Departments of Internal Medicine III and Laboratory Medicine, Kyoto
Prefectural University of Medicine, Kawaramachi-Hirokoji, Kamigyo-ku,
Kyoto 602-0841, Japan
We examined the blocking effects of terfenadine, an antihistaminic
agent, on the ATP-sensitive K+ current (IK,ATP)
in rabbit ventricular cells. IK,ATP was induced by
cromakalim or NaCN. Terfenadine blocked the IK,ATP with an IC50 of 1.7 µM at 
10 mV. This blockage was voltage
dependent; depolarization induced a stronger blockage. According to the
transmembrane electrical field model, terfenadine interacts with the
site located 15 to 18% from the cytoplasmic membrane surface. In line
with the assumption that the binding site is near the cytoplasmic
surface, terfenadine applied to the cytoplasmic solution potently
inhibited the single-channel activity for IK,ATP in the
inside-out configuration (IC50 0.19 µM). In contrast,
terfenadine applied to the external solution did not affect the channel
activity in the cell-attached configuration, but inhibited it when
applied into the pipette. The inhibition of the single channels by
terfenadine was accompanied by flickering of the channels. These
findings suggest that 1) terfenadine blocks the ATP-sensitive
K+ channel in the open state, 2) the binding site is near
the internal membrane surface and 3) terfenadine is poorly diffusible
into the lipid biomembrane and accesses the binding site via the
hydrophilic pathway. Terfenadine also inhibited the transient outward
K+ current, inward rectifier K+ current and
E4031-sensitive rectifier K+ current. However, the
inhibition of these repolarization currents by terfenadine at 1 µM
was not sufficient to prolong the action potential duration
significantly. Whereas, terfenadine (1 µM) prolonged the action
potential duration which had been shortened by cromakalim. Terfenadine
may modify the ischemia-induced arrhythmias by blocking
IK,ATP.
 |
Introduction |
Numerous
antihistaminic agents are used for the treatment of allergic diseases,
but their use is restricted because of their adverse effects on the
central nervous system; somnolence and diminished alertness (Simons and
Simons, 1994
). Agents showing only slight effects on the central
nervous system have been developed as second-generation antihistaminic
agents (Wiech and Martin, 1982; Rose et al., 1982).
Terfenadine is the most widely used one of these agents. However, it
has been found that terfenadine exerts cardiovascular actions, which
include life-threatening tachyarrhythmias (Monaham et al.,
1990; Woosley et al., 1993). Because the prolongation of QT
interval on the electrocardiogram precedes these arrhythmias (Jackman
et al., 1984), much attention has been paid to the effects
of terfenadine on the K+ currents that cause the action
potential repolarization of the myocardium. Experimental studies have
shown that terfenadine inhibits multiple cardiac K+
currents such as the delayed rectifiers (Woosley et al.,
1993; Rampe et al., 1993; Salata et al., 1995;
Crumb et al., 1995; Yang et al., 1995; Roy
et al., 1996), ITO (Crumb et al.,
1995; Berul and Morad, 1995) and IK1 (Berul and Morad,
1995; Salata et al., 1995).
Under pathological conditions, a decrease in the intracellular ATP
concentration induces another K+ current, the
IK,ATP, which carries a large outward current within the
plateau potential range of the ventricular action potentials (Noma,
1983; Nichols and Lederer, 1991). Activation of this current shortens
the APD to protect the energy consumption, and facilitates the ischemic
preconditioning (Parret, 1994). The K+ efflux through the
IK,ATP causes extracellular K+ accumulation
during acute ischemia (Wilde and Janse, 1994). Despite these important
roles of IK,ATP under pathological conditions, the effects
of antihistaminic agents on IK,ATP have not been clarified so far. It has been known that terfenadine is more likely to cause tachyarrhythmias in patients with underlying heart diseases; ischemic heart disease or congestive heart failure (Woosley et al.,
1993). Therefore, modulation of IK,ATP by terfenadine may
play some role in the adverse cardiovascular effects of this agent. The
aim of this study was to elucidate the effects of terfenadine on
IK,ATP, and to give additional insight to the mechanisms
underlying the terfenadine-induced arrhythmias.
| |
Materials and Methods |
Cell isolation.
Single cells were isolated as described
previously (Habuchi et al., 1996). Briefly, the heart
excised from rabbits (2-2.5 kg) was retrogradely perfused with
Ca++-free, phosphate-buffered solution containing (in mM):
NaCl, 142; KCl, 5.4; MgCl2, 1.0;
NaH2PO4, 0.33; Na2HPO4,
2.24 and glucose, 10 (pH 7.4). The solution was then switched to one
containing 0.02 mg/ml collagenase (Yakult, Tokyo, Japan) and 0.01 mg/ml
protease (Type 14, Sigma Chemical Co., St. Louis, MO). The temperature was 37°C and the solutions were bubbled with 100% O2.
