The use of DNA repair inhibitors and the comet assay—an overview

Abstract

The standard comet assay detects DNA strand breaks and alkali-labile sites, but these lesions are nonspecific. They may result directly from genotoxic agents or arise as intermediates during the repair of other DNA damage, such as oxidized bases or bulky DNA adducts. Different approaches have been developed to generate or trap these repair intermediates, making them detectable with the comet assay. Recently, the combination of the comet assay with DNA repair inhibitors like hydroxyurea and cytosine arabinoside has been proposed to detect bulky DNA adducts. These lesions are mainly repaired through nucleotide excision repair, a process that transiently produces strand breaks when damaged oligonucleotides are excised. Normally, these intermediates are rapidly repaired by DNA resynthesis and ligation. However, by inhibiting this repair step, strand breaks persist and can be detected by the comet assay. This strategy has been applied in various fields, including genotoxicity testing, environmental toxicology, human biomonitoring, and studies on DNA repair kinetics. This review focuses specifically on the use of hydroxyurea, cytosine arabinoside, and aphidicolin in in vitro experiments to evaluate the utility and specificity of this method for detecting different types of DNA lesions. Notably, in ~70% of studies reviewed, the inclusion of DNA repair inhibitors led to a significant increase in DNA damage, highlighting the added value of this approach. However, although the method enhances sensitivity to bulky adducts, it also responds to other types of damage, such as those induced by alkylating or oxidative agents.

Introduction

The comet assay is a simple and versatile method to detect DNA strand breaks (SBs) and alkali labile sites (ALS) that was first introduced in 1984 by Ostling and Johanson [1] and modified to its more common version a few years later by Singh et al. [2]. This assay can be applied to almost all eukaryotic cell types, tissue, or human samples from which a cell suspension can be obtained. Briefly, cells are embedded into agarose and lysed to obtain the nucleoids eliminating membranes and cell components other than DNA. Then the nucleoids are subjected to an alkaline treatment (pH > 13) where the double-stranded DNA unwinds, and ALS are converted into SBs, allowing their detection. Lastly, an electrophoresis field is applied. In the presence of SBs, DNA loops are relaxed and migrate from the scaffold creating what is known as the comet tails. Finally, gels are stained and scored using a fluorescence microscope. The tail intensity (% of DNA in the comet tail) is one of the parameters most used, and it is directly proportional to the number of SBs [3].

Importantly, the protocol can be modified in various ways to enhance sensitivity and specificity towards some DNA lesions, enabling the detection of DNA lesions beyond SBs and ALS [3–5]. These include (i) the incubation of nucleoids with lesion-specific enzymes to detect altered bases, (ii) the addition of a step involving DNA-breaking agents after cell treatment to reveal crosslinks (DNA migration will be reduced after treatment if crosslinks—especially intrastrand crosslinks—are present), and (iii) the use of DNA repair synthesis inhibitors to prevent strand rejoining during the repair process, thereby accumulating DNA breaks as incomplete repair sites. This publication focuses on the latter approach, which, strictly speaking, is a modification of the cell treatment.

Recently, the comet assay in combination with DNA repair inhibitors has been proposed as a simple method for the detection of bulky DNA adducts, which occur when a large chemical group attaches to the DNA molecule, distorting the DNA structure and interfering with DNA replication and transcription. These pre-mutagenic lesions do not affect DNA migration and, therefore, cannot be detected with the comet assay. However, the use of DNA repair inhibitors that block DNA resynthesis allows the accumulation of DNA repair intermediate breaks, which can then be detected with the comet assay. Ngo et al. [6], have demonstrated the effectiveness of the use of hydroxyurea (HU) and cytosine arabinoside (Ara-C) for detecting bulky DNA adducts induced by benzo(a)pyrene (B[a]P) and aflatoxin B₁ (AFB₁). Indeed, this publication is referred to as a method for the measurement and detection of bulky DNA adducts in the Key Event (KE 1879) ‘bulky DNA adducts increase’ that was included in the Adverse Outcome Pathway (AOP) ‘Bulky adducts leading to mutation’ in 2021 [7].

The most commonly used DNA repair inhibitors in combination with the comet assay are HU, Ara-C, and aphidicolin (APC) [8]. APC is an antimitotic and antiviral metabolite of the mould Cephalosporium aphidicola that inhibits DNA replication by specifically inhibiting B-family DNA polymerases, which includes Pol α, Pol δ, Pol ε, and Pol ζ [9]. Ara-C is a synthetic structural analogue of deoxycytidine, a natural nucleoside incorporated into DNA during replication. Ara-C competes with deoxycytidine triphosphate, the active form of deoxycytidine and a normal substrate for DNA polymerases, for incorporation into the growing DNA strand [10]. Once incorporated, Ara-C acts as a chain terminator preventing the elongation by DNA polymerases. It is also recognized as a potent inhibitor of polymerase α and a weak inhibitor of polymerase β [11]. HU inhibits ribonucleotide reductase, an enzyme responsible for producing deoxyribonucleotides (dNTPs), which results in a depletion of dNTP pools [12], mainly purine dNTPs. Additionally, it has been shown that HU generally disrupts the coordination between the excision and resynthesis steps of excision repair, resulting in the accumulation of DNA SBs [13]. Moreover, it has been demonstrated that HU increases the effectiveness of Ara-C; therefore these two inhibitors are commonly used together to enhance the inhibitory effect on DNA repair.

The use of DNA repair inhibitors together with the comet assay has been used for different purposes, such as evaluation of repair capacity and kinetics, genotoxicity testing, ecotoxicology, or human biomonitoring. This review focuses on the use of APC, Ara-C, HU, and Ara-C/HU in combination with the comet assay in in vitro studies, with the aim to investigate the specificity of the approach towards different DNA lesions and the experimental conditions used.

Materials and methods

Search strategy and data source

The database PubMed was used to search for scientific articles published up to December 2024 in which the comet assay in combination with APC, Ara-C, HU, or the combination Ara-C/HU was used. The following search strategy and keywords in the title and abstract were used: (‘comet assay’ OR ‘single cell gel electrophoresis’) AND (‘APC’ OR ‘aphidicolin’ OR ‘ARA-C’ OR ‘cytosine arabinoside’ OR ‘1-beta-D-arabino-furanosylcytosine’ OR ‘HU’ OR ‘hydroxyurea’). The search strategy flowchart is presented in Fig. 1.

Figure 1.

Search strategy flowchart in PubMed.

Study selection

Phase 1. The selection process was carried out by screening the titles and abstracts of all the scientific articles resulting from the search using the following inclusion and exclusion criteria:

Inclusion criteria:

1. Original research articles written in English.

2. In vitro studies in which at least one inhibitor was used in combination with the comet assay. This includes the studies in which human samples (e.g. lymphocytes or whole blood from healthy individuals) were treated in vitro to assess genotoxicity or to study DNA repair.

