- Email: [email protected]
anoxic injury in vitro and in vivo
Neuroscience 144 (2007) 1509 –1515
Kir6.2-CONTAINING ATP-SENSITIVE POTASSIUM CHANNELS PROTECT CORTICAL NEURONS FROM ISCHEMIC/ANOXIC INJURY IN VITRO AND IN VIVO H.-S. SUN,a Z.-P. FENG,b P. A. BARBER,c A. M. BUCHANc AND R. J. FRENCHa*
Cerebral strokes, a leading cause of morbidity and mortality in industrialized countries, cause acute cerebral ischemia and reperfusion injuries that result in severe neurological damage (Lipton, 1999). A major repercussion of the loss of oxygen and nutrients caused by ischemic stroke is a concomitant massive release of the excitatory neurotransmitter glutamate, leading to a intracellular Ca2⫹ overload and cell death (Lipton, 1999). This discovery led to clinical trials to determine the efficacy of anti-excitotoxic therapies (Davis et al., 1997, 2000; Morris et al., 1999; Lees et al., 2000); unfortunately, these therapies failed to benefit stroke patients (Ikonomidou and Turski, 2002; Hoyte et al., 2004), underscoring the still unmet need for effective neuroprotective agents and therapeutic strategies to protect the brain from the damage caused by cerebral stroke, and improve the recovery of stroke patients. Recently, neuroprotective effects of ATP-sensitive potassium channels (KATP channels) against hypoxic induced generalized seizures have been demonstrated (Yamada et al., 2001; Seino and Miki, 2004). KATP channels, which are hetero-octamers composed of pore-forming Kir6.x (6.1 or 6.2) subunits and sulfonylurea receptor (SUR1 or SUR2) regulatory subunits (Babenko et al., 1998; Miki et al., 1999; Seino, 1999), are expressed in many types of tissues and cells. Kir6.2-containing KATP channels are present in a variety of cell types in the brain, including pyramidal cells, interneurons, granule cells and glial cells of the hippocampus (Zawar et al., 1999; Zhou et al., 2002; Sun et al., 2006), neurons of the substantia nigra pars reticulata (Yamada et al., 2001), and also in other parts of the CNS, including the hypothalamus (Miki et al., 2001). KATP channel activation, which is regulated by intracellular ATP and ADP concentrations, has been associated with acute and delayed protective responses to ischemia in a variety of organs and tissues. An increase in the ratio of [ATP] to [ADP] decreases the probability of KATP channel opening, whereas a decrease in this ratio promotes opening of these channels. KATP channel activation tends to hyperpolarize the cell by shifting the membrane potential toward the potassium equilibrium potential. Thus, KATP channels couple cell metabolism to cell membrane potential, thereby regulating many cellular activities. Recently we (Sun et al., 2006), identified a fundamental, neuroprotective role for KATP channels in the hippocampus, and discovered a KATP-dependent electrophysiological response of hippocampal neurons to anoxic stress. Here, we have focused our attention on the neuroprotective effects of KATP channels against damage from
a
Department of Physiology and Biophysics, Faculty of Medicine, University of Calgary, 3330 Hospital Drive Northwest, Calgary, Alberta, Canada T2N 4N1
b
Department of Physiology, University of Toronto, 1 King’s College Circle, Toronto, Ontario, Canada M5S 1A8
c
Department of Clinical Neuroscience, University of Calgary, Calgary, Alberta, Canada
Abstract—ATP-sensitive potassium (KATP) channels are weak inward rectifiers that appear to play an important role in protecting neurons against ischemic damage. Cerebral stroke is a major health issue, and vulnerability to stroke damage is regional within the brain. Thus, we set out to determine whether KATP channels protect cortical neurons against ischemic insults. Experiments were performed using Kir6.2ⴚ/ⴚ KATP channel knockout and Kir6.2ⴙ/ⴙ wildtype mice. We compared results obtained in Kir6.2ⴚ/ⴚ and wildtype mice to evaluate the protective role of KATP channels against focal ischemia in vivo, and, using cortical slices, against anoxic stress in vitro. Immunohistochemistry confirmed the presence of KATP channels in the cortex of wildtype, but not Kir6.2ⴚ/ⴚ, mice. Results from in vivo and in vitro experimental models indicate that Kir6.2-containing KATP channels in the cortex provide protection from neuronal death. Briefly, in vivo focal ischemia (15 min) induced severe neurological deficits and large cortical infarcts in Kir6.2ⴚ/ⴚ mice, but not in wildtype mice. Imaging analyses of cortical slices exposed briefly to oxygen and glucose deprivation (OGD) revealed a substantial number of damaged cells (propidium iodide–labeled) in the Kir6.2ⴚ/ⴚ OGD group, but few degenerating neurons in the wildtype OGD group, or in the wildtype and Kir6.2ⴚ/ⴚ control groups. Slices from the three control groups had far more surviving cells (anti-NeuN antibodylabeled) than slices from the Kir6.2ⴚ/ⴚ OGD group. These findings suggest that stimulation of endogenous cortical KATP channels may provide a useful strategy for limiting the damage that results from cerebral ischemic stroke. © 2006 IBRO. Published by Elsevier Ltd. All rights reserved. Key words: cerebral cortex, cerebral ischemia, focal cerebral ischemia, neuronal death, neuroprotection. *Corresponding author. Tel: ⫹1-403-220-8389 or ⫹1-403-220-6893; fax: ⫹1-403-283-8731. E-mail address: [email protected] (R. J. French). Abbreviations: CCA, common carotid artery; ECA, external carotid artery; ICA, internal carotid artery; KATP, ATP-sensitive potassium; Kir6.2, subunit of the ATP-sensitive potassium channel; Kir6.2⫺/⫺, mutant mice lacking the Kir6.2 subunit of the ATP-sensitive potassium channel; LDF, laser-Doppler flowmetry; MCA, middle cerebral artery; MCAO, middle cerebral artery occlusion; NS, not significant; OGD, oxygen and glucose deprivation; PI, propidium iodide; rCBF, regional cerebral blood flow; TTC, triphenyltetrazolium.
0306-4522/07$30.00⫹0.00 © 2006 IBRO. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.neuroscience.2006.10.043
1509
1510
H.-S. Sun et al. / Neuroscience 144 (2007) 1509 –1515
focal ischemia and oxygen and glucose deprivation (OGD) in the cerebral cortex. It is generally accepted that global and focal ischemia induces specific cell injuries in different regions of the brain (Lipton, 1999). For example, a 15 min global ischemia can cause damage selectively to the CA1 region of the hippocampus (Nelson et al., 1997; Modo et al., 2000). There are at least two reasons why the consequences of focal ischemia might differ from those of global ischemia. First, in the ischemic core of a focal lesion, blood flow is generally higher than it is in global ischemia. Thus, longer episodes might be needed to induce a similar amount of damage. Second, in a focal lesion, there is a progression in the severity of ischemia from the penumbra to the core, resulting in different metabolic conditions within the affected regions. This diversity makes focal ischemia more complex than global ischemia, and underlines the difficulty of predicting its effects. Although we have shown that KATP channels provide significant protection against hypoxia and ischemia in hippocampal neurons (Sun et al., 2006), whether, and how, KATP channels contribute to neuroprotection in the cortex has not been examined in detail. We postulated that KATP channels play a less important role in the cortex since hippocampal neurons are especially sensitive to global ischemia. Contrary to our hypothesis, we report that Kir6.2containing KATP channels protect neurons in all cortical layers to a similar extent to that seen for neurons in our in vitro studies in the hippocampus (Sun et al., 2006). These results indicate that Kir6.2-containing KATP channels play a similarly fundamental role in neuroprotection in cortical and hippocampal neurons.
EXPERIMENTAL PROCEDURES Mice Mutant mice lacking the Kir6.2 subunit of the ATP-sensitive potassium channel (Kir6.2⫺/⫺) (Miki et al., 1998) and wildtype mice with same background were used in these experiments, as previously described (Miki et al., 1998; Li et al., 2000; Yamada et al., 2001). Adult mice used were male, aged between 10 and 12 weeks, and had a body weight of 25–35 g. Professor Susumu Seino and Dr. Takashi Miki, Division of Cellular and Molecular Medicine, Kobe University Graduate School of Medicine, Kobe, Japan provided breeding stock for the Kir6.2⫺/⫺ mice. Knockout of the Kir6.2 gene was confirmed by PCR genotyping. Kir6.2⫺/⫺ mice develop normally with no apparent abnormalities in general appearance, and are fertile (Miki et al., 1998). The brain cytoarchitecture of wildtype and Kir6.2⫺/⫺ mice appears to be comparable (Miki et al., 1998; Yamada et al., 2001). Basic physiological parameters, including body weight, body temperature, blood glucose, blood gas, pH, blood pressure, heart rate, ECG, respiratory rate, and O2 consumption, of the Kir6.2⫺/⫺ mice and wildtype mice were not significantly different (Miki et al., 1998; Li et al., 2000; Yamada et al., 2001). All mice used in this study were bred and maintained in the Animal Resources Centre at the University of Calgary, with an ambient temperature of 20⫾1 °C and a 12-h light/dark cycle. In preparation for immunohistochemistry (Sun et al., 2006), mice were deeply anesthetized, and transcardiacally perfused with heparinized saline followed by 4% paraformaldehyde in PBS. Brains were placed in the same fixative at 4 °C overnight and subsequently cut into 50 m sections for immunostaining.
