Epigenetic control of topoisomerase 1 activity presents a cancer vulnerability

Abstract

DNA transactions introduce torsional constraints that pose an inherent risk to genome integrity. While topoisomerase 1 (TOP1) activity is essential for DNA supercoil removal, the aberrant stabilization of TOP1:DNA cleavage complexes (TOP1ccs) can result in cytotoxic DNA lesions. What protects genomic hot spots of topological stress from excessive TOP1cc accumulation remains unknown. Here, we identify chromatin context as an essential means to coordinate TOP1cc resolution. Through its ability to bind poly(ADP-ribose) (PAR), the histone variant macroH2A1.1 facilitates TOP1cc repair factor recruitment and lesion turnover, thereby preventing DNA damage in response to transcription-associated topological stress. The alternatively spliced macroH2A1.2 isoform is unable to bind PAR or protect from TOP1ccs. Impaired macroH2A1.1 splicing, a frequent cancer feature, was predictive of increased sensitivity to TOP1 poisons in a pharmaco-genomic screen in breast cancer cells, and macroH2A1.1 inactivation mirrored this effect. We propose macroH2A1 alternative splicing as an epigenetic modulator of TOP1-associated genome maintenance and a potential cancer vulnerability.

Fig. 1. MacroH2A1.1 defines TOP1 permissive chromatin domains.

a Heatmaps and profile plots for FLAG-macroH2A1.1 CUT&RUN signal centered on TSS-proximal (TSS) and TSS-distal (non-TSS) TOP1 peaks in MDA-MB-231 macroH2A1.1 KO cells reconstituted with FLAG-macroH2A1.1. Cells without FLAG-macroH2A1.1 served as a negative control (no FLAG), a representative CUT&RUN experiment is shown. b IGV browser shots of distinct macroH2A1.1 chromatin environments associated with TOP1 domains (left) or facultative heterochromatin (HC) marked by H3K27me3 (right). c Jaccard indices for observed peak overlap between a feature of interest (top) and a reference feature (bottom), or randomly shuffled reference feature peaks of equal size (permuted). Values on the y-axis represent the intersection divided by the union in base pairs. d Western blot for the indicated proteins in nuclear lysates (input) or anti-FLAG IP samples from parental and FLAG-macroH2A1.1 knock-in 293 cells (F-1.1) in the presence or absence of CPT and the PARPi Olaparib. Similar results were obtained in a second, independent experiment. * ubiquitinated macroH2A1. Source data are provided as a Source Data file.

Fig. 2. MacroH2A1.1 protects from TOP1cc accumulation.

a Schematic for RADAR and TOP1 CAD-Seq assays. b Representative TSS profile plots for TOP1 CUT&RUN (green) and TOP1cc CAD-Seq (purple: IP, gray: input) in MDA-MB-231 cells. Y axes depict Z-normalized read counts. c TSS-associated TOP1cc turnover following CPT-induced damage in the presence or absence of macroH2A1.1. Schematic depicts CPT treatment times used to generate the damage. TOP1cc turnover (ΔTOP1cc) is defined as the ratio of prolonged damage (30’ CPT) over steady state (5’ CPT). Profile plot depicts Z-normalized ΔTOP1cc log₂ ratios for sh-RFP control (black) and macroH2A1.1 knockdown (1.1 KD, blue). A ΔTOP1cc log₂ ratio >0 reflects an accumulation of TOP1cc in response to damage. Two independent 30’ CPT and 5’ CPT TOP1 CAD-Seq replicates were combined for average ΔTOP1cc ratios, see Supplementary Fig. 2d for 30’ CPT and 5’ CPT TOP1 CAD-Seq profiles. d Comparison of TSS-proximal ΔTOP1cc ratios from (c), separated by RNA-Seq-derived gene expression quartiles; Q1: bottom 25%, Q4: top 25%. Boxplots show the distribution of mean ΔTOP1cc signal (TSS ± 250 bp) within each expression quartile, each data point represents the TSS of one gene. P values are based on two-sided Mann–Whitney U test for the indicated, pairwise comparisons between TSS quartiles. Box limits represent upper and lower quartiles, whiskers minimum to maximum and center lines represent the median. Similar results were obtained when analyzing replicate experiments independently. e TOP1 peak-associated ΔTOP1cc ratios as in (c). TOP1 peaks were separated into top (1.1^high) and bottom tertiles (1.1^low) based on macroH2A1.1 (FLAG-1.1) enrichment, see TOP1-centered FLAG-1.1 heatmap. f TOP1 RADAR assay with genomic DNA from MDA-MB-231 cells expressing the indicated shRNAs, prior to and at the indicated timepoints after CPT treatment (1 µM, 30 min). See Supplementary Fig. 2f for a quantification of two independent experiments. Source data are provided as a Source Data file.

