Structural and biological aspects of carotenoid cleavage.

Apo-carotenoid compounds such as retinol (vitamin A) are involved in a variety of cellular processes and are found in all kingdoms of life. Instead of being synthesized from small precursors, they are commonly produced by oxidative cleavage and subsequent modification of larger carotenoid compounds. The cleavage reaction is catalyzed by a family of related enzymes, which convert specific substrate double bonds to the corresponding aldehydes or ketones. The individual family members differ in their substrate preference and the position of the cleaved double bond, giving rise to a remarkable number of products starting from a limited number of carotenoid substrate molecules. The recent determination of the structure of a member of this family has provided insight into the reaction mechanism, showing how substrate specificity is achieved. This review will focus on the biochemistry of carotenoid oxygenases and the structural determinants of the cleavage reaction.

Phenylalanine ammonia-lyase modified with polyethylene glycol: potential therapeutic agent for phenylketonuria.

Phenylketonuria (PKU) is an autosomal recessive genetic disease caused by the defects in the phenylalanine hydroxylase (PAH) gene. Individuals homozygous for defective PAH alleles show elevated levels of systemic phenylalanine and should be under strict dietary control to reduce the risk of neuronal damage associated with high levels of plasma phenylalanine. Researchers predict that plant phenylalanine ammonia-lyase (PAL), which converts phenylalanine to nontoxic t-cinnamic acid, will be an effective therapeutic enzyme for the treatment of PKU. The problems of this potential enzyme therapy have been the low stability in the circulation and the antigenicity of the plant enzyme. Recombinant PAL originated from parsley (Petroselinum crispum) chemically conjugated with activated PEG2 [2,4-bis(O-methoxypolyethyleneglycol)-6-chloro-s-triazine] showed greatly enhanced stability in the circulation and was effective in reducing the plasma concentration of phenylalanine in the circulation of mice. PEG-PAL conjugate will be an effective therapeutic enzyme for the treatment of PKU.

The structure of CMP:2-keto-3-deoxy-manno-octonic acid synthetase and of its complexes with substrates and substrate analogs.

The enzyme CMP-Kdo synthetase (CKS) catalyzes the activation of the sugar Kdo (2-keto-3-deoxy-manno-octonic acid) by forming a monophosphate diester. CKS is a pharmaceutical target because CMP-Kdo is used in the biosynthesis of lipopolysaccharides that are vital for Gram-negative bacteria. We have refined the structure of the unligated capsule-specific CKS from Escherichia coli at 1.8 A resolution (1 A=0.1 nm) and we have established the structures of its complexes with the substrate CTP, with CDP and CMP as well as with the product analog CMP-NeuAc (CMP-sialate) by X-ray diffraction analyses at resolutions between 2.1 A and 2.5 A. The N-terminal domains of the dimeric enzyme bind CTP in a peculiar nucleotide-binding fold, whereas the C-terminal domains form the dimer interface. The observed binding geometries together with the amino acid variabilities during evolution and the locations of a putative Mg(2+) and of a very strongly bound water molecule suggest a pathway for the catalysis. The N-terminal domain shows sequence homology with the CMP-NeuAc synthetases. Moreover, the chain fold and the substrate-binding position of CKS resemble those of other enzymes processing nucleotide-sugars.

The effect of substrate binding on the conformation and structural stability of Herpes simplex virus type 1 thymidine kinase.

The structure of Herpes simplex virus type 1 thymidine kinase (TK(HSV1)) is known at high resolution in complex with a series of ligands and exhibits important structural similarities to the nucleoside monophosphate (NMP) kinase family, which are known to show large conformational changes upon binding of substrates. The effect of substrate binding on the conformation and structural stability of TK(HSV1), measured by thermal denaturation experiments, far-UV circular dichroism (CD) and fluorescence is described, and the results indicate that the conformation of the ligand-free TK(HSV1) is less ordered and less stable compared to the ligated enzyme. Furthermore, two crystal structures of TK(HSV1) in complex with two new ligands, HPT and HMTT, refined to 2.2 A are presented. Although TK(HSV1):HPT does not exhibit any significant deviations from the model of TK(HSV1):dT, the TK(HSV1):HMTT complex displays a unique conformationally altered active site resulting in a lowered thermal stability of this complex. Moreover, we show that binding affinity and binding mode of the ligand correlate with thermal stability of the complex. We use this correlation to propose a method to estimate binding constants for new TK(HSV1)substrates using thermal denaturation measurements monitored by CD spectroscopy. The kinetic and structural results of both test substrates HPT and HMTT show that the CD thermal denaturation system is very sensitive to conformational changes caused by unusual binding of a substrate analog.

