Aflatoxin B1 – a potential endocrine disruptor – up-regulates CYP19A1 in JEG-3 cells

Toxicology Letters 202 (2011) 161–167

Contents lists available at ScienceDirect

Toxicology Letters journal homepage: www.elsevier.com/locate/toxlet

Research paper

Aflatoxin B1 – a potential endocrine disruptor – up-regulates CYP19A1 in JEG-3 cells Markus Storvik a,1 , Pasi Huuskonen a,1 , Taija Kyllönen a , Sarka Lehtonen b , Hani El-Nezami c , Seppo Auriola a , Markku Pasanen a,∗ a b c

School of Pharmacy, University of Eastern Finland, FIN-70211, Kuopio, Finland A.I. Virtanen Institute for Molecular Sciences, University of Eastern Finland, FIN-70211, Kuopio, Finland School of Biological Sciences, University of Hong Kong, Hong Kong, China

a r t i c l e

i n f o

Article history: Received 29 November 2010 Received in revised form 26 January 2011 Accepted 31 January 2011 Available online 4 February 2011 Keywords: Aflatoxin B1 Aflatoxicol Aromatase Endocrine disruptor JEG-3 Placenta

a b s t r a c t Previous studies have indicated that aromatase (CYP19A1) is involved in the metabolism of aflatoxin B1 (AFB1). We hypothesized that exposure to AFB1 contaminated food during pregnancy could disrupt the normal production of steroid hormones in placenta. We examined the capability of AFB1 exposure to disrupt CYP19A1 expression as a putative endocrine disrupter, and to investigate the metabolism of AFB1 by CYP19A1. JEG-3 cells, as model for placental cells, were exposed alone and in combination to AFB1 and estrogen receptor ligands for 24–96 h. AFB1 (0.3–1.0 ␮M) induced the expression of CYP19A1 by 163%–339% compared to control at the 96 h time point, although no induction was observed at 24 h. AFB1 concentrations higher than 1 ␮M were cytotoxic to JEG-3 cells, and the cytotoxicity was inhibited by the aromatase inhibitor, finrozole. AFB1 was metabolized to aflatoxicol (AFL) by JEG-3 cells and CYP19A1 recombinant protein. AFL formation was partially inhibited by addition of tamoxifen and finrozole to the JEG-3 cells. AFB1 had no effect on the expression of CYP1A2 and CYP3A4 in JEG-3 cells. These results reveal that AFB1 can affect the expression of aromatase enzyme, indicating that chronic exposure to AFB1 may cause endocrine disruption in the foetoplacental unit. © 2011 Elsevier Ireland Ltd. All rights reserved.

1. Introduction Placenta has metabolic and endocrine functions, and these functions can be affected by endogenous and xenobiotic compounds present in the maternal circulation. The expression of steroid metabolizing enzymes in placenta can be disrupted by a variety of xenobiotic pollutants, including cigarette smoke (Huuskonen et al., 2008) and maternal drug abuse (Paakki et al., 2000) and ambient air pollutants (Obolenskaya et al., 2010). Thus far, most studies examining the hormonal functions on the human foetoplacental unit have been carried out in countries where the environmental impact or nutritional hazardous effects are minimized. In countries with less well developed hygiene standards the chemical stress in foetoplacental unit may be high, which then may evoke alterations in metabolizing capacity. In Africa, South East Asia and South America exposure to crops contaminated by mould containing aflatoxin B1

∗ Corresponding author at: School of Pharmacy, Faculty of Health Sciences, University of Eastern Finland, P.O. Box 1627, FIN-70211 Kuopio, Finland. Tel.: +358 40 7199 346; fax: +358 17 162 424. E-mail address: markku.pasanen@uef.fi (M. Pasanen). 1 These two authors share the first authorship. 0378-4274/$ – see front matter © 2011 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.toxlet.2011.01.028

(AFB1) is a major public health problem (Abdulrazzaq et al., 2002, 2004; Polychronaki et al., 2006). AFB1 is well known as a mycotoxin and carcinogen, often present as a contaminant in grains and nuts (IARC, 2002). AFB1 requires metabolic activation in order to elicit its carcinogenic properties. The carcinogenicity of AFB1 is mainly attributable to the metabolite AFB-8,9-epoxide which can bind to DNA. The other known metabolites include carcinogenic metabolites aflatoxin M1, P1, Q1 and aflatoxicol (AFL) (Salhab and Edwards, 1977; Wong and Hsieh, 1976). AFB1 is metabolized mainly by cytochrome P450 (CYP) enzymes CYP1A2 and CYP3A isoforms into several metabolites (Eaton and Gallagher, 1994; Guengerich et al., 1998; IARC, 2002) in adult liver (Gallagher et al., 1994; IARC, 2002). It is also known that neither CYP1A2 nor CYP3A4-7 is active in placenta at term (Hakkola et al., 1996a,b; Myllynen et al., 2007). Additionally, high AFB1 concentrations in umbilical cord have been associated with low birth weight, kernicterus, and in some cases also with death of the foetus (Abdulrazzaq et al., 2002, 2004). Furthermore, in a recent perfusion study, aflatoxicol was the only metabolite detected in placental perfusion models, and it was demonstrated that AFB1 is transferred through the placenta (Partanen et al., 2010). The formation of AFL is reported to be mediated by a NADPHdependent reductase (Fig. 1.) (Salhab and Edwards, 1977). In the

