Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Reversible airway obstruction, airway hyperreactivity to nonspecific stimuli, and pulmonary inflammation are distinguishing features of the asthmatic condition. Within minutes of an experimental exposure to antigen, alterations in lung function occur and quickly resolve (early phase response). This phase is often associated with mast cell degranulation. Four to 12 h later, lung function again declines and can remain impaired for hours (late phase response). Often the late phase is associated with airway hyperreactivity which can persist for days (1-3). During the late response, inflammation can be detected as the increased presence, differentiation, and activation of eosinophils, mast cells, and T cells (4). In addition, damage of the bronchial mucosa eventually develops because of repeated exposure of lung tissue to toxic granule proteins derived from inflammatory cells (5). Although there have been significant advances in our understanding of this disease, the interrelationships among cellular infiltration, cytokine production, and airway reactivity continues to be the focus of intense research.

Cytokines associated with a Th2-like response appear to play a role in disease progression, in particular interleukin (IL)-4 and IL-5 (6-8). IL-4 enhances B cell secretion of IgE, mast cell growth, and endothelial cell upregulation of adhesion molecules, especially VCAM-1 (vascular cell adhesion molecule-1) which is involved in selective recruitment of eosinophils (9-13). IL-5 acts on eosinophils by enhancing differentiation, adhesion to vascular endothelial cells, chemotaxis, cytotoxic activity, and degranulation (13, 14).

IL-4 and IL-5 proteins and the frequencies of cells expressing these mRNAs are increased in bronchoalveolar lavage (BAL) fluid and lung tissue samples from asthmatic patients but not from normal control subjects (13). Their levels correlate with increased eosinophil numbers, increased airway reactivity, and decreased FEV₁. In animal models, antibodies to IL-5 are effective in inhibiting eosinophil accumulation and airway hyperreactivity depending on the genetic background of the animal (15-22). In addition, IL-5-deficient mice do not develop eosinophilia or airway hyperreactivity in response to aerosol challenge (23). Antibodies to IL-4 can prevent the development of allergy if given during sensitization and thus inhibit the antigen-induced eosinophilic response (24, 25). However, strain-dependent differences in these responses to antibody-mediated inhibition have been noted (26). Taken together, these results suggest that cytokines participate in multiple pathways that can be distinguished by various animal models (27).

Thus, the development of experimental animal models is useful for elucidating the multiple mechanisms that contribute to asthma. To this end, a number of models utilizing monkeys, guinea pigs, and mice have been established (16-22, 28). We and others have used a paradigm in which allergy is induced in mice by intraperitoneal (i.p.) injection of antigen adsorbed to alum (19, 25, 28, 29). Within 12 days, serum IgE levels are increased significantly and mice subsequently are challenged with aerosolized antigen. Animals treated in this way exhibit many of the characteristics observed in human asthma, including eosinophil and activated T cell infiltration and elevated Th2-like cytokine mRNA levels in the lungs as well as increased serum IgE (19, 28-30).

One significant difference between this model and asthma is the absence of pulmonary inflammation prior to challenge. Therefore, we have modified our paradigm to include a second challenge at a time when there exists well characterized pulmonary inflammation. In this study, we examine the cellular infiltrate and pulmonary reactivity that occurs in rechallenged mice. In brief, the eosinophil and T cell responses as well as the pulmonary hyperreactivity that developed following a single challenge with antigen were augmented significantly by a second challenge. However, unlike results from clinical studies (31, 32), there was no correlation between the extent of hyperreactivity and the numbers of eosinophils in individual mice, although double-challenged mice had the greatest eosinophilia and exhibited the most airway reactivity. This modified model has many of the features observed in human asthma and should be useful in further defining the relationship between cellular infiltration, cytokine dysregulation, and airway hyperreactivity.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Sensitization and Challenge Protocol

