Transcribed human ribosomal RNA genes are attached to the nuclear matrix

Myc-induced anchorage of the rDNA IGS region to nucleolar matrix modulates growth-stimulated changes in higher-order rDNA architecture

Chromatin domain organization and the compartmentalized distribution of chromosomal regions are essential for packaging of deoxyribonucleic acid (DNA) in the eukaryotic nucleus as well as regulated gene expression. Nucleoli are the most prominent morphological structures of cell nuclei and nucleolar organization is coupled to cell growth. It has been shown that nuclear scaffold/matrix attachment regions often define the base of looped chromosomal domains in vivo and that they are thereby critical for correct chromosome architecture and gene expression. Here, we show regulated organization of mammalian ribosomal ribonucleic acid genes into distinct chromatin loops by tethering to nucleolar matrix via the non-transcribed inter-genic spacer region of the ribosomal DNA (rDNA). The rDNA gene loop structures are induced specifically upon growth stimulation and are dependent on the activity of the c-Myc protein. Matrix-attached rDNA genes are hypomethylated at the promoter and are thus available for transcriptional activation. rDNA genes silenced by methylation are not recruited to the matrix. c-Myc, which has been shown to induce rDNA transcription directly, is physically associated with rDNA gene looping structures and the intergenic spacer sequence in growing cells. Such a role of Myc proteins in gene activation has not been reported previously.

Large-scale organization of ribosomal DNA chromatin is regulated by Tip5

The DNase I accessibility and chromatin organization of genes within the nucleus do correlate to their transcriptional activity. Here, we show that both serum starvation and overexpression of Tip5, a key regulator of ribosomal RNA gene (rDNA) repression, dictate DNase I accessibility, facilitate the association of rDNA with the nuclear matrix and thus regulate large-scale rDNA chromatin organization. Tip5 contains four AT-hooks and a TAM (Tip5/ARBP/MBD) domain, which were proposed to bind matrix-attachment regions (MARs) of the genome. Remarkably, the TAM domain of Tip5 functions as nucleolar localization and nuclear matrix targeting module, whereas AT-hooks do not mediate association with the nuclear matrix, but they are required for nucleolar targeting. These findings suggest a dual role for Tip5's AT-hooks and TAM domain, targeting the nucleolus and anchoring to the nuclear matrix, and suggest a function for Tip5 in the regulation of higher-order rDNA chromatin structure.

c-Myc induces changes in higher order rDNA structure on stimulation of quiescent cells

c-Myc is an oncogenic transcription factor capable of activating transcription by all three nuclear RNA polymerases, thus acting as a high-level coordinator of protein synthesis capacity and cell growth rate. c-Myc recruits RNA polymerase I-related transcription factors to the rDNA when quiescent cells are stimulated to re-enter the cell cycle. Using a model system of cell lines with variable c-Myc status, we show that on stimulation c-Myc rapidly induces gene loop structures in rDNA chromatin that juxtapose upstream and downstream rDNA sequences. c-Myc activation is both necessary and sufficient for this change in rDNA chromatin conformation. c-Myc activation induces association of TTF-1 with the rDNA, and c-Myc is physically associated with induced rDNA gene loops. The origins of two or more rDNA gene loops are closely juxtaposed, suggesting the possibility that c-Myc induces nucleolar chromatin hubs. Induction of rDNA gene loops may be an early step in the reprogramming of quiescent cells as they re-enter the growth cycle.

Rearrangement of chromatin domains during development in Xenopus

A dynamic change in the organization of different gene domains transcribed by RNA polymerase I, II, or III occurs during the progression from quiescent [pre-midblastula transition (pre-MBT)] to active (post-MBT) embryos during Xenopus development. In the rDNA, c-myc, and somatic 5S gene domains, a transition from random to specific anchorage to the nuclear matrix occurs when chromatin domains become active. The keratin gene domain was also randomly associated to the nuclear matrix before MBT, whereas a defined attachment site was found in keratinocytes. In agreement with this specification, ligation-mediated (LM)-PCR genomic footprinting carried out on the subpopulation of 5S domains specifically attached to the matrix reveals the hallmarks of determined chromatin after the midblastula transition. In contrast, the same analysis performed on the total 5S gene population does not reveal specific chromatin organization, validating the use of nuclear matrix fractionation to unveil active chromatin domains. These data provide a means for the determination of active chromosomal territories in the embryo and emphasize the role of nuclear architecture in regulated gene expression during development.

