Protein phosphatase 1 recruitment by Rif1 regulates DNA replication origin firing by counteracting DDK activity

Abstract

The firing of eukaryotic origins of DNA replication requires CDK and DDK kinase activities. DDK, in particular, is involved in setting the temporal program of origin activation, a conserved feature of eukaryotes. Rif1, originally identified as a telomeric protein, was recently implicated in specifying replication timing in yeast and mammals. We show that this function of Rif1 depends on its interaction with PP1 phosphatases. Mutations of two PP1 docking motifs in Rif1 lead to early replication of telomeres in budding yeast and misregulation of origin firing in fission yeast. Several lines of evidence indicate that Rif1/PP1 counteract DDK activity on the replicative MCM helicase. Our data suggest that the PP1/Rif1 interaction is downregulated by the phosphorylation of Rif1, most likely by CDK/DDK. These findings elucidate the mechanism of action of Rif1 in the control of DNA replication and demonstrate a role of PP1 phosphatases in the regulation of origin firing.

Graphical abstract

Figure 1

Rif1 Interacts with PP1 and Recruits It to Telomeres (A) Left: N-terminal sequence of ScRif1 spanning the putative PP1 docking motifs (top), which were mutated in the rif1-PP1 allele (bottom). Right: N-terminal sequence of SpRif1 spanning the putative PP1 docking motifs (top), mutated in the rif1-PP1 allele (bottom). (B) Protein extracts from budding yeast cells of the indicated genotypes were immunoprecipitated with anti-Myc and analyzed by western blotting against Flag (Rif1) and Myc (Glc7). (C) Protein extracts from fission yeast cells of the indicated genotypes were immunoprecipitated with anti-GFP and analyzed by western blotting against Myc (Rif1) and GFP (Sds21 and Dis2). (D) ChIP analysis of the association of ScGlc7 with the indicated chromosomal loci in the indicated strains, in exponentially growing asynchronous cultures. Fold enrichment was obtained by normalization against the PDI1 locus. SD values were derived from triplicates, and statistical significance was assessed by determining p values calculated from two-tailed t tests (in all cases, each mutant versus wild-type). ^∗p < 0.05. (E) Association of N-terminally GFP-tagged SpSds21 and SpDis2 from exponentially growing asynchronous cultures at the indicated loci as determined by ChIP and quantified as fold enrichment over the ars2004 locus. SDs and p values were calculated from four replicates. (F) ChIP analysis of SpDis2 chromatin binding as in (E); SDs and data are from four replicates. See Figure S1 for expression levels of mutant alleles.

Figure 2

The PP1 Docking Motifs in Rif1 Are Required to Establish the Replication Timing of Budding Yeast Telomeres and Fission Yeast DNA Replication Origins (A) Analysis of the association of C-terminally Myc-tagged Pol2 with selected telomeres and origins in RIF1 wild-type, rif1-PP1, and rif1-Δ budding yeast cells after synchronous release from G1 arrest. ARS607 (blue) and ARS1412 (orange) were used as markers of early and late S phase, respectively. To account for differences in efficiencies in the immunoprecipitations among different experiments, each profile for each amplicon was normalized against its highest peak. The data represent the average of three independent experiments for each strain. The significance of the change in the position of the telomere peak for each rif1 mutant against the wild-type was assessed by applying a Wilcoxon test (one sided; p < 0.024). (B) Analysis of the replication timing of ARS607 and the VI-R and XV-L telomeres, in reference to ARS1412. DNA amounts for cells after release from G1 arrest were quantified by quantitative PCR (qPCR). For each of the three loci analyzed, normalization was first carried out against the ARS1412 locus at the same time point, and subsequently against the G1 time point (0 min). A minimum of three experiments were averaged for the analysis. Two-tailed t tests were carried out for significance for each mutant against wild-type at the same time point (p < 0.05 is indicated by asterisks). See also Figure S2. (C) Replication efficiency of early and late origins in fission yeast, in wild-type, rif1-Δ, rif1-PP1, and sds21-Δ strains. Log-phase cultures were arrested in G2 at 36°C using the cdc25-22 temperature-sensitive allele and then released into 25 mM hydroxyurea for 140 min. Genomic DNA was prepared for the G2 (0 min time point) and late S phase (140 min time point) cells and quantified by qPCR. The ratio of the amount of genomic DNA in late S phase to that in G2 was calculated for each locus. The non-ori1 locus was used for normalization (Hayano et al., 2012). Two-tailed t test for each mutant against wild-type were performed from at least eight replicates. A p value < 0.05 was deemed significant and is indicated by an asterisk in the graph. SDs are indicated in all panels.