The sensitization and initial challenge protocol have been described in detail (28). Briefly, male B6D2F1 (C57BL/6 X DBA/2) mice between the ages of 6 and 10 wk (Jackson Laboratory, Bar Harbor, ME) were injected i.p. with 7.5 µg ovalbumin adsorbed to 2 mg alum in 0.5 ml saline on days 0 and 5. One week later, mice were challenged with ovalbumin (0.5%) aerosolized with an ultrasonic nebulizer (Ultra-Neb 99; DeVilbiss, Somerset, PA) for 1 h and again 4 h later for 1 h. One, 2, or 3 wk after the first challenge the mice were challenged again (rechallenged) with the same protocol. In some studies, anti-mouse IL-5 monoclonal antibody (TRFK-5) or an isotype-matched control antibody (anti--galactosidase, GL113) were injected i.p. 2 h before rechallenge (33, 34). Both antibodies were produced at Schering-Plough (Union, NJ).

Bronchoalveolar Lavage (BAL)

On the indicated days after challenge or rechallenge, lungs from groups of five mice were perfused through the pulmonary artery via the right ventricle with saline to remove peripheral blood and were lavaged with 0.5 to 0.6 ml of saline containing 0.1% ethylenediaminetetraacetic acid. The total number of cells in each BAL sample was determined using standard hematologic procedures. Cytospins of BAL samples were prepared and stained with Leukostat stain (Fisher Scientific, Pittsburgh, PA) for differential cell counts. Eosinophils, neutrophils, and monocytes-macrophages were enumerated. Lymphocytes were quantitated by flow cytometric analysis as described below.

Histologic Evaluation of Lung Tissues

Tissues were prepared for histologic evaluation as described (28). Briefly, perfused and lavaged lungs were removed and fixed in 10% phosphate-buffered formalin. Lung tissues were embedded in paraffin, sectioned (5 µm), and stained with hematoxylin and eosin. Eosinophils surrounding the bronchi and bronchioles were enumerated at ×500 magnification. As previously described (15, 28), damage to respiratory epithelium and the presence of airway lumenal mucus was assessed using a semiquantitative numerical scale (0 = none, 1 = slight, 2 = moderate, 3 = severe). Five randomly chosen fields per tissue section were examined with at least 5 animals in each treatment group.

Phenotypic Analysis

Phenotypic analysis of BAL fluid cells has been described in detail (30). Briefly, BAL fluid cells from groups of five mice were pooled and washed in phosphate-buffered saline containing 2% fetal bovine serum. Cells were stained with unlabeled anti-FcII receptor monoclonal antibody (clone 2.4G2; PharMingen, San Diego, CA) in order to block Fc- mediated and nonspecific binding. They were then stained with the following monoclonal antibodies (PharMingen): anti-Thy1.2-fluorescein (clone 53-2.1), anti-B220-phycoerythrin (clone RA3-6B2), anti-CD4-phycoerythrin (clone RM4-5), anti-CD8a-fluorescein (clone 53-6.7), anti-CD44-biotin (PGP-1; clone IM7), and anti-CD45RB-biotin (clone 16A) followed by streptavidin-peridinin chlorophyll protein (PerCP). Stained cells fixed in 1% paraformaldehyde were analyzed in a FACSort flow cytometer (Becton-Dickinson Immunocytometry Systems, San Jose, CA). Lymphocytes were delineated on the basis of light scatter properties indicating relative size (forward light scatter) and granularity (side angle light scatter). The percentage of cells expressing a given surface phenotype and the total cell counts were used to calculate the absolute numbers of cells per volume of BAL fluid.

