Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Neutrophil-dominated airway inflammation is implicated in tissue damage and in airway gland hypersecretion associated with various airway diseases, such as chronic bronchitis and cystic fibrosis (1). More recently, it has been reported that the most characteristic feature of inflammation in the cystic fibrosis lung is a persistent infiltration into the airways of large numbers of neutrophils (4). Furthermore, inflammatory cell products, such as human neutrophil elastase (HNE), have been shown to induce further inflammation by initiating production of chemotactic cytokines, such as interleukin 8 (5, 6). In this way, HNE may perpetuate cycles of inflammation and hypersecretion in the lungs. Previous studies have shown that HNE leads to a marked increase of glycoprotein secretion in bovine airway gland cells (7) and in human tracheal explants (8), and causes degranulation in canine and human airway submucosal glands (9). Numerous other mediators of inflammation have been reported to stimulate mucus secretion in airway secretory cells, including cholinergic and adrenergic stimuli (10), platelet-activating factor, and histamine and adenosine-5'-triphosphate (ATP) (11). Despite evidence that HNE is the most potent secretagogue described so far for airway gland cells (7, 15), the molecular and ionic mechanisms responsible for regulating secretory activity of human airway gland cells still remain unclear.

Mechanisms of stimulus-secretion coupling have been extensively investigated in secretory cells with respect to the pivotal role of Ca²⁺ in the exocytotic release process. The regulation of exocytosis appears to depend on the secretory cell type and marked differences in the sensitivity of these cells to Ca²⁺, Ca²⁺-binding proteins, and protein phosphorylation have been reported (16, 17). Human airway tissue cells and more specifically airway submucosal gland cells, may respond differently to various agonists known to stimulate the release of mucus from the airways.

We have recently developed a new method combining Indo-1/AM loading and quantitative UV laser microspectrofluorometry, which permits the analysis of dynamic changes in the intracellular free Ca²⁺ concentration, [Ca²⁺]_i, in human tracheal submucosal serous gland (HTG) cells at the single-cell level in response to externally applied secretagogues (18, 19). Using a combination of this new method and measurements of the exocytotic degranulation rate by the videomicroscopic monitoring of quinacrine fluorescence, we have examined the relationship between the secretagogue-induced dynamic changes in [Ca²⁺]_i and the exocytotic release rates of individual secretory granule (SG) in HTG cells in response to HNE stimulation. To test the hypothesis that dynamic changes of [Ca²⁺]_i are a critical messenger in the regulation of the exocytotic response, we manipulated the [Ca²⁺]_i in HTG cells by either modifying the concentration of extracellular Ca²⁺ or by depleting internal Ca²⁺ stores using thapsigargin, an irreversible endoplasmic reticulum (ER) Ca²⁺-ATPase inhibitor (20). We have also examined changes in [Ca²⁺]_i and exocytosis in HTG cells in response to two other potent secretagogues, histamine and ATP, which have recently been reported to induce changes in Ca²⁺-signaling mechanisms and to increase secretion rates in different airway mucous secretory cells (12). We report here that the regulation of secretagogue-induced SG exocytosis in HTG cells appears to be linked to dynamic changes in [Ca²⁺]_i which is dependent on an influx of extracellular Ca²⁺. Furthermore, we show that, in comparison to histamine and ATP, very low concentrations (0.01 µM to 1 µM) of HNE that give rise to asynchronous oscillations in [Ca²⁺]_i and influx of extracellular Ca²⁺, promoting a fast exocytotic response in HTG cells. We propose that the dynamic changes in [Ca²⁺]_i in human airway submucosal serous gland cells are implicated in the fast initiation of SG exocytotic process and consequently, may play an important role in the fast exocytotic process of mucus release in human airway inflammation.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Chemicals

HNE isolated from human purulent sputum was purchased from Elastin Products (Pacific, MO). The molar concentration of active HNE was calculated using the molecular weight of 29 kD and an extinction coefficient (E 1%, 280 nm), and was determined spectrophometrically with the subtrate succinyl trianaline-p-nitroanilide as previously described (21). We found that the purified HNE used in this study was 96% active, based on an active-site titration curve. Histamine, ATP, thapsigargin and quinacrine were supplied by Sigma Chemical Co. (St. Louis, MO). Indo-1/AM were purchased from Molecular Probes, Inc. (Eugene, OR).

