The endocrine disruptors (EDs) are heterogeneous and widely distributed group of hormone-like chemicals, mostly of anthropogenic origin. A common feature of these compounds is their ability to interfere with natural hormones at concentrations below their toxicity threshold (i.e., in the μM range). Consequently, EDs trigger the onset of several pathologies, particularly but not only, related to reproduction and development (1).
Some pesticides, phtalates, and several phenolic compounds, such as alkylphenoles and biphenyls including biphenyl methane (also known as bisphenol A (2,2-bis(4-hydroxyphenyl) propane, BPA), are in the list of EDs (2, 3). BPA is produced at a value of ∼1.7 billion kg per annum worldwide and used in the manufacture of several very ordinary products (4). BPA discharged into the environment is expected to break down fairly rapidly in air, whereas it is more persistent in water (5–7). Thus, environmental BPA pollution is of great impact on human beings as well as on domestic and wildlife fauna. Human exposure can arise from several sources, particularly from the leaching of babies' feeding bottles, from the plastic material used to line food and drink cans, and from dental fillings and sealants (8, 9). Traces of BPA have been found in fetal serum and full-term amniotic fluid, confirming its passage through the placenta (4). Moreover, people are mainly exposed to BPA through water consumption or by eating fish in which this estrogenic substance accumulates and increases in concentration (1).
Enzymes could degrade EDs leaving the ecosystem unchanged. Indeed, laccase and tyrosinase have been reported to degrade BPA in isothermal and nonisothermal reactors (10, 11). However, the endocrine mimetic activity of laccase-treated BPA, to validate BPA biodegradation, is unknown.
BPA significantly affects human and animal health. Neurochemical changes and abnormalities in sperm production, oocyte maturation, hormone release, immune system, cell proliferation, and sexual maturation have been reported (see1, 12–14). Of particular relevance, epidemiological studies have highlighted the correlation between the increase of BPA level in the environment and the incidence of cancer in humans (see15, 16). These data suggest that BPA can act as a mitogenic substance inducing cell proliferation and, consequently, the incidence and susceptibility to neoplastic transformations (17, 18).
Several mechanisms have been proposed to be implicated in BPA effects. Among others, estrogen receptors α and β (ERα and ERβ, respectively) have been reported as the foremost molecular mediators of the in vitro and in vivo effects exerted by BPA (see12, 14, 19–21). ERα and ERβ are ligand-activated transcription factors belonging from nuclear receptor super-family (22). Research has been mainly focused on the ability of BPA to modulate nuclear transcriptional activity of these receptor (see12, 19–21, 23, 24). However, ERα and ERβ are also localized at the plasma membrane where they initiate 17β-estradiol (E2)-induced rapid signals (22). Evidence accumulated suggests that ERα-dependent rapid membrane-starting actions are crucial for the E2-induced cancer promoting response in several target cells (22). Therefore, it is reasonable to suggest that BPA, similarly to E2, binds to ERα and produces changes in these rapid signals. Few data address the ability of BPA to mediate nongenomic estrogenic actions (see25–31) and, as far we know, no information focuses on the involvement of these pathways in the proliferative effect of BPA.
Aims of this work are to elucidate the ERα-dependent mechanism(s) underlying the proliferative effect of BPA and to evaluate if its estrogenic activity persists after laccase-catalyzed oxidation. Human cervix adenocarcinoma cells (HeLa cells) both devoid of ERs or transiently transfected with the ERα expression vector have been used as experimental model to discriminate between ERα-dependent and putative ERα-independent effects.
MATERIALS AND METHODS
E2, BPA, L-glutamine, gentamicin, trypsin, penicillin, DMEM (without phenol red), charcoal-stripped fetal calf serum, and GenElute plasmid maxiprep kit were purchased from Sigma-Aldrich (St. Louis, MO). Liquid scintillation cocktail was obtained from Perkin Elmer (Cambridge, UK). Recombinant human ERα was purchased from PanVera (Madison, WI). The ER inhibitor ICI 182,780 was obtained from Tocris (Ballwin, MO). The AKT inhibitor and the extracellular regulated kinase (ERK) inhibitor, PD98059, were obtained from Calbiochem (San Diego, CA). Lipofectamine reagent was obtained from GIBCO-BRL Life-technology (Galthersburg, MD). The XTT assay kit was purchased from Roche (Basel, Switzerland). The polyclonal anti-phospho-AKT antibody was obtained from New England Biolabs (Beverly, MA). The polyclonal anti-ERK, the monoclonal anti-phospho-ERK and the anti-AKT antibodies were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). The anti-β-tubulin antibody was obtained from MP Biomedicals (Solon, OH). The chemiluminescence reagent for Western blot CDP-Star and [6,7-3H(N)]estradiol (44.8 Ci/mmol; [3H]-E2) were obtained from NEN Life Science (Boston, MA). All other products were obtained from Sigma-Aldrich. Analytical or reagent grade products were used without further purification.
