The genome and the nucleus: a marriage made by evolution
Genomes are housed within cell nuclei as individual chromosome territories. Nuclei contain several architectural structures that interact and influence the genome. In this review, we discuss how the genome may be organised within its nuclear environment with the position of chromosomes inside nuclei being either influenced by gene density or by chromosomes size. We compare interphase genome organisation in diverse species and reveal similarities and differences between evolutionary divergent organisms. Genome organisation is also discussed with relevance to regulation of gene expression, development and differentiation and asks whether large movements of whole chromosomes are really observed during differentiation. Literature and data describing alterations to genome organisation in disease are also discussed. Further, the nuclear structures that are involved in genome function are described, with reference to what happens to the genome when these structures contain protein from mutant genes as in the laminopathies.
There is increasing interest in how a genome is spatially and temporally organised within the busy environment of an interphase nucleus. The level of organisation found in nuclei strongly implies that organisation and positioning is involved on some level with the precise regulation of genomic function. The level of interest and understanding of genome organisation increases almost weekly at present with a plethora of published original articles and reviews coming to the fore. To keep pace, our group is bringing relevant new data and a review article to the international nuclear biology forum. Thus, this review will encompass modern ideas on how and why genomes are so organised within interphase nuclei and how data presented at the recent Genome Organisation symposium at the 15th International Chromosome Conference at Brunel University fits into these theories and moulds new ideologies.
The cell nucleus is highly complex but regulated and organised cellular organelle and has been described in the past as a ‘black box’ (van Driel et al. 1991) due to difficulties in determining what was occurring within it. In the past 25 years and especially the last decade, thanks to increasingly sophisticated microscopy, in vivo methodologies and interest, the nucleus is revealing its secrets.
How is the genome organised in the interphase nucleus?
Once chromosome territories had been rediscovered, researchers either looked more closely at intrachromosome organisation, or others examined the distribution of chromosomes in their nuclear environment. In the latter scenario, Bickmore and colleagues determined that chromosomes were radially organised within interphase nuclei of lymphocytes and fibroblasts (Croft et al. 1999; Bridger et al. 2000; Boyle et al. 2001). This was a major advance in understanding interphase genome organisation although non-random nuclear positioning of specific chromosomes had been reported before (Stahl et al. 1976; Manuelidis and Borden 1988; Hulspas et al. 1994; Armstrong et al. 1994; Park and De Boni 1998).
Bickmore and colleagues demonstrate a strong correlation between the gene density of a chromosome and its radial position within an interphase nucleus (Bridger and Bickmore 1998; Gilbert et al. 2005). However, this theory of radial interphase chromosome positioning is being challenged in ellipsoid nuclei such as those in dermal fibroblasts, and it is being purported that chromosome size is the factor that influences mammalian interphase chromosome position (Sun et al. 2000; Bolzer et al. 2005). Each theory has far-reaching implications on how and why the genome is organised.
Interphase chromosome position gene-density theory
The gene-density theory, whereby chromosomes are positioned radially within the nuclear environment according to the density of genes within a chromosome, was initially proposed due to the disparate nuclear positioning of two similarly sized human chromosomes, HSA 18 and HSA19, in lymphoblasts and proliferating human dermal fibroblasts (Croft et al. 1999). These chromosomes represented two extremes of the human genome, with chromosome 18 being a gene-poor chromosome and chromosome 19 being very gene rich. When the nuclear position of these two chromosomes was assessed, both in two (2D) and three dimensions (3D), it was found that chromosome 18 was predominantly positioned at the nuclear periphery, and chromosome 19 was localised towards the nuclear interior (Croft et al. 1999; Cremer et al. 2003). This means that a particular chromosome has a high probability of being positioned in a certain region of the nucleus, with some chromosome territories being located in an alternative area of the nucleus. Excitingly, this gene density-related correlation with interphase nuclear positioning has been seen in primates (Tanabe et al. 2002a) and in Old World monkeys (Tanabe et al. 2005). Further, this type of organisation of the interphase genome may even be evolutionary conserved in birds, because chicken chromosomes also segregate in the interphase nucleus with gene-poor chromosomes found at the nuclear periphery with gene-rich chromosomes found at the nuclear interior (Habermann et al. 2001; Tanabe et al. 2002b).
The study by Bickmore and colleagues was extended to all human chromosomes in lymphoblasts and many in dermal fibroblasts to test the gene-density hypothesis, and indeed, an overwhelming majority of chromosomes fitted the gene-density theory of chromosome positioning in nuclei (Boyle et al. 2001). An issue to be aware of with this model is that the gene-density profile of chromosomes has been estimated by a combination of assays and data such as replication timing, gene-mapping assignments and predictions of genes from expressed sequence tags (ESTs). Therefore, subsequent analysis of the human genome sequence data may reveal some alterations to this assignment of genes present on particular chromosomes.
This theory has implications for the compartmentalisation of transcriptional activity within the nucleus. Since gene-poor chromosomes are sequestered at the nuclear periphery, it implies that this is an area of transcriptional silencing, whereas the gene-rich chromosomes are localised at a more transcriptionally active region within the nuclear interior, as is early replicating DNA (Ferreira et al. 1997). Although organisation of the nucleus in this way has been discussed for many years (Bridger and Bickmore 1998), there is evidence of transcriptionally activated genes being localised at nuclear pore complexes, and thus the nuclear periphery (Casolari et al. 2004). Although others have demonstrated that sites of active transcription are distributed throughout the nucleus (Wansink et al. 1993). This is still an area of active debate within the field—whether or not active genes are more likely to be found in the nuclear interior.
Interestingly, human chromosomes 13 and 18 (both small chromosomes) are organised quite differently in quiescent cells (that have been serum-starved for 7 days) and senescent cells (that have completed their replicative life span) and exhibited an altered spatial positioning towards the nuclear interior (Bridger et al. 2000; Meaburn 2005). Whereas, larger chromosomes such as chromosome 4 and the X chromosome do not alter their position and remain at the nuclear periphery (Meaburn 2005) in quiescent or senescent cells. In quiescence (Zink et al. 1999; Brown et al. 1999; Bridger et al. 2000) and senescence (Bridger et al. 2000), genome organisation appears to be altered. It is possible that the reorganisation of the genome during quiescence and senescence is due to major modifications of the nuclear architecture leading to alterations in the interaction with the genome which in turn could lead to extensive changes in transcriptional status.
