MATERIALS AND METHODS

Cell culture

Smooth muscle cell line (HTSMC), originally derived from human trachea, was obtained from Cell Applications Inc (San Diego, CA, USA).²⁰ Cells were maintained in smooth muscle cell growth medium (Cell Applications Inc) and equilibrated in a humidified atmosphere of 5% CO₂/95% air at 37°C. The cells were subcultured weekly after detachment by using culture medium containing 1% trypsin. The experiments were performed after the cells reached confluence (usually 5–7 days).

Electrophysiological measurements

Immediately before each experiment cells were dissociated and an aliquot of the cell suspension was transferred to a recording chamber mounted on the stage of an inverted microscope (Diaphot 200; Nikon, Tokyo, Japan). Cells were bathed at room temperature (20–25°C) in normal Tyrode's solution containing 1.8 mM CaCl₂. Patch pipettes were prepared from Kimax capillary tubes (Vineland, NJ, USA) using a vertical two step electrode puller (PP-83; Narishige, Tokyo, Japan) and the tips were fire polished with a microforge (MF-83; Narishige). The pipette had a resistance of 3–5 MΩ when immersed in normal Tyrode's solution. A programmable stimulator (SMP-311; Biologic, Claix, France) was used for the digital generation of rectangular or linear voltage pulses. Ionic currents were recorded in either whole cell or inside out configuration of the patch clamp technique by using a patch clamp amplifier (RK-400; Biologic).^21,22 All potentials were corrected for liquid junction potential, a value that develops at the tip of the pipette when the composition of the pipette solution was different from that of the bath.

Data recording and analysis

The signals consisting of voltage and current tracings were monitored with a digital storage oscilloscope (model 1602; Gould, Valley View, OH, USA) and an LCD projector (AV600; Delta, Taipei, Taiwan). The data were simultaneously recorded using a digital audiotape recorder (model ZA5ES; Sony, Tokyo, Japan). Currents were low pass filtered at 1 or 3 kHz. A Digidata 1320A interface (Axon Instruments, Foster City, CA, USA) was used for the analogue-to-digital/digital-to-analogue conversion. This interface device was connected to a Pentium III grade laptop computer (Slimnote VX₃; Lemel, Taipei, Taiwan) through a USB port and was then controlled with the aid of pCLAMP 8.02 software package (Axon Instruments). Ionic currents were analysed off line with the aid of the Clampfit subroutine in the pCLAMP software or the Origin 6.0 software (Microcal Software Inc, Northampton, MA, USA).

To calculate the percentage increase in BK_Ca channel activity resulting from magnolol, the potential was held at the level of +30 mV and the bath solution contained 0.1 μM Ca²⁺. The probability of channel openings measured after application of magnolol (30 μM) was taken as 100%. The concentration dependent relation of magnolol on channel activity was fitted to a Hill function:where [C] is the concentration of magnolol, EC₅₀ and n are the half maximal concentration of magnolol required to increase the probability of channel openings and the Hill coefficient, respectively, and E_max is the magnolol induced maximal increase in channel activity. The EC₅₀ value of magnolol and its 95% confidence limits were calculated with the Origin program (Microcal Software Inc) using non-linear regression methods to obtain the best fit dose-response curve.

Single channel currents of the BK_Ca channel were analysed with Fetchan and Pstat subroutines in the pCLAMP software (Axon Instruments). Multigaussian adjustments of the amplitude distributions between channels were used to determine single channel currents. Functional independence between channels was then verified by comparing the observed stationary probabilities with the values calculated according to the binomial law.

To determine the effect of magnolol on the activation curve of BK_Ca channels, the ramp pulses from 0 to +80 mV with a duration of 1 second were delivered from the Clampex subroutine of pCLAMP software. The activation curve was calculated by averaging current traces in response to 20 voltage ramps and dividing each point of the mean current by the unitary amplitude of each potential after the leakage component was corrected.^22,23 The number of active channels in the patch (N) was counted at the end of each experiment when the solution with 100 μM Ca²⁺ was perfused This number was used to normalise the probability of channel openings at each potential. The activation curves obtained before and after the addition of magnolol were fitted with Boltzmann function of the form:where n is the maximal relative N • P₀, b is the slope factor of the voltage dependent activation (that is, the change in the potential required to result in an e-fold increase in the activation), and a is the voltage at which there is half maximal activation.

