Curcumin protects neurons against ... - Wiley Online Library

Journal of Neuroscience Research 92:1549–1559 (2014)

Curcumin Protects Neurons Against Oxygen-Glucose Deprivation/ReoxygenationInduced Injury Through Activation of Peroxisome Proliferator-Activated Receptor-c Function Zun-Jing Liu,1 Hong-Qiang Liu,2 Cheng Xiao,3* Hui-Zhen Fan,4 Qing Huang,5 Yun-Hai Liu,5 and Yu Wang1 1

Department of Neurology, China-Japan Friendship Hospital, Beijing, China Department of Pharmacy, Jining No.1 People’s Hospital, Jining, China 3 Institute of Clinical Medicine, China-Japan Friendship Hospital, Beijing, China 4 Department of Gastroenterology, The People’s Hospital of Yichun City, Yichun, China 5 Department of Neurology, Xiangya Hospital Central-South University, Changsha, China 2

The turmeric derivative curcumin protects against cerebral ischemic injury. We previously demonstrated that curcumin activates peroxisome proliferator-activated receptor-g (PPARg), a ligand-activated transcription factor involved in both neuroprotective and anti-inflammatory signaling pathways. This study tested whether the neuroprotective effects of curcumin against oxygen–glucose deprivation/reoxygenation (OGD/R)-induced injury of rat cortical neurons are mediated (at least in part) by PPARg. Curcumin (10 lM) potently enhanced PPARg expression and transcriptional activity following OGD/R. In addition, curcumin markedly increased neuronal viability, as evidenced by decreased lactate dehydrogenase release and reduced nitric oxide production, caspase-3 activity, and apoptosis. These protective effects were suppressed by coadministration of the PPARg antagonist 2-chloro-5nitrobenzanilide (GW9662) and by prior transfection of a small-interfering RNA (siRNA) targeting PPARg, treatments that had no toxic effects on healthy neurons. Curcumin reduced OGD/R-induced accumulation of reactive oxygen species and inhibited the mitochondrial apoptosis pathway, as indicated by reduced release of cytochrome c and apoptosis-inducing factor and maintenance of both the mitochondrial membrane potential and the Bax/Bcl-2 ratio. Again, GW9662 or PPARg siRNA transfection mitigated the protective effects of curcumin on mitochondrial function. Curcumin suppressed IjB kinase phosphorylation and IjB degradation, thereby inhibiting nuclear factor-j B (NF-jB) nuclear translocation, effects also blocked by GW9662 or PPARg siRNA. Immunoprecipitation experiments revealed that PPARg interacted with NFjB p65 and inhibited NF-jB activation. The present study provides strong evidence that at least some of the neuroprotective effects of curcumin against OGD/R are mediated by PPARg activation. VC 2014 Wiley Periodicals, Inc. C 2014 Wiley Periodicals, Inc. V

Key words: oxygen2glucose deprivation; mitochondria; peroxisome proliferator-activated receptor gamma; curcumin; nuclear factor-jB

Cerebral ischemia is caused by sudden extreme restriction in blood supply that leads to metabolic stress, resulting from oxygen and glucose deprivation to the brain. Neurons are very vulnerable to hypoxic injury, and permanent loss of neuronal function can occur within minutes of ischemia onset. Under hypoxic conditions, cellular energy charge falls rapidly, resulting in disruption of ionic homeostasis, neuronal depolarization, calcium influx and intracellular overload, free radical generation, excessive glutamate receptor stimulation, cell swelling, and mitochondrial damage, leading ultimately to cell death through necrotic or apoptotic pathways (Lee et al., 2000). Additional Supporting Information may be found in the online version of this article. Contract grant sponsor: National Natural Science Foundation of China, contract grant numbers: 81173595; 81373794; Contract grant sponsor: Beijing Natural Science Foundation, contract grant number: 7112121; Contract grant sponsor: China2Japan Friendship Hospital Scientific Research Foundation, contract grant number: 2010-QN-07; Contract grant sponsor: China2Japan Friendship Hospital Youth Science and Technology Excellence Project, contract grant number: 2014-QNYC-A04 Z.-J. Liu and H.-Q. Liu contributed equally to this work. *Correspondence to: Dr. Cheng Xiao, China-Japan Friendship Hospital, 2 Yinghua Dongjie, Hepingli, Beijing, China 100029. E-mail: [email protected] Received 29 November 2013; Revised 13 May 2014; Accepted 27 May 2014 Published online 26 June 2014 in Wiley Online (wileyonlinelibrary.com). DOI: 10.1002/jnr.23438

Library

1550

Liu et al.

