B R A I N R E S E A R C H 1 3 0 2 ( 2 0 0 9 ) 2 4 0 – 2 4 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 i n res

Research Report

Marine compound Xyloketal B protects PC12 cells againstOGD-induced cell damage

Jia Zhaoa,1, Ling Lia,1, Chen Linga, Jie Lib, Ji-Yan Pangc,d, Yong-Cheng Linc,d, Jie Liud,Ruxun Huanga, Guan-Lei Wangc,⁎, Zhong Peia,⁎, Jinsheng Zenga

aDepartment of Neurology, The First Affiliated Hospital, Sun Yat-Sen University, Guangzhou 510080, PR ChinabDepartment of Anesthesiology, The Second Affiliated Hospital, Sun Yat-Sen University, Guangzhou 510080, PR ChinacDepartment of Applied Chemistry and Department of Pharmacy, Sun Yat-Sen University, Guangzhou 510080, PR ChinadKey Laboratory of Functional Molecules from Oceanic Microorganisms (Sun Yat-sen University),Department of Education of Guangdong Province 510080, PR China

A R T I C L E I N F O

⁎ Corresponding authors. Z. Pei is to be contaZhongshan Road 2, Guangzhou 510080, ChinMedicine, Sun Yat-Sen University, 74 Zhongs

E-mail addresses: wangglei@mail.sysu.eduAbbreviations: PC12 cells, rat pheochromoc

2-picrylhydrazyl; Drp1, GTPase dynamin-reladensity lipoprotein1 These authors contributed equally to this

0006-8993/$ – see front matter © 2009 Elsevidoi:10.1016/j.brainres.2009.09.034

A B S T R A C T

Article history:Accepted 6 September 2009Available online 16 September 2009

Xyloketal B is a novel marine compound with unique chemical structure isolated frommangrove fungus Xylaria sp. (no. 2508). Recently, we have demonstrated that Xyloketal B isan antioxidant and can protect against oxidized low density lipoprotein (LDL)-induced cellinjury. In the present study, we investigated whether Xyloketal B can protect againstischemia-induced cell injury in an in vitro oxygen glucose deprivation (OGD) model ofischemic stroke in PC12 cells. We found that Xyloketal B could directly scavenge 1,1-diphenyl-2-picrylhydrazyl (DPPH) free radical and protect PC12 cells against OGD insult.Furthermore, Xyloketal B alleviated OGD-induced mitochondria superoxide, mitochondriafragmentation and GTPase dynamin-related protein 1 (Drp1) overexpression as well asreduction of mitochondrial membrane potential. All together, the present studydemonstrates that Xyloketal B protects PC12 cells against OGD-induced cell injury andthat the anti-oxidative property and protective action on mitochondria may account for itsneuroprotective actions.

© 2009 Elsevier B.V. All rights reserved.

Keywords:Xyloketal BOxygen glucose deprivationAntioxidantMitochondriaStrokeDrp1

1. Introduction

Stroke is one of the leading causes of disability and deathin developing countries. Approximately 40% of stroke deathsin developing countries were in China and the majority of

cted at Department of Na. Fax: +86 20 87335935.han Road II, Guangzhou.cn (G.-L. Wang), peizhonytoma; OGD, oxygen-glucted protein 1; MTT, 3-(4,5-

work.

er B.V. All rights reserved

survivors are disabled (Reddy and Yusuf, 1998). However, thetherapeutic option for stroke is very limited. Althoughrecombinant tissue plasminogen activator has been approvedby US Food and Drug Administration as an effective stroketherapy, it is only beneficial to 5–10% of acute ischemic stroke

eurology, First Affiliated Hospital, Sun Yat-Sen University, No. 58G.-L. Wang, Department of Pharmacology, Zhongshan School of510080, PR China. Fax: +86 20 87335935.g@mail.sysu.edu.cn (Z. Pei).ose deprivation; ROS, reactive oxygen species; DPPH, 1,1-diphenyl-dimethylthiazole-2-yl)-2,5-biphenyl tetrazolium bromide; LDL, low

.

Fig. 1 – Effect of Xyloketal B on DPPH free radical. (A) Thestructure of Xyloketal B. (B) Xyloketal B significantly inhibitedDPPH in a concentration-dependent manner. (P<0.01 vs.control, n=6 wells for each group, Kruskal–Wallis test). Theresults were obtained from three independent experiments.

241B R A I N R E S E A R C H 1 3 0 2 ( 2 0 0 9 ) 2 4 0 – 2 4 7

patients (Brown et al., 2005). Therefore, it is an urgent taskto develop effective and safe therapies for acute ischemicstroke.

