Download ATP-Sensitive K+ Channels in the Brain: Sensors of

ATP-Sensitive K+ Channels in the Brain: Sensors of
Hypoxic Conditions
Katsuya Yamada1 and Nobuya Inagaki1,2
1
Department of Physiology, Akita University School of Medicine, and 2Core Research for Evolutional Science and Technology, Japan Science and
Technology Cooperation, 1-1-1 Hondo, Akita 010-8543, Japan
T
he brain is an unresting assembly of cells continually
receiving and routing information to maintain the integrity
of the individual organism. The aerobic metabolism of glucose
is critical in this process. Indeed, although the brain represents
only ~2% of body weight, it accounts for ~20% of total body
resting O2 consumption. Because of the high metabolic rate
and limited energy stores, interruption of the O2 or glucose
supply by stroke, global ischemia, or pulmonary failure readily
causes loss of consciousness and, if unheeded, generalized
convulsive seizure.
During seizure, the cerebral metabolic rates of O2 and glucose uptake increase more than under any other circumstance
(2). This massive energy demand causes a rapid fall in ATP that,
if prolonged, leads ultimately to irreversible cell damage (5)
due to intracellular ionic derangements such as Na+ and Ca2+
overload (2). To prevent the development in the brain of
energy-demanding seizure during metabolic stress, the ATPsensitive K+ (KATP) channel, the molecule that controls membrane potentials by sensing intracellular ATP levels, may play a
pivotal role. In this brief review, recent progress in accord with
this hypothesis is discussed, together with other views.
KATP channels, discovered in cardiac myocytes and then
found in many other excitable cells, including hormonesecreting cells, skeletal and smooth muscle cells, and neurons,
alter open probability as the cytosolic ATP concentration
changes (for review, see Ref. 14). The function of KATP channels
has been described best in insulin-secreting pancreatic --cells
(Fig. 1). Elevated blood glucose increases the intracellular
ATP/ADP ratio in --cells to close the channels, depolarizing
the plasma membrane and activating the voltage-dependent
Ca2+ channels, allowing Ca2+ influx to induce exocytosis of
insulin. The sulfonylureas used in the treatment of diabetes
mellitus also close the KATP channels to stimulate insulin secretion. In heart cells, on the other hand, the decreased cytosolic
ATP concentration during ischemia or hypoxia promotes K+
efflux from the cells by activating the KATP channels, which
rapidly dampens excitability by shortening the action potential
duration.
KATP channels are also expressed in the brain, but their functional role is poorly understood (19). Binding studies using
radiolabeled sulfonylureas show that most brain areas, including basal ganglia, thalamus, hippocampus, and cerebral cor0886-1714/02 5.00 © 2002 Int. Union Physiol. Sci./Am. Physiol. Soc.
www.nips.org
tex, express KATP channels with different affinities for sulfonylureas (12). In the CA1 area of the hippocampus in response to
brief oxygen deprivation, the neurons are hyperpolarized by
activation of K+ conductance (5, 9). The KATP channel was proposed to be directly responsible for this change in conductance (19). Indeed, intracellular ATP decreases to 15% after ~2
min of hypoxic challenge (15), and hypoxia-induced hyperpolarization is depressed by the sulfonylureas glibenclamide and
tolbutamide in some CA1 neurons (19). However, in other CA1
neurons (for review, see Ref. 19) that are insensitive to KATP
channel blockers, a rise in the intracellular Ca2+ concentration
as well as a decrease in the cytosolic ATP concentration followed by activation of Ca2+-activated K+ (KCa) channels is
reported (9). Yamamoto and coworkers (19) have suggested
that both KATP and KCa channels contribute to hypoxia-induced
hyperpolarization in CA1 neurons and that the ratio of the contribution of these two channels differs in individual CA1 neurons.
