Comparison of the evolution of heme binding and NAD binding proteins.

Of the 85 three-dimensionally characterized residues of cytochrome b5, 51 are structurally and topologically equivalent to the globin fold. When these proteins have been superimposed, the heme irons are found to be less than 1.4 A separated and the heme normals are inclined by less than 9.5 degrees. Comparison of minimum base changes per codon between heme binding and NAD binding proteins are of the same order.     

A comparison of the heme binding pocket in globins and cytochrome b5.

Of the 85 three-dimensionally characterized residues of cytochrome b5, 51 are found to be structurally and topologically equivalent to the globin fold. When these proteins have been superimposed, the heme irons are found to be less than 1.4 A separated and the heme normals are inclined by less than 9.5 degrees. The proximal histidine of the globins and two adjacent helices are equivalent to the sixth iron ligand and adjacent helices of cytochrome b5. Larger differences in structure are observed on the distal side of the heme, coincident with the most changeable part of the globin structures. The heme itself is rotated by 53 degrees about its normal but such a change is energetically minimal and conservative as the heme side groups are not directly involved in the function of the molecules. The beta-sheet of cytochrome b5 is inserted into a corresponding cavity of the globins forming an additional lining to the heme pocket. The roughly 50 residues missing at the carboxy end of the known cytochrome b5 fragment could correspond in part to the H helix in the globins. While it would seem probable that these similarities represent divergent evolution from a primordial heme-binding protein, the possibility of structural convergence to a functionally satisfactory protein cannot be excluded.     

Comparison of beef liver and Penicillium vitale catalases

The structures of Penicillium vitale and beef liver catalase have been determined to atomic resolution. Both catalases are tetrameric proteins with deeply buried heme groups. The amino acid sequence of beef liver catalase is known and contains (at least) 506 amino acid residues. Although the sequence of P. vitale catalase has not yet been determined chemically, 670 residues have been built into the 2 A resolution electron density map and have been given tentative assignments. A large portion of each catalase molecule (91% of residues in beef liver catalase and 68% of residues in P. vitale catalase) shows structural homology. The root-mean-square deviation between 458 equivalenced C alpha atoms is 1.17 A. The dissimilar parts include a small fragment of the N-terminal arm and an additional "flavodoxin-like" domain at the carboxy end of the polypeptide chain of P. vitale catalase. In contrast, beef liver catalase contains one bound NADP molecule per subunit in a position equivalent to the chain region, leading to the flavodoxin-like domain, of P. vitale catalase. The position and orientation of the buried heme group in the two catalases, relative to the mutually perpendicular molecular dyad axes, are identical within experimental error. A mostly hydrophobic channel leads to the buried heme group. The surface opening to the channel differs due to the different disposition of the amino-terminal arm and the presence of the additional flavodoxin-like domain in P. vitale catalase. Possible functional implications of these comparisons are discussed.     

Exploring structural homology of proteins.

A method for systematically comparing the folding of the three-dimensional structures of proteins has been developed. A search function, plotted in terms of three Eulerian angles, represents the number of sequentially equivalenced amino acids. For each orientation one protein structure is rotated about its center of mass with respect to the other and probabilities are calculated which estimate the degree of structural parallelism. The structurally equivalent residues with highest probabilities are then selected for the best common topology. It was observed that, when structures containing about 150 residues were compared, the random background had a mean value of around 14 residues and the standard deviation was approximately nine residues. The method has been shown to be successful in determining the similarity of the NAD binding domains of lactate dehydrogenase and glyceraldehyde-3-phosphate dehydrogenase, and in comparing the heme binding fold of cytochrome b₅ with the globins.Application of the method to compare hen egg white lysozyme and T4 phage lysozyme led to a single significant peak of 62 residues. The structural homology indicated by this peak showed that the substrate, as bound to hen egg white lysozyme, has a corresponding binding site in the large cleft of the phage lysozyme. The predicted binding site of N-acetyl glucosamine at position C compares well with an N-acetyl glucosamine center observed to bind to crystalline phage lysozyme (B. W. Matthews, personal communication).Some results for the comparison of the two Fe-S cage binding domains of ferredoxin are also presented.     

