The 3D nuclear organization of telomeres marks the transition from Hodgkin to Reed-Sternberg cells

To get an insight into the transition from mononuclear Hodgkin cells (H cells) to diagnostic multinuclear Reed-Sternberg cells (RS cells), we performed an analysis of the three-dimensional (3D) structure of the telomeres in the nuclei of the Hodgkin cell lines HDLM-2, L-428, L-1236 and lymph node biopsies of patients with Hodgkin's disease. Cellular localization of key proteins of the telomere-localized shelterin complex, the mitotic spindle and double-stranded DNA breaks was also analyzed. RS cells show significantly shorter and significantly fewer telomeres in relation to the total nuclear volume when compared with H cells; in particular, telomere-poor 'ghost' nuclei are often adjacent to one or two nuclei displaying huge telomeric aggregates. Shelterin proteins are mainly cytoplasmic in both H and RS cells, whereas double-stranded DNA breaks accumulate in the nuclei of RS cells. In RS cells, multipolar spindles prevent proper chromosome segregation. In conclusion, a process of nuclear disorganization seems to initiate in H cells and further progresses when the cells turn into RS cells and become end-stage tumor cells, unable to divide further because of telomere loss, shortening and aggregate formation, extensive DNA damage and aberrant mitotic spindles that may no longer sustain chromosome segregation. Our findings allow a mechanistic 3D understanding of the transition of H to RS cells.

Inactivation of 14-3-3sigma influences telomere behavior and ionizing radiation-induced chromosomal instability

Telomeres are complexes of repetitive DNA sequences and proteins constituting the ends of linear eukaryotic chromosomes. While these structures are thought to be associated with the nuclear matrix, they appear to be released from this matrix at the time when the cells exit from G(2) and enter M phase. Checkpoints maintain the order and fidelity of the eukaryotic cell cycle, and defects in checkpoints contribute to genetic instability and cancer. The 14-3-3sigma gene has been reported to be a checkpoint control gene, since it promotes G(2) arrest following DNA damage. Here we demonstrate that inactivation of this gene influences genome integrity and cell survival. Analyses of chromosomes at metaphase showed frequent losses of telomeric repeat sequences, enhanced frequencies of chromosome end-to-end associations, and terminal nonreciprocal translocations in 14-3-3sigma(-/-) cells. These phenotypes correlated with a reduction in the amount of G-strand overhangs at the telomeres and an altered nuclear matrix association of telomeres in these cells. Since the p53-mediated G(1) checkpoint is operative in these cells, the chromosomal aberrations observed occurred preferentially in G(2) after irradiation with gamma rays, corroborating the role of the 14-3-3sigma protein in G(2)/M progression. The results also indicate that even in untreated cycling cells, occasional chromosomal breaks or telomere-telomere fusions trigger a G(2) checkpoint arrest followed by repair of these aberrant chromosome structures before entering M phase. Since 14-3-3sigma(-/-) cells are defective in maintaining G(2) arrest, they enter M phase without repair of the aberrant chromosome structures and undergo cell death during mitosis. Thus, our studies provide evidence for the correlation among a dysfunctional G(2)/M checkpoint control, genomic instability, and loss of telomeres in mammalian cells.

A short C-terminal domain of Yku70p is essential for telomere maintenance

The Yku heterodimer from Saccharomyces cerevisiae, comprising Yku70p and Yku80p, is involved in the maintenance of a normal telomeric DNA end structure and is an essential component of nonhomologous end joining (NHEJ). To investigate the role of the Yku70p subunit in these two different pathways, we generated C-terminal deletions of the Yku70 protein and examined their ability to complement the phenotypes of a yku70(-) strain. Deleting only the 30 C-terminal amino acids of Yku70p abolishes Yku DNA binding activity and causes a yku(-) phenotype; telomeres are shortened, and NHEJ is impaired. Using conditions in which at least as much mutant protein as full-length protein is normally detectable in cell extracts, deleting only 25 C-terminal amino acids of Yku70p results in no measurable effect on DNA binding of the Yku protein, and the cells are fully proficient for NHEJ. Nevertheless, these cells display considerably shortened telomeres, and significant amounts of single-stranded overhangs of the telomeric guanosine-rich strands are observed. Co-overexpression of this protein with Yku80p could rescue some but not all of the telomere-related phenotypes. Therefore, the C-terminal domain in Yku70p defines at least one domain that is especially involved in telomere maintenance but not in NHEJ.