A key question remains from these extensive studies: what is
the physiological role of brain KATP channels? Recent investigations of KATP channels by molecular approaches provide many
insights. The structure of the KATP channel was initially determined in pancreatic --cells to be an octameric complex of two
types of subunit (7): a pore-forming channel subunit (Kir6.2)
and a regulatory subunit, the sulfonylurea receptor (SUR1),
belonging to the ATP-binding cassette superfamily. Later, additional Kir and SUR subunits were identified that form complexes with different pharmacological properties: the cardiac
and skeletal muscle types are composed of Kir6.2 and SUR2A,
an isoform of SUR1 (8); the vascular smooth muscle type is
composed of Kir6.1 and SUR2B, a splice variant of SUR2A (for
review, see Ref. 14). SUR2A and SUR2B have low affinity for
sulfonylurea, whereas SUR1 (--cell type) has high affinity.
The role of brain KATP channels during hypoxic challenge
was investigated by using mutant mice lacking Kir6.2
[Kir6.2(
/
) mice] (18). The midbrain nucleus, called the substantia nigra pars reticulata (SNr), which consists mostly of
GABAergic neurons, was selected as a focus partly because the
nucleus expresses the highest binding densities for sulfonylureas with high affinity, suggesting that the KATP channels in
SNr neurons are --cell-type KATP channels (Kir6.2/SUR1).
Expression of --cell-type KATP channels in GABAergic SNr neuNews Physiol Sci 17: 127
130, 2002;
10.1152/nips.01384.2001
127
Downloaded from http://physiologyonline.physiology.org/ by 10.220.32.246 on June 15, 2017
Rapid minimization of energy consumption in excitable tissues is effective protection from lethal
effects of extreme metabolic stress. The ATP-sensitive K+ channels in the brain respond in ATPdepleted metabolic states such as hypoxia and may be involved in the protection mechanism
against energy-consuming generalized seizure.
rons has been verified functionally by single-channel analyses
(see Ref. 18 and references therein). In addition, Liss and
coworkers (10) have demonstrated by single-cell RT-PCR that
the SNr GABAergic neurons solely express pancreatic --celltype KATP channels. Thus the function of SNr GABAergic neurons in Kir6.2(
/
) mice should yield clues to the roles of -cell-type KATP channels in the brain. Another reason is that the
SNr is thought to act as a central gate in the propagation of
generalized seizure (3). Pharmacological inhibition or selective bilateral lesions of the SNr suppress seizure spread in most
animal models of epilepsy. In addition, anticonvulsant drugs
that enhance the GABA-mediated inhibition of seizures and
the blockade of excitatory neurotransmission in the nucleus
raise the threshold for seizures (for review, see Ref. 3). The possible involvement of the KATP channels of the SNr in seizure
control was first mentioned by Amoroso and colleagues in
1990 (1). They suggested that a decrease in blood glucose suppresses GABA release from the nerve terminals in the substantia nigra by hyperpolarization due to the opening of presynaptic KATP channels and proposed that this decrease in the
inhibitory capacity of the GABA system during hypoglycemia
might affect seizure protection by the SN. Thus altered seizure
susceptibility might be expected in Kir6.2(
/
) mice.
Daily behavior and basal physiological parameters of
Kir6.2(
/
) mice in resting conditions were not significantly different from those of wild-type mice. However, during brief
hypoxia (150 s, 5.4% O2), Kir6.2(
/
) mice all exhibited a
myoclonic jerk in <10 s followed by severe tonic-chronic convulsion and death at 22 s on average, whereas wild-type mice
all remained sedated during the challenge and revived normally. Electroencephalogram (EEG) and electromyogram
(EMG) revealed a sequence of seizure patterns in conscious
knockout mice: very-low-voltage EEG traces for a few seconds
indicating loss of consciousness, then fast waves for several
seconds after an abrupt, sharp deflection corresponding to the
128
News Physiol Sci • Vol. 17 • June 2002 • www.nips.org
tonic convulsion and myoclonus, followed by bilateral, highvoltage sharp wave bursts. In wild-type mice under the same
conditions, a medium-to-low-voltage EEG trace predominated
during the hypoxic challenge, suggesting that the KATP channels
participate critically in protection from seizure. This is supported by a recent study using SUR1-overexpressing animals
(6). Under the control of a Ca2+-calmodulin kinase promoter,
these transgenic mice overexpress the SUR1 subunit in forebrain. These mice show a significant increase in the threshold
for kainate-induced seizures together with increased survival
rate, suggesting that KATP channels play a pivotal role in raising
the threshold for seizures caused not only by metabolic deficiency but also by excitotoxicity.
