deplete the alternative pathway components of the complement system, leaving C1, C4 and C2 intact. Activation of the alternative pathway generates C3a, C5a and the membrane attack complex. C3a and C5a can cause increases in microvascular leakage, smooth muscle contraction, release of lysosomal products and formed amines stored in cells and increased adherence of neutrophils. Nonlethal concentrations of the membrane attack complex can stimulate the in vitro production and release of inflammatory and procoagulant mediators, including oxygen radicals generated by leukocytes, release of prostaglandins, interleukin-1 and tumor necrosis factor from glomerular mesangial cells and expression of procoagulant activity in platelets and endothelial cells (Nicholson-Weller and Halperin, 1993).
The mediators of the TMA-induced bronchoconstriction in the guinea pig have been determined by Hayeset al.(1992a) and Arakawa et al. (1993) using the guinea pig model of occupational asthma. The major mediator of the TMA-induced bronchoconstriction was found to be histamine with cyclooxygenase products also playing a role. Our study tested whether complement system activation was necessary for the TMA-induced bronchoconstriction. Complement system activation has been shown to occur after i.v. antigen in guinea pigs actively sensitized to ovalbumin (Regal et al., 1993b) and studies have demonstrated that C5a can cause airway contraction via the release of histamine and arachidonate metabolites (Regal and Fraser, 1990). Our study tested whether both of these phenomena occurred to cause the TMA-induced bronchoconstriction. CVF pretreatment did not inhibit the TMA-induced bronchoconstriction, indicating that products of complement system activation are not important in the generation of histamine and cyclooxygenase products necessary for the TMA-induced bronchoconstriction.
Measurements of Evans blue dye can be used to estimate lung permeability changes and have been shown to correlate highly with the extravasation of radiolabeled albumin into the guinea pig airways (Rogers et al., 1989). Several researchers previously demonstrated an increase in Evans blue from lung tissue after TMA challenge (Hayes et al., 1992c; Tao et al., 1991). Using antagonists and synthesis inhibitors, the TMA-induced microvascular permeability was demonstrated to be due to histamine with a minor role for lipoxygenase products. Evans blue dye extravasation was measured in these studies (Hayes et al., 1992c) at 6 min post TMA challenge and did not differ significantly from values we measured at 20 min. In our study, pretreatment with CVF did not inhibit the TMA-induced increase in airway microvascular permeability, suggesting that the complement system is not involved in this aspect of the asthmatic response. It appears that histamine is the major mediator of this phenomenon and even though products of complement system activation can cause release of histamine from mast cells, the complement system is not essential for TMA-induced microvascular leakage. Another indication of the increase in microvascular leakage is the appearance of increased amounts of protein in the BAL. Pretreating sensitized guinea pigs with CVF to deplete the complement system had no effect on the amount of protein recovered from the BAL.
Circulating cell changes accompanying instillation or aerosol antigen have been documented in our laboratory and in the laboratory of Vannieret al. (1991).Vannier et al.(1991) found that after aerosol or instillation of antigen, no changes in circulating platelets occurred with an initial drop in circulating WBC. In the present study, we have shown that both WBC and platelets transiently disappeared from the circulation after intratracheal TMA-GPSA challenge in sensitized guinea pigs. In those animals depleted of the complement system by pretreatment with CVF, no significant difference was measured in circulating WBC after antigen challenge when compared to control animals. However, in those animals pretreated with CVF there was significantly less of a change in circulating platelets compared to the platelets in the control group over the same time period. Whether these cells play any role in the TMA-induced bronchoconstriction has not been determined. Experiments by Regal (1988) involving selective depletion of platelets and/or WBC in the guinea pig showed no effect of platelet depletion on C5a-induced bronchoconstriction suggesting that if the complement system is activated in our study a C5a-induced platelet change would not influence the concomitant bronchoconstriction. Lellouch-Tubiana et al.(1988) found that depletion of platelets in passively sensitized guinea pigs caused an elimination of the i.v. antigen-induced bronchoconstriction in some of the guinea pigs and a significant reduction in granulocytes found in the lung tissue 24 hr later. Inasmuch as both the TMA-induced circulating platelet drop and cell infiltration were attenuated by pretreating guinea pigs with CVF, we might speculate that the platelets in this model of occupational asthma are involved in the later TMA-induced cell infiltration into the lung BAL.
