The unexpected structural role of glutamate synthase [4Fe-4S](+1,+2) clusters as demonstrated by site-directed mutagenesis of conserved C residues at the N-terminus of the enzyme beta subunit

Azospirillum brasilense glutamate synthase (GltS) is a complex iron-sulfur flavoprotein whose catalytically active alphabeta protomer (alpha subunit, 162kDa; beta subunit, 52.3 kDa) contains one FAD, one FMN, one [3Fe-4S](0,+1), and two [4Fe-4S](+1,+2) clusters. The structure of the alpha subunit has been determined providing information on the mechanism of ammonia transfer from L-glutamine to 2-oxoglutarate through a 30 A-long intramolecular tunnel. On the contrary, details of the electron transfer pathway from NADPH to the postulated 2-iminoglutarate intermediate through the enzyme flavin co-factors and [Fe-S] clusters are largely indirect. To identify the location and role of each one of the GltS [4Fe-4S] clusters, we individually substituted the four cysteinyl residues forming the first of two conserved C-rich regions at the N-terminus of GltS beta subunit with alanyl residues. The engineered genes encoding the beta subunit variants (and derivatives carrying C-terminal His6-tags) were co-expressed with the wild-type alpha subunit gene. In all cases the C/A substitutions prevented alpha and beta subunits association to yield the GltS alphabeta protomer. This result is consistent with the fact that these residues are responsible for the formation of glutamate synthase [4Fe-4S](+1,+2) clusters within the N-terminal region of the beta subunit, and that these clusters are implicated not only in electron transfer between the GltS flavins, but also in alphabeta heterodimer formation by structuring an N-terminal [Fe-S] beta subunit interface subdomain, as suggested by the three-dimensional structure of dihydropyrimidine dehydrogenase, an enzyme containing an N-terminal beta subunit-like domain.     

Structure-function studies of glutamate synthases: a class of self-regulated iron-sulfur flavoenzymes essential for nitrogen assimilation

Glutamate synthases play with glutamine synthetase an essential role in nitrogen assimilation processes in microorganisms, plants, and lower animals by catalyzing the net synthesis of one molecule of L-glutamate from L-glutamine and 2-oxoglutarate. They exhibit a modular architecture with a common subunit or region, which is responsible for the L-glutamine-dependent glutamate synthesis from 2-oxoglutarate. Here, a PurF- (Type II- or Ntn-) type amidotransferase domain is coupled to the synthase domain, a (beta/alpha)8 barrel containing FMN and one [3Fe-4S]0,+1 cluster, through a approximately 30 angstroms-long intramolecular tunnel for the transfer of ammonia between the sites. In bacterial and eukaryotic GltS, reducing equivalents are provided by reduced pyridine nucleotides thanks to the stable association with a second subunit or region, which acts as a FAD-dependent NAD(P)H oxidoreductase and is responsible for the formation of the two low potential [4Fe-4S]+1,+2 clusters of the enzyme. In photosynthetic cells, reduced ferredoxin is the physiological reductant. This review focus on the mechanism of cross-activation of the synthase and glutaminase reactions in response to the bound substrates and the redox state of the enzyme cofactors, as well as on recent information on the structure of the alphabeta protomer of the NADPH-dependent enzyme, which sheds light on the intramolecular electron transfer pathway between the flavin cofactors.     

Quaternary structure of Azospirillum brasilense NADPH-dependent glutamate synthase in solution as revealed by synchrotron radiation x-ray scattering

Azospirillum brasilense glutamate synthase (GltS) is the prototype of bacterial NADPH-dependent enzymes, a class of complex iron-sulfur flavoproteins essential in ammonia assimilation processes. The catalytically active GltS alpha beta holoenzyme and its isolated alpha and beta subunits (162 and 52 kDa, respectively) were analyzed using synchrotron radiation x-ray solution scattering. The GltS alpha subunit and alpha beta holoenzyme were found to be tetrameric in solution, whereas the beta subunit was a mixture of monomers and dimers. Ab initio low resolution shapes restored from the scattering data suggested that the arrangement of alpha subunits in the (alpha beta)4 holoenzyme is similar to that in the tetrameric alpha 4 complex and that beta subunits occupy the periphery of the holoenzyme. The structure of alpha 4 was further modeled using the available crystallographic coordinates of the monomeric alpha subunit assuming P222 symmetry. To model the entire alpha beta holoenzyme, a putative alpha beta protomer was constructed from the coordinates of the alpha subunit and those of the N-terminal region of porcine dihydropyrimidine dehydrogenase, which is similar to the beta subunit. Rigid body refinement yielded a model of GltS with an arrangement of alpha subunits similar to that in alpha 4, but displaying contacts also between beta subunits belonging to adjacent protomers. The holoenzyme model allows for independent catalytic activity of the alpha beta protomers, which is consistent with the available biochemical evidence.     