162

M. Storvik et al. / Toxicology Letters 202 (2011) 161–167

Fig. 1. Aflatoxin B1 (AFB1) metabolism in human placenta to aflatoxicol (AFL) by an unidentified reductase enzyme.

placenta metabolism of AFB1 is suggested to be catalyzed by CYP1A and CYP19A1, but the actual metabolites were not identified (Sawada et al., 1993). The main CYP enzyme in placenta is CYP19A1, a reductase enzyme, which is relevant to estrogen production. Another potential CYP candidate for catalyzing the metabolism of AFB1 is CYP1A1, an enzyme which can be induced by maternal cigarette smoking in placenta (Huuskonen et al., 2008). The human placenta at term expresses only a few functional xenobiotic-metabolizing CYP enzymes, depending also on the stage of development (Pasanen, 1999). The placental CYP enzymes are responsible for the synthesis and metabolism of steroid hormones such as estrogen and progesterone (Myllynen et al., 2007). The placenta maintains balanced supply of steroid hormones that the development of foetus requires (Sanderson, 2009). Since AFB1 is metabolized to AFL, and transferred through placenta (Partanen et al., 2010), we hypothesise that the metabolism is performed by aromatase enzyme, and aim to study what might be the impact of AFB1 and/or its metabolites on the expression of CYP19A1. AFB1 and other contaminants may affect the physiological aromatase function and also estrogen synthesis in placenta (Ibrahim and Abul-Hajj, 1990; Monteiro et al., 2006; Numazawa et al., 2008; Saarinen et al., 2001). There are no studies aimed at clarifying the effects of AFB1 on placental steroid or xenobiotic metabolism (IARC, 2002; Partanen et al., 2010). Should the function of aromatase be altered in placentas exposed to AFB1, the altered estrogen production may affect foetal development. The steroid exposure during pregnancy is known to cause epigenetic alterations in foetus (Newbold et al., 2006; Prins, 2008), with consequences that may not be predictable at the moment. In addition, estrogenic exposure during pregnancy is known to increase the subsequent risk for breast cancer (Saarinen et al., 2001). This study aimed to evaluate whether AFB1 can alter the expression of steroid-metabolizing enzymes, CYP1A1 or CYP19A1, in JEG-3 cells. 2. Materials and methods 2.1. Chemicals Dulbecco’s modified Eagle’s medium (DMEM BE-12-604F with glutamine), foetal bovine serum (FBS), penicillin, streptomycin, MycoAlert Assay (Lonza, Switzerland), trypsin (Gibco/Invitrogen, CA, USA), phosphate buffered saline (PBS), ethylenediaminetetraacetic acid (EDTA) (Merck, Germany), aflatoxin B1 (AFB1), aflatoxin M1 (AFM1), aflatoxicol (AFL), ␣-naphthoflavone (ANF), ␤-naphthoflavone (BNF), TRI-Reagent, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-tetrazolium bromide (MTT), tamoxifen (T5648), sodium dodecyl sulphate (SDS), isopropanol, chloroform, methyl tert-buthyl ether (MTBE), dimethyl sulphoxide (DMSO), diethylpyrocarbonate (DECP) (Sigma–Aldrich, MO, USA), finrozole (Orion Pharma, Finland, donated by Professor Olavi Pelkonen, University of Oulu, Finland), dimethylformamide (DMF), formic acid (Riedel-de Haën, Germany), ribonuclease inhibitor, M-MuLV reverse transcriptase (Fermentas, MD, USA), Turbo DNA-Free Kit (Ambion/Applied Biosystems, TX, USA), CYP1A1, CYP1A2, CYP3A4, CYP19A1 and ␤actin Taqman Primer Probe Sets (Applied Biosystems, CA, USA), human CYP19A1 recombinant enzyme (Gentest/BD Biosciences, MA, USA), methanol (HPLC grade) (J.T. Baker, Holland), acetonitrile (HPLC grade) (Lab-Scan, Poland), nicotinamide adenine dinucleotide phosphate (NADPH) (Roche, Germany).

