Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Pulmonary eosinophilia is one of the distinguishing features of asthma (1, 2). Appropriate activation of these recruited leukocytes, by a number of agonists including antigen, will result in the release of their granule constituents, i.e., degranulation (3, 4). Eosinophil granules contain several proinflammatory cationic proteins, with major basic protein (MBP) being the most prevalent (5). It has been postulated that, in human asthma, the extracellular release of these granule proteins may contribute to the ongoing pathologic process (6). Specifically, MBP has been linked to bronchial hyperreactivity, to epithelial damage, and to promoting the formation of edema by effecting an increase in vascular permeability (9).

There are animal models of allergic pulmonary inflammation that display several of the pathologic features of human asthma (12). Notably, in our murine model, we have documented the presence of pulmonary eosinophilia, tissue damage, and mild edema in the lungs of antigen-challenged allergic mice (12). Similar to the human condition, we have provided evidence for the requirement of interleukin-5 (IL-5) as a factor for regulating the pulmonary eosinophilia (16). Nevertheless, the activation and the subsequent release of granule constituents from the recruited eosinophils has not yet been detailed in this murine model.

Therefore, we have examined bronchoalveolar lavage (BAL) fluids and tissue sections for the presence and localization of mouse MBP (mMBP). To perform these studies, we used a rabbit anti-mouse MBP (rb -mMBP), combined with immunoblot analysis and indirect immunofluorescence, on samples obtained from mice used in paradigms of allergic pulmonary inflammation.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Collection and Processing of Samples

Lung tissues and BAL samples were obtained from ovalbumin-challenged allergic mice as previously described (12). Briefly, male B6D2F₁ mice (Jackson Laboratories, Bar Harbor, ME) were sensitized by an intraperitoneal injection of 15 µg chicken egg albumin (OA) (Sigma Chemical Co., St. Louis, MO) adsorbed to 2 mg aluminum hydroxide gel (Rehydragel; Reheis Inc., Berkeley Heights, NJ) in 0.9% saline. An intraperitoneal booster injection at the same concentration was administered 5 d later. Nonsensitized animals received injections of only the aluminum hydroxide gel. Twelve days after the initial sensitization, the mice were challenged in a Plexiglas^TM chamber with aerosolized 0.5% OA produced by an ultrasonic nebulizer (DeVilbiss, Somerset, PA). Aerosol challenge was in two 1-h sessions that were 4 h apart. For some of the studies, after the two OA challenges in a single day, the animals were returned to their cages and rested for 1 wk. Animals were then rechallenged with aerosol OA, twice in a single day (20).

In both paradigms, 24 h after aerosol challenge, the mice were killed by CO₂ asphyxiation and samples were obtained. BAL fluid was collected with a 24-gauge ball-tipped needle inserted through a tracheal incision. First, the lungs were perfused, in situ, with 2.5 ml of 1.1% saline by inserting a 24-gauge ball-tipped needle into an incision in the right ventricle. Then, the lungs were lavaged with 250 to 300 µl of 1.1% saline/0.1% EDTA. Afterwards, the lungs were removed, fixed in 10% formalin (Fisher Scientific, Pittsburgh, PA), and processed according to standard paraffin-embedding procedures (21, 22). Sections (5 µm) from the left lobe of each lung were mounted on methanol-cleaned slides stained with hematoxylin and eosin, according to established procedures (21, 22).

Immunostaining for eosinophil granule murine major basic protein (mMBP), was adapted from Filley and colleagues (23). Briefly, in order to free up any formalin-fixed epitopes, the slides were incubated in 0.1% trypsin and 0.1% CaCl₂ · 2 H₂O (pH 7.8) with 0.1 N NaOH for 4 h at 37°C. Sections were rinsed in distilled water followed by a rinse with 0.05% Tween 20 (Bio-Rad Laboratories, Richmond, CA) in phosphate-buffered saline (GIBCO, Grand Island, NY) (PBS/Tw20). Tissues or cytospin preparations were then encircled with a PAP^TM pen (Research Products International, Mt. Prospect, IL) to form a reservoir for reagents.