RNA Isolation and Semiquantitative Polymerase Chain Reaction (PCR)

Perfused lungs were immediately frozen on dry ice and stored at 70°C. Lung tissue was solubilized in guanidinium isothiocyanate and RNA was purified by cesium chloride ultracentrifugation (35). Semiquantitative PCR was used to determine IFN-, IL-4, and IL-5 steady-state mRNA levels as described (36, 37). Briefly, RNA (5 µg) was reverse transcribed with AMV-reverse transcriptase (RT; Boehringer Mannheim, Indianapolis, IN) in duplicate and serially diluted. Aliquots of each dilution were amplified by AmpliTaq DNA polymerase (Perkin Elmer, Norwalk, CT) with a GeneAmp PCR System 9600 Thermal Cycler (Perkin Elmer) under conditions determined in earlier experiments to be optimal for exponential amplification. Cycling profiles included an initial 30-s denaturization step at 94°C followed by two-step cycles of 94°C for 15 s and 60°C for 25 s, and completed with a final incubation at 72°C for 7 min. Oligonucleotide primers and internal probes (37) were synthesized on a Milligen/Biosearch 8750 DNA Synthesizer (Milligen/Biosearch, Burlington, MA).

Gene-specific amplification products were quantitated by hybridization to [³²P]-labeled internal oligonucleotide probes and detected with a Betascope 603 Blot Analyzer (Betagen, Waltham, MA). Background radiation was determined for each blot and subtracted from all samples. A standard curve for each analyzed gene was generated using a sample of RNA containing all cytokines examined and unknown sample dilutions quantitated relative to this curve. RNA used for the standard curve was from BALB/c splenocytes stimulated in vitro with Con A. Derived values were normalized by the constitutively expressed mRNA encoding hypoxanthine phosphoribosyltransferase.

Airway Responsiveness

Twenty-four hours after antigen challenge, mice were anesthetized with sodium pentobarbital (80 mg/kg, i.p.) and the trachea was intubated for mechanical ventilation. A jugular vein was cannulated for intravenous administration of drugs. Pancuronium bromide (65 mg/kg, i.v.), a paralytic agent, was administered to eliminate spontaneous respirations. Pulmonary insufflation pressure (PIP) was measured from a side port in the tracheal cannula using a pressure transducer (P23; Statham-Gould, San Juan, PR). Heart rate (HR) was monitored continuously from real time electrocardiography (ECG). Analog PIP and HR wave forms were displayed and recorded on an MI² Data Acquisition System (M-3000; Modular Instruments, Malvern, PA). Unprocessed data derived from the ECG measurement was sampled at 500 Hz, amplified (10 K), and processed with a low/high band pass filter at 1 KHz/100 Hz. Bronchoconstrictor responses to intravenous acetylcholine (0.1 and 10 mg/kg at 5-min intervals) were measured as the maximal increase in PIP due to acetylcholine in each animal. The peak increase in PIP was determined relative to baseline PIP before acetylcholine challenge. PIP returned to baseline levels between doses. The time-integrated change in peak airway pressure (area under the curve; airway pressure index) over a 2-min period following challenge with acetylcholine was determined and is expressed as cm H₂O · s.

Ex Vivo Cholinergic Sensitivity of Trachea

Mice were euthanized with CO₂ and the cervical trachea removed. Tracheal ring segments (3-4 mm) were mounted on supports suspended from transducers (FT-03; Grass Instruments, Quincy, MA) and anchored in a 15-ml water-jacketed tissue bath. Transducers were connected to a linearcorder physiograph (WR3310; Western Graphtec, Irvine, CA) for continuous hard copy recording of isometric tension. The linearcorder was wired in series to an MI² interface (M100; Modular Instruments) and a computer for data analysis (MI² Bioreport software; Modular Instruments). The baths (37°C) were filled with modified Krebs buffer (118 mM NaCl, 4.7 mM KCl, 1.2 mM MgSO₄, 1.2 mM KH₂PO₄, 24.9 mM NaHCO₃, 2.55 mM CaCl₂, 11.1 mM glucose; pH 7.4) and continuously aerated with 95% O₂-5% CO₂. Preparations were subjected to 1.5 g initial tension and challenged at 45 and 90 min with 10 µM carbachol for equilibration and conditioning. The baths were washed and tissues were allowed to return to baseline tension after each challenge. A rising cumulative acetylcholine dose response (100 nM-300 µM, at 2-min intervals) was performed. Upon completion of the dose response, tissues were treated with 1 mM carbachol to obtain a tension maximum for normalizing the acetylcholine responses.

