Histone H3K56 acetylation, CAF1, and Rtt106 coordinate nucleosome assembly and stability of advancing replication forks

Abstract

Chromatin assembly mutants accumulate recombinogenic DNA damage and are sensitive to genotoxic agents. Here we have analyzed why impairment of the H3K56 acetylation-dependent CAF1 and Rtt106 chromatin assembly pathways, which have redundant roles in H3/H4 deposition during DNA replication, leads to genetic instability. We show that the absence of H3K56 acetylation or the simultaneous knock out of CAF1 and Rtt106 increases homologous recombination by affecting the integrity of advancing replication forks, while they have a minor effect on stalled replication fork stability in response to the replication inhibitor hydroxyurea. This defect in replication fork integrity is not due to defective checkpoints. In contrast, H3K56 acetylation protects against replicative DNA damaging agents by DNA repair/tolerance mechanisms that do not require CAF1/Rtt106 and are likely subsequent to the process of replication-coupled nucleosome deposition. We propose that the tight connection between DNA synthesis and histone deposition during DNA replication mediated by H3K56ac/CAF1/Rtt106 provides a mechanism for the stabilization of advancing replication forks and the maintenance of genome integrity, while H3K56 acetylation has an additional, CAF1/Rtt106-independent function in the response to replicative DNA damage.

Figure 1. Defective replication-coupled chromatin assembly causes accumulation of recombinogenic DNA damage and checkpoint activation.

Effect of asf1Δ, H3K56R, asf1Δ H3K56R, rtt109Δ, hir1Δ, cac1Δ, rtt106, cac1Δ rtt106Δ and asf1Δ cac1Δ rtt106Δ on the frequency of genetic recombination between inverted repeats (A, C) and budded cells with Rad52-YFP foci (B, D). Asterisks and circles indicate statistically significant differences compared to wild type and mutants asf1Δ and cac1Δ rtt106Δ, respectively, according to an Anova one-way (Tukey) test, where one asterisk/circle represents a P-value<0.001 and two represents <0.05. Note that strains in panels (A, B) and (C, D) have different genetic backgrounds (MSY421 and BY4701, respectively). For the frequency of genetic recombination the average and standard deviation of 3–16 fluctuation tests performed with 3–8 independent transformants of each strain are shown. For the percentage of budded cells with Rad52-YFP foci 600–900 cells for each strain were analyzed, and the average and standard deviation of 6–9 independent measures are shown. Rad53 phosphorylation in the indicated strains under unperturbed conditions by western blot (E) and in situ kinase assay (F). The wild-type strain treated with 0.033% MMS for 2 h was used as a control of checkpoint activation.

Figure 2. Histone H3K56 acetylation is required for preventing the loss of replication forks.

(A) Schematic representation of the telomere-proximal region replicated from the early origin ARS305 (black oval). The position of dormant origins (grey ovals) and restriction fragments analyzed by 2D-gel electrophoresis is shown. (B) Schematic representation of the migration pattern of the bubble-, single Y-, double Y- and X-shaped RIs by 2D-gel electrophoresis. (C, D) Analysis of RIs at the ARS305 and two adjacent EcoRV-HindIII regions of cells synchronized in G1 and released into S phase. A representative kinetics with its quantification is shown. Quantification of the RIs was normalized to the total amount of DNA, including linear monomers (n), to the size of the restriction fragment, and to the percentage of cells synchronized in G1. The percentage of RIs at the ARS305 during the kinetics was calculated as the sum of bubbles, Ys and Xs at region Or of all time points combined, taking the total amount of wild-type RIs as 100. The average and standard deviation of 5 (asf1Δ) and 3 (rtt109Δ and H3K56R) independent experiments are shown.

Figure 3. Defective CAF1/Rtt106-dependent chromatin assembly causes a loss of replication forks.

(A, B) Analysis of RIs at the ARS305 and two adjacent EcoRV-HindIII regions of cells synchronized in G1 and released into S phase. See Legend Figure 2 for details. The average and standard deviation of 3 independent experiments are shown.

Figure 4. Chromatin assembly mutants are not affected in ARS305 replication firing.