How do KATP channels control the seizure threshold? To
investigate the cellular and ionic mechanisms, single unit
activities were recorded in the SNr by acute slice preparations
in Kir6.2(
/
) and wild-type mice (18). The spontaneous firing
rate of the SNr neurons under resting conditions was similar in
both mice. However, during brief hypoxic challenge, the wildtype neurons showed a marked decrease in the firing rate to
about one-third, whereas the firing rate of knockout neurons
increased ~1.8-fold. In addition, the sulfonylurea tolbutamide
reversed the hypoxia-induced inhibition of the firing of wildtype neurons to facilitation just as in Kir6.2(
/
) neurons,
although tolbutamide had no effect on the firing rate or the
membrane potential of SNr neurons under resting conditions.
During brief hypoxia, the membrane potentials of wild-type
SNr neurons were shifted in the hyperpolarized direction,
whereas Kir6.2(
/
) SNr neurons showed no hyperpolarization
but rather were depolarized in nystatin perforated-patch
recordings using dissociated SNr neurons (18). These results
indicate that the opening of the KATP channels exerts a strong
suppressive effect on wild-type SNr neuronal activity during
hypoxic challenge by shifting membrane potentials in the
hyperpolarized direction sufficiently to reverse the facilitation
Downloaded from http://physiologyonline.physiology.org/ by 10.220.32.246 on June 15, 2017
FIGURE 1. Simplified Kir6.2-containing ATP-sensitive K+ (KATP) channel alternative functions in insulin-secreting pancreatic --cells and midbrain GABAergic substantia nigra pars reticulata (SNr) neurons. Left: in hyperglycemia, glucose metabolism increases the intracellular ATP/ADP ratio in pancreatic --cells by oxidative
phosphorylation and closes the KATP channels, which depolarizes the plasma membrane to allow cellular excitation to induce insulin secretion. Right: in hypoxia,
decreased oxygen and the resultant decrease in cytosolic ATP/ADP ratio opens the KATP channels in SNr neurons, promoting K+ outflow, hyperpolarizing the membrane, and inactivating the neuronal spike activity to suppress generalized seizure.
gic SNc neurons express different types of KATP channel with
differing sensitivities to metabolic inhibition and proposed a
novel mechanism of the selective vulnerability of some
dopaminergic neurons in Parkinson’s disease. They showed
that neurons with --cell-type KATP channels, which comprise
Kir6.2 and SUR1, have the highest metabolic sensitivity and
that these and not neurons with other types of KATP channels
survive in weaver mice, suggesting that the --cell-type KATP
channels might have the strongest neuroprotective effect.
Zawar and colleagues (20) also reported heterogeneous
expression profiles of KATP channels in the hippocampal CA1
area: functional KATP channels (Kir6.1 plus SUR1, Kir6.2 plus
SUR1 or SUR2) are expressed in 17% of the pyramidal cells
and 75% of the interneurons. Especially interesting, 58% of
CA1 interneurons express --cell-type KATP channels. Clarification of the involvement of these channels in energy-depleted
conditions should provide clues to understanding why certain
sets of pyramidal neurons are extremely vulnerable to ischemic
stress but others are not.
What then is the specific role of the KATP channels in the
SNr? It is widely known that the neurons of the SNr show the
highest spontaneous activity (up to 100 Hz) in the brain, indicating a very high metabolic rate in these neurons in the normoxic condition. Indeed, SNr neurons are extremely sensitive
to hypoxia (18). On the other hand, it has been reported that
the potentials evoked by electrical stimulation in hippocampal
(5) and cerebral cortical (18) neurons are not altered during
brief hypoxia. Thus the KATP channels in SNr neurons are likely
to act as the sensors in hypoxia, responding before the general
self-defense reaction to hypoxic conditions in other neuron
types (Fig. 1). In addition, SNr neurons innervate various distant nuclei in diverse motor-related functions, including the
ventral thalamic nuclei, superior colliculus, and pedunculopontine nucleus in the brain stem. Abrupt silence in the SNr
GABAergic projection neurons during an early phase of brain
metabolic emergency might well exert the nigral protection
mechanism by conveying a signal of massive disinhibition to
all of these targets simultaneously, which should protect the
whole brain from generalized seizure. In addition, these studies suggest that KATP channels may be a target of site-specific
treatment of brain disorders associated with ATP insufficiency
such as stroke and metabolic encephalopathies.
