Mammalian gene promoters and enhancers share many properties. They are composed of a unified promoter architecture of divergent transcripton initiation and gene promoters may exhibit enhancer function. However, it is currently unclear how expression strength of a regulatory element relates to its enhancer strength and if the unifying architecture is conserved across Metazoa. Here we investigate the transcription initiation landscape and its associated RNA decay in Drosophila melanogaster. We find that the majority of active gene-distal enhancers and a considerable fraction of gene promoters are divergently transcribed. We observe quantitative relationships between enhancer potential, expression level and core promoter strength, providing an explanation for indirectly related histone modifications that are reflecting expression levels. Lowly abundant unstable RNAs initiated from weak core promoters are key characteristics of gene-distal developmental enhancers, while the housekeeping enhancer strengths of gene promoters reflect their expression strengths. The seemingly separable layer of regulation by gene promoters with housekeeping enhancer potential is also indicated by chromatin interaction data. Our results suggest a unified promoter architecture of many D. melanogaster regulatory elements, that is universal across Metazoa, whose regulatory functions seem to be related to their core promoter elements.

The promoters of immediate early genes (IEGs) are rapidly activated in response to an external stimulus. These genes, also known as primary response genes, have been identified in a range of cell types, under diverse extracellular signals and using varying experimental protocols. Whereas genomic dissection on a case-by-case basis has not resulted in a comprehensive catalogue of IEGs, a rigorous meta-analysis of eight genome-wide FANTOM5 CAGE (cap analysis of gene expression) time course datasets reveals successive waves of promoter activation in IEGs, recapitulating known relationships between cell types and stimuli: we obtain a set of 57 (42 protein-coding) candidate IEGs possessing promoters that consistently drive a rapid but transient increase in expression over time. These genes show significant enrichment for known IEGs reported previously, pathways associated with the immediate early response, and include a number of non-coding RNAs with roles in proliferation and differentiation. Surprisingly, we also find strong conservation of the ordering of activation for these genes, such that 77 pairwise promoter activation orderings are conserved. Using the leverage of comprehensive CAGE time series data across cell types, we also document the extensive alternative promoter usage by such genes, which is likely to have been a barrier to their discovery until now. The common activation ordering of the core set of early-responding genes we identify may indicate conserved underlying regulatory mechanisms. By contrast, the considerably larger number of transiently activated genes that are specific to each cell type and stimulus illustrates the breadth of the primary response.

The work of the FANTOM5 Consortium has brought forth a new level of understanding of the regulation of gene transcription and the cellular processes involved in creating diversity of cell types. In this study, we extended the analysis of the FANTOM5 Cap Analysis of Gene Expression (CAGE) transcriptome data to focus on understanding the genetic regulators involved in mouse cerebellar development.

Modified parental histones are segregated symmetrically to daughter DNA strands during replication and can be inherited through mitosis. How this may sustain the epigenome and cell identity remains unknown. Here we show that transmission of histone-based information during DNA replication maintains epigenome fidelity and embryonic stem cell plasticity. Asymmetric segregation of parental histones H3-H4 in MCM2-2A mutants compromised mitotic inheritance of histone modifications and globally altered the epigenome. This included widespread spurious deposition of repressive modifications, suggesting elevated epigenetic noise. Moreover, H3K9me3 loss at repeats caused derepression and H3K27me3 redistribution across bivalent promoters correlated with misexpression of developmental genes. MCM2-2A mutation challenged dynamic transitions in cellular states across the cell cycle, enhancing naïve pluripotency and reducing lineage priming in G1. Furthermore, developmental competence was diminished, correlating with impaired exit from pluripotency. Collectively, this argues that epigenetic inheritance of histone modifications maintains a correctly balanced and dynamic chromatin landscape able to support mammalian cell differentiation.

