Assessment of Global DNA Double-Strand End Resection using BrdU-DNA Labeling coupled with Cell Cycle Discrimination Imaging

Summary

In the present protocol, we demonstrate how to visualize DNA double-strand end resection during S/G2 phase of the cell cycle using an immunofluorescence-based method.

Abstract

The study of the DNA damage response (DDR) is a complex and essential field, which has only become more important due to the use of DDR-targeting drugs for cancer treatment. These targets are poly(ADP-ribose) polymerases (PARPs), which initiate various forms of DNA repair. Inhibiting these enzymes using PARP inhibitors (PARPi) achieves synthetic lethality by conferring a therapeutic vulnerability in homologous recombination (HR)-deficient cells due to mutations in breast cancer type 1 (BRCA1), BRCA2, or partner and localizer of BRCA2 (PALB2).

Cells treated with PARPi accumulate DNA double-strand breaks (DSBs). These breaks are processed by the DNA end resection machinery, leading to the formation of single-stranded (ss) DNA and subsequent DNA repair. In a BRCA1-deficient context, reinvigorating DNA resection through mutations in DNA resection inhibitors, such as 53BP1 and DYNLL1, causes PARPi resistance. Therefore, being able to monitor DNA resection in cellulo is critical for a clearer understanding of the DNA repair pathways and the development of new strategies to overcome PARPi resistance. Immunofluorescence (IF)-based techniques allow for monitoring of global DNA resection after DNA damage. This strategy requires long-pulse genomic DNA labeling with 5-bromo-2′-deoxyuridine (BrdU). Following DNA damage and DNA end resection, the resulting single-stranded DNA is specifically detected by an anti-BrdU antibody under native conditions. Moreover, DNA resection can also be studied using cell cycle markers to differentiate between various phases of the cell cycle. Cells in the S/G2 phase allow the study of end resection within HR, whereas G1 cells can be used to study non-homologous end joining (NHEJ). A detailed protocol for this IF method coupled to cell cycle discrimination is described in this paper.

Introduction

Modulation of DNA repair factors is an ever-evolving method for cancer therapy, particularly in DNA DSB repair-deficient tumor environments. The inhibition of specific repair factors is one of the ingenious strategies used to sensitize cancer cells to DNA-damaging agents. Decades of research led to the identification of various mutations of DNA repair genes as biomarkers for therapeutic strategy choices¹. Consequently, the DNA repair field has become a hub for drug development to ensure a wide range of treatments, empowering the personalized medicine concept.

DSBs are repaired by two main pathways: NHEJ and HR². The NHEJ pathway is error-prone, rapidly ligating the two DNA ends with little to no DNA end-processing and involving the protein kinase (DNA-PKcs), the Ku70/80 complex, 53BP1, and RIF1 proteins³. In contrast, HR is a faithful mechanism initiated by BRCA1⁴. An essential step in HR repair is the DNA-end resection process, which is the degradation of the broken ends leading to single-stranded (ss) DNA with 3'-OH ends. BRCA1 facilitates the recruitment of the downstream proteins that form the resectosome MRN/RPA/BLM/DNA2/EXO1, which are involved in the 5' to 3' DNA resection⁵.

The initial end-resection is accomplished through the endonuclease activity of MRE11, allowing for further processing by the DNA2 and EXO1 nucleases. The generated ssDNA overhangs are quickly coated by Replication Protein A (RPA) to protect them from further processing. Subsequently, BRCA2, PALB2, and BRCA1 engage to mediate the displacement of RPA and the assembly of the RAD51 nucleofilament required for homology-directed repair mechanism. A fine balance between the usage of NHEJ and HR is necessary for the optimal maintenance of genomic integrity. The pathway choice depends on the cell cycle phase. HR is preferentially used during the S to G2 phases wherein DNA resection is at the highest level, and the sister chromatids are available to ensure proper repair.