Open or closed lifetime distributions were fitted with logarithmically scaled bin width using the method of McManus et al.^22,24 To estimate all transition rates between states, single channel data were idealised and converted to an ASCII format—that is, a dwell time file format—using the SKM program of the QuB software suite that was downloaded from the web site (http://www.qub.buffalo.edu). This type of format was readily read by the maximum interval likelihood (MIL) program in the QuB suite.^25,26 The data were then used to determine single channel kinetic parameters by means of a maximum likelihood algorithm.^25,26 The highest log likelihood for the observed activity of BK_Ca channels in this study was obtained with the gating scheme: C₁ ↔ C₂ ↔ O. The single channel data were modelled on the basis of this simplified scheme. Simulated single channel data were also obtained using the determined transition rates and SIMU program. K and L values represent the equilibrium dissociation constant that equals the transition rate constant for C₁ ← C₂ divided by that for C₁ → C₂, and the equilibrium gating constant that equals opening rate constant divided by closing rate constant, respectively. This type of kinetic analysis was based on the assumption that the channel can be described by a finite state Markovian model and that the channels in the patch are mutually independent.

All values are reported as mean (SE). The paired or unpaired Student's t test and one way ANOVA with the least significance difference method for multiple comparisons were used for the statistical evaluation of differences in the mean values. Differences between values were considered significant at a value of p<0.05.

Drugs and solutions

Magnolol (5,5′-di-2-propenyl-1,1′-biphenyl-2,-2′-dol), an amphipathic molecule, is composed of two phenol rings with one polar hydroxyl and one non-polar hydrocarbon domain on each phenol ring (fig 1).²⁷ Because of its unique property, magnolol can freely permeate through cell membranes. Niflumic acid was purchased from Sigma Chemical (St Louis, MO, USA); 18α-glycyrrhetinic acid and baicalein were obtained from Aldrich Chemical (Milwaukee, WI, USA); iberiotoxin, glibenclamide, apamin, and sodium nitroprusside were obtained from Research Biochemicals (Natick, MA, USA); paxilline was purchased from Alomone (Jerusalem, Israel); and magnolol was a kind gift of Professor Chien-Chich Chen, National Institute of Chinese Medicine, Taipei City, Taiwan. All other chemicals were commercially available and of reagent grade.

The composition of normal Tyrode's solution was 136.5 mM NaCl, 5.4 mM KCl, 1.8 mM CaCl₂, 0.53 mM MgCl₂, 5.5 mM glucose, and 5.5 mM HEPES NaOH buffer, pH 7.4. To record K⁺ currents the patch pipette was filled with a solution consisting of 140 mM KCl, 1 mM MgCl₂, 3 mM Na₂ATP, 0.1 mM Na₂GTP, 0.1 mM EGTA, and 5 mM HEPES KOH buffer, pH 7.2. In the single channel recording, high K⁺ bathing solution contained 145 mM KCl, 0.53 mM MgCl₂, and 5 mM HEPES KOH, pH 7.4, and the pipette solution contained 145 mM KCl, 2 mM MgCl₂, and 5 mM HEPES KOH, pH 7.2. The value of free Ca²⁺ concentration was calculated assuming a dissociation constant for EGTA and Ca²⁺ (at pH 7.2) of 0.1 μM.²⁸

RESULTS

Effect of magnolol on whole cell K⁺ outward current (I_K) in cultured smooth muscle cells of human trachea

