Subdivision of the MDR superfamily of medium-chain dehydrogenases/reductases through iterative hidden Markov model refinement

Background  

The Medium-chain Dehydrogenases/Reductases (MDR) form a protein superfamily whose size and complexity defeats traditional means of subclassification; it currently has over 15000 members in the databases, the pairwise sequence identity is typically around 25%, there are members from all kingdoms of life, the chain-lengths vary as does the oligomericity, and the members are partaking in a multitude of biological processes. There are profile hidden Markov models (HMMs) available for detecting MDR superfamily members, but none for determining which MDR family each protein belongs to. The current torrential influx of new sequence data enables elucidation of more and more protein families, and at an increasingly fine granularity. However, gathering good quality training data usually requires manual attention by experts and has therefore been the rate limiting step for expanding the number of available models.     

An improved hidden Markov model for transmembrane protein detection and topology prediction and its applications to complete genomes

MOTIVATION: Knowledge of the transmembrane helical topology can help identify binding sites and infer functions for membrane proteins. However, because membrane proteins are hard to solubilize and purify, only a very small amount of membrane proteins have structure and topology experimentally determined. This has motivated various computational methods for predicting the topology of membrane proteins. RESULTS: We present an improved hidden Markov model, TMMOD, for the identification and topology prediction of transmembrane proteins. Our model uses TMHMM as a prototype, but differs from TMHMM by the architecture of the submodels for loops on both sides of the membrane and also by the model training procedure. In cross-validation experiments using a set of 83 transmembrane proteins with known topology, TMMOD outperformed TMHMM and other existing methods, with an accuracy of 89% for both topology and locations. In another experiment using a separate set of 160 transmembrane proteins, TMMOD had 84% for topology and 89% for locations. When utilized for identifying transmembrane proteins from non-transmembrane proteins, particularly signal peptides, TMMOD has consistently fewer false positives than TMHMM does. Application of TMMOD to a collection of complete genomes shows that the number of predicted membrane proteins accounts for approximately 20-30% of all genes in those genomes, and that the topology where both the N- and C-termini are in the cytoplasm is dominant in these organisms except for Caenorhabditis elegans. AVAILABILITY: http://liao.cis.udel.edu/website/servers/TMMOD/     

Prediction of cell wall sorting signals in gram-positive bacteria with a hidden markov model: application to complete genomes

Surface proteins in Gram-positive bacteria are frequently implicated in virulence. We have focused on a group of extracellular cell wall-attached proteins (CWPs), containing an LPXTG motif for cleavage and covalent coupling to peptidoglycan by sortase enzymes. A hidden Markov model (HMM) approach for predicting the LPXTG-anchored cell wall proteins of Gram-positive bacteria was developed and compared against existing methods. The HMM model is parsimonious in terms of the number of freely estimated parameters, and it has proved to be very sensitive and specific in a training set of 55 experimentally verified LPXTG-anchored cell wall proteins as well as in reliable data sets of globular and transmembrane proteins. In order to identify such proteins in Gram-positive bacteria, a comprehensive analysis of 94 completely sequenced genomes has been performed. We identified, in total, 860 LPXTG-anchored cell wall proteins, a number that is significantly higher compared to those obtained by other available methods. Of these proteins, 237 are hypothetical proteins according to the annotation of SwissProt, and 88 had no homologs in the SwissProt database--this might be evidence that they are members of newly identified families of CWPs. The prediction tool, the database with the proteins identified in the genomes, and supplementary material are available online at http://bioinformatics.biol.uoa.gr/CW-PRED/.     

A block-free hidden Markov model for genotypes and its application to disease association.

We present a new stochastic model for genotype generation. The model offers a compromise between rigid block structure and no structure altogether: It reflects a general blocky structure of haplotypes, but also allows for "exchange" of haplotypes at nonboundary SNP sites; it also accommodates rare haplotypes and mutations. We use a hidden Markov model and infer its parameters by an expectation-maximization algorithm. The algorithm was implemented in a software package called HINT (haplotype inference tool) and tested on 58 datasets of genotypes. To evaluate the utility of the model in association studies, we used biological human data to create a simple disease association search scenario. When comparing HINT to three other models, HINT predicted association most accurately.     

