Neuroprotective effects of yokukansan, a traditional japanese medicine, on glutamate-mediated excitotoxicity in cultured cells

Neuroscience 159 (2009) 1397–1407

NEUROPROTECTIVE EFFECTS OF YOKUKANSAN, A TRADITIONAL JAPANESE MEDICINE, ON GLUTAMATE-MEDIATED EXCITOTOXICITY IN CULTURED CELLS Z. KAWAKAMI,* H. KANNO, T. UEKI, K. TERAWAKI, M. TABUCHI, Y. IKARASHI AND Y. KASE

fare of Japan as a remedy for neurosis, insomnia, and irritability in children. Recently, TJ-54 was reported to improve behavioral and psychological symptoms of dementia (BPSD) such as hallucinations, agitation, and aggressiveness in patients with Alzheimer’s disease, dementia with Lewy bodies, and other forms of senile dementia (Iwasaki et al., 2005a,b). However, the mechanism underlying the effectiveness of the medicine is still unclear. Cognitive dysfunction and BPSD are thought to be associated with neurofunctional and neuropathological abnormalities in the brain. In several animal models used for studies in the pathogenesis of and therapy for dementia, an increase in the extracellular levels of the excitatory amino acids such as glutamate in the brain has been demonstrated (Todd and Butterworth, 2001; Han et al., 2008; Harkany et al., 2000; Behrens et al., 2002). Glutamate is well-known to contribute not only to induction of excitation of post-synaptic neurons but also to excitotoxic neuronal death due to the intensity and duration of glutamate exposure (Choi, 1988; Cheung et al., 1998). Under physiological conditions, neurotransmission to postsynaptic receptors is terminated by its clearance from the synaptic cleft by transporter proteins located in neuronal and astroglial cells to prevent an overload of glutamate to neurons. In particular, astrocytes take an important role in the efficient removal of glutamate from the extracellular space via two glutamate transporters, glutamate aspartate transporter (GLAST) and glutamate transporter 1 (GLT-1) (Kanai et al., 1993; Schlag et al., 1998). These findings suggest that a failure of the glutamatergic system is related to the appearance of cognitive dysfunction and BPSD. Previous studies have demonstrated that BPSD-like behaviors such as anxiety, depression, muricide, attacking, and startle responses as well as impairment of learning and memory are observed in thiamine-deficient (TD) rats and mice (Nakagawasai et al., 2000; Murata et al., 2004). Collins (1967) and Robertson et al. (1968) have demonstrated that astrocytes are among the first cells to be affected by thiamine deficiency in advance of neuronal cell death. Symptomatic TD is associated with increased extracellular glutamate concentrations in focal regions of the brain (Hazell et al., 1993; Langlais and Zhang, 1993) and TD in cultured astrocytes led to a decrease in the uptake of glutamate (Hazell et al., 2003). Therefore, dysfunction of the astrocyte glutamate transporter may be a major contributing factor to the increase in extracellular glutamate concentration and resulting excitotoxicity. In addition, excessive extracellular glutamate levels are believed to evoke neuronal dysfunction and in turn, neuronal

Tsumura Research Laboratories, Tsumura & Co., 3586 Yoshiwara, Ami-machi, Inashiki-gun, Ibaraki 300-1192, Japan

Abstract—To clarify the mechanism of yokukansan (TJ-54), a traditional Japanese medicine, against glutamate-mediated excitotoxicity, the effects of TJ-54 on glutamate uptake function were first examined using cultured rat cortical astrocytes. Under thiamine-deficient conditions, the uptake of glutamate into astrocytes, and the levels of proteins and mRNA expressions of glutamate aspartate transporter of astrocytes significantly decreased. These decreases were ameliorated in a dose-dependent manner by treatment with TJ-54 (100 – 700 ␮g/ml). The improvement of glutamate uptake with TJ-54 was completely blocked by the glutamate transporter inhibitor DL-threo-␤-hydroxyaspartic acid. Effects of TJ-54 on glutamate-induced neuronal death were next examined by using cultured PC12 cells as a model for neurons. Addition of 17.5 mM glutamate to the culture medium induced an approximately 50% cell death, as evaluated by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay. TJ-54 (1–1000 ␮g/ml) inhibited the cell death in a dose-dependent manner. Furthermore, competitive binding assays to glutamate receptors showed that TJ-54 bound potently to N-methyl-D-aspartate receptors, in particular, to its glutamate and glycine recognition sites. These results suggest that TJ-54 may exert a neuroprotective effect against glutamate-induced excitotoxicity not only by amelioration of dysfunction of astrocytes but also by direct protection of neuronal cells. © 2009 IBRO. Published by Elsevier Ltd. All rights reserved. Key words: astrocytes, GLAST, GLT-1, mRNAs, NMDA, PC12.

Yokukansan (TJ-54) is one of the traditional Japanese medicines called “Kampo” medicine in Japan. It is composed of seven kinds of dried medical herbs. TJ-54 has been approved by the Ministry of Health, Labor and Wel*Corresponding author. Tel: ⫹81-29-889-3850; fax: ⫹81-29-889-2158. E-mail address: [email protected] (Z. Kawakami). Abbreviations: ANOVA, analysis of variance; BPSD, behavioral and psychological symptoms of dementia; cpm, counts per minute; DHK, dihydrokainate; DMEM, Dulbecco’s modified Eagle’s medium; ER, endoplasmic reticulum; GFAP, glial fibrillary acidic protein; GLAST, glutamate aspartate transporter; GLT-1, glutamate transporter 1; GSH, glutathione; IC50, half-maximal inhibitory concentration; mGluR, metabotropic glutamate receptor; MTT, 3-(4,5-dimethylthiazol-2yl)-2,5-diphenyltetrazolium bromide; NMDA, N-methyl-D-aspartate; PBS(–), Ca-and Mg-free phosphate-buffered saline; PCP, phencyclidine; PKC, protein kinase C; RT, reverse transcription; RT-PCR, reverse-transcription polymerase chain reaction; SDS, sodium dodecyl sulfate; TBHA, DL-threo-␤-hydroxy-aspartic acid; TD, thiamine deficiency; TJ-54, yokukansan.

0306-4522/09 $ - see front matter © 2009 IBRO. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.neuroscience.2009.02.004

1397

1398

Z. Kawakami et al. / Neuroscience 159 (2009) 1397–1407

death, by excessive activation of N-methyl-D-aspartate (NMDA) receptors that leads to an intracellular cascade of cytotoxic events (Weikert et al., 1997). Recently, we also confirmed that an increase in the extracellular level of glutamate and degeneration of neuronal and astroglial cells were observed in the brain of TD rats, and found that TJ-54 inhibited not only the TD-induced increase in extracellular level of glutamate but also the degeneration of cerebral neurocytes and astrocytes in the vulnerable brain regions (Kanno et al., 2008). Recently, Uncaria rhynchophylla, a constituent herb of TJ-54, was shown to protect against NMDA-induced neuronal excitotoxicity in rat hippocampus (Lee et al., 2003) or to block NMDA-induced current in cortical neurons (Sun et al., 2003). These results suggest that one of the mechanisms underlying the effects of TJ-54 may be closely related to improvement of glutamate excitatory neurotoxicity, including glutamate transporters and NMDA receptors as causative factors. To clarify this hypothesis, we investigated the effects of TJ-54 on glutamate-mediated excitotoxicity by examining the effects on TD-induced decreases in glutamate uptake and glutamate transporters in cultured astrocytes, glutamate-induced PC12 cell death, and competitive binding to glutamate receptors in the rat brain membrane.