Statistical Analysis

Data is expressed as mean ± standard error (SE). Statistically significant differences (P  0.05) between groups were determined by analysis of variance using Fisher's protected least significant difference test (StatView, Abacus Concepts Inc., Berkeley, CA).

Animal Care and Use

This study was conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and the Animal Welfare Act in a program approved by the American Association for the Accreditation of Laboratory Animal Care.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Effect of Rechallenge on Eosinophil Accumulation

The accumulation of eosinophils in BAL fluid and lung tissue following a single antigen challenge has been characterized extensively and the most commonly reported endpoint of studies in mice is pulmonary inflammation. Our goal was to develop a model in which mice that had generated significant pulmonary inflammation were rechallenged. Therefore, measurements of eosinophil accumulation in BAL fluid were used to establish the optimal time to rechallenge mice. Groups of allergic animals were challenged, rested for 1, 2, or 3 wk, and challenged a second time. Beginning 24 h after challenge, BAL fluid was collected and eosinophils were enumerated. Consistent with previous studies, eosinophils accumulated to maximal levels in BAL fluid 48 to 72 h after a single challenge, remained elevated for at least 1 wk, and declined to baseline levels after approximately 2 wk (28; Figure 1, line A). When mice were rested for 1 wk and then rechallenged, 3-fold more eosinophils appeared 24 h later in BAL fluid than were present at that time in mice challenged once (Figure 1, line B; P < 0.001). In contrast to the kinetics of accumulation after a single challenge, peak eosinophil levels after a second challenge after 1 wk were reached within 24 h and declined to low levels within 1 wk.

View larger version (17K): [in this window] [in a new window]   Figure 1.   Eosinophils in BAL fluid from allergic mice rechallenged 1, 2, or 3 wk after an initial challenge. Mice were challenged once (line A, squares) and then 1 wk (line B, circles), 2 wk (line C, diamonds), or 3 wk (line D, triangles) later were challenged again. BAL fluid was collected at the indicated times and eosinophils were enumerated. The first BAL fluid samples were collected from rechallenged mice 1 d after the second challenge. Asterisks show significant differences from single-challenged groups at that time (P < 0.05). Data indicate the mean ± SE (n = 5 to 40 mice per group).

The magnitude of the eosinophil response was dependent on the time between the first and second challenges. Mice challenged once and rested for 2 wk prior to rechallenge exhibited modest increases in eosinophil levels in BAL fluid (Figure 1, line C) which reached a maximum 48 h after the second challenge. The number of eosinophils in BAL fluid from mice challenged 3 wk after the initial challenge was less than in the previous two cases (B and C) and was maximal after 24 h (Figure 1, line D).

Although the absolute numbers of eosinophils and the kinetics of accumulation were different between single- and double-challenged mice, the relative cellular composition of BAL fluid was similar. Three days after the first challenge (28) or 1 day after the rechallenge (Table 1), when maximal eosinophil levels were attained, BAL fluid consisted of approximately 90% eosinophils, 10% monocytes/macrophages, and 1% neutrophils. Therefore, in order to satisfy the criteria of rechallenge in the presence of existing inflammation and to make observations under conditions that resulted in maximal eosinophil accumulation, the remaining experiments were conducted with mice that were rested for 1 wk prior to the second challenge.

