Toxicology and Applied Pharmacology 251 (2011) 155–162

Contents lists available at ScienceDirect

Toxicology and Applied Pharmacology

j ourna l homepage: www.e lsev ie r.com/ locate /ytaap

XPC is essential for nucleotide excision repair of zidovudine-induced DNA damage inhuman hepatoma cells☆

Qiangen Wu a, Frederick A. Beland a, Ching-Wei Chang b, Jia-Long Fang a,⁎a Division of Biochemical Toxicology, National Center for Toxicological Research, Food and Drug Administration, Jefferson, AR 72079, USAb Division of Personalized Nutrition and Medicine, National Center for Toxicological Research, Food and Drug Administration, Jefferson, AR 72079, USA

☆ The views presented in this article do not necessariland Drug Administration.⁎ Corresponding author. Fax: +1 870 543 7136.

E-mail address: jia-long.fang@fda.hhs.gov (J.-L. Fang

0041-008X/$ – see front matter. Published by Elsevierdoi:10.1016/j.taap.2010.12.009

a b s t r a c t

a r t i c l e i n f o

Article history:Received 18 November 2010Revised 14 December 2010Accepted 16 December 2010Available online 28 December 2010

Keywords:DNA repairNucleotide excision repairXPCZidovudine

Zidovudine (3′-azido-3′-dexoythymidine, AZT), a nucleoside reverse transcriptase inhibitor, can beincorporated into DNA and cause DNA damage. The mechanisms underlying the repair of AZT-inducedDNA damage are unknown. To investigate the pathways involved in the recognition and repair of AZT-induced DNA damage, human hepatoma HepG2 cells were incubated with AZT for 2 weeks and the expressionof DNA damage signaling pathways was determined using a pathway-based real-time PCR array. Compared tocontrol cultures, damaged DNA binding and nucleotide excision repair (NER) pathways showed significantlyincreased gene expression. Further analysis indicated that AZT treatment increased the expression of genesassociated with NER, including XPC, XPA, RPA1, GTF2H1, and ERCC1. Western blot analysis demonstrated thatthe protein levels of XPC and GTF2H1 were also significantly up-regulated. To explore further the function ofXPC in the repair of AZT-induced DNA damage, XPC expression was stably knocked down by 71% using shorthairpin RNA interference. In the XPC knocked-down cells, 100 μM AZT treatment significantly increased [

3H]

AZT incorporation into DNA, decreased the total number of viable cells, increased the release of lactatedehydrogenase, induced apoptosis, and caused a more extensive G2/M cell cycle arrest when compared tonon-transfected HepG2 cells or HepG2 cells transfected with a scrambled short hairpin RNA sequence.Overall, these data indicate that XPC plays an essential role in the NER repair of AZT-induced DNA damage.

y reflect those of the U.S. Food

).

Inc.

Published by Elsevier Inc.

Introduction

Zidovudine (3′-azido-3′-dexoythymidine, AZT), a nucleosidereverse transcriptase inhibitor (NRTI), was the first anti-HIV agentapproved by the U.S. Food and Drug Administration. Over the past twodecades, significant advances have been made in the treatment ofAIDS patients and the prevention of mother-to-child transmission ofHIV-1 by the use of NRTIs. The Pediatric AIDS Clinical Trials Group(PACTG) Protocol 076 demonstrated that a regimen consisting of AZTgiven to HIV-1 infected women during pregnancy and to their infantsafter birth reduced the risk of HIV-1 vertical transmission byapproximately two thirds (Connor et al., 1994). Combining antire-troviral therapy with elective cesarean delivery and the avoidance ofbreastfeeding has reduced maternal–infant HIV infection to 1%–2%(Mofenson, 2010). The success in reducing HIV vertical transmissionthrough the use of NRTIs has led to concerns regarding known andunknown health risks to the drug-exposed mothers and, in particular,their fetuses.

Structurally, AZT is an analogue of thymidine in which the 3′-hydroxyl group is replaced by an azido group. AZT is phosphorylatedby successive cellular kinases to give AZT 5′-triphosphate, which canbe incorporated into host nuclear and mitochondrial DNA, andsubsequently cause DNA chain termination (Sommadossi et al.,1989; Tosi et al., 1992; Sussman et al., 1999). The toxic manifestationsof AZT–DNA incorporation are well documented and include theinduction of micronuclei, sister chromatid exchange, chromosomalaberrations, mitochondrial damage, decreases in telomere length, andincreases in intracellular production of reactive oxygen species(Gonzalez Cid and Larripa, 1994; Bialkowska et al., 2000; IARC,2000; Bishop et al., 2004a; Von Tungeln et al., 2004; Ji et al., 2005;Desai et al., 2009).

Chronic exposure to AZT induces abnormal differentiation invaginal epithelium and vaginal tumors in experimental animals(Olivero et al., 1994; Ayers et al., 1996; NTP, 1999), and mice exposedtransplacentally to AZT have increased incidences of tumors in anumber of organs including the liver, lung, mammary gland, andovaries (Olivero et al., 1997; Diwan et al., 1999; NTP, 2006; Walker etal., 2007). In other studies, transplacental or neonatal administrationof AZT to mice resulted in an increase mutation frequency and aunique pattern of mutations in the Tk gene (Von Tungeln et al., 2002,2004, 2007; Mittelstaedt et al., 2004). The genotoxicity and carcino-genicity of AZT in experimental animals raise concern about the

156 Q. Wu et al. / Toxicology and Applied Pharmacology 251 (2011) 155–162

potential of similar toxicities in humans, and indicate the need forlong-term follow-up of AZT exposures (Bishop et al., 2004b;Walker etal., 2007).

We previously reported significant dose-dependent decreases inthe number of viable cells in a human hepatoma cell line (HepG2)after 4 weeks AZT treatment and a 1-week recovery period (Fangand Beland, 2009). The decrease in cell viability was attributed to acombination of effects, including a delay of cell cycle progression,the induction of apoptosis, and a decrease in telomerase activity. Thegenotoxicity of AZT has also been demonstrated in the humanlymphoblastoid TK6 cells, and human peripheral or cord blood cells(Meng et al., 2000; Escobar et al., 2007).

The presence of DNA damage induced by AZT may lead to cellcycle checkpoint arrest to allow time for DNA repair processes tooccur (Humer et al., 2008; Fang et al., 2009; Fang and Beland,2009). If the AZT incorporated into DNA is not removed, the cellsmay become cancerous due to the accumulation of mutationsresulting from replication of damaged DNA. Therefore, it isimportant to understand the mechanisms by which DNA isrepaired upon the incorporation of AZT into DNA. Vazquez-Paduaet al. (1990) demonstrated the presence of an exonuclease thatremoved AZT in human leukemic K562 cells in vitro, and Skalski etal. (1995) reported the presence of another 3′–5′ exonuclease thatremoved AZT incorporated into DNA in human acute lymphoblas-tic leukemia H9 cells; however, additional information wouldclearly be useful.

