The generation of myeloid cells from their progenitors is regulated at the level of transcription by combinatorial control of key transcription factors influencing cell-fate choice. To unravel the global dynamics of this process at the transcript level, we generated transcription profiles for 91 human cell types of myeloid origin by use of CAGE profiling. The CAGE sequencing of these samples has allowed us to investigate diverse aspects of transcription control during myelopoiesis, such as identification of novel transcription factors, miRNAs, and noncoding RNAs specific to the myeloid lineage. We further reconstructed a transcription regulatory network by clustering coexpressed transcripts and associating them with enriched cis-regulatory motifs. With the use of the bidirectional expression as a proxy for enhancers, we predicted over 2000 novel enhancers, including an enhancer 38 kb downstream of IRF8 and an intronic enhancer in the KIT gene locus. Finally, we highlighted relevance of these data to dissect transcription dynamics during progressive maturation of granulocyte precursors. A multifaceted analysis of the myeloid transcriptome is made available (www.myeloidome.roslin.ed.ac.uk). This high-quality dataset provides a powerful resource to study transcriptional regulation during myelopoiesis and to infer the likely functions of unannotated genes in human innate immunity.

The analysis of CAGE (Cap Analysis of Gene Expression) time-course has been proposed by the FANTOM5 Consortium to extend the understanding of the sequence of events facilitating cell state transition at the level of promoter regulation. To identify the most prominent transcriptional regulations induced by growth factors in human breast cancer, we apply here the Complexity Invariant Dynamic Time Warping motif EnRichment (CIDER) analysis approach to the CAGE time-course datasets of MCF-7 cells stimulated by epidermal growth factor (EGF) or heregulin (HRG). We identify a multi-level cascade of regulations rooted by the Serum Response Factor (SRF) transcription factor, connecting the MAPK-mediated transduction of the HRG stimulus to the negative regulation of the MAPK pathway by the members of the DUSP family phosphatases. The finding confirms the known primary role of FOS and FOSL1, members of AP-1 family, in shaping gene expression in response to HRG induction. Moreover, we identify a new potential regulation of DUSP5 and RARA (known to antagonize the transcriptional regulation induced by the estrogen receptors) by the activity of the AP-1 complex, specific to HRG response. The results indicate that a divergence in AP-1 regulation determines cellular changes of breast cancer cells stimulated by ErbB receptors.

Here, we review the role of sucrose nonfermenting (SNF2) family enzymes in blood cell development. The SNF2 family comprises helicase-like ATPases, originally discovered in yeast, that can remodel chromatin by changing chromatin structure and composition. The human genome encodes 30 different SNF2 enzymes. SNF2 family enzymes are often part of multisubunit chromatin remodeling complexes (CRCs), which consist of noncatalytic/auxiliary subunit along with the ATPase subunit. However, blood cells express a limited set of SNF2 ATPases that are necessary to maintain the pool of hematopoietic stem cells (HSCs) and drive normal blood cell development and differentiation. The composition of CRCs can be altered by the association of specific auxiliary subunits. Several auxiliary CRC subunits have specific functions in hematopoiesis. Aberrant expressions of SNF2 ATPases and/or auxiliary CRC subunit(s) are often observed in hematological malignancies. Using large-scale data from the International Cancer Genome Consortium (ICGC) we observed frequent mutations in genes encoding SNF2 helicase-like enzymes and auxiliary CRC subunits in leukemia. Hence, orderly function of SNF2 family enzymes is crucial for the execution of normal blood cell developmental program, and defects in chromatin remodeling caused by mutations or aberrant expression of these proteins may contribute to leukemogenesis.

Although it is generally accepted that cellular differentiation requires changes to transcriptional networks, dynamic regulation of promoters and enhancers at specific sets of genes has not been previously studied en masse. Exploiting the fact that active promoters and enhancers are transcribed, we simultaneously measured their activity in 19 human and 14 mouse time courses covering a wide range of cell types and biological stimuli. Enhancer RNAs, then messenger RNAs encoding transcription factors, dominated the earliest responses. Binding sites for key lineage transcription factors were simultaneously overrepresented in enhancers and promoters active in each cellular system. Our data support a highly generalizable model in which enhancer transcription is the earliest event in successive waves of transcriptional change during cellular differentiation or activation.

The DEK gene is highly expressed in a wide range of cancer cells, and a recurrent translocation partner in acute myeloid leukemia. While DEK has been identified as one of the most abundant proteins in human chromatin, its function and binding properties are not fully understood.