After 10-min superfusion with the enzyme solution, the right
ventricular free wall was cut into small pieces. We did not use the
left ventricle because the cell isolation from the left ventricle
sometimes resulted in a poor yield of single cells, and because the
left ventricle may have large regional differences in the electrical
properties (Antzelevitch et al., 1991; Fedida and Giles,
1991). The pieces were stirred in the second enzyme solution bubbled
with 100% O2 at 37°C. The second enzyme solution
contained 1 mg/ml collagenase (Sigma, type H). The supernatant was
collected every 5 min. After centrifugation at 70 × g
for 1 min, isolated cells were placed in the stock solution at 4°C
with 0.1% bovine serum albumin (Sigma, fraction V). The stock solution
contained (in mM): K-glutamate, 90; oxalate, 10; KCl, 25;
KH2PO4, 10; MgSO4, 1; taurine, 10;
EGTA, 0.5; HEPES, 5 and glucose, 10 (pH 7.2 adjusted with KOH).
For the single channel recordings in cell-attached configuration, the
cells were loaded with NaCN to reduce the intracellular ATP storage.
NaCN (10 mM) was added to the stock solution (4°C) 2 to 4 hr before
the use of the cells.
Electrical measurements.
All the experiments were carried
out at 37°C. The amplifier used was an Axopatch-1D (Axon Instruments,
Foster, CA) or TM-1000 (ACT ME, Tokyo, Japan). For the whole-cell
current measurement, cells were superfused with glucose-free Tyrode
solution containing (in mM): NaCl, 140; KCl, 5.4; CaCl2,
1.0; MgCl2, 1.0 and HEPES, 5 (pH 7.4 adjusted with HCl),
and the pipette solution contained (in mM): K-aspartate, 110; KCl, 20;
CaCl2, 1.0; MgCl2, 1.0; EGTA, 10;
K2-ATP, 0.4; Na3-GTP, 0.2 and HEPES, 5 (pH 7.2 adjusted with KOH). The pipette used had a tip resistance between 2 and
2.5 M
. A liquid junction potential of 10 mV was corrected, and the series resistance was compensated to minimize the capacitive surge in
response to 5 mV step repolarization. The cell capacitance was measured
by digitally integrating the capacitive surge. The cells used had a
membrane capacitance of 76 ± 23 pF (n = 110, mean ± S.D.). IK,ATP was induced by application of
cromakalim (a K+ channel opener) at 10 µM or NaCN at 10 mM. For the induction of IK,ATP by NaCN, the cells in the
recording chamber were pretreated with 1 mM NaCN-containing,
glucose-free Tyrode solution for 10 to 20 min at 37°C. After the cell
was clamped with the 0.4 mM ATP containing solution, the concentration
of NaCN was increased to 10 mM, which usually caused a rapid activation
of IK,ATP (ref. fig. 1A). The
addition of 10 mM NaCN to the external solution yielded a change in the
liquid junction potential of 0.56 mV (n = 8). This
small change was not corrected.
View larger version (21K):
[in this window]
[in a new window]
|
Fig. 1.
Effects of terfenadine on IK,ATP. A,
Representative record of the holding current at 10 mV. The pipette
solution contained 0.4 mM ATP. Horizontal bars indicate the period of
extracellular application of the drugs. The concentrations of
cromakalim, terfenadine and glibenclamide were 10, 10 and 1 µM,
respectively. NaCN was first applied at a concentration of 1 mM. After
the whole-cell configuration was established, the concentration of NaCN
was increased to 10 mM. The dotted line indicates the baseline current
level obtained by the application of glibenclamide. B,
Concentration-dependence of the inhibition of IK,ATP by
terfenadine. The IK,ATP in the presence of the drug
relative to that immediately before the drug application is plotted
against the drug concentration. IK,ATP was measured as the
glibenclamide-sensitive current. Parentheses indicate the number of
cells (#, data for the NaCN-induced IK,ATP). A single
concentration of the drug was tested on one cell.  , Data for the
cromakalim-induced IK,ATP.  , Data for the NaCN-induced
IK,ATP. The curves are the fit to equation (2) given in the
text. IC50 = 1.7 µM.
|
|
The effects of terfenadine on IK,ATP were examined
regarding either the holding current (10 mV) or the
quasi-steady-state current-voltage (I-V) relationship. The
quasi-steady-state current was recorded using a ramp-pulse between
130 and +30 mV at a rate of 160 mV/sec in the presence of
nifedipine (1 µM). At the end of all these experiments glibenclamide
(1 µM) was perfused, and IK,ATP was measured as the
glibenclamide-sensitive current.
