Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

An important feature of asthma is the development of airway inflammation (1, 2), which is associated with the presence of several cell types such as eosinophils, neutrophils, metachromatic cells, and lymphocytes (3). In addition, airway inflammation is related to the presence of airway hyperresponsiveness (AHR), another important characteristic of asthma (7, 8).

We have previously provided evidence that a potentially important aspect of allergic inflammatory responses is the induction of increases in inflammatory-cell progenitors, which may then contribute to disease through the ongoing, enhanced production of inflammatory effector cells (9, 10). Larger numbers of both circulating eosinophil/basophil colony-forming units (Eo/B-CFU) and CD34⁺ hemopoietic progenitor cells are demonstrable in the blood of atopic subjects than in that of normals (9, 11). In addition, the numbers of Eo/B-CFU in the circulation of asthmatic subjects at the time of an acute exacerbation are significantly greater than those measured after resolution of the exacerbation (12). Studies with atopic subjects have shown that there are fluctuating numbers of Eo/B-CFU during seasonal exposure to allergen (13, 14) and significantly larger numbers 24 h after allergen inhalation (15). Additionally, bone-marrow (BM) inflammatory progenitors are significantly increased after allergen challenge both in humans and dogs (16, 17). In humans, Eo/B-CFU are significantly increased in subjects with mild asthma after allergen inhalation (16), whereas in dogs with allergen-induced AHR and airway inflammation, granulocyte-macrophage colony forming units (GM-CFU) are increased (17), which is reflected by the predominant neutrophilia seen in bronchoalveolar lavage fluid (BALF) after allergen challenge (18). In addition, the increase in BM GM-CFU has been shown to be associated with the production of a serum hemopoietic factor generated after allergen challenge (18). Although these studies demonstrate increased production of inflammatory progenitor cells in association with allergen inhalation, little is known about the release and trafficking of these cells and their progeny in response to allergen challenge.

The development of a monoclonal antibody to the thymidine analogue 5'-bromo-2'-deoxyuridine (BrdU) (19) has been a major development in studies of cell kinetics. BrdU is incorporated into the nuclei of S-phase cells (20), and can be identified by immunohistochemical staining with anti-BrdU antibody (21, 22). In the present study, we used BrdU to label proliferating myeloid and other hemopoietic cells in vivo in a canine model of allergen-induced AHR, to evaluate the effect of allergen challenge on trafficking of these, and by implication inflammatory cells and their progenitors, from the BM through the circulation and into the lungs.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Study Design

Two groups of dogs (random-source mongrels of 21 to 36 kg) were paired for study, based on changes in airway responsiveness after a screening inhalation of Ascaris suum extract. After a 4-wk period, one group inhaled A. suum (n = 8) and the other inhaled A. suum diluent (n = 8). Prior to challenge, baseline measurements of airway responsiveness to acetylcholine (ACh) were made. All dogs received a bolus injection of BrdU before and at 5 h after challenge. Blood samples were taken before challenge and at 5 h and 24 h after challenge, and BM aspirate and BALF samples were taken 24 h after challenge. Development of AHR was assessed by changes in airway responsiveness to ACh at 24 h after challenge.

Procedures

Dogs were anesthetized with intravenous pentobarbitol sodium (30 mg/kg; Somnotol; MTC Pharmaceuticals, Mississauga, Ontario, Canada). Additional anesthetic was administered as required during the experiment. An endotracheal tube (10 mm inner diameter) was inserted and connected to a constant-volume ventilator (Model 551; Harvard Apparatus, South Natick, MA) set at a VT of 10 ml/kg and a rate of 30 breaths/min. An esophageal balloon catheter was inflated as previously described (23), and was placed in the esophagus at the point of most negative end-expiratory pressure. The esophageal catheter and a port at the proximal end of the endotracheal tube were connected to a differential pressure transducer (Model 267; Hewlett Packard, Inc., Palo Alto, CA).