Gene expression programs change during cellular transitions. It is well established that a network of transcription factors and chromatin modifiers regulate RNA levels during embryonic stem cell (ESC) differentiation, but the full impact of post-transcriptional processes remains elusive. While cytoplasmic RNA turnover mechanisms have been implicated in differentiation, the contribution of nuclear RNA decay has not been investigated. Here, we differentiate mouse ESCs, depleted for the ribonucleolytic RNA exosome, into embryoid bodies to determine to which degree RNA abundance in the two states can be attributed to changes in transcription versus RNA decay by the exosome. As a general observation, we find that exosome depletion mainly leads to the stabilization of RNAs from lowly transcribed loci, including several protein-coding genes. Depletion of the nuclear exosome cofactor RBM7 leads to similar effects. In particular, transcripts that are differentially expressed between states tend to be more exosome sensitive in the state where expression is low. We conclude that the RNA exosome contributes to down-regulation of transcripts with disparate expression, often in conjunction with transcriptional down-regulation.

The PLZF transcription factor is essential for osteogenic differentiation of hMSCs; however, its regulation and molecular function during this process is not fully understood. Here, we revealed that the ZBTB16 locus encoding PLZF, is repressed by Polycomb (PcG) and H3K27me3 in naive hMSCs. At the pre-osteoblast stage of differentiation, the locus lost PcG binding and H3K27me3, gained JMJD3 recruitment, and H3K27ac resulting in high expression of PLZF. Subsequently, PLZF was recruited to osteogenic enhancers, influencing H3K27 acetylation and expression of nearby genes important for osteogenic function. Furthermore, we identified a latent enhancer within the ZBTB16/PLZF locus itself that became active, gained PLZF, p300 and Mediator binding and looped to the promoter of the nicotinamide N-methyltransferase (NNMT) gene. The increased expression of NNMT correlated with a decline in SAM levels, which is dependent on PLZF and is required for osteogenic differentiation.

Smooth muscle cells (SMC) in blood vessels are normally growth quiescent and transcriptionally inactive. Our objective was to understand promoter usage and dynamics in SMC acutely exposed to a prototypic growth factor or pro-inflammatory cytokine. Using cap analysis gene expression (FANTOM5 project) we report differences in promoter dynamics for immediate-early genes (IEG) and other genes when SMC are exposed to fibroblast growth factor-2 or interleukin-1β. Of the 1871 promoters responding to FGF2 or IL-1β considerably more responded to FGF2 (68.4%) than IL-1β (18.5%) and 13.2% responded to both. Expression clustering reveals sets of genes induced, repressed or unchanged. Among IEG responding rapidly to FGF2 or IL-1β were FOS, FOSB and EGR-1, which mediates human SMC migration. Motif activity response analysis (MARA) indicates most transcription factor binding motifs in response to FGF2 were associated with a sharp induction at 1 h, whereas in response to IL-1β, most motifs were associated with a biphasic change peaking generally later. MARA revealed motifs for FOS_FOS{B,L1}_JUN{B,D} and EGR-1..3 in the cluster peaking 1 h after FGF2 exposure whereas these motifs were in clusters peaking 1 h or later in response to IL-1β. Our findings interrogating CAGE data demonstrate important differences in promoter usage and dynamics in SMC exposed to FGF2 or IL-1β.

Mammals are composed of hundreds of different cell types with specialized functions. Each of these cellular phenotypes are controlled by different combinations of transcription factors. Using a human non islet cell insulinoma cell line (TC-YIK) which expresses insulin and the majority of known pancreatic beta cell specific genes as an example, we describe a general approach to identify key cell-type-specific transcription factors (TFs) and their direct and indirect targets. By ranking all human TFs by their level of enriched expression in TC-YIK relative to a broad collection of samples (FANTOM5), we confirmed known key regulators of pancreatic function and development. Systematic siRNA mediated perturbation of these TFs followed by qRT-PCR revealed their interconnections with NEUROD1 at the top of the regulation hierarchy and its depletion drastically reducing insulin levels. For 15 of the TF knock-downs (KD), we then used Cap Analysis of Gene Expression (CAGE) to identify thousands of their targets genome-wide (KD-CAGE). The data confirm NEUROD1 as a key positive regulator in the transcriptional regulatory network (TRN), and ISL1, and PROX1 as antagonists. As a complimentary approach we used ChIP-seq on four of these factors to identify NEUROD1, LMX1A, PAX6, and RFX6 binding sites in the human genome. Examining the overlap between genes perturbed in the KD-CAGE experiments and genes with a ChIP-seq peak within 50 kb of their promoter, we identified direct transcriptional targets of these TFs. Integration of KD-CAGE and ChIP-seq data shows that both NEUROD1 and LMX1A work as the main transcriptional activators. In the core TRN (i.e., TF-TF only), NEUROD1 directly transcriptionally activates the pancreatic TFs HSF4, INSM1, MLXIPL, MYT1, NKX6-3, ONECUT2, PAX4, PROX1, RFX6, ST18, DACH1, and SHOX2, while LMX1A directly transcriptionally activates DACH1, SHOX2, PAX6, and PDX1. Analysis of these complementary datasets suggests the need for caution in interpreting ChIP-seq datasets. (1) A large fraction of binding sites are at distal enhancer sites and cannot be directly associated to their targets, without chromatin conformation data. (2) Many peaks may be non-functional: even when there is a peak at a promoter, the expression of the gene may not be affected in the matching perturbation experiment.

