B R A I N R E S E A R C H 1 2 8 5 ( 2 0 0 9 ) 1 9 6 – 2 0 7

ava i l ab l e a t www.sc i enced i r ec t . com

www.e l sev i e r. com/ loca te /b ra in res

Research Report

Immediate neuronal preconditioning by NS1619

Tamás Gáspára,⁎, Ferenc Domokia,b, Laura Lentia,b, Prasad V.G. Katakama,James A. Snipesa, Ferenc Barib, David W. Busijaa

aDepartment of Physiology and Pharmacology, Wake Forest University Health Sciences, Medical Center Blvd, Winston-Salem, NC 27157, USAbDepartment of Physiology, University of Szeged, Faculty of Medicine, Dóm tér 10, Szeged, 6720, Hungary

A R T I C L E I N F O

⁎ Corresponding author. Fax: +1 336 716 0504.E-mail address: tamas74@freemail.hu (T.

0006-8993/$ – see front matter © 2009 Elsevdoi:10.1016/j.brainres.2009.06.008

A B S T R A C T

Article history:Accepted 5 June 2009Available online 11 June 2009

The objectives of our present experiments were to determine whether the BKCa channelagonist NS1619 is able to induce immediate preconditioning in cultured rat cortical neuronsand to elucidate the role of BKCa channels in the initiation of immediate preconditioning.NS1619 depolarized mitochondria and increased reactive oxygen species (ROS) generation,but neither of these effects was inhibited by BKCa channel antagonists. NS1619 alsoactivated the extracellular signal-regulated kinase signaling pathways. One-hour treatmentwith NS1619 induced immediate protection against glutamate excitotoxicity (viability 24 hafter glutamate exposure: control, 58.45±0.95%; NS1619 50 μM, 78.99±0.90%⁎; NS1619100 μM, 86.89±1.20%⁎; NS1619 150 μM, 93.23±1.23%⁎; mean±SEM; ⁎p<0.05 vs. control;n=16–32). Eliminating ROS during the preconditioning phase effectively blocked thedevelopment of cytoprotection. In contrast, the BKCa channel blockers iberiotoxin andpaxilline, the phosphoinositide 3-kinase inhibitor wortmannin, the protein kinase C blockerchelerythrine, and the mitogen activated protein kinase antagonist PD98059 were unable toantagonize the immediate neuroprotective effect. Finally, preconditioning with NS1619reduced the calcium load and ROS surge upon glutamate exposure and increasedsuperoxide dismutase activity. Our results indicate that NS1619 is an effective inducer ofimmediate neuronal preconditioning, but the neuroprotective effect is independent of theactivation of BKCa channels.

© 2009 Elsevier B.V. All rights reserved.

Keywords:NeuroprotectionMitochondriaBKCa channelReactive oxygen speciesGlutamate

1. Introduction

Preconditioning (PC) is a naturally occurring phenomenon inwhich a sublethal or nontoxic stimulus renders cells, tissues,and organs tolerant to a subsequent, otherwise lethal insult.This endogenous protective mechanism was first recognizedmore than 60 years ago by Noble (1943), and later by Dahl andBalfour (1964). The term ‘preconditioning’ was first used fortolerance induction in the same year (Janoff, 1964). Despitethese early reports, research in this field evolved only from thesecond half of the 1980s (Kitagawa et al., 1990; Murry et al.,

Gáspár).

ier B.V. All rights reserve

1986; Schurr et al., 1986). Subsequent studies demonstrated anearly or classical phase occurring virtually immediately afterthe triggering stimulus and lasting 2–3 h, and a late phase orsecond window of protection appearing 24 h later and lasting1–3 days (Kuzuya et al., 1993; Marber et al., 1993).

Mitochondrial potassium channels are generally acceptedto be major components of PC-induced cytoprotection. Sincethe first demonstration of ATP-sensitive potassium channelsin the inner membrane of mitochondria (Inoue et al., 1991),numerous studies, including our own, have demonstratedneuroprotection using selective channel openers (Busija et al.,

d.

197B R A I N R E S E A R C H 1 2 8 5 ( 2 0 0 9 ) 1 9 6 – 2 0 7

2004; Domoki et al., 1999; Gaspar et al., 2008b; Kis et al., 2003,2004; Mayanagi et al., 2007). Recently, another type ofpotassium channel, the large conductance calcium activatedpotassium (BKCa) channel was reported to be present in themitochondria (mitoBKCa) of a glioma cell line (Siemen et al.,1999), heart muscle (Xu et al., 2002), brain (Douglas et al., 2006),and skeletal muscle cells (Skalska et al., 2008). Studies havealso demonstrated that cardiac PC with the BKCa channelopener 1,3-dihydro-1-[2-hydroxy-5-(trifluoromethyl)phenyl]-5-(trifluoromethyl)-2H-benzimidazol-2-one (NS1619) could beinhibited by BKCa channel antagonists (Cao et al., 2005; Satoet al., 2005; Shintani et al., 2004; Wang et al., 2004; Xu et al.,2002). In contrast to the heart studies, our recent publicationprovides convincing evidence that delayed neuronal PC withNS1619 is independent of BKCa channels (Gaspar et al., 2008a).In that study we found that NS1619 induced depolarizationand reactive oxygen species (ROS) generation in isolatedmitochondria and subsequent activation of the phosphoinosi-tide-3-kinase (PI3K)— Akt signal transduction pathway whichultimately led to delayed neuroprotection against varioustoxic insults. Of the numerous effects, however, only cellmembrane hyperpolarization was effectively counteracted byBKCa channel inhibitors.We also looked for the constituents ofthe BKCa channel in mitochondrial preparations, but wereunable to demonstrate the presence of the pore forming αsubunit. Nevertheless, no immediate neuronal PC studies arecurrently available using BKCa channel activators; thus, thefunction of BKCa channels in immediate neuronal PC is yet tobe determined.

The purpose of our present study was to examine whetherNS1619 induces immediate PC in rat cortical neuronalcultures. We also investigated the role of BKCa channels andseveral signaling molecules in the development of immediatecytoprotection. Finally, we tested the effect of immediate PCwith NS1619 on ROS generation and cytosolic free calciumlevels in neurons.

2. Results

Preliminary experiments showed that treatment with NS1619for 30 min, or less, did not induce immediate PC (viability24 h after exposure to 200 μM glutamate for 1 h: untreatedcontrol, 53.75±1.62%; NS1619 150 μM for 15 min, 55.09±2.37%;NS1619 150 μM for 30 min, 58.67±2.27%; percent of untreatedcontrol unexposed to glutamate; mean±SEM; n=8 in eachgroup). However, treatment with NS1619 for 60 min inducedimmediate PC.

Treatment of quiescent cells with concentrations ofNS1619 up to 150 μM for 1 h had no detectable effect onneuronal survival whereas higher doses resulted indecreased viability (Fig. 1A). Therefore, a 1-hour treatmentwith 25–150 μM NS1619 was used to elicit immediate PC insubsequent experiments. Within this range, the compoundinduced a dose-dependent protection against glutamateexcitotoxicity (Fig. 1B) and, to a lesser extent, againstcombined oxygen and glucose deprivation (OGD) (Fig. 1C)and exogenous hydrogen peroxide (Fig. 1D). Because the bestprotection was observed against glutamate, this experimen-tal paradigm was chosen for further experiments investigat-

ing the mechanism of immediate PC. To determine theduration of the protection, neuronal cultures were treatedwith 150 μM NS1619 for 1 h, then the compound was washedout and the cells were incubated in the regular culturemedium for 0, 1, or 2 h before exposure to glutamate(200 μM; 60 min). These experiments showed that thewindow of immediate PC is very narrow and providesprotection only for about 2 h (viability 24 h after exposureto glutamate: untreated, 64.09±1.27%; NS1619 150 μM+0 hincubation before glutamate exposure, 94.66±1.05%⁎; NS1619150 μM+1 h incubation, 73.18±1.66%⁎; NS1619 150 μM+2 hincubation, 63.42±1.42%; mean±SEM; ⁎p<0.05 vs. untreated;n=12–24).

