Benchmarking of structure refinement methods for protein complex models

Protein structure docking is the process in which the quaternary structure of a protein complex is predicted from individual tertiary structures of the protein subunits. Protein docking is typically performed in two main steps. The subunits are first docked while keeping them rigid to form the complex, which is then followed by structure refinement. Structure refinement is crucial for a practical use of computational protein docking models, as it is aimed for correcting conformations of interacting residues and atoms at the interface. Here, we benchmarked the performance of eight existing protein structure refinement methods in refinement of protein complex models. We show that the fraction of native contacts between subunits is by far the most straightforward metric to improve. However, backbone dependent metrics, based on the Root Mean Square Deviation proved more difficult to improve via refinement.     

Structural evidence for ligand-induced sequential conformational changes in glyceraldehyde 3-phosphate dehydrogenase

Glyceraldehyde 3-phosphate dehydrogenase is a tetramer of four chemically identical subunits which requires the cofactor nicotinamide adenine dinucleotide (NAD) for activity. The structure of the holo-enzyme from Bacillus stearothermophilus has recently been refined using X-ray data to 2.4 A resolution. This has facilitated the structure determination of both the apo-enzyme and the enzyme with one molecule of NAD bound to the tetramer. These structures have been refined at 4 A resolution using the constrained-restrained parameter structure factor least-squares refinement program CORELS. When combined with individual atomic temperature factors from the holo-enzyme, these refined models give crystallographic R factors of 30.2% and 30.4%, respectively, for data to 3 A resolution. The apo-enzyme has 222 molecular symmetry, and the subunit structure is related to that of the holo-enzyme by an approximate rigid-body rotation of the coenzyme binding domain by 4.3 degrees with respect to the catalytic domains, which form the core of the tetramer. The effect of this rotation is to shield the coenzyme and active site from solvent in the holo-enzyme. In addition to the rigid-body rotation, there is a rearrangement of several residues involved in NAD binding. The structure of the 1 NAD enzyme is asymmetric. The subunit which contains the bound NAD adopts a conformation very similar to that of a holo-enzyme subunit, while the other three unliganded subunits are very similar to the apo-enzyme conformation. This result provides unambiguous evidence for ligand-induced sequential conformational changes in B. stearothermophilus glyceraldehyde 3-phosphate dehydrogenase.     

Structure of quinoprotein methylamine dehydrogenase at 2.25 A resolution.

The three-dimensional structure of quinoprotein methylamine dehydrogenase from Thiobacillus versutus has been determined at 2.25 A resolution by a combination of multiple isomorphous replacement, phase extension by solvent flattening and partial structure phasing using molecular dynamics refinement. In the resulting map, the polypeptide chain for both subunits could be followed and an X-ray sequence was established. The tetrameric enzyme, made up of two heavy (H) and two light (L) subunits, is a flat parallellepiped with overall dimensions of approximately 76 x 61 x 45 A. The H subunit, comprising 370 residues, is made up of two distinct segments: the first 31 residues form an extension which embraces one of the L subunits; the remaining residues are found in a disc-shaped domain. This domain is formed by a circular arrangement of seven topologically identical four-stranded antiparallel beta-sheets, with approximately 7-fold symmetry. In spite of distinct differences, this arrangement is reminiscent of the structure found in influenza virus neuraminidase. The L subunit consists of 121 residues, out of which 53 form a beta-sheet scaffold of a central three-stranded antiparallel sheet flanked by two shorter two-stranded antiparallel sheets. The remaining residues are found in segments of irregular structure. This subunit is stabilized by six disulphide bridges, plus two covalent bridges involving the quinone co-factor and residues 57 and 107 of this subunit. The active site is located in a channel at the interface region between the H and L subunits, and the electron density in this part of the molecule suggests that the co-factor of this enzyme is not pyrrolo quinoline quinone (PQQ) itself, but might be instead a precursor of PQQ.     

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.     

Subunits of RNA polymerase in function and structure. 7. Structure of premature core enzyme.

The structure of premature core enzyme, an obligatory intermediate in both in vivo and in vitro assembly of Escherichia coli DNA-dependent RNA polymerase, was compared with that of native core enzyme. Though this assembled but inactive form of core enzyme harbors the gross conformation similar to that of native enzyme, minor and presumably local differences exist, which were identified by near-ultraviolet circular dichroism spectra, tritium-hydrogen exchange rate, protease sensitivity, intersubunit cross-linking rate by bifunctional reagents, sedimentation behavior, and elution profile from phosphocellulose. Taken together these results indicate that the core enzyme subunits are loosely associated in the premature core. The temperature-dependent maturation is required for the core subunits to be tightly associated, leading to the formation of structurally stable and functionally active RNA polymerase.     

