Modulation of K / Channels by Intracellular ATP in Human
Neocortical Neurons
CHUN JIANG 1,2 AND GABRIEL G. HADDAD 1
1
Department of Pediatrics, Section of Respiratory Medicine, Yale University School of Medicine, New Haven,
Connecticut 06520, and 2 Department of Biology, Georgia State University, Atlanta, Georgia 30322-4010
change in intermediary metabolism to membrane excitability. During certain metabolic stress such as hypoxia and
hypoglycemia, this ATP-mediated modulation of membrane
excitability becomes specially important in cells that depend
strongly on an exogenous energy supply such as central
neurons. A decrease in membrane excitability resulting from
the activation of ATP-inhibited K / (KATP ) channels can
slow down energy consumption and thus may influence cell
survival during energy deprivation (Haddad and Jiang
1993).
Because these KATP channels seem to play a role in modulating cellular membrane excitability during a decrease in
intracellular ATP level, detailed studies of these channels
may have implications on several neurological diseases or
conditions. However, such studies have rarely been carried
out in central neurons (Ashford et al. 1988, 1990a; Jiang et
al. 1994; Ohno-Shosaku and Yamamoto 1992), in contrast
with a rich documentation in a number of peripheral tissues
(Ashcroft et al. 1984, 1990a; Cook and Hales 1984; Noma
1983; Spruce et al. 1987; Standen et al. 1989; Wang and
Giebisch 1991). In fact, we do not even know if these KATP
channels are present in human central nervous system
(CNS). To determine whether ATP-modulated K / channels
are present in human neurons and study their properties, we
performed patch-clamp experiments and took advantage of
a rather rare opportunity to obtain human neocortical tissue
from neurosurgery. In addition to human neurons, we studied
rat neocortical neurons to see if ion channels with similar
properties are present in both species. Our results have not
only demonstrated that KATP channels are present in human
neocortical neurons, but also have shown that physiological
concentrations of ATP modulates coexisting K / channels
differently in both human and rat CNS.
METHODS
INTRODUCTION
Adenosine triphosphate (ATP) is an important metabolite
and an inter- and intracellular messenger (Ashcroft 1988;
Quayle and Standen 1994; Standen 1992; Surprenant et al.
1995; Takeno and Noma 1993). Among a number of target
molecules that are modulated by intracellular ATP are ion
channels whose activity determines the overall plasma membrane excitability and cellular responses to external stimuli
(Hille 1992). The modulation of ion channel activity by
ATP in excitable cells appears to have major physiological
significance, because it enables a direct coupling of the
Methods for dissociation of human neocortical neurons were
similar to those previously described (Cummins et al. 1993; Jiang
and Haddad 1992). Briefly, a small portion of human cortical tissue
was obtained from patients with epilepsy who had suffered from
recurring seizures identified by long-term electroencephalograph or
corticoencephalograph. The tissue was obtained from the anterior
portion of the temporal lobe (mostly from the middle temporal
gyrus). The temporal tissue obtained for these studies was removed
solely to gain access to the hippocampus and was judged to have
normal excitability from intracranial chronic depth and subdural
electrode recordings and/or intraoperative cortical surface electrocorticography. Previous studies from this laboratory and others
have demonstrated that nerve cells in such excised human neocorti-
0022-3077/97 $5.00 Copyright q 1997 The American Physiological Society
/ 9k0b$$ja25
J-061-6
08-13-97 17:53:12
neupa
LP-Neurophys
93
Downloaded from http://jn.physiology.org/ by 10.220.32.246 on June 17, 2017
Jiang, Chun and Gabriel G. Haddad. Modulation of K / channels
by intracellular ATP in human neocortical neurons. J. Neurophysiol. 77: 93–102, 1997. ATP-modulated K / channels play an important role in regulating membrane excitability during metabolic
stress. To characterize such K / channels from the human brain,
single channel currents were studied in excised inside-out patches
from freshly dissociated human neocortical neurons. Three currents
that were sensitive to physiological concentrations of ATP and
selectively permeable to K / were identified. One of these currents
had a unitary conductance of Ç47 pS and showed a strong inward
rectification with symmetric K / concentrations across the membrane. This K / current was inhibited by ATP in a concentrationdependent manner with an IC50 (half-inhibition of channel activity)
of Ç130 mM. Channel activity also was suppressed by ADP, nonhydrolyzable ATP analogue AMP-PNP, and sulfonylurea receptor/
channel blocker glibenclamide. The second K / current had a unitary conductance of Ç200 pS and showed a weak inward rectification. Similarly, this current was inhibited by ATP (IC50 Å 350
mM), AMP-PNP, and glibenclamide. Unlike the small-conductance
ATP-inhibitable K / channel (S-KATP ), activation of this largeconductance K / channel (L-KATP ) required the presence of micromolar concentration of Ca 2/ in the internal solution, but charybdotoxin did not inhibit this channel. The third K / current was also
Ca 2/ dependent and had a large conductance ( Ç280 pS). It was
inhibited by external charybdotoxin, iberiotoxin, and tetraethylammonium. In contrast to the other two KATP channels, ATP enhanced
channel open-state probability and unitary conductance, and glibenclamide at concentration of 10–20 mM had no inhibitory effect
on this current. K / channels that have single-channel and pharmacological properties similar to these three human ATP-modulated
K / channels also were observed in experiments on rat neocortical
neurons. These results therefore indicate that KATP channels are
expressed in human neocortical neurons, and two distinct KATP
channels (S-KATP and L-KATP ) exist in the human and rat neurons.
The observation that ATP at different concentrations modulates
different K / channels suggests that metabolic rate may be continuously sensed in neurons with resulting alterations in neuronal membrane excitability.
94
C. JIANG AND G. G. HADDAD
/ 9k0b$$ja25
J-061-6
FIG . 1. Effect of levcromakalim and glibenclamide on whole cell currents in a human neocortical neuron. A perforated patch was used in recording of whole cell currents with membrane potential held at 060 mV.
Solutions of identical concentrations of K / (150 mM) were applied to both
bath and pipette. These solutions contained low Na / (5 mM) plus Co 2/
(1 mM). Thus Na / and Ca 2/ currents were inhibited. BRL 38227 (levcromakalim, 5 mM), a KATP channel activator, markedly increased inward
currents, and this effect was blocked by glibenclamide (20 mM), a sulfonylurea KATP blocker. Note that effect of glibenclamide was seen in the presence
of same concentration of BRL 38227.