Single-channel recording was conducted in both the cell-attached and
inside-out patch configurations (Hamill et al., 1981). Pipettes having a relatively high resistance (around 10 M) were used
to record as small a number of channels in the patch as possible. The
pipette was heat-polished immediately before use, with the shank coated
with Sylgard (Dow Corning Co., Midland, MI). The pipette solution
contained (in mM): NaCl, 140; KCl, 5.4 and HEPES, 5 (pH 7.2 adjusted
with KOH). In the cell-attached mode, the activation of
KATP channels was maintained by applying NaCN and
cromakalim to the NaCN-loaded cells (ref. fig. 5); the NaCN-loaded
cells were perfused with the bathing solution containing (in mM): KCl, 135; MgCl2, 2.0; EGTA, 5; NaCN, 10; cromakalim, 0.01 and
HEPES, 5 (pH 7.4 adjusted with KOH). In the inside-out mode, the
excised patch from the non-loaded cell was exposed to a bathing
solution containing (in mM): KCl, 135; MgCl2, 1.0; EGTA, 5;
NaCN, 10; HEPES, 5; ATP, 0.02 and ADP, 0.1 (pH 7.2 adjusted with KOH).
The effects of terfenadine on other repolarization K+
currents (ITO and IK1) and on the action
potentials were also evaluated. The solutions were sodium-free for the
measurement of ITO. The external solution contained (in
mM): choline Cl, 140; KCl, 5.4; CaCl2, 1.8;
MgCl2, 1.0; CdCl2, 0.1; HEPES, 5; glucose, 10 and atropine, 0.001 (pH 7.4 adjusted with Tris base). The pipette solution had the same composition as that used for the measurement of
IK,ATP in whole cells, but the ATP concentration was
increased to 5 mM (K-aspartate = 90 mM). ITO was
elicited by a step depolarization from a holding potential of 80 to
+20 mV for 350 ms (0.1 Hz). The same pipette solution was used for
recording IK1 and the action potentials. The external
solution was normal Tyrode. IK1 was measured with the ramp
pulse described above. Action potentials were elicited in the current
clamp mode by applying suprathreshold current pulses of 3-msec duration
at a rate of 0.1 Hz. The APD was measured between the action potential
upstroke and the time at which the repolarizing membrane potential
crossed 60 mV.
Data acquisition and analysis.
The frequency of the low-pass
filter (E-3201B, NF Electrics, Osaka, Japan) was set at 0.5 KHz for
IK,ATP and IK1 in the whole-cell recording and
5 KHz for ITO. The whole-cell data digitized on a digital
oscilloscope (Nicolet 310C, Madison, WI) were subsequently analyzed on
a computer (NEC 98, Tokyo, Japan). In most of the single-channel
experiments, the current data were filtered at 2 KHz, and fed into
pClamp system via DIGIDATA 1200 interface (Axon Instruments, Foster,
CA) with a sampling frequency of 1 KHz. The half amplitude threshold
method was used (Coquhoun and Sigworth, 1983), and the probability of
the channel opening (PO) was obtained based on the
following equation:
|
(1)
|
where I is the time-averaged current carried by the
KATP channels in the patch for a certain period. N and i
are the number of functioning channels and the unitary current
amplitude, respectively. The 0-current level was determined by applying
1 µM glibenclamide at the end of each experiment. The time-averaged
current (I) was measured every 15 sec.
For the analyses of the opening and closing kinetics of the
KATP channel, the patches which exhibited only one-channel
activity were used (see fig. 4). The inside-out configuration was made after the one-channel activity was confirmed in the cell-attached recording on a NaCN-loaded cell. The recording was then made during perfusion with the control bathing solution and after the application of terfenadine. When any plural-channel opening was observed during this series of experiment, the data were discarded. The filtering and
sampling frequencies were set at 5 KHz. The open- and closed-time distributions were analyzed from a continuous recording for 1 min in
the absence and presence of terfenadine.
The data are presented as means ± S.E., unless otherwise
specified. Student's t test was used for statistical
analysis and P < .05 were considered to be significant.
Drugs and chemicals.
Terfenadine, glibenclamide, cromakalim,
nifedipine, Na2-ATP, K2-ATP, ADP,
Na2-phosphocreatine and Na3-GTP were purchased
from Sigma. Cromakalim provided from Taisho Pharmaceutical Co. (Tokyo, Japan) was also used. E4031 was a gift from Eisai Pharmaceutical Co.
(Tokyo, Japan). All other chemicals were from Wako Pure Chemicals (Osaka, Japan). Terfenadine, glibenclamide and cromakalim were dissolved in dimethylsulfoxide as a 10 or 100 mM stock solution.
| |
Results |
Terfenadine blockage of IK,ATP.
