An In Vivo Multiwell-Based Fluorescent Screen for Monitoring

Jul 20, 2007 - L. Préau , K. Le Blay , E. Saint Paul , G. Morvan-Dubois , B.A. Demeneix ... potential risks of emerging contaminants to human and env...
0 downloads 0 Views 399KB Size
Download PDF
Environ. Sci. Technol. 2007, 41, 5908-5914

An In Vivo Multiwell-Based Fluorescent Screen for Monitoring Vertebrate Thyroid Hormone Disruption JEAN-BAPTISTE FINI,† SE ´ BASTIEN LE ME ´ VEL,† NATHALIE TURQUE,† KARIMA PALMIER,† DANIEL ZALKO,‡ JEAN-PIERRE CRAVEDI,‡ AND B A R B A R A A . D E M E N E I X * ,† UMR CNRS 5166, Evolution des Re´gulations Endocriniennes, Department of Regulations, Development and Molecular Diversity, Muse´um National d’Histoire Naturelle, 7 Rue Cuvier, 75231 Paris Cedex 05, France, INRA, UMR 1089 Xe´nobiotiques, 180 chemin de Tournefeuille, BP 3, 31931 Toulouse Cedex 9, France

There is a pressing need for high throughput methods to assess potential effects of endocrine disrupting chemicals (EDCs) released into the environment. Currently our ability to identify effects in vitro exceeds that for in vivo monitoring. However, only in vivo analysis provides the full spectrum of physiological impacts exerted by a given chemical. With the aim of finding a physiological system compatible with automatic plate reading we tested the capacity of early embryonic stage Xenopus laevis tadpoles to monitor thyroid hormone (TH) disruption. Fluorescent transgenic X. laevis embryos bearing a TH/bZIP-eGFP construct, placed in 96 well plates, were used for a physiological-based screen for potential TH signaling disruptors. Using stage NF-45 embryos (time of thyroid gland formation) allowed rapid detection of chemical interference with both peripheral TR signaling and production of endogenous TH. Nanomolar concentrations of TH receptor agonists could be detected within 72 h. Moreover, when testing against a 5nM T3 challenge, the effects of inhibitors of TH production were revealed, including inhibitors of TH synthesis, (methimazole: 1 mM or sodium perchlorate: 3.56 µM), as well as antagonists acting at the receptor level (NH3: 2 µM) and a deiodinase inhibitor (iopanoic acid: 10 µM). Finally, we show that the thyroid disrupting activities of BPA (10 µM) and TBBPA (1 µM) can also be detected in this rapid screening protocol. Finally, this noninvasive technology using an automatic reading system shows low variability (around 5%) and permits detection of subtle changes in signaling by EDCs that either inhibit or activate TH signaling in vivo.

Introduction The number of chemicals currently released into the environment has been estimated at over 30 000 (1). Few of * Corresponding author phone: +33140793607; fax: +33140793618; e-mail: [email protected]. † MNHN/CNRS Paris. ‡ INRA Toulouse. 5908

