Institute of Physiology (M.M., U.M., E.-J.S.) and Department of
Cardiology and Angiology (W.H., G.B.), University of Münster,
Münster, Germany; Hertie Foundation, Department of
Neuroscience/Multiple Sclerosis, Frankfurt, Germany (M.M.); Center of
Molecular Neurobiology, University of Hamburg, Hamburg, Germany (T.L.);
and Department of Anesthesiology, University Hospital Hamburg, Hamburg,
Germany (P.F., M.A.P.)
The effects of the antiarrhythmic drug propafenone at Kv2.1
channels were studied with wild-type and mutated channels expressed in
Xenopus laevis oocytes. Propafenone decreased the Kv2.1
currents in a time- and voltage-dependent manner (decrease of the time constants of current rise, increase of block with the duration of
voltage steps starting from a block of less than 19%, increase of
block with the amplitude of depolarization yielding a fractional electrical distance 
of 0.11 to 0.16). Block of Kv2.1 appeared with
application to the intracellular, but not the extracellular, side of
membrane patches. In mutagenesis experiments, all parts of the Kv2.1
channel were successively exchanged with those of the Kv1.2 channel,
which is much more sensitive to propafenone. The intracellular amino
and carboxyl terminus and the intracellular linker S4-S5 reduced the
blocking effect of propafenone, whereas the linker S5-S6, as well as
the segment S6 of the Kv1.2 channel, abolished it to the value of the
Kv1.2 channel. In the linker S5-S6, this effect could be narrowed down
to two groups of amino acids (groups 372 to 374 and 383 to 384), which
also affected the sensitivity to tetraethylammonium. In segment S6,
several amino acids in the intracellularly directed part of the helix significantly reduced propafenone sensitivity. The results suggest that
propafenone blocks the Kv2.1 channel in the open state from the
intracellular side by entering the inner vestibule of the channel.
These results are consistent with a direct interaction of propafenone
with the lower part of the pore helix and/or residues of segment S6.
 |
Introduction |
Voltage-operated
potassium currents play important roles in shaping and terminating
cardiac action potentials. Although several potassium current
components have been isolated, two types of currents can be
distinguished: fast activating and inactivating currents, referred to
as Ito, and delayed, more slowly inactivating currents, referred to as IK (Snyders, 1999
). The
molecular correlates underlying these current components, however, have
not yet been clearly determined. For IK, in
particular, several candidates have been identified or discussed. The
mRNA of the potassium channel families Kv1, Kv2, Kv3, and Kv4 are
present in significant amounts in cardiomyocytes (Dixon and McKinnon,
1994; Barry et al., 1995; Brahmajothi et al., 1996) and, after
expression in heterologous systems, yield currents sharing functional
characteristics with native IK (Snyders, 1999).
Much attention has been focused on the Kv1.5 potassium channel, which
is assumed to be the molecular correlate of the very rapidly activating
subcomponent of IK, referred to as
IKur (Snyders, 1999).
Another possible molecular correlate for IK,
however, might be Kv2.1. Cardiomyocytes of transgenic mice expressing
dominant negative Kv2.1 subunits showed a reduction of
IK and a marked prolongation of action potentials
(Xu et al., 1999). Accordingly, spontaneously triggered activity was
found in some myocytes of these transgenic mice. This correlated with
the observation that Kv2.1 mRNA is only present in part of wild-type
myocytes (Brahmajothi et al., 1996; Schultz et al., 2001). The
expression of Kv2.1 channels in heart is altered under pathologic
conditions associated with arrhythmias. Thus, 1) hyperthyroidism
drastically decreased the mRNA level of Kv2.1 in rat ventricular
myocytes (Nishiyama et al., 1998), 2) rats with infracted myocardium
showed increased action potential durations and reduced protein levels
of Kv2.1 (Huang et al., 2000), 3) cardiac hypertrophy in rats (induced by phorbol esters) increased the density of Kv2.1 (Walsh et al., 2001),
and 4) diabetic cardiomyopathy (induced by streptozotocin) yielded a
decrease of Kv2.1 protein density in rat left ventricular myocytes (Qin
et al., 2001).
Several drugs have been developed to prevent arrhythmias (Singh, 1997).
A widely used substance is
2'-[3'(propylamino)-2-(hydroxy)propoxy]-3-phenylpropiophenone hydrochloride; [propafenone (PROP)]. Electrophysiological studies using animal models have shown that PROP reduces the maximum rate of
rise and the amplitude of the action potential (Ledda et al., 1981).
Accordingly, sodium channel blocking effects of PROP were described
previously (Kohlhardt, 1984). Also, PROP increased the duration of the
action potential (Satoh and Hashimoto, 1984; Delgado et al., 1985) and
depressed the transient outward current (Ito) in
atrial myocytes of the rabbit and ventricular myocytes of the rat (Duan
et al., 1993), the hyperpolarization-activated inward current
(If) in isolated human atrial myocytes (Hoppe and
Beuckelmann, 1998), and the delayed rectifier current
(IK) in sinoatrial node cells and atrial myocytes
of rabbits and ventricular myocytes of guinea pigs (Satoh and
Hashimoto, 1984; Duan et al., 1993). In in vitro expression studies,
Kv1.5 and Kv2.1 currents are sensitive to micromolar PROP
concentrations (Franqueza et al., 1998; Zhu et al., 1999; Rolf et al.,
2000). The sensitivity of Kv2.1 channels to PROP increased 15-fold with
coexpression of a Kv6.2 potassium channel subunit (Zhu et al., 1999).
Thus, PROP in therapeutic concentrations may significantly affect
voltage-operated potassium channels.
In a recent study, we found that among cardiac Kv channels, Kv2.1
channels are particularly sensitive to PROP, whereas Kv1.2 channels are
relatively insensitive. To study the molecular site of PROP action at
the Kv2.1 channel, we constructed chimeric channels between Kv2.1 and
Kv1.2. All constructed chimeras gave functional channels; thus, a
systematic exchange of every part of Kv2.1 channel subunits could be
performed to identify the PROP site of action. In summary, we found
several Kv2.1 channel domains that affect PROP action. The results
suggest that PROP binds to the open Kv2.1 channel from the
intracellular side. Single-site mutations identified residues of the
lower part of the pore helix and the segment S6 most likely to be
involved in PROP interaction. The PROP binding site seems to be similar
but not identical to the ones reported for other antiarrhythmic agents
and potassium channel blockers.
| |
Materials and Methods |
In Vitro Mutagenesis and RNA Synthesis.
