2-Antioxidant Response Element

Jan 29, 2013 - Their DNA binding site is termed “the antioxidant response element” (ARE; also known as “the electrophile response element” or ...
1 downloads 0 Views 4MB Size
Article pubs.acs.org/crt

Bisphenol A Activates the Nrf1/2-Antioxidant Response Element Pathway in HEK 293 Cells Nikolai L. Chepelev,† Mutiat I. Enikanolaiye,† Leonid L. Chepelev,† Abdulrahman Almohaisen,† QiXuan Chen,§ Kylie A. Scoggan,§,∥ Melanie C. Coughlan,⊥ Xu-Liang Cao,# Xiaolei Jin,⊥ and William G. Willmore*,†,‡ †

Department of Biology, and ‡Institute of Biochemistry, Carleton University, 1125 Colonel By Drive, K1S 5B6 Ottawa, Ontario, Canada § Nutrition Research Division, Food Directorate, HPFB, Health Canada, Ottawa, Ontario K1A OK9, Canada ∥ Department of Biochemistry, Microbiology and Immunology, University of Ottawa, Ottawa, Ontario K1H 8M5, Canada ⊥ Toxicology Research Division, Food Directorate, HPFB, Health Canada, Ottawa, Ontario K1A OL2, Canada # Food Research Division, Food Directorate, HPFB, Health Canada, Ottawa, Ontario K1A OL2, Canada ABSTRACT: Bisphenol A (BPA) is used in the production of polycarbonate plastics and epoxy resins for baby bottles, liners of canned food, and many other consumer products. Previously, BPA has been shown to reduce the activity of several antioxidant enzymes, which may contribute to oxidative stress. However, the underlying mechanism of the BPAmediated effect upon antioxidant enzyme activity is unknown. Antioxidant and phase II metabolizing enzymes protect cells from oxidative stress and are transcriptionally activated by Nrf1 and Nrf2 factors through their cisregulatory antioxidant response elements (AREs). In this work, we have assessed the effect of BPA on the Nrf1/2-ARE pathway in cultured human embryonic kidney (HEK) 293 cells. Surprisingly, glutathione and reactive oxygen species (ROS) assays revealed that BPA application created a more reduced intracellular environment in cultured HEK 293 cells. Furthermore, BPA increased the transactivation activity of ectopic Nrf1 and Nrf2 and increased the expression of ARE-target genes ho-1 and nqo1 at high (100−200 μM) BPA concentrations only. Our study suggests that BPA activates the Nrf1/2-ARE pathway at high (>10 μM) micromolar concentrations.



Asn/Ser/Thr-rich (NST) domain,6,7 and proteolysis-mediated translocation from the ER to the nucleus.8 It is thought to be negatively controlled by beta-TrCP and by ER-associated degradation (ERAD).9 Nrf1 and Nrf2 control a number of overlapping target genes, but a specific set of Nrf1-controlled genes has been described.2 BPA has been reported to reduce the activities of catalase and glutathione peroxidase in cultured cells,10 mice,11 and rats,12 resulting in oxidative stress conditions. Other pro-oxidant effects of BPA include DNA breakage in MCF-7 cells.13 In this article, we describe the effects of BPA on the ARE/EpRE pathway in human embryonic kidney (HEK 293) cells. Our data suggests that BPA activates the Nrf1/2-ARE pathway, as reflected by increased Nrf1 transactivation activity and gene expression of ARE target genes ho-1 and nqo1.

INTRODUCTION Nrf (nuclear factor-erythroid 2 p45 subunit-related factor) 1 and 2 belong to the cap’n’collar (CNC) subfamily of basic leucine zipper (bZIP) transcription factors. The CNC-bZIP factors bind to small Maf proteins prior to DNA-binding to either activate (Nrf1, Nrf2, Nrf3, and p45NFE2) or repress (Bach 1 and Bach 2) gene expression. Their DNA binding site is termed “the antioxidant response element” (ARE; also known as “the electrophile response element” or EpRE). The ARE-controlled genes include the following phase II metabolizing and antioxidant enzymes: heme oxygenase-1 (ho-1), NAD(P)H dehydrogenase quinone 1 (nqo1), the glutamatecysteine ligase regulatory (gclm) and catalytic (gclc) subunits, catalase (cat), and glutathione peroxidases (gpx).1,2 Nrf2 is a master regulator of oxidative stress-inducible gene expression.3 Similarly, Nrf1 is essential for cell survival under oxidative stress conditions.4 Despite the fact that Nrf1 can be as important to the antioxidant defense and to human health as Nrf2 is, Nrf1 regulation has been studied to a significantly lesser extent than Nrf2.5 Nrf1 is thought to be controlled by its Nterminal endoplasmic reticulum (ER) targeting sequence anchoring Nrf1 to the ER membrane,6 glycosylation at its © 2013 American Chemical Society



MATERIALS AND METHODS

Plasmids. Firefly luciferase reporter with three AREs from chicken β-globin enhancer (3× ARE-luciferase) was a kind gift of Dr. Masayuki Received: January 25, 2013 Published: January 29, 2013 498

dx.doi.org/10.1021/tx400036v | Chem. Res. Toxicol. 2013, 26, 498−506

Chemical Research in Toxicology

Article

Table 1. Enhanced Expression of the Nrf1/2-ARE Pathway Regulatory and Target Genes at Different BPA Dosesa gene name

TaqMan assay

nqo1 ho-1 nrf1 nrf 2

Hs00168547_m1 Hs00157965_m1 Hs00231457_m1 Hs00975960_m1

BPA dose 0 μM 1.00 1.00 1.00 1.00

± ± ± ±

0.19 0.12 0.16 0.07

a a a a

10 μM

100 μM

1.07 ± 0.16 a 1.04 ± 0.25 ab 1.08 ± 0.35 a 1.14 ± 0.08 b

1.20 ± 0.20 a 1.39 ± 0.28 b 1.18 ± 0.32 a 1.21 ± 0.18 b

200 μM 1.88 2.23 1.69 2.04

± ± ± ±

0.48 0.44 0.23 0.22

p-value (one-way ANOVA) b c b c

0.000112 0.000001 0.001514 0.000001*

a

Relative expression levels of the genes were measured by RT-qPCR analysis from total RNA isolated from HEK 293 cells treated with the above doses of BPA for 24 h and normalized to GAPDH (TaqMan primer Hs 99999905_m1) expression. Values are presented as relative expression means ± SD (n = 6). Means in a row not sharing a lowercase letter are significantly different, p < 0.05 (Duncan’s test, one-way ANOVA). * Logarithmically transformed prior to statistical analysis. Values, significantly different from the control are presented in bold font.