Isolation and Culture of Airway Submucosal Gland Cells

The isolation and culture of HTG cells were performed using human tracheal tissue collected during lung transplant operations, as previously described (19, 22). Briefly, cells were isolated by enzymatic digestion from tracheal submucosa and were grown onto type I collagen coated flasks in a Dulbecco's modified Eagle's medium/Ham's F₁₂-mixture (50/50%, v/v) supplemented with 100 U/ml penicillin G and 100 µg/ml streptomycin as antibiotics and with 1% Ultroser G (a serum substitute from Sepracor; Villeneuve-la-Garenne, France), 3 µM epinephrine, glucose (10 g/l) and sodium pyruvate (0.33 g/l). After 14 d in primary culture, cells were passed by treatment with 0.25% trypsin, 0.5 mM EDTA in Ca²⁺, Mg²⁺-free phosphate-buffered saline solution. Using monolayer cultures between passages 1 and 3, HTG cells proliferated and exhibited characteristics of epithelial and submucosal secretory cells, including two secretory protein markers specific to the glandular serous type cell: lysozyme and antileukoprotease (19, 22).

Antileukoprotease ELISA and Immunofluorescence Procedure

In order to assess the effects of HNE on the secretory activity of HTG cells, the cells were exposed to HNE (1.0 µM) containing RPMI 1640 medium for a 10 min-treatment period. Antileukoprotease (ALP), a serous gland cell secretory protein, was measured using a standard and well-characterized non-competitive ELISA as previously described (22, 23). ALP concentrations were determined on supernatant aliquots (20 µl) taken every 2 min during the 10-min incubation period. The effects of HNE were analyzed by determining the relative secretion rate (i.e., ALP released in the assay versus those released in a control experiment at the same time).

To assess cell viability, lactate dehydrogenase (LDH) released into the cell supernatant was measured immediately after incubation using a Sigma LDH kit. LDH release never exceeded 5% of the total LDH content of cells under these experimental conditions.

An anti-ALP IgG fraction (at a 1.0 µg/ml concentration) purified from immune rabbit serum was used for immunofluorescence staining of ALP in methanol-fixed HTG cells (22). In all immunofluorescence experiments, bound antibodies were detected using the biotin-streptavidin fluorescein isothiocyanate (FITC) system (Amersham International, Amersham, UK). Biotinylated sheep anti-mouse IgG, biotinyled anti-rabbit IgG and streptavidin-FITC were used at a 1:50 dilution in a PBS-1% BSA solution. After rinsing, all specimens were counterstained with hematoxylin, mounted in antifading glycerol (DABCO) and examined with a Zeiss Axiophot microscope (Le Pecq, France) using successive epifluorescence interference and Nomarski differential interference illumination.

[Ca²⁺]_i Measurements by UV-Microspectrofluorometry

HTG cells were plated on type 1 collagen coated glass coverslips and maintained in an Ultroser G-free, hormonally defined cell culture medium for three days prior to [Ca²⁺]_i measurements. Cells were loaded with 2 µM-Indo-1/AM in red-phenol-free RPMI 1640 medium (Sigma) for 30 min at 37°C, washed three times and bathed with 1 ml of the same medium. Dynamic changes in [Ca²⁺]_i in individual cells of the monolayer culture were measured by using a new procedure as recently described (18, 19). Fields of 5-8 adjacent cells were visualized through a light microscope (Olympus BH2; Olympus Optical Co., Tokyo, Japan) equipped with a water immersion objective lens (magnification ×100; N.A. 0.95) (State Optical Institute of St. Petersburg, Russia) which was specially developed for the total transmission of UV radiation down to 300 nm and corrected for axial chromatic aberrations. Fluorescent emission spectra within each selected cell were recorded using a UV confocal microspectrofluorometer (XY model; DILOR, Lille, France). The 351 nm laser line (Ar+, 2065A model; Spectra Physics, Mountain View, CA) was focused on the sample with a measured power of 0.5 µW. The fluorescent emission in the 380-580 nm range was spectrally dispersed by diffraction grating, and was detected with a two dimensional CCD detector. Dynamic changes in [Ca²⁺]_i were measured as follows: a computer-driven (X,Y) motorized stage (model MCL-2 with increments of 0.1 µm; Märzhäuser, Wetzlan, Germany), permitted the storage of the (X,Y) positions of several (up to 10) chosen points in different locations, either from one individual cell or from different cells. The standard duration of measurement for each point was fixed to 1 s, so that [Ca²⁺]_i could be measured at each point every 5 s, when 5 different adjacent cells were analyzed. Basal fluorescence emission was recorded for the first 5 measurements from each cell prior to the addition of secretagogue. Cells were then monitored for about 10 min after secretagogue addition at room temperature, and successive measurements were stored in the computer.

