INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Mast cells, as well as eosinophils and T lymphocytes, have been implicated in the pathogenesis of the inflammatory process in the airways of asthmatic subjects (1). Mast cells trigger the early airway response (EAR) to allergen challenge through their release of preformed or de novo synthesized mediators such as histamine, tryptase, prostaglandin D₂ (PGD₂), platelet-activating factor (PAF), leukotriene (LT) C/D₄ (2, 3), and B₄ which can provoke bronchospasm, microvascular leakage, mucosal edema, and mucous gland hypersecretion (4, 5). Moreover, recent observations that mast cells are cellular sources of proinflammatory cytokines, such as tumor necrosis factor alpha (TNF-), interleukin-4 (IL-4), IL-5 (6, 7), and IL-13 (8), suggest that these cells might play an important role not only in the initiation but also in the maintenance of airway inflammation and the associated changes in pulmonary function. The activation of mast cells localized in the airway is mainly mediated by interaction between antigen-specific IgE and high-affinity IgE receptors expressed on the surface of these cells but also through low-affinity IgG receptors (9) and stem cell factor (10).

Over the last few years, adhesion molecules have been shown to play a critical role not only in leukocyte interaction and migration to the sites of airway inflammation but also in cell activation as demonstrated with eosinophils and T lymphocytes (11). Among them, the very late antigen-4, VLA-4 (CD49d/CD29) integrin, which is predominantly expressed on lymphocytes, monocytes, and eosinophils interacts with either the vascular cell adhesion molecule-1 (VCAM-1) on target tissue (12, 13) or extracellular matrix proteins such as fibronectin (14). VLA-4 is believed to modulate airway inflammation after allergen challenge by contributing to the recruitment of lymphocytes and eosinophils into the airways and/or by inducing the release of proinflammatory mediators (15).

A number of recent studies reported that blocking VLA-4 with specific antibodies could prevent both allergic EAR and late airway responses (LAR) (15, 16) and reduce so-called nonspecific airway hyperresponsiveness (16) in allergen-challenged animals. The inhibitory actions of anti-VLA-4 monoclonal antibodies (mAb) have been attributed mainly to preventing migration and/or activation of eosinophils and lymphocytes as well as mast cells in the sites of airway inflammation. However, the mechanism by which anti-VLA-4 mAb can affect the allergic EAR is still not well understood. The objectives of this study were to investigate whether blocking ₄-integrins before allergen-induced airway responses and lung inflammation is related to mast cell degranulation and inflammatory mediator release. This study was performed in the Brown Norway rat, sensitized and challenged with ovalbumin (OA) which is a well-established animal model of allergic airway inflammation (20).

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animals and Sensitization Procedure

Highly inbred male Brown Norway rats (SSn substrain), 6 to 8 wk of age and weighing 200 to 260 g at the time of study, were purchased from Harlan Sprague-Dawley UK Inc. (Blackthorn, UK) and maintained in conventional animal facilities at McGill University (Montreal, PQ, Canada). Rats were actively sensitized with a subcutaneous injection of 1 ml of normal saline containing 1 mg ovalbumin (OA; Sigma Immuno-chemicals, St. Louis, MO) and 4.28 mg of aluminum hydroxide gel (Anachemia Chemicals, Montreal, PQ, Canada) as an adjuvant. Simultaneously, 0.5 ml of Bordetella pertussis vaccine containing 6 × 10⁹ heat-killed bacilli (Armand-Frappier Institute; Laval-Des-Rapides, PQ, Canada) was injected intraperitoneally. Animals were studied 14 d after sensitization.

Animal Preparation

The rats were anesthetized with ethyl carbamate (urethane; Sigma Immuno-chemicals) administered intraperitoneally (1.1 g/kg), and placed on a heating pad (36° C) after orotracheal intubation (6 cm length of PE240 polyethylene tube) and surgical procedures. The external jugular vein was cannulated with a small polyethylene catheter (PE20) for the purpose of intravenous injection of antibodies. The common bile duct was exposed and cannulated (PE20 polyethylene catheter) after ligation of the duodenal end. For consecutive periods of 1 h starting 1 h before challenge bile samples were collected on ice under a stream of argon and were stored at 80° C prior to analysis. The rats were allowed to stabilize for a period of 2 h prior to challenge.

