Tetrameric dipeptidyl peptidase I directs substrate specificity by use of the residual pro-part domain.

The crystal structure of mature dipeptidyl peptidase I reveals insight into the unique tetrameric structure, substrate binding and activation of this atypical papain family peptidase. Each subunit is composed of three peptides. The heavy and light chains form the catalytic domain, which adopts the papain fold. The residual pro-part forms a beta-barrel with the carboxylate group of Asp1 pointing towards the substrate amino-terminus. The tetrameric structure appears to stabilize the association of the two domains and encloses a 12700 A3 spherical cavity. The tetramer contains six chloride ions, one buried in each S2 pocket and two at subunit interfaces.

Structures of beta-ketoacyl-acyl carrier protein synthase I complexed with fatty acids elucidate its catalytic machinery.

BACKGROUND: beta-ketoacyl-acyl carrier protein synthase (KAS) I is vital for the construction of the unsaturated fatty acid carbon skeletons characterizing E. coli membrane lipids. The new carbon-carbon bonds are created by KAS I in a Claisen condensation performed in a three-step enzymatic reaction. KAS I belongs to the thiolase fold enzymes, of which structures are known for five other enzymes. RESULTS: Structures of the catalytic Cys-Ser KAS I mutant with covalently bound C10 and C12 acyl substrates have been determined to 2.40 and 1.85 A resolution, respectively. The KAS I dimer is not changed by the formation of the complexes but reveals an asymmetric binding of the two substrates bound to the dimer. A detailed model is proposed for the catalysis of KAS I. Of the two histidines required for decarboxylation, one donates a hydrogen bond to the malonyl thioester oxo group, and the other abstracts a proton from the leaving group. CONCLUSIONS: The same mechanism is proposed for KAS II, which also has a Cys-His-His active site triad. Comparison to the active site architectures of other thiolase fold enzymes carrying out a decarboxylation step suggests that chalcone synthase and KAS III with Cys-His-Asn triads use another mechanism in which both the histidine and the asparagine interact with the thioester oxo group. The acyl binding pockets of KAS I and KAS II are so similar that they alone cannot provide the basis for their differences in substrate specificity.

Structural basis for the function of Bacillus subtilis phosphoribosyl-pyrophosphate synthetase.

Here we report the first three-dimensional structure of a phosphoribosylpyrophosphate (PRPP) synthetase. PRPP is an essential intermediate in several biosynthetic pathways. Structures of the Bacillus subtilis PRPP synthetase in complex with analogs of the activator phosphate and the allosteric inhibitor ADP show that the functional form of the enzyme is a hexamer. The individual subunits fold into two domains, both of which resemble the type I phosphoribosyltransfereases. The active site is located between the two domains and includes residues from two subunits. Phosphate and ADP bind to the same regulatory site consisting of residues from three subunits of the hexamer. In addition to identifying residues important for binding substrates and effectors, the structures suggest a novel mode of allosteric regulation.

The X-ray crystal structure of beta-ketoacyl [acyl carrier protein] synthase I.

The crystal structure of the fatty acid elongating enzyme beta-ketoacyl [acyl carrier protein] synthase I (KAS I) from Escherichia coli has been determined to 2.3 A resolution by molecular replacement using the recently solved crystal structure of KAS II as a search model. The crystal contains two independent dimers in the asymmetric unit. KAS I assumes the thiolase alpha(beta)alpha(beta)alpha fold. Electrostatic potential distribution reveals an acyl carrier protein docking site and a presumed substrate binding pocket was detected extending the active site. Both subunits contribute to each substrate binding site in the dimer.

Native carboxypeptidase A in a new crystal environment reveals a different conformation of the important tyrosine 248.

Native carboxypeptidase A has been crystallized in a new crystal form, and the structure has been refined with X-ray data to 2.0 A resolution. In contrast to the previously published structure [Rees, D. C., Lewis, M., and Lipscomb, W. N. (1983) J. Mol. Biol. 168, 367-387], no active-site amino acids are involved in the crystal packing. The important Tyr248 is stabilized inside the active site by a hydrogen bond and by interactions with Ile247. The proposed role of Tyr248 in the induced fit mechanism is therefore not supported by the findings in this structure of native carboxypeptidase A. The structure has a partly populated inhibitory Zn2+ site in close proximity to the catalytic Zn2+ as evident from X-ray anomalous dispersion data. A hydroxo bridge is found between the catalytic Zn2+ and the inhibitory Zn2+ with a Zn2+-Zn2+ distance of 3.48 A. In addition, the inhibitory Zn2+ has Glu270 as a monodentate ligand. No other protein ligands to the inhibitory Zn2+ are seen. The crystals were grown at 0.3 M LiCl and weak evidence for a binding site for partly competitive inhibitory anions is observed.

