Skip to main content
NCBI home page
UKPMC Funders Author Manuscripts logoLink to UKPMC Funders Author Manuscripts
. Author manuscript; available in PMC: 2024 Dec 5.
Published in final edited form as: Anal Chem. 2020 Jul 8;92(14):9887–9894. doi: 10.1021/acs.analchem.0c01393

Rapid Quantification of Oxidative and UV induced DNA Damage by Repair Assisted Damage Detection-(Rapid RADD)

Noa Gilat 1, Dmitry Torchinsky 1, Sapir Margalit 1, Yael Michaeli 1, Sigal Avraham 1, Hila Sharim 1, Nadav Elkoshi 2, Carmit Levy 2, Shahar Zirkin 1,*, Yuval Ebenstein 1,*
PMCID: PMC7616909  EMSID: EMS95043  PMID: 32578422

Abstract

Knowing the amount and type of DNA damage is of great significance for a broad range of clinical and research applications. However, existing methods either lack in their ability to distinguish between types of DNA damage, or are limited in their sensitivity and reproducibility. The method described herein enables rapid and robust quantification of type-specific single-strand DNA damage. The method is based on Repair-Assisted Damage Detection (RADD) by which fluorescent nucleotides are incorporated into DNA damage sites using type-specific repair enzymes. Up to 90 DNA samples are then deposited on a multi-well glass slide, and analyzed by a conventional slide scanner for quantification of DNA damage levels. Accurate and sensitive measurements of oxidative or UV-induced DNA damage and repair levels are presented for both in-vitro and in-vivo models.

Introduction

The human body suffers thousands of DNA damage events each day, associated with both endogenous and exogenous damaging factors 1,2. Normally, DNA damage is rapidly repaired via extensive cellular enzymatic machinery, and failure to repair damaged DNA can result in mutagenesis. This in turn may lead to loss of vital genomic information, genomic instability and the manifestation of diseases such as Parkinson’s, Alzheimer’s and cancer 3,4.

Reactive oxygen species (ROS) and reactive nitrogen species (RNS), which are produced under oxidative stress during normal metabolic activity or inflammatory response 1,5 are a major cause of endogenous DNA damage. Other triggers for DNA damage are exogenous factors such as exposure to environmental pollutants, chemicals in food and drugs as well as ionizing and solar UV 6,7. Understanding the type and extent of genomic DNA damage is crucial for both basic research and clinical intervention. The detection and quantification of damage may ultimately be used in determining predisposition to disease, early diagnostics and assessment of response to therapy.

The two main approaches for DNA damage detection rely either on lesion specific antibodies or on the detection of DNA integrity. Antibodies against specific damage lesions are used for colorimetric or fluorescence-based detection of both single and double strand DNA damage markers. Among the methods utilizing this approach are enzyme-linked immunosorbent assays (ELISA), Dot-blot, flow cytometry, and immunohistochemistry 8–12. The physical integrity of DNA due to single or double strand breaks (SSB & DSB) can be quantified by assays utilizing DNA unwinding such as the comet assay and other electrophoresis based techniques 13–15. Despite the wide acceptance of these methods, DNA damage detection remains challenging mostly due to lack in sensitivity and poor reproducibility. Moreover, in clinically relevant syndromes, minor changes in the amounts of DNA damage lesions can lead to sever outcomes 16,17. Addressing such changes requires high sensitivity and quantitative analysis not easily attained by existing methods. Single molecule DNA analysis may offer an alternative to sensitive quantification of various DNA lesions 18–21. This approach is based on excising DNA damage lesions enzymatically, followed by in-vitro incorporation of fluorescent nucleotides into the gap. Individual DNA molecules are then stretched on a microscope slide for imaging. DNA damage lesions are detected as fluorescent spots along the DNA contour. Despite being highly sensitive, the main limitations of the single molecule approach are its complexity and low throughput which prevent it from being broadly utilized.

