Experimental Neurology 263 (2015) 161–171

Contents lists available at ScienceDirect

Experimental Neurology journal homepage: www.elsevier.com/locate/yexnr

Regular Article

Neuronal KATP channels mediate hypoxic preconditioning and reduce subsequent neonatal hypoxic–ischemic brain injury Hong-Shuo Sun a,b,c,d,⁎, Baofeng Xu a,b, Wenliang Chen a,b, Aijiao Xiao b, Ekaterina Turlova a,b, Ammar Alibraham a,b, Andrew Barszczyk b, Christine Y.J. Bae a,b, Yi Quan b, Baosong Liu a,b, Lin Pei a,b, Christopher L.F. Sun b,e, Marielle Deurloo b, Zhong-Ping Feng b,⁎⁎ a

Department of Surgery, Faculty of Medicine, University of Toronto, Toronto, Ontario M5S 1A8, Canada Department of Physiology, Faculty of Medicine, University of Toronto, Toronto, Ontario M5S 1A8, Canada c Department of Pharmacology & Toxicology, Faculty of Medicine, University of Toronto, Toronto, Ontario M5S 1A8, Canada d Institute of Medical Science, Faculty of Medicine, University of Toronto, Toronto, Ontario M5S 1A8, Canada e Faculty of Applied Science & Engineering, University of Toronto, Toronto, Ontario M5S 1A4, Canada b

a r t i c l e

i n f o

Article history: Received 12 June 2014 Revised 23 August 2014 Accepted 10 October 2014 Available online 18 October 2014 Keywords: Hypoxic preconditioning Neonatal hypoxic–ischemic brain injury KATP channel Neuroprotection PKC

a b s t r a c t Neonatal hypoxic–ischemic brain injury and its related illness hypoxic–ischemic encephalopathy (HIE) are major causes of nervous system damage and neurological morbidity in children. Hypoxic preconditioning (HPC) is known to be neuroprotective in cerebral ischemic brain injury. KATP channels are involved in ischemic preconditioning in the heart; however the involvement of neuronal KATP channels in HPC in the brain has not been fully investigated. In this study, we investigated the role of HPC in hypoxia–ischemia (HI)-induced brain injury in postnatal seven-day-old (P7) CD1 mouse pups. Specifically, TTC (2,3,5-triphenyltetrazolium chloride) staining was used to assess the infarct volume, TUNEL (Terminal deoxynucleotidyl transferase mediated dUTP nick end-labeling) to detect apoptotic cells, Western blots to evaluate protein level, and patch-clamp recordings to measure KATP channel current activities. Behavioral tests were performed to assess the functional recovery after hypoxic–ischemic insults. We found that hypoxic preconditioning reduced infarct volume, decreased the number of TUNEL-positive cells, and improved neurobehavioral functional recovery in neonatal mice following hypoxic–ischemic insults. Pre-treatment with a KATP channel blocker, tolbutamide, inhibited hypoxic preconditioning-induced neuroprotection and augmented neurodegeneration following hypoxic–ischemic injury. Pre-treatment with a KATP channel opener, diazoxide, reduced infarct volume and mimicked hypoxic preconditioning-induced neuroprotection. Hypoxic preconditioning induced upregulation of the protein level of the Kir6.2 isoform and enhanced current activities of KATP channels. Hypoxic preconditioning restored the HI-reduced PKC and pAkt levels, and reduced caspase-3 level, while tolbutamide inhibited the effects of hypoxic preconditioning. We conclude that KATP channels are involved in hypoxic preconditioning-induced neuroprotection in neonatal hypoxic–ischemic brain injury. KATP channel openers may therefore have therapeutic effects in neonatal hypoxic–ischemic brain injury. © 2014 Elsevier Inc. All rights reserved.

Introduction Perinatal and neonatal hypoxic–ischemic brain injury (Nelson, 2007; Vannucci and Hagberg, 2004), also referred to as neonatal stroke, can lead to hypoxic–ischemic encephalopathy or cerebral palsy. Both hypoxic–ischemic encephalopathy and cerebral palsy are early onset Abbreviations: KATP channel, ATP-sensitive potassium channel. ⁎ Correspondence to: H.-S. Sun, Department of Surgery, Faculty of Medicine, University of Toronto, 1132 Medical Sciences Building, 1 King's College Circle, Toronto, Ontario M5S 1A8, Canada. Fax: +1 416 978 4373. ⁎⁎ Correspondence to: Z.-P. Feng, Department of Physiology, Faculty of Medicine, University of Toronto, 3306 Medical Sciences Building, 1 King's College Circle, Toronto, Ontario M5S 1A8, Canada. Fax: +1 416 978 4373. E-mail addresses: [email protected] (H.-S. Sun), [email protected] (Z.-P. Feng).

http://dx.doi.org/10.1016/j.expneurol.2014.10.003 0014-4886/© 2014 Elsevier Inc. All rights reserved.

brain and behavioral disorders in children. Hypoxic–ischemic encephalopathy (Fatemi et al, 2009) is a group of impairments including neurodevelopmental delay, cognitive and motor deficits, and epilepsy. Cerebral palsy (Nelson and Grether, 1999) is often associated with abnormal brain function including learning, memory, thinking, hearing, seeing and movement. Hypoxic–ischemic brain injury and its related illnesses hypoxic–ischemic encephalopathy and cerebral palsy are major causes of acute mortality and chronic neurological morbidity in infants and children (Nelson, 2007; Nelson and Grether, 1999; Nelson and Lynch, 2004; Vannucci, 2007; Vannucci and Hagberg, 2004). Currently, there is no effective treatment for neonatal hypoxic–ischemic brain injury (Wachtel and Hendricks-Munoz, 2011). While the causes of hypoxic–ischemic brain disorders have not yet been fully understood, it is generally accepted that the pathophysiological