The capacity for plasticity in the adult brain is limited by the anatomical traces laid down during early postnatal life. Removing certain molecular brakes, such as histone deacetylases (HDACs), has proven to be effective in recapitulating juvenile plasticity in the mature visual cortex (V1). We investigated the chromatin structure and transcriptional control by genome-wide sequencing of DNase I hypersensitive sites (DHSS) and cap analysis of gene expression (CAGE) libraries after HDAC inhibition by valproic acid (VPA) in adult V1.

High-throughput sequencing is becoming the standard tool for investigating protein-DNA interactions or epigenetic modifications. However, the data generated will always contain noise due to e.g. repetitive regions or non-specific antibody interactions. The noise will appear in the form of a background distribution of reads that must be taken into account in the downstream analysis, for example when detecting enriched regions (peak-calling). Several reported peak-callers can take experimental measurements of background tag distribution into account when analysing a data set. Unfortunately, the background is only used to adjust peak calling and not as a pre-processing step that aims at discerning the signal from the background noise. A normalization procedure that extracts the signal of interest would be of universal use when investigating genomic patterns.

The Fulani ethnic group has relatively better protection from Plasmodium falciparum malaria, as reflected by fewer symptomatic cases of malaria, lower infection rates, and lower parasite densities compared to sympatric ethnic groups. However, the basis for this lower susceptibility to malaria by the Fulani is unknown. The incidence of classic malaria resistance genes are lower in the Fulani than in other sympatric ethnic populations, and targeted SNP analyses of other candidate genes involved in the immune response to malaria have not been able to account for the observed difference in the Fulani susceptibility to P.falciparum. Therefore, we have performed a pilot study to examine global transcription and DNA methylation patterns in specific immune cell populations in the Fulani to elucidate the mechanisms that confer the lower susceptibility to P.falciparum malaria. When we compared uninfected and infected Fulani individuals, in contrast to uninfected and infected individuals from the sympatric ethnic group Mossi, we observed a key difference: a strong transcriptional response was only detected in the monocyte fraction of the Fulani, where over 1000 genes were significantly differentially expressed upon P.falciparum infection.

Genetic variants underlying complex traits, including disease susceptibility, are enriched within the transcriptional regulatory elements, promoters and enhancers. There is emerging evidence that regulatory elements associated with particular traits or diseases share similar patterns of transcriptional activity. Accordingly, shared transcriptional activity (coexpression) may help prioritise loci associated with a given trait, and help to identify underlying biological processes. Using cap analysis of gene expression (CAGE) profiles of promoter- and enhancer-derived RNAs across 1824 human samples, we have analysed coexpression of RNAs originating from trait-associated regulatory regions using a novel quantitative method (network density analysis; NDA). For most traits studied, phenotype-associated variants in regulatory regions were linked to tightly-coexpressed networks that are likely to share important functional characteristics. Coexpression provides a new signal, independent of phenotype association, to enable fine mapping of causative variants. The NDA coexpression approach identifies new genetic variants associated with specific traits, including an association between the regulation of the OCT1 cation transporter and genetic variants underlying circulating cholesterol levels. NDA strongly implicates particular cell types and tissues in disease pathogenesis. For example, distinct groupings of disease-associated regulatory regions implicate two distinct biological processes in the pathogenesis of ulcerative colitis; a further two separate processes are implicated in Crohn's disease. Thus, our functional analysis of genetic predisposition to disease defines new distinct disease endotypes. We predict that patients with a preponderance of susceptibility variants in each group are likely to respond differently to pharmacological therapy. Together, these findings enable a deeper biological understanding of the causal basis of complex traits.

The Wilms' tumor gene 1 (WT1) is recurrently mutated in acute myeloid leukemia. Mutations and high expression of WT1 associate with a poor prognosis. In mice, WT1 cooperates with the RUNX1/RUNX1T1 (AML1/ETO) fusion gene in the induction of acute leukemia, further emphasizing a role for WT1 in leukemia development. Molecular mechanisms for WT1 are, however, incompletely understood. Here, we identify the transcriptional coregulator NAB2 as a target gene of WT1. Analysis of gene expression profiles of leukemic samples revealed a positive correlation between the expression of WT1 and NAB2, as well as a non-zero partial correlation. Overexpression of WT1 in hematopoietic cells resulted in increased NAB2 levels, while suppression of WT1 decreased NAB2 expression. WT1 bound and transactivated the proximal NAB2 promoter, as shown by ChIP and reporter experiments, respectively. ChIP experiments also revealed that WT1 can recruit NAB2 to the IRF8 promoter, thus modulating the transcriptional activity of WT1, as shown by reporter experiments. Our results implicate NAB2 as a previously unreported target gene of WT1 and that NAB2 acts as a transcriptional cofactor of WT1.

