- Email: [email protected]
Oxidative DNA damage in the in utero initiation of postnatal neurodevelopmental deficits by normal fetal and ethanol-enhanced oxidative stress in oxoguanine glycosylase 1 knockout mice
Author's Accepted Manuscript
Oxidative DNA damage in the in utero initiation of postnatal neurodevelopmental deficits by normal fetal and ethanol-enhanced oxidative stress in oxoguanine glycosylase 1 (ogg1) knockout mice Lutfiya Miller-Pinsler, Daniel J. Pinto, Peter G. Wells www.elsevier.com/locate/freeradbiomed
PII: DOI: Reference:
S0891-5849(14)00445-6 http://dx.doi.org/10.1016/j.freeradbiomed.2014.09.026 FRB12163
To appear in:
Free Radical Biology and Medicine
Received date: 25 February 2014 Revised date: 15 September 2014 Accepted date: 24 September 2014 Cite this article as: Lutfiya Miller-Pinsler, Daniel J. Pinto, Peter G. Wells, Oxidative DNA damage in the in utero initiation of postnatal neurodevelopmental deficits by normal fetal and ethanol-enhanced oxidative stress in oxoguanine glycosylase 1 (ogg1) knockout mice, Free Radical Biology and Medicine, http://dx.doi.org/10.1016/j.freeradbiomed.2014.09.026 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
15th September 2014 Oxidative DNA damage in the in utero initiation of postnatal neurodevelopmental deficits by normal fetal and ethanol-enhanced oxidative stress in oxoguanine glycosylase 1 (ogg1) knockout mice Running title: Fetal DNA repair & ethanol behavioural deficits Lutfiya Miller-Pinsler†, Daniel J. Pinto† and Peter G. Wells*† *Division of Biomolecular Sciences, Faculty of Pharmacy, and †Department of Pharmacology and Toxicology, Faculty of Medicine University of Toronto Toronto, Ontario, Canada
Number of: • Text pages (excluding references, footnotes, tables and figure legends): 19 • Tables: 0 • Figures: 4 plus 2 supplemental online figures • References: 35 • Words: - Abstract: 150 - Introduction: 683 - Discussion: 1,299 Corresponding Author: Peter G. Wells, Pharm.D. Division of Biomolecular Sciences Faculty of Pharmacy, University of Toronto 144 College St., Toronto, Ontario, Canada M5S 3M2 Tel. 416-978-3221; Email: [email protected]
a. This work was supported by a grant from the Canadian Institutes of Health Research (CIHR). The authors declare no competing financial interests.
Non-standard Abbreviations: 8-oxodGuo, 8-oxo-2’-deoxyguanosine; FASD, fetal alcohol spectrum disorders; OGG1, oxoguanine glycosylase 1; PAT, passive avoidance test; PBN, alpha-phenyl-N-tert-butylnitrone; GD, gestational day; EtOH, ethanol; ROS, reactive oxygen species
1
ABSTRACT Studies in mice with deficient antioxidative enzymes have shown that physiological levels of reactive oxygen species (ROS) can adversely affect the developing embryo and fetus. Herein, DNA repair-deficient progeny of oxoguanine glycosylase 1 (ogg1) knockout mice lacking repair of the oxidative DNA lesion 8-oxo-2’-deoxyguanosine (8-oxodGuo) exhibited enhanced postnatal neurodevelopmental deficits, revealing the pathogenic potential of 8-oxodGuo initiated by physiological ROS production in fetal brain, and providing the first evidence of a pathological phenotype for ogg1 knockout mice. Moreover, when exposed in utero to ethanol (EtOH), ogg1 knockout progeny exhibited higher levels of 8-oxodGuo in fetal brain and more severe postnatal neurodevelopmental deficits than wild-type littermates, both of which were blocked by pretreatment with the free radical trapping agent phenylbutylnitrone. These results suggest ROSinitiated DNA oxidation, as distinct from altered signal transduction, contributes to neurodevelopmental deficits caused by in utero EtOH exposure, and fetal DNA repair is a determinant of risk. Key Words: Ethanol; phenylbutylnitrone; PBN; passive avoidance; reactive oxygen species; DNA oxidation; OGG1 mice; behavioural deficits; neurodevelopmental deficits
2
INTRODUCTION Embryonic or fetal formation of reactive oxygen species (ROS) can cause oxidative stress, which can potentially dysregulate development by altering signal transduction pathways or oxidatively damaging cellular macromolecules including proteins, lipids and DNA, resulting in structural birth defects, in utero or neonatal death, and postnatal neurodevelopmental deficits (Hansen and Harris, 2013; Wells et al., 2009). In untreated mutant mice, progeny with deficiencies in glucose-6-phosphate dehydrogenase (G6PD) or catalase, which are necessary for ROS detoxification, exhibit a higher incidence of structural birth defects and in utero and neonatal death (Abramov and Wells, 2011; Nicol et al., 2000), suggesting that physiological levels of ROS formation can be embryopathic. It has been estimated that the cause of human birth defects is unknown in up to 70% of cases (Hansen and Harris, 2013), and physiological ROS formation may contribute to these adverse developmental outcomes, particularly when the progeny are deficient in pathways important for ROS detoxification, or possibly the repair of oxidatively damaged DNA. The consequences of in utero EtOH exposure are well characterized, and include a spectrum of anomalies termed the Fetal Alcohol Spectrum Disorder (FASD), which includes both structural and functional birth defects, with the complete phenotype including characteristic craniofacial dysmorphology, growth retardation and behavioural deficits (Jones, 2011). The estimated incidence is 1 out of 100 live births in Canada (Stade et al., 2009), and is postulated to be one of the leading preventable causes of neurodevelopmental disorders in the Western world (Mattson et al., 2011). In humans, prenatal EtOH exposure has been shown to initiate a broad spectrum of behavioural anomalies in children including decreased IQ, motor coordination deficits, hyperactivity and deficits in executive function, verbal language, memory and attention
3
(Mattson et al., 2011). The mechanisms underlying these deficits are unclear, however enhanced ROS formation has been implicated (Brocardo et al., 2012; Miller et al., 2013), and EtOH can induce embryonic ROS-generating NADPH oxidase (NOX) enzymes (Dong et al., 2010), and may also enhance ROS via free radical formation during its metabolism by cytochromes P450 (Koop, 2006). Oxoguanine glycosylase 1 (OGG1) is a critical DNA repair enzyme involved in the excision and repair of the 8-oxo-2-deoxyguanosine (8-oxodGuo) lesion (Boiteux and Radicella, 2000), a biomarker of oxidative stress as well as a developmentally pathogenic lesion (McCallum et al., 2011; Wells et al., 2009; Wong et al., 2008). OGG1 may be particularly important in the developing brain, where its activity is double that in maternal tissues (Wong et al., 2008). The progeny of CD-1 mice exposed in utero to EtOH exhibited an increase in oxidatively damaged DNA (8-oxodGuo) in fetal brain DNA, as well as postnatal learning and memory deficits (Miller et al., 2013). Similarly, CD-1 mice exposed in utero to the ROSinitiating drug methamphetamine exhibit increased fetal brain levels of 8-oxodGuo as well as postnatal motor coordination deficits (Jeng et al., 2005). The pathogenic role of 8-oxodGuo in methamphetamine-initiated neurodevelopmental deficits was revealed in studies using knockout mice lacking ogg1 (Wong et al., 2008) or csb (McCallum et al., 2011), which also repairs the 8oxodGuo lesion. In these studies, knockout progeny exposed in utero to methamphetamine exhibited an ogg1 or csb gene dose-dependent increase in 8-oxodGuo in fetal brain of +/- and -/pups compared to wild-type littermates, and increased motor coordination deficits measured by rotarod testing. ROS-initiating teratogens in addition to methamphetamine, including phenytoin and thalidomide, similarly increase 8-oxodGuo formation in the developing embryo and fetus (Abramov and Wells, 2011; Lee et al., 2011; Parman et al., 1999; Winn and Wells, 1995).
4
Taken together, these data suggest that 8-oxodGuo is a pathogenic molecular lesion in the embryo, and ogg1 variability is a determinant of risk for behavioural deficits caused by oxidative stress. However, there is no published evidence of a phenotype in these ogg1 knockout mice, so the developmental importance of fetal OGG1 is similarly not known. Moreover, the roles of oxidatively damaged DNA and OGG1 in EtOH-initiated behavioural deficits are similarly unknown. Herein, we sought to determine the pathogenic contribution of ROS-mediated formation of the 8-oxodGuo lesion, and the protective role of fetal DNA repair (Figure 1) in the development of untreated mice, and in the mechanism of EtOH-initiated behavioural deficits, using DNA repair-deficient ogg1 knockout mice. Our results in untreated mice provide the first demonstration of a behavioural phenotype for the ogg1 knockout mouse, suggesting that even the low level of fetal DNA oxidation from normal physiological processes can be developmentally toxic when DNA repair is inadequate. Our EtOH studies reveal that deficient fetal DNA repair and enhanced 8-oxodGuo levels contribute to the mechanism of EtOH-initiated postnatal behavioural deficits.
5
MATERIALS AND METHODS
Chemicals. Alpha-phenyl-N-tert-butylnitrone (phenylbutylnitrone, PBN), minimum 98% purity by gas chromatography (GC), and all other reagents unless otherwise specified were purchased from Sigma Aldrich (St. Louis, MO). Saline (0.9 %, sterile) was purchased from Baxter Corporation (Mississauga, ON) and ethanol from Commercial Alcohol Inc. (Brampton, ON).
Animals and diet. All animal protocols were approved by the institutional animal care committee in conformance with the guidelines established by the Canadian Council on Animal Care. Ogg1 knockout mice on a 129SV/C57BL/6J background strain were originally generated by Klungland and coworkers (Klungland et al., 1999), and generously provided by Dr. Tomas Lindahl (Imperial Cancer Research Fund, UK) through Dr. Christi A. Walter at the University of Texas Health Science Center at San Antonio. Mice were housed in vented plastic cages from Allentown, Inc. (Allentown, NJ) with ground corncob bedding (Bed-O’Cobs Laboratory Animal Bedding, The Andersons Industrial Products Group, Maumee, OH). Mouse cages were maintained in a room with controlled light (14 hr light-10 hr dark cycle) and climate (20ºC, 50% humidity), and provided with rodent chow (Harlan Labs: 2018, Harlan Teklad, Montreal, QC) and tap water ad libitum. Mice were acclimatized for 1 week prior to use. Virgin +/- ogg1 females were mated with a +/- male overnight from 5:00 P.M. to 9:00 A.M. by housing one male in a cage containing 1-3 females. Heterozygous matings were employed to generate progeny of all 3 genotypes (+/+, +/-, -/-) within the same litter. The presence of a vaginal plug the next morning was designated as gestational day (GD) 1, and plugged females were separated, weighed and housed in groups of up to four dams per cage until use.
