Chromatin organisation in the nucleus

The nuclear envelope confines the cell nucleus and separates the genome and the factors needed for replication and transcription in the nucleoplasm from the cytoplasm. The nucleoplasm is further compartmentalised into a number of subnuclear organelles, in which certain protein and nucleic acid components with specific biological activities are concentrated. The most prominent example is the nucleolus, which contains regions with ribosomal RNA genes from several chromosomes, and also contains the machinery for the assembly of ribosomal subunits. Other nuclear substructures include the SC35 domains (“Speckles”), Cajal and promyelocytic leukaemia (PML) bodies (Spector 2001; Gall 2003; Lamond and Sleeman 2003; Lamond and Spector 2003; Spector 2003). The localisation, organisation, dynamics and biological activities of these suborganelles appear to be closely related to gene expression and are, therefore, an important subject of current research.

Using whole chromosome painting probes and fluorescence in situ hybridisation (FISH), a territorial organisation of interphase chromosomes has been demonstrated (Lichter et al. 1988; Pinkel et al. 1988; Cremer et al. 1993). Chromosome territories have irregular shapes and occupy discrete nuclear positions with little overlap (Fig. 1). In general, gene-rich chromosomes are located more in the nuclear interior while gene-poor chromosome territories are located at the nuclear periphery (Croft et al. 1999; Boyle et al. 2001; Cremer et al. 2001; Tanabe et al. 2002; Dietzel et al. 2004). In agreement with this, non-transcribed sequences were predominantly found at the nuclear periphery or perinucleolar while active genes and gene-rich regions tended to localise on chromosome surfaces exposed to the nuclear interior or on loops extending from the territories (Kurz et al. 1996; Dietzel et al. 1999; Volpi et al. 2000; Mahy et al. 2002; Scheuermann et al. 2004).

Fig. 1
figure 1

Nuclear organisation. a U2OS cell nucleus DAPI counterstained with FISH-painted chromosome territories 3 (green) and 18 (red). Scale bar represents 10 μm. Image courtesy of S. Geiger; DKFZ Heidelberg. b Schematic illustration of the compartmentalisation of an interphase nucleus showing nuclear bodies, Speckles and NLS-vimentin located outside of chromosome territories.

Full size image

Promyelocytic leukaemia and Cajal bodies are preferably located outside of chromosome territories (Zirbel et al. 1993; Bridger et al. 1998). In addition, specific RNA transcripts are found in track-like or spherical accumulations within the nucleus (Lawrence et al. 1989; Xing et al. 1993; Lampel et al. 1997; Bridger et al. 2005). Nuclear spaces accessible to large supramolecular complexes were examined with vimentin modified to carry a nuclear localisation sequence (NLS) as well as nuclear crystalline Mx1-YFP particles. NLS-vimentin and Mx1-YFP were both found in the vicinity of nuclear bodies (Bridger et al. 1998; Görisch et al. 2004) (Fig. 2a, b) and also next to each other (Richter et al. 2005) (Fig. 2c, d). NLS-vimentin was predominantly located outside of FISH-labelled chromosome territories (Bridger et al. 1998), and similar results were observed for NLS-vimentin and Mx1-YFP in BrdU-labelled cells (Fig. 2e, f). These experimental findings support the concept of a functional nuclear space, the interchromosomal domain (ICD) compartment (Cremer et al. 1993; Zirbel et al. 1993; Herrmann and Lichter 1999). According to the ICD model, the interface between chromosome territories is more easily accessible to large nuclear complexes than regions within the territory (Fig. 1b) (reviewed in Cremer and Cremer 2001). More recently, it has been proposed that chromosome territories are further organised into 1-Mb domains (Zink et al. 1999; Berezney et al. 2000; Visser et al. 2000; Cremer and Cremer 2001; Cremer et al. 2004; Kreth et al. 2004), extending the more accessible space to open intra-chromosomal regions surrounded by denser chromatin domains. Using high-resolution light microscopy, an apparent bead-like structure of chromatin can be visualised in which ~1-Mb domains of chromatin are more densely packed into an approximately spherical subcompartment structure with dimensions of 300–400 nm (Müller et al. 2004). As indicated in Fig. 3, these domains are thought to be formed by a specific folding of the 30-nm chromatin fibre, to which the chain of nucleosomes associates under physiological salt concentrations (Woodcock and Dimitrov 2001; Hansen 2002; Adkins et al. 2004). However, the resolution of the light microscope is not sufficient to identify the organisation of the 30-nm chromatin fibre that leads to dense chromatin regions and the formation of the ~1-Mb domains. The different models that have been proposed are shown in Fig. 3: The radial-loop models propose small loops of roughly 100 kb arranged in rosettes (Hamkalo and Rattner 1980; Pienta and Coffey 1984; Münkel et al. 1999; Paul and Ferl 1999) (Figure 3a) while the random-walk/giant-loop (RW/GL) model proposes large loops of chromatin back-folded to an underlying structure (Sachs et al. 1995) (Fig.3b). In the chromonema model (Sedat and Manuelidis 1978; Belmont and Bruce 1994; Tumbar et al. 1999), the compaction of the 30-nm fibre is achieved by its folding into 60- to 80-nm fibres that undergo additional folding to 100- to 130-nm chromonema fibres (Figure 3c).