Measurement of Total Pulmonary Resistance

Total pulmonary resistance (RL) was measured as described by Woolley and colleagues (17). Briefly, transpulmonary pressure was measured as the differential pressure between the endotracheal tube and the esophageal pressure (Pes). Flow was measured with a pneumotachometer (Fleisch No. 1; Instrumentation Associates, New York, NY), a differential pressure transducer (Hewlett Packard 270), and a pressure amplifier (Hewlett Packard 8805C). A continuous measurement of total RL was computed from the flow and transpulmonary pressure, using a respiratory analyzer (Hewlett Packard 8816A) that utilizes the method described by Mead and Whittenberger (24).

Measurement of Airway Responsiveness

Airway responsiveness was determined as described by Woolley and associates (17). Briefly, a dose-response relationship of RL against doubling concentrations of ACh (0.7 to 80.0 mg/ml; Sigma Chemicals, St. Louis, MO) was established. After baseline RL was measured, the dogs inhaled normal saline and then increasing concentrations of ACh at 5 min intervals until an increase was obtained of at least 5 cm H₂O/liter/s above the postsaline value. The response was expressed as the concentration of ACh causing an increase in RL of 5 cm H₂O/liter/s above the baseline measurement, which was termed the ACh provocative concentration. A decrease in this value represents an increase in airway responsiveness.

Allergen/Diluent Challenge

Allergen challenges involved inhalation of A. suum (stock extract, 10¹, wt/vol; Greer Laboratories, Lenoir, NC) as previously described (18). During the initial screening challenge, increasing concentrations of A. suum (10⁵, 10⁴, 10³, 10², and 10¹, wt/vol) were inhaled until RL increased by 10 cm H₂O · liter¹ · s¹ above preallergen levels. The concentration of A. suum producing this change in resistance was used for the allergen challenge during the study (for some dogs, resistance did not increase by 10 cm H₂O · liter¹ · s¹, in which case a concentration of 10¹ [wt/vol] was used). A. suum was delivered in 50 inhalations of 3-s duration each, using the same nebulizer as used for ACh challenges. A total of 10 min was allowed between doses during screening, and the postchallenge resistance was taken as the peak value in the 10 min after the inhalation. The diluent used in the A. suum preparations (0.4% phenol) was inhaled in the same concentration and manner as allergen.

Group-matching Protocol

To ensure that the study groups contained dogs with similar allergen-induced changes in airway responsiveness, dogs were matched for this attribute. Changes in airway responsiveness to ACh during a screening allergen challenge were expressed as shifts (prechallenge provocative concentration/postchallenge provocative concentration). Dogs with shifts that differed by less than 15% were paired, and each dog was randomly assigned to either the allergen or diluent group. However, dogs were not treated as pairs for statistical analysis.

Administration of BrdU

BrdU (Sigma Chemicals) was administered as equal intravenous bolus injections at 30 min before and 5 h after allergen challenge (total BrdU = 25 mg/kg). The total amount of BrdU per dog was calculated and dissolved in 60 ml endotoxin-free 0.9% sodium chloride solution (Baxter Corporation, Toronto, Canada). The solution was then filter-sterilized, and 30 ml was given per injection. The dose of BrdU was similar to that used by Bicknell and colleagues in a rabbit model of Streptococcus pneumoniae-induced inflammation (21).

Blood Samples

Venous blood samples were obtained from each dog before and at 5 h and 24 h after inhalation. The sample taken before challenge was also taken before the first injection of BrdU, and the 5-h sample was taken before the second BrdU injection. Samples were collected into heparin sodium-containing Vacutainer tubes (Hausser Scientific, Blue Bell, PA) for total and differential white blood cell (WBC) counts. WBC counts were made with a Neubauer hemocytometer, and differential cell counts were made from blood smears stained with the Diff-Quik method (American Scientific Products, McGaw Park, IL). Differential cell counts were made by one investigator in a blinded fashion, and the mean of two slides (300 cells counted per slide) was obtained. Cells were classified according to standard morphologic criteria. Results were expressed as absolute counts (× 10⁶ cells/ml). One milliliter of blood was used to make cytospin preparations for immunohistochemistry (IHC), as subsequently discussed.