The active conformation of glutamate synthase and its binding to ferredoxin

Glutamate synthases (GltS) are crucial enzymes in ammonia assimilation in plants and bacteria, where they catalyze the formation of two molecules of L-glutamate from L-glutamine and 2-oxoglutarate. The plant-type ferredoxin-dependent GltS and the functionally homologous alpha subunit of the bacterial NADPH-dependent GltS are complex four-domain monomeric enzymes of 140-165 kDa belonging to the NH(2)-terminal nucleophile family of amidotransferases. The enzymes function through the channeling of ammonia from the N-terminal amidotransferase domain to the FMN-binding domain. Here, we report the X-ray structure of the Synechocystis ferredoxin-dependent GltS with the substrate 2-oxoglutarate and the covalent inhibitor 5-oxo-L-norleucine bound in their physically distinct active sites solved using a new crystal form. The covalent Cys1-5-oxo-L-norleucine adduct mimics the glutamyl-thioester intermediate formed during L-glutamine hydrolysis. Moreover, we determined a high resolution structure of the GltS:2-oxoglutarate complex. These structures represent the enzyme in the active conformation. By comparing these structures with that of GltS alpha subunit and of related enzymes we propose a mechanism for enzyme self-regulation and ammonia channeling between the active sites. X-ray small-angle scattering experiments were performed on solutions containing GltS and its physiological electron donor ferredoxin (Fd). Using the structure of GltS and the newly determined crystal structure of Synechocystis Fd, the scattering experiments clearly showed that GltS forms an equimolar (1:1) complex with Fd. A fundamental consequence of this result is that two Fd molecules bind consecutively to Fd-GltS to yield the reduced FMN cofactor during catalysis.     

Determination of the midpoint potential of the FAD and FMN flavin cofactors and of the 3Fe-4S cluster of glutamate synthase

Glutamate synthase is a complex iron-sulfur flavoprotein that catalyzes the reductive transfer of the L-glutamine amide group to C(2) of 2-oxoglutarate, forming two molecules of L-glutamate. The bacterial enzyme is an alphabeta protomer, which contains one FAD (on the beta subunit, approximately 50 kDa), one FMN (on the alpha subunit, approximately 150 kDa), and three different Fe-S clusters (one 3Fe-4S center on the alpha subunit and two 4Fe-4S clusters at an unknown location). To address the problem of the intramolecular electron pathway, we have measured the midpoint potential values of the flavin cofactors and of the 3Fe-4S cluster of glutamate synthase in the isolated alpha and beta subunits and in the alphabeta holoenzyme. No detectable amounts of flavin semiquinones were observed during reductive titrations of the enzyme, indicating that the midpoint potential value of each flavin(ox)/flavin(sq) couple is, in all cases, significantly more negative than that of the corresponding flavin(sq)/flavin(hq) couple. Association of the two subunits to form the alphabeta protomer does not alter significantly the midpoint potential value of the FMN cofactor and of the 3Fe-4S cluster (approximately -240 and -270 mV, respectively), but it makes that of FAD some 40 mV less negative (approximately -340 mV for the beta subunit and -300 mV for FAD bound to the holoenzyme). Binding of the nonreducible NADP(+) analogue, 3-aminopyridine adenine dinucleotide phosphate, made the measured midpoint potential value of the FAD cofactor approximately 30-40 mV less negative in the isolated beta subunit, but had no effect on the redox properties of the alphabeta holoenzyme. This result correlates with the formation of a stable charge-transfer complex between the reduced flavin and the oxidized pyridine nucleotide in the isolated beta subunit, but not in the alphabeta holoenzyme. Binding of L-methionine sulfone, a glutamine analogue, had no significant effect on the redox properties of the enzyme cofactors. On the contrary, 2-oxoglutarate made the measured midpoint potential value of the 3Fe-4S cluster approximately 20 mV more negative in the isolated alpha subunit, but up to 100 mV less negative in the alphabeta holoenzyme as compared to the values of the corresponding free enzyme forms. These findings are consistent with electron transfer from the entry site (FAD) to the exit site (FMN) through the 3Fe-4S center of the enzyme and the involvement of at least one of the two low-potential 4Fe-4S centers, which are present in the glutamate synthase holoenzyme, but not in the isolated subunits. Furthermore, the data demonstrate a specific role of 2-oxoglutarate in promoting electron transfer from FAD to the 3Fe-4S cluster of the glutamate synthase holoenzyme. The modulatory role of 2-oxoglutarate is indeed consistent with the recently determined three-dimensional structure of the glutamate synthase alpha subunit, in which several polypeptide stretches are suitably positioned to mediate communication between substrate binding sites and the enzyme redox centers (FMN and the 3Fe-4S cluster) to tightly control and coordinate the individual reaction steps [Binda, C., et al. (2000) Structure 8, 1299-1308].     

