The transcriptional interactome: gene expression in 3D

Genome-wide map of regulatory interactions in the human genome

Increasing evidence suggests that interactions between regulatory genomic elements play an important role in regulating gene expression. We generated a genome-wide interaction map of regulatory elements in human cells (ENCODE tier 1 cells, K562, GM12878) using Chromatin Interaction Analysis by Paired-End Tag sequencing (ChIA-PET) experiments targeting six broadly distributed factors. Bound regions covered 80% of DNase I hypersensitive sites including 99.7% of TSS and 98% of enhancers. Correlating this map with ChIP-seq and RNA-seq data sets revealed cohesin, CTCF, and ZNF143 as key components of three-dimensional chromatin structure and revealed how the distal chromatin state affects gene transcription. Comparison of interactions between cell types revealed that enhancer-promoter interactions were highly cell-type-specific. Construction and comparison of distal and proximal regulatory networks revealed stark differences in structure and biological function. Proximal binding events are enriched at genes with housekeeping functions, while distal binding events interact with genes involved in dynamic biological processes including response to stimulus. This study reveals new mechanistic and functional insights into regulatory region organization in the nucleus.

A functional insulator screen identifies NURF and dREAM components to be required for enhancer-blocking

Chromatin insulators of higher eukaryotes functionally divide the genome into active and inactive domains. Furthermore, insulators regulate enhancer/promoter communication, which is evident from the Drosophila bithorax locus in which a multitude of regulatory elements control segment specific gene activity. Centrosomal protein 190 (CP190) is targeted to insulators by CTCF or other insulator DNA-binding factors. Chromatin analyses revealed that insulators are characterized by open and nucleosome depleted regions. Here, we wanted to identify chromatin modification and remodelling factors required for an enhancer blocking function. We used the well-studied Fab-8 insulator of the bithorax locus to apply a genome-wide RNAi screen for factors that contribute to the enhancer blocking function of CTCF and CP190. Among 78 genes required for optimal Fab-8 mediated enhancer blocking, all four components of the NURF complex as well as several subunits of the dREAM complex were most evident. Mass spectrometric analyses of CTCF or CP190 bound proteins as well as immune precipitation confirmed NURF and dREAM binding. Both co-localise with most CP190 binding sites in the genome and chromatin immune precipitation showed that CP190 recruits NURF and dREAM. Nucleosome occupancy and histone H3 binding analyses revealed that CP190 mediated NURF binding results in nucleosomal depletion at CP190 binding sites. Thus, we conclude that CP190 binding to CTCF or to other DNA binding insulator factors mediates recruitment of NURF and dREAM. Furthermore, the enhancer blocking function of insulators is associated with nucleosomal depletion and requires NURF and dREAM.

Is an observed non-co-linear RNA product spliced in trans, in cis or just in vitro?

Global transcriptome investigations often result in the detection of an enormous number of transcripts composed of non-co-linear sequence fragments. Such ‘aberrant’ transcript products may arise from post-transcriptional events or genetic rearrangements, or may otherwise be false positives (sequencing/alignment errors or in vitro artifacts). Moreover, post-transcriptionally non-co-linear (‘PtNcl’) transcripts can arise from trans-splicing or back-splicing in cis (to generate so-called ‘circular RNA’). Here, we collected previously-predicted human non-co-linear RNA candidates, and designed a validation procedure integrating in silico filters with multiple experimental validation steps to examine their authenticity. We showed that >50% of the tested candidates were in vitro artifacts, even though some had been previously validated by RT-PCR. After excluding the possibility of genetic rearrangements, we distinguished between trans-spliced and circular RNAs, and confirmed that these two splicing forms can share the same non-co-linear junction. Importantly, the experimentally-confirmed PtNcl RNA events and their corresponding PtNcl splicing types (i.e. trans-splicing, circular RNA, or both sharing the same junction) were all expressed in rhesus macaque, and some were even expressed in mouse. Our study thus describes an essential procedure for confirming PtNcl transcripts, and provides further insight into the evolutionary role of PtNcl RNA events, opening up this important, but understudied, class of post-transcriptional events for comprehensive characterization.

Loss of neuronal 3D chromatin organization causes transcriptional and behavioural deficits related to serotonergic dysfunction

The interior of the neuronal cell nucleus is a highly organized three-dimensional (3D) structure where regions of the genome that are linearly millions of bases apart establish sub-structures with specialized functions. To investigate neuronal chromatin organization and dynamics in vivo, we generated bitransgenic mice expressing GFP-tagged histone H2B in principal neurons of the forebrain. Surprisingly, the expression of this chimeric histone in mature neurons caused chromocenter declustering and disrupted the association of heterochromatin with the nuclear lamina. The loss of these structures did not affect neuronal viability but was associated with specific transcriptional and behavioural deficits related to serotonergic dysfunction. Overall, our results demonstrate that the 3D organization of chromatin within neuronal cells provides an additional level of epigenetic regulation of gene expression that critically impacts neuronal function. This in turn suggests that some loci associated with neuropsychiatric disorders may be particularly sensitive to changes in chromatin architecture.

The impact of chromosomal rearrangements on regulation of gene expression

The effects that coding region single-nucleotide polymorphisms or mutations have on gene expression have been well documented, predominantly owing to their association with disease. The effects of structural chromosomal rearrangements are also receiving increasing attention with the development of new techniques that allow accurate, high-resolution data, whether genomic interaction or transcriptome data, to be generated right down to the single-cell level. Over the past 18 months, these advances in experimental techniques have been used to further confirm and delineate the substantial effects that chromosome rearrangements can have on the regulation of gene expression and provide evidence of direct links between the two.