Interphase chromosome position-size theory
The data from our limited studies in non-proliferating primary human fibroblasts fit with another theory, the size theory, whereby chromosomes are positioned according to the their size, with small chromosomes towards the nuclear interior and large chromosomes at the nuclear periphery; as could another study in quiescent cells (Nagele et al. 1999). Indeed, Cremer and colleagues have elegantly analysed the position of all human chromosomes in quiescent fibroblast nuclei in 3D using 24-colour FISH. They have found that in quiescent nuclei, chromosomes are positioned in their nuclear environment according to size (Bolzer et al. 2005). These data could imply that a massive change in transcription as would be seen in cells that were quiescent or senescent could lead to genome organisational changes or a relaxation of tethering of specific chromosomes to specific nuclear architecture. Thus, the alternative theory on chromosome positioning in interphase nuclei is the size-dependent theory of chromosome positioning. This theory also has major support within the field of genome organisation (Sun et al. 2000; Cremer et al. 2001; Habermann et al. 2001; Bolzer et al. 2005).
The first evidence for the size theory was presented by Yokota and colleagues (Sun et al. 2000). They revealed using telomeric probes a distribution whereby the smaller chromosome q arm telomeres of chromosome 19 and chromosome 21 were preferentially distributed towards the interior of the nucleus; whereas the q arms of larger chromosomes, including chromosome 1 and chromosome 2, were more peripheral in location. It should be noted that although these authors used primary fibroblasts, they only used them when they were 90–95% confluent. Many of the primary cells in this situation would be quiescent/G0. However, these authors offer us two explanations as to why chromosome size would affect nuclear positioning of chromosomes. (1) This size-related positioning was due to the volume a chromosome possesses, known as the volume exclusion model or (2) its position on the metaphase plate known as the mitotic pre-set model. Indeed, in support of the volume exclusion model, Cremer et al. (2001) suggested a correlation between the volume a chromosome territory possessed and its spatial positioning in ellipsoid nuclei (fibroblasts and amniotic fluid cells), but not spherical nuclei (B and T lymphocytes) (Cremer et al. 2001). Thus, there appeared to be specific differences between cell types derived from different lineages. The mitotic pre-set model suggests that the position of each chromosome on a metaphase plate is the determinant of its subsequent nuclear positioning. On the wheel-shaped rosette of the metaphase plate, chromosomes are attached via their kinetochores, this would result in the smaller chromosomes' arms being more centrally located on the rosette and the larger chromosomes' arms being further away in distance from the centre of the rosette (Sun et al. 2000). Theoretically, this pattern would be preserved in G1 (Sun et al. 2000), and it has been shown by others that chromosome position can change dramatically in G1 (Bridger et al. 2000; Walter et al. 2003; Thomson et al. 2004). In the Sun study, no telomeric probes were used for the p arms of chromosomes and because it is possible that the p and q arms may occupy different positions in regard to each other (Dietzel et al. 1998a,b), especially in the larger sized chromosomes the positioning of telomeres does not necessarily provide accurate information about the whole positioning of a chromosome territory, which is usually elicited when delineating a whole chromosome territory. Such an example is human chromosome 18, where the telomeres occupy a more internal position to that of the actual chromosome arms (Croft et al. 1999).
The nuclear positioning studies on chicken chromosomes also fit very well with the size theory of nuclear chromosome positioning. The large macrochromosomes were found to be preferentially located at the nuclear periphery, and small microchromosomes towards the interior in both proliferating embryonic fibroblasts and embryonic non-proliferating neurons (Habermann et al. 2001). In a way it is unfortunate that this study fits both theories because advocates of either can use it to support their ideology (Tanabe et al. 2002b, this review). Murine chromosome position analyses are also not providing support for either model, but this maybe has more to do with how similar murine chromosomes are to each other with respect to size and gene distribution (Meaburn 2005; Shiels and Bridger, unpublished data). The situation of data fitting both theories is now occurring in analyses of porcine nuclei. Indeed, in a study from our group (Foster, Griffin and Bridger, unpublished data), we have found that the nuclear positioning of the majority of porcine chromosomes fits both theories, i.e. gene density and size, when analysing embryonic and adult fibroblasts, lymphocytes and mesenchymal stem cells. However, we also found that three chromosomes fit the gene-density model, two fit the size theory and two did not fit either model. The assessment of gene density of porcine chromosomes has been made by the distribution of porcine CpG islands (McQueen et al. 1997) and H3-rich isochores (Federico et al. 2004) and synteny with the human genome with respect to CpG island distribution, late replicating DNA and nuclear chromosome position (Craig and Bickmore 1994; Boyle et al. 2001). However, relatively little is known about porcine genome as compared to other organisms that have had their chromosomes sequenced, i.e. human (Venter et al. 2001) and chicken (Hillier et al. 2004). This will hopefully change in the next few years due to initiatives to sequence the pig genome (Schook et al. 2005). We would certainly relish this because the authors of this review have mapped the positioning of porcine interphase chromosomes in a variety of cell types making the pig the organism that has the most whole chromosomes mapped to their interphase position than any other organism (Foster et al. 2004, 2005; Foster, Griffin and Bridger, manuscript in preparation).
Although there is controversy over how chromosomes are positioned within nuclei, i.e. by gene density or by size, it is possible that these two theories are not mutually exclusive, and chromosome positioning depends on the status of the cell and/or chromosome. Further studies on chromosome positioning and interaction with nuclear architecture in several different organisms will help us to understand genome behaviour and function during evolution and development.
Evolutionary differences and similarities in genome organisation
Mammals, birds and reptiles diverged from a common ancestor 350 million years ago (Lundin 1993). Homologous chromosomal segments between several different species via genetic and physical mapping have been identified. Indeed, much evolutionary information can be gleaned from using complete sets of species-specific chromosome painting probes (Ferguson-Smith et al. 2005). For example, this has been achieved using human chromosome paints on dolphin (Bielec et al. 1998), lemurs (Muller et al. 1997), pig (Fronicke et al. 1996) and many other diverse species (Tanabe et al. 2002b).