The cDNAs for
chimeras between rat Kv1.2 (GenBank accession number X16003) and
human Kv2.1 (GenBank accession number X68302) were obtained using an
overlap polymerase chain reaction and were cloned into the pGEM vector
(Liman et al., 1992). For the chimeras, fragments containing the
nucleotides (nt) given below were fused together. The numbers refer to
the open reading frame of the respective channel.
| |
Mu1: nt 1-492 of Kv1.2, nt 571-Stop of Kv2.1.
|
| |
Mu2: nt 1-555 of Kv2.1, nt 478-843 of Kv1.2, nt 862-Stop of
Kv2.1.
|
| |
Mu3: nt 1-882 of Kv2.1, nt 874-1230 of Kv1.2, nt 1240-Stop of
Kv2.1.
|
| |
Mu4: nt 1-1239 of Kv2.1, nt 1231-Stop of Kv1.2.
|
| |
Mu5: nt 1-882 of Kv2.1, nt 874-936 of Kv1.2, nt 946-Stop of
Kv2.1.
|
| |
Mu6: nt 1-954 of Kv2.1, nt 946-996 of Kv1.2, nt 1006-Stop of
Kv2.1.
|
| |
Mu7: nt 1-1005 of Kv2.1, nt 997-1035 of Kv1.2, nt 1042-Stop of
Kv2.1.
|
| |
Mu8: nt 1-1059 of Kv2.1, nt 1051-1164 of Kv1.2, nt 1171-Stop of
Kv2.1.
|
| |
Mu9: nt 1-1182 of Kv2.1, nt 1174-1230 of Kv1.2, nt 1240-Stop of
Kv2.1.
|
| |
Mu10: nt 1-1059 of Kv2.1, nt 1051-1077 of Kv1.2, nt 1087-1155
of Kv2.1, nt 1147-1164 of Kv1.2, nt 1174-Stop of Kv2.1.
|
| |
Mu11: nt 1-1095 of Kv2.1, nt 1087-1092 of Kv1.2, nt 1102-Stop of
Kv2.1.
|
| |
Mu12: nt 1-1113 of Kv2.1, nt 1105-1113 of Kv1.2, nt 1123-Stop of
Kv2.1.
|
| |
Mu13: nt 1-1146 of Kv2.1, nt 1138-1143 of Kv1.2, nt 1153-Stop of
Kv2.1.
|
| |
Mu14: nt 1-1182 of Kv2.1, nt 1174-1176 of Kv1.2, nt 1186-1191
of Kv2.1, nt 1183-1185 of Kv1.2, nt 1195-Stop of Kv2.1.
|
| |
Mu15: nt 1-1209 of Kv2.1, nt 1201-1203 of Kv1.2, nt 1213-Stop of
Kv2.1.
|
| |
Mu16: nt 1-1224 of Kv2.1, nt 1216-1218 of Kv1.2, nt 1228-1230
of Kv2.1, nt 1222-1224 of Kv1.2, nt 1234-1239 of Kv2.1, nt 1231-1233
of Kv1.2, nt 1243-Stop of Kv2.1
|
| |
Mu17: nt 1-1248 of Kv2.1, nt 1240-1245 of Kv1.2, nt 1255-Stop of
Kv2.1
|
DNA sequences amplified by the polymerase chain reaction were
verified by sequencing using the BigDye terminator cycle sequencing kit
(Applied Biosystems, Foster City, CA). The sequence reactions were analyzed on an ABI 377 or Prism 310 automated sequencer (Applied Biosystems).
The cDNA encoding for the potassium channel subunits rat Kv1.2, human
Kv2.1, and their chimeras were transcribed to cRNA using a commercial
kit (mMessage mMachine; Ambion, Austin, TX) and T7 RNA polymerase.
Denaturating agarose gel electrophoresis was used to check the quality
of cRNA product of each reaction and to quantify the yield.
Preparation of Oocytes and Cell Lines.
South African clawed
frogs (Xenopus laevis) were anesthetized in ethyl
m-aminobenzoate (Sandoz, Basel, Switzerland) and small sections of the ovary were removed surgically. Oocytes (stage V or VI;
Dumont, 1972) were injected with 0.1 or 1.0 ng of cRNA in 50 nl of
distilled water and were maintained under tissue culture conditions at
20°C until used for experiments. The tissue culture solution was a
modified Barth medium (88 mM NaCl, 1 mM KCl, 1.5 mM
CaCl2, 2.4 mM NaHCO3, 0.8 mM MgSO4, 5 mM HEPES, pH 7.4) that was
supplemented with penicillin (100 IU/ml) and streptomycin (100 µg/ml).
For the electrophysiological recordings on membrane patches, part of or
all of the follicular tissues were removed from the oocytes. The outer
part of the follicular tissues was stripped off some hours after
injection of cRNA with forceps (defolliculation). The remaining
follicular tissues were removed manually just before the
electrophysiological investigation after shrinkage in a "stripping solution" (200 mM K-aspartate, 20 mM KCl, 1 mM
MgCl2, 10 mM EGTA, and 10 mM HEPES, pH 7.4 (Methfessel et al., 1986).
For heterologous expression in cell lines, we used CHO cells expressing
human Kv2.1 and rat Kv1.2 channels (Zhu et al., 1999). The cDNA were
subcloned into pcDNA3, and CHO cells were transfected using
LipofectAMINE following the manufacturer's protocol (Invitrogen, Carlsbad, CA). In all experiments, pcDNA3-GFP was cotransfected together with pcDNA3-Kv2.1 to mark the CHO cells expressing Kv channel
subunits with green fluorescent protein fluorescence. Briefly, 200 µl
of Opti-MEM I containing a total of 0.1 µg DNA (Kv2.1) or 1.0 µg
DNA (Kv1.2) together with 3 µl of LipofectAMINE and 0.5 µg of
enhanced green fluorescent protein were used to transfect 4 × 105 cells in a 35-mm tissue culture plate. The
lipid-DNA complex solution was replaced after 5 to 6 h of
incubation by minimal essential medium-
.
Electrophysiological Techniques.
For investigations of
oocytes with the two-electrode voltage-clamp technique, microelectrodes
were made from borosilicate glass and had resistances of 0.5 to 1 M
for the current electrodes and 1 to 2 M for the potential electrodes
when filled with 3 M KCl. The holding potential was 
80 mV, and
command potentials were applied up to a potential of +80 mV. Tail
currents were obtained by stepping from +20 mV to potentials of 40 to
120 mV. The control bath fluid was a Ringer solution (115 mM NaCl, 2 mM KCl, 1.8 mM CaCl2, and 10 mM HEPES; pH 7.2. Propafenone (PROP as chloride salt; Sigma, Deisenhofen, Germany) in
concentrations of 1 to 2000 µM and tetraethylammonium (TEA as
chloride salt; Merck-Schuchardt, Hohenbrunn, Germany) was added to the
bath solution and applied at least 30 s before eliciting currents.