Much research has been conducted in an attempt to identify potential therapeutics to ameliorate neural damage caused by ischemic insult. However, currently there is no treatment that is broadly efficacious when administered in the hours following ischemia. Curcumin is a yellow phenolic pigment obtained from the powdered rhizome of Curcumin longa Linn (turmeric). Curcumin has multiple pharmacological actions, such as antioxidant (Feinstein et al., 2005), anti-inflammation (Barinaga, 1998), and anticancer (Dutta et al., 2005) effects. Recent studies have demonstrated that curcumin protects neurons from cerebral ischemic injury (Lim et al., 2005; Zhao et al., 2010). In addition, our previous studies demonstrated that curcumin is a natural agonist of peroxisome proliferator-activated receptor-g (PPARg; Liu et al., 2011). PPARg is a member of the nuclear receptor superfamily of ligand-activated transcription factors. It is the most thoroughly characterized isoform, in part because of its therapeutic potential for treatment of diabetes and related consequences, such as metabolic syndrome. Studies have shown that PPARg is a crucial transcription factor for neuroprotection (Bordet et al., 2006) and antiinflammation (Lehrke and Lazar, 2005). Moreover, earlier studies implicated PPARg in several brain diseases, including stroke (Xu et al., 2013), Alzheimer’s disease (Fakhfouri et al., 2012), and amyotrophic lateral sclerosis (Benedusi et al., 2012). Because PPARg is both a neuroprotective factor and a target of curcumin, we speculated that curcumin protects against cerebral ischemic injury by activating PPARg signaling. The present study was designed to investigate the therapeutic effects and mechanisms of curcumin in a cellular model of oxygen–glucose deprivation/ reoxygenation (OGD/R) that specifically mimics the rapid depletion of oxygen and glucose observed under ischemic conditions in vivo (Cimarosti and Henley, 2008). The results show that curcumin markedly increases the viability of primary cultured neurons by maintaining mitochondrial function, suppressing the nuclear factor-jB (NF-jB) pathway, and preventing apoptosis, effects blocked by PPARg inhibition. Thus, the beneficial effects of curcumin on neurons are dependent on activation of PPARg. MATERIALS AND METHODS Reagents Curcumin, 2-chloro-5-nitrobenzanilide (GW9662), 3-(4,5-dimethylthiazol-2-yl)- 2,5-diphenyl tetrazolium bromide (MTT), Griess reagent, and a caspase-3 assay kit were purchased from Sigma (St. Louis, MO). A small-interfering RNA (siRNA) targeting PPARg was synthesized by Invitrogen (Carlsbad, CA), and a PPARg transcription factor assay kit was obtained from Cayman Chemical (Ann Arbor, MI). A coimmunoprecipitation (Co-IP) kit was purchased from Pierce (Rockford, IL). Antibodies against NF-jB p65, IKK, IjB-a, and cytochrome c (Cyt c) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Apoptosis-inducing factor (AIF), Bcl2, Bax, and PPARg antibodies were obtained from

Millipore (Temecula, CA), and caspase-3 antibody was obtained from Abcam (Cambridge, MA). A lactate dehydrogenase (LDH) assay kit was supplied by Nanjing Jiancheng Bioengineering Institute (Nanjing City, China). A JC-1 assay kit, dichlorofluorescein diacetate (DCFDA), and Lipofectamine LTX Plus reagent were purchased from Invitrogen. Cell Isolation and Culture Cortical neurons were obtained from Sprague Dawley prenatal rats at 19 days of gestation (Beijing Vital River Animal Center, Beijing, China). Briefly, the cortex was isolated and incubated in dissociation buffer containing 0.14 M NaCl, 5.4 mM KCl, 24 mM Hepes, 4.2 mM NaHCO3, 2.5 mM NaH2PO4, and 14 mM glucose. The tissue was digested with 0.25 mg/ml trypsin, triturated in Neurobasal medium by pipetting, and filtered to remove large debris. Isolated cortical neurons were suspended in Neurobasal medium containing 2% B27, 0.5 mM GlutaMAX-I, 100 U/ml penicillin, and 100 mg/ ml streptomycin. Cells were plated at 106 cells/cm2 on plastic dishes coated with 10 mg/ml poly-D-lysine and maintained in a humidified incubator with 95% air/5% CO2 at 37 C. After 72 hr, the cell medium was replaced by fresh medium containing 10 mM cytosine arabinoside to prevent the growth of glial cells. Cell viability was tested by the trypan blue exclusion method. All experimental procedures were performed in accordance with the guidelines of the Beijing Municipal Ethic Committee for the Care and Use of Laboratory Animals. OGD/R and Curcumin Treatments For OGD/R treatment, the regular Neurobasal culture medium was replaced by glucose-free Earle’s balanced salt solution, pH 7.4. The cells were then placed in a humidified anaerobic chamber containing a mixture of 95% N2 and 5% CO2 and maintained at 37 C for 60 min. Ischemic treatment was terminated by replacing the anoxic medium with Neurobasal medium, and the culture plates were returned to a normoxic incubator for an additional 24 hr. Curcumin (10 lM) was incubated with cells 1 hr before OGD/R. For the inhibition of PPARg function, GW9662 (1 lM) was added into the cultures, or cells were transfected with PPARg siRNA 1 hr before OGD/R. Targeting Silencing of PPARc by RNA Interference To silence PPARg gene expression, cortical neurons were grown to 60% confluence and transfected with Lipofectamine LTX Plus reagent plus 25 nM of the appropriate PPARg siRNA 1 hr before OGD/R, according to the manufacturer’s instructions. The concentration of PPARg siRNA was optimized in pilot experiments by testing the efficacy of different concentrations for gene silencing. Briefly, whole-cell lysates from siRNA-transfected culture plates were prepared, and the level of PPARg expression was measured by Western blot analysis. MTT Reduction Assay Cell viability was quantified by the MTT colorimetric assay. At the end of OGD/R, MTT was added at a final Journal of Neuroscience Research

