Structural role for a conserved region in the CTP synthetase glutamine amide transfer domain.

Site-directed mutations were introduced into a conserved region of the Escherichia coli CTP synthetase glutamine amide transfer domain. The amino acid replacements, valine 349 to serine, glycine 351 to alanine, glycine 352 to proline, and glycine 352 to cysteine, all increased the lability of CTP synthetase. The proline 352 replacement abolished the capacity to form the covalent glutaminyl-cysteine 379 catalytic intermediate, thus preventing glutamine amide transfer function; NH3-dependent CTP synthetase activity was retained. In CTP synthetase (serine 349), both glutamine and NH3-dependent activities were increased approximately 30% relative to that of the wild type. CTP synthetase mutants alanine 351 and cysteine 352 were not overproduced because of apparent instability and proteolytic degradation. We conclude that the conserved region between residues 346 and 355 in the CTP synthetase glutamine amide transfer domain has an important structural role.     

A cysteine-histidine-aspartate catalytic triad is involved in glutamine amide transfer function in purF-type glutamine amidotransferases

A family of four glutamine amidotransferases has a homologous glutamine amide transfer domain, designated purF-type, that is named after purF-encoded glutamine phosphoribosylpyrophosphate amidotransferase. The glutamine amide transfer domain of approximately 194 amino acid residues is at the NH2 terminus of the protein chain. Site-directed mutagenesis was used to replace several of the 9 invariant amino acids in the glutamine amide transfer domain of glutamine phosphoribosylpyrophosphate amidotransferase. The results indicate that a Cys1-His101-Asp29 catalytic triad is involved in the glutamine amide transfer function of this enzyme. The evidence suggests that His101 functions to increase the nucleophilicity of Cys1, which is used to form a glutamine-enzyme covalent intermediate. Asp29 has a role subsequent to formation of the covalent intermediate. The Cys-His-Asp catalytic triad is implicated in the glutamine amide transfer function of purF-type amidotransferases.     

Replacement by site-directed mutagenesis indicates a role for histidine 170 in the glutamine amide transfer function of anthranilate synthase

Anthranilate synthase is a glutamine amidotransferase that catalyzes the first reaction in tryptophan biosynthesis. Conserved amino acid residues likely to be essential for glutamine-dependent activity were identified by alignment of the glutamine amide transfer domains in four different enzymes: anthranilate synthase component II (AS II), p-aminobenzoate synthase component II, GMP synthetase, and carbamoyl-P synthetase. Conserved amino acids were mainly localized in three clusters. A single conserved histidine, AS II His-170, was replaced by tyrosine using site-directed mutagenesis. Glutamine-dependent enzyme activity was undetectable in the Tyr-170 mutant, whereas the NH3-dependent activity was unchanged. Affinity labeling of AS II active site Cys-84 by 6-diazo-5-oxonorleucine was used to distinguish whether His-170 has a role in formation or in breakdown of the covalent glutaminyl-Cys-84 intermediate. The data favor the interpretation that His-170 functions as a general base to promote glutaminylation of Cys-84. Reversion analysis was consistent with a proposed role of His-170 in catalysis as opposed to a structural function. These experiments demonstrate the application of combining sequence analyses to identify conserved, possibly functional amino acids, site-directed mutagenesis to replace candidate amino acids, and protein chemistry for analysis of mutationally altered proteins, a regimen that can provide new insights into enzyme function.     

Identification of a trpG-related glutamine amide transfer domain in Escherichia coli GMP synthetase

An improved method was developed to align related protein sequences and search for homology. A glutamine amide transfer domain was identified in an NH2-terminal segment of GMP synthetase from Escherichia coli. Amino acid residues 1-198 in GMP synthetase are homologous with the glutamine amide transfer domain in trpG X D-encoded anthranilate synthase component II-anthranilate phosphoribosyltransferase and the related pabA-encoded p-aminobenzoate synthase component II. This result supports a model for gene fusion in which a trpG-related glutamine amide transfer domain was recruited to augment the function of a primitive NH3-dependent GMP synthetase. Sequence analyses emphasize that glutamine amide transfer domains are thus far found only at the NH2 terminus of fused proteins. Two rules are formulated to explain trpG and trpG-related fusions. (i) trpG and trpG-related genes must have translocated immediately up-stream of genes destined for fusion in order to position a glutamine amide transfer domain at the NH2 terminus after fusion. (ii) trpG and trpG-related genes could not translocate adjacent to a regulatory region at the 5' end of an operon. These rules explain known trpG-like fusions and explain why trpG and pabA are not fused to trpE and pabB, respectively. Alignment searches of GMP synthetase with two other enzymes that bind GMP, E. coli amidophosphoribosyltransferase and human hypoxanthine-guanine phosphoribosyltransferase, suggest a structurally homologous segment which may constitute a GMP binding site.     