Intranuclear anchoring of repetitive DNA sequences: centromeres, telomeres, and ribosomal DNA

Centromeres, telomeres, and ribosomal gene clusters consist of repetitive DNA sequences. To assess their contributions to the spatial organization of the interphase genome, their interactions with the nucleoskeleton were examined in quiescent and activated human lymphocytes. The nucleoskeletons were prepared using "physiological" conditions. The resulting structures were probed for specific DNA sequences of centromeres, telomeres, and ribosomal genes by in situ hybridization; the electroeluted DNA fractions were examined by blot hybridization. In both nonstimulated and stimulated lymphocytes, centromeric alpha-satellite repeats were almost exclusively found in the eluted fraction, while telomeric sequences remained attached to the nucleoskeleton. Ribosomal genes showed a transcription-dependent attachment pattern: in unstimulated lymphocytes, transcriptionally inactive ribosomal genes located outside the nucleolus were eluted completely. When comparing transcription unit and intergenic spacer, significantly more of the intergenic spacer was removed. In activated lymphocytes, considerable but similar amounts of both rDNA fragments were eluted. The results demonstrate that: (a) the various repetitive DNA sequences differ significantly in their intranuclear anchoring, (b) telomeric rather than centromeric DNA sequences form stable attachments to the nucleoskeleton, and (c) different attachment mechanisms might be responsible for the interaction of ribosomal genes with the nucleoskeleton.

Role of nuclear architecture in the initiation of eukaryotic DNA replication

The eukaryotic genome is compacted in the cell nucleus, in a way that allows its faithful and ordered replication each cell cycle. Chromatin is organized into topologically constrained loops that are anchored to the nuclear matrix by specific attachment regions (SARs). Chromatin loops were proposed to correspond to replication units. In particular, it has been suggested that replication origins coincide with SARs. Critical examination of these hypotheses has long been hampered by the elusive nature of higher eukaryotic DNA replication origins and termini. In recent years, however, a number of loci have been mapped for both SARs and replication units, and studies on the nuclear localization of replicating DNA and replication proteins have begun. We review these data and argue that they question this model. We then try to delineate other aspects of chromosome compartmentalization and cell-cycle remodeling which might be responsible for the specification and activation of metazoan DNA replication origins.

Recruitment of damaged DNA to the nuclear matrix in hamster cells following ultraviolet irradiation

We examined the relationship between the nuclear matrix and DNA in the dihydrofolate reductase domain following irradiation of Chinese hamster cells with UV light. The fraction of matrix-bound DNA increased in transcribed and non-transcribed regions during a 3 h period after irradiation. However, no increase was observed with excision repair-deficient cells mutant for the ERCC1 gene. The major UV-induced lesion, the cyclobutane pyrimidine dimer, increased in frequency in the matrix-bound DNA 1 h after irradiation, in both transcribed and non-transcribed regions, but decreased subsequently. This phenomenon was also lacking in excision repair-deficient cells. These data demonstrate that recruitment of lesion-containing DNA to the nuclear matrix occurs following UV irradiation and suggest that this recruitment is dependent upon nucleotide excision repair. This is consistent with the concept of a 'repair factory' residing on the nuclear matrix at which excision repair occurs.

The architectural organization of nuclear metabolism

Nucleic acid metabolism is structurally organized in the nucleus. DNA replication and transcription have been localized to particular nuclear domains. Additional domains have been identified by their morphology or by their composition; for example, by their high concentration of factors involved in RNA splicing. The domain organization of the nucleus is maintained by the nuclear matrix, a nonchromatin nuclear scaffolding that holds most nuclear RNA and organizes chromatin into loops. The nuclear matrix is built on a network of highly branched core filaments that have an average diameter of 10 nm. Many of the intermediates and the regulatory and catalytic factors of nucleic acid metabolism are retained in nuclear matrix preparations, suggesting that nucleic acid synthesis and processing are structure-bound processes in cells. Tissue-specific and malignancy-induced variations in nuclear structure and metabolism may result from altered matrix architecture and composition.

Contributions of nuclear architecture to transcriptional control

Three parameters of nuclear structure contribute to transcriptional control. The linear representation of promoter elements provides competency for physiological responsiveness within the contexts of development as well as cycle- and phenotype-dependent regulation. Chromatin structure and nucleosome organization reduce distances between independent regulatory elements providing a basis for integrating components of transcriptional control. The nuclear matrix supports gene expression by imposing physical constraints on chromatin related to three-dimensional genomic organization. In addition, the nuclear matrix facilitates gene localization as well as the concentration and targeting of transcription factors. Several lines of evidence are presented that are consistent with involvement of multiple levels of nuclear architecture in cell growth and tissue-specific gene expression during differentiation. Growth factor and steroid hormone responsive modifications in chromatin structure, nucleosome organization, and the nuclear matrix that influence transcription of the cell cycle-regulated histone gene and the bone tissue-specific osteocalcin gene during progressive expression of the osteoblast phenotype are considered.