2.2. JEG-3 cells – human choriocarcinoma cell line JEG-3 (ATCC HTB-36) cells were maintained at 37 ◦ C as monolayers in DMEM BE-12-604F (with glutamine) supplemented with 10% inactivated FBS, 100 U/ml penicillin and 100 U/ml streptomycin in a humidified atmosphere of 5% CO2 and 95% air. Cells were subcultured by the treatment with 0.25% trypsin in PBS with EDTA, seeded and grown to the 70% confluence stage. We treated JEG-3 cells (2 million cells/dish in triplicate) with compounds as AFB1, finrozole, ANF, BNF, and tamoxifen (T5648) ≥99% pure. Cells were treated with these compounds for 0 h, 24 h and 72 h, or 96 h to examine long term effects. To ensure that contamination free cells were used, a mycoplasma test (MycoAlert Assay) was conducted. 2.3. MTT assay The cytotoxicity of AFB1 and finrozole treatment on the JEG-3 cells was analyzed using the MTT assay. The JEG-3 cells (2 × 105 cells/well) were planted onto the culture plate, incubated with AFB1 and finrozole for 72 h at 37 ◦ C, and then the unbound compounds were washed away. MTT reagent was added into each well, and after incubating for 4 h, and SDS–DMF was added and the plate was maintained overnight at 37 ◦ C. The optical density was measured using a BioTek ELx800 reader (BioTek, VT, USA) at a wavelength of 570 nm. 2.4. Aromatase activity measurement Aromatase CYP19A1 activities were measured with Wallac 1450 MicroBeta Trilux (Perkin Elmer, MA, USA) counter by a method of Pasanen (1985). 2.5. Extraction of total RNA Total RNA was extracted with TRI-Reagent followed with centrifugation in chloroform, isopropanol precipitation and 75% ethanol washing. The RNA pellet was diluted with DEPC–H2 O containing ribonuclease inhibitor. After extraction, total RNA was treated with DNase (Ambion Turbo DNA-Free Kit). The integrity of the isolated RNA was monitored by gel electrophoresis. The quantity of RNA was analyzed with NanoVue spectrophotometer (GE Healthcare, NJ, USA). The total RNA was stored at −80 ◦ C. 2.6. Quantitative real-time reverse transcription polymerase chain reaction (RT-QPCR) Complementary DNA (cDNA) was synthesized with the First-Strand cDNA synthesis technique (M-MuLV Reverse Transcriptase) where 2 ␮g of total RNA was used in each synthesis. The cDNA was stored at −80 ◦ C. RT-QPCR assay was performed using ABI prism 7500 instrument (Applied Biosystems, CA, USA). Detection was performed using CYP1A1 and CYP19A1 Taqman primer probe sets. Gene expression was normalized with reference gene ␤-actin and each sample was measured in triplicate. 2.7. Liquid chromatography–mass spectrometry detection of AFB1 and AFL Cell culture mediums were extracted with MTBE and evaporated to dryness with an evaporating centrifuge in room temperature. Evaporated residues were reconstructed with 300 ␮l of 20% methanol, filtered and transferred to HPLC-vials. AFB1, AFM1 and AFL were used as external standard chemicals. LC–MS system was equipped with an Agilent 1200 series HPLC system (Agilent Technologies, CA, USA) coupled to a Finnigan LTQ ion-trap mass spectrometry (Thermo Electron Corp., MA, USA) with an atmospheric pressure chemical ionization (APCI) source. The LC separation was carried out with a Waters XBridge C18 column (50 mm × 2.1 mm, 2.5 ␮m) (Waters Corp., MA, USA). Mobile phase consisted of (A) 0.1% formic acid and (B) methanol with 0.1% formic acid. Flow rate was 0.2 ml/min and column temperature was maintained at 30 ◦ C. The injection volume was 10 ␮l. In the gradient elution, the proportion of methanol was linearly increased from 10% to 90% in 14 min and then returned to 10%. The run time was 18 min. AFM1, AFB1 and AFL retention times were 8.5 min, 9.7 min and 10.8 min, respectively. In MS detection parent ions and

M. Storvik et al. / Toxicology Letters 202 (2011) 161–167

Fig. 2. (A) The effect of AFB1 treatment on CYP19A1 expression in JEG-3 cells. The CYP19A1 expression in JEG-3 cells treated by 0.1–1 ␮M AFB1, compared to controls, at 24 h, 72 h, and 96 h time points. Values shown are mean ± SD. (B) The activity of aromatase (CYP19A1) enzyme after AFB1 treatment in JEG-3 cells, measured as scintillation counts of metabolized 3 H-labelled androstenedione. Values shown are mean ± SD.

163

Fig. 3. The dose response curve of AFB1 and the effect of ER ligands measured by QPCR. (A) CYP19A1 mRNA expression at 96 h time point. (B) CYP1A1 expression at 96 h time point. Left: AFB1 or DMSO only; middle, AFB1 or DMSO with tamoxifen; right, AFB1 or DMSO with estriol. Values shown are mean ± SD. *Significant difference (p < 0.05, ANOVA) in expression values when compared to DMSO-only – control.

