l -Carnitine attenuates H2O2-induced neuron apoptosis via inhibition of endoplasmic reticulum stress

Neurochemistry International 78 (2014) 86–95

Contents lists available at ScienceDirect

Neurochemistry International journal homepage: www.elsevier.com/locate/nci

L-Carnitine

attenuates H2O2-induced neuron apoptosis via inhibition of endoplasmic reticulum stress Junli Ye a,⇑,1, Yantao Han b,1, Xuehong Chen b, Jing Xie b, Xiaojin Liu c, Shunhong Qiao d, Chunbo Wang b a

Department of Pathophysiology, Medical College, Qingdao University, Qingdao, Shandong 266071, China Department of Pharmacology, Medical College, Qingdao University, Qingdao 266071, China c Department of Pharmacy, Dezhou People’s Hospital, Dezhou, China d Department of Biological Sciences, University of Memphis, TN 38111, USA b

a r t i c l e

i n f o

Article history: Received 18 April 2014 Received in revised form 15 August 2014 Accepted 27 August 2014 Available online 9 September 2014 Keywords: Endoplasmic reticulum stress Oxidative stress SH-SY5Y cells L-Carnitine Apoptosis

a b s t r a c t Both oxidative stress and endoplasmic reticulum stress (ER stress) have been linked to pathogenesis of neurodegenerative diseases. Our previous study has shown that L-carnitine may function as an antioxidant to inhibit H2O2-induced oxidative stress in neuroblastoma SH-SY5Y cells. To further explore the neuroprotection of L-carnitine, here we study the effects of L-carnitine on the ER stress response in H2O2-induced SH-SY5Y cell injury. Our results showed that L-carnitine pretreatment could increase cell viability; inhibit apoptosis and ROS accumulation caused by H2O2 or tunicamycin (TM). L-carnitine suppress the endoplasmic reticulum dilation and activation of ER stress-associated proteins including glucose-regulated protein 78 (GRP78), CCAAT/enhancer-binding protein-homologous protein (CHOP), JNK, Bax and Bim induced by H2O2 or TM. In addition, H2O2-induced cell apoptosis and activation of ER stress can also be attenuated by antioxidant N-acetylcysteine (NAC), CHOP siRNA and the inhibitor of ER stress 4-phenylbutyric acid (4-PBA). Taken together, our results demonstrated that H2O2 could trigger both oxidative stress and ER stress in SH-SY5Y cells, and ER stress participated in SH-SY5Y apoptosis mediated by H2O2-induced oxidative stress. CHOP/Bim or JNK/Bim-dependent ER stress signaling pathways maybe related to the neuroprotective effects of L-carnitine against H2O2-induced apoptosis and oxidative injury. Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction Numerous studies have demonstrated the oxidative stress mediated by reactive oxygen substances (ROS) involved in pathogenesis of neurodegenerative diseases such as Alzheimer’s disease, Parkinson’s disease, and ischemic and hemorrhagic stroke and antioxidant strategy has shown promise in the treatment of both acute and chronic neurodegenerative diseases (Halliwell, 2006; Calabrese et al., 2010; Ghosh et al., 2011). In addition to oxidative stress, endoplasmic reticulum (ER) stress characterized by Abbreviations: H2O2, hydrogen peroxide; TM, tunicamycin; ROS, reactive oxygen species; DCFDA, 20 ,70 -dichlorodihydrofluorescein diacetate; GRP78/Bip, glucoseregulated protein 78; MAPK, mitogen-activated protein kinase; CHOP, CCAAT/ enhancer-binding protein-homologous protein; NAC, N-acetylcysteine; ER stress, endoplasmic reticular stress; ERK, extracellular-signal regulated kinase; JNK, c-Jun N-terminal protein kinase; ESR, electron spin resonance technology. ⇑ Corresponding author at: Department of Pathophysiology, Medical College, Qingdao University, 423 Room, Boya Building, 308 Ningxia Road, Qingdao 266071, China. Tel.: +86 532 83780035; fax: +86 532 83780029. E-mail address: [email protected] (J. Ye). 1 These authors contributed equally to this work. http://dx.doi.org/10.1016/j.neuint.2014.08.009 0197-0186/Ó 2014 Elsevier Ltd. All rights reserved.

unfolded protein accumulation and up-expression of glucose-regulated protein 78 (GRP78), also serves as an important role in neuron apoptosis and a number of classic death signals may involve ER gateways (Logue et al., 2013; Scheper and Hoozemans, 2009). The ER-mediated pathway triggered by ER stress will leads to proapoptotic unfolded protein response including induction of CCAAT/ enhancer-binding protein-homologous protein (CHOP), activation of the apoptosis signal-regulating kinase 1 (ASK1)–c-Jun-N-terminal kinase (JNK) pathway and Bim upregulation (Liu et al., 2013; Stefani et al., 2012). As an important organelle for neuronal survival and normal cellular function, ER is sensitive to alterations in cellular homeostasis such as redox imbalance in neuronal oxidative injury (Malhotra and Kaufman, 2007). When the client protein load is excessive compared with the reserve of ER chaperones, the ER stress occurred. Meanwhile, unfolded protein accumulation could produce more ROS, which may lead to oxidative stress (Gibson and Huang, 2004). Studies have demonstrated that both oxidative stress and endoplasmic reticulum stress (ER stress) have been linked to pathogenesis of many neurodegenerative diseases and lead to neuron apoptosis (Higgins et al., 2010; Kanekura

J. Ye et al. / Neurochemistry International 78 (2014) 86–95

et al., 2009). However, the cross-talk between oxidative and ER stress and whether and how ER stress participated in the neuronal oxidative injury need further study. L-Carnitine (4-N-trimethylammonium-3-hydroxybutyric acid) is an endogenous mitochondrial membrane compound and natural dietary additives (Nalecz and Nalecz, 1996). Recently studies have reported that L-carnitine could effectively protect various cells against oxidative injury both in vitro and in vivo (Virmani and Binienda, 2004; Binienda et al., 2004; Gülçin, 2006; Binienda and Ali, 2001; Rani and Pannerelvam, 2001; Dhitavat and Ortiz, 2005; Mazzio et al., 2003), which is a potential antioxidant for oxidative stress related neurodegenerative diseases. In our previous study, we used the human neuroblastoma SH-SY5Y cell line as an in vitro model and assessed the effect of L-carnitine on hydrogen peroxide (H2O2)-mediated oxidative stress and neurotoxicity. Our results showed that the neuroprotective effect of L-carnitine were mediated, at least, through scavenging oxygen free radicals, prevention of oxidation of lipids, enforcement of endogenous antioxidant defense, inhibition of cell apoptosis and regulation apoptosis related gene expression of Bcl-2 and Bax (Yu et al., 2011). Given the ER location of Bcl-2 and Bax proteins and their key role in the apoptotic cascade, it is reasonable to speculate that H2O2 may cause ER stress and thereby contributes to apoptosis of SH-SY5Y cells and ER stress pathways may be related to the antioxidative effects of L-carnitine on H2O2-induced neuronal apoptosis. To investigate the above hypothesis, here we observed the endoplasmic reticular ultra structural changes to verify the ER stress response in H2O2-induced or TM-induced SH-SY5Y cell apoptosis. Furthermore, the effects of L-carnitine on ER stress response, including the expression of glucose-regulated protein 78 (GRP78), CCAAT/enhancer-binding protein-homologous protein (CHOP), JNK, Bcl-2, Bax and Bim were studied. Meanwhile, the neuroprotective effects of N-acetylcysteine (NAC, ROS scavenger) and 4-PBA (a chemical chaperones and inhibitor of ER stress) on H2O2-induced SH-SY5Y cell injury were also observed.