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 16, 2007

these chemicals have been tested for their toxic or physiologic effects. Certain, such as plastifiers and flame retardants, are already known to have endocrine disrupting effects, particularly on signaling via nuclear receptors, including the estrogen (ER) and TH receptors (TRs) (2). The testing of the many others requires reliable, reproducible, and preferably rapid screening methods. We tested the possibility of using very early stage transgenic X. laevis embryos that can be placed in microwell plates for screening of TH inhibitors and agonists. The X. laevis embryo is particularly suited for chemical testing in general. Indeed, amphibians are the only tetrapod vertebrates to have free-living embryos, i.e., in which embryonic development is neither in utero nor in ovo, thus enabling a pertinent physiological approach on an intact vertebrate organism in a multiwell assay. As most EDCs modulate nuclear receptor signaling (3) it is advantageous to use an in vivo transcriptional based assay to detect potential EDCs. To date, one of the most used in vivo systems for studying TH actions and, more recently, for detecting the effects of TH disruption is the metamorphic process in anuran amphibians. This choice is determined by the fact that metamorphosis is totally dependent on and entirely orchestrated by TH (4). Moreover, the previous focus on the premetamorphic (stage NF51-NF55 (5)) and metamorphic (stage NF58-NF60) periods for testing TH disruption is logical, not only because of the robust responses to TH during this developmental phase, but because the tadpole thyroid gland appears from stage NF 43 onward (6, 7). However, recent work has shown that all the main components of thyroid signaling, ligand, receptors, deiodinase, and peroxidase enzymes are present well before the formation of the thyroid gland (7-9). These results raise the possibility that tissue specific TH responses followed by reporter gene assays should be detectable in tadpoles well before the premetamorphic and metamorphic periods. Precocious responses are advantageous in that the size of the tadpoles is amenable to more intensive screening protocols in multiwell plates. We tested this hypothesis using early stage transgenic X. laevis tadpoles in an in vivo transcriptionalbased assay to detect potential EDCs. The transgenic X. laevis line used carries a 376 bp portion of the X. laevis TH/bZIP promoter cloned upstream of the eGFP coding sequence and a portion of the γ-Crystallin promoter coupled to a RedFP plasmid (10). The TH/bZIP sequence contains a consensus thyroid hormone element (TRE). TH/bZIP is an optimal TH response gene (11), being an even more sensitive biomarker than TRβ for TH regulation (12). The fact that these transgenic X. laevis tadpoles can be used at later, premetamorphic stages to follow the effects of TH agonists and some types of EDCs has already been reported (10). However, these initial methodologies have two main limitations. First, the size of the pre-metamorphic tadpoles used precludes any multiple well-based assays (see Supporting Information Figure 1) and second, only chemicals acting on peripheral TH signaling can be detected in the short time frame (98% Sigma), thyroxine (T4, 98% Sigma), 3,5,3′-triiodothyroacetic acid (TRIAC, 98%, Sigma), methimazole (>98%, Sigma), BPA (99%, Aldrich), TBBPA (97%, Aldrich), iopanoic acid (98%, Aldrich), and sodium perchlorate (Fluka) were from Sigma-Aldrich (St. Quentin Fallavier, France). GC-1 and NH-3 were provided by T. Scanlan. T3, T4, and iopanoic acid were dissolved in 30% NaOH 1M, 70% sterilized water. Methimazole and sodium perchlorate were dissolved in MilliQ water. BPA, TBBPA, NH-3, GC-1, and TRIAC were dissolved in EtOH. Optimization of Stages to be Used for MFA. To optimize stages for MFA we quantified transcriptional activity of F1 transgenic tadpoles, from embryonic stage NF40 up to stage larval NF45 (see “Imaging” section below). Induction by exogenous TH at different developmental points between stage 42 and stage 52 was then assessed. To this end, tadpoles were sorted according to stage and fluorescence level and then placed into six well plates from TPP (Switzerland) or 5 liter aquaria for stage NF52 tadpoles. Images of tadpoles with or without T3 at 5.10-9M or 10-8M were recorded every 24 h over 4 days. Treatment Of Animals. F0 or F1 transgenic males were bred with wild type females. Fertilized eggs were transferred to aquaria for 4 days (light/dark: 12/12; T°:21 °C). Transgenic animals were sorted according to stage (5) and fluorescence, 24 h before treatment. Transgenic animals were placed at 10 or 15 tadpoles per well in transparent flat six well plates from TPP (Switzerland) at 24° ( 0.5 °C. Aliquots of stock solutions (10-2M) for T3, T4, TRIAC, NH3, and GC-1 were stored at -20 °C. All other solutions were made extemporaneously: BPA at 10-1M, TBBPA at 10-1M, IOP at 10-1M, methimazole at 10-1M, and sodium perchlorate at 1 g/L. Chemicals were diluted in 1.5 mL silanized tubes (Eppendorf). Dilutions were made by adding 5 µL of the dilution into 50 mL Greiner capped tubes filled with water. Solvents were used as controls at the highest dose used (1/10000 for EtOH). Dilutions were added directly to the well of six well plates used in treatment. Medium was renewed every 24 h. For fluorescent readings, by either imaging or plate readings, tadpoles were anesthetized in Ethyl 3-aminobenzoate methanesulfonate salt (MS-222) 0.01% (Sigma). Tadpoles cannot feed until stage NF46 (6 days pf) so no food was needed during the experiments. Additional experiments were carried out using solutions spiked with radio-labeled molecules, to ensure that the bisphenols tested were soluble under our experimental conditions. [3H]-BPA, with a specific activity of 185 GBq.mmol-1 and purity greater than 98% was purchased from Moravek Biochemicals (CA). [14C]-TBBPA, with a specific activity of 3.6 GBq‚mmol-1 and purity greater than 99% was synthesized using methods detailed elsewhere (15). The radioactivity in the solution was determined by counting of aliquots on a Packard scintillation analyzer (model Tricarb 2200CA; Packard Instruments, Meriden, CT) using Packard Ultima Gold as the scintillation cocktail (Packard Instruments, Downer Grove, IL). Imaging. Photographs were taken using an Olympus fluorescent dissecting microscope, equipped with an Olympus video camera DP50 (Olympus, Rungis, France) with GFP filter unit (Olympus SVZ-MGFP). All pictures were taken with the same parameters (32× objective and 4 s exposure time, intensity 100). Acquisitions were made with Studiolite software. Quantifications were done using ImageJ software (Rasband 1997) with the region of interest (ROI) delimited to the head, excluding intestine (to avoid any autofluorescence due to the yolk still present at this stage in this tissue). Data are expressed in relative units of fluorescence (RFU). VOL. 41, NO. 16, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

