Activation of Adenosine Triphosphate-Sensitive Potassium Channels ConfersProtection Against Rotenone-Induced CellDeath: Therapeutic Implications forParkinson’s Disease

Kwok-Keung Tai* and Daniel D. TruongThe Parkinson’s and Movement Disorder Institute, Long Beach Memorial Medical Center,Long Beach, California

It is anticipated that further understanding of the protectivemechanism induced by ischemic preconditioning will im-prove prognosis for patients of ischemic injury. It is notknown whether preconditioning exerts beneficial actions inneurodegenerative diseases, in which ischemic injury playsa causative role. Here we show that transient activation ofATP-sensitive potassium channels, a trigger in ischemicpreconditioning signaling, confers protection in PC12 cellsand SH-SY5Y cells against neurotoxic effect of rotenoneand MPTP, mitochondrial complex I inhibitors that havebeen implicated in the pathogenesis of Parkinson’s dis-ease. The degree of protection is in proportion to the boutsof exposure to an ATP-sensitive potassium channel opener,a feature reminiscent of ischemic tolerance in vivo. Protec-tion is sensitive to a protein synthesis inhibitor, indicatingthe involvement of de novo protein synthesis in the protec-tive processes. Pretreatment of PC12 cells with precondi-tioning stimuli FeSO4 or xanthine/xanthine oxidase alsoconfers protection against rotenone-induced cell death.Our results demonstrate for the first time the protective roleof ATP-sensitive potassium channels in a dopaminergicneuronal cell line against rotenone-induced neurotox-icity and conceptually support the view that ischemicpreconditioning-derived therapeutic strategies may havepotential and feasibility in therapy for Parkinson’s disease.© 2002 Wiley-Liss, Inc.

Key words: ischemic preconditioning; ATP-sensitive po-tassium channels; rotenone; pheochromocytoma cells;Parkinson’s disease

Ischemic preconditioning (IPC) is a well-knownphenomenon in which episodes of mild cellular ischemicstress induce protection against subsequent prolonged andlethal ischemic insults. IPC was first reported for themyocardium (Murry et al., 1986) and was subsequentlydemonstrated in a variety of organ systems, including thebrain (Kitagawa et al., 1990; Ishida et al., 1997). Thecascade of events underlying cardiac IPC is being eluci-

dated (Rubino and Yellen, 2000; Cohen et al., 2000).Although the mechanism underlying IPC in the brain maynot be identical to those of cardiac IPC (Tauskela et al.,1999), it is generally acknowledged that ATP-sensitivepotassium (K-ATP) channels (Heurteaux et al., 1995;Reshef et al., 1998; Domoki et al., 1999; Blondeau et al.,2000) and reactive oxygen species (ROS; Gidday et al.,1999; Gonazalez-Zulueta et al., 2000; Ravati et al., 2000)play an important role in triggering IPC-induced protec-tion in both the heart and the brain. IPC-induced protec-tion against ischemic injury can be simulated in vivo byactivating K-ATP channels and can be attenuated by ad-ministration of K-ATP channel blockers (Perez-Pinzonand Born, 1999). It is anticipated that further understand-ing of the molecular mechanism underlying IPC will leadto the development of novel therapeutic strategies for thetreatment of ischemic injury. However, whether thesestrategies can also be applied to therapy for neurodegen-erative diseases, in which ischemic injury plays a causativerole (Chorsky et al., 2001), is not known. Thus, thedemonstration that IPC confers protection in such diseasemodels is the first step toward utilizing this natural defensemechanism of the body to combat neurodegenerativeprocesses.

Inhibition of mitochondrial complex I activity hasbeen implicated in the pathogenesis of Parkinson’s disease(PD; Jenner, 2001) and has been widely used as modelsystem with which to elicit neurochemical alterations thatare associated with PD. In this study, we sought to deter-mine whether transient activation of K-ATP channels and

Contract grant sponsor: Parkinson’s and Movement Disorder Institute.

*Correspondence to: Kwok-Keung Tai, The Parkinson’s and MovementDisorder Institute, 2625 Pasadena Ave., Long Beach, CA 90806.E-mail: kktai@yahoo.com

Received 22 January 2002; Revised 15 April 2002; Accepted 19 April 2002

Published online 27 June 2002 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jnr.10309

Journal of Neuroscience Research 69:559–566 (2002)

© 2002 Wiley-Liss, Inc.

other preconditioning stimuli can precondition rat pheo-chromocytoma (PC12) cells against the neurotoxic effectof rotenone, a mitochondrial complex I inhibitor thatrecapitulates behavioral (Betarbet et al., 2000) and neuro-pathological (Ferrante et al., 1997; Betarbet et al., 2000)hallmarks of PD, including acceleration of �-synucleinformation (Uversky et al., 2001) in rodents followingsystemic administration.

