The Enzyme-Modified Neutral Comet (EMNC) Assay for Complex DNA Damage Detection

The comet assay is a versatile, simple, and sensitive gel electrophoresis–based method that can be used to measure and accurately quantify DNA damage, particularly single and double DNA strand breaks, in single cells. While generally this is used to measure variation in DNA strand break levels and repair capacity within a population of cells, the technique has more recently been adapted and evolved into more complex analysis and detection of specific DNA lesions, such as oxidized purines and pyrimidines, achieved through the utilization of damage-specific DNA repair enzymes following cell lysis. Here, we detail a version of the enzyme-modified neutral comet (EMNC) assay for the specific detection of complex DNA damage (CDD), defined as two or more DNA damage lesions within 1–2 helical turns of the DNA. CDD induction is specifically relevant to ionizing radiation (IR), particularly of increasing linear energy transfer (LET), and is known to contribute to the cell-killing effects of IR due to the difficult nature of its repair. Consequently, the EMNC assay reveals important details regarding the extent and complexity of DNA damage induced by IR, but also has potential for the study of other genotoxic agents that may induce CDD.


Introduction
Over the past decade, the single-cell gel electrophoresis, or comet assay, has become one of the standard methods for assessing DNA damage, with applications in genotoxicity testing, human biomonitoring, and molecular epidemiology, as well as fundamental research in DNA damage and repair, mainly due to its simplicity, sensitivity, versatility, speed and cheapness. When first devised, its simple approach consisted of embedding cells in an agarose matrix on a microscope slide and lysing the cells with non-ionic detergent and high-molarity sodium chloride. This causes the removal of membranes, cytoplasm and nucleoplasm, and the disruption of nucleosomes, leaving an intact nuclear matrix or scaffold composed of ribonucleic acid and proteins, with the DNA wrapped around it in its supercoiled form [1]. When the negative DNA supercoiling is subsequently unwound by the relatively neutral pH buffer (pH = 9.5), this causes the loops expanded out from the nucleoid core following gel electrophoresis to form a comet tail, visualized using ethidium bromide staining and fluorescence microscopy. This seems to be simply a halo of relaxed loops pulled to one side by the electrophoretic field [2]. The procedure was then modified with the use of high pH treatment (>pH 13) pre-and during electrophoresis, which is necessary to reveal DNA single-strand breaks (SSBs) [3,4]. Since its establishment, the comet assay has proven to be one of the most versatile methods for studying cellular DNA repair capacity, as it allows to quantitatively measure the actual DNA damage induced, as well as the damage remaining at intervals after treatment, thus allowing a study of the kinetics of cellular repair [5,6]. This consequently can avoid the interference of other the cleavage of the phosphodiester backbone at AP sites via hydrolysis, leaving a onenucleotide gap with 3'-hydroxyl and 5'-deoxyribose phosphate (dRP) termini, OGG1 is involved in the excision repair of predominantly 8-oxoguanine (8-oxoG) and NTH1 excises oxidized pyrimidines (e.g., thymine glycol and 5-hydroxycytosine) from DNA. The use of these three enzymes, particularly in combination, allows the conversion of DNA base damage-associated CDD into additional DNA DSBs, and ultimately, an increase in the visible tail intensity following electrophoresis.   Plate 50 µL of each mix onto separate pre-prepared 10 cm LB agar plates (containing the appropriate antibiotics) using bacterial spreaders. The remainder of the mix can be stored at 4 • C, in case further required. • Invert the plates and incubate overnight in a static incubator set at 37 • C. •

Experimental Design
The following day, select a single bacterial colony, add to 5 mL of LB containing the appropriate antibiotics, and incubate at 37 • C with shaking at 225 rpm overnight. We would advise selecting 2-3 different colonies and incubating in separate tubes to ensure at least one efficiently grown culture. • Add 300 µL of overnight culture to feed a 30 mL culture (i.e. 1:100) containing the appropriate antibiotics and 60 µL of glucose (20%), and grow at 37 • C with shaking at 225 rpm until an OD600nm of 0.6-0.8 is achieved (~3 h).