Refined atomic model of glutamine synthetase at 3.5 A resolution

An atomic model of 43,692 non-hydrogen atoms has been determined for the 12-subunit enzyme glutamine synthetase from Salmonella typhimurium, by methods of x-ray diffraction including restrained least-squares atomic refinement against 65,223 unique reflections. At 3.5 A resolution the crystallographic R-factor (on 2 sigma data) is 25.8%. As reported earlier for the unrefined structure, the 12 subunits are arranged in two layers of six; at the interface of pairs of subunits within each layer, cylindrical active sites are formed by six anti-parallel beta strands contributed by one subunit and two strands by the neighboring subunit. This interpretation of the electron density map has now been supported by comparison with glutamine synthetase from Escherichia coli by the Fourier difference method. Each active site cylinder holds two Mn2+ ions, with each ion having as ligands three protein side chains and two water molecules (one water shared by both metals), as well as a histidyl side chain just beyond liganding distance. The protein ligands to Mn2+ 469 are Glu-131, Glu-212, and Glu-220; those to Mn2+ 470 are Glu-129, His-269, and Glu-357. The two layers of subunits are held together largely by the apolar COOH terminus, a helical thong, which inserts into a hydrophobic pocket formed by two neighboring subunits on the opposite ring. Also between layers, there is a hydrogen-bonded beta sheet interaction, as there is between subunits within a ring, but hydrophobic interactions account for most of the intersubunit stability. The central loop, which extends into the central aqueous channel, is subject to attack by at least five enzymes and is discussed as an enzyme "passive site."     

X-ray structure of nucleoside diphosphate kinase.

The X-ray structure of a point mutant of nucleoside diphosphate kinase (NDP kinase) from Dictyostelium discoideum has been determined to 2.2 A resolution. The enzyme is a hexamer made of identical subunits with a novel mononucleotide binding fold. Each subunit contains an alpha/beta domain with a four stranded, antiparallel beta-sheet. The topology is different from adenylate kinase, but identical to the allosteric domain of Escherichia coli ATCase regulatory subunits, which bind mononucleotides at an equivalent position. Dimer contacts between NDP kinase subunits within the hexamer are similar to those in ATCase. Trimer contacts involve a large loop of polypeptide chain that bears the site of the Pro----Ser substitution in Killer of prune (K-pn) mutants of the highly homologous Drosophila enzyme. Properties of Drosophila NDP kinase, the product of the awd developmental gene, and of the human enzyme, the product of the nm23 genes in tumorigenesis, are discussed in view of the three-dimensional structure and of possible interactions of NDP kinase with other nucleotide binding proteins.     

Purification and Properties of Gentisate 1,2-Dioxygenase from Rhodococcus erythropolis S-1

Gentisate 1,2-dioxygenase, which participates in salicylate and m-hydroxybenzoate metabolism, was purified from cell-free extracts of Rhodococcus erythropolis S-1, a Gram-positive bacterium. The purified enzyme gave a single band on native PAGE and SDS–PAGE. The molecular mass of the enzyme was estimated to be 328 kDa. The structure of the enzyme appears to be an octamer of identical subunits. The enzyme from this bacterium was similar in general enzymatic properties to a gentisate 1,2-dioxygenase from a Gram-negative bacterium except for molecular mass and structure.     

Three-dimensional structure of Cu,Zn-superoxide dismutase from spinach at 2.0 A resolution

The three-dimensional structure of Cu,Zn-superoxide dismutase from spinach leaves has been determined by X-ray crystal structure analysis. The atomic coordinates were refined at 2.0 A resolution using the Hendrickson and Konnert program for stereochemically restrained refinement against structure factors, which allowed the use of non-crystallographic symmetry. The crystallographic residual error for the refined model was 24.9%, with a root mean square deviation of 0.03 A from the ideal bond length and an average atomic temperature factor of 9.6 A. A dimeric molecule of the enzyme is comprised of two identical subunits related by a non-crystallographic 2-fold axis. Each subunit of 154 amino acid residues is composed primarily of eight anti-parallel beta-strands that form a flattened cylinder, plus three external loops. The main-chain hydrogen bonds primarily link the beta-strands. The overall structure of this enzyme is quite similar to that of the bovine dismutase except for some parts. The single disulfide bridge (Cys57-Cys146) and the salt bridge (Arg79-Asp101) may stabilize the loop regions of the structure. The Cu2+ and Zn2+ ions in the active site lie 6.1 A apart at the bottom of the long channel. The Cu2+ ligands (ND1 of His-46, and NE2 of His-48, -63, and -120) show an uneven tetrahedral distortion from a square plane. The Zn2+ ligands (ND1 of His-63, -71, and -80 and OD1 of Asp-83) show an almost tetrahedral geometry. The imidazole ring of His-63 forms a bridge between the Cu2+ and Zn2+ ions.(ABSTRACT TRUNCATED AT 250 WORDS)     