Focal cerebral ischemia We used a modified monofilament intraluminal middle cerebral artery occlusion (MCAO) procedure to produce focal ischemia (Clark et al., 1997; Majid et al., 2000). Mice were anesthetized with isoflurane (3% initial, 1–1.5% maintenance) in O2 and air (80: 20%). Briefly, under an operating microscope, the left common carotid artery (CCA), the left external carotid artery (ECA), and the left internal carotid artery (ICA) were isolated, and a 6 – 0 suture was tied at the origin of the ECA and at the distal end of the ECA. The left CCA and the ICA were also temporarily occluded. A silicon-coated nylon suture (8 – 0) was introduced into the ECA and pushed through the ICA until resistance was felt; the filament was inserted approximately 9 –10 mm from the carotid bifurcation, effectively blocking the origin of the middle cerebral artery (MCA) (Barber et al., 2004). The diameter of the tip of the coated suture measured between 180 and 220 m. The suture was removed after 15 min, and the ECA was permanently tied off. Shamoperated mice underwent a surgical procedure that involved permanent ligation of the ECA and temporary occlusion of the CCA, without occlusion of the MCA. During the procedure, body temperature was monitored and maintained at 37⫾0.5 °C. Transcranial measurements of cerebral blood flow (CBF) were made by laser-Doppler flowmetry (LDF). An 0.5 mm diameter micro-fiber laser-Doppler probe (Perimed, Järfälla, Sweden) was attached to the skull with cyanoacrylate glue 6 mm lateral and 1 mm posterior to the bregma. While the mice were under general anesthesia, we monitored regional cerebral blood flow (rCBF) within the infarct core and in the parietal cortex (penumbra). The surgical procedure was considered adequate if a ⱖ70% reduction in rCBF occurred immediately after placement of the intraluminal occluding suture (Barber et al., 2004); otherwise mice were excluded. Blood flow velocity was measured prior to ischemia, and during MCAO and reperfusion using Perisoft (Perimed, Inc.). Mice were permitted to recover from the anesthesia at room temperature. Twenty-four hours after the procedure, we assessed neurobehavioral parameters and then removed the brains from deeply anesthetized mice. Infarct volume was assessed using a standard triphenyltetrazolium (TTC) staining technique.
Neurobehavioral evaluation Following ischemia, upon recovery from the anesthesia, mice were assessed for focal functional deficit to determine the impact of the MCA occlusion. At 24-hours post-ischemia, each mouse was scored for neurological function using a six-point standard scale (Clark et al., 1997) (neurological deficit scores: 0: no neurological deficit; 1: retracts left forepaw when lifted by the tail; 2: circles to the left; 3: falls while walking; 4: does not walk spontaneously; 5: dead).
TTC staining and infarct volume assessment In MCAO experiments, we performed infarct analysis using standard TTC (Sigma, St. Louis, MO, USA) staining techniques (Clark et al., 1997; Majid et al., 2000), and NIH Image J software (National Institutes of Health, Bethesda, MD, USA) to estimate infarct volume. Brains were sliced into 1-mm coronal sections with a matrix at defined bregma locations (Paxinos and Franklin, 2001). Each slice was incubated in 2% TTC in phosphate buffer, and stained at 37 °C for 30 min. Each TTC-stained section was digitally photographed, and infarct areas were traced onto templates representing 1 mm coronal slices. Templates were used to correct any brain edema produced by the infarction, allowing for a more accurate assessment of infarct volume. The infarct area from each template was then digitally traced using NIH Image J. The infarct volume was calculated by summing infarct areas from individual slices (Majid et al., 2000). Infarct volumes of both wildtype and
H.-S. Sun et al. / Neuroscience 144 (2007) 1509 –1515 Kir6.2⫺/⫺ were compared statistically to determine any significant difference.
Cortical slice preparations Under deep anesthesia, the brains were rapidly removed and horizontal slices (250 m) were cut using a vibratome. Brain slices were immediately transferred to a holding chamber, containing ACSF solution saturated with carbogen (95% O2 and 5% CO2), where they were held for at least 1 h at 30 °C⫾2 °C prior to OGD. The composition of the ACSF was (in mM): 124 NaCl, 3 KCl, 1.25 NaH2PO4, 2 MgCl2, 26 NaHCO3, 2 CaCl2, and 10 dextrose (pH 7.4). ACSF was saturated with 95% O2 and 5% CO2 at room temperature.
1511
was used to avoid any mechanical damage of the sectioning and consequent distortion of the images. For imaging, the slices were scanned to determine a depth halfway between the slice surfaces and the midslice section was used for confocal imaging (Sun et al., 2006). For a region of interest, we produced 3D digital reconstructions from a series of confocal images taken at 0.5 m intervals, and generated optical stacks of 10 images for the figures.
Cell counting We used NIH Image J to determine the densities of injured (PI stained) and surviving (anti-NeuN antibody stained) cells in cortical brain slices from each experimental group. Cell counts were routinely obtained from a reconstructed volume measuring 400⫻400⫻5 m.
Immunostaining using Kir6.2 antibody in cortex Brain slice sections were blocked in 3% normal goat serum/0.3% Triton X-100/0.1% BSA in PBS at room temperature for 1 h then incubated with rabbit anti-Kir6.2 antibody (1:500; kindly supplied by Professor S. Seino, Division of Cellular and Molecular Medicine, Kobe University Graduate School of Medicine, Kobe, Japan) at 4 °C overnight. The characteristics of this antibody have been reported elsewhere (Suzuki et al., 1997; Zhou et al., 2002). Subsequently, sections were incubated with affinity-purified secondary antibody, goat anti-rabbit Cy3 antibody (1:200; Chemicon, Temecula, CA, USA). Sections were then rinsed, dried, and coverslipped with Dako fluorescence mounting medium. Images were viewed under a confocal laser scanning microscope.
OGD in cortical slices We determined the survival of cortical neurons under experimental ischemic conditions caused by OGD as previously described (Sun et al., 2006). Four sets of experiments were performed. In control experiments, cortical slices from either wildtype or Kir6.2⫺/⫺ mice were incubated in ACSF saturated with carbogen (95% O2 and 5% CO2) for 30 min. In OGD experiments, cortical slices from either wildtype or Kir6.2⫺/⫺ mice were incubated under experimental ischemic conditions in ACSF lacking glucose and saturated with 95% N2 and 5% CO2 for the 30 min. All cortical brain slices were allowed to recover for 60 min in normal ACSF buffer saturated with 95% O2 and 5% CO2. Injured and surviving neurons were identified by laser confocal microscopy using double staining (PI and anti-NeuN antibody). The number of injured and surviving neurons were counted and compared under each experimental condition.
PI labeling Nuclear debris and nonviable neurons in cortical brain slices after OGD were labeled by propidium iodide (PI) (5 g/ml) (Macklis and Madison, 1990; Laake et al., 1999; Bonde et al., 2002; Yin et al., 2002).
Immunocytochemistry: NeuN In double labeling experiments, following PI staining, sections were incubated in mouse anti-NeuN antibody (1:100; Chemicon) as previously described (Sun et al., 2006). These sections were then reacted with goat anti-mouse fluorescein-tagged secondary antibody (1:200; Chemicon). Sections were rinsed, dried, and coverslipped with Dako fluorescence mounting medium.
Confocal imaging Antibody labeling and double labeling were imaged using a laser confocal scanning microscope (Olympus LSM-GB200; Olympus, Tokyo, Japan) and analyzed with a three-dimensional (3D) constructor (Image-Pro Plus). During the confocal imaging, caution
Statistical analysis Data were expressed as the mean⫾S.D. and significance was determined using ANOVA and multiple comparison tests (SigmaStat3.0). The numbers of damaged or surviving cells in the four experimental groups were first tested for any statistically significant difference among the groups using a one-way ANOVA test. Since there were statistically significant differences (P⬍0.05, ANOVA) among the four groups, we used Dunnett’s and Duncan’s tests for multiple comparisons to further evaluate the differences between the individual groups. For all tests, P⬍0.05 was considered significant.
RESULTS In this study, (a) we used immunohistochemistry to confirm that Kir6.2 protein is absent from cortex in the Kir6.2⫺/⫺ mice, but is present throughout the cortex of wildtype mice; (b) we showed that Kir6.2⫺/⫺ mice suffer severe neurological consequences, with associated grossly defined cerebral infarcts, following a 15 min MCAO, while wildtype mice and shams show no ill effects; and (c) we used PI and anti-NeuN antibody double labeling to show that 30 min OGD causes extensive cell death in cortical slices from Kir6.2⫺/⫺ mice, but causes little or no damage in wildtype cortical slices. These results lead us to conclude that cortical Kir6.2-containing KATP channels play an important role for normal neuroprotective responses in the cortex. Expression of Kir6.2 protein in cortex: anti-Kir6.2 antibody staining Using an anti-Kir6.2 antibody, we demonstrated by immunohistochemistry that Kir6.2 protein is expressed in pyramidal-shaped neurons of the cortex in adult wildtype mice (Fig. 1). In contrast, we observed no immunoreactivity to the Kir6.2 protein in the cortex of the Kir6.2⫺/⫺ mice (Fig. 1). In vivo experiments: focal cerebral ischemia (MCAO) We found no significant difference in body weights or blood glucose levels between the Kir6.2⫺/⫺ and wildtype mice. Body weights were 30.65⫾2.51 g for the wildtype and 31.23⫾2.58 g for the Kir6.2⫺/⫺ (n⫽10/group, NS, not significant: P⫽0.987). Blood glucose levels were 6.62⫾1.02 mmol/L for the wildtype group and 7.88⫾1.79 mmol/L for the Kir6.2⫺/⫺ group (n⫽10/group, NS: P⫽0.069). The rCBF, estimated by LDF, was reduced by more than 70%
1512
H.-S. Sun et al. / Neuroscience 144 (2007) 1509 –1515
15 min MCAO. Infarct volume was significantly larger in Kir6.2⫺/⫺ brains exposed to 15-min MCAO than in wildtype brains treated similarly (n⫽10 per group) (Fig. 2). In addition, we detected no damage in the brains of sham-operated mice, whether Kir6.2⫺/⫺ and wildtype (n⫽5/each group).
