Chromatin Controls DNA Replication Origin Selection, Lagging-Strand Synthesis, and Replication Fork Rates

Abstract

The integrity of eukaryotic genomes requires rapid and regulated chromatin replication. How this is accomplished is still poorly understood. Using purified yeast replication proteins and fully chromatinized templates, we have reconstituted this process in vitro. We show that chromatin enforces DNA replication origin specificity by preventing non-specific MCM helicase loading. Helicase activation occurs efficiently in the context of chromatin, but subsequent replisome progression requires the histone chaperone FACT (facilitates chromatin transcription). The FACT-associated Nhp6 protein, the nucleosome remodelers INO80 or ISW1A, and the lysine acetyltransferases Gcn5 and Esa1 each contribute separately to maximum DNA synthesis rates. Chromatin promotes the regular priming of lagging-strand DNA synthesis by facilitating DNA polymerase α function at replication forks. Finally, nucleosomes disrupted during replication are efficiently re-assembled into regular arrays on nascent DNA. Our work defines the minimum requirements for chromatin replication in vitro and shows how multiple chromatin factors might modulate replication fork rates in vivo.

Keywords: DNA replication; biochemistry; chromatin.

Graphical abstract

Figure 1

Loading of the MCM Complex on Chromatin (A) Reaction scheme for chromatin assembly and MCM loading on ARS1-containing 2.8-kb linear DNA coupled to paramagnetic beads. (B) Silver-stained gels of MCM loading reactions on naked DNA (left) compared to chromatin (right). In this and all subsequent experiments on bead-coupled DNA, ORC was added to the chromatin assembly reaction, followed by two 0.3-M K-acetate washes prior to MCM loading. Reactions were performed in the presence of 2 mM ATP or ATPγS. Loading reactions were washed either with 0.3 M K-acetate (low salt wash) or with 0.5 M NaCl (high salt wash). (C) MCM loading in the presence and absence of ORC and Cdc6. Loading reactions were conducted as shown in (A). ORC binding was assessed after two low salt washes by immunoblotting using an antibody recognizing the Orc6 subunit. (D) MCM loading and ORC binding on naked DNA and chromatin using wild-type (WT) and mutant ARS1 (A⁻B2⁻) origin DNA sequences. Reactions were performed as in (A) with indicated amounts of ORC. See also Figures S1 and S2.

Figure 2

Chromatin Inhibits DNA Replication In Vitro (A) Reaction scheme for replication reactions and CMG recruitment on 5.8-kb circular bead-bound ARS1-containing templates using S phase extract or purified proteins. Chromatin assembly and MCM loading were performed as in Figure 1A. Loaded MCMs were further phosphorylated with DDK before they were added to an S phase extract or purified replication proteins. (B) CMG recruitment on naked DNA and on chromatin in the presence of purified initiation and replication factors (Sld3/7, Cdc45, Dpb11, Polε, GINS, Sld2, Mcm10, and S-CDK). Reactions were performed as in Figure 3A. Beads were collected and washed with 0.3 M KCl. Recruitment of the CMG with or without DDK was assessed by immunoblotting using antibodies recognizing Cdc45, Psf1 (GINS), and Mcm7 (MCM) subunits. (C) Replication reactions on naked DNA and on chromatin conducted as shown in (A) using the minimal replication system (Yeeles et al., 2015). In this and all subsequent replication reactions, DNA was visualized by incorporation of [α³²P] deoxycytidine triphosphate (dCTP) into nascent DNA and products were separated through a 0.7% alkaline agarose gel. (D) Replication reactions on naked DNA and on chromatin using the complete replication system as described in Yeeles et al., 2016. (E) CMG recruitment in S phase extract on either naked DNA or chromatin in the presence and absence of DDK phosphorylation. (F) Replication reactions on naked DNA compared to chromatin using S phase extract in the presence or absence of DDK.