Molecular dynamics simulation of the interaction between the complex iron-sulfur flavoprotein glutamate synthase and its substrates

Glutamate synthase (GltS) is a complex iron-sulfur flavoprotein that catalyzes the reductive transfer of L-glutamine amide group to the C2 carbon of 2-oxoglutarate yielding two molecules of L-glutamate. Molecular dynamics calculations in explicit solvent were carried out to gain insight into the conformational flexibility of GltS and into the role played by the enzyme substrates in regulating the catalytic cycle. We have modelled the free (unliganded) form of Azospirillum brasilense GltS alpha subunit and the structure of the reduced enzyme in complex with the L-glutamine and 2-oxoglutarate substrates starting from the crystallographically determined coordinates of the GltS alpha subunit in complex with L-methionine sulphone and 2-oxoglutarate. The present 4-ns molecular dynamics calculations reveal that the GltS glutaminase site may exist in a catalytically inactive conformation unable to bind glutamine, and in a catalytically competent conformation, which is stabilized by the glutamine substrate. Substrates binding also induce (1) closure of the loop formed by residues 263-271 with partial shielding of the glutaminase site from solvent, and (2) widening of the ammonia tunnel entrance at the glutaminase end to allow for ammonia diffusion toward the synthase site. The Q-loop of glutamate synthase, which acts as an active site lid in other amidotransferases, seems to maintain an open conformation. Finally, binding of L-methionine sulfone, a glutamine analog that mimics the tetrahedral transient species occurring during its hydrolysis, causes a coordinated rigid-body motion of segments of the glutaminase domain that results in the inactive conformation observed in the crystal structure of GltS alpha subunit.     

Glutamate synthase: identification of the NADPH-binding site by site-directed mutagenesis

To contribute to the understanding of glutamate synthase and of beta subunit-like proteins, which have been detected by sequence analyses, we identified the NADPH-binding site out of the two potential ADP-binding regions found in the beta subunit. The substitution of an alanyl residue for G298 of the beta subunit of Azospirillum brasilense glutamate synthase (the second glycine in the GXGXXA fingerprint of the postulated NADPH-binding site) yielded a protein species in which the flavin environment and properties are unaltered. On the contrary, the binding of the pyridine nucleotide substrate is significantly perturbed demonstrating that the C-terminal potential ADP-binding fold of the beta subunit is indeed the NADPH-binding site of the enzyme. The major effect of the G298A substitution in the GltS beta subunit consists of an approximately 10-fold decrease of the affinity of the enzyme for pyridine nucleotides with little or no effect on the rate of the enzyme reduction by NADPH. By combining kinetic measurements and absorbance-monitored equilibrium titrations of the G298A-beta subunit mutant, we conclude that also the positioning of its nicotinamide portion into the active site is altered thus preventing the formation of a stable charge-transfer complex between reduced FAD and NADP(+). During the course of this work, the Azospirillum DNA regions flanking the gltD and gltB genes, the genes encoding the GltS beta and alpha subunits, respectively, were sequenced and analyzed. Although the Azospirillum GltS is similar to the enzyme of other bacteria, it appears that the corresponding genes differ with respect to their arrangement in the chromosome and to the composition of the glt operon: no genes corresponding to E. coli and Klebsiella aerogenes gltF or to Bacillus subtilis gltC, encoding regulatory proteins, are found in the DNA regions adjacent to that containing gltD and gltB genes in Azospirillum. Further studies are needed to determine if these findings also imply differences in the regulation of the glt genes expression in Azospirillum (a nitrogen-fixing bacterium) with respect to enteric bacteria.     

Functional properties of recombinant Azospirillum brasilense glutamate synthase, a complex iron-sulfur flavoprotein.