Slides from cytospin or lung sections were then overlaid with 10% normal goat serum (Accurate Chemical & Scientific Corp., Westbury, NY) and incubated overnight at 4°C in a humid chamber. Sera and antibodies were diluted in 0.1% bovine serum albumin (Sigma) in PBS/Tw20. The optimal tissue reactivity of the rb -mMBP was at a dilution of 1:200 (data not shown). Slides were incubated 30 min at 37°C in a humidity chamber. Following incubation, slides were washed in PBS/Tw20 for 15 min at room temperature. The bath was changed three times over the course of 15 min (every 5 min).

To prevent nonspecific staining of eosinophils with fluorescein, the slides were stained in 1% chromotrope 2R (Sigma) for 30 min at room temperature and washed. Samples were incubated with fluorescein-labeled goat anti-rabbit IgG (1:500) for 30 min at 37°C, and then subsequently washed. The slides were mounted in a 50% 0.01 M Tris, 50% glycerin solution, cover-slipped, and sealed. Slides stored in the dark at 4°C retained a fluorescent signal for at least 3 wk.

Lung tissue samples were also assayed with a silver- enhanced immunogold technique. Consecutive sections were cut and processed as described previously for fluorescein isothiocyanate (FITC) immunohistology, with the exceptions of omitting the chromotrope 2R step and substituting Bio-Rad's colloidal gold-conjugated goat anti-rabbit immunoglobulin for the FITC-conjugate anti-rabbit immunoglobulin. Silver enhancement of the bound colloidal gold conjugate was accomplished with reagents from Bio-Rad's silver enhancement kit, according to established procedures (24). Sections were counterstained with Mayer's hematoxylin (21, 22). Sensitivity for the detection of antigen using this system is approximately 10 pg.

Microscopy

The slides were evaluated on a Zeiss microscope equipped for epi-fluorescence. Enumeration, identification, and photography of eosinophil leukocytes was performed at ×250 magnification. Photos were taken using ASA 400 film with a microscope-mounted 35 mm camera. Sample auto-fluorescence was compared by reviewing samples treated with unconjugated goat anti-rabbit IgG.

Immunoblot Analysis

BAL cell suspensions were centrifuged (350 × g; 4°C) for 5 min to pellet cells. An aliquot of the supernatant (100- 200 µl) was removed and placed in a sterile screw-cap cryovial. The cell pellet was resuspended in the remaining fluid and pipetted into a separate sterile screw-cap cryovial. Samples were stored at 20°C until assayed by immunoblot analysis.

To perform dot blot analysis, nitrocellulose membranes (Bio-Rad Laboratories) were presoaked for 30 min, in Tris-buffered saline (TBS; Bio-Rad Laboratories) and then the membrane was mounted in the immunoblot apparatus (8 × 12 array; Bio-Rad Laboratories). Sample preparation consisted of sonicating thawed samples for 15 s at room temperature. All samples were then diluted 1:4 with TBS followed by serial 1:2 dilutions conducted in a microtiter plate. Samples were then applied, in duplicate, to the membrane by vacuum filtration.

After application of samples, the membrane was removed and consecutively blocked, 30 min each, with 1% normal goat serum in TBS and 1% nonfat dry milk in TBS. Membranes were then incubated for 1 h with the rb -mMBP (1:400, in TBS with 1% normal goat serum). Following this incubation, the membrane was washed with 0.05% Tw20 in TBS. Membranes were incubated overnight at 4°C with the gold-conjugated detecting antibody (goat anti-rabbit Ig, 1:25; Bio-Rad Laboratories) in TBS supplemented with 1% nonfat dry milk and 0.05% Tw20. The membrane was processed for gold enhancement with and according to Bio-Rad's gold enhancement kit. Positive reactions are defined by the presence of black precipitate.

