QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: jpet.104.070912v1 311/2/836    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Katsuki, H. Articles by Akaike, A. Search for Related Content PubMed PubMed Citation Articles by Katsuki, H. Articles by Akaike, A.

NEUROPHARMACOLOGY

Endogenous D-Serine Is Involved in Induction of Neuronal Death by N-Methyl-D-aspartate and Simulated Ischemia in Rat Cerebrocortical Slices

Department of Pharmacology, Graduate School of Pharmaceutical Sciences, Kyoto University, Kyoto, Japan

Received May 2, 2004; accepted July 7, 2004.

	Abstract

Top Abstract Materials and Methods Results Discussion References

 
Emerging evidence indicates that D-serine rather than glycine serves as an endogenous agonist at glycine site of the N-methyl-D-aspartate (NMDA) subtype of glutamate receptors, in several nervous tissues, including the developing cerebellum and the retina. Here, we examined whether endogenous D-serine plays a significant role in neuronal damage resulting from excitotoxic insults in the cerebral cortex, using rat brain slices maintained in a defined salt solution. Neuronal cell death induced by application of NMDA or by oxygen-glucose deprivation (simulated ischemia) was markedly suppressed by a competitive glycine site antagonist 2,7-dichlorokynurenic acid. Addition of glycine or D-serine did not augment neuronal damage by NMDA or simulated ischemia, indicating that sufficient amount of glycine site agonist(s) is supplied endogenously within the slices. Application of D-amino acid oxidase, an enzyme that degrades D-serine, markedly inhibited neuronal damage by NMDA and simulated ischemia, which was reversed by addition of excess D-serine or glycine. Sensitivity to the glycine site antagonist of NMDA- or ischemia-induced damage was not affected by the presence of a non-NMDA receptor antagonist, suggesting that kainate receptor-stimulated D-serine release as demonstrated in primary cultured astrocytes does not contribute significantly to the extent of neuronal injury in these settings. The present results suggest that endogenous supply of D-serine as a glycine site agonist is important for neuronal injury involving NMDA receptor overactivation in the cerebral cortex.

A remarkable feature of NMDA receptor channels is that, in addition to the agonist glutamate, they require a coagonist glycine for their activation (Johnson and Ascher, 1987; Kleckner and Dingledine, 1988). The idea that glycine site activation is essential for NMDA receptor-mediated physiological processes is supported by the findings that glycine site antagonists block the induction of long-term potentiation of hippocampal synaptic transmission (Izumi et al., 1990) as well as acquisition of spatial memory in rats (Watanabe et al., 1992). Apart from the physiological aspects of NMDA receptor functions, several studies have demonstrated remarkable protective effects of glycine site antagonists against NMDA receptor-mediated neuronal injury in vitro (McNamara and Dingledine, 1990; Patel et al., 1990) and in vivo (Foster et al., 1990), indicating that glycine site stimulation is also crucial for neuropathological events as a consequence of NMDA receptor activation.

Although earlier pharmacological experiments have already shown that D-serine is a potent agonist at glycine site of NMDA receptors (Kleckner and Dingledine, 1988), the compound was not initially regarded as an endogenous ligand because D-amino acid was assumed to be absent in mammalian tissues and organs. Since the discovery of substantial quantities of brain D-serine (Hashimoto et al., 1992; Hashimoto and Oka, 1997), however, emerging evidence supports the hypothesis that D-serine, rather than glycine, may be important as an endogenous "glycine site" ligand at least in several brain regions (Hashimoto et al., 1993; Schell et al., 1995; Snyder and Ferris, 2000). For example, NMDA receptor-mediated neurotransmission in neonatal rat cerebellar slices is markedly attenuated by application of D-amino acid oxidase (DAAOX), an enzyme that efficiently degrades extracellular D-serine (Mothet et al., 2000). DAAOX was also able to attenuate NMDA receptor-mediated synaptic transmission in hippocampal slices and dissociated hippocampal cultures (Mothet et al., 2000), suggesting that endogenous D-serine is essential for maintaining normal synaptic transmission. Recent reports using a similar approach proposed that endogenous D-serine supplied from glial cells may set the sensitivity of NMDA receptors at glutamatergic synapses in the retina (Stevens et al., 2003) and regulate the induction of long-term synaptic plasticity in the hippocampus (Yang et al., 2003).

Biochemical and immunohistochemical observations revealed that D-serine is present in rat cerebral cortex at high levels (Hashimoto et al., 1995; Schell et al., 1995, 1997). The patterns of regional variations and postnatal changes in brain D-serine levels show strong correlations with those of NMDA receptors (Hashimoto et al., 1993). The major cell types containing high levels of D-serine in the telencephalon seem to be astrocytes that are located in close vicinity to NMDA receptor NR2A/B subunits expressed on neuronal dendritic spines (Schell et al., 1995, 1997). Existence of serine racemase, a D-serine synthesizing enzyme, has also been demonstrated in astrocytes of the cerebral cortex (Wolosker et al., 1999). In addition, activation of non-NMDA subtypes of glutamate receptors stimulates efflux of D-serine from cultured astrocytes prepared from rat cerebral cortex (Schnell et al., 1995; Ribeiro et al., 2002). Despite these lines of evidence, however, roles of endogenous D-serine in the cerebral cortex have not been proven. Particularly, the issue concerning whether endogenous D-serine plays a significant role in pathological events such as excitotoxic injury has not been addressed in any brain regions. Using brain slice preparation, we examined potential pathophysiological roles of endogenous D-serine in the cerebral cortex.

	Materials and Methods

Top Abstract Materials and Methods Results Discussion References

 
Materials. MK-801 and 2,3-dihydroxy-6-nitro-7-sulfamoyl-benzo-[f]quinoxaline (NBQX) were purchased from Sigma-Aldrich (St. Louis, MO). Homoquinolinic acid and 2,7-dichlorokynurenic acid (DCKA) were obtained from Tocris Cookson Inc. (Bristol, UK). DAAOX was from MP Biomedicals (Irvine, CA) and -aminoisobutyric acid (AIB) was from Wako Pure Chemicals (Osaka, Japan). Other chemicals and reagents including NMDA, glycine, D-serine, and L-serine were obtained from Nacalai Tesque (Kyoto, Japan).