EXPERIMENTAL PROCEDURES Drugs and reagents TJ-54 used in the present study was a dry powdered extract from a mixture of Atractylodes lancea rhizome (4.0 g, rhizome of Atractylodes lancea De Candolle), Hoelen (4.0 g, sclerotium of Poria cocos Wolf), Cnidii rizoma (3.0 g, rhizome of Cnidium officinale Makino), Uncaria thorn (3.0 g, thorn of Uncaria rhynchophylla Miquel), Japanese angelica root (3.0 g, root of Angelica acutiloba Kitagawa), Bupleurum root (2.0 g, root of Bupleurum falcatum Linné), and Glycyrrhiza root (1.5 g, root and stolon of Glycyrrhiza uralensis Fisher) that was supplied by Tsumura & Co. (Tokyo, Japan). Reagents used in cell culture experiments, DL-threo-␤-hydroxy-aspartic acid (TBHA), dehydrokainate (DHK), pyrithiamine hydrobromide, Dulbecco’s modified Eagle’s medium (DMEM), DNase, glutamate dehydrogenase, ␤-nicotinamide adenine dinucleotide, 1-methoxyphenazine methosulfate, and Triton X-100, were purchased from Sigma-Aldrich (St. Louis, MO, USA). EDTA, Hepes, and 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) were purchased from Dojindo (Kumamoto, Japan). Purified rabbit polyclonal antiserum to glial fibrillary acidic protein (GFAP) was purchased from Chemicon (Temecula, CA, USA). Vectastain Elite ABC (rabbit IgG) and DAB substrate kits for peroxidase were purchased from Vector Laboratories (Burlingame, CA, USA). Other chemicals were purchased from commercial sources. For the Western blot analysis, a protease inhibitor cocktail, blocking buffer (SuperBlock), and Can Get Signal Solution were purchased from Sigma-Aldrich, Thermo Fisher Scientific (Rockford, IL, USA) and Toyobo (Osaka, Japan), respectively. Rabbit anti-GLT-1 and -GLAST were purchased from Wako (Osaka, Japan). Goat antiactin was purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Rabbit anti-goat and sheep antirabbit horseradish peroxidase (HRP)– conjugated secondary antibodies were purchased from Sigma-Aldrich and GE Healthcare (Buckinghamshire, UK), respectively. ECL plus Western blotting detection regents were also obtained from GE Healthcare. Protein

assay kit using Lowry method was purchased from Bio-Rad Laboratories (Hercules, CA, USA). For the real time reverse-transcription polymerase chain reaction (RT-PCR) assay, a Qiagen RNeasy mini kit was purchased from Qiagen (Hilden, Germany), and a TaqMan Gold RT-PCR Kit, TaqMan Universal PCR Master Mix, and TaqMan probes for detection of GLT-1, GLAST, and Rps29 were purchased from Applied Biosystems (Foster City, CA, USA). Reagents used in the receptor binding assay, [3H]CGP-39653 (NET-1050, specific radioactivity: 50 Ci/mmol), [3H]TCP (NET-886, specific radioactivity: 41 Ci/mmol), and [3H]ifenprodil (NET-1089, specific radioactivity: 43 Ci/mmol), were purchased from PerkinElmer (Waltham, MA, USA). [3H]MDL-105519 (TRK-1043, specific radioactivity: 71 Ci/mmol) was purchased from GE Healthcare. Tris(hydroxymethyl)amino-methane (Tris), L-glutamic acid, ifenprodil, dizolcipin; (⫹)-MK-801, MDL-105519, 4-(2-hydroxyethyl)piperazine-(1) ethanesulfonic acid N-(2-hydroxyethyl)piperazine-N=-(2-ethanesulfonic acid) and Hepes, were purchased from Sigma-Aldrich.

Preparation of primary astrocyte culture Astrocytes were cultured by a modification of a procedure described previously (Juurlink and Walz, 1998). In brief, newborn rat neopallia were mechanically disrupted by pipetting in DMEM/Caand Mg-free phosphate-buffered saline [PBS(⫺)] (1:1). The suspension was filtered through a sterile nylon mesh with 100 ␮m pores (BD Falcon, Bedford, MA, USA). The filtrate was passed through a sterile lens cleaning paper (Fujifilm, Tokyo, Japan). Cells (3.75⫻106) were seeded into a 75 cm2 culture flask (Corning, Corning, NY, USA) with DMEM containing 7.5 mM glucose, 2 mM glutamine, 25 mM NaNCO3, and 10% horse serum. The next day, the culture medium was replaced with DMEM containing 25 mM sorbitol, 2 mM glutamine, 25 mM NaNCO3, and 10% dialyzed horse serum, and the cells were incubated at 37 °C and 95% relative humidity provided with mixture of 5% CO2 and 95% air for 2 weeks. After the incubation period, the cultures were returned to the glucose-containing DMEM medium, and the purity of astrocytes in the culture was examined by an immunocytochemical stain using an antibody to GFAP, which is a specific marker for astrocytes. We confirmed that at least 95% of cells were astrocytes. The high purity– cultured astrocytes were removed from the substratum with Puck’s solution, pH 7.2, containing 3 mM EDTA, 2 mM pyruvate, 7.5 mM glucose, 0.02% DNase and 10 mM Hepes. Approximately 20,000 cells per cm2 were seeded into 96-well primaria culture plates (Becton Dickinson, Franklin Lakes, NJ, USA) and used in the following experiments after the cells became confluent.

Examination of effects of TJ-54 on glutamate uptake in cultured astrocytes TD astrocytes were prepared according to the procedure described by Hazell et al. (2003) to induce TD: the confluent astrocytes were transferred to 96 well-plate and cultured for 7 days in a custom-designed DMEM medium lacking in thiamine (CSTI, Sendai, Japan) and containing 5% horse serum in the presence of 10 ␮M pyrithiamine, which inhibits the enzyme responsible for production of the active form of thiamine. Control astrocytes were cultured in DMEM medium including thiamine and containing 5% horse serum for the same period. TD astrocytes were treated with TJ-54 as follows: astrocytes were cultured for 7 days in the TD medium with various concentrations of TJ-54 (final concentration: 100, 300, 500, or 700 ␮g/ml), which was filtered through a 0.22 ␮m filter. Glutamate uptake ability was evaluated in cells cultured for 7 days. Namely, a 100 ␮M glutamate was exogenously added to astrocyte cultures. The aliquot of the culture medium was carefully

Z. Kawakami et al. / Neuroscience 159 (2009) 1397–1407

1399

collected after incubation for 5 h. The extracellular concentration of glutamate was determined by modifying the colorimetric method described previously (Abe et al., 2000). In brief, 50 ␮l of the medium was mixed with the same volume of a substrate mixture consisting of 20 U/ml glutamate dehydrogenase, 2.5 mg/ml ␤-nicotinamide adenine dinucleotide, 0.25 mg/ml MTT, 100 ␮M 1-methoxyphenazine methosulfate and 0.1% Triton X-100 in 0.05 M Tris–HCl buffer, pH 8.2. The mixture was incubated at 37 °C for 10 min, and the reaction was stopped by adding 100 ␮l of a stop solution, pH 4.7, containing 50% dimethylformamide and 20% sodium dodecyl sulfate (SDS). The amount of formazan produced from MTT by glutamate was colorimetrically determined at a test wavelength of 540 nm and a reference wavelength of 690 nm using a microplate reader. The glutamate concentrations in the samples were estimated from the standard curve that was constructed for each assay using cell-free medium containing known concentrations of L-glutamate.