Treatment with anti-CD20 antibodies is only moderately efficient in chronic lymphocytic leukemia (CLL), a feature which has been explained by the inherently low CD20 expression in CLL. It has been shown that CD20 is epigenetically regulated and that histone deacetylase inhibitors (HDACis) can increase CD20 expression in vitro in CLL. To assess whether HDACis can upregulate CD20 also in vivo in CLL, the HDACi valproate was given to three del13q/NOTCH1wt CLL patients and CD20 levels were analysed (the PREVAIL study). Valproate treatment resulted in expected global activating histone modifications suggesting HDAC inhibitory effects. However, although valproate induced expression of CD20 mRNA and protein in the del13q/NOTCH1wt I83-E95 CLL cell line, no such effects were observed in the patients studied. In contrast to the cell line, in patients valproate treatment resulted in transient recruitment of the transcriptional repressor EZH2 to the CD20 promoter, correlating to an increase of the repressive histone mark H3K27me3. This suggests that valproate-mediated induction of CD20 may be hampered by EZH2 mediated H3K27me3 in vivo in CLL. Moreover, valproate treatment resulted in induction of EZH2 and global H3K27me3 in patient cells, suggesting transcriptionally repressive effects of valproate in CLL. Our results suggest new in vivo mechanisms of HDACis which may have implications on the design of future clinical trials in B-cell malignancies.

Heterochromatin regulation is critical for genomic stability. Different H3K9 methylation states have been discovered, with distinct roles in heterochromatin formation and silencing. However, how the transition from H3K9me2 to H3K9me3 is controlled is still unclear. Here, we investigate the role of the conserved bromodomain AAA-ATPase, Abo1, involved in maintaining global nucleosome organisation in fission yeast. We identified several key factors involved in heterochromatin silencing that interact genetically with Abo1: histone deacetylase Clr3, H3K9 methyltransferase Clr4, and HP1 homolog Swi6. Cells lacking Abo1 cultivated at 30 °C exhibit an imbalance of H3K9me2 and H3K9me3 in heterochromatin. In abo1∆ cells, the centromeric constitutive heterochromatin has increased H3K9me2 but decreased H3K9me3 levels compared to wild-type. In contrast, facultative heterochromatin regions exhibit reduced H3K9me2 and H3K9me3 levels in abo1∆. Genome-wide analysis showed that abo1∆ cells have silencing defects in both the centromeres and subtelomeres, but not in a subset of heterochromatin islands in our condition. Thus, our work uncovers a role of Abo1 in stabilising directly or indirectly Clr4 recruitment to allow the H3K9me2 to H3K9me3 transition in heterochromatin.

The response of the human acute myeloid leukemia cell line THP-1 to phorbol esters has been widely studied to test candidate leukemia therapies and as a model of cell cycle arrest and monocyte-macrophage differentiation. Here we have employed Cap Analysis of Gene Expression (CAGE) to analyze a dense time course of transcriptional regulation in THP-1 cells treated with phorbol myristate acetate (PMA) over 96 h. PMA treatment greatly reduced the numbers of cells entering S phase and also blocked cells exiting G2/M. The PMA-treated cells became adherent and expression of mature macrophage-specific genes increased progressively over the duration of the time course. Within 1-2 h PMA induced known targets of tumor protein p53 (TP53), notably CDKN1A, followed by gradual down-regulation of cell-cycle associated genes. Also within the first 2 h, PMA induced immediate early genes including transcription factor genes encoding proteins implicated in macrophage differentiation (EGR2, JUN, MAFB) and down-regulated genes for transcription factors involved in immature myeloid cell proliferation (MYB, IRF8, GFI1). The dense time course revealed that the response to PMA was not linear and progressive. Rather, network-based clustering of the time course data highlighted a sequential cascade of transient up- and down-regulated expression of genes encoding feedback regulators, as well as transcription factors associated with macrophage differentiation and their inferred target genes. CAGE also identified known and candidate novel enhancers expressed in THP-1 cells and many novel inducible genes that currently lack functional annotation and/or had no previously known function in macrophages. The time course is available on the ZENBU platform allowing comparison to FANTOM4 and FANTOM5 data.