Poly (ADP-ribose) polymerase 1 (PARP-1) is one of the earliest proteins recruited to the DSB. It regulates both resection activity and the assembly of downstream effectors involved in the NHEJ^5,6. PARP-1 is also required for DNA single-stranded break repair during replication^7,8. Due to its important role in DNA repair, PARP inhibitors (PARPi) are used as cancer therapies. In several HR-deficient cancers, PARPi treatment leads to a synthetic lethal response due to the incapacity of HR-deficient cells to repair the accumulated damage via an alternative pathway^9,10. There are currently four FDA approved PARPi: Olaparib, Rucaparib, Niriparib, and Talazoparib (also called BMN 673), which are used for various breast and ovarian cancer treatments¹¹. However, PARPi resistance is common, and one potential cause arises through the reacquisition of HR proficiency¹². Loss or inhibition of PARP-1 in the presence of irradiation dysregulates the resectosome machinery, leading to the accumulation of longer ssDNA tracts¹³. Therefore, an in-depth study of DNA resection in vivo is critical for a clearer understanding of the DNA repair pathways and the subsequent development of new strategies to treat cancer and to overcome PARPi resistance.

There have been several methods employed to detect DNA resection events⁵. One such method is the classical IF-based technique allowing for indirect staining and visualization of the resected DNA after stress-induced DSB by using an anti-RPA antibody. Labelling genomic DNA with 5-bromo-2′-deoxyuridine (BrdU) and detecting only ssDNA is a direct measurement of DNA resection events. It circumvents the monitoring of RPA, which is involved in multiple cellular processes such as DNA replication. In the method described here, cells incubated with BrdU for a single cell cycle allow BrdU to be incorporated into one strand of the replicating cellular DNA. Following resection, IF staining is performed under conditions allowing wherein BrdU detection only in the ssDNA form, with the use of an anti-BrdU antibody. This antibody can only access exposed BrdU nucleotides and will not detect those integrated into double-stranded DNA. Using fluorescence microscopy, the resected DNA can be visualized in the form of punctate BrdU/ssDNA foci. The nuclear intensity of these foci can be used as a readout to quantify resection following DNA damage. This paper describes step-by-step the processes of this method, which can be applied to most mammalian cell lines. This method should be of broad utility as a simple way of monitoring DNA end resection in cellulo, as a proof of concept.

Protocol

1. Cell culture, treatments, and coverslip preparation

NOTE: All cell plating, transfections, and treatments, aside from irradiation, should take place under a sterile cell-culture hood.

1. Day 1
   1. In a 6-well plate, place a single coverslip in each well for as many conditions as needed. Plate ~150,000 HeLa cells for transfection or drug treatment, as desired.
      NOTE: If transfecting, it is recommended to do a reverse transfection at the time of plating, or it is possible to do a forward transfection several hours after plating to allow for adherence. A reverse transfection is accomplished by adding the transfection mix to the coverslip before the cells are added; thus, transfection begins before the cells start adhering. A forward transfection, by contrast, is the addition of the transfection mix post-adherence to the surface usually on the following day. However, it is possible to do this on the same day, provided enough time is given for the cells to adhere.
   2. For this method, use 4 wells: Well 1: siRNA control (termed siCTRL); Well 2: siRNA against PARP-1 (siPARP-1); Well 3: untreated; Well 4: Cells to be treated with 5 µM BMN673 1 h prior to irradiation.
      NOTE: In this protocol, all the tests were conducted under irradiated conditions (see section 1.3).
   3. Incubate the cells at 37 °C in a 5% CO₂ humidified incubator overnight, although the transfection protocol allows for up to 3 days of incubation. The incubation time prior to BrdU treatment will depend on the siRNA transfection efficiency. If there is no transfection, incubation for 16 h is sufficient to allow for adherence to the coverslip and some cell growth.
      NOTE: Incubator conditions can be changed according to the cell line used.
2. Day 2
   1. Add BrdU at a final concentration of 10 µM in the appropriate culturing media, and incubate for 16 h (one cell cycle).
      NOTE: The BrdU solution is prepared in dimethylsulfoxide; the stock solution used is 10 mM and is stored in aliquots at -20 °C.
3. Day 3
   1. Irradiate the plates with a total dose of 5 Gy of X-ray irradiation (vary the dose of irradiation depending on the irradiator type, for example, small animal irradiators vs. benchtop irradiators). See the Table of Materials for the brand and model of irradiator used.
   2. Return the plates to the incubator, and release the cells for 3 h.
   3. During the 3 h incubation period, prepare the two buffers, A and B, and 4% paraformaldehyde (PFA) in 1x phosphate-buffered saline (PBS).
      NOTE: Buffers A and B and the sucrose should be prepared fresh the day of fixation, using fresh sucrose solution to prevent contamination.
      1. Prepare buffer A (Pre-Extraction Buffer), according to the order (30 mL) described in Table 1.
         NOTE: The 1M sucrose should be prepared the day of use to prevent contamination.
      2. Prepare buffer B (Cytoskeleton Stripping Buffer) according to the order described in Table 1 (30 mL).
         ​NOTE: Buffer B must be prepared in the cited order to prevent precipitation of Tween-20 and sodium deoxycholate solution.
      3. Prepare 4% PFA, 2 mL per condition (10 mL), under a chemical hood (Table 1).