Chromosome length and DNA loop size during early embryonic development of Xenopus laevis

The looped organization of the eukaryotic genome mediated by a skeletal framework of non-histone proteins is conserved throughout the cell cycle. The radial loop/scaffold model envisages that the higher order architecture of metaphase chromosomes relies on an axial structure around which looped DNA domains are radially arranged through stable attachment sites. In this light we investigated the relationship between the looped organization and overall morphology of chromosomes. In developing Xenopus laevis embryos at gastrulation, the bulk of the loops associated with histone-depleted nuclei exhibit a significant size increase, as visualized by fluorescence microscopy of the fully extended DNA halo surrounding high salt treated, ethidium bromide stained nuclei. This implies a reduction in the number of looped domains anchored to the supporting nucleoskeletal structure. The cytological analysis of metaphase plates from acetic acid fixed whole embryos, carried out in the absence of drugs inducing chromosome condensation, reveals a progressive thickening and shortening of metaphase chromosomes during development. We interpret these findings as a strong indication that the size and number of DNA loops influence the thickness and length of the chromosomes, respectively. The quantitative analysis of chromosome length distributions at different developmental stages suggests that the shortening is timed differently in different embryonic cells.

Increased frequency of acrocentric chromosome association during colcemid treatment

The effect of Colcemid on satellite association frequencies was investigated in human cell lines U-937 GTB, SC-N-MC, HL-60, and Raji by silver staining method. Cell cultures were exposed to Colcemid at a concentration of 0.02 mg/ml for 15-60 minutes. As exposure time to Colcemid was increased, both the frequency of cells with satellite associations and the number of chromosomes involved in satellite associations increased significantly. Furthermore, the mean number of silver-stained chromosomes decreased with longer exposure time. Without Colcemid, we were able to obtain an adequate number of good-quality metaphase spreads. The data obtained may be of help in furthering research concerning satellite associations and in obtaining better quality chromosome preparations.

Binding of sequences from the 5'- and 3'-nontranscribed spacers of the rat rDNA locus to the nucleolar matrix

Nucleolar matrix structures were obtained under different extraction conditions from highly purified isolated nucleoli. Their ultrastructural appearance, protein composition and capacity to bind rDNA preferentially were studied in a model binding system. A region spanning approximately 25 kb in the rat ribosomal gene locus was screened for DNA sites capable of specifically interacting with the proteins of the nucleolar matrix (MARs). Two such sites were identified: one is located on an EcoRV-KpnI fragment in the 5'-nontranscribed spacer region, between two repetitive elements and close to the transcription initiation site; the other MAR is on a PvuII-BamHI fragment located in the 3'-nontranscribed region, encompassing an element 85% homologous to a B2-sequence. The two MARs are located in regions rich in polypyrimidine/polypurine tracks and contain a few elements homologous to the consensus sequence for topoisomerase II. This indicates that the "attachment sites" for the ribosomal genes belong to the same class of sequences as the MARs attaching the chromosomal DNA to the nuclear matrix.

The association of the human epsilon-globin gene with the nuclear matrix: a reconsideration

The association of the human epsilon-globin gene with the nuclear matrix was studied in erythroid and non-erythroid cell lines. Using a high salt method to prepare histone depleted nuclei we studied the association of variety of fragments covering a 7.8 kb region which contains the human epsilon-globin gene. We furthermore studied the association of a set of DNA fragments covering the 13 kb human G gamma/A gamma-globin gene domain, the 16 kb psi beta/delta-globin gene domain and the 10 kb beta-globin gene domain with the nuclear matrix of K562 and Raji cells. The results show that all fragments studied are easily released from the nuclear matrix, indicating no specific association. Summarizing our results we could say that a region starting 5.7 kb 5' to the human epsilon-globin gene and ending 4 kb 3' to the human beta-globin gene seems to contain no attachment sites with the nuclear matrix of both erythroid and non-erythroid cells.

