Predicting the cofactors of oxidoreductases based on amino acid composition distribution and Chou's amphiphilic pseudo-amino acid composition

Predicting the cofactors of oxidoreductases plays an important role in inferring their catalytic mechanism. Feature extraction is a critical part in the prediction systems, requiring raw sequence data to be transformed into appropriate numerical feature vectors while minimizing information loss. In this paper, we present an amino acid composition distribution method for extracting useful features from primary sequence, and the k-nearest neighbor was used as the classifier. The overall prediction accuracy evaluated by the 10-fold cross-validation reached 90.74%. Comparing our method with other eight feature extraction methods, the improvement of the overall prediction accuracy ranged from 3.49% to 15.74%. Our experimental results confirm that the method we proposed is very useful and may be used for other bioinformatical predictions. Interestingly, when features extracted by our method and Chou's amphiphilic pseudo-amino acid composition were combined, the overall accuracy could reach 92.53%.     

Genome-wide profile of oxidoreductases in viruses, prokaryotes, and eukaryotes

Enzymes that utilize nicotinamide adenine dinucleotide (NAD) or its 2'-phosphate derivative (NADP) are found throughout the kingdoms of life. These enzymes are fundamental to many biochemical pathways, including central intermediary metabolism and mechanisms for cell survival and defense. The complete genomes of 25 organisms representing bacteria, protists, fungi, plants, and animals, and 811 viruses, were mined to identify and classify NAD(P)-dependent enzymes. An average of 3.4% of the proteins in these genomes was categorized as NAD(P)-utilizing proteins, with highest prevalence in the medium-chain oxidoreductase and short-chain oxidoreductase families. In general, the distribution of these enzymes by oxidoreductase family was correlated to the number of different catalytic mechanisms in each family. Organisms with smaller genomes encoded a larger proportion of NAD(P)-dependent enzymes in their proteome (approximately 6%) as compared to the larger genomes of eukaryotes (approximately 3%). Among viruses, those with large, double-strand DNA genomes were shown to encode oxidoreductases. Gram-positive and gram-negative bacteria showed some differences in the distribution of NAD(P)-dependent proteins. Several organisms such as M. tuberculosis, P. falciparum, and A. thaliana showed unique distributions of oxidoreductases corresponding to some phenotypic features.     

Enzyme family classification by support vector machines

One approach for facilitating protein function prediction is to classify proteins into functional families. Recent studies on the classification of G-protein coupled receptors and other proteins suggest that a statistical learning method, Support vector machines (SVM), may be potentially useful for protein classification into functional families. In this work, SVM is applied and tested on the classification of enzymes into functional families defined by the Enzyme Nomenclature Committee of IUBMB. SVM classification system for each family is trained from representative enzymes of that family and seed proteins of Pfam curated protein families. The classification accuracy for enzymes from 46 families and for non-enzymes is in the range of 50.0% to 95.7% and 79.0% to 100% respectively. The corresponding Matthews correlation coefficient is in the range of 54.1% to 96.1%. Moreover, 80.3% of the 8,291 correctly classified enzymes are uniquely classified into a specific enzyme family by using a scoring function, indicating that SVM may have certain level of unique prediction capability. Testing results also suggest that SVM in some cases is capable of classification of distantly related enzymes and homologous enzymes of different functions. Effort is being made to use a more comprehensive set of enzymes as training sets and to incorporate multi-class SVM classification systems to further enhance the unique prediction accuracy. Our results suggest the potential of SVM for enzyme family classification and for facilitating protein function prediction. Our software is accessible at http://jing.cz3.nus.edu.sg/cgi-bin/svmprot.cgi.     

A new role for coenzyme F420 in aflatoxin reduction by soil mycobacteria

Hepatotoxic aflatoxins have found a worthy adversary in two new families of bacterial oxidoreductases. These enzymes use the reduced coenzyme F420 to initiate the degradation of furanocoumarin compounds, including the major mycotoxin products of Aspergillus flavus. Along with pyridoxal 5'-phosphate synthases and aryl nitroreductases, these proteins form a large and versatile superfamily of flavin and deazaflavin-dependent oxidoreductases. F420-dependent members of this family appear to share a common mechanism of hydride transfer from the reduced, low-potential deazaflavin to the electron-deficient ring systems of their substrates.     

