When DNA repair goes wrong: BER-generated DNA-protein crosslinks to oxidative lesions

Free radicals generate an array of DNA lesions affecting all parts of the molecule. The damage to deoxyribose receives less attention than base damage, even though the former accounts for ∼20% of the total. Oxidative deoxyribose fragments (e.g., 3′-phosphoglycolate esters) are removed by the Ape1 AP endonuclease and other enzymes in mammalian cells to enable DNA repair synthesis. Oxidized abasic sites are initially incised by Ape1, thus recruiting these lesions into base excision repair (BER) pathways. Lesions such as 2-deoxypentos-4-ulose can be removed by conventional (single-nucleotide) BER, which proceeds through a covalent Schiff base intermediate with DNA polymerase β (Polβ) that is resolved by hydrolysis. In contrast, the lesion 2-deoxyribonolactone (dL) must be processed by multinucleotide (“long-patch”) BER: attempted repair via the single-nucleotide pathway leads to a dead-end, covalent complex with Polβ cross- linked to the DNA by an amide bond. We recently detected these stable DNA-protein crosslinks (DPC) between Polβ and dL in intact cells. The features of the DPC formation in vivo are exactly in keeping with the mechanistic properties seen in vitro: Polβ-DPC are formed by oxidative agents in line with their ability to form the dL lesion; they are not formed by non-oxidative agents; DPC formation absolutely requires the active-site lysine-72 that attacks the 5′-deoxyribose; and DPC formation depends on Ape1 to incise the dL lesion first. The Polβ-DPC are rapidly processed in vivo, the signal disappearing with a half-life of 15–30 min in both mouse and human cells. This removal is blocked by inhibiting the proteasome, which leads to the accumulation of ubiquitin associated with the Polβ-DPC. While other proteins (e.g., topoisomerases) also form DPC under these conditions, 60–70% of the trapped ubiquitin depends on Polβ. The mechanism of ubiquitin targeting to Polβ-DPC, the subsequent processing of the expected 5′-peptidyl-dL, and the biological consequences of unrepaired DPC are important to assess. Many other lyase enzymes that attack dL can also be trapped in DPC, so the processing mechanisms may apply quite broadly.     

Role of RNase H enzymes in maintaining genome stability in Escherichia coli expressing a steric-gate mutant of pol VICE391

pol V_ICE391 (RumAʹ₂B) is a low-fidelity polymerase that promotes considerably higher levels of spontaneous “SOS-induced” mutagenesis than the related E. coli pol V (UmuDʹ₂C). The molecular basis for the enhanced mutagenesis was previously unknown. Using single molecule fluorescence microscopy to visualize pol V enzymes, we discovered that the elevated levels of mutagenesis are likely due, in part, to prolonged binding of RumB to genomic DNA leading to increased levels of DNA synthesis compared to UmuC.We have generated a steric gate pol V_ICE391 variant (pol V_ICE391_Y13A) that readily misincorporates ribonucleotides into the E. coli genome and have used the enzyme to investigate the molecular mechanisms of Ribonucleotide Excision Repair (RER) under conditions of increased ribonucleotide-induced stress. To do so, we compared the extent of spontaneous mutagenesis promoted by pol V and pol V_ICE391 to that of their respective steric gate variants. Levels of mutagenesis promoted by the steric gate variants that are lower than that of the wild-type enzyme are indicative of active RER that removes misincorporated ribonucleotides, but also misincorporated deoxyribonucleotides from the genome.Using such an approach, we confirmed that RNase HII plays a pivotal role in RER. In the absence of RNase HII, Nucleotide Excision Repair (NER) proteins help remove misincorporated ribonucleotides. However, significant RER occurs in the absence of RNase HII and NER. Most of the RNase HII and NER-independent RER occurs on the lagging strand during genome duplication. We suggest that this is most likely due to efficient RNase HI-dependent RER which recognizes the polyribonucleotide tracts generated by pol V_ICE391_Y13A. These activities are critical for the maintenance of genomic integrity when RNase HII is overwhelmed, or inactivated, as ΔrnhB or ΔrnhB ΔuvrA strains expressing pol V_ICE391_Y13A exhibit genome and plasmid instability in the absence of RNase HI.     

