www.elsevier.com/locate/yabio

ANALYTICAL

BIOCHEMISTRY

Analytical Biochemistry 362 (2007) 83–88

Simultaneous measurement of D-serine dehydratase and D-aminoacid oxidase activities by the detection of 2-oxo-acid formationwith reverse-phase high-performance liquid chromatography q

Hiroyuki Tanaka *, Atsushi Yamamoto, Tetsuo Ishida, Kihachiro Horiike

Department of Biochemistry and Molecular Biology, Shiga University of Medical Science, Seta, Ohtsu, Shiga 520-2192, Japan

Received 12 September 2006Available online 20 December 2006

Abstract

N-methyl-D-aspartate receptors (NMDARs) play critical roles in excitatory synaptic transmission in the vertebrate central nervoussystem. NMDARs need D-serine for their channel activities in various brain regions. In mammalian brains, D-serine is produced fromL-serine by serine racemase and degraded by D-amino acid oxidase (DAO) to 3-hydroxypyruvate. In avian organs, such as the kidney,in addition to DAO, D-serine is also degraded to pyruvate by D-serine dehydratase (DSD). To examine the roles of these two enzymes inavian brains, we developed a method to simultaneously measure DAO and DSD activities. First, the keto acids produced from D-serinewere derivatized with 3-methyl-2-benzothiazolinone hydrazone to stable azines. Second, the azine derivatives were quantified by meansof reverse-phase high-performance liquid chromatography using 2-oxoglutarate as an internal standard. This method allowed the simul-taneous detection of DAO and DSD activities as low as 100 pmol/min/mg protein. Chicken brain showed only DSD activities(0.4 ± 0.2 nmol/min/mg protein) whereas rat brain exhibited only DAO activities (0.7 ± 0.1 nmol/min/mg protein). This result stronglysuggests that DSD plays the same role in avian brains, as DAO plays in mammalian brains. The present method is applicable to otherketo acids producing enzymes with minor modifications.� 2006 Elsevier Inc. All rights reserved.

Keywords: D-Amino acid oxidase; 3-Hydroxypyruvate; 3-Methyl-2-benzothiazolinone hydrazone; D-Serine; D-Serine dehydratase; HPLC

N-Methyl-D-aspartate receptors (NMDARs)1 are ubiq-uitously found in vertebrate brains. NMDARs play criticalroles in excitatory synaptic transmission, involving aspectssuch as learning and memory [1–3 and references cited

0003-2697/$ - see front matter � 2006 Elsevier Inc. All rights reserved.

doi:10.1016/j.ab.2006.12.025

q This study was supported by a Grant-in Aid for Scientific Research (C)(No. 18590528) (to T. I. and H. T.) from the Ministry of Education,Culture, Sports, Sciences and Technology of Japan, by Grants-in-Aid(Heisei era 13, 14, and 18) from Shiga University of Medical Science, andby a Grant-in-Aid (Heisei era 18) from The Shiga Medical ScienceAssociation for International Cooperation.

* Corresponding author. Fax: +81 77 548 2157.E-mail address: tanakah@belle.shiga-med.ac.jp (H. Tanaka).

1 Abbreviations used: NMDARs, N-methyl-D-aspartate receptors; DAO,D-amino acid oxidase; DSD, D-serine dehydratase; MBTH, 3-methyl-2-benzothiazolinone hydrazone; TCA, trichloroacetic acid; TFA, trifluoro-acetic acid.

therein]. Glutamate, the neurotransmitter for the receptor,cannot activate NMDAR in the absence of glycine and/orD-serine [4]. In mammalian brains, glial cells produce D-ser-ine from L-serine by serine racemase [5–8] and degrade it byD-amino acid oxidase (DAO; EC 1.4.3.3) [9]. The regionaldistribution of D-serine in the brain is the same as that ofserine racemase [8] and shows an inverse correlation withthat of DAO [9–11]. However, it is not known how theregional expression of these enzymes is controlled to mod-ulate NMDAR activities.

Unlike mammals, birds have another D-serine-degradingenzyme, D-serine dehydratase (DSD; EC. 4.3.1.18), togeth-er with DAO [12]. As shown in Fig. 1, DSD catalyzes thedehydration of D-serine to pyruvate, whereas DAO catalyz-es the oxidative deamination of D-serine to 3-hydroxypyru-vate. D-Serine exists in significant levels in both avian and

Fig. 1. Reactions catalyzed by D-serine dehydratase and D-amino acidoxidase. The primary enzymatic products undergo rapid nonenzymaticreaction with water to give 2-keto acids and ammonium ions.