Solutions were applied with a concentration-clamp technique (Madeja et
al., 1991), allowing an exchange of more than 90% of the extracellular
solution within <10 ms. All experiments were performed at days 3 and 4 after injection of cRNA and were carried out at room temperature
(22 ± 1°C).
Investigations of oocytes with the patch-clamp technique were done on
excised membrane patches in the outside-out or inside-out configurations. The patch pipettes had tip diameters between 1 and 4 µm and resistances between 2 and 5 M. The holding potential was
80 mV, and command potentials were applied up to a potential of +20
mV. Solutions were applied with the double-barrel flow pipette method
(Johnson and Ascher, 1987). In outside-out membrane patch
investigations, the bath fluid was the above-mentioned Ringer solution,
and the patch pipettes were filled with a solution of the composition
100 mM KCl, 10 mM EGTA, and 10 mM HEPES, pH 7.2. For investigations on
inside-out membrane patches, the solutions were exchanged correspondingly.
Channels expressed in CHO cells were studied with the patch-clamp
technique in the whole-cell configuration. The holding potential was
80 mV, and command potentials were applied to a potential of +60 mV.
The patch pipettes (resistances between 1.8 and 3 M) were filled
with a solution of the composition 160 mM KCl, 0.5 mM
MgCl2, 10 mM HEPES, and 2 mM
ATP-Na2, pH 7.2. The bath solution was composed
of 135 mM NaCl, 5 mM KCl, 2 mM MgCl2, 2 mM
CaCl2, 10 mM sucrose, 5 mM HEPES, and 0.01 g/liter phenol red, pH 7.4. PROP was added to the bath solution, giving
concentrations from 0.3 to 100 µM and was applied to the cells using
a hydrostatic superfusion system (50 to 100 µm distance to the cell
under investigation). Currents were recorded at room temperature at
days 2 and 3 after transfection of the cells.
Data Acquisition and Analysis.
The potassium currents
obtained in two-electrode recordings were low-pass filtered at 1 kHz
and were transferred to a computer system (pClamp; Axon Instruments,
Union City, CA). The amplitudes of the total outward currents were
corrected for leakage. Leakage currents and capacitive transients were
subtracted online using a p/4 pulse protocol.
The potassium current amplitude was measured at the peak of current
obtained during the depolarizing voltage step. Conductance-voltage relations were obtained by normalizing the conductance data to the
maximal value under control conditions and by fitting the data to the
Boltzmann equation y = Gmax (1 + exp
[(V1/2 V)/b]), where y is the normalized conductance,
Gmax is the normalized maximal
conductance, V1/2 is the potential of
the half-maximal conductance, V is the voltage, and
b is the slope factor. The decay of tail currents was fitted
with monoexponential functions. Concentration-response curves were
determined by fitting the mean current values at different PROP
concentrations to the Langmuir equation y = (KD/c)nH/[1 + (KD/c)nH],
where y is the fraction of control current,
KD is the half-blocking concentration,
c is the concentration of PROP, and
nH is the Hill coefficient.
The voltage dependence of block was determined using the
KD values that were obtained for these
calculations from the fractional current (f), measured as
the current in the presence of propafenone (IPROP) at a concentration [D] of 100 µM or 2 mM and under control conditions (ICTRL) at the
end of the voltage step: f = IPROP/ICTRL and
KD = [D] × f/(1 f). The fractional electrical distance () (i.e., the
fraction of the transmembranous electrical field sensed by a single
positive charge at the binding site) was determined by fitting the
KD values with the equation
KD = KD(0mV) × exp(zFV/RT). KD(0mV) represents the half-blocking
concentration at the reference potential of 0 mV, V is the
membrane potential, z is the charge of the molecule,
F is the Faraday constant, R is the real gas constant, and T is the absolute temperature.
The potassium currents of CHO cells were low-pass filtered at 3 kHz,
digitized using an analog-to-digital converter (HEKA Electronic,
Lambrecht, Germany) and were stored with a sampling rate of 10 kHz.
PROP-induced inhibition of the currents was quantified as inhibition of
the steady-state current. KD values
were determined with the above-mentioned Langmuir equation.
The measured values are given as mean or mean ± S.E.M.
Statistical significance was tested using a t test or a
Mann-Whitney rank sum test. Values of p < 0.01 were
taken as statistically significant. Curve fitting and all statistical
procedures were performed using the program SigmaPlot (SPSS Science,
Chicago, IL).
| |
Results |
Conductance Parameters.
In agreement with previous results,
outward currents mediated by Kv2.1 channels expressed in X. laevis oocytes appeared at potentials positive to 30 mV. The
currents increased relatively slowly and did not significantly
inactivate during the 500 ms test pulses (Fig.
1A, CTRL). PROP at a concentration of 100 µM decreased the outward currents over the entire potential range. At
the most positive test pulses (+50 and +60 mV), a change in the
activation kinetics became apparent (Fig. 1A, PROP; see the small hump
at the beginning of the current traces). A fitting of the mean
conductance values revealed a decrease of the maximal conductance
(Gmax) by more than 60% (Fig. 1B).
The conductance-voltage relation was not significantly affected with a
shift of the potential of half-maximal conductance
(V1/2) of less than 3 mV (mean
values: CTRL, 2.4mV; slope factor, 11.4 mV; PROP, 0.5 mV; slope
factor, 11.4 mV).
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 1.
Effect of PROP on Kv2.1 potassium currents. X.
laevis oocytes. Potential steps from 80 to +60 mV under
control conditions (CTRL,  ) and with PROP (100 µM,  ). A,
original recordings. B, open probability curves of mean conductance
values from 10 experiments. The conductance values
(Grel) were normalized to the respective
maximal value under control conditions. The error bars indicate SEM.
MP, membrane potential.
|
|
Activation and Deactivation Kinetics.
As can be seen from a
comparison of the normalized current traces at +40 mV in Fig.
