• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of molcellbPermissionsJournals.ASM.orgJournalMCB ArticleJournal InfoAuthorsReviewers
Mol Cell Biol. Sep 2008; 28(17): 5265–5274.
Published online Jun 23, 2008. doi:  10.1128/MCB.00181-08
PMCID: PMC2519731

In Xenopus Egg Extracts, DNA Replication Initiates Preferentially at or near Asymmetric AT Sequences[down-pointing small open triangle]


Previous observations led to the conclusion that in Xenopus eggs and during early development, DNA replication initiates at regular intervals but with no apparent sequence specificity. Conversely, here, we present evidence for site-specific DNA replication origins in Xenopus egg extracts. Using λ DNA, we show that DNA replication origins are activated in clusters in regions that contain closely spaced adenine or thymine asymmetric tracks used as preferential initiation sites. In agreement with these data, AT-rich asymmetric sequences added as competitors preferentially recruit origin recognition complexes and inhibit sperm chromatin replication by increasing interorigin spacing. We also show that the assembly of a transcription complex favors origin activity at the corresponding site without necessarily eliminating the other origins. Thus, although Xenopus eggs have the ability to replicate any kind of DNA, AT-rich domains or transcription factors favor the selection of DNA replication origins without increasing the overall efficiency of DNA synthesis. These results suggest that asymmetric AT-rich regions might be default elements that favor the selection of a DNA replication origin in a transcriptionally silent complex, whereas other epigenetic elements linked to the organization of domains for transcription may have further evolved over this basal layer of regulation.

Eukaryotic DNA replication begins with the assembly of a prereplication complex at DNA replication origins that are distributed throughout the genome. However, a genetically required consensus sequence has been identified only in Saccharomyces cerevisiae replication origins (26). The most striking feature outside this consensus sequence is an A-rich region that encompasses the “DNA-unwinding element,” where the first RNA primers are synthesized (5). In the fission yeast Schizosaccharomyces pombe, replication origins consist of two or more genetically required AT-rich asymmetric sequences, but a consensus has not yet been identified (38).

Replication origins in metazoans rely on internal DNA sequences, but sequence specificity appears to be less rigid (2, 29). In the rapidly cleaving frog or fly embryos, the initiation of DNA replication can occur at any sequence, but later in development, initiation events are restricted to specific sites (15, 37). Differences in origin usage may result from epigenetic parameters such as nucleotide pool levels (3), transcription factor binding sites (10), ratio of initiation proteins to DNA (21), gene transcription (22), chromosome structure (1), DNA topology (36), or DNA methylation (12, 35). The complexity of mammalian replication origins may reflect the fact that DNA contains many potential origins, which can all be activated in embryos undergoing rapid cell cleavage but, as development progresses, become regulated as both genetic and epigenetic parameters conspire to repress initiation at some of these sites while activating it at others (11, 29).

Because of its well-known sequence and large size, λ DNA is a convenient template for analyzing the initiation of DNA replication in Xenopus egg extracts. It assembles chromatin, forms nuclei, and replicates in a semiconservative fashion only once per cell cycle (9, 27, 28, 32). Using this defined template, we provide evidence that in Xenopus laevis, the initiation of DNA replication does not occur randomly but occurs preferentially at asymmetric AT-rich islands or at chromatin domains assembled for transcription. Interestingly, such elements provide site specificity without increasing the overall efficiency of replication of the template DNA.


Preparation of Xenopus interphasic egg extracts (low-speed extracts) and quantification of replicated DNA.

Xenopus low-speed egg extracts were prepared as previously described (30; www.igh.cnrs.fr/equip/mechali/). Upon thawing, egg extracts were supplemented with cycloheximide (250 μg/ml) and an energy regeneration system (10 μg/ml creatine kinase, 10 mM creatine phosphate, 1 mM ATP, 1 mM MgCl2). Replication of sperm chromatin in the presence of DNA competitors was performed with bromo-dUTP (BrdUTP) at a final concentration of 40 μM and aphidicolin (final concentration of 3 μg/ml) to slow down elongation of replication forks. Salmon sperm DNA was sheared to an average length of 1 kbp (Invitrogen), and poly(dA-dT) × poly(dA-dT) and poly(dA) × poly(dT) were obtained from Amersham Biosciences.

Purification of nascent-strand DNA and real-time PCR amplification.

λ DNA was incubated in low-speed interphasic egg extracts supplemented with cycloheximide (250 μg/ml) and the energy regeneration system. The final concentration of DNA ranged from between 2 ng and 10 ng per μl of egg extract. After 90 to 120 min of incubation at 22°C, replication of λ DNA was stopped with a solution containing 20 mM Tris (pH 7.9), 0.5% Triton X-100, 30 mM EDTA, 1% sarkosyl, and 600 ng/ml proteinase K. The reaction mixture was incubated for 3 h at 45°C, and DNA was then extracted with phenol-chloroform and precipitated with ethanol. Purified DNA was gently resuspended in TEN20 (10 mM Tris [pH 7.9], 1 mM EDTA [pH 8.0], and 20 mM NaCl), heat denatured (5 min at 96°C), loaded onto 5 to 20% sucrose gradients made in TEN20, and centrifuged in a SW50 rotor (45,000 rpm at 4°C for 7 h). All steps were done under RNase-free conditions. 32P-labeled molecular weight DNA markers were processed in parallel. Twenty-six fractions of 200 μl were collected from the top of the sucrose gradient, and the size of DNA in each fraction was estimated by agarose gel electrophoresis. Sucrose gradient fractions in the range of 600 to 1,500 bp DNA were treated with T4 polynucleotide kinase and then with 5′-to-3′ λ-exonuclease (Invitrogen), which degraded DNA but not RNA-primed DNA. RNA was removed from the reaction mixtures just before the quantification step.

