NORMAL GENOMIC

• Fiedler E, Thorell S, Sandalova T, Golbik R, Konig S, Schneider G
• Snapshot of a key intermediate in enzymatic thiamin catalysis: crystalstructure of the alpha-carbanion of (alpha,beta-dihydroxyethyl)-thiamindiphosphate in the active site of transketolase from Saccharomycescerevisiae.
• Proc Natl Acad Sci U S A. 2002; 99: 591-5
• Display abstract

• Kinetic and spectroscopic data indicated that addition of the donorsubstrate hydroxypyruvate to the thiamin diphosphate (ThDP)-dependentenzyme transketolase (TK) led to the accumulation of thealpha-carbanion/enamine of (alpha,beta-dihydroxyethyl) ThDP, the keyreaction intermediate in enzymatic thiamin catalysis. Thethree-dimensional structure of this intermediate trapped in the activesite of yeast TK was determined to 1.9-A resolution by usingcryocrystallography. The electron density suggests a planaralpha-carbanion/enamine intermediate having the E-configuration. Thereaction intermediate is firmly held in place through direct hydrogenbonds to His-103 and His-481 and an indirect hydrogen bond via a watermolecule to His-69. The 4-NH(2) group of the amino-pyrimidine ring of ThDPis within 3 A distance to the alpha-hydroxy oxygen atom of thedihydroxyethyl moiety but at an angle unfavorable for a strong hydrogenbond. No structural changes occur in TK on formation of the reactionintermediate, suggesting that the active site is poised for catalysis andconformational changes during the enzyme reaction are not very likely. Theintermediate is present with high occupancy in both active sites, arguingagainst previous proposals of half-of-the-sites reactivity in yeast TK.

• Meshalkina L, Nilsson U, Wikner C, Kostikowa T, Schneider G
• Examination of the thiamin diphosphate binding site in yeast transketolaseby site-directed mutagenesis.
• Eur J Biochem. 1997; 244: 646-52
• Display abstract

• The role of two conserved amino acid residues in the thiamin diphosphatebinding site of yeast transketolase has been analyzed by site-directedmutagenesis. Replacement of E162, which is part of a cluster of glutamicacid residues at the subunit interface, by alanine or glutamine results inmutant enzymes with most catalytic properties similar to wild-type enzyme.The two mutant enzymes show, however, significant increases in the K0.5values for thiamin diphosphate in the absence of substrate and in the lagof the reaction progress curves. This suggests that the interaction ofE162 with residue E418, and possibly E167, from the second subunit isimportant for formation and stabilization of the transketolase dimer.Replacement of the conserved residue D382, which is buried upon binding ofthiamin diphosphate, by asparagine and alanine, results in mutant enzymesseverely impaired in thiamin diphosphate binding and catalytic efficiency.The 25-80-fold increase in K0.5 for thiamin diphosphate suggests that D382is involved in cofactor binding, probably by electrostatic compensation ofthe positive charge of the thiazolium ring and stabilization of a flexibleloop at the active site. The decrease in catalytic activities in the D382mutants indicates that this residue might also be important in subsequentsteps in catalysis.

• Wang JJ, Martin PR, Singleton CK
• Aspartate 155 of human transketolase is essential for thiaminediphosphate-magnesium binding, and cofactor binding is required for dimerformation.
• Biochim Biophys Acta. 1997; 1341: 165-72
• Display abstract

• Active human transketolase is a homodimeric enzyme possessing two activesites, each with a non-covalently bound thiamine diphosphate andmagnesium. Both subunits contribute residues at each site which areinvolved in cofactor binding and in catalysis. His-tagged transketolase,produced in E. coli, was similar to transketolase purified from humantissues with respect to Km apps for cofactor and substrates and withrespect to cofactor-dependent hysteresis. Mutation of aspartate 155,corresponding to a conserved aspartate residue among thiaminediphosphate-binding proteins, resulted in an inactive protein which couldnot bind the cofactor-magnesium complex and which could not dimerize. Theresults are consistent with the suggestion that aspartate 155 is animportant coordination site for magnesium. In support of thisinterpretation, binding of cofactor by wild type apo-transketolaserequired the presence of magnesium. Additionally, monomericapo-his-transketolase required both magnesium and cofactor binding fordimer formation.

