Skip Navigation

This Article
Right arrow Abstract Freely available
Right arrow Print PDF (213K) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (4)
Right arrowRequest Permissions
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Benard, M.
Right arrow Articles by Pierron, G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Benard, M.
Right arrow Articles by Pierron, G.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Nucleic Acids Research Pages 2091-2098  

Early activated replication origins within the cell cycle-regulated histone H4 genes in Physarum
Introduction
Materials And Methods
   Strains and cultures
   DNA isolation and agarose gel electrophoreses
   Southern hybridisation and probes
   DNA sequence analysis
Results
   The H4-1 gene is linked to a replication origin
   Kinetics of DNA replication at the H4-1 locus
   Stalling of the rightward-moving replication fork downstream of the H4-1 gene
   2-Dimensional gel mapping of the H4-1 replication origin
   Sequence analysis
   2-Dimensional gel mapping of the H4-2 replication origin
Discussion
Acknowledgements
References

Early activated replication origins within the cell cycle-regulated histone H4 genes in Physarum

Early activated replication origins within the cell cycle-regulated histone H4 genes in Physarum

Marianne Bénard* and Gérard Pierron

Laboratoire Organisation Fonctionnelle du Noyau, CNRS UPR-1983, IFR-1221, F-94801 Villejuif, France

Received February 11, 1999; Revised and Accepted March 24, 1999

ABSTRACT

It was previously shown that the two members of the cell cycle-regulated histone H4 gene family, H4-1 and H4-2, are replicated at the onset of S phase in the naturally synchronous plasmodium of Physarum polycephalum, suggesting that they are flanked by replication origins. It was further shown that a DNA fragment upstream of the H4-1 gene is able to confer autonomous replication of a plasmid in the budding yeast. In this paper, we re-investigated replication of the unlinked Physarum histone H4 genes by mapping the replication origin of these two loci using alkaline agarose gel and neutral/neutral 2-dimensional agarose gel electrophoreses. We showed that the two replicons containing the H4 genes are simultaneously activated at the onset of S phase and we mapped an efficient, bidirectional replication origin in the vicinity of each gene. Our data demonstrated that the Physarum sequence that functions as an ARS in yeast is not the site of replication initiation at the H4-1 locus. We also observed a stalling of the rightward moving replication fork downstream of the H4-1 gene, in a region where transient topoisomerase II sites were previously mapped. Our results further extend the concept of replication/transcription coupling in Physarum to cell cycle-regulated genes.

INTRODUCTION

In eukaryotic cells, multiple sites are coordinately used to initiate replication of the chromosomes. The characterisation of these replication origins is still under progress: only in eukaryotic viruses and in the yeast Saccharomyces cerevisiae are they precisely defined at the level of the nucleotidic sequence. In yeast, genetic assays allowed the identification of ARS (autonomously replicating sequence) elements, which promote the replication of a plasmid lacking its intrinsic origin. Their structure has been extensively studied and a consensus sequence (ACS) delineated (1). The physical mapping of replication origins within the chromosomes was achieved later by the development of 2-dimensional agarose gel electrophoresis techniques (2,3): most, but not all, ARS elements function as origins of replication in their chromosomal context (4). Therefore, the ARS assay is not sufficient to define a chromosomal replication origin. These data also imply that the ARS nucleotide sequence by itself is not the only determinant for the selection of replication initiation sites in yeast and show that the mechanism of origin activation is not fully understood. In metazoan cells, genetic assays have failed to identify specific DNA sequences that would drive replication initiation. Thus, any DNA sequences allow the replication of an EBV-derived vector, provided they are long enough (5). Similarly, replication of exogenous DNA in Xenopus eggs depends on the size of DNA and on the nuclear structure, rather than on the DNA sequence itself (6). Various mapping strategies of replication origins in metazoan cells have resulted in controversial data, as exemplified by the dihydrofolate reductase locus in Chinese hamster ovary cells, where either a 0.5 kb origin of bidirectional replication (7,8) or a 55 kb initiation zone (9,10) were described. This suggests that metazoan origins are complex and perhaps do not require any specific sequence.

However, the origin recognition complex (ORC), first identified in yeast as a trans-acting protein complex that binds to the ARS (11,12) and promotes the assembling of a prereplication complex at the origin, is conserved among the eukaryotes. The discovery in different organisms of proteins homologous to the prereplication complex proteins cdc6 and mcm2-7 (13) also suggests that the mechanism of initiation is conserved and based on specific cis-acting DNA sequences. This underlines the usefulness of model systems in which replication origins can be localised, in order to recapitulate the evolutionary history of the origins. In the yeasts Schizosaccharomyces pombe and Yarrowia lipolytica, ARS assays have been successful and 2-dimensional gel mapping confirmed the identification of chromosomal replication origins (14-17). The nucleotidic sequence responsible for origin activity is not yet fully defined and is different from the budding yeast ACS (14).

In the slime mould Physarum polycephalum there is no ARS assay available at this time, but the natural synchrony of the nuclei in the plasmodial stage facilitated the physical mapping of replication origins. Fibre autoradiography and electron microscope studies have shown that the mean size of replicons is 35 kb (18). This is similar to the size of the budding yeast replicons. Yet, the two genomes are different: in Physarum, the C value is 3 × 108 bp (20 times larger than the yeast genome), so that 18 000 replicons are expected to duplicate the DNA of each of the diploid nuclei of the plasmodium, whereas about 400 origins are contained in the yeast genome. From the kinetics of replication throughout the 3 h S phase of the plasmodium, one can calculate that the subset of replication origins activated at the onset of S phase contains around 4000 origins (19). Interestingly, the observation of replicating chromatin under the electron microscope, as Miller spreads prepared from an early S phase plasmodium, has shown newly replicated genes in the middle of some nascent replication bubbles (19). As replication proceeds bidirectionally in Physarum (18), these results strongly suggest that each of these genes is situated in the vicinity of a replication origin. This link was further demonstrated at the level of particular loci: by the neutral 2-dimensional gel method (2), we mapped three bidirectional, efficient and site-specific replication origins activated at the onset of S phase, each of them located in the promoter region of an abundantly expressed gene. One was found upstream of the developmentally regulated profilin P gene (20), a gene highly transcribed in the diploid plasmodium but not in the haploid amoeba; the other ones were found upstream of the ardB and ardC actin genes (21), which are constitutively expressed in both developmental stages. With the aim of defining replication origins in Physarum, comparing them with the yeast ARS elements and further analysing their relationship with the genes, we now study the replication of genes showing another expression pattern, the cell cycle-regulated histone H4 gene family.

