Mutagenesis vol. 26 no. 5 pp. 619–628, 2011 doi:10.1093/mutage/ger023Advance Access Publication 3 June 2011

Aristolochic acid-induced carcinogenesis examined by ACB-PCR quantification ofH-Ras and K-Ras mutant fraction

Yiying Wang*, Fanxue Meng, Volker M. Arlt1, Nan Mei,Tao Chen and Barbara L. Parsons

Division of Genetic and Molecular Toxicology, National Center for ToxicologicalResearch, US Food and Drug Administration, HFT-120, 3900 NCTR Road,Jefferson, AR 72079, USA and 1Section of Molecular Carcinogenesis, Institute ofCancer Research, Sutton, Surrey SM2 5NG, UK

*To whom correspondence should be addressed. Tel: þ870 543 7839;Fax: þ870 543 7393; Email: yiying.wang@fda.hhs.gov

Received on February 2, 2011; revised on April 8, 2011;accepted on April 13, 2011

Aristolochic acid (AA) is a strong cytotoxic nephrotoxinand carcinogen associated with the development ofurothelial cancer in humans. AA induces forestomach,kidney and urothelial tract tumours in rats and mice.This study was conducted to characterise AA’s carcino-genic mechanism of action and compare allele-specificcompetitive blocker-polymerase chain reaction (ACB-PCR)-based early detection of carcinogenic effect usingtwo different tumour-relevant endpoints. H-Ras codon 61CAA/CTA mutation was analysed because it is found inrodent forestomach tumours and A:T/T:A transversionis the predominant mutational specificity induced by AA.K-Ras codon 12 GGT/GAT mutation was analysedbecause it is a common spontaneous mutation presentin various rodent tissues and may be a useful genericbiomarker for carcinogenic effect. DNA samples from BigBlue rats treated with 0, 0.1, 1.0 or 10.0 mg AA/kg bodyweight (bw) by gavage, 5 days/week for 12 weeks wereused in ACB-PCR in order to examine the induction ofthe two specific mutations. A significant dose-dependentinduction of H-Ras mutant fraction (MF) was observed inliver and kidney. Statistically significant correlationswere observed between AA-induced DNA adduct levelsor cII mutant frequencies (previously measured in thesame rats) and H-Ras MF measurements. No correlationbetween AA dose and K-Ras MF was found in liver orkidney, although there was a significant induction ofK-Ras mutation in kidneys exposed to 0.1 mg/kg bw AArelative to controls. Thus, the data establish a straightfor-ward dose-related increase in H-Ras MF due to fixation ofAA-induced DNA adducts, whereas the common sponta-neous K-Ras mutation showed a non-monotonic dose–response, consistent with loss of non-targeted mutation atcytotoxic doses.

Introduction

Aristolochic acid (AA), an extract of Aristolochia species, isa mixture of structurally related nitrophenanthrene carboxylic

acids, mainly aristolochic acid I (AAI) and aristolochic acid II(AAII) (1). In alternative medicine, Aristolochia plant extractshave been used as anti-inflammatory agents (2). However, AA isa strong genotoxic nephrotoxin and carcinogen in humans. AAhas been associated with the development of a progressive renalfibrosis [aristolochic acid nephropathy (AAN)] and urothelialcancer in AAN patients (3–6). Also, AA exposure is linked toBalkan endemic nephropathy (BEN), a similar type of kidneyfibrosis with malignant transformation of the urothelium (7,8).

The carcinogenic effects of AA also have been observed inrodents. AA induces tumours in the forestomach, kidney andurothelial tract of rats (9–13). In mice, AA treatment results inforestomach carcinoma, adenocarcinoma in the glandularstomach, kidney adenomas, lung carcinomas and haemangio-mas of the uterus (10–12). Based on evidence from humansand rodents, the International Agency for Research on Cancerhas classified herbal remedies containing AA and AA itself asGroup I human carcinogens (14,15). In 2001, the U.S. Foodand Drug Administration issued an advisory warning healthcare professionals that consumption of products with AAresulted in several life-threatening adverse events and pub-lished a list of botanical products that contain AA (16).

AA-induced cytotoxicity has been observed in severalspecies. Acute tubular damage occurred after a single in-travenous injection of 1 mg/kg body weight (bw) in rabbits andafter a dose of ,1 mg/kg bw/day given for 3 days in humans(17,18). In rats and mice, acute renal failure followedadministration of single doses of 20 or 30 mg/kg bw or higher,respectively, indicating that kidney is the target organ fortoxicity in both species. In Wistar rats, renal lesions developedin a dose-dependent manner within 3 days of receiving a singledose of 10, 50 or 100 mg/kg bw by gastric tube (19). Incontrast, renal dysfunction and hypocellular interstitial sclero-sis developed at 4 and 6 months in Wistar rats given anintraperitoneal (i.p.) AA dose of 5 mg/kg bw/day for 4 months(20). Renal failure with interstitial fibrosis was observed insalt-depleted Wistar rats receiving daily subcutaneous injec-tions of 10 mg/kg bw AA for 35 days, whereas rats receiving 1mg/kg bw AA exihibited urothelial dysplasia and slight tubularatrophy (12). This indicates that 1 and 10 mg/kg bw AA arecytotoxic doses for Wistar rats. New Zealand White rabbitsgiven 0.10 mg/kg bw/day AA i.p. developed impaired renalfunction at 16 months and extensive interstitial fibrosis at 17months (21).

Both in vitro and in vivo metabolism studies have identified theprinciple metabolites of AAI and AAII as aristolactam I and II,respectively (22,23). During bioactivation, AAI and AAIIundergo reduction of a nitro group to form reactive cyclicnitrenium ions with delocalised charge, which can interact with theexocyclic amino groups of deoxyadenosine and deoxyguanosine,resulting in the preferential formation of purine adducts (1). Studiesfrom rodents and AAN patients demonstrated that AA formscovalent DNA adducts in vitro and in vivo (24–27). The major

Published by Oxford University Press on behalf of the UK Environmental Mutagen Society. 2011 619

at FD

A Library on S

eptember 1, 2011

mutage.oxfordjournals.org

Dow

nloaded from

AA–DNA adducts have been identified as 7-(deoxyadenosin-N6-yl)aristolactam I (dA-AAI), 7-(deoxyguanosin-N2-yl)aristolactam I(dG-AAI), 7-(deoxyadenosin-N6-yl)aristolactam II (dA-AAII) and 7-(deoxyguanosin-N2-yl)aristolactamII (dG-AAII) (24,26,27). The predominant DNA adduct producedin vivo, dA-AAI, is also the most persistent adduct in target tissues(forestomach, bladder and kidney), as well as in non-target tissues(liver and lung) (28). Also, dA–AAI adducts were found in thekidney, ureter, bladder and other tissues of AAN patients at timeslong after AA exposure (5,24,25,29). This adduct has been shownto produce predominately A:T/T:A transversions (30,31).Sequence analysis and DNA-binding studies confirmed that AAsbind primarily to adenine residues in neutral reporter genes (LacZand cII) and genes involved in the carcinogenic process (TP53 andH-Ras), producing A:T/T:A transversion mutations in vivo(8,27,32–38).

