INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Different mechanisms may be involved in fatal asthma, and the specific role of these mechanisms is still controversial. Mucus plugging has long been recognized as a major factor contributing to the mortality associated with acute severe asthma (1, 2). Nevertheless, the mechanisms of hypersecretion are not well understood, and no specific therapy for airway hypersecretion is available. Although eosinophilic infiltrates predominate in chronic asthma, acute severe asthma is often associated with neutrophilic infiltration of airways (3). Neutrophil elastase has been shown to be a potent secretagogue in airway submucosal glands (4) and in goblet cells (5, 6) in various species, including humans (7), thus providing a potential mechanism for hypersecretion by neutrophils. In allergic humans and in sensitized animals, delivery of antigen into the airways results in leukocyte recruitment. In the early phase, recruitment is mainly neutrophilic (8). We hypothesized that neutrophil recruitment results in elastase release, which in turn causes goblet cell degranulation. To examine this hypothesis, we delivered antigen to the trachea of ovalbumin-sensitized guinea pigs, and we measured goblet cell degranulation by assessing the volume density of Alcian blue/PAS-stained mucosubstances on the mucosal surface of the tracheal epithelium using a semiautomatic imaging system. Similarly, the number of leukocytes stained with 3,3'-diaminobenzidine in the epithelial layer were counted. Instillation of ovalbumin into the trachea of ovalbumin-sensitized guinea pigs caused both time- dependent goblet cell degranulation and neutrophil recruitment. In subsequent studies, we examined the effect of an inhibitor of leukocyte recruitment, a blocking antibody to ICAM-1, and a selective elastase inhibitor, on antigen-induced goblet cell degranulation. The results implicate elastase from recruited neutrophils in antigen-induced goblet cell degranulation in guinea pigs.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animals

Specific pathogen-free male Dunkin-Hartley guinea pigs weighing 300 to 550 mg (Simonsen Laboratories, Gilroy, CA) were used. The animals were housed in pathogen-free rooms and maintained on laboratory chow with free access to food and water.

Sensitization Procedure

All guinea pigs were sensitized intraperitoneally with an injection of ovalbumin, 2.5 mg (Grade V; Sigma Chemical, St. Louis, MO), complexed with 50 mg of alum in 0.5 ml of sodium chloride, 0.9% on Days 0 and 10.

Intratracheal Challenge with Ovalbumin

Intratracheal challenge with ovalbumin or saline was performed between Days 20 and 22 (10 to 12 d after the second intraperitoneal injection). To avoid fatal anaphylactic shock, all sensitized animals received an intraperitoneal injection of the H1 antagonist, pyrilamine maleate (10 mg/kg), 30 min before ovalbumin or saline challenge. Animals were anesthetized intraperitoneally with sodium pentobarbital (35 mg/kg) (Anthony Products Co., Arcadia, CA) and placed in a frame so that the pharynx, larynx, and trachea were aligned. The airway was then illuminated externally using a high intensity illuminator (FiberLite; Dolan Jenner Industries, Inc., Lawrence, MA), and the trachea was cannulated using a 22-gauge Angiocath catheter (Becton Dickinson, Sandy, UT). In "challenged" animals, a 2% ovalbumin solution in 150 µl of saline was then instilled; sensitized "control" animals received an intratracheal injection of 150 µl of saline only. The animals were then placed in a heated chamber, and the body temperature was maintained at 37° C.

Tracheal Preparation

The heart of the animal was exposed, a blunt-ended needle was inserted from the apex of the left ventricle into the ascending aorta, and the systemic circulation was perfused with 1% paraformaldehyde. An incision in the right ventricle provided an outlet for the fixative. The trachea was then removed and placed in 4% paraformaldehyde for 24 h. After fixation, the tracheal segments were cut longitudinally in the membranous portion, dehydrated, and embedded in JB-4 Plus (Polysciences, Inc., Warrington, PA). Sections 4 µm thick were placed on slides and stained with 3,3'-diaminobenzidine (Sigma Chemical) to visualize leukocytes that had migrated into the epithelium. Other sections were stained with Alcian blue/Periodic Acid Schiff (PAS) and counterstained with hematoxylin-eosin to visualize mucosubstances. Slides were observed at magnification ×400, using an Axioplan microscope (Zeiss Inc., Thornwood, NY) equipped with a Plan-NEOFLUAR ×40/0.75 objective lens.

