High-voltage Electrophoresis Research Articles

Of Arginine-Specific Adenosine-5′-Diphosphate-Ribosyltransferase by Capillary Electrophoresis

Using high-voltage capillary electrophoresis we detected ADP-ribosylarginine, a product of ADP-ribosylation reaction catalyzed by arginine-specific ADP-ribosyltransferase in the presence of NAD and L-arginine. The authentic ADP-ribosylarginine, detected by its ultraviolet absorbance at 254 nm, had a different retention time from NAD or nicotinamide. When the ADP-ribosylation reaction products were analyzed, the peak corresponding to ADP-ribosylarginine increased with incubation time and in an enzyme-dose-dependent manner. The lower limit of detection was 0.3 pmol, a value 100 times lower than that obtained with the reversed-phase high-performance liquid chromatography assay described previously. Using the capillary electrophoresis system, a thiol-independent ADP-ribosyltransferase activity was detected in chicken spleen cell membrane. Since the capillary electrophoresis assay for ADP-ribosylarginine is simpler, faster, and less expensive than the high-performance liquid chromatography assay, determination of arginine-specific ADP-ribosyltransferase activity in animal tissues will be facilitated.

Meningeal Involvement in Acute Leukaemia and High-Grade Non-Hodgkin’s Lymphoma is Associated with Elevated Activities of Galactosyltransferases in the Cerebrospinal Fluid

Background: Galactosyltransferases (Gal-T) are intracellular and cell surface enzymes involved in the biosynthesis of glycoconjugates. Gal-T were suggested to be implicated in oncogenesis, since their levels are elevated in various body fluids of patients with different types of malignoma; the invasiveness of certain neoplastic cell lines is correlated with their cell surface Gal-T activities, and cancer-associated Gal-T-isoforms were identified. Material and Methods: Gal-T activities in the cerebrospinal fluid (CSF) of patients with meningeal involvement in acute leukaemia or high-grade lymphoma (n = 8), normal controls (n = 54), patients with acute viral meningitis/encephalitis (n = 8) and multiple sclerosis (n = 10) were compared. CSF Gal-T activities were measured with uridine diphosphate-14C-galactose as substrate after isolation of the enzymatically transferred 14C-galactose by high-voltage electrophoresis. Simultaneously, the permeability of the blood-CSF barrier was analysed by the Q-albumin method. Results: In case of an intact blood-CSF barrier, Gal-T activities were elevated exclusively in patients with meningeal involvement in malignancies, but declined rapidly during intrathecal polychemotherapy. Conclusions: CSF Gal-T activities might prove to be a marker for early or residual involvement of the central nervous system in acute leukaemia or high-grade lymphoma.

Selective screening for amino acid disorders

The analysis of amino acids is the most frequently applied technique in the selective screening of inborn errors of metabolism. When urine is used as a starting material, simple techniques such as thin-layer chromatography or high-voltage electrophoresis is preferred as a first approach. The quantitative analysis requires instrumentation, usually an amino acid analyser. Both plasma and urine are needed for establishing renal transport defects. Apart from the accumulation of the 'usual' amino acids, the presence of unusual amino acids may be of diagnostic significance. Furthermore the finding of decreased plasma concentrations of specific amino acids may pinpoint several inherited defects. No amino acid screening procedure is complete without the availability of an organic acid and a purine/pyrimidine analytical system, both yielding important additional diagnostic information. Considerable clinical problems may occur in subjects with a decreased tolerance to protein amino acids without being homozygous for any inherited defect. Examples of these disorders that need further studies are homocysteinaemia associated with vascular disease and carriers of ornithine transcarbamylase deficiency.

High voltage electrophoresis of amino acids in urine containing ampicillin

Analyis of amino acids is complicated by treatment with ampicillin. High voltage electrophoresis, which is convenient for the qualitative assessment of metabolic diseases, yields smears of ampicillin that mask the bands of citrulline, homocitrulline, phenylallanine, cystine, and homocystine. The addition of penicillinase prior to high voltage electrophoresis eliminates ampicillin and other penicillins and reveals these key amino acids.