2A, the Kv2.1 current increased more
rapidly with PROP compared with control conditions (Fig. 2A, left). The rise of current could be well fitted with monoexponential functions (except for the first few milliseconds of the test pulse; see Fig. 2A,
left) at test potentials of +10 to +60 mV. A plot of the time constants
as a function of the test voltage (Fig. 2A, right) showed that the time
constants decreased from 44 ± 3 ms at +10 mV to 11 ± 1 ms
at +60 mV (n = 5). In the presence of PROP, the time
constants were decreased at all test potentials. The decrease was
voltage-dependent and ranged from 30% at +10 mV to 42% at +60 mV. In
line with this voltage dependence, a linear fit of the logarithm of the
time constants in the range from +30 to +60 mV revealed 162 and 108 mV
for a 10-fold change in time constants under control conditions and
PROP, respectively (n = 5; data not shown). PROP did
not alter the deactivation kinetics of Kv2.1 channels (Fig. 2B, left),
which could be well described with monoexponential functions. Stepping
from +20 mV to test voltages of 40 to 120 mV, the voltage
dependence of deactivation time constants decreased from 17 ± 1 ms at 40 mV to 4 ± 1 ms at 120 mV (n = 5). In
the presence of PROP, the time constants were not significantly
different from the control values (Fig. 2B, right).
View larger version (23K):
[in this window]
[in a new window]
|
Fig. 2.
Effect of propafenone (PROP) on current kinetics of
Kv2.1 potassium channels. X. laevis oocytes. Currents
(I) under control conditions (CTRL, ) and with propafenone (100 µM, ). MP, membrane potential. A, time course of current increase.
Left, original recordings of potassium currents at +40 mV. The solid
lines indicate the fit of currents with monoexponential functions.
norm. PROP, currents with PROP normalized to the maximal current under
control. Right, graphical evaluation of the time constants ( ) of
current increase at different potentials. The data are given as
mean ± S.E.M. of five experiments and are fitted with
monoexponential functions. Asterisks mark statistically significant
differences. B, time course of tail currents. Left, tail currents at a
potential of 40 mV after 50-ms voltage steps to +20 mV. Except for
the first 3 ms after the end of the voltage step, the tail currents
were fitted with monoexponential functions. Right, graphical evaluation
of the time constants () of tail currents at different potentials;
five experiments. Further explanation as in A.
|
|
Kinetic and Voltage Dependence of Block.
We analyzed the
development of Kv2.1 current block in the presence of 100 µM PROP in
comparison with control currents. Plotting this relative current
against test pulse duration showed that the Kv2.1 current block
developed rapidly within the first milliseconds of the test pulse and
then increased more slowly during the rest of the pulse (Fig.
3A). Correspondingly, the decay could not
be fitted with a monoexponential function but was described with the
sum of two monoexponential functions with time constants of 14 and 901 ms at +60 mV (fit of normalized mean currents of five experiments).
Extrapolating this fit of the development of block to zero time at the
beginning of the test pulse indicated an initial block of <19% (Fig.
3A, hatched line). This observation supports the view that PROP mainly
blocks the open Kv2.1 channel and that the efficacy of block increases
with open probability. The block was apparently not use-dependent
because the same block of current was obtained after 1, 10, or 30 voltage pulses during a 30-s application of PROP (data not shown,
n = 5).
View larger version (11K):
[in this window]
[in a new window]
|
Fig. 3.
Time course and voltage dependence of the block of
Kv2.1 potassium currents by PROP. X. laevis oocytes. A,
change of fractional current (Irel) [i.e., fraction of
current with PROP (100 µM) relative to current under control
conditions (CTRL)] with the duration (t) of voltage
steps to +60 mV (except for the first 8 ms). The trace is the mean of
six experiments. Hatched line, fit of Irel with
biexponential functions. The fit is shown from 0 to 500 ms of the
voltage step. Dotted line, fraction of Irel obtained from
the faster time constant of the biexponential fit. Inset, original
recording of currents at +60 mV. B, voltage dependence of time
constants () of biexponential fits used to describe the time course
of block as shown in A. Fast () and slow time constants () are
shown as mean ± S.E.M. of five experiments in the potential range
from +20 to +80 mV. The solid line indicates the linear regression for
the correlation between the logarithm of the time constant and the
voltage (P = 0.006; regression coefficient
r = 0.90). MP, membrane potential; C, voltage
dependence of fraction of block. The half-blocking concentrations
(KD) at different potentials are given as
mean ± S.E.M. of five experiments (, dotted line). The hatched
line gives the open probability curve obtained by fitting the relative
conductance values (Grel) current amplitude
with a Boltzmann equation. The fractional electrical distance
(z) (i.e., the fraction of the transmembranous
electrical field sensed by a substance with the positive charge z at
the binding site) was determined as described under Data
Acquisition and Analysis. The fit was performed with the
measured KD(0mV) value of 111 ± 13 µM (z1 = 0.13 ± 0.02) and
with an unknown KD(0mV)
(z2 = 0.11 ± 0.02, KD(0mV) = 107 ± 17 µM;
n = 5, solid lines).
|
|
Next, we studied the contribution of the two different components of
the block. At +60 mV, the fast component of block (Fig. 3A, dotted
line) represented more than 89% of the block computed as the
difference of the block after 0 and 500 ms. At +20 mV, the fast
component slightly decreased to 82% of the block, thus suggesting a
voltage dependence of the two components of block (data not shown). We
therefore measured the two time constants of the block in the potential
range from +20 to +80 mV (Fig. 3B). The fast time constant decreased
steadily from 25 ± 7 ms at +20 mV to 14 ± 1 ms at +80 mV,
whereas the slow time constants ranged between 1035 ± 256 ms and
1309 ± 282 ms. A linear regression was obtained for the
correlation between the logarithm of the fast time constant and the
voltage (p = 0.006; regression coefficient r = 0.90; Fig. 3B), whereas the slow time constants
showed no voltage dependence (n = 5). Thus, the
voltage- and time-dependent increase in block described by the fast
time constants is consistent with a block of the channel in the open
state. The additional slowly increasing block is probably independent
of the channel state and might be caused, for example, by a
diffusion-dependent increase of PROP concentration at the binding site
(see Discussion).
In the open channel, the whole electric field has to decline from the
internal to the external mouth of the channel pore, and because this
field might affect the action of a blocking agent, we performed
investigations on the voltage dependence of PROP block. The voltage
dependence of block was calculated by the fraction of control current
with propafenone in the potential range from 20 to +80 mV (Fig. 3C).