A nuclear matrix protein binds very tightly to DNA in the avian beta-globin gene enhancer

Current evidence suggests that DNA is covalently attached to proteins in the nuclear matrix of eukaryotic cells and that specific DNA sequences are tightly associated with the nuclear matrix. However, it has not been documented that specific DNA sequences can become covalently attached to nuclear matrix protein. We have examined the binding of cloned DNA sequences that contain the avian beta-globin gene enhancer, a region previously shown to be matrix associated in erythroid cells in vivo, with nuclear matrices from several avian tissue sources to determine if covalent DNA-protein bonds are formed. Our results indicate that sequence-specific DNA-protein complexes that are resistant to denaturation by SDS, boiling, and phenol and disulfide reduction are formed. Excess protein, capable of forming very tight bonds with DNA that contains the beta-globin gene enhancer, is present in cells in which matrix attachment of this DNA sequence is not detected in vivo. Evidence is presented that suggests that the protein to which DNA forms very tight bonds is not topoisomerase II. These results are discussed in relation to current models of the nuclear matrix and the utility of in vitro assays of matrix attachment regions using cloned DNA.

Preferential association of a transcriptionally active gene with the nuclear matrix of rat fibroblasts transformed by a simian-virus-40-pBR322 recombinant plasmid

To study the relationship between the structural organization and function of the eukaryotic genome, DNA associated with nuclear matrix was analysed by using a transformed rat fibroblast cell line. The nuclear matrices were prepared from the isolated nuclei of pSV1-FR, a rat fibroblast cell line transformed by a pBR322-based recombinant plasmid containing an early gene region, which codes for large T-antigen, of simian virus 40. This transformed cell contained a single copy of the plasmid sequence integrated into the chromosomal DNA of the host cell. The early gene of this plasmid was constitutively expressed, as demonstrated by positive immunofluorescence staining of the cell for large T-antigen and by RNA-blot analysis for its specific mRNA. DNAs were extracted from whole isolated nuclei and nuclear-matrix preparations of the cells, and the relative amounts of the sequence similar to that of the plasmid were compared between these DNA preparations. By employing dot hybridization and Southern-blot analyses we found that the plasmid sequence was more enriched in the DNA extracted from the nuclear matrices than in the DNA extracted from the whole nuclei. When an albumin gene sequence that was not transcribed in this cell line was compared similarly as a control, we found no significant enrichment of this sequence in the DNA associated with the nuclear matrix. Our results strongly support the concept that a transcriptionally active gene is preferentially associated with the nuclear matrix.

Chromatin loop structure of the human X chromosome: relevance to X inactivation and CpG clusters

Part of the higher-order structure of chromatin is achieved by constraining DNA in loops ranging in size from 30 to 100 kilobase pairs; these loops have been implicated in defining functional domains and replicons and possibly in facilitating transcription. Because the human active and inactive X chromosomes differ in transcriptional activity and replication, we looked for differences in their chromatin loop structures. Since the islands of CpG-rich DNA at the 5' ends of X-linked housekeeping genes are the regions where functional differences in DNA methylation and nuclease sensitivity are found, we looked for scaffold association of these sequences after extraction of histones with lithium diiodosalicylate. Specifically, we examined the 5' CpG islands within the hypoxanthine phosphoribosyltransferase, glucose 6-phosphate dehydrogenase, P3, GdX, phosphoglycerate kinase type 1, and alpha-galactosidase loci in human lymphoblasts obtained from individuals with 1 to 4 X chromosomes. Although we detected no scaffold-associated regions near these genes, we found several such regions at the ornithine transcarbamylase and blood clotting factor IX loci. Our results suggest that the CpG islands are excluded from the nuclear scaffold and that even though transcriptionally active, housekeeping genes are less likely than X-linked tissue-specific genes to be scaffold associated. In all cases, the pattern of scaffold association was the same for loci on active and inactive X chromosomes.

Chromatin loop structure of the human X chromosome: relevance to X inactivation and CpG clusters

Part of the higher-order structure of chromatin is achieved by constraining DNA in loops ranging in size from 30 to 100 kilobase pairs; these loops have been implicated in defining functional domains and replicons and possibly in facilitating transcription. Because the human active and inactive X chromosomes differ in transcriptional activity and replication, we looked for differences in their chromatin loop structures. Since the islands of CpG-rich DNA at the 5' ends of X-linked housekeeping genes are the regions where functional differences in DNA methylation and nuclease sensitivity are found, we looked for scaffold association of these sequences after extraction of histones with lithium diiodosalicylate. Specifically, we examined the 5' CpG islands within the hypoxanthine phosphoribosyltransferase, glucose 6-phosphate dehydrogenase, P3, GdX, phosphoglycerate kinase type 1, and alpha-galactosidase loci in human lymphoblasts obtained from individuals with 1 to 4 X chromosomes. Although we detected no scaffold-associated regions near these genes, we found several such regions at the ornithine transcarbamylase and blood clotting factor IX loci. Our results suggest that the CpG islands are excluded from the nuclear scaffold and that even though transcriptionally active, housekeeping genes are less likely than X-linked tissue-specific genes to be scaffold associated. In all cases, the pattern of scaffold association was the same for loci on active and inactive X chromosomes.

