Oxidatively generated complex DNA damage: Tandem and clustered lesions

Cancer Letters 327 (2012) 5–15

Contents lists available at SciVerse ScienceDirect

Cancer Letters journal homepage: www.elsevier.com/locate/canlet

Mini-review

Oxidatively generated complex DNA damage: Tandem and clustered lesions Jean Cadet a,b,⇑, Jean-Luc Ravanat a, Marisa TavernaPorro a, Hervé Menoni c,d, Dimitar Angelov c a

Laboratoire ‘‘Lésions des Acides Nucléiques’’, SCIB-UMR-E no 3 (CEA/UJF), Institut Nanosciences et Cryogénie, CEA/Grenoble, F-38054 Grenoble Cedex 9, France Département de Médecine Nucléaire et Radiobiologie, Faculté de médecine de des sciences de la santé, Université de Sherbrooke, Sherbrooke, Québec, Canada J1H 5N4 c Laboratoire de Biologie Moléculaire de la Cellule, UMR-CNRS 5239, Ecole Normale Supérieure de Lyon, 69364 Lyon Cedex 07, France d Department of Genetics, Erasmus MC, Rotterdam, The Netherlands b

a r t i c l e

i n f o

Keywords: Oxidative stress Ionizing radiation DNA single and double strand breaks Purine 50 ,8-cyclonucleosides Intrastrand cross-links Interstrand cross-links Tandem base lesions DNA repair Mutagenesis

a b s t r a c t There is an increasing interest for oxidatively generated complex lesions that are potentially more detrimental than single oxidized nucleobases. In this survey, the recently available information on the formation and processing of several classes of complex DNA damage formed upon one radical hit including mostly hydroxyl radical and one-electron oxidants is critically reviewed. The modifications include tandem base lesions, DNA-protein cross-links and intrastrand (purine 50 ,8-cyclonucleosides, adjacent base cross-links) and interstrand cross-links. Information is also provided on clustered lesions produced essentially by exposure of cells to ionizing radiation and high energetic heavy ions through the involvement of multiple radical events that induce several lesions DNA in a close spatial vicinity. These consist mainly of double strand breaks (DSBs) and non-DSB clustered lesions that are referred as to oxidatively generated clustered DNA lesions (OCDLs). Ó 2012 Elsevier Ireland Ltd. All rights reserved.

1. Introduction Oxidative stress results from the imbalance between endogenous generation of reactive oxygen species (ROS) and anti-oxidant defence systems [1] that involve scavenging of low reactive ROS such as superoxide radical (O 2 ) and hydrogen peroxide (H2O2), the precursors of highly damaging hydroxyl radical (OH) [2]. These oxygen species may be triggered by inflammation reactions [3,4] that also lead to the exalted production of nitric oxide [5,6], one of the main nitrogen reactive species (NOS). Oxidative stress have been shown to be associated with physical exercise [7], metal toxicities [8], aging [9,10] and several pathologies including mellitus diabetis [11], cardiovascular diseases [12,13], neurological disorders [14,15] and cancers [16–20]. Exacerbated generation of ROS and other oxidizing processes such as one-electron oxidation of biomolecules are also provided by ionizing radiation [21,22] and the UVA component of solar light [23,24], two well established physical carcinogens. DNA is the main critical cellular target to oxidatively generated damage that may participate in the initiation and/or propagation processes leading to carcinogenesis through mutagenesis [17– 20,25]. An abundant literature is now available on the mechanisms of oxidative degradation of nucleobases [26–29] and 2-

⇑ Corresponding author at: Laboratoire ‘‘Lésions des Acides Nucléiques’’, SCIB-UMR-E no 3 (CEA/UJF), Institut Nanosciences et Cryogénie, CEA/Grenoble, F38054 Grenoble Cedex 9, France. Tel.: +33 4 38 78 49 87; fax: +33 4 38 78 50 90. E-mail address: [email protected] (J. Cadet). 0304-3835/$ - see front matter Ó 2012 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.canlet.2012.04.005

deoxyribose [30] in isolated DNA and model compounds that are mediated by OH, singlet oxygen (1O2) and one-electron oxidants, the three main identified reactive oxidizing species and agents. The measurement of oxidized bases in cellular DNA has been hampered until the end of the 1990s by the use of inappropriate methods that has led to overestimation of the levels of oxidized bases up to three orders of magnitude [31]. Accurate data are now available on the formation of several oxidatively generated base lesions including ubiquitous 8-oxo-7,8-dihydroguanine (8-oxoGua) and thirteen single oxidized purine and pyrimidine bases in cellular DNA. This may be achieved for electrochemically active lesions including 8-oxoGua, 8-oxo-7,8dihydroadenine (8-oxoAde) and 5-hydroxycytosine using highperformance liquid chromatography coupled to electrochemical detection (HPLC-ECD) as the analytical method. However, the method of choice appears to be HPLC associated with electrospray ionization tandem mass spectrometry (ESI-MS/MS) as a versatile and accurate method [31]. Information on the mutagenic features of several single base lesions has been gained from shuttle vector experiments [32,33] and polymerase-mediated incorporation into DNA of oxidized precursors present in the nucleotide pools [34]. Major efforts have been also devoted to the determination of substrate specificity and removal mechanisms of repair enzymes that mostly operate for single lesions through the base excision repair pathways [35,36]. The radiation-induced formation of locally multiply damaged sites, also referred as to oxidatively generated clustered lesions (ODCLs) [37,38] was suggested more than 25 years ago in addition to

6

J. Cadet et al. / Cancer Letters 327 (2012) 5–15

previously identified double strand breaks (DSBs) [39,40]. The formation of these complex lesions that may include DSBs, single strand breaks (SSBs), oxidized bases and abasic sites, is accounted for by the occurrence of at least two radical hits within one or two helix turns [41,42]. Following the pioneering contributions of Box and collaborators in the 1990s [43] it was found that tandem base modifications can be generated in DNA consecutively to the initial formation of peroxyl pyrimidine radicals that subsequently react with vicinal bases [44]. Other types of complex damage whose formation involves a single initial radical hit include DNA-protein cross-links, intra- and interstrand DNA cross-links [28]. In this short review article emphasis is placed on recent aspects of the formation of non-single oxidatively generated damage in cellular DNA that receive increasing attention due to their deleterious biological potential [45]. 2. Hydroxyl radical and one-electron oxidants as inducers of complex DNA damage As already briefly mentioned, the oxidation reactions indentified so far in DNA may be mostly explained in terms of initial involvement of OH, one-electron oxidant or 1O2. It is now well documented that 1O2 reacts specifically with guanine producing 8-oxoGua in cellular DNA at the exclusion of rearrangement products [23,27,46]. As a result, the lack of formation of the reactive quinonoid intermediate that is observed in free 20 -deoxyguanosine upon [4 + 2] cycloaddition of 1O2 to the purine base [47] would prevent the formation of DNA-protein cross-links. So far only OH and one-electron oxidants have been shown to generate organic radicals that are able in a second step to react either as radical cations, carbon centred radical or peroxyl radical with other DNA constituents or proteins. Typically OH may be produced in cells as the result of Fenton reactions that involve the reduction of H2O2 by ferrous ions and with a lower efficiency and specificity by copper ions [48]. Radiolysis of water molecules through the indirect effects of ionizing radiation is another convenient source of OH [21]. However as a relatively minor process, direct interaction of ionizing radiation with either the 2-deoxyribose moieties or any of the four bases of DNA leads in cells to the generation of the corresponding radical cations [21]. A suitable way to produce purine and pyrimidine base radical cations in DNA consists in exposing cells to high intensity nanosecond 266 nm laser pulses [49]. Recently it was found that 6-thioguanine (TG), once metabolized and inserted into cellular DNA is able under absorption of UVA light to specifically oxidize proximal guanine bases by one-electron abstraction [50]. One may also mention that one-electron oxidation of guanine may occur in cells during inflammation processes that give rise  to O 2 and NO [48]. Recombination of the two later species give rise to peroxynitrite anion (ONOO-) that is converted upon reaction with CO2 into nitrosoperoxycarbonate, the precursor of carbonate anion radical (CO 3 ), a strong one-electron oxidant [51,52]. 3. One radical hit-mediated intra- and inter-strand DNA damage 3.1. Tandem base modifications Two main types of tandem base modifications have been shown to be generated by one radical hit that may involve OH and/or oneelectron oxidants. Subsequently, either pyrimidine centered radicals or related pyrimidine peroxyl radicals thus generated are able to react with the adjacent base. The efficiency of the intramolecular reaction is higher when the target base is located on the 50 -side with respect to the reactive pyrimidine radicals. This may be rationalized in terms of shorter distances between the pyrimidine rad-

ical and the vicinal purine base involved in the addition or oxidation reactions. 3.1.1. Addition of pyrimidine carbon centered radicals to adjacent guanine Several model studies have shown that 5-(2-deoxyuridylyl) methyl and 6-hydroxy-5,6-dihydro-20 -deoxycytid-5-yl radicals are able to bind to the carbon 8 of the 50 -adjacent purine bases giving rise to tandem base lesions [53–56]. However a limiting factor to the intramolecular reaction is the presence of O2 that efficiently reacts with carbon centered radicals. This explains why only the predominant G[8-5m]T and G[8-5]C lesions (Fig. 1) whose formation involves the generation of a covalent bond between either the methyl group of thymine or the C5 carbon of cytosine and 50 -guanine has been detected in cellular DNA upon exposure to H2O2, the likely precursor of highly reactive OH through Fenton type reactions [55,56]. The measurement of G[8-5m]T and G[8-5]C that are generated in very low yields, typically 0.050 and 0.037 lesions per 109 normal nucleosides in about 30–50 lg extracted DNA has required the use of an accurate and highly sensitive HPLC/MS3 assay in order to prevent artefactual detection of erratic ions. Relevant information on the biochemical processing of both G[8-5m]T and G[8-5]C intrastrand cross-links [57] has been gained from in vitro replication studies [55,58–60] and shuttle vector experiments combined with HPLC-MS/MS analysis [56]. Both tandem lesions are able to block high-fidelity DNA polymerases [55,58–60] while replication of G[8-5]C in AB1157 Escherichia coli cells led to an elevated level of G ? T and G ? C mutations [56]. It was recently shown that in E. coli strains, polymerase IV (pol IV) and polymerase V (pol V) would allow bypassing G[8-5m]T cross-link that may be error prone. In that respect pol IV and pol V are implicated in the observed T deletions and most G to T transversions respectively [61]. It was also reported that yeast and human DNA g polymerases are able to bypass G[8-5]C lesion with however a lower efficiency and a reduced fidelity in nucleotide incorporation with respect to control [55,62,63]. Evidence has been provided for the implication of the nucleotide excision repair (NER) pathway in the removal G[8-5m]T and G[8-5]C at least in bacterial cells as shown by the recognition and incision of the tandem base lesions inserted into site-specific oligonucleotides by UvrABC nuclease [64,65]. 3.1.2. Implication of pyrimidine peroxyl radicals in addition reactions with vicinal bases Earlier observations have shown that X-irradiation of dinucleoside monophosphates and d(CpGpTpA) in aerated aqueous solutions gave rise to tandem base lesions consisting of 8-oxo7,8-dihydro-20 -deoxyguanosine (8-oxodGuo) and N-(2-deoxy-bD-erythro-pentofuranosyl)formylamine (dF) as the result of one initial OH hit on either thymine or cytosine [66,67]. It may be remembered that dF is one of the main OH degradation products of thymidine [68] or 20 -deoxycytidine [29] in aerated aqueous solutions. As observed for the formation of pyrimidine-guanine intrastrand cross-links there is strong sequence dependence on the OH-mediated formation of the latter vicinal base lesions in isolated DNA since the yield of 8-oxodGuo/dF (Fig. 2) is about 20-fold higher than that of dF/8-oxodGuo [69]. The mechanism of formation of 8-oxodGuo/dF and dF/8-oxodGuo that was inferred from [18O]-labeling experiments was rationalized in terms of addition of 5-(6)-hydroxy-6-(5)-peroxy-5,6-pyrimidyl radicals to C8 of guanine with subsequent cleavage of the peroxide bond thus generated and further rearrangement [44]. It was recently estimated that about 50% of 8-oxodGuo and 8-oxo-7,8-dihydro-20 -deoxyadenosine formed in aerated aqueous solutions of isolated DNA upon exposure to OH were part of tandem base lesions as the result of initial addition of peroxyl pyrimidine radicals at C8 of