Real-time PCR amplification was performed with a MyIQ real-time PCR apparatus (Bio-Rad), using the iQ Sybr green supermix (Bio-Rad), or with a LightCycler (Roche). Primers used for quantification are presented in Table Table11.

Primers used for quantification

The following PCR program was used: 5 min at 95°C and then 40 cycles with 30 s at 95°C, 30 s at 58°C and 60°C or 62°C, and 30 s at 72°C, including the melting curve analysis at the end of the run.

Statistical analysis of data.

The statistical significance of all PCR quantification data was estimated with the Excel program using TTEST. All data were obtained as copy numbers of nascent-strand molecules (usually 105 to 106) detected in the PCRs with each pair of primers using a standard curve made from wild-type (wt) λ molecules. All fractions (usually four to five fractions) containing nascent strands of 0.6 to 1.5 kbp were measured separately, and the average abundance of nascent strands for each pair of primers (one measurement) was then calculated. At least four measurements of nascent-strand quantity for every primer pair were done, and the average abundance was then calculated for each independent experiment. The results presented here are the average results from three independent experiments, with error bars presenting variations from minimum to maximum or only maximum variations. The results were normalized by considering the lowest amount of nascent strands to be equal to 1 (region B). The normalized values for each pair of primers in three independent experiments were estimated by TTEST for statistical significance (P value).

Statistical analysis of DNA combing data was done using TTEST or one-way analysis of variance column statistics (GraphPad Prism 5 software). The measurements of interorigin distance from three independent experiments are presented.

Primer extension.

RNA extraction and primer extension analyses were performed as described previously (10). We used the T7 primer (5′-GTAATACGACTCACTATAGGGC-3′) from the pBS/SK− region for the extension of RNA. Samples were analyzed by resolving them on a 6% denaturing polyacrylamide gel.

Chromatin immunoprecipitation (ChIP).

The DNA fragment containing the TATA box of the Xenopus c-myc gene adjacent to five binding sites for the transcription factor Gal4-VP16 was inserted into the λzapII vector (Stratagene) derived from wt λ DNA. This vector was preincubated with recombinant Gal4-VP16 and TATA binding protein (TBP) as described previously (10). Briefly, 500 ng of the constructed transcription-inducible vector was incubated with or without recombinant transcription factors in a molar ratio of 1:20 per binding site in 6 μl of binding buffer (10 mM HEPES [pH 7.7], 0.5 mM MgCl2, 30 mM KCl, 0.1 mg/ml bovine serum albumin [BSA], 20 μM ZnCl2, 2 mM dithiothreitol, and 5% glycerol). Reaction mixtures were incubated at 30°C for 30 min and then added to 60 μl of Xenopus low-speed egg extracts supplemented with cycloheximide (250 μg/ml) and the energy regeneration system. After an additional incubation at 22°C for 60 or 90 min, reaction mixtures were cross-linked for 5 min at 22°C with freshly diluted formaldehyde (Merck) at a 0.5% final concentration in XB buffer (20 mM HEPES [pH 7.9], 50 mM KCl, 2.5 mM MgCl2, 250 mM sucrose, 0.2 mM ATP, 0.1% Triton X-100, and protease inhibitors). λ DNA was then centrifuged for 10 min at 25,000 × g, and pellets were washed with XB buffer, dissolved in TEN buffer (10 mM Tris [pH 8.0], 150 mM NaCl, 1 mM EDTA) with 0.5% sodium dodecyl sulfate (SDS), and fragmented by sonication to give a DNA length range of between 300 and 1,500 bp.

Typical ChIP reaction mixtures contained 1/10 of cross-linked material in 100 μl of radioimmunoprecipitation assay (RIPA) buffer (50 mM Tris-HCl [pH 8.0], 150 mM NaCl, 0.5% NP-40, 0.5% Na-deoxycholate, 0.1% SDS, 1 mM EDTA, 0.2% BSA) and the Complete cocktail of protease inhibitors (Roche). We used anti-acetyl-H3 (Abcam), anti-H3 (Abcam), and anti-Gal4 antibodies (described in reference 10). Incubation with antibodies was done overnight at 4°C, 12 μl of protein A-Sepharose (Amersham) equilibrated in RIPA buffer was then added, and samples were incubated for an additional 2 h at 20°C. Pellets were washed three times for 15 min in RIPA buffer, once in RIPA-Li buffer (10 mM Tris-HCl [pH 8.0], 250 mM LiCl, 0.5% NP-40, 0.5% Na-deoxycholate, 1 mM EDTA), and, finally, three times for 5 min in TEN buffer. Immunoprecipitated DNA was incubated for 4 h at 65°C in a solution containing 10 mM Tris (pH 8.0), 1% SDS, and 1 mM EDTA and then incubated for 1 h at 45°C with proteinase K. After phenol-chloroform extraction, DNA was precipitated, resuspended in Tris-EDTA buffer, and used for quantitative real-time PCR amplification. As controls for input, samples of cross-linked material were processed as the DNA from immunoprecipitated samples. Samples processed without the addition of antibodies were used as “mock” immunoprecipitations. DNA analyses were performed by real-time PCR as described above for the quantification of nascent strands using the same primer pairs. The output signals obtained after PCR amplification with each primer set were presented as the percentage of the input DNA after the subtraction of the value obtained for “mock.” The statistical significance of all PCR quantification data was estimated with the Excel program using TTEST.