Partial characterization of an ∼20 K Mr retinal protein whose phosphorylation is inhibited by taurine

It has been demonstrated previously that taurine (2-aminoethanesulfonic acid) is an inhibitor of protein phosphorylation in a mitochondrial fraction of the rat retina. It appears that taurine is most effective in inhibiting the phosphorylation of an ∼20 K apparent molecular weight ( M r ) protein found in the retinal tissue. This study further characterizes the location of the ∼20 K phosphoprotein by phase separation using Triton X-l 14 and also characterizes the nature of the phosphate bond by various solvent extractions and by exposure to acid and base conditions. Triton X-114 experiments indicated that the ∼20 K phosphoprotein is located in the aqueous phase and, consequently, is probably not an integral protein of the mitochondrial membranes. Treatment of the phosphoprotein with solvents, acid, and/or base determined the phosphate linkage to be through a phosphoester bond rather than an acylphosphate bond. The ∼20 K M r phosphoprotein was also isolated from one-dimensional polyacrylamide gels and subsequently digested with trypsin and hydrolyzed with 1 M HCl to break all peptide bonds. Analysis of the phosphoamino acids by two-dimensional high voltage electrophoresis on cellulose plates revealed that it is both the serine and threonine residues that are phosphorylated. However, phosphorylation of the serine residue(s) is predominant.

In vitro phosphorylation of proteins tightly bound to DNA by protein kinase NII

1. 1. Highly purified DNA from calf thymus was phosphorylated with protein kinase NII. 2. 2. Digestion with proteinase K of this DNA demonstrates proteins as phosphorylated component. 3. 3. Gel filtration chromatography on Bio-Gel A-0.5m gel column shows a major protein peak between 50 and 70 kDa. 4. 4. SDS gel electrophoresis, after hydrolysis, to digest completely DNA, shows three major phosphorylated bands corresponding to polypeptides of Mr between 31 and 21 kDa. 5. 5. After high voltage electrophoresis on TLC plates tryptic digested polypeptides show very similar phosphopeptides patterns.

Structural determination of the glycolipid anchors of human and porcine membrane dipeptidases

Glycosyl-phosphatidylinositol (glycolipid) membrane anchors are responsible for anchoring many eukaryotic cell-surface proteins in the plasma membrane [reviewed in 1,2] and may be thought of as an alternative anchoring mechanism to the hydrophobic transmembrane polypeptide anchor. Although more than 100 examples of glycolipid anchored proteins have been described, very few have been structurally characterised in detail as the chemical analysis is a major commitment [2]. Of these only three are mammalian in origin: rat brain Thy-1 antigen [3], human erythrocyte acetylcholinestem [4] and scrapie prion protein [51. From such studies the glycolipid anchor is thought to consist of an apparently highly conserved motif of ethanolamine phosphate6Manal-2Manal-6Mana 1 -4GlcNa 1 -6mypinositol1 -phosphatelipid with variable side-chain modifications to the glycan core [21. The work presented here reports the preliminary structural determination of the glycan structures of porcine and human membrane dipeptidases (dehydmpeptidase-I; EC 3.4.13.11). This will provide us with more information about mammalian glycolipid anchors and also represents the first inter-species comparison of mammalian glycolipid anchor structures on the same protein. Membrane dipeptidase is an ectoenzyme found abundantly in kidney cortex and was purified in essentially two steps. Membranes prepared by differential centrifugation were incubated with bacterial phosphatidylinositol-specifk phospholipase C to yield the hydrophilic form of the enzyme and this was then purified by affhity chromatography on cilastatindepharose [6]. The analytical strategy and techniques used to characterise the glycan core have been developed at The University of Dundee and are described in detail in [7]. First the neutral glycan was isolated directly from the whole glycoprotein. Hydrophilic membrane dipeptidase (lOnmol) from either porcine or human origin was desialated (40mM triflumacetic acid; lh at 800C), subjected to mild base hydrolysis with concentrated ammonia:methanol (1: 1 v/v) to remove any inositol palmitoylation and dephosphorylated with 48% (v/v) aqueous hydrogen fluoride. The protein present was removed by trichloroacetic acid precipitation and the sample was deaminated with sodium nitrite and reduced with tritiated sodium borohydride. Radiochemical purification was then employed to remove the substantial radiochemical contamination present and involved downward paper chromatography and highvoltage electrophoresis. The resulting neutral glycan is the glycan structure present in the glycolipid anchor except that the glucosamine has been converted to a [3H]-labelled 2,5anhydromannitol (AHM) species. The presence of the radiolabel greatly facilitates identification of the structures present in subsequent analyses. However, the neutral glycan lacks any sialic acid residues or additional ethanolamine phosphate residues that might have been present in the original anchor structure. Liquid chromatography was then performed on 4% (1x106 c.p.m.) of each of the two neutral glycans to separate the different structures present. The samples were firstly separated by Dionex

