Prediction of transcription factors associated with DNA demethylation during human cellular development.

DNA methylation of CpG dinucleotides is an important epigenetic modification involved in the regulation of mammalian gene expression, with each type of cell developing a specific methylation profile during its differentiation. Recently, it has been shown that a small subgroup of transcription factors (TFs) might promote DNA demethylation at their binding sites. We developed a bioinformatics pipeline to predict from genome-wide DNA methylation data TFs that promote DNA demethylation at their binding site. We applied the pipeline to International Human Epigenome Consortium methylome data and selected 393 candidate transcription factor binding motifs and associated 383 TFs that are likely associated with DNA demethylation. Validation of a subset of the candidate TFs using an in vitro assay suggested that 28 of 49 TFs from various TF families had DNA-demethylation-promoting activity; TF families, such as bHLH and ETS, contained both TFs with and without the activity. The identified TFs showed large demethylated/methylated CpG ratios and their demethylated CpGs showed significant bias toward hypermethylation in original cells. Furthermore, the identified TFs promoted demethylation of distinct sets of CpGs, with slight overlap of the targeted CpGs among TF family members, which was consistent with the results of a gene ontology (GO) term analysis of the identified TFs. Gene expression analysis of the identified TFs revealed that multiple TFs from various families are specifically expressed in human cells and tissues. Together, our results suggest that a large number of TFs from various TF families are associated with cell-type-specific DNA demethylation during human cellular development.

FANTOM enters 20th year: expansion of transcriptomic atlases and functional annotation of non-coding RNAs.

The Functional ANnoTation Of the Mammalian genome (FANTOM) Consortium has continued to provide extensive resources in the pursuit of understanding the transcriptome, and transcriptional regulation, of mammalian genomes for the last 20 years. To share these resources with the research community, the FANTOM web-interfaces and databases are being regularly updated, enhanced and expanded with new data types. In recent years, the FANTOM Consortium's efforts have been mainly focused on creating new non-coding RNA datasets and resources. The existing FANTOM5 human and mouse miRNA atlas was supplemented with rat, dog, and chicken datasets. The sixth (latest) edition of the FANTOM project was launched to assess the function of human long non-coding RNAs (lncRNAs). From its creation until 2020, FANTOM6 has contributed to the research community a large dataset generated from the knock-down of 285 lncRNAs in human dermal fibroblasts; this is followed with extensive expression profiling and cellular phenotyping. Other updates to the FANTOM resource includes the reprocessing of the miRNA and promoter atlases of human, mouse and chicken with the latest reference genome assemblies. To facilitate the use and accessibility of all above resources we further enhanced FANTOM data viewers and web interfaces. The updated FANTOM web resource is publicly available at https://fantom.gsc.riken.jp/.

Gestational weight gain mediates the effects of energy intake on birth weight among singleton pregnancies in the Japan Environment and Children's Study.

BACKGROUND: Extra energy intake is commonly recommended for pregnant women to support fetal growth. However, relevant data regarding variations in energy intake and expenditure, body mass index and gestational weight gain (GWG) are frequently not considered. This study aimed to investigate how energy intake during pregnancy and gestational weight gain (GWG) are associated with birth weight. METHODS: Early pregnant women were recruited into a Japanese nationwide prospective birth cohort study between 2011 and 2014. We analysed data of 89,817 mother-child pairs of live-born non-anomalous singletons after excluding births before 28 weeks or after 42 weeks. Energy intake during pregnancy was estimated from self-administered food frequency questionnaires (FFQ) and was stratified into low, medium, and high. Participants completed the FFQ in mid-pregnancy (mean 27.9 weeks) by recalling food consumption at the beginning of pregnancy. Effects of energy intake on birth weight and mediation by GWG were estimated using the Karlson-Holm-Breen method; the method separates the impact of confounding in the comparison of conditional and unconditional parameter estimates in nonlinear probability models. Relative risks and risk differences for abnormal birth size were calculated. RESULTS: Mean daily energy intake, GWG, and birth weight were 1682.1 (533.6) kcal, 10.3 (4.0) kg, and 3032.3 (401.4) g, respectively. 6767 and 9010 women had small-for-gestational-age and large-for-gestational-age infants, respectively. Relative to low energy intake, moderate and high intakes increased adjusted birth weights by 13 g and 24 g, respectively: 58 and 69% of these effects, respectively, were mediated by GWG. Compared with the moderate energy intake group, the low energy intake group had seven more women per 1000 women with a small-for-gestational-age birth, whereas the high energy intake group had eight more women per 1000 women with a large-for-gestational-age birth. CONCLUSION: GWG mediates the effect of energy intake on birth weight. All pregnant women should be given adequate nutritional guidance for optimal GWG and fetal growth.

