Research ArticleGENETICS

Single-molecule imaging reveals control of parental histone recycling by free histones during DNA replication

See allHide authors and affiliations

Science Advances  18 Sep 2020:
Vol. 6, no. 38, eabc0330
DOI: 10.1126/sciadv.abc0330
  • Fig. 1 Fluorescent nucleosomes on λ DNA are discretely distributed in a beads-on-a-string manner.

    (A) Crystal structure of the Xenopus nucleosome (Protein Data Bank 1AOI) illustrating the location and type of fluorescent dye (Cy5 or Alexa Fluor 647, abbreviated as A647) used to label histones. (B) SDS–polyacrylamide gel electrophoresis analysis of wild-type (WT) and fluorescently labeled histones and histone octamers. MW, molecular weight. (C and D) Native EMSA (top) and MNase protection assay (bottom) for nucleosomes labeled at H2A-K119CCy5 (C) and H4-E63CA647 (D) reconstituted on λ DNA at increasing DNA:octamer ratios (1:0, 1:40, 1:120, and 1:200). kbp, kilobase pair. (E) Schematic of fluorescent λ nucleosomes immobilized in the microfluidic device for single-molecule imaging. (F and G) Single-molecule imaging of nucleosomes labeled at H2A-K119CCy5 (F) and H4-E63CA647 (G) reconstituted on λ DNA at increasing DNA:octamer ratios (1:0, 1:50, 1:125, 1:200, 1:275, and 1:350). (H and I) Single-molecule quantification of the DNA contour length for nucleosomes labeled at H2A-K119CCy5 (H) and H4-E63CA647 (I) reconstituted on λ DNA at increasing DNA:octamer ratios (1:0, 1:40, 1:120, and 1:200). The DNA length of individual molecules was measured on the basis of SYTOX Orange staining of the DNA (approximately 400 molecules at each histone octamer concentration).

  • Fig. 2 Histone dynamics during DNA licensing in HSS.

    (A) Schematic of the experimental setup for real-time single-molecule imaging of nucleosome dynamics during replication in X. leavis egg extracts. The immobilized DNA is licensed in high-speed supernatant (HSS). Bidirectional replication is initiated upon introduction of nucleoplasmic extract (NPE) supplemented with a fluorescent fusion protein Fen1-KikGR, which decorates replication bubbles. (B and C) Kymograms and corresponding intensity profiles for fluorescent λ nucleosomes during incubation in HSS. Nucleosomes labeled at H2A-K119CCy5 and H2B-T112CA647 (B) show faster loss of fluorescence than nucleosomes labeled at H3-K36CCy5, H3-T80CA647, and H4-E63CA647 (C). (D) Plot showing the mean loss of fluorescent signal for λ nucleosomes during incubation in HSS. More than 100 molecules were analyzed for each histone template. Individual fluorescence decay traces were normalized to background (“0”) and maximum value of fluorescence (“1”). A mean fluorescence value and SD were calculated and plotted for each time point. The mean value traces were then fitted to a single exponential function. (E) Summary of the fluorescence decay rate constants (K) and half-lives (t0.5) extracted from the single exponential fit to the data presented in (D). See table S1 for detailed fitting parameters.

  • Fig. 3 Heterogeneous dynamics of parental histones upon replication fork arrival.

    For each specified outcome, data are presented as kymograms of nucleosome-associated fluorescence (H4-E63CA647; yellow, left), Fen1-KikGR signal indicating nascent DNA (red, middle) and both signals together (merge, right). Time and size scales are presented. The white triangles mark the point of initial encounter between the replication fork and nucleosome. Dotted lines indicate sliding events, whereas solid lines correspond to replication fork stalling. For clarity, a schematic representation of each outcome is shown in gray borders. (A) Histone eviction is manifested by the loss of histone fluorescence at the point of collision with the replication fork. (B) Histone transfer is observed when the histone-associated fluorescence is retained and incorporated into the track of replicated DNA. (C) Histone sliding is observed when the histone-associated fluorescence moves together with the tip of the replication bubble (marked as a dotted white line). (D) Replication fork stalling occurs when nucleosome constitutes a roadblock preventing the replication fork from further movement. It is manifested in the kymogram as an arrested tip of the replication bubble next to a static histone signal (indicated as a solid line).

  • Fig. 4 Secondary outcomes of the replication fork collision with nucleosomes during DNA replication in Xenopus egg extracts.

    For each specified outcome, data are presented as kymograms of nucleosome-associated fluorescence (yellow, left), Fen1-KikGR signal indicating nascent DNA (red, middle) and both signals together (merge, right). Time and size scales are presented. The white triangles mark the point of initial encounter between the replication fork and nucleosome. Dotted lines indicate sliding events, whereas solid lines correspond to replication fork stalling. Replication-independent histone loss is marked in (E) with a white asterisk. (A, C, and E) Histone sliding can terminate in histone eviction (A), histone transfer (C), and replication fork stalling (E). (B, D, and F) Replication fork stalling can lead to histone eviction (B), histone transfer (D), and histone sliding (F).

  • Fig. 5 Effect of free histones on parental histone dynamics at the replication fork.

    (A and B) Quantification of the four basic outcomes of replication fork collision with nucleosomes labeled at H4-E63CA647 (A) and H3-K36CCy5 (B) in regular extracts, extracts depleted of histone H4 and H3 (ΔH4/H3), and depleted extracts supplemented with recombinant histones H4 and H3 (ΔH4/H3 + rH4/H3). n indicates the total number of analyzed collisions. All fork-nucleosomes collisions observed during the 60-min replication reaction were analyzed. Data from at least two biological repeats were pooled in the analysis for each tested condition. (C and D) Western blots used to estimate the concentration of histone H4 (C) and H3 (D) in extracts. (E and F) Tukey plot of replication fork velocities measured in extracts for λ nucleosomes containing H4-E63CA647 (E) and H3-K36CCy5 (F). Values above the box plots indicate the mean replication fork velocity extracted from the Gaussian fit (±SD). The number of values analyzed per dataset (n) is also shown. (G and H) Quantification of histone eviction versus transfer for nucleosomes labeled at H4-E63CA647 (G) and H3-K36CCy5 (H). Analysis for primary (eviction versus transfer) and secondary (slide/stall-eviction versus slide/stall-transfer) outcomes is presented.

  • Fig. 6 Model of parental histone transfer at high and low concentrations of newly synthesized histones.

    (A) At high concentrations of free histones, upon the encounter with the replication machinery, most of the parental histones are evicted from the DNA and released into the histone pool. (B) When the concentration of newly synthesized histones is low, most of the parental histones are recycled at the replication fork. Upon nucleosome disassembly ahead of the replication fork, parental histones are released from the DNA but are kept in the vicinity of the replisome, most likely through a concerted action of histone chaperones, replisome components, and DNA looping. Parental histones are rapidly ushered behind the replication fork where they are deposited onto daughter strands.

Supplementary Materials

Stay Connected to Science Advances

Navigate This Article