Reactive Oxygen Species May Participate in the Mutagenicity and Mutational Spectrum of Cadmium in...

Reactive Oxygen Species May Participate in theMutagenicity and Mutational Spectrum of Cadmium in

Chinese Hamster Ovary-K1 Cells

Jia-Ling Yang,* Jui-I Chao, and Jin-Guo LinMolecular Carcinogenesis Laboratory, Department of Life Sciences, National Tsing Hua University,

Hsinchu 300, Taiwan, Republic of China

Received July 24, 1996X

The molecular nature of mutations induced by Cd was investigated in this study to elucidatethe role of Cd in the initiation of carcinogenesis. Exposing Chinese hamster ovary (CHO)-K1cells to cadmium acetate markedly decreased the colony-forming ability of cells and inducedmutation frequency in the hypoxanthine (guanine) phosphoribosyltransferase (hprt) gene. Themutation frequency induced by Cd at LD30-LD20 doses was approximately 20 times that ofuntreated cells. D-Mannitol, a scavenger of reactive oxygen species (ROS), significantly protectscells against Cd cytotoxicity and mutagenicity. Furthermore, non-cytotoxic doses of 3-amino-1,2,4-triazole, a catalase inhibitor, potentiates Cd cytotoxicity and mutagenicity. The cellularCd uptake ability was not altered by the combined treatment with either D-mannitol or 3-amino-1,2,4-triazole. The GSH level and the activities of GSH peroxidase, GSSG reductase, andcatalase in cells treated with Cd (4 µM, 4 h) decreased to 78%, 47%, 40%, and 22% of theuntreated cells, respectively. Those enzymatic activities recovered to normal levels 8 h afterremoving Cd. Polymerase chain reaction and DNA sequencing analysis of 54 independent Cdmutants revealed Cd-induced base substitutions, splice mutations, and large genomic deletions.All six types of base substitutions were observed; however, base transversions (22/27; 81%)occurred more frequently than transitions (5/27; 19%). The frequencies of mutations occurringat T‚A or G‚C base pairs were roughly equal. Results in this study strongly suggest that Cdmutagenicity in CHO-K1 cells is ROS-dependent. Moreover, the unique mutational spectruminduced by Cd implies that specific DNA adducts generated through the interaction of Cd-DNA and ROS may play a role in the mutational specificity.

Introduction

Cd is a ubiquitous environmental contaminant thatinduces cytotoxicity, chromosomal aberrations, DNAstrand breakage, and mutagenicity in mammalian cells(1-4). A significant body of evidence confirms Cdcarcinogenicity in animal models; e.g., Cd has been shownto induce tumors in testes, lungs, prostates, hemato-poietic systems, and at injection sites (5, 6). Recently,Cd has been evaluated as a potential human carcinogen(7).Cd genotoxicity and carcinogenicity are rather complex

because this metal interacts with both DNA and proteins.Cd interaction with nucleic acids favor bases over DNAphosphates (8-10). Moreover, it has been reported thatCd can inhibit DNA replication and decrease the fidelityof DNA polymerases (11). Cd can also inhibit DNA repairpathways such asO6-alkylguanine-DNA-alkyltransferaseand nucleotide-excision repair (12). Furthermore, sulf-hydryl groups in metal-enzyme complexes have a higheraffinity for Cd than for the essential metal Zn (13); Cdcan also substitute for Zn in the zinc-finger motif oftranscription factors (14). Metallothionein is the mostimportant metal-binding protein accounting for the re-sistance to Cd genotoxicity (15-17). Moreover, Cd altersCa homeostasis (18) and consequently interferes with theexpression of certain functional genes and signal trans-duction pathways (19, 20). Thus, Cd plays direct orindirect roles in genotoxicity.

Intracellular reactive oxygen species (ROS),1 e.g., H2O2,superoxide, singlet oxygen, and hydroxyl radicals, areproduced by cellular redox cycling systems that areimportant mediators of the cytotoxicity of many chemi-cals (21, 22). Oxygen radicals can induce genotoxicity,including DNA damage and gene mutations in culturedcells (23-27). DNA strand breakage and chromosomalaberrations induced by Cd can be suppressed by ROSscavengers, implying that ROS mediate Cd genotoxicity(28, 29). In this study, we explore whether ROS areinvolved in Cd-induced mutagenesis of mammalian cells.Additionally, the molecular nature of mutations inducedby Cd is investigated to provide more thorough informa-tion about Cd mutagenicity mechanism. Chinese ham-ster ovary-K1 (CHO) cells were treated with Cd incombination with either D-mannitol (a ROS scavenger)or 3-amino-1,2,4-triazole (3-AT; a catalase inhibitor) andassayed for cytotoxicity and mutagenicity. Also, theintracellular activities of catalase, GSH peroxidase, andGSSG reductase and the GSH levels were measured toelucidate oxidative stress induced by Cd. The mutationalspectrum in the hprt gene induced by Cd was investi-gated by directly sequencing cDNA and/or genomic DNAamplified by polymerase chain reaction (PCR). Further-more, the Cd-induced mutational spectrum was com-pared with that occurring spontaneously (30) and with

* Author to whom correspondence should be addressed. Fax: 886-3-5721746. E-mail: [email protected]

X Abstract published in Advance ACS Abstracts, November 1, 1996.

1 Abbreviations: ROS, reactive oxygen species; CHO cells, Chinesehamster ovary cells; 3-AT, 3-amino-1,2,4-triazole; HPRT, hypoxanthine(guanine) phosphoribosyltransferase; cDNA; complementary DNA;PCR, polymerase chain reaction; RT, reverse transcriptase HAT, 1 ×10-4 M hypoxanthine, 4 × 10-7 M aminopterin, and 1.6 × 10-6 Mthymidine; 6-TG, 6-thioguanine; PBS, phosphate-buffered saline.

1360 Chem. Res. Toxicol. 1996, 9, 1360-1367

S0893-228x(96)00122-1 CCC: $12.00 © 1996 American Chemical Society

spectra induced by two other metal carcinogens, Cr(VI)and Pb (31, 32).