MATERIALS AND METHODS

Cell Culture

PC12 cells were cultured in high-glucose Dulbecco’smodified Eagle’s medium (DMEM; BioWhittaker, Bethesda,MD) with pyruvate and supplemented with 5% (v/v) fetalbovine serum, 10% (v/v) horse serum, 100 g/ml penicillin, and50 U/ml streptomycin in a 75 ml vented culture flask (Corning,Acton, MA). SH-SY5Y neuroblastoma cells were cultured withthe same culture media as for the PC12 cells, and rho 0 SH-SY5Ycells, which are depleted of mitochondrial DNA and lack afunctional mitochondrial electron transport chain (King andAttardi, 1989), were cultured with the same culture media butsupplemented with uridine (1 mg/ml). Cells were maintained ina humidified incubator at 37°C and 5% CO2. Medium waschanged every 3 days. By 6–8 days of incubation in the flask,cells had reached �60–70% confluence. Cells were harvestedand seeded onto a 10-mm-diameter 96-well plate at a celldensity of �50,000 cells/ml. After the addition of rotenone or1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) to thecell cultures, medium was exchanged every 3 days.

Preconditioning and Rotenone-Induced Cell Death

Preconditioning was induced as follows: PC12 cells inDMEM were exposed to pinacidil (10 �M) or FeSO4 (100 �M)or xanthine/xanthine oxidase (X/XO; 10 �M/0.5 mU ml–1)for 15 min at 37°C. After 15 min of exposure to any one of thesestimuli, media were aspirated and cells were rinsed with DMEMonce and then switched back to DMEM. Cells were allowed torecover for 24 hr before evaluation of the protective effectagainst rotenone toxicity. An aliquot of cells was taken at thetime specified and was subjected to cell viability analysis asdescribed below.

To quantify the protective effect of multiple exposures ofthe cells to pinacidil against the toxic effect of rotenone, cellswere exposed to10 �M pinacidil once every 24 hr for 1–3 days.Cell viability was assessed using trypan blue exclusion assays onday 7 after incubation with 20 �M rotenone.

The procedures for determining whether the K-ATPchannel-mediated protective effect requires expression of newlysynthesized protein were as follows: PC12 cells were incubatedwith 50 �M cycloheximide for 4 hr. One hour after exposure tocycloheximide, cells were preconditioned with 10 �M pinacidilfor 15 min, followed by rotenone treatment 24 hr later. Expo-sure of PC12 cells to cycloheximide (50 �M) for 4 hr had noeffect on cell viability as indicated by trypan blue exclusionassays.

To evaluate the protective effect of FeSO4 and X/XOpreconditioning, PC12 cells were exposed to 100 �M FeSO4 or10 �M/0.5 mU ml–1 X/XO for 15 min. Cells were then rinsed

with DMEM. Twenty-four hours after the preconditioningstimulus was removed, PC12 cells were incubated with 20 �Mrotenone for 7 days, and cell viability was then determined usingtrypan blue exclusion assays.

Cell Viability

Cell viability was evaluated on the basis of membraneleakage using trypan blue exclusion assays similar to those pre-viously described (Tai et al., 2001). In brief, after exposure torotenone for the time specified, cells were dislodged by gentlepipetting. A 20-�l aliquot of cells was then collected from eachwell of the plate and mixed with 20 �l 0.4% trypan blue dye(Sigma, St. Louis, MO). Cells were incubated for 5 min at roomtemperature. The number of trypan blue-stained cells and thetotal cell numbers were counted per square millimeter in anImproved Nebauer Hemacytometer under a microscope.Trypan blue-stained cells are considered as not viable regardlessof the degree of blue. The number of cells counted from day 1to day 7 following the addition of rotenone was between 10 and30 per square millimeter. Nine hundred sixty cells were countedfor each concentration of rotenone tested on day 7 followingtheir addition. Percentage of cell death is expressed as thenumber of trypan blue-stained cells divided by the total numberof cells counted.