2. Methods 2.1. Chemicals and reagents H2O2, NAC, L-carnitine, 4-phenyl-butyrate (4-PBA), dichlorofluorescein-diacetate (DCFH-DA), Hoechst33258, 3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyl-tetrazolium bromide (MTT) and tunicamycin (TM) were purchased from Sigma (St. Louis, MO, USA). Dulbecco’s modified Eagle’s medium (DMEM) and fetal bovine serum (FBS) were purchased from Gibco BRL (Grand Island, NY, USA). All other chemicals were of analytic grade.

2.2. Cell culture and treatment Human neuronal-like cells, SH-SY5Y, were routinely grown at 37 °C in a humidified incubator with 5% CO2 in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS), 2 mM glutamine, 1 mM sodium pyruvate, 100 units/ ml penicillin, and 100 lg/ml streptomycin. To determine the effects of L-carnitine on H2O2-exposed SH-SY5Y, the subconfluent (70–80%) cells were treated with indicated doses of L-carnitine or NAC (5 mM) for 3 h or 4-PBA (5 mM) for 6 h before H2O2 (400 lM) or TM (10 mM) exposure. Thereafter, cells were washed with PBS to remove the extracellular L-carnitine or NAC or 4-PBA and then cells in fresh medium were exposed to the desired doses of H2O2 or TM. Afterwards, cells were rinsed with fresh medium (without H2O2) and incubated. Cells were harvested for further analysis.

87

2.3. Cell viability analysis Cell viability and proliferation was determined using the MTT assay, which is a sensitive measurement of the normal metabolic status of cells. Briefly, cultured SH-SY5Y cells were initially plated in triplicate at a density of 1  104 cells/100 ll in 96 well plates for 24 h. The cells were pre-incubated with or without L-carnitine, NAC or 4-PBA following incubation with H2O2 or TM for 24 h. The cells were then incubated with 0.5 mg/ml MTT at 37 °C for 4 h. The formazan crystals generated by viable mitochondrial succinate dehydrogenase from MTT were extracted using an equal volume of the solubilizing buffer (0.01 N HCl and 10% SDS). Absorbance was measured at a wavelength of 490 nm using a Molecular Devices VersaMax microplate reader (Molecular devices, Sunnyvale, CA, USA). All experiments were performed in triplicate.

2.4. Apoptosis detection by nuclear Hoechst staining and Annexin VFITC/PI assay Apoptotic morphological criteria were observed by including shrinkage of the cytoplasm (round shape), nuclear condensation and membrane blebbing assayed by Hoechst 33258 staining. SHSY5Y cells were grown in 24-well plates on poly-L-lysine coated cover slips. After different treatment for 24 h as described above, cells were fixed with 4% paraformaldehyde in PBS (120 mM NaCl, 19 mM Na2HPO4, 6 mM KH2PO4), pH 7.4, for 30 min. Cells were washed with PBS and permeabilized with 0.1% Triton X-100 in PBS for 10 min and washed again. Cover slips were incubated with Hoechst 33258 (60 ng/ml) for 10 min. Slides were rinsed briefly with PBS, air-dried, then mounted in antifluorescein fading medium (Perma Flour, Immunon, PA, USA). Slides were analyzed under a fluorescence microscope (BX50-FLA, Olympus, Tokyo, Japan). Cells with condensed nuclei condensations were scored as apoptotic. The percentage of apoptotic cells in relation to the total number of cells was determined from 10 random fields per slide, from three independent experiments. Apoptosis quantification was also observed by Annexin V-FITC/ PI assay (McCullough et al., 2001). Briefly, after different treatment for 24 h as described above, cells were harvested by 0.25% trypsin, washed twice with cold PBS, and resuspended in 1  binding buffer (10 mM HEPES/NaOH, pH 7.4, 140 mM NaCl, 2.5 mM CaCl2) at a concentration of 1  106 cells/ml. Then the cells were incubated with AnnexinV- FITC and PI for 15 min at 20 °C in the dark. Samples were acquired on a FACScan flow cytometer (FACSCalibur, Becton Dickinson) and analyzed using CELLQuest software with in 1 h. Cells that were Annexin V/PI+ were counted as necrotic, those that showed up as Annexin V+/PI+ were counted as late apoptotic or secondarily necrotic, and Annexin V+/PI cells were recognized as apoptotic. Each measurement was carried out at least in triplicate.

2.5. Transmission electron microscopy Cultured SH-SY5Y cells were initially plated in triplicate at a density of 5  105 cells/well in 6 well plates for 24 h and then pre-incubated with or without L-carnitine following incubation with H2O2 for 24 h. The cultured SH-SY5Y cells were trypsinized and collected into Eppendorff tube after washing. They were fixed by 2.5% glutaraldehyde at 4 °C and washed by PBS, fixed by osmic acid, then washed by distilled water, and dehydrated by dimethylketone. After embedment in Epon-812, the sample was cut into ultrathin sections (70 nm). The ultrathin sections were dyed with uranium acetate and plumbum citrate and examined with JEM-1200EX electron microscopy (Le Bel et al., 1992).

88

J. Ye et al. / Neurochemistry International 78 (2014) 86–95

2.6. Measurement of ROS accumulation by electron spin resonance technology (ESR) and CM-H2DCFDA staining The changes of intracellular ROS level were detected by electron spin resonance technology (ESR) or CM-H2DCFDA staining. For ESR, the cells were pre-incubated with or without L-carnitine following incubation with H2O2. At the indicated time points, the cells (5  107) were harvested and homogenized at 0 °C in phosphoHepes buffer (0.038 M NaH2PO4, 0.162 M Na2HPO4, 0.01 mM EDTA, 10 mM Hepes, 0.32 M sucrose, 5 mM mercaptoethanol, 10 mM Ntert-Butyl-a-phenylnitrone [PBN], 0.5% Tween-80, and 2 mM diethylenetriamine pentacetic acid [DETAPAC]). The fragments were centrifuged at 12,000g for 10 min at 4 °C. 1.4 ml of supernatant was removed and placed in a warm bath for 30 min in the presence of 30 ll of a buffer containing 0.5 M Na2S2O4, 0.6 M diethyldithiocarbamate (DETC), 10 mM L-Argine, 0.3 M FeSO4. Then, 300 ll of acetic ether was added and the mixture was vortexed for 15 s and centrifuged at 12,000g for 8 min at 4 °C. The organic phase was assayed for ESR (Central Magnetic Field, 3385 G; sweep width, 400 G; power, 20 mW; magnification 4  105). For CM-H2DCFDA staining, cultured SH-SY5Y cells were initially plated in triplicate at a density of 5  105 cells/well in 6 well plates for 24 h and then treated as mentioned above. At the indicated time points, to monitor intracellular accumulation of ROS, the fluorescent probe cell-permeable fluorescent dyes, CM-H2DCFDA was used to monitor intracellular H2O2 (Le Bel et al., 1992). Briefly, cells were washed twice with PBS and loaded with CM-H2DCFDA for 30 min in the dark at 37 °C. Cells were subsequently washed twice with D-Hanks and collected, DCF fluorescence intensity of 100 ll cell suspension was observed under fluorescence microscopy and quantified with a fluorometer (GENios, USA) using 480 nm excitation and 530 nm emission filtER stress. The results are given as percents relative to the oxidative stress of the control cells set to 100%. All experiments were performed in triplicate. 2.7. Transient transfection of SH-SY5Y cells with CHOP small interfering RNA (siRNA) For knockdown experiments, SH-SY5Y cells grown to 50% confluence in six-well plates were transiently transfected with CHOP specific siRNA or control siRNA, a scrambled sequence non-specific for any cellular mRNA using Lipofectamine PLUS (Invitrogen, Carlsbad, CA) according to the manufacturer’s instructions. Predesigned siRNAs against human CHOP (catalog no. SC-35437) and control scrambled siRNA (catalog no. SC-37007) were purchased from Santa Cruz Biotechnology. The sense strands of siRNAs against CHOP are as follows: 50 -GAAGGCUUGGAGUAGACAA-30 , 50 -GGAAAGGUCUCAGCUUGUA-30 , and 50 -GUCUCAGCUUGUAUAU AGA-30 . SH-SY5Y cells were transfected with 100 nM CHOP siRNA or control siRNA for 24 h before H2O2 treatment. 2.8. Isolation of total RNA and reverse transcriptase-polymerase chain reaction (RT-PCR)

