Astroglial expression of ceramide in Alzheimer's disease brains: A role during neuronal apoptosis

Neuroscience 130 (2005) 657– 666

ASTROGLIAL EXPRESSION OF CERAMIDE IN ALZHEIMER’S DISEASE BRAINS: A ROLE DURING NEURONAL APOPTOSIS H. SATOI,a1 H. TOMIMOTO,a R. OHTANI,a T. KITANO,b T. KONDO,b M. WATANABE,b N. OKA,a I. AKIGUCHI,a S. FURUYA,c Y. HIRABAYASHIc AND T. OKAZAKIb*

These results suggest a regulatory mechanism of intracellular ceramide through inhibition of GlcCer synthase and a possible role of ceramide as an extracellular/intercellular mediator for neuronal apoptosis. The increased ceramide level in the CSF from AD patients, which may be derived from astroglia, raises a possibility of neuronal apoptosis by the response to intercellular ceramide in AD. © 2005 Published by Elsevier Ltd on behalf of IBRO.

a Department of Neurology, Graduate School of Medicine, Kyoto University, 54-Shogoin-Kawahara-cho, Sakyo-ku, Kyoto 606-5807, Japan b Department of Hematology/Oncology, Graduate School of Medicine, Kyoto University, 54-Shogoin-Kawahara-cho, Sakyo-ku, Kyoto 6065807, Japan c

Neuronal Circuit Mechanism Research Group, RIKEN Brain Science Institute, Wako 351-0198, Japan

Key words: ceramide, apoptosis, cerebrospinal fluid, Alzheimer’s disease, astroglia.

Abstract—Accumulating evidences indicate that ceramide is closely involved in apoptotic cell death in neurodegenerative disorders and aging. We examined ceramide levels in the cerebrospinal fluid (CSF) or brain tissues from patients with neurodegenerative disorders and the mechanism of how intra- and extracellular ceramide was regulated during neuronal apoptosis. We screened the ceramide levels in the CSF of patients with neurodegenerative disorders, and found that ceramide was significantly increased in patients with Alzheimer’s disease (AD) than in patients with age-matched amyotrophic lateral sclerosis (ALS) and other neurological controls. With immunohistochemistry in AD brains, ceramide was aberrantly expressed in astroglia in the frontal cortices, but not detected in ALS and control brains. To explore for the regulation of ceramide in astroglia in Alzheimer’s disease brains, we examined the metabolism of ceramide during neuronal apoptosis. In retinoic acid (RA)induced neuronal apoptosis, RA slightly increased de novo synthesis of ceramide, but interestingly, RA dramatically inhibited conversion of [14C] ceramide to glucosylceramide (GlcCer), suggesting that the increase of ceramide mass is mainly due to inhibition of the ceramide-metabolizing enzyme GlcCer synthase. In addition, a significant increase of the [14C] ceramide level in the culture medium was detected by chasing and turnover experiments without alteration of extracellular [14C] sphingomyelin levels. A 2.5-fold increase of ceramide mass in the supernatant was also detected after 48 h of treatment with RA.

Apoptosis is an important part of normal development in CNS. For example, more than 40% of neurons in the CNS undergo apoptosis as a part of natural pattern formation over a few days in developing vertebrate embryos (Oppenheim, 1991). Neuronal death by apoptotic process may be involved also in neurodegenerative disorders such as Alzheimer’s disease (AD), amyotrophic lateral sclerosis (ALS) and Parkinson’s disease (Ariga et al., 1998; Thompson, 1995; Waggie et al., 1999). This possibility is substantiated by the fact that astroglia or hippocampal neurons exposed to ␤-amyloid show characteristic changes of apoptosis (Loo et al., 1993; Malchiodi-Albedi et al., 2001). In addition, intracellular incorporation of mutant superoxide dismutase, an enzyme responsible for familial ALS, by liposomes causes extensive nitrotyrosine accumulation and apoptotic death in motor neurons (Beckman et al., 2001). Neuronal apoptosis is triggered by many extracellular stresses such as ionizing radiation, ultraviolet (UV) light, growth factor withdrawal and low K⫹. A major mechanism in these disorders is thought to involve reactive oxygen species (ROS)-related pathways (Ferrante et al., 1997; Goswami and Dawson, 2000); however, the precise mechanism leading to neuronal apoptosis in vitro and in vivo remains unclear. Sphingolipids such as ceramide, sphingomyelin (SM) and glycosphingolipids are a family of cellular lipids which contain the sphingosine structure as the hydrophobic moiety (Spiegel et al., 1996). Although sphingolipids are expressed abundantly in the vertebrate CNS (Tettamanti and Riboni, 1993), their functions have been confined to a structural role, particularly in the nonmitotic cell population of neurons. However, the SM/ceramide signaling cascade called the SM cycle, in which extracellular signaling agents results in the activation of sphingomyelinase, cleavage of membrane SM, formation of ceramide and the activation of multiple cellular and biochemical targets, was recently proposed as one of the important mechanisms involved in apoptosis, cell differentiation, cell growth and senescence

1

Present address: Department of Neurology, Takeda General Hospital, Kyoto 601-1495, Japan. *Corresponding author. Tel: ⫹11-81-75-751-3154; fax: ⫹11-81-75751-3154. E-mail address: [email protected] (T. Okazaki). Abbreviations: AD, Alzheimer’s disease; ALS, amyotrophic lateral sclerosis; C6-NBD-Cer, 6-{[(N-7-nitrobenz-2-oxa-1,3-diazol-4-yl) amino] caproyl}; CDR, clinical dementia rating; CSF, cerebrospinal fluid; DAB, diaminobenzidine tetrahydro-chloride; DAG, diacylglycerol; DAPI, 4=,6-diamidino-2-phenylindole dihydrochloride; FBS, fetal bovine serum; GCS, glucosylceramide synthase; GFAP, glial fibrillary acidic protein; GlcCer, glucosylceramide; MEM, minimal essential medium; NB2a, neuroblastoma Neuro2A; PBS, phosphate-buffered saline; RA, retinoic acid; ROS, reactive oxygen species; S-1-P, sphingosine-1-phosphate; SM, sphingomyelin; SMS, sphingomyelin synthase; TLC, thin layer chromatography; UDP-Glc, uridine diphosphate glucose; UV, ultraviolet. 0306-4522/05$30.00⫹0.00 © 2005 Published by Elsevier Ltd on behalf of IBRO. doi:10.1016/j.neuroscience.2004.08.056

