FEBS Letters 582 (2008) 4185–4190

Alcohol enhances Ab42-induced neuronal cell deaththrough mitochondrial dysfunction

Do Yeon Leea,b,1, Kyu-Sun Leec,1, Hyun Jung Leea, Hee-Yeon Jungd,Jun Young Leed, Sang Hyung Leee, Young Chul Younf, Kyung Mook Seog, Jang Han Leeh,

Won Bok Leea, Sung Su Kima,*

a Department of Anatomy and Cell biology, College of Medicine, Chung-Ang University, Seoul 156-756, Republic of Koreab Department of Herbal Resources Research, Korea Institute of Oriental Medicine, Daejeon 305-811, Republic of Korea

c Center for Regenerative Medicine, Korea Research Institute of Bioscience and Biotechnology (KRIBB), Daejeon 305-806, Republic of Koread Department of Psychiatry, College of Medicine, Seoul National University, Seoul 151-742, Republic of Korea

e Department of Neurosurgery, College of Medicine, Seoul National University, Seoul 151-742, Republic of Koreaf Department of Neurology, College of Liberal Arts, Chung-Ang University, Seoul 156-756, Republic of Korea

g Department of Rehabilitation medicine, College of Medicine, Chung-Ang University, Seoul 156-756, Republic of Koreah Department of Psychology, College of Liberal Arts, Chung-Ang University, Seoul 156-756, Republic of Korea

Received 5 October 2008; revised 2 November 2008; accepted 7 November 2008

Available online 19 November 2008

Edited by Jesus Avila

Abstract Mitochondrial dysfunction is a hallmark of beta-amy-loid (Ab)-induced neuronal toxicity in Alzheimer�s disease (AD).Epidemiological studies have indicated that alcohol consumptionplays a role in the development of AD. Here we show that alco-hol exposure has a synergistic effect on Ab-induced neuronal celldeath. Ab-treated cultured neurons displayed spontaneous gener-ation of reactive oxygen species (ROS), disruption of their mito-chondrial membrane potential, induction of caspase-3 and p53activities, and loss of cell viability. Alcohol exposure facilitatedAb-induced neuronal cell death. Our study shows that alcoholconsumption enhances Ab-induced neuronal cell death byincreasing ROS and mitochondrial dysfunction.Crown Copyright � 2008 Published by Elsevier B.V. on behalfof the Federation of European Biochemical Societies. All rightsreserved.

Keywords: : Alzheimer�s disease; Ab; Ethanol; Mitochondria;ROS

1. Introduction

Alzheimer�s disease (AD), a neurodegenerative disorder clin-

ically characterized by progressive dementia, is associated with

neurofibrillary tangles and amyloid beta (Ab) plaques [1,2].

Since the discovery of the 4 kDa Ab peptide, much research

has focused on understanding Ab toxicity and its relationship

with AD progression and pathogenesis. Ab peptides are

cleaved from amyloid precursor protein (APP) through the

sequential proteolysis of aspartyl b-secretase and presenilin-

dependent c-secretase in AD brains [3–5]. The progressive

accumulation of Ab aggregates eventually triggers a cascade

of cellular changes, including mitochondrial oxidative damage,

hyperphosporylation of tau, synaptic failure, and inflamma-

tion [6–8].

*Corresponding author. Fax: +82 2 826 1265.E-mail address: sungsu@cau.ac.kr (S.S. Kim).

1These authors contributed equally to this work.

0014-5793/$34.00 Crown Copyright � 2008 Published by Elsevier B.V. on beha

doi:10.1016/j.febslet.2008.11.007

Mitochondrial dysfunction in AD patients has been well

documented [6,7]. Several in vitro studies showed that Ab af-

fects mitochondrial DNA and proteins, leading to impair-

ments of the electronic transport chain (ETC) and,

ultimately, to mitochondrial dysfunction [9,10]. Recently,

Lustbader et al. [11] reported that Ab-binding alcohol dehy-

drogenase directly interacts with Ab in the mitochondria of

AD patients and transgenic mice, and that this interaction pro-

motes the leakage of reactive oxygen species (ROS), resulting

in mitochondrial dysfunction.

