INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Histamine plays an important role in triggering an asthmatic reaction. The human lung is a considerably rich source of histamine stored mainly in the secretory granules of mast cells. These cells are widely distributed throughout the human respiratory tract, located within the walls of conducting airways and of alveoli. In the bronchial wall mast cells are located in the airway mucosa rather close to the basement membrane and surface epithelium. Mast cells are even found between epithelial cells and can also abut directly on the bronchial lumen (1, 2). This latter mast cell population is placed in a strategically most prominent location to be among the first cells to interact with inhaled antigens and other agents. The same applies for basophils whose number in the airway mucosa is enhanced in asthmatics (3). Upon activation mast cells and basophils release histamine, which causes bronchial obstruction, vascular leakage, and mucus secretion. It is generally accepted that mast cells are involved in the early asthmatic reaction (4), whereas basophils appear to play a role in the late reaction (5). A frequent activation by antigens and environmental agents of superficially located mast cells and basophils may play a key role in bronchial obstruction. Importantly, it has been shown recently that bacteria like Hemophilus influenzae can activate mast cells and basophils directly by an IgE-independent mechanism (8). A regulatory pathway that controls the releasability of these mucosal mast cells and basophils may represent a most important mechanism to limit histamine release in the airways.

A regulatory pathway between the cholinergic system and histamine stores has been established in the alimentary tract. Cholinergic agonists stimulate histamine release from submaxillary and parotid glands and from mast cells isolated from the peritoneal and pleural cavity (9, 10). Vagal stimulation leads to the release of gastric histamine (11, 12). Likewise, acetylcholine evokes histamine release from human isolated adenoidal mast cells (13). In one report it was demonstrated that cholinergic agonists enhance the immunologically evoked histamine release from human lung tissue (14). In this latter study lung fragments, i.e., parenchymal tissue, was used. Studies with conducting bronchi have, so far, not been performed to investigate the effects of cholinergic agonists on the releasability of mast cells localized in the airway mucosa. In the last years, however, evidence has been accumulated showing that mast cells from different origin (peritoneal, mucosal, lung parenchyma, and skin mast cells) exhibit important biochemical as well as functional heterogeneity (15). Activators as well as inhibitors effective, for example, on peritoneal mast cells can no longer be predicted to affect mucosal mast cell function in the same manner. Therefore, the present study was performed to study the effects of acetylcholine on the release of histamine from human isolated conducting bronchi. Mast cells are known to become activated by a transient rise in cytosolic calcium. This can be mediated experimentally by a calcium ionophore or by the activation of the high affinity IgE receptor (FcRI) (15, 16). Both stimuli were used in the present study, i.e., histamine release from human isolated bronchi was stimulated either by the calcium ionophore A23187 or immunologically by the application of antihuman IgE antibody (ab).