Figure 1 shows
representative effects of terfenadine on IK,ATP. In figure
1A, the exposure of the cell to cromakalim (10 µM) or NaCN (5 mM)
induced an outward shift in the holding current at 10 mV, which
subsequently decayed and reached a plateau. This development of the
outward current was completely abolished by glibenclamide at 1 µM
(not shown), indicating that these changes in the holding current
represented the activation and subsequent run-down of
IK,ATP. Terfenadine at a selected concentration and glibenclamide at 1 µM were applied sequentially during the plateau phase. Figure 1A represents that terfenadine at 10 µM blocked both
the cromakalim- and NaCN-induced IK,ATP nearly completely. As shown in figure 1B, terfenadine inhibited the cromakalim- and NaCN-induced IK,ATP equally. The effects of terfenadine are
well expressed by the equation:
![<UP>Relative current</UP>=1/(1+[<UP>terfenadine</UP>]/<UP>IC</UP><SUB>50</SUB>)](/content/vol287/issue1/fulltext/293/img002.gif)
|
(2)
|
which indicates that terfenadine blocks IK,ATP by
interacting with the single binding site in one-to-one stoichiometry
(IC50 = 1.7 µM).
Figure 2 shows the quasi-steady-state
current-voltage (I-V) relationship for IK,ATP. Cromakalim
(10 µM) induced a large current (IK,ATP) having a
reversal potential of 85 ± 2 mV (n = 9, trace b
in fig. 2A). This current was partially inhibited by terfenadine at 1 µM (trace c), and the subsequent application of glibenclamide (1 µM) restored the control, N-shaped I-V relationship (trace d). The
glibenclamide-sensitive current in both the absence and presence of
terfenadine showed a linear I-V relationship (fig. 2B). The traces
shown in figure 2 C and D illustrate that the terfenadine blockage of
IK,ATP was slightly voltage-dependent; i.e.,
stronger blockage with depolarization. The mean slope between the
membrane potentials of 50 and +30 mV was 0.72 ± 0.17%/10 mV
with 1 µM concentration of terfenadine (n = 5). Use
of a higher concentration of terfenadine (5 µM) steepened the slope
to 1.11 ± 0.17%/10 mV (n = 6). The dotted lines
are an empirical fit based on the equation (3) shown in the
"Discussion" and will be described later.
View larger version (25K):
[in this window]
[in a new window]
|
Fig. 2.
A, Effect of terfenadine on the quasi-steady-state
I-V relationship for IK,ATP. The ramp protocol is shown in
the top inset. a, Control; b, cromakalim (10 µM); c, cromakalim plus
terfenadine (1 µM); d, glibenclamide (1 µM). B, The I-V
relationship for the glibenclamide-sensitive current in the absence
(b-d) and presence (c-d) of terfenadine. The data shown in A, B and C
were from the same cell. C and D, Voltage-dependent blockage of
IK,ATP by terfenadine. The relative current in the presence
of terfenadine [(c-d)/(b-d)] is plotted as a function of the
membrane potential. Note that the voltage-dependence was steepened with
a higher concentration (5 µM) of terfenadine. The dotted line
represents the fit to equation (3). The fractional electrical distance
(z ) from the cytoplasm was 0.16 in C, and 0.20 in D.
|
|
Effects of terfenadine on single KATP channels.
In
the following experiments, we examined the effects of terfenadine on
IK,ATP on the single-channel level. Figure
3A shows the effects of terfenadine (0.1 µM) on the single-channel current. Although this concentration of
terfenadine blocked the whole-cell IK,ATP by only 8.4 ± 5.0% (n = 5, fig. 1B), it halved the channel openings in the inside-out configuration. These current traces also
show that terfenadine did not change the amplitude of the single-channel current. In figure 3B, the amplitude of the
single-channel current is compared at various potentials between before
and after the application of terfenadine (0.3 µM). As previously
reported (Noma, 1983; Horie et al., 1992), the I-V
relationship for the single-channel current was linear at potentials
below +30 mV, and was inwardly rectified slightly at more positive
potentials. The I-V relationships clearly show that terfenadine did not
change the conductance of the single KATP channel
(48.2 ± 2.4 pS during control and 46.4 ± 3.6 pS in the
presence of terfenadine, n = 6). Therefore, the
inhibition of IK,ATP by terfenadine should be ascribed to
an attenuation of the open probability of the channels. Figure 3C
illustrates that when applied to the bathing solution, terfenadine
rapidly and reversibly reduced the open probability of the channels.
The concentration-dependent effect of terfenadine on the averaged
NPO is shown in figure 3D, which indicates that terfenadine
blocks IK,ATP with an IC50 of 0.19 µM at the
cytoplasmic surface.
View larger version (28K):
[in this window]
[in a new window]
|
Fig. 3.
