Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Since endothelin was identified in the culture medium of vascular endothelial cells (1), its effects on a wide variety of tissues, including airway smooth muscle, have been reported (2, 3). Endothelin is classified into the three subtypes of endothelin-1, -2, and -3, and the expression of endothelin-1, which is the most potent constrictor of airway smooth muscle among the subtypes (4), is upregulated in asthmatic bronchial epithelial cells (5). Furthermore, the concentration of endothelin-1 in bronchoalveolar lavage fluid (BALF) from patients with symptomatic asthma is increased (6), indicating the possible relevance of endothelin-1 in the pathophysiology of bronchial asthma. Meanwhile, histamine is a classical chemical mediator that is released from mast cells in asthmatic airways and causes contraction of airway smooth muscle (7), and may play an important role in initiating pathologic constriction of the airways.

Previous studies (8, 9) have shown some differences in the contractile response to endothelin or histamine and that to cholinergic muscarinic agonists, which are all well-known constrictors of airway smooth muscle. To date, many experiments have been done on the electrophysiologic mechanisms of airway smooth-muscle contraction in response to cholinergic agonists (10). Some investigators (12, 13) have reported that cholinergic stimulation causes membrane depolarization and stimulates voltage-dependent calcium channels (VDCCs) or Ca²⁺ influx, resulting in smooth-muscle-cell contraction. In addition, muscarinic agonists bound to muscarinic receptors are known to regulate multiple transduction pathways in airway smooth muscle, and are reported to play a role in contraction and relaxation of this muscle, indicating that they play a role in the pharmacomechanical coupling mechanism (11). However, information about the ionic and electrophysiologic mechanism of airway smooth-muscle contraction in response to chemical mediators, including histamine and endothelin, is limited. Therefore, we examined the effects of endothelin-1 and histamine on single airway smooth-muscle cells from bovine trachea, using the patch-clamp technique and comparing these effects with those of cholinergic agonists.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Cell Preparation

Fresh bovine tracheae were obtained from a local abattoir and immediately transferred to an ice-cold extracellular (bath) solution consisting of (in mM): 140 NaCl, 4.7 KCl, 1.13 MgCl₂, 1.2 CaCl₂, 10 glucose, and 10 N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (Hepes). The smooth-muscle layer was isolated from the posterior membranous portion of the trachea by removing surrounding connective-tissue and epithelium, and the isolated smooth-muscle strips were cut into small pieces. These smooth pieces were then dispersed enzymatically into single cells with collagenase (100 U/ml) and elastase (5 U/ml) for 40 min at 37°C. After incubation, the tissue pieces were gently agitated, and single smooth-muscle cells were resuspended in the bath solution.

Electrical Recordings

Standard patch-clamp recording techniques were used. A List LM-EPC7 (List Electronics, Darmstadt, Germany) or EPC9 (HEKA Electronics, Lambrecht/Pfalz, Germany) patch-clamp amplifier was used. Patch electrodes (Drummond Scientific Co., Broomall, PA) had a tip resistance of 2 to 5 M, and seal formation was obtained by gentle suction applied to the patch pipette. Typical seal resistances were 2 to 5 G. The liquid junction potential, membrane capacitance, and series resistance were compensated with the amplifier circuitry. The establishment of a whole-cell configuration under a clamp voltage of 40 mV was confirmed by an abrupt decrease in access resistance (3 to 8 M) and an increase in capacitive current, and 70 to 85% series resistance compensation was employed. The access resistance was estimated from the value required for compensation of series resistance, whereas the input resistance was measured by the steady-state change in potential induced by a hyperpolarizing current (duration: 120 ms). The plateau potential was measured for routine data analysis by examining the response to each agonist or agent. The membrane potential when contraction of a single smooth-muscle cell began on the television monitor coupled to the microscope used for cell magnification was also obtained, to determine the approximate value of the threshold potential for the contraction. Current signals were recorded with a thermal pen-recorder (Recti-Horiz-8K; NEC-SanEi, Tokyo, Japan), the bandwidth of which was 0 to 300 Hz. Membrane currents were digitized at 5 kHz and low-pass filtered at 1 kHz in most samples. However, in a few samples, data were analogously recorded and filtered at 100 Hz.

