Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2016 Jun 8;7:11748.
doi: 10.1038/ncomms11748.

A Replicator-Specific Binding Protein Essential for Site-Specific Initiation of DNA Replication in Mammalian Cells

Affiliations
Free PMC article

A Replicator-Specific Binding Protein Essential for Site-Specific Initiation of DNA Replication in Mammalian Cells

Ya Zhang et al. Nat Commun. .
Free PMC article

Abstract

Mammalian chromosome replication starts from distinct sites; however, the principles governing initiation site selection are unclear because proteins essential for DNA replication do not exhibit sequence-specific DNA binding. Here we identify a replication-initiation determinant (RepID) protein that binds a subset of replication-initiation sites. A large fraction of RepID-binding sites share a common G-rich motif and exhibit elevated replication initiation. RepID is required for initiation of DNA replication from RepID-bound replication origins, including the origin at the human beta-globin (HBB) locus. At HBB, RepID is involved in an interaction between the replication origin (Rep-P) and the locus control region. RepID-depleted murine embryonic fibroblasts exhibit abnormal replication fork progression and fewer replication-initiation events. These observations are consistent with a model, suggesting that RepID facilitates replication initiation at a distinct group of human replication origins.

Figures

Figure 1
Figure 1. Two distinct DNA–protein interactions at replicator sequences.
(a) Map of the HBB locus (top row), Rep-P (second row) and sequences of the intact (AG WT) and mutant (AG1, AG2 and AG mut) oligonucleotides used in this study. Only one strand is shown. The unshaded nucleotides indicate changes from the AG WT oligo. (b) EMSA analyses were used to measure interactions between proteins from K562 cells and biotin-labelled oligos of AG WT, mutated AG1 and mutated AG2 with sequences shown in a. Two DNA–protein complexes were detected with AG WT oligos, but only one complex was detected for AG1 mutant oligos (lower motility—interaction at the AG2 site) and AG2 mutant oligos (higher motility—interaction with the AG1 site). Arrowheads point to specific activities termed AG1 and AG2 and to free oligonucleotides. (c) Specificity of AG1 complex formation. Biotin-labelled double-stranded AG2-mutated oligonucleotides, which contain an intact AG1 site and a mutated AG2 site, interacted with K562 nuclear protein extracts in the presence and absence of specific competing unlabelled oligonucleotides (AG2) and nonspecific competing unlabelled oligonucleotides (AG mut, which could not participate in either AG1 or AG2 complexes). Increasing concentrations of unlabelled AG2, but not AG mut, competed for the AG1 complex. The molecular ratios of specific competitor and probe were 1:1, 1:10 and 1:100, and unspecific competitor and probe were 1:1 and 1:100. The ‘+' symbol indicates that the reagent was added to the binding reaction, whereas the ‘−' symbol indicates that the reagent was not included.
Figure 2
Figure 2. RepID interacts with replication-initiation sites.
(a) An antibody against RepID supershifted a DNA–protein complex. EMSAs in K562 nuclear protein extracts are shown in the presence and absence of the indicated antibodies. (The additional band above the AG1 complex is not specific.) (b) ChIP-Seq analysis of RepID binding the HBB locus. Top row, a chromosome map. Below the map, First row, nascent strands reads obtained from K562 cells, second row, RepID ChIP-Seq reads obtained from K562 cells aligned to the indicated region (see ‘Methods' for details). (c) ChIP analysis of RepID binding in Simian CV-1 cells harbouring Rep-P variants inserted by site-specific recombination into a constant site. Rep-P WT, unmutated Rep-P; Rep-P AG1, Rep-P carrying the AG1 mutation; Rep-P ΔAG, Rep-P with the entire AG domain deleted. The LacZ primer/probe served as a negative control. All data were normalized versus amplification by the LacZ primer/probe. FRT, the fLP recombinase target (FRT) site; Amp, ampicillin; Hyg, hygromycin; LacZ-Zeocin ORF, the LacZ-Zeocin open reading frame. Statistical significance (—P<0.05) was calculated (t-test) versus Rep-P WT. (d) ChIP of