84 Assay for D-serine degradation / H. Tanaka et al. / Anal. Biochem. 362 (2007) 83–88

mammalian brains [10]. The distribution of DAO and DSDactivities in avian brains remains to be revealed. Becausemammalian brains contain only DAO for the degradationof D-serine, a comparative study of avian and mammalianbrains is important to better elucidate the region-specificmodulation of NMDAR activities by D-serine in vertebratebrains.

In the present study, we developed a method to simulta-neously determine both DSD and DAO activities bymeasuring the enzymatic formation of pyruvate and 3-hydroxypyruvate from D-serine. Spectrophotometricdetermination of aldehydes and keto acids with 3-methyl-2-benzothiazolinone hydrazone (MBTH) has been wellestablished [13,14]. The chromatographic behavior ofMBTH–azines of various aldehydes has also been examinedin detail by Chiavari and Facchini [15]. On the basis of theseprevious studies, we first developed a high-performanceliquid chromatographic quantification method for ketoacids with precolumn MBTH derivatization. Using thismethod, we found that chicken brain contains only DSDactivities. Because mammalian brains contain only DAOactivities, this result has important evolutionary implica-tions for the degradation of D-serine in vertebrate brains.

Materials and methods

Materials

Sodium pyruvate and 2-oxoglutaric acid were purchasedfrom Nacalai Tesque (Kyoto, Japan). 3-Hydroxypyruvic

acid was from Sigma. D-Serine, D-alanine, and hydroxyl-amine were from Wako Pure Chemical Industries (Osaka,Japan). 3-Methyl-2-benzothiazolinone hydrazone hydro-chloride was from Aldrich. Catalase was obtained fromBoehringer Mannheim. All other chemicals were of analyt-ical grade.

Male rats (Sprague–Dawley, 6 weeks old, 200–250 g)were obtained from Japan Clea (Tokyo, Japan). Chickenorgans (male White Leghorn, Gallus domesticus) were froma local slaughter house and kept at about �35 �C until use.

Preparation of organ homogenates

Rats were anesthetized with an intraperitoneal injectionof sodium pentobarbital (50 mg/kg body weight) and per-fused with 10 mM sodium phosphate buffer (pH 7.4) con-taining 0.9% NaCl [9,10]. Then, the organs were removedand washed with buffer. Each organ was homogenized onice with 9 volumes of 20 mM Tris–HCl (pH 7.6) using aglass homogenizer. All these procedures were approvedby the animal care committee of our university. The chick-en organs were homogenized using the same method asdescribed above.

Assay for DSD and DAO

Reactions were performed in 1.5-mL polypropylenetubes with a final assay volume of 200 lL. Reactions werecarried out at 37 �C in 50 mM Tris–HCl, pH 7.6. A typicalreaction mixture contained 50 mM D-serine and 20 lg cat-alase. The reaction was initiated by the addition of 2–10 lLof the organ homogenates and stopped by the addition of40 lL of 12.5% trichloroacetic acid (TCA) after 0–60 minincubation. We added 10 lL of 1 mM 2-oxoglutaric acid(10 nmol) to each TCA-stopped reaction mixture as aninternal standard. Then, the solutions were centrifuged at5700g for 5 min. Aliquots of the supernatants (100 lLeach) were transferred into 1.5-mL polypropylene tubesand treated with MBTH as follows according to themethod described by Soda [14] with slight modifications.The supernatants were mixed with 200 lL of 1 M sodiumacetate buffer (pH 5.0) and 80 lL of 0.1% MBTH (aqueoussolution). The mixtures were incubated at 50 �C for 30 minand then allowed to cool to room temperature. Then 5 lLof each aliquot of the MBTH-derivatized samples wasapplied to a Cosmosil 3C18 column (4.6 · 100 mm; NacalaiTesque). The column was preequilibrated with 20% aceto-nitrile in water, which contained 0.1% trifluoroacetic acid(TFA). The elution of the azines was carried out at 25 �Cusing a linear gradient of acetonitrile (from 20 to 90% over8 min) in the presence of 0.1% TFA. The flow rate was1.2 mL/min, and the absorbance at 350 nm was continu-ously monitored. A Tosoh HPLC system (Tokyo, Japan),incorporating DP-8020 pumps, a PX-8020 controller, anda Shimadzu SPD-10Avp detector, was used in this study.The peaks were integrated using a Smart Chrom data pro-cessor (Kya Tech Corp., Tokyo, Japan).