Here, we present a robust, high-throughput and highly sensitive assay for the detection and quantification of type-specific single-strand DNA (ssDNA) damage. The assay is based on Repair Assisted Damage Detection (RADD). Specifically, single strand breaks and damage-lesions are replaced with fluorescent nucleotides followed by adsorption of the DNA to a defined compartment on a custom designed multi-well slide. The slide is then imaged by a standard slide scanner, and image analysis is performed to quantify the extent of DNA damage 22. This new technique, termed Rapid-RADD, enables quantifying a large number of samples in a fast and accurate manner, opening an avenue for large-scale DNA damage and repair screens for research and clinical use.

Materials and Methods

Method overview

Herein is an overview of the entire assay’s steps, described in details through the experimental section. (1) DNA extraction → (2) DNA labeling for oxidative\UV-induced damage with ATTO-550 fluorophore → (3) Slide preparation → (4) Applying DNA samples onto the activated slide → (5) Slide imaging for ATTO-550/TAMRA fluorophore → (6) Total DNA staining with EvaGreen intercalator → (7) Slide imaging EvaGreen intercalator→(8) Data Analysis

Cell culture

U2OS (human osteosarcoma) cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM), supplemented with 10% fetal bovine serum, L-glutamine (2 mM) and 1% Penicillin-Streptomycin (Gibco). Cells were incubated at 37 °C with 5% CO2.

UV irradiation

Cells were washed once with PBS, which was removed from the dish prior to exposure to UVB irradiation (302 nm). Cells were irradiated at doses of 200-1200 J/m2 in BLX chamber (Vilber). Cell medium was added to the culture prior to DNA extraction.

KBrO3 treatment

Final concentration of 50 mM or 100 mM KBrO3 was added to culture medium for one hour. Cells were washed with PBS twice, and then cell medium was added. DNA was extracted at various time points post KBrO3 treatment.

DNA extraction

For each sample, DNA was extracted from approximately 106 cells using the “GenElute-Mammalian Genomic DNA Miniprep Kit” (Sigma) according to manufacturer’s instructions.

Mice irradiation and DNA extraction

All experiments were carried out in accordance with guidelines of the Tel Aviv University Institutional Animal Care and Use Committee (IACUC approval 01-16-038). A total of three C57BL\6JOlaHsd mice at the age of 12-20 weeks were purchased from Harlan Labs. Dorsal hair was shaved two days prior to the experiment, mice were sacrificed and exposed to 10,000 J/m2 UVB (302 nm). Next, skin samples were flash frozen in liquid nitrogen at the following times points: 5 minutes, 30 minutes, 6 hours and 24 hours. DNA was extracted from thawed samples using the “GenElute-Mammalian Genomic DNA Miniprep Kit” (Sigma) according to manufacturer’s instructions.

Labeling oxidative DNA damage

KBrO3-treated DNA samples were labeled for oxidative damage in three consecutive enzymatic reactions. In the first step, each reaction tube contained 500 ng of DNA sample, 1.5 μl of 10x buffer 4 (New England Biolabs, NEB), 1.5 μl of 1 mg/ml bovine serum albumin solution (NEB), 0.3 μl of hOGG1 (ProSpec TechnoGene Ltd.) and ultrapure water to a final volume of 15 μl. The reaction mixture was incubated for 30 minutes at 37 °C. In the second step, 0.5 μl of Endo nuclease IV (NEB) enzyme was added to the reaction, and it was incubated for additional 30 minutes at 37 °C. In the final step, the following were added into each reaction tube: 1.5 μl of 10x buffer 4 (NEB), 1.5 μl of 1 mg/ml bovine serum albumin solution (NEB), 0.2 μl of 50 mM NAD+ (NEB), deoxynucleotides (A,G,C (sigma) and fluorescent ATTO-550-UTP (Jena biosciences GMBH)) to a final concentration of 100 nM, 0.4 μl of Bst DNA polymerase, Large fragment (NEB), 0.2 μl of Taq DNA ligase (NEB) and ultrapure water to a final volume of 30 μl. The reaction mixture was incubated for 30 minutes at 65 °C. The labeled DNA samples were purified from excess fluorophores using Oligo Clean & Concentrator columns (Zymo research), according to manufacturer’s recommendations, with two washing steps for optimal results.