58 °C for 1 min, 72 °C for 1 min). The PCR products were mixed with 2 ml of gel loading buffer, electrophoresed through a 1% agarose gel and visualized by ethidium bromide staining. The intensity of each band was calibrated to the standard molecular marker on the same gel and was normalized against the intensity of GAPDH. 2.9. Western blot analysis For the quantification of protein expression, Western blot analysis was used. The cells were placed in lysis buffer (1% NP40, 0.5% sodium deoxycholate, 0.1% SDS in normal PBS; pH 6.8) containing a protease inhibitor cocktail (10 ll/ml; Sigma–Aldrich). The homogenate was centrifuged at 12,000g at 4 °C for 20 min, and the supernatant was stored at 80 °C. The protein concentration was measured using a Bradford protein assay kit (BioRad). Equal amounts of protein (10 lg) from the supernatant fraction of the same sample were separated on a 10% SDS–PAGE and transferred electrophoretically to the nitrocellulose membranes. The membrane was blocked by 5% skim milk in Tris-buffered saline containing 0.1% Triton X-100 (TBS-T) for 2 h at room temperature. Immunoblots were performed with appropriate antibodies respectively: primary antibodies for GRP78 (1:200, Santa Cruz), CHOP (1:200, Santa Cruz), ERK (1:200, Santa Cruz), p-ERK (1:200, Santa Cruz), p38 (1:200, Santa Cruz), p-p38 (1:200, Santa Cruz), JNK(1:200, Santa Cruz), p-JNK (1:200, Santa Cruz), Bim (1:200, Santa Cruz) and b-actin (1:1,000, Santa Cruz). Immunoblot analysis was performed with horseradish peroxidase-conjugated antimouse and anti-rabbit IgG using enhanced chemiluminescence Western blotting detection reagents (Amersham Bioscience, Piscataway, NJ, USA). The bands corresponding to GRP78, CHOP, Bcl-2, Bax, ERK, p-ERK, p38, p-p38, JNK, p-JNK, Bim or b-actin, were scanned and densitometrically analyzed using an automatic image analysis system (Alpha Innotech Corporation, San Leandro, CA, USA). These quantitative analyses were normalized to b-actin from the same sample (after stripping). 2.10. Reproducibility of experiments and statistical analysis All quantitative data and experiments described in this study were repeated at least three times. All analyses were performed blinded such that experimenters performing data analysis were unaware of the treatments. Data were expressed as means ± S.E.M. The criterion for statistical significance was P < 0.05. Bartlett’s tests showed no significant differences in group variances; therefore, data were evaluated using parametric statistics. Comparisons between the different groups were performed by either Student’s t-test or one-way ANOVA followed by post hoc Bonferroni tests for comparison among means. All data were analyzed using GRAPHPAD PRISM (GraphPad Software Inc., La Jolla, CA, USA) data analysis software. 3. Results

The expression of CHOP mRNA in SH-SY5Y cells was assayed 12 h after H2O2 treatment. Total RNA was extracted using Beyozol reagent, in accordance with the manufacturer’s instructions (Beyozol Beyotime Biotechnology, China). The sequences of the specific oligonucleotide primer were as follows: forward 50 -GCA CCT CCC AGA GCC CTC ACT CTC C-30 and reverse 50 -GTC TAC TCC AAG CCT TCC CCC TGC G-30 for CHOP (Sangon Biological Engineering Technology, Shanghai, China); forward 50 -CGT GGA AGG ACT CAT GAC CA-30 and reverse 50 -TCC AGG GGT CTTACT CCT TG-30 for GAPDH (Sangon Biological Engineering Technology, Shanghai, China). DNA was amplified immediately with an initial hold step (94 °C for 5 min) and 35 cycles of a three step PCR (94 °C for 1 min,

3.1. Induction of oxidative stress and ER stress in SH-SY5Y cells by H2O2 In our previous studies, we have observed the oxidative cytotoxicity of H2O2 in SH-SY5Y cells and we choose the dose and duration of exposure to H2O2 to reduce SH-SY5Y cell viability by 50% (Yu et al., 2011). The results showed that treatment of SH-SY5Y cells with 400 lM of H2O2 for 20 min resulted in an approximately 50% loss of cellular viability, as compared with control cells, therefore that concentration and time were also used for the present study. To verify oxidative stress in SH-SY5Y cells, here we assayed

J. Ye et al. / Neurochemistry International 78 (2014) 86–95

the changes of intracellular ROS level at different time-points after H2O2 treatment in SH-SY5Y cells by ESR. As shown in Fig. 1A, the intracellular ROS production showed as the mean ROS relatively intensity increased significantly in a time-dependent manner after H2O2 treatment compared with untreated cells (1.5537 ± 0.5033 RIU) and the differences were significant (one-way ANOVA with post hoc Bonferroni test, F = 38.108, P < 0.05–0.001). At 3 h and 24 h after H2O2 treatment, the ROS relatively intensity reached two peak levels (5.1762 ± 0.7549 RIU at 3 h and 3.6333 ± 0.3457 RIU at 24 h, respectively). In tunicamycin-treated cells, the intracellular ROS production also increased in a time-dependent manner and compared with untreated cells, the differences were significant (one-way ANOVA with post hoc Bonferroni test, F = 32.254, P < 0.05–0.001). Different from H2O2 treatment, the time for ROS production is later and the peak level of ROS relatively intensity in tunicamycin-treated SH-SY5Y cells was only at 24 h (3.2014 ± 0.2516 RIU). At 3 h after TM treatment, the difference of ROS level between TM-treated and untreated cells were not significant (1.5688 ± 0.4286 vs 1.5537 ± 0.5033 RIU, P > 0.05). These

