Published papers

The human skin surface harbors huge numbers of microbes. The skin microbiota interacts with its host and forms a skin microbiome profile that is specific for each individual. It has been reported that the skin microbiota that is left on an individual’s possessions can act as a sort of “fingerprint” and be used for owner identification. However, this approach needs to be improved to take into account any long-term instability of skin microbiota and contamination from nonspecific bacteria. Here, we took advantage of single-nucleotide polymorphisms (SNPs) in the 16S-encoding rRNA gene of Cutibacterium acnes, the most common and abundant bacterium on human skin, to perform owner identification. We first developed a high-throughput genotyping method based on next-generation sequencing to characterize the SNPs of the C. acnes 16S rRNA gene and found that the genotype composition of C. acnes 16S rRNA is individual specific. Owner identification accuracy of around 90% based on random forest machine learning was achieved by using a combination of C. acnes 16S rRNA genotype and skin microbiome profile data. Furthermore, our study showed that the C. acnes 16S rRNA genotype remained more stable over time than the skin microbiome profile. This characteristic of C. acnes was further confirmed by the analysis of publicly available human skin metagenome data. Our approach, with its high precision, good reproducibility, and low costs, thus provides new possibilities in the field of microbiome-based owner identification and forensics in general. IMPORTANCE Cutibacterium acnes is the most common and abundant bacterial species on human skin, and the gene that encodes its 16S rRNA has multiple single-nucleotide polymorphisms. In this study, we developed a method to efficiently determine the C. acnes 16S rRNA genotype composition from microbial samples taken from the hands of participants and from their possessions. Using the C. acnes 16S rRNA genotype composition, we could predict the owner of a possession with around 90% accuracy when the 16S rRNA gene-based microbiome profile was included. We also showed that the C. acnes 16S rRNA genotype composition was more stable over time than the skin microbiome profile and thus is more suitable for owner identification.

Lactoferrin (LF), a multifunctional glycoprotein found in mammalian milk, is reported to have immunoregulatory effects. The present study aimed to evaluate whether enteric-coated LF (eLF) could improve symptoms in patients with atopic keratoconjunctivitis (AKC). This randomized double-blind placebo-controlled single-center trial comprised Japanese patients (n = 20; aged 22–60 years) with AKC. Patients treated with 0.1% tacrolimus ophthalmic suspension (TALYMUS®) were administered eLF (400 mg/d of bovine LF) or placebo tablets for 12 weeks. Conjunctival injection was examined, papillae formation in the palpebral conjunctiva was evaluated, and corneal fluorescein score, itchy sensation in end-point itching scale, and serum allergic parameters were assessed. Conjunctival injection was significantly reduced in the LF group than in the placebo group (p = 0.0017, Mann–Whitney U-test). Papillae formation in the palpebral conjunctiva showed a statistical decrease in the LF group than in the placebo group (p = 0.010, unpaired T-test). LF combined with TALYMUS® could be a promising treatment strategy to mitigate AKC.

Background: We have recently reported that eugenol exerted indiscriminate cytotoxicity towards normal oral cells and oral squamous cell carcinoma (OSCC) cell lines without induction of apoptosis markers. In order to investigate the underlying mechanisms of cytotoxicity induction, we investigated the effect of short-term treatment with eugenol on the metabolic profiles of a human OSCC cell line (HSC-2). Materials and Methods: The viable cell number was determined by direct cell counting with a hemocytometer after trypsinization. After washing with 5% D-mannitol solution (found to retain the highest amounts of intracellular metabolites among several washing conditions), cellular metabolites were extracted with methanol with internal markers and then subjected to metabolomic analysis. Results: Cytotoxic concentrations of eugenol induced the reduction of ATP utilization (assessed by a significant reduction of the AMP/ATP and ADP/ATP ratio), of oxidative stress (assessed by the increase in oxidized form of glutathione, cysteine-glutathione disulfide and methionine sulfoxide), and an increase in the polyamines and glycolytic metabolites. Conclusion: The metabolic changes observed in this study suggest the induction of non-apoptotic cell death by eugenol.