BPA Oxidation by Laccase and Characterization by Mass Spectrometry
Oxidized BPA was obtained from BPA aqueous solutions, after 120 min of treatment in a bioreactor, containing immobilized Trametes versicolor laccase (Sigma-Aldrich), operating under nonisothermal conditions as already described (10, 11). The molecular weight of oxidized BPA was determined by mass spectrometry in negative mode ([M-H]−) with a API 2,000-triple-quadrupole (Applied Biosystem, Foster City, CA).
HeLa cells were routinely grown in air containing 5% CO2, in modified, phenol red-free, DMEM medium containing 10% (v/v) charcoal-stripped fetal calf serum, L-glutamine (2 mM), gentamicin (100 μg/mL), and penicillin (100 U/mL). Cells were passaged every 2 days with 0.5 mL trypsin, and the media changed every 2 days.
Plasmids and Cell Transfection
The expression vectors for pCR3.1-β-galactosidase and human pSG5-hERα have been used (32). Furthermore, an empty vector, pCMV5 was used as control (32). Plasmids were purified for transfection using the GenEluted Plasmid Maxiprep Kit according to manufacturer's instructions (Sigma-Aldrich). A dose–response curve showed that the maximum effect was observed when 1 μg of plasmid was transfected together with 1 μg of pCR3.1-β-galactosidase to normalize for transfection efficiency (∼55–65%). HeLa cells were grown to ∼70% confluence, and then transfected with the expression vectors using lipofectamine reagent according to the manufacturer's instructions (GIBCO-BRL Life-technology). Six hours after transfection, the medium was changed and 24 h thereafter cells were stimulated.
Cell Proliferation Assay (XTT Assay)
HeLa cells were plated in 96-well culture plates at a density of 4,000 cells per well and stimulated with either vehicle (DMSO:PBS 1:1) or different concentrations of E2 (0.01–1,000 nM) or BPA (0.01–1,000 μM) or oxidized BPA (0.01–1,000 μM). In some experiments, 1 μM AKT inhibitor or 10 μM ERK inhibitor or 1 μM ER inhibitor ICI 182,780 was added before BPA treatment. After 24 h, cell growth was assessed by using the XTT reaction solution (sodium 3′-[1-(phenyl-aminocarbonyl)-3,4-tetrazolium]-bis(4-methoxy-6-nitro) benzene sulfonic acid hydrate and N-methyl dibenzopyrazine methyl sulfate; mixed in the 50:1 (v/v) ratio) according to the manufacturer's instructions (Roche). The optical density was determined at λ = 490 nm in the Multilabel Reader Victor3 (Perkin Elmer) after 4 h.
Electrophoresis and Immunoblotting
After stimulation, HeLa cells were lysed and solubilized in 0.125 M Tris HCl (pH 6.8), containing 10% SDS (w/v), 1 mM phenylmethylsulfonyl fluoride, and 5 μg/mL leupeptin, and boiled for 2 min. Twenty micrograms of solubilized proteins were resolved using 10% SDS-PAGE at 100 V for 1 h. The proteins were then electrophoretically transferred to nitrocellulose for 45 min at 150 V and 4 °C. The nitrocellulose was treated with 3% w/v bovine serum albumin in 138 mM NaCl, 26.8 mM KCl, 25 mM Tris HCl (pH 8.0), 0.05% Tween-20, 0.1% BSA, and then probed at 4 °C overnight with either one of anti-phospho-ERK or anti-phospho-AKT antibodies. The nitrocellulose was stripped by Restore Western Blot Stripping Buffer (Pierce Chemical Company, Rockford, IL) for 10 min at room temperature and then probed with either one of anti-ERK or anti-AKT antibodies. Anti-β-tubulin antibody (1 μg/mL) was used to normalize the sample loading. The antibody reaction was visualized with chemiluminescence reagent for Western blot. The densitometric analysis was performed by ImageJ software for Windows.