6
Genotyping. DNA was isolated from tail snips of fetal progeny by heating the sample at 95°C in 75 µL alkaline lysis reagent (25 mM NaOH, 0.2 mM disodium EDTA, pH 12) for 1 hr. Samples were then neutralized by addition of 75 µL of neutralizing buffer (40 mM Tris-HCl, pH 5), and genotyped using a PCR-based assay. Primers (synthesized by The Center for Applied Genomics, The Hospital for Sick Children, Toronto, ON) used to amplify the 500 base pair (bp) band for the ogg1 gene were: ogg1-sense (5’-ACTGCATCTGCTTAATGGCC-3’) (forward primer) and ogg1-antisense (5’-CGAAGGTCAGCACTGAACAG-3’) (reverse primer). Primers used to amplify the 300 bp band for the neo-cassette responsible for disruption of the ogg1 gene in the ogg1 knockout mice were neo-sense (5’-CTGAATGAACTGCAGGACGA-3’) (forward primer) and neo-antisense (5’-CTCTTCGTCCAGATCATCCT-3’) (reverse primer). PCR reaction conditions were: 8 ȝl genomic DNA, 5 ȝl per sample of 10X Hotstart Buffer (Fermentas Life Sciences, Burlington, ON), 3 ȝl per sample of MgCl2 (Fermentas Life Sciences), 0.25 ȝl Maxima Hotstart Taq polymerase (Fermentas Life Sciences), 1 ȝl per sample of 10 mM deoxyribonucleotides (dNTP) (Fermentas Life Sciences), 0.75 ȝL per sample of each 20 ȝM primer and 29.75 ȝl per sample of ddH2O for a final volume of 50 ȝl. Cycling conditions were: 95°C for 5 min; 35 cycles of: 94°C for 1 min, 55°C for 1.5 min, 72°C for 2 min and completed with a final extension at 72°C for 10 min and then placed on hold at 4°C. 6X DNA loading dye (0.03% bromophenol blue/0.03% xylene cyanol FF) was added to each sample. The PCR products were separated on a gel consisting of 1.5% (w/v) agarose, 40 mM Tris, 19.4 mM glacial acetic acid, 2.5 mM EDTA and 8 ȝl ethidium bromide. The agarose gel was run at 100 V for 60 min, and then visualized and photographed under an ultraviolet light (Supplemental Figure S1).
7
Herein, we refer to “gene dose-dependent” not as the dose of the drug, but rather to the dose of the gene (zero, one or two alleles). Dosing for measurement of DNA oxidation and postnatal behavioural testing. On GD 17 (within the fetal period), pregnant +/- ogg1 dams were pretreated with PBN (40 mg/kg i.p.) or its 0.79% saline vehicle control at 9:00 AM using a 26 gauge (G) 3/8 needle. Thirty min later, dams were treated with a 25% v/v solution of EtOH (2 g/kg i.p.) or its 0.9% saline vehicle. We based the 2 g/kg EtOH dose in part on a previously published neurodevelopmental study (Miller et al., 2013) in which a 4 g/kg dose was used successfully in an outbred CD-1 strain. However, the 4 g/kg dose in inbred genetically modified ogg1 mice (129SV/C57BL/6J) proved too maternally toxic for this strain (increased maternal lethality and reduced body weight). Accordingly, the dose was reduced by half, which eliminated the maternal toxicity, and the 2 g/kg dose was used in the current study. Food was removed from the cages for 6 hr post injection to avoid potential effects on EtOH pharmacokinetics. One group of pregnant dams was sacrificed 6 hr after EtOH exposure, fetal brains were removed, snap frozen and stored at -80°C until analysis for DNA oxidation. A second group was allowed to deliver spontaneously for postnatal behavioural testing of their EtOH-exposed offspring. Three to four litters per treatment group were used to examine fetal brain DNA oxidation.
DNA extraction and digestion. DNA was extracted from GD 17 fetal brains using a modified version of a chaotropic sodium iodide (NaI) method previously described (Ravanat et al., 2002). Briefly, samples were homogenized in 1 mL of cold lysis buffer (320 mM sucrose, 5 mM MgCl2, 10mM Tris, 0.1 mM desferoxamine, 1% Triton X-100, pH 7.5) and centrifuged twice at 10,000
8
X g for 10 min at 4°C to isolate the nuclear pellet containing DNA. The pellet was then incubated with 200 uL enzyme reaction solution (1% w/v SDS, 5 mM EDTA-NA2, 0.15 mM desferoxamine 10 mM Tris-HCl pH 8.0) for 1 hr at 50ºC with RNase A/T1 mix (624 and 312 U/ml final activities, respectively). Proteinase K (1.8 mg/ml final concentration) was added for an additional 1 hr. The DNA pellet obtained after extraction was then washed 5 times with 70% EtOH. Samples were resuspended in 200 ȝl of sodium acetate buffer (20 mM, pH 4.8) and sonicated into solution. DNA purity was determined by measuring the absorbance ratio at 260/280 nm of 2 ȝL of sample in Na-acetate buffer (total volume 200 ȝL). Samples were digested with nuclease P1 (5 U/sample, 1 hr, 37°C) and calf intestinal alkaline phosphatase (6 U/sample, 1 hr, 37°C) and filtered through Amicon UltraTM filter units (YM-10, 10,000 MW cutoff; Millipore, Billerica, MA, USA) to remove DNA digestion enzymes and large particulates.
Quantification of oxidative damage to DNA using 8-OHdG ELISA. Oxidatively damaged DNA was quantified using the 8-OHdG ELISA kit (JaICA, Fukuroi, Japan) according to the manufacturer’s instructions.
Passive avoidance test. At 6, 9, 12 and 16 weeks of age, ogg1 wild-type, and +/- and -/knockout offspring were tested using a passive avoidance test as previously described (Miller et al., 2013). Briefly, the test apparatus consisted of a plastic box with two equal sized chambers, one with clear plastic walls designated as the ‘light chamber’, and another with red walls designated as the ‘dark chamber’, separated by a guillotine door. The floors of both chambers
9
consisted of stainless steel rods, wherein the rods of only the dark chamber were connected to a power supply that provided a mild 1.0 milliampere (mA) shock. In trial #1, the mouse was placed in the light chamber and was allowed to explore for 20 sec. The guillotine door was lifted, and once the mouse entered the dark chamber, the guillotine door was closed, and a shock of 1.0 mA was administered for 4 sec. The mouse was then returned back to its home cage, the apparatus was wiped with 70% EtOH to remove any olfactory cues, and the same procedure was repeated 24 and 48 hr later for trials #2 and #3, respectively. The time it took the mouse to enter the dark chamber in both trials after opening the guillotine door was recorded as “latency to enter the dark chamber”. To examine the effect of treatment and ogg1 genotype on learning, the latency to enter the dark chamber was plotted for trial 3 only. After trial #3, the pups were weighed at 6, 9 and 12 weeks to assess whether or not in utero EtOH exposure during the fetal period had any effect on growth. Six to eight litters per treatment group were used to examine learning and memory using the passive avoidance test.
Statistical analysis. Statistical analysis was performed using GraphPad Prism, Version 5 (GraphPad Software, Inc., San Diego, CA). DNA oxidation data, passive avoidance data and fetal weight data were analyzed using a two-way analysis of variance (ANOVA) with a post-hoc Bonferroni test. The minimal significance level used throughout was p < 0.05.
10
RESULTS Maternal Effects There were no treatment- or genotype-dependent differences in maternal weight (data not shown). The single 2 g/kg dose of EtOH resulted in temporary sedation lasting approximately 2 hours, after which time the dam recovered and behaved normally. Litter sizes ranged from 3 – 9 pups, and no pups were found dead. The pups were cared for by their biological mother for 3 weeks until they were weaned, at which point they lived among same-sex littermates.
DNA oxidation in fetal brains EtOH-exposed +/+, +/- and -/- ogg1 progeny exhibited 2.4-fold (p < 0.001), 2.5-fold (p < 0.001) and 2.3-fold (p < 0.001) increases in oxidatively damaged DNA, measured as the 8-oxodGuo lesion, in fetal brain compared to their respective, genotypically matched saline-exposed littermates (Figure 2). PBN pretreatment of saline-exposed fetuses did not change baseline DNA oxidation. Pretreatment with PBN decreased EtOH-enhanced DNA oxidation in fetal brains by 46% in +/+ ogg1 littermates (p < 0.001), 23% in +/- littermates (marginal, p 0.05 < p < 0.1), and 23% in -/- littermates (p < 0.05) littermates. EtOH-exposed -/- OGG1-deficient progeny had 25% higher DNA oxidation than EtOH-treated +/+ ogg1 progeny (p < 0.05). A similar pattern for increasing DNA oxidation in PBN-pretreated, EtOH-exposed fetal brains was observed in +/+ and -/- ogg1 fetuses, with 48% higher DNA oxidation in -/- vs. +/+ fetuses (p < 0.001), while the 36% increase in +/- ogg1 fetuses was not significant.
Passive avoidance test
11
Ogg1 genotype effect: Saline-exposed progeny Saline-exposed +/+ ogg1 DNA repair-normal progeny had a greater latency time compared to +/and -/- DNA repair-deficient progeny at 6, 9 and 12 weeks (p < 0.001), while saline-exposed +/progeny had a greater latency time compared to -/- littermates at 9 weeks (p < 0.05) (Figure 3, upper panel). Latency time for +/- progeny at 12 weeks was higher than that at 6 weeks (p < 0.01), while latency time for -/- was higher at 9 weeks (p < 0.01) and 12 weeks (p < 0.001) than that at 6 weeks (Figure 3, lower panel). Ogg1 genotype effect: PBN-exposed progeny At 6 weeks, PBN-treated +/+ ogg1 progeny had a greater latency time compared to +/- (p < 0.001) and -/- (p < 0.001) progeny (Figure 3). At 9 weeks, +/+ (p < 0.001) and +/- (p < 0.01) progeny had higher latency times compared to -/- littermates. Latency time for +/- progeny at 9 weeks (p < 0.001) and 12 weeks (p < 0.001) was higher than that at 6 weeks, while latency time for -/- was higher at 12 weeks than that at 6 weeks (p < 0.001) and 9 weeks (p < 0.001) (Figure 3, lower panel). Ogg1 genotype effect: EtOH-exposed progeny At 6 weeks, EtOH-treated +/+ ogg1 progeny had a greater latency time compared to -/- (p < 0.05) littermates (Figure 3). At 9 weeks, EtOH-exposed +/+ progeny had a greater latency time compared to similarly exposed +/- (p < 0.001) and -/- (p < 0.001) littermates, and +/- progeny had a greater latency time than -/- littermates (p < 0.001). At 12 weeks, +/+ (p < 0.001) and +/(p < 0.01) progeny had a greater latency time than -/- littermates exposed to EtOH. Latency time for +/+ at 9 weeks (p < 0.01) and 12 weeks (p < 0.01) was higher than that at 6 weeks, while
12
latency time for +/- progeny at 12 weeks was higher than that at 6 weeks (p < 0.001). (Figure 3, lower panel). Ogg1 genotype effect: PBN-pretreated EtOH-exposed progeny At 6 weeks, there were no differences in latency times among +/+, +/- and -/- ogg1 progeny exposed in utero to PBN followed by EtOH (Figure 3). At 9 weeks, +/+ ogg1 progeny had a greater latency time compared to similarly exposed +/- (p < 0.001) and -/- littermates (p < 0.001), and +/- progeny had a greater latency time compared to -/- littermates (p < 0.05). At 12 weeks, +/+ (p < 0.01) and +/- (p < 0.05) progeny had a greater latency time than -/- progeny exposed to PBN followed by EtOH. Latency time for +/+ progeny at 9 weeks (p < 0.001) and 12 weeks (p < 0.001) was higher than that at 6 weeks. Latency time for +/- progeny at 9 weeks (p < 0.01) and 12 weeks (p < 0.001) was higher than that at 6 weeks. Latency time for -/- was higher at 12 weeks than that at 6 weeks (p < 0.05) and 9 weeks (p < 0.05) (Figure 3, lower panel). Treatment effect Saline-exposed +/+, +/- and -/- ogg1 progeny had greater latency times compared to EtOHtreated +/+ (p < 0.05), +/- (p < 0.05) and -/- (p < 0.01) progeny, respectively (Figure 4). Similarly, PBN-pretreated EtOH-exposed +/+ and +/- ogg1 progeny had greater latency times compared to EtOH-treated +/+ (p < 0.05) and +/- (p < 0.01) progeny, respectively, while there was no difference in the -/- progeny. However, when analyzed in isolation by a one-way ANOVA with a post-hoc Bonferroni test, PBN-pretreated EtOH exposed -/- progeny had greater latency times compared to EtOH-treated -/- progeny (p < 0.01). There were no differences in latency time among saline-exposed and PBN-exposed progeny for any of the genotypes.