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Tissue Preparation

The protocol for obtaining human tissue was approved by the local ethical review board for human studies (Landersärztekammer Rheinland-Pfalz, Germany). Human tissue was obtained at surgery from patients with lung cancer (n = 30) but without a history of chronic airway disease. Therefore, in the present experiments histamine release originated predominantly from mast cells because basophils are nearly absent in the airway mucosa of nonasthmatics (3). Atropine (1 mg) was routinely applied as premedication. The time between atropine application and the start of the experiment was at least 6 h. This long time period and vigorous washing before the start of the experiments minimized atropine remaining in the tissue. Immediately after lobectomy, tumor-free human bronchi were longitudinally opened and dissected down to a diameter of 4 mm, carefully cleaned from adhering parenchyma, and placed in oxygenated, ice-cold salt solution (composition in mmol/L: 125 NaCl, 23.8 NaHCO₃, 5.05 glucose, 2.68 KCl, 1.80 CaCl₂, 1.04 MgCl₂, 0.54 NaH₂PO₄, 0.057 ascorbic acid, and 0.001 choline chloride at pH 7.3). In the laboratory opened bronchi (diameter from 10 to 4 mm) were washed several times in the oxygenated salt solution, and cut into pieces with a width of about 3 to 5 mm, a length of 1.5 to 2 cm, and a weight of roughly 150 mg. Bronchi were set up vertically in 2-ml organ baths under a tension of 12 g. This tension was applied by a 12 g weight, which was connected via cotton thread and a reel with the tissue. In some experiments contractions were recorded by a force displacement transducer and displayed on a Hellige recorder. The surface epithelium of human bronchi was mechanically removed when indicated. In order to remove the surface epithelium, longitudinally opened bronchi were fixed in a Petri dish with the luminal surface facing upwards, and a cotton-tipped applicator (Q-tip) was gently rubbed for 5 s along the luminal surface. It has repeatedly been demonstrated that this manipulation did not penetrate the basal membrane, i.e., the underlying lamina propria remained intact (17, 18). In the experiments with electrical field stimulation the organ bath contained two platinum electrodes, and the bronchi were placed between these electrodes. Biphasic pulses (pulse duration, 1 ms; current strength, 250 mA) were delivered from a Grass S6 stimulator (Grass Instruments, Quincy, MA). In some experiments bronchi were homogenized (17) to determine the histamine content of the airway wall.

Superfusion, Incubation and Stimulation Protocol

For equilibration and washing, bronchi were at first superfused (2 ml/min for 30 min) with the oxygenated salt solution warmed to a constant temperature of 36° C. Thereafter superfusion was stopped and bronchi were incubated in 1.1 ml of the salt solution, which was exchanged in 5-min periods. The first three incubates were discarded and the subsequent (time zero) (see Figure 1) collected for measuring histamine content. In each individual experiment 12 samples were collected, i.e., incubation lasted 60 min. The first three samples represent the spontaneous histamine release collected in the absence of the test substance (acetylcholine, oxotremorine, physostigmine, atropine). The test substances were added from the sixteenth minute of incubation and remained. When atropine was used in interaction experiments with acetylcholine, the antagonist was already present from the start of the incubation (time zero) and acetylcholine added at the sixteenth minute of incubation. When the effect of atropine alone was tested, the antagonist was present from the sixteenth minute of incubation.

View larger version (38K): [in this window] [in a new window]   Figure 1.   Control experiments showing spontaneous and stimulated histamine release from human bronchi. Human isolated bronchi were incubated in physiologic salt solution, and the medium was exchanged in 5-min intervals. Surface epithelium remained intact (panels A, B, and C) or was removed mechanically at the start of the experiments (D; EPI-). Histamine release is expressed as a percentage of the control release (16th to 30th min; mean ± SEM). (A) Spontaneous histamine release (control release: 3.5 ± 0.5 nmol/g × 5 min; n = 9). (B) The calcium ionophore A23187 was added at the 35th min by a bolus injection given a final bath concentration of 10 µmol/L and washed out 1 min later (control release: 2.2 ± 0.3 nmol/g × 5 min; stimulated increase above control release: 148 ± 28%; n = 11). (C) Antihuman IgE ab (final dilution, 1:3,000) was present from the 31st to 45th min of incubation (control release: 1.4 ± 0.4 nmol/g × 5 min; stimulated increase above control release: 127 ± 32%, n = 6). (D) Same experiments as shown in B using epithelium-denuded bronchi (EPI-, control release: 3.4 ± 1.0 nmol/g × 5 min; n = 6).

Mast cells were stimulated either by the calcium ionophore A23187 or by antihuman IgE ab. A23187 was added by a 50-µl bolus injection at the thirty-fifth minute of incubation (final bath concentration, 10 µmol/L) and washed out 1 min later with the subsequent medium exchange (see Figure 1B). Antihuman IgE ab (dilution 1 to 3,000) was exposed from thirty-first minute to the forty-fifth minute of incubation (see Figure 1C). In some experiments spontaneous histamine release was measured throughout the incubation without applying any test substance (see Figure 1A). At the end of the incubation period bronchi were dried and weighed.