Application of NS1619 resulted in a dose-dependentdepolarization of in situ mitochondria in cultured corticalneurons (Fig. 2A). The BKCa channel inhibitors, iberiotoxin(IbTx; 5 μM) and paxilline (Pax; 20 μM) could not block thiseffect (Fig. 2B). Control neurons showed a moderate increasein hydroethidine (HEt) fluorescence over time which resultedfrom the basal ROS formation by the cells (Fig. 2C). NS1619induced a dose-dependent elevation of ROS production (Fig.2C) which, similar to the effect on the mitochondrialmembrane potential, was unresponsive to BKCa channelantagonists (Fig. 2D). To determine whether ROS generationinduced by NS1619 was responsible for mitochondrial depo-larization, we measured TMRE fluorescence in the presence ofeither catalase or the combination of the superoxide dismu-tase (SOD) mimetic M40401 and catalase and found that theantioxidants did not influence the depolarizing effect ofNS1619 (TMRE fluorescence: NS1619 150 μM, 69.51±1.98%;NS1619 150 μM+catalase 100 U/ml, 66.58±1.84%; NS1619150 μM+catalase 100 U/ml+M40401 50 μM, 68.46±1.90%;percent of untreated control; mean±SEM, n=14 in eachgroup). Removal of NS1619 quickly reversed effects onmitochondrial membrane potential and ROS generation.Washing out the compound resulted in a rapid repolarization,reaching a steady state slightly below baseline within 30 min(Fig. 3A). Stimulated levels of ROS generation fell to and even alittle bit below baseline immediately after washout (Fig. 3B).

To determine the role of BKCa channels in the induction ofimmediate PC, neuronal cultures were co-treated with NS1619(150 μM), and with either IbTx (5 μM) or Pax (20 μM) for 60 minand then were exposed to glutamate. The selected concentra-tions of BKCa antagonists showed no toxicity. Additionally,neither of them could antagonize the protective effect ofNS1619 against glutamate (Fig. 4A). We also tested whetherROS contributed to the immediate PC effect of NS1619. Thecells were treated with NS1619, with or without the combina-tion of M40401 (50 μM) and catalase (100 U/ml) or catalase(100 U/ml) alone for 60 min, then the compounds werethoroughly washed and the cultures were exposed to gluta-mate (200 μM, 60 min). The ROS scavengers, alone or withNS1619, did not affect cell viability. However, whereas catalasealone did not inhibit the protection the combination of M40401and catalase significantly reduced the neuroprotective effectof NS1619 against glutamate (Fig. 4B).

To elucidate the function of the PI3K and protein kinase C(PKC) signaling pathways in immediate PC, we tested whetherthe PI3K inhibitor wortmannin (Wrt) and the PKC blockerchelerythrine (Chel) would antagonize NS1619-induced

Fig. 1 – Immediate preconditioning with NS1619 protects neuronal cultures against various toxic insults. Neuronal cultureswere treated with increasing doses of NS1619 (25–300 μM) for 60 min then the compound was thoroughly washed out.Panel A shows the direct effect of NS1619 on cell viability 24 h after treatment. To test the protective effect of NS1619, othercultures were exposed to 200 μM glutamate (Glut) for 60 min (Panel B), combined oxygen and glucose deprivation (OGD) for180min (Panel C), or 50 μM exogenous hydrogen peroxide (H2O2) in the presence of glucose oxidase (1 U/l) for 60 min (Panel D),immediately after washing out NS1619. In all experiments, viability was measured by assaying lactate dehydrogenasereleased from living cells 24 h after the insults and was expressed as a percent of the viability of control cultures whichwere not treated with NS1619 and were not exposed to any of the toxic insults (white bars). *Significant difference (p<0.05)compared to untreated control neuronswhichwere not exposed to any of the insults. #Significant difference (p<0.05) comparedto untreated cultures which were exposed to the corresponding toxic insult. Data are expressed as mean±SEM; cells fromat least two individual cultures (n=16–32).

198 B R A I N R E S E A R C H 1 2 8 5 ( 2 0 0 9 ) 1 9 6 – 2 0 7

tolerance against glutamate exposure. The doses of Wrt andChel were selected so that they did not significantly affect theviability of control neurons unexposed to glutamate. Thepresence of antagonists, however, did not change the neuro-protection elicited by NS1619 against glutamate excitotoxicity(Fig. 5A). We also tested whether the compound would alsoaffect mitogen activated protein kinases (MAPK) and found arapid phosphorylation of extracellular signal-regulated kinase2 (ERK-2) which reached itsmaximumwithin 15min. A similartendency was observed with ERK-1; however, the changeswere not significant (Fig. 5B). We then treated neuronalcultures with the MAPK kinase inhibitor PD98059 (PD; 50 μM)along with NS1619 for 1 h before exposure to glutamate. Thedose of PD was chosen so that it did not influence the viabilityof control neurons. In addition, blockingMAPKs did not inhibitimmediate PC with NS1619 (Fig. 5C).

In the last set of experiments we attempted to identify theeffectors of immediate PC by NS1619. We first incubatedneuronswith NS1619 to induce PC thenmeasured the changesof intracellular free calcium ([Ca2+]i) and ROS levels inresponse to exposure to glutamate using the fluorescentdyes, Fluo-4 and HEt, respectively. The Fluo-4 fluorescent

signal of quiescent control and NS1619-treated neurons didnot show any difference and remained stable over time.Whenexposed to glutamate (200 μM), the peak [Ca2+]i value of thecultures pretreated with NS1619 was significantly lower thanthat of the control cultures, whereas the rate of calciumelimination did not seem to be affected by NS1619 treatment(Fig. 6A). Since reduced calcium response to glutamate may bea consequence of NMDA receptor down-regulation, weperformed western blot analysis of the NR1 subunit of theNMDA receptor using the membrane fraction isolated fromneurons after 60 min treatment with NS1619. However,NS1619 did not change the expression of the examinedsubunit (Fig. 6B).

Challenging quiescent neurons with glutamate (200 μM)resulted in an increase in free radical production detected asan increased rate of conversion of HEt to the fluorescent endproduct, ethidium (Fig. 6C). Immediate PC with NS1619 causeda mild dose-dependent decrease in non-stimulated ROSlevels. Additionally, a dose-dependent reduction in gluta-mate-induced ROS generation was observed in NS1619-preconditioned cultures. Incubation with 200 μM NS1619resulted in a higher basal ROS level and no further decrease

Fig. 2 – NS1619 induces mitochondrial depolarization and reactive oxygen species (ROS) generation in intact neurons,independently of large conductance calcium activated potassium channels. To assess mitochondrial membrane potential,cultured neurons in black-walled 96-well plates were treated with NS1619 (NS; 25–200 μM) and loaded withtetramethylrhodamine ethyl ester (TMRE) for 20 min (Panel A). Another set of neurons was pretreated with iberiotoxin(IbTx; 5μM) or paxilline (Pax; 20μM) for 5min before loadingwith TMRE and treatmentwith NS (150μM) (Panel B). After 20min,the cells were washed, and TMRE fluorescence was measured using a fluorescent microplate reader. *Significant difference(p<0.05) compared to untreated control cultures (n=16–24). ROS production was monitored using the fluorescent dyehydroethidine (HEt). Cultured neurons were treated with NS1619 (10–150 μM) and loaded with HEt at the same time, in thesame buffer, 1 min before ROS determination (Panel C). HEt-fluorescence was recorded every minute for 30 min using afluorescent microplate reader. NS1619 at 10 μM (○) did not induce any changes in HEt-fluorescence; therefore, this curveoverlies untreated control (●). *Significant difference (p<0.05) compared to the fluorescent intensity of untreated controlneurons (n=16 in each group). To evaluate the effect of K+ channel inhibitors on NS1619 induced ROS generation, neurons werepretreated with IbTx (5 μM), or Pax (20 μM) for 5 min and then were treated with NS1619 (150 μM) and loaded with HEt(5μM) 1 min before ROS determination (Panel D). HEt fluorescent intensity was measured every minute for 30 min with afluorescent microplate reader. *Significant difference (p<0.05) compared to the fluorescent intensity of untreated controlneurons. #Significant difference (p<0.05) compared to the fluorescent intensity of neurons treated with NS1619 (150 μM). Dataare expressed as mean±SEM (n=24 in each group).