162

H.-S. Sun et al. / Experimental Neurology 263 (2015) 161–171

events that occur during cerebral hypoxia and ischemia lead to ionic imbalance, which causes neuronal excitotoxicity and eventually neuronal cell death (Besancon et al, 2008; Dirnagl et al, 1999; Lipton, 1999; Tymianski, 2011). Ischemic preconditioning (IPC) of the heart (Murry et al, 1986; Terzic, 1999) and ischemic tolerance of the brain (Blondeau et al, 2000; Dirnagl et al, 1999) are natural adaptive processes induced by a range of sublethal insults (such as brief ischemia or transient hypoxia) preceding sustained ischemia and can decrease infarct size (Murry et al, 1986; Terzic, 1999). Preconditioning increases the resistance of tissue against subsequent and potentially lethal ischemic attack (Murry et al, 1986; Terzic, 1999). This adaptive cytoprotective phenomenon is a fundamental capability of living cells, allowing cells to survive exposure to recurrent stressors. Preconditioning induces two different time windows of tolerance: (1) a rapid phase that lasts for an hour, and (2) a delayed phase, which is the robust state of tolerance normally detectable after 1 day, peaking at 3 days and fading after 7 days (Obrenovitch, 2008). Many studies have focused on understanding the delayed phase of ischemic tolerance that provides the time window for preconditioning-induced neuroprotection. Ischemic tolerance has been successfully induced in experimental models of cerebral ischemia in both adult and neonatal animals (Autheman et al, 2012; Prass et al, 2003; Sheldon et al, 2007), and is considered a promising approach for neuroprotection. Thus, it is important to understand the molecular mechanisms contributing to neonatal brain injury so that novel therapeutic methods can be established. The ATP-sensitive potassium (KATP) channel (Noma, 1983) is a member of the inward rectifier K+ (Kir) channel superfamily. It opens upon reduction of the ATP/ADP ratio in the cell to reduce membrane excitability. A functional KATP channel requires (SUR/Kir)4, a heterooctamer complex (Babenko et al, 1998; Miki et al, 1999; Seino, 1999), which consists of the heterotetramer of pore-forming subunits (Kir6.1/Kir6.2)4 and modulatory sulfonylurea receptor (SUR1/SUR2)4 subunits (Ashcroft et al, 1984; Cook and Hales, 1984) in a 1:1 ratio of stoichiometry. Both sarcolemmal and mitochondrial KATP channels have been described previously (Babenko et al, 1998; Bajgar et al, 2001; Garlid, 2000; Miki et al, 1999; Seino, 1999). The neuronal isoform of sarcolemmal KATP channels is Kir6.2/SUR1 (Ashford et al, 1988; Bajgar et al, 2001; Seino, 1999). Sarcolemmal KATP channels are weak inward rectifier K+-selective channels, and play many important roles in cellular functions by linking cellular metabolism to the regulation of cell membrane electrical activity under ischemic conditions (Babenko et al, 1998; Miki et al, 1999; Seino, 1999; Sun et al, 2006, 2007; Yamada et al, 2001). KATP channels are involved in ischemic preconditioning in the heart (Horimoto et al, 2000; Murry et al, 1986; Terzic, 1999). KATP channels have also long been proposed to play a role in ischemic tolerance in stroke, but their role in ischemic tolerance in neonatal hypoxic– ischemic brain injury has not been fully studied. KATP channels are expressed in a large number of central neurons and are presumably closed under physiological conditions. During hypoxia and ischemia, energy failure reduces the ATP to ADP ratio, which activates the neuronal KATP channels in the brain and causes hyperpolarization of the cell membrane, which in turn suppresses neuronal activity and excitability (Fujimura et al, 1997; Sun et al, 2006, 2007; Yamada et al, 2001). We previously reported that this hypoxia–ischemia induced membrane hyperpolarization is a cellular mechanism underlying neuroprotection against stroke (Sun et al, 2006, 2007). We thus hypothesize that activation of KATP channels induces ischemic tolerance in the brain during hypoxic–ischemic insult, and serves as a potential therapeutic target for neuroprotection against neonatal hypoxic–ischemic brain injury. In this study, we showed that the infarct volume after hypoxic– ischemic brain injury in neonatal mice subjected to prior hypoxic preconditioning (HPC) was significantly reduced. HPC also increased the functional KATP channels in the brain and suppressed caspase-3 dependent apoptosis. KATP channel blocker prevented the neuroprotective

effect of hypoxic preconditioning; KATP channel opener mimicked hypoxic preconditioning. Thus, neuronal KATP channels may contribute to the underlying cellular and molecular mechanisms of ischemic tolerance against neonatal hypoxic–ischemic brain injury and neonatal stroke. Materials and methods Animals Timed pregnant CD-1 mice (Charles River Laboratory, Quebec, Canada) and seven- to nine-day old pups (P7 or P9) were used. All procedures conformed to guidelines established by the Canadian Council on Animal Care and were approved by the University of Toronto animal care committee. Materials Drugs used in the study were from Sigma-Aldrich Canada: tolbutamide, diazoxide, chelerythrine chloride, 2,3,5-triphenyltetrazolium chloride (TTC), and cresyl violet. Tissue sample preparation To determine Kir6.2 protein level at 1, 2, and 3 days after completion of hypoxia treatment, P7–9 pups (n = 5–7) were sacrificed under deep anesthesia with isoflurane and the brain was extracted. The brains were then flash frozen in dry ice and stored at −80 °C until protein extraction. Tissues from untreated animals (n = 4) were also collected. The brain tissue was used for Western blot. Protein extraction and Western blots (WB) Frozen mouse brains were homogenized in RIPA buffer (20 mM TrisCl, pH 7.5, 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1 mM PMSF plus 1% Igepal CA-630, 1% sodium deoxycholate) with a protease inhibitor cocktail (5 μl/100 mg tissue; Sigma-Aldrich), and incubated at 4 °C for at least 1 h, then centrifuged at 4 °C at 13,000 rpm for 15 min. The supernatant was extracted and protein concentrations were measured (Bio-Rad, Hercules, CA). Protein samples (mouse brain 30 μg) were boiled for 5 min in SDS sample buffer, and subjected to SDS-PAGE. Protein was transferred to polyvinylidene difluoride membranes (400 mA, 1.5 h) using a liquid transfer system (transfer buffer: 50 mM boric acid, 2.5 mM EDTA, pH 8.9). Blots were blocked with 5% non-fat dried milk dissolved in TBST buffer (10 mM Tris, 150 mM NaCl, and 0.1% Tween 20) for 1 h at room temperature, washed three times with TBST buffer, incubated with the appropriate primary antibody: rabbit antibody for Kir6.2 (1:200; Sigma-Aldrich), rabbit antibody for PKC (1:1000, Santa Cruz), and rabbit antibodies for Bax (1:1000), Bcl-2 (1:1000, D17C4), antiphospho-Akt (1:1000, Ser473) and anticleaved caspase-3 (1:500, Cell Signaling Technology, Danvers, Mass), and mouse antibody for βactin (1:3000; Sigma-Aldrich) diluted in 1% milk in TBST overnight at 4 °C plus 1 h at room temperature. The blots were washed again with TBST buffer three times and then incubated with appropriate horseradish peroxidase-conjugated secondary antibody accordingly (1:7500; Jackson ImmunoResearch Laboratories) for 1.5 h at room temperature. Antibody-labeled protein bands were visualized using enhanced chemiluminescent reagents (PerkinElmer) and analyzed by exposure to film (HyBlot CL). Hypoxic preconditioning (HPC) Hypoxic preconditioning (HPC) was induced by placing P7 mouse pups in a hypoxic chamber (A-Chamber A-15274 with ProOx 110 Oxygen Controller/E-720 Sensor, Biospherix, NY, USA) containing 7.5% oxygen balanced with 92.5% nitrogen for 60 min. One pup was monitored

H.-S. Sun et al. / Experimental Neurology 263 (2015) 161–171

to ensure that its rectal temperature did not exceed 36.5 °C using a homeothermic blanket control unit (K-017484 Harvard Apparatus, Massachusetts, USA). Pups not subjected to hypoxic preconditioning were kept in a chamber exposed to room air. The duration of exposure and oxygen concentration used for HPC were consistent with those used for inducing hypoxic–ischemic (HI) brain injury (Sheldon et al, 2007). Drug administration Twenty minutes prior to the onset of ischemia, either tolbutamide (100 mg/kg in dimethyl sulfoxide (DMSO)), diazoxide (20 mg/kg in distilled water, pH = 11.0), Chelerythrine Chloride (5 mg/kg in PBS), or corresponding vehicle controls (0.5% DMSO for tolbutamide, distilled water adjusted to pH 11.0 by NaOH for diazoxide and PBS for Chelerythrine Chloride) were administered to pups intraperitoneally (i.p.). The total volume injected per animal was 0.1 mL. Hypoxic–ischemic (HI) brain injury model Two days after hypoxic preconditioning (HPC), neonatal hypoxic– ischemic (HI) brain injury using the Vannucci method (Levine, 1960; Rice et al., 1981) was induced in P9 mouse pups as described previously (Alibrahim et al, 2013). Mice of either sex were anesthetized with isoflurane (3% for induction and 1.5% for maintenance) in a balance of oxygen during the procedure. A stereo dissecting microscope (SMZ-2B Nikon, Japan) with fiber-optic ring light illumination was used for assisting the dissection. The skin on the neck was cleaned with iodine followed by 75% alcohol. After a midline cervical incision 0.5 cm in length, the common carotid artery was carefully exposed on one side, detached from the surrounding vagus and sympathetic nerves, and permanently occluded by bipolar electrical coagulation (Vetroson V-10 Bi-polar electrosurgical unit). For model development and subsequent experiments, local cerebral blood flow was measured using the PeriMed PeriScan System PIM II Laser Doppler Blood Perfusion Imager, or randomly checked using the PeriMed PeriFlux System 5000/5010 Laser Doppler Perfusion Monitor (PeriMed, Stockholm, Sweden) to ensure sufficient local cerebral blood flow reduction in the ipsilateral hemisphere. Body temperature was monitored and maintained by using a homeothermic heating blanket. The whole procedure took approximately 5 min to complete for each pup. After the surgical procedure, the neck incision was closed using tissue glue (Tissue Adhesive, 3M Vetbond, No. 1469SB, St. Paul, MN, USA). The pups were placed in an incubator at 37 °C for about 10 min until fully awake, and then returned to the dam for 90 min to recover and feed. For the hypoxic challenge, the pups were then transferred into an airtight, transparent door hypoxia chamber (A-Chamber with ProOx 110 Oxygen Controller/E-720 Sensor, Biospherix, NY USA) and injected with 7.5% oxygen balanced with 92.5% nitrogen for approximately 60 min. Oxygen concentration and flow rate were regulated by a compact oxygen controller (ProOx 110 controller, Biospherix, NY USA) connected to a compressed nitrogen gas source (Linde, Mississauga, ON Canada). Throughout hypoxia treatment, the pups were placed on a heating pad, and one pup was monitored to ensure that body temperature did not exceed 37 °C using a homeothermic blanket control unit (K-017484 Harvard Apparatus, Massachusetts, USA). After hypoxia exposure, all mouse pups were allowed to recover in room air on a heating pad (37 °C) for 30 min inside the chamber with the door open, and later returned to their mother in the dam. For sham animals, an incision was made at the midline in the neck to expose the right common carotid artery and the incision was then closed using tissue glue. Sham animals were not subjected to hypoxia. Measurement of infarct volume and histology Twenty-four hours after hypoxic–ischemic injury, pups were sacrificed under deep anesthesia with isoflurane, the brains were

