Positive supercoiling of mitotic DNA drives decatenation by topoisomerase II in eukaryotes.

DNA topoisomerase II completely removes DNA intertwining, or catenation, between sister chromatids before they are segregated during cell division. How this occurs throughout the genome is poorly understood. We demonstrate that in yeast, centromeric plasmids undergo a dramatic change in their topology as the cells pass through mitosis. This change is characterized by positive supercoiling of the DNA and requires mitotic spindles and the condensin factor Smc2. When mitotic positive supercoiling occurs on decatenated DNA, it is rapidly relaxed by topoisomerase II. However, when positive supercoiling takes place in catenated plasmid, topoisomerase II activity is directed toward decatenation of the molecules before relaxation. Thus, a topological change on DNA drives topoisomerase II to decatenate molecules during mitosis, potentially driving the full decatenation of the genome.

Plasmid DNA replication and topology as visualized by two-dimensional agarose gel electrophoresis.

During the last 20 years, two-dimensional agarose gel electrophoresis combined with other techniques such as Polymerase Chain Reaction, helicase assay and electron microscopy, helped to characterize plasmid DNA replication and topology. Here we describe some of the most important findings that were made using this method including the characterization of uni-directional replication, replication origin interference, DNA breakage at the forks, replication fork blockage, replication knotting, replication fork reversal, the interplay of supercoiling and catenation and other changes in DNA topology that take place as replication progresses.

A redundancy of processes that cause replication fork stalling enhances recombination at two distinct sites in yeast rDNA.

DNA recombination was investigated by monitoring integration at the rDNA of a circular minichromosome containing a 35S minigene and a replication fork barrier (RFB). The effects of replication fork stalling on integration were studied in wild-type, FOB1Delta, SIR2Delta and the double mutant FOB1DeltaSIR2Delta cells. The results obtained confirmed that Sir2p represses and replication fork stalling enhances integration of the minichromosome. This integration, however, only took place at two distinct sites: the RFB and the 3' end of the 35S gene. For integration to take place at the 35S gene, replication fork stalling must occur at the 3' end of the gene in both the minichromosome and the chromosomal repeats. Integration at the RFB, on the other hand, occurred readily in FOB1Delta cells, indicating that more than a single mechanism triggers homologous recombination at this site. Altogether, these observations strongly suggest that the main role for replication fork stalling at the rDNA locus is to promote homologous recombination rather than just to prevent head-on collision of transcription and replication as originally thought.

Differential expression of Ran GTPase during HMBA-induced differentiation in murine erythroleukemia cells.

Murine erythroleukemia (MEL) cells undergo erythroid differentiation in vitro when treated with hexamethylene bisacetamide (HMBA). To identify genes involved in the commitment of MEL cells to differentiate, we screened a cDNA library constructed from HMBA-induced cells by differential hybridization and isolated GTPase Ran as a down-regulated gene. We observed that Ran was expressed in a biphasic mode. Following a decrease in mRNA level during the initial hours of induction, Ran re-expressed at 24-48 h, and gradually declined again. To investigate the role of Ran during MEL differentiation we constructed MEL transfectants capable to express or block Ran mRNA production constitutively. No effects were observed on cell growth and proliferation. Blockage of Ran, however, interfered with MEL cell differentiation resulting in a decrease of cell survival in the committed population.

Knotting dynamics during DNA replication.

The topology of plasmid DNA changes continuously as replication progresses. But the dynamics of the process remains to be fully understood. Knotted bubbles form when topo IV knots the daughter duplexes behind the fork in response to their degree of intertwining. Here, we show that knotted bubbles can form during unimpaired DNA replication, but they become more evident in partially replicated intermediates containing a stalled fork. To learn more about the dynamics of knot formation as replication advances, we used two-dimensional agarose gel electrophoresis to identify knotted bubbles in partially replicated molecules in which the replication fork stalled at different stages of the process. The number and complexity of knotted bubbles rose as a function of bubble size, suggesting that knotting is affected by both precatenane density and bubble size.

DNA knotting caused by head-on collision of transcription and replication.

Collision of transcription and replication is uncommon, but the reason for nature to avoid this type of collision is still poorly understood. In Escherichia coli pBR322 is unstable and rapidly lost without selective pressure. Stability can be rescued if transcription of the tetracycline-resistance gene (Tet(R)), progressing against replication, is avoided. We investigated the topological consequences of the collision of transcription and replication in pBR322-derived plasmids where head-on collision between the replication fork and the RNA polymerase transcribing the Tet(R) gene was allowed or avoided. The results obtained indicate that this type of collision triggers knotting of the daughter duplexes behind the fork. We propose this deleterious topological consequence could explain the instability of pBR322 and could be also one of the reasons for nature to avoid head-on collision of transcription and replication.