89

results confirmed the successful establishment of cell model in SH-SY5Y cells and confirmed that oxidative stress in SH-SY5Y cells could be induced by H2O2 directly or TM, an inducer of ER stress. Meanwhile, we examined whether 400 lM of H2O2 for 20 min has the potential to induce ER stress in SH-SY5Y cells by testing the endoplasmic reticular ultrastructural changes and the expression of endogenous marker for ER stress. As shown in Fig. 1B, SH-SY5Y cells treated with H2O2 presented obvious endoplasmic reticular proliferation, dilation and degranulation, accompanied with some apoptosis characteristic ultrastructural features including chromatin condensation and appearance of chromatin crescent, which morphological changes could also be observed in TM treated cells (data not shown). Western blotting analysis showed that exposure of SH-SY5Y cells to H2O2 resulted in significant induction of ER stress markers, which are positive in TM-treated cells (Fig. 1C). Expression of GRP78, an important molecular chaperone located in the ER, could be detected as early as 2 h after H2O2 treatment, which was markedly upregulated in a time-dependent

Fig. 1. H2O2 treatment induces oxidative stress and endoplasmic reticulum stress in SH-SY5Y cells. SH-SY5Y cells were treated with TM for 24 h or 400 lM H2O2 for 20 min, and further incubated in fresh growth medium for 24 h. (A) Time-dependent intracellular ROS accumulation was measured by ESR assay. The data were represented as intensity of ROS production of cells. Each bar represents the mean ± S.E.M of triplicate microcultures. ⁄P < 0.05 compared to non-treated cells by Student’s t test. (B) The morphological endoplasmic reticular ultrastructural changes in SH-SY5Y cells were observed under transmission electric microscopy. (a) And (c) for control non-treated cells. (b) and (d) H2O2-treated cells. Cells displayed endoplasmic reticulum proliferation, dilation and degranulation, accompanied with condensed chromatin and apoptotic nuclei. (c) And (d) are the enlargement of the indicated area in (a) and (b) Scale bars = 1 lm. (C) Time-course changes of the expression of GRP78 and CHOP were assayed by Western blotting. ⁄P < 0.05 compared to non-treated cells at time 0 h.

90

J. Ye et al. / Neurochemistry International 78 (2014) 86–95

manner (one-way ANOVA with post hoc Bonferroni test, F = 34.075, P < 0.05–0.001, Fig. 1C). Similar time-dependent results were observed for CHOP expression (one-way ANOVA with post hoc Bonferroni test, F = 469.858, P < 0.05–0.001, Fig. 1C), which could be detected as early as 4 h after H2O2 treatment. Time-lapse experiments revealed that substantial induction of CHOP was observed within 24 h, whereas induction of GRP78 was observed after 12 h (Fig. 1C). These results suggest that H2O2 triggered ER stress in SH-SY5Y cells and the pro-apoptotic pathway involving CHOP may play an important role in H2O2-induced apoptosis.

3.2. CHOP is essential for H2O2-induced SH-SY5Y cells apoptosis CHOP is well known for its proapoptotic role during ER stress (McCullough et al., 2001; Oyadomari and Mori, 2004). Therefore, we used siRNA to downregulate CHOP expression and confirm the role of CHOP in H2O2-induced SH-SY5Y cell apoptosis. As shown in Fig. 2A, after treatment with 400 lM of H2O2 for 20 min and incubation for 24 h, the increased expression of CHOP mRNA was inhibited significantly in CHOP siRNA cells (1.49 ± 0.11-fold vs 2.17 ± 0.12-fold, t test, t = 7.2351, P = 0.002) . The MTT assay showed that CHOP

Fig. 2. CHOP is essential for H2O2-induced apoptosis. SH-SY5Y cells were transfected with 100 nM CHOP siRNA or control siRNA (non-targeting siRNA). For 24 h before treatment with H2O2 (400 lM, 20 min). The cells were analyzed in a number of assays at 24 h, (A) the expression of CHOP mRNA was detected in the cells by RT-PCR and normalized against the intensity of GAPDH. ⁄P < 0.01 vs control siRNA, #P < 0.01 vs H2O2 + Control siRNA-treated group. (B) Effects of CHOP siRNA on H2O2-induced cell viability was detected by MTT assays. ⁄P < 0.01 vs control siRNA, #P < 0.01 vs H2O2 + Control siRNA-treated group. (C) Effects of CHOP siRNA on H2O2-induced apoptosis was detected by Hoechst 33258. Each bar presents the mean percentage of apoptotic cells ± SEM for three independent experiments. ⁄P < 0.01 vs control siRNA, #P < 0.01 vs H2O2 + Control siRNA-treated group. (D) Whole-cell lysates were used for Western blotting (n = 3) with antibodies specific for CHOP, Bax, Bcl-2, Bim and b-actin (loading control). Left: representative immunoblots. Right: densitometric analysis. #P < 0.01 vs H2O2 + Control siRNA-treated group.

J. Ye et al. / Neurochemistry International 78 (2014) 86–95

knockdown effectively increased cell viability (38%) compared with the H2O2 alone-treatment group (0.7697 ± 0.0.0518 vs 0.5010 ± 0.0205, t test, t = 8.3542, P = 0.0011, Fig. 2B). Hoechst 33258 staining demonstrated that H2O2 induced 31.8% apoptosis in the H2O2 alone-treatment group, whereas only 15.2% of the cells in the CHOP knockdown group were apoptotic (t test, t = 8.2637, P = 0.0012, Fig. 2C). Previous work has determined that Bcl-2 and Bim are downstream of CHOP (Ghosh et al., 2012; Puthalakath et al., 2007). Our present study showed that CHOP knockdown increased Bcl-2 expression (88.7 ± 12.95 vs 40.29 ± 12.57, t test, t = 4.6460, P = 0.009) and decreased Bax (205.3 ± 17.02 vs 365.9 ± 10.46, t test, t = 13.8180, P = 0.0001) and Bim levels (BimEL, 151.6 ± 12.82 vs 229.3 ± 16.38, t test, t = 5.5601, P = 0.0051; BimL, 161.7 ± 16.58 vs 209.5 ± 15.27, t test, t = 3.6731, P = 0.0214) in SH-SY5Y cells at 24 h after H2O2 treatment (Fig. 2D). Taken together, these data suggest that H2O2 induced SH-SY5Y cells apoptosis through a CHOP-dependent pathway. 3.3. Inhibition of L-carnitine on H2O2-induced cell apoptosis The effects of L-carnitine on cell cytotoxicity and cell apoptosis were observed by MTT assays and Annexin V-FITC/PI assay. Equal numbers of SH-SY5Y cells were pretreated with L-carnitine (100 lM) for 3 h before H2O2 exposure or TM treatment and then incubation for 24 h for further assay. As shown in Fig. 3A, H2O2 decreased SH-SY5Y cells viability significantly (0.4867 ± 0.0223 vs 0.9610 ± 0.0190, t test, t = 28.0412, P < 0.0001), which could be inhibited by L-carnitine (0.735 ± 0.0386 vs 0.4867 ± 0.0223, t test, t = 9.6474, P = 0.0006). Treatment of SH-SY5Y cells with 10 lM TM for 24 h also resulted in an approximately 37% loss of cellular viability, as compared with control cells (0.604 ± 0.0195 vs 0.9610 ± 0.0190, t test, t = 22.7115, P < 0.0001); while L-carnitine pretreatment significantly increased the viability of SH-SY5Y cells against TM-induced cytotoxicity (0.6807 ± 0.0211 vs 0.604 ± 0.0195, t test, t = 4.6357, P = 0.0098). Meanwhile, cell apoptosis induced by H2O2 were also inhibited by L-carnitine (15.382 ± 3.7766 vs 35.4867 ± 2.6545, t test, t = 7.5436, P = 0.0016) as shown in Fig. 3B. L-carnitine pretreatment could also protect cells against TM-induced cell apoptosis (21.5887 ± 2.0479 vs 29.7443 ± 1.1898, t test, t = 5.0475, P = 0.004). To understand the involvement of ER stress signaling pathway in H2O2-induced cell injury, cells were pretreated with inhibitor of ER stress 4PBA (5 mM, for 6 h) before H2O2 exposure for 24 h. We found that 4-PBA could significantly improve cell viability (0.6337 ± 0.0399 vs 0.4867 ± 0.0223, t test, t = 5.5703, P = 0.0050) and reduced the cell apoptosis (18.3003 ± 1.5161 vs 35.4867 ± 2.6545, t test, t = 9.7377, P = 0.0006) in H2O2-treated SH-SY5Y cells (Fig. 3A and B). Compared with non-treated cells, no notable changes in cell viability of SH-SY5Y cells were observed with addition of L-carnitine, NAC or 4-PBA alone (P > 0.05, data not shown). These results showed that both oxidative stress and ER stress contribute to SH-SY5Y cell injury even cell apoptosis, which could be antagonized by L-carnitine. Antioxidant, NAC and ER stress inhibitor, 4-PBA were also effective for protecting cell viability of SH-SY5Y cells against H2O2 exposure. The effects of 100 lM L-carnitine pretreatment on cell viability and cell apoptosis induced by H2O2 were relatively equal to 5 mM NAC, a known strong ROS scavenger (P > 0.05, Fig. 3A and B). 3.4. Inhibition of L-carnitine on H2O2-induced intracellular ROS accumulation By using ESR and CM-H2DCFDA, a ROS-sensitive fluorescence indicator, the accumulation of intracellular ROS in SH-SY5Y cells was further confirmed. Pretreatment with antioxidant L-carnitine and NAC abrogated ROS accumulation in a time-dependent man-