Barley alpha-amylase bound to its endogenous protein inhibitor BASI: crystal structure of the complex at 1.9 A resolution.

BACKGROUND: Barley alpha-amylase is a 45 kDa enzyme which is involved in starch degradation during barley seed germination. The released sugars provide the plant embryo with energy for growth. The major barley alpha-amylase isozyme (AMY2) binds with high affinity to the endogenous inhibitor BASI (barley alpha-amylase/subtilisin inhibitor) whereas the minor isozyme (AMY1) is not inhibited. BASI is a 19.6 kDa bifunctional protein that can simultaneously inhibit AMY2 and serine proteases of the subtilisin family. This inhibitor may therefore prevent degradation of the endosperm starch during premature sprouting and protect the seed from attack by pathogens secreting proteases. RESULTS: The crystal structure of AMY2 in complex with BASI was determined and refined at 1.9 A resolution. BASI consists of a 12-stranded beta-barrel structure which belongs to the beta-trefoil fold family and inhibits AMY2 by sterically occluding access of the substrate to the active site of the enzyme. The AMY2-BASI complex is characterized by an unusual completely solvated calcium ion located at the protein-protein interface. CONCLUSIONS: The AMY2-BASI complex represents the first reported structure of an endogenous protein-protein complex from a higher plant. The structure of the complex throws light on the strict specificity of BASI for AMY2, and shows that domain B of AMY2 contributes greatly to the specificity of enzyme-inhibitor recognition. In contrast to the three-dimensional structures of porcine pancreatic alpha-amylase in complex with proteinaceous inhibitors, the AMY2-BASI structure reveals that the catalytically essential amino acid residues of the enzyme are not directly bound to the inhibitor. Binding of BASI to AMY2 creates a cavity, exposed to the external medium, that is ideally shaped to accommodate an extra calcium ion. This feature may contribute to the inhibitory effect, as the key amino acid sidechains of the active site are in direct contact with water molecules which are in turn ligated to the calcium ion.

Molecular structure of a barley alpha-amylase-inhibitor complex: implications for starch binding and catalysis.

alpha-Amylases are widely occurring, multidomain proteins with a catalytic (beta/alpha)8-barrel. In barley alpha-amylase, insight into the catalytic mechanism is gained from the X-ray crystal structure of its molecular complex with acarbose, a pseudotetrasaccharide that acts like a transition-state analogue and which is shown to bind at two specific regions of the enzyme. The structure of the complex has been refined to an R-factor of 15.1% for all observations with Fo>sigma(Fo) between 10 and 2.8 A resolution. A difference Fourier map produced after refinement of the native structure against the data of the acarbose complex clearly revealed density corresponding to two oligosaccharide-binding sites. One of these is defined as the surface-located starch granule-binding site characteristic of cereal alpha-amylases. It involves stacking of two acarbose rings on Trp276 and Trp277. The other binding region is the active site covering subsites -1, +1 and +2. Here, Glu204 is positioned to act in general acid/base catalysis protonating the glucosidic oxygen atom assisted by Asp289. A water molecule that bridges Glu204 and Asp289 is found at the entrance cavity containing a total of five water molecules. This water molecule is proposed to reprotonate Glu204 and supply the hydroxyl ion for nucleophilic attack on the glucosyl C1 atom. Asp 179 acts as the nucleophile that can bind covalently to the substrate intermediate after bond cleavage. The present complex structure together with the conservation of active-site residues among alpha-amylases and related enzymes, are consistent with a common catalytic mechanism for this class of retaining carbohydrases.

Crystal structure of the dihaem cytochrome c4 from Pseudomonas stutzeri determined at 2.2A resolution.