Coordinate replication of dispersed repetitive sequences in Physarum polycephalum

The synchronous macroplasmodial growth phase of the slime mould Physarum polycephalum was used to study the in vivo replication of large chromosomal DNA segments. Newly replicated DNA was isolated at various points in S-phase by its preferential association with the nuclear matrix. This DNA was then used to probe cosmid clones of the Physarum genome. The results indicate that certain dispersed repetitive sequences in the genome are coordinately replicated. The observed pattern of replication may be due either to the presence of a replication origin within each repetitive sequence or to the systematic arrangement of these sequences around a replication origin. The latter appears more likely since the repetitive sequences are probably not randomly scattered within the genome.

Location and methylation pattern of a nuclear matrix associated region in the human pro alpha 2(I) collagen gene

Using both a 25 mM Lithium di-iodosalicylic acid (LIS) and a 2M NaCl extraction procedure to extract nuclear matrices from white cells we have identified a 0.9 kb nuclear matrix associated region (MAR) in the human pro alpha 2(I) collagen gene. The MAR is located towards the 3' coding end of the gene, it is completely associated with the matrix in transcriptionally inactive white cells but is incompletely associated with the matrix in transcriptionally active fibroblasts. Furthermore the methylation state of the fibroblast gene in the region coinciding with the MAR showed unique differences when compared to adjacent sites in the fibroblast gene and corresponding sites of the white cell gene.

Nuclear matrix associated DNA is preferentially repaired in normal human fibroblasts, exposed to a low dose of ultraviolet light but not in Cockayne's syndrome fibroblasts

In this study we addressed the questions as to whether repair is confined to the nuclear matrix compartment, analogous to replication and transcription and how repair events are distributed in DNA loops associated with the nuclear matrix. Pulse labelling of ultraviolet (254 nm) irradiated confluent human fibroblasts revealed that repair was preferentially located in nuclear matrix associated DNA in cells exposed to 5 J/m2. However, in cells exposed to 30 J/m2 repair approached a random distribution. The non-random distribution of repair label at 5 J/m2 was most pronounced directly after irradiation and gradually changed to a more random distribution within two hours after treatment. The results of pulse-chase experiments exclude the possibility of transient binding of repair sites to the matrix and favour the model of preferential repair of DNA sequences permanently associated with the nuclear matrix. Pronounced differences in distribution pattern of repair events in DNA loops were found among normal and UV-sensitive cell lines exposed to 5 J/m2. Repair in nuclear matrix associated DNA was 1.7 fold more efficient than in loop DNA in normal and xeroderma pigmentosum group D cells and over 3 fold in xeroderma pigmentosum group C cells. In Cockayne's syndrome fibroblasts repair in nuclear matrix DNA was found to be 2 fold less efficient than in loop DNA. This heterogeneity in distribution of repair correlates well with preferential removal of pyrimidine dimers from transcriptionally active DNA in normal and xeroderma pigmentosum group C cells and its absence in Cockayne's syndrome cells as recently reported by Mayne et al., 1988. The results suggest that Cockayne's syndrome cells have a defect in excision of UV-damage from transcriptionally active genes located proximal to the nuclear matrix. Xeroderma pigmentosum group C cells may possess a defect in DNA repair associated with chromatin regions outside transcriptionally active DNA.

Localization, in human placenta, of the tightly bound form of DNA methylase in the higher order of chromatin organization

In human placenta, the DNA of all subfractions of the third level of chromatin organization exhibits similar values of the methylcytosine-to-cytosine ratio. The tightly bound form of DNA methyltransferase is mostly recovered in the 'stripped loop' fraction, although, on the basis of the DNA content, the 'stripped loops' and the 'stripped matrix' appear to possess a similar amount of the enzyme. DNA methyltransferase activity is instead totally absent from the 'digested matrix', i.e., from the fraction remaining after digestion of the 'stripped matrix' with DNAase I. Upon addition of exogenous DNA methyltransferase, however, the DNA of this fraction, which is only 1% (in weight) of the total chromatin DNA and which has a length of approx. 9 kbp, can readily undergo methylation.