199B R A I N R E S E A R C H 1 2 8 5 ( 2 0 0 9 ) 1 9 6 – 2 0 7

in glutamate stimulated ROS generation (Fig. 6C). Besidesreduced calcium influx upon glutamate exposure, anothercause of reduced ROS response might be an increasedelimination of radicals by antioxidant systems. Therefore,we examined superoxide dismutase activity and expression.Incubation of neurons with NS1619 for 1 h resulted in a mildbut significant and dose-dependent increase in total SODactivity (untreated control, 100.00±2.07%; NS1619 50 μM,108.66±5.62%; NS1619 100 μM, 117.49±6.37%; NS1619 150 μM,124.66±5.24%⁎; ⁎p<0.05 vs. untreated control; mean±SEM;n=10 in each group), whereas western blot analysis did notdetect any increase in protein levels (data not shown).

Cellular ATP content is a sensitive indicator of stressinduced by glutamate exposure. Pretreatment with 150 μMNS1619 did not influence resting ATP levels, but NS1619-treated neurons contained significantly higher ATP levels

after a 60-minute exposure to 200 μM glutamate as comparedwith control cells (ATP levels in 10−18 M/cell measuredimmediately after challenge: untreated control unexposed toglutamate, 2893±23; NS1619 150 μM unexposed to glutamate,2951±21; untreated cells exposed to glutamate, 1706±14⁎;NS1619 150 μM exposed to glutamate, 2146±25⁎#; ⁎p<0.05 vs.untreated control unexposed to glutamate; #p<0.05 vs.untreated cells exposed to glutamate; mean±SEM, n=16 ineach group).

3. Discussion

This is the first study demonstrating immediate neuronal PCwith NS1619. Here we show that 1) 1 h incubation with NS1619induces immediate neuronal PC that is protective against

Fig. 4 – Reactive oxygen species (ROS) but not largeconductance calcium activated potassium (BKCa) channelactivation is required for the immediate preconditioningeffect of NS1619. Cultured cortical neurons were treated withNS1619 (150μM)with or without the BKCa channel inhibitors,iberiotoxin (IbTx; 5 μM) or paxilline (Pax; 20 μM) to blockBKCa activity (Panel A). In a second experiment, catalase(100 U/ml) alone (C) or the superoxide dismutase mimeticM40401 (50 μM) and 100 U/ml catalase (M+C) were appliedtogether with NS1619 (150 μM) to eliminate ROS generatedduring the initiating phase of preconditioning (Panel B). After60 min the compounds were washed out and the cultureswere exposed to 200 μM glutamate for 60 min. Cell viabilitywas measured 24 h later and was expressed as a percent ofcontrol cultures which were not treated with any of thecompounds and were not exposed to glutamate. *Significantdifference (p<0.05) compared to untreated control cultureswhichwere not exposed to glutamate. #Significant difference(p<0.05) compared to untreated cultureswhichwere exposedto glutamate. Data are expressed asmean±SEM; cells from atleast two individual cultures (n=8–32).

Fig. 3 – Washing out NS1619 quickly restores mitochondrialmembrane potential and reactive oxygen species (ROS)generation. Mitochondrial membrane potential wasmonitored with tetramethylrhodamine ethyl ester (TMRE).TMRE fluorescence was determined before (open bar;baseline), during (grey bar), and 0, 15, 30, and 60 min after(black bars) washout following 1 h of NS1619 treatment(150 μM) using a fluorescent microplate reader (Panel A).*Significant difference (p<0.05) compared to the fluorescentintensity of untreated control neurons. #Significant difference(p<0.05) compared to the fluorescent intensity of neuronsduring treatment with 150 μM NS1619 (grey bar). Data areexpressed as mean±SEM (n=16 in each group). ROSgeneration before (open bar; baseline), during (grey bar), 0, 15,30, and 60 min after (black bars) washout following 1 h ofNS1619 treatment (150 μM) was determined with thesuperoxide sensitive fluorescent dye, hydroethidine (HEt) ina fluorescent microplate reader (Panel B). HEt-fluorescencewas detected every minute for 30 min then changes influorescent intensity perminutewere calculated. *Significantdifference (p<0.05) compared to the fluorescent intensity ofuntreated control neurons. #Significant difference (p<0.05)compared to the fluorescent intensity of neurons duringtreatmentwith 150μMNS1619 (grey bar). Data are expressedas mean±SEM (n=16 in each group).

200 B R A I N R E S E A R C H 1 2 8 5 ( 2 0 0 9 ) 1 9 6 – 2 0 7

various neurotoxic insults; 2) the immediate effects of NS1619are transient mitochondrial depolarization and increased ROSgeneration in neurons; 3) elimination of ROS during the PCphase significantly reduces the neuroprotective effect ofNS1619; 4) antagonizing BKCa channels, and blocking the

PI3K, PKC, and ERK signal transduction pathways do notcounteract immediate PC; and finally, 5) consequences ofimmediate NS1619 PC include decreased Ca2+ influx throughglutamate receptors, increased SOD activity, a consequentreduced ROS response during glutamate stimulation, andbetter preservation of ATP.

The first report on the neurovascular protective effect of aBKCa channel openerwas published by our laboratory a decadeago in which we showed that pretreatment with NS1619preserved cerebrovascular response to NMDA after combined

201B R A I N R E S E A R C H 1 2 8 5 ( 2 0 0 9 ) 1 9 6 – 2 0 7

hypoxia and ischemia (Veltkamp et al., 1998). However, in thatstudy we did not identify the exact mechanism of theneuroprotection; nevertheless, our results suggested thatthe protection was independent of the vasodilator effect ofthe potassium channel openers. Parallel with the recentreports on BKCa channels in the inner mitochondrial mem-brane of several cell types (Douglas et al., 2006; Siemen et al.,1999; Skalska et al., 2008; Xu et al., 2002), a surge of studiesappeared reporting cardiac PC using BKCa channel activators,most frequently NS1619 (Cao et al., 2005; Sato et al., 2005;Stowe et al., 2006; Wang et al., 2004; Xu et al., 2002). Althoughclaimed to be selective by many authors, NS1619 is wellknown to have numerous BKCa channel-independent effects(Debska et al., 2003; Edwards et al., 1994; Holland et al., 1996;Kicinska and Szewczyk, 2004; Korper et al., 2003; Park et al.,2007; Yamamura et al., 2001) whichmight, at least partially, beresponsible for the protective effect. Additionally, in ourprevious study we could not confirm the presence of the αsubunit of the BKCa channel whereas we could detect themodulatory β4 subunit in our mitochondrial preparationsisolated from the rat cortex using commercially available

antibodies (Gaspar et al., 2008a). Since then we have testedother antibodies directed against the pore forming subunit(from Sigma and BD) but were unable to demonstrate itspresence in mitochondria (unpublished data). These resultsare seemingly in contrast with the report of Douglas et al.(2006) who demonstrated the presence of the α subunit of thechannel in brain mitochondria. However, the pore formingsubunit was only present in cerebellar neurons, includingPurkinje cells and granular cells, and only in a small fraction ofmitochondria which questions its role as an inducer ofneuronal PC and supports our view.

