Rtt109 slows replication speed by histone N-terminal acetylation

The wrapping of DNA around histone octamers challenges processes that use DNA as their template. In vitro, DNA replication through chromatin depends on histone modifiers, raising the possibility that cells modify histones to optimize fork progression. Rtt109 is an acetyl transferase that acetylates histone H3 before its DNA incorporation on the K56 and N-terminal residues. We previously reported that, in budding yeast, a wave of histone H3 K9 acetylation progresses ∼3–5 kb ahead of the replication fork. Whether this wave contributes to replication dynamics remained unknown. Here, we show that the replication fork velocity increases following deletion of RTT109, the gene encoding the enzyme required for the prereplication H3 acetylation wave. By using histone H3 mutants, we find that Rtt109-dependent N-terminal acetylation regulates fork velocity, whereas K56 acetylation contributes to replication dynamics only when N-terminal acetylation is compromised. We propose that acetylation of newly synthesized histones slows replication by promoting replacement of nucleosomes evicted by the incoming fork, thereby protecting genome integrity.

Eukaryotic cells protect their DNA by wrapping it within chromatin. The basic building blocks of chromatin are nucleosomes, which are histone octamers wrapped by ∼147 bp of DNA. During replication, DNA must dissociate from histones to allow progression of the replication fork. Once the fork has passed, nucleosomes rapidly reassemble, using both available and newly synthesized histones (Groth et al. 2007). This disassembly–reassembly process offers opportunities for regulating replication dynamics and coordinating replication with other processes that use DNA as their template (Hamperl and Cimprich 2016). Understanding the interplay between chromatin and DNA replication is therefore of great interest.Nucleosome stability depends on the interactions between DNA and histones. Determinants of these interactions are the DNA sequence and the various modifications added to histones. Prominent among histone modifications is lysine acetylation, observed in different positions and, in particular, on the disordered histone N termini (Hong et al. 1993; Lee et al. 1993; Bauer et al. 1994). Histone modification regulates nucleosome stability by altering its electrostatic interactions with DNA and by recruiting regulatory proteins that assist in the assembly or disassembly of nucleosomes (Bannister and Kouzarides 2011; Owen-Hughes and Gkikopoulos 2012).DNA replication is initiated from hundreds of sites that are distributed across the genome (Bell and Dutta 2002). These replication origins are bound by the origin recognition complex (ORC), which recruits the MCM complex and thus initiates assembly of the replication machinery (Bell and Stillman 1992). Upon firing, replication forks proceed bidirectionally until approaching forks that emanated from neighboring origins. The temporal replication dynamics therefore depend on the positioning of origins along the genome, the time in S phase when different origins fire, and the velocity at which replication forks proceed along the genome. Replication origins are relatively depleted of nucleosomes, and their surrounding chromatin is enriched with histone H3K4 methylation and H3K9/27 acetylation (MacAlpine et al. 2010; Eaton et al. 2011; Lubelsky et al. 2014; Miotto et al. 2016). Evidence from budding yeast suggests that histone modifications regulate origin firing. For example, tethering of Gcn5 (the catalytic subunit of the SAGA complex) to late origins advanced origin firing time (Vogelauer et al. 2002). Gcn5 acetylates H3 N-terminal lysines, implicating this acetylation in the regulation of origin firing.Cells may regulate fork progression by modifying histones ahead of the progressing fork. In vitro reconstitution experiments have shown that replication through chromatin is significantly slower than replication of naked DNA but can be accelerated by the addition of chromatin chaperones and modifiers (Kurat et al. 2017). Among the factors facilitating fork progression was Gcn5, suggesting a role of H3 N-terminal acetylation in this process. Gcn5 was not sufficient for allowing fork progression through chromatin but promoted the function of the histone chaperone FACT (Orphanides et al. 1999; Belotserkovskaya et al. 2003). Of note, although FACT was essential for fork progression in this assay, CAF-1, which functions as the principal chromatin assembly factor on replicated DNA (Smith and Stillman 1989), was dispensable.Chromatin modification may regulate fork velocity in vivo. We previously reported that in budding yeast, a wave of acetylated H3K9 precedes the replication fork by 3–5 kb (Bar-Ziv et al. 2016a). This wave coincides with the increased chromatin accessibility found ∼7 kb ahead of the fork (Rodriguez and Tsukiyama 2013). This prereplication was not dependent on Gcn5 but instead required Rtt109 (Bar-Ziv et al. 2016a). Rtt109 is an acetyltransferase that modifies H3 on its internal K56 residue and on several N-terminal lysine residues including nine, 14, 23, and 27 (Berndsen et al. 2008; Fillingham et al. 2008). Rtt109 is a replication-specific enzyme that modifies H3 before its DNA incorporation (Han et al. 2007a; Tsubota et al. 2007). This is in contrast to Gcn5, which acts primarily (although not exclusively (Burgess et al. 2010) on DNA-bound histones. Rtt109''s roles during DNA replication include the protection of genomic stability, promotion of nucleosome assembly, and preservation of expression homeostasis (Driscoll et al. 2007; Li et al. 2008; Voichek et al. 2016b). To all these roles, Rtt109 contributes by acetylating H3K56. Acetylation of H3K56 promotes, for example, the interaction of H3 with CAF-1 and, in this way, increases the efficiency of nucleosome assembly at the wake of the fork (Li et al. 2008). H3K56ac is also important for buffering gene expression against the unbalanced gene dosage introduced by replication in S phase (Bar-Ziv et al. 2016b; Voichek et al. 2016a,b, 2018; Bar-Ziv et al. 2020). In contrast, the possible functional consequences of Rtt109 activity on the H3 N-terminal acetylation remain unknown. Our study was set to examine whether this modification contributes to DNA replication dynamics.