Labeling UV-induced DNA damage

UV-irradiated DNA samples were labeled for UV damage in three consecutive enzymatic reactions. In the first step, each reaction tube contained 500 ng of DNA sample, 1.5 μl of 10x buffer 4 (NEB), 1.5 μl of 1 mg/ml bovine serum albumin solution (NEB), 0.5 μl of T4-PDG (NEB) and ultrapure water to a final volume of 15 μl. The reaction mixture was incubated for 30 minutes at 37 °C. The second and third labeling steps and the purification step are identical to respective steps in the oxidative damage labeling procedure described above.

Slides preparation

Teflon coated microscope slides (Tekdon, customized well formation, 2 mm diameter wells, 90 wells per slide) were immersed in 0.005% poly-L-lysine solution in water (Sigma), in order to positively charge the surface. The immersed slides were incubated for one hour at 37 °C with mild shaking (25 rpm) and then incubated overnight at 4°C. The following day a blocking step was performed; the slides were washed twice with PBST (0.05% Tween 20) solution and twice with PBS (Sigma), and immersed in a 5% w/v bovine serum albumin (Sigma) solution in PBS. The immersed slides were incubated for one hour at 37 °C with mild shaking (25 rpm) and then incubated overnight at 4 °C. In the final step, slides were washed with water and dried under a flow of nitrogen gas. The slides were used immediately upon drying.

Applying DNA samples onto the activated slides

1 μl labeled-DNA samples were placed in each well. The optimal DNA concentration for attachment is 10-20 ng per well. Three to five replicates of each sample were placed on the slide. Slides were incubated for 14 minutes at 42 °C and then for 24 minutes at 30 °C, in humid conditions to avoid rapid drying of the wells (Thermoshaker, Eppendorf). The slides were then washed with water and dried under a flow of nitrogen gas.

Total DNA staining

Total DNA was stained with EvaGreen DNA binding dye (Biotium). 1 μl of 1.25 μM dye (90% water, 10% DMSO) was added to each well containing the bound DNA. Wells containing only water and no DNA were also stained, in order to obtain the background signal of the EvaGreen dye in the absence of DNA. Slides were incubated for 30 minutes at room temperature in the dark. The slides were then washed with water and dried under a flow of nitrogen gas.

Slide imaging

Slides were imaged using InnoScan1100 slide scanner (Innopsys). A 532 nm green laser was used to image the ATTO-550 fluorophore. A 635 nm red laser was used to image the ATTO-647 fluorophore. A 488 nm blue laser was used to image the EvaGreen stain. The ATTO -550-labeled DNA was imaged before EvaGreen staining to avoid their co-excitation by the green laser.

Data analysis

Images were analyzed using ImageJ 30. The mean fluorescence intensity inside each well in both channels was extracted. The background signal was determined from the control replicates and subtracted from the ATTO-550 fluorescence signal (DNA damage labels) in each sample well. To account for background noise in the EvaGreen signal (total DNA), a mean fluorescence signal of all wells containing EvaGreen and no DNA was calculated and subtracted from the EvaGreen signal in each sample well. We divided the calculated ATTO-550 signal in each well by the fluorescence intensity calculated in the EvaGreen channel of the same well, in order to normalize the signal to the actual amount of DNA in the well. Next, the average and standard deviation for each sample were calculated over three to five replicates.

Quantification of DNA damage in single DNA molecules

The detailed conditions and protocol for the single molecule DNA damage-labeling assay are described in 23. Briefly, the DNA was extracted from HEK293 cells and subjected to DNA damage labeling using a type specific repair cocktail, similar to the methods described above for either UV-induced or oxidative DNA damage. Following the labeling procedure, the DNA was extended over activated glass slides. The slides were visualized using a fluorescence microscope (TILL photonics GmbH) and analyzed by an in-house software that counts and calculates the amount of DNA damage sites per length of DNA. 24