Investigations with animal models of fetal alcohol syndrome

(FAS) suggest that neuronal vulnerability to ethanol-induced

cell death may correlate with specific developmental events

[12]. Ethanol exposure is a consistent and reliable producer

of neuronal toxicity, particularly when the exposure occurs

during periods of enhanced neuronal vulnerability. Neuronal

dendritic shrinkage has been documented in alcoholics [13].

There does not appear to be a link between alcohol-related

brain damage and AD [14], although some studies suggest a

relationship between alcohol and aging [15].

To test the synergistic effect of ethanol in an Ab-rich envi-

ronment in cultured neurons, we co-treated human neuroblas-

toma cells with Ab and ethanol. Here, we demonstrate that

alcohol exposure is a risk factor facilitating Ab-induced neuro-

nal cell death. Alcohol exposure had a synergistic effect on Ab-

induced cell stress via mitochondrial dysfunction. Inhibition of

caspase-3 or p53 reduced the synergistic neurotoxicity of Aband ethanol, whereas inhibition of Akt facilitated the suscepti-

bility of cultured neurons to Ab and ethanol. Thus, our study

shows that alcohol consumption enhances Ab-induced neuro-

nal cell death by increasing ROS production and disruption of

mitochondrial integrity.

2. Materials and methods

2.1. Cell culture and pharmacological treatmentsSK-N-SH human neuroblastoma cells were cultured at 37 �C in Dul-

becco�s Modified Eagle�s Medium (DMEM) (Gibco-BRL) supple-mented with 10% heat-inactivated FBS in a humidified 95% air, 5%CO2 incubator. For aging, Ab1-42 (Biosource Inc.) was incubated

lf of the Federation of European Biochemical Societies. All rights reserved.

4186 D.Y. Lee et al. / FEBS Letters 582 (2008) 4185–4190

for 1 week at 37 �C before use. In experiments requiring a 48 hincubation with ethanol (Sigma), the culture plates were sealed withplastic tape and wrapped in plastic wrap to prevent ethanol evapora-tion.

2.2. Cell viability assayAb- and ethanol-treated cells were stained with 10 ll of alamarBlue

(Serotec) [16]. For assessment of apoptosis, cells were fixed with 4%paraformaldehyde and then stained with 8 mg/ml of Hoechst dye33258 (Sigma) for 5 min. Nuclear morphology was visualized using afluorescence microscope (IX70 microscope, Olympus).

2.3. Determination of ROS generationAb- or ethanol-induced hydrogen peroxide generation was measured

by incubating the cells with the fluorescent probe 2 0,7 0-dichlorofluores-cein diacetate (DCF-DA) (Sigma). DCF fluorescein intensity was mea-sured with a fluorometer (TECAN, GENios).

2.4. Effect of p53 antisense oligonucleotides and Akt inhibitor onneuronal viability

To access the role of p53 and Akt inhibitor on cell viability inAb- and ethanol-mediated neuronal cell death, antisense 20-merphosphorothioate oligonucleotides were synthesized and added tothe culture medium at a final concentration of 10 lM 6 h prior to

Fig. 1. Ab and EtOH synergistically enhance neuronal apoptosis in SK-N-SHwith the untreated control. (B) Morphological changes in cultured SK-N-SHinduction of caspase-3 activity. (D) Cells were stained with Hoechst 33258 snuclei condensation and fragmentation (arrow), compared with normal cell nthree independent experiments. Scale bar is 20 lm. *P < 0.05, **P < 0.001.

Ab and/or ethanol addition. The p53 sequence of the p53 antisenseolignucleotides (5 0-CGCTAGGATCTGACTGC-3 0) was complemen-tary to the translation initiation site of the human p53 mRNA. Aktinhibitor, LY294002 (CalBiochem), also pre-treated for 2 h beforeAb and/or ethanol addition in culture medium.