In sections where neutrophils were counted (using 3,3'-diaminobenzidine stain) and goblet cell degranulation was studied (Alcian blue/PAS stain), five animals were evaluated for each condition. To confirm the presence of neutrophils in the airway tissue, sections were stained with an antielastase antibody and adjacent sections were stained with an anti-CD-16 antibody (n = 4 for each condition).

Counting of Neutrophils

Neutrophils were counted in 20 consecutive high power fields of the epithelial layer stained with 3,3'-diaminobenzidine and counterstained with hematoxylin-eosin (from the basement membrane to cell apices) at magnification ×400.

Quantification of Goblet Cell Degranulation

To assess goblet cell degranulation, we determined the volume density of Alcian blue/PAS-stained mucosubstances on the mucosal surface epithelium using a semiautomatic imaging system. We examined the stained slides with an Axioplan microscope (Zeiss, Inc.), which was connected to a video camera control unit (DXC7550MD; Sony Corp. of America, Park Ridge, NJ). Images of the tracheal epithelium were recorded from 10 consecutive high power fields with a phase contrast lens at ×400, using an IMAXX Video System (PDI, Redmond, WA). The intracellular mucin in superficial epithelial secretory cells appears as oval-shaped, purple granules of varying sizes. We measured Alcian blue/PAS-positive-stained area and total epithelial area, and we expressed the data as the percentage of Alcian blue/PAS area to total area. The analysis was performed on a Macintosh 9500/120 computer (Apple Computer, Inc., Cupertino, CA), using the public domain NIH Image program (developed at the National Institutes of Health and available from the Internet by anonymous FTP from or on floppy disk from the National Technical Information Service, Springfield, VA, part no. PB 95-500195 GEI). Two slides per animal were evaluated, and the quantification was performed by a single observer without knowledge of the condition of the animal. This morphometric technique to quantify airway goblet cell degranulation has been validated by Harkema and colleagues (12, 13) and used in previous studies (14).

Immunohistochemical Staining for Neutrophils

Airway tissues were fixed with 4% paraformaldehyde and washed in PBS. The tissues were cryoprotected with 30% sucrose, embedded in ornithine carbamyl transferase compound (Miles, Elkhart, IN), and frozen. Frozen sections 8 µm thick were cut and mounted on positively charged glass slides (Fisher Scientific, Pittsburgh, PA).

To confirm the role of neutrophils in goblet cell degranulation after antigen challenge, we used an indirect immunofluorescence method for staining neutrophil elastase; the primary antibody was a monoclonal mouse antihuman neutrophil elastase (DAKO Corp., Carpinteria, CA) diluted 1:100. Frozen sections 8 µm thick of tracheal epithelium were obtained in four sensitized animals killed 1 h after ovalbumin challenge and in four control animals (sensitized animals killed 1 h after saline challenge). To further define the role of neutrophils in goblet cell degranulation after antigen challenge, we used an indirect immunofluorescence method for staining with an anti-CD-16 monoclonal antibody. CD-16 is a surface marker that has been used to distinguish neutrophils (CD-16+) from eosinophils (CD-16) (15). The primary antibody was a monoclonal mouse antihuman CD-16 (BioSource International, Camarillo, CA) diluted 1:50. The staining was assessed by fluorescence microscopy using adjacent sections from the four experimental animals and from four control animals.

Protocol of Experiments

To determine the effect of stimulation with antigen (ovalbumin) on goblet cell degranulation and on neutrophil migration into the tracheal epithelium during the allergic response, ovalbumin-sensitized animals were exposed to 150 µl of 2% ovalbumin instilled into the trachea. Similarly sensitized animals were exposed to saline only (150 µl) instilled into the trachea (sensitized control animals). The animals were killed and their tracheas were removed 1 or 2 h after the injection.

To evaluate the effect of an inhibitor of neutrophil recruitment on goblet cell degranulation, animals were pretreated intravenously with NPC 15669 (10 mg/kg) (generously provided by Scios. Inc, Sunnyvale, CA) 1 h before the intratracheal instillation of 150 µl of 2% ovalbumin. NPC 15669 is a member of a class of anti-inflammatory molecules called "Leumedins" that inhibit neutrophil recruitment in animal models of inflammation (16).

To evaluate the role of the adhesion molecule intercellular adhesion molecule-1 (ICAM-1) in goblet cell degranulation, animals were pretreated with an anti-ICAM-1 antibody (monoclonal mouse antihuman ICAM-1, CD-54; Genzyme, Cambridge, MA), 150 µg, diluted in 0.2 ml of saline (17, 18).