Experimental Procedures

Cell Culture. CHO-K1 cells obtained from American TypeCulture Collection (Rockville, MD) were grown in a completemedium that consisted of F12/Dulbecco’s modified Eagle me-dium (1:1; Gibco, Life Technologies Co., Grand Island, NY)supplemented with sodium bicarbonate (2.2%, w/v), L-glutamine(0.03%, w/v), penicillin (100 units/mL), streptomycin (100 µg/mL), and fetal calf serum (10%). Cell cultures were maintainedat 37 °C in a humidified incubator containing 5% CO2 in air.Approximately 104-105 cells were grown in a complete mediumcontaining 1 × 10-4 M hypoxanthine, 4 × 10-7 M aminopterin,and 1.6 × 10-6 M thymidine (HAT medium) for 2-3 days toreduce the spontaneous HPRT-deficient cells. Cultured cellswere then placed in complete medium and cultured for 3-5days.Cytotoxicity and Mutagenicity Assay. The HAT pre-

treated cells were used for cytotoxicity and mutagenicity assayas described previously (31). Cells in exponential growth weretrypsinized, and 1 × 106 cells were plated onto a 100-mm Petridish 18 h before Cd treatment. Cadmium acetate (Merck,Darmstadt, Germany) was dissolved in MilliQ-purified water(Millipore Co., Bedford, MA). Cells were treated with Cd for24 h in a complete medium. In experiments to determine theeffect of ROS on Cd-induced genotoxicity, cells were incubatedwith either 50 mM D-mannitol (Merck) for 30 min or 80 mM3-AT (Sigma, St. Louis, MO) for 1 h before co-exposure to Cdfor 4 h. Next, the drug-containing medium was removed, andthe cells were washed twice with phosphate-buffered saline(PBS) before trypsinization. Cells were plated at a density of200-1000 cells/60 mm Petri dish in triplicate. The cells werecultured for 7 days and stained with Giemsa solution, and thenthe colony numbers were counted for cytotoxicity determination.The Cd-treated and untreated cells were maintained in

exponential growth for 7 days to allow for expression ofresistance to 6-thioguanine (6-TG). One million cells from eachgroup were plated onto 10 100-mm Petri dishes in a selectivemedium containing 11 µg/mL 6-TG and then incubated for 1week. Plating efficiency of cells at the time of selection wasalso assayed in a non-selective medium to correct for theobserved mutation frequency. The mutation frequency wascalculated to be the total number of 6-TG-resistant coloniesdivided by the total number of clonable cells at selection time(31). Individual 6-TG-resistant colonies derived from eachpopulation were isolated using trypsin and then expanded formolecular analysis.Determination of Cellular Cd Levels. Cells (1× 106) were

exposed to various Cd concentrations in complete media for 4h. Following treatment, the cells were washed four times withPBS and trypsinized, and the numbers of cells were determined.Next, cells were centrifuged, and the cell pellet was sonicatedin MilliQ-purified water. Total Cd concentration was analyzedusing a polarized Zeeman atomic absorption spectrophotometer(Spectr AA-30, Varian) equipped with an autosampler and agraphite furnace. The analytical conditions for Cd uptake wereset as follows: absorption wavelength of 228.8 nm, lamp currentof 4 mA, and an atomizing temperature of 1800 °C.Determination of Enzyme Activity and GSH Levels.

Cells were washed 3 times with PBS after Cd treatment andharvested using a rubber policeman. The cell suspension wascentrifuged at 1000 rpm for 5 min at 4 °C. The cell pellets weresonicated in 0.5 mL of 50 mM phosphate buffer (pH 7.0), andthe debris was removed by centrifuging at 10 000 rpm for 30min at 4 °C. The supernatant was then used for enzyme assay.Catalase activity was measured as described by Aebi (33).

Assay was performed at room temperature in a 1-mL mixturecontaining clear cell lysate, 50 mM phosphate buffer (pH 7.0),and 10 mM H2O2. The decomposition of H2O2 was followeddirectly by a decrease in absorbance at 240 nm. The enzymeactivity was expressed in nanomoles of H2O2 decrease min-1 (mgof protein)-1.

GSH peroxidase activity was determined as described byFlohe and Gunzler (34). Briefly, the 1 mL of reaction mixturecontained clear cell lysate, 50 mM phosphate buffer (pH 7.0),0.24 U of GSSG reductase, 1 mM GSH, 0.15 mM NADPH, and1.2 mM tert-butyl hydroperoxide. The decrease in absorptionat 340 nm was monitored for 200 s. A unit of the enzyme wasdefined as the amount of enzyme that catalyzed the reductionof 1 µmol of NADPH per min.GSSG reductase activity was assayed as described by Carl-

berg and Mannervik (35). The 1-mL reaction mixture containedclear cell lysate, 100 mM phosphate buffer (pH 7.0), 1 mMGSSG, and 0.1 mMNADPH. The decrease in absorption at 340nm was recorded for 200 s. Also, a unit of the enzyme wasdefined as the amount of enzyme that catalyzed the reductionof 1 µmol of NADPH per min.The procedure for GSH determination was as described by

Cohn and Lyle (36). The cell pellets were sonicated in MilliQ-purified water (ice-cold), and the debris was removed bycentrifuging at 10 000 rpm for 30 min at 4 °C. A finalconcentration of 6.25% metaphosphoric acid was added to thesupernatant. Following centrifugation, 40 µL of the clearsupernatant was diluted with 2 mL of MilliQ-purified water andthen mixed with 0.5 mL of 0.1 M phosphate buffer (pH 8.0) and0.1 mL of 0.1%O-phthalaldehyde. The mixture was maintainedat room temperature for 20 min. Fluorescence (excitationwavelength 350 nm and emission wavelength 420 nm) wasmeasured with a fluorescence spectrophotometer. Proteinconcentrations were determined using bovine serum albuminas a standard (37).Reverse Transcriptase (RT) PCR Amplification. First-

strand cDNA was synthesized directly from hprtmRNA accord-ing to the method described by Yang et al. (38). Briefly, 250cells in PBS were microcentrifuged for 5 min at 4 °C. Thesupernatant was removed, and the cell pellet was resuspendedin 5 µL of cDNA cocktail containing buffer, bovine serumalbumin, RNase inhibitor, Nonidet P-40, 4 deoxynucleotides,oligo(dT)12-18, and Moloney murine leukemia virus reversetranscriptase. The reaction was performed at 37 °C for 1 h toallow the lysis of cell membranes and the synthesis of first-strand cDNA from poly(A) mRNA. The synthesized first-strandcDNA served as a template for two 30-cycle PCR amplificationsusing two nested sets of primers (38). The 5′ to 3′ nucleotidesequences of the four primers for cDNA amplification were P1,-68CTCGGCGCCTCCTCTGCGGG-49; P2, 721GGTAATTTTAC-TGGGAACAT702; P3, -44CTCCTCACACCGCTCTTCGC-25; andP4, 693CTCCTCGTGTTTGCAGATTC674 designed after the ham-ster hprt cDNA sequence reported by Konecki et al. (39). AllRT-PCR amplifications were run in parallel with negativecontrols to which no template or cell lysate was added to thereaction mixture. Positive controls were also included in eachamplification experiment using normal cells (6-TG sensitive) asthe starting template to monitor the success of amplifications.Further purification was not required between the two 30-cyclePCR amplifications.DNA-PCR Amplification. The intron-exon boundary re-

gion was amplified directly from the hprt gene in 1 × 104 cells.Cells in PBS were microcentrifuged, and cell pellets wereresuspended in 10 µL of the lysis cocktail as described (40). Thereaction was carried out at 50 °C for 1 h and was followed byincubation at 94 °C for 10 min. The DNA sample was used asthe template in a 30-cycle PCR amplification (40). The nucleo-tide sequences of primers used for DNA-PCR amplification weredesigned according to Rossiter et al. (41) (EMBO data libraryX50376, X50379, and X50380) and their 5′ to 3′ sequences areshown as follows:

Direct DNA Sequencing. The 5′ to 3′ nucleotide sequencesof primers S1, 264ATTTCTATTCAGT252; S2, 169ATGGGAGGC-