• Arjunan P et al.
• Crystal structure of the thiamin diphosphate-dependent enzyme pyruvatedecarboxylase from the yeast Saccharomyces cerevisiae at 2.3 A resolution.
• J Mol Biol. 1996; 256: 590-600
• Display abstract

• The crystal structure of pyruvate decarboxylase (EC 4.1.1.1), a thiamindiphosphate-dependent enzyme isolated from Saccharomyces cerevisiae, hasbeen determined and refined to a resolution of 2.3 A. Pyruvatedecarboxylase is a homotetrameric enzyme which crystallizes with twosubunits in an asymmetric unit. The structure has been refined by acombination of simulated annealing and restrained least squares to an Rfactor of 0.165 for 46,787 reflections. As in the corresponding enzymefrom Saccharomyces uvarum, the homotetrameric holoenzyme assembly hasapproximate 222 symmetry. In addition to providing more accurate atomicparameters and certainty in the sequence assignments, the high resolutionand extensive refinement resulted in the identification of several tightlybound water molecules in key structural positions. These water moleculeshave low temperature factors and make several hydrogen bonds with proteinresidues. There are six such water molecules in each cofactor bindingsite, and one of them is involved in coordination with the requiredmagnesium ion. Another may be involved in the catalytic reactionmechanism. The refined model includes 1074 amino acid residues (twosubunits), two thiamin diphosphate cofactors, two magnesium ionsassociated with cofactor binding and 440 water molecules. From the refinedmodel we conclude that the resting state of the enzyme-cofactor complex issuch that the cofactor is already deprotonated at the N4' position of thepyrimidine ring, and is poised to accept a proton from the C2 position ofthe thiazolium ring.

• Konig S, Schellenberger A, Neef H, Schneider G
• Specificity of coenzyme binding in thiamin diphosphate-dependent enzymes.Crystal structures of yeast transketolase in complex with analogs ofthiamin diphosphate.
• J Biol Chem. 1994; 269: 10879-82
• Display abstract

• The three-dimensional structures of complexes of yeast apotransketolasewith the coenzyme analogs 6'-methyl, N1'-pyridyl, and N3'-pyridyl thiamindiphosphate, respectively, were determined with protein crystallographicmethods. All three coenzyme analogs bind to the enzyme in a fashion highlysimilar to the cofactor thiamin diphosphate. Thus, either one of thehydrogen bonds of the pyrimidine ring nitrogens to the protein issufficient for proper binding and positioning of the cofactor. The lack ofcatalytic activity of the N3'-pyridyl analog is not due to incorrectorientation of the pyrimidine ring, but results from the absence of thehydrogen bond between the N1' nitrogen atom and the conserved residueGlu418. The structure analysis provides further evidence for theimportance of this conserved interaction for enzymatic thiamin catalysis.

• Dirr H, Reinemer P, Huber R
• Refined crystal structure of porcine class Pi glutathione S-transferase(pGST P1-1) at 2.1 A resolution.
• J Mol Biol. 1994; 243: 72-92
• Display abstract

