| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1-subunit
Department of Pharmacology and Program in Neuroscience, University of Tennessee Health Science Center, Memphis, Tennessee 38163
 |
ABSTRACT |
|---|
 TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Ethanol inhibition of large-conductance,
Ca2+-activated K+ (BKCa) channels
in aortic myocytes may contribute to the direct contraction of aortic
smooth muscle produced by acute alcohol exposure. In this tissue,
BKCa channels consist of pore-forming (bslo) and modulatory (
) subunits. Here, modulation of aortic myocyte
BKCa channels by acute alcohol was explored by expressing
bslo subunits in Xenopus oocytes, in the absence
and presence of 1-subunits, and studying channel
responses to clinically relevant concentrations of ethanol in excised
membrane patches. Overall, average values of bslo channel
activity (NPo, with N = no. of
channels present in the patch; Po = probability of a single channel being open) in response to ethanol
(3-200 mM) mildly decrease when compared with pre-ethanol,
isosmotic controls. However, channel responses show qualitative
heterogeneity at all ethanol concentrations. In the majority of patches
(42/71 patches, i.e., 59%), a reversible reduction in
NPo is observed. In this subset, the maximal
effect is obtained with 100 mM ethanol, at which
NPo reaches 46.2 ± 9% of control. The
presence of 1-subunits, which determines channel sensitivity to dihydrosoyaponin-I and 17-estradiol, fails to modify
ethanol action on bslo channels. Ethanol inhibition of bslo channels results from a marked increase in the mean
closed time. Although the voltage dependence of gating remains
unaffected, the apparent effectiveness of Ca2+ to gate the
channel is decreased by ethanol. These changes occur without
modifications of channel conduction. In conclusion, a new molecular
mechanism that may contribute to ethanol-induced aortic smooth muscle
contraction has been identified and characterized: a functional
interaction between ethanol and the bslo subunit and/or its
lipid microenvironment, which leads to a decrease in BKCa
channel activity.
maxi-potassium channel; alcohol; aorta; vasoconstriction
| |
INTRODUCTION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
ACUTE EXPOSURE TO clinically relevant concentrations of ethanol (EtOH) causes contraction of peripheral, cerebral, coronary, and umbilical arteries in several species, including humans (2-4, 23). There is substantial evidence that EtOH produces vasocontriction by acting on the smooth muscle itself (2, 26, 35, 55). However, the ion channel populations involved in EtOH action are largely unknown.
Large-conductance, Ca2+-activated K+ (BKCa) channel activity controls depolarization and contraction in vascular smooth muscle. Activation of BKCa channels gives rise to positive outward currents, which, in turn, drive the smooth muscle cell membrane potential in a negative direction and, eventually, counteract contraction (5, 8, 30). Therefore, EtOH inhibition of BKCa channels may be the mechanism underlying, or at least contributing to, alcohol-induced contraction of vascular smooth muscle.
BKCa channels consist of pore-forming (
, encoded by
slo genes) and modulatory () subunits (33,
51). Heterologous expression of slo channels produces
currents that bear most of the features of native BKCa
channels: large unitary conductance and selectivity for K+,
and Ca2+ and voltage dependence of gating. The presence of
-subunits shifts the voltage dependence of activation to more
negative potentials by increasing the apparent Ca2+
sensitivity of slo channels (1, 9, 41, 42) and
modifies channel pharmacology (41, 50).
Acute exposure to clinically relevant EtOH concentration ([EtOH]) reversibly increases BKCa channel activity in neurohypophysial terminals (20), GH3 cells (31), and small neurons in dorsal root ganglia, where alcohol action leads to a reduction in neuronal excitability (27). EtOH activation of BKCa channels is maintained when studied in isolated membrane patches of Xenopus oocytes expressing slo subunits cloned from mouse brain (mslo, mbr5 variant; see Refs. 11 and 18) or in lipid bilayers in which slo subunits from human brain (hslo; see Refs. 1 and 49) were reconstituted (12). These studies suggest that slo subunits and their immediate lipid microenvironment are sufficient targets for EtOH activation of neuronal BKCa channels.
Potentiation of BKCa channels by EtOH, however, is not a
universal finding. EtOH (10-100 mM) reversibly increases the
activity of BKCa channels in isolated membrane patches from
the nerve endings but not from the somata of supraoptic neurons
(22). BKCa channel phenotypes in these two
areas of the same neuron markedly differ in not only EtOH sensitivity
but also gating kinetics, Ca2+ sensitivity, ion conduction,
and charybdotoxin sensitivity (22). These differences are
likely determined by the presence of different slo isoforms
and/or -subunits in the two neuronal domains, which could also
contribute to differential EtOH sensitivity of the isochannels.
In contrast to its reversible potentiation of several BKCa
channels of neuronal origin, EtOH (5-80 mM) significantly inhibits the activity of native bovine aortic smooth muscle BKCa
channels after their reconstitution into artificial lipid bilayers
(52). It is quite possible that native BKCa
channels remain in these bilayers as heterooligomers formed by the
tight association of bslo (46) and
1-subunits, as found in their native smooth muscle membrane (24). Thus differential responses to EtOH between
native bovine aortic smooth muscle BKCa channels and
slo channels of neuronal origin when studied in cell-free
membranes or bilayers could be attributed to differential alcohol
response between bslo and other slo channels
and/or control of slo responses to EtOH by the proteolipid
microenvironment of the slo subunit, including introduction
of EtOH inhibition of bslo channel activity by the presence
of 1-subunits.