Terfenadine blockage of single KATP
channels in the inside-out configuration. A, Single-channel recording
in the absence and presence of terfenadine (0.1 µM). The membrane
potential was held at 10 mV. Terfenadine was added to the bathing
(cytoplasmic) solution. The dotted line indicates the 0-current level
that was determined by applying 1 µM glibenclamide at the end of the
experiment. B, Conductance of the single KATP channel
current. and  denote the single-channel amplitude at various
holding potentials before and after the application of terfenadine (0.3 µM), respectively (n = 6). The slope conductance was
measured from the linear regression between 50 and +30 mV. C,
Temporal change in the open probability of the KATP
channels following the application of terfenadine. The NPO
measured every 15 sec is plotted as a function of time. The horizontal
bar indicates the period of terfenadine (5 µM) application to the
bathing solution. C, The concentration-dependent effect of terfenadine
on NPO. The NPO was averaged for a period of 1 min immediately before and after the application of terfenadine. The
relative NPO denotes the ratio of the averaged
NPO during the terfenadine perfusion to that during
control. The curve is the fit to equation (2). IC50 = 0.19 µM.
|
|
To elucidate the blocking mode of KATP channels by
terfenadine, we examined the effects of terfenadine on the opening and closing kinetics of the single channels. In the experiments shown in
figure 4, only the patches that showed
one channel opening throughout the experiment were used
(n = 4). Terfenadine was applied at 0.3 µM in
inside-out configuration. The representative records shown in figure 4A
clearly indicate that terfenadine abolished the long-lasting opening
with a development of flickering of the channel. Figure 4B shows the
distribution of the open- and closed-time. Both the open- and
closed-time histograms were expressed by a sum of two exponentials.
Note that terfenadine eliminated the events showing a long open-time
(>10 msec) and increased the incidence of the events having an open
time shorter than 10 msec. As a result, the slower time constant
(
slow) for the open-time distribution was significantly
shortened by terfenadine from 11.5 ± 3.0 to 4.5 ± 0.8 msec
(n = 4, paired t test). Both the faster and
slower time constants for the closed-time distribution were
significantly increased by terfenadine (table
1).
View larger version (28K):
[in this window]
[in a new window]
|
Fig. 4.
Effects of terfenadine on the kinetics of the
KATP channel. A, Terfenadine-induced flickering of the
channel opening. The data were obtained in the inside-out
configuration, and was low-pass filtered at 5 KHz. Note that only
one-channel activity is seen in the patch. The concentration of
terfenadine was 0.3 µM. B, Open- and closed-time histograms of the
single KATP current at 10 mV. The curve is the
least-squares two-exponential fit. The faster and slower time constants
(fast and slow, respectively) are
indicated.
|
|
We then examined the effect of terfenadine applied to external membrane
surface (fig. 5). The records were made
using cell-attached configuration. When terfenadine was applied to the
bathing solution outside the pipette, a 15-min exposure to terfenadine
at 5 µM did not significantly alter the NPO (fig. 5A).
This finding indicates that terfenadine is poorly diffusible into the
cytoplasm through the lipid membrane. In the experiments shown in
figure 5B, the terfenadine-free solution was sucked in from the pipette
tip, and the terfenadine-(5 µM) containing solution was applied to the shank from the back. The lower left diagram shows a representative temporal change in the NPO after making the seal on a
NaCN-loaded cell. An initial period of stable channel activity for
several min was followed by a decline of the channel opening,
presumably reflecting the diffusion of terfenadine molecules to the
membrane surface in the patch. The mean NPO was decreased
by 42 ± 5% during the recording of 10 min (n = 6, right lower panel of fig. 5B). This degree of the blockage was
significantly smaller than that induced by 5 µM terfenadine in
whole-cell configuration (68 ± 4%, n = 18, fig.
1B), which could be ascribed to a technical limitation in that the
access of the drug to the membrane depended on simple diffusion inside
the pipette. The current records shown in figure 5B demonstrate that
terfenadine applied to the outer surface also caused a flickering of
the channels.
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 5.
Effects of terfenadine on single KATP
channels in the cell-attached configuration. Cells were NaCN-loaded.
The bathing solution contained NaCN (10 mM) and cromakalim (10 µM).
In A, terfenadine was applied to the bathing solution. The temporal
plot of the NPO is shown in the lower left diagram. The
relative NPO in the right diagram represents the averaged
NPO 15 min after the commencement of the terfenadine (5 µM) perfusion with reference to that obtained before the terfenadine
application. Representative current records are shown in the upper
traces. In B, the tip of the pipette was filled with the
terfenadine-free solution with a negative pressure, whereas the shank
was filled with the terfenadine-(5 µM) containing solution. The lower
left diagram shows the temporal change in the NPO after the
formation of the seal. The relative NPO in the lower right
diagram was measured at 10 min after the commencement of the
recording.
|
|
Prolongation of the APD by terfenadine.
In rabbit ventricular
myocytes, ITO and IK1 play important roles in
the repolarization of the action potentials (Giles and Imaizumi, 1988).