Yamamoto.14 The human Nrf1 gene, obtained from the mammalian gene collection (accession number BC010623), was inserted into modified pCR3.1 using NdeI and EcoRI and introduced into CMV-5aFLAG (Sigma-Aldrich) using EcoRI and Kpn1 to generate Cterminally FLAG-tagged Nrf1. N-Terminally tagged Nrf2 (pCDNA3myc3-Nrf2) was from Addgene (Addgene plasmid 21555).15 The GCLC- and GCLM-ARE-luciferase reporter plasmids (-3802GCLC 5′luc) were a kind gift from Professor Dale A. Dickinson, and its construction has been described elsewhere.16 Cell Culture and BPA Treatments. Human embryonic kidney (HEK) 293 cells were grown in DMEM with 10% horse serum and 3% penicillin/streptomycin/antimycotic (10,000 units/mL penicillin G/ 10,000 μg/mL streptomycin sulfate/25 μg/mL Fungizone in 0.85% saline, Invitrogen) in a humidified atmosphere (5% CO2, 37 °C). Cells were treated with the indicated concentrations of BPA for the indicated amounts of time. Shorter treatments were initiated after longer ones so that all time points were harvested simultaneously to ensure similar cell densities. Cell Viability (MTT) Assay. Cells were seeded at a density of 50,000 cells/mL and grown for 48 h and treated with the indicated BPA concentrations for 24 h. Methylthiazolyldiphenyl-tetrazolium bromide (MTT) was then added (0.5 mg/mL) for 2 h, and the absorbance was read at 570 nm with correction at 630 nm. Subcellular Fractionation. The subcellular fractionation involved sequential separation of the subcellular compartments using successively higher centrifugation forces and suitable buffers as described previously.8 Briefly, cells were lysed in the buffer, containing 10 mM Tris-HCl, pH 7.4, 150 mM NaCl, 5 mM EDTA, 250 mM sucrose, 0.5 mM PMSF, and a complete protease inhibitor cocktail tablet (Roche) by passing the samples through a 21 1/2 gauge needle. The lysates were first centrifuged at 1,000g for 10 min at 4 °C, to obtain the nuclear fraction (NUC) in the pellet. The remaining supernatant was then centrifuged at 12,000g for 15 min at 4 °C. The resulting pellet was referred to as the mitochondrial fraction (MITO), and the supernatant was centrifuged at 100,000g for 60 min at 4 °C to yield the cytosolic fraction (CYT) and the endoplasmic reticulum fraction (ER). Gas Chromatography. Following subcellular fractionation,8 samples were extracted, derivatized, and analyzed as described.17 The diester BPA derivative was extracted with isooctane followed by methyl t-butyl ether and analyzed using an Agilent 6890 gas chromatograph (GC) coupled to a 5973 mass selective detector (MSD) in selected ion monitoring mode. The amount of BPA in ng per each fraction was reported. ROS Measurements. Intracellular ROS were detected using the Image-iT LIVE Green Reactive Oxygen Species Kit (Molecular Probes). Cells were grown for 48 h as described above, washed in PBS, incubated in 10 μM of 5-(and-6)-carboxy-2′,7′-dichlorodihydrofluorescein diacetate (carboxy-H2DCFDA; an intracellular ROS probe) for 30 min, washed in PBS, and treated with BPA for 24 h using phenol red-free medium. 2,2′-azobis (2-amidinopropane) dihydrochloride (AAPH), a generator of peroxyl and alkoxyl radicals,18 served as a positive control. Determination of Glutathione Content. Free and total reduced glutathione (GSH) was measured using the DetectX Glutathione

Fluorescent Detection Kit (Arbor Assays). Butathione sulfoxamine (BSO), an inhibitor of GSH synthesis,19 served as a positive control. Electrophoretic Mobility Shift Assay (EMSA). Seven micrograms of nuclear protein was incubated with a biotinylated dsDNA probe containing the ARE of the human glutamate-cysteine ligase modifier subunit (gclm) promoter,4 and subjected to nondenaturing electrophoresis according to the manufacturer’s protocol (Panomics). Protein−DNA complexes were transferred onto an Amersham Hybond-N+ membrane (GE Healthcare) and visualized using streptavidin-HRP, enhanced chemiluminescence substrate (Millipore), and Kodak X-Omat blue film (Perkin-Elmer). The film was scanned using a CanoScan LIDE 80 scanner (Canon). Band density was quantified by AlphaEaseFC software, version 3.1.2 (Alpha Innotech). Relative band density was obtained by setting the arbitrary intensity units, provided by the software at 100% for controls. Transient Transfections and Luciferase Assays. Approximately 150,000 cells/mL, seeded in 6-well tissue culture plates and grown for 24 h, were transiently transfected with Lipofectamine 2000 following the manufacturer’s protocol (Invitrogen). Per well, 0.08 μg of βgalactosidase, 3.2 μg of 3× ARE-luciferase, and 1.2 μg of Nrf1-FLAG, myc-Nrf2, or pCR3.1 (empty vector) plasmids were added. Twentyfour hours later, cells were treated with BPA for 24 h and harvested. Firefly luciferase reporter assays were performed on cell lysates using a FLUOstar OPTIMA (BMG Labtech) microplate reader. The results were normalized to β-galactosidase activity, in which lysates were incubated in 0.2 mg/mL of chlorophenol red-β-D-galactopyranoside for 5 to 10 min, and the absorbance was measured at 580 nm.20 Total RNA Isolation and Real-Time Quantitative PCR (RealTime qPCR). Total RNA from frozen cell pellets was extracted using the RNeasy Mini Kit (Qiagen). Procedures for cDNA synthesis and real-time qPCR were as previously described.21 Briefly, 1 μg of total RNA was reverse-transcribed to synthesize cDNA with the Retroscript Kit (Applied Biosystems/Ambion). Real-time qPCR was performed on a Mx4000 Multiplex Quantitative PCR System using TaqMan Gene Expression Assays (Applied Biosystems; see Table 1 for assay information). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) expression was unaffected by BPA treatments (data not shown) and was considered a valid housekeeping gene. The fold change in expression for the gene-of-interest relative to GAPDH (gene-ofinterest/GAPDH) was calculated and compared to the control group (set as 1.0). Western Blot Analysis. Harvested cells were lysed in a buffer containing 20 mM HEPES (pH 7.9), 420 mM NaCl, 1.5 mM MgCl2, 0.2 mM EDTA, 25% glycerol, 0.5 mM DTT, and 0.5 mM PMSF and the complete protease inhibitor cocktail tablet (Roche). Twenty to thirty micrograms of total cell protein was loaded on a 10 or 12% SDS−PAGE, run at 120 to 150 V for 1.5 to 2 h, and transferred onto a Immobilon PVDF membrane (Millipore). Membranes were incubated in 5% nonfat milk in TBST for 1 h. All primary antibodies (mouse antihuman Nrf1, Nrf2, HO-1, Nqo1, actin, and histone H2B (Santa Cruz Biotechnology)) and HRP-labeled antimouse (DAKO Cytomation) or antirabbit (Santa Cruz Biotechnology) IgG secondary antibody were used at 1:1,000 and 1:2,000 dilutions, respectively. Membranes were also stained with Ponceau S red as the loading control. Band density was quantified as described above for EMSA. 499

