Oxidative stress, a deleterious processes resulting from animbalance between formation of ROS and antioxidative de-fense, is considered to play a critical role in the pathogenesisof various neurodegenerative diseases that are characterizedby selective neuronal death.1—3) Astrocytes support neuronalsurvival through inactivation of ROS by the activity of an-tioxidant and glutathione (GSH).4—8) Deficiency of antioxi-dantive function in astrocyte has been implicated in manyneurodegenerative processes.9—15) Neuronal loss occurred inspinal cord, such as amyotrophic lateral sclerosis (ALS) andspinalmuscularatrophy (SMA), has been attributed to the de-terioration of the ability of spinal cord astrocytes to supportneurons.16,17) On the other hand, compromised astrocytesfunction may increase the neuron’s susceptibility to oxidativestress.18) The involvement of ROS in neuronal toxicity andthe important role of astrocytes in neurodegenerativeprocesses make it important to determine the detoxificationpathways and regulation of antioxidants in astrocytes. An in-creasing number of antioxidants and phase II enzymes arefound to be essential in detoxification of reactive oxygenspecies.19)

Antioxidant response element (ARE) is an enhancer ofmany phase II antioxidant enzymes genes, such as heme oxygenase 1 (HO1), glutamate cysteine ligase (GCL) and reduced nicotinamide adenine dinucleotide phosphate(NAD(P)H): quinine oxidoreductase 1 (NQO1).20) Nuclearfactor erythroid 2-related factor 2 (Nrf2) is normally boundto its suppressor keap1 in cytoplasm but is freed from the lat-ter under certain conditions, and free Nrf2 translocates to nu-cleus. Once in the nucleus, Nrf2 binds to ARE and drives thetranscription of downstream genes.21)

Curcumin (1,7-bis[4-hydroxy-3-methoxyphenyl]-1,6-hep-tadiene-3,5-dione), a member of the curcuminoid family, isthe major active component of turmeric, a yellow compoundoriginally isolated from the plant Curcuma longa L. Cur-cumin has a number of biological and pharmacological activ-ities such as anti-carcinogen, anti-angiogenic22,23) anti-malar-ial,24) antioxidant, anti-mutagenic, antibacterial,25) immune-

modulatory,26) and anti-inflammatory.27) However, whether itactivates Nrf2 cytoprotective signaling and reduces the sus-ceptibility of spinal cord astrocytes to oxidative damage havenot been previously reported.

MATERIALS AND METHODS

Animals and Genotyping Genotypes (Nrf2�/� andNrf2�/�) of the animals were determined by polymerasechain reaction (PCR) amplification of genomic DNA fromtails. PCR amplification was carried out using three differentprimers, 5�-TGGACGG-GACTATTGAAGGCTG-3� (sensefor Nrf2�/� and Nrf2�/�), 5�-GCCGCCTTTTCAGTAGATG-GAGG-3� (antisense for Nrf2�/�) and 5�-GCGGATTGAC-CGTAATGGGATAGG-3� (antisense for Nrf2�/�). The geno-types of mice were verified by examining the sizes of thePCR products: Nrf2�/� (700 bp) and Nrf2�/� (400 bp).

Primary Cell Culture Primary astrocytes were preparedfrom spinal cords of postnatal day 1 to 3 Nrf2�/� male miceand Nrf2�/� male mice. In brief, animals were killed. Spinalcords were removed and placed in Ca2� and Mg2�-free phos-phate buffered saline pH 7.4. Spinal cord pieces weretrypsined for 15 min and mechanically dissociated. Aftercentrifugation at 300 g for 5 min, the pellets were re-sus-pended in Dulbecco’s modified Eagle’s medium (DMEM)(pH 7.4) supplemented with 100 U/ml penicillin, 100 mg/mlstreptomycin. The cells were seeded (at a density of 2�105

cells per cm2) into cell culture flasks.Curcumin and H2O2 Treatment In the solvent group,

cells were incubated in serum-free DMEM with 0.1% di-methylsulfoxide (DMSO) for 24 h. In the curcumin groups,cells were pre-incubated in serum-free DMEM with differentconcentrations of curcumin for 24 h. In the H2O2 plus cur-cumin group, cells were first pretreated with curcumin for24 h, followed by washing and then treatment with H2O2 inserum-free DMEM for 30 min.