It appears that territorial organisation of chromosomes does indeed go back a long way in evolution. To really study specific chromosome organisation within interphase nuclei, it is best to have individual painting probes, although sub-chromosomal probes can be informative. Therefore, interphase chromosome organisation has only been studied in several different organisms: these include human (Cremer et al. 1988; Lichter et al. 1988; Pinkel et al. 1988; Boyle et al. 2001), mouse (Haaf and Ward 1995; Garagna et al. 2001; Parada et al. 2002; Mahy et al. 2002a,b; Ragoczy et al. 2003; Chambeyron and Bickmore 2004a,b; Chambeyron et al. 2005), pig (Foster et al. 2004, 2005), primates (Tanabe et al. 2002a), Old World monkeys (Tanabe et al. 2005), birds (Habermann et al. 2001), marsupials and montremes (Rens et al. 2003; Greaves et al. 2003), Chinese and Indian Muntjacs (Scheuermann et al. 2005), reptiles (Rabl 1885), parasites (Ogbadoyi et al. 2000), Hydra (Alexandrova et al. 2003), plants (Leitch et al. 1990; Lysak et al. 2001), insects (Hochstrasser and Sedat 1987), nematodes (Boveri 1909) and yeast (Bystricky et al. 2005). Even bacteria organise their chromosome into functional domains in its environment (Sherratt 2003). Thus, it appears that chromosome territories or domains are found in more or less isolated entities within their nuclear environment in a wide range of species. However, some are organised radially and others are in a Rabl conformation such as plants (Cowan et al. 2001), although mainly in wheat, rye, barley and oats (Dong and Jiang 1998), yeast (Goto et al. 2001) and Drosophila (Marshall et al. 1996). It is also possible that radial organisation of chromosome territories is at least 600 million years old because Hydra also contain chromosome territories that appear to be organised similarly to higher metazoans (Alexandrova et al. 2003).
Other genomic features are also conserved between species. For example, mammalian, chicken and Hydra chromosomes have early replicating regions of the genome that tend to be preferentially located within the nuclear interior, whereas the later replicating DNA is commonly associated with the nuclear periphery (Alexandrova et al. 2003, Sadoni et al. 1999; Habermann et al. 2001; Tanabe et al. 2002a; Rens et al. 2003; Bolzer et al. 2005).
Positioning within the nucleus of particular chromosomes and/or their comparable syntenic regions is evolutionarily preserved at least between humans and higher primates. This has been demonstrated in lymphoblastoid cells from humans and apes, New World monkeys and Old World monkeys for primate chromosomes homologous to HSA18 and HSA19 (Tanabe et al. 2002a, 2005). This conservation of positioning is despite the presence of extensive chromosomal rearrangements that have occurred throughout the evolution of primates. However, we have found that mouse chromosomes are positioned at different nuclear locations within mouse nuclei when compared to the nuclear position of syntenic human chromosomes in human nuclei. Further, when analysing the nuclear position of human chromosomes in a mouse nuclear background (Meaburn et al. 2004), most of them are also positioned differently, which is not by gene density or by size (Meaburn, Newbold and Bridger, manuscript in preparation). Thus, it appears that the mouse genome, probably due to its uniformity in chromosome size and gene density, is organised quite differently in interphase than to humans for which it is a universal model organism. Even in porcine genome organisation, some syntenic regions of chromosomes are not positioned as they would be in humans (Foster, Griffin and Bridger, manuscript in preparation).
Interchromosomal domain compartment and chromatin loops
The functional significance of interphase territorial chromosome organisation and gene transcription has been addressed by the interchromosome domain (ICD) model (Cremer et al. 1993, 2000; Zirbel et al. 1993; Kurz et al. 1996; Bridger et al. 2005; Richter et al. 2005). This model postulates that transcription and RNA processing occurs within a network of space surrounding and forming inlets into chromosomes, known as the ICD compartment. The ICD compartment may also act as a transportation or channelling network for newly synthesised RNA to leave the nucleus (Politz and Pederson 2000; Cremer and Cremer 2001; Bridger et al. 2005). This ideology of nuclear organisation and function was supported by the observations that RNA transcripts predominantly accumulated in the space surrounding chromosomes and possessed a track-like appearance (Zirbel et al. 1993; Jimenez-Garcia and Spector 1993; Lawrence et al. 1989). Further credence was given to a functional space between interphase chromosomes by the introduction of temperature-sensitive Xenopus vimentin into human cells, engineered so that it contained a nuclear location sequence (Bridger et al. 1998a; Richter et al. 2005; Scheuermann et al. 2005). At 28°C, Xenopus vimentin formed filaments that co-localised with coiled bodies, nuclear RNAs and PML bodies, which are constituents of the ICD compartment (Zirbel et al. 1993; Bridger et al. 1998a, 2005; Matera 1999) and was found to be located exclusively outside of chromosome territories (Bridger et al. 1998a; Scheuermann et al. 2005). The extent of the ICD compartment has also been assessed by microinjection of fluorescently labelled macromolecules (Gorisch et al. 2003; Verschure et al. 2003).
Instead of using non-nuclear molecules in the nucleus to delineate the ICD, others have employed RNA, and Singer and colleagues have shown that introduced polyA RNA moves by diffusion through an interchromatin space (Politz et al. 1999), i.e. the ICD, and Lichter and colleagues have also shown that integrated or episomally transcribed viral genes produce branched networks of mRNA in between chromosomes (Bridger et al. 2005). Both studies presumably reveal the path taken by active message through the nucleoplasm to export (Bridger et al. 2004a). Indeed, the study by Bridger et al. shows that the RNA accumulations produced from the integrated viral genes are derived from gene signals positioned at the edges of chromosome territories (Bridger et al. 2005). Moreover, when RNA from specific genes are visualised by FISH, it is always seen outwith chromosome territories in a chromatin-free space (see Clemson and Lawrence 1996; Zirbel et al. 1993; Bridger et al. 2005). However, when other groups have assessed the nuclear distribution of all active sites of transcription by incorporation of bromo-uridine (BrUTP) in mammalian cells or the distribution of transcription factors, transcription sites have been found deep within the chromosome territories (Verschure et al. 1999, 2002; Sadoni and Zink 2004). Indeed, in plant nuclei, BrUTP incorporation is found also in the interior of plant chromosome territories (Abranches et al. 1998). If the new lattice model of intrachromosome organisation described by Bazzett-Jones and colleagues is found to be correct, then it will be that transcription deep inside chromosome territories will occur since the existence of invaginating functional channels of the ICD within chromosome territories will be called into question (Dehghani et al. 2005).