2. Pre-extraction and fixation

NOTE: All pre-extraction and fixation are performed with the coverslips remaining in the tissue culture plate on ice or at 4 °C; the coverslip is only lifted in the last step of mounting (see discussion).

1. Pre-extraction
   1. Aspirate the medium, carefully wash the cells twice with 1x PBS, and remove the PBS. Add 2 mL of Pre-extraction Buffer A and immediately incubate at 4 °C for 10 min.
      NOTE: It is important that incubation time in buffer A is not extended. Extended incubation in the pre-extraction buffers will result in an increased number of detached cells from the coverslip. Alternatively, instead of incubation at 4°C, incubation can be done on ice.
   2. Remove Buffer A through aspiration.
      NOTE: Do not wash the coverslips after removing buffer A. Proceed to the next step.
   3. Add 2 mL of Cytoskeleton Stripping Buffer B and immediately incubate at 4 °C for 10 min. Carefully aspirate Buffer B, and carefully wash the cells once with 1x PBS. Carefully aspirate the 1x PBS.
2. Fixation
   1. Fix the cells by adding 2 mL of 4% PFA under the chemical hood.
      NOTE: PFA is toxic and must be manipulated under a chemical hood.
   2. Incubate the cells at room temperature for 20 min. Wash the coverslips twice with 1x PBS, and aspirate the excess.
   3. Cover the coverslips with 100% cold methanol, and incubate the coverslips at -20 °C for 5 min. Wash twice with 1x PBS.
      ​NOTE: The protocol can be paused here. The coverslips can be stored at 4 °C in 1x PBS and the plate wrapped in aluminum foil if necessary. The cells should not be kept more than 5 days before continuing the IF protocol.

3. Permeabilization

1. Incubate the cells with 2 mL of 1x PBS containing 0.5% Triton X-100 at room temperature for 15 min. Wash the coverslips three times with 2 mL of 1x PBS.

4. Immunostaining

1. Blocking step
   1. Prepare enough fresh blocking buffer (3% BSA in 1x PBS) to use 2 mL per coverslip.
      ​NOTE: Prepare an extra 5mL of blocking buffer to make the antibody solutions.
   2. Add 2 mL of blocking buffer to each well and incubate at room temperature for 1 h.
2. Primary antibody incubation
   1. Prepare the primary antibody solution in fresh blocking buffer: BrdU RPN202 1:1000 and proliferating cell nuclear antigen (PCNA) 1:500, 100 µL per coverslip.
      NOTE: To preserve the antibody, a smaller volume can be used, 75-100 µL for a 22 x 22 mm coverslip, 50-75 µL for an 18 x 18 mm coverslip.
   2. On each coverslip, add 100 µL of primary antibody solution. Cover the coverslips with a square of parafilm using tweezers, and carefully position the parafilm to not create bubbles.
   3. Cover the plate in aluminum foil, and incubate the primary antibody overnight at 4 °C in a humidified chamber.
   4. Remove the parafilm squares, and wash the coverslips three times with 2 mL of sterile 1x PBS.
3. Secondary antibody incubation
   1. Prepare enough secondary antibody solution in fresh blocking buffer: anti-mouse 488 fluorescent secondary A11011 (for BrdU) dilution 1:800 and anti-rabbit 568 fluorescent secondary A11011 (for PCNA) dilution 1:800.
      NOTE: The colors used can be changed to suit the experimental needs. The specific brand used can be found in the Table of Materials.
   2. Add on each coverslip 100 µL of secondary antibody solution, cover the coverslips with a square of parafilm using tweezers, and carefully position the parafilm without bubbles.
   3. Incubate the secondary antibody at room temperature for 1 h with the plate covered in aluminum foil.
   4. Wash the coverslips three times with 2 mL of 1x PBS.
4. Nuclear staining
   1. Prepare a volume of 2 mL per coverslip of 1x PBS containing 4',6-diamidino-2-phenylindole (DAPI) at a final concentration of 1 µg/mL (1:1000).
   2. Add on each coverslip 2 mL of the DAPI solution. Incubate the coverslips at room temperature for 10 min with the plate covered in aluminum foil. Wash the coverslips twice with 1x PBS.
   3. Cover the coverslips with 1x PBS, and use either a needle or fine tweezers to lift the coverslip from the bottom of the well.Carefully blot the off the excess liquid by tapping one edge gently on a paper towel.
      NOTE: Be careful not to drop the slide or allow the flat surface with the adherent cells to touch the paper.
   4. Mount the coverslips on slides using 10-20 µL of IF-specific mounting media.