3. DNA repair studies using human samples (e.g. lymphocytes or whole blood) from healthy volunteers exposed to a known genotoxic agent since these studies will also give information related to the aim of this review.

4. Studies that included the standard comet assay (without inhibitors) in the same samples as part of their methodology.

5. Studies evaluating the effects of defined, individual chemical or physical agents (i.e. not mixtures).

Exclusion criteria:

1. Studies investigating DNA repair inhibitors primarily as therapeutic agents for cancer or other diseases.

2. Studies focused specifically on the APC gene.

3. In vivo studies involving animals.

4. Human biomonitoring studies in which the only objective was to compare the DNA repair capacity of groups with different exposure (e.g. comparing smokers vs. non-smokers).

5. Systematic reviews and meta-analyses (not related to the aim of this paper), conference abstracts, or other forms of secondary literature.

If the title and abstract did not provide enough clear information to decide on the inclusion or exclusion of an article, the article was labelled as ‘doubtful’ and moved to Phase 2 for a more detailed full-text review.

Phase 2. Articles selected during Phase 1, along with those categorized as ‘doubtful,’ were retrieved in full and thoroughly examined. The same inclusion and exclusion criteria used in Phase 1 were applied during this stage.

Data extraction

Data from the full-text articles selected in the previous stages were extracted and organized into four main tables, each categorized based on the DNA repair inhibitor used: Table 1—Comet assay with APC; Table 2—Comet assay with HU and Ara-C; Table 3—Comet assay with Ara-C alone; Table 4—Comet assay with HU alone.

Table 1.

Comet assay with APC.

Reference	Experimental system	Agent (treatment duration)	Inhibitor treatment	Results standard comet assay	Results with inhibitors	Cytotoxicity information
(A) Genotoxicity studies
[14]	Human skin model (3D)	2-AAF (48 h) N-OH-2-AAF (3 h) N-OH-2-AF (3 h) B[a]P (PC) (48 h)	5 μg/ml APC added 4 h before the end of the incubation	Negative Positive Positive Negative	Positive/↑ Positive/↑ Positive/↑ Positive/↑	AK release and ATP content
[15] Interlaboratory study: each compound was tested in 3 laboratories. Discrepancies between laboratories’ results are included in the corresponding column. In principle only compounds that tested negative in the standard comet assay were further tested at one concentration with APC.	Full thickness skin models: EpiDerm™ and Phenion®	MMC (48 h) Cadmium chloride (48 h) DMBA (48 h) Propyl gallate (48 h) Eugenol (48 h) DEHP (48 h) Cyclohexanone (48 h) B[a]P (PC) (48 h)	5 μg/ml APC added 4 h before the end of the 48 h exposure period	Negative Positive (CR) Negative (Laboratory B: positive) Negative Negative Negative Negative Negative	Positive Negative/↔ (The concentration tested with APC was also negative without APC) Positive/↑ (Laboratory B was the only one that tested DMBA with APC) Negative/↔ Negative/↔ Negative/↔ Negative/↑ (Laboratory D did not observe an increase) Positive	ATP content
[16]	A549 HeLa TK6 V79	BPDE (2 h) MMS (2 h) BPDE (2 h) MMS (2 h) BPDE (2 h) MMS (2 h) BPDE (2 h) MMS (2 h) BPDE (2 h) MMS (2 h) BPDE (2 h) MMS (2 h) BPDE (2 h) MMS (2 h) BPDE (2 h) MMS (2 h)	15 μM APC for 2 h 15 μM APC 30 min before treatment 15 μM APC for 2 h 15 μM APC 30 min before treatment 3 μM APC for 2 h 3 μM APC 30 min before treatment 15 μM APC for 2 h 15 μM APC 30 min before treatment	+ + + + + + + + + + + + + + + +	Equal Equal Increase Equal Increase Equal Increase Equal Increase Increase Increase Increase Equal Equal Equal Equal	In TK6 cells, 15 μM APC could not be evaluated because this concentration induced a very high percentage of hedgehogs.
[17]	A549	BPDE (2 h)	15 μM APC the last 30 min before analysis 15 μM APC for 2 h	+ +	Equal Equal
[18]	PBMC PHA-stimulated	BPDE (2 h)	3 μM APC the last 2 h of post-incubation period (2 h treatment+18 h post-incubation)	Positive	Increase
[19]	PBMC newborns PHA-stimulated PBMC mothers PHA-stimulated	BPDE (2 h)	0.5 μg/ml APC added 30 min before treatment +2 h co-exposure	Positive Positive	Increase/Positive Increase/Positive
[20]	Lymphocytes PHA-stimulated Whole blood PHA-stimulated	BPDE (2 h)	0.5 μg/ml APC added 30 min before exposure to BPDE +2 h co-treatment	Positive Positive	Increase Increase
[21]	PBMC PHA-stimulated	BPDE (2 h)	0.5 μg/ml APC added 30 min before exposure to BPDE +2 h co-treatment	Positive	Increase	Cell proliferation
[22]	Whole blood PHA-stimulated Whole blood unstimulated	BPDE (2 h) BCNU (2 h) MMS (2 h) BPDE (2 h) BCNU (2 h) MMS (2 h) BPDE (2 h) Gamma radiation (not clear) BPDE (2 h) MMS (2 h) BPDE (2 h) MMS (2 h)	0.5 μg/ml APC for 2 h 5 μg/ml APC for 2 h 5 μg/ml APC only 30 min before treatment 0.5 and 5 μg/ml APC added 30 min before treatment and 30 min post incubation 0.5 μg/ml APC for 2 h 5 μg/ml APC for 2 h	+ + + + + + + + + + + +	Increase Increase Increase Increase Increase Increase Increase Equal Increase Increase Increase Increase
[23]	Heparinized blood (unstimulated) from smokers and non smokers	BPDE (2 h)	0.5 μg/ml APC for 2 h	+	↑
[24]	CHO	UV light (not clear) RF (not clear)	10 μM APC added immediately prior to exposure	Positive Positive	Increase Increase	Trypan blue exclusion
[25]	MRC5CB1 XP12ROSV (XP cell line)	UV (not clear) 4-NQO (2 h) B[a]P + S9 (2 h) DMBA+S9 (2 h) UV (not clear) 4-NQO (2 h) B[a]P + S9 (2 h) DMBA+S9 (2 h)	15 μM APC for 30 min after UV treatment 15 μM APC for 2 h 15 μM APC for 30 min after UV treatment 15 μM APC for 2 h	Positive Positive Positive (CR) Positive (CR) Negative Negative Negative Positive	Increase Increase Increase Increase Equal Equal Equal Equal	Plating efficiency: XP cell line showed increased cell killing
(B) DNA repair studies
[26]	PBMC unstimulated	H2O2 (10 min)	3 μM APC for 45 min after treatment 6 μM APC for 45 min after treatment	+ +	↑ ↑
[27]	PHA-stimulated human lymphocytes	BPDE (2 h) MMS (2 h)	3 μM APC the last 30 min of lymphocyte stimulation	+ +	↑ ↑
[28]	Human lymphocytes PHA-stimulated	UVC (15 s)	5 μM APC for 1 h after UV irradiation 5 μM APC for 4 h after UV irradiation	Positive Negative	Increase Increase
[29]	Peripheral blood lymphocytes	UVC (not clear)	1 μM APC for 90 after UV irradiation 1 μM APC for 240 min after UV irradiation	+ =	Equal Increase
[30]	NC37 B-lymphoblast	UVA (from 1 to 30)	15 μM APC for 80 min after irradiation	-	↑