Here, a possible inhibition of bslo channel activity in
response to acute exposure to clinically relevant concentrations of EtOH was explored by expressing bslo subunits in
Xenopus oocytes and studying their alcohol responses in
excised membrane patches under conditions identical to those used to
evoke EtOH activation of mslo channels. In addition, a
possible modulation of bslo channel responses to EtOH by the
presence of 1-subunits was determined. Preliminary data
were published in abstract form (17).
| |
MATERIALS AND METHODS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
RNA preparation.
BKCa channels were expressed in Xenopus oocytes
by injection of cRNA transcribed in vitro from cDNA derived from the
Slowpoke locus of bovine aortic smooth muscle
(bslo; see Ref. 46). Briefly, bslo
cDNA inserted in the Kpn I/Not I sites of the
pcDNAIII expression vector was linearized with Xba I and
transcribed in vitro using T7 polymerase. BKCa channel
1-subunit cDNA inserted in the EcoR I/Sal I sites of the pCI-neo expression vector was
linearized with Not I and transcribed in vitro using T7
polymerase. Restriction enzymes were obtained from Promega (Madison,
WI). In vitro transcriptions were performed using mMessage mMachine
transcription kits (Ambion, Austin, TX). Full-length bslo
and BKCa channel 1-subunit cDNAs were generous gifts from Dr. Edward Moczydlowski (Department of Pharmacology, Yale University) and Dr. María García
(Merck Research Laboratories), respectively.
Use of Xenopus oocytes and RNA injection.
Xenopus laevis were purchased from commercial
breeders and kept in artificial pond water on a 12:12-h light-dark
cycle. Oocytes were removed and defolliculated, as previously described
(18). RNA injection was conducted 36-72 h before
patch-clamp recordings, using an automated nanoliter injector (Nanojet
II; Drummond Scientific, Broomal, PA). The 1-subunit
mRNA was coinjected with the (bslo)-subunit mRNA at a
concentration of 0.25 µg/µl each for a total volume of 36.8 nl.
These subunits were coinjected at a molar ratio >5:1 to ensure that
all bslo subunits were interacting with
1-subunits. BKCa channel activity after
bslo and 1 coexpression was recorded and
compared with that after injection of 36.8 nl (0.25 µg/µl) of the
bslo subunit alone in the same batch of oocytes. Before starting recordings, oocytes were placed in a dish containing a
hypertonic solution containing (in mM) 200 K-aspartate, 20 KCl, 1 MgCl2, 10 EGTA, and 10 HEPES, pH 7.4, for 10-15 min.
With this treatment, the oocytes shrunk, which facilitates the removal
of the vitelline layer with forceps, exposing the cell membrane for patch-clamp recording. Next, the oocytes were placed in ND-96 saline
containing (in mM) 96 NaCl, 2 KCl, 1.8 CaCl2, and 5 HEPES, pH 7.5, for 15 min before recording.
Data acquisition and analysis.
Single channel recordings were obtained from excised inside-out (I/O)
or outside-out (O/O) membrane patches using standard patch-clamp
techniques. Currents were recorded using a EPC8 patch-clamp amplifier
(List Electronics, Darmstadt, Germany), low-pass filtered at either 3 kHz (for dwell time analysis) or 1 kHz [for display and
NPo (N = no. of channels present
in the patch; Po = probability of a single
channel being open) determinations] using an eight-pole Bessel filter
(model 900C; Frequency Devices, Haverhill, MA), digitized at a sampling
rate of 10 kHz, and stored on an IBM-compatible computer hard drive.
Data acquisition and analysis were performed using pClamp software
(version 8; Axon Instruments, Union City, CA). An agar bridge
containing high-K+ gluconate solution (for composition, see
Table 1) was used as a ground electrode.
Patch pipettes were pulled from glass microcapillaries (Drummond
Scientific). The shank of each patch pipette was coated with Sylgard
184 (Dow Corning, Midland, MI) to reduce capacitance and noise.
Immediately before recording, the tip of each electrode was
fire-polished on a microforge (MF 200; World Precision Instruments, Sarasota, FL) to give resistances of 5-10 M
when filled with high-K+ solution (Table 1). All experiments were carried
out at room temperature.
|
). Values for
i were obtained from the Gaussian fit of all-points
amplitude histograms by measuring the distance between the modes
corresponding to the closed state(s) and the first opening level
(20).
In patches when N was low, its value could be determined
from the maximal number of unitary current levels under recording conditions that effectively raised Po to ~1.
This was achieved by holding I/O patches at positive potentials (+40 to
+60 mV) and applying solutions containing 30 µM free Ca2+
concentration to the cytosolic side of the patch. In single-channel patches, open time (to) and closed times
(tc) were measured with half-amplitude threshold
analysis. A maximum-likelihood minimization routine was used to fit
curves to the distribution of to and
tc. Determination of the minimum number of terms
for adequate fit was established using a standard F
statistic table (significance level <0.01). Changes in mean
to from multichannel patches of unknown
N were calculated as previously described (18).
NPo and dwell-time distribution data under a
particular experimental condition (drug exposure or isosmotic controls)
were calculated from periods of continuous channel recording that
typically ranged 20-60 s. For all experiments, voltages given
correspond to the potential at the intracellular side of the membrane.
Data are expressed as means ± SE. Statistical analysis was
performed with paired or unpaired Student's t-tests,
according to experimental design (25). Data plotting and
fitting were performed using Origin software (version 6.1; OriginLab,
Northampton, MA).