Measurements of Secretory Granule Exocytosis by Fluorescence Videomicroscopy

Cultured HTG cells on glass coverslips were loaded overnight with quinacrine by adding 0.1 µM of the dye to the culture medium. After loading, cells were washed two times and bathed with 1 ml of red-phenol-free RPMI 1640 medium. Earlier experiments demonstrated that the fluorescent quinacrine may be used to monitor the exocytosis of SG in various secretory cell types such as mast cells and pancreatic acinar cells (24, 25). Quinacrine fluorescence emission is restricted to individual SG, where it accumulates due to the slightly acidic nature of these subcellular organelles (24). Consequently, stimulation of SG release can be monitored as an accelerated loss of quinacrine fluorescence. To detect changes in quinacrine fluorescence emission, the measurements were made using an Axiovert IM35 inverted microscope equipped with a plan-Neofluorar 40×/1.30 A objective (Carl Zeiss, Oberkochen, Germany) and a heating stage (37°C) on which culture dishes were placed. Excitation was carried out at 340 nm and emitted light was collected at 510 nm. The fluorescence microscope was coupled with a silicon intensified target camera (SIT 4036; Lhesa, Cergy-Pontoise, France) which was connected to a Sun Sparc-Classic workstation equipped with a digitizing card (X-video card; Parallax Graphics, Santa Clara, CA). Fluorescence pictures were displayed in a split configuration as 512 × 512 pixel images on the screen of a Sun 486/33 MHZ computer. Videocamera control parameters (i.e., gain, offset, and sensitivity) were adjusted by using the image of a control HTG cell on the monitor. Control parameters were set up to obtain a clear image of the cell on the monitor and a fluorescence intensity of 250 (arbitrary units) in a secretory granule region of the cell. Time-lapse sequences were recorded at scanning rates of 1 s per image every 2 min for a 30-min period on a video tape, and then played back, digitized, and processed off-line by computer. Regions of interest were defined as bright areas with as little background as possible. Changes in fluorescence intensity as the function of time, before, during, and after secretagogue addition were stored and plotted. Data analysis was performed with the help of software we have developed for the analysis of fluorescence image series (modeling, parametric imaging, multivariate statistical analysis, etc.).

Partial and total release of each fluorescent SG was defined as the magnitude of differences in fluorescence emission intensity before (Fb) and after (Fa) the secretagogue addition. More precisely, the rate of SG exocytosis was expressed as follows: % of SG exocytosis = 100 × (Fb-Fa)/Fb. Control fluorescence emission measurements from HTG cells bathed in secretagogue-free red-phenol RPMI 1640 medium were carried out before each series of experiments.

Statistical Analysis

For [Ca²⁺]_i measurements, the differences ([Ca²⁺]_i) between the mean basal level and the maximum peak increase in [Ca²⁺]_i after exposure to secretagogues were calculated. Inasmuch as a low level of spontaneous noise in the fluorescence signals could possibly have been interpreted as an alteration in [Ca²⁺]_i, a positive response to an applied secretagogue was defined as a [Ca²⁺]_i increase of at least 20 nM. All data describing [Ca²⁺]_i and quinacrine fluorescence intensities are expressed as means ± SD. Significance of difference was determined by an unpaired Student's test for comparison between two groups. Values of P < 0.01 were considered significant.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Visualization of the SG Exocytotic Process in HTG Cells

The exocytosis of fluorescent SG from single HTG cells was visualized in real-time during the continuous exposure of cells to various secretagogues over a 20- to 30-min period. Upon addition of 1.0 µM HNE (Figure 1), the pattern of SG distribution preferentially formed a ring beneath the plasma membrane, which then lost fluorescence as dye diffused out of the exocytosed granules. This videomicroscopic observation reinforces our previous ultrastructural experiments showing that most of the SG are located beneath the plasma membrane of gland secretory cells in human tracheal gland cell line MM39 after HNE treatment (26).