BACKGROUND: . Cytochromes c4 are dihaem cytochromes c found in a variety of bacteria. They are assumed to take part in the electron-transport systems associated with both aerobic and anaerobic respiration. The cytochrome c4 proteins are located in the periplasm, predominantly bound to the inner membrane, and are able to transfer electrons between membrane-bound reduction systems and terminal oxidases. Alignment of cytochrome c4 sequences from three bacteria, Pseudomonas aeruginosa, Pseudomonas stutzeri and Azotobacter vinelandii, suggests that these dihaem proteins are composed of two similar domains. Two distinctly different redox potentials have been measured for the Ps. stutzeri cytochrome c4, however. RESULTS: . The crystal structure of the dihaem cytochrome c4 from Ps. stutzeri has been determined to 2.2A resolution by isomorphous replacement. The model, consisting of two entire cytochrome c4 molecules and 138 water molecules in the asymmetric unit, was refined to an R value of 20.1% for all observations in the resolution range 8-2.2A. The molecule is organized in two cytochrome c-like domains that are related by a pseudo-twofold axis. The symmetry is virtually perfectly close to the twofold axis, which passes through a short hydrogen bond between the two haem propionic acid groups, connecting the redox centre of each domain. This haem-haem interaction is further stabilized by an extensive symmetrical hydrogen-bond network. The twofold symmetry is not present further away from the axis, however, and the cytochrome c4 molecule can be considered to be a dipole with charged residues unevenly distributed between the two domains. The haem environment in the two domains show pronounced differences, mainly on the methionine side of the haem group. CONCLUSIONS: . The structure, in conjunction with sequence alignment, suggests that the cytochrome protein has evolved by duplication of a cytochrome c gene. The difference in charge distribution around each haem group in the two domains allows the haem group in the N-terminal domain to be associated with the lower redox potential of 241 mV and the C-terminal haem group with the higher potential of 328 mV. The molecular dipole characteristic of cytochrome c4 is important for its interaction with, and recognition of, its redox partners. In cytochrome c4, the hydrogen-bond network (between residues that are conserved in all known cytochrome c4 subspecies) seems to provide an efficient pathway for an intramolecular electron transfer that can ensure cooperativity between the two redox centres. The C-pyrrole corners of the haem edges are potential sites for external electron exchange.

Overexpression of Bacillus subtilis phosphoribosylpyrophosphate synthetase and crystallization and preliminary X-ray characterization of the free enzyme and its substrate-effector complexes.

Bacillus subtilis phosphoribosylpyrophosphate (PRPP) synthetase has been expressed to high levels in an Escherichia coli host strain devoid of endogenous PRPP synthetase. A rapid and efficient purification protocol has been developed allowing production of enzyme preparations with purity conforming to the stringent criteria required for crystallization. Crystallization experiments, in combination with dynamic light scattering studies, have led to the production of three crystal forms of the enzyme. These forms include the free enzyme, the enzyme in a binary complex with the substrate adenosine triphosphate (ATP), and the enzyme in a quaternary complex with the substrate analog alpha, beta-methylene adenosine triphosphate (mATP), the substrate ribose-5-phosphate (Rib-5-P), and the allosteric inhibitor adenosine diphosphate (ADP). Diffraction data showed that all three crystal forms are suitable for structure determination. They crystallize in the same hexagonal space group, P6(3), with virtually identical unit cell dimensions of a = b = 115.6 angstroms, c = 107.8 angstrom, and with two molecules in the asymmetric unit. The self-rotation function showed the existence of a non-crystallographic twofold axis perpendicular to the c axis. The availability of the different complexes should allow questions regarding the molecular mechanisms of catalysis and allostery in PRPP synthetase to be addressed.

Crystallization and preliminary crystallographic investigations of cytochrome c4 from Pseudomonas stutzeri.

Cytochrome c(4) from the bacterium Pseudomonas stutzeri has been crystallized and X-ray diffraction data have been collected to 2.2 A resolution. The crystals belong to the monoclinic system, space group P2(1) with cell parameters a = 49.49, b = 58.58, c = 63.51 A and beta = 96.96 degrees. The crystals contain two molecules it, the asymmetric unit. Cytochrome c(4) is known to contain two covalently bound haem groups per molecule and positions of the four haem Fe atoms in the asymmetric unit were determined from native anomalous-dispersion differences. The shortest Fe-Fe distances found in the crystal were 15.8, 16.9 and 19.0 A.

Crystal and molecular structure of barley alpha-amylase.