2.5. RT-PCR analysis and Western blottingTotal RNA was isolated from the cells by using Trizol Reagent

(Invitrogen) according to the manufacturer�s instructions. cDNA wassynthesized by using Superscript II Reverse Transcription system(Invitrogen). For RT-PCR, AccuPower PCR premix (Bioneer) wasmixed with each primer. Catalase primers 5 0-TTAATCCA-TTCGATCTCACC-3 0 (forward), 5 0-GGCGGTGAGTGTCAGGA-TAG-3 0 (reverse); Cu/ZnSOD primers 5 0-CAGTGCAGGT-CCTCACTTTA-3 0 (forward), 5 0-CCTGTCTTTGTACTTTCTTC-30

(reverse); MnSOD primers 5 0-GTTTTGGGGTATCTGGGCTC-3 0

(forward), 5 0-TCCCTGATTTGGACAAGCAGC-3 0 (reverse); GAP-DH-specific primers 5 0-GGGGCTCTCCAGAACATCAT-30 (for-ward), 5 0-AAGTGGTCGTTGAGGGCAAT-30 (reverse). Tomeasure the expression levels of p53 and phospho-Akt, immunoblot-ting was performed with anti-p53 and anti-phospho-Akt (Santa CruzBiotech, CA, USA). An anti-b-actin antibody (SantaCruz Biotecholo-gy; 1:1000 dilution) was used as a loading control. Protein preparationand immunoblotting were performed as described [16].

cells. (A) The relative cell viability was decreased in all cases comparedcells treated with Ab and/or EtOH. (C) Effect of Ab and EtOH on

taining. Dead cells were identified by morphological changes, such asuclei (arrowhead). Data were presented as means ± S.D. from at least

D.Y. Lee et al. / FEBS Letters 582 (2008) 4185–4190 4187

2.6. Analysis of mitochondrial membrane potential (DWm) and activityof caspase-3

Changes in DWm were estimated using tetramethylrhodamine ethylester (TMRE) (Molecular Probes). Carbonyl cyanide p-(trifluorometh-oxy)-phenylhydrazone (FCCP) was used as a positive control for com-plete mitochondrial depolarization. Caspase-3 activities were measuredas described [16].

2.7. Statistical analysisAll data are expressed as the means ± S.D. Statistical significance

was determined via ANOVA followed by Scheffe�s post hoc tests. AP-value of <0.05 was considered to be significant.

Fig. 2. Effects of Ab and EtOH on ROS production by andantioxidant enzyme mRNA expression in SK-N-SH cells. (A) Ab-and EtOH-induced ROS generation was measured by incubation withDCF-DA. Data were presented as means ± S.D. from three indepen-dent experiments. *P < 0.05. (B) The expression of antioxidant geneswas measured by RT-PCR analysis. GAPDH is used for a loadingcontrol.

3. Results

3.1. Characterization of the synergistic neurotoxicity effects of

combined treatment with Ab and ethanol

We first performed a series of experiments to evaluate the

neurotoxicity effects of single agent treatment with Ab or

ethanol on cell survival in human neuroblastoma cell line,

SK-N-SH. To confirm the synergistic cytotoxic effect of Aband ethanol, we also co-treated with Ab and ethanol under

the same conditions. At a concentration of 1 lM, Ab showed

no cell-killing effects, even after 48 h of treatment, compared

with un-treated control; however, this concentration did selec-

tively inhibit neurite growth. In contrast, cell viability decreased

with time when cells were treated with 15 mM ethanol, which is

a similar concentration of the drink/drive limit in the USA and

UK (0.08% w/v, 17.4 mM) (Fig. 1A). Interestingly, co-treat-

ment with Ab and ethanol showed a significant synergistic effect

on cell death compared to treatment with Ab or ethanol alone

for 48 h (Fig. 1A). This synergistic effect on neuronal cell death

was also detected in the morphological changes (Fig. 1B and D)

and caspase-3 activation (Fig. 1C).

3.2. Synergistic effects of Ab and ethanol in triggering oxidative

stress

To determine whether Ab and/or ethanol treatment induces

oxidative stress in cultured neurons, we measured the level of

ROS production using a DCF-DA-dependent fluorescence as-

say. When SK-N-SH neurons were exposed to Ab or ethanol

alone, intracellular ROS levels did not significantly change.

However, ROS production was significantly increased in co-

treated with both Ab and ethanol (Fig. 2A).