Solutions, chemicals, and construction of concentration-response
curves to EtOH.
Single-channel recordings were obtained from excised I/O or O/O patches
using a variety of solutions (Table 1). The extracellular solutions
used in I/O recordings (pipette solution) or O/O recordings (bath
solution) contained 
1 µM free [Ca2+], which helped
gigaohm seal formation. Calculation of the nominal free
[Ca2+] in these highly buffered solutions was done
according to the Max Chelator Sliders software (C. Patton, Stanford
University). For solutions with a nominal free [Ca2+]
300 nM, the actual free [Ca2+] was checked with a
Ca2+-selective and a reference electrode (models 476041 and
476416; Corning; see Table 1).
water (Milli-Q Water System; Millipore, Bedford,
MA). Deionized, 100% pure, EtOH (American Bioanalytical, Natick, MA)
was freshly diluted in bath solution immediately before recordings.
Urea was diluted in bath solution to the desired final concentration
from a concentrated stock solution (1 M) of ionic composition identical
to the final bath solution. Urea-containing solutions did not usually
have any effect on BKCa channel activity when compared with
urea-free bath solution, although in a very few cases some inhibition
of activity was observed. Dihydrosoyaponin-I (DHS-I; a generous gift
from Dr. Owen McManus, Merck Research Laboratories) and 17-estradiol
(Sigma Chemical) were diluted in bath solution to the desired final
concentrations from concentrated stock solutions (1 and 5 mM,
respectively) made with 100% DMSO. DMSO-containing solutions (final
DMSO concentration <0.001%) were used as controls for
17-estradiol- or DHS-I-containing solutions. Handling and
application of 17-estradiol and corresponding controls were
conducted in the dark.
After excision from the oocyte, the intracellular side of the I/O patch
was alternatively placed in the mouth of several "sewer" micropipettes (1 mm diameter; WPI), each delivering a solution that
contained the desired concentration of EtOH and free
[Ca2+]. Bath solution with urea isosmotically replacing
EtOH with the corresponding free [Ca2+] was used as a
control solution. This gravity-fed delivery system avoids the dilution
of agent-containing solutions in the bath solution that continuously
perfuses the dish where oocytes/patches are placed. For the
construction of concentration-response curves to EtOH action on
bslo channels in I/O patches, only "first applications" of EtOH (i.e., acute EtOH exposure to a naive patch) were considered. Thus different EtOH concentrations were applied to different patches.
For the construction of concentration-response curves to EtOH action on
bslo vs. bslo + 1 channels in
O/O patches, both control bath and EtOH-containing solutions were
applied to the extracellular surface of the same patch for 20-60
s, using an automated, pressurized, DAD-12 superfusion system (ALA
Scientific Instruments, Westbury, NY). NPo in
EtOH was compared with its value in control solution applied
immediately before EtOH. A similar method of solution application was
used for the construction of concentration-response curves to
intracellular Ca2+ (Ca
| |
RESULTS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Expression of bslo subunits in Xenopus oocytes renders unitary
current events that display all major features of BKCa
channels.
Single-channel current recordings obtained in I/O patches after
expression of bslo subunits in Xenopus oocytes
showed all major characteristics of BKCa unitary currents.
First, steady-state activity (NPo) reversibly
increased with increases in intracellular Ca2+
concentration ([Ca2+]i). At constant voltage
(e.g., +40 mV), when the cytosolic side of the same I/O patch was
consecutively exposed to solutions with a free [Ca2+] of
0.1, 0.3, and 1 µM, bslo channel
NPo reached 0.002, 0.008, and 0.9, respectively
(also see Fig. 4). Second, at constant free [Ca2+] at the
intracellular side of the patch, NPo increased
as the membrane potential was made more positive. From 
60 to +40 mV, the voltage dependence of activation could be described by a Boltzmann relationship, such that a plot of the natural log of
NPo (or Po) as a function
of voltage was linear at low values of Po. From this plot, the reciprocal of the slope is the potential needed to
produce an e-fold change in NPo
(20, 47). For bslo channels expressed in
oocytes, this reciprocal was 12.3 ± 1.6 mV/e-fold increase in NPo (n = 5; a
representative patch is shown in Fig. 3). This value is similar to
values reported for other slo channels expressed in
Xenopus oocytes (11, 16, 18, 28), after
reconstitution into artificial lipid bilayers (13), and
for arterial smooth muscle BKCa channels studied either in
situ (19, 43) or in lipid bilayers (52). From
the slope factor, the effective valence (z; see Ref.
48) was obtained: z = 2.6 ± 0.3 (n = 5), which indicates that a minimum of three
elementary charges are moved across the electrical field to gate the
bslo channel.
60 to +40 mV, with a
slope conductance of 259 ± 20 pS (n = 5). When
the bath solution was switched from 145 mM K+ to 100 mM
K+/45 mM Na+, the reversal potential shifted
from ~0 to 28 ± 4 mV (n = 3), almost identical
to the value predicted by the Nernst equation for a channel highly
selective for K+ over Na+ (i.e., 30.6 mV). This
combination of large conductance and high selectivity for
K+ over Na+ is a typical feature of the
slo pore (reviewed in Ref. 36).
The majority of bslo channels in excised, I/O patches respond to
acute EtOH exposure with a reversible decrease in steady-state activity. EtOH inhibition in this subset of channels is concentration dependent. The acute exposure of the cytosolic surface of I/O patches to 50 mM EtOH usually caused a reversible decrease in bslo channel steady-state activity compared with isoosmotic
controls (Fig.