Evaluation of Intracellular ROS Production and Via-bility Assay Intracellular ROS production was detected

1194 Vol. 34, No. 8Regular Article

Activation of Nuclear Factor Erythroid 2-Related Factor 2 CytoprotectiveSignaling by Curcumin Protect Primary Spinal Cord Astrocytes againstOxidative Toxicity

Hong JIANG,a,# Xinying TIAN,a,b,# Yansu GUO,a,b Weisong DUAN,a,b Hui BU,a,b and Chunyan LI*,a,b

a Department of Neurology, the Second Hospital of Hebei Medical University; Shijiazhuang, Hebei 050000, People’sRepublic of China: and b Institute of Cardiocerebrovascular Disease; Shijiazhuang, Hebei 050000, People’s Republic ofChina. Received December 1, 2010; accepted March 19, 2011; published online May 31, 2011

Oxidative damage plays a critical role in many neurodegenerative diseases. Astrocytes are involved in sup-porting the survival and protection of neurons against oxidative damage. The dysfunction of antioxidant in astro-cytes has been implicated in a variety of neurodegenerative disorders, such as amyotrophic lateral sclerosis(ALS), spinalmuscularatrophy (SMA). The loss of motor neuron in spinal cord has been attributed to deteriora-tion of astrocytes. The activation of antioxidantive function in astrocytes may serve as a therapeutic strategy forneurodegenerative diseases. Nuclear factor erythroid 2-related factor 2 (Nrf2) is a master transcriptional regula-tor of phase II antioxidantive genes. We report herein that curcumin significantly activates Nrf2 target genes inprimary spinal cord astrocytes, decreases the level of intracellular reactive oxygen species (ROS), and attenuatesoxidative damage and mitochondrial dysfunction.

Key words nuclear factor erythroid 2-related factor 2; curcumin; astrocyte; oxidative stress; reactive oxygen species

Biol. Pharm. Bull. 34(8) 1194—1197 (2011)

© 2011 Pharmaceutical Society of Japan∗ To whom correspondence should be addressed. e-mail: hblicy5@yahoo.com.cn# These authors contributed equally to this work.

using the nonfluorescent cell permeating compound, 2�,7�-dichlorofluorescein diacetate (DCF-DA). Astrocytes weretreated with DCF-DA (10 mM) for 30 min at 37 °C and rinsedwith serum-free DMEM. The cell viability assay involves theuse of 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxy-phenyl)-2-(4-sulfophenyl)-2H-tetrazolium inner salt (MTS),which forms a formazan product in cells.

Detection of Cytotoxicity by Measuring Lactate Dehy-drogenase (LDH) Release and Flow Cytometry Cytotox-icity was evaluated by measuring LDH release from the cy-tosol of damaged cells into the medium using a standard col-orimetric assay procedures. Mitochondrial transmembranepotential were measured using flow cytometry. One millilitercell suspension (106 cells) was incubated with rhodamin 123at 37 °C for 30 min.

Statistical Analysis Numerical data are expressed asmean�S.E.M. Differences between mean values of multiplegroups were analyzed by one-way analysis of variance(ANOVA) followed by Student–Newman–Keuls multiplerange test. Statistical significance was considered at p�0.05.

RESULTS

Curcumin Activates Nrf2 and Nrf2 Target Genes in As-trocytes As shown in Fig. 1A, curcumin at 5, 10, 15 mM

caused significant increase in expression of three Nrf2 targetgenes in primary Nrf2�/� astrocytes. In contrast, curcumintreatment did not result in significant increases in Nrf2 targetgenes in Nrf2�/� astrocytes in Fig. 1B. Curcumin caused sig-nificant increase in nuclear Nrf2 in primary Nrf2�/� astro-cytes, while the total Nrf2 was not significantly altered asshown in Fig. 1C. Nrf2�/� cells in the control group showedthe cytosolic distribution pattern of Nrf2, as the green fluo-

rescence of Nrf2 was mainly confined in the cytoplasm. Cur-cumin treatment converted the distribution pattern, as Nrf2accumulated in the nucleus as shown in Fig. 2. These resultsshow that not only is the Nrf2-ARE signaling pathway re-sponsive in astrocytes but also it can be significantly acti-vated by curcumin.