•
Add 30 mL of culture to feed a 300 mL culture (i.e., 1:10) containing the appropriate antibiotics and 600 µL of glucose (20 %), and grow at 37 • C with shaking at 225 rpm until an OD600 nm of 0.6-0.8 is achieved (~1.5 h). Any culture from the original 30 mL culture can be used to create a glycerol stock by removing 150 µL and adding to 50 µL of 50% glycerol, which can then be stored at −80 • C. Agarose solution can be used immediately (once cooled to 37 °C) or stored at RT.

Slide Coating (10 min)
 Prepare slides by adding 800 µ L of molten normal melting point agarose to a microscope slide, add a 22 × 50 mm coverslip, and leave agarose to set (~2-5 min) on a flat surface. Remove coverslip, carefully sliding sideward, and air dry slides overnight.

Cell Embedding and Lysis (30 min)
 Trypsinise cells and dilute to ~1 × 10 5 cells/mL. Add 250 µ L of cell suspension per well of a 24-well plate on ice to prevent repair and adhesion. Induce DNA damage by chemical or physical stress within plate. Note that chemicals with a long half-life will continue to induce DNA damage in cells in suspension, so these are not recommended to be used using this method. Alternatively, cells can be grown as monolayer and exposed separately to stress according to the experiment design (compound concentrations, time exposure) and then trypsinized. The amount of DNA damaging agent should be determined empirically, but should induce a DNA damage level of <35% tail DNA to ensure this is not too extensive for the cell to repair, if analyzing DNA repair efficiency.  Add 500 µ L of low melting point agarose (previously melted and maintained at 37 °C ) to cells, mix gently, and add 70 µ L of this cell suspension to two areas on a normal melting point agarose-coated slide (equivalent to duplicate treatments). Add two 22 × 22 mm coverslips and place on the ice tray for ~2 min for the agarose to set. Two slides per treatment should be prepared (one for the buffer treatment only detecting DSBs, and one for the enzyme treatment detecting both DSBs and CDD).  A negative control (no DNA damage treatment) should be prepared. Ideally, the experiment would also include a positive control with a treatment known to induce CDD (e.g. high-LET radiation • Trypsinise cells and dilute to~1 × 10 5 cells/mL. Add 250 µL of cell suspension per well of a 24-well plate on ice to prevent repair and adhesion. Induce DNA damage by chemical or physical stress within plate. Note that chemicals with a long halflife will continue to induce DNA damage in cells in suspension, so these are not recommended to be used using this method. Alternatively, cells can be grown as monolayer and exposed separately to stress according to the experiment design (compound concentrations, time exposure) and then trypsinized. The amount of DNA damaging agent should be determined empirically, but should induce a DNA damage level of <35% tail DNA to ensure this is not too extensive for the cell to repair, if analyzing DNA repair efficiency. • Add 500 µL of low melting point agarose (previously melted and maintained at 37 • C) to cells, mix gently, and add 70 µL of this cell suspension to two areas on a normal melting point agarose-coated slide (equivalent to duplicate treatments). Add two 22 × 22 mm coverslips and place on the ice tray for~2 min for the agarose to set. Two slides per treatment should be prepared (one for the buffer treatment only detecting DSBs, and one for the enzyme treatment detecting both DSBs and CDD).

•
A negative control (no DNA damage treatment) should be prepared. Ideally, the experiment would also include a positive control with a treatment known to induce CDD (e.g., high-LET radiation).

•
If analyzing DNA repair efficiency, place slides in a humidified chamber at 37 • C to allow repair according to experiment time-point (e.g., up to 6 h).