To evaluate the contribution of elastase released from neutrophils on goblet cell degranulation, animals were pretreated intravenously with the potent and selective elastase inhibitor ICI 200,355 (700 mg/ kg) (19) (generously provided by Zeneca Pharmaceuticals Group, Wilmington, DE) 1 h before the intratracheal instillation of 150 µl of 2% ovalbumin. The animals were treated with either NPC 15669, ICAM-1 antibody or ICI 200,355, and they were killed 1 h after the intratracheal instillation of ovalbumin.

Statistics

All data are expressed as mean ± SEM. The Mann-Whitney test for unpaired data was used for comparison between groups. A probability of less than 0.05 was accepted as indicating a statistically significant difference.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Effects of Ovalbumin Challenge

The intratracheal instillation of ovalbumin in nonsensitized guinea pigs had no effect on goblet cell degranulation: the area of the tracheal epithelium occupied by goblet cells in nonsensitized guinea pigs challenged with 2% ovalbumin was 15 ± 3% (mean ± SEM); the area in nonsensitized guinea pigs challenged with saline was 15 ± 7% (n = 5, p = 0.94). Intratracheal instillation of ovalbumin to sensitized guinea pigs caused intense goblet cell degranulation, which was significant at 1 h and which persisted at 2 h (Figure 1, upper panel). Instillation of ovalbumin into the trachea of sensitized guinea pigs also caused a marked recruitment of neutrophils into the tracheal epithelium that was greatest at 1 h and continued at 2 h after instillation of ovalbumin (Figure 1, lower panel). When we examined all of the sensitized guinea pigs that had been challenged with ovalbumin (both at 1 and at 2 h), we observed a significant correlation between the number of neutrophils located in the tracheal epithelium and the intensity of the goblet cell degranulation (r = 0.76, p < 0.01, n = 10). Instillation of saline to ovalbumin-sensitized animals did not cause goblet cell degranulation or neutrophil recruitment (Figure 1).

View larger version (25K): [in this window] [in a new window]   Figure 1.   Effect of ovalbumin instillation (2%, 150 µl) in the trachea on goblet cell degranulation (expressed as percentage area of tracheal epithelium stained by Alcian blue/PAS) (upper panel ) and on neutrophil recruitment in the trachea (expressed as the number of neutrophils per millimeter of epithelium) (lower panel ) in ovalbumin-sensitized guinea pigs. Compared with "control" animals sensitized with ovalbumin and killed without stimulation (closed columns), ovalbumin challenge (open columns) caused time-dependent goblet cell degranulation (decreased stained area). The number of neutrophils present in the tracheal epithelium increased and was maximum at 1 h but remained elevated at 2 h. In another group of ovalbumin-sensitized animals, saline alone (150 µl) (hatched columns) was instilled; saline caused no goblet cell degranulation or neutrophil recruitment; n = 5 for each group. Results are expressed as mean ± SEM. *p < 0.05; **p < 0.01 compared with the saline-challenged group.

Effect of Pretreatment with the Inhibitor of Leukocyte Migration, NPC 15669

We evaluated the effect of neutrophil migration into the airway epithelium on goblet cell degranulation by pretreating animals intravenously with a drug that prevents leukocyte migration (NPC 15669, 10 mg/kg) (16) 1 h before the intratracheal instillation of 2% ovalbumin. NPC 15669 inhibited neutrophil recruitment (Figure 2, lower panel) and goblet cell degranulation (Figure 2, upper panel) caused by intratracheal challenge with ovalbumin. Both the Alcian blue/PAS-stained area of epithelium and the number of neutrophils in the NPC 15669-treated animals were not significantly different from those of the control animals sensitized with ovalbumin and given saline alone by instillation.

View larger version (22K): [in this window] [in a new window]   Figure 2.   Effect of drug pretreatment on the effect of ovalbumin (OVA) instillation (2%, 150 µl) in the trachea on goblet cell degranulation (expressed as percentage area of tracheal epithelium stained by Alcian blue/PAS) (upper panel ) and on neutrophil recruitment (expressed as the number of neutrophils per millimeter of epithelium) (lower panel ) at 1 h in ovalbumin-sensitized guinea pigs. Compared with control animals sensitized with ovalbumin and given saline instillation alone (150 µl) (diagonally hatched columns), ovalbumin instillation (open columns) caused goblet cell degranulation and neutrophil recruitment. Pretreatment intravenously with the selective elastase inhibitor ICI 200,355 (700 µg/ kg) (closed columns), inhibited degranulation but had no effect on neutrophil recruitment. Pretreatment intravenously with the inhibitor of neutrophil recruitment NPC 15669 (10 mg/kg) (horizontally hatched columns) or pretreatment intravenously with anti-ICAM-1 antibody (150 µg) (vertically hatched columns) prevented both degranulation and neutrophil recruitment; n = 5 for each group. Results reported as mean ± SEM; *p < 0.05; **p < 0.01 compared with ovalbumin-challenged groups.