Characterization of the glutamine site of Escherichia coli guanosine 5'-monophosphate synthetase.

Alkylation of guanosine 5'-monophosphate (GMP) synthetase with the glutamine analogs L-2-amino-4-oxo-5-chloropentanoic acid (chloroketon) and 6-diazo-5-oxonorleucine (DON) inactivated glutamine- and NH3-dependent GMP synthetase. Inactivation exhibited second order kinetics. Complete inactivation was accompanied by covalent attachment of 0.4 to 0.5 equivalent of chloroketon/subunit. Alkylation of GMP synthetase with iodacetamide selectively inactivated glutamine-dependent activity. The NH3-dependent activity was relatively unaffected. Approximately 1 equivalent of carboxamidomethyl group was incorporated per subunit. Carboxymethylcysteine was the only modified amino acid hydrolysis. Prior treatment with chloroketone decreased the capacity for alkylation by iodacetamide, suggesting that both reagents alkylate the same residue. GMP synthetase exhibits glutaminase activity when ATP is replaced by adenosine plus PPi. Iodoacetamide inactivates glutaminase concomitant with glutamine-dependent GMP synthetase. Analysis of pH versus velocity and Km data indicates that the amide of glutamine remains enzyme bound and does not mix with exogenous NH3 in the synthesis of GMP.     

Glutamine amidotransferase function. Replacement of the active-site cysteine in glutamine phosphoribosylpyrophosphate amidotransferase by site-directed mutagenesis

Site-directed mutagenesis was employed to replace cysteine 12 with phenylalanine in Bacillus subtilis glutamine phosphoribosylpyrophosphate amidotransferase (amidophosphoribosyltransferase). Glutamine-dependent amidophosphoribosyltransferase activity was abolished as a consequence of the mutation. The mutant enzyme, however, exhibited NH3-dependent activity, contained Fe-S, and was normally regulated by AMP. These results document the role of the active site cysteine in activation of glutamine for amide transfer. NH3-dependent amidophosphoribosyltransferase was utilized for de novo purine nucleotide synthesis. Cells containing the mutant enzyme grew at nearly the wild-type rate in media containing a high concentration of NH4Cl. The Phe-12 mutation was used to study NH2-terminal processing. Whereas the wild-type Cys-12 enzyme is processed correctly in Escherichia coli by removal of 11 amino acid residues from the NH2 terminus, the Phe-12 mutant enzyme was not subject to undecapeptide processing. Neither the mutant nor wild-type enzyme made in vitro was correctly processed. Alternative enzymatic and autocatalytic processing mechanisms were considered. The available evidence favors autocatalytic NH2-terminal undecapeptide processing.     

Dual role for the glutamine phosphoribosylpyrophosphate amidotransferase ammonia channel. Interdomain signaling and intermediate channeling

Glutamine phosphoribosylpyrophosphate (PRPP) amidotransferase catalyzes the first reaction of de novo purine nucleotide synthesis in two steps at two sites. Glutamine is hydrolyzed to glutamate plus NH(3) at an N-terminal glutaminase site, and NH(3) is transferred through a 20-A hydrophobic channel to a distal PRPP site for synthesis of phosphoribosylamine. Binding of PRPP is required to activate the glutaminase site (termed interdomain signaling) to prevent the wasteful hydrolysis of glutamine in the absence of phosphoribosylamine synthesis. Mutations were constructed to analyze the function of the NH(3) channel. In the wild type enzyme, NH(3) derived from glutamine hydrolysis was transferred to the PRPP site, and little or none was released. Replacement of Leu-415 at the PRPP end of the channel with an alanine resulted in a leaky channel and release of NH(3) to the solvent. Mutations in five amino acids that line the channel and two other residues required for the reorganization of phosphoribosyltransferase domain "flexible loop" that leads to formation of the channel perturbed channel function as well as interdomain signaling. The data emphasize the role of the NH(3) channel in coupling interdomain signaling and NH(3) transfer.     