Despite their well-established functional roles, histone modifications have received less attention than DNA methylation in the cancer field. In order to evaluate their importance in colorectal cancer (CRC), we generated the first genome-wide histone modification profiles in paired normal colon mucosa and tumor samples.

Classically or alternatively activated macrophages (M1 and M2, respectively) play distinct and important roles for microbiocidal activity, regulation of inflammation and tissue homeostasis. Despite this, their transcriptional regulatory dynamics are poorly understood. Using promoter-level expression profiling by non-biased deepCAGE we have studied the transcriptional dynamics of classically and alternatively activated macrophages. Transcription factor (TF) binding motif activity analysis revealed four motifs, NFKB1_REL_RELA, IRF1,2, IRF7 and TBP that are commonly activated but have distinct activity dynamics in M1 and M2 activation. We observe matching changes in the expression profiles of the corresponding TFs and show that only a restricted set of TFs change expression. There is an overall drastic and transient up-regulation in M1 and a weaker and more sustainable up-regulation in M2. Novel TFs, such as Thap6, Maff, (M1) and Hivep1, Nfil3, Prdm1, (M2) among others, were suggested to be involved in the activation processes. Additionally, 52 (M1) and 67 (M2) novel differentially expressed genes and, for the first time, several differentially expressed long non-coding RNA (lncRNA) transcriptome markers were identified. In conclusion, the finding of novel motifs, TFs and protein-coding and lncRNA genes is an important step forward to fully understand the transcriptional machinery of macrophage activation.

Inflammatory bowel disease (IBD) is a chronic intestinal disorder, with two main types: Crohn's disease (CD) and ulcerative colitis (UC), whose molecular pathology is not well understood. The majority of IBD-associated SNPs are located in non-coding regions and are hard to characterize since regulatory regions in IBD are not known. Here we profile transcription start sites (TSSs) and enhancers in the descending colon of 94 IBD patients and controls. IBD-upregulated promoters and enhancers are highly enriched for IBD-associated SNPs and are bound by the same transcription factors. IBD-specific TSSs are associated to genes with roles in both inflammatory cascades and gut epithelia while TSSs distinguishing UC and CD are associated to gut epithelia functions. We find that as few as 35 TSSs can distinguish active CD, UC, and controls with 85% accuracy in an independent cohort. Our data constitute a foundation for understanding the molecular pathology, gene regulation, and genetics of IBD.

Epigenetic information is available from contemporary organisms, but is difficult to track back in evolutionary time. Here, we show that genome-wide epigenetic information can be gathered directly from next-generation sequence reads of DNA isolated from ancient remains. Using the genome sequence data generated from hair shafts of a 4000-yr-old Paleo-Eskimo belonging to the Saqqaq culture, we generate the first ancient nucleosome map coupled with a genome-wide survey of cytosine methylation levels. The validity of both nucleosome map and methylation levels were confirmed by the recovery of the expected signals at promoter regions, exon/intron boundaries, and CTCF sites. The top-scoring nucleosome calls revealed distinct DNA positioning biases, attesting to nucleotide-level accuracy. The ancient methylation levels exhibited high conservation over time, clustering closely with modern hair tissues. Using ancient methylation information, we estimated the age at death of the Saqqaq individual and illustrate how epigenetic information can be used to infer ancient gene expression. Similar epigenetic signatures were found in other fossil material, such as 110,000- to 130,000-yr-old bones, supporting the contention that ancient epigenomic information can be reconstructed from a deep past. Our findings lay the foundation for extracting epigenomic information from ancient samples, allowing shifts in epialleles to be tracked through evolutionary time, as well as providing an original window into modern epigenomics.