3.2. CYP19A1 and CYP1A1 induction by AFB1 in JEG-3 cells fragments were for AFM1 m/z 329 → 273, 301; for AFB1 m/z 313 → 285; and for AFL m/z 297 → 241, 269 (Supplementary Fig. 1). The parent ion of AFL was formed by a loss of the hydroxyl group. Data acquisition and quantification were conducted using XCalibur 2.0/LCquan 2.5 software (Thermo Electron Corp., MA, USA).

2.8. Recombinant aromatase incubation Since the only metabolite observed in placenta has been AFL (Partanen et al., 2010), and since the aromatase is the main CYP enzyme expressed in placenta, we tested whether human CYP19A1 recombinant enzyme possessed the ability to convert AFB1 to AFL. The incubation mixture contained 100 mM potassium phosphate buffer, pH 7.4, 2.5 mM MgCl2 , 2 ␮M AFB1, 10 pmol of recombinant aromatase, 0.5 mM NADPH and H2 O up to 500 ␮l. The incubation mixture was preheated at 37 ◦ C for 10 min, the reaction started by adding NADPH and the reaction was terminated by addition of an equal amount of acetonitrile. The incubation times were 0, 20, 60, and 180 min. After the termination of the reaction samples were centrifuged for 20 min at 2000 × g. A total of 500 ␮l supernatant was extracted with 1 ml of MTBE and evaporated at room temperature in an evaporating centrifuge. The residues were dissolved in 300 ␮l of 20% methanol, and measured with LC–MS, similar to the cell culture samples.

3. Results 3.1. AFB1 induces CYP19A1 expression in JEG-3 cells at 72–96 h time points Although the CYP19A1 expression remained at basal level at 24 h time point, the CYP19A1 messenger RNA (mRNA) increased at the 72 and 96 h time points in a dose-dependent manner (Fig. 2 A). The CYP19A1 mRNA expression has been reported to correlate directly with the aromatase activity (Wang and Leung, 2007), and this was confirmed by the present results. Enzyme activity assay with 3 H-labelled androstenedione, demonstrated that the aromatase activity had increased in AFB1 treated JEG-3 cells (Fig. 2B).

A dose–response curve for AFB1 induction of CYP19A1 was established. The 96 h time point was selected based on the delayed expression seen in the pilot experiment. The peak for the induction of CYP19A1 mRNA was observed when the cells were incubated with 0.3–1 ␮M AFB1 (Fig. 3A). Additional QPCR runs were performed in order to study the effect of AFB1 on CYP1A1 expression. There was a nonlinear CYP1A1 induction by AFB1 (Fig. 3B). This nonlinear induction was observed even when the experiment was repeated. There was a trend of 0.1 ␮M AFB1 to induce CYP1A1 but no induction was noted at 0.3 ␮M AFB1. Nonetheless, 1 ␮M AFB1 induced CYP1A1 in a statistically significant manner but the expressions of CYP1A2 and CYP3A4 were not affected by AFB1 (data not shown). Since the CYP1A1 expression had been induced by AFB1, and because CYP1A2 is one of the major enzymes involved in the metabolism of AFB1 in liver, we also evaluated the effect of compounds known to affect CYP1A1 expression on the AFB1 induction of CYPs. The CYP1A1 is strongly regulated by aryl hydrocarbon receptor (AHR) activity. The AHR mediates the effects of many endocrine disrupters, including cigarette smoke, and polycyclic aromatic hydrocarbons. Although AHR has not been reported to play a major effect on CYP19A1 mRNA expression, AHR and estrogen receptor (ER) nuclear signalling have cross-talk (Swedenborg and Pongratz, 2010) which we hypothesized could to affect the outcome of both CYP1A1 and CYP19A1 expression under AFB1 exposure. AHR agonist BNF, and AHR antagonist/partial agonist ANF were introduced in 5 ␮M co-treatment with AFB1. The AHR-antagonist ANF was observed to prevent the CYP19A1 induction by AFB1 slightly (p < 0.05). The AHR agonist BNF induced CYP1A1 mRNA dramatically, but there was no significant effect of BNF on AFB1-regulated induction of CYP19A1 (data not shown).

164

M. Storvik et al. / Toxicology Letters 202 (2011) 161–167

In the MTT assay, AFB1 evoked cytotoxicity in a dose-dependent manner (Fig. 5AB). The measured viability for 3 ␮M AFB1 was 70% of untreated control (Fig. 5A). 1 ␮M finrozole effectively prevented the increase of cytotoxicity seen with 0.1–3 ␮M AFB1 concentrations, i.e., viability was elevated to ∼90% of control in all measurements (Fig. 5AB). The measured cytotoxicity with AFB1 3 ␮M was confirmed to be prevented by 1 ␮M finrozole and difference reached significance (p < 0.001) in 2-way ANOVA with Bonferroni post-test (Fig. 5A). Dead cell plaques were observed at all AFB1 concentrations (0.1–3 ␮M) in a dose dependent manner (Fig. 5B).