7

J. Cadet et al. / Cancer Letters 327 (2012) 5–15

H N

H2N

O

N

N

HO

N

O

H N

H2N

O

N

N

NH2

N

O

N

HO O

O O P O O

O

N

O

O

-

N

G[8-5m]T

O

O O P O O

HO

HO

O

N

N

NH

O O P O O

O

N

HO

O

N

O

H N

H2 N

N O HO

GT

G[8-5]C

O HN

O

H2N

N N

N HO

O

N

N

H

P

O

O

N

O

HO

N

NH2 N

O

N

N

NH

N

OO

NH2

O

NH2

N

O

HO

N

N

O

H

O O-

O P

GCT

O

OH

O

OH

CyclodGuo

O

CyclodAdo

OH

Intrastrand cross-links Fig. 1. Three types of oxidatively generated intrastrand cross-links including tandem base cross-links, guanine–thymine adduct and 50 ,8-cyclo-20 -deoxyadenosine.

H N

H2N

O

-

O

N O

N

NH

O O P O O

O HO

N

O2

O

N

O

•

OO

O O P O O

HO

-

O

O

O

NH N O

HO

NH

HO

N

HO HO •

H N

H2N N

N

N HO O

H N

H2 N

O

N O

O -O P O O

O

HN

O

HO

Fig. 2. Oxidatively generated tandem guanine–thymine lesion through peroxyl pyrimidine radicals.

guanine or adenine [70]. However so far, attempts to detect 8oxodGuo/dF in cellular DNA have been unsuccessful. This is likely due to a lack of sensitivity of the HPLC-MS/MS method to cope with the expected formation of low levels of the tandem lesions, even for the predominant 8-oxodGuo/dF that has been shown to be 10-fold lower than that of 8-oxodGuo in isolated DNA [69]. It has been also shown that pyrimidine peroxyl radicals can give rise to the formation of 8-oxodGuo within DNA through one-electron transfer reactions [70]. Evidence has been also provided for the participation of initially generated pyrimidine radicals in reactions with adjacent thymine base in oligonucleotides giving rise to tandem base lesions [71,72]. Another reaction of peroxyl pyrimidine radical in DNA strands is hydrogen atom abstraction from the vicinal 2-deoxyribose moiety at C1 of 50 -nucleoside that leads to the formation of tandem modifications consisting of 2-deoxyribonolactone and 5,6-dihydroxy-5,6-dihydrothymidine (ThGl) [73]. The mutagenicity of ThGl that was inferred from replication studies in E. coli cell hosts was found to be significantly enhanced by the presence of the flanking 2-deoxyribonolactone [73]. The mutagenic features of 8-oxodGuo/dF upon replication of a unique lesion containing single-stranded DNA shuttle vector into simian COS7 cells were found to be a combination of the contribution of the two

isolated base lesions [74]. The mutagenic ability of two other tandem base lesions consisting of 8-oxodGuo and thymidine glycol (ThGl) whose mechanism of formation remains to be established were assessed on the basis of in vitro replication studies [75] and shuttle vector investigations with E. coli cells as the hosts [76]. As noted for the 8-oxodGuo and dF tandem lesions, either 8-oxodGuo/ThGl or ThGl/8-oxodGuo were reported to be more mutagenic than single base lesions [76]. More specifically the presence of ThGl led to a significant increase in G ? T transversions frequency [76]. It was reported that the defined sequence DNA fragments in which either 8-oxodGuo/dF or dF/8-oxodGuo were site specifically inserted when paired with complementary strands were prone to enzymatic strand cleavage by DNA repair glycosylases including bacterial endonuclease III and formamidopyrimidine DNA N-glycosylase [77]. Strong sequence effects were observed on the repair efficiency of the oxidized purine base of 8-oxodGuo/ThGl and ThGl/8-oxodGuo tandem lesions by human 8-oxoguanine DNA glycosylase (hOGG1) [76]. The presence of a 50 -adjacent ThGl lesion considerably affects the hOGG1-mediated excision of the oxidized purine base while an enhancement of the release of 8-oxo-7,8dihydroguanine was observed when the thymine glycol (ThyGl) was on the 30 -end [76] (See Table 1).

8

J. Cadet et al. / Cancer Letters 327 (2012) 5–15

Table 1 Classes of oxidatively base damage to cellular DNA: oxidizing species involved in the formation in the lesions and assays for their monitoring. DNA damage

Reactive species and oxidizing agents

Single oxidized bases Tandem base lesions Purine 50 ,8-cyclonucleosides DNA-protein cross-links DNA interstrand cross-links (G) DNA interstrand cross-links (C40 ) DNA double strand breaks Bistranded OCDLs



Methods of measurement

1

OH, O2, one-electron oxidants OH, one-electron oxidants OH One-electron oxidants One-electron oxidants  OH  OH, ionization  OH, ionization

HPLC-MS/MS HPLC/ECD enzymatic assays HPLC-MS/MS HPLC-MS3 HPLC-MS/MS SDS Page Denaturing PAGE HPLC-MS/MS Neutral comet assay, PFGE, cH2AX foci Enzymatic assays

 

3.1.3. Nucleophilic addition As it will be discussed later, one of the two main reactions of the guanine radical cation is its ability to undergo reactions with nucleophiles that may involve nucleobases or bound proteins. Recent evidence was provided for the formation of intrastrand crosslink between guanine at C8 and nearby thymine at N3 in either single-stranded oligonucleotides [78] or DNA duplexes [79] upon selective one-oxidation of the guanine bases by the carbonate radical anion. Nucleophilic addition through N3 of either adjacent thymine or a more distant one separated by one cytosine base, gave rise to intrastrand guanine-thymine lesions in a 5 G-T 30 and 50 GCT30 sequence context (Fig. 3). However the search for the formation of the latter intrastrand cross-links in cellular DNA that could be used as relevant biomarkers of inflammation is awaiting further experiments. Relevant structural features on the guaninethymine cross-links when inserted into DNA were gained from modeling, molecular dynamics, energy calculations and thermal melting studies [80]. It was found that both lesions dynamically distort and destabilize the DNA duplex structure, the effects being more pronounced in GCT lesion than in the GT intrastrand cross-link. Both cross-linked intrastrand base lesions are substrates for the enzymes of the NER machinery as shown from the results of dual incision assay using a site-specific modified 135-mer as the probe and human HeLa cell extracts as the source of repair enzymes. The more flexible GCT cross-link was excised 4-fold more efficiently than the GT intrastrand tandem base lesion [80]. 3.2. Tandem base-sugar lesions: purine 50 ,8-cyclonucleosides Increasing interest is currently devoted to chemical and biological aspects of purine 50 ,8-cyclonucleosides initially characterized in the 1970s in model studies and isolated DNA [for a recent comprehensive review, see 81]. It was postulated that these intrastrand cross-links could be implicated in neurological disorders associated with pathologies such as xeroderma pigmentosum as the result of dysfunction in the NER pathway [82–84]. A large body of information is now available on the mechanism of formation of 50 ,8-cyclo20 -deoxyadenosine (cdAdo) and 50 ,8-cyclo-20 -deoxyguanosine (cdGuo) that both are present in isolated DNA as pairs of 50 R and 50 S diastereomers [85]. This was rationalized in terms of OH-mediated hydrogen abstraction from the sugar 5-hydroxymethyl group

followed by intramolecular radical addition to the C8 of the purine bases at rate constants of kc = (1.6 ± 0.2)  105 s1 and (6.9 ± 0.8)  105 s1 for adenine and guanine respectively [86,87]. The purinyl radicals thus formed with the impair electron mostly localized at N7 [88] are oxidized in aerated aqueous solutions by O2 giving rise in DNA to the predominant 50 R diastereomer for both adenine and guanine 50 ,8-cyclonucleosides in a ratio  4:1 with respect to the 50 S diastereomer [85]. It was shown that the presence of oxygen that efficiently adds to C50 sugar radical leads to a significant decrease in the formation of both cdAdo (Fig. 1) and cdGuo tandem lesions in isolated and cellular DNA [85]. The accurate measurement of the purine 50 ,8-cyclonucleosides has been hampered for a decade by the use of inappropriate methods including GC-MS and HPLC-MS [89,90] that has led to strong overestimation of the yields for reasons that were recently discussed [85]. HPLC-MS/MS, a highly quantitative method, has been found to be a much suitable alternative. A 2 kGy dose of gamma rays has been, however, necessary to allow detection in cellular DNA of the most expected (5R0 )-cdAdo, by considering information gained from model studies. The measured radiation yield of (5R0 )-cdAdo, 0.2 lesion per 109 normal nucleosides and per Gy in THP1 human monocytes is about two orders of magnitude lower than that of 8-oxodGuo [85]. It was recently confirmed in an independent study involving accurate HPLC-MS3 measurements that purine 50 ,8-cyclonucleosides represent a minor class of oxidatively generated damage to DNA. Thus the level of any of the four diastereomers of cdAdo and cdGuo was found on the average to be two orders of magnitude lower than that of single oxidized nucleosides including 5-formyl-20 -deoxyuridne and 5-(hydroxymethyl)-20 -deoxyuridine in the DNA of brain and liver tissues of Long-Evans Cinnamon rats that were exposed to strong endogenous oxidative conditions [91]. As expected from the tandem base-sugar structure of purine 50 ,8-cyclonucleoside, it was found that 50 R and 50 S diastereomers of cdAdo were not substrates for enzymes of the base excision repair (BER) pathway [82,92]. Evidence was provided that a single (50 S)-cdAdo site-specifically inserted into a 140-mer duplex DNA was excised as oligonucleotides varying between 24 and 32 nucleotides through the NER pathway by extracts from Chinese hamster ovary cells [93]. This is indicative of occurrence of NER pathway as also independently demonstrated by another set of experiments [94] that have involved the specific synthesis of 50 R- and 50 S-cdAdo

O

O

N

HN H2N

N

N dR

N

HN H2 N

N

HN

•+

O

O

N

K

H2 N

N

N

N H

HN

dR guanine-lysine crosslink

N dR

Nucleophilic base

DNA-DNA crosslinks?