View larger version (144K): [in this window] [in a new window]   Figure 1.   HNE-induced SG movements and exocytosis. Videomicroscopic series of fluorescence changes (A-F ) in HNE-treated HTG cells preloaded with quinacrine. (A) Resting HTG cells prior to the addition of 1.0 µM HNE. (B-F  ) Images of the same field of cells at sequential time points, as indicated in the lower right (in min) following the addition of HNE at time zero. The loss of quinacrine fluorescence, reflecting SG exocytosis responses was analyzed as described in MATERIALS AND METHODS. Note that one pool of SG is capable of responding quickly with a fast release (a 50% fluorescence decrease is observed after an 8-min period, as demonstrated in the [FR] square in A and D) whereas the other SG pool requires a longer time with a slow release (a 50% fluorescence decrease is observed after a 16-min period, as demonstrated in the SR circle in A and F  ).

Secretagogue-Induced Changes in the Dynamics of SG Exocytosis

Stimulation of HTG cells with 1.0 µM HNE, 100 µM histamine or 100 mM ATP induced a time-dependent release of SG (Figure 2). Continuous exposure with HTG cells to HNE (1.0 µM) induced a fast and brief response. Compared with the 1.0 µM HNE-induced release, the rate of the SG release following the addition of 100 µM histamine or 100 µM ATP produced a significantly slower initial rate of SG release. When the total amount of released SG during the first 8 min of the stimulation period was analyzed (inset, Figure 2), it was apparent that 1.0 µM HNE promoted far more release (52%) than 100 µM histamine (10%) and 100 µM ATP (6%). In control conditions (i.e., without any added secretagogue), the basal rate of SG exocytosis was less than 4% as assessed throughout the 30 min observation period.

View larger version (16K): [in this window] [in a new window]   Figure 2.   Time-related changes of SG exocytosis in HTG cells induced by HNE (1.0 µM), histamine (His, 100 µM) and ATP (100 µM) treatment. Data are expressed as % change in quinacrine fluorescence following secretagogues addition at time zero (100%). Each point represents the mean of 56, 92, and 72 SG analyzed after ATP, histamine or HNE treatment, respectively, over the 30-min observation period. The bars in the inset represent the actual percentage of SG release obtained by each secretagogue after an 8-min incubation period with the secretagogue.

Correlation between SG Exocytosis Reponses and ALP Secretion

To be certain that the rate of fluorescent SG release monitored by fluorescence videomicroscopy was associated with an active degranulation of secretory products by HTG cells, we analyzed both the intracellular ALP immunostaining and the amount of stimulated ALP recovered in degranulation supernatants by ELISA from HNE-treated HTG cells over a 10 min period. The staining of untreated cells (Figure 3A, time 0) showed an intense staining of ALP within numerous cytoplasmic SG, thus confirming the serous gland phenotype of HTG cells in culture. The addition of 1.0 µM HNE in cultured cells provoked a marked decrease of intracellular ALP granular staining after an 8-min incubation period (Figure 3B) and completly abolished staining after the 30-min incubation period (Figure 3C).

View larger version (75K): [in this window] [in a new window]   Figure 3.   Time-related changes in ALP immunofluorescent staining in HTG cells after treatment with 1.0 µM HNE. (A) Granular staining of ALP in methanol-fixed HTG cells is evident in the cytoplasm of most cells in the absence of HNE (time 0); (B) a faint staining for ALP is observed after an 8-min treatment; and (C) completely disappears after a 30-min treatment. Respective Normaski images of (0 min) untreated (D), 8 min (E), and 30 min (F  ) treated HTG cells in culture. A-F: magnification ×400.

As shown in Figure 4, the kinetics of the decrease of the SG fluorescence was significantly and negatively correlated (r = 0.86, P < 0.01) with the increase in the amount of ALP released in degranulation supernatants as assessed by ELISA. Over a 10-min observation period, the effects of 1.0 µM HNE treatment on the ALP secretion was time-dependent and caused an increase from 150% to 300% in the ALP release over baseline after 2 min and 8 min of treatment, respectively. No evidence of HNE-mediated cell detachment and cytotoxicity was detected by optical microscopy (Figures 3E and 3F) and LDH determination (data not shown) even after a 30-min treatment period.

View larger version (11K): [in this window] [in a new window]   Figure 4.   Correlation between quinacrine fluorescence in SG (% change due to 1.0 µM HNE added at zero time: 100%) and ALP relative secretory rate of HTG cells exposed to 1.0 µM HNE treatment for a 10-min period. The ALP relative secretory rate (ALP secreted in the degranulation supernatant/ALP secreted in the control experiment) is determined at increasing times (from 0 to 10 min) on quadruplicate assays. Results are expressed as means ± SD.

