Disruption of dioxin-inducible phase I and phase II gene expression patterns by cadmium, chromium,...

Disruption of Dioxin-Inducible Phase Iand Phase II Gene Expression Patternsby Cadmium, Chromium, and Arsenic

Andrew Maier,* Timothy P. Dalton, and Alvaro Puga

Center for Environmental Genetics and Department of Environmental Health, University of Cincinnati Medical Center,Cincinnati, Ohio

Recent work suggesting that cellular oxidative stress exerts an inhibitory effect on aromatic hydrocarbon receptor(AHR)±dependent gene expression led us to test the hypothesis that pro-oxidant environmental pollutants might alterthe induction of detoxification genes by 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD), an AHR ligand. We found that,in mouse hepatoma Hepa-1 cells, TCDD-inducible cytochrome P450, Cyp1a1, and nicotinamide adenine dinucleotidephosphate±quinone oxidoreductase (Nqo1) mRNA accumulation were differentially affected by cadmium (Cd2+),chromium (Cr6+), and arsenic (As3+). Cadmium or arsenic did not change Cyp1a1 mRNA levels but did enhance TCDD-inducible levels of Nqo1 mRNA, an effect that paralleled the ability of these metals to activate a b-galactosidase genereporter system regulated by an electrophile response promoter element. Chromium inhibited mRNA accumulation forboth Cyp1a1 and Nqo1. Manipulation of cellular thiol status did not modify the response to combined chromium-TCDD exposure, suggesting that the response was not caused by oxidative stress. Chromium did not block DNA-binding competence of the AHR and did not have an effect on mRNA stability, but it inhibited Cyp1a1 genetranscription and the expression of an AHR-dependent luciferase reporter. These data indicate that coexposure to pro-oxidant metals and AHR ligands, which is common in the environment, can disrupt the regulation of phase I and phaseII detoxification genes, leading to imbalances in gene expression that may have important consequences for thetoxicity of complex mixtures. Mol. Carcinog. 28:225±235, 2000. ß 2000 Wiley-Liss, Inc.

Key words: aromatic hydrocarbon receptor; 2,3,7,8-tetrachlorodibenzo-p-dioxin; cadmium; chromium; arsenic;mixtures; oxidative stress; electrophile response element; DNA crosslinks; transcription

INTRODUCTION

The aromatic hydrocarbon receptor (AHR) plays acentral role in the toxicity of numerous aromaticcompounds by controlling the expression of abattery of detoxi®cation enzymes responsible fortheir metabolism [1]. The coordinated regulation ofthese detoxi®cation genes suggests that, to avoid aserious imbalance in detoxi®cation homeostasis,increases in production of reactive metabolites byphase I enzymes need to be accompanied bycomparable increases in conjugation of hydrophilicgroups by phase II enzymes. Hence, coordinatedexpression of the genes encoding phase I enzymes,such as the cytochromes P450 (CYP1A1, CYP1A2,and CYP1B1), and phase II enzymes, such asglutathione-S-transferase A1 (GSTA1), nicotinamideadenine dinucleotide phosphate±quinone oxi-doreductase (NQO1), and UDP-dependent glucur-onosyltransferease (UGT1A6), is an importantdeterminant of toxicant fate.

The importance of the AHR in the toxic responsesinduced by many of its ligands, such as benzo(a)-pyrene (BaP) and 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD), has been well demonstrated inmouse strains harboring Ahr alleles that encode

AHRs of different ligand af®nities [2,3] and in Ahrknockout mice [4,5], but the role of nonligands asmodi®ers of these responses has been less wellstudied. Recent reports have demonstrated thatmodi®cation of AHR-dependent gene expressioncan result from oxidative stress, suggesting thatcoexposure to AHR ligands and pro-oxidant envir-onmental pollutants may disrupt the coordinatedregulation of detoxi®cation genes. The precise stepin the AHR signaling pathway that is acted on by

MOLECULAR CARCINOGENESIS 28:225±235 (2000)

ß 2000 WILEY-LISS, INC.

*Correspondence to: Toxicology Excellence for Risk Assessment,1757 Chase Avenue, Cincinnati, OH 45223.

Received 29 December 1999; Revised 17 May 2000; Accepted 7June 2000

Abbreviations: AHR, aromatic hydrocarbon receptor; CYP, cyto-chrome P450; Gsta1, glutathione-S-transferase A1; Nqo1, nicotina-mide adenine dinucleotide phosphate±quinone oxidoreductase;UGT1A6, UDP-glucuronosyltransferase; BaP, benzo(a)pyrene;TCDD, 2,3,7,8-tetrachlorodibenzo-p-dioxin; BSO, L-buthionine(S,R)-sulfoximine; NAC, N-acetyl cysteine; EpRE, electrophileresponse element; LDH, lactate dehydrogenase; SSC, sodiumchloride/sodium citrate; SDS, sodium dodecyl sulfate; Ho, hemeoxygenase; Sod1, Cu,Zn superoxide dismutase; Mt-I, metallothio-nein I; Gclr, glutamate cysteine ligase regulatory subunit; AhRE,aromatic hydrocarbon receptor response element; DMSO, dimethyl-sulfoxide.

oxidative stress is not clear. Sulfhydryl-modifyingagents have been shown to block binding of TCDDto the AHR [6±8] and AHR DNA binding [9,10],suggesting that AHR activation may be directlyredox regulated. In addition, H2O2 has been shownto inhibit Cyp1a1 inducibility [11] and disruptreporter gene expression driven by AHR-responsivepromoter sequences [12]. It remains unclear, how-ever, whether these effects are mediated through theAHR or through other transcription factors, such asNF-1, which may be an important determinant ofthe oxidative regulation of the Cyp1a1 gene [13,14].

The ability of cadmium, chromium, and arsenicto generate oxidative stress has been well documen-ted [15,16]. Both cadmium and arsenic are able tobind sulfhydryl groups avidly, and changes in thelevels of cellular thiols such as glutathione, eitherdirectly or through formation of reactive oxygenspecies, may be an important avenue for thebiological activity of these metals [17±21]. Inaddition, increases in production of reactive oxygenspecies and lipid peroxidation products have beenshown in both cadmium- [19,21±23] and arsenic-[18,20,24,25] treated cells and tissues. Increasedformation of reactive oxygen species and reactivechromium intermediates has also been reportedafter chromium treatment and is thought to resultfrom the intracellular reduction of hexavalentchromium by cellular reductants, such as glu-tathione or ascorbic acid [26±30]. Based on thesediverse mechanisms of oxidative stress generation,we chose to evaluate markers for cellular oxidativestress encompassing multiple pathways, includingglutathione depletion and induction of oxidativestress±responsive genes.

Many sources of environmental exposure to AHRligands involve metal coexposure. For example,carcinogenic metals and AHR ligands are commoncontaminants of hazardous waste sites and arecoreleased from sources such as fossil fuel combus-tion and municipal waste incineration and ascomponents of cigarette smoke. Estimated humantissue concentrations in the low micromolar rangefor cadmium, chromium [31,32], and arsenic[reviewed in 33] have been reported, even inindividuals without known exposure, thus con®rm-ing the high prevalence of background exposure.Based on the potential for exposure and seriousnature of their toxic effects, cadmium, chromium,arsenic, BaP, and polychlorinated biphenyls, includ-ing the AHR-activating mixture arochlor 1254,continue to be high priorities for environmentalremediation [34].

To investigate further the potential interactionsbetween these environmental pollutants, we testedthe effect of cadmium, chromium, and arsenic onAHR-dependent gene expression and on oxidativestress responses in mouse hepatoma Hepa-1 cells.We found that metal coexposure led to changes in

the expression patterns of detoxi®cation genesinduced by TCDD.