In the patch clamp technique the whole cell configuration was used to investigate the effect of magnolol on macroscopic ionic currents in these cells. The cells were bathed in normal Tyrode's solution containing 1.8 mM CaCl₂ and the pipette solution contained a low concentration (0.1 mM) of EGTA and 3 mM ATP. As shown in fig 2, when the cell was held at –40 mV and various potentials ranging from –50 to +50 mV with 20 mV increments were applied, a family of K⁺ outward currents were elicited and the current/voltage (I/V) relationship for these currents had an outward rectification. Within 1 minute of exposing cells to magnolol (3 μM) the amplitude of the outward currents was greatly increased throughout the entire range of voltage clamp steps. For example, when the depolarising pulses from –40 to +50 mV were applied, magnolol (3 μM) significantly increased current amplitude from 1105 (150) pA to 2794 (280) pA (n=8). This stimulatory effect was readily reversed when magnolol was washed out. Figure 2B shows the mean I/V relationships for these currents in the absence and presence of magnolol (3 μM).

To examine the nature of I_K stimulated by magnolol, the values of reversal potential obtained during cell exposures to magnolol (3 μM) were measured at different concentrations of extracellular K⁺. The data from each cell were then pooled and plotted as a function of extracellular K⁺. The line of best fit through the mean data revealed a slope of 58 mV per tenfold increase in extracellular K⁺. The outward current measured after application of magnolol therefore indicates that the K⁺ selective channel exhibits Nernstian behaviour.

The effects of various K⁺ channel blockers including glibenclamide, iberiotoxin, paxilline, and apamin on magnolol stimulated I_K in these cells were also examined and compared. As shown in fig 3, glibenclamide (10 μM) did not affect the magnolol mediated increase in the amplitude of I_K significantly. However, both iberiotoxin (200 nM) and paxilline (1 μM) was able to suppress the increase in I_K caused by magnolol (10 μM). Paxilline, an indole alkaloid isolated from Penicillium paxilli, was reported to be a potent inhibitor of BK_Ca channels.²⁹ These results indicate that the observed increase in outward current by magnolol is due to its stimulation of I_K which is sensitive to inhibition by iberiotoxin or paxilline.

Stimulatory effect of magnolol on BK_Ca channel activity in cultured smooth muscle cells of human trachea

Since macroscopic I_K stimulated by the presence of magnolol appears to be a large noisy voltage dependent, Ca²⁺ sensitive K⁺ current, the next series of experiments were designed to investigate whether magnolol interacts with the BK_Ca channel to stimulate Ca²⁺ activated K⁺ currents. Unless otherwise specified, all experiments of single channel recordings performed in this study were conducted with a symmetrical K⁺ concentration (145 mM) and an inside out configuration, and the bath medium contained 0.1 μM Ca²⁺. To provide 0.1 μM free Ca²⁺ in the bath solution, 0.5 mM CaCl₂ and 1 mM EGTA were added.²⁸ As shown in fig 4A, when the potential was held at +30 mV, application of magnolol (10 μM) to the cytosolic surface of an inside out patch greatly enhanced single channel activity which occurred in rapid open-close transitions. When membrane patches were exposed to magnolol (10 μM) the opening probability of BK_Ca channels was significantly increased to 0.622 (0.022) from a control of 0.013 (0.005) (n=8). Channel activity was reduced to 0.018 (0.007) (n=6) after washout of magnolol. However, there was no significant difference in the amplitude of single channel current between the absence and presence of magnolol (10 μM; fig 4C).

The relationship between the concentration of magnolol and the opening probability of BK_Ca channels was further examined using a holding potential of +30 mV. As shown in fig 4D, magnolol (0.1–30 μM) effectively increased the probability of channel openings in a concentration-dependent manner. The EC₅₀ value for magnolol stimulated channel activity was 1.5 μM. In addition, the Hill coefficient was found to be 2.2, suggesting that there was positive cooperation for the stimulation of BK_Ca channels.