Patterns of [Ca²⁺]_i Responses to Various Secretagogues

Figure 5 shows the 3 typical profiles of dynamic changes in [Ca²⁺]_i observed by quantitative UV laser microspectrofluorometry from 5 adjacent HTG cells in monolayer culture visualized by video-optical microscopy (Figure 5A) following the addition of 1.0 µM HNE (Figure 5B), 100 µM histamine (Figure 5C) or 100 µM ATP (Figure 5D), respectively. Despite the cell to cell variability, dynamic changes in [Ca²⁺]_i were reproducible for each of the secretagogues assessed. Although the [Ca²⁺]_i response pattern was variable in magnitude and duration, the response to three secretagogues was as follows: (1) During exposure to 1.0 µM HNE (Figure 5B), the HTG cells exhibited [Ca²⁺]_i asynchronous oscillations with or without a detectable time lag (up to 3 min). This was followed in some cells by an elevated plateau with repetitive spikes in [Ca²⁺]_i which continued for 2 min to 8 min, while in other cells, [Ca²⁺]_i levels returned to prestimulated values despite the continued presence of HNE. The number of cells displaying oscillations in [Ca²⁺]_i increased with increasing HNE concentrations (0.01 µM, 31%, n = 36; 0.1 µM, 44%, n = 30; and 1.0 µM, 52%, n = 42; Table 1); (2) In contrast, when HTG cells were exposed to 100 µM histamine (Figure 3C) or 100 µM ATP (Figure 3D), no asynchronous oscillations in [Ca²⁺]_i was observed, but rather, only a single and rapid transient rise in [Ca²⁺]_i occurred over the 10-min treatment period.

View larger version (108K): [in this window] [in a new window]   Figure 5.   Secretagogue-induced [Ca2+]i transients in single HTG cells. (A) Optical microscopic videoimage of HTG cells in monolayer culture during the [Ca2+]i measurement period. Time courses of [Ca2+]i transients are recorded in 5 adjacent (column 1-5) HTG cells, before and after the addition of 1.0 µM HNE (B), 100 µM histamine (C), and 100 µM ATP (D). In each cell, [Ca2+]i has been measured for 1 s, every 5 s. The onset of HNE, histamine and ATP addition is indicated by arrows. Of note is the difference in the patterns of [Ca2+]i response according to the added secretagogues: addition of HNE (B) provokes repetitive asynchronous oscillations in [Ca2+]i (particularly in cells 3 and 5); addition of histamine (C) only induces a rapid rise to peak in [Ca2+]i followed by a smooth decay back to baseline in all cells; while the addition of ATP produces a delayed onset of the rise to peak [Ca2+]i in a synchronous fashion in all analyzed cells. Recordings are representative of results obtained following the analysis of more than 15 fields of HTG cells under similar conditions. The color scale represents [Ca2+]i variations.

The effects of various concentrations of HNE, histamine, and ATP on the peak rises in [Ca²⁺]_i in HTG cells are given in Table . From basal [Ca²⁺]_i level of 65 ± 38 nM (n = 70), the exposure of HTG cells to HNE (1 µM) produced an increase in [Ca²⁺]_i of 109 ± 52 nM in the 52% (n = 45) of cells that responded (i.e., with an increase in [Ca²⁺]_i  20 nM). At lower concentrations of HNE (0.01 µM and 0.1 µM), the [Ca²⁺]_i increases remained similar in the 31-44% of the cells that responded. Upon exposure to 100 µM histamine or 100 µM ATP addition, the [Ca²⁺]_i increases were 240 ± 127 nM (n = 40) and 121 ± 19 nM (n = 18), respectively. Of note was the short delay time of 18 ± 15 s (n = 40) from the addition of 100 µM histamine to the peak increase in [Ca²⁺]_i seen in 100% of cells, compared with a delay of 160 ± 31 s (n = 18) in 76% of responding cells following exposure of the cells to 100 µM ATP (Table ).