Slice Preparation. Cerebrocortical slices were prepared from 10- to 16-day-old Wistar rats of both sexes (Nihon SLC, Shizuoka, Japan). Procedures in this study were approved by our institutional animal experimentation committee, and animals were treated in accordance with the Guide for the Care and Use of Laboratory Animals by the National Institutes of Health. Rats under deep anesthesia with diethyl ether were decapitated, and the brains were rapidly removed from the skull and immersed into ice-cold bathing solution saturated with 95% O₂, 5% CO₂. The standard bathing solution consisted of 124 mM NaCl, 3 mM KCl, 1.2 mM NaH₂PO₄, 2.4 mM CaCl₂, 1.3 mM MgSO₄, 26 mM NaHCO₃ and 10 mM glucose. Coronal brain slices (400 µm in thickness) were cut with the use of a vibrating blade microtome (Leica, Nussloch, Germany), and then they were separated into two hemispheres. Slices containing the parietal cortex and the striatum were used for the following experiments, after other brain structures such as the septum and the basal forebrain were removed from each slice. Slices were transferred to a reservoir filled with the standard bathing solution continuously bubbled with 95% O₂, 5% CO₂ and were maintained at 4°C for 1 to 2 h.

Drug Treatment and Simulated Ischemia. Nylon mesh beds (Izumi et al., 1995) of 25 mm in diameter were used for treatment of slices with NMDA or homoquinolinic acid. Slices were transferred into nylon mesh beds (four slices per each bed) that were inserted into a six-well plate, each well containing 6 ml of the standard bathing solution. After preincubation for 1 h, exposure of slices to excitotoxins was achieved by transfer of nylon mesh beds to another six-well plate filled with 6 ml of the bathing solution containing excitotoxins. After 30-min treatment, slices were transferred into yet another plate filled with fresh standard solution and incubated for further 6 h. All solutions were continuously bubbled with 95% O₂,5% CO₂ and maintained at 34°C. In the case of simulated ischemia, slices preincubated for 1 h in the standard oxygenated bathing solution were transferred into glucose-free bathing solution saturated with 95% N₂, 5% CO₂ at 34°C. After oxygen/glucose deprivation for 20 min (unless otherwise indicated), slices were transferred into fresh, glucose-containing standard solution bubbled with 95% O₂, 5% CO₂ and incubated for further 6 h. MK-801, DCKA, and NBQX were applied from the preincubation period of 1 h and were also added in the bathing solution during excitotoxin or ischemic treatment and postincubation for 6 h. D-Serine and glycine were applied only during excitotoxin or ischemic treatment. DAAOX, as well as MK-801 and DCKA in several experiments, was applied from the preincubation period and during excitotoxin or ischemic treatment. Heat inactivation of DAAOX was done by 5-min incubation of DAAOX solution in distilled water at 100°C.

Histological Assessment of Neuronal Damage. At termination of postincubation, slices were fixed with 4% paraformaldehyde in phosphate-buffered saline (PBS) overnight at 4°C. Then, slices were dehydrated with graded ethanol and xylene, embedded in paraffin, and cut into sections of 5 µm in thickness with a sliding microtome. Nissl staining with toluidine blue was performed on these sections after deparaffinization, and the sections were examined by bright-field optical microscopy. Positively stained round or oval cell bodies showing clear boundary with surrounding parenchyma were judged as viable neurons, and the number of viable neurons in an area of 170 x 245 µm² within the parietal cortex of individual slices was counted. The grid for cell counting was positioned at layer III of the forelimb/hindlimb area, the most dorsome-dial parts of the parietal cortex (Zilles and Wree, 1995). This region was chosen because it contained a high and homogenous density of neurons, and also because it could be easily identified by the presence of a well developed layer IV underneath, a characteristic of the parietal cortex. Data shown in figures are from a representative set of experiments. Reproducibility of the results was confirmed by two or three different sets of experiments.

In several experiments, morphology of cell nuclei was examined by staining with Hoechst 33342. Sections were treated with 2 µg/ml Hoechst 33342 in PBS at room temperature, washed three times in PBS, and viewed under fluorescence microscope with a 4,6-dia-midino-2-phenylindole filter set.

Statistics. Data are expressed as means ± S.E.M (n = 4 for each condition). Statistical significance was evaluated by one-way analysis of variance followed by Student-Newman-Keuls test. Probability values less than 0.05 were considered significant.

	Results

Top Abstract Materials and Methods Results Discussion References

 
NMDA- and Ischemia-Induced Neuronal Injury in Rat Cerebrocoritical Slices. Cerebrocortical slices were prepared from 10- to 16-day-old rats. When these slices were maintained in an oxygenated balanced salt solution for 7.5 h and examined histologically thereafter, the majority of cells inside slices displayed healthy looking, large round or oval cell bodies, which were judged as viable neurons (Fig. 1A). On the other hand, when slices were treated with NMDA for 30 min and incubated for further 6 h in drug-free solution, the number of viable neurons markedly decreased. The cytotoxic effect of NMDA was concentration-dependent within a range of 10 to 300 µM and was abolished by 10 µM MK-801, a noncompetitive NMDA receptor antagonist (Fig. 1B). Simulated ischemia, which was achieved by incubation of slices in a solution devoid of oxygen and glucose followed by postincubation for 6 h, also resulted in a marked decrease in the number of viable neurons (Fig. 1A). The extent of injury was dependent on the duration of oxygen/glucose deprivation (Fig. 1C). The decrease in the number of viable neurons induced by 20 min of simulated ischemia was markedly suppressed by 10 µM MK-801, whereas the compound did not provide a significant protective effect against injury by 30-min ischemia. Although expression levels and subunit compositions of NMDA receptors in the cerebral cortex may undergo developmental changes (Sheng et al., 1994), we did not find obvious changes in NMDA- and ischemia-induced neuronal death within the ages ranging from postnatal 10 to 16 days.

