Limits...
USF binding sequences from the HS4 insulator element impose early replication timing on a vertebrate replicator.

Hassan-Zadeh V, Chilaka S, Cadoret JC, Ma MK, Boggetto N, West AG, Prioleau MN - PLoS Biol. (2012)

Bottom Line: This shift requires the presence of HS4 on both sides of the replication origin and results in an advance of replication timing of the target locus from the second half of S-phase to the first half when a transcribed gene is positioned nearby.Moreover, we find that the USF transcription factor binding site is the key cis-element inside the HS4 insulator that controls replication timing.Taken together, our data identify a combination of cis-elements that might constitute the basic unit of multi-replicon megabase-sized early domains of DNA replication.

View Article: PubMed Central - PubMed

Affiliation: Institut Jacques Monod, Centre National de la Recherche Scientifique, Université Paris Diderot, Paris, France.

ABSTRACT
The nuclear genomes of vertebrates show a highly organized program of DNA replication where GC-rich isochores are replicated early in S-phase, while AT-rich isochores are late replicating. GC-rich regions are gene dense and are enriched for active transcription, suggesting a connection between gene regulation and replication timing. Insulator elements can organize independent domains of gene transcription and are suitable candidates for being key regulators of replication timing. We have tested the impact of inserting a strong replication origin flanked by the β-globin HS4 insulator on the replication timing of naturally late replicating regions in two different avian cell types, DT40 (lymphoid) and 6C2 (erythroid). We find that the HS4 insulator has the capacity to impose a shift to earlier replication. This shift requires the presence of HS4 on both sides of the replication origin and results in an advance of replication timing of the target locus from the second half of S-phase to the first half when a transcribed gene is positioned nearby. Moreover, we find that the USF transcription factor binding site is the key cis-element inside the HS4 insulator that controls replication timing. Taken together, our data identify a combination of cis-elements that might constitute the basic unit of multi-replicon megabase-sized early domains of DNA replication.

Show MeSH
Genome-wide analysis of replication timing in DT40 cells.(A) Replication timing profile along a portion of Chromosome 1 is shown. This profile was obtained with cells sorted into three fractions (early, mid-, and late S-phase). After immunoprecipitation, BrdU pulse-labeled nascent DNA from early and late fractions was differentially labeled and cohybridized to a chicken whole-genome oligonucleotide microarray at a density of one probe every 5.6 kb. The log2-ratio (early/late) of the abundance of each probe in the early and late S-phase is shown. Early and late domains are in red and green, respectively. The chosen site of insertion is indicated. Below annotated genes, human proteins and CpG islands are shown. (B) Replicate experiments with nascent strands extracted from cells sorted into two fractions, early and late-replicating DNA, were reciprocally labeled (“dye-switch”) and hybridized. The log2-ratio timing profiles were smoothed using the Moving Average option of the Agilent Genomic Workbench 5.0 software with the linear algorithm and 200 kb windows (log2(Early/Late) blue and log2(Late/Early) pink). The mirror image shows the high degree of correlation between replicates.
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC3295818&req=5

pbio-1001277-g001: Genome-wide analysis of replication timing in DT40 cells.(A) Replication timing profile along a portion of Chromosome 1 is shown. This profile was obtained with cells sorted into three fractions (early, mid-, and late S-phase). After immunoprecipitation, BrdU pulse-labeled nascent DNA from early and late fractions was differentially labeled and cohybridized to a chicken whole-genome oligonucleotide microarray at a density of one probe every 5.6 kb. The log2-ratio (early/late) of the abundance of each probe in the early and late S-phase is shown. Early and late domains are in red and green, respectively. The chosen site of insertion is indicated. Below annotated genes, human proteins and CpG islands are shown. (B) Replicate experiments with nascent strands extracted from cells sorted into two fractions, early and late-replicating DNA, were reciprocally labeled (“dye-switch”) and hybridized. The log2-ratio timing profiles were smoothed using the Moving Average option of the Agilent Genomic Workbench 5.0 software with the linear algorithm and 200 kb windows (log2(Early/Late) blue and log2(Late/Early) pink). The mirror image shows the high degree of correlation between replicates.