Fig. 3. Lysine–guanine adduct and DNA–DNA cross-links through nucleophilic reactions of the guanine radical cation.

J. Cadet et al. / Cancer Letters 327 (2012) 5–15

containing duplex closed-circular plasmids [95,96]. Incubation of the modified DNA probes with human cell tracts led to NER synthesis and the release of oligonucleotides that contain cyclodAdo [94]. Interestingly it was found that (50 S)-cdAdo was three to four times more efficiently excised that the more abundantly formed 50 R diastereomer. However the excision of cdAdo diastereomers is much less efficient by factors varying between 40 and 150 than that of 1,3-intrastrand cis-platin adduct, one of the reference NER substrates [94]. Evidence has been provided through measurement of DNA lesions in liver rats for the implication of NEILI, a BER DNA N-glycosylase in the NER of (50 R)- band (50 S)-cdAdo according to a mechanism that remains to be elucidated [97]. Additionally it was recently reported that (50 S)-cdGuo when site-specifically inserted into a DNA plasmid was a relatively poor substrate for E. coli Uvra nuclease that operates through the NER pathway with a twofold lower efficiency than the 50 S diastereomer of cyclodAdo [98]. Relevant information on local structural perturbations induced by the presence of (50 S)-cdGuo opposite to dCyd in a DNA duplex, likely to be implicated in the recognition of the lesion by NER enzymes, has been gained from detailed NMR study and molecular mechanics calculations [99]. Both cdAdo diastereomers are expected to be cytotoxic as suggested for their observed ability to block primer extension by mammalian DNA Pol d and T7 DNA polymerase using site-specific modified templates [82]. One may also point out that (50 S)-cdAdo is more easily bypassed than the 50 R diastereomer by human pol g [94], an error-prone translesional synthesis polymerase. This constitutes a second example in addition to the different ability for NER enzymes to remove (50 R)- and (50 S)-cyclodAd, of the role played by conformation DNA changes on the biochemical processing of diastereomeric intrastrand cross-linked lesions. A recently developed next-generation sequencing method [100] together with the CRAB assay [101] and a shuttle vector based approach have been used for assessing the mutagenic and cytotoxic features of the 50 S diastereomers of cdAdo and cdGuo in different strains of E. coli cells. It was concluded that both purine 50 ,8-cyclonucleosides were strong blocks to DNA replication and highly mutagenic in E. coli cells, the most frequent mutations being G ? T transitions and A ? T transversions [100]. (50 S)-cdGuo is strongly lethal when replicating in wild type E. coli strains. The induction of SOS response [102] was found to increase survival from less than 1–5.5% and enhance bypass of the tandem lesion [98]. This was accompanied as inferred from progeny analysis by a significant elevation of mutation frequency, predominantly G ? A transitions as the result of translesion synthesis errors [98]. (50 S)-cdAdo is able to inhibit gene expression by blocking transcription by RNA polymerase II (pol II) as inferred from the reduced luciferase activity in transformed fibroblast cells of human XPA patients [93]. In addition (50 S)-cdAdo was found to enhance transcriptional mutagenesis [103] while inhibiting binding of the TATA binding protein [104] and multiple specific transcription factors to DNA including cyclic AMP response element binder (CREB) and NF-kappa B [105]. 3.3. DNA-protein cross-links DNA-proteins cross-links (DPCs) represent an important class of DNA damage that may be produced according to different mechanism by several chemical agents including among others acrolein, arsenic compounds, cis-platinum, copper, chromate, diepoxybutane, formaldehyde, melphalan, mitomycin C and nickel [106,107]. One may add to this list, ionizing radiation that, through the likely direct effect, is able to generate about 150 DPCs/cell per Gy. This yield has to be compared with the lower radiation-induced levels of DSBs and DNA-DNA cross-links that are estimated to be 20–40 and 30 lesions/cell per Gy respectively [108]. Relevant

9

insights in the radiation-induced formation of DPCs in mammalian cells have been gained from a proteomic study that has involved the isolation of the cross-linked lesions [109] using the suitable and sensitive DNAzol-silica method [110]. Thus in addition to actin and histone H2B, a large variety of proteins including structural proteins, regulatory proteins and GDP/GTP-binding proteins were shown to be implicated in DPC formation [109]. It was also found that in contrast to previous observations [for a review article, see 108] that have necessitated high dose of ionizing radiation, oxygen does not exert inhibitory effects on the formation of DPCs [109]. It has been proposed that the radiation-induced formation of DPC in nucleohistone and in cells may involve recombination reactions of pyrimidine base radicals with either phenoxyl radical derived from tyrosine or lysine radicals [111–113]. The presence of oxygen that is known to react efficiently, at least, with all pyrimidine base radicals would prevent significant formation of the proposed DPCs even under hypoxic conditions. An alternative mechanism that would implicate addition of 5-(uracilyl)methyl radical to tyrosine followed by one-oxidation of the resulting radical appears to be more reasonable to explain the formation of 3-[1,3-dihydro-2,4dioxopyrimidine-5-yl)-methyl]-L-tyrosine, the thymine-tyrosine (Thy-Tyr) cross-link [113]. However the formation of Thy-Tyr cross-link in cells that was detected using the questionable GCMS assay [31] remains a pending issue that would need further accurate measurements in order to further support the proposed mechanism. An alternative pathway for explaining the radiation-induced formation of DPCs is based on the observation that the guanine radical cation (G+) is able to react efficiently with nucleophiles such as water [114]. It was found that the free e-amino group of central lysine residue of trilysine peptide bound to TGT trinucleotide is able to covalently attach to C8 of G+ generated by type I photosensitization reaction [115]. Subsequent one-electron oxidation of the aminyl radical thus generated leads to a lysine-guanine cross-link that has been unambiguously characterized on the basis of extensive and accurate NMR and mass spectrometry measurements [115]. In addition arginine and serine but not tyrosine are able to act as nucleophiles and react with G+ to form cross-links with guanine (unpublished results). This independent O2 mechanism that involves as the key step the nucleophilic addition of a free amino group to G+ has received further support from several experiments for which isolated DNA and several proteins were the targets with either flash-quench [116,117] or type I photosensitizers [118] as the agents of one-electron oxidation of guanine. The use of restriction endonuclease assay for mapping DPCs in a pBR322 DNA plasmid has established that the guanine bases involved in the cross-links with histone III-S were the most susceptible DNA targets of nucleophilic addition of proteins [117]. Interestingly, the formation of DPCs upon exposure of human necrosis factor NF-kappaB p50 homodimer and a 37-mer duplex DNA to high intensity nanosecond 266 nm laser pulses has been used to map contact points between the two biomolecules that consist mostly of guanines in the DNA fragment [119]. Another example of oxidative formation of DPCs is provided by UVA-mediated photochemical reactions of 6-thioguanine (TG) [120], once metabolized and inserted into DNA following incubation of cells with 6-mercaptopurine, azathioprine or TG that are efficient anticancer drugs and immunosuppressants [121,122]. TG that strongly absorbs in the UVA range has been found to be an efficient photosensitizer operating through both type I and type II mechanisms allowing one-electron oxidation of guanine and the release of 1O2 respectively. Therefore it is likely that the stable DPCs that were measured in human cells are accounted for by the covalent attachment of nucleophilic residues, such as the lysine e-amino groups, of bound proteins as discussed above (Fig. 3). This hypothesis that applied also to the high-intensity UVC laser induced formation of

10

J. Cadet et al. / Cancer Letters 327 (2012) 5–15

DPCs remains to be challenged in cells by the search of the guanine-lysine adduct, previously identified in a model study [115]. There is still a paucity of information of the DNA repair of DPCs, particularly for those induced by ionizing radiation. It has been suggested that these bulky tandem lesions, considered to be strong blocking damage, are repaired by NER in bacteria when small DPCs are concerned and through the homologous recombination pathway in mammalian cells for the processing of oversized DPCs [123]. It has also been postulated that proteolytic digestion of the covalently bound enzyme to DNA could be implicated in the initial repair process before to be completed by NER enzymes [124]. In addition other types of DPCs than those discussed above could be generated under oxidative stress conditions. Model studies have shown that 2-deoxyribonolactone [30], an oxidized abasic site, is able to undergo cross-link formation with enzymes of the BER pathway including endonuclease III and polymerase b [125– 127]. Another example of cross-link is illustrated by the covalent attachment of binding proteins to oxanine, a nitric oxide-mediated decomposition product of guanine [128]. Evidence has been also provided for the efficient formation of transient DPCs in nucleosome core particle between the lysine residues through amino group of H3 and H4 histone proteins and abasic sites [129] that may arise from either base excision repair of oxidized bases [130] or spontaneous hydrolysis of modified bases such as formamidopyrimidines derived from OH reaction with guanine and adenine [48]. The DPC thus formed has been shown to promote through b-elimination reaction induction of a strand break on the 30 -end [129]. However the formation of these various DPCs that have been observed in model systems remains to be established in cells. 3.4. DNA interstrand cross-links Three main oxidative pathways involved in the formation of interstrand cross-links (ICLs) between the two opposite DNA strands have been identified so far with two of them having been shown to occur in cells. 3.4.1. Nucleophilic addition to guanine radical cation Evidence has been provided for the radiation-induced formation of ICLs in cells [108,131] in a similar yield to DSBs [108]. However information is still lacking on the nature and the formation mechanism of these deleterious types of DNA damage. In addition ICLs have been found to be efficiently generated in DNA duplex upon exposure to high intensity UVC laser irradiation [132] and UVA irradiation of TG containing double-stranded DNA fragments [133,134]. A common mechanism for the formation of ICLs under these different conditions of irradiation would involve initial generation of G+ followed by reaction of this intermediate with opposite guanine or cytosine (Fig. 3) as shown by polyacrylamide agarose gel analysis of UVC laser irradiated double-stranded oligodeoxynucleotides (unpublished data). The photo-induced formation of ICLs in cells pre-incubated with TG has led to the induction of chromosome aberrations together with activation of the Fanconi anemia (FA) pathway [134]. Interestingly FA defective cells were found to be hypersensitive to killing, an observation that is suggestive of a strong cytotoxicity induced by the photosensitized formation of ICLs. 3.4.2. Implication of C40 oxidized abasic site Evidence for the OH-mediated formation of ICLs that involves the transient generation of the so-called ‘‘C40 oxidized abasic site’’ arising from hydrogen abstraction at the C4 of the 2-deoxyribose moiety has been gained from detailed studies performed on isolated DNA and polynucleotides [135,136]. This has led to the