•
After incubation, carefully remove coverslips by sliding sideward and place slides in coplin jars containing fresh cold lysis buffer. Lyse cells for at least 1 h at 4 • C.
 Agarose solution can be used immediately (once cooled to 37 °C) or stored at RT.

Slide Coating (10 min)
 Prepare slides by adding 800 µ L of molten normal melting point agarose to a microscope slide, add a 22 × 50 mm coverslip, and leave agarose to set (~2-5 min) on a flat surface. Remove coverslip, carefully sliding sideward, and air dry slides overnight.

Cell Embedding and Lysis (30 min)
 Trypsinise cells and dilute to ~1 × 10 5 cells/mL. Add 250 µ L of cell suspension per well of a 24-well plate on ice to prevent repair and adhesion. Induce DNA damage by chemical or physical stress within plate. Note that chemicals with a long half-life will continue to induce DNA damage in cells in suspension, so these are not recommended to be used using this method. Alternatively, cells can be grown as monolayer and exposed separately to stress according to the experiment design (compound concentrations, time exposure) and then trypsinized. The amount of DNA damaging agent should be determined empirically, but should induce a DNA damage level of <35% tail DNA to ensure this is not too extensive for the cell to repair, if analyzing DNA repair efficiency.  Add 500 µ L of low melting point agarose (previously melted and maintained at 37 °C ) to cells, mix gently, and add 70 µ L of this cell suspension to two areas on a normal melting point agarose-coated slide (equivalent to duplicate treatments). Add two 22 × 22 mm coverslips and place on the ice tray for ~2 min for the agarose to set. Two slides per treatment should be prepared (one for the buffer treatment only detecting DSBs, and one for the enzyme treatment detecting both DSBs and CDD).  A negative control (no DNA damage treatment) should be prepared. Ideally, the experiment would also include a positive control with a treatment known to induce CDD (e.g. high-LET radiation).  If analyzing DNA repair efficiency, place slides in a humidified chamber at 37 °C to allow repair according to experiment time-point (e.g. up to 6 h).  After incubation, carefully remove coverslips by sliding sideward and place slides in coplin jars containing fresh cold lysis buffer. Lyse cells for at least 1 h at 4 °C . CRITICAL STEP: duration of cell lysis may vary among different cell lines and must be empirically determined. PAUSE STEP: slides can be kept in lysis buffer overnight at 4 °C .

Enzyme Treatment (100 min)
 Wash slides three times, for 5 min each, in PBS in coplin jars, at 4 °C (to remove lysis buffer).  Lay slides on a flat surface and place 50 µ L of enzyme solution per agarose/cell area, or buffer alone for the untreated slides, and cover with a 22 × 22 mm coverslip. Note that the amount of enzyme to use should be predetermined beforehand, ideally by titrating each enzyme (NTH1, OGG1, and APE1) against a positive control (e.g. high-LET radiation) versus a negative control (e.g. low-LET radiation) to ensure that CDD is only revealed largely under the former conditions.  Place slides in a humidified chamber and incubate at 37 °C for 1 h to allow enzyme processing.

Electrophoresis (105 min + Overnight Dry)
 Carefully remove coverslips by sliding sideward and wash slides three times for 5 min each in coplin jars containing 1× PBS.
CRITICAL STEP: duration of cell lysis may vary among different cell lines and must be empirically determined.
 Agarose solution can be used immediately (once cooled to 37 °C) or stored at RT.

Slide Coating (10 min)
 Prepare slides by adding 800 µ L of molten normal melting point agarose to a microscope slide, add a 22 × 50 mm coverslip, and leave agarose to set (~2-5 min) on a flat surface. Remove coverslip, carefully sliding sideward, and air dry slides overnight.