ig. 3. (A) HPLC separation of 3-hydroxypyruvate (peaks 1 and 1 0),-oxoglutarate (peak 2), and pyruvate (peaks 3 and 3 0) as MBTH–azines.

reaction mixture that contained authentic 2-oxo acids (10 nmol each)as subjected to the MBTH treatment (see Materials and methods). AlL aliquot of the final solution (which contained about 60 pmol eacherivative) was applied to a Cosmosil 3C18 column (4.6 · 100 mm).B) Kinetics of the isomerization of the peak 1 fraction in 52% acetonitrile,.1% TFA (E-isomer of MBTH–3-hydroxypyruvate). A freshly preparedeak 1 fraction was incubated at room temperature, and at indicated timesliquots of the solution were subjected to HPLC analysis. The ratio of theeak 1 area (A1) to the sum of the peak 1 and 1 0 areas (A1 + A1 0), the-isomer fraction (fE(t) in Eq. (2)), was plotted against the incubation

ime. The curve is the best fit to Eq. (2), (k1 = 0.14 h�1, k�1 = 0.25 h�1).C) Kinetics of the isomerization of the peak 3 fraction in 66% acetonitrile,.1% TFA (E-isomer of MBTH–pyruvate). The curve is the best fit to Eq.2) (k1 = 0.092 h�1, k�1 = 0.38 h�1).

Assay for D-serine degradation / H. Tanaka et al. / Anal. Biochem. 362 (2007) 83–88 85

To examine the O2 dependence of the enzymatic reac-tions, the reaction mixtures were argon-saturated in a glassvial sealed with a rubber septum and an aluminum cap, asdescribed previously [16].

Analysis of MBTH–keto acids isomerization

As shown in Fig. 2, both E- and Z-isomers are formedfrom 2-keto acid by reaction with MBTH. The two isomersundergo slow interconversion,

E-isomer�! �k1

k�1

Z-isomer; ð1Þ

where k1 and k�1 are the reaction rate constants. The frac-tion of E-isomer at time t, fE(t), is defined as the ratio of themolar concentration of E-isomer to the total concentrationof the MBTH–azine at time t. Then, fE(t) is given accordingto the following equations.

fEðtÞ ¼ aþ ðfEð0Þ � aÞe�ðk1þk�1Þt: ð2Þa ¼ k�1=ðk1 þ k�1Þ: ð3Þ

First, we collected the E-isomers of MBTH–3-hydroxy-pyruvate (peak 1 in Fig. 3) and MBTH–pyruvate (peak 3in Fig. 3). Second, after t h incubation at room tempera-ture, an aliquot of the collected fraction was analyzed byHPLC, as described above, to determine the relativeamounts of peaks 1 and 1 0 or peaks 3 and 3 0 at time t.The experimentally determined fE(t) was fitted to Eq. (2and 3) by nonlinear least squares method using a programmade in-house.

Other analytical methods

The protein concentrations of organ homogenates weredetermined using a bicinchoninic acid protein assay kit(Pierce) and bovine serum albumin as a standard. Spectro-scopic experiments were performed with a Shimadzu UV2200 spectrometer. Mass spectra of the azine derivativesof keto acids were obtained with a Voyager-DE RP massspectrometer (Applied Biosystems).

Fig. 2. Reaction of 3-methyl-2-benzothiazolinone hydrazone (MBTH)with 3-hydroxypyruvate (R = CH2OH) and pyruvate (R = CH3). Twoazine isomers are formed by the present MBTH derivatization.

F2Aw5d(0papE

t(0(

Results

Fig. 2 shows the reaction scheme of 2-oxo acid withMBTH. We initially tested the MBTH derivatization con-ditions for 2-oxo acids. As judged by the change in absorp-tion at 350 nm during the derivatization of authenticsamples (pyruvate, 3-hydroxypyruvate, and 2-oxoglutar-ate), the reactions were completed at 50 �C within 30 min(data not shown). All obtained azine derivatives were sta-ble at room temperature over several days.