Protein classification based on text document classification techniques

The need for accurate, automated protein classification methods continues to increase as advances in biotechnology uncover new proteins. G-protein coupled receptors (GPCRs) are a particularly difficult superfamily of proteins to classify due to extreme diversity among its members. Previous comparisons of BLAST, k-nearest neighbor (k-NN), hidden markov model (HMM) and support vector machine (SVM) using alignment-based features have suggested that classifiers at the complexity of SVM are needed to attain high accuracy. Here, analogous to document classification, we applied Decision Tree and Naive Bayes classifiers with chi-square feature selection on counts of n-grams (i.e. short peptide sequences of length n) to this classification task. Using the GPCR dataset and evaluation protocol from the previous study, the Naive Bayes classifier attained an accuracy of 93.0 and 92.4% in level I and level II subfamily classification respectively, while SVM has a reported accuracy of 88.4 and 86.3%. This is a 39.7 and 44.5% reduction in residual error for level I and level II subfamily classification, respectively. The Decision Tree, while inferior to SVM, outperforms HMM in both level I and level II subfamily classification. For those GPCR families whose profiles are stored in the Protein FAMilies database of alignments and HMMs (PFAM), our method performs comparably to a search against those profiles. Finally, our method can be generalized to other protein families by applying it to the superfamily of nuclear receptors with 94.5, 97.8 and 93.6% accuracy in family, level I and level II subfamily classification respectively.     

MUPRED: a tool for bridging the gap between template based methods and sequence profile based methods for protein secondary structure prediction

Predicting secondary structures from a protein sequence is an important step for characterizing the structural properties of a protein. Existing methods for protein secondary structure prediction can be broadly classified into template based or sequence profile based methods. We propose a novel framework that bridges the gap between the two fundamentally different approaches. Our framework integrates the information from the fuzzy k-nearest neighbor algorithm and position-specific scoring matrices using a neural network. It combines the strengths of the two methods and has a better potential to use the information in both the sequence and structure databases than existing methods. We implemented the framework into a software system MUPRED. MUPRED has achieved three-state prediction accuracy (Q3) ranging from 79.2 to 80.14%, depending on which benchmark dataset is used. A higher Q3 can be achieved if a query protein has a significant sequence identity (>25%) to a template in PDB. MUPRED also estimates the prediction accuracy at the individual residue level more quantitatively than existing methods. The MUPRED web server and executables are freely available at http://digbio.missouri.edu/mupred.     

Structural basis for the exocellulase activity of the cellobiohydrolase CbhA from Clostridium thermocellum

Numerous bacterial and fungal organisms have evolved elaborate sets of modular glycoside hydrolases and similar enzymes aimed at the degradation of polymeric carbohydrates. Presently, on the basis of sequence similarity catalytic modules of these enzymes have been classified into 90 families. Representatives of a particular family display similar fold and catalytic mechanisms. However, within families distinctions occur with regard to enzymatic properties and type of activity against carbohydrate chains. Cellobiohydrolase CbhA from Clostridium thermocellum is a large seven-modular enzyme with a catalytic module belonging to family 9. In contrast to other representatives of that family possessing only endo- and, in few cases, endo/exo-cellulase activities, CbhA is exclusively an exocellulase. The crystal structures of the combination of the immunoglobulin-like module and the catalytic module of CbhA (Ig-GH9_CbhA) and that of an inactive mutant Ig-GH9_CbhA(E795Q) in complex with cellotetraose (CTT) are reported here. The detailed analysis of these structures reveals that, while key catalytic residues and overall fold are conserved in this enzyme and those of other family 9 glycoside hydrolases, the active site of GH9_CbhA is blocked off after the -2 subsite. This feature which is created by an extension and altered conformation of a single loop region explains the inability of the active site of CbhA to accommodate a long cellulose chain and to cut it internally. This altered loop region is responsible for the exocellulolytic activity of the enzyme.     