DNA combing.

Sperm nuclei or λ DNA embedded in agarose plugs was stained with YOYO-1 (Molecular Probes) and resuspended in 50 mM MES (morpholineethanesulfonic acid) (pH 5.7) after digestion of the plugs with β-agarase (Biolabs). DNA combing was performed as described previously (19, 31). Combed DNA fibers were denatured for 20 min with 1 N NaOH, and bromodeoxyuridine (BrdU) incorporation was detected with a rat monoclonal antibody (Sera Lab) and a secondary antibody coupled to Alexa 488 (Molecular Probes). DNA molecules were counterstained with an anti-deoxythymidine mouse monoclonal antibody (clone MAB3034; Chemicon) and an anti-mouse immunoglobulin G coupled to Alexa 546 (Molecular Probes). Center-to-center distances between BrdU tracks were measured with MetaMorph (Universal Imaging Corp.) using adenovirus DNA molecules as size standards (1 pixel equals 340 bp). DNA combing images are presented on figures after removal of the background in order to better distinguish the fluorescent signals.

DNA combing combined with fluorescent in situ hybridization (FISH).

Two DNA probes were generated by PCR amplification using the pairs of primers shown in Table Table22.

Probes used for DNA combing combined with FISH

These probes were biotinylated by random priming (Prime-a-Gene labeling system; Promega) and hybridized to combed λ DNA previously denatured in a 1 N NaOH solution. Hybridization was carried out as previously described (34). Briefly, hybridization was done overnight in a humid chamber at 37°C with 25 ng/μl of each probe in a solution containing 2× SSC (1× SSC is 0.15 M NaCl plus 0.015 M sodium citrate), 50% formamide, 10% dextran sulfate, 50 mM Na-phosphate buffer (pH 7.0), and 1% Tween 20. Slides were washed sequentially with 2× SSC, 50% formamide, and then 2× SSC, followed by blocking in a solution containing 1× phosphate-buffered saline (pH 7.4), 1% Triton, and 1% BSA for 30 min. Immunodetection was done by sequential incubations with antibodies separated by phosphate-buffered saline washings, as follows: (i) rat anti-BrdU antibody (Sera Lab); (ii) Texas Red-conjugated NeutrAvidin (Molecular Probes), rat anti-BrdU antibody (Sera Lab), and mouse antideoxythymidine antibody (clone MAB3034; Chemicon); (iii) biotin-conjugated goat antiavidin (Sera Lab), Alexa Fluor 488-goat anti-rat antibody (Molecular Probes), and Cy5-conjugated anti-mouse antibody (Interchim); (iv) Texas Red-conjugated NeutrAvidin (Molecular Probes); (v) biotin-conjugated goat antiavidin (Sera Lab); and (vi) Texas Red-conjugated NeutrAvidin (Molecular Probes).


In Xenopus egg extracts, λ DNA replication initiates at clustered origins, which present an average spacing of 23 kbp.

Bacteriophage λ DNA replicates in Xenopus eggs and egg extracts at a slower rate than sperm chromatin and in a reaction that can take up to 10 h (6, 9). We found that only 4 to 5% of the input λ DNA was replicated after 150 min of incubation in egg extracts (Fig. (Fig.1A,1A, left), while the replication of sperm chromatin was completely finished at this time point (Fig. (Fig.1A,1A, right), in agreement with previous observations (6, 32). The explanation given for this relatively low efficiency was that sperm chromatin forms fully functional nuclei much more rapidly than λ DNA in these extracts (9, 32). However, a less efficient firing of DNA replication origins would also result in a lower efficiency of DNA replication.

FIG. 1.
Clusters of origins on replicating λ DNA in Xenopus egg extracts. (A) Replication kinetics of wt λDNA (left) and sperm chromatin (right) in Xenopus egg extracts. (B) Representative images of combed λ DNA after 120 min of incubation ...

To address this question, we analyzed the interorigin distance using dynamic molecular combing (31). BrdUTP was added to extracts with λ DNA, and samples were then collected at regular time points to detect replicating molecules. DNA was purified and uniformly stretched onto silanized glass slides, and BrdU incorporation (green) was detected on the DNA fibers (in red) (see Materials and Methods). λ DNA molecules formed stable concatemers in egg extracts (Fig. (Fig.1B),1B), as previously reported after microinjection in Xenopus eggs (27). In this experiment, we were not able to detect the significant incorporation of BrdU (i.e., tracks longer than 3 kbp) earlier than 2 h after the beginning of incubation. We found that a minority of λ molecules initiated DNA replication, but within this population, initiation was rather synchronous. Analysis of 2,800 combed molecules (for a total DNA length of 138 Mb) showed that replicating molecules exhibited defined BrdU tracks similar to those presented in Fig. Fig.1B,1B, while less than 0.07% molecules were entirely replicated after 2 h. We also observed that after 3 h of incubation, a new population of λ DNA molecules entered into the replication pool, with a pattern similar to the one presented in Fig. Fig.1B.1B. Finally, the analysis of the interorigin spacing showed a mean replicon size of 22.8 kbp (Fig. (Fig.1C),1C), similar to what was observed in sperm chromatin (19). Using TTEST (see Materials and Methods), we compared the measurements of the interorigin distance of λ DNA and sperm chromatin and obtained a P value of 0.2560 (see Fig. S1 in the supplemental material), confirming that there is no significant difference between these two interorigin distances. Comparison of sperm chromatin replication with the less efficient λ DNA replication allowed us to conclude, first, that the interorigin spacing is similar in both templates and, second, that the activation of origins is happening almost simultaneously within clusters when DNA replication starts on these molecules. This result suggests that the limiting factor controlling the initiation of DNA replication is the selection of the DNA molecule as a template rather than the selection of the origins inside the template.