Fig. 4. Effect of aromatase inhibitor finrozole on AFB1 induced CYP19A1 expression at 96 h time point in JEG-3 cells. AFB1 (0–3 ␮M) incubated without (left) or together with 0.1 ␮M (middle) or 1.0 ␮M (right) finrozole. Values shown are mean ± SD. *Significant difference (p < 0.05, ANOVA) in expression values when compared to DMSO-only – control.

3.3. Effects of estrogen receptor ligands on AFB1-induced CYP expression in JEG-3 cells Based on the literature, it was hypothesized that the CYP19A1 induction by AFB1 could be a result of ER activity which is known to drive CYP19A1 expression (Kumar et al., 2009). It is possible that either AFB1 can alter ERs, or AFB1 may be converted to a metabolite with this kind of activity. In order to study the effect of alterations in ER activity during AFB1 exposure, the JEG-3 cells were co-treated with AFB1 and estrogenic ligands. Tamoxifen is ER␣ and ER␤ receptor antagonist, while estriol is the major placental estrogen, acting as an ER␤ agonist. Both of the ligands were observed to be able to prevent both the basal, and the AFB1 induced CYP19A1 expression (Fig. 3A), probably through a ER mediated mechanism. The CYP19A1 expression was significantly decreased in 50 nM estriol (a concentration similar to the foetal side physiological concentration) treated JEG-3 cells as compared to control. The 50 nM estriol concentration caused even lower CYP19A1 expression than 500 nM estriol (concentration close to the maternal side physiological concentration) (Fig. 3A), although there was no major difference between the ability of 50 nM and 500 nM estriol to inhibit the AFB1 induced CYP19A1 expression. Although CYP1A1 is not considered to be regulated by estrogens, the expression of CYP1A1 was assayed from the same samples because nuclear receptor cross-talk may have lead to alterations. Estriol did not have any effect on CYP1A1 expression. However, the estrogen antagonist tamoxifen alone, but not with co-treatment with AFB1, did induce CYP1A1 to slight extent (Fig. 3B). 3.4. Effect of aromatase inhibitor on CYP19A1 expression Since the ER ligands were observed to modulate the effect of AFB1-induced and basal expression of CYP19A1, we studied if aromatase inhibitor finrozole could alter the CYP19A1 expression in AFB1 affected JEG-3 cells (Fig. 4). We observed that the aromatase inhibitor had no effect on basal expression of CYP19A1. However, finrozole effectively inhibited the induction caused by AFB1 (0.3 ␮M) indicating that AFB1 or its metabolite may have different potential ways to activate ERs. In addition, finrozole protected the JEG-3 cells from 3 ␮M AFB1 induced cytotoxicity (Fig. 5 A). When the cytotoxicity was prevented by finrozole, superinduction of CYP19A1 mRNA was observed with 3 ␮M AFB1 (Fig. 4). The CYP19A1 mRNA was +73% and +60% higher with 3 ␮M AFB1 when combined with 0.1 ␮M or 1 ␮M finrozole (respectively) as compared to otherwise highest consistent CYP19A1 induction observed at 0.3 ␮M AFB1. In addition, at 1 ␮M AFB1 concentration there was a trend towards a higher expression of CYP19A1 when co-treated with finrozole.

3.5. AFL is formed from AFB1 in JEG-3 cells, and in recombinant aromatase enzyme incubation As the only metabolite observed in placenta has been AFL (Partanen et al., 2010), and since the aromatase is the main CYP enzyme expressed in placenta, the ability of recombinant aromatase enzyme to convert AFB1 to AFL was tested. Increasing AFL formation was observed when recombinant aromatase enzyme was incubated with 2 ␮M AFB1 for 20–180 min (Fig. 6). Finrozole inhibited AFL formation 16%–28% at 90 min time point in recombinant aromatase enzyme incubation (data not shown). Finrozole reduced the conversion of AFB1 to AFL in JEG-3 cells (Fig. 7). In addition, the AFL formation was attenuated by a reduction of aromatase expression by treatment with tamoxifen (Fig. 8). No formation of the other AFB1 metabolite AFM1 was observed.