Animal Care and Use

These studies were done in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and with 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

Dot Blot Analysis of BAL Supernatants and Lysates

Immunodot blot analysis, for the presence of mMBP, was performed on BAL supernatants and cell pellet lysates derived from nonsensitized and sensitized OA-challenged mice (Figure 1). Immunoreactive mMBP could not be detected in the BAL supernatant from either nonsensitized or sensitized OA-challenged animals. In addition, the lysates of BAL cells recovered from nonsensitized but challenged animals lacked eosinophil leukocytes and had no detectable mMBP. In contrast, the cell lysates derived from sensitized and OA-challenged mice, displaying a prominent eosinophilia, revealed immunopositive dot blot results (Figure 1).

View larger version (45K): [in this window] [in a new window]   Figure 1.   Representative immunodot-blot analysis of BAL samples for mMBP from nonsensitized and sensitized ovalbumin (OA)-challenged mice. Individual samples were derived from either non-sensitized/challenged animals (Group 1) or from sensitized/challenged animals (Group 2). Neither animal in Group 1 had recoverable eosinophils. In contrast, the animal from Group 2 had 2.43 × 105 recoverable eosinophils. Supernatants (supt.) or cell pellet (pellet) lysates were blotted for mMBP, in duplicate, according to the MATERIALS AND METHODS section. Controls consist of blotting 20 µl of rb -mMBP (1:400) to the membrane, as defined in the MATERIALS AND METHODS section. rb -mMBP was detected with a gold-conjugated goat anti-rabbit IgG. Membrane reactions were enhanced using Bio-Rad's silver enhancement kit.

Additional immunodot blot analyses were performed on antigen-challenged allergic mice that displayed varying numbers of BAL-recoverable eosinophils. The immunodot blot results suggested a trend for the presence of mMBP with the number of eosinophils (data not shown). Specifically, the more eosinophils present in the original BAL lysate, the greater the mMBP immunoreactivity.

Immunohistochemical Staining of Lung Tissue

Lung tissue sections from nonsensitized but OA-challenged mice showed neither a pulmonary infiltration of eosinophils or demonstrable immunofluorescent positive cells (Figure 2). In contrast, antigen challenge of allergic mice consistently resulted in a profound peribronchial eosinophil leukocyte infiltration (32.8 ± 3.6 eosinophils/high power field; Figure 3, upper panel) (17). Nevertheless, when looking for mMBP immunoreactivity using the rb -mMBP, all fluorescent activity was localized to intact eosinophils (Figure 3, lower panel). Moreover, when allergic mice were challenged with OA on 2 d separated by 1 wk, a more pronounced pulmonary eosinophilia could be observed, but still no detectable release of mMBP was noted (Figure 4).

View larger version (118K): [in this window] [in a new window]   Figure 2.   Representative peribronchiolar region of lung tissue derived from a nonsensitized, OA-challenged mouse. Tissue preparation and staining procedures are as defined in the MATERIALS AND METHODS section. (Upper panel) Hematoxylin and eosin-stained section (5 µm) showing the lack of any demonstrable pulmonary eosinophilic inflammation. (Lower panel ) Using the rb -mMBP antibody on a consecutive tissue section from the same mouse, there is a lack of any measurable immunofluorescent activity. (Original magnification: ×250.)

View larger version (110K): [in this window] [in a new window]   Figure 3.   Representative peribronchiolar region of lung tissue derived from a sensitized, OA-challenged mouse. (Upper panel) Five-micron section stained with hematoxylin and eosin. Section displays a prominent pulmonary influx of peribronchiolar eosinophils. (Lower panel) Consecutive section stained by indirect immunofluorescence with rb -mMBP. Fluorescence activity is only associated within the eosinophil leukocytes, i.e., no extracellular fluorescence is seen. Tissue preparation and staining procedures are as defined in the MATERIALS AND METHODS section. (Original magnification: ×250.)