657

658

H. Satoi et al. / Neuroscience 130 (2005) 657– 666

(Obeid et al., 1993; Okazaki et al., 1990; Gottschalk et al., 1995). In many cell systems, intracellular ceramide generated by stresses such as tumor necrosis factor-␣, anti-Fas cross-linkage, ionizing radiation, UV, anti-cancer agents and heat shock has been shown to be an intracellular mediator for apoptosis through ceramide generation (Hannun and Obeid, 1995; Jayadev et al., 1995; Venable et al., 1995; Hannun, 1996; Verheij et al., 1996). Signaling molecules including ceramide-activated protein kinases and phosphatases, N-terminal jun kinase, protein kinase C ␨, transcription factors and the caspase family have been shown to act downstream of the ceramide signal. However, the precise nature of the intracellular targets of ceramide is controversial and seems to vary depending on the cell types and stresses (Hannun, 1996; Okazaki et al., 1998). Ceramide is believed to have beneficial effects at low levels, whereas at higher levels it may cause cell death (Kolesnick and Kronke, 1998; Hannun and Luberto, 2000). Indeed, intracellular ceramide was required for dendritic differentiation and survival of Purkinje cells (Furuya et al., 1998; Irie and Hirabayashi, 1998). On the contrary, exogenous ceramide induces apoptosis in neurons and astroglia in culture (Brugg et al., 1996; Mangoura and Dawson, 1998; Sastry and Rao, 2000). There are accumulating evidences that ceramide is intimately involved in causing neurodegenerative disorders, stroke and aging (Chen et al., 2001; Han et al., 2002; Yu et al., 2000). Based on these findings, we believe that the ceramide regulation plays an important role in neurological diseases. We here examine the levels of ceramide in CSF and brain tissues in AD and ALS patients, and further investigate how ceramide and its regulation are involved in neuronal apoptosis in culture. Since we found the increase of ceramide level in CSF and brain tissue astroglia obtained from AD patients as well as the increase of intra- and extracellular ceramide levels in retinoic acid (RA)-induced neuronal apoptosis, a possible role of ceramide as an extracellular or intercellular proapoptotic mediator is discussed.

EXPERIMENTAL PROCEDURES Reagents and solvents [␥-32P] ATP (6000 Ci/mM) and L-[U-14C] serine (150 mCi/mM) were purchased from Amersham Life Science (Arlington Heights, IL, USA). Dr. Y. Hannun (Duke University) kindly provided diacylglycerol (DAG) kinase. Uridine diphosphate glucose (UDP-Glc) was obtained from NEN Life Science Products, Inc. (Boston, MA, USA). 6-{[(N-7-nitrobenz-2-oxa-1,3-diazol-4-yl) amino] caproyl} sphingosine (C6-NBD-Cer) was purchased from Molecular Probes (Eugene, OR, USA), and silica gel thin layer chromatography (TLC) plates from Whatman (Clifton, NJ, USA). Other reagents were obtained from Sigma (St. Louis, MO, USA). Solvents were purchased from Nacalai Tesque Co. (Kyoto, Japan).

Cerebrospinal fluid (CSF) samples Samples of CSF were obtained by lumbar puncture. The diagnosis of 16 AD patients was made based on the DSM-IV and graded by clinical dementia rating (CDR). The diagnosis of 13 ALS patients was made clinico-pathologically, and retrospectively met the

clinical criteria proposed by the Ministry of Health and Welfare in Japan. Fourteen control patients consisted of three with cervical spondylosis, three with psychosomatic disease, three with asymptomatic lacunar infarct, two with tension-type headache, two with metabolic encephalopathy and one with past history of transient ischemic attack. None of these patients had evidence of general inflammation or pleocytosis in the CSF. All CSF samples, while clear to inspection, were centrifuged at 100,000⫻g for 15 min before ceramide measurement.

Brain samples and immunohistochemistry Two patients with AD and two with ALS were compared with two non-neurological control patients, in which one had hepatic cancer and the other had renal cancer. The diagnosis of AD was made based on the Consortium to Establish a Registry for Alzheimer’s Disease (CERAD) diagnostic neuropathologic criteria (Mirra et al., 1991). The ALS patients conformed to the electrophysiological diagnostic criteria for ALS in a premortem period (Brooks et al., 2000). These brains were perfusion fixed in 4% paraformaldehyde and 0.35% glutaraldehyde in 0.1 M PB, pH 7.4 for 20 min and postfixed in 4% paraformaldehyde in 0.1 M PB for 24 – 48 h. Tissue blocks from the frontal lobe were snap frozen and cut with a freezing microtome at 30 ␮m thickness. Frozen sections were incubated with a primary antibody against ceramide (14 ␮g/ml), which was kindly provided by Azwell Co., Ltd., Osaka, Japan (Kawase et al., 2002) overnight at 4 °C. Astroglia were stained with a monoclonal antibody against glial fibrillary acidic protein (GFAP; Dakoppats; ⫻500) and activated microglia with a monoclonal antibody against HLA-DR (Dakopatts; ⫻100). After incubation with the individual primary antibodies, sections were treated with appropriate biotinylated secondary antibodies (Vector Laboratories; ⫻200) and avidin biotin complex (Vector; ⫻200) in 0.1 M PBS containing 0.3% Triton X-100. The sections were finally incubated in 0.01% diaminobenzidine tetrahydro-chloride (DAB) and 0.005% H2O2 in 50 mM Tris-HCl. For the double labeling immunohistochemistry, the first staining cycle for ceramide was colorized for 10 min with a solution of 0.02% DAB, 0.6% nickel ammonium sulfate and 0.00005% H2O2 in 50 mM Tris-HCl. After incubation in 0.5% H2O2 in 50 mM Tris-HCl for 20 min, the sections were then processed for the second staining cycle for either GFAP or HLA-DR, and colorized for 5 min in 0.02% DAB and 0.1% H2O2 in 50 mM Tris-HCl. To test the specificity of the immunohistochemical reaction, the control sections were treated with normal mouse IgG or normal rabbit serum instead of the primary antibodies.

Cell culture The murine neuroblastoma cell line clone NB2a (Neuro2A, CCL131; American Cell Type Culture Collection) was used. Cells were cultured in dishes in minimal essential medium (MEM) supplemented with 10% fetal bovine serum (FBS), 4 mM L-glutamine, 100 units/ml potassium penicillin G, and 100 ␮g/ml streptomycin sulfate in a humidified 5% CO2 and 95% air atmosphere. Cells were plated at 1.75⫻105 cells/ml, and cell numbers were assessed by the 0.05% Trypan Blue dye exclusion method under microscopic observation. After cells were cultured for 24 h, they were resuspended in serum-free MEM containing 40 ␮M RA in absolute ethanol (final concentration, 0.2%), which was found to be the optimal condition for detecting RA-induced apoptosis in preliminary experiments, and incubated for the indicated times. At the concentration used, ethanol alone was neither toxic nor protective for the cells (data not shown).