Fig. 1. Anti-Kir6.2 antibody immunostaining of the cortex of the brain (60⫻ objective). The section shown spans approximately from layer III to layer V, from top to bottom. Most cells of the cortical cells express Kir6.2 protein in wildtype mice. In contrast, the cortex of Kir6.2⫺/⫺ mice did not show any immunoreactivity for the Kir6.2 protein. Scale bars⫽50 m.
from pre-occlusion values after introduction of the suture, indicating that the procedure resulted in adequate occlusion (Barber et al., 2004). rCBF measurements were similar in Kir6.2⫺/⫺ and wildtype groups before, during, and after, the MCAO procedure. Neurobehavioral scores In general, as compared with the wildtype mice, Kir6.2⫺/⫺ mice did not respond well to MCAO. Kir6.2⫺/⫺ mice had a mortality rate over 50% (10 survivors, n⫽20) 24 h after a 15 min MCAO, and survivors showed signs of severe neurological deficits (NB score 4, n⫽10). We observed no cerebral hemorrhage or other bleeding around the operated area in the neck of these Kir6.2⫺/⫺ mice. No neurological deficits were detected in wildtype mice exposed to a 15 min MCAO (NB score 0, n⫽10) or in sham-operated mice, whether Kir6.2⫺/⫺ or wildtype (NB score 0, n⫽5/ group). Infarct volumes for the MCAO mice Twenty-four hours after the MCAO, the TTC staining revealed significant damage in the brains of Kir6.2⫺/⫺ mice exposed to a 15 min MCAO (Fig. 2); no detectable damage was observed in the brains of wildtype mice exposed to a
Fig. 2. TTC-stained slices of brains. (A) Slices from Kir6.2⫺/⫺ mouse. (B) Slices from a wildtype mouse exposed to brief focal ischemia. Mice were killed and slices prepared 24 h after exposure to a 15 min MCAO. Kir6.2⫺/⫺ mouse brains had extreme damage (unstained white area) in the cortex. The wildtype mouse brain showed no gross damage with the TTC stain. The thickness of each slice was 1 mm. Scale bars⫽2 mm. Slices were consecutive in the order right, bottom to top, followed by left bottom to top, at the following locations: bregma 1.34, 1.14, 0.14, ⫺0.94, ⫺2.06, ⫺3.08, ⫺4.04 mm (Paxinos and Franklin, 2001). They are viewed from the anterior side. (C) Comparison of the infarct volumes from Kir6.2⫺/⫺ and wildtype mice after 15 min MCAO. Infarct volumes in the brains of Kir6.2⫺/⫺ mice were significantly larger than that of Kir6.2 wildtype mice (n⫽10 per group, P⬍0.001).
H.-S. Sun et al. / Neuroscience 144 (2007) 1509 –1515
1513
Due to the all-or-nothing nature of the damage caused by brief focal ischemia, which caused severe gross injury (in TTC stain) in Kir6.2⫺/⫺ mice, and no gross infarction in the wildtype mice, we did not further evaluate the brains of these mice using other histological staining methods, such as H&E staining. In vitro experimental ischemia (ODG) on cortical brain slices Following 30 min OGD, cortical slices, double stained with PI and anti-NeuN antibody, revealed extensive cell damage in the Kir6.2⫺/⫺ OGD group, but minimal damage in the wildtype OGD group. In the Kir6.2⫺/⫺ OGD group, damage was apparent in all regions of the cortex (Fig. 3A, B). The numerous injured cells in the cortical fields were revealed as the PI-labeled cells in the confocal images. Relatively few surviving cells, labeled with anti-NeuN antibody, were visible in the confocal images obtained from Kir6.2⫺/⫺ cortical slices. In contrast, we observed few PIlabeled injured cells in the wildtype OGD, or the wildtype, and Kir6.2⫺/⫺ control groups; anti-NeuN antibody labeling indicated that surviving cells were abundant in all fields in cortical slices for these three groups. PI staining for injured cells: cell counts The numbers of injured cells stained with PI in all four experimental groups (Kir6.2⫺/⫺ and wildtype OGD groups, and Kir6.2⫺/⫺ and wildtype control groups) were counted and statistically compared. The results are shown in Fig. 3C. There were significantly more PI-stained, damaged cells in all cortical fields from the Kir6.2⫺/⫺ OGD group than in the wildtype OGD group, or the Kir6.2⫺/⫺, and wildtype control groups. Anti-NeuN antibody staining for surviving cells: cell counts To make a more thorough comparison, we used anti-NeuN antibody to label surviving cells in the same cortical slices. The numbers of anti-NeuN antibody-labeled surviving cells for all four experimental groups (Kir6.2⫺/⫺ and wildtype OGD groups, and Kir6.2⫺/⫺ and wildtype control groups) were counted and statistically compared. These results are shown in Fig. 3C. We observed substantially fewer surviving neurons in all cortical fields from the Kir6.2⫺/⫺ OGD group than were detected in the wildtype OGD, or Kir6.2⫺/⫺ and wildtype control groups. Statistical comparison tests of the four groups Results from ANOVA and multiple comparison tests showed a statistically significant difference among the four experimental groups (ANOVA: P⬍0.001; multiple comparison test: P⬍0.05). The numbers of injured and surviving cells in the cortical slices from the Kir6.2⫺/⫺ OGD group differed significantly from those seen in the three other experimental groups (wildtype OGD, Kir6.2⫺/⫺ and wildtype controls). Results were consistent in all cortical fields. Thus, we confirmed quantitatively that there were substantially more damaged neurons, and fewer surviving neu-
Fig. 3. (A) Double labeling, with anti-NeuN antibody (green), for surviving neurons, and PI (red), for injured cells, in cortex (4⫻ objective) exposed to OGD for 30 min. In general, there were more damaged cells labeled with PI in the Kir6.2⫺/⫺ OGD group than in the three other experimental groups (wildtype OGD, Kir6.2⫺/⫺ control, and wildtype control). In contrast, the three other experimental groups (wildtype OGD, Kir6.2⫺/⫺ control, and wildtype control) showed mostly surviving neurons labeled with anti-NeuN antibody. Scale bars⫽200 m. (B) Double labeling under higher magnification (40⫻) after 30 min of OGD. The sections shown span approximately from cortical layer III to layer V (c.f. Fig. 1A.). The Kir6.2⫺/⫺ OGD group showed more PI-labeled, damaged cells than the other three experimental groups (wildtype OGD, or Kir6.2⫺/⫺, and wildtype controls), which had mostly anti-NeuN antibody-labeled surviving neurons labeled. Scale bars⫽50 m. (C) Numbers of PI (black) and anti-NeuN antibody (gray) stained cells. The Kir6.2⫺/⫺ OGD group had the highest numbers of damaged cells and the lowest numbers of surviving cells (* P⬍0.05) compared with the three other experimental groups (wildtype OGD, or controls for either the Kir6.2⫺/⫺ or the wildtype).
1514
H.-S. Sun et al. / Neuroscience 144 (2007) 1509 –1515
rons, in all fields of the cortex in the Kir6.2⫺/⫺ OGD group compared with the three other groups (wildtype OGD, Kir6.2⫺/⫺ and wildtype controls).
DISCUSSION Kir6.2 expression Using an anti-Kir6.2 antibody, we confirmed that Kir6.2 protein is expressed in all regions of the cerebral cortex in adult wildtype mice; we did not detect Kir6.2 expression in Kir6.2⫺/⫺ mice, paralleling results we obtained for the expression of these Kir6.2 in the mouse hippocampus (Sun et al., 2006). In other studies, Kir6.2 protein has also been detected in the mouse pancreas (Suzuki et al., 1997) and the rat brain (Zhou et al., 2002). In addition, single-cell RT-PCR has detected KATP mRNA in a variety of neuronal populations in rat brains (Zawar et al., 1999). In this study, we showed that the Kir6.2 protein is widely distributed in the cortical neurons of the wildtype mice. Thus, we found that overall distribution of Kir6.2 in the different regions of the mouse brain resembles that reported for rats. KATP channels and neuroprotection To date, there have been few studies detailing the role played by KATP channels in protecting neurons from the harmful effects of ischemia and hypoxia (Fujimura et al., 1997; Yamamoto et al., 1997; Garcia et al., 1999; Blondeau et al., 2000; Yamada et al., 2001). Recently, one group (Shimizu et al., 2002) tested the neuroprotective effectiveness of diazoxide, used as a mitochondrial KATP channel agonist, against the effects of transient focal cerebral ischemia induced by MCAO. Diazoxide reduced infarct volume by 40%; in other experiments, 5-hydroxydecanoate (5-HD), another putative selective mitochondrial KATP channel blocker, completely prevented the diazoxide’s effect. Although the role of mitochondrial KATP channels remains controversial, results obtained in other systems, and based on similar pharmacological criteria, were consistent with complementary roles in cytoprotection for plasmalemmal and mitochondrial KATP channels (Light et al., 2001). Kir6.2 is thought to be the predominant pore-forming subunit of plasmalemmal KATP channels in skeletal muscle and neurons. Kir6.2 has been shown to provide protection during hypoxia-induced generalized seizure (Yamada et al., 2001) and electrical recordings in hippocampal neurons have indicated that plasmalemmal Kir6.2-containing KATP channels protect against metabolic stress by exerting a hyperpolarizing influence on plasma membranes, thereby attenuating electrical activity (Sun et al., 2006). Our results are entirely consistent with the view that a similar mechanism underlies Kir6.2-mediated neuroprotection in the cortex. This does not preclude an additional, complementary protective contribution by glial cells that also express Kir6.2. In this study, to evaluate in detail a neuroprotective role for Kir6.2-containing KATP channels in the cerebral cortex, we focused on determining rates of cell death and survival in cortical neurons exposed to ischemia. Notwithstanding
suggestions of differences in regional sensitivity to ischemic stress in the brain, we report here that Kir6.2-containing KATP channels protect neurons in all layers of the cortex to a degree similar to that seen in the hippocampus (Sun et al., 2006), indicating a fundamental role for Kir6.2containing KATP channels in neuroprotection in cortical and hippocampal neurons, and likely throughout the brain. In future, KATP channel modulators may prove to be clinically useful neuroprotective agents, as part of a combination therapy for stroke management. Acknowledgments—This work was supported by a grant from the Canadian Institutes of Health Research (to R.J.F.). H.-S.S. was recipient of Doctor Research Awards from (1) the Heart and Stroke Foundation of Canada–Focus on Stroke Training Initiative Program, in partnership with the Canadian Stroke Network, Heart and Stroke Foundation of Canada, the CIHR Institute of Circulatory and Respiratory Health, the CIHR/Rx&D Program, and AstraZeneca Canada, and (2) a Studentship Award from the Alberta Heritage Foundation for Medical Research (AHFMR). R.J.F. is an AHFMR Medical Scientist. Z.-P.F. is a CIHR New Investigator. P.A.B is an AHFMR Clinical Investigator and a Heart and Stroke Foundation of Canada New Investigator. We are grateful to Drs. Susumu Seino and Takashi Miki for making available the knockout mice. We thank Dr. Zonghang Zhao for helpful discussions on in vivo experiments and Lisa Hoyte for assisting in the initial in vivo setup. The work presented here formed part of a thesis presented by H.-S.S. in partial fulfillment of the requirements for a Ph.D. in Neuroscience at the University of Calgary.