Cell Embedding and Lysis (30 min)
 Trypsinise cells and dilute to ~1 × 10 5 cells/mL. Add 250 µ L of cell suspension per well of a 24-well plate on ice to prevent repair and adhesion. Induce DNA damage by chemical or physical stress within plate. Note that chemicals with a long half-life will continue to induce DNA damage in cells in suspension, so these are not recommended to be used using this method. Alternatively, cells can be grown as monolayer and exposed separately to stress according to the experiment design (compound concentrations, time exposure) and then trypsinized. The amount of DNA damaging agent should be determined empirically, but should induce a DNA damage level of <35% tail DNA to ensure this is not too extensive for the cell to repair, if analyzing DNA repair efficiency.  Add 500 µ L of low melting point agarose (previously melted and maintained at 37 °C ) to cells, mix gently, and add 70 µ L of this cell suspension to two areas on a normal melting point agarose-coated slide (equivalent to duplicate treatments). Add two 22 × 22 mm coverslips and place on the ice tray for ~2 min for the agarose to set. Two slides per treatment should be prepared (one for the buffer treatment only detecting DSBs, and one for the enzyme treatment detecting both DSBs and CDD).  A negative control (no DNA damage treatment) should be prepared. Ideally, the experiment would also include a positive control with a treatment known to induce CDD (e.g. high-LET radiation).  If analyzing DNA repair efficiency, place slides in a humidified chamber at 37 °C to allow repair according to experiment time-point (e.g. up to 6 h).  After incubation, carefully remove coverslips by sliding sideward and place slides in coplin jars containing fresh cold lysis buffer. Lyse cells for at least 1 h at 4 °C . CRITICAL STEP: duration of cell lysis may vary among different cell lines and must be empirically determined. PAUSE STEP: slides can be kept in lysis buffer overnight at 4 °C .

Enzyme Treatment (100 min)
 Wash slides three times, for 5 min each, in PBS in coplin jars, at 4 °C (to remove lysis buffer).  Lay slides on a flat surface and place 50 µ L of enzyme solution per agarose/cell area, or buffer alone for the untreated slides, and cover with a 22 × 22 mm coverslip. Note that the amount of enzyme to use should be predetermined beforehand, ideally by titrating each enzyme (NTH1, OGG1, and APE1) against a positive control (e.g. high-LET radiation) versus a negative control (e.g. low-LET radiation) to ensure that CDD is only revealed largely under the former conditions.  Place slides in a humidified chamber and incubate at 37 °C for 1 h to allow enzyme processing.

Electrophoresis (105 min + Overnight Dry)
 Carefully remove coverslips by sliding sideward and wash slides three times for 5 min each in coplin jars containing 1× PBS.
PAUSE STEP: slides can be kept in lysis buffer overnight at 4 • C.

Enzyme Treatment (100 min)
• Wash slides three times, for 5 min each, in PBS in coplin jars, at 4 • C (to remove lysis buffer). • Lay slides on a flat surface and place 50 µL of enzyme solution per agarose/cell area, or buffer alone for the untreated slides, and cover with a 22 × 22 mm coverslip. Note that the amount of enzyme to use should be predetermined beforehand, ideally by titrating each enzyme (NTH1, OGG1, and APE1) against a positive control (e.g., high-LET radiation) versus a negative control (e.g., low-LET radiation) to ensure that CDD is only revealed largely under the former conditions. • Place slides in a humidified chamber and incubate at 37 • C for 1 h to allow enzyme processing.

Electrophoresis (105 min + Overnight Dry)
• Carefully remove coverslips by sliding sideward and wash slides three times for 5 min each in coplin jars containing 1× PBS. • Transfer slides to an electrophoresis tank (can be placed at 4 • C if ambient temperature is relatively high), organize them into two separate columns, and cover with~1.2 L of the fresh cold Comet Electrophoresis Buffer. • Incubate slides for 30 min. • Electrophorese slides at 25 V (1 V/cm,~20 mA; note that volume of buffer may need to be removed/added to adjust to the correct current) for 25 min. • Carefully remove slides from the electrophoresis tank to minimize movement of the slides (and potential loss of agarose/cells) and lay on a flat surface covered with paper roll. Cover agarose/gel areas with cold 1× PBS buffer (~500 µL per slide) for 5 min. Repeat twice.