isolation and characterization of the four diastereomers of 6-(2deoxy-b-D-erythro-pentofuranosyl)-2-hydroxy-3(3-hydroxy-2oxopropyl)-2,6-dihydroimidazo[1,2-c]-pyrimidin-5(3H)-one that derive from dCyd and may exist in the closed form (Fig. 4). A key step in the sequence of reactions leading to ICLs [137] is the generation through a b-elimination reaction of a highly reactive unsaturated keto-aldehyde that is able to undergo efficient cycloaddition across the 5,6-pyrimidine bond of opposite dCyd. In subsequent studies it was confirmed that the formation of ICLs is strongly sequence dependent [138,139]. Thus the b-elimination reaction that leads to the cleavage of the 30 -phosphodiester bond of the C40 abasic site occurs when 20 -deoxyadenosine (dAdo) and to a lesser extent dCyd are located on the complementary strand. In contrast opposite 20 -deoxyguanosine (dGuo) was not able to promote the formation of ICLs [138] as inferred from experiments in which the C40 radical was site-specifically generated in defined sequence doublestranded oligonucleotides from a photolabile precursor. The four dCyd adducts released after enzymatic digestion of ICL containing DNA have been detected by HPLC-MS/MS measurements in human monocytes either exposed to gamma rays or incubated with bleomycin [136], a therapeutic agent that mostly acts on DNA by abstracting hydrogen at C40 . The radiation-induced formation efficiency of the ICL is low since the frequency of this complex lesion that includes an adjacent strand break on the 30 -end represents about 1% of the level of predominant 8-oxodGuo. The repair of the dCyd adducts in the DNA of human cells that were treated with bleomycin is a slow process since 24 h were required to remove about 90% of the bistranded lesions [136]. 3.4.3. Implication of C50 sugar radical In addition to the main conversion pathways C5-yl radical of 2deoxyribose moiety of DNA in aerated aqueous solution that includes intramolecular cyclization to the C8 of purine bases and the formation of 5-aldehyde [30], there is a third reaction that gives rise according to a still not fully elucidated mechanism to 50 -(2phosphoryl-1,4-dioxobutane) (DOB) [140]. The precursor of DOB that has been proposed to be a C5 oxyl radical when neocarzinostatin is the DNA damaging agent, would generate a C4 centred radical upon cleavage of C4–C5 bond through a b-scission reaction [140]. The acyclic form of DOB has been found to be implicated in the formation of ICL in a strong sequence manner since only opposite dAdo to a 30 -adjacent thymidine (Thd) is able to significant react in order to form an ICL, although dCyd exhibits a similar reactivity [141]. The condensation product has been unambiguously characterized as arising from the addition of the bis-electrophilic DOB to the 6-amino group of dAdo [140]. Validation of this oxidative pathway in cell is awaiting further experiments that may involve the measurement of the dAdo addition product. 4. Multiple radical hit-induced clustered DNA damage It is well documented that energy deposition along the track of 200 keV–1 MeV photons leads to the generation of OH, ionization events and secondary electrons in a very close vicinity. This gives rise when the nucleus is the target of either ionizing radiation or heavy ions to the formation of complex types of damage including double strand breaks (DSBs) and non-DSB oxidatively generated clustered DNA lesions (OCDLs) through transient multiple radical and excitation events [21,22,41,142]. 4.1. Double strand breaks DSBs are cytotoxic lesions [40,143] that arise from the formation of at least one nick on each of the two DNA strands within one or two helix turns. DSBs that may be also generated by

11

J. Cadet et al. / Cancer Letters 327 (2012) 5–15

OH N O N

O

O O O

O

Base O

O

N

O

O

O

Base elimination β elimination

O

NH2

OH

C4' radical O

O

N

N

O

N O N

O O

dCyd

O

N

O

O O

ICL

Fig. 4. Interstrand DNA crosslinks involving reaction of the oxidized C40 abasic site with opposite cytosine.

sub-micron radiation microbeams from near infrared femtosecond lasers [144], enediyne drugs [145] and biochemical pathways including topoisomerase II failures [146], collapse of stalled DNA replication forks [147,148] and repair of OCDLs [149–151] are considered as molecular signatures of the effects of ionizing radiation on cellular DNA. In the latter case, DSBs constitute a heterogeneous class of clustered DNA damage that may differ in size and by the presence of other lesions such as oxidized bases, abasic sites and modified sugar extremities. It may be added that the complexity of DSB structure increases with that of linear energy transfer (LET) of photons and heavy ions [40,45,151,152]. Several methods including pulsed-field gel electrophoresis (PFGE), neutral comet assay, histo-immunofluorescence detection of cH2AX foci and fluorescence detection of GFP-tagged proteins as DSB response markers are available for the measurements of DSBs in cells [153–156]. The cH2AX based assays whose specificity for detecting DSBs has been questioned [157–160] are highly sensitive since foci have been detected upon exposure to a dose of ionizing radiation as low as 1.2 mGy [161]. It has been estimated that one Gy of gamma or X-rays is capable of inducing 40 DSBs in nuclear DNA. This number has to be compared with the 1000 SSBs and 2000 modified bases that are also generated under these conditions [41]. The repair of DSBs that is crucial to maintain genome stability and prevent chromosome aberrations to occur is mainly operated in mammalian cells by the non-homologous DNA end joining (NHEJ) pathway in an error-prone fashion [162,163]. In addition less frequent homologous recombination (HR) that is error free is specifically involved in the removal of DSBs localized in heterochromatic DNA regions in S/G2 phase cells [164]. In addition NER, mismatch repair and cell cycle regulation have been suggested to participate in the repair of DSBs through a coordination network with NHEJ and HR [165]. It has been shown that the presence in DNA of upstream lesions including oxidized bases and abasic sites may constitute severe blocks to NHEJ [166]. 4.2. Non-DSB oxidatively generated clustered DNA lesions A second class of complex DNA damage that includes oxidation products of nucleobases and 2-deoxyribose moieties of DNA is generated upon exposure of cells to ionizing radiation and heavy ions [37,42,167]. None of these clustered lesions have been so far identified X- or gamma irradiated cellular DNA since the measurement of such complex types of clustered DNA damage remains a

highly challenging analytical issue. In fact, each of these radiation-induced lesions is expected to be unique if one considers sequence effects and the high number of possibility in terms of modified bases and sugar residues including single strand breaks (SSBs) that may be produced according to available data from model studies [27–30] and accurate HPLC-MS/MS measurement of single lesions in cellular DNA [168,169]. However information on the nature and distribution of non-DSB clustered lesions has been gained mostly from Monte Carlo calculations [22,142,170,171] and biochemical measurements that involved the use of DNA repair enzymes including Fpg, hOGG1, endo III and related human NTH1 to convert base lesions into SSBs [37,38,142,172]. In addition endonuclease IV (endo IV) and human AP endonuclease (Ape1) have been used to cleave DNA fragments at the sites of abasic modifications [37,38,172]. Therefore enzymatic incubation of DNA embedded in agarose plugs is expected to generate DSBs within one or two DNA helix turns when sub-classes of bi-stranded clustered lesions exhibit at least one SSB on one DNA strand and modified bases and/or abasic sites on the opposite strand. In a subsequent step, the newly generated DSBs are detected by either constant electrophoresis or PFGE [37,38,142,172]. Among numerous data that have been reported on the quantitative formation of non-DSB clustered lesions, one may quote a recent study in which a good agreement was obtained between the levels of radiation-induced Fpg- and Endo III-clusters in normoxic HeLa cells and Monte Carlo Damage Simulation (MCDS) calculations [142]. The yields of Fpg clusters and Endo III clusters that were assessed using an optimized version of the combined DNA glycosylase-PFGE assay were found to be 27 and 25 clustered lesions per Gy and per 109 nucleobase respectively upon exposure to 5 Gy of c-rays [142]. This has to be compared with HPLC-MS/MS measurement of the overall yield of the main radiation-induced purine base lesions including 8-oxoGua, FapyGua and FapyAde, all three excellent substrates for Fpg enzymes, that was estimated to be 67 lesions per Gy and 109 bases in the DNA of human monocytes [168]. Therefore one may note that the frequency of reported Fpg clusters [168] that represents about 40% of the damage measured by HPLC-MS/MS as single oxidized purine nucleosides appears to be high. It is also intriguing that the steady-state levels of non-DSBs clustered lesions were found to be elevated in two human hematopoietic cell lines and human cancer tissues [173–175]. It has been suggested that under conditions of high stress OCDLs may be generated [18]. However there is no mechanistic rationale to explain the

12

J. Cadet et al. / Cancer Letters 327 (2012) 5–15

exacerbated formation of clustered damage under conditions where mostly Fenton reactions are implicated in oxidative reactions to cellular DNA. Another questionable data concern the formation of non-DSB clustered lesions and DSBs upon exposure of supercoiled pUC18 DNA to a combination of UVB and UVA radiations [176]. In that respect, the formation of significant amounts of DSBs in the DNA of cells exposed either to UVB or UVA radiations has been recently ruled out [177–179]. Further experiments including the development of more specific methods are required to further resolve these somewhat controversial issues and also to gain more quantitative data on the radiation-induced generation of non-DSB bistranded clustered DNA lesions. A large body of information is available on the processing of non-DSB clustered DNA lesions using site-specific modified oligonucleotides that contain either two or three lesions including altered bases, an abasic site and/or a SSB [for a recent comprehensive review, see 45]. Several modified bases including 8-oxoGua, 8-oxoAde, ThyGl, 5-formyluracil (5-ForUra), 5-hydroxyuracil (5-OHUra), 5,6-dihydrothymine (hThy) and 5,6-dihydrouracil (hUra) have been used for the construction of multiply damaged sites [150,151,180–184]. One may however note that 5-OHUra is not generated at least in detectable amounts in irradiated cellular DNA as well as hThy and hUra that are not oxidatively generated base lesions. As a general observation, it was shown by several groups that the removal of the two or three lesion-bistranded clustered damage, usually slower than that of related single modifications, is mostly achieved by the BER machinery including DNA Nglycosylases and AP endonucleases. There is also a hierarchy in the DNA N-glycosylase-mediated repair of bi-stranded OCDLs that is strongly dependent on the nature and the distribution of the modified DNA components within the clusters. It may be added that DSBs could be generated during the repair of some three-lesion clustered damage [150,151] whereas mutagenesis could be strongly increased [180,181] due to the impairment of DNA N-glycosylase activities [185]. Evidence was recently provided that clustered DNA lesions induced by Fe ions are refractory to repair, thus promoting genome instability through chromosome breakage [186].

5. Conclusions Relevant information has been gained on the oxidative formation of complex types of DNA damage as the result of either one hit or several simultaneous radical events within a localized part of DNA. Relevant mechanistic insights have been obtained on the formation of DNA-protein and DNA-DNA cross-links that is based on a reasonable hypothesis involving in both cases nucleophilic addition to the guanine radical cation. This is further supported by the identification of intrastrand cross-links between guanine and thymine that may be also rationalized by a similar mechanism. Efforts have now to be made to develop appropriate methods, like the HPLC-MS/MS approaches, to search for the presence of these complex lesions and also of tandem base modifications in cellular DNA. As already mentioned there is also a strong need of better assessing the formation of non-DSBs oxidatively generated clustered lesions in cells exposed to ionizing radiation. The somewhat intriguing issue that concerns the endogenous formation of OCDLs in cells needs to be carefully addressed. Complementary information on the repair and mutagenic properties of most of the identified so far complex DNA damage is also waiting further experiments based in particular on the use of shuttle vector techniques. Altogether this should provide the necessary information for a better assessment of the biological role on the various classes of oxidatively generated lesions to DNA.

Acknowledgements The authors are grateful to the ‘‘Association pour la Recherche sur le Cancer-ARC’’ (Grant 1424/2011 to DA), Agence Nationale de la Recherche (grant ANR-09-PIRI-0022 (to J-LR and MTP), Electricité de France (to J-LR) and CEA Eurotalents program (to MTP) for financial support. JC is member of EU network COST Action CM0603 ‘‘Free Radicals in Chemical Biology’’.