Initially, from the data describing specific RNA species being found outwith chromosome territories, it was predicted that active genes would be found at the outside of chromosome territories enabling RNA transcripts to be exported straight into the ICD compartment for processing, and with studies by Kurz et al. (1996) and Dietzel et al. (1999) from the Lichter and Cremer laboratories, respectively, this appeared to be the case. In detail, in careful two-colour 3D analyses, three coding regions were positioned at the edge of their respective human chromosome territories, whereas a non-coding sequence was positioned randomly throughout its relevant territory (Kurz et al. 1996). A similar finding was established when the positioning of X chromosome ANT2 and ANT3 genes were compared in active (Xa) and inactive (Xi) human X chromosome territories (Dietzel et al. 1999). ANT2 is transcriptionally active on Xa, but inactive on Xi, whereas ANT3 is located on the pseudo-autosomal region and thus evades inactivation. Transcriptionally active ANT2 and ANT3 had peripheral locations on the Xa territory, whereas ANT2 on Xi had a more interior positioning. In the latest genetic intrachromosomal study from the Lichter group (Scheuermann et al. 2004), the position of many more genes has been analysed in 3D preserved nuclei and with sophisticated 3D analysis programmes (Eils et al. 1995). All gene sequences analysed whether they were active or inactive were predominately found at the periphery of chromosome territories (Scheuermann et al. 2004). The distribution of non-coding sequences being located towards the middle of a chromosome territory is supported by the distribution of gene-poor L1 and L2 isochores being found more within the interior of chromosome territories (Tajbakhsh et al. 2000). The discussion of Scheuermann et al. (2004) also held up the observation that regions low in GC content were to be found towards the interior of chromosomes, and genomic regions high in GC content were found towards the edges of chromosome territories. In combination with studies showing active transcription deep within chromosome territories, Bickmore's group have found active and inactive gene sequences at both the chromosome interior and periphery. Mahy et al. (2002a) performed 2D and 3D FISH on a stretch of genomic DNA located on human chromosome 11p13 and 11p15 in different cell types. The 11p13 region contains a mixture of house-keeping and tissue-specific genes, in addition to intergenic DNA. Unlike the study by Kurz and colleagues, the entire 11p13 locus was positioned internally in chromosome territories, even those genes that were actively expressed. Nevertheless, preferential positioning was observed within a territorial sub-domain whereby a ubiquitously expressed gene occupied a peripheral position, and non-coding DNA was found within the interior of the sub-domain. RNA was not simultaneously visualised, therefore, it was unknown if specific genes were actually active. Thus, these observations complicate the simple ICD model and have led to an adaptation to the model that is now described as the interchromatin (IC) compartment model (Cremer and Cremer 2001; Williams 2003). This model postulates that although transcription appears to be internal, it is in fact at the surface of an inlet of the ICD protruding towards the interior of the chromosome territory (Verschure et al. 1999). This fits with the different morphological observations between Xa and Xi, which have irregular furrowed surfaces and a smooth rounded territory appearance, respectively (Eils et al. 1996). However, Dehghani et al. (2005), using energy-filtered transmission electron microscopy, suggest that chromosome territories have more of a lattice organization of chromatin, indicating that the ICD space may be more extensive than revealed by fluorescence microscopy.
A very exciting study from the group of Fraser has demonstrated that indeed genes may migrate to active pre-assembled transcription factories. Because distal genes can occupy the same transcription factory, and this dynamic organisation of transcription is predicted to happen either in cis or in trans, then this study could also support gene sequences not being found deep within chromosome territories (Osborne et al. 2004).
Differentiation is an intricate process of specialisation, where cells develop unique tissue-specific functions or roles and thus become committed to particular cell lineages. The numerous cell and tissue types that constitute a multicellular organism are all derived from a single totipotent zygote. Through coordinated control of gene expression and nuclear and chromatin remodelling, the zygote can give rise to all cell types. With these properties in mind, the differentiation process is an ideal system to study the influence of genome organisation on gene expression. Indeed, in vitro differentiation systems using progenitor cells such as embryonic or adult stem cells are ideal models to investigate chromatin organisation and gene expression when induced to specialise to particular lineages. Some studies have even analysed genome organisation in vivo in sections of tissues and embryos.
Cellular differentiation is accompanied by modifications in gene expression patterns. Transition from one genetic profile to another is highly dynamic, where specific genes are actively expressed and others silenced. This is thought to be mediated through alterations in chromatin conformation in conjunction with epigenetic changes that influence gene expression and collectively initiate differentiation. Some of the underlying mechanisms necessary for differentiation were examined by inducing Hoxb gene expression in mouse ES cells (Chambeyron and Bickmore 2004a) and in mouse embryos (Chambeyron et al. 2005). Stimulation of the Hoxb genes corresponded with increased chromatin remodelling via nucleosome modification, chromatin decondensation and the protrusion of the Hoxb gene locus in a loop away from the chromosome territory towards the nuclear interior (Chambeyron and Bickmore 2004a). The coordinated choreography of these events suggests that epigenetic alterations of histone lysine residues, via acetylation or methylation, are a prerequisite that set genes in preparation for expression (Chambeyron and Bickmore 2004b). Therefore, epigenetic alterations were not synonymous with gene expression per se (Chambeyron and Bickmore 2004b). Together, chromatin decondensation and epigenetic modifications appear to induce a transcriptionally permissive chromatin state (Chambeyron and Bickmore 2004a). In contrast, it was suggested that the repositioning of the Hoxb1 gene from the chromosome territory interior in undifferentiated ES cells, to a loop extruding away from the territory with the induction of differentiation, is representative of a transcriptionally active state (Chambeyron and Bickmore 2004a).
A recent study investigated the position of the lysozyme gene (cLys) located on chicken chromosome 1 during macrophage differentiation (Stadler et al. 2004). Lysozyme expression is low in granulocyte–macrophage precursors, but increases in differentiated macrophages (Huber et al. 1995). Both the lysozyme gene domain and centromere were repositioned from an interior chromosome position in precursor cells to the chromosome territory periphery in myeloblasts and differentiated macrophages (Stadler et al. 2004). Repositioning of the locus was particularly prominent in stimulated macrophages, where cLys and closely positioned gene, cGas41, occupied their most peripheral location, sometimes extending from the chromosome territory. However, the loci were also repositioned to the territory periphery in proerythroblasts, which like precursor cells do not express cLys and cGas41 (Stadler et al. 2004). This indicates that other factors also influence the positioning of genes relative to the territory and not just transcriptional activity.
Gene silencing is also a critical event associated with differentiation and genome reorganisation, and gene repositioning has been found to be important in gene expression silencing and down-regulation. The heritable silencing of particular genes associated with differentiation events has been investigated during T-cell development (Brown et al. 1999) and thymocyte differentiation (Su et al. 2004; Merkenschlager et al. 2004). During T-cell maturation, the transcriptional activity of Rag and TdT genes was silenced through differentiation-induced mechanisms. Following gene silencing, these genes were subsequently repositioned to centromeric heterochromatin at more peripheral regions of the nucleus (Brown et al. 1997, 1999). Furthermore, it has been demonstrated that silencing of the co-receptor loci CD4 and CD8 is coincident with positioning of these genes with centromeric heterochromatin (Merkenschlager et al. 2004). Association of inactive genes with perinuclear heterochromatin has also been seen in several transformed human cell types. Thus, it seems that dynamic association of genes with heterochromatic regions of the nucleus is a mechanism to induce cell-type specific genome expression profiles.