Chemicals

Pinacidil, rotenone, xanthine, xanthine oxidase, FeSO4,and MPTP were purchased from Sigma; glibenclamide anddimethyl sulfoxide (DMSO) were purchased from Tocris Cook-son (St. Louis, MO); DMEM was purchased from BioWhit-taker; fetal bovine serum, horse serum, penicillin/streptomycin,and glutamine were purchased from Gibco RBL (Grand Island,NY). All other chemicals and reagents purchased from Sigma.Pinacidil, rotenone, xanthine, glibenclamide were dissolved inDMSO. The final concentration of DMSO in cell cultures didnot exceed 0.1%. DMSO at this concentration has no harmfuleffect on the cells. All other chemicals were dissolved in waterfor cell culture from Sigma.

Data Analysis

Data are presented as mean � SEM. Statistical significancefor comparisons were determined using one-way analysis ofvariance (ANOVA) followed by the Tukey’s post hoc test tocompare group means. Data were collected from four separateexperiments with eight replicate samples from each experiment.P values �0.05 were considered statistically significant.

RESULTS

Rotenone Induces Cell Death in PC12 Cells in aTime- and Concentration-Dependent Manner

To establish the time course of rotenone-inducedcell death, PC12 cells were exposed to 0.03–60 �M ofrotenone over a period of 7 days. Cell viability was as-sessed using trypan blue exclusion assays on day 1, day 2,day 3, day 4, and day 7 after rotenone was added to the cellcultures. Cell death caused by 0.03–1 �M rotenone (2–5%) was not significantly different from that of untreatedcontrol cells (2–4%) over a period of 7 days (Fig. 1).

560 Tai and Truong

The results shown in Figure 1 indicate that 0.03–60 �M rotenone has no effect on PC12 cells viability onday 1. Between 17% and 27% of the cells were dead 2 daysafter incubation with 5–60 �M rotenone. The percent-ages of cell death increased to 31%, 42%, and 43%, 3 daysafter exposure to 5 �M, 20 �M, and 60 �M rotenone,respectively. By day 7, 5 �M, 20 �M, and 60 �M rote-none killed 65%, 85%, and 88% of the cells, respectively.In a similar manner, 60 �M rotenone killed �90% ofSH-SY5Y cells. However, the same concentration of ro-tenone killed only �5% of rho 0 SH-SY5Y cells.

Activation of K-ATP Channels Confers Protectionin PC12 Cells Against Rotenone-Induced CellDeath

PC12 cells were exposed to pinacidil, a K-ATPchannels opener, for 15 min. Twenty-four hours later,both untreated and pinacidil-treated PC12 cells wereexposed to various concentrations of rotenone (0.03–60 �M) for 7 days. The percentage of rotenone-inducedcell death was assessed on day 7 after incubation withrotenone. PC12 cells exposed to 20 �M rotenone for 7days showed extensive cell death as indicated by trypanblue exclusion assays (Fig. 2A). Cell viability was signifi-cantly increased by preconditioning with pinacidil (Fig.2A), and the protective effect of preconditioning withpinacidil was almost abolished when preconditioning tookplace in the presence of glibenclamide (Fig. 2A). A 15-min

exposure to pinacidil significantly reduced cell death (P �0.01) induced by 5–60 �M rotenone 7 days after incuba-tion with rotenone (Fig. 2B). To determine whether thisbeneficial effect of preconditioning with pinacidil was theresult of activation of K-ATP channels in PC12 cells,PC12 cells were preconditioned with 10 �M pinacidil inthe absence and presence of 10 �M glibenclamide, aK-ATP channel blocker, for 15 min. The protective effectof pinacidil was attenuated by glibenclamide; cell deathincreased from 32–47% to 67–82% after exposure of thePC12 cells to 5–60 �M rotenone on day 7 after theincubation (Fig. 2B). Preconditioning of PC12 cells andSH-SY5Y cells to pinacidil also increased cell viability inresponse to MPTP toxicity; cell death was significantlydecreased from 80–85% to 47–50% (Table I). Thepinacidil-mediated protective effect was attenuated in thepresence of 10 �M glibenclamide (Table I). These resultssuggest that the observed protective effect of brief expo-sure to pinacidil is mediated by activation of K-ATPchannels.