To determine whether a combined treatment with Ab and

ethanol disturbs the internal defense mechanism for ROS, we

examined the expression of antioxidant genes. ROS are regu-

lated by endogenous antioxidant defense systems, which in-

clude catalase, superoxide dismutase (SOD), and glutathione

peroxidase [17]. Of these antioxidant genes, catalase and mito-

chondrial manganese-containing MnSOD, but not cytosolic

Cu/ZnSOD, decreased following combination treatment with

Ab and ethanol (Fig. 2B). Expression of catalase and MnSOD

were decreased by 3-fold and 5-fold, respectively. These results

indicate that Ab and ethanol have a synergistic effect on ROS

generation in SK-N-SH cells.

3.3. Synergistic effects of Ab and ethanol on apoptotic or

survival pathways

To investigate whether Ab and ethanol have a synergistic ef-

fect on apoptosis, we measured the expression of pro-apoptotic

marker protein p53. Treatment of Ab or ethanol alone did not

cause a change in p53 expression. In contrast, p53 was signif-

icantly increased after 3 h following co-treatment of cells with

Ab and ethanol (Fig. 3A). To further demonstrate that p53

induction correlates with cell death, we blocked the expression

of the p53 protein by pre-incubating with p53-specific anti-

sense oligonucleotides. Inhibition of p53 (p53 AS) significantly

rescued the synergistic cell death effect of Ab and ethanol in

SK-N-SH cells (Fig. 3B).

We also investigated whether a combined treatment with Aband ethanol influences neuronal cell survival signaling by mea-

suring the activity of Akt with a phospho-specific Akt anti-

body. The level of pAkt in neurons was significantly

decreased after 3 h with a combined treatment with Ab and

ethanol (Fig. 3C). In addition, pre-treatment of Akt specific

inhibitor, LY294002, significantly enhanced the synergistic

Ab- and ethanol-induced neuronal cell death (Fig. 3D). To-

gether, these findings suggest that Ab and ethanol show syner-

gistic effects on neuronal cell death through activating the p53-

dependent apoptotic pathway and inhibiting Akt-dependent

survival signaling.

3.4. Effects of p53 antisense oligonucleotide and pAkt inhibitor

on ROS generation

To demonstrate that p53 and Akt signaling directly influence

the ROS production, we pre-treated cultured neurons with a

p53 antisense oligonucleotide or an Akt inhibitor before mea-

Fig. 3. p53 and pAkt are involved in Ab- and EtOH-induced neuronal cell death in SK-N-SH cells. Western blot analysis was performed using anti-p53 (A) and anti-pAkt (C) antibodies. Beta-actin is used for a protein loading control. Involvement of p53 and pAkt in Ab- and EtOH-induced cellviability. Cells were pre-treated with p53 antisense oligonucleotides (B, open squares) or with LY294002 (D, open squares). Data were represented asmeans ± S.D. from three independent experiments. *P < 0.05 versus vehicle alone, Ab alone, or EtOH alone; **P < 0.05 versus Ab/EtOH together.

4188 D.Y. Lee et al. / FEBS Letters 582 (2008) 4185–4190

suring the ROS level. As above, we confirmed that combined

treatment with Ab and ethanol significantly increased ROS

production in SK-N-SH cells (Fig. 2A). Inhibition of p53 sig-

nificantly reduced ROS production in Ab plus ethanol-treated

cells, similar to the positive control group where cells were pre-

treated for 2 h with a well-known glutathione precursor, NAC

(Fig. 4A). Glutathione plays an important role in the detoxifi-

cation of ROS in brain [18]. In contrast, Akt inhibition did not

significantly change ROS production.

3.5. Effects of p53 antisense oligonucleotide and pAkt inhibitor

on mitochondrial membrane potential

To investigate the synergistic effect of Ab and ethanol on

mitochondrial integrity, we examined the mitochondrial mem-

brane potential (DWm) with TMRE staining. A slight disrup-

tion of DWm was observed in neurons treated with Ab or

ethanol alone. Co-treatment with Ab and ethanol triggered

the collapse of DWm, whereas pre-treatment with a p53 anti-

sense oligonucleotide significantly blocked DWm depolarization

(Fig. 4B). In contrast, Akt inhibition slightly enhanced the

depolarization of DWm in cultured neurons co-treated with

Ab and ethanol. Treatment with NAC, the p53 antisense oligo-

nucleotide, or the Akt inhibitor alone did not affect DWm.