Pea leaf glutamine synthetase: regulatory properties

Of a variety of purine and pyrimidine nucleotides tested, only ADP and 5'AMP significantly inhibited the Mg(2+)-dependent activity of pea leaf glutamine synthetase. They were less effective inhibitors where Mn(2+) replaced Mg(2+). They were competitive inhibitors with respect to ATP, with inhibition constant (Ki) values of 1.2 and 1.8 mm, respectively. The energy charge significantly affects the activity of glutamine synthetase, especially with Mg(2+). Of a variety of amino acids tested, l-histidine and l-ornithine were the most inhibitory, but significant inhibition was seen only where Mn(2+) was present. Both amino acids appeared to compete with l-glutamate, and the Ki values were 1.9 mm for l-histidine (pH 6.2) and 7.8 mm for l-ornithine (pH 6.2). l-Alanine, glycine, and l-serine caused slight inhibition (Mn(2+)-dependent activity) and were not competitive with ATP or l-glutamate.Carbamyl phosphate was an effective inhibitor only when Mn(2+) was present, and did not compete with substrates. Inorganic phosphate and pyrophosphate caused significant inhibition of the Mg(2+)-dependent activity.     

Glutamine metabolism regulation in Chlorella pyrenoidosa. Regulation of Chlorella glutamine synthetase activity by amino acids]

Effect of glutamine and its metabolites (amino acids) on Chlorella glutamine synthetase (GS) (E.C.6.3.1.2) in the presence of Mg or Mn was studied. Purified GS preparation was used, isolated from Chlorella grown in the presence of NH as a sole nitrogen source. Glutamate, aspartate, alanine and glycine inhibit GS activity in the presence of both Mg and Mn. Tryptophane and valine (up to 15 mM) activate GS in the presence of Mn. Tryptophane inhibits GS in the system with Mg. Sinergistic inhibition was observed under the combined effect of amino acids on GS in the presence of Mn and aspartate or alanine. The change of GS activity observed is supposed to be due to the inhibitory effect of glutamine and amino acids studied, since the glutamine content is increased (in 2.5 times for 5 min) and that of alanine and dicarbonic amino acids (for the following 15 min) under NH assimilation in Chlorella cells.     

Regulation of glutamine synthetase in the blue-green alga Anabaena L-31

In N2-grown cultures of Anabaena L-31, in which protein synthesis was prevented by chloramphenicol, presence of NH+4 caused a drastic decrease of glutamine synthetase (L-glutamate:ammonia ligase (ADP-forming), EC 6.3.1.2) activity indicating NH+4-mediated inactivation or degradation of the enzyme. The half-life of glutamine synthetase was more than 24 h, whereas that of nitrogenase (reduced ferredoxin:dinitrogen oxidoreductase (ATP-hydrolysing), EC 1.18.2.1) was less than 4 h, suggesting that glutamine synthetase may not act as positive regulator of nitrogenase synthesis in Anabaena. Glutamine synthetase purified to homogeneity was subject to cumulative inhibition by alanine, serine and glycine. The amino acids, however, exhibited partial antagonism in this behaviour. Glyoxylate, an intermediate in photorespiration, virtually prevented the amino acid inhibition. Kinetic studies revealed inhibition of the enzyme activity by high Mg2+ concentration under limiting glutamate level and by high glutamate in limiting Mg2+. Maximum enzyme activity occurred when the ratio of glutamate to free Mg2+ was 0.5 to 1.0. The results demonstrate that the enzyme is subject to multiple regulation by various metabolites involved in nitrogen assimilation.     

Metabolism of glutamine and glutamate by rat renal tubules. Study with 15N and gas chromatography-mass spectrometry