Degree of Protection Is in Proportion to theNumber of Bouts of K-ATP Channel Activation

To evaluate whether our system simulates IPC invivo, we sought to investigate whether the protectiveeffect of preconditioning with pinacidil has characteristicsthat resemble those of IPC in vivo. One unique feature ofIPC in vivo is that the level of protection correlates withthe number of bouts of IPC (Downey et al., 1994; Saitohet al., 1995; Sandhu et al., 1997). In this series of exper-iments, PC12 cells were subjected to multiple exposures to10 �M pinacidil. The protective effects that resulted frommultiple exposures to pinacidil against the neurotoxic ef-fect of 20 �M rotenone were determined on day 7 afterincubation. Transient exposure of PC12 cells to pinacidilfor one to three times did not affect cell viability; thepercentage of cell death was similar to that of controls(2–4%; Fig. 3B). A single, transient exposure to 10 �Mpinacidil for 15 min significantly reduced the percentageof cell death caused by 20 �M rotenone from �90% to�49% (P � 0.01; Fig. 3B). Two or three exposures topinacidil further reduced the percentage of rotenone-induced cell death to �37% (P � 0.05) or �25% (P �0.01), respectively (Fig. 3B). These results demonstratethat the degree of protection is in proportion to thenumber of exposures to pinacidil and is cumulative, afeature reminiscent of IPC in vivo.

K-ATP Channel-Mediated Protective EffectRequires Expression of Newly Synthesized Protein

To determine whether K-ATP channel-mediatedprotection in PC12 cells requires de novo protein synthe-sis, PC12 cells were pretreated with 50 �M cyclohexi-mide, a protein synthesis inhibitor. It has been shown thatcycloheximide at this concentration prevents the hypoxia-induced protein kinase B phosphorylation cell survivalpathway in PC12 cells by inhibiting protein synthesis(Alvarez-Tejado et al., 2001). The percentage of cell deathin cycloheximide-treated cells is higher than those pre-

Fig. 1. Rotenone induces cell death in PC12 cells in a time- andconcentration-dependent manner. Percentage of cell death on day 1,day 2, day 3, day 4, and day 7 following exposure to 0.03–60 �Mrotenone. Each point represents the mean � SEM of four independentexperiments. *,**Significant higher percentage of cell death comparedwith groups treated with 0.03–1 �M rotenone at P � 0.05 and P �0.01, respectively.

K-ATP Channel-Mediated Protection From Cell Death 561

conditioned with pinacidil alone (Fig. 4A), suggesting thatthe protective effect from K-ATP channel activation re-quires de novo protein synthesis. In support of this notion,the protective effect of pinacidil preconditioning was sig-nificantly attenuated when the PC12 cells were treatedwith pinacidil and rotenone simultaneously (Fig. 4B).

Preconditioning With FeSO4 or X/XO ConfersProtection in PC12 Cells Against Rotenone-Induced Cell Death

In addition to K-ATP channels, ROS are also im-portant mediators in the mechanism of preconditioningprotection. It has been established that exogenously pro-duced ROS can be used as a preconditioning stimulus onneuronal cultures against glutamate-induced excitotoxicdamage as well as staurosporine-induced apoptosis(Gonzalez-Zulueta et al., 2000; Ravati et al., 2000). Pre-conditioning of PC12 cells with Fe2� (100 �M) for15 min significantly reduced (P � 0.01) percentage of cell

TABLE I. Transient Exposure to Pinacidil Confers Protectionin PC12 Cells and SH-SY5Y Cells Against MPTP Toxicity†

Treatment PC12 cells SH-SY5Y cells

Control 2 � 1 3 � 2MPTP 80 � 6 85 � 5Pinacidil preconditioning,

followed by MPTP47 � 4** 50 � 4**

Pinacidil preconditioning in thepresence of glibenclamide,followed by MPTP

75 � 5‡ 80 � 6‡

†Activation of K-ATP channels protects PC12 cells and SH-SY5Y cellsagainst MPTP toxicity. The percentage of cell death induced by 50 �MMPTP on day 7 in control, untreated cells and those preconditioned with10 �M pinacidil in the absence or presence of 10 �M glibenclamide. Dataare means � SEM of four independent determinations.**, ‡Significant difference at P � 0.01 compared with the MPTP-treatedgroup and cells with pinacidil preconditioning in the absence of gliben-clamide, respectively.

Fig. 2. Activation of K-ATP channels confers protection in PC12 cellsagainst rotenone-induced cell death. A: Trypan blue exclusion assays ofcontrol, untreated PC12 cells and PC12 cells treated with 20 �Mrotenone for 7 days with and without prior preconditioning with10 �M pinacidil alone or in the presence of 10 �M glibenclamide.B: Percentage of cell death induced by 0.03–60 �M rotenone on day

7 in control untreated cells (squares) and those preconditioned with10 �M pinacidil in the absence (solid triangles) or presence (opentriangles) of 10 �M glibenclamide. Each point represents the mean �SEM of four independent experiments. **Significant decrease in per-centage of cell death from those cells without pinacidil preconditioningat P � 0.01.