In type 2 diabetes mellitus, glucose homeostasis is tightly maintained through insulin secretion and insulin sensitivity. Therefore, finding an accurate method to assess insulin secretion and sensitivity using clinically available data would enhance the quality of diabetic medical care. In an effort to find such a method, we developed a computational approach to derive indices of these factors using a 2-h oral glucose tolerance test (OGTT). To evaluate our method, clinical data from subjects who received an OGTT and a glucose clamp test were examined. Our insulin secretion index was significantly correlated with an analogous index obtained from a hyperglycemic clamp test (r = 0.90, n = 46, p < 0.001). Our insulin sensitivity index sensitivity was also significantly correlated with an analogous index obtained from a hyperinsulinemic-euglycemic clamp test (r = 0.56, n = 79, p < 0.001). These results suggest that our method can potentially provide an accurate and convenient tool toward improving the management of diabetes in clinical practice by assessing insulin secretion and insulin sensitivity. © 2011 The Author(s).

Remineralization of organic matter in deep-sea sediments is important in oceanic biogeochemical cycles, and bacteria play a major role in this process. Shewanella violacea DSS12 is a psychrophilic and piezophilic γ-proteobacterium that was isolated from the surface layer of deep sea sediment at a depth of 5110 m. Here, we report the complete genome sequence of S. violacea and comparative analysis with the genome of S. oneidensis MR-1, isolated from sediments of a freshwater lake. Unlike S. oneidensis, this deep-sea Shewanella possesses very few terminal reductases for anaerobic respiration and no c-type cytochromes or outer membrane proteins involved in respiratory Fe(iii) reduction, which is characteristic of most Shewanella species. Instead, the S. violacea genome contains more terminal oxidases for aerobic respiration and a much greater number of putative secreted proteases and polysaccharases, in particular, for hydrolysis of collagen, cellulose and chitin, than are encoded in S. oneidensis. Transporters and assimilatory reductases for nitrate and nitrite, and nitric oxide-detoxifying mechanisms (flavohemoglobin and flavorubredoxin) are found in S. violacea. Comparative analysis of the S. violacea genome revealed the respiratory adaptation of this bacterium to aerobiosis, leading to predominantly aerobic oxidation of organic matter in surface sediments, as well as its ability to efficiently use diverse organic matter and to assimilate inorganic nitrogen as a survival strategy in the nutrient-poor deep-sea floor. © The Royal Society of Chemistry 2010.

Many important cellular processes are regulated by reaction-diffusion (RD) of molecules that takes place both in the cytoplasm and on the membrane. To model and analyze such multicompartmental processes, we developed a lattice-based Monte Carlo method, Spatiocyte that supports RD in volume and surface compartments at single molecule resolution. Stochasticity in RD and the excluded volume effect brought by intracellular molecular crowding, both of which can significantly affect RD and thus, cellular processes, are also supported. We verified the method by comparing simulation results of diffusion, irreversible and reversible reactions with the predicted analytical and best available numerical solutions. Moreover, to directly compare the localization patterns of molecules in fluorescence microscopy images with simulation, we devised a visualization method that mimics the microphotography process by showing the trajectory of simulated molecules averaged according to the camera exposure time. In the rod-shaped bacterium Escherichia coli, the division site is suppressed at the cell poles by periodic pole-to-pole oscillations of the Min proteins (MinC, MinD and MinE) arising from carefully orchestrated RD in both cytoplasm and membrane compartments. Using Spatiocyte we could model and reproduce the in vivo MinDE localization dynamics by accounting for the previously reported properties of MinE. Our results suggest that the MinE ring, which is essential in preventing polar septation, is largely composed of MinE that is transiently attached to the membrane independently after recruited by MinD. Overall, Spatiocyte allows simulation and visualization of complex spatial and reaction-diffusion mediated cellular processes in volumes and surfaces. As we showed, it can potentially provide mechanistic insights otherwise difficult to obtain experimentally. © 2009 The Author(s).