After the RT, real-time PCR was performed using an ABI Prism 7900HT sequence detection system with a 384 format (Applied Biosystems); the PCR was performed in a 20 ␮l reaction mixture containing 4.0 ␮l of cDNA, 10.0 ␮l of TaqMan Universal PCR Master Mix, and 1.0 ␮l of each TaqMan probe for detection of GLT-1, GLAST, and Rps29. Amplification was performed at 50 °C for 2 min with AmpErase UNG to remove any uracil incorporated into the cDNA, 95 °C for 10 min for AmpliTaq Gold activation, and then a PCR run of 40 cycles at 95 °C for 15 s for denaturation, and 60 °C for 1 min for annealing and extension. The real-time PCR of each sample was run in triplicate. The GLAST or GLT-1 mRNA level of each sample was calculated as the relative ratio of each target gene expression to that of Rps29, the reference gene.

Examination of effects of TJ-54 on glutamate transporter proteins in cultured astrocytes

PC12 cells obtained from Dainippon Sumitomo Pharma (Osaka, Japan) were maintained at 37 °C in 95% air and 5% CO2 with 95% relative humidity in RPMI 1640 (Invitrogen, Grand Island, NY, USA) supplemented with 5% fetal calf serum, 10% heat-inactivated horse serum, penicillin (50 U/ml), and streptomycin (50 ␮g/ml) until used in experiments. On the day of experiments, PC12 cells were seeded into 96-well microplates (5000 cells/well in 100 ␮l medium) in RPMI 1640 supplemented with 5% dialyzed fetal calf serum, 10% dialyzed horse serum, penicillin (50 U/ml), and streptomycin (50 ␮g/ml) without Phenol Red. The cultures were incubated at 37 °C in 95% relative humidity with a mixture of 5% CO2 and 95% air. Forty-eight hours after seeding, the media were renewed with one of three types of fresh culture media (the medium without glutamate as a control, medium including 17.5 mM glutamate, and medium including 17.5 mM glutamate plus various concentrations of TJ-54), and then the cells were incubated for 24 h. The survival of cells was evaluated by a MTT reduction assay (Sakai et al., 1994) after the incubation. The MTT reduction assay was performed as follows: 20 ␮l of 5 mg/ml MTT dissolved in PBS(⫺) was added into each individual well and incubated at 37 °C for 5 h. The reaction was stopped by the addition of 100 ␮l of solubilization solution (10% SDS in 0.01 N HCl), and the blue formazan formed from MTT by the reaction was dissolved by additional incubation for 18 h at 37 °C. Absorbance of the formazan solution was measured using a microplate reader at a test wavelength of 540 nm and a reference wavelength of 690 nm.

Control, TD, and TJ-54-treated TD astrocytes were prepared according to the procedure described in the previous section. In this experiment, GLT-1 and GLAST proteins in the astrocytes cultured for 7 days were analyzed by Western blotting. In brief, the astrocytes were washed with PBS(⫺) and harvested in lysis buffer containing 50 mM Tris, 150 mM NaCl, 0.1% SDS, 1% Nonidet P-40, 0.5% sodium deoxycholate (pH 8.0), and a protease inhibitor cocktail. The cellular debris was removed by centrifugation at 10,000⫻g for 15 min at 4 °C, and the total protein content was measured by a Bio-Rad protein kit. The samples were subjected to SDS–10% polyacrylamide gel electrophoresis and transferred to polyvinylidene difluoride membranes (Hybond-P, GE Healthcare). Nonspecific binding was blocked by incubation with SuperBlock blocking buffer at room temperature for 1 h, followed by incubation overnight at 4 °C in Can Get Signal solution 1 (Toyobo) with rabbit anti-GLT-1 (1:10,000), rabbit anti-GLAST (1:20,000) or goat antiactin (1:10,000). The blots were washed and incubated for 1 h at room temperature in Can Get Signal solution 2 (Toyobo) with sheep antirabbit (1:10,000) or rabbit anti-goat HSRconjugated secondary antibody (1:80,000), and then visualized by ECL plus Western blotting Detection Regents. Image obtained using Typhoon Imaging System was analyzed by ImageQuant TL software (GE Healthcare).

Examination of effects of TJ-54 on glutamate transporter mRNAs in cultured astrocytes Control, TD, and TJ-54-treated TD astrocytes were prepared according to the procedure described in the previous section. In this experiment, the astrocytes were cultured for 6 days, and mRNA of GLT-1, GLAST, or Rps29 as the reference gene, was determined by real time RT-PCR. In brief, total RNA from the astrocytes was extracted using a Qiagen RNeasy mini kit according to the manufacturer’s specification. The extracted total RNA was finally eluted with 50 ␮l of RNase-free water. The RNA concentration was determined spectrophotometrically at 260 nm. Reverse transcription (RT) was carried out using a TaqMan Gold RT-PCR Kit according to the manufacturer’s specification with a TAK-TP400 Thermal Cycler; 1.0 ␮g of total RNA, which was included in a total volume of 50 ␮l final reaction mixture containing 1⫻ TaqMan RT buffer, 5.5 mM MgCl2, 500 ␮M of each dATP, dGTP, dCTP, and dTTP, 2.5 ␮M random hexamers, 20 U RNAse inhibitor, and 62.5 U MultiScribe reverse transcriptase, was used for the RT reaction. This reaction was carried out by preincubation at 25 °C for 10 min and incubation at 48 °C for 30 min, and it was stopped by inactivating the reverse transcriptase by heating at 95 °C for 5 min.

Examination of effect of TJ-54 on glutamate-induced PC12 cell death

Examination of binding of TJ-54 to NMDA receptors Competitive binding of TJ-54 to NMDA receptors was assayed on the recognition sites of glutamate, glycine, phencyclidine (PCP), and polyamine. They were assayed by the methods described by Sills et al. (1991); Siegel et al. (1996); Goldman et al. (1985) and Schoemaker et al. (1990), respectively. The membrane fraction prepared from rat cerebral cortex was used for the binding assay. In brief, the cortex was dissected from brains of normal male Wistar rats (175⫾25 g body weight) after decapitation and immediately homogenized in 20 volumes of ice-cold 50 mM Tris–HCl (pH 7.4) or 50 mM Hepes (pH 7.7) using a Polytron homogenizer. After the pellets were washed by resuspension in the buffer and subsequent centrifugation (20 min, 48,000⫻g), the final pellet (membrane) was suspended in 20 volumes of the buffer. In the binding assay, 500 ␮l of the membrane suspension (glutamate or glycine site: 2.5 mg, polyamine site: 5.0 mg, or PCP site: 6.25 mg) was incubated in duplicate with 20 ␮l of a specific radioligand for each recognition site and 5.25 ␮l of various concentrations of TJ-54 at 4 °C for 20 min (glutamate site), 30 min (glycine site) or 120 min (polyamine site), or at 25 °C for 45 min (PCP site). The specific radioligands used for glutamate, glycine, PCP, or polyamine site were 2.0 nM [3H]CGP-39653: D,L-(E)-2-

1400

Z. Kawakami et al. / Neuroscience 159 (2009) 1397–1407

Fig. 1. Glutamate uptake abilities of astrocytes cultured in T(⫹): thiamine-containing normal medium or T(⫺): TD medium. Astrocytes cultured for 7 d were used. Changes in extracellular glutamate concentrations were measured 1, 3, 5 and 7 h after the addition of 100 ␮M glutamate to the medium. Each value represents the mean⫾SEM of six separate samples. * P⬍0.05, ** P⬍0.01, and *** P⬍0.001 vs. 0 h (one-way ANOVA⫹Dunnett’s test). ††† P⬍0.001 vs. corresponding T(⫹) controls (Student’s t-test).