Results

Rapid-RADD takes advantage of the ability of DNA repair enzymes to specifically recognize damage lesions and create a single-strand gap in the DNA at target damage sites. This is followed by the insertion of fluorescently labeled nucleotides by a DNA-polymerase to fill the gap. The result of this reaction is fluorescence intensity, which is proportional to the level of DNA damage, and can be quantified using a commercial slide scanner. The assay workflow is illustrated in Figure 1. First, DNA is extracted from the studied cells of choice. DNA is treated with a cocktail of specific DNA repair enzymes that recognize and excise different DNA damage lesions (in this case, UV or oxidative DNA damage, Figure 1.A,B). Next, DNA polymerase and fluorescent nucleotides are introduced, leading to incorporation of fluorescent nucleotides into the gap (Figure 1.C). The labeled DNA is deposited onto a partitioned poly-L-lysine-coated glass slide ( Figure 1.D), which is then imaged by a standard slide scanner or other fluorescence imager (Figure 1.E). Finally, automatic image analysis is used to accurately quantify the amount of DNA damage (Figure 1.F). Figure S1.A shows a picture of the array slide and the hydrophobic mask that is printed on it in order to separate the different samples. This custom-made slide was designed to define 90 wells, each holding 0.5-2.0 μl of sample by surface tension. Figure S1.B shows an example of a slide scan, illustrating a constant amount of DNA (blue) and an increasing amount of UV damage (green).

Figure 1.

Figure 1

Open in a new tab

Assay workflow diagram. (A) DNA lesions of either UV-induced DNA damage or oxidative DNA damage are recognized by a specific repair enzyme. (B) The damaged lesion is excised by the repair enzyme, leaving a gap in the DNA chain. (C) DNA polymerase is introduced into the tube, along with fluorescent nucleotides to fill the formed gap. (D) The labeled DNA samples are loaded into wells on a positively charged glass slide. (E) The slide is imaged on a slide scanner, followed by image analysis (F).

In order to optimize the enzymatic repair reactions we first used single molecule detection, an approach that allows for sensitive and accurate quantification of DNA damage 19,23. Labeled DNA molecules are stretched by flow on a microscope slide and damage sites appear as fluorescent spots along the DNA molecule contour 19. Despite its complexity, single molecule detection offers direct counting of individual damage sites and is a powerful tool for optimization experiments. HEK293 cells were exposed to UVB irradiation (302 nm, 200 J/m2) or treated with 50 mM H2O2 to induce oxidative stress. Genomic DNA was labeled using a repair cocktail based on T4 phage β-glucosyltransferase (T4-PDG) to target UV-induced lesions or based on the human Oxoguanine Glycosylase 1 (hOGG1) for labeling oxidative damage. DNA samples were stretched and imaged by a sensitive fluorescence microscope and the damage sites were detected and quantified (Figure 2).

Figure 2.

Figure 2

Open in a new tab

Optimization of labeling reactions by sSingle- molecule detection and quantificationofquantification of DNA damage. HEK293 cells were exposed to DNA damage agents. The DNA samples were labeled using either the T4-PDG repair cocktail (for UVB-irradiated cells) or hOGG1 repair cocktail (for H2O2-treated cells), and the damage level was quantified.The DNA was then extracted and the damage sites were labeled with ATTO-647 fluorophores (red). The DNA molecules were stained with YOYO-1 (blue) followed by stretching on microscope cover-slip glass slides for imaging and analysis. The damage sites are seen as red dots (ATTO-647 fluorophore) on the blue (YOYO-1) DNA molecules in the representative images in A-C. (A) Control cells that were not exposed to any type of DNA damaging agent;e (B) UV-B induced DNA damage (302nm, 100 J/m2); (C) Oxidationve induced DNA damage (0.5 hours of treatment with 100 mM nM H2O2)KBrO3 for oxidative damage. (D) A plotbar graph of the calculated damage siteslevels per Mb of versus the DNA length (in base pairs). Scale bar = 20 μm.

The formation of DNA damage for both UV and H2O2 exposures is clearly visualized when comparing to a control sample that was not exposed to any of these damage agents. In order to check the reproducibility of the assay we have conducted a series of eight biological repeat experiments for both H2O2 and UV induced DNA damage. We measured a variation of 7% and 9.6% for oxidative and UV damage levels respectively, confirming the robustness of the enzymatic labeling reaction (Figure S2).