View larger version (44K): [in this window] [in a new window]   Fig. 1. Neuronal damage induced by NMDA or simulated ischemia in rat cerebrocortical slices. A, representative images showing Nissl-stained sections obtained from rat cerebrocortical slices. From left to right, a slice incubated for 7.5 h under control conditions, a slice treated with 30 µM NMDA for 30 min and incubated for further 6 h in a drug-free standard solution, and a slice treated with oxygen/glucose-free solution for 20 min and incubated for a further 6 h in an oxygenated standard solution. After incubation, slices were fixed and processed for preparation of 5-µm-thick sections that were stained with toluidine blue. Scale bar, 50 µm. B, concentration-dependent neurotoxicity of NMDA and its reversal by MK-801. Cerebrocortical slices were treated with NMDA at indicated concentrations for 30 min and then incubated in the standard solution for further 6 h. MK-801 (10 µM) was applied from 1 h before NMDA treatment and was present during the entire course of incubation. ***, P < 0.001 versus control; ###, P < 0.001 versus NMDA alone. C, ischemia-induced neuronal damage and effect of MK-801. Slices were incubated in oxygen/glucose-free solution for indicated periods and then returned to oxygenated standard solution and incubated for a further 6 h. Where indicated, MK-801 (10 µM) was applied from 1 h before ischemic treatment and was present during the entire course of incubation. ***, P < 0.001 versus control; ###, P < 0.001 versus ischemia alone.

Endogenous Supply of Glycine Site Ligand(s) Is Both Necessary and Sufficient for Neuronal Damage by NMDA and Simulated Ischemia. Because in the present experimental settings the cortical slices were bathed in a balanced salt solution, ligand(s) for glycine site of NMDA receptors should be provided endogenously within the slices. To ascertain whether the presence of endogenous glycine site ligand(s) is critical for NMDA receptor-mediated neuronal damage, we examined the effect of DCKA on NMDA- and ischemia-induced injury. DCKA is a potent antagonist acting at the glycine site of NMDA receptor complex (Baron et al., 1990). DCKA attenuated NMDA cytotoxicity onto cortical neurons in a concentration-dependent manner (Fig. 2A). Significant neuroprotective effect was observed with DCKA at concentrations of 3 µM or higher, and at 100 µM the drug almost abolished the cytotoxicity of 30 µM NMDA. DCKA was also effective in protecting cortical neurons from ischemic injury. At concentrations of 10 µM or higher, DCKA markedly and significantly increased the number of surviving neurons after ischemic insults in a concentration-dependent manner (Fig. 2B). These results confirm that endogenous glycine site ligand(s) are essential for NMDA receptor-mediated injury. MK-801 (10 µM) and DCKA (30-100 µM) were fully protective even if they were omitted from postincubation solution after NMDA or ischemic treatment (Fig. 2C), indicating that NMDA receptor activation during these insults is crucial for neuronal damage.

View larger version (31K): [in this window] [in a new window]   Fig. 2. DCKA prevents neuronal damage induced by NMDA or simulated ischemia. A, effect of DCKA on NMDA-induced neuronal damage. NMDA (30 µM) was applied to cortical slices for 30 min, which was followed by incubation of slices in the standard solution for 6 h. DCKA at indicated concentrations was applied from 1 h before NMDA treatment and was present during the entire course of incubation. ***, P < 0.001 versus control; ##, P < 0.01; ###, P < 0.001 versus NMDA alone. B, effect of DCKA on ischemia-induced neuronal damage. Slices were incubated in oxygen/glucose-free solution for 20 min and then returned to oxygenated standard solution and incubated for a further 6 h. DCKA at indicated concentrations was applied from 1 h before ischemic treatment and was present during the entire course of incubation. ***, P < 0.01 versus control; ###, P < 0.01 versus ischemia alone. C, MK-801 and DCKA are fully protective against NMDA or ischemic insults even when they are absent during postincubation. MK-801 (MK, 10 µM) and DCKA (100 µM) were applied 1 h before and during 30 min NMDA (30 µM) or 20 min ischemic treatment. Slices were then returned to drug-free standard solution and incubated for a further 6 h. ***, P < 0.001 versus control; ###, P < 0.001.

Several lines of evidence suggest that the glycine site of NMDA receptors is not saturated in in vitro brain slice preparations (Bergeron et al., 1998; Chen et al., 2003). Accordingly, we examined whether exogenous application of glycine site agonists could exacerbate neurotoxic consequences of NMDA and simulated ischemia. As shown in Fig. 3A, addition of glycine (10-30 µM) or D-serine (100-300 µM) did not provide significant influences on the decrease in the number of surviving neurons induced by 30 µM NMDA. In addition, glycine and D-serine showed no significant effect on ischemia-induced neuronal injury (Fig. 3B). Glycine (100 µM) and D-serine (1 mM) alone applied for 30 min did not affect the viability of neurons assessed after 6 h (control, 177.3 ± 3.0 cells/field; glycine, 184.0 ± 11.3 cells/field; D-serine, 182.3 ± 6.0 cells/field). These results suggest that sufficient amount of glycine site ligand(s) is supplied endogenously to allow NMDA receptor overactivation leading to neuronal injury.

View larger version (29K): [in this window] [in a new window]   Fig. 3. Exogenous glycine or D-serine does not exacerbate neuronal damage induced by NMDA or simulated ischemia. A, effects of glycine and D-serine on NMDA-induced neuronal damage. Cortical slices were treated with 30 µM NMDA for 30 min and then incubated in the standard solution for a further 6 h. Glycine or D-serine at indicated concentrations was applied during NMDA treatment. ***, P < 0.001 versus control. B, effects of glycine and D-serine on ischemia-induced neuronal damage. Slices were incubated in oxygen/glucose-free solution for 20 min and then returned to oxygenated standard solution and incubated for a further 6 h. Glycine or D-serine at indicated concentrations was applied during ischemic treatment. ***, P < 0.001 versus control.