Thus, from several studies, it appears that intranuclear position of genes is perhaps common in differentiation, and this presents us with a mechanism by which regions of the genome can become extensively expressed or heritably silenced. These studies may also give us more indirect evidence for genomic regions being extruded at a distance from chromosome territories because there is very little really convincing evidence to show that whole chromosomes move during differentiation, thus maybe only specific regions of chromosomes move.
In a study that utilises the differentiation of cells into specialised cell types, i.e. adipocytes, the radial and relative nuclear positioning of human chromosomes 12 and 16 were investigated in both pre-adipocytes and adipocytes to determine whether repositioning of chromosome territories is a characteristic of adipogenesis (Kuroda et al. 2004). Interestingly, the radial distribution of chromosomes 12 and 16 remained unaltered; however, their positions relative to one another did change with a slight increase in proximity upon differentiation to adipocytes (Kuroda et al. 2004). Having chromosomes in closer proximity to each other could influence the frequencies of specific chromosomal translocations such as those frequently observed in liposarcomas, such as the t(12;16) present in approximately 95–98% of myxoid and round-cell liposarcomas (Kuroda et al. 2004).
The Bridger group have developed novel differentiation model systems using porcine mesenchymal stem cells and have induced them to follow adipogenesis or myogenesis differentiation pathways. However, we have seen no major change in radial position of the pig chromosomes so far analysed. On the other hand, during erthyroid differentiation of human bone marrow cells, the territory of chromosome 11 moved to a more peripheral radial position, but specific genes on chromosome 11, i.e. beta-like globin gene clusters, remained in the same relative nuclear position on a chromosome loop (Galiova et al. 2004); and in T-cell differentiation, murine chromosome 6 slightly altered its nuclear position (Kim et al. 2004). Genome organisation and chromosome territory repositioning during differentiation could be implied in the studies showing that the nuclear distribution of centromeres alters during differentiation of promyelocytic leukemia cells (Beil et al. 2002) and T-cell differentiation (Kim et al. 2004). In in vitro cultured cells derived from specific mouse tissues that share a common differentiation pathway, similar nuclear positioning of corresponding chromosome territories was observed, with the exception of murine chromosome 5 that could be found at differing nuclear positions depending on cell type (Parada et al. 2004).
The most obvious redistribution of chromosome territories during differentiation seen so far is observed in the porcine testes, with the sex chromosomes moving from a peripheral nuclear positioning to the most interior nuclear positioning, an autosome moving from an interior nuclear location to a peripheral position and another autosome not repositioning at all, as cells differentiate from spermatocytes through spermatids to mature sperm. Interestingly, this occurs as cells are becoming less and less transcriptionally active. The haploid spermatozoa cells are being prepared to meet the oocyte, and chromosome positioning within a gamete may have consequences on future gene expression patterns in the developing embryo (Foster et al. 2005). We propose here that although intranuclear positioning of specific gene loci can change dramatically during differentiation, whole chromosome territory position does not change fundamentally until there are major alterations in transcriptional status in a cell. This view would be supported by data from the Gilbert group who do not observe large changes in the spatial organisation of chromosomes in murine cells of different types (Panning and Gilbert 2005) and from the Silver group, whereby yeast chromosome III does move in response to developmental changes after mating, which is linked to major alterations to the gene expression profile (Casolari et al. 2005). Alterations in transcriptional status of cells during differentiation may be more subtle, be relatively temporally slow and thus only give rise to sub-chromosomal regions moving within the nucleus.
Thus, to determine the true extent of cell-type specific genome organisation, it is important to carry out further experiments in differentiating cells and cells dramatically changing status.
Although chromosomes may not change intranuclear position that dramatically during most differentiation, there is evidence of large-scale morphological changes. Conformational changes in territory compaction are apparent during chicken macrophage differentiation, where territories are in a compacted state in chicken myeloid precursor cells and become increasingly diffuse with differentiation, appearing more ‘blurry’ with furrowed exteriors (Stadler et al. 2004). Similar alterations in territory size during differentiation were also determined during human adipogenesis, where the volume occupied by chromosome 16 increased by 50% after differentiation to adipocytes (Kuroda et al. 2004). However, this was not a global phenomenon inasmuch as chromosome 12 possessed a similar territory size throughout adipogenesis (Kuroda et al. 2004).
The nuclear lamina
The nuclear membrane is a lipid bilayer structure that encloses the contents of the nucleus during interphase. Subjacent to the inner nuclear membrane is the nuclear lamina that has a close and probably regulative relationship with the genome. The mammalian nuclear lamina comprised type V intermediate filaments known as lamins (A-type and B-type) that polymerise to form an interwoven lattice-like network of fibrils, and it also contains many integral membrane proteins (IMPs) with various enigmatic functions. Mammalian A-type lamins have been reported as being regulated developmentally and so are expressed predominantly in differentiated cells (Lehner et al. 1987; Rober et al. 1989), with a possible role in tissue specificity maintenance (Lehner et al. 1987; Goldman et al. 2002). A-type lamins are encoded by the LMNA gene and are differentially spliced to produce several different isoforms, A, AΔ10, C and C2, which are found in a variety of ratios in different cell types (see Mounkes et al. 2003 for review). The stalwart B-type lamins are expressed constitutively in both embryonic and somatic cells and appear to be essential, although they can be compensated for by A-type lamins for a short period of time (Steen and Collas 2001). Defects in B-type lamins are considered lethal because in mice generated with a lamin B1 mutation, the pups died at birth (Vergnes et al. 2004), and using RNA interference of lamin during Caenorhabditis elegans development resulted in embryonic lethality (Liu et al. 2000). However, mutations in the A-type lamins and certain inner membrane proteins give rise to various inherited tissue-specific diseases with quite diverse phenotypes termed laminopathies.
Lamins have been found in every metazoan so far assessed (Cohen et al. 2001). However, nuclear lamins have yet to be discovered in yeast, although exogenous lamin does polymerise at the nuclear periphery (Enoch et al. 1991). Lamins could be present in plants (McNulty and Saunders 1992; Minguez and Moreno Diaz de la Espina 1993; Chen et al. 2000; Moreno Diaz de la Espina et al. 2003; Blumenthal et al. 2004) although no definitive data has yet been seen; but they are definitely present in insects (Jensen and Brasch 1985; Smith et al. 1987), nematodes (Riemer et al. 1993, 2000; Erber et al. 1999), echinoderms (Holy et al. 1995), fish (Zimek et al. 2003) and most interestingly in cnidarians such as Hydra (Erber et al. 1999).
Lamins have a range of enticing roles in nuclear function and interact with a variety of molecules involved in organising nuclear structure, chromatin or gene function, which leads to their recognised, reputed or alleged roles in maintaining nuclear integrity, nuclear reassembly after division, DNA replication, DNA repair, apoptosis, transcription, genome organisation, cell cycle control, proliferation and life span.