Agent abbreviations are described in Table 5. ‘PC’ indicates that the tested agent was used as a positive control. ‘CR’ indicates a significant concentration–response relationship. If ‘CR’ is not indicated, it is either because no such response was observed (negative result) or because it was not tested (e.g. when only two concentrations of the agent were evaluated). ‘Positive’ indicates that at least one concentration of the tested substance showed a statistically significant increased level of DNA damage compared to the negative control; and ‘negative’ indicates that no significant differences were found compared to the negative control. If no statistical analysis was provided, the results were recorded using symbols: ‘+’ to indicate an increase, ‘−’ a decrease, and ‘=’ no change in DNA damage relative to the negative control. ‘Increase’ indicates that the inclusion of DNA repair inhibitors led to a statistically significant increase in DNA damage at least at one concentration of the tested substance, compared to the assay without inhibitors; and ‘equal’ indicates that no significant differences were observed between the assay with and without DNA repair inhibitors. If no statistical analysis was reported, results were recorded using symbols: ‘↑’ to indicate an increase, and ‘↔’ for no apparent increase in DNA damage when DNA repair inhibitors were used.

Table 2.

Comet assay with HU and Ara-C.

Reference	Experimental system	Agent (treatment duration)	Inhibitor treatment	Results standard comet assay	Results with inhibitors	Cytotoxicity information
(A) Genotoxicity studies
[6]	TK6 XPG mutants TK6 HepaRG HepG2 Primary mouse hepatocytes HepaRG Comet Chip	UV (not clear) UV (not clear) B[a]P (24 h) AB1 (24 h) B[a]P (24 h) AB1 (24 h) B[a]P(24 h) AB1 (24 h) B[a]P (24 h) AB1 (24 h) Etoposide (24 h) 2,4-DAT (24 h) CP (24 h) PCA (24 h) NDMA (24 h) HQ (24 h) B[a]P (24 h) CAM (24 h) Cisplatin (24 h)	1 mM HU and 10 μM Ara-C 40 min before and 1 h after exposure 1 mM HU and 10 μM Ara-C 40 min before and 1 h after exposure 1 mM HU and 10 μM Ara-C for 24 h 1 mM HU and 10 μM Ara-C for 24 h 1 mM HU and 10 μM Ara-C for 24 h 1 mM HU and 10 μM Ara-C for 24 h 1 mM HU and 10 μM Ara-C for 24 h	Negative Negative Negative Negative Negative Negative Negative Negative Positive Positive Negative Negative Negative Negative Positive Negative Negative Negative Negative	Positive/↑ Positive/↑ Negative/↑ Negative/↑ Positive/↑ Positive/↑ Positive/↑ Negative/↑ Positive/↑ Positive/↑ Positive/↑ Positive/↑ Positive/↑ Negative/↔ Positive/↔ Positive/↑ Positive/↑ Positive/↑ Negative/↔	Trypan blue exclusion and CellTiter-Glo® assay (CTG®) High levels of cytotoxicity in HQ and CAM
[31]	PBMC unstimulated	PhIP (30 min) IQ (30 min) MeIQx (30 min) DiMeIQx (30 min)	10 mM HU and 1.8 mM Ara-C for 30 min	Positive (CR) Positive (CR) Positive (CR) Positive (CR)	Equal Equal Increase Increase	Trypan blue exclusion
[32]	TK6	MNU (2 h) ENU (2 h) EMS (2 h) MMS (2 h) BLM (2 h) UVC (not clear)	10 mM HU and 1.8 mM Ara-C for 2 h 10 mM HU and 1.8 mM Ara-C for 2 h after irradiation	Positive Positive Positive Positive Positive Positive	Positive/↑ Positive/↑ Positive/↑ Positive/↑ Positive/↔ Positive/↑	Trypan blue exclusion
[33]	MRC-5 TK6 Whole blood unstimulated	EMS (30 min) MMS (30 min) EMS (30 min) MMS (30 min) EMS (30 min) MMS (30 min)	10 mM HU and 1.8 mM Ara-C for 30 min	Positive Positive Positive Positive Negative Positive	Increase Increase Increase Increase Equal Increase	Fluorescein diacetate and ethidium bromide
[34]	MCF-10A MCF-10AT MCF-10ATG3B	H2O2 (5 min)	10 mM HU and 1.8 mM Ara-C for 30 min prior to H2O2 treatment	Positive (CR) Positive (CR) Positive (CR)	Increase Increase Increase	Trypan blue exclusion
[35]	Transfected BE(2)-M17 neuroblastoma cells	Fe(II) (2 h)	1 mM HU and 120 μM Ara-C for 2 h	Positive	Increase	Note: different sensitivities between cell lines
[36]	Primary keratinocytes	UVR (not clear)	100 μmol l−1 HU and 10 mmol l−1 Ara-C directly in the gels during 1 h after irradiation	Negative	Increase (CR)
[37]	Primary culture of cells from prostate tissues	PhIP (30 min) N-OH-PhIP (30 min) B[a]P (30 min)	10 mM HU and 1.8 mM Ara-C for 30 min	+ + +	Increase (CR) Increase (CR) Increase (CR)
[38]	V79 V79 genetically modified	2-AAF (24 h) PhIP (24 h) 2-AAF (24 h) PhIP (24 h)	10 mM HU and 1.8 mM Ara-C for 24 h	+ Negative Positive Negative	↔ Negative/↔ ↔ Positive/↑	Fluorescein diacetate and ethidium bromide
[39]	MCL-5	BPDE (1 h) MNNG (1 h) B[a]P (25 h)	10 mM HU and 1.8 mM Ara-C for 1 h 10 mM HU and 1.8 mM Ara-C the last hour of 25 h treatment	Positive Positive Positive	Increase Increase Increase	Cell proliferation
[40]	MKCK	OTA (up to 400 min)	10 nM HU and 1.8 mM Ara-C up to 400 min	+	Increase	Neutral red uptake assay
[41]	Exfoliated cells from breast milk	PhIP (30 min) Trp-P-1 (30 min) Trp-P-2 (30 min) B[a]P (30 min) 1-NP (30 min) o-toluidine (30 min) p-chloramide (30 min)	10 mM HU and 1.8 mM Ara-C for 30 min	Positive Positive Positive Negative Positive Positive Positive	Increase Increase Increase Increase Increase Increase Increase	Trypan blue exclusion
[42]	V79	MNNG (120 min)	2 × 10−3 mol/l HU and 2 × 10−5 mol/l Ara-C for 120 min	+	Equal
[43]	MCL-5	MeIQx (30 min) Di-MeIQx (30 min) IQ (30 min) A[α]C (30 min) MeA[α]C (30 min) PhIP (30 min) B[a]P (30 min) 3-MCA (30 min) DMBA (30 min) 1-NP (30 min) 2-NF (30 min) Aniline (30 min) o-toluidine (30 min) benzene (30 min) lindane (30 min) DES (30 min) Bleomycin (30 min) Cisplatin (30 min) MNNG (30 min) Sodium -chromate (30 min) Chromic chloride (30 min)	10 mM HU and 1.8 mM Ara-C for 30 min	Negative Negative Positive Negative Positive Positive Positive Negative Positive Negative Positive Positive Positive Positive Positive Negative Positive Positive Positive Positive Positive	Negative/Equal Positive/Increase Equal Positive/Increase Increase Increase Increase Positive/Increase Equal Negative/Equal Equal Increase Increase Increase Equal Negative/Equal Increase Equal Increase Increase Increase	Trypan blue exclusion
[44]	VH10 Hep G2 V79	MNNG (120 min)	2 mM HU and 20 μM Ara-C for 120 min	+ + +	↔ ↔ ↔
[45]	HL-60 cells	MX (1 h)	2 mM HU and 0.1 mM Ara-C for 1 h	Negative	Positive	Trypan blue exclusion
(B) DNA repair studies
[46]	Earthworms testis cells	UVC (not clear) H2O2 (30 min)	5 mM HU and 50 μM Ara-C added 30 min before and 6 h after UVC exposure 5 mM HU and 50 μM Ara-C added 30 min before +30 min co-exposure+6 h inhibitors	Negative Negative	Increase Increase
[30]	NC37 B-lymphoblast	UVA (from 1 to 30)	2 mM HU and 5 μM Ara-C for 80 min after irradiation	-	↑