1A),
confirming an early finding (21). This [EtOH] was initially chosen because it was reported to cause vasoconstriction in
most in vitro artery preparations (see references in 55) and to
decrease both amplitude and frequency of spontaneous transient outward
currents in aortic smooth muscle cells (56). EtOH action on bslo channel activity was routinely recorded 15 min after
patch excision, with both sides of the patch exposed to solutions
containing no nucleotides. Thus EtOH-induced channel inhibition is
neither secondary to changes in diffusible second messenger levels nor mediated by cellular processes such as GTP-binding protein modulation, NAD(P)H-dependent metabolism/modulation, etc. Furthermore, because bslo channel inhibition by EtOH was observed in I/O patches
with both sides of the patch exposed to highly buffered
[Ca2+] and at positive potentials, this inhibition is not
secondary to a possible decrease in [Ca2+] at the
cytosolic side of the patch (which would cause BKCa channel inhibition) resulting from putative EtOH inhibition of Ca2+
transmembrane influx. Thus EtOH inhibition of bslo channels
appears to result from a direct interaction of EtOH with these
BKCa subunits and/or some closely associated component of
the cell membrane.
|
-subunits are associated with the slo channel (see
below), it is unlikely that heterogeneity in bslo responses
to EtOH could be explained by the hypothetical presence of endogenous
-subunits in some patches and not in others.
The overall pattern of bslo responses to acute EtOH exposure
markedly differs from the responses to acute EtOH (3-200 mM) of
mslo (mbr5 isoform) channels after expression in
Xenopus oocytes under recording conditions identical to
those used in the present study. Reversible increases in the activity
of mslo (mbr5) channels (18) were found in 36 out of 37 I/O patches, the remaining being insensitive. Furthermore,
when bslo and mslo (mbr5) were expressed in the
same batch of oocytes and their activity was recorded in response to 50 mM EtOH exposure, differential responses were still observed:
mslo (mbr5) channel NPo was typically
increased to an average of 210 ± 21.2% of controls
(P < 0.005; paired, 2-tailed, Student's
t-test; n = 6, all patches being EtOH
sensitive), whereas bslo activity was decreased to an
average of 52.8 ± 4% of controls (P < 0.005;
paired, 2-tailed, Student's t-test; n = 5, where 5 patches out of 6 tested were EtOH sensitive). Thus differential responses to acute EtOH exposure between bslo and
mslo (mbr5) channels when inserted in a similar proteolipid
microenvironment may be linked to the existence of nonconserved regions
between these slo channel proteins (see
DISCUSSION).
The overall pattern of bslo channel responses to acute EtOH
exposure shown above (i.e., 59% of channels are inhibited, 8% are
potentiated, and 32% are insensitive) also markedly differs from the
responses evoked by similar [EtOH] acutely applied to native aortic
smooth muscle channels reconstituted in bilayers of controlled lipid
composition, where EtOH shows very high potency and efficacy in >90%
of bilayers (50). Next, we decided to determine EtOH
potency and efficacy on bslo channels from the selective subset of patches that are inhibited by the drug and compare these parameters of drug action with those obtained with native aortic BKCa channels. EtOH-induced inhibition of bslo
channel activity, as determined from the subset of sensitive patches,
was concentration dependent from 3 to 100 mM (Fig. 1C).
Maximal inhibition of bslo activity was obtained with 100 mM, at which average NPo in the presence of EtOH
reached 46.2 ± 9% of controls (P < 0.005;
paired, 2-tailed, Student's t-test; n = 4).
The highest concentration routinely tested, 200 mM, had no additional
effect over that obtained at 100 mM (Fig. 1C). This range of
concentrations at which EtOH causes a decrease in bslo
channel activity is similar to those reportedly causing vascular smooth
muscle contraction (2, 55) and inhibition of native bovine
aortic smooth muscle BKCa channels in artificial lipid
bilayers (52).
EtOH action on bslo channels is unmodified in the presence of
functional 1-subunits.
Data obtained with native BKCa channels in bilayers,
however, differ from current results with bslo channels in
two major aspects. First, 90% of native BKCa channels in
bilayers (52), but only 59% of bslo channels,
are inhibited by EtOH exposure; second, 7.5-10 mM EtOH are enough
to totally shut down the activity of native BKCa channels
in bilayers, while the maximal alcohol effect (achieved with 100 mM
EtOH) on bslo channels, determined from the subset of
sensitive patches (Fig. 1C), only decreases activity by
~54%. Thus the "effectiveness" (i.e., no. of patches wherein
BKCa channel activity was decreased/no. of patches tested), potency, and efficacy of EtOH action on native bovine aortic smooth muscle BKCa channels reconstituted into artificial lipid
bilayers are significantly higher than those from current EtOH data
obtained with bslo channels in I/O patches of natural
membranes. Because in the bilayer preparation the native
BKCa channel most likely exists as a heterooligomeric
bslo + 1 complex, it is conceivable that
the presence of modulatory 1-subunits in the
bilayer-aortic smooth muscle BKCa channel preparation
determines, or at least contributes to, the differences in EtOH effects
on native aortic BKCa channels in bilayers vs.
bslo channels expressed in oocytes. Thus a possible
modification of EtOH action on bslo channels in Xenopus oocytes by the presence of modulatory
1-subunits was explored next.