The clustering of CpG islands may constitute an important determinant of the 3D organization of interphase chromosomes

We used the 4C-Seq technique to characterize the genome-wide patterns of spatial contacts of several CpG islands located on chromosome 14 in cultured chicken lymphoid and erythroid cells. We observed a clear tendency for the spatial clustering of CpG islands present on the same and different chromosomes, regardless of the presence or absence of promoters within these CpG islands. Accordingly, we observed preferential spatial contacts between Sp1 binding motifs and other GC-rich genomic elements, including the DNA sequence motifs capable of forming G-quadruplexes. However, an anchor placed in a gene/CpG island-poor area formed spatial contacts with other gene/CpG island-poor areas on chromosome 14 and other chromosomes. These results corroborate the two-compartment model of the spatial organization of interphase chromosomes and suggest that the clustering of CpG islands constitutes an important determinant of the 3D organization of the eukaryotic genome in the cell nucleus. Using the ChIP-Seq technique, we mapped the genome-wide CTCF deposition sites in the chicken lymphoid and erythroid cells that were used for the 4C analysis. We observed a good correlation between the density of CTCF deposition sites and the level of 4C signals for the anchors located in CpG islands but not for an anchor located in a gene desert. It is thus possible that CTCF contributes to the clustering of CpG islands observed in our experiments.

p63 and Brg1 control developmentally regulated higher-order chromatin remodelling at the epidermal differentiation complex locus in epidermal progenitor cells

Chromatin structural states and their remodelling, including higher-order chromatin folding and three-dimensional (3D) genome organisation, play an important role in the control of gene expression. The role of 3D genome organisation in the control and execution of lineage-specific transcription programmes during the development and differentiation of multipotent stem cells into specialised cell types remains poorly understood. Here, we show that substantial remodelling of the higher-order chromatin structure of the epidermal differentiation complex (EDC), a keratinocyte lineage-specific gene locus on mouse chromosome 3, occurs during epidermal morphogenesis. During epidermal development, the locus relocates away from the nuclear periphery towards the nuclear interior into a compartment enriched in SC35-positive nuclear speckles. Relocation of the EDC locus occurs prior to the full activation of EDC genes involved in controlling terminal keratinocyte differentiation and is a lineage-specific, developmentally regulated event controlled by transcription factor p63, a master regulator of epidermal development. We also show that, in epidermal progenitor cells, p63 directly regulates the expression of the ATP-dependent chromatin remodeller Brg1, which binds to distinct domains within the EDC and is required for relocation of the EDC towards the nuclear interior. Furthermore, Brg1 also regulates gene expression within the EDC locus during epidermal morphogenesis. Thus, p63 and its direct target Brg1 play an essential role in remodelling the higher-order chromatin structure of the EDC and in the specific positioning of this locus within the landscape of the 3D nuclear space, as required for the efficient expression of EDC genes in epidermal progenitor cells during skin development.

A Crowdsourced nucleus: understanding nuclear organization in terms of dynamically networked protein function

The spatial organization of the nucleus results in a compartmentalized structure that affects all aspects of nuclear function. This compartmentalization involves genome organization as well as the formation of nuclear bodies and plays a role in many functions, including gene regulation, genome stability, replication, and RNA processing. Here we review the recent findings associated with the spatial organization of the nucleus and reveal that a common theme for nuclear proteins is their ability to participate in a variety of functions and pathways. We consider this multiplicity of function in terms of Crowdsourcing, a recent phenomenon in the world of information technology, and suggest that this model provides a novel way to synthesize the many intersections between nuclear organization and function. This article is part of a Special Issue entitled: Chromatin and epigenetic regulation of animal development.

A high-resolution map of the three-dimensional chromatin interactome in human cells

A large number of cis-regulatory sequences have been annotated in the human genome^1,2, but defining their target genes remains a challenge³. One strategy is to identify the long-range looping interactions at these elements with the use of chromosome conformation capture (3C) based techniques⁴. However, previous studies lack either the resolution or coverage to permit a whole-genome, unbiased view of chromatin interactions. Here, we report a comprehensive chromatin interaction map generated in human fibroblasts using a genome-wide 3C analysis method (Hi-C)⁵. We determined over one million long-range chromatin interactions at 5–10kb resolution, and uncovered general principles of chromatin organization at different types of genomic features. We also characterized the dynamics of promoter-enhancer contacts upon TNF-α signaling in these cells. Unexpectedly, we found that TNF-α responsive enhancers are already in contact with their target promoters prior to signaling. Such pre-existing chromatin looping, which also exists in other cell types with different extra-cellular signaling, is a strong predictor of gene induction. Our observations suggest that the three-dimensional chromatin landscape, once established in a particular cell type, is rather stable and could influence the selection or activation of target genes by a ubiquitous transcription activator in a cell-specific manner.

Integrative transcriptome sequencing identifies trans-splicing events with important roles in human embryonic stem cell pluripotency

Trans-splicing is a post-transcriptional event that joins exons from separate pre-mRNAs. Detection of trans-splicing is usually severely hampered by experimental artifacts and genetic rearrangements. Here, we develop a new computational pipeline, TSscan, which integrates different types of high-throughput long-/short-read transcriptome sequencing of different human embryonic stem cell (hESC) lines to effectively minimize false positives while detecting trans-splicing. Combining TSscan screening with multiple experimental validation steps revealed that most chimeric RNA products were platform-dependent experimental artifacts of RNA sequencing. We successfully identified and confirmed four trans-spliced RNAs, including the first reported trans-spliced large intergenic noncoding RNA (“tsRMST”). We showed that these trans-spliced RNAs were all highly expressed in human pluripotent stem cells and differentially expressed during hESC differentiation. Our results further indicated that tsRMST can contribute to pluripotency maintenance of hESCs by suppressing lineage-specific gene expression through the recruitment of NANOG and the PRC2 complex factor, SUZ12. Taken together, our findings provide important insights into the role of trans-splicing in pluripotency maintenance of hESCs and help to facilitate future studies into trans-splicing, opening up this important but understudied class of post-transcriptional events for comprehensive characterization.

Cycles in spatial and temporal chromosomal organization driven by the circadian clock

Dynamic transitions in the epigenome have been associated with regulated patterns of nuclear organization. The accumulating evidence that chromatin remodeling is implicated in circadian function prompted us to explore whether the clock may control nuclear architecture. We applied the chromosome conformation capture on chip technology in mouse embryonic fibroblasts (MEFs) to demonstrate the presence of circadian long-range interactions using the clock-controlled Dbp gene as bait. The circadian genomic interactions with Dbp were highly specific and were absent in MEFs whose clock was disrupted by ablation of the Bmal1 gene (also called Arntl). We establish that the Dbp circadian interactome contains a wide variety of genes and clock-related DNA elements. These findings reveal a previously unappreciated circadian and clock-dependent shaping of the nuclear landscape.

Dynamic trans interactions in yeast chromosomes

Three-dimensional organization of the genome is important for regulation of gene expression and maintenance of genomic stability. It also defines, and is defined by, contacts between different chromosomal loci. Interactions between loci positioned on different chromosomes, i.e. "trans" interactions are one type of such contacts. Here, we describe a case of inducible trans interaction in chromosomes of the budding yeast S. cerevisiae. Special DNA sequences, inserted in two ectopic chromosomal loci positioned in trans, pair with one another in an inducible manner. The spatial proximity diagnostic of pairing is observable by both chromosome capture analysis (3C) and epifluorescence microscopy in whole cells. Protein synthesis de novo appears to be required for this process. The three-dimensional organization of the yeast nucleus imposes a constraint on such pairing, presumably by dictating the probability with which the two sequences collide with one another.