amino-4-propyl-5-phosphono-3-pentenoic acid, 0.33 nM [3H]MDL105519; (E)-3-(2-phenyl-2-carboxyethenyl)-4,6-dichloro1H-indole-2carboxylic acid, 4.0 nM [3H]TCP: N-[1-(2-thienyl)cyclohexyl] piperidine, or 2.0 nM [3H]ifenprodil, respectively. To define non-specific binding, 1000 ␮M L-glutamic acid, 10 ␮M MDL-105519, 1.0 ␮M dizolcipin, or 10 ␮M ifenprodil was used to bind glutamate, glycine, PCP, and polyamine sites, respectively. After incubation, the receptor–ligand complexes were isolated by rapid filtration through a Whatman GF/B filter using a Cell Harvester (Brandel MLR-48, Skatron micro 96, PerkinElmer). The trapped radioactive complex was washed four times by 3 ml of ice-cold 50 mM Tris–HCl buffer and dried. Radioactivity (counts per minute, cpm) trapped on the dried filter was measured by a liquid scintillation counter (PerkinElmer). The specific binding was defined by subtracting nonspecific binding from total binding, and expressed as the percentage inhibition using the following formula: Inhibition (%)⫽[1–(c–a)/ (b–a)]⫻100, where “a” is the average cpm of non-specific binding, “b” is the average cpm of total binding, and “c” is cpm in the presence of the test substance. IC50 (50% inhibition concentration of the specific binding) values were determined by a nonlinear least squares regression analysis using MathIQ (ID Business Solutions, Ltd., UK).

mal medium or TD medium for 7 days. As shown in Fig. 1, when 100 ␮M glutamate was added to the normal medium, the extracellular concentration of glutamate gradually decreased in a time-dependent manner, and reached at the lowest level (5 ␮M) at 3 h after the addition of glutamate, indicating that most (⬎95%) of the added glutamate was taken up by normal astrocytes. In contrast, no significant changes in the extracellular concentration of glutamate added to TD medium were observed through 7 h incubation time, indicating that the uptake of glutamate into astrocytes was inhibited in the TD condition. Fig. 2 shows changes in the uptake of glutamate by astrocytes when they were cultured in TD or normal condition for 10 days. The glutamate uptake ability on each culture day (day 1, 4, 7, or 10) was evaluated 5 h after the addition of glutamate to the medium. Normal astrocytes took up most of the glutamate added to the medium on days 1, 4, 7 and 10, as indicated by extracellular glutamate levels of less than 20 ␮M during the culture period of 10 days. In contrast, the uptake of TD astrocytes was dysfunctional from the fourth day, as indicated by the result that the extracellular concentrations gradually increased in comparison with those of the corresponding control culture. Furthermore, a glutamate concentration higher than the added glutamate was detected in TD astrocytes cultured 10 days, which is thought to originate in an outflow of glutamate from astrocytes. Thus, the effects of TJ-54 on glutamate uptake ability were evaluated by measuring extracellular glutamate concentration 5 h after addition of glutamate to the medium using astrocytes cultured for 7 days.

Statistical analysis Data are presented as the mean⫾SEM. The statistical significance of differences between groups in cell culture experiments was assessed by Student’s t-test or one-way analysis of variance (ANOVA) followed by Dunnett’s post hoc test. For the levels of proteins and mRNA expressions of glutamate transporters, the statistical significance of differences between groups was assessed by Student’s t-test or one-way ANOVA followed by Bonferroni multiple comparisons test. The significance level in each statistical analysis was accepted at P⬍0.05.

RESULTS Relationships between TD and glutamate uptake in astrocytes Optimal conditions for evaluation of glutamate uptake ability were examined using astrocytes cultured in nor-

Fig. 2. Culture period– dependent decrease in glutamate uptake ability in TD astrocytes. Astrocytes were cultured for 1, 4, 7 or 10 days either in T(⫹): thiamine-containing normal medium or T(⫺): TD medium. Glutamate uptake by astrocytes was evaluated 5 h after addition of 100 ␮M glutamate to the medium. No cells: 100 ␮M glutamate was added to a well without cells, and the glutamate level was measured after incubation for 5 h. The glutamate concentration was the same as the added level, implying that the added glutamate was not decomposed in the medium and not adsorbed to wells in the present experimental condition. Each value represents the mean⫾SEM of six separate samples. ** P⬍0.01 and *** P⬍0.001 vs. corresponding T(⫹) controls, ns: not significant (Student’s t-test).

Z. Kawakami et al. / Neuroscience 159 (2009) 1397–1407

1401

Fig. 3. Morphological changes of cultured rat cortical astrocytes. Astrocytes were cultured for 7 days in T(⫹): thiamine-containing normal medium (A), T(⫺): TD medium (B), or T(⫺) medium containing 1000 ␮g/ml of TJ-54 (C). Astrocytes were fixed with paraformaldehyde and stained with monoclonal antibodies to GFAP and counterstained with hematoxylin. Scale bar⫽200 ␮m.

Effect of TJ-54 on TD-induced dysfunction of glutamate uptake in astrocytes To clarify the effects of TJ-54 on glutamate uptake ability of TD astrocytes, astrocytes were cultured in TD media containing various concentrations of TJ-54 (final concentration: 100, 300, 500 or 700 ␮g/ml) for 7 days. Fig. 3 shows the morphological changes in GFAP-stained astrocytes. The TD astrocyte (Fig. 3B) was a slim spindle shape in comparison with those (flat and polygonal shape) in control cells (Fig. 3A). TJ-54 ameliorated the TD-induced morphological changes (Fig. 3C). The results of the glutamate uptake evaluated by measuring glutamate concentration in the medium are shown in Fig. 4. The glutamate uptake in TD astrocytes was completely inhibited because the extracellular glutamate concentration was approximately 100 ␮M, which was the final concentration added to the medium. TJ-54 significantly decreased the extracellular glutamate in a concentration-dependent manner, suggesting that TJ-54 ameliorated TD-induced dysfunction of glutamate transport.

Fig. 4. Effects of TJ-54 on TD-induced decrease of glutamate uptake in cultured astrocytes. Astrocytes were cultured for 7 days in T(⫹): thiamine-containing normal medium, T(⫺): TD medium, or T(⫺) medium containing various concentrations of TJ-54 (100 –700 ␮g/ml). Extracellular glutamate concentration was measured 5 h after the addition of 100 ␮M glutamate to the medium. Each value represents the mean⫾SEM of six separate samples. *** P⬍0.001 vs. T(⫹), and ††† P⬍0.001 vs. T(⫺): one-way ANOVA⫹Dunnett’s test.

Fig. 5 shows the effects of the glutamate transporter inhibitors TBHA (non-specific inhibitor of the transporters) and DHK (a specific inhibitor of GLT-1) on glutamate uptake in normal astrocytes. Normal glutamate transport function as reflected by a decrease in the extracellular glutamate concentration was completely blocked by TBHA (1000 ␮M), but it was not blocked absolutely by DHK (1000 ␮M). Fig. 6 shows the effects of TBHA and DHK on the amelioration by TJ-54 (500 ␮g/ml) of TD-induced glutamate uptake dysfunction. The increased concentration of extracellular glutamate in the TD culture medium, that is, the dysfunction of glutamate uptake in TD astrocytes, was significantly ameliorated by treatment with TJ-54 (500 ␮g/ ml), similar to the result shown in Fig. 4. The amelioration by TJ-54 was significantly blocked by TBHA (10 –1000 ␮M) in a concentration-dependent manner (Fig. 6A), but it was not blocked even if DHK was increased to a very high concentration (300 –10,000 ␮M) (Fig. 6B).