13
Fetal Weight There were no treatment- or genotype-dependent differences in fetal weight on GD 17, nor in the postnatal weight of progeny at any time point, suggesting that the single dose of EtOH administered was minimally toxic (Supplemental Figure S2). Males were heavier than females at 3 weeks (p < 0.05), 6 weeks (p < 0.001), 9 weeks (p < 0.001) and 12 weeks (p < 0.001).
14
DISCUSSION Our study in DNA repair-deficient ogg1 knockout mice demonstrates for the first time: (1) a behavioural phenotype for these mice in the absence of drug exposure; (2) that a single prenatal EtOH exposure on GD 17 during fetal development increases DNA oxidation measured as the 8oxodGuo lesion in fetal brain; (3) this single EtOH exposure decreases postnatal performance in the passive avoidance test at a minimally toxic dose that does not affect postnatal body weight; and, (4) that EtOH-initiated DNA oxidation in fetal brain and postnatal neurodevelopmental deficits are enhanced in DNA repair-deficient mice. These results, reflecting a deficit in learning and memory, reveal the exquisite susceptibility of developing brain function in the fetal period to impairment by both physiological oxidative stress in untreated fetuses, and in utero exposure to what we previously have shown is a relatively non-teratogenic dose of EtOH with respect to structural birth defects (Miller et al., 2013), consistent with the absence of any effect of EtOH in progeny body weight observed herein. Perhaps most noteworthy, our results provide the most direct evidence to date of a causal role for oxidatively damaged DNA in the fetal brain, and protective role for DNA repair, in the mechanism of neurodevelopmental deficits initiated by either physiological fetal ROS formation, or EtOH-enhanced ROS formation. While the 8-oxodGuo lesion is a biomarker for oxidative stress, several studies suggest that it is also a pathogenic lesion, likely via altered transcription as opposed to increased mutagenesis (Wells et al., 2010). Interestingly, although EtOH caused an ogg1 gene-dose-dependent increase in DNA oxidation in fetal brain, consistent with its enhancement of postnatal neurodevelopmental deficits, an increase in the 8-oxodGuo lesion was not observed in the brains of saline-exposed -/- ogg1 knockout progeny, which exhibited greater neurodevelopmental deficits than their saline-exposed +/+ littermates. This may in part reflect the impact of a more
15
tissue- or cell-selective macromolecular injury not detected by our analysis of whole brain homogenates, as tissue-selective oxidative DNA damage has been observed in untreated aged mice deficient in glucose-6-phosphate dehydrogenase (Jeng et al., 2013), as well as in adult mice treated with the ROS-initiating agents methamphetamine, 3,4-methylenedioxyamphatamine (Jeng et al., 2006) and 3,4-methylenedioxymethamphetamine (Ecstasy) (Jeng and Wells, 2010). Such selective macromolecular damage may be further augmented by postnatal accumulation of the 8-oxodGuo lesion in the saline-exposed DNA repair-deficient mice. Several potential mechanisms exist whereby in utero formation of the 8-oxodGuo lesion in fetal brain may initiate postnatal behavioural deficits, including altering the expression and activity of proteins required for normal embryonic development, alteration of gene transcription or expression via its ability to regulate binding affinity of various transcription factors including nuclear factor kappa B (NFțB) to specific promoter elements, and/or apoptosis resulting from 8-oxodGuo accumulation, none of which are mutually exclusive (Wells et al., 2009). Likely downstream consequences include altered cellular division, differentiation, migration, function and intercellular communication in the brain. Saline-exposed +/+ progeny exhibited a consistently better learning and memory performance than +/- and -/- ogg1 knockout littermates in an ogg1 gene dose-dependent fashion. This neurodevelopmental deficit in untreated animals constitutes the first demonstration of a phenotype for ogg1 DNA repair-deficient mice, which somewhat surprisingly do not exhibit an increase in cancer (Klungland et al., 1999), suggesting that the fundamental importance of this gene lies in development rather than providing lifelong protection against cancer. The OGG1dependent enhanced susceptibility to neurodevelopmental deficits was similarly observed in EtOH-exposed progeny, which exhibited a shift in the ogg1 gene dose-response curve,
16
corroborating the hypothesis that EtOH-initiated behavioural deficits are initiated at least in part by ROS-mediated oxidatively damaged DNA in the developing fetal brain which, if not repaired by OGG1, can result in abnormal postnatal brain function. The results observed herein for EtOH, whereby fetal brain DNA oxidation and learning deficits were increased in an ogg1 gene dose-dependent fashion, are consistent with studies showing that CD-1 progeny exposed in utero to EtOH exhibited enhanced fetal brain DNA oxidation and learning deficits (Miller et al., 2013), ogg1 knockout progeny exposed in utero to methamphetamine exhibited enhanced motor coordination deficits (Wong et al., 2008), and knockout mice deficient in Cockayne Syndrome B (CSB), another DNA repair protein also exhibited enhanced motor coordination deficits compared to their DNA repair-normal wild-type littermates (McCallum et al., 2011). Fetal weights were not affected by EtOH exposure, indicating that the behavioural deficit observed was not secondary to a structural growth deficit, and that the single dose administered was minimally fetotoxic. Additionally, we previously found that the single dose used herein does not cause morphological birth defects when administered during the embryonic period, and when administered during the fetal period is the lowest reported single dose that results in postnatal neurobehavioural deficits in a passive avoidance model (Miller et al., 2013). As observed in our single-exposure threshold model herein, previous studies have observed similar postnatal neurodevelopmental deficits with chronic EtOH exposure throughout pregnancy in both rats (Abel, 1982; Mattson et al., 1993; Riley et al., 1979) and mice (Becker and Randall, 1989; Fiore et al., 2009; Gilliam et al., 1987), whereby EtOH-exposed progeny required more trials to learn the passive avoidance task than control progeny.
17
Phenylbutylnitrone (PBN) is a free radical spin trapping agent that traps free radicals thereby minimizing the damaging effect of excess ROS (Janzen et al., 1985). The protective effect of PBN in blocking EtOH-initiated 8-oxodGuo formation in +/+, +/- and -/- ogg1 fetal brains and postnatal behavioural deficits in +/+ and -/+ ogg1 progeny suggests that ROS are involved in the mechanism of EtOH-initiated DNA oxidation and neurodevelopmental deficits, and that PBN is protecting via a mechanism proximal to DNA repair. PBN similarly protects against birth defects initiated by other ROS-initiating teratogens including thalidomide (Lee et al., 2011; Parman et al., 1999) and phenytoin (Liu and Wells, 1994; Wells et al., 1989), and reduce the associated oxidative damage to cellular macromolecules including DNA, protein and lipids (Wells et al., 2009). The increased susceptibility of ogg1 DNA repair-deficient knockout mice to fetal DNA oxidation and postnatal behavioral deficits caused by physiological fetal ROS formation or in utero EtOH exposure indicates both the pathogenic importance of the 8-oxodGuo lesion, and the protective importance of fetal DNA repair enzymes in preventing behavioral abnormalities. It therefore is possible that physiological ROS formation in the embryo or fetus combined with a genetically or environmentally reduced capacity for DNA repair contribute to the estimated 70% of human birth defects of undetermined origin. Similarly, if alcohol is consumed by a pregnant woman whose child has a functionally relevant DNA repair variant, inherited from either the mother or father, that renders OGG1 inactive and/or deficient, her child could be at increased risk of developing FASD. Recent studies have associated the Ser326Cys polymorphism in the human OGG1 gene with delayed enzyme function (Kershaw and Hodges, 2012), lung cancer in Chinese population (Kirkali et al., 2009) and non-small cell lung cancer (Janik et al., 2011). Given that DNA damage and repair have been implicated in teratogenesis as well as
18
carcinogenesis (Wells et al., 2010), our mouse model suggests that adverse developmental outcomes would be expected in the DNA repair-deficient children of mothers drinking ethanol during pregnancy. In summary, OGG1-deficient fetuses exposed once in utero to only saline vehicle exhibit enhanced postnatal cognitive deficits, revealing the first evidence of a phenotype for these knockout mice. DNA oxidation and postnatal neurodevelopmental deficits are further increased by in utero exposure of ogg1 knockout fetuses to a single dose of EtOH that does not affect body weight. These results reveal the exquisite sensitivity of the developing brain to both physiological fetal and/or neonatal ROS formation, and EtOH-enhanced oxidative stress. The protection afforded by pretreatment with the free radical spin trapping agent PBN suggests that both the macromolecular damage and consequential cognitive deficits are ROS-mediated. The increased susceptibility of untreated ogg1 knockout progeny, and the ogg1 gene dose-dependent exacerbation of EtOH-initiated 8-oxodGuo formation in fetal brain and postnatal cognitive deficits among littermates with deficient DNA repair provides the most direct evidence to date that: (1) the 8-oxodGuo lesion plays a pathogenic role in the mechanism of postnatal neurodevelopmental deficits initiated by in utero physiological fetal ROS formation or EtOHenhanced oxidative stress; and, (2) OGG1 activity in the fetal brain constitutes a determinant of risk.
19
ACKNOWLEDGEMENT The authors are grateful to Kyla Lam in the Department of Pharmacology & Toxicology at the University of Toronto for developing the passive avoidance apparatus and testing procedure used in this study.
20
REFERENCES Abel, E. L. (1982). In utero alcohol exposure and developmental delay of response inhibition. Alcohol Clin Exp Res 6, 369-76. Abramov, J. P. and Wells, P. G. (2011). Embryonic catalase protects against endogenous and phenytoin-enhanced DNA oxidation and embryopathies in acatalasemic and human catalaseexpressing mice. Faseb J 25, 2188-200. Becker, H. C. and Randall, C. L. (1989). Effects of prenatal ethanol exposure in C57BL mice on locomotor activity and passive avoidance behavior. Psychopharmacology (Berl) 97, 40-4. Boiteux, S. and Radicella, J. P. (2000). The human OGG1 gene: structure, functions, and its implication in the process of carcinogenesis. Arch Biochem Biophys 377, 1-8. Brocardo, P. S., Boehme, F., Patten, A., Cox, A., Gil-Mohapel, J. and Christie, B. R. (2012). Anxiety- and depression-like behaviors are accompanied by an increase in oxidative stress in a rat model of fetal alcohol spectrum disorders: Protective effects of voluntary physical exercise. Neuropharmacology 62, 1607-18. Dong, J., Sulik, K. K. and Chen, S. Y. (2010). The role of NOX enzymes in ethanol-induced oxidative stress and apoptosis in mouse embryos. Toxicol Lett 193, 94-100. Fiore, M., Laviola, G., Aloe, L., di Fausto, V., Mancinelli, R. and Ceccanti, M. (2009). Early exposure to ethanol but not red wine at the same alcohol concentration induces behavioral and brain neurotrophin alterations in young and adult mice. Neurotoxicology 30, 59-71. Gilliam, D. M., Stilman, A., Dudek, B. C. and Riley, E. P. (1987). Fetal alcohol effects in longand short-sleep mice: activity, passive avoidance, and in utero ethanol levels. Neurotoxicol Teratol 9, 349-57. Hansen, J. M. and Harris, C. (2013). Redox control of teratogenesis. Reprod Toxicol 35, 165-79. Janik, J., Swoboda, M., Janowska, B., Ciesla, J. M., Gackowski, D., Kowalewski, J., Olinski, R., Tudek, B. and Speina, E. (2011). 8-Oxoguanine incision activity is impaired in lung tissues of NSCLC patients with the polymorphism of OGG1 and XRCC1 genes. Mutat Res 709-710, 21-31. Janzen, E. G., Stronks, H. J., Dubose, C. M., Poyer, J. L. and McCay, P. B. (1985). Chemistry and biology of spin-trapping radicals associated with halocarbon metabolism in vitro and in vivo. Environ Health Perspect 64, 151-70. Jeng, W., Wong, A. W., Ting, A. K. R. and Wells, P. G. (2005). Methamphetamine-enhanced embryonic oxidative DNA damage and neurodevelopmental deficits. Free Radic Biol Med 39, 317-26.