Curcumin Protects Neurons Via PPARg concentration of 0.5 mg/ml for 4 hr at 37 C. After removing the medium, the insoluble formazan crystals produced from MTT by viable cells were dissolved in 100 ll dimethyl sulfoxide, and the absorbance was read at 490 nm. The number of viable cells is proportional to the optical density. Results are expressed as percentage of the optical density measured from vehicle-treated control cultures. LDH Assay Neuronal injury was quantitatively assessed by measurement of LDH released into the culture medium after OGD/R treatment by using a commercially available kit. The optical density was measured at 492 nm. Results were expressed as percentage of the optical density measured in vehicle-treated control cultures. Nitric Oxide Assay Accumulation of nitrite (NO 2 2 ), a stable breakdown product of nitric oxide (NO), was measured in culture media after OGD/R by using the Griess reagent. Briefly, Griess reagent was added to an equal volume of supernatant (100 ll) and incubated for 20 min at room temperature. The optical density was measured at 540 nm. A standard curve was established by using NO 2 2 at a range of 1–100 lM dissolved in fresh culture medium. DAPI Staining To observe cells undergoing apoptosis, 4C ¸ ,6-diamidino2-phenylindole (DAPI) staining was performed. Briefly, cells were fixed with 4% paraformaldehyde for 15 min at room temperature, washed with PBS, stained with DAPI (1 lg/ml) for 10 min at room temperature, and then washed with PBS. Stained cells were examined by using the In Cell Analyzer 1000 (GE Healthcare, Piscataway, NJ). Reactive Oxygen Species Measurement Reactive oxygen species (ROS) production was measured by using the cell-permeant probe DCFDA. Cells were loaded with DCFDA for 30 min at 37 C in the dark. The samples were pelletized by centrifugation and resuspended. ROS production was determined by measuring the fluorescence emission with a spectrofluorometer. Measurement of Mitochondrial Membrane Potential Treated cultures were incubated with 5 lM JC-1 dye for 30 min at 37 C. After incubation, the cells were washed with PBS. The emission signals at 590 and 527 nm elicited by excitation at 485 nm were measured by using a fluorometer. The ratio of the signal at 590 nm to that at 527 nm was calculated as an estimate of mitochondrial membrane potential (DWm). Caspase-3 Activity Assay Caspase-3 activity was measured in lysates of treated cortical neurons by using a commercial caspase-3 assay kit, according to the manufacturer’s instructions. Briefly, neurons were lysed by freeze–thaw. Cell lysates were centrifuged at Journal of Neuroscience Research

1551

12,000 rpm for 10 min at 4 C, and the supernatant fraction was collected. An aliquot of supernatant was incubated with 200 lM of Ac-DEVD-pNA substrate at 37 C for 4 hr, and the absorbance was measured at 405 nm. Optical density was converted to nanamoles of p-nitroaniline by using a standard curve generated with free p-nitroaniline. Western Blot Analysis Rat cortical neurons were lysed in nondenaturing lysis buffer. The 30-mg sample proteins were separated by SDSpolyacrylamide gel electrophoresis (SDS-PAGE) in a 10% polyacrylamide gel and transferred to polyvinylidene difluoride (PVDF) membranes. The membranes were blocked in 5% skim milk-TBS-T (20 mM Tris-HCl, pH 7.5, 500 mM NaCl, 0.1% Tween 20) at 4 C overnight. Blocked membranes were then incubated in 5% skim milk-TBS-T containing primary antibodies for 2 hr at room temperature, washed with 5% skim milkTBS-T, and then incubated with horseradish peroxidaseconjugated secondary antibody in skim milk-TBS-T for 2 hr at room temperature. The blot was developed with an LAS3000 chemiluminescence system (Fujifilm, Tokyo, Japan), and the densities of the bands were determined in Gel-Pro Analyzer 4.0 software. PPARc Transcriptional Activity Assay The transcriptional activity of PPARg was measured by using a commercial assay according to the manufacturer’s instructions. Briefly, nuclear extracts from control or treated cultures were prepared by using a nuclear-cytosol extraction kit. Samples of the nuclear extract were added to 96-well plates with PPARg primary antibody (1:100), and the mixture was incubated for 1 hr at room temperature. After washing, horseradish peroxidase-conjugated secondary antibody was added to the plate and incubated for 1 hr. A transcription factor developing solution was added to each well and absorbance was measured at 450 nm, which correlates with PPARg transcriptional activity. Co-IP Assay Nuclear extracts for detection of PPARg-NF-jB binding and for detection of Bax-Bcl2 binding were prepared by using nuclear-cytosol and membrane extraction kits, respectively. The Co-IP assay was performed following the protocol of the Co-IP kit. Briefly, 50 lg of purified PPARg or Bax antibody was immobilized in 100 ll of 50% antibody-coupling gel. Protein extracts (300 lg per trial) were incubated with gentle, end over end mixing for 2 hr at room temperature. Immunoprecipitated complexes were eluted three times with 50 ll elution buffer, boiled, separated by SDS-PAGE, and transferred to PVDF membranes. Membranes were then incubated with an NF-jB or Bcl2 antibody, as previously described, and detected by using the enhanced LAS3000 chemiluminescence system. Statistical Analysis All calculations and statistical evaluations were performed in SPSS 13.0. Differences among group means were tested by one-way ANOVA with post hoc LSD tests for pair-wise

1552

Liu et al.