Activation of proto-oncogenes and inactivation of tumoursuppressor genes are considered to be critical molecular eventsin chemical carcinogenesis. Mutations in oncogenes andtumour suppressor genes have been implicated in the mech-anism of AA-induced tumour development (7,8,32,33,37). AA-initiated carcinogenesis in rodents is associated with activationof H-Ras by a specific A:T/T:A transversion mutation at thesecond position of codon 61 (CAA). H-Ras codon 61CAA/CTA mutations have been found in all forestomachand ear duct tumours of rats treated with AAI, as well as inforestomach and lung tumours of treated mice (32,33).Moreover, A:T/T:A transversions have been identified inTP53 codon 139 of urothelial tumours of AAN and BENpatients, whose DNA also contains AA–DNA adducts (8,37).In vitro studies identified AA-induced A:T/T:A transversionmutations in the human TP53 gene of an embryonic cell linederived from the Hupki (human TP53 knock-in) mouse strain(39,40). Thus, the mutational profile observed in rodent neutralreporter gene assays and in human tumours is consistent withthe AA–DNA adducts profile observed in vitro and in vivo inrodents and humans (41).

Animal studies have been conducted to elucidate the site-specificity of AA-induced carcinogenesis. AA is activated inkidney, liver, forestomach and other tissues; however, it mainlyinduces tumours in kidney and forestomach (11). Studiesreporting kidney and forestomach tumour incidence after AAexposure did not observe an increase in liver tumours, unlessthe rats also received a partial hepatectomy (42). Rats treatedorally with 0.1, 1.0 or 10 mg AA/kg bw/day for 3 monthsexhibited high incidences of tumours in the forestomach,kidney and urinary tract (72, 28 and 17%, respectively) (11).When the same treatment regimen was used to treat male BigBlue transgenic rats (5 days/week for 12 weeks), up to 4-foldhigher levels of total AA–DNA adducts were found in kidneythan in liver. Also, an �2-fold higher cII gene mutationinduction was observed in kidney as compared to liver.Sequence analysis of the cII gene mutants revealed thatA:T/T:A transversion was the predominant type of mutationin the AA-treated rats, whereas GC/AT transition was thepredominant mutation in the control rats (27,34).

Allele-specific competitive blocker-polymerase chain reaction(ACB-PCR) is a sensitive approach for directly quantifyingspecific rare base substitution mutations within a population ofDNA molecules (43). ACB-PCR has been used to characterise theinduction of oncogene point mutations in response to carcinogentreatment. For example, ACB-PCR was used to quantify K-Rascodon 12 GGT/GAT and GGT/TGT mutations in A/J mouse

lung following acute exposure to different doses of benzo[a]py-rene (B[a]P) (44). Also, ACB-PCR was used to describe thedose–response relationship between simulated solar light exposureand induction of a ultraviolet (UV)-specific TP53 codon 270CGT/TGT mutation in mouse skin tissues and to relate thosemeasurements to the induction of simulated solar light-inducedmouse skin tumours (45). Thus, the ACB-PCR approach isa quantitative and sensitive tool to detect specific tumour-associated mutations as early reporters of tumour response (43).

The current study employed the ACB-PCR approach toanalyse H-Ras codon 61 CAA/CTA and K-Ras codon 12GGT/GAT mutations in liver and kidney DNA of male BigBlue rats treated with 0, 0.1, 1.0 or 10.0 mg AA/kg bw. Thespecific goals of this study were (i) to characterise the AAdose–response using H-Ras and K-Ras mutation as endpointsin rat liver and kidney, as a means to elucidate themechanism(s) of AA-induced carcinogenesis and (ii) toinvestigate whether spontaneous K-Ras mutation can be usedas a generic reporter of AA-induced carcinogenesis.

Materials and methods

Animals

The liver and kidney tissues used in this study were those collected as part ofstudies conducted by Chen and Mei (27,34). Briefly, 6-week-old male Big Bluetransgenic rats were purchased from Taconic Laboratories (Germantown, NY,USA). All animal procedures followed the recommendations of the NationalCenter for Toxicological Research (NCTR) Institutional Animal Care and UseCommittee for handling, maintenance, treatment and sacrifice. AA (CAS 313-67-7) was purchased from Sigma–Aldrich (St Louis, MO, USA) and the content ofthe test agent was 96% (40% AAI and 56% AAII). The structures of AAI andAAII can be found in (2). Six rats per group were treated with AA (as its sodiumsalt) at concentrations of 0.1, 1.0 and 10.0 mg/kg bw by gavage, five times/weekfor 12 weeks and were sacrificed 1 day after the last treatment. Control rats weregavaged with 0.9% sodium chloride using the same schedule as for the AA-treated rats. The livers and kidneys were removed, immediately frozen in liquidnitrogen and stored at �80�C.

DNA isolation

Rat liver and kidney tissues were homogenised in an extraction buffer consistingof 1 mg/ml proteinase K (Sigma–Aldrich), 100 mM NaCl, 25 mM EDTA and 1%sodium dodecyl sulphate. The homogenate was incubated �16 h at 37�C,extracted with an equal volume of phenol/chloroform/isoamyl alcohol (25:24:1)and ethanol precipitated. DNA samples were resuspended in 200 ll of a solutioncontaining 10 mg/ml RNAse A (Sigma–Aldrich), 600 units/ml RNAse T1(Sigma–Aldrich), 100 mM sodium acetate and 50 mM Tris–HCl, pH 8. For liversamples, this buffer also included 1 mg/ml a-amylase (Fluka Chemical,Ronkonkoma, NY, USA). Samples were incubated �16 h at 37�C and thenextracted as described above. Each DNA sample was precipitated and thenresuspended in 20 ll of 0.5� TE buffer (5 mM Tris, 0.5 mM EDTA, pH 7.5).DNA samples were digested with EcoRI according to the manufacturer’sinstructions (New England Biolabs, Beverly, MA, USA). The digested DNAswere extracted as described above, ethanol precipitated, resuspended in 20 ll of0.5� TE buffer and concentrations were measured by UV absorbance at 260 nm.

Generating first-round PCR products

EcoRI-digested liver and kidney genomic DNAs were used as the template forfirst-round PCR amplification of a 319 bp H-Ras gene segment encompassingexon 2 and part of intron 2, using primers RHu (5#-CCGGAAACAGGTAGT-CATTGA-3#) and RHd (5#-GAGGGGGATGGGGTGTA-3#). Each PCRreaction contained 10 mM KCl, 10 mM (NH4)2SO4, 20 mM Tris–HCl (pH8.75), 2 mM MgSO4, 0.1% Triton X-100, 0.1 mg/ml bovine serum albumin,0.2 mM dNTPs, 0.2 lM RD1 (5#-TTAAGCGTCGATGGAGGAGTT-3#), 0.2lM RD2 (5#-GTCCTGCACCAGTAATATGC-3#) and 4 units of ClonedPfuUltra� Hotstart DNA Polymerase (Stratagene, La Jolla, CA, USA). Thecycling conditions included heating for 2 min at 94�C, followed by 35 cycles of1 min at 94�C, 1 min at 52�C and 1 min at 72�C.

EcoRI-digested liver and kidney genomic DNAs were used for first-roundPCR amplification of a 304 bp K-Ras gene segment encompassing 50 bp of 5#flanking sequence, exon 1 and part of intron 1. The PCR conditions wereidentical to those described above, except the primers used were RKR2

Y. Wang et al.

620

at FD

A Library on S

eptember 1, 2011

mutage.oxfordjournals.org

Dow

nloaded from

(5#-CTGTACCGATGGTTCCCT-3#) and RKR5 (5#-ATATTTTATTATTTT-TATTATAAGGCCTG-3#). The PCR products were agarose gel purified usinga Geneclean� Spin Kit (Q-Biogene, Vista, CA, USA). Single-use aliquots offirst-round PCR products were prepared and repeatedly quantified usinga Nanodrop ND-1000 spectrophotometer (Nanodrop Technologies, Wilming-ton, DE, USA), until three measurements that varied ,10% from the groupmean were obtained. All primers were purchased from Sigma-Genosys (TheWoodlands, TX, USA).