The Physarum histone H4 genes are of low iteration: two unlinked genes, H4-1 and H4-2, have been identified. The coding sequences are 87% homologous and both are interrupted by a short intron (86 and 90 bp, respectively). The deduced amino acid sequences are identical and highly similar to the H4 histones from other organisms. The two genes are equally transcribed (22). The level of H4 mRNA was followed by northern blot analysis: the highest amounts of H4 mRNA were found in late G2 and early S phase, with a maximum at stage +30 min, which corresponds to a high DNA synthesis rate in the plasmodium of Physarum (23). Run-on assays have shown that the S phase transcription is dependent on DNA synthesis activity (24). The G2 transcription is not immediately followed by translation, suggesting that a pool of histone H4 mRNA is stored for the subsequent S phase (23).

The replication timing of H4-1 and H4-2 has been previously investigated in the plasmodium by in vivo incorporation of bromodeoxyuridine: the two genes were found to replicate with the early component of the genome (25). Interestingly, a Physarum DNA fragment located 2 kb upstream of the H4-1 gene enhances the transformation efficiency in yeast of a plasmid devoid of a replication origin. It was consequently hypothesised that this element would play a role in replication in Physarum (26). Here, we analysed replication of the cell cycle-regulated histone H4 genes using the neutral/neutral 2-dimensional gel method (2). We mapped a bidirectional replication origin confined to a small region (1-2 kb) encompassing each gene and we showed that the H4-1 gene is not replicated from the DNA segment exhibiting ARS activity in yeast.

MATERIALS AND METHODS

Strains and cultures

Plasmodia of strain M3cIV were grown from microplasmodia subcultures as described (21). The plasmodium is a multinucleated cell in which millions of nuclei, in a common cytoplasm, are dividing simultaneously every 10 h, entering a 30 min long mitosis, immediately followed by a 3 h S phase and a 7 h G2 phase. Plasmodia of 5 cm diameter were harvested after mitosis II or III, as deduced from the observation of ethanol-fixed smears under a phase contrast microscope. We used the typical morphological structures of mitosis stages to determine the onset of S phase. A time was attributed to each DNA sample, which refers to the minutes after telophase at which the plasmodium was harvested. As indicated by previous analysis of specific replication intermediates, this time defines a temporal window of a few minutes only.

DNA isolation and agarose gel electrophoreses

Nuclei from a whole plasmodium were isolated as described (21). For the alkaline gel electrophoresis experiment (Fig. 1), DNA was extracted with organic solvent as previously reported (20). DNA samples were denatured and loaded onto a 0.6% alkaline agarose gel run at 1 V/cm for 22 h (20). DNA was transferred on a nylon membrane (GeneScreen Plus; NEN) with a VacuGene apparatus (Pharmacia).

For 2-dimensional gel experiments, isolated nuclei were embedded in 1 vol of 1% low melting point agarose (21). Plugs 1 cm long were digested with restriction enzymes and the structure of replication intermediates was analysed by neutral 2-dimensional gel electrophoresis (2). The first electrophoresis was performed on a 0.4% agarose gel run at 0.5-1 V/cm for 24-60 h and the second one on a 1% agarose gel run at 4°C at 3 V/cm for 15 h. The DNA was transferred onto a nylon membrane and hybridised.

Southern hybridisation and probes

The membrane was prehybridised in 1% SDS, 10% dextran sulphate, 1 M NaCl for 1 h at 65°C. Hybridisation was initiated by adding the labelled probe and 0.1 mg/ml heat-denaturated salmon testes DNA. After 16 h at 65°C, the filter was washed in successive baths of increasing stringency for 15 min at 65°C (final bath, 15 mM NaCl, 1.5 mM sodium citrate, 0.1% SDS).

Because of sequence similarities between the two genes, we used genomic probes under stringent hybridisation conditions. In each case, the restriction fragments containing the other H4 gene appeared as faint bands on Southern blots (data not shown), which did not interfere with 2-dimensional gel analyses. The H4-1 probe is a 0.6 kb HindIII fragment (Figs 1-4) while the H4-2 probe is a 1.4 kb HindIII-EcoRI fragment (Fig. 6). They have been derived from genomic clones kindly provided by M. and F. X. Wilhelm. The probes were radiolabelled with 50 µCi of [[alpha]-32P]dCTP by random primed labelling (NEN).

DNA sequence analysis

The sequence of the 2.8 kb EcoRI-HindIII fragment upstream of the H4-1 gene was determined on both strands by the chain termination method (Fidelity; Appligene). The contiguity of the subclone containing the gene and the one containing upstream sequences (26) was demonstrated by PCR amplification of genomic DNA with one oligonucleotide located in the H4-1 gene coding sequence (5[prime]-ATACTTCGTTTCATCAAAGG) and one oligonucleotide located in the promoter-containing sequence (5[prime]-GATAGAGGGATTAGAAATAGAT). The expected 1 kb PCR product was obtained and sequenced.

Sequence analysis was performed using the GCG packageand Thermodyn program. The latter was kindly provided by D. Kowalski.

The GenBank accession numbers for the reported sequences are: H4-1 coding region, X15142; H4-1 upstream region, AJ132393; proP promoter region, M38038; ardB and ardC promoter regions, X60788 and M73459, respectively.