• The crystal structure of class Pi glutathione S-transferase from porcinelung (pGST P1-1) in complex with glutathione sulphonate has been refinedat 2.11 A resolution, to a crystallographic R-factor of 16.5% for 21, 165unique reflections. The refined structure includes 3314 protein atoms, 46inhibitor (glutathione sulphonate) atoms and 254 water molecules. Themodel shows good stereochemistry, with root-mean-square deviations fromideal bond lengths and bond angles of 0.011 A and 2.8 degrees,respectively. The estimated root-mean-square co-ordinate error is 0.2 A.The protein is a dimer assembled from identical subunits of 207 amino acidresidues. The tertiary structure of the pGST P1 subunit is organized astwo domains, the N-terminal domain (domain I, residues 1 to 74) and thelarger C-terminal domain (domain II, residues 81 to 207). Glutathionesulphonate, a competitive inhibitor, binds to the G-site region (i.e. theglutathione-binding region) of the active site located on each subunit.Each G-site is, however, structurally dependent of the neighbouringsubunit as structural elements forming a fully functional G-site areprovided by both subunits, with domain I as the major supportingframework. A number of direct and water-mediated polar interactions areinvolved in sequestering the glutathione analogue at the G-site. Theextended conformation assumed by the enzyme-bound inhibitor as well as thestrategic interactions between inhibitor and protein, closely resemblethose observed for the physiological substrate, reduced glutathione boundat the active site of class Mu glutathione S-transferase 3-3. Hydrogenbonding between the sulphonyl moiety of the inhibitor and the hydroxylgroup of an evolutionary conserved tyrosine residue, Tyr7, provides thefirst direct structural evidence for a catalytic protein group inglutathione S-transferases that is involved in the activation of thesubstrate glutathione. The catalytic role for Tyr7 has subsequently beenconfirmed by mutagenesis and kinetic studies. Comparison of the knowncrystal structures for class Pi, class Mu and class Alpha isoenzymes,indicates that the cytosolic glutathione S-transferases share a commonfold and that the structural features for catalysis are similar.

• Dyda F, Furey W, Swaminathan S, Sax M, Farrenkopf B, Jordan F
• Catalytic centers in the thiamin diphosphate dependent enzyme pyruvatedecarboxylase at 2.4-A resolution.
• Biochemistry. 1993; 32: 6165-70
• Display abstract

• The crystal structure of brewers' yeast pyruvate decarboxylase, a thiamindiphosphate dependent alpha-keto acid decarboxylase, has been determinedto 2.4-A resolution. The homotetrameric assembly contains two dimers,exhibiting strong intermonomer interactions within each dimer but morelimited ones between dimers. Each monomeric subunit is partitioned intothree structural domains, all folding according to a mixed alpha/betamotif. Two of these domains are associated with cofactor binding, whilethe other is associated with substrate activation. The catalytic centerscontaining both thiamin diphosphate and Mg(II) are located deep in theintermonomer interface within each dimer. Amino acids important incofactor binding and likely to participate in catalysis and substrateactivation are identified.

• Robinson BH, Chun K
• The relationships between transketolase, yeast pyruvate decarboxylase andpyruvate dehydrogenase of the pyruvate dehydrogenase complex.
• FEBS Lett. 1993; 328: 99-102
• Display abstract

• The amino acid sequences of four thiamine pyrophosphate-requiring enzymeswere aligned with the published amino acid sequence of the transketolaseof Hansenula polymorpha. Sequences of the combined alpha and beta subunitsof the E1 enzyme of the pyruvate dehydrogenase complexes of Homo sapiensand Bacillus stearothermophilus aligned well with the transketolase whilethe E1 of the pyruvate dehydrogenase complex of Escherichia coli alignedeasily provided a non-aligning segment of 77 amino acids was omitted. Thenon-acetylating pyruvate decarboxylase of Saccharomyces cerevisiae couldonly be aligned if the sequence was cut in two with the C-terminuscorresponding to the N-terminus of the other TPP-dependent enzymes. Usingthe published 2.5 A resolution of the X-ray crystal structure ofSaccharomyces cerevisiae transketolase as a template we show that ahydrophobic region of the beta-subunit of the PDH E1 alpha beta enzymeslikely contains a binding site for the thiazolium ring of TPP and keymotifs are retained in common by all the TPP-dependent enzymes considered,which are essential for catalysis.