Azospirillum brasilense glutamate synthase is a complex iron-sulfur flavoprotein that catalyses the NADPH-dependent reductive transfer of glutamine amide group to the C(2) carbon of 2-oxoglutarate to yield L-glutamate. Its catalytically active alphabeta protomer is composed of two dissimilar subunits (alpha subunit, 164.2 kDa; beta subunit, 52.3 kDa) and contains one FAD (at Site 1, the pyridine nucleotide site within the beta subunit), one FMN (at Site 2, the 2-oxoglutarate/L-glutamate site in the alpha subunit) and three different iron-sulfur clusters (one 3Fe-4S center on the alpha subunit and two 4Fe-4S clusters of unknown location). A plasmid harboring the gltD and gltB genes, the genes encoding the glutamate synthase beta and alpha subunits, respectively, each one under the control of the T7/lac promoter of pET11a was found to be suitable for the overproduction of glutamate synthase holoenzyme in Escherichia coli BL21(DE3) cells. Recombinant A. brasilense glutamate synthase could be purified to homogeneity from overproducing E. coli cells by ion exchange chromatography, gel filtration and affinity chromatography on a 2',5' ADP-Sepharose 4B column. The purified enzyme was indistinguishable from that prepared from Azospirillum cells with respect to cofactor content, N-terminal sequence of the subunits, aggregation state, kinetic and spectroscopic properties. The study of the recombinant holoenzyme allowed us to establish that the tendency of glutamate synthase to form a stable (alphabeta)4 tetramer at high protein concentrations is a property unique to the holoenzyme, as the isolated beta subunit does not oligomerize, while the isolated glutamate synthase alpha subunit only forms dimers at high protein concentrations. Furthermore, the steady-state kinetic analysis of the glutamate synthase reaction was extended to the study of the effect of adenosine-containing nucleotides. Compounds such as cAMP, AMP, ADP and ATP have no effect on the enzyme activity, while the 2'-phosphorylated analogs of AMP and NADP(H) analogs act as inhibitors of the reaction, competitive with NADPH. Thus, it can be ruled out that glutamate synthase reaction is subjected to allosteric modulation by adenosine containing (di)nucleotides, which may bind to the putative ADP-binding site at the C-terminus of the alpha subunit. At the same time, the strict requirement of a 2'-phosphate group in the pyridine nucleotide for binding to glutamate synthase (GltS) was established. Finally, by comparing the inhibition constants exhibited by a series of NADP+ analogs, the contribution to the binding energy of the various parts of the pyridine nucleotide has been determined along with the effect of substituents on the 3 position of the pyridine ring. With the exception of thio-NADP+, which binds the tightest to GltS, it appears that the size of the substituent is the factor that affects the most the interaction between the NADP(H) analog and the enzyme.     

Purification and properties of glutamate synthase from Bacillus licheniformis

Glutamate synthase [L-glutamate:NADP+ oxidoreductase (transaminating); EC 1.4.1.13](GltS) was purified to homogeneity from Bacillus licheniformis A5. The native enzyme had a molecular weight of approximately 220,000 and was composed of two nonidentical subunits (molecular weights, approximately 158,000 and approximately 54,000). The enzyme was found to contain 8.1 +/- 1 iron atoms and 8.1 +/- 1 acid-labile sulfur atoms per 220,000-dalton dimer. Two flavin moieties were found per 220,000-dalton dimer, with a ratio of flavin adenine dinucleotide to flavin mononucleotide of 1.2. The UV-visible spectrum of the enzyme exhibited maxima at 263,380 and 450 nm. The GltS from B. licheniformis had a requirement for NADPH, alpha-ketoglutarate, and glutamine. Classical hyperbolic kinetics were seen for NADPH affinity, which resulted in an apparent Km value of 13 microM. Nonhyperbolic kinetics were obtained for alpha-ketoglutarate and glutamine affinities, and the reciprocal plots obtained for these substrates were biphasic. The apparent Km values obtained for glutamine were 8 and 100 microM, and the apparent Km values obtained for alpha-ketoglutarate were 6 and 50 microM. GltS activity was found to be relatively insensitive to inhibition by amino acids, keto acids, or various nucleotides. L-Methionine-DL-sulfoximine, L-methionine sulfone, and DL-methionine sulfoxide were found to be potent inhibitors of GltS activity, yielding I0.5 values of 150, 11, and 250 microM, respectively. GltSs were purified from cells grown in the presence of ammonia and nitrate as sole nitrogen sources and were compared. Both yielded identical final specific activities and identical physical (UV-visible spectra, flavin, and iron-sulfur composition) and kinetic characteristics.     