No alpha-lactalbumin-like activity detected in a low molecular mass protein fraction of rat epididymal extract

The same low M(r) protein fraction of epididymal luminal fluid as that studied previously by Hamilton in 1981 was assayed for alpha-lactalbumin (alpha-lac) activity with bovine milk galactosyltransferase (GalTase) as the enzyme source and bovine serum albumin (BSA) to make the total protein concentrations of all the samples in each assay constant. No alpha-lac activity could be detected in this fraction. When glucose was used as an acceptor, radioactivity passing through the ion exchange column used to retain the remaining UDP-galactose did increase with the amount of the proteins of the low M(r) fraction, but this increase was observed not only for samples with an acceptor but also for samples without. The increased radioactive product co-migrated in high-voltage electrophoresis with galactose, not lactose. When GlcNAc was used as an acceptor, the situation was the same, and the increased radioactive product co-migrated with galactose, not LacNAc. No inhibition of bovine milk GalTase activity by the low M(r) proteins was observed. Addition of protein, either BSA or the low M(r) fraction of epididymal luminal fluid, to assay medium that contained no proteins other than GalTase enhanced the GalTase activity both in forming lactose in the presence of glucose and also LacNAc in the presence of GlcNAc. It appears that the earlier reports of alpha-lac-like activity in epididymal fluids and extracts may have been due to the presence of enzymes liberating free galactose from UDP-galactose and/or a stimulatory non-specific effect of the protein in the solutions on the lactose synthesis activity of the GalTase.

Protein kinase C in rat brain myelin.

It has been suggested that phosphorylation of myelin basic protein (MBP) in CNS is catalyzed by protein kinase C (PKC). In order to demonstrate that PKC in the myelin phosphorylates MBP, PKC was partially purified from rat CNS myelin by solubilization with Triton X-100 followed by a DEAE-cellulose column. MBP and histone III-S were phosphorylated in the presence of Ca2+ and phospholipid by rat myelin PKC. High voltage electrophoresis revealed that the phosphoamino acids in MBP by this kinase was serine residue, which is known to be the amino acid phosphorylated by PKC. The activity of PKC extracted from myelin was inhibited by the addition of psychosine to the incubation mixture. To confirm the presence of PKC molecule and to identify the isoform of PKC in the myelin, the solubilized myelin fraction was applied on SDS-PAGE, transferred to a nitrocellulose sheet and stained with anti-PKC monoclonal antibodies. Rat CNS myelin contained the PKC of about 80 kDa (intact PKC), and no proteolytic fragments were observed. PKC isozymes in myelin were type II and III. A developmental study from 14 to 42 postnatal days showed that PKC activity in CNS myelin seemed to parallel the deposition of myelin protein.

Glycation of human tissue and serum creatine kinase

Human creatine kinase (CK) was demonstrated to be partly present as a glycated molecule. Sialic acid, galactose and sulfate were also found to be present on the molecule. The glycated forms were characterized by higher activation energies and were thermally unstable. Skeletal muscle CK showed lower relative binding towards the lectin concanavalin A (Con A) in comparison with the heart tissue forms. After skeletal muscle trauma, CK in serum was found to be less glycated than in tissue. After acute myocardial infarction (AMI), no glycated CK could be detected in serum. Following injury, it appears that the transfer from tissue to plasma is accompanied by a loss of glycated isoforms. High-voltage electrophoresis showed no differences in the distribution of CK isoforms between the glycated and non-glycated forms.

Creatine kinase isoforms: investigation of inhibitors of in vitro degradation and establishment of a reference range.

The in vitro stability of creatine kinase isoforms was examined by separation with high voltage electrophoresis. The effect of inhibitors of carboxypeptidase was evaluated. Preservation of samples is essential to inhibit in vitro changes in isoform pattern. EDTA at a final concentration of 15 mmol/L is recommended. Using appropriately preserved samples, normal reference intervals for the MM isoforms have been established.