Ischemic stroke accounts for 80% to 90% of all stroke casesand oxidative stress plays an important role in the patho-physiology of ischemic stroke (Kontos, 2001; Andersen, 2004).Reactive oxygen species (ROS) formed during ischemia andreperfusion contribute to neuronal injury whereas suppres-sion of ROS alleviates ischemia-induced neuronal damage(McCord, 1985; Greenlund et al., 1995). Mitochondria are animportant therapeutic target for ischemic stroke. Duringischemia, mitochondria generate huge amounts of ROS. Onthe other hand, excessive ROS also damage mitochondria(Schild and Reiser, 2005; Yamamoto et al., 2007). Consequently,the injured mitochondria in turn produce more ROS, causinga vicious cycle.

Antioxidants have been shown to neutralize hazardousfree radicals and prevent cell death in animalmodels of strokeeven when given after stroke. When combined with othertherapies, antioxidants can have additional benefits in strokemodels. Thus, antioxidants are thought to be a promisingtreatment for strokes. Despite the fact that lots of antioxida-tive agents have failed to show significant clinical benefits inthe treatment of stroke (Kamat et al., 2008), antioxidantEdaravone has been approved in Japan for clinical use instroke patients. The efficacy of Edaravone in acute cerebralinfarction has been proved in a randomized, placebo-con-trolled, double-blind clinical trail conducted at multiplecenters (Edaravone Acute Infarction Study Group, 2003).

Xyloketal B is a novel marine compound with uniquechemical structure isolated from mangrove fungus Xylaria sp.(no. 2508). Recently, we have demonstrated that Xyloketal Bprotects against oxidized LDL-induced cell injury mainlythrough its antioxidative activity (Chen et al., 2009). XyloketalB is quite safe and high concentrations of Xyloketal B do notcause toxic or proliferative effects on cells. Structurally,Xyloketal B has a hydroxyl-phenol radical structure, suggest-ing that Xyloketal B may have direct free radical scavengingactivity. Collectively, Xyloketal B is an attractive candidate forstroke drug development.

In the present study, we first investigated whetherXyloketal can directly scavenge DPPH free radical in a cellfree system and then studied the neuroprotective potential ofXyloketal B in an in vitromodel of ischemic stroke in PC12 cells.Additionally, we also explored themitochondrial mechanismsunderlying protective potential of Xyloketal B.

2. Results

2.1. Free radical scavenging ability of Xyloketal B

DPPH assaywas used to investigate the free radical scavengingability of Xyloketal B without the involvement of cells oranimals. DPPH is a stable free radical and can accept anelectron or hydrogen radical to become a stable diamagneticmolecule. This feature made it a sensitive agent for detectingantioxidant activities (Prasad et al., 2009). DPPH assay showedthat Xyloketal B at a concentration range of 12.5 to 800 μMsignificantly reduced DPPH free radical (Fig. 1B).

2.2. Effect of Xyloketal B on OGD-induced injury inPC12 cells

3-(4,5-dimethylthiazole-2-yl)-2,5-biphenyl tetrazolium bro-mide (MTT) assay and nuclear morphological analysis wereused to evaluate the cell viability. To test weather Xyloketal Balone is toxic to PC12 cells, different doses (12.5 to 200 μM) ofXyloketal B were added to normal PC12 cell cultures,respectively. There was no significant difference in cell-viability between Xyloketal B-treated groups and vehicle-treated group (data not shown). We then examined whetherXyloketal B can protect PC12 cells against OGD insult. OGD 4 hplus 24 h reperfusion led to approximately 40% reduction ofMTT value. Xyloketal B at a concentration range of 50 to200 μM significantly prevented OGD-induced decrease of MTTvalues in a concentration-dependentmanner (Fig. 2B). Nuclearmorphological analysis yielded similar results. The morpho-logical changes were almost invisible in the normal controlcells. In contrast, OGD 4 h plus 24 h reperfusion led tosubstantial morphological changes such as crenation andcondensation in PC12 cells. The extents of cell damage were2.87±0.62%, 25.63±1.92%, and 6.30±0.86% in control group,OGD group and Xyloketal B+ OGD group, respectively.Pretreatment with Xyloketal B significantly attenuated thenumber of cells with abnormal nuclear morphology inducedby OGD (Fig. 2A). According to the concentration-effect trendof Xyloketal B, Xyloketal B at 100 μM was applied to allsubsequent experiments.

Fig. 2 – Effect of Xyloketal B on cell viability in PC12 model of OGD. (A) Digital photomicrograph under fluorescent illuminationshowingmorphology of nuclei. PC12 were stained with DAPI and nuclei were imaged. Crenation of nuclei and condensation ofchromatin were evident in PC12 cells exposed to OGD-induced (marked with arrows) and OGD-induced nucleusmorphologicalchanges were obviously reduced in PC12 cells receiving treatment with Xyloketal B (100 μM) before exposure to OGD (P<0.01,n>500, χ2 test). (B) Bar graph showing the effect of pretreatment with different concentrations of Xyloketal B on the survival ofcultured PC12 cells at 24 h following 4 h of OGD. Cell viability is indicated by the MTT values of each experimental groupexpressed as percentages of that of the sham-OGD group. Values are means ± SEM. The cell viability was determined by MTTassay. Pretreatment with Xyloketal B attenuated OGD-induced cell injury in a concentration dependent manner (P<0.01, n=6wells for each group, Kruskal–Wallis test). The results were obtained from three independent experiments.