Gas chromatography-mass spectrometry was utilized to study the metabolism of [15N]glutamate, [2-15N]glutamine, and [5-15N]glutamine in isolated renal tubules prepared from control and chronically acidotic rats. The main purpose was to determine the nitrogen sources utilized by the kidney in various acid-base states for ammoniagenesis. Incubations were performed in the presence of 2.5 mM 15N-labeled glutamine or glutamate. Experiments with [5-15N]glutamine showed that in control animals approximately 90% of ammonia nitrogen was derived from 5-N of glutamine versus 60% in renal tubules from acidotic rats. Experiments with [2-15N]glutamine or [15N]glutamate indicated that in chronic acidosis approximately 30% of ammonia nitrogen was derived either from 2-N of glutamine or glutamate-N by the activity of glutamate dehydrogenase. Flux through glutamate dehydrogenase was 6-fold higher in chronic acidosis versus control. No 15NH3 could be detected in renal tubules from control rats when [2-15N]glutamine was the substrate. The rates of 15N transfer to other amino acids and to the 6-amino groups of the adenine nucleotides were significantly higher in normal renal tubules versus those from chronically acidotic rats. In tubules from chronically acidotic rats, 15N abundance in 15NH3 and the rate of 15NH3 appearance were significantly higher than that of either the 6-amino group of adenine nucleotides or the 15N-amino acids studied. The data indicate that glutamate dehydrogenase activity rather than glutamate transamination is primarily responsible for augmented ammoniagenesis in chronic acidosis. The contribution of the purine nucleotide cycle to ammonia formation appears to be unimportant in renal tubules from chronically acidotic rats.     

Structure-function relation of the NH2-terminal domain of the Semliki Forest virus capsid protein.

The capsid (C) protein of alphaviruses consists of two protein domains: a serine protease at the COOH terminus and an NH2-terminal domain which is thought to interact with RNA in the virus nucleocapsid (NC). The latter domain is very rich in positively charged amino acid residues. In this work, we have introduced large deletions into the corresponding region of a full-length cDNA clone of Semliki Forest virus, expressed the transcribed RNA in BHK-21 cells, and monitored the autoprotease activity of C, the formation of intracellular NCs, and the release of infectious virus. Our results show that if the gene region encoding the whole NH2-terminal domain is removed, the expressed C protein fragment cannot assemble into NCs and virus particles but it is still able to function as an autoprotease. Thus, these results underline the general importance of the NH2-terminal domain in the virus assembly process and furthermore show that the serine protease domain can function independently of the NH2 terminus. Surprisingly, analysis of additional C protein deletion variants showed that not all of the NH2-terminal domain is required for virus assembly, but large deletions involving up to one-third of its positively charged residues are still compatible with NC and virus formation. The fact that so much flexibility is allowed in the structure of the NH2-terminal domain of C suggests that most of this region is involved in nonspecific interactions with the encapsidated RNA, probably through its positively charged amino acid residues.     

o-Phosphotyrosyl glutamine synthetase: modification of the nucleotide ligation site of adenylylated glutamine synthetase

The nucleotide ligation site of adenylylated glutamine synthetase, which contains a unique tyrosyl residue linked through a phosphodiester bond to 5'-AMP, was studied by digestion with three hydrolytic enzymes. The products on micrococcal nuclease digestion were adenosine and o-phosphotyrosyl glutamine synthetase. The Km for this macromolecular substrate with the nuclease was 40 microM, at pH 8.9. The glutamine synthetase activity was not affected by deadenosylation with the nuclease, in contrast to SVPDE digestion, with which the glutamine synthetase activity was markedly increased. The Km for the native adenylylated glutamine synthetase with the SVPDE was 36 microM, i.e., similar to that for the nuclease. When the isolated o-phosphotyrosyl enzyme was incubated with alkaline phosphatase at pH 7.2, the glutamine synthetase activity rapidly increased to the same level as that of the SVPDE treated enzyme. Furthermore, kinetic properties of the o-phosphotyrosyl glutamine synthetase were compared with those of the adenylylated enzyme. The optimum pH, apparent Km for each of three substrates, glutamate, ATP, and NH3, and Vmax were in good agreement, as to either Mg2+- or Mn2+-dependent biosynthetic activity. From these results we can conclude that the regulation of glutamine synthetase activity simply requires the phosphorylation of the tyrosyl residue in each subunit, without recourse to adenylylation.     