Effect of Pretreatment with an anti-ICAM-1 Antibody

Pretreatment of animals intravenously with an anti-ICAM-1 antibody (150 µg) 1 h before the intratracheal instillation of 2% ovalbumin significantly inhibited neutrophil recruitment (Figure 2, lower panel) and goblet cell degranulation (Figure 2, upper panel) caused by intratracheal challenge with ovalbumin. Both the Alcian blue/PAS-stained area of epithelium and the number of neutrophils in the anti-ICAM-1 antibody-treated animals were not significantly different from control animals sensitized with ovalbumin and given saline alone by instillation.

Effect of Pretreatment with the Elastase Inhibitor ICI 200,355

We evaluated the effects of elastase secreted by neutrophils on goblet cell degranulation by pretreating the animals intravenously with the selective elastase inhibitor ICI 200,355 (700 mg/kg) 1 h before the intratracheal instillation of 2% ovalbumin. ICI 200,355 significantly inhibited goblet cell degranulation studied at 1 h after intratracheal challenge with ovalbumin (Figure 2, upper panel) but had no effect on the number of neutrophils recruited into the epithelium (Figure 2, lower panel). The Alcian blue/PAS-stained area of epithelium in the ICI 200,355-treated animals was not significantly different from control animals sensitized with ovalbumin and given saline alone by instillation.

Identification of Leukocytes Recruited after Ovalbumin Challenge

In all sensitized guinea pigs challenged with ovalbumin, we observed elastase-positive cells in submucosal blood vessels, within the tracheal epithelium, and also, occasionally, in the airway lumen. Adjacent cross sections immunolabeled with an anti-CD-16 antibody showed many cells that were immunofluorescent positive, indicating the presence of neutrophils (Figure 3).

View larger version (54K): [in this window] [in a new window]   Figure 3.   Immunofluorescence histochemistry of frozen sections of guinea pig tracheal mucosa. Sections were stained with neutrophil elastase antibody (left panels) and anti-CD-16 antibody (right panels). In the ovalbumin-challenged animals (left upper panels), numerous elastase-positive cells are located in the submucosal blood vessels, tracheal epithelium, and the airway lumen (arrows). In an adjacent section (right upper panel ), staining with an antibody to CD-16 demonstrates the presence of neutrophils in the tissue (arrows). Control animals (lower panels) show no immunolabeling for elastase or for CD-16. Bar = 50 µm.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In the present study, instillation of antigen (ovalbumin) in the trachea of sensitized guinea pigs caused profound goblet cell degranulation. It has previously been shown that allergic challenge in the airways of sensitized animals causes mucus secretion (8, 10, 11, 16, 20). The novel observation in this study relates to the mechanism of goblet cell degranulation: goblet cell degranulation was present within an hour after instillation of antigen and was associated with the recruitment of neutrophils into the airways. Both the recruitment of neutrophils in the tracheal epithelium and the goblet cell degranulation were prevented by pretreatment with NPC 15669, a molecule that inhibits leukocyte recruitment (16), implicating leukocytes in the goblet cell response. Degranulation was also prevented by pretreatment with a blocking antibody to ICAM-1, a member of the immunoglobulin superfamily of proteins expressed by endothelial cells that serve as a counterreceptor for the beta₂-integrins on leukocytes (25). ICAM-1 is involved in leukocyte adhesion and transmigration through the blood vessel wall (17, 18), and there is evidence that ICAM-1 is involved in the development of antigen-induced lung inflammation (26, 27).