4. Discussion Although there is an extensive body of literature available on receptor-binding mediated endocrine disruption, a recent trend has been to expand the studies also to compounds which can alter the activities of enzymes in steroid production and metabolism. The present data suggests that AFB1 and/or its metabolites can have effects on enzymes which are important for placental hormonal production, i.e., addition of potential endocrine disruption into the repertoire of AFB1 in addition to its well known effect as a mutagen and carcinogen. We observed CYP19A1 to be strongly induced by AFB1 treatment. The CYP19A1 induction was delayed which points to the accumulation of a metabolite, new protein synthesis, or epigenetic changes such as histone modifications may be required before the induction can occur. The AFB1 had no effect on CYP19A1 expression at the 24 h time point. Since AFB1 exposure is often chronic, we continued to test the effect of AFB1 for up to 72 and 96 h. The CYP19A1 mRNA was induced at both time points. This may indicate that at least some of the effect is caused by the accumulation of metabolites. AFB1 evoked CYP19A1 induction in a time and dose-dependent manner. The highest consistent induction was observed at 0.3 ␮M. There was a slight variation in CYP19A1 induction between the experiments probably due to long cell culture experiments. However, the data suggests that at concentrations higher than 1 ␮M, AFB1 is cytotoxic to JEG-3 cells. This cytotoxicity could be augmented by addition of an aromatase inhibitor, finrozole, suggesting that the toxicity was increased either by the natural estrogen production by the cells or more likely by the accumulation of the AFB1 metabolite produced by CYP19A1 enzyme. The cytotoxicity of 1–3 ␮M AFB1 was confirmed in MTT assay and it was prevented by finrozole. This suggests that metabolites attributable to aromatase activity are essential for the cytotoxicity of AFB1. The data may mean that 0.1 ␮M finrozole was sufficient to inhibit cytotoxicity of 3 ␮M AFB1 without fully preventing aromatase activity, whereas 1 ␮M finrozole evoked more complete aromatase inhibition and less extensive super-induction of CYP19A1 mRNA.

M. Storvik et al. / Toxicology Letters 202 (2011) 161–167

165

Fig. 5. The cytotoxicity assessment of AFB1 and the protection by aromatase inhibitor finrozole in JEG-3 cells, using MTT assay. (A) Cell viability measured in MTT assay calculated as comparing to untreated DMSO control. ***Significance in 2-way ANOVA with Bonferroni post-test. (B) Light microscope pictures. Large dead cell plaques appear in a dose dependent manner. AFB1 concentrations (0–3 ␮M) are as indicated.

Fig. 6. The time–course metabolism of 2 ␮M AFB1 to AFL by recombinant aromatase enzyme. The values shown are mean ± SD.

The only metabolite observed in placenta has been AFL (Partanen et al., 2010) which we here report to be formed by the aromatase from AFB1. The major hepatic AFB1 metabolite, AFM1, was not observed in either cell culture mediums or with aromatase recombinant enzyme incubation. AFL accumulation was observed to be partially prevented by tamoxifen or finrozole in JEG-3 cells. The present data suggests that CYP19A1 induction by AFB1 or its metabolite AFL is probably dependent on ER activity. The CYP19A1 expression is driven by tissue-specific promoters located upstream from the tissue specific first exons (Bulun and Simpson, 1994; Kumar et al., 2009; Meinhardt and Mullis, 2002; Rawn and Cross, 2008). In placenta, transcription begins from exon I.1. In JEG-3 cells, CYP19A1 expression is regulated by the same promoter as in placental cells (Huang and Leung, 2009; Wang and Leung, 2007). Based on this similarity, JEG-3 cells can present a mechanistic model for placental cells. The hormonal regulators, transcription factors, and detailed mechanisms of CYP19A1 expression in placental tissue are still poorly understood. Kumar et al. (2009) inhibited CYP19A1 expression in COS-7 cells with ICI 182,780 which is an antagonist of classical ERs. ER␣ and ER␤ proteins have been detected in cultured syncytiotrophoblast cells

166

M. Storvik et al. / Toxicology Letters 202 (2011) 161–167

graphical areas. Furthermore, we need to evaluate the effects of chronic low-dose AFB1 exposure in placenta and in the developing foetus, and to determine whether some of the harmful effects of chronic AFB1 exposure may be mediated through hormonal disruption. However, in vivo research will be required to answer those questions. In addition, in general there is very little data on the effects of prenatal estrogenic alterations on later health of the newborns, caused by environmental factors. 5. Conclusions

Fig. 7. The AFL formation in AFB1 treated JEG-3 cells was partially inhibited by finrozole treatment at 72 h time point. The values shown are mean ± SD.

using immunohistochemistry (Schiessl et al., 2005). The regulation of CYP19A1 by the ERs and its relationship to epigenetic mechanisms has been further discussed by Kumar et al. (2009). However, aromatase has a complex regulation by the alternative promoters and interacting transcription factors (Bulun et al., 2003; Simpson, 2004). The complexity of the regulation suggests that one should not try to over-speculate on the estrogenic mechanisms on the observed induction of CYP19A1, or AFL accumulation role in it. From an evolutionary point of view, it seems that CYP19A1 has acquired a complex promoter structure and tissue dependent expression due its critical role in development, preventing viability of deletions or duplications. In addition, the ER function is still not completely understood, i.e., non-classical estrogen signalling and certain fast intracellular signalling-related actions of estrogens have been reported, suggesting that our understanding of the functions of estrogenic ligands may still change. In addition, according to the literature, even cyclic adenosine monophosphate (cAMP) accumulation can induce CYP19A1 mRNA highly as confirmed here by forskolin treatment in JEG-3 cells (data not shown). This would provide possibilities for both membrane bound receptors and the nuclear ERs to regulate aromatase expression in the placenta. One limitation which should be considered is that the AFB1 concentrations used in the experiments are higher than measured placental concentrations in populations in badly contaminated geo-