Although single molecule detection provides ultimate sensitivity, it is limited for broad use by the research and clinical communities. Specifically, the assay cannot be used without appropriate expertise and equipment, nor can it deal with a large number of samples simultaneously. In order to lift these limitations we applied the optimized labeling reactions to the partitioned slide platform thus providing a high-throughput, cheap and easy to use assay that can be adapted rapidly to any clinical or research setting.

Rapid-RADD was optimized for quantifying two types of ssDNA damage, namely UV-induced and oxidative DNA damage. Most of the existing assays dealing with UV-induced DNA damage use highly energetic UVC radiation, a portion of the UV spectrum that is almost completely blocked by the atmosphere, and is hence less significant in real-life scenarios. UVB radiation is more relevant to environmental exposures and also generates measurable UV-induced DNA photolesions 23 (See Figure S3 for comparison). We therefore used this type of radiation for further Rapid-RADD experiments. The second labeling reaction was aimed at detecting oxidative damage, with 8-Oxoguanine being the most common and important damage lesion. In order to address this damage type, we used the hOGG1 based repair cocktail. This reaction efficiently replaces 8-Oxoguanine with a fluorophore for optical detection. To test the assay, we used KBrO3 which is a strong oxidation agent also known as the food additive Formolene® or E924. The use of KBrO3 in food products is banned in most of the world (but not in the US) as it was shown to induce renal tumors in rats 25 and is classified as a category 2B carcinogen by the International Agency for Research on Cancer (IARC).

U2OS cells were irradiated with increasing levels of UVB light, and damage levels were assessed as a function of irradiation intensity (302 nm, 200-1200 J/m2). DNA was immediately extracted following irradiation and the T4-PDG repair cocktail was used to excise and label UV-induced lesions. Figure 3.A shows a linear dose response for the T4-PDG-labeled samples. Oxidative DNA damage was induced in U2OS cells by exposure to KBrO3 at increasing concentrations for one hour (0, 50 and 100 mM). DNA was extracted from the cells and labeled using the hOGG1 repair cocktail. Figure 3.B shows the increase in oxidative DNA damage with exposure to increasing concentrations of KBrO3. These experiments validated the utility of the optimized Rapid-RADD assay for quantifying UV induced and oxidative damage lesions.

Figure 3.

Figure 3

Open in a new tab

Quantification of oxidative damage and UV-induced damage in U2OS cells. (A) Linear increase in damage levels for U2OS cells exposed to increasing doses of UVB light (200-1200 J/m2) or (B) Linear increase in oxidative damage with increasing concentrations of KBrO3 (0-100 mM), followed by DNA extraction. The DNA samples were labeled using either T4-PDG repair cocktail (for UVB-irradiated cells) or hOGG1 repair cocktail (for KBrO3-treated cells), and the damage level was quantified. Data represent mean ± standard deviation (n = 5). All data was collected from two independent experiments.

Rapid-RADD is also an attractive tool for assessing the function and efficiency of the cellular DNA repair machinery by probing the levels of damage as a function of time after exposure. We next set out to measure the repair dynamics of U2OS cells exposed to 50 mM KBrO3 for one hour. DNA was extracted from cells at various time points up to 24 hours post treatment, allowing native DNA repair to occur. DNA samples from each point were labeled using the hOGG1 repair cocktail and assayed as described above. As expected, the measured damage level immediately after exposure was the highest, and it gradually decreased at later time points indicating repair of the damaged DNA (Figure 4.A, B). Notably, although significantly reduced after 24 hours, DNA damage did not return to its basal level as measured for non-treated samples.

Figure 4.