Functional role of a highly repetitive DNA sequence in anchorage of the mouse genome

The major portion of the eukaryotic genome consists of various categories of repetitive DNA sequences which have been studied with respect to their base compositions, organizations, copy numbers, transcription and species specificities; their biological roles, however, are still unclear. A novel quality of a highly repetitive mouse DNA sequence is described which points to a functional role: All copies (approximately 50,000 per haploid genome) of this DNA sequence reside on genomic Alu I DNA fragments each associated with nuclear polypeptides that are not released from DNA by proteinase K, SDS and phenol extraction. By this quality the repetitive DNA sequence is classified as a member of the sub-set of DNA sequences involved in tight DNA-polypeptide complexes which have been previously shown to be components of the subnuclear structure termed 'nuclear matrix'. From these results it has to be concluded that the repetitive DNA sequence characterized in this report represents or comprises a signal for a large number of site specific attachment points of the mouse genome in the nuclear matrix.

Chromosomal loop/nuclear matrix organization of transcriptionally active and inactive RNA polymerases in HeLa nuclei

The relative distribution of transcriptionally active and inactive RNA polymerases I and II between the nuclear matrix/scaffold and chromosomal loops of HeLa cells was determined. Total RNA polymerase was assessed by immunoblotting and transcribing RNA polymerase by a photoaffinity labeling technique in isolated nuclei. Nuclear matrix/scaffold was isolated by three methods using high-salt, intermediate-salt or low-salt extraction. The distribution of RNA polymerases I and II were very similar within each of the methods, but considerable differences in distributions were found between the different preparation methods. Either intermediate-salt or high-salt treatment of DNase I-digested nuclei showed significant association of RNA polymerases with the nuclear matrix. However, intermediate-salt followed by high-salt treatment released all transcribing and non-transcribing RNA polymerases. Nuclear scaffolds isolated with lithium diiodosalicylate (low-salt) contained very little of the RNA polymerases. This treatment, however, caused the dissociation of RNA polymerase II transcription complexes. These results show unambiguously that RNA polymerases, both in their active and inactive forms, are not nuclear matrix proteins. The data support models in which the transcriptional machinery moves around DNA loops during transcription.

Topoisomerase I cleavage sites identified and mapped in the chromatin of Dictyostelium ribosomal RNA genes

Sites of an endogenous activity that has the properties of a DNA topoisomerase I have been identified on the palindromic ribosomal RNA genes of the slime mould Dictyostelium discoideum. This was done in vitro, by treating isolated nuclei with sodium dodecyl sulphate, which denatures topoisomerase during its cycle of nicking, strand passing and resealing, and hence reveals the DNA cleavages. It was also done in vivo using the drug camptothecin, which is believed to stabilize the cleavable complex of topoisomerase I plus DNA, hence increasing the chances of cleavage when sodium dodecyl sulphate is subsequently added. The cleavages in vitro and in vivo were mapped by indirect end-labelling. Both treatments cause what appear to be strong double-stranded cleavages at 200 and 2200 base-pairs and at 17 X 10(3) base-pairs upstream from the rRNA transcription start. The cleavage at 200 base-pairs was analysed in greater detail using RNA hybridization probes specific for single DNA strands. The cleavage is in fact composed of three closely spaced nicks on each DNA strand. The DNA sequence at each of the nicks is strongly homologous across 15 base-pairs. Sodium dodecyl sulphate-induced cleavage by eukaryotic topoisomerase I is known to yield enzyme covalently attached to the 3' cut end of the DNA. We show that protein-linked DNA restriction fragments with their 3' ends at the cleavage sites are selectively retarded on denaturing gels, which provides strong evidence that the unusual cluster of cleavages is caused by a topoisomerase I. Additionally, the camptothecin results revealed cleavages not only at the specific upstream sites, but also across the transcribed region. Interestingly, the zone of camptothecin-assisted cleavage does not extend as far at the 3' end of the gene as the zone of endogenous nuclease sensitivity.