• Nilsson U, Lindqvist Y, Kluger R, Schneider G
• Crystal structure of transketolase in complex with thiamine thiazolonediphosphate, an analogue of the reaction intermediate, at 2.3 Aresolution.
• FEBS Lett. 1993; 326: 145-8
• Display abstract

• The crystal structure of the complex of transketolase and thiaminethiazolone diphosphate has been determined at 2.3 A resolution. Thecomplex has a structure which closely resembles that of this enzyme withthe cofactor ThDP. This is consistent with the observation that thebinding of the analogue to transketolase involves ground state rather thantransition state interactions. Since thiamine thiazolone diphosphateresembles an expected intermediate in the catalytic pathway, the structureof the intermediate was modelled from the crystal structure. Based on thismodel, enzymic groups responsible for binding of the intermediate andproton transfer during catalysis are suggested.

• Muller YA, Lindqvist Y, Furey W, Schulz GE, Jordan F, Schneider G
• A thiamin diphosphate binding fold revealed by comparison of the crystalstructures of transketolase, pyruvate oxidase and pyruvate decarboxylase.
• Structure. 1993; 1: 95-103
• Display abstract

• BACKGROUND: The crystal structures of three thiamin diphosphate-dependentenzymes that catalyze distinct reactions in basic metabolic pathways areknown. These enzymes--transketolase, pyruvate oxidase and pyruvatedecarboxylase--also require metal ions such as Ca2+ and Mg2+ as cofactorsand have little overall sequence similarity. Here, the crystal structuresof these three enzymes are compared. RESULTS: The three enzymes share asimilar pattern of binding of thiamin diphosphate and the metal ioncofactors. The enzymes function as multisubunit proteins, with eachpolypeptide chain folded into three alpha/beta domains. Two of thesedomains are involved in binding of the thiamin diphosphate and the metalion. These domains have the same topology of six parallel beta-strands andsurrounding alpha-helices. The thiamin diphosphate is bound in a cleft,formed by two domains from two different subunits. Only a few residues areconserved in all three enzymes and these are responsible for properbinding of the cofactors. CONCLUSIONS: Despite considerable differences inquaternary structure and lack of overall sequence homology, thiamindiphosphate binds to the three enzymes in a very similar fashion, and ageneral thiamin-binding fold can be revealed.

• Messerschmidt A et al.
• Refined crystal structure of ascorbate oxidase at 1.9 A resolution.
• J Mol Biol. 1992; 224: 179-205
• Display abstract

• The crystal structure of the fully oxidized form of ascorbate oxidase (EC1.10.3.3) from Zucchini has been refined at 1.90 A (1 A = 0.1 nm)resolution, using an energy-restrained least-squares refinement procedure.The refined model, which includes 8764 protein atoms, 9 copper atoms and970 solvent molecules, has a crystallographic R-factor of 20.3% for 85,252reflections between 8 and 1.90 A resolution. The root-mean-squaredeviation in bond lengths and bond angles from ideal values is 0.011 A and2.99 degrees, respectively. The subunits of 552 residues (70,000 Mr) arearranged as tetramers with D2 symmetry. One of the dyads is realized bythe crystallographic axis parallel to the c-axis giving one dimer in theasymmetric unit. The dimer related about this crystallographic axis issuggested as the dimer present in solution. Asn92 is the attachment sitefor one of the two N-linked sugar moieties, which has defined electrondensity for the N-linked N-acetyl-glucosamine ring. Each subunit is builtup by three domains arranged sequentially on the polypeptide chain andtightly associated in space. The folding of all three domains is of asimilar beta-barrel type and related to plastocyanin and azurin. Ananalysis of intra- and intertetramer hydrogen bond and van der Waalsinteractions is presented. Each subunit has four copper atoms bound asmononuclear and trinuclear species. The mononuclear copper has twohistidine, a cysteine and a methionine ligand and represents the type-1copper. It is located in domain 3. The bond lengths of the type-1 coppercentre are comparable to the values for oxidized plastocyanin. Thetrinuclear cluster has eight histidine ligands symmetrically supplied fromdomain 1 and 3. It may be subdivided into a pair of copper atoms withhistidine ligands whose ligating N-atoms (5 NE2 atoms and one ND1 atom)are arranged trigonal prismatic. The pair is the putative type-3 copper.The remaining copper has two histidine ligands and is the putativespectroscopic type-2 copper. Two oxygen atoms are bound to the trinuclearspecies as OH- or O2- and bridging the putative type-3 copper pair and asOH- or H2O bound to the putative type-2 copper trans to the copper pair.The bond lengths within the trinuclear copper site are similar tocomparable binuclear model compounds. The putative binding site for thereducing substrate is close to the type-1 copper.(ABSTRACT TRUNCATED AT400 WORDS)