Supercoiling, knotting and replication fork reversal in partially replicated plasmids.

To study the structure of partially replicated plasmids, we cloned the Escherichia coli polar replication terminator TerE in its active orientation at different locations in the ColE1 vector pBR18. The resulting plasmids, pBR18-TerE@StyI and pBR18-TerE@EcoRI, were analyzed by neutral/neutral two-dimensional agarose gel electrophoresis and electron microscopy. Replication forks stop at the Ter-TUS complex, leading to the accumulation of specific replication intermediates with a mass 1.26 times the mass of non-replicating plasmids for pBR18-TerE@StyI and 1.57 times for pBR18-TerE@EcoRI. The number of knotted bubbles detected after digestion with ScaI and the number and electrophoretic mobility of undigested partially replicated topoisomers reflect the changes in plasmid topology that occur in DNA molecules replicated to different extents. Exposure to increasing concentrations of chloroquine or ethidium bromide revealed that partially replicated topoisomers (CCCRIs) do not sustain positive supercoiling as efficiently as their non-replicating counterparts. It was suggested that this occurs because in partially replicated plasmids a positive DeltaLk is absorbed by regression of the replication fork. Indeed, we showed by electron microscopy that, at least in the presence of chloroquine, some of the CCCRIs of pBR18-Ter@StyI formed Holliday-like junction structures characteristic of reversed forks. However, not all the positive supercoiling was absorbed by fork reversal in the presence of high concentrations of ethidium bromide.

Premature termination of DNA replication in plasmids carrying two inversely oriented ColE1 origins.

In Escherichia coli plasmids carrying two inversely oriented ColE1 origins, DNA replication initiates at only one of the two potential origins. The other silent origin acts as a replication fork barrier. Whether this barrier is permanent or simply a pausing site remains unknown. Here, we used a repeated primer extension assay to map in vivo, at the nucleotide level, the 5' end of the nascent strand where initiation and blockage of replication forks occurs. Initiation occurred primarily at the previously defined origin, however, an alternative initiation site was detected 17 bp upstream. At the barrier, the lagging strand also terminated at the main initiation site. Therefore, the 5' end of the nascent strand at the barrier was identical to that generated during initiation. This observation strongly suggests that blockage of the replication fork at the silent origin is not just a pausing site but permanent, and leads to a premature termination event.

Bi-directional replication and random termination.

Two-dimensional (2D) agarose gel electrophoresis was used to study termination of DNA replication in a shuttle vector, YRp7', when it replicated in Escherichia coli, Saccharomyces cerevisiae and Xenopus egg extracts. In E. coli, the 2D gel patterns obtained were consistent with uni-directional replication initiated at a specific site, the ColE1 origin. In consequence, termination also occurred precisely at the ColE1 origin. In Xenopus egg extracts, the particular shape of the bubble arc as well as the triangular smear detected to the left of the simple-Y pattern indicated random initiation and termination. In S.cerevisiae, initiation occurred at the ARS1 origin and replication proceeded in a bi-directional manner. However, termination did not always occur at a specific site 180 degrees across from the origin, but almost all along the south hemisphere of the plasmid. Inversion, deletion or replacement of DNA sequences located throughout this hemisphere did not eliminate random termination. Analysis of the replication intermediates of another yeast plasmid bearing a different origin, ARS305, also exhibited random termination. We propose that the random termination events observed in S.cerevisiae could be due to an asynchronous departure of both forks from the bi-directional origin in addition to differences in the rate of fork progression. These observations could be extended to all bi-directional origins.

Visualisation of plasmid replication intermediates containing reversed forks.

Blockage of replication forks can have deleterious consequences for the cell as it may prompt premature termination of DNA replication. Moreover, the blocked replication intermediate (RI) could be particularly sensitive to recombination processes. We analysed the different populations of RIs generated in vivo in the bacterial plasmid pPI21 after pausing of replication forks at the inversely oriented ColE1 origin. To achieve this goal, a new method was developed based on two-dimensional agarose gel electrophoresis. This method allows the isolation of specific RIs, even when they were rather scarce, from the total DNA. Here we describe the occurrence of RI restriction fragments containing reversed forks. These Holliday-like structures have been postulated but never observed before.

Characterization of the pea rDNA replication fork barrier: putative cis-acting and trans-acting factors.