Overlapping gene coexpression patterns in human medullary thymic epithelial cells generate self-antigen diversity

Promiscuous expression of numerous tissue-restricted self-antigens (TRAs) in medullary thymic epithelial cells (mTECs) is essential to safeguard self-tolerance. A distinct feature of promiscuous gene expression is its mosaic pattern (i.e., at a given time, each self-antigen is expressed only in 1-3% of mTECs). How this mosaic pattern is generated at the single-cell level is currently not understood. Here, we show that subsets of human mTECs expressing a particular TRA coexpress distinct sets of genes. We identified three coexpression groups comprising overlapping and complementary gene sets, which preferentially mapped to certain chromosomes and intrachromosomal gene clusters. Coexpressed gene loci tended to colocalize to the same nuclear subdomain. The TRA subsets aligned along progressive differentiation stages within the mature mTEC subset and, in vitro, interconverted along this sequence. Our data suggest that single mTECs shift through distinct gene pools, thus scanning a sizeable fraction of the overall repertoire of promiscuously expressed self-antigens. These findings have implications for the temporal and spatial (re)presentation of self-antigens in the medulla in the context of tolerance induction.

Chromatin loop organization of the junb locus in mouse dendritic cells

The junb gene behaves as an immediate early gene in bacterial lipopolysaccharide (LPS)-stimulated dendritic cells (DCs), where its transient transcriptional activation is necessary for the induction of inflammatory cytokines. junb is a short gene and its transcriptional activation by LPS depends on the binding of NF-κB to an enhancer located just downstream of its 3′ UTR. Here, we have addressed the mechanisms underlying the transcriptional hyper-reactivity of junb. Using transfection and pharmacological assays to complement chromatin immunoprecipitation analyses addressing the localization of histones, polymerase II, negative elongation factor (NELF)-, DRB sensitivity-inducing factor (DSIF)- and Positive Transcription Factor b complexes, we demonstrate that junb is a RNA Pol II-paused gene where Pol II is loaded in the transcription start site domain but poorly active. Moreover, High salt-Recovered Sequence, chromosome conformation capture (3C)- and gene transfer experiments show that (i) junb is organized in a nuclear chromatin loop bringing into close spatial proximity the upstream promoter region and the downstream enhancer and (ii) this configuration permits immediate Pol II release on the junb body on binding of LPS-activated NF-κB to the enhancer. Thus, our work unveils a novel topological framework underlying fast junb transcriptional response in DCs. Moreover, it also points to a novel layer of complexity in the modes of action of NF-κB.

DNase I-hypersensitive exons colocalize with promoters and distal regulatory elements

The precise splicing of genes confers an enormous transcriptional complexity to the human genome. The majority of gene splicing occurs cotranscriptionally, permitting epigenetic modifications to affect splicing outcomes. Here we show that select exonic regions are demarcated within the three-dimensional structure of the human genome. We identify a subset of exons that exhibit DNase I hypersensitivity and are accompanied by 'phantom' signals in chromatin immunoprecipitation and sequencing (ChIP-seq) that result from cross-linking with proximal promoter- or enhancer-bound factors. The capture of structural features by ChIP-seq is confirmed by chromatin interaction analysis that resolves local intragenic loops that fold exons close to cognate promoters while excluding intervening intronic sequences. These interactions of exons with promoters and enhancers are enriched for alternative splicing events, an effect reflected in cell type-specific periexonic DNase I hypersensitivity patterns. Collectively, our results connect local genome topography, chromatin structure and cis-regulatory landscapes with the generation of human transcriptional complexity by cotranscriptional splicing.

High-throughput identification of long-range regulatory elements and their target promoters in the human genome

Enhancer elements are essential for tissue-specific gene regulation during mammalian development. Although these regulatory elements are often distant from their target genes, they affect gene expression by recruiting transcription factors to specific promoter regions. Because of this long-range action, the annotation of enhancer element-target promoter pairs remains elusive. Here, we developed a novel analysis methodology that takes advantage of Hi-C data to comprehensively identify these interactions throughout the human genome. To do this, we used a geometric distribution-based model to identify DNA-DNA interaction hotspots that contact gene promoters with high confidence. We observed that these promoter-interacting hotspots significantly overlap with known enhancer-associated histone modifications and DNase I hypersensitive sites. Thus, we defined thousands of candidate enhancer elements by incorporating these features, and found that they have a significant propensity to be bound by p300, an enhancer binding transcription factor. Furthermore, we revealed that their target genes are significantly bound by RNA Polymerase II and demonstrate tissue-specific expression. Finally, we uncovered that these elements are generally found within 1 Mb of their targets, and often regulate multiple genes. In total, our study presents a novel high-throughput workflow for confident, genome-wide discovery of enhancer-target promoter pairs, which will significantly improve our understanding of these regulatory interactions.

In vivo live imaging of RNA polymerase II transcription factories in primary cells

Transcription steps are marked by different modifications of the C-terminal domain of RNA polymerase II (RNAPII). Phosphorylation of Ser5 and Ser7 by cyclin-dependent kinase 7 (CDK7) as part of TFIIH marks initiation, whereas phosphorylation of Ser2 by CDK9 marks elongation. These processes are thought to take place in localized transcription foci in the nucleus, known as "transcription factories," but it has been argued that the observed clusters/foci are mere fixation or labeling artifacts. We show that transcription factories exist in living cells as distinct foci by live-imaging fluorescently labeled CDK9, a kinase known to associate with active RNAPII. These foci were observed in different cell types derived from CDK9-mCherry knock-in mice. We show that these foci are very stable while highly dynamic in exchanging CDK9. Chromatin immunoprecipitation (ChIP) coupled with deep sequencing (ChIP-seq) data show that the genome-wide binding sites of CDK9 and initiating RNAPII overlap on transcribed genes. Immunostaining shows that CDK9-mCherry foci colocalize with RNAPII-Ser5P, much less with RNAPII-Ser2P, and not with CDK12 (a kinase reported to be involved in the Ser2 phosphorylation) or with splicing factor SC35. In conclusion, transcription factories exist in living cells, and initiation and elongation of transcripts takes place in different nuclear compartments.