Saliva is a readily accessible and informative biofluid, making it ideal for the early detection of a wide range of diseases including cardiovascular, renal, and autoimmune diseases, viral and bacterial infections and, importantly, cancers. Saliva-based diagnostics, particularly those based on metabolomics technology, are emerging and offer a promising clinical strategy, characterizing the association between salivary analytes and a particular disease. Here, we conducted a comprehensive metabolite analysis of saliva samples obtained from 215 individuals (69 oral, 18 pancreatic and 30 breast cancer patients, 11 periodontal disease patients and 87 healthy controls) using capillary electrophoresis time-of-flight mass spectrometry (CE-TOF-MS). We identified 57 principal metabolites that can be used to accurately predict the probability of being affected by each individual disease. Although small but significant correlations were found between the known patient characteristics and the quantified metabolites, the profiles manifested relatively higher concentrations of most of the metabolites detected in all three cancers in comparison with those in people with periodontal disease and control subjects. This suggests that cancer-specific signatures are embedded in saliva metabolites. Multiple logistic regression models yielded high area under the receiver-operating characteristic curves (AUCs) to discriminate healthy controls from each disease. The AUCs were 0.865 for oral cancer, 0.973 for breast cancer, 0.993 for pancreatic cancer, and 0.969 for periodontal diseases. The accuracy of the models was also high, with cross-validation AUCs of 0.810, 0.881, 0.994, and 0.954, respectively. Quantitative information for these 57 metabolites and their combinations enable us to predict disease susceptibility. These metabolites are promising biomarkers for medical screening. © The Author(s) 2009.

Chaos Game Representation (CGR) is a generalized scale-independent Markov transition table, which is useful for the visualization and comparative study of genomic signature, or for the study of characteristic sequence motifs. However, in order to fully utilize the scale-independent properties of CGR, it should be accessible through scale-independent user interface instead of static images. Here we describe a web server and Perl library for generating zoomable CGR images utilizing Google Maps API, which is also easily searchable for specific motifs. The web server is freely accessible at http://www.g-language.org/wiki/cgr/, and the Perl library as well as the source code is distributed with the G-language Genome Analysis Environment under GNU General Public License. © 2009 Arakawa et al; licensee BioMed Central Ltd.

We have developed a simple and unbiased method for membrane proteome analysis using cyanocysteine-mediated cleavage in combination with trypsin digestion. In our previous study, application of the trypsin-based phase-transfer surfactants (PTS) protocol for membrane proteome analysis provided a substantial improvement in the identification of the membrane proteome, but the task remains challenging, because trypsin often generates peptides larger than the observable m/z range. Here, we predict computationally that the combination of Cys cleavage with tryptic digestion would be more effective than trypsin digestion alone for membrane proteome analysis. To validate this prediction, we applied a combined Cys cleavage-trypsin approach to 14 μg of Escherichia coli membrane-enriched pellet. By using two-dimensional LC-MS/MS, we identified a total of 1530 proteins, of which 667 were membrane proteins. This represents a 10% increase over the number identified with the PTS protocol using optimized trypsin-based digestion in our previous study [Masuda, T., et al. J. Proteome Res. 2008, 7 (2), 731-40]. The coverage of the E. coli membrane proteome was approximately 40%, ranging from 37% to 42% in various subcategories. Further, the distribution of the number of transmembrane domains per protein was unbiased compared with that in the GenoBase database. These results indicate that the combination Cys chemical cleavage-assisted trypsin digestion protocol will be a powerful tool for membrane proteome analysis. © 2009 American Chemical Society.

Motivation: Liquid chromatography-tandem mass spectrometry (LC-MS/MS) is a powerful tool in proteomics studies, but when peptide retention information is used for identification purposes, it remains challenging to compare multiple LC-MS/MS runs or to match observed and predicted retention times, because small changes of LC conditions unavoidably lead to variability in retention times. In addition, non-contiguous retention data obtained with different LC-MS instruments or in different laboratories must be aligned to confirm and utilize rapidly accumulating published proteomics data. Results: We have developed a new alignment method for peptide retention times based on linear solvent strength (LSS) theory. We found that log k0 (logarithm of retention factor for a given organic solvent) in the LSS theory can be utilized as a 'universal' retention index of peptides (RIP) that is independent of LC gradients, and depends solely on the constituents of the mobile phase and the stationary phases. We introduced a machine learning-based scheme to optimize the conversion function of gradient retention times (tg) to log k0. Using the optimized function, tg values obtained with different LC-MS systems can be directly compared with each other on the RIP scale. In an examination of Arabidopsis proteomic data, the vast majority of retention time variability was removed, and five datasets obtained with various LC-MS systems were successfully aligned on the RIP scale. © The Author 2008. Published by Oxford University Press. All rights reserved.