It was previously shown that in pea (Pisum sativum), rDNA repeats contain a polar replication fork barrier that blocks progression of the replication machinery moving in the direction opposite to transcription. This barrier maps in the untranscribed spacer close to the 3' end of the 25S gene. Very similar barriers are also found in the rDNA of yeast, Xenopus and mammalian cultured cells. This high conservation indicates that the rDNA barrier plays a relevant biological role. Progression of replication forks through the DNA sequence where the barrier maps in pea was investigated in plasmids replicating in Escherichia coli and Saccharomyces cerevisiae. No barrier was detected in these heterologous systems, indicating that the DNA sequence by itself was insufficient to block the replication machinery. Therefore, trans-acting factors were likely to be required. Taking advantage of the natural sequence heterogeneity in pea rDNA, we obtained evidence that a 27 bp imperfect tandem repeat is involved in the arrest of replication. Moreover, nuclear protein(s) specifically bound to this repeat suggesting that this DNA/protein complex is responsible for the polar arrest of replication forks.

Formation of knots in partially replicated DNA molecules.

Bacterial plasmids with two origins of replication in convergent orientation are frequently knotted in vivo. The knots formed are localised within the newly replicated DNA regions. Here, we analyse DNA knots tied within replication bubbles of such plasmids, and observe that the knots formed show predominantly positive signs of crossings. We propose that helical winding of replication bubbles in vivo leads to topoisomerase-mediated formation of knots on partially replicated DNA molecules.

DnaB helicase is unable to dissociate RNA-DNA hybrids. Its implication in the polar pausing of replication forks at ColE1 origins.

A series of plasmids were constructed containing two unidirectional ColE1 replication origins in either the same or opposite orientations and their replication mode was investigated using two-dimensional agarose gel electrophoresis. The results obtained showed that, in these plasmids, initiation of DNA replication occurred at only one of the two potential origins per replication round regardless of origins orientation. In those plasmids with inversely oriented origins, the silent origin act as a polar pausing site for the replication fork initiated at the other origin. The distance between origins (up to 5.8 kilobase pairs) affected neither the interference between them to initiate replication nor the pausing function of the silent origin. A deletion analysis indicated that the presence of a transcription promoter upstream of the origin was the only essential requirement for it to initiate replication as well as to account for its polar pausing function. Finally, in vitro helicase assays showed that Escherichia coli DnaB is able to melt DNA-DNA homoduplexes but is very inefficient to unwind RNA-DNA hybrids. Altogether, these observations strongly suggest that replication forks pause at silent ColE1 origins due to the inability of DnaB helicase, which leads the replication fork in vivo, to unwind RNA-DNA hybrids.

A computer model for the analysis of DNA replication intermediates by two-dimensional agarose gel electrophoresis.

We present a computer model to predict the patterns expected for the replication intermediates (RIs) of DNA fragments analyzed by neutral/neutral two-dimensional (2D) agarose gel electrophoresis. The model relies on the mode of replication (uni- or bi-directional), the electrophoretic mobility of linear DNA fragments and the retardation caused by the three-dimensional shape of non-linear molecules. The utility of this model is demonstrated with two examples: replication analysis of the plasmids pBR322 and pHH5.8 in Escherichia coli after digestions with EcoRI and HindIII, respectively.

Topological complexity of different populations of pBR322 as visualized by two-dimensional agarose gel electrophoresis.

Neutral/neutral two-dimensional (2D) agarose gelelectrophoresis was used to investigate populations of the different topological conformations that pBR322 can adopt in vivo in bacterial cells as well as in Xenopus egg extracts. To help in interpretation and identification of all the different signals, undigested as well as DNA samples pretreated with DNase I, topoisomerase I and topoisomerase II were analyzed. The second dimension of the 2D gel system was run with or without ethidium bromide to account for any possible changes in the migration behavior of DNA molecules caused by intercalation of this planar agent. Finally, DNA samples were isolated from a recA-strain of Escherichia coli , as well as after direct labeling of the replication intermediates in extracts of Xenopus laevis eggs. Altogether, the results obtained demonstrated that 2D gels can be readily used to identify most of the complex topological populations that circular molecules can adopt in vivo in both bacteria and eukaryotic cells.

Co-localization of polar replication fork barriers and rRNA transcription terminators in mouse rDNA.