163

removed and sectioned coronally into approximately 1 mm slices. These sections were then immersed in 2% 2,3,5-triphenyltetrazolium chloride (TTC) in PBS at 37 °C in a dark place for 20–30 min (Alibrahim et al, 2013; Sun et al, 2006, 2007, 2008). After the TTC staining, brain slices were scanned using a photo scanner and the infarct area was measured. The areas of the ipsilateral and contralateral hemispheres for each coronal section were measured using an imageanalysis system (NIH ImageJ software). Infarct volume was calculated by summing the representative areas in all brain sections and multiplying by the slice thickness. After correcting for edema, the volumes of infarction were calculated as follows: Corrected infarct volume (CIV), (%) = [contralateral hemisphere volume − (ipsilateral hemisphere volume − infarct volume)] /contralateral hemisphere volume × 100 (Qi et al, 2004). Seven days after hypoxic–ischemic brain injury, whole brains in the 3 groups (HI only, HPC + HI, and HPC + HI + tolbutamide) were digitally imaged to observe whether they exhibited any differences in gross anatomy. In addition, the 7 day brains were sectioned coronally, stained with Nissl stain (0.1% cresyl violet), and imaged for histological comparison across all 3 groups.

Behavioral tests Age-matched mouse pups subjected to hypoxic–ischemic injury and different drug treatments were tested in multiple neurobehavioral tests at 1, 3 and 7 days following the HI procedure. These tests evaluated: 1) geotaxis reflex (Sanches et al, 2012) to evaluate vestibular and/or proprioceptive functions; 2) cliff avoidance reaction (Bouslama et al, 2007) to check for the presence of maladaptive impulsive behavior; and 3) grip ability (Liu et al, 2013) to assess grip force and fatigability. After completion of the tests, mice were deeply anesthetized with isoflurane and their brains were removed and fixed with 4% paraformaldehyde in PBS for immunohistochemical evaluation.

Immunohistochemistry and confocal fluorescence imaging Brains were post-fixed in 30% sucrose/4% paraformaldehyde in PBS solution at 4 °C overnight (Sun et al, 2009) and subsequently sectioned into 50 μm-thick coronal slices using a Vibratome (Vibratome Series 1000, Warner Instruments, USA). Immunohistochemical staining was performed as described previously (Sun et al, 2006, 2009). In brief, coronal brain sections were blocked using 3% normal goat serum (vol/vol), 1% BSA (vol/vol) and 0.3% Triton X-100 (vol/vol) in PBS at room temperature for 60 min. Samples were labeled with mouse anti-NeuN antibody (MAB377, 1:100, Chemicon) overnight at 4 °C with rocking. NeuN is a broadly used neuron-specific cell marker (Mullen et al, 1992). Sections were subsequently washed in PBS and blocked briefly with the blocking solution. Sections were then incubated with affinity-purified goat antimouse secondary antibody Alexa 488 (1:200, Molecular Probes) for 1 h at room temperature. TUNEL staining (Sun et al, 2009) was performed according to Chemicon's protocol using the Apoptag kit (S7165, ApopTag® Red In Situ Apoptosis Detection Kit, Chemicon). For double staining (Sun et al, 2009), slices were labeled with NeuN, and either Kir6.2 (Sigma) or PKC (Sigma). Finally, brain sections were mounted on glass coverslips and sealed with ProLong Gold antifade reagent (Invitrogen/Molecular Probes). Immunolabeled brain slices were imaged with a confocal laser scanning microscope (Zeiss LSM700, Zeiss, Germany) and images were analyzed with a three-dimensional (3D) constructor (ImageJ software; http://imagej.nih.gov/ij/). In the 50 μm slice preparation, only the middle sections of the brain slice were used for imaging to avoid any mechanical damage and consequent distortion of the images. 3D digital reconstructions were produced from a series of confocal images taken at 1 μm intervals through the region of interest, and optical stacks of 10 images were produced for the final figures, as described previously (Sun et al, 2006, 2009).

164

H.-S. Sun et al. / Experimental Neurology 263 (2015) 161–171

Data analysis Data are presented as mean ± S.E.M. Statistical analysis was performed using SigmaStat (3.0, Jandel Scientific). Differences between experimental groups were evaluated using Student's t-test for two groups, and one-way analysis of variance (ANOVA) followed by Fisher-LSD post hoc test for multiple experimental groups. Statistical significance of the differences was defined by probability level of lower than 0.05 (p b 0.05). Results Hypoxic preconditioning reduced subsequent hypoxic–ischemic brain injury Neonatal mouse pups were subjected to hypoxic preconditioning (HPC, 7.5% O2, 1 h) two days prior to hypoxic–ischemic (HI) brain injury. The study was conducted by two independent teams, each with a different set of HI conditions (7.5% O2 for 60 min and 100 min, the latter to produce a more severe brain injury). All experiments and analyses were performed in a blinded manner, including the induction of HI. Brain infarct was detected by TTC staining (Fig. 1). The infarct volume of the HI group treated with 7.5% O2 for 60 min was 36.87 ±1.91% (n = 5); the infarct volume of the animals that received HPC pre-treatment followed by HI was significantly reduced (HPC + HI group: 23.42 ± 1.15%; n = 6, p b 0.05) (Fig. 1A). To further confirm our findings, a different group of experimenters that were also blind to treatment groups conducted independent experiments to determine the effect of HPC on animals subjected to HI with more severe hypoxia (7.5% O2 for 100 min). The infarct volume after more severe hypoxia (7.5% O2 for 100 min) was 56.13 ± 5.67% (n = 9) in animals without HPC and 38.72 ± 2.93% (n = 7, p b 0.05) in animals subjected to HPC before HI (Fig. 1B). These results indicate that HPC significantly reduces HI brain injury in neonates. Sham control and HPC sham control groups did not show any damage as in TTC stain in matching age (P10) (Supplementary Fig. 1). KATP channels contribute to the protective effect of hypoxic preconditioning To investigate the involvement of KATP channels in the neuroprotective effect of HPC, we applied the KATP channel blocker tolbutamide after HPC but before HI. Blood glucose level before the tolbutamide administration and 30 min after the administration were 8.4 ± 0.89 mmol/L and 7.44 ± 1.06 mmol/L, respectively (using Accu-Chek glucose meter; not significant, n = 5). The involvement of KATP channel in HPC would be suggested if the KATP channel blocker prevented the neuroprotective effect of the HPC. We thus first compared the infarct areas of the animals subjected to HPC + HI procedure with or without tolbutamide application. As anticipated, infarct volumes were larger when tolbutamide (100 mg/kg, i.p., 20 min prior to HPC + HI) was administered, as indicated by TTC staining (Fig. 2A). Specifically, HPC + HI (60 min) animals had an infarct area of 21.41 ± 3.45% (n = 5), which significantly increased to 45.32 ± 1.67% (n = 6) in the HPC + HI + tolbutamide group (Fig. 2A1). Similarly, in the 100 min HI group, HI-treated animals had an infarct volume of 32.88 ± 3.01% (n = 11), which increased to 56.14 ± 8.29% (n = 7) with tolbutamide application (Fig. 2A2). These findings suggest that neuronal KATP channels contribute to the neuroprotective effects of hypoxic preconditioning and ischemic tolerance against subsequent neonatal hypoxic–ischemic brain injury, as application of tolbutamide blocked the neuroprotective effect of hypoxic preconditioning in HI. We next asked whether a KATP channel agonist could mimic HPC. To address this question, the KATP channel opener diazoxide (20 mg/kg, i.p.) was administered to animals 20 min prior to HI. Blood glucose level before the diazoxide administration and 30 min after the administration were 7.88 ± 0.39 mmol/L and 8.97 ± 0.66 mmol/L, respectively (using Accu-Chek glucose meter; not significant, n = 4). HI-induced infarct