Ceramide measurement After harvesting the cells and supernatant at the indicated times, the mass of ceramide was measured as previously described

H. Satoi et al. / Neuroscience 130 (2005) 657– 666

659

(Okazaki et al., 1990; Takeda et al., 1999) with a slight modification. Extraction of lipids by the Bligh-Dyer method (Bligh and Dyer, 1959) and ceramide measurement using DAG kinase were performed as described (Takeda et al., 1999). Labeled lipids (ceramide 1-phosphate) were separated by TLC in chloroform:acetone:methanol:acetic acid:water (10:4:3:2:1, v/v). Quantification of ceramide 1-phosphate was carried out by autoradiography and densitometric scanning using an Imaging Analyzer (Model BAS2000; Fuji, Japan) and software provided with the instrument by the manufacturer. Although we measured the phosphate in each sample in a preliminary test, the levels in the cell supernatant and CSF were below the limit of detection. Therefore, we did not need to correct the measured level of ceramide.

Cytochemical staining of apoptotic cells Morphological changes in the nuclear chromatin of cells undergoing apoptosis were detected by staining with 4=,6-diamidino-2phenylindole dihydrochloride (DAPI; Nacalai Tesque). After the cells and supernatant were harvested at the indicated times, the mixture of scraped cells and supernatant was centrifuged at 3000 r.p.m. for 5 min. The ceramide level in the supernatant after centrifugation was measured. The pelleted cells were washed twice with phosphate-buffered saline (PBS), resuspended in 100 ␮l of 1% glutaraldehyde in PBS and incubated for 30 min at room temperature. The fixative was removed, and cells were washed once in PBS and were resuspended in 50 ␮l of PBS containing 2 ␮g/ml DAPI. A 10 ␮l aliquot was placed on a glass slide, and 500 cells per slide were scored for the incidence of apoptotic chromatin changes observed under a fluorescence microscope with a UV filter.

Metabolic labeling of lipids with [14C] serine in NB2a cells NB2a cells were seeded at 1.75⫻105 cells/ml in 4 ml of MEM supplemented with 10% FBS in 60-mm dishes and cultured at 37 °C for 24 h. Then either 1) the cells were resuspended in 2 ml of serum-free MEM with or without 40 ␮M RA in absolute ethanol and supplemented with 1 ␮Ci of L-[U-14C] serine. The incubation was continued for the indicated times, or 2) the cells were resuspended in 2 ml of serum-free MEM containing 1 ␮Ci of L-[U-14C] serine, and incubated for 24 h. The medium was removed and replaced with 2 ml of serum-free MEM with 40 ␮M RA, and cells were incubated for the indicated times. After the cells and supernatant were harvested at the indicated times, the mixture of scraped cells and supernatant were centrifuged at 100,000⫻g for 5 min. Lipids extracted from the pelleted cells and supernatant (1 ml/dish) by the Bligh-Dyer method were separated on TLC plates with methylacetate, propanol-1, chloroform, methanol, and 0.25% KCl (25:25:25:10:9 v/v). Radioactive lipids on the TLC plates were visualized, and their relative radioactivity was determined by using a BAS2000 Image Analyzer (Hanada et al., 1997; van Echten et al., 1990).

Glucosylceramide (GlcCer) synthase (GCS) enzyme assay Cells were harvested from RA-supplemented dishes and control dishes at the indicated times. GCS activity was assayed according to the method previously described (Riboni et al., 1995). C6-NBDCer, a synthetic fluorescent substrate (50 ␮g), and lecithin (500 ␮g) were mixed in 100 ␮l of ethanol, and the solvent was evaporated. Water (1 ml) was added and the mixture was sonicated to form liposomes. A standard reaction mixture (100 ␮l) composed of 20 mM Tris-HCl (pH 7.5), 500 mM UDP-Glc, 20 ␮l of liposomes, and 50 mg of cell protein was incubated at 37 °C for 2 h. After the incubation, lipids were extracted and applied to silica gel TLC plates. NBD lipids were separated in chloroform:metha-

Fig. 1. Ceramide content in CSF from the patients with AD and ALS, and from control individuals. The ceramide mass in 1 ml of centrifuged CSF samples was measured by the DAG kinase method as described in Experimental Procedures. Each value represents the mean of duplicate determinations. Statistical analysis was performed using ANOVA followed by Scheffe’s F test.

nol:12 mM MgCl2 in water (65:25:4 v/v) and visualized with UV illumination. NBD lipids were quantified by measurement of relative fluorescence intensities using FluorImager SI (Amersham) and software provided with the instrument by the manufacturer.

RESULTS Ceramide content in CSF from the patients with AD and ALS The patient groups for AD (n⫽16) and ALS (n⫽13), and other neurological disease patients as the control group (n⫽14) were age-matched (68.2⫾7.9, 59.4⫾10.4, 62.8⫾10.8 yo, respectively). The number of white blood cells (3⫾2, 4⫾4, 3⫾2/mm3, respectively) and the protein levels (37⫾10, 39⫾15, 31⫾7 mg/dl, respectively) were not different significantly among these three groups. As shown in Fig. 1, the mean content of ceramide in the CSF in the AD group (53.1⫾13.1 nM/ml) was significantly higher (P⬍0.01) compared with that in the control group (30.2⫾15.0 nM/ml). On the contrary, in the ALS group, the ceramide content (44.3⫾16.3 nM/ml) tended to be elevated, but the elevation was not significant as compared with the controls. In the AD group, the mass of ceramide in moderate dementia (CDR score, 2) was significantly

660

H. Satoi et al. / Neuroscience 130 (2005) 657– 666

Fig. 2. Immunohistochemistry for ceramide in brains from the patients with AD and control individuals. There are ceramide-immunoreactive glia in the layers 2 and 3 of the cerebral cortex in the brains from AD patients (B, C), but not in control individuals. The rectangle in (B) is enlarged in (C). Double labeling immunohistochemistry for ceramide and ␤-protein showed a regional coexistence of these proteins (D). Ceramide-immunoreactive glia were of astroglial lineage, but not of microglial one, as revealed by double labeling for ceramide and GFAP (E) and ceramide and HLA-DR (F). Scale bars⫽50 ␮m in (A, B), 10 ␮m in (C), 100 ␮m in (D) and 30 ␮m (E, F).

higher (68.1⫾6.5 nM/ml; n⫽4) than that in mild (CDR score, 0.5–1; 48.1⫾6.6 nM/ml; n⫽6) and severe dementia (CDR score, 3; 48.7⫾15.8 nM/ml; n⫽5; P⬍005, one pa-