Histologic Evaluation of Lung Tissue from Mice Challenged Twice

To further characterize the eosinophilic infiltrate in the lungs of rechallenged mice and to estimate respiratory epithelial damage and mucus accumulation, formalin-fixed lung tissues were examined histologically (Figure 2). The greatest number of eosinophils surrounding the bronchi were reached within 24 h and began declining within 48 h in both single- and double-challenged mice (Figure 2A). However, the second challenge increased the number of eosinophils already in the lungs by 3.5-fold within 24 h (P < 0.0001). Eosinophils surrounding bronchioles also increased following the second challenge, but the magnitude of the response was less than that seen for bronchi-associated eosinophils (Figure 2B).

View larger version (24K): [in this window] [in a new window]   Figure 2.   Eosinophils in lung tissue from allergic mice rechallenged one week after an initial challenge. Mice were challenged once (squares) and 1 wk later were challenged again (circles). Eosinophils within peribronchial (A) and peribronchiolar (B) regions of the lung were enumerated. The first tissue samples from rechallenged mice were collected 1 day after the second challenge and arrows indicate the time of rechallenge. Asterisks show significant differences from single-challenged groups at that time (P < 0.05). Data indicate the mean ± SE (n = 5 to 15 mice per group).

Allergic mice challenged once exhibit slight to moderate epithelial damage and mucus accumulation within the lungs (15, 28; Figure 3). Although mice challenged twice had epithelial damage and mucus accumulation surrounding and within bronchi, the magnitude was less than in single-challenged mice at the same time (P < 0.05). The same trend existed within bronchiole-associated tissue, although the difference between that seen in single- and double-challenged mice was not significant.

View larger version (23K): [in this window] [in a new window]   Figure 3.   Tissue damage and mucus formation in lung tissue from allergic mice rechallenged one week after an initial challenge. Allergic mice were treated as described in Figure 2 (single challenge, squares; double challenge, circles). Respiratory epithelial damage (A and C) and mucus formation (B and D) within peribronchial (A and B) and peribronchiolar (C and D) regions of the lung were assessed using a semiquantitative numerical scale (0 = none, 1 = slight, 2 = moderate, 3 = severe). Arrows indicate the time of second challenge. Asterisks denote significant differences from single-challenged groups one or two weeks after the first challenge (P < 0.05). Data indicate the mean ± SE (n = 5 to 15 mice per group).

Phenotypic Evaluation of T Cells in BAL Fluid from Mice Challenged Twice

T cells, especially CD4⁺ T cells expressing cell surface markers consistent with activated-memory cells (CD44⁺, CD45RB^lo), are detectable in the lungs of challenged allergic mice within 24 h of antigen administration (29, 30). Unlike eosinophils, Thy1⁺ T cells continued to accumulate and reached maximum levels within 1 wk (Figure 4A). Within 3 wk, T cell levels in BAL fluid declined to levels seen in BAL fluid obtained at 24 h. The same pattern was seen for the T cell subsets CD4⁺ and CD8⁺ (Figures 4B and 4C). In BAL fluid, from mice challenged a second time, the number of T cells, including both T cell subsets CD4⁺ and CD8⁺, rapidly increased and within 3 days returned to levels found at that time in mice challenged once.

View larger version (19K): [in this window] [in a new window]   Figure 4.   T cell accumulation in BAL fluid from allergic mice rechallenged 1 wk after an initial challenge. Allergic mice were challenged (squares) and 1 wk later challenged again (circles). Three-color immunofluorescence staining and total cell counts were used to determine the number of Thy1+, CD4+, and CD8+ T cells in BAL fluid. Arrows indicate the time of second challenge. The first BAL fluid samples from rechallenged mice were collected 1 day after the second challenge. BAL fluid from five mice per group was combined for analysis. Asterisks show significant differences from single-challenged groups at that time (P < 0.05). Data indicate the mean ± SE (n = 2 to 7 experiments).