MMEETTHHOODDSS

CCeellll LLiinneess,, GGrroowwtthh CCoonnddiittiioonnss,, aanndd CChheemmiiccaall TTrreeaattmmeennttss

Mouse hepatoma Hepa-1 cells were seeded toapproximately 25±35% con¯uence and grown over-night in a-minimal essential medium (GibcoBRLLife Technologies, Grand Island, NY) containing 5%fetal bovine serum and 1% antibiotics, as describedpreviously [35]. Cadmium chloride, potassiumdichromate, and sodium arsenite (hereinafterreferred to simply as cadmium, chromium, andarsenic; Sigma Co., St. Louis, MO) were dissolvedfresh in sterile deionized water. L-buthionine (S,R)-sulfoximine (BSO) and N-acetyl cysteine (NAC) weredissolved in serum-free medium and stored atÿ20�C until use. Speci®c treatment regimens aredescribed in the captions for each ®gure.

To assess AHR-dependent responses, we used aHepa-1±derived cell line stably transformed withthe luciferase reporter gene regulated by the AHR-binding domain in the promoter of the mouseCyp1a1 gene, pAhRDTKLuc3. The pAhRDTKLuc3plasmid was constructed by subcloning three AHR-binding elements from the C57BL6 mouse Cyp1a1gene promoter (base pairs ÿ1100 to ÿ869) and thethymidine kinase minimal promoter into thereporter plasmid pGL3 basic (Promega, Madison,WI) [35]. Similarly, to determine electrophile-dependent responses, we used a Hepa-1 cell linestably transformed with a b-galactosidase±neomy-cin-resistance reporter plasmid, pEpREbgeo. ThepEpREbgeo plasmid was generated by subcloning asingle electrophile response element (EpRE) fromthe mouse Gsta1 enhancer (base pairs ÿ754 toÿ714) [36] in front of the metallothionein geneminimal promoter to drive expression of a b-galactosidase- and neomycin-resistance gene. Thesestably transfected cell lines were grown underconditions similar to those of Hepa-1 cells, withthe addition of geneticin for cell selection.

CCyyttoottooxxiicciittyy AAnnaallyyssiiss

The effect of metal and TCDD treatments on cellviability was determined by measuring lactatedehydrogenase (LDH) release into the medium.Cells were plated and grown for 24 h in tissueculture medium as described above, the mediumwas subsequently removed, and the cells were rinsedtwice with phosphate-buffered saline. Fetal bovineserum has signi®cant LDH activity, and phenol redabsorbs strongly at 420 nm even in the absence ofcells. As a result, for the metal treatments, themedium was replaced with D-minimal essentialmedium without phenol red and supplementedwith 0.25% fetal bovine serum, 2.5 mg/mL bovineserum albumin, and 1% antibiotics. Positive con-

222266 MMAAIIEERR EETT AALL..

trols included cells incubated in sterile water for 30min before the addition of medium. Treatmentswere added to the medium, and 100 mL of mediumwas removed from each culture at 12 and 24 h aftertreatment. LDH activity in the medium was deter-mined by using a TOX-7 assay kit as recommendedby the manufacturer (Sigma Co.).

RRNNAA IIssoollaattiioonn aanndd NNoorrtthheerrnn BBlloott AAnnaallyyssiiss

RNA was isolated with Tri-reagent (MRC, Inc.,Cincinnati, OH). Northern blotting and hybridiza-tions were performed as described elsewhere [37].Brie¯y, aliquots of 5±10 mg of total RNA wereseparated in a 1% agarose gel and transferred toNYTRAN Plus membranes (Schleicher & Schuell,Keene, NH) in 20� sodium chloride/sodium citrate(SSC; 3 M NaCl and 0.3 M sodium citrate, pH 7.0).Blots were incubated for 1 h at 65�C in prehybridi-zation buffer (0.45 M NaCl; 90 mM Tris-HC; pH 8;6 mM EDTA; 0.1% sodium dodecyl sulfate (SDS);2 mg/mL each Ficoll, polyvinylpyrrolidone; andbovine serum albumin; and 250 mg/mL yeast tRNA)and hybridized in the same buffer for at least 12 hwith 1 � 106 dpm/mL of the appropriate probes. Fora Cyp1a1 probe, we used the StuI fragment derivedfrom the 3 0 untranslated region of the mouse cDNA.For a Nqo1 probe, we used the full-length mousecDNA, as described elsewhere [38]. Probes for hemeoxygenase (Ho) and Cu,Zn superoxide dismutase(Sod1) were prepared from full-length rat cDNAclones, and the probe for metallothionein I (Mt-I)was prepared from a mouse cDNA clone, asdescribed previously [39]. The primers GATTTGGT-CAGGGAGTTTCC (sense) and AATGCAGTCAAA-TCTGGTGG (antisense) were used to amplify cDNAsequences of the regulatory subunit of glutamatecysteine ligase (Gclr) to generate a probe correspond-ing to nucleotides 180±607 relative to the transla-tion initiation site. After hybridization, blots werewashed in 0.3� SSC and 0.1% SDS until thebackground was negligible and visualized by expo-sure to X-omat-AR ®lm (Kodak, Rochester, NY).Intensity of the resulting signal was quanti®ed on aStorm 860 phosphorimager (Molecular Dynamics,Sunnyvale, CA).

EElleeccttrroopphhoorreettiicc MMoobbiilliittyy SShhiifftt AAssssaayy

Nuclear extract preparation and binding condi-tions were performed as described previously [40].Brie¯y, 15 mg of nuclear protein extract waspreincubated in binding buffer (10% glycerol, 1mM EDTA, 1 mM dithiothreitol, 240 mM KCl, 1 mgof poly[dI-dC], and 20 mM HEPES, pH 7.8) for 15min at room temperature. An AHR response ele-ment (AhRE) double-stranded probe end-labeledwith [g-32P]ATP (Amersham Life Science, ArlingtonHeights, IL) was added (10 000 dpm/reaction), andthe reaction was incubated for an additional 15 minat room temperature. The reaction products were

separated by electrophoresis in 4.5% nondenaturingacrylamide gels. Gels were dried at 80�C for 1 h andexposed overnight to X-omat-AR ®lm.

NNuucclleeaarr RRuunn--OOnn AAssssaayyss

Nuclei were prepared as described elsewhere [41].Brie¯y, cells were harvested in ice-cold phosphate-buffered saline, spun down at 500 � g, and resus-pended in lysis buffer (10 mM Tris-HCl, pH 7.4;3 mM CaCl2; and 2 mM MgCl2). The cells wererecentrifuged as previously described and resus-pended in 2 mL of lysis buffer and 2 mL of the samelysis buffer containing 1% NP-40. The cells weredisrupted by 10 strokes in a dounce homogenizer,and the nuclei were pelleted by centrifugation asbefore. The supernatant was removed, and thenuclei were resuspended in storage buffer (50 mMTris-HCl, pH 8.3; 40% glycerol; 5 mM MgCl2; and0.1 mM EDTA) and stored at ÿ70�C. The hybridiza-tion targets were the full-length Cyp1a1 cDNA [42]and, for actin, was derived from rat cDNA [43]. Thehybridization targets were denatured in 100 mMNaOH for 30 min at room temperature, and 5 mg ofeach DNA was applied to a NYTRAN Plus membraneby ®ltration. Blots were neutralized with 6� SSC,and the DNA was crosslinked to the support byultraviolet irradiation. Run-on reactions were con-ducted as described previously [37], with minormodi®cations. For labeling of newly transcribedRNA, approximately 5 � 107 nuclei in 100 mL wereadded to 100 mL of 2� reaction buffer (10 mM Tris-HCl, pH 8; 5 mM MgCl2; 0.16 M KCl; 500 mM eachATP, GTP, and CTP; 5mM dithiothreitol; and150 mCi/mL [a-32P]UTP). The reaction was incubatedat 30�C for 30 min and was stopped by the additionof 30 U of RQ1 DNase (Promega). After incubationfor 10 min at 37�C, 600 mL of stop solution (2% SDS,7 M urea, 0.35 M LiCl, 1 mM EDTA, and 10 mM Tris-HCl, pH 8) was added. The reaction was incubatedfor 1 h at 56�C, and the labeled RNA was isolated bysubsequent trichloroacetic acid and ethanol preci-pitations. Hybridization was performed in the samebuffers as for Northern blot analysis, with 7.5 � 106

dpm/mL for 30 h at 65�C.