1-subunits associated with the
bslo channel before checking the EtOH sensitivity of
heterooligomeric channels. Extracellular application of 5 µM
17-estradiol increases BKCa currents in isolated
membrane patches of Xenopus oocytes that coexpress
hslo + -subunits but not in those solely expressing hslo subunits (50). Bath application of
17-estradiol (5 µM) to O/O patches significantly and reversibly
increased the NPo of heterooligomeric
bslo + 1 channels (Fig.
2A), which reached an average
of 184 ± 22% of control (P < 0.01; paired,
1-tailed, Student's t-test). This potentiation was observed
in all O/O patches tested (n = 9). In contrast, 5 µM
17-estradiol had no effect on the activity of homomeric
bslo channels in all O/O patches tested (n = 7; Fig. 2A). Thus introduction of 17-estradiol
sensitivity by the presence of -subunits is not restricted to
hslo + -subunit complexes but may represent a common
phenomenon among slo channels.
|
1-subunits
associated with bslo channels in the oocyte membrane was
determined by evaluating NPo changes in response
to the application of 100 nM DHS-I to the cytosolic side of I/O
patches, including patches excised from the same oocytes wherein
17-estradiol and EtOH effects were being tested. As found with the
application of 17-estradiol to the extracellular side of O/O
patches, application of DHS-I (100 nM) to the intracellular side of I/O
patches reversibly increased the NPo of
heterooligomeric bslo + 1 channels in
all cases (n = 9) while having no effect on the
activity of homomeric bslo channels in all cases
(n = 10; Fig. 2A). As previously reported with bovine tracheal smooth muscle BKCa channels
incorporated into planar lipid bilayers (40) and
mslo + complexes expressed in Xenopus
oocytes (41), DHS-I action was characterized by the introduction of intervals of high activity that lasted a few seconds separated by intervals of low activity similar to those in control (data not shown). This resulted in an overall increase of
severalfold in steady-state activity when recorded for longer periods
of time (typically >60 s), with NPo reaching an
average of 1,597 ± 576% of control (P < 0.05;
paired, 1-tailed, Student's t-test; n = 9; Fig. 2A, left). Control solutions (final
DMSO concentration <0.001%) for both 17-estradiol and DHS-I had no
effect on channel NPo (n = 35).
Acute exposure of O/O patches expressing homomeric bslo
channels to 3-100 mM EtOH resulted in a reversible reduction of
steady-state activity in some, but not all, cases (4 out of 7 O/O
patches, the other 3 being insensitive). For the construction of these concentration-response curves to EtOH, NPo in
the presence of a given [EtOH] was compared with
NPo under isosmotic control conditions recorded
immediately before alcohol application from the same O/O patch (see
MATERIALS AND METHODS). This
heterogeneity in EtOH action from patch to patch (i.e., only in 57% of
patches was bslo activity reduced by EtOH) was similar to
that previously found in a much higher number of I/O patches, where
inhibition was observed in 56% of 66 patches (Fig. 1B). A
maximal effect was obtained with 100 mM EtOH, at which bslo
channel NPo in O/O patches reached 49.2 ± 7% of control (P < 0.005; paired, 2-tailed,
Student's t-test). This value does not differ from that
obtained with I/O patches (46.2 ± 9%). These similarities in the
modulation of bslo channel activity by EtOH when the drug is
applied to either the cytosolic surface of I/O patches or the
extracellular side of O/O patches would be expected, since EtOH is a
small amphiphile thought to readily cross biological membranes to
access its site(s) of action.
Acute exposure of O/O patches expressing heterooligomeric
bslo + 1 channels to 3-100 mM EtOH
also resulted in a reversible reduction of NPo
in some, but not all, cases (5 out of 9 O/O patches; 55%). From
responsive patches, a maximal effect was obtained with 100 mM EtOH, at
which NPo reached 53.8 ± 8.9% of control
(P < 0.005; paired, 2-tailed, Student's
t-test). These findings are almost identical to those found
with excised membrane patches expressing bslo channels
alone. Furthermore, concentration-response curves from responsive
patches to EtOH action on channel NPo of bslo + 1 channels vs. bslo
channels demonstrate that EtOH potency is similar for heterooligomeric
and homomeric channels (Fig. 2B). Overall, this pattern of
alcohol action is practically identical to that found in a much larger
number of cases with bslo channels from I/O patches (Fig.
1C). Altogether, these data clearly indicate that the
presence of functional modulatory 1-subunits does not significantly modify EtOH action on bslo channel activity.
EtOH inhibition of bslo channel activity is primarily determined by alcohol-induced reduction in the channel mean tc. The possibility that EtOH causes a decrease in NPo by sequestering channel-forming bslo subunits from the patch membrane is unlikely because of the fast and reversible nature of the inhibition (Fig. 1A). In the vast majority of cases (65 out of 66 I/O patches, all 16 O/O patches), EtOH action on bslo activity was tested in multichannel patches of unknown N. Dilution of cRNA and/or lengthening the interval between cRNA injection and recordings rendered patches containing no channels or the usual multichannel patches. This may be explained by clustering of bslo channels in the oocyte membrane, which is similar to previous findings with mslo channel expression in this system (18). In the single case in which N was certainly equal to one (I/O patch exposed to 50 mM EtOH), the decrease in Po (46% of control) was within the range for the decrease in Po produced by this [EtOH] in multichannel patches of unknown N (Fig. 1B). Thus the decrease in Po produced by EtOH appears to totally account for EtOH-induced reduction in bslo channel NPo.
