DNA Damage: Measurement of Markers
Analytical methods aiming at measuring oxidative DNA lesions have been developed during the past decades and can roughly be grouped into direct and indirect approaches, respectively 381. Direct approaches require extraction and hydrolytic cleavage of cellular DNA, followed by chromatographic separation and detection of the constituents. Characteristic direct approaches include high-performance liquid chromatography with electro-chemical detection (HPLC-ECD), and high-performance liquid chromatography coupled with tandem mass spectroscopy (HPLC-MS/MS). Indirect methods require alkaline cell lysis in conjunction with N-glycosylases and encompass, for instance, enzymatic/immunological methods such as immunoassays as well as polymerase chain reaction (PCR)-based reactions and single-cell gel electrophoresis (comet assay) 381.
High performance liquid chromatography-electrochemical detection (HPLC-ECD)
HPLC coupled to electrochemical detection (HPLC-ECD) operating in the oxidative mode is a highly selective method for the detection of 8-OH-dG as a marker for oxidative DNA damage 444-446. This technique is based on the lower redox potential of 8-OH-dG in comparison to that of the unmodified nucleoside. It is well-documented that 8-OH-dG is oxidized only when the relevant potential has been adopted 381. It has been shown previously that the combination of HPLC with ECD markedly enhances the sensitivity of the assay 447. Notably, DNA preparation and processing preceding HPLC measurement constitute crucial steps as oxidative processes can occur to nucleosides and nucleotides. Therefore, various approaches had been performed aiming at optimizing the extraction step using the “chaotropic” method (i.e., with NaI as a salting-out agent) and minimizing the artificial increase of DNA adducts such as 8-oxo-dG and 8-oxo-G 448. An important further aspect relates to the nature of the biological material from which DNA should be isolated. With respect to 8-OH-dG, the modified DNA base can be found not only in cells and tissues but also in urine, where it is supposed to indicate DNA repair mechanisms in the body 449. In order to determine the urinary 8-OH-dG content in a sample, solid-phase extraction technique is recommended enabling the enrichment of the specific parameter 450. For an overview of the chromatographic conditions used for the detection of 8-OH-dG, see Dabrowska and Wiczkowski (2017) 381. Likewise, further electrochemically actively modified bases and nucleosides can be determined by HPLC-ECD using higher oxidation potentials such as 8-oxo-A and FapyG, 5-hydroxyuridine (5-OH-U), and 5-hydroxycytidine (5-OH-Cyt) 451.
High performance liquid chromatography coupled with tandem mass spectrometry (HPLC-MS/MS)
HPLC-ESI-MS/MS and the sensitive HPLC-MS3 can be considered as being the gold standard methods for the detection of modified nucleosides in biological samples from different sources 266, 452-454. This method, which allows for the detection of various oxidized DNA components including 8-OH-dG, takes advantage of the electrospray ionization technique whose high sensitivity and flexibility ensures the identification of the required DNA adducts 455, 456. It is of note that an option to adjust for any deprivation through DNA digestion and changes in ionization efficiency during analysis is given by the availability of synthetic [13C]- and [15N]-labeled internal standards, facilitating the utilization of isotope dilution 457, 458. Meanwhile, several HPLC-ESI-MS/MS approaches have been established to determine single and clustered DNA lesions in biological samples in addition to the primary determination of 8-OH-dG and 8-oxo-dG, respectively 459, 460. Artificial oxidation of DNA could be reduced significantly by the addition of transition metal chelators and anti-oxidants prior to HPLC-MS/MS 461, 462. In addition, a minimum of 30 µg of DNA is required in order to avoid artificial DNA autoxidation 463. As outlined by Cadet and colleagues in 2017, artificial autoxidation restricts the applicability of HPLC-MS/MS measurements as they are only valuable for DNA samples with a markedly enhanced damage degree 463. It should be noted that the HPLC-MS/MS technology is inappropriate to determine oxidatively damaged DNA in mitochondria as the amount of mitochondrial DNA is rather low and thus highly susceptible to autoxidation. One exception is 8-oxoG and 8-oxo-dG that is protected against artificial autoxidation during preparation 453.
Gas chromatography-mass spectrometry (GC-MS)
Gas chromatography coupled with mass spectrometry (GC-MS) has been established as a technique to determine the formation of radiation-induced modifications of nucleobases and nucleosides in the mid 1980s. This technique involves the acidic hydrolysis of DNA culminating in the release of free bases 464, 465. In the initial step, intact and oxidatively damaged nucleobases are transformed into volatile trimethylsilylated derivatives by heating at 140°C for 30 min prior to GC-MS analysis in order to enhance performance. On the contrary, the determination of 5’,8-cyclo-2’-deoxyguanosine (cdG) in isolated and cellular DNA was achieved after enzymatic digestion of the analyte by using a mixture of certain exonucleases 466-468. Such derivatization is a prerequisite for the mass-spectrometric detection of various oxidative DNA damages which uses at least two specific ions to safeguard suitable specificity detection 469. However, a major disadvantage of the derivatization step obviously relates to the concomitant degradation of oxidatively damaged bases such as FapyA and FapyG 470. In an alternative approach, the repair endonucleases Fpg and Endo III were used instead of acid hydrolysis to avoid artifact formation and to liberate the base products from unmodified DNA samples, allowing for the accurate detection of FapyA, FapyG, and 8-oxo-G in unmodified DNA samples 471. However, only a few studies have been described in the literature that utilize this auspicious approach 472, 473.