SSttaattiissttiiccaall AAnnaallyyssiiss

Group comparisons were made by one-wayanalysis of variance. Differences were consideredsigni®cant at P < 0.05.

RREESSUULLTTSS

DDoossee-- aanndd TTiimmee--DDeeppeennddeenntt DDeeccrreeaasseess iinn CCeellll VViiaabbiilliittyyGGeenneerraatteedd bbyy MMeettaall TTrreeaattmmeennttss

Studies of the biologic mechanisms of toxicity ofcadmium, chromium, and arsenic in the existingliterature have employed a wide dose range, perhapsre¯ecting different species and tissue sensitivity. Todetermine the appropriate concentrations for use in

MMEETTAALLSS AALLTTEERR PPHHAASSEE II AANNDD PPHHAASSEE IIII GGEENNEE IINNDDUUCCTTIIOONN 222277

our cell model (Hepa-1 cells), we evaluated the effectof metal treatments on release of LDH activity in thecell-culture medium relative to that of positivecontrols of cells lysed by hypotonic disruption.The cytotoxicity of the metals was dose and timedependent. After a 12-h incubation, only 50 mMchromium increased LDH activity. In contrast,treatment for 24 h resulted in signi®cantly increasedLDH release at 50 mM cadmium, 50 mM chromium,and 25 mM arsenic (Figure 1). In general, thecytotoxicity assay results corresponded well withdirect observation of the cells because no changes incell morphology were noted in any of the geneexpression or reporter gene experiments aftercadmium or chromium treatments at these dosesfor up to 8 h. In some cases, concentrations of 10 mMor greater of arsenic for 8 h increased cell roundingand the number of detached cells, although noeffect of arsenic was noted for shorter treatmentdurations. These results indicate that, under theconditions of the treatments reported in this study,the effects of metals were not likely due to decreasesin cell viability, although exposure of Hepa-1 cells atthe higher doses could result in overt cytotoxicityfor longer exposures.

AAlltteerreedd TTCCDDDD--IInndduucciibbllee mmRRNNAA LLeevveellss wwiitthhMMeettaall CCoottrreeaattmmeenntt

To test the effect of metals on TCDD-induciblegene expression, Cyp1a1 and Nqo1 mRNA accumu-lation were measured in cells treated with 1 nMTCDD plus different concentrations of cadmium,chromium, or arsenic. To preclude effects of treat-ment due to cytotoxicity, cells were exposed to pro-oxidants for a maximum of 4 h. TCDD treatmentinduced both Cyp1a1 and Nqo1, an effect that was

signi®cantly decreased by 50 mM chromium to 23%and 36% of TCDD-inducible levels for these twogenes, respectively (Figures 2 and 3). Cadmium orarsenic treatment had no effect on Cyp1a1 induc-tion by TCDD. Cadmium at 40 mM and arsenic at3 mM and 10 mM appeared to increase the TCDD-inducible levels of Nqo1 mRNA, but the magnitudeof the enhancing effect of the metals was variable.

We also determined the ability of these metals toinduce characteristic oxidative stress and metalresponses by measuring mRNA levels of Gclr, Ho,and Mt-I (Figures 2 and 3). The effects of cadmiumplus TCDD treatment on Gclr were similar to thoseobserved for Nqo1, with cadmium generating only aweak induction (2.1-fold) at 40 mM that was notsigni®cantly different from TCDD treatment alone.Arsenic was a potent inducer of Gclr, with treatmentat 10 mM plus TCDD resulting in a 28-fold increasein mRNA levels versus TCDD treatment alone.Chromium treatment had no effect on Gclr mRNA.The levels of Ho mRNA were greatly increased, 39-fold, over TCDD treatment by 10 mM arsenic plusTCDD and were not affected by the other metals.Mt-I mRNA levels were signi®cantly increased byeither 20 mM cadmium or 30 mM arsenic, whereaschromium treatment had no effect. It is unlikelythat TCCD had any effect on the observed inductionof oxidative stress or metal-responsive genes bycadmium (Mt-I) or arsenic (Gclr, Ho, and Mt-I)because TCDD treatment alone did not affectexpression of these genes and increased mRNAlevels were also observed for the metal-only treat-ments. In all cases, levels of Sod1 mRNA remainedunchanged, indicating that the observed mRNAchanges for other genes were not the result ofnonspeci®c changes in mRNA expression or differ-ences in gel loading.

To determine whether pro-oxidant treatmentsinduced electrophile-driven responses, Hepa-1 cellsstably transformed with an EpRE-responsive b-galactosidase reporter were treated with differentmetal concentrations (Figure 4). b-Galactosidaseactivity was increased 4.1-fold by 25 mM cadmiumand 5.6-fold by 5 mM arsenic. No response occurredafter chromium treatment at any of the concentra-tions used. These results, which paralleled thepattern of mRNA accumulation for EpRE-regulatedgenes (Nqo1 and Gclr), suggested that cadmium andarsenic, but not chromium, effectively inducedoxidative stress, as indicated by activation of theelectrophile response pathway.

DDeeccrreeaasseedd AAccttiivvaattiioonn ooff aann AAHHRR--RReessppoonnssiivvee LLuucciiffeerraasseeRReeppoorrtteerr wwiitthh CChhrroommiiuumm aanndd AArrsseenniicc CCoottrreeaattmmeenntt

Induction of Cyp1a1 has frequently been used as amarker for AHR activation. Because chromium wasable to inhibit Cyp1a1 induction partially and theAHR has been reported to be regulated by redoxchanges, we tested the effects of metals on induc-

FFiigguurree 11.. LLDDHH rreelleeaassee ffrroomm mmeettaall--ttrreeaatteedd HHeeppaa--11 cceellllss.. CCeellllss wweerreettrreeaatteedd wwiitthh ccaaddmmiiuumm,, cchhrroommiiuumm,, oorr aarrsseenniicc aatt tthhee iinnddiiccaatteeddccoonncceennttrraattiioonnss ffoorr 1122 hh oorr 2244 hh,, aanndd LLDDHH aaccttiivviittyy wwaass ddeetteerrmmiinneeddrreellaattiivvee ttoo cceellllss iinnccuubbaatteedd ffoorr 3300 mmiinn uunnddeerr hhyyppoottoonniicc ccoonnddiittiioonnss..EEaacchh ppooiinntt rreepprreesseennttss tthhee mmeeaann�� SSDD ooff dduupplliiccaattee ssaammpplleess ffrroomm ttwwooiinnddeeppeennddeenntt eexxppeerriimmeennttss.. TThhee rreellaattiivvee LLDDHH aaccttiivviittyy wwaass ccoommppaarreeddwwiitthh tthhee llooww ddoossee ffoorr eeaacchh ttrreeaattmmeenntt,, aanndd ssiiggnnii®®ccaanntt ddiiffffeerreenncceess((PP<< 00..0055)) aarree ddeennootteedd wwiitthh aasstteerriisskkss..


tion of an AhRE-responsive luciferase reporter stablytransformed in Hepa-1 cells. Treatment of these cellswith 100 pM TCDD generated a signi®cant 8.8-foldinduction in luciferase activity (Figure 5). This con-centration of TCDD causes submaximal gene induc-tion effects and was used to determine whether anyof the metal treatments would increase or decreaseTCDD-inducible activation of the AHR. Addition ofchromium or arsenic 30 min before TCDD decreasedTCDD-inducible luciferase activity in a concentra-tion-dependent fashion. Consistent with the north-ern blot analysis, cadmium did not generate anyeffect on TCDD-inducible luciferase activity.