A decrease in channel Po by EtOH could be determined by a drug-induced decrease in the mean to and/or an increase in the mean tc. Calculated from EtOH-sensitive, multichannel I/O patches ([Ca2+]i
300
nM and V = 40 mV), to was found
mildly modified by alcohol exposure, reaching 87% of control values:
2.94 ± 0.46 and 2.55 ± 0.39 ms in the absence and presence
of 50 mM EtOH, respectively (P < 0.05; paired,
2-tailed, Student's t-test; n = 4). This
[EtOH], however, markedly reduced Po to 50.5%
of control values. Thus a major increase in the channel mean
tc is the main determinant of EtOH action on
Po.
Because multichannel patches of unknown N were found in the
vast majority of patches (81 out of 82 I/O or O/O patches), dwell-time distributions were not routinely analyzed. In a single case, the distribution of channel openings could be well fitted with two exponentials, with time constants of 1.5 and 9.7 ms, which contributed 77 and 23% to the total fit area (i.e., to the total time spent in
open states). In the presence of 50 mM EtOH, openings could also be
well fitted with two exponentials with time constants of 2.9 and 8 ms,
which contributed 95 and 5% to the total time spent in open states,
respectively. Thus, in the presence of EtOH, the channel population is
driven away from longer toward briefer open states. However, the mean
duration of brief events is increased, not decreased. Thus, when the
duration of the exponential components was weighted,
to in the presence of EtOH reaches only 91.2%
of control (3.4 and 3.1 ms in the absence and presence of 50 mM EtOH, respectively). This mild decrease in to
calculated from a single channel patch is similar to the average change
in to obtained from several multichannel patches
under identical conditions (see above).
In the absence of EtOH, the distribution of closed times in this single
channel patch could be well fitted with two exponentials, with time
constants of 1.8 and 35.7 ms and a contribution to the total fit area
(i.e., the total time spent in closed states) of 33 and 67%,
respectively. Thus the bslo channel exhibits at least two
open and two closed states, which is consistent with a previous model
proposed for other slo channels (16). In the
presence of 50 mM EtOH, time constants were 1.9 (22%) and 65.1 (78%)
ms. Thus EtOH slightly shifts the closed channel population toward longer closures and, more important, markedly increases the average duration of these long events. These EtOH-induced changes result in a
marked increase in tc from 24.5 to 51.2 ms
(209% of control). Therefore, EtOH decreases bslo
Po by causing a small decrease in
to and a marked increase in
tc that is secondary to alcohol-induced stabilization of channel long closed state(s).
EtOH modifies bslo channel gating without markedly affecting the voltage dependence of gating. EtOH targets specific channel dwell states, leading to a reversible reduction in Po (see above). In BKCa channels, whether native or cloned slo subunits, the gating process(es) that determine steady-state Po are controlled by transmembrane voltage (14, 16, 38). Thus we explored whether EtOH inhibition of bslo channel activity occurs by (or with) a drug-induced modification of the voltage dependence of channel gating.
The voltage dependence of channel activity was studied in I/O patches in the presence and absence of 100 mM EtOH, a concentration that caused the maximal effect on bslo channel NPo, as previously shown in Fig. 1C. A logarithmic plot of bslo channel NPo as a function of membrane voltage in the presence of EtOH shows a linear increase as potentials are made more positive at low Po values, as found in controls. Furthermore, the rate-limiting factor was essentially the same in the presence and absence of EtOH: 13.8 (r = 0.97) vs. 12.7 (r = 0.97) mV/e-fold change in NPo when evaluated in the same I/O patch (Fig. 3). Data from this and two other patches rendered an average slope factor of 10.8 ± 1.4 mV/e-fold change in NPo (z = 2.28 ± 0.29) in the presence of 100 mM [EtOH], which is not significantly different (paired, 2-tailed Student's t-test, n = 3) from the average value obtained in the absence of EtOH (see above). Thus EtOH produces a parallel shift in the NPo-voltage relationship toward positive potentials, the number of elementary charges moved across the electrical field to gate the bslo pore being essentially the same in the absence and presence of EtOH.
|
EtOH reduces the Ca2+ dependence of
bslo channel gating.
A parallel shift in the voltage activation curve of BKCa
channels, whether native (6, 44) or cloned from
slo genes (14, 16), toward positive potentials
is observed when the Ca
10 mV in the presence and
absence of 100 mM EtOH, evaluated in the same I/O patch (Fig.
4), indicates that EtOH decreases the
"apparent" Ca2+ sensitivity of bslo
channels, since channels in the presence of EtOH require more cytosolic
free Ca2+ for any given level of activity. Furthermore,
EtOH decreased the slope of the
NPo-[Ca2+]i
relationship: 1.19 (r = 0.99) vs. 0.84 (r = 0.97) in the absence and presence of 100 mM EtOH.
The slope value in the absence of EtOH suggests that bslo
channel activation requires either the strong cooperative binding of
two Ca2+ ions or the weaker cooperativity of more than two
Ca2+ ions. These data are consistent with previous findings
obtained in describing the Ca2+ dependence of channel
gating after injection of dslo or hslo (16) and mslo subunits (18) in
Xenopus oocytes. Similar data to those shown in Fig. 4 were
obtained in three other patches from different oocytes, which rendered
average slopes of 1.06 ± 0.15 vs. 0.76 ± 0.12 in
the absence and presence of 100 mM EtOH (n = 4;
P < 0.05, paired, 2-tailed, Student's
t-test). This decrease in slope suggests that EtOH alters
the capability of the Ca2+-sensing site(s) to respond to
increases in [Ca2+]i. These data also
indicate that EtOH inhibition of bslo activity is a direct
function of [Ca2+]i.
|
EtOH modifies bslo channel steady-state activity without affecting basic ion conduction properties of the bslo pore. The association of large unitary conductance and high selectivity for K+ over Na+ is a typical feature of the slo pore (36), which allows BKCa channels to contribute effectively to membrane repolarization in vascular smooth muscle. The possibility that EtOH-induced reduction in bslo NPo was accompanied by drug-induced modifications in channel ion conduction properties was evaluated next.