The three-dimensional structure of barley malt alpha-amylase (isoform AMY2-2) was determined by multiple isomorphous replacement using three heavy-atom derivatives and solvent flattening. The model was refined using a combination of simulated annealing and conventional restrained least-squares crystallographic refinement to an R-factor of 0.153 based on 18,303 independent reflections with F(o) > sigma(F(o)) between 10 and 2.8 A resolution, with root-mean-square deviations of 0.016 A and 3.3 degrees from ideal bond lengths and bond angles, respectively. The final model consists of 403 amino acid residues, three calcium ions and 153 water molecules. The polypeptide chain folds into three domains: a central domain forming a (beta alpha)8-barrel of 286 residues, with a protruding irregular structured loop domain of 64 residues (domain B) connecting strand beta 3 and helix alpha 3 of the barrel, and a C-terminal domain of 53 residues forming a five stranded anti-parallel beta-sheet. Unlike the previously known alpha-amylase structures, AMY2-2 contains three Ca2+ binding sites co-ordinated by seven or eight oxygen atoms from carboxylate groups, main-chain carbonyl atoms and water molecules, all calcium ions being bound to domain B and therefore essential for the structural integrity of that domain. Two of the Ca2+ sites are located only 7.0 A apart with one Asp residue serving as ligand for both. One Ca2+ site located at about 20 A from the other two was found to be exchangeable with Eu3+. By homology with other alpha-amylases, some important active site residues are identified as Asp179, Glu204 and Asp289, and are situated at the C-terminal end of the central beta-barrel. A starch granule binding site, previously identified as Trp276 and Trp277, is situated on alpha-helix 6 in the central (beta alpha)8-barrel, at the surface of the enzyme. This binding site region is associated with a considerable disruption of the (beta alpha)8-barrel 8-fold symmetry.

Characterization, crystallization and preliminary X-ray crystallographic analysis of the complex between barley alpha-amylase and the bifunctional alpha-amylase/subtilisin inhibitor from barley seeds.

The complex between a member of the barley malt alpha-amylase isozyme 2 family (AMY2-2), and the endogenous bifunctional alpha-amylase/subtilisin inhibitor, BASI, has been crystallized by the hanging drop vapour diffusion technique at a AMY2-2: BASI molar ratio of 1:1. Crystals have been grown within 4 days from solutions containing polyethylene glycol and calcium chloride. Analysis of single crystals by gel electrophoresis showed the presence of both proteins in the crystal lattice. The crystals belong to the orthorhombic space group P2(1)2(1)2(1), with unit cell dimensions a = 74.5 A, b = 96.9 A, c = 171.3 A and they diffract to 2.0 A resolution. The presence of two molecules of the 1:1 complex in the asymmetric unit gives a solvent content of 45% by volume. The 1:1 stoichiometry of the complex was confirmed by the molecular replacement method, using as a search model the recently determined three-dimensional structure of the barley alpha-amylase.

Three-dimensional structure of a recombinant peroxidase from Coprinus cinereus at 2.6 A resolution.

The structure of a recombinant peroxidase from the ink cap, Coprinus cinereus, CiP, is reported to 2.6 A resolution and refined to a R-value of 18.1%. The structure was solved by molecular replacement using the coordinates from a newly published ligninase structure. LiP. CiP crystallizes in space group P2(1)2(1)2(1) with two independent molecules in the asymmetric unit related by the vector 0.29b + 0.5c. The two CiP molecules are structurally identical; each contains two Ca2+ ions in positions equivalent to those found in the LiP structure. Two N-acetylglucosamines and one mannose residue were fitted into the density adjacent to two of the three predicted glycosylation sites. The refinement also included 40 and 41 water molecules, respectively, in the two CiP molecules. The structure of CiP displays a folding very similar to that of LiP. The active sites are almost identical in the LiP and CiP structures. CiP has a much larger opening to the active site than LiP.

Site-directed mutagenesis of histidine 93, aspartic acid 180, glutamic acid 205, histidine 290, and aspartic acid 291 at the active site and tryptophan 279 at the raw starch binding site in barley alpha-amylase 1.

The pseudotetrasaccharide acarbose has high affinity for the active site (Ki,app = 1 microM) and low affinity for a secondary site (Kd = 2.3 mM) in barley alpha-amylase 1, distinguished by inhibition kinetics and spectral perturbation. Mutants of putative catalytic residues, D180N, E205Q, and D291N, are inactive and display low affinity for acarbose-Sepharose. H93N and H290N mutants, at invariant residues, have kcat/Km for p-nitrophenylmaltoheptaoside of 0.3 and 1.2% of wild-type. A corresponding 370- and 85-fold increased Ki,app for acarbose and a lack of shifts in pH activity profiles indicate that these histidines participate in transition state stabilization but not directly in catalysis. This finding agrees with H bonding to OH groups of the valienamine ring of acarbose in the three-dimensional structure. Loss of inhibition above pH 6 supports that acarbose is most potent in protonated form. The low affinity site contains Trp278 and Trp279, known to bind cyclomaltoheptaose. While the W279A mutant has 10-fold decreased affinity for starch granules, production of Trp278 mutants failed. The invariant Trp278 is perhaps critical for stability or folding in cereal alpha-amylases.