E4F 6TTCTTACATCTAATAATTTTTTGTG30E4R 221CCAAGTGAGTGATTGAAAGCA201E7/8F 6GTTCTATTGTCTTTCCCATATG27E7/8R 455CAGTCTGGTCAAATGACGAG436E9F 5GTTTGGTAGGAACCAGACAATT26E9R 750AGAGTTCTATAAGAGACAGTCC729

ROS Participate in Cd-Induced Mutational Spectrum Chem. Res. Toxicol., Vol. 9, No. 8, 1996 1361

CATC181; S3, 424ACAATGCAAACTCC436; and S4, 521TATCCAA-CACTTC509 were designed for sequencing the entire codingregion of hprt cDNA. The 5′ to 3′ nucleotide sequences of fourprimers E4S, 200CAGTTACTAATTG188; E7S, 163AGCAATTT-TAGTC151; E8S, 423GAAAACAATTCTT411; and E9S, 210GAAT-GCTTGGTAA198 were designed according to EMBO data libraryX50376, X50379, and X50380 for sequencing genomic hprt DNAof splice mutants. The amplified cDNA or DNA was ethanol(70%, w/v) precipitated, dried and resuspended in distilled waterbefore DNA sequencing. Direct DNA sequencing was performedby dideoxy DNA sequencing of double-strand linear templatesusing 5′ end-labeled primers as described by Yang et al. (38).

Results

Cytotoxicity and Mutagenicity of Cd. Figure 1a,bshows the survival curves and the mutation frequenciesof CHO-K1 cells exposed to 0-1.25 µM cadmium acetatefor 24 h. The Cd concentration required to reduce thecolony-forming ability to 37% (D0) was estimated to be0.8 µM (Figure 1a). The Cd-treated and untreated cellswere cultured for 7 days and then assayed for theirresistance to 6-TG. The background mutation frequencywas 1.2 ( 0.5 per 106 clonable cells and increased inlinear proportion when cells were treated with 0.75-1µM Cd; at 1 µM, the induced mutation frequency was28.0 ( 1.8 per 106 clonable cells (Figure1b). However,cells exposed to 1.25 µM Cd for 24 h exhibited severegrowth retardation and low mutation frequencies. This

phenomenon is possibly attributable to high cytotoxicdosages of Cd inhibiting DNA replication, i.e., the processfor mutation fixation.Effect of ROS on Cytotoxicity and Mutagenicity

of Cd. To explore the role of ROS in Cd mutagenicity,cells were exposed to either 50 mM D-mannitol for 30 minor 80 mM 3-AT for 1 h before co-treatment with Cd for 4h. The cytotoxicity and mutagenicity curves induced byexposing cells to Cd for 4 h were similar to those inducedby 24-h treatments (Figure 2). D-Mannitol markedlyprotected CHO-K1 cells against cytotoxicity and muta-genicity induced by Cd (Figure 2). The mutation fre-quency induced by 4 µM Cd (4-h treatment) was 26.0 (2.6 per 106 clonable cells. This frequency decreasedmarkedly to 6.8 ( 1.6 per 106 clonable cells aftercombined treatment with 50 mM D-mannitol (Student’st-test, p < 0.001). In contrast, 3-AT potentiated thecytotoxicity and mutagenicity of Cd (Figure 2). Themutation frequency induced by 4 µMCd was significantlyenhanced to 36.8 ( 2.5 per 106 clonable cells by 80 mM3-AT (Student’s t-test, p < 0.001). The 3-AT concentra-tion used here had neither a cytotoxic nor mutageniceffect on CHO-K1 cells.

Figure 1. Cytotoxicity (a) and mutagenicity (b) induced bycadmium acetate in CHO-K1 cells. The cytotoxicity curve wasdetermined from nine experiments involving 24-h Cd treat-ments. Each mutation experiment was accompanied by acytotoxicity experiment. The mutation frequency was deter-mined from at least five independent experiments, and the totalclonable cells assayed were 4.8-9.4 × 106 cells per dose. Theplating efficiencies at the time of selection were 60-97%. Thebars denote population SE.

Figure 2. Effects of D-mannitol and 3-AT on the cytotoxicity(a) and mutagenicity (b) induced by cadmium acetate. CHO-K1cells were treated with 50 mM D-mannitol (open triangle), 80mM 3-AT (filled triangle), or H2O (open circle) before they wereexposed to Cd. The cytotoxicity and mutagenicity curves werederived from at least four experiments. D-Mannitol or 3-AT alonedid not significantly influence the colony-forming abilities ofcells. The total clonable cells assayed were 1.7-7.3 × 106 cellsper dose. The bars denote population SE. Statistical analysiswas performed by the Student’s t-test; *, p < 0.01, **, p < 0.001.

1362 Chem. Res. Toxicol., Vol. 9, No. 8, 1996 Yang et al.

Cd Uptake Ability. Attenuation of Cd uptake mayreduce Cd genotoxicity (42). Therefore, cells’ Cd uptakeability was determined by atomic absorption spectropho-tometry. Table 1 shows that the intracellular amountsof Cd were nearly equal in cells treated with Cd, Cd plusD-mannitol, and Cd plus 3-AT. The results have sug-gested that differential efficiency of Cd uptake by cellsis not involved in significant differences in cytotoxicityand mutagenicity among these treatments.Determination of Catalase Activity. The intra-

cellular catalase activity was assayed immediately and8 h after exposing cells to 3-AT, Cd, or Cd plus 3-AT.Exposing cells to 80 mM 3-AT for 5 h reduced theintracellular catalase activity to 10% of the untreatedlevel; catalase activity recovered to 25% of the untreatedlevel 8 h after removal of 3-AT (Figure 3). This resultconfirmed that 3-AT blocks catalase activity. Further-more, Cd treatment (4 µM, 4 h) also reduced intracellularcatalase activity to 22% of the untreated level. However,catalase activity recovered to the untreated level 8 h afterremoving Cd. Intracellular catalase activity after Cd plus3-AT treatment was 10% of the untreated level; thisactivity remained at a low level 8 h after removing bothagents (Figure 3).Determination of GSH Peroxidase, GSSG Reduc-

tase, and GSH Levels. GSH peroxidase and GSSG

reductase participate in the reduction of intracellularH2O2. Therefore, the intracellular levels of these en-zymes in Cd-treated cells were examined. Table 2 revealsthat exposing cells to 4 µM Cd for 4 h reduced GSHperoxidase and GSSG reductase activities to 47.5% and39.7% of the levels in untreated cells, respectively.Again, those enzymatic activities reached untreatedlevels 8 h after removal of Cd. In addition, the intra-cellular GSH level in Cd-treated cells was reduced to 78%of the level in the untreated control.Characterization of hprt Mutants. Independent