The Helicase-related protein 3 (Hrp3), an ATP-dependent chromatin remodeling enzyme from the CHD family, is crucial for maintaining global nucleosome occupancy in Schizosaccharomyces pombe (S. pombe). Although the ATPase domain of Hrp3 is essential for chromatin remodeling, the contribution of non-ATPase domains of Hrp3 is still unclear. Here, we investigated the role of non-ATPase domains using in vitro methods. In our study, we expressed and purified recombinant S. pombe histone proteins, reconstituted them into histone octamers, and assembled nucleosome core particles. Using reconstituted nucleosomes and affinity-purified wild type and mutant Hrp3 from S. pombe we created a homogeneous in vitro system to evaluate the ATP hydrolyzing capacity of truncated Hrp3 proteins. We found that all non-ATPase domain deletions (∆chromo, ∆SANT, ∆SLIDE, and ∆coupling region) lead to reduced ATP hydrolyzing activities in vitro with DNA or nucleosome substrates. Only the coupling region deletion showed moderate stimulation of ATPase activity with the nucleosome. Interestingly, affinity-purified Hrp3 showed co-purification with all core histones suggesting a strong association with the nucleosomes in vivo. However, affinity-purified Hrp3 mutant with SANT and coupling regions deletion showed complete loss of interactions with the nucleosomes, while SLIDE and chromodomain deletions reduced Hrp3 interactions with the nucleosomes. Taken together, nucleosome association and ATPase stimulation by DNA or nucleosomes substrate suggest that the enzymatic activity of Hrp3 is fine-tuned by unique contributions of all four non-catalytic domains.

Aberrant DNA methylation, dysregulation of chromatin-modifying enzymes, and microRNAs (miRNAs) play a crucial role in haematological malignancies. These epimutations, with an impact on chromatin accessibility and transcriptional output, are often associated with genomic instability and the emergence of drug resistance, disease progression, and poor survival. In order to exert their functions, epigenetic enzymes utilize cellular metabolites as co-factors and are highly dependent on their availability. By affecting the expression of metabolic enzymes, epigenetic modifiers may aid the generation of metabolite signatures that could be utilized as targets and biomarkers in cancer. This interdependency remains often neglected and poorly represented in studies, despite well-established methods to study the cellular metabolome. This review critically summarizes the current knowledge in the field to provide an integral picture of the interplay between epigenomic alterations and the cellular metabolome in haematological malignancies. Our recent findings defining a distinct metabolic signature upon response to enhancer of zeste homolog 2 (EZH2) inhibition in multiple myeloma (MM) highlight how a shift of preferred metabolic pathways may potentiate novel treatments. The suggested link between the epigenome and the metabolome in haematopoietic tumours holds promise for the use of metabolic signatures as possible biomarkers of response to treatment.

Transposable elements are an abundant source of transcription factor binding sites, and favorable genomic integration may lead to their recruitment by the host genome for gene regulatory functions. However, it is unclear how frequent co-option of transposable elements as regulatory elements is, to which regulatory programs they contribute and how they compare to regulatory elements devoid of transposable elements. Here, we report a transcription initiation-centric, in-depth characterization of the transposon-derived regulatory landscape of mouse embryonic stem cells. We demonstrate that a substantial number of transposable element insertions, in particular endogenous retroviral elements, are associated with open chromatin regions that are divergently transcribed into unstable RNAs in a cell-type specific manner, and that these elements contribute to a sizable proportion of active enhancers and gene promoters. We further show that transposon subfamilies contribute differently and distinctly to the pluripotency regulatory program through their repertoires of transcription factor binding site sequences, shedding light on the formation of regulatory programs and the origins of regulatory elements.

Standard culture of human induced pluripotent stem cells (hiPSCs) requires basic Fibroblast Growth Factor (bFGF) to maintain the pluripotent state, whereas hiPSC more closely resemble epiblast stem cells than true naïve state ES which requires LIF to maintain pluripotency. Here we show that chemokine (C-C motif) ligand 2 (CCL2) enhances the expression of pluripotent marker genes through the phosphorylation of the signal transducer and activator of transcription 3 (STAT3) protein. Moreover, comparison of transcriptomes between hiPSCs cultured with CCL2 versus with bFGF, we found that CCL2 activates hypoxia related genes, suggesting that CCL2 enhanced pluripotency by inducing a hypoxic-like response.Further, we show that hiPSCs cultured with CCL2 can differentiate at a higher efficiency than culturing withjust bFGF and we show CCL2 can be used in feeder-free conditions [corrected]. Taken together, our finding indicates the novel functions of CCL2 in enhancing its pluripotency in hiPSCs.