We first examined the effects of curcumin on PPARg expression and transcriptional activity in primary cultured cortical neurons subjected to OGD/R. Both PPARg expression and transcriptional activity were markedly decreased upon OGD/R, whereas curcumin treatment reversed these effects, in agreement with our previous study showing that curcumin is a potent PPARg agonist (Liu et al., 2011) . Cells coadministrated curcumin plus GW9662 exhibited lower PPARg activity than control cells after OGD/R but exhibited similar expression, suggesting that GW9662 acts mainly to inhibit PPARg activity. Transfection of PPARg siRNA markedly decreased both PPARg expression and activity compared with control cells (Fig. 2).

Fig. 1. Effects of GW9662 on neuronal viability and PPARg transcriptional activity. Different concentrations (0210 lM) of GW9662 were incubated with neurons for 24 hr. A: Effects of GW9662 on cortical neuronal viability. B: Inhibitory effects of GW9662 on PPARg activity.

comparisons. Results were presented as mean 6 SD and P < 0.05 was considered significant.

RESULTS Curcumin Upregulated PPARc Expression and Transcriptional Activity in Primary Cultured Cortical Neurons Subjected to OGD/R PPARg antagonist GW9662 was used to assess the involvement of PPARg in curcumin-mediated neuroprotection. To find the most effective and nontoxic dose of GW9662, we added a series of concentrations of GW9662 to primary rat cortical neurons and assessed toxicity. The results showed that GW9662 at 1 lM for 24 hr had no measureable neurotoxicity and exhibited good inhibition of PPARg activity (Fig. 1). Moreover, GW9662 at 1 lM did not alter OGD/R-induced neurotoxicity when applied alone (Supp. Info. Fig. 1). In the following experiments, therefore, the concentration of GW9662 was set at 1 lM. We silenced PPARg expression by using a targeted PPARg siRNA. In pilot experiments, we optimized the concentration of PPARg siRNA and found that transfection of 25 nM of PPARg siRNA for 24 hr had no influence on neuronal viability (Supp. Info. Fig. 2) and did not alter the neurotoxic effects of OGD/R (Supp. Info. Fig. 1).

Curcumin Enhanced Neuronal Viability and Reduced Apoptosis To investigate the neuroprotective effects of curcumin against ischemic-like injury, rat cortical neurons were incubated with 10 mM curcumin 1 hr before OGD/R. The results showed that OGD/R elicited remarkable neuron injury detected by MTT and LDH assay. Curcumin significantly increased cell viability, as evidenced by LDH and MTT assays, compared with cultures subjected to OGD/R alone (Fig. 3A,B). In addition, curcumin pretreatment reduced apoptosis and NO release induced by OGD/R (Fig. 3C,D). All of these neuroprotective actions were attenuated by copretreatment with GW9662 or transfection of PPARg siRNA (Fig. 2), suggesting that the beneficial effects of curcumin on neurons were PPARg mediated. Curcumin Decreased Caspase-3 Activity Caspase-3 has been suggested to mediate the terminal stages of neuronal apoptosis. Both caspase-3 expression and activity increased markedly following OGD/R, whereas curcumin pretreatment significantly reduced caspase-3 expression and activity. The inhibitory effects of curcumin on caspase-3 were attenuated by coadministration of GW9662 or silencing of PPARg (Fig. 4), suggesting that PPARg mediates the neuroprotective effects of curcumin, at least in part, by blocking activation of the caspase-3-dependent apoptosis pathway. Curcumin Improved Mitochondrial Function Mitochondria are important sources of intracellular ROS within most mammalian cells, and ROS production contributes to mitochondrial damage. As shown in Figure 5A, the production of ROS increased after OGD/R, whereas curcumin pretreatment significantly reduced ROS accumulation. The DWm is an important parameter of mitochondrial function and is used as an indicator of cell health. The present data also showed that curcumin ameliorated the loss of DWm caused by OGD/R (Fig. 5B). Cyt c and AIF are localized to mitochondria and released in response to OGD/R stimulus. Consistent with the enhanced cell viability and suppression of caspase-3 Journal of Neuroscience Research

Curcumin Protects Neurons Via PPARg

1553

Fig. 2. Curcumin increased PPARg expression and activation in primary cultured cortical neurons. Curcumin (10 lM) was incubated with cells 1 hr before OGD/R. GW9662 (1 lM) was added to the cultures or cells were transfected with PPARg siRNA 1 hr before OGD/R. A: Western blot assay of PPARg expression. Western blot

images are representative of four independent experiments. Data are expressed as mean 6 SD. B: PPARg transcriptional activity assay. Data are expressed as mean 6 SD of six independent experiments. * P < 0.05, **P < 0.01 vs. control cells; ##P < 0.01 vs. OGD/R-treated cells; DP < 0.05, DDP < 0.01 vs. curcumin-treated cells.