Mechanistic and active-site studies on d(–)-mandelate dehydrogenase from Rhodotorula graminis

D(--)-Mandelate dehydrogenase, the first enzyme of the mandelate pathway in the yeast Rhodotorula graminis, catalyses the NAD(+)-dependent oxidation of D(--)-mandelate to phenylglyoxylate. D(--)-2-(Bromoethanoyloxy)-2-phenylethanoic acid ['D(--)-bromoacetylmandelic acid'], an analogue of the natural substrate, was synthesized as a probe for reactive and accessible nucleophilic groups within the active site of the enzyme. D(--)-Mandelate dehydrogenase was inactivated by D(--)-bromoacetylmandelate in a psuedo-first-order process. D(--)-Mandelate protected against inactivation, suggesting that the residue that reacts with the inhibitor is located at or near the active site. Complete inactivation of the enzyme resulted in the incorporation of approx. 1 mol of label/mol of enzyme subunit. D(--)-Mandelate dehydrogenase that had been inactivated with 14C-labelled D(--)-bromoacetylmandelate was digested with trypsin; there was substantial incorporation of 14C into two tryptic-digest peptides, and this was lowered in the presence of substrate. One of the tryptic peptides had the sequence Val-Xaa-Leu-Glu-Ile-Gly-Lys, with the residue at the second position being the site of radiolabel incorporation. The complete sequence of the second peptide was not determined, but it was probably an N-terminally extended version of the first peptide. High-voltage electrophoresis of the products of hydrolysis of modified protein showed that the major peak of radioactivity co-migrated with N tau-carboxymethylhistidine, indicating that a histidine residue at the active site of the enzyme is the most likely nucleophile with which D(--)-bromoacetylmandelate reacts. D(--)-Mandelate dehydrogenase was incubated with phenylglyoxylate and either (4S)-[4-3H]NADH or (4R)-[4-3H]NADH and then the resulting D(--)-mandelate and NAD+ were isolated. The enzyme transferred the pro-R-hydrogen atom from NADH during the reduction of phenylglyoxylate. The results are discussed with particular reference to the possibility that this enzyme evolved by the recruitment of a 2-hydroxy acid dehydrogenase from another metabolic pathway.

Alterations of the glycan moiety of human αl-acid glycoprotein in late-term pregnancy

The carbohydrate moiety of purified αl-acid glycoprotein (AGP) from healthy male adults (AGP n) and late-term pregnant women (AGP p) was analysed. Polyacrylamide gel electrophoresis with sodium dodecyl sulfate before and after N-glycanase treatment showed that AGP p had a slightly higher molecular mass due to an enriched carbohydrate moiety. BIO-GEL P-4 and Concanavalin A (Con A)-Sepharose chromatography of the oligosaccharides released by hydrazinolysis and fractionated by high-voltage electrophoresis indicated a progression towards Con A-unbound oligosaccharides and towards larger glycans in pregnancy. Carbohydrate analysis of purified AGP p and AGP n and of the most increased oligosaccharide fraction (F4A) evidenced a decrease in the fucosyl molar ratio and a slight increase in the galactosyl, N-acetyl-glucosaminyl and N-acetyl neuraminyl ratios. These results suggest that AGP contains more highly branched oligosaccharides and/or additional N-acetyllactosamine-type oligosaccharides in pregnancy.

Synthesis of glycosylated tuftsins and tuftsin‐containing IgG fragment undecapeptide

Syntheses are described of two new tuftsin derivatives containing a 2-acetamido-2-deoxy-D-galactopyranosyl unit alpha- or beta-glycosidically linked to the threonine's hydroxy side chain function and of the glycosylated undecapeptide corresponding to the tuftsin region of the heavy chain of IgG (amino acid sequence 289-299). The glycosylated tuftsins were synthesized by the solution procedure. Fmoc-[Gal NAc(Ac)3 alpha]Thr-OH and Fmoc-[GalNAc(Ac)3 beta]Thr-OH were allowed to react with H-Lys(Z)-Pro-Arg(NO2)-OBzl by the mixed anhydride procedure and the resulting glycosylated tetrapeptides were fully deblocked by catalytic hydrogenation followed by treatment with potassium cyanide, purified by ion exchange chromatography and characterized by analytical HPLC, elemental and amino acid analyses, optical rotation, and proton NMR spectroscopy. Synthesis of the glycosylated undecapeptide was achieved by the continuous flow solid phase procedure on 4-hydroxymethylphenoxyacetyl-norleucyl derivatized Kieselguhr-supported resin. Fmoc-amino acid symmetrical anhydrides or pentafluorophenyl esters, in the presence of N-hydroxybenzotriazole, were used as the acylating agents. To mimic the native sequence of the tuftsin region at the Fc-domain of immunoglobulin G a 2-acetamido-2-deoxy-beta-D-glucopyranosyl unit was N-glycosidically linked to the amide side chain of Asn 297. The glycosylated asparagine residue was introduced as N2-fluorenylmethyloxycarbonyl-N4-(2-acetamido-3,4,6-tri-O-acetyl-2 -deoxy-beta-D - glucopyranosyl)-asparagine pentafluorophenyl ester. After cleavage from the resin the glycopeptide was deprotected, purified by ion exchange chromatography, and characterized by analytical HPLC, amino acid analysis, high voltage electrophoresis, and proton NMR. The conformational features of the glyco-undecapeptide were determined by circular dichroism measurements both in water and in 98% trifluoroethanol. Results of biological assays will be published elsewhere.