Agent abbreviations are described in Table 5. ‘CR’ indicates a significant concentration–response relationship. If ‘CR’ is not indicated, it is either because no such response was observed (negative result) or because it was not tested (e.g. when only two concentrations of the agent were evaluated). ‘Positive’ indicates that at least one concentration of the tested substance showed a statistically significant increased level of DNA damage compared to the negative control; and ‘negative’ indicates that no significant differences were found compared to the negative control. If no statistical analysis was provided, the results were recorded using symbols: ‘+’ to indicate an increase, ‘−’ a decrease, and ‘=’ no change in DNA damage relative to the negative control. ‘Increase’ indicates that the inclusion of DNA repair inhibitors led to a statistically significant increase in DNA damage at least at one concentration of the tested substance, compared to the assay without inhibitors; and ‘equal’ indicates that no significant differences were observed between the assay with and without DNA repair inhibitors. If no statistical analysis was reported, results were recorded using symbols: ‘↑’ to indicate an increase, and ‘↔’ for no apparent increase in DNA damage when DNA repair inhibitors were used.

Table 3.

Comet assay with Ara-C.

Reference	Experimental system	Agent (treatment duration)	Inhibitor treatment	Results standard comet assay	Results with inhibitors	Cytotoxicity information
(A) Genotoxicity studies
[47]	AT21RM AT28RM FLEBV (wild type ATM genotype)	X-rays (not clear)	50 μM Ara-C for 15 min before irradiation and 3 h after irradiation	Positive Positive Positive	Increase Increase Increase
[48]	Peripheral lymphocytes PHA-stimulated	HQ (90 min after stimulation)	10 μg/ml Ara-C from 32 h after stimulation to the treatment time 46 h	Positive	Increase	Trypan blue exclusion
(B) DNA repair studies
[49]	Unstimulated lymphocytes	UVC (not clear) BCNU (30 min)	10 μM Ara-C for 2 h before UVC treatment +1 h incubation without any treatment (repair) 10 μM Ara-C for 2 h before UVC treatment +4 h incubation without any treatment (repair) 10 μM Ara-C for 2 h before BCNU treatment +4 incubation without treatment	+ — +	↔ ↑ ↑	Apoptotic cell death after 24 h of treatment: cytotoxicity from 20 μM Ara-C
[50]	Human lymphocytes PHA-stimulated	MMS (2 h) H2O2 (30 min on ice)	1 μg/ml Ara-C during the first 16 h of incubation after treatment	0 h: Positive 16 h: Positive 48 h: Negative 0 h: Positive 16 h: Negative 48 h: Negative	16 h: Positive 48 h: Negative 16 h: Positive 48 h: Negative	Fluorochrome-mediated viability test Note: different incubation hours for stimulation
[30]	NC37 B-lymphoblast	UVA (from 1 to 30)	5 μM Ara-C for 80 min after irradiation	-	↑

Agent abbreviations are described in Table 5. ‘Positive’ indicates that at least one concentration of the tested substance showed a statistically significant increased level of DNA damage compared to the negative control; and ‘negative’ indicates that no significant differences were found compared to the negative control. If no statistical analysis was provided, the results were recorded using symbols: ‘+’ to indicate an increase, ‘−’ a decrease, and ‘=’ no change in DNA damage relative to the negative control. ‘Increase’ indicates that the inclusion of DNA repair inhibitors led to a statistically significant increase in DNA damage at least at one concentration of the tested substance, compared to the assay without inhibitors; and ‘equal’ indicates that no significant differences were observed between the assay with and without DNA repair inhibitors. If no statistical analysis was reported, results were recorded using symbols: ‘↑’ to indicate an increase, and ‘↔’ for no apparent increase in DNA damage when DNA repair inhibitors were used.

Table 4.

Comet assay with HU.