Fig. 1. Hypoxic preconditioning (HPC) reduced subsequent hypoxic–ischemic (HI) brain injury. A. Brain infarctions detected 24 h after the HI injury (7.5% O2 for 60 min) with or without HPC (7.5% O2, 1 h, 2 days prior to HI). Left: Representative brain slices stained with TTC (damage shown in white areas); Right: Summary of infarct areas. The HPC + HI group has smaller infarct volume compared with that of HI only group. *p b 0.05 (t test); the numbers of animals tested are indicated in the bar graphs. B. Brain infarctions detected 24 h after the HI injury (7.5% O2 for 100 min) with or without HPC (7.5% O2, 1 h, 2 days prior to HI). Left: Representative brain slices stained with TTC; right: summary of the brain infarct areas. The HPC + HI group has smaller infarct volume compared with that of HI only group. *p b 0.05 (t test); the numbers of animals tested are indicated in the bar graphs.

volumes were then compared between diazoxide and vehicle control treated groups. TTC staining showed that the infarct volume in the HI + diazoxide treated group (HI 60 min, 26.10 ± 1.31%, n = 6) was significantly smaller than that of the HI only group (60 min, 35.06 ± 1.77%, n = 4) (Fig. 2B). These findings indicate that the opening of neuronal KATP channels by the KATP channel opener diazoxide alone may induce a neuroprotective effect similar to that of HPC. Taken together, our data showing that a KATP channel antagonist reduced the effect of HPC, and a KATP channel agonist mimicked the effect of HPC suggest the involvement of KATP channels in HPC and neuroprotection in HI brain injury in neonatal animals. We next investigated whether HPC reduces apoptotic cell death following HI insult and whether a KATP channel blocker will mitigate this effect. As shown in Fig. 2C, TUNEL stain demonstrated that brains in the HPC + HI group had fewer TUNEL positive cells (21.50 ± 2.43 per × 10 image field, n = 6) compared to brains in the HI only group (110.17 ± 11.48 per × 10 field, n = 6). These findings support the idea that HPC reduces apoptotic cell death caused by HI. Tolbutamide application after HPC but before HI (HPC + HI + tolbutamide group) resulted in more TUNEL-positive cells (220 ± 9.88, n = 4) (Fig. 2C), suggesting that tolbutamide blocked the neuroprotective effect of

H.-S. Sun et al. / Experimental Neurology 263 (2015) 161–171

165

Fig. 2. Neuroprotective effects of hypoxic preconditioning (HPC) were inhibited by KATP channel blocker. A. KATP channel blocker increased brain infarct volume. Brain infarct volume was detected 24 h after (A1) 60 min or (A2) 100 min HI injury (7.5% O2) with HPC (7.5% O2, 1 h, 2 days prior to HI) followed by either vehicle or tolbutamide (100 mg/kg) treatment before HI. Left: Representative brain slices stained with TTC; right: summary of infarct areas. Application of tolbutamide following HPC increased the infarct volume compared to that of vehicle control group. *p b 0.05 (t test); the numbers of animals tested are indicated in the bar graphs. B. Pre-treatment of KATP channel agonist reduced brain infarct volume. Brain infarct areas detected 24 h after HI only or diazoxide treated groups. Left: Representative brain slices stained with TTC; right: summary of the infarct areas. Application of diazoxide reduced the infarct volume compared to that of vehicle control group. *p b 0.05 (t test); the numbers of animals tested are indicated in the bar graphs. C. Hypoxic preconditioning (HPC) reduced TUNEL positive cell counts 3 days after the hypoxic–ischemic (HI) brain injury. Left: Representative confocal images showing TUNEL positive cells observed from HI-only, HPC + HI, and HPC + HI + Tol (tolbutamide) groups. Scale bars = 200 μm. Right: Summary showing TUNEL positive cell counts in three different groups shows that HPC + HI group has significantly less TUNEL positive cells. *p b 0.05 to “HI only” group: p b 0.05; one-way ANOVA + Fisher LSD test; n = 6 HI; n = 6 HPC + HI; n = 4 HPC + HI + Tol (tolbutamide). (Note: sham and HPC sham had no countable TUNEL positive cells, thus not shown.) D. Long-term effect of the hypoxic preconditioning in reducing brain damage. Left: Whole brains in 7 days after HI under different treatments: hypoxic preconditioning reduced brain damage (showing complete brain) in comparison to HI only and HPC + HI + Tol (tolbutamide) groups (showing missing brain tissues). (Note: sham and HPC sham brains had no detectable cross damage, thus not shown.) Right: cresyl violet staining for HI, HPC + HI, and HPC + HI + Tol (tolbutamide). Summary of infarct areas: The HPC + HI group has smaller infarct volume compared with that of HI only and HPC + HI + Tol (tolbutamide) groups (*p b 0.05; one-way ANOVA + Fisher LSD test; n = 4/group). The hypoxic preconditioning reduced brain damage (showing complete brain slice) in comparison to HI only and HPC + HI + Tol (tolbutamide) groups (showing missing brain parts). *p b 0.05; the numbers of animals tested are indicated in the bar graphs.