References [1] D. Ziech, R. Franco, A. Pappa, M.I. Panayiotidis, Reactive oxygen species (ROS) – induced genetic and epigenetic alterations in human carcinogenesis, Mutat. Res. 711 (2011) 167–173. [2] S.V. Avery, Molecular targets of oxidative stress, Biochem. J. 434 (2011) 201– 210. [3] S. Reuter, S.C. Gupta, M.M. Chaturvedi, B.B. Aggarwal, Oxidative stress, inflammation, and cancer: how are they linked?, Free Radic Biol. Med. 49 (2010) 1603–1610. [4] L.R. Ferguson, Chronic inflammation and mutagenesis, Mutat. Res. 690 (2010) 3–11. [5] F. Fannati, M. Piciocchi, E. Lavezzo, M. Bortolami, R. Cardin, Oxidative stress and inducible nitric oxide synthase induction in carcinogenesis, Dig. Dis. 28 (2010) 579–584. [6] P. Lonkar, P.C. Dedon, Reactive species and DNA damage in chronic inflammation: reconciling chemical mechanisms and biological fates, Int. J. Cancer 128 (2011) 1999–2009. [7] S.K. Powers, W.B. Nelson, M.B. Hudson, Exercise-induced oxidative stress in humans: cause and consequences, Free Radic. Biol. Med. 51 (2011) 942–950. [8] K. Jomova, M. Valko, Advances in metal-induced oxidative stress and human disease, Toxicology 283 (2011) 65–87. [9] D.B. Lombard, K.F. Chua, R. Mostoslavsky, S. Franco, M. Gostissa, F.W. Alt, DNA repair, genome stability, and aging, Cell 120 (2005) 497–512. [10] R.S. Sohal, W.C. Orr, The redox hypothesis of aging, Free Radical. Biol. Med. 52 (2012) 539–555. [11] R. Takayanagi, T. Inoguchi, K. Ohnaka, Clinical and experimental evidence for oxidative stress as an exacerbating factor of diabetes mellitus, J. Clin. Biochem. Nutr. 48 (2011) 72–77. [12] M.M. Elahi, Y.X. Kong, B.M. Matata, Oxidative stress as a mediator of cardiovascular disease, Oxid. Med. Cell. Longev. 2 (2009) 259–269. [13] K. Satoh, B.C. Berk, H. Shimokawa, Vascular-derived reactive oxygen species for homeostasis and diseases, Nitric Oxide 25 (2011) 211–215. [14] M.A. Lovell, S. Soman, M.A. Bradley, Oxidatively modified nucleic acids in preclinical Alzheimer’s diseases (PCAD) brain, Mech. Ageing Dev. 132 (2011) 443–448. [15] M.J. Devine, H. Plun-Favreau, N.W. Wood, Parkinson’s disease and cancer: two wars, one front, Nat. Rev. Cancer 11 (2011) 812–823. [16] S. Toyokuni, Molecular mechanisms of oxidative stress-induced carcinogenesis: from epidemiology to oxygenomics, IUBMB Life 60 (2008) 441–447. [17] O.A. Sedelnikova, C.E. Redon, J.S. Dickey, A.J. Nakamura, A.G. Georgakilas, W.M. Bonnner, Role of oxidatively induced DNA lesions in human pathogenesis, Mutat. Res. 704 (2010) 152–159. [18] T.B. Kryston, A.B. Georgiev, P. Pissi, A.G. Georgakilas, Role of oxidative stress and DNA damage in human carcinogenesis, Mutat. Res. 711 (2011) 193–201. [19] J.E. Klaunig, Z. Wang, X. Pu, S. Zhou, Oxidative stresss and oxidative damage in chemical carcinogenesis, Toxicol. Appl. Pharmacol. 254 (2011) 86–99. [20] A. Kotsinas, V. Aggarwal, E.-J. Tan, B. Levy, V.G. Gorgoulis, PIG3: a novel link between oxidative stress and DNA damage response in cancer, Cancer Lett., (2012), http://dx.doi.org/10.1016/J.canlet.2011.12.009. [21] C. von Sonntag, Free-Radical-Induced DNA Damage and its Repair. A Chemical Perspective, Springer-Verlag, Berlin, Heidelberg, New York, 2006. [22] H. Nikjoo, P. Girard, A model of the cell nucleus for DNA damage calculations, Int. J. Radiat. Biol. 88 (2012) 87–97. [23] J. Cadet, T. Douki, E. Sage, Ultraviolet radiation-mediated damage to cellular DNA, Mutat. Res. 571 (2005) 3–17. [24] B. Epe, DNA damage spectra induced by photosensitization, Photochem. Photobiol. Sci. 11 (2011) 98–106. [25] R. Visconti, D. Grieco, New insights on oxidative stress in cancer, Curr. Opin. Drug Discovery Dev. 12 (2009) 240–245. [26] G. Pratviel, B. Meunier, Guanine oxidation: one- and two-electron reactions, Chemistry 12 (2006) 6018–6030. [27] J. Cadet, T. Douki, J.-L. Ravanat, Oxidatively generated damage to the guanine moiety of DNA: mechanistic aspects and formation in cells, Acc. Chem. Res. 41 (2008) 1075–1083. [28] J. Cadet, T. Douki, J.-L. Ravanat, Oxidatively generated base damage to cellular DNA, Free Radic. Biol. Med. 40 (2010) 9–21. [29] J.R. Wagner, J. Cadet, Oxidation reactions of cytosine DNA components by hydroxyl radical and one-electron oxidants in aerated aqueous solutions, Acc. Chem. Res. 43 (2010) 564–571. [30] P.C. Dedon, The chemical toxicology of 2-deoxyribose oxidation in DNA, Chem. Res. Toxicol. 21 (2008) 206–219.

J. Cadet et al. / Cancer Letters 327 (2012) 5–15 [31] J. Cadet, T. Douki, J.-L. Ravanat, Measurement of oxidatively generated base damage in cellular DNA, Mutat. Res. 711 (2011) 3–12. [32] J.N. Patro, C.J. Wiederholt, Y.L. Jiang, J.C. Delaney, J.M. Essigmann, M.M. Greenberg, Studies on the replication of the ring opened formamidopyrimidine, Fapy.dG in Escherichia coli, Biochemistry 46 (2007) 10202–10212. [33] M. Münzel, U. Lischke, D. Stathis, T. Pfaffeneder, F.A. Gnerlich, C.A. Deiml, S.C. Koch, K. Karaghiosoff, T. Carell, Improved synthesis and mutagenicity of oligonucleotides containing 5-hydroxymethylcytosine, 5-formylcytosine and 5-carboxylcytosine, Chemistry 17 (2011) 13782–13788. [34] H. Kamiya, Mutagenicity of oxidized DNA precursors in living cells: roles of nucleotide pool sanitization and DNA repair enzymes, and translesion synthesis DNA polymerases, Mutat. Res. 703 (2010) 32–36. [35] S. Maynard, S.H. Schurman, C. Harboe, N.C. de Souza-Pinto, V.A. Bohr, Base excision repair of oxidative DNA damage and association with cancer and aging, Carcinogenesis 30 (2009) 2–10. [36] D. Svilar, E.M. Goellner, K.H. Almeida, R.W. Sobol, Base excision repair and lesion-dependent subpathways for repair of oxidative DNA damage, Antioxid. Redox Signal. 14 (2011) 2491–2497. [37] B.M. Sutherland, P.V. Bennett, O. Sidorkina, J. Laval, Clustered DNA damages induced in isolated DNA and in human cells by low doses of ionizing radiation: oxidized bases and strand breaks, Proc. Natl. Acad. Sci. USA 97 (2000) 103–108. [38] M. Hada, A.G. Georgakilas, Formation of clustered DNA damage after high-LET irradiation: a review, J. Radiat. Res. (Tokyo) 49 (2008) 203–210. [39] J.F. Ward, The complexity of DNA damage: relevance to biological consequences, Int. J. Radiat. Biol. 66 (1994) 427–432. [40] J.F. Ward, DNA damage produced by ionizing radiation in mammalian cells: identities, mechanisms of formation and repairability, Prog. Nucleic Acids Mol. Biol. 35 (1988) 95–125. [41] D.T. Goodhead, Initial events in the cellular effects of ionizing radiations: clustered damage in DNA, Int. J. Radiat. Biol. 65 (1994) 7–17. [42] P. O’Neill, P. Wardman, Radiation chemistry comes before radiation biology, Int. J. Radiat. Biol. 85 (2009) 9–25. [43] H.C. Box, J.B. Dawidzk, E.E. Budzinski, Free radical-induced double lesions in DNA, Free Radic. Biol. Med. 31 (2001) 856–868. [44] T. Douki, J. Rivière, J. Cadet, DNA tandem lesions containing 8-oxo-7,8dihydroguanine and formamido residues arise from intermolecular addition of thymine peroxyl radical to guanine, Chem. Res. Toxicol. 15 (2002) 445– 454. [45] E. Sage, L. Harrison, Clustered DNA lesion repair in eukaryotes: relevance to mutagenesis in cell survival, Mutat. Res. 711 (2011) 123–133. [46] J.-L. Ravanat, P. Di Mascio, G.R. Martinez, M.H. Medeiros, J. Cadet, Singlet oxygen induces oxidation of cellular DNA, J. Biol. Chem. 275 (2000) 40601– 40604. [47] X. Xu, J.G. Muller, Y. Ye, C.J. Burrows, DNA-protein cross-links between guanine and lysine depend on the mechanism of oxidation for formation of C5 vs C8 guanosine adducts, J. Am. Chem. Soc. 130 (2008) 703–709. [48] J. Cadet, S. Loft, R. Olinski, M.D. Evans, K. Bialkowski, J.R. Wagner, P.C. Dedon, P. Moeller, M.M. Greenberg, M.S. Cooke, Biologically relevant oxidants and terminology, classification and nomenclature of oxidatively generated damage to nucleobases and 2-deoxyribose in nucleic acids, Free Radic. Res. 48 (2012) 367–381. [49] T. Douki, J.-L. Ravanat, D. Angelov, J.R. Wagner, J. Cadet, Effects of duplex stability on charge transfer efficiency within DNA, Top. Curr. Chem. 236 (2004) 1–25. [50] R. Brem, P. Karran, Multiple forms of DNA damage caused by UVA photoactivation of DNA 6-thioguanine, Photochem. Photobiol. 88 (2012) 5– 13. [51] D.B. Medinas, G. Cerchiaro, D.F. Trindade, O. Augusto, The carbonate radical and related oxidants derived from bicarbonate buffer, IUBMB Life 59 (2007) 255–262. [52] V. Shafirovich, A. Dourandin, W. Huang, N.E. Geacintov, The carbonate radical is a site-selective oxidizing agent of guanine in double-stranded oligonucleotides, J. Biol. Chem. 276 (2001) 24621–24626. [53] H.C. Box, E.E. Budzinski, J.B. Dawidzik, J.C. Wallace, M.S. Evans, J.S. Gobey, Radiation-induced formation of a crosslink between base moiety of deoxyguanosine and thymidine in deoxygenated solutions of d(CpGpTpA), Radiat. Res. 145 (1996) 641–643. [54] S. Bellon, J.-L. Ravanat, D. Gasparutto, J. Cadet, Cross-linked thymine-purine base tandem lesions: synthesis, characterization, and measurements in gamma-irradiated isolated DNA, Chem. Res. Toxicol. 15 (2002) 598–606. [55] Y. Jiang, H. Hong, H. Cao, Y. Wang, In vivo formation and in vitro replication of a guanine-thymine intrastrand cross-link lesion, Biochemistry 46 (2007) 12757–12763. [56] H. Hong, H. Cao, Y. Wang, Formation and genotoxicity of a guanine-cytosine intrastrand cross-link lesion in vivo, Nucleic Acids Res. 35 (2007) 7118–7127. [57] Y. Wang, Bulky DNA lesions induced by reactive oxygen species, Chem. Res. Toxicol. 21 (2008) 276–281. [58] S. Bellon, D. Gasparutto, C. Saint-Pierre, J. Cadet, Guanine-thymine intrastrand cross-linked lesion containing oligonucleotides: from chemical synthesis to in vitro enzymatic replication, Org. Biomol. Chem. 4 (2006) 3831–3837. [59] C. Gu, Y. Wang, LC-MS/MS identification and yeast polymerase g bypass of a novel c-irradiation-induced intrastrand cross-link lesion G[8-5]C, Biochemistry 43 (2004) 6745–6750.