Nucleosomes play important roles in a cell beyond their basal functionality in chromatin compaction. Their placement affects all steps in transcriptional regulation, from transcription factor (TF) binding to messenger ribonucleic acid (mRNA) synthesis. Careful profiling of their locations and dynamics in response to stimuli is important to further our understanding of transcriptional regulation by the state of chromatin. We measured nucleosome occupancy in human hepatic cells before and after treatment with transforming growth factor beta 1 (TGFβ1), using massively parallel sequencing. With a newly developed method, SuMMIt, for precise positioning of nucleosomes we inferred dynamics of the nucleosomal landscape. Distinct nucleosome positioning has previously been described at transcription start site and flanking TF binding sites. We found that the average pattern is present at very few sites and, in case of TF binding, the double peak surrounding the sites is just an artifact of averaging over many loci. We systematically searched for depleted nucleosomes in stimulated cells compared to unstimulated cells and identified 24 318 loci. Depending on genomic annotation, 44-78% of them were over-represented in binding motifs for TFs. Changes in binding affinity were verified for HNF4α by qPCR. Strikingly many of these loci were associated with expression changes, as measured by RNA sequencing.

CAGE (cap analysis gene expression) and RNA-seq are two major technologies used to identify transcript abundances as well as structures. They measure expression by sequencing from either the 5' end of capped molecules (CAGE) or tags randomly distributed along the length of a transcript (RNA-seq). Library protocols for clonally amplified (Illumina, SOLiD, 454 Life Sciences [Roche], Ion Torrent), second-generation sequencing platforms typically employ PCR preamplification prior to clonal amplification, while third-generation, single-molecule sequencers can sequence unamplified libraries. Although these transcriptome profiling platforms have been demonstrated to be individually reproducible, no systematic comparison has been carried out between them. Here we compare CAGE, using both second- and third-generation sequencers, and RNA-seq, using a second-generation sequencer based on a panel of RNA mixtures from two human cell lines to examine power in the discrimination of biological states, detection of differentially expressed genes, linearity of measurements, and quantification reproducibility. We found that the quantified levels of gene expression are largely comparable across platforms and conclude that CAGE and RNA-seq are complementary technologies that can be used to improve incomplete gene models. We also found systematic bias in the second- and third-generation platforms, which is likely due to steps such as linker ligation, cleavage by restriction enzymes, and PCR amplification. This study provides a perspective on the performance of these platforms, which will be a baseline in the design of further experiments to tackle complex transcriptomes uncovered in a wide range of cell types.

Transcriptional regulation is tightly coupled with chromosomal positioning and three-dimensional chromatin architecture. However, it is unclear what proportion of transcriptional activity is reflecting such organisation, how much can be informed by RNA expression alone and how this impacts disease. Here, we develop a computational transcriptional decomposition approach separating the proportion of expression associated with genome organisation from independent effects not directly related to genomic positioning. We show that positionally attributable expression accounts for a considerable proportion of total levels and is highly informative of topological associating domain activities and organisation, revealing boundaries and chromatin compartments. Furthermore, expression data alone accurately predict individual enhancer-promoter interactions, drawing features from expression strength, stabilities, insulation and distance. We characterise predictions in 76 human cell types, observing extensive sharing of domains, yet highly cell-type-specific enhancer-promoter interactions and strong enrichments in relevant trait-associated variants. Overall, our work demonstrates a close relationship between transcription and chromatin architecture.