242 B R A I N R E S E A R C H 1 3 0 2 ( 2 0 0 9 ) 2 4 0 – 2 4 7

2.3. Effects of Xyloketal B on OGD-induced mitochondrialsuperoxide production and protein oxidative damage inPC12 cells

To further explore Xyloketal B's protective mechanisms, weinvestigated its effect on the mitochondrial ROS, and proteinoxidative damage. MitoSOX signal was localized in Mito-Tracker-stained organelles, indicating that the MitoSOXstaining was specific to mitochondria (data not shown).Compared with control (1.23±0.08), OGD significantly in-creased the intensity of MitoSOX signal (1.76±0.01) whereasXyloketal B (100 μM ) significantly ameliorated OGD-inducedmitochondrial ROS up to 27% (Fig. 3A).

The oxidative processes might affect the side chains ofmost amino acid residues, and the most widely studiedmodification induced by oxidative stress is the formation ofcarbonyl groups. Thus, protein carbonyl formation has beenthought to be a sensitive marker of oxidative stress-inducedprotein denaturation (Singhal et al., 2002). Based on theirreaction with DNPH, the carbonyls of the oxidized protein canbe detected by Western blot analysis with anti-dinitrophenyl

(DNP) antibodies, a procedure known as Oxyblot (Levine et al.,1994).

There was very low level of protein oxidation both incontrol group and in Xyloketal B alone-treated group. Asshown in Fig. 4C, the quantification of protein oxidationrevealed that there was no statistical difference in level ofprotein oxidation between the control group and Xyloketal Balone-treated group. The increase in protein oxidation levelwas evident in OGD group compared with control group(P<0.01). In contrast, pretreatment with Xyloketal B (100 μM)significantly reduced OGD-induced protein oxidation (P<0.01).

2.4. Effect of Xyloketal B on mitochondrial morphology andfunction under OGD

In normal PC 12 cell cultures, mitochondria displayed typicalelongated filamentous morphology. When cells exposed toOGD, a lot of filamentous mitochondria become small andround organelles. The percentage of fragmented mitochon-dria relative to intact mitochondria was 3.90±0.11, 4.23±0.01,23.50±0.20 and 10.78±0.10 in control, Xyloketal B alone, OGD

Fig. 3 – Effect of Xyloketal B on OGD-induced mitochondria dysfunction. (A) Bar graph showing mitochondrial superoxideproduction which is expressed as the ratio of MitoSOX to Mitotracker. Xyloketal B (100 μM) significantly attenuatedOGD-induced mitochondrial superoxide production (P<0.01, n=6 wells for each group, Kruskal–Wallis test). (B). Bar graphshowing mitochondrial membrane potential which is expressed as Rhodamine-123 to Mitotracker. Xyloketal B (100 μM)significantly attenuated OGD-induced mitochondrial superoxide production (P<0.01, n=6 wells for each group, ANOVA withLSD). The results were obtained from three independent experiments. (C) Digital photograph showing the level of proteinoxidation detected by Oxyblot and Bar graph showing quantitative analysis of the total protein oxidation. Xyloketal B (100 μM)significantly attenuated OGD-induced protein oxidation (P<0.01, Kruskal–Wallis test). Representative images were selectedfrom three independent experiments.

243B R A I N R E S E A R C H 1 3 0 2 ( 2 0 0 9 ) 2 4 0 – 2 4 7

and Xyloketal B+OGD, respectively. OGD significantly in-creased fragmented mitochondria (OGD vs. control; 23.50±0.20vs. 3.90±0.11, P<0.01). Treatment with Xyloketal B (100 μM)significantly attenuated OGD-induced mitochondrial fragmen-tation (Xyloketal B+OGD vs. OGD; 10.78±0.10 vs. 23.50±0.20,P<0.01).

To further analyze the effects of Xyloketal B on OGD-induced mitochondrial morphology, we measured Drp1 pro-tein expression in PC12 cells exposed to different treatments.Drp1 is regarded as an important factor which has a closerelationship to apoptosis and mitochondria dysfunction(Barsoum et al., 2006). Overexpression of Drp1 increasesneurons loss through promoting mitochondrial fission inmany neurological diseases including stroke whereas inhibi-tion of Drp1 prevents mitochondria-dependent cell death(Smirnova et al., 2001; Barsoum et al., 2006).

