Comprehensive Mapping of Histone Modifications at DNA Double-Strand Breaks Deciphers Repair Pathway Chromatin Signatures

Abstract

Double-strand breaks (DSBs) are extremely detrimental DNA lesions that can lead to cancer-driving mutations and translocations. Non-homologous end joining (NHEJ) and homologous recombination (HR) represent the two main repair pathways operating in the context of chromatin to ensure genome stability. Despite extensive efforts, our knowledge of DSB-induced chromatin still remains fragmented. Here, we describe the distribution of 20 chromatin features at multiple DSBs spread throughout the human genome using ChIP-seq. We provide the most comprehensive picture of the chromatin landscape set up at DSBs and identify NHEJ- and HR-specific chromatin events. This study revealed the existence of a DSB-induced monoubiquitination-to-acetylation switch on histone H2B lysine 120, likely mediated by the SAGA complex, as well as higher-order signaling at HR-repaired DSBs whereby histone H1 is evicted while ubiquitin and 53BP1 accumulate over the entire γH2AX domains.

Keywords: 53BP1; ChIP-seq; DNA double-strand breaks; DSB repair; chromatin; histone H1; histone modifications; homologous recombination; non-homologous end joining; γH2AX.

Graphical abstract

Figure 1

BLESS Mapping of AsiSI-Induced DSB (A) Genome browser screenshot representing γH2AX ChIP-seq and BLESS signal at a single DSB located on human chromosome 17. (B) Heatmap representation of the BLESS profile for the 1,211 predicted AsiSI sites in the human genome in DIvA cells untreated (−4OHT) or treated for 4 hr (+4OHT). DSBs are ordered based on decreasing read count in a 1 kb window. (C) Boxplot representing BLESS signals (1 kb window) for the 80 cut (pink) and the 1,131 uncut (gray) AsiSI sites in untreated (−4OHT) or treated (+4OHT) DIvA cells. p values were calculated using two-sample Wilcoxon tests. (D) Average BLESS count profile for the 80 induced DSBs before (−4OHT, gray line) and after induction (+4OHT, black line).

Figure 2

Large-Scale DSB-Induced Chromatin Changes (A) Boxplot representing the ChIP-seq enrichment ratio between 4OHT-treated and untreated DIvA cells (expressed as a log₂ ratio) for 80 DSBs (upper panel) or 80 randomly picked genomic regions (lower panel) over a 1 Mb window. Boxes are colored according to p values (two-sample Wilcoxon tests) and the nature of the change. (B) Left: Average profile of the enrichment between 4OHT-treated and untreated DIvA cells for ubiquitin and H1 over 80 DSBs in a 10 Mb window. Values are expressed as log₂ ratios. Right: Average profile of γH2AX in 4OHT-treated DIvA cells.

Figure 3

Narrow-Scale DSB-Induced Chromatin Changes (A) Boxplot representing the ChIP-seq enrichment ratio between treated and untreated DIvA cells (expressed as a log₂ ratio) for 80 DSBs (upper panel) or 80 randomly picked genomic regions (lower panel) over a 1 kb window. Boxes are colored according to p values (two-sample Wilcoxon tests) and the nature of the change. (B) Average profile of the enrichment between treated and untreated DIvA cells for 12 modifications, showing significant differences between each condition over 80 DSBs in a 10 kb window. Values are expressed as log₂ ratios.

Figure 4

Chromatin Features Associated with HR-Prone DSBs (A) Average profile for RAD51 ChIP-seq, XRCC4 ChIP-seq, and BLESS for 30 HR (yellow) or 30 NHEJ (gray) DSBs. HR and NHEJ DSBs were defined based upon the RAD51/XRCC4 binding ratio (see STAR Methods). (B) Boxplot representing BLESS read count (1 kb window) around 30 HR or 30 NHEJ DSBs. p value was calculated using two-sample Wilcoxon test. (C) Heatmap representing the XRCC4 and DNA Ligase IV ChIP-seq signals on a 40 kb window centered around all AsiSI sites, ordered based on the BLESS level. (D) Same as (A) for DNA Ligase IV ChIP-seq. (E) Boxplot representing the ChIP-seq read count (4 kb window) for each histone modification in untreated cells for 30 HR (yellow) and 30 NHEJ (gray) DSBs. p values were calculated using two-sample Wilcoxon test. ^∗p < 0.05, ^∗∗p < 0.01; p > 0.05 is not significant (ns). (F) Circle plot representing p values (from two-sample Wilcoxon test) when comparing ChIP-seq signal for HR and NHEJ DSB using increasing window size.

Figure 5

HR and NHEJ-Induced Chromatin Changes (A) Boxplot representing the ChIP-seq enrichment ratio between treated and untreated DIvA cells (expressed as a log₂ ratio) for 30 HR (upper panel) or 30 NHEJ (lower panel) over a 1 kb window. Boxes are colored according to p values (two-sample Wilcoxon tests) and the nature of the change. (B) Same as (A) for a 1 Mb window.

Figure 6

Acute Signaling at HR-Prone DSBs (A) Genome Browser screenshots representing ChIP-seq signals for XRCC4, RAD51, γH2AX, ubiquitin, H1, and BLESS for two HR-DSBs (left) and two NHEJ DSBs (right). Data are expressed as read count (from 4OHT-treated samples) for XRCC4, RAD51, and γH2AX ChIP-seq and BLESS. Data for ubiquitin (FK2) and H1 are expressed as a log₂ ratio between treated and untreated cells. (B) Average profile for γH2AX ChIP-seq (read count in OHT treated samples), ubiquitin, and H1 (as a log₂ ratio between treated and untreated cells) for 30 HR (yellow) and 30 NHEJ (gray) DSBs in a 10 Mb window. (C) Boxplot representing ChIP-seq read count from treated cells (γH2AX) or log₂ ratio between treated and untreated cells (ubiquitin [FK2] and H1) for 30 HR (yellow) or 30 NHEJ (gray) DSBs in a 1 Mb window. p values were calculated using two-sample Wilcoxon tests.

Figure 7

53BP1 Preferentially Accumulates at HR-Prone DSBs Specifically in G1 (A) Genome Browser screenshots representing ChIP-seq signals (from 4OHT-treated samples) for 53BP1, γH2AX, and ubiquitin. (B) Boxplot representing 53BP1 enrichment in a 1 Mb window for 30 HR (yellow) and 30 NHEJ (gray) DSBs. p value was calculated using two-sample Wilcoxon test. (C) Average profile for 53BP1 ChIP-seq (read count in 4OHT-treated cells) for 30 HR (yellow) and 30 NHEJ (gray) DSBs in a 1.6 Mb window (upper panel). (D) Heatmap representing 53BP1 ChIP-seq count obtained in G1- and G2-synchronized cells as indicated, on a 1 Mb window surrounding 80 DSBs induced by AsiSI. DSBs are sorted by BLESS read counts. (E) Boxplot representing 53BP1 enrichment in a 1 Mb window for 30 HR (yellow) and 30 NHEJ (gray) DSBs in G1- and G2-synchronized cells as indicated. p values were calculated using two-sample Wilcoxon test.