Prediction of protein solvent accessibility using fuzzy k-nearest neighbor method

MOTIVATION: The solvent accessibility of amino acid residues plays an important role in tertiary structure prediction, especially in the absence of significant sequence similarity of a query protein to those with known structures. The prediction of solvent accessibility is less accurate than secondary structure prediction in spite of improvements in recent researches. The k-nearest neighbor method, a simple but powerful classification algorithm, has never been applied to the prediction of solvent accessibility, although it has been used frequently for the classification of biological and medical data. RESULTS: We applied the fuzzy k-nearest neighbor method to the solvent accessibility prediction, using PSI-BLAST profiles as feature vectors, and achieved high prediction accuracies. With leave-one-out cross-validation on the ASTRAL SCOP reference dataset constructed by sequence clustering, our method achieved 64.1% accuracy for a 3-state (buried/intermediate/exposed) prediction (thresholds of 9% for buried/intermediate and 36% for intermediate/exposed) and 86.7, 82.0, 79.0 and 78.5% accuracies for 2-state (buried/exposed) predictions (thresholds of each 0, 5, 16 and 25% for buried/exposed), respectively. Our method also showed slightly better accuracies than other methods by about 2-5% on the RS126 dataset and a benchmarking dataset with 229 proteins. AVAILABILITY: Program and datasets are available at http://biocom1.ssu.ac.kr/FKNNacc/ CONTACT: jul@ssu.ac.kr.     

Discovering the electronic circuit diagram of life: structural relationships among transition metal binding sites in oxidoreductases

Oxidoreductases play a central role in catalysing enzymatic electron-transfer reactions across the tree of life. To first order, the equilibrium thermodynamic properties of these proteins are governed by protein folds associated with specific transition metals and ligands at the active site. A global analysis of holoenzyme structures and functions suggests that there are fewer than approximately 500 fundamental oxidoreductases, which can be further clustered into 35 unique groups. These catalysts evolved in prokaryotes early in the Earth''s history and are largely responsible for the emergence of non-equilibrium biogeochemical cycles on the planet''s surface. Although the evolutionary history of the amino acid sequences in the oxidoreductases is very difficult to reconstruct due to gene duplication and horizontal gene transfer, the evolution of the folds in the catalytic sites can potentially be used to infer the history of these enzymes. Using a novel, yet simple analysis of the secondary structures associated with the ligands in oxidoreductases, we developed a structural phylogeny of these enzymes. The results of this ‘composome’ analysis suggest an early split from a basal set of a small group of proteins dominated by loop structures into two families of oxidoreductases, one dominated by α-helices and the second by β-sheets. The structural evolutionary patterns in both clades trace redox gradients and increased hydrogen bond energy in the active sites. The overall pattern suggests that the evolution of the oxidoreductases led to decreased entropy in the transition metal folds over approximately 2.5 billion years, allowing the enzymes to use increasingly oxidized substrates with high specificity.     

Autocatalytic mechanism and consequences of covalent heme attachment in the cytochrome P4504A family

The prosthetic heme group in the CYP4A family of cytochrome P450 enzymes is covalently attached to an I-helix glutamic acid residue. This glutamic acid is conserved in the CYP4 family but is absent in other P450 families. As shown here, the glutamic acid is linked, presumably via an ester bond, to a hydroxyl group on the heme 5-methyl group. Mutation of the glutamic acid to an alanine in CYP4A1, CYP4A3, and CYP4A11 suppresses covalent heme binding. In wild-type CYP4A3 68% of the heme is covalently bound to the heterologously expressed protein, but in the CYP4A3/E318D mutant, 47% of the heme is unchanged, 47% is present as noncovalently bound 5-hydroxymethylheme, and only 6% is covalently bound to the protein. In the CYP4A3/E318Q mutant, the majority of the heme is unaltered, and <2% is covalently linked. The proportion of covalently bound heme in the recombinant CYP4A proteins increases with time under turnover conditions. The catalytic activity is sensitive in some, but not all, CYP4A enzymes to the extent of covalent heme binding. Mutations of Glu(318) in CYP4A3 decrease the apparent k(cat) values for lauric acid hydroxylation. The key conclusions are that (a) covalent heme binding occurs via an ester bond to the heme 5-methyl group, (b) covalent binding of the heme is mediated by an autocatalytic process, and (c) fatty acid oxidation is sensitive in some CYP4A enzymes to the presence or absence of the heme covalent link.     