DNA replication initiates preferentially at A or T asymmetric islands on λ DNA.

In Xenopus eggs and early embryos, DNA replication occurs at regular intervals but with no apparent sequence specificity (6, 7, 14, 18, 28, 32), and specific initiation sites were not observed in small plasmid molecules (10, 13, 25). However, small plasmid DNA uses only one origin per molecule (13, 23, 25), whereas DNA replication occurs at 10- to 25-kbp intervals on sperm chromatin (7, 19, 42). We asked whether preferential initiation sites could be detected on longer DNA molecules such as λ DNA, where multiple origins are activated (Fig. (Fig.1).1). The localization of DNA replication origins was investigated by quantification of short nascent DNA strands. RNA-primed nascent strands (0.6 to 1.5 kbp) were purified and quantified by real-time PCR amplification (see Materials and Methods). Unexpectedly, we observed significant variations in the quantities of nascent strands from different regions of λ DNA (Fig. (Fig.2A;2A; also see Table S1 in the supplemental material for statistical analysis). The two extreme cases were reproducibly regions E and B. Indeed, region E was nine times more efficiently used as an origin than region B, clearly showing the presence of preferred sites of initiation of DNA replication. The primary sequence of λ DNA has an average AT content of 50%, but the left arm of λ DNA is less AT rich (43.5%) than the right arm (56%). Region E is within a 3.5-kbp sequence, which contains 57% of AT, but more importantly, it encompasses four very prominent poly(dA) or poly(dT) asymmetric islands, with mostly A on one strand and mostly T on the other and where the A or T content exceeds 70% (Fig. (Fig.2B).2B). Interestingly, this region also encompasses the λ bacteriophage origin that is used during its replication in Escherichia coli cells. In contrast, region B, where the lowest quantity of nascent strands was detected, is within a 1-kbp sequence with only 42% AT content and without any prominent AT-rich asymmetric sequences. Another region enriched for nascent strands is region A that contains a very prominent asymmetric island with 65% poly(dT).

FIG. 2.
DNA replication initiates preferentially at AT-rich asymmetric sequences. (A) Quantification of short nascent strands by real-time PCR amplification. Average measurements from three independent experiments with error bars are presented after normalization ...

In order to further confirm this observation, we analyzed the nascent-strand abundance at regions containing over 60% poly(dA) or poly(dT) asymmetric sequences. Primer pairs at these positions (C1, C2, C3, E1, E2, E3, E4, and F1) detected much higher levels of nascent strands than primer B, B1, or B2 (Fig. (Fig.2B).2B). The P values (TTEST) comparing region B, B1, or B2 with the other regions confirmed these results. In addition, primers B3 and B4, which are closer to regions A and C, respectively, exhibit a slightly increased efficiency compared to those of B1 and B2. This is in agreement with their position close to asymmetric sequences, with 40 to 50% asymmetric poly(dT) × poly(dA). As we were analyzing short nascent DNA of 0.6 to 1.5 kbp, we also expected that moving the primers closer to region A or C will result in an increased enrichment of nascent-strand DNA. We concluded that in λ DNA molecules, the selection of DNA replication origins is not entirely random and sequence independent and that poly(dA) or poly(dT) asymmetric regions are favored as initiation efficiency is proportionally increasing with the percentage of A or T.

Single-molecule analysis of λ DNA replication confirms preferential initiation sites.

We employed another approach to confirm the uneven use of DNA replication origins along λ DNA by combining DNA combing with FISH. Initiation sites on λ molecules were labeled with BrdU during incubation in egg extracts, and DNA was then purified and stretched on silanized glass slides (19, 31). Hybridization of the stretched λ molecules with two λ DNA-specific biotinylated probes permitted the correct orientation of the molecules. A simplified scheme of this experimental procedure is presented in Fig. Fig.3A.3A. The left panel of Fig. Fig.3B3B shows the combed and aligned λ DNA molecules (blue) after orientation using the two FISH probes (red), whereas the right panel shows only the BrdU signals (green) of the neosynthesized DNA. Due to the length of BrdU tracks needed to score them, the localization of origins by the DNA combing method is less precise than the nascent-strand quantification, but initiation regions can be mapped. A total of 94 DNA molecules were randomly chosen for the analysis. Each λ DNA molecule (48.5 kbp) was divided into five regions of equal size (vertical lines), and the center of each BrdU stretch was considered to be the initiation site of a replication origin. More than 52% of the analyzed λ molecules had origin activity in region 4, while the lowest activity was in region 2 (only 18%) (Fig. (Fig.3C),3C), confirming our previous observations. Interestingly, the preferential incorporation of BrdU also correlated with the number of poly(dA) or poly(dT) asymmetric sequences within these five regions (Fig. (Fig.3D).3D). Thus, region 4, which showed the highest level of incorporation of BrdU, had five asymmetric sequences with over 60% A or T, and four of them had an A or T content equal to or higher than 70%. In contrast, region 2, which showed the lowest origin activity, was very poor in A or T asymmetric sequences. Five examples of asymmetric sequences, in which the A or T content exceeds 70%, and their positions on λ DNA are presented in Table Table33.