Analytic Procedure

Histamine content of the medium was determined after derivatization with o-phthaldialdehyde by HPLC and fluorometric detection. Medium (900 µl) was alkalized with 200 µl NaOH solution (1 mol/L), for derivatization 200 µl o-phthaldialdehyde in methanol were added at room temperature (0.1%, vol/vol in methanol). The derivatization was stopped after 3 min by adding 35 µl phosphoric acid (50%, vol/ vol) and by adjusting pH to 4. The samples were passed by automatic injector (Promis II; Spark Holland, Emmen, The Netherlands) onto a reverse phase HPLC column (RP-select B; 5 µm column, which was fitted with a precolumn) by a recirculating mobile phase of the following composition: acetic acid, 90 mmol/L; 1-pentanesulfonic acid, 6.6 mmol/L; acetonitrile, 15% (vol/vol). The reaction products between histamine and o-phthaldialdehyde were detected by a Shimadzu RF-530 fluorescence monitor (wavelength of excitation and emission, 360 nm and 450 nm, respectively) and recorded by an integrator. Quantification was done by comparison with external histamine standard.

Calculations and Statistical Analysis

Histamine content of the incubation medium was normalized to 1 g bronchus. The mean histamine content of three prestimulation periods (16th to 30th min of incubation) (see Figure 1) was regarded as 100% (individual control release), and histamine release was expressed as a percentage of this individual control. Stimulated histamine release was calculated by comparing three subsequent samples after mast cell activation (A23187: 36th to 50th min of incubation; antihuman IgE ab: 31st to 45th min of incubation) with three prestimulation samples (16th to 30th min of incubation; control release). The results are given as means ± standard error of the mean (SEM). The significance of differences was evaluated by Student's t test (unpaired), for multiple comparison the modified t test according to Bonferroni was used (19). For statistical analysis and calculation of IC₅₀ value the computer program InStat^® was used; p values < 0.05 were regarded as significant.

Drugs and Special Chemicals

The following drugs were used: acetylcholine chloride (Sigma Chemie, Munic, Germany); anti-human IgEab (Sigma Chemie); atropine sulfate (Merck, Darmstadt, Germany); bradykinin (Sigma Chemie); calcium ionophore A23187 (Sigma Chemie); choline chloride (Sigma Chemie); compound 48/80 (Sigma Chemie); histamine diphosphate (Sigma Chemie); methanol (Roth, Karlsruhe, Germany); o-phthaldialdehyde (Sigma Chemie); oxotremorine sesquifumarate (Sigma Chemie); physostigmine hemisulfate (Sigma Chemie).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Control Experiments

Human epithelium-intact segmental bronchi contained 260 ± 30 nmol/g (n = 4) histamine. Removal of the surface epithelium reduced the histamine tissue content to 140 ± 30 nmol/g (n = 4). Spontaneous histamine release from the isolated epithelium-intact bronchi showed some tendency to increase over the 60-min incubation period (see Figure 1A), starting with 2.7 ± 0.45 nmol/g × 5 min and ending with 4.1 ± 0.8 nmol/g × 5 min (n = 9). The increase in spontaneous histamine release occurred preferentially within the first 15 min of incubation, but it kept rather constant over the remaining period (Figure 1A). None of the test substances (acetylcholine, oxotremorine, physostigmine, atropine) affected the spontaneous histamine release (see Figures 2 and 4).

View larger version (37K): [in this window] [in a new window]   Figure 2.   Inhibition by applied acetylcholine (A) or physostigmine (B) of A23187-evoked histamine release from human isolated bronchi. The experimental protocol is described in Figure 1. (A) Acetylcholine (ACh, 10 nmol/L) was added at the 16th min and remained; given are the means ± SEM of six experiments (control release: 2.0 ± 0.6 nmol/g × 5 min; stimulated increase above control release: 27 ± 5%, n = 6). (B) physostigmine (PHY, 100 nmol/L) was added at the 16th min and remained (control release: 1.8 ± 0.2 nmol/g × 5 min; stimulated increase above control release 60 ± 15%, n = 12). Open columns (dotted lines) represent percent increase under control conditions, i.e., in the absence of applied acetylcholine or physostigmine.