These results indicate that Ab and ethanol synergistically in-

duce depolarization of DWm. Pre-treatment with p53 antisense

oligonucleotides attenuated and an Akt inhibitor enhanced

this synergistic effect on DWm reduction (Fig. 4B), indicating

that p53 and/or Akt plays a role in Ab- and ethanol-induced

depolarization of DWm.

4. Discussion

The results of this study are the first to show the synergistic

effects of ethanol exposure and an Ab-rich environment, rela-

tive to ROS production, disruption of mitochondrial integrity,

and cytotoxicity. In late-onset, sporadic AD, an age-dependent

increase of ROS has been identified as a key factor in the devel-

opment and progression of AD [1,8]. In this study, we suggest

that alcohol poisoning increases the risk of sporadic AD, with

damage to the neuronal tissues or cells through enhanced ROS

production.

It is well recognized that one of the main effects of ROS

exposure is damage to DNA molecules. Among the proteins

that play an important role in the cellular response to DNA

damage is the tumor suppressor p53 [19]. Neuronal cells in

which p53 was inhibited were less vulnerable to ROS-induced

Fig. 4. Effect of p53 and pAkt on Ab- and EtOH-induced synergistic induction of ROS production and loss of mitochondrial membrane potential(DWm) in SK-N-SH cells. ROS (A) and DWm (B) were analyzed using DCF-DA and TMRE. Data were presented as means ± S.D. from threeindependent experiments. FCCP was used as a positive control for complete mitochondrial depolarization. *P < 0.05 versus vehicle alone, Ab alone,or EtOH alone; **P < 0.05 versus Ab/EtOH together.

D.Y. Lee et al. / FEBS Letters 582 (2008) 4185–4190 4189

oxidative injury. This protective mechanism involved the

impairment of ROS-activated, p53-dependent apoptosis.

Over the past decade, the Akt pathway has been shown to be

sufficient, and in some cases necessary, for the growth factor-

induced cell survival of several neuronal cell types [20]. Akt

may induce the expression of survival genes, such as Bcl-XL

and A1, by activating CREB or NF-jB, and directly inhibits

members of the apoptotic machinery [20]. In this study, we

show that neuronal cell death induced by Ab and ethanol

through decreasing Akt activity.

Recently, Smith et al. [21] proposed that Ab toxicity is elic-

ited through the generation of ROS, leading to activation of

the JNK pathway, which in turn activates p66Shc by phos-

phorylation at Ser36. Activated p66Shc then triggers the phos-

phorylation of FKH, thereby down-regulating target genes,

such as MnSOD, and leading to an even greater accumulation

of cellular ROS. This biochemical cycle causes cell death. Here,

we found that MnSOD expression was significantly decreased

by combined treatment with Ab and ethanol.

Mitochondria are unique organelles in the consumption of

oxygen, production of ATP, generation of oxygen radicals,

and mobilization of calcium [22]. Mitochondrial dysfunction

is evaluated by multiple parameters of mitochondrial integrity.

Regulation of changes in mitochondrial membrane potential

(DWm), the release of cytochrome c from the mitochondria into

the cytoplasm and the activation of caspase-9 are important

processes during mitochondria-dependent apoptotic process

[23]. In this study, depolarization of DWm was increased by

over half in combined treatment with Ab and ethanol.

This study demonstrates that a low-dose of Ab or ethanol

alone does not affect cell viability, whereas the combined expo-

sure to both two agents activates apoptotic pathways through

ROS production and the disruption of mitochondrial integrity.

This study also indicates that the p53 and Akt pathways con-

tribute to pro- or anti-apoptosis in cultured neurons.

Acknowledgement: This work was supported by Grant(20080401034034) from BioGreen 21 Program, Rural DevelopmentAdministration, Republic of Korea.

4190 D.Y. Lee et al. / FEBS Letters 582 (2008) 4185–4190

References

[1] Mattson, M.P. (2004) Pathways towards and away from Alzhei-mer�s disease. Nature 430, 631–639.

[2] Tanzi, R.E. and Bertram, L. (2005) Twenty years of theAlzheimer�s disease amyloid hypothesis: a genetic perspective.Cell 120, 545–555.

[3] Hardy, J. and Selkoe, D.J. (2002) The amyloid hypothesis ofAlzheimer�s disease: progress and problems on the road totherapeutics. Science 297, 353–356.