Fig. 4. Intracellular content of ceramide in NB2a cells undergoing RA-induced apoptosis. Cells were plated at 1.75⫻105 cells/ml in 60 mm dishes. After 24 h, cells were resuspended in serum-free media containing 40 ␮M RA, further cultured and harvested at the indicated times (0, 12, 24, 36, and 48 h). Then the lipids were extracted by the Bligh and Dyer method, and ceramide contents were measured by the DAG kinase method as described in Experimental Procedures. (A) Intracellular ceramide was detected as ceramide-1-phosphate on TLC. The results shown are representative of more than three separate experiments. (B) Ceramide contents were calculated using a BAS2000 Image Analyzer (Fuji Co., Japan). Each value represents the mean⫾S.D. (bars) of duplicate determinations.

tient was excluded since the CDR score was not determined). Ceramide immunoreactive astroglia in AD brains

Fig. 3. Apoptosis induction by RA in NB2a cells. (A) The incidence of apoptosis was examined in cells cultured in the presence or absence of RA for various times. Cells were plated at 1.75⫻105 cells/ml in 60-mm dishes. After 24 h, cells were resuspended in serum-free medium containing various concentrations of RA (0, 20 or 40 ␮M) further cultured for the indicated times, harvested and stained by the DAPI method. At least 500 cells were counted under a fluorescent microscopy. Each value represents the mean⫾S.D. (bars) of duplicate determinations. Results are representative of more than three separate experiments. (B) Morphological changes of apoptotic cells were examined by the DAPI method at the indicated times (0, 24 and 48 h).

In the brains with AD, there were ceramide-immunoreactive glia in the layers 2 and 3 of the frontal cortices, which were regionally colocalized with senile plaques (Fig. 2B–D). Without the primary antibody against ceramide, no immunoreactive structures were detected (Fig. 2A). These astroglia were not detected in the brains either with ALS or control brains. In the spinal cord of ALS patients, no immunoreactivities for ceramide were observed in glia or neurons (photo not shown). Double labeling immunohistochemistry revealed that ceramide-immunoreactive glia were also positive for GFAP, but not for HLA-DR, thereby indicating that ceramide was aberrantly expressed in astroglia (Fig. 2E, F). RA-induced apoptosis in NB2a cells When NB2a cells were treated with RA in serum-free conditions, apoptosis was increased in a time- and dosedependent manner (Fig. 3A). Treatment with 40 ␮M RA for 48 h caused an increase of the percent apoptotic cells from

H. Satoi et al. / Neuroscience 130 (2005) 657– 666

661

Fig. 5. Chasing (A, B) and turnover (C, D) of intracellular ceramide by [14C] serine in NB2a cells with or without RA treatment. After the culture medium was replaced with 2 ml of serum-free MEM containing 1 ␮Ci of L-[U-14C] serine and cells were cultured with or without 40 ␮M RA for the indicated times, the lipids were extracted from the harvested cells, and separated by TLC as described in Experimental Procedures. [14C] serine-labeled ceramide on the TLC plates was visualized by autoradiography (A), and the amount was calculated with a BAS 2000 Image Analyzer (B). In turnover experiments, the incubation was continued at least for 24 h after the culture medium was replaced with 2 ml of serum-free MEM containing 1 ␮Ci of L-[U-14C] serine, and then the cells were resuspended in serum-free MEM with or without 40 ␮M RA for the indicated times. [14C] serine-labeled ceramide on the TLC plates was visualized by autoradiography (C), and the amount was calculated with a BAS 2000 Image Analyzer (D). Results of TLC are representative of more than three separate experiments. Each value represents the mean⫾S.D. (bars) of triplicate determinations.

3% to 47%. As shown in Fig. 3B, apoptosis was clearly confirmed by morphological changes such as condensation of chromatin and fragmentation of the nucleus as detected by staining with DAPI, but differentiated features such as dendrite-like formation were not detected. Thus, although it was reported previously that apoptosis in neuroblastoma cells could be induced by prolonged treatment with an inducer of terminal differentiation (Honma et al., 1996), we showed here that RA induced NB2a cell apoptosis in serum-free conditions, probably without inducing differentiation. Generation of intracellular ceramide in NB2a cells undergoing apoptosis In NB2a cells treated with 40 ␮M RA, the ceramide content measured by the DGK assay in NB2a cells was increased approximately three-fold as compared with the control level (11 nM/105 cells) 12 h after treatment, and the ceramide content continued to increase until 48 h of treatment (Fig. 4). To investigate the mechanism by which RA increased the intracellular ceramide content, we next studied the synthesis and metabolism of ceramide in NB2a cells undergoing apoptosis. To examine whether ceramide was generated through the de novo serine-palmitoyl CoA transferase pathway, we labeled sphingolipids with [14C] serine in cells treated with RA to induce apoptosis (Fig. 5A, B). Previously, RA-induced differentiation in NB2a cells was reported to increase constantly due to de novo synthesis of ceramide (Riboni et al., 1995). However, our chasing experiment showed that the treatment with 40 ␮M RA in-

creased the amount of [14C] serine label in ceramide from 79 arbitrary units (A.U.) to 103 A.U. after 24 h, but did not at 48 h (Fig. 5A, B). In contrast to ceramide synthesis, measurement of the turnover in a chase experiment using [14C] serine-labeled ceramide showed a significant decrease of the [14C] serine-labeled ceramide pool (from 100 A.U. to 42 A.U.) in the control, and a significant inhibition of metabolism (from 100 A.U. to 122 A.U.) by treatment with 40 ␮M RA (Fig. 5C, D). The results clearly indicated that, under apoptosis-inducing conditions, the ceramide content increased not only by enhancing de novo ceramide synthesis but also by inhibiting the ceramide-metabolizing pathway. Effects of RA on ceramide metabolizing enzymes, GCS and SM synthase (SMS) Next, we examined whether there were the changes of [14C] serine-labeled sphingolipids such as SM and GlcCer, which are produced by metabolism of ceramide by SMS and GCS, respectively, in RA-induced NB2a apoptosis. As shown in Fig. 6A and B, the amount of [14C] serine-labeled GlcCer significantly decreased (from 37 A.U. to 15 A.U.), whereas the amount of SM increased (from 27 A.U. to 39 A.U.) after 24 h of treatment with 40 ␮M RA, suggesting the inhibition of GCS or the enhancement of GlcCermetabolizing enzymes. Indeed, 40 ␮M RA significantly decreased the activity of GCS (from 100 A.U. to 81 A.U. and 45 A.U., respectively, after 24 and 48 h of treatment), but not the activity of SMS (Fig. 6C), indicating that the inhibition of GCS activity was closely involved in keeping

662

H. Satoi et al. / Neuroscience 130 (2005) 657– 666

the level of ceramide released outside of the cells, and that this extracellular release was specific for ceramide because no extracellular increase of SM was detected under the same experimental conditions. These notions were confirmed by the data shown in Fig. 8 that the extracellular ceramide content increased approximately 2.5-fold by 40 ␮M RA as compared with the control, while the content of DAG, which was judged as a glycerolipid counterpart of ceramide based on structural similarity, was not increased significantly.