Update of the FANTOM web resource: expansion to provide additional transcriptome atlases.

The FANTOM web resource (http://fantom.gsc.riken.jp/) was developed to provide easy access to the data produced by the FANTOM project. It contains the most complete and comprehensive sets of actively transcribed enhancers and promoters in the human and mouse genomes. We determined the transcription activities of these regulatory elements by CAGE (Cap Analysis of Gene Expression) for both steady and dynamic cellular states in all major and some rare cell types, consecutive stages of differentiation and responses to stimuli. We have expanded the resource by employing different assays, such as RNA-seq, short RNA-seq and a paired-end protocol for CAGE (CAGEscan), to provide new angles to study the transcriptome. That yielded additional atlases of long noncoding RNAs, miRNAs and their promoters. We have also expanded the CAGE analysis to cover rat, dog, chicken, and macaque species for a limited number of cell types. The CAGE data obtained from human and mouse were reprocessed to make them available on the latest genome assemblies. Here, we report the recent updates of both data and interfaces in the FANTOM web resource.

SCPortalen: human and mouse single-cell centric database.

Published single-cell datasets are rich resources for investigators who want to address questions not originally asked by the creators of the datasets. The single-cell datasets might be obtained by different protocols and diverse analysis strategies. The main challenge in utilizing such single-cell data is how we can make the various large-scale datasets to be comparable and reusable in a different context. To challenge this issue, we developed the single-cell centric database 'SCPortalen' (http://single-cell.clst.riken.jp/). The current version of the database covers human and mouse single-cell transcriptomics datasets that are publicly available from the INSDC sites. The original metadata was manually curated and single-cell samples were annotated with standard ontology terms. Following that, common quality assessment procedures were conducted to check the quality of the raw sequence. Furthermore, primary data processing of the raw data followed by advanced analyses and interpretation have been performed from scratch using our pipeline. In addition to the transcriptomics data, SCPortalen provides access to single-cell image files whenever available. The target users of SCPortalen are all researchers interested in specific cell types or population heterogeneity. Through the web interface of SCPortalen users are easily able to search, explore and download the single-cell datasets of their interests.

High fasting insulin concentrations may be a pivotal predictor for the severity of hepatic fibrosis beyond the glycemic status in non-alcoholic fatty liver disease patients before development of diabetes mellitus.

BACKGROUND: Insulin resistance and type 2 diabetes mellitus (T2DM) contribute to the progression of non-alcoholic fatty liver disease (NAFLD). However, the relationship between glucose metabolic factors and the histological severity in NAFLD patients before development of T2DM is not well known. METHODS: In 103 biopsy-proven NAFLD patients (68 men and 35 women) with hemoglobin A1c of <6.5% and fasting blood glucose of <126 mg/dL, we investigated whether glucose metabolic factors influenced the severity of hepatic fibrosis without prior known T2DM. RESULTS: Female gender, age, serum aspartate aminotransferase, the aspartate aminotransferase/alanine aminotransferase ratio, fasting immunoreactive insulin (f-IRI), homeostasis model assessment - insulin resistance, hemoglobin A1c, hyaluronic acid, and type IV collagen 7 s were significantly higher, and 1,5-anhydroglucitol was significantly lower, in the fibrosis stage F3 group than in the F0-2 group. Multiple logistic regression analysis showed that only f-IRI (P = 0.006; odds ratio, 1.15151; 95% confidence interval, 1.04198-1.27254) was significantly indicated as a predictive factor for F3. As determined by both forward and backward stepwise selection analyses to optimize the model, f-IRI (P = 0001; odds ratio, 1.16788) remained an independent predictive factor for F3. To discriminate the F3 group from the F0-2 group, the area under the receiver operating characteristic curves showed that fasting insulin was 0.7219, and the best cut-off value of f-IRI was 13.2 µU/mL in the receiver operating characteristic curve analysis. CONCLUSIONS: High fasting insulin concentrations may be a pivotal glucose metabolic predictor for the severity of hepatic fibrosis beyond the glycemic status in NAFLD patients before development of T2DM.

Fibrotic gene expression coexists with alveolar proteinosis in early indium lung.

Occupational inhalation of indium compounds can cause the so-called "indium lung disease". Most affected individuals show pulmonary alveolar proteinosis (PAP) and fibrotic interstitial lung disease. In animal experiments, inhalation of indium tin oxide or indium oxide has been shown to cause lung damage. However, the mechanisms by which indium compounds lead to indium lung disease remain unknown. In this study, we constructed a mouse model of indium lung disease and analyzed gene expression in response to indium exposure. Indium oxide (In2O3, 10 mg/kg, primary particle size <100 nm) was administered intratracheally to C57BL/6 mice (male, 8 weeks of age) twice a week for 8 weeks. Four weeks after the final instillation, histopathological analysis exhibited periodic acid-Schiff positive material in the alveoli, characteristic of PAP. Comprehensive gene expression analysis by RNA-Seq, however, revealed expression of fibrosis-related genes, such as surfactant associated protein D, surfactant associated protein A1, mucin 1, and collagen type I and III, was significantly increased, indicating that fibrotic gene expression progresses in early phase of indium lung. These data supported the latest hypothesis that PAP occurs as an acute phase response and is replaced by fibrosis after long-term latency.

Regulation of Gene Expression by Sodium Valproate in Epithelial-to-Mesenchymal Transition.

PURPOSE: Epithelial-to-mesenchymal transition (EMT) is an important mechanism in cancer metastasis and pulmonary fibrosis. Previous studies demonstrated effect of histone H3 and H4 acetylation in cancer and pulmonary fibrosis, so we hypothesized that histone modification might play a crucial role in gene regulation during EMT. In this study, we investigated the mechanism behind EMT by analyzing comprehensive gene expression and the effect of sodium valproate (VPA), a class I histone deacetylase inhibitory drug, on histone modification. METHODS: EMT was induced in human alveolar epithelial cells (A549) using 5 ng/mL of transforming growth factor (TGF)-ß1. Various concentrations of VPA were then administered, and Western blotting was used to analyze histone acetylation or methylation. Comprehensive gene expression analysis was carried out by RNA sequencing, and chromatin immunoprecipitation was performed with an anti-acetyl histone H3 lysine 27 antibody. RESULTS: TGF-ß1 stimulation led to a decrease in histone acetylation, especially that of histone H3K27, and H3K27ac localization was decreased at particular gene loci. This decrease was recovered by VPA treatment, which also up-regulated the mRNA expression of genes down-regulated by TGF-ß1, and correlated with the localization of H3K27ac. However, genes up-regulated by TGF-ß1 stimulation were not suppressed by VPA, with the exception of COL1A1. CONCLUSIONS: Histone acetylation was down-regulated by TGF-ß1 stimulation in A549 cells. VPA partially inhibited EMT and the decrease of histone acetylation, which plays an important role in the progression of EMT.

Global analysis for functional residues of histone variant Htz1 using the comprehensive point mutant library.

Histone variants perform unique functions and are deposited onto DNA by mechanisms distinct from those of canonical histones. The H2A variant, H2A.Z, also known as Htz1 in Saccharomyces cerevisiae, is not uniformly distributed across the genome but facilitates transcriptional activation at target gene promoters and anti-silencing at heterochromatin loci. Htz1 is also involved in DNA replication, DNA repair, chromosome segregation and cell cycle control. Its sequence identity to canonical H2A is only ∼60%, and it is likely that the nonconserved residues are responsible for Htz1-specific functions. However, precise roles of these variant-specific residues are not well understood. To gain insights into the molecular basis underlying the functional differences between canonical and variant histones, 117 alanine-scanning point mutants of Htz1 were constructed for this study, and chemical genetic screens were carried out. Consequently, seven Htz1 residues that conferred one or more abnormal phenotypes when mutated were identified. Based on primary sequence and functional conservation between H2A and Htz1, two of these residues (F32 and I109) appear to have an Htz1-specific role, whereas the rest seem to have functions shared between H2A and Htz1. This study provides a useful resource for future investigations into functional convergence and divergence between canonical and variant histones.