dx.doi.org/10.1021/tx400036v | Chem. Res. Toxicol. 2013, 26, 498−506

Chemical Research in Toxicology

Article

Figure 1. Effect of BPA on cell viability and intracellular redox status. HEK 293A cells were treated with the indicated BPA concentrations for 24 h, and cell viability (A), subcellular BPA content in ng per each fraction (B), ROS generation (C), total glutathione content (micromolar) per sample (D), and the GSH/GSSG ratio (E) were assessed as described in Materials and Methods. Results are the means ± SEM (n ≥ 3). An asterisk (*) represents a significant difference from controls (Student’s paired t test, p < 0.05). TOT, total cell lysate; NUC, nuclear fraction; MITO, mitochondrial fraction; ER, endoplasmic reticulum fraction CYT, cytoplasmic fraction.



Statistical Analyses. Data points were considered statistically significant compared to the control when the p-value of paired Student’s t test was lower than 0.05 (p < 0.05). Data are shown as the means of at least three independent experiments ± standard error of the mean (SEM). For real-time qPCR, all data were evaluated for equality of variance prior to analysis and tested with one-way ANOVA followed by posthoc Duncan’s multiple range test. All statistical analyses were performed using SigmaPlot 11.0 (Systat Software, Inc. San Jose, CA, USA).

RESULTS First, we wanted to investigate the effect of BPA on cell viability. The MTT assay results indicated that BPA is not cytotoxic and, instead, promotes cell proliferation at high (50 to 100 μM for 24 h) doses (Figure 1A). We wanted to test the hypothesis of Ooe and co-workers22 that the compounds with a BPA-like structure are partitioned into the mitochondria, and we employed subcellular fractionation followed by gas 500

dx.doi.org/10.1021/tx400036v | Chem. Res. Toxicol. 2013, 26, 498−506

Chemical Research in Toxicology

Article

chromatography to look at the relative levels of BPA in different organelles. According to the data in Figure 1B, BPA was mostly found in the mitochondrial and cytosolic fractions. Typically, the purity of each fraction obtained during the subcellular fractionation is assessed using specific antibodies against marker proteins, present in a specific subcellular fraction, which often shows some cross-contamination between the fractions.8 Given clear difference between the BPA levels even in the adjacent (e.g., NUC and MITO; ER and CYT in Figure 1B) fractions, we thought that the BPA contamination between the fractions was unlikely in our case. In addition, BPA treatment reduced the basal rate of ROS production by almost 2-fold (Figure 1C). This effect was already evident after 10 h (data not shown). We considered a possibility that antioxidant-rich serum could mask the production of ROS by BPA; however, the same trend (lowered ROS signal) was noticed even in the cells cultured without serum (data not shown). Similar results were obtained for the GSH/GSSG ratio, an index of intracellular redox balance. Thus, at 100 μM BPA, the ratio was about 30% higher compared to that of the controls (Figure 1D and E). Further, we employed EMSAs (gel-shift assays) to look at protein binding to a typical ARE sequence. As seen in Figure 2A, aside from the free probe, three distinct bands were present in our EMSA gels. As these were depleted with the unlabeled (cold) probe (Figure 2A) or with anti-Nrf1, -Bach1, or Nrf2 antibodies (data not shown), they represent specific protein binding to the ARE. The effect of BPA on DNA binding to the labeled ARE sequence was very subtle, reducing the band density of the ARE-protein binding to 60% of the control at 200 μM BPA. The ARE−protein complexes seen in our EMSA assays have been found to contain relatively high levels of Bach1, compared to Nrf1 and Nrf2 (data not shown); Bach1 is an inhibitor of the Nrf1/2-ARE pathway, whose removal stimulates the binding of Nrf2 to the ho-1 ARE.23 While EMSAs are informative in terms of the overall understanding of factors binding to a given DNA sequence, such a binding event can either have a stimulatory or an inhibitory effect on the transcription of downstream genes. To this end, we employed a firefly luciferase reporter harboring three AREs upstream of the luciferase gene (3× AREluciferase), which was cotransfected with pCR3.1 (empty vector), Nrf1-FLAG, or myc-Nrf2, and the transfected cells were treated with BPA. Nrf1-FLAG activity was significantly higher at 100 μM BPA treatment compared to that with the control (Figure 2B). We used the oxidative stressor tertbutylhydroquinone (tBHQ), a compound known to activate Nrf2,24 as a positive control. This analysis showed the increase of about 30% compared to the control for the Nrf2/AREluciferase reporter signal at 200 μM BPA (Figure 3B). These data suggest that BPA acts on both Nrf1 and Nrf2 to activate their transactivational activity, which is consistent with the BPA-provoked increase in the expression of the Nrf1/2-ARE controlled genes measured by real-time qPCR as described below. Next, we analyzed the expression of typical ARE-Nrf1/2 target genes upon BPA treatment using real-time qPCR. One hundred micromolar BPA increased mRNA levels of ho-1 by 1.39-fold (p < 0.05, Table 1), and 200 μM BPA treatment increased mRNA levels of both ho-1 and nqo1 by 2.23- and 1.88-fold, respectively (p < 0.05, Table 1), when compared to all other doses. A similar, stimulatory effect of 200 μM BPA was noted for both nrf1 and nrf 2.