Fig. 2
figure 2

A common nuclear compartment is shared by ectopically expressed proteins and nuclear bodies. Crystalline Mx1-YFP particles are located next to a Cajal bodies and b nuclear Speckles. Within this compartment other ectopically expressed proteins are also localised, as shown for Mx1-YFP and Xenopus laevis NLS-vimentin at c moderate and d high expression levels of Xenopus NLS-vimentin. These proteins are localised to regions with low chromatin density, namely, outside of chromosome territories or in intra-chromosomal openings, as demonstrated with BrdU-labelled chromosome territories for e Mx1-YFP particles and f human NLS-vimentin. Confocal sections of HeLa cells (a –d) and MCF7 cells (e, f). Scale bars represent 10 μm.

Full size image
Fig. 3
figure 3

Hypothetical models for the folding of the chromatin fibre during interphase, leading to the formation of 1-Mb chromatin domains. a the radial-loop model, b the random-walk/giant-loop (RW/GL) model or c the chromonema model.

Full size image

The local variations of the chromatin density in the interphase nucleus are apparent when staining the DNA with fluorescent dyes like DAPI and Hoechst 33258 or labelling of the chromatin by incorporation of fluorescent histones (reviewed in Cremer and Cremer 2001). Nucleosome concentrations from 10 μM to 250 μM with an average value of 140 μM have been measured in the HeLa nucleus (Weidemann et al. 2003). Besides the chromatin density variations on a length scale of several hundred nanometers that have been assigned to the existence of the above-mentioned 1-Mb domains, regions in the micrometre-length scale at the nuclear periphery around the nucleolus and at the centromeres also display a high degree of compaction. These dense chromatin regions are often referred to as heterochromatin as opposed to the less dense euchromatin (reviewed in Dillon and Festenstein 2002; Dillon 2004). Heterochromatin was originally characterised by its intense staining in light microscopy images (Heitz 1928) and tight coiling observed in electron micrographs (Davies 1967). It has been described as containing increased DNA methylation at cytosines, specific histone modification patterns like methylation of lysine 9 on histone H3 and histone hypoacetylation, binding of heterochromatin protein 1 (HP1), interaction with non-coding RNA and activities of the RNAi-mediated silencing machinery (Bannister et al. 2001; Lachner et al. 2001; Chadwick and Willard 2003; Hall and Lawrence 2003; Fejes Tóth et al. 2004; Grewal and Rice 2004; Maison and Almouzni 2004). Regions of so-called facultative heterochromatin are known, which display a transition from a more open transcriptionally active conformation into a biologically inactive conformation (Dillon and Festenstein 2002; Dillon 2004). The most prominent example of facultative heterochromatin is the inactivation of one of the two X chromosomes in mammalian females, which shows a strong chromatin compaction during embryogenesis into a dense structure referred to as the Barr body (Chadwick and Willard 2003; Hall and Lawrence 2003; Dillon 2004). The relation of a dense heterochromatic state with a biologically inactive chromatin conformation has led to the concept that the biological activity of chromatin is regulated via its accessibility to protein factors (reviewed in Dillon and Festenstein 2002).