Properties of the recombinant ferredoxin-dependent glutamate synthase of Synechocystis PCC6803. Comparison with the Azospirillum brasilense NADPH-dependent enzyme and its isolated alpha subunit

The properties of the recombinant ferredoxin-dependent glutamate synthase of Synechocystis PCC6803 were determined by means of kinetic and spectroscopic approaches in comparison to those exhibited by the bacterial NADPH-dependent enzyme form. The ferredoxin-dependent enzyme was found to be similar to the bacterial glutamate synthase alpha subunit with respect to cofactor content (one FMN cofactor and one [3Fe-4S] cluster per enzyme subunit), overall absorbance properties, and reactivity of the FMN N(5) position with sulfite, as expected from the similar primary structure of ferredoxin-dependent glutamate synthase and of the bacterial NADPH-dependent glutamate synthase alpha subunit. The ferredoxin- and NADPH-dependent enzymes were found to differ with respect to the apparent midpoint potential values of the FMN cofactor and of the [3Fe-4S] cluster, which are less negative in the ferredoxin-dependent enzyme form. This feature is, at least in part, responsible for the efficient oxidation of L-glutamate catalyzed by this enzyme form, but not by the bacterial NADPH-dependent counterpart. At variance with earlier reports on ferredoxin-dependent glutamate synthase, in the Synechocystis enzyme the [3Fe-4S] cluster is not equipotential with the flavin cofactor. The present studies also demonstrated that binding of reduced ferredoxin to ferredoxin-dependent glutamate synthase is essential in order to activate reaction steps such as glutamine binding, hydrolysis, or ammonia transfer from the glutamine amidotransferase site to the glutamate synthase site of the enzyme. Thus, ferredoxin-dependent glutamate synthase seems to control and coordinate catalytic activities taking place at its subsites by regulating the reactions of the glutamine amidotransferase site. Association with reduced ferredoxin appears to be necessary, but not sufficient, to trigger the required activating conformational changes.     

Effect of NBD chloride (4-chloro-7-nitrobenzo-2-oxa-1,3-diazole) on the pyridine nucleotide transhydrogenase of Escherichia coli.

Pyridine nucleotide transhydrogenases of bacterial cytosolic membranes and mitochondrial inner membranes are proton pumps in which hydride transfer between NADP(+) and NAD(+) is coupled to proton translocation across cytosolic or mitochondrial membranes. The pyridine nucleotide transhydrogenase of Escherichia coli is composed of two subunits (alpha and beta). Three domains are recognized. The extrinsic cytosolic domain 1 of the amino-terminal region of the alpha subunit bears the NAD(H)-binding site. The NADP(H)-binding site is present in domain 3, the extrinsic cytosolic carboxyl-terminal region of the beta subunit. Domain 2 is composed of the membrane-intrinsic carboxyl-terminal region of the alpha subunit and the membrane-intrinsic amino-terminal region of the beta subunit. Treatment of the transhydrogenase of E. coli with 4-chloro-7-nitrobenzo-2-oxa-1,3-diazole (NBD chloride) inhibited enzyme activity. Analysis of inhibition revealed that several sites on the enzyme were involved. NBD chloride modified two (betaCys-147 and betaCys-260) of the seven cysteine residues present in the transhydrogenase. Modification of betaCys-260 in domain 2 resulted in inhibition of enzyme activity. Modification of residues other than cysteine residues also resulted in inhibition of transhydrogenation as shown by use of a cysteine-free mutant enzyme. The beta subunit was modified by NBD chloride to a greater extent than the alpha subunit. Reaction of domain 2 and domain 3 was prevented by NADPH. Modification of domain 3 is probably not associated with inhibition of enzyme activity. Modification of domain 2 of the beta subunit resulted in a decreased binding affinity for NADPH at its binding site in domain 3. The product resulting from the reaction of NBD chloride with NADPH was a very effective inhibitor of transhydrogenation. In experiments with NBD chloride in the presence of NADPH it is likely that all of the sites of reaction described above will contribute to the inhibition observed. The NBD-NADPH adduct will likely be more useful than NBD chloride in investigations of the pyridine nucleotide transhydrogenase.     

Organization of the multiple coenzymes and subunits and role of the covalent flavin link in the complex heterotetrameric sarcosine oxidase