A further direct method for measuring oxidative DNA damage is represented by the [32P]post-labeling method. This method was originally described by Randerath and colleagues for the precise detection of DNA lesions induced by reactive chemicals or genotoxins 474. This method is characterized by enzymatic hydrolysis of the DNA yielding nucleoside 3′-monophosphates that are subsequently radiolabeled using [γ-32P]ATP and T4 polynucleotide kinase. DNA adducts are separated from their normal counterparts by thin layer chromatography (TLC) 474-476. During the past years, the original procedure has been revised manifoldly but only the following modifications have attracted much interest of the scientific community 381. This includes, on the one hand, the application of nuclease P1 prior to the labeling process allowing for the dephosphorylation of unmodified nitrogenous bases and, on the other hand, the isolation of hydrophobic DNA adducts using the phase transfer agent tert-butylammonium chloride. Consequently, hydrophobic adducts are exclusively found in the organic phase and are subjected to adjacent isotopic labeling 475.
Alkaline elution, comet assay, DNA unwinding, ligation-mediated polymerase chain reaction (LM-PCR)
Indirect, enzymatic methods wed the use of specific DNA glycosylases with the detection of DNA strand breaks. These DNA repair glycosylases incise oxidized DNA lesions to form a strand break which can be detected by an appropriate technique such as the comet assay, alkaline elution, and DNA unwinding assays 477. The comet assay (single-cell gel electrophoresis) is widely utilized to determine oxidized bases and DNA strand breaks. For the comet assay, cells are embedded in agarose and transferred onto a microscope slide followed by alkaline hydrolysis using specific DNA N-glycosylases (recognizing damaged purines, particularly 8-OH-dG) or uvrABC (for bulky lesions). Digests are separated via gelelectrophoresis in which DNA migration occurs in the presence of strand breaks only. Stained gels are analyzed by fluorescence microscopy in which migrated DNA loops are visualized as a ‘comet tail’. Results are quantitated by calculating the percentage of cells with tails as a measure of the oligonucleotide break frequency 477-480. A modified version of the comet assay has recently been used to monitor genotoxic actions of environmental stressors in the form of nanomaterials 481.
Another indirect approach is represented by alkaline elution in which cells are subjected to alkaline hydrolysis using the formamidopyrimidine DNA N-glycosylase (FPG) that induces the formation of DNA single-strand breaks (SSB). SSB are then eluted at alkaline pH, the elution time correlating directly to the number of strand breaks 477, 482. Alkaline unwinding of DNA has also been utilized to quantitate DNA strand breaks, providing a fast and sensitive tool to determine the amount of oxidative DNA lesions in mammalian cells 483. Herein, DNA is allowed to unwind at alkaline pH in the presence of FPG. After neutralization and sonication, single- and double-stranded DNA are separated and quantified. The number of DNA strand breaks is inversely correlated to the percentage of double-stranded DNA. A common feature in measuring oxidatively damaged DNA is the use of specific DNA N-glycosylases which transform a DNA lesion into a strand break. The most widely used glycosylases comprise FPG and endonuclease III (Endo III) from E. coli. While Endo III is specific for oxidized pyrimidines, FPG specifically cleaves oxidized purines, including 8-oxo-dG and its corresponding formamidopyrimidine derivative 2,6-diamino-4-hydroxy-5-formamidopyrimidine (FapyG) 484-486. As 8-oxo-dG is predominantly produced in DNA and FPG can excise that lesion efficiently, it is consistently assumed that the amount of FPG-sensitive sites correlates directly to the amount of 8-oxo-dG. However, one drawback is that FPG has also been shown to detect and cleave alkylated DNA adducts 487. This drawback might be avoided by the use of more specific DNA glycosylases such as the human 8-oxoguanine glycosylase, hOGG1, because 8-oxo-G and FapyG are the only substrates for this enzyme 488.
Ligation-mediated polymerase chain reaction (LM-PCR) adds to the indirect approaches in measuring oxidative DNA lesions. As outlined by Ravanat in 2005, this method allows for the detection of oxidized DNA bases such as 8-OH-dG together with the identification of the chromosomal lesion location 477. Moreover, LM-PCR has also been found as being suitable for the mapping of UV-induced DNA photoproducts such as cyclobutane pyrimidine dimers 489.