Progressive combinatorial algorithm for multiple structural alignments: application to distantly related proteins

MALECON is a progressive combinatorial procedure for multiple alignments of protein structures. It searches a library of pairwise alignments for all three-protein alignments in which a specified number of residues is consistently aligned. These alignments are progressively expanded to include additional proteins and more spatially equivalent residues, subject to certain criteria. This action involves superimposing the aligned proteins by their hitherto equivalent residues and searching for additional Calpha atoms that lie close in space. The performance of MALECON is illustrated and compared with several extant multiple structure alignment methods by using as test the globin homologous superfamily, the OB and the Jellyrolls folds. MALECON gives better definitions of the common structural features in the structurally more diverse proteins of the OB and Jellyrolls folds, but it yields comparable results for the more similar globins. When no consistent multiple alignments can be derived for all members of a protein group, our procedure is still capable of automatically generating consistent alignments and common core definitions for subgroups of the members. This finding is illustrated for proteins of the OB fold and SH3 domains, believed to share common structural features, and should be very instrumental in homology modeling and investigations of protein evolution.     

The N-terminal region of the heme-regulated eIF2alpha kinase is an autonomous heme binding domain.

The N-terminal domain (NTD) of the heme-regulated eukaryotic initiation factor (eIF)2alpha kinase (HRI) was aligned to sequences in the NCBI data base using ENTREZ and a PAM250 matrix. Significant similarity was found between amino acids 11-118 in the NTD of rabbit HRI and amino acids 16-120 in mammalian alpha-globins. Several conserved amino acid residues present in globins are conserved in the NTD of HRI. His83 of HRI was predicted to be equivalent to the proximal heme ligand (HisF8) that is conserved in all globins. Molecular modeling of the NTD indicated that its amino acid sequence was compatible with the globin fold. Recombinant NTD (residues 1-159) was expressed in Escherichia coli. Spectral analysis of affinity purified recombinant NTD indicated that the NTD contained stably bound hemin. Mutational analysis indicated that His83 played a critical structural role in the stable binding of heme to the NTD, and was required to stabilize full length HRI synthesized de novo in the rabbit reticulocyte lysate. These results indicate that the NTD of HRI is an autonomous heme-binding domain, with His83 possibly serving as the proximal heme binding ligand.     

pH dependence of heme electrochemistry in cytochromes investigated by multiconformation continuum electrostatic calculations

Cytochromes belong to a diverse family of heme-containing redox proteins that function as intermediaries in electron transfer chains. They can be soluble, extrinsic, or intrinsic membrane proteins, and are found in different structural motifs (globin, 4-helix bundles, alpha beta roll, beta sandwich). Measured electrochemical midpoint potentials vary over a wide range even though the basic redox reaction at the heme is the same for all cytochromes. The perturbation of the heme electrochemistry is induced by the protein structure. Also, the pH dependence varies since it depends on the strength of interaction between the heme and surrounding residues as well as the ionization states of these groups. Multiconformation continuum electrostatics (MCCE) has been used to investigate the pH dependence of heme electrochemistry in cytochromes with different folds. Often propionates are the primary contributors for pH dependence especially if they are partially protonated in the reduced heme as it is shown for globin cytochrome c551 P. aeruginosa and cytochrome b5 R. norvegicus (alpha beta roll). However, if the propionates are already fully ionized at a certain pH they do not contribute to the pH dependence even if they have big interaction with the heme. At pH 7 there is no propionate contribution for cytochrome f C. reinhardtii (beta sandwich) and the 4-helix bundle c' R. palustris. Other residues can also change their ionization significantly during heme oxidation and therefore be involved in proton release and pH dependence. These residues have been identified for different cytochrome types.     

Mapping protein matrix cavities in human cytoglobin through Xe atom binding

Cytoglobin is the fourth recognized globin type, almost ubiquitously distributed in human tissues; its function is still poorly understood. Cytoglobin displays a core region of about 150 residues, structurally related to hemoglobin and myoglobin, and two extra segments, about 20 residues each, at the N- and C-termini. The core region hosts a large apolar cavity, held to provide a ligand diffusion pathway to/from the heme, and/or ligand temporary docking sites. Here we report the crystal structure (2.4A resolution, R-factor 19.1%) of a human cytoglobin mutant bearing the CysB2(38) --> Ser and CysE9(83) --> Ser substitutions (CYGB*), treated under pressurized xenon. Three Xe atoms bind to the heme distal site region of CYGB* mapping the protein matrix apolar cavity. Despite the conserved globin fold, the cavity found in CYGB* is structured differently from those recognized to play a functional role in myoglobin, neuroglobin, truncated hemoglobins, and Cerebratulus lacteus mini-hemoglobin.     