View larger version (40K): [in this window] [in a new window]   Figure 4.   Effect of exogenous acetylcholine (A) and atropine (B) on antihuman IgE ab-evoked histamine release from human isolated bronchi. The experimental protocol is described in Figure 1. (A) Acetylcholine (ACh, 100 nmol/L) was added at the 16th min and remained (control release: 2.2 ± 0.6 nmol/g; stimulated increase above control release: 39 ± 9%; means ± SEM of six experiments). Open columns (dotted lines) indicate percent increase under control conditions, i.e., in the absence of applied acetylcholine. (B) atropine (ATR) (0.1 µmol/L) was added at the 16th min and remained (control release: 1.1 ± 0.2 nmol/g; stimulated increase above control release: 138 ± 28%, n = 4).

A 1-min exposure of human bronchi to the calcium ionophore A23187 caused a sharp increase in the release of histamine (Figure 1B). Histamine release (36th to 50th min) increased by 148 ± 28% (n = 11) compared with the prestimulation level (16th to 30th min) (Figure 1B). When the calcium ionophore was added a second time to the same preparation later on (120 min of incubation) only a marginal increase in histamine release occurred (n = 4, not shown). Therefore, each individual preparation was stimulated for one time only.

In some experiments the surface epithelium was removed mechanically before the bronchi were placed into the organ bath. When A23187 was applied to epithelium-denuded bronchi the calcium ionophore was ineffective, i.e., an increase in histamine release was not observed in epithelium-denuded bronchi (Figure 1D). This observation indicates that A23187 preferentially activates mast cells localized in or close to the airway surface epithelium.

Mast cells were immunologically activated by the application of antihuman IgE ab present from the 31st to 45th min of incubation (see Figure 1C). The release kinetic was somewhat slower with antihuman IgE ab than with the calcium-ionophore, but the percentage increase in histamine release corresponded between both stimuli (antihuman IgE ab: increase by 127 ± 32%, n = 6) (compare Figure 1B and C). Finally, other release stimuli were tested. Neither bradykinin (1 µmol/L; n = 3) nor compound 48/80 (10 µg/ml; n = 4) triggered a significant increase in histamine release from human isolated epithelium-intact bronchi (not shown). Also, transmural electrical stimulation (1,200 pulses at 15 Hz) did not affect the release of histamine into the bath medium (1.9 ± 0.4 and 2.0 ± 0.5 nmol/g × 5 min before and after electrical stimulation, respectively, n = 6).

Effect of Acetylcholine, Oxotremorine, and Physostigmine on A23187-evoked Histamine Release

Acetylcholine, added at the 16th min of incubation to the medium, strongly inhibited A23187-evoked histamine release, an effect that was mediated by rather low concentrations. For example, 10 nmol/L acetylcholine reduced A23187-induced histamine release by 82% (Figures 2A and 3). The complete concentration-response curve is shown in Figure 3. Acetylcholine significantly inhibited A23187-evoked histamine release in concentrations as low as 0.1 nmol/L and was maximally effective at a concentration of 100 nmol/L (inhibition by 89%). The degree of inhibition attenuated with acetylcholine concentrations higher than 0.1 µmol/L. Thus, a bell-shaped concentration-response curve was obtained (Figure 3). According to the descending part of the curve an IC₅₀ value of 0.03 nmol/L (95% confidence interval, 0.003 to 0.2 nmol/L) was calculated for the inhibitory action of acetylcholine on evoked histamine release. Likewise, oxotremorine, a stable agonist at muscarinic receptors, inhibited A23187-evoked histamine release. A concentration as low as 1 nmol/L suppressed A231287-evoked histamine release completely, whereas concentrations lower or higher than 1 nmol/L were less effective (Table 1).