activity, curcumin pretreatment significantly inhibited the release of Cyt c and AIF following OGD/R (Fig. 5C). The mitochondrial apoptotic pathway is largely mediated through Bcl-2 family proteins. These proteins include both proapoptotic members such as Bax, which promotes mitochondrial permeability transition leading to loss of DWm and release of Cyt c with ensuing caspase-3 activation, and antiapoptotic members such as Bcl-2, which inhibits their effects or inhibits the mitochondrial release of Cyt c. The Bax/Bcl-2 ratio is a well-established determinant of apoptosis, with an increase in hypoxic neurons predictive of apoptosis. Indeed, expression of Bax increased and Bcl2 expression decreased following OGD/R, whereas curcumin treatment markedly reduced the Bax/Bcl-2 ratio (Fig. 5D). Bcl-2 binds to Bax at the membrane and inhibits apoptosis, but Co-IP revealed that this interaction was very weak in OGD/R-treated cells, whereas curcumin pretreatment markedly increased Bax– Bcl-2 binding (Fig. 5E). In agreement with results showing that coadministration of GW9662 or transfection of PPARg siRNA blocked the neuroprotective effects of curcumin, these treatments exacerbated ROS accumulation, DWm loss, Cyt c and ATF release, and Bax/Bcl-2 ratio and reduced Bax–Bcl-2 binding in the presence of curcumin (Fig. 5).

the inhibitory IjB protein, which results in the dissociation of IjB from NF-jB. This study shows that, after cells were subjected to OGD/R, increased phosphorylation of IKK and degradation of IjB-a were observed. Pretreatment with curcumin decreased IKK phosphorylation and IjB-a degradation. Coadministration of GW9662 counteracted the inhibitory effect of curcumin on IKK phosphorylation and IjB-a degradation (Fig. 6A,B). It has been reported that PPARg can also inhibit NF-jB activity, so we further investigated whether curcumin-induced suppression of the IKK–IjB-a–NFjB pathway is mediated by PPARg. Translocation of the NF-jB p65 subunit after OGD/R was inhibited by curcumin pretreatment but restored by GW9662 or PPARg siRNA (Fig. 6C). Moreover, PPARg interacted with NF-jB p65 in curcumin-treated cells as revealed by CoIP, whereas blocking PPARg with GW9662 or silencing of PPARg reduced this interaction (Fig. 6D). These results suggest that activated PPARg interacted with NFjB p65 and that this interaction may suppress NF-jB activation in curcumin-treated neuronal cells subjected to OGD/R.

Curcumin Suppressed NF-jB Signaling Pathway IjB kinase (IKK) is the upstream of the NF-jB signal transduction cascade, and it specifically phosphorylates Journal of Neuroscience Research

DISCUSSION PPARg is a transcription factor with well-established neuroprotective features. Pharmacological activation of PPARg confers neuroprotection in experimental models of ischemic injury (Lehrke et al., 2005), Alzheimer’s disease (Xu et al., 2013), and amyotrophic lateral

1554

Liu et al.

Fig. 3. Curcumin protected neurons from death and apoptosis caused by OGD/R. Curcumin (10 lM) was incubated with cells 1 hr before OGD/R. GW9662 (1 lM) was added to the cultures or cells were transfected with PPARg siRNA 1 hr before OGD/R. A: MTT assay of cell viability. B: LDH release to the culture medium. C: NO levels

in the culture medium. Data are expressed as mean 6 SD of six individual experiments. D: DAPI staining of apoptosis cells. **P < 0.01 vs. control cells; ##P < 0.01 vs. OGD/R-treated cells; DP < 0.05 vs. curcumin-treated cells. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

sclerosis (Fakhfouri et al., 2012). Our previous study demonstrated that curcumin is a potent promoter of PPARg and that curcumin-induced neuroprotection against cerebral ischemic injury in rats was mediated by PPARg through suppression of the inflammatory response (Liu et al., 2013). This study examined the molecular mechanisms underlying curcumin neuroprotection against OGD/R-induced ischemic-like injury. The decrease of neuronal viability and the increase of LDH, NO release, and caspase-3 activity, in addition to the dysfunction of mitochondria and activated NF-jB

signaling, were observed in cells subjected to OGD/R, suggesting that OGD/R caused serious damage to the primary neuronal culture. Pretreatment of curcumin protected neurons from apoptosis, improved mitochondrial function, and suppressed the NF-jB pathway. That these neuroprotective effects were blocked by antagonists of PPARg indicates that PPARg signaling is necessary for curcumin-induced neuroprotection against ischemia-like damage and strongly confirms our previous study on cerebral ischemia in rats (Liu et al., 2013). Journal of Neuroscience Research


1555

Fig. 4. Curcumin decreased the activation of caspase-3. Curcumin (10 lM) was incubated with cells 1 hr before OGD/R. GW9662 (1 lM) was added to the cultures or cells were transfected with PPARg siRNA 1 hr before OGD/R. A: Caspase-3 activity. B: Active caspase-3 and procaspase-3 expression. Data are expressed as

mean 6 SD. For measurement of caspase-3 activity, six independent experiments were performed. Western blot images are representative of four independent experiments. **P < 0.01 vs. control cells; ## P < 0.01 vs. OGD/R-treated cells; DP < 0.05 vs. curcumin-treated cells.

