Studying DNA damage repair kinetics requires a system to induce lesions at defined sub-nuclear regions. We describe a method to create localized double-stranded breaks using a laser-scanning confocal microscope equipped with a 405 nm laser and provide automated procedures to quantify the dynamics of repair factors at these lesions.
The DNA Damage Response (DDR) uses a plethora of proteins to detect, signal, and repair DNA lesions. Delineating this response is critical to understand genome maintenance mechanisms. Since recruitment and exchange of proteins at lesions are highly dynamic, their study requires the ability to generate DNA damage in a rapid and spatially-delimited manner. Here, we describe procedures to locally induce DNA damage in human cells using a commonly available laser-scanning confocal microscope equipped with a 405 nm laser line. Accumulation of genome maintenance factors at laser stripes can be assessed by immunofluorescence (IF) or in real-time using proteins tagged with fluorescent reporters. Using phosphorylated histone H2A.X (γ-H2A.X) and Replication Protein A (RPA) as markers, the method provides sufficient resolution to discriminate locally-recruited factors from those that spread on adjacent chromatin. We further provide ImageJ-based scripts to efficiently monitor the kinetics of protein relocalization at DNA damage sites. These refinements greatly simplify the study of the DDR dynamics.
Cells are constantly exposed to endogenous and exogenous sources of DNA damage that threaten their genomic integrity. The DDR is an ensemble of signaling pathways that detect, signal, and repair DNA lesions to sustain genome stability. At DNA double-stranded breaks (DSBs), the DDR occurs mainly on two complementary platforms: γ-H2A.X-labeled chromatin and a resected single-stranded DNA (ssDNA) region typically coated with the ssDNA-binding complex RPA1,2.
UV-laser micro-irradiation of cells pre-sensitized with the thymidine analogue 5-bromo-2'deoxyuridine (BrdU) or the bisbenzimide ethoxide trihydrochloride (BBET, Hoechst 33342) DNA dye creates a mixture of DNA lesions including single-stranded breaks (SSBs) and DSBs which elicit a localized DDR on both the chromatin and ssDNA platforms3,4. Previous work showed that recruitment of genome maintenance factors to these two distinct platforms at DSB can be discriminated using high energy UV-A lasers (335-365 nm) combined with IF2,5. Microscopes equipped with such lasers are costly as they require a high energy laser and dedicated UV-A transmitting objectives making them far less prevalent in academic and pharmaceutical settings than laser-scanning confocal microscopes with 405 nm laser lines. The study of protein recruitment and exchange at micro-irradiated sites is also precluded by the laborious manual image analysis required to describe the dynamic behavior of genome maintenance factors.
Here, we show that the pre-sensitization of cells with BrdU or BBET nucleic acid stain followed by micro-irradiation using a 405 nm laser on a common confocal microscope allows the monitoring of genome maintenance factor dynamics at DNA lesions. Using γ-H2A.X or RPA complex subunits as platform markers together with z-stacking for greater depth of field and deconvolution for improved resolution allows the experimenter to discriminate factors that are locally recruited to DSBs from those that spread to large chromatin domains surrounding the initial lesion. This sub-classification according to distinct intra-nuclear compartments helps refine the potential roles of uncharacterized proteins recruited to micro-irradiated sites. Moreover, we provide convenient protocols and optimized pipelines to rapidly analyze the dynamics of genome maintenance factors using the open-source software Fiji (an ImageJ distribution)6,7,8. These refinements to current micro-irradiation methods render the study of the DDR possible in virtually any laboratory setting.
1. Pre-sensitization of Cells
2. Micro-irradiation of Cells
3. Live-imaging of Fluorescently-labeled Proteins
4. Recruitment Kinetics Analysis
5. IF Protocol
Following micro-irradiation, cells are allowed to recover for a specific period of time depending on the nature of the genome maintenance proteins1,12. DSBs will be processed by nucleases, most extensively during the S and G2 phases of the cell cycle, to create a limited region of ssDNA which is rapidly coated by RPA and other genome maintenance factors9. This ssDNA region is surrounded by chromatin which is extensively modified during the DDR to regulate the recruitment and exchange of other DDR factors13. Protein recruitment to micro-irradiated regions can be monitored by IF (Figure 1A). Typically, chromatin-associated factors (e.g., γ-H2A.X, 53BP1, etc.) are detectable at earlier (1–5 min) time points than ssDNA-bound proteins because nuclease-mediated resection of DSB ends is required to generate the RPA-ssDNA platform (≥10 min). To facilitate image analysis and publication, we have devised a Fiji script (The Outliner)10 which automatically outlines cell nuclei using a DNA dye-associated channel (e.g., DAPI), making it easier to distinguish micro-irradiation stripes and individual cells by leaving out uninformative DNA staining from the final images while preserving nuclear definition (Figure 1B).