Figure 4

Open in a new tab

Repair dynamics of oxidative damage and UV-induced damage. (A) U2OS cells were incubated with 50 mM KBrO3, then washed and allowed to undergo native DNA damage repair. The DNA was extracted at several time points post-treatment, labeled using the hOGG1 repair cocktail, and further assayed as described above. Representative fluorescence image of wells for each time point (pre-treatment and 0-24 hours post-treatment); (B) Quantification results derived from the slide images and showing gradual repair of the damaged DNAr, where accurate quantification of the damage level in each sample was calculated. Data represent mean ± standard deviation (n = 5). Each bar represents the averaged results of two independent experiment. (C) C57BL\6JOlaHsd mice were irradiated at 10,000 J/m2 UVB light (302 nm), then allowed to undergo native DNA damage repair. DNA was extracted at several time points post-radiation, then labeled using the T4-PDG repair cocktail, and further assayed as described above. Each different-colored dot represents a different mouse. DataofData of Eacheach dot represents the mean damage level mean ± standard deviation (n = 5).

As a final proof of concept, we aimed to study DNA damage in the context of an in-vivo model. DNA damage and repair dynamics in a live organism are extremely challenging to study. However, despite the complexity of these biological processes in-vivo, such studies offer more comprehensive information regarding the repair process and its relation to other traits on the organism level such as nutrition, environmental exposures and disease. We used live mice as a model for quantifying DNA repair in-vivo. Three mice were subjected to UVB irradiation, followed by skin DNA extraction at different time points post irradiation, allowing native DNA repair to occur. As presented in Figure 4.C, the highest level of damage was measured immediately after UVB irradiation (5 minutes post radiation), and the damage level gradually decreased at later time points providing the dynamic repair profile.

In addition to direct quantification of DNA damage levels, the simplicity of the proposed detection concept offers facile multiplexing. By using additional fluorescent colors, the presented labeling reactions may be combined with the quantification of other genomic observables such as epigenetic DNA modifications. As a proof of concept we used two colors of fluorophores to label UV-induced damage in green (TAMRA fluorophores) and the epigenetic marker 5-hydroxymethylcytosine (5hmC) in red (ATTO-647 fluorophores) 19,22,26–28. Both signals were reliably measured simultaneously for the same DNA sample, as seen in Figure S4.

Discussion

ssDNA damage is the most common form of damage that occurs in DNA, and if not efficiently repaired, can lead to the generation of mutations and the manifestation of diseases, primarily cancer, as well as cell senescence and aging processes. A quick and cost effective method for determining DNA damage levels may complement sequencing-based mutation burden analyses for personalized treatment.

In this work, we present Rapid-RADD, a simple, high-throughput and highly sensitive method for the quantification of type-specific ssDNA damage. We have shown that oxidative or UV-induced ssDNA damage can be fluorescently labeled by specific oxidative or UV damage repair enzymes, respectively. The developed assay is facile and can be performed rapidly and with high throughput owing to compartmentalized slide-based analysis using commercially available reagents and equipment.

In the case of cancer, gene alterations are common, and the loss of DNA damage response function is linked to the accumulation of DNA damage 29. In colorectal cancer, oxidative DNA damage can promote mismatch-repair gene defects, and damage detection can predict the success of specific therapy 30,31. An example for such therapy is the synthetic lethality approach, which is now under clinical trial, for mismatch-repair defective cancers 32. In tumor cells, homologous recombination defects can be tested by measuring the RAD51 response. This can be done by administering patients with a DNA-damaging agent ex-vivo 31. Different patients respond differently to the administration of the DNA-damaging agent, and present different DNA damage levels. Our technique has the potential to detect and quantify these levels of DNA damage, and hence lead to accurate and reliable choice of therapy and dosing.

At its current constellation, Rapid-RADD enables the quantification of up to 90 samples on a single slide. Our method was found suitable for the detection and quantification of both UV and oxidative-induced ssDNA damage, and was successful in assessing ssDNA damage in conjunction with quantification of the epigenetic modification 5hmC in the same samples. Finally, we were able to characterize and quantify the repair dynamics of DNA damage and specifically, to follow the repair of UV-induced damage in vivo. The developed method is therefore suitable for analyzing a wide variety of DNA samples, and could be easily applied for both clinical and research purposes.