Size-dependent accessibility of chromatin

Depending on the degree of compaction, chromatin regions have different accessibilities. This has been visualised by progressive exclusion of microinjected and fluorescently labelled dextrans with increasing size from 4 kDa to 2.5 MDa (Fig. 4) (Görisch et al. 2003; Verschure et al. 2003). Dextrans are polymers that adopt a dynamic random-coil conformation. The size of such a particle can be described by its radius of gyration RG (Seksek et al. 1997; Lukacs et al. 2000), which characterises the average distance of atoms from the centre of mass of the molecule. The corresponding diameter (2×RG) serves as an estimate for the minimal extensions of open regions to which the molecule has access (Lénárt et al. 2003). Approximating that the RG value of a random coil is equivalent to the radius r of a much more rigid spherical protein, a corresponding protein molecular weight M can be calculated according to Eq. 1.

$$ M = \frac{4} {3}\,\pi \,r^3 \frac{{N_{\text{A}} }} {{\bar v}} $$
(1)
Fig. 4
figure 4

Nuclear accessibility can be examined by an increase in granularity and accumulation of microinjected FITC dextrans with increasing size in HeLa cells: a 42 kDa, b 148 kDa, c 464 kDa and d 2.5 MDa FITC-dextrans. Scale bars represent 10 μm.

Full size image

The average value of the partial specific protein volume is \(\bar v\)=0.73 ml g−1 (Durchschlag 1986) and NA is the Avogadro number 6.022×1023 mol−1. Up to a size of ~70 kDa (RG~11 nm (Seksek et al. 1997)), a mostly unlimited nuclear access for dextrans has been found (Görisch et al. 2003; Verschure et al. 2003), i.e. the dextrans are homogenously distributed. According to Eq. 1, this would correspond to a molecular weight of M=4.6 MDa for a spherical protein complex, suggesting that within the resolution of the light microscope (~250 nm lateral), all chromatin regions are accessible to even very large protein complexes. In agreement with this estimate, it has been shown that GFP (M=27 kDa) (Wachsmuth et al. 2000) and the RNA polymerase II complex (M=600 kDa) (Verschure et al. 2003) are homogenously distributed throughout the nucleus. At a dextran size of ~150 kDa, nuclear access becomes increasingly restricted. Dextrans of ~500 kDa are clearly excluded from putative heterochromatic regions while the 2.5 MDa FITC dextrans are excluded from chromatin in general and are confined to nuclear regions referred to as dextran patches (Görisch et al. 2003). These observations are in agreement with a diffusion analysis where FITC dextran particles up to 580 kDa are still mobile within the nucleoplasm whereas 2 MDa FITC-dextran particles are essentially immobile (Lukacs et al. 2000). Upon temperature shifts, no change in this behaviour can be measured (Braga et al. 2004).

It can be estimated that the average spacing of the 30-nm chromatin fibre for a nucleosome concentration in the nucleus of ~100 μM is around 100 nm; i.e. would allow the passage of complexes up to a size of ~70 nm. Even the dense chromatin regions, where a nucleosome concentration of ~260 μM has been measured, would still be accessible to particles of ~30 nm (Weidemann et al. 2003). Indeed, an exclusion from dense chromatin regions has been observed for FITC dextrans of ~150 kDa (Görisch et al. 2003), which corresponds to a size of ≥30 nm (Seksek et al. 1997). The nuclear accessibility is not only dependent on the size of the particle but will also reflect interactions with nuclear structures. This is apparent from the analysis of the effect of negative charges on the dextran distribution. While small anionic dextrans are distributed like non-charged dextrans, large anionic dextrans of 500 kDa are concentrated in nuclear foci and almost completely excluded from chromatin in contrast to non-charged dextrans of the same size (Görisch et al. 2003).