γ-H2A.X and RPA complex subunits function as markers for chromatin- and ssDNA-associated factors, respectively. RPA-ssDNA appears as punctate foci surrounded by large fluorescent chromatin stripes decorated by the γ-H2A.X antibody (Figure 1C). Co-localization with these markers can be used to precisely define the position of genome maintenance factors at DNA lesions. For instance, 53BP1, a chromatin-associated protein which controls the DSB repair pathway choice, co-localizes well with the γ-H2A.X signal13. Conversely, PRP19, an E3 ubiquitin ligase that functions on RPA-ssDNA to promote ATR activation and DNA repair is recruited as punctate foci that closely match the RPA32 distribution14,15,16,17 (Figure 1D). Of note, some factors may also be recruited on the chromatin adjacent to the initial lesions but do not spread to large domains surrounding the DSBs like γ-H2A.X and 53BP1. In such cases, these factors would appear as punctate foci but would not perfectly co-localize with ssDNA-bound proteins.
For live cell imaging, cells expressing a POI tagged with a fluorescent marker can be micro-irradiated and imaged in real-time to obtain highly detailed recruitment kinetics (Figure 2A–C). Multiple different proteins can be observed simultaneously provided that their fluorescent labels can be spectrally separated from each other. To facilitate analysis of time series, we have created two Fiji scripts which allow the user to produce movie files containing time-series information (Movie Maker, see movies of Figure 2C as Online Supplementary Material) and to quantify protein recruitment at micro-irradiated sites (Damage Analyzer)10. Instead of the time-consuming manual image curation normally required to describe the behavior of a protein, the Damage Analyzer script guides the user through streamlined steps which monitor signal enrichment at micro-irradiated sites, while automatically correcting for cell movement, image drift, background signal, and bleaching at all time points of the experiment. The generated data can then be easily compiled into a spreadsheet application and plotted as needed (Figure 2D).
Figure 1: Localization of genome maintenance factors to the chromatin and RPA-ssDNA platforms at micro-irradiation-induced lesions. (A) Laser micro-irradiation in pre-sensitized cells leads to the production of DNA double-stranded breaks, which are resected by endo- and exo-nucleases to generate ssDNA. The ssDNA sub-compartment can be visualized as punctate foci using antibodies against Replication Protein A or other ssDNA-localized factors. In contrast, proteins or modifications that spread to large chromatin domains appear as large stripes at the micro-irradiated site similar to the pattern observed with antibodies targeting the phosphorylated histone variant H2A.X (γ-H2A.X). (B) Cells pre-sensitized with BBET or BrdU were micro-irradiated, pre-extracted, fixed, and stained for the indicated proteins. 12-bit images were collected at 1X zoom and processed using the Outliner Fiji script. (C) Z-stacking of BBET or BrdU pre-sensitized cells allows the resolution of ssDNA- and chromatin-associated factors (maximum intensity Z-projection shown). (D) Using this method, 53BP1 and PRP19 can be classified as chromatin- and ssDNA-associated genome maintenance factors, respectively. Scale bars = 10 µm; images in D are all at the same scale. Please click here to view a larger version of this figure.
Figure 2: Live-imaging of RPA complex recruitment to sites of DNA damage. (A) Stable or transient transfection of plasmids encoding candidate proteins tagged with fluorescent reporters followed by micro-irradiation of transfected cells allows the monitoring of protein recruitment to DNA lesions in real-time. (B) Cells were transfected with a pDEST47-RPA32-GFP plasmid and lysed 24 h later. Protein expression was confirmed by immunoblotting. (C) Cells transfected with pDEST-SFB EGFP or pDEST47 RPA32-GFP plasmids pre-sensitized with BrdU were micro-irradiated and imaged at 12-bit at 1X zoom once per minute post-irradiation. Timing is in minutes post-irradiation. The 00:00 time point image was taken prior to micro-irradiation. Movie files were created using the Movie Maker Fiji script and key frames are shown. Scale bars = 10 µm. (D) Recruitment of RPA32-GFP to micro-irradiation stripes was quantitatively monitored using the Damage Analyzer Fiji script. Output data were plotted using Microsoft Excel. The black line is the mean enrichment-fold of RPA32-GFP at micro-irradiated sites relative to the pre-irradiation signal and the error bars represent the standard error of the mean of 15 independently micro-irradiated cells. Please click here to view a larger version of this figure.