A one-hour treatment with NS1619 induced a dose-dependent immediate protection against several toxic chal-lenges. Shorter treatment periods were insufficient to evokesignificant tolerance providing evidence for the time-depen-dence of the development of immediate PC. Whereas in ourprevious study the best delayed protection was seen againsthydrogen peroxide (Gaspar et al., 2008a), here we found thatthe immediate PC effect of NS1619 provided the highest levelof protection against glutamate excitotoxicity. Previously we

Fig. 5 – The activation of the phosphoinositide 3-kinase(PI3K), protein kinase C (PKC), and extracellularsignal-regulated kinase (ERK) pathways does not contributeto the immediate preconditioning effect of NS1619. Culturedcortical neurons were treatedwith NS1619 (150μM) andwiththe PI3K inhibitor wortmannin (Wrt; 300 nM), or the PKCinhibitor chelerythrine (Chel; 5μM) for 60min (Panel A). Thenthe compounds were washed out and the cells wereexposed to glutamate (200 μM) for 60 min. Cell viability wasmeasured 1 day later and was expressed as a percent ofcontrol cultures which were not treated with any of thecompounds and were not exposed to glutamate.*Significantdifference (p<0.05) compared to untreated control cultureswhichwere not exposed to glutamate. #Significant difference(p<0.05) compared to untreated cultureswhichwere exposedto glutamate. Data are expressed as mean±SEM; cells fromat least two individual cultures (n=8–16). Neuronal culturesin 35 mm dishes were treated with NS1619 (100 μM) for15 min, 30 min, 1 h, and 3 h after which proteins wereextracted and subjected to western blot analysis for total andphosphorylated ERKs (Panel B). The levels of phosphorylatedforms were normalized to the total amount of the sameprotein. Inset shows representative blots. *Significantdifference (p<0.05) compared to the normalized protein leveof untreated control. Data are expressed as mean±SEM (n=4in each group). Cultured cortical neurons were treated withNS1619 (150 μM) with or without the mitogen activatedprotein kinase inhibitor PD98059 (PD98; 50 μM) for 60 min(Panel C). Subsequently, the cultures were exposed toglutamate (200 μM) for 60 min. Cell viability was measured24 h later and was expressed as a percent of control cultureswhich were not treated with any of the compounds and werenot exposed to glutamate. *Significant difference (p<0.05)compared to untreated control cultures which were notexposed to glutamate. #Significant difference (p<0.05)compared to untreated cultures which were exposed toglutamate. Data are expressed as mean±SEM; cells from atleast two individual cultures (n=12–16).

l

202 B R A I N R E S E A R C H 1 2 8 5 ( 2 0 0 9 ) 1 9 6 – 2 0 7

reported similar neuroprotection against glutamate afterimmediate PC using the mitochondrial ATP-sensitive potas-sium channel opener BMS-191095 (Kis et al., 2004). Our presentresults also confirm that the protective effect of immediate PClasts only for a short period.

Compared to heart studies in which 10–30 μM NS1619 wasusually used to induce cardiac PC (Sato et al., 2005;Wang et al.,2004; Xu et al., 2002), in our current and previous experiments,optimal neuroprotection by immediate and delayed neuronalPC was achieved using 100–150 μM NS1619. The need forhigher concentrations to induce neuronal PC might beexplained by the fact that our experiments were performedin protein containingmedia and buffers whichmay reduce theavailability of the compound. This hypothesis is supported bythe fact that at least 100 μM NS1619 was needed to inducesignificant membrane hyperpolarization in a protein contain-ing transport medium (Gaspar et al., 2008a) as compared with10–20 μM in a protein-free buffer (Gaspar et al., 2007; Mayanagiet al., 2007). This effect of NS1619 could be antagonized by BKCa

channel inhibitors, providing pharmacological evidence forthe presence of plasma membrane BKCa channels in ourneuronal preparations (Gaspar et al., 2008a). Similar to ourrecent experiments using NS1619, in previous studies withdiazoxide, a higher concentration of the compound wasneeded to induce PC in neurons (Kis et al., 2003) than thatreported in the heart (Garlid et al., 1997).

In our previous studywe examined the effects of NS1619 onisolatedmitochondria and found that it caused depolarizationand ROS generation that were not antagonized by BKCa

channel blockers (Gaspar et al., 2008a). However, the isolationprocedure may influence the integrity of mitochondria andexperiments on isolated mitochondria do not always corre-spond to findings in intact cells. Therefore, in the presentstudy, we examined the direct effects of NS1619 onmitochon-drial function in intact neurons. Verifying our previous resultswith isolated mitochondria, NS1619 induced mitochondrialdepolarization and elevated ROS production in cultured cells.These findings are in partial agreement with those of Stowe etal. (2006) who demonstrated that NS1619-induced cardiopro-tection could be abolished with a ROS scavenger, confirmingincreased ROS generation after NS1619 application. Theseauthors also reported that the protection was similarlyantagonized by Pax. In our experiments, however, neithermitochondrial depolarization nor enhanced ROS formationcould be blocked with BKCa channel antagonists even usingthe lipophilic Pax as well as the polypeptide IbTx. To

Fig. 6 – Immediate preconditioning with NS1619 results inreduced Ca2+ load and decreased ROS availability uponexposure to glutamate without altering the expression of theNR1 subunit of the NMDA receptor. Intracellular free calciumlevels in cultured neurons were measured by a confocalmicroscope using Fluo-4AM during exposure to 200 μMglutamate (panel A). To induce preconditioning, neuronalcultureswere treatedwithNS1619 (150μM) for 60minprior tomeasurement. Fluorescent intensity of the untreated controlat the starting point was regarded as 100%. *Significantdifference (p<0.05) compared to untreated control. Data areexpressed as mean±SEM (n=50–112). Neuronal cultures in35 mm dishes were treated with NS1619 (50, 150 μM) for 1 hthen protein was extracted and subjected to western blotanalysis for the NMDA receptor subunit NR1 (Panel B). Bandswere normalized toβ-actin andwere expressed as percent ofuntreated control. Inset demonstrates a representative blot.Data are shown as mean±SEM (n=4 in each group). Culturedneurons in poly-D-lysine coated black-walled 96-well plateswere treatedwithNS1619 (25–200μM) 1 h before loadingwithROS-sensitive fluorescent dye hydroethidine (HEt) (Panel C).Neurons were exposed to glutamate (200 μM) then HEtfluorescent intensity was measured for 30 min with afluorescentmicroplate reader andwas expressed as a changein relative fluorescent units/minute (ΔRFU/min). *Significantdifference (p<0.05) compared to the fluorescent intensity ofuntreated control neurons unexposed to glutamate.#Significant difference (p<0.05) compared to untreatedneurons that were exposed to glutamate. Data are expressedas mean±SEM; cells from at least two individual cultures(n=16–32).

203B R A I N R E S E A R C H 1 2 8 5 ( 2 0 0 9 ) 1 9 6 – 2 0 7

determine whether ROS generation is upstream to mitochon-drial depolarization, we performed TMRE measurements inthe presence of antioxidants. Catalase and M40401, however,did not influence the depolarizing effect of NS1619 suggestingthat both mitochondrial ROS generation and depolarizationmay be the consequences of the inhibitory effect of NS1619 onthe respiratory chain (Debska et al., 2003; Kicinska andSzewczyk, 2004). Mitochondria quickly repolarized after wash-ing out NS1619, but reached a slightly lower-than-baselineresting potential after about 30 min. Washout also caused arapid fall in ROS generation, resulting in an about 25% lower-than-baseline level of free radical production. This reducedrate of ROS generation was dose-dependent (Fig. 6C) andreturned to control levels 1 h after washout (Fig. 3B). Theseexperiments demonstrated that the mitochondrial effects ofNS1619 were only transient and reversible by washout.

To identify the underlyingmechanisms of tolerance induc-tion, we applied BKCa channel antagonists along with NS1619before glutamate exposure. Consistent with our results onmitochondrial depolarization and ROS generation, we foundthat the BKCa blockers were ineffective in terms of inhibition ofprotection. Consequently, our results provide strong evidencethat BKCa channel activation (mitochondrial and/or plasma-lemmal) does not induce tolerance in neurons and does notplay any role in the immediate neuronal PC effect of NS1619.Since PI3K activation was shown to be crucial in delayed PC byNS1619 (Gaspar et al., 2008a), we also tested whether the pro-survival signal transduction pathways known to be activatedby the compound play any particular role in the immediateneuroprotective effect. Thus, we co-applied NS1619 with thePI3K blocker Wrt, the PKC inhibitor Chel, and the MAPK kinaseinhibitor PD. However, none of these compounds blockedimmediate PC, suggesting that the activation of these path-ways is not essential in immediate tolerance induction.