5909

TABLE 1. Comparison of Experimental Advantages and Limitations for Experimentation for Stages 42, 45, and 52 X. laevis Tadpolesa stage 42 time to reach this stage (24 °C) size possible automatic fluorescence reading thyroid gland activity interference with feeding T3 response within 3 days intra assay variability in GFP signal

4 days

stage 45 5 days

stage 52 21 days

small (mm) small (mm) big (cm) yes yes not possible beginning low activity moderate no beginning yes variable robust variable yes

no

no

a Stages 42, 45, and 52 were analyzed and compared for different endpoints. Note that stage 45 offers the most advantages for rapidly studying TH responses.

Automatic 96 Well Plate Readings (MFA). Tadpoles were placed one per well of a black, conic based-96 well plate (Greiner Bio-one, France), with the head in the center of the well and the dorsal surface in contact with the plate so as to limit interference with the GFP signal by melanocytes present on the dorsal surface. Fluorescence was measured with the Ultra Evolution high-end multifunctional microplate reader (TECAN France SA, Lyon), equipped with excitation (475495 nm) and emission filters (495-545 nm). Readings were done on nine points per well, spatially distinct (each repeated 10 times) with z-optimal on each plate. Gain was optimized in each experiment on T3 positive treated tadpoles. X Fluor macro (under Microsoft Excel tabler) was used for integration of results. Statistical Analysis of Results. In vivo results are expressed as mean (SE per group. Differences between means were analyzed when one concentration per product was tested by nonparametric Mann-Whitney t test. When multiple concentrations were analyzed simultaneously, significance was assessed according to experimental plan with either twoway ANOVA or one-way ANOVA, (two tailed with R ) 5%, followed with Dunnet’s test). Differences were considered significant at p < 0.05. In all cases, typical experiments are shown, each experiment having been repeated at least three times (with n g 15 tadpoles/group) and providing the same results. Statistics were done with Instat 2.0.