Diffusion of large particles in the nucleus

The mobility of protein complexes can be studied by a number of techniques, including FRAP (fluorescence recovery after photobleaching) and fluorescence correlation spectroscopy (FCS) (Bastiaens and Pepperkok 2000; Lippincott-Schwartz et al. 2001), e.g. the RNA polymerase II (Kimura et al. 2002) or the above-mentioned GFP (Wachsmuth et al. 2000). For larger particles that can be detected in the light microscope, the mobility is analysed by methods of single particle tracking (SPT) in which a time series of images is collected to determine the movement of the particle over time (Qian et al. 1991). In these experiments, nuclear particles usually displayed no directed movements in the nucleus. In addition, the observed accessibility and mobility of the particles are size-dependent and frequently reflect features of the chromatin environment. Furthermore, the apparent speed with which a particle translocates from its start position depends on the observation time if the velocity of the movement is defined as the distance travelled by the body divided by the corresponding time period. A more useful parameter to describe this random movement, independent of the observation time, is the diffusion coefficient D of the particle. This parameter is determined from the mean square displacement (MSD) or <d2> a molecule travels in a given time period Δt. It is computed on a time series of fluorescence images after corrections for any movements of the whole cell (Bacher et al. 2004; Rieger et al. 2004), according to Eq. 2.

$$ {\text{MSD}} = \left\langle {{\text{d}}^2 \left( {\Delta t} \right)} \right\rangle = \left\langle {\left[ {{\text{d}}\left( t \right) - {\text{d}}\left( {t + \Delta t} \right)} \right]^2 } \right\rangle $$
(2)

The squared distance d2 between the two positions d(t) and d(t + Δt) of the particle before and after a time Δt is measured, averaged for all identical values of Δt and then plotted against time Δt. For a diffusion process in n dimensions according to Eq. 3 it is

$$ {\text{MSD}} = \left\langle {{\text{d}}^2 \left( {\Delta t} \right)} \right\rangle = 2n\,\Gamma \,\Delta t^\alpha $$
(3)

with Γ being a generalised transport coefficient, which for a value of α=1 equals the diffusion coefficient D. In this case, a plot of MSD versus time will yield a straight line with a slope of 2nD. It is noted that for an isotropic movement of the particle in three dimensions (n=3), the value determined for D from projections to two dimensions (n=2) will be the same as that obtained from the analysis of 3D data sets, as shown, for example, in the supplementary material of reference (Görisch et al. 2004). For α≠1, an apparent time dependence of the diffusion coefficient is observed, as the slope of the MSD versus Δt plot is no longer constant and one speaks of “anomalous diffusion”. These deviations of the mobility from that of free diffusion can be further distinguished as confined diffusion, obstructed diffusion or directed motion (Sako and Kusumi 1994; Simson et al. 1998; Wachsmuth et al. 2000; Görisch et al. 2004) (Fig.5). Using a logarithmic plot of the MSD versus time relationship, a graphic test for anomalous diffusion can be conducted (Saxton 1994; Platani et al. 2002). An alternative approach is to simply fit the data to Eq. 3 and then characterise the type of diffusion by the value of the coefficient α. Experimental values of α<0.1 may be considered as characteristic of confined diffusion (Fig. 5a): 0.9<α<1.1 indicates free diffusion (Fig. 5c); 0.1<α<0.9 is interpreted as obstructed diffusion (Fig. 5b); α>1.1 represents a directed motion (Fig. 5d). It should be noted that due to random variations even for the case of free diffusion, a value close to α=1 will only be obtained if a sufficiently high number of data points are averaged. For the movement of a free-diffusing particle, time periods can be selected from a single trajectory in which, due to random fluctuation, the particle moves with an apparent higher or lower MSD versus time. In order to distinguish these variations from the existence of distinct classes of particles, a histogram of the values of α or other parameters derived from SPT can be a useful diagnostic tool. If systematic deviations of the MSD data versus time are observed for a fit to Eq. 3, more complex descriptions have to be derived for a quantitative description of the particle mobility. Recently, we introduced the so-called “moving corral” (MC) model that describes the mobility of nuclear particles according to Eq. 4.