Figure 6A shows the effect of terfenadine
on ITO elicited by a step depolarization to +20 mV and
indicates that terfenadine at 1 µM only slightly inhibited the
ITO (17 ± 1%, n = 7, fig. 6B). These
traces also show that the steady-state current at the end of the test
pulse is almost superimposable between before and after the addition of
terfenadine (1 µM) or 4-aminopyridine (5 mM). On average,
4-aminopyridine shifted the steady-state current inwardly by only
0.16 ± 0.07 pA/pF, but this change was not significant (fig. 6B).
Figure 6C illustrates the effect of terfenadine on IK1.
Terfenadine inhibited the background current, as measured by the ramp
experiments, at potentials negative to its reversal potential. Namely,
terfenadine (1 µM) blocked the inward background current by 14 ± 7% at 110 mV and by 15 ± 7% at 130 mV (n = 5). Whereas, terfenadine (1 µM) did not significantly affect the
steady-state outward net current at potentials between 85 and 30
mV. At more depolarized potentials (between 30 and +20 mV), we found
that terfenadine (1 µM) shifted the quasi-steady-state current
slightly (n = 5, fig. 6C). This terfenadine-sensitive outward current had a peak at 10 mV (0.22 ± 0.07 pA/pF,
n = 5), and was absent when the cells were treated with
E4031 (2 µM), a blocker of the rapidly activating delayed rectifier
K+ current (IKr) (n = 4, fig.
6D).
View larger version (30K):
[in this window]
[in a new window]
|
Fig. 6.
Effects of terfenadine on the repolarization currents
in rabbit ventricular cells. In A, left, superimposed are the currents
elicited by 350-msec depolarization to +20 mV from a holding potential
of 80 mV. Terfenadine (1 µM, ) and 4-aminopyridine (5 mM,  )
were applied sequentially. The right traces show the
4-aminopyridine-sensitive currents in an expanded time scale. B shows
the effects of terfenadine (1 µM) on the 4-aminopyridine-sensitive
current measured at the outward peak (left) and at the end of the test
depolarization (right) (n = 7). C shows the effect of
terfenadine (1 µM) on the background current as measured by the ramp
clamp. , control; , terfenadine. The currents between 60 and
+30 mV are expanded in the right diagram. D shows the result of the
same ramp experiment carried out in the presence of E4031 (1 µM).
|
|
Figure 7 shows the effects of terfenadine
of the action potentials. In these experiments, we used an ATP
concentration of 5 mM in the pipette, because with the use of the low
(0.4 mM) ATP-containing pipette solution, cromakalim shortened the
action potential progressively and usually abolished the action
potential plateau completely. Terfenadine did not prolong the APD
significantly during the control perfusion (n = 4, fig.
7A). As shown in figure 7B, after cromakalim at 10 µM shortened the
APD from the control value of 390 to 56 msec, the addition of 1 µM
terfenadine resulted in a recovery of the APD to 88 msec. Terfenadine
did not change the resting potential in the absence or presence of
cromakalim. The bottom right panel in figure 6B indicates that
terfenadine at 1 µM significantly prolonged the APD by 33 ± 5%
in the presence of cromakalim (n = 6).
View larger version (26K):
[in this window]
[in a new window]
|
Fig. 7.
Effects of terfenadine on the action potentials. In
A, the pipette solution contained 5 mM ATP, and the external solution
was normal Tyrode. The APD was measured from the onset of the action
potential upstroke to the repolarization to 60 mV (see
"Methods"), and is plotted in the lower diagrams. n = 4 for the right bar graph. The action potentials indicated by a and b
are shown in the upper traces. In B, the cell was first treated with
cromakalim at 10 µM (trace b), which was followed by the
coapplication of terfenadine at 1 µM (trace c). n = 6 for the right lower bar graph. NS and *indicate that the difference is
insignificant or significant (P < .05), respectively.
|
|
| |
Discussion |
Terfenadine blockage of IK,ATP.
We here showed
that terfenadine inhibited IK,ATP in rabbit ventricular
myocytes. The IC50 for the terfenadine blockage was 1.7 µM at a plateau potential of 10 mV. Thereby, terfenadine begins to
inhibit IK,ATP at a concentration of 0.1 µM and abolished it almost completely at 10 µM. Terfenadine (1 µM) significantly reversed the action potential shortening induced by cromakalim. The
therapeutic plasma concentration of terfenadine is reportedly less than
0.02 µM (Coutant et al., 1991). However, a high plasma concentration of approximately 0.3 µM was reported in patients who
underwent ventricular tachycardia or Torsade de pointes
(Coutant et al., 1991). Therefore, blockage of
IK,ATP by terfenadine can occur when patients with ischemic
heart disease take an overdose of terfenadine or take other drugs that
suppress the degradation of terfenadine by occupying the cytochrome
P-450 enzymes in the liver (Brown et al., 1985).