Thymidine glycol (5,6-dihydro-5,6-dihydroxythymidine; Tg) represents one of the most abundant oxidation products of thymidine and can be formed by a great variety of exogenous and endogenous stressors 383, 384. Early attempts of measuring radiation-induced Tg include the indirect assessment of radiolabeled and γ-irradiated Tg by determining 2-methylglycerol or acetol that arises from borohydride reduction 490 and alkaline degradation, respectively 491. In the course of the past years, a broad spectrum of chromatographic analyses as well as immunological and enzymatic approaches has been designed. Amongst them are next-generation sequencing and array-based hybridization methods enabling the mapping of modified bases and its oxidation derivatives at the nucleoside level in the genome (for a review see 455, 492, 493).
Immunoassays (ELISA/EIA, RIA, immunoprecipitation, immunohistochemistry)
Indirect approaches used for the detection of oxidatively damaged DNA bases include immunologically based methods such as radioimmunoassay (RIA) and enzyme-linked immunosorbent assay (ELISA). Original attempts in the 1980s referred to the design of mono- and polyclonal antibodies directed against Tg 494, 495. Afterwards, efforts focussed on the development of immunoassays for 5-OH-dG (8-oxo-dG), and 5-oxo-dA, respectively 496-498. Amongst them, ELISA technology evolved into the most widely used test for the measurement of oxidative DNA lesions. For the detection of 8-OH-dG in biological material, a competitive ELISA using a capture antigen has been established which is commercially available 498-503. Antibodies against 8-OH-dG have also been utilized for sensitive immunofluorescent detection in various tissues with numerous clinical applications 504-510. Monoclonal antibodies against 8-OH-dG and 1,N6-propano-2′-deoxyadenosine (acrolein-dA) were also successfully used to immunoprecipitate DNA fragments containing the respective modified base 511. Yoshihara and colleagues analyzed the genome-wide distribution profile of 8-oxo-G in normal rat kidney by combining immunoprecipitation by anti-8-oxo-G antibodies with a microarray that covers rat genome 512. The data raised clearly revealed that 8-oxo-G is preferentially located at gene deserts. In current applications mentioned above, the monoclonal antibody N45.1 is predominantly used for the immunological detection of 8-oxo-G in human tissues and body fluids 513, 514. The group of Toshio Mori generated a monoclonal antibody specific for 5’,8-cyclo-2’-deoxyadenosine (cdA) in single-stranded DNA which effectively recognizes this bulky tandem base-sugar modification in situ 515. A novel electrochemical immunosensor was generated by Yang et al. for the quantitation of 5-hydroxymethylcytosine (5-hmC) in genomic DNA using an anti-5-hmC antibody 516. The findings clearly demonstrated a marked reduction of the 5-hmC levels in human breast cancer tissues compared to normal tissues. A commercially available antibody against 5-hmC was used by the group of Petra Hajkova for the quantitation of 5-hmC in mouse zygotes using immunofluorescence analysis and ultra-sensitive LC/MS 517.
Recent data revealed that 5-hmC can be oxidized to 5-formylcytosine (5-fC) and 5-carboxycytosine (5-caC) 518. Immunological detection of 5-caC in somatic cells could be achieved by the use of a respective polyclonal antibody 519. A biotin-avidin mediated enzyme-based immunoassay (EIA) was designed by Chowdhury and collaborators in order to evaluate the levels of 5-hmC, 5-caC, and 5-fC in the genomic DNA from the blood of cancer patients 520. The authors found that this biotin-avidin mediated EIA which uses polyclonal antibodies specific to 5-hmC, 5-caC, and 5-fC, facilitates sensitive detection of these rare DNA methylation derivatives. They also observed a prominent reduction in the 5-hmC levels in the blood of cancer patients. However, further studies are warranted in order to verify the applicability of this approach in measuring DNA methylation products in human body fluids.
New tools to investigate DNA DSB in vivo include, for instance, the H2AX/MDC1 luciferase reporter system in which H2AX and the C-terminal BRCT domains of MDC1 are fused to half of luciferase 521. Induction of DSB induces H2AX phosphorylation around the lesion followed by the reconstitution of a functional luciferase whose activation is detectable subsequently. Several approaches aiming at replacing classical antibodies have also been introduced and comprise genetically encoded fluorescent single-chain variable fragments (scFV) representing a fusion of the variable regions of the heavy and light chains of immunoglobulins connected to a linker peptide, e.g. DNA-PKcs 522. A further interesting tool poses the use of genetically encoded fluorescent single-domain antibodies (sdAb) 523. One example is anti-PCNA sdAb fused to a GFP tag which recognizes PCNA and enables the replication analysis in vivo and in real-time 524. To study DNA DSB induction and repair, alternative techniques are related to the immunological detection of certain post-translational modifications, such as phosphorylation of H2AX. Radio- or fluorescently labeled anti-γH2AX, fused to the HIV transcription activator (Tat) as an entry peptide, has been shown to penetrate living cells and to bind to DSB 525. However, much effort has to be invested in the development of improved and more specific DDR markers. Considerable progress has been made in the development of more accurate ELISAs for the determination of DNA damage. Improvements could be achieved by tyramide signal amplification (TSA), enhanced polymer one-step staining (EPOS), and time resolved amplified cryptate emission (TRACE). For more detailed information in this regard see review by Boguszewska et al. (2019) 526.