DAAOX Inhibits NMDA- and Ischemia-Induced Neuronal Injury. To verify whether D-serine plays an important role in the cerebral cortex as an endogenous glycine site ligand, we used DAAOX. DAAOX is an enzyme that catalyzes oxidative deamination of D-amino acids (D'Aniello et al., 1993). At physiological pH, this enzyme is highly selective for D-serine and does not act on other amino acids, including glycine. Indeed, the specificity of the enzyme has been demonstrated in cerebellar slice preparation, in that DAAOX treatment reduces D-serine levels by 90%, without affecting significantly the levels of other amino acids including glycine, L-serine, and L-glutamate (Mothet et al., 2000). When DAAOX at 1 to 30 U/liter was applied 1 h before and during 30-min treatment of cortical slices with 30 µM NMDA, it showed a marked protective effect against NMDA cytotoxicity in a concentration-dependent manner. At 30 U/liter, DAAOX almost completely inhibited the decrease in the number of surviving neurons induced by NMDA (Fig. 4A).

View larger version (30K): [in this window] [in a new window]   Fig. 4. DAAOX prevents neuronal damage by NMDA, homoquinolinic acid, or simulated ischemia. A, effect of DAAOX on NMDA-induced neuronal damage. Cortical slices were treated with 30 µM NMDA for 30 min and then incubated in the standard solution for a further 6 h. DAAOX at indicated concentrations was applied 1 h before and during NMDA treatment. ***, P < 0.001 versus control; ##, P < 0.01; ###, P < 0.001 versus NMDA alone. B, effects of MK-801 and DAAOX on homoquinolinic acid-induced neuronal damage. Cortical slices were treated with 30 µM homoquinolinic acid for 30 min and then incubated in the standard solution for a further 6 h. MK-801 (10 µM) was applied from 1 h before homoquinolinic acid treatment and was present during the entire course of incubation. DAAOX at indicated concentrations was applied 1 h before and during homoquinolinic acid treatment. ***, P < 0.001 versus control; ###, P < 0.001 versus homoquinolinic acid alone. C, effect of DAAOX on ischemia-induced neuronal damage. Slices were incubated in oxygen/glucose-free solution for 20 min and then returned to oxygenated standard solution and incubated for a further 6 h. DAAOX at indicated concentrations was applied 1 h before and during ischemic treatment. ***, P < 0.01 versus control; #, P < 0.05; ###, P < 0.001 versus ischemia alone.

Although DAAOX is highly selective for neutral D-amino acids and is unlikely to degrade NMDA (Mothet et al., 2000), we further examined the effect of DAAOX against neurotoxicity of homoquinolinic acid, another selective NMDA receptor agonist that does not possess amino acid structure (Patneau and Mayer, 1990). Homoquinolinic acid at 30 µM was cytotoxic to cortical neurons, which was abolished by 10 µM MK-801 (Fig. 4B). Notably, DAAOX afforded a remarkable protective effect against homoquinolinic acid cytotoxicity, being significantly effective at 3 and 10 U/liter. Protective effect of DAAOX was also evident in the case of ischemic neuronal injury. Thus, DAAOX at 1 to 30 U/liter provided concentration-dependent protection against neuronal damage induced by 20 min of simulated ischemia (Fig. 4C).

We also performed Hoechst 33342 staining (Pizzi et al., 2000) to verify the neuroprotective effect of DAAOX against NMDA- and ischemia-induced injury. Nuclear condensation and fragmentation were clearly observed in slices treated with 30 µM NMDA. Changes in nuclear morphology such as shrinkage and distorted appearance were also evident in slices that received 20 min of simulated ischemia. These deleterious effects were markedly prevented by application of 30 U/liter DAAOX before and during NMDA or ischemic treatment (Fig. 5).

View larger version (84K): [in this window] [in a new window]   Fig. 5. Changes in nuclear morphology induced by NMDA or simulated ischemia, and effect of DAAOX. Shown are representative photomicrographs of Hoechst 33342 staining on a slice incubated under control conditions (control), a slice treated with 30 µM NMDA (NMDA), a slice treated with 30 µM NMDA + 30 U/l DAAOX (NMDA + DAAOX), a slice that received ischemic treatment for 20 min (ischemia), and a slice that received ischemic treatment for 20 min in the presence of 30 U/l DAAOX (ischemia + DAAOX). Nuclear morphology was assessed after 6 h of postincubation in the standard buffer. In NMDA-treated slices, nuclear condensation (arrows) and fragmentation (arrowheads) were prominent. In slices that received ischemic treatment, shrinkage (arrows) and distorted morphology (arrowhead) of nuclei were typical. These changes were markedly inhibited by DAAOX.

Specificity of the effect of DAAOX was examined further. The protective effect of 10 U/liter DAAOX against NMDA neurotoxicity was reversed by exogenous application of 1 mM D-serine or 100 µM glycine (Fig. 6A), which strongly suggest that the effect of DAAOX is attributable to the removal of endogenous D-serine by its enzymatic activity. Moreover, DAAOX protein at equivalent amount with 10 U/liter, when heat-inactivated before application, showed no significant effect on NMDA neurotoxicity (Fig. 6A). We obtained similar results when we did these examinations onto the protective effect of DAAOX against ischemic neuronal injury. That is, the effect of DAAOX was abolished either by concomitant application of excess D-serine and glycine or by prior heat inactivation (Fig. 6B). Moreover, the protective effect of DAAOX against NMDA and ischemic insults was not significantly affected by L-serine (Fig. 7).