Since nucleotide excision repair (NER) is the major repair systemfor removing bulky DNA lesions formed by exposure to chemicals orradiation (Sancar et al., 2004), we postulated that the DNA damageinduced by AZT may be recognized and removed by NER ineukaryotic cells. In this study, we initially investigated theexpression of genes related to DNA damage and repair response inHepG2 cells treated with AZT by using a DNA damage signalingpathway-based PCR array, which indicated the importance of theNER pathway in the repair of AZT-induced DNA damage. Thisresponse was confirmed by Western blotting. Subsequently, weknocked down XPC, a key damage recognition factor in the humanNER pathway, by using a short hairpin (sh) RNA interferencetechnique and studied the effects of the XPC knock down on thetoxicities associated with AZT treatment.

Materials and methods

Chemicals. AZT was purchased from Cipla Ltd. (Mumbai, India).[3H]AZT (12.7 Ci/mmol) was obtained from Moravek Biochemicals,Inc. (Brea, CA). Williams’ Medium E, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), propidium iodide (PI), 5-bromo-2′-deoxyuridine (BrdU), bovine serum albumin (BSA), RNaseA, and the lactate dehydrogenase (LDH) kit were acquired fromSigma-Aldrich Co. (St. Louis, MO). Penicillin–streptomycin and 2.5%trypsin were purchased from Fisher Scientific (Pittsburgh, PA).Dulbecco's phosphate buffered saline (PBS) and SuperScript III first-strand synthesis kits were obtained from Invitrogen Life Technolo-gies (Carlsbad, CA). Fetal bovine serum (FBS) was acquired fromAtlanta Biologicals (Lawrenceville, GA). RNeasy Mini kits werepurchased from Qiagen Sciences (Germantown, MD). RT2 Profiler™PCR arrays, SureSilencing shRNA plasmids, and SureFECT transfec-tion reagent were bought from SABioscience (Frederick, MD).Neomycin (G418) was obtained from MP Biomedicals (Solon, OH).The BCA Protein Assay kit and RIPA buffer were obtained from PierceBiotechnology (Rockford, IL). The APO-BrdU kit was acquired fromBD Biosciences (San Jose, CA). Complete protease inhibitor cocktailwas purchased from Roche Applied Science (Mannheim, Germany).All other chemicals and biochemicals were of analytical grade andused without further purification.

Antibodies. Antibodies to XPC, XPA, RPA1, and ERCC1 were pur-chased from Santa Cruz Biotechnology (Santa Cruz, CA). The antibodyto GTF2H1 was obtained from Abcam Inc. (Cambridge, MA). Thefluorescein isothiocyanate (FITC)-conjugated anti-BrdU monoclonalantibody (clone B44) was acquired from BD Biosciences.

Cell culture and treatments. HepG2 cells (American Type CultureCollection, VA) were cultured in Williams’ E medium supplementedwith 10% FBS and antibiotics at 37 °C in a humidified atmospherecontaining 95% air and 5% CO2. AZT was dissolved in PBS. The cellswere treated with 2, 20, or 100 μM AZT for two weeks. Our previousstudy (Fang and Beland, 2009) indicated a 2-week exposure gave themaximum responses in HepG2 cells treated with AZT. HepG2 cellstransfected with a scrambled shRNA sequence (HepG2/Vector) or ashRNA plasmid targeting XPC (HepG2/XPCKD) were treated with100 μM AZT for 48 h or 1 week depending upon the specificmeasurement to be made. The incorporation of AZT into DNA wasmeasured after 48 h to allow comparisons with our previous results(Fang and Beland, 2009), while a 1-week exposure was selected forcomparing the knocked-down cells with the normal parental cellsbased upon our previous experiments that showed this is a sufficientlength of time to assess biochemical changes induced by exposure toAZT (Fang and Beland, 2009). Control cells were fed completemedium free of AZT. Each of the incubations was performed intriplicate.

Pathway-specific real-time PCR array. At the end of the AZTtreatments, total RNA was isolated from the treated and controlcells using RNeasy Mini kits. First strand cDNA synthesis wasconducted with SuperScript III first-strand synthesis system for RT-PCR according to the manufacturer's protocol. Subsequently, a humanDNA damage signaling RT2 Profile PCR array was used to determinethe effect of AZT treatment on the expression of a total of 84 geneswith known roles in the response to repair of DNA damage.Quantitative real-time PCR was performed using a Bio-Rad iCyclerreal-time PCR system (Bio-Rad laboratories, Hercules, CA) followingthe manufacturer's instructions. The Ct values for the 84 genes werenormalized to five housekeeping genes. Comparisons between AZTtreated groups and control groups were conducted by the ΔΔCt

method to assess the relative changes in gene expression levels.

Western blotting. At the end of the AZT treatment, the cells weretrypsinized and washed three times with PBS. Forty microliters of cellpellets was lysed in five volumes of RIPA buffer supplemented withcomplete protease inhibitor cocktail. After incubation on ice for30 min, followed by sonication for 30 s with a 50% pulse, thesupernatants were collected by centrifugation at 14,000 g for 20 minand stored in aliquots at −65 °C for further analysis. The amount ofprotein in the cell lyses was determined by the BCA protein assay withBSA as the standard. Forty micrograms of cell lysates was separated by12% SDS-polyacrylamide gel electrophoresis and electrophoreticallytransferred onto PVDF membranes. The blots were blocked with 5%milk and probed with anti-XPC, XPA, RPA1, GTF2H1, or ERCC1antibodies followed by a secondary antibody conjugated to HRP. β-Actin protein levels were detected in the same manner and used as aloading control. The intensity of each bandwas quantifiedwith ImageJsoftware (NIH, Bethesda, MD).

Stable suppression of XPC expression with shRNA. HepG2 cells weretransfected with the SureSilencing shRNA plasmids that target humanXPC by using SureFECT transfection reagent according to themanufacturer's protocol. A vector with a scrambled artificial sequencewas transfected as a negative control. Single cell clones were selectedand propagated in the presence of 800 μg/ml G418 and then screenedfor XPC expression by real-time PCR and Western blotting. The clonethat exhibited the lowest expression level of XPC (HepG2/XPCKD) and

Table 1Analysis of DNA repair pathways, as assessed by MANOVA, in HepG2 cells.