6-TG-resistant mutants derived from CHO-K1 cellstreated with Cd for 24 h were cloned for RT-PCRamplification of the entire coding region of hprt cDNA.The PCR amplifications were performed at least twicein all analyzed mutants. Those amplified cDNAs weresubsequently analyzed using agarose gel electrophoresisand DNA sequencing. Identical mutations observed inmutants derived from the same dish of Cd treatmentwere counted as single mutational events, thus avoidingoverestimation of mutational hot spots induced by Cd.Agarose gel electrophoresis analysis of RT-PCR prod-

ucts derived from a total of 54 independent mutantsshowed that 24 (44%) had transcripts equal to normalsize, 10 (19%) were slightly smaller than normal size,seven (13%) had large deletions, and 13 (24%) could notbe recovered. Mutant cDNAs amplified by RT-PCR werecharacterized using the dideoxy DNA sequencing proce-dure. Table 3 summarizes that sequence alterations inthe hprt cDNA of 29 independent Cd mutants had hprttranscripts equal to or slightly smaller than normal size.The sequence alterations in the entire hprt cDNA of thesemutants included single-base substitutions (14 mutants),two-base substitutions (three mutants), small duplica-tions (two mutants), and splice mutations (10 mutants)(Table 3). Mutations in the remaining mutants were notdetermined because they were lost during analysis. Thegenomic DNAs of putative splice mutants were amplifiedby DNA-PCR and further characterized by DNA sequenc-ing to reveal mutations in the intron-exon boundaryregions. All six analyzed splice mutants contained basesubstitutions in the 5′ or 3′ consensus splice donor oracceptor sites; five were single-base substitutions and onehad tandem-base substitutions (Table 3). The cDNA oftwo mutants having 7-bp duplications (ATTCCAG) in theexon 8 (cDNA position 547-553) were further confirmedby DNA-PCR and sequencing of their genomic DNA.Recurrent single-base substitutions were observed atcDNA positions 119 and 293 and at the splice site I3:-1.Recurrent mutations were also observed in mutantshaving two-base substitutions (Cd105 and Cd121) andduplications (Cd25 and Cd37). Small-sequence duplica-tions have been observed in several hprt mutants andoncogenic genes in human cancers (43-46). A slipped-

Table 1. Cadmium Uptake Ability in CHO-K1 Cells

treatmenta Cd (ng/106 cells)

control 0.51 ( 0.12 (3)bCd 45.19 ( 2.40 (3)Cd + D-mannitol 45.96 ( 1.61 (3)Cd + 3-AT 45.78 ( 1.14 (3)

a The concentrations of Cd, D-mannitol, and 3-AT were 4 µM,50 mM, and 80 mM, respectively. The cells were treated, and theintracellular Cd concentrations were determined as described inMaterials and Methods. b Mean ( SE. Numbers in parenthesesindicate number of experiments.

Figure 3. Effects of Cd and 3-AT on intracellular catalaseactivity in CHO-K1 cells. One million cells in exponential growthwere plated onto a 100-mm dish 18 h before drug treatment.The cells were treated with 80 mM 3-AT for 1 h at 37 °C. Next,cells were treated or not treated with 4 µM Cd for 4 h. Cellswere then washed with PBS, and one set of cells was culturedin complete medium for another 8 h. Catalase activity in cellsimmediately after treatment or 8 h after removal of Cd and/or3-AT was assayed as described in Materials and Methods. Theresults were obtained from the average of three experiments,and the bars denote population SE.

Table 2. Intracellular GSH Peroxidase and GSSGReductase Activities, and GSH Levels in CHO-K1 Cells

treatmenta

Cdrecoverytime (h)

GSHperoxidase(mU/mgof protein)

GSSGreductase(mU/mgof protein)

GSH(nmol/mgof protein)

- 0 12.70 ( 1.73 (5)b 154.07 ( 9.42 (5) 27.74 ( 3.14 (6)- 8 16.95 ( 1.64 (2) 137.50 ( 11.83 (2) ndc+ 0 7.44 ( 1.21 (5) 79.80 ( 12.30 (5) 21.61 ( 1.51 (6)+ 8 15.70 ( 3.95 (2) 133.08 ( 17.70 (2) nd

a Cells were treated with 4 µM Cd for 4 h (+) or not treated(-). Assay was performed immediately after treatment and 8 hafter removal of Cd. b Mean ( SE. Numbers in parenthesesindicate the number of experiments. c No data available.


mispairing model proposed by Kunkel can account for thegeneration of duplications/deletions (47).Seven Cd-induced mutants had large deletions in their

hprt transcripts. Sequencing analysis of five of thesemutants showed that they had deleted two (exons 2-3)or three (exons 2-4) exons. The hprt cDNA in 13 of the54 Cd-induced independent mutants could not be ampli-fied by RT-PCR. The inability to amplify hprt cDNAcould be due to large deletions in the hprt gene, tomutations located at the sequences for PCR primers and/or at the 5′ promoter region, or to very low mRNA copynumbers in mutants. This mutation type was catego-rized as large genomic DNA deletions in Table 4 tocalculate the frequencies of each type of mutation.Table 4 also summarizes the specific types of base

substitutions derived from Cd-treated cells. All six typesof base substitutions were observed. The frequency oftransversions (22/27; 81%) was significantly higher thanthat of transitions (5/27; 19%). The most frequentmutations observed were G‚C f C‚G (30%), T‚A f A‚T(26%), and T‚A f G‚C (19%) tranversions (Table 4). Thefrequency of mutations occurring at T‚A base pairs (15/27; 56%) was approximately the same as that occurringat G‚C base pairs (12/27; 44%).Several recent studies have demonstrated that strand-

and sequence-specific mutagenesis exists in mutants

induced by some chemical carcinogens, e.g., benzo[a]-pyrene (48) and N-methyl-N′-nitro-N-nitrosoguanidine(40, 49). Although, specific bases damaged by Cd areunknown, the frequency of T‚A base substitutions on thetwo strands exhibited strand specificity, i.e., 11 of the15 mutations (73%) had thymine located on the non-transcribed strand, which was significantly higher thanthe available thymines (44%) located on the non-transcribed strand (goodness-of-fits for ø2 test, p < 0.05).Conversely, as observed on the non-transcribed andtranscribed strands, the distribution of Cd-induced G‚Cbase substitutions (75%:25%) was not significantly dif-ferent from that of the available target sites on the twostrands (70%:30%). In addition, ∼50% of the Cd-inducedbase substitutions occurred at A/T-rich sequences.