91

ner (Fig. 3C and D). At 3 h and 24 h, ESR assays showed that the intracellular ROS level in L-carnitine and NAC pretreated cells are significantly less than H2O2 alone-treated cells (t tests, P < 0.05, respectively, Fig. 3C), which confirmed the antioxidative effects of L-carnitine. For TM-treated cells, the significant inhibitory effects of L-carnitine on intracellular ROS level were observed at 24 h by ESR assay and CM-H2DCFDA staining (t tests, P < 0.05, respectively, Fig. 3C and D), which partly indicated that ER stress could induce oxidative stress and further confirmed the inhibitory effects of L-carnitine on ER stress. 4-PBA pretreatment have no significant effects on ROS accumulation at 3 h (5.0767 ± 0.1521 vs 5.1262 ± 0.7004 RIU, t test, t = 0.1692, P = 0.8690, Fig. 3C), while at 24 h 4-PBA pretreatment showed milder inhibitory effects on ROS accumulation than L-carnitine and NAC, which were demonstrated by ESR assays (2.6667 ± 0.2843 vs 3.0666 ± 0.1415 RIU, t test, t = 3.0846, P = 0.0116, Fig. 3C) and CM-H2DCFDA staining (279.03 ± 8.43 vs 394.15 ± 7.63, t test, t = 24.8, P < 0.0001, Fig. 3D). The effects of 100 lM L-carnitine on the ROS production were relatively equal to 5 mM NAC (P > 0.05, Fig. 3C and D). These results also suggested that the peak changes of intracellular ROS accumulation at 3 h mainly caused by direct action of exogenous H2O2 and later parts of ROS at 24 h come from secondary unfolded protein response induced by H2O2. Also, ER stress was the downstream events of H2O2-induced oxidative stress in this model. 3.5. Inhibition of L-carnitine on H2O2-induced ER stress markers To enhance the hypothesis that ER stress is related to protection of L-carnitine, we observed the effects of L-carnitine on H2O2-induced or TM-induced ER stress markers. As shown in Fig. 4 assayed by Western blotting at 12 h, GRP78 and CHOP proteins were upregulated significantly in H2O2-induced or TM-treated group, which could be significantly inhibited by L-carnitine (t tests, P < 0.05, respectively, Fig. 4). Similar to L-carnitine, NAC and 4-PBA also attenuated the expression of GRP78 and CHOP in H2O2-treated SH-SY5Y cells (t tests, P < 0.05, Fig. 4). As an important downstream signal involved in CHOP-dependent ER stress apoptosis pathways, the expression of Bim were observed by Western blotting. The results showed that two longer isoforms of Bim, BimEL and BimL in SH-SY5Y cells treated with H2O2 or TM were highly expressed compared to control cells (t tests, P < 0.05, respectively, Fig. 4). L-carnitine suppressed the expression of BimEL and BimL in SH-SY5Y cells treated with H2O2 or TM for 12 h, as shown in Fig. 4. The effects of NAC and 4-PBA on Bim expression induced by H2O2 are similar to L-carnitine, which suggest that Bim is the downstream signal for H2O2-induced ER stress and involved in H2O2-induced apoptosis and L-carnitine protection. L-Carnitine treatment alone has no significant effects on the expression of GRP78, CHOP, BimEL and BimL in untreated control SH-SY5Y cells (t tests, P > 0.05, respectively, Fig. 4). 3.6. Effects of L-carnitine on H2O2-induced MAPK activation MAPKs are important upstream regulators for cellular proliferation and death, which are sensitive to ROS and related to oxidative stress or ER stress-induced cell apoptosis. Our previous study has shown that H2O2 could induce phosphorylation of proteins of the MAPK family, such as ERK1/2 and JNK, but not p38 in SHSY5Y cells (Ye et al., 2009). In the present study, we observed the effects of L-carnitine, NAC and 4-PBA on H2O2-induced MAPK activation. As shown in Fig. 5, treatment of SH-SY5Y cells with 400 lM H2O2 for 20 min resulted in ERK and JNK phosphorylation. Pretreatment of cells with 100 lM L-carnitine significantly inhibited JNK phosphorylation (t tests, P < 0.05). Less inhibitory effect on H2O2-induced ERK1/2 activation (t tests, P < 0.05) and no effect on p38 activation (t tests, P > 0.05, Fig. 5) were observed in L-carni-

92

J. Ye et al. / Neurochemistry International 78 (2014) 86–95

Fig. 3. L-Carnitine inhibits cell apoptosis and ROS accumulation in H2O2 or TM-treated SH-SY5Y cells. SH-SY5Y cells were pretreated with L-carnitine (100 lM) or NAC (5 mM) for 3 h or 4-PBA (5 mM) for 6 h before H2O2 or TM exposure and then evaluated for cell viability by MTT assays (A), apoptosis measured by Annexin V-FITC/PI assay (B), intracellular ROS accumulation measured by ESR assays (C) and analysis of CM-H2DCFCA-staining (D). The data were expressed as means ± S.E.M of the percentage of untreated control cells from three independent experiments. ⁄P < 0.05 compared to non-treated control cells, #P < 0.05 compared to H2O2-treated cells, NP < 0.05 compared to TM-treated cells, (Student’s t-test).