21
Jeng, W., Ramkissoon, A., Parman, T. and Wells, P. G. (2006). Prostaglandin H synthasecatalyzed bioactivation of amphetamines to free radical intermediates that cause CNS regional DNA oxidation and nerve terminal degeneration. Faseb J 20, 638-50. Jeng, W. and Wells, P. G. (2010). Reduced 3,4-methylenedioxymethamphetamine (MDMA, Ecstasy)-initiated oxidative DNA damage and neurodegeneration in prostaglandin H synthase-1 knockout mice. ACS Chem Neurosci 1, 366-80. Jeng, W., Loniewska, M. M. and Wells, P. G. (2013). Brain glucose-6-phosphate dehydrogenase protects against endogenous oxidative DNA damage and neurodegeneration in aged mice. ACS Chem Neurosci 4, 1123-32. Jones, K. L. (2011). The effects of alcohol on fetal development. Birth Defects Res C Embryo Today 93, 3-11. Kershaw, R. M. and Hodges, N. J. (2012). Repair of oxidative DNA damage is delayed in the Ser326Cys polymorphic variant of the base excision repair protein OGG1. Mutagenesis 27, 50110. Kirkali, G., de Souza-Pinto, N. C., Jaruga, P., Bohr, V. A. and Dizdaroglu, M. (2009). Accumulation of (5'S)-8,5'-cyclo-2'-deoxyadenosine in organs of Cockayne syndrome complementation group B gene knockout mice. DNA Repair (Amst) 8, 274-8. Klungland, A., Rosewell, I., Hollenbach, S., Larsen, E., Daly, G., Epe, B., Seeberg, E., Lindahl, T. and Barnes, D. E. (1999). Accumulation of premutagenic DNA lesions in mice defective in removal of oxidative base damage. Proc Natl Acad Sci U S A 96, 13300-5. Koop, D. R. (2006). Alcohol metabolism's damaging effects on the cell: a focus on reactive oxygen generation by the enzyme cytochrome P450 2E1. Alcohol Res Health 29, 274-80. Lee, C. J., Goncalves, L. L. and Wells, P. G. (2011). Embryopathic effects of thalidomide and its hydrolysis products in rabbit embryo culture: evidence for a prostaglandin H synthase (PHS)dependent, reactive oxygen species (ROS)-mediated mechanism. Faseb J 25, 2468-83. Liu, L. and Wells, P. G. (1994). In vivo phenytoin-initiated oxidative damage to proteins and lipids in murine maternal hepatic and embryonic tissue organelles: potential molecular targets of chemical teratogenesis. Toxicol Appl Pharmacol 125, 247-55. Mattson, S. N., Carlos, R. and Riley, E. P. (1993). The behavioral teratogenicity of alcohol is not affected by pretreatment with aspirin. Alcohol 10, 51-7. Mattson, S. N., Crocker, N. and Nguyen, T. T. (2011). Fetal alcohol spectrum disorders: neuropsychological and behavioral features. Neuropsychol Rev 21, 81-101.
22
McCallum, G. P., Wong, A. W. and Wells, P. G. (2011). Cockayne syndrome B protects against methamphetamine-enhanced oxidative DNA damage in murine fetal brain and postnatal neurodevelopmental deficits. Antioxid Redox Signal 14, 747-56. Miller, L., Shapiro, A. M., Cheng, J. and Wells, P. G. (2013). The free radical spin trapping agent phenylbutylnitrone reduces fetal brain DNA oxidation and postnatal cognitive deficits caused by in utero exposure to a non-structurally teratogenic dose of ethanol: A role for oxidative stress. Free Radic Biol Med 60, 223-32. Nicol, C. J., Zielenski, J., Tsui, L. C. and Wells, P. G. (2000). An embryoprotective role for glucose-6-phosphate dehydrogenase in developmental oxidative stress and chemical teratogenesis. Faseb J 14, 111-27. Parman, T., Wiley, M. J. and Wells, P. G. (1999). Free radical-mediated oxidative DNA damage in the mechanism of thalidomide teratogenicity. Nat Med 5, 582-5. Ravanat, J. L., Douki, T., Duez, P., Gremaud, E., Herbert, K., Hofer, T., Lasserre, L., SaintPierre, C., Favier, A. and Cadet, J. (2002). Cellular background level of 8-oxo-7,8-dihydro-2'deoxyguanosine: an isotope based method to evaluate artefactual oxidation of DNA during its extraction and subsequent work-up. Carcinogenesis 23, 1911-8. Riley, E. P., Lochry, E. A. and Shapiro, N. R. (1979). Lack of response inhibition in rats prenatally exposed to alcohol. Psychopharmacology (Berl) 62, 47-52. Stade, B., Ali, A., Bennett, D., Campbell, D., Johnston, M., Lens, C., Tran, S. and Koren, G. (2009). The burden of prenatal exposure to alcohol: revised measurement of cost. Can J Clin Pharmacol 16, e91-102. Wells, P. G., Zubovits, J. T., Wong, S. T., Molinari, L. M. and Ali, S. (1989). Modulation of phenytoin teratogenicity and embryonic covalent binding by acetylsalicylic acid, caffeic acid, and alpha-phenyl-N-t-butylnitrone: implications for bioactivation by prostaglandin synthetase. Toxicol Appl Pharmacol 97, 192-202. Wells, P. G., McCallum, G. P., Chen, C. S., Henderson, J. T., Lee, C. J., Perstin, J., Preston, T. J., Wiley, M. J. and Wong, A. W. (2009). Oxidative stress in developmental origins of disease: teratogenesis, neurodevelopmental deficits, and cancer. Toxicol Sci 108, 4-18. Wells, P. G., McCallum, G. P., Lam, K. C., Henderson, J. T. and Ondovcik, S. L. (2010). Oxidative DNA damage and repair in teratogenesis and neurodevelopmental deficits. Birth Defects Res C Embryo Today 90, 103-9. Winn, L. M. and Wells, P. G. (1995). Phenytoin-initiated DNA oxidation in murine embryo culture, and embryo protection by the antioxidative enzymes superoxide dismutase and catalase: evidence for reactive oxygen species-mediated DNA oxidation in the molecular mechanism of phenytoin teratogenicity. Mol Pharmacol 48, 112-20.
23
Wong, A. W., McCallum, G. P., Jeng, W. and Wells, P. G. (2008). Oxoguanine glycosylase 1 protects against methamphetamine-enhanced fetal brain oxidative DNA damage and neurodevelopmental deficits. J Neurosci 28, 9047-54.
24
LEGENDS Figure 1. Postulated role of oxidatively damaged DNA in the fetal brain, and repair by fetal oxoguanine glycosylase 1 (OGG1), in the mechanism of postnatal behavioural deficits initiated by physiological fetal and/or neonatal formation of reactive oxygen species (ROS), or by in utero exposure to ethanol (EtOH). EtOH can induce ROS-producing fetal NADPH oxidases. The free radical spin trapping agent phenylbutylnitrone (PBN) can detoxify drug and ROS free radical intermediates. OGG1 is the primary enzyme for repairing the DNA lesion 8oxo-2’-deoxyguanosine (8-oxodGuo). Figure 2. Oxidatively damaged DNA in fetal brains from oxoguanine glycosylase 1 (OGG1) wild-type (+/+), and heterozygous (+/-) and homozygous (-/-) knockout mice. Pregnant ogg1 +/- mice bred with +/- males were treated on gestational day (GD) 17 with ethanol (2 g/kg i.p.) or saline vehicle (VH), with or without pretreatment with the free radical spin trapping agent phenylbutylnitrone (PBN) (40 mg/kg i.p.) and sacrificed 6 hr later. Litters containing embryos from all 3 genotypes were analyzed for 8-oxo-2'-deoxyguanosine (8-oxodGuo) formation using an ELISA kit with an 8-oxodGuo-specific antibody. Data were analyzed by two-way ANOVA with a post-hoc Bonferroni test. Single daggers indicate a difference from the vehicle control group for the same genotype (††† = p < 0.001). Asterisks indicate a difference from +/+ fetuses for the same treatment group (* = p < 0.05, *** = p < 0.05). Psi symbols indicate a difference from EtOH-treated fetuses for the same genotype (ȥ = p < 0.05, ȥȥȥ = p < 0.01). Alpha symbol indicates a difference from EtOH-treated fetuses for the same genotype (Į = 0.05 < p < 0.1) The number of fetal brains analyzed is given in parentheses. For each treatment group, fetuses were randomly selected from 3-4 litters. Figure 3. Time course for postnatal learning development in oxoguanine glycosylase 1 (OGG1) wild-type (+/+), heterozygous (+/-) and knockout (-/-) progeny exposed in utero to EtOH with or without PBN pretreatment. Pregnant mice were treated as described in fig. 2, and the progeny were tested postnatally for a passive avoidance test at weeks 6, 9 and 12. Data are shown for trial #3 at 6, 9 and 12 weeks. Data in the bottom four panels are different representations of the same data presented in the top four panels. Data were analyzed by twoway ANOVA with a post-hoc Bonferroni test. Alpha symbols indicate a difference from week 6 of the same genotype (Į = p < 0.05, ĮĮ = p < 0.01, ĮĮĮ = p < 0.001). Beta symbols indicate a difference from week 9 of the same genotype (ȕ = p < 0.05, ȕȕȕ = p < 0.001). Asterisks indicate a difference from +/+ progeny for the same time point (* = p < 0.05, ** = p < 0.01, *** = p < 0.001). Pound symbols indicate a difference from +/- progeny for the same time point (# = p < 0.05, ## = p < 0.01, ### = p < 0.001). The number of progeny tested is given in parentheses. For each treatment group, fetuses were randomly selected from 6-8 litters. No gender differences were observed, therefore the N values represent a combination of male and female progeny. Figure 4. Effect of PBN pretreatment on EtOH-initiated learning deficits in oxoguanine glycosylase 1 (OGG1) wild-type (+/+), and heterozygous (+/-) and homozygous (-/-) knockout progeny. Pregnant mice were treated as described in fig. 2, and the progeny were tested postnatally for a passive avoidance test at weeks 6, 9 and 12. Data are shown for trial #3 at 12 weeks. Data in the bottom panel are a different representation of the same results shown in the
25
top panels. Data were analyzed by two-way ANOVA with a post-hoc Bonferroni test, and symbols indicate the significance of treatment effects. For the statistical significance of genotypic effects, please refer to fig. 3. Daggers indicate a difference from saline for the same genotype († = p < 0.5, †† = p < 0.01, ††† = p < 0.001). Psi symbols indicate a difference from EtOH-treated fetuses for the same genotype (ȥ = p < 0.05, ȥȥ = p < 0.01). The number of progeny tested is given in parentheses. For each treatment group, fetuses were randomly selected from 6-8 litters. No gender differences were observed, therefore the N values represent a combination of male and female progeny.
