Cooperative Binding of Transcription Factors Orchestrates Reprogramming.

Oct4, Sox2, Klf4, and cMyc (OSKM) reprogram somatic cells to pluripotency. To gain a mechanistic understanding of their function, we mapped OSKM-binding, stage-specific transcription factors (TFs), and chromatin states in discrete reprogramming stages and performed loss- and gain-of-function experiments. We found that OSK predominantly bind active somatic enhancers early in reprogramming and immediately initiate their inactivation genome-wide by inducing the redistribution of somatic TFs away from somatic enhancers to sites elsewhere engaged by OSK, recruiting Hdac1, and repressing the somatic TF Fra1. Pluripotency enhancer selection is a stepwise process that also begins early in reprogramming through collaborative binding of OSK at sites with high OSK-motif density. Most pluripotency enhancers are selected later in the process and require OS and other pluripotency TFs. Somatic and pluripotency TFs modulate reprogramming efficiency when overexpressed by altering OSK targeting, somatic-enhancer inactivation, and pluripotency enhancer selection. Together, our data indicate that collaborative interactions among OSK and with stage-specific TFs direct both somatic-enhancer inactivation and pluripotency-enhancer selection to drive reprogramming.

X chromosome reactivation dynamics reveal stages of reprogramming to pluripotency.

Reprogramming to iPSCs resets the epigenome of somatic cells, including the reversal of X chromosome inactivation. We sought to gain insight into the steps underlying the reprogramming process by examining the means by which reprogramming leads to X chromosome reactivation (XCR). Analyzing single cells in situ, we found that hallmarks of the inactive X (Xi) change sequentially, providing a direct readout of reprogramming progression. Several epigenetic changes on the Xi occur in the inverse order of developmental X inactivation, whereas others are uncoupled from this sequence. Among the latter, DNA methylation has an extraordinary long persistence on the Xi during reprogramming, and, like Xist expression, is erased only after pluripotency genes are activated. Mechanistically, XCR requires both DNA demethylation and Xist silencing, ensuring that only cells undergoing faithful reprogramming initiate XCR. Our study defines the epigenetic state of multiple sequential reprogramming intermediates and establishes a paradigm for studying cell fate transitions during reprogramming.

Capturing cell type-specific chromatin compartment patterns by applying topic modeling to single-cell Hi-C data.

Single-cell Hi-C (scHi-C) interrogates genome-wide chromatin interaction in individual cells, allowing us to gain insights into 3D genome organization. However, the extremely sparse nature of scHi-C data poses a significant barrier to analysis, limiting our ability to tease out hidden biological information. In this work, we approach this problem by applying topic modeling to scHi-C data. Topic modeling is well-suited for discovering latent topics in a collection of discrete data. For our analysis, we generate nine different single-cell combinatorial indexed Hi-C (sci-Hi-C) libraries from five human cell lines (GM12878, H1Esc, HFF, IMR90, and HAP1), consisting over 19,000 cells. We demonstrate that topic modeling is able to successfully capture cell type differences from sci-Hi-C data in the form of "chromatin topics." We further show enrichment of particular compartment structures associated with locus pairs in these topics.

Novel features of telomere biology revealed by the absence of telomeric DNA methylation.

Cytosine methylation regulates the length and stability of telomeres, which can affect a wide variety of biological features, including cell differentiation, development, or illness. Although it is well established that subtelomeric regions are methylated, the presence of methylated cytosines at telomeres has remained controversial. Here, we have analyzed multiple bisulfite sequencing studies to address the methylation status of Arabidopsis thaliana telomeres. We found that the levels of estimated telomeric DNA methylation varied among studies. Interestingly, we estimated higher levels of telomeric DNA methylation in studies that produced C-rich telomeric strands with lower efficiency. However, these high methylation estimates arose due to experimental limitations of the bisulfite technique. We found a similar phenomenon for mitochondrial DNA: The levels of mitochondrial DNA methylation detected were higher in experiments with lower mitochondrial read production efficiencies. Based on experiments with high telomeric C-rich strand production efficiencies, we concluded that Arabidopsis telomeres are not methylated, which was confirmed by methylation-dependent restriction enzyme analyses. Thus, our studies indicate that telomeres are refractory to de novo DNA methylation by the RNA-directed DNA methylation machinery. This result, together with previously reported data, reveals that subtelomeric DNA methylation controls the homeostasis of telomere length.

Mediator coordinates PIC assembly with recruitment of CHD1.