Gene set Subgroup pathway Numberof gene

AZT (P valuea)

2 μM 20 μM 100 μM

DNA repair 52 0.027 0.004 0.001Damaged DNA binding 26 0.097 0.037 0.034Base excision repair 7 0.173 0.092 0.087Double strand break repair 9 0.095 0.095 0.060Mismatch repair 14 0.209 0.029 0.084Nucleotide excision repair 5 0.019 0.001 0.001

a Compare with that of control. Values in bold are significant at pb0.05.

157Q. Wu et al. / Toxicology and Applied Pharmacology 251 (2011) 155–162

a negative plasmid transfected clone (HepG2/Vector) were chosen forthe subsequent experiments.

[3H]AZT incorporation into DNA. The incorporation of [3H]AZT intonuclear DNA of HepG2, HepG2/Vector, and HepG2/XPCKD cells wasquantified as described previously (Fang and Beland, 2009). Briefly,the cells were cultured for 48 hwith 0 or 100 μM [3H]AZT (diluted to aspecific activity of 20 μCi/μmol with unlabeled AZT). After the AZTtreatment, the cells were harvested and cellular DNA was isolated byconventional phenol–chloroform extraction. The incorporation of [3H]AZT into the DNA was quantified by liquid scintillation techniques.

Cell proliferation. The number of viable cells was assessed by anMTT assay as previously described (Fang et al., 2009) with minormodifications. Briefly, HepG2, HepG2/Vector, and HepG2/XPCKD cellswere treated with 100 μMAZT in 96-well microtitration plates for oneweek. Control cells were fed complete medium free of AZT. After thetreatment, the 96-well cultures were incubated with 500 μg/ml MTTfor 4 h at 37 °C. Subsequently, the MTT solution was removed, and thereduced MTT crystals were dissolved in 200 μl DMSO following theaddition of 25 μl Sorensen's glycine buffer (100 mM glycine, 100 mMNaCl, pH 10.5). The absorbance was determined with a BioTekSynergy 2 reader (BioTek Instruments, Inc., Winooski, VT) at 570 nm,with 620 nm as the reference.

Cell necrosis. The percent of necrotic cells was measured with anLDH kit according to the manufacturer's instructions. Briefly,HepG2, HepG2/Vector, and HepG2/XPCKD cells were incubatedwith or without 100 μM AZT for one week. The cell culturemedium was collected after centrifugation at 250 g for 4 min. LDHreleased into the collected medium was measured by determiningthe absorbance at 490 nm, with 690 nm as the reference, in aBioTek Synergy 2 reader. The total LDH level was quantified in thesame manner to calculate the percent of LDH released into the cellculture medium.

Cell apoptosis. Apoptotic cells were analyzed after AZT treatmentusing an APO-BrdU kit as described previously (Fang et al., 2009; Fangand Beland, 2009). Briefly, HepG2, HepG2/Vector, and HepG2/XPCKD

cells were treated with or without 100 μM AZT for 1 week. Followingtreatment, cells were harvested, fixed in 1% paraformaldehyde, andpermeabilized using ice-cold 70% ethanol. The fixed cells wereincubated with a DNA labeling solution containing BrdU-triphosphateand terminal dexoynucleotidyl transferase at 37 °C for 1 h. The cellswere then doubly stained with an FITC-labeled anti-BrdU monoclonalantibody solution and PI/RNase A solution before being analyzed on aBecton Dickinson FACSCalibur flow cytometer (BD Biosciences, SanJose, CA).

Cell cycle analysis. The effects of AZT on cell cycle progression weredetermined by flow cytometry analysis. HepG2, HepG2/Vector, andHepG2/XPCKD cells were treated with or without 100 μM AZT for1 week. At the end of the treatment, the cells were collected and2×106 cells were incubated in completed medium containing 10 μMBrdU at 37 °C for 1 h. The cells were then fixed and stained with FITC-conjugated anti-BrdU monoclonal antibody and PI as previouslydescribed (Fang and Beland, 2009). The stained cells were analyzedwith CellQuest software on a Becton Dickinson FACSCalibur flowcytometer.

Statistical analysis. Data are presented as X±SD. Two-sidedmultivariate analysis of variance (MANOVA) was used to assess thesignificance of increases in the expression of pathways from real-timePCR array data (Tsai and Chen, 2009). Dose-related trends wereinvestigated by regression analysis. One-way analysis of variance(ANOVA) followed by pairwise-comparisons using Dunnett's tests

was used to determine the significance between specific AZTtreatments and control cells. Data were logarithm transformed ifthey failed in variance homogeneity tests. The difference wasconsidered statistically significant when the P value was less than0.05.

Results

Expression of DNA damage and repair related genes after AZT treatment

The effect of a two-week AZT treatment on the expression profileof 84 genes related to DNA damage and repair was investigated inHepG2 cells. After the treatment, seven genes (8.3%) were up-regulated (N2-fold) at 2 μM AZT, 35 genes (41.7%) at 20 μM AZT, and22 genes (26.2%) at 100 μM AZT. None of the genes were down-regulated (N2-fold) in AZT treated HepG2 cells.

Of the 84 genes on the DNA damage signaling pathway PCRarray, 52 are directly associated with DNA repair. MANOVAanalysis indicated that the expression of the DNA repair pathwaywas significantly increased at 2, 20, and 100 μM AZT in the HepG2cells (Table 1). The genes associated with DNA repair pathwaywere subdivided into groups involved in damaged DNA binding,base excision repair, double strand break repair, mismatch repair,and NER. MANOVA analysis of these subgroups indicated that inHepG2 cells the expression of the damaged DNA binding pathwaywas increased significantly at 20 and 100 μM AZT and the NERpathway was increased significantly at 2, 20, and 100 μM AZT(Table 1).

Relative gene expression of NER subpathway in AZT-treated HepG2 cells

The relative level expression of genes involved in nucleotideexcision repair (XPC, XPA, RPA1, GTF2H1, and ERCC1) was furtheranalyzed in the cells after two weeks of AZT treatment. As shown inFig. 1, AZT induced a dose-related increasing trend in the expressionof each gene. The expression of XPC was increased significantly at 20and 100 μMAZT (2.0- and 3.2-fold, respectively), while the expressionof XPAwas increased significantly at all three dose levels of AZT (2.5 to4.2-fold). Significant increases in the expression of RPA1 (2.0-fold)and GTF2H1 (3.0-fold) were observed at 20 and 100 μM AZT,respectively.