Thy1⁺ T cells comprised a majority of the lymphocytes found in BAL fluid from double-challenged mice with few B cells or other non-T cells present (Figure 5A). Approximately 75% of the T cells were CD4⁺ and 25% were CD8⁺ (Figures 4B, 4C, and 5B). The CD4⁺ cells were CD44⁺ (PGP-1), CD45RB^lo while the CD8⁺ cells were CD44⁺ and CD45RB^lo or CD45RB^hi (Figures 5C-5F). The expression pattern CD44⁺ CD45RB^lo is associated with activated-memory cells (38, 39). This pattern was the same as that observed in mice challenged once (30, 37).

View larger version (39K): [in this window] [in a new window]   Figure 5.   CD44 and CD45RB expression by BAL fluid cells from allergic mice rechallenged 1 wk after an initial challenge. Three-color immunofluorescence staining was used to determine the phenotype of T cells in BAL fluid from mice 24 h after they were rechallenged. Lymphocytes gated on the basis of forward and side light scatter were analyzed for their expression of Thy1.2 (T cells; A), B220 (B cells; A), CD4 (B-D), and CD8 (B, E, F). CD4+ and CD8+ lymphocytes were analyzed further for their expression of CD44 (PGP-1; C and E) and CD45RB (CD45R; D and F). A representative experiment is shown.

Cytokine Production by Rechallenged Mice

Cytokines are thought to play a significant role in the development and progression of allergy and the recruitment of eosinophils into the lungs. As we have shown previously by semiquantitative RT-PCR analysis (30; Figure 6), lung tissue from allergic mice challenged once express increased levels of IL-4 and IL-5 steady-state mRNA within 6 h of challenge (10- and 20-fold of unchallenged levels, respectively) with relatively little change in IFN- steady-state mRNA levels (2-fold). This pattern is consistent with a Th2-like response. Six hours after the second challenge, IL-4 and IL-5 steady-state mRNA levels were 30- and 50-fold higher, respectively, than those levels measured in unchallenged allergic mice (Figure 6). In contrast, IFN- steady-state mRNA levels only increased 4-fold above the level measured in unchallenged allergic mice.

View larger version (30K): [in this window] [in a new window]   Figure 6.   Cytokine steady-state mRNA analysis of lung tissue from mice rechallenged 1 wk after an initial challenge. Allergic mice were challenged and rechallenged 1 wk later. Total RNA was prepared from lung tissue at the indicated times after one or two challenges. IL-5 (A), IL-4 (B), and IFN- (C) steady-state mRNAs were analyzed by RT-PCR. The level of steady-state mRNA for each cytokine compared with that detected in sensitized unchallenged (0 h) mice is expressed in relative units. Lung tissue from five mice per group were combined for analysis. Asterisks denote significant differences from double-challenged 6 h group (P < 0.05). Data indicate the mean ± SE (n = 2 to 6 experiments).

Pulmonary eosinophilia in allergic mice challenged once can be inhibited by treatment with anti-IL-5 monoclonal antibodies (15, 18, 19). In the model presented here, the change in steady-state mRNA levels following a second challenge was greatest for IL-5, thus suggesting that it might play a significant role in inflammation. In order to demonstrate the biological significance of IL-5, allergic mice challenged once were rested 1 wk, and treated with an anti-IL-5 monoclonal antibody (TRFK-5) 2 h prior to the beginning of the second challenge. One day later, BAL fluid was collected and eosinophils were enumerated. Anti-IL-5 dose-dependently inhibited eosinophilia by as much as 85% to levels at or below those seen at this time in mice given one challenge (Table 2).