Influence of Extracellular Ca²⁺ on [Ca²⁺]_i and Dynamics of SG Exocytosis

An increase in intracellular Ca²⁺ could be due to the release of Ca²⁺ from internal stores, or from an influx of the extracellular Ca²⁺ across the plasma membrane, or both. The relative contribution of extracellular Ca²⁺ to the increase in [Ca²⁺]_i on one hand, and to the rate of SG exocytotic release on the other, was examined by incubating cells in a Ca²⁺ and red phenol-free RPMI 1640 mM medium containing 5 mM EGTA. As shown in Table 2, an analysis of the first 8-min period of response showed that lowering extracellular Ca²⁺ significantly reduced (P < 0.01) both the rate of [Ca²⁺]_i increase and the extent of SG exocytotic release from HTG cells exposed to HNE. In contrast, because of the relatively low levels of secretory response to histamine and ATP, no significant increase in the rate of SG exocytosis was observed in HTG cells when exposed to histamine or ATP in the presence of a Ca²⁺-free medium.

Influence of Thapsigargin on [Ca²⁺]_i and SG Exocytosis Dynamics

The relative contribution of internal Ca²⁺ stores to the increase in [Ca²⁺]_i and the rate of SG release was directly assessed by depleting internal Ca²⁺ stores via the inhibition of the ER Ca²⁺-ATPase (20). The addition of 500 nM thapsigargin to the medium resulted in a slow increase in [Ca²⁺]_i that required 8-12 min to reach its maximal value (Figure 6). The response to thapsigargin was transient, with [Ca²⁺]_i returning to basal levels in HTG cells within 20 min. Despite the transient Ca²⁺ release following the thapsigargin addition, sufficient Ca²⁺ was retained 20 min after the addition of thapsigargin to support both an attenuated increase in [Ca²⁺]_i, in secretagogue-stimulated HTG cells and SG exocytosis. When HTG cells were stimulated for 10 min or more after the addition of thapsigargin, the rises in [Ca²⁺]_i after the addition of 1.0 µM HNE, 100 µM histamine or 100 µM ATP were significantly reduced (P < 0.01) compared with control values (Figure 6 and Table 3). These results indicate that [Ca²⁺]_i responses to HNE, histamine and ATP stimulation were critically dependent on the filling status of the internal Ca²⁺ stores of HTG cells. However, exposure to each of the secretagogues of thapsigargin-treated cells in a Ca²⁺ containing RPMI 1640 medium did not greatly affect the rate of SG exocytosis (Table ). For example, in the case of stimulation by 1.0 µM HNE or 100 µM histamine, 48% and 7% of SG were released in the first 8-min period, respectively, compared with 52% and 10% of SG exocytosis in untreated-HTG cells (Table ).

View larger version (16K): [in this window] [in a new window]   Figure 6.   Changes in [Ca2+]i in HTG cells in response to exposure to HNE (1.0 µM), histamine (His, 100 µM) and ATP (100 µM) addition in the absence or presence of thapsigargin (TSG, 500 nM). Note that the addition of TSG alone induces transient and slow increases in [Ca2+]i and markedly reduces the effects of both HNE, histamine and ATP.

In order to verify that the source of Ca²⁺ in untreated-HTG cells contributing to the SG exocytotic process in stimulated cells was extracellular, the addition of each of the secretagogues was also analyzed in thapsigargin-treated cells after removal of extracellular Ca²⁺ by incubating cells in a Ca²⁺-free RPMI 1640 medium containing 5 mM EGTA. Under these Ca²⁺-free conditions, no significant increase in either the [Ca²⁺]_i increase or the SG exocytotic release could be detected following addition of the secretagogue (Table ). Thus, secretagogue-induced increases in the rate the SG exocytotic in HTG cells were clearly dependent on an influx of extracellular Ca²⁺.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

The present study demonstrates that the exposure of HTG cells to three important secretagogues associated with human airway inflammation and diseases (i.e., HNE, histamine, and ATP) leads to different patterns of SG exocytotic responses associated with different intracellular Ca²⁺ mobilization processes. Although HNE has thus far been considered as the most potent secretagogue of airway cells (7, 15), little is known of the signal transduction pathways studied at the single-cell level by which HNE induces an exocytotic release of SG from HTG cells. The results presented here rule out the possibility that a rapid onset of exocytotic responses obtained in HTG cells after HNE treatment is secondary to the generation of asynchronous oscillations in [Ca²⁺]_i coupled with an influx of extracellular Ca²⁺, rather than to a single and transient [Ca²⁺]_i peak rise within the cytoplasm of the HTG cells.