Characterization of the nickel-rich extract from the nickel hyperaccumulator Dichapetalum gelonioides

Abstract Following the discovery of the hyperaccumulation of nickel by the Philippine plant Dichapetalum gelonioides subsp. tuberculatum , the nature of the nickel in aqueous plant extracts has been investigated by gel chromatography, ion exchange chromatography, high voltage electrophoresis and GC-MS. The purified extract contained 18% Ni, 24% citric acid and 43% malic acid(mole ratio 1:0.4:1), the remainder consisting of small amounts of Ca, Mg, K, Na and a trace of tartaric acid. Experimental observations are discussed in terms of the stabilities of complexes of citrate and malate with Ni. By using X-ray crystallography it was shown that crystals obtained from a Ni-citrate-malate solution simulating the extract, contain Ni exclusively in the form of an anionic Ni(II)-citrato complex.

Biochemical and Functional Effects of Sulfate Restriction in the Marine Sponge, Microciona prolifera.

The functional and biochemical consequences of sulfate restriction were studied in chemically dissociated Microciona sponge cells maintained in artificial seawater with or without SO42-. In cells pre-treated to reduce preformed secretions, SO42- deprivation reduced cell motility judged by the lack of aggregates in rotating or stationary cultures in comparison with controls. Microscopic examination showed that cells that customarily demonstrate cytoplasmic processes, such as filopodia and pseudopodia, exhibited marked decreases in these cellular processes when maintained in SO42--deprived artificial seawater. Uptake and incorporation of 35SO42- by disaggregated and pre-treated cells was higher under SO42--free conditions relative to controls; this effect was time dependent, rising to a maximum at 12 h, when a three- to seven-fold difference could be demonstrated. 3H-leucine incorporation indicated that protein synthesis was similar in test and control populations. Comparative high voltage electrophoresis of supernatants containing 35SO4 macromolecules from chemically dissociated cells indicated deficiencies of such 35SO4 macromolecules if the rotated cells that released these secretions had been pre-treated in SO42--free artificial seawater. The results of SO42- restriction suggest that secretion of macromolecules or Microciona aggregation factor (MAF), and aggregation and locomotion of Microciona cells depend upon an adequate extracellular source of SO42-, sulfate transport, and sulfation of macromolecules such as polysaccharides.

Gamma-Glutamylglutamine identified in plasma and cerebrospinal fluid from hyperammonaemic patients

gamma-Glutamylglutamine has been identified in plasma and cerebrospinal fluid. Preparative high voltage electrophoresis was used to isolate the compound from two to three ml of sample. Analysis of archival material from eight patients with urea cycle disorders showed that plasma γ-glutamylglutamine was 20–214 μmol/l when the plasma glutamine was greater than 1000 μmol/l and that the relationship was strongly positive ( P < 0.0005). Plasma γ-glutamylglutamine concentration was also raised in one patient with secondary hyperammonaemia, but not in three other cases, one of whom had a plasma glutamine of 10000 μmol/l. Cerebrospinal fluid from the last patient contained 97 μmol/l of γ-glutamylglutamine. Samples of cerebrospinal fluid from two patients with urea cycle disorders were available and the γ-glutamylglutamine levels were 203 and 313 μmol/l. gamma-Glutamylglutamine and 5-oxoproline concentrations were not significantly increased in urine from any of these patients.