Study Protocol

Two series of experiments were performed using two different methods of allergen challenge, by aerosolization and by insufflation. The OA-sensitized rats were divided into three different groups, which were treated as follows: (1) Group 1 (n = 27) challenged with OA 1 h after receiving irrelevant isotype-matched control IgG₁ monoclonal antibody (B9, 2 mg/0.5 ml), (2) Group 2 (n = 15) challenged with OA 1 h after receiving anti-rat VLA-4 monoclonal antibody (TA-2, 2 mg/ 0.5 ml), and (3) Group 3 (n = 13) challenged with bovine serum albumin (BSA) (Sigma Immuno-chemicals) after receiving 0.5 ml of normal saline. Fourteen days after the sensitization, all rats were challenged with either aerosolized OA or BSA (5% wt/vol in saline) using a Hudson nebulizer (Model 1400; Hudson, Temecula, CA) with an airflow of 8 L/min for 5 min, or tracheal insufflation of 0.1 ml of OA or BSA solution (1%) through the endotracheal tube. The level of biliary cysteinyl-LTs and the physiological changes of lung resistance were measured during 8 h in rats challenged with aerosolized OA with TA-2 pretreatment (n = 7) or without TA-2 (n = 13), or BSA (n = 7). The level of histamine and tryptase-like activity in bronchoalveolar lavage (BAL) as well as the cell populations in the lavage were determined in rats challenged with insufflation of OA (n = 15) or BSA (n = 8). The administration of allergen by insufflation was chosen in order to maximize the magnitude of the allergic EAR but was not performed for more prolonged observations because of the increased risk of mortality in sensitized rats following this form of challenge.

Monoclonal Antibodies

An anti-VLA-4 mAb, TA-2, was prepared by immunizing BALB/c mice with the rat peritoneal lymphocytes as previously described (20). This IgG₁ mAb reacts with all rat lymphocytes and blocks adhesion to rat microvascular endothelial cells stimulated with cytokines (21) and reacts with lymphocytes and eosinophils of Brown Norway rats (unpublished observations). An isotype matched mAb against pertussis toxin B9 (IgG₁) was used as control. Stock mAbs were diluted with saline and filter sterilized through a 0.22-µm filter (Millipore Co., Bedford, MA).

Measurement of Pulmonary Mechanics

Measurement of airway responses was performed in animals that were anesthetized with urethane and placed in the supine position. The end of the endotracheal tube of orotracheally intubated animals was placed inside a small Plexiglas box (volume 265 ml) and a Fleisch pneumotachograph coupled to a differential pressure transducer (Micro-Switch 163PC01D36; Honeywell, Scarborough, ON, Canada) was attached to the other end of the box to measure airflow. Volume was obtained by numerical integration of the flow signal. Changes in esophageal pressure (Pes) were measured using a saline-filled catheter placed in the lower third of the esophagus and connected to one port of a differential pressure transducer (Transpac II disposable transducer; Sorenson, Salt Lake City, UT); the other port was connected to the Plexiglas box. Transpulmonary pressure was obtained by subtraction of Pes from the pressure in the Plexiglas box. Pulmonary resistance (RL) was determined by fitting the equation of motion of the lung to the data by multiple linear regression analysis (15) using a commercial software package (RHT Infodat Inc., Montreal, PQ, Canada).

Pulmonary resistance was measured just before challenge (baseline) and at 2, 5, 7, 10, 15, 20, 25, and 30 min after challenge, and every 15 min thereafter until 8 h after challenge. In the rats challenged with insufflation of OA or BSA solution, the measurement was performed until 10 min after challenge, just before the first BAL.

