| Nucleic Acids Research | Pages |
Efficient in vitro repair of 7-hydro-8-oxodeoxyguanosine by human cell extracts: involvement of multiple pathways
Introduction
Materials And Methods
Materials
Plasmid DNA
Preparation of whole cell extracts
Fpg nicking assay
Repair incorporation assay
Kinetics experiments
Data analysis
Analysis of the ratio of dGMP to dCMP incorporation by DNA repair pathways
Results
Effect of increased dose of visible light during methylene-blue treatment of DNA on in vitro repair synthesis by normal WCE
Determination of the number of Fpg protein-sensitive sites
Efficiency of repair incorporation by cell extracts into methylene blue and UV-damaged DNA
Specific nucleotide differences in DNA repair incorporation into methylene blue-damaged DNA by human whole cell extracts
Kinetics of repair of methylene blue induced damage
Effect of DNA polymerase inhibitors on incorporation of radiolabeled dGMP and dCMP by normal human WCE into methylene blue-damaged DNA
Incorporation of radiolabeled dCMP into methylene blue-damaged DNA by extracts from normal and repair deficient cells
Models of DNA repair pathways
Discussion
Acknowledgements
References
Efficient in vitro repair of 7-hydro-8-oxodeoxyguanosine by human cell extracts: involvement of multiple pathways
ABSTRACT
INTRODUCTION
Oxidative damage to DNA is an inevitable consequence of metabolic activity and the chemical instability of DNA (1-3). Oxidative damage in DNA has often been suggested as a contributing factor in the process of aging and the development of cancer (4). The repair of oxidative damage in DNA by human cells is not well understood due to the multitude of lesions produced by oxidative treatments (5) and the complexity of the response. The repair of damage induced by ionizing radiation and other agents producing base modifications as well as single-strand breaks has both aphidicolin-sensitive and -resistant components (6), which has been interpreted in terms of long patch and short patch repair (7). In permeablized human fibroblasts, the repair of damage induced by the radiomimetic agent bleomycin involves short patch repair with polymerase [beta] and long patch repair with polymerase [delta] (8). However, because the above treatments produce a variety of lesions in DNA, heterogeneity in the repair has been attributed to the repair of different lesions by different repair pathways.
One of the most studied and important lesions produced in DNA by reactive oxygen species is 7-hydro-8-oxodeoxyguanosine (8-oxodG). Due to its mispairing with deoxyadenosine, 8-oxodG is mutagenic (9,10). In Escherichia coli, 8-oxodG is removed from DNA by either a specific glycosylase, known as formamidopyrimidine glycosylase (Fpg protein), in the initial step of base excision repair (BER), or by the UvrABC enzymes through nucleotide excision repair (NER) (11). Cells deficient in either pathway do not show an appreciable defect in the repair of 8-oxodG, but the double mutants exhibit sensitivity to agents producing 8oxodG lesions (11). Eukaryotic cells may also employ multiple pathways for the removal of this common lesion. Separable activities of an 8-oxodG endonuclease and an 8-oxodG-specific glycosylase have been identified in human cells (12). The yeast (13,14), mouse (15) and human (16,17) genes encoding homologous 8-oxodG-specific glycosylase/apurinic lyases have recently been cloned. A similar mitochondrial-specific activity has also been characterized (18). The Drosophila ribosomal protein S3 displays specific 8-oxodG and AP activities (19). The human, but not the mouse, N-methylpurine DNA glycosylase exhibits an 8-oxodG incising activity (20). In addition, the involvement of the human NER system in the repair of 8-oxodG has recently been demonstrated (21). However, the relative importance of these activities in the overall repair of 8-oxodG in human cells has not been evaluated.
