Metabolomics of Escherichia coli Treated with the Antimicrobial Carbon Monoxide-Releasing Molecule CORM-3 Reveals Tricarboxylic Acid Cycle as Major Target.

In the last decade, carbon monoxide-releasing molecules (CORMs) have been shown to act against several pathogens and to be promising antimicrobials. However, the understanding of the mode of action and reactivity of these compounds on bacterial cells is still deficient. In this work, we used a metabolomics approach to probe the toxicity of the ruthenium(II) complex Ru(CO)3Cl(glycinate) (CORM-3) on Escherichia coli By resorting to 1H nuclear magnetic resonance, mass spectrometry, and enzymatic activities, we show that CORM-3-treated E. coli accumulates larger amounts of glycolytic intermediates, independently of the oxygen growth conditions. The work provides several evidences that CORM-3 inhibits glutamate synthesis and the iron-sulfur enzymes of the tricarboxylic acid (TCA) cycle and that the glycolysis pathway is triggered in order to establish an energy and redox homeostasis balance. Accordingly, supplementation of the growth medium with fumarate, α-ketoglutarate, glutamate, and amino acids cancels the toxicity of CORM-3. Importantly, inhibition of the iron-sulfur enzymes glutamate synthase, aconitase, and fumarase is only observed for compounds that liberate carbon monoxide. Altogether, this work reveals that the antimicrobial action of CORM-3 results from intracellular glutamate deficiency and inhibition of nitrogen and TCA cycles.

Glutamate synthase, but not GABA shunt enzymes, contributes to nitrogen metabolism of the sheep abomasal nematode parasites Haemonchus contortus and Teladorsagia circumcincta.

Glutamate synthase (E.C. 1.4.1.14) (GOGAT) activity was not detectable in L3 Haemonchus contortus, but was present in L3 Teladorsagia circumcincta and adult worms of both species. GOGAT activity was inhibited by 80% by azaserine. Activity (nmol min(-1) mg(-1) protein) was 33-59 in adult H. contortus, 51-91 in adult T. circumcincta and 24-41 in L3 T. circumcincta, probably depending on exposure to ammonia, as incubation with 1mM NH(4)Cl doubled GOGAT activity. The pH optimum was 7.5 in both species. Either NAD or NADP acted as co-factor. The mean apparent K(m) for 2-oxoglutarate was 0.7 (0.5-0.9) mM and for glutamine was 1.0 (0.5-1.7) mM for different homogenates. There was no detectable activity in whole parasite homogenates of glutamate decarboxylase (E.C. 4.1.1.15) or succinic semialdehyde dehydrogenase (E.C. 1.2.1.24), the first and third enzymes of the GABA shunt, respectively, suggesting that the GABA shunt is not important in general metabolism in these species.

Short- and long-term enzymatic regulation secondary to metabolic insult in the rat retina.

Changes in oxygen and/or glucose availability may result in altered levels of ATP production and amino acid levels, and alteration in lactic acid production. However, under certain metabolic insults, the retina demonstrates considerable resilience and maintains ATP production, and/or retinal function. We wanted to investigate whether this resilience would be reflected in alterations in the activity of key enzymes of retinal metabolism, or enzymes associated with amino acid production that may supply their carbon skeleton for energy production. Enzymatic assays were conducted to determine the activity of key retinal metabolic enzymes total ATPase and Na(+)/K(+)-ATPase, aspartate aminotransferase and lactate dehydrogenase. In vitro anoxia led to an increase in retinal lactate dehydrogenase activity and to a decrease in retinal aspartate aminotransferase activity, without significant changes in Na(+)/K(+)-ATPase activity. In vivo inhibition of glutamine synthetase resulted in a short-term significant decrease in retinal aspartate aminotransferase activity. An increase in retinal aspartate aminotransferase and lactate dehydrogenase activities was accompanied by altered levels of amino acids in neurons and glia after partial inhibition of glial metabolism, implying that short- and long-term up- and down-regulation of key metabolic enzymes occurs to supply carbon skeletons for retinal metabolism. ATPase activity does not appear to fluctuate under the metabolic stresses employed in our experimental procedures.

Synthesis and biological evaluation of new amino acids structurally related to the antitumor agent acivicin.