The western blot analysis demonstrated that the densityof Drp1 bands was 13161.50±1513.76, 14457.83±1377.54,

18399.41±2083.78, 24963.38±2970.74 in control, Xyloketal Balone, Xyloketal B+OGD and OGD group, respectively (Fig. 4B).Compared with control, OGD induced a significant increase inexpression of Drp1. In contrast, pretreatment with XyloketalB (100 μM) significantly attenuated OGD-induced overexpres-sion of Drp1.

Mitochondrial membrane potential was assayed usingRhodamine-123. The ratio between the intensity of mitochon-drial membrane potential andmass ofmitochondria was usedfor comparison among groups. The ratio of mitochondriamembrane potential relative to mass of mitochondria was7.72±0.18 in control group and 8.15±0.29 in XyloketalB-treated group. OGD significantly decreased mitochondrialmembrane potential (OGD vs. control; 5.18±0.31 vs. 7.72±0.18,P<0.01). Treatment with Xyloketal B (100 μM) significantlyattenuated OGD-induced reduction of mitochondrial mem-brane potential (Xyloketal B+ OGD vs. OGD; 6.69±0.37 vs. 5.18±0.31, P<0.01, Fig. 3B).

Fig. 4 – Effect of Xyloketal B on mitochondrial morphology and function under OGD in PC12. (A) Digital photomicrograph underfluorescent illumination showing the morphology of mitochondria was detected using Mitotracker staining. Small roundmitochondrial organelles were evident in PC12 cells exposed to OGD insult whereas Xyloketal B significantly attenuated thenumber of OGD-induced small round mitochondrial organelles (P<0.05, ANOVA with LSD). The represent images were fromthree independent experiments (60×). (B) Digital photograph showing the expression of Drp1 detected by Western blot. Blotswere probedwith antibodies against Drp1 andβ-actin was used as a control. Bar graph showing semi-quantified densitometryfrom Western blot. Drp1 were significantly increased in PC12 cells exposed to OGD insult whereas Xyloketal B significantlyattenuated OGD-induced Drp1 overexpression (P<0.05, ANOVA with LSD). The representative images were selected from fourindependent experiments.

244 B R A I N R E S E A R C H 1 3 0 2 ( 2 0 0 9 ) 2 4 0 – 2 4 7

3. Discussion

In the present study, we have revealed that Xyloketal B candirectly scavenge DPPH free radical production and protectPC12 cells against OGD-induced cell injury in a concentration-dependent manner. Furthermore, Xyloketal can alleviateOGD-induced mitochondria superoxide, mitochondria frag-mentation, and reduction of mitochondrial membrane poten-tial as well as protein oxidation.

OGD in PC12 cell line has been used as a rapid andsensitive in vitro model of ischemic stroke in development ofpotential neuroprotective agents. To mimic cerebral ischemia–reperfusion injury, PC12 cells are first subjected to a shortperiod of OGD (ischemia) followed by a prolonged period ofre-oxygenation and return of normal culture medium (reper-fusion). Thus this model is believed to better mimic thepathological conditions of stroke.

Previously, we have found that Xyloketal B can attenuateNADPH oxidase-derived ROS generation in oxLDL-treatedendothelial cells. We here investigated whether it can directlyscavenge free radicals in a cell free system. Xyloketal B at 12.5to 800 μM inhibited DPPH free radical production in aconcentration-dependent manner. Together with previous

observation, our data suggest that Xyloketal B may have abroad spectrum of antioxidant activity.

Free radicals are major mediators of ischemic neuronaldamage (Kontos 2001). Superoxide is the first ROS generated inthe oxygen free radical chain during the early phase ofreperfusion and contributes to neuronal injury. Consistently,we found that mitochondrial superoxide production, a potentfree radical, was significantly induced in OGD-treated PC12cells. As we expected, pretreatment with Xyloketal B signifi-cantly reduced OGD-induced superoxide production. Giventhat excessive generation of ROS is caused by disruptedbalance between the production and clearance of ROS(Noshita et al., 2001), further study is needed to investigatewhether Xyloketal B can directly scavenge mitochondrialsuperoxide or indirectly act on mitochondrial enzymesinvolved in the clearance of superoxide. Nevertheless, thisresult provides further evidence to demonstrate the antioxi-dant action of Xyloketal B.