Results and Discussion We describe a screening method that combines transgenic and imaging technologies with our most recent knowledge of thyroid signaling and physiology in X. laevis. At the outset we hypothesised that young tadpoles with very low levels of circulating TH are competent to respond to a TH stimulus, as they express both TRs ands deiodinases (8, 9). We tested this hypothesis by following the transcriptional activity of a TH response gene in tadpoles from an established line of transgenic TH/bZip-eGFP X. laevis. We first tested which stages gave optimal fluorescent TH response (Table 1). The earliest stage to show strong, robust T3 induction was larval stage NF45 (approximately 5 days post-hatching), so all subsequent experiments used this stage. Initially, fluorescence was quantified by individual quantification of computer images or automatic plate reading on 96 well plates (microplate fluorescence assay, MFA). Both methods gave exactly the same result (Figure 1) when used for testing with 2.10-8M T3 or 2.10-8M TRIAC (triiodothyroacetic acid), a naturally occurring T3 analogue. As the 96 well plate MFA allows for high throughput screening we used MFA in all following experiments. 5910

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 16, 2007

FIGURE 2. Dose dependent responses to TH, T3 are seen at 48, 72, and 96 h. Germinally transgenic TH/bZIP-eGFP tadpoles (F1 generation) were treated with or without T3 at different concentrations for 4 days. Each day, the medium was renewed and fluorescence of a batch of 15 tadpoles quantified by automatic well plate reading Ultra Evolution TECAN. In each case, the experiment was repeated three times, providing similar results. Two-way ANOVA, two tailed, r) 5% followed by Dunnet’s test ***p < 0.001. Dose-dependent responses to T3 were analyzed at different time points. T3 concentrations as low as 5.10-10M gave significant inductions and dose-dependent responses were recorded at all time points examined, i.e., at 48, 72, and 96 h (Figure 2). At 72 h after a 5 nM T3 stimulus, a significant, robust response was clearly visible, so this time point was used from then on. We next assessed responses to various TH agonists including T4, as compared to a standardized 5.10-9M T3 response. T4 is the precursor of T3, the biologically active form of TH, which is produced by deiodination in target tissues. Figure 3 A and B shows the induction in fluorescence seen with 5.10-9M T3 for 72 h. TH/bZIP-eGFP expression is found in the head region, mainly in the central nervous system and gills, known sites of TH action during metamorphosis. Figure 3C shows the tadpoles positioned in the 96 well plates for monitoring. Dose-dependent responses to physiological levels of T4 were recorded, with 2-fold inductions seen with 10-8M T4 within 72 h (Figure 3D). Notably, a nearly 3-fold induction of signal was seen with 5.10-9M T3 (Figure 3 A, B, and D). Two well-known TH agonists, GC-1, a TRβ isoform specific agonist (16) and TRIAC were tested and each produced strong, significant inductions of the GFP signal. TRIAC (10-8M) gave a 3-fold induction (Figure 3D) being equipotent with the same concentration of T3 (data not shown). GC-1, induced a 3-fold induction at 10-7M, reflecting the lower levels of TRβ present in these early stage embryos (17). Responses to agonists were robust and reproducible with only small intraand inter-assay variations (e5% in each case) when using the same generation of tadpoles. We also assessed whether thyroid inhibitors could be detected in this transcriptional assay. Two categories of antagonists were tested, either alone or against a T3 (5.10-9M) or a T4 (10-8M) pulse. First, two substances acting in the periphery, at the level of the receptor, were tested: NH-3, a synthetic TH antagonist and iopanoic acid (IOP), a deiodinase inhibitor. NH-3, at 2.10-6 M the optimal antagonist concentration for inhibiting T3 induced metamorphosis (18), totally abrogated both the T3 and the T4 dependent inductions, but had no effect alone (Figure 4A). Next, as expected, IOP produced differential effects according to whether it was tested against a T3 or a T4 stimulus (Figure 4B). These differential effects reflect the fact that IOP inhibits all deiodinases, affecting both activation and inactivation of TH. When tested against T3, IOP increased the signal, reflecting the fact that degradation of T3 through deiodinase 3 was inhibited, thus increasing local concentrations of T3. In contrast, when tested against T4, IOP reduced the fluorescent signal, as in this case the activating step of conversion of T4 to T3 was inhibited.