The function of DNA polymerase alpha at telomeric G tails is important for telomere homeostasis

Telomere length control is influenced by several factors, including telomerase, the components of telomeric chromatin structure, and the conventional replication machinery. Although known components of the replication machinery can influence telomere length equilibrium, little is known about why mutations in certain replication proteins cause dramatic telomere lengthening. To investigate the cause of telomere elongation in cdc17/pol1 (DNA polymerase alpha) mutants, we examined telomeric chromatin, as measured by its ability to repress transcription on telomere-proximal genes, and telomeric DNA end structures in pol1-17 mutants. pol1-17 mutants with elongated telomeres show a dramatic loss of the repression of telomere-proximal genes, or telomeric silencing. In addition, cdc17/pol1 mutants grown under telomere-elongating conditions exhibit significant increases in single-stranded character in telomeric DNA but not at internal sequences. The single strandedness is manifested as a terminal extension of the G-rich strand (G tails) that can occur independently of telomerase, suggesting that cdc17/pol1 mutants exhibit defects in telomeric lagging-strand synthesis. Interestingly, the loss of telomeric silencing and the increase in the sizes of the G tails at the telomeres temporally coincide and occur before any detectable telomere lengthening is observed. Moreover, the G tails observed in cdc17/pol1 mutants incubated at the semipermissive temperature appear only when the cells pass through S phase and are processed by the time cells reach G(1). These results suggest that lagging-strand synthesis is coordinated with telomerase-mediated telomere maintenance to ensure proper telomere length control.

Identification of ribonucleoprotein (RNP)-specific protein interactions using a yeast RNP interaction trap assay (RITA)

We describe an adaptation of the yeast three-hybrid system that allows the reconstitution in vivo of tripartite (protein-RNA-protein) ribonucleoproteins (RNPs). To build and try this system that we called RNP interaction trap assay (RITA), we used as a model the autoantigenic Ro RNPs. hY RNAs bear distinct binding sites for Ro60 and La proteins, and Ro RNPs are thus physiologically tripartite (Ro60/hY RNA/La). Using recombinant La (rLa) and Ro60 (rRo60) proteins and recombinant hY RNAs (rhY) co-expressed in yeast, we found that RNPs made of rRo60/rhY/rLa were readily reassembled. Reconstitution of tripartite RNPs was critically dependent on the presence of an appropriate Ro60 binding site on the recombinant RNA. The RITA assay was further used to detect (rRo60/rhY RNP)-binding proteins from a HeLa cell cDNA library, allowing specific identification of La and of a novel Ro RNP-binding protein (RoBPI) in more than 70% of positive clones. RITA assay may complement already available two- and three-hybrid systems to characterize RNP-binding proteins by allowing the in vivo identification of interactions strictly dependent upon the simultaneous presence of a protein and of its cognate RNA.

Accumulation of single-stranded DNA and destabilization of telomeric repeats in yeast mutant strains carrying a deletion of RAD27

The Saccharomyces cerevisiae RAD27 gene encodes the yeast homologue of the mammalian FEN-1 nuclease, a protein that is thought to be involved in the processing of Okazaki fragments during DNA lagging-strand synthesis. One of the predicted DNA lesions occurring in rad27 strains is the presence of single-stranded DNA of the template strand for lagging-strand synthesis. We examined this prediction by analyzing the terminal DNA structures generated during telomere replication in rad27 strains. The lengths of the telomeric repeat tracts were found to be destabilized in rad27 strains, indicating that naturally occurring direct repeats are subject to tract expansions and contractions in such strains. Furthermore, abnormally high levels of single-stranded DNA of the templating strand for lagging-strand synthesis were observed in rad27 cells. Overexpression of Dna2p in wild-type cells also yielded single-stranded DNA regions on telomeric DNA and caused a cell growth arrest phenotype virtually identical to that seen for rad27 cells grown at the restrictive temperature. Furthermore, overexpression of the yeast exonuclease Exo1p alleviated the growth arrest induced by both conditions, overexpression of Dna2p and incubation of rad27 cells at 37 degrees C. However, the telomere heterogeneity and the appearance of single-stranded DNA are not prevented by the overexpression of Exo1p in these strains, suggesting that this nuclease is not simply redundant with Rad27p. Our data thus provide in vivo evidence for the types of DNA lesions predicted to occur when lagging-strand synthesis is deficient and suggest that Dna2p and Rad27p collaborate in the processing of Okazaki fragments.