Structures of D-xylose isomerase from Arthrobacter strain B3728 containing the inhibitors xylitol and D-sorbitol at 2.5 A and 2.3 A resolution, respectively

The structures of D-xylose isomerase from Arthrobacter strain B3728 containing the polyol inhibitors xylitol and D-sorbitol have been solved at 2.5 A and 2.3 A, respectively. The structures have been refined using restrained least-squares refinement methods. The final crystallographic R-factors for the D-sorbitol (xylitol) bound molecules, for 43,615 (32,989) reflections are 15.6 (14.7). The molecule is a tetramer and the asymmetric unit of the crystal contains a dimer, the final model of which, incorporates a total of 6086 unique protein, inhibitor and magnesium atoms together with 535 bound solvent molecules. Each subunit of the enzyme contains two domains: the main domain is a parallel-stranded alpha-beta barrel, which has been reported in 14 other enzymes. The C-terminal domain is a loop structure consisting of five helical segments and is involved in intermolecular contacts between subunits that make up the tetramer. The structures have been analysed with respect to molecular symmetry, intersubunit contacts, inhibitor binding and active site geometry. The refined model shows the two independent subunits to be similar apart from local deviations due to solvent contacts in the solvent-exposed helices. The enzyme is dependent on a divalent cation for catalytic activity. Two metal ions are required per monomer, and the high-affinity magnesium(II) site has been identified from the structural results presented here. The metal ion is complexed, at the high-affinity site, by four carboxylate side-chains of the conserved residues, Glu180, Glu216, Asp244 and Asp292. The inhibitor polyols are bound in the active site in an extended open chain conformation and complete an octahedral co-ordination shell for the magnesium cation via their oxygen atoms O-2 and O-4. The active site lies in a deep pocket near the C-terminal ends of the beta-strands of the barrel domain and includes residues from a second subunit. The tetrameric molecule can be considered to be a dimer of "active" dimers, the active sites being composed of residues from both subunits. The analysis has revealed the presence of several internal salt-bridges stabilizing the tertiary and quaternary structure. One of these, between Asp23 and Arg139, appears to play a key role in stabilizing the active dimer and is conserved in the known sequences of this enzyme.(ABSTRACT TRUNCATED AT 400 WORDS)     

Cloning and sequence analysis of a cDNA for human glycosylasparaginase. A single gene encodes the subunits of this lysosomal amidase

We have isolated a full-length cDNA (HPAsn.6) for human placenta glycosylasparaginase using a 221-bp PCR amplified fragment containing rat liver asparaginase gene sequences. The deduced amino acid sequence from the human clone showed sequence identity to both the alpha and beta subunits of the rat enzyme. The human enzyme is encoded as a 34.6 kDa polypeptide that is post-translationally processed to generate two subunits of approx. 19.5 (alpha) and 15 (beta) kDa. A charge enriched region is present at the predicted site where cleavage occurs. Using polyclonal antibodies against the alpha and beta subunits of rat liver asparaginase, we have shown that the human enzyme is similar in structure to the rat enzyme.     

Subunit location and sequences of the cysteinyl peptides of pig heart NAD-dependent isocitrate dehydrogenase

Pig heart NAD-dependent isocitrate dehydrogenase has a subunit structure consisting of alpha 2 beta gamma, with the alpha subunit exhibiting a molecular weight of 39,000 and the beta and gamma each having molecular weights of 41,000. The amino-terminal sequences (33-35 residues) and the cysteinyl peptide sequences have now been determined by using subunits separated by chromatofocusing or isoelectric focusing and electroblotting. Displacement of the N-terminal sequence of the alpha subunit by 11-12 amino acids relative to that of the larger beta and gamma subunits reveals a 17 amino acid region of great similarity in which 10 residues are identical in all three subunits. The complete enzyme has 6.0 free SH groups per average subunit of 40,000 daltons, but yields 15 distinguishable cysteines in isolated tryptic peptides. Six distinct cysteines in sequenced peptides have been located in the alpha subunit. The beta and gamma subunits contain seven and five cysteines, respectively, with tryptic peptides containing three cysteines being common to the beta and gamma subunits. The three subunits appear to be closely related, but beta and gamma are more similar to each other than either is to the alpha subunit. The NAD-specific isocitrate dehydrogenase from pig heart has been shown to have 2 binding sites/enzyme tetramer for isocitrate, manganous ion, NAD+, and the allosteric activator ADP [Colman, R. F. (1983) Pept. Protein Rev. 1, 41-69]. It is proposed that the catalytically active tetrameric enzyme is organized as a dimer of dimers in which the alpha beta and alpha gamma dimers are nonidentical but functionally similar.     