Movie 1: Micro-irradiation time-lapse of SFB-GFP expressing cells. Cells transiently transfected with a pDEST-SFB GFP plasmid were pre-sensitized 24 h with BrdU, micro-irradiated, and imaged once per minute for 29 min. Please click here to download this movie.
Movie 2: Micro-irradiation time-lapse of RPA32-GFP expressing cells. Cells transiently transfected with a pDEST47 RPA32-GFP plasmid were pre-sensitized 24 h with BrdU, micro-irradiated, and imaged once per minute for 29 min. Please click here to download this movie.
Using the methods outlined above, 405 nm laser micro-irradiation of cells pre-sensitized with BBET or BrdU allows the localized generation of DNA lesions including DSBs within the nuclei of adherent human cells. Kinetics of DDR at these DSBs are similar to those generated with other methods in eukaryotic cells9,18,19. DSB resection at micro-irradiated sites leads to ssDNA-generation which can be followed using a RPA32-targeting antibody or an RPA32-GFP fusion. Chromatin-associated factors, on the other hand, can be located using γ-H2A.X as reference. Some chromatin-associated factors may not spread away from the initial DSB as much as γ-H2A.X and will appear as punctate foci that may be distinct from ssDNA-binding proteins foci.
Once DSBs are generated, protein recruitment and clearance from damage sites can be monitored by IF on fixed cells. Performing these experiments requires only a suitable IF antibody against the POI. If IF-compatible antibodies are unavailable, plasmids can be engineered carrying genes of interest in frame with specific epitopes (e.g.: FLAG, His-tag) or fluorescent proteins. Transient or stable transfection/transduction of these plasmids into adherent cells can then be used to determine whether a specific factor is actively recruited to damaged DNA. Overexpression of proteins can lead to mis-localization and may artificially boost their recruitment to micro-irradiated stripes. To obtain results that are more representative of endogenous proteins, CRISPR-Cas9 technology can now be used routinely to tag any gene of interest with suitable epitopes20,21.
The Python scripts provided in the micro-irradiation Analysis Fiji Plugin facilitate the visualization, presentation, publication, and analysis of DDR proteins kinetics. This analysis package is open-source and functions irrespective of the microscope platform used to perform micro-irradiation experiments and image acquisition. All that is required is that the images taken on the microscopy platform be saved as Bio-Formats compatible files22. Since background, drifting, and bleaching corrections are automatically performed, analysis of protein levels at DNA damage sites with the provided scripts is straightforward and very fast compared with the typical manual signal quantification performed with microscope image acquisition software.
As described in the protocol, users must carefully configure their respective laser-scanning confocal microscope systems in order to achieve the appropriate amount of damage and induce DSBs that follow normal DDR kinetics. Indeed, the quantity and type of DNA lesions generated by micro-irradiation will depend on the laser wavelength and dose as well as on the pre-sensitization methods. Users need to keep in mind that even though DSBs are generated by the conditions described here, 405 nm laser micro-irradiation of photosensitized cells will also generate a mixture of other DNA lesions, including cyclopyrimidine dimers (CPDs), SSBs, and oxidized bases, that will vary between microscope systems and settings23,24,25. Thus, a careful monitoring of the lesions at micro-irradiated sites needs to be performed to obtain a comprehensive picture of the damage generated and the response elicited. Importantly, to study the response to DSBs, each microscope system should be optimized by monitoring the kinetics of canonical DNA damage proteins and post-translational modifications at micro-irradiated sites. For instance, γ-H2A.X and the RPA complex subunits (RPA70, RPA32 or RPA14) can be used conveniently as they are quite easy to detect and rapidly accumulate at DSBs. As a rule of thumb, γ-H2A.X signal should be visualized as stripes at early (5 min) and late (up to 3–4 h) time points post-damage whereas RPA should start accumulating as punctate foci approximately 10-15 min post-micro-irradiation. If a pan-nuclear γ-H2A.X signal is observed, the amount of energy received by the cells is non-physiological and will result in rapid cell death.
There are some limitations that have to be considered when using a 405 nm laser for the study of specific DNA repair pathways. For example, although the BBET-405 nm combination readily produces a combination of SSBs and DSBs, it does not generate 6–4 photoproducts (6–4 PPs) and consequently, the kinetics of recruitment of nucleotide excision repair proteins are different than those observed using UV-C micro-irradiation. The kinetics and extent of recruitment of various proteins involved in the response to DSBs may also vary according to the DNA damaging method that is being used23. Because of these variations, we recommend that users adapt their systems to maximize the type of lesions that are potentially recognized by their POIs. Although such systems are less broadly available, using other types of lasers including UV-A (337–355 nm) or femtosecond near-infrared (800 nm) along with different sensitization reagents (e.g.: tri-methyl psoralen) can allow users to generate more specific DNA lesions depending on the DNA repair pathways under study24,26.
