Telomere-binding protein Taz1 controls global replication timing through its localization near late replication origins in fission yeast

Abstract

In eukaryotes, the replication of chromosome DNA is coordinated by a replication timing program that temporally regulates the firing of individual replication origins. However, the molecular mechanism underlying the program remains elusive. Here, we report that the telomere-binding protein Taz1 plays a crucial role in the control of replication timing in fission yeast. A DNA element located proximal to a late origin in the chromosome arm represses initiation from the origin in early S phase. Systematic deletion and substitution experiments demonstrated that two tandem telomeric repeats are essential for this repression. The telomeric repeats recruit Taz1, a counterpart of human TRF1 and TRF2, to the locus. Genome-wide analysis revealed that Taz1 regulates about half of chromosomal late origins, including those in subtelomeres. The Taz1-mediated mechanism prevents Dbf4-dependent kinase (DDK)-dependent Sld3 loading onto the origins. Our results demonstrate that the replication timing program in fission yeast uses the internal telomeric repeats and binding of Taz1.

Figure 1.

Replication timings of ars2004 and AT2088 origin fragments are maintained at ectopic loci. (A) The locations of early and late replication origins on Schizosaccharomyces pombe chromosome II are presented schematically. The positions of early origins AT2024 (gray) and ars2004 (red); cen2; late origins AT2035 (green) and AT2088 (blue); and a subtelomeric origin, TAS59 (purple) are shown. For the ars2004 and AT2088 loci, the locations of the genes, along with the direction of transcription (arrow) and fragments (3.2-kb ars2004[red]; 3.6-kb AT2088 [blue]) used for translocation, are presented. Relevant restriction fragments (EcoT22I [E], FbaI [F], and NcoI [N]) analyzed by two-dimensional (2D) gel electrophoresis are shown below the maps. (B) Replication kinetics of early and late origins. Fission yeast cdc25-22 nmt1-TK⁺ cells arrested at the G2/M boundary for 3 h at 36°C were released at 25°C in the presence of BrdU (200 μM). At the indicated time points, the replicated heavy–light (HL) DNA was separated from light–light (LL) DNA using cesium chloride (CsCl) density gradient centrifugation, and the amount of DNA of AT2024 (black), AT2035 (green), ars2004 (red), AT2088 (blue), and TAS59 (purple) in the LL and HL densities was determined by qPCR. The replication kinetics of each origin are presented. The results of biologically independent experiments are shown in Supplemental Figure S1B. (C,D) The replication kinetics of ars2004 (red), inserted at the AT2088 locus (C), and AT2088, inserted at the ars2004 locus (D), were obtained as described in B. The results of biologically independent experiments are shown in Supplemental Figure S1, C and D. The right panels show the results of 2D gel analysis of ars2004 (FbaI–NcoI fragment) at the AT2088 locus (C) and of AT2088 (EcoT22I fragment) at the ars2004 locus (D) prepared at 90 min after G2/M release in the presence of HU (10 mM). (D) The membrane was rehybridized with the early origin AT2046 probe. An arrowhead indicates the bubble arc. (E) Comparison of the time required for replication in half of the cell population (T1/2). T1/2 (in minutes) for each origin was obtained from the results presented in B–D and Supplemental Figure. S1B–D. The difference in T1/2 between an origin and AT2024 (early origin control) is presented. AT2035 and AT2088 replicate 6–12 min later than ars2004 and AT2024, regardless of their locations on the chromosome.

Figure 2.

Repression of early initiation from the ars2004 origin by the adjoining AT2088 fragment. (A) Schematic presentation of the ars2004 locus carrying the ura4⁺-AT2088 or ura4⁺ insertion. (B) cdc25-22 nmt1-TK⁺ derivatives with or without the insertion of ura4⁺-AT2088 or ura4⁺ alone next to the ars2004 locus were released from the G2/M block and labeled with BrdU for 120 min in the presence of HU (10 mM). Replication (percentage) in the presence of HU was determined for ars2004 (red) as well as the control early and late origins AT2024 (gray) and AT2035 (green), respectively. The mean ± SD obtained from multiple measurements in qPCR is presented. The results of a biologically independent experiment are presented in Supplemental Figure S2A. (C) The replication intermediates of ars2004 (StyI fragment) and a control early origin, AT2047 (StyI fragment), prepared at 90 min after G2/M release of the HU-treated AT2088-ars2004 cells were analyzed by 2D gel electrophoresis. An arrowhead shows the bubble arc.