562 Tai and Truong

death from �88% to �57% (Fig. 4C). In a similar manner,preconditioning of PC 12 cells with X/XO (10 �M/0.5 mU ml–1) for 15 min also significantly reduced (P �0.05) cell death to �69% (Fig. 4C).

DISCUSSIONIn this study, we tested whether preconditioning can

increase cell survival against the neurotoxic effect of rote-none, a mitochondrial toxin implicated in the pathogen-esis of PD. Preconditioning was mimicked in vitro bytransient exposure of PC12 cells to Fe2� or X/XO, whichinduces the generation of ROS (Ravati et al, 2000), orpinacidil, a K-ATP channel opener, which activatesK-ATP channels but not other ion channels in membranepatches of isolated neurons (Lee et al., 1995). Systemicadministration of pinacidil in the rat does not alter theregulation of calcium ion channels and small Ca2�-activated K� channels in the brain (Gopalakrishnan and

Triggle, 1992), indicating that pinacidil is a specificK-ATP channel opener.

Our results indicate that PC12 cells are not sensitiveto rotenone at 0.03–1 �M. This appears to be inconsistentwith findings from a previous study (Hartley et al., 1994)that demonstrated rotenone, at 50–300 nM, to causemaximal lactate dehydrogenase release into the medium.This discrepancy could be due to the conditions underwhich the PC12 cells were cultured. For example, it iswell documented that the concentration of glucose in theculture medium would dramatically affect the susceptibil-ity of PC12 cells to the toxic effect of rotenone (Basma etal., 1992). Rotenone was much less effective in killing thecells incubated in high concentrations of glucose medium,as were used in our studies, because PC12 cells are neo-plastic in nature and have a relatively high rate of glycolysisand low consumption of glucose carbon via the Krebs

Fig. 3. Degree of K-ATP channel-mediated protection in PC12 cells is inproportion to the number of exposures to pinacidil. A: Trypan blueexclusion assays showing control, untreated PC12 cells or PC12 cellstreated with 20 �M rotenone for 7 days with or without preconditioningof three exposures to 10 �M pinacidil. B: Percentage of cell death in PC12

cells preconditioned with 10 �M pinacidil for different numbers of timeswith (squares) or without (triangles) 20 �M rotenone treatment for 7 days.Each point represents the mean � SEM of four independent determina-tions. *,**Significantly lower percentage of cell death than in those cellspreconditioned with pinacidil once at P � 0.05 and P � 0.01, respectively.

K-ATP Channel-Mediated Protection From Cell Death 563

cycle (Racker and Spector, 1981; Morelli et al., 1986).Rotenone acts by inhibiting complex I activity in themitochondrial respiration chain. As a result, both oxidativephosphorylation and the glycolytic pathway must come toa halt before total depletion of ATP stores and rotenone-induced cell damage becomes apparent. This is consistentwith our observation that rotenone kills PC12 cells slowlyand gradually in a time- and concentration-dependentmanner. Significantly, rotenone-induced cell death inPC12 cells is primarily a consequence of complex I inhi-bition, in that rotenone has no significant deleterious effecton rho 0 SH-SY5Y cells, in which complex I activity isabsent. This finding is in line with the earlier observation

that 5 mM methylpyridinium ion (MPP�), another com-plex I inhibitor implicated in the pathogenesis of PD(Langston et al., 1983), does not cause apoptosis in rho 0SH-SY5Y cells (Fall and Bennett, 1999).

It has been known for some time that activation ofK-ATP channels acts an endogenous defense mechanismagainst neuronal injury induced by an array of stresses,including hypoxia/ischemia (Wind et al., 1997; Ettaiche etal., 2001), oxidative insults, and amyloid beta-peptidetoxicity (Goodman and Mattson, 1996). The present studyis the first demonstration that transient activation ofK-ATP channels can precondition PC12 cells against theneurotoxic effect of a mitochondrial complex I inhibitor,