5. Image acquisition and analysis

NOTE: Image acquisition can be done on several types of fluorescent microscopes. An epifluorescence microscope with a 63x oil objective was used; see the Table of Materials for the brand and model. Z-stacks are not required, although they may be of use depending on the cell line and level of mitochondrial staining.

1. Image analysis
   1. For each image, create multiple tiff files; merge all planes, but keep the color channels separate. Import these files into Cell Profiler (The Broad Institute https://cellprofiler.org/home) and analyze using the Speckle Counting pipeline (Supplemental Video 1).
   2. Upload the images (Supplemental Figure S1A).
      1. To begin creating the project, load the speckle counting pipeline into the program and the images to be analyzed in the provided area.
      2. Use the NamesAndTypes module to assign a meaningful name to each image by which other modules will refer to it-C01: Representing BrdU Channel; C02: Representing PCNA Channel; C03: Representing DAPI Channel.
         NOTE: These identifiers will depend on the microscope; there should be an identifier in the file name to distinguish between the channels.
   3. Identify which file will be used to identify the nuclei and name it (change this name as required) (see Supplemental Figure S1B and Supplemental Figure S1C).
      1. Select the diameter of the object between 50 and 300 pixel units, which is the acceptable size range for nuclei, but change the value to fit the nuclei in the image.
         NOTE: It may be that 50 is too large a value and prevents smaller nuclei from being identified; likewise, 300 may allow for large groupings of nuclei to be counted as 1.
      2. Discard objects outside of the diameter range to ensure only those that match the criteria will be counted.
      3. Discard objects touching borders to remove cells that may only be partially in the field.
      4. Apply the threshold to fit the specific images. Note the following settings as an example. Change the thresholding correction factor based on how stringent the thresholding strategy will be. Other settings include threshold strategy: Global; thresholding method: Otsu; two class or three class thresholding: Two class; threshold smoothing scale: 1; threshold smoothing factor: 1; lower and upper bounds on threshold: 0.0 and 1.0; method to distinguish clumped objects: Shape; and method to draw dividing lines between clumped objects: Propagate.
         NOTE: For each parameter, cell profiler provides complementary definitions and available possibility for the variable setting.
   4. Identify the primary object to identify the foci (Supplemental Figure S2A).
      1. Use the following settings: select the input image: Maskedgreen; named the primary object to be identified: BrdUFoci; typical diameter of the object: 1-15; discard the object outside the diameter range: Yes; discard the object touching the border of the image : No; threshold strategy: Adaptative; thresholding method: Otsu; two class or three class thresholding: Three class; threshold smoothing scale: 1; threshold smoothing factor: 3; and size of smoothing filter: 4.
   5. Measure intensity (Supplemental Figure S2B).
      1. Select the image to be measured: OrigGreen.
      2. Click on Add another image. Select the image to be measured: PCNA. Select objects to measure: Nuclei.
         NOTE: This will be used to measure the BrdU and PCNA intensity-the final data that will be graphed.

Representative Results

In this protocol, the bromodeoxyuridine (BrdU)-based assay was used to quantitatively measure the resection response of HeLa cells to irradiation-induced damage. The generated ssDNA tracks are visualized as distinct foci after immunofluorescence staining (Figure 1A). The identified foci were then quantified and expressed as the total integrated intensity of the BrdU staining in the nuclei (Figure 1B, Supplemental Figure S1, Supplemental Figure S2, and Supplemental Figure S3). It is possible to measure the foci number or mean foci intensity, although this can be less reliable than the total nuclear intensity, largely in part to the variable size of the BrdU foci.

To differentiate between the short-range resection as a result of NHEJ and the long-range resection of HR, co-staining was performed using an anti-PCNA antibody to identify cells going through S-phase (Figure 1 and Supplemental Figure S4). PCNA constitutes the DNA clamp that acts as a processivity factor for DNA polymerase and is essential for replication. PCNA is prominent in the nucleus and reaches maximal expression during the S-phase of the cell cycle. Hence, in early S-phase, the PCNA signal is low and has a granular distribution. In contrast, in late S-phase, the PCNA staining is quite strong (Figure 1A). When first analyzing the PCNA (S-phase) signal, the nuclei must be identified and the PCNA intensity measured.