Figure 2. BPA alters protein binding to the ARE consensus sequence probe. (A) EMSAs were run on the nuclear extracts obtained from HEK 293 cells treated with 0, 100, or 200 μM BPA as described in Materials and Methods. A lane without protein extract (probe only) and a lane with an excess of unlabeled extract (cold probe) were used to determine the mobility of the free probe and the specific AREprotein complexes, respectively. The location of the free probe and the specific ARE-protein bands are marked with brackets. Two representative results of four independent experiments are shown. (B) The band densitometry was performed on the specific AREprotein band indicated in A and is graphed here as the means ± SEM (n = 4) of the percent of the control (0 μM BPA) ARE-protein band. An asterisk (*) indicates that the data are statistically significantly different from controls (p < 0.05).

To confirm the real-time qPCR results and to explore the possibility that BPA activates the ARE pathway through the alteration of Nrf1 and Nrf2 protein expression, we used immunoblotting to determine the relative expression levels of both regulatory (Nrf1 and Nrf2) and target (Nqo1 and HO-1) proteins of the ARE-Nrf1/2 pathway. Counterintuitively, the expression of the HO-1 monomer was markedly inhibited at high BPA doses (Figure 4A and B) despite over 2-fold increase in the mRNA level of ho-1 (Table 1). At the same time, we observed the increase in the band, migrating as an approximately 51-kDa polypeptide (Figure 4A and B). We think that the BPA-inducible band, revealed upon HO-1 immunostaining represents the HO-1 dimer, which is consistent with the HO-1 dimerization reported recently,25 but more experiments are required to support this claim. No significant change of the protein expression was noted for Nqo1 (Figure 4C and D). Two major bands were observed for Nrf1, corresponding to the reported 95 and 23 kDa forms of Nrf1 (Figure 5A). The 95 kDa form (p95) is thought to represent the active, nuclear and nonglycosylated form of Nrf1,6,7 while the 23 kDa band (p23) is thought to represent the product of Nrf1 cleavage near the N-terminus.41 The expression of both Nrf1 forms was unaltered 501

dx.doi.org/10.1021/tx400036v | Chem. Res. Toxicol. 2013, 26, 498−506

Chemical Research in Toxicology

Article

Figure 3. BPA activates the transactivation activity of exogenous Nrf1 and Nrf2 in HEK 293 cells. HEK 293 cells were transiently transfected with Nrf1 (A), Nrf2 (B), or pCR3.1 (transfection control) along with 3× ARE-luciferase (B) or GCLC-ARE-luciferase (C) reporter plasmids and β-galactosidase (transfection efficiency control) and treated with 0, 10, 100, or 200 μM BPA for 24 h. The cells were then harvested, lysed, and analyzed using of the luciferase reporter assays as described in the Materials and Methods section. As a positive control for Nrf2 induction, cells transfected with Nrf2-FLAG or pCR3.1 were treated with dimethylsulfoxide (DMSO, vehicle control) or 100 μM tBHQ, a known inducer of the Nrf2-mediated ARE-driven gene expression.24 The means ± SEM (n ≥ 3) are shown. An asterisk (*) indicates that the data are statistically significantly different from controls (p < 0.05).

by BPA treatment (Figure 5B). For Nrf2, we observed increased stability of a polypeptide, migrating at approximately 100 kDa (Figure 5A).

Figure 4. Protein expression of the Nrf1/2-ARE pathway target proteins HO-1 and Nqo1 as a result of BPA treatment. HEK 293 cells were treated with 0, 100, and 200 μM BPA for 24 h, and Western blots (A) were performed as described in Materials and Methods. For semiquantitative densitometric analysis (B), band density was normalized to the density of actin bands (loading control). The means ± SEM (n ≥ 3) are shown. An asterisk (*) indicates that the data are statistically significantly different from controls (p < 0.05).



DISCUSSION The BPA-induced proliferation of HEK 293 cells, observed in this study, could be explained by the effect of the Nrf1/2-ARE pathway activation on cell proliferation. For instance, constitutive expression of Nrf2 in human lung carcinoma cells26 may equip the cells with higher levels of the antioxidant enzymes HO-1 and Nqo1, favoring their unrestrained proliferation.26 Given the activation of the Nrf1/2-ARE pathway observed in our study, it is plausible to suggest that BPA acted on the pathway to favor cell proliferation. A similar result was reported by Oh and Lim in Chang liver cells,10 but the exact mechanisms of BPA-induced cell proliferation and the contributions of Nrf1 and Nrf2 to the process await further investigation.

The fact that BPA decreased the basal rate of ROS production is contrary to previous publications.10−12,22 This effect of BPA is in line with the reported antioxidant effects of estrogens, phytoestrogens, and polyphenols.27 It has been shown that these compounds can activate antioxidant defenses through the MAPK-Nrf2-Keap1 pathway. Similar to our RTqPCR data, microarray analysis of gene expression in human fetal lung fibroblasts suggested an upregulation of Nrf1/2502