Novel higher-order epigenetic regulation of the Bdnf gene upon seizures

Studies in cultured cells have demonstrated the existence of higher-order epigenetic mechanisms, determining the relationship between expression of the gene and its position within the cell nucleus. It is unknown, whether such mechanisms operate in postmitotic, highly differentiated cell types, such as neurons in vivo. Accordingly, we examined whether the intranuclear positions of Bdnf and Trkb genes, encoding the major neurotrophin and its receptor respectively, change as a result of neuronal activity, and what functional consequences such movements may have. In a rat model of massive neuronal activation upon kainate-induced seizures we found that elevated neuronal expression of Bdnf is associated with its detachment from the nuclear lamina, and translocation toward the nucleus center. In contrast, the position of stably expressed Trkb remains unchanged after seizures. Our study demonstrates that activation-dependent architectural remodeling of the neuronal cell nucleus in vivo contributes to activity-dependent changes in gene expression in the brain.

DNA looping facilitates targeting of a chromatin remodeling enzyme

ATP-dependent chromatin remodeling enzymes are highly abundant and play pivotal roles regulating DNA-dependent processes. The mechanisms by which they are targeted to specific loci have not been well understood on a genome-wide scale. Here, we present evidence that a major targeting mechanism for the Isw2 chromatin remodeling enzyme to specific genomic loci is through sequence-specific transcription factor (TF)-dependent recruitment. Unexpectedly, Isw2 is recruited in a TF-dependent fashion to a large number of loci without TF binding sites. Using the 3C assay, we show that Isw2 can be targeted by Ume6- and TFIIB-dependent DNA looping. These results identify DNA looping as a mechanism for the recruitment of a chromatin remodeling enzyme and define a function for DNA looping. We also present evidence suggesting that Ume6-dependent DNA looping is involved in chromatin remodeling and transcriptional repression, revealing a mechanism by which the three-dimensional folding of chromatin affects DNA-dependent processes.

Remodeling of three-dimensional organization of the nucleus during terminal keratinocyte differentiation in the epidermis

The nucleus of epidermal keratinocytes (KCs) is a complex and highly compartmentalized organelle, whose structure is markedly changed during terminal differentiation and transition of the genome from a transcriptionally active state seen in the basal and spinous epidermal cells to a fully inactive state in the keratinized cells of the cornified layer. Here, using multicolor confocal microscopy, followed by computational image analysis and mathematical modeling, we demonstrate that in normal mouse footpad epidermis, transition of KCs from basal epidermal layer to the granular layer is accompanied by marked differences in nuclear architecture and microenvironment including the following: (i) decrease in the nuclear volume; (ii) decrease in expression of the markers of transcriptionally active chromatin; (iii) internalization and decrease in the number of nucleoli; (iv) increase in the number of pericentromeric heterochromatic clusters; and (v) increase in the frequency of associations between the pericentromeric clusters, chromosomal territory 3, and nucleoli. These data suggest a role for nucleoli and pericentromeric heterochromatin clusters as organizers of nuclear microenvironment required for proper execution of gene expression programs in differentiating KCs, and provide important background information for further analyses of alterations in the topological genome organization seen in pathological skin conditions, including disorders of epidermal differentiation and epidermal tumors.

Chromatin structure and organization: the relation with gene expression during development and disease

The elementary level of chromatin fiber, namely the nucleofilament, is known to undergo a hierarchical compaction leading to local chromatin loops, then chromatin domains and ultimately chromosome territories. These successive folding levels rely on the formation of chromatin loops ranging from few kb to some Mb. In addition to a packaging and structural role, the high-order organization of genomes functionally impacts on gene expression program. This review summarises to which extent each level of chromatin compaction does affect gene regulation. In addition, we point out the structural and functional changes observed in diseases. Emphasis will be mainly placed on the large-scale organization of the chromatin.

Nuclear RNA sequencing of the mouse erythroid cell transcriptome

In addition to protein coding genes a substantial proportion of mammalian genomes are transcribed. However, most transcriptome studies investigate steady-state mRNA levels, ignoring a considerable fraction of the transcribed genome. In addition, steady-state mRNA levels are influenced by both transcriptional and posttranscriptional mechanisms, and thus do not provide a clear picture of transcriptional output. Here, using deep sequencing of nuclear RNAs (nucRNA-Seq) in parallel with chromatin immunoprecipitation sequencing (ChIP-Seq) of active RNA polymerase II, we compared the nuclear transcriptome of mouse anemic spleen erythroid cells with polymerase occupancy on a genome-wide scale. We demonstrate that unspliced transcripts quantified by nucRNA-seq correlate with primary transcript frequencies measured by RNA FISH, but differ from steady-state mRNA levels measured by poly(A)-enriched RNA-seq. Highly expressed protein coding genes showed good correlation between RNAPII occupancy and transcriptional output; however, genome-wide we observed a poor correlation between transcriptional output and RNAPII association. This poor correlation is due to intergenic regions associated with RNAPII which correspond with transcription factor bound regulatory regions and a group of stable, nuclear-retained long non-coding transcripts. In conclusion, sequencing the nuclear transcriptome provides an opportunity to investigate the transcriptional landscape in a given cell type through quantification of unspliced primary transcripts and the identification of nuclear-retained long non-coding RNAs.

Large-scale functional organization of long-range chromatin interaction networks

Chromatin interactions play important roles in transcription regulation. To better understand the underlying evolutionary and functional constraints of these interactions, we implemented a systems approach to examine RNA polymerase-II-associated chromatin interactions in human cells. We found that 40% of the total genomic elements involved in chromatin interactions converged to a giant, scale-free-like, hierarchical network organized into chromatin communities. The communities were enriched in specific functions and were syntenic through evolution. Disease-associated SNPs from genome-wide association studies were enriched among the nodes with fewer interactions, implying their selection against deleterious interactions by limiting the total number of interactions, a model that we further reconciled using somatic and germline cancer mutation data. The hubs lacked disease-associated SNPs, constituted a nonrandomly interconnected core of key cellular functions, and exhibited lethality in mouse mutants, supporting an evolutionary selection that favored the nonrandom spatial clustering of the least-evolving key genomic domains against random genetic or transcriptional errors in the genome. Altogether, our analyses reveal a systems-level evolutionary framework that shapes functionally compartmentalized and error-tolerant transcriptional regulation of human genome in three dimensions.

