Mutation Research 596 (2006) 43–56

7H-Dibenzo[c,g]carbazole and 5,9-dimethyldibenzo[c,g]carbazoleexert multiple toxic events contributing to tumor promotion

in rat liver epithelial ‘stem-like’ cells

Jan Vondracek a,b, Lenka Svihalkova-Sindlerova a, Katerina Pencıkova b, Pavel Krcmar b,Zdenek Andrysık a,b, Katerina Chramostova a, Sona Marvanova b, Zuzana Valovicova c,

Alois Kozubık a, Alena Gabelova c, Miroslav Machala b,∗a Laboratory of Cytokinetics, Institute of Biophysics, Academy of Sciences of the Czech Republic, Kralovopolska 135,

612 65 Brno, Czech Republicb Department of Chemistry and Toxicology, Veterinary Research Institute, Hudcova 70, 621 32 Brno, Czech Republic

c Department of Mutagenesis and Carcinogenesis, Cancer Research Institute, Slovak Academy of Sciences,Vlarska 7, 833 91 Bratislava, Slovak Republic

Received 30 June 2005; received in revised form 22 September 2005; accepted 30 November 2005Available online 9 January 2006

Abstract

Immature liver progenitor cells have been suggested to be an important target of hepatotoxins and hepatocarcinogens. The goal

of the present study was to assess the impact of 7H-dibenzo[c,g]carbazole (DBC) and its tissue-specific carcinogenic N-methyl(N-MeDBC) and 5,9-dimethyl (DiMeDBC) derivatives on rat liver epithelial WB-F344 cells, in vitro model of liver progenitor cells.We investigated the cellular events associated with both tumor initiation and promotion, such as activation of aryl hydrocarbonreceptor (AhR), changes in expression of enzymes involved in metabolic activation of DBC and its derivatives, effects on cellcycle, cell proliferation/apoptosis and inhibition of gap junctional intercellular communication (GJIC). N-MeDBC, a tissue-specificsarcomagen, was only a weak inhibitor of GJIC or inducer of AhR-mediated activity, and it did not affect either cell proliferationor apoptosis. DBC was efficient GJIC inhibitor, while DiMeDBC manifested the strongest AhR inducing activity. Accordingly,DiMeDBC was also the most potent inducer of cytochrome P450 1A1 (CYP1A1) and CYP1A2 expression among the three com-pounds tested. Both DBC and DiMeDBC induced expression of CYP1B1 and aldo-keto reductase 1C9 (AKR1C9). N-MeDBC failedto significantly upregulate CYP1A1/2 and it only moderately increased CYP1B1 or AKR1C9. Only the potent liver carcinogens,DBC and DiMeDBC, caused a significant increase of p53 phosphorylation at Ser15, an increased accumulation of cells in S-phaseand apoptosis at micromolar concentrations. In addition, DiMeDBC was found to stimulate cell proliferation of contact-inhibitedWB-F344 cells at 1 �M concentration, which is a mode of action that might further contribute to its hepatocarcinogenicity. Thepresent data seem to suggest that the AhR activation, induction of enzymes involved in metabolic activation, inhibition of GJIC orstimulation of cell proliferation might all contribute to the hepatocarcinogenic effects of DBC and DiMeDBC.© 2005 Elsevier B.V. All rights reserved.

Keywords: Dibenzocarbazoles; Gap junctional intercellular communication (GJIC); AhR activation; Cell proliferation; Apoptosis; CytochromesP450

∗ Corresponding author. Tel.: +420 533331813; fax: +420 541211229.E-mail address: machala@vri.cz (M. Machala).

0027-5107/$ – see front matter © 2005 Elsevier B.V. All rights reserved.doi:10.1016/j.mrfmmm.2005.11.005

44 J. Vondracek et al. / Mutation Research 596 (2006) 43–56

1. Introduction

7H-dibenzo[c,g]carbazole (DBC), an ubiquitousenvironmental pollutant, is a common component of avariety of complex mixtures derived from the incom-plete combustion of organic matter, such as cigarettesmoke, soot, tars and diesel engine exhaust (reviewedin [1]). DBC is a potent multi-species and multi-site car-cinogen in in vivo animal studies with both local andsystemic effects [1,2]. Based on these facts, the Inter-national Agency for Research against Cancer (IARC)[3,4] listed DBC as a possible human carcinogen (Group2B), although the extent of human exposure is unknown[5]. DBC is a member of an important class of environ-mental compounds known as N-heterocyclic aromaticcompounds (NHAs). The chemical structure of DBCshares the features of polycyclic aromatic hydrocarbons(PAHs), such as the aromatic skeleton with a typicalbay region, while the pyrrolic NH group has a chemi-cal behavior similar to that of the arylamides (Fig. 1).Although NHAs represent a relatively minor fractionof the crude organic mixtures of environmental pol-lutants, they may pose a serious carcinogenic risk tohuman due their intrinsic biological activity. DBC, forexample, was shown to be a more potent pulmonarytumorigen in hamsters than benzo[a]pyrene [6]. Thisagent is unusual in its carcinogenic response; it canproduce significant numbers of skin and liver tumorswhen applied topically to the back skin of mice [7].On the other hand, its methyl derivatives, N-methyl-

in the liver, but it is devoid of any activity in the skin[10–12].

Like many other chemical carcinogens, DBC requiresmetabolic activation to electrophilic species beforethey can interact with DNA and other macromoleculesand exert their mutagenic and carcinogenic effects.It has been suggested that formation of bay regiondihydrodiol epoxides and o-quinones are the princi-ple metabolic pathways for DBC [1,13], while for-mation of radical cation is not the major route ofDBC metabolic activation [14]. Both types of metabo-lites, dihydrodiol epoxides and o-quinones, may con-tribute to PAH-induced carcinogenesis and a relativeimportance of each pathway depends on tissue typeand the level and inducibility of activation enzyme(s)[13,15]. Cytochromes P450 1A1 (CYP1A1), CYP1A2and CYP1B1 are key enzymes involved in formation ofPAH dihydrodiol epoxides, whereas dihydrodiol dehy-drogenases (aldo-keto reductases—AKR) are responsi-ble for formation of o-quinones and induction of oxida-tive stress and reactive semiquinones [13,15]. The roleof CYP1 enzymes in vivo is more complex, as theymight contribute to both bioactivation and detoxifica-tion of PAHs and related compounds, depending on celltype- and tissue-specific context [16]. Although the rela-tive importance of CYP1A1 and CYP1A2 for metabolicactivation of DBC and its derivatives has been studiedin some detail [17–19], the information on inducibilityof individual enzymes in specific target cells or tissuesis mostly missing.

DBC (N-MeDBC) and 5,9-dimethyl-DBC (DiMeDBC)manifested specific tropism for the skin and liver, respec-tively. N-MeDBC produces DNA adducts, tumors andmutations mainly in the skin and lacks hepatocarcino-genic potential [8,9], whereas the strict hepatocarcino-gen DiMeDBC produces tumors, adducts and mutations

Fig. 1. Chemical structures.

Upregulation of phases I and II xenobiotic-metabolizing enzymes, which are involved in biotrans-formation of PAHs and NHAs, is largely mediated viabinding of ligands to the aryl hydrocarbon receptor(AhR), and it is currently the most clearly understoodprocess of AhR signaling. Although the capacity of DBCand its derivatives to activate AhR could significantlyaffect their genotoxicity, there is no comprehensive infor-mation about the AhR activation by DBC and its deriva-tives. Their potencies to activate AhR might also dif-fer in various liver cell populations, as their genotoxiceffects are also different, e.g. in parenchymal and non-parenchymal liver cells [20]. Nevertheless, apart fromregulation of expression of metabolizing enzymes, theAhR plays also an important role in cell cycle regulation,mitogenesis, oxidative stress and apoptosis [21–23]; pro-cesses that might significantly affect the carcinogenicproperties of genotoxins.

