Large-scale chromatin structure and function.

Recent results in living cells have now established the existence of levels of chromatin folding above the 30 nm fiber within interphase chromosomes. We discuss the potential functional impact of this large-scale chromatin organization, including its possible role in regulating gene expression.

Reproducible but dynamic positioning of DNA in chromosomes during mitosis.

How DNA is folded into chromosomes is unknown. Mitotic chromosome banding shows reproducibility in longitudinal compaction at a resolution of several megabase pairs, but it is less clear whether DNA sequences are targeted laterally to specific locations. The in vitro chromosome assembly of prokaryotic DNA suggests that there is a lack of sequence requirements for chromosome condensation, implying an absence of DNA targeting. Protein extraction experiments indicate, however, that specific DNA sequences may bind to a chromosome scaffold. Chromosome banding patterns, using dyes with differential sequence specificity, have been interpreted to result from the alignment of AT-rich sequences in a partially helically folded chromosome scaffold. But fluorescence in situ hybridization experiments, perhaps owing to technical limitations, have shown at best only slight deviation from a random, lateral sequence distribution. Here we show that there is highly reproducible targeting of specific chromosome segments to the metaphase chromatid axis, but that these segments localize to the periphery of prophase and telophase chromosomes. Unfolding intermediates during anaphase and telophase suggest that sequence repositioning occurs through the global uncoiling of an underlying chromatid structure.

Three-dimensional reconstruction of painted human interphase chromosomes: active and inactive X chromosome territories have similar volumes but differ in shape and surface structure.

This study provides a three-dimensional (3D) analysis of differences between the 3D morphology of active and inactive human X interphase chromosomes (Xa and Xi territories). Chromosome territories were painted in formaldehyde-fixed, three-dimensionally intact human diploid female amniotic fluid cell nuclei (46, XX) with X-specific whole chromosome compositive probes. The colocalization of a 4,6-diamidino-2-phenylindole dihydrochloride-stained Barr body with one of the two painted X territories allowed the unequivocal discrimination of the inactive X from its active counterpart. Light optical serial sections were obtained with a confocal laser scanning microscope. 3D-reconstructed Xa territories revealed a flatter shape and exhibited a larger and more irregular surface when compared to the apparently smoother surface and rounder shape of Xi territories. The relationship between territory surface and volume was quantified by the determination of a dimensionless roundness factor (RF). RF and surface area measurements showed a highly significant difference between Xa and Xi territories (P < 0.001) in contrast to volume differences (P > 0.1). For comparison with an autosome of similar DNA content, chromosome 7 territories were additionally painted. The 3D morphology of the chromosome 7 territories was similar to the Xa territory but differed strongly from the Xi territory with respect to RF and surface area (P < 0.001).

DNA-repair protein distribution along the tracks of energetic ions.

A simple model of homogenous chromatin distribution in HeLa-cell nuclei suggests that the track of an energetic ion hits 30 nm chromatin fibers with a mean distance of 0.55 mum. To test this assumption, living HeLa-cells were irradiated at the irradiation setup of the ion microprobe SNAKE using the ion beams provided by the Munich 14 MV tandem accelerator. After irradiation, the distribution of 53BP1 protein foci was studied by immunofluorescence. The observed 53BP1 distribution along the tracks of 29 MeV (7)Li ions and 24 MeV (12)C ions differed significantly from the expectations resulting from the simple chromatin model, suggesting that the biological track structure is determined by cell nuclear architecture with higher order organisation of chromatin.

Compartmentalization of interphase chromosomes observed in simulation and experiment.

Human interphase chromosomes were simulated as a flexible fiber with excluded volume interaction, which represents the chromatin fiber of each chromosome. For the higher-order structures, we assumed a folding into 120 kb loops and an arrangement of these loops into rosette-like subcompartments. Chromosomes consist of subcompartments connected by small fragments of chromatin. Number and size of subcompartments correspond with chromosome bands in early prophase. We observed essentially separated chromosome arms in both our model calculations and confocal laser scanning microscopy, and measured the same overlap in simulation and experiment. Overlap, number and size of chromosome 15 subcompartments of our model chromosomes agree with subchromosomal foci composed of either early or late replicating chromatin, which were observed at all stages of the cell cycle and possibly provide a functionally relevant unit of chromosome territory compartmentalization. Computed distances of chromosome specific markers both on Mb and 10-100 Mb scale agree with fluorescent in situ hybridization measurements under different preparation conditions.

Nuclear architecture and the induction of chromosomal aberrations.

Progress in fluorescence in situ hybridization, three dimensional microscopy and image analysis has provided the means to study the three-dimensional structure and distribution of chromosome territories within the cell nucleus. In this contribution, we summarize the present state of knowledge of the territorial organization of interphase chromosomes and their topological relationships with other macromolecular domains in the human cell nucleus, and present data from computer simulations of chromosome territory distributions. On this basis, we discuss models of chromosome territory and nuclear architecture and topological consequences for the formation of chromosome exchanges.

The nuclear distribution of Polycomb during Drosophila melanogaster development shown with a GFP fusion protein.

The chromatin protein Polycomb (PC) is necessary for keeping homeotic genes repressed in a permanent and heritable manner. PC is part of a large multimeric complex (PcG proteins) involved in generating silenced chromatin domains at target genes, thus preventing their inappropriate expression. In order to assess the intranuclear distribution of PC during mitosis in different developmental stages as well as in the germ line we generated transgenic fly lines expressing a PC-GFP (Green Fluorescent Protein) fusion protein. Rapidly dividing nuclei were found to display a rather homogeneous PC-GFP distribution. However, with increasing differentiation a pronounced subnuclear pattern was observed. In all investigated diploid somatic tissues the bulk of PC-GFP fusion protein is depleted from the chromosomes during mitosis: however, a detectable fraction remains associated. In the male germ line in early spermatogenesis, PC-GFP was closely associated with the chromosomal bivalents and gradually lost at later stages. Interestingly, we found that PC is associated with the nucleolus in spermatocytes, unlike somatic nuclei. In contrast to mature sperm showing no PC-GFP signal the female germ line retains PC in the germinal vesicle.

Evidence against a looped structure of the inactive human X-chromosome territory.

Multicolor fluorescence in situ hybridization with a whole chromosome composite probe for the X-chromosome and microdissection probes for the Xp and Xq arms, as well as for the Xp terminal, Xq terminal, and X centromer specific subregional probes, was applied to three-dimensional (3D) preserved human female amniotic fluid cell nuclei. Confocal laser scanning microscopy and three-dimensional image analysis demonstrated distinctly separated Xp arm and Xq arm domains. 3D distance measurements revealed a high variability of intrachromosomal distances between Xpter, Xcen, and Xqter specific probes within both X territories. A 3D distance measurement error of +/- 70 nm was found in control experiments using quartz glass microspheres labeled with different fluorochromes. Our data argue against the hypothesis of Walker et al. (1991, Proc. Natl. Acad. Sci. USA 88, 6191-6195) that a looped structure of the inactive X territory is formed by tight telomere-telomere associations.

Microirradiation of cells with energetic heavy ions.

The ion microprobe SNAKE at the Munich 14 MV tandem accelerator achieves beam focussing by a superconducting quadrupole doublet and can make use of a broad range of ions and ion energies, from 20 MeV protons to 200 MeV gold ions. Because of these properties, SNAKE is particularly attractive for biological microbeam experiments. Here we describe the adaptation of SNAKE for microirradiation of cell samples. This includes enlarging of the focal distance in order to adjust the focal plane to the specimen stage of a microscope, construction of a beam exit window in a flexible nozzle and of a suitable cell containment, as well as development of procedures for on-line focussing of the beam, preparation of single ions and scanning by electrostatic deflection of the beam. When irradiating with single 100 MeV (16)O ions, the adapted set-up permits an irradiation accuracy of 0.91 microm (full width at half maximum) in the x-direction and 1.60 microm in the y-direction, as demonstrated by retrospective track etching of polycarbonate foils. Accumulation of the repair protein Rad51, as detected by immunofluorescence, was used as a biological track detector after irradiation of HeLa cells with geometric patterns of counted ions. Observed patterns of fluorescence foci agreed reasonably well with irradiation patterns, indicating successful adaptation of SNAKE. In spite of single ion irradiation, we frequently observed split fluorescence foci which might be explained by small-scale chromatin movements.

The 3D positioning of ANT2 and ANT3 genes within female X chromosome territories correlates with gene activity.

The three-dimensional positioning of the X-chromosomal adenine nucleotide translocase genes, ANT2 and ANT3, were compared in the active and inactive X chromosome territories (Xa and Xi) of female human amniotic fluid cell nuclei. ANT2 is located in Xq24-q25 and is transcriptionally active on Xa, but inactive on Xi. ANT3 is located in the pseudoautosomal region Xp22.3 and escapes X-inactivation. Three-color fluorescence in situ hybridization, confocal laser scanning microscopy, and three-dimensional image analysis revealed that transcriptionally active ANT2 and ANT3 genes were positioned more peripheral within their chromosome territory than the inactive ANT2 gene. The position of the latter was significantly more interior in the Xi territory. Although the volumes of both X territories were similar, 3D distances between ANT2 and ANT3 were significantly smaller in Xi compared to Xa territories reflecting different territory shapes. Our data show a correlation between 3D positioning and transcriptional activity of these X-specific genes.

Aspects of three-dimensional chromosome reorganization during the onset of human male meiotic prophase.

The three-dimensional morphology and distribution of human chromosomes 3 were studied in nuclei of spermatogonia and spermatocytes I from formaldehyde-fixed human testis sections. Chromosome arms, pericentromeres and telomeric regions were painted by a three-color, five-probe fluorescence in situ hybridization protocol. Light optical serial sections of premeiotic and meiotic nuclei obtained by confocal laser scanning microscopy revealed that premeiotic chromosomes 3 are separate from each other and occupy variably shaped territories, which are sectored in distinct 3 p- and q-arm domains. Three-dimensional reconstructions of the painted chromosome domains by a Voronoi tessellation approach showed that mean chromosome volumes did not differ significantly among the premeiotic and meiotic stages investigated. A significant increase in surface area and reduction of dimensionless 'roundness factor' estimates of arm domains indicated that the restructuring of spatially separate chromosome territories initiates during preleptotene. Telomeric regions, which in meiotic stem cells located predominantly in arm-domain chromatin, showed a redistribution towards the domain surface during this stage. At leptotene homologues were generally misaligned and displayed intimate intermingling of non-homologous chromatin. Pairing initiated at the ends of bent zygotene chromosomes, which displayed a complex surface structure with discernible sister chromatids. The results indicate that, in mammals, homology search is executed during leptotene, after remodeling of chromosome territories.

Separate and variably shaped chromosome arm domains are disclosed by chromosome arm painting in human cell nuclei.

Fluorescence in situ hybridization (FISH) with microdissection probes from human chromosomes 3 and 6 was applied to visualize arm and subregional band domains in human amniotic fluid cell nuclei. Confocal laser scanning microscopy and quantitative three-dimensional image analysis showed a pronounced variability of p- and q-arm domain arrangements and shapes. Apparent intermingling of neighbouring arm domains was limited to the domain surface. Three-dimensional distance measurements with pter and qter probes supported a high variability of chromosome territory folding.