Measurement of Cell Populations in BAL

BAL was performed in the rats challenged by insufflation of OA or BSA solution. The first lavage was performed 10 min after challenge, a time previously reported to be optimal for maximal histamine release from mast cells in vivo (22). It consisted of an instillation of 2 ml of sterile saline via the tracheal tube and reaspiration within 30 s. BAL fluid was centrifuged (1,400 rpm, 10 min, 4° C) and supernatants were aliquoted into 0.25 ml in Eppendorf tubes and stored at 80° C until assayed for histamine. The volume of BAL fluid recovered ranged from 0.8 to 1.2 ml. The second lavage was performed 60 min after challenge, at which time tryptase release from mast cells in vivo appears to be maximal (22). For this lavage, 10 ml of sterile apyrogenic saline was introduced through the endotracheal tube and, after centrifugation, supernatants were aliquoted into 0.5 ml in Eppendorf tubes and stored at 80° C until tryptase assay. The volume of BAL fluid recovered ranged from 6.3 to 8.4 ml. The pellets were resuspended in 5 ml of phosphate-buffered saline (PBS), the cells were counted using a hemacytometer, and cell viability was assessed by the trypan blue dye exclusion test. Cytospin slides were prepared using Cytospin model II (Shandon, Pittsburgh, PA) and air-dried for 5 min and stained with May-Grunwald Giemsa for the differential cell counts.

Measurement of Mast Cell-derived Preformed Mediators

The concentrations of histamine in BAL were measured by modification of a colorimetric assay using the Pauly reaction based on the presence of the imidazole group in histamine, as initially described by Marwaha and coworkers (23). Before performing the histamine assay, the proteins contained in the BAL fluids were precipitated to avoid the occurrence of the Pauli reaction with the imidazole group contained in histidine residues. Briefly, 0.125 ml of HCl 1.4 N were added to 0.5 ml of BAL fluids and incubated at room temperature for 2 min. After centrifugation (600 × g, 5 min, 20° C) 0.5 ml of the supernatant was placed in borosilicate-glass tubes and 0.1 ml of 1% sulfanilic acid (Sigma Immuno-chemicals) and 0.1 ml of 5% aqueous sodium nitrite (Sigma Immuno-chemicals) solution were added. After a 10-min incubation at room temperature, 1.3 ml of 5% aqueous sodium carbonate (Sigma Immuno-chemicals) was added and 2 min later 1 ml of 75% of ethanol solution was added. Aliquots of 200 µl of the reaction mixture were put in 96-well microplates and the absorbance of the orange- red-colored complex was measured at 530 nm within 20 min using an ELISA plate reader (400 ATC; SLT Labinstruments, Salzburg, Austria). The blank (0.5 ml of saline) and the standard curve done with different concentrations of histamine (Sigma Immuno-chemicals) (10 to 40 µg/ml) were treated following the same procedure. In some BAL samples, aliquots were diluted (usually 1:2) when the absorbance was beyond the linear range of the standard curve. Results were expressed as microgram of histamine per milliliter of BAL fluid.

The concentrations of tryptase in the BAL were assessed indirectly by the measurement of the intrinsic trypsin-like activity using a modification of the quantitative colorimetric assay described by Lavens and coworkers (24) where benzoyl-DL-arginine-p-nitroaniline (BAPNA) was demonstrated to be a specific substrate for tryptase without cleavage by either neutrophil lysates or chymotrypsin, and mast cell-derived tryptase measured colorimetrically was highly correlated with the level of tryptase determined by radioimmunoassay (24). Briefly, 0.1 ml of BAL samples were mixed with 0.1 ml of 0.8 mM BAPNA (Sigma, St. Louis, MO) diluted in TRIS buffer (0.1 M Trizma base, 1 mM calcium chloride, 1 mM magnesium chloride, and 1 M glycerol) and the plates were incubated at 37° C for up to 72 h in 96-well microplates. The appearance of nitroaniline was then measured at 410 nm using an ELISA plate reader (400 ATC; SLT Labinstruments). The blank (0.5 ml of saline) and the standard curve with concentrations of trypsin ranging from 0.01 to 1 ng/ml (type IX from porcine pancreas; Sigma Immuno-chemicals) were performed using the same procedure. In some BAL samples, aliquots were diluted (usually 1:2) when the absorbance was over the linear range of the standard curve. Results were expressed as microgram of tryptase-like activity per milliliter of BAL fluid.