Transcriptional Regulatory Networks (TRNs) coordinate multiple transcription factors (TFs) in concert to maintain tissue homeostasis and cellular function. The re-establishment of target cell TRNs has been previously implicated in direct trans-differentiation studies where the newly introduced TFs switch on a set of key regulatory factors to induce de novo expression and function. However, the extent to which TRNs in starting cell types, such as dermal fibroblasts, protect cells from undergoing cellular reprogramming remains largely unexplored. In order to identify TFs specific to maintaining the fibroblast state, we performed systematic knockdown of 18 fibroblast-enriched TFs and analyzed differential mRNA expression against the same 18 genes, building a Matrix-RNAi. The resulting expression matrix revealed seven highly interconnected TFs. Interestingly, suppressing four out of seven TFs generated lipid droplets and induced PPARG and CEBPA expression in the presence of adipocyte-inducing medium only, while negative control knockdown cells maintained fibroblastic character in the same induction regime. Global gene expression analyses further revealed that the knockdown-induced adipocytes expressed genes associated with lipid metabolism and significantly suppressed fibroblast genes. Overall, this study reveals the critical role of the TRN in protecting cells against aberrant reprogramming, and demonstrates the vulnerability of donor cell's TRNs, offering a novel strategy to induce transgene-free trans-differentiations.

Humans are composed of hundreds of cell types. As the genomic DNA of each somatic cell is identical, cell type is determined by what is expressed and when. Until recently, little has been reported about the determinants of human cell identity, particularly from the joint perspective of gene evolution and expression. Here, we chart the evolutionary past of all documented human cell types via the collective histories of proteins, the principal product of gene expression. FANTOM5 data provide cell-type-specific digital expression of human protein-coding genes and the SUPERFAMILY resource is used to provide protein domain annotation. The evolutionary epoch in which each protein was created is inferred by comparison with domain annotation of all other completely sequenced genomes. Studying the distribution across epochs of genes expressed in each cell type reveals insights into human cellular evolution in terms of protein innovation. For each cell type, its history of protein innovation is charted based on the genes it expresses. Combining the histories of all cell types enables us to create a timeline of cell evolution. This timeline identifies the possibility that our common ancestor Coelomata (cavity-forming animals) provided the innovation required for the innate immune system, whereas cells which now form the brain of human have followed a trajectory of continually accumulating novel proteins since Opisthokonta (boundary of animals and fungi). We conclude that exaptation of existing domain architectures into new contexts is the dominant source of cell-type-specific domain architectures.

The fibrillins and latent transforming growth factor binding proteins (LTBPs) form a superfamily of extracellular matrix (ECM) proteins characterized by the presence of a unique domain, the 8-cysteine transforming growth factor beta (TGFβ) binding domain. These proteins are involved in the structure of the extracellular matrix and controlling the bioavailability of TGFβ family members. Genes encoding these proteins show differential expression in mesenchymal cell types which synthesize the extracellular matrix. We have investigated the promoter regions of the seven gene family members using the FANTOM5 CAGE database for human. While the protein and nucleotide sequences show considerable sequence similarity, the promoter regions were quite diverse. Most genes had a single predominant transcription start site region but LTBP1 and LTBP4 had two regions initiating different transcripts. Most of the family members were expressed in a range of mesenchymal and other cell types, often associated with use of alternative promoters or transcription start sites within a promoter in different cell types. FBN3 was the lowest expressed gene, and was found only in embryonic and fetal tissues. The different promoters for one gene were more similar to each other in expression than to promoters of the other family members. Notably expression of all 22 LTBP2 promoters was tightly correlated and quite distinct from all other family members. We located candidate enhancer regions likely to be involved in expression of the genes. Each gene was associated with a unique subset of transcription factors across multiple promoters although several motifs including MAZ, SP1, GTF2I and KLF4 showed overrepresentation across the gene family. FBN1 and FBN2, which had similar expression patterns, were regulated by different transcription factors. This study highlights the role of alternative transcription start sites in regulating the tissue specificity of closely related genes and suggests that this important class of extracellular matrix proteins is subject to subtle regulatory variations that explain the differential roles of members of this gene family.