Fig. 5. Effects of TBHA and DHK on glutamate uptake in normal astrocytes. Astrocytes were cultured for 7 days in thiamine-containing normal medium. Extracellular glutamate concentration was measured 5 h after the addition of 100 ␮M glutamate with or without the transport inhibitor (100 –1000 ␮M TBHA or 100 –1000 ␮M DHK) to the medium. Each value represents the mean⫾SEM of six separate samples. *** P⬍0.001 vs. 0 ␮M TBHA. ns: Not significant vs. 0 ␮M DHK (one-way ANOVA⫹Dunnett’s test).

1402

Z. Kawakami et al. / Neuroscience 159 (2009) 1397–1407

Effect of TJ-54 on glutamate transporters GLAST and GLT-1 proteins To examine the effects of TJ-54 on glutamate transporter proteins, GLT-1 and GLAST immunoblottings were performed and the resulting bands were quantified by normalization to actin. While constitutive GLAST protein content was high in control cells, levels of GLT-1 protein were almost undetectable (data not shown here). Effects of TD and TJ-54 on GLAST transporter protein are shown in Fig. 7. The GLAST protein level in astrocytes cultured in TD medium for 7 days decreased significantly compared with those in control astrocytes. The decreased level was significantly ameliorated by treatment with TJ-54 (500 ␮g/ml). Effect of TJ-54 on mRNA expressions of glutamate transporters GLAST and GLT-1 GLAST and GLT-1 mRNA expressions in high-purity astrocytes cultured in normal medium for 6 days were examined, and the results are shown in Fig. 8. The expression level of GLT-1 mRNA was considerably lower (2.3%) than the GLAST mRNA level. The GLAST level in normal astrocytes was significantly decreased in astrocytes cultured

Fig. 7. Effects of TJ-54 on glutamate transporter GLAST proteins. Astrocytes were cultured for 7 days in T(⫹): thiamine-containing normal medium, T(⫺): TD medium, or T(⫺) medium containing TJ-54 (500 ␮g/ml). In Western blotting, GLAST protein (60 kDa) was quantified by normalization to actin (42 kDa). Each value represents the mean⫾SEM of three separate samples. * P⬍0.05 vs. T(⫹), and † P⬍0.05 vs. T(⫺): Student’s t-test.

in TD medium for 6 days. The TD-induced decrease in GLAST mRNA level was ameliorated by treatment with TJ-54 (125–500 ␮g/ml) in a concentration-dependent manner (Fig. 9A). The GLT-1 mRNA level was also decreased by TD treatment, and the decrease was ameliorated by the addition of TJ-54 in a concentration-dependent manner (Fig. 9B). Effect of TJ-54 on glutamate-induced PC12 cell death Prior to evaluation of the effects of TJ-54 on glutamateinduced cell death, which was evaluated by MTT activity, the optimal concentration of glutamate to induce cell death

Fig. 6. Effects of TBHA (A) and DHK (B) on amelioration of TDinduced dysfunction of glutamate uptake by TJ-54. Astrocytes were cultured for 7 days in T(⫺): TD medium or T(⫺) medium containing TJ-54 (500 ␮g/ml). Extracellular glutamate concentration was measured 5 h after addition of 100 ␮M glutamate with TBHA (10 –1000 ␮M) or DHK (300 –10,000 ␮M). Each value represents the mean⫾SEM of six separate samples. *** P⬍0.001 vs. T(⫺), and †† P⬍0.01 and ††† P⬍0.001 vs. T(⫺)⫹ TJ-54 (500 ␮g/ml): one-way ANOVA⫹ Dunnett’s test.

Fig. 8. GLAST and GLT-1 mRNA expressions in astrocytes cultured in thiamine-containing normal medium for 6 d. Each value is calculated as the ratio relative to an internal control gene Rps 29 and represents the mean⫾SEM of three measurements performed in triplicate.

Z. Kawakami et al. / Neuroscience 159 (2009) 1397–1407

1403

phological changes (Fig. 11C) and the reduced MTT activity in the cells (Fig. 11F). The quantitative results of MTT activities in PC12 cells are shown in Fig. 11G. The cell death, i.e. the reduction of the MTT activity, induced by glutamate was inhibited by treatment with TJ-54 (1–1000 ␮g/ml) in a concentrationdependent manner. Binding of TJ-54 to NMDA receptors The results of competitive binding assays of TJ-54 to NMDA receptors are shown in Fig. 12. TJ-54 (25– 800 ␮g/ml) bound to both glutamate and glycine recognition sites in NMDA receptors in a concentration-dependent manner, but did not bind to PCP and polyamine sites. The IC50 values of TJ-54 for the glutamate and glycine binding sites were estimated to be 37.1 and 172 ␮g/ml, respectively.

DISCUSSION

Fig. 9. Effects of TJ-54 on TD-induced decreases in mRNA expressions of GLAST (A) and GLT-1 (B) in cultured astrocytes. Astrocytes were cultured for 6 days in T(⫹): thiamine-containing normal medium, T(⫺): TD medium, or T(⫺) medium containing TJ-54 (125–500 ␮g/ml). The mRNAs of GLAST, GLT-1, and Rps 29 as an internal control gene were quantified by real time RT-PCR. Each value is expressed as a relative value of the T(⫹) control group after the expression of mRNAs in each sample was normalized to the Rps 29 mRNA level. Each value represents the mean⫾SEM of three measurements performed in triplicate. * P⬍0.05 and *** P⬍0.001 vs. T(⫹), and †† P⬍0.01 and ††† P⬍0.001 vs. T(⫺): one-way ANOVA⫹Bonferroni test.

of PC12 cells was determined, and the results are shown in Fig. 10. Cell death was induced by increasing concentrations of glutamate in the medium in a concentrationdependent manner. The concentration of glutamate that induced 50% cell death was defined as approximately 17.5 mM. Typical photographs (Fig. 11A–C: phase contrast, Fig. 11D–F: MTT activity) of PC12 cells in control, glutamate and glutamate⫹TJ-54 medium are shown in Fig. 11. The number of cells decreased, and the shrunken and birefringent cells reflecting cell death were observed in the glutamate group (Fig. 11B) compared with the control group (Fig. 11A). The MTT activity stained black in the cells decreased in the glutamate group (Fig. 11E) compared with the control group (Fig. 11D). TJ-54 ameliorated the glutamate-induced mor-

We previously demonstrated that TJ-54 ameliorates TDinduced degeneration of neuronal and astroglial cells in the rat brain stem, cerebral cortex, and hippocampus, which are responsible for learning, memory, and various psychological functions (Ikarashi et al., 2006). Numerous studies investigating TD-induced neuronal death or degeneration suggest glutamate excitotoxicity is one of the causative factors (Hazell et al., 1998; Todd and Butterworth, 1999, 2001; Hazell et al., 2001; Carvalho et al., 2006). From these results, we hypothesized that one of mechanisms underlying the effects of TJ-54 may be closely related to amelioration of glutamate excitatory neurotoxicity. The aim of the present study was to evaluate the neuroprotective effect of TJ-54 against glutamate-mediated excitotoxicity. In the present study, 7 days’ culture in TD medium produced a decrease in glutamate uptake by rat astrocytes. Because the glutamate uptake is affected by the activities of two subtypes of glutamate transporter, GLAST and GLT-1, both of which are localized predominantly in

Fig. 10. Relationship between glutamate concentration and PC12 cell death. PC12 cells were incubated for 24 h in control medium with (10 –20 mM) or without glutamate, and cell survival was evaluated by MTT assay. Each value calculated as a percentage of the MTT activity in control PC12 cells represents the mean⫾SEM of six separate samples. *** P⬍0.001 vs. control Glu(⫺): one-way ANOVA⫹Dunnett’s test.