We found that chromium inhibited AHR-mediated responses when administered 30 minbefore TCDD. Chromium generates oxidative stressrapidly [44±46], yet it inhibits BaP-induced muta-genesis when administered simultaneously with BaPbut not when added before or after BaP [47]. Hence,analysis of the timing of chromium addition was ofparticular interest. To test the effect of the timing ofchromium treatment on AHR-inducible responses,we measured the expression of AHR-dependentluciferase activity as a function of the time ofchromium addition relative to TCDD addition.

The greatest decrease in TCDD inducibility, to 14%of TCDD-inducible levels, occurred with the addi-tion of chromium 2 h before TCDD (Figure 6). Theeffect became less severe when the two reagentswere added simultaneously, and only a decrease to50% of control levels was observed when chromiumwas added 2 h after TCDD. These results suggest thatchromium interferes with an early step of induciblegene transcription.

RReeddooxx--IInnddeeppeennddeenntt CChhaannggeess iinn AAHHRR--RReessppoonnssiivveeGGeennee EExxpprreessssiioonn

To determine whether changes in AHR-depen-dent luciferase activity were due to metal-inducedoxidative stress, we tested the effect of BSO and NACaddition on AHR-dependent luciferase activity. BSOdepletes cellular glutathione levels in Hepa-1 cellsunder the present dosing regimen [48] and NACprotects against sulfhydryl-binding molecules bydirect inactivation of electrophilic species. There-fore, if the effects of these metals on AHR-dependentgene expression had an oxidative stress basis, BSOshould enhance the effect and NAC should decreaseit. Chromium, at a concentration of 25 mM,inhibited TCDD inducibility to 36% of the level

FFiigguurree 22.. NNoorrtthheerrnn bblloott aannaallyyssiiss ooff mmRRNNAA ffrroomm TTCCDDDD-- aanndd mmeettaall--ttrreeaatteedd HHeeppaa--11 cceellllss.. CCeellllss wweerree ttrreeaatteedd wwiitthh ccaaddmmiiuumm,, cchhrroommiiuumm,, oorraarrsseenniicc aatt tthhee iinnddiiccaatteedd ccoonncceennttrraattiioonnss ffoorr 3300 mmiinn bbeeffoorree ccoottrreeaatt--

mmeenntt wwiitthh 11 nnMM TTCCDDDD ffoorr 44 hh.. NNoorrtthheerrnn bblloottss wweerree ppeerrffoorrmmeedd aattlleeaasstt ttwwiiccee wwiitthh RRNNAA pprreeppaarreedd ffrroomm iinnddeeppeennddeennttllyy ttrreeaatteedd cceellllccuullttuurreess.. OOnnee rreepprreesseennttaattiivvee rreessuulltt iiss sshhoowwnn..


that occurred without chromium, and this effectwas not altered by either BSO or NAC (Figure 7). Incontrast, BSO enhanced the effect of arsenic on theTCDD response, decreasing luciferase activity from90% to 30% of the levels generated by TCDD alone,whereas NAC had no signi®cant effect. Neither BSOnor NAC treatment changed the lack of effect ofcadmium. These ®ndings support the conclusionthat arsenic, but not chromium, generated an

FFiigguurree 33.. QQuuaannttiittaattiioonn ooff nnoorrtthheerrnn bblloott aannaallyyssiiss.. NNoorrtthheerrnn bblloottsswweerree ppeerrffoorrmmeedd aass ddeessccrriibbeedd iinn FFiigguurree 11 aanndd wweerree qquuaannttiittaatteedd oonn aaSSttoorrmm pphhoosspphhoorriimmaaggeerr.. TThhee rreessuullttiinngg bbaanndd iinntteennssiittiieess wweerree ssttaann--ddaarrddiizzeedd ttoo SSoodd11 ttoo ccoonnttrrooll ffoorr ggeell llooaaddiinngg aanndd aarree pprreesseenntteedd rreellaattiivveettoo DDMMSSOO ccoonnttrroollss.. EEaacchh bbaarr rreepprreesseennttss tthhee mmeeaann�� SSDD ffrroomm aatt lleeaassttttwwoo iinnddeeppeennddeenntt eexxppeerriimmeennttss.. MMeettaall--oonnllyy ttrreeaattmmeennttss wweerree ccoomm--ppaarreedd wwiitthh DDMMSSOO ccoonnttrroollss,, aanndd ssiiggnnii®®ccaanntt ddiiffffeerreenncceess ((PP<< 00..0055)) aarreeddeennootteedd wwiitthh aa pplluuss ssiiggnn.. MMeettaall--TTCCDDDD ccoottrreeaattmmeennttss wweerree ccoommppaarreeddwwiitthh TTCCDDDD--ttrreeaatteedd cceellllss,, aanndd ssiiggnnii®®ccaanntt ddiiffffeerreenncceess ((PP<< 00..0055)) aarreeddeennootteedd wwiitthh aasstteerriisskkss..

FFiigguurree 44.. bb--GGaallaaccttoossiiddaassee aaccttiivviittyy iinn HHeeppaa--11 cceellllss ssttaabbllyy ttrraannss--ffoorrmmeedd wwiitthh ppEEppRREEbbggeeoo.. CCeellllss wweerree iinnccuubbaatteedd ffoorr 66 hh iinn tthheepprreesseennccee ooff iinnccrreeaassiinngg ccoonncceennttrraattiioonnss ooff mmeettaallss aass iinnddiiccaatteedd.. bb--GGaallaaccttoossiiddaassee aaccttiivviittyy wwaass ssttaannddaarrddiizzeedd ttoo ttoottaall pprrootteeiinn mmaassss aanndd iisspprreesseenntteedd aass tthhee ffoolldd iinnccrreeaassee iinn aaccttiivviittyy oovveerr uunnttrreeaatteedd cceellllss.. EEaacchhbbaarr rreepprreesseennttss tthhee mmeeaann�� SSDD ffrroomm tthhrreeee iinnddeeppeennddeenntt eexxppeerriimmeennttss..SSiiggnnii®®ccaanntt ddiiffffeerreenncceess ((PP<< 00..0055)) ffrroomm uunnttrreeaatteedd cceellllss aarree ddeennootteeddwwiitthh aasstteerriisskkss..

FFiigguurree 55.. LLuucciiffeerraassee aaccttiivviittyy iinn HHeeppaa--11 cceellllss ssttaabbllyy ttrraannssffoorrmmeeddwwiitthh ppAAhhRRDDTTKKLLuucc33.. CCeellllss wweerree ttrreeaatteedd wwiitthh mmeettaallss ffoorr 3300 mmiinn aatt tthheeiinnddiiccaatteedd ccoonncceennttrraattiioonn bbeeffoorree ccoottrreeaattmmeenntt wwiitthh 110000 ppMM TTCCDDDD ffoorr88 hh.. LLuucciiffeerraassee aaccttiivviittyy wwaass ssttaannddaarrddiizzeedd ttoo ttoottaall pprrootteeiinn mmaassss,, aannddtthhee rreessuullttss aarree rreeppoorrtteedd rreellaattiivvee ttoo TTCCDDDD aalloonnee.. EEaacchh bbaarr rreepprreesseennttsstthhee mmeeaann�� SSDD ffrroomm tthhrreeee iinnddeeppeennddeenntt eexxppeerriimmeennttss.. MMeettaall--oonnllyyttrreeaattmmeennttss wweerree ccoommppaarreedd wwiitthh DDMMSSOO ccoonnttrroollss,, aanndd ddiiffffeerreenncceesswweerree nnoott ssiiggnnii®®ccaanntt ((PP>> 00..0055)).. MMeettaall--TTCCDDDD ccoottrreeaattmmeennttss wweerreeccoommppaarreedd wwiitthh TTCCDDDD--ttrreeaatteedd cceellllss,, aanndd ssiiggnnii®®ccaanntt ddiiffffeerreenncceess((PP<< 00..0055)) aarree ddeennootteedd wwiitthh aann aasstteerriisskk..