References
1. Amoroso S, Schmid-Antomarchi H, Fosset M, and Lazdunski M. Glucose,
sulfonylureas, and neurotransmitter release: role of ATP-sensitive K+ channels. Science 247: 852
854, 1990.
2. Basic Neurochemistry (6th ed.), edited by Siegel GJ, Agranoff BW, Albers
RW, Fisher SK, and Uhler MD. Philadelphia: Lippincott-Raven, 1999.
3. Depaulis A, Vergnes M, and Marescaux C. Endogenous control of
epilepsy: the nigral inhibitory system. Prog Neurobiol 42: 33
52, 1994.
4. Haddad GG and Jiang C. O2-sensing mechanisms in excitable cells: role
of plasma membrane K+ channels. Annu Rev Physiol 59: 23
43, 1997.
5. Hansen AJ. Effect of anoxia on ion distribution in the brain. Physiol Rev 65:
101
148, 1985.
6. Hernandez-Sanchez C, Basile AS, Fedorova I, Arima H, Stannard B, Fernandez AM, Ito Y, and LeRoith D. Mice transgenically overexpressing
sulfonylurea receptor 1 in forebrain resist seizure induction and excitotoxic neuron death. Proc Natl Acad Sci USA 98: 3549
3554, 2001.
7. Inagaki N, Gonoi T, Clement JP 4th, Namba N, Inazawa J, Gonzalez G,
News Physiol Sci • Vol. 17 • June 2002 • www.nips.org
129
Downloaded from http://physiologyonline.physiology.org/ by 10.220.32.246 on June 15, 2017
of neuronal activity in the Kir6.2(
/
) neurons that is due to
membrane depolarization (Fig. 1). The mechanism of the spike
facilitation and membrane depolarization observed in
Kir6.2(
/
) SNr neurons is currently unknown; depression of
electrogenic Na+-K+ pump activity during hypoxic challenge is
most likely, but other mechanisms such as inactivation of O2sensitive K+ channels by decreased PO2 are possible (4).
As mentioned, presynaptic modulatory effects on neuronal
firing by brain KATP channels have been suggested during metabolic stress (1, 16, 17). In the substantia nigra pars compacta
(SNc), the KATP channels on the striatonigral terminals of SNc
dopaminergic neurons might become active due to reduced
intraterminal levels of ATP, and the resultant hyperpolarization
of the terminal should lead to a reduction of GABA release and
so facilitate SNc neuronal activity. However, as pointed out by
Watts and colleagues (17), it is difficult to reconcile this
hypothesis with the postsynaptic suppressive effects of KATP
channel activation. The contribution of postsynaptic KATP channels in neuronal responses to hypoxic challenge was investigated in acute slice preparations (18). In the condition of isolation from presynaptic effects by the presence of blockers of
both excitatory and inhibitory fast neurotransmissions, the firing rates of both wild-type and Kir6.2(
/
) GABAergic SNr
neurons increased ~35% in normoxia, indicating some facilitatory effect of blocked presynaptic GABAergic terminals on
the firings of SNr neurons. However, the responses to hypoxic
challenge in the presence and absence of blockers were similar: a striking contrast in the spontaneous firing rate of SNr neurons and a net decrease in wild-type and a net increase in
Kir6.2(
/
) neurons, indicating that the opening of the postsynaptic KATP channels is pivotal in the hypoxia-induced
responses of wild-type SNr neurons, at least in this experimental condition. The physiological significance of the presynaptic
KATP channels must be further investigated.