Figure 3

FACT Is Necessary and Sufficient for Chromatin Replication (A) Purified histone chaperones, Nhp6 protein, chromatin remodelers, and histone acetyltransferases analyzed by SDS-PAGE with Coomassie staining. (B) Immunodepletion of the Spt16 subunit of the FACT complex from an S phase extract (yCFK2) assessed by immunoblotting using an antibody recognizing the FLAG epitope. (C) Dependence of chromatin replication on in S phase extract. Spt16 was immunodepleted as in (B) and replication reactions on chromatin were performed as described in (A). Indicated amounts of purified FACT were added back to the immunodepleted extract. (D) Effect of FACT on chromatin replication using the complete replication system. See also Figures S3–S5 and Table S3.

Figure 4

Contribution of Chromatin Remodelers and Lysine Acetyltransferases to Chromatin Replication (A) Reaction scheme for soluble replication reaction on chromatin. Chromatin was assembled on ARS1-containing 10.6-kb plasmid DNA in solution in the presence of ORC. ORC-containing chromatinized circles were then isolated for subsequent steps by gel filtration. All subsequent replication reactions were performed using the complete replication system. (B) Time course of a soluble chromatin replication reaction using the scheme shown in (A) in the presence of FACT added at the beginning of the reaction. (C) Effect of histone chaperones and chromatin remodelers on chromatin replication assessed individually. (D) Effect on chromatin replication of adding histone chaperones, chromatin remodelers, or Nhp6 together with FACT. (E) Histone acetylation by catalytic subcomplexes of NuA4 (pNuA4) and SAGA (pSAGA) stimulate chromatin replication. Reactions were performed as described in (A). After DDK treatment, nucleosomes were acetylated using purified pNuA4, pSAGA, and acetyl-coenzyme A. (F) Effect of adding of pNuA4, pSAGA, and INO80 together with FACT and Nhp6 on chromatin replication. (G) Bulk chromatin with or without FACT, Nhp6, INO80, and histone acetylation was assessed by MNase digestion following native agarose gel electrophoresis and ethidium bromide staining. See also Figures S6 and S7.

Figure 5

FACT, Nhp6, INO80, and Histone Acetylation Together Result in In Vivo Rates of Replisome Progression through Chromatin (A) Time course of chromatin replication reactions as conducted in Figure 4A with FACT, Nhp6, INO80, and histone acetylation. (B) Lane profile for the 10-min time point in Figure 6A. (C) Pulse chase experiment to measure replication rates with the same reaction setup as in Figure 6A. For the pulse, the dCTP concentration was reduced to 4 mM. Following a 3-min incubation, unlabeled dCTP was added to a final concentration of 150 mM, and time points were taken every 50 s. (D) Maximum (front) and peak product length for the experiment shown in (C). To derive maximum and bulk leading-strand synthesis rates, data were fitted to a linear regression.

Figure 6

Chromatin Influences Lagging-Strand Size (A) Replication reactions on naked DNA and chromatin were conducted as in Figure 4A with the amounts of Pol α as indicated. For chromatin replication, nucleosomes were acetylated and FACT, Nhp6, and INO80 were added. (B) Replication reactions on chromatin as in (A) with indicated amounts of Pol α. (C) Replication reactions on naked DNA versus chromatin in the presence or absence of FACT and Nhp6. Nucleosomes were not acetylated and INO80 was omitted. (D) Replication reactions on naked DNA as in (A) in the presence or absence of FACT and Nhp6. Pol α was added at the indicated amounts. See also Figure S7.

Figure 7

Nucleosomes Are Re-deposited on Nascent DNA (A and B) Nucleosomes are re-deposited on nascent DNA. MNase digestion of replicated products of naked DNA compared to those of chromatin. Replication products were treated with 100 U MNase and samples were taken every minute, quenched with EGTA, and analyzed on a 1.3% alkaline agarose gel. Replication products were visualized by autoradiography. (C) Model of how chromatin influences origin selection. See the Discussion for details. (D) Model of FACT-dependent replisome progression through chromatin. Parental nucleosomes are in green/light blue. Nucleosomes including newly synthesized histones are in gray/light gray. Double-stranded DNA is in red and single-stranded DNA is in gray. See the Discussion for details.