Mentions: Our aim was to identify combinations of cis-elements capable of imposing early replication timing in a naturally late replicating region. To do so, we employed the efficient homologous recombination capacity of the chicken lymphoid DT40 cell line as a model system [38]. We decided to target an isogenic locus that is replicated in the second part of S-phase and is devoid of replication origins. Firstly, we determined the genome-wide replication timing profiles of DT40 cells following pulse labeling with BrdU and cell sorting into three fractions (early, mid, and late S-phase). BrdU labeled nascent DNA from early and late fractions were immunoprecipitated with anti-BrdU antibody, amplified, differentially labeled, and co-hybridized onto a whole chicken genome oligonucleotide microarray. The log2-ratio of the abundance of each genomic probe in the early and late S-phase fractions generates a replication-timing profile that reveals early and late replicated domains (Figure 1A). Exclusion of the mid-replicated fraction increases the intensity of the differences between early and late replicating domains without changing the global shape of the pattern (compare Figure 1A and 1B). We made replicate experiments with nascent strands extracted from cells sorted into early and late-replicating DNA fractions, which were reciprocally labeled (“dye-switch”) prior to hybridization. The mirror image of the timing profiles reflects a high degree of correlation (Figure 1B). We also constructed a whole genome map of replication origins in DT40 cells to allow the identification of origin free regions. We prepared four independent biological samples of short nascent strands (SNS) from 108 cells as described previously [17]. SNS, by contrast to NS, which are synthesized along the whole genome, are only enriched at replication starting points. SNS were pooled and made double stranded by random-priming and ligation. DNA was then fragmented and two different libraries were constructed and subject to high throughput sequencing. A total of 60 million uniquely mapped reads were generated. We arbitrarily selected a mid-late replicating region on chicken chromosome 1 that is devoid of replication origins (Figures 1A and 2). The chromosomal landscape of the chosen integration site in DT40 cells is AT-rich and lacks transcriptionally active genes (Figure 2A). The closest origins, detected by deep sequencing and validated by qPCR, are located 58 kb upstream and 80 kb downstream of the site of insertion (Figure 2A). We analyzed replication timing more precisely by sorting BrdU pulse-labeled cells into four S-phase fractions, from early to late (S1–S4). We quantified nascent strands (NS) across the chromosomal region surrounding the site of integration. The region from 140 kb upstream to 150 kb downstream of the site of integration was found to be mid-late replicating (Figure 2B), in agreement with our genome-wide profiling of replication timing (Figure 1A). In conclusion, we identified and selected a 100 kb intergenic region devoid of replication origins that is replicated in mid-late S-phase. The capacity to specifically target this region by homologous recombination gave us a model system in which we can test the effect of cis-regulatory elements on replication dynamics in a very controlled manner.


USF binding sequences from the HS4 insulator element impose early replication timing on a vertebrate replicator.

Hassan-Zadeh V, Chilaka S, Cadoret JC, Ma MK, Boggetto N, West AG, Prioleau MN - PLoS Biol. (2012)

Genome-wide analysis of replication timing in DT40 cells.(A) Replication timing profile along a portion of Chromosome 1 is shown. This profile was obtained with cells sorted into three fractions (early, mid-, and late S-phase). After immunoprecipitation, BrdU pulse-labeled nascent DNA from early and late fractions was differentially labeled and cohybridized to a chicken whole-genome oligonucleotide microarray at a density of one probe every 5.6 kb. The log2-ratio (early/late) of the abundance of each probe in the early and late S-phase is shown. Early and late domains are in red and green, respectively. The chosen site of insertion is indicated. Below annotated genes, human proteins and CpG islands are shown. (B) Replicate experiments with nascent strands extracted from cells sorted into two fractions, early and late-replicating DNA, were reciprocally labeled (“dye-switch”) and hybridized. The log2-ratio timing profiles were smoothed using the Moving Average option of the Agilent Genomic Workbench 5.0 software with the linear algorithm and 200 kb windows (log2(Early/Late) blue and log2(Late/Early) pink). The mirror image shows the high degree of correlation between replicates.
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC3295818&req=5