There is currently little information on non-genotoxicmodes of action that might contribute to carcinogenic-ity of DBC and its derivatives, including their potentialtumor promoting effects, such as perturbation of tissue

J. Vondracek et al. / Mutation Research 596 (2006) 43–56 45

homeostasis due to disruption of cell-to-cell commu-nication or proliferation/apoptosis balance. Inhibitionof gap-junctional intercellular communication (GJIC)plays an important role in these mechanisms [24] andin vitro determination of this effect has been found asa suitable tool for detection of many tumor promotersand carcinogens [24,25]. A number of low-molecular-weight PAHs are relatively efficient transient inhibitorsof GJIC, suggesting that inhibition of GJIC could be oneof modes of action they exert in target tissues [26,27].The induction of cell proliferation in contact-inhibitedrat liver epithelial cells after exposure to a number ofPAHs is another mechanism of action of PAHs possiblyassociated with tumor promotion, which has been doc-umented recently [28]. Stimulation of cell proliferationhas been shown to be a critical event for expression ofmutations in the mouse liver after DBC and DiMeDBCexposure [29–31]. The severe hepatocellular toxicityobserved in livers taken from DBC-treated mice [1] isfollowed by the liver regeneration [11], suggesting thatthe extensive apoptosis induced by genotoxic DBC inliver cells [32,33] can lead to regenerative cell prolifera-tion. The hepatotoxicity and resulting liver injury mightalso lead to necrosis associated with a release of fur-ther mediators, such as inflammatory cytokines, whichcould also significantly contribute to proliferation of spe-cific liver cell populations, including oval cells [34,35].Therefore, the induction of cell proliferation in liver pro-genitor cells observed in vitro after exposure to varioustypes of AhR ligands [28,36–38] could further contributetTceto

irtOhfGpairtssl

uate effects of hepatocarcinogenic dibenzocarbazoles,DBC and DiMeDBC, and sarcomagenic DBC derivative,N-MeDBC, on WB-F344 cells, focusing principally onthe effects that might promote the genotoxic potential ofthese compounds. Therefore, the impact of these agentson GJIC, induction of AhR-mediated activity, expres-sion of enzymes involved in their metabolic activation,or on cell proliferation, cell cycle and apoptosis, wereinvestigated in contact-inhibited WB-F344 cells.

2. Materials and methods

2.1. Chemicals

DBC (CAS No. 194-59-2) and its methylated derivatives(Fig. 1) were kindly provided by Dr. F. Perin, Institute Curie,France. 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD, CASNo. 1746-01-6) was from Cambridge Isotope Laboratories(Andover, MA). Antibodies to connexin43 and Ser-15phosphorylated p53 tumor suppressor were obtained fromSigma–Aldrich (Prague, Czech Republic) and Cell SignalingTechnology (Andover, MA), respectively. Horseradishperoxidase-conjugated anti-immunoglobulins were fromSigma–Aldrich; polyvinylidene difluoride (PVDF) mem-brane Hybond-P, and chemiluminescence detection reagents(ECL + Plus) were purchased from Ammersham (Aylesbury,UK). 4′,6-Diamidino-2-phenylindole dihydrochloride (DAPI)was purchased from Fluka (Buchs, Switzerland). All otherchemicals were provided by Sigma–Aldrich.

2.2. Cells

o proliferative effects of DBC and DiMeDBC in liver.he AhR-mediated activity of DBC and its derivativesould thus play significant roles both in regulation ofnzymes participating in formation of ultimate geno-oxic, apoptosis-inducing metabolites and in modulationf cell proliferation.

The rat liver epithelial stem-like WB-F344 cell line,solated from the liver of an adult male Fischer 344at, is considered to be an in vitro model of pluripo-ent oval cells, presumed liver progenitor cells [39].ur recent studies have shown that different PAHs oreterocyclic aromatic compounds are inducing multi-aceted responses in these cells, including inhibition ofJIC, stimulation of cell proliferation or induction ofrogrammed cell death [27,28,40], which makes themn interesting model for studies on cellular processesnvolved in carcinogenesis. The liver oval cells can giveise to both hepatocytes and biliary epithelial cells, andhey might play a significant role in hepatocarcinogene-is [41,42]. Therefore, these cells could represent a pos-ible target for DBC and DiMeDBC, both being potentiver carcinogens. The goal of this study was to eval-

WB-F344 rat liver epithelial cells were grown in mod-ified Eagle’s minimum essential medium (Sigma–Aldrich)with 50% increased concentrations of essential and non-essential amino acids, and supplemented with sodium pyru-vate (110 mg/l), 10 mM HEPES and 5% heat-inactivated fetalbovine serum (Sigma–Aldrich). Only the cells at passages15–22 were used throughout the study. The rat hepatomaH4IIEGud.Luc1 cells, licensed from BioDetection Systems(Amsterdam, The Netherlands), were grown in the alpha mod-ification of minimal essential medium (Sigma–Aldrich), sup-plemented with 10% of heat-inactivated fetal bovine serum.Cells were incubated in a humidified atmosphere of 5% CO2

at 37 ◦C. Cells were routinely maintained in 75 cm2 flasks andsubcultured twice a week.

2.3. Detection of AhR-mediated activity

The rat hepatoma H4IIEGud.Luc1.1 cell line, stably trans-fected with a luciferase reporter gene under the control ofdioxin responsive elements, was used to detect for AhR-mediated activity in the DR-CALUXTM assay [43,44]. Theassays were performed in 96-well cell culture plates. Briefly,24 h after seeding at split ratio 1:2, cells (at 90–100%

46 J. Vondracek et al. / Mutation Research 596 (2006) 43–56

confluency) were exposed to the tested or reference compounds(TCDD) dissolved in dimethyl sulfoxide (DMSO). Maximumconcentrations of DMSO did not exceed 0.4% (v/v). Follow-ing 6 or 24-h exposure, the medium was aspirated, cells werewashed with phosphate-buffered saline (PBS) and luciferasewas extracted with the low salt lysis buffer (10 mM Tris, 2 mMDTT, 2 mM 1,2-diamin cyclic hexane-N,N,N′,N′-tetraaceticacid, pH 7.8). The plates were frozen at −80 ◦C and luciferaseexpression was then measured on a microplate luminometerusing the Luciferase Monitoring Kit (Labsystems, Helsinki,Finland).

2.4. Real-time RT-PCR

The levels of CYP1A1, CYP1A2, CYP1B1 and AKR1C9mRNAs were determined by real-time RT-PCR. The sequencesof primers and probes are listed in Table 1. The primers weredesigned on the exon junction for amplification of cDNA only.Total RNA was isolated from cells using the RNeasy MiniKit (Qiagen, Valencia, CA), including treatment with DNase I(Qiagen). The amplifications of the samples were carried outusing QuantiTect Probe RT-PCR Kit (Qiagen) according tomanufacturer’s specifications. All probes were labeled withthe fluorescent reporter dye 6-carboxyfluorescein (6-FAM) onthe 5′-end, and with the Black Hole 1 (BH 1) fluorescentquencher dye on the 3′-end. The amplifications for CYP1A1,CYP1B1 and AKR1C9 were run on the LightCycler (RocheDiagnostics GmbH, Mannheim, Germany) using the followingprogram: reverse transcription at 50 ◦C for 20 min and initialactivation step at 95 ◦C for 15 min, followed by 40 cycles at95 ◦C for 0 s and 60 ◦C for 60 s. As the levels of CYP1A2mRNA in WB-F344 cells were very low and undetectable with

TM

mal cycler (Corbett Research, Sydney, Australia), using thefollowing program: reverse transcription at 50 ◦C for 30 minand initial activation step at 95 ◦C for 15 min, followed by 40cycles at 94 ◦C for 15 s and 60 ◦C for 60 s. Gene expressionfor each sample was expressed in terms of the threshold cycle(Ct), normalized to housekeeping gene porphobilinogen deam-inase (�Ct). �Ct values were then compared between controlsamples (DMSO 0.1%) and samples treated with dibenzocar-bazoles to calculate ��Ct (�Ct [control] − �Ct [xDBC]). Thefinal comparison of transcript ratios between samples is givenas 2−��Ct [45].

2.5. Assessment of cell proliferation and cell cycledistribution

The proliferative effects of DBC and its derivatives on con-fluent WB-F344 cells were determined as described previously[28]. Briefly, cells were seeded in 4-well cell-culture plates(Nunc, Roskilde, Denmark) and grown until they reached anapproximate confluency. The cells were exposed to tested com-pounds dissolved in DMSO for 72 h. The final concentration ofDMSO did not exceed 0.1% (v/v) in any of the samples. Themedium with tested compounds was changed daily. Follow-ing the treatment, medium was removed, cells were harvestedwith trypsin and counted with a Coulter Counter (Model ZM,Coulter Electronics, Luton, UK). Cells were then washed withPBS, and fixed in 70% ethanol at 4 ◦C overnight. Fixed cellswere washed once with PBS and resuspended in 0.5 ml of Vin-delov solution (1 M Tris–HCl, pH 8.0; 0.1% Triton X-100,v/v; 10 mM NaCl; propidium iodide 50 �g/ml; RNAse A 50Kunitz units/ml) [46] and incubated at 37 ◦C for 30 min. Cellswere then analyzed on FACSCalibur, using 488-nm (15 mW)

probe s

CAGCTTGCCATACAT

AACTATGTTTAGGAG

CCTCTATGGTGCAGG

TTTATCTTTTCTCGT

CCTGGCTGGGACACC

ward (Fprime

LightCycler, this reaction was run on the Rotor-Gene ther-

Table 1Primer and probe sequences for quantitative real-time RT-PCR

Genes/accession no. Primer and

CYP1A1 F980 5′-ATGTCNM 012540 R1122 5′-ATCCC

P1057 5′-CCAGG

CYP1A2 F886 5′-GTGAGNM 012541 R957 5′-GTGAC

P917 5′-CCCTC

CYP1B1 F1398 5′-CTCATNM 012940 R1491 5′-GACGT

P1463 5′-CTCAT

AKR1C9 F809 5′-TCACCD17310 R908 5′-CTGGA

P880 5′-TTTGT

PBG-D F480 5′-CCTCAX06827 R612 5′-TTCCT

P512 5′-CCTCA

a The primers and probes are identified by letters designating the forto the position of the base at the 5′-end of the positive strand of theaccession.

air cooled argon-ion laser for propidium iodide excitation, and

equencesa Product length (bp)

CTCAGATGATAAGGTC-3′ 167ATCACTGTGTCTAAC-3′GAGGCTCCAAGAGATAGC-3′

CAAAGACAACGGTG-3′ 94CAAATCCAGCTCC-3′AAGATTGTCAACATTGTC-3′

TTACCAGATACCCG-3′ 117AAGTTGGGTTGGTC-3′GCAGGCGGTCCCTCCCC-3′

CTCAACCAGAGCAA-3′ 122CTGATCCACCCAT-3′GAACTTCCCAGCGTGC-3′

AATTCAAGAGTATTCG-3′ 156TGCAAGATCTGGCC-3′CGCCTTCGGAAGCT-3′

), the reverse (R) primer or the probe (P), and a number correspondingr or probe in the gene reference sequences, according to number of

J. Vondracek et al. / Mutation Research 596 (2006) 43–56 47

CELLQuestTM software for data acquisition (Becton Dickin-son, San Jose, CA). A minimum of 15,000 events was collectedper sample. Data were analyzed using ModFit LT Version 2.0software (Verity Software House, Topsham, ME).

2.6. Cell death detection

Cell death induced by the tested compounds was determinedusing morphological criteria (fragmentation of nuclei). Cellswere incubated with test compounds or vehicle as describedabove. After indicated time period, cells were harvested bytrypsinization (including the floating cells) and prepared forDNA labeling with DAPI. For DAPI staining, 5 × 105 cellswere resuspended with 50 �l of methanol containing 1 �g/mlDAPI (final concentration) and incubated for 30 min at roomtemperature. After incubation, the cells were centrifuged andmixed with 15 �l of MOWIOL (10% MOWIOL 4-88 was pre-pared in 25% glycerol, 100 mM Tris–HCl, pH 8.5) solution andmounted for observation under a fluorescence microscope. Aminimum of 200 nuclei were counted per sample.

2.7. Western blotting

In order to determine effects of DBC and its two derivativeson Ser-15 phosphorylation of p53 protein, confluent WB-F344cells were exposed for 24 h to test compounds or to 0.1%DMSO as vehicle control. Effects of DBC and methylatedderivatives on connexin43 hyperphosphorylation have beendetermined in cells exposed to 30 �M DBC, DiMeDBC, N-MeDBC, 20 nM TPA (positive control) and DMSO (negativecontrol) for 30 min. Whole cell lysates were prepared by har-vesting cells in lysis buffer (1% SDS, TRIS, 10% glycerol,protease inhibitor cocktail), and protein concentrations weredFsPsate

2

fc(actlabmmw

Lucia image analysis software (Laboratory Imaging, Prague,Czech Republic). Three independent experiments were carriedout in duplicate and at least three scrapes per well were eval-uated. In a separate set of experiments, effects of compoundson GJIC were investigated after 60 min and 24-h incubation,respectively.

2.9. Statistical analysis

For each compound tested, relative AhR-inducing poten-cies were defined as the ability to induce luciferase activityusing concentration-response curves. Their relative inductionequivalency factors (IEFs) were estimated as described previ-ously [44]. The ratio of GJIC inhibition related to the negativecontrol was evaluated and expressed in percentage (fractionof control, FOC) [27]. Cell proliferation data were expressedas means ± standard deviation for at least three independentrepeats. The data were analyzed by Student t-test, or by the non-parametric Mann–Whitney U-test and Kruskal–Wallis analysisof variance, where appropriate. A p-value of less than 0.05 wasconsidered significant.

3. Results

3.1. DiMeDBC is a significantly more potent AhRagonist than its parental compound or N-MeDBC

The induction of AhR-mediated activity is an impor-tant effect of potent liver tumor promoters such asTCDD [37,47], and it leads to induction of enzymesinvolved both in metabolic activation and in detoxifi-cation of PAHs. Therefore, we investigated effects of

etermined with DC Protein Assay (BioRad, Hercules, CA).or Western blot analyses, equal amounts of total protein wereubjected to 10% SDS PAGE, electrotransferred onto Hybond-membrane, immunodetected using appropriate primary and

econdary antibodies, and visualized by ECL + Plus reagentccording to manufacturer’s instructions. After immunodetec-ion, each membrane was stained with amidoblack to confirmqual protein loading.

.8. Detection of GJIC

The modified scrape loading/dye transfer assay was per-ormed as described previously [27]. The confluent WB-F344ells, grown in 24-well plates, were exposed to test compoundsup to 50 �M concentration), 12-O-tetradecanoylphorbol-13-cetate (TPA) (20 nM, positive control), or DMSO (negativeontrol) for 30 min. After the exposure, the cells were washedwice with 0.5× PBS; fluorescent dye was added (lucifer yel-ow 0.05%, w/v in PBS) and the cells were scraped using

surgical blade. After 4 min, the cells were washed twicey 0.5× PBS and fixed with 4% formaldehyde (v/v) and theigration of the dye was evaluated using an epifluorescenceicroscope. The distance of dye migration from a scrape lineas measured at six randomly chosen spots per scrape, using

DBC and its derivatives on AhR activation, using rat hep-atoma cell line stably transfected with luciferase reportergene under control of dioxin responsive elements. Thecapacity of dibenzocarbazoles to activate the AhR wasmeasured using concentrations ranging from 100 pM to10 �M. TCCD, the most potent AhR ligand, was usedas a positive control. As shown in Fig. 2 and Table 2,

Table 2Relative potencies of DBC and its methylated derivatives to activateAhR (DR-CALUXTM assay), and to inhibit GJIC in WB-F344 cells

AhR-IEFsa

(6 h)AhR-IEFs(24 h)

GJIC inhibitionIC50 (30 min)

TCDD 1 1 –TPA – – 7.6 nMDBC 4.91E−06 2.64E−07 11.9 �MDiMeDBC 2.46E−03 3.17E−05 38.7 �MN-MeDBC 2.74E−06 8.99E−07 40.9 �M

a The numbers represent induction equivalency factors (IEFs) cal-culated as the ratio between the 25% effective concentration (EC25)of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and concentration ofselected compound inducing the 25% of maximum TCDD-inducedluciferase activity.

48 J. Vondracek et al. / Mutation Research 596 (2006) 43–56

Fig. 2. Aryl hydrocarbon receptor-mediated induction of luciferasereporter gene in H4IIEpGudLuc1.1 rat hepatoma cell line by TCDD,DBC, DiMeDBC and N-MeDBC after 6 and 24-h exposure. The datashown here are representative of three independent experiments; errorbars represent standard deviation from the mean value.

DiMeDBC was a relatively potent inducer of AhR-mediated activity following the 6-h incubation and itsactivity declined following the 24-h incubation period.Its relative potency, expressed as induction equivalencyfactor (IEF), was more than two orders of magnitudehigher than relative potencies of DBC and N-MeDBC.Both DBC and N-MeDBC induced less than 50% ofthe maximum TCDD-induced AhR-mediated activity(Fig. 2, Table 2).

3.2. Induction of mRNA expression of enzymesparticipating in metabolic activation

As the above results suggested that dibenzocarbazolesare AhR agonists, we next investigated the effects ofDBC and its methylated derivatives on expression of

enzymes that have been suggested to participate intheir metabolic activation in WB-F344 cells, using real-time quantitative RT-PCR. As summarized in Fig. 3A,DiMeDBC was the most potent inducer of CYP1A1expression among the three compounds tested and 1 �Mconcentration induced levels of mRNA similar to theprototypical AhR ligand TCDD. DBC also inducedCYP1A1 expression, however the mRNA levels weresignificantly lower than those induced by DiMeDBC.Compared to these two compounds, the capacity of N-MeDBC to induce CYP1A1 mRNA was very low. Thebackground levels of CYP1A2 in WB-F344 cells arevery low compared to rat hepatoma cells and induc-tion of CYP1A2 mRNA levels is limited (manuscriptin preparation). We found that DiMeDBC was inducinga significant increase in CYP1A2 mRNA, whereas DBCinduced only a minor increase of CYP1A2 mRNA level(Fig. 3B).

N-MeDBC was only a poor inducer of CYP1B1expression, while both DBC and DiMeDBC increasedlevels of CYP1B1 mRNA to a similar extent (Fig. 3C).Similarly, both DBC and DiMeDBC increased expres-sion of AKR1C9 mRNA (Fig. 3D). N-MeDBC againdid not significantly increase AKR1C9 expression. Wefound that 10 �M concentration of N-MeDBC inducedminor increases of CYP1A1, CYP1B1 and AKR1C9mRNAs expression, however these were still signifi-cantly lower that the levels induced by 1 �M doses ofDiMeDBC or DBC (data not shown).

3.3. Effects of DBC and its derivatives on cellproliferation, cell cycle and cell death

Our recent results suggested that PAHs may inducemultifaceted responses in rat liver epithelial cells, suchas the modulation of proliferation/apoptosis balance, inaddition to previously reported acute inhibition of GJIC[27,28]. Therefore, we investigated impact of both DBCand its two methylated derivatives on cell proliferation,cell cycle and apoptosis in contact-inhibited WB-F344cells. The confluent cells were treated with dibenzo-carbazoles at concentrations ranging from 100 pM to10 �M. As summarized in Fig. 4A, both DiMeDBCand DBC were found to induce a sharp increase inpercentage of S-phase cells, suggesting that, similar toother genotoxic polyaromatics, these compounds areinducing accumulation of cells in S-phase at concentra-tions 100 nM–10 �M (DBC) and 1–10 �M (DiMeDBC).DBC was most efficient at a concentration of 1 �M(∼40% cells in S-phase) while DiMeDBC at a con-centration 10 �M (∼60% cells in S-phase). A statis-tically significant increase in cell number was found

J. Vondracek et al. / Mutation Research 596 (2006) 43–56 49

Fig. 3. Induction of CYP1A1 (A), CYP1A2 (B), CYP1B1 (C) and AKR1C9 (D) mRNA following 24-h treatment with DBC or its two methylatedderivatives (1 �M) as compared to a prototypical AhR ligand, TCDD (1 nM). Total RNA was isolated and quantitative real-time RT-PCR wasperformed as described in Section 2. The results are expressed as mean ± S.D. of three independent experiments.

after 1 �M DiMeDBC treatment (Fig. 4B), while bothcompounds also markedly increased apoptosis, detectedas nuclear fragmentation, especially at 10 �M concen-tration (Fig. 4C). Unlike two hepatocarcinogenic com-pounds, N-MeDBC did not affect significantly the cellcycle, DNA fragmentation and cell proliferation. Aminor increase in percentage of cells in S-phase wasdetected only at 10 �M concentration.

3.4. DBC and DiMeDBC induce phosphorylation ofp53 at serine 15

The p53 tumor suppressor protein is a potent tran-scription factor that accumulates in cells in responseto signals arising from DNA-damaging agents, includ-ing DBC [32]. It can be post-translationally modifiedat various amino acid residues and a number of stud-ies have shown that phosphorylation of p53 protein atserine 15 may play a critical role in the stabilization,up-regulation and functional activation of p53 duringgenotoxic stress [48]. Phosphorylation of p53 at Ser-15 was measured after 24-h exposure to dibenzocar-

bazoles at concentrations ranging from 0.01 to 10 �M(Fig. 5). Dibenzo[a,l]pyrene, the strongest genotoxinamong PAH, was used as a positive control in theseexperiments. Both DBC and diMeDBC induced Ser-15phosphorylation of p53, although to a lower extent thandibenzo[a,l]pyrene. In contrast, no increase in Ser-15phosphorylation of p53 was determined in cells exposedto tissue-specific sarcomagen N-MeDBC.

3.5. DBC is a potent inhibitor of GJIC in WB-F344cells

The capacity of DBC and its methyl derivatives toinhibit the GJIC was measured in concentrations rang-ing from 1 to 50 �M after 30 min, 60 min and 24 h ofexposure. As shown in Fig. 6A and B and in Table 2,DBC was efficient inhibitor of GJIC in rat liver epithe-lial cells. The parent molecule decreased significantlydye transfer at concentration as low as 10 �M and itsIC50 after 30-min incubation was 11.6 �M. Interest-ingly, unlike many other PAHs that inhibit GJIC [26,27],the effect of DBC was not transient and it lasted at

50 J. Vondracek et al. / Mutation Research 596 (2006) 43–56

Fig. 4. Modulation of cell proliferation and cell death by DBC, DiMeDBC and N-MeDBC in WB-F344 cells assessed by measuring percentage ofcells in S-phase of cell cycle (A), by counting cell numbers (B) and by detecting of percentage of cells with fragmented nuclei considered apoptotic(C). Cells were cultured for 72 h prior to exposition, and then treated for another 72 h with TCDD or test compounds in a daily applied fresh medium.Cells were then trypsinized and counted on a Coulter Counter (Coulter Electronics, Luton, UK). Cells were then fixed in ethanol for a subsequentflow cytometric analysis of DNA content. The cell-cycle analysis was performed on a FACSCalibur flow cytometer equipped with CellQuest andModFit software (Becton Dickinson, San Jose, CA, USA). For detection of apoptosis, both floating and adherent cells were collected and stained withDAPI. A minimum of 200 cells were counted per each sample. Dibenzo[a,1]pyrene (DBalP; 100 nM) was used as a positive control for induction ofapoptosis. All data are expressed as mean ± S.D. of three independent experiments run in duplicates. *Significant difference between control (0.1%DMSO) and treated samples (p < 0.05). **Significant difference between control (0.1% DMSO) and treated samples (p < 0.01).

J. Vondracek et al. / Mutation Research 596 (2006) 43–56 51

Fig. 5. Induction of p53 phosphorylation at Ser15 by DBC andDiMeDBC. WB-F344 cells were treated with the indicated concentra-tions of test compound for 24 h. Dibenzo[a,1]pyrene (DBalP; 100 nM)was used as a positive control. Cell lysates were prepared and Westernblotting was performed as described in Section 2. The results shownhere are representative of three independent experiments.

least for 24-h (Fig. 6C). The longer time intervals werenot tested due to the extensive cell death induced bythis compound. In contrast, both DiMeDBC and N-MeDBC failed to produce a complete inhibition ofGJIC, and their respective IC50 doses were 38.7 and40.9 �M.

One of the potential mechanisms of GJIC inhibitionis the induction of hyperphosphorylation of connexin43,a major connexin species present in rat liver epithelialcells. It is characteristic for prototypical tumor promot-ers, such as TPA or growth factors [49,50]. Connexin43hyperphosphorylation was assessed in WB-F344 cellstreated with equimolar concentration (30 �M) of diben-zocarbazoles for 30 min. However, unlike TPA, noneof the tested compounds induced connexin43 hyper-phosphorylation in rat liver epithelial cells (Fig. 6D),suggesting a distinct mechanism of GJIC inhibition.

4. Discussion

A hallmark of numerous chemical carcinogens is theirstrict organ- or tissue-specificity; however, the precisemechanism of this event is still unknown. It is supposedthat the lack of drug-metabolizing enzymes responsiblefor production of ultimate mutagens in a target tissuecould play an important role in this phenomenon. Cova-lent binding of a xenobiotic into DNA or another formof DNA damage represents an essential step in chemical

F by DBCo 60 mine ing/dya indepenN 4 cellsu controld ree independent experiments.

ig. 6. Inhibition of gap junctional intercellular communication (GJIC)f GJIC after 30 min exposure. (C) Inhibition of GJIC after 30 min,ffects of test compounds on GJIC were determined by the scrape loads fraction of control (FOC) and they represent means ± S.D. of three-MeDBC do not induce connexin43 hyperphosphorylation. WB-F34sed as a solvent control, while TPA (20 nM) was used as a positiveescribed in Section 2. The results shown here are representative of th

and its methylated derivatives in WB-F344 cells. (A and B) Inhibitionand 24 h of incubation with increasing concentrations of DBC. Thee transfer method as described in Section 2. The results are expresseddent experiments carried out in triplicates. (D) DBC, DiMeDBC andwere treated with 30 �M of test compounds for 30 min. DMSO was. Cell lysates were prepared and Western blotting was performed as

52 J. Vondracek et al. / Mutation Research 596 (2006) 43–56

carcinogenesis, although a number of studies suggestthat the spontaneous mutations, arising from errors dur-ing turnover of undamaged DNA, might represent thecrucial initiating events [51]. Nevertheless, other biolog-ical pathways have to be triggered in the target tissue, inorder to promote and complete the process of neoplastictransformation. Removal of an initiated cell from a con-trol of neighboring cells, activation of a mitogenic path-ways or inhibition of programmed cell death representsome of the critical events in chemical carcinogenesis[24].

The genotoxic effects of DBC and its tissue andorgan specific derivatives are well documented both invivo and in vitro [1,10,20,52]. Nevertheless, there iscurrently little information on effects of DBC and itsderivatives on activation of AhR, the major regulatorof PAH-metabolizing enzymes, on regulation of cellproliferation/apoptosis or on intercellular communica-tion. The target organ of DBC and DiMeDBC, the liver,consists not only of hepatocytes and supporting non-parenchymal cells, but it also contains a population ofprogenitor cells, which in case of serious liver injury andinhibition of hepatocyte regenerative capacity, can giverise to oval cells [53]. The stem cell hypothesis of can-cer suggests that cancer may arise from maturation blockin progenitor cells and that stem or ‘stem-like’ cells arelikely to play an important role in carcinogenesis [54,55].The existence of stem cells in liver has been a matterof long debate, nevertheless it is now generally agreedthat oval cells of liver are ‘stem-like’ cells descending

oxidative stress, which may further contribute to DNAdamage [23]. We found that, in contrast to DBC andN-MeDBC, DiMeDBC is an efficient AhR ligand in rathepatoma cells (Fig. 3, Table 2), with a relative potencyclose to that of benzo[a]pyrene [44]. DiMeDBC wasthe most potent inducer of CYP1A1 mRNA expres-sion in WB-F344 cells among the tested compounds.Only DiMeDBC also induced a significant increase inCYP1A2 mRNA levels, comparable to TCDD in ratliver epithelial cells, although the levels of its mRNAwere still significantly lower than, e.g. in rat hepatomacells (manuscript in preparation). This is in agreementwith the study showing that DiMeDBC induces CYP1A2expression in the adult mice liver, whereas DBC is onlya poor inducer of CYP1A1/2 [52]. N-MeDBC, a tissue-specific sarcomagen, does not induce overexpression ofCYP1A subfamily isoforms [52]. We found that it didnot increase CYP1A2 levels in WB-F344 cells signifi-cantly, and it was also a poor inducer of CYP1A1 mRNAin these cells, and only at 10 �M concentration (datanot shown). Both DBC and DiMeDBC were efficientinducers of CYP1B1 mRNA expression. N-MeDBC is asubstrate for both CYP1A1 and CYP1A2, and it inducesthe highest level of gene mutations and DNA adductsdue to activation via CYP1A2, in comparison to the par-ent molecule and the strict hepatocarcinogen DiMeDBC[17–19,61]. Therefore, its failure to stimulate expressionof these enzymes might be responsible for the lack ofits effects on p53 phosphorylation, S-phase arrest/delayand apoptosis, observed in the present study. It should be

from a population of progenitor cells involved in liverregeneration [34,56,57]. The oval cells can develop intoboth hepatocytes and biliary epithelial cells, and theymight play a significant role in hepatocarcinogenesis[41,42,58,59]. Therefore, in the present study, we inves-tigated the effects of DBC and its methylated derivativeson cell proliferation, cell death and GJIC, as well as theirimpact on AhR activity and expression of metabolizingenzymes in rat liver epithelial WB-F344 cells, which areconsidered to represent an in vitro model of oval cells[39].

The outcome of PAH effects in rat liver epithelialcells seems to reflect both AhR activation and geno-toxic effects, the latter being induced by PAH metabolites[28]. Although the metabolic activation due to activityof enzymes expressed at basal levels might be suffi-cient for formation of DNA adducts [60], many PAHsand related compounds enhance their own activationvia binding to the Ah receptor, which mediates theselective overexpression of particular drug metaboliz-ing enzymes [16]. Induction of AhR gene battery hasbeen also reported to be associated with generation of

noted, however, that no direct relationship was observedbetween the capacity of DBC and its methylated deriva-tives to induce CYP1A1/2 and the tissue specificity ofcarcinogenesis or DNA binding in liver [52]. This sug-gests that other enzymes, such as CYP1B1, are alsoinvolved in their metabolic activation.

The role of CYP1B1 in metabolic activation of diben-zocarbazoles is presently unclear. Nevertheless, a num-ber of studies suggest that both CYP1A1 and CYP1B1are actively involved in formation of ultimate geno-toxic metabolites of PAHs and NHAs [13]. We foundthat both DBC and DiMeDBC can increase CYP1B1expression in rat liver epithelial cells, while N-MeDBConly partially upregulated the expression of this enzymeat 10 �M concentration. Although CYP1B1 is tradi-tionally viewed as being present mostly in extrahep-atic tissues, PAHs are known to increase its levels inrodent liver [62,63] and the present data suggest that itis induced by NHAs in the population of liver progeni-tor cells. Interestingly, it has been shown that some poorAhR ligands among PAHs, such as dibenzo[a,l]pyrene,can increase CYP1B1 in murine liver, suggesting that

J. Vondracek et al. / Mutation Research 596 (2006) 43–56 53

also AhR-independent mechanisms might participate inCYP1B1 regulation [63]. This is supported by the findingthat 3-methylcholanthrene increases CYP1B1 expres-sion in liver of AhR knock-out mice [60]. This mightexplain why DBC and DiMeDBC induced similar levelsof CYP1B1 mRNA, despite the latter one being far morepotent AhR ligand.

In contrast to formation of DNA adducts due toCYP1A1/2 and/or CYP1B1 activity, little is known aboutthe role of dihydrodiol dehydrogenases in metabolic acti-vation of DBC and its derivatives. These monomericcytosolic NADP(H)-dependent oxidoreductases havebeen suggested to play a significant role in themetabolism of xenobiotics, including PAHs [64]. RatAKR1C9 catalyzes oxidation of PAH trans-dihydrodiolsto reactive PAH o-quinones, and the formation of reactiveand redox active o-quinones has been suggested to con-tribute to PAH-induced carcinogenesis [15]. A metabo-lite of DBC, DBC-3,4-dione, has been synthesized andit has been shown to form adducts with nucleic bases ornucleosides in vitro [65]. A liver DNA adduct has beenidentified in mice treated with DBC-3,4-dione, which isidentical to one of the adducts identified in liver of mousetreated with DBC. This seems to suggest that AKR1C9might participate in DBC metabolic activation. In thepresent study, both DBC and DiMeDBC were found toincrease the expression of AKR1C9 mRNA in rat liverepithelial cells (Fig. 3D). To our knowledge, this is thefirst evidence that DBC or its derivative can increaseAKR1C9 expression in liver cells, indicating that thefiTltltmatfiae

itvoaea

maintenance of genetic integrity of multicellular organ-isms [68]. Induction of S-phase arrest and/or apopto-sis is observed with a number of genotoxic PAHs, aswell as with their proximate carcinogenic metabolites[28,66,69–72]. Both DBC and DiMeDBC induced apop-tosis and accumulation of cells in S-phase of cell cycleespecially at concentrations 1–10 �M after 72 h of treat-ment (Fig. 5). In contrast, induction of apoptosis oraccummulation of cells in S-phase was not observed withN-MeDBC, again suggesting that this compound was notinducing massive DNA damage in WB-F344 cells.

In rat liver epithelial cells, both PAHs and hetero-cyclic aromatic compounds can disrupt cell prolifera-tion/apoptosis balance and stimulate cell proliferation incontact-inhibited cells [28,40], a mode of action typicallyinduced by some tumor promoters [73]. In this study,DiMeDBC significantly increased cell numbers at 1 �Mconcentration, although to a lower extent than TCDD, aperstistent AhR ligand. However, higher concentrationsof DiMeDBC failed to potentiate cell proliferation, mostprobably due to induction of apoptosis. The induction ofcell proliferation at lower concentration and induction ofapoptosis at higher concentrations of DiMeDBC, mightreflect the fact that its high levels saturate DNA repairmechanisms, which consequently leads to accumulationof high levels of DNA adducts [30,74] and activationof apoptosis due to DNA damage. In vivo, the exten-sive cell death through apoptosis or necrosis may lead toregenerative cell proliferation in the liver, which seemsto be a critical event in DiMeDBC-mediated mutagenic-

ormation of o-quinones might contribute to cytotoxic-ty and mutagenicity of DBC and DiMeDBC in liver.he induction of AKR1C9 expression by DBC and its

iver-specific derivative, DiMeDBC, suggests that induc-ion of oxidative stress might also contribute to theiriver-specific carcinogenicity. It has been suggested thathe formation of reactive and redox active o-quinones

ight lead to formation of reactive oxygen species, suchs hydrogen peroxide, superoxide or hydroxyl radical,hrough redox cycling [15]. The oxidative stress mighturther contribute both to tumor initiation through inflict-ng oxidative DNA damage and to tumor promotion byctivating specific cell signaling pathways associated,.g. with regulation of cell proliferation [15].

PAHs and some other chemical carcinogens fail tonduce G1 arrest, which has been suggested to increasehe likelihood of fixation of mutations [66]. The acti-ation of S-phase checkpoint could be an integral partf cellular response to compounds forming bulky DNAdducts, such as PAH dihydrodiol epoxides [67], whilelimination of potential DNA lesions by removing dam-ged cells through apoptosis is a further option for

ity in mice [29–31]. The stimulation of cell proliferationobserved at lower dose might thus further contribute tohepatocarcinogenicity of DiMeDBC.

Induction of high amounts of DNA adducts can lead toupregulation of p53 tumor suppressor levels and its activ-ity by PAHs [72,75] and DBC [32]. The p53 protein canbe post-translationally modified at various amino acidresidues. Various studies have shown that phosphory-lation of p53 protein at serine 15 may is important forthe stabilization and functional activation of p53 dur-ing cellular stress [48,70]. We found that both DBC andDiMeDBC but not N-MeDBC significantly increasedp53 phosphorylation at serine 15 (Fig. 5). These resultsstrongly support the hypothesis that DBC and DiMeDBCare efficiently metabolized in WB-F344 cells to theirproximate genotoxic metabolites, which can then leadto p53 activation, as well as accumulation of cells inS-phase and extensive apoptosis.

Many environmentally important PAHs, especiallythose with lower molecular mass, are potent in vitroinhibitors of GJIC in WB-F344 rat liver epithelial cellline [27]. Disruption of GJIC is supposed to be involved

54 J. Vondracek et al. / Mutation Research 596 (2006) 43–56

in tumor promotion, as the reduced GJIC is a character-istic feature of tumor cells and a vast majority of tumorpromoters inhibit GJIC [24,25]. The parent compoundDBC was a powerful inhibitor of GJIC in WB-F344cells. Its IC50 value approached the efficiency of fluoran-thene, the most potent inhibitor of GJIC identified amongPAHs so far [26,27]. Unlike other PAHs, which are onlytransient inhibitors of GJIC, DBC induced a permanentdownregulation of GJIC, without signs of recovery for upto 24 h. Nevertheless, this long-term effect might also beassociated with induction of apoptosis by this compound.On the other hand, neither DBC nor N-MeDBC were ableto fully suppress GJIC; their IC50 values were consider-ably higher in comparison to DiMeDBC (Table 2). Theinhibition of GJIC was not associated with induction ofconnexin43 hyperphosphorylation. This suggested thatthe mode of action of DBC differs from that of TPA, amodel tumor promoter, which is known to induce con-nexin43 phosphorylation through activation of proteinkinase C and mitogen-activated protein kinases ERK1/2[49,50].

Taken together, significant differences among theeffects of DBC and its two tissue-specific methylatedderivatives on rat liver epithelial cells were observedin the present study. The sarcomagenic N-MeDBC wasa weak inhibitor of GJIC and a weak inducer of theAhR-mediated activity, which failed to induce expres-sion of enzymes involved in formation of its proximategenotoxic metabolites. It did not induce either apopto-sis or a significant accumulation of cells in S-phase, nor

by grant no. 525/03/1527 from the Czech Science Foun-dation, grant no. B6004407 from the Grant Agency ofthe Academy of Sciences of the Czech Republic andVEGA grant 2/6032 of the Slovak Academy of Sciences.The study was further supported by the Research Planof the Academy of Sciences of the Czech Republic, no.AVOZ50040507 and by the Ministry of Agriculture ofthe Czech Republic (grant no. 00002716201).

References

[1] D. Warshawsky, G. Talaska, W. Xue, J. Schneider, Compara-tive carcinogenicity, metabolism, mutagenicity, and DNA bindingof 7H-dibenzo[c,g]carbazole and dibenz[a,j]acridine, Crit. Rev.Toxicol. 26 (1996) 213–249.

[2] D. Szafarz, F. Perin, D. Valero, F. Zajdela, Structure and car-cinogenicity of dibenzo(c,g)carbazole derivatives, Biosci. Rep. 8(1988) 633–643.

[3] IARC, Certain Polycyclic Aromatic Hydrocarbons and Hetero-cyclic Compounds, vol. 3, International Agency for Research onCancer, Lyon, 1973.

[4] IARC, Polynuclear Aromatic Compounds, vol. 32, InternationalAgency for Research on Cancer, Lyon, 1983.

[5] P. Boffetta, N. Jourenkova, P. Gustavsson, Cancer risk fromoccupational and environmental exposure to polycyclic aromatichydrocarbons, Cancer Causes Control. 8 (1997) 444–472.

[6] A. Sellakumar, P. Shubik, Carcinogenicity of 7H-dibenzo[c,g]carbazole in the respiratory tract of hamsters,J. Natl. Cancer Inst. 48 (1972) 1641–1646.

[7] D. Warshawsky, W. Barkley, M.L. Miller, K. LaDow,A. Andringa, Carcinogenicity of 7H-dibenzo[c,g]carbazole,dibenz[a,j]acridine and benzo[a]pyrene in mouse skin andliver following topical application, Toxicology 93 (1994) 135–

[

[

[

[

[

did it increase p53 phosphorylation. In contrast, bothliver carcinogens, DBC and DiMeDBC, induced p53phosphorylation, S-phase arrest/delay and apoptosis inrat liver epithelial ‘stem-like’ cells. Nevertheless, therewere significant differences among their effects. DBCwas an effective GJIC inhibitor, while DiMeDBC was apotent AhR agonist, which induced significantly higherlevels of CYP1A1/2 mRNAs and stimulated cell prolif-eration. Both DBC and DiMeDBC increased expressionof CYP1B1 and AKR1C9, which might contribute totheir metabolic activation and their hepatocarcinogenic-ity. Thus, both compounds elicit multiple toxic eventsassociated with carcinogenicity, in this in vitro model ofliver progenitor cells. However, it remains to be estab-lished, whether the pluripotent oval cells are a target forDBC and DiMeDBC in vivo.

Acknowledgments

The authors thank Dr. F. Perin, Department of Geno-toxicity and Carcinogenicity, Institute Curie, France forproviding DBC derivatives. This study was supported

149.[8] F. Perin, D. Valero, J. Mispelter, F. Zajdela, In vitro metabolism

of N-methyl-dibenzo[c,g]carbazole a potent sarcomatogen devoidof hepatotoxic and hepatocarcinogenic properties, Chem. Biol.Interact. 48 (1984) 281–295.

[9] M.E. Schurdak, D.B. Stong, D. Warshawsky, K. Randerath,N-methylation reduces the DNA-binding activity of 7H-dibenzo[c,g]carbazole approximately 300-fold in mouse liver butonly approximately 2-fold in skin: possible correlation with car-cinogenic activity, Carcinogenesis 8 (1987) 1405–1410.

10] D. Valero, F. Perin, F. Zajdela, Binding of 5,9-dimethyldibenzo[c,g]carbazole, a potent hepatocarcinogen, to mouse hep-atic cytosolic proteins, Carcinogenesis 4 (1983) 1333–1340.

11] D. Valero, F. Perin, M.J. Plessis, F. Zajdela, Sexual differencesin the expression of gamma-glutamyl transpeptidase during 5,9-dimethyldibenzo[c,g]carbazole-induced hepatocarcinogenesis inmice, Cancer Lett. 27 (1985) 181–197.

12] D. Renault, F. Tombolan, D. Brault, F. Perin, V. Thybaud, Compar-ative mutagenicity of 7H-dibenzo[c,g]carbazole and two deriva-tives in MutaMouse liver and skin, Mutat. Res. 417 (1998)129–140.

13] W. Xue, D. Warshawsky, Metabolic activation of polycyclic andheterocyclic aromatic hydrocarbons and DNA damage: a review,Toxicol. Appl. Pharmacol. 206 (2005) 73–93.

14] H.V. Dowty, W. Xue, K. LaDow, G. Talaska, D. Warshawsky,One-electron oxidation is not a major route of metabolic activationand DNA binding for the carcinogen 7H-dibenzo[c,g]carbazole

J. Vondracek et al. / Mutation Research 596 (2006) 43–56 55

in vitro and in mouse liver and lung, Carcinogenesis 21 (2000)991–998.

[15] T.M. Penning, M.E. Burczynski, C.F. Hung, K.D. McCoull, N.T.Palackal, L.S. Tsuruda, Dihydrodiol dehydrogenases and poly-cyclic aromatic hydrocarbon activation: generation of reactive andredox active o-quinones, Chem. Res. Toxicol. 12 (1999) 1–18.

[16] D.W. Nebert, T.P. Dalton, A.B. Okey, F.J. Gonzalez, Role of arylhydrocarbon receptor-mediated induction of the CYP1 enzymesin environmental toxicity and cancer, J. Biol. Chem. 279 (2004)23847–23850.

[17] A. Gabelova, G. Bacova, L. Ruzekova, T. Farkasova, Role ofcytochrome P4501A1 in biotransformation of a tissue specificsarcomagen N-methyldibenzo[c,g]carbazole, Mutat. Res. 469(2000) 259–269.

[18] A. Gabelova, T. Farkasova, G. Bacova, S. Robichova, Mutagenic-ity of 7H-dibenzo[c,g]carbazole and its tissue specific derivativesin genetically engineered Chinese hamster V79 cell lines stablyexpressing cytochrome P450, Mutat. Res. 517 (2002) 135–145.

[19] A. Gabelova, B. Binkova, Z. Valovicova, R.J. Sram, DNA adductformation by 7H-dibenzo[c,g]carbazole and its tissue- and organ-specific derivatives in Chinese hamster V79 cell lines stablyexpressing cytochrome P450 enzymes, Environ. Mol. Mutagen.44 (2004) 448–458.

[20] O. Perin-Roussel, N. Barat, F. Zajdela, F. Perin, Tissue-specificdifferences in adduct formation by hepatocarcinogenic and sar-comatogenic derivatives of 7H-dibenzo[c,g]carbazole in mouseparenchymal and nonparenchymal liver cells, Environ. Mol.Mutagen. 29 (1997) 346–356.

[21] A. Puga, Y. Xia, C. Elferink, Role of the aryl hydrocarbon receptorin cell cycle regulation, Chem. Biol. Interact. 141 (2002) 117.

[22] C.J. Elferink, Aryl hydrocarbon receptor-mediated cell cycle con-trol, Prog. Cell Cycle. Res. 5 (2003) 261–267.

[23] D.W. Nebert, A.L. Roe, M.Z. Dieter, W.A. Solis, Y. Yang, T.P.Dalton, Role of the aromatic hydrocarbon receptor and [Ah] genebattery in the oxidative stress response, cell cycle control, and

[

[

[

[

[

[

[

lacz mutant frequency in the liver of MutaTMMice treated with5,9-dimethyldibenzo[c,g]carbazole, Carcinogenesis 20 (1999)1357–1362.

[31] O. Dorchies, O. Perin-Roussel, O. Gillardeaux, J.A. Vericat, N.O.Roome, A. Prenez, F. Perin, Induction of DNA synthesis in mouseliver following increases of DNA adduct levels elicited by verylow cumulative doses of the genotoxic hepatocarcinogen 7H-dibenzo[c,g]carbazole, Toxicol. Pathol. 29 (2001) 528–534.

[32] T. O’Brien, G. Babcock, J. Cornelius, M. Dingeldein, G. Talaska,D. Warshawsky, K. Mitchell, A comparison of apoptosis andnecrosis induced by hepatotoxins in HepG2 cells, Toxicol. Appl.Pharmacol. 164 (2000) 280–290.

[33] I. Martın-Burriel, N.O. Roome, O. Dorchies, A. Prenez,Histopathological and molecular changes during apoptosis pro-duced by 7H-dibenzo[c,g]-carbazole in mouse liver, Toxicol.Pathol. 32 (2004) 202–211.

[34] K.N. Lowes, E.J. Croager, J.K. Olynyk, L.J. Abraham, G.C.Yeoh, Oval cell-mediated liver regeneration: role of cytokinesand growth factors, J. Gastroenterol. Hepatol. 18 (2003) 4–12.

[35] B. Knight, V.B. Matthews, B. Akhurst, E.J. Croager, E. Klinken,L.J. Abraham, J.K. Olynyk, G.C. Yeoh, Liver inflammation andcytokine production,but not acute phase protein synthesis, accom-pany the adult liver progenitor (oval) cell response to chronic liverinjury, Immunol. Cell Biol. 83 (2005) 364–374.

[36] C. Kohle, H. Gschaidmeier, D. Lauth, S. Topell, H. Zitzer, K.W.Bock, 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD)-mediatedmembrane translocation of c-Src protein kinase in liver WB-F344cells, Arch. Toxicol. 73 (1999) 152–158.

[37] C. Dietrich, D. Faust, S. Budt, M. Moskwa, A. Kunz, K.W. Bock,F. Oesch, 2,3,7,8-tetrachlorodibenzo-p-dioxin-dependent releasefrom contact inhibition in WB-F344 cells: involvement of cyclinA, Toxicol. Appl. Pharmacol. 183 (2002) 117–126.

[38] J. Vondracek, M. Machala, V. Bryja, K. Chramostova, P. Krcmar,C. Dietrich, A. Hampl, A. Kozubık, Aryl hydrocarbon receptor-activating polychlorinated biphenyls and their hydroxylated

[

[

[

[

[

[

[

apoptosis, Biochem. Pharmacol. 59 (2000) 65–85.24] J.E. Trosko, B.L. Upham, The emperor wears no clothes in the

field of carcinogen risk assessment: ignored concepts in cancerrisk assessment, Mutagenesis 20 (2005) 81–92.

25] H.S. Rosenkranz, N. Pollack, A.R. Cunningham, Exploring therelationship between the inhibition of gap junctional intercellularcommunication and other biological phenomena, Carcinogenesis21 (2000) 1007–1011.

26] B.L. Upham, S.J. Masten, B.R. Lockwood, J.E. Trosko, Nongeno-toxic effects of polycyclic aromatic hydrocarbons and their oxy-genation by-products on the intercellular communication of ratliver epithelial cells, Fundam. Appl. Toxicol. 23 (1994) 470–475.

27] L. Blaha, P. Kapplova, J. Vondracek, B. Upham, M. Machala, Inhi-bition of gap-junctional intercellular communication by environ-mentally occurring polycyclic aromatic hydrocarbons, Toxicol.Sci. 65 (2002) 43–51.

28] K. Chramostova, J. Vondracek, L. Sindlerova, B. Vojtesek, A.Kozubık, M. Machala, Polycyclic aromatic hydrocarbons modu-late cell proliferation in rat hepatic epithelial stem-like WB-F344cells, Toxicol. Appl. Pharmacol. 196 (2004) 136–148.

29] F. Tombolan, D. Renault, D. Brault, M. Guffroy, O. Perin-Roussel,F. Perin, V. Thybaud, Kinetics of induction of DNA adducts,cell proliferation and gene mutations in the liver of MutaMicetreated with 5,9-dimethyldibenzo[c,g]carbazole, Carcinogenesis20 (1999) 125–132.

30] F. Tombolan, D. Renault, D. Brault, M. Guffroy, F. Perin, V. Thy-baud, Effect of mitogenic or regenerative cell proliferation on

metabolites induce cell proliferation in contact-inhibited rat liverepithelial cells, Toxicol. Sci. 83 (2005) 53–63.

39] M.S. Tsao, J.D. Smith, K.G. Nelson, J.W. Grisham, A diploidepithelial cell line from normal adult rat liver with phenotypicproperties of ’oval’ cells, Exp. Cell. Res. 154 (1984) 38–52.

40] J. Vondracek, K. Chramostova, M. Plıskova, L. Blaha, W. Brack,A. Kozubık, M. Machala, Induction of aryl hydrocarbon receptor-mediated and estrogen receptor-mediated activities, and modula-tion of cell proliferation by dinaphthofurans, Environ. Toxicol.Chem. 23 (2004) 2214–2220.

41] M.L. Dumble, E.J. Croager, G.C. Yeoh, E.A. Quail, Generationand characterization of p53 null transformed hepatic progenitorcells: oval cells give rise to hepatocellular carcinoma, Carcino-genesis 23 (2002) 435–445.

42] T.A. Roskams, L. Libbrecht, V.J. Desmet, Progenitor cells in dis-eased human liver, Semin. Liver Dis. 23 (2003) 385–396.

43] J.M. Aarts, M.S. Denison, M.A. Cox, M.A. Schalk, P.M. Garrison,K. Tullis, L.H. de Haan, A. Brouwer, Species-specific antagonismof Ah receptor action by 2,2′,5,5′- tetrachloro- and 2,2′,3,3′4,4′-hexachlorobiphenyl, Eur. J. Pharmacol. 293 (1995) 463–474.

44] M. Machala, J. Vondracek, L. Blaha, M. Ciganek, J.V. Neca, Arylhydrocarbon receptor-mediated activity of mutagenic polycyclicaromatic hydrocarbons determined using in vitro reporter geneassay, Mutat. Res. 497 (2001) 49–62.

45] K.J. Livak, T.D. Schmittgen, Analysis of relative gene expressiondata using real-time quantitative PCR and the 2(-Delta Delta C(T))Method, Methods 25 (2001) 402–408.

56 J. Vondracek et al. / Mutation Research 596 (2006) 43–56

[46] L.L. Vindelov, Flow microfluorometric analysis of nuclear DNAin cells from solid tumors and cell suspensions. A new methodfor rapid isolation and straining of nuclei, Virchows Arch. B CellPathol. 24 (1977) 227–242.

[47] J.P. Vanden Heuvel, G. Lucier, Environmental toxicology ofpolychlorinated dibenzo-p-dioxins and polychlorinated dibenzo-furans, Environ. Health Perspect. 100 (1993) 189–200.

[48] Q.B. She, N. Chen, Z. Dong, ERKs and p38 kinase phosphorylatep53 protein at serine 15 in response to UV radiation, J. Biol. Chem.275 (2000) 20444–20449.

[49] R.J. Ruch, J.E. Trosko, B.V. Madhukar, Inhibition of connexin43gap junctional intercellular communication by TPA requires ERKactivation, J. Cell. Biochem. 83 (2001) 163–169.

[50] E. Rivedal, H. Opsahl, Role of PKC and MAP kinase in EGF- andTPA-induced connexin43 phosphorylation and inhibition of gapjunction intercellular communication in rat liver epithelial cells,Carcinogenesis 22 (2001) 1543–1550.

[51] W.G. Thilly, Have environmental mutagens caused oncomuta-tions in people? Nat. Genet. 34 (2003) 255–259.

[52] D. Taras-Valero, O. Perin-Roussel, M.J. Plessis, F.Zajdela, F. Perin, Tissue-specific activities of methylateddibenzo[c,g]carbazoles in mice: carcinogenicity, DNA adductformation, and CYP1A induction in liver and skin, Environ.Mol. Mutagen. 35 (2000) 139–149.

[53] N. Fausto, Mouse liver tumorigenesis: models, mechanisms, andrelevance to human disease, Semin. Liver Dis. 19 (1999) 243–252.

[54] V.R. Potter, Phenotypic diversity in experimental hepatomas: theconcept of partially blocked ontogeny. The 10th Walter HubertLecture, Br. J. Cancer 38 (1978) 1–23.

[55] J.E. Trosko, C.C. Chang, Stem cell theory of carcinogenesis, Tox-icol. Lett. 49 (1989) 283–295.

[56] S.H. Sigal, S. Brill, A.S. Fiorino, L.M. Reid, The liver as a stemcell and lineage system, Am. J. Physiol. 263 (1992) G139–G148.

[57] J. Travis, The search for liver stem cells picks up, Science 259(1993) 1829.

[

[

[

[

[

[63] T. Shimada, K. Inoue, Y. Suzuki, T. Kawai, E. Azuma, T. Naka-jima, M. Shindo, K. Kurose, A. Sugie, Y. Yamagishi, Y. Fujii-Kuriyama, M. Hashimoto, Arylhydrocarbon receptor-dependentinduction of liver and lung cytochromes P450 1A1, 1A2, and1B1 by polycyclic aromatic hydrocarbons and polychlorinatedbiphenyls in genetically engineered C57BL/6J mice, Carcinogen-esis 23 (2002) 1199–1207.

[64] M.E. Burczynski, H.K. Lin, T.M. Penning, Isoform-specificinduction of a human aldo-keto reductase by polycyclic aromatichydrocarbons (PAHs), electrophiles, and oxidative stress: impli-cations for the alternative pathway of PAH activation catalyzedby human dihydrodiol dehydrogenase, Cancer Res. 59 (1999)607–614.

[65] W. Xue, A. Siner, M. Rance, K. Jayasimhulu, G. Talaska,D. Warshawsky, A metabolic activation mechanism of 7H-dibenzo[c,g]carbazole via o-quinone. Part 2: covalent adducts of7H-dibenzo[c,g]carbazole-3,4-dione with nucleic acid bases andnucleosides, Chem. Res. Toxicol. 15 (2002) 915–921.

[66] Q.A. Khan, A. Dipple, Diverse chemical carcinogens fail toinduce G(1) arrest in MCF-7 cells, Carcinogenesis 21 (2000)1611–1618.

[67] R.S. Weiss, P. Leder, C. Vaziri, Critical role for mouse Hus1 inan S-phase DNA damage cell cycle checkpoint, Mol. Cell Biol.23 (2003) 791–803.

[68] C.J. Norbury, B. Zhivotovsky, DNA damage-induced apoptosis,Oncogene 23 (2004) 2797–2808.

[69] B.D. Jeffy, E.U. Schultz, O. Selmin, J.M. Gudas, G.T. Bowden, D.Romagnolo, Inhibition of BRCA-1 expression by benzo[a]pyreneand its diol epoxide, Mol. Carcinogenesis 26 (1999) 100–118.

[70] Y.W. Kwon, S. Ueda, M. Ueno, J. Yodoi, H. Masutani, Mechanismof p53-dependent apoptosis induced by 3-methylcholanthrene:involvement of p53 phosphorylation and p38 MAPK, J. Biol.Chem. 277 (2002) 1837–1844.

[71] S. Chen, N. Nguyen, K. Tamura, M. Karin, R.H. Tukey,

[

[

[

[

58] Y.W. Tian, P.G. Smith, G.C. Yeoh, The oval-shaped cell as a candi-date for a liver stem cell in embryonic, neonatal and precancerousliver: identification based on morphology and immunohistochem-ical staining for albumin and pyruvate kinase isoenzyme expres-sion, Histochem. Cell Biol. 107 (1997) 243–250.

59] B. Knight, G.C. Yeoh, K.L. Husk, T. Ly, L.J. Abraham, C. Yu, J.A.Rhim, N. Fausto, Impaired preneoplastic changes and liver tumorformation in tumor necrosis factor receptor type 1 knockout mice,J. Exp. Med. 192 (2000) 1809–1818.

60] S.R. Kondraganti, P. Fernandez-Salguero, F.J. Gonzalez, K.S.Ramos, W. Jiang, B. Moorthy, Polycyclic aromatic hydrocarbon-inducible DNA adducts: evidence by 32P-postlabeling and useof knockout mice for Ah receptor-independent mechanisms ofmetabolic activation in vivo, Int. J. Cancer 103 (2003) 5–11.

61] T. Farkasova, A. Gabelova, D. Slamenova, Induction of micronu-clei by 7H-dibenzo[c,g]carbazole and its tissue specific deriva-tives in Chinese hamster V79MZh1A1 cells, Mutat. Res. 491(2001) 87–96.

62] N. Hatanaka, H. Yamazaki, R. Kizu, K. Hayakawa, Y. Aoki, M.Iwanari, M. Nakajima, T. Yokoi, Induction of cytochrome P4501B1 in lung, liver and kidney of rats exposed to diesel exhaust,Carcinogenesis 22 (2001) 2033–2038.

The role of the Ah receptor and p38 in benzo[a]pyrene-7,8-dihydrodiol and benzo[a]pyrene-7,8-dihydrodiol-9,10-epoxide-induced apoptosis, J. Biol. Chem. 278 (2003) 19526–19533.

72] B. Binkova, Y. Giguere, P. Rossner Jr., M. Dostal, R.J. Sram,The effect of dibenzo[a,1]pyrene and benzo[a]pyrene on humandiploid lung fibroblasts: the induction of DNA adducts, expressionof p53 and p21(WAF1) proteins and cell cycle distribution, Mutat.Res. 471 (2000) 57–70.

73] F. Oesch, A. Schafer, R.J. Wieser, 12-O-tetradecanoylphorbol-13-acetate releases human diploid fibroblasts from contact-dependent inhibition of growth, Carcinogenesis 9 (1988)1319–1322.

74] D. Renault, D. Brault, Y. Lossouarn, O. Perin-Roussel, D. Taras-Valero, F. Perin, V. Thybaud, Kinetics of DNA adduct formationand removal in mouse hepatocytes following in vivo exposureto 5,9-dimethyldibenzo[c,g]carbazole, Carcinogenesis 21 (2000)289–294.

75] M. Ramet, K. Castren, K. Jarvinen, K. Pekkala, T. Turpeenniemi-Hujanen, Y. Soini, P. Paakko, K. Vahakangas, p53 proteinexpression is correlated with benzo[a]pyrene-DNA adductsin carcinoma cell lines, Carcinogenesis 16 (1995) 2117–2124.