Published papers

Annotations of complete genome sequences submitted directly from sequencing projects are diverse in terms of annotation strategies and update frequencies. These inconsistencies make comparative studies difficult. To allow rapid data preparation of a large number of complete genomes, automation and speed are important for genome re-annotation. Here we introduce an open-source rapid genome re-annotation software system, Restauro-G, specialized for bacterial genomes. Restauro-G re-annotates a genome by similarity searches utilizing the BLAST-Like Alignment Tool, referring to protein databases such as UniProt KB, NCBI nr, NCBI COGs, Pfam, and PSORTb. Re-annotation by Restauro-G achieved over 98% accuracy for most bacterial chromosomes in comparison with the original manually curated annotation of EMBL releases. Restauro-G was developed in the generic bioinformatics workbench G-language Genome Analysis Environment and is distributed at http://restauro-g.iab.keio.ac.jp/under the GNU General Public License. © 2007 Beijing Genomics Institute.

Background. Gene Regulatory Networks (GRNs) have become a major focus in interest in recent years. A number of reverse engineering approaches have been developed to help uncover the regulatory networks giving rise to the observed gene expression profiles. However, this is an overspecified problem due to the fact that more than one genotype (network wiring) can give rise to the same phenotype. We refer to this phenomenon as "gene elasticity." In this work, we study the effect of this particular problem on the pure, data-driven inference of gene regulatory networks. Methodology. We simulated a four-gene network in order to produce "data" (protein levels) that we use in lieu of real experimental data. We then optimized the network connections between the four genes with a view to obtain the original network that gave rise to the data. We did this for two different cases: one in which only the network connections were optimized and the other in which both the network connections as well as the kinetic parameters (given as reaction probabilities in our case) were estimated. We observed that multiple genotypes gave rise to very similar protein levels. Statistical experimentation indicates that it is impossible to differentiate between the different networks on the basis of both equilibrium as well as dynamic data. Conclusions. We show explicitly that reverse engineering of GRNs from pure expression data is an indeterminate problem. Our results suggest the unsuitability of an inferential, purely data-driven approach for the reverse engineering transcriptional networks in the case of gene regulatory networks displaying a certain level of complexity. © 2007 Krishnan et al.

Background. A theoretical model of genetic redundancy has proposed that the fates of redundant genes depend on the degree of functional redundancy, and that functionally redundant genes will not be inherited together. However, no example of actual gene evolution has been reported that can be used to test this model. Here, we analyzed the molecular evolution of the ribonuclease H (RNase H) family in prokaryotes and used the results to examine the implications of functional redundancy for gene evolution. Results. In prokaryotes, RNase H has been classified into RNase HI, HII, and HIII on the basis of amino acid sequences. Using 353 prokaryotic genomes, we identified the genes encoding the RNase H group and examined combinations of these genes in individual genomes. We found that the RNase H group may have evolved in such a way that the RNase HI and HIII genes will not coexist within a single genome - in other words, these genes are inherited in a mutually exclusive manner. Avoiding the simultaneous inheritance of the RNase HI and HIII genes is remarkable when RNase HI contains an additional non-RNase H domain, double-stranded RNA, and an RNA-DNA hybrid-binding domain, which is often observed in eukaryotic RNase H1. This evolutionary process may have resulted from functional redundancy of these genes, because the substrate preferences of RNase HI and RNase HIII are similar. Conclusion. We provide two possible evolutionary models for RNase H genes in which functional redundancy contributes to the exclusion of redundant genes from the genome of a species. This is the first empirical study to show the effect of functional redundancy on changes in gene constitution during the course of evolution. © 2007 Kochiwa et al; licensee BioMed Central Ltd.

Proteins play a critical role in complex biological systems, yet about half of the proteins in publicly available databases are annotated as functionally unknown. Proteome-wide functional classification using bioinformatics approaches thus is becoming an important method for revealing unknown protein functions. Using the hyperthermophilic archaeon Pyrococcus furiosus as a model species, we used the support vector machine (SVM) method to discriminate DNA/RNA-binding proteins from proteins with other functions, using amino acid composition and periodicities as feature vectors. We defined this value as the composition score (CO) and periodicity score (PD). The P. furiosus proteins were classified into three classes (I-III) on the basis of the two-dimensional correlation analysis of CO score and PD score. As a result, approximately 87 of the functionally known proteins categorized as class I proteins (CO score + PD score > 0.6) were found to be DNA/RNA-binding proteins. Applying the two-dimensional correlation analysis to the 994 hypothetical proteins in P. furiosus, a total of 151 proteins were predicted to be novel DNA/RNA-binding protein candidates. DNA/RNA-binding activities of randomly chosen hypothetical proteins were experimentally verified. Six out of seven candidate proteins in class I possessed DNA/RNA-binding activities, supporting the efficacy of our method. © The Author 2007. Kazusa DNA Research Institute.

The Bacillus subtilis GenoMe (BGM) vector was designed as a versatile vector for the cloning of giant DNA segments. Cloned DNA in the BGM can be retrieved to a plasmid using our Bacillus recombinational transfer (BReT) method that takes advantage of competent cell transformation. However, delivery of the plasmid to a different B. subtilis strain by the normal transformation method is hampered by DNA size-related inefficiency. Therefore, we designed a novel method, conjugational plasmid-mediated DNA retrieval and transfer (CReT) from the BGM vector, and investigated conjugational transmission to traverse DNA between cells to circumvent the transformation-induced size limitation. pLS20, a 65-kb plasmid capable of conjugational transfer between B. subtilis strains, was modified to retrieve DNA cloned in the BGM vector by homologous recombination during normal culture. As the plasmid copy number was estimated to be 3, the retrieval plasmid was selected using increased numbers of marker genes derived from the retrieved DNA. We applied this method to retrieve Synechocystis genome segments up to 90 kb in length. We observed retrieved plasmid transfers between B. subtilis strains by conjugation in the absence of structural alterations in the DNA fragment. Our observations extend DNA transfer protocols over previously exploited size ranges. © 2007 Elsevier B.V. All rights reserved.

A method for the determination of nucleotides based on pressure-assisted capillary electrophoresis-electrospray ionization mass spectrometry (PACE-MS) is described. To prevent multi-phosphorylated species from adsorbing onto the fused-silica capillary, silanol groups were masked with phosphate ions by preconditioning the capillary with the background electrolyte containing phosphate. During preconditioning, nebulizer gas was turned off to avoid contamination of MS detector with phosphate ions. To detect nucleotides using the CE positive mode at a pH 7.5, it was necessary to apply air pressure to the inlet capillary during electrophoresis to supplement the electroosmotic flow (EOF) toward the cathode. Moreover, we exchanged the running electrolyte every analysis using the buffer replenishment system to obtain the required reproducibility. Under the optimized conditions, 14 phosphorylated species such as nucleotides, nicotinamide-adenine dinucleotides and coenzyme A (CoA) compounds were well determined in less than 20 min. The relative standard deviations (n = 6) of the method were better than 0.9% for migration times and between 1.7% and 8.1% for peak areas. The detection limits for these species were between 0.5 and 1.7 μmol/L with pressure injection of 50 mbar for 30 s (30 nL) at a signal-to-noise ratio of 3. This approach is robust and quantitative compared to the previous method, and its utility is demonstrated by the analysis of intracellular nucleotides and CoA compounds extracted from Escherichia coli wild type, pfkA and pfkB knockout mutants. The methodology was used to suggest that pfkA is the main functional enzyme. © 2007 Elsevier B.V. All rights reserved.

We have developed a metabolic flux analysis method that is based on 13C-labeling patterns of the intracellular metabolites directly measured by capillary electrophoresis time-of-flight mass spectrometry (CE-TOFMS). The flux distribution of the central carbon metabolism in Escherichia coli was determined by this new approach and the results were compared with findings obtained by conventional GC-MS analysis based on isotopomer of the proteinogenic amino acids. There were some differences in estimation results between new approach using CE-TOFMS and conventional approach using GC-MS. These were thought to be attributable to variations in measured mass distributions between amino acids and the corresponding precursors and to differences in the sensitivity of the exchange coefficients to mass distributions. However, our CE-TOFMS method facilitates high-throughput flux analysis without requiring complicated sample preparation such as hydrolysis of proteins and derivatization of amino acids. © 2007 Elsevier B.V. All rights reserved.

In the cyanobacterium, Synechococcus elongatus, most promoters are regulated by a circadian clock under continuous light (LL) conditions. Nevertheless, the basic circadian oscillation is primarily generated by alternating KaiC phosphorylation/dephosphorylation reactions at the posttranslational level. Indeed, the KaiC phosphorylation cycle was recently reconstituted in vitro by incubating KaiA, KaiB, and KaiC proteins with ATP. However, the molecular dynamics of this chemical oscillation and the mechanism that drives the circadian transcription/translation rhythms remain unknown. In this report, the KaiC phosphorylation cycle and the gene regulatory network in the cyanobacterial circadian system have been modeled. The model reproduces the robust KaiC phosphorylation cycle in the absence of de novo gene expression as is observed in vitro, as well as its coupling to transcriptional/translational feedback in LL conditions in vivo. Moreover, the model is consistent with most previous experiments, including various combinations of genetic knockout or overexpression of kai genes. It also predicts that multiple KaiC phosphorylation states and dynamic Kai protein interactions may be required for the cyanobacterial circadian system. © 2007 Sage Publications.

Bacterial chromosomes are highly polarized in their nucleotide composition through mutational selection related to replication. Using compositional skews such as the GC skew, replication origin and terminus can be predicted in silico by observing the shift points. However, the genome sequence is affected by myriad functional requirements and selection on numerous subgenomic features, and elimination of this "noise" should lead to better predictions. Here, we present a noise-reduction approach that uses low-pass filtering through Fast Fourier transform coupled with cumulative skew graphs. It increases the prediction accuracy of the replication termini compared with previously documented methods based on genomic base composition. © 2006 Federation of European Biochemical Societies.

RNA decay is thought to exert an important influence on gene expression by maintaining a steady-state level of transcripts and/or by eliminating aberrant transcripts. However, the sequence elements which control such processes have not been determined. Upstream open reading frames (uORFs) in the transcripts of several genes are reported to control translational initiation by stalling ribosomes and thereby promote RNA decay. We therefore performed bioinformatic analysis of the tissue-wide expression profiles and mRNA half-life of transcripts containing uORFs in humans and mice to assess the relationship between RNA decay and the presence of uORFs in transcripts. The expression levels of transcripts containing uORF were markedly lower than those not containing uORF. Moreover, the half-life of the uORF-containing transcripts was also shorter. These results suggest that uORFs are sequence elements that down-regulate RNA transcripts via RNA decay mechanisms. © 2007 Federation of European Biochemical Societies.

Background. In the process of constructing a dynamic model of a metabolic pathway, a large number of parameters such as kinetic constants and initial metabolite concentrations are required. However, in many cases, experimental determination of these parameters is time-consuming. Therefore, for large-scale modelling, it is essential to develop a method that requires few experimental parameters. The hybrid dynamic/static (HDS) method is a combination of the conventional kinetic representation and metabolic flux analysis (MFA). Since no kinetic information is required in the static module, which consists of MFA, the HDS method may dramatically reduce the number of required parameters. However, no adequate method for developing a hybrid model from experimental data has been proposed. Results. In this study, we develop a method for constructing hybrid models based on metabolome data. The method discriminates enzymes into static modules and dynamic modules using metabolite concentration time series data. Enzyme reaction rate time series were estimated from the metabolite concentration time series data and used to distinguish enzymes optimally for the dynamic and static modules. The method was applied to build hybrid models of two microbial central-carbon metabolism systems using simulation results from their dynamic models. Conclusion. A protocol to build a hybrid model using metabolome data and a minimal number of kinetic parameters has been developed. The proposed method was successfully applied to the strictly regulated central-carbon metabolism system, demonstrating the practical use of the HDS method, which is designed for computer modelling of metabolic systems. © 2007 Ishii et al; licensee BioMed Central Ltd.

The ergodic hypothesis, which assumes the independence of each cell of the ensemble from all the others, is a necessary prerequisite to attach single cell based explanations to the grand averages taken from population data. This was the prevailing view about the interpretation of cellular biology experiments that typically are performed on colonies of billions of cells. By analysing gene expression data of different cells going from yeast to mammalian cell cultures, we demonstrate that cell cultures display a sort of 'ecology-in-a-plate' giving rise to a rich dynamics of gene expression that are independent from reproductive cycles, hence contradicting simple ergodic assumptions The aspecific character of the observed coordinated gene expression activity inhibits any simple mechanistic hypothesis and highlights the need to consider population effects in the interpretation of data coming from cell cultures. © 2007 The Authors.

Metabolomics is expected to boost data driven research. In biomarker discovery, powerful filtering methods to remove noise and outliers are essential for screening significant candidates from the huge volume of omic data. Here we propose a post-measurement peak filtering method (named P-BOSS) for CE electrospray ionization-time-of-flight MS (CE-TOFMS) data. Combining outlier detection method functions in parallel, we applied P-BOSS to the data using Escherichia coli knockout mutants of the tryptophan and purine biosynthesis pathways. As the result, P-BOSS showed remarkably superior performance, reducing 65% of all peaks, while leaving significant peaks. © 2007 Elsevier B.V. All rights reserved.

When erythrocytes are exposed to hypoxia, hemoglobin (Hb) stabilizes in theT-state by capturing 2,3-bisphosphoglycerate. This process could reduce the intracellular pool of glycolytic substrates, jeopardizing cellular energetics. Recent observations suggest that hypoxia-induced activation of glycolytic enzymes is correlated with their release from Band III (BIII) on the cell membrane. Based on these data, we developed a mathematical model of erythrocyte metabolism and compared hypoxia-induced differences in predicted activities of the enzymes, their products, and cellular energetics between models with and without the interaction of Hb with BIII. The models predicted that the allostery-dependent Hb interaction with BIII accelerates consumption of upstream glycolytic substrates such as glucose 6-phosphate and increases downstream products such as phosphoenolpyruvate. This prediction was consistent with metabolomic data from capillary electrophoresis mass spectrometry. The hypoxia-induced alterations in the metabolites resulted from acceleration of glycolysis, as judged by increased conversion of [13C]glucose to [13C]lactate. The allostery-dependent interaction of Hb with BIII appeared to contribute not only to maintenance of energy charge but also to further synthesis of 2,3-bisphosphoglycerate, which could help sustain stabilization of T-state Hb during hypoxia. Furthermore, such an activation of glycolysis was not observed when Hb was stabilized in R-state by treating the cells with CO. These results suggest that Hb allostery in erythrocytes serves as an O2-sensing trigger that drives glycolytic acceleration to stabilize intracellular energetics and promote the ability to release O 2 from the cells. © 2007 by The American Society for Biochemistry and Molecular Biology, Inc.

Background. Cellular signaling involves a sequence of events from ligand binding to membrane receptors through transcription factors activation and the induction of mRNA expression. The transcriptional-regulatory system plays a pivotal role in the control of gene expression. A novel computational approach to the study of gene regulation circuits is presented, here. Methodology. Based on the concept of finite state machine, which provides a discrete view of gene regulation, a novel sequential logic model (SLM) is developed to decipher control mechanisms of-dynamic transcriptional regulation of gene expressions. The SLM technique is also used to systematically analyze the dynamic function of transcriptional inputs, the dependency and cooperativity, such as synergy effect, among the binding sites with respect to when, how much and how fast the gene of interest is expressed. Principal Findings. SLM is verified by a set of well studied expression data on endo 16 of Strongylocentrotus purpuratus (sea urchin) during the embryonic midgut development. A dynamic regulatory mechanism for endo 16 expression controlled by three binding sites, UI, R and Otx is identified and demonstrated to be consistent with experimental findings. Furthermore, we show that during transition from specification to differentiation in wild type endo16 expression profile, SLM reveals three binary activities are not sufficient to explain the transcriptional regulation of endo16 expression and additional activities of binding sites are required. Further analyses suggest detailed mechanism of R switch activity where indirect dependency occurs in between UI activity and R switch during specification to differentiation stage. Conclusions/Significatwe. The sequential logic formalism allows for a simplification of regulation network dynamics going from a continuous to a discrete representation of gene activation in time. In effect our SLM is non-parametric and model-independent, yet providing rich biological insight. The demonstration of the efficacy of this approach in endo 16 is a promising step for further application of the proposed method. © 2007 Yeo et al.

Methemoglobin (metHb), an oxidized form of hemoglobin, is unable to bind and carry oxygen. Erythrocytes are continuously subjected to oxidative stress and nitrite exposure, which results in the spontaneous formation of metHb. To avoid the accumulation of metHb, reductive pathways mediated by cytochrome b5 or flavin, coupled with NADH-dependent or NADPH-dependent metHb reductases, respectively, keep the level of metHb in erythrocytes at less than 1% of the total hemoglobin under normal conditions. In this work, a mathematical model has been developed to quantitatively assess the relative contributions of the two major metHb-reducing pathways, taking into consideration the supply of NADH and NADPH from central energy metabolism. The results of the simulation experiments suggest that these pathways have different roles in the reduction of metHb; one has a high response rate to hemoglobin oxidation with a limited reducing flux, and the other has a low response rate with a high capacity flux. On the basis of the results of our model, under normal oxidative conditions, the NADPH-dependent system, the physiological role of which to date has been unclear, is predicted to be responsible for most of the reduction of metHb. In contrast, the cytochrome b5-NADH pathway becomes dominant under conditions of excess metHb accumulation, only after the capacity of the flavin-NADPH pathway has reached its limit. We discuss the potential implications of a system designed with two metHb-reducing pathways in human erythrocytes. © 2007 The Authors.

Chi sites (5′-GCTGGTGG-3′) are homologous recombinational hotspot octamer sequences, which attenuate the exonuclease activity of RecBCD in Escherichia coli. They are overrepresented in the genome (1008 occurrences), preferentially located within coding regions (98%), oriented in the direction of replication (75%), and occur most commonly on the mRNA-synonymous sense strand of the double helix (79%). Previous statistical studies of the genome sequence suggested that these genomic properties of Chi sites appear to be related to their role in recombinational repair and therefore to replication and transcription. In this study, we employ three mathematical models to predict the properties of Chi sites from single nucleotide and multi-nucleotide compositions, and validate them statistically using the distribution of all octamer sequences in the entire genome, or exclusively within ORFs. The model based on the overall distribution of all octamers provided better predictions than the single nucleotide composition model, and the ORF and sense strand preference of Chi sites were shown to be within the standard deviation of all octamers. In contrast, the orientation bias of the Chi sites in the direction of replication was significant, although the bias was not as pronounced as with the single nucleotide composition model, suggesting a selective pressure related to the role of RecBCD in replication. © 2007 Elsevier B.V. All rights reserved.

The E-CELL System (http://www.e-cell.org) is a genetic software package developed, to model not only metabolic pathways but also other higher-order cellular processes such as protein synthesis and signal transduction. Using the system, we have successfully constructed a virtual cell with 127 genes sufficient for 'self-support'. The gene set was selected from the genome of Mycoplasma genitalium, and the metabolisms include transcription, translation, membrane transport, the glycolysis pathway for energy production, and the phospholipid biosynthesis pathway or membrane structure.

Tomita, M. and Carbonell. J.G.; in Machine Translation Summit, Nagao, M. (ed.), Ohmsha, (1989)