Reference	Experimental system	Agent (treatment duration)	Inhibitor treatment	Results standard comet assay	Results with inhibitors	Cytotoxicity information
(A) Genotoxicity studies
[51]	MCL-5	3-NBA (2 h)	5 mM HU for 2 h	Positive (CR)	Increase	Trypan blue exclusion
(B) DNA repair studies
[52]	Human lymphocytes unstimulated	OTA (3 h)	10 mM N-HU for 60 min after 3 h treatment	Positive	Increase	Fluorescein diacetate and ethidium bromide
[30]	NC37 B-lymphoblast	UVA (from 1 to 30)	2 mM HU for 80 min after irradiation	-	↑

Agent abbreviations are described in Table 5. ‘CR’ indicates a significant concentration–response relationship. If ‘CR’ is not indicated, it is either because no such response was observed (negative result) or because it was not tested (e.g. when only two concentrations of the agent were evaluated). ‘Positive’ indicates that at least one concentration of the tested substance showed a statistically significant increased level of DNA damage compared to the negative control; and ‘negative’ indicates that no significant differences were found compared to the negative control. If no statistical analysis was provided, the results were recorded using symbols: ‘+’ to indicate an increase, ‘−’ a decrease, and ‘=’ no change in DNA damage relative to the negative control. ‘Increase’ indicates that the inclusion of DNA repair inhibitors led to a statistically significant increase in DNA damage at least at one concentration of the tested substance, compared to the assay without inhibitors; and ‘equal’ indicates that no significant differences were observed between the assay with and without DNA repair inhibitors. If no statistical analysis was reported, results were recorded using symbols: ‘↑’ to indicate an increase, and ‘↔’ for no apparent increase in DNA damage when DNA repair inhibitors were used.

Each of these tables was further divided into two sections: A, for studies assessing genotoxicity, and B, for studies focused on DNA repair, for a clearer distinction between study objectives.

The tables included the following information:

• Reference: bibliographic reference of the article. Within each table, publications have been sorted by date of publication, with the most recent appearing first.

• Experimental system: the specific experimental system used is specified (cell line name, blood cells, or 3D model).

• Agent and treatment duration: the specific name of the evaluated physical or chemical agent is given. The treatment duration is indicated in brackets next to the agent.

• Inhibitor treatment: the concentration and time of the inhibitor exposure are given. The units are kept as in the original articles. If not specified, the inhibitor treatment was carried out at the same time as the agent treatment (co-exposure) at 37°C.

• Results standard comet assay: negative, positive, and no statistical results are reported. ‘Positive’ indicates that at least one concentration of the tested substance showed a statistically significant increased level of DNA damage compared to the negative control, and ‘negative’ indicates that no significant differences were found compared to the negative control. If no statistical analysis was provided, the results were recorded using symbols: ‘+’ to indicate an increase, ‘−’ a decrease, and ‘=’ no change in DNA damage relative to the negative control, based on the authors’ interpretation or, when not explicitly stated, after visually inspecting the data. In addition, ‘CR’ will be included when a significant concentration-response was observed, according to the authors.

• Results with inhibitors: results were interpreted in the same way as for the comet assay without inhibitors, by comparing DNA damage levels to the corresponding reference control, when statistics were performed. In addition, outcomes were classified as either ‘increase’ or ‘equal’ when comparing the modified assay (with inhibitors) to the standard version (without inhibitors). ‘Increase’ indicates that the inclusion of DNA repair inhibitors led to a statistically significant increase in DNA damage at least at one concentration of the tested substance, compared to the assay without inhibitors; and ‘equal’ indicates that no significant differences were observed between the assay with and without DNA repair inhibitors. If no statistical analysis was reported, results were recorded using symbols: ‘↑’ to indicate an increase, and ‘↔’ for no apparent increase in DNA damage when DNA repair inhibitors were used.

• Cytotoxicity information: since cytotoxicity itself can induce DNA damage [53, 54], it is important to consider its potential confounding effects when interpreting genotoxicity results. Therefore, it has been noted whether a cytotoxicity assay was performed in each study. Cytotoxicity results are reported only when measurable levels of cell damage or viability loss were observed.

Another table (Table 5) was prepared to summarize the main mechanisms of action of all the exposure agents analyzed, along with the number of studies in which each agent was evaluated and the number of studies reporting an increased sensitivity when DNA repair inhibitors were used. It is important to note that ‘articles’ and ‘studies’ are not synonymous; in this context, ‘studies’ refers to individual experiments reported within a single article that differ in key experimental conditions (e.g. cell line, exposure time, etc.).

Table 5.

Overview of genotoxicity, mechanisms, and study findings for exposure agents (ordered by mechanism of action).