Several studies, including those from our laboratory, haveprovided evidence for the role of ROS in the initiation of PC(Gaspar et al., 2007; Kis et al., 2003). ROS can act as secondmessengers and activate ROS-sensitive kinases (Otani, 2004).We and others have shown that NS1619 induces mitochon-drial ROS generation that is required for the initiation of PC(Gaspar et al., 2008a; Heinen et al., 2007; Stowe et al., 2006). Inthe present study, we were able to block the immediateprotective effect of NS1619 using ROS scavengers during theinitiation of PC. The fact that catalase alone did not but thecombination of a SOD mimetic and catalase did antagonizethe protection suggests a major role for superoxide in theinitiation of immediate PC. In contrast, Kulawiak et al. (2008)recently demonstrated that the direct effect of BKCa channelactivators in rat brainmitochondria is a potassium-dependentdecrease in ROS formation. However, the dose of NS1619applied in that study (3 μM) was well below that used in PCstudies. Thus, the relevance of putative mitochondrial BKCa

channels in the initiation of PC was not clarified. Finally,NS1619 is known for its ability to inhibit the mitochondrialelectron transport chain at Complex I (Debska et al., 2003;Kicinska and Szewczyk, 2004). The latter effect explains thefinding of increased ROS production and the subsequent chainof events leading to cytoprotection.

To identify the final effectors of NS1619-induced immedi-ate PC, we examined the Ca2+ and ROS response upon

glutamate stimulation. Glutamate exposure results in mas-sive calcium influx in cortical neurons with a consequentmitochondrial calcium accumulation, increased consump-tion and decreased generation of ATP, and increasedproduction of ROS (Nicholls, 2004). We found that incubationwith NS1619 significantly reduced Ca2+ influx as comparedwith untreated controls whereas the elimination phase of thecurves seemed to be unaffected. Because NS1619 wasthoroughly washed out in all experiments, a direct effect ofthe compound on glutamate receptors is not likely. Addi-tionally, to our knowledge, such an effect has not beendescribed. Earlier we demonstrated that glutamate-inducedneuronal cell death is almost exclusively dependent onNMDA receptor activation (Gaspar et al., 2008a). Therefore,we tested whether NS1619-treatment would down-regulateNMDA receptors but found that the compound did not affectthe expression of the NR1 subunit. NMDA receptors areknown to be sensitive to the oxidizing effect of free radicals(Aizenman et al., 1990). Thus, ROS produced in response toNS1619 may represent the link between the direct effect ofthe compound on mitochondria (i.e. inhibition of the electrontransport chain) and the subsequent indirect effect onglutamate receptors. The same authors reported similaralterations in NMDA receptor properties after chemicalpreconditioning with KCN (Aizenman et al., 2000). Thishypothesis of oxidative modulation is supported by theevidence that the elimination of ROS during NS1619 treat-ment abolished the protection. It has to be noted that, whilePI3K, PKC, and ERKs are not part of the cascade, theinvolvement of a supposedly ROS-sensitive kinase otherthan the ones tested in this study in the signaling cannotbe excluded.

Decreased ROS availability in response to potentially lethalstimuli is a major component of PC-induced cytoprotection.Similar to our previous results using different compounds andPC protocols (Gaspar et al., 2007; Kis et al., 2003, 2004), ourpresent study showed that immediate PC with NS1619resulted in a dose-dependent decrease in ROS levels uponexposure to glutamate. In neurons, the ROS response uponNMDA receptor stimulation is dependent on Ca2+ influx(Kahlert et al., 2005) which provides a direct link betweenreduced Ca2+ influx and the decreased ROS availabilityobserved in the present study and explains the reducedvulnerability of our neuronal cultures to glutamate afterimmediate PC by NS1619. In addition to decreased production,we also found that immediate PC by NS1619 mildly increasedSOD activity in the neurons. Others have also shown asuperoxide-induced rapid up-regulation of SOD in acutecardiac PC by the anesthetic isoflurane (Chen et al., 2008).

During NMDA receptor activation, mitochondria accumu-late calcium and, consequently, use more ATP to expelprotons. Excessive glutamate receptor stimulation ultimatelycauses mitochondrial calcium overload resulting in therupture of the mitochondrial inner membrane and a failureto generate ATP (for a review see Nicholls, 2004). Therefore, areduced calcium response in NS1619-treated neurons isexpected to result in better maintenance of mitochondrialintegrity and function. Indeed, we found that whereas short-term treatment with the compound did not influence cellularATP content, NS1619-treated neurons could better preserve

204 B R A I N R E S E A R C H 1 2 8 5 ( 2 0 0 9 ) 1 9 6 – 2 0 7

ATP during a glutamate challenge. Besides being the conse-quence of reduced calcium response to glutamate exposure,higher ATP levels may also allow a quicker restoration ofnormal cellular function after glutamate excitotoxicity.

In summary, our present study is the first demonstratingimmediate neuronal PC using NS1619. Similar to our previousstudy reporting delayed PC, the immediate PC effect is initiatedby ROS generation and is independent of BKCa channelactivation.While delayed PCbyNS1619 required PI3K activationwitha subsequent inhibitionofapoptosisandprotectionagainsthydrogen peroxide toxicity, immediate PC perhaps results fromdirect or indirectmodulation ofNMDA receptors by ROS and up-regulation of SOD activity followed by decreased Ca2+ influx anda reduction in oxidative stress upon glutamate exposure.

4. Experimental procedures

4.1. Materials

Cell culture plastics were purchased from Becton-Dickinson(San Jose, CA, USA). Dulbecco's modified Eagle medium(DMEM), Neurobasal medium, B27 Supplement, 2-mercap-toethanol, and horse serum were obtained from Gibco BRL(Grand Island, NY, USA). Percoll was purchased fromAmersham Biosciences (Uppsala, Sweden), dispase I fromRoche (Mannheim, Germany), and M40401 from MetaphorePharmaceuticals (St. Louis, MO, USA). NS1619, Pax, Wrt, andPDwerepurchased fromSigma (St. Louis,MO,USA), and IbTxfrom California Peptide Research (Napa, CA, USA). CellTiter96 AQueous One Solution Assay and CellTiter-Glo Lumines-centAssaywereprocured fromPromega (Madison,WI,USA),and the SOD Assay Kit from Fluka (Buchs, Switzerland). HEt,Fluo-4AM, Pluronic F-127, and tetramethylrhodamine ethylester (TMRE) were obtained from Molecular Probes (Eugene,OR, USA). The Bio-Rad DC Protein Assay was procured fromBio-Rad (Hercules, CA, USA), and the “Cytotoxicity DetectionKit (LDH)” fromRocheDiagnostics (Indianapolis, IN,USA).Allother chemicalswere fromSigma. Antibodieswere obtainedfrom the following sources: anti-glial fibrillary acidic proteinantibody from Chemicon (Temecula, CA, USA), anti-micro-tubule associated protein-2 and anti-NMDA-receptor sub-unit 1 (NR1) antibodies from Becton-Dickinson, anti-MAPKantibody from Sigma, anti-active MAPK from Promega, andanti-rabbit IgG and anti-mouse IgG from Jackson Immuno-Research (West Grove, PA, USA).