The mitochondrion is at the core of cellular energymetabolism and normal mitochondrial morphology is essen-tial to maintain mitochondrial metabolic stability. The majorevents involved in the ischemia-induced cell death such asenergy failure and oxidative stress are tightly associated withthe mitochondrial function (Fiskum, 2000). In models of

245B R A I N R E S E A R C H 1 3 0 2 ( 2 0 0 9 ) 2 4 0 – 2 4 7

ischemic stroke, mitochondrial fragmentation is an earlyevent of ischemia and ROS has been implicated in ischemia-induced mitochondrial fragmentation. Physiologically, Drp1regulates mitochondrial morphology and is responsible formitochondrial fission (Barsoumet al., 2006). Overexpression ofDrp1 increases neurons loss through promoting mitochondri-al fission in many neurological diseases including strokewhereas inhibition of Drp1 prevents mitochondria-dependentcell death (Barsoum et al., 2006). Consistently, OGD signifi-cantly induced mitochondrial fragmentation and Drp1 ex-pression, suggesting the involvement of Drp1 in OGD-inducedmitochondrial fragmentation.When cells received Xyloketal Bbefore exposure to OGD insult, both OGD-induced mitochon-drial fragmentation and Drp1 overexpression were reduced.Moreover, pretreatment with Xyloketal B significantly re-duced OGD-induced reduction of mitochondrial membranepotential. These actions of Xyloketal B were in parallel to itsprotective effects, suggesting that the mitochondria may bean important target of Xyloketal B.

Another interesting finding in the present study is thatpretreatment with Xyloketal B reduced OGD-induced increasein protein oxidation. This finding may have clinical implica-tions. Proteins are major components of most biologicalsystems and increased protein oxidation has been detectedin stroke patients (Requena et al., 2001). Oxidation-induceddestructive effects on protein structure can critically impairprotein function, finally leading to cell death. Therefore, theinhibitory action of Xyloketal B on protein oxidation may beable to reduce the consequences of stroke.

Xyloketal B may have potential for treatment of stroke.Edaravone was approved in 2001 for treating acute ischemicstroke in Japan and was recommended by the JapaneseGuidelines for the Management of Stroke 2004. The protectiveeffect of Edaravone has been documented in the PC12 in vitroOGD model (Song et al., 2006). In OGD model, maximumprotection of Edaravone was approximate 25% at 0.1 μM, withno further protection achieved at higher doses (Song et al.,2006). Xyloketal B reduced OGD-induced PC12 cell damage by17.5% to 28% at the concentrations between 100 μM, and200 μM. Therefore, the beneficial effects of Xyloketal B mightbe comparable to Edaravone in PC12 cell model of stroke.

In summary, we here demonstrate the neuroprotectiveactions of Xyloketal B in an in vitro PC12 cell model of stroke.The beneficial effects seem to associate with its free radicalscavenging ability, antioxidant property and protection ofmitochondria. Given that Xyloketal B is a natural marinecompound and is quiet safe in a variety of cell systems, it maybe a good candidate for stroke treatment. Future study isnecessary to test its therapeutic efficacy and time window inanimal models of stroke.

4. Experimental procedures

4.1. Cell culture and reagents

PC12 (rat pheochromocytoma) cells were cultured in RPMI-1640(Invitrogen, USA) supplemented with 10% heat-inactivatedhorse serum (Invitrogen, USA) and 5% heat-inactivated fetalcalf serum (Invitrogen, USA) in a humidified incubator (Thermo

electron corporation, USA) containing 5% CO2 at 37 °C. PC12cells were cultured on collagen-coated plates or dishes whendifferent experiments were performed.

Xyloketal B was isolated and purified by Department ofApplied Chemistry and Department of Pharmacy, Sun Yat-SenUniversity, China (Lin et al., 2001; Pettigrew andWilson, 2006).Xyloketal B was dissolved in dimethyl sulfoxide (DMSO) andstored at −20 °C until use (Chen et al., 2009). The solution formof Xyloketal B was then diluted by PBS to the concentrationneeded. Cells were pretreated with Xyloketal B for 30 minbefore any other stimuli were performed. All reagents werepurchased from Sigma (Sigma, Shanghai, China) unlessotherwise stated.

4.2. Measurement of stable free radicalscavenging activity

The scavenging activity of the stable DPPH free radical wasconducted according to the methods described elsewhere(Lee et al., 2007; Xu et al., 2007). Briefly, 0.2 ml of ethanolicsolution of 80 μM DPPH (DPPH solution) was mixed with0.6 ml of sample solution containing different concentrationsof Xyloketal B (12.5–800 μM) to make Asample, and 0.2 ml ofDPPH solution was mixed with 0.6 ml ethanol to make Acontrol.The mixture was incubated in the dark at room temperaturefor 30 min. Absorbance was measured at 517 nm by anautomated enzyme immunoassay analyzer (BIO-TEK, synergy2, USA). The percentage of DPPH discoloration of each samplewas calculated according to the following equation: % ofdiscoloration=[1− (Asample/Acontrol)]×100. The degree of deco-loration was proportional to the scavenging activities of thesubstances.