Pig heart short chain L-3-hydroxyacyl-CoA dehydrogenase revisited: sequence analysis and crystal structure determination.

Short chain L-3-hydroxyacyl CoA dehydrogenase (SCHAD) is a soluble dimeric enzyme critical for oxidative metabolism of fatty acids. Its primary sequence has been reported to be conserved across numerous tissues and species with the notable exception of the pig heart homologue. Preliminary efforts to solve the crystal structure of the dimeric pig heart SCHAD suggested the unprecedented occurrence of three enzyme subunits within the asymmetric unit, a phenomenon that was thought to have hampered refinement of the initial chain tracing. The recently solved crystal coordinates of human heart SCHAD facilitated a molecular replacement solution to the pig heart SCHAD data. Refinement of the model, in conjunction with the nucleotide sequence for pig heart SCHAD determined in this paper, has demonstrated that the previously published pig heart SCHAD sequence was incorrect. Presented here are the corrected amino acid sequence and the high resolution crystal structure determined for pig heart SCHAD complexed with its NAD+ cofactor (2.8 A; R(cryst) = 22.4%, R(free) = 28.8%). In addition, the peculiar phenomenon of a dimeric enzyme crystallizing with three subunits contained in the asymmetric unit is described.     

Crystal structure of the tetrameric inositol 1-phosphate phosphatase (TM1415) from the hyperthermophile, Thermotoga maritima

The structure of the first tetrameric inositol monophosphatase (IMPase) has been solved. This enzyme, from the eubacterium Thermotoga maritima, similarly to its archaeal homologs exhibits dual specificity with both IMPase and fructose-1,6-bisphosphatase activities. The tetrameric structure of this unregulated enzyme is similar, in its quaternary assembly, to the allosterically regulated tetramer of fructose-1,6-bisphosphatase. The individual dimers are similar to the human IMPase. Additionally, the structures of two crystal forms of IMPase show significant differences. In the first crystal form, the tetrameric structure is symmetrical, with the active site loop in each subunit folded into a beta-hairpin conformation. The second form is asymmetrical and shows an unusual structural change. Two of the subunits have the active site loop folded into a beta-hairpin structure, whereas in the remaining two subunits the same loop adopts an alpha-helical conformation.     

A prediction of the amino acids and structures involved in DNA recognition by type I DNA restriction and modification enzymes.

The S subunits of type I DNA restriction/modification enzymes are responsible for recognising the DNA target sequence for the enzyme. They contain two domains of approximately 150 amino acids, each of which is responsible for recognising one half of the bipartite asymmetric target. In the absence of any known tertiary structure for type I enzymes or recognisable DNA recognition motifs in the highly variable amino acid sequences of the S subunits, it has previously not been possible to predict which amino acids are responsible for sequence recognition. Using a combination of sequence alignment and secondary structure prediction methods to analyse the sequences of S subunits, we predict that all of the 51 known target recognition domains (TRDs) have the same tertiary structure. Furthermore, this structure is similar to the structure of the TRD of the C5-cytosine methyltransferase, Hha I, which recognises its DNA target via interactions with two short polypeptide loops and a beta strand. Our results predict the location of these sequence recognition structures within the TRDs of all type I S subunits.     

Cyanomet human hemoglobin crystallized under physiological conditions exhibits the Y quaternary structure

Cyanomet human hemoglobin has been crystallized at a chloride ion concentration and pH similar to physiological conditions. Molecular replacement calculations definitively show that the hemoglobin subunits are arranged in the Y quaternary form recently discovered in carbon monoxy hemoglobin Ypsilanti (99 Asp–Tyr), and subsequently observed in carbon monoxy normal human hemoglobin crystallized at low ionic strength and low pH. The structure has been refined at 2.09 Å resolution to an R-value of 0.232, and further refinement is currently underway. Although the refinement is not yet complete, our results are the first indication that the Y structure may represent an important quaternary form of liganded hemoglobin under physiological buffer conditions. These results suggest the need for a reexamination of structure–function correlations in the hemoglobin system. © 1994 John Wiley & Sons, Inc.     