The solutions used were of the following compositions. Extracellular (bath) solution: 140 mM NaCl, 4.7 mM KCl, 1.13 mM MgCl₂, 1.2 mM CaCl₂, 10 mM Hepes, 10 mM glucose. Intracellular (pipette) solution: 140 mM KCl, 1.13 mM MgCl₂, 10 mM Hepes, 10 mM glucose, 1 or 0.5 mM ethyleneglycol-bis-(-aminoethyl ether)-N,N,N',N'-tetraacetic acid (EGTA), 1 mM Na₂ATP. Because the free Ca²⁺ concentration of the intracellular solution was as low as 20 to 30 nM, which was determined with the fluorescent dye Fura-2, in some experiments 0.5 mM CaCl₂ was added to the solution, resulting in a free Ca²⁺ concentration of 156 nM. Ca²⁺-free bath solution was obtained by increasing the concentration of EGTA to 5 mM in the pipette solution. Both bath and pipette solutions were at pH 7.2, and all experiments were done at room temperature (20 to 25°C). Before each experiment, the pH of the bath and pipette solutions was measured and adjusted with NaOH (bath solution) or KOH (pipette solution) when it changed. The fluids were superfused over the cell(s) by hydrostatic pressure-driven application (40 to 50 cm H₂O) through polyethylene tubes (0.5 mm inner diameter).

For a few samples, recordings were made from the cells, using the nystatin perforated-patch configuration of whole-cell recordings to prevent the dilution of intracellular constituents in the patch-pipette (14). After dissolving it with methanol (10 mg/ml), the final concentration of nystatin in the pipette solution was 175 µg/ml.

Chemicals

Drugs used in the present experiments were endothelin-1 (human) (Peptide Institute, Osaka, Japan), Fura-2/AM (Dojin, Kumamoto, Japan), acetylcholine (ACh) chloride, acetyl--methylcholine chloride (methacholine [MCh]), histamine dihydrochloride, and tetraethylammonium chloride (TEA) (all from Wako Pure Chemicals, Osaka, Japan). The other chemicals and reagents were all purchased from Sigma Chemical Co., St. Louis, MO.

Statistical Analysis

Data are shown as means ± SE. For mean comparison, the two-tailed paired or unpaired Student's t test was used, and the Cochran-Cox t test was used when Bartlett's test for uniformity of variance showed it to be nonuniform. A value of P < 0.05 was considered statistically significant. Furthermore, Fisher's exact probability test was used for incidence comparison.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

K⁺ Channels

A large-conductance K⁺ channel was found in the patched membrane, and it showed voltage-dependence. The current-voltage (I-V) relationship of the channel showed that the channel had a conductance of 127 ± 4 pS (n = 33, in the standard extracellular and intracellular solutions) around a membrane potential of 0 mV. The reversal potential was 74 mV when the membrane potentials were limited to less than 0 mV, and it was close to the equilibrium potential of K⁺ calculated with the Nernst equation. Furthermore, the current was abolished by removal of Ca²⁺ from the intracellular aspect of an inside-out patch by addition of 5 mM EGTA, indicating the Ca²⁺ dependence of the channel. Introduction of TEA (10 mM) to the bath solution (140 mM NaCl, 4.7 mM KCl, 1.13 mM MgCl₂, 1.2 mM CaCl₂, 10 mM Hepes, 10 mM glucose, and 10 mM TEA) in an outside-out patch also abolished the channel opening. However, internal application of TEA had no substantial effect on the channel activity. The channel conductance, Ca²⁺ dependence, and TEA sensitivity of these channels corresponded very well with those of the large-conductance, Ca²⁺-activated K⁺ (K_Ca) channel that has previously been described in airway smooth muscle (15). Ninety percent of the patched membranes (79 of 88) contained this type of channel.