Evidence that the HNE-induced secretory response of HTG cells is due to exocytosis of SG is given as follows: (1) an immunocytochemical analysis of HTG cells demonstrates the presence of ALP, a specific secretory protein for serous gland cell-type, within cytoplasmic SG prior to HNE stimulation, and a complete loss of immunostained SG in response to HNE over a 10 min treatment period; and (2) kinetic studies of SG release detected in single cells by monitoring quinacrine fluorescence show a significant negative correlation (r = 0.86, P < 0.001) between the decrease in cytoplasmic fluorescent SG and an increase in the amount of ALP released in degranulation supernatants. The molecular mechanism by which HNE promotes the fast exocytosis of SG remains unexplained. It is possible that the mechanism is predominantly due to the active catalytic site of HNE, as observed for the cathepsin G-induced activation of platelets (27) and the chymase- induced degranulation of mast cells (28). This possibility is supported by recent studies showing that exogenous HNE and cathepsin G readily bind to sites on the plasma membrane of polymorphonuclear leukocytes (29) and alveolar macrophages (30). Interestingly, the binding sites of exogenously added HNE have been found to be numerous (> 10⁶ sites/cell), but exhibit low-affinity/high capacity-binding characteristics (31). If present on the plasma membrane of HTG cells, such low-affinity/high capacity-binding sites for HNE could explain the low but increasing percentage of HTG cells responding with rises in [Ca²⁺]_i in response to increasing HNE concentrations. Moreover, we have demonstrated in a previous study (26) that an HNE-specific inhibitor, elafin, which has recently been identified and characterized in bronchial secretions (32) significantly reduced both the HNE-induced rise in [Ca²⁺]_i and the number of HTG cells responding to HNE exposure. Taken together, the results suggest that the asynchronous oscillations in [Ca²⁺]_i generated in HTG cells exposed to HNE require a catalytic site for the enzyme on the cell plasma membrane of HTG cells and seem to involve a site-specific enzymatic process.

Our results also demonstrate that even when a high peak rise cytosolic in Ca²⁺occurs in HTG cells, the rate of SG release does not necessarily increase. As shown in Table , the peak rises in [Ca²⁺]_i are higher than those observed after HNE treatment following the addition of histamine or ATP, but without significantly increasing the rate of SG release. It clearly appears that the rate of SG release is dependent on the oscillations in [Ca²⁺]_i and an influx of extracellular Ca²⁺ elicited by HNE treatment, but not on the absolute level of the cytosolic rise in [Ca²⁺]_i per se. Only histamine elicited a rapid transient rise in [Ca²⁺]_i which was followed by a smooth decay back to baseline. This pattern of [Ca²⁺]_i response is in accordance with the histamine response described in other secretory cell types, including colonic (33) and immortalized airway surface epithelium cell lines (34, 35); the latter suggesting that histamine, acting through H1-type receptors, triggered a rapid [Ca²⁺]_i transient and sustained Ca²⁺ influx. In contrast to the rapid peak rise in [Ca²⁺]_i induced by histamine, the addition of ATP to HTG cells generated a delayed increase in the peak rise in [Ca²⁺]_i which was associated with a low rate of SG exocytotic release. Concentrations of ATP as high as 100 µM failed to elicit a significant SG release in HTG cells. Studies employing mucin-secreting primary cultures have shown that surface goblet cells are stimulated by ATP and platelet activating factor in cat and hamster airway tissues (11, 13). Lethem and associates (14) have demonstrated that ATP causes the activation of mucin granule release in goblet cells from human airway epithelial explants. Our cell preparation consists of a pure well-differentiated submucosal serous gland epithelial cell population, devoid of airway surface epithelial cells and exhibiting characteristics of tracheal submucosal serous gland cell-type (19, 21, and this study). It is possible that the mucous and serous secretory cell phenotypes may respond to secretagogues via different signal transduction pathways. Different sensitivities to ATP may reflect cell variability at one or more of the steps in the stimulus- secretion response pathway. This may explain why ATP failed to stimulate a significant SG exocytotic release in HTG cells.