Integral membrane protein
Although both types of lamins have chromatin/DNA-binding abilities, there are also a plethora of IMPs embedded at the nuclear envelope that could also have genome tethering roles at the nuclear periphery. The community have already characterised several of these IMPS, but there could be many more potential IMPs, given a proteomics study by Gerace and colleagues (Schirmer et al. 2003). Known IMPs include Lamin B receptor (LBR) which, as its name suggests, binds lamin B (Worman et al. 1988), but it also binds heterochromatin (Simos and Georgatos 1992; Ye and Worman 1996; Dreger et al. 2002) via a heterochromatin-associated protein HP1 (Ye and Worman 1996; Ye et al. 1997). Further, in vitro experiments suggest that LBR also binds chromatin and histones H3/H4 (Makatsori et al. 2004).
Lamina-associated polypeptide 1 (LAP1 A, B, C) and 2 (LAP2 β) are also found at the nuclear envelope and interact with the genome. Both LAP1 A and B can bind indirectly to mitotic chromosomes (Foisner and Gerace 1993).
LAP2 and two other IMPs called Emerin and MAN1 are related to each other inasmuch as they contain a conserved N-terminal domain called a LEM box. The LEM box is thought to be important in the interaction of these LEM proteins with a chromatin-binding protein barrier of autointegration factor (BAF) (Furukawa 1999). LAP2 has a high affinity for DNA (Cai et al. 2001), chromatin (Vlcek et al. 1999), BAF (Furukawa 1999) and the chromatin-associated protein HA95 (Martins et al. 2003). Like LBR (Pyrpasopoulou et al. 1996; Ellenberg et al. 1997), LAP2 also associates with mitotic chromosomes that indicates an important role in the initial events of nuclear reassembly (Foisner and Gerace 1993).
There is a recent surge of interest in a family of large proteins, termed nesprins (Zhang et al. 2001), that have a partial LEM box but do not bind BAF (Mislow et al. 2002). Nesprins come in many isoforms, and the smaller ones have been demonstrated to co-localise with heterochromatic regions of the genome (Zhang et al. 2001, 2005), implicating them in a tethering role or a communication route to the genome.
Thus, at the nuclear envelope, there seems to be many proteins (including lamins) that bind and potentially anchor the genome to the nuclear periphery. Several of these proteins associate specifically with heterochromatic or gene-poor regions of the genome, allowing us to postulate how the periphery of the nucleus got to be the most transcriptionally inert nuclear compartment.
The nucleolus is the archetypal nuclear compartment and is a sub-compartment involved in the biogenesis of ribosomes. Within the nucleolus, ribosomal DNA genes (namely, 18S, 5.8S and 28S) are transcribed by RNA polymerase I. The resultant precursor RNA is processed and assembled into pre-ribosomal subunits that are exported to the cytoplasm (Shaw and Doonan 2005). Ribosomal genes form arrays of tandem repeats on particular chromosomes and are termed nucleolar organiser regions (NORs). In humans, NORs are located on the acrocentric chromosomes 13, 14, 15, 21 and 22. The nucleoli are formed when NOR regions on one or more chromosomes cluster together. The ultrastructure of the nucleolus constitutes three distinct structures: the fibrillar centres (FCs), the dense fibrillar component (DFC), and the granular component (GC).
Recently, the nucleolus has been accredited as having an important role in nuclear regulatory functions. Particular proteins involved in regulating the cell cycle such as cdc14 and mdm2 that have roles in supporting mitosis exit and p53 inhibitor, respectively, are sequestered in the nucleolus (Pederson 1998; Visintin and Amon 2000).
The nucleolus not only has many important roles in nuclear function but is probably also involved in genome organisation and chromosome positioning. We now know that the NOR carrying acrocentric chromosomes are associated with or embedded within nucleoli (Weipoltshammer et al. 1999), even chromosomes carrying inactive NORs (Bridger et al. 1998b; Sullivan et al. 2001). Furthermore, the nucleolus has also been shown to house other chromosomes that do not contain NORs in humans (Manuelidis and Borden 1988; Bridger et al. 1998b, 2000; Chubb et al. 2002; Bridger, unpublished data) and in yeast (Thompson et al. 2003). This idea of non-NOR-containing chromosomes being associated is supported by studies that reveal clustering within nuclei after damage by radiation (e.g. Arsuaga et al. 2004). These authors reported two clusters of chromosomes that were closer to each other than any other chromosomes, and they contained HSAs 1, 16, 17, 19, 22 and HSAs 13, 14, 15, 21, 22. The former group is all chromosomes that have been reported to be the most interior chromosomes and probably associated with nucleoli. The latter group is of course the chromosomes one would expect to find associated with nucleoli.
With increasing combinatorial studies using both FISH and immunocytochemistry and biochemical extractions, we will uncover more about the relationship of the nucleolus with the non-NOR-containing regions of the genome.
The nuclear matrix
The nuclear matrix is a piece of nuclear architecture whose existence is still somewhat in question and appears to be an anatomising network of filaments throughout the nucleus. However, it seems with the publication of elegant modern-day studies, there are fewer and fewer nuclear biologists that challenge its authenticity. Increasing evidence of the existence of a nuclear matrix is being compiled via electron microscopy (Nickerson 2001) and by direct observations in living cells using fluorescently tagged proteins, indicating that although chemically undefined, a nuclear matrix does appear to exist (Oegema et al. 1997; Broers et al. 1999; Nalepa and Harper 2004).
However, even for stalwart believers, it is not easy for them to describe the nuclear matrix's exact nature, dimensions or roles. The nucleoskeleton (Jackson and Cook 1985) may also be what is termed the nuclear matrix, but the two structures do associate with different regions of the genome (Craig et al. 1997). What is more definite is a role for the nuclear matrix/nucleoskeleton in genome function, participating in support of chromosome structure, gene expression, DNA replication, RNA transport and processing, signal transduction and apoptosis (Bickmore and Oghene 1996; Bode et al. 2000; Djeliova et al. 2001; Wei et al. 1998, 1999).