$$ {\text{MSD}} = \left\langle {r_{\text{c}} ^2 } \right\rangle \,\left( {1 + \frac{{4D_{\text{c}} t}} {{\left\langle {r_{\text{c}} ^2 } \right\rangle }}} \right)\left\lfloor {1 - \exp \left( { - \frac{{4D_{\text{b}} t}} {{\left\langle {r_{\text{c}} ^2 } \right\rangle }}} \right)} \right\rfloor \, $$
(4)
Fig. 5
figure 5

According to the mean square displacement, four different types of mobility can be distinguished: a confined diffusion, b obstructed diffusion, c free diffusion and d directed motion. The black squares represent measured data using Mx1-YFP particles while the red line represents the fit according to Eq. 4.

Full size image

In the MC model, the overall mobility of the particle is described by a corral size of diameter rc in which the particle experiences confined diffusion with a diffusion coefficient Db. These chromatin corrals can then translocate according to a free diffusion model with a diffusion coefficient Dc. The derivation of this model has been described in detail previously (Görisch et al. 2004).

Chromatin dynamics

The diffusional properties of chromatin have been studied first by monitoring the mobility of a photobleached spot in dye-labelled chromatin (Abney et al. 1997). For individual tracking of subchromosomal regions, fluorescently labelled nucleotides were incorporated and diluted through subsequent segregation (Zink et al. 1998; Manders et al. 1999). The diffusion coefficients of 1.5×10−5 μm2 s−1 for foci located on different territories and 5.0×10−6 μm2 s−1 for foci located on the same territory were reported (Bornfleth et al. 1999).

As a technique to trace individual chromosome loci during interphase in living eukaryotic cells, the lac operator/LacI-GFP system was introduced by Belmont and co-workers (Robinett et al. 1996; Straight et al. 1996). It is based on the high-affinity binding of a bacterial lac-repressor protein fused with a GFP protein domain to multiple repeats of its recognition sequence stably integrated into the genome. Using the lacO/LacI-GFP system in mammalian cells as well as in yeast and Drosophila, rapid but locally restricted movements were found on a length scale of 0.4–0.5 μm (Marshall et al. 1997; Heun et al. 2001; Vazquez et al. 2001; Chubb et al. 2002). The movements of different sites in human living cells revealed that loci at nucleoli or the nuclear periphery are significantly less mobile than more nucleoplasmic loci. For the latter sites, a mean diffusion coefficient in human cells of 1.3×10−4 μm2 s−1 was reported (Chubb et al. 2002). As an alternative approach to the lac-repressor system, peptide nuclei acid probes (PNAs) directed against specific sequences were also used to analyse the diffusional properties of chromatin (Molenaar et al. 2003). Using PNAs to analyse the diffusion of telomeres in human cells showed that the majority of telomeres (~90%) revealed constrained diffusive movement (within confined nuclear regions of radius of ~230 nm) with a diffusion coefficient similar to that determined for nucleoplasmic lac arrays in human cells (Table 1).

Table 1 Corral size and diffusion coefficients of particles in the nucleus and chromatin loci.
Full size table

Mobility of inert particles and nuclear bodies in the nucleus

The diffusion properties of inert nuclear particles, like beads or crystalline protein particles within the chromatin environment, were determined using SPT in a number of studies. Fluorescent nanospheres with a diameter of 100 nm microinjected into mouse 3T3 cells demonstrated nuclear movements, which were briefly restricted to a small cage or corral with an average diameter of 290±50 nm. The MSD of the nanospheres in the nucleus showed a slight increase in time from 0 s to 0.1 s, an apparent plateau at 0.1–1 s (restricted motion within the corral) and a pronounced increase from 1 s to 10 s, which can be interpreted as escape of the nanosphere from the corral-type confinement (Tseng et al. 2004). Corralling was also observed using 100-nm polystyrene particles and NLS-vimentin bodies (Bacher et al. 2004). Crystalline nuclear particles (1.3±0.06 μm) formed by murine Mx1-YFP showed a displacement of the particle centre within a corral with an average diameter of 600 nm (Fig. 6). We speculate that the regions accessible for these particles correspond to regions of lower chromatin density, as visualised by the weak staining with Hoechst 33258 shown in Fig. 6a, b. In addition, analysis of the Mx1-YFP particles indicated the presence of a slower component. This is assigned to movements of the surrounding chromatin leading to a translocation of the particles. Accordingly, movement of the Mx1-YFP particles could be accurately described by the MC model in which the nuclear bodies diffuse in and with a chromatin “corral” in the nucleus (Görisch et al. 2004).