The resulting values are then plotted as a scatter plot to best determine the cut-off intensity to discriminate between the PCNA-positive and PCNA-negative nuclei (Supplemental Figure S4A). The PCNA-negative results will then be removed from the data set to allow for the BrdU intensity-based analysis because of the low BrdU signal regardless of the condition (Supplemental Figure S4B). In these experimental conditions, the PCNA-negative nuclei harbor a basal integrated intensity of BrdU foci of 900 arbitrary units (A.U.). However, this value reaches 1800 A.U. in PCNA-positive nuclei (Figure 1B). This represents a 100% increase in the amount of the integrated signal of the BrdU foci. Increase in BrdU intensity can be then correlated to an increase in DNA resection, which is more pronounced and efficient when cells go through S and G2 phases and are irradiated.

In this protocol, to demonstrate an increase in DNA resection after irradiation (5 Gy), PARP-1 activity was modulated either through the loss of PARP-1 using an siRNA knockdown condition or after potent inhibition of PARP-1 using Talazoparib (BMN673) (Figure 2 and Supplemental Figure S5). The amounts of resected DNA in each condition were then quantified after IF (Figure 2B). In unperturbed condition (siCTRL), the integrated intensity of BrdU foci per nucleus is approximately 1500 A.U. in the PCNA-positive nuclei (Figure 2A,B). After an efficient knockdown of PARP-1 using an siRNA (Figure 2C), a significant increase in the intensity of the BrdU foci to approximately 1900 A.U. was observed, which corresponds to 26% increase in intensity (p ˂ 0.0001). Similarly, after inhibition of PARP-1 using BMN673 (5 µM final concentration), the intensity of foci increased from approximately 1750 to 2500 A.U. corresponding to 43% increase of the foci intensity (p ˂ 0.0001). Thus, loss of PARP-1 dysregulates DNA resection as reported previously¹³.

It is important to remember that BMN-673 both inhibits the activity of PARP-1 and traps it in DNA, and the effect of this treatment is much more detrimental to the cell, resulting in the greater increase of BrdU intensity than in siRNA-treated cells. The slight difference in untreated and siCTRL intensities demonstrates how any treatment, such as siRNA transfection, may affect the response and output of an assay. It can further demonstrate the importance of using the proper controls for each condition.

Figure 1: BrdU foci formation is more prone to occur in PCNA-positive cells than in replicating cells. (A) Representative images of BrdU foci formation and PCNA staining following 5 Gy irradiation and 3 h release. A zoomed-in square showing marked BrdU foci is present in the corner of each BrdU image. Scale bars = 5 µm. (B) Quantification of the BrdU nuclear intensity in untreated (without BMN-673) HeLa cells. The data show the mean ± s.e.m (Mann-Whitney U-test). Abbreviations: BrdU = 5-bromo-2′-deoxyuridine; PCNA = proliferating cell nuclear antigen; IR = irradiation; DAPI = 4',6-diamidino-2-phenylindole; A.U. = arbitrary units; s.e.m. = standard error of the mean. Please click here to view a larger version of this figure.

Figure 2: Poly(ADP-ribose) polymerase-1 knockdown or inhibition results in increased BrdU foci formation in replicating cells. (A) Representative images of BrdU foci formation and PCNA staining in untreated HeLa cells, treated with 5 µM BMN-673, siCTRL, and siPARP-1, followed by 5 Gy irradiation and 3 h release. A zoomed-in square showing marked BrdU foci is present in the corner of each BrdU image. Scale bars = 5 µm. (B) Quantification of the BrdU nuclear intensity in untreated HeLa cells, treated with 5 µM BMN-673, siCTRL, and siPARP-1 followed by 5 Gy irradiation, the data show the mean ± s.e.m (Mann-Whitney U-test). (C) Western blot to validate the siRNA knockdown of PARP-1. The F1-23 PARP-1 antibody recognizes automodified PARP-1 resulting in the smeared appearance of the siCTRL band and the solid appearance of the catalytically inhibited PARP-1 in BMN-673-treated samples. Abbreviations: PARP-1 = poly(ADP-ribose) polymerase 1; BrdU = 5-bromo-2′-deoxyuridine; PCNA = proliferating cell nuclear antigen; IR = irradiation; DAPI = 4',6-diamidino-2-phenylindole; A.U. = arbitrary units; siCTRL = siRNA control; siPARP-1 = siRNA to knock down PARP-1; s.e.m. = standard error of the mean. Please click here to view a larger version of this figure.