SkewC: Identifying cells with skewed gene body coverage in single-cell RNA sequencing data.

The analysis and interpretation of single-cell RNA sequencing (scRNA-seq) experiments are compromised by the presence of poor-quality cells. For meaningful analyses, such poor-quality cells should be excluded as they introduce noise in the data. We introduce SkewC, a quality-assessment tool, to identify skewed cells in scRNA-seq experiments. The tool's methodology is based on the assessment of gene coverage for each cell, and its skewness as a quality measure; the gene body coverage is a unique characteristic for each protocol, and different protocols yield highly different coverage profiles. This tool is designed to avoid misclustering or false clusters by identifying, isolating, and removing cells with skewed gene body coverage profiles. SkewC is capable of processing any type of scRNA-seq dataset, regardless of the protocol. We envision SkewC as a distinctive QC method to be incorporated into scRNA-seq QC processing to preclude the possibility of scRNA-seq data misinterpretation.

Global analysis of functional relationships between histone point mutations and the effects of histone deacetylase inhibitors.

Comprehensive analyses of the histone-GLibrary in previous studies showed that most mutants of modification sites in the histone core regions show phenotypes, whereas those with modifications in the histone N-terminal unstructured tail regions (N-tails) do not. One possible reason is that modifications in N-tails are linked to each other to form a scale-free network termed histone 'modification web'. In the network, the compensatory pathways are created to acquire the robustness against the any defects. Because of this robustness, it is difficult to determine the significance of the individual histone modifications in N-tails in vivo. To overcome this problem, we used a strategy using drugs coordinately to inhibit modification enzymes and observed the mutant phenotypes when the compensatory pathways are largely interrupted. We analyzed histone-GLibrary using inhibitors of histone deacetylases (HDACs) and identified novel phenotypic mutants. We also examined the phenotypic changes through the combined use of an HDAC inhibitor and an inhibitor of DNA-mediated reactions. Mutation of modifiable sites H3-K4 and H4-K16 in histone N-tails, which are presumed to be the 'hubs' of the network, resulted in identifiable phenotypes. The data obtained provide valuable information for speculation on novel relationships between histone modification in N-tails and biological function and for predicting unknown modification sites in core histones.

Discovery of widespread transcription initiation at microsatellites predictable by sequence-based deep neural network.

Using the Cap Analysis of Gene Expression (CAGE) technology, the FANTOM5 consortium provided one of the most comprehensive maps of transcription start sites (TSSs) in several species. Strikingly, ~72% of them could not be assigned to a specific gene and initiate at unconventional regions, outside promoters or enhancers. Here, we probe these unassigned TSSs and show that, in all species studied, a significant fraction of CAGE peaks initiate at microsatellites, also called short tandem repeats (STRs). To confirm this transcription, we develop Cap Trap RNA-seq, a technology which combines cap trapping and long read MinION sequencing. We train sequence-based deep learning models able to predict CAGE signal at STRs with high accuracy. These models unveil the importance of STR surrounding sequences not only to distinguish STR classes, but also to predict the level of transcription initiation. Importantly, genetic variants linked to human diseases are preferentially found at STRs with high transcription initiation level, supporting the biological and clinical relevance of transcription initiation at STRs. Together, our results extend the repertoire of non-coding transcription associated with DNA tandem repeats and complexify STR polymorphism.

Global analysis of mutual interaction surfaces of nucleosomes with comprehensive point mutants.

The surfaces of core histones in nucleosome are exposed as required for factor recognition, or buried for histone-DNA and histone-histone interactions. To understand the mechanisms by which nucleosome structure and function are coordinately altered in DNA-mediated reactions, it is essential to define the roles of both exposed and buried residues and their functional relationships. For this purpose, we developed GLASP (GLobal Analysis of Surfaces by Point mutation) and GLAMP (GLobal Analysis of Mutual interaction surfaces of multi-subunit protein complex by Point mutation) strategies, both of which are comprehensive analyses by point mutagenesis of exposed and buried residues in nucleosome, respectively. Four distinct DNA-mediated reactions evaluated by Ty suppression (the Spt(-) phenotype), and sensitivities to 6-azauracil (6AU), hydroxyurea (HU), and methyl methanesulfonate (MMS), require common and different GLAMP residues. Mutated GLAMP residues at the interface between histones H2A and H2B mainly affect the Spt(-) phenotype but not HU and MMS sensitivities. Interestingly, among the mutated GLAMP residues surrounding the histone H3-H3' interface, some equally affect the Spt(-) phenotype, and HU and MMS sensitivities, whereas others differentially affect the Spt(-) phenotype, and HU and MMS sensitivities. Based on these and other results, the functional relationships among chromatin factors and GLASP and GLAMP residues provide insights into nucleosome disassembly/assembly processes in DNA-mediated reactions.