Lack of effect of magnolol on single channel conductance of BK_Ca channels

The effect of magnolol on the activity of BK_Ca channels at various membrane potentials was also examined (fig 5). The plots of single channel amplitude as a function of holding potential were constructed. Figure 5B illustrates the I/V relationship of BK_Ca channels in the absence and presence of magnolol (10 μM). The single channel conductance of BK_Ca channels calculated from a linear I/V relationship in control solutions (in the absence of magnolol) was 196 (8) pS (n=12) with a reversal potential of 0 (3) mV (n=12). This value for single channel conductance of BK_Ca channels was similar to that reported previously in tracheal smooth muscle cells,³⁰ but was not significantly different from that measured after application of magnolol (197 (8) pS; n=10). Thus, it is clear that the presence of magnolol effectively enhances the activity of BK_Ca channels but produces no significant change in single channel conductance in these cells.

Effect of magnolol on voltage dependent activation of BK_Ca channels from cultured smooth muscle cells of human trachea

Figure 6A shows the activation curves of BK_Ca channels in the absence and presence of 1 and 3 μM magnolol. In these experiments the activation curve of BK_Ca channels was obtained with the aid of voltage ramp protocols.^22,23 The ramp pulses were applied from 0 to +80 mV with a duration of 1 second. The plots of the probability of channel openings as a function of membrane potential were constructed and fitted with the Boltzmann function as described in the Methods section. In the control n=0.37 (0.04), a=49.1 (1.5) mV, and b=9.7 (0.6) mV (n=6); in the presence of 1 and 3 μM magnolol n=0.55 (0.05) and 1.02 (0.08), a=43.5 (1.1) and 35.2 (1.6) mV, and b=9.6 (0.7) and 9.5 (0.7) mV, respectively (n=6). Thus, application of magnolol not only produced an increase in the maximal opening probability of BK_Ca channels, but also significantly shifted the activation curve to the less positive potential. In contrast, magnolol had no effect on the slope (b value) of the activation curve. These data clearly indicate that the magnolol induced change in the activity of BK_Ca channels in these tracheal smooth muscle cells is dependent on the level of membrane potential.

Effect of internal Ca²⁺ concentration on magnolol stimulated activity of BK_Ca channels in cultured smooth muscle cells of human trachea

To determine whether the magnolol induced increase in channel activity involves the level of internal Ca²⁺ concentration, another series of experiments was conducted with various concentrations of Ca²⁺ in the bath and the probability of channel openings before and after application of magnolol was examined. As shown in fig 6B, at a given concentration of magnolol (3 μM) the magnitude of the magnolol induced increase in channel activity was increased as the internal Ca²⁺ concentration was increased. However, the observed effect of magnolol on channel activity was unaffected by a change in the internal Ca²⁺ concentration. At a holding potential of +30 mV the presence of magnolol (3 μM) at internal Ca²⁺ concentrations of 0.01, 0.1, and 1 μM enhanced channel activity to a similar extent (about a fourfold increase).

Effect of magnolol on kinetic behaviour of BK_Ca channels

Because magnolol produced no change in single channel conductance, the effect of magnolol on the gating of these channels was also analysed. As shown in fig 7, in a detached patch of control cells—that is, in the absence of magnolol—open and closed time histograms at the level of +30 mV can be fitted by a one or two exponential curve. In addition, the time constant for the open time histogram was 8.9 (1.2) ms, whereas those for the fast and slow components of the closed time histogram were 43.4 (4.5) and 721.0 (13.5) ms, respectively (n=7). Magnolol (3 μM) added to the bath increased the time constant of the open state to 32.2 (4.3) ms (n=6) and decreased the mean closed times to 16.6 (3.2) and 419.6 (10.8) ms (n = 6). Thus, magnolol produced an increase in channel open time and a decrease in channel closed time. The presence of magnolol may cause the channel to spend more time in the open state with a longer open time. Changes in mean open and closed times caused by magnolol might explain its stimulatory effect on the activity of BK_Ca channels.