Buffer A (Pre-Extraction Buffer) 30 mL
Reagent	Volume	Final Concentration
H2O	18 mL
PIPES (pH 7, 500 mM)	600 µL	10 mM
NaCl (4 M)	750 µL	100 mM
MgCl2 (1 M)	90 µL	3 mM
EGTA (500 mM)	60 µL	1 mM
Trition X-100 (10%)	1.5 mL	0.50%
Sucrose (1 M)	9 mL	300 mM

Buffer B (Cytoskeleton Stripping Buffer) 30 mL
Reagent	Volume	Final Concentration
H2O	25.035 mL
Tris pH 7.5 (1 M)	300 µL	10 mM
NaCl (4 M)	75 µL	10 mM
MgCl2 (1 M)	90 µL	3 mM
Tween20 (10%)	3 mL	1%
10% Sodium Deoxycholcate	1.5 mL	0.50%

Reagent	Volume
16% PFA	2.5 mL
PBS 1x	7.5 mL

Table 1.

Supplemental Figure S1: Visual representation of the Cell Profiler Software and Speckle counting pipeline part 1. (A) Screenshot shows the window used to identify the different channel files of the images to be analyzed. These shots show the cell profiler interface. (B) Screenshot of the Identify Primary Objects menu to identify the Nuclei. The highlighted values are the commonly altered values to best identify nuclei. Abbreviations: BrdU = 5-bromo-2′-deoxyuridine; DAPI = 4',6-diamidino-2-phenylindole. Please click here to download this File.

Supplemental Figure S2: Visual representation of the Cell Profiler Software and Speckle counting pipeline part 2. (A) Screenshot of the Identify Primary Objects menu to identify the BrdU foci within the nuclei. (B) Screenshot of Measure Object Intensity menu to measure the BrdU and PCNA nuclear intensity. Abbreviations: BrdU = 5-bromo-2′-deoxyuridine; PCNA = proliferating cell nuclear antigen. Please click here to download this File.

Supplemental Figure S3: Visual representation of the Cell Profiler Software and Speckle counting pipeline part 3. (A) Representation of the nuclei identification by cell profiler following optimization. (B) Representation of the BrdU foci identification. (C) Screenshot is a zoom-in on one cell within the initial window. Please click here to download this File.

Supplemental Figure S4: Scatter plot of PCNA-positive and PCNA-negative cells. (A) An example of a PCNA scatter plot used to distinguish PCNA-positive from PCNA-negative cells. The PCNA intensity from a single condition was plotted and the distinction between the two populations identified. The green highlights the PCNA-positive cells, the red represents the PCNA-negative cells. This is done for each individual experiment and condition. (B) Quantification of the BrdU nuclear intensity in PCNA-negative untreated HeLa cells, treated with 5 µM BMN-673, siCTRL, and siPARP-1; the data show the mean ± s.e.m (Mann-Whitney U-test). Abbreviations: PARP-1 = poly(ADP-ribose) polymerase 1; BrdU = 5-bromo-2′-deoxyuridine; IR = irradiation; PCNA = proliferating cell nuclear antigen; A.U. = arbitrary units; siCTRL = siRNA control; siPARP-1 = siRNA to knock down PARP-1; s.e.m. = standard error of the mean. Please click here to download this File.

Supplemental Figure S5: BrdU signal without irradiation. (A) Representative images of BrdU foci formation and PCNA staining without irradiation. Scale bars = 5 µm. A zoomed-in square showing marked BrdU foci is present in the corner of each BrdU image. (B) Quantification of the BrdU nuclear intensity in PCNA-positive untreated Hela cells, treated with 5 µM BMN-673, siCTRL, and siPARP-1; the data show the mean ± s.e.m (Mann-Whitney U-test). (C) Quantification of the BrdU nuclear intensity in PCNA-negative untreated HeLa cells, treated with 5 µM BMN-673, siCTRL, and siPARP-1; the data show the mean ± s.e.m (Mann-Whitney U-test). Abbreviations: PARP-1 = poly(ADP-ribose) polymerase 1; BrdU = 5-bromo-2′-deoxyuridine; PCNA = proliferating cell nuclear antigen; IR = irradiation; DAPI = 4',6-diamidino-2-phenylindole; A.U. = arbitrary units; siCTRL = siRNA control; siPARP-1 = siRNA to knock down PARP-1; s.e.m. = standard error of the mean. Please click here to download this File.