Measurement of Airway Reactivity in Rechallenged Mice

Unlike other common laboratory animals such as guinea pigs which experience anaphylaxis during the acute response to aerosolized antigen (40), mice were refractory to early responses. Therefore, to assess pulmonary function in mice, measurements were made 24 h after challenge. Bronchoconstrictor responses to intravenous (i.v.) acetylcholine were quantitated as the change in pulmonary insufflation pressure and airway pressure index as described in MATERIALS AND METHODS. At both high (10 mg/kg) and low (0.1 mg/kg) doses of acetylcholine studied, mice challenged twice generally exhibited greater changes in insufflation pressure and airway pressure index than those observed in unchallenged or single-challenged mice (Figure 7). In addition, the baseline insufflation pressure (measured before administration of the first dose of methylcholine) for double-challenged mice (14.1 ± 1.2 cm H₂O) was significantly different (P < 0.0005) than that measured in unchallenged mice (10.6 ± 0.2 cm H₂O) or in mice 24 h or 1 wk after the first challenge (11.1 ± 0.2 and 10.7 ± 0.3 cm H₂O, respectively). Although rechallenged mice accumulated significantly more eosinophils and exhibited significantly greater bronchoconstrictor responses than single- challenged or unchallenged mice, a linear correlation did not exist between eosinophil levels and indices of lung function (data not shown). Finally, the in vivo increase in bronchoconstrictor responses to acetylcholine did not appear to be due to direct effects on smooth muscle contraction. Isolated trachea from single- and double-challenged mice were equally responsive to acetylcholine in tissue bath measurements of tracheal contraction (Figure 8).

View larger version (49K): [in this window] [in a new window]   Figure 7.   Airway responsiveness of allergic mice rechallenged 1 wk after an initial challenge. Mice were challenged and rechallenged 1 wk later. Airway responsiveness to high (10 mg/kg) and low (0.1 mg/kg) doses of acetylcholine were determined at the indicated times. The change in pulmonary insufflation pressure (A) and airway pressure index (B) is shown. Asterisks show significant differences from rechallenged groups for a given dose of acetylcholine (P < 0.05). Crosses show significant differences from unchallenged groups for a given dose of acetylcholine (P < 0.05). Data indicate the mean ± SE (n = 20 to 31 mice per group).

View larger version (19K): [in this window] [in a new window]   Figure 8.   Ex vivo cholinergic sensitivity of trachea from allergic mice rechallenged 1 wk after an initial challenge. Mice were challenged and rechallenged 1 wk later. Trachea were removed 24 h after the second challenge and assessed for their ex vivo responsiveness to increasing doses of acetylcholine at the indicated times. Data indicate the mean ± SE (n = 11 to 17 mice per group).

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

Asthma is a complex disease involving immunologic and physiologic processes that combine to affect patients acutely and chronically (3). In order to rigorously study this disease, a number of animal models have been developed which imitate much of the associated pathophysiology (16-22, 25, 28, 29, 41). Models in which animals are exposed to antigen multiple times for days or weeks attempt to approximate the chronic nature of asthma. However, little is known regarding the stepwise development of pulmonary responses during these complex sensitization-challenge regimens. Therefore, additional, well characterized animal models are needed to study the details of disease progression.

In this report, we have characterized responses to aerosol challenge given to mice with existing pulmonary inflammation, in an attempt to develop a model that reproduces the chronic aspect of atopic asthma. The pre-existing inflammation included elevated levels of both eosinophils and T cells. When the rechallenge was performed 1 wk after the initial challenge, there was significant exacerbation of eosinophil and T cell accumulation, as well as increased nonspecific airway reactivity. The T cells were predominantly CD4⁺, CD44⁺, CD45RB^lo, a cell surface phenotype consistent with activated-memory T helper cells and prominent in BAL fluid and lung tissue observed after a single challenge (30, 37). In addition, mRNA levels for Th2-like cytokines were preferentially elevated. IL-5 played a significant role because treatment with anti-IL-5 monoclonal antibodies inhibited eosinophil accumulation in BAL fluid from mice challenged a second time.

BAL fluid eosinophils reached maximum levels within 24 h of a rechallenge, in contrast to that in single-challenged mice in which maximum levels of eosinophils were not attained until 48 to 72 h. However, the kinetics of eosinophil accumulation in lung tissue were remarkably similar, in both cases reaching maximum levels 24 h after challenge or rechallenge. It is possible that earlier measurements (prior to 24 h) might have revealed faster eosinophil accumulation in rechallenged mice compared with single-challenged mice. These observations are consistent with the expected faster relative kinetics of a secondary immune response.