Wild-type and mutant reference DNAs

Initially, in vitro mutagenesis was used to generate the mutant and wild-type (WT)reference DNAs. To generate an H-Ras WT reference DNA, in vitro mutagenesiswas performed with RHu and an H-Ras WT mutagenesis primer (5#-CCGGAAACAGGTAGTCATTGATGGGGAGACGTGTTTACTGGACATCTTAGACACAGCAGGTCAAGAAGAGTATAGTGCCATGCGGGAC-3#), usingcontrol rat genomic DNA as template. Primer RHu and an H-Ras mutantmutagenesis primer (5#-CCGGAAACAGGTAGTCATTGATGGGGAGACGTGTTTACTGGACATCTTAGACACAGCAGGTCTAGAAGAGTATAGTGCCATGCGGGAC-3#) were used to generate an H-Ras mutant reference DNA. Later, theH-Ras reference DNAs were used in an ACB-PCR to determine which controlsample had the lowest codon 61 CTA MF. Genomic DNA from that sample wasthen used with primers RHu and RHd to synthesise a second WT reference DNA,which was used in the ACB-PCR analysis of all the unknown samples. The cyclingconditions, purification and quantification were the same as described above.

K-Ras WT and mutant reference DNAs were generated by in vitro mutagenesisusing control rat genomic DNA as template. Primer RD1 and a K-Ras WTmutagenesis primer (5#-ATATTTTATTATTTTTATTATAAGGCCTGCTGAAAATGACTGAGTATAAAC TTGTGGTAGTTGGAGCTGGTGGCGTAGG-3#)were used to generate the WT reference DNA. Primer RD1 and a K-Ras mutantmutagenesis primer (5#-ATATTTTATTATTTTTATTATAAGGCCTGCTGAAAATGACTGAGTATAAACTTGTGGTAGTTGGAGCTGATGGCGTAGG-3#)were used to generate the mutant reference DNA.

Allele-specific competitive blocker-polymerase chain reaction

ACB-PCR quantification is based on the concurrent analysis of first-round PCRproducts generated from unknown samples and standards with defined ratios ofmutant:WT DNA (i.e. MF standards). Purified mutant and WT DNA sampleswere mixed to generate standards with MFs of 10�1, 10�2, 10�3, 10�4, 10�5 and0. These standards and a no-DNA control were analysed in parallel with equalnumbers of copies of unknown first-round PCR products. For H-Ras codon 61CAA/CTA ACB-PCR, 4 � 108 copies were analysed in each 50 ll reactioncontaining: 40 lM dNTPs, 1� Stoffel buffer, 1.5 mM MgCl2, 0.1 mg/ml gelatin,1.0 mg/ml Triton X-100, 500 nM primer UP-H (5#-GGAAACAGGTAGT-CATTGA-3#), 500 nM MSP-H (5#-fluorescein CATGGCACTA-TACTCTTCCA-3#) and 500 nM BP-H (5#-CATGGCACTATACTCTTCCdT-3#). Thermocycler conditions were 90 sec at 95�C, followed by 38 cycles of 30sec at 94�C, 45 sec at 50�C and 1 min at 72�C. The H-Ras codon 61 CTA ACB-PCR product is 78 bp in length.

For K-Ras codon 12 GGT/GAT ACB-PCR, 2 � 108 copies were analysedin each 50 ll reaction containing 80 lM dNTPs, 1� Stoffel buffer, 1.5 mMMgCl2, 0.1 mg/ml gelatin, 1.0 mg/ml Triton X-100, 400 nM primer P4 (5#-GATTTACCTCT ATTGTTGGA-3#), 400 nM MSP-A (5#-fluorescein-CTTGTGGTAGTTGGAGCTTA-3#) and 366 nM BP-A (5#-CTTGTGGTAGTTGGAGCTTdG-3#). Thermocycler conditions were 90 secat 95�C, followed by 36 cycles of 30 sec at 94�C, 45 sec at 42�C and 1 min at72�C. The K-Ras codon 12 GAT ACB-PCR product is 105 bp in length.

Gel electrophoresis, image analysis and data collectionEqual volumes of ACB-PCR products were analysed on non-denaturing 8%polyacrylamide gels. The fluorescent bands were visualised using a PharosFXMolecular Imager with an external blue laser (Bio-Rad Life Science, Hercules,CA, USA). Pixel intensities of the bands were quantified using Quantity One�software and a locally averaged background correction (Bio-Rad, Hercules,CA, USA). After fluorescein-based quantification, gels were stained with VistraGreen (GE Healthcare Life Sciences, Piscataway, NJ, USA) to image the co-electrophoresed 25-bp DNA length marker (Invitrogen) and confirm the size ofthe ACB-PCR product.

Data analysisThe pixel intensities determined for the MF standards were plotted against theirMFs on log–log plots. A trend line (exponential function) was fit to the data andthe formula of the exponential function was used to calculate the MF in eachunknown sample based on its pixel intensity. The arithmetic average of threeindependent MF measurements was calculated. The average MF in each DNAsample was log-transformed and the average log-transformed MF for the six rats

in each treatment group was calculated. This value, converted back to scientificnotation, is the geometric mean MF for each treatment group. Statisticaldifferences in the average log-transformed MFs among treatment groups wereanalysed using Fisher’s exact test, comparing number of samples with MFs .and ,10�5 (Graphpad Prism 5, Graphpad Software, Inc.). Linear regressionanalysis was performed using Graphpad Prism 5 software.

Results

The ACB-PCR approach is based on parallel analysis ofstandards with defined MFs and unknowns, each containingthe same total number of molecules

Genomic DNAs from AA-treated (0.1, 1.0 and 10.0 mg/kg bwby gavage, five times/week for 12 weeks) and control Big Bluerat livers and kidneys were isolated. Pure WT and mutantreference DNAs for H-Ras codon 61 CAA/CTA or K-Rascodon 12 GGT/GAT were generated by PCR and then werecombined to construct MF standards ranging from 10�1 to 10�5.Each MF standard was analysed in duplicate, along with no-DNA control (pure WT standard). Using the standards, theACB-PCR reaction conditions were optimised and then used tomeasure the H-Ras codon 61 CTA and K-Ras codon 12 GATMF in 48 Big Blue rat liver and kidney samples (six rats in eachgroup and four groups of 0, 0.1, 1.0 and 10.0 mg/kg bw AA),including three replicate ACB-PCR measurements per sample.

ACB-PCR preferentially amplifies the mutant allele usinga primer (mutant-specific primer, MSP) that has moremismatches to the WT allele than to the mutant allele. Theprimer design used in the mutant-specific amplification of the H-Ras codon 61 CAA/CTA mutation is shown in Figure 1A andthat for the K-Ras codon 12 GGT/GAT mutation is shown inFigure 1B. Because the MSP used in ACB-PCR is labelled witha 5#-fluorescein, ACB-PCR products are fluorescent and can bevisualised following polyacrylamide gel electrophoresis.