Comparative genomic and phylogenetic analysis of short-chain dehydrogenases/reductases with dual retinol/sterol substrate specificity

Human short-chain dehydrogenases/reductases with dual retinol/sterol substrate specificity (RODH-like enzymes) are thought to contribute to the oxidation of retinol for retinoic acid biosynthesis and to the metabolism of androgenic and neuroactive 3alpha-hydroxysteroids. Here, we investigated the phylogeny and orthology of these proteins to understand better their origins and physiological roles. Phylogenetic and genomic analysis showed that two proteins (11-cis-RDH and RDHL) are highly conserved, and their orthologs can be identified in the lower taxa, such as amphibians and fish. Two other proteins (RODH-4 and 3alpha-HSD) are significantly less conserved. Orthologs for 3alpha-HSD are present in all mammals analyzed, whereas orthologs for RODH-4 can be identified in some mammalian species but not in others due to species-specific gene duplications. Understanding the evolution and divergence of RODH-like enzymes in various vertebrate species should facilitate further investigation of their in vivo functions using animal models.     

Identification and characterization of retinoid-active short-chain dehydrogenases/reductases in Drosophila melanogaster

Background

In chordates, retinoid metabolism is an important target of short-chain dehydrogenases/reductases (SDRs). It is not known whether SDRs play a role in retinoid metabolism of protostomes, such as Drosophila melanogaster.

Methods

Drosophila genome was searched for genes encoding proteins with ∼ 50% identity to human retinol dehydrogenase 12 (RDH12). The corresponding proteins were expressed in Sf9 cells and biochemically characterized. Their phylogenetic relationships were analyzed using PHYLIP software.

Results

A total of six Drosophila SDR genes were identified. Five of these genes are clustered on chromosome 2 and one is located on chromosome X. The deduced proteins are 300 to 406 amino acids long and are associated with microsomal membranes. They recognize all-trans-retinaldehyde and all-trans-3-hydroxyretinaldehyde as substrates and prefer NADPH as a cofactor. Phylogenetically, Drosophila SDRs belong to the same branch of the SDR superfamily as human RDH12, indicating a common ancestry early in bilaterian evolution, before a protostome–deuterostome split.

Conclusions

Similarities in the substrate and cofactor specificities of Drosophila versus human SDRs suggest conservation of their function in retinoid metabolism throughout protostome and deuterostome phyla.

General significance

The discovery of Drosophila retinaldehyde reductases sheds new light on the conversion of β-carotene and zeaxantine to visual pigment and provides a better understanding of the evolutionary roots of retinoid-active SDRs.     

iHMMune-align: hidden Markov model-based alignment and identification of germline genes in rearranged immunoglobulin gene sequences

MOTIVATION: Immunoglobulin heavy chain (IGH) genes in mature B lymphocytes are the result of recombination of IGHV, IGHD and IGHJ germline genes, followed by somatic mutation. The correct identification of the germline genes that make up a variable VH domain is essential to our understanding of the process of antibody diversity generation as well as to clinical investigations of some leukaemias and lymphomas. RESULTS: We have developed iHMMune-align, an alignment program that uses a hidden Markov model (HMM) to model the processes involved in human IGH gene rearrangement and maturation. The performance of iHMMune-align was compared to that of other immunoglobulin gene alignment utilities using both clonally related and randomly selected IGH sequences. This evaluation suggests that iHMMune-align provides a more accurate identification of component germline genes than other currently available IGH gene characterization programs. AVAILABILITY: iHMMune-align cross-platform Java executable and web interface are freely available to academic users and can be accessed at http://www.emi.unsw.edu.au/~ihmmune/.     

Prediction of the coupling specificity of GPCRs to four families of G-proteins using hidden Markov models and artificial neural networks

MOTIVATION: G-protein coupled receptors are a major class of eukaryotic cell-surface receptors. A very important aspect of their function is the specific interaction (coupling) with members of four G-protein families. A single GPCR may interact with members of more than one G-protein families (promiscuous coupling). To date all published methods that predict the coupling specificity of GPCRs are restricted to three main coupling groups G(i/o), G(q/11) and G(s), not including G(12/13)-coupled or other promiscuous receptors. RESULTS: We present a method that combines hidden Markov models and a feed-forward artificial neural network to overcome these limitations, while producing the most accurate predictions currently available. Using an up-to-date curated dataset, our method yields a 94% correct classification rate in a 5-fold cross-validation test. The method predicts also promiscuous coupling preferences, including coupling to G(12/13), whereas unlike other methods avoids overpredictions (false positives) when non-GPCR sequences are encountered. AVAILABILITY: A webserver for academic users is available at http://bioinformatics.biol.uoa.gr/PRED-COUPLE2     

Inhibition of glutamate dehydrogenase and malate dehydrogenases by palmitoyl coenzyme A.