Measurement of Biliary N-acetyl-LTE₄

N-acetyl-LTE₄ concentrations were measured in the bile of four animals randomly selected from the OA-challenged groups with and without TA-2 treatment. N-acetyl-LTE₄ levels in the samples were measured for the following time periods: 1 h before challenge (baseline), zero to 1 h, 1 to 2 h, 2 to 4 h, and 4 to 8 h. N-acetyl-LTE₄ was measured using precolumn extraction/reversed-phase high-pressure liquid chromatography (RP-HPLC)/radioimmunoassay as previously described (25). Briefly, methanol was added to aliquots (0.2 ml) of bile to give a final concentration of 80%. After standing 30 min, the samples were centrifuged to remove protein and the supernatant adjusted to a concentration of methanol of 30% and a pH of 3.0 by addition of water and 1 M phosphoric acid, respectively. The samples were filtered and loaded onto a precolumn packed with a Bondapack C18 column (Millipore-Waters, Bedford, MA) using a Milton Roy minipump. Using a six-port switching valve, the sample was then injected onto a NOApack C18 column (Millipore-Waters), which was eluted with 55% methanol in aqueous buffer (1 mM EDTA and 0.1% acetic acid, adjusted to a pH of 5.4 by the addition of ammonium hydroxide). The flow rate was 1 ml/min. Ultraviolet (UV) absorbance was monitored by a variable wavelength UV detector (Model 481; Millipore-Waters). N-acetyl-LTE₄ (retention time 15 min) was measured by radioimmunoassay in column fractions after evaporation of the solvent in a Speed-Vac vacuum centrifuge. A monoclonal antibody to LTC₄ (kindly provided by Dr. I. Rodger, Merck-Frosst, Pointe Claire, Quebec, Canada), which cross-reacted to the extent of 60% with N-acetyl-LTE₄ was used for the assay. [14,15-3H]LTC₄ was used as the radioactive ligand. All values were corrected for the degree of cross-reactivity of LTs with the antibody.

Data Analysis

EAR was calculated from the peak value of RL within 30 min after challenge as a percentage of the baseline RL. LAR was calculated as the area under the curve against time from 3 to 8 h after challenge.

Data are presented as mean ± SEM. Statistical comparison was performed using Student's t test for unpaired variables, and an analysis of variance (ANOVA) followed by a Tukey test for comparisons among several means. A difference was considered to be statistically significant when the p value was less than 0.05. Linear regression analysis was done using the technique of least squares.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Effects of Anti-VLA-4 mAb on the Allergic Airway Responses

The baseline RL was not significantly different among the three animal groups (IgG₁-injected/OA-challenged group: 0.182 ± 0.009 cm H₂O/ml/s; BSA-challenged group: 0.162 ± 0.011; TA-2-injected/OA-challenged group: 0.168 ± 0.008; p = NS by ANOVA). Sensitized rats undergoing OA challenge by either insufflation or aerosolization demonstrated a significant increase in RL within 30 min after OA challenge. The increases in the RL were greater after insufflation than aerosol challenge (342.9 ± 24.4% baseline RL in rats challenged with OA insufflation, n = 14; versus 132.8 ± 5.7% in rats challenged with aerosolized OA, n = 13; p < 0.01; Figures 1 and 2). The changes after OA administration contrast with RL in the BSA-challenged group which did not change significantly after BSA challenge (108.6 ± 3.2% in rats challenged with BSA insufflation, n = 8; versus 115.0 ± 5.0% in rats challenged with aerosolized BSA, n = 7; p = NS). The EAR following OA challenge was significantly attenuated in the TA-2-treated groups, in comparison to the groups injected with the irrelevant antibody, and this was true whether animals were challenged by insufflation or aerosolization (153.3 ± 19.4% in rats challenged with OA insufflation, n = 6; versus 110.8 ± 1.9% in rats challenged with aerosolized OA, n = 8; p < 0.01). The LAR were also significantly decreased in the TA-2-injected/OA-challenged group (LAR: 2.83 ± 0.55) compared with the IgG₁-injected/OA-challenged group (LAR: 9.89 ± 1.37; p < 0.01) and no LAR was observed in the BSA-challenged rat group. Two animals died in the TA-2-treated group before the completion of the 8-h period of observations and were excluded from analysis of the LAR.