DISCUSSION

Fig. 6. Metabolism of [14C] serine-labeled GlcCer and SM, and changes of GCS and SMS activity in NB2a cells undergoing RAinduced apoptosis. After the culture medium was replaced with 2 ml of serum-free MEM containing 1 ␮Ci of L-[U-14C] serine, the incubation was continued at least for 24 h and then the cells were resuspended in serum-free MEM with or without 40 ␮M RA for the indicated times. [14C] serine-labeled ceramide on the TLC plates was visualized by autoradiography (A), and the amount was calculated with a BAS 2000 Image Analyzer (B). Activity of GCS and SMS was detected as described in Experimental Procedures (C). Results of TLC are representative of more than three separate experiments. Results in (C) are representative of three separate experiments assayed in duplicate.

the levels of ceramide higher in cells undergoing apoptosis than in the controls. Increase of extracellular ceramide during apoptosis To examine the possibility that ceramide was released to act intercellularly as a pro-apoptotic signal, we examined the changes of extracellular ceramide levels in apoptosis of NB2a cells. When we induced apoptosis by adding 40 ␮M RA to NB2a cells while labeling with [14C] serine, the labeling experiment showed that [14C] serine-labeled ceramide was increased from 1 A.U. to 24 A.U. in the medium after 48 h of treatment, but not [14C] serinelabeled SM (Fig. 7A, B). Similarly, it was found in the turnover experiments that RA significantly enhanced the increase of [14C] serine-labeled ceramide in the medium compared with the control level, but did not affect the [14C] serine-labeled SM level (Fig. 7C, D). These results suggest that apoptosis increased not only the intracellular ceramide content but also increased

Recently, elevation of ceramide levels has been reported in the total brain tissues from very early AD (Han et al., 2002). In addition, a recent study has shown that oxidative stress induced by ␤ protein upregulates ceramide and cholesterol levels in cell cultures of hippocampal neurons (Cutler et al., 2004). In the present study, we revealed an increase of ceramide synthesis and mass in the CSF by [14C]serine-labeling and DGK assay methods, respectively, and in astroglia, but not neurons, in AD brains by histochemical method using anti-ceramide antibody. As the mechanism by which the ceramide levels were increased outside as well as inside of NB2a cells during apoptosis, the involvement of ceramide generating enzymes such as ceramide synthase and sphingomyelinase was suggested during differentiation and apoptosis (Riboni et al., 1995; Wiesner and Dawson, 1996). We here demonstrated ceramide increase due to inhibition of the ceramide-metabolizing enzyme GlcCer synthase. An increase of ceramide levels by downregulation of GlcCer synthase has also been demonstrated in cerebral ischemia in vivo (Takahashi et al., 2004; Ohtani et al., 2004). In neurodegenerative disorders, the systems other than the ROS-related mechanism, which involve intercellular immunologic or paracrine-type effects of proapoptotic molecules, may play roles in neuronal apoptosis. For example, increase of the CSF levels of tau (Arai et al., 1995; Motter et al., 1995), ␤-amyloid (1– 42) (Jensen et al., 1999; Kanai et al., 1998) and AD7C (Monte et al., 1997) in AD, and 4-hydroxynonenal and 3-nitrotyrosine (Smith et al., 1998; Tohgi et al., 1999) in ALS have been reported to be biomarkers for disease progression, and suggested to be intercellular apoptotic signals. The present study also showed that ceramide may function in vitro and in vivo not only as an intracellular but also as an intercellular mediator of apoptosis. Previously, de novo ceramide generation through activation of neutral sphingomyelinase or ceramide synthase has been revealed in RA-induced NB2a cell differentiation (Riboni et al., 1995). A 230% increase in the intracellular ceramide level was induced also in GH4C1 cells by RA through de novo synthesis by ceramide synthase, but without any change in the mass of SM, sphingosine or phosphorylcholine (Kalen et al., 1992). An increase of ceramide levels through de novo synthesis was commonly observed in oligodendroglial apoptosis after stimulation with ␤amyloid (Lee et al., 2004). In line with these data, we here

H. Satoi et al. / Neuroscience 130 (2005) 657– 666

663

Fig. 7. Chasing (A, B) and turnover (C, D) of extracellular ceramide by [14C] serine with or without RA treatment. After the cell culture medium was replaced with 2 ml of serum-free MEM containing 1 ␮Ci of L-[U-14C] serine and cells were cultured with or without 40 ␮M RA for the indicated times, labeled lipids were extracted from the harvested cells, and separated by TLC as described in Experimental Procedures. [14C] serine-labeled ceramide on the TLC plates was visualized by autoradiography (A), and the amount was calculated with a BAS 2000 Image Analyzer (B). In turnover experiments, the incubation was continued for at least for 24 h after the cell culture medium was replaced with 2 ml of serum-free MEM containing 1 ␮Ci of L-[U-14C] serine, and then the cells were resuspended in serum-free MEM with or without 40 ␮M RA for the indicated times. [14C] serine-labeled ceramide on the TLC plates was visualized by autoradiography (C), and the amount was calculated with a BAS 2000 Image Analyzer (D). Results of TLC are representative of more than three separate experiments. Each value represents the mean⫾S.D. (bars) of triplicate determinations.

showed that when NB2 cells underwent apoptosis, the ceramide level increased approximately 300% as compared with the untreated control (Figs. 3 and 4). However, interestingly, de novo ceramide synthesis only increased slightly, and measurement of the turnover of [14C] serinelabeled sphingolipids did not show any decrease of SM after treatment with RA (Fig. 5A, B). Therefore, these results indicated only a minor involvement of de novo synthesis in RA-induced ceramide generation, and no involvement of SM hydrolysis by sphingomyelinase. In general, the pathway mediated by ceramidase is known to metabolize only a small amount of ceramide, and indeed, degradation of [14C] serine-labeled ceramide to sphingosine was not detected in the present study (data not shown). We herein thought that the ceramide mass might increase rather through the inhibition of ceramidemetabolizing enzymes such as GCS and SMS, because RA was shown to increase degradation of [14C]serinelabeled ceramide more than [14C]ceramide synthesis during apoptosis (Figs. 4 and 5). As shown in Fig. 6, the conversion of ceramide to GlcCer was inhibited by RA, although the conversion to SM was not affected. Moreover, the activity of GCS was, indeed, significantly inhibited by RA, whereas that of SMS was not changed. It remains unknown which aspect of ceramide generation, either SM hydrolysis, de novo ceramide synthesis, or ceramide metabolism, is most importantly involved in RA-induced apo-