Supplemental Video 1: Use of Cell Profiler and the Speckle Counting pipeline to analyze BrdU staining for the measurement of DNA resection. This video demonstrates the use of Cell Profiler and the Speckle counting pipeline. It shows how to import and analyze images with this tool specifically for this protocol.Abbreviation: BrdU =5-bromo-2′-deoxyuridine. Please click here to download this Video.

Discussion

This paper describes a method that makes use of IF staining to measure variations in DNA resection in cellulo. The current standard for observing an effect on DNA resection is through RPA staining; however, this is an indirect method that may be influenced by DNA replication. Previously, another BrdU incorporation-based DNA resection IF technique has been described for classifying the resulting intensities in BrdU-positive and BrdU-negative cells. This method allowed for cells that are not undergoing HR to be counted as positive due to background or mitochondrial staining resulting in a high BrdU intensity^14,15,16. The primary novelty of the method described in this paper is the addition of PCNA staining, which allows for selectivity for S and G2 phases of the cell cycle, ensuring that the resulting BrdU signal is due to resection and thus, to homology-directed repair.

The critical step in this protocol is the pre-extraction; without this, the BrdU foci will not be visible, and if done incorrectly, the cells will detach completely. Incorrectly prepared buffers will result in incomplete pre-extraction, such as partial cytoskeletal removal or increased background signal. It is very important to respect the duration of incubation with the pre-extraction buffers, increased duration in the buffers can result in the cells detaching from the coverslip. Importantly, if this protocol is performed with poorly adherent cells, pre-treat the coverslips with polylysine to reduce cell detachment during pre-extraction. Subsequently, without methanol fixation, the PCNA signal will not be visible.

Another important variable in this protocol is the blocking buffer: incompatible blocking buffers will result in loss of BrdU signal. For example, 10% fetal bovine serum in 1x PBS blocking buffer will function when not combined with methanol fixation; however, when combined with methanol fixation as required for the PCNA staining, the BrdU signal is considerably reduced. It is possible other blocking agents and methods will work with this protocol; 3% BSA in 1x PBS provided the best results. A final essential component to this technique is the use of a DNA damage source; without this, there will be little to no nuclear BrdU signal (Supplemental Figure S5). Experience has shown that irradiation is an excellent source of damage for this method. However, if there is no access to an irradiator, radiomimetic drugs such as neocarzinostatin can be used instead.

Cell profiler is a useful and versatile tool that allows easy and consistent analysis of the images produced in this technique. The settings can be customized for both the size of the objects identified and the threshold at which it does so (Supplemental Video 1). As the nuclear intensity of the BrdU staining is the desired readout, the key customization is in the identification of the nuclei from the DAPI files. The analysis of the images from each condition must be done separately as the results are obtained in the form of a single csv file containing cell and image numbers, but not condition names. To determine if the analysis conditions are appropriate, use the Start test mode feature, and make sure the eye icon is open in the desired steps. Close this icon when running the actual analysis as it will result in pop-up windows and slow down the program. In the Start test mode, users can go back and forth, altering the settings as needed to properly identify first the nuclei and then the BrdU foci.

Once satisfied with the resulting identification, use the same settings for each condition within the same experiment; this may require the acquisition of a test image for each condition prior to analyzing the whole batch to ensure that the settings comply with each condition. It is important to note that, depending on image acquisition, there may be variability in the Cell profiler settings between users and between software versions. Hence, it is essential to go through the testing stage to determine the ideal parameters for specific experiments. When preparing to graph the results, the PCNA (S phase)-positive cells must be identified. To do this, the PCNA intensity is measured, in the same fashion as the BrdU intensity, and the resulting values plotted as a scatter plot to best visualize the cut-off intensity value for PCNA-positive cells. The PCNA-negative results will then be removed from the data set to allow for the BrdU intensity graphing. The provided supplementary figures show screenshots from the program highlighting the most optimized settings (Supplemental Figure S1, Supplemental Figure S2, and Supplemental Figure S3).