BM Aspiration

BM aspirates were obtained from the iliac crest of anesthetized dogs with a 16-gauge Rosenthal needle. Three milliliters of BM were aspirated into a 10-ml syringe containing 1 ml of sterile heparin (1,000 U/ml; Leo Laboratories, Canada), and the syringe contents were then immediately resuspended in 50 ml of 1% bovine serum albumin (BSA) (Sigma Chemicals) in phosphate-buffered saline (PBS). Prior to cytospin preparation, the sample was spun for 10 min at 1,500 × g, and cytospins were then prepared for IHC from the cell pellet, as subsequently discussed.

Preparation of Cytospins from Blood and BM for IHC

One milliliter of each blood sample, and the BM pellet, were lysed for 30 s with 10 ml of cold 0.2% PBS to remove erythrocytes, and were then resuspended in 40 ml of BSA/ PBS to restore the normal concentration of PBS. The samples were allowed to stand for at least 30 min to allow the WBCs to recover from the lysing process, and were then resuspended in BSA/PBS. The cell concentration was adjusted to 2 × 10⁶/ml, and cytospins were prepared on APTEX-coated (Sigma Chemicals) slides.

Bronchoalveolar Lavage

Bronchoalveolar lavage (BAL) was performed as previously described (18). Briefly, a fiberoptic bronchoscope (BF-Be; Olympus, Tokyo, Japan) (optical density = 6 mm) was passed into a third-generation airway of the right middle lobe. Five 20-ml aliquots of PBS warmed to 37°C were injected into the airway via the bronchoscope, and immediately after injection of each aliquot, BALF was aspirated through the bronchoscope into collection traps. The BALF was then pooled and spun at 1,500 × g for 10 min, and the cell pellet was resuspended in BSA/PBS. A total cell count (TCC) was made, the cell count was adjusted to 2 × 10⁶/ml, and cytospin samples were prepared on APTEX-coated slides for IHC. Cytospin preparations were also made on glass slides, and differential counts were performed in a blinded fashion on Diff-Quik-stained slides. Mean counts from duplicate slides were obtained (500 cells counted per slide) and expressed as the number of cells per milliliter of BALF recovered (× 10⁶/ml BALF).

Immunohistochemical Staining for BrdU-labeled Cells

Immunohistochemical staining for BrdU-labeled cells was performed according to a method described by Bicknell and coworkers (21), with some modifications. All blood, BM, and BAL cytospin preparations were fixed for 10 min in 1% paraformaldehyde (BDH Inc., Toronto, Canada) in PBS, and then digested at 37°C for 5 min in 0.001% pepsin (Sigma Chemicals) solution acidified to pH 2.5. DNA in the cytospin samples was denatured at 37°C for 1 h in 2 N HCl, followed by neutralization in three washes of 0.1 M borate buffer, pH 8.5, each lasting 10 min. A final 10-min wash in 50 mM Tris(hydroxymethyl) aminomethane hydrochloride (Sigma Chemicals) and 150 mM NaCl, pH 7.6, containing 0.1% Tween 20 (TBS-Tween) (Sigma Chemicals), was used to restore neutrality. The alkaline phosphatase-anti-alkaline phosphatase (APAAP) technique (25) was used to detect BrdU-labeled DNA in cells. Briefly, cytospin preparations were incubated consecutively in 5% rabbit serum (GIBCO, Grand Island, NY) for 15 min, and then in 2 µg/ml mouse anti-BrdU antibody (DAKO Laboratories, Copenhagen, Denmark) prepared with 1% BSA in TBS-Tween at room temperature in a humidified chamber for 1 h. Nonimmune mouse IgG₁ (Sigma Chemicals) at 2 µg/ml was used as a negative control for each specimen. Incubation in a 1:20 dilution of rabbit antimouse IgG (DAKO Laboratories) for 30 min was followed by 30 min in a 1:50 dilution of a mouse monoclonal APAAP complex (DAKO Laboratories). Slides were washed three times (3 min each) in TBS-Tween following each antibody incubation. The alkaline phosphatase reaction was developed for 20 min, using the Fast Red Substrate System (DAKO Laboratories), and the resulting preparation was counterstained with Mayer's hematoxylin for 60 s and mounted in an aqueous medium. The nuclei of positive cells stained bright red, and duplicate slides from each sample were analyzed with light microscopy. An average number of cells per high power field (hpf) was calculated by counting 5 hpfs. The number of hpfs required to count 10,000 cells was then calculated, and the number of BrdU-positive cells in 10,000 cells was recorded and expressed as the percentage of BrdU-positive cells. For some slides, on which insufficient cells were present, the number of cells counted was always between 5,000 and 10,000.