Hz) with a Gaussian filter. Events õ200 ms were ignored. The
unitary conductance was calculated by measuring current amplitudes in the I/V plot using a slope voltage command. The openstate probability (Popen ) was calculated as described previously
(Davies et al. 1992; Jiang and Haddad 1994a,b; Jiang et al. 1994;
Quayle et al. 1988; Sigworth and Sine 1987; Standen et al. 1989):
1) Popen values were calculated from stretches of data (obtained
using the Fetchex software) having a total duration of 18–36 s.
The time, t j , spent at current levels corresponding to j Å 0,1,2,. . .N
channels open was first measured. The Popen was then expressed
as a function of tj
N
Popen Å ( ∑ tj j )/TN
jÅ1
where N is the number of channels active in the patch and T is the
duration of recordings. 2) For segmented data (obtained using the
Clampex software), stimulating artifact and leak currents were first
removed by subtracting these currents from each traces of records,
and 50–60 records then were averaged. Popen values were calculated
as Popen Å I/ni, where I is the amplitude of averaged currents, n
the number of channels active in the patch, and i the amplitude of
single channel current.
Chemicals were all purchased from Sigma Chemicals (St. Louis,
MO), except charybdotoxin and iberiotoxin, which were obtained
from RBI (Natick, MA) and levcromakalim (BRL 38227) which
was a gift of SmithKline Beecham Pharmaceuticals.
Data are presented as means { SD (n Å number of patches)
and differences in means were tested with the Student’s t-test and
the x 2 test and accepted as significant if P ° 0.05.
RESULTS
Glibenclamide- and levcromakalim-sensitive K / currents
in neocortical neurons
Whole cell currents were studied under voltage-clamp
mode with the membrane potential held at 060 mV. Symmetric K / concentrations (150 mM) were used in both bath
and pipette solutions. Na / and Ca 2/ currents were abolished
using low Na / (5 mM) solution plus 1 mM Co 2/ in the
bath. Figure 1 shows an example of the recording of whole
08-13-97 17:53:12
neupa
LP-Neurophys
Downloaded from http://jn.physiology.org/ by 10.220.32.246 on June 17, 2017
cal tissue are electrophysiologically and histologically normal
(Cummins et al. 1993; Jiang and Haddad 1992). Neocortical tissue
was removed by neurosurgeons (Drs. Dennis D. Spencer and
Charles Duncan) and a portion ( Ç0.5–1.0 cm3 ) of the human
brain tissue was placed in ice-cold artificial cerebrospinal fluid
solution (0–17C) containing (in mM) 124.0 NaCl, 2.5 KCl, 1.25
NaH2PO4 , 2.0 MgSO4 , 2.0 CaCl2 , 26.0 NaHCO3 , and 10.0 Dglucose; the solution was oxygenated with 95% O2-5% CO2 (pH
7.4) and rapidly transferred to the laboratory where the tissue was
then sectioned into 300 mm slices. The sections were incubated for
1 h with oxygenated N-2-hydroxyethylpiperazine-N *-2-ethanesulfonic acid (HEPES) buffer containing (in mM) 140 NaCl, 2.5
KCl, 1 MgCl2 , 1 CaCl2 , 25 D-glucose, 10 HEPES, and trypsin
(0.2–0.3%, Sigma type XI) at 357C (pH 7.40). Sections then were
washed in oxygenated HEPES-buffer and maintained for °6 h.
Neurons also were harvested from the temporal cortex of
Sprague-Dawley rats (10–20 days). The rats were killed with
inhalation of saturated methoxyflurane and decapitated. The brain
was removed rapidly, chilled in 0–17C Ringer solution and prepared as a tissue block. The tissue block from the temporal cortex
was sectioned transversely into 300-mm slices. These sections then
were incubated with a tyrosine-containing buffer as described
above.
Immediately before recording, individual tissue slices (human
and rat) were removed and placed in a low Na / (5 mM) HEPES
buffer. Using a dissecting microscope, the gray matter (presumably
including layers III, IV, and V) was cut free from the rest of the
slice. The tissue of interest then was dissociated by gentle trituration with fire-polished Pasteur pipettes. Cells were plated in 35mm Petri dishes and observed with Hoffman modulation optics.
Recordings were obtained only from cells that had a pyramidal
shape of soma with a single thick proximal dendrite (presumably
pyramidal neurons) and did not show visible evidence of injury.
Flat or swollen cells or cells with a grainy membrane appearance
were not studied as previously described (Cummins et al. 1993;
Jiang and Haddad 1994a,b; Kay and Wong 1986).
Patch-clamp experiments were performed at room temperature
( Ç247C). Fire-polished patch pipettes (2–4 MV ) were made from
1.2 mm borosilicate capillary glass using a Sutter P-84 puller (Sutter Instrument, Novato, CA). Whole cell currents were studies in
the voltage-clamp mode, and single channel currents were recorded
from inside-out patches (Hamill et al. 1981) using an Axo-Patch
200A amplifier (Axon Instruments, Foster City, CA). To study
whole cell currents, perforated patches were performed as described previously (Jiang et al. 1994). The tip of the recording
pipette was filled with a solution containing (in mM) 150 KCl,
0.2 MgCl2 , 10 D-glucose, and 10 HEPES (pH 7.4), and the remainder of the pipette was filled with the same solution plus 100 mg/
ml nystatin (Sigma, stock solution 25 mg/ml in dimethylsulfoxide). After formation of a gigohm seal, perforation was monitored
by measuring series resistance with an Axo-Patch 200A amplifier.
Patches were accepted when the series resistance was õ50 MV.
Current records were low-pass filtered (Bessel, 4-pole filter, 03
dB at 2 kHz), digitized (10 kHz, 12-bit resolution) with pClamp
5.5.1 software (Axon Instruments), and stored on computer disk
for later analysis (Jiang and Haddad 1994a,b; Jiang et al. 1994).
For single channel studies, identical solutions were applied to
the bath and recording pipettes, and these contained (in mM) 150
KCl, 1 ethylene glycol-bis( b-aminoethyl ether)-N,N,N *,N *-tetraacetic acid (EGTA), 0.2 ADP (Mg 2/ salt), 10 D-glucose, and 10
HEPES (pH Å 7.4, free Ca 2/ õ 10 nM). In some experiments,
CaCl2 was added to the internal solution to achieve a free Ca 2/
concentration of Ç1 mM at pH 7.4. A perfusion system was used
to administer agents in parallel to patches at a rate of Ç1 ml/min.
This perfusion system allows a fast exchange of perfusion solutions
in Ç200 ms.