Genotoxicity	Name	Abbreviation	Main mechanism of action	No. studies evaluating the exposure agent	No. of studies with increased sensitivity in the presence of DNA repair inhibitors (inhibitor used)	No. of studies without increased sensitivity in the presence of DNA repair inhibitors (inhibitor used)	No. of studies in which the compound tested negative with the standard assay and positive with inhibitors
Genotoxic agents	3-chloro-4-dichloromethyl-5-hydroxy-2 5H-furanone	MX	Bacterial mutagen	1	1 (HU and Ara-C)	-	1
	Etoposide		SBs	1	1 (HU and Ara-C)	-	1
	Bleomycin	BLM	SBs	1	-	1NS (HU and Ara-C)	-
	-	Fe(II)	Single SBs	1	1 (HU and Ara-C)	-	-
	Gamma radiation		DNA SBs + base damage	1	-	1 (APC)	-
	Hydrogen peroxide	H2O2	SB + oxidative stress	4	1NS (APC) 2 (HU and Ara-C) 1 (Ara-C)	-— -	1
	1-nitropyrene	1-NP	DNA adducts	2	1 (HU and Ara-C)	1 (HU and Ara-C)	-
	2-acetylaminofluorene	2-AAF	DNA adducts	3	1 (APC)	2NS (HU and Ara-C)	1
	2-nitrofluorene	2-NF	DNA adducts	1	-	1 (HU and Ara-C)	-
	3-amino-1,4-dimethyl-5H-pyrido[4,3-b]indole	Trp-P-1	DNA adducts	1	1 (HU and Ara-C)	-	-
	3-amino-1-methyl-5H-pyrido[4,3-b]indole	Trp-P-2	DNA adducts	1	1 (HU and Ara-C)	-	-
	3-nitrobenzanthrone	3-NBA	DNA adducts	1	1 (HU)	-	-
	4-nitroquinoline-1-oxide	4-NQO	DNA adducts	2	1 (APC)	1 (APC)*	-
	7,12-dimethylvenz(a)anthracene	DMBA	DNA adducts	4	2 (APC)	1 (APC)* 1 (HU and Ara-C)	-
	AFB1	AFB1	DNA adducts	4	4NS (HU and Ara-C)	-	1
	Aniline		DNA adducts	1	1 (HU and Ara-C)	-	-
	Benzo[a]pyrene	B[a]P	DNA adducts	13	3 (APC) 8 + 1NS (HU and Ara-C)	1 (APC)* -	5
	Benzo[a]pyrene-7,8-dihydrodiol-9,10-epoxide	BPDE	DNA adducts	21	13 + 2NS (APC) 1 (HU and Ara-C)	5 (APC) -	-
	N-hydroxy- 2-aminofluorene	N-OH-2-AF	DNA adducts	1	1NS (APC)	-	-
	N-hydroxy-2-acetylaminofluorene	N-OH-2-AAF	DNA adducts	1	1NS (APC)	-	-
	2,4-diamino-6-(2,4-diclorofenil)-1,3,5-triazina	2,4-DAT	DNA adducts + oxidative stress	1	1 (HU and Ara-C)	-	1
	2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine	PhIP	DNA adducts + oxidative stress	6	5 (HU and Ara-C)	1 (HU and Ara-C)	1
	2-amino-3,4,8-trimethyl-3H-imidazo[4,5-f]quinoxaline	DiMeIQx	DNA adducts + oxidative stress	2	2 (HU and Ara-C)	-	1
	2-amino-3,8-dimethyl-imidazo[4,5-f]quinoxaline	MeIQx	DNA adducts + oxidative stress	2	1 (HU and Ara-C)	1 (HU and Ara-C)	-
	2-amino-3-methyl-3H-imidazo[4,5-f]quinoline	IQ	DNA adducts + oxidative stress	2	-	2 (HU and Ara-C)	-
	2-amino-3-methyl-9H-pyrido[2,3-b]indole	MeA[α]C	DNA adducts + oxidative stress	1	1 (HU and Ara-C)	-	-
	2-amino-9H-pyrido[2,3-b]índole	A[α]C	DNA adducts + oxidative stress	1	1 (HU and Ara-C)	-	1
	3-methylcholanthrene	3-MCA	DNA adducts + oxidative stress	1	1 (HU and Ara-C)	-	1
	p-chloramide	-	DNA adducts + oxidative stress	1	1 (HU and Ara-C)	-	-
	o-toluidine	-	DNA adducts + oxidative stress	2	2 (HU and Ara-C)	-	-
	Chromic chloride		Oxidative stress	1	1 (HU and Ara-C)	-	-
	Sodium chromate		Oxidative stress	1	1 (HU and Ara-C)	-	-
	Mitomycin C	MMC	Cross link	1	1 (APC)	-	1
	Cisplatin	CisPt	Cross links	2	-	2 (HU and Ara-C)	-
	Ethyl methane sulfonate	EMS	Alkylation	4	3 (HU and Ara-C)	1 (HU and Ara-C)	-
	Methyl methane sulfonate	MMS	Alkylation	16	4 + 1NS (APC) 3 + 1NS (HU and Ara-C) -	6 (APC)— 1(Ara-C)	-
	Methyl nitrosourea	MNU	Alkylation	1	1NS (HU and Ara-C)	-	-
	N-ethyl-N-nitrosurea	ENU	Alkylation	1	1NS (HU and Ara-C)	-	-
	N-methyl-N′-nitro-N-nitrosoguanidine	MNNG	Alkylation (mainly ALS)	4	2 (HU and Ara-C)	2 (HU and Ara-C)	-
	Bischloroethylnitrosurea	BCNU	Alkylation and cross links	3	2 (APC) 1NS (Ara-C)	- -	-
	N-nitrosodimethylamine	NDMA	Methylation	1	-	1NS (HU and Ara-C)	-
	UVA, UVC, UVR	UV	CPDs	15	5 + 1NS (APC) 3 + 1NS (HU and Ara-C) 1NS (Ara-C) -	1 + 1*(APC)— 1NS (Ara-C) 1NS (HU)	2
	X-rays		Chromosome aberrations	3	3 (Ara-C)	-	-
Non-genotoxic	Diethylstilboestro	DES		1	-	1 (HU and Ara-C)	-
	Di-(2-ethylhexyl)phthalate	DEHP		1	-	1NS (APC)	-
	Eugenol		MN formation due to cytotoxicity	1	-	1NS (APC)	-
	Propyl gallate	-	Micronucleus formation due to cytotoxicity	1	-	1 (APC)	-
Unclear	Cadmium chloride		Unclear	1	-	1NS (APC)	-
	Chloramphenicol	CAM	Unclear	1	1 (HU and Ara-C)	-	1
	Cyclohexanone		Unclear	1	1NS (APC)	-	-
	Cyclophosphamide	CP	Unclear	1	1 (HU and Ara-C)	-	1
	Hydroquinone	HQ	Unclear	2	1 (HU and Ara-C) 1(Ara-C)	- -	1
	Lindane		Unclear	1	-	1 (HU and Ara-C)	-
	Ochratoxin A	OTA	Unclear	2	1 (HU and Ara-C) 1 (HU)	- -	-
	Radio frequency	RF	Unclear	1	1 (APC)	-	-

^*XP deficient cell line. ^NSis stated when no statistical analysis was reported.

Results

A total of 105 publications were retrieved from the PubMed search (Fig. 1). After Phase 1 of the selection process, 83 articles were selected for full-text examination. In Phase 2, after full text examination, 43 articles were excluded according to the inclusion and exclusion criteria. Consequently, 40 articles were ultimately included for data extraction.

The selected articles were organized into four separate tables based on the DNA repair inhibitor used in combination with the comet assay. Among the inhibitors, APC and the combination of HU with Ara-C were the most frequently employed. APC was used in 17 studies (Table 1), while the HU and Ara-C combination was applied in 18 studies (Table 2). These two DNA repair inhibitors, HU and Ara-C, were also used alone, but to a much lesser extent: Ara-C in 5 articles (Table 3) and HU only in 3 (Table 4). The work by Bock et al. [30] was included in all tables as it is the only article that used more than one DNA repair inhibitor.

At the same time, each table was subdivided into sections A and B, distinguishing between studies focused on genotoxicity (A) and those assessing DNA repair (B). Among the studies using APC, 12 investigated genotoxicity and 5 evaluated DNA repair (Table 1). For the combination of HU and Ara-C, 16 studies focused on genotoxicity and 2 on DNA repair (Table 2). In the case of Ara-C alone, 2 studies assessed genotoxicity and 3 DNA repair (Table 3), while with HU alone, 1 study addressed genotoxicity and 2 evaluated DNA repair (Table 4).

Concerning DNA repair inhibitor exposure, co-exposure with the tested agent was the most commonly used condition. However, the addition of the inhibitor before or after the treatment, sometimes together with co-exposures, was also reported in several of the reviewed publications. With respect to the exposure duration, different time frames were reported, including co-exposures ranging from 30 min to 24 h; addition of the inhibitor 4 h before the end of the treatment period with the tested agent; addition only 30 min before the treatment; or addition 30 min prior to a co-exposure with the test compound, among others. As for the concentrations of DNA repair inhibitors, a relatively narrow range was identified for APC (1–15 μM for APC) and HU (1–10 mM). In contrast, Ara-C was tested over a broader range (4.1 µM–1.8 mM).