J. Ye et al. / Neurochemistry International 78 (2014) 86–95

Fig. 4. Inhibitions of L-carnitine on ER stress markers in H2O2 or TM-treated SHSY5Y cells. SH-SY5Y cells were pretreated with L-carnitine (100 lM) or NAC (5 mM) for 3 h or 4-PBA (5 mM) for 6 h before H2O2 or TM exposure and then incubated for 12 h. The expression of GRP78, CHOP, Bcl-2, Bax and Bim were analyzed by western blot analysis. b-Actin from the same sample was detected for controls. Data were expressed as mean values ± S.E.M of a ratio of untreated controls from three independent experiments. ⁄P < 0.05 compared to non-treated control cells, #P < 0.05 compared to H2O2-treated cells, NP < 0.05 compared to TM-treated cells, (Student’s t-test).

tine-treated cells (Fig. 5). The same inhibitory effects were observed in NAC and 4-PBA pretreated cells. In TM treated SHSY5Y cells, ERK, JNK and p38 phosphorylation could be detected obviously, which were inhibited by L-carnitine pretreatment (t tests, P < 0.05, Fig. 5). Combined these together, we interfered that L-carnitine may affect MAPK activation, especially JNK activation and inhibited apoptosis in SH-SY5Y cells upon H2O2 or TM challenge, which confirmed that JNK is the critical mediator in causing neurons death under oxidative stress and ER stress.

4. Discussion As a naturally occurring antioxidant and energy precursor, has been found to be an interesting candidate to protect cells against oxidative stress for neurodegenerative disease (Virmani and Binienda, 2004; Binienda et al., 2004; Gülçin, 2006; Binienda and Ali, 2001; Rani and Pannerelvam, 2001; Dhitavat and Ortiz, 2005; Mazzio et al., 2003; Yu et al., 2011). However, the mechanism underlying its protection is still not fully understood. Our previous study have showed that the neuroprotective effects of L-carnitine against H2O2 toxicity in SH-SY5Y cells were mediated, at least, through scavenging oxygen free radicals, prevention of oxidation of lipids, enforcement of endogenous antioxidant defense, inhibition of cell apoptosis and regulation apoptosis related gene expression of Bcl-2 and Bax (Yu et al., 2011). As Bcl-2 and Bax are the molecular targets located on the cytoplasmic face of the outer mitochondrial membrane and endoplasmic reticulum for L-carnitine and there are cross-talk between oxidative stress

L-carnitine

93

Fig. 5. Inhibition of L-carnitine on the activation of MAPK in H2O2-treated SHSY5Ycells. SH-SY5Y cells were pretreated with L-carnitine (100 lM) or NAC (5 mM) for 3 h or 4-PBA (5 mM) for 6 h before H2O2 exposure and then incubated for 12 h. The activation of MAPK were analyzed by western blot analysis. b-Actin was detected for controls. Data were expressed as mean values ± S.E.M of a ratio of untreated controls from three independent experiments. ⁄P < 0.05 compared to non-treated control cells, #P < 0.05 compared to H2O2-treated cells, NP < 0.05 compared to TM-treated cells, (Student’s t-test).

and ER stress, we assumed that ER may relate to the protective function of L-carnitine. By using our previous cell model in SH-SY5Y cells, we first observe oxidative stress and ER stress caused by H2O2 in SHSY5Y cells. ESR assays showed that H2O2 treatment could induce intracellular ROS accumulation in a time-dependent manner, which could be blocked by L-carnitine or NAC pretreatment. These results confirmed the oxidative stress in SH-SY5Y cells and the antioxidative character of L-carnitine. Further, we found that SHSY5Y cells treated with H2O2 presented obvious endoplasmic reticular proliferation, dilation and degranulation, accompanied with some apoptosis characteristic ultrastructural features, which demonstrated that ER is a sensitive target organelle for H2O2 and oxidative stress, may interfere with ER function. To confirm ER stress, we observed the expression of GRP78, the key ER regulatory protein which functions as a molecular chaperone and plays an important role in the recognition of unfolded proteins. Under normal conditions, GRP78 proteins are expressed at a low level and are bound to ER transmembrane receptors. When ER stress is triggered, the unfolded or misfolded proteins combine with free GRP78 and lead to their activation. Our results showed that H2O2 induced similar effects as TM, an inducer of ER stress, on GRP78 expression in a time-dependent manner, which indicated that H2O2 caused accumulation of unfolded or misfolded proteins and lead to a cellular stress response, ER stress. Intense or persistent ER stress can

94

J. Ye et al. / Neurochemistry International 78 (2014) 86–95

induce apoptosis, in which CHOP functions as a transcriptional factor that regulates genes involved. In our experiment, we found that H2O2 treatment induced a significant upregulation of CHOP protein in SH-SY5Y cells, which can be observed in TM-treated cells. Furthermore, silencing of CHOP by siRNA markedly reduced cell apoptosis and ER-related protein expression induced by H2O2, suggesting that CHOP-mediated ER stress involved in H2O2 induced apoptosis in SH-SY5Y cells. Recent studies have suggested that both oxidative stress and ER stress contributed to the neuronal injury and neuron apoptosis (Higgins et al., 2010; Kanekura et al., 2009). Meanwhile, there are some interesting cross-talk pathways between oxidative stress and ER stress. Acting as a protein folding compartment, ER is a very important organelle for neurons and is susceptible to oxidative stress (Malhotra and Kaufman, 2007). Meanwhile, ER stress could lead to calcium release from the ER and stimulate mitochondrial metabolism to produce more ROS. In the present study, we found there were two peak levels of ROS accumulation at 3 h and 24 h after H2O2 treatment by ESR assays. We inferred the ROS accumulation at 3 h was mainly caused by exogenous H2O2 directly and ROS accumulation at 24 h may be induced by intracellular metabolism. We wonder if ER stress contributes to the later wave changes of ROS in SH-SY5Y cells after H2O2 treatment. In order to explain this, we observed the effect of TM or 4-PBA on the ROS production of SH-SY5Y cells after H2O2 treatment at 24 h, which point is the most marked time for ER stress showed by GRP78 and CHOP expression. ESR assays and CM-H2DCFDA staining showed that 4PBA pretreatment has inhibitory effects on ROS accumulation at 24 h rather than 3 h in H2O2-treated cells and this effect is less than NAC and L-carnitine. Also, ESR assays showed that the difference of ROS level between TM-treated and untreated cells were not significant at 3 h, while the peak level of ROS relatively intensity in tunicamycin treated SH-SY5Y cells were at 24 h. These results help us to infer that in the present cell model, the unfolded protein response pathways may participate in the production of the second peak of H2O2-induced intracellular ROS accumulation at 24 h and ER stress was the downstream events of H2O2-induced oxidative stress in SH-SY5Y cells. To explore the role of ER pathway in L-carnitine function, we observed the effect of L-carnitine on cell viability, apoptosis and the expression of ER stress marker in H2O2-treated or TM-treated cells respectively. The results showed that although L-carnitine treatment alone showed no direct effects on ER-related gene expression in non-treated cells, L-carnitine pretreatment could effectively inhibit cell cytotoxicity, apoptosis and up-expression of GRP78 and CHOP induced by H2O2 exposure in SH-SY5Y cells, which effects are similar to traditional ROS scavenger, NAC or chemical chaperones and inhibitor of ER stress, 4-PBA. These results suggested that H2O2-induced cytotoxicity and apoptosis in SH-SY5Y cells is partly mediated by ER stress and the neuroprotective effects of L-carnitine on H2O2-induced neuronal apoptosis may be related to CHOP-dependent ER stress pathways. Further studies in TM-treated cells confirmed the inhibitory effects of Lcarnitine on ER stress induced cell cytotoxicity, apoptosis and expression of GRP78 and CHOP directly. MAPKs are important upstream regulators of transcription factor activities and their signaling is critical to the transduction of extracellular oxidative stress stimuli into intracellular events and is tightly linked to ER stress (Runchel et al., 2011; Kim and Choi, 2010). JNK, one member of MAPK family, have been demonstrated to be involved in CHOP-dependent ER stress-dependent apoptosis pathways (Scott et al., 2008). In our previous study, we observed that H2O2 induced phosphorylation of proteins of the MAPK family, such as ERK1/2 and JNK, but not p38 (Ye et al., 2009) in SH-SY5Y cells. To further explore the ER stress role in neuroprotection of L-carnitine, we observed the effects of L-carnitine on MAPK

phosphorylation. Our results showed that L-carnitine, NAC and 4PBA treatment showed significant less JNK phosphorylation after H2O2 incubation. Compared to H2O2-treated cells, MAPK pathway shows different extent of activation in TM-treated cells. ERK, JNK and p38 phosphorylation in SH-SY5Y cells were induced by TM and L-carnitine-pretreated cells showed less MAPK activation than TM-treated cells. These results suggested the complicated role of MAPK pathways response to oxidative stress and ER stress. Considering different response of MAPK activation in H2O2 or TM-treated cells and the key role of MAPK pathway in apoptosis, we inferred that the inhibitory effects of L-carnitine on cell apoptosis upon H2O2 or TM challenge are related to MAPK activation. Although the exact mechanism of inhibition of H2O2-induced phosphorylation of MAPK proteins by L-carnitine may not be clearly explained on the basis of the present set of data, it seems that the neuroprotective property of L-carnitine mainly contributed to the inhibition of phosphorylation of JNK as JNK is the critical mediator in causing neurons death under oxidative stress and ER stress. The exact target of these pathways in inhibition of L-carnitine on MAPK activation of neuronal cells should be identified in our further study. Bcl-2, Bax and Bim are members of the large Bcl-2 family located on the cytoplasmic face of the outer mitochondrial membrane and endoplasmic reticulum, which were involved in cell apoptosis (Vela et al., 2013). In response to severe ER stress, specific Bcl-2 and Bcl-2 homology domain 3 (BH3)-only family members are targeted to initiate apoptosis (Jiang et al., 2004). Prototypical Bcl-2 inhibits cell death by binding and inactivating proapoptotic member such as Bax. BH3 only-containing proteins like Bim indirectly activate Bax by binding Bcl-2 (through the BH3 motif), thereby releasing Bax from the complex. Bax then permeabilizes the mitochondrial outer membrane, allowing cytochrome C release to the cytoplasm. Under ER stress, Bax also interacts with and activates IRE1a. IRE1a then signals to JNK to simultaneously activate Bim and inhibit Bcl-2 (Ariazi et al., 2011). A variety of ER stress inducer stimulates Bim expression, and Bim is essential in ER stress-induced apoptosis in a wide range of cell types (Puthalakath et al., 2007; Ariazi et al., 2011). We previously verified that H2O2 could downregulate Bcl-2 and upregu-

Fig. 6. Proposed schematic illustration of the possible mechanism underlying the protective effects of L-carnitine against ER stress induced by oxidative stress in SHSY5Y cells. There are four possible stages to the process. (1) H2O2 induces ER stress and oxidative stress; (2) oxidative stress leads to up-regulation of p-ERK and p-JNK; ER stress leads to the ER stress marker GRP78, CHOP and p-JNK; (3) CHOP and p-JNK induces apoptosis and up-regulation of BH3-only proapoptotic protein Bim and Bax and inhibits the expression of Bcl-2. (4) Bim and Bax induces cell apoptosis (showed by solid arrows). L-Carnitine exerts its protection against oxidative stress induced by ROS (H2O2). The protection of L-carnitine against ER stress induced by oxidative stress may be related to the expression of GRP78, CHOP, JNK, Bcl-2, Bax and Bim (possible effects showed by dotted arrows).

J. Ye et al. / Neurochemistry International 78 (2014) 86–95

late Bax protein expression, and this effect was blocked by the pretreatment of L-carnitine (Yu et al., 2011). In the present study, we observed the role of Bim in H2O2-induced neuronal apoptosis. The results showed that L-carnitine suppressed the expression of BimEL and BimL in SH-SY5Y cells induced by H2O2 or TM, which suggest that Bim is the downstream signal for H2O2-induced ER stress and involved in H2O2-induced apoptosis and L-carnitine may inhibit H2O2-induced SH-SY5Y cell apoptosis by affecting the expression of Bim. In the present study, our results showed that L-carnitine alone have no significant effects on GRP78, CHOP, Bax and Bim gene expression in non-stressed SH-SY5Y cells, while L-carnitine may affect stress-related gene expression as the cell is in stress-state, such as oxidative stress or ER stress. These results suggested that the effects of L-carnitine on ROS-related gene expression maybe related to its antioxidative character. Furthermore, as Bcl-2 family also located on mitochondrial, which are both the target and the source of ROS (Vela et al., 2013). The most biological function of L-carnitine is in the transport of fatty acids into the mitochondria for subsequent b-oxidation (Nalecz and Nalecz, 1996; Virmani and Binienda, 2004). Whether the effects of L-carnitine on stress-related gene expression are direct or through mitochondrial pathways need our further study. Based on the previous findings and the current results, we propose a possible model for the mechanism of L-carnitine protection against ER stress induced by oxidative stress in the present study. This possible model is presented in Fig. 6. In conclusion, the present study provides further evidences of the connection between oxidative stress and ER stress in neuronal injury. More importantly, we found that H2O2 could induce apoptosis in SH-SY5Y cells through the ER stress pathway and the protective function of L-carnitine may be via alleviation of ER stress by regulating the expression of CHOP, JNK, Bcl-2, Bax and Bim. Our results showed that the antioxidant effect may be a major mechanism for L-carnitine-mediated neuroprotection and ER stress contributed to it. These data provide new understanding into the mechanisms contributing to L-carnitine neuroprotection, a function with potential therapeutic relevance to many age-related neurodegenerative disorders. Declaration of interest All authors in this paper declare that there are no conflicts of interest in this research. Acknowledgments The work was supported by Research Award Fund for Outstanding Middle-aged and Young Scientist of Shandong Province (Grant No. BS2012YY004), the Shandong Provincial Natural Science Foundation, China (Grant Nos. Q2008C04 and ZR2010HL068) and the National Natural Science Foundation of China (Grant Nos. 31100824 and 81473384). References Ariazi, E.A., Cunliffe, H.E., Lewis-Wambi, J.S., Slifker, M.J., Willis, A.L., Ramos, P., Tapia, C., Kim, H.R., Yerrum, S., Sharma, C.G., Nicolas, E., Balagurunathan, Y., Ross, E.A., Jordan, V.C., 2011. Estrogen induces apoptosis in estrogen deprivation-resistant breast cancer through stress responses as identified by global gene expression across time. Proc. Natl. Acad. Sci. U.S.A. 108 (47), 18879– 18886. Binienda, Z.K., Ali, S.F., 2001. Neuroprotective role of L-carnitine in the 3nitropropionic acid induced neurotoxicity. Toxicol. Lett. 125, 67–73.

95

Binienda, Z., Virmani, A., Przybyla-Zawislak, B., 2004. Neuroprotective effect of Lcarnitine in the 3-nitropropionic acid (3-NPA)-evoked neurotoxicity in rats. Neurosci. Lett. 367, 264–267. Calabrese, V., Cornelius, C., Dinkova-Kostova, A.T., Calabrese, E.J., Mattson, M.P., 2010. Cellular stress responses, the hormesis paradigm, and vitagenes: novel targets for therapeutic intervention in neurodegenerative disorders. Antioxid. Redox Signal. 13 (11), 1763–1811. Dhitavat, S., Ortiz, D., Rogers, E., Rivera, E., Shea, T.B., 2005. Folate, vitamin E, and acetyl-L-carnitine provide synergistic protection against oxidative stress resulting from exposure of human neuroblastoma cells to amyloid-beta. Brain Res. 1061, 114–117. Ghosh, N., Ghosh, R., Mandal, S.C., 2011. Antioxidant protection: a promising therapeutic intervention in neurodegenerative disease. Free Radical Res. 45 (8), 888–905. Ghosh, A.P., Klocke, B.J., Ballestas, M.E., Roth, K.A., 2012. CHOP potentially cooperates with FOXO3a in neuronal cells to regulate PUMA and BIM expression in response to ER stress. PLoS One 7, e39586. Gibson, G.E., Huang, H.M., 2004. Mitochondrial enzymes and endoplasmic reticulum calcium stores as targets of oxidative stress in neurodegenerative diseases. J. Bioenerg. Biomembr. 36 (4), 335–340. Gülçin, I., 2006. Antioxidant and antiradical activities of L-carnitine. Life Sci. 78, 803–811. Halliwell, B., 2006. Oxidative stress and neurodegeneration: where are we now? J. Neurochem. 97, 1634–1658. Higgins, G.C., Beart, P.M., Shin, Y.S., Chen, M.J., Cheung, N.S., Nagley, P., 2010. Oxidative stress: emerging mitochondrial and cellular themes and variations in neuronal injury. J. Alzheimers 20 (Suppl. 2), S453–S473. Jiang, Z., Zheng, X., Lytle, R.A., Higashikubo, R., Rich, K.M., 2004. Lovastatin-induced up-regulation of the BH3-only protein, Bim, and cell death in glioblastoma cells. J. Neurochem. 89 (1), 168–178. Kanekura, K., Suzuki, H., Aiso, S., Matsuoka, M., 2009. ER stress and unfolded protein response in amyotrophic lateral sclerosis. Mol. Neurobiol. 39 (2), 81–89. Kim, E.K., Choi, E.J., 2010. Pathological roles of MAPK signaling pathways in human diseases. Biochim. Biophys. Acta 1802 (4), 396–405. Le Bel, C., Ischiropoulos, H., Bondy, S.C., 1992. Evaluation of the probe 20 ,70 dichlorofluorescin as an indicator of reactive oxygen species formation and oxidative stress. Chem. Res. Toxicol. 5, 227–231. Liu, D., Zhang, M., Yin, H., 2013. Signaling pathways involved in endoplasmic reticulum stress-induced neuronal apoptosis. Int. J. Neurosci. 123 (3), 155–162. Logue, S.E., Cleary, P., Saveljeva, S., Samali, A., 2013. New directions in ER stressinduced cell death. Apoptosis 18 (5), 537–546. Malhotra, J.D., Kaufman, R.J., 2007. Endoplasmic reticulum stress and oxidative stress: a vicious cycle or a double-edged sword? Antioxid. Redox Signal. 9 (12), 2277–2293. Mazzio, E., Yoon, K.J., Soliman, K.F., 2003. Acetyl-L-carnitine cytoprotection against 1-methyl-4-phenylpyridinium toxicity in neuroblastoma cells. Biochem. Pharmacol. 66, 297–306. McCullough, K.D., Martindale, J.L., Klotz, L.O., Aw, T.Y., Holbrook, N.J., 2001. Gadd153 sensitizes cells to endoplasmic reticulum stress by down-regulating Bcl-2 and perturbing the cellular redox state. Mol. Cell. Biol. 21, 1249–1259. Nalecz, K.A., Nalecz, M.J.I., 1996. Carnitine-a known compound, a novel function in neural cells. Acta Neurobiol. Exp. 56, 597–609. Oyadomari, S., Mori, M., 2004. Roles of CHOP/GADD153 in endoplasmic reticulum stress. Cell Death Differ. 11, 381–389. Puthalakath, H., O’Reilly, L.A., Gunn, P., Lee, L., Kelly, P.N., Huntington, N.D., Hughes, P.D., Michalak, E.M., McKimm-Breschkin, J., Motoyama, N., Gotoh, T., Akira, S., Bouillet, P., Strasser, A., 2007. ER stress triggers apoptosis by activating BH3only protein Bim. Cell 129 (7), 1337–1349. Rani, P.J., Pannerelvam, C., 2001. Carnitine as a free radical scavenger in aging. Exp. Gerontol. 36, 1713–1726. Runchel, C., Matsuzawa, A., Ichijo, H., 2011. Mitogen-activated protein kinases in mammalian oxidative stress responses. Antioxid. Redox Signal. 15 (1), 205–218. Scheper, W., Hoozemans, J.J., 2009. Endoplasmic reticulum protein quality control in neurodegenerative disease: the good, the bad and the therapy. Curr. Med. Chem. 16 (5), 615–626. Scott, D.W., Longpre, J.M., Loo, G., 2008. Upregulation of GADD153 by butyrate: involvement of MAPK. DNA Cell Biol. 27 (11), 607–614. Stefani, I.C., Wright, D., Polizzi, K.M., Kontoravdi, C., 2012. The role of ER stressinduced apoptosis in neurodegeneration. Curr. Alzheimer Res. 9 (3), 373–387. Vela, L., Gonzalo, O., Naval, J., Marzo, I., 2013. Direct interaction of Bax and Bak proteins with Bcl-2 homology domain 3 (BH3)-only proteins in living cells revealed by fluorescence complementation. J. Biol. Chem. 288 (7), 4935–4946. Virmani, A., Binienda, Z., 2004. Role of carnitine esters in brain neuropathology. Mol. Aspects Med. 25, 533–549. Ye, J., Han, Y., Wang, C., Yu, W., 2009. Cytoprotective effect of polypeptide from Chlamys farreri on neuroblastoma (SH-SY5Y) cells following HO exposure involves scavenging ROS and inhibition JNK phosphorylation. J. Neurochem. 111 (2), 441–451. Yu, J., Ye, J., Liu, X., Han, Y., Wang, C., 2011. Protective effect of L-carnitine against H(2)O(2)-induced neurotoxicity in neuroblastoma (SH-SY5Y) cells. Neurol. Res. 33 (7), 708–716.