Fig. 3A shows a typical chromatogram of the MBTH–azines of pyruvate, 3-hydroxypyruvate and 2-oxoglutarate.Double peaks appeared for 3-hydroxypyruvate (peaks 1and 1 0) and for pyruvate (peaks 3 and 3 0), whereas a singlepeak appeared for 2-oxoglutarate (peak 2). The variationsin the retention times of these MBTH–azines were less than1% within a series of assays. The molecular weights of thedouble peaks determined by mass spectrometry (265 for

86 Assay for D-serine degradation / H. Tanaka et al. / Anal. Biochem. 362 (2007) 83–88

peaks 1 and 1 0, 249 for peaks 3 and 3 0) were identical to thetheoretical values of the corresponding MBTH–azine deriv-atives (265.1 and 249.0, respectively). The appearance ofdouble peaks is known for the MBTH–azine derivativesof aldehydes [15]. The two isomers, E- and Z-isomers, wereproduced from each carbonyl compound by the reactionwith MBTH (Fig. 2). In the case of MBTH–aldehydes,the rate of interconversion between the two isomers is veryslow (on the order of hours) compared to the time neededfor a single chromatographic run (10–20 min), resulting inthe separation of the E- and Z-isomers on the chromatogra-phy [15].

We confirmed that the present appearance of the doublepeaks is due to the same mechanism as that observed forMBTH–aldehydes. As shown in Fig. 3B, the Z-isomer ofMBTH–3-hydroxypyruvate (peak 1 0) was gradually formedfrom the E-isomer (peak 1), which had been preparedimmediately before use by collecting the first peak of a largedose chromatography. The forward and reverse reactionrates were 0.14 ± 0.03 and 0.25 ± 0.05 h�1, respectively, atroom temperature. Fig. 3C shows the isomerization frompeak 3 to peak 3 0 of the MBTH–pyruvate. In this case,the forward and reverse reaction rates were 0.092 ± 0.026and 0.38 ± 0.10 h�1, respectively. The equilibrium ratiosof E- and Z-isomers were pH dependent (Fig. 3). The pres-ent chromatographic conditions could not separate the twoisomers of MBTH–2-oxoglutarate, giving a single peak witha slightly broader peak width (peak 2 in Fig. 3A).

Fig. 4 shows typical calibration curves of the azinederivatives of 3-hydroxypyruvate and pyruvate. The totalarea of the double peaks relative to that of 2-oxoglutarateexhibited a good linear dependence on the amounts of 3-hydroxypyruvate (Fig. 4A) and pyruvate (Fig. 4B), respec-tively, over a wide range of 0.3–50 nmol. As shown below,in the present assay for DSD and DAO activities, theamounts of 3-hydroxypyruvate, pyruvate, and 2-oxoglutar-ate introduced into reaction mixtures by the addition oforgan homogenates were maximally 1.3 nmol. Therefore,the sensitivity of the present method is sufficient for thesimultaneous quantification of DSD and DAO activitiesusing their physiological substrate of D-serine.

Fig. 4. Standard curves for 3-hydroxypyruvate (A) and pyruvate (B). The arearelative to the peak 2 area (2-OG), were plotted against the respective amoumixtures.

It has been reported that chicken kidney contains DSDactivities [12]. To verify the usefulness of the present assaymethod, we examined the enzymatic formation of 3-hydroxypyruvate and/or pyruvate from D-serine usingchicken and rat kidney homogenates (Fig. 5). After 30-min incubation at 37 �C with chicken kidney homogenates,significant amounts of both 3-hydroxypyruvate (peaks 1and 1 0) and pyruvate (peaks 3 and 3 0) were produced fromD-serine, while only 3-hydroxypyruvate was produced incomparable amounts in the case of the rat kidney homog-enates. Benzoate, a potent competitive inhibitor of DAO,completely inhibited the formation of 3-hydroxypyruvateby incubation with the chicken and rat homogenates.Under anaerobic conditions achieved by means of argonflushing [16], the production of 3-hydroxypyruvate wasalso nearly completely inhibited. These results indicatedthat the observed 3-hydroxypyruvate formation was dueto the DAO activities contained in both kidney homoge-nates. On the other hand, the production of pyruvate byincubation with chicken kidney homogenates was signifi-cantly inhibited by hydroxylamine, an inhibitor of pyridox-al-phosphate-dependent enzymes, whereas neitherbenzoate addition nor anaerobic conditions affected thepyruvate production. These results strongly suggested thatDSD is a pyridoxal-phosphate-dependent enzyme and thatthe observed pyruvate formation was due mainly to theDSD activities.

We examined the DSD and DAO activities of chickenand rat brains. Table 1 summarizes the results togetherwith those obtained for kidney and liver. Interestingly,chicken brain contained only DSD activities (0.40 nmol/min/mg protein), while the kidney and liver showed bothDSD and DAO activities. In contrast, all rat organs exam-ined contained only DAO activities and showed no detect-able levels of DSD activity.