In addition to the K_Ca channel, a lower-conductance channel was identified. TEA at low concentration did not inhibit the channel that was activated by both histamine and endothelin-1 (Figure 1). This channel was inhibited by an inhibitor of the delayed rectifier K⁺ (K_DR) channel (16), 4-aminopyridine (Figure 1A). As shown in Figure 2, the channel had a conductance of 21 ± 1 pS (n = 5), which is similar to that of the K_DR channel that has been reported in canine and porcine airway smooth-muscle cells (16). The I-V curve was constructed from the data for five channels, and three channels were discarded for the data analysis because of incomplete data (not all data was in the 40 to +60 mV range). The small-conductance K⁺ channel was observed in eight of 30 excised outside-out patches when stimulated by histamine and endothelin-1, although this incidence was significantly larger than the two of 30 patches in the nonstimulated condition in which this channel was found (P < 0.05). This suggested either a small population of these channels in the plasma membrane or the requirement of some cellular components for the channels' activity.

View larger version (15K): [in this window] [in a new window]   Figure 1.   Representative examples of (A) histamine- and (B) endothelin-1 (ET-1)-operated K+ channels of small conductance. The large-conductance K+ channel was abolished by the introduction of TEA to the bath in an outside-out patch (A). Histamine (10 µM) in addition to TEA activated a small-conductance channel that was blocked by 4-aminopyridine (4-AP, 5 mM) (A). Similarly, ET (100 nM) also activated the channel (B). The cell membrane potential was held at +60 mV. The external solution was 140 mM NaCl, 4.7 mM KCl, 1.13 mM MgCl2, 1.2 mM CaCl2, 10 mM glucose, 10 mM Hepes. The pipette solution (internal solution) was 140 mM KCl, 1.13 mM MgCl2, 10 mM Hepes, 10 mM glucose, 1 mM EGTA, and 1 mM Na2ATP. C and O represent closed and open states of channels, respectively.

View larger version (11K): [in this window] [in a new window]   Figure 2.   Current-voltage (I-V) curve of small-conductance K+ channels. The continuous line represents the regression line for all membrane potentials measured (y = 0.021× ± 0.95), indicating a conductance of 21 pS (n = 5) at a membrane potential of approximately 0 mV. Values given are means ± SE. Solutions are listed in Figure 1.

Membrane Potentials

The mean resting membrane potential was 48 mV in the whole-cell configuration, which was not significantly altered by differences in EGTA concentration or by the addition of 0.5 mM CaCl₂ and 1 mM EGTA to the pipette solution (i.e., by 0.5 mM EGTA [47 ± 3 mV, n = 15], 1.0 mM EGTA [47 ± 4 mV, n = 24], or the addition of CaCl₂ [49 ± 3 mV, n = 8]). Both TEA and ACh or MCh depolarized the membrane, and representative examples of this are shown in Figure 3. However, a definite difference in the membrane resistance was observed with the TEA- and MCh-induced depolarizations (i.e., in contrast to its increased membrane resistance in the presence of TEA [Figure 3A], ACh [or MCh] decreased the membrane resistance [Figure 3B], suggesting that MCh opens channels in the membrane). Furthermore, despite the continued stimulation by ACh or MCh, the depolarization reversed to one half the peak value with time, reaching a plateau potential, as shown in Figure 3B.

View larger version (16K): [in this window] [in a new window]   Figure 3.   Effects of TEA (A) and MCh (B) on membrane potential. Both 10 mM TEA and 10 µM MCh depolarized the membrane in a whole-cell configuration. Downward spikes in the records represent the resistance of the whole membrane. Hyperpolarizing currents of 40 pA and 20 pA, with a duration of 120 ms, were induced in A and B, respectively. Note the difference in change of resistance induced by the respective agents (i.e., TEA increased the membrane resistance, in contrast to a decline in the presence of MCh). The input resistance was measured by the steady-state change in potential induced by a hyperpolarizing current. Solutions are listed in Figure 1.

In contrast, both histamine and endothelin-1 produced a gradual hyperpolarization of the membrane, with the potential reaching a plateau, as shown in Figure 4. The histamine- or endothelin-1-induced hyperpolarization was accompanied by a decline in the membrane resistance, suggesting the activation of some channels in the membrane (Figure 4). Even when the membrane was treated with 10 mM TEA, histamine and endothelin-1 produced a gradual hyperpolarization, which was similar to that in the absence of TEA, as shown in Figure 5. The lack of TEA sensitivity is consistent with the hypothesis that K_DR or small-conductance K⁺ channels are involved in the membrane hyperpolarization induced by histamine and endothelin-1.