RESULTS

The H4-1 gene is linked to a replication origin

Physarum histone H4 is encoded by two unlinked genes, H4-1 and H4-2. A previous study using in vivo incorporation of bromodeoxyuridine showed that the H4-1 gene is replicated within the first 10 min of S phase (25). Considering that the replication forks progress at a rate of 1.2 kb/min/replicon in the plasmodium (18), the early replication of H4-1 suggested that the gene is at most a few kilobases from an origin of replication. To specify the distance between the H4-1 gene and its replication origin, we analysed the nascent strands of the H4-1 replicon by alkaline gel electrophoresis (Fig. 1). Synchronous S and G2 phase DNA samples were run on a denaturing gel and hybridised with the H4-1 probe. A signal corresponding to the high molecular weight parental DNA is seen in each lane. In S phase DNA samples, an additional signal was detected, which corresponds to the nascent strands of the growing replicon; their size ranged from 1.8 to 6 kb at the earliest stages (+5 and +7 min after the onset of S phase) and from 10 to 15 kb at +15 min. Hybridisation of the H4-1 probe with nascent strands at the beginning of S phase indicates early replication of the gene. This result, obtained in the absence of any treatment of the cell, extends the data previously obtained from density shift experiments (25). The short size of the nascent strands at +7 min in Figure 1 limits the location of the replication origin. Indeed, the nascent strands hybridizing with the H4-1 probe contain the replication origin and have reached the H4-1 gene (see the hypothetical positions of the nascent strands in Fig. 1). Since the shortest fragments that hybridise with H4-1 are only 1.8 kb long, the replication origin lies at most 1.8 kb from either end of the gene, and even less if the locus is bidirectionally replicated, as is generally the case in Physarum (see spotted rectangle in Fig. 1). As the DNA fragment that is active as an ARS in yeast is located 2 kb upstream of the gene (26), this first experiment does not exclude completely that this DNA site corresponds to the H4-1 gene replication origin (Fig. 1).


Figure 1. Detection of nascent strands of the H4-1 gene replicon by alkaline gel electrophoresis. DNA samples were obtained from plasmodia at different stages of the mitotic cycle as indicated at the top; the numbers refer to minutes after the onset of S phase. They were denaturated by alkali, loaded on a 0.6% alkaline agarose gel, hybridised with the H4-1 probe (black bar) and autoradiographed. In the G2 phase sample, only undegraded high molecular weight DNA hybridises with the genomic H4-1 probe. Nascent strands containing the H4-1 gene are detected at early stages of S phase. At +5 and +7 min, the nascent strands vary from 1.8 to 6 kb in size, with a mean size of 3 kb. At +15 min, they reach 10-15 kb with a mean size of 13 kb. The size marker on the left is a HindIII digest of [lambda] DNA. The more extreme possible locations with respect to the probe of the 1.8 kb nascent strands are depicted as rectangles with crosshatching (see text). Considering that replication is generally bidirectional in Physarum, as symbolised by arrowheads, these nascent strand positions define a 1.4 kb region in which the initiation is likely to occur: this maps the origin in the vicinity of the gene, as shown by the spotted rectangle. The grey box indicates the EcoRI-SmaI fragment having an ARS activity in yeast (26). E, EcoRI; S, SmaI; H, HindIII. The scale is indicated under the map.

Kinetics of DNA replication at the H4-1 locus

The neutral/neutral 2-dimensional gel technique reveals the structure of replication intermediates within restricted DNA fragments (2): it allows discrimination between DNA fragments that are replicated by two divergent replication forks (arc of bubble structures) and the ones that are replicated by a single replication fork (arc of Y-shaped molecules). Taking advantage of a restriction fragment length polymorphism, we followed the kinetics of appearance and disappearance of replication forks within the 6.6 and 12 kb EcoRI allelic DNA fragments containing H4-1 (Fig. 2). As these fragments encompass the gene, we anticipated that they contain the replication origin evidenced in the previous experiment and depicted under the autoradiograms. Indeed, as soon as +3 min, prominent nascent arcs of bubble structures were detected for both alleles (Fig. 2), which means that initiation events occur simultaneously and within these allelic fragments. At +5 min, the signal extending from the smallest allelic fragment (H4-1a in Fig. 2) has evolved into a clear bubble arc to Y arc transition. The replication intermediates in the largest fragment (H4-1b) also formed a discontinuous signal composed of the end of an arc of bubble structures and the apex region of a simple Y arc (see the interpretative scheme next to the autoradiogram). Thus, for both allelic fragments, the bubble arc switched into a simple Y arc, a non-composite pattern which demonstrates that the initiation of replication takes place at an efficient, bidirectional origin asymmetrically located in each DNA fragment. Its position is probably more asymmetric in H4-1b since the Y arc started from the apex, whereas a more restricted portion of the Y arc was seen for H4-1a. Unlike the 6.6 kb EcoRI fragment, not all of the 12 kb fragment was duplicated at +15 min; the replication intermediates were distributed on the downward part of a simple Y arc, indicating that the H4-1b fragment was then replicated by a single fork. The replication of both restriction fragments was completed at +40 min. These kinetics show the high level of synchrony of the nuclei in the plasmodium and define a temporal window of a few minutes for the timing of DNA sampling. As illustrated by the drawing of fork progression under the autoradiograms in Figure 2, these observations are consistent with our previous origin mapping (Fig. 1) and the linkage between the gene and its replication origin is reinforced by the demonstration that the locus is bidirectionally replicated.


Figure 2. Kinetics of replication at the H4-1 locus. DNA plugs prepared at various time points of the plasmodial 3 h S phase were digested by EcoRI, subjected to neutral/neutral 2-dimensional gel electrophoresis and hybridised with the H4-1 probe. For clarity, we have depicted below each neutral 2-dimensional gel a diagram of fork progression based on the mapping of the origin presented in Figure 1, reproduced as the spotted rectangle under the restriction maps of the two allelic loci (E, EcoRI; E*, the polymorphic EcoRI site). The 1x spots correspond to the linear allelic 6.6 and 12 kb fragments (a and b respectively on the autoradiograms). At +3 min, nascent arcs of bubble-shaped molecules, typical of an active replication of the restriction fragments, are observed for both alleles. At +5 min a transition from a bubble arc to a Y arc is seen for the H4-1a EcoRI fragment. A similar signal is obtained in the case of the H4-1b EcoRI fragment (an interpretative diagram is drawn on the right) but, because of the higher molecular weight of the replicating fragments, the end of the bubble arc appears with an unusual shape and turns into a simple Y arc of which only the apex is seen at this stage. At +15 min, only the H4-1a 1x spot is detected, as the replication forks disappear from within the fragment. The H4-1b fragment is then replicated by a single replication fork, as shown by the detection of an arc of Y-shaped molecules. At +40 min, both allelic fragments are duplicated.