Nucleoside binding site of herpes simplex type 1 thymidine kinase analyzed by X-ray crystallography.

The crystal structures of the full-length Herpes simplex virus type 1 thymidine kinase in its unligated form and in a complex with an adenine analogue have been determined at 1.9 A resolution. The unligated enzyme contains four water molecules in the thymidine pocket and reveals a small induced fit on substrate binding. The structure of the ligated enzyme shows for the first time a bound adenine analogue after numerous complexes with thymine and guanine analogues have been reported. The adenine analogue constitutes a new lead compound for enzyme-prodrug gene therapy. In addition, the structure of mutant Q125N modifying the binding site of the natural substrate thymidine in complex with this substrate has been established at 2.5 A resolution. It reveals that neither the binding mode of thymidine nor the polypeptide backbone conformation is altered, except that the two major hydrogen bonds to thymidine are replaced by a single water-mediated hydrogen bond, which improves the relative acceptance of the prodrugs aciclovir and ganciclovir compared with the natural substrate. Accordingly, the mutant structure represents a first step toward improving the virus-directed enzyme-prodrug gene therapy by enzyme engineering.

The mitochondrial import receptor Tom70: identification of a 25 kDa core domain with a specific binding site for preproteins.

The mitochondrial import receptor of 70 kDa, Tom70, preferentially recognizes precursors of membrane proteins with internal targeting signals. We report the identification of a stably folded 25 kDa core domain located in the middle portion of Tom70 that contains two of the seven tetratricopeptide repeat motifs of the receptor. The core domain binds non-cleavable and cleavable preproteins carrying internal targeting signals with a specificity indistinguishable from the full-length receptor. Competition studies indicate that both types of preproteins interact with overlapping binding sites of the core domain and that at least one additional interaction site is present in the full-length receptor. We suggest a model of Tom70 function in import of membrane proteins whereby a hydrophobic preprotein concomitantly interacts with several binding sites of the receptor.

Structures of l-fuculose-1-phosphate aldolase mutants outlining motions during catalysis.

The crystal structures of l-fuculose-1-phosphate aldolase (FucA) with and without a ligated analogue of dihydroxyacetone phosphate (DHAP) and of a number of active center mutants have resulted in a model of the catalytic mechanism. This model has now been confirmed by structural analyses of further mutations at the zinc coordination sphere and at the phosphate site. In addition, these mutants have revealed new aspects of the catalysis: the hydroxyl group of Tyr113' (from a neighboring subunit), which sits just outside the zinc coordination sphere, steers DHAP towards a productive binding mode at the zinc ion; Glu73 contacts zinc in between the two ligand positions intended for the DHAP oxygen atoms and thus avoids blocking of these positions by a tetrahedrally coordinated hydroxy ion; the FucA polypeptide does not assume its minimum energy state but oscillates between two states of elevated energy as demonstrated by a mutant in a minimum energy state. The back and forth motion involves a mobile loop connecting the phosphate site with intersubunit motions and thus with the Brownian motion of the solvent. The phosphate group is bound strongly at a given distance to the zinc ion, which prevents the formation of too tight a DHAP:zinc complex. This observation explains our failure to find mutants that accept phosphate-free substitutes for DHAP. The FucA zinc coordination sphere is compared with that of carbonic anhydrase.

Crystal structures of adrenodoxin reductase in complex with NADP+ and NADPH suggesting a mechanism for the electron transfer of an enzyme family.