Ligand Binding Analysis
Values of the apparent molar fraction of E2-, BPA-, and oxidized BPA- bound ERα (Yapp) were determined by competitive radiometric binding assays using 2 nM [3H]-E2 as the tracer as previously reported (24, 33, 34). Incubation was done at 25 °C for 2 h in the binding buffer (Tris-HCl 0.04 M, EDTA 1.5 mM, DDT 1 mM, yeast extract 1% w/v, and glycerol 10% v/v, pH 7.4). The free and ERα-bound radioligand were separated by vacuum filtration through a 12-samples Millipore filter manifold (Bedford, MA), holding glass microfiber filters (Whatman Ltd, UK) (24). In some experiments, ERα-transfected HeLa cells were treated with 5 nM [3H]-E2 in the presence of the same concentration of competitors (E2, BPA, and oxidized BPA) used for the in vitro binding assay. At the end of incubation (37°C for 45 min), the culture medium was removed and the cells were washed with PBS and incubated with 400 μL of the lysis buffer (10 mM Tris-HCl pH 6.8, 2% SDS w/v, 10% glycerol v/v) for 30 min (35). The radioactivity present in whole cell or in vitro assays was counted with a 2100TR Tri-Carb liquid scintillation analyzer (Packard Instrument).
Values of the intrinsic molar fraction of E2-, BPA-, and oxidized BPA-bound ERα (Y) were obtained at pH 7.4 and 25 °C from Yapp values according to eq. 1 (36):
where [B] is the fixed [3H]-E2 concentration (=2 nM or 5 nM) and H is the equilibrium dissociation constant for [3H]-E2 binding to ERα (=0.2 ± 0.05 nM).
Values of the intrinsic equilibrium dissociation constant for E2 and BPA binding to ERα (Kd) were obtained at pH 7.4 and 25 °C according to eq. 2 (36):
where [L] is the free E2, BPA, or oxidized BPA concentration.
A statistical analysis was performed by using Student t-test with the GraphPad INSTAT3 software system for Windows. In all cases, probability (P) values below 0.05 were considered significant.
As expected (32), 17β-estradiol (E2) stimulation increases cell proliferation only in ERα-transfected HeLa cells (Figs. 1A and 1B). Correspondingly, 24 h of BPA stimulation does not promote the proliferation of ERα devoid HeLa cells at any of the tested concentrations (Fig. 1C); whereas 1 and 10 μM BPA increases the cell viability with respect to the vehicle-stimulated cells (Fig. 1D). At BPA concentration lower than 1 μM, no significant variation of cell proliferation has been observed. Of note, 100 and 1,000 μM BPA decrease the cell viability by 75% and 96%, respectively, in both ERα-devoid and ERα-containing HeLa cells (Figs. 1C and 1D).
On the contrary, the stimulation of either empty vector (Fig. 1E) and ERα-transfected HeLa cells (Fig. 1F) with oxidized BPA (0.01–1,000 μM) does not induce any significant increase in cell proliferation. Thus, BPA oxidation by laccase impairs the E2-mimetic properties of this compound on cancer cell growth. Notably, high levels of oxidized BPA (i.e., 1,000 μM) decreases (∼40%) the cell viability (Figs. 1E and 1F).
To confirm the ERα-dependence of BPA-induced cell proliferation, ERα-transfected HeLa cells have been stimulated with 0.01 μM E2 or 10 μM BPA in the absence or presence of 1 μM ER inhibitor ICI 182,780. Both E2 and BPA double the cell number; however, this effect is prevented by the antagonist ICI 182,780 (Fig. 2).