Cytological and functional aspects of telomere maintenance

The fact that eukaryotic chromosomes are linear poses a special problem for their maintenance: the natural ends of chromosomes must be distinguished from ends generated by chromosomal breakage and somehow, the chromosome ends must also be fully replicated to maintain their integrity. Telomeres, the complex structures at the ends of chromosomes are thought to be instrumental for both of these functions. However, recent insights in telomere biology suggest that these terminal structures do much more than just fulfill these two basic functions. Cytological data demonstrate that telomeres may play leading roles in chromatin organization and nuclear architecture during mitosis and meiosis. Moreover, non-functional telomeres may lead to genetic instability, a common prelude to cancer. Here, we review the basic functions of telomeres during chromosome replication and discuss the cytological aspects of telomere function during mitosis and meiosis.

Processing of telomeric DNA ends requires the passage of a replication fork

During telomere replication in yeast, chromosome ends acquire a long single-stranded extension of the strand making the 3' end. Previous work showed that these 3' tails are generated late in S-phase, when conventional replication is virtually complete. In addition, the extensions were also observed in cells that lacked telomerase. Therefore, a model was proposed that predicted an activity that recessed the 5' ends at yeast telomeres after conventional replication was complete. Here, we demonstrate that this processing activity is dependent on the passage of a replication fork through yeast telomeres. A non-replicating linear plasmid with telomeres at each end does not acquire single-stranded extensions, while an identical construct containing an origin of replication does. Thus, the processing activity could be associated with the enzymes at the replication fork itself, or the passage of the fork through the telomeric sequences allows a transient access for the activity to the telomeres. We therefore propose that there is a mechanistic link between the conventional replication machinery and telomere maintenance.

Telomere elongation by hnRNP A1 and a derivative that interacts with telomeric repeats and telomerase

Telomeric DNA of mammalian chromosomes consists of several kilobase-pairs of tandemly repeated sequences with a terminal 3' overhang in single-stranded form. Maintaining the integrity of these repeats is essential for cell survival; telomere attrition is associated with chromosome instability and cell senescence, whereas stabilization of telomere length correlates with the immortalization of somatic cells. Telomere elongation is carried out by telomerase, an RNA-dependent DNA polymerase which adds single-stranded TAGGGT repeats to the 3' ends of chromosomes. While proteins that associate with single-stranded telomeric repeats can influence tract lengths in yeast, equivalent factors have not yet been identified in vertebrates. Here, it is shown that the heterogeneous nuclear ribonucleoprotein A1 participates in telomere biogenesis. A mouse cell line deficient in A1 expression harbours telomeres that are shorter than those of a related cell line expressing normal levels of A1. Restoring A1 expression in A1-deficient cells increases telomere length. Telomere elongation is also observed upon introduction of exogenous UP1, the amino-terminal fragment of A1. While both A1 and UP1 bind to vertebrate single-stranded telomeric repeats directly and with specificity in vitro, only UP1 can recover telomerase activity from a cell lysate. These findings establish A1/UP1 as the first single-stranded DNA binding protein involved in mammalian telomere biogenesis and suggest possible mechanisms by which UP1 may modulate telomere length.

Yeast Ku as a regulator of chromosomal DNA end structure

During telomere replication in yeast, chromosome ends acquire an S-phase-specific overhang of the guanosine-rich strand. Here it is shown that in cells lacking Ku, a heterodimeric protein involved in nonhomologous DNA end joining, these overhangs are present throughout the cell cycle. In vivo cross-linking experiments demonstrated that Ku is bound to telomeric DNA. These results show that Ku plays a direct role in establishing a normal DNA end structure on yeast chromosomes, conceivably by functioning as a terminus-binding factor. Because Ku-mediated DNA end joining involving telomeres would result in chromosome instability, our data also suggest that Ku has a distinct function when bound to telomeres.