1404

Z. Kawakami et al. / Neuroscience 159 (2009) 1397–1407

Fig. 11. Effects of TJ-54 on glutamate-induced PC12 cell death. PC12 cells were incubated for 24 h in control medium without glutamate, 17.5 mM glutamate-containing medium, or the glutamate medium including various concentrations of TJ-54 (1–1000 ␮g/ml). Typical photographs of PC12 cells in control, glutamate and glutamate⫹TJ-54 (500 ␮g/ml) medium are shown in A–C (phase contrast) and D–F (MTT activity). The number of cells decreased, and the shrunken and birefringent cells (arrows) were observed in the glutamate group (B) compared with the control group (A). The MTT activity stained black in the cells decreased in the glutamate group (E) compared with the control group (D). TJ-54 ameliorated the glutamate-induced morphological changes (C) and the reduced MTT activity in the cells (F). The cell morphology was observed by phase-contrast microscopy. Scale bar⫽100 ␮m. The quantitative results of MTT activities in PC12 cells are shown in G. Each value calculated as a percentage of the MTT activity in control PC12 cells represents the mean⫾SEM of six separate samples. *** P⬍0.001 vs. control Glu (⫺), and †† P⬍0.01 and ††† P⬍0.001 vs. Glu (⫹): one-way ANOVA⫹Dunnett’s test.

Z. Kawakami et al. / Neuroscience 159 (2009) 1397–1407

Fig. 12. Competitive binding of TJ-54 to glutamate, glycine, PCP, and polyamine recognition sites in NMDA receptors. The membrane fraction of rat cerebral cortex was used for the binding assay. Binding of TJ-54 (25– 800 ␮g/ml) is expressed as % inhibition of TJ-54 against total specific binding of a radioligand to each binding site. Each value is expressed as the mean of duplicate determinations.

astrocytes (Rothstein et al., 1994; Lehre et al., 1995), the TD-induced decrease in glutamate uptake by the astrocytes was inferred to involve dysfunction of these transporters. To clarify the hypothesis, changes in the transporters in TD astrocytes were examined at the protein and mRNA levels. While constitutive GLAST protein content was high in control cells, the level of GLT-1 protein was almost undetectable. In the mRNAs analysis, the expression of GLT-1 mRNA was less than 2.3% of GLAST mRNA. These results are consistent with the previous reports (Gegelashvili et al., 1997; Swanson et al., 1997; Schlag et al., 1998) demonstrating that the GLAST protein concentration in highly pure cultured astrocytes is comparatively high, while GLT-1 protein is almost undetectable or detectable at very low levels. The in vivo study using TD rats has demonstrated downregulation of GLAST and GLT-1 proteins in the medial thalamus which is a region vulnerable to TD (Hazell et al., 2001). In the present study, to confirm a functional role of the glutamate transporters, the effects of the glutamate transport inhibitor TBHA (effective doses: 100 –1000 ␮M), which is a non-specific inhibitor of GLAST and GLT-1 (Shimamoto et al., 1998), and DHK (effective doses: 100 –1000 ␮M), which is a specific inhibitor of GLT-1 (Kawahara et al., 2002), on the glutamate uptake function were examined in normal astrocytes. Although TBHA completely blocked glutamate transport (glutamate uptake), DHK did not affect it. These results suggest that the glutamate transport function in the cultured astrocytes used in the present study is regulated mainly by the function of GLAST. TJ-54 ameliorated TD-induced decreases in the glutamate uptake function, and the levels of GLAST protein and the mRNA. The amelioration of glutamate uptake dysfunction in TD astrocytes by TJ-54 was completely blocked by TBHA (1000 ␮M), but not blocked absolutely by DHK (even at 10,000 ␮M), suggesting that the amelioration by TJ-54 is predominantly reflected by ameliorating GLAST transporter dysfunction, under the present experimental conditions. We could not clarify the effect of TJ-54 on

1405

GLT-1 protein, because GLT-1 protein was undetectable in highly pure cultured astrocytes. However, because the TD-induced decrease in the mRNA level was ameliorated by treatment of TJ-54, we could not rule out the involvement of TJ-54 on GLT-1 in the present study. We think that detailed examinations on the possibility will be necessary in the future. Hazell et al. (2003) demonstrated that the loss of glutamate uptake function and downregulation of GLAST protein in cultured TD astrocytes were blocked by treatment with H7, which is a protein kinase C (PKC) inhibitor, or DCG IV, which is a group II metabotropic glutamate receptor (mGluR) agonist. The decreases in GLAST expression and glutamate transport activity due to activating PKC are also supported by the results demonstrating that phorbol esters, which activate PKC, decrease GLAST expression and glutamate transport activity (Conradt and Stoffel, 1997; Gonzalez et al., 1999). Glycyrrhizic acid in Glycyrrhiza root, which is a constituent herb of TJ-54, is reported to inhibit the Ca2⫹- and phospholipid-dependent phosphotransferase activity of PKC (O’Brian et al., 1990). On the other hand, the increase in glutamate uptake and improvement of GLAST levels by activation of mGluR with DCG IV are thought to be due to negative coupling to cAMP in the astrocytes (Petralia et al., 1996; Gegelashvili et al., 2000; Hazell et al., 2003). These mechanisms, including PKC and mGluR, may be involved in the underlying mechanisms of TJ-54, although further studies are necessary to confirm the hypothesis. The present results, at least, suggest that TJ-54 possesses an ameliorative effect on the transport function that protects neuronal cells from excessive extracellular glutamate. It has been known that glutamate is toxic to cultured neuronal cells via two different processes, both of which result in the production of free radicals. The classical pathway known as excitotoxicity, occurs through the activation of NMDA and non-NMDA glutamatergic receptors (Olney, 1969) and subsequent calcium influx into the cells (Michaels and Rothman, 1990). Recently, it has been reported that PC12 cells do not express a normal profile of NMDA receptors (Edwards et al., 2007). Glutamate toxicity on the PC12 cells is reported to be mediated by oxidative glutamatergic toxicity pathway, which is related to the cystine/glutamate antiporter system (Pereira et al., 2000; Lo et al., 2008). Because the competition by glutamate for cystine/glutamate antiporter induces an imbalance in the homeostasis of cysteine which is the precursor of glutathione (GSH), the inhibition of cystine uptake by the constant and high exposure to glutamate is supposed to give rise to an inability to maintain intracellular GSH levels, leading to a reduced ability to protect against oxidative injury of the cells, and ultimately cell death (Pereira et al., 2000). In addition, Wang et al. (2007) have reported that TD induces endoplasmic reticulum (ER) stress in neurons, which is suggested to be caused by oxidative stress. The ER stress has been implicated in various neurodegenerative processes, such as brain ischemia, Alzheimer’s disease, and Parkinson’s disease. More recently, TJ-54 was demonstrated to inhibit