RADICL-seq identifies general and cell type-specific principles of genome-wide RNA-chromatin interactions.

Mammalian genomes encode tens of thousands of noncoding RNAs. Most noncoding transcripts exhibit nuclear localization and several have been shown to play a role in the regulation of gene expression and chromatin remodeling. To investigate the function of such RNAs, methods to massively map the genomic interacting sites of multiple transcripts have been developed; however, these methods have some limitations. Here, we introduce RNA And DNA Interacting Complexes Ligated and sequenced (RADICL-seq), a technology that maps genome-wide RNA-chromatin interactions in intact nuclei. RADICL-seq is a proximity ligation-based methodology that reduces the bias for nascent transcription, while increasing genomic coverage and unique mapping rate efficiency compared with existing methods. RADICL-seq identifies distinct patterns of genome occupancy for different classes of transcripts as well as cell type-specific RNA-chromatin interactions, and highlights the role of transcription in the establishment of chromatin structure.

Combinatorial perturbation analysis reveals divergent regulations of mesenchymal genes during epithelial-to-mesenchymal transition.

Epithelial-to-mesenchymal transition (EMT), a fundamental transdifferentiation process in development, produces diverse phenotypes in different physiological or pathological conditions. Many genes involved in EMT have been identified to date, but mechanisms contributing to the phenotypic diversity and those governing the coupling between the dynamics of epithelial (E) genes and that of the mesenchymal (M) genes are unclear. In this study, we employed combinatorial perturbations to mammary epithelial cells to induce a series of EMT phenotypes by manipulating two essential EMT-inducing elements, namely TGF-ß and ZEB1. By measuring transcriptional changes in more than 700 E-genes and M-genes, we discovered that the M-genes exhibit a significant diversity in their dependency to these regulatory elements and identified three groups of M-genes that are controlled by different regulatory circuits. Notably, functional differences were detected among the M-gene clusters in motility regulation and in survival of breast cancer patients. We computationally predicted and experimentally confirmed that the reciprocity and reversibility of EMT are jointly regulated by ZEB1. Our integrative analysis reveals the key roles of ZEB1 in coordinating the dynamics of a large number of genes during EMT, and it provides new insights into the mechanisms for the diversity of EMT phenotypes.

refTSS: A Reference Data Set for Human and Mouse Transcription Start Sites.

Transcription starts at genomic positions called transcription start sites (TSSs), producing RNAs, and is mainly regulated by genomic elements and transcription factors binding around these TSSs. This indicates that TSSs may be a better unit to integrate various data sources related to transcriptional events, including regulation and production of RNAs. However, although several TSS datasets and promoter atlases are available, a comprehensive reference set that integrates all known TSSs is lacking. Thus, we constructed a reference dataset of TSSs (refTSS) for the human and mouse genomes by collecting publicly available TSS annotations and promoter resources, such as FANTOM5, DBTSS, EPDnew, and ENCODE. The data set consists of genomic coordinates of TSS peaks, their gene annotations, quality check results, and conservation between human and mouse. We also developed a web interface to browse the refTSS (http://reftss.clst.riken.jp/). Users can access the resource for collecting and integrating data and information about transcriptional regulation and transcription products.

OVOL2 induces mesenchymal-to-epithelial transition in fibroblasts and enhances cell-state reprogramming towards epithelial lineages.