In extension of a previous study with yeast glucose-6-P dehydrogenase (Kawaguchi, A., and Bloch, K. (1974) J. Biol. Chem. 249, 5793-5800), the structural changes accompanying the inhibition of glutamate dehydrogenase and several malate dehydrogenases by palmitoyl-CoA and by sodium dodecyl sulfate have been investigated. Palmitoyl-CoA converts liver glutamate dehydrogenase to enzymatically inactive dimeric subunits (Mr = 1.2 X 10(5)) and tightly binds to the dissociated enzyme. Removal of the inhibitor from the palmitoyl-CoA-dimer complex fails to regenerate enzyme activity. The Ki values for palmitoyl-CoA inhibition of malate dehydrogenases (oxalacetate reduction) are, for the enzyme from pig heart mitochondria, 1.8 muM, 500 muM from pig heart supernatant, and 10 muM from chicken heart supernatant. These inhibitions are readily reversible. Palmitoyl-CoA does not alter the quaternary structure of any of the malate dehydrogenases and binds only weakly to these enzymes. Mitochondrial malate dehydrogenase assayed in the direction malate to oxalacetate is much less sensitive to palmitoyl-CoA, with Ki values of 50 muM at pH 10 and greater than 50 muM at pH 7.4. While the differences in palmitoyl-CoA sensitivity in the forward and backward reactions catalyzed by mitochondrial dehydrogenase are unexplained, a physiological rationale for these differential effects is offered. Sodium dodecyl sulfate dissociates the various dehydrogenases to monomeric subunits in contrast to the more selective effects of palmitoyl-CoA.     

A new member of the short-chain dehydrogenases/reductases superfamily: Purification, characterization and substrate specificity of a recombinant carbonyl reductase from Pichia stipitis

A novel short-chain dehydrogenases/reductases superfamily (SDRs) reductase (PsCR) from Pichia stipitis that produced ethyl (S)-4-chloro-3-hydroxybutanoate with greater than 99% enantiomeric excess, was purified to homogeneity using fractional ammonium sulfate precipitation followed by DEAE-Sepharose chromatography. The enzyme purified from recombinant Escherichia coli had a molecular mass of about 35 kDa on SDS–PAGE and only required NADPH as an electron donor. The K_m value of PsCR for ethyl 4-chloro-3-oxobutanoate was 4.9 mg/mL and the corresponding V_max was 337 μmol/mg protein/min. The catalytic efficiency value was the highest ever reported for reductases from yeasts. Moreover, PsCR exhibited a medium-range substrate spectrum toward various keto and aldehyde compounds, i.e., ethyl-3-oxobutanoate with a chlorine substitution at the 2 or 4-position, or α,β-diketones. In addition, the activity of the enzyme was strongly inhibited by SDS and β-mercaptoethanol, but not by ethylene diamine tetra acetic acid.     

Purification and properties of acyl coenzyme A dehydrogenases from bovine liver. Formation of 2-trans,4-cis-decadienoyl coenzyme A

Three straight chain acyl-CoA dehydrogenases were purified to apparent homogeneity from bovine liver using 40-70% (NH4)2SO4 precipitation, gel filtration, DEAE-cellulose column chromatography, and preparative electrophoresis. Separation of the acyl-CoA dehydrogenases by these procedures has been efficiently monitored by two newly developed analytical methods: (i) native staining of acyl-CoA dehydrogenases following separation by electrophoresis in polyacrylamide gels and (ii) determination of general acyl-CoA dehydrogenase by means of a specific substrate, 4-cis-decenoyl-CoA. The three acyl-CoA dehydrogenases were classified into short chain, general, and long chain acyl-CoA dehydrogenases on the basis of their chain length specificities according to the nomenclature proposed by Hall and Kamin (Hall, C. L., and Kamin, H. (1975) J. Biol. Chem. 250, 3470-3486). The enzymes gave single protein bands in polyacrylamide gel electrophoresis under denaturing and nondenaturing conditions, and their subunit and native molecular weights were estimated to be 40,300 and 188,000 for short chain acyl-CoA dehydrogenase, 43,300 and 205,000 for general acyl-CoA dehydrogenase, and 45,200 and 172,000 for long chain acyl-CoA dehydrogenase. Long chain and general acyl-CoA dehydrogenases markedly differed in their substrate specificities toward unsaturated acyl-CoA esters with a double bond at position 4. The former oxidized 4-cis-decenoyl-CoA at a rate of only 2.7% of that obtained with decanoyl-CoA as substrate, while for the latter enzyme 4-cis-decenoyl-CoA was even a slightly better substrate than decanoyl-CoA. 2-trans,4-cis-Decenoyl-CoA was identified as the product of this reaction.     

2,4-Dienoyl coenzyme A reductases from bovine liver and Escherichia coli. Comparison of properties

2,4-Dienoyl-CoA reductases, enzymes of the beta-oxidation of unsaturated fatty acids which were purified from bovine liver and oleate-induced cells of Escherichia coli, revealed very similar substrate specificities but distinctly different molecular properties. The subunit molecular weights, estimated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis were 32,000 and 73,000 for the mammalian and the bacterial enzyme, respectively. The native molecular weights, calculated from sedimentation coefficients and Stokes radii yielded 124,000 for the bovine liver and 70,000 for the bacterial enzyme. Thus, bovine liver 2,4-dienoyl-CoA reductase is a tetramer consisting of four identical subunits. The E. coli 2,4-dienoyl-CoA reductase, however, possesses a monomeric structure. The latter enzyme contains 1 mol of FAD/mol of enzyme, whereas the former reductase is not a flavoprotein. The bovine liver reductase reduced 2-trans, 4-cis- and 2-trans,4-trans-decadienoyl-CoA to 3-trans-decenoyl-CoA. The E. coli reductase catalyzed the reduction of the same two substrates but in contrast yielded 2-trans-decenoyl-CoA as reaction product. Certain other properties of the two 2,4-dienoyl-CoA reductases are also presented. The localization of the reductase step within the degradation pathway of 4-cis-decenoyl-CoA, a metabolite of linoleic acid, is discussed.     

Mechanism and specificity of succinyl-CoA:3-ketoacid coenzyme A transferase.

(a) The reactivity of substituted acetates as substrates for CoA transferase increases sharply with increasing basicity and exhibits a slope of 1.0 in a plot of either log kappacat or log (kappacat/Km) against pKa (betanuc = 1.0). This result shows that the catalyzed reaction, which involves both carboxylate activation and leaving group transfer, does not proceed through a fully concerted reaction mechanism in the rate-determining step. The result is consistent with a stepwise reaction mechanism that proceeds through an anhydride intermediate. (b) Equilibrium constants for thiol ester formation, either bound to the enzyme or free in solution, show the same dependence on the basicity of carboxylate ions (betaeq = 1.0) and are independent of acidity when expressed in terms of the carboxylic acid. Thus, the polar environment around substituents on the acyl group is the same for carboxylic acids, thiol esters, and oxygen esters. (c) The interaction of the terminal CH3CO group of acetoacetate with the active site causes a 200,000-fold increase in kappacat/Km, corresponding to a decrease in delta G++ OF 7.2 kcal/mol compared with an unsubstituted acid of the same pK. The binding energy of the coenzyme A moiety of the substrate is utilized to interact with the active site and cause a 10(4) to 10(6)-fold increase in kappacat, corresponding to a decrease in delta G++ of 6 to 9 kcal/mol, compared with fragments of the coenzyme A moiety added separatly or together. (d) The exchange of labeled coenzyme A into acyl-CoA substrates was found to be greater than or equal to 10(5) slower than substrate turnover.     

Repetitive sequences in complete and defective genomes of Herpesvirus saimiri.

Two types of Herpesvirus saimiri genomes can be isolated from purified virions: (i) the M genome is a double-stranded, liniear DNA molecule with a mean contour length corresponding to 89 times 10-6 daltons. The M genome contains about 70% of unique sequences (light DNA, 36% guanine plus cytosine) and 30% reiterated sequences (heavy DNA, 71% guanine plus cytosine). (ii) the H genome is composed of heavy DNA only and is more heterogeneous in size. The sequences in the H genome are up to 40-fold reiterated, indicating defectiveness of this type of genome. The repetitions in the H genome and the M genome cross-hybridize almost completely and have identical kinetic complexity (2.8 times 10-6 daltons). DNA infectivity studies by using the calcium phosphate and the DEAE-dextran method gave further evidence that H genomes are defective: no infectious virus was recovered from permissive cells treated with heavy DNA, whereas M genome-infected cells developed cytopathic changes after 11 to 56 days. Defective H genomes were present in the progeny virus two passages after transfection.     

Localization of dehydrogenases, reductases, and electron transfer components in the sulfate-reducing bacterium Desulfovibrio gigas.

Various dehydrogenases, reductases, and electron transfer proteins involved in respiratory sulfate reduction by Desulfovibrio gigas have been localized with respect to the periplasmic space, membrane, and cytoplasm. This species was grown on a lactate-sulfate medium, and the distribution of enzyme activities and concentrations of electron transfer components were determined in intact cells, cell fractions prepared with a French press, and lysozyme spheroplasts. A significant fraction of formate dehydrogenase was demonstrated to be localized in the periplasmic space in addition to hydrogenase and some c-type cytochrome. Cytochrome b, menaquinone, fumarate reductase, and nitrite reductase were largely localized on the cytoplasmic membrane. Fumarate reductase was situated on the inner aspect on the membrane, and the nitrite reductase appeared to be transmembraneous. Adenylylsulfate reductase, bisulfite reductase (desulfoviridin), pyruvate dehydrogenase, and succinate dehydrogenase activities were localized in the cytoplasm. Significant amounts of hydrogenase and c-type cytochromes were also detected in the cytoplasm. Growth of D. gigas on a formate-sulfate medium containing acetate resulted in a 10-fold increase in membrane-bound formate dehydrogenase and a doubling of c-type cytochromes. Growth on fumarate with formate resulted in an additional increase in b-type cytochrome compared with lactate-sulfate-grown cells.     

Orientation of coenzyme A substrates, nicotinamide and active site functional groups in (Di)enoyl-coenzyme A reductases

The stereochemical course of reduction of dienoyl-coenzyme A (CoA) thiolesters catalyzed by the 2,4-dienoyl-CoA reductase from rat liver mitochondria was investigated. The configuration of the double bond in the 3-enoyl-CoA products was determined by (1)H NMR, and experiments to determine the stereochemical course of reduction at Calpha and Cdelta by use of 4-(2)H-labeled beta-nicotinamide adenine dinucleotide phosphate, reduced form (NADPH), were conducted in H(2)O and D(2)O. Defining the diastereoselectivity of the reaction, catalyzed by the Delta(3),Delta(2)-enoyl-CoA isomerase, facilitated the determination of the stereochemical course of reduction by 2, 4-dienoyl-CoA reductase. The absence of solvent exchange of the proton transferred during the Delta(3),Delta(2)-enoyl-CoA isomerase catalyzed equilibration of trans-2- and trans-3-enoyl-CoAs, coupled with the strong sequence homology to enoyl-CoA hydratase support the intramolecular suprafacial transfer of the pro-2R proton of trans-3-enoyl-CoA to the pro-4R position of trans-2-enoyl-CoA. The results indicate that the configuration of the double bond of the 3-enoyl-CoA product is trans and that a general acid-catalyzed addition of a solvent derived proton/deuteron occurs on the si face at Calpha of the dienoyl-CoA. The addition of the pro-4S hydrogen from NADPH occurs on the si face at Cdelta of trans-2, cis-4-dienoyl-CoA and on the re face at Cdelta of trans-2, trans-4-dienoyl-CoA. The stereochemical course of reduction of InhA, an enoyl-thiolester reductase from Mycobacterium tuberculosis, was also determined by use of ?4-(2)HNADH in D(2)O. The reduction of trans-2-octenoyl-CoA catalyzed by InhA resulted in the syn addition of (2)H(2) across the double bond yielding (2R,3S)-?2, 3-(2)H(2)?ctanoyl-CoA. In the crystal structure of the InhA ternary complex, the residue donating the proton to Calpha could not be identified ?Rozwarski, D. A., Vilcheze, C., Sugantino, M., Bittman, R., and Sacchettini, J. C. (1999) J. Biol. Chem. 274, 15582-15589. The current results place further restrictions on the source of the proton and suggest the reduction is stepwise.