FIG. 3.
Single-molecule analysis of DNA replication initiation on wt λ DNA by molecular combing combined with FISH. (A) Simplified scheme of the experimental procedure. wt λ DNA was incubated in Xenopus interphasic egg extracts supplemented with ...
DNA sequences of five preferred regions for initiation of DNA replication (A or T content equal to or over 70%) on wt λ DNA replicating in Xenopus egg extracts with their positionsa

Altogether, these data confirm that although the initiation of DNA replication can occur at many sites, regions containing AT-rich sequences with clustered adenine or thymine on one strand are preferred sites for λ DNA replication initiation in Xenopus egg extracts.

Single-molecule analysis of interorigin distance on sperm chromatin in the presence of different DNA competitors.

If AT-rich sequences favor the initiation of DNA replication in Xenopus egg extracts, the addition of AT-rich DNA, as a competitor, should inhibit sperm chromatin replication. In order to test this, we monitored the replication of sperm chromatin in Xenopus egg extracts containing different DNA competitors such as poly(dA) × poly(dT), poly(dA-dT) × poly(dA-dT), or sheared salmon sperm DNA. These competitors replicate themselves very poorly in egg extracts, 12% at 3 h for poly(dA) × poly(dT), 6% for poly(dA-dT) × poly(dA-dT), and less than 1% for sheared salmon sperm DNA. The addition of 1 ng/μl of poly(dA) × poly(dT) inhibited the kinetics of replication of sperm chromatin, while the same amount of sheared salmon sperm genomic DNA or poly(dA-dT) × poly(dA-dT) had no influence on sperm chromatin replication (Fig. (Fig.4A).4A). Similar results were obtained using different concentrations of the competitors (see Fig. S2 in the supplemental material). In all cases, poly(dA) × poly(dT) sequences were the strongest competitors. We then analyzed whether the interorigin distance could be affected by poly(dA) × poly(dT) DNA. Sperm nuclei were incubated in egg extracts with BrdUTP to label initiation sites and a low concentration of aphidicolin, which permits initiation but slows down elongation. Chromosomal DNA was purified and combed, and the center-to-center distance between adjacent BrdU tracks was measured. Sperm nuclei had a mean interorigin spacing of 21.9 kbp (Fig. (Fig.4B),4B), in agreement with previously reported values (19) and those reported in Fig. Fig.1.1. This value did not change after the addition of sheared salmon sperm DNA or poly(dA-dT) × poly(dA-dT). In contrast, a significant increase in the interorigin spacing occurred in the presence of poly(dA) × poly(dT) asymmetric sequences. We also observed that in the presence of AT-rich asymmetric DNA, the proportion of replicons above 20 kbp increased 1.7 times, while the percentage of replicons longer than 30 kbp increased more than 1.9 times compared to those in control reactions. Further statistical analysis of these measurements (P value) is presented in Fig. S3 and Table S2 in the supplemental material and confirms our conclusions.

FIG. 4.
Asymmetric AT-rich sequences compete with sperm nucleus DNA replication. (A) Kinetics of sperm (Sp.) nucleus replication in the presence of different DNA competitors. (B) Single-molecule analysis of the interorigin spacing of sperm nuclei in the presence ...

We then asked whether poly(dA) × poly(dT) sequences were favored sequences to initiate DNA replication because they assemble origin recognition complexes (ORCs) more efficiently than bulk DNA. This would explain the competitor effect of such sequences. We therefore performed a titration experiment in which various concentrations of sperm chromatin, λ DNA, or poly(dA) × poly(dT) sequences were incubated in egg extracts, and the corresponding amount of chromatin-bound ORC2 was measured by Western blot analysis. Indeed, ORC2 was titrated more rapidly by poly(dA) × poly(dT) than sperm chromatin (Fig. (Fig.55 and see Fig. S4 in the supplemental material). From this experiment, as well as the previous one, we conclude that poly(dA) × poly(dT) competitors first decrease DNA replication efficiency and then increase interorigin spacing and finally recruit more ORC2 proteins. Although we cannot exclude that other features of poly(dA) × poly(dT) sequences may interfere with the DNA replication of sperm chromatin, these data strongly suggest that asymmetric AT-rich sequences are preferred for DNA replication initiation in this system. This result is also in agreement with previous data showing that Xenopus ORC binds preferentially to asymmetric AT sequences (17).

FIG. 5.
Efficiency of ORC2 binding to chromatin in the presence of DNA competitors. Various concentrations of sperm chromatin (Sp.chrom), wt λ DNA, and poly(dA) × poly(dT) were incubated for 30 min in Xenopus egg extracts. Chromatin was purified ...

Specification of additional DNA replication origins by the preassembly of transcription factors.

Our data show that in contrast to previous reports, the initiation of DNA replication is not random along λ DNA and that some regions are more prone to serve as replication origins than others. We previously reported that a site-specific DNA replication origin could be induced by the assembly of a transcription complex on a small plasmid DNA (10). We therefore asked whether the same transcription complex could favor DNA replication initiation when placed on the λ DNA that already contains preferential sites of DNA replication initiation.

A DNA fragment containing the Xenopus c-myc TATA box adjacent to five binding sites for the transcription factor Gal4-VP16 was inserted into the λzapII vector derived from wt λ DNA (Fig. (Fig.6A).6A). The initiation of DNA replication was assessed by the quantification of short nascent DNA strands (Fig. (Fig.6B).6B). Primer pair P replaced primer pair C of wt λ DNA to quantify nascent strands at the promoter region. The initiation of DNA replication on this construct, without bound transcription factors, was very similar to that of λ DNA (compare Fig. Fig.6B6B with with2A).2A). Analysis of H3 histone acetylation by ChIP showed no correlation with the observed pattern of nascent strands' enrichment (Fig. (Fig.6B),6B), suggesting that the acetylation of histone H3 was not contributing to the selection of DNA replication origins in this system.