dx.doi.org/10.1021/tx400036v | Chem. Res. Toxicol. 2013, 26, 498−506

Chemical Research in Toxicology

Article

the inhibition of methacrylate polymerization30 and prevention of iron-stimulated lipid peroxidation in mice.31 An antioxidant is expected to sacrifice an electron, usually from a weak R-H bond to reduce the attacking species, after which an antioxidant would either be regenerated by compounds such as ascorbate or undergo a disproportionation reaction to form unreactive species, terminating the proliferation of oxidative damage. The strength of the sacrificial R-H bond has to be sufficiently low to allow for the quenching of peroxyl radicals (ROO·) through the regeneration of the ROOH bond, the bond dissociation energy (BDE) for which is approximately 88 kcal/mol. For the regeneration reaction to be favorable, the sacrificial R-H BDE has to be below that of the ROO-H bond. It is also desirable for the antioxidant in question to be quenched by cellular antioxidants, such as ascorbate, the BDE for which is 68.5 kcal/mol. It has been demonstrated that a compound would exhibit antioxidant capacity if the BDE of its sacrificial R-H bond falls within the lower part of this window.32 The weakest bond in BPA is the O−H bond, which has a BDE of 83.9 kcal/mol (as calculated using the MLM2 method33), indicating that BPA may be considered to be a potential antioxidant, albeit a weak one. For example, BPA is expected to be a weaker direct antioxidant compared to a known antioxidant resveratrol, whose BDE is 79.8 kcal/mol.33 It has been noted34 that, as a minor route of xenobiotic metabolism, BPA can also potentially be hydroxylated at the ortho position leading to a catechol which can subsequently undergo redox cycling, generating ROS (see Scheme 1A). These factors make BPA unfavorable as a direct antioxidant, meaning that the antioxidant capacity of BPA observed here is most likely due to an indirect effect, such as the activation of the Nrf1/2-ARE pathway described in this study. Our results strongly suggest that BPA induced an adaptive (or compensatory) response leading to more reduced conditions. The effects of BPA on Nrf1/2 ARE binding were subtle (Figure 2A). This might suggest that the activation of the ARENrf1/2 pathway reported in this manuscript does not proceed simply as a result of the increased protein binding to the ARE. What is also important for the ARE-controlled genes is not only the amount but also the identity of the proteins bound to the ARE. Thus, it can be envisioned that if the relative amount of the negative regulators of the ARE pathway could be somehow decreased as a result of the BPA treatment, this would lead to the activation of the pathway. This would be in agreement with the overall 40% decrease in the ARE-protein binding (Figure 2B) in combination with several pieces of evidence of the activation of the Nrf1/2-ARE pathway presented below. As a precedent, heme binding to Bach1, bound to ho-1 ARE under unstressed conditions (normal intracellular iron levels), displaces this inhibitor from the ARE, activating ho-1 transcription to cope with increased heme levels.23 The stimulatory effect of BPA on the Nrf1/2-ARE pathway was evident using Nrf1/2 transient overexpression and luciferase results (Figure 3) as well as the real-time qPCR results, showing higher mRNA content of ho-1 and nqo1 (Table 1). Despite the increase in ho-1 gene expression (Table 1), the protein level of the HO-1 monomer was approximately seven times lower at high (100−200 μM) BPA doses. The discrepancy between mRNA and protein expression of HO-1 during BPA treatment may be explained by HO-1 dimerization, observed clearly as the appearance of a higher molecular weight (∼ 51-kDa) band during cell treatment with 100 and 200 μM

Figure 5. Protein expression of the Nrf1/2-ARE pathway regulatory proteins Nrf1 and Nrf2 as a result of BPA treatment. HEK 293 cells were treated with 200 μM BPA for various times as indicated, harvested, and lysed, and the total cellular lysate was used for Western blot analysis (A) described in Materials and Methods. For semiquantitative densitometric analysis (B), band density was normalized to the density of histone H2B bands (loading control). Both p95 and p23 Nrf1 bands (95 and 23 kDa forms of Nrf1) are shown. The means ± SEM (n = 3) are shown. An asterisk (*) indicates that the data are statistically significantly different from controls (p < 0.05).

mediated stress response pathways by 100 μM BPA (unpublished data). Further studies of the activities of Nrf1/2 downstream targets will reveal if they are involved in BPAspecific responses as well as responses to similar phenolics in general. Halliwell reported some potential artifacts of cell culture studies.28 Among these is culturing cells at oxygen concentrations about 10−15 times greater than in vivo oxygen levels to which cells are exposed, which could increase the rate of ROS generation28 compared to cells treated at lower oxygen concentrations that are more representative of the in vivo oxygen milieu of the cell. Although BPA itself appears to be a rather weak direct antioxidant as described in greater detail below, BPA could act as an indirect antioxidant, affecting the Nrf1/2-ARE pathway to increase the expression of glutathione synthesis enzymes, such as glutamate-cysteine ligase to create more reduced conditions and relieve hyperoxia-mediated oxidative stress. Another interesting possibility is that BPA could reduce the rate of mitochondrial ROS production since it has been suggested to target the mitochondrial membrane and was shown to reduce complex I activity of the mitochondrial electron transport chain (ETC).22 The mitochondrial localization of BPA in our study is in good agreement with the aforementioned hypothesis that BPA targets mitochondrial membranes. Decreasing the flux through the ETC by lower substrate availability or reduced expression of ETC complex I components is thought to reduce the rate of mitochondrial ROS production.29 Known antioxidant actions of BPA include 503