RepID binding in K562 cells at different phases of the cell cycle. Primers and probe: bG59.8, bG61.3 and AG from the human Rep-P, hCollagen (human collagen VI), hLB2Nori (a non-initiating sequence near the LMNB2). Statistical significance (**P<0.01 or *P<0.05) was calculated versus G1. (e) The abundance of HBB sequences in nascent DNA strands from CV-1 cells harbouring Rep-P variants as described in the legend to c. Data were normalized to sequences amplified by the LacZ primer/probe. Statistical significance (****P<0.001 or ***P<0.005) was calculated versus Rep-P WT. (f) Depletion of RepID prevented replication initiation at beta-globin origin, but initiation was restored by re-introducing RepID. The abundance of sequences from the HBB locus was measured in nascent DNA strands isolated from U2OS cells harbouring RepID siRNA or expressing of pCMV-RepID-3 × FLAG. Statistical significance (***P<0.005 or **P<0.01) was calculated as indicated. RepID expression levels were detected using indicated antibodies as shown. Each chart in cf shows results from a representative experiment (n=3).
Figure 3
Figure 3. Genome-wide colocalization of RepID with replication-initiation sites.
(a) A screenshot of a sample genomic region showing replication-initiation profiles (NS-Seq) and protein-binding (ChIP-Seq) data. Top track, below the RepSeq genes: nascent-strand patterns from cells with WT RepID (Replication: WT NS). Middle track: ChIP-Seq patterns in WT cells (RepID ChIP). Lowest track: nascent-strand patterns from cells depleted of RepID (Replication: RepID KO NS). The shaded region delineates a RepID-binding origin adjacent to an origin that does not bind RepID. An expanded screenshot of the same region is shown in Supplementary Fig. 4d. (b) The distribution of replication-initiation events in RepID KO cells (overall, 58,656 NS-Seq peaks) that colocalized with initiation events in WT cells, plotted as a function of the distance from the centre of WT origins (overall, 78,859 NS-Seq peaks). (c) A screenshot of a sample genomic region showing replication-initiation profiles as in a. The shaded region delineates a replication origin that binds RepID in WT cells and does not initiate replication in KO cells. An expanded screenshot of the same region is shown in Supplementary Fig. 4e. (d) The distribution of replication-initiation events in KO cells plotted as a function of the distance from the centre of origins that were bound by RepID in WT cells (14,716 NS-Seq peaks). (e) A screenshot of a sample genomic region showing replication-initiation profiles as in a. The shaded region delineates a replication origin that does not bind RepID in WT cells and initiates replication in both WT and KO cells. An expanded screenshot of the same region is shown in Supplementary Fig. 4f. (f) The distribution of replication-initiation events in KO cells along genomic regions flanking replication origins that initiated replication but did not bind RepID in WT cells (64,143 peaks). (g) Consensus sequence for RepID binding identified using a subset of RepID-bound regions. Consensus is aligned with a 12-bp motif, which matches the AG1 site. Data showing the abundance of the motif in RepID-bound regions or randomized files are presented in Supplementary Table 4.
Figure 4
Figure 4. Depletion of RepID decreases the frequency of replication-initiation events.
(a) Cells were sequentially labelled with IdU followed by CldU. Top panel, a typical field with replication signals (IdU detected in green and CldU detected in red). Second panel, the same field with all fibres labelled with an antibody detecting single-strand DNA (ssDNA; grey). Third and fourth panels, an example of CldU-IdU (third) ssDNA (fourth) fibre tracks from RepID WT MEFs. Fifth and sixth panels, an example of CldU-IdU (fifth) ssDNA (sixth) fibre tracks from RepID −/− MEFs. Illustrations of replication fork patterns are shown below the ssDNA track. The lengths of fibres label associated with ldU and CIdU incorporation and inter-origin distances were measured (see Methods), and rates of replication fork progression were calculated based on these values. Ori, origin; ssDNA, DNA detected by anti-single strand antibody. (b,c) Measurements of the distribution of distances between replication origins in DNA fibres from WT MEFs and RepID −/− MEFs. (d,e) Measurements of the distribution of replication fork progression rates for WT and RepID −/−MEFs. The differences between measurements from fibres obtained from wild-type and RepID-deficient MEFs were significant at P<0.05 (P=0.0218 for inter-origin distance and P=0.0061 for replication fork speed as calculated using the Mann–Whitney test). Normality test by Kolmogorov–Smirnov test showed that the distributions of data for be are not normal (P<0.01).