Glucosamine-6-phosphate synthase from Escherichia coli yields two proteins upon limited proteolysis: identification of the glutamine amidohydrolase and 2R ketose/aldose isomerase-bearing domains based on their biochemical properties

The proteolysis of native glucosamine-6-phosphate synthase (Mr 67,000) from Escherichia coli was investigated using two nonspecific and five specific endoproteinases, alpha-chymotrypsin generated two nonoverlapping polypeptides CT1 and CT2 of Mr 40,000 and 27,000 lacking glucosamine-6P synthesizing activity. Amino terminal and carboxy terminal sequence analysis showed that cleavage occurred between positions 240 and 241 of the primary sequence without further degradation. The glutamine amidohydrolase activity was located in the CT2 N-terminal polypeptide which was capable of incorporating 0.7 equivalent of the glutamine site-directed affinity label [2-3H]-N3-(4-methoxyfumaroyl)-diaminopropionic acid indicating that it bears the amidotransferase function. CT1 which displayed a higher reactivity than CT2 for fructose-6P binding contains the ketose/aldose isomerase activity. These data suggest the existence of a hinge structure essential for the catalytically efficient coupling between the ammonia generating domain and the sugar binding domain and support the model recently proposed by Mei and Zalkin in which purF-type amidotransferases contain a glutamine hydrolase domain of approximately 200 amino acids fused to an ammonia-transfer domain.     

Muscle glutamine production in diabetic ketoacidotic rats.

The mechanism of activation of glutamine production by the hindlimb during diabetic ketoacidosis (DKA) was investigated in rats. Muscle glutamine production was estimated to account for over 90% of the total glutamine produced by the hindlimb. DKA produced significant increases in the concentrations of NH4+ and IMP in hindlimb muscles, suggesting that AMP deaminase is activated by DKA. NH4Cl- and HCl-induced acidosis did not produce these changes, indicating either that acidosis itself is not the stimulus for increased AMP deaminase activity or that the more severe degree of acidosis accompanying DKA is necessary for activation. Muscle glutamine concentrations were depressed in DKA. Experiments with isolated epitrochlearis muscle showed that the transport and permeability properties of the muscle cells (as judged by uptake and release of alpha-aminoisobutyrate and glutamine) were not altered by DKA. However, glutamine uptake by muscle cells was significantly inhibited by L-leucine, the concentration of which, along with other branched-chain amino acids, is markedly elevated in DKA.     

Identification of the reactive cysteinyl residue and ATP binding site in Bacillus cereus glutamine synthetase by chemical modification

Bacillus cereus glutamine synthetase was modified by reaction with a fluorescent SH reagent, N-[[(iodoacetyl)amino]ethyl]-5-naphthylamine-1-sulfonic acid (IAEDANS), or an ATP analog, 5'-p-fluorosulfonylbenzoyladenosine (FSBA). The locations of the specific binding sites of these reagents were identified. IAEDANS inactivated Mg2(+)-dependent activity and activated Mn2(+)-dependent activity. FSBA inactivated only Mn2(+)-dependent activity. Mg2+ plus Mn2(+)-dependent activity was inactivated by IAEDANS or FSBA. Amino acid sequence analysis of the single AEDANS-labeled proteolytic fragment showed the cysteinyl residue at position 306 to be the site of modification. Cys 306 is one of three cysteines that are unique to Bacillus glutamine synthetase. The result suggested that the cysteine has a role in the active site of the enzyme. We also report that the amino acid residue modified by FSBA was the lysyl residue at position 43.     

Formation of alanine and glutamine in chick (Gallus domesticus) skeletal muscle

1. Incubation of extensor digitorum communis muscles from fed chicks in the presence of plasma concentrations of branched-chain amino acids (BCAA) increased the formation of glutamate, glutamine and alanine. This effect was inhibited by 1.5 mM L-cycloserine. 2. 2-Oxoisocaproate (0.1 and 0.5 mM) increased the formation of leucine but decreased that of glutamate, glutamine and alanine. 3. NH4Cl (0.3 mM) increased the formation of glutamine, and decreased the release and intracellular concentrations of glutamate and alanine. 4. Our results demonstrate that alanine and glutamine are synthesized de novo in chick skeletal muscle and demonstrate the similarity in alanine and glutamine synthesis in skeletal muscle between the domestic fowl and mammals.     