View larger version (34K): [in this window] [in a new window]   Fig. 6. Protective effect of DAAOX is attributable to its enzymatic activity and reversed by excess D-serine or glycine. A, effects of D-serine and glycine on the protective effect of DAAOX, and effect of heat-inactivated DAAOX on NMDA-induced neuronal damage. Cortical slices were treated with 30 µM NMDA for 30 min and then incubated in the standard solution for a further 6 h. DAAOX (10 U/l) or the same amount of heat-inactivated DAAOX was applied 1 h before and during NMDA treatment. Glycine or D-serine at indicated concentrations was applied during NMDA treatment. ***, P < 0.001 versus control; #, P < 0.05; ###, P < 0.001. B, effects of D-serine and glycine on the protective effect of DAAOX and effect of heat-inactivated DAAOX on ischemia-induced neuronal damage. Slices were incubated in oxygen/glucose-free solution for 20 min and then returned to oxygenated standard solution and incubated for a further 6 h. DAAOX (10 U/l) or the same amount of heat-inactivated DAAOX was applied 1 h before and during ischemic treatment. Glycine or D-serine at indicated concentrations was applied during ischemic treatment. ***, P < 0.001 versus control; ###, P < 0.001.

View larger version (36K): [in this window] [in a new window]   Fig. 7. Protective effect of DAAOX was not affected by L-serine or AIB. A, effects of L-serine and AIB on the protective effect of DAAOX on NMDA-induced neuronal damage. Cortical slices were treated with 30 µM NMDA for 30 min and then incubated in the standard solution for a further 6 h. DAAOX (10 U/l) was applied 1 h before and during NMDA treatment. D-Serine (1 mM), L-serine, and AIB at indicated concentrations were applied during NMDA treatment. ***, P < 0.001 versus control; #, P < 0.05; ##, P < 0.01. B, effects of L-serine and AIB on the protective effect of DAAOX on ischemia-induced neuronal damage. Slices were incubated in oxygen/glucose-free solution for 20 min and then returned to oxygenated standard solution and incubated for a further 6 h. DAAOX (10 U/l) was applied 1 h before and during ischemic treatment. D-Serine (1 mM), L-serine, and AIB at indicated concentrations were applied during ischemic treatment. ***, P < 0.001 versus control; ###, P < 0.001.

A recent report has shown that D-serine is a substrate of a neutral amino acid transporter asc-1 expressed mainly on presynaptic nerve terminals in rodent brain (Helboe et al., 2003). To determine whether inhibition of asc-1-mediated amino acid transport by D-serine contributed to the observed effect of D-serine, we examined the effect of AIB, a well known inhibitor of system A/L transport that was shown to inhibit asc-1 (Helboe et al., 2003). AIB neither reversed the protective effect of DAAOX, nor affected the effect of D-serine (Fig. 7).

Blockade of non-NMDA Receptors Does Not Affect Sensitivity of NMDA- and Ischemia-Induced Neuronal Injury to a Glycine Site Antagonist. Application of non-NMDA receptor agonists has been shown to stimulate efflux of D-serine from cultured astrocytes (Schell et al., 1995; Ribeiro et al., 2002). If this regulatory mechanism of D-serine efflux makes significant contributions to D-serine supply during excitotoxic or ischemic insults, blockade of non-NMDA receptors is expected to decrease the amount of available D-serine and therefore provide neuroprotective effects by compromising NMDA receptor overactivation. According to these assumptions, we examined the influences of a non-NMDA receptor antagonist NBQX on NMDA- and ischemia-induced neuronal damage. NBQX, applied to cerebrocortical slices at 10 µM during the entire course of the incubation periods, did not show significant effect on neuronal damage induced by 30 µM NMDA. In addition, the concentration dependence of the neuroprotective effect of DCKA against NMDA cytotoxicity was not apparently affected by NBQX (Fig. 8A; compare with Fig. 2A). Similar results were obtained in the case of ischemic neuronal injury (Fig. 8B), namely, neuronal injury induced by 20 min of simulated ischemia was not affected by 30 µM NBQX. Moreover, concentration-dependent protective effect of DCKA against ischemic neuronal damage was not altered by NBQX (compare Fig. 8B with Fig. 2B), indicating that non-NMDA receptor-stimulated D-serine efflux does not contribute significantly to the outcome of ischemic neuronal injury under the present experimental conditions.

View larger version (30K): [in this window] [in a new window]   Fig. 8. Blockade of AMPA/kainate receptors does not affect sensitivity of NMDA- and ischemia-induced neuronal damage to a competitive glycine site antagonist. A, effect of DCKA on NMDA-induced neuronal damage in the presence of NBQX. Cortical slices were treated with 30 µM NMDA for 30 min and then incubated in the standard solution for a further 6 h. NBQX (30 µM) and DCKA at indicated concentrations were applied from 1 h before NMDA treatment and were present during the entire course of incubation. ***, P < 0.001 versus control; ##, P < 0.01; ###, P < 0.001 versus NMDA alone. B, effect of DCKA on ischemia-induced neuronal damage in the presence of NBQX. Slices were incubated in oxygen/glucose-free solution for 20 min and then returned to oxygenated standard solution and incubated for a further 6 h. NBQX (30 µM) and DCKA at indicated concentrations were applied from 1 h before ischemic treatment and were present during the entire course of incubation. ***, P < 0.001 versus control; ##, P < 0.01; ###, P < 0.001 versus ischemia alone.

	Discussion

Top Abstract Materials and Methods Results Discussion References

 
Neuronal damage resulting from overactivation of NMDA receptors contributes to various pathological processes in the central nervous system, including acute disorders such as ischemia and trauma as well as slowly progressing neurodegenerative diseases (Hardingham and Bading, 2003). In the present study, we used cerebrocortical slice preparation maintained in a defined salt solution, to verify unambiguously whether endogenous glycine site ligand, particularly D-serine, plays a key role in determining excitotoxic consequences in the cerebral cortex. DCKA, a potent glycine site antagonist, showed a remarkable protective effect against neuronal injury induced by NMDA or simulated ischemia. On the other hand, addition of D-serine or glycine did not potentiate cytotoxic effects of NMDA and ischemia. These results suggest that endogenous supply of glycine site agonist(s) is both necessary and sufficient for overactivation of NMDA receptors under pathological conditions. In contrast to our findings, a recent electrophysiological study on brain slice preparation suggested that glycine site of NMDA receptors on rat prefrontal cortical neurons was not saturated, because glycine site agonists potentiated synaptically evoked NMDA excitatory postsynaptic currents (Chen et al., 2003). A plausible explanation for the apparent discrepancy is that, unlike in the case of synaptic transmission under physiological conditions, partial occupancy of the glycine site may be sufficient for NMDA receptor overactivation leading to neuronal injury, under pathological conditions where a large amount of glutamate-binding site ligands are present. It is also possible that under pathological conditions such as ischemia, extracellular concentrations of glycine site ligand(s) are elevated, reaching levels sufficient for saturating the glycine site of NMDA receptors (Lo et al., 1998; see below).