Regarding the agents studied across the publications, various types were included, such as alkylating agents, oxidizing agents, and also bulky DNA adduct inducers, among others (Table 5).

In all the publications reviewed, for at least one agent evaluated, the use of DNA repair inhibitors revealed DNA lesions that were not detected with the standard comet assay, except for the work of Slameňová et al. [44] and Horváthová et al. [42], in which no increased DNA damage was observed with the DNA repair inhibitors HU and Ara-C, and the work of Bausinger et al. [17], in which no increased DNA damage was observed with APC (Tables 1–4). Among these 37 articles, 7 reported a change from ‘negative’ in the standard comet assay to ‘positive’ when combined with DNA repair inhibitors for at least one agent; 24 showed statistically significant differences between the two conditions in at least one agent (i.e. an ‘increase’); and in the remaining 6 articles, although no statistical analyses were reported, qualitative results indicated an increase in DNA damage (i.e. ↑) for at least one compound studied. In addition, cytotoxicity assays were conducted in a total of 22 articles. Regarding papers dealing with genotoxicity, in 19 out of 28, cytotoxicity was evaluated, most of them using common cytotoxicity assays (e.g. trypan blue assay, ethidium bromide staining, ATP …). Just one of the papers used the proliferation assay [21].

Among the 55 agents studied in the articles reviewed (Table 5), 43 were genotoxic, 4 were non-genotoxic, and the remaining 8 were unclear. Concerning the genotoxic compounds, all showed increased DNA damage levels when using DNA repair inhibitors with the comet assay, in comparison with the standard assay, in at least one study, except for IQ [6] [43], 2-NF [43], CisPt [6, 43], NDMA [6], and gamma radiation [22], for which no increase was observed in any of the studies. Among these exceptions, all tested positive in at least one study, except for the work of Ngo et al. [6], in which CisPt showed negative results in both comet assays (i.e. standard and with DNA repair inhibitors). Furthermore, the combination of HU and Ara-C was the most commonly used DNA repair inhibitor treatment on these exceptions, except for gamma radiation, which was tested with APC. Regarding the non-genotoxic compounds, as expected, negative results were observed both with and without DNA repair inhibitors. Finally, among the compounds classified as unclear, only lindane [43] did not show increased DNA damage in the comet assay combined with HU and Ara-C (although positive results were observed with the standard comet assay for this compound).

Discussion

The standard comet assay simply detects SBs and ALS. However, these are quite unspecific lesions as they can be directly formed by genotoxic compounds, but they can also be formed during the repair of several lesions such as oxidized and alkylated bases or DNA adducts. Using different approaches SBs intermediates can be generated or trapped and detected with the comet assay. For example, the use of lesion-specific enzymes has been widely used to increase the sensitivity of the comet assay allowing the detection of alkylated or oxidized bases [5, 55–57]. In this work we have been focused on another approach, the use of DNA repair inhibitors as a possibility to increase the sensitivity and specificity of the comet assay towards some DNA lesions and, as far as we are aware, this is the first review to address this approach.

As mentioned in the introduction section, the comet assay combined with HU and Ara-C has been included in the AOP ‘Bulky adducts leading to mutation’ for the detection of bulky DNA adducts (KE 1879) [7]. These pre-mutagenic lesions are mainly repaired by the nucleotide excision repair (NER) pathway, which involves the removal of an oligomer containing the damage; therefore, during this process SBs intermediates are generated. However, NER intermediates form gradually over time, so their presence at any given moment may be limited. Moreover, these breaks are short-lived, as they are rapidly and efficiently repaired through DNA resynthesis and ligation. By inhibiting DNA resynthesis using HU, Ara-C, or APC, SBs can be trapped and detected with the comet assay.

In this work, a total of 40 articles were reviewed in which both the standard and modified comet assays, using HU, Ara-C, or APC as DNA repair inhibitors, were used in vitro to evaluate different agents. DNA damage results were extracted from studies focused on genotoxicity testing and DNA repair assessment. Overall, the results indicate that generally the use of these DNA repair inhibitors (i.e. HU and Ara-C or APC) increases the DNA damage detected and thus the sensitivity of the assay. However, this has not been only observed for bulky adduct inducers but also for alkylating agents such as MMS [22, 33], EMS [36], MNGN [42], or BCNU [22] and oxidizing agents such as H₂O₂ [34], among others.

One of the most common agents studied is B[a]P or benzo[a]pyrene-7,8-dihydrodiol-9,10-epoxide (BPDE), its main active metabolite, which it is known to induce bulky DNA adducts. Our review showed a variety of results when applying the comet assay in combination with DNA repair inhibitors in this case. Some studies could not detect BPDE/B[a]P with the standard comet assay, but positive results were obtained when combining it with APC [14, 15] or HU/Ara-C [6, 41]. However, other works could detect this agent with the standard assay, although when DNA inhibitors were used the sensitivity significantly increased [19–21, 43]. Some other studies, such as Bausinger et al. [17] did not find any differences between the use or not of repair inhibitors. Another commonly studied agent is UV radiation, which mainly formed cyclobutene pyrimidine dimers (CPDs) and it is also mainly repaired by NER. Exactly as in the previous case, a wide range of results were found when analyzing UV. Also, some other bulky adduct inducers were studied and effectively detected with this approach, such as 2-acetylaminofluorene (2-AAF) [14], AFB₁ [6], 4-nitroquinoline-1-oxide (4-NQO) [25], or 2-amino-3,8-dimethyl-imidazo[4,5-f]quinoxaline (MeIQx) [31]. Speit and Hartman [25] demonstrated that NER is involved in the repair of DNA adducts induced by 4-NQO and B[a]P, using an XP-deficient cell line. Moreover, Ngo et al. [6] demonstrated by using XPG-deficient cells treated with UV in the presence or absence of HU/Ara-C and XPA-deficient cells treated with B[a]P and AFB₁ with or without HU/Ara-C that almost all of the SBs detected using these inhibitors are NER intermediates. Thus, NER intermediates contribute to most of the SBs induced by B[a]P and AFB₁.

The wide variety of results found along the review may be explained by the different experimental factors, including the experimental system and design. Among the various experimental conditions, co-exposure to 15 μM APC for 2 h or to 10 mM HU and 1.8 mM Ara-C for 30 min to 2 h were the most commonly used protocols. Although pre-treatment with DNA repair inhibitors was occasionally employed, Speit et al. [16] reported no differences between the addition of 15 μM APC 30 min before agent exposure and simultaneous addition.

Inhibitor concentration and treatment duration, as the SBs intermediates are gradually formed, would be determinant. Odongo et al. [26] observed a stronger APC-enhanced DNA effect in the comet assay when using 6 μM compared to 3 μM, and Speit et al. [22] also reported a greater effect with 5 μg/ml of APC than with 0.5 μg/ml. However, it should be taken into account that these enhanced effects at higher concentrations may eventually be associated with cytotoxicity (presence of hedgehogs) [16]. Regarding treatment duration, no studies have specifically addressed its influence.