Discussion

The present study has shown that Cd mutagenicity inCHO-K1 cells is markedly attenuated by removing in-tracellular ROS using D-mannitol; whereas inhibition ofcatalase activity in cells by 3-AT potentiates Cd muta-genicity. Moreover, the Cd uptake ability of cells isunaltered by those ROSmodulators. Metallothionein hasbeen reported to protect cells against Cd genotoxicity(15-17). However, our unpublished observation showed

Table 3. Types and Locations of Mutations Observed in the hprt Gene in Mutants Derived from CadmiumAcetate-Treated CHO-K1 Cells

mutant dose (µM) positiona exon type of mutation target sequenceb (5′f 3′) amino acid change

Single-Base Substitutions (14 Mutants)Cd47 1.0 40 2 G‚C f T‚A GAT GAA CCA Glu f stopCd64 1.0 119 2 G‚C f C‚G CAT GGA GTG Gly f AlaCd80 1.0 119 2 G‚C f C‚G CAT GGA GTG Gly f AlaCd33 1.0 125 2 T‚A f G‚C GTG ATT ATG Ile f SerCd83 0.75 145 3 C‚G f T‚A AGA CTT GCC Leu f PheCd15 1.0 220 3 T‚A f A‚T AAA TTC TTT Phe f IleCd5 0.75 293 3 A‚T f T‚A GTA GAT TTT Asp f ValCd32 1.0 293 3 A‚T f T‚A GTA GAT TTT Asp f ValCd36 1.0 293 3 A‚T f T‚A GTA GAT TTT Asp f ValCd81 1.0 296 3 T‚A f G‚C GAT TTT ATC Phe f CysCd96 1.25 488 7 T‚A f C‚G agC TTG CTG Leu f SerCd58 0.75 491 7 T‚A f G‚C TTG CTG GTG Leu f ArgCd112 0.75 527 7 C‚G f G‚C AGG CCA GAC Pro f ArgCd126 1.0 648 9 C‚G f G‚C AAA TAC AAA Tyr f stop

Two-Base Substitutions (3 Mutants)Cd77 1.0 62 2 T‚A f A‚T GAT TTA TTT Leu f stop

123 2 G‚C f A‚T GGA GTG ATT no changeCd105 0.75 185 3 T‚A f A‚T CAC ATT GTG Ile f Asn

188 3 T‚A f C‚G ATT GTG GCC Val f AlaCd121 1.25 185 3 T‚A f A‚T CAC ATT GTG Ile f Asn

188 3 T‚A f C‚G ATT GTG GCC Val f Ala

Duplications (2 Mutants)Cd25 0.75 547-553 8 a 7 bp duplication GAA ATTCCAG ACCd37 1.0 547-553 8 a 7 bp duplication GAA ATTCCAG AC

Splice Site Mutations (10 Mutants)

mutant dose mRNA alteration DNA alterationc target sequenced

Cd86 0.75 exon 2 missing UDCd88 0.75 exon 2 missing UDCd48 0.75 exon 4 missing I3:-1 g f c aactagAATGATCd100 1.0 exon 4 missing I3:-1 g f c aactagAATGATCd3 0.75 exon 4 missing I4:1 g f c GGAAAGgtaagtCd119 1.0 exon 5 missing UDCd118 1.0 exon 7 missing I6:-2 a f c taacagCTTGCTCd85 0.75 exon 8 missing UDCd44 1.0 5′ 21 bases of exon 8 missing I7:-1 g f t ttacagTTGTTG

E8:1 T f GCd55 0.75 5′ 17 bases of exon 9 missing I8:-1 g f c ttgcagCATATT

a Mutations in the coding region, A of AGU start codon is designated as position 1. b Sequence of non-transcribed strand. Underscoresindicate locations of base changes. c The positions in the intron-exon boundaries designated as I3:-1 represents the last base of intron 3and I4:1 represents the first base of intron 4. UD, undetermined. d Underscores indicate locations of base changes; lowercase lettersindicate intron bases.


that the endogenous metallothionein level in CHO-K1cells was low and that this level cannot be significantlyinduced by mutagenic Cd doses. Taken together, theabove results strongly suggest that ROS participate inCd mutagenicity in CHO-K1 cells. These observationsalso correspond with findings that DNA strand breakageand chromosomal aberrations induced by Cd are ROS-dependent (28, 29). Furthermore, we have demonstratedthat mutagenic Cd doses temporarily reduced intra-cellular catalase, GSH peroxidase, and GSSG reductaseactivities in CHO-K1 cells. Recently, Tabatabaie andFloyd reported that GSH peroxidase and GSSG reductaseare susceptible to oxidative damage that can result inthe loss of enzymatic activity (50). Cd may enhancecellular oxidative stress (51, 52) and subsequently leadto temporary loss of catalase, GSH peroxidase, and GSSGreductase activities. Alternatively, Cd may directlyinteract with those antioxidant enzymes or proteinsthrough the sulfhydryl groups of amino acids (13, 14),thereby inhibiting their activities. Additionally, theexpressions of gene coding for catalase, GSH peroxidase,and GSSG reductase may be functional because thoseenzymatic activities recovered to untreated levels afterCd removal.Results from previous studies on cellular antioxidant

defense systems against Cd treatment are somewhatcontradictory (51-54). For instance, GSSG reductase,catalase, and GSH were reduced, but GSH peroxidasewas increased in testicular Leydig cells after in vivotreatment of rats with Cd (51). GSH was reduced, butcatalase, GSH peroxidase, and GSSG reductase remainunaltered by a 2-h Cd treatment in Chinese hamster V79cells (52). Obviously, the differential sensitivities ofantioxidant enzymes to Cd is tissue- and cell-dependent,e.g., the levels of metallothionein or other Cd-bindingproteins may be different in those cells. The low metallo-thionein level in CHO-K1 cells may partly account forthe high sensitivity of these enzymes to Cd treatment.Analysis of the hprt cDNA amplified from mutants

induced by Cd revealed that the frequency of smallalterations (63%; including single-exon skips) was higherthan the frequency of unrecovered and large-deletion

cDNA (37%). The Cdmutational spectrumwas comparedwith that occurring spontaneously in the hprt gene ofCHO cells as reported by Xu et al. (30). Similar ratios ofsmall alterations, single-exons skips, and large genomicdeletions were observed in mutants induced by Cd andthose occurring spontaneously (Table 4). This findingsuggests that both point mutations and large deletionswere enhanced in Cd-treated cells. Cd induces higherfrequencies of G‚C f C‚G and T‚A f A‚T transversions;additionally, the positions of base substitutions in thehprt gene induced by Cd completely differ from thoseoccurring spontaneously (Table 3) (30). Base substitu-tions at position 208 (a six-Gs′ sequence) have beenreported to occur spontaneously and were significantlyenhanced by N-methyl-N′-nitro-N-nitrosoguanidine andbenzo[a]pyrene diol epoxide (30, 40, 48, 49). However,this mutational hot spot was not found in mutantsinduced by Cd. Taken together, these comparisonssuggest that Cd induces a unique mutational spectrum.Other specific features of Cd-induced base substitutionswere (1) ∼50% of them occurred at A/T-rich sequences,(2) strand bias mutagenicity was observed in T‚A basepairs, and (3) ∼80% of G‚C base substitutions occurredat G flanking A or T (5′AG, 5′TG, 5′GA, or 5′GT sites).We have previously demonstrated the hprtmutational

spectra of two other metal carcinogens, Cr(VI) and Pb,in CHO-K1 cells (31, 32). Table 4 also compares thekinds of mutations induced by these metals. Cd andCr(VI) induce single- and two-base substitutions, whereasPb causes single- but not two-base substitutions. More-over, base substitutions induced by Cd were distributedroughly equally at T‚A (56%) and G‚C (44%) base pairs.In contrast, Pb induces a high G‚C base substitutionproportion (77%), and Cr(VI) induces predominantly T‚Abase substitutions (92%). This comparison indicates thateach metal induces a specific mutational spectrum inCHO-K1 cells.It has been shown that Cd increases H2O2 and Fe levels

in rat testicular Leydig cells (51). However, Cd alone didnot cause DNA single-strand breakage nor did it enhancethose induced by H2O2 in Leydig cells (55). Interactionof Fe or Cu with H2O2 through Fenton reaction could