View larger version (17K): [in this window] [in a new window]   Figure 1.   EAR after allergen challenge by insufflation. OA-sensitized rats underwent OA challenge by insufflation after control antibody treatment (open circles) or TA-2 antibody or with a comparable BSA challenge (open squares). EAR was calculated as the peak value of RL (expressed as a percent of baseline) within 30 min after challenge. Horizontal lines indicate mean values. *p < 0.01.

View larger version (17K): [in this window] [in a new window]   Figure 2.   Early and late airway responses after allergen challenge by aerosol. EAR was calculated as in Figure 1. LAR was calculated as the area under the curve of RL versus time from 3 to 8 h after challenge corrected for the baseline value of RL. OA-challenged control animals (open circles) were treated with control antibody, and the treatment group received TA-2 antibody (closed circles) prior to challenge. OA-sensitized animals were sham challenged with BSA (open squares). *p < 0.01.

Effect of Anti-VLA-4 mAb on Cell Recruitment into the Airways

There were significant differences in the total cell number in BAL among the various groups (p < 0.05; ANOVA; Figure 3). The differences were attributable to the OA-challenged group which had 3.15 ± 0.46 × 10⁶ cells compared with 1.20 ± 0.08 × 10⁶ cells in the BSA-challenged group and 1.21 ± 0.15 × 10⁶ cells in the TA-2-treated group. Macrophages showed the greatest change in numbers after OA challenge and the numbers were significantly reduced by TA-2 treatment (p < 0.01). Neutrophils showed similar trends but the changes were not significant. The eosinophils were few in number in the OA-challenged group and were absent in both the BSA-challenged and the TA-2-treated groups. These changes were not significant.

View larger version (31K): [in this window] [in a new window]   Figure 3.   BAL fluid leukocyte numbers. There were significant differences among total cell counts (p < 0.05; ANOVA) which were attributable to the OA-challenged group (n = 13), which had 3.15 ± 0.46 (million) cells compared with the BSA-challenged group (1.20 ± 0.08; n = 7), and the TA-2-treated group (1.21 ± 0.15, n = 7). The macrophages increased signficantly after OA challenge and were reduced by TA-2. (ND: not detected). *p < 0.01.

Effect of Anti-VLA-4 mAb on the Release of Mast Cell-derived Mediators

The concentration of histamine in BAL increased by 240% in OA-challenged animals compared with BSA-challenged rats (p < 0.01; ANOVA; Figure 4). The injection of TA-2 antibodies inhibited by 85% the concentration of histamine in BAL. Linear regression analysis indicated a significant correlation between the magnitude of EAR and the log-transformed concentration of histamine in BAL (r = 0.732, n = 29; Figure 5). There was a significant difference in the concentration of tryptase in BAL from the OA-challenged group and the BSA-challenged group (OA-challenged: 0.143 ± 0.035 µg/ml; BSA-challenged: 0.043 ± 0.009 µg/ml; p < 0.01; ANOVA; Figure 4). Treatment with TA-2 antibodies reduced OA-induced tryptase-like activity in BAL (0.053 ± 0.022 µg/ml; p < 0.01). There was no significant correlation between EAR and the level of tryptase-like activity in BAL (not shown).