1406

Z. Kawakami et al. / Neuroscience 159 (2009) 1397–1407

amyloid ␤-induced cytotoxicity in a primary culture of rat cortical neurons (Tateno et al., 2008). Therefore, the results demonstrating that TJ-54 is protected glutamate-induced PC12 cell death may be due to anti-cytotoxic action not via NMDA receptors, such as antioxidative stress or anti-ER stress, although the detailed examination to clarify the hypothesis will be necessary in future. To date, oxyindole alkaloids such as isorhynchophylline, isocorynoxeine, and rhynchophylline and indole alkaloids such as hirsuteine and hirsutine contained in Uncaria rhynchophylla, which is a constituent herb of TJ-54, are known for their protective effects against glutamate-induced neuronal death in cultured cerebellar granule cells (Shimada et al., 1999). Uncaria rhynchophylla and its phenolic compounds, such as epicatechin, caffeic acid, and quercetin, have been reported to protect against oxidative damage by H2O2 in NG108-15 cells (Mahakunakorn et al., 2004). Obtaining information regarding these constituent herbs and their components is thought to be important to clarify the mechanism of TJ-54 in the future. The stimulation of NMDA receptors by glutamate induces an intracellular Ca2⫹ overload, and activates neuronal NOS, resulting in excessive production of NO and formation of reactive oxygen species and lipid peroxidation to induce neuronal death (Weikert et al., 1997). Receptor binding assays in the present study revealed that TJ-54 bound strongly to glutamate and glycine recognition sites in NMDA receptors (Fig. 12). To date, it has been shown that Uncaria rhynchophylla blocks NMDA-induced currents in cortical neurons and reduces NMDA-induced neuronal death (Sun et al., 2003; Lee et al., 2003). Rhynchophylline and isorhynchophylline, which are ingredients of Uncaria rhynchophylla, have been reported to be antagonists of the NMDA receptor, which reduced the maximal current responses evoked by NMDA and glycine, but showed no interaction with the polyamine-binding site, Zn2⫹ site, proton site, or redox modulatory site on the NMDA receptor (Kang et al., 2002). Taken together, our results of the NMDA binding assay suggests the possibility that TJ-54 might possess antagonistic binding to NMDA receptors, although a detailed examination will be necessary in the future. A traditional Japanese medicine, TJ-54 consists of a combination of several medical herbs and has various components. Although additional studies are required to clarify the underlying mechanisms of TJ-54, including it components, the present study suggests, at least, that TJ-54 may possess a neuroprotective effect against the glutamate-mediated excitotoxicity due not only to amelioration of dysfunction of astrocytes but also direct protection of neuronal cells.

REFERENCES Abe K, Abe Y, Saito H (2000) Evaluation of L-glutamate clearance capacity of cultured rat cortical astrocytes. Biol Pharm Bull 23: 204 –207. Behrens PF, Franz P, Woodman B, Lindenberg KS, Landwehrmeyer GB (2002) Impaired glutamate transport and glutamate-glutamine cycling: downstream effects of the Huntington mutation. Brain 125:1908 –1922.

Carvalho FM, Preira SRC, Pires RGW, Ferraz VP, Romano-Silva MA, Oliveira-Slva IF, Ribeiro AM (2006) Thiamine deficiency decreases glutamate uptake in the prefrontal cortex and impairs memory performance in a water maze test. Pharmacol Biochem Behav 83:481– 489. Cheung NS, Pacoe CJ, Giardina SF, John CA, Beart PM (1998) Micromolar L-glutamate induces extensive apoptosis in an apoptotic-necrotic continuum of insult-dependent, excitotoxic injury in cultured cortical cella. Neuropharmacology 37:1419 –1429. Choi DW (1988) Glutamate neurotoxicity and diseases of the nervous system. Neuron 1:623– 634. Collins GH (1967) Glial cell changes in the brain stem of thiaminedeficient rats. Am J Pathol 50:791– 814. Conradt M, Stoffel W (1997) Inhibition of the high-affinity brain glutamate transporter GLAST-1 via direct phosphorylation. J Neurochem 68:1244 –1251. Edwards MA, Loxley RA, Williams AJ, Connor M, Phillips JK (2007) Lack of functional expression of NMDA receptors in PC12 cells. Neurotoxicology 28:876 – 885. Gegelashvili G, Danbolt NC, Schousboe A (1997) Neuronal soluble factors differentially regulate the expression of the GLT-1 and GLAST glutamate transporters in cultured astroglia. J Neurochem 69:2612–2615. Gegelashvili G, Dehnes Y, Danbolt NC, Schousboe A (2000) The high-affinity glutamate transporters GLT-1, GLAST, EAAT4 are regulated via different signaling mechanisms. Neurochem Int 37:163–170. Goldman ME, Jacobson AE, Rice KC, Paul SM (1985) Differentiation of [3H]phencyclidine and (⫹)-[3H]SKF-10,047 binding sites in rat cerebral cortex. FEBS Lett 190:333–336. Gonzalez MI, Lopez-Colom AM, Ortega A (1999) Sodium-dependent glutamate transporter in Muller glial cells: regulation by phorbol esters. Brain Res 831:140 –145. Han F, Shioda N, Moriguchi S, Qin ZH, Fukunaga K (2008) Downregulation of glutamate transporters is associated with elevation in extracellular glutamate concentration following rat microsphere embolism. Neurosci Lett 430:275–280. Harkany T, Ábrahám I, Timmerman W, Laskay G, Tóth B, Sasvári M, Kónya C, Sebens JB, Korf J, Nyakas C, Zarándi M, Soós K, Penke B, Luiten PGM (2000) ␤-Amyloid neurotoxicity is mediated by a glutamate-triggered excitotoxic cascade in rat nucleus basalis. Eur J Neurosci 12:2735–2745. Hazell AS, Butterworth RF, Hakim AM (1993) Cerebral vulnerability is associated with selective increase in extracellular glutamate concentration in experimental thiamine deficiency. J Neurochem 61:1155–1158. Hazell AS, Todd KG, Butterworth RF (1998) Mechanism of neuronal cell death in Wernicke’s encephalopathy. Metab Brain Dis 13: 97–122. Hazell AS, Rao KV, Danbolt NC, Pow DV, Butterworth RF (2001) Selective downregulation of the astrocyte glutamate transporters GLT-1 and GLAST within the medial thalamus in experimental Wernicke’s encephalopathy. J Neurochem 78:560 –568. Hazell AS, Pannunzio P, Rama Rao KV, Pow DV, Rambaldi A (2003) Thiamine deficiency results in downregulation of the GLAST glutamate transporter in cultured astrocytes. Glia 43:175–184. Ikarashi Y, Iizuka S, Imamura S, Kanno H, Yamaguchi T, Yuzurihara M, Kase Y, Takeda S (2006) Yokukansan inhibits memory disturbance, anxiety and neurological symptoms in thiamine-deficient rats. Dementia 20(Suppl):49. Iwasaki K, Satoh-Nakagawa T, Maruyama M, Monma Y, Nemoto M, Tomita N, Tanji H, Fujiwara H, Seki T, Fujii M, Arai H, Sasaki H (2005a) A randomized, observer-blind, controlled trial of the traditional Chinese medicine Yi-Gan San for improvement of behavioral and psychological symptoms and activities of daily living in dementia patients. J Clin Psychiatry 66:248 –252. Iwasaki K, Maruyama M, Tomita N, Furukawa K, Nemoto M, Fujiwara H, Seki T, Fujii M, Kodama M, Arai H (2005b) Effects of the