As shown in Figure 7, every intervention that produced an increase in [Ca²⁺]_i levels in HTG cells following exposure to a secretagogue did not necessarily produce an increase in the rate of SG exocytosis. The highest increase in SG exocytosis appeared to be related to HNE treatment in the presence of extracellular Ca²⁺. It remains unclear, however, how changes in asynchronous oscillations in [Ca²⁺]_i are transduced into changes in the rate of SG exocytosis. To explain our observations of stimulus-secretion coupling in HTG cells, we propose that the onset of the SG release in HTG cells is initiated by Ca²⁺ entering the cell from the extracellular mediumpossibly though Ca²⁺ conducting channelsand acting in proximity to the SG sites where local [Ca²⁺]_i oscillations initiate the process leading to SG exocytotic release. Several recent reports agree on the fact that the pattern of dynamic changes in [Ca²⁺]_i referred to as Ca²⁺ spiking (36) plays a fundamental role in exocytosis, i.e., the last transport step in the secretory pathway, being a process that involves the specific interaction and fusion of SG with the plasma membrane. In other exocrine gland cells, such as mouse pancreatic cells (40, 41) where Ca²⁺ oscillations have been described, it is generally believed that the entry of Ca²⁺ from outside the cell is required to replenish depleted Ca²⁺ stores. However, the mechanisms by which Ca²⁺ oscillates seem to vary from one cell type to another (16, 36). In order to discriminate between the dynamic changes in [Ca²⁺]_i caused by a secretagogue-induced influx of extracellular Ca²⁺ and those brought by inositol triphosphate (IP₃)-dependent intracellular Ca²⁺ release, internal Ca²⁺ stores in HTG cells were emptied by the Ca²⁺-ATPase inhibitor, thapsigargin (20). In a Ca²⁺-containing medium, the exposure of each of the secretagogues to thapsigargin-treated cells resulted in a slight increase in [Ca²⁺]_i in stimulated cells. To date, it is well accepted that the depletion of the IP₃-sensitive Ca²⁺ stores causes an influx of Ca²⁺ across the plasma membrane to the cytosol, termed capacitative Ca²⁺ entry, that is thought to be the basis for sustained Ca²⁺ responses in cells (36, 42, 43). Thus, thapsigargin depletes intracellular Ca²⁺ stores and stimulates the influx of extracellular Ca²⁺ into HTG cells which is consistent with capacitative Ca²⁺ entry being operational in HTG cells. Since the oscillations in [Ca²⁺]_i in HTG cells do not completely disappear following exposure of the cells to thapsigargin, we propose that asynchronous oscillations in [Ca²⁺]_i that are observed in HTG cells after HNE treatment, and not after histamine and ATP treatment are due to an influx of extracellular Ca²⁺ rather than to the release of internal Ca²⁺ stores.

View larger version (11K): [in this window] [in a new window]   Figure 7.   Relationship between the [Ca2+]i increase and rate of SG exocytosis after the 8-min incubation period. Plotted are the means ± SD of the SG exocytosis data versus the data of [Ca2+]i increases from all experiments. Definitions of symbols are as follows: filled squares, peak histamine responses (1.0 and 100 µM) in the presence of extracellular Ca2+; filled circles, peak HNE responses (0.01, 0.1, and 1.0 µM) in the presence of extracellular Ca2+; filled triangle, peak ATP response (100 µM) in the presence of extracellular Ca2+; open square, peak histamine response (100 µM) in the absence of extracellular Ca2+; open circle, peak HNE response (1.0 µM) in the absence of extracellular Ca2+; open triangle, peak ATP response (100 µM) in the absence of extracellular Ca2+.

In conclusion, our observation of HNE-triggered [Ca²⁺]_i oscillations and a matching modulation of the rapid exocytosis response signal in HTG cells shows an interesting pathophysiologic consequence of asynchronous oscillations in [Ca²⁺]_i in human airway gland cells, and provides a further example that fast exocytosis responses can follow rapid dynamic changes in [Ca²⁺]_i. In in vivo pathologic situations, where the secretion rates of airway submucosal gland cells can be markedly increased by HNE (2, 8, 9), it is highly plausible that asynchronous oscillations in [Ca²⁺]_i may initiate stimulus-response coupling leading to exocytotic SG release and mediate the fast secretory responses into the human airway lumen. It seems clear that asynchronous oscillations in [Ca²⁺]_i coupled with an influx of extracellular Ca²⁺ play a central role in regulating the rate of SG exocytosis in airway serous gland cells. However, the Ca²⁺ sensor and molecular events related to intracellular Ca²⁺ mobilization in normal and pathologic human airway submucosal gland cells, together with mucus hypersecretion observed in patients with chronic bronchitis and cystic fibrosis, remain to be elucidated.