26
Fig. 1
Fig. 2
Fig. 3
Fig. 4
Graphical Abstract (for review)
Title: Oxidative DNA damage in the in utero initiation of postnatal neurodevelopmental deficits by normal fetal and ethanol-enhanced oxidative stress in oxoguanine glycosylase 1 (ogg1) knockout mice Authors: Lutfiya Miller, Daniel J. Pinto and Peter G. Wells
Highlights • • • • •
DNA repair-deficient ogg1 knockout mice have impaired neurodevelopment In utero ethanol (EtOH) enhances fetal DNA oxidation and impairs neurodevelopment Both EtOH outcomes are greater in ogg1 knockout progeny Both EtOH outcomes in ogg1 progeny are blocked by a free radical spin trap Fetal OGG1 and DNA oxidation are determinants of risk for in utero EtOH outcomes
Oxidative DNA damage in the in utero initiation of postnatal neurodevelopmental deficits by normal fetal and ethanol-enhanced oxidative stress in oxoguanine glycosylase 1 (ogg1) knockout mice Lutfiya Miller-Pinsler, Daniel J. Pinto, Peter G. Wells www.elsevier.com/locate/freeradbiomed
PII: DOI: Reference:
S0891-5849(14)00445-6 http://dx.doi.org/10.1016/j.freeradbiomed.2014.09.026 FRB12163
To appear in:
Free Radical Biology and Medicine
Received date: 25 February 2014 Revised date: 15 September 2014 Accepted date: 24 September 2014 Cite this article as: Lutfiya Miller-Pinsler, Daniel J. Pinto, Peter G. Wells, Oxidative DNA damage in the in utero initiation of postnatal neurodevelopmental deficits by normal fetal and ethanol-enhanced oxidative stress in oxoguanine glycosylase 1 (ogg1) knockout mice, Free Radical Biology and Medicine, http://dx.doi.org/10.1016/j.freeradbiomed.2014.09.026 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
15th September 2014 Oxidative DNA damage in the in utero initiation of postnatal neurodevelopmental deficits by normal fetal and ethanol-enhanced oxidative stress in oxoguanine glycosylase 1 (ogg1) knockout mice Running title: Fetal DNA repair & ethanol behavioural deficits Lutfiya Miller-Pinsler†, Daniel J. Pinto† and Peter G. Wells*† *Division of Biomolecular Sciences, Faculty of Pharmacy, and †Department of Pharmacology and Toxicology, Faculty of Medicine University of Toronto Toronto, Ontario, Canada
Number of: • Text pages (excluding references, footnotes, tables and figure legends): 19 • Tables: 0 • Figures: 4 plus 2 supplemental online figures • References: 35 • Words: - Abstract: 150 - Introduction: 683 - Discussion: 1,299 Corresponding Author: Peter G. Wells, Pharm.D. Division of Biomolecular Sciences Faculty of Pharmacy, University of Toronto 144 College St., Toronto, Ontario, Canada M5S 3M2 Tel. 416-978-3221; Email: [email protected]
a. This work was supported by a grant from the Canadian Institutes of Health Research (CIHR). The authors declare no competing financial interests.
Non-standard Abbreviations: 8-oxodGuo, 8-oxo-2’-deoxyguanosine; FASD, fetal alcohol spectrum disorders; OGG1, oxoguanine glycosylase 1; PAT, passive avoidance test; PBN, alpha-phenyl-N-tert-butylnitrone; GD, gestational day; EtOH, ethanol; ROS, reactive oxygen species
1
ABSTRACT Studies in mice with deficient antioxidative enzymes have shown that physiological levels of reactive oxygen species (ROS) can adversely affect the developing embryo and fetus. Herein, DNA repair-deficient progeny of oxoguanine glycosylase 1 (ogg1) knockout mice lacking repair of the oxidative DNA lesion 8-oxo-2’-deoxyguanosine (8-oxodGuo) exhibited enhanced postnatal neurodevelopmental deficits, revealing the pathogenic potential of 8-oxodGuo initiated by physiological ROS production in fetal brain, and providing the first evidence of a pathological phenotype for ogg1 knockout mice. Moreover, when exposed in utero to ethanol (EtOH), ogg1 knockout progeny exhibited higher levels of 8-oxodGuo in fetal brain and more severe postnatal neurodevelopmental deficits than wild-type littermates, both of which were blocked by pretreatment with the free radical trapping agent phenylbutylnitrone. These results suggest ROSinitiated DNA oxidation, as distinct from altered signal transduction, contributes to neurodevelopmental deficits caused by in utero EtOH exposure, and fetal DNA repair is a determinant of risk. Key Words: Ethanol; phenylbutylnitrone; PBN; passive avoidance; reactive oxygen species; DNA oxidation; OGG1 mice; behavioural deficits; neurodevelopmental deficits
2
INTRODUCTION Embryonic or fetal formation of reactive oxygen species (ROS) can cause oxidative stress, which can potentially dysregulate development by altering signal transduction pathways or oxidatively damaging cellular macromolecules including proteins, lipids and DNA, resulting in structural birth defects, in utero or neonatal death, and postnatal neurodevelopmental deficits (Hansen and Harris, 2013; Wells et al., 2009). In untreated mutant mice, progeny with deficiencies in glucose-6-phosphate dehydrogenase (G6PD) or catalase, which are necessary for ROS detoxification, exhibit a higher incidence of structural birth defects and in utero and neonatal death (Abramov and Wells, 2011; Nicol et al., 2000), suggesting that physiological levels of ROS formation can be embryopathic. It has been estimated that the cause of human birth defects is unknown in up to 70% of cases (Hansen and Harris, 2013), and physiological ROS formation may contribute to these adverse developmental outcomes, particularly when the progeny are deficient in pathways important for ROS detoxification, or possibly the repair of oxidatively damaged DNA. The consequences of in utero EtOH exposure are well characterized, and include a spectrum of anomalies termed the Fetal Alcohol Spectrum Disorder (FASD), which includes both structural and functional birth defects, with the complete phenotype including characteristic craniofacial dysmorphology, growth retardation and behavioural deficits (Jones, 2011). The estimated incidence is 1 out of 100 live births in Canada (Stade et al., 2009), and is postulated to be one of the leading preventable causes of neurodevelopmental disorders in the Western world (Mattson et al., 2011). In humans, prenatal EtOH exposure has been shown to initiate a broad spectrum of behavioural anomalies in children including decreased IQ, motor coordination deficits, hyperactivity and deficits in executive function, verbal language, memory and attention
3
(Mattson et al., 2011). The mechanisms underlying these deficits are unclear, however enhanced ROS formation has been implicated (Brocardo et al., 2012; Miller et al., 2013), and EtOH can induce embryonic ROS-generating NADPH oxidase (NOX) enzymes (Dong et al., 2010), and may also enhance ROS via free radical formation during its metabolism by cytochromes P450 (Koop, 2006). Oxoguanine glycosylase 1 (OGG1) is a critical DNA repair enzyme involved in the excision and repair of the 8-oxo-2-deoxyguanosine (8-oxodGuo) lesion (Boiteux and Radicella, 2000), a biomarker of oxidative stress as well as a developmentally pathogenic lesion (McCallum et al., 2011; Wells et al., 2009; Wong et al., 2008). OGG1 may be particularly important in the developing brain, where its activity is double that in maternal tissues (Wong et al., 2008). The progeny of CD-1 mice exposed in utero to EtOH exhibited an increase in oxidatively damaged DNA (8-oxodGuo) in fetal brain DNA, as well as postnatal learning and memory deficits (Miller et al., 2013). Similarly, CD-1 mice exposed in utero to the ROSinitiating drug methamphetamine exhibit increased fetal brain levels of 8-oxodGuo as well as postnatal motor coordination deficits (Jeng et al., 2005). The pathogenic role of 8-oxodGuo in methamphetamine-initiated neurodevelopmental deficits was revealed in studies using knockout mice lacking ogg1 (Wong et al., 2008) or csb (McCallum et al., 2011), which also repairs the 8oxodGuo lesion. In these studies, knockout progeny exposed in utero to methamphetamine exhibited an ogg1 or csb gene dose-dependent increase in 8-oxodGuo in fetal brain of +/- and -/pups compared to wild-type littermates, and increased motor coordination deficits measured by rotarod testing. ROS-initiating teratogens in addition to methamphetamine, including phenytoin and thalidomide, similarly increase 8-oxodGuo formation in the developing embryo and fetus (Abramov and Wells, 2011; Lee et al., 2011; Parman et al., 1999; Winn and Wells, 1995).
4
Taken together, these data suggest that 8-oxodGuo is a pathogenic molecular lesion in the embryo, and ogg1 variability is a determinant of risk for behavioural deficits caused by oxidative stress. However, there is no published evidence of a phenotype in these ogg1 knockout mice, so the developmental importance of fetal OGG1 is similarly not known. Moreover, the roles of oxidatively damaged DNA and OGG1 in EtOH-initiated behavioural deficits are similarly unknown. Herein, we sought to determine the pathogenic contribution of ROS-mediated formation of the 8-oxodGuo lesion, and the protective role of fetal DNA repair (Figure 1) in the development of untreated mice, and in the mechanism of EtOH-initiated behavioural deficits, using DNA repair-deficient ogg1 knockout mice. Our results in untreated mice provide the first demonstration of a behavioural phenotype for the ogg1 knockout mouse, suggesting that even the low level of fetal DNA oxidation from normal physiological processes can be developmentally toxic when DNA repair is inadequate. Our EtOH studies reveal that deficient fetal DNA repair and enhanced 8-oxodGuo levels contribute to the mechanism of EtOH-initiated postnatal behavioural deficits.
5
MATERIALS AND METHODS
Chemicals. Alpha-phenyl-N-tert-butylnitrone (phenylbutylnitrone, PBN), minimum 98% purity by gas chromatography (GC), and all other reagents unless otherwise specified were purchased from Sigma Aldrich (St. Louis, MO). Saline (0.9 %, sterile) was purchased from Baxter Corporation (Mississauga, ON) and ethanol from Commercial Alcohol Inc. (Brampton, ON).
Animals and diet. All animal protocols were approved by the institutional animal care committee in conformance with the guidelines established by the Canadian Council on Animal Care. Ogg1 knockout mice on a 129SV/C57BL/6J background strain were originally generated by Klungland and coworkers (Klungland et al., 1999), and generously provided by Dr. Tomas Lindahl (Imperial Cancer Research Fund, UK) through Dr. Christi A. Walter at the University of Texas Health Science Center at San Antonio. Mice were housed in vented plastic cages from Allentown, Inc. (Allentown, NJ) with ground corncob bedding (Bed-O’Cobs Laboratory Animal Bedding, The Andersons Industrial Products Group, Maumee, OH). Mouse cages were maintained in a room with controlled light (14 hr light-10 hr dark cycle) and climate (20ºC, 50% humidity), and provided with rodent chow (Harlan Labs: 2018, Harlan Teklad, Montreal, QC) and tap water ad libitum. Mice were acclimatized for 1 week prior to use. Virgin +/- ogg1 females were mated with a +/- male overnight from 5:00 P.M. to 9:00 A.M. by housing one male in a cage containing 1-3 females. Heterozygous matings were employed to generate progeny of all 3 genotypes (+/+, +/-, -/-) within the same litter. The presence of a vaginal plug the next morning was designated as gestational day (GD) 1, and plugged females were separated, weighed and housed in groups of up to four dams per cage until use.