In conclusion, it was noted that AFB1 could induce CYP19A1 expression with a delay. This indicates that the metabolites of AFB1 produced by aromatase may be responsible for the observed CYP19A1 induction, as the induction was inhibited by the aromatase inhibitor, finrozole. In addition, finrozole prevented cytotoxicity of AFB1. The accumulation of AFL was dependent on the aromatase and attenuated partially by finrozole by inhibiting aromatase reaction and by tamoxifen reducing the estrogenrelated aromatase expression. The data suggest that both ER␣ and ER␤ may participate in the induction of CYP19A1 caused by AFB1 or its metabolite AFL, but due to the complexity of estrogen signalling pathway, further studies will be needed to solve the mechanisms of action. The present data suggest for the first time that the AFB1 had effects on genes important in endocrine regulation in placental cells. The putative effects of AFB1 as hormonal disruptor at the population level in countries with chronic AFB1 contamination should be studied further. Conflict of interest The authors declare that there are no conflicts of interest. Acknowledgements We would like to thank Mrs. Pirjo Hänninen for her skilful laboratory assistance. This work was supported by The Academy of Finland [122859/2007] and by The Finnish Funding Agency for Technology and Innovation [40225/2008]. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.toxlet.2011.01.028. References

Fig. 8. The AFL formation was attenuated by tamoxifen JEG-3 cells. The values shown are single measurements from pooled samples.

Abdulrazzaq, Y.M., Osman, N., Ibrahim, A., 2002. Fetal exposure to aflatoxins in the United Arab Emirates. Ann. Trop. Paediatr. 22, 3–9. Abdulrazzaq, Y.M., Osman, N., Yousif, Z.M., Trad, O., 2004. Morbidity in neonates of mothers who have ingested aflatoxins. Ann. Trop. Paediatr. 24, 145–151. Bulun, S.E., Sebastian, S., Takayama, K., Suzuki, T., Sasano, H., Shozu, M., 2003. The human CYP19 (aromatase P450) gene: update on physiologic roles and genomic organization of promoters. J. Steroid Biochem. Mol. Biol. 86, 219–224. Bulun, S.E., Simpson, E.R., 1994. Regulation of aromatase expression in human tissues. Breast Cancer Res. Treat. 30, 19–29. Eaton, D.L., Gallagher, E.P., 1994. Mechanisms of aflatoxin carcinogenesis. Annu. Rev. Pharmacol. Toxicol. 34, 135–172. Gallagher, E.P., Wienkers, L.C., Stapleton, P.L., Kunze, K.L., Eaton, D.L., 1994. Role of human microsomal and human complementary DNA-expressed cytochromes P4501A2 and P4503A4 in the bioactivation of aflatoxin B1. Cancer Res. 54, 101–108. Guengerich, F.P., Johnson, W.W., Shimada, T., Ueng, Y.F., Yamazaki, H., Langouet, S., 1998. Activation and detoxication of aflatoxin B1. Mutat. Res. 402, 121–128. Hakkola, J., Pasanen, M., Hukkanen, J., Pelkonen, O., Maenpaa, J., Edwards, R.J., Boobis, A.R., Raunio, H., 1996a. Expression of xenobiotic-metabolizing cytochrome P450 forms in human full-term placenta. Biochem. Pharmacol. 51, 403–411. Hakkola, J., Raunio, H., Purkunen, R., Pelkonen, O., Saarikoski, S., Cresteil, T., Pasanen, M., 1996b. Detection of cytochrome P450 gene expression in human placenta in first trimester of pregnancy. Biochem. Pharmacol. 52, 379–383.