FIG. 6.
DNA replication also exhibits favored DNA replication origins on λzapII DNA. (A) Schematic representation of the λzapII+Gal4-TATA construct in which a 352-bp fragment containing the Xenopus c-myc TATA box adjacent to five binding ...

When this vector was preincubated with recombinant Gal4-VP16 and TBPs and then added to Xenopus egg extracts, transcription was induced at the expected promoter site (Fig. (Fig.7A).7A). ChIP experiments showed the presence of Gal4-VP16 at the expected position (Fig. (Fig.7B,7B, top). The binding of TBP and Gal4-VP16 transcription factors to the promoter increased DNA replication origin activity at the corresponding region without affecting the frequency of initiation events at other regions (Fig. (Fig.7B,7B, bottom) or the kinetics of DNA replication (data not shown). Relative to position B, the efficiency of DNA replication was then around 6 and therefore as efficient as in asymmetric AT-rich regions (region E, for example). Our interpretation is that the binding of transcription factors increases the origin activity of this region to a level comparable to that of an asymmetric AT-rich region. As for the stimulation of initiation due to the binding of transcription factors, the absolute enrichment is slightly smaller than the enrichment found with a small circular plasmid (10). However, AT-rich asymmetric islands were poorly represented in that plasmid, and therefore, the presence of an assembled transcription complex could result in a pronounced competitor effect. We also cannot exclude that supercoiling may favor site specificity for small plasmid DNA (43). We conclude that although preferential sites of initiation are already present in the λ genome, the assembly of a transcription domain favors origin selection.

FIG. 7.
Binding of a transcription complex favors new replication origins on λ DNA. (A) The λzapII+Gal4-TATA construct was incubated directly (lane 1) or after preincubation with recombinant TBP and GAL4-VP16 proteins (lane 2) in Xenopus ...


Bacteriophage λ DNA was previously reported to be a useful and defined template for the analysis of replication in Xenopus egg extracts. It reproduces nuclear assembly in discrete steps (32), and DNA replication occurs semiconservatively at discrete replication foci and only once per cell cycle (9, 27) but with low efficiency. We report here that replicating λ DNA and sperm nuclei have comparable mean interorigin distances. This result suggests that the low efficiency of λ DNA replication, compared to sperm nuclei, is not caused by an increased interorigin distance on replicating molecules but rather by selection of the DNA molecules that will start DNA replication. Possibly, only a minority of λ DNA molecules can assemble nuclei, but once they do it, they start DNA replication in a regular manner and with clustered replication origins, as usually observed in Xenopus and in other species (7).

The results presented here also reveal that the initiation of DNA replication in egg extracts is not so random and sequence independent as it was previously thought to be. We show that AT-rich asymmetric sequences (at least 25 bp) are present in regions where DNA replication initiation is favored. In contrast, the least favored sequences for initiation are those with low AT content and without any prominent A- or T-rich asymmetric sequence. A recent work on the human beta globin locus similarly showed that AT-rich sequences are essential for replicator activity and reported a low frequency of initiation in DNA fragments that included short stretches of AT-rich sequences, whereas the inclusion of additional AT-rich stretches increased initiation efficiency (44). On the contrary, in Xenopus, such elements do not confer an overall better efficiency of DNA replication. The efficiency of DNA replication itself is more correlated with the length of the molecule rather than with the primary sequence of DNA (28). However, when clusters of origins are activated, AT-rich asymmetric elements are favored among the others. Our competition experiments show that AT asymmetric DNA sequences inhibit DNA replication more strongly than alternative AT DNA sequences. Moreover, we found a significant correlation between asymmetric AT richness and ORC binding for the competitor DNA as well as the increase in the interorigin spacing. Although we cannot exclude other possible effects of the competitor, our results indicate that at three levels, i.e., efficiency of DNA replication, ORC binding, and interorigin spacing, asymmetric AT-rich regions are favored for the initiation of DNA replication in this system.

Metazoan DNA replication origins do not show a consensus sequence, but AT-rich regions have been identified in most of them. It is not yet clear whether these regions are simply important to facilitate the unwinding of the two DNA strands at the origin or whether they are used in more specific functions during origin recognition. In fission yeast, it was clearly demonstrated that SpOrc4 contains nine AT hook motifs that specifically bind AT-rich sequences (8, 16, 20, 33, 40). S. pombe DNA replication origins are also in AT-rich islands (38). Interestingly, the SpOrc4 protein can compete with Xenopus ORC binding sites (17). In metazoans, such AT hook domains are not present in ORC proteins, but AT-rich sequences are preferentially bound by ORC proteins (4, 24, 39, 41), including Xenopus ORC (17). In agreement, our data also show that poly(dA) × poly(dT) is titrating ORC from the extracts more efficiently than sperm chromatin. Our data reconcile several results obtained in different species and make a clear distinction between the efficiency of DNA replication and the site specificity of DNA replication origins. They suggest that within a region prone to initiate DNA replication, some elements may be favored as origins, although they are not absolutely necessary for initiation in the corresponding domain.

We also observed that in vitro, the assembly of a transcription complex at a specific site on λ DNA favors the specification of a DNA replication origin at that site. Therefore, the assembly of a transcription complex may compete with AT-rich regions as a potential initiation site of DNA replication.