The PROP-induced block increased with positive-going potentials
as indicated by a steady decrease of the half-blocking concentrations
(KD) with positive-going potentials. The fractional electrical distance (i.e., the fraction of the transmembranous electrical field sensed by the drug at the binding site, referenced to the intracellular side of the channel) was calculated at potentials from +50 to +80 mV. In this potential range,
the open probability was roughly maximal (change of maximal conductance
<7%; n = 5; Fig. 3C, hatched line) and the endogenous currents of the oocytes were small (<0.8 µA up to +80 mV in
water-injected oocytes; n = 3; data not shown). A fit
of the half-blocking concentrations (KD) at these potentials with the
measured KD of 111 ± 13 µM
PROP at the reference potential of 0 mV (KD
0mV) yielded an electrical distance
z1 = 0.13 ± 0.02 (n = 5; Fig. 3C). At the potential of 0 mV, however,
the open probability was low (45% of the maximal value), suggesting
that the theoretical KD for maximal
open probability could yield a smaller value than the computed one. The
values were therefore refitted with the assumption of an unknown
KD at 0 mV. The fit revealed an
electrical distance z2 = 0.11 ± 0.02 and
a KD 0mV = 107 ± 17 µM PROP
(Fig. 3C). As expected, the fitted KD
0mV value was lower than the measured one, but the
difference was small, indicating that at this (and at more positive)
potential, the increase in open probability did not contribute strongly
to the increase in block.
Because the block by PROP could only be well described with the sum of
two exponential functions and because only the fast component might be
caused by changes in the channel's state, we also calculated the
fractional electrical distance using the limit values of the fast
monoexponential fit of block. With these values, we obtained an
electrical distance z3 = 0.14 ± 0.07 (data not shown; n = 5).
Effects on Isolated Membrane Patches and Transfected Mammalian
Cells.
To test the hypothesis that PROP blocks open Kv2.1 channels
from the inside, we applied 10 µM PROP to membrane patches excised from oocytes expressing Kv2.1 channels. As a first step, we measured current-voltage relations from 80 to +20 mV in inside-out membrane patches (Fig. 4A). As observed for the
whole oocyte, the threshold for Kv2.1 current activation was between
40 and 30 mV. Under these conditions, PROP at a concentration of 10 µM blocked much more than half of the current (Fig. 4A). Thus, the
blocking effect seemed to be much larger than in whole oocytes (compare
Figs. 1 and 3C), suggesting that the reduced sensitivity in whole
oocytes is most probably caused by reduced access to the channel's
binding site (Madeja et al., 1997; Rolf et al., 2000). The persistence of the PROP effect in isolated membrane patches suggests, however, that
intracellular messengers or factors of the cell are not critically involved in the action of PROP and that the block of Kv2.1 currents is
induced by PROP directly.
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 4.
Effect of PROP on excised membrane patches and
mammalian cells with Kv2.1 and Kv1.2 potassium channels. CTRL, control
conditions. A, current-voltage relations of Kv2.1 channels in
inside-out patches of oocytes of X. laevis. PROP was
applied at a concentration of 10 µM. Inset, original recordings used
for the graphical evaluation. Calibrations, 20 pA and 90 ms. CTRL,
control conditions; I, potassium current. B, concentration-response
curves for PROP of Kv2.1 () and Kv1.2 currents () in
heterologously transfected CHO cells. The symbols represent fraction of
inhibition of potassium current as mean ± S.E.M. of 16 (Kv2.1)
and 21 experiments (Kv1.2). Data were fitted to a Langmuir equation.
Irel, fraction of current with PROP relative to current
under control conditions; 1 Irel, fraction of
inhibition. The insets show original recordings. Calibrations; 1 nA and
100 ms. C, effect of PROP on Kv2.1 potassium currents in membrane
patches of oocytes of X. laevis in the inside-out (1)
and outside-out (2) configurations. Superposed original recordings
before (CTRL) and with PROP (1 s after begin of application).
Calibrations, 6 pA and 90 ms.
|
|
To study the sensitivity to PROP in a mammalian expression system, we
measured the KD values in CHO cells
expressing Kv2.1 channels (and for comparison also Kv1.2 channels).
Voltage steps were elicited from 80 to +60 mV (Fig. 4B). The
KD values were 1.2 ± 0.3 µM
for Kv2.1 (n = 16) and 9.8 ± 4.9 µM for Kv1.2
(n = 21; Fig. 4B). In oocytes, however, the
KD values were 91 ± 12 µM for
the Kv2.1 channel (n = 8) and 953 ± 88 µM for
the Kv1.2 channel (n = 7; data not shown). Thus, the
two Kv channels expressed in the mammalian cells showed a much higher
sensitivity to PROP compared with their sensitivity in oocytes, but the
relative differences in sensitivity between Kv2.1 and Kv1.2 channels
were similar in both expression systems
(KDKv2.1/KDKv1.2:
0.12 in CHO cells and 0.10 in oocytes).
To test PROP action on the intracellular or extracellular side of the
membrane, we applied PROP to membrane patches of oocytes in both the
inside-out and outside-out configuration (Fig. 4C). The excised patches
contained only a few channels in our experiments; thus, it was not
possible to make conclusions about channel kinetics. We elicited Kv2.1
currents by a voltage step to +20mV. To avoid contaminating effects
caused by PROP traversing the cell membrane, we elicited the voltage
steps 1 s after begin of PROP application to the membrane patches.
With this protocol, 10 µM PROP applied to inside-out patches markedly
reduced Kv2.1 current amplitudes (Fig. 4C, 1; n = 4).
By contrast, PROP application to outside-out patches did not
significantly affect Kv2.1 current amplitudes (Fig. 4C, 2;
n = 5).
Mutagenesis Experiments.
As reported previously (Rolf et al.,
2000), Kv1.2 channels in oocytes of Xenopus laevis are
relatively resistant to block by PROP. Kv1.2 mean current amplitudes at
+40 mV (i.e., 50 mV positive to the threshold potential) were reduced
by only 9%, whereas Kv2.1 currents were decreased by 59% (Fig.
5A; n = 6 to 9). We took
advantage of this observation and constructed a systematic set of
Kv2.1/Kv1.2 chimeras to screen the Kv2.1 channel domains for
high-affinity PROP binding sites in the oocyte expression system. We
tested for insensitivity of the Kv2.1 channel to also discover a
possible composite site of action. In a first step, the Kv2.1 amino
terminus (Mu1), the first (Mu2) and second half (Mu3) of the
membrane-inserted Kv2.1 core domain and the Kv2.1 carboxy terminus
(Mu4) were replaced by Kv1.2 (Fig. 5B; n = 5 to 9).
Current amplitudes at +40 mV of chimeras Mu1, Mu2, and Mu4 were reduced
by PROP in a similar manner to those of wild-type Kv2.1 channels,
whereas the current amplitudes of chimera Mu3 were as insensitive to
PROP as wild-type Kv1.2 channels. These results suggest a high-affinity
binding site for PROP to domain(s) in the second half of the Kv2.1
membrane-spanning core region extending from segment S4 to S6.
