Toxicology and Applied Pharmacology xxx (2015) xxx–xxx

YTAAP-13307; No of Pages 10

Contents lists available at ScienceDirect

Toxicology and Applied Pharmacology

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

Metformin inhibits 7,12-dimethylbenz[a]anthracene-induced breastcarcinogenesis and adduct formation in human breast cells by inhibitingthe cytochrome P4501A1/aryl hydrocarbon receptor signaling pathway

Zaid H. Maayah a, Hazem Ghebeh b, Abdulqader A. Alhaider a,c, Ayman O.S. El-Kadi d, Anatoly A. Soshilov e,Michael S. Denison e, Mushtaq Ahmad Ansari a, Hesham M. Korashy a,⁎a Department of Pharmacology and Toxicology, College of Pharmacy, King Saud University, Riyadh 11451, Saudi Arabiab Stem Cell & Tissue Re-Engineering, King Faisal Specialist Hospital and Research Center, Riyadh 11211, Saudi Arabiac Camel Biomedical Research Unit, College of Pharmacy and Medicine, King Saud University, Riyadh 11451, Saudi Arabiad Faculty of Pharmacy & Pharmaceutical Sciences, University of Alberta, Edmonton, Canadae Department of Environmental Toxicology, University of California at Davis, Davis, CA 95616, USA

⁎ Corresponding author at: Department of PharmacolPharmacy, King Saud University, P.O. Box 2457, Riyadh 11 467 7200.

E-mail address: hkorashy@ksu.edu.sa (H.M. Korashy).

http://dx.doi.org/10.1016/j.taap.2015.02.0070041-008X/© 2015 Elsevier Inc. All rights reserved.

Please cite this article as: Maayah, Z.H., et aformation in human breast cells by inhi..., To

a b s t r a c t

a r t i c l e i n f o

Article history:Received 26 October 2014Revised 5 February 2015Accepted 6 February 2015Available online xxxx

Keywords:Breast cancerMCF10A cellsMetforminCytochrome P4501A1DNA damageAhR

Recent studies have established that metformin (MET), an oral anti-diabetic drug, possesses antioxidant activityand is effective against different types of cancer in several carcinogen-induced animalmodels and cell lines. How-ever, whether MET can protect against breast cancer has not been reported before. Therefore, the overall objec-tives of the present study are to elucidate the potential chemopreventive effect of MET in non-cancerous humanbreast MCF10A cells and explore the underlying mechanism involved, specifically the role of cytochromeP4501A1 (CYP1A1)/aryl hydrocarbon receptor (AhR) pathway. Transformation of theMCF10A cells into initiatedbreast cancer cells with DNA adduct formation was conducted using 7,12-dimethylbenz[a]anthracene (DMBA),an AhR ligand. The chemopreventive effect of MET against DMBA-induced breast carcinogenesis was evidencedby the capability ofMET to restore the induction of the mRNA levels of basic excision repair genes, 8-oxoguanineDNA glycosylase (OGG1) and apurinic/apyrimidinic endonuclease1 (APE1), and the level of 8-hydroxy-2-deoxyguanosine (8-OHdG). Interestingly, the inhibition of DMBA-induced DNA adduct formationwas associatedwith proportional decrease in CYP1A1 and in NAD(P)H:quinone oxidoreductase 1 (NQO1) gene expression.Mechanistically, the involvements of AhR and nuclear factor erythroid 2-related factor-2 (Nrf2) in the MET-mediated inhibition of DMBA-induced CYP1A1 and NQO1 gene expression were evidenced by the ability ofMET to inhibit DMBA-induced xenobiotic responsive element and antioxidant responsive element luciferase re-porter gene expressionwhich suggests an AhR- andNrf2-dependent transcriptional control. However, the inabil-ity of MET to bind to AhR suggests that MET is not an AhR ligand. In conclusion, the present work shows a strongevidence that MET inhibits the DMBA-mediated carcinogenicity and adduct formation by inhibiting the expres-sion of CYP1A1 through an AhR ligand-independent mechanism.

© 2015 Elsevier Inc. All rights reserved.

Introduction

Breast cancer is emerged as one of the most widespread and lethalform of cancer. The high prevalence of breast cancer provides a strongrationale for identifying new natural and synthetic cancer chemopre-ventive agents. Statistically, about 5–10% of breast cancer cases areinherited, whereas 85% occurs due to genetic mutations that happenas a result of the aging process and life in general, such as exposure toenvironmental toxicants. Among these environmental toxicants, 7,12-

ogy and Toxicology, College of1451, Saudi Arabia. Fax: +966

l., Metformin inhibits 7,12-dixicol. Appl. Pharmacol. (2015

dimethylbenz[a]anthracene (DMBA) is a common polycyclic aromatichydrocarbon (PAH) and a powerful organ-specific laboratory carcino-gen that is widely used as a chemical carcinogen in rat mammarytumor model (Chidambaram and Baradarajan, 1996; Dias et al., 1999).The carcinogenicity of DMBA is attributed to its ability to disturb the bal-ance between carcinogen-activating genes; such as the cytochromeP4501A1 (CYP1A1) and carcinogen-detoxifying genes; such as theNAD(P)H:quinone oxidoreductase 1 (NQO1).

Several studies on the carcinogenicity and mutagenicity of DMBAand other PAHs have elucidated a significant role for the CYP1A1 induc-tion in bio-activating these PAHs into their ultimate carcinogenic epox-ide and diol-epoxide intermediates, which bind covalently to the DNAforming DNA adduct (Shimada and Fujii-Kuriyama, 2004). The carcino-genic role of CYP1 in DMBA-induced carcinogenesis is further supported

methylbenz[a]anthracene-induced breast carcinogenesis and adduct), http://dx.doi.org/10.1016/j.taap.2015.02.007

2 Z.H. Maayah et al. / Toxicology and Applied Pharmacology xxx (2015) xxx–xxx

by the fact that DMBA induces cancer in wild type, but not cyp1 knock-out, mice (Buters et al., 1999). The biochemical and carcinogenic effectsof DMBA are primarily initiated by binding to and activation of a cyto-solic ligand-activated transcription factor, the aryl hydrocarbon recep-tor (AhR) (Sogawa and Fujii-Kuriyama, 1997). The activated AhR thendissociates from its inhibitory proteins and translocates to the nucleus,where it heterodimerizes with a nuclear transcription factor proteinknown as the AhR nuclear translocator (ARNT) (Whitelaw et al.,1994). The heterodimeric AhR–ARNT complex then binds to specificDNA recognition sequences, GCGTG, within the xenobiotic responsiveelement (XRE) located in the promoter region of all AhR-dependentgene, includingCYP1A1 andNQO1 resulting in the initiation of transcrip-tion and translation process (Denison et al., 1989; Nebert et al., 2004;Korashy and El-Kadi, 2006a).

NQO1, a constitutively expressed cytosolic flavoprotein, plays anessential role in the detoxification of xenobiotics and carcinogenic me-tabolites (Lee and Johnson, 2004; Xu et al., 2005), by catalyzing thetwo-electron reduction of several environmental contaminants thusprotecting cells against various chemical stresses and carcinogenesis(Chen and Kunsch, 2004; Pinaire et al., 2004). The protective role ofNQO1 against DMBA-induced carcinogenesis is supported by the find-ing that NQO1 knockout mice are more susceptible to DMBA-inducedcancer than their wild-type littermates (Long et al., 2001). The tran-scriptional regulation of NQO1 is primarily mediated by the activationof a labile transcriptional factor, nuclear factor erythroid 2-relatedfactor-2 (Nrf2) (Nioi and Hayes, 2004; Hayes et al., 2005). Upon activa-tion, Nrf2 dissociates from Kelch-like ECH associating protein 1 (Keap1)and then translocates to the nucleus, where it dimerizes with a smallMaf protein. The Nrf2–Maf complex then binds to the antioxidant re-sponsive element (ARE) consensus sequence located in the promoterregion of NQO1 gene, resulting in the initiation of the transcription pro-cess (Chen and Kunsch, 2004; Jaiswal, 2004; Nioi and Hayes, 2004).

In the light of the information described above, one of the strategiesfor protecting human cells and tissues from the toxic effects of carcino-genic and cytotoxic metabolites includes the attenuation of the CYP1A1gene expression and/or enhancing the adaptivemechanismsby increas-ing the expression of NQO1 gene. Although several cancer chemopre-ventive agents such as tamoxifen and celecoxib are available andclinically used, serious side effects with their use have been reported,thus, identifying safer drugs are needed. Recently, the oral anti-diabetic agent metformin (MET) has been shown to be effective againstdifferent types of cancer in several carcinogen-induced animal modelsand cell lines (Hosono et al., 2010; Aljada andMousa, 2012). In addition,epidemiological research has established a link between the use of METand a decrease in cancer incidence, in that patients with type II diabeteswho are prescribed with MET have a lower risk of breast cancer, com-pared with patients who do not take MET (Anon, 2012; Col et al.,2012). Moreover, experimental animal studies have demonstrated theability of MET to reduce the growth of tumor xenografts establishedfrombreast cancer cells and to repress the development of breast cancerin transgenic mice.

Although recent studies suggest that METmight be a promising can-didate for the chemoprevention of breast cancer, very few reports haveexamined the molecular mechanism involved. Thus, the current studywas designed to investigate: (a) the possibility that MET protectsagainst DMBA-induced breast carcinogenesis and adduct formation innon-cancerous human breast MCF10A cells and (b) the impact ofCYP1A1/AhR and NQO1/Nrf2 signaling pathways on the protective ef-fect of MET in DMBA-induced carcinogenesis.

Materials and methods

Chemicals and reagents. Metformin (N,N-dimethylimidodicarbonimidicdiamide hydrochloride) and 7,12-dimethylbenz[a]anthracene (DMBA)were obtained from Toronto Research Chemicals (Toronto, ON).7-Ethoxyresorufin, Dulbecco's Modified Eagle's Medium (DMEM),

Please cite this article as: Maayah, Z.H., et al., Metformin inhibits 7,12-diformation in human breast cells by inhi..., Toxicol. Appl. Pharmacol. (2015

anti-goat IgG peroxidase secondary antibody, and 3-(4,5-dimethyl-thiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) were purchasedfrom Sigma Chemical Co. (St. Louis, MO). 2,3,7,8-Tetrachlorodibenzo-p-dioxin, N99% pure, was purchased from Cambridge Isotope Laborato-ries (Woburn, MA). Resorufin was purchased from ICN BiomedicalsCanada (Montreal, QC). TRIzol reagent and Lipofectamine 2000 kitswere purchased from Invitrogen Co. (Grand Island, NY). High CapacitycDNA Reverse Transcription kit and SYBR® Green PCR Master Mixwere purchased from Applied Biosystems (Foster City, CA). Nitrocellu-lose membrane was purchased from Bio-Rad Laboratories (Hercules,CA). Goat polyclonal primary antibodies of CYP1A1 andNQO1were pur-chased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). Chemilu-minescence Western blot detection kits were obtained from GEHealthcare Life Sciences (Piscataway, NJ). Luciferase assay reagentswere obtained from Promega (Madison, WI). All other chemicals werepurchased from Fisher Scientific Co. (Toronto, ON).

Cell culture and treatment. Non-cancerous human breast (MCF10A) andmurine hepatoma (Hepa 1c1c7) cells (American Type Culture Collection,Rockville, MD)weremaintained in DMEMwith phenol red supplement-edwith 10% fetal bovine serum, 200 μML-glutamine, 100 IU/ml penicillinG, and 10 μg/ml streptomycin. In addition,MCF10A cells were alsomain-tainedwith 100 ng/ml epidermal growth factor, 100 ng/ml cholera toxinand 10 μg/ml insulin. The cells were grown in 75-cm2 tissue cultureflasks at 37 °C under a 5% CO2 humidified environment. MET solutionwas prepared fresh just before each experiment and dissolved indistilled water, whereas DMBA was dissolved in dimethyl sulfoxide(DMSO) in which the concentrations did not exceed 0.05% (V/V).

Determination of cell viability. TheMCF10A cell viability was determinedby measuring the capacity of reducing enzymes present in only viablecells to convert 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazoliumbromide (MTT) to colored formazan crystals as described previously(Korashy et al., 2012). Briefly, MCF10A cells were treated for 24 h withvarious concentrations of test compounds, thereafter, media were re-moved and cells were incubated with MTT for 2 h. The color intensityin each well was then measured at a wavelength of 550 nm using anEL 312e 96-well microplate reader, Bio-Tek Instruments Inc. (Winooski,VT). The percentage of cell viability was calculated relative to controlwells designated as 100% viable cells using the following formula:(Atreated) / (Acontrol) × 100%.

RNA extraction and cDNA synthesis. After incubation with the test com-pounds for the specified time periods, total cellular RNA was isolatedusing TRIzol reagent (Invitrogen®), according to the manufacturer's in-structions. RNA quality and quantity were determined by measuringthe absorbance at 260 nm and 260/280 ratio (~2) using a Nanodrop®

8000 spectrophotometer (Thermo Fisher Scientific Inc., Waltham, MA).Thereafter, first strand cDNA synthesis was performed using the High-Capacity cDNA reverse transcription kit (Applied Biosystems), accordingto the manufacturer's instructions and as described previously (Korashyet al., 2011). Briefly, 1.5 μg of total RNA from each samplewas added to amixture of 2.0 μl of 10× reverse transcriptase buffer, 0.8 μl of 25× dNTPmix (100mM), 2.0 μl of 10× reverse transcriptase randomprimers, 1.0 μlof MultiScribe reverse transcriptase, and 3.2 μl of nuclease-free water.The final reaction mixture was kept at 25 °C for 10 min, heated to37 °C for 120 min, heated for 85 °C for 5 min, and finally cooled to 4 °C.

Quantification of mRNA expression by real-time polymerase chain reaction(RT-PCR). Quantitative analysis of specific mRNA expression was per-formedby RT-PCRby subjecting the resulting cDNA to PCR amplificationusing96-well optical reaction plates in theABI 7500 Fast RT-PCR System(Applied Biosystems) (Korashy et al., 2012). The 25-μl reaction mixturecontained 0.1 μl of 10 μM forward primer and 0.1 μl of 10 μM reverseprimer, 12.5 μl of SYBR Green Universal Mastermix, 11.05 μl of

methylbenz[a]anthracene-induced breast carcinogenesis and adduct), http://dx.doi.org/10.1016/j.taap.2015.02.007

3Z.H. Maayah et al. / Toxicology and Applied Pharmacology xxx (2015) xxx–xxx

nuclease-free water, and 1.25 μl of cDNA sample. Human primers forCYP1A1, NQO1, 8-oxoguanine DNA glycosylase (OGG1), apurinic/apyrimidinic endonuclease1 (APE1), and β-ACTIN (Table 1) were syn-thesized and purchased from Integrated DNA technologies (IDT,Coralville, IA). The fold changes in the level of these genes betweentreated and untreated cells were corrected by the levels of β-ACTIN.Assay controls were incorporated onto the same plate, namely, no-template controls to test for the contamination of any assay reagents.The RT-PCR data were analyzed using the relative gene expression(i.e., ΔΔ CT) method (Livak and Schmittgen, 2001) using the followingequation: fold change = 2−Δ(ΔCt), where ΔCt = Ct(target) − Ct(β-ACTIN)and Δ(ΔCt) = ΔCt(treated) − ΔCt(untreated).

Determination of 8-hydroxy-2-deoxyguanosine (8-OHdG) level. The levelsof 8-OhdG, a biomarker for DNAdamage,were determined according to apreviously published method (Martin et al., 2009). Briefly, total genomicDNA fromMCF10A cell lysate was extracted using a commercially avail-able kit from Qiagen Inc. (Valencia, CA) according to the manufacturer'sinstructions. TheDNAquality and purityweremaintained in 260/280 ab-sorbance ratio range of 1.8–2 optical density. The extractedDNAwas thendigested by DNase-1 (1 U/1 μg DNA), and then subjected to the determi-nation of 8-OHdG according to the protocol of the commercially availableELISA Kit from Abcam (Cambridge, UK).

Determination of reactive oxygen species (ROS) production. IntracellularROS level was analyzed fluorometrically by measuring the oxidation ofnon-fluorescent probes 2′,7′-dichlorofluorescein diacetate (DCF-DA)to fluorescent metabolites dichlorofluorescein (DCF), as described pre-viously (Shankar et al., 2003). Briefly, MCF10A cells were incubatedwith test compound for 24 h, thereafter cells were incubated withDCF-DA (5 μM) for 1 h at 37 °C. Themean fluorescence intensity was di-rectly measured at excitation and emission wavelengths of 485 and535 nm, respectively using POLARstar Omega®, BMG LabTech(Offenburg, Germany).

Western blot analysis. Western blot analysis was performed using a pre-viously described method (Korashy and El-Kadi, 2004). Briefly, 25 μg ofprotein, extracted using a previously described method (Korashy andEl-Kadi, 2004), from each treatment group was separated by 10% sodi-um dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis (PAGE),and then electrophoretically transferred to nitrocellulose membrane.Protein blots were then blocked overnight at 4 °C in blocking solutioncontaining 0.15 M sodium chloride, 3 mM potassium chloride, 25 mMTris-base (TBS), 5% skim milk powder, 2% bovine serum albumin, and0.5% Tween-20. After blocking, the blots were washed several timeswith TBS–Tween-20 before being incubated with a primary goat poly-clonal CYP1A1 and NQO1 antibodies overnight at 4 °C in TBS solutioncontaining 0.05% (v/v) Tween-20 and 0.02% sodium azide. Incubationwith a peroxidase-conjugated rabbit anti-goat IgG secondary antibodywas carried out in blocking solution for 1 h at room temperature. Thebands were visualized using the enhanced chemiluminescence methodaccording to themanufacturer's instructions (GE Healthcare, Mississau-ga, ON). The intensity of CYP1A1 and NQO1 protein bands was quanti-fied relative to the signals obtained for glyceraldehyde-3-phosphatedehydrogenase (GAPDH) protein, using an ImageJ® image processing

Table 1Primer sequences used for RT-PCR reactions.

Gene Forward primer Reverse primer

CYP1A1 CCAAACGAGTTCCGGCCT TGCCCAAACCAAAGAGAATGANQO1 GGAGCCTGCGAAGGTCAA TATCTTCGGTACCGGAAGCTGTAPE1 GGTTAACCATGACCGGGAACT TGCCCAAACCAAAGAGAGTGAOGG1 CTCGCCATAGCCATGCTTATC CCTTCAGCTCATTCATGGCAATCβ-ACTIN CCAGATCATGTTTGAGACCTTCAA GTGGTACGACCAGAGGCATACA

Please cite this article as: Maayah, Z.H., et al., Metformin inhibits 7,12-diformation in human breast cells by inhi..., Toxicol. Appl. Pharmacol. (2015

program (National Institutes of Health, Bethesda, MD, http://rsb.info.nih.gov/ij).

Determination of CYP1A1 enzymatic activity. CYP1A1-dependent 7-ethoxyresorufin (7ER) O-deethylase (EROD) activity was performedon intact living MCF10A cells using 7ER as a substrate (Kennedy et al.,1993). Fluorescence measurements were recorded every 5 min for20 min interval at excitation/emission (545 nm/575 nm) using an EL312e 96-well microplate reader, Bio-Tek Instruments Inc. (Winooski,VT). The amount of resorufin formed in each well was determined bycomparison with a standard curve of known concentrations and nor-malized to protein levels determined using a modified fluorescentassay (Lorenzen and Kennedy, 1993). Resorufin formation wasexpressed as fold of change to the control.

Transient transfection and luciferase assay. MCF10A cells were trans-fected with 1.6 μg of the ARE- and XRE-driven luciferase reporter plas-mid pGudLuc 1.1, generously provided by Dr. M.S. Denison (Universityof California at Davis) and Hepa 1c1c7 cells were transfected with1.6 μg of the XRE-driven luciferase reporter plasmid pGudLuc 1.1 in aserumand antibiotic freemediumusing Lipofectamine 2000 reagent ac-cording to the manufacturer's instructions (Invitrogen). The luciferaseenzyme activities were determined using a luciferase reporter assaysystem from Promega (Madison, WI), as described previously(Korashy et al., 2007), and quantified using a TD-20/20 luminometer(Turner BioSystems, Sunnyvale, CA). The emitted light per well was re-ported as a percentage of the control.

Determination of NQO1 enzymatic activity. NQO1 activity was deter-mined by the continuous spectrophotometric assay to quantitate the re-duction of its substrate, 2,6-dichlorophenolindophenol (DCPIP) asdescribed previously (Korashy and El-Kadi, 2006a, 2006b). The rate ofDCPIP reduction was monitored over 1.5 min at 600 nmwith an extinc-tion coefficient (ϵ) of 2.1mM−1 cm−1. TheNQO1activitywas calculatedas the decrease in absorbance per minute per milligram of total proteinof the sample expressed as percentage of the control.

Competitive ligand binding assay. Ligand binding displacement assaywas performed using a hydroxyapatite (HAP) assay as previouslydescribed (Denison et al., 1986) with slight modifications. Untreatedguinea pig cytosolic protein was diluted to 2 mg/ml in MEDG [25 mM3-(N-morpholino)propanesulfonic acid, pH 7.5, 1 mM ethylenedi-aminetetraacetic acid, 1 mM dithiothreitol and 10% (v/v) glycerol]. Ali-quots of 100 μl were incubated with 2 nM [3H]-TCDD (total binding)in the presence and absence of either 200 nM 2,3,7,8-tetrachlo-rodibenzofuran (TCDF) (100 fold excess of competitor, non-specificbinding) or increasing concentrations of MET. All chemical stockswere prepared in DMSO, in which DMSO content in reactions wasadjusted to 2% (v/v) where necessary. After 1.5 h incubation at roomtemperature, reactions were further incubated with 250 μl of hydro-xyapatite suspension for additional 30 min with gentle vortexingevery 10 min. Thereafter, reactions were washed three times with1 ml MEGT buffer [25 mM 3-(N-morpholino)propanesulfonic acid,pH 7.5, 1 mM ethylenediaminetetraacetic acid, 10% (v/v) glycerol and0.5% (v/v) Tween 80]. The HAP pellets were transferred to 4 ml scintil-lation vials, scintillation cocktail was added and reactions were countedin a scintillation counter.

Electrophoretic mobility shift assay (EMSA). Nuclear protein extractswere prepared from Hepa 1c1c7 cells treated for 2 h with 20 nMTCDD in the absence and presence of 8mMMET as described previously(Rogers and Denison, 2002). XRE complementary oligonucleotides, 5-GGAGTTGCGTGAGAAGAGCC-3 and 5-GGCTCTTCTCACGCAACTCC-3,were synthesized, annealed, labeled with γ-32P-ATP at the 5-endusing T4 polynucleotide kinase, and used as a probe for EMSA reactions

methylbenz[a]anthracene-induced breast carcinogenesis and adduct), http://dx.doi.org/10.1016/j.taap.2015.02.007

Fig. 2. Effects ofMET on APE1 and OGG1mRNA expression (A), 8-OHdG (B), and ROS pro-duction (C) inMCF10A cells. (A)MCF10A cellswere treatedwith 2.5 μMDMBA in thepres-ence and absence of different concentrations of MET (2, 4 and 8 mM) for an additional18 h. APE1 and OGG1 mRNA levels were quantified by RT-PCR and normalized to β-ACTIN as a housekeeping gene. Duplicate reactions were performed for each experiment.The values representmean of fold change±SEM. (n=6);+ P b 0.05 comparedwith con-trol (0 mM); *P b 0.05 compared with DMBA. (B) MCF10A cells were treated with 2.5 μMDMBA in the presence and absence of 8 mM MET for an additional 24 h. The level of 8-OHdG was determined using ELISA Kit. (C) MCF10A cells were treated with 2.5 μMDMBA in the presence and absence of 8 mM MET for an additional 24 h. Thereafter, ROSproduction was determined using DCF-DA as a substrate. The values represent mean ±SEM (n= 6); +P b 0.05 compared with control (0mM); *P b 0.05 compared with DMBA.

4 Z.H. Maayah et al. / Toxicology and Applied Pharmacology xxx (2015) xxx–xxx

as described previously (Denison et al., 1989). Aliquots of the nuclearprotein (20mg)were incubated for 30min at room temperature in a re-action mixture (30 μl) containing 25 mM HEPES, pH 7.9, 80 mM KCl,1 mM EDTA, 1 mM dithiothreitol, 10% glycerol (v/v), and 400 ngpoly(dI·dC). Thereafter, ~1 ng (100,000 cpm) [32P]-labeled XREwas in-cubated with the mixture for another 30 min before being separatedthrough a 4% non-denaturing PAGE. The gel was dried at 80 °C for 1 h,and AhR–XRE complexes formed were visualized by autoradiography(Gharavi and El-Kadi, 2005).

Statistical analysis. The comparative analysis of the results from variousexperimental groups with their corresponding controls was performedusing SigmaStat® for Windows (Systat Software, Inc., CA). One-wayanalysis of variance (ANOVA) followed by the Student–Newman–Keuls test was carried out to assess which treatment groups showed asignificant difference from the control group. The differences were con-sidered significant when P b 0.05.

Results

Effect of MET on MCF10A cells viability

To determine the effect of MET on cell viability and hence the max-imum non-toxic concentrations of MET to be utilized in the in vitrostudy, MCF10A cells were incubated for 24 h with wide concentrationrange of MET (0, 1, 2, 4, 8, 16, 32mM). Thereafter, MCF10A cell viabilitywas determined using MTT assay. Fig. 1 shows that all tested MET con-centrations up to 8 mM range did not significantly affect cell prolifera-tion and viability as compared to control (0 mM). However, higherconcentrations 16 and 32 mM significantly decreased cell viability byapproximately 30% and 40%, respectively. Based on these findings,MET concentrations 2, 4, and 8 mM were utilized in all subsequentexperiments.

Effect ofMET onDMBA-mediated induction of APE1 and OGG1mRNA levels

First, we questioned whether the induction of DNA repair genes, amarker for DNA adduct formation, at the mRNA level by DMBA isblocked by MET. Therefore, MCF10A cells were treated with 2.5 μMDMBA in the presence and absence of different concentrations of MET(2, 4 and 8mM) for an additional 18 h. Thereafter, themRNAexpressionlevels of several DNA repair genes such as APE1, OGG1, MUTYH, XRCC1and NEIL were quantified by RT-PCR. Fig. 2A shows that treatment ofMCF10A cells with DMBA significantly induced DNA adduct as evi-denced by increased APE1 and OGG1 mRNA levels by approximately9- and 3-fold, respectively. MET alone did not significantly alter themRNAexpression levels of DNA repair genes at all tested concentrations(data not shown). However, the induction of APE1 and OGG1 mRNAin response to DMBA was completely prevented by MET in a

Fig. 1. Effect of MET on MCF10A cell viability. The effect of various concentrations of METon MCF10A cell viability was determined using MTT assay. Values are presented as per-centage of the control (mean ± SEM, n = 6). +P b 0.05 compared to control (0 mM).

Please cite this article as: Maayah, Z.H., et al., Metformin inhibits 7,12-diformation in human breast cells by inhi..., Toxicol. Appl. Pharmacol. (2015

concentration-dependent manner (Fig. 2A). The maximum inhibi-tion of APE1 and OGG1 (70%) mRNA was observed at the highestconcentration of MET (8 mM).

Effect of MET on DMBA-induced 8-OHdG level

To further confirm the protective effect of MET against DMBA-induced DNA adduct formation, we tested the effect of MET againstDMBA-induced 8-OHdG level, a biomarker for oxidative DNA damage.For this purpose, MCF10A cells were treated for 24 h with 2.5 μMDMBA in the presence and absence of a single concentration of MET8 mM (the MET concentration that showed the highest protection ef-fect), thereafter, 8-OHdG level was determined as described in theMaterials and methods section. Our result showed that DMBA signifi-cantly induced 8-OHdG level by approximately 165% as compare withcontrol level (Fig. 2B). AlthoughMET alone did not show any significant

methylbenz[a]anthracene-induced breast carcinogenesis and adduct), http://dx.doi.org/10.1016/j.taap.2015.02.007

Fig. 4. Effects of MET on XRE luciferase report gene in MCF10A (A) and Hepa 1c1c7(B) cells. MCF10A and Hepa 1c1c7 cells, transiently transfectedwith the XRE-driven lucif-erase reporter gene, were treatedwith 2.5 μMDMBA(A) or 2 nMTCDD (B) in thepresenceand absence of 8mMMET for an additional 24 h. Luciferase activitywasmeasured accord-ing to the manufacturer's instructions. The graph represents the mean ± SEM (n = 4).+P b 0.05 compared to control (0 mM); *P b 0.05 compared with DMBA.

Fig. 3. Effects of MET on CYP1A1 mRNA (A), protein (B), and activity (C) in MCF10A cells.(A)MCF10A cellswere treatedwith 2.5 μMDMBA in the presence and absence of differentconcentrations of MET (2, 4 and 8 mM) for an additional 18 h. CYP1A1 mRNA was quan-tified by RT-PCR and normalized to β-ACTIN as a housekeeping gene. Duplicate reactionswere performed for each experiment. (B) MCF10A cells were treated with 8 mM MET inthe presence and absence of 2.5 μM DMBA for an additional 24 h. CYP1A1 protein levelwas determined by Western blot analysis. The intensity of CYP1A1 protein bands wasquantified relative to the signals obtained for GAPDH protein, using ImageJ®. One of thethree representative experiments is shown. (C) MCF10A cells were treated with 8 mMMET in the presence and absence of 2.5 μMDMBA for an additional 24 h. CYP1A1 enzymeactivity (EROD) was measured in intact living cells using 7ER as a substrate. The valuesrepresent mean of fold change ± SEM (n = 6). +P b 0.05 compared to control (0 mM);*P b 0.05 compared with DMBA.

5Z.H. Maayah et al. / Toxicology and Applied Pharmacology xxx (2015) xxx–xxx

effect, it significantly inhibited the induction of 8-OHdG in response toDMBA by approximately 70% (Fig. 2B).

Effect of MET on DMBA-induced ROS production

To further investigate the protective role of MET against DMBA-induced oxidative stress, MCF10A cells were treated for 24 h with2.5 μM DMBA in the presence and absence of 8 mM MET. Thereafter,ROSproductionwasmeasuredfluorometerically usingDCF-DAas a sub-strate. Fig. 2C shows that treatment of the cells with DMBA significantlyincreased ROS production by approximately 3-fold. AlthoughMET alone

Please cite this article as: Maayah, Z.H., et al., Metformin inhibits 7,12-diformation in human breast cells by inhi..., Toxicol. Appl. Pharmacol. (2015

did not show any effect, it significantly decreased the DMBA-inducedRSO production by approximately 40%.

Effect of MET on DMBA-mediated induction of CYP1A1 mRNA, protein, andactivity levels

To determine the capacity of MET to alter the induction of CYP1A1gene expression by DMBA, human breast epithelial MCF10A cells weretreatedwith 2.5 μMDMBA in the presence and absence of different con-centrations of MET (2, 4 and 8 mM) for an additional 18 h. Thereafter,CYP1A1 mRNA levels were determined by RT-PCR. Fig. 3A shows thatDMBA significantly and markedly induced CYP1A1 mRNAs by approxi-mately 23-fold (Fig. 3A). Although MET alone did not significantly in-duce CYP1A1 mRNA at all tested concentrations (data not shown),pretreatment of MCF10A cells with MET significantly inhibited theDMBA-mediated induction of CYP1A1 mRNAs in a concentration-dependent manner. The maximum inhibitory effect of MET (70%) wasobserved at the highest concentration tested, 8 mM (Fig. 3A).

To further investigatewhether the inhibitory effect ofMET onCYP1A1mRNA is translated into an inhibition on the protein and catalytic activitylevels, MCF10A cells were treatedwith 2.5 μMDMBA in the presence andabsence of a single concentration of MET 8 mM (the concentration thatshowedmaximumeffect) for an additional 24 h. Thereafter, CYP1A1 pro-tein and catalytic activity levels were determined by Western blotanalysis and EROD assay, respectively. Fig. 3B and C show that DMBA sig-nificantly induced CYP1A1 protein and activity levels by approximately2- and 2.5-fold. Although, MET alone did not alter the expression ofCYP1A1, it significantly restored the DMBA-induced CYP1A1 proteinand activity levels by approximately 50 and 40%, respectively.

Effect of MET on DMBA-mediated induction of AhR-dependent reportergene expression

To address the question of whether the inhibitory effect of MET onDMBA-induced CYP1A1 gene expression is mediated through an AhR-

methylbenz[a]anthracene-induced breast carcinogenesis and adduct), http://dx.doi.org/10.1016/j.taap.2015.02.007

6 Z.H. Maayah et al. / Toxicology and Applied Pharmacology xxx (2015) xxx–xxx

dependentmechanism, MCF10A cells were transiently transfected withthe XRE-driven luciferase reporter gene that is only activated by theactivation of AhR before incubated for 24 h with DMBA (2.5 μM)in the presence and absence of MET (8 mM). Thereafter, luciferaseactivity was determined as described in the Materials and methodssection. Fig. 4A shows that DMBA significantly induced XRE-dependentluciferase activity by approximately 3.5-fold. Treatment of the transfectedcellswithMETalone slightly, but not significantly, increased the luciferaseactivity, importantly MET completely inhibited the increase in XRE lucif-erase activity by DMBA to its control level.

To further confirm the ability of MET to inhibit AhR-dependent re-porter gene, Hepa 1c1c7 cells, themost sensitive cell model for studyingAhR activation, was treated for 24 h with TCDD, themost potent AhR li-gand, in the presence and absence of 8mMMET. Thereafter, the lucifer-ase activitywasmeasured using a TD-20/20 luminometer. Fig. 4B showsthat TCDD significantly induced Hepa 1c1c7 XRE-dependent luciferaseactivity by approximately 7-fold. Importantly, MET completely restoredthe induction of XRE-dependent luciferase activity by TCDD to its con-trol level, suggesting an AhR-dependent mechanism.

Effect of MET on DMBA-mediated induction of NQO1 mRNA, protein, andactivity levels

Todetermine the capacity ofMET to alter the expression ofNQO1 genein response to DMBA, MCF10A cells were treated for 18 h with DMBA(2.5 μM) in the presence and absence of different concentrations of MET(2, 4 and 8 mM). Thereafter, NQO1 mRNA levels were determined byRT-PCR. Our results show that DMBA significantly induced NQO1 mRNA

Fig. 5. Effects ofMET onNQO1mRNA (A), protein (B), and activity (C), and ARE luciferase (D) levabsence of different concentrations of MET (2, 4 and 8 mM) for an additional 18 h. NQO1 mRNDuplicate reactionswere performed for each experiment. (B)MCF10A cellswere treatedwith 2.level was determined byWestern blot analysis. The intensity of NQO1 protein bands was quanrepresentative experiments is shown. (C)MCF10A cells were treatedwith 2.5 μMDMBA in the psured using DCPIP as a substrate. (D) MCF10A cells, transiently transfected with the ARE-driven8mMMET for an additional 24 h. Luciferase activity wasmeasured according to themanufacturcontrol (0 mM); *P b 0.05 compared with DMBA.

Please cite this article as: Maayah, Z.H., et al., Metformin inhibits 7,12-diformation in human breast cells by inhi..., Toxicol. Appl. Pharmacol. (2015

level by approximately 2-fold (Fig. 5A). Unexpectedly, pretreatment ofMCF10A cells with MET slightly but significantly decreased the DMBA-mediated induction of NQO1 mRNA by approximately 35% (Fig. 5A).

To further investigate whether the obtained inhibition on DMBA-mediated induction of NQO1 mRNA level by MET is associated with in-hibition at the protein and catalytic activity levels, MCF10A cells weretreated for 24 h with 2.5 μM DMBA in the presence and absence of8 mMMET, thereafter, NQO1 protein and catalytic activity levels weredetermined by Western blot analysis and DCPIP assay, respectively.Fig. 5B and C show that DMBA significantly induced NQO1 protein andcatalytic activity levels by approximately 1.9 and 2.4-fold, respectively.On the other hand, MET alone did not alter the constitutive expressionof NQO1 protein and activity, but significantly and completely restoredthe DMBA-mediated induction of NQO1 to control values, in a mannersimilar to mRNA levels (Fig. 5B and C).

Effect of MET on DMBA-mediated induction of Nrf2-dependent reportergene expression

To investigate the role of Nrf2 in the inhibitory effect of MET onDMBA-induced NQO1 gene expression, MCF10A cells were transientlytransfected with ARE-driven luciferase reporter gene that is only acti-vated by the activation of Nrf2. Thereafter, the cells were treated for24 h with DMBA (2.5 μM) in the presence and absence of MET(8 mM), and luciferase activity was determined as described in theMaterials and methods section. Fig. 5D shows that DMBA significantlyinduced ARE-dependent luciferase activity by approximately 2.7- fold.Treatment of the transfected cells with MET alone slightly, but not

els inMCF10A cells. (A)MCF10A cellswere treatedwith 2.5 μMDMBA in the presence andA levels were quantified by RT-PCR and normalized to β-ACTIN as a housekeeping gene.5 μMDMBA in thepresence and absence of 8mMMET for an additional 24 h. NQO1 proteintified relative to the signals obtained for GAPDH protein, using ImageJ®. One of the threeresence and absence of 8mMMET for an additional 24 h. NQO1 enzyme activity wasmea-luciferase reporter gene, were treated with 2.5 μMDMBA in the presence and absence of

er's instructions. The values representmean of fold change± SEM.+P b 0.05 compared to

methylbenz[a]anthracene-induced breast carcinogenesis and adduct), http://dx.doi.org/10.1016/j.taap.2015.02.007

Fig. 6. Effect of MET on AhR binding (A) and activation (B). (A) Untreated guinea pighepatic cytosol (2 mg/ml) was incubated with 2 nM [3H]-TCDD alone (total binding),2 nM [3H]-TCDD and 200 nMTCDF (100 fold excess of competitor) (non-specific binding),or 2 nM [3H]-TCDD in the presence of increasing concentrations ofMET (4, 8 and 16mM).The AhR binding dissociation was analyzed by the hydroxyapatite assay. The values wereadjusted for non-specific binding and expressed as a percentage of specific binding rela-tive to the absence of a competitor ligand. The values are presented as mean ± SEM,n= 9. *P b 0.05 compared to [3H]-TCDD. (B) Nuclear extracts fromHepa 1c1c7 cells treat-ed with TCDD in the presence and absence of MET 8 mM were incubated with [32P]-la-beled XRE. Thereafter AhR/XRE complex was determined by EMSA and then visualizedby autoradiography. The values are presented asmean±SEM, n=8.+P b 0.05 comparedto control; *P b 0.05 compared with TCDD.

7Z.H. Maayah et al. / Toxicology and Applied Pharmacology xxx (2015) xxx–xxx

significantly, increased the luciferase activity, however MET completelyrestored the induction of ARE-dependent luciferase activity by DMBA toits control level, suggesting an Nrf2-dependent mechanism.

Effect of MET on AhR binding and activation

The ability of MET to increase the AhR-dependent XRE luciferaseactivity (Fig. 4) raises the question of whether MET is an AhR ligand.To address this question, a ligand competition binding assay usinghydroxyapatite was performed. The total binding is the overall bindingof [3H]-TCDD to cytosolic AhR protein. However, part of this binding isnon-specific, i.e. not through the AhR, or not through ligand-bindingcenter of the AhR. To account for this non-specific binding, reactionsare conducted in the presence of 100-fold excess of competitor (TCDF).Therefore, difference between total and non-specific binding is specificbinding of [3H]-TCDD to the AhR. Our results demonstrat that MET atthe concentrations of 4, 8 and 16 mMwere not able to significantly dis-place [3H]-TCDD from its AhR binding site (Fig. 6A) suggesting that METis not a ligand for AhR.

To further test whether MET is able to inhibit the nuclear AhR trans-location and accumulation, nuclear extract of Hepa 1c1c7 cells treatedwith TCDD in the presence and absence of a single concentration MET(8mM)was subjected to EMSA. Our results show that TCDD significant-ly induced nuclear AhR accumulation by approximately 7-fold. Impor-tantly, MET 8 mM completely inhibited the AhR translocation byTCDD (Fig. 6B). Collectively, our data indicate that MET inhibited thetranslocation process rather than inhibiting the binding step in theAhR transduction pathway.

Discussion

The current manuscript provides the evidence that MET exhibits achemopreventive effect duringDMBA-induced initiation of breast carci-nogenesis and adduct formation in human breastMCF10A cells throughthe inhibition of CYP1A1 gene expression, a well-known carcinogen-ac-tivating gene, at the transcriptional level through ligand-independentAhR inhibition.

One of the strategies for protecting human cells and tissues from thetoxic effects of carcinogenic and cytotoxic metabolites includes the at-tenuation of the carcinogen activating genes signaling pathways and/or enhancing the adaptive mechanisms by increasing the expressionof detoxification and antioxidant genes. Accordingly, we hypothesizethat MET exhibits cancer chemopreventive effects by inhibiting the ex-pression of CYP1A1 gene and/or inducing the expression of NQO1 genein human breast MCF10A cells. The MCF10A cell lines are non-cancerous human breast cells that exhibits normal mammary cell mor-phology and thus are preferred in vitro model for studying early eventsin breast carcinogenesis (Zientek-Targosz et al., 2008). The in vitro con-centrations of MET used in the current study (2, 4 and 8 mM) weremaintained within the physiologically relevant dose of MET, and inagreement with several previous studies (Owen et al., 2000;Asensio-Lopez et al., 2011; Zhuang and Miskimins, 2011). However,these concentrations were above the feasible therapeutic plasma levelin human which could be attributed to the excessive concentrations ofinsulin, glucose, FBS, and growth factor in culture media (Wahdan-Alaswad et al., 2013), combined with the presence of oncogenic muta-tion in the cell culture system and variable expression of themembranetransporter OCT1. Each collectivelymay account for the reduced efficacyand elevated concentration of MET required to elicit cellular responsesin vitro to achieve the pharmacological effects ofMET in cell culture sys-tems (Will et al., 2008). In addition, it has been reported that MET accu-mulates in tissues at much higher concentrations than in the blood(Wilcock and Bailey, 1994; Owen et al., 2000), indicating that the con-centrations of MET employed in the current in vitro study (2–8 mM)might be attained during cancer treatment. Therefore, the present re-sults still seem clinically relevant.

Please cite this article as: Maayah, Z.H., et al., Metformin inhibits 7,12-diformation in human breast cells by inhi..., Toxicol. Appl. Pharmacol. (2015

In the current study, the initiation of breast carcinogenesis inMCF10A cells by DMBA, a chemical breast carcinogen, was achievedby increased DNA adduct formation as evidenced by the increase ina) the mRNA expression of basic excision DNA repair (BER) genes,APE1 and OGG1, b) the level of 8-OHdGwhich is a biomarker for the ox-idative DNA damage (Hwang andBowen, 2007), and c) ROS production.Our results are in agreement with previous studies that showed the in-duction of BER genes, APE1 and OGG1, and the level of 8-OHdG in re-sponse to DMBA in vitro and in vivo (Braithwaite et al., 1998; Sahinet al., 2011). Importantly, the chemopreventive effect of MET againstDMBA-induced breast carcinogenesis was evidenced by the capabilityof MET to restore the induction of the mRNA expression of APE1 andOGG1 genes, the level of 8-OHdG, and ROS production. In an agreementwith our findings, tamoxifen, an estrogen receptor modulator, inhibitsestrogen-induced DNA adduct formation and mammary cell tumori-genesis in MCF10A cells (Montano et al., 2007), whereas resveratrol, anaturally occurring polyphenolic compound, attenuates DMBA-inducedDNA damage in MCF10A cells (Leung et al., 2009). In addition, it hasbeen shown that the administration of lycopene and genistein combina-tion, chemopreventive agents, protects against DMBA-induced breast

methylbenz[a]anthracene-induced breast carcinogenesis and adduct), http://dx.doi.org/10.1016/j.taap.2015.02.007

8 Z.H. Maayah et al. / Toxicology and Applied Pharmacology xxx (2015) xxx–xxx

cancer development through the suppression of 8-OHdG levels (Sahinet al., 2011). These results not only indicate a protective effect of METbut also suggest that the reduced chelation of DMBA to DNA appearedto be the major route for protection.

Several studies on the carcinogenicity and mutagenicity of DMBAhave elucidated a significant role for the induction of CYP1A1 inbioactivating this compound into its ultimate carcinogenic epoxideand diol-epoxide intermediates (Shimada and Fujii-Kuriyama, 2004).These ultimate carcinogenic metabolites can bind covalently to DNAforming DNA adducts which can hinder the ability of a cell to carryout its function and can significantly increase the likelihood of tumorformation. Thus, CYP1A1 is considered as a key for DMBA-inducedbreast carcinogenicity. Accordingly, the current study showed for thefirst time the ability of MET to inhibit DMBA-induced CYP1A1 gene ex-pression at mRNA, protein and activity levels. This was associatedwith a proportional decrease in DNA adduct formation. In agreementwith our results, it has been reported before that resveratrol protectsagainst DMBA-induced oxidative DNA damage through CYP1A1 medi-ated effect (Leung et al., 2009). Importantly, the ability ofMET to inhibitDMBA-induced XRE luciferase reporter gene expression, that is onlymediated through AhR, suggests an AhR-dependent transcriptionalcontrol and excludes the possibility of any posttranscriptional mecha-nisms, such as mRNA stability (Pasco et al., 1988).

Modulation of antioxidant and detoxifying NQO1 gene is consideredone of the major line of defense against cytotoxic agents and known toplay essential roles in the detoxification and elimination of activatedcarcinogens during tumor initiation (Sheweita and Tilmisany, 2003;Chen and Kunsch, 2004; Jaiswal, 2004; Nioi and Hayes, 2004). In thisstudy, we showed that MET inhibits DMBA-mediated induction ofNQO1 gene expression at mRNA, protein and activity levels. This inhib-itory effect was mediated through an oxidative stress-dependent tran-scriptional control of NQO1 as evidenced by the inhibitory effect ofMET on the DMBA-induced ARE luciferase reporter gene expression.The inhibition of NQO1 by chemopreventive agents has also been re-ported before (Anwar-Mohamed and El-Kadi, 2009).

The unexpected inhibition inNQO1 gene expression byMET could beattributed, at least in part, to two postulations. First, the ability ofMET toinhibit the AhR/XRE luciferase activity, which partially contributed to

Fig. 7. Proposed mechanism for the chemopreventive effect of MET ag

Please cite this article as: Maayah, Z.H., et al., Metformin inhibits 7,12-diformation in human breast cells by inhi..., Toxicol. Appl. Pharmacol. (2015

the induction of NQO1 gene expression. This postulation is supportedby the fact that both AhR–XRE and Nrf2–ARE pathways play an integralrole in the regulation of NQO1 gene and crosstalk to each other(Radjendirane and Jaiswal, 1999; Miao et al., 2005). It has been furtherfound that the Nrf2 gene expression is directly regulated through AhRactivation, in that the gene promoter of Nrf2 contains at least one func-tional XRE (Miao et al., 2005), and the gene promoter of AhR containsseveral AREs (Ma et al., 2004; Nioi and Hayes, 2004; Shin et al., 2007),and hence NQO1 gene expression is controlled by CYP1A1 activitythrough an AhR-dependent pathway (Marchand et al., 2004). Thesecond postulation is the antioxidant effect of MET against DMBA-induced oxidative stress is through the inhibition of diol epoxidemetabolites formation, ROS production, and oxidative DNA damage,rather than the induction of detoxifying enzymes.

In the light of the information described above, the current resultssuggest a direct evidence for the involvement of AhR in the transcrip-tional regulation of CYP1A1 byMET. This raises the question of whetheror notMET is a ligand for the AhR. Therefore, we examined the ability ofMET to directly bind to AhRprotein using anHAP assay in guinea pig cy-tosol model. Guinea pig cytosol is extensively used to assess the bindingand affinity of ligands to the AhR and exhibited the greatest degree ofAhR transformation in response to AhR ligand. The inability ofMET to sig-nificantly displace [3H]-TCDD from its binding site in guinea pig cytosolicextract, while successfully inhibited the nuclear translocation of AhR byTCDD inHepa1c1c7 cells suggests a ligand-independent inhibitorymech-anism. Ligand-independent AhR inhibition has been reported by severaldrugs and chemicals. For example, harmalol has been shown to inhibitthe AhR-dependent gene expression such as CYP1A1without direct bind-ing to the AhR (El Gendy et al., 2012). Although the exact mechanismsgoverning the ligand-independent inhibition of AhR are still not clear, itcould be postulated that MET interacts with a second binding site onthe AhR rather than the TCDD-binding site (Ciolino et al., 1998). ThisAhR binding site has been proposed for primaquine AhR-dependent in-duction of CYP1A1 although it did not show a significant competitionwith [3H]-TCDD binding site on the AhR (Fontaine et al., 1999).

In conclusion (Fig. 7), the present work provides the mechanisticevidence that MET protects against DMBA-induced breast carcinogene-sis in human breast MCF10A cells, at least in part, via inhibiting the

ainst DMBA-induced breast carcinogenesis and adduct formation.

methylbenz[a]anthracene-induced breast carcinogenesis and adduct), http://dx.doi.org/10.1016/j.taap.2015.02.007

9Z.H. Maayah et al. / Toxicology and Applied Pharmacology xxx (2015) xxx–xxx

expression of CYP1A1, a carcinogen -activating gene, at transcriptionallevel through an AhR-ligand independent mechanism. These resultsare of potential clinical significance to humans in that it uncovers themolecular mechanism involved and could explain the anecdotal evi-dence for the successful use of MET in the prevention of breast cancer.

Conflict of interest statement

There are no financial or other interests with regard to this manu-script that might be construed as a conflict of interest.

Acknowledgments

This project was supported by NSTIP strategic technologies programnumber (12-MED3131-02) in the Kingdom of Saudi Arabia.

References

Aljada, A., Mousa, S.A., 2012. Metformin and neoplasia: implications and indications.Pharmacol. Ther. 133, 108–115.

Anon, 2012. Cancer: lower risk of breast cancer in women with diabetes mellitus treatedwith metformin. Nat. Rev. Endocrinol. 8, 508.

Anwar-Mohamed, A., El-Kadi, A.O., 2009. Down-regulation of the detoxifying enzymeNAD(P)H:quinone oxidoreductase 1 by vanadium in Hepa 1c1c7 cells. Toxicol.Appl. Pharmacol. 236, 261–269.

Asensio-Lopez, M.C., Lax, A., Pascual-Figal, D.A., Valdes, M., Sanchez-Mas, J., 2011. Metfor-min protects against doxorubicin-induced cardiotoxicity: involvement of theadiponectin cardiac system. Free Radic. Biol. Med. 51, 1861–1871.

Braithwaite, E., Wu, X., Wang, Z., 1998. Repair of DNA lesions induced by polycyclic aro-matic hydrocarbons in human cell-free extracts: involvement of two excision repairmechanisms in vitro. Carcinogenesis 19, 1239–1246.

Buters, J.T., Sakai, S., Richter, T., Pineau, T., Alexander, D.L., Savas, U., Doehmer, J., Ward,J.M., Jefcoate, C.R., Gonzalez, F.J., 1999. Cytochrome P450 CYP1B1 determines suscep-tibility to 7, 12-dimethylbenz[a]anthracene-induced lymphomas. Proc. Natl. Acad.Sci. U. S. A. 96, 1977–1982.

Chen, X.L., Kunsch, C., 2004. Induction of cytoprotective genes through Nrf2/antioxidantresponse element pathway: a new therapeutic approach for the treatment of inflam-matory diseases. Curr. Pharm. Des. 10, 879–891.

Chidambaram, N., Baradarajan, A., 1996. Influence of selenium on glutathione and someassociated enzymes in rats with mammary tumor induced by 7,12-dime-thylbenz(a)anthracene. Mol. Cell. Biochem. 156, 101–107.

Ciolino, H.P., Daschner, P.J., Yeh, G.C., 1998. Resveratrol inhibits transcription of CYP1A1in vitro by preventing activation of the aryl hydrocarbon receptor. Cancer Res. 58,5707–5712.

Col, N.F., Ochs, L., Springmann, V., Aragaki, A.K., Chlebowski, R.T., 2012. Metformin andbreast cancer risk: a meta-analysis and critical literature review. Breast Cancer Res.Treat. 135, 639–646.

Denison, M.S., Harper, P.A., Okey, A.B., 1986. Ah receptor for 2,3,7,8-tetrachlorodibenzo-p-dioxin. Codistribution of unoccupied receptor with cytosolic marker enzymes duringfractionation of mouse liver, rat liver and cultured Hepa-1c1 cells. Eur. J. Biochem.155, 223–229.

Denison, M.S., Fisher, J.M., Whitlock Jr., J.P., 1989. Protein–DNA interactions at recognitionsites for the dioxin–Ah receptor complex. J. Biol. Chem. 264, 16478–16482.

Dias, M., Cabrita, S., Sousa, E., Franca, B., Patricio, J., Oliveira, C., 1999. Benign and malig-nant mammary tumors induced by DMBA in female Wistar rats. Eur. J. Gynaecol.Oncol. 20, 285–288.

El Gendy, M.A., Soshilov, A.A., Denison, M.S., El-Kadi, A.O., 2012. Harmaline and harmalolinhibit the carcinogen-activating enzyme CYP1A1 via transcriptional and posttransla-tional mechanisms. Food Chem. Toxicol. 50, 353–362.

Fontaine, F., Delescluse, C., de Sousa, G., Lesca, P., Rahmani, R., 1999. Cytochrome 1A1 in-duction by primaquine in human hepatocytes and HepG2 cells: absence of binding tothe aryl hydrocarbon receptor. Biochem. Pharmacol. 57, 255–262.

Gharavi, N., El-Kadi, A.O., 2005. tert-Butylhydroquinone is a novel aryl hydrocarbon re-ceptor ligand. Drug Metab. Dispos. 33, 365–372.

Hayes, J.D., Flanagan, J.U., Jowsey, I.R., 2005. Glutathione transferases. Annu. Rev.Pharmacol. Toxicol. 45, 51–88.

Hosono, K., Endo, H., Takahashi, H., Sugiyama, M., Uchiyama, T., Suzuki, K., Nozaki, Y.,Yoneda, K., Fujita, K., Yoneda, M., Inamori, M., Tomatsu, A., Chihara, T., Shimpo, K.,Nakagama, H., Nakajima, A., 2010. Metformin suppresses azoxymethane-induced co-lorectal aberrant crypt foci by activating AMP-activated protein kinase. Mol. Carcinog.49, 662–671.

Hwang, E.S., Bowen, P.E., 2007. DNA damage, a biomarker of carcinogenesis: its measure-ment and modulation by diet and environment. Crit. Rev. Food Sci. Nutr. 47, 27–50.

Jaiswal, A.K., 2004. Nrf2 signaling in coordinated activation of antioxidant gene expres-sion. Free Radic. Biol. Med. 36, 1199–1207.

Kennedy, S.W., Lorenzen, A., James, C.A., Collins, B.T., 1993. Ethoxyresorufin-O-deethylaseand porphyrin analysis in chicken embryo hepatocyte cultures with a fluorescencemultiwell plate reader. Anal. Biochem. 211, 102–112.

Please cite this article as: Maayah, Z.H., et al., Metformin inhibits 7,12-diformation in human breast cells by inhi..., Toxicol. Appl. Pharmacol. (2015

Korashy, H.M., El-Kadi, A.O., 2004. Differential effects of mercury, lead and copper on theconstitutive and inducible expression of aryl hydrocarbon receptor (AHR)-regulatedgenes in cultured hepatoma Hepa 1c1c7 cells. Toxicology 201, 153–172.

Korashy, H.M., El-Kadi, A.O., 2006a. Transcriptional regulation of the NAD(P)H:quinoneoxidoreductase 1 and glutathione S-transferase ya genes by mercury, lead, and cop-per. Drug Metab. Dispos. 34, 152–165.

Korashy, H.M., El-Kadi, A.O., 2006b. The role of aryl hydrocarbon receptor in the patho-genesis of cardiovascular diseases. Drug Metab. Rev. 38, 411–450.

Korashy, H.M., Shayeganpour, A., Brocks, D.R., El-Kadi, A.O., 2007. Induction of cyto-chrome P450 1A1 by ketoconazole and itraconazole but not fluconazole in murineand human hepatoma cell lines. Toxicol. Sci. 97, 32–43.

Korashy, H.M., Anwar-Mohamed, A., Soshilov, A.A., Denison, M.S., El-Kadi, A.O., 2011. Thep38MAPK inhibitor SB203580 induces cytochrome P450 1A1 gene expression inmu-rine and human hepatoma cell lines through ligand-dependent aryl hydrocarbon re-ceptor activation. Chem. Res. Toxicol. 24, 1540–1548.

Korashy, H.M., Maayah, Z.H., Abd-Allah, A.R., El-Kadi, A.O., Alhaider, A.A., 2012. Camel milktriggers apoptotic signaling pathways in human hepatoma HepG2 and breast cancerMCF7 cell lines through transcriptionalmechanism. J. Biomed. Biotechnol. 2012, 593195.

Lee, J.M., Johnson, J.A., 2004. An important role of Nrf2–ARE pathway in the cellular de-fense mechanism. J. Biochem. Mol. Biol. 37, 139–143.

Leung, H.Y., Yung, L.H., Shi, G., Lu, A.L., Leung, L.K., 2009. The red wine polyphenol resver-atrol reduces polycyclic aromatic hydrocarbon-induced DNA damage in MCF-10Acells. Br. J. Nutr. 102, 1462–1468.

Livak, K.J., Schmittgen, T.D., 2001. Analysis of relative gene expression data usingreal-time quantitative PCR and the 2(−delta delta C(T)) method. Methods 25,402–408.

Long II, D.J., Waikel, R.L., Wang, X.J., Roop, D.R., Jaiswal, A.K., 2001. NAD(P)H:quinoneoxidoreductase 1 deficiency and increased susceptibility to 7,12-dimethylbenz[a]-an-thracene-induced carcinogenesis in mouse skin. J. Natl. Cancer Inst. 93, 1166–1170.

Lorenzen, A., Kennedy, S.W., 1993. A fluorescence-based protein assay for use with a mi-croplate reader. Anal. Biochem. 214, 346–348.

Ma, Q., Kinneer, K., Bi, Y., Chan, J.Y., Kan, Y.W., 2004. Induction of murine NAD(P)H:qui-none oxidoreductase by 2,3,7,8-tetrachlorodibenzo-p-dioxin requires the CNC (cap‘n’ collar) basic leucine zipper transcription factor Nrf2 (nuclear factor erythroid 2-related factor 2): cross-interaction between AhR (aryl hydrocarbon receptor) andNrf2 signal transduction. Biochem. J. 377, 205–213.

Marchand, A., Barouki, R., Garlatti, M., 2004. Regulation of NAD(P)H:quinone oxidoreduc-tase 1 gene expression by CYP1A1 activity. Mol. Pharmacol. 65, 1029–1037.

Martin, S.A., McCarthy, A., Barber, L.J., Burgess, D.J., Parry, S., Lord, C.J., Ashworth, A., 2009.Methotrexate induces oxidative DNA damage and is selectively lethal to tumour cellswith defects in the DNA mismatch repair gene MSH2. EMBO Mol. Med. 1, 323–337.

Miao, W., Hu, L., Scrivens, P.J., Batist, G., 2005. Transcriptional regulation of NF-E2 p45-related factor (NRF2) expression by the aryl hydrocarbon receptor–xenobioticresponse element signaling pathway: direct cross-talk between phase I and II drug-metabolizing enzymes. J. Biol. Chem. 280, 20340–20348.

Montano, M.M., Chaplin, L.J., Deng, H., Mesia-Vela, S., Gaikwad, N., Zahid, M., Rogan, E.,2007. Protective roles of quinone reductase and tamoxifen against estrogen-induced mammary tumorigenesis. Oncogene 26, 3587–3590.

Nebert, D.W., Dalton, T.P., Okey, A.B., Gonzalez, F.J., 2004. Role of aryl hydrocarbonreceptor-mediated induction of the CYP1 enzymes in environmental toxicity andcancer. J. Biol. Chem. 279, 23847–23850.

Nioi, P., Hayes, J.D., 2004. Contribution of NAD(P)H:quinone oxidoreductase 1 to protec-tion against carcinogenesis, and regulation of its gene by the Nrf2 basic-region leu-cine zipper and the arylhydrocarbon receptor basic helix–loop–helix transcriptionfactors. Mutat. Res. 555, 149–171.

Owen, M.R., Doran, E., Halestrap, A.P., 2000. Evidence that metformin exerts its anti-diabetic effects through inhibition of complex 1 of the mitochondrial respiratorychain. Biochem. J. 348 (Pt 3), 607–614.

Pasco, D.S., Boyum, K.W., Merchant, S.N., Chalberg, S.C., Fagan, J.B., 1988. Transcriptionaland post-transcriptional regulation of the genes encoding cytochromes P-450c andP-450d in vivo and in primary hepatocyte cultures. J. Biol. Chem. 263, 8671–8676.

Pinaire, J.A., Xiao, G.H., Falkner, K.C., Prough, R.A., 2004. Regulation of NAD(P)H:quininoneoxidoreductase by glucocorticoids. Toxicol. Appl. Pharmacol. 199, 344–353.

Radjendirane, V., Jaiswal, A.K., 1999. Antioxidant response element-mediated 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) induction of human NAD(P)H:quinone oxidore-ductase 1 gene expression. Biochem. Pharmacol. 58, 1649–1655.

Rogers, J.M., Denison, M.S., 2002. Analysis of the antiestrogenic activity of 2,3,7,8-tetrachlorodibenzo-p-dioxin in human ovarian carcinoma BG-1 cells. Mol.Pharmacol. 61, 1393–1403.

Sahin, K., Tuzcu, M., Sahin, N., Akdemir, F., Ozercan, I., Bayraktar, S., Kucuk, O., 2011. Inhib-itory effects of combination of lycopene and genistein on 7,12-dimethylbenz(a)anthracene-induced breast cancer in rats. Nutr. Cancer 63, 1279–1286.

Shankar, B., Kumar, S.S., Sainis, K.B., 2003. Generation of reactive oxygen species and ra-diation response in lymphocytes and tumor cells. Radiat. Res. 160, 478–487.

Sheweita, S.A., Tilmisany, A.K., 2003. Cancer and phase II drug-metabolizing enzymes.Curr. Drug Metab. 4, 45–58.

Shimada, T., Fujii-Kuriyama, Y., 2004. Metabolic activation of polycyclic aromatic hydro-carbons to carcinogens by cytochromes P450 1A1 and 1B1. Cancer Sci. 95, 1–6.

Shin, S., Wakabayashi, N., Misra, V., Biswal, S., Lee, G.H., Agoston, E.S., Yamamoto, M.,Kensler, T.W., 2007. NRF2 modulates aryl hydrocarbon receptor signaling: influenceon adipogenesis. Mol. Cell. Biol. 27, 7188–7197.

Sogawa, K., Fujii-Kuriyama, Y., 1997. Ah receptor, a novel ligand-activated transcriptionfactor. J. Biochem. (Tokyo) 122, 1075–1079.

Wahdan-Alaswad, R., Fan, Z., Edgerton, S.M., Liu, B., Deng, X.S., Arnadottir, S.S., Richer, J.K.,Anderson, S.M., Thor, A.D., 2013. Glucose promotes breast cancer aggression and re-duces metformin efficacy. Cell Cycle 12, 3759–3769.

methylbenz[a]anthracene-induced breast carcinogenesis and adduct), http://dx.doi.org/10.1016/j.taap.2015.02.007

10 Z.H. Maayah et al. / Toxicology and Applied Pharmacology xxx (2015) xxx–xxx

Whitelaw, M.L., Gustafsson, J.A., Poellinger, L., 1994. Identification of transactivation andrepression functions of the dioxin receptor and its basic helix–loop–helix/PAS partnerfactor Arnt: inducible versus constitutive modes of regulation. Mol. Cell. Biol. 14,8343–8355.

Wilcock, C., Bailey, C.J., 1994. Accumulation of metformin by tissues of the normal and di-abetic mouse. Xenobiotica 24, 49–57.

Will, Y., Dykens, J.A., Nadanaciva, S., Hirakawa, B., Jamieson, J., Marroquin, L.D., Hynes, J.,Patyna, S., Jessen, B.A., 2008. Effect of the multitargeted tyrosine kinase inhibitorsimatinib, dasatinib, sunitinib, and sorafenib on mitochondrial function in isolatedrat heart mitochondria and H9c2 cells. Toxicol. Sci. 106, 153–161.

Please cite this article as: Maayah, Z.H., et al., Metformin inhibits 7,12-diformation in human breast cells by inhi..., Toxicol. Appl. Pharmacol. (2015

Xu, C., Li, C.Y., Kong, A.N., 2005. Induction of phase I, II and III drug metabolism/transportby xenobiotics. Arch. Pharm. Res. 28, 249–268.

Zhuang, Y., Miskimins, W.K., 2011. Metformin induces both caspase-dependent andpoly(ADP-ribose) polymerase-dependent cell death in breast cancer cells. Mol. Can-cer Res. 9, 603–615.

Zientek-Targosz, H., Kunnev, D., Hawthorn, L., Venkov, M., Matsui, S., Cheney, R.T., Ionov,Y., 2008. Transformation of MCF-10A cells by random mutagenesis with frameshiftmutagen ICR191: a model for identifying candidate breast-tumor suppressors. Mol.Cancer 7, 51.

methylbenz[a]anthracene-induced breast carcinogenesis and adduct), http://dx.doi.org/10.1016/j.taap.2015.02.007