To determine specifically NMDA contained in an amphioxus homogenate, the sample was highly purified from other amino acids and amino compounds, then the NMDA content in the purified sample was determined by HPLC. The purification of NMDA from a crude homogenate was an essential step in order to avoid interference by the chromatographic peaks of other amino acids that could overlap with the peak of methylamine, CH₃NH₂ (generated from the oxidation of NMDA with D-AspO). In fact, after the sample has been purified with OPA, almost 95–98% of all NMDA is purified from other compounds. The use of the D-aspartate oxidase for the determination of NMDA was another essential condition, because NMDA as such does not react with OPA-mercaptoethanol to form a fluorescent complex. Therefore, NMDA was first oxidized with D-AspO to form CH₃NH₂ and then treated with OPA-mercaptoethanol. In this way the CH₃NH₂ reacts with OPA-mercaptoethanol giving a fluorescent chromatographic peak. Figure 2 shows a typical example of NMDA determination from a pool of amphioxus nerve cords. In panel A, an analysis of total amino acids obtained from 2 mg of tissues is shown. Panel B shows the same sample after OPA purification, in this case almost all the amino acid peaks have disappeared. Furthermore, when the sample is treated with D-AspO there is an increase of the peak at elution time of about 12 min due to the formation of CH₃NH₂ derived from the oxidation of NMDA (Fig. 2, panel C).

In this study, we have improved the previously described procedure for the purification of NMDA from biological samples 910, showing that purified NMDA from the nervous system of the amphioxus B. lanceolatum (Table 1) can be obtained with minimal loss during the procedure.

Table 2 compares the concentration of NMDA found in the amphioxus CNS with nervous tissues from different animal phyla, such as chordates: mammals (Rattus norvegicus), birds (Gallus gallus), amphibians (Rana esculenta), ray fish (Torpedo ocellata) and tunicates (Ciona intestinalis); arthropods: crustacean (Carcinus maenas); and mollusks: cephalopod (Octopus vulgaris).

The concentration of NMDA in the nervous tissue of several animals ranges normally between 1.4 and 7.6 nmol/g of tissue, except for the cerebral ganglion of Ciona intestinalis, where NMDA occurs at considerably higher concentrations, 13.8 nmol/g of tissue (Table 2). Interestingly also the amphioxus brain contains a very high NMDA concentration (10.52 ± 1.41 nmol/g).

In rat 7 and in ascidians 9 NMDA is biosynthesized from D-aspartic acid, which represents the precursor amino acid for the synthesis of this molecule, and SAM (S-Adenosyl Metionine) as the methyl group donor. In this study we carried out parallel experiments, finding a very high NMDA biosynthetic activity when D-Asp was used as substrate and when we used extracts from the same amphioxus tissues in which NMDA was found endogenously (Table 3). In fact, the nerve cord and particularly the brain were the tissues in which a higher amount of NMDA was synthesized: 33.4 ± 4.30 and 45.5 ± 5.4 nmol/g of tissue, respectively. However, muscular tissues show a poor NMDA synthase activity (1.5 ± 0.4 nmol/g of tissue). It seems, thus, that the amount of endogenous NMDA is proportionally comparable with in vitro activity of D-aspartate methyl transferase enzyme, able to transform the D-aspartic residue into NMDA.

In order to know if only D-Asp constitutes the NMDA precursor molecule, several different amino acids, in both D- and L-conformation, were used as potential substrate in the same reaction conditions as previously described for D-Asp (Table 3). The results obtained demonstrate that, except for D-Asp, no other amino acid is able to act as precursor for the synthesis of NMDA.

Adult amphioxus Branchiostoma lanceolatum were collected in the bay of Argelès-sur-Mer (southern coast of France) in January 2007 as described in Fuentes et al., 2007 15. Adult animals were kept for a week before dissection in glass tanks in the laboratory at a water temperature of 15°C and a 10/14 day-night light cycle (corresponding to natural conditions). Filtered seawater was changed daily and kanamycin antibiotic (10 mg/l sea water) was added to prevent bacterial contamination. Specimens were dissected under an optic microscope using a micro-fine knife with a 4 mm cutting edge (F.S.T.) on ice. No anaesthetic was used to immobilize the animals, in order to avoid drug-induced alterations in the measurement of NMDA. All salts, amino acids, OPA (o-phthaldialdehyde), NAC (N-acetyl-L-cysteine) and other chemicals were purchased from Sigma Chemical Company. The Sep-Pak^® ODS C₁₈ (octadecylsilyl-C₁₈) cartridges were purchased from Waters Corporation, Milford, MA, USA.