Rhesus macaque was the second non-human primate whose genome has been fully sequenced and is one of the most used model organisms to study human biology and disease, thanks to the close evolutionary relationship between the two species. But compared to human, where several previously unknown RNAs have been uncovered, the macaque transcriptome is less studied. Publicly available RNA expression resources for macaque are limited, even for brain, which is highly relevant to study human cognitive abilities. In an effort to complement those resources, FANTOM5 profiled 15 distinct anatomical regions of the aged macaque central nervous system using Cap Analysis of Gene Expression, a high-resolution, annotation-independent technology that allows monitoring of transcription initiation events with high accuracy. We identified 25,869 CAGE peaks, representing bona fide promoters. For each peak we provide detailed annotation, expanding the landscape of 'known' macaque genes, and we show concrete examples on how to use the resulting data. We believe this data represents a useful resource to understand the central nervous system in macaque.

We have identified a variant in ADCY3 (encoding adenylate cyclase 3) associated with markedly increased risk of obesity and type 2 diabetes in the Greenlandic population. The variant disrupts a splice acceptor site, and carriers have decreased ADCY3 RNA expression. Additionally, we observe an enrichment of rare ADCY3 loss-of-function variants among individuals with type 2 diabetes in trans-ancestry cohorts. These findings provide new information on disease etiology relevant for future treatment strategies.

In eukaryotes, capped RNAs include long transcripts such as messenger RNAs and long noncoding RNAs, as well as shorter transcripts such as spliceosomal RNAs, small nucleolar RNAs, and enhancer RNAs. Long capped transcripts can be profiled using cap analysis gene expression (CAGE) sequencing and other methods. Here, we describe a sequencing library preparation protocol for short capped RNAs, apply it to a differentiation time course of the human cell line THP-1, and systematically compare the landscape of short capped RNAs to that of long capped RNAs. Transcription initiation peaks associated with genes in the sense direction have a strong preference to produce either long or short capped RNAs, with one out of six peaks detected in the short capped RNA libraries only. Gene-associated short capped RNAs have highly specific 3' ends, typically overlapping splice sites. Enhancers also preferentially generate either short or long capped RNAs, with 10% of enhancers observed in the short capped RNA libraries only. Enhancers producing either short or long capped RNAs show enrichment for GWAS-associated disease SNPs. We conclude that deep sequencing of short capped RNAs reveals new families of noncoding RNAs and elucidates the diversity of transcripts generated at known and novel promoters and enhancers.

Myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) is a complex and debilitating disease with no molecular diagnostics and no treatment options. To identify potential markers of this illness, we profiled 48 patients and 52 controls for standard laboratory tests, plasma metabolomics, blood immuno-phenotyping and transcriptomics, and fecal microbiome analysis. Here, we identified a set of 26 potential molecular markers that distinguished ME/CFS patients from healthy controls. Monocyte number, microbiome abundance, and lipoprotein profiles appeared to be the most informative markers. When we correlated these molecular changes to sleep and cognitive measurements of fatigue, we found that lipoprotein and microbiome profiles most closely correlated with sleep disruption while a different set of markers correlated with a cognitive parameter. Sleep, lipoprotein, and microbiome changes occur early during the course of illness suggesting that these markers can be examined in a larger cohort for potential biomarker application. Our study points to a cluster of sleep-related molecular changes as a prominent feature of ME/CFS in our Japanese cohort.

Many genes are co-expressed and form genomic domains of coordinated gene activity. However, the regulatory determinants of domain co-activity remain unclear. Here, we leverage human individual variation in gene expression to characterize the co-regulatory processes underlying domain co-activity and systematically quantify their effect sizes. We employ transcriptional decomposition to extract from RNA expression data an expression component related to co-activity revealed by genomic positioning. This strategy reveals close to 1,500 co-activity domains, covering most expressed genes, of which the large majority are invariable across individuals. Focusing specifically on domains with high variability in co-activity reveals that contained genes have a higher sharing of eQTLs, a higher variability in enhancer interactions, and an enrichment of binding by variably expressed transcription factors, compared to genes within non-variable domains. Through careful quantification of the relative contributions of regulatory processes underlying co-activity, we find transcription factor expression levels to be the main determinant of gene co-activity. Our results indicate that distal trans effects contribute more than local genetic variation to individual variation in co-activity domains.