pbio-1001277-g001: Genome-wide analysis of replication timing in DT40 cells.(A) Replication timing profile along a portion of Chromosome 1 is shown. This profile was obtained with cells sorted into three fractions (early, mid-, and late S-phase). After immunoprecipitation, BrdU pulse-labeled nascent DNA from early and late fractions was differentially labeled and cohybridized to a chicken whole-genome oligonucleotide microarray at a density of one probe every 5.6 kb. The log2-ratio (early/late) of the abundance of each probe in the early and late S-phase is shown. Early and late domains are in red and green, respectively. The chosen site of insertion is indicated. Below annotated genes, human proteins and CpG islands are shown. (B) Replicate experiments with nascent strands extracted from cells sorted into two fractions, early and late-replicating DNA, were reciprocally labeled (“dye-switch”) and hybridized. The log2-ratio timing profiles were smoothed using the Moving Average option of the Agilent Genomic Workbench 5.0 software with the linear algorithm and 200 kb windows (log2(Early/Late) blue and log2(Late/Early) pink). The mirror image shows the high degree of correlation between replicates.
Mentions: Our aim was to identify combinations of cis-elements capable of imposing early replication timing in a naturally late replicating region. To do so, we employed the efficient homologous recombination capacity of the chicken lymphoid DT40 cell line as a model system [38]. We decided to target an isogenic locus that is replicated in the second part of S-phase and is devoid of replication origins. Firstly, we determined the genome-wide replication timing profiles of DT40 cells following pulse labeling with BrdU and cell sorting into three fractions (early, mid, and late S-phase). BrdU labeled nascent DNA from early and late fractions were immunoprecipitated with anti-BrdU antibody, amplified, differentially labeled, and co-hybridized onto a whole chicken genome oligonucleotide microarray. The log2-ratio of the abundance of each genomic probe in the early and late S-phase fractions generates a replication-timing profile that reveals early and late replicated domains (Figure 1A). Exclusion of the mid-replicated fraction increases the intensity of the differences between early and late replicating domains without changing the global shape of the pattern (compare Figure 1A and 1B). We made replicate experiments with nascent strands extracted from cells sorted into early and late-replicating DNA fractions, which were reciprocally labeled (“dye-switch”) prior to hybridization. The mirror image of the timing profiles reflects a high degree of correlation (Figure 1B). We also constructed a whole genome map of replication origins in DT40 cells to allow the identification of origin free regions. We prepared four independent biological samples of short nascent strands (SNS) from 108 cells as described previously [17]. SNS, by contrast to NS, which are synthesized along the whole genome, are only enriched at replication starting points. SNS were pooled and made double stranded by random-priming and ligation. DNA was then fragmented and two different libraries were constructed and subject to high throughput sequencing. A total of 60 million uniquely mapped reads were generated. We arbitrarily selected a mid-late replicating region on chicken chromosome 1 that is devoid of replication origins (Figures 1A and 2). The chromosomal landscape of the chosen integration site in DT40 cells is AT-rich and lacks transcriptionally active genes (Figure 2A). The closest origins, detected by deep sequencing and validated by qPCR, are located 58 kb upstream and 80 kb downstream of the site of insertion (Figure 2A). We analyzed replication timing more precisely by sorting BrdU pulse-labeled cells into four S-phase fractions, from early to late (S1–S4). We quantified nascent strands (NS) across the chromosomal region surrounding the site of integration. The region from 140 kb upstream to 150 kb downstream of the site of integration was found to be mid-late replicating (Figure 2B), in agreement with our genome-wide profiling of replication timing (Figure 1A). In conclusion, we identified and selected a 100 kb intergenic region devoid of replication origins that is replicated in mid-late S-phase. The capacity to specifically target this region by homologous recombination gave us a model system in which we can test the effect of cis-regulatory elements on replication dynamics in a very controlled manner.

Bottom Line: This shift requires the presence of HS4 on both sides of the replication origin and results in an advance of replication timing of the target locus from the second half of S-phase to the first half when a transcribed gene is positioned nearby.Moreover, we find that the USF transcription factor binding site is the key cis-element inside the HS4 insulator that controls replication timing.Taken together, our data identify a combination of cis-elements that might constitute the basic unit of multi-replicon megabase-sized early domains of DNA replication.

View Article: PubMed Central - PubMed

Affiliation: Institut Jacques Monod, Centre National de la Recherche Scientifique, Université Paris Diderot, Paris, France.

ABSTRACT
The nuclear genomes of vertebrates show a highly organized program of DNA replication where GC-rich isochores are replicated early in S-phase, while AT-rich isochores are late replicating. GC-rich regions are gene dense and are enriched for active transcription, suggesting a connection between gene regulation and replication timing. Insulator elements can organize independent domains of gene transcription and are suitable candidates for being key regulators of replication timing. We have tested the impact of inserting a strong replication origin flanked by the β-globin HS4 insulator on the replication timing of naturally late replicating regions in two different avian cell types, DT40 (lymphoid) and 6C2 (erythroid). We find that the HS4 insulator has the capacity to impose a shift to earlier replication. This shift requires the presence of HS4 on both sides of the replication origin and results in an advance of replication timing of the target locus from the second half of S-phase to the first half when a transcribed gene is positioned nearby. Moreover, we find that the USF transcription factor binding site is the key cis-element inside the HS4 insulator that controls replication timing. Taken together, our data identify a combination of cis-elements that might constitute the basic unit of multi-replicon megabase-sized early domains of DNA replication.

Show MeSH