Published papers

A method for simultaneous determination of anionic metabolites based on capillary electrophoresis (CE) coupled to electrospray ionization mass spectrometry is described. To prevent current drop by the system, electroosmotic flow (EOF) reversal by using a cationic polymer-coated capillary was indispensable. A mixture containing 32 standards including carboxylic acids, phosphorylated carboxylic acids, phosphorylated saccharides, nucleotides, and nicotinamide and flavin adenine coenzymes of glycolysis and the tricarboxylic acid cycle pathways were separated by CE and selectively detected by a quadrupole mass spectrometer with a sheath-flow electrospray ionization interface. Key to the analysis was EOF reversal using a cationic polymer-coated capillary and an electrolyte system consisting of 50 mM ammonium acetate, pH 9.0. The relative standard deviations of the method were better than 0.4% for migration times and between 0.9% and 5.4% for peak areas. The concentration detection limits for these metabolites were between 0.3 and 6.7 μmol/L with pressure injection of 50 mbar for 30 s (30 nL); i.e., mass detection limits ranged from 9 to 200 fmol, at a signal-to-noise ratio of 3. This method was applied to the comprehensive analysis of metabolic intermediates extracted from Bacillus subtilis, and 27 anionic metabolites could be directly detected and quantified.

Background: Genetic background of a fetus contributes to the abnormal development after teratogen exposure. In rodents, in utero exposure to dioxins affects male external genital development. The effects of dioxins are mediated via the aryl hydrocarbon receptor (AHR) and its binding protein, aryl hydrocarbon receptor nuclear translocator (ARNT). In mice, aryl hydrocarbon receptor repressor (AHRR), which binds to ARNT in competition with AHR, plays a critical negative regulatory role in AHR signaling. We attempt to characterize the human AHRR gene and investigate the relationship between AHRR polymorphisms and the incidence of micropenis, a phenotype of undermasculinization. Methods: We identified and characterized the human homolog of mouse AHRR, taking advantage of the publicly available draft version of the human genome sequence. After detecting an AHRR protein polymorphism by the direct sequencing of pooled human genomic DNA, we evaluated the association between the polymorphism and the presence or absence of micropenis (<-2.5 SD) in patients with micropenis and control subjects. Results: The deduced sequence for human AHRR (715 residues) and the mouse AHRR protein exhibited 81% sequence homology to each other. The Pro185Ala polymorphism was identified between the PAS-A region and the highly conserved arginine/cysteine-rich RCFRCRL/VRC region. Forty-six percent (27/59) of patients with micropenis and 27% (22/80) of the controls were homozygous for 185Pro; this difference in frequencies was significant (P = 0.03). Conclusions: Homozygosity for the 185Pro allele of AHRR may increase the susceptibility of a fetus to the undermasculinizing effects of dioxin exposure in utero, presumably through the diminished inhibition of AHR-mediated signaling. © 2002 Wiley-Liss, Inc.

The processing of 16S rRNA and 23S rRNA by RNase III in E.coli is known to involve stem structures formed by both ends of the rRNA. Indeed, complementary nucleotide sequences are usually found at both ends of 16S rRNA and 23S rRNA. However, whether or not this phenomenon exists in various other bacteria has not yet been adequately studied. We have conducted computer analyses of potential stem structures of rRNA operons in 12 bacterial and 3 archaeal genomes, and compared characteristics of the stem structures among these species. We systematically computed free energy values by exhaustively 'annealing' sequences around the 5′ end and sequences around the 3′ end of both 16S rRNA and 23S rRNA genes, in order to predict potential stem structures. The results suggest that rRNAs in most species form stem structures at both ends. Some species, such as A.aeolicus, seem to form unusually stable stem structures. On the other hand, some rRNAs, such as rRNAs of D.radiodurans, seem not to form solid stem structures. This suggests that rRNA processing in those species must employ a reliable targeting mechanism other than recognizing stem structures by RNase III. © 2000 Elsevier Science B.V.

Only a small proportion of the mouse genome is transcribed into mature messenger RNA transcripts. There is an international collaborative effort to identify all full-length mRNA transcripts from the mouse, and to ensure that each is represented in a physical collection of clones. Here we report the manual annotation of 60,770 full-length mouse complementary DNA sequences. These are clustered into 33,409 'transcriptional units', contributing 90.1% of a newly established mouse transcriptome database. Of these transcriptional units, 4,258 are new protein-coding and 11,665 are new non-coding messages, indicating that non-coding RNA is a major component of the transcriptome. 41% of all transcriptional units showed evidence of alternative splicing. In protein-coding transcripts, 79% of splice variations altered the protein product. Whole-transcriptome analyses resulted in the identification of 2,431 sense-antisense pairs. The present work, completely supported by physical clones, provides the most comprehensive survey of a mammalian transcriptome so far, and is a valuable resource for functional genomics.

Recent investigations into the translation termination sites of various organisms have revealed that not only stop codons but also sequences around stop codons have an effect on translation termination. To investigate the relationship between these sequence patterns and translation as well as its termination efficiency, we analysed the correlation between strength of consensus and translation efficiency, as predicted according to Codon Adaptation Index (CAI) value. We used RIKEN full-length mouse cDNA sequences and ten other eukaryotic UniGene datasets from NCBI for the analyses. First, we conducted sequence profile analyses following translation termination sites. We found base G and A at position +1 as a strong consensus for mouse cDNA. A similar consensus was found for other mammals, such as Homo sapiens, Rattus norvegicus and Bos taurus. However, some plants had different consensus sequences. We then analysed the correlation between the strength of consensus at each position and the codon biases of whole coding regions, using information content and CAI value. The results showed that in mouse cDNA, CAI value had a positive correlation with information content at positions +1. We also found that, for positions with strong consensus, the strength of the consensus is likely to have a positive correlation with CAI value in some other eukaryotes. Along with these observations, biological insights into the relationship between gene expression level, codon biases and consensus sequence around stop codons will be discussed. © 2002 Published by Elsevier Science B.V.

G-language Genome Analysis Environment (G-language GAE) is an open source generic software package aimed for higher efficiency in bioinformatics analysis. G-language GAE has an interface as a set of Perl libraries for software development, and a graphical user interface for easy manipulation. Both Windows and Linux versions are available.

With the sequencing and annotation of genomes and transcriptomes of several eukaryotes, the importance of noncoding RNA (ncRNA)-RNA molecules that are not translated to protein products - has become more evident. A subclass of ncRNA transcripts are encoded by highly regulated, multi-exon, transcriptional units, are processed like typical protein-coding mRNAs and are increasingly implicated in regulation of many cellular functions in eukaryotes. This study describes the identification of candidate functional ncRNAs from among the RIKEN mouse full-length cDNA collection, which contains 60,770 sequences, by using a systematic computational filtering approach. We initially searched for previously reported ncRNAs and found nine murine ncRNAs and homologs of several previously described nonmouse ncRNAs. Through our computational approach to filter artifact-free clones that lack protein coding potential, we extracted 4280 transcripts as the largest-candidate set. Many clones in the set had EST hits, potential CpG islands surrounding the transcription start sites, and homologies with the human genome. This implies that many candidates are indeed transcribed in a regulated manner. Our results demonstrate that ncRNAs are a major functional subclass of processed transcripts in mammals.

Motivation: Readthrough is an unusual process in which a stop codon is misread or skipped. Recently it has been shown that some translation is regulated by the readthrough reactions although the complete mechanism is not clear. Therefore, the discovery of 'readthrough genes' is important for further investigation of their cellular roles, which may provide additional insights into the mechanism of translational regulation. Results: We constructed a system that lists candidates of readthrough genes based on the existence of a 'protein motif' at the 3′ untranslated region (UTR). Using this system, we extracted 85 candidates from 4082 nucleic acid sequences of Drosophila melanogaster in GenBank database. The sequences of these candidates had a slightly more stable secondary structure and different base preferences compared to the non-candidates. As these features are known to have an effect on readthrough events, we would like to suggest that these candidates contain actual readthrough genes.

A reconstitution system that recapitulates the processing of Okazaki-primer RNA was established by the heat-stable recombinant enzymes RNase HII and FEN-1 (termed Pf-RNase HII and Pf-FEN-1, respectively) prepared from a hyperthermophilic archaeon, Pyrococcus furiosus. A 35-mer RNA-DNA/DNA hybrid substrate mimicking an Okazaki fragment was used to investigate the properties of the processing reaction in vitro at 50°C. Pf-RNase HII endonucleolytically cleaves the RNA primer region, but does not cut the junction between RNA and DNA. Removal of the RNA of the RNA-DNA junction was brought about by Pf-FEN-1 after Pf-RNase HII digestion. In the presence of 0.25-5mM MnCl2, Pf-FEN-1 alone weakly cleaved the junction. The addition of Pf-RNase HII to the reaction mixture increased removal efficiency and optimal Pf-FEN-1 activity was achieved at an equal amount of the two enzymes. These results indicate that there are at least two steps in the degradation of primer RNA requiring a step-specific enzyme. It is likely that Pf-RNase HII and Pf-FEN-1 cooperatively process Okazaki fragment during lagging-strand DNA replication. © 2003 Elsevier Inc. All rights reserved.

Motivation: Molecular biotechnology now makes it possible to build elaborate systems models, but the systems biology community needs information standards if models are to be shared, evaluated and developed cooperatively. Results: We summarize the Systems Biology Markup Language (SBML) Level 1, a free, open, XML-based format for representing biochemical reaction networks. SBML is a software-independent language for describing models common to research in many areas of computational biology, including cell signaling pathways, metabolic pathways, gene regulation, and others.

Motivation: The modeling of system dynamics of genetic networks, metabolic networks or signal transduction cascades from time-course data is formulated as a reverse-problem. Previous studies focused on the estimation of only network structures, and they were ineffective in inferring a network structure with feedback loops. We previously proposed a method to predict not only the network structure but also its dynamics using a Genetic Algorithm (GA) and an S-system formalism. However, it could predict only a small number of parameters and could rarely obtain essential structures. In this work, we propose a unified extension of the basic method. Notable improvements are as follows: (1) an additional term in its evaluation function that aims at eliminating futile parameters; (2) a crossover method called Simplex Crossover (SPX) to improve its optimization ability; and (3) a gradual optimization strategy to increase the number of predictable parameters. Results: The proposed method is implemented as a C program called PEACE1 (Predictor by Evolutionary Algorithms and Canonical Equations 1). Its performance was compared with the basic method. The comparison showed that: (1) the convergence rate increased about 5-fold; (2) the optimization speed was raised about 1.5-fold; and (3) the number of predictable parameters was increased about 5-fold. Moreover, we successfully inferred the dynamics of a small genetic network constructed with 60 parameters for 5 network variables and feedback loops using only time-course data of gene expression.

A computer-based analysis was conducted to assess the characteristics of microsatellites in transcribed regions of rice and Arabidopsis. In addition, two mammals were simultaneously analyzed for a comparative analysis. Our analyses confirmed a novel plant-specific feature in which there is a gradient in microsatellite density along the direction of transcription. It was also confirmed that pyrimidine-rich microsatellites are found intensively near the transcription start sites, specifically in the two plants, but not in the mammals. Our results suggest that microsatellites located at high frequency in the 5′-flanking regions of plant genes can potentially act as factors in regulating gene expression. © 2003 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.

Summary: A new version of the E-Cell simulation system, which runs on Windows as well as Linux, has been released as free software under the terms of the GNU General Public License.

We modeled and analyzed a signal transduction system of long-term potentiation (LTP) in hippocampal post-synapse. Bhalla and Iyengar [Science 283(1999) 381] have developed a hippocampal LTP model. In the conventional model, the concentration of protein phosphatase 2A (PP2A) was fixed. However, it was reported that dynamic inactivation of PP2A was essential for LTP [J. Neurochem. 74 (2000) 807]. We introduced a dynamic modeling of PP2A; inactivation (phosphorylation) of PP2A by calcium/calmodulin-dependent protein kinase II (CaMKII) in the presence of calcium/calmodulin, self-activation (autodephosphorylation) of PP2A, and inactivation (dephosphorylation) of CaMKII by PP2A. This model includes complex feedback loops; both CaMKII and PP2A are autoactivated, while they inactivate each other. Moreover, we proposed an analysis strategy for model validation by applying the results of sensitivity analysis. In our system, calcineurin (CaN) played an essential role, rather than the activation of protein kinase C (PKC) as documented in the conventional model. From results of the analysis of our model, we found the following robustness as characteristics of bistability in our model: (1) PP2A reactions against calcium ion (Ca2+) perturbation; (2) PP2A inactivation against PP2A increase; (3) protein phosphatase 1 (PP1) activation against PF2A increase; and (4) PP2A reactions against PP2A initial concentration. These properties facilitated LTP induction in our system. We showed that another mechanism could introduce bistable behavior by adding dynamic reactions of PP2A. © 2003 Elsevier Ltd. All rights reserved.

A genome-wide computational analysis of overlapping genes in various prokaryotic genomes was conducted to study the evolutionary rates of the formation and degradation of overlapping genes. Overlapping genes are adjacent genes that overlap partially or entirely. Overlapping gene pairs extracted from complete bacterial genome sequences were classified into three directional patterns, namely, 'convergent' (→←), 'unidirectional' (→→), and 'divergent' (←→). Divergent overlapping genes were rare compared to the other two patterns. We also compared the structures of overlapping genes that are not conserved in the homologous genes of different species. We found that overlapping genes are mainly generated from gene pairs with short intergenic regions. Overlapping genes caused by mutations at the 3′-end of genes occur more frequently than overlapping genes generated by mutations at the 5′-end of genes in closely related species. In contrast, gene overlaps due to mutations at the 5′-ends accumulate in distant species. These differences presumably correlate to the genomic divergence among species. The evolution of these overlapping gene structures is probably related to the evolutionary time scale, which suggests that the evolution of overlapping genes occurs at a universal mutation rate across species. © 2003 Elsevier B.V. All rights reserved.

Translation termination is catalyzed by release factors that recognize stop codons. However, previous works have shown that in some bacteria, the termination process also involves bases around stop codons. Recently, Ito et al, analyzed release factors and identified the amino acids therein that recognize stop codons. However, the amino acids that recognize bases around stop codons remain unclear. To identify the candidate amino acids that recognize the bases around stop codons, we aligned the protein sequences of the release factors of various bacteria and searched for amino acids that were conserved specifically in the sequence of bacteria that seemed to regulate translation termination by bases around stop codons. As a result, species having several highly conserved residues in RF1 and RF2 showed positive correlations between their codon usage bias and conservation of the bases around the stop codons. In addition, some of the residues were located very close to the SPF motif, which deciphers stop codons. These results suggest that these conserved amino acids enable the release factors to recognize the bases around the stop codons.

Summary: A computational tool for mitochondrial systems biology has been developed as a simulation model of E-Cell2, a publicly available simulation system. The general model consists of 58 enzymatic reactions and 117 metabolites, representing the respiratory chain, the TCA cycle, the fatty acid β-oxidation and the inner-membrane transport system. It is based on previously published enzyme kinetics studies in the literature; we have successfully integrated and packaged them into a single large model. The model can be easily extended and modified so that mitochondrial biologists/physiologists can integrate their own models and evaluate them in the context of the whole organelle metabolism. © Oxford University Press 2004; all rights reserved.

Shannon entropy from information theory has been applied to estimate the degree of deviation from equal usage of synonymous codons; however, previous attempts have failed to take into account all three aspects of amino acid usage, i.e. (i) the number of distinct amino acids, (ii) their relative frequencies, and (iii) their degree of codon degeneracy. A new index taking into account all of these aspects is proposed. The index, designated as the 'weighted sum of relative entropy' (Ew), is defined as the sum of the relative entropy of each amino acid weighted by its relative frequency in the sequence. In this paper, we demonstrate that Ew allows us to avoid some amino acid usage biases and can yield results contradictory to those obtained by previous methods. © 2004 Elsevier B.V. All rights reserved.