Acknowledgement

Y.E. acknowledges support from the BeyondSeq consortium (EC program 634890), the European Research Council Proof of Concept grant (grant no. 767931), the European Research Council starter grant (grant No. 337830), the US National Institute of Health (grant R21ES028015-011), and the Joint Israeli German R&D nanotechnology (grant no. 61976). C.L. acknowledges grant support from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programmer (grant agreement no. 726225)

Funding

This work was supported by the BeyondSeq consortium (EC program 634890), the European Research Council Proof of Concept grant (grant no. 767931), the European Research Council starter grant (grant No. 337830), the US National Institute of Health (grant R21ES028015-011), and the Joint Israeli German R&D nanotechnology (grant no. 61976).

Footnotes

Conflict of Interest

The authors declare that they have no competing interests. YE, SZ, YM are listed as inventors on a patent application related to the methods described in this manuscript.

References

  • 1.De Bont R, Van Larebeke N. Mutagenesis. 2004;19:169–185. doi: 10.1093/mutage/geh025. [DOI] [PubMed] [Google Scholar]
  • 2.Lindahl T, Barnes D. Cold Spring Harbor symposia on quantitative biology. Cold Spring Harbor Laboratory Press; 2000. pp. 127–134. [DOI] [PubMed] [Google Scholar]
  • 3.Markesbery WR, Lovell MA. Antioxidants & redox signaling. 2006;8:2039–2045. doi: 10.1089/ars.2006.8.2039. [DOI] [PubMed] [Google Scholar]
  • 4.Tsang AH, Chung KK. Biochimica et Biophysica Acta (BBA)-Molecular Basis of Disease. 2009;1792:643–650. doi: 10.1016/j.bbadis.2008.12.006. [DOI] [PubMed] [Google Scholar]
  • 5.Dedon PC, Tannenbaum SR. Archives of biochemistry and biophysics. 2004;423:12–22. doi: 10.1016/j.abb.2003.12.017. [DOI] [PubMed] [Google Scholar]
  • 6.Sinha RP, Häder D-P. Photochemical & Photobiological Sciences. 2002;1:225–236. doi: 10.1039/b201230h. [DOI] [PubMed] [Google Scholar]
  • 7.Wogan GN, Hecht SS, Felton JS, Conney AH, Loeb LA. Seminars in cancer biology. Elsevier; 2004. pp. 473–486. [DOI] [PubMed] [Google Scholar]
  • 8.Wani AA, Gibson-D’Ambrosio RE, D’Ambrosio SM. Photochemistry and photobiology. 1984;40:465–471. doi: 10.1111/j.1751-1097.1984.tb04619.x. [DOI] [PubMed] [Google Scholar]
  • 9.Yoshida R, Ogawa Y, Kasai H. Cancer Epidemiology and Prevention Biomarkers. 2002;11:1076–1081. [PubMed] [Google Scholar]
  • 10.Manis JP, Morales JC, Xia Z, Kutok JL, Alt FW, Carpenter PB. Nature immunology. 2004;5:481. doi: 10.1038/ni1067. [DOI] [PubMed] [Google Scholar]
  • 11.Cosaceanu D, Budiu R, Carapancea M, Castro J, Lewensohn R, Dricu A. Oncogene. 2007;6:2423. doi: 10.1038/sj.onc.1210037. [DOI] [PubMed] [Google Scholar]
  • 12.Bonner WM, Redon CE, Dickey JS, Nakamura AJ, Sedelnikova OA, Solier S, Pommier Y. Nature Reviews Cancer. 2008;8:957. doi: 10.1038/nrc2523. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Singh NP, McCoy MT, Tice RR, Schneider EL. Experimental cell research. 1988;175:184–191. doi: 10.1016/0014-4827(88)90265-0. [DOI] [PubMed] [Google Scholar]
  • 14.Potter AJ, Gollahon KA, Palanca BJ, Harbert MJ, Choi YM, Moskovitz AH, Potter JD, Rabinovitch PS. Carcinogenesis. 2002;23:389–401. doi: 10.1093/carcin/23.3.389. [DOI] [PubMed] [Google Scholar]