13

[60] C. Gu, Y. Wang, Thermodynamic and in-vitro replication studies of an intrastrand crosslink lesion G[8-5]C, Biochemistry 44 (2005) 8883–8889. [61] P. Raychaudhury, A.K. Basu, Genetic requirement for mutagenesis of the G[8,5-Me]T cross-link in Escherichia coli: DNA polymerases IV and V compete for error-prone bypass, Biochemistry 50 (2011) 2330–2338. [62] L.C. Colis, P. Raychaudhury, A.K. Basu, Mutational specificity of c-radiationinduced guanine–thymine and thymine–guanine intrastrand cross-links in mammalian cells and translesion synthesis past the guanine–thymine lesion by human DNA polymerase g, Biochemistry 47 (2008) 8070–8079. [63] P. Raychaudhury, A.K. Basu, Replication past the c-radiation-induced guanine-thymine cross-link G[8-5]T by human and yeast DNA polymerase g, J. Nucleic Acids (2010) (No. 101495). [64] Z. Yang, L.C. Colis, A.K. Basu, Y. Zou, Recognition and insertion of c-radiationinduced cross-linked guanine–thymine tandem lesion G[8-5Me]T by UvrABC nuclease, Chem. Res. Toxicol. 18 (2005) 1339–1346. [65] C. Gu, Q. Zhang, Z. Yang, Y. Wang, Y. Zou, Y. Wang, Recognition and incision of oxidative intrastrand cross-link lesions by UvrABC nuclease, Biochemistry 45 (2006) 10739–10746. [66] H.C. Box, H.G. Freund, E.E. Budzinski, J.C. Wallace, A.E. Maccubbin, Free radical-induced double base lesions, Radiat. Res. 141 (1995) 91–94. [67] E.E. Budzinski, J.D. Dawidzik, J.C. Wallace, H.G. Freund, H.C. Box, The radiation chemistry of d(CpGpTpA) in the presence of oxygen, Radiat. Res. 142 (1995) 107–109. [68] J. Cadet, T. Douki, D. Gasparutto, J.-L. Ravanat, J.R. Wagner, Oxidatively generated nucleobase modifications in isolated and cellular DNA, in: C. Chatgilialoglu, A. Studer (Eds.), Encyclopedia of Radicals in Chemistry, Biology and Materials, John Wiley & Sons, Chichester, UK, 2012, pp. 1319– 1344. [69] A.G. Bourdat, T. Douki, S. Frelon, D. Gasparutto, J. Cadet, Tandem base lesions are generated by hydroxyl radical within isolated DNA in aerated aqueous solution, J. Am. Chem. Soc. 122 (2000) 4549–4556. [70] F. Bergeron, F. Auvré, J.P. Radicella, J.-L. Ravanat, HO radicals induced an unexpected high proportion of tandem base lesions refractory to repair by DNA glycosylases, Proc. Natl. Acad. Sci. USA 107 (2010) 5528–5533. [71] S. Hong, K.N. Carter, K. Sato, M.M. Greenberg, Characterization and mechanism of formation of tandem lesions in DNA by a nucleobase peroxyl radical, J. Am. Chem. Soc. 129 (2007) 4089–4098. [72] A. Ghosh, A. Joy, G.B. Schuster, T. Douki, J. Cadet, Selective one-electron oxidation of duplex DNA oligomers: reactions at thymine, Org. Biomol. Chem. 6 (2008) 916–928. [73] H. Huang, S. Imoto, M.M. Greenberg, The mutagenicity of thymidine glycol in E. coli is increased when part of a tandem lesion, Biochemistry 48 (2009) 7833–7841. [74] A. Gentil, F. Le Page, J. Cadet, A. Sarasin, Mutation spectra induced by replication of two vicinal oxidative DNA lesions in mammalian cells, Mutat. Res. 452 (2000) 51–56. [75] Y. Jiang, Y. Wang, Y. Wang, In-vitro replication and repair studies of tandem lesions containing neighboring thymidine glycol and 8-oxo-7,8-dihydro-20 deoxyguanosine, Chem. Res. Toxicol. 22 (2009) 574–583. [76] B. Yuan, Y. Jian, Y. Wang, Y. Wang, Efficient formation of tandem thymine glycol/8-oxo-7,8-dihydroguanine lesion in isolated DNA and the mutagenic and cytotoxic properties of the tandem lesions in Escherichia coli cells, Chem. Res. Toxicol. 23 (2010) 11–19. [77] A.-G. Bourdat, D. Gasparutto, J. Cadet, Synthesis and enzymatic processing of oligodeoxynucleotides containing tandem base damage, Nucleic Acids Res. 27 (1999) 1015–1024. [78] C. Crean, Y. Uvaydov, N.E. Geacintov, V. Shafirovich, Oxidation of singlestranded oligonucleotides by carbonate anions: generating intrastrand crosslinks between guanine and thymine bases separated by cytosines, Nucleic Acids Res. 36 (2008) 742–755. [79] B.H. Yun, N.E. Geacintov, V. Shafirovich, Generation of guanine–thymidine cross-links in DNA by peroxynitrite/carbon dioxide, Chem. Res. Toxicol. 24 (2011) 1144–1152. [80] S. Ding, K. Kropachev, Y. Cai, M. Kolbanovskiy, S.A. Durandina, Z. Liu, V. Shafirovich, S. Broyde, N.E. Geacintov, Structural, energetic and dynamic properties of guanine (C8)-thymine(N3) cross-links in DNA provide insights on susceptibility to nucleotide excision repair, Nucleic Acids Res. 40 (2012) 2506–2517. [81] C. Chatgilialoglu, C. Ferreri C, M.A. Terzidis, Purine 50 ,8-cyclonucleoside lesions: chemistry and biology, Chem. Soc. Rev. 40 (2011) 1368–1382. [82] I. Kuruoka, C. Bender, A. Romieu, J. Cadet, R.D. Wood, T. Lindhal, Removal of oxygen free-radical-induced 50 ,8-purine cyclodeoxynucleosides from DNA by the nucleotide excision repair pathway in human cells, Proc. Natl. Acad. Sci. USA 97 (2000) 3832–3837. [83] P.J. Brooks, The case for 8,50 -cyclopurine-20 -deoxynucleosides as endogenous DNA lesions that cause neurodegeneration in xeroderma pigmentosum, Neuroscience 145 (2007) 1407–1417. [84] P.J. Brooks, The 8,50 -cyclopurine-20 -deoxynucleosides: Candidate neurodegenerative DNA lesions in xeroderma pigmentosum, and unique probes of transcription and nucleotide excision repair, DNA Repair (Amsterdam) 7 (2008) 1168–1179. [85] N. Belmadoui, F. Boussicault, M. Guerra, J.-L. Ravanat, C. Chatgilialoglu, J. Cadet, Radiation-induced formation of purine 50 ,8-cyclonucleosides in isolated and cellular DNA: high stereospecificity and modulating effect of oxygen, Org. Biomol. Chem. 8 (2010) 3211–3219.