Figure 3.

Identification of the essential sequence for RTC. (A) The structure of the ars2004 locus with the insertion of AT2088 fragment is presented schematically. (B) Cells carrying the insertion of various portions of the AT2088 fragment (blue horizontal bars) next to the ars2004 locus were synchronously released from G2/M and labeled with BrdU in the presence of HU (10 mM) for 120 min (top panel), 90 min (middle panel), or 80 min (bottom panel). Replication (percentage) of the ars2004 and that of the internal control origin AT2024 were determined as mean ± SD from multiple qPCR measurements, as described in Figure 2. To compare the effects of the insertions on early replication of ars2004, relative replication of ars2004 to that of AT2024 was obtained, and the ratio of relative replication (ars2004/AT2024) with insertion to value without insertion was determined (red horizontal bars). The results of biologically independent experiments are presented in Supplemental Figure S3A. (C) The profile of AT content of every 100-bp window in a 3.6-kb AT2088 region is presented together with two AT-rich regions (>75%, shaded rectangles), the pre-RC site (arrowhead) (Hayashi et al. 2007), and the locations of two ARS fragments, AT2088A (Supplemental Fig. S3B) and ars745 (Maundrell et al. 1988). The regions sufficient (gray) and essential (red) for RTC are shown above the AT content profile. (D) The 3.6-kb AT2088 fragment with base substitutions in the essential region from position 2617 to 2648, as determined in B, was inserted next to the ars2004, and the effects of base substitutions on the repression of early replication of ars2004 were analyzed as described in B. The nucleotide sequence of the complementary strand of 2617–2648 is presented together with base substitutions (lowercase letters). The RTC essential sequence is indicated. (E) Replication intermediates in the HU-treated wild-type (left panels) and AT2088Δ210 (right panels) cells lacking the 210-bp (2527–2736) sequence were analyzed by 2D gel electrophoresis for ars2004 (EcoT22I fragment) and AT2088 (FbaI–NcoI fragment) at 90 min after G2/M release. Arrowheads indicate the bubble arcs.

Figure 4.

Telomeric repeats that recruit Taz1 are essential for repression of late replication origins. (A) The positions of AT2035 (green), AT2088 (blue), and Tel-0.3 (purple) on chromosome II and of ade6 (gray) on chromosome III are presented schematically. (B, left) The nucleotide sequences of the relevant regions of AT2088 (2624–2648, complementary strand), containing two copies of telomere-like repeats (blue underlining); AT2088-S2632, carrying base substitutions (red); and AT2088-telF and AT2088-telR,with two telomeric repeats (blue underlining) in either orientation, respectively, are presented together with the consensus sequence of fission yeast telomeres (Cooper et al. 1997). (Right) Replication (percentage) of the AT2088 (blue), ars2004 (red), and AT2035 (green) was measured in the HU-treated wild type and in strains carrying base substitutions in AT2088 at 120 min after G2/M release as described in Figure 2. The mean ± SD obtained from multiple measurements in qPCR is presented. The results of biologically independent experiments are shown in Supplemental Figure S4A. (C) Replication of the AT2035 was examined as described in B. The nucleotide sequences of the region (1954–1993, complementary strand) containing telomere-like repeats (blue underlining) and base substitutions (red) in AT2035-S1959 and AT2035-S1985 are presented. The results of biologically independent experiments are shown in Supplemental Figure S4B. (D) Flag-Taz1 was immunoprecipitated from asynchronously cultured wild-type, AT2035-S1959, and AT2088-S2632 cells. Immunoprecipitation recovery of the chromosome locus against the total cellular DNA was measured by qPCR for ade6⁺ (gray), AT2035 (green), AT2088 (blue), and Tel-0.3 (purple). Bars indicate the mean of immunoprecipitation recovery (percentage) with standard error (SE) obtained from three biologically independent experiments. (Right) The results, except for those of Tel-0.3, are presented for a better comparison of Taz1 localization at the internal sites.

