Prevention of PC-12 Cell Death by Crocin via Sphingomyelinase- Ceramide Signaling by Increase of Glutathione Biosynthesis

Crocetin glycoside, crocin is a pharmacologically active component of Crocus sativus L. (saffron) that has been used in traditional Chinese medicine. Previously we demonstrated that crocin inhibited apoptosis in PC-I2 cells by affecting the function of tumor necrosis factor-α. In this study, we found that depriving cultured PC-12 cells of serum/glucose causes a rapid increase in cellular ceramide levels, followed by an increase in the phosphorylation of c-jun kinase (JNK). The accumulation of ceramide was found to depend on the activation of magnesium-dependent neutral sphingomyelinase (N-SMase), but not on de novo synthesis. The serum/glucosedeprived PC-12 cells also decreased the cellular levels of glutathione (GSH), which is the potent inhibitor of N-SMase. Treating the PC-12 cells with crocin prevented NSMase activation, ceramide production, and JNK phosphorylation. We also found that the chemical can enhance the activities of GSH reductase and glutamylcysteinyl synthase (γ-GCS), contributing to a stable GSH supply that blocks the activation of N-SMase. Thus our data suggest that crocin combats the serum/glucose deprivationinduced ceramide formation in PC-12 cells by increasing GSH levels and prevents the activation of JNK pathway, which is reported to have a role of the signaling cascade downstream ceramide for neuronal cell death. INTRODUCTION It has been evident that glutathione (GSH) is a major non-protein antioxidant that protects neurons from oxidative stress, and the depletion is an early event that precedes the onset of apoptosis induced by various agents (Beaver and Waring, 1995). A possibility of cell death mechanism may be occurred by the activation of magnesiumdependent neutral sphingomyelinase (N-SMase) caused by a drop in GSH levels, followed by a rapid increase in cellular ceramide levels (Liu and Hannun, l997; Liu et al., 1998; Yoshimura et al., 1999). Recently it becomes clear that ceramide stimulates the stress-activated protein kinase/c-jun kinase (SAPK/INK) signaling system that precedes induction of apoptosis (Verheij et al. 1996). Willaime-Morawek et al. (2003) have further investigated the role of JNK pathway in ceramide-induced cell death and demonstrated that p38 kinase and JNK/c-Jun act in parallel to induce neuronal apoptosis by ceramide. Several types of SMases have been identified; these include lysosomal and secreted acidic SMases (A-SMase) and cytosolic, magnesium-independent N-SMase (Levade and Jaffrezou, 1999; Perry, 1999). However, the membrane-bound magnesium-dependent NSMase may be the SMase most closely linked to the initial production of ceramide induced by depletion of GSH (Liu and Hannun, 1997; Liu et al., 1998; Yoshimura et al., 1999; Lavrentiadou et al., 2001). GSH is synthesized in theγ-glutamyl cycle, in which γglutamylcysteinyl synthase (γ-GCS) is the rate-limiting enzyme (Griffith et al., 1979; Pan and Pérez-Polo, 1993). Glutathione peroxidase (GPx) removes H202 and organic peroxides by converting GSH to the oxidized form (GSSG), whereas glutathione reductase (GR) regenerates GSH in the presence of NADPH. Nerve growth factor (NGF) Proc. I IS on Saffron Eds: J.-A. Fernández & F. Abdullaev Acta Hort 650, ISHS 2004 424 prevents the PC-12 cell death from oxidative stress by increasing theγ-GCS activity (Pan and Pérez-Polo, 1993). Interleukin-6 (IL-6) is also suggested to prevent PC-12 cell death by increasing the expression ofγ-GCS mRNA, which is followed by an increase in cellular GSH levels (Nakajima et al. 2002). Therefore, chemicals having roles of these neurotrophic factors may be clinically useful. Crocus sativus L. (saffron) is used for anodyne, traquid and emmenagogue in traditional Chinese medicine. One of the active components, crocin, is a carotenoid pigment, and has the structure of crocetin di-gentiobiose ester. We previously published that crocin exhibits a variety of pharmacological effects in mice including inhibition of skin tumor growth (Konoshima et al., 1998), improvement of learning behavior previously impaired by ethanol (Sugiura et al., 1995), and prevention of long-term potentiation caused by ethanol in rats (Sugiura et al., 1994). Here we have been investigating the effect of crocin on the N-SMase-ceramide signaling pathway. Furthermore, we show that crocin prevents the death of PC-12 cells induced by serum/glucose deprivation. The results suggest GSH-dependent inhibition of N-SMaseceramide signaling via the enhancement of both GR and γ-GCS activities MATERIALS AND METHODS The rat pheochromocytoma cell line (PC-12) was obtained from the RIKEN cell bank, Ibaragi, Japan. Stock cultures of undifferentiated PC-12 cells were maintained in DMEM supplemented with 10% (v/v) fetal bovine serum, 10% horse serum, 50 units/ml penicillin, and 100μg/ml streptomycin at 37oC in a humidified 5% CO2 atmosphere. Prior to experiments, the growing cells were seeded in collagen-coated six-well plates and differentiated to neuronal cells in the above medium plus 50ng/ml NGF at 37oC for 7 days. After removal of the culture media, the differentiated PC-12 cells were rinsed twice with glucose-free DMEM. To induce cell death, the cells were immersed in serum/glucose-free DMEM without NGF. The experimental cells were put in medium containing crocin at various concentrations. They were incubated at 37°C for set periods ranging from 1 to 24 h. The amounts of C16-ceramide in PC-12 cells were measured using the liquid chromatography/ion spray ionization mass spectrometry (LC/MS) procedure previously reported (Soeda et al., 2001a). Western blot analysis of phosphorylated JNK Both treated and untreated cells were centrifuged, then washed with ice-cold PBS. Extraction followed for 10mm in 100 p.1 of a lysis buffer, 10mM HEPES buffer (pH 7.4) 150mM KClI3 mM magnesium acetate/0.3 mM EDTA/l00 μ.M phenylmethylsulfonyl fluoride/10% glyceroll0.5% Nonidet P-40. After centrifugation at 12,000 x g for 20 mm at 4o C, the resulting supernatants were tested for the presence of phosphorylated JNK. Protein concentration was measured with a BCA Protein Assay kit (Sigma). Equal amounts of the proteins (20 p.g each) were separated by 8% SDS-polyacrylamide gel electrophoresis and electrophoretically transferred to polyvinylidene difluoride membrane (BIORAD). The transferred protein was reacted with goat polyclonal antibodies against phosphorylated JNK 1 (Thr 183/Tyr 185) obtained from Santa Cruz Biotechnology Inc. The resulting immunocomplex was further reacted with peroxidase-conjugated secondary antibodies and made visible with 4-chloro-l-naphthol as the peroxidase substrate. Intracellular GSH levels were measured by using a Glutathione Assay kit (Cayman Chemical Co., Ann Arbor, Ml). The cells were centrifuged, then washed with ice-cold PBS, and homogenized in 1 ml of MES buffer (pH 6.0)/i mM EDTA. After centrifugation at 10,000 x g for 15 mm at 4oC, the supernatants were collected. Following removal of the proteins from the supernatant solutions, GSH levels were measured by the rate of colorimetric change of 5,5’-dithiobis-2-nitrobenzoic acid (Eliman' s reagent) at 405 nm according to the manufacturer's instruction. Cells were centrifuged, then washed with ice-cold PBS and homogenized in 500 μ1 of a lysis solution containing 5mM EDTA/0.0l% digitoninl0.25% sodium cholate.