HPC. This further supports the notion that KATP channels are involved in HPC. The blocking effect of tolbutamide on HPC was also evident from morphological and gross anatomy assessments. Seven days post-HI, histology revealed a dramatic difference in whole-brain sizes between HI only, HPC + HI and HPC + HI + tolbutamide groups (Fig. 2D). Brains from the HPC + HI group exhibited near-normal gross anatomy of the brains 7 days after HI, while brains in the HI-only and HPC + HI + tolbutamide groups exhibited deformed/dented ipsilateral brain hemispheres 7 days after HI (Fig. 2D). In addition, brains in the HPC + HI group showed close to normal morphology in a low-power coronal section of the brain subjected to HI 7 days prior to Nissl stain (cresyl violet). The HI-only and HPC + HI + tolbutamide brain slices showed large missing pieces of cortical tissues in the area corresponding to the perfusion area of the middle cerebral artery 7 days after HI (Fig. 2D-right panel). These results show that HPC prior to HI insult significantly

reduces subsequent brain damage and apoptotic cell death, as well as preserve normal anatomical and morphological brain features. Hypoxic preconditioning improves functional recovery in hypoxic–ischemic brain injury via KATP channel-dependent mechanisms We next examined whether HPC improves neurological functional recovery by accessing the battery of geotaxis reflex, cliff avoidance and grip tests. Functional recovery in mice subjected to HI only, HPC + HI and HPC + HI + tolbutamide was compared on 1, 3, and 7 days with sham and HPC sham groups after HI. HPC + HI showed significant improvement in geotaxis scores on day 1 and trend of improvement in day 3 and day 7 post-HI in comparison to HI-only and HPC + HI + tolbutamide (Fig. 3A). Animals in the HPC + HI group also showed significant improvement in the cliff avoidance test on 1, 3 and 7 days post-HI in comparison to the other two groups of HI only

166

H.-S. Sun et al. / Experimental Neurology 263 (2015) 161–171

Hypoxic preconditioning up-regulates KATP channels in mouse neonatal brains To understand how HPC regulates KATP channels, we first measured the protein level of the Kir6.2 subunit of KATP channels. Our WB analysis showed that the protein level of Kir6.2 subunits increased significantly 2 days after HPC, respectively (Fig. 4A). Consistent with these findings, immunohistochemistry in conjunction with laser confocal microscopy showed that Kir6.2 expression in the hippocampal CA1 area increased two days after HPC (Fig. 4B). To determine whether the up-regulation of KATP channel proteins resulted in an increase in channel current, patch-clamp recording was carried out in acutely-dissociated hippocampal neurons. The current recorded in hippocampal neurons increased under high [ADP] and low [ATP] conditions 2 days after HPC (Fig. 4C). The increased current was inhibited by the KATP channel blocker tolbutamide (200 μM), suggesting that the increase in current activity was mediated through KATP channels. The representative I-V curves demonstrating the blocking effect of tolbutamide on the outward current from the control and HPC groups are shown in Figs. 4C1 and 4C2, respectively. The tolbutamide-blocked current component increased in HPC-treated groups. Furthermore, Kir6.2 protein expressions decreased in 1 day (P10; HPC 2 d + HI 1 d) after HI; HPC + HI restored the Kir6.2 level and HPC + HI + Tol blocked the HPC induced increased in Kir6.2 after HI in comparison to matching-aged sham control and HPC sham (Fig. 4D). Our data collectively demonstrate an HPC induced upregulation of neuronal KATP channels in the brain. Potential signaling pathways involved in the hypoxic preconditioning

Fig. 3. Hypoxic preconditioning (HPC) improved behavioral test scores in 1, 3 and 7 days after the hypoxic–ischemic (HI) injury. A. Geotaxis reflex test. HPC (7.5% O2, 1 h) 2 days prior to HI (7.5% O2, 1 h) significantly improved behavioral geotaxis test scores 1 day after HI in comparison with the HI only and HPC + HI + Tol (tolbutamide) groups. There was a trend of improvement for HPC at 3 and 7 days post HI in the geotaxic test. B. Cliff avoidance test. HPC (7.5% O2, 1 h) 2 days prior HI (7.5% O2, 1 h) significantly improved behavioral cliff avoidance test scores in 1, 3 and 7 days after HI in comparison with the HI only and HPC + HI + Tol (tolbutamide) groups. C. Grip test. HPC (7.5% O2, 1 h) 2 days prior HI (7.5% O2, 1 h) showed trend of improvement in behavioral grip test scores in 1, 3 and 7 days after HI in comparison with the HI only and HPC + HI + Tol (tolbutamide) groups. *p b 0.05; one-way ANOVA followed by Fisher LSD multiple comparison test was used in all the behavioral tests; numbers of pups used in the tests are indicated on the bars. Sham and HPC sham groups were used as baseline measurement in these behavioral tests.

and HPC + HI + tolbutamide (Fig. 3B). There was a trend of improvement for HPC at 1, 3 and 7 days post HI in the grip tests, but the difference was not statistically significant (Fig. 3C). The KATP channel blocker tolbutamide blocked the improvements induced by HPC in all three behavioral tests: geotaxis reflex, cliff avoidance test and grip test. These findings further support the involvement of KATP channels in the neuroprotection induced by HPC.

We next investigated the potential pro-survival signaling pathways that might be involved in HPC. PKC is known to play a role in HPCinduced cytoprotection. We asked whether PKC is also involved in HPC-induced neuroprotection in neonatal HI brain injury. Our WB analyses showed that the PKC level was increased in the hippocampus 6 h after HPC (Fig. 5A). In addition, HPC restored PKC protein level (WB) by 24 h after hypoxic–ischemic injury (HPC + HI) in comparison to HI only and HPC + HI + Tol (tolbutamide groups) (Fig. 5A). Consistent with the WB data, our IHC analysis of mouse brain slices showed that the PKC level in the NeuN-positive cells in the CA1 region increased following HPC treatment with respect to the control group (Fig. 5B). These findings suggest a potential involvement of PKC in HPC, in addition to KATP channels. To confirm this finding, the PKC inhibitor chelerythrine chloride (5 mg/kg, i.p., 20 min prior to HI) was applied to P9 mouse pups following HPC. Fig. 5C shows that the infarct volume with HPC followed by HI (7.5% O2 for 60 min) was 24.99 ± 3.25% (n = 5) in the HPC + HI group and 34.03 ± 2.15% (n = 6) in the HPC + HI + chelerythrine group (33.90 ± 1.89%, n = 4 in HI-only group) (Fig. 5C), suggesting that chelerythrine blocked the neuroprotective effect of HPC. These results indicate that neuronal KATP channels and the downstream signaling molecule PKC contribute to the neuroprotective effects of the hypoxic preconditioning and ischemic tolerance against subsequent neonatal hypoxic–ischemic brain injury. We next studied whether any pro-apoptotic signaling pathways might be involved in HPC. Cleaved caspase-3 protein expression (normalized to β-actin) was significantly increased 24 h after HI (2.08 ± 0.44, n = 4) as compared to that in the sham group (0.27 ± 0.04; p b 0.05). Hypoxic preconditioning significantly reduced the cleaved caspase-3 protein level in compared to the HI group (p b 0.05, n = 4) (Fig. 6A). The KATP channel blocker tolbutamide prevented the effect of HPC on cleaved caspase-3 levels (1.82 ± 0.38, p b 0.05), similar to that seen in the HI group (Fig. 6A). These results suggest that HPC reduces hypoxic–ischemic cell death by regulating a known pro-apoptotic signaling pathway in the neonatal mice. We then studied the involvement of the Akt dependent pro-survival signaling pathway. Specifically, phosphorylated and total Akt (p-Akt/tAkt) ratios were compared and a decreased p-Akt/t-Akt was found in

H.-S. Sun et al. / Experimental Neurology 263 (2015) 161–171

167

Fig. 4. Hypoxic preconditioning increased Kir6.2 protein expression and channel activity in the neonatal mouse brains. A. Left: Representative Western blot analysis of the expression levels of Kir6.2 protein in the neonatal mouse brains at 1 and 2 days after hypoxic preconditioning. Right: Kir6.2 protein expression was significantly upregulated in 2 days after hypoxic preconditioning in comparison to the matching-age control (*p b 0.05 to 1-day control and HPC, and also to 2 day control, one-way ANOVA + Fisher LSD test, n = 5/group). B. Kir6.2 proteins upregulated in 2 days after hypoxic preconditioning (HPC) as shown in immunohistochemistry (IHC) of CA1 neurons in conjunction with laser confocal imaging. (p b 0.05; t test, n = 5/group). Ctrl = control; HPC = hypoxic preconditioning. C. ATP-sensitive potassium current increased after hypoxic preconditioning (HPC) which was blocked by tolbutamide (200 μM) in acutely isolated mouse hippocampal neurons. Representative patch-clamp recording of whole-cell currents evoked by 400 ms-long voltage ramps from −100 to +100 mV in neurons of postnatal day 9 under control (C1) or following HPC treatment 2 days prior (C2). The trace labeled “1” is the baseline current in normal bath solution, trace “2” shows the current after perfusion of 200 μM tolbutamide into bath, trace “3” indicates the current after washing out tolbutamide with bath solution, and trace “1-2” is created by arithmetical subtraction of trace “2” from “1”. C3. Summary of currents of control and tolbutamide perfusion in P9 and P9 + HPC treatment groups. There is a significant difference for control currents among the two groups, whereas HPC significantly increased the KATP currents which were then blocked by tolbutamide. *p b 0.01; one-way ANOVA + Fisher LSD test, n of each group is shown in the bottom of column. Tolb: tolbutamide. D. Kir6.2 protein expressions decreased in 1 day (P10; HPC 2d + HI 1d) after HI; HPC + HI restored the Kir6.2 level and HPC + HI + Tol blocked the HPC induced increased in Kir6.2 after HI in comparison to matching-aged sham control and HPC sham (P10). *p b 0.01; one-way ANOVA + Fisher LSD test, n of each group is shown in the bottom of column. Tolb: tolbutamide.