NER protein expression in HepG2 cells treated with AZT for 2 weeks

The protein levels of XPC, XPA, RPA1, GTF2H1, and ERCC1 in HepG2cells after two weeks of AZT treatment are shown in Fig. 2. XPC andGTF2H1 showed a dose-related increasing trend in protein levels,with the difference being significant at 100 μM AZT.

Stable XPC knockdown with shRNA interference

The increase in gene expression and protein levels suggested thatthe NER pathway might be recruited to repair AZT-induced DNAdamage in HepG2 cells. Since XPC plays a key role in DNA damage

Fig. 1. Relative gene expression levels of the NER pathway in AZT-treated HepG2 cells.Relative gene levels ofNER inHepG2 cells treatedwith 2, 20, or 100 μMAZT for twoweeks.Data are expressed as the fold change (X±SD) for each gene as determined from threeindependent experiments. F, significant (Pb0.05) dose-related trend; *, significantly(Pb0.05) different from the control group.

158 Q. Wu et al. / Toxicology and Applied Pharmacology 251 (2011) 155–162

recognition, this protein was knocked down stably in HepG2 cells bythe use of shRNA. Analysis of the selected HepG2/XPCKD cloneindicated a reduction in the XPC mRNA levels of 71% (95% confidenceinterval: 60–79%) compared to HepG2 cells transfected with ascrambled shRNA sequence (HepG2/Vector). The knockdown of XPCin the HepG2/XPCKD cells was confirmed by Western blotting, asshown in Fig. 3.

XPC knockdown increases [3H]AZT DNA incorporation

AZT is phosphorylated intracellularly andbecomes incorporated intoDNA as a chain elongation terminator, which can have biologicalconsequences including decreased cell growth and a delay in cell cycleprogression (Roskrow and Wickramasinghe, 1990; Humer et al., 2008;Fang et al., 2009; Fang and Beland, 2009). The levels of AZTincorporation into the DNA of the HepG2, HepG2/Vector, and HepG2/XPCKD cells were determined by incubating the cells with 100 μM [3H]AZT for 48 h. In HepG2 and HepG2/Vector cells, the levels of AZTincorporation into DNAwere 228±34 and 287±32AZTmolecules/106

nucleotides (X±SD for 3 independent experiments), a difference thatwas not significant. In contrast, HepG2/XPCKD cells exposed to 100 μM

Fig. 2. NER protein levels in AZT-treated HepG2 cells. Protein levels of XPC, XPA, RPA1, GTF2two weeks. Data are expressed as the X±SD from three independent experiments. F, signigroup.

AZT had a significant increase of AZT–DNA incorporation (309±18 AZTmolecules/106 nucleotides) compared to parent HepG2 cells. Theincreased incorporation of AZT into the DNA of HepG2/XPCKD cellssuggests that down-regulation of XPC could contribute to an increasedsensitivity of HepG2/XPCKD cells to AZT.

AZT treatment reduces HepG2 cell proliferation to a greater extent in XPCdown-regulated cells

Our previous studies showed that AZT inhibited HepG2 cellproliferation in a dose-dependent manner (Fang and Beland, 2009).We investigated whether down-regulation of XPC expression wouldattenuate further cell proliferation in HepG2 cells by AZT treatment. Inparent HepG2 cells and HepG2/Vector cells treated with 100 μM AZTfor 1 week, the relative number of viable cells was 22% and 25%,respectively, when compared to their appropriate control cells(Fig. 4). In HepG2/XPCKD cells treated with 100 μM AZT for 1 week,the relative number of viable cells was 12%, a difference that wassignificant compared to parent HepG2 cells and HepG2/Vector cells.

AZT treatment increases cell necrosis in HepG2 cells after XPC down-regulation

To test whether XPC knock-down cells would be more likely tohave necrotic cell death, the relative amount of cytoplasmic LDHreleased into the culture medium was investigated in parent HepG2cells, HepG2/Vector cells, and HepG2/XPCKD cells treated with 100 μMAZT for one week. As shown in Fig. 5, a significant increase of relativeLDH release was observed in HepG2/XPCKD cells but not the parentHepG2 or HepG2/Vector cells after AZT treatment. The increased LDHrelease indicated that AZT could induce necrotic cell death in HepG2cells with down-regulated XPC expression.

AZT treatment increases apoptosis of HepG2 cell to a greater extent afterXPC knockdown

An increased apoptosis might also contribute to the reduced cellproliferation induced by exogenous DNA-damaging agents. Fig. 6shows the relative extent of apoptotic cell death, as measured by flowcytometry using an APO-BrdU kit, in HepG2, HepG2/Vector, andHepG2XPCKD cells treated for 1 week with 0 or 100 μM AZT. In HepG2and HepG2/Vector cells, the AZT treatment increased the relativelevels of apoptosis significantly to 170% and 155%, respectively. InHepG2/XPCKD cells treated with 100 μM AZT for 1 week, the relativelevels of apoptosis increased further to 212%, a difference that was

H1, and ERCC1, as determined by Western blotting, in HepG2 cells treated with AZT forficant (Pb0.05) dose-related trend; *, significantly (Pb0.05) different from the control

Fig. 3. XPC protein levels in HepG2, HepG2/Vector, and HepG2/XPCKD cells. HepG2 cellswere stably transfected with constructs encoding XPC shRNA or a scrambled sequenceas a control. Whole cell lysates were prepared from untransfected (HepG2), vectortransfected (HepG2/Vector), or XPC shRNA transfected (HepG2/XPCKD) cells andseparated on 12% SDS-PAGE.

Fig. 5. LDH release into culture medium from HepG2, HepG2/Vector, and HepG2/XPCKD

cells treated with 0 or 100 μM AZT for one week. The percentage of LDH release intoculture medium was measured relative to total LDH activity. The results shown are theX±SD of 3 independent experiments. *, significantly (Pb0.05) different from thecontrol group; #, significantly (Pb0.05) different from HepG2/XPCKD cells.

159Q. Wu et al. / Toxicology and Applied Pharmacology 251 (2011) 155–162

significant when compared to HepG2 cells or HepG2/Vector cells.Thus, AZT treatment appears to be more likely to induce apoptosis inthe XPC down-regulated HepG2 cells.

AZT treatment induce more extensive G2/M phase arrest of HepG2 cellswith XPC knockdown

To clarify further the role of cell cycle progression in the inhibitionof proliferation in HepG2/XPCKD cells induced by AZT, the cell cycledistributionwas assessed using a flow cytometric assay after a 1-weektreatment with 100 μM AZT in parent HepG2, HepG2/Vector, andHepG2/XPCKD cells. Previously, we reported that short-term AZTtreatment caused S and G2/M phase cell cycle arrest that contributedto the growth inhibition of HepG2 cells treated with AZT (Fang andBeland, 2009). In agreement with our previous data, we observed adecrease in G1/G0 phase and a simultaneous increase of cells in Sphase and G2/M population in parent HepG2 cells treated with100 μM AZT for one week (Fig. 7). A similar response was observedwith HepG2/Vector cells. With the HepG2/XPCKD cells, AZT treatmentinduced a greater decrease in G1/G0 phase cells and increase in G2/Mphase cells compared to parent HepG2 and HepG2/Vector cells treatedwith AZT.

Discussion

In this study, we investigated the pathways involved in therecognition and repair of AZT-induced DNA damage in humanhepatoma cells. Initially, we used a pathway-based array to examinethe effects of AZT treatment on the expression levels of genes relatedto DNA damage and repair pathways. The increased gene and proteinlevels of the NER pathway suggested that NER participated in AZT-induced DNA damage repair. We further demonstrated that the

Fig. 4. Cell proliferation (as measured by a MTT assay) in HepG2, HepG2/Vector, andHepG2/XPCKD cells treated with 0 or 100 μM AZT for 1 week. The results are presentedas the X±SD of 3 independent experiments. *, significantly (Pb0.05) different from thecontrol group; #, significantly (Pb0.05) different from HepG2/XPCKD cells.

knocking down of XPC led to increased AZT–DNA incorporation,decreased in cell growth, induced apoptosis and necrotic cell death,and increased cell cycle arrest after AZT treatment. These resultssuggest that XPC is crucial for DNA damage recognition and repairafter AZT treatment in human hepatoma cells.

As an analogue of thymidine, in which the 3'-hydroxyl group isreplaced by an azido group, AZT is phosphorylated by cellularkinases to AZT 5′-triphosphate, which can be incorporated intohost nuclear and mitochondrial DNA (Sommadossi et al., 1989;Tosi et al., 1992; Sussman et al., 1999). Using [3H]AZT, we found228 AZT molecules/106 nucleotides incorporated into the DNA ofthe parent HepG2 cells and 287 AZT molecules/106 nucleotides inHepG2/Vector cells treated with 100 μM [3H]AZT for 48 h, which issimilar to what we have reported previously with HepG2 cells(Fang and Beland, 2009). Since the levels of AZT–DNA incorpora-tion were not significantly different between the parent andshRNA vector transfected HepG2 cells, this suggests that shRNAtransfection, per se, does not effect AZT–DNA incorporation inHepG2 cells. The more extensive [3H]AZT–DNA incorporation (309AZT molecules/106 nucleotides ) in HepG2/XPCKD cells suggeststhat less removal of AZT occurs in the XPC down-regulated HepG2cells due to the deficiency of initial DNA damage recognition byXPC in the NER pathway. Only a few studies have highlighted thepossible role of DNA repair in the removal of AZT from DNA(Mamber et al., 1990; Vazquez-Padua et al., 1990; Skalski et al.,1995; Bialkowska et al., 2000; Slamenova et al., 2006; Fang and

Fig. 6. Relative apoptosis levels, as measured using an APO-BrdU kit, in HepG2, HepG2/Vector, and HepG2/XPCKD cells treated with 0 or 100 μM AZT for one week. The resultsshown are the X±SD of 3 independent experiments. *, significant (Pb0.05) differentfrom the control group; #, significant (Pb0.05) different from the AZT treated HepG2/XPCKD cells.

Fig. 7. Cell cycle distribution, as determined by flow cytometry, in HepG2, HepG2/Vector, and HepG2/XPCKD cells treated with 0 or 100 μM AZT for one week. The resultsshown are the X±SD of four independent experiments. *, significantly (Pb0.05)different from 0 μM AZT (control); #, significantly (Pb0.05) different from HepG2/XPCKD cells.

160 Q. Wu et al. / Toxicology and Applied Pharmacology 251 (2011) 155–162

Beland, 2009). The activation of a checkpoint pathway that arreststhe cell cycle to permit DNA repair and the fact that AZT effectswere partially or totally restored during the recovery periodsuggested that there was active removal of AZT from DNA (Fangand Beland, 2009). Further studies are needed to determine thepossible role of XPC in the AZT–DNA incorporation removalprocess following AZT treatment.

When AZT becomes incorporated into host DNA chains, it lacks thenecessary ribose moiety required for DNA chain extension andtherefore acts as a DNA chain terminator. The genotoxic manifesta-tions of this incorporation, including micronuclei (Von Tungeln et al.,2004, 2007; Borojerdi et al., 2009), sister chromatic exchange,chromosomal aberrations (Olivero et al., 1994; Wutzler and Thust,2001), and telomere shortening (Zhou et al., 2007), have been welldocumented (IARC, 2000). AZT treatment has been shown to induceapoptosis in mice skeletal muscle cells (Desai et al., 2009), micemyocardium (Purevjav et al., 2007), human adipocytes (McComsey etal., 2005), human liver HepG2 and THLE2 cells (Fang and Beland,2009), and human parathyroid cancer cells (Falchetti et al., 2005). Thecurrent study demonstrated increased apoptosis levels in parentHepG2 and HepG2/Vector cells following AZT treatment, at levels

comparable to our previous results (Fang and Beland, 2009).Furthermore, the greater extent of apoptotic cell death in XPCdown-regulated HepG2 cells indicates an important role of XPC viathe NER pathway in the apoptotic cell death as a result of AZT-inducedDNA damage. The defect in the initial DNA damage recognition by XPCappears to have increased the sensitivity of HepG2 cells to apoptosisfollowing AZT treatment. A similar essential role for XPC in the NERprocess has been demonstrated in the induction of apoptotic celldeath by the anticancer drug cisplatin (Colton et al., 2006).

Cell cycle perturbation as a consequence of AZT induced DNAdamage has been observed in eukaryotic cells, including NIH 3 T3 cells(Fang et al., 2009), human HepG2 cells, and THLE2 cells (Fang andBeland, 2009), human chronic myeloid leukemia cell lines (Wu et al.,2004), HL60 cells (Roskrow and Wickramasinghe, 1990), humancolon carcinoma cells (Chandrasekaran et al., 1995), human Tlymphocytes (Fridland et al., 1990), and human melanoma cells(Humer et al., 2008). S-phase arrest and a concomitant decrease of thepercentage of cells in G0/G1 were found after short-term AZTtreatment (Roskrow and Wickramasinghe, 1990; Chandrasekaran etal., 1995; Wu et al., 2004). Flow cytometry analysis of the cell cycleprofiles in HepG2 or HepG2/Vector cells exposed to 100 μM AZT for1 week revealed a slight, statistically insignificant, increase in thepercentage of cells in S and G2/M phase accompanied by a reductionof G1/G0 phase (Fig. 7). These data in combination with our previousstudies of AZT (Fang and Beland, 2009), suggest that S and G2/Mphase arrest of the cell cycle is involved in the AZT-dependentinhibition of cell proliferation in HepG2 cells exposed to 100 μM AZTfor 1 week. Several studies have indicated that the presence of DNAdamage induced by AZT can lead to cell cycle checkpoint arrest toallow for DNA repair (Humer et al., 2008; Fang et al., 2009; Fang andBeland, 2009). The more extensive G2/M phase arrest in HepG2/XPCKD cells indicates the important role of XPC in the process of cellcycle arrest for DNA repair and cell survival after AZT treatment.

Earlier studies have shown that the binding of the XPC-HR23Bcomplex to the site of DNA damage is essential for recruiting otherNER components and initiating the NER process (Sugasawa et al.,1998, 2001, 2002; Wood, 1999; Batty and Wood, 2000; Volker et al.,2001; Tapias et al., 2004). An initial recognition role by XPC of AZT-induced DNA damage was suggested by our pathway-based real-timePCR array and western results. Additionally, down-regulation of XPCresulted in a greater inhibition of cell proliferation, increased AZT-DNA incorporation, the induction of apoptotic and necrotic cell death,and G2/M cell cycle arrest in HepG2/XPCKD cells treated with AZT.These results provide insights into the mechanism by which the AZT–DNA incorporation is removed by the NER repair pathway afterrecognition by XPC of the DNA damage.

Although this study focused on the role of the XPC in NER responseto AZT treatment, other NER factors or pathwaysmay also be recruitedto repair the DNA damage after AZT treatment. After being recognizedby the XPC-HR23B complex, the damaged DNA structure is targetedby GTF2H1, which recruits the other NER factors upon the addition ofATP to repair the DNA damage (Sugasawa et al., 1998; Volker et al.,2001; Riedl et al., 2003; Tapias et al., 2004). To understand fully themechanisms involved in the response to AZT–DNA incorporation,further studies are being conducted to determine the effects of otherNER factors, in combinationwithXPC, on the repair of AZT-inducedDNAdamage. In conclusion, the results obtained from this study demon-strated for thefirst time that XPC, a crucial factor forNER pathway, playsan essential role in the repair AZT-induced DNA damage in humanhepatocytes.

Supplementarymaterials related to this article can be found onlineat doi:10.1016/j.taap.2010.12.009.

Conflict of interest statement

The authors declare that there are no conflicts of interest.

161Q. Wu et al. / Toxicology and Applied Pharmacology 251 (2011) 155–162

Acknowledgments

QWwas supported by an appointment to the Postgraduate ResearchProgram in the Division of Biochemical Toxicology at the NationalCenter for Toxicological Research administered by Oak Ridge Institutefor ScienceEducation through an interagency agreementbetween theU.S. Department of Energy and the FDA.

Funding. This research was supported by Interagency Agreement224-07-0007 between the National Center for Toxicological Research/U.S. Food and Drug Administration, and the National Institute forEnvironmental Health Sciences/National Toxicology Program.

References

Ayers, K.M., Clive, D., Tucker, W.E., Hajian, G., de Miranda, P., 1996. Nonclinical toxicologystudies with zidovudine: genetic toxicity tests and carcinogenicity bioassays in miceand rats. Fundam. Appl. Toxicol. 32, 148–158.

Batty, D.P., Wood, R.D., 2000. Damage recognition in nucleotide excision repair of DNA.Gene 241, 193–204.

Bialkowska, A., Bialkowski, K., Gerschenson, M., Diwan, B.A., Jones, A.B., Olivero, O.A.,Poirier, M.C., Anderson, L.M., Kasprzak, K.S., Sipowicz, M.A., 2000. Oxidative DNAdamage in fetal tissues after transplacental exposure to 3′-azido-3′-deoxythymidine(AZT). Carcinogenesis 21, 1059–1062.

Bishop, J.B., Tani, Y., Witt, K., Johnson, J.A., Peddada, S., Dunnick, J., Nyska, A., 2004a.Mitochondrial damage revealed by morphometric and semiquantitative analysis ofmouse pup cardiomyocytes following in utero and postnatal exposure to zidovudineand lamivudine. Toxicol. Sci. 81, 512–517.

Bishop, J.B., Witt, K.L., Tice, R.R., Wolfe, G.W., 2004b. Genetic damage detected in CD-1mouse pups exposed perinatally to 3′-azido-3′-deoxythymidine and dideox-yinosine via maternal dosing, nursing, and direct gavage. Environ. Mol. Mutagen.43, 3–9.

Borojerdi, J.P.,Ming, J., Cooch, C.,Ward, Y., Semino-Mora, C., Yu,M., Braun, H.M., Taylor, B.J.,Poirier, M.C., Olivero, O.A., 2009. Centrosomal amplification and aneuploidy inducedby the antiretroviral drug AZT in hamster and human cells. Mutat. Res.-Fundam. Mol.Mech. Mutagen. 665, 67–74.

Chandrasekaran, B., Kute, T.E., Duch, D.S., 1995. Synchronization of cells in the S-phase ofthe cell cycle by3′-azido-3′-deoxythymidine: implications for cell cytotoxicity. CancerChemother. Pharmacol. 35, 489–495.

Colton, S.L., Xu, X.X.S.,Wang,Y.A.,Wang,G., 2006. The involvement of ataxia-telangiectasiamutated protein activation in nucleotide excision repair-facilitated cell survival withcisplatin treatment. J. Biol. Chem. 281, 27117–27125.

Connor, E.M., Sperling, R.S., Gelber, R., Kiselev, P., Scott, G., O'Sullivan, M.J., VanDyke,R., Bey, M., Shearer, W., Jacobson, R.L., 1994. Reduction of maternal–infanttransmission of human immunodeficiency virus type 1 with zidovudinetreatment. Pediatric AIDS Clinical Trials Group Protocol 076 Study Group. N.Engl. J. Med. 331, 1173–1180.

Desai, V.G., Lee, T., Moland, C.L., Branham,W.S., Von Tungeln, L.S., Beland, F.A., Fuscoe,J.C., 2009. Effect of short-term exposure to zidovudine (AZT) on the expression ofmitochondria-related genes in skeletal muscle of neonatal mice. Mitochondrion9, 9–16.

Diwan, B.A., Riggs, C.W., Logsdon, D., Haines, D.C., Olivero, O.A., Rice, J.M., Yuspa, S.H., Poirier,M.C., Anderson, L.M., 1999.Multiorgan transplacental and neonatal carcinogenicity of 3′-azido-3′-deoxythymidine in mice. Toxicol. Appl. Pharmacol. 161, 82–99.