4.3. Oxygen-glucose deprivation (OGD)

OGD was achieved using methods published elsewhere.Briefly, 24 h after PC12 cells were seeded in different cultureplates, the culture medium was changed to the glucose-freeDMEM containing either different concentrations of XyloketalB in 0.2% (w/v) DMSO in the Xyloketal B-treated groups or 0.2%DMSO in the vehicle-treated groups for 30 min. Then cellswere placed into an anaerobic chamber that was flushed with5% CO2 and 95% N2 (v/v). The cell cultures within theanaerobic chamber were kept in a humidified incubator at37 °C for various time intervals in different experiments. Toterminate the OGD, the culture medium was changed tonormal medium containing the same concentration of Xylo-ketal B in DMSO or DMSO alone before returning to thenormoxic incubating conditions. In the normal control groups,the cell cultures were subjected to the same experimentalprocedures with vehicle only and without exposure to theglucose-free medium or anoxia.

4.4. Cellular viability and DAPI staining

Cellular viability was assessed using the MTT kit (MTT CellGrowth Kit, Chemicon, USA) according to the instruction fromthe company. Briefly, 10 μl of the MTT labeling reagent at afinal concentration of 0.5 mg/ml was added into each well atthe termination of OGD before incubation in a humidified

246 B R A I N R E S E A R C H 1 3 0 2 ( 2 0 0 9 ) 2 4 0 – 2 4 7

incubator at 37 °C with 5% CO2 and 95% air (v/v) at 90%humidity for 4 h to allow formation of purple formazan crystal.4 h later, 100 μl of the solubilization reagent was added intoeach well. Finally, the spectrophotometric absorbance of thesolubilized purple formazan crystals was measured using amicroplate reader at an absorbance wavelength of 570 nm.

Cell viability was also evaluated using nuclear morphologyfollowing staining with the fluorescent dye, DAPI (2.5 μg/ml,Sigma, USA). Briefly, PC12 cells were fixed in 4% polylysine onice for 15 min, and were then washed with PBS twice. Withoutthe rupture of cell membrane, the cells were incubated withDAPI in PBS for 20 min at room temperature. Nuclearmorphology stained by DAPI was visualized under a fluores-cence microscope (Olympus DP70, Japan) and quantificationwas performed by a person who was not aware of theexperimental treatment. Condensed, shrunken, fragmentedor irregular nuclei were considered to be the sign of celldeath.

To quantitative analysis the changes of nuclear morphol-ogy, 10 different pictures were selected randomly in differentgroups and more than 500 cells in total were observed in eachgroup. Ratio of dead cells to total cells was compared amongdifferent groups.

4.5. Measurement of mitochondrial ROS production andprotein oxidation

Intracellular measurement of mitochondrial ROS generationwas performed using MitoSox (Invitrogen, USA). MitoSOX Redis a fluorescent dye specific for the detection the mitochon-drial superoxide in live cells. Briefly, 4 h after the OGD, PC12cells were loaded with the MitoSOX Red (2.5 μM) for 10 min. Toconfirm the mitochondrial localization of MitoSOX Red, cellswere also incubated with MitoTracker Green (0.2 μM, Invitro-gen, USA) for 10min. Following the removal of excessmitoSOXreagent, cells planted in 24-plates were washed with Hanks'balanced salt solution (HBSS) and were then imaged with afluorescence microscope (Olympus DP70, USA) using a 40 ×objective. Quantitative measurements of fluorescence read-ings were determined at 485/538 nm (excitation/emission)using a SpectraMax GEMINI EM fluorescent plate reader(Molecular Devices, USA) from cells planted in 96-plates.Fluorescence data were given as the ratio between theintensity of MitoSOX and Mitotracker in each well to compen-sate for differences of mitochondrial mass and unequalMitoSOX loading, The ratio per well was calculated andaveraged over six coordinate wells of each group and theaverage ratio was used for the comparison.

Protein oxidation was measured using a protein-oxidationdetection kit (OxyBlot; Chemicon, USA). Based on the featureof carbonyls reactingwith 2,4-dinitrophenylhydrazine (DNPH),the derivative products could be immunodetected (Namuraet al., 2001). Briefly, 5 μl protein sample (20 μg) was addedwith 5 μl of 12% SDS and 10 μl of 1× DNPH solution into atube. Tubes were incubated at room temperature for 15 min.Neutralization solution (7.5 μL) was added to each tube, andone mixed sample per lane was loaded onto 12% SDS-polyacrylamide gel. The following procedure was similar toWestern Blot Assay mentioned below except for a rabbit anti-DNP antibody (1:150; Chemicon, USA) used as the primary

antibody and a horseradish peroxidase (HRP)-conjugatedanti-rabbit IgG (1:300; Chemicon, USA) used as the secondaryantibody. The intensity of the five given protein were addedtogether to compare among groups. To determine specificity,the oxidized proteins provided by the kit were included as apositive control. Treatment of samples with a controlsolution served as a negative control to the DNPH treatment.Data are given as percentage of the intensity in a certaingroup to control.