Different cell lines have been used in the studies reviewed, which may also account for variability in the results. In fact, Speit et al. [16] demonstrated different sensitivities along different cell lines detecting BPDE and MMS. In addition, it is known that the resynthesis step of the NER pathway is partially dependent on the growth state of the cell, which may also explain the different sensitivity between different cells [8]. Among the most employed cells are PHA-stimulated blood cells. It would be expected that stimulated blood cells show higher DNA damage since they are replicating; however, several studies have reported no statistically significant differences in DNA damage between stimulated and unstimulated cells [20, 22, 48]. For example, Speit et al. [20] compares DNA damage between stimulated and unstimulated blood cultures exposed to BPDE. In the higher BPDE concentration (2.5 μM) tested, higher DNA damage was found in unstimulated blood than in stimulated blood. However, none of the differences were statistically significant. Another important factor is if cells are metabolically competent or not when promutagens are used, and in this case, time exposure will also be a determinant. For instance, Ngo et al. [6] tested two pro-mutagens, B[a]P and AFB₁, in different cell lines after a 24 h exposure. They obtained negative results in the comet assay with HU and Ara-C in TK6 cells, which are metabolically incompetent, and positive results in HepaRG cells, which are metabolically competent.

As mentioned earlier, the use of HU, Ara-C, or APC increases the DNA damage detected by the comet assay; however, this effect is not specific to bulky adduct inducers, as it is also observed with alkylating and oxidizing agents. It is worth mentioning that NER not only repairs bulky DNA adducts caused by environmental mutagens or UV-photoproducts but also certain oxidative endogenous lesions such as cyclopurines and oxidative adducts formed by chemotherapeutic drugs [58]. These NER substrates have in common that they destabilize the DNA helix and that they are bulky. This may also explain the fact that other agents with different mechanisms of action are also repaired by NER and therefore detected with this approach. In addition, agents can activate more than one pathway where other DNA polymerases are involved.

Moreover, the DNA repair inhibitors used may not be selective for the NER pathway. DNA polymerase β is mainly involved in base excision repair (BER), but also polymerases δ or ε take part in BER processes to a lesser extent. Similarly, mismatch repair of replication errors appears to involve DNA polymerases δ or ε. On the other hand, polymerase α is required for replication of DNA but not for repairing processes [9]. As already mentioned in the introduction section, APC inhibits B-family DNA polymerases, including Pol α, Pol δ, Pol ε, and Pol ζ [11]; Ara-C, apart from being a chain terminator preventing the elongation by DNA polymerases, is a potent inhibitor of Pol α and a weak inhibitor of Pol β [9]; and HU inhibits ribonucleotide reductase and increases the effectiveness of Ara-C [12]. This means that the use of these DNA repair inhibitors will also affect these pathways and even the DNA semiconservative replication process, as APC inhibits DNA polymerase α. Moreover, the resynthesis step of the NER pathway is partially dependent on the growth state of the cell, which may also explain the different sensitivity between different cell lines [8]. DNA polymerases δ and κ and ligase III are mainly involved in non-proliferating cells, while in cycle cells pol ε and ligase I are used. For an efficient process, proliferating cell nuclear antigen (PCNA) and the replication factor C are needed [8].

Choosing which DNA repair inhibitor is better is a challenging task, since none of them is selective for a specific DNA repair pathway. The three DNA repair inhibitors studied in this review increase the sensitivity of the comet assay; however, it is difficult to determine which one (including the combination of Ara-C and HU) gives a higher response due to different chemical and different conditions used, including the variety of cell lines. Moreover, there is a need to evaluate a broader range of compounds, particularly non-genotoxic and cytotoxic agents, in order to assess the specificity of these approaches.

Although HU and Ara-C each can enhance comet assay sensitivity when used separately, they are more effective together, as their mechanisms of action are complementary. Interestingly, approximately the same number of studies did not observe an increase in DNA damage with HU and Ara-C (17 studies) or with APC (21 studies); however, this should be carefully analyzed, considering the specific factors of each study and the chemical being analyzed.

Cytotoxicity was only performed in 19 of 22 studies focused on genotoxicity and in all studies non-toxic concentrations were used. It is known that the use of cytotoxic concentrations can lead to false-positive results in the comet assay [54]. To address this issue and minimize its impact, it is recommended that publications reporting in vitro comet assay results include data on at least cytotoxicity, cell death, or cell proliferation [54]. Ideally, both cytotoxicity or cell death assays and proliferation or colony-forming efficiency assays should be included, particularly in actively proliferating cells. However, when DNA repair inhibitors are used in proliferating cells, proliferation or colony-forming efficiency assays may not be applicable, as cells exposed to such inhibitors might affect proliferation. Nevertheless, a cytotoxicity assay can still be performed after co-treatment; however, further research is needed to establish a cytotoxicity threshold that can be reliably used.

The use of these inhibitors has been used for both, genotoxicity testing and DNA repair evaluation. Considering this review, this approach can be used for genotoxicity testing since it enhances the sensitivity of the comet assay, allowing the classification of compounds that yield negative results in the standard version as genotoxic. Although, as mentioned before, the evaluation of a larger number of non-genotoxic and cytotoxic agents is needed. Regarding DNA repair evaluation, it can serve as an indicator of repair capacity, but the results should be interpreted with caution, as it primarily measures the cells’ ability to carry out the incision and/or excision step/s of repair. Although this step is considered the rate-limiting stage of the process, potential issues in the subsequent repair steps, such as DNA resynthesis and ligation, are not accounted for.

Although HU, Ara-C, and APC have been the focus of this work, as they are the most commonly DNA repair inhibitors used in combination with the comet assay, other inhibitors have also been applied. To inhibit BER, novobiocin, a topoisomerase II inhibitor [22, 28, 29] or methoxyamine, which reacts with abasic sites, protecting them from enzymatic incision [30], have been used. Fludarabine, which acts similarly to Ara-C, terminating DNA elongation after being incorporated into DNA [59] has been applied for the inhibition of NER. In all cases more studies are needed to study their specificity towards different DNA lesions.

To conclude, the value of the use of the DNA repair inhibitors HU, Ara-C and APC with the comet assay in terms of additional information is demonstrated by the fact that, in 70% of the studies reviewed, a significant increase in DNA damage was seen in the presence of DNA repair inhibitors. However, the use of this approach as a method for detecting bulky DNA adducts should be carefully studied as it has been seen that not only detects bulky adducts inducer compounds but also alkylating or oxidizing agents.

Conflict of interest statement

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Funding

This work was supported by the Spanish Ministry of Science, Innovation and Universities (PID2020-115348RB-I00, FPU21/03187).