Background: Clostridium difficile is a Gram-positive anaerobe and the leading cause of antibiotic-associated diarrhea worldwide. The emergence of ribotype 027 (RT027) strains is associated with increased incidence of infection and mortality. To further understand the relationship between C. difficile NCKUH-21, a RT027 strain isolated from a patient in Taiwan, and other RT027 strains, we performed whole-genome shotgun sequencing on NCKUH-21 and comparative genomic analyses. Results: The genome size, G+C content, and gene number for the NCKUH-21 strain were determined to be similar to those for other C. difficile strains. The core genome phylogeny indicated that the five RT027 strains R20291, CD196, NCKUH-21, BI1, and 2007855 formed a clade. A pathogenicity locus, tcdR-tcdB-tcdE-orf-tcdA-tcdC, was conserved in the genome. A genomic region highly similar to the Clostridium phage $$\upvarphi$$ φ CD38-2 was present in the NCKUH-21 strain but absent in the other RT027 strains and designated as the prophage $$\upvarphi$$ φ NCKUH-21. The prophage $$\upvarphi$$ φ NCKUH-21 genes were significantly higher in G+C content than the other genes in the NCKUH-21 genome, indicating that the prophage does not match the base composition of the host genome. Conclusions: This is the first whole-genome analysis of a RT027 C. difficile strain isolated from Taiwan. Due to the high identity with $$\upvarphi$$ φ CD38-2, the prophage identified in the NCKUH-21 genome has the potential to regulate toxin production. These results provide important information for understanding the pathogenicity of RT027 C. difficile in Taiwan.

As the physical properties and functionality of dipeptides differ from those of amino acids, they have attracted attention in metabolomics; however, their functions in vivo have not been clarified in detail. Hepatocellular carcinoma (HCC) is the most common type of primary liver cancer, and its major cause is chronic hepatitis. This study was conducted to explore tumor-specific dipeptide characteristics by performing comprehensive dipeptide analysis in the tumor and surrounding nontumor tissue of patients with HCC. Dipeptides were analyzed by liquid chromatography tandem mass spectrometry and capillary electrophoresis tandem mass spectrometry. Principal component analysis using 236 detected dipeptides showed differences in the dipeptide profiles between nontumor and tumor tissues; however, no clear difference was observed in etiological comparison. In addition, the N-and C-terminal amino acid compositions of the detected dipeptides significantly differed, suggesting the substrate specificity of enzyme proteins, such as peptidase. Furthermore, hepatitis-derived HCC may show a characteristic dipeptide profile even before tumor formation. These results provide insight into HCC pathogenesis and may help identify novel biomarkers for diagnosis.

Commensal microbiota colonize the surface of our bodies. The inside of the gastrointestinal tract is one such surface that provides a habitat for them. The gastrointestinal tract is a long organ system comprising of various parts, and each part possesses various functions. It has been reported that the composition of intestinal luminal metabolites between the small and large intestine are different; however, comprehensive metabolomic and commensal microbiota profiles specific to each part of the gastrointestinal lumen remain obscure. In this study, by using capillary electrophoresis time-of-flight mass spectrometry (CE-TOFMS)-based metabolome and 16S rRNA gene-based microbiome analyses of specific pathogen-free (SPF) and germ-free (GF) murine gastrointestinal luminal profiles, we observed the different roles of commensal microbiota in each part of the gastrointestinal tract involved in carbohydrate metabolism and nutrient production. We found that the concentrations of most amino acids in the SPF small intestine were higher than those in the GF small intestine. Furthermore, sugar alcohols such as mannitol and sorbitol accumulated only in the GF large intestine, but not in the SPF large intestine. On the other hand, pentoses, such as arabinose and xylose, gradually accumulated from the cecum to the colon only in SPF mice, but were undetected in GF mice. Correlation network analysis between the gastrointestinal microbes and metabolites showed that niacin metabolism might be correlated to Methylobacteriaceae. Collectively, commensal microbiota partially affects the gastrointestinal luminal metabolite composition based on their metabolic dynamics, in cooperation with host digestion and absorption.