Curcumin Inhibited Intracellular ROS Production In-duced by H2O2 in Nrf2��/�� Astrocytes, while Its ActivityWas Abolished in Nrf2��/�� Astrocytes Thirty minutesafter the insult with 200 mM H2O2, we observed an increase inROS production in both Nrf2�/� astrocytes and Nrf2�/� as-trocytes. But ROS production in Nrf2�/� astrocytes washigher than in Nrf2�/� astrocytes. Addition of 10 mM cur-cumin prevented this alteration in Nrf2�/� astrocytes, as ROSproduction decreased significantly. In contrast, no significantalterations were observed in Nrf2�/� cells, as shown in Fig.3. Drug treatment as described (see Fig. 3C for treatmentschedules).

Decreased Oxidative Damage in Curcumin-Treated As-trocytes Is Accompanied by Activation of Nrf2 Signalingand Protection of Mitochondrial Damage As shown inFig. 4A, exposure of Nrf2�/� astrocytes and Nrf2�/� astro-cytes to H2O2 resulted in a substantial loss in the protein lev-els of HO1 and NQO1. But pretreatment of these cells withcurcumin significantly attenuated this decrease. In contrast,curcumin pretreatment was ineffective in Nrf2�/� astrocytes.To gain further insight into the protective activity of cur-cumin, we measured mitochondrial integrity in Nrf2�/� as-trocytes and Nrf2�/� astrocytes. H2O2 treatment caused sig-nificant loss of mitochondrial trans-membrane potential inboth in Nrf2�/� astrocytes and Nrf2�/� astrocytes. Cells werefirst treated with curcumin at 10 mM for 24 h before combinedtreatment with H2O2. We found that curcumin efficiently pre-vented the loss of mitochondrial trans-membrane potential inNrf2�/� astrocytes, while failed to do so in Nrf2�/� astro-cytes, as shown in Fig. 4B.

Nrf2 Is a Crucial Determinant of Susceptibility of As-trocytes to Oxidative Stress, and Curcumin Decreases theSusceptibility to Oxidative Stress and Protects againstH2O2-Induced Oxidative Damage As shown in Fig. 5,H2O2 treatment markedly reduced the viability and increasedLDH release from Nrf2�/� astrocytes and Nrf2�/� astrocytes,

August 2011 1195

Fig. 1. Effect of Curcumin on Nrf2 and Nrf2 Target Genes in Nrf2�/� As-trocytes and Nrf2�/� Astrocytes

(A) Nrf2�/� astrocytes were incubated with solvent (0.1% DMSO) or 5, 10, 15, 20and 25 mM curcumin for 24 h. Cell lysates were examined for levels of HO1, NQO1,GCLc by Western blotting, b-actin was used as control. (B) Total cell lysates fromNrf2�/� astrocytes were prepared after treatment with curcumin at 5 or 10 mM for 24 h.Immunoblots were probed with Nrf2 specific antibody. (C) Cell lysates and nuclear ex-tract were probed with Nrf2 specific antibody for Western blotting. # Significantly dif-ferent from control.

Fig. 2. Curcumin Induced the Translocation of Nrf2 into Nucleus inNrf2�/� Astrocytes

Cells in the control group showed the cytosolic distribution pattern of Nrf2. Thegreen fluorescence of Nrf2 was mainly confined in the cytoplasm. Curcumin treatmentconverted the distribution pattern Nrf2 move to the nucleus.

but Nrf2�/� astrocytes were more susceptible than Nrf2�/�

astrocytes. Thus, Nrf2�/� cells exhibited a significantly in-creased sensitivity to H2O2-induced oxidative damage. Pre-treatment of Nrf2�/� astrocytes with 10 mM curcumin for 24 hled to a significant protection against and decreased the sen-sitivity to H2O2-induced oxidative stress. In contrast, thesame curcumin pretreatment in Nrf2�/� cells resulted in ei-ther none or only a slight cytoprotection against the cell in-jury and led to no significant alteration in the sensitivity to

H2O2-induced oxidative stress.

DISCUSSION

Phase II enzymes play an important role in protecting cellsagainst stress through the removal of free radicals and detoxi-cation of toxins. HO1 has been recognized as an importantcytoprotective enzyme in addition to its metabolic role in the breakdown of heme. NQO1, an enzyme that detoxifiesquinones, is reported to have antiapoptotic effects. GSH isthe most abundant low-molecular-weight thiol that regulatesthe redox state of the cell. The synthesis of GSH involvesGCLC.