4.6. Evaluation of mitochondrial morphology

For mitochondrial fission assay, cells were incubated with0.2 μM Mitotracker Green for 25 min and were washed withHBSS twice. Cells were then imaged with confocal microscope(FV500, Olympus, USA) at 40 × objective to study the alterna-tion of mitochondrial morphology.

4.7. Evaluation of Drp1 expression

Western blot assay was used to evaluate the level of Drp1.PC12 cells were lysed and centrifuged at 25,000 g for 3min. Thesupernatant was collected and protein concentration wasdetermined using a BCA Protein Assay Kit (Pierce, USA).Protein was subjected to 8% SDS-polyacrylamide gel forelectrophoresis. Following electrophoresis and transfer topolyvinylidene difluoride PVDF membranes (Millipore, USA),membranes were blocked in a buffer (50 mM Tris-HCl, 154mMNaCl, 0.1% Tween-20 [pH 7.5]) containing 5% nonfat dry milkfor 1 h, and incubated with appropriately diluted antibodies(1:1000 rabbit monoclonal antibody Drp1, Novus Biological Inc,USA; 1:5000 β-action, Sigma, China) for 16–18 h at 4°C. Afterthe primary antibody incubation, the membrane was washedand incubated with either horseradish peroxidase (HRP)-conjugated anti-mouse IgG or anti-rabbit IgG (R&D, USA) for1 h at room temperature. Chemiluminescence reactions(Milipore, USA) were conducted according to the manufac-turer's protocol. The intensity (INT×mm2) of each band wasmeasured and analyzed with a Chemi Doc XRS imagingsystem (BIO-RAD, USA). Data are given as percentage ofvehicle control and β-actin was used as an internal control(Namura et al., 2001).

4.8. Estimation of mitochondrial membrane potential

The mitochondrial membrane potential was measured usingthe fluorescence dye Rhodamine-123 (Invitrogen) (Baraccaet al., 2003). Cells were exposed to 10 μM Rhodamine-123 and0.2 μM Mitotracker Green for 25 min and were then washedtwice with HBSS before the fluorescence was determinedwith a SpectraMax GEMINI EM fluorescent plate reader(Molecular Devices, USA) at 511/534 nm (the excitation andemission wavelengths for Rhodamine-123, respectively) and490/516 nm (the excitation and emission wavelengths forMitotracker Green). Fluorescence data are given as the ratiobetween the average intensity of the mitochondrial mem-brane potential (Rhodamine-123) and the mitochondria(Mitotracker) to compensate for background differences andunequal Rhodamine-123 loading (Twig et al., 2008; Wang et al.,2008; Valle et al., 2005).

247B R A I N R E S E A R C H 1 3 0 2 ( 2 0 0 9 ) 2 4 0 – 2 4 7

4.9. Statistics

Data are shown at the form of mean±SEM. The comparisonsof multiple quantitative variables were analyzed using one-way analysis of variance (ANOVA) with LSD as the post hoc testor using Kruskal–Wallis test followed by Dunn's multiple-comparison test when the data were not normally distributed.Chi-square test was used to analyze categorical variables. Theanalyses were performed using SPSS software (13.0 forwindows). P<0.01 was considered to be statistically significantunless otherwise stated.

Acknowledgments

This work was supported by the grants from 863 project(2006AA09Z440) from the Ministry of Science and Technologyof the People's Republic of China, National Natural ScienceFoundation of China (No. 30873059), the National Key NewDrug Creation Program (2009ZX09103-039), and Science andTechnology Planning Project of Guangdong Province, China(2008B030303061).

R E F E R E N C E S

Andersen, J.K., 2004. Oxidative stress in neurodegeneration: causeor consequence. Nat. Med. 10, S18–25.

Baracca, A., Sgarbi, G., Solaini, G., 2003. Rhodamine 123 as a probeof mitochondrial membrane potential: evaluation of protonflux through F(0) during ATP synthesis. Biochim. Biophys. Acta1606, 137–146.

Barsoum, M.J., Yuan, H., Gerencser, A.A., 2006. Nitricoxide-induced mitochondrial fission is regulated bydynamin-related GTPases in neurons. EMBO J. 25, 3900–3911.

Brown, D.L., Barsan, W.G., Lisabeth, L.D., Gallery, M.E.,Morgenstern, L.B., 2005. Survey of emergency physiciansabout recombinant tissue plasminogen activator for acuteischemic stroke. Ann. Emerg. Med. 46, 56–60.

Chen, W.L., Qian, Y., Meng, W.F., Pang, J.Y., Lin, Y.C., Guan, Y.Y.,Chen, S.P., Liu, J., Pei, Z., Wang, G.L., 2009. A novel marinecompound xyloketal B protects against oxidizedLDL-induced cell injury in vitro. Biochem Pharmacol.78 (8), 941–950.

Edaravone Acute Infarction Study Group, 2003. Effect of a novelfree radical scavenger, edaravone (MCI-186), on acute braininfarction: randomized, placebo-controlled, double-blindstudy at multicenters. Cerebrovasc. Dis. 15, 222–229.