Altogether, our data suggest that in Xenopus eggs, although DNA replication can be initiated at any DNA site, multiple independent features such as structural elements (AT-rich sequences) or the presence of other regulatory factors on DNA molecules can select replication origins. Our results also suggest that the evolution from nonspecific to site-specific initiation might be achieved later in development (or in somatic cells) by the synergy of several factors. AT richness could be a sequence-dependent default element that favors the selection of origins in a transcriptionally silent context, whereas other chromatin features linked to gene expression may restrict DNA replication origins in the corresponding chromatin domains.

Supplementary Material

[Supplemental material]


We thank P. Pasero and E. Ralph for useful discussion and critical reading of the manuscript. We acknowledge the Montpellier DNA Combing Facility for providing the silanized surfaces. We also acknowledge N. Montel and S. Bocquet for technical help.

S.S. is supported by the Fondation pour la Recherche Medicale. This work is supported by the CNRS, the Association pour la Recherche sur le Cancer, the Ligue contre le Cancer, and the National Agency for Research (ANR).


[down-pointing small open triangle]Published ahead of print on 23 June 2008.

Supplemental material for this article may be found at http://mcb.asm.org/.


1. Aggarwal, B. D., and B. R. Calvi. 2004. Chromatin regulates origin activity in Drosophila follicle cells. Nature 430372-376. [PubMed]
2. Aladjem, M. I. 2007. Replication in context: dynamic regulation of DNA replication patterns in metazoans. Nat. Rev. Genet. 8588-600. [PubMed]
3. Anglana, M., F. Apiou, A. Bensimon, and M. Debatisse. 2003. Dynamics of DNA replication in mammalian somatic cells: nucleotide pool modulates origin choice and interorigin spacing. Cell 114385-394. [PubMed]
4. Austin, R. J., T. L. Orr-Weaver, and S. P. Bell. 1999. Drosophila ORC specifically binds to ACE3, an origin of DNA replication control element. Genes Dev. 132639-2649. [PMC free article] [PubMed]
5. Bielinsky, A. K., and S. A. Gerbi. 1998. Discrete start sites for DNA synthesis in the yeast ARS1 origin. Science 27995-98. [PubMed]
6. Blow, J. J., and R. A. Laskey. 1986. Initiation of DNA replication in nuclei and purified DNA by a cell-free extract of Xenopus eggs. Cell 47577-587. [PubMed]
7. Blow, J. J., P. J. Gillespie, D. Francis, and D. A. Jackson. 2001. Replication origins in Xenopus egg extract are 5-15 kilobases apart and are activated in clusters that fire at different times. J. Cell Biol. 15215-25. [PMC free article] [PubMed]
8. Chuang, R. Y., and T. J. Kelly. 1999. The fission yeast homologue of Orc4p binds to replication origin DNA via multiple AT-hooks. Proc. Natl. Acad. Sci. USA 962656-2661. [PMC free article] [PubMed]
9. Cox, L. S., and R. A. Laskey. 1991. DNA replication occurs at discrete sites in pseudonuclei assembled from purified DNA in vitro. Cell 66271-275. [PubMed]
10. Danis, E., K. Brodolin, S. Menut, D. Maiorano, C. Girard-Reydet, and M. Mechali. 2004. Specification of a DNA replication origin by a transcription complex. Nat. Cell Biol. 6721-730. [PubMed]
11. DePamphilis, M. L. 1999. Replication origins in metazoan chromosomes: fact or fiction? Bioessays 215-16. [PubMed]
12. Harvey, K. J., and J. Newport. 2003. CpG methylation of DNA restricts prereplication complex assembly in Xenopus egg extracts. Mol. Cell. Biol. 236769-6779. [PMC free article] [PubMed]
13. Hyrien, O., and M. Mechali. 1992. Plasmid replication in Xenopus eggs and egg extracts: a 2D gel electrophoretic analysis. Nucleic Acids Res. 201463-1469. [PMC free article] [PubMed]
14. Hyrien, O., and M. Mechali. 1993. Chromosomal replication initiates and terminates at random sequences but at regular intervals in the ribosomal DNA of Xenopus early embryos. EMBO J. 124511-4520. [PMC free article] [PubMed]
15. Hyrien, O., C. Maric, and M. Mechali. 1995. Transition in specification of embryonic metazoan DNA replication origins. Science 270994-997. [PubMed]
16. Kong, D., and M. L. DePamphilis. 2001. Site-specific DNA binding of the Schizosaccharomyces pombe origin recognition complex is determined by the Orc4 subunit. Mol. Cell. Biol. 218095-8103. [PMC free article] [PubMed]
17. Kong, D., T. R. Coleman, and M. L. DePamphilis. 2003. Xenopus origin recognition complex (ORC) initiates DNA replication preferentially at sequences targeted by Schizosaccharomyces pombe ORC. EMBO J. 223441-3450. [PMC free article] [PubMed]
18. Laskey, R. A., and R. M. Harland. 1982. Replication origins in the Xenopus egg. Cell 31503. [PubMed]
19. Lemaitre, J. M., E. Danis, P. Pasero, Y. Vassetzky, and M. Mechali. 2005. Mitotic remodeling of the replicon and chromosome structure. Cell 123787-801. [PubMed]
20. Lee, J. K., K. Y. Moon, Y. Jiang, and J. Hurwitz. 2001. Schizosaccharomyces pombe origin recognition complex interacts with multiple AT-rich regions of the replication origin DNA by means of the AT-hook domains of the spOrc4 protein. Proc. Natl. Acad. Sci. USA 9813589-13594. [PMC free article] [PubMed]