Evidence has shown that PPARg is critical for mitochondrial stabilization (Fuenzalida et al., 2007). Mitochondrion is a key modulator in the development of brain injury during cerebral ischemia (Sims and Muyderman, 2010). Under ischemic conditions in which the delivery of oxygen and glucose supply is inadequate, mitochondrial oxidative phosphorylation is inhibited, resulting in more ROS generation (Sanderson et al., 2013), DWm loss, and disruption of energy metabolism (Li et al., 2012). PPARg can stabilize mitochondrial function under pathological conditions (Fuenzalida et al., 2007). Mitochondrial function declined following OGD/ R, as evidenced by ROS accumulation, dissociation of Bax from Bcl2, increased Bax/Bcl2 ratio, DWm loss, and release of Cyt c and AIF. Several previous studies also have shown that curcumin protected neurons from apoptosis by maintaining mitochondrial function (Wang et al., 2005; Zhao et al., 2010; Pan et al., 2012), but the molecular mechanisms remained unclear. Activation of PPARg by the agonists rosiglitazone and pioglitazone prevented DWm loss and mitochondrial damage (Puigserver and Spiegelman, 2003; Wu et al., 2009), suggesting that PPARg activation promotes mitochondrial biogenesis and repair during cellular injury, thereby preventing oxidative stress and restoring energy charge. In agreement with these studies, improved mitochondrial function under curcumin treatment (e.g., stabilization of DWm) was blocked by an inhibitor of PPARg activity. More-

over, similar to cells subjected to OGD/R alone, inhibiting PPARg prior to OGD/R downregulated Bcl2, reduced Bax–Bcl2 binding, decreased DWm, enhanced Cyt c and AIF release, and increased the apoptosis rate. Our data were further supported by a study showing that rosiglitazone induced mitochondrial biogenesis in mouse brain (Strum et al., 2007) and were in agreement with the resistance to mitochondrial injury and cell death already described in Bcl2-overexpressing neural cells (Soane and Fiskum, 2005). However, other articles have reported curcumin-induced mitochondria-mediated apoptosis in cancer cells (Karunagaran et al., 2005; Ravindran et al., 2009), suggesting that curcumin has distinct effects in transformed cells. We demonstrated that curcumininduced protection of mitochondria is not direct but is mediated through PPARg. Thus, the differences in curcumin responses between different cell types or pathological conditions might reflect differences in intermediary signaling mechanisms, an idea that warrants further study. The NF-jB family of transcription factors responds to a variety of stressful stimuli associated with OGD/R, including ROS and excess glutamate, by activating genes involved in inflammation and apoptosis. There is also evidence that NF-jB has direct effects within mitochondria, the site of oxidative phosphorylation-dependent ATP generation. The vast majority of mitochondrial proteins are encoded by nuclear genes and imported from the cytosol (Schmidt et al., 2010). The activity of NF-jB is

Journal of Neuroscience Research

1556

Liu et al.

Fig. 5. Curcumin protected mitochondria from injury. Curcumin (10 lM) was incubated with cells 1 hr before OGD/R. GW9662 (1 lM) was added to the cultures or cells were transfected with PPARg siRNA 1 hr before OGD/R. A: ROS level in mitochondria. B: Membrane potential of mitochondria. Data are expressed as mean 6 SD of six independent experiments. C: Cyt c and AIF in

mitochondria. D: Bax/Bcl-2 ratio in mitochondria. Data are expressed as mean 6 SD. Western blot images are representative of four independent experiments. **P < 0.01 vs. control cells; ##P < 0.01 vs. OGD/R-treated cells; DP < 0.05, DDP < 0.01 vs. curcumin-treated cells. E: Interaction of Bax and Bcl2.



1557

Fig. 6. Curcumin suppressed the NF-jB signaling pathway and promoted the binding of PPARg to NF-jB p65. Curcumin (10 lM) was incubated with cells 1 hr before OGD/R. GW9662 (1 lM) was added to the cultures or cells were transfected with PPARg siRNA 1 hr before OGD/R. A: p-IKK and IKK expression. B: IjB-a expres-

sion. C: NF-jB p65 expression. Data are expressed as mean 6 SD. Western blot images are representative of four independent experiments. **P < 0.01 vs. control cells; #P < 0.05, ##P < 0.01 vs. OGD/ R-treated cells; DP < 0.05, DDP < 0.01 vs. curcumin-treated cells. D: Interaction of PPARg and NF-jB p65.

tightly regulated at multiple levels. The primary mechanism for regulating NF-jB is through inhibitory IjB and the kinase that phosphorylates IjBs, the IKK. In response to external signals, associated mainly with stress, IjB is phosphorylated by the IKK complex and subsequently degraded through ubiquitin-dependent proteolysis (Yamaoka et al., 1998). The NF-jB pathway is activated in cerebral ischemia (Ridder and Schwaninger, 2009), possibly by ROS-induced IjB degradation (Song et al., 2005) and by posttranslational modifications of NF-jB

p65 mandatory for full NF-jB-mediated transcription (Gloire and Piette, 2009). Curcumin has been found to attenuate NF-jB DNA-binding activity by inhibiting IjB degradation, thus preventing NF-jB translocation to the nucleus (Moon et al., 2006). These results are in agreement with our findings that OGD/R significantly enhanced IKK phosphoactivation and increased both IjB-a degradation and NF-jB p65 translocation, resulting in increased NF-jB p65 transcriptional activity, whereas curcumin effectively reversed NF-jB pathway


1558

Liu et al.