Cytochromec oxidase metal centers: Location and function

Cytochromec oxidase ofParacoccus denitrificans is spectroscopically and functionally very similar to the mammalian enzyme. However, it has a very much simpler quaternary structure, consisting of only three subunits instead of the 13 of the bovine enzyme. The known primary structure of theParacoccus denitrificans subunits, the knowledge of a large number of sequences from other species, and data on the controlled proteolytic digestion of the enzyme allow structural restrictions to be placed on the models describing the binding of the active metal centers to the polypeptide structure.     

The C-S Lyases of Higher Plants : Direct Comparison of the Physical Properties of Homogeneous Alliin Lyase of Garlic (Allium sativum) and Onion (Allium cepa)

Garlic and onion alliin lyases, although from closely related species, have many differences. The two enzymes differ in their K_m values, pH optima, and isoelectric points. There is a major difference in their molecular weight and subunit structure. The garlic holoenzyme has a molecular weight of 85,000 and consists of two subunits of molecular weight 42,000. The onion enzyme has a holoenzyme molecular weight of 200,000 composed of four subunits of molecular weight 50,000. The onion enzyme is much more difficult to dissociate into its subunits which suggests differences in subunit interaction between the two enzymes. The dimeric stucture of the garlic and the tetrameric structure of the onion enzyme is consistent with a coenzyme content (pyridoxal-5′-phosphate) equivalent to one mole per subunit. The two enzymes vary vastly in their spectra, the onion enzyme having a lower pyridoxal-5′-phosphate absorbance at 430 nanomoles and an inability to react with l-cysteine. Both enzymes are glycoproteins and bind to concanavalin A-Sepharose columns. The onion alliin lyase binds more tightly than the garlic enzyme. The amino acid content of both enzymes is similar as is the carbohydrate content. However, upon hydrolysis the onion lyase does yield more mannose units than the garlic enzyme which is consistent with the former's stronger affinity for concanavalin A.     

Crystal structure of Agaricus bisporus mushroom tyrosinase: identity of the tetramer subunits and interaction with tropolone

Tyrosinase catalyzes the conversion of phenolic compounds into their quinone derivatives, which are precursors for the formation of melanin, a ubiquitous pigment in living organisms. Because of its importance for browning reactions in the food industry, the tyrosinase from the mushroom Agaricus bisporus has been investigated in depth. In previous studies the tyrosinase enzyme complex was shown to be a H(2)L(2) tetramer, but no clues were obtained of the identities of the subunits, their mode of association, and the 3D structure of the complex. Here we unravel this tetramer at the molecular level. Its 2.3 ? resolution crystal structure is the first structure of the full fungal tyrosinase complex. The complex comprises two H subunits of ～392 residues and two L subunits of ～150 residues. The H subunit originates from the ppo3 gene and has a fold similar to other tyrosinases, but it is ～100 residues larger. The L subunit appeared to be the product of orf239342 and has a lectin-like fold. The H subunit contains a binuclear copper-binding site in the deoxy-state, in which three histidine residues coordinate each copper ion. The side chains of these histidines have their orientation fixed by hydrogen bonds or, in the case of His85, by a thioether bridge with the side chain of Cys83. The specific tyrosinase inhibitor tropolone forms a pre-Michaelis complex with the enzyme. It binds near the binuclear copper site without directly coordinating the copper ions. The function of the ORF239342 subunits is not known. Carbohydrate binding sites identified in other lectins are not conserved in ORF239342, and the subunits are over 25 ? away from the active site, making a role in activity unlikely. The structures explain how calcium ions stabilize the tetrameric state of the enzyme.     

ON THE MECHANISM OF STIMULATION OF AMP-AMINOHYDROLASE ACTIVITY BY HEXOKINASE IN BRAIN TISSUE

—A hexokinase has been isolated from brain tissue on Sephadex G-100 and DEAE cellulose which is similar to yeast enzyme in stimulating the AMP-aminohydrolase activity of rat brain soluble fractions. This effect of hexokinase is influenced neither by N-acetyl-glucosamine nor noradrenaline. An isoenzyme of hexokinase isolated from brain tissue on DEAE cellulose, having properties similar to that of the muscle enzyme, has no effect on AMP-aminohydrolase activity. The activating effect of yeast hexokinase is not due to its oligomeric structure. Enzyme subunits obtained by the treatment of native yeast enzyme by urea also activate AMP-aminohydrolase of rat brain soluble fractions.