Locations and contexts of sequences that hybridize to poly(dG-dT).(dC-dA) in mammalian ribosomal DNAs and two X-linked genes

Sequences located several kilobases both 5' and 3' of the stably transcribed portion of several genes hybridize to radio-labeled pure fragments of the alternating sequence poly (dG-dT) (dC-dA) ["poly(GT)"]. The genes include the ribosomal DNA of mouse, rat, and human, and also human glucose-6-phosphate dehydrogenase (G6PD) and mouse hypoxanthine-guanine phosphoribosyl transferase (HPRT). HPRT has additional hybridizing sequences in introns. Fragments that include the hybridizing sequences and up to 300 bp of adjoining DNA show perfect runs of poly(GT) (greater than 30bp) in all but the human 5' region of rDNA, which shows a somewhat different alternating purine:pyrimidine sequence, poly(GTAT) (36bp). Within 150 bp of these sequences in various instances are found a number of other sequences reported to affect DNA conformation in model systems. Most marked is an enhancement of sequences matching at least 67% to the consensus binding sequence for topoisomerase II. Two to ten-fold less of such sequences were found in other sequenced portions of the nontranscribed spacer or in the transcribed portion of rDNA. The conservation of the locations of tracts of alternating purine:pyrimidine between evolutionarily diverse species is consistent with a possible functional role for these sequences.

Anchorage of the Chinese hamster dihydrofolate reductase gene to the nuclear scaffold occurs in an intragenic region

We have identified a region near the center of the dihydrofolate reductase gene dhfr in Chinese hamster ovary cells that is attached to nuclear scaffolds isolated by extraction with lithium diiodosalicylate. Detailed analysis presented here reveals the presence of only two closely linked sites in 35,000 base-pairs scanned that mediate attachment of the dhfr gene to the nuclear scaffold. Sequence analysis of one of the sites reveals a high A + T content, the presence of cleavage consensus sequences for topoisomerase II, and direct and inverted repeated sequence motifs that are localized to a small region of the attachment site. Attachment of these two regions to the nuclear scaffold is observed in wild-type, hemizygous, and amplified cell lines. Attachment is also retained in dhfr mutants isolated in our laboratory, in which chromosomal lesions have occurred directly adjacent to the scaffold-associated regions. These two regions are not bound to scaffolds prepared from isolated metaphase chromosomes, suggesting that attachment of the dhfr gene is lost during mitosis.

Nuclear matrix association regions of rat alpha 2-macroglobulin gene

We have identified DNA fragments which bind specifically to the nuclear matrix in vitro, termed matrix association regions (MARs), in the first and fourth introns of rat alpha 2-macroglobulin gene. The MAR in the first intron is enriched with sequences closely related to the cleavage consensus of topoisomerase II, and contains the binding site of nuclear factor-alpha, a sequence-specific DNA binding protein reported previously.

Ultrastructural and biochemical comparisons of nuclear matrices prepared by high salt or LIS extraction

We have directly compared two independently published methods for isolating operationally defined nuclear matrices by studying EM ultrastructure, protein composition and distribution of replicating DNA. Nuclear matrices prepared by extraction with 2 M NaCl consisted of fibrous pore complex lamina, residual fibrillar and granular components of nucleoli and interchromatin granules, and an extensive anastomosing internal fibrous network. These matrices were enriched in high molecular weight nonhistone proteins but were virtually devoid of histones. Consistent with previously published data, newly-replicated DNA was resistant to this high salt extraction. Nuclear matrices prepared by extraction of nuclei with 25 mM lithium 3,5-diiodosalicylate, LIS, also contained fibrous pore complex lamina, but lacked morphologically distinct residual nucleoli and were markedly depleted in internal structure. The reduced amounts and complexity of proteins associated with the LIS matrix were consistent with the ultrastructural data. Moreover, much less newly-replicated DNA was recovered in LIS matrices. The data show that LIS dissociates nuclear ultrastructure and extracts both protein and DNA in proportion to the concentration used, regardless of whether nuclei or high salt nuclear matrices are used as starting material. While the data suggest that LIS may not necessarily be an optimal reagent for preparing nuclear matrices containing internal structural elements from all tissue sources, it may be useful for selectively solubilizing and analyzing components of the nuclear matrix.

Sub-set characteristics of DNA sequences involved in tight DNA/polypeptide complexes and their homology to nuclear matrix DNA

Polypeptides co-isolating with DNA induce the binding of a fraction of native DNA fragments to nitrocellulose filters. Southern analysis reveals a high intensity of self-hybridization of the DNA sequences retained on nitrocellulose filters. Consistently, the DNA fraction passing the filters shows only weak hybridization when probed with DNA retained on filters. This indicates that the DNA/polypeptide complexes reside on a non-random sub-set of DNA sequences. Moreover, a high degree of homology was found between residual nuclear matrix DNA sequences and the DNA sequences retained on nitrocellulose filters. This indicates that the DNA sequences associated with tightly bound polypeptides originate from sites where the genome is salt-stably anchored in the nuclear matrix.