We investigated the replication of the region where transcription terminates in mouse rDNA. It contains a replication fork barrier (RFB) that behaves in a polar manner, arresting only replication forks moving in the direction opposite to transcription. This RFB consists of several closely spaced fork arrest sites that co-localize with the transcription terminator elements, known as Sal boxes. Sal boxes are the target for mTTF-I (murine transcription termination factor I). These results suggest that both termination of rRNA transcription and replication fork arrest may share cis-acting as well as trans-acting factors.

Cloning, sequencing and expression in MEL cells of a cDNA encoding the mouse ribosomal protein S5.

We describe the isolation and characterization of a cDNA encoding the mouse S5 ribosomal protein. It was isolated from a MEL (murine erythroleukemia) cell cDNA library by differential hybridization as a down regulated sequence during HMBA-induced differentiation. Northern series analysis showed that S5 mRNA expression is reduced 5-fold throughout the differentiation process. The mouse S5 mRNA is 760 bp long and encodes for a 204 amino acid protein with 94% homology with the human and rat S5.

The ColE1 unidirectional origin acts as a polar replication fork pausing site.

Co-orientation of replication origins is the most common organization found in nature for multimeric plasmids. Streptococcus pyogenes broad-host-range plasmid pSM19035 and Escherichia coli pPI21 are among the exceptions. pPI21, which is a derivative of pSM19035 and pBR322, has two long inverted repeats, each one containing a potentially active ColE1 unidirectional origin. Analysis of pPI21 replication intermediates (RIs) by two-dimensional agarose gel electrophoresis and electron microscopy revealed the accumulation of a specific RI containing a single internal bubble. The data obtained demonstrated that initiation of DNA replication occurred at a single origin in pPI21. Progression of the replicating fork initiated at either of the two potential origins was transiently stalled at the other inversely oriented silent ColE1 origin of the plasmid. The accumulated RIs, containing an internal bubble, occurred as a series of stereoisomers with different numbers of knots in their replicated portion. These observations provide one of the first functional explanations for the disadvantage of head-to-head plasmid multimers with respect to head-to-tail ones.

The migration behaviour of DNA replicative intermediates containing an internal bubble analyzed by two-dimensional agarose gel electrophoresis.

Initiation of DNA replication in higher eukaryotes is still a matter of controversy. Some evidence suggests it occurs at specific sites. Data obtained using two-dimensional (2D) agarose gel electrophoresis, however, led to the notion that it may occur at random in broad zones. This hypothesis is primarily based on the observation that several contiguous DNA fragments generate a mixture of the so-called 'bubble' and 'simple Y' patterns in Neutral/neutral 2D gels. The interpretation that this mixture of hybridisation patterns is indicative for random initiation of DNA synthesis relies on the assumption that replicative intermediates (RIs) containing an internal bubble where initiation occurred at different relative positions, generate comigrating signals. The latter, however, is still to be proven. We investigated this problem by analysing together, in the same 2D gel, populations of pBR322 RIs that were digested with different restriction endonucleases that cut the monomer only once at different locations. DNA synthesis begins at a specific site in pBR322 and progresses in a uni-directional manner. Thus, the main difference between these sets of RIs was the relative position of the origin. The results obtained clearly showed that populations of RIs containing an internal bubble where initiation occurred at different relative positions do not generate signals that co-migrate all-the-way in 2D gels. Despite this observation, however, our results support the notion that random initiation is indeed responsible for the peculiar 'bubble' signal observed in the case of several metazoan eukaryotes.

Conserved features in the mode of replication of eukaryotic ribosomal RNA genes.

It was previously shown that a 1.5 kb fragment located in the non-transcribed spacer (NTS) is the earliest replicating region of pea (Pisum sativum) rDNA in synchronized root cells. In the present report the structure of this region was characterized. It contains a cluster of four 11 bp near matches to the Saccharomyces cerevisiae ARS consensus sequence (ACS). These near matches are embedded in an A+T rich domain located upstream from the transcription initiation site. We identified and mapped an intrinsic DNA bending locus 5' to the cluster of near matches. Several eukaryotic origins including the ARS from the budding yeast show very similar structural features. This observation strengthens the notion that pea rDNA replication initiates at or near this region. Replication of the entire pea rDNA repeat was analysed by two-dimensional (2D) agarose gel electrophoresis. The results obtained indicate that only a small fraction of the potential origins is used in each replication round. Forks moving in the direction opposite to rRNA transcription are stalled at a polar replication fork barrier (RFB), which mapped near the 3' end of the transcription unit. Consequently, most of pea rDNA appears to replicate in a unidirectional manner. These results show that the strategy used to replicate pea and yeast rRNA genes is very similar, suggesting that it has been conserved and might be common to most eukaryotes.