Transcriptional networks driving enhancer function in the CFTR gene

A critical cis-regulatory element for the CFTR (cystic fibrosis transmembrane conductance regulator) gene is located in intron 11, 100 kb distal to the promoter, with which it interacts. This sequence contains an intestine-selective enhancer and associates with enhancer signature proteins, such as p300, in addition to tissue-specific TFs (transcription factors). In the present study we identify critical TFs that are recruited to this element and demonstrate their importance in regulating CFTR expression. In vitro DNase I footprinting and EMSAs (electrophoretic mobility-shift assays) identified four cell-type-selective regions that bound TFs in vitro. ChIP (chromatin immunoprecipitation) identified FOXA1/A2 (forkhead box A1/A2), HNF1 (hepatocyte nuclear factor 1) and CDX2 (caudal-type homeobox 2) as in vivo trans-interacting factors. Mutation of their binding sites in the intron 11 core compromised its enhancer activity when measured by reporter gene assay. Moreover, siRNA (small interfering RNA)-mediated knockdown of CDX2 caused a significant reduction in endogenous CFTR transcription in intestinal cells, suggesting that this factor is critical for the maintenance of high levels of CFTR expression in these cells. The ChIP data also demonstrate that these TFs interact with multiple cis-regulatory elements across the CFTR locus, implicating a more global role in intestinal expression of the gene.

Interchromosomal clustering of active genes at the nuclear pore complex

Genomes are spatially organized on many levels and the positioning of genes within the nucleus contributes to their proper expression. This positioning can also result in the clustering of genes with similar expression patterns, a phenomenon sometimes called "gene kissing." We have found that yeast genes are targeted to the nuclear periphery through interaction of the nuclear pore complex with small, cis-acting "DNA zip codes" in their promoters. Our recent study demonstrated that genes with the same zip codes cluster together at the nuclear periphery. The zip codes were necessary and sufficient to induce interchromosomal clustering. Finally, we identified a transcription factor (Put3) that binds to the GRS I zip code. Put3 binds to GRS I and is required for both GRS I-dependent positioning at the nuclear periphery and interchromosomal clustering of GRS I-targeted genes. We speculate that our findings might provide insight into other types of gene kissing, some of which also require cis-acting DNA sequences and trans-acting proteins.

TFIIIC bound DNA elements in nuclear organization and insulation

tRNA genes (tDNAs) have been known to have barrier insulator function in budding yeast, Saccharomyces cerevisiae, for over a decade. tDNAs also play a role in genome organization by clustering at sites in the nucleus and both of these functions are dependent on the transcription factor TFIIIC. More recently TFIIIC bound sites devoid of pol III, termed Extra-TFIIIC sites (ETC) have been identified in budding yeast and these sites also function as insulators and affect genome organization. Subsequent studies in Schizosaccharomyces pombe showed that TFIIIC bound sites were insulators and also functioned as Chromosome Organization Clamps (COC); tethering the sites to the nuclear periphery. Very recently studies have moved to mammalian systems where pol III genes and their associated factors have been investigated in both mouse and human cells. Short interspersed nuclear elements (SINEs) that bind TFIIIC, function as insulator elements and tDNAs can also function as both enhancer - blocking and barrier insulators in these organisms. It was also recently shown that tDNAs cluster with other tDNAs and with ETCs but not with pol II transcribed genes. Intriguingly, TFIIIC is often found near pol II transcription start sites and it remains unclear what the consequences of TFIIIC based genomic organization are and what influence pol III factors have on pol II transcribed genes and vice versa. In this review we provide a comprehensive overview of the known data on pol III factors in insulation and genome organization and identify the many open questions that require further investigation. This article is part of a Special Issue entitled: Transcription by Odd Pols.

Long-range genomic interactions epigenetically regulate the expression of a cytokine receptor

Current research on the cytokine-mediated signalling towards the polarization and differentiation of a T-helper cell lineage lacks mechanistic insights on the transcriptional regulation of cytokine receptor genes. Here, we propose a new mechanism for the transcriptional regulation of the interferon gamma receptor 1 gene via long-range intrachromosomal interactions with the Ifnγ locus mediated by the protein CTCF. These interactions sustain the monoallelic expression of the differentially methylated IfnγR1 gene and are persistent on blockade of active transcription. Our findings suggest that regulatory elements for a cytokine gene locus can also positively regulate the transcription of its receptor.

Controlling long-range genomic interactions at a native locus by targeted tethering of a looping factor

Chromatin loops juxtapose distal enhancers with active promoters, but their molecular architecture and relationship with transcription remain unclear. In erythroid cells, the locus control region (LCR) and β-globin promoter form a chromatin loop that requires transcription factor GATA1 and the associated molecule Ldb1. We employed artificial zinc fingers (ZF) to tether Ldb1 to the β-globin promoter in GATA1 null erythroblasts, in which the β-globin locus is relaxed and inactive. Remarkably, targeting Ldb1 or only its self-association domain to the β-globin promoter substantially activated β-globin transcription in the absence of GATA1. Promoter-tethered Ldb1 interacted with endogenous Ldb1 complexes at the LCR to form a chromatin loop, causing recruitment and phosphorylation of RNA polymerase II. ZF-Ldb1 proteins were inactive at alleles lacking the LCR, demonstrating that their activities depend on long-range interactions. Our findings establish Ldb1 as a critical effector of GATA1-mediated loop formation and indicate that chromatin looping causally underlies gene regulation.

Molecular logic underlying chromosomal translocations, random or non-random?

Chromosomal translocations serve as essential diagnostic markers and therapeutic targets for leukemia, lymphoma, and many types of solid tumors. Understanding the mechanisms of chromosomal translocation generation has remained a central biological question for decades. Rather than representing a random event, recent studies indicate that chromosomal translocation is a non-random event in a spatially regulated, site-specific, and signal-driven manner, reflecting actions involved in transcriptional activation, epigenetic regulation, three-dimensional nuclear architecture, and DNA damage-repair. In this review, we will focus on the progression toward understanding the molecular logic underlying chromosomal translocation events and implications of new strategies for preventing chromosomal translocations.

Nature and function of insulator protein binding sites in the Drosophila genome

Chromatin insulator elements and associated proteins have been proposed to partition eukaryotic genomes into sets of independently regulated domains. Here we test this hypothesis by quantitative genome-wide analysis of insulator protein binding to Drosophila chromatin. We find distinct combinatorial binding of insulator proteins to different classes of sites and uncover a novel type of insulator element that binds CP190 but not any other known insulator proteins. Functional characterization of different classes of binding sites indicates that only a small fraction act as robust insulators in standard enhancer-blocking assays. We show that insulators restrict the spreading of the H3K27me3 mark but only at a small number of Polycomb target regions and only to prevent repressive histone methylation within adjacent genes that are already transcriptionally inactive. RNAi knockdown of insulator proteins in cultured cells does not lead to major alterations in genome expression. Taken together, these observations argue against the concept of a genome partitioned by specialized boundary elements and suggest that insulators are reserved for specific regulation of selected genes.