Single-channel data were analyzed by further filtering (0–1,000
ATP-MODULATED K / CHANNELS IN HUMAN NEURON
PNP (0.5–1.0 mM, n Å 4), also inhibited this channel when
applied to the internal surface. Similar to KATP channels
described in other tissues, this ATP-inhibited K / channel
in human neocortical neurons showed an ADT-dependent
rundown. Channel activity disappeared within 1–2 min after
patch excision from the cell unless Mg-ADP was present
in the internal solution. Activation of this K / channel is
independent of free Ca 2/ in the cytosol because this current
was observed using a Ca 2/ -free internal medium (free
Ca 2/ õ 10 nM, n Å 9) and an increase in free Ca 2/ to
micromolar concentration in this medium did not increase
channel activity (n Å 2). In addition, this K / channel was
inhibited by internal Cs / (1 mM, n Å 2).
An inward rectifying K / channel that had a similar conductance (46 { 4 pS, n Å 7) as the human K / channel
described above also was found in rat neocortical neurons.
This K / channel was inhibited by glibenclamide (n Å 3),
AMP-PNP (n Å 5; Fig. 4) and ATP in a concentrationdependent manner (Fig. 5). Although BRL38227 did not
Small conductance ATP-inhibitable K / channel
Single channel activity was studied with equal K / concentrations (150 mM) and Na / -free (or with 5 mM Na / in 2
patches) pipette and bath solutions. Using these solutions,
a small-conductance current was seen. This channel was
observed in nine patches from Ç80 human neocortical neurons studied (1 patch per cell). Figure 2 shows an example
of the effect of ATP on this small-conductance current in a
human neocortical neuron. This current had a high frequency
of openings and closures with a baseline Popen Å 0.299.
Channel activity was inhibited strongly and rapidly when
the patch was exposed to 1 mM of ATP (K / salt), and Popen
was reduced to 0.026. The inhibition by ATP was removed
rapidly after washout (Popen Å 0.263). Figure 3 shows an
example of K / selective permeability of this channel in a
human neocortical neuron. With equal concentrations of K /
(150 mM) applied to both internal and external solutions,
reversal potential of the current was 0 mV with a clear inward
rectification (Fig. 3A). Single channel conductance, which
was almost linear from 010 to 090 mV, averaged 47 { 3
pS (n Å 9). When the KCl concentration on the internal
surface was reduced to 54 mM (96 mM KCl was replaced
by the same amount of NaCl), the unitary conductance decreased to Ç40 pS, and the reversal potential was shifted to
23.4 { 1.0 mV, n Å 3 (Fig. 3B), which is close to the
calculated Nernst potential for K / (25.6 mV). Thus this
demonstrates that the channel is K / selective.
ATP inhibited this channel in a concentration-dependent
manner. Channel activity started to respond to ATP concentrations on the cytosolic side at 30 mM and was abolished
almost totally by 1.0 mM. The concentration of ATP that
produced half-inhibition of channel activity (IC50 ) was 130
mM. Channel activity also was inhibited significantly by
glibenclamide (10–20 mM, n Å 3), a sulfonylurea receptor/
channel blocker (Fig. 3C). Other nucleotides such as ADP
(1 mM, n Å 2) and nonhydrolyzable ATP analogue, AMP-
/ 9k0b$$ja25
J-061-6
FIG . 2. Small-conductance KATP current from a human neocortical neuron. Single channel currents were recorded from an inside-out patch using
symmetric concentrations of K / (150 mM). Membrane potential was held
at 0100 mV. Three channels were active during baseline. In absence of
ATP in internal solution, these currents had a channel open probability
(Popen ) of 0.299 with fast openings and closures. Channel activity was
inhibited strongly and rapidly when the patch was exposed to 1 mM of
ATP (K / salt; Popen Å 0.026). Washout of ATP resulted in a recovery of
channel activity to almost baseline level (Popen Å 0.263). Note that bottom
trace in each pair of traces is obtained from a segment of top trace between
2 arrows. Time scale calibrations are indicated for top and bottom, respectively.
08-13-97 17:53:12
neupa
LP-Neurophys
Downloaded from http://jn.physiology.org/ by 10.220.32.246 on June 17, 2017
cell currents in a perforated patch from a human neocortical
neuron. BRL 38227 (levcromakalim, A KATP channel activator) at 5 mM markedly increased the whole cell inward current in this cell. The increase in inward current was inhibited
by subsequent application of glibenclamide (20 mM), a sulfonylurea receptor blocker, to the bath solution (Fig. 1).
These glibenclamide- and BRL38227-sensitive currents are
not Na / currents, because the inward Na / currents were
inhibited strongly in our Na / -free or low Na / solutions
(Jiang et al. 1994) and because these currents were activated
with hyperpolarization rather than depolarization.
To characterize these levcromakalim and glibenclamide
sensitive currents, single channel currents were studied in
excised inside-out patches. Recording from giant inside-out
patches (using patch pipettes with 6–8 mm tip openings and
a resistance of 0.5–1 MV ) with symmetric K / concentrations, we found that ATP (1 mM) significantly inhibited K /
currents and AMP-PNP (0.5–1 mM), a nonhydrolyzable
ATP analogue, had a similar effect. Interestingly, we found
that currents that were inhibited by ATP seemed to have
different conductances, indicating that more than one type
of K / channels that are inhibited by ATP existed in these
human neurons.
95
96
C. JIANG AND G. G. HADDAD
affect baseline channel activity in excised patches, it enhanced Popen when these channels had been inhibited by ATP
(0.011 { 0.007 vs. 0.155 { 0.028, n Å 3, P õ 0.05).
These characteristics indicate that this inward rectifying K /
channel found in both human and rat neocortical neurons
belongs to the KATP channel family.
Large conductance ATP-inhibitable K / channel
Under the same experimental conditions, we observed a
large-conductance current in 5 patches out of Ç70 patches
studied in human neocortical neurons. This channel had a
very low baseline channel activity with the same internal
/ 9k0b$$ja25
J-061-6
and external solutions described above. When free Ca 2/ in
the internal solution was increased to micromolar level, however, channel activity markedly increased. As in the previously described small conductance current, we found that
ATP, glibenclamide and AMP-PNP also inhibited this current (n Å 5). Figure 6 shows an example of this current
from a human neocortical neuron, in which single channel
currents were recorded using symmetric K / concentrations
on either sides of the patch membrane (free Ca 2/ É 1 mM).