Table 4. Comparison of hprt Mutational Spectra in CHO Cells Induced by Cadmium Acetate, Lead Acetate, andChromium(VI) Oxide as Well as Those Derived Spontanously

mutation type Cd(CH3COO)2 spontaneousa CrO3b Pb(CH3COO)2c

total no. of mutantsd 54 (111) 64 (116) 27 (100) 56 (114)single-base substitutions 19 (35)e 34 (53)e 10 (37) 30 (54)etwo or more base substitutions 4 (7)e 0 6 (22) 0putative base substitutions 5 (9) 0 0 0small alterations 2 (4) 1 (2) 5 (19) 4 (7)single-exon skips 10f (19) 15 (23) 6 (22) 10 (18)two or three exon skips 7 (13) 0 0 6 (11)large genomic DNA deletions 13 (24)g 15 (23) NA 14 (25)large genonic duplications/insertions 0 4 (6) NA 0undefined mutations 0 5 (8) 0 0

total no. of base substitutions 27 34 24 30

transversions 22 (82) 22 (65) 20 (83) 16 (54)A‚T f T‚A 7 (26) 2 (6) 8 (33) 2 (7)A‚T f C‚G 5 (19) 4 (12) 10 (42) 0G‚C f C‚G 8 (30) 4 (12) 1 (4) 6 (20)G‚C f T‚A 2 (7) 12 (35) 1 (4) 8 (27)

transitions 5 (18) 12 (35) 4 (17) 14 (47)G‚C f A‚T 3 (11) 10 (29) 0 9 (30)A‚T f G‚C 2 (7) 2 (6) 4 (17) 5 (17)

a Data adopted from Xu et al. (30). b Data adopted from Yang et al. (31). NA, not available. c Data adopted from Yang et al. (32).d Numbers in parentheses are percentages. Total percentages may exceed 100% because some base substitutions leading to splice errorsare included in single-base substitutions or two-base substitutions. e Including base substitutions at splice boundaries. f Including twomutants having partial exon 8 or partial exon 9 deletions in their mRNA. g The hprt cDNA of these mutants could not be amplified byRT-PCR.


potentiate the generation of hydroxyl radicals and singletoxygen that is more reactive toward macromolecules (21,23). Accordingly, it has been suggested that Cd geno-toxicity is the result of its ability to enhance Fenton-typedamage in cells (55). Previous studies have shown themutational spectrum induced by ROS damage throughFenton metals (Fe, Cu, and Ni) during replication ofdamaged plasmids or phage single-stranded DNA in cells(25, 26). All of the results showed that G and/or C basesare the mutational sites. The G‚C f C‚G induced by Cdmay be derived from singlet-oxygen damage to DNA (24,26). However, nearly 50% of the Cd-induced base sub-stitutions in the endogenous hprt gene occurred at T‚Abase pairs that were rarely observed in mutants gener-ated by Fenton metals. Thus, the indirect DNA-damagemodel cannot completely explain Cd-induced mutagen-esis.The interaction between Cd and H2O2 in vitro under

physiological conditions can generate hydroxyl radicalsdetected using electron paramagnetic resonance spectro-photometry, but the intensity of hydroxyl radicals pro-duced by Cd plus H2O2 is weaker than those generatedby typical Fenton metals.2 Considering that Cd enhancesH2O2 levels and it also binds to DNA, the generation ofROS damage at specific Cd binding sites may account inpart for the unique mutational spectrum induced by Cd.Cd exhibits a higher affinity for T‚A pairs than othermetals, e.g., Zn and Cu (10). The interaction of Cd withT‚A pairs of DNA may account for the high frequency ofT‚A base substitutions induced by this metal.The exact nature of DNA adducts induced by Cd has

not been defined. Because Cd can enhance ROS levelsand its mutagenicity is ROS-dependent, the mutationalspectrum of Cd was thus compared with those derivedfrom ROS adducts. ROS produce a broad spectrum ofdamage in DNA (23). Each specific ROS-induced DNAadduct may have a different mutagenic effect. Some ofthe most stable forms of these adducts, such as 8-hy-droxydeoxyguanosine (56, 57), apurinic sites (58), 5-hy-droxydeoxycytidine (59), and two thymine ring fragmen-tation products, urea and â-ureidoisobutyric acid (60),have been studied for mutagenicity. Mutagenic replica-tion of 8-hydroxydeoxyguanosine as a template resultsin G f T transversions due to 8-hydroxydeoxyguanosine‚Amispairing (56, 57). Similarly, misincorporation of 8-hy-droxydeoxyguanosine as a substrate leads to A f Ctransversions (57). Apurinic sites can lead to basesubstitutions because DNA polymerases can insert abase, most often adenine, opposite an apurinic site (58).The low frequency of G‚C f T‚A transversions observedin Cd mutants indicates that 8-hydroxydeoxyguanosineand/or apurinic sites derived from guanine in DNA maynot be lesions accounting for Cd-induced mutagenesis inCHO-K1 cells. Similarly, 5-hydroxydeoxycytidine maybe not important for Cd mutagenesis because incorporat-ing this ROS-damaged substrate into DNA generates Cf T transitions (59) that were not significantly inducedby Cd. The predominant base substitutions derived fromthe thymine ring fragmentation product urea are T f Ctransitions and the â-ureidoisobutyric acid substitutionsare T f A transversions (60). â-Ureidoisobutyric acidmay be a potent ROS adduct in Cd mutagenesis becauseCd enhances T‚A f A‚T transversions.As discussed above, metal mutational specificity may

be related to DNA adducts generated through the inter-action of metal-DNA and ROS. Other factors may also

affect metal mutagenicity. The mutagenic metals mightalter the fidelity of DNA polymerases in different waysthat contribute to a unique mutational spectrum (11, 26).Furthermore, metals have been reported to alter cellularDNA repair systems (12). Further investigation into thespecific types of adducts generated by Cd in cells and thestudy of the mutagenic potential of those adducts wouldgreatly help elucidate the origins of base substitutionsinduced by this metal.

Acknowledgment. The authors would like to thankthe Department of Health, Republic of China, for finan-cial support of this manuscript under Contract DOH84-HR-407.

References

(1) Ochi, T., and Ohsawa, M. (1983) Induction of 6-thioguanine-resistant mutants and single-strand scission of DNA by cadmiumchloride in cultured Chinese hamster cells.Mutat. Res. 111, 69-78.

(2) Ochi, T., Mogi, M., Watanabe, M., and Ohsawa, M. (1984)Induction of chromosome aberrations in cultured Chinese hamstercells by short-term treatment with cadmium chloride.Mutat. Res.137, 103-109.