Stalling of the rightward-moving replication fork downstream of the H4-1 gene

Our data locate a bidirectional efficient replication origin within the centre of the 6.6 kb EcoRI fragment. This assumption is based on equivalent rates for both divergent forks of the H4-1 replicon. However, we detected a fork stalling within the H4-1 locus, as is illustrated in Figure 3. In Figure 3A, the allelic SacI fragments that encompass the H4-1 gene are shown. The replication intermediates of both alleles appeared as identical signals, consisting of an arc of bubble structures which was converted into an arc of Y-shaped molecules. This pattern was expected since the replication origin lies close to the centre of these restriction fragments. Interestingly, the intensity of the hybridisation signals was stronger in the final portion of the simple Y arcs. This stronger signal suggests that the rightward replication fork is stalled near the end of the fragment, as illustrated by the scheme under the autoradiogram. We therefore analysed a BamHI-SmaI restriction fragment that overlaps the 3[prime]-end of the gene: an accumulation of replication intermediates was detected in the apex region of the Y arc resulting from the rightward fork movement (Fig. 3B), demonstrating that the replication fork is slowing down near the centre of the fragment. Consistently, when we analysed the BamHI fragment which extends further downstream, the replication intermediates were mainly distributed in the ascending part of the Y arc, with a higher density just before the apex (Fig. 3C). These specific variations of signal intensities indicate that the rightward replication fork is stalled in the 3[prime]-region of the gene, close to the downstream SacI site (see the black rectangle in Fig. 3).


Figure 3. Stalling of the rightward replication fork downstream of the H4-1 gene. DNA samples prepared at +3 or +5 min from early S phase plasmodia were digested by the appropriate restriction enzymes, submitted to 2-dimensional gel electrophoresis and hybridised with the H4-1 probe. An accumulation of replication intermediates corresponding to a specific position of the replication forks in each restriction fragment is detected (arrows). In (A), identical signals are detected for the allelic 4.2 and 4.4 kb SacI fragments, consisting of a transition from bubble arcs to Y arcs. This confirms that an efficient bidirectional replication origin is located in the central third of each allelic fragment (spotted rectangle). In addition, we noticed that the replication intermediates are particularly dense in the downward part of the Y arc (arrow). This corresponds to the downstream part of the fragment (see below), as shown in scheme (A) by the hypothetical positions of the replication forks within the restriction fragment. In (B), the 5.4 kb BamHI-SmaI fragment is mainly replicated by a single fork, as revealed by the detection of a Y arc. The replication intermediates are aggregated in the apex region, showing that the rightward replication fork is stalled in the middle of the fragment (scheme B). In (C), a DNA sample extracted slightly earlier in S phase (+3 instead of +5 min) was cut with BamHI. A single replication fork moves through this 7.0 kb fragment; the replication intermediates accumulate in the upward part of the Y arc, which means that the rightward replication fork is stalling in the upstream half of the fragment (scheme C). Taken together, these results permitted mapping of the stalling of the replication fork downstream of the H4-1 gene, near the SacI site, as shown by the black rectangle under the map. Sa, SacI; Sm, SmaI; B, BamHI.


Figure 4. Mapping of a bidirectional replication origin coincident with the H4-1 gene. A DNA sample prepared from a plasmodium at stage +5 min was digested by restriction enzymes, run onto a 2-dimensional gel and probed with the H4-1 probe. Only the H4-1a allele, of which the restriction map is drawn at the bottom, is shown. (A) The EcoRI restricted DNA gives a discontinuous pattern from an arc of bubble structures to a simple Y arc (the descending part of the Y arc is partly obscured by the signal extending from the 1x spot of the H4-1b allele). It means that the fragment contains an efficient bidirectional replication origin in its central third (hatched bars). (B) Digestion with SmaI and KpnI enzymes results in a 3.0 kb SmaI-KpnI fragment (H4-1a) and a 6.7 kb SmaI fragment (H4-1b) because the KpnI site is an allele-specific site. The 3.0 kb SmaI-KpnI fragment is actively replicated since a transition from a bubble arc to a Y arc is seen; the position of the internal origin is represented by a hatched bar. This experiment definitively rules out the possibility that the fragment with an ARS activity in yeast (grey box), which is located outside the SmaI-KpnI fragment, is used as the site of replication initiation for the H4-1 locus in Physarum. (C) The 4.1 kb BamHI-EcoRI fragment that lies 3[prime] to the gene is analysed. The fragment is mainly replicated by a single replication fork since a simple Y arc is detected. This restricts the replication origin position in one third of the fragment, as shown by the hatched bar. These data delineate the most probable location of the origin as a 1 kb fragment centred on the transcription initiation site, as illustrated by the schematic replication bubble above the map. Notice that it is consistent with the mapping obtained from alkaline gel analysis (spotted rectangle). B, BamHI; E, EcoRI; K, KpnI; Sa, SacI; Sm, SmaI.