[4] Sisodia, S.S., Annaert, W., Kim, S.H. and De Strooper, B. (2001)Gamma-secretase: never more enigmatic. Trends Neurosci. 24, S2–S6.

[5] Walsh, D.M., Klyubin, I., Fadeeva, J.V., Cullen, W.K., Anwyl,R., Wolfe, M.S., Rowan, M.J. and Selkoe, D.J. (2002) Naturallysecreted oligomers of amyloid beta protein potently inhibithippocampal long-term potentiation in vivo. Nature 416, 535–539.

[6] Hirai, K. et al. (2001) Mitochondrial abnormalities in Alzhei-mer�s disease. J. Neurosci. 21, 3017–3023.

[7] Reddy, P.H. (2006) Amyloid precursor protein-mediated freeradicals and oxidative damage: implications for the developmentand progression of Alzheimer�s disease. J. Neurochem. 96, 1–13.

[8] Tanzi, R.E. (2005) The synaptic Abeta hypothesis of Alzheimerdisease. Nat. Neurosci. 8, 977–979.

[9] Cardoso, S.M., Santos, S., Swerdlow, R.H. and Oliveira, C.R.(2001) Functional mitochondria are required for amyloid beta-mediated neurotoxicity. FASEB J. 15, 1439–1441.

[10] Casley, C.S., Canevari, L., Land, J.M., Clark, J.B. and Sharpe,M.A. (2002) Beta-amyloid inhibits integrated mitochondrialrespiration and key enzyme activities. J. Neurochem. 80, 91–100.

[11] Lustbader, J.W. et al. (2004) ABAD directly links Abeta tomitochondrial toxicity in Alzheimer�s disease. Science 304, 448–452.

[12] Berman, R.F. and Hannigan, J.H. (2000) Effects of prenatalalcohol exposure on the hippocampus: spatial behavior, electro-physiology, and neuroanatomy. Hippocampus 10, 94–110.

[13] Harper, C. and Corbett, D. (1990) Changes in the basal dendritesof cortical pyramidal cells from alcoholic patients—a quanti-tative Golgi study. J. Neurol. Neurosurg. Psychiatr. 53, 856–861.

[14] Morikawa, Y. et al. (1999) Cerebrospinal fluid tau protein levelsin demented and nondemented alcoholics. Alcohol Clin. Exp.Res. 23, 575–577.

[15] Caine, D., Halliday, G.M., Kril, J.J. and Harper, C.G. (1997)Operational criteria for the classification of chronic alcoholics:identification of Wernicke�s encephalopathy. J. Neurol. Neuro-surg. Psychiatr. 62, 51–60.

[16] Lee do, Y. et al. (2008) Kynurenic acid attenuates MPP(+)-induced dopaminergic neuronal cell death via a Bax-mediatedmitochondrial pathway. Eur. J. Cell Biol. 87, 389–397.

[17] Finkel, T. and Holbrook, N.J. (2000) Oxidants, oxidative stressand the biology of ageing. Nature 408, 239–247.

[18] Dringen, R. (2000) Metabolism and functions of glutathione inbrain. Prog. Neurobiol. 62, 649–671.

[19] Grilli, M. and Memo, M. (1999) Possible role of NF-kappaB andp53 in the glutamate-induced pro-apoptotic neuronal pathway.Cell Death Differ. 6, 22–27.

[20] Brunet, A., Datta, S.R. and Greenberg, M.E. (2001) Transcrip-tion-dependent and -independent control of neuronal survival bythe PI3K-Akt signaling pathway. Curr. Opin. Neurobiol. 11, 297–305.

[21] Smith, W.W., Norton, D.D., Gorospe, M., Jiang, H., Nemoto, S.,Holbrook, N.J., Finkel, T. and Kusiak, J.W. (2005) Phosphor-ylation of p66Shc and forkhead proteins mediates Abeta toxicity.J. Cell Biol. 169, 331–339.

[22] Melov, S. (2000) Mitochondrial oxidative stress. Physiologicconsequences and potential for a role in aging. Ann. NY Acad.Sci. 908, 219–225.

[23] Kroemer, G. and Reed, J.C. (2000) Mitochondrial control of celldeath. Nat. Med. 6, 513–519.