Two channels with fast flickering activity were seen during
baseline in this patch. The slope conductance of this current
was 180 pS (Fig. 6A), and Popen 0.211 (Fig. 6B). Channel
activity was inhibited markedly when the cytosolic side of
08-13-97 17:53:12
neupa
LP-Neurophys
Downloaded from http://jn.physiology.org/ by 10.220.32.246 on June 17, 2017
FIG . 3. Characterization of small-conductance KATP channel from a human neocortical neuron. A: single channel current
was recorded in an inside-out patch. Patch was held at 0 mV, and a ramp potential was given from 0100 to 100 mV. When
internal and external solutions contained equal concentrations of K / (150 mM), channel showed an inward rectification
with a reversal potential at 0 mV. In negative membrane potential ( 010 to 090 mV), conductance is almost linear and it is
48 pS. B: when internal membrane surface was exposed to a solution containing 54 mM KCl and 96 mM NaCl, reversal
potential was shifted to /23 mV (f ) and unitary conductance was reduced to 40 pS. Note that initial baseline currents are
produced by inward K / gradient. C: inhibition of channel activity by ATP and glibenclamide. Same single channel currents
as in A and B were studied when membrane potential was held at 0 mV and stepped to 080 mV. Single channel activity is
shown in trace 2–6 in each panel. Closing levels are indicated by broken lines. Top traces in each panel show ‘‘macroscopic’’
) are seen during baseline (Control). Both of these
currents by averaging 60 individual records. Two active channels (
currents are inhibited when ATP (1 mM) was applied to internal solution (2nd panel labeled ATP). After washout of ATP,
channel activity rapidly returned to baseline level. Channel activity is also markedly and reversibly inhibited by glibenclamide
(10 mM).
ATP-MODULATED K / CHANNELS IN HUMAN NEURON
97
the patch membrane was exposed to 1.0 mM of ATP
(Popen Å 0.015). The inhibitory effect of ATP was reversible
and channel activity rapidly returned to baseline level after
washout (Popen Å 0.223).
With symmetric K / across the membrane, this channel
had a slope conductance of 145–230 pS (196 { 27 pS,
n Å 5) with a weak inward rectification (Fig. 7A). When
the internal K / concentration was reduced to 54 mM, the
reversal potential was shifted to near 24 mV (calculated
reversal potential is 25.6 mV), indicating that this channel
is highly K / -selective (Fig. 7B).
FIG . 5. Concentration-dependent inhibition of ATP. Data were obtained
from an inside-out patch of a rat neocortical neuron. Single channel currents
were recorded (Vm Å 0100 mV) with equal concentrations of K / in both
internal and external solutions. Popen is normalized to baseline level (Popen /
Popen.CTL ). Data are fitted using the Hill equation: y Å 1/{1 / ([ATP]/
K) h }, where y Å Popen /Popen.CTL ; [ATP] Å internal ATP concentration;
K Å IC50 (half-blocking concentration, which is 140 mM), and h (Hill
coefficient) Å 1.5. ATP at 30 mM has evident inhibitory effect on channel
activity and Popen reaches almost 0 when ATP concentration is 1.0 mM.
/ 9k0b$$ja25
J-061-6
It is worth noting that besides a relatively wide range of
conductances, two patterns of channel activity were seen.
The 180-pS KATP channel in Fig. 6 had a high frequency of
channel openings and closures, whereas the 230 pS one in
Fig. 7 showed a relatively low flickering activity. Because
they had a much larger single channel conductance (4–5
times larger) than the 47 pS strongly inward rectifying channel described above, we pooled them together as a single
group.
As for the small-conductance KATP channel, this largeconductance K / channel was inhibited by ATP in a concentration-dependent manner with an IC50 of 350 mM (Fig. 8A).
This K / channel also was inhibited by AMP-PNP (0.5–1.0
mM, n Å 4) and by glibenclamide (10 mM, n Å 2; Fig.
8B). Also, channel activity ceased within 3–4 min after
excision in the absence of MgADP in the internal solution.
In addition, this channel was sensitive to internal Cs / (1
mM, n Å 2) and modestly inhibited by internal tetraethylammonium (TEA, 20 mM; Fig. 8B). Unlike the small-conductance KATP channel, however, activation of this large-conductance K / channel required the presence of micromolar
concentration of Ca 2/ in the bath (cytosolic) solution. Although this channel was activated by Ca 2/ and had a large
conductance, this channel still was seen when charybdotoxin
(50 or 100 nM) was present in the pipette (extracellular)
solution (n Å 2).
In rat neocortical neurons, we also observed a K / channel
that was inhibited by ATP, glibenclamide and AMP-PNP. As
for its human neocortical counterpart, this K / channel in rat
had a conductance of 188 { 33 (n Å 5) and showed a weak
inward rectification. Therefore, these data indicate that this
large conductance K / channel also belongs to the KATP channel
family. To differentiate this channel in human and rat neocortical neurons from the other much smaller KATP channel, we
term the large-conductance channel as L-KATP channel and the
small-conductance one as S-KATP channel.
08-13-97 17:53:12
neupa
LP-Neurophys
Downloaded from http://jn.physiology.org/ by 10.220.32.246 on June 17, 2017
FIG . 4. Small-conductance KATP channel in a rat neocortical neuron. Single channel current was studied in same condition
as Fig. 7C. An active channel is seen in this inside-out patch. Channel activity is reversibly inhibited by ATP (1.0 mM)
and AMP-PNP (1 mM).
98
C. JIANG AND G. G. HADDAD
FIG . 6. Inhibition of large-conductance current by ATP in a human
neocortical neuron. Single channel currents were studied using solutions
containing symmetric K / (150 mM) with a free-Ca 2/ concentration of 1
mM. A: patch was held at 0 mV and a ramp potential was given from 075
to 75 mV. Under this condition, 2 active channels were seen, and they
showed an inward rectification. In negative membrane potential (0 to 050
mV), linear slope conductance is 180 pS. B: when membrane potential was
held at 030 mV, same channels as in A had a Popen of 0.211 during baseline.
Application of ATP (1 mM) to cytosolic membrane caused a marked inhibition of both of these currents (Popen Å 0.015). Channel activity recovers
after washout of ATP (Popen Å 0.223).
Large conductance Ca 2/-activated K / channel
We also observed a very large conductance outward current in inside-out patches from human neocortical neurons,
which was different from both channels described above.
This current had very low channel activity when the internal
solution contained no Ca 2/ (free Ca 2/ õ 10 nM). However,
as for the L-KATP channel, when Ca 2/ concentration in the
internal solution rose to micromolar concentrations, channel
activity markedly increased. This channel was observed in
eight patches from 15 human neocortical neurons (1 patch/
/ 9k0b$$ja25
J-061-6
FIG . 7. Potassium selectivity of large-conductance current in a human
neocortical neuron. A single channel current was recorded in an inside-out
patch. The patch was held at 0 mV and a ramp potential was given from
0100 to 100 mV. A: when internal and external solutions contained equal
concentrations of K / (150 mM), this current showed a weak inward rectification with a reversal potential at 0 mV. rrr, slope conductance of 230
pS. B: when K / in internal solution was reduced to 54 mM K / (with 96
mM NaCl added), reversal potential was shifted to /24 mV (f ) and unitary
conductance was reduced to Ç190 pS.