Epigenetic regulation of gene expression in keratinocytes

Nucleus is a complex and highly compartmentalized organelle, which organization undergoes major changes during cell differentiation allowing cells to become specialized and fulfill their functions.During terminal differentiation of the epidermal keratinocytes, nucleus undergoes programmed transformation from active status, associated with execution of the genetic programs of cornification and epidermal barrier formation, to fully inactive condition and becomes a part of the keratinized cells of the cornified layer. Tremendous progress achieved within the last two decades in understanding the biology of the nucleus and epigenetic mechanisms controlling gene expression allowed defining several levels in the regulation of cell differentiation-associated gene expression programs, including an accessibility of the gene regulatory regions to DNA-protein interactions, covalent DNA and histone modifications and ATP-dependent chromatin remodeling, as well as higher-order chromatin remodeling and nuclear compartmentalization of the genes and transcription machinery. Here, we integrate our current knowledge of the mechanisms controlling gene expression during terminal keratinocyte differentiation with distinct levels of chromatin organization and remodeling. We also propose the directions to further explore the role of epigenetic mechanisms and their interactions with other regulatory systems in the control of keratinocyte differentiation in normal and diseased skin.

Pairing of homologous regions in the mouse genome is associated with transcription but not imprinting status

Although somatic homologous pairing is common in Drosophila it is not generally observed in mammalian cells. However, a number of regions have recently been shown to come into close proximity with their homologous allele, and it has been proposed that pairing might be involved in the establishment or maintenance of monoallelic expression. Here, we investigate the pairing properties of various imprinted and non-imprinted regions in mouse tissues and ES cells. We find by allele-specific 4C-Seq and DNA FISH that the Kcnq1 imprinted region displays frequent pairing but that this is not dependent on monoallelic expression. We demonstrate that pairing involves larger chromosomal regions and that the two chromosome territories come close together. Frequent pairing is not associated with imprinted status or DNA repair, but is influenced by chromosomal location and transcription. We propose that homologous pairing is not exclusive to specialised regions or specific functional events, and speculate that it provides the cell with the opportunity of trans-allelic effects on gene regulation.

The prolactin gene: a paradigm of tissue-specific gene regulation with complex temporal transcription dynamics

Transcription of numerous mammalian genes is highly pulsatile, with bursts of expression occurring with variable duration and frequency. The presence of this stochastic or 'noisy' expression pattern has been relatively unexplored in tissue systems. The prolactin gene provides a model of tissue-specific gene regulation resulting in pulsatile transcription dynamics in both cell lines and endocrine tissues. In most cell culture models, prolactin transcription appears to be highly variable between cells, with differences in transcription pulse duration and frequency. This apparently stochastic transcription is constrained by a transcriptional refractory period, which may be related to cycles of chromatin remodelling. We propose that prolactin transcription dynamics result from the summation of oscillatory cellular inputs and by regulation through chromatin remodelling cycles. Observations of transcription dynamics in cells within pituitary tissue show reduced transcriptional heterogeneity and can be grouped into a small number of distinct patterns. Thus, it appears that the tissue environment is able to reduce transcriptional noise to enable coordinated tissue responses to environmental change. We review the current knowledge on the complex tissue-specific regulation of the prolactin gene in pituitary and extra-pituitary sites, highlighting differences between humans and rodent experimental animal models. Within this context, we describe the transcription dynamics of prolactin gene expression and how this may relate to specific processes occurring within the cell.

Epigenetic alterations in muscular disorders

Epigenetic mechanisms, acting via chromatin organization, fix in time and space different transcriptional programs and contribute to the quality, stability, and heritability of cell-specific transcription programs. In the last years, great advances have been made in our understanding of mechanisms by which this occurs in normal subjects. However, only a small part of the complete picture has been revealed. Abnormal gene expression patterns are often implicated in the development of different diseases, and thus epigenetic studies from patients promise to fill an important lack of knowledge, deciphering aberrant molecular mechanisms at the basis of pathogenesis and diseases progression. The identification of epigenetic modifications that could be used as targets for therapeutic interventions could be particularly timely in the light of pharmacologically reversion of pathological perturbations, avoiding changes in DNA sequences. Here I discuss the available information on epigenetic mechanisms that, altered in neuromuscular disorders, could contribute to the progression of the disease.

Did the modulation of expression noise shape the evolution of three dimensional genome organizations in eukaryotes?

The pervasive role of distant chromatin interactions in transcriptional regulation is increasingly becoming evident. There is a possibility that the greater diversity in chromatin interactions of a genomic locus could contribute to stochastic variation in its gene expression. However, this issue has not been addressed. Here, I present a few lines of evidence, which suggest that the variation in trans chromatin interactions might occur at the cost of expression noise. Genomic regions with nucleosome depletion, abundant and rapid transcription and with essential gene clusters exhibit relatively fewer trans chromatin interactions in the nucleus. Moreover, loci with greater number of interactions tend to show higher expression noise. Based on these observations, I hypothesize that the three dimensional organization of eukaryotic genomes might have evolved under a selective pressure to minimize the expression noise of essential gene clusters in the nucleus.

Internucleosomal interactions mediated by histone tails allow distant communication in chromatin

Action across long distances on chromatin is a hallmark of eukaryotic transcriptional regulation. Although chromatin structure per se can support long-range interactions, the mechanisms of efficient communication between widely spaced DNA modules in chromatin remain a mystery. The molecular simulations described herein suggest that transient binary internucleosomal interactions can mediate distant communication in chromatin. Electrostatic interactions between the N-terminal tails of the core histones and DNA enhance the computed probability of juxtaposition of sites that lie far apart along the DNA sequence. Experimental analysis of the rates of communication in chromatin constructs confirms that long-distance communication occurs efficiently and independently of distance on tail-containing, but not on tailless, chromatin. Taken together, our data suggest that internucleosomal interactions involving the histone tails are essential for highly efficient, long-range communication between regulatory elements and their targets in eukaryotic genomes.