The terminal DNA structure of mammalian chromosomes

In virtually all eukaryotic organisms, telomeric DNA is composed of a variable number of short direct repeats. While the primary sequence of telomeric repeats has been determined for a great variety of species, the actual physical DNA structure at the ends of a bona fide metazoan chromosome with a centromere is unknown. It is shown here that an overhang of the strand forming the 3' ends of the chromosomes, the G-rich strand, is found at mammalian chromosome ends. Moreover, on at least some telomeres, the overhangs are > or = 45 bases long. Such surprisingly long overhangs were present on chromosomes derived from fully transformed tissue culture cells and normal G0-arrested peripheral leukocytes. Thus, irrespective of whether the cells were actively dividing or arrested, a very similar terminal DNA arrangement was found. These data suggest that the ends of mammalian and possibly all vertebrate chromosomes consist of an overhang of the G-rich strand and that these overhangs may be considerably larger than previously anticipated.

The DNA structures at the ends of eukaryotic chromosomes

The sequence organisation of the telomeric regions is extremely similar for all eukaryotes examined to date. Subtelomeric areas may contain large sequence arrays of middle repetitive, complex elements that sometimes have similarities to retrotransposons. In between and within these complex sequences are short, satellite-like repeats. These areas contain very few genes and are thought to be organised into a heterochromatin-like domain. The terminal regions almost invariably consist of short, direct repeats. These repeats usually contain clusters of 2-4 G residues and the strand that contains these clusters (the G strand) always forms the extreme 3'-end of the chromosome. Thus, most telomeric repeats are clearly related to each other which in turn suggests a common evolutionary origin. A number of different structures can be formed by single-stranded telomeric G strand repeats and, as has been suggested recently, by the G strand. Since the main mechanism for the maintenance of telomeric repeats predicts the occurrence of single-stranded extensions of the G strand, the propensity of G-rich DNA to fold into alternative DNA structures may have implications for telomere biology.

Cell cycle-regulated generation of single-stranded G-rich DNA in the absence of telomerase

Current models of telomere replication predict that due to the properties of the polymerases implicated in semiconservative replication of linear DNA, the two daughter molecules have one end that is blunt and one end with a short 3' overhang. Telomerase is thought to extend the short 3' overhang to produce long single-stranded overhangs. Recently, such overhangs, or TG1-3 tails, were shown to occur on both telomeres of replicated linear plasmids in yeast. Moreover, indirect evidence suggested that the TG1-3 tails also occurred in a yeast strain lacking telomerase. We report herein a novel in-gel hybridization technique to probe telomeres for single-stranded DNA. Using this method, it is shown directly that in yeast strains lacking the TLC1 gene encoding the yeast telomerase RNA, TG1-3 single-stranded DNA was generated on chromosomal and plasmid telomeres. The single-stranded DNA only appeared in S phase and was sensitive to digestion with a single-strand-specific exonuclease. These data demonstrate that during replication of telomeres, TG1-3 tails can be generated in a way that is independent of telomerase-mediated strand elongation. In wild-type strains, these TG1-3 tails could subsequently serve as substrates for telomerase and telomere binding proteins on all telomeres.

Mutations in the homologous ZDS1 and ZDS2 genes affect cell cycle progression

The Saccharomyces cerevisiae ZDS1 and ZDS2 genes were identified as multicopy suppressors in distinct genetic screens but were found to encode highly similar proteins. We show that at semipermissive temperatures, a yeast strain with a cdc28-1N allele was uniquely deficient in plasmid maintenance in comparison with strains harboring other cdc28 thermolabile alleles. Quantitative analysis of plasmid loss rates in cdc28-1N strains carrying plasmids with multiple replication origins suggests that a defect in initiating DNA replication probably causes this plasmid loss phenotype. The ZDS1 gene was isolated as a multicopy suppressor of the cdc28-1N plasmid loss defect. A zds1 deletion exhibits genetic interactions with cdc28-1N but not with other cdc28 alleles. SIN4 encodes a protein which is part of the RNA polymerase II holoenzyme-mediator complex, and a sin4 null mutation has pleiotropic effects suggesting roles in transcriptional regulation and chromatin structure. The ZDS2 gene was isolated as a multicopy suppressor of the temperature-sensitive growth defect caused by the sin4 null mutation. Disruption of either ZDS1 or ZDS2 causes only modest phenotypes. However, a strain with both ZDS1 and ZDS2 disrupted is extremely slowly growing, has marked defects in bud morphology, and shows defects in completing S phase or entering mitosis.