Murine Chd1 (chromodomain helicase DNA-binding protein 1), a chromodomain-containing chromatin remodeling protein, is necessary for embryonic stem (ES) cell pluripotency. Chd1 binds to nucleosomes trimethylated at histone 3 Lys 4 (H3K4me3) near the beginning of active genes but not to bivalent domains also containing H3K27me3. To address the mechanism of this specificity, we reproduced H3K4me3- and CHD1-stimulated gene activation in HeLa extracts. Multidimensional protein identification technology (MuDPIT) and immunoblot analyses of purified preinitiation complexes (PICs) revealed the recruitment of CHD1 to naive chromatin but enhancement on H3K4me3 chromatin. Studies in depleted extracts showed that the Mediator coactivator complex, which controls PIC assembly, is also necessary for CHD1 recruitment. MuDPIT analyses of CHD1-associated proteins support the recruitment data and reveal numerous components of the PIC, including Mediator. In vivo, CHD1 and Mediator are recruited to an inducible gene, and genome-wide binding of the two proteins correlates well with active gene transcription in mouse ES cells. Finally, coimmunoprecipitation of CHD1 and Mediator from cell extracts can be ablated by shRNA knockdown of a specific Mediator subunit. Our data support a model in which the Mediator coordinates PIC assembly along with the recruitment of CHD1. The combined action of the PIC and H3K4me3 provides specificity in targeting CHD1 to active genes.

Comparison of reprogramming factor targets reveals both species-specific and conserved mechanisms in early iPSC reprogramming.

BACKGROUND: Both human and mouse fibroblasts can be reprogrammed to pluripotency with Oct4, Sox2, Klf4, and c-Myc (OSKM) transcription factors. While both systems generate pluripotency, human reprogramming takes considerably longer than mouse. RESULTS: To assess additional similarities and differences, we sought to compare the binding of the reprogramming factors between the two systems. In human fibroblasts, the OSK factors initially target many more closed chromatin sites compared to mouse. Despite this difference, the intra- and intergenic distribution of target sites, target genes, primary binding motifs, and combinatorial binding patterns between the reprogramming factors are largely shared. However, while many OSKM binding events in early mouse cell reprogramming occur in syntenic regions, only a limited number is conserved in human. CONCLUSIONS: Our findings suggest similar general effects of OSKM binding across these two species, even though the detailed regulatory networks have diverged significantly.

A Genome-Wide Chronological Study of Gene Expression and Two Histone Modifications, H3K4me3 and H3K9ac, during Developmental Leaf Senescence.

The genome-wide abundance of two histone modifications, H3K4me3 and H3K9ac (both associated with actively expressed genes), was monitored in Arabidopsis (Arabidopsis thaliana) leaves at different time points during developmental senescence along with expression in the form of RNA sequencing data. H3K9ac and H3K4me3 marks were highly convergent at all stages of leaf aging, but H3K4me3 marks covered nearly 2 times the gene area as H3K9ac marks. Genes with the greatest fold change in expression displayed the largest positively correlated percentage change in coverage for both marks. Most senescence up-regulated genes were premarked by H3K4me3 and H3K9ac but at levels below the whole-genome average, and for these genes, gene expression increased without a significant increase in either histone mark. However, for a subset of genes showing increased or decreased expression, the respective gain or loss of H3K4me3 marks was found to closely match the temporal changes in mRNA abundance; 22% of genes that increased expression during senescence showed accompanying changes in H3K4me3 modification, and they include numerous regulatory genes, which may act as primary response genes.

Reorganization of the host epigenome by a viral oncogene.

Adenovirus small e1a oncoprotein causes ~70% reduction in cellular levels of histone H3 lysine 18 acetylation (H3K18ac). It is unclear, however, where this dramatic reduction occurs genome-wide. ChIP-sequencing revealed that by 24 h after expression, e1a erases 95% of H3K18ac peaks in normal, contact-inhibited fibroblasts and replaces them with one-third as many at new genomic locations. The H3K18ac peaks at promoters and intergenic regions of genes with fibroblast-related functions are eliminated after infection, and new H3K18ac peaks are established at promoters of highly induced genes that regulate cell cycling and at new putative enhancers. Strikingly, the regions bound by the retinoblastoma family of proteins in contact-inhibited fibroblasts gain new peaks of H3K18ac in the e1a-expressing cells, including 55% of RB1-bound loci. In contrast, over half of H3K9ac peaks are similarly distributed before and after infection, independently of RB1. The strategic redistribution of H3K18ac by e1a highlights the importance of this modification for transcriptional activation and cellular transformation as well as functional differences between the RB-family member proteins.

Dynamic chromatin organization and regulatory interactions in human endothelial cell differentiation.