REFERENCES Babenko AP, Aguilar-Bryan L, Bryan J (1998) A view of sur/KIR6.X, KATP channels. Annu Rev Physiol 60:667– 687. Barber PA, Hoyte L, Colbourne F, Buchan AM (2004) Temperatureregulated model of focal ischemia in the mouse: a study with histopathological and behavioral outcomes. Stroke 35:1720 –1725. Blondeau N, Plamondon H, Richelme C, Heurteaux C, Lazdunski M (2000) K(ATP) channel openers, adenosine agonists and epileptic preconditioning are stress signals inducing hippocampal neuroprotection. Neuroscience 100:465– 474. Bonde C, Noraberg J, Zimmer J (2002) Nuclear shrinkage and other markers of neuronal cell death after oxygen-glucose deprivation in rat hippocampal slice cultures. Neurosci Lett 327:49 –52. Clark WM, Lessov NS, Dixon MP, Eckenstein F (1997) Monofilament intraluminal middle cerebral artery occlusion in the mouse. Neurol Res 19:641– 648. Davis SM, Albers GW, Diener HC, Lees KR, Norris J (1997) Termination of acute stroke studies involving selfotel treatment. ASSIST Steering Committee. Lancet 349:32. Davis SM, Lees KR, Albers GW, Diener HC, Markabi S, Karlsson G, Norris J (2000) Selfotel in acute ischemic stroke: possible neurotoxic effects of an NMDA antagonist. Stroke 31:347–354. Fujimura N, Tanaka E, Yamamoto S, Shigemori M, Higashi H (1997) Contribution of ATP-sensitive potassium channels to hypoxic hyperpolarization in rat hippocampal CA1 neurons in vitro. J Neurophysiol 77:378 –385. Garcia DA, Franke H, Pissarek M, Nieber K, Illes P (1999) Neuroprotection by ATP-dependent potassium channels in rat neocortical brain slices during hypoxia. Neurosci Lett 273:13–16. Hoyte L, Barber PA, Buchan AM, Hill MD (2004) The rise and fall of NMDA antagonists for ischemic stroke. Curr Mol Med 4:131–136. Ikonomidou C, Turski L (2002) Why did NMDA receptor antagonists fail clinical trials for stroke and traumatic brain injury? Lancet Neurol 1:383–386.
H.-S. Sun et al. / Neuroscience 144 (2007) 1509 –1515 Laake JH, Haug FM, Wieloch T, Ottersen OP (1999) A simple in vitro model of ischemia based on hippocampal slice cultures and propidium iodide fluorescence. Brain Res Brain Res Protoc 4:173– 184. Lees KR, Asplund K, Carolei A, Davis SM, Diener HC, Kaste M, Orgogozo JM, Whitehead J (2000) Glycine antagonist (gavestinel) in neuroprotection (GAIN International) in patients with acute stroke: a randomised controlled trial. GAIN International Investigators. Lancet 355:1949 –1954. Li RA, Leppo M, Miki T, Seino S, Marban E (2000) Molecular basis of electrocardiographic ST-segment elevation. Circ Res 87:837– 839. Light PE, Kanji HD, Fox JE, French RJ (2001) Distinct myoprotective roles of cardiac sarcolemmal and mitochondrial KATP channels during metabolic inhibition and recovery. FASEB J 15:2586 –2594. Lipton P (1999) Ischemic cell death in brain neurons. Physiol Rev 79:1431–1568. Macklis JD, Madison RD (1990) Progressive incorporation of propidium iodide in cultured mouse neurons correlates with declining electrophysiological status: a fluorescence scale of membrane integrity. J Neurosci Methods 31:43– 46. Majid A, He YY, Gidday JM, Kaplan SS, Gonzales ER, Park TS, Fenstermacher JD, Wei L, Choi DW, Hsu CY (2000) Differences in vulnerability to permanent focal cerebral ischemia among 3 common mouse strains. Stroke 31:2707–2714. Miki T, Liss B, Minami K, Shiuchi T, Saraya A, Kashima Y, Horiuchi M, Ashcroft F, Minokoshi Y, Roeper J, Seino S (2001) ATP-sensitive K⫹ channels in the hypothalamus are essential for the maintenance of glucose homeostasis. Nat Neurosci 4:507–512. Miki T, Nagashima K, Seino S (1999) The structure and function of the ATP-sensitive K⫹ channel in insulin-secreting pancreatic betacells. J Mol Endocrinol 22:113–123. Miki T, Nagashima K, Tashiro F, Kotake K, Yoshitomi H, Tamamoto A, Gonoi T, Iwanaga T, Miyazaki J, Seino S (1998) Defective insulin secretion and enhanced insulin action in KATP channel-deficient mice. Proc Natl Acad Sci U S A 95:10402–10406. Modo M, Sowinski P, Hodges H (2000) Conditional discrimination learning in rats with global ischaemic brain damage. Behav Brain Res 111:213–221. Morris GF, Bullock R, Marshall SB, Marmarou A, Maas A, Marshall LF (1999) Failure of the competitive N-methyl-D-aspartate antagonist selfotel (CGS 19755) in the treatment of severe head injury: results of two phase III clinical trials. The Selfotel Investigators. J Neurosurg 91:737–743.
1515
Nelson A, Lebessi A, Sowinski P, Hodges H (1997) Comparison of effects of global cerebral ischaemia on spatial learning in the standard and radial water maze: relationship of hippocampal damage to performance. Behav Brain Res 85:93–115. Paxinos G, Franklin KBJ (2001) The mouse brain in stereotaxic coordinates. San Diego, CA: Academic Press. Seino S (1999) ATP-sensitive potassium channels: a model of heteromultimeric potassium channel/receptor assemblies. Annu Rev Physiol 61:337–362. Seino S, Miki T (2004) Gene targeting approach to clarification of ion channel function: studies of Kir6.x null mice. J Physiol 554:295– 300. Shimizu K, Lacza Z, Rajapakse N, Horiguchi T, Snipes J, Busija DW (2002) MitoK(ATP) opener, diazoxide, reduces neuronal damage after middle cerebral artery occlusion in the rat. Am J Physiol Heart Circ Physiol 283:H1005–H1011. Sun HS, Feng ZP, Miki T, Seino S, French RJ (2006) Enhanced neuronal damage after ischemic insults in mice lacking Kir6.2containing ATP-sensitive K⫹ channels. J Neurophysiol 95:2590 – 2601. Suzuki M, Fujikura K, Inagaki N, Seino S, Takata K (1997) Localization of the ATP-sensitive K⫹ channel subunit Kir6.2 in mouse pancreas. Diabetes 46:1440 –1444. Yamada K, Ji JJ, Yuan H, Miki T, Sato S, Horimoto N, Shimizu T, Seino S, Inagaki N (2001) Protective role of ATP-sensitive potassium channels in hypoxia-induced generalized seizure. Science 292: 1543–1546. Yamamoto S, Tanaka E, Higashi H (1997) Mediation by intracellular calcium-dependent signals of hypoxic hyperpolarization in rat hippocampal CA1 neurons in vitro. J Neurophysiol 77:386 –392. Yin HZ, Sensi SL, Ogoshi F, Weiss JH (2002) Blockade of Ca2⫹permeable AMPA/kainate channels decreases oxygen-glucose deprivation-induced Zn2⫹ accumulation and neuronal loss in hippocampal pyramidal neurons. J Neurosci 22:1273–1279. Zawar C, Plant TD, Schirra C, Konnerth A, Neumcke B (1999) Celltype specific expression of ATP-sensitive potassium channels in the rat hippocampus. J Physiol 514 (Pt 2):327–341. Zhou M, Tanaka O, Suzuki M, Sekiguchi M, Takata K, Kawahara K, Abe H (2002) Localization of pore-forming subunit of the ATPsensitive K(⫹)-channel, Kir6.2, in rat brain neurons and glial cells. Brain Res Mol Brain Res 101:23–32.