View larger version (38K):
[in this window]
[in a new window]
|
Fig. 5.
Effect of PROP (100 µM) on potassium currents of
chimeric channels obtained from Kv1.2 and Kv2.1 wild-type channels.
X. laevis oocytes. The schematic drawings represent the
structure of the wild-type and mutant channels (Mu1 to Mu9). The bars
and numbers show the current percentage reduction 50 mV positive to the
threshold potential (mean ± S.E.M.). The numbers in brackets
indicate the number of experiments. Asterisks, no statistically
significant difference from the Kv1.2 channel; Daggers, statistically
significant difference from the Kv2.1 channel. Wild-type channels (A),
chimeras with exchanges of larger parts of the channel (B), and
chimeras with exchanges of single segments or linkers in the region S4
to S6 (C).
|
|
Next, we analyzed the S4 to S6 region in more detail (Fig. 5C;
n = 5 to 9). Replacement of segment S4 (Mu5), segment
S5 (Mu7), or the intracellular S4-S5 linker (Mu6) did not transfer PROP insensitivity to the Kv2.1 channel. Transfer of the S5-S6 linker region
(Mu8), as well as replacement of segment S6 (Mu9), reduced the
sensitivity of Kv2.1 channels to PROP to the level of wild-type Kv1.2
channels. Thus, the site of action can be assumed to be located at
those parts of the channel molecule that have been shown to form the
pore proper of a potassium channel (Doyle et al., 1998). It should be
noted, however, that in addition to the mutations mentioned above (Mu3,
Mu8, and Mu9), smaller but statistically significant reductions of the
PROP-induced decrease of current were found for the mutants Mu1, Mu4,
and Mu6 (Fig. 5, daggers), thus suggesting some effect of the
intracellular termini and the intracellular linker S4-S5.
The results suggest that the amino acid residues forming the S5-S6
linker region and segment contain the PROP binding site of the Kv2.1
channel. At first, we exchanged amino acid residues in the Kv2.1 S5-S6
linker region and measured the change in maximal conductance of the
mutants. Concerning the linker region between segments S5 and S6 (Fig.
6A, n = 6 to 10), the
exchange of the chains of extracellular amino acids flanking the
intramembraneous pore region (Doyle et al., 1998; Fig. 6A, p) did not
reduce the PROP sensitivity significantly (Mu10). Within the P-region,
however, the exchange of a group of three amino acids (threonine 371, isoleucine 372, and threonine 373; Mu12) and another group of two amino
acids (isoleucine 383 and tyrosine 384; Mu13) for those of the Kv1.2 channel strongly reduced the channel's sensitivity to PROP
(differences from Kv1.2 statistically not significant; Fig. 6A,
asterisks), whereas the other exchange in the P-region (alanine 366 and
serine 367; Mu11) had no statistically significant effect (Fig. 6A).
View larger version (30K):
[in this window]
[in a new window]
|
Fig. 6.
Effect of PROP (100 µM) on potassium conductance of
mutated Kv2.1 channels representing substitutions by the corresponding
residues of the Kv1.2 channel in the linker S5-S6 (A) and in the S6
segment (B). X. laevis oocytes. The left part shows the
amino acid sequences of the wild-type and mutated channels (Mu10 to
Mu17). For the Kv1.2 channel and the mutated channels, only the
residues differing from Kv2.1 are given. The bars and numbers show the
percentage reduction of maximal conductance (mean ± S.E.M.). The
numbers in brackets indicate the number of experiments. Asterisks, no
statistically significant difference from the Kv1.2 channel; Daggers,
statistically significant difference from the Kv2.1 channel.
|
|
To study the concentration dependence of the PROP action on these
mutants, we measured the KD values at
potentials of +60 mV by applying PROP in concentrations from 1 to 1000 µM. Both mutants showed an insensitivity to PROP (Mu12,
KD, 2909 ± 614 µM,
n = 7; Mu13, KD,
2180 ± 512 µM, n = 6) which was the same or
even greater than of the of the Kv1.2 channel
(KD = 953 µM) and clearly distinct
of the Kv2.1 channel (KD = 91 µM).
Because the residues 371 to 373 and 383 to 384 are located in regions
of the molecule that have been associated with the binding site of TEA
in other potassium channels (Yellen et al., 1991; see
Discussion), we tested the effect of TEA on the mutants.
Similar to PROP, the Kv2.1 channel was more sensitive to TEA than the Kv1.2 (reduction of Gmax by 20 mM TEA:
Kv2.1, 67 ± 3%, n = 6; Kv1.2, 20 ± 3%,
n = 5; data not shown). Both mutants of the P-region had a reduced PROP sensitivity (Mu12 and Mu13) and also a decreased sensitivity to TEA (reduction of Gmax:
Mu12, 37 ± 2%; Mu13, 2 ± 1%, n = 6 each;
differences to Kv2.1 statistically significant; data not shown).
In a last set of mutants, we successively replaced Kv2.1 amino acid
residues in S6 with those of Kv1.2 (Fig. 6B, Mu14 to Mu17). None of the
current amplitudes of these mutants was reduced by PROP application as
strongly as those of Mu9 or of wild-type Kv1.2 channels (Fig. 6B). Mu14
current amplitudes were as sensitive to PROP as wild-type Kv2.1
channels. In contrast, the current amplitudes associated with Mu15,
Mu16, and Mu17 were significantly reduced by PROP but did not reach the
PROP sensitivity of wild-type Kv1.2 channels.
Finally, we measured the fractional electrical distance
z1 (in the potential range from 0 to +60 mV) for Kv1.2 and Mu10 to Mu17. Effects of PROP on voltage
dependence of open probability were estimated by the shifts of
V1/2 in the presence of 100 µM PROP.