Heterotetrameric (alphabetagammadelta) sarcosine oxidase from Corynebacterium sp. P-1 (cTSOX) contains noncovalently bound FAD and NAD(+) and covalently bound FMN, attached to beta(His173). The beta(His173Asn) mutant is expressed as a catalytically inactive, labile heterotetramer. The beta and delta subunits are lost during mutant enzyme purification, which yields a stable alphagamma complex. Addition of stabilizing agents prevents loss of the delta but not the beta subunit. The covalent flavin link is clearly a critical structural element and essential for TSOX activity or preventing FMN loss. The alpha subunit was expressed by itself and purified by affinity chromatography. The alpha and beta subunits each contain an NH(2)-terminal ADP-binding motif that could serve as part of the binding site for NAD(+) or FAD. The alpha subunit and the alphagamma complex were each found to contain 1 mol of NAD(+) but no FAD. Since NAD(+) binds to alpha, FAD probably binds to beta. The latter could not be directly demonstrated since it was not possible to express beta by itself. However, FAD in TSOX from Pseudomonas maltophilia (pTSOX) exhibits properties similar to those observed for the covalently bound FAD in monomeric sarcosine oxidase and N-methyltryptophan oxidase, enzymes that exhibit sequence homology with beta. A highly conserved glycine in the ADP-binding motif of the alpha(Gly139) or beta(Gly30) subunit was mutated in an attempt to generate NAD(+)- or FAD-free cTSOX, respectively. The alpha(Gly139Ala) mutant is expressed only at low temperature (t(optimum) = 15 degrees C), but the purified enzyme exhibited properties indistinguishable from the wild-type enzyme. The much larger barrier to NAD(+) binding in the case of the alpha(Gly139Val) mutant could not be overcome even by growth at 3 degrees C, suggesting that NAD(+) binding is required for TSOX expression. The beta(Gly30Ala) mutant exhibited subunit expression levels similar to those of the wild-type enzyme, but the mutation blocked subunit assembly and covalent attachment of FMN, suggesting that both processes require a conformational change in beta that is induced upon FAD binding. About half of the covalent FMN in recombinant preparations of cTSOX or pTSOX is present as a reversible covalent 4a-adduct with a cysteine residue. Adduct formation is not prevented by mutating any of the three cysteine residues in the beta subunit of cTSOX to Ser or Ala. Since FMN is attached via its 8-methyl group to the beta subunit, the FMN ring must be located at the interface between beta and another subunit that contains the reactive cysteine residue.     

Cross-talk and ammonia channeling between active centers in the unexpected domain arrangement of glutamate synthase

INTRODUCTION: The complex iron-sulfur flavoprotein glutamate synthase catalyses the reductive synthesis of L-glutamate from 2-oxoglutarate and L-glutamine, a reaction in the plant and bacterial pathway for ammonia assimilation. The enzyme functions through three distinct active centers carrying out L-glutamine hydrolysis, conversion of 2-oxoglutarate into L-glutamate, and electron uptake from an electron donor. RESULTS: The 3.0 A crystal structure of the dimeric 324 kDa core protein of a bacterial glutamate synthase was solved by the MAD method, using the very weak anomalous signal of the two 3Fe-4S clusters present in the asymmetric unit. The 1,472 amino acids of the monomer fold into a four-domain architecture. The two catalytic domains have canonical Ntn-amidotransferase and FMN binding (beta/alpha)8 barrel folds, respectively. The other two domains have an unusual "cut (beta/alpha)8 barrel" topology and an unexpected novel beta-helix structure. Channeling of the ammonia intermediate is brought about by an internal tunnel of 31 A length, which runs from the site of L-glutamine hydrolysis to the site of L-glutamate synthesis. CONCLUSIONS: The outstanding property of glutamate synthase is the ability to coordinate the activity of its various functional sites to avoid wasteful consumption of L-glutamine. The structure reveals two polypeptide segments that connect the catalytic centers and embed the ammonia tunnel, thus being ideally suited to function in interdomain signaling. Depending on the enzyme redox and ligation states, these signal-transducing elements may affect the active site geometry and control ammonia diffusion through a gating mechanism.     

Structural evidence for ammonia tunneling across the (beta alpha)(8) barrel of the imidazole glycerol phosphate synthase bienzyme complex

Since reactive ammonia is not available under physiological conditions, glutamine is used as a source for the incorporation of nitrogen in a number of metabolic pathway intermediates. The heterodimeric ImGP synthase that links histidine and purine biosynthesis belongs to the family of glutamine amidotransferases in which the glutaminase activity is coupled with a subsequent synthase activity specific for each member of the enzyme family. Its X-ray structure from the hyperthermophile Thermotoga maritima shows that the glutaminase subunit is associated with the N-terminal face of the (beta alpha)(8) barrel cyclase subunit. The complex reveals a putative tunnel for the transfer of ammonia over a distance of 25 A. Although ammonia tunneling has been reported for glutamine amidotransferases, the ImGP synthase has evolved a novel mechanism, which extends the known functional properties of the versatile (beta alpha)(8) barrel fold.     