2-Dimensional gel mapping of the H4-1 replication origin

Taking into account the irregular rightward fork progression, we further mapped the replication origin with respect to the H4-1 gene and the ARS element by 2-dimensional gel analysis (Fig. 4). We used a single early S phase DNA sample that was first digested with EcoRI to estimate the distribution of the replication intermediates in the DNA preparation. We focused our analysis on the H4-1a allele, which is contained in the smallest EcoRI fragment (Fig. 4A): this 6.6 kb fragment presented a bubble to Y arc transition, a pattern very similar to the one obtained at stage +5 min in Figure 2. This shows the reproducibility of the sampling and of the temporal control of the replication origin contained within the fragment (see hatched bars in rectangle A). This EcoRI fragment overlaps the stalling region 3[prime] to the gene (see black bar in Fig. 4), but in the bubble arc, the accumulation of replication intermediates is reduced as a result of the ongoing elongation by the leftward moving replication fork. We then analysed the internal SmaI-KpnI fragment, a 3.0 kb fragment that does not contain the stalling region (Fig. 4B). A transition from a bubble arc to a Y arc was again observed: the central third of the fragment, where the origin can be located, corresponds to the H4-1 coding sequence and neighbouring upstream region (see hatched bars in rectangle B). The location of the origin was further assessed using a restriction enzyme that cuts near this location (Fig. 4C). The downstream BamHI-EcoRI fragment is replicated mainly by a single replication fork, since the replication intermediates pattern was a simple Y arc which limits the presence of an origin to the terminal third of the fragment (hatched bar in rectangle C). The 3.0 kb SmaI-KpnI fragment analysed in Figure 4B was the smallest one in which we could detect an arc of bubble-shaped molecules and the KpnI site is located ahead of the stalling of the replication fork. Therefore, we assume that the mapping of the origin based on the 2-dimensional gel analysis of this fragment is reliable: the initiation takes place within a 1 kb region comprising the promoter and the coding sequence of H4-1 (see the bubble structure schematised over the restriction map in Fig. 4). This also demonstrates that the fragment that replicates autonomously in yeast (grey box), which lies outside the 3 kb SmaI-KpnI fragment (Fig. 4), is not the site of replication initiation in Physarum.

Sequence analysis

We sequenced 2.8 kb upstream of the H4-1 gene, including the yeast ARS element. We found four consecutive near matches of the ARS consensus sequence (ACS) located 2600 bp upstream of the gene and an additional near match 140 bp closer to the gene (Fig. 5A). All differ from the 11 bp ACS (WTTTATRTTW) by one mismatch, yet they are likely to be involved in the ARS activity of the 1.2 kb EcoRI-SmaI fragment in yeast. A similar ability for autonomous replication has been shown for yeast nucleotide sequences nearly matching the ACS (1). At ~900 bp from the replication origin mapped at the H4-1 locus by the 2-dimensional gel method, we found a polypurine track composed of 20 repeats of the AGGAAAGGG motif. A stretch of purine has been previously identified in the promoter region of the profilin P gene that also contains a replication origin (20). Application of the Thermodyn program on the 4 kb nucleotide sequence encompassing the gene (Fig. 5B) showed a potential low helical stability in the vicinity of the replication origin mapped in this study. This is reminiscent of the DNA unwinding elements (DUEs) that have been identified close to oriC in Escherichia coli and to yeast replication origins (27). We compared the sequences upstream of the ardB and ardC genes, of the profilin P gene and of the H4-1 gene and we did not detect obvious homologies between these regions that contain origins of replication of the Physarum genome; the polypurine tracks and the potential DUEs are the only common features detected between them.


Figure 5. Sequence analysis of the H4-1 locus. We sequenced the EcoRI-HindIII fragment upstream of H4-1. On the nucleotides scale, +1 represents the transcription initiation site. The sequence is 46% GC-rich, compared with the 42% of the Physarum genome. We did not find other histone genes, which suggests that these genes are not organised in a tight cluster in Physarum. The particular features that were found in the sequence are depicted on a map in (A). yARS corresponds to five close matches to the yeast ACS; they are contained within the EcoRI-SmaI fragment which confers autonomous replication to a yeast plasmid (26). The rectangle named pur is a purine track composed of 20 repeats of the AGGAAAGGG motif. DUE is a region of potentially low helical stability (see B). The arrows indicate the topoisomerase II cleavage sites that were shown to appear in mitosis and early S phase at the H4-1 locus (34). Below the map are indicated the replication origin (ori) and stalling of the rightward replication fork (block) as they have been mapped by 2-dimensional gel analysis. In (B), the helical stability ([Delta]G) of the H4-1 gene region was plotted following Thermodyn program analysis (27). The scale at the top is the same as in (A). The most putative unstable region is a large valley with a minimum value of 70 kcal/mol, at -700 nt from the transcription initiation start, close to the region where we mapped the origin. The two other low free energy regions correspond to the yeast ACS, at -2600 nt, and to the 3[prime] flanking sequences.


Figure 6. Mapping of a bidirectional replication origin in the vicinity of the H4-2 gene. DNA samples in early S phase were digested with appropriate restriction enzymes, subjected to 2-dimensional gel electrophoresis and hybridised with the H4-2 probe. (A) A +5 min DNA sample was digested with EcoRV. Because of a restriction fragment length polymorphism, the two major 1x spots correspond to the allelic 7 and 10 kb EcoRV fragments (the minor 1x spots correspond to a partial restriction digestion). We focused on the well-resolved replication intermediates extending from the smallest fragment. A transition from a bubble arc to a Y arc is detected (see the interpretative scheme next to the autoradiogram), which means that an internal bidirectional origin is responsible for replication of the fragment (hatched bar). In (B), we further confirmed the mapping of the H4-2 origin by cutting close to the initiation site: another +5 min DNA sample was digested with HindIII and arcs of Y-shaped molecules are extending from the 1x spots corresponding to the linear 5.0 and 5.2 kb allelic fragments. In (C), a 12 kb EcoRI fragment is analysed and a partial simple Y arc is detected. This excludes the origin from the central third of the fragment. The DNA sample was prepared from a +15 min plasmodium, so that the replication forks have a different distribution than in (A) and (B), as illustrated by arrowheads under the rectangle depicting the restricted DNA fragment. The patterns of replication intermediates at the H4-2 locus allow mapping of the replication origin close to the gene as illustrated by the schematic bubble structure above the map. E, EcoRI; Ev, EcoRV; H, HindIII.