A set of racemic conformationally constrained analogues of the antitumor antibiotic acivicin (+)-1 has been prepared through a strategy based on 1,3-dipolar cycloaddition of bromonitrile oxide to suitable dipolarophiles. The bromo analogue (2) of acivicin was also synthesized and tested as a reference compound, together with its stereoisomer 3. The antitumor properties of novel amino acids 4-7 were evaluated in vitro against human tumor cell lines. Their efficacy to inhibit glutamate synthase (GltS) from Azospirillum brasilense was also assayed. None of the studied compounds, but 2, showed significant activity.

Regulation of amino acid uptake in conifers by exogenous and endogenous nitrogen.

Although an accumulating amount of research clearly indicates that plants are capable of taking up exogenous amino acids, the actual importance of such organic N sources for plant N nutrition is under debate. In this study, we show that amino acid uptake by Scots pine (Pinus sylvestris L.) is significantly decreased by elevated internal NH(4)(+) levels, while it increases following exposure to exogenous amino acids. Furthermore, amino acid uptake is larger in N-deficient plants than in plants grown with a large access of N. The regulatory pattern of amino acid uptake shows important similarities to the regulation of NO(3)(-) and NH(4)(+) transport as well as to the regulation of yeast amino acid transporters. In addition, our data suggest that uptake may be regulated by factors not originating from N metabolism. The up-regulation of uptake in response to N deficiency suggests that amino acid uptake may be a significant contributor to the N economy of P. sylvestris.

Anaerobic accumulation of amino acids in rice roots: role of the glutamine synthetase/glutamate synthase cycle.

Accumulation of amino acids was studied in rice roots of 3-day-old seedlings subjected for 48h to anaerobic conditions. Alanine and Gaba were the main amino acids accumulated under anoxia. Their synthesis was strongly inhibited by MSX and AZA, inhibitors of glutamine synthetase and glutamate synthase. These activities increased after 8h of anaerobic treatment and, by immunoprecipitation of 35S-labeled proteins, it was shown that glutamine synthetase and ferredoxin-dependent glutamate synthase were synthesized during the treatment. These findings indicate that the glutamine synthetase/glutamate synthase cycle play an important role in anaerobic amino acid accumulation.

Accumulation of glutamate by Salmonella typhimurium in response to osmotic stress.

Salmonella typhimurium accumulates glutamate in response to osmotic stress. Cells in aerobic exponential growth have an intracellular pool of approximately 125 nmol of glutamate mg of protein-1. When cells were grown in minimal medium with 500 mM NaCl, KCl, or sucrose, 290 to 430 nmol of glutamate was found to accumulate. Values were lower when cells were harvested in stationary phase. Cells were grown in conventional medium, harvested, washed, resuspended in the control medium or in medium with osmolytes, and aerated for 1 h. With aeration, glutamate was found to accumulate at levels comparable to those observed in exponential cultures. Antibiotics inhibiting protein synthesis did not affect glutamate accumulation when cells were aerated. Strains with mutations in glutamate synthase (glt) or in glutamate dehydrogenase (gdh) accumulated nearly normal levels of glutamate under these conditions. A double (gdh glt) mutant accumulated much less glutamate (63.9 nmol mg of protein-1), but a 1.9-fold excess accumulated when cells were aerated with osmotic stress. Methionine sulfone, an inhibitor of glutamate synthase, did not prevent accumulation of glutamate in cells aerated with osmotic stress. Glutamate dehydrogenase is thought to have minimum activity when ammonium is limiting. Resuspending cells with limiting ammonium reduced glutamate production but did not eliminate accumulation of excess glutamate when cells were osmotically stressed. Amino oxyacetic acid, an inhibitor of transamination reactions, did not prevent accumulation of excess glutamate.(ABSTRACT TRUNCATED AT 250 WORDS)

Novel 6,5-fused ring heterocyclic antifolates: biochemical and biological characterization.