Functional diversity of cysteine residues in proteins and unique features of catalytic redox-active cysteines in thiol oxidoreductases

Thiol-dependent redox systems are involved in regulation of diverse biological processes, such as response to stress, signal transduction, and protein folding. The thiol-based redox control is provided by mechanistically similar, but structurally distinct families of enzymes known as thiol oxidoreductases. Many such enzymes have been characterized, but identities and functions of the entire sets of thiol oxidoreductases in organisms are not known. Extreme sequence and structural divergence makes identification of these proteins difficult. Thiol oxidoreductases contain a redox-active cysteine residue, or its functional analog selenocysteine, in their active sites. Here, we describe computational methods for in silico prediction of thiol oxidoreductases in nucleotide and protein sequence databases and identification of their redox-active cysteines. We discuss different functional categories of cysteine residues, describe methods for discrimination between catalytic and noncatalytic and between redox and non-redox cysteine residues and highlight unique properties of the redox-active cysteines based on evolutionary conservation, secondary and three-dimensional structures, and sporadic replacement of cysteines with catalytically superior selenocysteine residues.     

Predicting enzyme family classes by hybridizing gene product composition and pseudo-amino acid composition

A new method has been developed to predict the enzymatic attribute of proteins by hybridizing the gene product composition and pseudo amino acid composition. As a demonstration, a working dataset was generated with a cutoff of 60% sequence identity to avoid redundancy and bias in statistical prediction. The dataset thus constructed contains 39989 protein sequences, of which 27469 are non-enzymes and 12520 enzymes that were further classified into 6 enzyme family classes according to their 6 main EC (Enzyme Commission) numbers (2314 are oxidoreductases, 3653 transferases, 3246 hydrolases, 1307 lyases, 676 isomerases, and 1324 ligases). The overall success rate by the jackknife test for the identification between enzyme and non-enzyme was 94%, and that for the identification among the 6 enzyme family classes was 98%. It is anticipated that, with the rapid increase of protein sequences entering into databanks, the current method will become a useful automated tool in identifying the enzymatic attribute of a newly found protein sequence.     

Affinity purification of bacterial luciferase and NAD(P)H:FMN oxidoreductases by FMN-sepharose for analytical applications

A modified purification method for bacterial luciferases and NAD(P)H:FMN oxidoreductases is described which uses FMN-Sepharose alone or coupled to DEAE ion exchange chromatography for the simultaneous purification of luciferase and the various oxidoreductases from Vibrio harveyi, a bright mutant of Vibrio harveyi, Vibrio fischeri, and Photobacterium phosphoreum. This purification method is compared with DEAE-Sepharose CI 6B fractionations from these organisms. Both methods allow the separation of oxidoreductases specific for either NADH or NADPH. The use of FMN-Sepharose coupled to DEAE-Sepharose fractionation allows the isolation of highly purified enzymes. Lacking interfering factors, these are very suitable for various analytical applications based on bacterial bioluminescence enzymes. The partially purified enzymes from the affinity column have higher specific activities than those obtained using DEAE-Sepharose.     