Nerve cord, brain and muscle tissues were obtained from adults' animals and divided in five pools (each pool containing tissue from 20 animals). Samples were homogenized in 0.1 M TCA (trichloroacetic acid) in a ratio of 1:10 and centrifuged at 13,000 g for 5 minutes. The supernatants were purified to separate NMDA from other free amino acids and amino compounds. The sample purification was necessary to avoid interference of the NMDA peak by the peaks of other amino acids or amino compounds at HPLC. The principle of purification was based on the fact that free amino acids and amino compounds react with OPA (o-phthaldialdehyde) to give a complex "amino acid-OPA" which, if passed on an ODS-C₁₈ Sep-Pak cartridge, binds strongly to the resin. NMDA instead, since it does not possess a primary amino group (NH₂), does not react with OPA, therefore it does not form a complex with OPA and passes through the ODS-C₁₈ column in purified form 9101112. Briefly, the procedure is the following: 0.3 ml of TCA supernatant of tissue homogenate, as obtained above, is brought to pH 9.5 with 0.1 M NaOH and then mixed with 100 μl of 0.1 M sodium borate-boric acid buffer, pH 9.5 (to stabilize the pH), 20 μl of 1 M OPA and 2 μl of mercaptoethanol and incubated for 30 min at 37°C to complete the formation of the "OPA-amino acids" complex. After incubation, the mixture is brought to pH 8.2 with 0.1 M HCl and passed through an ODS-C₁₈ cartridge containing about 800 mg of resin (the Sep-Pak had been previously regenerated by treatment with 10 ml of methanol and washed with 10 ml of distilled water). After the sample has been loaded on the cartridge, the eluent is collected. The cartridge is washed with 5 ml distilled water and this last eluent is pooled with the previous eluent and dried in a Petri dish, placed on a warm plate at 40–50°C and under aspiration in a hood until dried. The residue is dissolved in 0.3 ml distilled water and used for the HPLC determination of NMDA.

This method is based on the measurement by HPLC of CH₃NH₂ (methylamine), which is generated by the oxidation of NMDA with D-aspartate oxidase (D-AspO) according to the reaction in Figure 391011.

The CH₃NH₂ formed is then treated with OPA-mercaptoethanol to generate a fluorescent compound and analyzed by HPLC. The method is specific for NMDA since the purified beef kidney D-AspO oxidizes only D-Asp and NMDA 161718 and only NMDA can generate CH₃NH₂ when NMDA is oxidized by the D-AspO 78. Briefly, the procedure is as follows: 100 μl of purified sample is mixed with 50 μl of 0.2 M borate buffer, pH 8.2, and 5 μl of purified D-AspO (1–2 mg/ml; 200–300 U/ml) 161718 and incubated at 37°C for 30 min. After incubation, 100 μl of 0.1 M sodium borate buffer, pH 10.0, and 5.0 μl of OPA-mercaptoethanol reagent (20 mg of OPA plus 40 μl of 2-mercaptoethanol in 2 ml of methanol) are added and mixed. After 2 min, 100 μl of this mixture is injected into an ODS-C₁₈ Supelcosil column (μ ◂ 5 μm, 250 × 4.6 mm) connected to a Beckman-Gold HPLC system and to a fluorescence detector with excitation wavelength 330 nm and emission wavelength 450 nm. The column is eluted at 1.2 ml/min with a gradient consisting of 2 mobile phases, A and B. Solvent A consists of: 920 ml of double distilled water, 50 ml acetonitrile and 30 ml of 1.0 M citrate-phosphate buffer, pH 5.3 (final pH 5.6). Solvent B consists of: 900 ml acetonitrile and 100 ml double distilled water. The column is equilibrated with solvent A and then the following gradient program is carried out: 0–30% B over 3 min; 30–70% B over 11 min; 70–100% B over 4 min, 100% B for 4 min and return to 0% B in 1 min. The CH₃-NH₂ elutes as a sharp peak at the retention time of 11.8–12.0 min.

In the same way they are carried out a blank sample, a general standard and a general blank. The blank sample is prepared as the sample, but no D-AspO was added. The standard is prepared using 100 μl of NMDA at a concentration of 0.01 nmol/ml instead of the sample. The general blank is prepared using 100 μl of distilled water instead of the sample.

D-Aspartate oxidase (D-AspO: EC 1.4.3.1) is an oxidative enzyme, which specifically catalyzed the oxidation of only the dicarboxylic amino acids in D-form (D-Asp, D-Glu and NMDA). However, it has been shown that the purified enzyme obtained from beef kidney oxidizes primarily D-Asp and NMDA and only minimally D-Glu (D-Glu is oxidized in a ratio of 2–3% compared to D-Asp or NMDA). On the other hand, the purified D-AspO obtained from the hepatopancreas of the mollusc Octopus vulgaris oxidizes only D-Asp and D-Glu, and only minimally NMDA (4–6% compared to D-Asp or D-Glu 161718). In this study we have used purified beef kidney D-aspartate oxidase obtained by recombinant expression in E. coli according to the described procedure 16.

Recently, we reported that the amphioxus nervous system contains considerable amounts of free D-aspartic acid 14. Since this amino acid constitutes the precursor for the in vivo biosynthesis of NMDA 91011, in the present study we have designed experiments to reveal if NMDA is biosynthesized from D-Asp in the amphioxus nervous system. To verify this, we used the described procedure 9 modified as follows: pools of nerve cord, cephalic vesicle/hindbrain and muscle, obtained from 10 animals were homogenized with 200 μl of 0.05 M phosphate buffer, pH 7.5, and centrifuged at 13,000 rpm for 10 min. Then, 100 μl of the supernatant was incubated with 10 μl of 0.2 M D-Asp (used as precursor for the biosynthesis of NMDA) and with 10 μl of 10 mM SAM (S-AdenosylMethionine, used as methyl group donor) for 2 h at 37°C. A blank sample was carried out in the same way as the sample, but without D-Asp. After incubation, the sample was mixed with 200 μl of 0.2 M TCA and centrifuged at 13,000 rpm for 5 min. The supernatant was neutralized with 0.1 M NaOH and was subjected to the purification of NMDA by using the above method for the determination of NMDA by HPLC. In the same assay conditions other amino acids (L-Asp, D-Glu, L-Glu, D-Ala, L-Ala, D-Ser, L-Ser and Gly) instead of D-Asp were used.

Statistical analyses were performed using Statistical Pachage (Statsoft, 98^th Edition, 1977).