Escobar, P.A., Olivero, O.A., Wade, N.A., Abrams, E.J., Nesel, C.J., Ness, R.B., Day, R.D., Day, B.W.,Meng,Q.X., O'Neill, J.P.,Walker, D.M., Poirier,M.C.,Walker, V.E., Bigbee,W.L., 2007.Genotoxicity assessed by the Comet and GPA assays following in vitro exposure ofhuman lymphoblastoid cells (H9) or perinatal exposure of mother-child pairs to AZTor AZT-3TC. Environ. Mol. Mutagen. 48, 330–343.

Falchetti, A., Franchi, A., Bordi, C.,Mavilia, C.,Masi, L., Cioppi, F., Recenti, R., Picariello, L.,Marini,F., Del Monte, F., Ghinoi, V., Martineti, V., Tanini, A., Brandi, M.L., 2005. Azidothymidineinduces apoptosis and inhibits cell growth and telomerase activity of humanparathyroidcancer cells in culture. J. Bone Miner. Res. 20, 410–418.

Fang, J.-L., Beland, F.A., 2009. Long-term exposure to zidovudine delays cell cycleprogression, induces apoptosis, and decreases telomerase activity in humanhepatocytes. Toxicol. Sci. 111, 120–130.

Fang, J.-L., McGarrity, L.J., Beland, F.A., 2009. Interference of cell cycle progression byzidovudine and lamivudine in NIH 3T3 cells. Mutagenesis 24, 133–141.

Fridland, A., Connelly, M.C., Ashmun, R., 1990. Relationship of deoxynucleotide changes toinhibition of DNA synthesis induced by the antiretroviral agent 3′-azido-3′-deoxythymidine and release of its monophosphate by human lymphoid cells(CCRF-CEM). Mol. Pharmacol. 37, 665–670.

Gonzalez Cid, M., Larripa, I., 1994. Genotoxic activity of azidothymidine (AZT) in in vitrosystems. Mutat. Res. 321, 113–118.

Humer, J., Ferko, B., Waltenberger, A., Rapberger, R., Pehamberger, H., Muster, T., 2008.Azidothymidine inhibits melanoma cell growth in vitro and in vivo. Melanoma Res.18, 314–321.

IARC, 2000. Zidovudine (AZT). IARCMonographs on the evaluation of carcinogenic risks tohumans. Some Antiviral and Antineoplastic Drugs, and Other Pharmaceutical Agents,vol. 76. International Agency for Research on Cancer, Lyon, France, pp. 73–127.

Ji, H.J., Rha, S.Y., Jeung, H.C., Yang, S.H., An, S.W., Chung, H.C., 2005. Cyclic induction ofsenescencewith intermittent AZT treatment accelerates both apoptosis and telomereloss. Breast Cancer Res. Treat. 93, 227–236.

Mamber, S.W., Brookshire, K.W., Forenza, S., 1990. Induction of the SOS response inEscherichia coli by azidothymidine and dideoxynucleosides. Antimicrob. AgentsChemother. 34, 1237–1243.

McComsey, G.A., Paulsen, D.M., Lonergan, J.T., Hessenthaler, S.M., Hoppel, C.L.,Williams, V.C., Fisher, R.L., Cherry, C.L., White-Owen, C., Thompson, K.A., Ross, S.T.,Hernandez, J.E., Ross, L.L., 2005. Improvements in lipoatrophy, mitochondrialDNA levels and fat apoptosis after replacing stavudine with abacavir orzidovudine. AIDS 19, 15–23.

Meng, Q.X., Su, T., Olivero, O.A., Poirier, M.C., Shi, X.C., Ding, X.X., Walker, V.E., 2000.Relationships between DNA incorporation, mutant frequency, and loss of heterozy-gosity at the TK locus in human lymphoblastoid cells exposed to 3′-azido-3′-deoxythymidine. Toxicol. Sci. 54, 322–329.

Mittelstaedt, R.A., Von Tungeln, L.S., Shaddock, J.G., Dobrovolsky, V.N., Beland, F.A., Heflich,R.H., 2004. Analysis of mutations in the Tk gene of Tk(+/−) mice treated as neonateswith 3′-azido-3′-deoxythymidine (AZT). Mutat. Res.-Fundam. Mol. Mech. Mutagen.547, 63–69.

Mofenson, L.M., 2010. Prevention in neglected subpopulations: prevention of mother-to-child transmission of HIV infection. Clin. Infect. Dis. 50, S130–S148.

NTP, 1999. NTP technical report on the toxicology and carcinogenesis studies of AZT (CASNo. 30516-87-1) and AZT/α-interferon A/D B6C3F1 mice (gavage studies). NatlToxicol. Program Tech. Rep. Ser. 469, 1–361.

NTP, 2006. NTP technical report on the toxicology and carcinogenesis studies oftransplacental AZT (CAS No. 30516-87-1) in Swiss (CD-1®) mice (in utero studies).Natl Toxicol. Program Tech. Rep. Ser. 522, 1–186.

Olivero, O.A., Beland, F.A., Fullerton, N.F., Poirier, M.C., 1994. Vaginal epithelial DNAdamage and expression of preneoplastic markers in mice during chronic dosingwith tumorigenic levels of 3′-azido-2′,3′-dideoxythymidine. Cancer Res. 54,6235–6242.

Olivero, O.A., Anderson, L.M., Diwan, B.A., Haines, D.C., Harbaugh, S.W., Moskal, T.J., Jones,A.B., Rice, J.M., Riggs, C.W., Logsdon, D., Yuspa, S.H., Poirier, M.C., 1997. Transplacentaleffects of 3′-azido-2′,3′-dideoxythymidine (AZT): tumorigenicity in mice andgenotoxicity in mice and monkeys. J. Natl Cancer Inst. 89, 1602–1608.

Purevjav, E., Nelson, D.P., Varela, J.J., Jimenez, S., Kearney, D.L., Sanchez, X.V., DeFreitas, G.,Carabello, B., Taylor, M.D., Vatta, M., Shearer, W.T., Towbin, J.A., Bowles, N.E., 2007.Myocardial Fas ligand expression increases susceptibility to AZT-induced cardiomy-opathy. Cardiovasc. Toxicol. 7, 255–263.

Riedl, T., Hanaoka, F., Egly, J.M., 2003. The comings and goings of nucleotide excision repairfactors on damaged DNA. EMBO J. 22, 5293–5303.