(3) Friberg, L., Kjellstrom, T., and Nordberg, G. F. (1986) Cadmium.In Handbook on the Toxicology of Metals, 2nd ed. (Friberg, L.,Nordberg, G. F., and Vouk, V., Eds.) pp 130-184, Elsevier Science,Amsterdam.

(4) Howard, W., Leonard, B., Moody, W., and Kochhar, T. S. (1991)Induction of chromosome changes by metal compounds in culturedCHO cells. Toxicol. Lett. 56, 179-186.

(5) Waalkes, M. P., and Oberdorster, G. (1990) Cadmium carcino-genesis. In Biological Effects of Heavy Metals (Foulkes, E. C., Ed.)Vol. 2, pp 129-157, CRC Press, Boca Raton, FL.

(6) Waalkes, M. P., and Rehm, S. (1994) Chronic toxic and carcino-genic effects of cadmium chloride in male DBA/2NCr and NFS/NCr mice: strain-dependent association with tumors of thehematopoietic system, injection site, liver, and lung. Fundam.Appl. Toxicol. 23, 21-31.

(7) IARC. (1993) Cadmium and cadmium compounds. In IARCMonographs on the evaluation of carcinogenic risks to humans:Beryllium, cadmium, mercury, and exposures in the glass manu-facturing industry, Vol. 58 pp 119-237, International Agency forResearch on Cancer, Lyon.

(8) Jacobson, K. B., and Turner, J. E. (1980) The interaction ofcadmium and certain other metal ions with proteins and nucleicacids. Toxicology 16, 1-37.

(9) Koizumi, T., and Waalkes, M. P. (1990) Effects of zinc on thebinding of cadmium to DNA: assessment with testicular inter-estitial cell and calf thymus DNAs. Toxicol. in Vitro 4, 51-55.

(10) Langlais, M., Tajmir-Riahi, H. A., and Savoie, R. (1990) Ramanspectroscopic study of the effects of Ca2+, Mg2+, Zn2+, and Cd2+

ions on calf thymus DNA: Binding sites and conformationalchanges. Biopolymers 30, 743-752.

(11) Zakour, R. A., Kunkel, T. A., and Loeb, L. A. (1981) Metal-inducedinfidelity of DNA synthesis. Environ. Health Perspect. 40, 197-205.

(12) Hartwig, A. (1994) Role of DNA repair inhibition in lead- andcadmium-induced genotoxicity: a review. Environ. Health Perpect.102 (Suppl. 3), 45-50.

(13) Vallee, B. L., and Ulmer, D. D. (1972) Biochemical effects ofmercury, cadmium and lead. Annu. Rev. Biochem. 41, 91-128.

(14) Predki, P. F., and Sarkar, B. (1994) Metal replacement in "zincfinger" and its effect on DNA binding. Environ. Health Perspect.102 (Suppl. 3), 195-198.

(15) Waalkes, M. P., and Goering, P. L. (1990) Metallothionein andother cadmium-binding proteins: recent developments. Chem.Res. Toxicol. 3, 281-288.

(16) Masters, B. A., Kelly, E. J., Quaife, C. J., Brinster, R. L., andPalmiter, R. D. (1994) Targeted disruption of metallothionein Iand II genes increases sensitivity to cadmium. Proc. Natl. Acad.Sci. U.S.A. 91, 584-588.

(17) Lazo, J. S., Kondo, Y., Dellapiazza, D., Michalska, A. E., Choo,K. H. A., and Pitt, B. R. (1995) Enhanced sensitivity to oxidativestress in cultured embryonic cells from transgenic mice deficientin metallothionein I and II genes. J. Biol. Chem. 270, 5506-5510.

(18) Cheung, W. Y. (1988) Calmodulin and its activation by cadmiumion. Ann. N.Y. Acad. Sci. 522, 74-87.

(19) Beyersmann, D., Block, C., and Malviya, A. N. (1994) Effects ofcadmium on nuclear protein kinase C. Environ. Health Perspect.102 (Suppl. 3), 177-180.2 Unpublished data.


(20) Smith, J. B., Smith, L., Pijuan, V., Zhuang, Y., and Chen, Y.-C.(1994) Transmembrane signals and protooncogene inductionevoked by carcinogenic metals and prevented by zinc. Environ.Health Perspect. 102 (Suppl. 3), 181-189.

(21) Imlay, J. A., and Linn, S. (1988) DNA damage and oxygen radicaltoxicity. Science 240, 1302-1309.

(22) Sinha, B. K., and Mimnaugh, E. G. (1990) Free radicals andanticancer drug resistance: oxygen free radicals in the mecha-nisms of drug cytotoxicity and resistance by certain tumors. FreeRadical Biol. Med. 8, 567-581.

(23) Halliwell, B., and Aruoma, O. I. (1991) DNA damage by oxygen-derived species: its mechanism and measurement in mammaliansystems. FEBS Lett. 281, 9-19.

(24) Oliveira, R. C., Ribeiro, D. T., Nigro, R. G., Mascio, P. D., andMenck, C. F. M. (1992) Singlet oxygen induced mutation spectrumin mammalian cells. Nucleic Acids Res. 16, 4319-4323.

(25) Akasaka, S., and Yamamoto, K. (1994) Hydrogen peroxide inducesG:C to T:A and G:C to C:G transversions in the supF gene ofEscherichia coli. Mol. Gen. Genet. 243, 500-505.

(26) Reid, T. M., Feig, D. I., and Loeb, L. A. (1994) Mutagenesis bymetal-induced oxygen radicals. Environ. Health Perspect. 102(Suppl. 3), 57-61.

(27) Feig, D. I., and Loeb, L. A. (1994) Oxygen radical inducedmutagenesis is DNA polymerase specific. J. Mol. Biol. 235, 33-41.

(28) Ochi, T., Ishiguro, T., and Ohsawa, M. (1983) Participation ofactive oxygen species in the induction of DNA single-strandscissions by cadmium chloride in cultured Chinese hamster cells.Mutat. Res. 122, 169-175.

(29) Ochi, T., and Ohsawa, M. (1985) Participation of active oxygenspecies in the induction of chromosomal aberrations by cadmiumchloride in cultured Chinese hamster cells.Mutat. Res. 143, 137-142.

(30) Xu, Z., Yu, Y., Schwartz, J. L., Meltz, M. L., and Hsie, A. W. (1995)Molecular nature of spontaneous mutations at the hypoxanthine-guanine phosphoribosyltransferase (hprt) locus in Chinese ham-ster ovary cells. Environ. Mol. Mutagen. 26, 127-138.

(31) Yang, J.-L., Hsieh, Y.-C. , Wu, C.-W., and Lee, T.-C. (1992)Mutational specificity of chromium(VI) compounds in the hprtlocus Chinese hamster ovary-K1 cells. Carcinogenesis 13, 2053-2057.

(32) Yang, J.-L., Yeh, S.-C., and Chang, C.-Y. (1996) Lead acetatemutagenicity and mutational spectrum in the hprt gene ofChinese hamster Ovary-K1 cells. Mol. Carcinogen. In press.

(33) Aebi, H. (1984) Catalase in vitro. Methods Enzymol. 105, 121-126.