Identification of the transcriptional initiation site of ribosomal RNA genes in the crustacean Artemia

The proximal part of the Intergenic Spacer, as well as most of the External Transcribed Spacer of the ribosomal RNA type I genes from the crustacean Artemia have been sequenced. We have identified in the Intergenic Spacer five repeats of around 600 bp in length and, possibly, two imperfect or truncated repeats, derived from the principal ones. These sequences are separated by 485 bp from the 17S rRNA coding sequence. We have also identified the start point of transcription by S1 nuclease analysis. This start point is found 248 bp inside the first repeat. The sequence around the start point shows homology with that described for other members of the same phylum, mostly insects. The most conserved regions are from -1 to +25, and the G residue at position -16. At least the three 600-bp repeats upstream from that containing the promoter also contain the start point sequence, and could therefore act as initiation sites for snPIRNA and/or as enhancer sequences for ribosomal RNA gene transcription.

Disintegration of nucleoskeletal elements by metrizamide/2 M salt isopyknic centrifugation

Supramolecular structures that remain bound to chromosomal DNA under high salt conditions are believed to anchor DNA in the interphase nuclear skeleton. In order to identify these anchorage structures, the non-DNA materials that remain firmly bound to chromosomal DNA under conditions that disintegrate the high salt-stable architecture of nuclei were investigated. Nuclei of Ehrlich ascites cells were histone-depleted by treatment with 2 M salt. The residual halo structures were gently sheared and subjected to metrizamide isopyknic centrifugation in the presence of 2 M salt. By this combined treatment the high salt stable nuclear skeleton becomes disintegrated and three main fractions are resolved. A light fraction comprises the DNA which appears to be essentially depleted of other nuclear components. The only non-DNA material that could be identified in the DNA band is a fraction of (nascent) RNP. No other materials which could reflect nucleoskeletal elements (e.g. lamina proteins) were found together with DNA. A peak of intermediate density comprises RNA/RNP dissociated from DNA. The heavy fraction contains the proteins that become dissociated from DNA by high-salt and/or centrifugal forces, e.g. histones and the major nuclear lamina proteins. The results indicate that nascent RNP is more tightly bound to chromosomal DNA than other components that may be involved in nuclear skeletons. This suggest that transcription complexes represent at least one type of anchorage structure of DNA, which is consistent with results indicating that nascent RNA and actively transcribed DNA sequences are preferentially retained in high-salt-treated nuclei.

Chicken histone genes retain nuclear matrix association throughout the cell cycle

The association between histone genes and the nuclear matrix (NM) during periods of high (S-phase) and low (non-S-phase) transcriptional activity has been investigated with synchronized cells from a chicken erythroid cell line (abbreviated ts34). By DNase I and restriction enzyme analysis, these studies reveal that both core and linker histone genes (represented by H2A and H1 genes respectively) are attached to the NM independent of their transcriptional activity during the cell-cycle. The tissue-specific histone gene H5, expressed constitutively, is nuclear matrix (NM)-associated in ts34 cells but is found in the supernatant (S/N) fractions of a non-erythroid T-cell line. Furthermore, we show that DNA sequences necessary for NM-attachment of the H5 gene lie within a 780 base pair region spanning part of the coding and 5' non-translated region. Of the three non-histone genes investigated, beta-actin sequences are expressed and are NM-attached, feather keratin genes are not expressed and predominate in the S/N, and beta-globin genes although not expressed in the ts34 cell line used were found in the NM fraction. In this case the association may be fortuitous or may reflect an early event prior to transcription of globin genes in differentiating erythroid cells. These results generally support the notion that actively transcribed genes are NM-attached, but that attachment per se is not synonymous with transcription.

Characterization of the region around the start point of transcription of ribosomal RNA in the Chinese hamster

The initiation site for ribosomal RNA transcription in the Chinese hamster was identified and the sequence around and upstream determined. The start point region shows considerable homology with comparable regions in the mouse and rat. In the Chinese hamster, the region between bp -700 and -200 consists of imperfect repeats approximately 120-130 bp in length which are flanked by T-rich regions. The region within each repeat which is homologous with an adjacent repeat decreases in length as the start point is approached. The final promoter-proximal repeat preserves only an 11-bp region of the promoter-distal repeats. This short sequence, termed the repeat core, appears with a periodicity of about 120-130 bp in the Chinese hamster, and is conserved in both mouse and rat. In humans, a short repeated sequence occupies similar positions, suggesting that while complete 120-130-bp repeats are not a feature of all mammalian RNA polymerase I promoter-proximal r X DNA spacers, a short sequence repeating with approximate 120-130-bp periodicity may be such a feature.