Roskrow,M.,Wickramasinghe, S.N., 1990. Acute effects of 3′-azido-3′-deoxythymidine onthe cell-cycle of HL60 cells. Clin. Lab. Haematol. 12, 177–184.

Sancar, A., Lindsey-Boltz, L.A., Unsal-Kacmaz, K., Linn, S., 2004. Molecular mechanisms ofmammalian DNA repair and the DNA damage checkpoints. Annu. Rev. Biochem. 73,39–85.

Skalski, V., Liu, S.H., Cheng, Y.C., 1995. Removal of anti-human-immunodeficiency-virus2′, 3′-dideoxynucleoside monophosphates from DNA by a novel human cytosolic3′–N5′-exonuclease. Biochem. Pharmacol. 50, 815–821.

Slamenova, D., Horvathova, E., Bartkova, M., 2006. Nature of DNA lesions induced inhuman hepatoma cells, human colonic cells and human embryonic lung fibroblastsby the antiretroviral drug 3′-azido-3′-deoxythymidine. Mutat. Res.-Fundam. Mol.Mech. Mutagen. 593, 97–107.

Sommadossi, J.P., Carlisle, R., Zhou, Z., 1989. Cellular pharmacology of 3′-azido-3′-deoxythymidine with evidence of incorporation into DNA of human bone marrowcells. Mol. Pharmacol. 36, 9–14.

Sugasawa, K., Ng, J.M.Y., Masutani, C., Iwai, S., van der Spek, P.J., Eker, A.P.M., Hanaoka, F.,Bootsma, D., Hoeijmakers, J.H.J., 1998. Xeroderma pigmentosum group C proteincomplex is the initiator of global genome nucleotide excision repair. Mol. Cell 2,223–232.

Sugasawa, K., Okamoto, T., Shimizu, Y., Masutani, C., Iwai, S., Hanaoka, F., 2001. Amultistep damage recognition mechanism for global genomic nucleotide excisionrepair. Genes Dev. 15, 507–521.

Sugasawa, K., Shimizu, Y., Iwai, S., Hanaoka, F., 2002. A molecular mechanism for DNAdamage recognition by the xeroderma pigmentosum group C protein complex.DNA Repair 1, 95–107.

Sussman, H.E., Olivero, O.A., Meng, Q., Pietras, S.M., Poirier, M.C., O'Neill, J.P., Finette, B.A.,Bauer, M.J., Walker, V.E., 1999. Genotoxicity of 3′-azido-3′-deoxythymidine in thehuman lymphoblastoid cell line, TK6: relationships between DNA incorporation,mutant frequency, and spectrum of deletion mutations in HPRT. Mutat. Res. 429,249–259.

Tapias, A., Auriol, J., Forget, D., Enzlin, J.H., Scharer, O.D., Coin, F., Coulombe, B., Egly, J.M.,2004. Ordered conformational changes in damaged DNA induced by nucleotideexcision repair factors. J. Biol. Chem. 279, 19074–19083.

Tosi, P., Calabresi, P., Goulette, F.A., Renaud, C.A., Darnowski, J.W., 1992. Azidothy-midine-induced cytotoxicity and incorporation into DNA in the human colontumor cell line HCT-8 is enhanced by methotrexate in vitro and in vivo. CancerRes. 52, 4069–4073.

Tsai, C.-A., Chen, J.J., 2009. Multivariate analysis of variance test for gene set analysis.Bioinformatics 25, 897–903.

Vazquez-Padua, M.A., Starnes, M.C., Cheng, Y.C., 1990. Incorporation of 3′-azido-3′-deoxythymidine into cellular DNA and its removal in a human leukemic cell line.Cancer Commun. 2, 55–62.

Volker, M., Mone, M.J., Karmakar, P., van Hoffen, A., Schul, W., Vermeulen, W.,Hoeijmakers, J.H.J., van Driel, R., van Zeeland, A.A., Mullenders, L.H.F., 2001.Sequential assembly of the nucleotide excision repair factors in vivo. Mol. Cell 8,213–224.

Von Tungeln, L.S., Hamilton, L.P., Dobrovolsky, V.N., Bishop, M.E., Shaddock, J.G., Heflich, R.H., Beland, F.A., 2002. Frequency of Tk andHprt lymphocytemutants and bonemarrow

162 Q. Wu et al. / Toxicology and Applied Pharmacology 251 (2011) 155–162

micronuclei in B6C3F(1)/Tk(+/−) mice treated neonatally with zidovudine andlamivudine. Carcinogenesis 23, 1427–1432.

Von Tungeln, L.S., Dobrovolsky, V.N., Bishop,M.E., Shaddock, J.G., Heflich, R.H., Beland, F.A.,2004. Frequency of Tk andHprt lymphocytemutants and bonemarrowmicronuclei inmice treated neonatally with zidovudine and didanosine. Mutagenesis 19, 307–311.

Von Tungeln, L.S.,Williams, L.D., Doerge, D.R., Shaddock, J.G., McGarrity, L.J., Morris, S.M.,Mittelstaedt, R.A., Heflich, R.H., Beland, F.A., 2007. Transplacental drug transfer andfrequency of Tk and Hprt lymphocyte mutants and peripheral blood micronuclei inmice treated transplacentally with zidovudine and lamivudine. Environ. Mol.Mutagen. 48, 258–269.

Walker, D.M., Malarkey, D.E., Seilkop, S.K., Ruecker, F.A., Funk, K.A., Wolfe, M.J., Treanor, C.P.,Foley, J.F., Hahn, F.F., Hardisty, J.F.,Walker, V.E., 2007. Transplacental carcinogenicity of 3′-

azido-3′-deoxythymidine in B6C3F1 mice and F344 rats. Environ. Mol. Mutagen. 48,283–298.

Wood, R.D., 1999. DNA damage recognition during nucleotide excision repair inmammalian cells. Biochimie 81, 39–44.

Wu, Y.W., Xiao, Q., Jiang, Y.Y., Fu, H., Ju, Y., Zhao, Y.F., 2004. Synthesis, in vitro anticancerevaluation, and interferencewithcell cycleprogressionofN-phosphoaminoacid estersofzidovudine and stavudine. Nucleosides Nucleotides Nucleic Acids 23, 1797–1811.

Wutzler, P., Thust, R., 2001. Genetic risks of antiviral nucleoside analogues—a survey.Antivir. Res. 49, 55–74.

Zhou, F.X., Liao, Z.K., Dai, J., Xiong, J., Xie, C.H., Luo, Z.G., Liu, S.Q., Zhou, Y.F., 2007.Radiosensitization effect of zidovudine on human malignant glioma cells. Biochem.Biophys. Res. Commun. 354, 351–356.