View larger version (126K): [in this window] [in a new window]   Figure 4.   Representative peribronchiolar region of lung tissue derived from a sensitized mouse challenged twice with OA. (Upper panel) Five-micron section stained with hematoxylin and eosin. Compared with that observed in the single challenged mouse, this section displays an increase in the pulmonary influx of peribronchiolar eosinophils. (Lower panel) In a consecutive section stained by indirect immunofluorescence with rb -mMBP, fluorescence activity was still only associated with peribronchiolar eosinophils. Despite the further recruitment of eosinophils, no detectable extracellular deposition of mMBP can be seen. Tissue preparation as described in Figure 3. (Original magnification: ×250.) Scale bar = 50 µm.

We further investigated MBP release using a silver- enhanced immunogold technique. This procedure affords sensitivity of antigen detection to approximately 10 pg of MBP. As seen with the immunofluorescent assays, no demonstrable extracellular release of MBP could be found in the lung tissues (Figure 5). In the single OA-challenged animals, we found all dark black precipitates of silver, which reacts exclusively to immunogold conjugate, particles strictly associated with intact eosinophils. These results are consistent and indicative of a lack of eosinophil degranulation. Despite that all of the sections derived from OA-challenged animals contained significant numbers of eosinophils, no evidence of extracellular MBP deposition could be discerned.

View larger version (134K): [in this window] [in a new window]   Figure 5.   Representative peribronchiolar region of lung tissue derived from a sensitized, OA-challenged mouse. (Upper panel) Five-micron section stained with hematoxylin and eosin. Section displays a prominent pulmonary influx of peribronchiolar eosinophils. (Lower panel ) Consecutive section stained by silver-enhanced indirect immunogold using rb -mMBP. Section was then counterstained with Mayer's hematoxylin. Tissue preparation and stain-ing procedures are as defined in the MATERIALS AND METHODS section. (Original magnification: ×200.)

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

By using an antibody to murine major basic protein (rb -mMBP), we were able to assess the state of activation of eosinophils that were recruited to the lungs in a murine model of allergic inflammation. Based on the pattern of mMBP distribution in BAL samples and in lung tissue sections, within the level of sensitivity of our assay systems, we conclude that the recruited eosinophils have not substantially degranulated. This is supported by our immunoblot analysis where mMBP was only detected in cell lysates and not in the supernatant fractions of BAL fluids, despite containing significant numbers of eosinophils. Furthermore, lung sections from antigen-rechallenged allergic mice, with existing pulmonary eosinophilia, consistently showed immunofluorescent mMBP reactivity. However, all fluorescence was limited to the intact eosinophil leukocyte.

We chose our current methodology because alternative immunologic probes for murine peroxidase or other eosin-ophil granule constituents are currently unavailable. Furthermore, alternative assays employing immunoperoxidase techniques require the full inactivation of substantial levels of endogenous eosinophil peroxidase activity to ensure no artifacts. Treatments used to successfully block this level of activity might render the antigenic sites of the MBP protein unrecognizable by our antibody.

This rb -mMBP bound specifically to eosinophils. We confirmed the utility of this antibody in cytospin preparations of BAL samples from aerosol-challenged allergic mice, yielding a high percentage of eosinophils. Only that portion equivalent to the number of eosinophils would immunostain positive for mMBP. To further document the specificity of this antibody, we evaluated its indirect immunostaining pattern on peritoneal cells enriched for neutrophils. While greater than 75% of the recoverable peritoneal cells were neutrophils (78 ± 4%; n = 4), the relative immunofluorescent activity corresponded to the percentage of recoverable eosinophils (11 ± 3%; n = 4; data not shown). Moreover, in some of the BAL and peritoneal preparations, both T lymphocytes and macrophages could be observed. Yet, we did not observe any immunopositive mononuclear cells. Finally, only lung sections from the allergic OA-challenged mice displayed immunofluorescent activity or silver enhanced particle deposition, correlating to the influx of eosinophils.