Fig. 6
figure 6

Examples of corralled motion of nuclear particles. a and b show two different in vivo trajectories of an Mx1-YFP particle in a HeLa cell nucleus with Hoechst-33258-stained chromatin. The first and the last image of a time series taken over 80 s (a) and 270 s (b) and a maximum intensity projection of all images of the time series are shown. The expected average MSD determined from the trajectories of 100 Mx1-YFP particles (Görisch et al. 2004) is outlined by the dashed white circle. At short times (<1 min), the centre of mass of the body displays relatively fast movements in a corral with a radius rc=0.28±0.03 μm. At longer observation times, a slow translocation of the region is observed in which the particle displays a corralled movement.

Full size image

Promyelocytic leukaemia and Cajal bodies displayed very similar average mobilities, as the biologically inert Mx1-YFP particles of similar size and their translocation in the nucleus could also be quantitatively described by the MC model (Görisch et al. 2004). This suggests that the body mobility reflects to a large extent the dynamics and accessibility of the surrounding chromatin environment. This view is supported by the observed co-localisation of nuclear bodies (discussed in Misteli 2001). In addition, chromatin regions around the nucleolus appeared to be devoid of nuclear bodies, consistent with the finding that the chromatin around the nucleolus is frequently more strongly condensed. In this context, it is also noteworthy that a significant fraction of mRNA molecules displayed corralled movements (Shav-Tal et al. 2004).

It has been shown that Cajal bodies are associated to several different snRNA and snoRNA gene loci as well as histone gene loci (Gall et al. 1981; Callan et al. 1991; Frey and Matera 1995; Schul et al. 1998; Jacobs et al. 1999). For PML bodies, an occasional association to the MHC gene region (Shiels et al. 2001), telomeres (Molenaar et al. 2003) and actively transcribed regions has been reported (Wang et al. 2004). These preferred localisations of nuclear bodies could be explained either by differences in the accessibility of nuclear regions and/or the binding to chromatin as it has been proposed for Cajal bodies (Platani et al. 2002). The analysis of a hamster cell line revealed that a fraction of PML bodies moved over longer distances in a directed and energy-dependent manner, suggesting that in this case also, active transport processes are involved. However, these faster-moving bodies were not observed in HeLa cells (Muratani et al. 2002; Görisch et al. 2004). The somewhat different interpretations with respect to the effect of energy depletion using sodium azide—increased Cajal body diffusion (Platani et al. 2002), decreased PML body (Muratani et al. 2002) and mRNP diffusion (Shav-Tal et al. 2004)—might indicate that multiple processes are effected. We and others observed that ATP depletion by treatment with sodium azide with or without deoxyglucose induced reversible chromatin condensation (Görisch et al. 2004; Shav-Tal et al. 2004). This could also be related to the ATP-dependent activity of chromatin remodelling complexes (Corona and Tamkun 2004).