X-ray structural evidence for a local helix-loop transition in alpha-lactalbumin.

The three-dimensional structure of human alpha-lactalbumin for two crystal forms has been determined by x-ray analysis. One crystal (the form LT) was obtained at pH 4.2 and room temperature, while the other crystal (the form HT) was grown at pH 6.5 and 37 degrees C. The backbone structure for Lys1-Ile95 residues is almost conserved between the two structures as indicated by the root mean square difference of 0.30 A for the superposition of equivalent C alpha atoms. The calcium ion is surrounded by seven oxygen atoms of three carboxyl groups, two carbonyl groups, and two water molecules, which form a distorted pentagonal bipyramid in both structures. A large difference in polypeptide folding is found in the region of Leu96-Leu123 residues. Especially in the region of Trp104-Cys111 residues, a distorted alpha-helix is observed in the form HT while a loop structure is formed in the other crystal. The fact that the crystals of both forms appeared in the same batch at pH 6.5 and room temperature indicates that the human alpha-lactalbumin structure is highly fluctuated in solution and the folding and unfolding of the alpha-helix of Trp104-Cys111 residues are in equilibrium. Since the crystal of the form HT exclusively appeared around the physiological temperature, the structure of this form can be considered as the native structure. The partially unfolded structure in the form LT indicates that the local denaturation occurs even at room temperature.     

Comparison of solvent-inaccessible cores of homologous proteins: definitions useful for protein modelling

The three-dimensional structures of 41 homologous proteins (belonging to eight families) were compared by pairwise superposition. A subset of 'core' residues was defined as those whose side chains have less than 7% of their surface exposed to solvent. This subset has significantly higher sequence identity and lower root mean square (RMS) alpha carbon separation than for all topologically equivalent residues in the structure, when members of a protein family are superposed. For such superpositions the relationship between RMS distance and percentage sequence identity of this subset of residues is similar to that for all equivalent residues, although some variation is observed between families of proteins which are predominantly beta sheet and those which are mainly alpha helix. The definition of a structurally more conserved core may be useful in model building proteins from an homologous family. The RMS differences of coordinates of structures of proteins with identical sequences are found to be related to the resolutions of the structures.     

Primary sequence and solution conformation of ferrocytochrome c-552 from Nitrosomonas europaea.

Cytochrome c-552 from Nitrosomonas europaea is a 9.1-kDa monoheme protein that is a member of the bacterial cytochrome c-551 family. The gene encoding for c-552 has been cloned and sequenced and the primary sequence of the product deduced. Proton resonance assignments were made for all main-chain and most side-chain protons in the diamagnetic, reduced form by two-dimensional NMR techniques. Distance constraints (1056) were determined from nuclear Overhauser enhancements, and torsion angle constraints (88) were determined from scalar coupling estimates. Solution conformations for the protein were computed by the hybrid distance geometry-simulated annealing approach. For 20 computed structures, the root mean squared deviation from the average position of equivalent atoms was 0.84 A (sigma = 0.12) for backbone atoms over all residues. Analysis by residue revealed there were three regions clearly less well defined than the rest of the protein: the first two residues at the N-terminus, the last two at the C-terminus, and a loop region from residues 34 to 40. Omitting these regions from the comparison, the root mean squared deviation was 0.61 A (sigma = 0.13) for backbone atoms, 0.86 A (sigma = 0.12) for all associated heavy atoms, and 0. 43 A (sigma = 0.17) for the heme group. The global folding of the protein is consistent with others in the c-551 family. A deletion at the N-terminus relative to other family members had no impact on the global folding, whereas an insertion at residue 65 did affect the way the polypeptide packs against the methionine-ligated side of the heme. The effects of specific substitutions will be discussed. The structure of c-552 serves to delineate essential features of the c-551 family.     