View larger version (16K): [in this window] [in a new window]   Figure 4.   Histamine and tryptase concentration in BALF. The concentration of histamine in BAL fluid (left panel ) was measured by a colorimetric assay on fluid obtained 10 min after challenge. The concentration of tryptase-like activity was measured by colorimetric assay in BAL fluid (right panel ) obtained at 60 min after challenge. *p < 0.01.

View larger version (16K): [in this window] [in a new window]   Figure 5.   Relationship between EAR and concentration of histamine in BAL. Linear regression analysis disclosed a significant correlation between the magnitude of the EAR and the log-transformed value of histamine in BAL fluid (r = 0.732; n = 29; p < 0.01).

Effect of Anti-VLA-4 mAb on the Level of Biliary N-acetyl-LTE₄

There was no significant difference in the concentration of N-acetyl-LTE₄ in bile collected for 1 h before challenge (baseline) between groups (IgG₁-injected/OA-challenged: 5.63 ± 0.43 pmol/h [n = 4]; TA-2-injected/OA-challenged: 4.44 ± 1.18: n = 4; p > 0.05; Figure 6). OA challenge caused a sharp increase in the concentration of biliary N-acetyl-LTE₄ in bile during the first 2 h after challenge (0-1 h; 16.20 ± 3.90 pmol/h, 1 to 2 h; 15.03 ± 2.45 pmol/h; p < 0.05 compared with control). This increase was significantly prevented by pretreatment of animals with TA-2 (0-1 h; 9.31 ± 2.44, 1 to 2 h; 9.06 ± 1.13). The biliary levels of N-acetyl-LTE₄ obtained during 2 to 4 h after challenge were similar to baseline, and were not significantly different either between the two groups. Although bile was collected from the BSA-challenged rats, the concentration of N-acetyl-LTE₄ was not assayed in these samples.

View larger version (23K): [in this window] [in a new window]   Figure 6.   The amount of N-acetyl-LTE4 in bile. The concentration of N-acetyl-LTE4 was measured by HPLC and radioimmunoassay in bile collected during 1 h before challenge (baseline) and for 1-h periods up to 4 h after challenge in OA-challenged rats (open bars; n = 4) and OA-challenged and TA-2-pretreated rats (hatched bars; n = 4). Samples collected from 2 to 3 h and 3 to 4 h were pooled. The means + 1 SE are shown. *p < 0.05.

Preliminary analysis of biliary N-acetyl-LTE₄ in a sample of sensitized animals after aerosol challenge with OA did not show a significant elevation during the LAR (from 4 to 8 h after OA challenge). The bile from the remaining animals was therefore not analyzed.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Consistent with previous reports (15), our results show that the LAR is completely abrogated in the rat after treatment with a specific anti-VLA-4 mAb. Our results also indicate that this treatment inhibits the EAR following either insufflated or aerosolized OA in sensitized Brown Norway rats despite the substantially greater dose of OA administered by insufflation and the substantially greater airway response elicited by that method of challenge. Such an inhibitory effect on the EAR has been reported previously in the rat and in the rabbit (15, 16). Although an early report of responses of allergic sheep to challenge (17) failed to show inhibition of the EAR by anti-VLA-4 mAb, a more recent study in the same model (26) using a potent inhibitor of VLA-4 which prevents VLA-4-mediated binding to fibronectin (CS-1 ligand mimic), showed inhibition of the EAR.

Several studies have reported the potential role of certain adhesion molecules in cellular activation mediated by high- affinity IgE receptors on mast cells (27). In addition to the report that crosslinking receptor-bound IgE leads to an increase of cell adherence to fibronectin (27), such studies have suggested direct involvement of extracellular matrix proteins in cellular activation such as described for rat basophilic leukemia (RBL-2H3) cells (27, 29) and human basophils (30). The accumulation of mast cells in the airway subepithelial region and the increase in deposition of extracellular matrix proteins in asthmatic patients (31) are consistent with the hypothesis of a potential interaction of both IgE cross-bridging receptor and ₄-integrins in mast cell degranulation in vivo.