Vascular endothelial cells are a mesoderm-derived lineage with many essential functions, including angiogenesis and coagulation. The gene-regulatory mechanisms underpinning endothelial specialization are largely unknown, as are the roles of chromatin organization in regulating endothelial cell transcription. To investigate the relationships between chromatin organization and gene expression, we induced endothelial cell differentiation from human pluripotent stem cells and performed Hi-C and RNA-sequencing assays at specific time points. Long-range intrachromosomal contacts increase over the course of differentiation, accompanied by widespread heteroeuchromatic compartment transitions that are tightly associated with transcription. Dynamic topologically associating domain boundaries strengthen and converge on an endothelial cell state, and function to regulate gene expression. Chromatin pairwise point interactions (DNA loops) increase in frequency during differentiation and are linked to the expression of genes essential to vascular biology. Chromatin dynamics guide transcription in endothelial cell development and promote the divergence of endothelial cells from cardiomyocytes.

CTCF-mediated insulation and chromatin environment modulate Car5b escape from X inactivation.

Background: The number and escape levels of genes that escape X chromosome inactivation (XCI) in female somatic cells vary among tissues and cell types, potentially contributing to specific sex differences. Here we investigate the role of CTCF, a master chromatin conformation regulator, in regulating escape from XCI. CTCF binding profiles and epigenetic features were systematically examined at constitutive and facultative escape genes using mouse allelic systems to distinguish the inactive X (Xi) and active X (Xa) chromosomes. Results: We found that escape genes are located inside domains flanked by convergent arrays of CTCF binding sites, consistent with the formation of loops. In addition, strong and divergent CTCF binding sites often located at the boundaries between escape genes and adjacent neighbors subject to XCI would help insulate domains. Facultative escapees show clear differences in CTCF binding dependent on their XCI status in specific cell types/tissues. Concordantly, deletion but not inversion of a CTCF binding site at the boundary between the facultative escape gene Car5b and its silent neighbor Siah1b resulted in loss of Car5b escape. Reduced CTCF binding and enrichment of a repressive mark over Car5b in cells with a boundary deletion indicated loss of looping and insulation. In mutant lines in which either the Xi-specific compact structure or its H3K27me3 enrichment was disrupted, escape genes showed an increase in gene expression and associated active marks, supporting the roles of the 3D Xi structure and heterochromatic marks in constraining levels of escape. Conclusion: Our findings indicate that escape from XCI is modulated both by looping and insulation of chromatin via convergent arrays of CTCF binding sites and by compaction and epigenetic features of the surrounding heterochromatin.

Single-cell landscape of nuclear configuration and gene expression during stem cell differentiation and X inactivation.

BACKGROUND: Mammalian development is associated with extensive changes in gene expression, chromatin accessibility, and nuclear structure. Here, we follow such changes associated with mouse embryonic stem cell differentiation and X inactivation by integrating, for the first time, allele-specific data from these three modalities obtained by high-throughput single-cell RNA-seq, ATAC-seq, and Hi-C. RESULTS: Allele-specific contact decay profiles obtained by single-cell Hi-C clearly show that the inactive X chromosome has a unique profile in differentiated cells that have undergone X inactivation. Loss of this inactive X-specific structure at mitosis is followed by its reappearance during the cell cycle, suggesting a "bookmark" mechanism. Differentiation of embryonic stem cells to follow the onset of X inactivation is associated with changes in contact decay profiles that occur in parallel on both the X chromosomes and autosomes. Single-cell RNA-seq and ATAC-seq show evidence of a delay in female versus male cells, due to the presence of two active X chromosomes at early stages of differentiation. The onset of the inactive X-specific structure in single cells occurs later than gene silencing, consistent with the idea that chromatin compaction is a late event of X inactivation. Single-cell Hi-C highlights evidence of discrete changes in nuclear structure characterized by the acquisition of very long-range contacts throughout the nucleus. Novel computational approaches allow for the effective alignment of single-cell gene expression, chromatin accessibility, and 3D chromosome structure. CONCLUSIONS: Based on trajectory analyses, three distinct nuclear structure states are detected reflecting discrete and profound simultaneous changes not only to the structure of the X chromosomes, but also to that of autosomes during differentiation. Our study reveals that long-range structural changes to chromosomes appear as discrete events, unlike progressive changes in gene expression and chromatin accessibility.

Interactions between core histone marks and DNA methyltransferases predict DNA methylation patterns observed in human cells and tissues.