Critical roles of glycine site stimulation in induction of NMDA receptor-mediated neuronal damage have been repeatedly demonstrated in several brain regions such as the hippocampus (Foster et al., 1990; Newell et al., 1995) and the striatum (Foster et al., 1990). What we revealed here beyond these previous studies is that D-serine plays an important role as an endogenous glycine site agonist in determining excitotoxic consequences in the cerebral cortex. Potential endogenous agonists at the glycine site of NMDA receptors are D-serine and glycine, and at present, DAAOX is an only available tool to discriminate which compound is actually working physiologically and pathophysiologically. DAAOX is highly selective for neutral D-amino acids (D'Aniello et al., 1993) and is shown to deplete D-serine in cerebellar slices without affecting the levels of other amino acids, including glycine (Mothet et al., 2000). NMDA receptor-mediated synaptic transmission is significantly attenuated by DAAOX in hippocampal slices and immature cerebellar slices (Mothet et al., 2000) as well as in the isolated retina (Stevens et al., 2003), indicating that endogenous D-serine is crucial for supporting synaptic transmission under physiological conditions. Using DAAOX, the present study demonstrated that endogenous D-serine is also important for pathological processes mediated by NMDA receptor overactivation in the cerebral cortex. This conclusion is supported by the findings that neuronal damage induced by NMDA, homoquinolinic acid, or simulated ischemia was markedly inhibited by application of DAAOX to cortical slices. The neuroprotective effect of DAAOX was abolished by prior heat inactivation and also was reversed by exogenous addition of excess D-serine or glycine, confirming that the enzymatic degradation of D-serine by DAAOX is responsible for the observed neuroprotective effects. The fact that AIB did not influence the protective effect of DAAOX suggests that the reversal of neuroprotection by excess D-serine was indeed mediated by glycine site stimulation rather than blockade of neutral amino acid transport (Helboe et al., 2003). The reversal of the effect of DAAOX by glycine is also most likely due to stimulation of glycine site of NMDA receptors, although possible involvement of other actions of glycine, including activation of inhibitory glycine receptors or influences on amino acid transport, cannot be excluded completely. A previous in vivo microdialysis study demonstrating that extracellular concentration of D-serine in the frontal cortex may be high enough to saturate the glycine site of NMDA receptors (Matsui et al., 1995) is also consistent with the critical role of D-serine in this brain structure.

In contrast to D-serine, L-serine showed little effect on the protection afforded by DAAOX against NMDA and ischemic insults. This is not surprising because L-serine itself is a poor agonist at the glycine site of NMDA receptors (McBain et al., 1989). Moreover, although L-serine may be converted to D-serine by serine racemase (Wolosker et al., 1999) after its cellular uptake, or may stimulate D-serine release by direct exchange through neutral amino acid transporters (Ribeiro et al., 2002), a short period of L-serine application is unlikely to elevate extracellular D-serine levels sufficient to overwhelm the rate of degradation of D-serine by DAAOX.

Although the above-mentioned results strongly suggest that D-serine acts as an endogenous glycine site ligand in the cerebral cortex, we do not entirely exclude the possible contribution of glycine as another endogenous ligand, because we have not examined the consequences of removing endogenous glycine. Utilization of glycine-degrading enzymes such as glycine oxidase (Nishiya and Imanaka, 1998) may provide useful information concerning this issue.

D-Serine immunoreactivity is concentrated in astrocytes (Schell et al., 1995, 1997; Wolosker et al., 1999). In hippocampal CA1 region, the distribution of D-serine immunoreactivity roughly parallels that of NR2A/B immunoreactivity (Schell et al., 1997), supporting the view that the glycine site of NMDA receptors is occupied primarily by glia-derived D-serine. Interestingly, efflux of D-serine from cultured astrocytes is stimulated by agonists of non-NMDA subtypes of glutamate receptors (Schell et al., 1995; Ribeiro et al., 2000). In the present study, however, blockade of non-NMDA receptors by NBQX neither affected NMDA- and ischemia-induced injury, nor altered the concentration dependence of the protective effect of DCKA. These results suggest that non-NMDA receptor-mediated enhancement of D-serine efflux, even if it occurs, has little impact on neuronal injury under the present experimental conditions. Our results do not necessarily mean that concentrations of D-serine show no changes during excitotoxic or ischemic insults, because subtle changes in extracellular D-serine levels may have been overlooked by estimation relying on the effect of DCKA. In addition, mechanisms other than those mediated by non-NMDA receptors may also regulate the rate of D-serine efflux from astrocytes. Indeed, extracellular concentrations of D-serine in rabbit cerebral cortex have been shown to increase during transient forebrain ischemia in vivo (Lo et al., 1998). Although the pathways of D-serine efflux across plasma membranes remain to be established, ASCT-like neutral amino acid transporters seem to play an important role (Hayashi et al., 1997; Ribeiro et al., 2002). Transporter-mediated exchange with other neutral amino acids may be among important mechanisms contributing to the efflux of D-serine because L-serine and L-threonine are potent inducers of D-serine efflux from cultured astrocytes, being more effective than kainate (Ribeiro et al., 2002). Also noteworthy are the findings of Hashimoto et al. (2000), who demonstrated that intrastriatal perfusion of NMDA or kainate caused a significant decrease, rather than an increase, in extracellular concentrations of D-serine. Clearly, there remains much to be done for elucidation of the mechanisms regulating extracellular levels of D-serine.