ptosis. However, we clearly demonstrated that ceramide metabolism through GCS was, at least in part, involved in regulating the ceramide levels in RA-induced NB2a cell apoptosis. Changes of extracellular ceramide levels during apoptosis induction have not been investigated previously in vitro or in vivo. We here showed an increase of ceramide levels in the CSF of AD patients, and as well, secreted [14C]serine-labeled ceramide and the mass of extracellular ceramide in apoptosis. Both chasing and turnover experiments using [14C]serine-labeled sphingolipids showed an increase of ceramide in the supernatant of NB2 cells after 48 h of treatment with RA, but no increase of extracellular SM (Fig. 7). The mass of extracellular ceramide was significantly increased by treatment with RA along with an increase of intracellular ceramide, whereas the mass of DAG showed only a slight increase, suggesting a specific increase of extracellular ceramide during neuronal apoptosis (Fig. 8). Recently, sphingosine-1-phosphate (S-1-P) was reported to function not only as an intracellular but also as an extracellular/intercellular mediator of cell growth through EDG family receptors (Spiegel and Milstien, 2000). Therefore, it may be conceivable that ceramide works outside the cells after being converted to sphingosine or S-1-P. However, when we examined the release of sphingolipids after labeling the cells with [14C]serine, other sphingolipids except for ceramide, SM and a small amount of glyco-

664

H. Satoi et al. / Neuroscience 130 (2005) 657– 666

Fig. 8. Extracellular content of ceramide and DAG released into the medium in RA-induced NB2a cell apoptosis. Cells were plated at 1.75⫻105 cells/ml in 60 mm dishes. After 24 h, cells were resuspended in serum-free medium containing 40 ␮M RA, further cultured, and harvested at the indicated times (0, 12, 24, 36, and 48 h). Then lipids in the supernatant were extracted by the Bligh and Dyer method, and the masses of ceramide and DAG were measured by the DAG kinase method as described in Experimental Procedures. Extracellular ceramide and DAG were detected as ceramide-1-phosphate and phosphatidic acid, respectively, on TLC plates (A). Results of TLC are representative of more than three separate experiments. Ceramide and DAG masses were calculated with a BAS 2000 Image Analyzer (B). Each value represents the mean⫾S.D. (bars) of triplicate determinations.

sphingolipids were hardly detected in the supernatant. Since sphingosine and S-1-P usually exert their biological effects at concentrations of 1–10 ␮M, which are similar to bioactive ceramide concentrations, the lack of [14C]labeled sphingosine or S-1-P meant that there was no sufficient conversion of ceramide to produce bioactive amounts of other sphingolipids in the supernatant. The mechanism for the extracellular release of sphingolipid remains unknown. Recently, it was reported that multidrug resistance-related protein-1 transported C6NBD-GlcCer and C6-NBD-SM from the Golgi apparatus to the basolateral membrane in LLC-PK1 cells (Raggers et al., 1999). Chemokine connective tissue-activating peptide III was also shown to transfer SM between cells (Stoeckelhuber et al., 2000). These data may indicate possible mechanisms for the release of sphingolipids, although further investigations are required. As far as we know, the present study is the first to show an increase of extracellular ceramide during neuronal apoptosis, suggesting a possible role of ceramide as an extracellular/intercellular mediator. This notion is not contradictory to the fact that exogenous long-chain ceramide induced apoptosis at a concentration of 10 –100 nM (Hung et al., 1999), which

was in a similar range to the extracellular ceramide level in the present study (approximately 10 nM/105 cells). Apoptosis is believed to represent a beneficial mechanism essential to normal development of the vertebrate nervous system. On the contrary, inappropriate initiation of apoptosis has been proposed to underlie the neurological diseases. Indeed, a 17-kDa glycoprotein named gliotoxin, which induces glial cell apoptosis, and soluble Fas were found to increase in the CSF from multiple sclerosis patients (Menard et al., 1998). Neuronal apoptosis has been indicated in various neurodegenerative diseases including AD, ALS and Parkinson’s disease, all of which are characterized by the gradual loss of specific populations of neurons (Ariga et al., 1998). In early AD patients, intrathecal ganglioside GM1 increases, probably, due to the release of lipids constituting the plasma membrane, a process called “cell surface shedding” (Yanagisawa et al., 1995). Therefore, it is likely that ceramide is released into the extracellular space mainly from astroglia, and may be involved in the neuronal apoptosis, since we verified here that ceramide was aberrantly expressed in AD brain tissues and increased in CSF (Fig. 2). In support for this, astroglia show phenotypic transformation after pretreatment with ␤-amyloid or peroxynitrite, a toxic metabolite of nitric oxide. These astroglia may upregulate intracellular ceramide, release it extracellularly and induce neuronal apoptosis in culture (Cassina et al., 2002; Malchiodi-Albedi et al., 2001). It remained unclear in the present work why only insignificant increase of ceramide was detected in the CSF or the spinal cord of patients with ALS, since Cutler et al. (2002) reported an increased level of ceramide in the spinal cord tissue of ALS patients. The reason for this might be related to the difference of samples used for the analysis, or the sensitivity of the method for detection. Finally, we made two novel findings here: one is the possibility that the intracellular level of ceramide increased through inhibition of GlcCer synthase, and the other is that ceramide was aberrantly expressed in vivo in astroglia in AD brains as well as in vitro in the supernatant of neuronal cells in response to apoptotic stress. It is worth mentioning that the increase of extracellular ceramide does not seem to be merely the result of apoptotic collapse of the cells, but may have a more positive implication on the function of ceramide as an extracellular/intercellular mediator for neuronal apoptosis. Acknowledgments—This work was supported by Grants-in-Aid for Peripheral Neuropathy from the Japanese Ministry of Health, Labor and Welfare, and Grants-in-Aid for Scientific Research from the Japanese Ministry of Education, Culture, Sports, Science and Technology.

REFERENCES Arai H, Terajima M, Miura M, Higuchi S, Muramatsu T, Machida N, Seiki H, Takase S, Clark CM, Lee VM, Trojanowski JQ, Sasaki H (1995) Tau in cerebrospinal fluid: a potential diagnostic marker in Alzheimer’s disease. Ann Neurol 38:649 – 652. Ariga T, Jarvis WD, Yu RK (1998) Role of sphingolipid-mediated cell death in neurodegenerative disease. J Lipid Res 39:1–16.