View larger version (12K): [in this window] [in a new window]   Figure 3.   Concentration-response curve of acetylcholine on A23187-evoked histamine release. The experimental protocol is shown in Figure 2. Closed squares: effect of acetylcholine alone; closed triangles: effect of 10 nmol/L acetylcholine in the presence of 300 nmol/L atropine, which was present from time zero. Given are the means ± SEM of four to nine experiments. Significance of the difference from the control: ** p < 0.01.

Finally, it was tested whether physostigmine, an inhibitor of acetylcholinesterase enzyme activity, affected A23187-evoked histamine release. A low concentration of 0.1 µmol/L physostigmine was applied to inhibit cholinesterase activity only partly because already rather low concentrations of acetylcholine were effective (see bell-shaped concentration-response curve of acetylcholine) (Figure 3). Physostigmine, present from the 16th min of incubation, significantly (p < 0.01) reduced A23187-evoked histamine release by 58% (Figure 2B).

When contractions were recorded it was found that neither 10 nmol/L acetylcholine nor 1 nmol/L oxotremorine, i.e., strongly effective concentrations in inhibiting histamine release (Figure 3 and Table 1), increased the tone of the isolated bronchi. A high concentration of oxotremorine (1,000 µmol/L) induced a contractile response of 14 ± 2 mN (n = 8), but it was without any effect on spontaneous histamine release (2.8 ± 0.6 and 2.7 ± 0.4 nmol/g = 5 min before and after the application of oxotremorine, respectively; n = 5).

Effect of Acetylcholine on Antihuman IgE ab-evoked Histamine Release

Applied acetylcholine (0.1 µmol/L) potently inhibited immunologically evoked histamine release from human isolated bronchi by about 70% (Figure 4). The effects of 0.01 and 1 µmol/L acetylcholine on antihuman IgE ab-evoked histamine release are given in Table 2. Again the inhibitory potency declined with high acetylcholine concentration, a pattern observed already in the experiments with the calcium ionophore.

Effect of Atropine on Evoked Histamine Release

To unmask a possible endogenous activation of the inhibitory muscarinic receptors the nonselective receptor antagonist atropine was present from the 16th min of incubation onwards. Atropine (300 nmol/L) did not increase the release of histamine evoked by A23187 or by antihuman IgE ab (120 ± 30%, n = 7, and 140 ± 28%, n = 4, respectively, compare with the respective controls above). However, atropine increased the kinetic of histamine release evoked by antihuman IgE ab. In the absence of atropine, immunologically evoked histamine release peaked in the second stimulation period (see Figure 1C; histamine release_{[36th-40th min]} versus histamine release _{[31st-35th min]}: 1.50 ± 0.21, n = 6), whereas in the presence of atropine histamine release peaked already in the first stimulation period (see Figure 4B; histamine release_{[36th-40th min]} versus histamine release_{[31st-35th min]}: 0.92 ± 0.06, n = 4; p < 0.05).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The present experiments demonstrate for the first time an inhibitory link between muscarinic receptors and histamine release in human isolated bronchi. Applied acetylcholine, oxotremorine, a stable agonist at muscarinic receptors, and physostigmine, an acetylcholinesterase inhibitor, substantially inhibited the calcium ionophore-evoked histamine release. For example, 10 nmol/L acetylcholine reduced A23187-evoked histamine release by 89%, an effect that was antagonized by atropine. Oxotremorine, 1 nmol/L, even suppressed A23187-evoked histamine release completely. All these results indicate that muscarinic receptors are involved in inhibiting histamine release in human airways. In the present experiments bronchi from nonasthmatics have been used. Therefore, basophils are not expected to contribute significantly to histamine release. We conclude that in human airways stimulation of muscarinic receptors strongly inhibits the releasability of mucosally localized mast cells. Importantly, acetylcholine was also effective in inhibiting histamine release evoked by antihuman IgE ab, i.e., by cross-linking the high affinity Fc-receptor, a physiologic activation pathway. Therefore, we additionally conclude that in vivo activation of mucosal mast cells by inhaled antigens can be inhibited by stimulation of muscarinic receptors. At increasing concentrations acetylcholine and oxotremorine lost their inhibitory potency. Probably, a second facilitatory receptor type with lower affinity exists on mucosal mast cells. Alternatively, high agonist concentrations may cause receptor desensitization or may trigger additional intracellular signals counteracting the inhibitory pathway. These questions as well as the characterization of the subtype of receptors involved, will be a matter for further research.