FFiigguurree 66.. EEffffeecctt ooff ttiimmee ooff cchhrroommiiuumm aaddddiittiioonn oonn lluucciiffeerraasseeaaccttiivviittyy iinn HHeeppaa--11 cceellllss ssttaabbllyy ttrraannssffoorrmmeedd wwiitthh ppAAhhRRDDTTKKLLuucc33.. CCeellllsswweerree ttrreeaatteedd wwiitthh 2255 mmMM cchhrroommiiuumm aatt ddiiffffeerreenntt ttiimmeess rraannggiinngg ffrroomm 22hh bbeeffoorree ttoo 22 hh aafftteerr aaddddiittiioonn ooff TTCCDDDD ttoo 11 nnMM.. TThhee cceellllss wweerreeiinnccuubbaatteedd iinn tthhee pprreesseennccee ooff TTCCDDDD ffoorr 88 hh iinn aallll ccaasseess.. LLuucciiffeerraasseeaaccttiivviittyy wwaass ssttaannddaarrddiizzeedd ttoo ttoottaall pprrootteeiinn mmaassss aanndd iiss rreeppoorrtteeddiinn rreellaattiioonn ttoo cceellllss ttrreeaatteedd wwiitthh TTCCDDDD aalloonnee.. EEaacchh bbaarr rreepprreesseennttsstthhee mmeeaann�� SSDD ffrroomm ttwwoo iinnddeeppeennddeenntt eexxppeerriimmeennttss.. MMeettaall--oonnllyyttrreeaattmmeennttss wweerree ccoommppaarreedd wwiitthh DDMMSSOO ccoonnttrroollss aanndd wweerree nnoottssiiggnnii®®ccaannttllyy ddiiffffeerreenntt ((PP>>00..0055)).. MMeettaall--TTCCDDDD ccoottrreeaattmmeennttss wweerreeccoommppaarreedd wwiitthh TTCCDDDD--ttrreeaatteedd cceellllss,, aanndd ssiiggnnii®®ccaanntt ddiiffffeerreenncceess((PP<< 00..0055)) aarree ddeennootteedd wwiitthh aann aasstteerriisskk..


oxidative stress response responsible for alterationsof AHR-dependent gene expression.

TTrraannssccrriippttiioonnaall IInnhhiibbiittiioonn ooff CCyypp11aa11 aannddNNqqoo11 bbyy CChhrroommiiuumm

The observation that chromium can inhibitinducible gene expression by directly interacting

with DNA [reviewed in 26] prompted us to test theeffect of chromium on transcriptional regulationof the genes under study. To test the effect ofchromium on the DNA-binding competence of theAHR, we used electrophoretic mobility shift assaysof nuclear extracts from Hepa-1 cells. TCDD treat-ment resulted in the formation of DNA-bindingcompetent complexes of the AHR with its dimeriza-tion partner the AHR nuclear translocator asdemonstrated by the shift of a 32P-labeled AhREprobe. This effect was not altered by cotreatmentwith different concentrations of chromium (Figure8). In contrast, a positive control with the proteinkinase C inhibitor staurosporine blocked formationof DNA-protein complexes, consistent with pre-

FFiigguurree 77.. EEffffeecctt ooff BBSSOO aanndd NNAACC oonn lluucciiffeerraassee aaccttiivviittyy iinn HHeeppaa--11cceellllss ssttaabbllyy ttrraannssffoorrmmeedd wwiitthh ppAAhhRRDDTTKKLLuucc33.. CCeellllss wweerree ttrreeaatteedd wwiitthh2200 mmMM BBSSOO ffoorr 1122 hh oorr wwiitthh 55 mmMM NNAACC 22 hh bbeeffoorree tthhee aaddddiittiioonn ooffTTCCDDDD ttoo 11 nnMM.. AArrsseenniicc ((55 mmMM)),, ccaaddmmiiuumm ((2255 mmMM)),, oorr cchhrroommiiuumm ((2255mmMM)) wwaass aaddddeedd 3300 mmiinn bbeeffoorree TTCCDDDD ttrreeaattmmeenntt.. TThhee cceellllss wweerreeiinnccuubbaatteedd iinn tthhee pprreesseennccee ooff TTCCDDDD ffoorr 88 hh.. LLuucciiffeerraassee aaccttiivviittyy wwaassssttaannddaarrddiizzeedd ttoo ttoottaall pprrootteeiinn mmaassss aanndd iiss pprreesseenntteedd iinn rreellaattiioonn ttoo cceellllssttrreeaatteedd wwiitthh TTCCDDDD aalloonnee.. EEaacchh bbaarr rreepprreesseennttss tthhee mmeeaann�� SSDD ffrroommttwwoo iinnddeeppeennddeenntt eexxppeerriimmeennttss.. MMeettaall--oonnllyy ttrreeaattmmeennttss wweerree ccoomm--ppaarreedd wwiitthh DDMMSSOO ccoonnttrroollss,, aanndd ddiiffffeerreenncceess wweerree nnoott ssiiggnnii®®ccaanntt((PP>> 00..0055)).. MMeettaall--TTCCDDDD ccoottrreeaattmmeennttss wweerree ccoommppaarreedd wwiitthh TTCCDDDD--ttrreeaatteedd cceellllss,, aanndd ddiiffffeerreenncceess wweerree nnoott ssiiggnnii®®ccaanntt ((PP>> 00..0055))..

FFiigguurree 88.. DDeetteerrmmiinnaattiioonn ooff AAHHRR aaccttiivvaattiioonn bbyy TTCCDDDD iinn tthheepprreesseennccee ooff cchhrroommiiuumm.. NNuucclleeaarr eexxttrraaccttss wweerree pprreeppaarreedd ffrroomm HHeeppaa--11 cceellllss ttrreeaatteedd wwiitthh cchhrroommiiuumm aatt tthhee iinnddiiccaatteedd ccoonncceennttrraattiioonnss ffoorr 3300mmiinn oorr wwiitthh 00..22 mmMM ssttaauurroossppoorriinnee ffoorr 11 hh bbeeffoorree tthhee aaddddiittiioonn ooffTTCCDDDD ttoo 11 nnMM.. CCeellllss wweerree iinnccuubbaatteedd ffoorr 22 hh iinn tthhee pprreesseennccee ooff TTCCDDDDiinn aallll ccaasseess.. NNuucclleeaarr eexxttrraacctt pprrootteeiinnss ffrroomm cceellllss ttrreeaatteedd wwiitthh TTCCDDDDaalloonnee,, TTCCDDDD pplluuss iinnccrreeaassiinngg ccoonncceennttrraattiioonnss ooff cchhrroommiiuumm,, oorr TTCCDDDDpplluuss ssttaauurroossppoorriinnee wweerree iinnccuubbaatteedd wwiitthh aa 3322PP--llaabbeelleedd AAhhRREE pprroobbee ffoorrtthhee eelleeccttrroopphhoorreettiicc mmoobbiilliittyy sshhiifftt aassssaayyss..

FFiigguurree 99.. NNuucclleeaarr rruunn--oonn aannaallyyssiiss ooff cchhrroommiiuumm aanndd TTCCDDDD--ttrreeaatteeddHHeeppaa--11 cceellllss.. CCeellllss wweerree ttrreeaatteedd wwiitthh 00,, 55,, 2255,, oorr 5500 mmMM cchhrroommiiuummffoorr 3300 mmiinn aanndd tthheenn ccoottrreeaatteedd wwiitthh 11 nnMM TTCCDDDD ffoorr 22 hh bbeeffoorreehhaarrvveessttiinngg ffoorr nnuucclleeii pprreeppaarraattiioonn.. ((AA)) RRuunn--oonn aannaallyyssiiss wwaass ppeerrffoorrmmeeddffoouurr ttiimmeess;; oonnee rreepprreesseennttaattiivvee rreessuulltt iiss sshhoowwnn.. ((BB)) HHyybbrriiddiizzaattiioonniinntteennssiittiieess ffoorr CCyypp11aa11 wweerree qquuaannttii®®eedd aanndd aarree sshhoowwnn rreellaattiivvee ttoo tthhee

iinntteennssiittyy ooff tthhee aaccttiinn mmRRNNAA hhyybbrriiddiizzaattiioonn ffoorr cceellllss rreecceeiivviinngg tthhee ssaammeettrreeaattmmeennttss.. EEaacchh bbaarr rreepprreesseennttss tthhee mmeeaann�� SSDD ffrroomm ffoouurr iinnddeeppeenn--ddeenntt eexxppeerriimmeennttss.. MMeettaall--TTCCDDDD ccoottrreeaattmmeennttss wweerree ccoommppaarreedd wwiitthhTTCCDDDD--ttrreeaatteedd cceellllss,, aanndd ssiiggnnii®®ccaanntt ddiiffffeerreenncceess ((PP<< 00..0055)) aarreeddeennootteedd wwiitthh aasstteerriisskkss..


vious observations [40]. These data suggest thatchromium does not impair AHR activation.