Mesenchymal-to-epithelial transition (MET) is an important step in cell reprogramming from fibroblasts (a cell type frequently used for this purpose) to various epithelial cell types. However, the mechanism underlying MET induction in fibroblasts remains to be understood. The present study aimed to identify the transcription factors (TFs) that efficiently induce MET in dermal fibroblasts. OVOL2 was identified as a potent inducer of key epithelial genes, and OVOL2 cooperatively enhanced MET induced by HNF1A, TP63, and KLF4, which are known reprogramming TFs to epithelial lineages. In TP63/KLF4-induced keratinocyte-like cell-state reprogramming, OVOL2 greatly facilitated the activation of epithelial and keratinocyte-specific genes. This was accompanied by enhanced changes in chromatin accessibility across the genome. Mechanistically, motif enrichment analysis revealed that the target loci of KLF4 and TP63 become accessible upon induction of TFs, whereas the OVOL2 target loci become inaccessible. This indicates that KLF4 and TP63 positively regulate keratinocyte-associated genes whereas OVOL2 suppresses fibroblast-associated genes. The exogenous expression of OVOL2 therefore disrupts fibroblast lineage identity and facilitates fibroblast cell reprogramming into epithelial lineages cooperatively with tissue-specific reprogramming factors. Identification of OVOL2 as an MET inducer and an epithelial reprogramming enhancer in fibroblasts provides new insights into cellular reprogramming improvement for future applications.

The FANTOM5 Computation Ecosystem: Genomic Information Hub for Promoters and Active Enhancers.

The Functional Annotation of the Mammalian Genome 5 (FANTOM5) project conducted transcriptome analysis of various mammalian cell types and provided a comprehensive resource to understand transcriptome and transcriptional regulation in individual cellular states encoded in the genome.FANTOM5 used cap analysis of gene expression (CAGE) with single-molecule sequencing to map transcription start sites (TSS) and measured their expression in a diverse range of samples. The main results from FANTOM5 were published as a promoter-level mammalian expression atlas and an atlas of active enhancers across human cell types. The FANTOM5 dataset is composed of raw experimental data and the results of bioinformatics analyses. In this chapter, we give a detailed description of the content of the FANTOM5 dataset and elaborate on different computing applications developed to publish the data and enable reproducibility and discovery of new findings. We present use cases in which the FANTOM5 dataset has been reused, leading to new findings.

FANTOM5 CAGE profiles of human and mouse reprocessed for GRCh38 and GRCm38 genome assemblies.

The FANTOM5 consortium described the promoter-level expression atlas of human and mouse by using CAGE (Cap Analysis of Gene Expression) with single molecule sequencing. In the original publications, GRCh37/hg19 and NCBI37/mm9 assemblies were used as the reference genomes of human and mouse respectively; later, the Genome Reference Consortium released newer genome assemblies GRCh38/hg38 and GRCm38/mm10. To increase the utility of the atlas in forthcoming researches, we reprocessed the data to make them available on the recent genome assemblies. The data include observed frequencies of transcription starting sites (TSSs) based on the realignment of CAGE reads, and TSS peaks that are converted from those based on the previous reference. Annotations of the peak names were also updated based on the latest public databases. The reprocessed results enable us to examine frequencies of transcription initiations on the recent genome assemblies and to refer promoters with updated information across the genome assemblies consistently.

Differences in characteristics of glucose intolerance between patients with NAFLD and chronic hepatitis C as determined by CGMS.

Glucose intolerance frequently develops in accordance with the progression of chronic liver disease. However, differences in the characteristics of glucose intolerance between patients with nonalcoholic fatty liver disease (NAFLD) and those with chronic hepatitis C (C-CH) remain incompletely understood. To clarify these differences, patients with NAFLD (n = 37) and C-CH (n = 40) were evaluated with a continuous glucose monitoring system (CGMS). In the patients with NAFLD, Maximum blood glucose concentration and blood glucose swings were significantly correlated with hepatic fibrosis markers. In the patients with C-CH, however, those two CGMS parameters were negatively correlated with the serum albumin (ALB) concentration. Furthermore, in the patients with C-CH with an ALB concentration of ≤4.0 g/dl, those two CGMS parameters were negatively correlated with the ALB concentration with greater statistical significance. In conclusion, obvious differences in the characteristics of glucose intolerance between patients with NAFLD and those with C-CH were clarified. In patients with NAFLD, glucose intolerance gradually progressed in accordance with the progression of hepatic fibrosis. In those with C-CH, glucose intolerance suddenly developed upon the appearance of hypoalbuminaemia.