Model for nuclear body movement in the nucleus

In summary, the analysis of dextrans and nanospheres of a defined size indicates that the size-dependent accessibility, the confinement of the movement (“corralling”) and the dynamics of chromatin are important determinants for the mobility of large particles in the nucleus. Several conclusions can be arrived at:

The accessibility of chromatin for even very large protein complexes up to several MDa molecular weight (~20 nm in diameter) is essentially unrestricted within the resolution of the light microscope of ~250 nm. Particles of a size ≥30 nm, like 150 kDa and larger FITC dextrans, are progressively excluded from dense chromatin regions. This suggests that these bona fide heterochromatic regions have a dense chromatin fibre organisation with mesh sizes ≤30 nm in diameter, as discussed above. Larger particles with sizes around 100 nm (100-nm-diameter nanospheres, 2.5-MDa dextrans) are completely excluded from this chromatin mesh (Görisch et al. 2003; Tseng et al. 2004). This is certainly true also for the nuclear Cajal and PML bodies with a size of ~1 μm, which will therefore have access to only a subspace of the nucleus. Any movement of these bodies over distances above a few hundred nanometers will require a chromatin reorganisation that allows the separation of chromatin subdomains to create accessible regions within and through the chromatin network. The interface between chromosome territories (ICD) would provide such a subcompartment. This has been experimentally demonstrated for cells with distinctly different karyotypes by the above-mentioned ability of NLS-vimentin filaments to form between chromosome territories (Bridger et al. 1998; Reichenzeller et al. 2000; Scheuermann et al. 2005). In agreement with this model, a co-localisation of nuclear bodies and NLS vimentin in this ICD space has been observed (Zirbel et al. 1993; Bridger et al. 1998; Richter et al. 2005). Several experiments discussed above are summarised in Table 1 and indicate that the chromatin environment allows a random translocation of large particles in the nucleus by several hundred nanometers in the time scale of a few minutes or less. The movement can be described by a diffusion-type mobility of the particles’ centre of mass in a corral with a radius of 200–300 nm (Table 1). This value is strikingly similar to the region in which the chromatin loci itself display a random movement. Furthermore, the chromatin locus mobility occurs with an apparent diffusion coefficient of up to ~10−4 μm2 s−1 (Chubb et al. 2002; Molenaar et al. 2003; Görisch et al. 2004), which is very similar to the diffusion coefficient calculated for the translocation of the chromatin corrals.

The experiments reviewed here, involving inert fluorescent macromolecules like dextrans or ectopically expressed proteins like NLS-vimentin filaments, as well as measurements of the mobility of nuclear bodies, nanospheres and chromatin loci, are included in the model presented in Fig. 7. As shown schematically on the left panel of Fig. 7, certain restrictions of the accessibility are imposed by the folding of the interphase chromatin fibre, as discussed above. The characteristics of the chromatin environment that define the mobility of PML and Cajal nuclear bodies are included in the MC model shown on the right side of Fig. 7. The size-dependent accessibility is reflected by the radius of the corral rc, in which the body displays relatively fast movements with a diffusion coefficient Db (Eq. 4). The movement of these corrals due to chromatin rearrangements is described by the diffusion coefficient Dc. This is obviously a simplification valid only for observation times in the order of several minutes since the maximal translocation of the corrals is limited by the finite size of the nucleus.

Fig. 7
figure 7

Models for nuclear accessibility and body diffusion. The left image shows a section of a cell nucleus with macromolecules of different sizes that have, according to their sizes, unlimited nuclear access or are excluded from heterochromatic regions. The right image illustrates the moving corral model: nuclear bodies are enclosed in a chromatin “corral”, which is defined by the chromatin accessibility. Within this accessible domain, they diffuse with the diffusion coefficient Db while the chromatin corrals translocate within the nucleus with the diffusion coefficient Dc, as indicated by the blue arrow.

Full size image

As discussed above, the nuclear subspace accessible for nuclear bodies is likely to include the interface between chromosome territories that allows a movement of the bodies by a transient separation of chromatin domains. By random movements, the nuclear bodies explore this accessible space and are expected to be localised more frequently in these more open chromatin regions. In as much as they coincide with regions of active gene transcription, they would also constitute possible biological targets of PML and Cajal bodies. This hypothesis can be tested by analysing the intra-nuclear mobility and localisation of multiple nuclear components simultaneously. With the availability of improved methods for imaging and data analysis in conjunction with new labelling techniques, it becomes possible to study nuclear bodies, RNA and chromatin loci in parallel within the same living cell and to identify their functional relations.