H. Satoi et al. / Neuroscience 130 (2005) 657– 666 Beckman JS, Estevez AG, Crow JP, Barbeito L (2001) Superoxide dismutase and the death of motoneurons in ALS. Trends Neurosci 24(suppl):S15–20. Bligh EA, Dyer WJ (1959) A rapid method of total lipid extraction and purification. Can J Biochem Physiol 37:911–917. Brooks BR, Miller RG, Swash M, Munsat TL (2000) El Escorial revisited: revised criteria for the diagnosis of amyotrophic lateral sclerosis. Amyotroph Lateral Scler Other Motor Neuron Disord 1:293–299. Brugg B, Michel PP, Agid Y, Ruberg M (1996) Ceramide induces apoptosis in cultured mesencephalic neurons. J Neurochem 66:733–739. Cassina P, Peluffo H, Pehar M, Martinez-Palma L, Ressia A, Beckman JS, Estevez AG, Barbeito L (2002) Peroxynitrite triggers a phenotypic transformation in spinal cord astrocytes that induces motor neuron apoptosis. J Neurosci Res 67:21–29. Cutler RG, Pedersoen WA, Camandola S, Rothstein JD, Mattson MP (2002) Evidence that accumulation of ceramides and cholesterol esters mediates oxidative stress-induced death of motor neurons in amyotrophic lateral sclerosis. Ann Neurol 52:448 – 457. Cutler RG, Kelly J, Storie K, Pedersoen WA, Tammara A, Hatanpaa K, Troncoso JC, Mattson MP (2004) Involvement of oxidative stressinduced abnormalities in ceramide and cholesterol metabolism in brain aging and Alzheimer’s disease. Proc Natl Acad Sci USA 101:2070 –2075. Chen Y, Ginis I, Hallenbeck JM (2001) The protective effect of ceramide in immature rat brain hypoxia-ischemia involves upregulation of Bcl-2 and reduction of TUNEL-positive cells. J Cereb Blood Flow Metab 21:34 – 40. Ferrante RJ, Browne SE, Shinobu LA, Bowling AC, Baik MJ, MacGarvey U, Kowall NW, Brown RH Jr, Beal MF (1997) Evidence for oxidative damage in both sporadic and familial amyotrophic lateral sclerosis. J Neurochem 69:2064 –2074. Furuya S, Mitoma J, Makino A, Hirabayashi Y (1998) Ceramide and its interconvertible metabolite sphingosine function as indispensable lipid factors involved in survival and dendritic differentiation of cerebellar Purkinje cells. J Neurochem 71:366 –377. Goswami R, Dawson G (2000) Does ceramide play a role in neural cell apoptosis? J Neurosci Res 60:141–149. Gottschalk AR, McShan C, Kilkus J, Dawson G, Quintans J (1995) Physiological cell death in B-lymphocytes: II. Resistance to antiIgM-induced apoptosis in a WEHI-231 subline is due to deficient generation of the killer lipid ceramide. Eur J Immunol 25: 1032–1038. Han X, Holtzman DM, McKeel DW Jr, Kelley J, Morris JC (2002) Substantial sulfatide deficiency and ceramide elevation in very early Alzheimer’s disease: potential role in disease pathogenesis. J Neurochem 82:809 – 818. Hanada K, Hara T, Nishijima M, Kuge O, Dickson RC, Nagiec MM (1997) A mammalian homologue of the yeast LCB1 encodes a component of serine palmitoyltransferase, the enzyme catalyzing the first step in sphingolipid synthesis. J Biol Chem 51: 32108 –32114. Hannun YA, Obeid LM (1995) Ceramide: an intracellular signal for apoptosis. Trends Biochem Sci 20:73–77. Hannun YA (1996) Functions of ceramide in coordinating cellular responses to stress. Science 274:1855–1859. Hannun YA, Luberto Ch (2000) Ceramide in the eukaryotic stress response. Trends Cell Biol 10:73– 80. Honma N, Uchida A, Hirose H, Srsen V, Kishimoto T, Hisanaga S (1996) Two types of apoptotic cell death of rat central nervous system-derived neuroblastoma B50 and B104 cells: apoptosis induced during proliferation and after differentiation. J Neurochem 67:1856 –1865. Hung WC, Chang HC, Chuang LY (1999) Activation of caspase-3-like proteases in apoptosis induced by sphingosine and other longchain bases in Hep3B hepatoma cells. Biochem J 338:161–166.

665

Irie F, Hirabayashi Y (1998) Application of exogenous ceramide to cultured rat spinal motoneurons promotes survival or death by regulation of apoptosis depending on its concentrations. J Neurosci Res 54:475– 485. Jayadev S, Liu B, Bielawska AE, Lee JY, Nazaire F, Pushkareva MY, Obeid LM, Hannun YA (1995) Role for ceramide in cell cycle arrest. J Biol Chem 270:2047–2052. Jensen M, Schroder J, Blomberg M, Engvall B, Pantel J, Ida N, Basun H, Wahlund LO, Werle E, Jauss M, Beyreuther K, Lannfelt L, Hartmann T (1999) Cerebrospinal fluid A beta 42 is increased early in sporadic Alzheimer’s disease and declines with disease progression. Ann Neurol 45:504 –511. Kalen A, Borchardt RA, Bell RM (1992) Elevated ceramide levels in GH4C1 cells treated with retinoic acid. Biochim Biophys Acta 1125:90 –96. Kanai M, Matsubara E, Isoe K, Urakami K, Nakashima K, Arai H, Sasaki H, Abe K, Iwatsubo T, Kosaka T, Watanabe M, Tomidokoro Y, Shizuka M, Mizushima K, Nakamura T, Igeta Y, Ikeda Y, Amari M, Kawarabayashi T, Ishiguro K, Harigaya Y, Wakabayashi K, Okamoto K, Hirai S, Shoji M (1998) Longitudinal study of cerebrospinal fluid levels of tau, A beta1-40, and A beta1-42(43) in Alzheimer’s disease: a study in Japan. Ann Neurol 44:17–26. Kawase M, Watanabe M, Kondo T, Yabu T, Taguchi Y, Umehara H, Uchiyama T, Mizuno K, Okazaki T (2002) Increase of ceramide in adriamycin-induced HL-60 cell apoptosis: detection by a novel anti-ceramide antibody. Biochim Biophys Acta 1584:104 –114. Kolesnick RN, Kronke M (1998) Regulation of ceramide production and apoptosis. Annu Rev Physiol 60:643– 665. Lee J-T, Xu J, Lee J-M, Ku G, Han X, Yang D-I, Chen S, Hsu CY (2004) Amyloid-␤ peptide induces oligodendrocyte death by activating the neutral sphingomyelinase-ceramide pathway. J Cell Biol 164:123–131. Loo DT, Copani A, Pike CJ, Whittemore ER, Walencewicz AJ, Cotman CW (1993) Apoptosis is induced by beta-amyloid in cultured central nervous system neurons. Proc Natl Acad Sci USA 99: 7951–7955. Mangoura D, Dawson G (1998) Programmed cell death in cortical chick embryo astrocytes is associated with activation of protein kinase PK60 and ceramide formation. J Neurochem 70:130 –138. Malchiodi-Albedi F, Domenici MR, Paradisi S, Bernardo A, AjmoneCat MA, Minghetti L (2001) Astrocytes contribute to neuronal impairment in beta A toxicity increasing apoptosis in rat hippocampal neurons. Glia 34:68 –72. Menard A, Amouri R, Dobransky T, Charriaut-Marlangue C, Pierig R, Cifuentes-Diaz C, Ghandour S, Belliveau J, Gascan H, Hentati F, Lyon-Caen O, Perron H, Rieger F (1998) A gliotoxic factor and multiple sclerosis. J Neurol Sci 154:209 –221. Mirra SS, Heyman A, McKeel D, Sumi SM, Crain BJ, Brownlee LM, Vogel FS, Hughes JP, van Belle G, Berg L (1991) The Consortium to Establish a Registry for Alzheimer’s Disease (CERAD): Part II. Standardization of the neuropathologic assessment of Alzheimer’s disease. Neurology 41:479 – 486. Monte SM, Ghanbari K, Frey WH, Beheshti I, Averback P, Hauser SL, Ghanbari HA, Wands JR (1997) Characterization of the AD7CNTP cDNA expression in Alzheimer’s disease and measurement of a 41-kD protein in cerebrospinal fluid. J Clin Invest 100: 3093–3104. Motter R, Vigo-Pelfrey C, Kholodenko D, Barbour R, Johnson-Wood K, Galasko D, Chang L, Miller B, Clark C, Green R, et al (1995) Reduction of beta-amyloid peptide 42 in the cerebrospinal fluid of patients with Alzheimer’s disease. Ann Neurol 38:643– 648. Obeid LM, Linardic CM, Karolak LA, Hannun YA (1993) Programmed cell death induced by ceramide. Science 259:1769 –1771. Ohtani R, Tomimoto H, Kondo T, Wakita H, Akiguchi I, Shibasaki H, Okazaki T (2004) Upregulation of ceramide and its regulating mechanism in a rat model of chronic cerebral ischemia. Brain Research 1023:31– 40.