• Schneider G, Lindqvist Y, Lundqvist T
• Crystallographic refinement and structure of ribulose-1,5-bisphosphatecarboxylase from Rhodospirillum rubrum at 1.7 A resolution.
• J Mol Biol. 1990; 211: 989-1008
• Display abstract

• The amino acid sequence of ribulose-1,5-bisphosphate carboxylase/oxygenasefrom Rhodospirillum rubrum has been fitted to the electron density maps.The resulting protein model has been refined to a nominal resolution of1.7 A using the constrained-restrained least-squares refinement program ofSussman and the restrained least-squares refinement program of Hendrickson& Konnert. The crystallographic refinement, based on 76,452 reflectionswith F greater than sigma (F) in the resolution range 5.5 to 1.7 Aresulted in a crystallographic R-factor of 18.0%. The asymmetric unitcontains one dimeric ribulose-1,5-biphosphate carboxylase molecule,consisting of 869 amino acid residues and 736 water molecules. Thegeometry of the refined model is close to ideal, with root-mean-squaredeviations of 0.018 A in bond lengths and 2.7 degrees in bond angles. Twoloop regions, comprising residues 54 to 63 and 324 to 335, and the lastten amino acid residues at the C terminus are disordered in our crystals.The expected trimodal distribution is obtained for the side-chain chi1-angles with a marked preference for staggered conformation. Thehydrogen-bonding pattern in the N-terminal beta-sheet and the parallelsheet in the beta/alpha-barrel is described. A number of hydrogen bondsand salt bridges are involved in domain-domain and subunit-subunitinteractions. The subunit-subunit interface in the dimer covers an area of2800 A2. Considerable deviations from the local 2-fold symmetry are foundat both the N terminus (residues 2 to 5) and the C terminus (residues 422to 457). Furthermore, loop 8 in the beta/alpha-barrel domain has adifferent conformation in the two subunits. A number of amino acidside-chains have different conformations in the two subunits. Most ofthese residues are located at the surface of the protein. An analysis ofthe individual temperature factors indicates a high mobility of theC-terminal region and for some of the loops at the active site. Thepositions and B-factors for 736 solvent sites have been refined (averageB: 45.9 A2). Most of the solvent molecules are bound as clusters to theprotein. The active site of the enzyme, especially the environment of theactivator Lys191 in the non-activated enzyme is described.Crystallographic refinement at 1.7 A resolution clearly revealed thepresence of a cis-proline at the active site. This residue is part of thehighly conserved region Lys166-Pro167-Lys168.

• Katti SK, LeMaster DM, Eklund H
• Crystal structure of thioredoxin from Escherichia coli at 1.68 Aresolution.
• J Mol Biol. 1990; 212: 167-84
• Display abstract

