INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Nasal allergen challenge results in the local release of cysteinyl leukotrienes (LTC₄, LTD₄, and LTE₄) and in the release of leukotriene B₄ (LTB₄) (1, 2). The cysteinyl leukotrienes are reported to induce bronchoconstriction and to stimulate bronchial secretion (3, 4), whereas LTB₄ is known primarily for its effect as a chemoattractant for neutrophils (5) and, to a lesser extent, for eosinophils and monocytes (6). Purified neutrophil elastase is a potent secretagogue in airways of various species, including dogs and humans (7, 8), and allergen-induced secretion in sensitized canine airways is partially mediated by neutrophil elastase (9). Recently, neutrophil-dependent hypersecretion in guinea pigs and human airways has been shown to occur by a novel mechanism (10). Therefore, we developed a method for isolating and superfusing the canine nasal segment in vivo.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animal Preparation

Five dogs weighing 17.2 to 34 kg were selected randomly from an inbred colony of dogs, anesthetized intravenously with a 5-mg/kg bolus of Ketalar (ketamine hydrochloride; Parke-Davis, Warner-Lambert Co., Morris Plains, NJ) and 0.5 mg/kg Diazepam (diazepam; Elkins-Sinn Inc., NJ), intubated with an endotracheal tube, and ventilated by a constant-volume respirator (Model 607A; Harvard Apparatus, Dover, MA) set at a tidal volume of 10 ml/kg and at a frequency of 20 breaths/min. Anesthesia was maintained with halothane (1 to 1.5%; Halocarbon Laboratories, River Edge, NJ). Intravenous fluid (Lactated Ringer's Injection USP; Baxter, Deerfield, IL) was administered (50 ml/h) to prevent dehydration. The dogs were kept on a heating pad to maintain body temperature at 37° C. The dogs' vital signs were monitored continuously during the study.

Continuous Superfusion of the Nose

A modified Sheridan endotracheal tube equipped with a latex double balloon system was used to create an isolated segment of the nasal cavity. The balloons were positioned in the vestibulum nasi and epipharynx, respectively. The two balloons were inflated, and the segment was superfused through an open system consisting of two small-bore silastic tubes (11). Care was taken to ensure that the isolated nasal segment was patent and that no leaks had occurred. The superfusate was allowed to drip into a polypropylene syringe barrel, which was used as a reservoir, and the fluid was pumped back into the nasal segment via a peristaltic roller pump (Buchler, Fort Lee, NJ) at 14 ml/min (Figure 1).

View larger version (28K): [in this window] [in a new window]   Figure 1.   An isolated segment of nasal cavity is created with a modified endotracheal tube equipped with a double-balloon system. One balloon is positioned in the vestibulum nasi and the other in the epipharynx. The isolated segment of the nasal cavity, in between the two balloons, is superfused continuously via a double lumen catheter system. The buffer solution is pumped into the segment through one of the catheters (stippled ). The buffer leaves the segment and drips into a reservoir through the other catheter (hatched ) and is recirculated. An increase of the amount of fluid in the reservoir is regarded as a reflection of the total secretion (black bar). The dripping fluid as well as the fluid in the container is sampled.

Experimental Protocol

Experiments were conducted in five dogs, which served as their own controls. The isolated nasal segment was superfused with 20 ml of (1) sterile Hanks' balanced salt solution supplemented with 100 U/ml penicillin and 100 µg/ml streptomycin (HBSS) to prevent bacterial growth during the experiments, (2) LTB₄ supernatant diluted in HBSS (final concentration, 10⁷ M), or (3) LTB₄ diluted in HBSS with the addition of the neutrophil elastase inhibitor ICI 200,355 (final concentration, 10⁵ M). Exposure to the different superfusates occurred at 3-wk intervals. The amount of fluid in the reservoir was checked continuously, and an increase was regarded as a reflection of increased total secretion (Figure 1). Samples from the superfusate (1 ml) were removed from the superfusate as it returned from the isolated segment. A specimen was taken at the beginning of the experiment and at 1-h intervals for 6 h. These samples were replaced with 1 ml each of the superfusate used in the ongoing experiment, and this dilution was taken into account in calculating results. Ten microliters of the superfusate was stained with methylene blue to assess the superfusate for cell counts on a hemocytometer. A Giemsa stain was performed on the cytospin of cytocentrifuge-concentrated samples. At least 200 cells per sample were examined for differential counts. If the superfusate after 10 min (the "control" period) contained > 50 cells/µl, the dog was considered to have an upper respiratory infection and was excluded from the study.