To study the possibility that mast cell activation is impaired by anti-VLA-4 mAb, we have measured both the levels of histamine and "tryptase" activity in BAL following OA challenge with pretreatment with TA-2. Our results indicate that allergen-stimulated increases in levels of histamine and tryptase-like activity in BAL are both inhibited by treatment with TA-2 which strongly suggests an interaction between IgE cross-bridging receptor and ₄-integrins during antigen-induced mast cell degranulation. The anti-VLA-4 mAb, TA-2, also decreased the concentrations of biliary N-acetyl-LTE₄ excreted following the acute allergic response and is consistent with interference with mast cell function. Although in the present study the mechanism by which VLA-4 modulates antigen-induced mast cell activation was not determined, previous reports have indicated that integrin-induced and high-affinity IgE receptor-induced mast cell degranulation share common intracellular pathways signaling through tyrosine kinase activation (i.e., pp125^FAK) (34, 35). Perhaps cross bridging of the VLA-4 molecule in vivo in some as yet undefined manner leads to a refractoriness of the mast cell to other stimuli.

The LAR was also inhibited by TA-2 pretreatment. Although it is possible that inhibition of the EAR which is often considered proemial to the LAR accounts for the reduction of the LAR, an independent effect on the LAR is also possible. Indeed evidence of inhibition of the LAR in the absence of effects on the EAR in earlier studies (17) is supportive of this idea. Several studies have suggested that eosinophils and lymphocytes are involved in the development of LAR (36, 37) and these cells express VLA-4 (11). We did not anticipate effects of anti-VLA-4 mAb treatment on eosinophil and lymphocyte recruitment into the airways of sensitized Brown Norway rats because we had previously found no change in allergen- induced airway eosinophilia after anti-VLA-4 treatment in rats (15). Similar findings have been reported for the sheep (17, 26). Interestingly an inhibitory effect on the degree of allergen-induced BAL eosinophilia has been observed in the guinea pig (18, 38) so that species dependence on various adhesion pathways may account for the discrepancies. The absence of an effect on cell numbers does not necessarily exclude an effect on activation so that the decrease in the magnitude of the LAR could result from the effect of anti-VLA-4 mAb on activation of resident airway eosinophils (17) and lymphocytes. Indeed, recent studies have shown interactions between VLA-4 and either VCAM-1 or extracellular matrix proteins in the cellular pathway of activation of both eosinophils and lymphocytes (11, 39- 42).

We have shown previously that allergen challenge in sensitized Brown Norway rats increases cys-LT synthesis during the LAR (43, 44). Pharmacological experiments have confirmed the dependence of the LAR in the rat on cys-LT synthesis in the airways; the LAR is completely inhibited by selective LTD₄ antagonists (44). We did not confirm an increase in cys-LTs in the current experiments, presumably because the LAR was relatively small compared with those evoked in previous studies of these compounds in LAR (43). However it seems likely that the inhibition of the LAR reflected a reduction of cys-LT synthesis by TA-2 during this phase of the airway response to allergen, also. Recently, Munoz and coworkers (45) have shown that ligation of human eosinophils with fibronectin increased PAF-induced LTC₄ synthesis and that effect was blocked by anti-VLA-4 mAb. In rodents the eosinophil does not produce LTC₄ so that it seems more likely that the cellular source of cys-LTs is the macrophage (46) but similar mechanisms are likely to apply.

In summary, we have shown that treatment with a specific anti-VLA-4 mAb prevents allergen-induced airway responses, the release of mast cell mediators, and cys-LT synthesis in sensitized Brown Norway rats which confirms a role for ₄- integrins in the modulation of airway allergic inflammation. Previous studies have failed to show any modification of airway responsiveness to inhaled aerosols of methacholine by TA-2 treatment (19) indicating that all of the observed inhibition of allergic airway responses is attributable to inhibition of mediator release rather than through a modification of the properties of airway smooth muscle, the primary effector of allergen-induced airway narrowing. Whether blocking VLA-4 molecules is related to inhibitory effects on the interactions between immune effector cells and the extracellular matrix proteins and/or due to inhibition of cell-cell interactions remains to be determined.