For all channels, PROP induced shifts to negative potentials. The shift
ranged between 0.3 ± 0.2 mV (Mu12) and 2.7 ± 0.3 mV (Mu17); the only exception was Mu14 with a shift of 6.2 ± 0.7 mV. The calculations of the fractional electrical distance yielded almost unchanged values for Mu14
(z1 = 0.14 ± 0.01, n = 8) and Mu17 (z1 = 0.13 ± 0.02, n = 6) compared with the Kv2.1
wild-type (z1 = 0.13, see above),
but reductions for Mu10 (z1 = 0.06 ± 0.01, n = 7), Mu11
(z1 = 0.05 ± 0.01, n = 8), Mu15 (z1 = 0.06 ± 0.02, n = 8), and Mu16
(z1 = 0.10 ± 0.04, n = 7). Because Mu12 and Mu13 showed only small effects
with 100 µM PROP, measurements were made with the concentration of
2000 µM PROP. The fractional electrical distances so obtained were
larger than those of Kv2.1 and yielded
z1 = 0.15 ± 0.03 for Mu12
(n = 6) and z1 = 0.21 ± 0.03 for Mu13 (n = 7). For the
interpretation of these values, however, it must be noted that,
especially for Mu13, the high PROP concentration caused a significant
negative shift (Mu13, 7.5 ± 1.2 mV, n = 7;
Mu12, 3.1 ± 1.0, n = 6), which, in principle, could lead to increased values of
z1.
In summary, amino acid residues in the P-region and in the S6 segment
reduced the PROP sensitivity of the Kv2.1 channel. A complete transfer
of PROP insensitivity was obtained by exchanging the amino acids 371 to
373 and 383 to 384 of the P-region of the Kv2.1 channel. Furthermore,
several amino acids of the S6 segment (404, group 409/411/414 and group
417/418) were found to reduce the sensitivity of the Kv2.1 channel to
PROP.
| |
Discussion |
The present study suggests 1) that PROP decreases the Kv2.1
potassium current by an open-channel block, 2) that PROP acts from the
intracellular side of the membrane, and 3) that several residues of the
pore region and of segment S6 determine the sensitivity to PROP.
Open-Channel Block.
Three findings of the present study
suggest a blockade in the open state of the channel: 1) the increase of
block with time of activation, 2) the kinetics of current increase, and
3) the voltage dependence of block. Thus, almost no block of current was found at the beginning of the voltage pulse, and the block increased with time. The current increase with PROP was faster than
under control conditions, suggesting a blocking effect of PROP that
increases with time (Mergenthaler et al., 2001), although a direct
effect of PROP on channel kinetics cannot be ruled out. Finally, the
increase in block with positive-going potentials suggests a correlation
between block and open probability of the channel. Taken together,
these findings allow the conclusion that PROP blocks the Kv2.1 channel
in the open state. This is consistent with blocking mechanisms of PROP
found for Kv1.5 (Franqueza et al., 1998) and HERG channels
(Mergenthaler et al., 2001).
Intracellular Side of Action.
The conclusion that PROP
acts from the intracellular side of the membrane is based on two
findings: 1) the calculation of the electrical distance and 2) the
efficacy of PROP in excised inside-out patches.
The calculation of the fractional electrical distance () (i.e., the
fraction of the transmembranous electrical field sensed by a single
positive charge at the binding site) yielded a value of 0.11 to
0.14 for PROP. However, PROP can be assumed to be only partly charged
in the cytoplasm of the oocyte; this fact needs to be considered in
estimating the electrical distance. With a pK value of 8.8 for PROP
(Mergenthaler et al., 2001) and an intracellular pH value of 8.2 in the
oocyte (Kauder et al., 1991), only 80% of the intracellular
propafenone is positively charged. Assuming that PROP blocks the
potassium channel in both the charged and uncharged forms with the same
efficacy, a mean charge z = 0.8 could increase the
electrical distance to 1 = 0.16, 2 = 0.14, and 3 = 0.18. These values are similar to those found for TEA (internal
application; = 0.16; Choi et al., 1993), quinidine ( = 0.19; Snyders et al., 1992), and bupivacaine ( = 0.16;
Valenzuela et al., 1995), whose site of action has been attributed to
the internal mouth of the potassium channel based on the results of
mutagenesis experiments (Yellen et al., 1991; Yeola et al., 1996;
Franqueza et al., 1997).
In experiments with excised patches, we obtained a blocking effect
after rapid application to the intracellular side of the cell membrane
but not to the extracellular side. Because this experimental approach
minimized possible contributing effects of diffusion and/or transport
across the cell membrane, the findings suggest an action of PROP at
intracellular regions of the channel molecule.
It was not the aim of the present study to determine by which mechanism
PROP can enter the cell. The finding that the amount of block at the
Kv2.1 channel is independent of the time the channel is activated,
however, suggests that drug transport across the cell membrane is not
associated with channel activity; thus, structures of the channel
molecule are most probably not involved in this transport. A simple
diffusion process through the cell membrane, however, might be possible
and is in line with the quite slow increase in block with time (less
than 15% of block after 1 s of PROP application and 60% after
10 s; data not shown). Thus, uncharged PROP molecules (pK value
8.8) might enter the cell membrane compartment and might move
transversely into the intracellular space as charged molecules because
of pH-dependent equilibria and driven by the concentration gradients.
Such a mechanism is supported by the finding that an increase of
extracellular pH improved the presumably intracellular blocking
efficacy of PROP at the HERG channel (Mergenthaler et al., 2001).
Molecular Sites of Action of the Channel Molecule.
Several
regions have been found that when replaced by the corresponding parts
of the Kv1.2 channel, reduced the sensitivity of the Kv2.1 channel.
Replacement of the intracellular termini and the intracellular linker
S4-S5 induced moderate effects. Because there is growing evidence that
these parts of the channel form an intracellular compartment placed
near the membrane-associated part of the channel pore (hanging gondola
model; Kobertz et al., 2000), the intracellular termini and the S4-S5
linker might impair the access of PROP to its proper site of action in
the internal mouth of the channel.
A complete abolition of the PROP sensitivity of Kv2.1 (to the level of
Kv1.2) was found with exchanges of the linker S5-S6, as well as with
the S6 segment itself. Based on X-ray analysis of the KcsA potassium
channel, these parts can be assumed to form the inner wall of the pore
proper (Doyle et al., 1998).
Within the linker between segments S5 and S6, the replacement of amino
acids 372 to 374 by the corresponding residues of Kv1.2 abolished the
PROP sensitivity to the level of the Kv1.2 channel. If the amino acid
sequence of the Kv2.1 channel is transferred onto the structure of the
KcsA channel (Doyle et al., 1998), these residues form the
intracellularly orientated part of the pore helix (Fig.
7, black points). Although these residues
have no direct access to the internal vestibule of the analyzed KcsA
channel, which is most probably in a closed state, they are near the
vestibule and might be accessible after transition to the open state.
On the other hand, it cannot be excluded that the three-dimensional structure of the Kv2.1 channel differs slightly from that of KcsA channel, thus exposing these residues into the inner vestibule, or that
these residues only affect or shield the binding site.
View larger version (38K):
[in this window]
[in a new window]
|
Fig. 7.