Discussion

D-Serine is a physiological agonist needed for NMDARsto function properly in vertebrate brains, including avianbrains [3,17–19]. Therefore, it is important to elucidate

sums of the double peaks (see Fig. 3A), (3HPYR) and (PYR), respectively,nts of 3-hydroxypyruvate and pyruvate contained in the 200-lL reaction

Table 1Comparison of DSD and DAO activities in the brain, kidney, and liver ofchickens and rats

Organs DSD (nmol/min/mg protein) DAO (nmol/min/mg protein)

Chicken Rat Chicken Rat

Brain 0.40 ± 0.20 NDa NDa 0.70 ± 0.10Kidney 4.4 ± 1.4 NDa 14.0 ± 2.3 87.0 ± 15.0Liver 0.40 ± 0.10 NDa 4.5 ± 1.3 4.0 ± 0.5

Values are means ± SD (n = 4 or 5).a ND, not detectable.

Fig. 5. DSD and DAO activities in chicken and rat kidney homogenates. A 10-lL aliquot of chicken kidney homogenates (110 lg protein) and a 2-lLaliquot of rat kidney homogenates (20 lg protein) were used for the assay, respectively. After 30-min incubation at 37 �C, the amounts of 3-hydroxypyruvate and pyruvate were quantified by HPLC after precolumn derivatization with MBTH. For benzoate inhibition, sodium benzoate (5 mM)was added to the reaction mixture. For hydroxylamine inhibition, the homogenates were treated at room temperature with the inhibitor (5 mM) for 5 minbefore the start of the reaction. The peak numbers are the same as those in Fig. 3A.

Assay for D-serine degradation / H. Tanaka et al. / Anal. Biochem. 362 (2007) 83–88 87

how D-serine is degraded in brains in a strictly controlledmanner. In the present study, we developed a quantitativemethod to directly examine the enzymatic degradation ofD-serine in organ homogenates. Because the present methodmeasures the keto acid products from D-serine after conver-sion to stable MBTH–azine derivatives, we can specificallyexamine both DSD and DAO activities in a single assay.The present method is applicable to assays for other ketoacids producing enzymes, such as oxidases (e.g., D-aspartateoxidase), dehydrogenases, or transaminases, with slightmodifications to the MBTH derivatization conditions andthe solvent system for the HPLC analysis of relevantMBTH–azines. For example, in the case of D-aspartate oxi-dase assay, oxalacetate is enzymatically formed from D-as-partate. We can completely decarboxylate the product topyruvate by incubation with MBTH at 80 �C for 15 min.With this modification, we have determined D-aspartateoxidase activities in various mammalian tissues using thepresent method (unpublished).

It is thought that in human and rat brains, D-serine isdegraded by DAO [10,11]. However, because D-serine is apoor substrate for DAO compared to D-proline and D-ala-nine [20], it is difficult to demonstrate the D-serine degrada-

tion by DAO using conventional assay methods, such asoxygen electrode [21] and a peroxidase-coupled procedure[9,22]. By applying the present method, we showed forthe first time the degradation of D-serine by DAO in ratbrain homogenate. On the other hand, DSD activity wasnot detected in the rat brain homogenate. These resultssuggest that, in the rat brain, DAO is the only enzymeresponsible for the degradation of D-serine.

In the context of the metabolism of D-serine, as shown inTable 1, chicken brain contained D-serine degrading activ-ity (DSD activity) at levels comparable to the DAO activ-ities in mammalian brains. To investigate the physiologicalrole of DSD in avian brains, enzymatic characterizationand elucidation of the regional distribution of DSD arecurrently underway in our laboratory.

References

[1] S. Cull-Candy, S. Brickley, M. Farrant, NMDA receptor subunits:diversity, development and disease, Curr. Opin. Neurobiol. 11 (2001)327–335.

[2] I. Perez-Otano, M.D. Ehlers, Homeostatic plasticity and NMDAreceptor trafficking, Trends Neurosci. 28 (2005) 229–238.

[3] A. Panatier, D.T. Theodosis, J.P. Mothet, B. Touquet, L. Pollegioni,D.A. Poulain, S.H.R. Oliet, Glia-derived D-serine controls NMDAreceptor activity and synaptic memory, Cell 125 (2006) 775–784.

[4] N.W. Kleckner, R. Dingledine, Requirement for glycine in activationof NMDA-receptors expressed in Xenopus oocytes, Science 241 (1988)835–837.

