POLE3-POLE4 Is a Histone H3-H4 Chaperone that Maintains Chromatin Integrity during DNA Replication

Abstract

Maintenance of epigenetic integrity relies on coordinated recycling and partitioning of parental histones and deposition of newly synthesized histones during DNA replication. This process depends upon a poorly characterized network of histone chaperones, remodelers, and binding proteins. Here we implicate the POLE3-POLE4 subcomplex of the leading-strand polymerase, Polε, in replication-coupled nucleosome assembly through its ability to selectively bind to histones H3-H4. Using hydrogen/deuterium exchange mass spectrometry and physical mapping, we define minimal domains necessary for interaction between POLE3-POLE4 and histones H3-H4. Biochemical analyses establish that POLE3-POLE4 is a histone chaperone that promotes tetrasome formation and DNA supercoiling in vitro. In cells, POLE3-POLE4 binds both newly synthesized and parental histones, and its depletion hinders helicase unwinding and chromatin PCNA unloading and compromises coordinated parental histone retention and new histone deposition. Collectively, our study reveals that POLE3-POLE4 possesses intrinsic H3-H4 chaperone activity, which facilitates faithful nucleosome dynamics at the replication fork.

Keywords: POLE3-POLE4 complex; epigenome stability; histone chaperone; replication-coupled nucleosome assembly.

Graphical abstract

Figure 1

The POLE3-POLE4 Complex Is a Bona Fide H2A-H2B Histone Fold Complex (A) Cartoon depicting human POLE3 and POLE4 proteins and their predicted structural motifs. α helices are represented as squared boxes, while histone fold motifs are indicated by brackets. (B) Modeled POLE3-POLE4 dimer with conserved Phe44 and Phe74 in stick format. Consurf was used to color residues by conservation from blue (most conserved) to red (least conserved). (C) In vitro GST pull-downs of the indicated GST-tagged proteins in the presence of His-POLE4 (left) or His-POLE3 (right). (D) Western blot analysis of FLAG IPs from whole-cell extracts of HeLa TRex-expressing empty FLAG, FLAG-POLE3 WT, or F44D mutant. (E) Western blot analysis of FLAG IPs from whole-cell extracts of HeLa TRex-expressing empty FLAG, FLAG-POLE4 WT, or F74D mutant.

Figure 2

The POLE3-POLE4 Complex Interacts In Vitro with H3-H4 (A) In vitro GST pull-down of the GST-POLE4/His-POLE3 complex in the presence of H3-H4 at the described NaCl concentrations. (B) In vitro GST pull-down of the indicated GST-tagged proteins in the presence of H3-H4. Experiments were performed in 150 mM NaCl. (C) In vitro GST pull-down of the indicated GST-tagged proteins in the presence of H2A-H2B dimers. Experiments were performed in 150 mM NaCl. (D) Limited trypsin digestion of recombinant POLE3-POLE4 complex in the presence of the indicated trypsin concentrations. (E) Analytical gel filtration of POLE3-POLE4, H3-H4, and POLE3-POLE4-H3-H4 complexes in 300 mM NaCl concentrations. (F) SDS-PAGE analysis of gel filtration fractions from (E).

Figure 3

H/D Exchange Mass Spectrometry and Protein Pull-Downs Identify Interaction Domains between POLE3-POLE4 and H3-H4 Complexes (A) Difference plot of hydrogen/deuterium (H/D) exchange data from POLE3 (upper left), POLE4 (upper right), H3 (lower left), and H4 (lower right). A cartoon depicting the domains of the analyzed proteins is shown on the top of the graph. Experimental error is reported in gray, while different colors represent different D₂O incubation times. More exposed and protected regions are located, respectively, on the higher and lower part of the graph. Peptide aa numbers from highly protected or exposed regions of POLE3 and H4 are shown. ±0.5 Dalton difference is indicated with red/blue lines, which represents 98% confidence limit, so that peptides above or below these lines can be considered significantly changed. Difference heatmaps are provided in Figure S3A. (B) Analytical gel filtration of the described protein complexes in 300 mM NaCl concentrations. (C) SDS-PAGE analysis of gel filtration fractions from the described protein complexes.

Figure 4

The POLE3-POLE4 Complex Binds to H3-H4 Dimers and Tetramers and Promotes Tetrasome Formation and Supercoiling In Vitro (A) DSS crosslinking experiments performed in 150 mM NaCl. Lane 4 shows control without crosslinking; lanes 1, 2, and 5 show samples incubated with 1 mM DSS for 30 min at 23°C and resolved by SDS-PAGE and Coomassie staining. Lane 3 was left empty. Two black arrows indicate the molecular weight species identified upon crosslinking of POLE3-POLE4/H3-H4 co-complex. T (tetramers) indicates POLE3-POLE4 binding to H3-H4 tetramers, while D (dimers) indicates POLE3-POLE4 binding to dimers. (B) Western blot analysis of DSS crosslinking experiments performed in 150 mM NaCl. Lane 1 shows control without crosslinking; lane 2 shows sample incubated with 1 mM DSS for 30 min at 23°C. Samples were resolved by SDS-PAGE, transferred to nitrocellulose membranes, and incubated with α-H3, α-H4, α-POLE3, and α-POLE4 antibodies, from left to right. (C) Analytical gel filtration of POLE3-POLE4 complex alone or in combination with H3-H4 or H3(EE)-H4. Experiments were performed in 300 mM NaCl to prevent histone aggregation. (D) SDS-PAGE analysis and Coomassie staining of gel filtration fractions from (C). (E) Tetrasome assembly on linear DNA (Widom 601 sequence) monitored by native PAGE. Lane 2 shows tetrasome preassembled by salt dilution method; lanes 3–5 show linear DNA incubated with the indicated proteins. (F and G) Tetrasome assembly on linear DNA, performed as in (E), with the indicated proteins. (H) Plasmid supercoiling assay resolved by native agarose gel electrophoresis. Lane 1 shows supercoiled control phix174 RF1 DNA; lane 2, phix174 RF1 DNA relaxed by TOPO I; and lanes 3–7, phix174 RF1 DNA incubated, in the presence of TOPO I, with histones and increasing concentrations of POLE3-POLE4. (I) Plasmid supercoiling assay performed as described in (H) using increasing concentrations of POLE3ΔC-POLE4 or POLE3-POLE4.