Six novel antifolates with 2,4-diaminopyrimidine-fused five-membered rings containing either pyrrole or cyclopentene rings were characterized at the cellular and biochemical level. Five of these antifolates were more growth inhibitory to the CCRF-CEM human leukemia cell line than methotrexate [MTX; drug concentration effective at inhibiting cell growth by 50% relative to untreated control (EC50), 12 nM], the antifolate used in the clinic, and two were more potent than 10-ethyl-10-deazaaminopterin (EC50, 2.7 nM); similar patterns of response were obtained in the FaDu and A253 squamous carcinoma cell lines. In addition, the growth inhibitory potency of these antifolates was generally less dependent on exposure time than was MTX. Growth inhibitory effects could be reversed by leucovorin, indicating an antifolate mechanism. These antifolates targeted dihydrofolate reductase (DHFR) based on direct human DHFR inhibition assays [drug concentration inhibiting enzyme activity by 50% (IC50), 0.6-28 nM; MTX IC50, 0.8 nM] and the cross-resistance of MTX-resistant CCRF-CEM cells containing elevated DHFR. Inhibition of human thymidylate synthase was generally weak. These 6,5-fused ring heterocyclic antifolates utilized the reduced folate/MTX transporter for uptake, based on the cross-resistance of MTX uptake-impaired CCRF-CEM cells, and were efficient substrates for this uptake system, based on inhibition of [3H]MTX uptake (IC50, 0.3-5.8 microM; aminopterin IC50, 2.6 microM). These analogues were substrates for CCRF-CEM folylpolyglutamate synthetase, with several being among the most active substrates now known (highest Vrel/Km 0.73; MTX and 10-ethyl-10-deazaaminopterin, 0.013 and 0.24, respectively). Substrate activity for murine intestinal folylpolyglutamate synthetase was also assayed, and a different specificity pattern was observed. These new antifolates are apparently not substrates for aldehyde oxidase. Analogues containing the fused cyclopentene ring are preferred to those containing the fused pyrrole ring based on growth inhibitory potency, effectiveness against decreased uptake mutants and apparent affinity for transport, and inhibition of DHFR. In addition, fused cyclopentene-containing analogues are efficiently polyglutamylated. The data indicate that antifolates with 2,4-diaminopyrimidine-fused five-membered rings, especially those containing the fused cyclopentene ring, are an important new class of antifolates which warrant further exploration at the synthetic and preclinical levels.

The pH-dependent behavior of catalytic activities of Azospirillum brasilense glutamate synthase and iodoacetamide modification of the enzyme provide evidence for a catalytic Cys-His ion pair.

The pH dependence of the kinetic parameters of the glutamine- and ammonia-dependent reactions of Azospirillum brasilense glutamate synthase revealed the presence of ionizable groups with pKa values between 6 and 10 involved in the binding of the substrates and in catalytic steps. The V profile of the glutamine-dependent reaction is complicated by a deviation from a simple bell-shaped curve between pH 8 and pH 10, which may suggest that deprotonation of a group with pKa value in this region decreases but does not abolish glutamine-dependent enzyme activity. This group does not seem to be required in the ammonia-dependent reaction of GltS, which decreases on the acidic and alkaline sides as groups with pKa values of about 8.8 and 9.9 dissociate. The V/K profile for ammonia exhibits a single pKa value of about 8.7, suggesting that ammonia is the actual substrate of the enzyme, and that ammonia binding to glutamate synthase is largely pH independent. The hypothesis that a group with pKa between 8 and 10 is involved in the glutaminase segment of the glutamine-dependent glutamate synthase activity was supported by studies of the modification of the enzyme by 6-diazo-5-oxo-L-norleucine, a glutamine analog, and iodoacetamide, a cysteine-directed reagent. Analyses of the kinetics of inactivation of the enzyme in the presence and absence of enzyme substrates and their analogs at different pH values demonstrated that iodoacetamide reacts with a group involved in glutamine binding and/or activation, most likely the cysteine residue at the N-terminus of glutamate synthase alpha subunit, which may form a Cys-His ion pair in the active site of glutamate synthase, as suggested for other amidotransferases (Mei, B., and Zalkin, H. (1989) J. Biol. Chem. 264, 16613-16619).

Some properties of glutamate dehydrogenase, glutamine synthetase and glutamate synthase from Corynebacterium callunae.