In conclusion, we provided evidence that D-serine acts as an endogenous ligand at the glycine site of NMDA receptors in the cerebral cortex. The present study also demonstrated for the first time that reduction in extracellular D-serine levels affords remarkable protective effect against NMDA receptor-mediated neurotoxic insults. Hence, controlling brain D-serine levels may constitute a novel strategy for neuroprotection against various neuropathological and neurodegenerative conditions (Snyder and Ferris, 2000; Wolosker et al., 2002).

	Footnotes

 
This study was supported in part by a grant-in-aid for scientific research from The Ministry of Education, Culture, Sports, Science and Technology, Japan, and from the Japan Society for the Promotion of Science. H.S. is supported as a teaching assistant by the 21st Century Center of Excellence Program "Knowledge Information Infrastructure for Genome Science".

doi:10.1124/jpet.104.070912.

ABBREVIATIONS: NMDA, N-methyl-D-aspartate; DAAOX, D-amino acid oxidase; MK-801, (-)-5-methyl-10,11-dihydro-5H-dibenzo[a,d]cyclohepten-5,10-imine maleate; NBQX, 2,3-dihydroxy-6-nitro-7-sulfamoyl-benzo[f]quinoxaline; DCKA, 2,7-dichlorokynurenic acid; AIB, -aminoisobutyric acid; PBS, phosphate-buffered saline.

Address correspondence to: Dr. Akinori Akaike, Department of Pharmacology, Graduate School of Pharmaceutical Sciences, Kyoto University, 46-29 Yoshida-shimoadachi-cho, Sakyo-ku, Kyoto 606-8501, Japan. E-mail: aakaike{at}pharm.kyoto-u.ac.jp

	References

Top Abstract Materials and Methods Results Discussion References

 

Baron BM, Harrison BL, Miller FP, McDonald IA, Salituro FG, Schimidt CJ, Sorensen SM, White HS, and Palfreyman MG (1990) Activity of 5,7-dichlorokynurenic acid, a potent antagonist at the N-methyl-D-aspartate receptor-associated glycine binding site. Mol Pharmacol 38: 554-561.[Abstract]

Bergeron R, Meyer TM, Coyle JT, and Greene RW (1998) Modulation of N-methyl-D-aspartate receptor function by glycine transport. Proc Natl Acad Sci USA 95: 15730-15734.[Abstract/Free Full Text]

Chen L, Muhlhauser M, and Yang CR (2003) Glycine transporter-1 blockade potentiates NMDA-mediated responses in rat prefrontal cortical neurons in vitro and in vivo. J Neurophysiol 89: 691-703.[Abstract/Free Full Text]

Contestabile A (2000) Roles of NMDA receptor activity and nitric oxide production in brain development. Brain Res Rev 32: 476-509.[CrossRef][Medline]

D'Aniello A, Vetere A, and Petrucelli L (1993) Further study on the specificity of D-amino acid oxidase and D-aspartate oxidase and time course for complete oxidation of D-amino acids. Comp Biochem Physiol B 105: 731-734.[CrossRef][Medline]

Foster AC, Willis CL, and Tridgett R (1990) Protection against N-methyl-D-aspartate receptor-mediated neuronal degeneration in rat brain by 7-chlorokynurenate and 3-amino-1-hydroxypyrrolid-2-one, antagonists at the allosteric site for glycine. Eur J Neurosci 2: 270-277.[CrossRef][Medline]

Hardingham GE and Bading H (2003) The Yin and Yang of NMDA receptor signalling. Trends Neurosci 26: 81-89.[CrossRef][Medline]

Hashimoto A, Kanda J, and Oka T (2000) Effects of N-methyl-D-aspartate, kainate or veratridine on extracellular concentrations of free D-serine and L-glutamate in rat striatum: an in vivo microdialysis study. Brain Res Bull 53: 347-351.[CrossRef][Medline]

Hashimoto A, Nishikawa T, Hayashi T, Fujii N, Harada K, Oka T, and Takahashi K (1992) The presence of free D-serine in rat brain. FEBS Lett 296: 33-36.[CrossRef][Medline]

Hashimoto A, Nishikawa T, Oka T, and Takahashi K (1993) Endogenous D-serine in rat brain: N-methyl-D-aspartate receptor-related distribution and aging. J Neurochem 60: 783-786.[Medline]

Hashimoto A and Oka T (1997) Free D-aspartate and D-serine in the mammalian brain and periphery. Prog Neurobiol 52: 325-353.[CrossRef][Medline]

Hashimoto A, Oka T, and Nishikawa T (1995) Anatomical distribution and postnatal changes in endogenous free D-aspartate and D-serine in rat brain and periphery. Eur J Neurosci 7: 1657-1663.[CrossRef][Medline]

Hayashi F, Takahashi K, and Nishikawa T (1997) Uptake of D- and L-serine in C6 glioma cells. Neurosci Lett 239: 85-88.[CrossRef][Medline]

Helboe L, Egebjerg J, Moller M, and Thomsen C (2003) Distribution and pharmacology of alanine-serine-cysteine transporter 1 (asc-1) in rodent brain. Eur J Neurosci 18: 2227-2238.[CrossRef][Medline]

Izumi Y, Benz AM, Kurby CO, Labruyere J, Zorumski CF, Price MT, and Olney JW (1995) An ex vivo rat retinal preparation for excitotoxicity studies. J Neurosci Methods 60: 219-225.[CrossRef][Medline]

Izumi Y, Clifford DB, and Zorumski CF (1990) Glycine antagonists block the induction of long-term potentiation in CA1 of rat hippocampal slices. Neurosci Lett 112: 251-256.[CrossRef][Medline]

Johnson JW and Ascher P (1987) Glycine potentiates the NMDA response in cultured mouse brain neurones. Nature (Lond) 325: 529-531.[CrossRef][Medline]

Kleckner NW and Dingledine R (1988) Requirement for glycine in activation of NMDA-receptors in Xenopus oocytes. Science (Wash DC) 241: 835-837.[Abstract/Free Full Text]

Lo EH, Pierce AR, Matsumoto K, Kano T, Evans CJ, and Newcomb R (1998) Alterations in K⁺ evoked profiles of neurotransmitter and neuromodulator amino acids after focal ischemia-reperfusion. Neuroscience 83: 449-458.[Medline]