dx.doi.org/10.1021/tx400036v | Chem. Res. Toxicol. 2013, 26, 498−506

Chemical Research in Toxicology

Article

nqo1 gene expression. BPA treatment may exert differential effects upon transcription (through Nrf1/2) by acting through translational (e.g., reduced mRNA stability, storage in polysomes, stress granules and P-bodies37) or post-translational (e.g., protein aggregation or increased proteolytic degradation) mechanisms, which may result in stable protein levels. BPA effects on the overall activity of Nqo1 remain to be investigated. On the basis of the ARE-luciferase reporter results (Figure 3B) and increased Nrf2 gene expression (Table 1) and protein levels (Figure 5) as a result of BPA treatment, it is logical to propose that the increased expression of the Nrf1/2-AREcontrolled genes is the consequence of greater abundance of Nrf2 and thus its increased transactivational activity. It is worthy to note that our experiments did not distinguish between increased stabilization of Nrf2 and its increased gene expression as potential reasons for the increased pool of cellular Nrf2. In addition to the well-known stabilization of Nrf2 protein by decreased degradation, there are a number of factors that induce nrf 2, which BPA may be affecting. These include the aryl hydrocarbon receptor38 and Nrf2 itself.39 As for the BPA-provoked activation of Nrf1 (Figure 3B), which was not accompanied by any apparent change in the Nrf1 protein expression (Figure 5), this is in accord with previous reports, describing the alteration of the transactivational activity of Nrf1 without any accompanying effects of the Nrf1 protein expression.7 It would be tempting to speculate that the increase of the transactivation activity of the transiently overexpressed Nrf1 occurs via some covalent modification, not readily detectable by immunoblotting. This could potentially be phosphorylation since a closely related family member, Nrf2, is activated by this modification.40 Similarly, Zhang and coworkers6 suggested that Nrf1 phosphorylation may weaken the association of Nrf1 with the ER membrane, stimulating its nuclear translocation and, hence, Nrf1 transactivation function. Furthermore, we observed an increase in Nrf1-FLAG activity following phosphatase inhibitor treatment,41 suggesting that the phosphorylation status of Nrf1 or some upstream signaling factor, acting on Nrf1, might be affected by the BPA treatment. We must add that, unlike the case for Nrf2 activation, presented here, the evidence for Nrf1 activation by BPA is limited only to a modest (30%) increase in the transactivation activity of the luciferase reporter when co-overexpressed with Nrf1 (Figure 3A). Therefore, further studies are required to verify Nrf1 activation by BPA by analyzing the expression of Nrf1-specific, ARE-controlled genes, such as metallothioneins 1 and 2,2 for example. For Nrf2, a band of approximately 100 kDa was observed to increase with BPA treatment (Figure 5). The migration of this oxidative stress-inducible band is in stark contrast to the predicted molecular weight of Nrf2 of 66 kDa, but the exact nature of the Nrf2 post-translational modification, reflected by this band is still unknown.42 It has been suggested that the 100 kDa polypeptide represents tetra-ubiquitinated Nrf2,43 Nrf2 covalently bound to actin,44 or Nrf2 alone migrating abnormally.45 Regardless of its nature, the stabilization of this band by BPA (Figure 5B), along with the increased transactivation activity of Nrf2 upon the treatment with BPA (Figure 3B), suggests that BPA also has the potential to somehow stabilize the 100 kDa form of Nrf2. This might be the mechanism responsible for the activation of Nrf2 (Figure 3B), as well as the Nrf2-ARE pathway target genes (Table 1), observed in this study.

Scheme 1. Pro-Oxidant and Antioxidant Properties of BPA and Our Proposed Model for the BPA Effect on the ARENrf1/2 Pathwaya

a

(A) BPA has both pro-oxidant (redox cycling) and antioxidant (H atom donation) potential. Because of its relatively high BDE of 83.9 kcal/mol (see text for more details), BPA is expected to act poorly as a direct antioxidant. (B) BPA treatment results in more reduced (i.e., lower rate of ROS production) intracellular conditions in HEK 293 cells (Figure 1C−E). This could be due to the activation of the ARENrf1/2 pathway (Figure 3) by stabilization of Nrf2 and by some yet unknown modification of Nrf1 (e.g, phosphorylation41) other than its stabilization (Figure 5). Increased ARE-Nrf1/2-dependent gene (Table 1) and protein (Figure 4A and B) expression could enhance cell proliferation (Figure 1A10,26) and reduce the steady-state ROS levels arising from various intracellular sources such as mitochondria through ROS quenching and detoxification by the antioxidant and cytoprotective enzymes. In addition, BPA could reduce intracellular ROS levels as mitochondrial localization of BPA (Figure 1B) could dampen the flux of metabolites through the mitochondrial ETC by interfering with the activity of the complex I of the ETC,22 which is thought to lower the rate of mitochondrial ROS production.29.

BPA (Figure 4A and B), which could be responsible for lowered HO-1 monomer protein expression despite its increased gene expression. The dimerization and oligomerization of HO-1 is thought to be important for the catalytic activity of HO-1, stabilizing HO-1 presumably by rendering it resistant to the ubiquitin-mediated proteasomal degradation.25,35 To our best knowledge, our study is the first to report HO-1 dimerization due to a chemical or any other treatment of cultured mammalian cells. Given the wide interest of the scientific community regarding HO-1 and the multiplicity of clinical applications (e.g., molecular therapy of tissue injury36) resulting from our enhanced knowledge of the HO-1 structure and function, the BPA-induced dimerization of HO-1 merits further investigation. For Nqo1, corresponding changes in the NQO1 protein levels were not seen despite increases in 504