Fig. 4. K-ATP channel-mediated protection requires protein synthesis,and rotenone-induced cell death can be attenuated by other precondi-tioning stimuli, FeSO4 or X/XO. A: Percentage of cell death inducedby rotenone (0.03–60 �M) on day 7 after incubation in PC12 cells(squares) and in PC12 cells preconditioned with pinacidil in the absence(solid triangles) and presence (open triangles) of 50 �M cycloheximide.B: Percentage of cell death induced by rotenone (0.03–60 �M) on day7 in control PC12 cells (squares) and in PC12 cells preconditioned with

pinacidil (triangles) or in PC12 cells treated with rotenone and pinacidilpreconditioning simultaneously (circles). C: Percentage of cell death incontrol, untreated PC12 cells and in FeSO4 (100 �M)- or X/XO(10 �M/0.5 mU ml–1)- preconditioned PC12 cells incubated with 20�M rotenone for 7 days. Values are mean � SEM of four independentdeterminations. *,**Significant difference in percentage of cell deathvs. the corresponding control group at P � 0.05 and P � 0.01,respectively.

564 Tai and Truong

rotenone. The results reveal a new neuronal protectiverole of K-ATP channel signaling in a cell culture model ofPD.

It is not known by which mechanism(s) activation ofK-ATP channels induces protection in PC12 cells againstrotenone-induced cell death. The acquisition of protec-tion in PC12 cells by pinacidil preconditioning requires atime window of hours or days, and the protective effectcan be attenuated by concurrent exposure to a proteinsynthesis inhibitor. In addition, the degree of protectioncorrelates with the number of bouts of exposure to pinaci-dil. These results seem to suggest that the induced protec-tive effects are probably associated with transcriptionalprocesses and result from an enhanced expression of genesthat promote cell survival or repressed expression of genesthat cause cell death. It has been shown that K-ATPchannel activation can induce the expression of the 70,000molecular weight class of heat shock proteins (HSPs;Blondeau et al., 2000), which act as chaperones to restorethe normal functioning of stress-damaged proteins by re-folding or reassembling them. Further studies are neededto verify whether HSPs are involved in the K-ATPchannel-mediated protection against rotenone toxicity.

Whether K-ATP channels are the final effector inIPC is still a point of contention. The time lag of daysbetween initial K-ATP channel activation and the mani-festation of its beneficial actions indicates that an additionalfactor(s) must be involved in mediating the observed pro-tection following activation of K-ATP channels. Thisfinding is in line with the notion that the K-ATP channelserves as an upstream trigger rather than an end effector ofIPC (Pain et al., 2000).

We also found that a brief exposure to Fe2� orX/XO can precondition PC12 cells against rotenone tox-icity. Recent study showed that Fe2� or X/XO activatesthe transcription factor nuclear factor-�B (NF-�B) andinduces an increase in the protein level of manganese-dependent superoxidase dismutase (Mn-SOD; Ravati etal., 2001), an antioxidant whose gene expression is ele-vated in the dopaminergic neurons of the nigrostriatalpathway following 6-hydroxydopamine lesioning (Kuni-kowska and Jenner, 2001). In addition to its role in im-munity, NF-�B is also known to inhibit programmed celldeath (apoptosis), a process by which MPTP kills SH-SY5Y cells (Sheehan et al., 1997). Thus, it is likely that theobserved Fe2�- or X/XO-induced protection against ro-tenone toxicity involves NF-�B signaling cascades andantioxidant enzymes activities.

Despite the identification of genetic abnormality inhereditary PD and Huntington’s disease in recent years,the precise roles of these genes and proteins leading to theselective neuronal death in specific regions of the brainremain a mystery. Emerging evidence reveals that thepathogenesis of cell death in a number of neurodegenera-tive diseases is associated with excitotoxicity, energy de-fects, and oxidative damage processes. One therapeuticstrategy is to increase the capacity of the cells to resist orprevent cell injury from stress-related insults. Our results

show that activation of K-ATP channels and two otherpreconditioning stimuli can boost natural defense mecha-nisms to combat stress-induced injury. Because K-ATPchannel signaling and Fe2�- or X/XO-induced ROSboth are involved in eliciting tissue protection from IPC,our findings also conceptually support the view that IPC-derived therapeutic strategies may be feasible for PD andpossibly other neurodegenerative diseases in which mito-chondrial dysfunction plays a role in the pathogenesis,although this concept requires validation in vivo.

ACKNOWLEDGMENTSWe thank Natalie Gude and Dr. Maurice Montal

(University of California, San Diego) for the PC12 cells.We also thank Kate Borland, Dr. James P. Bennett, Dr.P.A. Trimmer, and J.A. Almeida (University of VirginiaCenter for the Study of Neurodegenerative Diseases) forthe SH-SY5Y and rho 0 SH-SY5Y cell lines. We aregrateful to Gregory B. Quinn and Mary Ann Chapman forediting the manuscript.