08-13-97 17:53:12
neupa
LP-Neurophys
Downloaded from http://jn.physiology.org/ by 10.220.32.246 on June 17, 2017
cell). This channel had a slope conductance of 277 { 24
pS (n Å 8), was voltage activated and did not show inward
rectification (Fig. 9, A,C, and D). This channel was almost
completely blocked by charybdotoxin (50 nM), iberiotoxin
(100 nM) or external TEA (2 mM) but was not sensitive
to internal TEA °40 mM (n Å 4). These characteristics
indicated that this current belonged to the BK (or Maxi K)
type of K / channels.
Unlike its effect on S-KATP and L-KATP channels, ATP
enhanced channel activity of the BK channel in human neocortical neurons when applied to the cytosolic side in the
presence of Mg 2/ . Popen was increased from 0.096 { 0.011
to 0.172 { 0.014 (n Å 5, P õ 0.05; Fig. 9B). Also, ATP
enhanced the unitary conductance of this BK channel. The
slope conductance increased from 267 { 6 to 295 { 7
(n Å 5, P õ 0.01; Fig. 9, C and D). AMP-PNP at concentrations of 0.5–1.0 mM did not consistently affect this channel,
although we saw a slight increase in channel activity in three
out of eight patches. Glibenclamide at a concentration of
10–20 mM had no effect on this BK channel (n Å 4).
ATP-MODULATED K / CHANNELS IN HUMAN NEURON
This BK channel, which had a large conductance and was
blockable by charybdotoxin, also was seen in rat neocortical
neurons (n Å 4). Its channel activity (both Popen and conductance) was enhanced by intracellular ATP as in the human
neuronal BK channel.
Besides these three K / channels, we have observed several other types of K / channels whose unitary conductances
and other single channel properties were different from the
three ATP-modulated K / channels. Because the modulation
of these channels by intracellular ATP was not seen, we did
not perform detailed studies on these channels.
DISCUSSION
KATP channels are present in a number of tissues and cell
types, including pancreatic b-cells, cardiac myocytes, skeletal muscles, smooth muscles, and rodent central neurons
(Ashcroft at al. 1984; Ashford et al. 1988, 1990a; Cook
/ 9k0b$$ja25
J-061-6
and Hales 1984; Noma 1983; Ohno-Shosaku and Yamamoto
1992; Spruce et al. 1987; Standen et al. 1989; Wang and
Giebisch 1991). Our current studies have shown for the first
time that at least two distinct KATP channels exist in the CNS
of humans and rats. Evidence for the existence of more than
one subtype of KATP channels also can be found from a large
number of previous studies indicating that KATP channels are
different in their biophysical and pharmacological properties, such as conductance (ranging from 7 to 250 pS) (Ashcroft et al. 1984; Ashford et al. 1988, 1990a; Bonev and
Nelson 1993; Cook and Hales 1984; Jiang et al. 1994; Kajioka et al. 1991; Kirsch et al. 1990; Lorenz et al. 1992; Noma
1983; Ohno-Shosaku and Yamamoto 1992; Sellers et al.
1992; Spruce et al. 1987; Standen et al. 1989; Tung and
Kurachi 1991; Wakatuki et al. 1992; Wang and Giebisch
1991), rectification (both inward and outward rectification)
(Ashford et al. 1990a; Cook and Hales 1984; Kakei and
Noma 1984; Noma 1983; Standen et al. 1989; Trube and
Hescheler 1984), sensitivity to ATP (IC50 level spans from
50 mM to up to mM) (Ashcroft et al. 1984; Ashford et al.
1988; Cook and Hales 1984; Kakei et al. 1985; Noma 1983;
Noma and Shibasaki 1985; Schmid-Antomarchi et al. 1987;
Spruce et al. 1987; Standen et al. 1989; Wang and Giebisch
1991) and receptor affinity to sulfonylureas (Gopalakrishnan
et al. 1991; Treherne and Ashford 1991; Xia and Haddad,
1991; Zini et al. 1991).
The S-KATP channel found in the present study appears to
be a novel neuronal type of KATP channel. Its biophysical
and pharmacological properties are different from those of
KATP channels that have previously been found in central
neurons. Although Ashford et al. (1988) have observed a
KATP channel in cultured rat cortical neurons that has a conductance of 50 pS, the current shows a clear outward rectification and a low sensitivity to ATP. Also in cultured rat
cortical neurons, Ohno-Shosaku and Yamamoto (1992) have
identified another KATP channel, whose rectification and
ATP-sensitivity are comparable with the S-KATP channel.
However, its unitary conductance (65 pS) is Ç40% larger
than that of the S-KATP channel. It is not clear as yet whether
differences in these channel properties derive from a different protein structure or result from differences in experimental conditions. We do not believe that different recording
solutions have contributed to these differences because similar symmetric K / concentrations were used in both these
studies. Though Ashford et al. (1988) used asymmetric K /
concentrations in their internal (40 mM) and external solutions (140 mM), a substitution of K / with Na / in the internal solution should produce more inward rectification but
not an outward rectification, suggesting that the difference
in rectification between our S-KATP channel and that described by Ashford et al. (1988) is even greater than it
appears. It is interesting to note however that the S-KATP
channel shown in the present study has similarity to a KATP
channel previously described in pancreatic b-cells. Both of
these KATP channels have a unitary conductance of Ç50 pS,
show inward rectification, and are highly sensitive to ATP
and sulfonylureas (Ashcroft et al. 1984; Cook and Hales
1984).
The L-KATP channel is evidently different from the S-KATP
channel, especially in its conductance and Ca 2/ dependence.
It appears to belong to the large-conductance KATP channels
08-13-97 17:53:12
neupa
LP-Neurophys
Downloaded from http://jn.physiology.org/ by 10.220.32.246 on June 17, 2017
/
FIG . 8. A: concentration-dependent inhibition of large-conductance K
current by ATP. Data were obtained from same human neuronal patch as
in Fig. 7. Popen (normalized to baseline Popen ) is plotted as function of
internal ATP concentration. Half-blocking concentration (IC50 ) was 350
mM and Hill coefficient h was 1.2. B: characterization of same largeconductance KATP channel as in A. ATP, AMP-PNP, and glibenclamide all
inhibited this channel. Internal tetraethylammonium at a concentration of
20 mM suppressed slightly this channel.
99
100
C. JIANG AND G. G. HADDAD
that have previously been described in rat cortical, hypothalamic and substantia nigra neurons (Jiang et al. 1994; Sellers
at al. 1992; Treherne and Ashford 1991). These KATP channels have a unitary conductance of 220–250 pS and are
Ca 2/ -activated. KATP channels with a large conductance and
Ca 2/ sensitivity also have been described in other tissues.