The nuclear matrix/nucleoskeleton structural constituents are complex and as of yet not totally defined. Some known components include lamins (Hozak et al. 1995; Neri et al. 1999; Barboro et al. 2002), intermediate filament-like core filaments, nuclear bodies, nucleolar remnants and interchromatin granule clusters (Philimonenko et al. 2001). The integrity of the nuclear matrix is dependent on RNA, and treatment with RNase A and high-salt concentrations (2 M NaCl) can disrupt both the nuclear matrix (He et al. 1990, 1991) and chromosome territories (Ma et al. 1999). Thus, insinuating that the nuclear matrix is a structural component in genome organisation. The nuclear matrix is also thought to play an important role in setting up functional domains within nuclei (Bode et al. 2000) and facilitates important cellular processes such as initiation of DNA replication (Radichev et al. 2005; Jenke et al. 2004; Girard-Reydet et al. 2004; for review of earlier work see Stein et al. 2003 and Jackson 2003) and transcription (for review see Stein et al. 2004). Nevertheless, the matrix appears morphologically similar in early embryos that exhibit very little transcription compared to later embryos that do transcribe heavily and those treated with transcription inhibitors (Philimonenko et al. 2001).
Many different regions of the genome have been identified as having interactions with the nuclear matrix, i.e. matrix attachment regions (MARs) (Dijkwel and Hamlin 1988), telomeres (de Lange 1992; de Lara et al. 1993; Markova et al. 1994; Luderus et al. 1996; Okabe et al. 2004), and centromeres/kinetochores (Chaly et al. 1985; Markova et al. 1994; Liao et al. 1995). One of the most interesting differential interactions of genome with the nuclear matrix was observed by Bickmore and colleagues who prepared nuclear matrices from human lymphoblasts and fibroblasts without the DNA digestion (DNA halos) and found that gene-rich HSA19 remained associated tightly with the nuclear matrix, whereas gene-poor HSA18 was distributed outwith the residual nucleus in the DNA halo (Croft et al. 1999).
Genome organisation and disease
If the spatial organisation and positioning of the genome is important in the regulation of gene expression and cellular health, then we would expect to see perturbations of normal genome organisation in diseased cells. Several studies have assessed chromosome positioning in transformed, immortalised or cancer cells, and chromosome positioning has seemed similar and comparable to normal cells (Croft et al. 1999; Boyle et al. 2001; Parada and Misteli 2002; Cremer et al. 2003). To highlight this, a comparison of neighbouring chromosome arrangements was investigated in normal mouse splenocyte cells and lymphoma cells. This revealed that two chromosomes that were commonly found to be translocated with each other in murine lymphoma were spatially positioned in close proximity in 50% of lymphoma cells, but this positioning was also conserved in normal splenocyte cells (Parada and Misteli 2002). However, in another study by Cremer and colleagues, altered radial positioning of human chromosome 18 was evident in several tumour cell lines, where HSA18 exhibited a more internal nuclear location (Cremer et al. 2003). Indeed, chromosome 18's central locality was even more prominent than chromosome 19, indicating a partial loss in radial chromosome organisation in nuclei derived from specific tumours (Cremer et al. 2003). These findings highlight the need for careful consideration of cell types used in genome organisation studies.
Chromosomal translocations are especially associated with many cancers. Several studies have highlighted that specific translocations could be generated because those chromosomes are in close proximity with each other (Kozubek et al. 1999; Parada et al. 2002; Bickmore and Teague 2002; Roix et al. 2003), which maybe one of the reasons why the Robertsonian translocations are so common given that these are translocations between the nucleolar-associated acrocentric chromosomes and would be in very close proximity. Other studies have generated translocations by creating double-strand breaks in the interphase genome using radiation (Lukasova et al. 1999; Nikiforova et al. 2000; Arsuaga et al. 2004) and have intimated that spatial proximity may be important in the generation of such chromosomal aberrations as translocation.
The first real observation that chromosomes could be repositioned in cells associated with illness came from a study whereby the positioning of chromosomes has been examined in neurons within human brain cortex (Borden and Manuelidis 1988). Chromosome positioning was investigated in brain samples derived from epilepsy sufferers and compared to normal brain samples. Small positional changes of chromosomes 1, 9 and Y were detected in cells in electrophysiologically defined seizure foci within the brain. However, drastic repositioning of the centromere of the X chromosome was seen (Borden and Manuelidis 1988). Epilepsy has a genetic element associated with it and is a gene on the X chromosome called Arx (Stromme et al. 2002). Although this gene is on the X chromosome but as yet no link has been made with X chromosome repositioning and the expression of this gene.
There are several diseases that have perturbed nuclear architecture and have a wide range of mutations in the gene encoding A-type lamins, LMNA. These diseases that are commonly termed laminopathies are often characterised by the manifestation of disease in particular tissues derived from mesenchymal origin including muscle, adipose tissue and neurons (Hutchison and Worman 2004). They can be broadly categorised as muscular dystrophy, lipodystrophy, neurodystrophy and progeroid disorders. The causative mechanisms of disease associated with A-type lamin mutations have also been postulated. One explanation is that mutations in A-type lamins weaken the structural integrity of the nucleus, making them fragile and prone to physical stress (Sullivan et al. 1999; Hutchison et al. 2001). Another hypothesis proposes that lamins are fundamental for correct gene expression patterns, and that mutations produce perturbations in genome organisation and signal pathways that affect the correct nuclear functioning of the cell (Cohen et al. 2001; Mounkes et al. 2001). The underlying disease mechanism may also be due to a combination of factors hypothesised in both the structural and gene expression model.
Evidence of increased nuclear envelope fragility is demonstrated by Lmna−/− mice that exhibit altered nuclear envelope structures with deformations including ‘herniations’ and blebbing (Sullivan et al. 1999). Lmna−/− mice are also prone to mechanical stress, as they are less stiff and rigid in comparison with wild-type cells (Lammerding et al. 2004). However, weakened mechanical integrity of the cell only partially elucidates phenotypes observed in different laminopathies. Thus, perturbation of lamin interaction with the genome with consequent alterations to gene expression and regulation of signal pathways could also account for the tissue-specific laminopathy disease.
The gene expression model appears to be supported by the observation that mice that are emerin or lamin A/C deficient have difficulties forming interactions with chromatin and in fact lack the heterochromatin layer normally associated with the NE (Sullivan et al. 1999).
We have found that normal chromosome positioning is altered in all laminopathy patients so far assessed, and this is also true for non-symptomatic heterozygous carriers. The repositioning of the genome is similar to that seen in quiescent or senescent cells, although the laminopathy cells are definitely proliferating as determined by specific markers (Bridger et al. 2004a,b,c; Meaburn 2005; Meaburn and Bridger, manuscript in preparation). These data link mutation in a known nuclear structural protein with altered genome organisation and chromosome positioning. The consequences of this are not yet clear.
The idea that the spatial and temporal organisation of genomes in their nuclear environment helps to control their function and behaviour is becoming widely accepted, especially when we see genome mis-organisation in disease cells such as cancer and nuclear envelope diseases such as laminopathies. Studies on functional genome organisation should include how the genome interacts with the underlying architecture making up the nucleus. Thus, understanding aspects of the genome and its interphase home in 4D will lead to 21st century treatments and therapies for a range of ills from cancer to ageing.