(A) Cell cycle progression by DNA content analysis of cells synchronized in G1 and released into S phase. (B) Percentage of G1 synchronized cells that reach G2/M. This value was obtained by FACS analysis of cells synchronized in G1 and released into S phase in the presence of NCD until the number of cells in G2/M did not change. It was calculated as (%G2_f-%G2_i)/%G1_i. The average and standard deviation of 3 independent experiments is shown. Statistically significant differences were not obtained according to an Anova one-way (Tukey) test. (C) Cell cycle progression by budding analysis of cells synchronized in G1 and released into S phase in the absence (top and middle) or presence (bottom) of NCD. The presence of NCD prevented G2/M cells at time cero from re-entering a new cell cycle thus allowing budding analysis in mutants in which α-factor synchronization led to less than 90% cells in G1. The average and standard deviation of 3 independent experiments are shown. Statistically significant differences compared to wild type (P-value<0.05) were obtained only in cac1Δ rtt106Δ, asf1Δ cac1Δ rtt106Δ, asf1Δ rad52Δ and H3K56R at times 45 and 60 minutes, according to an Anova one-way (Tukey) test. (D) Efficiency of ARS305 replication firing determined as the amount of DNA at the origin in cells arrested in S phase with HU relative to cells arrested in G1 with α-factor. The average and standard deviation of 3 independent measures are shown. Statistically significant differences were not obtained according to an Anova one-way (Tukey) test.

Figure 5. Homologous recombination is required for replication fork rescue in asf1Δ.

(A) Analysis of RIs at the ARS305 and two adjacent EcoRV-HindIII regions of cells synchronized in G1 and released into S phase. See Legend Figure 2 for details. The average and standard deviation of 3 independent experiments are shown. (B) Effect of asf1Δ, rad52Δ and asf1Δ rad52Δ on cell growth.

Figure 6. Chromatin assembly is not required for the stability of stalled replication forks.

(A) Analysis of stalled RIs at the ARS305 and two adjacent EcoRV-HindIII regions of cells synchronized in G1 and released into the S phase in the presence of 0.2 M HU for different times. The percentage of RIs over the whole region during the kinetics was calculated as the sum of bubbles, Ys and Xs of all time points combined, taking the total amount of wild-type RIs as 100. (B) Analysis of stalled RIs at the ARS305 and two adjacent EcoRV-HindIII regions of cells synchronized in G1 and released into the S phase in the presence of 0.2 M HU for 1 hour. A representative kinetics with its quantification is shown. The percentage of RIs over the whole region was calculated as the sum of bubbles, Ys and Xs in the three fragments (Or, A and B), taking the total amount of wild-type RIs as 100. The average and standard deviation of 7 (asf1Δ) and 3 (rest) independent experiments are shown. (C) Amount of X-shaped molecules relative to total RIs (bubbles, Ys and Xs) at the EcoRV-HindIII ARS305 fragment from cells synchronized in G1 and released into the S-phase in the presence of 0.2 M HU for 30 and 60 minutes. The average and standard deviation of 10 (asf1Δ), 6 (rad52Δ) and 7 (asf1Δ rad52Δ) values are shown. Only increases in asf1Δ (P-value<0.001), asf1Δ rad52Δ (P-value<0.001) and rad52Δ (P-value<0.01) relative to wild type, and in asf1Δ relative to rad52Δ (P-value<0.005) are statistically significant, according to an Anova one-way (Tukey) test.

Figure 7. Roles of H3K56 acetylation and CAF1/Rtt106 on response to replication inhibition and replicative DNA damage.

(A) DNA damage sensitivity to genotoxic agents as determined by ten-fold serial dilutions from the same number of mid-log phase cells onto medium containing drugs at the indicated concentrations. (B) Cell-cycle progression by FACS analysis of cells synchronized in G1 and released into the S-phase in the presence of 0.2 M HU (left) or 0.033% MMS (right) for 1 hour, and then released into fresh media for the indicated times. (C) Kinetics of checkpoint activation and deactivation upon replicative DNA damage as determined by western blot against phosphorylated Rad53 from selected samples in (B).