Characteristics of the three major ammonia assimilatory enzymes, glutamate dehydrogenase (GDH), glutamine synthetase (GS) and glutamate synthase (GO-GAT) in Corynebacterium callunae (NCIB 10338) were examined. The GDH of C. callunae specifically required NADPH and NADP+ as coenzymes in the amination and deamination reactions, respectively. This enzyme showed a marked specificity for alpha-ketoglutarate and glutamate as substrates. The optimum pH was 7.2 for NADPH-GDH activity (amination) and 9.0 for NADP(+)-GDH activity (deamination). The results showed that NADPH-GDH and NADP(+)-GDH activities were controlled primarily by product inhibition and that the feedback effectors alanine and valine played a minor role in the control of NADPH-GDH activity. The transferase activity of GS was dependent on Mn+2 while the biosynthetic activity of the enzyme was dependent on Mg2+ as essential activators. The pH optima for transferase and biosynthetic activities were 8.0 and 7.0, respectively. In the transfer reaction, the Km values were 15.2 mM for glutamine, 1.46 mM for hydroxylamine, 3.5 x 10(-3) mM for ADP and 1.03 mM for arsenate. Feedback inhibition by alanine, glycine and serine was also found to play an important role in controlling GS activity. In addition, the enzyme activity was sensitive to ATP. The transferase activity of the enzyme was responsive to ionic strength as well as the specific monovalent cation present. GOGAT of C. callunae utilized either NADPH or NADH as coenzymes, although the latter was less effective. The enzyme specifically required alpha-ketoglutarate and glutamine as substrates.(ABSTRACT TRUNCATED AT 250 WORDS)

The kinetic mechanism of the reactions catalyzed by the glutamate synthase from Azospirillum brasilense.

The reactions catalyzed by glutamate synthase from Azospirillum brasilense have been investigated by a combination of absorption spectroscopy, steady-state kinetic measurements and experiments with stereospecifically labelled substrate. The data show that both L-glutamine-dependent and ammonia-dependent reactions of the glutamate synthase from A. brasilense follow an identical two-site uni-uni bi-bi kinetic mechanism, in which the enzyme is alternately reduced by NADPH and oxidized by the iminoglutarate formed on addition of ammonia to the C2 of 2-oxoglutarate. The spectroscopic experiments support the involvement of the enzyme chromophores (flavins and iron-sulfur centers) in both reactions. Finally, using stereospecifically labelled NADPH, we showed that the enzyme from Azospirillum is specific for the transfer of the 4S hydrogen of NADPH. During the catalysis of both L-glutamine-dependent and ammonia-dependent reactions, this hydrogen atom equilibrates with the solvent. The data obtained with glutamate synthase from A. brasilense, a diazotroph, differ significantly from those regarding the ammonia-dependent reaction of other glutamate synthases. The ammonia-dependent activity of glutamate synthase from Azospirillum is not physiologically significant, representing only a segment of the overall physiological L-glutamine-dependent activity and requiring the enzyme flavins and iron-sulfur centers. Finally, the data are not consistent with the hypothesis [Geary, L. E. & Meister, A. (1977) J. Biol. Chem. 252, 3501-3508] that the small subunit of glutamate synthase is endowed with a glutamate-dehydrogenase-like activity.

Ammonia assimilation by Aspergillus nidulans: [15N]ammonia study.

15N kinetic labelling studies were done on liquid cultures of wild-type Aspergillus nidulans. The labelling pattern of major amino acids under 'steady state' conditions suggests that glutamate and glutamine-amide are the early products of ammonia assimilation in A. nidulans. In the presence of phosphinothricin, an inhibitor or glutamine synthetase, 15N labelling of glutamate, alanine and aspartate was maintained whereas the labelling of glutamine was low. This pattern of labelling is consistent with ammonia assimilation into glutamate via the glutamate dehydrogenase pathway. In the presence of azaserine, an inhibitor of glutamate synthase, glutamate was initially more highly labelled than any other amino acid, whereas its concentration declined. Isotope also accumulated in glutamine. Observations with these two inhibitors suggest that ammonia assimilation can occur concurrently via the glutamine synthetase/glutamate synthase and the glutamate dehydrogenase pathways in low-ammonia-grown A. nidulans. From a simple model it was estimated that about half of the glutamate was synthesized via the glutamate dehydrogenase pathway; the other half was formed from glutamine via the glutamate synthase pathway. The transfer coefficients of nine other amino acids were also determined.

Regulation of carbon and nitrogen flow by glutamate synthase in Neurospora crassa.