Figure 5

POLE3 and POLE4 Interact with H3-H4 In Vivo (A) IP of endogenous POLE3 from human HeLa cells performed after CSK-Triton (0.5%) extraction on soluble and DNaseI-digested chromatin fractions. After SDS-PAGE, western blotting was performed using antibodies against the indicated proteins. (B) FLAG IP experiments from HeLa TRex-expressing empty FLAG or FLAG-POLE4 under tetracycline-regulated promoter. Cells were induced with doxycycline for 24 hr and lysed in CSK-Triton 0.5%. FLAG IPs were performed on soluble and DNaseI-digested chromatin fractions. After SDS-PAGE and nitrocellulose transfer, membranes were incubated with antibodies against the indicated proteins. (C) FLAG IP experiments from HeLa TRex-expressing empty FLAG, FLAG-POLE3 WT, or POLE3ΔC mutants under a tetracycline-regulated promoter. Cells induced with doxycycline for 24 hr were lysed in CSK-Triton 0.5%, and FLAG IP was performed on DNaseI-digested chromatin. After SDS-PAGE, western blotting was performed using antibodies against the indicated proteins. (D) HA tag IP experiments from HeLa S3 cells stably expressing HA-H3.1 and HA-H3.3 from soluble and DNaseI-digested chromatin fractions. After SDS-PAGE, western blotting was performed using antibodies against HA tag, POLE3, and POLE4. (E) FLAG IPs from HeLa TRex-expressing empty FLAG or POLE4-FLAG mutants under tetracycline-regulated promoter. Cells were induced with doxycycline for 24 hr and lysed in CSK-Triton 0.5%. FLAG IPs were performed on DNaseI-digested chromatin fractions; after SDS-PAGE and western blotting, membranes were incubated with antibodies against marks specific of newly synthesized or parental histones. (F) Sequential IP experiments performed on empty FLAG or FLAG-POLE3 HeLa SNAP-HA-H3.1-transfected cells. After cell lysis in CSK-Triton 0.5%, chromatin was solubilized with benzonase and incubated with anti-FLAG agarose beads. After subsequent FLAG bead elution in 3xFLAG peptides (1 mg/mL concentration), HA IPs were performed, followed by SDS-PAGE and western blotting using antibodies against the indicated proteins.

Figure 6

Constitutive or Transient Depletion of the POLE3-POLE4 Complex Affects Nucleosome Disruption/Maturation at the Replication Fork (A) FACS analysis of endogenous chromatin-bound RPA from POLE4^+/+ and POLE4^−/− MEFs treated or not with 2 mM hydroxyurea for 2 hr. (B) FACS analysis of chromatin from U2OS cells stably expressing GFP-RPA, subjected or not to HU treatment as described in (A). Cells were transfected with siRNAs against the indicated genes, and chromatin purification and FACS analysis were performed 48 hr later. (C) Bar graph showing percentage of chromatin RPA in cells transfected with the indicated siRNA and treated or not with 2 mM HU for 2 hr. Gates for quantification were selected as shown in (B). Biological triplicates (n = 3) are reported with mean, standard deviation, and p value (^∗p < 0.05, ^∗∗p < 0.01). (D) Bar graph showing the median of PCNA staining intensity (arbitrary units) from U2OS RFP-PCNA transiently transfected with siRNAs against the indicated genes. Biological triplicates (n = 3) are reported with mean, standard deviation, and p value (^∗p < 0.05, ^∗∗p < 0.01). (E) Western blot analysis of cell lysates from HeLa TRex cells expressing empty FLAG, FLAG-POLE3, or FLAG-POLE3ΔC and transfected with the indicated siRNAs in the presence of doxycycline. After SDS-PAGE and nitrocellulose transfer, membranes were incubated with antibodies against the indicated proteins. (F) FACS analysis of endogenous chromatin RPA from HeLa TRex FLAG-POLEΔC cells transfected with the indicated siRNAs and treated or not with 2 mM hydroxyurea for 2 hr.

Figure 7

Transient Depletion of the POLE3-POLE4 Complex Affects Histone Deposition in SNAP Tag H3.1-Expressing Cells (A) In HeLa H3.1-SNAP: at left, quench-chase-pulse experiment to follow new H3.1; at right, pulse-chase experiment to follow parental H3.1. A quenching step labels all pre-existing histones with a non-fluorescent dye. A chase step allows synthesis and deposition of new unlabeled H3.1-SNAP. A pulse using the fluorophore TMR (red) labels available H3.1-SNAP. EdU incorporation at the end of the assay allows the detection of replicated DNA (green). A Triton extraction step is performed prior to fixation to eliminate soluble histones and analyze chromatin-bound H3.1. (B) Representative images of new (left) or parental (right) H3.1 (TMR, red) and replication sites (EdU, green) in control, POLE1-depleted, POLE3-depleted, or POLE4-depleted conditions. Scale bars, 10 μm. (C) Quantification of TMR fluorescence signal per nucleus normalized to control mean for new (left) or parental (right) H3.1 in control, POLE1-depleted, POLE3-depleted, or POLE4-depleted conditions (n = 3). For new H3.1 (left), only cells in S phase were quantified.