4.2. Primary rat cortical neuronal culture

Timed pregnant Sprague–Dawley rats were obtained fromHarlan (Indianapolis, IN, USA) and were maintained andused in compliance with the principles set forth by theAnimal Care and Use Committee ofWake Forest UniversityHealth Sciences. Primary rat cortical neurons were isolatedfrom E18 Sprague–Dawley fetuses, as described elsewhere(Gaspar et al., 2006). Briefly, the cellswereplatedat adensityof 2×105 cells/cm2 onto poly-D-lysine coated glass cover-slips for confocal microscopic analysis and 106 cells/cm2

were placed onto poly-D-lysine coated plates or dishes forthe other experiments in platingmedium consisting of 60%

DMEM, 20%Ham's F-12 NutrientMixture, 20% horse serum,and L-glutamine (0.5mM). After cell attachment, the platingmedium was replaced with regular cell culture medium[feeding medium (FM)] consisting of Neurobasal mediumsupplemented with B27 (2%), L-glutamine (0.5 mM), 2-mercaptoethanol (55 μM), and KCl (25 mM). Positiveimmunostaining for microtubule-associated protein-2 andnegative immunostaining for glial fibrillary acidic proteinverified that the cultures were composed ofmore than 98%of neurons on day 7 in vitro. Experiments were carried outon 7–9-day old cultures, during which period neuronsexpressed NMDA, α-amino-3-hydroxy-5-methylisoxazole-4-propionate, and kainate receptors andwere vulnerable toglucose deprivation (Mattson et al., 1991, 1993). Formediumchanges, fresh FM was used in all experiments.

4.3. Treatment with NS1619

To induce immediate PC, 9-day old neuronal cultures weretreatedwith increasing doses ofNS1619 (10, 25, 50, 100, 150,and 200 μM) for 60 min in FM. In other experiments, thecellswere treated for 60minwithNS1619 (150 μM) andwiththe BKCa channel antagonists IbTx (5 μM) or Pax (20 μM), thePI3K inhibitorWrt (300 nM), the MAPK inhibitor PD (50 μM),the PKC blocker Chel (5 μM), or with catalase (100 U/ml) andM40401 (50 μM). After each treatment, the plates werewashed twice with phosphate buffered saline (PBS) andthen the cells were kept in FM.

4.4. Combined oxygen and glucose deprivation

Neurons in 96-well plateswere exposed to OGD for 180minat 37 °Cusing aprotocol describedpreviously (Goldberg andChoi, 1993; Kis et al., 2003). Briefly, after the treatment withNS1619, the cells were rinsed, the medium was replacedwith glucose-free Earle's balanced salt solution (EBSS) andthe cultures were placed in a ShelLab Bactron AnaerobicChamber (SheldonManufacturing Inc., Cornelius, OR, USA)filled with a humidified anaerobic gas mixture (5% CO2, 5%H2and90%N2) at 37 °C.Oxygen levelwas kept below0.1% inthe chamber. Control cell cultures were incubated inglucose-containing (5.5 mM) EBSS in a regular 5% CO2 cellculture incubator for 180 min. OGD was terminated byreoxygenation and replacing the glucose-free EBSS withFM. Thereafter the cultures weremaintained in the regular5% CO2 incubator within normoxic conditions.

4.5. Glutamate excitotoxicity

After 60-min treatment with NS1619, cell cultures in 96-wellplates were washed and then exposed to glutamate (200 μM)for60min inFMat37°C in the5%CO2 incubator.Afterwardthecells were rinsed and returned to the 5% CO2 incubator in FM.

4.6. Exogenous hydrogen peroxide toxicity

Neuronal cultures in 96-well plates were rinsed and thecells were incubated in Neurobasal medium containingH2O2 (50 μM) and glucose oxidase (1 U/l) at 37°C in the 5%CO2 incubator for 1 h. The continuous generation of H2O2

205B R A I N R E S E A R C H 1 2 8 5 ( 2 0 0 9 ) 1 9 6 – 2 0 7

by glucose oxidase was shown to compensate for theconsumption of the initial dose of H2O2 by neurons(Antunes and Cadenas, 2001; Vauzour et al., 2007). Afterthe challenge, the cultures were washed, FM was restored,and the cells were returned to the 5% CO2 incubator.

4.7. Quantification of neuronal survival

The viability of neuronal cultures was determined 24 h aftertheneurotoxic insults. Theplateswerewashed twicewithPBSwhichwasthenreplacedwithNeurobasalmediumcontaining1%TritonX to release lactate dehydrogenase (LDH) from livingcells. LDH levels were then determined using a commerciallyavailable LDH detection kit according to the manufacturer'sinstructions. Absorbance at 492 nm was measured with amicroplate reader (FLUOstar OPTIMA, BMG Labtech GmbH,Offenburg, Germany). Comparisons were always made in thesame manner, between sister cultures exposed to theneurotoxic stimulus on the same day, and cell viability wasexpressedasapercentageof thecorrespondingcontrol culture(untreated, and not exposed to the lethal insult) as follows: %viability=(absorbanceTREATED−absorbanceBACKGROUND)×100/(absorbanceCONTROL−absorbanceBACKGROUND).

4.8. Fluorescent measurements

The changes of mitochondrial membrane potential weredetermined using TMRE. Neurons in FM were loaded withTMRE (0.5 μM) and treated with NS1619 (25–200 μM). Asecond set of neurons was pretreated with IbTx (5 μM) orPax (20 μM) for 5min before loading andNS1619 application(150 μM). After 20min of incubation in the dark at 37°C thatallowed the redistribution of the dye in proportion to themembranepotential ofmitochondria, thedrugsand thedyewere washed out and TMRE fluorescence was measured ineach well using the same microplate reader that was usedfor viability measurements (λex=510 nm, λem=590 nm).Measurements were performed in PBS containing 1 mg/mlglucose at 37 °C. Data were expressed as a percentage of theintensity of the untreated control culture.

ROS generation was assessed in black-walled 96-wellplateswithHEt using the samemicroplate reader used in theTMRE measurements. Neurons in PBS containing 1 mg/mlglucose were loaded with HEt (5 μM) and were treated withNS1619 (10–150 μM) at the same time, in the same buffer. Inanother experiment, neurons were pretreated with eitherIbTx (5μM)or Pax (20μM) for 5min thenwere loadedwithHEt(5 μM) and treated with NS1619 (150 μM). ROS determinationwas started 1min after NS1619 application. HEt-fluorescence(λex=510 nm, λem=590 nm) in each well was measured at37 °C every minute for 30 min. Data were expressed aspercent of the starting intensity of theuntreated control or asa change in relative fluorescent intensity/minute (ΔRFU/min).

Changes of intracellular free calcium ([Ca2+]i) weremonitored using the Ca-indicator dye Fluo-4AM in glu-cose-containing PBS (1 mg/ml). Neuronal cultures wereloaded with 2 μM Fluo-4AM and 1 μM Pluronic F-127 in PBSin the dark for 60 min at room temperature (22 °C). In asecond group, cells were treated with NS1619 (150 μM) orvehicle during loading with Fluo-4AM. After loading,

neurons were rinsed three times with PBS and confocalimages of cellular Fluo-4AM fluorescence (λex=488 nm,λem=520 nm) were acquired using a laser scanning micro-scope (LSM 510; Zeiss, Jena, Germany) with a 63× waterimmersion objective (Zeiss). Glutamate (200 μM)was addedto themediumafter 1min. Fluorescent images of randomlyselected fields were recorded every 20 s for 10 min and theaverage pixel intensity of individual cell bodies wasdetermined using the software supplied by the manufac-turer (LSM 510, Zeiss). Datawere expressed as a percentageof the starting intensity of the untreated control culture.

4.9. Western blotting for extracellular signal-regulatedkinases and for the NR1 subunit of the NMDAreceptor complex

Cultured cells were washed twice in ice-cold PBS and wereharvested by scraping in ice-cold Nonidet P40 lysis buffersupplemented with a protease inhibitor cocktail and a phos-phatase inhibitor cocktail (both fromSigma).Equalamountsofprotein for each sample were separated by 4–20% sodiumdodecyl sulfate-polyacrylamide gel electrophoresis and trans-ferredontoapolyvinylidinedifluoride sheet (PolyscreenPVDF;Perkin Elmer Life Sciences, Boston, MA, USA). Membraneswere incubated inablockingbuffer (Tris-bufferedsaline, 0.05%Tween 20, and 1% bovine serum albumin) for 1 h at roomtemperature after which the blots were incubated withpolyclonal anti-MAPK (1:10000), polyclonal anti-active MAPK(1:5000), and monoclonal anti-NR1 (1:1000) antibodies over-night at 4 °C. The membranes were then washed three timesin Tris-buffered salinewith 0.05%Tween 20 and incubated for1 h in the blocking buffer with anti-rabbit IgG (1:100000) oranti-mouse IgG (1:100000) conjugated to horseradish peroxi-dase. The final reaction products were visualized usingenhanced chemiluminescence (SuperSignalWest Pico; Pierce,Rockford, IL, USA) and recorded on X-ray film. Films weredigitalized using a FOTO/Analyst® Investigator-PC ElectronicDocumentation System with the provided Image v5.00 soft-ware (FOTODYNE Inc., Hartland, WI, USA). Densitometry ofdigitalizedblotswasperformedwith the ImageJ1.30vsoftware(http://rsb.info.nih.gov/ij; National Institutes of Health, USA).