Adrenodoxin reductase is a flavoenzyme that shuffles electrons for the biosynthesis of steroids. Its chain topology belongs to the glutathione reductase family of disulfide oxidoreductases, all of which bind FAD at equivalent positions. The three reported structures of adrenodoxin reductase were ligated with reduced and oxidized NADP and have now confirmed this equivalence also for the NADP-binding site. Remarkably, the conformations and relative positions of the prosthetic group FAD and the cofactor NADP have been conserved during protein evolution despite very substantial changes in the polypeptide. The ligated enzymes showed small changes in the domain positions. When compared with the structure of the NADP-free enzyme, these positions correspond to several states of the domain motion during NADP binding. On the basis of the observed structures, we suggest an enzymatic mechanism for the subdivision of the received two-electron package into the two single electrons transferred to the carrier protein adrenodoxin. The data banks contain 10 sequences that are closely related to bovine adrenodoxin reductase. Most of them code for gene products with unknown functions. Within this family, the crucial residues of adrenodoxin reductase are strictly conserved. Moreover, the putative docking site of the carrier is rather well conserved. Five of the family members were assigned names related to ferredoxin:NADP(+) reductase, presumably because adrenodoxin reductase was considered a member of this functionally similar family. Since this is not the case, the data bank entries should be corrected.

beta-Barrel membrane proteins.

beta-Barrel proteins are found in the outer membranes of bacteria, mitochondria and chloroplasts. The presently known sizes range from small eight-stranded to large twenty-two-stranded beta barrels existing as monomers and oligomers. Their functions are as diverse as active ion transport, passive nutrient intake, membrane anchors, membrane-bound enzymes and defense against attack proteins. Of further interest are the folding process, the crystallization, the observed limited structural diversity and the manifold channel engineering options of these beta-barrel proteins.

Kinetics and crystal structure of the wild-type and the engineered Y101F mutant of Herpes simplex virus type 1 thymidine kinase interacting with (North)-methanocarba-thymidine.

Kinetic and crystallographic analyses of wild-type Herpes simplex virus type 1 thymidine kinase (TK(HSV1)) and its Y101F-mutant [TK(HSV1)(Y101F)] acting on the potent antiviral drug 2'-exo-methanocarba-thymidine (MCT) have been performed. The kinetic study reveals a 12-fold K(M) increase for thymidine processed with Y101F as compared to the wild-type TK(HSV1). Furthermore, MCT is a substrate for both wild-type and mutant TK(HSV1). Its binding affinity for TK(HSV1) and TK(HSV1)(Y101F), expressed as K(i), is 11 microM and 51 microM, respectively, whereas the K(i) for human cytosolic thymidine kinase is as high as 1.6 mM, rendering TK(HSV1) a selectivity filter for antiviral activity. Moreover, TK(HSV1)(Y101F) shows a decrease in the quotient of the catalytic efficiency (k(cat)/K(M)) of dT over MCT corresponding to an increased specificity for MCT when compared to the wild-type enzyme. Crystal structures of wild-type and mutant TK(HSV1) in complex with MCT have been determined to resolutions of 1.7 and 2.4 A, respectively. The thymine moiety of MCT binds like the base of dT while the conformationally restricted bicyclo[3.1.0]hexane, mimicking the sugar moiety, assumes a 2'-exo envelope conformation that is flatter than the one observed for the free compound. The hydrogen bond pattern around the sugar-like moiety differs from that of thymidine, revealing the importance of the rigid conformation of MCT with respect to hydrogen bonds. These findings make MCT a lead compound in the design of resistance-repellent drugs for antiviral therapy, and mutant Y101F, in combination with MCT, opens new possibilities for gene therapy.

Comparative structural analysis and substrate specificity engineering of the hyperthermostable beta-glucosidase CelB from Pyrococcus furiosus.

The substrate specificity of the beta-glucosidase (CelB) from the hyperthermophilic archaeon Pyrococcus furiosus, a family 1 glycosyl hydrolase, has been studied at a molecular level. Following crystallization and X-ray diffraction of this enzyme, a 3.3 A resolution structural model has been obtained by molecular replacement. CelB shows a homo-tetramer configuration, with subunits having a typical (betaalpha)(8)-barrel fold. Its active site has been compared to the one of the previously determined 6-phospho-beta-glycosidase (LacG) from the mesophilic bacterium Lactococcus lactis. The overall design of the substrate binding pocket is very well conserved, with the exception of three residues that have been identified as a phosphate binding site in LacG. To verify the structural model and alter its substrate specificity, these three residues have been introduced at the corresponding positions in CelB (E417S, M424K, F426Y) in different combinations: single, double, and triple mutants. Characterization of the purified mutant CelB enzyme revealed that F426Y resulted in an increased affinity for galactosides, whereas M424K gave rise to a shifted pH optimum (from 5.0 to 6.0). Analysis of E417S revealed a 5-fold and a 3-fold increase of the efficiency of hydrolyzing o-nitrophenol-beta-D-galactopyranoside-6-phosphate, in the single and triple mutants, respectively. In contrast, their activity on nonphosphorylated sugars was largely reduced (30-300-fold). The residue at position E417 in CelB seems to be the determining factor for the difference in substrate specificity between the two types of family 1 glycosidases.

