A new method for high-resolution imaging of Ku foci to decipher mechanisms of DNA double-strand break repair

Abstract

DNA double-strand breaks (DSBs) are the most toxic of all genomic insults, and pathways dealing with their signaling and repair are crucial to prevent cancer and for immune system development. Despite intense investigations, our knowledge of these pathways has been technically limited by our inability to detect the main repair factors at DSBs in cells. In this paper, we present an original method that involves a combination of ribonuclease- and detergent-based preextraction with high-resolution microscopy. This method allows direct visualization of previously hidden repair complexes, including the main DSB sensor Ku, at virtually any type of DSB, including those induced by anticancer agents. We demonstrate its broad range of applications by coupling it to laser microirradiation, super-resolution microscopy, and single-molecule counting to investigate the spatial organization and composition of repair factories. Furthermore, we use our method to monitor DNA repair and identify mechanisms of repair pathway choice, and we show its utility in defining cellular sensitivities and resistance mechanisms to anticancer agents.

Figure 1.

Coupling RNase preextraction with high-resolution microscopy allows detection of NHEJ proteins at laser-induced DNA damage. (A) Localizations of Ku80 and nucleolin were analyzed in undamaged U2OS cells by immunofluorescence without CSK (no CSK), with preextraction using CSK only (CSK), or CSK combined with RNase A (CSK+R). (B) Analysis by immunoblotting of NHEJ proteins in whole-cell extracts (WCE) and in fractions retained after CSK extraction (−R) or CSK+R extraction (+R). (C) U2OS cells were microirradiated, postincubated for 5 min, and fixed without CSK or with preextraction using CSK or CSK+R. (D) Analysis by immunoblotting of expression levels of GFP-FLAG-Ku70 and GFP-FLAG-Ku80 in U2OS stable cells using antibodies against Ku70 and Ku80. (E) U2OS cells stably expressing GFP-FLAG-Ku70 (top) or GFP-FLAG-Ku80 (bottom) were microirradiated, postincubated for 5 min, and preextracted with CSK+R. Immunofluorescence was performed with an anti-GFP antibody. (F) Analysis by immunoblotting of GFP-FLAG-XRCC4 expression levels in U2OS stable cells with an antibody against XRCC4. (G) U2OS cells stably expressing GFP-FLAG-XRCC4 (clone 01) were microirradiated, postincubated for 5 min, and fixed without preextraction (top, no CSK) or preextracted with CSK+R (middle and bottom rows, CSK+R). In this figure, insets represent twofold zoom to highlight microfoci (boxed regions) formed by NHEJ proteins. The position of each nucleus, as defined by DAPI staining, is highlighted by a dotted line. Bars: (white) 10 µm; (green) 1 µm.

Figure 2.

The CSK+R extraction protocol reveals Ku at DNA ends generated by various DNA-damaging agents. (A) U2OS cells were untreated (left columns) or irradiated with 10 Gy of IR (right columns), postincubated for 5 min, and fixed without CSK or with CSK or CSK+R preextraction. Quantifications of Ku foci generated 5 min after irradiation are provided in Fig. 5 A. (B) U2OS cells stably expressing GFP-FLAG-XRCC4 (clone 01) were untreated (NT) or treated with 10 Gy of IR, postincubated for 5 min, preextracted with CSK+R, and processed for immunofluorescence with an anti-GFP antibody to boost the GFP signal. (C) U2OS cells were untreated (top row, DMSO) or treated for 30 min with 100 µM etoposide (middle and bottom rows) before being preextracted with CSK+R and processed for immunofluorescence. (D) U2OS Tet-On cells were transiently transfected with an empty plasmid (pICE; control) or with a plasmid expressing HA-I-PpoI (pICE-HA-NLS-I-PpoI; I-PpoI), and after 24 h, I-PpoI expression was induced by doxycycline for 5 h. Whole-cell extracts were collected and analyzed by immunoblotting (top), and cells on coverslips were preextracted with CSK+R and analyzed by immunofluorescence (bottom). In this figure, insets represent a twofold zoom to highlight Ku80 microfoci (boxed regions). The position of each nucleus, as defined by DAPI staining, is highlighted by a dotted line. Bars: (white) 10 µm; (green) 1 µm.

Figure 3.

CSK+R extraction allows super-resolution imaging of DNA damage sensors and chromatin marks at DNA ends. (A) U2OS cells were treated with IR, postincubated for 5 min, preextracted with CSK+R, and processed for immunofluorescence. Cells were analyzed by high-resolution (top left, conventional) or SIM (top right, 3D-SIM). At the bottom, images are magnifications of the boxed regions highlighted by arrowheads in the top images, highlighting the resolution gain between high-resolution (bottom left) and super-resolution (bottom right) microscopy. (B) Graph showing the fluorescence profile, in arbitrary units (AU), of an individual 3D-SIM Ku focus, corresponding to the dashed line in the bottom right panel of A. The data shown are representative of 30 measured foci in two independent experiments. (C) Frequency distribution of the number of Ku foci per γ-H2AX focus. U2OS cells were treated with 2 Gy of IR, postincubated for 5 min, preextracted with CSK+R, and processed for immunofluorescence. 3D-SIM pictures were acquired and manually analyzed for Ku foci, as represented by a frequency distribution. The data shown are from a single representative experiment out of two repeats. For the experiment shown, n = 535. (D) Ku and γ-H2AX spatial distributions as analyzed by 3D-SIM. Representative 3D-SIM pictures analyzed in C are presented. (E) U2OS cells were untreated (NT) or treated with 10 Gy of IR, postincubated for 5 min, and analyzed by immunofluorescence for NBS1 or γ-H2AX, without CSK preextraction (no CSK) or with CSK+R preextraction. (F) U2OS cells were untreated (NT) or treated with 10 Gy of IR, postincubated for 5 min, preextracted with CSK+R, and processed for immunofluorescence. Insets represent initial and 3.3-fold magnifications from boxed regions. (G) U2OS cells were treated with 2 Gy of IR, postincubated for 5 min, preextracted with CSK+R, and processed for immunofluorescence. 3D-SIM pictures were acquired, and representative foci are presented. The position of each nucleus, as defined by DAPI staining, is highlighted by a dotted line. Bars: (white) 10 µm; (green) 1 µm; (red) 0.2 µm.