For instance, previous studies have shown that the activity
of KATP channels in skeletal and smooth muscles is stimulated by an increase in free Ca 2/ concentrations in the cytosol (Kajioka et al. 1990; Krippeit-Drews and Uönnendonker
1992; Silberberg and van Breemen 1990). In spite of the
fact that the L-KATP channel is Ca 2/ -dependent and has a
large conductance, our data do not support the idea that it
belongs to the classical BK type of Ca 2/ -activated K / channels, because it has a number of biophysical and pharmacological properties different from those of BK channels (see
/ 9k0b$$ja25
J-061-6
below). It may be worth mentioning that the L-KATP channel
found in the present study has a range of conductances.
Two examples shown in Figs. 6 and 7 are 180 and 230
pS, respectively. The 180-pS KATP channel shows a high
frequency of channel open and closed activity, whereas the
230 pS one has a relatively low flickering activity. Because
we do not have other evidence suggesting that each of them
belongs to a distinct channel family, we tentatively pooled
them together as a single group.
Besides the KATP channels, our current studies have shown
that a large conductance Ca 2/ -activated K / channel present
in human neocortical neurons is also modulated by ATP.
This channel is similar to the BK channel previously described in a number of tissues, since it has a large conductance ( Ç280 pS) and a high sensitivity to charybdotoxin,
iberiotoxin, and external TEA. As with previous studies
08-13-97 17:53:12
neupa
LP-Neurophys
Downloaded from http://jn.physiology.org/ by 10.220.32.246 on June 17, 2017
FIG . 9. A: voltage-dependent activation of BK channel in 2 rat neocortical neurons. Currents were recorded with symmetric
concentrations of K / applied to both sides of inside-out patches (free Ca 2pl É 1 mM). Channel activity (Popen , normalized
to maximum level) started to increase near a membrane potential (Vm ) of 20 mV. Full activation was reached when Vm was
ú80 mV. Relation of Popen to Vm can be described with Boltzmann expression: y Å 1/{1 / exp[ Kv 0 Vm )/k]}, where y Å
normalized Popen , Vm Å membrane potential, Kv Å 52 mV (Vm at 50% of Popen ), and k Å 9 mV. Note that data at 0 mV point
are missing, because these potentials are very close to reversal potentials. B: enhancement of channel activity of 300-pS
outward current by ATP in a human neocortical neuron. Symmetric concentrations of K / were applied to both sides of
inside-out patch, and membrane potential was held at /40 mV. One active channel was seen during baseline with Popen Å
0.077. When patch was exposed to an internal solution containing 1 mM of ATP, 2nd channel that had a similar conductance
was activated and Popen increased to 0.171. Channel activity decreased after washout of ATP, and Popen returned to baseline
level within 1–2 min. Note that bottom traces in each pair of traces showed expanded segments of top traces located between
2 arrows. C: augmentation of unitary conductance of BK currents by ATP. Single channel currents were recorded from a
rat neocortical neuron in an inside-out patch. Using symmetric K / concentrations and a ramp command potential ( 050 to
50 mV), 2 active channels were observed. Both of these channels had same conductance of 270 pS. D: when patch was
exposed to an internal solution containing 1 mM ATP, unitary conductance increased by Ç10% to 300 pS.
ATP-MODULATED K / CHANNELS IN HUMAN NEURON
/ 9k0b$$ja25
J-061-6
In conclusion, our results have shown that three ATPmodulated K / channels exist in human and rat neocortical
neurons. By modulating these different species of K / channels, ATP could serve as a regulator of cellular membrane
excitability by changing K / channel activity in temporospatial-specific manner depending on ATP availability and
demand in nerve cells.
We thank SmithKline Beecham Pharmaceuticals for a gift of BRL 38227
(levcromakalim).
This work was supported by the United Cerebral Palsy Foundation, the
American Lung Association of Connecticut, and National Institutes of
Health Grants RO1 HL-51256-01 (C. Jiang) and PO1 HD-32573, P20 NS32578, and HL-39924 (G. G. Haddad).
Address for reprint requests: C. Jiang, Dept. of Biology, Georgia State
University, 24 Peachtree Central Ave., Atlanta, GA 30322-4010.
Received 29 January 1996; accepted in final form 4 September 1996.
REFERENCES
ASHCROFT, F. M. Adenosine 5 *-triphosphate-sensitive potassium channels.
Annu. Rev. Neurosci. 11: 97–118. 1988.
ASHCROFT, F. M., HARRISON, D. E., AND S. ASHCROFT, J. H. Glucose induces closure of single potassium channels in isolated rat pancreatic bcells. Nature Lond. 312: 446–448, 1984.
ASHFORD, M.L.J., BODEN, P. R., AND TREHERNE, J. M. Glucose-induced
excitation of hypothalamic neurons is mediated by AT-sensitive K / channels. Pfluegers Arch. 415: 479–483, 1990a.
ASHFORD, M.L.J., BODEN, P. R., AND TREHERNE, J. M. Tolbutamide excites
rat glucoreceptive ventromedial hypothalamic neurones by indirect inhibition of ATP-K / channels. Br. J. Pharmacol. 101: 531–540, 1990b.
ASHFORD, M.L.J., STURGESS, N. C., TROUT, N. J., GARDNER, N. J., AND
HALES, C. N. Adenosine-5-triphosphate-sensitive ion channels in neonatal rat cultured central neurons. Pfluegers Arch. 412: 297–304, 1988.
BONEV, A. D. AND NELSON, M. K. ATP-sensitive potassium channels in
smooth muscle cells from guinea pig urinery bladder. Am. J. Physiol.
264 (Cell Physiol. 33): C1190–C1200, 1993.
CHUNG, S. K., REINHART, P. H., MARTIN, B. L., BRAUTIGAN, D., AND LEVITAN, L. B. Protein kinase activity closely associated with a reconstituted
calcium-activated potassium channel. Science Wash. DC 253: 560–562,
1991.
COOK, D. L. AND HALES, C. N. Intracellular ATP directly blocks K / channels in pancreatic b-cells. Nature Lond. 311: 271–273, 1984.
CUMMINS, T. R., JIANG, C., AND HADDAD, G. G. Decrease in human neocortical excitability during anoxia via Na / channel modulation. J. Clin.
Invest. 91: 968–915, 1993.
DAVIES, N. W., STANDEN, N. B., AND STANFIELD, P. R. The effect of intracellular pH on ATP-dependent potassium channels of frog skeletal muscle. J. Physiol. Lond. 445: 549–568, 1992.
DE WEILLE, J. R. AND LAZDUNSKI, M. Regulation of the ATP-sensitive
potassium channel. In: Ion Channels, edited by T. Narahashi. New York:
Plenum, 1990, p. 205–222.