activation. Inhibition of the NF-jB signaling pathway by curcumin is thus implicated in the molecular regulation of ischemia-triggered oxidative stress. Recently, the inhibitory effects of PPARg on NF-jB activation have been demonstrated in multiple cell systems; for example, PPARg suppresses proinflammatory gene expression at the transcriptional level by inhibiting NF-jB activation (Jennewein et al., 2008). Our previous study showed that NF-jB activity was inhibited by PPARg in a rat cerebral ischemia model, and the present data show that PPARg physically interacts with the NF-jB p65 subunit, thereby blocking NF-jB activation and inhibiting downstream activation of p65-dependent gene expression. Blockade of this pathway by GW9662 or transfection of PPARg siRNA strongly suggests that curcumin inhibits the NFjB pathway through PPARg. However, because curcumin is a multitarget drug with multiple pharmacological actions, we speculate that there may be additional mechanisms responsible for neuroprotection by curcumin. For example, the BDNF/TrkB2MAPK/PI-3K2CREB signaling pathway (Wang et al., 2010) and Nrf2 and HO-1 (Yang et al., 2009) have been found to contribute to curcumin neuroprotection. Nonetheless, we consider the PPARg pathway to be a critical mediator of the neuroprotective effects of curcumin as evidenced by the neartotal reversal of neuroprotection by PPARg inhibition. The precise molecular mechanisms through which PPARg mediates the neuroprotective effects of curcumin require further study. This study has investigated the effects of curcumin on neuronal ischemic insults in vitro to gain a better understanding of both the potential neuroprotective efficacy of curcumin and the molecular signaling pathways. These data provide evidence that curcumin protects primary neurons from OGD/R-induced apoptosis by activating PPARg. In turn, PPARg sustains mitochondrial function, at least in part, by suppressing the NF-jB pathway. ACKNOWLEDGMENTS We are most grateful to Prof. Gai-Hui Guo for the helpful suggestions and generous comments. The authors have no conflicts of interest. REFERENCES Barinaga M. 1998. Stroke-damaged neurons may commit cellular suicide. Science 281:1302–1303. Benedusi V, Martorana F, Brambilla L, Maggi A, Rossi D. 2012. The peroxisome proliferator-activated receptor g (PPARg) controls natural protective mechanisms against lipid peroxidation in amyotrophic lateral sclerosis. J Biol Chem 287:35899–35911. Bordet R, Ouk T, Petrault O, Gele P, Gautier S, Laprais M, Deplanque D, Duriez P, Staels B, Fruchart JC, Bastide M. 2006. PPAR: a new pharmacological target for neuroprotection in stroke and neurodegenerative diseases. Biochem Soc Trans 34:1341–1346. Cimarosti H, Henley JM. 2008. Investigating the mechanisms underlying neuronal death in ischemia using in vitro oxygen2glucose deprivation: potential involvement of protein SUMOylation. Neuroscientist 14:626– 636.

Dutta S, Padhye S, Priyadarsini KI, Newton C. 2005. Antioxidant and antiproliferative activity of curcumin semicarbazone. Bioorg Med Chem Lett 15:2738–2744. Fakhfouri G, Ahmadiani A, Rahimian R, Grolla AA, Moradi F, Haeri A. 2012. WIN55212-2 attenuates amyloid-beta-induced neuroinflammation in rats through activation of cannabinoid receptors and PPAR-g pathway. Neuropharmacology 63:653–666. Feinstein DL, Spagnolo A, Akar C, Weinberg G, Murphy P, Gavrilyuk V, Dello Russo C. 2005. Receptor-independent actions of PPAR thiazolidinedione agonists: is mitochondrial function the key? Biochem Pharmacol 70:177–188. Fuenzalida K, Quintanilla R, Ramos P, Piderit D, Fuentealba RA, Martinez G, Inestrosa NC, Bronfman M. 2007. Peroxisome proliferator-activated receptor gamma upregulates the Bcl-2 anti-apoptotic protein in neurons and induces mitochondrial stabilization and protection against oxidative stress and apoptosis. J Biol Chem 282: 37006–37015. Gloire G, Piette J. 2009. Redox regulation of nuclear post-translational modifications during NF-kappaB activation. Antioxid Redox Signal 11: 2209–2222. Jennewein C, Kuhn AM, Schmidt MV, Meilladec-Jullig V, von Knethen A, Gonzalez FJ, Br€ une B. 2008. Sumoylation of peroxisome proliferator-activated receptor gamma by apoptotic cells prevents lipopolysaccharide-induced NCoR removal from kappaB binding sites mediating transrepression of proinflammatory cytokines. J Immunol 181:5646–5652. Karunagaran D, Rashmi R, Kumar TR. 2005. Induction of apoptosis by curcumin and its implications for cancer therapy. Curr Cancer Drug Targets 5:117–129. Lee JM, Grabb MC, Zipfel GJ, Choi DW. 2000. Brain tissue responses to ischemia. J Clin Invest 106:723–731. Lehrke M, Lazar MA. 2005. The many faces of PPAR gamma. Cell 123: 993–999. Li J, Ma X, Yu W, Lou Z, Mu D, Wang Y, Shen B, Qi S. 2012. Reperfusion promotes mitochondrial dysfunction following focal cerebral ischemia in rats. PLoS One 7:e46498. Lim CS, Jin DQ, Mok H, Oh SJ, Lee JU, Hwang JK, Ha I, Han JS. 2005. Antioxidant and anti-inflammatory activities of xanthorrhizol in hippocampal neurons and primary cultured microglia. J Neurosci Res 82:831–838. Liu ZJ, Bao XQ, Jia SH, Hu YQ, Zhang Y, Zhang YH, Jiao JS, Tian ZH. 2011. Constructing the screening cell model of PPARg and to validate whether the curcumin is the natural agonist of PPARg. J Apoplexy Nervous Dis 28:872–874. Liu ZJ, Liu W, Liu L, Xiao C, Wang Y, Jiao JS. 2013. Curcumin protects neuron against cerebral ischemia-induced inflammation through improving PPAR-gamma function. Evid Based Complement Alternat Med 2013:470975. doi:10.1155/2013/470975. Moon DO, Jin CY, Lee JD, Choi YH, Ahn SC, Lee CM, Jeong SC, Park YM, Kim GY. 2006. Curcumin decreases binding of Shiga-like toxin-1B on human intestinal epithelial cell line HT29 stimulated with TNF-alpha and IL-1beta: suppression of p38, JNK and NF-kappaB p65 as potential targets. Biol Pharm Bull 29:1470–1475. Pan J, Li H, Ma JF, Tan YY, Xiao Q, Ding JQ, Chen SD. 2012. Curcumin inhibition of JNKs prevents dopaminergic neuronal loss in a mouse model of Parkinson’s disease through suppressing mitochondria dysfunction. Transl Neurodegener 1:16. Puigserver P, Spiegelman BM. 2003. Peroxisome proliferator-activated receptor-gamma coactivator 1 alpha (PGC-1 alpha): transcriptional coactivator and metabolic regulator. Endocr Rev 24:78–90. Ravindran J, Prasad S, Aggarwal BB. 2009. Curcumin and cancer cells: how many ways can curry kill tumor cells selectively? AAPS J 11:495– 510. Ridder DA, Schwaninger M. 2009. NF-kappaB signaling in cerebral ischemia. Neuroscience 158:995–1006. Journal of Neuroscience Research