A possible modification to the protocol is the PCNA antibody used; two different PCNA antibodies were successfully tested, the one listed here and another which is no longer in production-a rat monoclonal antibody (Bulldog Bio PCA16D10). Finally, users should cover the coverslips with parafilm during antibody incubation. It is not strictly necessary, although a significantly larger volume of antibody solution would be required to cover the coverslip entirely without this. Another method is to place a sheet of parafilm surrounding a glass plate, place drops of the antibody solution onto the parafilm, and carefully place the coverslip cell-side down onto the drop of antibody solution. This glass plate can then be placed in a humid chamber at 4 °C overnight for primary incubation and in the dark at room temperature for secondary incubation. This method is completely functional although it does significantly increase the time of handling the coverslips and as a result, increases the probability of dropping or breaking the coverslip.

An alternative method to measure resection is through the use of the DNA-combing SMART method; while this technique provides very clear results, it is much more complicated, requires more time, and is more expensive¹⁷. The SMART method requires the extraction of the DNA without damaging it, followed by stretching this DNA onto coverslips to be followed by an IF. The BrdU IF method, by comparison, is both simple and cost-effective, not requiring a greater investment than the cost of the antibody. This method provides an initial way to measure resection that is more accurate than simply measuring RPA foci formation, which can be related to its effects on many cellular functions. However, RPA is phosphorylated on specific residues during resection as part of the DNA damage response.

Single-molecule imaging recently revealed phosphorylated RPA (pRPA) as a negative resection regulator¹⁸. Hence, in the long term, this method could be made even more precise by coupling PCNA/phosphorylated RPA staining with BrdU imaging with the use of appropriate antibodies and optimized staining conditions. However, the method presented provides a representation of resection that is observed independently from the proteins involved in the resection process, thus negating bias that can occur from changes in their signal in response to the treatment. This technique provides not only a method to determine if a protein is involved in resection, but also to determine if cell death resulting from drug treatment is related to hypo/hyper-resection. This information can be beneficial in understanding not only the mechanistic aspects of DNA repair, but also the biological mechanism underlying the drug-induced cell death.

Materials

Name	Company	Catalog Number	Comments
Alexa 568 goat anti-rabbit	Molecular probes	A11011	1:800
Alexa Fluor 488 goat anti-mouse	Molecular probes	A11001	1:800
Anti PARP1 (F1-23)	Homemade		1:2500
Anti PCNA (SY12-07)	Novus	NBP2-67390	1:500
Anti-Alpha tubulin (DM1A)	Abcam	Ab7291	1:100000
anti-BrdU	GE Healthcare	RPN202	1:1000
Benchtop X-ray Irradiator	Cell Rad
BMN673	MedChem Express	HY-16106
Bromodeoxyuridine (BrdU)	Sigma	B5002
BSA	Sigma	A7906
Cell profiler	Broad Institute	V 3.19	https://cellprofiler.org/
Curwood Parafilm M Laboratory Wrapping Film 4in / 250 ft	Fisher scientific	13-374-12
DAPI (4',6-Diamidino-2-Phenylindole, Dihydrochloride)	Invitrogen Life Technology	D1306
DMEM high glucose	Fisher scientific	10063542
EGTA	Sigma-Aldrich	E3889
Fetal Bovine serum	Gibco	12483-020
Fisherbrand Cover Glasses: Squares 22 x 22	Fisher scientific	12 541B
Fluorescent microscope	Leica	DMI6000B	63x immersion objective
HeLa	ATCC	CCL-2
HERACELL 160I CO2 INCUBATOR CU 1-21 TC 120V	VWR	51030408	37% CO2
MgCl2	BioShop Canada	MAG520.500
NaCl	BioShop Canada	SOD002.10
Needle
PBS 1x	Wisent Bio Products	311-010-CS
PFA 16%	Cedarlane Labs	15710-S(EM)
PIPES	Sigma-Aldrich	P6757-100G
ProLong Gold Antifade Mountant	Invitrogen Life Technology	P-36930
RNAiMAX	Invitrogen	13778-075
siPARPi	Dharmacon		AAG AUA GAG CGU GAA GGC GAA dTdT
siRNA control	Dharmacon		UUCGAACGUGUCACGUCAA
Sodium Deoxycholcate	Sigma-Aldrich	D6750-100G
Sucrose	BioShop Canada	SUC507.5
Tris-base	BioShop Canada	TRS001.5
Trition X-100	Millipore Sigma	T8787-250ML
Tween20	Fisher scientific	BP337500
Tweezers