Fig. 3. TOP1cc clearance depends on the macroH2A1.1 PAR binding domain.

a TOP1 RADAR assay with genomic DNA from macroH2A1.1 knockout (1.1KO) MDA-MB-231 cells (–) and 1.1KO cells reconstituted with WT (+1.1WT) or G224E mutant FLAG-macroH2A1.1 (+1.1GE), prior to and at the indicated timepoints after CPT treatment as in Fig. 2f, a representative of three independent experiments is shown. b Quantification of RADAR analysis in (a), depicting the percentage of remaining TOP1cc relative to 0’ after CPT treatment (n = 3 independent replicates), values are expressed as mean and SD. P values are based on two-sided Student’s t-test. Source data are provided as a Source Data file.

Fig. 4. XRCC1 recruitment to TOP1ccs depends on macroH2A1.1.

a Western blot for the indicated proteins in nuclear lysates (input) or IP lysates from parental (P) and FLAG-macroH2A1.1 (F-1.1) or FLAG-macroH2A1.2 (F-1.2) knock-in 293 cells in the presence or absence of PARPi. A representative of two independent experiments is shown. b Quantification of XRCC1 foci in MCF7 WT and macroH2A1.1 KO cells in the presence or absence of CPT treatment (1 µM, 30 min), y-axis depicts foci per nucleus (n > 350, see source data for exact n). Similar results were obtained in a second, independent experiment, see Supplementary Fig. 4a for representative images. c Western blot for the indicated proteins in macroH2A1.1 KO (1.1 KO) MCF7 cells reconstituted with empty vector (EV), FLAG-macroH2A1.1 (1.1WT) or FLAG-macroH2A1.1 G224E (1.1GE). One of two independent experiments is shown. d XRCC1 IF in cells from (c) treated with CPT (1 µM, 30 min) or DMSO, scale bar: 10 µm. Foci were quantified as in (b), n > 300 nuclei per sample, see source data for exact n. A representative of two independent 1.1 KO reconstitutions is shown. For violin plots in (b) and (d), center lines (red) reflect the median, white dotted lines depict upper and lower quartiles, and p values are based on two-sided Mann-Whitney U test for the indicated, pairwise comparisons. e Western blot for the indicated proteins in chromatin-bound fractions from WT or TDP1 KO HCT116 cells expressing a control shRNA (C) or sh-macroH2A1.1 (1.1). Cells were treated with DMSO or 1 µM CPT for 30 min, in the presence or absence of PARPi. See Supplementary Fig. 4g for whole cell extracts. Similar results were obtained in a second, independent experiment. Source data for (a–e) are provided as a Source Data file. f CUT&RUN NGS profile plots for XRCC1 or IgG in MDA-MB-231 cells expressing sh-RFP (sh-ctrl) or sh-macroH2A1.1 (sh-1.1), treated with DMSO or 1 µM CPT for 30 min. Averaged RPKM profiles from two independent experiments are shown, centered on TOP1 peaks. g Profile plots of CPT-treated cells from (f), centered on the TSS and separated based on RNA-Seq-derived gene expression as in Fig. 2d.

Fig. 5. MacroH2A1.1 promotes TOP1cc repair at sites of nascent transcription.

a Schematic of U2OS cell-based reporter for transcription-associated DNA damage repair. An MS2 repeat-containing transcript is induced by Dox and detected with YFP-MCP, PARG inhibitor (PARGi) treatment served to stabilize PAR chains. b YFP-MCP, XRCC1 and TOP1 IF images of a representative nucleus 5 h after Dox-treatment, PARGi was added for 30 min prior to analysis. Squares depict the MS2 site or a control region used for analyses in (d–f), scale bar: 10 µm. Similar results were obtained in three independent experiments. c TOP1 CAD-qPCR at the MS2 locus in the presence or absence of Dox in MS2 reporter cells expressing sh-RFP (sh-ctrl) or sh-macroH2A1.1 (sh-1.1). TSS-proximal and upstream qPCR primer pairs are shown in (a). IgG was used as a negative control. The y-axis depicts relative enrichment, normalized to TOP1cc levels in untreated sh-RFP cells, average and SEM are from three independent experiments. d Averaged signal intensities for the indicated proteins following MS2 induction as in (a), n = 82 MS2 or matched control loci. e XRCC1 intensity distributions 5 h after Dox/PARGi treatment in cells transfected with non-targeting control (si-ctrl), or macroH2A1.1 siRNA (si-1.1). For each MS2 site, mean fluorescence intensity was measured at the MS2 peak (white circle) or a corresponding control region as defined in (b) (n = 82 MS2 or matched control loci). A representative of two independent experiments is shown in (d, e); si-ctrl cells were used for analyses shown in (d). f Average TOP1 intensities and corresponding violin plots as in (e) (n > 55 MS2 loci, see source data for exact n). TOP1 signal was measured following 1 h of Dox treatment to avoid potential changes in TOP1 retention due to prolonged repair activity. For all violin plots, center lines (red) represent the median, white dotted lines depict upper and lower quartiles, p values are based on two-sided Mann-Whitney U test for the indicated, pairwise comparisons. Scale bars in (d–f) are 1 µm. Source data for (c, e, f) are provided as a Source Data file.