- Bolzer A, Kreth G, Solovei I, Saracoglu K, Fauth C, Müller S, Eils R, Cremer C, Speicher MR, Cremer T (2005) Complete 3D-maps of chromosome positions in human male fibroblast nuclei and prometaphase rosettes demonstrate a chromosome size dependent, probabilistic arrangement. PLoS Biology, (in press)Google Scholar
- Boveri T (1888) Zellenstudien II. Die befruchtung und teilung des eies von ascaris megalocephala. Jena Zeit Naturw 22:685–882Google Scholar
- Boveri T (1909) Ueber geschlechstschromosomen bei nematoden. Arch Zellf 4:132–141Google Scholar
- Bridger JM, Meaburn KJ, Foster HA, Figgitt M, Bonne G, Levy N, Griffin DK, Kill IR (2004a) The role of nuclear structure in genomic health. Chromosome Res 12(suppl 1)Google Scholar
- Bridger JM, Reichenzeller M, Lichter P, Herrmann H (2004b) In: Hemmerich P, Diekmann S (eds) The interchromosomal domain compartment: an active space within the nucleus. Chapter in visions of the cell nucleusGoogle Scholar
- Cremer T, Kurz A, Zirbel R, Dietzel S, Rinke B, Schrock E, Speicher MR, Mathieu U, Jauch A, Emmerich P, Schertan H, Reid T, Cremer C, Lichter P (1993) Role of chromosome territories in the functional compartmentalization of the cell nucleus. Cold Spring Harb Symp Quant Biol 58:777–792PubMedGoogle Scholar
- Eils R, Bertin E, Saracoglu K, Rinke B, Schrock E, Parazza F, Usson Y, Robert-Nicoud M, Stelzer EH, Chassery JM et al (1995) Application of confocal laser microscopy and three-dimensional Voronoi diagrams for volume and surface estimates of interphase chromosomes. J Microsc 177:150–161PubMedGoogle Scholar
- Eils R, Dietzel S, Bertin E, Schrock E, Speicher MR, Ried T, Robert-Nicoud M, Cremer C, Cremer T (1996) Three-dimensional reconstruction of painted human interphase chromosomes: active and inactive X chromosome territories have similar volumes but differ in shape and surface structure. J Cell Biol 135:1427–1440CrossRefPubMedGoogle Scholar
- Foster HA, Griffin DK, Leese HJ, Sturmey RG, Stokes PJ, Abeydeera LR, Bridger JM (2004) Genome and nuclear architecture organisation during development and differentiation using the pig as a model organism. Chromosome Res 12(supp 1):3–4Google Scholar
- Hillier LW, Miller W, Birney E, Warren W, Hardison RC, Ponting CP, Bork P, Burt DW, Groenen MA, Delany ME, Dodgson JB, Chinwalla AT, Cliften PF, Clifton SW, Delehaunty KD, Fronick C, Fulton RS, Graves TA, Kremitzki C et al (2004) Sequence and comparative analysis of the chicken genome provide unique perspectives on vertebrate evolution. Nature 432:695–716CrossRefPubMedGoogle Scholar
- Jenke AC, Stehle IM, Herrmann F, Eisenberger T, Baiker A, Bode J, Fackelmayer FO, Lipps HJ (2004) Nuclear scaffold/matrix attached region modules linked to a transcription unit are sufficient for replication and maintenance of a mammalian episome. Proc Natl Acad Sci U S A 101:11322–11327CrossRefPubMedGoogle Scholar
- Knoch A, Münkel C, Langowski J (1999) Three-dimensional organization of chromosome territories and the human interphase nucleus. In: high performance computing in science and engineering. Springer, Berlin Heidelberg New YorkGoogle Scholar
- Kozubek S, Lukasova E, Mareckova A, Skalnikova M, Kozubek M, Bartova E, Kroha V, Krahulcova E, Slotova J (1999) The topological organization of chromosomes 9 and 22 in cell nuclei has a determinative role in the induction of t(9,22) translocations and in the pathogenesis of t(9,22) leukemias. Chromosoma 108:426–435CrossRefGoogle Scholar
- Makatsori D, Kourmouli N, Polioudaki H, Shultz LD, McLean K, Theodoropoulos PA, Singh PB, Georgatos SD (2004) The inner nuclear membrane protein lamin B receptor forms distinct microdomains and links epigenetically marked chromatin to the nuclear envelope. J Biol Chem 279:25567–25573CrossRefPubMedGoogle Scholar
- Meaburn K (2005) The role of nuclear architecture in genomic stability. PhD thesis, Brunel UniversityGoogle Scholar
- Meaburn KJ, Newbold RF, Cox H, Bridger JM (2004) Using the monochromosome hybrid cell lines to identify nuclear factors required for correct human genome organisation. Chromosome Res 12(supp 1):99Google Scholar
- Rabl K (1885) Ûber Zelltheilung. Gegenbaurs Morphol Jahrb 10:214–330Google Scholar
- Stromme P, Mangelsdorf ME, Shaw MA, Lower KM, Lewis SM, Bruyere H, Lutcherath V, Gedeon AK, Wallace RH, Scheffer IE, Turner G, Partington M, Frints SG, Fryns JP, Sutherland GR, Mulley JC, Gecz J (2002) Mutations in the human ortholog of Aristaless cause X-linked mental retardation and epilepsy. Nat Genet 30:441–445CrossRefPubMedGoogle Scholar
- Venter JC, Adams MD, Myers EW, Li PW, Mural RJ, Sutton GG, Smith HO, Yandell M, Evans CA, Holt RA, Gocayne JD, Amanatides P, Ballew RM, Huson DH, Wortman JR, Zhang Q, Kodira CD, Zheng XH, Chen L, Skupski M et al (2001) The sequence of the human genome. Science 291:1304–1351CrossRefPubMedGoogle Scholar
- Volpi EV, Chevret E, Jones T, Vatcheva R, Williamson J, Beck S, Campbell RD, Goldsworthy M, Powis SH, Ragoussis J, Trowsdale J, Sheer D (2000) Large-scale chromatin organization of the major histocompatibility complex and other regions of human chromosome 6 and its response to interferon in interphase nuclei. J Cell Sci 113:1565–1576PubMedGoogle Scholar
- Zhang Q, Ragnauth CD, Skepper JN, Worth NF, Warren DT, Roberts RG, Weissberg PL, Ellis JA, Shanahan CM (2005) Nesprin-2 is a multi-isomeric protein that binds lamin and emerin at the nuclear envelope and forms a subcellular network in skeletal muscle. J Cell Sci 118:673–687CrossRefPubMedGoogle Scholar