Crystal structure of cytoglobin: the fourth globin type discovered in man displays heme hexa-coordination

Cytoglobin is a recently discovered hemeprotein belonging to the globin superfamily together with hemoglobin, myoglobin and neuroglobin. Although distributed in almost all human tissues, cytoglobin has not been ascribed a specific function. Human cytoglobin is composed of 190 amino acid residues. Sequence alignments show that a protein core region (about 150 residues) is structurally related to hemoglobin and myoglobin, being complemented by about 20 extra residues both on the N and C termini. In the absence of exogenous ligands (e.g. O2), the cytoglobin distal HisE7 residue is coordinated to the heme Fe atom, thus decreasing the ligand affinity. The crystal structure of human cytoglobin (2.1 A resolution, 21.3% R-factor) highlights a three-over-three alpha-helical globin fold, covering residues 18-171; the 1-17 N-terminal, and the 172-190 C-terminal residue segments are disordered in both molecules of the crystal asymmetric unit. Heme hexa-coordination is evident in one of the two cytoglobin chains, whereas alternate conformation for the heme distal region, achieving partial heme penta-coordination, is observed in the other. Human cytoglobin displays a large apolar protein matrix cavity, next to the heme, not related to the myoglobin cavities recognized as temporary ligand docking stations. The cavity, which may provide a heme ligand diffusion pathway, is connected to the external space through a narrow tunnel nestled between the globin G and H helices.     

Detecting local structural similarity in proteins by maximizing number of equivalent residues

A new algorithm for superimposing protein structures based on maximizing the number of spatially equivalent residues is introduced. The algorithm works in three distinct steps. First, the optimal residue map is calculated by structural alignment. By default, the double dynamic programming algorithm, as implemented in the program ASH, was used for the structure alignment step, but we also present results based on alignments imported from three other programs (Dali, CE, and VAST).Second, the structures are spatially superimposed such that the effective number of equivalent residues (NER)--aligned residue pairs that can be spatially overlapped--is maximized. The NER score is an analytic, differentiable similarity function that rewards spatially equivalent residues but ignores non-equivalent ones. Maximization of the NER score results in accurate superpositions in cases where root mean square deviation (RMSD) minimization fails. Third, the NER function is used in conjunction with traditional dynamic programming to realign the structures based on the proximity of residues in the superposition. Results are presented for a wide range of superposition problems and compared to results from Dali, CE, and VAST. In addition, several structure-structure pairs that show only partial similarity are discussed, and results are compared to those from the LGA, SARF2, and ThreeCa programs.     

Crystal structures of recombinant and native soybean beta-conglycinin beta homotrimers.

The crystal structures of recombinant and native beta homotrimers of soybean beta-conglycinin were determined by X-ray crystallography at 2.7 and 2.8 A resolutions, respectively. The crystals of the recombinant and native beta homotrimers belong to space group P21 with cell parameters a = 80.51 A, b = 63.48 A, c = 131.43 A, and beta = 90.01 degrees and with cell parameters a = 82.78 A, b = 69.47 A, c = 125.33 A and beta = 97.22 degrees, respectively. The beta monomers consist of amino-terminal and carboxyl-terminal modules that are very similar to each other and are related by a pseudo-dyad axis. Each module of the beta monomer is subdivided into a core and a loop domain. These structural features of both beta homotrimers are consistent with those of canavalin and phaseolin, which are similar vicilin class proteins. The superposition of the models of the native and recombinant beta monomers shows a root mean square deviation of 0.43-0.51 A for 343 common Calpha atoms within 2.0 A. This result indicates that the N-linked glycans do not influence the final structure of the beta homotrimer. Comparison of the models of beta-conglycinin, phaseolin and canavalin indicates that beta-conglycinin resembles canavalin rather than phaseolin, and that canavalin and phaseolin differ the most among them. The evolutional relationships are discussed.     

Solution conformation of cytochrome c-551 from Pseudomonas stutzeri ZoBell determined by NMR.