Model of the pore region of the Kv2.1 channel derived
from the structure of the KcsA channel after Doyle et al. (1998). The
presentation of the structure is based on the Cn3D program (National
Center for Biotechnology Information, Bethesda, MD). The backbone
is shown lined with the schematic helices. One of the four subunits is
shown in red. The residues corresponding to the amino acids in Kv2.1
abolishing and reducing the sensitivity to propafenone are marked as
and , respectively. The numbers of these amino acids are given
on the right together with corresponding residues in other channels
affecting drug sensitivity to bupivacaine, cisapride (cis), dofetilide,
MK-499, quinidine, terfenadine (terf), tetrabutylammonium (TBA), TEA,
verapamil (Yellen et al., 1991; Choi et al., 1993; Yeola et al., 1996;
Franqueza et al., 1997, 1998; Ficker et al., 1998; Zhang et al., 1999;
Mitcheson et al., 2000; Zhou et al., 2001). The gray dot and arrow mark
residue 376, shown to be important for drug sensitivities in other
channels (Yellen et al., 1991; Choi et al., 1993; Yeola et al., 1996;
Franqueza et al., 1997; Mitcheson et al., 2000).
|
|
Studies of other substances on the HERG channel showed a reduction of
the sensitivity to dofetilide and verapamil when the serine residue 620 corresponding to residue 373 of the Kv2.1 was mutated (Ficker et al.,
1998; Zhang et al., 1999; Fig. 7). A neighboring residue corresponding
to amino acid 376 in Kv2.1 (Fig. 7, gray point) and forming the lowest
end of the pore helix in the KcsA channel (Doyle et al., 1998),
however, has been found to affect the sensitivity to several drugs,
including bupivacaine and quinidine (Franqueza et al., 1997; Yeola et
al., 1996) in the Kv1.5 channel and the antiarrhythmic drug MK-499 in
the HERG channel (Mitcheson et al., 2000). The region corresponding to
residues 372 to 374 in Kv2.1, which, in parallel to PROP action might
also affect the sensitivity to these drugs, has unfortunately not been probed.
Results of TEA support the hypothesis of slightly varying sites of
action for drugs in different channels. In the Shaker channel, amino
acids corresponding to residues 372 to 374 of Kv2.1 have not been found
to be associated with the sensitivity to TEA (Choi et al., 1993),
whereas strong effects appeared for the amino acids 440 and 441, corresponding to the residues 375 and 376 of Kv2.1 (Yellen et al.,
1991; Choi et al., 1993). The opposite is found for the Kv2.1 channel.
The residues 375 and 376 cannot be responsible for differences in TEA
sensitivity because the corresponding residues are identical in Kv1.2
and Kv2.1 channels (Fig. 6), whereas in our study, mutations of the
residues 372 to 374 strongly affected the TEA sensitivity. These data,
however, do not prove varying binding sites for TEA and PROP in the
Shaker and Kv2.1 channel because it cannot be excluded that the
residues 375 and 376 form the binding site, with nearby residues 372 to
374 allowing or preventing access of the drugs.
The second group of amino acids in the S5-S6 linker able to abolish the
sensitivity of Kv2.1 to PROP concerns the residues 383 and 384 (Fig. 7,
black points). The residue corresponding to the amino acid 384 in Kv2.1
has been found to affect the sensitivity to externally applied TEA in
Shaker (residue 449; Yellen et al., 1991). Correspondingly, the
mutation of these residues in Kv2.1 strongly reduced the effect of TEA
in the present study. Thus, it is likely that this region is important
for the blocking effects of both TEA and PROP. Based on the structure
of the KcsA channel, however, these residues can be assumed to be
located on the extracellular side of the channel molecule near the
outer channel mouth. This is not in agreement with our findings in
excised patches, because in these experiments, an application of PROP
to the extracellular side did not yield a blocking effect. Although the
reason for this discrepancy is not clear, that the site of mutation
might not necessarily be the site of drug binding has to be considered. Indirect effects such as changes in channel conformation (allosteric mechanisms) or opening kinetics might be induced, which could have
strong effects on drug action. Furthermore, although unlikely, it
cannot be excluded totally that these amino acids become accessible from the intracellular side with the opening of the pore. In summary, however, the data do not allow postulation of an external binding site
for PROP (or TEA) at the Kv2.1 channel.
We have found several amino acids of the intracellularly orientated
part of the segment S6 of the Kv2.1 channel that decrease the PROP
sensitivity strongly (residue 404, group 409/411/414 and group 417/418;
Fig. 7, ). Some of these amino acids affected the sensitivity to
quinidine, bupivacaine, TEA, MK-499, cisapride, terfenadine, and
tetrabutylammonium in other channels (Choi et al., 1993; Yeola et al.,
1996; Franqueza et al., 1997; Mitcheson et al., 2000; Zhou et al.,
2001). In these studies from the literature, however, other
mutations of S6 have also been described to affect drug sensitivity
(e.g., mutations of residues corresponding to residues 401, 402, 405, 407, 408, and 413). It cannot be concluded based on our data whether
the changes in S6 affect the binding site of PROP or the access of PROP
to the site of action. The changes of the fractional electrical
distance for some of these mutants, however, suggest that the binding
pocket for PROP might be impaired with residues facing the inner
vestibule of the channel.
Taking all the findings into consideration, we suggest that PROP enters
the cell by diffusion across the cell membrane and acts from the
intracellular side. PROP blocks the Kv2.1 channel in the inner
vestibule of the channel while it is in the open state. The access of
PROP to its binding site can be impaired by intracellular parts of the
channel (linker S4-S5, intracellular C and N terminus). Residues of
segment S6 might contribute to this impairment or to the formation of
the binding site, which most probably includes the lower part of the
pore helix.
We are thankful to Prof. Olaf Pongs for supporting us with cRNA
of the wild-type and chimeric channels, for fruitful discussions, and
for reviewing the manuscript. We are grateful to B. J. Corrette for English corrections.
PROP, propafenone;
nt, nucleotide;
CHO, Chinese
hamster ovary;
TEA, tetraethylammonium chloride;
HERG, human
ether-a-go-go-related gene;
MK-499, (+)-N-[1'-(6-cyano-1,2,3,4-tetrahydro-2(R)-naphthalenyl)-3,4-dihydro-4(R)-hydroxyspiro(2H-1-benzopyran-2,4'-piperidin)-6-yl]methanesulfonamide] monohydrochloride.
![[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
-subunit of voltage-gated potassium channel.
Recept Channels
6:
337-350[Medline].