(Accepted 19 October 2006) (Available online 18 December 2006)
Kir6.2-CONTAINING ATP-SENSITIVE POTASSIUM CHANNELS PROTECT CORTICAL NEURONS FROM ISCHEMIC/ANOXIC INJURY IN VITRO AND IN VIVO H.-S. SUN,a Z.-P. FENG,b P. A. BARBER,c A. M. BUCHANc AND R. J. FRENCHa*
Cerebral strokes, a leading cause of morbidity and mortality in industrialized countries, cause acute cerebral ischemia and reperfusion injuries that result in severe neurological damage (Lipton, 1999). A major repercussion of the loss of oxygen and nutrients caused by ischemic stroke is a concomitant massive release of the excitatory neurotransmitter glutamate, leading to a intracellular Ca2⫹ overload and cell death (Lipton, 1999). This discovery led to clinical trials to determine the efficacy of anti-excitotoxic therapies (Davis et al., 1997, 2000; Morris et al., 1999; Lees et al., 2000); unfortunately, these therapies failed to benefit stroke patients (Ikonomidou and Turski, 2002; Hoyte et al., 2004), underscoring the still unmet need for effective neuroprotective agents and therapeutic strategies to protect the brain from the damage caused by cerebral stroke, and improve the recovery of stroke patients. Recently, neuroprotective effects of ATP-sensitive potassium channels (KATP channels) against hypoxic induced generalized seizures have been demonstrated (Yamada et al., 2001; Seino and Miki, 2004). KATP channels, which are hetero-octamers composed of pore-forming Kir6.x (6.1 or 6.2) subunits and sulfonylurea receptor (SUR1 or SUR2) regulatory subunits (Babenko et al., 1998; Miki et al., 1999; Seino, 1999), are expressed in many types of tissues and cells. Kir6.2-containing KATP channels are present in a variety of cell types in the brain, including pyramidal cells, interneurons, granule cells and glial cells of the hippocampus (Zawar et al., 1999; Zhou et al., 2002; Sun et al., 2006), neurons of the substantia nigra pars reticulata (Yamada et al., 2001), and also in other parts of the CNS, including the hypothalamus (Miki et al., 2001). KATP channel activation, which is regulated by intracellular ATP and ADP concentrations, has been associated with acute and delayed protective responses to ischemia in a variety of organs and tissues. An increase in the ratio of [ATP] to [ADP] decreases the probability of KATP channel opening, whereas a decrease in this ratio promotes opening of these channels. KATP channel activation tends to hyperpolarize the cell by shifting the membrane potential toward the potassium equilibrium potential. Thus, KATP channels couple cell metabolism to cell membrane potential, thereby regulating many cellular activities. Recently we (Sun et al., 2006), identified a fundamental, neuroprotective role for KATP channels in the hippocampus, and discovered a KATP-dependent electrophysiological response of hippocampal neurons to anoxic stress. Here, we have focused our attention on the neuroprotective effects of KATP channels against damage from
a
Department of Physiology and Biophysics, Faculty of Medicine, University of Calgary, 3330 Hospital Drive Northwest, Calgary, Alberta, Canada T2N 4N1
b
Department of Physiology, University of Toronto, 1 King’s College Circle, Toronto, Ontario, Canada M5S 1A8
c
Department of Clinical Neuroscience, University of Calgary, Calgary, Alberta, Canada
Abstract—ATP-sensitive potassium (KATP) channels are weak inward rectifiers that appear to play an important role in protecting neurons against ischemic damage. Cerebral stroke is a major health issue, and vulnerability to stroke damage is regional within the brain. Thus, we set out to determine whether KATP channels protect cortical neurons against ischemic insults. Experiments were performed using Kir6.2ⴚ/ⴚ KATP channel knockout and Kir6.2ⴙ/ⴙ wildtype mice. We compared results obtained in Kir6.2ⴚ/ⴚ and wildtype mice to evaluate the protective role of KATP channels against focal ischemia in vivo, and, using cortical slices, against anoxic stress in vitro. Immunohistochemistry confirmed the presence of KATP channels in the cortex of wildtype, but not Kir6.2ⴚ/ⴚ, mice. Results from in vivo and in vitro experimental models indicate that Kir6.2-containing KATP channels in the cortex provide protection from neuronal death. Briefly, in vivo focal ischemia (15 min) induced severe neurological deficits and large cortical infarcts in Kir6.2ⴚ/ⴚ mice, but not in wildtype mice. Imaging analyses of cortical slices exposed briefly to oxygen and glucose deprivation (OGD) revealed a substantial number of damaged cells (propidium iodide–labeled) in the Kir6.2ⴚ/ⴚ OGD group, but few degenerating neurons in the wildtype OGD group, or in the wildtype and Kir6.2ⴚ/ⴚ control groups. Slices from the three control groups had far more surviving cells (anti-NeuN antibodylabeled) than slices from the Kir6.2ⴚ/ⴚ OGD group. These findings suggest that stimulation of endogenous cortical KATP channels may provide a useful strategy for limiting the damage that results from cerebral ischemic stroke. © 2006 IBRO. Published by Elsevier Ltd. All rights reserved. Key words: cerebral cortex, cerebral ischemia, focal cerebral ischemia, neuronal death, neuroprotection. *Corresponding author. Tel: ⫹1-403-220-8389 or ⫹1-403-220-6893; fax: ⫹1-403-283-8731. E-mail address: [email protected] (R. J. French). Abbreviations: CCA, common carotid artery; ECA, external carotid artery; ICA, internal carotid artery; KATP, ATP-sensitive potassium; Kir6.2, subunit of the ATP-sensitive potassium channel; Kir6.2⫺/⫺, mutant mice lacking the Kir6.2 subunit of the ATP-sensitive potassium channel; LDF, laser-Doppler flowmetry; MCA, middle cerebral artery; MCAO, middle cerebral artery occlusion; NS, not significant; OGD, oxygen and glucose deprivation; PI, propidium iodide; rCBF, regional cerebral blood flow; TTC, triphenyltetrazolium.
0306-4522/07$30.00⫹0.00 © 2006 IBRO. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.neuroscience.2006.10.043
1509
1510
H.-S. Sun et al. / Neuroscience 144 (2007) 1509 –1515
focal ischemia and oxygen and glucose deprivation (OGD) in the cerebral cortex. It is generally accepted that global and focal ischemia induces specific cell injuries in different regions of the brain (Lipton, 1999). For example, a 15 min global ischemia can cause damage selectively to the CA1 region of the hippocampus (Nelson et al., 1997; Modo et al., 2000). There are at least two reasons why the consequences of focal ischemia might differ from those of global ischemia. First, in the ischemic core of a focal lesion, blood flow is generally higher than it is in global ischemia. Thus, longer episodes might be needed to induce a similar amount of damage. Second, in a focal lesion, there is a progression in the severity of ischemia from the penumbra to the core, resulting in different metabolic conditions within the affected regions. This diversity makes focal ischemia more complex than global ischemia, and underlines the difficulty of predicting its effects. Although we have shown that KATP channels provide significant protection against hypoxia and ischemia in hippocampal neurons (Sun et al., 2006), whether, and how, KATP channels contribute to neuroprotection in the cortex has not been examined in detail. We postulated that KATP channels play a less important role in the cortex since hippocampal neurons are especially sensitive to global ischemia. Contrary to our hypothesis, we report that Kir6.2containing KATP channels protect neurons in all cortical layers to a similar extent to that seen for neurons in our in vitro studies in the hippocampus (Sun et al., 2006). These results indicate that Kir6.2-containing KATP channels play a similarly fundamental role in neuroprotection in cortical and hippocampal neurons.
EXPERIMENTAL PROCEDURES Mice Mutant mice lacking the Kir6.2 subunit of the ATP-sensitive potassium channel (Kir6.2⫺/⫺) (Miki et al., 1998) and wildtype mice with same background were used in these experiments, as previously described (Miki et al., 1998; Li et al., 2000; Yamada et al., 2001). Adult mice used were male, aged between 10 and 12 weeks, and had a body weight of 25–35 g. Professor Susumu Seino and Dr. Takashi Miki, Division of Cellular and Molecular Medicine, Kobe University Graduate School of Medicine, Kobe, Japan provided breeding stock for the Kir6.2⫺/⫺ mice. Knockout of the Kir6.2 gene was confirmed by PCR genotyping. Kir6.2⫺/⫺ mice develop normally with no apparent abnormalities in general appearance, and are fertile (Miki et al., 1998). The brain cytoarchitecture of wildtype and Kir6.2⫺/⫺ mice appears to be comparable (Miki et al., 1998; Yamada et al., 2001). Basic physiological parameters, including body weight, body temperature, blood glucose, blood gas, pH, blood pressure, heart rate, ECG, respiratory rate, and O2 consumption, of the Kir6.2⫺/⫺ mice and wildtype mice were not significantly different (Miki et al., 1998; Li et al., 2000; Yamada et al., 2001). All mice used in this study were bred and maintained in the Animal Resources Centre at the University of Calgary, with an ambient temperature of 20⫾1 °C and a 12-h light/dark cycle. In preparation for immunohistochemistry (Sun et al., 2006), mice were deeply anesthetized, and transcardiacally perfused with heparinized saline followed by 4% paraformaldehyde in PBS. Brains were placed in the same fixative at 4 °C overnight and subsequently cut into 50 m sections for immunostaining.