14

J. Cadet et al. / Cancer Letters 327 (2012) 5–15

[86] F. Boussicault, P. Kaloudis, C. Caminal, Q.G. Mulazzani, C. Chatgilialoglu, The fate of C50 radicals of purine nucleosides under oxidative conditions, J. Am. Chem. Soc. 130 (2008) 8377–8385. [87] C. Chatgilialoglu, M. D’Angelantonio, G. Kciuk, K. Bobrowski, New insights into the reaction paths of hydroxyl radicals with 20 -deoxyguanosine, Chem. Res. Toxicol. 24 (2011) 2200–2206. [88] R.b. Zhang, L.A. Eriksson, Theoretical study of the tandem cross-linkage lesion in DNA, Chem. Phys. Lett. 417 (2006) 303–308. [89] M. D’Errico, E. Parlanti, M. Teson, B.M. Bernades de Jesus, P. Degan, A. Calcagnile, P. Jaruga, M. Bjøras, M. Crescenzi, A.M. Pedrini, J.M. Egly, G. Zambruno, M. Stefanini, M. Dizdaroglu, E. Dogliotti, New functions of XPC in the protection of human skin cells from oxidative damage, EMBO J. 25 (2006) 4305–4315. [90] M. D’Errico, E. Parlanti, M. Teson, P. Degan, T. Lemma, A. Calcagnile, I. Iavarone, P. Jaruga, M. Ropolo, A.M. Pedrini, D. Orioli, G. Frosina, G. Zambruno, M. Dizdaroglu, M. Stefanini, E. Dogliotti, The role of CSA in the response to oxidative DNA damage in human cells, Oncogene 26 (2007) 4336–4343. [91] J. Wang, B. Yuan, C. Guerrero, R. Bahde, S. Gupta, Y. Wang, Quantification of oxidative DNA lesions in tissues of Long–Evans cinnamon rats by capillary high-performance liquid chromatography–tandem mass spectrometry coupled with stable isotope-dilution method, Anal. Chem. 83 (2011) 2201– 2209. [92] D. Gasparutto, A.G. Bourdat, C. D’Ham, V. Duarte, A. Romieu, J. Cadet, Repair and replication of oxidized DNA bases using modified oligodeoxyribonucleotides, Biochimie 82 (2000) 19–24. [93] P.J. Brooks, D.S. Wise, D.A. Berry, J.V. Komoski, M.J. Smerdon, R.L. Somers, H. Mackie, A.Y. Spoonde, E.J. Ackerman, K. Coleman, R.E. Tarone, J.H. Robbins, The oxidative DNA lesion 8,50 -cyclo-20 -deoxyadenosine is repaired by the nucleotide excision repair pathway and blocks gene expression in mammalian cells, J. Biol. Chem. 275 (2000) 22355–22362. [94] I. Kuraoka, P. Robins, C. Masutani, F. Hanaoka, D. Gasparutto, J. Cadet, R.D. Wood, T. Lindahl, Oxygen free radical damage to DNA. Translesion synthesis by human DNA polymerase g and resistance to exonucleases action at cyclopurine deoxynucleoside residues, J. Biol. Chem. 276 (2001) 49283– 49288. [95] A. Romieu, D. Gasparutto, D. Molko, J. Cadet, Site-specific introduction of (50 S)-50 ,8-cyclo-20 -deoxyadenosine into oligodeoxynucleotides, J. Org. Chem. 63 (1998) 5245–5249. [96] A. Romieu, D. Gasparutto, J. Cadet, Synthesis and characterization of oligonucleotides containing 50 ,8-cyclopurine 20 -deoxyribonucleosides : (50 R)-50 ,8-cyclo-20 -deoxyadenosine, (50 S)-50 ,8-cyclo-20 -deoxyguanosine, and (50 R)-50 ,8-cyclo-20 -deoxyguanosine, Chem. Res. Toxicol. 12 (1999) 412–421. [97] P. Jaruga, Y. Xiao, V. Vartanian, R.S. Lloyd, M. Dizdaroglu, Evidence for the involvement of DNA repair enzyme NEIL1 in nucleotide excision repair of (50 R)- and (50 S)-8,50 -cyclo-20 -deoxyadenosine, Biochemistry 49 (2010) 1053– 1055. [98] V.P. Jasti, R.S. Das, B.A. Hilton, S. Weerasooriya, Y. Zou, A.K. Basu, (50 S)-8,50 cyclo-20 -deoxyguanosine is a strong block to replication, a potent pol Vdependent mutagenic lesion, and it inefficiently repaired in Escherichia coli, Biochemistry 50 (2011) 3862–3865. [99] H. Huang, R.S. Das, A.K. Basu, M.P. Stone, Structure of (50 S)-8,50 -cyclo-20 deoxyguanosine in DNA, J. Am. Chem. Soc. 133 (2011) 20357–20368. [100] B. Yuan, J. Wang, H. Cao, R. Sun, Y. Wang, High-throughput analysis of the mutagenic and cytotoxic properties of DNA lesions by next-generation sequencing, Nucleic Acids Res. 39 (2011) 5945–5954. [101] J.C. Delaney, J.M. Essigmann, Assay for determining lesion bypass efficiency and mutagenicity of site-specific DNA lesions in vivo, Methods Enzymol. 408 (2006) 1–15. [102] C. Janion, Inducible SOS response system of DNA repair and mutagenesis in Escherichia coli, Int. J. Biol. Sci. (2008) 338–344. [103] C. Marietta, P.J. Brooks, Transcriptional bypass of bulky DNA lesions causes new mutant RNA transcripts in human cells, EMBO Rep. 8 (2007) 388–393. [104] C. Marietta, H. Gulam, P.J. Brooks, A single 8,50 -cyclo-20 -deoxyadenosine lesion in a TATA box prevents binding of the TATA binding protein and strongly reduces transcription in vivo, DNA Repair (Amsterdam) 1 (2002) 967–975. [105] J. Abraham, P.J. Brooks, Divergent effects of oxidatively induced modification to the C8 of 20 -deoxyadenosine on transcription factor binding: 8;50 (S)-cyclo-20 -deoxyadenosine inhibits the binding of multiple sequence specific transcription factors, while 8-oxo-20 -deoxyadenosine increases binding of CREB and NF-kappa B to DNA, Environ. Mol. Mutagen. 52 (2011) 287–295. [106] M. Costa, A. Zhitkovich, M. Harris, D. Paustenbach, M. Gargas, DNA-protein cross-links produced by various chemical in cultured lymphoma cells, J. Toxicol. Environ. Health 50 (1997) 433–449. [107] A. Zecevic, E. Hagan, M. Reynolds, G. Poage, T. Johnston, A. Zhitkovich, XPA impacts formation of DNA-protein cross-links induced by chromate, Mutagenesis 25 (2010) 381–388. [108] S. Barker, M. Weinfeld, D. Murray, DNA-protein crosslinks: their induction, repair, and biological consequences, Mutat. Res. 580 (2005) 111–135. [109] S. Barker, M. Weinfeld, J. Zheng, L. Li, D. Murray, Identification of mammalian proteins cross-linked to DNA by ionizing radiation, J. Biol. Chem. 280 (2005) 33826–33838. [110] S. Barker, D. Murray, J. Zheng, L. Li, M. Weinfeld, A method for the isolation of covalent DNA-protein crosslinks suitable for proteomics analysis, Anal. Biochem. 344 (2005) 204–215.

[111] R. Olinski, Z.Z. Nackerdien, M. Dizdaroglu, DNA-protein cross-linking between thymine and tyrosine in chromatin of c-irradiated or H2O2-treated cultured human cells, Arch. Biochem. Biophys. 297 (1992) 139–143. [112] M. Dizdaroglu, E. Gajewski, Structure and mechanism of hydroxyl radicalinduced formation of a DNA- protein cross-link involving thymine and lysine in nucleohistone, Cancer Res. 49 (1989) 3463–3467. [113] M. Dizdaroglu, P. Jaruga, Mechanisms of free radical-induced damage to DNA, Free Radic. Res. 46 (2012) 382–419. [114] H. Kasai, Z. Yamaizumi, M. Berger, J. Cadet, Photosensitized formation of 7,8dihydro-8-oxo-20 -deoxyguanosine (8-hydroxy-20 -deoxyguanosine) in DNA by riboflavin: a non-singlet oxygen mediated reaction, J. Am. Chem. Soc. 114 (1992) 9692–9694. [115] S. Perrier, J. Hau, D. Gasparutto, J. Cadet, A. Favier, J.-L. Ravanat, Characterization of lysine–guanine cross-links upon one-electron oxidation of a guanine-containing oligonucleotide in the presence of a trilysine peptide, J. Am. Chem. Soc. 128 (2006) 5703–5710. [116] K.L. Nguyen, M. Steryo, K. Kurbanyan, K.M. Nowitzki, S.M. Butterfield, S.R. Ward, E.D.A. Stemp, DNA protein cross-linking from oxidation of guanine via the flash-quench technique, J. Am. Chem. Soc. 122 (2000) 3585–3594. [117] A.L. Madison, Z.A. Perez, P. To, T. Maisonet, E.V. Rios, Y. Trejo, C. OchoaPaniagua, A. Reno, E.D.A. Stemp, Dependence of DNA-protein cross-linking via guanine oxidation upon local DNA sequence as studied by restriction endonuclease inhibition, Biochemistry 51 (2012) 362–369. [118] K. Kurbanyan, K.L. Nguyen, P. To, E.V. Rivas, A.M.K. Lueras, C. Kosinski, M. Steryo, A. Gonzalez, D.A. Mah, E.D.A. Stemp, DNA-protein cross-linking via guanine oxidation. Dependence upon protein and photosensitizer, Biochemistry 42 (2003) 10269–10281. [119] D. Angelov, M. Charra, C.W. Müller, J. Cadet, S. Dimitrov, Solution study of the NF-kappa B p50-DNA complex by UV laser protein-DNA cross-linking, Photochem. Photobiol. 77 (2003) 592–596. [120] P. O’Donovan, C.M. Perrett, X. Zhang, B. Montaner, Y.Z. Xu, C.A. Harwood, J.M. McGregor, S.L. Walker, F. Hanaoka, P. Karran, Azathiopurine and UVA light generate mutagenic oxidative DNA damage, Science 309 (2005) 1871–1874. [121] X. Zhang, G. Jeffs, X. Ren, P. O’Donovan, B. Montaner, C.M. Perret, P. Karran, Y.Z. Xu, Novel lesions generated by the interaction between therapeutic thiopurines and UVA light, DNA Repair (Amsterdam) 6 (2007) 344–354. [122] Q. Gueranger, A. Kia, D. Frith, P. Karran, Crosslinking of DNA repair and replication proteins to DNA in cells treated with 6-thioguanine and UVA, Nucleic Acids Res. 39 (2011) 5057–5060. [123] H. Ide, M.I. Shoulkamy, T. Nakano, M. Miyamoto-Matsubara, A.M.H. Salem, Repair and biochemical effects of DNA-protein crosslinks, Mutat. Res. 711 (2011) 113–122. [124] J.T. Reardon, Y. Cheng, A. Sancar, Repair of DNA-protein cross-links in mammalian cells, Cell Cycle 5 (2006) 1366–1370. [125] M. Hashimoto, M.M. Greenberg, Y.W. Kow, J.T. Hwang, R.P. Cunningham, The 2-deoxyribonolactone lesion produced in DNA by neocarzinostatin and other damaging agents forms cross-links with the base-excision repair enzyme endonuclease III, J. Am. Chem. Soc. 123 (2001) 3161–3162. [126] M.S. DeMott, E. Beyret, D. Wong, B.C. Bales, J.T. Hwang, M.M. Greenberg, B. Demple, Covalent trapping of human DNA polymerase b by the oxidative DNA lesion 2-deoxyribonolactone, J. Biol. Chem. 277 (2002) 7637–7640. [127] J.S. Sung, B. Demple, Analysis of base excision DNA repair of the oxidative lesion 2-deoxyribonolactone and the formation of DNA-protein cross-links, Methods Enzymol. 408 (2006) 48–64. [128] T. Nakano, H. Terato, K. Asagoshi, A. Masaoka, M. Mukuta, Y. Ohyama, T. Suzuki, K. Makino, H. Ide, DNA-protein cross-link formation mediated by oxanine, J. Biol. Chem. 278 (2003) 25264–25272. [129] J.T. Sczepanski, R.S. Wong, J.N. McKnight, G.D. Bowman, M.M. Greenberg, Rapid DNA-protein cross-linking and strand scission by an abasic site in a nucleosome core particle, Proc. Natl. Acad. Sci. USA 107 (2010) 22475–22480. [130] S.S. David, V.L. O’Shea, S. Kundu, Base-excision repair of oxidative DNA damage, Nature 447 (2007) 941–950. [131] M.E. Dextraze, T. Gantchev, S. Girouard, D. Hunting, DNA interstrand crosslinks induced by ionizing radiation: an unsung lesion, Mutat. Res. 704 (2010) 101–107. [132] S. Cuesta-López, H. Menoni, D. Angelov, M. Peyrard, Guanine radical chemistry reveals the effect of thermal fluctuations in gene promoter regions, Nucleic Acids Res. 39 (2011) 5276–5283. [133] N.R. Attard, P. Karran, UVA photosensitization of thiopurines and skin cancer in organ transplant recipients, Photochem. Photobiol. Sci. 11 (2011) 62–68. [134] R. Brem, I. Daehn, P. Karran., Efficient DNA interstrand crosslinking by 6thioguanine and UVA radiation, DNA Repair (Amsterdam) 10 (2011) 869–876. [135] P. Regulus, S. Spessotto, M. Gateau, J. Cadet, A. Favier, J.-L. Ravanat, Detection of new radiation-induced DNA lesions by liquid chromatography coupled to tandem mass spectrometry, Rapid Commun. Mass Spectrom. 18 (2004) 2223–2228. [136] P. Regulus, B. Duroux, P.A. Bayle, A. Favier, J. Cadet, J.-L. Ravanat, Oxidation of the sugar moiety of DNA by ionizing radiation or bleomycin could induce the formation of a cluster DNA lesion, Proc. Natl. Acad. Sci. USA 104 (2006) 14032–14037. [137] J.T. Sczepanski, A.C. Jacobs, M.M. Greenberg, Self-promoted DNA interstrand cross-link formation by an abasic site, J. Am. Chem. Soc. 130 (2008) 9646– 9647. [138] J.T. Sczepanski, A.C. Jacobs, A. Majumdar, M.M. Greenberg, Scope and mechanism of interstrand cross-link formation by the C40 -oxidized abasic site, J. Am. Chem. Soc. 131 (2009) 11132–11139.