Catalytic action of fuculose 1-phosphate aldolase (class II) as derived from structure-directed mutagenesis.

Previous analyses established the structures of unligated L-fuculose 1-phosphate aldolase and of the enzyme ligated with an inhibitor mimicking the substrate dihydroxyacetone phosphate. These data allowed us to suggest a catalytic mechanism. On the basis of this proposal, numerous mutations were now introduced at the active center and tested with respect to their catalytic rates and their product distributions. For several mutants, the structures were determined. The results demonstrate the catalytic importance of some particular residues in defined conformations and in the mobile C-terminal chain end. Moreover, they led to a modification of the proposed mechanism. The effect of some mutations on enantioselectivity and on the ratio of diastereomer formation indicates clearly the binding site of the aldehyde moiety in relation to the other substrate dihydroxyacetone phosphate.

High-resolution structure of the OmpA membrane domain.

The membrane domain of OmpA consists of an eight-stranded all-next-neighbor antiparallel beta-barrel with short turns at the periplasmic barrel end and long flexible loops at the external end. The structure analysis has been extended from medium resolution to 1. 65 A (1 A=0.1 nm), and the molecular model has been refined anisotropically to show oriented mobilities of the structural elements. The improved data allowed us to locate five further detergent molecules and 11 more water molecules. Moreover, the two large non-polar packing contacts have now been defined in detail. The analysis indicates that the beta-barrel constitutes a solid scaffold such that the long external loops need not contribute to stability. These loops are highly mobile and thus cause a major problem during the crystallization process. The beta-barrel was related to those of lipocalins. Two further crystal forms with exceptionally dense packing arrangements were established at medium resolution.

The structure of the outer membrane protein OmpX from Escherichia coli reveals possible mechanisms of virulence.

BACKGROUND: The integral outer membrane protein X (OmpX) from Escherichia coli belongs to a family of highly conserved bacterial proteins that promote bacterial adhesion to and entry into mammalian cells. Moreover, these proteins have a role in the resistance against attack by the human complement system. Here we present the first crystal structure of a member of this family. RESULTS: The crystal structure of OmpX from E. coli was determined at 1.9 A resolution using multiple isomorphous replacement. OmpX consists of an eight-stranded antiparallel all-next-neighbor beta barrel. The structure shows two girdles of aromatic amino acid residues and a ribbon of nonpolar residues that attach to the membrane interior. The core of the barrel consists of an extended hydrogen-bonding network of highly conserved residues. OmpX thus resembles an inverse micelle. The structure explains the dramatically improved crystal quality of OmpX containing the mutation His100-->Asn, which made the X-ray analysis possible. The coordination spheres of two bound platinum ions are described. CONCLUSIONS: The OmpX structure shows that within a family of virulence-related membrane proteins, the membrane-spanning part of the protein is much better conserved than the extracellular loops. Moreover, these loops form a protruding beta sheet, the edge of which presumably binds to external proteins. It is suggested that this type of binding promotes cell adhesion and invasion and helps defend against the complement system. Although OmpX has the same beta-sheet topology as the structurally related outer membrane protein A (OmpA), their barrels differ with respect to the shear numbers and internal hydrogen-bonding networks.

Overexpression of a designed 2.2 kb gene of eukaryotic phenylalanine ammonia-lyase in Escherichia coli.

Phenylalanine ammonia-lyase (EC 4.3.1.5) is a key enzyme in the secondary metabolism of higher plants catalyzing the non-oxidative conversion of L-phenylalanine into transcinnamate. The nucleotide sequence of its 2.2 kb gene was designed for expression in Escherichia coli and synthesized in a single reaction from 108 oligonucleotides using assembly PCR. After amplification, the gene was cloned into the expression vector pT7-7 and coexpressed with the chaperone HSP-60 system. The expression system yielded 70 mg of fully active enzyme per liter culture.

The structure of adrenodoxin reductase of mitochondrial P450 systems: electron transfer for steroid biosynthesis.