2-Dimensional gel mapping of the H4-2 replication origin

We then analysed the replication pattern of the second histone H4 gene, H4-2. This gene was reported to replicate between 20 and 30 min after the onset of S phase, according to a bromodeoxyuridine density shift experiment (25). Yet, we detected replication intermediates in restriction fragments encompassing H4-2 as early as +5 min after the onset of S phase for both alleles of H4-2 (Fig. 6). At this stage, we observed a clear transition from a bubble arc to a Y arc for the shortest of the two allelic EcoRV fragments (Fig 6A). This demonstrates that a localised, efficient and bidirectional replication origin is contained within this 7 kb fragment (hatched bar in rectangle A). To confirm the localisation of the origin, we analysed the 5.0 and 5.2 kb allelic HindIII fragments that contain the gene and downstream sequences (Fig. 6B). We obtained complete arcs of Y-shaped molecules, showing that the fragments were mainly replicated by a single replication fork. Similarly, at a later time in S phase, a partial simple Y arc was extending from the 12 kb EcoRI fragment that includes the H4-2 gene and a large upstream region (Fig. 6C). These data allow us to map a bidirectional replication origin within a 2 kb region that contains the H4-2 coding sequence and proximal upstream region (see the schematic bubble drawn above the map in Fig. 6). Furthermore, this experiment enables us to compare the replication of both H4 genes. Indeed, the EcoRI-digested DNA sample analysed in Figure 6C is the +15 min sample presented in Figure 2. After rehybridisation with the H4-2 probe, similar replicative signals are detected, which further confirms the early replication of H4-2 in our strain. Therefore, we conclude that the two H4 genes are contained within replicons concomitantly activated at the onset of S phase. The bidirectional replication origins of each replicon are coincident with these cell cycle-regulated genes.

DISCUSSION

We have used both alkaline gel and 2-dimensional gel electro-phoresis analyses to demonstrate that the two unlinked histone H4 genes of Physarum are contained within replicons activated at the very beginning of the plasmodium S phase. Denaturing gel experiments indicated that the H4-1 gene is tightly linked to a replication origin (Fig. 1). Further studies by 2-dimensional gel analysis showed that the H4-1 locus is replicated bidirectionally and confirmed that the origin and the gene are coincident (Fig. 4); similar properties were found for the H4-2 gene (Fig. 6). The H4-1 origin has been mapped in a 1 kb region centred on the site of transcription initiation, whereas a slightly larger 2 kb region was identified from the analysis of H4-2 replication. Thus, for both loci, the initiation of DNA replication is confined to a short, specific DNA region.

Our 2-dimensional gel analysis also allows visualisation of the replication fork progression and to map a stalling of the rightward-moving fork downstream of the H4-1 gene. This was deduced from the abrupt increase in the hybridisation signal intensity along the arcs of replicating restriction fragments (Fig. 3); this accumulation of replication intermediates extends over a region of ~1.3 kb where the progression of the rightward-moving replication fork is hampered. Unlike the replication fork barriers found in the ribosomal gene cluster in S.cerevisiae (28,29), stalling of the replication fork downstream of H4-1 is transient, since the fork passes through the site, as evidenced by completion of the Y arc in Figure 4C. This resembles the pause sites found upstream of some genes of the Drosophila histone gene repeats (30), in the 5[prime]-non-transcribed spacer of Tetrahymena and Physarum rDNA (31,32) and downstream of tRNA genes in S.cerevisiae (33). Our data indicate that stalling of the replication fork is located close to the H4-1 origin (at ~1.5 kb), as is the case of the pause sites located upstream of Tetrahymena and Physarum rRNA genes. The stalling of a replication fork right after it has been generated is paradoxal and it emphasises the fact that DNA replication is not a smooth process and is submitted to regulation of fork rate elongation. The pause sites that were identified downstream of several tRNA genes in yeast are related to the opposite orientations of replication and transcription at these highly expressed loci (33). We cannot propose the same explanation for the case of the Physarum H4-1 gene, since the pause region is downstream of both the origin and the gene, which is co-directionally replicated and transcribed. Interestingly, topoisomerase II sites have been mapped in the same region where fork stalling takes place (34; Fig. 5). Moreover, the enzyme was trapped at these sites in early S phase (34), at a time that coincides with the progression of replication forks through the H4-1 locus (Fig. 2). Therefore, these results suggest that topoisomerase II is involved in the resolution of a particular chromatin structure, downstream of the gene, which slows down the replication fork. The involvement of a particular DNA-protein structure has also been postulated to explain the replication pause sites found in the untranscribed centromeres of S.cerevisiae and Y.lipolytica (35,17).

Although stalling of the fork downstream of H4-1 complicated the origin mapping, several arguments support the localisation of the origin in the promoter region of the gene. First, the 0.6 kb probe encompassing H4-1 hybridised to nascent strands no shorter than 1.8 kb (Fig. 1), which is best explained by an origin outside the gene. Further, 2-dimensional gel mapping is consistent with a 5[prime] location of the origin with respect to the H4-1 gene. Such a location corresponds to the centre of the 3.0 kb SmaI-KpnI fragment, the smallest DNA fragment that contains bubble-shaped molecules and is not affected by pausing (Fig. 4B). Moreover, cutting at the internal BamHI site converts the signal into a simple Y arc (Figs 3C and 4C), demonstrating that nascent bubbles overlapped this restriction site located 0.4 kb upstream of the gene. We therefore conclude that the histone H4-1 origin lies most likely immediately upstream of the H4-1 gene. It is of note that a computer analysis indicates that the lowest intrinsic stability of the DNA helix at this locus is found 0.7 kb 5[prime] to the H4-1 gene (27; Fig. 5). Similarly, our 2-dimensional gel analysis of the H4-2 locus is compatible with an origin located in the promoter region of the H4-2 gene (Fig. 6). This contrasts with the histone H4 locus in S.cerevisiae, where an ARS element was found downstream of the gene (36), although the origin activity of this ARS within the yeast chromosome has not been demonstrated. In fact, in the budding yeast genome, several origins are found 3[prime] to genes; therefore its functional organisation appears different from the Physarum genome, even though both organisms have site-specific replication origins.