Whenever new conditions are being tested, it is convenient to use fluorescently-labeled well-established DNA repair proteins to test doses, wavelengths, and pre-sensitizer combinations, and to ensure that the lesion of interest is efficiently generated. For example, PARP1 and XRCC1 can function as markers for SSB and base-excision repair, XPA is a good indicator of nucleotide excision repair, and 53BP1 will be recruited to DSBs and function in non-homologous end-joining (references23,25,27,28). Antibodies that recognize modifications associated with specific types of DNA damage can also be used to quantify the levels of DNA adducts or breaks. It is thus possible to monitor oxidized bases, cyclopyrimidine dimers, 6–4 PPs, SSBs (using poly-ADP ribose antibody), and DSBs (γ-H2A.X and 53BP1) by IF and determine the parameters that will achieve the highest level of the desired lesions23,27,29. With careful monitoring, users should be able to study their repair pathway of interest using micro-irradiation.
Altogether, the use of wide-spread laser-scanning confocal microscopes equipped with 405 nm laser lines to generate localized DSBs, combined with platform-specific DNA damage markers to sub-classify genome maintenance factors, and streamlined data processing with the help of the micro-irradiation Analysis Fiji plugin, should render the study of DDR factors dynamics more accessible than ever before to both experienced and novice investigators of the genome maintenance field.
The authors have nothing to disclose.
This work was supported by the Natural Sciences and Engineering Research Council of Canada [Discovery grant 5026 to A.M] and by a Next-Generation of Scientists Scholarship from the Cancer Research Society [21531 to A.M]. We thank Jean-François Lucier for providing excellent quality control of the Fiji Python scripts and advice on statistical analysis. We thank Dr. Stephanie A. Yazinski for the IF fixation solution recipe. We thank Paul-Ludovic Karsenti for help with laser power measurements.
Olympus FLUOVIEW FV3000 resonant laser scanning confocal microscope | Olympus | ||
FluoView FV3000 software (version 2.1) | Olympus | ||
405 nm laser | Coherent | Radiation source | |
Microscope incubator AIO-UNO-OLYUS-1 | Okolab | OKO-UNO-1 | |
60X /1.4 Objective | Olympus | Oil immersion objective | |
8-well culture slides on coverglass (X-well) | Sarstedt | 94.6190.802 | |
U2OS osteosarcoma human cell line | ATCC | HTB-96 | Stable cell lines |
4’,6-Diamidino-2-Phenylindole (DAPI) | ThermoFisher | D1306 | Caution toxic |
5-bromo-2′-deoxyuridine (BrdU) | Sigma-Aldrich | 59-14-3 | |
Paraformaldehyde | Sigma-Aldrich | 30525-89-4 | Caution toxic |
Bisbenzimide H 33342 (Hoechst 33342) | ThermoFisher | 62249 | |
Bovine serum albumin (BSA) | ThermoFisher | BP1600-100 | |
Trypsin EDTA 0.1% | ThermoFisher | 15400-054 | |
Triton X-100 | BioShop | TRX777.100 | |
Tween-20 | ThermoFisher | BP337-500 | |
Sucrose | BioShop | SUC507 | |
JetPRIME transfection reagent | Polyplus | 114-07 | |
ProLong Diamond Antifade mountant with DAPI | ThermoFisher | P36962 | |
DMEM 1X, + phenol red | Wisent | 319-005-CL | |
DMEM 1X, – phenol red | Wisent | 319-051-CL | |
Fetal bovine serum (FBS) | Wisent | 080-150 | |
Mouse anti-RPA32 antibody | Abcam | ab2175 | 1:500 for IF |
Rabbit anti-γ-H2A.X antibody | Cell Signaling | 9718S | 1:500 for IF |
Rabbit Anti-53BP1 antibody | Cell Signaling | 4937S | 1:500 for IF |
Rabbit Anti-PRP19 antibody | Abcam | ab27692 | 1:500 for IF |
Goat anti-rabbit IgG Alexa Fluor 647 | Cell Signaling | 4414S | 1:250 for IF |
Goat anti-mouse IgG Alexa Fluor 488 | Cell Signaling | 4408S | 1:250 for IF |
Fiji image analysis open-source software | https://fiji.sc | ||
1918-K Power meter with a 918D-SL-0D3 sensor | Newport |