666

H. Satoi et al. / Neuroscience 130 (2005) 657– 666

Okazaki T, Bielawska A, Bell RM, Hannun YA (1990) Role of ceramide as a lipid mediator of 1 alfa, 25-dihydroxyvitamin D3-induced HL-60 cell differentiation. J Biol Chem 265:15823–15831. Okazaki T, Kondo T, Kitano T, Tashima M (1998) Diversity and complexity of ceramide signalling in apoptosis. Cell Sig 10:685– 692. Oppenheim RW (1991) Cell death during development of the nervous system. Annu Rev Neurosci 14:453–501. Raggers RJ, van Helvoort A, Evers R, van Meer G (1999) The human multidrug resistance protein MRP1 translocates sphingolipid analogs across the plasma membrane. J Cell Sci 112:415– 422. Riboni L, Prinetti A, Bassi R, Caminti A, Tettamanti G (1995) A mediator role of ceramide in the regulation of neuroblastoma neuro2a cell differentiation. J Biol Chem 270:26868 –26875. Sastry PS, Rao KS (2000) Apoptosis and the nervous system. J Neurochem 74:1–20. Smith RG, Henry YK, Mattson MP, Appel SH (1998) Presence of 4-hydroxynonenal in cerebrospinal fluid of patients with sporadic amyotrophic lateral sclerosis. Ann Neurol 44:696 – 699. Spiegel S, Foster D, Kolesnick R (1996) Signal transduction through lipid second messengers. Curr Opin Cell Biol 8:159 –167. Spiegel S, Milstien S (2000) Sphingosine-1-phosphate: signaling inside and out. FEBS Lett 476:55–57. Stoeckelhuber M, Dobner P, Baumgartner P, Ehlert J, Brandt E, Mentele R, Adam D, Engelmann B (2000) Stimulation of cellular sphingomyelin import by the chemokine connective tissue-activating peptide III. J Biol Chem 275:37365–37372. Takahashi K, Ginis I, Nishioka R, Klimanis D, Barone FC, White RF, Chen Y, Hallenbeck JM (2004) Glucosylceramide synthase activity and ceramide levels are modulated during cerebral ischemia after ischemic preconditioning. J Cereb Blood Flow Metab 24:623– 627. Takeda Y, Tashima M, Takahashi A, Uchiyama T, Okazaki T (1999) Ceramide generation in nitric oxide-induced apoptosis. J Biol Chem 274:10654 –10660.

Tettamanti G, Riboni L (1993) Gangliosides and modulation of the function of neural cells. Adv Lipid Res 25:235–267. Thompson CB (1995) Apoptosis in the pathogenesis and treatment of disease. Science 267:1456 –1462. Tohgi H, Abe T, Yamazaki K, Murata T, Ishizaki E, Isobe C (1999) Remarkable increase in cerebrospinal fluid 3-nitrotyrosine in patients with sporadic amyotrophic lateral sclerosis. Ann Neurol 46:129 –131. van Echten G, Birk R, Brenner-Weiss G, Schmidt RR, Sandhoff K (1990) Modulation of sphingolipid biosynthesis in primary cultured neurons by long chain bases. J Biol Chem 265:9333–9339. Venable ME, Lee JY, Smyth MJ, Bielawska A, Obeid LM (1995) Role of ceramide in cellular senescence. J Biol Chem 270:30701– 30708. Verheij M, Bose R, Lin XH, Yao B, Jarvis WD, Grant S, Birrer MJ, Szabo E, Zon LI, Kyriakis JM, Haimovitz-Friedman A, Fuks Z, Kolesnick RN (1996) Requirement for ceramide-initiated SAPK/JNK signalling in stress-induced apoptosis. Nature 380:75–79. Waggie KS, Kahle PJ, Tolwani RJ (1999) Neurons and mechanisms of neuronal death in neurodegenerative diseases: a brief review. Lab Anim Sci 49:358 –362. Wiesner DA, Dawson G (1996) Staurosporine induces programmed cell death in embryonic neurons and activation of the ceramide pathway. J Neurochem 66:1418 –1425. Yanagisawa K, Okada A, Suzuki N, Ihara Y (1995) GM1 gangliosidebound amyloid beta-protein (A beta): a possible form of preamyloid in Alzheimer’s disease. Nat Med 1:1062–1066. Yu ZF, Nikolova-Karakashian M, Zhou D, Cheng G, Schuchman EH, Mattson MP (2000) Pivotal role for acidic sphingomyelinase in cerebral ischemia-induced ceramide and cytokine production, and neuronal apoptosis. J Mol Neurosci 15:85–97.

(Accepted 15 August 2004) (Available online 23 November 2004)