The single-channel experiments using the inside-out configuration
revealed that terfenadine applied to the cytoplasmic surface blocked
the KATP channel with a lower IC50 of 0.19 µM. However, terfenadine applied to the bathing solution in the
cell-attached configuration did not block the KATP channels
inside the patch, but inhibited the channel opening when it was applied
into the pipette. These findings suggest that although terfenadine
easily accesses the binding site in the KATP channel from
the inner surface of the cell membrane, the terfenadine molecules are
poorly diffusible into the lipid membrane and into the cytoplasm.
Terfenadine is a base with a pKa of 8.6, and thus most of terfenadine
molecules are charged at the pH of 7.4. Accordingly, terfenadine is
considered to access the binding site via the hydrophilic pathway as a
charged form. Terfenadine interrupted the long-lasting opening of the KATP channel, and caused a flickering of the channels
during the open state. This type of blockage, suggesting that
terfenadine preferentially blocks KATP channels in the open
state, was observed when terfenadine was applied either to the
cytoplasmic membrane surface or to the outer membrane surface in the
pipette.
The different sensitivity of IK,ATP to terfenadine applied
from the outer and inner membrane surface may reflect the location of
the binding site within the membrane; the binding site is located close
to the inner membrane surface. To test this hypothesis, we applied the
Woodhull model (1973) to our data shown in figure 2. The I-V
relationship for the single KATP current is rectified at
positive potentials (Horie et al., 1987). Voltage-dependent changes in the channel kinetics are also suggested to contribute the
rectifying property of the whole-cell IK,ATP (Zilberter
et al., 1988). However, we showed that the I-V relationships
for both the whole-cell IK,ATP and the single
KATP channel current are linear at potentials negative to
+30 mV (figs. 2B and 3B; see also Wu et al., 1992; Noma,
1983; Kakei et al., 1985; Horie et al., 1992).
That is, the open probability of the KATP channel must not
be affected by the membrane potential at the potentials used in our
experiments. The voltage-dependence of the IK,ATP blockage
by terfenadine (fig. 2 C and D) would therefore reflect the
interactions of the charged molecules with the binding site in the
electrical field, not the state-dependent blockage of
IK,ATP. According to Woodhull (1973), the drug-receptor
interaction within the electrical field is expressed as:
![f=[<UP>D</UP>]/([<UP>D</UP>]+K<SUB>d</SUB>×<UP>e</UP><SUP><UP>−</UP>&dgr;<UP>zF</UP>(<UP>E</UP>+10)/<UP>RT</UP></SUP>)](/content/vol287/issue1/fulltext/293/img003.gif)
|
(3)
|
where f denotes the fractional block, z, the valency,
F, Faraday constant, R, gas constant, T, absolute temperature and , the fraction of the transmembrane field sensed by a single charge at
the receptor site. Kd denotes the apparent
dissociation constant at the reference voltage (10 mV). The dotted
lines in figures 2 C and D represent the fit to this equation between
50 and +30 mV. The fractional electrical distance (z) from the
cytoplasm was 0.15 ± 0.04 with 1 µM terfenadine
(n = 5), and 0.18 ± 0.03 with 5 µM terfenadine
(n = 6), confirming that the terfenadine binding site
is near the internal surface of the membrane. Interestingly, a similar
value of 0.21 was reported for the interactions between terfenadine and
the recombinant human K+ channel (KV1.5, Yang
et al., 1995).
Effects of terfenadine on repolarizing K currents.
We also
found that terfenadine blocks K+ currents other than
IK,ATP in rabbit ventricular cells. Terfenadine at 1 µM
blocked ITO by only 17%. A similar degree of 1 µM
terfenadine-induced inhibition of ITO was reported in rat
and human cardiac myocytes (Berul and Morad, 1995; Crumb et
al., 1995). Although terfenadine blocked IK1,
terfenadine did not significantly affect the net outward current at
potentials between the reversal potential (85 mV) and 30 mV. Berul
and Morad (1995) also found that terfenadine (1 µM) blocked
IK1 only at potentials negative to 100 mV in guinea pig
ventricular myocytes. Salata et al. (1995) and Crumb
et al. (1995) reported that terfenadine (1 µM) did not
affect IK1. Together with our present data, these studies
suggest that the blockage of these major repolarizing currents (in
rabbit ventricular cells) by terfenadine at clinically relevant
concentrations is small.
More prominent blocking effects of terfenadine were reported on the
delayed rectifier-type K+ currents. Woosley et
al. (1993) and Salata et al. (1995) showed that in cat
and guinea pig ventricular cells, terfenadine blocked IKr
with IC50 values of 0.2 and 0.05 µM, respectively.