The ability of chromium to interact with DNA andselectively inhibit the expression of inducible geneswas consistent with the effects observed in theNorthern blot analysis of TCDD-responsive genes.Hence, the decrease in Cyp1a1 and Nqo1 mRNAlevels could be due to changes in mRNA stability orin transcription rates. To determine RNA stability,we measured decay rates of Cyp1a1, Nqo1, and Sod1mRNA in TCDD-treated cells after blocking furthertranscription initiation with actinomycin D. TheCyp1a1 mRNA half-life was not signi®cantlyaffected by treatment with 25 mM chromium andneither were the half-lives of Nqo1 and Sod1 (datanot shown). However, transcriptional run-on ana-lyses showed that Cyp1a1 transcription was inducedby 1 nM TCDD and that induction was signi®cantlydecreased approximately to 50% by 25 or 50 mMchromium (Figure 9A and B). These data suggestthat chromium acts by blocking gene transcriptionand not through direct inactivation of the AHR.

DDIISSCCUUSSSSIIOONN

In this study we report that treatment withcadmium, chromium, or arsenic can differentiallyalter the TCDD-inducible expression of two genes,Cyp1a1 and Nqo1, involved in phase I and phase IIdetoxi®cation, respectively. Chromium decreasedTCDD-inducible expression of both genes. Theeffects of chromium were not modulated by treat-ment with BSO or NAC, and chromium did notinduce expression of Gclr, Ho, or an EpRE b-galactosidase reporter. Thus, the mechanism ofaction of chromium on AHR-dependent induciblegene expression does not appear to be related tooxidative stress or to electrophile-driven responses.Rather, our results are consistent with observationsof chromium-mediated transcriptional inhibitionreported for other inducible genes engendered bydirect interactions of chromium with DNA or DNA-associated proteins [49±51]. It seems plausible thatdisruption of normal chromatin architecture wouldexplain the inhibitory results of chromium in ourstudies because DNA conformation changes in theCyp1a1 gene promoter may in part drive its AHR-dependent induction [52].

The selective action of chromium on induciblepromoters may result from a more open chromatinstructure that offers a better target for chromiumbinding than the more closed chromatin structureof constitutive promoters [53]. Chromium-inducedDNA-protein crosslinks are found preferentially inassociation with the nuclear matrix [54]. Becausemany transcriptional proteins associate with thenuclear matrix, crosslinks between DNA and theseproteins may block their function. For example,intracellular reduction of hexavalent chromium,although stimulating DNA binding of NF-kB, did

not lead to an increase in NF-kB±dependent geneexpression [55]. This discordance between DNAbinding of the transcription factor and gene induc-tion was due to the ability of chromium to block thebinding of the p65 NF-kB subunit to the cyclicAMP±responsive element-binding protein (p300), atranscriptional coactivator with intrinsic histoneacetyl transferase activity [55].

In contrast to the inhibition observed withchromium, both cadmium and arsenic moderatelyenhanced the effect of TCDD on Nqo1 mRNA levels.The fact that they did not in¯uence TCDD induci-bility of Cyp1a1 suggests that they act through amechanism distinct to phase II gene promoters. Thisobservation is consistent with the ability of cad-mium and arsenic to induce electrophile-drivenresponses in our experiments (Figure 4) and asdemonstrated by others for the Nqo1 [56,57] andGsta1 genes [36,58,59]. We observed only a limitedenhancement of TCDD inducibility in our experi-ments, possibly because electrophile responsesdriven by these metals were masked by the highpotency of TCDD as a Nqo1 inducer through theAhRE in the promoter of this gene [1] at theconcentrations used in our experiments. If this weretrue, then the effect of these metals on phase I tophase II balance may be more dramatic at lowerconcentrations of TCDD than might be expectedfrom environmental exposures.

Our results suggest that combined exposure tometals may alter phase I metabolism and phase IIconjugation of polycyclic aromatic hydrocarbons.The potential of cadmium, chromium, and arsenicto modify downstream biologic effects, includingxenobiotic metabolism resulting from phase I orphase II gene expression, however, has not beenwell de®ned and is an area of major interest.Cadmium has been shown to both decrease andincrease P450-dependent metabolism [60±62], withresponses differing in a dose- and sex-dependentmanner. Treatment of cells in culture with BaP andcadmium resulted in additive mutagenic effects[63]. Cadmium has been recently proposed toenhance mutagenicity of Bap-diol epoxide byfavoring formation of N2-guanine adducts ratherthan N7-guanine adducts [64]. Chromium hasshown potentially inhibitory effects on AHR ligandinducible responses [65,66] and, in one recentreport, inhibited the formation of BaP-inducedmutations when cells were treated simultaneouslybut not when chromium was added before or afterBaP [47]. Antioxidants blocked the inhibitory effectof chromium, suggesting that chromium-mediatedoxidative stress was responsible for decreased yieldof BaP-induced mutations. In vitro, arsenic inhibitsligand-inducible CYP1A1 and CYP1A2 activation[67], but in vivo mixed results have been reported,perhaps because of tissue- and species-speci®cdifferences [68,69]. In addition, arsenic and BaP


appear to have a positive interaction on lungtumorigenesis in animals [70,71].

The widely different biologic responses to metaland polycyclic aromatic hydrocarbon coexposuresdescribed in the existing literature highlight theneed for additional studies to understand theparameters driving these responses. We have shownthat metals can modulate biologic responses to AHRligands at the gene expression level. The resultssuggest that metal and halogenated aromatic hydro-carbon coexposures represent signi®cantly differenthealth risks than would be anticipated based onevaluation of the toxic mechanisms of each class ofcompound evaluated singly. By extension, it islikely that the same argument applies to mixturesof metals and polycyclic aromatic hydrocarbonligands of the AHR.

AACCKKNNOOWWLLEEDDGGMMEENNTTSS

We thank M. Z. Dieter for the Hepa-1 cell linestably transformed with pEpREbgeo. We also thankP. Gartside and M. Medvedovic for assistance withstatistical analysis. This work was supported byNIEHS grants ES06273 and P30 ES06096 and by aUS EPA STAR predoctoral fellowship to AM.

RREEFFEERREENNCCEESS

1. Nebert DW, Puga A, Vasiliou V. Role of the Ah receptor andthe dioxin-inducible [Ah] gene battery in toxicity, cancer,and signal transduction. Ann NY Acad Sci 1993;685:624±640.

2. Poland AP, Glover E, Robinson JR, Nebert DW. Geneticexpression of aryl hydrocarbon hydroxylase activity. Induc-tion of monooxygenase activities and cytochrome P1-450formation by 2,3,7,8- tetrachlorodibenzo-p-dioxin in micegenetically ``nonresponsive'' to other aromatic hydrocar-bons. J Biol Chem 1974;249:5599±5606.

3. Nebert DW. The Ah locus: Genetic differences in toxicity,cancer, mutation, and birth defects. Crit Rev Toxicol 1989;20:153±174.