Figure 3

Recruitment of PP1 by Rif1 Counteracts DDK Activity on Mcm4 (A) Suppression of the temperature-sensitivity phenotype of the budding yeast cdc7-1 allele by rif1-PP1. 5-fold serial dilutions of log-phase cultures of budding yeast strains of the indicated genotypes were plated onto YPAD media and incubated at temperatures ranging from 25°C to 33°C. Plates were imaged following 2 day incubations. (B) Suppression of the temperature sensitivity of the fission yeast hsk1-89 allele by rif1-PP1. 10-fold serial dilutions of log-phase cultures of the indicated genotypes were spotted on rich medium and incubated for 4 days at 25°C or 3 days at 30°C and 37°C (the latter is a permissive temperature for hsk1-89). (C) Analysis of budding yeast Mcm4 phosphorylation. Budding yeast strains bearing Flag-tagged Mcm4 were arrested in the G1 phase with α-factor at 25°C or 37°C for 2 hr, as indicated. Western analysis of protein samples was performed with anti-Flag (top) and anti-Pgk1 (bottom). The phosphorylated fraction of the Mcm4 protein and the total Mcm4 signal were quantified using ImageQuant software and normalized to the loading control (Pgk1), and then the percentage of phosphorylation was calculated. The p values from two-tailed t tests are reported in the graph. At least three replicates were used for the analysis and SDs are indicated. See also Figure S3.

Figure 4

Rif1/PP1 Is Affected by Mutations at Putative CDK and DDK Phosphorylation Sites in the Rif1 N Terminus (A) N-terminal sequence of ScRif1 spanning the putative PP1 docking motifs (top). The RVxF- and SILK-type motifs are indicated in purple and green, respectively. Potential DDK sites are indicated in orange; putative CDK sites are indicated in blue (top). Phosphomimic changes to aspartic acid present in the rif1-9D allele are indicated in red (bottom). (B) Suppression of the temperature-sensitivity phenotype of the budding yeast cdc7-1 allele by rif1-9D. 5-fold serial dilutions of log-phase cultures of budding yeast strains of the indicated genotypes were plated onto YPAD media and incubated at temperatures ranging from 25°C to 33°C. Plates were imaged following 2 day incubations. (C) Protein extracts from budding yeast cells of the indicated genotypes were immunoprecipitated with anti-Myc and analyzed by western blotting against Flag (Rif1) and Myc (Glc7). (D) N-terminal sequence of SpRif1 spanning the putative PP1 docking motifs (top). The RVxF- and SILK-type motifs, and the putative DDK and CDK sites are indicated as in (A). Changes to aspartic acid or alanine present in the rif1-12D, rif1-7A, and rif1-7APP1 alleles are indicated in red (bottom). (E) Suppression of the temperature sensitivity of the fission yeast hsk1-89 allele by various rif1 alleles. Ten-fold serial dilutions of log-phase cultures of the indicated genotypes were spotted on rich medium and incubated for 4 days at 25°C or 3 days at 30°C and 37°C. (F) Association of N-terminally GFP-tagged SpSds21 from exponentially growing asynchronous cultures at the indicated loci as determined by ChIP and quantified as fold enrichment over the ars2004 locus. SDs and p values for each mutant versus wild-type were calculated from three replicates. See Figure S1 for expression levels of mutant alleles.