Transcription of the Physarum H4 genes was previously found to vary during the cell cycle, with high levels of H4 mRNA and of transcriptional activity of the genes in late G2 and in S phase (23,24). There is no G1 phase in the plasmodium; considering that transcription is interrupted during mitosis (19), replication of the H4 genes, which occurs immediately after decondensation of the chromosomes in telophase, is likely to take place before resumption of transcription. We found a coincidence between several abundantly expressed genes, including the cell cycle-regulated histone H4 genes, and origins activated at the onset of S phase (20,21). Similar features, like an open chromatin state, are needed for both transcription and replication and observations of their coupling were also reported for different organisms. For instance, replication was found to initiate upstream of one of the Drosophila chorion genes, upstream of the Sciara DNA puffII/9A and the human c-myc and [beta]-globin genes (37-40). This might reflect the sharing of regulating factors and cis-acting sequences, as is the case for eukaryotic viruses (41).

Our sequence analysis of the region upstream of the Physarum H4-1 gene (Fig. 6) revealed that multiple ACS near matches are contained in the DNA fragment active as an ARS in budding yeast (26). This element was previously identified as a putative origin in Physarum. However, our 2-dimensional gel analysis of the H4-1 locus clearly showed that the replication origin is distinct from the yeast ARS element (Fig. 4). To our knowledge, it is the first time that a foreign DNA sequence having an ARS activity in yeast is shown not to coincide with an active origin in its native chromosomal context. In the same way, the origin-containing region upstream of the profilin P gene of Physarum is unable to confer autonomous replication to a plasmid in yeast (20; unpublished results). This further establishes the lack of ACS-like sequences in the replication origins of Physarum and indicates that the molecular definition of the yeast replicators is of little help in defining the primary sequence of replicators in other organisms. However, it is a formal possibility that the Physarum DNA fragment located 2 kb upstream of the replication origin of the H4-1 locus and that is active as an ARS in yeast might play a role in controlling initiation in the H4-1 replicon of Physarum. Yet, this situation does not exist in yeast, where replicators correspond to the physical sites of initiation of DNA replication (42). We have evidence that they are also coincident in Physarum, since the origin linked to the 1 kb promoter region of the actin C gene is still active in an ectopic chromosomal site (43). These observations raise the question of the nature of Physarum replication origins: sequence comparison did not allow the identification of a common motif among the origin-containing regions that we have mapped so far, i.e. the 1 kb regions upstream of the ardB and ardC actin genes (21) and the 1.3 kb region upstream of the profilin P gene (20).

Finally, our data underline the strict timing of replication during early S phase in the plasmodium of Physarum (Figs 2, 4 and 6). Recent results on the replication of the entire yeast chromosome VI demonstrated that origins are rarely active in 100% of the cells; only one origin is very efficient and fires very early (44). These two properties are exemplified for the origins of the two H4 gene loci. In this context, it would be informative to compare the origins that fire at the onset of S phase in the plasmodium nuclei with the ones that fire later on. We previously identified several loci that are replicated in mid and late S phase (45). The study of replication initiation at these loci would allow specification of whether or not the temporal order of DNA replication is as strictly controlled throughout the S phase of Physarum.

ACKNOWLEDGEMENTS

We thank M. Wilhelm and F. X. Wilhelm for kindly providing the H4 plasmids, D. Kowalski for the gift of the Thermodyn program, J. Pédron and Y. Florentin for expert technical assistance, V. Borde for sharing information on the H4-1 locus sequence and D. Pallotta for helpful comments on the manuscript. This work was supported by general funding from the CNRS and grant 1301 from the Association pour la Recherche sur le Cancer.

REFERENCES

1. Newlon,C.S. (1996) In DePamphilis,M. (ed.), DNA Replication in Eukaryotic Cells. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, pp. 873-914.

2. Brewer,B.J. and Fangman,W.L. (1987) Cell, 51, 463-471. MEDLINE Abstract

3. Huberman,J.A., Spotila,L.D., Nawotka,K.E., El-Assouli,S.M. and Davis,L.R. (1987) Cell, 51, 473-481. MEDLINE Abstract

4. Fangman,W.L. and Brewer,B.J. (1991) Annu. Rev. Cell. Biol., 7, 375-402. MEDLINE Abstract

5. Krysan,P.J. and Calos,M.P. (1991) Mol. Cell. Biol., 11, 1464-1472. MEDLINE Abstract

6. Coverley,D. and Laskey,R.A. (1994) Annu. Rev. Biochem., 63, 745-776. MEDLINE Abstract

7. Burhans,W.C., Vassilev,L.T., Caddle,M.S., Heintz,N.H. and DePamphilis,M.L. (1990) Cell, 62, 955-965. MEDLINE Abstract

8. Pelizon,C., Diviacco,S., Falaschi,A. and Giacca,M. (1996) Mol. Cell. Biol., 16, 5358-5364. MEDLINE Abstract

9. Vaughn,J.P., Dijkwell,P.A. and Hamlin,J.L. (1990) Cell, 61, 1075-1087. MEDLINE Abstract

10. Wang. S., Dijkwel,P.A. and Hamlin,J.L. (1998) Mol. Cell. Biol., 18, 39-50. MEDLINE Abstract

11. Bell,S.P. and Stillman,B. (1992) Nature, 357,128-134. MEDLINE Abstract

12. Diffley,J.F.X. and Cocker,J.H. (1992) Nature, 357, 169-172. MEDLINE Abstract

13. Dutta,A. and Bell,S.P. (1997) Annu. Rev. Cell Dev. Biol., 13, 293-332.

14. Clyne,R.K. and Kelly,T.J. (1995) EMBO J., 14, 6348-6357. MEDLINE Abstract

15. Dubey,D.D., Kim,S.M., Todorov,I.T. and Huberman,J.A. (1996) Curr. Biol., 6, 467-473. MEDLINE Abstract

16. Okuno,Y., Okasaki,T. and Masukata,H. (1997) Nucleic Acids Res., 25, 530-537. MEDLINE Abstract

17. Vernis,L., Abbas,A., Chasles,M., Gaillardin,C.M., Brun,C., Huberman,J.A. and Fournier,P. (1997) Mol. Cell. Biol., 17, 1995-2004. MEDLINE Abstract

18. Funderud,S., Andreassen,A. and Haugli,F. (1978) Cell, 15, 1519-1527. MEDLINE Abstract

19. Pierron,G. and Bénard,M. (1996) In DePamphilis,M. (ed.), DNA Replication in Eukaryotic Cells. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, pp. 933-946.