Epigenome sequencing comes of age in development, differentiation and disease mechanism research

In eukaryotic organisms, changes in cell phenotype are tightly associated with dynamic changes in the epigenome. Over the past few years, sequencing-based genome-wide approaches to generate, analyze, interpret and integrate epigenetic information have been applied to investigate the mechanisms behind the changes in cell status, such as those which are seen in differentiation, disease and reprogramming. This article focuses on the four types of epigenomic information (i.e., nucleosome positioning, histone modification, DNA methylation and chromatin higher-order structure). We summarize the distinct high-throughput sequencing applications used to generate the different types of epigenomic profiles and the bioinformatic software available for performing routine analysis. With the dramatic improvement of sequencing technology and bioinformatic analysis, epigenome sequencing has gradually become the common approach to study a variety of biological issues.

Specific changes in the expression of imprinted genes in prostate cancer--implications for cancer progression and epigenetic regulation

Epigenetic dysregulation comprising DNA hypermethylation and hypomethylation, enhancer of zeste homologue 2 (EZH2) overexpression and altered patterns of histone modifications is associated with the progression of prostate cancer. DNA methylation, EZH2 and histone modifications also ensure the parental-specific monoallelic expression of at least 62 imprinted genes. Although it is therefore tempting to speculate that epigenetic dysregulation may extend to imprinted genes, expression changes in cancerous prostates are only well documented for insulin-like growth factor 2 (IGF2). A literature and database survey on imprinted genes in prostate cancer suggests that the expression of most imprinted genes remains unchanged despite global disturbances in epigenetic mechanisms. Instead, selective genetic and epigenetic changes appear to lead to the inactivation of a sub-network of imprinted genes, which might function in the prostate to limit cell growth induced via the PI3K/Akt pathway, modulate androgen responses and regulate differentiation. Whereas dysregulation of IGF2 may constitute an early change in prostate carcinogenesis, inactivation of this imprinted gene network is rather associated with cancer progression.

Nuclear retinoic acid receptors: conductors of the retinoic acid symphony during development

The vitamin A derivative, retinoic acid (RA), is essential for embryonic development through the activation of cognate nuclear receptors, RARs, which work as ligand dependent regulators of transcription. In vitro studies revealed how RARs control gene expression at the molecular level and now it appears that it is fine-tuned by a phosphorylation code. In addition, several genetic approaches provided valuable insights on the functions of RARs during development and on the influence of other actors such as the enzymes involved in RA synthesis and degradation and other signaling pathways. It appears that RARs are the conductors of the RA signaling symphony through controlling the dynamics and the coordination of the different players and development steps.

Rapid recent human evolution and the accumulation of balanced genetic polymorphisms

All evolutionary change can be traced to alterations in allele frequencies in populations over time. DNA sequencing on a massive scale now permits us to follow the genetic consequences as our species has diverged from our close relatives and as we have colonized different parts of the world and adapted to them. But it has been difficult to disentangle natural selection from many other factors that alter frequencies. These factors include mutation and intragenic reciprocal recombination, gene conversion, segregation distortion, random drift, and gene flow between populations (these last two are greatly influenced by splits and coalescences of populations over time). The first part of this review examines recent studies that have had some success in dissecting out the role of natural selection, especially in humans and Drosophila. Among many examples, these studies include those that have followed the rapid evolution of traits that may permit adaptation to high altitude in Tibetan and Andean populations. In some cases, directional selection has been so strong that it may have swept alleles close to fixation in the span of a few thousand years, a rapidity of change that is also sometimes encountered in other organisms. The second part of the review summarizes data showing that remarkably few alleles have been carried completely to fixation during our recent evolution. Some of the alleles that have not reached fixation may be approaching new internal equilibria, which would indicate polymorphisms that are maintained by balancing selection. Finally, the review briefly examines why genetic polymorphisms, particularly those that are maintained by negative frequency dependence, are likely to have played an important role in the evolution of our species. A method is suggested for measuring the contribution of these polymorphisms to our gene pool. Such polymorphisms may add to the ability of our species to adapt to our increasingly complex and challenging environment.

Innate structure of DNA foci restricts the mixing of DNA from different chromosome territories

The distribution of chromatin within the mammalian nucleus is constrained by its organization into chromosome territories (CTs). However, recent studies have suggested that promiscuous intra- and inter-chromosomal interactions play fundamental roles in regulating chromatin function and so might define the spatial integrity of CTs. In order to test the extent of DNA mixing between CTs, DNA foci of individual CTs were labeled in living cells following incorporation of Alexa-488 and Cy-3 conjugated replication precursor analogues during consecutive cell cycles. Uniquely labeled chromatin domains, resolved following random mitotic segregation, were visualized as discrete structures with defined borders. At the level of resolution analysed, evidence for mixing of chromatin from adjacent domains was only apparent within the surface volumes where neighboring CTs touched. However, while less than 1% of the nuclear volume represented domains of inter-chromosomal mixing, the dynamic plasticity of DNA foci within individual CTs allows continual transformation of CT structure so that different domains of chromatin mixing evolve over time. Notably, chromatin mixing at the boundaries of adjacent CTs had little impact on the innate structural properties of DNA foci. However, when TSA was used to alter the extent of histone acetylation changes in chromatin correlated with increased chromatin mixing. We propose that DNA foci maintain a structural integrity that restricts widespread mixing of DNA and discuss how the potential to dynamically remodel genome organization might alter during cell differentiation.

p63 regulates Satb1 to control tissue-specific chromatin remodeling during development of the epidermis

During development, multipotent progenitor cells establish tissue-specific programs of gene expression. In this paper, we show that p63 transcription factor, a master regulator of epidermal morphogenesis, executes its function in part by directly regulating expression of the genome organizer Satb1 in progenitor cells. p63 binds to a proximal regulatory region of the Satb1 gene, and p63 ablation results in marked reduction in the Satb1 expression levels in the epidermis. Satb1(-/-) mice show impaired epidermal morphology. In Satb1-null epidermis, chromatin architecture of the epidermal differentiation complex locus containing genes associated with epidermal differentiation is altered primarily at its central domain, where Satb1 binding was confirmed by chromatin immunoprecipitation-on-chip analysis. Furthermore, genes within this domain fail to be properly activated upon terminal differentiation. Satb1 expression in p63(+/-) skin explants treated with p63 small interfering ribonucleic acid partially restored the epidermal phenotype of p63-deficient mice. These data provide a novel mechanism by which Satb1, a direct downstream target of p63, contributes in epidermal morphogenesis via establishing tissue-specific chromatin organization and gene expression in epidermal progenitor cells.