M. Storvik et al. / Toxicology Letters 202 (2011) 161–167 Huang, H., Leung, L.K., 2009. Bisphenol A downregulates CYP19 transcription in JEG-3 cells. Toxicol. Lett. 189, 248–252. Huuskonen, P., Storvik, M., Reinisalo, M., Honkakoski, P., Rysa, J., Hakkola, J., Pasanen, M., 2008. Microarray analysis of the global alterations in the gene expression in the placentas from cigarette-smoking mothers. Clin. Pharmacol. Ther. 83, 542–550. Ibrahim, A.R., Abul-Hajj, Y.J., 1990. Aromatase inhibition by flavonoids. J. Steroid Biochem. Mol. Biol. 37, 257–260. IARC Working Group on the Evaluation of Carcinogenic Risks to Humans, 2002. Some traditional herbal medicines, some mycotoxins, naphthalene and styrene. IARC Monogr. Eval. Carcinog. Risks Hum. 82, 1–556. Kumar, P., Kamat, A., Mendelson, C.R., 2009. Estrogen receptor alpha (ERalpha) mediates stimulatory effects of estrogen on aromatase (CYP19) gene expression in human placenta. Mol. Endocrinol. 23, 784–793. Meinhardt, U., Mullis, P.E., 2002. The aromatase cytochrome P-450 and its clinical impact. Horm. Res. 57, 145–152. Monteiro, R., Azevedo, I., Calhau, C., 2006. Modulation of aromatase activity by diet polyphenolic compounds. J. Agric. Food Chem. 54, 3535–3540. Myllynen, P., Pasanen, M., Vahakangas, K., 2007. The fate and effects of xenobiotics in human placenta. Expert Opin. Drug Metab. Toxicol. 3, 331–346. Newbold, R.R., Padilla-Banks, E., Jefferson, W.N., 2006. Adverse effects of the model environmental estrogen diethylstilbestrol are transmitted to subsequent generations. Endocrinology 147, S11–S17. Numazawa, M., Komatsu, S., Tominaga, T., Yamashita, K., 2008. Structure–activity relationships of estrogen derivatives as aromatase inhibitors. Effects of heterocyclic substituents. Chem. Pharm. Bull. (Tokyo) 56, 1304–1309. Obolenskaya, M.Y., Teplyuk, N.M., Divi, R.L., Poirier, M.C., Filimonova, N.B., Zadrozna, M., Pasanen, M.J., 2010. Human placental glutathione S-transferase activity and polycyclic aromatic hydrocarbon DNA adducts as biomarkers for environmental oxidative stress in placentas from pregnant women living in radioactivity- and chemically-polluted regions. Toxicol. Lett. 196, 80–86. Paakki, P., Stockmann, H., Kantola, M., Wagner, P., Lauper, U., Huch, R., Elovaara, E., Kirkinen, P., Pasanen, M., 2000. Maternal drug abuse and human term placental xenobiotic and steroid metabolizing enzymes in vitro. Environ. Health Perspect. 108, 141–145. Partanen, H.A., El-Nezami, H.S., Leppanen, J.M., Myllynen, P.K., Woodhouse, H.J., Vahakangas, K.H., 2010. Aflatoxin B1 transfer and metabolism in human placenta. Toxicol. Sci. 113, 216–225.

View publication stats

167

Pasanen, M., 1999. The expression and regulation of drug metabolism in human placenta. Adv. Drug Deliv. Rev. 38, 81–97. Pasanen, M., 1985. Human placental aromatase activity: use of a C18 reversed-phase cartridge for separation of tritiated water or steroid metabolites in placentas from both smoking and non-smoking mothers in vitro. Biol. Res. Pregnancy Perinatol. 6, 94–99. Polychronaki, N., Turner, C.P., Mykkanen, H., Gong, Y., Amra, H., Abdel-Wahhab, M., El-Nezami, H., 2006. Determinants of aflatoxin M1 in breast milk in a selected group of Egyptian mothers. Food Addit. Contam. 23, 700–708. Prins, G.S., 2008. Estrogen imprinting: when your epigenetic memories come back to haunt you. Endocrinology 149, 5919–5921. Rawn, S.M., Cross, J.C., 2008. The evolution, regulation, and function of placentaspecific genes. Annu. Rev. Cell Dev. Biol. 24, 159–181. Saarinen, N., Joshi, S.C., Ahotupa, M., Li, X., Ammala, J., Makela, S., Santti, R., 2001. No evidence for the in vivo activity of aromatase-inhibiting flavonoids. J. Steroid Biochem. Mol. Biol. 78, 231–239. Salhab, A.S., Edwards, G.S., 1977. Comparative in vitro metabolism of aflatoxicol by liver preparations from animals and humans. Cancer Res. 37, 1016–1021. Sanderson, J.T., 2009. Placental and fetal steroidogenesis. Methods Mol. Biol. 550, 127–136. Sawada, M., Kitamura, R., Norose, T., Kitada, M., Itahashi, K., Kamataki, T., 1993. Metabolic activation of aflatoxin B1 by human placental microsomes. J. Toxicol. Sci. 18, 129–132. Schiessl, B., Mylonas, I., Hantschmann, P., Kuhn, C., Schulze, S., Kunze, S., Friese, K., Jeschke, U., 2005. Expression of endothelial NO synthase, inducible NO synthase, and estrogen receptors alpha and beta in placental tissue of normal, preeclamptic, and intrauterine growth-restricted pregnancies. J. Histochem. Cytochem. 53, 1441–1449. Simpson, E.R., 2004. Aromatase: biologic relevance of tissue-specific expression. Semin. Reprod. Med. 22, 11–23. Swedenborg, E., Pongratz, I., 2010. AhR and ARNT modulate ER signaling. Toxicology 268, 132–138. Wang, Y., Leung, L.K., 2007. Pharmacological concentration of resveratrol suppresses aromatase in JEG-3 cells. Toxicol. Lett. 173, 175–180. Wong, J.J., Hsieh, D.P., 1976. Mutagenicity of aflatoxins related to their metabolism and carcinogenic potential. Proc. Natl. Acad. Sci. U.S.A. 73, 2241–2244.