Fiskum, G., 2000. Mitochondrial participation in ischemic andtraumatic neural cell death. J. Neurotrauma. 17, 843–855.

Greenlund, L.J., Deckwerth, T.L., Johnson, E.M., 1995. Superoxidedismutase delays neuronal apoptosis: a role for reactiveoxygen species in programmed neuronal death. Neuron 14,303–315.

Kamat, C.D., Gadal, S., Mhatre, M., 2008. Antioxidants in centralnervous system diseases: preclinical promise and translationalchallenges. J. Alzheimers Dis. 15, 473–493.

Kontos, H.A., 2001. Oxygen radicals in cerebral ischemia. The 2001Willis Lecture. Stroke 32, 2712–2716.

Lee, H.H., Lin, C.T., Yang, L.L., 2007. Neuroprotection andfree radical scavenging effects of Osmanthus fragrans.J. Biomed. Sci. 14, 819–827.

Levine, R.L., Williams, J.A., Stadtman, E.R., Shacter, E, 1994.Carbonyl assays for determination of oxidatively modifiedproteins. Methods Enzymol. 233, 346–357.

Lin, Y., Wu, X., Feng, S., 2001. Five unique compound: xyloketalsfrommangrove fungus Xylaria sp. From South China Sea coast.J. Org. Chem. 66, 6252–6256.

McCord, J.M., 1985. Oxygen-derived free radicals in postischemictissue injury. N. Engl. J. Med. 312, 159–163.

Namura, S., Nagata, I., Takami, S., 2001. Ebselen reducescytochrome c release from mitochondria and subsequent DNAfragmentation after transient focal cerebral ischemia in mice.Stroke 32, 1906–1911.

Noshita, N., Sugawara, T., Fujimura, M., 2001. Manganesesuperoxide dismutase affects cytochrome c release andcaspase-9 activation after transient focal cerebral ischemia inmice. J. Cereb. Blood Flow Metab. 21, 557–567.

Pettigrew, J.D., Wilson, P.D., 2006. Synthesis of xyloketal A, B, C, Dand G analogues. J. Org. Chem. 71, 1620–1625.

Prasad, K.N., Hao, J., Yi, C., 2009. Antioxidant and anticanceractivities of Wampee. J. Biomed. Biotechnol. 2009, 1–6.

Reddy, K.S., Yusuf, S., 1998. Emerging epidemic of cardiovasculardisease in developing countries. Circulation 97, 596–601.

Requena, J.R., Chao, C.C., Levine, R.L., 2001. Glutamic andaminoadipic semialdehydes are the main carbonyl products ofmetalcatalyzed oxidation of proteins. Proc. Natl. Acad. Sci. 98,69–74.

Schild, L., Reiser, G., 2005. Oxidative stress is involved in thepermeabilization of the inner membrane of brainmitochondria exposed to hypoxia/reoxygenation and lowmicromolar Ca2+. FEBS J. 272, 3593–3601.

Singhal, A.B., Wang, X., Sumii, T., 2002. Effects ofnormobaric hyperoxia in a rat model of focal cerebralischemia–reperfusion. J. Cereb. Blood Flow Metab. 22, 861–868.

Smirnova, E., Griparic, L., Shurland, D.L., 2001. Dynamin-relatedprotein Drp1 is required for mitochondrial division inmammalian cells. Mol. Biol. Cell 12, 2245–2256.

Song, Y., Li, M., Li, J.C., 2006. Edaravone protects PC12 cellsfrom ischemic-like injury via attenuating the damage tomitochondria. J. Zhejiang Univ. Sci. B 7, 749–756.

Twig, G., Elorza1, A., Molina, A.J.A., 2008. Fission and selectivefusion govern mitochondrial segregation and elimination byautophagy. EMBO J. 27, 433–446.

Valle, I., lvarez-Barrientos, A.A., Arza, E., 2005. PGC-1a regulatesthe mitochondrial antioxidant defense system in vascularendothelial cells. Cardiovasc. Res. 66, 562–573.

Wang, X., Su, B., Siedlak, S.L., 2008. Amyloid-β overproductioncauses abnormal mitochondrial dynamics via differentialmodulation of mitochondrial fission/fusion proteins. Proc.Natl. Acad. Sci. 105, 19318–19323.

Xu, B.J., Yuan, S.H., Chang, S.K.C., 2007. Comparative analyses ofphenolic composition, antioxidant capacity, and color of coolseason legumes and other selected food legumes. J. Food Sci.72, S167–S177.

Yamamoto, N., Sawada, H., Izumi, Y., 2007. Proteasome inhibitioninduces glutathione synthesis and protects cells fromoxidative stress: relevance to Parkinson disease. J. Biol. Chem.282, 4364–4372.