1H NMR spectroscopy and solution structure computations have been used to examine ferrocytochrome c-551 from Pseudomonas stutzeri ZoBell (ATCC 14405). Resonance assignments are proposed for all main-chain and most side-chain protons. Stereospecific assignments were also made for some of the beta-methylene protons and valine methyl protons. Distance constraints were determined based upon nuclear Overhauser enhancements between pairs of protons. Dihedral angle constraints were determined from estimates of scalar coupling constants and intra-residue NOEs. Twenty structures were calculated by distance geometry and refined by energy minimization and simulated annealing on the basis of 1012 interproton distance and 74 torsion angle constraints. Both the main-chain and side-chain atoms are well defined except for two terminal residues, and some side-chain atoms located on the molecular surface. The average root mean squared deviation in the position for equivalent atoms between the 20 individual structures and the mean structure obtained by averaging their coordinates is 0.56 +/- 0.10 A for the main-chain atoms, and 0.95 +/- 0.09 A for all nonhydrogen atoms of residue 3 to 80 plus the heme group. The average structure was compared with an analogous protein, cytochrome c-551 from pseudomonas stutzeri. The main-chain folding patterns are very consistent, but there are some differences, some of which can be attributed to the loss of normally conserved aromatic residues in the ZoBell c-551.     

Models for the complexes formed between cytochrome b5 and the subunits of methemoglobin

Computer graphics-generated models for the electron transfer complexes formed between cytochrome b5 and the subunits of methemoglobin are proposed. For both complexes, the orientation allowing optimal hydrogen bonding involves interaction between negatively charged residues on cytochrome b5 and positively charged residues on methemoglobin. In each complex, the heme groups of the interacting species are coplanar with the edges of the heme groups separated by 7-8 A and with the iron atoms 16 A apart. For the alpha-chain X cytochrome b5 complex, alpha-chain residues 56 (Lys), 60 (Lys), and 90 (Lys) interact with cytochrome b5 residues 44 (Glu), 43 (Glu), and 60 (Asp) respectively. A fourth hydrogen bond involves alpha-61 (Lys) bridging between a heme propionate from cytochrome b5 and a heme propionate from the alpha-chain. The contacts present in the beta-chain X cytochrome b5 complex involve hydrogen-bonding between beta-chain lysyl residues 59, 61, 65, and 95, and cytochrome b5 residues 48 (Glu), 44 (Glu), 43 (Glu), and 60 (Asp) respectively. An additional hydrogen bond can be formed by bridging of the epsilon-amino group of beta-66 (Lys) between a heme propionate from cytochrome b5 and a beta-chain heme propionate. In each complex, two nonionic interactions, one on each side of the heme groups, are also suggested. These interactions appear to effectively exclude external water molecules from the center of the protein-protein interaction domain. Comparison of the proposed binding loci for cytochrome b5 on the methemoglobin subunits with those proposed on cytochrome c reveals considerable structural homology between the cytochrome b5 binding sites.     

Significantly Enhanced Heme Retention Ability of Myoglobin Engineered to Mimic the Third Covalent Linkage by Nonaxial Histidine to Heme (Vinyl) in Synechocystis Hemoglobin

Heme proteins, which reversibly bind oxygen and display a particular fold originally identified in myoglobin (Mb), characterize the “hemoglobin (Hb) superfamily.” The long known and widely investigated Hb superfamily, however, has been enriched by the discovery and investigation of new classes and members. Truncated Hbs typify such novel classes and exhibit a distinct two-on-two α-helical fold. The truncated Hb from the freshwater cyanobacterium Synechocystis exhibits hexacoordinate heme chemistry and bears an unusual covalent bond between the nonaxial His¹¹⁷ and a heme porphyrin 2-vinyl atom, which remains tightly associated with the globin unlike any other. It seems to be the most stable Hb known to date, and His¹¹⁷ is the dominant force holding the heme. Mutations of amino acid residues in the vicinity did not influence this covalent linkage. Introduction of a nonaxial His into sperm whale Mb at the topologically equivalent position and in close proximity to vinyl group significantly increased the heme stability of this prototype globin. Reversed phase chromatography, electrospray ionization-MS, and MALDI-TOF analyses confirmed the presence of covalent linkage in Mb I107H. The Mb mutant with the engineered covalent linkage was stable to denaturants and exhibited ligand binding and auto-oxidation rates similar to the wild type protein. This indeed is a novel finding and provides a new perspective to the evolution of Hbs. The successful attempt at engineering heme stability holds promise for the production of stable Hb-based blood substitute.     

Investigation of the solution conformation of cytochrome c-551 from Pseudomonas stutzeri.