•
Pour off excess PBS, lay slides flat, and allow to dry overnight.

Rehydration and DNA Staining (75 min + Overnight Dry)
• Place dried slides in coplin jars containing dH 2 O (pH = 8.0) for 30 min to rehydrate agarose. • Lay slides on a flat surface covered with paper roll and add enough SYBR Gold (diluted 1:20,000) to cover each slide (~500 µL) for DNA staining. Cover the slides to protect from light and incubate for 30 min. • Remove excess stain from the slides, lay flat, and allow slides to dry overnight (while protecting from light) prior to analysis or storage in a sealed box.

Analysis
• Capture images of stained DNA from the dried slides using a fluorescent microscope equipped with a 10× objective. There is no standard procedure for measuring the intensity of light and correcting the level of DNA migration accordingly, therefore, this should be optimized empirically. • Images can be analyzed live or offline using a validated image analysis software and by scoring 50 cells per agarose/cell square across multiple images, which are present in duplicate on each slide. In our experience, we recommend the Komet 6.0 image analysis software, although other commercial and free software is available (e.g., Comet Assay IV, OpenComet, and CometScore).

•
When analyzing images, it is important to recognize the presence of apoptotic cells (also called "hedgehogs") that have extremely high DNA damage levels (>90% of DNA in the comet tail), and that these should be quantified separately from cells containing levels of DNA damage (0-40 %) that are amenable to cellular repair.

Principles of the EMNC Assay
Here, we developed and described a protocol for the detection of CDD using the EMNC assay, which incorporates the use of three different recombinant DNA repair enzymes (APE1, OGG1, and NTH1). The neutral version of the classic comet assay provides details regarding the quantitative levels of DNA DSBs and the capacity of cells to repair these, but does not allow an assessment of DNA damage complexity ( Figure 1A). Therefore, the addition of an enzyme treatment step post cell lysis will recognize and incise any residual DNA base damage (based on the specificity of the enzyme used), and which, if in close proximity, will create additional DSBs that can be separated by electrophoresis and detected following DNA staining and image analysis ( Figure 1B). Usually, the EMNC assay is performed both in the absence and presence of enzymes, therefore, this assay advantageously reveals the relative levels and repair of both DSBs and CDD in a single experiment. In the EMNC assay, following cell lysis, the DNA is treated with recombinant DNA repair enzymes that recognize and incise the DNA at unrepaired DNA base damage sites to create additional DSBs that can then be detected following electrophoresis.

Detection of CDD Following Proton Irradiation
We have validated that the EMNC assay detects CDD using cells irradiated with protons at lower energies, and therefore increasing LET compared to high energy (low-LET) protons. High-LET radiation, including protons at and around the Bragg peak where the radiation is deposited, are well-known to induce increased levels of CDD through the densely ionising track structure, whereas low-LET radiation generates more spacely-separated DNA damage [14]. When the EMNC assay was performed in the absence of an enzyme (APE1, OGG1, and NTH1) treatment, this revealed that the levels and induction of DSBs (shown as % tail DNA) are the same immediately following proton irradiation with cells positioned at the relatively high-LET at the Bragg peak distal end, versus cells In the EMNC assay, following cell lysis, the DNA is treated with recombinant DNA repair enzymes that recognize and incise the DNA at unrepaired DNA base damage sites to create additional DSBs that can then be detected following electrophoresis.