Focal cerebral ischemia We used a modified monofilament intraluminal middle cerebral artery occlusion (MCAO) procedure to produce focal ischemia (Clark et al., 1997; Majid et al., 2000). Mice were anesthetized with isoflurane (3% initial, 1–1.5% maintenance) in O2 and air (80: 20%). Briefly, under an operating microscope, the left common carotid artery (CCA), the left external carotid artery (ECA), and the left internal carotid artery (ICA) were isolated, and a 6 – 0 suture was tied at the origin of the ECA and at the distal end of the ECA. The left CCA and the ICA were also temporarily occluded. A silicon-coated nylon suture (8 – 0) was introduced into the ECA and pushed through the ICA until resistance was felt; the filament was inserted approximately 9 –10 mm from the carotid bifurcation, effectively blocking the origin of the middle cerebral artery (MCA) (Barber et al., 2004). The diameter of the tip of the coated suture measured between 180 and 220 m. The suture was removed after 15 min, and the ECA was permanently tied off. Shamoperated mice underwent a surgical procedure that involved permanent ligation of the ECA and temporary occlusion of the CCA, without occlusion of the MCA. During the procedure, body temperature was monitored and maintained at 37⫾0.5 °C. Transcranial measurements of cerebral blood flow (CBF) were made by laser-Doppler flowmetry (LDF). An 0.5 mm diameter micro-fiber laser-Doppler probe (Perimed, Järfälla, Sweden) was attached to the skull with cyanoacrylate glue 6 mm lateral and 1 mm posterior to the bregma. While the mice were under general anesthesia, we monitored regional cerebral blood flow (rCBF) within the infarct core and in the parietal cortex (penumbra). The surgical procedure was considered adequate if a ⱖ70% reduction in rCBF occurred immediately after placement of the intraluminal occluding suture (Barber et al., 2004); otherwise mice were excluded. Blood flow velocity was measured prior to ischemia, and during MCAO and reperfusion using Perisoft (Perimed, Inc.). Mice were permitted to recover from the anesthesia at room temperature. Twenty-four hours after the procedure, we assessed neurobehavioral parameters and then removed the brains from deeply anesthetized mice. Infarct volume was assessed using a standard triphenyltetrazolium (TTC) staining technique.
Neurobehavioral evaluation Following ischemia, upon recovery from the anesthesia, mice were assessed for focal functional deficit to determine the impact of the MCA occlusion. At 24-hours post-ischemia, each mouse was scored for neurological function using a six-point standard scale (Clark et al., 1997) (neurological deficit scores: 0: no neurological deficit; 1: retracts left forepaw when lifted by the tail; 2: circles to the left; 3: falls while walking; 4: does not walk spontaneously; 5: dead).
TTC staining and infarct volume assessment In MCAO experiments, we performed infarct analysis using standard TTC (Sigma, St. Louis, MO, USA) staining techniques (Clark et al., 1997; Majid et al., 2000), and NIH Image J software (National Institutes of Health, Bethesda, MD, USA) to estimate infarct volume. Brains were sliced into 1-mm coronal sections with a matrix at defined bregma locations (Paxinos and Franklin, 2001). Each slice was incubated in 2% TTC in phosphate buffer, and stained at 37 °C for 30 min. Each TTC-stained section was digitally photographed, and infarct areas were traced onto templates representing 1 mm coronal slices. Templates were used to correct any brain edema produced by the infarction, allowing for a more accurate assessment of infarct volume. The infarct area from each template was then digitally traced using NIH Image J. The infarct volume was calculated by summing infarct areas from individual slices (Majid et al., 2000). Infarct volumes of both wildtype and
H.-S. Sun et al. / Neuroscience 144 (2007) 1509 –1515 Kir6.2⫺/⫺ were compared statistically to determine any significant difference.
Cortical slice preparations Under deep anesthesia, the brains were rapidly removed and horizontal slices (250 m) were cut using a vibratome. Brain slices were immediately transferred to a holding chamber, containing ACSF solution saturated with carbogen (95% O2 and 5% CO2), where they were held for at least 1 h at 30 °C⫾2 °C prior to OGD. The composition of the ACSF was (in mM): 124 NaCl, 3 KCl, 1.25 NaH2PO4, 2 MgCl2, 26 NaHCO3, 2 CaCl2, and 10 dextrose (pH 7.4). ACSF was saturated with 95% O2 and 5% CO2 at room temperature.
1511
was used to avoid any mechanical damage of the sectioning and consequent distortion of the images. For imaging, the slices were scanned to determine a depth halfway between the slice surfaces and the midslice section was used for confocal imaging (Sun et al., 2006). For a region of interest, we produced 3D digital reconstructions from a series of confocal images taken at 0.5 m intervals, and generated optical stacks of 10 images for the figures.
Cell counting We used NIH Image J to determine the densities of injured (PI stained) and surviving (anti-NeuN antibody stained) cells in cortical brain slices from each experimental group. Cell counts were routinely obtained from a reconstructed volume measuring 400⫻400⫻5 m.
Immunostaining using Kir6.2 antibody in cortex Brain slice sections were blocked in 3% normal goat serum/0.3% Triton X-100/0.1% BSA in PBS at room temperature for 1 h then incubated with rabbit anti-Kir6.2 antibody (1:500; kindly supplied by Professor S. Seino, Division of Cellular and Molecular Medicine, Kobe University Graduate School of Medicine, Kobe, Japan) at 4 °C overnight. The characteristics of this antibody have been reported elsewhere (Suzuki et al., 1997; Zhou et al., 2002). Subsequently, sections were incubated with affinity-purified secondary antibody, goat anti-rabbit Cy3 antibody (1:200; Chemicon, Temecula, CA, USA). Sections were then rinsed, dried, and coverslipped with Dako fluorescence mounting medium. Images were viewed under a confocal laser scanning microscope.
OGD in cortical slices We determined the survival of cortical neurons under experimental ischemic conditions caused by OGD as previously described (Sun et al., 2006). Four sets of experiments were performed. In control experiments, cortical slices from either wildtype or Kir6.2⫺/⫺ mice were incubated in ACSF saturated with carbogen (95% O2 and 5% CO2) for 30 min. In OGD experiments, cortical slices from either wildtype or Kir6.2⫺/⫺ mice were incubated under experimental ischemic conditions in ACSF lacking glucose and saturated with 95% N2 and 5% CO2 for the 30 min. All cortical brain slices were allowed to recover for 60 min in normal ACSF buffer saturated with 95% O2 and 5% CO2. Injured and surviving neurons were identified by laser confocal microscopy using double staining (PI and anti-NeuN antibody). The number of injured and surviving neurons were counted and compared under each experimental condition.
PI labeling Nuclear debris and nonviable neurons in cortical brain slices after OGD were labeled by propidium iodide (PI) (5 g/ml) (Macklis and Madison, 1990; Laake et al., 1999; Bonde et al., 2002; Yin et al., 2002).
Immunocytochemistry: NeuN In double labeling experiments, following PI staining, sections were incubated in mouse anti-NeuN antibody (1:100; Chemicon) as previously described (Sun et al., 2006). These sections were then reacted with goat anti-mouse fluorescein-tagged secondary antibody (1:200; Chemicon). Sections were rinsed, dried, and coverslipped with Dako fluorescence mounting medium.
Confocal imaging Antibody labeling and double labeling were imaged using a laser confocal scanning microscope (Olympus LSM-GB200; Olympus, Tokyo, Japan) and analyzed with a three-dimensional (3D) constructor (Image-Pro Plus). During the confocal imaging, caution
Statistical analysis Data were expressed as the mean⫾S.D. and significance was determined using ANOVA and multiple comparison tests (SigmaStat3.0). The numbers of damaged or surviving cells in the four experimental groups were first tested for any statistically significant difference among the groups using a one-way ANOVA test. Since there were statistically significant differences (P⬍0.05, ANOVA) among the four groups, we used Dunnett’s and Duncan’s tests for multiple comparisons to further evaluate the differences between the individual groups. For all tests, P⬍0.05 was considered significant.
RESULTS In this study, (a) we used immunohistochemistry to confirm that Kir6.2 protein is absent from cortex in the Kir6.2⫺/⫺ mice, but is present throughout the cortex of wildtype mice; (b) we showed that Kir6.2⫺/⫺ mice suffer severe neurological consequences, with associated grossly defined cerebral infarcts, following a 15 min MCAO, while wildtype mice and shams show no ill effects; and (c) we used PI and anti-NeuN antibody double labeling to show that 30 min OGD causes extensive cell death in cortical slices from Kir6.2⫺/⫺ mice, but causes little or no damage in wildtype cortical slices. These results lead us to conclude that cortical Kir6.2-containing KATP channels play an important role for normal neuroprotective responses in the cortex. Expression of Kir6.2 protein in cortex: anti-Kir6.2 antibody staining Using an anti-Kir6.2 antibody, we demonstrated by immunohistochemistry that Kir6.2 protein is expressed in pyramidal-shaped neurons of the cortex in adult wildtype mice (Fig. 1). In contrast, we observed no immunoreactivity to the Kir6.2 protein in the cortex of the Kir6.2⫺/⫺ mice (Fig. 1). In vivo experiments: focal cerebral ischemia (MCAO) We found no significant difference in body weights or blood glucose levels between the Kir6.2⫺/⫺ and wildtype mice. Body weights were 30.65⫾2.51 g for the wildtype and 31.23⫾2.58 g for the Kir6.2⫺/⫺ (n⫽10/group, NS, not significant: P⫽0.987). Blood glucose levels were 6.62⫾1.02 mmol/L for the wildtype group and 7.88⫾1.79 mmol/L for the Kir6.2⫺/⫺ group (n⫽10/group, NS: P⫽0.069). The rCBF, estimated by LDF, was reduced by more than 70%
1512
H.-S. Sun et al. / Neuroscience 144 (2007) 1509 –1515
15 min MCAO. Infarct volume was significantly larger in Kir6.2⫺/⫺ brains exposed to 15-min MCAO than in wildtype brains treated similarly (n⫽10 per group) (Fig. 2). In addition, we detected no damage in the brains of sham-operated mice, whether Kir6.2⫺/⫺ and wildtype (n⫽5/each group).