J. Cadet et al. / Cancer Letters 327 (2012) 5–15 [139] J.T. Sczepanski, C.N. Hiemstra, M.M. Greenberg, Probing DNA interstrand cross-link formation by an oxidized abasic site using nonnative nucleotides, Biorg. Med. Chem. 19 (2011) 5788–5793. [140] I.H. Goldberg, Mechanism of neocarzinostatin action: role of DNA microstructure in determination of chemistry of bistranded oxidative damage, Acc. Chem. Res. 24 (1991) 191–198. [141] L. Guan, M.M. Greenberg, DNA interstrand cross-link formation by the 1,4dioxobutane abasic lesion, J. Am. Chem. Soc. 131 (2009) 15225–15231. [142] R.D. Stewart, V.K. Yu, A.G. Georgakilas, C. Koumenis, J.H. Park, D.J. Carlson, Effects of radiation quality and oxygen on clustered DNA lesions and cell death, Radiat. Res. 176 (2011) 587–602. [143] G. Iliakis, The role of DNA double strand breaks in ionizing radiation-induced killing of eukaryotic cells, Bioessays 13 (1991) 185–195. [144] S.W. Botchway, P. Reynolds, A.W. Parker, P. O’Neill, Mutat. Res. Rev. 704 (2010) 38–44. [145] L.F. Povirk, DNA damage and mutagenesis by radiomimetic DNA-cleaving agents: bleomycin, neocarzinostatin and other enediynes, Mutat. Res. 355 (1996) 71–89. [146] P. Heisig, Type II topoisomerase – inhibitors, repair mechanisms and mutations, Mutagenesis 24 (2009) 465–469. [147] C. Allen, A.K. Ashley, R. Hromas, J.A. Nickoloff, More forks on the road to replication stress recovery, J. Mol. Cell Biol. 3 (2011) 4–12. [148] J.V. Harper, J.A. Anderson, P. O’Neill, Radiation induced DNA DSBs: contribution from stalled replication forks?, DNA Repair (Amsterdam) 9 (2010) 907–913 [149] A.G. Georgakilas, Processing of DNA damage clusters in human cells: current status of knowledge, Mol. Biosyst. 4 (2008) 30–35. [150] S.G. Kozmin, Y. Sedletska, A. Reynaud-Angelin, D. Gasparutto, E. Sage, The formation of double-strand breaks at multiply damaged sites is driven by the kinetics of excision/incision at base damage in eukaryotic cells, Nucleic Acids Res. 37 (2009) 1767–1777. [151] L.J. Eccles, M.E. Lomax, P. O’Neill, Hierarchy of lesions processing governs the repair, double-strand break formation and mutability of three-lesion clustered DNA damage, Nucleic Acids Res. 38 (2010) 1123–1134. [152] I.R. Radford, DNA lesion complexity and induction of apoptosis by ionizing radiation, Int. J. Radiat. Biol. 78 (2002) 457–466. [153] P.L. Olive, J.P. Banáth, The comet assay: a method to measure DNA damage in individual cells, Nat. Protocol. 1 (2006) 23–29. [154] L.-J. Mah, C. Orlowski, K. Ververis, R. Vasireddy, A. El-Osta, T.C. Karagiannis, Evaluation of the efficacy of radiation-modifyning compounds using cH2AX as a molecular marker of DNA double-strand breaks, Genome Integr. 2 (2011) 3. [155] A. Ivashkevich, C.E. Redon, A.J. Nakamura, R.F. Martin, O.A Martin, Use of cH2AX assay to monitor DNA damage and repair in translational cancer research, Cancer Lett., (2012), http://dx.doi.org/10.1016/J.canlet.2011.12.05. [156] J. Stap, P.M. Krawczyk, C.H. Van Oven, G.W. Barendsen, J. Essers, R. Kanaar, J.A. Aten, Induction of linear tracks of DNA double-strand breaks by a-particle irradiation of cells, Nat. Methods 5 (2008) 261–266. [157] F. Bouquet, C. Muller, B. Salles, The loss of cH2AX signal is a marker of DNA double strand breaks repair only at low levels of DNA damage, Cell Cycle 5 (2006) 1116–1122. [158] P.L. Olive, Endogenous DNA breaks: gamma H2AX and the role of telomers, Aging (Albany NY) 1 (2009) 154–156. [159] Y. Revet, L. Feeney, S. Bruguera, W. Wilson, T.K. Dong, D.H. Oh, D. Dankort, J.E. Cleaver, Functional relevance of the histone cH2AX in the response to DNA damaging agents, Proc. Natl. Acad. Sci. USA 108 (2011) 8663–8667. [160] J.E. Cleaver, L. Feeny, I. Revet, Phosphorylated H2AX is not an unambiguous marker for DNA double-strand breaks, Cell Cycle 10 (2011) 3223–3224. [161] K. Rothkamm, M. Lobrich, Evidence for a lack of DNA double-strand break repair in human cells exposed to very lox X-ray doses, Proc. Natl. Acad. Sci. USA 100 (2003) 5057–5062. [162] M.R. Lieber, The mechanism of double-strand DNA break repair by the nonhomologous DNA end joining pathway, Annu. Rev. Biochem. 79 (2010) 181–211. [163] E. Mladenov, G. Iliakis, Induction and repair of DNA double strand breaks: the increasing spectrum of non-homologous end joining pathways, Mutat. Res. 711 (2011) 61–72.

15

[164] P.A. Jeggo, V. Geuting, M. Löbrich, The role of homologous recombination in radiation-induced double strand break repair, Radiother. Oncol. 101 (2011) 7–12. [165] Y. Zhang, L.H. Rohde, H. Wu, Involvement of nucleotide excision repair and mismatch repair mechanisms in double strand break repair, Curr. Genomics 10 (2009) 250–258. [166] K. Datta, S. Purkayastha, R.D. Neumann, E. Pastwa, T.A. Winters, Base damage immediately upstream from double-strand break ends is a more severe impediment to nonhomologous end joining than blocked 30 -termini, Radiat. Res. 175 (2011) 97–112. [167] D. Tsao, P. Kalogerinis, I. Tabrizi, M. Dingfelder, R.D. Stewart, A.G. Georgakilas, Induction and processing of oxidative clustered DNA lesions in 56Fe-ionirradiated human monocytes, Radiat. Res. 168 (2007) 87–97. [168] J.-P. Pouget, S. Frelon, J.-L. Ravanat, I. Testard, F. Odin, J. Cadet, Formation of modified DNA bases in cells exposed either to gamma radiation or to highLET particles, Radiat. Res. 157 (2002) 589–595. [169] T. Douki, J.-L. Ravanat, J.-P. Pouget, I. Testard, J. Cadet, Minor contribution of direct ionization to DNA base damage induced by heavy ions, Int. J. Radiat. Biol. 82 (2006) 119–127. [170] H. Nikjoo, P. O’Neill, W.E. Wilson, D.T. Goodhead, Computational approach for determining the spectrum of DNA damage induced by ionizing radiation, Radiat. Res. 156 (2001) 577–583. [171] V.A. Semenenko, R.D. Stewart, A fast Monte Carlo algorithm to simulate the spectrum of DNA damages formed by ionizing radiation, Radiat. Res. 161 (2004) 451–457. [172] A.G. Georgakilas, P.V. Bennett, D.M. Wilson III, B.M. Sutherland, Processing of bistranded abasic DNA clusters in c-irradiated human hematopoietic cells, Nucleic Acids Res. 32 (2004) 5609–5620. [173] P.V. Bennett, N.S. Cintron, L. Gros, J. Laval, B.M. Sutherland, Are endogenous clustered DNA damages induced in human cells?, Free Radic Biol. Med. 37 (2004) 488–499. [174] P. Bennett, A.A. Ishchenko, J. Laval, B. Paap, B.M. Sutherland, Endogenous DNA damage clusters in human meatopoietic stem and progenitor cells, Free Radic. Biol. Med. 45 (2008) 1352–1359. [175] P.V. Bennett, N.L. Cuomo, S. Paul, S.T. Tafrov, B.M. Sutherland, Endogenous DNA damage clusters in human skin, 3-D model, and cultured skin cells, Free Radic. Biol. Med. 39 (2005) 832–839. [176] J.M. Song, J.R. Milligan, B.M. Sutherland, Bistranded oxidized purine damage clusters: induced in DNA by long-wavelength ultraviolet (290–400 nm) radiation?, Biochemistry 41 (2002) 8683–8688 [177] J.L. Rizzo, J. Dunn, A. Rees, T.M. Rünger, No formation of DNA double-strand breaks and no activation of recombination repair with UVA, J. Invest. Dermatol. 131 (2010) 1139–1148. [178] J. Cadet, T. Douki, Oxidatively generated damage to DNA radiation in cells and human skin, J. Invest. Dermatol. 131 (2011) 1005–1007. [179] J.E. Cleaver, CH2AX: biomarker of damage or functional participant in DNA repair ‘‘all that glitters is not gold’’, Photochem. Photobiol. 87 (2011) 1230–1239. [180] G. Eot-Houllier, S. Eon-Marchais, D. Gasparutto, E. Sage, Processing of a complex multiply damaged DNA site by human cell extracts and purified repair proteins, Nucleic Acids Res. 33 (2005) 260–271. [181] G. Eot-Houllier, M. Gonera, D. Gasparutto, C. Giustranti, E. Sage, Interplay between DNA N-glycosylases/AP lyases at multiply damaged sites and biological consequences, Nucleic Acids Res. 35 (2007) 3355–3366. [182] H. Budworth, G. Matthemann, P. O’Neill, G.L. Dianov, Repair of tandem base lesions in DNA by human cell extracts generates persisting single-strand breaks, J. Mol. Biol. 351 (2005) 1020–1029. [183] S. Bellon, N. Shikazono, S. Cunniffe, M. Loma, P. O’Neill, Processing of thymine glycol in a clustred DNA damage site: mutagenic or cytotoxic, Nucleic Acids Res. 37 (2009) 4430–4440. [184] M. Noguchi, A. Urushibara, A. Yokoya, P. O’Neill, N. Shikazono, The mutagenic potential 8-oxoG/single strand break-containing clusters of their relative positions, Mutat. Res. 732 (2012) 34–42. [185] L.J. Eccles, P. O’Neill, M.E. Lomax, Delayed repair of radiation-induced clustered DNA damage: friend of foe?, Mutat Res. 711 (2011) 134–141. [186] A. Asaithamby, B. Hu, D.J. Chen, Unrepaired clustered DNA lesions induce chromosome breakage in human cells, Proc. Natl. Acad. Sci. USA 108 (2011) 8293–8298.