AA-induced H-Ras codon 61 CAA/CTA mutation

Six gels were used to quantify all the study samples for H-Rascodon 61 CAA/CTA MFs, with 12 MF standards per gel (i.e.replicate 10�1, 10�2, 10�3, 10�4, 10�5 and 0 standard). Theaverage r2 value for the set of H-Ras codon 61 CTA standardcurves was 0.9764 (range from 0.9704 to 0.9885). Results offour of the six replicate H-Ras codon 61 CTA ACB-PCRmeasurements are shown in Figure 2. Results from liver andkidney of control rats and of rats exposed to 0.1 mg/kg bw AA,along with MF standards, are presented in Figure 2A. Resultsfrom liver and kidney of rats exposed to 1.0 and 10.0 mg/kg bwAA, along with MF standards, are presented in Figure 2B. Arepresentative standard curve is presented in Figure 2C.

The H-Ras codon 61 CAA/CTA MF measurements incontrol and treated livers and kidneys are summarised in Table 1and Figure 3. Although some of the H-Ras codon 61 CTA ACB-PCR MFs measurements were below the lowest MF standard(10�5), the calculated ACB-PCR MFs were used to estimate thegeometric mean MF, the median MF and mutation prevalence foreach treatment group (Table 1). The spontaneous level of H-Rascodon 61 CTA mutation was low in Big Blue control rat liverand kidney tissues. The control liver and kidney samples hadgeometric mean MFs of 1.63 � 10�6 and 1.01 � 10�6,respectively. No control sample from rat liver or kidney hada spontaneous H-Ras codon 61 CTA MF .10�5.

In liver, H-Ras codon 61 CTA MFs were .10�5 in zero of sixrats in the 0.01 mg/kg bw AA dose group, in two of six rats in the1.0 mg/kg bw AA dose group and in six of six rats in the 10.0

AA-induced carcinogenesis examined by ACB-PCR

621

at FD

A Library on S

eptember 1, 2011

mutage.oxfordjournals.org

Dow

nloaded from

mg/kg bw AA dose group (Table 1, mutation prevalence). TheH-Ras codon 61 CTA geometric mean MFs were 1.30 � 10�6,1.14 � 10�5 and 5.81 � 10�4 for the 0.01, 1.0 and 10.0 mg/kgbw AA treatment groups, respectively (Table 1). A significant

trend of increasing H-Ras MF with increasing dose was observed(slope of linear regression line in log–log plot was significantlynon-zero, P 5 0.0001, Figure 4A). Because some of the H-Rascodon 61 CTA MF measurements were below the lowest MFstandard, the distribution of samples . and ,10�5 was testedusing Fisher’s exact test. Statistically significant differences werefound between the 10 mg/kg bw dose group and each of the othertreatment groups (for control and 0.1 mg/kg bw, P 5 0.0180; for1.0 mg/kg bw, P 5 0.0303, using one-tailed tests).

In kidney, the H-Ras codon 61 CTA MF was .10�5 in zeroof six rats in the 0.01 mg/kg bw AA dose group, in two of sixrats in the 1.0 mg/kg bw AA dose group and in five of six rats inthe 10.0 mg/kg bw AA dose group (see Table 1). The H-Rascodon 61 CTA geometric mean MFs for the AA treatmentgroups were 5.05 � 10�6, 1.72 � 10�5 and 1.32 � 10�4 for the0.01, 1.0 and 10.0 mg/kg bw AA treatment groups, respectively(Table 1). A significant trend of increasing H-Ras MF withincreasing dose was observed (slope of linear regression line inlog–log plot was significantly non-zero, P5 0.0325, Figure 4B).Treatment groups were analysed for the distribution of samples. and ,10�5 relative to control samples. Statistically significantinduction of H-Ras mutation was observed in kidneys of ratsexposed to 10.0 mg/kg bw AA and each of the other treatmentgroups (for control and 0.1 mg/kg bw, P 5 0.0180; for 1.0 mg/kg bw, P 5 0.0303, using one-tailed tests).

Fig. 1. Diagram of the ACB-PCR priming strategy. The approximate positionsof the upstream primer (UP), mutant-specific primer (MSP) and the blockerprimer (BP) are depicted, along with the predicted preferential annealing of theMSP and the BP to the mutant and WT templates, respectively, for the H-Rascodon 61 CAA/CTA ACB-PCR (A) and the K-Ras codon 12 GGT/GATACB-PCR (B). Extendable and non-extendable primers are drawn ending with‘,’ or ‘‘’, respectively.

Fig. 2. ACB-PCR measurement of H-Ras codon 61 CAA/CTA MF. PCR products generated from 48 Big Blue rat liver and kidney DNA samples were analysedin triplicate by ACB-PCR, along with MF standards, a no mutant control and a no-DNA control. The ACB-PCR results of two of the three replicate analyses(Experiments 1 and 2) are shown, for samples 1–24 (A) and samples 25–48 (B). A representative example of a standard curve relating pixel intensity of ACB-PCRproduct to H-Ras MF is shown in (C).

Y. Wang et al.

622

at FD

A Library on S

eptember 1, 2011

mutage.oxfordjournals.org

Dow

nloaded from

K-Ras codon 12 GGT/GAT mutation

Six gels were used to quantify K-Ras codon 12 GGT/GATMF. The average r2 value for the set of K-Ras codon 12 GATstandard curves was 0.9651 (range from 0.9600 to 0.9755).Results of four of the six replicate K-Ras codon 12 GAT ACB-PCR measurements are shown in Figure 5. Samples from liverand kidney of control rats and rats exposed to 0.1 mg/kg bw AA,along with MF standards, are presented in Figure 5A. Samplesfrom liver and kidney of rats exposed to 1.0 and 10.0 mg/kg bwAA, along with MF standards, are presented in Figure 5B. Arepresentative standard curve is presented in Figure 5C.

The K-Ras codon 12 GGT/GAT MF measurements incontrol and AA-treated livers and kidneys are summarised inTable 1 and Figure 6. None of the control kidney samples hada K-Ras codon 12 GAT MF .10�5, whereas one control liver didhave a K-Ras codon 12 GAT MF .10�5. In liver, K-Ras codon12 GAT MFs were .10�5 in three of six rats in the 0.01 mg/kgbw AA dose group, in zero of six rats in the 1.0 mg/kg bw AAdose group and one of six rats in the 10.0 mg/kg bw AA dosegroup (Table 1). The K-Ras codon 12 GAT geometric mean MFswere 1.36 � 10�5, 8.58 10�7 and 3.76 x 10�6 for the 0.01, 1.0and 10.0 mg/kg bw AA treatment groups, respectively (Table 1).No significant trend of increasing GAT MF with increasing dosewas observed. The distribution of samples . and ,10�5 for eachtreatment group was compared to controls using Fisher’s exacttest. No statistically significant differences between the controlgroup and any AA treatment group were observed.

In kidney, K-Ras codon 12 GAT MFs were .10�5 in four ofsix rats in the 0.01 mg/kg bw AA dose group, in zero of six rats in

the 1.0 mg/kg bw AA dose group and in one of six rats in the 10.0mg/kg bw AA dose group. The K-Ras codon 12 GAT geometricmean MFs were 2.62 � 10�5, 3.04 � 10�6 and 6.02 � 10�6 forthe 0.01, 1.0 and 10.0 mg/kg bw AA treatment groups,respectively (Table 1). Analysis of samples . and ,10�5 usingFisher’s exact test indicated that a significant induction of K-Rasmutation was observed in kidneys of rats exposed to 0.1 mg/kg bwAA, as compared to control rats (P5 0.0303, in a one-sided test).

Correlation between DNA adducts, cII gene mutantfrequencies and Ras MF

Levels of DNA adducts and cII gene mutation were measuredpreviously for each Big Blue rat used in the current study. Thisenabled direct comparisons of the levels of DNA adducts and cIIgene mutation with the levels of H-Ras or K-Ras mutation

Table I. Summary of H-Ras and K-Ras MF measurements in Big Blue rat liver and kidney DNA samples

Organ AA treatment(mg/kg bw)

H-Ras codon 61 CTA mutation K-Ras codon 12 GAT mutation

Geometric mean MF Median MF Mutation prevalence (%)a Geometric mean MF Median MF Mutation prevalence (%)a

Liver 0 1.63 � 10�6 2.28 � 10�6 0 5.19 � 10�6 5.32 � 10�6 170.1 1.30 � 10�6 2.57 � 10�6 0 1.36 � 10�5 1.09 � 10�5 501.0 1.14 � 10�5 1.50 � 10�5 17 8.58 � 10�7 6.70 � 10�7 010.0 5.81 � 10�4 6.40 � 10�4 100 3.07 � 10�6 1.43 � 10�6 17

Kidney 0 1.01 � 10�6 1.24 � 10�6 0 2.86 � 10�7 4.40 � 10�7 00.1 5.05 � 10�6 5.99 � 10�6 0 2.62 � 10�5 4.30 � 10�5 671.0 1.72 � 10�5 1.49 � 10�5 34 3.04 � 10�6 1.56 � 10�6 010.0 1.32 � 10�4 5.12 � 10�4 83 6.02 � 10�6 2.97 � 10�6 17

aMutation prevalence is defined as the percentage of samples with a Ras MF .10�5.

Fig. 3. H-Ras codon 61 CAA/CTA MF. The average log10-transformed MFsfor the six rats in each treatment group (the geometric mean MF) were plottedrelative to dose. Error bars represent the standard deviation.

Fig. 4. Correlation between H-Ras codon 61 CAA/CTA MF and AA dose.(A) The base 10 log of H-Ras codon 61 CTA MF in liver DNA is plottedrelative to the base 10 log of dose. Error bars indicate the standard error of thegeometric mean MF. Dashed lines indicate the 95% confidence interval. (B)The base 10 log of H-Ras codon 61 CTA MF in the kidney DNA is plottedrelative to the base 10 log of dose. Error bars indicate the standard error of thegeometric mean MF. Dashed lines indicate the 95% confidence interval.

AA-induced carcinogenesis examined by ACB-PCR

623

at FD

A Library on S

eptember 1, 2011

mutage.oxfordjournals.org

Dow

nloaded from

induced in each rat. The total burden of DNA adducts induced ina particular rat by a given dose of AA was plotted against the H-Ras codon 61 CTA MF or K-Ras codon 12 GAT MF induced inthe liver and kidney tissues of the same rat. Significantcorrelations between the total burden of DNA adducts and H-Ras MF measurements were observed for both liver (linear

regression, P 5 0.0001) and kidney (linear regression, P 50.0325). No significant correlations between K-Ras codon 12GAT MF and DNA adduct burden were observed (Figure 7).cII gene mutant frequency induced in a particular rat was

plotted relative to the H-Ras codon 61 CTA MF or K-Rascodon 12 GAT MF induced in the liver and kidney of the samerat (Figure 8). Figure 8 shows significant correlations betweenthe cII mutant frequency and H-Ras MF were observed in bothliver and kidney (linear regression, P5 0.0001). No significantcorrelations between K-Ras codon 12 GAT MF and cII mutantfrequency were observed.

Discussion

In this study, ACB-PCR was employed to examine the ability ofAA to induce two different oncogene mutations (H-Ras codon 61CTA and K-Ras codon 12 GAT mutations) in order to determinewhether the carcinogenic effect of AA can be measured at anearly time point and to investigate whether spontaneous mutationscan be used as generic reporters of carcinogenesis.

For H-Ras codon 61 CAA/CTA mutation, a low sponta-neous background level was observed in the Big Blue rat liverand kidney and a significant dose-dependent induction of H-Ras MF was observed in AA-treated liver (linear regression, P

Fig. 5. ACB-PCR measurement of K-Ras codon 12 GGT/GAT MF. PCR products generated from 48 Big Blue rat liver and kidney DNA samples were analysedin triplicate by ACB-PCR, along with MF standards, a no mutant control and a no-DNA control. The ACB-PCR results of two of the three replicate analyses(Experimets 1 and 2) are shown, for samples 1–24 (A) and samples 25–48 (B). A representative example of a standard curve relating pixel intensity of ACB-PCRproduct to K-Ras MF is shown in (C).

Fig. 6. K-Ras codon 12 GGT/GAT MF. The average log10-transformed MFsfor the six rats in each treatment group (the geometric mean MF) were plottedrelative to dose. Error bars represent the standard deviation.

Y. Wang et al.

624

at FD

A Library on S

eptember 1, 2011

mutage.oxfordjournals.org

Dow

nloaded from

, 0.001) and kidney (linear regression, P, 0.05). This findingis consistent with previous reports that H-Ras codon 61CAA/CTA mutation is frequently induced by AA in rodentforestomach tumours and A:T/T:A transversion is the

predominant mutational specificity induced by AA. Becausethis induction was measured only 12 weeks after exposure, thisadds to the body of evidence that ACB-PCR can detectpotentially carcinogenic effects at times considerably earlier

Fig. 7. Correlation between DNA adducts and Ras MF. AA–DNA adducts were determined by 32P-postlabelling as reported (27). The total burden of DNA adductsinduced in each rat was plotted against the base 10 log of H-Ras codon 61 CTA MF induced in the liver (A) and kidney (B) of the same rat. The total burden of DNAadducts induced in each rat was plotted against the base 10 log of K-Ras codon 12 GAT MF induced in the liver (C) and kidney (D) of the same rat.

Fig. 8. Correlation between cII gene mutant frequencies and Ras MF. The cII mutant frequency induced in each rat was plotted against the base 10 log of H-Rascodon 61 CTA MF induced in the liver (A) and kidney (B) of the same rat. The cII gene mutant frequency induced in each rat was plotted against the K-Ras codon12 GAT MF induced in the liver (C) and kidney (D) of the same rat.

AA-induced carcinogenesis examined by ACB-PCR

625

at FD

A Library on S

eptember 1, 2011

mutage.oxfordjournals.org

Dow

nloaded from

than that needed for tumours to develop (43). While it is notclear that every AA-induced H-Ras codon 61 CTA mutationwill lead to the development of a tumour, the fact that thismutation is prevalent in AA-induced tumours argues that theearly induction of the mutation is likely to be carcinogenic.

ACB-PCR measurements in control- and vehicle-treatedanimals published to date indicate Ras oncomutations arepresent in various tissues of control rodents at high spontaneousmutant fractions (MFs) relative to background measurements inneutral reporter gene (43,44,46–49). Approximately 60% (88/146) of analysed tissues carry oncogene or tumour suppressorgene mutations at a level .10�5 and some have MFs as high as10�3 or 10�4 (43). In most cases, the K-Ras codon 12GGT/GAT MF in control animal tissues is .10�5, rangingfrom 10�6 to 10�3 in different rodent tissues. Furthermore, allhuman colonic mucosa samples analysed by ACB-PCR hadmeasurable levels of the K-RAS codon 12 GAT mutation, witha geometric mean MF of 1.03 � 10�4. These findings areconsistent with the idea that oncomutations confer a selectivegrowth or survival advantage, allowing them to accumulate tohigher levels than neutral reporter gene mutations. Therefore,pre-existing spontaneous oncomutations could potentially besensitive generic reporters of carcinogenic effect.

For a number of mutagenic carcinogens, induction ofspontaneous mutation (distinct from the primary mutationalspecificity of the chemical) has been noted. For example, in A/Jmouse lung, the K-Ras codon 12 GAT MF increased ina B[a]P-dose-dependent manner, even though B[a]P primarilyinduces G/T mutation (44). In Big Blue/F344 rat liver, the K-Ras codon 12 GAT MF significantly increased after N-hydroxy-2-acetylaminofluorene (N-OH-AAF) treatment, eventhough N-OH-AAF primarily induces G/T mutation (49–51).Observations like these were the basis for investigatingwhether K-Ras codon 12 GAT mutation, a frequent spontane-ous mutation, could be used as a generic reporter ofcarcinogenic effect. In the current study, K-Ras codon 12GAT mutation did not increase in a dose-dependent manner.An increase was observed at low AA dose (0.1 mg/kg bw),which was statistically significant in kidney, even thoughinduction of spontaneous mutations was not observed in the 1or 10 mg/kg bw AA treatment groups.

The absence of an AA dose-dependent increase in K-Ras MFmay be due to AA’s cytotoxicity and/or ability to induceapoptosis. AA is known to form DNA adducts after metabolicactivation and the resulting DNA adducts can block transcriptionand DNA replication (52,53). These effects may, in turn, inducecell cycle arrest or trigger apoptosis (52). In vitro studies confirmthat porcine and human kidney cell lines (LLC-PK1 cells andHKC cells, respectively) treated with 20 or 40 mg/l AAI havesignificantly higher percentages of apoptotic cells than controls(54,55). Increase in intracellular calcium ion concentration andcaspase 3 and 7 activities are involved in the mechanism of AAI-induced TP53-dependent apoptosis (54,56,57).

Research on the cytotoxicity of AA consistently indicatesthat 1 or 10 mg/kg bw AA are cytotoxic doses in rats (12).Although one study found that 0.1 mg/kg bw/day for 16months could induce chronic toxicity in rabbits, there is noevidence that 0.1 mg/kg bw/day for 3 months is cytotoxic inBig Blue rat liver or kidney (21). Thus, at non-toxic doses, AAmay induce the amplification of pre-existing K-Ras mutation aswell as cause de novo induction of AA-specific mutation. Attoxic doses, however, AA causes significant necrosis and/orapoptosis selecting against the pre-existing K-Ras mutation,

whereas DNA adduct and de novo AA-specific mutationcontinue to accumulate despite the cytotoxicity. In this regard,it is important to note that H-Ras mutation induction wascorrelated with the DNA adduct levels induced at different AAdoses, but no such correlation was observed for K-Ras.

The target organ specificity of AA carcinogenesis is ofinterest because AA induces DNA adducts and mutations inliver and kidney but induces tumours only in kidney(10,11,58). Previously, Chen et al. determined AA-inducedDNA adduct level and cII mutant frequency in liver and kidneyfor the same rats used in the current study (27,34). Theyreported AA dose-dependent increases in total DNA adductsand cII mutant frequency in both liver and kidney, althoughhigher levels of both DNA adducts and cII mutants weredetected in kidney. In the current study, H-Ras codon 61 CTAmutation was induced to similar levels in liver and kidney. Thissuggests factors in addition to DNA damage and mutation arenecessary for tumour induction and that those factors accountfor the tissue specificity. Changes in gene expression, forexample, may also drive tumour development. Microarrayanalysis was used to characterise differential responses to AAin kidney and liver (58–62). Significant alterations of genesassociated with defense response, apoptosis and immuneresponse were found in kidney but not in liver, which mayexplain the tissue-specific carcinogenicity of AA.

The data obtained on AA-induced H-Ras mutation, viewedalongside other available data, provide definitive evidence thatAA is a mutagenic and cytotoxic carcinogen. Bioactivated AAinteracts with the exocyclic amino groups of deoxyadenosineand deoxyguanosine and leads to the preferential formation ofadenine adducts (1,16,26). A:T/T:A transversion mutationsare the primary mutation specificity induced by AA and AA-induced tumours frequently carry H-Ras CAA/CTA muta-tions, which were detected by ACB-PCR shortly after exposure.Together, this dataset provides strong evidence that AA isa mutagenic carcinogen. Applying the same approaches to othercarcinogens, therefore, may be a productive approach forunderstanding mode of action long before tumours may develop.

In conclusion, ACB-PCR measurement of H-Ras codon 61CTA mutation provides an early measure of AA-inducedcarcinogenic effect. The K-Ras codon 12 GAT MF resultsindicate that frequent spontaneous mutations are not usefulgeneric reporters of carcinogenic effects at cytotoxic doses.Nevertheless, the approach did detect a significant induction ofK-Ras GAT at low dose (0.1 mg/kg bw), 12 weeks afterexposure in the target tissue, providing additional support forthe hypothesis that spontaneous mutation can be a genericreporter of carcinogenic effect when not measured undercytotoxic conditions. Finally, the present study illustrates howthe ACB-PCR assay, when combined with other endpoints,has the potential to elucidate a chemical’s cancer mode ofaction.

Funding

This work was supported by an appointment (Y.W.) to thePostgraduate Research Program at the NCTR administered bythe Oak Ridge Institute for Science and Education through aninteragency agreement between the U.S. Department of Energyand the U.S. Food and Drug Administration (FDA). Workat the Institute of Cancer Research was supported byAssociation for International Cancer Research (AICR) andCancer Research UK.

Y. Wang et al.

626

at FD

A Library on S

eptember 1, 2011

mutage.oxfordjournals.org

Dow

nloaded from

Acknowledgements

The authors thank Drs Page McKinzie, Robert Heflich and Martha Moore fortheir critical review of this manuscript. The contents of this manuscript do notnecessarily reflect the views or polices of the U.S. FDA nor does the mention oftrade names or commercial products constitute endorsement or recommendationfor use.

Conflict of interest statement: None declared.

References

1. Schmeiser, H. H., Stiborova, M. and Arlt, V. M. (2009) Chemical and

molecular basis of the carcinogenicity of Aristolochia plants. Curr. Opin.Drug Discov. Devel., 12, 141–148.

2. Arlt, V. M., Stiborova, M. and Schmeiser, H. H. (2002) Aristolochic acid as

a probable human cancer hazard in herbal remedies: a review. Mutagenesis,17, 265–277.

3. Vanherweghem, J. L., Depierreux, M., Tielemans, C. et al. (1993) Rapidly

progressive interstitial renal fibrosis in young women: association with

slimming regimen including Chinese herbs. Lancet, 341, 387–391.4. Cosyns, J. P., Jadoul, M., Squifflet, J. P., Wese, F. X. and van Ypersele de

Strihou, C. (1999) Urothelial lesions in Chinese-herb nephropathy. Am. J.Kidney Dis., 33, 1011–1017.

5. Nortier, J. L., Martinez, M. C., Schmeiser, H. H. et al. (2000) Urothelial

carcinoma associated with the use of a Chinese herb (Aristolochia fangchi).

N. Engl. J. Med., 342, 1686–1692.6. Debelle, F. D., Vanherweghem, J. L. and Nortier, J. L. (2008) Aristolochic

acid nephropathy: a worldwide problem. Kidney Int., 74, 158–169.7. Arlt, V. M., Stiborova, M., vom Brocke, J., Simoes, M. L., Lord, G. M.,

Nortier, J. L., Hollstein, M., Phillips, D. H. and Schmeiser, H. H. (2007)

Aristolochic acid mutagenesis: molecular clues to the aetiology of Balkan

endemic nephropathy-associated urothelial cancer. Carcinogenesis, 28,

2253–2261.8. Grollman, A. P., Shibutani, S., Moriya, M. et al. (2007) Aristolochic acid

and the etiology of endemic (Balkan) nephropathy. Proc. Natl Acad. Sci.USA, 104, 12129–12134.

9. Mengs, U. (1983) On the histopathogenesis of rat forestomach carcinoma

caused by aristolochic acid. Arch. Toxicol., 52, 209–220.10. Mengs, U. (1988) Tumour induction in mice following exposure to

aristolochic acid. Arch. Toxicol., 61, 504–505.11. Mengs, U., Lang, W. and Poch, J. A. (1982) The carcinogenic action of

aristolochic acid in rats. Arch. Toxicol., 51, 107–119.12. Debelle, F. D., Nortier, J. L., De Prez, E. G., Garbar, C. H., Vienne, A. R.,

Salmon, I. J., Deschodt-Lanckman, M. M. and Vanherweghem, J. L. (2002)

Aristolochic acids induce chronic renal failure with interstitial fibrosis in

salt-depleted rats. J. Am. Soc. Nephrol., 13, 431–436.13. Cosyns, J. P., Goebbels, R. M., Liberton, V., Schmeiser, H. H.,

Bieler, C. A. and Bernard, A. M. (1998) Chinese herbs nephropathy-

associated slimming regimen induces tumours in the forestomach but no

interstitial nephropathy in rats. Arch. Toxicol., 72, 738–743.14. IARC. (2002) IARC Monographs on the Evaluation of Carcinogenic Risks

to Humans. Some Traditional Herbal Medicines, Some Mycotoxins,Naphthalene and Styrene: World Health Organisation. IARC Press, Lyon,

France.15. Grosse, Y., Baan, R., Straif, K. et al. (2009) WHO International Agency for

Research on cancer Monograph working group A review of human

carcinogens-Part A: pharmaceuticals. Lancet Oncol., 10, 13–14.16. Schwetz, B. A. (2001) Safety of aristolochic acid. JAMA, 285, 2705–2726.17. Jackson, L., Kofman, S., Weiss, A. and Brodovsky, H. (1964) Aristolochic

acid (Nsc-50413): phase I Clinical study. Cancer Chemother. Rep., 42, 35–37.18. Mehes, J., Decsi, L., Varga, F. and Kovacs, S. (1958) [Selective chemical

disconnection of the uriniferous tubes of the first order in rabbits]. NaunynSchmiedebergs Arch. Exp. Pathol. Pharmakol., 234, 548–565.

19. Mengs, U. and Stotzem, C. D. (1993) Renal toxicity of aristolochic acid in

rats as an example of nephrotoxicity testing in routine toxicology. Arch.Toxicol., 67, 307–311.

20. Zheng, F., Zhang, X. and Huang, Q. (2001) [Establishment of model of

aristolochic acid-induced chronic renal interstitial fibrosis in rats].

Zhonghua Yi Xue Za Zhi, 81, 1095–1100.21. Cosyns, J. P., Dehoux, J. P., Guiot, Y., Goebbels, R. M., Robert, A.,

Bernard, A. M. and van Ypersele de Strihou, C. (2001) Chronic aristolochic

acid toxicity in rabbits: a model of Chinese herbs nephropathy? Kidney Int.,59, 2164–2173.

22. Krumbiegel, G., Hallensleben, J., Mennicke, W. H., Rittmann, N. andRoth, H. J. (1987) Studies on the metabolism of aristolochic acids I and II.Xenobiotica, 17, 981–991.

23. Schmeiser, H. H., Pool, B. L. and Wiessler, M. (1986) Identification andmutagenicity of metabolites of aristolochic acid formed by rat liver.Carcinogenesis, 7, 59–63.

24. Schmeiser, H. H., Bieler, C. A., Wiessler, M., van Ypersele de Strihou, C.and Cosyns, J. P. (1996) Detection of DNA adducts formed by aristolochicacid in renal tissue from patients with Chinese herbs nephropathy. CancerRes., 56, 2025–2028.

25. Arlt, V. M., Alunni-Perret, V., Quatrehomme, G. et al. (2004) Aristolochicacid (AA)-DNA adduct as marker of AA exposure and risk factor for AAnephropathy-associated cancer. Int. J. Cancer, 111, 977–980.

26. Stiborova, M., Fernando, R. C., Schmeiser, H. H., Frei, E., Pfau, W. andWiessler, M. (1994) Characterization of DNA adducts formed byaristolochic acids in the target organ (forestomach) of rats by 32P-postlabelling analysis using different chromatographic procedures. Carci-nogenesis, 15, 1187–1192.

27. Mei, N., Arlt, V. M., Phillips, D. H., Heflich, R. H. and Chen, T. (2006)DNA adduct formation and mutation induction by aristolochic acid in ratkidney and liver. Mutat. Res., 602, 83–91.

28. Fernando, R. C., Schmeiser, H. H., Schert, H. and Wiessler, M. (1993)Formation and persistence of specific purine DNA adducts by 12P-postlabelling in target and nontarget organs of rats treated with aristolochicacid I. In Phillips, D. H., Castegnaro, M. and Bartsch, H. (eds),Postlabeling Methods for Detection of DNA Adducts, Vol. 124. IARC,Lyon, France, pp. 167–171.

29. Lord, G. M., Cook, T., Arlt, V. M., Schmeiser, H. H., Williams, G. andPusey, C. D. (2001) Urothelial malignant disease and Chinese herbalnephropathy. Lancet, 358, 1515–1516.

30. Broschard, T. H., Wiessler, M. and Schmeiser, H. H. (1995) Effect of site-specifically located aristolochic acid DNA adducts on in vitro DNAsynthesis by human DNA polymerase alpha. Cancer Lett., 98, 47–56.

31. Broschard, T. H., Wiessler, M., von der Lieth, C. W. and Schmeiser, H. H.(1994) Translesional synthesis on DNA templates containing site-specifically placed deoxyadenosine and deoxyguanosine adducts formedby the plant carcinogen aristolochic acid. Carcinogenesis, 15, 2331–2340.

32. Schmeiser, H. H., Janssen, J. W. G., Lyons, J., Scherf, H. R., Pfau, W.,Buchmann, A., Bartram, C. R. and Wiessler, M. (1990) Aristolochic acidactivates ras genes in rat tumors at deoxyadenosine residues. Cancer Res.,50, 5464–5469.

33. Schmeiser, H. H., Scherf, H. R. and Wiessler, M. (1991) Activatingmutations at codon 61 of the c-Ha-ras gene in thin-tissue sections of tumorsinduced by aristolochic acid in rats and mice. Cancer Lett., 59, 139–143.

34. Chen, L., Mei, N., Yao, L. and Chen, T. (2006) Mutations induced bycarcinogenic doses of aristolochic acid in kidney of Big Blue transgenicrats. Toxicol Lett., 165, 250–256.

35. Kohara, A., Suzuki, T., Honma, M., Ohwada, T. and Hayashi, M. (2002)Mutagenicity of aristolochic acid in the lambda/lacZ transgenic mouse(Muta(TM)Mouse). Mutat. Res., 515, 63–72.

36. Arlt, V. M., Wiessler, M. and Schmeiser, H. H. (2000) Using polymerasearrest to detect DNA binding specificity of aristolochic acid in the mouseH-ras gene. Carcinogenesis, 21, 235–242.

37. Lord, G. M., Hollstein, M., Arlt, V. M., Roufosse, C., Pusey, C. D.,Cook, T. and Schmeiser, H. H. (2004) DNA adducts and p53 mutations ina patient with aristolochic acid-associated nephropathy. Am. J. Kidney Dis.,43, e11–e17.

38. Arlt, V. M., Schmeiser, H. H. and Pfeifer, G. P. (2001) Sequence-specificdetection of aristolochic acid-DNA adducts in the human p53 gene byterminal transferase-dependent PCR. Carcinogenesis, 22, 133–140.

39. Liu, Z., Hergenhahn, M., Schmeiser, H. H., Wogan, G. N., Hong, A. andHollstein, M. (2004) Human tumor p53 mutations are selected for in mouseembryonic fibroblasts harboring a humanized p53 gene. Proc. Natl Acad.Sci. USA, 101, 2963–2968.

40. Nedelko, T., Arlt, V. M., Phillips, D. H. and Hollstein, M. (2009) TP53mutation signature supports involvement of aristolochic acid in the aetiologyof endemic nephropathy-associated tumours. Int. J. Cancer, 124, 987–990.

41. Kucab, J. E., Phillips, D. H. and Arlt, V. M. (2010) Linking environmentalcarcinogen exposure to TP53 mutations in human tumours using the humanTP53 knock-in (Hupki) mouse model. FEBS J., 277, 2567–2583.

42. Rossiello, M. R., Laconi, E., Rao, P. M., Rajalakshmi, S. andSarma, D. S. R. (1993) Induction of hepatic nodules in the rat byaristolochic acid. Cancer Lett., 71, 83–87.

43. Parsons, B. L., Myers, M. B., Meng, F., Wang, Y. and McKinzie, P. B. (2010)Oncomutations as biomarkers of cancer risk. Environ. Mol. Mutagen, 51,836–850.

AA-induced carcinogenesis examined by ACB-PCR

627

at FD

A Library on S

eptember 1, 2011

mutage.oxfordjournals.org

Dow

nloaded from

44. Meng, F., Knapp, G. W., Green, T., Ross, J. A. and Parsons, B. L. (2010)K-Ras mutant fraction in A/J mouse lung increases as a function ofbenzo(a)pyrene dose. Environ. Mol. Mutagen., 51, 146–155.

45. Verkler, T. L., Delongchamp, R. R., Miller, B. J., Webb, P. J.,Howard, P. C. and Parsons, B. L. (2008) Simulated solar light-inducedp53 mutagenesis in SKH-1 mouse skin: a dose-response assessment. Mol.Carcinog., 47, 599–607.

46. Parsons, B. L., Beland, F. A., Von Tungeln, L. S., Delongchamp, R. R.,Fu, P. P. and Heflich, R. H. (2005) Levels of 4-aminobiphenyl-inducedsomatic H-ras mutation in mouse liver DNA correlate with potential forliver tumor development. Mol. Carcinog., 42, 193–201.

47. McKinzie, P. B. and Parsons, B. L. (2011) Accumulation of K-Ras codon12 mutations in the F344 rat distal colon following azoxymethaneexposure. Environ. Mol. Mutagen, in press.

48. Parsons, B. L., Marchant-Miros, K. E., Delongchamp, R. R., Verkler, T. L.,Patterson, T. A., McKinzie, P. B. and Kim, L. T. (2010) ACB-PCRquantification of K-ras codon 12 GAT and GTT mutant fraction in colontumor and non-tumor tissue. Cancer Invest., 28, 364–375.

49. McKinzie, P. B., Delongchamp, R. R., Chen, T. and Parsons, B. L. (2006)ACB-PCR measurement of K-ras codon 12 mutant fractions in livers of BigBlue rats treated with N-hydroxy-2-acetylaminofluorene. Mutagenesis, 21,391–397.

50. Chen, T., Mittelstaedt, R. A., Aidoo, A., Hamilton, L. P., Beland, F. A.,Casciano, D. A. and Heflich, R. H. (2001) Comparison of hprt and lacImutant frequency with DNA adduct formation in N-hydroxy-2-acetylami-nofluorene-treated Big Blue rats. Environ. Mol. Mutagen., 37, 195–202.

51. Chen, T., Mittelstaedt, R. A., Shelton, S. D., Dass, S. B.,Manjanatha, M. G., Casciano, D. A. and Heflich, R. H. (2001) Gene-and tissue-specificity of mutation in Big Blue rats treated with thehepatocarcinogen N-hydroxy-2-acetylaminofluorene. Environ. Mol. Muta-gen., 37, 203–214.

52. Roos, W. P. and Kaina, B. (2006) DNA damage-induced cell death byapoptosis. Trends Mol. Med., 12, 440–450.

53. Stiborova, M., Frei, E., Arlt, V. M. and Schmeiser, H. H. (2008) Metabolicactivation of carcinogenic aristolochic acid, a risk factor for Balkanendemic nephropathy. Mutat. Res., 658, 55–67.

54. Gao, R., Zheng, F., Liu, Y., Zheng, D., Li, X. and Bo, Y. (2000)Aristolochic acid I-induced apoptosis in LLC-PK1 cells and amelioration ofthe apoptotic damage by calcium antagonist. Chin. Med. J. (Engl.), 113,418–424.

55. Su, Z., Xu, S., Zheng, F. and Li, Y. (2002) [Aristolochic acid inducedtransdifferentiation and apoptosis in human tubular epithelial cells in vitro].Zhonghua Yu Fang Yi Xue Za Zhi, 36, 301–304.

56. Balachandran, P., Wei, F., Lin, R. C., Khan, I. A. and Pasco, D. S. (2005)Structure activity relationships of aristolochic acid analogues: toxicity incultured renal epithelial cells. Kidney Int., 67, 1797–1805.

57. Simoes, M. L., Hockley, S. L., Schwerdtle, T., da Costa, G. G.,Schmeiser, H. H., Phillips, D. H. and Arlt, V. M. (2008) Gene expressionprofiles modulated by the human carcinogen aristolochic acid I in humancancer cells and their dependence on TP53. Toxicol. Appl. Pharmacol.,232, 86–98.

58. Arlt, V. M., Zuo, J., Trenz, K., Roufosse, C. A., Lord, G. M., Nortier, J. L.,Schmeiser, H. H., Hollstein, M. and Phillips, D. H. (2010) Gene expressionchanges induced by the human carcinogen aristolochic acid I in renal andhepatic tissue of mice. Int. J. Cancer, 128, 21–32.

59. Chen, T., Guo, L., Zhang, L., Shi, L., Fang, H., Sun, Y., Fuscoe, J. C. andMei, N. (2006) Gene expression profiles distinguish the carcinogeniceffects of aristolochic acid in target (kidney) and non-target (liver) tissues inrats. BMC Bioinformatics, 7 (Suppl. 2), S20.

60. Guo, L., Lobenhofer, E. K., Wang, C. et al. (2006) Rat toxicogenomicstudy reveals analytical consistency across microarray platforms. Nat.Biotechnol., 24, 1162–1169.

61. Stemmer, K., Ellinger-Ziegelbauer, H., Ahr, H. J. and Dietrich, D. R.(2007) Carcinogen-specific gene expression profiles in short-term treatedEker and wild-type rats indicative of pathways involved in renaltumorigenesis. Cancer Res., 67, 4052–4068.

62. Stemmer, K., Ellinger-Ziegelbauer, H., Ahr, H. J. and Dietrich, D. R.(2009) Molecular characterization of preneoplastic lesions providesinsight on the development of renal tumors. Am. J. Pathol., 175,1686–1698.

Y. Wang et al.

628

at FD

A Library on S

eptember 1, 2011

mutage.oxfordjournals.org

Dow

nloaded from