the HI group (0.47 ± 0.02 arbitrary units in the HI group versus 0.64 ± 0.01 arbitrary units in the sham group, p b 0.05, n = 4), and HPC significantly restored the p-Akt/t-Akt ratio (0.56 ± 0.03 arbitrary unit versus HI group, p b 0.05, n = 4, Fig. 6B) to the level similar to sham group. The

KATP channel blocker tolbutamide (0.42 ± 0.03 arbitrary units, p b 0.05, n = 4, Fig. 6B) administered to the animals before HPC + HI reduced the p-Akt/Akt level to that seen in the HI-only group, suggesting that the Akt pathway may also be involved in KATP-mediated hypoxic

168

H.-S. Sun et al. / Experimental Neurology 263 (2015) 161–171

Fig. 5. Involvement of PKC in hypoxic preconditioning. A. Western blot (WB) analysis showing PKC level. Left: PKC level in brain from P7 animals was upregulated 6 h after hypoxic preconditioning (HPC) in comparison to control (Ctrl). Right: comparison of the PKC level in brain from P10 animals reduced in 24 h after hypoxic–ischemic injury (HI), hypoxic preconditioning followed by HI (HPC + HI), and tolbutamide application (i.p.) prior to HPC (HPC + HI + Tol). B. Immunohistochemistry (IHC) in conjunction with laser confocal imaging, showing PKC antibody labeled CA1 neurons have higher fluorescent intensity level than the control neurons, two days after hypoxic preconditioning (HPC). Left: representative confocal images; right: summary of PKC expression levels. (*p b 0.05; n = 5/group). C. TTC staining showing brain infarctions in HI only, HPC + HI, and HPC + HI + chelerythrine chloride treated groups. Effect of HPC treated two days prior to HI was offset by PKC inhibitor chelerythrine chloride as the brain shows similar damages with HI group. Chelerythrine chloride was administrated i.p. after HPC but before the HI injury. The brains were sliced 24 h after HI and stained with TTC. *p b 0.05 to “HI only” and “HPC + HI + chelerythrine chloride” groups; one-way ANOVA + Fisher LSD test; n = 4 HI; n = 5 HPC + HI; n = 6 HPC + HI + chelerythrine chloride.

preconditioning. These results indicate that HPC reduces hypoxic– ischemic cell death by partially mediating a known pro-survival signaling pathway in the neonatal mouse brain. Discussion In this study, we reported that hypoxic preconditioning plays an important neuroprotective role in neonatal hypoxic–ischemic brain injury in mice, and that neuronal KATP channels contribute to HPC-induced neuroprotection. Specifically, we showed that: 1) HPC reduces infarct volume from HI; 2) neuroprotection induced by HPC is mediated, at least partially, by neuronal KATP channels, as the KATP channel blocker tolbutamide prevented the protective effect induced by HPC, and the KATP channel opener diazoxide caused neuroprotection; HPC reduced TUNEL-positive cells and HI-induced brain damage, and a KATP channel blocker prevents HPC effects; 3) HPC improves functional and

behavioral recovery following subsequent HI insults, and a KATP channel blocker prevents HPC effects; 4) HPC induces up-regulation of the protein level and current of KATP channels; and 5) the downstream signaling pathway for HPC in neonatal HI brain injury may include PKC and Akt. Ischemic preconditioning was first described in the heart (Murry et al, 1986; Terzic, 1999), where brief periods of ischemia preceding sustained ischemia could markedly decrease infarct size. Similar to the heart, a phenomenon of induced ischemic tolerance has also been demonstrated in the brain (Blondeau et al, 2000; Dirnagl et al, 1999). However, the underlying mechanism of preconditioning and/or ischemic tolerance is largely unknown. It is well established that KATP channels play a crucial role in cytoprotection as part of ischemic preconditioning of the heart (Horimoto et al, 2000; Murry et al, 1986; Terzic, 1999), though the involvement of KATP channels in ischemic preconditioning in the brain has not been investigated. In this study, we first reported

H.-S. Sun et al. / Experimental Neurology 263 (2015) 161–171

169

Fig. 6. Western blotting detected both pro-apoptotic and pro-survival related protein expressions 24 h after hypoxic–ischemic injury under different conditions in the mouse brains. A. Cleaved caspase-3 proteins decreased in hypoxic preconditioning (HPC) followed by hypoxic–ischemic (HI) injury group (HPC + HI) in comparison to HI only and HPC + HI + Tol (tolbutamide) groups (*p b 0.05, one-way ANOVA + Fisher LSD test; n = 4/group). B. PI3K/Akt signal pathway protein expression: hypoxic preconditioning (HPC) restored the p-Akt(s473) and t-Akt protein expressions 24 h after hypoxic–ischemic (HI) injury in comparison to the HI only and HPC + HI + Tol (tolbutamide) groups (*p b 0.05, one-way ANOVA + Fisher LSD test; n = 4/group).

that HPC-induced neuroprotection is mediated by Kir6.2 KATP channels in the neonatal hypoxic–ischemic brain injury model. KATP channel (Noma, 1983) is a member of the inward rectifier K+ (Kir) channel superfamily, and serves as a metabolism-sensing channel, because of its ability to link cellular metabolism to the regulation of electrical activity of cell membranes under physiological or ischemic conditions (Babenko et al, 1998; Miki et al, 1999; Seino, 1999). KATP channels are expressed in the cortical and hippocampal neurons of the brain (Sun et al, 2006, 2007). Opening of neuronal KATP channels during hypoxia and/or ischemia with the reduction of the ATP/ADP ratio in the cell induces membrane hyperpolarization, stabilization of cell membrane potential, and reduction of membrane excitability, and thus, making it a key therapeutic target for cytoprotection and ischemic preconditioning strategies against cerebral ischemia and stroke (Sun et al, 2006, 2007). In this study, we showed not only that KATP channel activity increased, but also that protein levels of Kir6.2 subunits increased in hippocampal neurons following HPC, supporting the notion that KATP channels contribute to neuroprotection during the preconditioning phase. There are two major subunit isoforms, SUR1 and Kir6.2, that constitute KATP channels in the brain. Knockout of SUR1 did not affect HPCinduced neuroprotection in adult mice (Munoz et al, 2003), suggesting that KATP channel-independent mechanisms are involved in HPC in cerebral ischemia in adult mice. In this study, we demonstrated that HPC is involved in a KATP channel-dependent pathway, since blocking KATP channels by tolbutamide prevented HPC induced neuroprotection in a neonatal HI brain injury mouse model. Kir6.2 KO (Yamada et al, 2001) and SUR1 KO mice are highly susceptible to hypoxia, and brief hypoxia rapidly induces lethal seizures in these animals (Munoz et al, 2003; Yamada et al, 2001). Thus, it is likely that the KO mice preclude the development of hypoxic preconditioning. In our study, we used pharmacological tools to test the involvement of Kir6.2 KATP channels in HPC, and demonstrated that blocking Kir6.2 KATP channels prevented HPC-induced neuroprotection. Ischemic preconditioning can be chemically mimicked by the KATP channel opener diazoxide, both in the heart (Horimoto et al, 2000) and in the brain (Munoz et al, 2003). While the site of action for diazoxide on KATP channels remains controversial, it is generally accepted that diazoxide is a KATP channel opener (Seino, 1999). Consistent with previous findings (Horimoto et al, 2000; Munoz et al, 2003), we report