Curcumin Protects Neurons Via PPARg Sanderson TH, Reynolds CA, Kumar R, Przyklenk K, H€ uttemann M. 2013. Molecular mechanisms of ischemia–reperfusion injury in brain: pivotal role of the mitochondrial membrane potential in reactive oxygen species generation. Mol Neurobiol 47:9–23. Schmidt O, Pfanner N, Meisinger C. 2010. Mitochondrial protein import: from proteomics to functional mechanisms. Nat Rev Mol Cell Biol 11:655–667. Sims NR, Muyderman H. 2010. Mitochondria, oxidative metabolism, and cell death in stroke. Biochim Biophys Acta 1802:80–91. Soane L, Fiskum G. 2005. Inhibition of mitochondrial neural cell death pathways by protein transduction of Bcl-2 family proteins. J Bioenerg Biomembr 37:179–190. Song YS, Lee YS, Chan PH. 2005. Oxidative stress transiently decreases the IKK complex (IKK alpha, beta, and gamma), an upstream component of NF-kappaB signaling, after transient focal cerebral ischemia in mice. J Cereb Blood Flow Metab 25:1301–1311. Strum JC, Shehee R, Virley D. 2007. Rosiglitazone induces mitochondrial biogenesis in mouse brain. J Alzheimers Dis 11:45–51. Wang Q, Sun AY, Simonyi A, Jensen MD, Shelat PB, Rottinghaus GE, MacDonald RS, Miller DK, Lubahn DE, Weisman GA, Sun GY. 2005. Neuroprotective mechanisms of curcumin against cerebral ischemia-induced neuronal apoptosis and behavioral deficits. J Neurosci Res 82:138–148.


1559

Wang R, Li YH, Xu Y, Li YB, Wu HL, Guo H, Zhang JZ, Zhang JJ, Pan XY, Li XJ. 2010. Curcumin produces neuroprotective effects via activating brain-derived neurotrophic factor/TrkB-dependent MAPK and PI-3K cascades in rodent cortical neurons. Prog Neuropsychopharmacol Biol Psychiatry 34:147–153. Wu JS, Lin TN, Wu KK. 2009. Rosiglitazone and PPAR-g overexpression protect mitochondrial membrane potential and prevent apoptosis by upregulating antiapoptotic Bcl-2 family proteins. J Cell Physiol 220: 58–71. Xu F, Li J, Ni W, Shen YW, Zhang XP. 2013. Peroxisome proliferatoractivated receptor-g agonist 15d-prostaglandin J2 mediates neuronal autophagy after cerebral ischemia–reperfusion injury. PLoS One 8: e55080. Yamaoka S, Courtois G, Bessia C, Whiteside ST, Weil R, Agou F, Kirk HE, Kay RJ, Isra€el A. 1998. Complementation cloning of NEMO, a component of the IjB kinase complex essential for NF-jB activation. Cell 93:1231–1240. Yang C, Zhang X, Fan H, Liu Y. 2009. Curcumin upregulates transcription factor Nrf2, HO-1 expression and protects rat brains against focal ischemia. Brain Res 1282:133–141. Zhao J, Yu S, Zheng W, Feng G, Luo G, Wang L, Zhao Y. 2010. Curcumin improves outcomes and attenuates focal cerebral ischemic injury via antiapoptotic mechanisms in rats. Neurochem Res 35:374–379.