1H NMR spectroscopy and solution structure computations have been used to examine ferrocytochrome c-551 from Pseudomonas stutzeri (ATCC 17588). Resonance assignments are proposed for all main-chain and most side-chain protons. Distance constraints were determined on the basis of nuclear Overhauser enhancements between pairs of protons. Dihedral angle constraints were determined from estimates of scaler coupling constants. Twenty-four structures were calculated by distance geometry and refined by energy minimization and simulated annealing on the basis of 1033 interproton distance and 57 torsion angle constraints. Both the main-chain and side-chain atoms are well defined except for a loop region around residues 34-40, the first two residues at the N-terminus and the last two at the C-terminus, and some side chains located on the molecular surface. The average root mean squared deviation in position for equivalent atoms between the 24 individual structures and the mean structure obtained by averaging their coordinates is 0.54 +/- 0.08 A for the main-chain atoms and 0.97 +/- 0.09 A for all non-hydrogen atoms of residues 3-80 plus the heme group. These structures were compared to the X-ray crystallographic structure of an analogous protein, cytochrome c-551 from Pseudomonas aeruginosa [Matsuura, Takano, & Dickerson (1982) J. Mol. Biol. 156, 389-409). The main-chain folding patterns are very consistent, but there are some differences. The largest difference is in a surface loop segment from residues 34 to 40.     

Cross-reactivity and 1.4-A crystal structure of Malassezia sympodialis thioredoxin (Mala s 13), a member of a new pan-allergen family

We have identified thioredoxins (Trx) of Malassezia sympodialis, a yeast involved in the pathogenesis of atopic eczema, and of Aspergillus fumigatus, a fungus involved in pulmonary complications, as novel IgE-binding proteins. We show that these Trx, including the human enzyme, represent cross-reactive structures recognized by serum IgE from individuals sensitized to M. sympodialis Trx. Moreover, all three proteins were able to elicit immediate-type allergic skin reactions in sensitized individuals, indicating a humoral immune response based on molecular mimicry. To analyze structural elements involved in these reactions, the three-dimensional structure of M. sympodialis Trx (Mala s 13) has been determined at 1.4-A resolution by x-ray diffraction analysis. The structure was solved by molecular replacement and refined to a crystallographic R factor of 14.0% and a free R factor of 16.8% and shows the typical Trx fold. Mala s 13 shares 45% sequence identity with human Trx and superposition of the solved Mala s 13 structure with those of human Trx reveals a high similarity with a root mean square deviation of 1.11 A for all Calpha atoms. In a detailed analysis of the molecular surface in combination with sequence alignment, we identified conserved solvent-exposed amino acids scattered over the surface in both structures which cluster to patches, thus forming putative conformational B cell epitopes potentially involved in IgE-mediated cross- and autoreactivity.     

Structure of the Clade 1 catalase,CatF of Pseudomonas syringae,at 1.8 A resolution

Catalase CatF of Pseudomonas syringae has been identified phylogenetically as a clade 1 catalase, closely related to plant catalases, a group from which no structure has been determined. The structure of CatF has been refined at 1.8 A resolution by using X-ray synchrotron data collected from a crystal flash-cooled with liquid nitrogen. The crystallographic agreement factors R and R(free) are, respectively, 18.3% and 24.0%. The asymmetric unit of the crystal contains a whole molecule that shows accurate 222-point group symmetry. The crystallized enzyme is a homotetramer of subunits with 484 residues, some 26 residues shorter than predicted from the DNA sequence. Mass spectrometry analysis confirmed the absence of 26 N-terminal residues, possibly removed by a periplasmic transport system. The core structure of the CatF subunit was closely related to seven other catalases with root-mean-square deviations (RMSDs) of 368 core Calpha atoms of 0.99-1.30 A. The heme component of CatF is heme b in the same orientation that is found in Escherichia coli hydroperoxidase II, an orientation that is flipped 180 degrees with respect the orientation of the heme in bovine liver catalase. NADPH is not found in the structure of CatF because key residues required for nucleotide binding are missing; 2129 water molecules were refined into the model. Water occupancy in the main or perpendicular channel of CatF varied among the four subunits from two to five in the region between the heme and the conserved Asp150. A comparison of the water occupancy in this region with the same region in other catalases reveals significant differences among the catalases.