A glycine-resistant Neurospora crassa mutant (am-132;glyr), derived from the am-132 mutant, was isolated and characterized. [am-132 itself has a deletion in the structural gene for NADP-dependent glutamate dehydrogenase (GDH).] This new mutation also conferred resistance to serine and methionine sulphoximine (MS), which are inhibitors of glutamine synthetase (GS). In addition, the mutant obtained grew better on ammonium than the am-132 parental strain. Resistance to glycine was not due to increased synthesis of glutamine by an altered or induced GS, nor to increased glutamate synthesis by induction of the catabolic NAD-dependent GDH, nor to NADH-dependent glutamate synthase (GOGAT), which was as sensitive to inhibitors as the GOGAT from the parental strain. The glycine-resistance mutation lowered but did not abolish the carbon flow; this resulted in a lower content of tricarboxylic acid cycle intermediates. GOGAT activity was inhibited in vitro by several organic acids and methionine sulphone (MSF). The higher growth rate of the glycine-resistant mutant on ammonium or on ammonium plus glycine, serine or MS was explained by an increased capacity of GOGAT to synthesize glutamate in vivo due to a lower content of inhibitory tricarboxylic acid cycle intermediates; the higher glutamate content overcomes the effect of the GS inhibitors and explains the MSF resistance of the mutant.

Glutamate synthase from Bacillus subtilis PCI 219.

Reduced pyridine nucleotide dependent glutamate synthase [L-glutamate: NADP+ oxidoreductase (transaminating); EC 1.4.1.13] was purified to homogeneity from Bacillus subtilis PCI 219. The molecular weight of the enzyme was 210,000, and the enzyme was composed of two nonidentical subunits with molecular weights of 160,000 and 56,000. The absorption and CD spectra of the enzyme indicated that the enzyme is an iron-sulfur flavoprotein. The enzyme was found to contain 1:1:7.4:8.7 mol of FMN, FAD, iron atoms, and acid-labile sulfur atoms per mol (MW 210,000). EPR measurements of the NADPH-reduced enzyme at 77K revealed the formation of a stable flavin semiquinone intermediate; however, none of the signals originating from the iron-sulfur cluster was observed. Still at 4.2K the EPR signals in the region of g = 2, which may originate from the paramagnetic iron-sulfur cluster, were clearly observed for both the isolated and dithionite-reduced states of the enzyme. The enzyme exhibited a wide coenzyme specificity, and either NADPH or NADH could be used as electron donor, although the latter was less effective. The enzyme activity was also expressed when ammonium chloride was substituted for L-glutamine. The optimum pHs for NADPH-Gln-, NADH-Gln-, and NADPH-NH3-dependent reactions were 7.8, 6.9, and 9.4, respectively. The apoenzyme exhibited substantial inactivation of the Gln-dependent activities but still retained the NH3-dependent activities. Enzyme reduction-oxidation experiments, initial velocity experiments, and product inhibition patterns revealed that both the NADPH-Gln- and NADH-Gln-dependent reactions coincided with the two-site ping-pong uni-uni bi-bi kinetic mechanism, while the NADPH-NH3-dependent reaction deviated from Michaelis-Menten kinetics. The Gln-dependent activities were inhibited by several TCA cycle members, especially L-malate and fumarate, as well as L-methionine-SR-sulfoximine, pyridoxal-5'-phosphate, and pCMB. The regulation of the glutamate synthase, glutamine synthetase [EC 6.3.1.2], and glutamate dehydrogenase [EC 1.4.1.3] activities was examined with cultures of cells grown with various nitrogen and carbon sources.

Inhibition of nitrate utilization by amino acids in intact Anacystis nidulans cells.

In an attempt to establish the nature of the ammonium-assimilation products which mediate the inhibition by ammonium of nitrate uptake in cyanobacteria, the effect of different amino acids on nitrate utilization by intact Anacystis nidulans cells has been assayed. To exclude an indirect inhibition of nitrate uptake through the ammonium which the amino acids might release, the cells were pretreated with L-methionine-D,L-sulfoximine (MSX), a potent inactivator of glutamine synthetase. Under these conditions, several L-amino acids, but not the corresponding D-isomers, affected nitrate utilization to a variable extent, causing inhibitions ranging between 20 and 80% when added at 20 mM concentration. For most of the inhibitory amino acids, including L-isoleucine, L-leucine and L-valine, a correlation was found between their ability to act as amino group donors to alpha-ketoglutarate, in reactions catalyzed by A. nidulans cell-free extracts, and their inhibitory effect on nitrate utilization. L-Glutamine, L-asparagine and glycine, being effective inhibitors of nitrate utilization, were poor substrates for the transaminating activity to alpha-ketoglutarate, however. The possible role of the latter amino acids as mediators in the ammonium-promoted inhibition of nitrate uptake is discussed.