Our study demonstrated that incubation of Nrf2�/� astro-cytes with curcumin led to a significant induction of phase IIenzymes. The up-regulation of phase II enzymes was abol-ished in Nrf2�/� cells. This observation suggested that Nrf2cytoprotective signaling can be significantly activated by curcumin in astrocytes. Curcumin effectively intercepts andneutralizes ROS induced by H2O2, which is accompanied byactivation of Nrf2 signaling.

In normally situation, Nrf2 resides in the cytoplasm as partof a complex with the redox-sensitive protein Keap-1. After ashort period, this Nrf2 is ubiquitinated and released fromKeap-1 for proteasomal degradation. After the curcumintreatment, Keap-1 releases Nrf2, allowing it to translocate tothe nucleus. In the nucleus, Nrf2 binds to ARE containingpromoter of many phase II antioxidant enzymes genes. Theseenzymes and molecules act together to reduce the ROS loadand to restore and maintain redox balance.

Many observations suggest that the lack of Nrf2 in astro-cytes may be a common contributing factor to neurodegener-ative disorders and the activation of Nrf2 in astrocytes confer

1196 Vol. 34, No. 8

Fig. 3. Effect of Curcumin on Intracellular ROS Accumulation

(A) ROS production decreased significantly in Nrf2�/� astrocytes. In contrast, nosignificant alterations were observed in Nrf2�/� cells. All images are representativefields of at least three independent experiments (�10, scale bar, 50 mm). (B) The fluo-rescence was quantified. # Significantly different from control. (C) Drug treatmentschedule. After one week of culture, primary astrocytes were treated with solvent (0.1%DMSO), 200 mM H2O2 or a combination of H2O2 (200 mM) and curcumin (10 mM). In thecombination treatment, cells were first incubated with curcumin for 24 h before switch-ing to the combination treatment.

Fig. 4. Decreased Oxidative Damage in Curcumin-Treated Astrocytes IsAccompanied by Activation of Nrf2 Signaling and Mitochondrial Protection

Nrf2�/� astrocytes and Nrf2�/� astrocytes were treated with solvent (0.1% DMSO),200 mM H2O2 or combination of H2O2 plus curcumin (10 mM). Cell lysates were exam-ined for levels of HO1 and NQO1 by Western blotting, b-actin was used as a control(A). Mitochondrial trans-membrane potential was measured in astrocytes by flow cytometry (B). # Significantly different from control.

Fig. 5. Nrf2 Modulates the Susceptibility to Oxidative Stress and Cur-cumin Decreases the Susceptibility of Astrocytes

Nrf2�/� astrocytes and Nrf2�/� astrocytes were incubated with or without 10 mM cur-cumin for 24 h, followed by incubation with various concentrations of H2O2 for another12 h. After this incubation, cytotoxicity was determined by MTT reduction (A) andLDH release assay (B). # Significantly different from Nrf2�/�. ∗ Significantly differentfrom Nrf2�/�.

significant protection to neurons in vivo.28—31) Chemical/drugs that activate this pathway may have efficacy in block-ing neuronal cell death. A focused effort to find those thatcross the blood–brain barrier efficiently, activate the pathwayin astrocytes, and have efficacy against neurotoxicity is un-derway.

Activation of phase II enzymes by curcumin offer a greatadvantage for therapeutic purposes, as curcumin could be-come part of the human diet and be consumed daily as herbalsupplements. It was low toxicity and able to enter the centralnervous system. Curcumin activates of Nrf2-ARE pathway inspinal cord astrocytes may serve as a therapeutic strategy forneurodegenerative diseases.

Acknowledgments This work was supported in part byGrants from the Natural Science Foundation of China(30670732, 30870882) and a Grant from the Science andTechnological Department (08966105D).

REFERENCES

1) Butterfield D. A., Lauderback C. M., Free Radic. Biol. Med., 32,1050—1060 (2002).

2) Zhu X., Raina A. K., Lee H. G., Casadesus G., Smith M. A., Perry G.,Brain Res., 1000, 32—39 (2004a).