Matsui T, Sekiguchi M, Hashimoto A, Tomita U, Nishikawa T, and Wada K (1995) Functional comparison of D-serine and glycine in rodents: the effect on cloned NMDA receptors and the extracellular concentrations. J Neurochem 65: 454-458.[Medline]

McBain CJ, Kleckner NW, Wyrick S, and Dingledine R (1989) Structural requirements for activation of the glycine coagonist site of N-methyl-D-aspartate receptors expressed in Xenopus oocytes. Mol Pharmacol 36: 556-565.[Abstract]

McNamara D and Dingledine R (1990) Dual effect of glycine on NMDA-induced neurotoxicity in rat cortical cultures. J Neurosci 10: 3970-3976.[Abstract]

Mothet J-P, Parent AT, Wolosker H, Brady RO Jr, Linden DJ, Ferris CD, Rogawski MA, and Snyder SH (2000) D-Serine is an endogenous ligand for the glycine site of the N-methyl-D-aspartate receptor. Proc Natl Acad Sci USA 97: 4926-4931.[Abstract/Free Full Text]

Newell DW, Barth A, and Malouf AT (1995) Glycine site NMDA receptor antagonists provide protection against ischemia-induced neuronal damage in hippocampal slice cultures. Brain Res 675: 38-44.[CrossRef][Medline]

Nishiya Y and Imanaka T (1998) Purification and characterization of a novel glycine oxidase from Bacillus subtilis. FEBS Lett 438: 263-266.[CrossRef][Medline]

Patel J, Zinkland WC, Thompson C, Keith R, and Salama A (1990) Role of glycine in the N-methyl-D-aspartate-mediated neuronal cytotoxicity. J Neurochem 54: 849-854.[CrossRef][Medline]

Patneau DK and Mayer ML (1990) Structure-activity relationships for amino acid transmitter candidates acting at N-methyl-D-aspartate and quisqualate receptors. J Neurosci 10: 2385-2399.[Abstract]

Pizzi M, Benarese M, Boroni F, Goffi F, Valerio A, and Spano PF (2000) Neuroprotection by metabotropic glutamate receptor agonists on kainate-induced degeneration of motor neurons in spinal cord slices from adult rat. Neuropharmacology 39: 903-910.[CrossRef][Medline]

Ribeiro CS, Reis M, Panizzutti R, de Miranda J, and Wolosker H (2002) Glial transport of the neuromodulator D-serine. Brain Res 929: 202-209.[CrossRef][Medline]

Schell MJ, Brady RO Jr, Molliver ME, and Snyder SH (1997) D-Serine as a neuromodulator: regional and developmental localizations in rat brain glia resemble NMDA receptors. J Neurosci 17: 1604-1615.[Abstract/Free Full Text]

Schell MJ, Molliver ME, and Snyder SH (1995) D-Serine, an endogenous synaptic modulator: localization to astrocytes and glutamate-stimulated release. Proc Natl Acad Sci USA 92: 3948-3952.[Abstract/Free Full Text]

Sheng M, Cummings J, Roldan LA, Jan YN, and Jan LY (1994) Changing subunit composition of heteromeric NMDA receptors during development of rat cortex. Nature (Lond) 368: 144-147.[CrossRef][Medline]

Silva AJ (2003) Molecular and cellular cognitive studies of the role of synaptic plasticity in memory. J Neurobiol 54: 224-237.[CrossRef][Medline]

Snyder SH and Ferris CD (2000) Novel neurotransmitters and their neuropsychiatric relevance. Am J Psychiatry 157: 1738-1751.[Abstract/Free Full Text]

Stevens ER, Esguerra M, Kim PM, Newman EA, Snyder SH, Zahs KR, and Miller RF (2003) D-Serine and serine racemase are present in the vertebrate retina and contribute to the physiological activation of NMDA receptors. Proc Natl Acad Sci USA 100: 6789-6794.[Abstract/Free Full Text]

Watanabe Y, Himi T, Saito H, and Abe K (1992) Involvement of glycine site associated with the NMDA receptor in hippocampal long-term potentiation and acquisition of spatial memory in rats. Brain Res 582: 58-64.[CrossRef][Medline]

Wolosker H, Blackshaw S, and Snyder SH (1999) Serine racemase: a glial enzyme synthesizing D-serine to regulate glutamate-N-methyl-D-aspartate neurotransmission. Proc Natl Acad Sci USA 96: 13409-13414.[Abstract/Free Full Text]

Wolosker H, Panizzutti R, and de Miranda J (2002) Neurobiology through the looking-glass: D-serine as a new glial-derived transmitter. Neurochem Int 41: 327-332.[CrossRef][Medline]

Yang Y, Ge W, Chen Y, Zhang Z, Shen W, Wu C, Poo M, and Duan S (2003) Contribution of astrocytes to hippocampal long-term potentiation through release of D-serine. Proc Natl Acad Sci USA 100: 15194-15199.[Abstract/Free Full Text]

Zilles K and Wree A (1995) Cortex: areal and laminar structure, in The Rat Nervous System, 2nd ed. (Paxinos G ed) pp 649-685, Academic Press, San Diego.

This article has been cited by other articles:

				Z. Zhang, N. Gong, W. Wang, L. Xu, and T.-L. Xu Bell-Shaped D-Serine Actions on Hippocampal Long-Term Depression and Spatial Memory Retrieval Cereb Cortex, February 14, 2008; (2008) bhn008v1. [Abstract] [Full Text] [PDF]

				M. Shleper, E. Kartvelishvily, and H. Wolosker D-Serine Is the Dominant Endogenous Coagonist for NMDA Receptor Neurotoxicity in Organotypic Hippocampal Slices J. Neurosci., October 12, 2005; 25(41): 9413 - 9417. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: jpet.104.070912v1 311/2/836    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Katsuki, H. Articles by Akaike, A. Search for Related Content PubMed PubMed Citation Articles by Katsuki, H. Articles by Akaike, A.