Mechanisms Involved in the BPA-Induced Cell Proliferation
At present it is well accepted that E2 exerts proliferative effects in several cell lines by rapid ERα-dependent ERK and AKT activation (22). Thus, the BPA ability to induce ERK and AKT phosphorylation in ERα and in empty vector-transfected HeLa cells has been evaluated. The Western blot analysis indicates that neither E2 or BPA or oxidized BPA increase the ERK phosphorylation in empty vector-transfected HeLa cells (Figs. 3A and 3A′) at any ligand concentrations (0.01–1,000 μM) and at any stimulation time (6–48 h) tested (data not shown). On the contrary, in ERα-transfected cells, E2 and BPA, but not oxidized BPA, trigger ERK activation 60 min after stimulation (Figs. 3B and 3B′). Notably, the addition of the ER inhibitor ICI 182,780 15 min before the treatment, completely prevents the E2- and BPA-induced ERK phosphorylation (Figs. 3B and 3B′). A similar behavior has been observed for the E2- and BPA-induced AKT phosphorylation. Indeed, neither E2 or BPA increases AKT phosphorylation when HeLa cells are transfected with the empty vector (Figs. 4A and 4A′). In the ERα-transfected HeLa cells, E2 and BPA induce AKT phosphorylation 60 min after stimulation (Figs. 4B and 4B′). The addition of ICI 182,780 completely impairs BPA-induced AKT phosphorylation. Lastly, oxidized BPA does not affect AKT phosphorylation neither in HeLa cells transfected with the empty vector nor in ERα-transfected HeLa cells (Fig. 4).
In order to determine whether ERK and AKT phosphorylation are involved in BPA-induced cell proliferation, the BPA- induced HeLa cell growth has been assessed in the presence of specific ERK or AKT inhibitors. Figure 5 shows that both ERK and AKT inhibitors impair proliferation induced by BPA in ERα-transfected HeLa cells.
The reported results indicate that laccase-catalyzed oxidation of BPA converts an ERα agonist into a much less reactive substance. To evaluate whether laccase treatment impairs the BPA ability to bind ERα, competitive binding assays have been performed. Data reported in Fig. 6 indicate that Kd values for E2 and BPA binding to ERα in vitro are 0.21 ± 0.05 nM and 1.2 ± 0.3 μM, respectively. These Kd values are in good agreement with those previously reported (33, 34). Moreover, Kd values obtained in vitro by using human recombinant ERα (Fig. 6, filled symbols) are confirmed in whole ERα-transfected HeLa cells (Fig. 6, open symbols). On the other hand, oxidized BPA does not bind to ERα neither in vitro nor in the whole cell over the ligand concentration range explored (i.e., from 10 nM to 30 μM) (Fig. 6). According to thermodynamic considerations (36), the Kd value for oxidized BPA binding to ERα is > 300 μM.
It is well established that BPA binds to ERα and ERβ inducing estrogen mimetic signals that modify E2-responsive gene expression (14, 20, 21, 24). Interestingly, the set of genes induced by BPA and E2 seems to be quite different. Most of them being unique for BPA (12, 19–21, 23, 24). During the past decade, there has been an enhanced understanding of the complexity of ERα and ERβ action. Indeed, the idea that ERα and ERβ act only via a genomic signaling mechanism must now be modified since both ERs are present at the plasma membrane to be part of the rapid phosphorylation signal transduction mechanism. Evidence accumulated indicates that ERα-dependent rapid signals account for E2-induced cancer promoting response in several target cells (22, 37), whereas ERβ is associated with a protective role in carcinogenesis (38, 39).
Intriguingly, BPA effects in inducing cell proliferation could be related to its ability to rapidly activate ERα-dependent “nongenomic” signals. Rapid modulation of ERK and AKT signaling in response to BPA has been observed in neurons, immune cells, and during brain development (14, 40–42). As far as we know, no information is available on the involvement of these BPA-induced rapid responses on cancer cell proliferation.
Our results confirm that BPA is a weak ERα agonist. In fact, the Kd value for BPA binding to ERα, either in vitro or overexpressed in cellular systems, is about 10,000-fold higher than that of E2. No BPA effect is observed in the presence of empty vector at any BPA concentration tested. The ability of the ER competitive inhibitor ICI 182,780 to prevent the BPA- and E2-induced cell proliferation further outlines that this effect of BPA is ERα-dependent.
BPA-induced cell proliferation appears very close to that of E2 although the required BPA concentration is about three orders of magnitude higher than that of E2 (i.e., 10 μM BPA versus 10 nM E2). This BPA concentration is compatible with the maximum estrogenic activity observed in MCF-7 cells (43, 44).