that diazoxide mimics HPC-induced neuroprotection, suggesting the involvement of KATP channels sensitive to diazoxide in HPC. Thus, KATP channels may serve as a potential target for neuroprotection and improving functional recovery from HI brain injury. Further studies are required to demonstrate the usefulness of KATP channel openers in stroke treatment. Induced preconditioning includes both the rapid phase (lasts for an hour), and the robust delayed phase (from 1 to 7 days after induction) (Obrenovitch, 2008). The delayed phase of IPC has been well studied in the heart as it provides the time window for cellular and molecular regulation. In this study, HPC 2 days prior to HI reduced subsequent HI-induced infarct volume and TUNEL positive cell counts in the neonatal brains. The KATP channel expression and function were optimal 2 days after the HPC, consistent with the time course for the delayed phase of ischemic tolerance (Obrenovitch, 2008). There are indications that PKC is involved in the KATP channel signaling pathway in the heart (Light et al., 2001). In this study, we found that KATP channels and the PKC dependent signaling pathway are also involved in the delayed phase of ischemic tolerance. The PKC inhibitor (chelerythrine chloride) blocked the HPC's protective effect in the hypoxic–ischemic brain injury when treated 2 days after the preconditioning and before the HI. Further study is needed to elucidate the role of PKC for HPC in the brain. While the turnover rate of the functional KATP channels in central neurons remains unclear, induction, synthesis and insertion of the ion channel proteins into the cell membrane may require optimal time following HPC. HPC enhances the pro-survival and suppresses the pro-apoptotic signaling pathways (Brunet et al, 1999; Datta et al, 1997; Downward, 2004). In this study, HPC 2 days prior to HI restored the Akt pro-survival signaling and inhibited the caspase-3 proapoptotic signaling, further suggesting the time window for HPCinduced neuroprotection in neonatal stroke. Functional recovery from cerebral ischemic insults is a key goal of therapeutic strategies for stroke. In this study, we adapted a number of behavioral tests, including the geotaxis reflex (Sanches et al, 2012), cliff avoidance reaction (Bouslama et al, 2007) and grip test (Liu et al, 2013) to test whether HPC improves motor functional recovery after HI. In addition to the morphological evidence from TTC staining at 24 h and brain histology at 7 days post-HI, we reported that HPC improved functional and behavioral recovery. In the adult brain, the

170

H.-S. Sun et al. / Experimental Neurology 263 (2015) 161–171

brain–blood barrier (BBB) seems to be disrupted under ischemic conditions (Aarts et al, 2002) and thus may be more permeable to drugs in systemic circulation. The BBB undergoes developmental changes at an early postnatal age, and an immature BBB favors drug delivery into the neonatal brain. We recently reported that the volume-regulated anion channel (VRAC) blocker DCPIB passes through the less developed BBB in neonatal mice (Alibrahim et al, 2013) but not the BBB of adult rats (Zhang et al, 2008). In this study, we further demonstrated that the KATP channel blocker and agonist could cross the BBB after i.p. administration in the neonatal animals following HI brain injury. The ischemic tolerance induced by HPC was blocked by the KATP channel blocker tolbutamide in both morphological and functional recovery experiments, further supporting the notion that HPC in neonatal HI brain injury is mediated by neuronal KATP channels. Neonatal hypoxic–ischemic brain injury and its related brain disorders hypoxic–ischemic encephalopathy and cerebral palsy are a major health problem in children, and the related disabilities cause a significant social and economic burden to society worldwide (Centers for Disease Control and Prevention (CDC), 2004; Kruse et al, 2009). Finding the causes and eventual cure of intellectual impairment for this childhood disease can assist in establishing potential therapeutic strategy. Our study potentially has strong clinical implications with respect to neonatal HI brain injury and its subsequent illness hypoxic-ischemic encephalopathy and cerebral palsy. KATP channels are a potential therapeutic target for neonatal hypoxic–ischemic brain injury, and KATP channel openers may be useful in preventing damage and enhancing functional recovery in neonatal stroke in the future. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.expneurol.2014.10.003. Acknowledgments This work was supported by operating grants to HSS from the Heart and Stroke Foundation of Canada (G-13-0003069) and The Scottish Rite Charitable Foundation of Canada (#12104), and to ZPF from the National Sciences and Engineering Research Council of Canada (NSERC) (NSERC249962-09). Acknowledgments for a Saudi Arabian Cultural Bureau of Canada Scholarship for AA; China Scholarship Council (CSC) Fellowships for BX and AJX; CIHR studentship for ET (CIHR-CGS-M) and Ontario Graduate Studentship for AB and CYB. ZPF was a recipient of a New Investigator Award from the Heart and Stroke Foundation of Canada. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. References Aarts, M., Liu, Y., Liu, L., Besshoh, S., Arundine, M., Gurd, J.W., Wang, Y.T., Salter, M.W., Tymianski, M., 2002. Treatment of ischemic brain damage by perturbing NMDA receptor-PSD-95 protein interactions. Science 298, 846–850. Alibrahim, A., Zhao, L.Y., Bae, C.Y., Barszczyk, A., Lf, S.C., Wang, G.L., Sun, H.S., 2013. Neuroprotective effects of volume-regulated anion channel blocker DCPIB on neonatal hypoxic–ischemic injury. Acta Pharmacol. Sin. 34, 113–118. Ashcroft, F.M., Harrison, D.E., Ashcroft, S.J., 1984. Glucose induces closure of single potassium channels in isolated rat pancreatic beta-cells. Nature 312, 446–448. Ashford, M.L., Sturgess, N.C., Trout, N.J., Gardner, N.J., Hales, C.N., 1988. Adenosine-5′triphosphate-sensitive ion channels in neonatal rat cultured central neurones. Pflugers Arch. 412, 297–304. Autheman, D., Sheldon, R.A., Chaudhuri, N., von, A.S., Siegenthaler, C., Ferriero, D.M., Christen, S., 2012. Glutathione peroxidase overexpression causes aberrant ERK activation in neonatal mouse cortex after hypoxic preconditioning. Pediatr. Res. 72, 568–575. Babenko, A.P., Aguilar-Bryan, L., Bryan, J., 1998. A view of sur/KIR6.X, KATP channels. Annu. Rev. Physiol. 60, 667–687. Bajgar, R., Seetharaman, S., Kowaltowski, A.J., Garlid, K.D., Paucek, P., 2001. Identification and properties of a novel intracellular (mitochondrial) ATP-sensitive potassium channel in brain. J. Biol. Chem. 276, 33369–33374. Besancon, E., Guo, S., Lok, J., Tymianski, M., Lo, E.H., 2008. Beyond NMDA and AMPA glutamate receptors: emerging mechanisms for ionic imbalance and cell death in stroke. Trends Pharmacol. Sci. 29, 268–275. 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.