Statistical Analysis

AHR. ACh provocative concentrations were log₁₀-transformed prior to analysis. Pre- versus postchallenge comparisons were made for each group of dogs, using paired Student's t tests. Values are expressed as geometric means (antilog of the mean logged data) and % SEM (antilog of the SEM of the logged data). The pre- versus postchallenge change in logged provocative concentrations was compared between groups with an independent Student's t test.

Blood. Differences in blood neutrophil counts and the percentage of BrdU-positive cells before and 5 h and 24 h after challenge were assessed for each group of dogs, using a mixed-model analysis of variance (ANOVA) (nonrepeated factor: allergen versus diluent; repeated factors: pre- versus 5 h post- versus 24 h postchallenge values) (26). Post hoc comparisons were made where indicated, using the Newman-Keuls test.

BM. The percentage of BrdU-positive cells in the BM in different groups was compared through an independent Student's t test.

BALF. Differences between groups in the TCC, differential count, and percentage of BrdU-positive cells in the BALF were compared through an independent Student's t test. Regression analysis was used to detect significant relationships between BALF neutrophils and the percentage of BrdU-positive cells in the BALF in the allergen-challenge group only (27). Statistical significance was assumed at P < 0.05.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

AHR

AHR to ACh developed in the allergen-challenged group but not in the diluent group at 24 h after challenge. The geometric mean provocative concentration values for ACh in the allergen-challenge group fell from 2.31 mg/ml (%SEM = 1.29) before to 0.94 mg/ml (%SEM = 1.26) after allergen challenge (P < 0.05), whereas in the diluent group the values were 3.68 mg/ml (%SEM = 1.38) before and 3.49 mg/ ml (%SEM = 1.36) after diluent (P = 0.81) (Figure 1). In addition, there was a significant difference between the two groups in the challenge-induced shift in logged ACh provocative concentration values (P < 0.05) (Figure 1).

View larger version (15K): [in this window] [in a new window]   Figure 1.   Acetylcholine airway responsiveness. ACh provocative concentration values before (circles) and after (squares) diluent (open symbols) and allergen (closed symbols) inhalation. Geometric means are indicated by solid bars. There was a significant reduction in ACh provocative concentration values following allergen (P < 0.05) but not diluent inhalation. In addition, there was a significant difference between the pre- and postinhalation values of ACh provocative concentrations between the two groups (P < 0.05).

Blood and BALF Neutrophils

Blood neutrophils increased at 24 h after allergen challenge to 11.19 ± 1.18 × 10⁶/ml from 6.18 ± 0.62 × 10⁶/ml before challenge (P < 0.0005), but did not increase at 5 h after allergen challenge, being 5.98 ± 1.12 × 10⁶/ml (P = 0.97) (Figure 2, Table 1). In addition, blood neutrophils at 24 h after allergen challenge were significantly increased as compared with their values at the same time point after diluent (P < 0.001) (Figure 2, Table 1). There were no significant differences in blood neutrophils at 5 h or 24 h after diluent inhalation (Figure 2, Table 1).