Most agents producing oxidative damage in DNA result in the formation of multiple DNA base lesions. However, photoactivated methylene blue produces almost exclusively the base modification 8-oxoguanosine in double-stranded DNA (22,23) without significant conversion of the other three bases (24). With 1% of the guanosines modified, 8-oxoguanosine and formamidopyrimidine are detected in a ratio of 20:1 (23). At twice that level of modification, an additional product is detected, but at <0.3% of the level of 8-oxodG (25). At higher levels of modification, eight additional species are formed (24), although 8-oxodG remains the most prevalent. In addition to 8-oxodG, the formation of another lesion in single-stranded DNA, but not double-stranded DNA, has been proposed based on the mutation spectrum (26). Thus under conditions in which <1% of the guanines are modified, the reaction of double-stranded DNA with photoactivated methylene blue produces predominantly 8-oxodG.
Although the presence of 8-oxodG-specific glycosylases in mammalian cells has been amply demonstrated, little is known about the overall repair of this lesion. DNA damaged by exposure to methylene blue and visible light represents a useful model system for in vitro investigation of the repair of a common oxidative lesion in DNA due to the near-exclusive formation of the lesion 8-oxodG. We have used plasmid DNA damaged by methylene blue and visible light to investigate the repair of 8-oxodG by extracts from human cells.
MATERIALS AND METHODS
Materials
Radioisotopes [[alpha]-32P]dCTP and [[alpha]-32P]dGTP (3000 Ci/mmol) were purchased from Amersham or DuPont NEN Research Products. Radiolabeled dGTP was used within 5 days of the reference date. Methylene blue was obtained from Ricca Chemical. ATP, dNTPs and restriction enzymes were obtained from Boehringer Mannheim. Aphidicolin, HEPES, creatine phosphokinase (rabbit muscle), phosphocreatine and protease inhibitors were purchased from Sigma. BSA, cell culture media and supplements were purchased from GIBCO BRL. Fpg enzyme (5 × 107 U/mg) was obtained from Dr A.P.Grollman (SUNY, Stony Brook, NY). Endonuclease III was purchased from Trevigen Inc.
Plasmid DNA
pBluescript II KS(+) (Stratagene), referred to herein as pKS(+), and pRS, a 4470 bp derivative of pUC19 constructed by Dr G.Dianov (NIA, Baltimore MD), were propagated in XL1-Blue cells (Stratagene) and prepared by the lysozyme-Triton X-100 method (27) with modifications. Phenol and chloroform were not used during plasmid purification. pKS(+) DNA was purified by cesium chloride-ethidium bromide equilibrium centrifugation prior to treatment with damaging agents. Undamaged pRS DNA was purified by a total of three cesium chloride-ethidium bromide centrifugation steps and one sucrose gradient sedimentation step.
UV-damaged pKS(+) plasmid was prepared by irradiating DNA at 50 µg/ml in 40 mM HEPES-KOH (pH 7.9), 0.1 M KCl, 0.5 mM EDTA with UV-C (254 nm) at a fluence of 1.01 W/m2 for a total of 450 J/m2. The irradiated plasmid was treated with E.coli endonuclease III to remove pyrimidine hydratesand further purified by two cesium chloride-ethidium bromide gradients and one sucrose gradient. The resulting plasmid is free of nicks and serves as a standard substrate for NER activity.
Oxidatively damaged pKS(+) was prepared by treating 0.1 mg/ml DNA with 10 µM methylene blue, in 0.01 M sodium phosphate buffer (pH 7.4). The plasmid DNA was kept on ice and exposed to visible light at a fluence of 117 W/m2, from a 100 W tungsten bulb, for varying amounts of time. In addition, pKS(+) DNA treated with methylene blue but kept in the dark was prepared as an external control. Methylene blue was removed by three consecutive ethanol precipitation steps. Except for the visible light exposure, all manipulations with methylene blue were performed in the dark or under a dim blue safelight. Plasmid DNA was further purified by two cesium chloride-ethidium bromide gradient centrifugation steps and one sucrose gradient step. None of the plasmids had a detectable population of nicked molecules and the fraction of nicked molecules did not increase significantly during storage at 4°C for 4 months.
Preparation of whole cell extracts
Human lymphoblastoid cell lines were obtained from NIGMS Human Genetic Mutant Cell Repository and NIA Aging Cell Repository (Coriell Institute for Medical Research, Camden, NJ). The cell lines were: GM01310b (normal), AG09387 (normal), GM02250e (XPA), GM02252a (XPB), GM02253e (XPD), GM01857 (CSA) and GM01712a (CSB). Cells were grown in RPMI 1640 media, supplemented with 13% fetal bovine serum, 2 mM l-glutamine, 100 U/ml penicillin G and 100 µg/ml streptomycin sulfate, and maintained at 37°C with 5% CO2 and 95% humidity. All cell lines were negative for mycoplasm contamination. The cells were harvested at a density of [sim]6 × 105 cells/ml. Whole cell extracts (WCE) were prepared as described (28). WCE exhibited the expected proficiency or deficiency for repair incorporation into UV-damaged plasmid and XP WCE exhibited the expected pattern of complementation (29).
Fpg nicking assay
DNA damaged by exposure to methylene blue and visible light was treated with Fpg protein to determine the number of enzyme-sensitive sites. DNA at a concentration of 2.5 µg/ml in 50 mM Tris-HCl (pH 8.0), 50 mM KCl and 1 mM EDTA was incubated with various amounts of Fpg protein in a 20 µl volume at 375C for 25 min. The products were separated on a 1% agarose gel containing 0.5 µg/ml ethidium bromide. The relative fluorescence of the supercoiled and nicked bands was determined using a FluorImager (Molecular Dynamics) and analyzed using ImageQuant software (Molecular Dynamics). Assuming a Poisson distribution of the damage, the average number of enzyme sensitive sites per plasmid is given by the expression: -ln (1.4 × Fs/(1.4 × Fs + Fn)), where Fs and Fn are the fluorescence of the supercoiled and nicked bands, respectively, and the factor 1.4 corrects for the difference in ethidium bromide binding to supercoiled compared to linear DNA.
Repair incorporation assay
Repair incorporation assays were performed essentially as described (28). Briefly, reactions contained 300 ng each of supercoiled damaged pKS(+) and undamaged pRS, 100 µg of WCE protein in a volume of 50 µl in a solution of 44 mM HEPES-KOH (pH 7.9), 70 mM KCl, 7.5 mM MgCl2, 1.2 mM DDT, 0.5 mM EDTA, 2 mM ATP, 250 µg/ml BSA, 40 mM phosphocreatine, 50 µg/ml creatine phosphokinase, and 50 µM each of dATP and dTTP. In addition, reactions contained either 50 µM dGTP, 5 µM dCTP and 1 µCi of [[alpha]-32P]dCTP or 50 µM dCTP, 5 µM dGTP and 1 µCi of [[alpha]-32P]dGTP. The reactions were incubated at 30°C for 2 h, stopped by addition of EDTA, and then treated with RNaseA followed by proteinase K. The DNA was recovered by phenol-chloroform extraction and ethanol precipitation. Plasmids were linearized with EcoRI endonuclease and separated by electrophoresis through a 1% agarose gel, containing 0.5 µg/ml ethidium bromide. The gel was scanned to measure the intensity of ethidium bromide fluorescence (FluorImager, Molecular Dynamics). The gel was dried under vacuum and exposed to a storage screen (PhosphorImager, Molecular Dynamics) for 41-42 h.
Kinetics experiments
Reactions to determine the rate of nucleotide incorporation were performed similarly, except that the reaction mix containing cell extracts was preincubated at 30°C for 5 min before the reaction was initiated by addition of DNA. WCE nicking experiments were performed similarly, except that dNTPs were omitted from the reaction and the DNA was not linearized.
Data analysis
Autoradiograms and images of ethidium bromide fluorescence were analyzed using ImageQuant software (Molecular Dynamics). Gel image intensities are displayed with a linear response to the signal. Fluorescence intensities of linearized pKS(+) and pRS bands, in comparison with known amounts of DNA, were used to determine the DNA loading. Background-corrected values of the radioactivity incorporated for the damaged and undamaged plasmids were normalized for the amount of DNA. The damage-specific radioactivity incorporated was calculated by subtracting the background-corrected and normalized radioactivity incorporated in undamaged pRS from that incorporated into damaged pKS(+). In calculating the moles of nucleotide incorporated, the amount of 32P-labeled nucleotide was corrected for decay. The counts per minute of radioactivity incorporated may be related to the moles of the specific nucleotide incorporated by the use of the equation:
 |
1 |
The radioactivity incorporated (d.p.m.) was determined by liquid scintillation counting of the excised band or by using a particular phosphorus storage screen for which the counting efficiency had been determined.
Analysis of the ratio of dGMP to dCMP incorporation by DNA repair pathways
Assuming that the damage is localized exclusively to guanosine and that the surrounding sequence does not affect the extent of damage formation, then the average size of the repair patch can be deduced from the differential incorporation of dGMP to dCMP. We analyzed the sequence of pKS(+) (GenBank #X52327) using a custom program written in Visual Basic. The fraction of all Gs having a G immediately 3[prime] was q2,G = 0.257 while the fraction with a C in that position was q2,C = 0.300. Similarly for the next position to the 3[prime], q3,G = 0.279 and q3,C = 0.216, and so on. Averaging up to the patch size, n, leads to an expression for the ratio of dGMP to dCMP incorporated,
 |
2 |
For patch sizes of 2-6, the ratio of dGMP to dCMP incorporated is predicted to take the values: 4.19, 2.98, 2.35, 2.03, 1.78.
However, if repair proceeds by two pathways, one replacing only the modified base and a second replacing the modified G and a characteristic number of adjacent nucleotides, then the proportion of the lesions that are repaired by each pathway can be determined if the patch size of the second pathway, m, is known. The fraction of the lesions repaired by the second pathway, fb is given by the expression:
 |
3 |
RESULTS
Effect of increased dose of visible light during methylene-blue treatment of DNA on in vitro repair synthesis by normal WCE
Figure 1. Increasing light exposure to methylene-blue treated DNA increases in vitro repair incorporation by human lymphoblast cell extracts. Standard reactions containing WCE from GM01310b (normal) cells, undamaged DNA (- damage) and DNA treated with 10 µM methylene blue and exposed to various amounts of visible light (+ damage). (A) Gel images show ethidium bromide staining of DNA and incorporation of [[alpha]-32P]dCMP. (B) Damage-specific incorporation of [[alpha]-32P]CMP is expressed as counts per ng and increases with increasing exposure to light. The arrow marks the exposure used for subsequent experiments. This experiment was repeated with similar results but the values were not averaged because PhosphoImager screens with different efficiency were used. As shown in Figure 
Determination of the number of Fpg protein-sensitive sites
Fpg protein possesses 8-oxodG glycosylase and AP lyase activities (23,31). We determined the number of Fpg protein-sensitive sites per plasmid molecule created by exposing the pKS(+) plasmid to 21 kJ/m2 visible light in the presence of 10 µM methylene blue. As shown in Figure Figure 2. Determination of the number of Fapy glycosylase-sensitive sites in pKS(+) DNA treated with methylene blue and 21 kJ/m2 of visible light. Ethidium bromide stained gel showing conversion of supercoiled to nicked DNA by Fpg protein. The number of enzyme sensitive sites per plasmid was determined using the Poisson distribution.
Efficiency of repair incorporation by cell extracts into methylene blue and UV-damaged DNA
The incorporation of [[alpha]-32P]dCMP by normal WCE into the methylene blue-damaged plasmid and the UV-damaged plasmid is shown in Figure
Figure 3. Efficient DNA repair synthesis into DNA containing methylene blue and visible light-induced damage compared with UV-damaged DNA. Gel images show ethidium bromide staining of DNA and incorporation of [[alpha]-32P]dCMP. The larger plasmid (- damage) is the undamaged internal control. WCE from normal human cells (GM01310b) incorporate [[alpha]-32P]CMP into DNA damaged by methylene blue and 21 kJ/m2 of visible light more efficiently than into DNA damaged by 450 J/m2 UVC.
Specific nucleotide differences in DNA repair incorporation into methylene blue-damaged DNA by human whole cell extracts
The amount of radioactivity incorporated by human cell extracts during the repair of methylene blue-induced damage depends on the identity of the labeled nucleotide. As shown in Figure Figure 4. Nucleotide-specific incorporation by normal human whole cell extracts (GM01310b) into DNA damaged by exposure to methylene blue and visible light. Gel images show ethidium bromide staining of DNA and incorporation of [[alpha]-32P]dGMP or [[alpha]-32P]dCMP into damaged or undamaged DNA. The larger plasmid (- damage) was an untreated internal control. The smaller plasmid (+ damage) was treated with 10 µM methylene blue and 21 kJ/m2 of visible light. We have performed DNA repair synthesis experiments with the concentration of the limiting nucleotide varied between 2 and 20 µM and the amount of labeled nucleotide constant at 1 µCi (data not shown). The amount of nucleotide incorporated, when calculated according to equation 1 and corrected for isotopic dilution, is essentially constant over this range. Thus residual nucleotides in the cell extracts do not contribute significantly in the reaction. In addition, experiments with radiolabeled nucleotide used immediately after delivery or 10 days later give the same ratio of incorporated dGMP to dCMP (data not shown), indicating that degradation products in the labeled nucleotide stock do not significantly affect incorporation. 
Kinetics of repair of methylene blue induced damage
To determine the rate limiting step in the repair of 8-oxodG, the kinetics of nicking and resynthesis by the WCE was investigated. Damage specific nicking of the methylene blue-damaged plasmid by the WCE occurred rapidly, with a limiting value of [sim]0.5 nicks per plasmid being attained within 15 min (data not shown). The undamaged plasmid, pRS, was also nicked by the WCE, exhibiting [sim]0.07 non-specific nicks per 3 kb in that time.
In contrast, the resynthesis step is slower. As shown in Figure Figure 5. Kinetics of nucleotide incorporation by GM01310 (normal) WCE into DNA damaged by methylene blue and visible light. (A) Gel images showing DNA loading and incorporation of radioactivity for reactions containing [[alpha]-32P]CTP and [[alpha]-32P]GTP. (B) Time-dependence of damage specific incorporation of radioactivity. Incorporation of radioactivity was determined by liquid scintillation counting of excised bands and is expressed as fmoles of dNMP incorporated per 300 ng of damaged DNA: closed circles, [[alpha]-32P]GTP; open circles, [[alpha]-32P]CTP.
Effect of DNA polymerase inhibitors on incorporation of radiolabeled dGMP and dCMP by normal human WCE into methylene blue-damaged DNA
The effects of specific DNA polymerase inhibitors were examined and the results are summarized in Table 1. The incorporation of dCTP is moderately sensitive to the presence of aphidicolin, exhibiting 58% of the activity compared to the control. Addition of ddTTP alone has no effect on the incorporation, but in the presence of both inhibitors, the activity is reduced to 33%. The incorporation of dGTP is comparatively less sensitive to aphidicolin and more sensitive to ddTTP, giving 94 and 77% of the activity, respectively. In the presence of both inhibitors, the incorporation of dGTP is reduced to 45% of the control.
With purified proteins, aphidicolin and ddTTP are selective inhibitors of pol [delta] (or pol [epsis]) and pol [beta], respectively. However, whole cell extracts or permeabilized cells have often displayed only partial inhibition (30,32,33). Although the inhibitors are only partially effective in the present system, the differences in the sensitivities of dGMP and dCMP incorporation to the inhibitors suggests differences in the repair pathways detected by the two nucleotides.
Table 1
| Inhibitor | fmoles dCMP incorporateda |
relative dCMP incorporated |
fmoles dGMP incorporateda |
relative GCMP incorporated |
| none | 325 ± 36 | 1.00 | 1139 ± 34 | 1.00 |
| aphidicolin | 194 ± 31 | 0.60 | 1071 ± 9 | 0.94 |
| ddTTP | 322 ± 33 | 0.99 | 875 ± 19 | 0.77 |
| both | 88 ± 17 | 0.27 | 467 ± 52 | 0.41 |
Incorporation of radiolabeled dCMP into methylene blue-damaged DNA by extracts from normal and repair deficient cells
To examine the effects of defective NER on the repair of this oxidatively damaged DNA, we determined the in vitro DNA repair synthesis performed by extracts from apparently normal donors, xeroderma pigmentosum (XP) and Cockayne syndrome patients. The results are shown in Figure
The observation that cell extracts from two differentXP complementation groups exhibit a deficiency in repair suggests that some XP proteins may directly or indirectly influence the repair of 8-oxodG detected by the incorporation of dCMP. However, the proficiency of XPD extracts in this repair suggests that it is not NER per se. CS extracts, shown here to be proficient or nearly proficient in the repair of methylene blue-induced damage, are also proficient in the in vitro repair of UV-induced damage (29,34).
Models of DNA repair pathways
The considerable specificity of damage produced by methylene blue permits differences in dGMP and dCMP incorporation to be interpreted in terms of the size of the repair patch. Based on the average ratio of dGMP to dCMP incorporation of robs = 2.7 ± 0.5, the average repair patch size of n = 3.3 ± 0.7 is calculated by interpolation using equation 2. The use of a model assuming a random sequence in the repair patch 3[prime] to the modified G, with an equal proportion of each nucleotide, does not produce a significantly different result. This implies that the in vitro repair of the oxidative damage lesion, 8-oxodG, when present in plasmid DNA, does not simply occur by the single nucleotide replacement pathway. Either the repair is dominated by a pathway that replaces approximately two adjacent nucleotides in addition to the modified guanosine, or two pathways contribute significantly to the repair. An average repair patch size of [sim]4 was determined for the repair of a synthetic (reduced) abasic site in Xenopus laevis oocyte extracts (35).
Figure 6. Damage-specific incorporation of [[alpha]-32P]CMP by WCE from normal and repair defective cells into DNA damaged by methylene blue and visible light. (A) Gel images showing ethidium bromide staining of DNA and incorporation of [[alpha]-32P]CMP. (B) Histogram showing damage-specific incorporation into methylene-blue damaged DNA. Data represent the mean ± standard deviation of two experiments and are normalized to GM01310b. Differences in the kinetics of incorporation of dCMP compared to dGMP (Fig. 
DISCUSSION
We have inferred characteristics of the repair of 8-oxodG in human cells by examining the damage-specific nicking and DNA repair synthesis activity of whole cell extracts on covalently closed circular DNA containing damage induced by methylene blue and visible light. Most of the damage is repaired by a BER process rather than NER as indicated by several observations, including the overall efficiency of repair of methylene blue-induced damage compared to UV-induced damage, the identification of DNA synthesis rather than incision as rate-limiting and the proficiency in the repair of this oxidative damage by XPD extracts, which are defective in the repair of UV-induced damage. These observations do not preclude a small contribution from NER (21). The involvement of more than one pathway in this efficient repair is suggested by the differences in the kinetics of dGMP and dCMP incorporation and differences in the sensitivity of incorporation of the two nucleotides to the inhibitors aphidicolin and ddTTP. Deficient repair of 8-oxodG lesions by XPA and XPB extracts, as detected by incorporation of dCMP, suggests that some XP proteins may directly or indirectly influence the repair of this lesion.
Many studies indicate that 8-oxodG is repaired rapidly in human cells. Fpg protein-sensitive sites produced by photoactivated acridine orange, which include 8-oxodeoxyguanosine and formamidopyrimidine, are rapidly repaired in both the nucleus and mitochondria, with >80% of the damage removed from actively transcribed DNA within 2 h, based on the gene-specific repair assay (36). An examination of the repair of specific base lesions induced by H2O2 in human cells, using the GC/MS technique, has shown that 8-oxoguanosine and 8-OH-adenine were removed more slowly than pyrimidine lesions and some other purine lesions (37). About 50% of the 8-oxodG is removed in 1 h (37). The rapid repair of the oxidative damage can be compared with the considerably slower repair of cyclobutane dimers in actively transcribed genes, which requires 24 h to repair 80% of the lesions (38), suggesting that most 8-oxodG lesions are removed by base excision repair processes.
However, some evidence implicates the mammalian nucleotide excision repair pathway in the repair of 8-oxodG. A shuttle vector containing a single 8-oxodG was defective in a plasmid replication assay after passage through XPA cells compared with normal cells, suggesting that XPA cells are defective in the repair of 8-oxodG (39). However, no significant impairment in the repair of methylene blue plus light induced damage was detected in XPA cells using a host cell reactivation assay (40). Plasmid DNA damaged by [gamma]-irradiation or H2O2 and Cu2+ treatments, which produce a variety of base damages, did not show any significant difference in DNA repair synthesis between normal and XP cell extracts (41). However, after treatment of the plasmids with the E.coli enzymes Fpg protein and endonuclease III, the remaining damage was repaired in normal but not XP extracts (41). Recently, the excision of the characteristic 8-oxodG-containing oligonucleotide by the NER pathway using cell extracts or purified repair proteins has been demonstrated (21). The overall efficiency of that process is low and does not account for the efficient repair we observe here. Based on the 2-fold greater efficiency of excision of 8-oxodeoxyguanosine by the NER system compared with a cyclobutane dimer (21), we estimate that only a small fraction of the incorporated dCMP can be attributed to nucleotide excision repair.
Although cells from XP patients are considered to have normal activity in BER, some reports suggest a compromised ability to repair oxidative damage. Here we have shown that cell extracts from the XPA lymphoblastoid cell line, GM02250, exhibited considerably less [[alpha]-32P]dCMP incorporation into DNA containing methylene blue induced damage compared with normal cell extracts. As discussed above, extracts from this cell line are also defective in the repair of Fpg protein- and endonuclease III-resistant lesions produced by [gamma]-irradiation or treatment with H2O2 and Cu2+ (41). Fibroblast cells derived from the same patient (XP12BE) exhibited [sim]80% of normal repair of [gamma]-irradiation induced damage in MT genes, based on the incorporation of BrdU, and were slightly sensitive to [gamma]-irradiation (42). However, repair of thymine glycols in the same fibroblast cells is normal (43).
Although CS has been characterized as being defective in transcription-coupled NER, various CS lines also exhibit radiosensitivity (42). CS3BE fibroblasts derived from the same patient as the CSA line studied here exhibits only a slight increase in sensitivity (42), while a different CSB patient (CS1AN) was quite sensitive. The radiosensitivity of CS cells has recently been attributed to the requirement for CSB protein in the transcription-coupled repair of thymine glycols, another oxidative damage lesion in DNA also thought to be repaired predominantly by BER (43). The global repair of thymine glycols was not defective in CSB cells (43), which is similar to the absence of a significant defect in the in vitro repair of 8-oxodG lesions shown here.
BER of uracil from DNA in mammalian cells proceeds predominantly by a single nucleotide replacement mechanism involving DNA polymerase [beta] (32). Recently, the existence of an alternative, PCNA-dependent pathway has been demonstrated for the resolution of repair initiated by glycosylases in Xenopus (44), hamster (45) and human extracts (46). The dependence of this second pathway on PCNA may arise either through the involvement of DNA polymerase [delta] or [epsis] (33,47), in which PCNA loads polymerase [delta] or [epsis] onto the DNA, or through PCNA stimulation of FEN-1activity (46). Although this second pathway has been demonstrated to operate on reduced or oxidized AP sites (46), it may also be an essential feature of the repair of oxidized base damage recognized by eukaryotic glycosylases that proceed by a mechanism similar to E.coli endonuclease III, creating a gap bounded by a DNA strand with a 3[prime] unsaturated aldehyde (14).
In this paper, we have presented evidence that a common oxidative lesion in DNA is efficiently repaired by at least two pathways in human cells. These pathways may be initiated by a glycosylase specific for the removal of 8-oxodG such as the hOGG1 protein. Following the initial nicking, the repair may proceed through single base replacement or by a slower process that involves the replacement of the 8-oxodG and several surrounding bases. These repair pathways may be similar to the PCNA-dependent and independent pathways which have been previously described.
ACKNOWLEDGEMENTS
We appreciate the interaction with the Danish Center for Molecular Gerontology. Insightful comments on the manuscript were provided by R. M. Anson, D. Croteau, N. Hoehr and R. Stierum. G. Dianov generously provided samples of WCE and plasmid DNA for comparison in initial studies.
REFERENCES
This article has been cited by other articles:
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 17 Apr 1998
Copyright©Oxford University Press, 1998.
CiteULike 
Connotea 
Del.icio.us What's this?
M. S. Cooke, R. Olinski, S. Loft, and members of the European Standards Committee on Uri
Measurement and Meaning of Oxidatively Modified DNA Lesions in Urine
Cancer Epidemiol. Biomarkers Prev.,
January 1, 2008;
17(1):
3 - 14.
[Abstract]
[Full Text]
[PDF]
N. R. Bianco, G. Perry, M. A. Smith, D. J. Templeton, and M. M. Montano
Functional Implications of Antiestrogen Induction of Quinone Reductase: Inhibition of Estrogen-Induced Deoxyribonucleic Acid Damage
Mol. Endocrinol.,
July 1, 2003;
17(7):
1344 - 1355.
[Abstract]
[Full Text]
[PDF]
A. Li, D. Schuermann, F. Gallego, I. Kovalchuk, and B. Tinland
Repair of Damaged DNA by Arabidopsis Cell Extract
PLANT CELL,
January 1, 2002;
14(1):
263 - 273.
[Abstract]
[Full Text]
[PDF]
M. Jaiswal, N. F. LaRusso, and G. J. Gores
Nitric oxide in gastrointestinal epithelial cell carcinogenesis: linking inflammation to oncogenesis
Am J Physiol Gastrointest Liver Physiol,
September 1, 2001;
281(3):
G626 - G634.
[Abstract]
[Full Text]
[PDF]
M. Jaiswal, N. F. LaRusso, L. J. Burgart, and G. J. Gores
Inflammatory Cytokines Induce DNA damage and Inhibit DNA repair in Cholangiocarcinoma Cells by a Nitric Oxide-dependent Mechanism
Cancer Res.,
January 1, 2000;
60(1):
184 - 190.
[Abstract]
[Full Text]
G. Dianov, C. Bischoff, J. Piotrowski, and V. A. Bohr
Repair Pathways for Processing of 8-Oxoguanine in DNA by Mammalian Cell Extracts
J. Biol. Chem.,
December 11, 1998;
273(50):
33811 - 33816.
[Abstract]
[Full Text]
[PDF]
This Article 
Abstract
Print PDF (289K)
Alert me when this article is cited
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Search for citing articles in:
ISI Web of Science (28)
Request Permissions
Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Articles by Jaiswal, M.
Articles by Mazur, S. J.
Search for Related Content
PubMed
PubMed Citation
Articles by Jaiswal, M.
Articles by Mazur, S. J.
Social Bookmarking
What's this?