All-points amplitude histograms of data collected at10 mV in the
absence and presence of a maximally inhibitory concentration of EtOH
show that alcohol, while decreasing channel NPo
from 0.071 to 0.030, did not change the unitary current amplitude of
the channel [2.66 ± 0.43 vs. 2.64 ± 0.57 pA, in the
absence and presence of 100 mM EtOH (Fig.
5A)]. From these histograms,
unitary current values at different potentials were obtained for
constructing i/V plots. The
i/V relationship of averaged data obtained from I/O patches in symmetric 145 mM [K+] in the presence and
absence of 100 mM EtOH indicates that alcohol neither modified the
reversal potential (near 0 mV) nor introduced any rectification or
shift from 60 to +40 mV (Fig. 5B). The slope conductance,
obtained by linear regression of data, was 255 ± 22 pS
(n = 3), a value similar to that found in controls (see above).
|
| |
DISCUSSION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
An EtOH-bslo subunit functional interaction leads to
alcohol inhibition of aortic smooth muscle BKCa channels.
Data presented here demonstrate that EtOH inhibits BKCa
channel activity in the majority of membrane patches after expression
of bslo subunits in Xenopus oocytes. EtOH action
on channel function was studied in excised membrane (I/O and O/O)
patches in solutions containing no nucleotides 10-15 min after
patch excision from the cell. Thus EtOH inhibition of bslo
channels is not mediated by cell metabolism or freely diffusible
cytosolic messengers. Rather, it results from a direct interaction
between EtOH and bslo subunits and/or their immediate
proteolipid environment. Furthermore, the presence of modulatory
1-subunits, while introducing sensitivity to DHS-I and
17-estradiol, failed to affect the EtOH sensitivity of the bslo channel, stressing the fact that bslo
subunits seem sufficient for alcohol action.
Differential EtOH sensitivity of bslo channels expressed in oocyte
membranes vs. native aortic smooth muscle BKCa channels
reconstituted into artificial lipid bilayers.
EtOH (5-60 mM) inhibition of native aortic BKCa
channels incorporated into artificial bilayers (52) is
characterized by higher effectiveness, potency, and efficacy compared
with data presented here with bslo channels in excised
patches of oocyte membranes. These differences may be potentially
attributed to the presence of modulatory 1-subunits in
the bilayer preparation, different recording conditions (in particular,
[Ca2+]i), the presence of modulatory lipid
species in the oocyte membrane that are lost in the artificial bilayer,
and/or the existence of polymorphism in the slo subunit of
aortic BKCa channels. The lack of effect of
1-subunit expression on EtOH action on bslo channel activity (Fig. 2B) suggests that these modulatory
subunits unlikely contribute to the differential EtOH sensitivity
between native bovine aortic BKCa and bslo channels.
2.5 µM (52)
and the bulk of our results from I/O patches were obtained at
[Ca2+]i 300 nM (Fig. 1B), it
is possible that the different levels of Ca1-subunits in isolated oocyte membrane patches should be
focused on the lipid microenvironment of the slo subunit and/or the putative existence of polymorphism in slo
subunits of aortic BKCa channels.
Interestingly, we have recently demonstrated that EtOH potentiation of
hslo (hbr1 variant; see Ref. 49) channels
incorporated in planar lipid bilayers is modulated by different
membrane lipid species. Increases in bilayer cholesterol content or
decreases in bilayer phosphatidylserine, in particular, blunt EtOH
action on hslo activity (13). Thus the presence
of modulatory lipid species in the oocyte membrane, but absence in an
artificial bilayer made of phosphatidylethanolamine and
phosphatidylcholine (52), might contribute to the
diminished alcohol action found with bslo channels in the
oocyte membrane.
Natural membranes contain lipid microdomains with a lipid composition
and physical properties different from the bulk membrane (10,
54), where slo channels have been reported to cluster (7). The association of bslo channels with
distinct lipid species in the oocyte membrane may also help to explain
the probable clustering of functional slo channels in the
oocyte membrane noted here with bslo and previously
(18) with mslo channels. Heterogeneity among these lipid microdomains in the oocyte membrane might also contribute to the intraoocyte variability in the EtOH responses of a homogenous population of cloned channel proteins (Fig. 1B).
A major role for the proteolipid microenvironment of an ion channel
protein as probable modulator of EtOH action on channel function as
postulated here may not be restricted to BKCa but extended
to other ion channels. For example, EtOH potentiation of
GABAA-mediated currents was observed with
122S-subunits when
expressed in Xenopus oocytes but not when the same subunit combination was expressed in HEK cells (reviewed in Ref.
45).