3) Cen L., Hutzen B., Ball S., DeAngelis S., Chen C. L., Fuchs J. R., LiC., Li P. K., Lin J., BMC Cancer, 30, 9—99 (2009).

4) Desagher S., Glowinski J., Premont J., J. Neurosci., 16, 2553—2562(1996).

5) Jakel R. J., Kern J. T., Johnson D. A., Johnson J. A., Toxicol. Sci., 87,176—186 (2005).

6) Peuchen S., Bolaños J. P., Heales S. J., Almeida A., Duchen M. R.,Clark J. B., Prog. Neurobiol., 52, 261—281 (1997).

7) Shih A. Y., Johnson D. A., Wong G., Kraft A. D., Jiang L., Erb H.,Johnson J. A., Murphy T. H., J. Neurosci., 23, 3394—3406 (2003).

8) Takeshima T., Johnston J. M., Commissiong J. W., J. Neurosci., 14,4769—4779 (1994).

9) Lee J. M., Calkins M. J., Chan K., Kan Y. W., Johnson J. A., J. Biol.Chem., 278, 12029—12038 (2003).

10) Ishii T., Itoh K., Yamamoto M., Methods Enzymol., 348, 182—190(2002).

11) Chen P. C., Vargas M. R., Pani A. K., Smeyne R. J., Johnson D. A.,Kan Y. W., Johnson J. A., Proc. Natl. Acad. Sci. U.S.A., 106, 2933—2938 (2009).

12) Makar T. K., Nedergaard M., Preuss A., Gelbard A. S., Perumal A. S.,Cooper A. J., J. Neurochem., 62, 45—53 (1994).

13) Mishra S., Karmodiya K., Surolia N., Surolia A., Bioorg. Med. Chem.,16, 2894—2902 (2008).

14) Nagai M., Re D. B., Nagata T., Chalazonitis A., Jessell T. M.,Wichterle H., Przedborski S., Nat. Neurosci., 10, 615—622 (2007).

15) Parvathy K. S., Negi P. S., Srinivas P., Food Chem., 115, 265—271(2009).

16) Jeffrey A., Delinda A., Lee J. M., Int. Congr. Ser., 1302, 143—153(2007).

17) Stewart V. C., Heales S. J. R., Free Radic. Biol. Med., 34, 287—303(2003).

18) Kraft A. D., Johnson D. A., Johnson J. A., J. Neurosci., 24, 1101—1112 (2004).

19) Lee J. M., Johnson J. A., J. Biochem. Mol. Biol., 37, 139—143 (2004).20) Wilson J. X., Can J., Physiol. Pharmacol., 75, 1149—1163 (1997).21) Itoh K., Chiba T., Takahashi S., Ishii T., Igarashi K., Katoh Y., Oyake

T., Hayashi N., Satoh K., Hatayama I., Yamamoto M., Nabeshima Y.,Biochem. Biophys. Res. Commun., 236, 313—322 (1997).

22) Zhu X., Raina A. K., Perry G., Smith M. A., Lancet Neurol., 3, 219—226 (2004b).

23) Yoysungnoen P., Wirachwong P., Changtam C., Suksamrarn A., Patum-raj S., World J. Gastroenterol., 14, 2003—2009 (2008).

24) Simonian N. A., Coyle J. T., Annu. Rev. Pharmacol. Toxicol., 36, 83—106 (1996).

25) Shih A. Y., Imbeault S., Barakauskas V., Erb H., Jiang L., Li P., Mur-phy T. H., J. Biol. Chem., 280, 22925—22936 (2005).

26) Allam G., Immunobiology, 214, 712—727 (2009).27) Chau L. T., Choi Y. L., Kok W. L., Eur. J. Pharmacol., 628, 247—254

(2010).28) Calkins M. J., Jakel R. J., Johnson D. A., Chan K., Kan Y. W., Johnson

J. A., Proc. Natl. Acad. Sci. U.S.A., 102, 244—249 (2005).29) Calkins M. J., Vargas M. R., Johnson D. A., Johnson J. A., Toxicol.

Sci., 115, 557—568 (2010).30) Jakel R. J., Townsend J. A., Kraft A. D., Johnson J. A., Brain Res.,

1144C, 192—201 (2007).31) Takuma K., Baba A., Matsuda T., Prog. Neurobiol., 72, 111—127

(2004).

August 2011 1197