Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2010 Sep 2;6(9):e1001092.
doi: 10.1371/journal.pgen.1001092.

Diversity of Eukaryotic DNA Replication Origins Revealed by Genome-Wide Analysis of Chromatin Structure

Free PMC article

Diversity of Eukaryotic DNA Replication Origins Revealed by Genome-Wide Analysis of Chromatin Structure

Nicolas M Berbenetz et al. PLoS Genet. .
Free PMC article


Eukaryotic DNA replication origins differ both in their efficiency and in the characteristic time during S phase when they become active. The biological basis for these differences remains unknown, but they could be a consequence of chromatin structure. The availability of genome-wide maps of nucleosome positions has led to an explosion of information about how nucleosomes are assembled at transcription start sites, but no similar maps exist for DNA replication origins. Here we combine high-resolution genome-wide nucleosome maps with comprehensive annotations of DNA replication origins to identify patterns of nucleosome occupancy at eukaryotic replication origins. On average, replication origins contain a nucleosome depleted region centered next to the ACS element, flanked on both sides by arrays of well-positioned nucleosomes. Our analysis identified DNA sequence properties that correlate with nucleosome occupancy at replication origins genome-wide and that are correlated with the nucleosome-depleted region. Clustering analysis of all annotated replication origins revealed a surprising diversity of nucleosome occupancy patterns. We provide evidence that the origin recognition complex, which binds to the origin, acts as a barrier element to position and phase nucleosomes on both sides of the origin. Finally, analysis of chromatin reconstituted in vitro reveals that origins are inherently nucleosome depleted. Together our data provide a comprehensive, genome-wide view of chromatin structure at replication origins and suggest a model of nucleosome positioning at replication origins in which the underlying sequence occludes nucleosomes to permit binding of the origin recognition complex, which then (likely in concert with nucleosome modifiers and remodelers) positions nucleosomes adjacent to the origin to promote replication origin function.

Conflict of interest statement

The authors have declared that no competing interests exist.