Although Leumedin NPC 15669 and anti-ICAM-1 antibody inhibit the recruitment of both neutrophils and eosinophils and cannot differentiate between effects of the two cell types, neutrophils are implicated as the inflammatory cells involved in early antigen-induced goblet cell degranulation for several reasons. First, instillation of ovalbumin in sensitized animals caused a selective, early recruitment of neutrophils into the tracheal epithelium. The higher the number of neutrophils found in the epithelium, the greater was the intensity of the goblet cell degranulation. Furthermore, the leukocytes recruited into the trachea 1 h after instillation of ovalbumin stained positively for CD-16, a surface marker present on neutrophils but not on eosinophils (15). This confirms the finding that ovalbumin causes early selective recruitment of neutrophils at a time when goblet cell degranulation occurs. Second, various studies have shown that after introduction of antigen into the airway, recruitment of neutrophils occurs first (starting as early as 1 h after antigen challenge), and eosinophils only begin to appear in the airway epithelium several hours later (8); in those studies, eosinophil recruitment did not occur at the time that goblet cell degranulation was found in the present study (1 and 2 h). In fact, a previous study showing evidence of early mucus secretion after antigen challenge (maximal at 2 h) could not correlate this finding with the presence of eosinophils in the airway, which was maximal at 24 h (20). In the present study, antigen-induced goblet cell degranulation was inhibited by ICI 200,355, suggesting further that the degranulation was due to neutrophil elastase. Elastase is a serine protease present in the azurophilic granules in neutrophils. It is a potent secretagogue for airway glands (4) and goblet cells (5, 6) in various species, including humans (7). The presence of elastase-postive immunostaining in the animals challenged with ovalbumin but not in control animals further supports a role for elastase in goblet cell degranulation.

Although not specifically addressed in the present study, multiple mechanisms may be involved in neutrophil recruitment after antigen challenge. Lipid mediators liberated by activation of mast cells (e.g., LTB4, PAF) can promote neutrophil chemotaxis (21). Also, it has been shown that TNF- contributes to mast cell-dependent leukocyte infiltration (28). Finally, introduction of ragweed antigen into the trachea in ragweed-sensitive dogs induces interleukin-8 (IL-8) production in airway epithelial cells (23), and IL-8 is a potent neutrophil chemoattractant.

In the present study we have demonstrated the importance of the neutrophils and of neutrophil elastase in goblet cell degranulation after antigen challenge. Preincubation with elastase inhibitors or with inhibitors of neutrophil recruitment prevented antigen-induced goblet cell degranulation. However, we cannot rule out small effects of other mediators that could be involved in this process (e.g., lipoxygenases, mast cell chymase).

The present study specifically addressed the role of neutrophils in goblet cell degranulation after antigen challenge, and consequently we cannot exclude the possibility that other inflammatory cells such as eosinophils could also be involved in antigen-induced degranulation at later times. Eosinophils produce mediators that cause mucus secretion, and various studies have related eosinophils to the mucous secretory response that occurs during the late periods of the allergic response (11, 21). Our results confirm the data of Savoie and colleagues (20) regarding the presence of early goblet cell degranulation after ovalbumin challenge. For the first time, we link this early goblet cell degranulation with neutrophils. The coincidence of goblet cell degranulation with recruitment of neutrophils into the tracheal epithelium, the correlation between the number of neutrophils and the intensity of the goblet cell degranulation, the inhibition of the secretory effect by a specific elastase inhibitor, and the demonstration of the presence of both CD-16-positive cells and elastase staining in the airways by immunohistochemistry confirm the importance of neutrophils in early goblet cell degranulation after antigen challenge.

In summary, within an hour after ovalbumin was instilled into the trachea, degranulation of goblet cells occurred in guinea pig airways, an event associated with neutrophil recruitment in the airway epithelium. Goblet cell degranulation was inhibited when neutrophil recruitment was prevented or when neutrophil elastase activity was inhibited. These findings have important potential pathophysiologic implications in severe, acute asthma where goblet cell degranulation and mucus obstruction in peripheral airways are believed to play important roles: in asthmatics who die of acute severe attacks, postmortem studies of lungs show mucous plugs that are believed to be responsible for extensive airway obstruction (1, 2). In asthmatics who die from other causes, mucous obstruction was not found (29). The peripheral airway obstruction with mucus present in these patients is most likely due to goblet cell degranulation (30). Although eosinophilic infiltrates predominate in chronic asthma, acute severe asthma is often associated with neutrophilic infiltration of airways (3). It has been reported that in patients with acute severe asthma seen in the emergency ward, sputum samples contained a predominance of neutrophils over eosinophils in the majority of them (31). This neutrophil recruitment could be due to such stimuli as respiratory viral infection or to inhalation of allergens. Regardless of the stimulus, neutrophilic infiltration of airways could lead to rapid goblet cell degranulation, mucus obstruction, and death. The design of the present study does not allow us to compare goblet cell degranulation with other mechanisms influencing asthma attacks such as bronchospasm or other airway inflammatory effects. However, the present results could have important implications concerning mechanisms of airway obstruction in acute, severe asthma and could suggest strategies for therapeutic intervention.