[5] M.J. Schell, M.E. Molliver, S.H. Snyder, D-Serine, an endogenoussynaptic modulator: localization to astrocytes and glutamate-stimu-lated release, Proc. Natl. Acad. Sci. USA 92 (1995) 3948–3952.

[6] D.S. Dunlop, A. Neidle, The origin and turnover of D-serine in brain,Biochem. Biophys. Res. Commun. 235 (1997) 26–30.

[7] H. Wolosker, K.N. Sheth, M. Takahashi, J.P. Mothet, R.O. BradyJr., C.D. Ferris, S.H. Snyder, Purification of serine racemase:biosynthesis of the neuromodulator D-serine, Proc. Natl. Acad. Sci.USA 96 (1999) 721–725.

88 Assay for D-serine degradation / H. Tanaka et al. / Anal. Biochem. 362 (2007) 83–88

[8] E.R. Stevens, M. Esguerra, P.M. Kim, E.A. Newman, S.H. Snyder,K.R. Zahs, R.F. Miller, D-Serine and serine racemase are present inthe vertebrate retina and contribute to the physiological activation ofNMDA receptors, Proc. Natl. Acad. Sci. USA 100 (2003) 6789–6794.

[9] K. Horiike, H. Tojo, R. Arai, M. Nozaki, T. Maeda, D-Amino-acidoxidase is confined to the lower brain stem and cerebellum in ratbrain: regional differentiation of astrocytes, Brain Res. 652 (1994)297–303.

[10] Y. Nagata, K. Horiike, T. Maeda, Distribution of free D-serine invertebrate brains, Brain Res. 634 (1994) 291–295.

[11] K. Horiike, T. Ishida, H. Tanaka, R. Arai, Distribution of D-amino-acid oxidase and D-serine in vertebrate brains, J. Mol. Catal. 12B(2001) 37–41.

[12] M.A. Grillo, T. Fossa, M. Coghe, D-Serine dehydratase of chickenkidney, Enzymologia 28 (1965) 377–388.

[13] M.A. Paz, O.O. Blumenfeld, M. Rojkind, E. Henson, C. Furfine,P.M. Gallop, Determination of carbonyl compounds with N-methylbenzothiazolone hydrazone, Arch. Biochem. Biophys. 109 (1965)548–559.

[14] K. Soda, A spectrophotometric microdetermination of keto acidswith 3-methyl-2-benzothiazolone hydrazone, Agr. Biol. Chem. 31(1967) 1054–1060.

[15] G. Chiavari, M.C. Facchini, Behaviour of 3-methyl-2-benzothiazo-lone azines of carbonyl compounds in high-performance liquidchromatography, J. Chromatogr. 387 (1987) 459–466.

[16] H. Nakajima, T. Ishida, H. Tanaka, K. Horiike, Accurate measure-ment of near-micromolar oxygen concentrations in aqueous solutionsbased on enzymatic extradiol cleavage of 4-chlorocatechol: applica-tions to improved low-oxygen experimental systems and quantitativeassessment of back diffusion of oxygen from the atmosphere,J. Biochem. 131 (2002) 523–531.

[17] R.J. Steele, C.R. Dermon, M.G. Stewart, D-Cycloserine causestransient enhancement of memory for a weak aversive stimulus inday-old chicks (Gallus domesticus), Neurobiol. Learn. Mem. 66 (1996)236–240.

[18] S.A. White, Learning to communicate, Curr. Opin. Neurobiol. 11(2001) 510–520.

[19] A. Zarain-Herzberg, I. Lee-Rivera, G. Rodrıguez, A.M. Lopez-Colome, Cloning and characterization of the chick NMDA receptorsubunit-1 gene, Mol. Brain Res. 137 (2005) 235–251.

[20] M. Dixon, K. Kleppe, D-Amino-acid oxidase. II. specificity, compet-itive inhibition and reaction sequence, Biochim. Biophys. Acta 96(1965) 368–382.

[21] M.S. Huynh, K. Horiike, H. Tojo, M. Katagiri, T. Yamano, Kineticproperties of rat kidney D-amino acid oxidase associated withperoxisomes, Comp. Biochem. Physiol. 80B (1985) 425–430.

[22] K. Horiike, R. Arai, H. Tojo, T. Yamano, M. Nozaki, T. Maeda,Histochemical staining of cells containing flavoenzyme D-amino-acidoxidase based on its enzymatic activity: application of a coupledperoxidation method, Acta Histochem. Cytochem. 18 (1985) 539–550.