It is important to note that wild-type mice exhibited generalized convulsion in very severe hypoxic conditions, such as
4.3% O2 for 150 s (18). In milder hypoxic conditions, such as
7.3% O2, most of the Kir 6.2(
/
) mice showed no convulsion
and none died. Secher and Wilhjelm (13) have reported that
the tolerance of conscious animals to anoxia increases very
rapidly when the O2 concentration is >5%, whereas at <4% O2
survival time is extremely short (<10 min). There is a rapid
deflection in the O2 concentration vs. survival time curve at
~4
5% O2. Thus the KATP channel may fully exert its protective
role only in limited severities of hypoxic challenge just above
this critical transition. In addition, the EEG and EMG of
Kir6.2(
/
) mice responded within several seconds after the
hypoxic condition was achieved, suggesting involvement of
the KATP channels in the initial stage of the response to hypoxic
challenge.
To investigate the contribution of the KATP channels in other
brain nuclei to the hypoxia-induced response is important.
Although KATP channels are functionally expressed in various
nuclei, such as cerebral cortex (11), hippocampus (19), hypothalamus, and SNc (reviewed in Ref. 10), the molecular
makeup of neuronal KATP channels appears not to be homogeneous. Liss and colleagues (10), using a combined approach of
patch-clamp and single-cell RT-PCR, reported that dopaminer-
8.
9.
10.
11.
12.
130
News Physiol Sci • Vol. 17 • June 2002 • www.nips.org
14. Seino S. ATP-sensitive potassium channels: a model of heteromultimeric
potassium channel/receptor assemblies. Annu Rev Physiol 61: 337
362,
1999.
15. Siesjö BK. Mechanisms of ischemic brain damage. Crit Care Med 16: 954
963, 1988.
16. Stanford IM and Lacey MG. Electrophysiological investigation of adenosine triphosphate-sensitive potassium channels in the rat substantia nigra
pars reticulata. Neuroscience 74: 499
509, 1996.
17. Watts AE, Hicks GA, and Henderson G. Putative pre- and postsynaptic
potassium channels in the rat substantia nigra in vitro. J Neurosci 15:
3065
3074, 1995.
18. Yamada K, Ji JJ, Yuan H, Miki T, Sato S, Horimoto N, Shimizu T, Seino S,
and Inagaki N. Protective role of ATP-sensitive potassium channels in
hypoxia-induced generalized seizure. Science 292: 1543
1546, 2001.
19. Yamamoto S, Tanaka E, and Higashi H. Mediation by intracellular calcium-dependent signals of hypoxic hyperpolarization in rat hippocampal
CA1 neurons in vitro. J Neurophysiol 77: 386
392, 1997.
20. Zawar C, Plant TD, Schirra C, Konnerth A, and Neumcke B. Cell-type specific expression of ATP-sensitive potassium channels in the rat hippocampus. J Physiol 514: 327
341, 1999.
Downloaded from http://physiologyonline.physiology.org/ by 10.220.32.246 on June 15, 2017
13.
Aguilar-Bryan L, Seino S, and Bryan J. Reconstitution of IKATP: an inward
rectifier subunit plus the sulfonylurea receptor. Science 270: 1166
1170,
1995.
Inagaki N, Gonoi T, Clement JP, Wang CZ, Aguilar-Bryan L, Bryan J, and
Seino S. A family of sulfonylurea receptors determines the pharmacological properties of ATP-sensitive K+ channels. Neuron 16: 1011
1017,
1996.
Leblond J and Krnjevic K. Hypoxic changes in hippocampal neurons. J
Neurophysiol 62: 1
14, 1989.
Liss B, Bruns R, and Roeper J. Alternative sulfonylurea receptor expression
defines metabolic sensitivity of K-ATP channels in dopaminergic midbrain
neurons. EMBO J 18: 833
846, 1999.
Luhman HJ 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
sulfonylureas: localization of binding sites in the brain and effects on the
hyperpolarization induced by anoxia in hippocampal slices. Brain Res
486: 159
164, 1989.
Secher O and Wilhjelm B. The protective action of anaesthetics against
hypoxia. Can Anaesth Soc J 15: 423
440, 1968.