FINDLAY, I. Calcium-dependent inactivation of the ATP-sensitive K / channel of rat ventricular myocytes. Biochim. Biophys. Acta 943: 297–304,
1988.
GOPALAKRISHNAN, M., JOHNSON, D. E., JANIS, R. A., AND TRIGGLE, D. J.
Characterization of binding of the ATP-sensitive potassium channel ligand, [ 3H]-glyburide, to neuronal and muscle preparations. J. Pharmacol. Exp. Ther. 257: 1162–1171, 1991.
GRIGG, J. J. AND ANDERSON, E. G. Glucose and sulfonylurea modify different phases of the membrane potential change during hypoxia in rat hippocampal slices. Brain Res. 489: 302–310, 1989.
HADDAD, G. G. AND JIANG, C. O2 deprivation in the central nervous system:
on mechanisms of neuronal response, differential sensitivity and injury.
Prog. Neurobiol. 40: 277–318, 1993.
HAMILL, O. P., MARTY, A., NEHER, E., SAKMANN, B., AND SIGWORTH , F. J.
Improved patch-clamp techniques for high-resolution current recording
from cells and cell-free membrane patches. Pfluegers Arch. 391: 85–
100, 1981.
HILLE, B. Ionic Channels of Excitable Membranes. Potassium Channels
and Chloride Channels. Sunderland, MD: Sinauer, 1992, p. 115–139.
JIANG, C. AND HADDAD, G. G. Effect of anoxia on intracellular and extracel-
08-13-97 17:53:12
neupa
LP-Neurophys
Downloaded from http://jn.physiology.org/ by 10.220.32.246 on June 17, 2017
(Chung et al. 1991), our results have shown that ATP enhances Popen and the unitary conductance of this BK channel.
This enhancement of Popen by ATP appears to be mediated
by a protein phosphorylation process since the effect can
not be mimicked by the nonhydrolyzable ATP analogue
AMP-PNP (see also Chung et al. 1991).
Although both BK channel and L-KATP channel are Ca 2/ dependent and have a large conductance, there are clear
differences between these two channels. For example, 1)
ATP at micromolar concentration can suppress completely
this channel, whereas the BK channels are stimulated by
ATP. 2) The effect of ATP on the L-KATP channel is not
mediated by protein phosphorylation, but phosphorylation is
required for the ATP effect on the BK channels (Chung et
al. 1991). 3) The L-KATP channel is sensitive to KATP channel
blockers and insensitive to charybdotoxin, whereas the BK
channel is not sensitive to glibenclamide. It is also known
that single channel conductance of KATP currents varies
among tissues and cell types, and KATP channels with large
conductances have been described previously in other tissues
(Ashford et al. 1990a,b; Jiang et al. 1994; Krippeit-Drews
and Uönnendonker 1992; Lorenz et al. 1992; Sellers et al.
1992).
The modulation of ion channel activity by intracellular
metabolites such as ATP can have major impact on cellular
function and response to metabolic stress (de Weille and
Lazdunski 1990; Haddad and Jiang 1993; Rorsman et al.
1990; Standen 1992; Takeno and Noma 1993). In central
neurons, KATP channels are activated during hypoxia as we
and others have previously shown (Grigg and Anderson
1989; Jiang and Haddad 1991; Jiang et al. 1994; Luhmann
and Heinemann 1992; Mourre et al. 1989; Murphy and
Greenfield 1992). Activation of these KATP channels produces hyperpolarization or attenuation of the hypoxia-induced depolarization in neurons, with a resulting decrease
in energy consumption and thus an increased likelihood of
cell survival (Jiang and Haddad 1991; Mourre et al. 1991;
Murphy and Greenfield 1992).
One important question that can be raised is why nerve
cells are endowed with KATP channels that have very different biophysical and pharmacological properties. Because SKATP can be activated around resting membrane potential,
the modulation of this inward rectifier KATP channel may be
critical for constant adjustment of resting membrane potential (Quayle and Standen 1994; Standen 1992; Surprenant
et al. 1995; Takeno and Noma 1993). The L-KATP channel
may be an important modulator of membrane potential when
the cell is depolarized and when Ca 2/ increases such as
during hypoxia or hypoglycemia. In addition, both S-KATP
and L-KATP may play a role in determining membrane potential when intracellular ATP level drops. Because these channels have different sensitivities to ATP, it is possible that
L-KATP may be activated at an earlier time when ATP is
declining. It is important to note that since the BK channel
is activated with ATP, ATP can activate K / channels not
only when it is abnormally low but also when it occurs at
physiologic concentrations ( ú1 mM). Hence, it would seem
that the different biophysical properties of K / channels described can be useful in various circumstances and levels of
stress. Thus ATP (or other nucleotides) may be considered
as sensors that can adjust neuronal membrane excitability.
101
102
C. JIANG AND G. G. HADDAD
/ 9k0b$$ja25
J-061-6
K / channel in rat cultured cortical neurons. Pfluegers Arch. 422: 260–
266, 1992.
QUAYLE, J. M., STANDEN, N. B., AND STANFIELD, P. B. The voltage-dependent block of ATP-sensitive potassium channels of frog skeletal muscle
by cesium and barium ions. J. Physiol. Lond. 405: 677–697, 1988.
QUAYLE, J. M. AND STANDEN, N. B. KATP channels in vascular smooth
muscle. Cardiovasc. Res. 28: 797–804, 1994.
RORSMAN, P., BERGGREN, P. O., BOKVIST, K., AND EFENDIC, S. ATP-regulated K / channels and diabetes mellitus. News Physiol. Sci. 5: 143–147,
1990.
SCHMID-ANTOMARCHI, H., DE WEILLE, J. R., FOSSET, M., AND LAZDUNSKI,
M. The receptor for antidiabetic sulfonylureas controls the activity of the
ATP-modulated K / channel in insulin-secreting cells. J. Biol. Chem.
262: 15840–15844, 1987.
SELLERS, A. J., BODEN, P. R., AND ASHFORD, M. L. Lack of effect of potassium channel openers on ATP-modulated potassium channels recorded
from rat ventromedial hypothalamic neurons. Br. J. Pharmacol. 107:
1069–1074, 1992.
SIGWORTH, F. J. AND SINE, S. M. Data transformations for improved display
and fitting of single-channel dwell time histograms. Biophys. J. 52: 1047–
1054, 1987.
SILBERBERG, S. D. AND VAN BREEMEN, C. An ATP, Calcium and voltage
sensitive potassium channel in porcine coronary artery smooth muscle
cells. Biochem. Biophys. Res. Commun. 172: 517–522, 1990.