Figure 5
Figure 5. RepID-deficient MEFs exhibit replication fork asymmetry.
(a) Examples of DNA fibres derived from wild-type and RepID −/− MEFs that contain symmetric and asymmetric replication forks. (b,c) Scatter plots of left and right fork lengths in RepID WT (b) and RepID −/− (c) cells. The percentages of the asymmetric forks (outside the red lines) and the number of replication forks measured in both cells are presented on the plots, demonstrating that 31% of forks were asymmetric in RepID−/− MEFs, whereas 8% of forks were asymmetric in RepID WT MEFs.
Figure 6
Figure 6. RepID is present in a complex between LCR and Rep-P in early replicating HBB loci.
(a) Schematic illustration of the beta-globin locus and the outline of ChIP-3C procedure. Cells were lysed and digested with HindIII. Crosslinked chromatin fragments were immunoprecipitated with anti-RepID antibody and ligated. Crosslinks were reversed and DNA was isolated and amplified by PCR. The primers and TaqMan probe used for the ChIP-3C analysis are listed in the ‘Methods' section. The red arrows indicate the location of AG and HS2 on the beta-globin locus, which are more than 50 kb away from each other. (b) ChIP-3C analysis of long-distance RepID-associated chromatin interactions at the HBB locus in U2OS WT and RepID knockout cells. The 3C primers correspond to sequences near the downstream sticky ends of the 3C fragments. Primer/probe combinations were designated p6 to p13, and their locations are indicated as half arrows. The x axis shows the positions of the restriction fragments on the genomic scale (vertical bars). The diagram represents the extent of PCR amplification of each primer/probe pair. Values represent the average of triplicate samples (bars represent s.e.'s). Data were obtained after subtraction of no ligation controls and were normalized to the AG site loading control. Primer efficiencies were normalized using a single bacterial artificial chromosome (BAC) clone covering the genome segment under study.
Figure 7
Figure 7. Spatial organization of DNA–protein interactions within Rep-P.
The asymmetric region within the Rep-P replication of the HBB locus is involved in two distinct DNA–protein interactions. An interaction with the LARC complex facilitates transcription and maintains an open chromatin conformation in erythroid cells, whereas the interaction of adjacent sequences with RepID facilitates the initiation of DNA replication from Rep-P. Top diagram, a schematic of the HBB locus, illustrating the location of the Rep-P replicator; middle, a schematic of the Rep-P replicator, illustrating the location of the AG sequence (sequence shown in Fig. 1a); lower diagram, two adjacent protein–DNA interactions within AG.

Similar articles

See all similar articles

Cited by 16 articles

See all "Cited by" articles

References

    1. Mechali M. Eukaryotic DNA replication origins: many choices for appropriate answers. Nat. Rev. Mol. Cell Biol. 11, 728–738 (2010). - PubMed
    1. Bielinsky A. K. Replication origins: why do we need so many? Cell Cycle 2, 307–309 (2003). - PubMed
    1. Aladjem M. I. Replication in context: dynamic regulation of DNA replication patterns in metazoans. Nat. Rev. Genet. 8, 588–600 (2007). - PubMed
    1. Martin M. M. et al. . Genome-wide depletion of replication initiation events in highly transcribed regions. Genome Res. 21, 1822–1832 (2011). - PMC - PubMed
    1. Gilbert D. M. et al. . Space and time in the nucleus: developmental control of replication timing and chromosome architecture. Cold Spring Harb. Symp. Quant. Biol. 75, 143–153 (2010). - PubMed

Publication types

Substances

LinkOut - more resources

Feedback