Fig. 6. macroH2A1.1 protects from CPT-induced DNA damage.

a Alkaline comet assay in WT and 1.1 KO MCF7 cells treated with 1 µM CPT or DMSO for 1 h; y-axis depicts Olive tail moment (n > 50 nuclei per sample, see source data for exact n), a representative of two independent experiments is shown, scale bar: 100 µm. b γH2AX IF in EdU-negative (EdU^–) macroH2A1.1 KO (1.1 KO) MCF7 cells reconstituted with empty vector (EV), FLAG-macroH2A1.1 (1.1WT) or FLAG-macroH2A1.1 G224E (1.1GE). Cells were treated with CPT (1 µM, 30 min) with or without 2 h release. Representative EdU^– nuclei are shown, scale bar: 10 µm. Violin plots depict γH2AX foci per nucleus (n > 150 nuclei per sample, see source data for exact n). For all violin plots, center lines (red) reflect the median, white dotted lines depict upper and lower quartiles, p values are based on two-sided Mann-Whitney U test. One of two independent experiments is shown. See Supplementary Fig. 7 for EdU⁺ cells. Source data are provided as a Source Data file.

Fig. 7. macroH2A1.1 drives TOP1i resistance in cancer cells.

a Western blot for the indicated proteins in MDA-MB-453 (gray) and NCI60 (black) breast cancer cell lines. TNBC cell lines are italicized. A representative of two independent experiments is shown. b Drug activity levels in NCI60 breast cancer cell lines based on data from the National Cancer Institute NCI60 drug screening program, arranged by increasing macroH2A1.1 expression, n: number of compounds tested per drug target. Box plots depict average compound activity Z score distributions for each cell line, each data point represents one drug. The majority of compounds were tested in at least two independent experiments, see https://discover.nci.nih.gov/cellminer/ for details. Box limits represent upper and lower quartiles, whiskers the 10 to 90 percentile range, and center lines the median. P values are based on the two-sided Pearson’s Correlation Coefficient of median drug activity scores. c Cell viability of the indicated cell lines in response to CPT treatment measured by MTT assay, macroH2A1.1^high cells are in gray, macroH2A1.1^low cells in blue. Data are presented as mean and SD (n = 4 independent replicates); p values are based on two-sided Student’s t-test. d TOP1 RADAR assay for the indicated cell lines as in Fig. 2f. Similar results were obtained in three independent experiments. e Western blot in MCF7 cells expressing siRNAs against macroH2A1.1 (si-1.1), macroH2A1.2 (si-1.2) or a control sRNA (si-ctrl). A representative of two independent experiments is shown. f Clonogenic survival of cells from (e) in response to the indicated drug combinations. Survival was normalized to untreated cells for each siRNA transfection. Representative images are shown, scale bar: 1 cm. Data are presented as mean and SD (n = 3 independent replicates); p values are based on two-sided Student’s t-test. g Cell viability in the indicated cell lines in response to CPT treatment in the presence or absence of PARPi, measured by MTT after 10 days of clonogenic growth, data are presented as mean and SD (n = 3 independent replicates); p values are based on two-sided Student’s t-test. Source data for a-g are provided as a Source Data file. h Kaplan–Meier survival analysis of TCGA ovarian cancer patient subgroups where treatment regimens contained Topotecan (n = 119 patients) or taxol (n = 821 patients). Patients were stratified by macroH2A1.1 mRNA expression based on an isoform-specific Affymetrix microarray probe (214500_at), the bottom 25% were considered macroH2A1.1 low expressors.