Our study has shown that complement system activation is essential for the eosinophil and mononuclear infiltration 24 hr after TMA instillation. Hayes et al.(1992c) and Obata et al.(1992) demonstrated a TMA-induced eosinophil infiltration into the lungs of sensitized guinea pigs after exposure to TMA without determining the chemotactic mediators. Hayeset al.(1992c) showed that eosinophils accumulate in the lung tissue of sensitized guinea pigs 8 hr after inhalation challenge with TMA. They made no mention of neutrophils in their study but did state that they measured a small but significant increase in mononuclear cells in their sensitized group compared with the control group. Obata et al.(1992) also measured eosinophil infiltration in the guinea pig with a different sensitization protocol and different inhalation concentration. They detected no increase in neutrophils in the lung tissue in this study and speculated that eosinophils were attracted by the chemotactic substances released during the mast cell degranulation. Identity of the mediators responsible for the TMA-induced eosinophil infiltration was not investigated. We have evidence from this study that components of the complement system are responsible in part for the WBC infiltration. TMA-GPSA instillation into sensitized guinea pigs caused an increase in PMN and mononuclear cells, as well as eosinophils in the recovered BAL fluid. When pretreated with CVF to deplete the complement system, significantly fewer mononuclear cells and eosinophil peroxidase activity were measured 24 hr after TMA-GPSA challenge. The number of eosinophils in the BAL was reduced by CVF pretreatment but the reduction was not statistically significant. Our data demonstrate that eosinophil peroxidase activity correlates closely with the numbers of eosinophils in the BAL. EPO activity may be a more accurate and less variable measure of the numbers of eosinophils than differential counting because the sampling volume is more representative of the whole. Our results are consistent with studies by Roska et al.(1977) showing that CVF pretreatment inhibited an ovalbumin-induced granulocyte infiltration into the lungs of guinea pigs.
Mononuclear cells increased in the BAL after TMA challenge similarly to the increase seen by Tarayre et al.(1991) after ovalbumin challenge. In our study, pretreating guinea pigs with CVF inhibited the TMA-induced mononuclear cell and eosinophil increase in the BAL, indicating that complement system activation was necessary for this cellular infiltration. TMA-induced lung hemorrhage as measured by O.D.412 was also significantly inhibited by pretreating guinea pigs with CVF. Activated mononuclear cells as well as eosinophils can release cationic proteins and active oxygen products via NADPH oxidase. These released substances can cause cell damage leading to hemorrhage (McCarthy and Henson, 1979; Bender and Van Epps, 1985). Both cell types appeared to accumulate in the lung after TMA challenge due in part to activation of the complement system. Products of complement system activation have been shown to activate the granulocyte and macrophage NADPH oxidase in vitro (Gerard et al., 1986; Marshall and Lands, 1986). Although not tested directly, these data suggest that mononuclear cells and eosinophils in the lung are responsible for the TMA-induced hemorrhage.
The purpose of this study was to determine if the complement system is important in TMA-induced bronchoconstriction, circulating cell changes, cellular infiltration into the lungs and airway microvascular leakage. Depletion of the complement system with CVF did not affect the immediate TMA-induced bronchoconstriction or the TMA-induced microvascular leakage. However, we found that products of complement system activation are involved in TMA-induced circulating platelet changes, TMA-induced WBC infiltration into the lungs, TMA-induced increases in EPO activity in the BAL and TMA-induced hemorrhage. Thus, our results suggest that in the guinea pig, the complement system is an important source of mediators for cellular infiltration into the lung after exposure to this acid anhydride and that inhibiting complement activation may be useful in preventing the inflammatory cell infiltration in TMA-induced asthma.
The authors thank Ms. S. Kurki for secretarial assistance.
1Supported by grants from the Minnesota Medical Foundation and the Minnesota Affiliate of the American Heart Association.