GH10 XynF1 and Xyn11A: the predominant xylanase identified in the profiling of extracellular proteome of <Emphasis Type="Italic">Aspergillus oryzae</Emphasis> LC1

Advanced techniques of enzyme production and purification have become prerequisite due to their diverse industrial applications. There is an utmost requirement for screening of new strains capable of synthesising industrially useful enzymes. The present study reports the production and profiling of extracellular proteins expressed by the newly isolated strain of a filamentous fungus, Aspergillus oryzae LC1. The extracellular enzyme production was done by submerged fermentation using Mendel’s and Sternberg’s medium (MSM), and its optimisation was done using one factor at a time (OFAT). The presence of xylanase was confirmed by sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) and zymography. In addition, the profiling of extracellular proteome of Aspergillus oryzae LC1 was carried out by liquid chromatography coupled tandem mass spectrometry (LC-MS/MS). In this study, media optimisation showed 5.7-fold increase in xylanase activity. The multiple bands observed in zymography revealed the presence of various forms of xylanase. A total of 73 proteins were identified in LC-MS/MS analysis. Functional classification showed that the hydrolytic enzymes consisted of 48% glycoside hydrolase, 11% proteases, 1% polysaccharide lyase and esterase’s, 9% oxidoreductases and 30% other proteins. A total of 26 families of glycosidic hydrolase were detected with other protein families such as serine peptidase, S, LysM, G-D-S-L, M35, carboxyl esterase (CE1), pectate lyase (PL) and oxidoreductases. Among the huge diversity of synergistically acting biomass cleaving enzymes, endo-1, 4-β xylanase with isoforms: xyn F1, xyn B, β xylanase and xyn 11A belonging to GH10 family covered the major portion of the total percentage of identified proteins. As per our knowledge, this is the first report of extracellular proteome analysis of Aspergillus oryzae LC1 suggesting its capability for recombinant expression and evaluation in hemicellulose deconstruction applications.     

Rapid fluorescent visualization of multiple NAD(P)H oxidoreductases in homogenate,permeabilized cells,and tissue slices

Intracellular NAD(P)H oxidoreductases are a class of diverse enzymes that are the key players in a number of vital processes. The method we present and validate here is based on the ability of many NAD(P)H oxidoreductases to reduce the superoxide probe lucigenin, which is structurally similar to flavins, to its highly fluorescent water-insoluble derivative dimethylbiacridene. Two modifications of the method are proposed: (i) an express method for tissue homogenate and permeabilized cells in suspensions and (ii) a standard procedure for cells in culture and acute thin tissue slices. The method allows one to assess, visualize, and localize, using fluorescent markers of cellular compartments, multiple NADH and NADPH oxidoreductase activities. The application of selective inhibitors (e.g., VAS2870, a NOX2 inhibitor; plumbagin, a NOX4 inhibitor) allows one to distinguish and compare specific NAD(P)H oxidoreductase activities in cells and tissues and to attribute them to known enzymes. The method is simple, rapid, and flexible. It can be easily adapted to a variety of tasks. It will be useful for investigations of the role of various NAD(P)H oxidoreductases in a number of physiological and pathophysiological processes.     

Homology between egg white sulfhydryl oxidase and quiescin Q6 defines a new class of flavin-linked sulfhydryl oxidases.

The flavin-dependent sulfhydryl oxidase from chicken egg white catalyzes the oxidation of sulfhydryl groups to disulfides with the reduction of oxygen to hydrogen peroxide. Reduced proteins are the preferred thiol substrates of this secreted enzyme. The egg white oxidase shows an average 64% identity (from randomly distributed peptides comprising more than 30% of the protein sequence) to a human protein, Quiescin Q6, involved in growth regulation. Q6 is strongly expressed when fibroblasts enter reversible quiescence (Coppock, D. L., Cina-Poppe, D., Gilleran, S. (1998) Genomics 54, 460-468). A peptide antibody against Q6 cross-reacts with both the egg white enzyme and a flavin-linked sulfhydryl oxidase isolated from bovine semen. Sequence analyses show that the egg white oxidase joins human Q6, bone-derived growth factor, GEC-3 from guinea pig, and homologs found in a range of multicellular organisms as a member of a new protein family. These proteins are formed from the fusion of thioredoxin and ERV motifs. In contrast, the flavin-linked sulfhydryl oxidase from Aspergillus niger is related to the pyridine nucleotide-dependent disulfide oxidoreductases, and shows no detectable sequence similarity to this newly recognized protein family.     