Throwing transcription for a loop: expression of the genome in the 3D nucleus

The functional output of the genome is closely dependent on its organization within the nucleus, which ranges from the 10-nm chromatin fiber to the three-dimensional arrangement of this fiber in the nuclear space. Recent observations suggest that intra- and inter-chromosomal interactions between distant sequences underlie several aspects of transcription regulatory processes. These contacts can bring enhancers close to their target genes or prevent inappropriate interactions between regulatory sequences via insulators. In addition, intra- and inter-chromosomal interactions can bring co-activated or co-repressed genes to the same nuclear location. Recent technological advances have made it possible to map long-range cis and trans interactions at relatively high resolution. This information is being used to develop three-dimensional maps of the arrangement of the genome in the nucleus and to understand causal relationships between nuclear structure and function.

Human tRNA genes function as chromatin insulators

Insulators help separate active chromatin domains from silenced ones. In yeast, gene promoters act as insulators to block the spread of Sir and HP1 mediated silencing while in metazoans most insulators are multipartite autonomous entities. tDNAs are repetitive sequences dispersed throughout the human genome and we now show that some of these tDNAs can function as insulators in human cells. Using computational methods, we identified putative human tDNA insulators. Using silencer blocking, transgene protection and repressor blocking assays we show that some of these tDNA-containing fragments can function as barrier insulators in human cells. We find that these elements also have the ability to block enhancers from activating RNA pol II transcribed promoters. Characterization of a putative tDNA insulator in human cells reveals that the site possesses chromatin signatures similar to those observed at other better-characterized eukaryotic insulators. Enhanced 4C analysis demonstrates that the tDNA insulator makes long-range chromatin contacts with other tDNAs and ETC sites but not with intervening or flanking RNA pol II transcribed genes.

Copy-number variations, noncoding sequences, and human phenotypes

Whereas single-nucleotide polymorphisms and their role in predisposition to disease have been studied extensively, the analysis of structural variants--genomic changes such as insertions, deletions, inversions, duplications, and translocations--is still in its infancy. Changes in copy number, also known as copy-number variations (CNVs), constitute one such group of these structural variants. CNVs are structural genomic variants that arise from deletions (loss) or duplications (gain), and as a consequence result in a copy-number change of the respective genomic region. CNVs may include entire genes or regions of transcribed sequence, or, indeed, comprise only nontranscribed sequences. Whereas the duplication or deletion of a gene can be expected to have an effect on gene dosage, the consequences of CNVs in nontranscribed sequences are less obvious. Here we review CNVs that involve regulatory nontranscribed regions of the genome, describe the associated human phenotypes, and discuss possible disease mechanisms.

Nuclear organisation of sperm remains remarkably unaffected in the presence of defective spermatogenesis

Organisation of chromosome territories in interphase nuclei has been studied in many systems and positional alterations have been associated with disease phenotypes (e.g. laminopathies, cancer) in somatic cells. Altered nuclear organisation is also reported in developmental processes such as mammalian spermatogenesis where a "chromocentre" model is proposed with the centromeres and sex chromosomes repositioning to the nuclear centre. The purpose of this study was to test the hypothesis that alterations in nuclear organisation of human spermatozoa are associated with defects upstream in spermatogenesis (as manifest in certain infertility phenotypes). The nuclear address of (peri-) centromeric loci for 18 chromosomes (1-4, 6-12, 15-18, 20, X and Y) was assayed in 20 males using established algorithms for 3D extrapolations of 2D data. The control group comprised 10 fertile sperm donors while the test group was 10 patients with severely compromised semen parameters including high sperm aneuploidy. All loci examined in the control group adopted defined, interior positions thus providing supporting evidence for the presence of a chromocentre and interior sex chromosome territories. In the test group however there were subtle alterations in the nuclear address for certain centromeres in individual patients and, when all patient results were pooled, some different nuclear addresses were observed for chromosomes 3, 6, 12 and 18. Considering the extensive impairment of spermatogenesis in the test group (evidenced by compromised semen parameters and increased chromosome abnormalities), the observed differences in nuclear organisation for centromeric loci compared to the controls were modest. A defined pattern of nuclear reorganisation of centromeric loci in sperm heads therefore appears to be a remarkably robust process, even if spermatogenesis is severely compromised.

Escaping the cut by restriction enzymes through single-strand self-annealing of host-edited 12-bp and longer synthetic palindromes

Palindromati, the massive host-edited synthetic palindromic contamination found in GenBank, is illustrated and exemplified. Millions of contaminated sequences with portions or tandems of such portions derived from the ZAP adaptor or related linkers are shown (1) by the 12-bp sequence reported elsewhere, exon Xb, 5' CCCGAATTCGGG 3', (2) by a 22-bp related sequence 5' CTCGTGCCGAATTCGGCACGAG 3', and (3) by a longer 44-bp related sequence: 5' CTCGTGCCGAATTCGGCACGAGCTCGTGCCGAATTCGGCACGAG 3'. Possible reasons for why those long contaminating sequences continue in the databases are presented here: (1) the recognition site for the plus strand (+) is single-strand self-annealed; (2) the recognition site for the minus strand (-) is not only single-strand self-annealed but also located far away from the single-strand self-annealed plus strand, rendering impossible the formation of the active EcoRI enzyme dimer to cut on 5' G/AATTC 3', its target sequence. As a possible solution, it is suggested to rely on at least two or three independent results, such as sequences obtained by independent laboratories with the use, preferably, of independent sequencing methodologies. This information may help to develop tools for bioinformatics capable to detect/remove these contaminants and to infer why some damaged sequences which cause genetic diseases escape detection by the molecular quality control mechanism of cells and organisms, being undesirably transferred unchecked through the generations.

Transcription factories in the context of the nuclear and genome organization

In the eukaryotic nucleus, genes are transcribed in transcription factories. In the present review, we re-evaluate the models of transcription factories in the light of recent and older data. Based on this analysis, we propose that transcription factories result from the aggregation of RNA polymerase II-containing pre-initiation complexes assembled next to each other in the nuclear space. Such an aggregation can be triggered by the phosphorylation of the C-terminal domain of RNA polymerase II molecules and their interaction with various transcription factors. Individual transcription factories would thus incorporate tissue-specific, co-regulated as well as housekeeping genes based only on their initial proximity to each other in the nuclear space. Targeting genes to be transcribed to protein-dense factories that contain all factors necessary for transcription initiation and elongation through chromatin templates clearly favors a more economical utilization and better recycling of the transcription machinery.