dx.doi.org/10.1021/tx400036v | Chem. Res. Toxicol. 2013, 26, 498−506

Chemical Research in Toxicology

Article

(7) Zhang, Y., Lucocq, J. M., and Hayes, J. D. (2009) The Nrf1 CNC/bZIP protein is a nuclear envelope-bound transcription factor that is activated by t-butyl hydroquinone but not by endoplasmic reticulum stressors. Biochem. J. 418, 293−310. (8) Wang, W., and Chan, J. Y. (2006) Nrf1 is targeted to the endoplasmic reticulum membrane by an N-terminal transmembrane domain. J. Biol. Chem. 281, 19676−19687. (9) Tsuchiya, Y., Morita, M. K., Kim, M., Lemura, S.-I., Natsume, T., Yamamoto, M., and Kobayashi, A. (2011) Dual regulation of the transcriptional activity of Nrf1 by β-TrCP and Hrd1-dependent degradation mechanisms. Mol. Cell. Biol. 31, 4500−4512. (10) Oh, P. S., and Lim, K. T. (2008) Blocking of intracellular ROS production by phytoglycoprotein (30 kDa) causes anti-proliferation in bisphenol A-stimulated Chang liver cells. J. Appl. Toxicol. 28, 749−758. (11) Kabuto, H., Hasuike, S., Minagawa, N., and Shishibori, T. (2003) Effects of bisphenol A on the metabolisms of active oxygen species in mouse tissues. Environ. Res. 93, 31−35. (12) Chitra, K. C., Latchoumycandane, C., and Mathur, P. P. (2003) Induction of oxidative stress by bisphenol A in the epididymal sperm of rats. Toxicology 185, 119−127. (13) Iso, T., Watanabe, T., Iwamoto, T., Shimamoto, A., and Furuichi, Y. (2006) DNA damage caused by bisphenol A and estradiol through estrogenic activity. Biol. Pharm. Bull. 29, 206−210. (14) Igarashi, K., Kataoka, K., Itoh, K., Hayashi, N., Nishizawa, M., and Yamamoto, M. (1994) Regulation of transcription by dimerization of erythroid factor NF-E2 p45 with small Maf proteins. Nature 367, 568−572. (15) Furukawa, M., Xiong, Y. BTB protein Keap1 targets antioxidant transcription factor Nrf2 for ubiquitination by the Cullin 3-Roc1 ligase. Mol. Cell. Biol. 25, 162-167. (16) Mulcahy, R. T., Wartman, M. A., Bailey, H. H., and Gipp, J. J. (1997) Constitutive and beta-naphthoflavone-induced expression of the human gamma-glutamylcysteine synthetase heavy subunit gene is regulated by a distal antioxidant response element/TRE sequence. J. Biol. Chem. 272, 7445−7454. (17) Cao, X.-L., Corriveau, J., Popovic, S., Coughlan, M. C., Chepelev, N., Willmore, W., and Jin, X. (2010) Background bisphenol A in experimental materials and its implication to low dose in vitro study. Chemosphere 81, 817−820. (18) Chepelev, N. L., Bennitz, J. D., Wright, J. S., Smith, J. C., and Willmore, W. G. (2009) Oxidative modification of citrate synthase by peroxyl radicals and protection with novel antioxidants. J. Enzyme Inhib. Med. Chem. 24, 1319−1331. (19) Yim, C. Y., Hibbs, J. B., McGregor, J. R., Galinsky, R. E., and Samlowski, W. E. (1994) Use of N-acetyl cysteine to increase intracellular glutathione during the induction of antitumor responses by IL-2. J. Immunol. 152, 5796−5805. (20) Eustice, D. C., Feldman, P. A., Colberg-Poley, A. M., Buckery, R. M., and Neubauer, R. H. (1991) A sensitive method for the detection of beta-galactosidase in transfected mammalian cells. Biotechniques 11 (739−740), 742−743. (21) Chen, Q., Gruber, H., Swist, E., Pakenham, C., Ratnayake, W. M., and Scoggan, K. A. (2009) Influence of dietary phytosterols and phytostanols on diastolic blood pressure and the expression of blood pressure regulatory genes in SHRSP and WKY inbred rats. Br. J. Nutr. 102, 93−101. (22) Ooe, H., Taira, T., Iguchi-Ariga, S. M., and Ariga, H. (2005) Induction of reactive oxygen species by bisphenol A and abrogation of bisphenol A-induced cell injury by DJ-1. Toxicol. Sci. 88, 114−126. (23) Sun, J., Hoshino, H., Takaku, K., Nakajima, O., Muto, A., Suzuki, H., Tashiro, S., Takahashi, S., Shibahara, S., Alam, J., Taketo, M. M., Yamamoto, M., and Igarashi, K. (2002) Hemoprotein Bach1 regulates enhancer availability of heme oxygenase-1 gene. EMBO J. 21, 5216− 5224. (24) Wang, W., and Jaiswal, A. K. (2006) Nuclear factor Nrf2 and antioxidant response element regulate NRH:quinone oxidoreductase 2 (NQO2) gene expression and antioxidant induction. Free Radical Biol. Med. 40, 1119−1130.

In conclusion, we propose that BPA stimulates the Nrf1/2ARE pathway as shown by increased mRNA levels of ho-1 and nqo1 (Table 1) and increased transactivation activity of Nrf1FLAG (see Scheme 1B). It is conceivable that BPA acted on the mitochondrial ETC to reduce the flow of metabolites through the ETC, reducing the rate of mitochondrial ROS generation and altering the intracellular redox status. Finally, our study suggests that HEK 293 cells can withstand even high micromolar BPA concentrations without any significant cytotoxic effects.



AUTHOR INFORMATION

Corresponding Author

*Phone: (613) 520-2600 ext. 1211. Fax: 613-520-3539. E-mail: [email protected]. Funding

This research was funded by the Chemical Management Plan Research Network of Health Canada on Bisphenol A. N.L.C. and W.G.W. were supported by a CGS D Scholarship and Discovery Grant from the Natural Sciences and Engineering Research Council of Canada (NSERC), respectively. Q.C. was a recipient of a Visiting Postdoctoral Fellowship from NSERC. Notes

The authors declare no competing financial interest.

■ ■

ACKNOWLEDGMENTS X.-L.C. thanks Svetlana Popovic (Health Canada) for sample analysis. ABBREVIATIONS AAPH, 2,2′-azobis (2-amidinopropane) dihydrochloride; ARE, antioxidant response element; BPA, bisphenol A; BSO, butathione sulfoxamine; CNC-bZIP, cap’n’collar-basic leucine zipper; HEK, human embryonic kidney cells; EpRE, electrophile response element; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; GCLC, glutamate-cysteine ligase catalytic subunit; GCLM, glutamate-cysteine ligase regulatory subunit; GSSG, oxidized glutathione; GSH, reduced glutathione; HO-1, heme oxygenase-1; MTT, methylthiazolyldiphenyl-tetrazolium bromide; NQO1, NAD(P)H dehydrogenase quinone 1; Nrf, nuclear factor-erythroid 2 p45 subunit-related factor; ROS, reactive oxygen species



REFERENCES

(1) Zhu, H., Itoh, K., Yamamoto, M., Zweier, J. L., and Li, Y. (2005) Role of Nrf2 signaling in regulation of antioxidants and phase 2 enzymes in cardiac fibroblasts: protection against reactive oxygen and nitrogen species-induced cell injury. FEBS Lett. 579, 3029−3036. (2) Ohtsuji, M., Katsuoka, F., Kobayashi, A., Aburatani, H., Hayes, J. D., and Yamamoto, M. (2008) Nrf1 and Nrf2 play distinct roles in activation of antioxidant response element-dependent genes. J. Biol. Chem. 283, 33554−33562. (3) Li, W., and Kong, A. N. (2009) Molecular mechanisms of Nrf2mediated antioxidant response. Mol. Carcinog. 48, 91−104. (4) Kwong, M., Kan, Y. W., and Chan, J. Y. (1999) The CNC basic leucine zipper factor, Nrf1, is essential for cell survival in response to oxidative stress-inducing agents. J. Biol. Chem. 274, 37491−37498. (5) Sykiotis, G. P., and Bohmann, D. (2010) Stress-activated cap’n’collar transcription factors in aging and human disease. Sci. Signaling 3, 1−22. (6) Zhang, Y., Lucocq, J. M., Yamamoto, M., and Hayes, J. D. (2007) The NHB1 (N-terminal homology box 1) sequence in transcription factor Nrf1 is required to anchor it to the endoplasmic reticulum and also to enable its asparagine-glycosylation. Biochem. J. 408, 161−172. 505

