Project description:In mammalian cells transcription factors (TFs) preferentially bind sites contained in regions of high nucleosomal occupancy, as determined by nucleotide-dependent computational analysis. This observation suggests that nucleosomes may act as gatekeepers of TF binding sites. We hypothesized that in mammalian genomes the information controlling nucleosome assembly may partially coincide with the information that enables TFs to recognize cognate sites in cis-regulatory elements while ignoring the myriad of non-functional, randomly occurring consensus binding sites. This way, nucleosome-mediated masking would be coupled to TF binding site functionality. The hematopoietic master regulator Pu.1 maintained nucleosome depletion at macrophage-specific enhancers that were otherwise occupied by nucleosomes in other cell types and in reconstituted chromatin. We identified a minimal set of DNA sequence and shape features that predicted Pu.1 binding with 78% accuracy. The same features predicted nucleosome occupancy in cells where Pu.1 was not expressed with higher accuracy than specifically designed models. Control of nucleosome deposition by DNA sequence and shape features that also specify TF consensus site functionality may allow maintaining the gatekeeper function of nucleosomes during evolution of cis-regulatory elements.
Project description:In mammalian cells transcription factors (TFs) preferentially bind sites contained in regions of high nucleosomal occupancy, as determined by nucleotide-dependent computational analysis. This observation suggests that nucleosomes may act as gatekeepers of TF binding sites. We hypothesized that in mammalian genomes the information controlling nucleosome assembly may partially coincide with the information that enables TFs to recognize cognate sites in cis-regulatory elements while ignoring the myriad of non-functional, randomly occurring consensus binding sites. This way, nucleosome-mediated masking would be coupled to TF binding site functionality. The hematopoietic master regulator Pu.1 maintained nucleosome depletion at macrophage-specific enhancers that were otherwise occupied by nucleosomes in other cell types and in reconstituted chromatin. We identified a minimal set of DNA sequence and shape features that predicted Pu.1 binding with 78% accuracy. The same features predicted nucleosome occupancy in cells where Pu.1 was not expressed with higher accuracy than specifically designed models. Control of nucleosome deposition by DNA sequence and shape features that also specify TF consensus site functionality may allow maintaining the gatekeeper function of nucleosomes during evolution of cis-regulatory elements. Chromatin immuno-precipitations of the transcription factor Pu.1 followed by multiparallel sequencing performed in murine bone marrow-derived macrophages. Experiments carried out in cells infected either with a retroviral vector containing a short hairpin targeting Pu.1 or with the empty vector as control. The shPU.1 hairpin (sequence available upon request) was selected among five designed using a publicly available software (http://katahdin.mssm.edu/siRNA) and was cloned in a modified version of TtRMPVIR inducible retroviral vector (Genbank HQ456318) in which the puromycin resistance gene was inserted. The empty vector, containing an sh-Renilla sequence, was used as control. Chromatin immuno-precipitations of the transcription factor PU.1 binding to in vitro reconstituted chromatin followed by multiparallel sequencing Micrococcal nuclease digestion of chromatin extracted from bone marrow derived macrophages followed by multiparallel sequencing . Experiments were carried out in untreated cells (4 replicates) and cells infected either with a retroviral vector containing a short hairpin targeting Pu.1 or with the empty vector as control (2 replicates). The shPU.1 hairpin (sequence available upon request) was selected among five designed using a publicly available software (http://katahdin.mssm.edu/siRNA) and was cloned in a modified version of TtRMPVIR inducible retroviral vector (Genbank HQ456318) in which the puromycin resistance gene was inserted. The empty vector, containing an sh-Renilla sequence, was used as control. Micrococcal nuclease digestion of in vitro reconstituted chromatin followed by multiparallel sequencing
Project description:Precise nucleosome positioning is an increasingly recognized feature of promoters and enhancers, reflecting complex contributions of DNA sequence, nucleosome positioning, histone modification and transcription factor binding to enhancer activity and regulation of gene expression. Changes in nucleosome position and occupancy, histone variants and modifications, and chromatin remodeling are also critical elements of dynamic transcriptional regulation, but poorly understood at enhancers. We investigated glucocorticoid receptor-associated (GR) nucleosome dynamics at enhancers in acute lymphoblastic leukemia. For the first time, we demonstrate functionally distinct modes of nucleosome remodeling upon chromatin binding by GR, which we term central, non-central, phased, and minimal. Central and non-central remodeling reflect nucleosome eviction by GR and cofactors, respectively. Phased remodeling involves nucleosome repositioning and is associated with rapidly activated enhancers and induction of gene expression. Minimal remodeling sites initially have low levels of enhancerassociated histone modification, but the majority of these regions gain H3K4me2 or H3K27Ac to become de novo enhancers. Minimal remodeling regions are associated with gene ontologies specific to decreased B cell number and mTOR inhibition and may make unique contributions to glucocorticoid-induced leukemia cell death. Our findings form a novel framework for understanding the dynamic interplay between transcription factor binding, nucleosome remodeling, enhancer function, and gene expression in the leukemia response to glucocorticoids.