View larger version (15K): [in this window] [in a new window]   Figure 4.   Effect of histamine (A) and endothelin (ET)-1 (B) on membrane potential in the whole-cell configuration. Both histamine (10 µM) and endothelin-1 (100 nM) hyperpolarized the membrane, as shown in the figures. Hyperpolarizing currents of 20 pA and 10 pA were induced in A and B, respectively. Solutions are listed in Figure 1.

View larger version (23K): [in this window] [in a new window]   Figure 5.   Effect of histamine (A) and endothelin (ET)-1 (B) on membrane potential following depolarization by TEA in the whole-cell configuration. As shown in the figures, both histamine (10 µM) and endothelin-1 (100 nM) repolarized the membrane, which had been depolarized by 10 mM of TEA. Hyperpolarizing currents of 20 pA and 10 pA were induced in A and B, respectively. Solutions are listed in Figure 1.

The summarized data for the changes in membrane potential are shown in Figure 6. TEA (10 mM) and ACh (100 µM) significantly depolarized the membrane by +28 ± 4 mV (n = 12) and +21 ± 2 mV (n = 7), respectively (Figure 6). The TEA-induced depolarization was accompanied by a significant increase in the membrane resistance, whereas the ACh-induced depolarization was accompanied by a decrease in the membrane resistance (Figure 6). In contrast, histamine (10 µM) and endothelin-1 (100 nM) significantly hyperpolarized the membrane by 21 ± 6 mV (n = 8) and 15 ± 2 mV (n = 8), respectively (Figure 6). The histamine- and endothelin-1-induced hyperpolarizations were accompanied by significant decreases in the membrane resistance (Figure 6B). The histamine-induced hyperpolarization was similarly observed in the nystatin perforated-patch experiment. Here, histamine (10 µM) significantly hyperpolarized the resting membrane (potential = 46 ± 9 mV) by 19 ± 7 mV (P < 0.05, n = 4), with a decrease in the membrane resistance (98 ± 13 M) in the nystatin perforated-patch configuration (14).

View larger version (22K): [in this window] [in a new window]   Figure 6.   Summarized data of changes in membrane potential (A) and membrane resistance (B), expressed by  values from the prior baseline values. Both TEA (10 mM) and ACh (10 µM) significantly depolarized the membrane (A). TEA increased the membrane resistance, whereas ACh decreased it, as shown in B. Both histamine (10 µM) and endothelin-1 (100 nM) hyperpolarized the membrane (A), with decreases in membrane resistance (B). Values are means ± SE. ***P < 0.001; **P < 0.01; *P < 0.05 versus baseline values.

The membrane potentials when the smooth-muscle cells began to contract in response to histamine or endothelin-1 were much more negative than those in response to TEA or ACh. The cells contracted 23 ± 7 s and 29 ± 10 s, after stimulation by histamine (10 µM) and endothelin-1 (100 nM), showing membrane potentials of 59 ± 9 mV (n = 7) and 52 ± 10 mV (n = 6), respectively. In contrast, the cells contracted 10 ± 3 s and 15 ± 6 s, respectively, after stimulation by TEA (10 mM) and ACh (100 µM), showing membrane potentials of 18 ± 3 mV (n = 12) and 16 ± 8 mV (n = 7), respectively. In three samples, the cell contraction was not clearly observed, although the membrane potentials clearly changed, and these samples were discarded for this analysis.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