Lysozyme Activity

Because low concentrations of neutrophil enzymes cause gland secretion (12), lysozyme activity was studied as a reflection of gland secretion (13) using a radial-diffusion plate method (detection limit, 0.25 µg/ml) (14). Briefly, 150 mg of Micrococcus lysodeicticus was suspended in 1% melted agarose in 0.07 M triphosphate buffer (pH 7) and poured into Petri dishes. Wells 3 mm in diameter were made using a punch, where 10 µl of perfusate samples or egg-white lysozyme standards (0.1 to 25 µg/ml) were added. After incubation at 37° C for 14 h, the zones of lysis of the samples were compared with the standards on a semilogarithmic scale. Incubation with HBSS plus antibiotics or LTB₄ showed no detectable lysozyme activity. The lysozyme activity reflected the concentration of lysozyme concentration in the superfusate, and the total amount of lysozyme secreted could be calculated using the total volume of the superfusate (the 20 ml of Hanks' buffer plus the amount of secretion eventually obtained).

Chemical Reagents and Drugs

Lactated Ringer's Injection USP was obtained from Baxter, Ketalar from Parke-Davis, Warner-Lambert Co., Diazepam from Elkins-Sinn Inc., and halothane from Halocarbon Laboratories. Sterile HBSS, TSBD, penicillin, and streptomycin were all obtained from the Tissue Culture Facility (UCSF, San Francisco, CA), and LTB₄ from Biomol Research Laboratories, Inc. (Plymouth Meeting, PA). ICI 200,355 (15) was generously provided by the Department of Medical Chemistry (ICI Americas, Wilmington, DE).

Statistical Analysis

All data are presented as means ± SEM and compared using Student's two-tailed t test. Values of p < 0.05 were considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Before the start of the experiment, the nose was always rinsed with 20 ml of HBSS buffer for 10 min. At the start of the experiment this solution was replaced with the test solution. All dogs tested showed a low neutrophil count (> 0.1 × 10⁶ neutrophils) in the rinse solution. All cell populations counted during the experiment consisted of more than 95% neutrophils. The remaining cells were epithelial cells, macrophages, and eosinophils.

During control conditions, an ongoing secretion was seen in all dogs (1.8 ml/h). In experiments where LTB₄ was added to the superfusate, the secretion was markedly increased (4.6 ml/h), whereas in experiments using a combination of LTB₄ and ICI 200,355 the secretory rate was equal to the rate seen during control conditions (1.3 ml/h) (Figure 2). The SEM values reported mainly reflect differences between animals, whereas differences recorded in repeated experiments using the same dog, during unaltered conditions, were relatively small (not shown). Low numbers of neutrophils could be detected in the nasal superfusate of the control dogs during the first 3 h. However, during the last 3 h there was a small, slow increase in neutrophils. In contrast, in experiments with LTB₄, both with or without addition of ICI 200,355, there was an exponential increase in neutrophils in the superfusate (Figure 3). ICI 200,355 alone had no significant effect on basal secretion and neutrophil recruitment during control conditions (n = 2, not shown).

View larger version (13K): [in this window] [in a new window]   Figure 2.   Total secretion in the isolated nasal segment continuously superfused with Hanks' balanced salt solution (HBSS) (open circles), LTB4 suspended in HBSS (107 M) (closed squares), and LTB4 + ICI 200,355 suspended in HBSS (107 M and 105 M, respectively) (closed triangles). Data are mean ± SEM (n = 5 for all groups). *p < 0.05 and **p < 0.01 for LTB4 compared with LTB4 + ICI 200,355.

View larger version (13K): [in this window] [in a new window]   Figure 3.   Total number of neutrophils counted in the isolated nasal segment continuously superfused with Hanks' balanced salt solution (HBSS) (open circles), LTB4 suspended in HBSS (107 M) (closed squares), and LTB4 + ICI 200,355 suspended in HBSS (107 M and 105 M, respectively) (closed triangles). Data are mean ± SEM (n = 5 for all groups). * p < 0.05 for LTB4 compared with control values.

The total amount of lysozyme found in the superfusate was also increased by LTB₄. However, in combination with ICI 200,355 the lysozyme secretion was decreased to the basal level (Figure 4, left panel ). There was a direct correlation between the amount of lysozyme in the superfusate and the resulting total secretion. When the lysozyme levels were corrected for the volume increase, there was no difference between animals exposed to only HBSS buffer, LTB₄, and LTB₄ + ICI 200,355 (Figure 4, right panel ). To compare the amount of lysozyme released from the neutrophils with the glandular lysozyme release, neutrophils were isolated from dog venous blood. The neutrophils were incubated with LTB₄ (10⁷ M), subsequently sonicated, and the total amount of lysozyme measured was 3 µg in 22 × 10⁶ neutrophils. In contrast, 70 µg of lysozyme was obtained in LTB₄ superfusate containing a similar number of neutrophils (14 × 10⁶).