dx.doi.org/10.1021/tx400036v | Chem. Res. Toxicol. 2013, 26, 498−506

Chemical Research in Toxicology

Article

(25) Hwang, H. W., Lee, J. R., Chou, K. Y., Suen, C. S., Hwang, M. J., Chen, C., Shieh, R. C., and Chau, L. Y. (2009) Oligomerization is crucial for the stability and function of heme oxygenase-1 in the endoplasmic reticulum. J. Biol. Chem. 284, 22672−22679. (26) Padmanabhan, B., Tong, K. I., Ohta, T., Nakamura, Y., Scharlock, M., Ohtsuji, M., Kang, M. I., Kobayashi, A., Yokoyama, S., and Yamamoto, M. (2006) Structural basis for defects of Keap1 activity provoked by its point mutations in lung cancer. Mol. Cell 21, 689−700. (27) Mitchell, J. H., Gardner, P. T., McPhail, D. B., Morrice, P. C., Collins, A. R., and Duthie, G. G. (1998) Antioxidant efficacy of phytoestrogens in chemical and biological model systems. Arch. Biochem. Biophys. 360, 142−148. (28) Halliwell, B. (2009) The wanderings of a free radical. Free Radical Biol. Med. 46, 531−542. (29) Semenza, G. (2010) Hypoxia-inducible factor 1: Regulator of mitochondrial metabolism and mediator of ischemic preconditioning. Biochim. Biophys. Acta 1813, 1263−1268. (30) Kadoma, Y., and Fujisawa, S. (2000) Kinetic evaluation of reactivity of bisphenol A derivatives as radical scavengers for methacrylate polymerization. Biomaterials 21, 2125−2130. (31) Kabuto, H., Hasuike, S., Minagawa, N., and Shishibori, T. (2003) Effects of bisphenol A on the metabolisms of active oxygen species in mouse tissues. Environ. Res. 93, 31−35. (32) Hussain, H. H., Babic, G., Durst, T., Wright, J. S., Flueraru, M., Chichirau, A., and Chepelev, L. L. (2003) Development of novel antioxidants: design, synthesis, and reactivity. J. Org. Chem. 68, 7023− 7032. (33) Wright, J. S., Johnson, E. R., and DiLabio, G. A. (2001) Predicting the activity of phenolic antioxidants: Theoretical method, analysis of substituent effects, and application to major families of antioxidants. J. Am. Chem. Soc. 123, 1173−1183. (34) Kovacic, P. (2010) How safe is bisphenol A? Fundamentals of toxicity: metabolism, electron transfer and oxidative stress. Med. Hypotheses 75, 1−4. (35) Kim, H. P., Pae, H. O., Back, S. H., Chung, S. W., Woo, J. M., Son, Y., and Chung, H. T. (2011) Heme oxygenase-1 comes back to endoplasmic reticulum. Biochem. Biophys. Res. Commun. 404, 1−5. (36) Ryter, S. W., and Choi, A. M. (2009) Heme oxygenase-1/carbon monoxide: from metabolism to molecular therapy. Am. J. Respir. Cell Mol. Biol. 41, 251−260. (37) Brengues, M., Teixeira, D., and Parker, R. (2005) Movement of eukaryotic mRNAs between polysomes and cytoplasmic processing bodies. Science 310, 486−489. (38) Miao, W., Hu, L., Scrivens, P. J., and Batist, G. (2005) Transcriptional regulation of NF-E2 p45-related factor (NRF2) expression by the aryl hydrocarbon receptor-xenobiotic response element signaling pathway: direct cross-talk between phase I and II drug-metabolizing enzymes. J. Biol. Chem. 280, 20340−20348. (39) Kwak, M. K., Itoh, K., Yamamoto, M., and Kensler, T. W. (2002) Enhanced expression of the transcription factor Nrf2 by cancer chemopreventive agents: role of antioxidant response element-like sequences in the nrf2 promoter. Mol. Cell. Biol. 22, 2883−2892. (40) Numazawa, S., Ishikawa, M., Yoshida, A., Tanaka, S., and Yoshida, T. (2003) Atypical protein kinase C mediates activation of NF-E2-related factor 2 in response to oxidative stress. Am. J. Physiol. Cell Physiol. 285, C334−C342. (41) Chepelev, N. L., Bennitz, J. D., Huang, T., McBride, S., and Willmore, W. G. (2011) The Nrf1 CNC-bZIP protein is regulated by the proteasome and activated by hypoxia. PLoS One 6, e29167 DOI: 10.1371/journal.pone.0029167. (42) Taguchi, K., Motohashi, H., and Yamamoto, M. (2011) Molecular mechanisms of the Keap1−Nrf2 pathway in stress response and cancer evolution. Genes Cells 16, 123−140. (43) Li, J., Johnson, D., Calkins, M., Wright, L., Svendsen, C., and Johnson, J. (2005) Stabilization of Nrf2 by tBHQ confers protection against oxidative stress-induced cell death in human neural stem cells. Toxicol. Sci. 83, 313−328.

(44) Kang, K. W., Lee, S. J., Park, J. W., and Kim, S. G. (2002) Phosphatidylinositol 3-kinase regulates nuclear translocation of NFE2-related factor 2 through actin rearrangement in response to oxidative stress. Mol. Pharmacol. 62, 1001−1010. (45) Lau, A., Tian, W., Whitman, S. A., and Zhang, D. D. (2013) The predicted molecular weight of Nrf2: It is what it is not. Antioxid. Redox Signaling 18, 91−93.

506

dx.doi.org/10.1021/tx400036v | Chem. Res. Toxicol. 2013, 26, 498−506