In the present study, the conductance of the predominant K_Ca channels in bovine tracheal smooth-muscle cells was 127 pS, which is similar to the values (120 to 155 pS) found in previous studies of guinea pig, rabbit, canine, and porcine airway smooth-muscle cells, done with experimental solutions similar to ours (i.e., a low K⁺ concentration in the outer cell-membrane solution [5 to 6 mEq/liter] and a high K⁺ concentration in the inner cell-membrane solution [126 to 140 mEq/liter]) (17). In experiments done with the same high K⁺ concentration in both the outer and inner cell-membrane solutions (135 to 150 mEq/liter), the conductance of the predominant K_Ca channels was reported to be 220 to 266 pS in bovine, canine, and porcine airway smooth-muscle cells (15, 21, 22). The experimental conditions may affect the absolute value of the ion-channel conductance. In addition, small-conductance K⁺ channels were observed in bovine tracheal smooth-muscle cells in the present experiments. The conductance, voltage- dependence, and 4-aminopyridine sensitivity of these K⁺ channels are similar to those of the K_DR channels (16), although the ionic selectivity or Ca²⁺ independence was not fully examined in the present study because of the rarity of the small-conductance channels.

In the present study with the patch-clamp method, both TEA and ACh (or MCh) depolarized the membrane of bovine tracheal smooth-muscle cells. It is possible that cholinergic muscarinic stimulation induces membrane depolarization, leading to the activation of dihydropyridine (DHP)-sensitive VDCCs or Ca²⁺ influx, and resulting in airway smooth-muscle contraction (11). However, there has been some controversy about the degree of depolarization of the membrane required for activation of VDCCs in airway smooth-muscle cells. Some investigators (10) have reported that VDCCs in airway smooth-muscle cells require potentials in excess of 20 mV for activation, whereas others (23) have found that ACh activates VDCCs with a membrane potential rarely exceeding 30 mV, which is close to that found in the present experiment. Therefore, it remains unresolved whether membrane depolarization plays a central role in the contraction of airway smooth muscle by cholinergic stimulation. Cholinergic stimulation has been shown to stimulate phosphoinositide metabolism, which is linked to Ca²⁺-signaling events (11); furthermore, in our experiments, membrane depolarization by ACh or MCh reversed with time, which contrasted with the depolarization induced by TEA. These findings suggest the presence of other mechanism(s), in addition to the activation of VDCCs, in ACh- or MCh-evoked contraction of airway smooth muscle.

In contrast to ACh- (or MCh)-induced membrane depolarization, both histamine and endothelin-1 induced membrane hyperpolarization for a few minutes after stimulation. Previous experiments with glass microelectrodes have shown that histamine induces a significant depolarization in pig (24), guinea pig (25), and canine airway smooth muscles (26). For example, Lee and colleagues (27), using a glass microelectrode method, have reported endothelin-induced depolarization of the membrane in ferret bronchial and tracheal smooth muscle. Furthermore, Souhrada and Souhrada (25) reported that the membrane potential showed a gradual depolarization that reached a maximum value 10 to 15 min after histamine exposure in guinea pig tracheal smooth muscle. Species differences seem unlikely to be the cause of the differences in changes induced in the membrane potential by histamine or endothelin, since Kirkpatrick, using microelectrode and sucrose gap techniques, reported membrane depolarization by histamine in bovine airway smooth-muscle cells (28). A possible explanation is the difference in methodology used in the experiments. The difference in changes in membrane potential induced by histamine and endothelin may have resulted from differences in the duration of observation. In the whole-cell patch-clamp method the membrane potential can be observed for only a few minutes. It seems possible that histamine or endothelin induces an initial hyperpolarization followed by a depolarization in the membrane of bovine tracheal smooth-muscle cells.