4. Fernandez-Salguero PM, Hilbert DM, Rudikoff S, WardJM, Gonzalez FJ. Aryl-hydrocarbon receptor±de®cientmice are resistant to 2,3,7,8-tetrachlorodibenzo-p-dioxin-induced toxicity. Toxicol Appl Pharmacol 1996;140:173±179.

5. Shimizu Y, Nakatsuru Y, Ichinose M, et al. Benzo[a]pyrenecarcinogenicity is lost in mice lacking the aryl hydrocarbonreceptor. Proc Natl Acad Sci USA 2000;97:779±782.

6. Denison MS, Vella LM, Okey AB. Structure and function ofthe Ah receptor: Sulfhydryl groups required for binding of2,3,7,8-tetrachlorodibenzo-p-dioxin to cytosolic receptorfrom rodent livers. Arch Biochem Biophys 1987;252:388±395.

7. Kester JE, Gasiewicz TA. Characterization of the in vitrostability of the rat hepatic receptor for 2,3,7,8-tetrachloro-dibenzo-p-dioxin (TCDD). Arch Biochem Biophys 1987;252:606±625.

8. Henry EC, Kester JE, Gasiewicz TA. Effects of SH-modifyingreagents on the rat hepatic Ah receptor: Inhibition of ligandbinding and transformation, and disruption of the ligand±receptor complex. Biochim Biophys Acta 1988;964:361±376.

9. Ireland RC, Li SY, Dougherty JJ. The DNA binding of puri®edAh receptor heterodimer is regulated by redox conditions.Arch Biochem Biophys 1995;319:470±480.

10. Xu C, Siu CS, Pasco DS. DNA binding activity of the arylhydrocarbon receptor is sensitive to redox changes in intactcells. Arch Biochem Biophys 1998;358:149±156.

11. Barker CW, Fagan JB, Pasco DS. Down-regulation ofP4501A1 and P4501A2 mRNA expression in isolatedhepatocytes by oxidative stress. J Biol Chem 1994;269:3985±3990.

12. Xu C, Pasco DS. Suppression of CYP1A1 transcription byH2O2 is mediated by xenobiotic-response element. ArchBiochem Biophys 1998;356:142±150.

13. Morel Y, Barouki R. Down-regulation of cytochrome P4501A1 gene promoter by oxidative stress. Critical contributionof nuclear factor 1. J Biol Chem 1998;273:26969±26976.

14. Morel Y, Mermod N, Barouki R. An autoregulatory loopcontrolling CYP1A1 gene expression: Role of H2O2 and NF1.Mol Cell Biol 1999;19:6825±6832.

15. Stohs SJ, Bagchi D. Oxidative mechanisms in the toxicity ofmetal ions. Free Radic Biol Med 1995;18:321±336.

16. Nieboer E, Fletcher GG. Determinants of reactivity in metaltoxicology. In: Chang LW, editor. Toxicology of metals. BocaRaton: CRC Press; 1996. p 113±132.

17. Applegate LA, Luscher P, Tyrrell RM. Induction of hemeoxygenase: A general response to oxidant stress in culturedmammalian cells. Cancer Res 1991;51:974±978.

18. Ramos O, Carrizales L, Yanez L, et al. Arsenic increased lipidperoxidation in rat tissues by a mechanism independentof glutathione levels. Environ Health Perspect 1995;103(suppl 1):85±88.

19. Bagchi D, Bagchi M, Hassoun EA, Stohs SJ. Cadmium-induced excretion of urinary lipid metabolites, DNAdamage, glutathione depletion, and hepatic lipid peroxida-tion in Sprague-Dawley rats. Biol Trace Elem Res 1996;52:143±154.

20. Flora SJ. Arsenic-induced oxidative stress and its reversibilityfollowing combined administration of N-acetylcysteine andmeso 2,3-dimercaptosuccinic acid in rats. Clin Exp Pharma-col Physiol 1999;26:865±869.

21. Shaikh ZA, Vu TT, Zaman K. Oxidative stress as a mechanismof chronic cadmium-induced hepatotoxicity and renaltoxicity and protection by antioxidants. Toxicol ApplPharmacol 1999;154:256±263.

22. Hassoun EA, Stohs SJ. Cadmium-induced production ofsuperoxide anion and nitric oxide, DNA single strand breaksand lactate dehydrogenase leakage in J774A.1 cell cultures.Toxicology 1996;112:219±226.

23. Yang CF, Shen HM, Shen Y, Zhuang ZX, Ong CN.Cadmium-induced oxidative cellular damage in human fetallung ®broblasts (MRC-5 cells). Environ Health Perspect1997;105:712±716.

24. Chen YC, Lin-Shiau SY, Lin JK. Involvement of reactiveoxygen species and caspase 3 activation in arsenite-inducedapoptosis. J Cell Physiol 1998;177:324±333.

25. Barchowsky A, Roussel RR, Klei LR, et al. Low levels ofarsenic trioxide stimulate proliferative signals in primaryvascular cells without activating stress effector pathways.Toxicol Appl Pharmacol 1999;159:65±75.

26. Klein CB. Carcinogenicity and genotoxicity of chromium. In:Chang LW, editor. Toxicology of metals. Boca Raton: CRCPress; 1996. p 205±219.

27. Kortenkamp A, Casadevall M, Da Cruz Fresco P. Thereductive conversion of the carcinogen chromium (VI) andits role in the formation of DNA lesions. Ann Clin Lab Sci1996;26:160±175.

28. Tsou TC, Chen CL, Liu TY, Yang JL. Induction of 8-hydroxydeoxyguanosine in DNA by chromium(III) plushydrogen peroxide and its prevention by scavengers.Carcinogenesis 1996;17:103±108.

29. Sugden KD, Wetterhahn KE. Direct and hydrogen peroxide-induced chromium(V) oxidation of deoxyribose in single-stranded and double-stranded calf thymus DNA. Chem ResToxicol 1997;10:1397±1406.


30. Tsou TC, Lai HJ, Yang JL. Effects of mannitol or catalase onthe generation of reactive oxygen species leading to DNAdamage by chromium(VI) reduction with ascorbate. ChemRes Toxicol 1999;12:1002±1009.

31. Gerhardsson L, Brune D, Nordberg GF, Wester PO. Multi-elemental assay of tissues of deceased smelter workers andcontrols. Sci Total Environ 1988;74:97±110.

32. Raithel HJ, Ebner G, Schaller KH, Schellmann B, Valentin H.Problems in establishing norm values for nickel andchromium concentrations in human pulmonary tissue. AmJ Intern Med 1987;12:55±70.

33. Clewell HJ, Gentry PR, Barton HA, Shipp AM, Yager JW,Andersen ME. Requirements for a biologically realisticcancer risk assessment for inorganic arsenic. Int J Toxicol1999;18:131±147.

34. ATSDR. ATSDR/EPA priority list for 1999. Atlanta:ATSDR; 1999. Available from: http://www.atsdr.cdc.gov/99list.html.

35. Chang CY, Puga A. Constitutive activation of the aromatichydrocarbon receptor. Mol Cell Biol 1998;18:525±535.

36. Prestera T, Holtzclaw WD, Zhang Y, Talalay P. Chemical andmolecular regulation of enzymes that detoxify carcinogens.Proc Natl Acad Sci USA 1993;90:2965±2969.

37. Dalton T, Palmiter RD, Andrews GK. Transcriptional induc-tion of the mouse metallothionein-I gene in hydrogenperoxide-treated Hepa cells involves a composite major latetranscription factor/antioxidant response element and metalresponse promoter elements. Nucleic Acids Res 1994;22:5016±5023.

38. Vasilou V, Theurer MJ, Puga A, Reuter SF, Nebert DW.Mouse dioxin-inducible NAD(P)H:menandione oxidoreduc-tase: NMO1 cDNA sequence and genetic differences inmRNA levels. Pharmacogenetics 1994;4:341±348.