DNA methylation and histone modifications are two major epigenetic marks in mammalian cells. Previous studies have revealed that these two mechanisms interact although a quantitative model of these is still lacking in mammalian cells. Here we sought to develop such a model by systematically evaluating the quantitative relationship between DNA methylation and the core histone modification marks in human epigenomes. This model reflects the interactions of ADD and PWWP domains of DNA methyltransferase (DNMTs) with histone 3 lysine tails. Our analysis integrated 35 whole genome bisulphite sequencing data sets (about 800 million CpG sites), 35 chromatin states and 175 ChIP-Seq histone modification profiles across 35 human cell types. The logistic regression model we built shows that more than half of the variance across DNA methylomes can be explained by the five-core histone modification across varied types of human cell and tissue samples. Importantly, we find that H3K4me3 has a dramatic effect in DNA methylation patterning, highlighting the essential interaction between ADD domain of DNMTs and histone 3 lysine 4 in human. Moreover, our model suggests DNA methylation is generally inhibited by the presence of H3K4me3, H3K4me1 and H3K27me3, while increased levels are found in regions that are marked by H3K9me3 and H3K36me3. In summary, our results provide a comprehensive evaluation of the crosstalk between DNA methylation and histone modification in a variety of human cell types, and shows that DNA methylation patterns can be largely explained by interactions between histone 3 lysine tails and specific domains of DNA methyltransferases.

Unsupervised manifold alignment for single-cell multi-omics data.

Integrating single-cell measurements that capture different properties of the genome is vital to extending our understanding of genome biology. This task is challenging due to the lack of a shared axis across datasets obtained from different types of single-cell experiments. For most such datasets, we lack corresponding information among the cells (samples) and the measurements (features). In this scenario, unsupervised algorithms that are capable of aligning single-cell experiments are critical to learning an in silico co-assay that can help draw correspondences among the cells. Maximum mean discrepancy-based manifold alignment (MMD-MA) is such an unsupervised algorithm. Without requiring correspondence information, it can align single-cell datasets from different modalities in a common shared latent space, showing promising results on simulations and a small-scale single-cell experiment with 61 cells. However, it is essential to explore the applicability of this method to larger single-cell experiments with thousands of cells so that it can be of practical interest to the community. In this paper, we apply MMD-MA to two recent datasets that measure transcriptome and chromatin accessibility in ~2000 single cells. To scale the runtime of MMD-MA to a more substantial number of cells, we extend the original implementation to run on GPUs. We also introduce a method to automatically select one of the user-defined parameters, thus reducing the hyperparameter search space. We demonstrate that the proposed extensions allow MMD-MA to accurately align state-of-the-art single-cell experiments.

Trans- and cis-acting effects of Firre on epigenetic features of the inactive X chromosome.

Firre encodes a lncRNA involved in nuclear organization. Here, we show that Firre RNA expressed from the active X chromosome maintains histone H3K27me3 enrichment on the inactive X chromosome (Xi) in somatic cells. This trans-acting effect involves SUZ12, reflecting interactions between Firre RNA and components of the Polycomb repressive complexes. Without Firre RNA, H3K27me3 decreases on the Xi and the Xi-perinucleolar location is disrupted, possibly due to decreased CTCF binding on the Xi. We also observe widespread gene dysregulation, but not on the Xi. These effects are measurably rescued by ectopic expression of mouse or human Firre/FIRRE transgenes, supporting conserved trans-acting roles. We also find that the compact 3D structure of the Xi partly depends on the Firre locus and its RNA. In common lymphoid progenitors and T-cells Firre exerts a cis-acting effect on maintenance of H3K27me3 in a 26 Mb region around the locus, demonstrating cell type-specific trans- and cis-acting roles of this lncRNA.

Computational methods for analyzing and modeling genome structure and organization.

Recent advances in chromosome conformation capture technologies have led to the discovery of previously unappreciated structural features of chromatin. Computational analysis has been critical in detecting these features and thereby helping to uncover the building blocks of genome architecture. Algorithms are being developed to integrate these architectural features to construct better three-dimensional (3D) models of the genome. These computational methods have revealed the importance of 3D genome organization to essential biological processes. In this article, we review the state of the art in analytic and modeling techniques with a focus on their application to answering various biological questions related to chromatin structure. We summarize the limitations of these computational techniques and suggest future directions, including the importance of incorporating multiple sources of experimental data in building a more comprehensive model of the genome. This article is categorized under: Analytical and Computational Methods > Computational Methods Laboratory Methods and Technologies > Genetic/Genomic Methods Models of Systems Properties and Processes > Mechanistic Models.

DNA methylation estimation using methylation-sensitive restriction enzyme bisulfite sequencing (MREBS).