Heterotetrameric sarcosine oxidase: structure of a diflavin metalloenzyme at 1.85 A resolution

The crystal structure of heterotetrameric sarcosine oxidase (TSOX) from Pseudomonas maltophilia has been determined at 1.85 A resolution. TSOX contains three coenzymes (FAD, FMN and NAD+), four different subunits (alpha, 103 kDa; beta, 44 kDa; gamma, 21 kDa; delta, 11 kDa) and catalyzes the oxidation of sarcosine (N-methylglycine) to yield hydrogen peroxide, glycine and formaldehyde. In the presence of tetrahydrofolate, the oxidation of sarcosine is coupled to the formation of 5,10-methylenetetrahydrofolate. The NAD+ and putative folate binding sites are located in the alpha-subunit. The FAD binding site is in the beta-subunit. FMN is bound at the interface of the alpha and beta-subunits. The FAD and FMN rings are separated by a short segment of the beta-subunit with the closest atoms located 7.4 A apart. Sulfite, an inhibitor of oxygen reduction, is bound at the FMN site. 2-Furoate, a competitive inhibitor with respect to sarcosine, is bound at the FAD site. The sarcosine dehydrogenase and 5,10-methylenetetrahydrofolate synthase sites are 35 A apart but connected by a large internal cavity (approximately 10,000 A3). An unexpected zinc ion, coordinated by three cysteine and one histidine side-chains, is bound to the delta-subunit. The N-terminal half of the alpha subunit of TSOX (alphaA) is closely similar to the FAD-binding domain of glutathione reductase but with NAD+ replacing FAD. The C-terminal half of the alpha subunit of TSOX (alphaB) is similar to the C-terminal half of dimethylglycine oxidase and the T-protein of the glycine cleavage system, proteins that bind tetrahydrofolate. The beta-subunit of TSOX is very similar to monomeric sarcosine oxidase. The gamma-subunit is similar to the C-terminal sub-domain of alpha-TSOX. The delta-subunit shows little similarity with any PDB entry. The alphaA domain/beta-subunit sub-structure of TSOX closely resembles the alphabeta dimer of L-proline dehydrogenase, a heteroctameric protein (alphabeta)4 that shows highest overall similarity to TSOX.     

Structural studies on the synchronization of catalytic centers in glutamate synthase

The complex iron-sulfur flavoprotein glutamate synthase (GltS) plays a prominent role in ammonia assimilation in bacteria, yeasts, and plants. GltS catalyzes the formation of two molecules of l-glutamate from 2-oxoglutarate and l-glutamine via intramolecular channeling of ammonia. GltS has the impressive ability of synchronizing its distinct catalytic centers to avoid wasteful consumption of l-glutamine. We have determined the crystal structure of the ferredoxin-dependent GltS in several ligation and redox states. The structures reveal the crucial elements in the synchronization between the glutaminase site and the 2-iminoglutarate reduction site. The structural data combined with the catalytic properties of GltS indicate that binding of ferredoxin and 2-oxoglutarate to the FMN-binding domain of GltS induce a conformational change in the loop connecting the two catalytic centers. The rearrangement induces a shift in the catalytic elements of the amidotransferase domain, such that it becomes activated. This machinery, over a distance of more than 30 A, controls the ability of the enzyme to bind and hydrolyze the ammonia-donating substrate l-glutamine.     

Isolation and characterization of recombinant human casein kinase II subunits alpha and beta from bacteria

cDNA encoding the casein kinase II (CKII) subunits alpha and beta of human origin were expressed in Escherichia coli using expression vector pT7-7. Significant expression was obtained with E. coli BL21(DE3). The CKII subunits accounted for approximately 30% of the bacterial protein; however, most of the expressed proteins were produced in an insoluble form. The recombinant CKII alpha subunit was purified by DEAE-cellulose chromatography, followed by phosphocellulose and heparin-agarose chromatography. The recombinant CKII beta subunit was extracted from the insoluble pellet and purified in a single step on phosphocellulose. From 10 g bacterial cells, the yield of soluble protein was 12 mg alpha subunit and 5 mg beta subunit. SDS/PAGE analysis of the purified recombinant proteins indicated molecular masses of 42 kDa and 26 kDa for the alpha and beta subunits, respectively, in agreement with the molecular masses determined for the subunits of the native enzyme. The recombinant alpha subunit exhibited protein kinase activity which was greatest in the absence of monovalent ions. With increasing amounts of salt, alpha subunit kinase activity declined rapidly. Addition of the beta subunit led to maximum stimulation at a 1:1 ratio of both subunits. Using a synthetic peptide (RRRDDDSDDD) as a substrate, the maximum protein kinase stimulation observed was fourfold under the conditions used. The Km of the reconstituted enzyme for the synthetic peptide (80 microM) was comparable to the mammalian enzyme (40-60 microM), whereas the alpha subunit alone had a Km of 240 microM. After sucrose density gradient analysis, the reconstituted holoenzyme sedimented at the same position as the mammalian CKII holoenzyme.     