4.10. SOD activity

SOD activity was measured using the tetrazolium-basedSOD Assay Kit as directed by the manufacturer. Briefly,neurons on 96-well plates were incubated with differentdosesofNS1619 (50, 100, and150μM) for 1h. Then theplateswere rinsed with PBS, and enzyme activity was measuredwith the same microplate reader that was used for othermeasurements (λabs=460 nm). An SOD activity standardcurve was generated and then data were expressed as apercentage of the activity of the nontreated control culture.

4.11. ATP assay

The ATP level of neurons wasmeasured with the glowtypeCellTiter-Glo Luminescent Assay, as directed by the man-ufacturer. Cortical neurons cultured in opaque-walled 96-well plates were equilibrated to room temperature (21 °C)

206 B R A I N R E S E A R C H 1 2 8 5 ( 2 0 0 9 ) 1 9 6 – 2 0 7

for 30 min. CellTiter-Glo was added to each well and theplates were incubated at room temperature for 10 min tostabilize the luminescent signal, whichwas thenmeasuredwith the same microplate reader that was used for othermeasurements. An ATP standard curve was generated foreach measurement to calculate the ATP contents of cells.

4.12. Statistical analysis

Statistical analyses were performed with SigmaStat (SPSS,Chicago, IL, USA). Data are presented as means±SEM.Differences between groups were assessed by one wayANOVA or two way repeated measures ANOVA, whereappropriate, followed by Tukey comparison tests. A valueof p<0.05 was considered to be statistically significant.

Acknowledgments

The authors gratefully thankNancyBusija,M.A., for editing themanuscript. This work was supported by the National Insti-tutes of Health Grants (HL-030260, HL-065380, and HL-077731),and by the Hungarian Science Research Fund (OTKA K63401,K68976, and IN69967).

R E F E R E N C E S

Aizenman, E., Hartnett, K.A., Reynolds, I.J., 1990. Oxygen freeradicals regulate NMDA receptor function via a redoxmodulatory site. Neuron 5, 841–846.

Aizenman, E., Sinor, J.D., Brimecombe, J.C., Herin, G.A., 2000.Alterations of N-methyl-D-aspartate receptor properties afterchemical ischemia. J. Pharmacol. Exp. Ther. 295, 572–577.

Antunes, F., Cadenas, E., 2001. Cellular titration of apoptosis withsteady state concentrations of H(2)O(2): submicromolar levelsof H(2)O(2) induce apoptosis through Fenton chemistryindependent of the cellular thiol state. Free Radic. Biol. Med. 30,1008–1018.

Busija, D.W., Lacza, Z., Rajapakse, N., Shimizu, K., Kis, B., Bari, F.,Domoki, F., Horiguchi, T., 2004. Targeting mitochondrialATP-sensitive potassium channels—a novel approach toneuroprotection. Brain Res. Brain Res. Rev. 46, 282–294.

Cao, C.M., Xia, Q., Gao, Q., Chen, M., Wong, T.M., 2005.Calcium-activated potassium channel triggerscardioprotection of ischemic preconditioning. J. Pharmacol.Exp. Ther. 312, 644–650.

Chen, C.H., Liu, K., Chan, J.Y., 2008. Anesthetic preconditioningconfers acute cardioprotection via up-regulation ofmanganesesuperoxide dismutase and preservation of mitochondrialrespiratory enzyme activity. Shock 29, 300–308.

Dahl, N.A., Balfour, W.M., 1964. Prolonged anoxic survival due toanoxia pre-exposure: brain ATP, lactate, and pyruvate. Am. J.Physiol. 207, 452–456.

Debska, G., Kicinska, A., Dobrucki, J., Dworakowska, B.,Nurowska, E., Skalska, J., Dolowy, K., Szewczyk, A., 2003.Large-conductance K+ channel openers NS1619 and NS004 asinhibitors of mitochondrial function in glioma cells.Biochem. Pharmacol. 65, 1827–1834.

Domoki, F., Perciaccante, J.V., Veltkamp, R., Bari, F., Busija, D.W.,1999. Mitochondrial potassium channel opener diazoxidepreserves neuronal-vascular function after cerebral ischemiain newborn pigs. Stroke 30, 2713–2718 discussion 2718-9.

Douglas, R.M., Lai, J.C., Bian, S., Cummins, L., Moczydlowski, E.,Haddad, G.G., 2006. The calcium-sensitive large-conductance

potassium channel (BK/MAXI K) is present in the innermitochondrial membrane of rat brain. Neuroscience 139,1249–1261.

Edwards, G., Niederste-Hollenberg, A., Schneider, J., Noack, T.,Weston, A.H., 1994. Ion channel modulation by NS 1619, theputative BKCa channel opener, in vascular smooth muscle. Br.J. Pharmacol. 113, 1538–1547.

Garlid, K.D., Paucek, P., Yarov-Yarovoy, V., Murray, H.N.,Darbenzio, R.B., D'Alonzo, A.J., Lodge, N.J., Smith, M.A., Grover,G.J., 1997. Cardioprotective effect of diazoxide and itsinteraction with mitochondrial ATP-sensitive K+ channels.Possible mechanism of cardioprotection. Circ. Res. 81,1072–1082.

Gaspar, T., Kis, B., Snipes, J.A., Lenzser, G., Mayanagi, K., Bari, F.,Busija, D.W., 2006. Transient glucose and amino aciddeprivation induces delayed preconditioning in cultured ratcortical neurons. J. Neurochem. 98, 555–565.

Gaspar, T., Kis, B., Snipes, J.A., Lenzser, G., Mayanagi, K., Bari, F.,Busija, D.W., 2007. Neuronal preconditioning with theantianginal drug, bepridil. J. Neurochem. 102, 595–608.

Gaspar, T., Katakam, P., Snipes, J.A., Kis, B., Domoki, F., Bari, F.,Busija, D.W., 2008a. Delayed neuronal preconditioning byNS1619 is independent of calcium activated potassiumchannels. J. Neurochem. 105, 1115–1128.

Gaspar, T., Snipes, J.A., Busija, A.R., Kis, B., Domoki, F., Bari, F.,Busija, D.W., 2008b. ROS-independent preconditioning inneurons via activation of mitoK(ATP) channels by BMS-191095.J. Cereb. Blood Flow Metab. 28, 1090–1103.

Goldberg, M.P., Choi, D.W., 1993. Combined oxygen and glucosedeprivation in cortical cell culture: calcium-dependent andcalcium-independent mechanisms of neuronal injury.J. Neurosci. 13, 3510–3524.

Heinen, A., Camara, A.K., Aldakkak, M., Rhodes, S.S., Riess, M.L.,Stowe, D.F., 2007. Mitochondrial Ca2+-induced K+ influxincreases respiration and enhances ROS production whilemaintaining membrane potential. Am. J. Physiol. Cell. Physiol.292, C148–C156.

Holland, M., Langton, P.D., Standen, N.B., Boyle, J.P., 1996. Effects ofthe BKCa channel activator, NS1619, on rat cerebral arterysmooth muscle. Br. J. Pharmacol. 117, 119–129.

Inoue, I., Nagase, H., Kishi, K., Higuti, T., 1991. ATP-sensitive K+channel in the mitochondrial inner membrane. Nature 352,244–247.