In order to obtain more information on the channel kinetics, single channel currents in the absence and presence of magnolol were idealised and modelled using the kinetic scheme described in the Methods section. This gating scheme is composed of one open and two closed states. Transition rates between states were derived from the maximum likelihood estimation.^25,26,29 Figure 8 shows examples of simulated single channel data that were generated using the transition rates of BK_Ca channels in the absence and presence of magnolol. The equilibrium dissociation constants (K values) obtained in the absence and presence of magnolol (10 μM) were estimated to be 0.44 (0.03) and 1.37 (0.04), respectively (n=6), a difference of about threefold. However, the equilibrium gating constants (L values) in the absence and presence of magnolol (10 μM) were 0.184 (0.003) and 3.09 (0.005), respectively (n=6), which differ by about 17-fold. These results suggest that the difference in the gating constant in the absence and presence of magnolol is greater than in the dissociation constant.

Comparison between effect of magnolol and those of baicalein, glycyrrhetinic acid, niflumic acid and Evans blue

The effects of magnolol, baicalein, glycyrrhetinic acid, glibenclamide, and paxilline on the activity of BK_Ca channels in cultured smooth muscle cells of human trachea were also examined and compared. As shown in fig 9, neither baicalein (10 μM) nor glycyrrhetinic acid (10 μM) applied intracellularly had any significant effect on the probability of channel openings. These compounds are known to be other phenolic compounds contained in Saiboko-to, a herbal medicine used to treat bronchial asthma.^1,13,16 However, both niflumic acid (30 μM) and Evans blue (30 μM) enhanced the channel activity significantly. These two compounds have previously been reported to stimulate BK_Ca channel activity in vascular endothelial cells.^29,31 In addition, paxilline effectively suppressed the increased activity of BK_Ca channels caused by magnolol which is consistent with the finding in whole cell current recordings (fig 3). Thus, it is clear that magnolol is more potent than niflumic acid or Evans blue in stimulating the activity of BK_Ca channels in these cells.

DISCUSSION

The major findings in the present study are as follows: (a) in cultured smooth muscle cells of human trachea magnolol can reversibly increase the amplitude of K⁺ outward current; (b) magnolol stimulates the activity of BK_Ca channels in a concentration and voltage dependent manner but does not modify single channel conductance of these channels; (c) the magnolol induced increase in the probability of channel openings is accompanied by an increase of the time constant in the open state and a decrease in the time constant in the closed state; and (d) magnolol can modify both dissociation and gating constants of these channels. The stimulatory effect of this compound on BK_Ca channel activity would be expected to affect cell excitability if similar results are present in airway myocytes in vivo.^18,19,30

The single channel conductance of BK_Ca channels measured with the use of a symmetrical K⁺ concentration (145 mM) on both sides of the membrane was 196 (8) pS (n=12). This value is similar to those of typical BK_Ca channels reported previously in swine tracheal smooth myocytes.³⁰ It was also found that, in these cells, the channel activity stimulated by magnolol is sensitive to inhibition by paxilline. However, the density of BK_Ca channels reported here in cultured smooth muscle cells of human trachea appears to be lower than that in freshly dissociated airway myocytes, since we did not clearly observe a unique type of spontaneous transient outward K⁺ current as shown previously.^30,32 On the other hand, because there is a lower density of BK_Ca channels, single channel properties may be easily identified and examined in these smooth muscle cells of human trachea.

It seems likely that magnolol did not exert its effect on BK_Ca channel activity via an increase in the affinity of Ca²⁺ ions or the Ca²⁺ binding site in the membrane because magnolol stimulated channel activity was found to be independent of internal Ca²⁺. The EC₅₀ value of magnolol required for the stimulation of BK_Ca channels was 1.5 μM, a value that is lower than those used to produce the antioxidative actions.^5–8 BK_Ca channels expressed in airway smooth myocytes may therefore be a relevant “target” for the action of magnolol. Regardless of the mechanism of its actions, findings from the present study demonstrating the stimulatory effect of magnolol on BK_Ca channels should be noted with caution in relation to its increasing use as an antioxidant agent.^4,6–8