Project description:We report the comparative investigation of genome-wide chromatin state maps, transcription factor (TF) occupancy, and gene expression profiles from developing red cell precursors at two developmental stages. Contrasting the similarities and differences between fetal and adult erythropoiesis provides important insights into the erythroid gene expression programs and gene regulatory networks. Specifically, comparative analyses of human erythropoiesis identify developmental stage-specific enhancers as primary determinants of stage-specific gene expression programs. We find that master regulators, such as GATA1 and TAL1, cooperatively act within active enhancers but have little predictive value for stage-specific enhancer activity. Instead, a set of stage-specific co-regulators collaborates with master regulators and contributes to differential gene expression. We further identify and validate IRF2, IRF6, and MYB as effectors of adult-stage expression program. Thus, the combinatorial assembly of master regulators and transcriptional co-regulators at developmental stage-specific enhancers controls gene expression programs and temporal regulation of transcriptional networks in a mammalian genome. Examination of various histone modifications and transcription factor occupancy by ChIP-seq in fetal and adult proerythroblasts.
Project description:It is widely believed that reorganization of nucleosomes result in availability of binding sites that engage transcription factors during eukaryotic gene regulation. Recent findings, on the other hand, suggest that transcription factors induced as a result of physiological perturbations directly (or in association with chromatin modifiers) may alter nucleosome occupancy to facilitate DNA binding. Although, together these suggest a close relationship between transcription factor binding and nucleosome reorganization, the nature of the inter-dependency, or to what extent it influences regulatory transcription is not clear. Moreover, since most studies used physiolgical pertubations that induced multiple transcription factor chromatin modifiers, the relatively local (or direct) effect of transcription factor binding on nucleosome occupancy remains unclear. With these in mind, we used a single transcription factor to induce physiological changes, representing metastatic (aggressive cancer) and the corresponding non-metastatic state, in human cancer cells. Following characterization of the two states (before and after induction of the transcription factor) we determined: (a) genome wide binding sites of the transcription factor, (b) promoter nucleosome occupancy and (c) transcriptome profiles, independently in both conditions. Interestingly, we find only ~20% of TF binding results from nucleosome reorganization - however, almost all corresponding genes were transcriptionally altered. Whereas, in cases where TF-occupancy was independent of nucleosome repositioning (in close vicinity), or co-occurred with nucleosomes, only a small fraction of the corresponding genes were expressed/repressed. Together, these indicate a model where TF occupancy only when coupled with nucleosome repositioning in close proximity is transcriptionally active. This, to our knowledge, for the first time also helps explain why genome wide TF occupancy (e.g., from ChIP-seq) is typically associated with only a small fraction of genes that change expression. For expression profiling of cells in NME2-induced conditions, A549 cells were transfected with pcDNA-NME2-MYC or pcDNA-MYC (control). RNA was isolated from the cells 48h after transfection using the trizol method (Sigma) as per manufacturerM-bM-^@M-^Ys protocol. Total RNA was processed to hybridize to Illumina Human HT-12 v4 Expression BeadChip as per manufacturerM-bM-^@M-^Ys instructions. Three biological replicates were averaged and data was analyzed using BeadStudio (P <0.05 of fold change).