Analysis of 15 primary open-angle glaucoma families from Australia identifies a founder effect for the Q368STOP mutation of myocilin

Primary open-angle glaucoma (POAG) is a leading cause of blindness in the world. A number of mutations in the myocilin gene have been identified that predispose to glaucoma. The most frequent of these is the Glutamine368STOP (Q368STOP) mutation. It has been postulated that individuals with the Q368STOP mutation are derived from a common founder. To clarify this situation, we studied 15 unrelated POAG families who carried the Q368STOP mutation, from south eastern Australia. In one large family, nine affected and ten unaffected individuals were identified with the Q368STOP mutation. Closely linked polymorphic microsatellite markers were used to establish a disease haplotype in this family. Additional genotyping of markers in another 14 unrelated Q368STOP families revealed the presence of the same disease haplotype. These findings indicate that the Q368STOP mutation in all 15 families shared a common origin prior to the European settlement of Australia in the early 1800s.     

Aryl-alcohol oxidase protein sequence: a comparison with glucose oxidase and other FAD oxidoreductases

Aryl-alcohol oxidase (AAO), an FAD-dependent enzyme involved in lignin degradation, has been cloned from Pleurotus eryngii. The AAO protein is composed of 593 amino acids, 27 of which form a signal peptide. It shows 33% sequence identity with glucose oxidase from Aspergillus niger and lower homology with other oxidoreductases. The predicted secondary structures of both enzymes are very similar. For AAO, it is predicted to contain 13 putative alpha-helices and two major beta-sheets, each of the putative beta-sheets formed by six beta-strands. The ADP binding site and the signature-2 consensus sequence of the glucose-methanol-choline (GMC) oxidoreductases were also present. Moreover, residues potentially involved in catalysis and substrate binding were identified in the vicinity of the flavin ring. They include two histidines (H502 and H546) and several aromatic residues (Y78, Y92 and F501), as reported in other FAD oxidoreductases.     

Octaheme nitrite reductases: Structure and properties

Octaheme oxidoreductases are widespread among various bacterial taxa involved in the biogeochemical nitrogen cycle. The evolution of octaheme oxidoreductases of the nitrogen cycle from the evolutionarily more ancient pentaheme nitrite reductases was accompanied by changes in function from reduction of nitrogen oxides to their oxidation under changing environmental conditions. Octaheme nitrite reductases, which are the subject of the present review, are of a transitional form that combines structural and functional characteristics of pentaheme reductases and octaheme oxidases and possesses a number of unique features typical of only this family of enzymes. The review summarizes data on structure-function investigations of the family of octaheme nitrite reductases. Emphasis is given to comparison of the structures and functions of octaheme nitrite reductases and other multiheme oxidoreductases of the nitrogen cycle.     

N-linked oligosaccharides of Aspergillus awamori feruloyl esterase are important for thermostability and catalysis

A unique N-linked glycosylation motif (Asn(79)-Tyr-Thr) was found in the sequence of type-A feruloyl esterases from Aspergillus spp. To clarify the function of the flap, the role of N-linked oligosaccharides located in the flap region on the biochemical properties of feruloyl esterase (AwFAEA) from Aspergillus awamori expressed in Pichia pastoris was analyzed by removing the N-linked glycosylation recognition site by site-directed mutagenesis. N79 was replaced with A or Q. N-glycosylation-free N79A and N79Q mutant enzymes had lower activity than that of the glycosylated recombinant AwFAEA wild-type enzyme toward alpha-naphthylbutyrate (C4), alpha-naphthylcaprylate (C8), and phenolic acid methyl esters. Kinetic analysis of the mutant enzymes indicated that the lower catalytic efficiency was due to a combination of increased Km and decreased k(cat) for N79A, and to a considerably decreased k(cat) for N79Q. N79A and N79Q mutant enzymes also exhibited considerably reduced thermostability relative to the wild-type.