Z. Kawakami et al. / Neuroscience 159 (2009) 1397–1407 traditional Chinese herbal medicine Yi-Gan San for cholinesterase inhibitor-resistant visual hallucinations and neuropsychiatric symptoms in patients with dementia with Lewy bodies. J Clin Psychiatry 66:1612–1613. Juurlink BHJ, Walz W (1998) Neural cell culture techniques. In: Neuromethods, Vol. 33, cell neurobiology techniques (Boulton AA, Baker GB, Bateson, AN, eds), pp 53–102. Totowa, NJ: Humana Press. Kanai Y, Smith CP, Hediger MA (1993) The elusive transporters with a high affinity for glutamate. Trends Neurosci 16:365–370. Kang TH, Murakai Y, Matsumoto K, Takayama H, Kitajima M, Aimi N, Watanabe H (2002) Rhynchophylline and isorhynchophylline inhibit NMDA receptors expressed in Xenopus oocytes. Eur J Pharmacol 455:27–34. Kanno H, Iizuka S, Kawakami Z, Yamaguchi T, Sekiguchi K, Yuzurihara M, Ikarashi Y, Kase Y (2008) Effects of yokukansan on extracellular glutamate in the ventral posterior medial thalamus of thiamine-deficient rats. J Pharmacol Sci 106(Suppl):136. Kawahara K, Hosoya R, Sato H, Tanaka M, Nakajima T, Iwabuchi S (2002) Selective blockade of astrocytic glutamate transporter GLT-1 with dihydrokainate prevents neuronal death during ouabain treatment of astrocyte/neuron cocultures. Glia 40:337–349. Langlais PJ, Zhang SX (1993) Extracellular glutamate is increased in thalamus during thiamine deficiency-induced lesions and is blocked by MK-801. J Neurochem 61:2175–2182. Lee J, Son D, Lee P, Kim D-K, Shin M-C, Jang M-H, Kim C-J, Kim Y-S, Kim H (2003) Protective effect of methanol extract of Uncaria rynchophylla against excitotoxicity induced by N-methyl-D-aspartate in rat hippocampus. J Pharmacol Sci 92:70 –73. Lehre KP, Levy LM, Ottersen OP, Storm-Mathisen J, Danbort NC (1995) Differential expression of two glutamate transporters in rat brain: quantitative and immunocytochemical observations. J Neurosci 15:1836 –1853. Lo M, Wang YZ, Gout PW (2008) The x(c)-cystine/glutamate antiporter: a potential target for therapy of cancer and other diseases. J Cell Physiol 215:593– 602. Mahakunakorn P, Tohda M, Murakami Y, Matsumoto K, Watanabe H (2004) Antioxidant and free radical-scavenging activity of chotosan and its related constituents. Biol Pharm Bull 27:38 – 46. Michaels RL, Rothman SM (1990) Glutamate neurotoxicity in vitro: antagonist pharmacology and intracellular calcium concentrations. J Neurosci 10:283–292. Murata A, Nakagawasai O, Yamadera F, Oba A, Wakui K, Arai Y, Tadano T (2004) Dysfunction of serotonergic systems in thiaminedeficient diet fed mice: effect of SSRI on abnormality induced by thiamine deficiency. Jpn J Neuropsychopharmacol 24:93–99. Nakagawasai O, Tadano T, Niijima K, Tan-no F, Kisara K (2000) Immunohistochemical estimation of rat brain somatostatin on avoidance learning impairment induced by thiamine deficiency. Brain Res Bull 51:47–55. O’Brian CA, Ward NE, Vogel VG (1990) Inhibition of protein kinase C by the 12-O-tetradecanoylphorbol-13-acetate antagonist glycyrrhetic acid. Cancer Lett 49:9 –12. Olney JW (1969) Brain lesions, obesity, and other disturbances in mice treated with monosodium glutamate. Science 164:719 –721. Pereira CF, Oliveira CR (2000) Oxidative glutamate toxicity involves mitochondrial dysfunction and perturbation of intracellular Ca2⫹ homeostasis. Neurosci Res 37:227–236. Petralia RS, Wang YX, Niedzielski AS, Wenthhold RJ (1996) The metabotropic glutamate receptors, mGluR2 and mGluR3, show

1407

unique postsynaptic, presynaptic and glial localizations. Neuroscience 71:949 –976. Robertson DM, Wasan SM, Skinner DB (1968) Ultrastructural features of early brain stem lesions of thiamine-deficient rats. Am J Pathol 52:1081–1097. Rothstein JD, Martin L, Levey AL, Dykes-Hoberg M, Jin L, Wu D, Nash N, Kuncl RW (1994) Localization of neuronal and glial glutamate transporters. Neuron 13:713–725. Sakai M, Miyazaki A, Hakamata H, Sasaki T, Yui S, Yamazaki M, Shichiri M, Horiuchi S (1994) Lysophosphatidylcholine plays an essential role in the mitogenic effect of oxidized low density lipoprotein on murine macrophages. J Biol Chem 269:31430 –31435. Schlag BD, Vondrasek JR, Munir M, Kalandadze A, Zelenaia OA, Rothstein JD, Robinson MB (1998) Regulation of the glial Na⫹dependent glutamate transporters by cyclic AMP analogs and neurons. Mol Pharmacol 53:355–369. Schoemaker H, Allen J, Langer SZ (1990) Binding of [3H]ifenprodil, a novel NMDA antagonist, to a polyamine-sensitive site in the rat cerebral cortex. Eur J Pharmacol 176:249 –250. Shimada Y, Goto H, Itoh T, Sakakibara I, Kubo M, Sasaki H, Terasawa K (1999) Evaluation of the protective effects of alkaloids from the hooks and stems of Uncaria sinensis on glutamate-induced neuronal death in cultured cerebellar granule cells from rats. J Pharm Pharmacol 51:715–722. Shimamoto K, LeBrun B, Yasuda-Kamatani Y, Sakaitani M, Shigeri Y, Yumoto N, Nakajima T (1998) DL-threo-beta-benzyloxyaspartate, a potent blocker of excitatory amino acid transporters. Mol Pharmacol 53:195–201. Siegel BW, Sreekrishna K, Baron BM (1996) Binding of the radiolabeled glycine site antagonist [3H]MDL 105,519 to homomeric NMDA-NR1a receptors. Eur J Pharmacol 312:357–365. Sills MA, Fagg G, Pozza M, Angst C, Brundish DE, Hurt SD, Wilusz EJ, Williams M (1991) [3H]CGP 39,653: a new N-methyl-D-aspartate antagonist radioligand with low nanomolar affinity in rat brain. Eur J Pharmacol 192:19 –24. Sun X, Chan LN, Gong X, Sucher NJ (2003) N-methyl-D-aspartate receptor antagonist activity in traditional Chinese stroke medicines. Neurosignals 12:31–38. Swanson RA, Liu J, Miller JW, Rothstein JD, Farrell K, Stein BA, Longuemare MC (1997) Neuronal regulation of glutamate transporter subtype expression in astrocytes. J Neurosci 17:932–940. Tateno M, Ukai W, Ono T, Saito S, Hashimoto E, Saito T (2008) Neuroprotective effects of Yi-Gan San against beta amyloid-induced cytotoxicity on rat cortical neurons. Prog Neuropsychopharmacol Biol Psychiatry 32:1704 –1707. Todd KG, Butterworth RF (1999) Early microglial response in experimental thiamine deficiency: an immunohistochemical analysis. Glia 25:190 –198. Todd KG, Butterworth RF (2001) In vivo microdialysis in an animal model of neurological disease: thiamine deficiency (Wernicke) encephalopathy. Methods 23:55– 61. Wang X, Wang B, Fan Z, Shi X, Ke ZJ, Luo J (2007) Thiamine deficiency induces endoplasmic reticulum stress in neurons. Neuroscience 144:1045–1056. Weikert S, Freyer D, Weh M, Lsaev N, Busch C, Schultze J, et al. (1997) Rapid Ca2⫹ dependent NO-production from central nervous system in culture measured by NO-nitric/ozone chemoluminescence. Brain Res 748:1–11.

(Accepted 2 February 2009) (Available online 7 February 2009)