Metastasis is responsible for approximately 90% of cancer-associated mortality and proceeds through multiple steps. Several herbal medicines are reported to inhibit primary tumor growth, but the suppressor effects of the medicines on metastasis progression are still not fully elucidated. Here we report that cinnamon bark extract (CBE) has a suppressor effect on metastatic dissemination of cancer cells. Through a phenotypic screening using zebrafish embryos, CBE was identified to interfere with the gastrulation progression of zebrafish embryos, of which the molecular mechanisms are conserved in metastasis progression. A Boyden chamber assay showed that CBE decreased cell motility and invasion of MDA-MB-231 human breast cancer cells without affecting their cell viability. Furthermore, CBE suppressed metastatic dissemination of the cells in a zebrafish xenotransplantation model. Quantitative metabolome analyses revealed that the productions of glucose-6-phosphate (G6P) and fructose 6-phosphate which are intermediate metabolites of glycolytic metabolism were interrupted in CBE-treated cells. qPCR and western-blotting analyses revealed that CBE-treated cells showed decreased expression of hexokinase 2 (HK2) which yields G6P. Pharmacological inhibition of HK2 with 2-deoxy-D-glucose suppressed cell invasion and migration of the cells without affecting their cell viability. Taken together, CBE suppresses metastatic dissemination of cancer cells through inhibition of glycolysis metabolism.

The increasing number of patients with osteoporosis and sarcopenia is a global concern among countries with progressively aging societies. The high medical costs of treating those patients suggest that prevention rather than treatment is preferable. We enrolled 729 subjects who attended both the second and third surveys of the Research on Osteoarthritis/Osteoporosis Against Disability (ROAD) study. Blood samples were collected from subjects at the second survey, and then a comprehensive metabolomic analysis was performed. It was found that 35 had newly developed osteoporosis at the third survey performed four years later, and 39 were newly diagnosed with sarcopenia at the third survey. In the second survey, we found that serum Gly levels were significantly higher even after adjustment for age, sex, and BMI in subjects with newly developed osteoporosis relative to those who remained osteoporosis-negative during the four-year follow-up. We also show that serum taurine levels were significantly lower at the second survey, even after adjustment for age, sex, and BMI in subjects with newly developed sarcopenia during the four-year follow-up compared with those not diagnosed with sarcopenia at the second or third surveys. Though our sample size and odds ratios were small, increased Gly and decreased taurine levels were found to be predictive of new development of osteoporosis and sarcopenia, respectively, within four years.

While most ant species establish a colony independently, some socially parasitic ants build the foundation of their colony by invading other ant (host) colonies and utilizing their labor forces. Many socially parasitic ants disguise their cuticular hydrocarbon (CHC) profile, which is also known as signature mixture for nestmate discrimination, when invading the host colony. Since the strategy of chemical disguise is widespread in socially parasitic ants, elucidating the mechanism of chemical disguise will promote knowledge on the evolutionary history of social parasitism. However, detailed knowledge is still lacking, as the relevant information has only originated from circumstantial evidence, which was obtained from ecological observations. In this study, we investigated the mechanism of chemical disguise in a new queen of a temporary socially parasitic spiny ant (Polyrhachis lamellidens) by measuring its CHC profile, performing a tracing assay with labeled substances, and analyzing gene expression levels. First, after rubbing behavior was observed against the host workers, the CHC profile in P. lamellidens shifted to pronounced peaks that closely resembling that of the host workers. We also observed a reduction in aggressive behaviors by the host ant against P. lamellidens after rubbing behavior was performed. In addition, P. lamellidens acquired artificially-applied labeling substances from host workers through their rubbing behaviors, while gene expression profiling showed the expression of CHC synthesis-related genes did not change during this behavior. These results suggest that P. lamellidens directly obtains host CHCs through rubbing behavior, and these host CHCs enables P. lamellidens to remain disguised during colony invasion.

Background: Comprehensive metabolomic analyses have been conducted in various institutes and a large amount of metabolomic data are now publicly available. To help fully exploit such data and facilitate their interpretation, metabolomic data obtained from different facilities and different samples should be integrated and compared. However, large-scale integration of such data for biological discovery is challenging given that they are obtained from various types of sample at different facilities and by different measurement techniques, and the target metabolites and sensitivities to detect them also differ from study to study. Results: We developed iDMET, a network-based approach to integrate metabolomic data from different studies based on the differential metabolomic profiles between two groups, instead of the metabolite profiles themselves. As an application, we collected cancer metabolomic data from 27 previously published studies and integrated them using iDMET. A pair of metabolomic changes observed in the same disease from two studies were successfully connected in the network, and a new association between two drugs that may have similar effects on the metabolic reactions was discovered. Conclusions: We believe that iDMET is an efficient tool for integrating heterogeneous metabolomic data and discovering novel relationships between biological phenomena.