Figure 1
Figure 1. Average views of nucleosome occupancy at replication origins and transcription start sites.
(A) Nucleosome maps at 222 replication origins were aligned by the ACS and oriented by the T-rich strand. The average is shown in black, overlaid on a bivariate histogram in which color indicates the density of the data at each point. (B) Nucleosome maps at 222 randomly-selected promoters, aligned by the transcription start site. (C) The average NDR widths for the ACS (N = 222) and the random subset of TSSs (N = 222). Distances in bp between nucleosome midpoints are indicated for the ACSs (red) and the TSSs (blue).
Figure 2
Figure 2. Correlation of nucleosome position with DNA sequence features.
(A) GC content (red line) and nucleosome occupancy are plotted (blue line), with the Pearson correlation indicated. (B) Average dinucleotide profile is plotted for each of the six groups of dinucleotide properties (I–VI). The average nucleosome occupancy is also plotted for comparison (blue line). The average DNA dinucleotide profiles were partitioned into 6 groups using k-means clustering. The subcluster average DNA dinucleotide profile, re-scaled to a range of +1 to −1, is shown for each. The number of dinucleotide properties in each group (N) is indicated along with the Pearson correlation of each group with the average ACS profile.
Figure 3
Figure 3. The diversity of nucleosome occupancy patterns at replication origins.
(A) Heatmap of k-means clustered replication origins. ARSs are aligned on the Y axis, and the distance from the ACS is indicated on the X axis. Colors correspond to the log2 value of data points at a given position: in general, nucleosome occupancy is indicated in yellow and nucleosome depletion is indicated in blue/green. (B) Cluster averages for each of the four groups from the k-means clustering are plotted. Within each subcluster average plot the LOESS-smoothed moving sum of gene ends or TSSs located within 25 probe windows are plotted in blue (TSSs) or red (gene ends). The number of origins (N), TSS elements (T), and gene ends (G) in each group is indicated.
Figure 4
Figure 4. Representative nucleosome profiles and nucleosome calls for 4 origins.
(A) A nucleosome profile (ARS VII-112) similar to the average ACS profile. (B) A nucleosome profile (ARS II-170) that contains a second NDR to the left of ACS-proximal NDR with a single nucleosome gap. (C) A nucleosome profile (ARS IV-1166) that contains a two nucleosome gap between two NDRs. (D) A nucleosome profile (ARS X-737) that lacks an NDR at the ACS. Dotted lines indicate the positions of nucleosome midpoints.
Figure 5
Figure 5. The relationship among replication timing and the locations of ACS-proximal TSSs and gene ends, and NDR widths.
Distribution of 107 TSS locations (A,B) or 152 gene end locations (C,D) within 800bp of the ACS is plotted with early origin ratio (A,C) or replication time (B,D). TSS or gene end positions were counted within a moving window of 25 probes. Each position within the TSS or gene end distribution corresponds to the midpoint of the moving window and includes the total number of gene end or TSS locations counted within that window. The TSS or gene end distributions were LOESS-smoothed. Replication time was determined by identifying origins that contain a TSS or gene end within each 25 probe window and calculating the proportion of early origins (mean early origin ratio, (A,C)) or replication time (mean Trep, panel B and D) . Each position within the replication timing distribution corresponds to the midpoint of the 25 probe window and includes all of the origins which contained either a TSS or gene end within these probes. Windows with fewer than 5 origins are shaded grey so that effects from small numbers of origins are not considered. (E) NDR widths were divided into 7 quantiles and the proportion (μ early) of early origins was compared to 10,000 similar sized samples of the original 222 origins in order to determine how many groups contained a less extreme proportion of early origins. P-values indicate the % of randomly re-sampled groups that had the same proportion of early origins. Extreme values, either close to 0 or close to 100, indicate the early origin proportion is at the low end (late firing) or the high end (early firing) of the distribution.
Figure 6
Figure 6. The effect of ORC depletion on nucleosome occupancy.
(A) Nucleosome occupancy map of the wild-type control strain, plotted as in Figure 1. (B) Nucleosome occupancy map following ORC depletion in the GAL:orc2-1 strain. (C) Average nucleosome occupancy plotted for wild-type (blue), GAL:orc2-1 (red), and a difference plot comparing nucleosomal DNA from GAL:orc2-1 to that from the wild-type strain (green). (D) NDR width distributions were calculated using a moving sum for windows containing 9 probes (36 bp) and LOESS-smoothed. The distribution in wild-type (blue line) and following Orc2 depletion (red line) is shown.
Figure 7
Figure 7. Nucleosome occupancy at replication origins in chromatin assembled in vitro.
(A) Nucleosome maps at 174 replication origins in the in vitro nucleosome dataset were aligned by the ACS and oriented by the T-rich strand. The average is shown in black, overlayed on a bivariate histogram in which color indicates the density of the data at each point. (B) Model of stepwise establishment of nucleosome positioning at replication origins. The DNA sequence surrounding the ACS specifies a low nucleosome occupancy, creating a permissive environment for ORC binding. Upon binding by ORC and recruitment of chromatin remodeling and modification activities the +1 and −1 nucleosomes are positioned. The adjacent nucleosomes then pack in uniformly-spaced arrays.

Similar articles

See all similar articles

Cited by 69 articles

See all "Cited by" articles


    1. Marahrens Y, Stillman B. A yeast chromosomal origin of DNA replication defined by multiple functional elements. Science. 1992;255:817–823. - PubMed
    1. Bell SP, Stillman B. ATP-dependent recognition of eukaryotic origins of DNA replication by a multiprotein complex. Nature. 1992;357:128–134. - PubMed
    1. Raghuraman MK, Winzeler EA, Collingwood D, Hunt S, Wodicka L, et al. Replication dynamics of the yeast genome. Science. 2001;294:115–121. - PubMed
    1. Yabuki N, Terashima H, Kitada K. Mapping of early firing origins on a replication profile of budding yeast. Genes Cells. 2002;7:781–789. - PubMed
    1. MacAlpine DM, Bell SP. A genomic view of eukaryotic DNA replication. Chromosome Res. 2005;13:309–326. - PubMed

Publication types

MeSH terms

LinkOut - more resources