Evidence for a new step in telomere maintenance

The strand of telomeric DNA that runs 5'-3' toward a chromosome end is typically G rich. Telomerase-generated G tails are expected at one end of individual DNA molecules. Saccharomyces telomeres acquire TG1-3 tails late in S phase. Moreover, the telomeres of linear plasmids can interact when the TG1-3 tails are present. Molecules that mimic the structures predicted for telomere replication intermediates were generated in vitro. These in vitro generated molecules formed telomere-telomere interactions similar to those on molecules isolated from yeast, but only if both ends that interacted had a TG1-3 tail. Moreover, TG1-3 tails were generated in vivo in cells lacking telomerase. These data suggest a new step in telomere maintenance, cell cycle-regulated degradation of the C1-3A strand, which can generate a potential substrate for telomerase and telomere-binding proteins at every telomere.

Origin activation and formation of single-strand TG1-3 tails occur sequentially in late S phase on a yeast linear plasmid

In order to understand the mechanisms leading to the complete duplication of linear eukaryotic chromosomes, the temporal order of the events involved in replication of a 7.5-kb Saccharomyces cerevisiae linear plasmid called YLpFAT10 was determined. Two-dimensional agarose gel electrophoresis was used to map the position of the replication origin and the direction of replication fork movement through the plasmid. Replication began near the center of YLpFAT10 at the site in the 2 microns sequences that corresponds to the 2 microns origin of DNA replication. Replication forks proceeded bidirectionally from the origin to the ends of YLpFAT10. Thus, yeast telomeres do not themselves act as origins of DNA replication. The time of origin utilization on YLpFAT10 and on circular 2 microns DNA in the same cells was determined both by two-dimensional gel electrophoresis and by density transfer experiments. As expected, 2 microns DNA replicated in early S phase. However, replication of YLpFAT10 occurred in late S phase. Thus, the time of activation of the 2 microns origin depended upon its physical context. Density transfer experiments established that the acquisition of telomeric TG1-3 single-strand tails, a predicted intermediate in telomere replication, occurred immediately after the replication forks approached the ends of YLpFAT10. Thus, telomere replication may be the very last step in S phase.

Saccharomyces telomeres acquire single-strand TG1-3 tails late in S phase

Saccharomyces telomeres consist of approximately 300 bp of C1-3A/TG1-3 DNA. Nondenaturing Southern hybridization, capable of detecting approximately 60 to approximately 300 bases of TG1-3 DNA, revealed that yeast telomeres acquired and lost TG1-3 tails, the predicted intermediate in telomere replication, in a cell cycle-dependent manner. TG1-3 tails were also detected on the ends of a linear plasmid isolated from late S phase cells. In addition, a nonlinear form of this plasmid was detected: this structure migrated in two-dimensional agarose gels like a nicked circle of the same size as the linear plasmid, but had considerably more single-stranded character than a conventional nicked circle. The evidence indicates that these circles were formed by telomere-telomere interactions involving the TG1-3 tails. These data provide evidence for a cell cycle-dependent change in telomere structure and demonstrate that TG1-3 tails, generated during replication of a linear plasmid in vivo, are capable of mediating telomere-telomere interactions.

Use of non-denaturing Southern hybridization and two dimensional agarose gels to detect putative intermediates in telomere replication in Saccharomyces cerevisiae

Telomeres are required for the complete duplication of the ends of linear chromosomes. Saccharomyces telomeres bear approximately 350 bps of C1-3A/TG1-3 sequences. Previous work using non-denaturing Southern blotting has demonstrated the cell cycle controlled appearance of single stranded TG1-3 tails on chromosomal and plasmid telomeres (Wellinger et al. submitted). Furthermore it was shown that short linear plasmids carrying an origin of replication derived from 2 microns DNA can circularize at the time of telomere replication (Wellinger et al. submitted). Here we demonstrate that those loci previously shown to acquire single stranded tails are indeed telomeres and that single stranded TG1-3 cannot be observed in non-telomeric C1-3A/TG1-3-tracts. Moreover, we demonstrate that the formation of circular DNA by short linear plasmids is not restricted to plasmids containing a 2 microns origin of replication but can also be detected for plasmids containing ARS1.