20. Bénard,M. and Pierron,G. (1992) Nucleic Acids Res., 20, 3309-3315.

21. Bénard,M., Lagnel,C., Pallotta,D. and Pierron,G. (1996) Mol. Cell. Biol., 16, 1968-1976.

22. Wilhelm,M.L. and Wilhelm,F.X. (1989) J. Mol. Evol., 28, 322-326. MEDLINE Abstract

23. Wilhelm,M.L., Toublan,B., Jalouzot,R. and Wilhelm,F.X. (1984) EMBO J., 3, 2659-2662.

24. Carrino,J.J., Kueng,V., Braun,R. and Laffler,T.G. (1987) Mol. Cell. Biol., 7, 1933-1937. MEDLINE Abstract

25. Jalouzot,R., Toublan,B., Wilhelm,M.L. and Wilhelm,F.X. (1985) Proc. Natl Acad. Sci. USA, 82, 6475-6479. MEDLINE Abstract

26. Wilhelm,M.L, Demouveau,M.F. and Wilhelm,F.X. (1986) Curr. Genet., 10, 459-462.

27. Natale,D.A., Schubert,A.E. and Kowalski,D. (1992) Proc. Natl Acad. Sci. USA, 89, 2654-2658. MEDLINE Abstract

28. Brewer,B.J. and Fangman,W.L. (1988) Cell, 55, 637-643. MEDLINE Abstract

29. Linskens,M.H.K. and Huberman,J.A. (1988) Mol. Cell. Biol., 8, 4927-4935. MEDLINE Abstract

30. Shinomiya,T. and Ina,S. (1993) Mol. Cell. Biol., 13, 4098-4106. MEDLINE Abstract

31. McAlpine,D.M., Zhang,Z. and Kapler,G.M. (1997) Mol. Cell. Biol., 17, 4517-4525. MEDLINE Abstract

32. Bénard,M., Lagnel,C. and Pierron,G. (1995) Nucleic Acids Res., 23, 1447-1453. MEDLINE Abstract

33. Deshpande,A.M. and Newlon,C.S. (1996) Science, 272, 1030-1033. MEDLINE Abstract

34. Borde,V. and Duguet,M. (1998) Nucleic Acids Res., 26, 2042-2049. MEDLINE Abstract

35. Greenfeder,S.A. and Newlon,C.S. (1992) Mol. Cell. Biol., 12, 4056-4066. MEDLINE Abstract

36. Bouton,A.H. and Smith,M.M. (1986) Mol. Cell. Biol., 6, 2354-2363. MEDLINE Abstract

37. Heck,M.M.S. and Spradling,A.C. (1990) J. Cell Biol., 4, 903-914.

38. Liang,C., Spitzer,J.D., Smith,H.S. and Gerbi,S.A. (1993) Genes Dev., 7, 1072-1084. MEDLINE Abstract

39. Vassilev,L.T. and Johnson,E.M. (1990) Mol. Cell. Biol., 10, 4899-4904. MEDLINE Abstract

40. Kitsberg,D., Selig,S., Keshet,I. and Cedar,H. (1993) Nature, 366, 588-590. MEDLINE Abstract

41. DePamphilis,M.L. (1993) Trends Cell Biol., 3, 161-167.

42. Bielinski,A.K. and Gerbi,S.A. (1998) Science, 279, 95-98. MEDLINE Abstract

43. Pierron,G., Pallotta,D. and Bénard,M. (1999) Mol. Cell. Biol., 19, in press. MEDLINE Abstract

44. Friedman,K.L., Brewer,B.J. and Fangman,W.J. (1997) Genes Cell, 2, 667-678.

45. Pierron,G., Bénard,M., Puvion,E., Flanagan,R., Sauer,H.W. and Pallotta,D. (1989) Nucleic Acids Res., 17, 553-566. MEDLINE Abstract

*To whom correspondence should be addressed. Tel: +33 1 49 58 33 73; Fax: +33 1 49 58 33 81; Email: benard@infobiogen.fr

This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: jnl.info{at}oup.co.uk
Last modification: 27 Apr 1999
Copyright©Oxford University Press, 1999.
Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Nucleic Acids ResHome page
M. Benard, C. Maric, and G. Pierron
Low rate of replication fork progression lengthens the replication timing of a locus containing an early firing origin
Nucleic Acids Res., September 27, 2007; 35(17): 5763 - 5774.
[Abstract] [Full Text] [PDF]

Home page
 Nucleic Acids ResHome page
C. Maric, E. Swanston, J. Bailey, and G. Pierron
Replicational organization of three weakly expressed loci in Physarum polycephalum
Nucleic Acids Res., June 1, 2002; 30(11): 2261 - 2269.
[Abstract] [Full Text] [PDF]

Home page
 Mol. Cell. Biol.Home page
P.-H. Chen, W.-B. Tseng, Y. Chu, and M.-T. Hsu
Interference of the Simian Virus 40 Origin of Replication by the Cytomegalovirus Immediate Early Gene Enhancer: Evidence for Competition of Active Regulatory Chromatin Conformation in a Single Domain
Mol. Cell. Biol., June 1, 2000; 20(11): 4062 - 4074.
[Abstract] [Full Text]

This Article
Right arrow Abstract Freely available
Right arrow Print PDF (213K) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (4)
Right arrowRequest Permissions
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Benard, M.
Right arrow Articles by Pierron, G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Benard, M.
Right arrow Articles by Pierron, G.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?