Figure 4.

CSK+R extraction permits definition of repair complex composition by single-molecule counting. (A) Fluorescence intensity profiles of a hypothetical Ku focus, containing one or two molecules, plotted over several data acquisition frames. (B) Schematic representation of endogenous tagging of Ku70. RPE-1 cells were infected with an rAAV construct consisting of the GFP-FLAG sequence surrounded by two homology arms targeting HR after the start codon of the XRCC6 (Ku70) gene. (C) Immunoblotting of extracts from untagged cells or cells tagged with GFP on one or both XRCC6 alleles. (D) RPE-1 GFP tagged on both XRCC6 alleles (clone E4) were untreated (NT) or treated with 10 Gy of IR, postincubated for 5 min, preextracted with CSK+R, and processed. The position of each nucleus, as defined by DAPI staining, is highlighted by a dotted line. Bar, 10 µm. (E) RPE-1 cells or two RPE-1 cell clones (E4 and B9) GFP tagged on both XRCC6 alleles were treated with the indicated IR doses, and their survivals were analyzed by clonogenic assay (ns indicates a nonsignificant difference to sensitivity of untagged RPE-1). (F and G) RPE-1 cells endogenously tagged on both XRCC6 alleles (clone E4) were treated with 10 Gy of IR, postincubated for 5 min, preextracted with CSK+R, and processed. A small nuclear volume was continuously imaged, and fluorescence profiles of multiple individual foci were plotted to determine the number of bleaching steps for each focus. The data shown are from a single representative experiment out of two repeats. For the experiment shown, n = 263. (F) Representative fluorescence profiles analyzed in G of focus containing one, two, or more than two Ku molecules. (G) Numbers of bleaching steps, corresponding to the number of GFP-Ku70 molecules per focus, are represented as a frequency distribution.

Figure 5.

CSK+R extraction allows assessment of DSB repair through software-assisted Ku focus quantification. (A) Graph representing a dose–response analysis of the number of Ku foci 5 min after the indicated x-ray doses. U2OS cells were treated, postincubated for 5 min, preextracted with CSK+R, and processed. High-resolution pictures of >20 cells were acquired for each condition and submitted to automated focus detection by Volocity software. The slope given by linear regression of these data (R² = 0.962) equated to 24 Ku foci per Gy, which is less than the expected 30–40 DSBs per Gy per mammalian cell (Ciccia and Elledge, 2010). This apparent discrepancy might reflect some DSBs being blocked and unable to recruit Ku within the time frame of our experiments. Such blocks could represent certain DSBs possessing secondary structures or DNA base adducts, which must be removed for Ku to bind, and/or those residing within chromatin structures, which must be remodeled before Ku and other NHEJ components can gain access to the associated DNA ends. (B) Kinetic analysis of Ku IRIF numbers after 10 Gy of IR. U2OS cells were untreated (NT) or treated with 10 Gy of IR and postincubated for the indicated times. Cells were then processed and analyzed as in A. (C) Kinetic analysis of Ku IRIF numbers after 10 Gy of IR. U2OS cells were preincubated with DMSO, NU7441 (DNA-PKi), and/or KU55933 (ATMi) and then untreated (NT) or treated and analyzed as in B. (D) Impact of ATMi and/or DNA-PKi on survival after IR. U2OS cells were preincubated with DMSO, DNA-PKi, and/or ATMi and treated with the indicated IR doses. Inhibitors were washed away 18 h after treatment, and cell survival was determined by colony formation. For all graphs, each point corresponds to at least three independent experiments, vertical bars correspond to standard deviations, and asterisks indicate a significant difference to DMSO control (*, P < 0.05; **, P < 0.01; ***, P < 0.001).

Figure 6.

CSK+R extraction enables observations into mechanisms controlling repair pathway choice. (A) U2OS cells were preincubated for 1 h with DMSO or ATMi and then treated for 1 h with camptothecin (CPT) before being preextracted with CSK+R and processed for immunofluorescence. Cells in S phase were identified as showing pronounced γ-H2AX staining upon camptothecin treatment. The position of each nucleus, as defined by DAPI staining, is highlighted by a dotted line. Insets represent twofold zoom to highlight Ku80 microfoci (boxed regions). Bars: (white) 10 µm; (green) 1 µm. (B) U2OS cells were preincubated 1 h with DMSO, ATMi, and/or DNA-PKi before being treated with DMSO (NT, not treated) or 1 µM camptothecin for the indicated times (minutes). At the end of the treatment, cells were preextracted with CSK+R and processed for immunofluorescence. Ku foci were quantified as in Fig. 5 A in cells in S phase untreated or treated with camptothecin. NonS columns represent Ku focus numbers in non–S-phase cells treated for 1 h with CPT. Each bar corresponds to at least three independent experiments. (C) Total extracts from MEFs derived from wild-type (WT) or XRCC5^−/− mice were analyzed by immunoblotting. (D) MEF wild type or XRCC5^−/− were preincubated for 1 h with DMSO or 10 µM ATMi and then treated with the indicated doses of camptothecin. Inhibitors and camptothecin were washed away 18 h later, and cell survival was determined by colony formation. Each point corresponds to three independent experiments. Error bars correspond to standard deviations.