Dideoxynucleoside triphosphate-sensitive DNA polymerase from rice is involved in base excision repair and immunologically similar to mammalian DNA pol beta

A single polypeptide with ddNTP-sensitive DNA polymerase activity was purified to near homogeneity from the shoot tips of rice seedlings and analysis of the preparations by SDS-PAGE followed by silver staining showed a polypeptide of 67 kDa size. The DNA polymerase activity was found to be inhibitory by ddNTP in both in vitro DNA polymerase activity assay and activity gel analysis. Aphidicolin, an inhibitor of other types of DNA polymerases, had no effect on plant enzyme. The 67 kDa rice DNA polymerase was found to be recognized by the polyclonal antibody (purified IgG) made against rat DNA polymerase beta (pol beta) both in solution and also on Western blot. The recognition was found to be very specific as the activity of Klenow enzyme was unaffected by the antibody. The ability of rice nuclear extract to correct G:U mismatch of oligo-duplex was observed when oligo-duplex with 32P-labeled lower strand containing U (at 22nd position) was used as substrate. Differential appearance of bands at 21-mer, 22-mer, and 51-mer position in presence of dCTP was visible only with G:U mismatch oligo-duplex, but not with G:C oligo-duplex. While ddCTP or polyclonal antibody against rat-DNA pol beta inhibits base excision repair (BER), aphidicolin had no effect. These results for the first time clearly demonstrate the ability of rice nuclear extract to run BER and the involvement of ddNTP-sensitive pol beta type DNA polymerase. Immunological similarity of the ddNTP-sensitive DNA polymerase beta of rice and rat and its involvement in BER revealed the conservation of structure and function of ddNTP-sensitive DNA pol beta in plant and animal.     

Coordination of MYH DNA glycosylase and APE1 endonuclease activities via physical interactions

MutY homologue (MYH) is a DNA glycosylase which excises adenine paired with the oxidative lesion 7,8-dihydro-8-oxoguanine (8-oxoG, or G^o) during base excision repair (BER). Base excision by MYH results in an apurinic/apyrimidinic (AP) site in the DNA where the DNA sugar–phosphate backbone remains intact. A key feature of MYH activity is its physical interaction and coordination with AP endonuclease I (APE1), which subsequently nicks DNA 5′ to the AP site. Because AP sites are mutagenic and cytotoxic, they must be processed by APE1 immediately after the action of MYH glycosylase. Our recent reports show that the interdomain connector (IDC) of human MYH (hMYH) maintains interactions with hAPE1 and the human checkpoint clamp Rad9–Rad1–Hus1 (9–1–1) complex. In this study, we used NMR chemical shift perturbation experiments to determine hMYH-binding site on hAPE1. Chemical shift perturbations indicate that the hMYH IDC peptide binds to the DNA-binding site of hAPE1 and an additional site which is distal to the APE1 DNA-binding interface. In these two binding sites, N212 and Q137 of hAPE1 are key mediators of the MYH/APE1 interaction. Intriguingly, despite the fact that hHus1 and hAPE1 both interact with the MYH IDC, hHus1 does not compete with hAPE1 for binding to hMYH. Rather, hHus1 stabilizes the hMYH/hAPE1 complex both in vitro and in cells. This is consistent with a common theme in BER, namely that the assembly of protein–DNA complexes enhances repair by efficiently coordinating multiple enzymatic steps while simultaneously minimizing the release of harmful repair intermediates.     

Human DNA polymerases catalyze lesion bypass across benzo[a]pyrene-derived DNA adduct clustered with an abasic site

The combined action of oxidative stress and genotoxic polycyclic aromatic hydrocarbons derivatives can lead to cluster-type DNA damage that includes both a modified nucleotide and a bulky lesion. As an example, we investigated the possibility of repair of an AP site located opposite a minor groove-positioned (+)-trans-BPDE-dG or a base-displaced intercalated (+)-cis-BPDE-dG adduct (BP lesion) by a BER system. Oligonucleotides with single uracil residue in the certain position were annealed with complementary oligonucleotides bearing either a cis- or trans-BP adduct. Digestion with uracil DNA glycosylase was utilized to generate an AP site which was then hydrolyzed by APE1, and the resulting gap was processed by X-family DNA polymerases β (Polβ) and λ (Polλ), or Y-family polymerase ι (Polι). By varying reaction conditions, namely, Mg²⁺/Mn²⁺ replacement/combination and ionic strength decrease, we found that under certain conditions both Polβ and Polι can catalyze lesion bypass across both cis- and trans-BP adducts in the presence of physiological dNTP concentrations. Polβ and Polι catalyze gap filling trans-lesion synthesis in an error prone manner. By contrast, Polλ selectively introduced the correct dCTP opposite the modified dG in the case of cis-BP-dG adduct only, and did not bypass the stereoisomeric trans-adduct under any of the conditions examined. The results suggest that Polλ is a specialized polymerase that can process these kinds of lesions.     

DNA polymerase X of African swine fever virus: insertion fidelity on gapped DNA substrates and AP lyase activity support a role in base excision repair of viral DNA

DNA polymerase X (pol X) from African swine fever virus (ASFV) is the smallest naturally ocurring DNA-directed DNA polymerase (174 amino acid residues) described so far. Previous biochemical analysis has shown that ASFV pol X is a highly distributive, monomeric enzyme, lacking a proofreading 3'-5' exonuclease. Also, ASFV pol X binds intermediates of the single-nucleotide base excision repair (BER) process, and is able to efficiently repair single-nucleotide gapped DNA. In this work, we perform an extensive kinetic analysis of single correct and incorrect nucleotide insertions by ASFV pol X using different DNA substrates: (i) a primer/template DNA; (ii) a 1nt gapped DNA; (iii) a 5'-phosphorylated 1nt gapped DNA. The results obtained indicate that ASFV pol X exhibits a general preference for insertion of purine deoxynucleotides, especially dGTP opposite template C. Moreover, ASFV pol X shows higher catalytic efficiencies when filling in gapped substrates, which are increased when a phosphate group is present at the 5'-margin of the gap. Interestingly, ASFV pol X misinserts nucleotides with frequencies from 10(-4) to 10(-5), and the insertion fidelity varies depending on the substrate, being more faithful on a phosphorylated 1nt gapped substrate. We have analyzed the capacity of ASFV pol X to act on intermediates of BER repair. Although no lyase activity could be detected on preincised 5'-deoxyribose phosphate termini, ASFV pol X has lyase activity on unincised abasic sites. Altogether, the results support a role for ASFV pol X in reparative BER of damaged viral DNA during ASFV infection.     

AP endonuclease 1 prevents the extension of a T/G mismatch by DNA polymerase β to prevent mutations in CpGs during base excision repair

Dynamics of DNA methylation and demethylation at CpG clusters are involved in gene regulation. CpG clusters have been identified as hot spots of mutagenesis because of their susceptibility to oxidative DNA damage. Damaged Cs and Gs at CpGs can disrupt a normal DNA methylation pattern through modulation of DNA methylation and demethylation, leading to mutations and deregulation of gene expression. DNA base excision repair (BER) plays a dual role of repairing oxidative DNA damage and mediating an active DNA demethylation pathway on CpG clusters through removal of a T/G mismatch resulting from deamination of a 5mC adjacent to a guanine that can be simultaneously damaged by oxidative stress. However, it remains unknown how BER processes clustered lesions in CpGs and what are the consequences from the repair of these lesions. In this study, we examined BER of an abasic lesion next to a DNA demethylation intermediate, the T/G mismatch in a CpG dinucleotide, and its effect on the integrity of CpGs. Surprisingly, we found that the abasic lesion completely abolished the activity of thymine DNA glycosylase (TDG) for removing the mismatched T. However, we found that APE1 could still efficiently incise the abasic lesion leaving a 3-terminus mismatched T, which was subsequently extended by pol β. This in turn resulted in a C to T transition mutation. Interestingly, we also found that APE1 3′–5′ exonuclease activity efficiently removed the mismatched T, thereby preventing pol β extension of the mismatched nucleotide and the resulting mutation. Our results demonstrate a crucial role of APE1 3′–5′ exonuclease activity in combating mutations in CpG clusters caused by an intermediate of DNA demethylation during BER.     

In vitro eradication of abasic site-mediated DNA–peptide/protein cross-links by Escherichia coli long-patch base excision repair

Apurinic/apyrimidinic (AP or abasic) sites are among the most abundant DNA lesions. Numerous proteins within different organisms ranging from bacteria to human have been demonstrated to react with AP sites to form covalent Schiff base DNA–protein cross-links (DPCs). These DPCs are unstable due to their spontaneous hydrolysis, but the half-lives of these cross-links can be as long as several hours. Such long-lived DPCs are extremely toxic due to their large sizes, which physically block DNA replication. Therefore, these adducts must be promptly eradicated to maintain genome integrity. Herein, we used in vitro reconstitution experiments with chemically synthesized, stable, and site-specific Schiff base AP-peptide/protein cross-link analogs to demonstrate for the first time that this type of DPC can be repaired by Escherichia coli (E. coli) long-patch base excision repair. We demonstrated that the repair process requires a minimum of three enzymes and five consecutive steps, including: (1) 5′-DNA strand incision of the DPC by endonuclease IV; (2 to 4) strand-displacement DNA synthesis, removal of the 5′-deoxyribose phosphate-peptide/protein adduct-containing flap, and gap-filling DNA synthesis by DNA polymerase I; and (5) strand ligation by a ligase. We further demonstrated that endonuclease IV plays a major role in incising an AP-peptide cross-link within E. coli cell extracts. We also report that eradicating model AP-protein (11.2–36.1 kDa) DPCs is less efficient than that of an AP-peptide_10mer cross-link, supporting the emerging model that proteolysis is likely required for efficient DPC repair.     

A domain of the Klenow fragment of Escherichia coli DNA polymerase I has polymerase but no exonuclease activity

The Klenow fragment of DNA polymerase I from Escherichia coli has two enzymatic activities: DNA polymerase and 3'-5' exonuclease. The crystal structure showed that the fragment is folded into two distinct domains. The smaller domain has a binding site for deoxynucleoside monophosphate and a divalent metal ion that is thought to identify the 3'-5' exonuclease active site. The larger C-terminal domain contains a deep cleft that is believed to bind duplex DNA. Several lines of evidence suggested that the large domain also contains the polymerase active site. To test this hypothesis, we have cloned the DNA coding for the large domain into an expression system and purified the protein product. We find that the C-terminal domain has polymerase activity (albeit at a lower specific activity than the native Klenow fragment) but no measurable 3'-5' exonuclease activity. These data are consistent with the hypothesis that each of the three enzymatic activities of DNA polymerase I from E. coli resides on a separate protein structural domain.     

Sources of thymidine and analogs fueling futile damage-repair cycles and ss-gap accumulation during thymine starvation in Escherichia coli

Thymine deprivation in thyA mutant E. coli causes thymineless death (TLD) and is the mode of action of popular antibacterial and anticancer drugs, yet the mechanisms of TLD are still unclear. TLD comprises three defined phases: resistance, rapid exponential death (RED) and survival, with the nature of the resistance phase and of the transition to the RED phase holding key to TLD pathology. We propose that a limited source of endogenous thymine maintains replication forks through the resistance phase. When this source ends, forks undergo futile break-repair cycle during the RED phase, eventually rendering the chromosome non-functional. Two obvious sources of the endogenous thymine are degradation of broken chromosomal DNA and recruitment of thymine from stable RNA. However, mutants that cannot degrade broken chromosomal DNA or lack ribo-thymine, instead of shortening the resistance phase, deepen the RED phase, meaning that only a small fraction of T-starved cells tap into these sources. Interestingly, the substantial chromosomal DNA accumulation during the resistance phase is negated during the RED phase, suggesting futile cycle of incorporation and excision of wrong nucleotides. We tested incorporation of dU or rU, finding some evidence for both, but DNA-dU incorporation accelerates TLD only when intracellular [dUTP] is increased by the dut mutation. In the dut ung mutant, with increased DNA-dU incorporation and no DNA-dU excision, replication is in fact rescued even without dT, but TLD still occurs, suggesting different mechanisms. Finally, we found that continuous DNA synthesis during thymine starvation makes chromosomal DNA increasingly single-stranded, and even the dut ung defect does not completely block this ss-gap accumulation. We propose that instability of single-strand gaps underlies the pathology of thymine starvation.     

Base excision repair activities required for yeast to attain a full chronological life span

The chronological life span of yeast, the survival of stationary (G0) cells over time, provides a model for investigating certain of the factors that may influence the aging of non-dividing cells and tissues in higher organisms. This study measured the effects of defined defects in the base excision repair (BER) system for DNA repair on this life span. Stationary yeast survives longer when it is pre-grown on respiratory, as compared to fermentative (glucose), media. It is also less susceptible to viability loss as the result of defects in DNA glycosylase/AP lyases (Ogg1p, Ntg1p, Ntg2p), apurinic/apyrimidinic (AP) endonucleases (Apn1p, Apn2p) and monofunctional DNA glycosylase (Mag1p). Whereas single BER glycosylase/AP lyase defects exerted little influence over such optimized G0 survival, this survival was severely shortened with the loss of two or more such enzymes. Equally, the apn1delta and apn2delta single gene deletes survived as well as the wild type, whereas a apn1delta apn2delta double mutant totally lacking in any AP endonuclease activity survived poorly. Both this shortened G0 survival and the enhanced mutagenicity of apn1delta apn2delta cells were however rescued by the over-expression of either Apn1p or Apn2p. The results highlight the vital importance of BER in the prevention of mutation accumulation and the attainment of the full yeast chronological life span. They also reveal an appreciable overlap in the G0 maintenance functions of the different BER DNA glycosylases and AP endonucleases.     

A role for p53 in base excision repair

Wild-type p53 protein can markedly stimulate base excision repair (BER) in vitro, either reconstituted with purified components or in extracts of cells. In contrast, p53 with missense mutations either at hot-spots in the core domain or within the N-terminal transactivation domain is defective in this function. Stimulation of BER by p53 is correlated with its ability to interact directly both with the AP endonuclease (APE) and with DNA polymerase beta (pol beta). Furthermore, p53 stabilizes the interaction between DNA pol beta and abasic DNA. Evidence that this function of p53 is physiologically relevant is supported by the facts that BER activity in human and murine cell extracts closely parallels their levels of endogenous p53, and that BER activity is much reduced in cell extracts immunodepleted of p53. These data suggest a novel role for p53 in DNA repair, which could contribute to its function as a key tumor suppressor.     

Modulation of DNA glycosylase activities in mesenchymal stem cells

Adipose-tissue derived mesenchymal stem cells (AT-MSCs) are a promising tool for use in cell-based therapies. However, in vitro expansion is required to obtain clinically relevant cell numbers, and this might increase the chance of genomic instability. DNA repair is crucial for maintaining DNA integrity. Here we have compared the initial step of base excision repair in uncultured and cultured AT-MSCs by analysis of base removal activities and expression levels of relevant DNA glycosylases. Uracil, 5-hydroxyuracil and ethenoadenine removal activities were upregulated in cultured cells compared to uncultured cells. In contrast, both the 8-oxo-7,8-dihydroguanine (8-oxoG) removal activity and the concentration of 8-oxoG bases in the DNA were reduced in the cultured cells. Gene expression analysis showed no substantial changes in mRNA expression. The glycosylase activities remained stable through at least 12 passages, suggesting that DNA repair is proficient through the period required for in vitro expansion of AT-MSCs to clinically relevant numbers.     

The excision of AP sites by the 3'-5' exonuclease activity of the Klenow fragment of Escherichia coli DNA polymerase I

The 3' AP endonucleases (class I) are said to hydrolyze the phosphodiester bond 3' to AP sites yielding 3'-OH and 5'-phosphate ends; on the other hand, the resulting 3' terminal AP site is not removed by the 3'-5' exonuclease activity of the Klenow fragment [1]. We show that AP sites in DNA are easily removed by the 3'-5' exonuclease activity of the Klenow fragment and that they are excised as deoxyribose-5-phosphate. It is suggested that the 3' AP endonucleases are perhaps not the hydrolases they are supposed to be.     

Interactions of replication versus repair DNA substrates with the Pol I DNA polymerases from Escherichia coli and Thermus aquaticus

Different DNA polymerases partition differently between replication and repair pathways. In this study we examine if two Pol I family polymerases from evolutionarily distant organisms also differ in their preferences for replication versus repair substrates. The DNA binding preferences of Klenow and Klentaq DNA polymerases, from Escherichia coli and Thermus aquaticus respectively, have been studied using a fluorescence competition binding assay. Klenow polymerase binds primed-template DNA (the replication substrate) with up to 50× higher affinity than it binds to nicked DNA, DNA with a 2 base single-stranded gap, blunt-ended DNA, or to a DNA end with a 3′ overhang. In contrast, Klentaq binds all of these DNAs almost identically, indicating that Klenow has a stronger ability to discriminate between replication and repair substrates than Klentaq. In contrast, both polymerases bind mismatched primed-template and blunt-ended DNA tighter than they bind matched primed-template DNA, suggesting that these two proteins may share a similar mechanism to identify mismatched DNA, despite the fact that Klentaq has no proofreading ability. In addition, the presence or absence of 5′- or 3′-phosphates has slightly different effects on DNA binding by the two polymerases, but again reinforce Klenow's more effective substrate discrimination capability.     

Transient protein–protein complexes in base excision repair

Abstract

Transient protein–protein complexes are of great importance for organizing multiple enzymatic reactions into productive reaction pathways. Base excision repair (BER), a process of critical importance for maintaining genome stability against a plethora of DNA-damaging factors, involves several enzymes, including DNA glycosylases, AP endonucleases, DNA polymerases, DNA ligases and accessory proteins acting sequentially on the same damaged site in DNA. Rather than being assembled into one stable multisubunit complex, these enzymes pass the repair intermediates between them in a highly coordinated manner. In this review, we discuss the nature and the role of transient complexes arising during BER as deduced from structural and kinetic data. Almost all of the transient complexes are DNA-mediated, although some may also exist in solution and strengthen under specific conditions. The best-studied example, the interactions between DNA glycosylases and AP endonucleases, is discussed in more detail to provide a framework for distinguishing between stable and transient complexes based on the kinetic data.

Communicated by Ramaswamy H. Sarma     

A comprehensive approach to determining BER capacities and their change with aging in Drosophila melanogaster mitochondria by oligonucleotide microarray

DNA repair mechanisms are key components for the maintenance of the essential mitochondrial genome. Among them, base excision repair (BER) processes, dedicated in part to oxidative DNA damage, are individually well known in mitochondria. However, no large view of these systems in differential physiological conditions is available yet. Combining the use of pure mitochondrial fractions and a multiplexed oligonucleotide cleavage assay on a microarray, we demonstrated that a large range of glycosylase activities were present in Drosophila mitochondria. Most of them were quantitatively different from their nuclear counterpart. Moreover, these activities were modified during aging.     

Deployment of DNA polymerases beta and lambda in single-nucleotide and multinucleotide pathways of mammalian base excision DNA repair

There exist two major base excision DNA repair (BER) pathways, namely single-nucleotide or “short-patch” (SP-BER), and “long-patch” BER (LP-BER). Both pathways appear to be involved in the repair of small base lesions such as uracil, abasic sites and oxidized bases. In addition to DNA polymerase β (Polβ) as the main BER enzyme for repair synthesis, there is evidence for a minor role for DNA polymerase lambda (Polλ) in BER. In this study we explore the potential contribution of Polλ to both SP- and LP-BER in cell-free extracts. We measured BER activity in extracts of mouse embryonic fibroblasts using substrates with either a single uracil or the chemically stable abasic site analog tetrahydrofuran residue. The addition of purified Polλ complemented the pronounced BER deficiency of POLB-null cell extracts as efficiently as did Polβ itself. We have developed a new approach for determining the relative contributions of SP- and LP-BER pathways, exploiting mass-labeled nucleotides to distinguish single- and multinucleotide repair patches. Using this method, we found that uracil repair in wild-type and in Polβ-deficient cell extracts supplemented with Polλ was ∼80% SP-BER. The results show that recombinant Polλ can contribute to both SP- and LP-BER. However, endogenous Polλ, which is present at a level ˜50% that of Polβ in mouse embryonic fibroblasts, appears to make little contribution to BER in extracts. Thus Polλ in cells appears to be under some constraint, perhaps sequestered in a complex with other proteins, or post-translationally modified in a way that limits its ability to participate effectively in BER.     

Cytotoxicity and mutagenesis induced by singlet oxygen in wild type and DNA repair deficient Escherichia coli strains

Singlet oxygen ((1)O(2)) is a product of several biological processes and can be generated in photodynamic therapy, through a photosensitization type II mechanism. (1)O(2) is able to interact with lipids, proteins and DNA, leading to cell killing and mutagenesis, and can be directly involved with degenerative processes such as cancer and aging. In this work, we analyzed the cytotoxicity and mutagenesis induced after direct treatment of wild type and the DNA repair fpg and/or mutY deficient Escherichia coli strains with disodium 3,3'-(1,4-naphthylidene) diproprionate endoperoxide (NDPO(2)), which releases (1)O(2) by thermodissociation. The treatment induced cell killing and mutagenesis in all strains, but the mutY strain showed to be more sensitive. These results indicate that even (1)O(2) generated outside bacterial cells may lead to DNA damage that could be repaired by pathways that employ MutY protein. As (1)O(2) is highly reactive, its interaction with cell membranes may generate secondary products that could react with DNA, leading to mutagenic lesions.