View larger version (12K): [in this window] [in a new window]   Figure 2.   Blood and BALF inflammatory cells following challenge. Changes in neutrophil numbers in blood (× 106/ml) and BALF (× 106/ml BALF) following allergen (solid bars) and diluent (open bars) inhalation. Significant increases in blood neutrophils from baseline were apparent at 24 h after allergen (*P < 0.0005) but not diluent inhalation. In addition, blood neutrophils at 24 h after allergen inhalation were significantly different from the neutrophil count at the same time point following diluent (**P < 0.001). There were significantly more BALF neutrophils at 24 h after allergen challenge than at the same time point after diluent (***P < 0.05).

BALF neutrophils increased at 24 h after allergen challenge as compared with their values at 24 h after diluent, being 0.63 ± 0.24 × 10⁶/ml BALF after allergen versus 0.03 ± 0.005 × 10⁶/ml BALF after diluent (P < 0.05) (Figure 2, Table 1). There were no significant differences between the two groups in numbers of other cell types (Table 1).

Trafficking of BrdU-positive Cells

BrdU-positive nuclear staining could be demonstrated with IHC in BM, blood, and BALF samples (Figures 3 and 4) after both allergen and diluent challenge. Positive-staining cells in the blood were of the band and metamyelocyte forms typical of granulocytes (Figures 3B and 3C). By contrast, cells staining positive for BrdU in the BM consisted of various nucleated cell types, including both mononuclear and polymorphonuclear cells (Figures 4A and 4B). The cellular morphology of BrdU-positive cells in the BALF following diluent was mainly mononuclear (Figure 4C), whereas after allergen inhalation, polymorphonuclear cells were also apparent (Figure 4D).

View larger version (79K): [in this window] [in a new window]   Figure 3.   BrdU-positive blood cells. BrdU-positive nuclear staining in the blood before challenge (A), 24 h after diluent (B), and 24 h after allergen challenge (C). Cells staining positive for BrdU are of the metamyelocyte type and band forms typical of granulocytes. Bar = 25 µm.

View larger version (105K): [in this window] [in a new window]   Figure 4.   BrdU-positive BM and BALF cells. BrdU-positive nuclear staining in BM and BALF. Cells staining positive for BrdU in the BM following diluent (A) and allergen challenge (B) consist of various nucleated cell types, including both mononuclear and polymorphonuclear cells. Cellular staining in the BALF following diluent is mainly mononuclear (C ), whereas that following allergen inhalation is mainly polymorphonuclear (D). Bar = 25 µm.

There was a statistically insignificant increase in the percentage of BrdU-positive cells in the postallergen BM as compared with the postdiluent BM, at: 27.0 ± 3.4% versus 18.9 ± 3.2%, respectively (P = 0.1) (Figure 5). BrdU-positive cells in the blood increased at 24 h after challenge in both groups, from 0 ± 0% to 5.7 ± 0.6% after allergen challenge (P < 0.0005) and from 0 ± 0 to 2.5 ± 0.7% after diluent (P < 0.0005) (Figure 5). There was a significant difference between the two groups in the percentage of BrdU-positive cells in the blood at 24 h after challenge (P < 0.0005) (Figure 5). There were no significant increases in the percentage of BrdU-positive cells in the blood at 5 h after challenge. Additionally, there was a greater percentage of BrdU-positive cells in the postallergen BALF sample than in the postdiluent BALF sample, at 3.25 ± 0.39% versus 0.84 ± 0.25%, respectively (P < 0.0005) (Figure 5). There was a significant correlation between the number of BALF neutrophils and the percentage of BrdU-positive BALF cells in the allergen group (r² = 0.54, P < 0.05) (Figure 6).