21. Li, C. J., J. A. Bogan, D. A. Natale, and M. L. DePamphilis. 2000. Selective activation of pre-replication complexes in vitro at specific sites in mammalian nuclei. J. Cell Sci. 113887-898. [PubMed]
22. Lin, C. M., H. Fu, M. Martinovsky, E. Bouhassira, and M. I. Aladjem. 2003. Dynamic alterations of replication timing in mammalian cells. Curr. Biol. 131019-1028. [PubMed]
23. Lucas, I., M. Chevrier-Miller, J. M. Sogo, and O. Hyrien. 2000. Mechanisms ensuring rapid and complete DNA replication despite random initiation in Xenopus early embryos. J. Mol. Biol. 296769-786. [PubMed]
24. MacAlpine, D. M., H. K. Rodriguez, and S. P. Bell. 2004. Coordination of replication and transcription along a Drosophila chromosome. Genes Dev. 183094-3105. [PMC free article] [PubMed]
25. Mahbubani, H. M., T. Paull, J. K. Elder, and J. J. Blow. 1992. DNA replication initiates at multiple sites on plasmid DNA in Xenopus egg extracts. Nucleic Acids Res. 201457-1462. [PMC free article] [PubMed]
26. Marahrens, Y., and B. Stillman. 1992. A yeast chromosomal origin of DNA replication defined by multiple functional elements. Science 255817-823. [PubMed]
27. Mechali, M., F. Mechali, and R. A. Laskey. 1983. Tumor promoter TPA increases initiation of replication on DNA injected into Xenopus eggs. Cell 3563-69. [PubMed]
28. Mechali, M., and S. Kearsey. 1984. Lack of specific sequence requirement for DNA replication in Xenopus eggs compared with high sequence specificity in yeast. Cell 3855-64. [PubMed]
29. Mechali, M. 2001. DNA replication origins: from sequence specificity to epigentics. Nat. Rev. Genet. 2640-645. [PubMed]
30. Menut, S., J. M. Lemaitre, A. Hair, and M. Méchali. 1999. DNA replication and chromatin assembly using Xenopus egg extracts, p. 196-226. In J. D. Richter (ed.), Advances in molecular biology. Oxford University Press, Oxford, United Kingdom.
31. Michalet, X., R. Ekong, F. Fougerousse, S. Rousseaux, C. Schurra, N. Hornigold, M. van Slegtenhors, S. Wolfe, J. Povey, J. S. Beckmann, et al. 1997. Dynamic molecular combing: stretching the whole human genome for high-resolution studies. Science 2771518-1523. [PubMed]
32. Newport, J. 1987. Nuclear reconstitution in vitro: stages of assembly around protein-free DNA. Cell 48205-217. [PubMed]
33. Okuno, Y., H. Satoh, M. Sekiguchi, and H. Masukata. 1999. Clustered adenine/thymine stretches are essential for function of a fission yeast replication origin. Mol. Cell. Biol. 196699-6709. [PMC free article] [PubMed]
34. Pasero, P., A. Bensimon, and E. Schwob. 2002. Single-molecule analysis reveals clustering and epigenetic regulation of replication origins at the yeast rDNA locus. Genes Dev. 162479-2484. [PMC free article] [PubMed]
35. Rein, T., T. Kobayashi, M. Malott, M. Leffak, and M. DePamphilis. 1999. DNA methylation at mammalian replication origins. J. Biol. Chem. 27425792-25800. [PubMed]
36. Remus, D., E. L. Beall, and M. R. Botchan. 2004. DNA topology, not DNA sequence, is a critical determinant for Drosophila ORC-DNA binding. EMBO J. 23897-907. [PMC free article] [PubMed]
37. Sasaki, T., T. Sawado, M. Yamaguchi, and T. Shinomiya. 1999. Specification of regions of DNA replication initiation during embryogenesis in the 65-kilobase DNApolalpha-dE2F locus of Drosophila melanogaster. Mol. Cell. Biol. 19547-555. [PMC free article] [PubMed]
38. Segurado, M., A. de Luis, and F. Antequera. 2003. Genome-wide distribution of DNA replication origins at A+T-rich islands in Schizosaccharomyces pombe. EMBO Rep. 41048-1053. [PMC free article] [PubMed]
39. Stefanovic, D., S. Stanojcic, A. Vindigni, A. Ochem, and A. Falaschi. 2003. In vitro protein-DNA interactions at the human lamin B2 replication origin. J. Biol. Chem. 27842737-42743. [PubMed]
40. Takahashi, T., and H. Masukata. 2001. Interaction of fission yeast ORC with essential adenine/thymine stretches in replication origins. Genes Cells 6837-849. [PubMed]
41. Vashee, S., C. Cvetic, W. Lu, P. Simancek, T. J. Kelly, and J. C. Walter. 2003. Sequence-independent DNA binding and replication initiation by the human origin recognition complex. Genes Dev. 171894-1908. [PMC free article] [PubMed]
42. Walter, J., and J. W. Newport. 1997. Regulation of replicon size in Xenopus egg extracts. Science 275993-995. [PubMed]
43. Walter, J., and J. W. Newport. 2000. Initiation of eukaryotic DNA replication: origin unwinding and sequential chromatin association of Cdc45, RPA, and DNA polymerase alpha. Mol. Cell 5617-627. [PubMed]
44. Wang, L., C. M. Lin, J. O. Lopreiato, and M. I. Aladjem. 2006. Cooperative sequence modules determine replication initiation sites at the human beta-globin locus. Hum. Mol. Genet. 152613-2622. [PubMed]

Articles from Molecular and Cellular Biology are provided here courtesy of American Society for Microbiology (ASM)
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...