Janoff, A., 1964. Alterations in lysosomes (intracellular enzymes)during shock; effects of preconditioning (tolerance) andprotective drugs. Int. Anesthesiol. Clin. 2, 251–269.

Kahlert, S., Zundorf, G., Reiser, G., 2005. Glutamate-mediatedinflux of extracellular Ca2+ is coupled with reactive oxygenspecies generation in cultured hippocampal neurons but not inastrocytes. J. Neurosci. Res. 79, 262–271.

Kicinska, A., Szewczyk, A., 2004. Large-conductance potassiumcation channel opener NS1619 inhibits cardiac mitochondriarespiratory chain. Toxicol. Mech. Methods 14, 59–61.

Kis, B., Rajapakse, N.C., Snipes, J.A., Nagy, K., Horiguchi, T.,Busija, D.W., 2003. Diazoxide induces delayedpre-conditioning in cultured rat cortical neurons.J. Neurochem. 87, 969–980.

Kis, B., Nagy, K., Snipes, J.A., Rajapakse, N.C., Horiguchi, T., Grover,G.J., Busija, D.W., 2004. The mitochondrial K(ATP) channelopener BMS-191095 induces neuronal preconditioning.Neuroreport 15, 345–349.

Kitagawa, K., Matsumoto, M., Tagaya, M., Hata, R., Ueda, H.,Niinobe, M., Handa, N., Fukunaga, R., Kimura, K., Mikoshiba, K.,Kamada, T., 1990. ‘Ischemic tolerance’ phenomenon found inthe brain. Brain Res. 528, 21–24.

Korper, S., Nolte, F., Rojewski, M.T., Thiel, E., Schrezenmeier, H.,2003. The K+ channel openers diazoxide and NS1619 inducedepolarization of mitochondria and have differential effects oncell Ca2+ in CD34+ cell line KG-1a. Exp. Hematol. 31, 815–823.

207B R A I N R E S E A R C H 1 2 8 5 ( 2 0 0 9 ) 1 9 6 – 2 0 7

Kulawiak, B., Kudin, A.P., Szewczyk, A., Kunz, W.S., 2008. BKchannel openers inhibit ROS production of isolated rat brainmitochondria. Exp. Neurol. 212, 543–547.

Kuzuya, T., Hoshida, S., Yamashita, N., Fuji, H., Oe, H., Hori, M.,Kamada, T., Tada, M., 1993. Delayed effects of sublethalischemia on the acquisition of tolerance to ischemia. Circ. Res.72, 1293–1299.

Marber, M.S., Latchman, D.S., Walker, J.M., Yellon, D.M., 1993.Cardiac stress protein elevation 24 hours after brief ischemia orheat stress is associated with resistance to myocardialinfarction. Circulation 88, 1264–1272.

Mattson, M.P., Wang, H., Michaelis, E.K., 1991. Developmentalexpression, compartmentalization, and possible role inexcitotoxicity of a putative NMDA receptor protein in culturedhippocampal neurons. Brain Res. 565, 94–108.

Mattson, M.P., Zhang, Y., Bose, S., 1993. Growth factors preventmitochondrial dysfunction, loss of calcium homeostasis, andcell injury, but not ATP depletion in hippocampal neuronsdeprived of glucose. Exp. Neurol. 121, 1–13.

Mayanagi, K., Gaspar, T., Katakam, P.V., Kis, B., Busija, D.W., 2007.Themitochondrial K(ATP) channel opener BMS-191095 reducesneuronal damage after transient focal cerebral ischemia inrats. J. Cereb. Blood Flow Metab. 27, 348–355.

Murry, C.E., Jennings, R.B., Reimer, K.A., 1986. Preconditioningwithischemia: a delay of lethal cell injury in ischemic myocardium.Circulation 74, 1124–1136.

Nicholls, D.G., 2004. Mitochondrial dysfunction and glutamateexcitotoxicity studied in primary neuronal cultures. Curr. Mol.Med. 4, 149–177.

Noble, R.L., 1943. The development of resistance by rats andguinea pigs to amounts of trauma usually fatal. Am. J. Physiol.138, 346–351.

Otani, H., 2004. Reactive oxygen species as mediators of signaltransduction in ischemic preconditioning. Antioxid. Redox.Signal. 6, 449–469.

Park,W.S., Kang, S.H., Son, Y.K., Kim, N., Ko, J.H., Kim, H.K., Ko, E.A.,Kim, C.D., Han, J., 2007. The mitochondrial Ca2+-activated K+channel activator, NS 1619 inhibits L-type Ca2+ channels in ratventricular myocytes. Biochem. Biophys. Res. Commun. 362,31–36.

Sato, T., Saito, T., Saegusa, N., Nakaya, H., 2005. MitochondrialCa2+-activated K+ channels in cardiacmyocytes: a mechanismof the cardioprotective effect andmodulation by protein kinaseA. Circulation 111, 198–203.

Schurr, A., Reid, K.H., Tseng, M.T., West, C., Rigor, B.M., 1986.Adaptation of adult brain tissue to anoxia and hypoxia in vitro.Brain Res. 374, 244–248.

Shintani, Y., Node, K., Asanuma, H., Sanada, S., Takashima, S.,Asano, Y., Liao, Y., Fujita, M., Hirata, A., Shinozaki, Y.,Fukushima, T., Nagamachi, Y., Okuda, H., Kim, J., Tomoike, H.,Hori, M., Kitakaze, M., 2004. Opening of Ca2+-activated K+channels is involved in ischemic preconditioning in caninehearts. J. Mol. Cell. Cardiol. 37, 1213–1218.

Siemen, D., Loupatatzis, C., Borecky, J., Gulbins, E., Lang, F., 1999.Ca2+-activated K channel of the BK-type in the innermitochondrial membrane of a human glioma cell line.Biochem. Biophys. Res. Commun. 257, 549–554.

Skalska, J., Piwonska, M., Wyroba, E., Surmacz, L., Wieczorek, R.,Koszela-Piotrowska, I., Zielinska, J., Bednarczyk, P., Dolowy, K.,Wilczynski, G.M., Szewczyk, A., Kunz, W.S., 2008. A novelpotassium channel in skeletal muscle mitochondria. Biochim.Biophys. Acta 1777, 651–659.

Stowe, D.F., Aldakkak, M., Camara, A.K., Riess, M.L., Heinen, A.,Varadarajan, S.G., Jiang, M.T., 2006. Cardiac mitochondrialpreconditioning by Big Ca2+-sensitive K+ channel openingrequires superoxide radical generation. Am. J. Physiol. HeartCirc. Physiol. 290, H434–H440.

Vauzour, D., Vafeiadou, K., Rice-Evans, C., Williams, R.J.,Spencer, J.P., 2007. Activation of pro-survival Akt and ERK1/2signalling pathways underlie the anti-apoptotic effects offlavanones in cortical neurons. J. Neurochem. 103,1355–1367.

Veltkamp, R., Domoki, F., Bari, F., Busija, D.W., 1998. Potassiumchannel activators protect the N-methyl-D-aspartate-inducedcerebral vascular dilation after combined hypoxia andischemia in piglets. Stroke 29, 837–842 discussion 842-3.

Wang, X., Yin, C., Xi, L., Kukreja, R.C., 2004. Opening ofCa2+-activated K+ channels triggers early and delayedpreconditioning against I/R injury independent of NOS inmice.Am. J. Physiol. Heart Circ. Physiol. 287, H2070–H2077.

Xu, W., Liu, Y., Wang, S., McDonald, T., Van Eyk, J.E., Sidor, A.,O'Rourke, B., 2002. Cytoprotective role of Ca2+-activated K+channels in the cardiac inner mitochondrial membrane.Science 298, 1029–1033.

Yamamura, H., Ohi, Y., Muraki, K., Watanabe, M., Imaizumi, Y.,2001. BK channel activation by NS-1619 is partiallymediated byintracellular Ca2+ release in smooth muscle cells of porcinecoronary artery. Br. J. Pharmacol. 132, 828–834.