Figure 5.

Two copies of the RTC essential sequence are sufficient for recruitment of Taz1 and repression of the nearby early origin. (A) Two copies of the 25-bp fragment (RTC25) containing the RTC essential sequence (purple boxes), inserted next to the ars2004 locus, are presented schematically. A fragment (ars2004*) used for qPCR analysis is shown below the map. (B) cdc25-22 nmt1-TK⁺ cells with (2RTC25-ars2004) or without (wild type [WT]) the insertion of 2RTC25 were released synchronously from G2/M and labeled with BrdU in the presence of HU (10 mM) for 120 min. The ratios of the replicated population of non-ARS, AT2035, and ars2004 loci to that of the AT2024 locus were obtained as described in Figure 3B. (C) Localization of Taz1 at the inserted 2RTC25 was analyzed by ChIP assay. Flag-Taz1 was immunoprecipitated from asynchronously cultured wild-type (WT) and 2RTC25-ars2004 cells. Immunoprecipitation recovery of the chromosome loci against the total cellular DNA was measured by qPCR for ade6⁺ (gray), AT2088 (blue), and ars2004* (orange). The mean ± SD obtained from multiple measurements in qPCR is presented.

Figure 6.

Taz1 is required for repression of a subset of late replication origins. (A) Replication (percentage) of AT2035 (green), ars2004 (red), AT2088 (blue), ARS727 (orange), and TAS59 (purple) was determined in HU-treated wild-type and taz1Δ cells at 120 min after release from G2/M. The mean ± SD obtained from multiple measurements in qPCR is presented. (B) BrdU incorporation profiles of six late origins associated with the telomeric repeats in wild type (top) and taz1Δ (middle) are presented together with profiles of Taz1-ChIP-seq (bottom). The sequence read from the HL density DNA prepared from a 120-min sample after G2/M release in HU-treated wild type (top) and taz1Δ (middle) was compared with the sequence read from the total cellular DNA of the respective cells, and the relevant enrichment was plotted over the chromosome region (Supplemental Fig. S5). (Bottom) Flag-Taz1 was immunoprecipitated as described in Figure 4D, and the relevant enrichment, the sequence read from ChIP in Flag-taz1 cells versus the sequence read from ChIP in the nontag strain, was plotted. The positions of late origins and telomeric repeats are shown by blue arrows and purple triangles, respectively. The results covering the entire genome are presented in Supplemental Figure S5 and summarized in Supplemental Table S1. (C) Replication profile of the subtelomere 2R is presented. Blue arrows show late origins. (D) Comparison of late origins that are activated in taz1Δ. A Venn diagram shows 72 late origins activated in taz1Δ (purple) that contain 46 subtelomeric and 26 internal origins (white), including 16 internal origins where Taz1 was localized (blue).

Figure 7.

Recruitment of Sld3 to a subset of late origins in taz1Δ. Localizations of Mcm6 (A) and Sld3-5Flag (B) at replication origins was analyzed by ChIP assay. Wild-type, AT2088-S2632, and taz1Δ strains carrying the nda4-108/mcm5 and cdc25-22 mutations were synchronously released from the G2/M boundary and incubated for 120 min at 20°C, causing arrest before initiation of replication due to the nda4-108 mutation (Yamada et al. 2004). DNA of the indicated loci recovered by immunoprecipitation with Mcm6 or Sld3-5Flag was measured by qPCR for non-ARS (white), AT2035 (green), ars2004 (red), AT2080 (yellow), AT2088 (blue), ARS727 (orange), and TAS59 (purple). The mean ± SD obtained from multiple measurements in qPCR is presented. The results of biologically independent experiments are shown in Supplemental Figure S6.