Previous reports have shown that magnolol can reduce the level of intracellular cyclic AMP in rat neutrophils.² It was believed that the magnolol mediated inhibition of the respiratory burst in formylmethionyl-leucyl-phenylalanine activated neutrophils appeared to involve the attenuation of tyrosine kinase phosphorylation or mitogen activated protein kinase activation.² The inhibition of phorbol ester induced neutrophil aggregation by magnolol is found to be associated with the inhibition of the activity of protein kinase C.³³ Previous studies have also shown that magnolol can stimulate Ca²⁺ release from internal stores and Ca²⁺ influx across plasma membrane in rat neutrophils.³⁴ However, in our experimental conditions, the effect of magnolol on ionic currents was observed within only 2 minutes of application of this compound. Furthermore, in an inside out configuration magnolol added to the bath is effective in stimulating the activity of BK_Ca channels. It is thus possible that the magnolol stimulated increase in BK_Ca channel activity in these cells is direct and does not require cytosolic diffusible substances inside the cells. Magnolol mediated increases in the activity of BK_Ca channels observed in these airway smooth muscle cells are also unlikely to result from an increase in intracellular Ca²⁺ that is induced by Ca²⁺ release from internal stores, Ca²⁺ entry from the cell exterior, or both.³⁵

It seems to be uncertain whether the presence of magnolol affects the production of nitric oxide. A previous report suggested that the antiarrhythmic effect of magnolol during acute coronary occlusion might involve an increase in the activity of nitric oxide synthase.³⁶ On the other hand, in lipopolysaccharide activated macrophages magnolol was found to suppress the production of nitric oxide.³ However, in our study, when tracheal smooth muscle cells were preincubated with sodium nitroprusside, magnolol mediated stimulation of outward currents was not significantly affected (data not shown). The magnolol induced effect on the generation of nitric oxide does not therefore appear to be primarily responsible for the stimulation of BK_Ca channels observed in the present study.

In addition to magnolol, both baicalein and glycyrrhetinic acid are contained in Saiboku-to which has been found to possess antiasthmatic activity.^1,13 However, in our study neither baicalein nor glycyrrhetinic acid affected the activity of BK_Ca channels. Magnolol thus appears to be an important ingredient present in Saiboko-to that can effectively stimulate BK_Ca channels. Moreover, it will be of interest to determine whether magnolol contained in Saiboko-to¹ decreases the release of leukotrienes through a mechanism of stimulatory action on BK_Ca channel activity.¹³

The increase in I_K caused by magnolol is apparently not a result of an increase in single channel amplitude of BK_Ca channels because no significant difference was found in single channel conductance in the absence or presence of magnolol. However, this compound increased the activity of BK_Ca channels in tracheal smooth muscle cells in a voltage dependent fashion and possibly by acting at a site that is accessible from the intracellular side of the channel. We also found that magnolol produced an increase in mean open time and a decrease in mean closed time. These findings can account for the increased probability of channel openings. The concentration-response curve of channel activation by magnolol had a Hill coefficient of 2.2, suggesting that more than one magnolol molecule may bind to the channel, although the stoichiometry of magnolol binding to the channel is currently unknown. More interestingly, on the basis of the simple gating scheme described in the Methods section, the mechanism of action of magnolol on BK_Ca channel activity was also deduced. The results showed that the stimulatory actions of magnolol appeared to lie not only in the binding step, but also in the gating of channels because both the equilibrium dissociation constant and the gating constant were significantly altered after application of magnolol. However, we cannot rule out the possibility that the binding of magnolol shifts the channel to longer open state(s). In addition, how these binding events are functionally coupled to the opening of channel pores has yet to be elucidated.

In summary, the data reported here provide evidence that, in addition to stimulating steroidogenesis in adrenal cells,³⁷ magnolol is potent in stimulating the activity of BK_Ca channels in cultured smooth muscle cells of human trachea. The experiments did not determine whether the modulatory effect on the BK_Ca channel is the primary mechanism by which magnolol reduces asthmatic symptoms or whether it is secondary to effects mediated through other cell types. However, the present finding will be of great help in the study of the underlying mechanisms through which magnolol interacts with BK_Ca channels expressed in smooth muscle cells. The potent effect of magnolol on the stimulation of BK_Ca channels seems to make it a promising compound to induce relaxation of airway smooth muscle.