Adrenodoxin reductase is a monomeric 51 kDa flavoenzyme that is involved in the biosynthesis of all steroid hormones. The structure of the native bovine enzyme was determined at 2.8 A resolution, and the structure of the respective recombinant enzyme at 1.7 A resolution. Adrenodoxin reductase receives a two-electron package from NADPH and converts it to two single electrons that are transferred via adrenodoxin to all mitochondrial cytochromes P 450. The structure suggests how the observed flavin semiquinone is stabilized. A striking feature is the asymmetric charge distribution, which most likely controls the approach of the electron carrier adrenodoxin. A model for the interaction is proposed. Adrenodoxin reductase shows clear sequence homology to half a dozen proteins identified in genome analysis projects, but neither sequence nor structural homology to established, functionally related electron transferases. Yet, the structure revealed a relationship to the disulfide oxidoreductases, permitting the assignment of the NADP-binding site.

Crystal structure of histidine ammonia-lyase revealing a novel polypeptide modification as the catalytic electrophile.

Histidine ammonia-lyase (EC 4.3.1.3) catalyzes the nonoxidative elimination of the alpha-amino group of histidine and is closely related to the important plant enzyme phenylalanine ammonia-lyase. The crystal structure of histidase from Pseudomonas putida was determined at 2.1 A resolution revealing a homotetramer with D2 symmetry, the molecular center of which is formed by 20 nearly parallel alpha-helices. The chain fold, but not the sequence, resembles those of fumarase C and related proteins. The structure shows that the reactive electrophile is a 4-methylidene-imidazole-5-one, which is formed autocatalytically by cyclization and dehydration of residues 142-144 with the sequence Ala-Ser-Gly. With respect to the first dehydration step, this modification resembles the chromophore of the green fluorescent protein. The active center is clearly established by the modification and by mutations. The observed geometry allowed us to model the bound substrate at a high confidence level. A reaction mechanism is proposed.

Homogenization and crystallization of histidine ammonia-lyase by exchange of a surface cysteine residue.

Histidase (histidine ammonia-lyase, EC 4.3.1.3) from Pseudomonas putida was expressed in Escherichia coli and purified. In the absence of thiols the tetrameric enzyme gave rise to undefined aggregates and suitable crystals could not be obtained. The solvent accessibility along the chain was predicted from the amino acid sequence. Among the seven cysteines, only one was labeled as 'solvent-exposed'. The exchange of this cysteine to alanine abolished all undefined aggregations and yielded readily crystals diffracting to 1.8 A resolution.

Structure and mechanism of the amphibolic enzyme D-ribulose-5-phosphate 3-epimerase from potato chloroplasts.

Ribulose-5-phosphate 3-epimerase (EC 5.1.3.1) catalyzes the interconversion of ribulose-5-phosphate and xylulose-5-phosphate in the Calvin cycle and in the oxidative pentose phosphate pathway. The enzyme from potato chloroplasts was expressed in Escherichia coli, isolated and crystallized. The crystal structure was elucidated by multiple isomorphous replacement and refined at 2.3 A resolution. The enzyme is a homohexamer with D3 symmetry. The subunit chain fold is a (beta alpha)8-barrel. A sequence comparison with homologous epimerases outlined the active center and indicated that all members of this family are likely to share the same catalytic mechanism. The substrate could be modeled by putting its phosphate onto the observed sulfate position and its epimerized C3 atom between two carboxylates that participate in an extensive hydrogen bonding system. A mutation confirmed the crucial role of one of these carboxylates. The geometry together with the conservation pattern suggests that the negative charge of the putative cis-ene-diolate intermediate is stabilized by the transient induced dipoles of a methionine sulfur "cushion", which is proton-free and therefore prevents isomerization instead of epimerization.

Locating proper non-crystallographic symmetry in low-resolution electron-density maps with the program GETAX.

Non-crystallographic symmetry averaging for improving and extending an initial set of phases can be crucial at an early stage of a protein structure analysis. A method is described which detects the position of a proper rotation axis in a surprisingly poor electron-density map and is fast enough to run through a large number of axis orientations. It uses a simple multimer mask to define the searching unit, which is then shifted through the whole unit cell looking for the position with the highest correlation coefficient between the interrelated parts. Appropriate weighting and averaging enhances the signal-to-noise ratio. Examples of the application of this algorithm are given. The use of the local rotation axis for phasing is commented on. A search of the Protein Data Bank showed that 27% of the unique crystal forms contain proper local n-fold axes, which could have been located with the presented method.