6
Genotyping. DNA was isolated from tail snips of fetal progeny by heating the sample at 95°C in 75 µL alkaline lysis reagent (25 mM NaOH, 0.2 mM disodium EDTA, pH 12) for 1 hr. Samples were then neutralized by addition of 75 µL of neutralizing buffer (40 mM Tris-HCl, pH 5), and genotyped using a PCR-based assay. Primers (synthesized by The Center for Applied Genomics, The Hospital for Sick Children, Toronto, ON) used to amplify the 500 base pair (bp) band for the ogg1 gene were: ogg1-sense (5’-ACTGCATCTGCTTAATGGCC-3’) (forward primer) and ogg1-antisense (5’-CGAAGGTCAGCACTGAACAG-3’) (reverse primer). Primers used to amplify the 300 bp band for the neo-cassette responsible for disruption of the ogg1 gene in the ogg1 knockout mice were neo-sense (5’-CTGAATGAACTGCAGGACGA-3’) (forward primer) and neo-antisense (5’-CTCTTCGTCCAGATCATCCT-3’) (reverse primer). PCR reaction conditions were: 8 ȝl genomic DNA, 5 ȝl per sample of 10X Hotstart Buffer (Fermentas Life Sciences, Burlington, ON), 3 ȝl per sample of MgCl2 (Fermentas Life Sciences), 0.25 ȝl Maxima Hotstart Taq polymerase (Fermentas Life Sciences), 1 ȝl per sample of 10 mM deoxyribonucleotides (dNTP) (Fermentas Life Sciences), 0.75 ȝL per sample of each 20 ȝM primer and 29.75 ȝl per sample of ddH2O for a final volume of 50 ȝl. Cycling conditions were: 95°C for 5 min; 35 cycles of: 94°C for 1 min, 55°C for 1.5 min, 72°C for 2 min and completed with a final extension at 72°C for 10 min and then placed on hold at 4°C. 6X DNA loading dye (0.03% bromophenol blue/0.03% xylene cyanol FF) was added to each sample. The PCR products were separated on a gel consisting of 1.5% (w/v) agarose, 40 mM Tris, 19.4 mM glacial acetic acid, 2.5 mM EDTA and 8 ȝl ethidium bromide. The agarose gel was run at 100 V for 60 min, and then visualized and photographed under an ultraviolet light (Supplemental Figure S1).
7
Herein, we refer to “gene dose-dependent” not as the dose of the drug, but rather to the dose of the gene (zero, one or two alleles). Dosing for measurement of DNA oxidation and postnatal behavioural testing. On GD 17 (within the fetal period), pregnant +/- ogg1 dams were pretreated with PBN (40 mg/kg i.p.) or its 0.79% saline vehicle control at 9:00 AM using a 26 gauge (G) 3/8 needle. Thirty min later, dams were treated with a 25% v/v solution of EtOH (2 g/kg i.p.) or its 0.9% saline vehicle. We based the 2 g/kg EtOH dose in part on a previously published neurodevelopmental study (Miller et al., 2013) in which a 4 g/kg dose was used successfully in an outbred CD-1 strain. However, the 4 g/kg dose in inbred genetically modified ogg1 mice (129SV/C57BL/6J) proved too maternally toxic for this strain (increased maternal lethality and reduced body weight). Accordingly, the dose was reduced by half, which eliminated the maternal toxicity, and the 2 g/kg dose was used in the current study. Food was removed from the cages for 6 hr post injection to avoid potential effects on EtOH pharmacokinetics. One group of pregnant dams was sacrificed 6 hr after EtOH exposure, fetal brains were removed, snap frozen and stored at -80°C until analysis for DNA oxidation. A second group was allowed to deliver spontaneously for postnatal behavioural testing of their EtOH-exposed offspring. Three to four litters per treatment group were used to examine fetal brain DNA oxidation.
DNA extraction and digestion. DNA was extracted from GD 17 fetal brains using a modified version of a chaotropic sodium iodide (NaI) method previously described (Ravanat et al., 2002). Briefly, samples were homogenized in 1 mL of cold lysis buffer (320 mM sucrose, 5 mM MgCl2, 10mM Tris, 0.1 mM desferoxamine, 1% Triton X-100, pH 7.5) and centrifuged twice at 10,000
8
X g for 10 min at 4°C to isolate the nuclear pellet containing DNA. The pellet was then incubated with 200 uL enzyme reaction solution (1% w/v SDS, 5 mM EDTA-NA2, 0.15 mM desferoxamine 10 mM Tris-HCl pH 8.0) for 1 hr at 50ºC with RNase A/T1 mix (624 and 312 U/ml final activities, respectively). Proteinase K (1.8 mg/ml final concentration) was added for an additional 1 hr. The DNA pellet obtained after extraction was then washed 5 times with 70% EtOH. Samples were resuspended in 200 ȝl of sodium acetate buffer (20 mM, pH 4.8) and sonicated into solution. DNA purity was determined by measuring the absorbance ratio at 260/280 nm of 2 ȝL of sample in Na-acetate buffer (total volume 200 ȝL). Samples were digested with nuclease P1 (5 U/sample, 1 hr, 37°C) and calf intestinal alkaline phosphatase (6 U/sample, 1 hr, 37°C) and filtered through Amicon UltraTM filter units (YM-10, 10,000 MW cutoff; Millipore, Billerica, MA, USA) to remove DNA digestion enzymes and large particulates.
Quantification of oxidative damage to DNA using 8-OHdG ELISA. Oxidatively damaged DNA was quantified using the 8-OHdG ELISA kit (JaICA, Fukuroi, Japan) according to the manufacturer’s instructions.
Passive avoidance test. At 6, 9, 12 and 16 weeks of age, ogg1 wild-type, and +/- and -/knockout offspring were tested using a passive avoidance test as previously described (Miller et al., 2013). Briefly, the test apparatus consisted of a plastic box with two equal sized chambers, one with clear plastic walls designated as the ‘light chamber’, and another with red walls designated as the ‘dark chamber’, separated by a guillotine door. The floors of both chambers
9
consisted of stainless steel rods, wherein the rods of only the dark chamber were connected to a power supply that provided a mild 1.0 milliampere (mA) shock. In trial #1, the mouse was placed in the light chamber and was allowed to explore for 20 sec. The guillotine door was lifted, and once the mouse entered the dark chamber, the guillotine door was closed, and a shock of 1.0 mA was administered for 4 sec. The mouse was then returned back to its home cage, the apparatus was wiped with 70% EtOH to remove any olfactory cues, and the same procedure was repeated 24 and 48 hr later for trials #2 and #3, respectively. The time it took the mouse to enter the dark chamber in both trials after opening the guillotine door was recorded as “latency to enter the dark chamber”. To examine the effect of treatment and ogg1 genotype on learning, the latency to enter the dark chamber was plotted for trial 3 only. After trial #3, the pups were weighed at 6, 9 and 12 weeks to assess whether or not in utero EtOH exposure during the fetal period had any effect on growth. Six to eight litters per treatment group were used to examine learning and memory using the passive avoidance test.
Statistical analysis. Statistical analysis was performed using GraphPad Prism, Version 5 (GraphPad Software, Inc., San Diego, CA). DNA oxidation data, passive avoidance data and fetal weight data were analyzed using a two-way analysis of variance (ANOVA) with a post-hoc Bonferroni test. The minimal significance level used throughout was p < 0.05.
10
RESULTS Maternal Effects There were no treatment- or genotype-dependent differences in maternal weight (data not shown). The single 2 g/kg dose of EtOH resulted in temporary sedation lasting approximately 2 hours, after which time the dam recovered and behaved normally. Litter sizes ranged from 3 – 9 pups, and no pups were found dead. The pups were cared for by their biological mother for 3 weeks until they were weaned, at which point they lived among same-sex littermates.
DNA oxidation in fetal brains EtOH-exposed +/+, +/- and -/- ogg1 progeny exhibited 2.4-fold (p < 0.001), 2.5-fold (p < 0.001) and 2.3-fold (p < 0.001) increases in oxidatively damaged DNA, measured as the 8-oxodGuo lesion, in fetal brain compared to their respective, genotypically matched saline-exposed littermates (Figure 2). PBN pretreatment of saline-exposed fetuses did not change baseline DNA oxidation. Pretreatment with PBN decreased EtOH-enhanced DNA oxidation in fetal brains by 46% in +/+ ogg1 littermates (p < 0.001), 23% in +/- littermates (marginal, p 0.05 < p < 0.1), and 23% in -/- littermates (p < 0.05) littermates. EtOH-exposed -/- OGG1-deficient progeny had 25% higher DNA oxidation than EtOH-treated +/+ ogg1 progeny (p < 0.05). A similar pattern for increasing DNA oxidation in PBN-pretreated, EtOH-exposed fetal brains was observed in +/+ and -/- ogg1 fetuses, with 48% higher DNA oxidation in -/- vs. +/+ fetuses (p < 0.001), while the 36% increase in +/- ogg1 fetuses was not significant.
Passive avoidance test
11
Ogg1 genotype effect: Saline-exposed progeny Saline-exposed +/+ ogg1 DNA repair-normal progeny had a greater latency time compared to +/and -/- DNA repair-deficient progeny at 6, 9 and 12 weeks (p < 0.001), while saline-exposed +/progeny had a greater latency time compared to -/- littermates at 9 weeks (p < 0.05) (Figure 3, upper panel). Latency time for +/- progeny at 12 weeks was higher than that at 6 weeks (p < 0.01), while latency time for -/- was higher at 9 weeks (p < 0.01) and 12 weeks (p < 0.001) than that at 6 weeks (Figure 3, lower panel). Ogg1 genotype effect: PBN-exposed progeny At 6 weeks, PBN-treated +/+ ogg1 progeny had a greater latency time compared to +/- (p < 0.001) and -/- (p < 0.001) progeny (Figure 3). At 9 weeks, +/+ (p < 0.001) and +/- (p < 0.01) progeny had higher latency times compared to -/- littermates. Latency time for +/- progeny at 9 weeks (p < 0.001) and 12 weeks (p < 0.001) was higher than that at 6 weeks, while latency time for -/- was higher at 12 weeks than that at 6 weeks (p < 0.001) and 9 weeks (p < 0.001) (Figure 3, lower panel). Ogg1 genotype effect: EtOH-exposed progeny At 6 weeks, EtOH-treated +/+ ogg1 progeny had a greater latency time compared to -/- (p < 0.05) littermates (Figure 3). At 9 weeks, EtOH-exposed +/+ progeny had a greater latency time compared to similarly exposed +/- (p < 0.001) and -/- (p < 0.001) littermates, and +/- progeny had a greater latency time than -/- littermates (p < 0.001). At 12 weeks, +/+ (p < 0.001) and +/(p < 0.01) progeny had a greater latency time than -/- littermates exposed to EtOH. Latency time for +/+ at 9 weeks (p < 0.01) and 12 weeks (p < 0.01) was higher than that at 6 weeks, while
12
latency time for +/- progeny at 12 weeks was higher than that at 6 weeks (p < 0.001). (Figure 3, lower panel). Ogg1 genotype effect: PBN-pretreated EtOH-exposed progeny At 6 weeks, there were no differences in latency times among +/+, +/- and -/- ogg1 progeny exposed in utero to PBN followed by EtOH (Figure 3). At 9 weeks, +/+ ogg1 progeny had a greater latency time compared to similarly exposed +/- (p < 0.001) and -/- littermates (p < 0.001), and +/- progeny had a greater latency time compared to -/- littermates (p < 0.05). At 12 weeks, +/+ (p < 0.01) and +/- (p < 0.05) progeny had a greater latency time than -/- progeny exposed to PBN followed by EtOH. Latency time for +/+ progeny at 9 weeks (p < 0.001) and 12 weeks (p < 0.001) was higher than that at 6 weeks. Latency time for +/- progeny at 9 weeks (p < 0.01) and 12 weeks (p < 0.001) was higher than that at 6 weeks. Latency time for -/- was higher at 12 weeks than that at 6 weeks (p < 0.05) and 9 weeks (p < 0.05) (Figure 3, lower panel). Treatment effect Saline-exposed +/+, +/- and -/- ogg1 progeny had greater latency times compared to EtOHtreated +/+ (p < 0.05), +/- (p < 0.05) and -/- (p < 0.01) progeny, respectively (Figure 4). Similarly, PBN-pretreated EtOH-exposed +/+ and +/- ogg1 progeny had greater latency times compared to EtOH-treated +/+ (p < 0.05) and +/- (p < 0.01) progeny, respectively, while there was no difference in the -/- progeny. However, when analyzed in isolation by a one-way ANOVA with a post-hoc Bonferroni test, PBN-pretreated EtOH exposed -/- progeny had greater latency times compared to EtOH-treated -/- progeny (p < 0.01). There were no differences in latency time among saline-exposed and PBN-exposed progeny for any of the genotypes.