Bouslama, M., Renaud, J., Olivier, P., Fontaine, R.H., Matrot, B., Gressens, P., Gallego, J., 2007. Melatonin prevents learning disorders in brain-lesioned newborn mice. Neuroscience 150, 712–719. Brunet, A., Bonni, A., Zigmond, M.J., Lin, M.Z., Juo, P., Hu, L.S., Anderson, M.J., Arden, K.C., Blenis, J., Greenberg, M.E., 1999. Akt promotes cell survival by phosphorylating and inhibiting a Forkhead transcription factor. Cell 96, 857–868. Centers for Disease Control and Prevention (CDC), 2004. Economic costs associated with mental retardation, cerebral palsy, hearing loss, and vision impairment—United States, 2003. MMWR Morb. Mortal. Wkly Rep. 53, 57–59. Cook, D.L., Hales, C.N., 1984. Intracellular ATP directly blocks K+ channels in pancreatic Bcells. Nature 311, 271–273. Datta, S.R., Dudek, H., Tao, X., Masters, S., Fu, H., Gotoh, Y., Greenberg, M.E., 1997. Akt phosphorylation of BAD couples survival signals to the cell-intrinsic death machinery. Cell 91, 231–241. Dirnagl, U., Iadecola, C., Moskowitz, M.A., 1999. Pathobiology of ischaemic stroke: an integrated view. Trends Neurosci. 22, 391–397. Downward, J., 2004. PI 3-kinase, Akt and cell survival. Semin. Cell Dev. Biol. 15, 177–182. Fatemi, A., Wilson, M.A., Johnston, M.V., 2009. Hypoxic–ischemic encephalopathy in the term infant. Clin. Perinatol. 36, 835–858 (vii). 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. Garlid, K.D., 2000. Opening mitochondrial K(ATP) in the heart—what happens, and what does not happen. Basic Res. Cardiol. 95, 275–279. Horimoto, H., Gaudette, G.R., Saltman, A.E., Krukenkamp, I.B., 2000. The role of nitric oxide, K(+)(ATP) channels, and cGMP in the preconditioning response of the rabbit. J. Surg. Res. 92, 56–63. Kruse, M., Michelsen, S.I., Flachs, E.M., Bronnum-Hansen, H., Madsen, M., Uldall, P., 2009. Lifetime costs of cerebral palsy. Dev. Med. Child Neurol. 51, 622–628. Levine, S., 1960. Anoxic–ischemic encephalopathy in rats. Am. J. Pathol. 36, 1–17. Light, P.E., Kanji, H.D., Fox, J.E., French, R.J., 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. Liu, X.H., Yan, H., Xu, M., Zhao, Y.L., Li, L.M., Zhou, X.H., Wang, M.X., Ma, L., 2013. Hyperbaric oxygenation reduces long-term brain injury and ameliorates behavioral function by suppression of apoptosis in a rat model of neonatal hypoxia–ischemia. Neurochem. Int. 62, 922–930. Miki, T., Nagashima, K., Seino, S., 1999. The structure and function of the ATP-sensitive K+ channel in insulin-secreting pancreatic beta-cells. J. Mol. Endocrinol. 22, 113–123. Mullen, R.J., Buck, C.R., Smith, A.M., 1992. NeuN, a neuronal specific nuclear protein in vertebrates. Development 116, 201–211. Munoz, A., Nakazaki, M., Goodman, J.C., Barrios, R., Onetti, C.G., Bryan, J., guilar-Bryan, L., 2003. Ischemic preconditioning in the hippocampus of a knockout mouse lacking SUR1-based K(ATP) channels. Stroke 34, 164–170. Murry, C.E., Jennings, R.B., Reimer, K.A., 1986. Preconditioning with ischemia: a delay of lethal cell injury in ischemic myocardium. Circulation 74, 1124–1136. Nelson, K.B., 2007. Perinatal ischemic stroke. Stroke 38, 742–745. Nelson, K.B., Grether, J.K., 1999. Causes of cerebral palsy. Curr. Opin. Pediatr. 11, 487–491. Nelson, K.B., Lynch, J.K., 2004. Stroke in newborn infants. Lancet Neurol. 3, 150–158. Noma, A., 1983. ATP-regulated K+ channels in cardiac muscle. Nature 305, 147–148. Obrenovitch, T.P., 2008. Molecular physiology of preconditioning-induced brain tolerance to ischemia. Physiol. Rev. 88, 211–247. Prass, K., Scharff, A., Ruscher, K., Lowl, D., Muselmann, C., Victorov, I., Kapinya, K., Dirnagl, U., Meisel, A., 2003. Hypoxia-induced stroke tolerance in the mouse is mediated by erythropoietin. Stroke 34, 1981–1986. Qi, X., Hosoi, T., Okuma, Y., Kaneko, M., Nomura, Y., 2004. Sodium 4-phenylbutyrate protects against cerebral ischemic injury. Mol. Pharmacol. 66, 899–908. Rice III, J.E., Vannucci, R.C., Brierley, J.B., 1981. The influence of immaturity on hypoxic– ischemic brain damage in the rat. Ann. Neurol. 9, 131–141. Sanches, E.F., Arteni, N.S., Spindler, C., Moyses, F., Siqueira, I.R., Perry, M.L., Netto, C.A., 2012. Effects of pre- and postnatal protein malnutrition in hypoxic–ischemic rats. Brain Res. 1438, 85–92. Seino, S., 1999. ATP-sensitive potassium channels: a model of heteromultimeric potassium channel/receptor assemblies. Annu. Rev. Physiol. 61, 337–362. Sheldon, R.A., Aminoff, A., Lee, C.L., Christen, S., Ferriero, D.M., 2007. Hypoxic preconditioning reverses protection after neonatal hypoxia–ischemia in glutathione peroxidase transgenic murine brain. Pediatr. Res. 61, 666–670. Sun, H.S., Feng, Z.P., Miki, T., Seino, S., French, R.J., 2006. Enhanced neuronal damage after ischemic insults in mice lacking Kir6.2-containing ATP-sensitive K+ channels. J. Neurophysiol. 95, 2590–2601. Sun, H.S., Feng, Z.P., Barber, P.A., Buchan, A.M., French, R.J., 2007. Kir6.2-containing ATPsensitive potassium channels protect cortical neurons from ischemic/anoxic injury in vitro and in vivo. Neuroscience 144, 1509–1515. Sun, H.S., Doucette, T.A., Liu, Y., Fang, Y., Teves, L., Aarts, M., Ryan, C.L., Bernard, P.B., Lau, A., Forder, J.P., Salter, M.W., Wang, Y.T., Tasker, R.A., Tymianski, M., 2008. Effectiveness of PSD95 inhibitors in permanent and transient focal ischemia in the rat. Stroke 39, 2544–2553. Sun, H.S., Jackson, M.F., Martin, L.J., Jansen, K., Teves, L., Cui, H., Kiyonaka, S., Mori, Y., Jones, M., Forder, J.P., Golde, T.E., Orser, B.A., Macdonald, J.F., Tymianski, M., 2009. Suppression of hippocampal TRPM7 protein prevents delayed neuronal death in brain ischemia. Nat. Neurosci. 12, 1300–1307. Terzic, A., 1999. New frontiers of cardioprotection. Clin. Pharmacol. Ther. 66, 105–109. Tymianski, M., 2011. Emerging mechanisms of disrupted cellular signaling in brain ischemia. Nat. Neurosci. 14, 1369–1373.

H.-S. Sun et al. / Experimental Neurology 263 (2015) 161–171 Vannucci, S.J., 2007. Cerebral ischemia and the developing brain: introduction. Stroke 38, 723. Vannucci, S.J., Hagberg, H., 2004. Hypoxia–ischemia in the immature brain. J. Exp. Biol. 207, 3149–3154. Wachtel, E.V., Hendricks-Munoz, K.D., 2011. Current management of the infant who presents with neonatal encephalopathy. Curr. Probl. Pediatr. Adolesc. Health Care 41, 132–153.

171

Yamada, K., Ji, J.J., 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. Zhang, Y., Zhang, H., Feustel, P.J., Kimelberg, H.K., 2008. DCPIB, a specific inhibitor of volume regulated anion channels (VRACs), reduces infarct size in MCAo and the release of glutamate in the ischemic cortical penumbra. Exp. Neurol. 210, 514–520.