Effects of excess centromeres and excess telomeres on chromosome loss rates

The linear chromosomes of eukaryotes contain specialized structures to ensure their faithful replication and segregation to daughter cells. Two of these structures, centromeres and telomeres, are limited, respectively, to one and two copies per chromosome. It is possible that the proteins that interact with centromere and telomere DNA sequences are present in limiting amounts and could be competed away from the chromosomal copies of these elements by additional copies introduced on plasmids. We have introduced excess centromeres and telomeres into Saccharomyces cerevisiae and quantitated their effects on the rates of loss of chromosome III and chromosome VII by fluctuation analysis. We show that (i) 600 new telomeres have no effect on chromosome loss; (ii) an average of 25 extra centromere DNA sequences increase the rate of chromosome III loss from 0.4 x 10(-4) events per cell division to 1.3 x 10(-3) events per cell division; (iii) centromere DNA (CEN) sequences on circular vectors destabilize chromosomes more effectively than do CEN sequences on 15-kb linear vectors, and transcribed CEN sequences have no effect on chromosome stability. We discuss the different effects of extra centromere and telomere DNA sequences on chromosome stability in terms of how the cell recognizes these two chromosomal structures.

Lack of positional requirements for autonomously replicating sequence elements on artificial yeast chromosomes

In the yeast Saccharomyces cerevisiae, origins of replication (autonomously replicating sequences; ARSs), centromeres, and telomeres have been isolated and characterized. The identification of these structures allows the construction of artificial chromosomes in which the architecture of eukaryotic chromosomes may be studied. A common feature of most, and possibly all, natural yeast chromosomes is that they have an ARS within 2 kilobases of their physical ends. To study the effects of such telomeric ARSs on chromosome maintenance, we introduced artificial chromosomes of approximately 15 and 60 kilobases into yeast cells and analyzed the requirements for telomeric ARSs and the effects of ARS-free chromosomal arms on the stability of these molecules. We find that terminal blocks of telomeric repeats are sufficient to be recognized as telomeres. Moreover, artificial chromosomes containing telomere-associated Y' sequences and telomeric ARSs were no more stable during both mitosis and meiosis than artificial chromosomes lacking terminal ARSs, indicating that yeast-specific blocks of telomeric sequences are the only cis-acting requirement for a functional telomere during both mitotic growth and meiosis. The results also show that there is no requirement for an origin of replication on each arm of the artificial chromosomes, indicating that a replication fork may efficiently move through a functional centromere region.

Exogenous mouse mammary tumor virus proviral DNA isolated from a kidney adenocarcinoma cell line contains alterations in the U3 region of the long terminal repeat

Mouse mammary tumor virus (MMTV) is a B-type retrovirus which induces predominantly mammary carcinomas after a relatively long latency period. To date, very little is known about the reasons for the strict tissue specificity of MMTV. The BALB/cf/Cd strain of mice, which was infected with milk-borne MMTV (C3H), shows a high incidence of kidney adenocarcinomas, and our data suggest that MMTV might be involved in the formation of these tumors. Newly integrated exogenous MMTV proviruses were found in the genome of transplanted tumor cells as well as in the DNA of a cell line derived from one tumor, but not in normal cells of BALB/cf/Cd mice. The MMTV DNA in these tumor cells was transcribed and viral RNA synthesis was strongly stimulated by glucocorticoid hormones. Viral structural polypeptides, comparable in size and antigenicity to MMTV polypeptides of infected mammary tumor cells were synthesized and processed normally in the cell line and were organized correctly into intracytoplasmic particles. Heteroduplex analysis of the molecularly cloned MMTV proviral DNAs of kidney and mammary tumor origin revealed a high degree of homology in the gag, pol, and env genes. A striking difference, however, was observed in the U3 region of the two LTRs that might relate to the different tissue specificity of the two viruses.