View larger version (13K): [in this window] [in a new window]   View larger version (12K): [in this window] [in a new window]   Figure 4.   Total amount of lysozyme (left panel ) and lysozyme concentration (right panel ) in the isolated nasal segment continuously superfused with Hanks' balanced salt solution (HBSS) (open circles), LTB4 suspended in HBSS (107 M) (closed squares) and LTB4 + ICI 200,355 suspended in HBSS (107 M and 105 M, respectively) (closed triangles). Data are mean ± SEM. n = 5 for all groups. *p < 0.05 and ** p < 0.01 for LTB4 compared to LTB4 + ICI 200,355.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Recently, we described a novel method for continuous evaluation of cellular and biochemical events in the nasal cavity (11). This method has now been further developed to allow a direct estimate of total nasal secretion. Using the new method, we show that LTB₄ increases the total nasal fluid secretion, as well as the influx of neutrophils. ICI 200,355, a neutrophil elastase inhibitor, reduces the LTB₄-induced secretion to a level slightly below the basal level without affecting the influx of neutrophils.

Nasal secretion is a mixture of mucus from goblet cells, fluid secretion from seromucous glands, and watery secretion from anterior serous glands. In addition, tears, condensed water from expired air, and a transudate from blood plasma contribute to nasal fluid (16, 17). The basal fluid secretion in the present experiments probably represents the normal continuous sol production in the nasal chamber. However, a minor reflex-mediated contribution to this secretion cannot be excluded (18). Repeated control experiments showed only small differences in the total fluid secretion in each dog on different occasions, although relatively large differences were seen among them, demonstrating the reproducibility of the procedure. Neutrophils originate in the bone marrow, circulate in the blood for a few days, and can be viewed as "enzymatic storehouses" capable of destroying engulfed microorganisms and antigen-antibody complexes (19). In the present study, some neutrophils were always found in the nasal superfusate, even during control conditions, probably reflecting the fact that neutrophils are normally recruited in the nasal mucosa. From the lamina propria, they continue through the epithelium into the nasal cavity (20).

LTB₄ increased both the total fluid secretion and the influx of neutrophils in the superfused nasal segment. The ability of LTB₄ to recruit neutrophils is supported by previous findings in the lower airways (21). Neutrophils can also synthesize LT_A, which is converted to LT_B and may participate in the production of a positive feedback loop for neutrophil recruitment (22). This may explain the increased amount of neutrophils seen during the last 3 h of the present control experiments. The parallel increase of neutrophils and fluid secretion seen in the present experiments as a result of LTB₄ stimulation indicate a potential role neutrophils in nasal secretion. This assumption is also supported by previous findings in the lower airways (12). To further investigate the role of neutrophils, ICI 200,355, a well-characterized, selective elastase inhibitor was used (15). Human neutrophil elastase is one of several proteolytic enzymes normally constrained within granules of the neutrophils (23). Elastase has been reported to be a more potent secretagogue than other neutrophil proteases and secretory agonists (e.g., cathespin G, histamine) (12). In the present experiments, ICI 200,355 reduced LTB₄-stimulated fluid secretion to a level slightly lower than the level observed during control conditions, but neutrophil recruitment was unaffected. This result suggests a role for elastase in neutrophil-dependent airway secretion in the nose. Such a role for elastase has already been suggested for the lower airways (9). However, the exact molecular mechanism of elastase-induced secretion remains a mystery. In contrast to other receptor-coupled secretagogues such as histamine, the elastase-induced secretion does not involve identified second messengers (24). It has been suggested that elastase may substitute for or mimic the action of an endogenous metalloprotease that appears to be activated intracellularly during receptor-mediated secretion. Thus, elastase may activate secretion directly, bypassing the signal transduction mechanisms necessary for secretion caused by other agonists (12). Furthermore, recent studies have shown that chemoattractants in airways not only cause neutrophil recruitment but also neutrophil-dependent goblet cell degranulation involving adhesive interaction between neutrophils and goblet cells, with elastase actively moving from cytoplasmic granules to the neutrophil surface where it is free to interact with adherent goblet cells (10).

Changes in lysozyme secretion mimicked the changed in total fluid secretion induced by LTB₄. Thus, lysozyme levels increased during LTB₄ stimulation and returned to the basal level in the presence of ICI 200,355. However, the concentration of lysozyme was the same in all three groups, indicating a direct correlation between the secretory rate and the lysozyme levels. Lysozyme is synthesized in the nasal mucosa and used as a marker for serous gland secretion (13). Serous and seromucous glands may therefore be responsible for the major part of the total fluid secretion induced by LTB₄. Lysozyme can also be released from neutrophils (25), but in the present study the contribution of lysozyme released from neutrophils was less than 5% of the total lysozyme level obtained.

In conclusion, LTB₄ recruits neutrophils into the nasal lumen, and neutrophil elastase stimulates secretion from serous and seromucous glands. These results suggest a possible role for leukotriene-antagonists, elastase inhibitors, and inhibitors of neutrophil recruitment in the treatment of nasal hypersecretion.