Fig. 1. Anti-Kir6.2 antibody immunostaining of the cortex of the brain (60⫻ objective). The section shown spans approximately from layer III to layer V, from top to bottom. Most cells of the cortical cells express Kir6.2 protein in wildtype mice. In contrast, the cortex of Kir6.2⫺/⫺ mice did not show any immunoreactivity for the Kir6.2 protein. Scale bars⫽50 m.
from pre-occlusion values after introduction of the suture, indicating that the procedure resulted in adequate occlusion (Barber et al., 2004). rCBF measurements were similar in Kir6.2⫺/⫺ and wildtype groups before, during, and after, the MCAO procedure. Neurobehavioral scores In general, as compared with the wildtype mice, Kir6.2⫺/⫺ mice did not respond well to MCAO. Kir6.2⫺/⫺ mice had a mortality rate over 50% (10 survivors, n⫽20) 24 h after a 15 min MCAO, and survivors showed signs of severe neurological deficits (NB score 4, n⫽10). We observed no cerebral hemorrhage or other bleeding around the operated area in the neck of these Kir6.2⫺/⫺ mice. No neurological deficits were detected in wildtype mice exposed to a 15 min MCAO (NB score 0, n⫽10) or in sham-operated mice, whether Kir6.2⫺/⫺ or wildtype (NB score 0, n⫽5/ group). Infarct volumes for the MCAO mice Twenty-four hours after the MCAO, the TTC staining revealed significant damage in the brains of Kir6.2⫺/⫺ mice exposed to a 15 min MCAO (Fig. 2); no detectable damage was observed in the brains of wildtype mice exposed to a
Fig. 2. TTC-stained slices of brains. (A) Slices from Kir6.2⫺/⫺ mouse. (B) Slices from a wildtype mouse exposed to brief focal ischemia. Mice were killed and slices prepared 24 h after exposure to a 15 min MCAO. Kir6.2⫺/⫺ mouse brains had extreme damage (unstained white area) in the cortex. The wildtype mouse brain showed no gross damage with the TTC stain. The thickness of each slice was 1 mm. Scale bars⫽2 mm. Slices were consecutive in the order right, bottom to top, followed by left bottom to top, at the following locations: bregma 1.34, 1.14, 0.14, ⫺0.94, ⫺2.06, ⫺3.08, ⫺4.04 mm (Paxinos and Franklin, 2001). They are viewed from the anterior side. (C) Comparison of the infarct volumes from Kir6.2⫺/⫺ and wildtype mice after 15 min MCAO. Infarct volumes in the brains of Kir6.2⫺/⫺ mice were significantly larger than that of Kir6.2 wildtype mice (n⫽10 per group, P⬍0.001).
H.-S. Sun et al. / Neuroscience 144 (2007) 1509 –1515
1513
Due to the all-or-nothing nature of the damage caused by brief focal ischemia, which caused severe gross injury (in TTC stain) in Kir6.2⫺/⫺ mice, and no gross infarction in the wildtype mice, we did not further evaluate the brains of these mice using other histological staining methods, such as H&E staining. In vitro experimental ischemia (ODG) on cortical brain slices Following 30 min OGD, cortical slices, double stained with PI and anti-NeuN antibody, revealed extensive cell damage in the Kir6.2⫺/⫺ OGD group, but minimal damage in the wildtype OGD group. In the Kir6.2⫺/⫺ OGD group, damage was apparent in all regions of the cortex (Fig. 3A, B). The numerous injured cells in the cortical fields were revealed as the PI-labeled cells in the confocal images. Relatively few surviving cells, labeled with anti-NeuN antibody, were visible in the confocal images obtained from Kir6.2⫺/⫺ cortical slices. In contrast, we observed few PIlabeled injured cells in the wildtype OGD, or the wildtype, and Kir6.2⫺/⫺ control groups; anti-NeuN antibody labeling indicated that surviving cells were abundant in all fields in cortical slices for these three groups. PI staining for injured cells: cell counts The numbers of injured cells stained with PI in all four experimental groups (Kir6.2⫺/⫺ and wildtype OGD groups, and Kir6.2⫺/⫺ and wildtype control groups) were counted and statistically compared. The results are shown in Fig. 3C. There were significantly more PI-stained, damaged cells in all cortical fields from the Kir6.2⫺/⫺ OGD group than in the wildtype OGD group, or the Kir6.2⫺/⫺, and wildtype control groups. Anti-NeuN antibody staining for surviving cells: cell counts To make a more thorough comparison, we used anti-NeuN antibody to label surviving cells in the same cortical slices. The numbers of anti-NeuN antibody-labeled surviving cells for all four experimental groups (Kir6.2⫺/⫺ and wildtype OGD groups, and Kir6.2⫺/⫺ and wildtype control groups) were counted and statistically compared. These results are shown in Fig. 3C. We observed substantially fewer surviving neurons in all cortical fields from the Kir6.2⫺/⫺ OGD group than were detected in the wildtype OGD, or Kir6.2⫺/⫺ and wildtype control groups. Statistical comparison tests of the four groups Results from ANOVA and multiple comparison tests showed a statistically significant difference among the four experimental groups (ANOVA: P⬍0.001; multiple comparison test: P⬍0.05). The numbers of injured and surviving cells in the cortical slices from the Kir6.2⫺/⫺ OGD group differed significantly from those seen in the three other experimental groups (wildtype OGD, Kir6.2⫺/⫺ and wildtype controls). Results were consistent in all cortical fields. Thus, we confirmed quantitatively that there were substantially more damaged neurons, and fewer surviving neu-
Fig. 3. (A) Double labeling, with anti-NeuN antibody (green), for surviving neurons, and PI (red), for injured cells, in cortex (4⫻ objective) exposed to OGD for 30 min. In general, there were more damaged cells labeled with PI in the Kir6.2⫺/⫺ OGD group than in the three other experimental groups (wildtype OGD, Kir6.2⫺/⫺ control, and wildtype control). In contrast, the three other experimental groups (wildtype OGD, Kir6.2⫺/⫺ control, and wildtype control) showed mostly surviving neurons labeled with anti-NeuN antibody. Scale bars⫽200 m. (B) Double labeling under higher magnification (40⫻) after 30 min of OGD. The sections shown span approximately from cortical layer III to layer V (c.f. Fig. 1A.). The Kir6.2⫺/⫺ OGD group showed more PI-labeled, damaged cells than the other three experimental groups (wildtype OGD, or Kir6.2⫺/⫺, and wildtype controls), which had mostly anti-NeuN antibody-labeled surviving neurons labeled. Scale bars⫽50 m. (C) Numbers of PI (black) and anti-NeuN antibody (gray) stained cells. The Kir6.2⫺/⫺ OGD group had the highest numbers of damaged cells and the lowest numbers of surviving cells (* P⬍0.05) compared with the three other experimental groups (wildtype OGD, or controls for either the Kir6.2⫺/⫺ or the wildtype).
1514
H.-S. Sun et al. / Neuroscience 144 (2007) 1509 –1515
rons, in all fields of the cortex in the Kir6.2⫺/⫺ OGD group compared with the three other groups (wildtype OGD, Kir6.2⫺/⫺ and wildtype controls).
DISCUSSION Kir6.2 expression Using an anti-Kir6.2 antibody, we confirmed that Kir6.2 protein is expressed in all regions of the cerebral cortex in adult wildtype mice; we did not detect Kir6.2 expression in Kir6.2⫺/⫺ mice, paralleling results we obtained for the expression of these Kir6.2 in the mouse hippocampus (Sun et al., 2006). In other studies, Kir6.2 protein has also been detected in the mouse pancreas (Suzuki et al., 1997) and the rat brain (Zhou et al., 2002). In addition, single-cell RT-PCR has detected KATP mRNA in a variety of neuronal populations in rat brains (Zawar et al., 1999). In this study, we showed that the Kir6.2 protein is widely distributed in the cortical neurons of the wildtype mice. Thus, we found that overall distribution of Kir6.2 in the different regions of the mouse brain resembles that reported for rats. KATP channels and neuroprotection To date, there have been few studies detailing the role played by KATP channels in protecting neurons from the harmful effects of ischemia and hypoxia (Fujimura et al., 1997; Yamamoto et al., 1997; Garcia et al., 1999; Blondeau et al., 2000; Yamada et al., 2001). Recently, one group (Shimizu et al., 2002) tested the neuroprotective effectiveness of diazoxide, used as a mitochondrial KATP channel agonist, against the effects of transient focal cerebral ischemia induced by MCAO. Diazoxide reduced infarct volume by 40%; in other experiments, 5-hydroxydecanoate (5-HD), another putative selective mitochondrial KATP channel blocker, completely prevented the diazoxide’s effect. Although the role of mitochondrial KATP channels remains controversial, results obtained in other systems, and based on similar pharmacological criteria, were consistent with complementary roles in cytoprotection for plasmalemmal and mitochondrial KATP channels (Light et al., 2001). Kir6.2 is thought to be the predominant pore-forming subunit of plasmalemmal KATP channels in skeletal muscle and neurons. Kir6.2 has been shown to provide protection during hypoxia-induced generalized seizure (Yamada et al., 2001) and electrical recordings in hippocampal neurons have indicated that plasmalemmal Kir6.2-containing KATP channels protect against metabolic stress by exerting a hyperpolarizing influence on plasma membranes, thereby attenuating electrical activity (Sun et al., 2006). Our results are entirely consistent with the view that a similar mechanism underlies Kir6.2-mediated neuroprotection in the cortex. This does not preclude an additional, complementary protective contribution by glial cells that also express Kir6.2. In this study, to evaluate in detail a neuroprotective role for Kir6.2-containing KATP channels in the cerebral cortex, we focused on determining rates of cell death and survival in cortical neurons exposed to ischemia. Notwithstanding
suggestions of differences in regional sensitivity to ischemic stress in the brain, we report here that Kir6.2-containing KATP channels protect neurons in all layers of the cortex to a degree similar to that seen in the hippocampus (Sun et al., 2006), indicating a fundamental role for Kir6.2containing KATP channels in neuroprotection in cortical and hippocampal neurons, and likely throughout the brain. In future, KATP channel modulators may prove to be clinically useful neuroprotective agents, as part of a combination therapy for stroke management. Acknowledgments—This work was supported by a grant from the Canadian Institutes of Health Research (to R.J.F.). H.-S.S. was recipient of Doctor Research Awards from (1) the Heart and Stroke Foundation of Canada–Focus on Stroke Training Initiative Program, in partnership with the Canadian Stroke Network, Heart and Stroke Foundation of Canada, the CIHR Institute of Circulatory and Respiratory Health, the CIHR/Rx&D Program, and AstraZeneca Canada, and (2) a Studentship Award from the Alberta Heritage Foundation for Medical Research (AHFMR). R.J.F. is an AHFMR Medical Scientist. Z.-P.F. is a CIHR New Investigator. P.A.B is an AHFMR Clinical Investigator and a Heart and Stroke Foundation of Canada New Investigator. We are grateful to Drs. Susumu Seino and Takashi Miki for making available the knockout mice. We thank Dr. Zonghang Zhao for helpful discussions on in vivo experiments and Lisa Hoyte for assisting in the initial in vivo setup. The work presented here formed part of a thesis presented by H.-S.S. in partial fulfillment of the requirements for a Ph.D. in Neuroscience at the University of Calgary.