• The crystal structure of thioredoxin from Escherichia coli has beenrefined by the stereochemically restrained least-squares procedure to acrystallographic R-factor of 0.165 at 1.68 A resolution. In the finalmodel, the root-mean-square deviation from ideality for bond distances is0.015 A and for angle distances 0.035 A. The structure contains 1644protein atoms from two independent molecules, two Cu2+, 140 watermolecules and seven methylpentanediol molecules. Ten residues have beenmodeled in two alternative conformations. E. coli thioredoxin is a compactmolecule with 90% of its residues in helices, beta-strands or reverseturns. The molecule consists of two conformational domains, beta alphabeta alpha beta and beta beta alpha, connected by a single-turnalpha-helix and a 3(10) helix. The beta-sheet forms the core of themolecule packed on either side by clusters of hydrophobic residues.Helices form the external surface. The active site disulfide bridgebetween Cys32 and Cys35 is located at the amino terminus of the secondalpha-helix. The positive electrostatic field due to the helical dipole isprobably important for stabilizing the anionic intermediate during thedisulfide reductase function of the protein. The more reactive cysteine,Cys32, has its sulfur atom exposed to solvent and also involved in ahydrogen bond with a backbone amide group. Residues 29 to 37, whichinclude the active site cysteine residues, form a protrusion on thesurface of the protein and make relatively fewer interactions with therest of the structure. The disulfide bridge exhibits a right-handedconformation with a torsion angle of 81 degrees and 72 degrees about theS-S bond in the two molecules. Twenty-five pairs of water molecules obeythe noncrystallographic symmetry. Most of them are involved inestablishing intramolecular hydrogen-bonding interactions between proteinatoms and thus serve as integral parts of the folded protein structure.Methylpentanediol molecules often pack against the loops and stabilizetheir structure. Cu2+ used for crystallization exhibit a distortedoctahedral square bipyramid co-ordination and provide essential packinginteractions in the crystal. The two independent protein molecules arevery similar in conformation but distinctly different in atomic detail(root-mean-square = 0.94 A). The differences, which may be related to thecrystal contacts, are localized mostly to regions far from the activesite.

• Wright CS
• Refinement of the crystal structure of wheat germ agglutinin isolectin 2at 1.8 A resolution.
• J Mol Biol. 1987; 194: 501-29
• Display abstract

• The crystal structure of wheat germ agglutinin isolectin 2 has beenrefined by the restrained least-squares method of Hendrickson & Konnert(1980). The asymmetric unit of the C2 crystals contains two chemicallyidentical promoters related by a non-crystallographic 2-fold screwoperation. A total of 2290 protein atoms and 186 ordered water sitesrefined to a final R-factor of 0.179 and an average B-value of 21.6 A2,using 54% (15,601) of the total possible number of reflections in theresolution range 8 to 1.8 A with Fo greater than 3 sigma (Fo). The finalmodel conforms to stereochemically correct bond distances and angles withroot-mean-square (r.m.s.) values of 0.018 A and 3.3 degrees, respectively.Accuracy of this model is estimated to be 0.20 A on the basis of a Luzzatiplot. Main-chain atomic positions in the two independent promoters,designated I and II, agree with an r.m.s. deviation of 0.30 A (0.58 A forall atoms), indicating identical backbone conformation. The largestdiscrepancies are seen at flexible surface residues. One error wasdetected in the amino acid sequence at position 41 (Ser), which refinedsatisfactorily as a Trp. Loss of electron density for residue A171 duringthe course of refinement suggests either disorder or absence of thisC-terminal residue. The conformation of the polypeptide chain, which isfolded into four homologous 43-residue domains (A, B, C and D), wasanalyzed in terms of dihedral angles, backbone hydrogen bond lengths andCA-atom positions. The four domains were found to be very similaraccording to all these criteria and superposition of their CA-atomsyielded r.m.s. distances ranging from 0.36 to 0.72 A for the six possiblecomparisons [corrected]. Large deviations (greater than 1.0 A) are onlyseen in the five-residue segments that link adjacent domains and at the Nand C termini. Refinement has also allowed critical examination of each ofthe two unique sugar binding sites, referred to as "primary" and"secondary" sites, in different lattice environments. While the essentialtyrosyl side-chain in each of these sites (Y73, Y159) assumes preciseorientation for optimum hydrophobic contact with the N-acetyl methyl groupof the sugar ligand, side-chains involved in hydrogen bonds (S62, E115;and S148, D29) were found to be relatively flexible and able to adapttheir conformation to changes in environment. Ordered water structurepresent in these binding sites is not completely analogous in thedifferent environments.(ABSTRACT TRUNCATED AT 400 WORDS)