  • 15.Gudmundsson B, Thormar HG, Sigurdsson A, Dankers W, Steinarsdottir M, Hermanowicz S, Sigurdsson S, Olafsson D, Halldorsdottir AM, Meyn S. Nucleic acids research. 2018;46:e118–e118. doi: 10.1093/nar/gky645. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Jackson SP, Bartek J. Nature. 2009;461:1071. doi: 10.1038/nature08467. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Richter C. The international journal of biochemistry & cell biology. 1995;27:647–653. doi: 10.1016/1357-2725(95)00025-k. [DOI] [PubMed] [Google Scholar]
  • 18.Lee J, Park HS, Lim S, Jo K. Chemical Communications. 2013;49:4740–4742. doi: 10.1039/c3cc38884k. [DOI] [PubMed] [Google Scholar]
  • 19.Zirkin S, Fishman S, Sharim H, Michaeli Y, Don J, Ebenstein Y. Journal of the American Chemical Society. 2014;136:7771–7776. doi: 10.1021/ja503677n. [DOI] [PubMed] [Google Scholar]
  • 20.Lee J, Kim Y, Lim S, Jo K. Analyst. 2016;141:847–852. doi: 10.1039/c5an01875g. [DOI] [PubMed] [Google Scholar]
  • 21.Müller V, Dvirnas A, Andersson J, Singh V, KK S, Johansson P, Ebenstein Y, Ambjörnsson T, Westerlund F. Nucleic acids research. 2019 doi: 10.1093/nar/gkz489. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Margalit S, Avraham S, Shahal T, Michaeli Y, Gilat N, Magod P, Caspi M, Loewenstein S, Lahat G, Friedmann-Morvinski D. International journal of cancer. 2019 doi: 10.1002/ijc.32519. [DOI] [PubMed] [Google Scholar]
  • 23.Torchinsky D, Michaeli Y, Gassman NR, Ebenstein Y. Chemical Communications. 2019;55:11414–11417. doi: 10.1039/c9cc05198h. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Jain N, Shahal T, Gabrieli T, Gilat N, Torchinsky D, Michaeli Y, Vogel V, Ebenstein Y. bioRxiv. 2019 doi: 10.1080/15592294.2019.1638700. 587311. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Kurokawa Y, Maekawa A, Takahashi M, Hayashi Y. Environmental health perspectives. 1990;87:309–335. doi: 10.1289/ehp.9087309. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Shahal T, Gilat N, Michaeli Y, Redy-Keisar O, Shabat D, Ebenstein Y. Analytical chemistry. 2014;86:8231–8237. doi: 10.1021/ac501609d. [DOI] [PubMed] [Google Scholar]
  • 27.Nifker G, Levy-Sakin M, Berkov-Zrihen Y, Shahal T, Gabrieli T, Fridman M, Ebenstein Y. ChemBioChem. 2015;16:1857–1860. doi: 10.1002/cbic.201500329. [DOI] [PubMed] [Google Scholar]
  • 28.Gilat N, Tabachnik T, Shwartz A, Shahal T, Torchinsky D, Michaeli Y, Nifker G, Zirkin S, Ebenstein Y. Clinical epigenetics. 2017;9:70. doi: 10.1186/s13148-017-0368-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Knijnenburg TA, Wang L, Zimmermann MT, Chambwe N, Gao GF, Cherniack AD, Fan H, Shen H, Way GP, Greene CS. Cell reports. 2018;23:239–254. doi: 10.1016/j.celrep.2018.03.076. e236. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Martin SA, Hewish M, Sims D, Lord CJ, Ashworth A. Cancer research. 2011;71:1836–1848. doi: 10.1158/0008-5472.CAN-10-2836. [DOI] [PubMed] [Google Scholar]
  • 31.Lord CJ, Ashworth A. Nature. 2012;481:287. doi: 10.1038/nature10760. [DOI] [PubMed] [Google Scholar]
  • 32.Martin SA, McCarthy A, Barber LJ, Burgess DJ, Parry S, Lord CJ, Ashworth A. EMBO molecular medicine. 2009;1:323–337. doi: 10.1002/emmm.200900040. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES