Transcript
[CANCER RESEARCH 64, 1050 –1057, February 1, 2004]
Expression of Base Excision DNA Repair Genes Is a Sensitive Biomarker for in Vivo Detection of Chemical-induced Chronic Oxidative Stress: Identification of the Molecular Source of Radicals Responsible for DNA Damage by Peroxisome Proliferators Ivan Rusyn,1,2 Shoji Asakura,2 Brian Pachkowski,2 Blair U. Bradford,1 Mikhail F. Denissenko,3 Jeffrey M. Peters,4 Steven M. Holland,5 Janardan K. Reddy,6 Michael L. Cunningham,7 and James A. Swenberg2 1 Laboratory of Environmental Genomics, and 2Laboratory of Molecular Carcinogenesis and Mutagenesis, Department of Environmental Sciences and Engineering, University of North Carolina School of Public Health, Chapel Hill, North Carolina 27599; 3Sequenom, Inc., San Diego, California 92121; 4Department of Veterinary Science, Pennsylvania State University, University Park, Pennsylvania; 5Laboratory of Host Defenses, National Institute of Allergy and Infectious Diseases, NIH, Bethesda, Maryland; 6Department of Pathology, Feinberg School of Medicine, Northwestern University, Chicago, Illinois; and 7National Center for Toxicogenomics and National Toxicology Program, National Institute of Environmental Health Sciences, Research Triangle Park, North Carolina
ABSTRACT Oxidative stress to DNA is recognized as one of the mechanisms for the carcinogenic effects of some environmental agents. Numerous studies have been conducted in an attempt to document the fact that chemical carcinogens that are thought to induce production of oxidants also cause the formation of oxidative DNA lesions. Although many DNA adducts continue to be useful biomarkers of dose/effect, changes in gene expression have been proposed to be a practical novel tool for studying the role of chemically induced oxidative DNA damage. Here, we hypothesized that expression of base excision DNA repair genes is a sensitive biomarker for in vivo detection of chemically induced chronic oxidative stress. To test this hypothesis, mice were treated with a known rodent carcinogen and peroxisome proliferator, WY-14,643 (500 ppm, 1 month). A number of end points that are commonly used to assess oxidative DNA damage were considered. Our data demonstrate that no difference in 8-oxoguanine, the number of abasic sites, or single strand breaks can be detected in genomic DNA from livers of control or WY-treated animals. However, a concordant marked induction of genes specific for the long-patch base excision DNA repair, a predominant pathway that removes oxidized DNA lesions in vivo, was observed in livers of WY-treated mice. Kupffer cell NADPH oxidase, and peroxisomes in parenchymal cells have been proposed as the potential sources of peroxisome proliferator-induced oxidants. The analysis of expression of base excision DNA repair genes was used to assess whether this biomarker of oxidative stress can be used to determine the source of oxidants. The data suggest that DNA-damaging oxidants are generated by enzymes that are induced after activation of peroxisome proliferator activator receptor ␣, such as those involved in lipid metabolism in peroxisomes, and are not the result of activation of NADPH oxidase in Kupffer cells. We conclude that expression of base excision DNA repair genes is a sensitive in vivo biomarker for chemically induced oxidative stress to DNA that can be successfully used for the identification of the molecular source of radicals responsible for DNA damage in vivo.
INTRODUCTION It is generally recognized that mutations are required for the process of carcinogenesis. Both correct replication of the whole genome and maintenance of meaningful sequences of individual genes are at risk when even slight modifications of DNA bases, referred to as DNA damage, occur. All cells possess elaborate machinery for the removal Received 9/25/03; revised 11/14/03; accepted 11/19/03. Grant support: NIH Grants ES11391, ES11660, ES05948, ES10126, GM23750, and CA16086. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Notes: I. Rusyn was a recipient of a Transition to Independent Position Award ES11660 and an individual postdoctoral National Research Service Award ES05920 from the National Institute of Environmental Health Sciences. Requests for reprints: Ivan Rusyn, Department of Environmental Sciences and Engineering, 357 Rosenau Hall CB 7431, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599-7431. Phone/Fax: (919) 843-2596; E-mail:
[email protected].
of modified bases, to minimize the introduction of mutations (1). Still, these protective mechanisms occasionally fail and trigger apoptosis, or cause mutations leading to growth of a selected cell population. Oxidative stress to DNA is recognized as a common pathway to mutations and is suspected to be one of the major causes of cancer (2– 4). Many environmental agents that are classified as “non-genotoxic” carcinogens are known to induce the production of free radicals (5). Oxidative stress to DNA is thought to be one of the primary modes of action of such chemicals (6), whereas other effects of oxidative stress, such as damage to proteins, lipids, and other cellular components, are also potentially involved. Despite wide appreciation of this fact, clear experimental evidence for persistent oxidative DNA damage after treatment with carcinogenic agents remains elusive. Peroxisome proliferators, a class of nongenotoxic rodent carcinogens (7), are a good example where a controversy over the role of oxidative stress to DNA in the mechanism of carcinogenesis has existed for decades. Evidence that oxidants are generated in liver after treatment with these agents has been repeatedly demonstrated and suggests that indirect oxidative DNA damage may occur. Still, conflicting reports on measurements of oxidative DNA bases in this model are plentiful (reviewed in Ref. 8). The difficulties in detecting persistent oxidative damage at the level of the base, or sugar, in vivo are both technical and biological. The former reflects the high potential for ex vivo adduct formation during DNA isolation (artifact) that most likely hampered many published reports (9), whereas the latter reflects the constant removal of DNA damage in a living system where numerous repair pathways are active (1). Recently, the historical ⬎100-fold difference in reported control levels of 8-oxo-deoxyguanosine in genomic DNA was experimentally accredited to erroneous DNA oxidation during isolation and/or sample processing in some of the analytical techniques used to measure this ubiquitous oxidative lesion (10, 11). The data from European Standards Committee on Oxidative DNA Damage show that it is possible to minimize DNA oxidation during isolation and that some of the procedures to measure oxidized bases in purified DNA (e.g., high-performance liquid chromatography/electrochemical detection) agree with techniques that do not require processing of DNA (e.g., Comet assay). Still, the overall utility of measuring oxidant-derived DNA adducts under the conditions of chronic low-level oxidative stress in vivo remains a challenging approach to unequivocally demonstrate the role of free radicals in the mechanisms of action of environmental agents. Because induction of DNA repair is a known response to damage (12), we hypothesized that changes in expression of specific DNA repair pathways responsible for removal of oxidized DNA bases could be used as a sensitive and ex vivo artifact-free measure of in vivo chemically induced oxidative DNA damage. Importantly, our previous observations in rodent liver in vivo (13) demonstrated that per-
1050
DNA REPAIR GENES AS A BIOMARKER OF OXIDATIVE STRESS
oxisome proliferators cause time- and dose-dependent increases in mRNA for base excision repair genes (OGG1, APE, MPG, and Pol ), whereas no effect was observed on DNA repair genes that were not related to oxidative DNA damage (e.g., O6-methylguanine methyl transferase, MGMT). Furthermore, peroxisome proliferator-induced changes in expression of DNA repair genes correlated with the carcinogenic potency of these compounds, supporting a role of oxidative stress in the mechanism of action. Here, we further investigated what repair pathways are induced by peroxisome proliferators in vivo and asked what possible molecular sources of reactive oxygen species exist for DNA damage after treatment with these compounds. A variety of analytical approaches, such as detection of 8-oxoguanine, intact and cleaved abasic sites (i.e., single-strand breaks), and gene expression for DNA repair pathways, were used to assess the level of oxidative DNA damage after the treatment with a rodent carcinogen known to invoke oxidative stress in liver. Collectively, these studies suggest that expression of DNA base excision repair genes can be used as a sensitive end point for determining whether and how oxidative stress to DNA plays a role in the mechanism of action of environmental carcinogens in vivo. Most importantly, this biomarker, coupled with mouse knockout models, showed conclusively that peroxisome proliferator activator receptor ␣ (PPAR␣)inducible enzyme(s) in liver parenchymal cells is the source of DNAdamaging free radicals in the mechanism of action of peroxisome proliferators. MATERIALS AND METHODS
incubated in lysis buffer (Applied Biosystems) overnight at 4°C with proteinase K (500 mg/ml; Applied Biosystems). DNA was then extracted twice with a mixture of phenol:chloroform:water and once with chlorophorm:isoamyl alcohol (24:1), followed by ethanol precipitation. The extracted DNA was incubated in PBS (pH 7.4) with RNase A for 1 h, followed by DNA precipitation with cold ethanol. Then, the DNA pellet was resuspended in sterilized double distilled water. The DNA solutions were stored at ⫺80°C until assayed. The DNA extraction method used in this study is unlikely to modify the original number of AP sites and single-strand breaks in genomic DNA from intact tissues or cells, based on re-extraction data of DNA exposed to high concentrations of methylmethane sulfonate (0.5 mM; Sigma, St. Louis, MO).8 Measurement of 8-Oxoguanine. Chromatographic identification of 8-oxoguanine and release of 8-oxoguanine from genomic DNA by purified formamidopyrimidine glycosylase (Fpg) were performed as detailed in (18). Fpg was a kind gift of Dr. K. . Beckman (University of California at Berkeley). Briefly, 100 g of genomic DNA were incubated with Fpg (2 g/reaction; 1 h at 37°C) in Fpg reaction buffer [50 mM phosphate (pH 7.5), 10 mM NaCl, and 0.5 mM EDTA]. Then, the solution was filtered through an ultrafiltration column with a Mr 10,000 molecular weight cutoff (Microcon 10; Amicon). The volumes of ultrafiltrates that contained excised bases and other small molecules were adjusted to 100 l and were used for measurements using an high-performance liquid chromatography/electrochemical detection method using an electrochemical array detector (ESA, Chelmsford, MA). Running conditions and buffer concentrations were essentially as detailed by Beckman et al. (18). The reference amounts of 8-oxoguanine standard (10 –500 fmol; Sigma-Aldrich Rare Chemicals) were used to construct a calibration curve. The results were expressed in terms of 8-oxoguanine/106 G, based on the initial amount of DNA used (100 g) and assuming 660 g/mol bp in DNA and the conversion factor of fmol/g to ppm of 1.57. Apurinic/Apyrimidinic (AP) Site Assay. AP sites were measured following a procedure reported by Nakamura et al. (19). Briefly, 8 g of DNA in 150 l of PBS were incubated with 1 mM aldehyde reactive probe at 37°C for 10 min. After precipitation using cold ethanol, DNA was resuspended in TE buffer (10 mM Tris-HCl, pH 7.4, containing 1 mM EDTA). After dialysis using Microcon centrifugation, DNA was resuspended in double-distilled water. DNA (250 ng) in TE buffer was heat denatured and loaded on a nitrocellulose membrane (110 ng DNA/slot, Hybond-C Super; Amersham Pharmacia Biotech). The nitrocellulose membrane was soaked with 5⫻ SSC and then baked in a vacuum oven for 30 min. The membrane was preincubated with 10 ml of Tris-HCl containing BSA for 15 min and then incubated in the same solution containing streptavidin-conjugated horseradish peroxidase at room temperature for 45 min. After rinsing the nitrocellulose membrane, the enzymatic activity on the membrane was visualized by enhanced chemiluminescence reagents. The nitrocellulose filter was exposed to an X-ray film, and the developed film was analyzed using a Kodak Image Station 440. Quantitation was based on comparisons to internal standard DNA containing a known amount of AP sites. AP Site Cleavage Assay. To determine the number of cleaved AP sites in DNA (i.e., single-strand breaks), the AP site cleavage assay was performed as reported previously by Lin et al. (20). For exonuclease (Exo) III treatment, aldehyde reactive probe-reacted DNA (275 ng) and 30 units of Exo III (New England BioLabs, Beverley, MA) were incubated in 10 l of 50 mM HEPES/ KOH buffer (pH 7.5) containing 50 mM NaCl, 100 g/ml BSA, 2 mM DTT, and 5 mM MgCl2 for 10 min on ice. Immediately after the reaction, 210 l of TE buffer were added to the samples, followed by the AP site assay as detailed above. For T7 gene 6 exonuclease treatment, DNA (275 ng) pre-reacted with aldehyde reactive probe and 25 units of T7 Exo (United States Biochemical Corp., Cleveland, OH) were incubated in 10 l of 50 mM HEPES-KOH buffer (pH 7.5) for 30 s on ice, followed by addition of 210 l of TE buffer and measurement by the AP site assay. RNA Isolation and RNase Protection Assays. Total RNA was isolated using QuickPrep extraction kit (Amersham Pharmacia Biotech, Inc.), followed by RNeasy total RNA (Qiagen, Valencia, CA) extraction and dissolved in RNase-free water. Samples were stored at ⫺80°C. The quality of preparations was determined using Agilent Bio-Analyzer (Agilent Technologies, Palo Alto, CA). Expression of base excision DNA repair genes was analyzed with an
Animals and Treatments. The following mouse strains were used in this study: C57BL/6J (from The Jackson Laboratory), 129S4/Svter (from the laboratory of Dr. F. J. Gonzalez, National Cancer Institute), p47phox-null (on a C57BL/6J background, from the laboratory of Dr. S.M. Holland, National Institute of Allergy and Infectious Diseases; Ref. 14), PPAR␣-null (on 129S4/ Svter background, from the laboratory of Dr. F. J. Gonzalez, National Cancer Institute; Ref. 15), and acyl-CoA oxidase (AOX)-null (on C57BL/6J background, from the laboratory of Dr. J. K. Reddy, Northwestern University; Ref. 16). Mice were housed in sterilized microbarrier cages in special facilities with a 12-h night/day cycle designed to handle knockout animals (i.e., isolation cubicles) at the University of North Carolina at Chapel Hill. Temperature and relative humidity were maintained at 22 ⫾ 2°C and 50 ⫾ 5%, respectively. Animals were maintained on NIH-07 chow and purified water ad libitum. Treated animals were given either ground rodent chow or the same ground chow blended with WY-14,643 (500 ppm) for 1 month. One month after initiation of the treatment, mice were euthanized, and livers were removed, weighed, and after a portion of the liver was harvested for fixation and paraffin embedding, snap frozen for further analysis. The Division of Laboratory Animal Medicine maintains the facilities, and veterinarians were always available to insure animal health. All animals were given humane care in compliance with NIH and institutional guidelines. All procedures and treatments that involved live animals were approved by the Institutional Animal Care and Use Committee, and the appropriate protocols are on file. NADPH oxidase-deficient mice lack a critical cytosolic component, p47phox, and develop granulomatous disease after 8 –10 months (14). These mice, which have been characterized extensively at Dr. Holland’s laboratory at National Institute of Allergy and Infectious Diseases, were housed four per cage in a pathogen-free barrier facility with filtered air in sterilized cages and given sterilized feed and water. They were used before 4 months of age (i.e., well before granulomatous disease develops). Frozen liver tissue samples from AOX-null mice, 3– 4 months of age, were obtained from the laboratory of Dr. J. K. Reddy, Northwestern University. DNA Isolation. Genomic DNA was extracted by a procedure slightly modified from the method reported previously (17). To minimize formation of oxidative artifacts during isolation, 2,2,6,6-tetramethylpiperidinoxyl (TEMPO; final concentration, 20 mM) was added to all solutions, and all procedures were performed on ice. Briefly, frozen tissues were thawed and homogenized in PBS. After centrifugation at 2,000 ⫻ g for 10 min, the nuclear pellets were 1051
8
J.A. Swenberg and J. Nakamura, unpublished data.
DNA REPAIR GENES AS A BIOMARKER OF OXIDATIVE STRESS
RNase protection assay using mouse multi-probe RNA probe template sets (mBER-1 and mBER-2; BD PharMingen, San Diego, CA) essentially as described by Rusyn et al. (13). Riboprobes were synthesized in the presence of [32P]dUTP to yield labeled antisense RNA probes. The RNase protection assays were performed on 10 g of individual total RNA samples using a RiboQuant multi-probe RNase protection assay kit (BD PharMingen). Protected fragments were separated on 5% polyacrylamide denaturing nucleic acid separation gels, dried, and exposed to a phosphor-imaging screen. The intensity of protected bands was quantified using a phosphorimage analyzer (Cyclone, Perkin-Elmer, Wellesley, MA) and Kodak 1D Image Analysis Software (Scientific Imaging Systems, Eastman Kodak Company, Rochester, NY). The expression of individual gene transcripts was normalized to the intensity of housekeeping genes L32 and GAPDH and expressed in arbitrary units.
RESULTS Two groups of wild type mice (n ⫽ 5 per group) were fed either ground NIH-07 rodent chow or the same diet containing WY-14,643 (500 ppm) for 1 month. Morphological changes characteristic of the effects of peroxisome proliferators in rodent liver, such as ⬎ 2.5-fold increase in liver:body weight ratio, were observed in WY-treated animals (data not shown). No change in gross liver morphology or liver weight was evident in a control group. Genomic DNA was isolated and used for measurements of 8-oxoguanine and AP sites as detailed in “Materials and Methods.” The high-performance liquid chromatography/electrochemical detection method, coupled with Fpg pretreatment of DNA to release oxidatively damaged DNA bases, was used. Using this technique, the control levels of 8-oxoguanine in HeLa cell extracts have been reported to be less than one 8-oxoguanine lesion/million guanines (18). These values were corroborated recently by a multi-laboratory effort to validate measurements of oxidized lesions in genomic DNA, where it was shown that the levels of 8-oxoguanine in untreated HeLa cells are about 0.3–1.0 lesion/million guanines (11). In this study, when DNA samples from livers of control and WY-treated animals were compared, no significant difference in the amount of 8-oxoguanine was observed: 0.36 ⫾ 0.03 and 0.29 ⫾ 0.02 8-oxoguanine/106 guanine, respectively. Importantly, when commercially available calf thymus DNA was processed using identical procedures, the amount was more than seven 8-oxoguanine/ 106 guanine (data not shown). Next, we compared the amounts of AP sites, both total, cleaved, and intact, between control and WY-treated animals (Fig. 1). AP sites in DNA can be detected by the aldehyde reactive probe (19). Furthermore, it is possible to distinguish between intact (i.e., uncleaved) and
Fig. 1. Treatment with WY-14,643 (500 ppm, 1 month) has no effect on the accumulation of apurinic/apyrimidinic (AP) sites. The number of AP sites in genomic DNA isolated from livers of control (f) and WY-14,643-treated (䡺) mice was determined as described in “Materials and Methods.” Both total, 3⬘-nicked, or 5⬘-nicked, intact AP sites and residual aldehyde-reactive lesions were measured. Data are reported as means from four to five animals/group; bars, SE.
cleaved AP sites (17). Similar to the results of 8-oxoguanine measurements, no statistically significant difference in number of total, 3⬘-nicked, 5⬘-nicked, or intact AP sites was detected between control and WY-treated groups. It has been shown that treatment with some peroxisome proliferators causes an increase in expression of base excision DNA repair (BER) genes, an effect that correlates with a carcinogenic potency of these compounds (13). Because no measurable effect on the commonly used markers of oxidative stress to DNA was detected after a rather lengthy treatment with a potent rodent carcinogen that is thought to cause production of reactive oxygen species, the expression of BER genes was further analyzed (Table 1). In addition to dramatic increases in expression of a number of oxidative lesion-removing glycosylases and AP endonuclease, we found that treatment with WY-14,643 led to a preferential induction of the long-patch base excision repair genes: PCNA, Pol , FEN1, and Lig1 (Table 1). Importantly, expression of genes involved in one-nucleotide replacement branch of BER, such as XRCC1, PARP, and Lig3, did not change. Similarly, expression of MGMT, a protein that participates in a single-step repair of alkylated guanine residues, was also unaffected. Two major molecular sources of reactive oxygen species (ROS) have been suggested to be induced by peroxisome proliferators, lipid metabolizing enzymes in the parenchymal cell peroxisomes, and NADPH oxidase in Kupffer cells, the resident hepatic macrophages [reviewed by Rusyn et al. (8)]. Current risk assessments of these chemicals have focused on the important differences between humans and rodents in expression of PPAR␣, a nuclear receptor that regulates expression of peroxisomal and other genes (21). The analysis of expression of DNA repair genes was used to assess whether this end point could be used as a biomarker of oxidative stress to determine the source of DNA-damaging ROS in the carcinogenic mechanism of peroxisome proliferators. Here, two knockout mouse models that have been used to address the role of PPAR␣-dependent pathways and NADPH oxidase [PPAR␣-null (15) and p47phox-null (14), respectively] were used. Knockout and corresponding wild-type mice were fed control, or WY-containing (500 ppm) diets for 1 month, similar to the experimental design detailed above. Total RNA was isolated from liver samples, and expression of BER genes was compared. An increase in expression of DNA repair genes was observed in wild-type mice fed WY-14,643 for 1 month; however, no change was detected in PPAR␣-null animals (Fig. 2). Interestingly, in mice deficient in NADPH oxidase (p47phox-null), DNA repair genes were induced by WY-14,643 treatment as much as in wild-type animals (Fig. 3). Although gene expression analysis clearly demonstrated differences between groups, the measurements of total AP sites in genomic DNA from the same animals failed to show a biological response (Table 2). It was originally hypothesized that AOX, a peroxisomal enzyme that catalyzes -oxidation of fatty acids and reduces molecular oxygen to hydrogen peroxide, is a key source of ROS in livers of peroxisome proliferator-treated rodents (22). The AOX-null mice develop spontaneous tumors via mechanisms that potentially involve oxidative stress (16). Here, we analyzed liver samples from wild-type and AOX-null mice (3– 4 months of age) to assess BER gene expression and number of AP sites. Expression of long-patch BER genes was found to be up-regulated significantly in livers of naı¨ve AOX-null mice (Table 3) and was increased similarly to the effect of WY14,643. No difference in the number of AP sites in genomic DNA of wild-type and AOX-null mice was found (Table 2). DISCUSSION There are several primary points for discussion based on the data presented in this study:
1052
DNA REPAIR GENES AS A BIOMARKER OF OXIDATIVE STRESS
Table 1 Expression of DNA repair genes in livers of wild-type 129S4/Svter mice after treatment with control (NIH-07) or WY-14,643-containing diets Total RNA was isolated from liver samples and used in the RNase protection assay. The results are mean ⫾ SE (from 4 to 5 animals/group) of the relative expression of each gene as normalized to the expression of housekeeping gene L32. DNA repair gene
Control
WY-14,643 (500 ppm, 1 month)
OGG1, 8-oxoguanine DNA glycosylase 1 UNG, uracil DNA glycosylase MPG, N-methylpurine DNA glycosylase TDG, thymine DNA glycosylase NTH1, endonuclease III homolog 1 APE, apurinic/apyrimidinic endonuclease 1 Pol , polymerase (DNA directed), beta FEN1, flap structure-specific endonuclease 1 PCNA, proliferating cell nuclear antigen PARP, poly(ADP-ribose) polymerase XRCC1, X-ray repair complementing defective repair in Chinese hamster cells 1 Lig1, ligase I, DNA, ATP-dependent Lig3, ligase III, DNA, ATP-dependent MGMT, O6-methylguanine-DNA methyltransferase
3.9 ⫾ 0.3 3.8 ⫾ 0.3 9.3 ⫾ 0.7 6.9 ⫾ 0.4 3.2 ⫾ 0.4 13.4 ⫾ 0.3 9.1 ⫾ 0.1 1.3 ⫾ 0.4 30.8 ⫾ 1.3 25.5 ⫾ 1.8 6.0 ⫾ 0.5
8.8 ⫾ 0.6a 5.8 ⫾ 0.1a 22.9 ⫾ 2.1a 9.5 ⫾ 1.4a 4.0 ⫾ 0.2 27.9 ⫾ 1.0a 13.1 ⫾ 0.4a 4.9 ⫾ 0.8a 40.7 ⫾ 1.4a 25.4 ⫾ 0.3 6.6 ⫾ 0.3
1.7 ⫾ 0.4 10.0 ⫾ 1.2 36.1 ⫾ 2.1
10.6 ⫾ 1.4a 8.8 ⫾ 0.5 34.0 ⫾ 2.3
utility of measurements of oxidative DNA lesions between short term in vitro and in vivo experiments and chronic animal studies. If exposure to an environmental agent leads to production of massive amounts of oxidants in many cells simultaneously, the end effect most likely would be nonspecific cell killing, potentially leading to severe tissue injury, organ failure, and even death of the exposed subject. This scenario is often reproduced in vitro when cells are treated with
a Statistical difference (P ⬍ 0.05) from the corresponding control group by Student’s t-test.
(a) Is the issue of the utility of various biomarkers of oxidative stress to DNA as applied to in vivo studies on the role of oxidants in the mechanisms of environmental agent-induced carcinogenesis? (b) If reliable and sensitive biomarkers for oxidative DNA damage can be established, how can they be used to determine whether oxidative stress to DNA takes place when animals are treated with peroxisome proliferators? (c) What cellular and molecular sources of ROS are most likely involved in the peroxisome proliferator-induced carcinogenesis? Establishing a Link between Environmental Agent-induced Carcinogenesis and Oxidative DNA Damage. The possible scenarios for increased production of ROS after environmental exposure include: direct formation of free radicals [e.g., radiolysis of water by ionizing radiation (23)]; induction of metabolizing enzymes that may “leak” reactive intermediates [e.g., peroxisome proliferators, (24); and reaction with intracellular constituents to produce ROS and DNAreactive lipid peroxides [e.g., pentachlorophenol metabolites, (25, 26)]. ROS induce a variety of lesions in DNA, including chemical alterations of purine and pyrimidine bases, 2⬘-deoxyribose, and the formation of abasic (AP) sites and DNA strand breaks (2). Oxidation of guanine residues at the 8-position to produce 8-hydroxy-2⬘-deoxyguanosine is considered one of the most common forms of oxidative damage to DNA (27, 28). It was shown that this adduct is promutagenic because it could lead to induction of G:C to T:A transversions during DNA replication (29, 30). Improved analytical tools have advanced our knowledge of both exogenous DNA adducts induced by a variety of environmental factors and DNA lesions formed in the course of normal metabolic processes and thus deemed endogenous. Normal oxidative metabolism leads to constant endogenous production of ROS and formation of DNA damage, and the environmental agents that are thought to alter the oxidant/antioxidant balance would primarily affect the amounts, but not necessarily the chemical entity, of the adducts that persist, even under normal conditions. Thus, the very issue at stake in making a conclusion that any particular chemical agent leads to the formation of oxidative DNA damage is about knowing the precise endogenous or “control” levels of the adducts that are used as a “biomarker.” Remarkably, no consensus on the baseline levels of a commonly used biomarker such as 8-oxo-deoxyguanosine existed until very recent efforts by the European Standards Committee on Oxidative DNA Damage (11). Another important point is the profound difference in the potential
Fig. 2. PPAR␣ is critical for up-regulation of DNA repair gene expression by peroxisome proliferators. Expression of base excision DNA repair enzymes in livers of wild-type (⫹/⫹) or PPAR␣-null (⫺/⫺) mice fed control NIH-07 (CON) or WY-14,643containing diet (500 ppm, 1 month) was analyzed by the RNase protection assay with multi-probe template mBER-1 (A and C) and mBER-2 (B and D) representative data. The intensity of protected bands was quantified using phosphorimaging and normalized to the intensity of housekeeping genes. Data shown (C and D) are the results of densitometry analysis of images (mean values from four to five animals/group) from the experiments detailed in A and B, respectively.
1053
DNA REPAIR GENES AS A BIOMARKER OF OXIDATIVE STRESS
Fig. 3. Oxidant-producing NADPH oxidase plays no role in peroxisome proliferatorinduced oxidative DNA damage. Expression of base excision DNA repair enzymes in livers of wild type (⫹/⫹) or p47phox-null (⫺/⫺) mice fed control NIH-07 (CON) or WY-14,643-containing diet (500 ppm, 1 month) was analyzed by the RNase protection assay with multi-probe template mBER-1 (A and C) and mBER-2 (B and D) representative data. The intensity of protected bands was quantified using phosphorimaging and normalized to the intensity of housekeeping genes. Data shown (C and D) are the results of densitometry analysis of images (mean values from four to five animals/group) from the experiments detailed in A and B, respectively.
utility is rather limited. Most environmental agent exposures last a long time, and the doses are such that no immediate and even short-term adverse phenotype associated with increased production of oxidants can be detected. Nonetheless, chronic low-level exposures, and ensuing chronic low-level increase in ROS production and/or DNA damage, could have considerable long-term effects, such as cancer, but it is difficult to provide an undisputed link between exposure, ROS, and disease. Furthermore, the potential role for DNA repair that works tirelessly to remove both endogenous and exogenous DNA lesions, as a biomarker of DNA damage, has not been fully recognized. This study attempted to resolve the formidable difficulties in drawing a link between exposure to a chemical carcinogen and oxidative stress to DNA in a target organ. Here, the measurements of the damage to DNA per se (i.e., DNA adducts, AP sites, and strand breaks) were considered together with studies on changes in DNA repair gene expression (Scheme 1). Peroxisome proliferators have been shown to cause production of ROS using a variety of direct (e.g., electron spin resonance spectroscopy) and indirect measurements (8); however, the reported levels of oxidative DNA damage have been interpreted with caution because of high levels of adducts in control samples. In this study, DNA was isolated from livers of mice treated for 1 month with a potent peroxisome proliferator and rodent carcinogen WY-14,643. When DNA isolation was performed using a procedure that minimizes potential ex vivo oxidation of DNA bases, the levels of 8-oxoguanine in isolates from control samples were ⬍1 adduct/million guanine residues, which is within the range of baseline values reported recently by other laboratories (11). Most importantly, no increase in the amount of 8-oxoguanine was observed in WYtreated mice. It was reasoned that oxidized DNA bases, if produced in greater amounts because of exposure to WY-14,643, could be rapidly repaired and converted to AP sites or further to strand breaks. When the numbers of total, 3⬘-nicked or 5⬘-nicked, and intact AP sites in liver genomic DNA were compared between control and WY-treated mice, again no significant increase in these markers of oxidative DNA damage was detected (Fig. 1). One potential conclusion from these measurements is that there is no evidence for oxidative DNA damage in liver of peroxisome proliferator-treated mice. However, it is also possible that the analytical techniques that were used are simply not sensitive enough to detect subtle increases in oxidative DNA damage; or, the lesions that were considered are not the ones that are formed by peroxisome proliferators in vivo. Oxidative stress can lead to formation of scores of potentially mutagenic base modifications (2), and DNA lesion(s) that would demonstrate conclusively a significant increase in oxidative DNA damage in WY-treated animals are yet to be identified. Conversely, oxidative DNA damage may be increased, but through up-regulation of DNA repair similar steady-state numbers of oxidized DNA lesions are maintained. Because DNA repair machinery is
Table 2 The number of apurinic/apyrimidinic (AP) sites per 106 nucleotides in genomic DNA isolated from mouse whole liver samples DNA was isolated, and the number of AP sites in each frozen liver sample was determined as detailed in “Materials and Methods.” Data shown are mean ⫾ SD from four to six animals/group.
up to mM concentrations of pro-oxidants (e.g., H2O2), or test chemical agents, and the conclusions about the mechanism of action are based on the measurements of DNA adducts and other lesions that arise as a result of these unphysiological conditions. Although such studies provide useful “proof-of-principle” results attesting to a potential for oxidative DNA damage as a mechanism of carcinogenicity, their 1054
PPAR␣ (⫹/⫹) PPAR␣ (⫺/⫺) p47phox (⫹/⫹) p47phox (⫺/⫺) AOX (⫹/⫹) AOX (⫺/⫺) a
Control diet
WY-14,643 (500 ppm, 1 month)
6.5 ⫾ 1.3 6.2 ⫾ 0.5 7.0 ⫾ 2.1 6.9 ⫾ 2.0 5.2 ⫾ 0.7 5.0 ⫾ 0.6
5.3 ⫾ 0.5 6.6 ⫾ 4.0 7.5 ⫾ 1.5 6.7 ⫾ 2.9 NAa NAa
Wild-type and acyl-CoA oxidase (AOX)-null mice were not treated with WY-14,643.
DNA REPAIR GENES AS A BIOMARKER OF OXIDATIVE STRESS
Table 3 Expression of DNA repair genes in livers of wild-type (C57BL/6J) and AOX-null mice 3 to 4 months of age Total RNA was isolated from liver samples and used in the RNase protection assay. The results are mean ⫾ SE (from five to six animals/group) of the relative expression of each gene as normalized to the expression of housekeeping gene L32.
OGG1, 8-oxoguanine DNA glycosylase 1 UNG, uracil DNA glycosylase MPG, N-methylpurine DNA glycosylase TDG, thymine DNA glycosylase NTH1, endonuclease III homolog 1 APE, apurinic/apyrimidinic endonuclease 1 Pol , polymerase (DNA directed), beta FEN1, flap structure-specific endonuclease 1 PCNA, proliferating cell nuclear antigen PARP, poly(ADP-ribose) polymerase XRCC1, X-ray repair complementing defective repair in Chinese hamster cells 1 Lig1, ligase I, DNA, ATP-dependent Lig3, ligase III, DNA, ATP-dependent MGMT, O6-methylguanine-DNA methyltransferase
Wild-type (C57BL/6J)
AOX-null
4.3 ⫾ 0.4 3.9 ⫾ 0.1 4.2 ⫾ 0.2 5.9 ⫾ 0.3 5.5 ⫾ 0.7 7.7 ⫾ 0.6 13.2 ⫾ 0.7 1.9 ⫾ 0.4 26.2 ⫾ 1.2 21.1 ⫾ 1.7 9.5 ⫾ 0.8
7.5 ⫾ 0.2a 6.8 ⫾ 0.7a 7.8 ⫾ 0.7a 8.6 ⫾ 0.4a 5.1 ⫾ 0.7 19.4 ⫾ 1.3a 19.9 ⫾ 1.2a 6.4 ⫾ 1.0a 37.9 ⫾ 0.7a 23.9 ⫾ 1.5 11.3 ⫾ 0.7
2.6 ⫾ 0.4 17.5 ⫾ 1.9 9.3 ⫾ 0.6
8.8 ⫾ 0.6a 19.0 ⫾ 1.1 8.9 ⫾ 0.3
a Statistical difference (P ⬍ 0.05) from the corresponding control group by Student’s t test.
designed to deal with dozens of DNA adducts via a broad specificity of lesion-recognizing enzymes and several alternative repair pathways are available, it has been hypothesized that changes in repair could serve as a marker for DNA damage. Furthermore, because certain types of damage (e.g., oxidative lesions) are known to be predominantly removed by one of the branches of repair (e.g., base excision DNA repair), we reasoned that if a chemical treatment results in concordant and selective changes in gene expression for a particular repair mechanism, an inference for the presence of a unique type of DNA damage can be made. Gene Expression Changes in BER as a Marker of Oxidative Stress to DNA. It has been shown that peroxisome proliferators cause a time- and dose-dependent increase in expression of BER genes that correlates with the carcinogenic potency of a compound and is independent of changes in cell proliferation (13). BER is considered to be a predominant pathway for the removal of oxidized and many alkylated DNA bases and is initiated by an array of DNA glycosylases, which are often promiscuous as far as their substrate specificities are concerned (31, 32). After removal of a damaged base, a potentially mutagenic AP site is formed that can be further processed by AP endonuclease, followed by either short-patch BER or long-patch BER. One nucleotide replacement in a short-patch repair is dependent on the activity of Pol , PARP, and XRCC1/Lig3 complex (33). Long-patch BER occurs by the excision of at least 2 nucleotides, and often 6 –13 nucleotides, and DNA synthesis can be catalyzed by Pol  or by Pol ␦/⑀ and is dependent on the PCNA, FEN1, and Lig1 [Scheme 1 (33, 34)]. Several in vitro studies examining the preferential use of polymerases in BER, as well as a recent analysis of kinetic parameters, have suggested a predominant role for short-patch BER in the removal of oxidized DNA bases (35, 36). However, direct evidence that, after the excision of 8-oxoguanine or incision at an AP site, a significant proportion of the damage is processed by the long-patch BER in vivo, was presented recently (37). Importantly, in this study, a specific induction of long-patch BER genes was detected in vivo in rodent liver after exposure to WY-14,643 (Table 1). At the same time, no change in expression of short-patch BER genes, or MGMT, a repair protein that has no role in removal of oxidative DNA damage, was seen. Collectively, we conclude that a concordant marked induction of genes specific for the long-patch BER could be used as a sensitive and specific biomarker for chemically induced DNA damage in vivo when other markers directed at detection of oxidized DNA bases, AP sites, or strand breaks provide ambiguous evidence for oxidative stress to DNA.
Sources of Reactive Oxygen Species in the Mechanism of DNA Damage by Peroxisome Proliferators. Next, we tested the utility of BER gene expression changes as a biomarker for addressing an important mechanistic question in the carcinogenesis by peroxisome proliferators: What are the molecular and cellular sources for DNA damaging ROS? Several potential mechanisms for increased production of ROS in liver of peroxisome proliferator-treated rodents have been proposed, such as Kupffer cell NADPH oxidase and parenchymal cell peroxisomes (Scheme 2). Direct evidence for Kupffer cell and NADPH oxidase-mediated formation of ROS in liver in response to peroxisome proliferators was presented using NADPH oxidasedeficient p47phox-null mice (38). At the same time, -oxidation of fatty acids in peroxisomes reduces molecular oxygen to hydrogen peroxide, and this mechanism was postulated to be a key source of ROS in livers of peroxisome proliferator-treated rodents (22). Peroxisome proliferation is dependent on activation of PPAR␣, a nuclear receptor (39), and PPAR␣-null mice are resistant to phenotypic effects of peroxisome proliferators, such as cell proliferation and liver cancer (40). Thus, to address a potential role of NADPH oxidase and PPAR␣-mediated peroxisomal induction as the sources of DNAdamaging ROS, two knockout mouse models were used. Our results strongly suggest that DNA-damaging oxidants are generated by enzymes that are induced after activation of PPAR␣, such as those involved in lipid metabolism in peroxisomes, because WY14,643 treatment failed to induce expression of BER genes in livers of PPAR␣-null mice (Fig. 2). At the same time, studies in p47phox-null mice demonstrate that NADPH oxidase-generated ROS are not involved in DNA damage in liver (Fig. 3), although this pathway plays a key role in activation of Kupffer cell-mediated release of mitogenic cytokines (41). Importantly, the measurements of AP sites (Table 2) failed to provide any evidence of DNA damage and thus could not be used as a biomarker for mechanistic studies in peroxisome proliferator-induced carcinogenesis. To further investigate the role of lipid-metabolizing proteins in the peroxisome as a potential source of ROS in liver carcinogenesis, AOX-null mice were used (16). AOX is a major oxidant-producing protein in the peroxisome, and it is markedly induced after treatment
Scheme 1. The schematic diagram of base excision repair pathways for removal of oxidized DNA lesions formed as a result of exposure to chemical agents that cause oxidative stress. The sequence of modifications of DNA is depicted in bold, whereas the molecular events that occur at each step are shown in italics. The gene products that participate at each step of repair and their expressions are provided in boxes. ⴱ, expression of genes significantly increased in WY-14,643-treated mice.
1055
DNA REPAIR GENES AS A BIOMARKER OF OXIDATIVE STRESS
REFERENCES
Scheme 2. Molecular mechanisms and cell-cell interactions in peroxisome proliferatorinduced effects in rodent liver. Kupffer cells have been shown to be rapidly activated by peroxisome proliferators, generate oxidants, and release mitogenic cytokines that stimulate proliferation of parenchymal cells. Concomitantly, activation of PPAR␣-mediated events in parenchymal cells leads to potentiation of cell proliferation signaling and production of ROS by peroxisomal enzymes. In this study, the cellular and molecular mechanisms of production of DNA-damaging ROS were studied. Data presented here demonstrate that PPAR␣-dependent peroxisomal proteins, but not Kupffer cell NADPH oxidase, are responsible for peroxisome proliferator-induced DNA damage in mouse liver.
with peroxisome proliferators (42). In addition to AOX, there are several additional oxidant-producing enzymes that could possibly be involved, such as oxidases located in the peroxisome matrix (24), CYP4A family enzymes (43), and other proteins (44). AOX-null mice, in the absence of any treatment, develop severe steatohepatitis that is followed by increased hepatic H2O2 levels, hepatocyte death, and hepatocellular regeneration accompanied by profound generalized spontaneous peroxisome proliferation and increased mRNA levels of genes that are regulated by PPAR␣, and ultimately leading to liver carcinomas (45). Recently, it was shown that liver tumors from naive AOX-null mice or wild-type mice treated with WY-14,643 have similar gene expression profiles (46). Furthermore, it was demonstrated by proteomics analysis that global protein profiles of livers from AOX-null mice and WY-treated wild-type mice look more similar to each other than to wild-type untreated mice (47). Our data demonstrate that changes in BER gene expression in livers of AOXnull mice (3– 4 months of age) are similar to that of the wild-type animals that received WY-14,643 for 1 month (Table 3), whereas no change in AP sites was observed (Table 2). These results corroborate the hypothesis that dysregulation of hepatic lipid metabolism caused by exogenous peroxisome proliferators or overabundance of natural substrates (e.g., in AOX-null mice) leads to hepatic oxidative stress and DNA damage. Further studies are required to determine which of the hydrogen peroxide-generating peroxisomal oxidases is responsible for oxidative DNA damage after induction of peroxisomes in liver parenchymal cells. In conclusion, these studies provide further evidence for up-regulation of BER genes, specifically of the long-patch pathway, in peroxisome proliferator-treated animals. We also show that this concordant induction in the mechanism of removal of oxidized DNA lesions can be used as a sensitive and specific marker for oxidative stress to DNA when other commonly used indicators, such as 8oxoguanine and AP sites, fail to provide evidence for DNA damage. Importantly, gene expression for BER can be used to address the mechanisms of formation of DNA-damaging ROS in chronic chemical exposure animal studies.
ACKNOWLEDGMENTS We thank Dr. Jun Nakamura (University of North Carolina-Chapel Hill) for providing reagents for an AP site assay.
1. Friedberg, E. C. DNA damage and repair. Nature (Lond.), 421: 436 – 440, 2003. 2. Marnett, L. J. Oxyradicals and DNA damage. Carcinogenesis (Lond.), 21: 361–370, 2000. 3. Hussain, S. P., Hofseth, L. J., and Harris, C. C. Radical causes of cancer. Nat. Rev. Cancer, 3: 276 –285, 2003. 4. Cooke, M. S., Evans, M. D., Dizdaroglu, M., and Lunec, J. Oxidative DNA damage: mechanisms, mutation, and disease. FASEB J., 17: 1195–1214, 2003. 5. Klaunig, J. E., Xu, Y., Isenberg, J. S., Bachowski, S., Kolaja, K. L., Jiang, J., Stevenson, D. E., and Walborg, E. F., Jr. The role of oxidative stress in chemical carcinogenesis. Environ. Health Perspect. 106 Suppl 1: 289 –295, 1998. 6. Helbock, H. J., Beckman, K. B., Shigenaga, M. K., Walter, P. B., Woodall, A. A., Yeo, H. C., and Ames, B. N. DNA oxidation matters—the HPLC-electrochemical detection assay of 8-oxo-deoxyguanosine and 8-oxo-guanine. Proc. Natl. Acad. Sci. USA, 95: 288 –293, 1998. 7. Reddy, J. K., Azarnoff, D. L., and Hignite, C. E. Hypolipidaemic hepatic peroxisome proliferators form a novel class of chemical carcinogens. Nature (Lond.), 283: 397–398, 1980. 8. Rusyn, I., Rose, M. L., Bojes, H. K., and Thurman, R. G. Novel role of oxidants in the molecular mechanism of action of peroxisome proliferators. Antiox. Redox Signal., 2: 607– 621, 2000. 9. Collins, A. R. Oxidative DNA damage, antioxidants, and cancer. Bioessays, 21: 238 –246, 1999. 10. ESCODD. European Standards Committee on Oxidative DNA Damage. Comparative analysis of baseline 8-oxo-7,8-dihydroguanine in mammalian cell DNA, by different methods in different laboratories: an approach to consensus. Carcinogenesis (Lond.), 23: 2129 –2133, 2002. 11. ESCODD. European Standards Committee on Oxidative DNA Damage. Measurement of DNA oxidation in human cells by chromatographic and enzymic methods. Free Radic. Biol. Med. 34: 1089 –1099, 2003. 12. Samson, L., and Cairns, J. A new pathway for DNA repair in Escherichia coli. Nature (Lond.), 267: 281–283, 1977. 13. Rusyn, I., Denissenko, M. F., Wong, V. A., Butterworth, B. E., Cunningham, M. L., Upton, P. B., Thurman, R. G., and Swenberg, J. A. Expression of base excision repair enzymes in rat and mouse liver is induced by peroxisome proliferators and is dependent upon carcinogenic potency. Carcinogenesis (Lond.), 21: 2141–2145, 2000. 14. Jackson, S. H., Gallin, J. I., and Holland, S. M. The p47phox mouse knock-out model of chronic granulomatous disease. J. Exp. Med., 182: 751–758, 1995. 15. Lee, S. S., Pineau, T., Drago, J., Lee, E. J., Owens, J. W., Kroetz, D. L., FernandezSalguero, P. M., Westphal, H., and Gonzalez, F. J. Targeted disruption of the alpha isoform of the peroxisome proliferator- activated receptor gene in mice results in abolishment of the pleiotropic effects of peroxisome proliferators. Mol. Cell. Biol., 15: 3012–3022, 1995. 16. Fan, C. Y., Pan, J., Chu, R., Lee, D., Kluckman, K. D., Usuda, N., Singh, I., Yeldandi, A. V., Rao, M. S., Maeda, N., and Reddy, J. K. Hepatocellular and hepatic peroxisomal alterations in mice with a disrupted peroxisomal fatty acyl-coenzyme A oxidase gene. J. Biol. Chem., 271: 24698 –24710, 1996. 17. Nakamura, J., La, D. K., and Swenberg, J. A. 5⬘-Nicked apurinic/apyrimidinic sites are resistant to -elimination by -polymerase and are persistent in human cultured cells after oxidative stress. J. Biol. Chem., 275: 5323–5328, 2000. 18. Beckman, K. B., Saljoughi, S., Mashiyama, S. T., and Ames, B. N. A simpler, more robust method for the analysis of 8-oxoguanine in DNA. Free Radic. Biol. Med., 29: 357–367, 2000. 19. Nakamura, J., Walker, V. E., Upton, P. B., Chiang, S. Y., Kow, Y. W., and Swenberg, J. A. Highly sensitive apurinic/apyrimidinic site assay can detect spontaneous and chemically induced depurination under physiological conditions. Cancer Res., 58: 222–225, 1998. 20. Lin, P. H., Nakamura, J., Yamaguchi, S., Asakura, S., and Swenberg, J. A. Aldehydic DNA lesions induced by catechol estrogens in calf thymus DNA. Carcinogenesis (Lond.), 24: 1133–1141, 2003. 21. Gonzalez, F. J., Peters, J. M., and Cattley, R. C. Mechanism of action of the nongenotoxic peroxisome proliferators: role of the peroxisome proliferator-activator receptor alpha. J. Natl. Cancer Inst., 90: 1702–1709, 1998. 22. Reddy, J. K., and Lalwani, N. D. Carcinogenesis by hepatic peroxisome proliferators: evaluation of the risk of hypolipidemic drugs and industrial plasticizers in humans. CRC Crit. Rev. Toxicol., 12: 1–58, 1983. 23. von Sonntag, C. The Chemical Basis of Radiation Biology. London: Taylor and Francis, 1987. 24. Yeldandi, A. V., Rao, M. S., and Reddy, J. K. Hydrogen peroxide generation in peroxisome proliferator-induced oncogenesis. Mutat. Res., 448: 159 –177, 2000. 25. Naito, S., Ono, Y., Somiya, I., Inoue, S., Ito, K., Yamamoto, K., and Kawanishi, S. Role of active oxygen species in DNA damage by pentachlorophenol metabolites. Mutat. Res., 310: 79 – 88, 1994. 26. Lin, P. H., Nakamura, J., Yamaguchi, S., La, D. K., Upton, P. B., and Swenberg, J. A. Induction of direct adducts, apurinic/apyrimidinic sites, and oxidized bases in nuclear DNA of human HeLa S3 tumor cells by tetrachlorohydroquinone. Carcinogenesis (Lond.), 22: 635– 639, 2001. 27. Kasai, H., Crain, P. F., Kuchino, Y., Nishimura, S., Ootsuyama, A., and Tanooka, H. Formation of 8-hydroxyguanine moiety in cellular DNA by agents producing oxygen radicals and evidence for its repair. Carcinogenesis (Lond.), 7: 1849 –1851, 1986. 28. Nishimura, S. 8-Hydroxyguanine, DNA adduct formed by oxygen radicals. In: M.
1056
DNA REPAIR GENES AS A BIOMARKER OF OXIDATIVE STRESS
29.
30.
31. 32. 33. 34.
35.
36.
37. 38.
39.
Dizdaroglu and A. Karakaya (eds.), Advances in DNA Damage and Repair, pp. 319 –327. New York: Kluwer Academic/Plenum Publishers, 1999. Michaels, M. L., and Miller, J. H. The GO system protects organisms from the mutagenic effect of the spontaneous lesion 8-hydroxyguanine (7,8-dihydro-8-oxoguanine). J. Bacteriol., 174: 6321– 6325, 1992. Thomas, D., Scot, A. D., Barbey, R., Padula, M., and Boiteux, S. Inactivation of OGG1 increases the incidence of G:C to T:A transversions in Saccharomyces cerevisiae: evidence for endogenous oxidative damage to DNA in eukaryotic cells. Mol. Gen. Genet., 254: 171–178, 1997. Scharer, O. D., and Jiricny, J. Recent progress in the biology, chemistry and structural biology of DNA glycosylases. Bioessays, 23: 270 –281, 2001. Wood, R. D., Mitchell, M., Sgouros, J., and Lindahl, T. Human DNA repair genes. Science (Wash. DC), 291: 1284 –1289, 2001. Krokan, H. E., Nilsen, H., Skorpen, F., Otterlei, M., and Slupphaug, G. Base excision repair of DNA in mammalian cells. FEBS Lett., 476: 73–77, 2000. Klungland, A., and Lindahl, T. Second pathway for completion of human DNA base excision-repair: reconstitution with purified proteins and requirement for DNase IV (FEN1). EMBO J., 16: 3341–3348, 1997. Dianov, G. L., Souza-Pinto, N., Nyaga, S. G., Thybo, T., Stevnsner, T., and Bohr, V. A. Base excision repair in nuclear and mitochondrial DNA. Prog. Nucleic Acid Res. Mol. Biol., 68: 285–297, 2001. Sokhansanj, B. A., Rodrigue, G. R., Fitch, J. P., and Wilson, D. M., III A quantitative model of human DNA base excision repair. I. Mechanistic insights. Nucleic Acids Res., 30: 1817–1825, 2002. Sattler, U., Frit, P., Salles, B., and Calsou, P. Long-patch DNA repair synthesis during base excision repair in mammalian cells. EMBO Rep., 4: 363–367, 2003. Rusyn, I., Kadiiska, M. B., Dikalova, A., Kono, H., Yin, M., Tsuchiya, K., Mason, R. P., Peters, J. M., Gonzalez, F. J., Segal, B. H., Holland, S. M., and Thurman, R. G. Phthalates rapidly increase production of reactive oxygen species in vivo: role of Kupffer cells. Mol. Pharmacol., 59: 744 –750, 2001. Issemann, I., and Green, S. Activation of a member of the steroid hormone receptor superfamily by peroxisome proliferators. Nature (Lond.), 347: 645– 650, 1990.
40. Peters, J. M., Cattley, R. C., and Gonzalez, F. J. Role of PPAR␣ in the mechanism of action of the nongenotoxic carcinogen and peroxisome proliferator Wy-14,643. Carcinogenesis (Lond.), 18: 2029 –2033, 1997. 41. Rusyn, I., Yamashina, S., Segal, B. H., Schoonhoven, R., Holland, S. M., Cattley, R. C., Swenberg, J. A., and Thurman, R. G. Oxidants from nicotinamide adenine dinucleotide phosphate oxidase are involved in triggering cell proliferation in the liver due to peroxisome proliferators. Cancer Res., 60: 4798 – 4803, 2000. 42. Nemali, M. R., Usuda, N., Reddy, M. K., Oyasu, K., Hashimoto, T., Osumi, T., Rao, M. S., and Reddy, J. K. Comparison of constitutive and inducible levels of expression of peroxisomal -oxidation and catalase genes in liver and extrahepatic tissues of rat. Cancer Res., 48: 5316 –5324, 1988. 43. Reddy, J. K., and Hashimoto, T. Peroxisomal -oxidation and peroxisome proliferator-activated receptor ␣: an adaptive metabolic system. Annu. Rev. Nutr., 21: 193– 230, 2001. 44. Cherkaoui-Malki, M., Meyer, K., Cao, W. Q., Latruffe, N., Yeldandi, A. V., Rao, M. S., Bradfield, C. A., and Reddy, J. K. Identification of novel peroxisome proliferator-activated receptor ␣ (PPAR␣) target genes in mouse liver using cDNA microarray analysis. Gene Expr., 9: 291–304, 2001. 45. Fan, C. Y., Pan, J., Usuda, N., Yeldandi, A. V., Rao, M. S., and Reddy, J. K. Steatohepatitis, spontaneous peroxisome proliferation and liver tumors in mice lacking peroxisomal fatty acyl-CoA oxidase. Implications for peroxisome proliferatoractivated receptor ␣ natural ligand metabolism. J. Biol. Chem., 273: 15639 –15645, 1998. 46. Meyer, K., Lee, J. S., Dyck, P. A., Cao, W. Q., Rao, M. S., Thorgeirsson, S. S., and Reddy, J. K. Molecular profiling of hepatocellular carcinomas developing spontaneously in acyl-CoA oxidase deficient mice: comparison with liver tumors induced in wild-type mice by a peroxisome proliferator and a genotoxic carcinogen. Carcinogenesis (Lond.), 24: 975–984, 2003. 47. Chu, R., Zhang, W., Lim, H., Yeldandi, A. V., Herring, C., Brumfield, L., Reddy, J. K., and Davison, M. Profiling of acyl-CoA oxidase-deficient and peroxisome proliferator Wy14,643-treated mouse liver protein by surface-enhanced laser desorption/ionization Protein Chip Biol. Syst. Gene Expr., 10: 165–177, 2002.
1057