39. Dalton T, Pazdernik TL, Wagner J, Samson F, Andrews GK.Temporal spatial patterns of expression of metallothionein-Iand -III and other stress related genes in rat brain after kainicacid±induced seizures. Neurochem Int 1995;27:59±71.

40. Carrier F, Owens RA, Nebert DW, Puga A. Dioxin-dependent activation of murine Cyp1a-1 gene transcriptionrequires protein kinase C±dependent phosphorylation. MolCell Biol 1992;12:1856±1863.

41. Greenberg ME, Bender TP. Identi®cation of newly tran-scribed RNA. In: Ausubel FM, et al., editors. Currentprotocols in molecular biology. New York: John Wiley &Sons; 1997. p 4.10.1±4.10.11.

42. RayChaudhuri B, Nebert DW, Puga A. The murine Cyp1a-1gene negatively regulates its own transcription and thatof other members of the aromatic hydrocarbon-res-ponsive [Ah] gene battery. Mol Endocrinol 1990;4:1773±1781.

43. Borchers MT, Wert SE, Leikauf GD. Acrolein-inducedMUC5ac expression in rat airways. Am J Physiol 1998;274:L573±L581.

44. Ye J, Zhang X, Young HA, Mao Y, Shi X. Chromium(VI)-induced nuclear factor-kappa B activation in intact cells viafree radical reactions. Carcinogenesis 1995;16:2401±2405.

45. Kim G, Yurkow EJ. Chromium induces a persistentactivation of mitogen-activated protein kinases by aredox-sensitive mechanism in H4 rat hepatoma cells. CancerRes 1996;56:2045±2051.

46. Chen F, Ye J, Zhang X, Rojanasakul Y, Shi X. One-electronreduction of chromium(VI) by alpha-lipoic acid and relatedhydroxyl radical generation, dG hydroxylation and nucleartranscription factor-kappaB activation. Arch Biochem Bio-phys 1997;338:165±172.

47. Tesfai Y, Davis D, Reinhold D. Chromium can reduce themutagenic effects of benzo[a]pyrene diolepoxide in normalhuman ®broblasts via an oxidative stress mechanism. MutatRes 1998;416:159±168.

48. Shertzer HG, Vasiliou V, Liu RM, Tabor MW, Nebert DW.Enzyme induction by L-buthionine (S,R)-sulfoximine in

cultured mouse hepatoma cells. Chem Res Toxicol 1995;8:431±436.

49. Hamilton JW, Wetterhahn KE. Differential effects ofchromium(VI) on constitutive and inducible gene expressionin chick embryo liver in vivo and correlation with chro-mium(VI)-induced DNA damage. Mol Carcinog 1989;2:274±286.

50. Alcedo JA, Misra M, Hamilton JW, Wetterhahn KE. Thegenotoxic carcinogen chromium(VI) alters the metal-indu-cible expression but not the basal expression of themetallothionein gene in vivo. Carcinogenesis 1994;15:1089±1092.

51. McCaffrey J, Wolf CM, Hamilton JW. Effects of thegenotoxic carcinogen chromium(VI) on basal and hor-mone-inducible phosphoenolpyruvate carboxykinase geneexpression in vivo: Correlation with glucocorticoid- anddevelopmentally regulated expression. Mol Carcinog 1994;10:189±198.

52. Okino ST, Whitlock JP Jr. Dioxin induces localized, gradedchanges in chromatin structure: implications for Cyp1A1gene transcription. Mol Cell Biol 1995;15:3714±3721.

53. Manning FC, Xu J, Patierno SR. Transcriptional inhibition bycarcinogenic chromate: Relationship to DNA damage. MolCarcinog 1992;6:270±279.

54. Xu J, Manning FC, Patierno SR. Preferential formation andrepair of chromium-induced DNA adducts and DNA±proteincrosslinks in nuclear matrix DNA. Carcinogenesis 1994;15:1443±1450.

55. Shumilla JA, Broderick RJ, Wang Y, Barchowsky A.Chromium(VI) inhibits the transcriptional activity of nuclearfactor-kappaB by decreasing the interaction of p65 withcAMP-responsive element-binding protein-binding protein.J Biol Chem 1999;274:36207±36212.

56. Falkner KC, McCallum GP, Bend JR. Effects of arsenitetreatment on NAD(P)H:quinone acceptor oxidoreductaseactivity in liver, lung, kidney, and heart of the rat.Comparison to induction by the polyaromatic hydrocarbon,beta-naphtho¯avone. Drug Metab Dispos 1993;21:334±337.

57. Jaiswal AK. Antioxidant response element. Biochem Phar-macol 1994;48:439±444.

58. Prestera T, Talalay P. Electrophile and antioxidant regulationof enzymes that detoxify carcinogens. Proc Natl Acad SciUSA 1995;92:8965±8969.

59. Talalay P, Fahey JW, Holtzclaw WD, Prestera T, Zhang Y.Chemoprotection against cancer by phase 2 enzymeinduction. Toxicol Lett 1995;82±83:173±179.

60. Furst A, Mogannam JD. Inhibition of aryl hydrocarbonhydroxylase by cadmium ion. Proc West Pharmacol Soc1975;18:326±329.

61. Fair PH. Interaction of benzo(a)pyrene and cadmium onGSH-S-transferase and benzo(a)pyrene hydroxylase in theblack sea bass Centropristis striata. Arch Environ ContamToxicol 1986;15:257±263.

62. George SG, Young P. The time course of effects of cadmiumand 3-methylcholanthrene on activities of enzymes ofxenobiotic metabolism and metallothionein levels in theplaice, Pleuronectes platessa. Comp Biochem Physiol C1986;83:37±44.

63. Ochi T, Ohsawa M. Induction of 6-thioguanine-resistantmutants and single-strand scission of DNA by cadmiumchloride in cultured Chinese hamster cells. Mutat Res1983;111:69±78.

64. Prakash AS, Tran H, Peng C, Koyalamudi SR, Dameron CT.Kinetics of DNA alkylation, depurination and hydrolysis ofanti diol epoxide of benzo(a)pyrene and the effect ofcadmium on DNA alkylation. Chem Biol Interact. 2000;125:133±150.

65. Dixon JR, Lowe DB, Richards DE, Cralley LJ, Stokinger HE.The role of trace metals in chemical carcinogenesis:Asbestos cancers. Cancer Res 1970;30:1068±1074.


66. Williams SJ, Karis MA, Menzel DB. Interactions of heavymetals with the pulmonary metabolism of [3H]benzo[a]pyr-ene. Environ Res 1984;34:212±226.

67. Jacobs J, Roussel R, Roberts M, et al. Effect of arsenite oninduction of CYP1A and CYP2H in primary cultures of chickhepatocytes. Toxicol Appl Pharmacol 1998;150:376±382.

68. Falkner KC, McCallum GP, Cherian MG, Bend JR. Effects ofacute sodium arsenite administration on the pulmonarychemical metabolizing enzymes, cytochrome P-450 mono-oxygenase, NAD(P)H:quinone acceptor oxidoreductase andglutathione S-transferase in guinea pig: Comparison witheffects in liver and kidney. Chem Biol Interact 1993;86:51±68.

69. Albores A, Sinal CJ, Cherian MG, Bend JR. Selective increaseof rat lung cytochrome P450 1A1 dependent monooxy-genase activity after acute sodium arsenite administration.Can J Physiol Pharmacol 1995;73:153±158.

70. Ishinishi N, Kodama Y, Nobutomo K, Hisanaga A. Pre-liminary experimental study on carcinogenicity of arsenictrioxide in rat lung. Environ Health Perspect 1977;19:191±196.

71. Pershagen G, Nordberg G, Bjorklund NE. Carcinomas of therespiratory tract in hamsters given arsenic trioxide and/orbenzo[a]pyrene by the pulmonary route. Environ Res1984;34:227±241.


Disruption of dioxin-inducible phase I and phase II gene expression patterns by cadmium, chromium,...

Documents

Transcript of Disruption of dioxin-inducible phase I and phase II gene expression patterns by cadmium, chromium,...