In mice allowed to rest for 2 or 3 wk before they were rechallenged, the accumulations of eosinophils in BAL fluid were markedly less than those seen in mice rechallenged 1 wk after the initial challenge. This was not due to reduced levels of total serum IgE (data not shown). It is interesting to note that 1 wk after the first challenge, the time at which the greatest rechallenge response was detected, both eosinophils and T cells were at maximum levels just prior to the second challenge. This was not the case 2 or 3 wk later when rechallenge-induced eosinophilia was minimal. Two weeks after the first challenge, only T cells were at maximum levels while eosinophils had declined to low levels. Three weeks after the first challenge, there were virtually no eosinophils and the number of T cells were declining in BAL fluid. It is possible that within 2 wk after the first exposure to aerosolized-antigen, the mice had engaged a mechanism to resolve the pulmonary inflammatory reaction. The signals generated by an aerosol challenge at this time might not have been strong enough or were not synchronized to overcome this process. Consistent with this idea is the observation that tissue damage and mucus accumulation in mice challenged 1 wk after the initial challenge had improved. Rechallenge at later times enabled this mechanism to dominate the response and only minimal inflammation developed. Pursuit of this concept might lead to insights into the mechanisms responsible for recovery from asthmatic attacks in humans.

There was a significant increase in airway reactivity to acetylcholine in vivo in mice challenged twice compared with that observed in mice challenged once. In addition, baseline airway pressure was slightly higher in rechallenged mice compared with all other groups tested. It does not appear that the in vivo hypersensitivity was due to increased contractile sensitivity to acetylcholine because the in vivo responsiveness of smooth muscle derived from double-challenged mice was similar to the other experimental groups and naive mice. It is unlikely that edema was responsible for the increased responsiveness because the amount of mucus histologically detected in the airways of rechallenged mice was the same or slightly less than that observed in single-challenged mice. Finally, although double-challenged mice exhibited pulmonary epithelial damage, the extent of damage was not greater than that seen in single-challenged mice, suggesting that damage might contribute to airway reactivity, but other factors can also influence this response. Thus, there appears to be a factor(s) present in the intact lung of allergic mice that results in altered in vivo airway reactivity when mice are challenged in the presence of ongoing inflammation.

There was no direct correlation between the level of airway reactivity and the number of eosinophils in BAL fluid found individually in double-challenged or single-challenged allergic mice. However, our results do indicate that bronchial hyperreactivity accompanies eosinophilia in the groups of allergic antigen-challenged mice. In contrast, such a correlation has been reported in humans (31, 32, 42). If eosinophils are directly responsible for reactivity, the demonstration of a correlation might require the ability to distinguish between different functional subtypes of eosinophils. In addition, increased airway reactivity generally correlated with an increased number of T cells. However, due to the relatively low number of T cells in BAL fluid from individual mice, determinations of T cell counts were made on the basis of pooled samples. Therefore, a direct correlation between T cell numbers and airway responsiveness in individual mice could not be determined.

The data presented in this report confirms and further characterizes in B6D2F1 mice the general observations that rechallenge augments both eosinophil and T cell levels in the lung and airway reactivity to acetylcholine (41, 43-46). However, the subtly different outcomes described in these reports indicate that the modes of sensitization and challenge and the genetic backgrounds of the mice have significant effects on the ensuing responses. These results are consistent with the concept that eosinophilia and hyperreactivity can be evoked by multiple mechanisms, each of which might be represented by different experimental models (27). Thus, the development and characterization of various animal models of asthma is essential for the study of human disease. This system provides significantly more pulmonary eosinophils, T cells, cytokines, and airway reactivity than the single-challenge model. Combined with the presence of established pulmonary inflammation, these responses are similar to those observed in asthma.