These studies were conducted 24 h after antigen challenge because the pathologic conditions that occur at this time-point have been characterized previously (12, 17). In addition to the tissue eosinophilia, modest edema, mucus hypersecretion, epithelial damage, and pulmonary hyperreactivity were observed (17, 20). Nevertheless, based on our results, mMBP was not released in substantial amounts and probably was not contributing to the observed pathology. However, there are limits to the sensitivity of our immunologic assays and, under the best of conditions we would expect the detection of 10 pg of antigen (mMBP).

We do not believe that we are missing demonstrable extracellular release of MBP in these studies. Experimental observations generated using similar murine models to ours support the notion that the recruited pulmonary eosinophils have not degranulated. In these two studies, electron microscopic techniques were used, and it was observed that there was a lack of eosinophil degranulation (25, 26). Furthermore, another well-established animal model, the guinea pig, has presented conflicting data regarding eosinophil degranulation measured by eosinophil peroxidase (27). Taken together, when comparing animal models to the human condition, caution is advised until the proper parameters of eosinophil activation and involvement are addressed.

Interestingly, after incubating those suspensions that contained significant numbers of eosinophils overnight, mMBP could be detected in the supernatant fluids (data not shown). This suggests that eosinophils incubated overnight may auto-activate degranulatory events. One plausible mechanism for this extracellular recovery may come from the experiments of in vitro cytokine exposure. It has been documented that IL-5 supports not only eosinophil survival but also stimulates degranulation after antigen challenge (28). It is feasible that the recovered BAL fluid eosinophils and mononuclear cells, from antigen challenged mice, are producing such a cytokine cocktail (29). In support of this, we have documented an increase in mRNA levels for IL-5 in this paradigm (20, 30). Alternatively, the recovered eosinophils could be undergoing autolysis. This could result in the release of their granule contents. Nevertheless, in either case, the extracellular release of the granule protein, mMBP, was detected by, and thereby validates, our immunodot blot assay.

We also performed studies involving a second antigen challenge, 1 wk after the first. In this paradigm, animals are rechallenged in the presence of established pulmonary eosinophilia (20). Compared with singly challenged mice, these rechallenged animals displayed a significantly larger increase in the number of pulmonary eosinophils, both in the BAL and in the peribronchiolar regions of the lung. However, all immunoreactive mMBP was still located intracellularly.

In this model of antigen rechallenge, there is evidence of an increase in the bronchoconstrictor responses to acetylcholine (20). However, the overall degree of pulmonary hyperreactivity in these mice was a modest 2 to 3-fold increase. We believe the lack of extracellular mMBP deposition provides a plausible explanation as to why there is only a modest degree of airway hyperreactivity. Indeed, it has been postulated that eosinophil activation, with subsequent degranulation and release of granule constituents, may contribute to the hyperreactivity in guinea pigs, monkeys, and humans (31). Furthermore, there were only infrequent changes in epithelial damage, edema formation, and mucus hypersecretion in the antigen-rechallenged animals (20).

These observations are in contrast to human pulmonary inflammation associated with asthma, where extracellular MBP has been documented (33, 35). It is interesting to speculate what factors, be they eosinophil derived or otherwise, might be contributing to the pulmonary pathology observed in these animals. Putative candidates include leukotrienes, mast cell granule-derived constituents, platelet-activating factor, and nitric oxide (36, 37). From previous experiments, we have found that nitric oxide and mast cells also contribute to the pulmonary allergic response in mice (38, 39).

In summary, our results show that, despite the presence of tissue eosinophilia, no immunohistologic evidence of substantial degranulation of the recruited eosinophils could be found in either BAL fluids or lung tissue sections obtained from aerosol antigen-challenged allergic mice. In addition, even after repeated antigen challenge, mMBP was not released. Our results are the first to comprehensively document that degranulation and the extracellular release of MBP has not occurred. Furthermore, we conclude that in this model of allergic pulmonary inflammation, mMBP does not contribute to the observed histopathogic changes and bronchial hyperreactivity.