View larger version (14K): [in this window] [in a new window]   Figure 5.   Changes in BrdU-positive cells following challenge. Changes in the percentage of BrdU-positive cells in the BM, blood, and BALF after allergen (solid bars) and diluent (open bars) inhalation. There was an increase in the percentage of BrdU-positive cells in the BM following allergen as compared with diluent, but this difference was not significant. BrdU-positive cells in the blood increased 24 h after challenge in both groups, and there was a significant difference in the percentage of BrdU- positive cells in the blood between the two groups at 24 h after challenge (P < 0.0005). There was also an increase in the percentage of BrdU-positive cells in the BALF following allergen as compared with diluent (P < 0.0005).

View larger version (11K): [in this window] [in a new window]   Figure 6.   There was a significant correlation between the number of BALF neutrophils and the percentage of BrdU-positive BALF cells (r2 = 0.54, P < 0.05).

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

This study demonstrates that allergen-induced AHR and airway inflammation are associated with increased numbers of newly formed cells in the blood and airways, suggesting trafficking into the airways of inflammatory cells from the blood or BM. The results are consistent with those in our previous studies, which have shown an increased production of BM neutrophil progenitors (GM-CFU) after allergen inhalation in sensitized dogs (17, 18). Taken together, studies in this canine model suggest that allergen inhalation may initiate a cascade of events, with a stimulatory effect on the BM to produce and release newly formed inflammatory cells, followed by trafficking of these cells, through the circulation and into the airways.

The BM is the site of proliferation and terminal differentiation of neutrophilic granulocytes (28). In humans, neutrophil proliferation and differentiation consist of approximately five divisions, which take place during the first three stages of the neutrophil maturation process. There is one mitosis each at the myeloblast and promyelocyte stages, two successive mitoses in the myelocyte stage, and a minimum of 3 h from terminal myelocyte mitosis to the appearance of labeled cells in the metamyelocyte compartment (29). An additional mitosis may occur at the myelocyte level during times of granulocytic inflammation (30). After the myelocyte stage, the cells become "end-stage" cells (no longer capable of mitosis), a compartment that includes metamyelocytes, band forms, and mature neutrophils.

Studies with dogs have found a mean transit time of tritiated-thymidine-labeled cells from the myelocyte stage to their appearance in the circulation of 102 h in the steady state (31). The appearance of labeled granulocytes in our study began at 5 h and was significantly increased at 24 h after allergen inhalation. The kinetics appear much faster than in the steady state, but are in accordance with the findings in other studies in which neutrophilic inflammation was initiated. For example, Cronkite and colleagues (30) induced a sterile inflammation in the uterus of dogs, and showed that the myeloid-to-erythroid ratio of cells was 3-fold greater than in control dogs, peaking at 24 h. The time to appearance of tritiated thymidine-labeled cells in the blood was reduced from 40 h in the control dogs to 16 h in the dogs with inflammation. Also, infusion of starch into the peritoneal cavity of dogs caused a rapid rise in tritiated-thymidine-labeled cells in the blood within 10 h after infusion (32). In addition, the induction of pneumococcal pneumonia in dogs has been shown to increase the rate of production of neutrophils from their precursors, shortening their maturation time and decreasing their storage time in the marrow, with release of mature and immature neutrophils into the circulation (33). BrdU has also been used in rabbits, in which, after the introduction of Streptococcus pneumoniae into the lung, the transit times of labeled neutrophils through the proliferating and nonproliferating pools in the BM were shortened considerably (22).

In the current study, we demonstrated the appearance of labeled cells in the circulation 24 h after diluent challenge, which represents the steady-state turnover of cells for this model and is in contrast with the finding in studies using tritiated thymidine, in which the first appearance of labeled cells in the circulation was at 40 h (30). This suggests that BrdU labeling of dividing cells may be a more sensitive indicator of proliferating cells than tritiated thymidine labeling. Our findings are supported by studies with rabbits, in which BrdU-labeled cells began to appear in the circulation at between 24 h and 48 h in the steady state (21).