(34) Flohe, L., and Gunzler, W. A. (1984) Assays of glutathioneperoxidase. Methods Enzymol. 105, 114-121.

(35) Carlberg, I., and Mannervik, B. (1984) Glutathione reductase.Methods Enzymol. 105, 484-495.

(36) Cohn, V. H., and Lyle, J. (1966) A fluorometric assay forglutathione. Anal. Biochem. 14, 434-440.

(37) Bradford, M. M. (1976) A rapid and sensitive method for thequantitation of microgram quantities of protein utilizing theprinciple of protein-dye binding. Anal. Biochem. 72, 248-254.

(38) Yang, J.-L., Maher, V. M., and McCormick, J. J. (1989) Amplifica-tion and direct nucleotide sequencing of cDNA from the lysate oflow numbers diploid human of cells. Gene 83, 347-354.

(39) Konecki, D. S., Brennand, J., Fuscoe, J. C., Caskey, C. T. andChinault, A. C. (1982) Hypoxanthine-guanine phosphoribosyl-transferase genes of mouse and Chinese hamster: constructionand sequence analysis of cDNA recombinants. Nucleic Acids Res.10, 6762-6775.

(40) Yang, J.-L., Hu, M.-C., and Wu, C.-W. (1991) Novel mutationalspectrum induced byN-methyl-N′-nitro-N-nitrosoguanidine in thecoding region of the hypoxanthine (guanine) phosphoribosyltrans-ferase gene in diploid human fibroblasts. J. Mol. Biol. 221, 421-430.

(41) Rossiter, B. J. F., Fuscoe, J. C., Muzny, D. M., Fox, M., andCaskey, C. T. (1991) The Chinese hamster HPRT gene: restrictionmap, sequence analysis, and multiplex PCR deletion screen.Genomics 9, 247-256.

(42) Tsuchiya, N., and Ochi, T. (1994) Mechanism for the decrease inthe accumulation of cadmium (Cd) in Cd-resistant Chinesehamster V79 cells. Arch. Toxicol. 68, 325-331.

(43) Lichtenauer-Kaligis, E. G. R., Thijssen, J. C. P., Den Dulk, H.,Van de Putte, P., Giphart-Gassler, M., and Tasseron-de Jong, J.G. (1995) Spontaneous mutation spectrum in the hprt gene inhuman lymphoblastoid TK6 cells. Mutagenesis 10, 137-143.

(44) Giver, C. R., Nelson, S. L., and Grosovsky, A. J. (1993) Spectrumof spontaneous HPRT- mutations in TK6 human lymphoblasts.Environ. Mol. Mutagen. 22, 138-146.

(45) Hou, S.-M., Steen, A.-M., Falt, S., and Andersson, B. (1993)Molecular spectrum of background mutation at the hprt locus inhuman T-lymphocytes. Mutagenesis 8, 43-49.

(46) Jego, N., Thomas, G., and Hamelin, R. (1993) Short direct repeatsflanking deletions, and duplicating insertions in p53 gene inhuman cancers. Oncogene 8, 209-213.

(47) Kunkel, T. A. (1990) Misalignment-mediated DNA synthesiserrors. Biochemistry 29, 8003-8011.

(48) Wei, S.-J. C., Chang, R. L., Bhachech, N., Cui, X. X., Merkler, K.A., Wong, C-Q., Hennig, E., Yagi, H., Jerina, D. M., and Conney,A. H. (1993) Dose-dependent differences in the profile of muta-tions induced by (+)-7R,8S-dihydroxy-9S,10R-epoxy-7,8,9,10-tet-rahydrobenzo(a)pyrene in the coding region of the hypoxanthine(guanine) phosphoribosyltransferase gene in Chiense hamsterV-79 cells. Cancer Res. 53, 3294-3301.

(49) Yang, J.-L., Hsieh F.-P., Lee, P.-C., and Tseng, H.-J. R. (1994)Strand- and sequence-specific attenuation of N-methyl-N′-nitro-N-nitrosoguanidine-induced G‚C to A‚T transitions by expressionof human O6-methylguanine-DNA methyltransferase in Chinesehamster ovary cells. Cancer Res. 54, 3857-3863.

(50) Tabatabaie, T., and Floyd, R. A. (1994) Susceptibility of gluta-thione peroxidase and glutathione reductase to oxidative damageand the protective effect of spin traping agents. Arch. Biochem.Biophys. 314, 112-119.

(51) Koizumi, T., and Li, Z. G. (1992) Role of oxidative stress in single-dose, cadmium-induced testicular cancer. J. Toxicol. Environ.Health 37, 25-36.

(52) Ochi, T., Takahashi, K., and Ohsawa, M. (1987) Indirect evidencefor the inducton of a prooxidant state by cadmium chloride incultured mammalian cells and a possible mechanism for theinduction. Mutat. Res. 180, 257-266.

(53) Muller, L. (1986) Consequences of cadmium toxicity in rathepatocytes: effect of cadmium on the glutathione peroxidasesystem. Toxicol. Lett. 30, 259-265.

(54) Splittgerber, A. G., and Tappel, A. L. (1979) Inhibition ofglutathione peroxidase by cadmium and other metal ions. Arch.Biochem. Biophys. 197, 534-542.

(55) Koizumi, T., Li, Z. G., and Tatsumoto, H. (1992) DNA damagingactivity of cadmium in Leydig cells, a target cell population forcadmium carcinogenesis in the rat testis. Toxicol. Lett. 63, 211-220.

(56) Wood, M. L., Dizdaroglu, M., Gajewski, E., and Essigmann, J. E.(1990) Mechanistic studies of ionizing radiation and oxidativemutagenesis: genetic effects of a single 8-hydroguanine (7-hydro-8-oxoguanine) residue inserted at a unique site in a viral genome.Biochemistry 29, 7024-7032.

(57) Cheng, K. C., Cahill, D. S., Kasai, H., Nishimura, S., and Loeb,L. (1992) 8-Hydroxyguanine, an abundant form of oxidative DNAdamage, cause G f T and A f C substitutions. J. Biol. Chem.267, 166-172.

(58) Loeb, L. A., and Preston, B. D. (1986) Mutagenesis by apurinic/apyrimidinic sites. Annu. Rev. Genet. 20, 201-230.

(59) Feig, D. I., Sowers, L. C., and Loeb, L. A. (1994) Reverse chemicalmutagenesis: identification of the mutagenic lesions resultingfrom reactive oxygen species-mediated damage to DNA. Proc.Natl. Acad. Sci. U.S.A. 91, 6609-6613.

(60) Maccabee, M., Evans, J. S., Glackin, M. P., Hatahet, Z., andWallace, S. S. (1994) Pyrimidine ring fragmentation products:effects of lesion structure and sequence context on mutagenesis.J. Mol. Biol. 236, 514-530.

TX960122Y


Reactive Oxygen Species May Participate in the Mutagenicity and Mutational Spectrum of Cadmium in...

Documents

Transcript of Reactive Oxygen Species May Participate in the Mutagenicity and Mutational Spectrum of Cadmium in...