BPA at a concentration of at least 10-fold higher than that inducing cell proliferation exerts severe cytotoxic effects in a ER-independent manner. In fact, both empty vector- and ERα-transfected HeLa cells showed the same dose-dependent decrease in cell number. Similar results regarding both the cytotoxicity and the independency of ERα were previously reported in MCF-7 cells (11, 45). The mechanism by which BPA induces cytotoxicity on both HeLa and MCF-7 cells could be related to the production of reactive oxygen species and the consequent oxidative stress (46). Furthermore, several in vitro studies have shown that the treatment of Chinese hamster ovary cells with high BPA concentration induces chromosomal aberrations and is correlated with cell cytotoxicity. It is currently unclear whether the chromosomal alterations observed in these cells are a direct result of cytotoxicity or if cyotoxicity and aneuploidy are independent, but coincidentally occur at similar mM BPA concentration (16).
Present results correlate, for the first time, the BPA-induced proliferative response to the ERα-mediated nongenomic signals activation. In fact, upon BPA stimulation ERα mediates signal transduction pathways which culminates with ERK and AKT phosphorylation. The BPA-induced ERK and AKT phosphorylation are not present in empty vector-transfected cells and are completely prevented by the antiestrogen ICI 182,780. Moreover, similarly to E2, BPA-induced ERK and AKT activation is necessary for its proliferative effect. As a whole, this study indicates that BPA acts as an E2 agonist by the activation of rapid and nongenomic pathways that drive cells into proliferation.
Owing to the potential noxious effects of BPA on cell proliferation, special attention should be given to remediation of polluted waters. The bioremediation of BPA could became an accessible goal because of the fulfilment of a bioreactor carrying a membrane-immobilized laccase able to oxidize BPA (10). Our results indicate that laccase-catalyzed oxidation of BPA converts this ER agonist and cytotoxic compound into a much less reactive substance; indeed, enzymatically oxidized BPA loses any ERα-dependent activity. The stimulation of empty vector- and ERα-transfected HeLa cells with oxidized BPA does not induce cell proliferation under all the experimental conditions. The inability of oxidized BPA to induce the ERα-dependent cell proliferation is in good agreement with the inability of this compound to bind ERα either in vitro or in transfected HeLa cells. In line with this observation and according to the molecular weight obtained by mass spectrometry ([M-H]− = 227 ± 1), the most probable product obtained by laccase-catalyzed oxidation of BPA is 2,2-bis(4-phenylquinone)propane (Fig. 7). The oxidation of both BPA hydroxyl groups by laccase impairs ERα recognition. Indeed, according to X-ray studies (47–51), E2 recognition by ERα is achieved through hydrogen bonding between the hydroxyl group of the ligand (H-bond donor) and the receptor amino acidic residues (H-bond acceptors).
Intriguingly, the cytotoxic effect of BPA decreases after laccase-mediated oxidation. In fact, 100 μM BPA determines a decrease of 70% of the cell viability, while 1 mM oxidized BPA causes only 40% decrease of cell viability. The carbonylic groups present in oxidized BPA could render this compound less prone to the production of reactive oxygen species compared with BPA and, thus, less able to cause an oxidative stress into the cells.
In recent years, it has become evident that environmental EDs, even in minute amounts (part per trillion), could interfere with the synthesis, secretion, transport, metabolism, binding, action, or elimination of natural hormones, which, in turn, are responsible for homeostasis maintenance, reproduction, and developmental processes (52). Nowadays, endocrine disruption is one of the topics receiving much attention throughout all sectors of the society; thus, the need to identify the appropriate remediation actions is increasing. Our results indicate that bioreactors could be useful tools for the bioremediation of EDs in aqueous solutions.
Authors thank Dr. Stefania Notari (National Institute for Infectious Diseases I.R.C.C.S. “Lazzaro Spallanzani,” Roma, Italy) for performing mass spectroscopy analysis. Authors also thank Mr. Peter De Muro for helping in preparing this manuscript. Dr. Alessandro Bolli is supported by a grant from National Institute of Biostructures and Biosystems (INBB). This work was financially supported by MIUR (COFIN-PRIN 2006) to M.M.