REFERENCES Babenko AP, Aguilar-Bryan L, Bryan J (1998) A view of sur/KIR6.X, KATP channels. Annu Rev Physiol 60:667– 687. Barber PA, Hoyte L, Colbourne F, Buchan AM (2004) Temperatureregulated model of focal ischemia in the mouse: a study with histopathological and behavioral outcomes. Stroke 35:1720 –1725. Blondeau N, Plamondon H, Richelme C, Heurteaux C, Lazdunski M (2000) K(ATP) channel openers, adenosine agonists and epileptic preconditioning are stress signals inducing hippocampal neuroprotection. Neuroscience 100:465– 474. Bonde C, Noraberg J, Zimmer J (2002) Nuclear shrinkage and other markers of neuronal cell death after oxygen-glucose deprivation in rat hippocampal slice cultures. Neurosci Lett 327:49 –52. Clark WM, Lessov NS, Dixon MP, Eckenstein F (1997) Monofilament intraluminal middle cerebral artery occlusion in the mouse. Neurol Res 19:641– 648. Davis SM, Albers GW, Diener HC, Lees KR, Norris J (1997) Termination of acute stroke studies involving selfotel treatment. ASSIST Steering Committee. Lancet 349:32. Davis SM, Lees KR, Albers GW, Diener HC, Markabi S, Karlsson G, Norris J (2000) Selfotel in acute ischemic stroke: possible neurotoxic effects of an NMDA antagonist. Stroke 31:347–354. Fujimura N, Tanaka E, Yamamoto S, Shigemori M, Higashi H (1997) Contribution of ATP-sensitive potassium channels to hypoxic hyperpolarization in rat hippocampal CA1 neurons in vitro. J Neurophysiol 77:378 –385. Garcia DA, Franke H, Pissarek M, Nieber K, Illes P (1999) Neuroprotection by ATP-dependent potassium channels in rat neocortical brain slices during hypoxia. Neurosci Lett 273:13–16. Hoyte L, Barber PA, Buchan AM, Hill MD (2004) The rise and fall of NMDA antagonists for ischemic stroke. Curr Mol Med 4:131–136. Ikonomidou C, Turski L (2002) Why did NMDA receptor antagonists fail clinical trials for stroke and traumatic brain injury? Lancet Neurol 1:383–386.
H.-S. Sun et al. / Neuroscience 144 (2007) 1509 –1515 Laake JH, Haug FM, Wieloch T, Ottersen OP (1999) A simple in vitro model of ischemia based on hippocampal slice cultures and propidium iodide fluorescence. Brain Res Brain Res Protoc 4:173– 184. Lees KR, Asplund K, Carolei A, Davis SM, Diener HC, Kaste M, Orgogozo JM, Whitehead J (2000) Glycine antagonist (gavestinel) in neuroprotection (GAIN International) in patients with acute stroke: a randomised controlled trial. GAIN International Investigators. Lancet 355:1949 –1954. Li RA, Leppo M, Miki T, Seino S, Marban E (2000) Molecular basis of electrocardiographic ST-segment elevation. Circ Res 87:837– 839. Light PE, Kanji HD, Fox JE, French RJ (2001) Distinct myoprotective roles of cardiac sarcolemmal and mitochondrial KATP channels during metabolic inhibition and recovery. FASEB J 15:2586 –2594. Lipton P (1999) Ischemic cell death in brain neurons. Physiol Rev 79:1431–1568. Macklis JD, Madison RD (1990) Progressive incorporation of propidium iodide in cultured mouse neurons correlates with declining electrophysiological status: a fluorescence scale of membrane integrity. J Neurosci Methods 31:43– 46. Majid A, He YY, Gidday JM, Kaplan SS, Gonzales ER, Park TS, Fenstermacher JD, Wei L, Choi DW, Hsu CY (2000) Differences in vulnerability to permanent focal cerebral ischemia among 3 common mouse strains. Stroke 31:2707–2714. Miki T, Liss B, Minami K, Shiuchi T, Saraya A, Kashima Y, Horiuchi M, Ashcroft F, Minokoshi Y, Roeper J, Seino S (2001) ATP-sensitive K⫹ channels in the hypothalamus are essential for the maintenance of glucose homeostasis. Nat Neurosci 4:507–512. Miki T, Nagashima K, Seino S (1999) The structure and function of the ATP-sensitive K⫹ channel in insulin-secreting pancreatic betacells. J Mol Endocrinol 22:113–123. Miki T, Nagashima K, Tashiro F, Kotake K, Yoshitomi H, Tamamoto A, Gonoi T, Iwanaga T, Miyazaki J, Seino S (1998) Defective insulin secretion and enhanced insulin action in KATP channel-deficient mice. Proc Natl Acad Sci U S A 95:10402–10406. Modo M, Sowinski P, Hodges H (2000) Conditional discrimination learning in rats with global ischaemic brain damage. Behav Brain Res 111:213–221. Morris GF, Bullock R, Marshall SB, Marmarou A, Maas A, Marshall LF (1999) Failure of the competitive N-methyl-D-aspartate antagonist selfotel (CGS 19755) in the treatment of severe head injury: results of two phase III clinical trials. The Selfotel Investigators. J Neurosurg 91:737–743.
1515
Nelson A, Lebessi A, Sowinski P, Hodges H (1997) Comparison of effects of global cerebral ischaemia on spatial learning in the standard and radial water maze: relationship of hippocampal damage to performance. Behav Brain Res 85:93–115. Paxinos G, Franklin KBJ (2001) The mouse brain in stereotaxic coordinates. San Diego, CA: Academic Press. Seino S (1999) ATP-sensitive potassium channels: a model of heteromultimeric potassium channel/receptor assemblies. Annu Rev Physiol 61:337–362. Seino S, Miki T (2004) Gene targeting approach to clarification of ion channel function: studies of Kir6.x null mice. J Physiol 554:295– 300. Shimizu K, Lacza Z, Rajapakse N, Horiguchi T, Snipes J, Busija DW (2002) MitoK(ATP) opener, diazoxide, reduces neuronal damage after middle cerebral artery occlusion in the rat. Am J Physiol Heart Circ Physiol 283:H1005–H1011. Sun HS, Feng ZP, Miki T, Seino S, French RJ (2006) Enhanced neuronal damage after ischemic insults in mice lacking Kir6.2containing ATP-sensitive K⫹ channels. J Neurophysiol 95:2590 – 2601. Suzuki M, Fujikura K, Inagaki N, Seino S, Takata K (1997) Localization of the ATP-sensitive K⫹ channel subunit Kir6.2 in mouse pancreas. Diabetes 46:1440 –1444. Yamada K, Ji JJ, Yuan H, Miki T, Sato S, Horimoto N, Shimizu T, Seino S, Inagaki N (2001) Protective role of ATP-sensitive potassium channels in hypoxia-induced generalized seizure. Science 292: 1543–1546. Yamamoto S, Tanaka E, Higashi H (1997) Mediation by intracellular calcium-dependent signals of hypoxic hyperpolarization in rat hippocampal CA1 neurons in vitro. J Neurophysiol 77:386 –392. Yin HZ, Sensi SL, Ogoshi F, Weiss JH (2002) Blockade of Ca2⫹permeable AMPA/kainate channels decreases oxygen-glucose deprivation-induced Zn2⫹ accumulation and neuronal loss in hippocampal pyramidal neurons. J Neurosci 22:1273–1279. Zawar C, Plant TD, Schirra C, Konnerth A, Neumcke B (1999) Celltype specific expression of ATP-sensitive potassium channels in the rat hippocampus. J Physiol 514 (Pt 2):327–341. Zhou M, Tanaka O, Suzuki M, Sekiguchi M, Takata K, Kawahara K, Abe H (2002) Localization of pore-forming subunit of the ATPsensitive K(⫹)-channel, Kir6.2, in rat brain neurons and glial cells. Brain Res Mol Brain Res 101:23–32.
(Accepted 19 October 2006) (Available online 18 December 2006)