Detection of CDD Following Proton Irradiation
We have validated that the EMNC assay detects CDD using cells irradiated with protons at lower energies, and therefore increasing LET compared to high energy (low-LET) protons. High-LET radiation, including protons at and around the Bragg peak where the radiation is deposited, are well-known to induce increased levels of CDD through the densely ionising track structure, whereas low-LET radiation generates more spacelyseparated DNA damage [14]. When the EMNC assay was performed in the absence of an enzyme (APE1, OGG1, and NTH1) treatment, this revealed that the levels and induction of DSBs (shown as % tail DNA) are the same immediately following proton irradiation with cells positioned at the relatively high-LET at the Bragg peak distal end, versus cells positioned at the low-LET entrance (Figure 2A, compare dark blue and dark green bars at time 0; Figure 2B,C). Additionally, the efficient repair of DSBs under both of these conditions is apparent 4 h post-irradiation. In contrast, performing the EMNC assay in the presence of an enzyme (APE1, OGG1, and NTH1) treatment reveals that additional DSBs (corresponding to CDD) are immediately generated only following relatively high-LET protons, and not low-LET protons (Figure 2A, compare light green and light blue bars at time 0; Figure 2B,C). Furthermore, the increased persistence of CDD is shown through the observation that these exist for at least 4 h post-irradiation with relatively high-LET protons, consistent with the theory that CDD represents a challenge to the cellular DNA repair machinery. Note that the levels of DSBs were not significantly increased in unirradiated cells in the presence versus the absence of enzyme treatment (Figure 2A, compare dark blue/green and light blue/green bars at control; Figure 2B,C). More comprehensive analysis has previously been performed [21,22], which has clearly demonstrated that relatively high-LET protons generated at the distal end of the Bragg peak can induce CDD in cells that persists for several hours (>4 h) post-irradiation, and that this contributes significantly to the increased cell-killing effects observed under these conditions.

Conclusions
CDD is a major factor involved in the cell-killing effects of ionising radiation, as this

Conclusions
CDD is a major factor involved in the cell-killing effects of ionising radiation, as this persists in cells post-treatment and is significantly more difficult to repair than isolated DNA lesions, given that CDD will consist of multiple DNA damage types (e.g., base damage, SSBs and DSBs). However, a quantitative assay to measure and visualize CDD has been a long-standing challenge in the radiobiology field. We demonstrate that the EMNC assay can clearly be used to quantify the levels of CDD (DSB-associated) in cultured cells in the absence and presence of irradiation, and indeed, could be utilized for the assessment of other DNA damaging agents for their ability to induce CDD. However, the EMNC assay can also be employed to establish the capacity of individual cells and cell populations to repair CDD. Therefore, the EMNC is a valuable methodology and resource for quantitatively determining CDD levels following different sources of ionising radiation, particularly high-LET protons and heavy ions, but also to reveal specific details regarding the CDD repair efficiency of different cell models. Additionally, the enzymes and critical DNA repair mechanisms that are responsive to CDD are still debatable [14], and are very much dependent on the radiation source (and LET) itself. To this effect, we have recently demonstrated that CDD induced by relatively high-LET protons triggers a specific cellular DNA damage response mediated by histone H2B ubiquitylation catalyzed by the E3 ubiquitin ligases ring finger protein 20/40 (RNF20/40) and male-specific lethal-2 (MSL2) [21]. We have also utilized an siRNA screen for identifying deubiquitylation enzymes that modulate radiosensitivity following relatively high-LET protons, which revealed a critical role for ubiquitin specific protease (USP6) in this process. Here, we revealed evidence that USP6 is essential for maintaining the stability of the SSB repair protein poly(ADP-ribose) polymerase-1 (PARP-1), and that targeting the PARP-1 protein itself using siRNA, or inhibiting PARP-1-dependent poly(ADP-ribosyl)ation using olaparib, is able to significantly decrease the survival of cells in response to relatively high-LET protons [22]. Consequently, the utilization of the EMNC has allowed us to establish new mechanistic evidence supporting the roles for changes at the histone and chromatin level, as well as in identifying a critical role for the SSB repair pathway in responding to proton-induced CDD, which requires further research and development.

Troubleshooting
Causes and solutions to a number of potential problems encountered during execution of the EMNC assay (Table 1).

Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.