Whole-genome bisulfite sequencing (WGBS) and reduced representation bisulfite sequencing (RRBS) are widely used for measuring DNA methylation levels on a genome-wide scale. Both methods have limitations: WGBS is expensive and prohibitive for most large-scale projects; RRBS only interrogates 6-12% of the CpGs in the human genome. Here, we introduce methylation-sensitive restriction enzyme bisulfite sequencing (MREBS) which has the reduced sequencing requirements of RRBS, but significantly expands the coverage of CpG sites in the genome. We built a multiple regression model that combines the two features of MREBS: the bisulfite conversion ratios of single cytosines (as in WGBS and RRBS) as well as the number of reads that cover each locus (as in MRE-seq). This combined approach allowed us to estimate differential methylation across 60% of the genome using read count data alone, and where counts were sufficiently high in both samples (about 1.5% of the genome), our estimates were significantly improved by the single CpG conversion information. We show that differential DNA methylation values based on MREBS data correlate well with those based on WGBS and RRBS. This newly developed technique combines the sequencing cost of RRBS and DNA methylation estimates on a portion of the genome similar to WGBS, making it ideal for large-scale projects of mammalian genomes.

Dynamics of genome reorganization during human cardiogenesis reveal an RBM20-dependent splicing factory.

Functional changes in spatial genome organization during human development are poorly understood. Here we report a comprehensive profile of nuclear dynamics during human cardiogenesis from pluripotent stem cells by integrating Hi-C, RNA-seq and ATAC-seq. While chromatin accessibility and gene expression show complex on/off dynamics, large-scale genome architecture changes are mostly unidirectional. Many large cardiac genes transition from a repressive to an active compartment during differentiation, coincident with upregulation. We identify a network of such gene loci that increase their association inter-chromosomally, and are targets of the muscle-specific splicing factor RBM20. Genome editing studies show that TTN pre-mRNA, the main RBM20-regulated transcript in the heart, nucleates RBM20 foci that drive spatial proximity between the TTN locus and other inter-chromosomal RBM20 targets such as CACNA1C and CAMK2D. This mechanism promotes RBM20-dependent alternative splicing of the resulting transcripts, indicating the existence of a cardiac-specific trans-interacting chromatin domain (TID) functioning as a splicing factory.

X-Chromosome Inactivation and Escape from X Inactivation in Mouse.

X chromosome inactivation silences one X chromosome in female mammals. However, this silencing is incomplete, and some genes escape X inactivation. We describe methods to determine the chromosome-wide X inactivation status of genes in tissues or cell lines derived from mice using a combination of skewing of X inactivation and allele-specific analyses of gene expression based on RNA-seq.

X Chromosome Dosage Influences DNA Methylation Dynamics during Reprogramming to Mouse iPSCs.

A dramatic difference in global DNA methylation between male and female cells characterizes mouse embryonic stem cells (ESCs), unlike somatic cells. We analyzed DNA methylation changes during reprogramming of male and female somatic cells and in resulting induced pluripotent stem cells (iPSCs). At an intermediate reprogramming stage, somatic and pluripotency enhancers are targeted for partial methylation and demethylation. Demethylation within pluripotency enhancers often occurs at ESC binding sites of pluripotency transcription factors. Late in reprogramming, global hypomethylation is induced in a female-specific manner. Genome-wide hypomethylation in female cells affects many genomic landmarks, including enhancers and imprint control regions, and accompanies the reactivation of the inactive X chromosome. The loss of one of the two X chromosomes in propagating female iPSCs is associated with genome-wide methylation gain. Collectively, our findings highlight the dynamic regulation of DNA methylation at enhancers during reprogramming and reveal that X chromosome dosage dictates global DNA methylation levels in iPSCs.

Structural aspects of the inactive X chromosome.

A striking difference between male and female nuclei was recognized early on by the presence of a condensed chromatin body only in female cells. Mary Lyon proposed that X inactivation or silencing of one X chromosome at random in females caused this structural difference. Subsequent studies have shown that the inactive X chromosome (Xi) does indeed have a very distinctive structure compared to its active counterpart and all autosomes in female mammals. In this review, we will recap the discovery of this fascinating biological phenomenon and seminal studies in the field. We will summarize imaging studies using traditional microscopy and super-resolution technology, which revealed uneven compaction of the Xi. We will then discuss recent findings based on high-throughput sequencing techniques, which uncovered the distinct three-dimensional bipartite configuration of the Xi and the role of specific long non-coding RNAs in eliciting and maintaining this structure. The relative position of specific genomic elements, including genes that escape X inactivation, repeat elements and chromatin features, will be reviewed. Finally, we will discuss the position of the Xi, either near the nuclear periphery or the nucleolus, and the elements implicated in this positioning.This article is part of the themed issue 'X-chromosome inactivation: a tribute to Mary Lyon'.