Topological analysis of the pyridine nucleotide transhydrogenase of Escherichia coli using proteolytic enzymes

The pyridine nucleotide transhydrogenase of Escherichia coli has an alpha 2 beta 2 structure (alpha: Mr, 54,000; beta: Mr, 48,700). Hydropathy analysis of the amino acid sequences suggested that the 10 kDa C-terminal portion of the alpha subunit and the N-terminal 20-25 kDa region of the beta subunit are composed of transmembranous alpha-helices. The topology of these subunits in the membrane was investigated using proteolytic enzymes. Trypsin digestion of everted cytoplasmic membrane vesicles released a 43 kDa polypeptide from the alpha subunit. The beta subunit was not susceptible to trypsin digestion. However, it was digested by proteinase K in everted vesicles. Both alpha and beta subunits were not attacked by trypsin and proteinase K in right-side out membrane vesicles. The beta subunit in the solubilized enzyme was only susceptible to digestion by trypsin if the substrates NADP(H) were present. NAD(H) did not affect digestion of the beta subunit. Digestion of the beta subunit of the membrane-bound enzyme by trypsin was not induced by NADP(H) unless the membranes had been previously stripped of extrinsic proteins by detergent. It is concluded that binding of NADP(H) induces a conformational change in the transhydrogenase. The location of the trypsin cleavage sites in the sequences of the alpha and beta subunits were determined by N- and C-terminal sequencing. A model is proposed in which the N-terminal 43 kDa region of the alpha subunit and the C-terminal 30 kDa region of the beta subunit are exposed on the cytoplasmic side of the inner membrane of E. coli. Binding sites for pyridine nucleotide coenzymes in these regions were suggested by affinity chromatography on NAD-agarose columns.     

Identification of a novel proline-rich peptide-binding domain in prolyl 4-hydroxylase.

Prolyl 4-hydroxylase (EC 1.14.11.2) catalyzes the hydroxylation of -X-Pro-Gly- sequences and plays a central role in the synthesis of all collagens. The [alpha(I)]2beta2 type I enzyme is effectively inhibited by poly(L-proline), whereas the [alpha(II)]2beta2 type II enzyme is not. We report here that the poly(L-proline) and (Pro-Pro-Gly)10 peptide substrate-binding domain of prolyl 4-hydroxylase is distinct from the catalytic domain and consists of approximately 100 amino acids. Peptides of 10-19 kDa beginning around residue 140 in the 517 residue alpha(I) subunit remained bound to poly(L-proline) agarose after limited proteolysis of the human type I enzyme tetramer. A recombinant polypeptide corresponding to the alpha(I) subunit residues 138-244 and expressed in Escherichia coli was soluble, became effectively bound to poly(L-proline) agarose and could be eluted with (Pro-Pro-Gly)10. This polypeptide is distinct from the SH3 and WW domains, and from profilin, and thus represents a new type of proline-rich peptide-binding module. Studies with enzyme tetramers containing mutated alpha subunits demonstrated that the presence of a glutamate and a glutamine in the alpha(II) subunit in the positions corresponding to Ile182 and Tyr233 in the alpha(I) subunit explains most of the lack of poly(L-proline) binding of the type II prolyl 4-hydroxylase. Keywords: collagen/dioxygenases/peptide-binding domain/ proline-rich/prolyl hydroxylase     

Functional split and crosslinking of the membrane domain of the beta subunit of proton-translocating transhydrogenase from Escherichia coli

Proton pumping nicotinamide nucleotide transhydrogenase from Escherichia coli contains an alpha subunit with the NAD(H)-binding domain I and a beta subunit with the NADP(H)-binding domain III. The membrane domain (domain II) harbors the proton channel and is made up of the hydrophobic parts of the alpha and beta subunits. The interface in domain II between the alpha and the beta subunits has previously been investigated by cross-linking loops connecting the four transmembrane helices in the alpha subunit and loops connecting the nine transmembrane helices in the beta subunit. However, to investigate the organization of the nine transmembrane helices in the beta subunit, a split was introduced by creating a stop codon in the loop connecting transmembrane helices 9 and 10 by a single mutagenesis step, utilizing an existing downstream start codon. The resulting enzyme was composed of the wild-type alpha subunit and the two new peptides beta1 and beta2. As compared to other split membrane proteins, the new transhydrogenase was remarkably active and catalyzed activities for the reduction of 3-acetylpyridine-NAD(+) by NADPH, the cyclic reduction of 3-acetylpyridine-NAD(+) by NADH (mediated by bound NADP(H)), and proton pumping, amounting to about 50-107% of the corresponding wild-type activities. These high activities suggest that the alpha subunit was normally folded, followed by a concerted folding of beta1 + beta2. Cross-linking of a betaS105C-betaS237C double cysteine mutant in the functional split cysteine-free background, followed by SDS-PAGE analysis, showed that helices 9, 13, and 14 were in close proximity. This is the first time that cross-linking between helices in the same beta subunit has been demonstrated.