Inhibition of homocysteine sulfonamide of glutamate synthase purified from Saccharomyces cerevisiae.

Glutamate synthase, isolated in apparently homogeneous form (Mr approximately 265,000) from Saccharomyces cerevisiae after 7500-fold purification, is markedly inhibited by homocysteine sulfonamide. Inhibitions competitive with respect to L-glutamine; the apparent Ki value calculated for L-homocysteine sulfonamide is 3.6 microM; the apparent Km value for L-glutamine is 280 microM. The very high affinity of the inhibitor for the enzyme, as well as structural considerations, suggest that homocysteine sulfonamide is a transition state inhibitor. The previously reported growth inhibitory properties of homocysteine sulfonamide (Reisner, D. B. (1958) J. Am. Chem. soc. 78, 5102-5104) may be due, at least in part, to inhibition of glutamate synthase. L-Methionine sulfone is also a potent competitive inhibitor, whereas L-albizziin, L-methionine-SR-sulfoximine, and L-methionine-SR-sulfoxide are much less effective inhibitors. S. cerevisiae glutamate synthase, which is composed of two dissimilar subunits, uses NADH exclusively, and exhibits low but definite activity when NH3 is substituted for glutamine.

Kinetic mechanism of NADH-dependent glutamate synthase from lupin nodules.

From initial-rate studies, a partially random kinetic mechanism has been deduced for NADH-dependent glutamate synthase from lupin nodules. The mechanism involves compulsory binding of NADH as first substrate, followed by random-order binding of glutamine and 2-oxoglutarate. Patterns of inhibition by glutamate substantiate the mechanism. Dithionite was incapable of acting as an alternative reducing substrate although it is known to reduce the flavine groups of the enzyme. The implications of these results are discussed. Published rate equations for this type of mechanism were found to be unsatisfactory for this enzyme and suitable new equations are produced. These equations should have general application where the obligatory first substrate binds very tightly.

Inactivation of glutamate dehydrogenase and glutamate synthase from Bacillus megaterium by phenylglyoxal, butane-2,3-dione and pyridoxal 5'-phosphate.

Reaction of phenylglyoxal with glutamate dehydrogenase (EC 1.4.1.4), but not with glutamate synthase (EC 2.6.1.53), from Bacillus megaterium resulted in complete loss of enzyme activity. NADPH alone or together with 2-oxoglutarate provided substantial protection from inactivation by phenylglyoxal. Some 2mol of [14C]Phenylglyoxal was incorporated/mol of subunit of glutamate dehydrogenase. Addition of 1mM-NADPH decreased incorporation by 0.7mol. The Ki for phenylglyoxal was 6.7mM and Ks for competition with NADPH was 0.5mM. Complete inactivation of glutamate dehydrogenase by butane-2,3-dione was estimated by extrapolation to result from the loss of 3 of the 19 arginine residues/subunit. NADPH, but not NADH, provided almost complete protection against inactivation. Butane-2,3-dione had only a slight inactivating effect on glutamate synthase. The data suggest that an essential arginine residue may be involved in the binding of NADPH to glutamate dehydrogenase. The enzymes were inactivated by pyridoxal 5'-phosphate and this inactivation increased 3--4-fold in the borate buffer. NADPH completely prevented inactivation by pyridoxal 5'-phosphate.

Enzymes of nitrogen metabolism in legume nodules. Purification and properties of NADH-dependent glutamate synthase from lupin nodules.

An NADH-dependent glutamate synthase has been purified 500-fold from the plant cytoplasm fraction of Lupinus angustifolius nodules. It consists of a single polypeptide chain, Mr 235000. The optimum pH is 8.5, at which Km values for 2-oxoglutarate, glutamine and NADH are 39 micrometer, 400 micrometer and 1.3 micrometer respectively. The catalytic centre activity is of the order of 70 s-1 and is independent of pH between 6.5 and 9.5. Glutamate synthase is inhibited by glutamic acid, oxaloacetic acid, aspartic acid and asparagine, all competitive with 2-oxoglutarate; and by NAD+, which is competitive with NADH. There is evidence of two flavine prosthetic groups per enzyme molecule.