13
Fetal Weight There were no treatment- or genotype-dependent differences in fetal weight on GD 17, nor in the postnatal weight of progeny at any time point, suggesting that the single dose of EtOH administered was minimally toxic (Supplemental Figure S2). Males were heavier than females at 3 weeks (p < 0.05), 6 weeks (p < 0.001), 9 weeks (p < 0.001) and 12 weeks (p < 0.001).
14
DISCUSSION Our study in DNA repair-deficient ogg1 knockout mice demonstrates for the first time: (1) a behavioural phenotype for these mice in the absence of drug exposure; (2) that a single prenatal EtOH exposure on GD 17 during fetal development increases DNA oxidation measured as the 8oxodGuo lesion in fetal brain; (3) this single EtOH exposure decreases postnatal performance in the passive avoidance test at a minimally toxic dose that does not affect postnatal body weight; and, (4) that EtOH-initiated DNA oxidation in fetal brain and postnatal neurodevelopmental deficits are enhanced in DNA repair-deficient mice. These results, reflecting a deficit in learning and memory, reveal the exquisite susceptibility of developing brain function in the fetal period to impairment by both physiological oxidative stress in untreated fetuses, and in utero exposure to what we previously have shown is a relatively non-teratogenic dose of EtOH with respect to structural birth defects (Miller et al., 2013), consistent with the absence of any effect of EtOH in progeny body weight observed herein. Perhaps most noteworthy, our results provide the most direct evidence to date of a causal role for oxidatively damaged DNA in the fetal brain, and protective role for DNA repair, in the mechanism of neurodevelopmental deficits initiated by either physiological fetal ROS formation, or EtOH-enhanced ROS formation. While the 8-oxodGuo lesion is a biomarker for oxidative stress, several studies suggest that it is also a pathogenic lesion, likely via altered transcription as opposed to increased mutagenesis (Wells et al., 2010). Interestingly, although EtOH caused an ogg1 gene-dose-dependent increase in DNA oxidation in fetal brain, consistent with its enhancement of postnatal neurodevelopmental deficits, an increase in the 8-oxodGuo lesion was not observed in the brains of saline-exposed -/- ogg1 knockout progeny, which exhibited greater neurodevelopmental deficits than their saline-exposed +/+ littermates. This may in part reflect the impact of a more
15
tissue- or cell-selective macromolecular injury not detected by our analysis of whole brain homogenates, as tissue-selective oxidative DNA damage has been observed in untreated aged mice deficient in glucose-6-phosphate dehydrogenase (Jeng et al., 2013), as well as in adult mice treated with the ROS-initiating agents methamphetamine, 3,4-methylenedioxyamphatamine (Jeng et al., 2006) and 3,4-methylenedioxymethamphetamine (Ecstasy) (Jeng and Wells, 2010). Such selective macromolecular damage may be further augmented by postnatal accumulation of the 8-oxodGuo lesion in the saline-exposed DNA repair-deficient mice. Several potential mechanisms exist whereby in utero formation of the 8-oxodGuo lesion in fetal brain may initiate postnatal behavioural deficits, including altering the expression and activity of proteins required for normal embryonic development, alteration of gene transcription or expression via its ability to regulate binding affinity of various transcription factors including nuclear factor kappa B (NFțB) to specific promoter elements, and/or apoptosis resulting from 8-oxodGuo accumulation, none of which are mutually exclusive (Wells et al., 2009). Likely downstream consequences include altered cellular division, differentiation, migration, function and intercellular communication in the brain. Saline-exposed +/+ progeny exhibited a consistently better learning and memory performance than +/- and -/- ogg1 knockout littermates in an ogg1 gene dose-dependent fashion. This neurodevelopmental deficit in untreated animals constitutes the first demonstration of a phenotype for ogg1 DNA repair-deficient mice, which somewhat surprisingly do not exhibit an increase in cancer (Klungland et al., 1999), suggesting that the fundamental importance of this gene lies in development rather than providing lifelong protection against cancer. The OGG1dependent enhanced susceptibility to neurodevelopmental deficits was similarly observed in EtOH-exposed progeny, which exhibited a shift in the ogg1 gene dose-response curve,
16
corroborating the hypothesis that EtOH-initiated behavioural deficits are initiated at least in part by ROS-mediated oxidatively damaged DNA in the developing fetal brain which, if not repaired by OGG1, can result in abnormal postnatal brain function. The results observed herein for EtOH, whereby fetal brain DNA oxidation and learning deficits were increased in an ogg1 gene dose-dependent fashion, are consistent with studies showing that CD-1 progeny exposed in utero to EtOH exhibited enhanced fetal brain DNA oxidation and learning deficits (Miller et al., 2013), ogg1 knockout progeny exposed in utero to methamphetamine exhibited enhanced motor coordination deficits (Wong et al., 2008), and knockout mice deficient in Cockayne Syndrome B (CSB), another DNA repair protein also exhibited enhanced motor coordination deficits compared to their DNA repair-normal wild-type littermates (McCallum et al., 2011). Fetal weights were not affected by EtOH exposure, indicating that the behavioural deficit observed was not secondary to a structural growth deficit, and that the single dose administered was minimally fetotoxic. Additionally, we previously found that the single dose used herein does not cause morphological birth defects when administered during the embryonic period, and when administered during the fetal period is the lowest reported single dose that results in postnatal neurobehavioural deficits in a passive avoidance model (Miller et al., 2013). As observed in our single-exposure threshold model herein, previous studies have observed similar postnatal neurodevelopmental deficits with chronic EtOH exposure throughout pregnancy in both rats (Abel, 1982; Mattson et al., 1993; Riley et al., 1979) and mice (Becker and Randall, 1989; Fiore et al., 2009; Gilliam et al., 1987), whereby EtOH-exposed progeny required more trials to learn the passive avoidance task than control progeny.
17
Phenylbutylnitrone (PBN) is a free radical spin trapping agent that traps free radicals thereby minimizing the damaging effect of excess ROS (Janzen et al., 1985). The protective effect of PBN in blocking EtOH-initiated 8-oxodGuo formation in +/+, +/- and -/- ogg1 fetal brains and postnatal behavioural deficits in +/+ and -/+ ogg1 progeny suggests that ROS are involved in the mechanism of EtOH-initiated DNA oxidation and neurodevelopmental deficits, and that PBN is protecting via a mechanism proximal to DNA repair. PBN similarly protects against birth defects initiated by other ROS-initiating teratogens including thalidomide (Lee et al., 2011; Parman et al., 1999) and phenytoin (Liu and Wells, 1994; Wells et al., 1989), and reduce the associated oxidative damage to cellular macromolecules including DNA, protein and lipids (Wells et al., 2009). The increased susceptibility of ogg1 DNA repair-deficient knockout mice to fetal DNA oxidation and postnatal behavioral deficits caused by physiological fetal ROS formation or in utero EtOH exposure indicates both the pathogenic importance of the 8-oxodGuo lesion, and the protective importance of fetal DNA repair enzymes in preventing behavioral abnormalities. It therefore is possible that physiological ROS formation in the embryo or fetus combined with a genetically or environmentally reduced capacity for DNA repair contribute to the estimated 70% of human birth defects of undetermined origin. Similarly, if alcohol is consumed by a pregnant woman whose child has a functionally relevant DNA repair variant, inherited from either the mother or father, that renders OGG1 inactive and/or deficient, her child could be at increased risk of developing FASD. Recent studies have associated the Ser326Cys polymorphism in the human OGG1 gene with delayed enzyme function (Kershaw and Hodges, 2012), lung cancer in Chinese population (Kirkali et al., 2009) and non-small cell lung cancer (Janik et al., 2011). Given that DNA damage and repair have been implicated in teratogenesis as well as
18
carcinogenesis (Wells et al., 2010), our mouse model suggests that adverse developmental outcomes would be expected in the DNA repair-deficient children of mothers drinking ethanol during pregnancy. In summary, OGG1-deficient fetuses exposed once in utero to only saline vehicle exhibit enhanced postnatal cognitive deficits, revealing the first evidence of a phenotype for these knockout mice. DNA oxidation and postnatal neurodevelopmental deficits are further increased by in utero exposure of ogg1 knockout fetuses to a single dose of EtOH that does not affect body weight. These results reveal the exquisite sensitivity of the developing brain to both physiological fetal and/or neonatal ROS formation, and EtOH-enhanced oxidative stress. The protection afforded by pretreatment with the free radical spin trapping agent PBN suggests that both the macromolecular damage and consequential cognitive deficits are ROS-mediated. The increased susceptibility of untreated ogg1 knockout progeny, and the ogg1 gene dose-dependent exacerbation of EtOH-initiated 8-oxodGuo formation in fetal brain and postnatal cognitive deficits among littermates with deficient DNA repair provides the most direct evidence to date that: (1) the 8-oxodGuo lesion plays a pathogenic role in the mechanism of postnatal neurodevelopmental deficits initiated by in utero physiological fetal ROS formation or EtOHenhanced oxidative stress; and, (2) OGG1 activity in the fetal brain constitutes a determinant of risk.
19
ACKNOWLEDGEMENT The authors are grateful to Kyla Lam in the Department of Pharmacology & Toxicology at the University of Toronto for developing the passive avoidance apparatus and testing procedure used in this study.
20
REFERENCES Abel, E. L. (1982). In utero alcohol exposure and developmental delay of response inhibition. Alcohol Clin Exp Res 6, 369-76. Abramov, J. P. and Wells, P. G. (2011). Embryonic catalase protects against endogenous and phenytoin-enhanced DNA oxidation and embryopathies in acatalasemic and human catalaseexpressing mice. Faseb J 25, 2188-200. Becker, H. C. and Randall, C. L. (1989). Effects of prenatal ethanol exposure in C57BL mice on locomotor activity and passive avoidance behavior. Psychopharmacology (Berl) 97, 40-4. Boiteux, S. and Radicella, J. P. (2000). The human OGG1 gene: structure, functions, and its implication in the process of carcinogenesis. Arch Biochem Biophys 377, 1-8. Brocardo, P. S., Boehme, F., Patten, A., Cox, A., Gil-Mohapel, J. and Christie, B. R. (2012). Anxiety- and depression-like behaviors are accompanied by an increase in oxidative stress in a rat model of fetal alcohol spectrum disorders: Protective effects of voluntary physical exercise. Neuropharmacology 62, 1607-18. Dong, J., Sulik, K. K. and Chen, S. Y. (2010). The role of NOX enzymes in ethanol-induced oxidative stress and apoptosis in mouse embryos. Toxicol Lett 193, 94-100. Fiore, M., Laviola, G., Aloe, L., di Fausto, V., Mancinelli, R. and Ceccanti, M. (2009). Early exposure to ethanol but not red wine at the same alcohol concentration induces behavioral and brain neurotrophin alterations in young and adult mice. Neurotoxicology 30, 59-71. Gilliam, D. M., Stilman, A., Dudek, B. C. and Riley, E. P. (1987). Fetal alcohol effects in longand short-sleep mice: activity, passive avoidance, and in utero ethanol levels. Neurotoxicol Teratol 9, 349-57. Hansen, J. M. and Harris, C. (2013). Redox control of teratogenesis. Reprod Toxicol 35, 165-79. Janik, J., Swoboda, M., Janowska, B., Ciesla, J. M., Gackowski, D., Kowalewski, J., Olinski, R., Tudek, B. and Speina, E. (2011). 8-Oxoguanine incision activity is impaired in lung tissues of NSCLC patients with the polymorphism of OGG1 and XRCC1 genes. Mutat Res 709-710, 21-31. Janzen, E. G., Stronks, H. J., Dubose, C. M., Poyer, J. L. and McCay, P. B. (1985). Chemistry and biology of spin-trapping radicals associated with halocarbon metabolism in vitro and in vivo. Environ Health Perspect 64, 151-70. Jeng, W., Wong, A. W., Ting, A. K. R. and Wells, P. G. (2005). Methamphetamine-enhanced embryonic oxidative DNA damage and neurodevelopmental deficits. Free Radic Biol Med 39, 317-26.