Project description:It is widely believed that reorganization of nucleosomes result in availability of binding sites that engage transcription factors during eukaryotic gene regulation. Recent findings, on the other hand, suggest that transcription factors induced as a result of physiological perturbations directly (or in association with chromatin modifiers) may alter nucleosome occupancy to facilitate DNA binding. Although, together these suggest a close relationship between transcription factor binding and nucleosome reorganization, the nature of the inter-dependency, or to what extent it influences regulatory transcription is not clear. Moreover, since most studies used physiolgical pertubations that induced multiple transcription factor chromatin modifiers, the relatively local (or direct) effect of transcription factor binding on nucleosome occupancy remains unclear. With these in mind, we used a single transcription factor to induce physiological changes, representing metastatic (aggressive cancer) and the corresponding non-metastatic state, in human cancer cells. Following characterization of the two states (before and after induction of the transcription factor) we determined: (a) genome wide binding sites of the transcription factor, (b) promoter nucleosome occupancy and (c) transcriptome profiles, independently in both conditions. Interestingly, we find only ~20% of TF binding results from nucleosome reorganization - however, almost all corresponding genes were transcriptionally altered. Whereas, in cases where TF-occupancy was independent of nucleosome repositioning (in close vicinity), or co-occurred with nucleosomes, only a small fraction of the corresponding genes were expressed/repressed. Together, these indicate a model where TF occupancy only when coupled with nucleosome repositioning in close proximity is transcriptionally active. This, to our knowledge, for the first time also helps explain why genome wide TF occupancy (e.g., from ChIP-seq) is typically associated with only a small fraction of genes that change expression.
Project description:Precise nucleosome positioning is an increasingly recognized feature of promoters and enhancers, reflecting complex contributions of DNA sequence, nucleosome positioning, histone modification and transcription factor binding to enhancer activity and regulation of gene expression. Changes in nucleosome position and occupancy, histone variants and modifications, and chromatin remodeling are also critical elements of dynamic transcriptional regulation, but poorly understood at enhancers. We investigated glucocorticoid receptor-associated (GR) nucleosome dynamics at enhancers in acute lymphoblastic leukemia. For the first time, we demonstrate functionally distinct modes of nucleosome remodeling upon chromatin binding by GR, which we term central, non-central, phased, and minimal. Central and non-central remodeling reflect nucleosome eviction by GR and cofactors, respectively. Phased remodeling involves nucleosome repositioning and is associated with rapidly activated enhancers and induction of gene expression. Minimal remodeling sites initially have low levels of enhancerassociated histone modification, but the majority of these regions gain H3K4me2 or H3K27Ac to become de novo enhancers. Minimal remodeling regions are associated with gene ontologies specific to decreased B cell number and mTOR inhibition and may make unique contributions to glucocorticoid-induced leukemia cell death. Our findings form a novel framework for understanding the dynamic interplay between transcription factor binding, nucleosome remodeling, enhancer function, and gene expression in the leukemia response to glucocorticoids. We used microarrays to measure changes in gene expression following treatment of cells with dexamethasone for 0, 8, or 24 hours.
Project description:The ability to measure epigenetic features, such as histone modifications and occupancy by transcription factors and co-activators, on a genome-wide scale is advancing the accuracy of CRM predictions. While integration of signals from multiple features is expected to improve predictions, the contribution of each feature to prediction accuracy is not known. We began with predictions of 4,915 erythroid enhancers based on genomic occupancy by TAL1, a key hematopoietic transcription factor that is strongly associated with gene induction in erythroid cells. Seventy of these DNA segments occupied by TAL1 (TAL1 OSs) were tested by transient transfections of cultured hematopoietic cells, and 56% of these were active as enhancers. Sixty-six TAL1 OSs were evaluated in transgenic mouse embryos, and 65% of these were active enhancers in various tissues. Inclusion of additional epigenetic features improved the prediction accuracy, with combinations of TAL1, GATA1, EP300, H3K4me1, and H3K27ac giving high accuracy of enhancer prediction (70%-75% success depending on method of clustering) while maintaining good sensitivity and specificity. Motifs that distinguish active from inactive TAL1 OSs implicate IRFs, STATs, and FOX protein families as candidate positive co-factors with TAL1, while REST (NRSF) and HOX family proteins are implicated in inactivity. While signals for evolutionary constraint were weak over the entire TAL1-bound DNA segments regardless of activity in either assay, phylogenetic preservation of a TF-binding site motif was associated with enhancer activity. The contribution of 8 epigenetic features including H3K27ac to identification of enhancers in 24h-induced G1E-ER4 cells.
Project description:The binding patterns of some transcription factors have been shown to diverge substantially between closely related species. Here, we show that the binding pattern of the developmental transcription factor Twist is highly conserved across six Drosophila species, revealing strong functional constraints at developmental enhancers. Conserved binding correlates with sequence motifs for Twist and its partners, permitting the de novo discovery of their cooperative binding. It also includes over 10,000 low-occupancy sites near the detection limit, which tend to mark enhancers of later developmental stages. We predict that conservation, dynamic occupancy, and combinatorial regulation will be generally true for developmental enhancers.