Domain structure in actin-binding proteins: expression and functional characterization of truncated severin

Severin from Dictyostelium discoideum is a Ca2(+)-activated actin-binding protein that severs actin filaments, nucleates actin assembly, and caps the fast growing ends of actin filaments. Sequence comparison with functionally related proteins, such as gelsolin, villin, or fragmin revealed highly conserved domains which are thought to be of functional significance. To attribute the different activities of the severin molecule to defined regions, progressively truncated severin polypeptides were constructed. The complete cDNA coding for 362 (DS362) amino acids and five 3' deletions coding for 277 (DS277), 177 (DS177), 151 (DS151), 117 (DS117), or 111 (DS111) amino acids were expressed in Escherichia coli. The proteins were purified to homogeneity and then characterized with respect to their effects on the polymerization or depolymerization kinetics of G- or F-actin solutions and their binding to G-actin. Furthermore, the Ca2+ binding of these proteins was investigated with a 45Ca-overlay assay and by monitoring Ca2(+)-dependent changes in tryptophan fluorescence. Bacterially expressed DS362 showed the same Ca2(+)-dependent activities as native severin. DS277, missing the 85 COOH-terminal amino acids of severin, had lost its strict Ca2+ regulation and displayed a Ca2(+)-independent capping activity, but was still Ca2+ dependent in its severing and nucleating activities. DS151 which corresponded to the first domain of gelsolin or villin had completely lost severing and nucleating properties. However, a residual severing activity of approximately 2% was detectable if 26 amino acids more were present at the COOH-terminal end (DS177). This locates similar to gelsolin the second actin-binding site to the border region between the first and second domain. Measuring the fluorescence enhancement of pyrene-labeled G-actin in the presence of DS111 showed that the first actin-binding site was present in the NH2-terminal 111 amino acids. Extension by six or more amino acids stabilized this actin-binding site in such a way that DS117 and even more pronounced DS151 became Ca2(+)-independent capping proteins. In comparison to many reports on gelsolin we draw the following conclusions. Among the three active actin-binding sites in gelsolin the closely neighboured sites one and two share the F-actin fragmenting function, whereas the actin-binding sites two and three, which are located in far distant domains, collaborate for nucleation. In contrast, severin contains two active actin-binding sites which are next to each other and are responsible for the severing as well as the nucleating function. The single actin-binding site near the NH2-terminus is sufficient for capping of actin filaments.     

The involvement of glutamine synthetase/glutamate synthase in ammonia assimilation by Aspergillus nidulans

Wild-type Aspergillus nidulans grew equally well on NH4Cl, KNO3 or glutamine as the only nitrogen source. NADP+-dependent glutamate dehydrogenase (EC 1.4.1.4) and glutamine synthetase (GS; EC 6.3.1.2) activities varied with the type and concentration of nitrogen source supplied. Glutamate synthase (GOGAT) activity (EC 1.4.7.1) was detected but it was almost unaffected by the type and concentration of nitrogen source supplied. Ion exchange chromatography showed that the GOGAT activity was due to a distinct enzyme. Azaserine, an inhibitor of the GOGAT reaction, reduced the glutamate pool by 60%, indicating that GOGAT is involved in ammonia assimilation by metabolizing the glutamine formed by GS.     

Formation of glutamine from [13n]ammonia, [13n]dinitrogen, and [14C]glutamate by heterocysts isolated from Anabaena cylindrica.

A method is described for the isolation of metabolically active heterocysts from Anabaena cylindrica. These isolated heterocysts accounted for up to 34% of the acetylene-reducing activity of whole filaments and had a specific activity of up to 1,560 nmol of C2H4 formed per mg of heterocyst chlorphyll per min. Activity of glutamine synthetase was coupled to activity of nitrogenase in isolated heterocysts as shown by acetylene-inhibitable formation of [13N]NH3 and of amidelabeled [13N]glutamine form [13N]N2. A method is also described for the production of 6-mCi amounts of [13N]NH3. Isolated heterocysts formed [13N]glutamine from [13N]NH3 and glutamate, and [14C]glutamine from NH3 and [14C]glutamate, in the presence of magnesium adenosine 5'-triphosphate. Methionine sulfoximine strongly inhibited these syntheses. Glutamate synthase is, after nitrogenase and glutamine synthetase, the third sequential enzyme involved in the assimilation of N2 by intact filaments. However, the kinetics of solubilization of the activity of glutamate synthase during cavitation of suspensions of A. cylindrica indicated that very little, if any, of the activity of that enzyme was located in heterocysts. Concordantly, isolated heterocysts failed to form substantial amounts of radioactive glutamate from either [13N]glutamine or alph-[14C]ketoglutarate in the presence of other substrates and cofactors of the glutamate synthase reaction. However, they formed [14C]glutamate rapidly from alpha-[14C]ketoglutarate by aminotransferase reactions, with various amino acids as the nitrogen donor. The implication of these findings with regard to the identities of the substances moving between heterocysts and vegetative cells are discussed.