Depletion of cystine in cystinotic fibroblasts by homocysteine: Synergism of cysteamine with various reducing agents in depletion of cystine from cystinotic fibroblasts

The present study shows that homocysteine depleted cystine from cystinotic fibroblasts in vitro. No toxic effects were noted as judged by morphology and growth patterns. Efflux of radioactivity from cystinotic cells prelabeled with [ 35S]cystine was greater in homocysteine-treated cystinotic cells than in untreated controls. This radioactivity was found, by high voltage electrophoresis separation of effluxed products, to consist mainly of [ 35S]cystine, along with smaller amounts of [ 35S]homocysteine-cysteine mixed disulfide. When homocysteine and cysteamine were presented together to cystinotic cells at dose levels individually ineffective in removing cystine from these cells, a marked synergistic effect was observed and cystine content fell to 10% of that seen in untreated cystinotic fibroblasts. Similarly, synergistic effects of cystine depletion from cystinotic cells were demonstrated when cells were treated with a combination of cysteamine and dithiothreitol or glutathione. Incubation of cystinotic cells with homocysteine, dithiothreitol, or cysteamine in combination with vitamin C did not yield synergistic effects. The above findings suggest a novel way to probe metabolic processes in these mutant cells. Exploration of these synergistic effects may lead to more efficacious therapeutic protocols for cystinosis.

Tyrosine phosphorylation of the asialoglycoprotein receptor.

The asialoglycoprotein (ASGP) receptor undergoes constitutive endocytosis through the coated pit/coated vesicle pathway in hepatocytes. Studies on HepG2 cells have shown that the receptor is phosphorylated at serine under control conditions and following protein kinase C stimulation. This study examined whether the ASGP receptor could also serve as a substrate for a tyrosine kinase in HepG2 cells. 32P labeling was performed in membrane preparations, in permeabilized cells at 4 degrees C, and in intact cells at 37 degrees C. The phosphorylated ASGP receptor was isolated by immunoprecipitation, hydrolyzed in 6 N HCl at 110 degrees C, and analyzed by two-dimensional high voltage electrophoresis. The receptor isolated from a membrane preparation incubated in vitro with [gamma-32P]ATP incorporated radiolabel predominantly (greater than 90%) into phosphotyrosine. ASGP receptor phosphorylation at both tyrosine and serine was detected in intact cells incubated with phosphatase inhibitors for 60 min at 37 degrees C. The presence of both phenylarsine oxide (20 microM) and sodium orthovanadate (200 microM) was required for tyrosine phosphorylation. Use of these inhibitors together resulted in a 16.4-fold increase in phosphorylation of the immunoprecipitated ASGP receptor, whereas phosphorylation of total HepG2 membrane proteins was not significantly augmented by this procedure. Selective proteolytic digestion of ASGP receptors in isolated vesicles demonstrated that the phosphorylation site identified in these studies is located at tyrosine 5 in the cytoplasmic tail.

A study in rat brain cortex synaptic vesicles of endogenous ligands for N-methyl- d-aspartate receptors

The presence of endogenous ligands for the N-methyl- d-aspartate receptor was looked for in highly purified rat brain cortex synaptic vesicles, the contents of which were extracted and fractionated by gel filtration on Sephadex G-10, or by three different high-voltage electrophoresis procedures. The presence of endogenous ligands was detected by their ability to compete with 50 nM l-[ 3H]glutamate for binding to whole rat brain N-methyl- d-aspartate receptors. The receptor preparations used were those present in purified postsynaptic densities, in which the quisqualate receptors were blocked by 10 μM quisqualate. Synaptic vesicles had a high content of N-methyl- d-aspartate receptor ligands, which on fractionation always coincided with glutamate or aspartate. A variable and very small amount of a highly acidic endogenous ligand was also found. The latter substance did not coincide in the electrophoresis with homocysteic, cysteic, quinolinic, cysteine sulphinic or homocysteine sulphinic acids, or with N-acetyl-aspartyl-glutamic acid, S-sulphocysteine or sulphoserine. We also found that a single centrifugation, in 0.25 M sucrose, 25 mM Tris-citrate, pH 7.1, of purified synaptic vesicles, at 135,000 g max for 45 min, led to a 51% loss of endogenous glutamate, but did not change their aspartate content. Thus, in uncentrifuged vesicles the glutamate/aspartate ratio was 9.4, while in centrifuged ones the ratio was 3.9. ATP markedly enhanced l-[ 3H]glutamate uptake into synaptic vesicles, but did not change the binding of l-[ 3H]aspartate. Differences in labelled aspartate and glutamate efflux from the vesicles were also found. These marked differences between glutamate and aspartate suggest that aspartate is not present inside synaptic vesicles and, therefore, does not play a direct role in central neurotransmission. This leaves l-glutamate as the only endogenous ligand for the N-methyl- d-aspartate receptor which can be detected in synaptic vesicles by the procedures used.