REFERENCESAlvarez-Tejado M, Naranjo-Suarez S, Jimenez C, Carrera AC, Landazuri

MO, del Peso L. 2001. J Biol Chem 276:22368–22374.Basma AN, Heikkila RE, Saporito MS, Philbert M, Geller HM, Nicklas

WJ. 1992. 1-Methyl-4-(2�-ethylphenyl)-1,2,3,6-tetrahydropyridine-induced toxicity in PC12 cells is enhanced by preventing glycolysis.J Neurochem 58:1052–1059.

Betarbet R, Sherer TB, MacKenzie G, Garcia-Osuna M, Panov AV,Greenamyre JT. 2000. Chronic systemic pesticide exposure reproducesfeatures of Parkinson’s disease. Nat Neurosci 3:1301–1306.

Blondeau N, Plamondon H, Richelme C, Heurteaux C, Lazdunski M.2000. K(ATP) channel openers, adenosine agonists and epileptic precon-ditioning are stress signals inducing hippocampal neuroprotection. Neu-roscience 100:465–474.

Chorsky RL, Yaghmai F, Hill WD, Stopa EG. 2001. Alzheimer’s disease:a review concerning immune response and microischemia. Med Hypoth56:124–127.

Cohen MV, Baines CP, Downey JM. 2000. Ischemic preconditioning:From adenosine receptor to KATP channels. Annu Rev Physiol 62:79–109.

Domoki F, Perciaccante JV, Veltkamp R, Bari F, Busija DW.1999. Mitochondrial potassium channel opener diazoxide preservesneuronal-vascular function after cerebral ischemia in newborn pigs.Stroke 30:2713–2718.

Downey JM, Cohen MV, Ytrehus K, Liu Y. 1994. Cellular mechanisms inischemic preconditioning: the role of adenosine and protein kinase C.Ann N Y Acad Sci 723:82–98.

Ettaiche M, Heurteaux C, Blondeau N, Borsotto M, Tinel N, LazdunskiM. 2001. ATP-sensitive potassium channels (K-ATP) in retina: a key rolefor delayed ischemic tolerance. Brain Res 890:118–129.

Fall CP, Bennett JP Jr. 1999. Characterization and time course of MPP�-induced apoptosis in human SH-SY5Y neuroblastoma cells. J NeurosciRes 55:620–628.

Ferrante RJ, Schulz JB, Kowall NW, Beal MF. 1997. Systemic adminis-tration of rotenone produces selective damage in the striatum and globuspallidus, but not in the substantia nigra. Brain Res 753:157–162.

Gidday JM, Shah AR, Maceren RG, Wang Q, Pelligrino DA, HoltzmanDM, Park TS. 1999. Nitric oxide mediates cerebral ischaemic tolerance ina neonatal rat model of hypoxic preconditioning. J Cereb Blood FlowMetab 19:331–340.

K-ATP Channel-Mediated Protection From Cell Death 565

Gonzalez-Zulueta M, Feldman AB, Klesse LJ, Kalb RG, Dillman JF, ParadaLF, Dawson TM, Dawson VL. 2000. Requirements for nitric oxideactivation of p21ras/extracellular regulated kinase in neuronal ischemicpreconditioning. Proc Natl Acad Sci USA 97:436–441.

Goodman Y, Mattson MP. 1996. K� channel openers protect hippocampalneurons against oxidant injury and amyloid beta-peptide toxicity. BrainRes 706:328–332.

Gopalakrishnan M, Triggle DJ. 1992. Regulation of ATP-sensitive K�

channels by chronic glyburide and pinacidil administration. BiochemPharmacol 44:1843–1847.

Hartley A, Stone JM, Heron C, Cooper JM, Schapira AHV. 1994. Com-plex I inhibitors induce dose-dependent apoptosis in PC 12 cells: rele-vance to Parkinson’s disease. J Neurochem 63:1987–1990.

Heurteaux C, Lauritzen I, Widmann C, Lazdunski M. 1995. Essential roleof adenosine, adenosine A1 receptors, and ATP-sensitive K� channels incerebral ischemic preconditioning. Proc Natl Acad Sci USA 92:4666–4670.

Ishida T, Yarimizu K, Gute DC, Korthuis RJ. 1997. Mechanisms ofischemic preconditioning. Shock 8:86–94.