IKr in human atrial cells was also shown to be blocked by
terfenadine at 0.2 µM (Crumb et al., 1995). In this study,
we found that terfenadine (1 µM) blocks the quasi-steady-state
outward current at potentials compatible with the IKr
activation, and that this component was E4031-sensitive (fig. 6 C and
D). Although IKr is known to contribute to the action
potential repolarization in rabbit ventricular cells (Veldkamp et
al., 1993), the 1 µM terfenadine-sensitive current was small in
amplitude (0.22 pA/pF at 10 mV).
A potent inhibitory effect of terfenadine was also found on the
4-aminopyridine-sensitive ultra-rapidly activating delayed rectifier
K+ current (IKur). Crumb et al.
(1995) showed that terfenadine at 1 µM blocked IKur in
human atrial cells by 42%. Similarly, terfenadine (0.01-0.3 µM) was
found to block the cloned K+ channels that represent
IKur (Kv1.5a or fHK) (Crumb et al., 1995; Yang
et al., 1995; Rampe et al., 1993). In our study,
with step depolarization to +20 mV, we did not observe a significant
4-aminopyridine-sensitive component at the end of the test pulse (fig.
6A). In addition, terfenadine did not affect the quasi-steady-state
current at potentials between +20 and +30 mV (fig. 6C). Therefore,
IKur seems to be very small if present in rabbit
ventricular cells.
Effects of terfenadine on the action potentials.
Depending on
the types of the outward currents contributing to the action potential
repolarization, the effects of terfenadine can vary with species. A
marked prolongation of APD was reported in the guinea pig ventricle
(Pinney et al., 1995; Salata et al., 1995),
whereas terfenadine (
1 µM) barely changed the APD in canine Purkinje fibers (Lang et al., 1993). In this study, we found
that in rabbit ventricular cells, terfenadine (1 µM) did not affect the APD significantly (fig. 7A). This is probably because
ITO and IK1 are the major repolarizing currents
in these cells (Giles and Imaizumi, 1988). In addition, terfenadine
reportedly blocks the fast Na+ and L-type Ca++
current (Lang et al., 1993; Ming and Nordin, 1995; Liu
et al., 1997). Thus, in rabbit ventricular cells, the
blocking effects of terfenadine on ITO, IK1 and
IKr may have been counteracted by a concomitant inhibition
of these inward currents. However, IK,ATP carries a large
outward current at the plateau potentials because of the poor
rectification property. The net outward current during the action
potential plateau is small (for instance 10 pA/cell based on the
assumption that the plateau potential has a slope of 0.1 V/sec and
the cell has a membrane capacitance of 100 pF). It then follows that in
our experiments, cromakalim largely shortened the action potential, and
that terfenadine significantly reversed the action potential
shortening. However, the 1 µM terfenadine-induced reversal of the
action potential shortening was small (fig. 7B). This is probably
because the residual IK,ATP still had an amplitude sufficient to increase the net outward current during the action potential plateau.
The activation of IK,ATP was found to prevent the
arrhythmias induced by the early or delayed afterdepolarization
(Spinelli et al., 1991; Carlsson et al., 1992;
Takahashi et al., 1991). However, the roles of
IK,ATP in the development of ventricular arrhythmias during
ischemia are still controversial. The action potential shortening in
the ischemic zone induces dispersion of the effective refractory
period. The prolongation of the action potential due to the
IK,ATP blockage in the ischemic zone would attenuate the
dispersion of the refractory period. Blockage of IK,ATP is
supposed to contribute to the antiarrhythmic actions of some class 1 antiarrhythmic agents (Horie et al., 1992; Wu et
al., 1992). However, the action potential shortening due to activation of IK,ATP conserves the energy and protects the
myocardium during ischemia (Nichols and Lederer, 1991). It also plays a
key role in the ischemic preconditioning (Parret and Kane, 1994). Therefore, the blockage of IK,ATP by terfenadine perturbs
these protecting mechanisms of IK,ATP, and may eventually
aggravate the ischemia and the ischemia-induced arrhythmias. However,
terfenadine at clinically relevant concentrations blocks less than 50%
of IK,ATP (fig. 1). This degree of blockage only slightly
prolonged the action potential duration once the IK,ATP was
activated (fig. 7B). Although antihistaminic agents may modify the
ischemia-induced arrhythmias by blocking IK,ATP, our
results indicate that the blockage of this current is not a major cause
of antihistaminic agent-induced ventricular arrhythmias, which are
usually predisposed to by a prolongation of the action potential
duration or QT interval.
The authors thank Prof. Manabu Yoshimura for his support and
encouragement during this project.
![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Top
Abstract
Introduction
-aminoethyl ether)-N,N,N',N'-tetraacetic acid;
HEPES, N-2-hydroxyethyl-piperazine-N'-2-ethanesulfonic acid.