In asthmatic subjects, it is known that the influx of inflammatory cells into the airways begins at 3 to 4 h after allergen challenge. Since BrdU is cleared rapidly from the circulation, it was administered in two doses, to ensure availability of this thymidine analogue over the period in which inflammatory-cell production and trafficking may occur in response to allergen in the dog model. However, because of this study design, no inference can be made from the study data with regard to the true time course of the appearance in the circulation of labeled cells from the BM. This could only be addressed by administration of one dose of BrdU, at or shortly after the challenge procedure.

We have demonstrated a significantly higher number of BrdU-positive cells in the BALF at 24 h after allergen as compared with diluent inhalation. Morphologically, these cells appear to be of the granulocyte series, and there was a significant correlation between the number of BALF neutrophils and the percentage of BrdU-positive cells in the BALF. This response is in accordance with the study by Bicknell and colleagues, in which there were significantly larger numbers of labeled neutrophils in the inflammatory foci and alveolar spaces at 4 h after instillation of S. pneumoniae than in the control lung (21). Unfortunately, we did not examine the BALF at earlier time points to follow the time course of this response.

Previous results from our laboratory (18) have shown that increases in BALF neutrophils after allergen inhalation in dogs are associated with the production of a serum factor that has a hemopoietic effect on the marrow, increasing the number of myeloid progenitors. In the current study there was an increase in BrdU-labeled BM cells after allergen inhalation, but this was not significant. Since the marrow is a site of ongoing cell proliferation, it is probable that the cells staining positive for BrdU in this compartment include precursor cells for other hemopoietic lineages in addition to the myeloid series. The study did, however, demonstrate an allergen-induced increase in the number of proliferating cells appearing in the circulation and BALF. It is likely that BrdU-positive cells in the circulation and BALF are derived from progenitor cells in the BM. However, we cannot exclude the possibility that some of these cells were derived from progenitor cells that were already present and divided in those compartments, although this has never been previously shown to occur. Future studies, done with markers of hemopoietic progenitor cells and/or lineage-specific markers, are required to confirm the BM involvement in this process.

The present study does not provide information on the specificity of the trafficking of cells in response to allergen. It is feasible that allergen deposited in another site, such as the skin or nose, would induce similar trafficking of inflammatory cells from the BM in association with the inflammatory reaction at that site.

The study does not make clear the extent to which the inflammatory response contributes to the development of allergen-induced AHR in dogs. However, our previous studies, in which we have pretreated dogs with the inhaled corticosteroid budesonide before allergen inhalation, have shown that this corticosteroid attenuates allergen-induced increases in BM progenitors, as well as airway inflammation and AHR (17). However, it is not yet known whether the trafficking of these inflammatory cells contributes to the AHR. There was no significant correlation between the percentage of BrdU-positive cells in the lung and the development of AHR. An estimated 27 dogs would have to be studied in order to achieve significance for these parameters.

On the basis of our findings, we can hypothesize that there is both a more rapid production of neutrophils in the BM and a movement of newly divided neutrophils through the circulation into the lung in states of neutrophilic inflammation during the development of AHR. Similar events may occur in allergen-induced asthma, in which we have observed increased eosinophil progenitors in the BM concurrent with the appearance of eosinophilic inflammation and hyperresponsiveness in the airways (16). In addition, eosinophils have been shown to be increased in the BM, circulation, and BALF in a murine model of allergen-induced eosinophilic inflammation of the airways (34). The precise mechanisms responsible for these types of inflammatory-cell trafficking remain to be elucidated, but probably involve mediators released as a result of the immune response to allergen, or secondarily, in response to the ensuing inflammation in the airway. In either of these scenarios, it is possible that the trafficking of cells is important in initiating and/or supporting airway inflammation and the ongoing physiologic disturbance. Uncovering such possible mechanisms for inflammatory-cell trafficking may reveal new avenues for therapeutic intervention in diseases with chronic allergic inflammation.