21
Jeng, W., Ramkissoon, A., Parman, T. and Wells, P. G. (2006). Prostaglandin H synthasecatalyzed bioactivation of amphetamines to free radical intermediates that cause CNS regional DNA oxidation and nerve terminal degeneration. Faseb J 20, 638-50. Jeng, W. and Wells, P. G. (2010). Reduced 3,4-methylenedioxymethamphetamine (MDMA, Ecstasy)-initiated oxidative DNA damage and neurodegeneration in prostaglandin H synthase-1 knockout mice. ACS Chem Neurosci 1, 366-80. Jeng, W., Loniewska, M. M. and Wells, P. G. (2013). Brain glucose-6-phosphate dehydrogenase protects against endogenous oxidative DNA damage and neurodegeneration in aged mice. ACS Chem Neurosci 4, 1123-32. Jones, K. L. (2011). The effects of alcohol on fetal development. Birth Defects Res C Embryo Today 93, 3-11. Kershaw, R. M. and Hodges, N. J. (2012). Repair of oxidative DNA damage is delayed in the Ser326Cys polymorphic variant of the base excision repair protein OGG1. Mutagenesis 27, 50110. Kirkali, G., de Souza-Pinto, N. C., Jaruga, P., Bohr, V. A. and Dizdaroglu, M. (2009). Accumulation of (5'S)-8,5'-cyclo-2'-deoxyadenosine in organs of Cockayne syndrome complementation group B gene knockout mice. DNA Repair (Amst) 8, 274-8. Klungland, A., Rosewell, I., Hollenbach, S., Larsen, E., Daly, G., Epe, B., Seeberg, E., Lindahl, T. and Barnes, D. E. (1999). Accumulation of premutagenic DNA lesions in mice defective in removal of oxidative base damage. Proc Natl Acad Sci U S A 96, 13300-5. Koop, D. R. (2006). Alcohol metabolism's damaging effects on the cell: a focus on reactive oxygen generation by the enzyme cytochrome P450 2E1. Alcohol Res Health 29, 274-80. Lee, C. J., Goncalves, L. L. and Wells, P. G. (2011). Embryopathic effects of thalidomide and its hydrolysis products in rabbit embryo culture: evidence for a prostaglandin H synthase (PHS)dependent, reactive oxygen species (ROS)-mediated mechanism. Faseb J 25, 2468-83. Liu, L. and Wells, P. G. (1994). In vivo phenytoin-initiated oxidative damage to proteins and lipids in murine maternal hepatic and embryonic tissue organelles: potential molecular targets of chemical teratogenesis. Toxicol Appl Pharmacol 125, 247-55. Mattson, S. N., Carlos, R. and Riley, E. P. (1993). The behavioral teratogenicity of alcohol is not affected by pretreatment with aspirin. Alcohol 10, 51-7. Mattson, S. N., Crocker, N. and Nguyen, T. T. (2011). Fetal alcohol spectrum disorders: neuropsychological and behavioral features. Neuropsychol Rev 21, 81-101.
22
McCallum, G. P., Wong, A. W. and Wells, P. G. (2011). Cockayne syndrome B protects against methamphetamine-enhanced oxidative DNA damage in murine fetal brain and postnatal neurodevelopmental deficits. Antioxid Redox Signal 14, 747-56. Miller, L., Shapiro, A. M., Cheng, J. and Wells, P. G. (2013). The free radical spin trapping agent phenylbutylnitrone reduces fetal brain DNA oxidation and postnatal cognitive deficits caused by in utero exposure to a non-structurally teratogenic dose of ethanol: A role for oxidative stress. Free Radic Biol Med 60, 223-32. Nicol, C. J., Zielenski, J., Tsui, L. C. and Wells, P. G. (2000). An embryoprotective role for glucose-6-phosphate dehydrogenase in developmental oxidative stress and chemical teratogenesis. Faseb J 14, 111-27. Parman, T., Wiley, M. J. and Wells, P. G. (1999). Free radical-mediated oxidative DNA damage in the mechanism of thalidomide teratogenicity. Nat Med 5, 582-5. Ravanat, J. L., Douki, T., Duez, P., Gremaud, E., Herbert, K., Hofer, T., Lasserre, L., SaintPierre, C., Favier, A. and Cadet, J. (2002). Cellular background level of 8-oxo-7,8-dihydro-2'deoxyguanosine: an isotope based method to evaluate artefactual oxidation of DNA during its extraction and subsequent work-up. Carcinogenesis 23, 1911-8. Riley, E. P., Lochry, E. A. and Shapiro, N. R. (1979). Lack of response inhibition in rats prenatally exposed to alcohol. Psychopharmacology (Berl) 62, 47-52. Stade, B., Ali, A., Bennett, D., Campbell, D., Johnston, M., Lens, C., Tran, S. and Koren, G. (2009). The burden of prenatal exposure to alcohol: revised measurement of cost. Can J Clin Pharmacol 16, e91-102. Wells, P. G., Zubovits, J. T., Wong, S. T., Molinari, L. M. and Ali, S. (1989). Modulation of phenytoin teratogenicity and embryonic covalent binding by acetylsalicylic acid, caffeic acid, and alpha-phenyl-N-t-butylnitrone: implications for bioactivation by prostaglandin synthetase. Toxicol Appl Pharmacol 97, 192-202. Wells, P. G., McCallum, G. P., Chen, C. S., Henderson, J. T., Lee, C. J., Perstin, J., Preston, T. J., Wiley, M. J. and Wong, A. W. (2009). Oxidative stress in developmental origins of disease: teratogenesis, neurodevelopmental deficits, and cancer. Toxicol Sci 108, 4-18. Wells, P. G., McCallum, G. P., Lam, K. C., Henderson, J. T. and Ondovcik, S. L. (2010). Oxidative DNA damage and repair in teratogenesis and neurodevelopmental deficits. Birth Defects Res C Embryo Today 90, 103-9. Winn, L. M. and Wells, P. G. (1995). Phenytoin-initiated DNA oxidation in murine embryo culture, and embryo protection by the antioxidative enzymes superoxide dismutase and catalase: evidence for reactive oxygen species-mediated DNA oxidation in the molecular mechanism of phenytoin teratogenicity. Mol Pharmacol 48, 112-20.
23
Wong, A. W., McCallum, G. P., Jeng, W. and Wells, P. G. (2008). Oxoguanine glycosylase 1 protects against methamphetamine-enhanced fetal brain oxidative DNA damage and neurodevelopmental deficits. J Neurosci 28, 9047-54.
24
LEGENDS Figure 1. Postulated role of oxidatively damaged DNA in the fetal brain, and repair by fetal oxoguanine glycosylase 1 (OGG1), in the mechanism of postnatal behavioural deficits initiated by physiological fetal and/or neonatal formation of reactive oxygen species (ROS), or by in utero exposure to ethanol (EtOH). EtOH can induce ROS-producing fetal NADPH oxidases. The free radical spin trapping agent phenylbutylnitrone (PBN) can detoxify drug and ROS free radical intermediates. OGG1 is the primary enzyme for repairing the DNA lesion 8oxo-2’-deoxyguanosine (8-oxodGuo). Figure 2. Oxidatively damaged DNA in fetal brains from oxoguanine glycosylase 1 (OGG1) wild-type (+/+), and heterozygous (+/-) and homozygous (-/-) knockout mice. Pregnant ogg1 +/- mice bred with +/- males were treated on gestational day (GD) 17 with ethanol (2 g/kg i.p.) or saline vehicle (VH), with or without pretreatment with the free radical spin trapping agent phenylbutylnitrone (PBN) (40 mg/kg i.p.) and sacrificed 6 hr later. Litters containing embryos from all 3 genotypes were analyzed for 8-oxo-2'-deoxyguanosine (8-oxodGuo) formation using an ELISA kit with an 8-oxodGuo-specific antibody. Data were analyzed by two-way ANOVA with a post-hoc Bonferroni test. Single daggers indicate a difference from the vehicle control group for the same genotype (††† = p < 0.001). Asterisks indicate a difference from +/+ fetuses for the same treatment group (* = p < 0.05, *** = p < 0.05). Psi symbols indicate a difference from EtOH-treated fetuses for the same genotype (ȥ = p < 0.05, ȥȥȥ = p < 0.01). Alpha symbol indicates a difference from EtOH-treated fetuses for the same genotype (Į = 0.05 < p < 0.1) The number of fetal brains analyzed is given in parentheses. For each treatment group, fetuses were randomly selected from 3-4 litters. Figure 3. Time course for postnatal learning development in oxoguanine glycosylase 1 (OGG1) wild-type (+/+), heterozygous (+/-) and knockout (-/-) progeny exposed in utero to EtOH with or without PBN pretreatment. Pregnant mice were treated as described in fig. 2, and the progeny were tested postnatally for a passive avoidance test at weeks 6, 9 and 12. Data are shown for trial #3 at 6, 9 and 12 weeks. Data in the bottom four panels are different representations of the same data presented in the top four panels. Data were analyzed by twoway ANOVA with a post-hoc Bonferroni test. Alpha symbols indicate a difference from week 6 of the same genotype (Į = p < 0.05, ĮĮ = p < 0.01, ĮĮĮ = p < 0.001). Beta symbols indicate a difference from week 9 of the same genotype (ȕ = p < 0.05, ȕȕȕ = p < 0.001). Asterisks indicate a difference from +/+ progeny for the same time point (* = p < 0.05, ** = p < 0.01, *** = p < 0.001). Pound symbols indicate a difference from +/- progeny for the same time point (# = p < 0.05, ## = p < 0.01, ### = p < 0.001). The number of progeny tested is given in parentheses. For each treatment group, fetuses were randomly selected from 6-8 litters. No gender differences were observed, therefore the N values represent a combination of male and female progeny. Figure 4. Effect of PBN pretreatment on EtOH-initiated learning deficits in oxoguanine glycosylase 1 (OGG1) wild-type (+/+), and heterozygous (+/-) and homozygous (-/-) knockout progeny. Pregnant mice were treated as described in fig. 2, and the progeny were tested postnatally for a passive avoidance test at weeks 6, 9 and 12. Data are shown for trial #3 at 12 weeks. Data in the bottom panel are a different representation of the same results shown in the
25
top panels. Data were analyzed by two-way ANOVA with a post-hoc Bonferroni test, and symbols indicate the significance of treatment effects. For the statistical significance of genotypic effects, please refer to fig. 3. Daggers indicate a difference from saline for the same genotype († = p < 0.5, †† = p < 0.01, ††† = p < 0.001). Psi symbols indicate a difference from EtOH-treated fetuses for the same genotype (ȥ = p < 0.05, ȥȥ = p < 0.01). The number of progeny tested is given in parentheses. For each treatment group, fetuses were randomly selected from 6-8 litters. No gender differences were observed, therefore the N values represent a combination of male and female progeny.
26
Fig. 1
Fig. 2
Fig. 3
Fig. 4
Graphical Abstract (for review)
Title: Oxidative DNA damage in the in utero initiation of postnatal neurodevelopmental deficits by normal fetal and ethanol-enhanced oxidative stress in oxoguanine glycosylase 1 (ogg1) knockout mice Authors: Lutfiya Miller, Daniel J. Pinto and Peter G. Wells
Highlights • • • • •
DNA repair-deficient ogg1 knockout mice have impaired neurodevelopment In utero ethanol (EtOH) enhances fetal DNA oxidation and impairs neurodevelopment Both EtOH outcomes are greater in ogg1 knockout progeny Both EtOH outcomes in ogg1 progeny are blocked by a free radical spin trap Fetal OGG1 and DNA oxidation are determinants of risk for in utero EtOH outcomes