Jenner P. 2001. Parkinson’s disease, pesticides and mitochondrial dysfunc-tion. Trends Neurosci 24:245–247.

King MP, Attardi G. 1989. Human cells lacking mtDNA: repopulationwith exogenous mitochondria by complementation. Science 246:500–503.

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 in the brain. Brain Res 528:21–24.

Kunikowska G, Jenner P. 2001. 6-Hydroxydopamine lesioning of thenigrostriatal pathway in rats alters basal ganglia mRNA for copper-, zinc-and manganese-superoxidase dismutase, but not glutathione peroxidase.Brain Res 922:51–64.

Langston JW, Ballard P, Tetrud JW, Irwin I. 1983. Chronic parkinsonismin humans due to a product of meperidine-analog synthesis. Science219:979–980.

Lee K, Rowe IC, Ashford ML. 1995. NS 1619 activates BKCa channelactivity in rat cortical neurons. Eur J Pharmacol 280:215–219.

Morelli A, Grasso M, Calissano P. 1986. Effect of nerve growth factor onglucose utilization and nucleotide content of pheochromocytoma cells(clone PC12). J Neurochem 47:375–381.

Murry CE, Jennings RB, Reimer KA. 1986. Preconditioning with isch-emia: a delay of lethal cell injury in ischemic myocardium. Circulation74:1124–1136.

Pain T, Yang XM, Critz SD, Yue Y, Nakano A, Liu GS, Heusch G, Cohen

MV, Downey JM. 2000. Opening of mitochondrial K(ATP) channelstriggers the preconditioned state by generating free radicals. Circ Res87:460–466.

Perez-Pinzon MA, Born JG. 1999. Rapid preconditioning neuroprotectionfollowing anoxia in hippocampal slices: role of the K� ATP channel andprotein kinase C. Neuroscience 89:453–459.

Racker E, Spector M. 1981. Warburg effect revisited: merges of biochem-istry and molecular biology. Science 213:303–307.

Ravati A, Ahlemeyer B, Becker A, Krieglstein J. 2000. Preconditioning-induced neuroprotection is mediated by reactive oxygen species. BrainRes 866:23–32.

Ravati A, Ahlemeyer B, Becker A, Klumpp S, Krieglstein J. 2001.Preconditioning-induced neuroprotection is mediated by reactive oxygenspecies and activation of the transcription factor nuclear factor-kappa B.J Neurochem 78:909–919.

Reshef A, Sperling O, Zoref-Shani E. 1998. Opening of ATP-sensitivepotassium channels by cromakalim confers tolerance against chemicalischemia in rat neuronal cultures. Neurosci Lett 250:111–114.

Rubino A, Yellen DM. 2000. Ischemic preconditioning of the vasculature:an overlooked phenomenon for protecting the heart? Trends PharmacolSci 21:225–230.

Saitoh Y, Gu K, Kin S, Yamauchi M, Sasaki T, Nosaka S, Hashimoto M,Nakayama K, Tamura K. 1995. Ischemic preconditioning improves car-diac functional recovery following preservation with University of Wis-consin solution. Transplantation 60:1079–1083.

Sandhu R, Diaz RJ, Mao GD, Wilson GJ. 1997. Ischemic preconditioning:differences in protection and susceptibility to blockade with single-cycleversus multicycle transient ischemia. Circulation 96:984–995.

Sheehan JP, Palmer PE, Helm GA, Tuttle JB. 1997. MPP� inducedapoptotic cell death in SH-SY5Y neuroblastoma cells: an electron micro-scope study. J Neurosci Res 48:226–237.

Tai KK, Blondelle SE, Ostresh JM, Houghten RA, Montal M. 2001. AnNMDA receptor channel blocker with neuroprotective activity. ProcNatl Acad Sci USA 98:3519–3524.

Tauskela JS, Chakravarthy BR, Murray CL, Wang Y, Comas T, Hogan M,Hakim A, Morley P. 1999. Evidence from cultured cortical neurons ofdifferences in the mechanism of ischemic preconditioning of brain andheart. Brain Res 827:143–151.

Uversky VN, Li J, Fink AL. 2001. Pesticides directly accelerate the rate ofalpha-synuclein fibril formation: a possible factor in Parkinson’s disease.FEBS Lett 500:105–108.

Wind T, Prehn JHM, Peruche B, Krieglstein J. 1997. Activation of ATP-sensitive potassium channels decreases neuronal injury caused by chemicalhypoxia. Brain Res 751:295–299.

566 Tai and Truong