Various findings suggest the mechanism of membrane hyperpolarization by histamine or endothelin-1. There have been reports of the inhibitory effect of spasmogens, especially cholinergic agonists, on K⁺ channels of airway smooth-muscle cells. For example, Kume and Kotlikoff (29) found a direct inhibition of K_Ca channels via pertussis toxin-sensitive G-protein linked to the muscarinic receptor. Following this, Janssen and Sims (18) reported that ACh activates both nonselective cation and Cl conductances in guinea pig tracheal smooth-muscle cells, inducing depolarization accompanied by an increase in the membrane conductance. The depolarization with an increase in membrane conductance induced by cholinergic stimulation was also observed in the present experiments with bovine tracheal smooth-muscle cells. Janssen and Sims (8), using a whole-cell recording technique, found that histamine activated the K⁺ current in guinea pig tracheal smooth-muscle cells, and produced a transient Cl current with no contribution from nonselective cation conductance, and which returned to the basal level within a few seconds. We have found in our previous experiments (30) that both histamine and endothelin-1 evoke a marked increase in the outward K⁺ current, whereas ACh induces a sustained decrease after an initial transient increase in the K⁺ current in bovine tracheal smooth-muscle cells. It is therefore possible that the dominant and sustained outward K⁺ current, with a relatively small transient inward Cl current evoked by histamine, results in membrane hyperpolarization. Activation of small-conductance K⁺ channels has been suggested as playing a role in or causing the histamine- or endothelin-1-induced hyperpolarization, because a marked increase in the outward K⁺ current was observed with 10 mM TEA, which is known to eliminate K_Ca channels (15). However, the TEA concentration of 10 mM might not be adequate to eliminate K_Ca channels, especially when agonists induce increases in [Ca²⁺]_i. Consequently, it is possible that K⁺ channels other than small-conductance K⁺ channels play a partial role in histamine- or endothelin-1-induced hyperpolarization. Furthermore, there may be alternative explanations, which include the involvement of other TEA-insensitive K⁺ channels (i.e., ATP-dependent channels, inward rectifier K⁺ channels, and certain Ca²⁺-dependent K⁺ channels) (30) and/or the suppression of an ionic conductance.

However, as described previously, our findings, taken together with those of earlier studies using microelectrodes (24), show that it is likely that histamine and endothelin induce an initial hyperpolarization with late depolarization. It is unknown whether or how the initial hyperpolarization contributes to the contraction of airway smooth muscle. However, we can speculate on its role in such contraction as follows. Fluorescent dye studies with cultured human airway smooth-muscle cells indicate that histamine causes an influx of Ca²⁺ that might not involve VDCCs (32). Furthermore, Advenier and colleagues (4), in examining the effect of DHP on the contractile responses, speculated that in human airway smooth muscle, endothelin acts through not only DHP-sensitive Ca²⁺-channel activation, but also through other mechanisms, and particularly through a direct effect on [Ca²⁺]_i via activation of hydrolysis of phosphatidylinositol. Previous studies (3, 33) have shown that histamine and endothelin activate phosphatidylinositol hydrolysis. Further, it is possible that VDCCs are modulated by G-protein-coupled receptors, as has been shown in other tissues and cells (34). Taken together with our observation of membrane hyperpolarization, these findings are consistent with the idea that airway smooth-muscle contraction induced by histamine or endothelin is due to an increase in [Ca²⁺]_i that occurs through a pathway distinct from VDCCs. Besides coming from the release of Ca²⁺ from intracellular stores, an increase in [Ca²⁺]_i caused by these agonists may involve so-called "receptor-operated calcium channels" (ROCCs) (35). Previous studies (9, 36) found some differences between histamine- or endothelin-induced and ACh- or carbachol-induced contractile responses in airway smooth muscle (i.e., histamine- or endothelin-1-induced responses showed a gradual contraction, as compared with a rapid response to ACh). Generally, ROCCs in various other cells are considered to be related to a slower response than that mediated by VDCCs (37). Therefore, it is possible that the observations in the present study reflect a difference in the cellular mechanism of the contractile response induced by endothelin or histamine and that induced by cholinergic agonists.

In guinea pigs, sensitization and resensitization to allergens have been reported to cause a significant membrane hyperpolarization of airway smooth muscle, which is accompanied by airway hyperreactivity or hyperresponsiveness (25). Murray and associates (35) observed that membrane hyperpolarization caused a marked increase in [Ca²⁺]_i through ROCC activation after histamine exposure in human airway smooth muscle. Furthermore, induction of hyperpolarizing currents has been shown to increase agonist-induced contraction in canine trachealis muscle (10). Taken together with these observations, our finding of membrane hyperpolarization in airway smooth-muscle cells in response to histamine and endothelin-1 may be important for understanding airway hyperreactivity or hyperresponsiveness in bronchial asthma.