SPRUCE, A. E., STANDEN, N. B., AND STANFIELD, P. R. Studies on unitary
properties of adenosine-5 triphosphate-regulated potassium channels of
frog skeletal muscle. J. Physiol. Lond. 382: 213–237, 1987.
STANDEN, N. B. Potassium channels metabolism and muscle. Exp. Physiol.
77: 1–25, 1992.
STANDEN, N. B., QUAYLE, J. M., DAVIES, J. E., BRAYDEN, N. W., HUANG,
Y., AND NELSON, M. T. Hyperpolarizing vasodilators activate ATP-sensitive K / channels in arterial smooth muscle. Science Wash. DC 345: 177–
180, 1989.
SURPRENANT, A., BUELL, G., AND NORTH, R. A. P-2X receptors bring new
structure to ligand-gated ion channels. Trends Neurosci. 18: 224–229,
1995.
TAKENO, M. AND NOMA, A. The ATP-sensitive K / channel. Prog. Neurobiol. 41: 21–30, 1993.
TREHERNE, J. M. AND ASHFORD, M. L. The regional distribution of sulphonylurea binding sites in rat brain. Neuroscience 40: 523–531, 1991.
TRUBE, G. AND HESCHELER, J. Inward-rectifying channels in isolated patches
of the heart cell membrane: ATP-dependence and comparison with cellattached patches. Pfluegers Arch. 401: 178–184, 1984.
TUNG, R. T. AND KURACHI, Y. On the mechanism of nucleotide diphosphate
activation of the ATP-sensitive K / channel in ventricular cell of guineapig. J. Physiol. Lond. 437: 239–256, 1991.
WAK ATUKI, T., NAK AYA, Y., AND INOUE, I. Vasopressin modulates K / channel activities of cultured smooth muscle cells from porcine coronary
artery. Am. J. Physiol. 263 (Heart Circ. Physiol. 31): H491–H496, 1992.
WANG, W. H. AND GIEBISCH, G. Dual effect of adenosine triphosphate on
the apical small conductance K / channel of the rat cortical collecting
duct. J. Gen. Physiol. 98: 35–61, 1991.
XIA, Y. AND HADDAD, G. G. Major differences in CNS sulfonylurea receptor
distribution between rat (newborn, adult and turtle). J. Comp. Neurol.
314: 278–289, 1991.
ZINI, S., TREMBLAY, E., ROISIN, M. P., AND BEN-ARI, Y. Two binding sites
for [ 3H]glibenclamide in rat brain. Brain Res. 542: 151–154, 1991.
08-13-97 17:53:12
neupa
LP-Neurophys
Downloaded from http://jn.physiology.org/ by 10.220.32.246 on June 17, 2017
lular potassium activity in hypoglossal neurons in vitro. J. Neurophysiol.
66: 103–111, 1991.
JIANG, C. AND HADDAD, G. G. Differential responses of neocortical neurons
to O2 and/or glucose deprivation in human and rat. J. Neurophysiol. 68:
2165–2173, 1992.
JIANG, C. AND HADDAD, G. G. A direct mechanism for sensing low oxygen
levels by central neurons. Proc. Natl. Acad. Sci. USA 91: 7198–7201,
1994.
JIANG, C. AND HADDAD, G. G. Oxygen deprivation inhibits a K / channel
independently of cytosolic factors in central neurons. J. Physiol. Lond.
481: 15–26, 1994.
JIANG, C., SIGWORTH, F. J., AND HADDAD, G. G. Oxygen deprivation activates an ATP-inhibitable K / channel in substantia nigra neurons. J.
Neurosci. 14: 5590–5602, 1994.
KAJIOK A, S., KITAMURA, K., AND KURIYAMA, H. Guanosine diphosphate
activates an adenosine-5 *-triphosphate-sensitive K / channel in the rabbit
portal vein. J. Physiol. Lond. 444: 397–418, 1991.
KAJIOK A, S., OIKE, M., AND KITAMURA, K. Nicorandil opens a calcium
dependent potassium channel in smooth muscle cells of the rat portal
vein. J. Pharmacol. Exp. Ther. 254: 905–913, 1990.
KAKEI, M. AND NOMA, A. Adenosine 5 *-triphosphate-sensitive single potassium channels in atrioventricular node cell of the rabbit heart. J. Physiol.
Lond. 352: 265–284, 1984.
KAKEI, M., NOMA, A., AND SHIBASAKI, T. Adenosine-triphosphate-regulated
potassium channels in guinea-pin ventricular cells. J. Physiol. Lond. 363:
441–462, 1985.
KAY, A. R. AND WONG, R. K. S. Isolation of neurons suitable for patchclamping from adult mammalian central nervous system. J. Neurosci.
Methods 16: 227–238, 1986.
KIRSCH, G. E., CODINA, J., BIRNBAUMER, L., AND BROWN, A. M. Coupling
of ATP-sensitive K / channels to A1 receptors by G proteins in rat ventricular myocytes. Am. J. Physiol. 259 (Heart Circ. Physiol. 27): H820–
H826, 1990.
KRIPPEIT-DREWS, P. AND UÖ NNENDONKER, U. Dual effects of calcium on
ATP-sensitive potassium channels of frog skeletal muscle. Biochim. Biophys. Acta 1108: 119–122, 1992.
LORENZ, J. N., SCHNERMANN, J., BROSIUS, F. C., BRIGGS, J. P., AND FURSPAN, P. B. Intracellular ATP can regulate afferent arteriolar tone via
ATP-sensitive K / channels in the rabbit. J. Clin. Invest. 90: 733–740,
1992.
LUHMANN, H. J. AND HEINEMANN, U. Hypoxia-induced functional alterations in adult rat neocortex. J. Neurophysiol. 67: 798–811, 1992.
MOURRE , C., BEN ARI, Y., BERNARDI, H., FOSSET, M., AND LAZDUNSKI,
M. Antidiabetic sulfonylurea: location of binding sites in the brain and
effects on the hyperpolarization induced by anoxia in hippocampal slices.
Brain Res. 486: 159–164, 1989.
MURPHY, K. P. S. AND GREENFIELD, S. A. Neuronal selectivity of ATPsensitive potassium channels in guinea-pig substantia nigra revealed by
responses to anoxia. J. Physiol. Lond. 453: 167–183, 1992.
NOMA, A. ATP-regulated K / channels in cardiac muscle. Nature Lond.
305: 147–148, 1983.
NOMA, A. AND SHIBASAKI, T. Membrane current through adenosine-triphosphate-regulated potassium channels in guinea-pig ventricular cells. J.
Physiol. Lond. 363: 463–480, 1985.
OHNO-SHOSAKU, T. AND YAMAMOTO, C. Identification of an ATP-sensitive