Differential and common responses to EtOH between bslo and other slo channels. One striking finding of the present work is the differential responses of bslo vs. mslo (mbr5) channels to the same range of [EtOH] when studied in the excised oocyte membrane under identical recording conditions. These differences are maintained even when bslo and mslo channel responses are evaluated in the same batch of oocytes. Overall, alcohol activation of mslo channels was found in 42 out of 43 patches (98%), the remaining patch being nonresponsive, while activation of bslo channels was observed only in 6 out of 71 patches (8%), the remaining being either inhibited (58%) or nonresponsive (34%). It is possible that nonconserved regions between bslo and mslo (mbr5) contribute to their differential responses to EtOH, whether by directly interacting with this compound, by "sensing" EtOH action at lipid-protein interfaces, and/or by being differentially targeted after posttranslational processing that modifies EtOH action. Bslo and mslo (mbr5) are 98% identical proteins (11, 46), the least conserved area being the COOH-terminal region. This region is represented by a sequence of 60 amino acids, the first 8 of which are substituted, and the remaining absent in bslo, resulting in a shorter COOH terminus. Thus the longer COOH-terminal end in the mslo protein might be responsible for EtOH potentiation of BKCa channel activity. Alternatively, differential EtOH sensitivity between bslo and mslo (mbr5) could be linked to other nonconserved regions, such as three residues in the linker between S8 and S9.
Bslo and other slo channel responses to EtOH, however, share strong similarities. Despite causing qualitatively opposite effects (i.e., inhibition and activation), in all cases EtOH acts at similar concentrations and modulates NPo primarily by targeting channel long closed states. Furthermore, EtOH-induced changes in NPo occur without major modification of the voltage dependence of gating and the characteristic high-K+ permeability and selectivity over Na+. These data indicate that EtOH does not modify functional properties associated with the slo channel "core" domain [S1-S8 (53) or S0-S8 (51)], a region highly conserved between mslo (mbr5) and bslo subunits. Finally, for both mslo and bslo channels, EtOH modulates activity with a reduction in the Ca2+ dependence of channel gating. For both channels, the slope of the ln(NPo-[Ca2+]i) relationship is shifted from control values greater than one to values below one in the presence of EtOH. Because EtOH is unlikely to change the number of slo subunits (see RESULTS and also Ref. 18), this reduction in slope suggests that alcohol introduces apparent negative cooperativity in the interaction between Ca2+ in the cytosolic side of the patch and Ca2+-sensing site(s) near or in the slo subunit. Thus the EtOH-induced decrease in the Ca2+ dependence of slo channel gating may be determined by a reduced availability of Ca2+-binding sites without a decrease in the number of channel proteins, introduction of negative cooperativity in the actual binding of Ca2+ to a conserved number of Ca2+ recognition sites, and/or alteration in the Ca2+ dependence of channel gating per se (i.e., EtOH modifies steps subsequent to Ca2+ binding). EtOH action on mslo (mbr5) channel activity and its Ca2+ dependence is best described by a model in which EtOH behaves as a partial agonist of the mslo channel, for which CaPathophysiological implications of the present results.
There is wide evidence that changes in vascular tone in response to
acute alcohol exposure are primarily the result of EtOH action on the
vascular smooth muscle itself (2, 26, 34, 35, 55). At this
level, EtOH-induced relaxation and contraction have both been reported,
depending upon [EtOH] and, primarily, vessel type (2, 3,
34). In vitro exposure to clinically relevant [EtOH] produces
contraction of aortic smooth muscle (32, 37, 39, 55), this
effect being partially mediated by protein kinase C (32,
55) and a caffeine-sensitive pool of Ca
| |
ACKNOWLEDGEMENTS |
|---|
I gratefully acknowledge Dr. Joshua J. Singer for critical reading of the manuscript, Dr. Weihua Huang for advice and help with cDNA cloning and linearization, Dr. Suleiman Bahouth for helpful discussion, and Maria Asuncion-Chin for excellent technical assistance.
| |
FOOTNOTES |
|---|
This work was supported by National Institute on Alcohol and Alcohol-Related Diseases Grant AA-11560.
Address for reprint requests and other correspondence: A. Dopico, Dept. Pharmacology, Univ. Tennessee Hlth. Sci. Ctr., 874 Union Ave., Memphis, TN 38163 (E-mail: adopico{at}utmem.edu).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
First published February 5, 2003;10.1152/ajpcell.00421.2002
Received 13 September 2002; accepted in final form 30 January 2003.
| |
REFERENCES |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
1.
Ahring, PK,
Strobæk D,
Christophersen P,
Olessen SP,
and
Johansen TE.
Stable expression of the human large-conductance Ca2+-activated K+ channel alpha- and beta-subunits in HEK293 cells.
FEBS Lett
415:
67-70,
1997[ISI][Medline].
2.
Altura, B,
and
Altura B.
Microvascular and vascular smooth muscle actions of ethanol acetaldehyde, and acetate.
Fed Proc
41:
2447-2451,
1982[ISI][Medline].
3.
Altura, B,
and
Altura B.
Peripheral vascular actions of ethanol and its interaction with neurohumoral substances.
Neurobehav Toxicol Teratol
5:
211-220,
1983[ISI][Medline].
4.
Altura, B,
Li Y,
Altura B,
Jelicks L,
Wittenberg B,
and
Gupta R.
Beneficial vs. detrimental actions of ethanol on heart and coronary vascular sm
) vs. bslo
channels (
), obtained from responsive patches,
demonstrate that EtOH potency is similar for heterooligomeric and
homomeric channels. EtOH effectiveness and efficacy were also identical
between heterooligomeric and homomeric complexes (see
RESULTS). Thus the presence of
functional 
) and absence
(
) of a maximal inhibitory concentration of EtOH (100 mM), obtained in the same I/O patch. Given a fixed intracellular
Ca2+ concentration ([Ca2+]; in this case free
[Ca2+] 