It might be argued that the inhibitory effect of muscarinic agonists on histamine release reflects a purely mechanical phenomenon, i.e., the contractile response might have retarded histamine from being released into the medium and the ionophore from penetrating into the tissue. This possibility can be excluded for several reasons. Firstly, strongly effective inhibitory concentrations of acetylcholine and oxotremorine were too low to induce a contractile response. Secondly, the experiments with 1,000 nmol/L oxotremorine indicated that the spontaneous histamine release was not affected despite the induction of a contractile response. Thirdly, electrical transmural stimulation did not affect the release of histamine in the present experiments. Taken together, all these observations clearly indicate that the inhibition of histamine release was not caused by a change of the smooth muscle tone. In this context it should also be considered that the ionophore-induced histamine release originated from the surface epithelium, i.e., could be liberated directly into the bath medium.

Mast cells are a heterogenous cell population, cells of the mucosal type and connective tissue mast cells represent two different cell populations (15). What type of mast cell has been characterized in the present study? To answer this question several experimental observations have to be considered. The surface epithelium was removed in some experiments without penetrating the basal membrane, i.e., the underlying lamina propria remained intact (see 16-18). Removal of the epithelium reduced the histamine content of the airway wall by 44%, but abolished A23187-evoked histamine release. Two important conclusions can be drawn from these results: (1) histamine appears to be concentrated in the surface epithelium in relation to the remaining airway wall; (2) A23187-stimulated histamine release originates preferentially from mast cells localized in or rather close to the airway surface epithelium, i.e., from mucosal mast cells. In line with this conclusion we found in the present study that neither bradykinin nor compound 48/80 triggered a significant release of histamine. Both secretagogues are also ineffective on mast cells recovered by bronchoalveolar lavage (20) or on intestinal mucosal mast cells. In contrast, these secretagogues stimulate histamine release from peritoneal and skin mast cells, which represent connective tissue mast cells (15). Moreover, it has been reported that mucosally localized mast cells that are recovered by bronchoalveolar lavage contain about 1 pg/cell histamine (21), whereas connective tissue mast cells store considerably more histamine (about 20 pg/cell) (15). This biochemical difference indicates that mast cells localized superficially in human conducting airways represent the mucosal cell population. Thus, to answer the aforementioned question we conclude that in the present in vitro experiments mast cells of the mucosal type have been characterized. The releasability of these mucosal mast cells is strongly inhibited by stimulation of muscarinic receptors.

Can endogenous acetylcholine stimulate the muscarinic receptors mediating the inhibitory effect? The experiments with physostigmine indicate that, at least after partial blockade of cholinesterase activity, endogenous acetylcholine can establish a threshold biophase concentration to stimulate the inhibitory muscarinic receptors. What are the sources of acetylcholine responsible for this inhibitory effect? It is generally accepted that in human airways the surface epithelium is not innervated by cholinergic neurons (22). Therefore, one might propose diffusion of neuronal acetylcholine released from intramural nerve terminals. Very recently, however, it had been demonstrated that the human airway epithelial cells express the synthesizing enzyme choline acetyltransferase, synthesize acetylcholine, and contain significant amounts of acetylcholine (17, 18, 26). Therefore, we assume that non-neuronal, epithelial acetylcholine is released in intimate contact to mucosal mast cells and directly controls their releasability. The high potency of acetylcholine (IC₅₀ value: 0.03 nmol/L) in inhibiting histamine release facilitates a direct interaction between epithelial cells and mast cells. This cross-talk appears as a most important regulatory pathway to keep mast cell activation below a threshold level and to prevent human lung from overstimulation by mast cell mediators. This mechanism may maintain a homeostasis in the airway mucosa, which is exposed continuously to indoor and outdoor environmental pollutants and infectious material.

An inhibitory tone should become unmasked by a facilitatory action of atropine on evoked histamine release. In the present in vitro experiments, atropine did not increase the amount of histamine released in response to A23187 or antihuman IgE ab. But in experiments with antihuman IgE ab atropine increased the release kinetic, showing a significantly higher histamine release upon the first 5-min incubation with antihuman IgE ab than in the absence of atropine (see Figure 4). This may be indicative for a facilitatory effect of atropine on cell releasability, i.e., the threshold level for exocytotic meditor release appears to be reduced in the presence of atropine. Muscarinic receptor antagonists are applied in the treatment of chronic bronchitis and asthma. Assuming an endogenous cholinergic tone under in vivo conditions, the treatment with atropine-like drugs would prevent acetylcholine from controlling mast cell function with the consequence of an enhanced mediator release. Importantly enough, it had been shown recently that local histamine release triggered by nasal antigen challenge was increased after atropine pretreatment but reduced by methacholine (27). Paradoxical reactions observed sometimes with inhaled atropine-like drugs and the ineffectiveness of these drugs in the treatment of allergic asthma may be related to a disinhibition of mast cell function in human conducting airways.

The present experiments demonstrate the first example for an inhibitory link between muscarinic receptors and histamine release. As already outlined above, a facilitatory link between acetylcholine and histamine release has repeatedly been shown. For example, histamine release from rat isolated peritoneal mast cells is facilitated by acetylcholine (10). The facilitatory effect occurred in a similar concentration range (0.01 to 1 nmol/L) as the inhibitory effect observed in the present study. To explain the opposite effects one has to consider that peritoneal mast cells and mucosal airway mast cells represent two different cell populations exhibiting morphologic and biochemical heterogeneity (15). A facilitatory effect of acetylcholine was found with mast cells originating from the peritoneal cavity, lung parenchyma, and glandular tissue. In the present experiments, however, mast cells of the mucosal type have been investigated. Thus, acetylcholine regulates the releasability of both mast cell populations differently: acetylcholine inhibits the releasability of airway mucosal mast cells but facilitates the releasability of connective tissue mast cells.

At this stage of research one can only speculate about the intracellular mechanisms by which muscarinic receptors mediate such a profound inhibition of the secretory histamine release. Phospholipase C activation, increased calcium and activation of GTP-proteins trigger exocytotic histamine release. Acetylcholine inhibited the calcium ionophore-induced histamine release. Thus, the inhibitory signal must operate downstream the calcium signal. Recent progress in illuminating the role of GTP-proteins has shown that at least one heterotrimeric G protein (Gi3) expressed in mast cell membrane and one small GTP-protein of the Rab family appear to be involved in membrane fusion and exocytosis (28). Probably, inhibitory muscarinic receptors may interfere with the GTP-binding proteins and therewith reduce releasability. The inhibitory effect was strongly restricted to the stimulated histamine release, whereas the spontaneous release was not affected. This pattern indicates that muscarinic receptors trigger a rather specific signal to inhibit stimulated exocytotic histamine release only.

In conclusion, the present experiments have shown that the release of histamine evoked from human isolated conducting bronchi is strongly inhibited by acetylcholine via stimulation of muscarinic receptors. Endogenous acetylcholine, originating possibly from epithelial cells, can stimulate these receptors to inhibit mucosal mast cell function. We propose that the cross-talk between epithelial cells and mucosal mast cells represents an important communication to limit mast cell activation and mediator release in human airways and to maintain mucosal homeostasis.