FIGURE 3. Stage 45 TH reporter tadpoles show dose-dependent responses to thyroid agonists. (A) Stage 45, control, TH/bZIP-GFP/γ-crystalline-RedFP plasmid transgenic tadpole viewed under blue light showing fluorescence in the eyes and background fluorescence in the intestine. (B) Stage 45, TH/bZIP-GFP/γ-crystalline-RedFP plasmid transgenic tadpole exposed to 5.10-9 M T3 for 72 h, showing markedly induced fluorescence in the central nervous system and gills. (C) Stage 45 tadpoles placed in 96 wells viewed under natural light. (D) Action of TH agonists assessed using F2 generation of germinally transgenic tadpoles. Tadpoles were treated for 72 h (with daily renewal) with T3, T4 (10-10 M, 10-8 M, 10-6 M) or thyroid hormone agonists GC-1 (10-7 M) and TRIAC (10-8 M). TH/bZIP-GFP transcription was measured using an Ultra Evolution multiple well plate reader at gain 56, z: 6100 µm. Values shown are mean ( SE (n ) 20/group). The experiment was repeated three times, providing similar results. One way ANOVA was used for analysing statistical differences. *p < 0.05, **p < 0.01, ***p < 0.001. Next, two chemicals inducing hypothyroidism by acting at the level of the thyroid gland itself were tested: methimazole and sodium perchlorate (Figure 4C and D). Methimazole, a thioamide, inhibits thyroid peroxidase (TPO) and thus formation of T4 and T3 (19); whereas perchlorate ions competitively block the sodium-iodine symporter (NIS) and iodine uptake by thyroid gland follicles (20). These substances were tested at concentrations known to slow X. laevis tadpole development (21). Methimazole (1 mM) exerted a significant inhibitory effect against the T3 stimulus, but had no effect either alone or combined with T4 (Figure 4C). Similarly, sodium perchlorate (3.56 µM or 500 µg/L) produced a slight (10%), but significant (p < 0.05) reduction in the signal produced by a T3 pulse. The concentration of perchlorate used was the lowest effective dose known to significantly delay development in the OECD X. laevis metamorphosis assay XEMA (14). This assay requires three weeks exposure to perchlorate, starting at late, premetamorphic stages (NF 51), to see the inhibitory effects of perchlorate at this concentration. However, lower concentrations (65-125 µg/ L) can be detected over three weeks using finer morphological parameters (Moser, personal communication). We saw effects on transgene expression levels at these doses within 72 h when using computer imaging of individual tadpoles (data not shown). When measuring signals by MFA we could only detect antagonists acting at the level of the thyroid gland when using the T3 challenge. We had hypothesised that using embryos at this early stage, when the thyroid gland was just forming, would make the thyroid follicles particularly sensitive to anti-thyroidal agents. This hypothesis would imply

that interference with thyroid gland formation by the antagonists and thus early, low level production of TH, would be read out by decreased basal signaling from the reporter gene, in the absence or presence of an exogenous T3 stimulus. However, using MFA, we only saw effects in the presence of the T3 challenge. A possible explanation is that some of the genes involved in thyroid gland formation are actually stimulated by T3 and that the antagonists are playing out their effects against such a scenario. One argument that bolsters this hypothesis is the fact that the follicular NIS gene is activated by T3 (22), providing an explanation for the antiT3 effects of perchlorate. Methimazole could also block this enhancement of thyroid formation. Having established that both agonists and different categories of antagonists could be detected by MFA, we next analyzed whether the action of known TH EDCs could be confirmed. We analyzed bisphenol A, a monomer used in plastic production and tetrabromobisphenol A, a brominated derivative, one of the most used flame-retardants. Both chemicals disrupt estrogen and TH action, as these chemicals are known to bind ER and TR, at high, but potentially environmentally relevant, doses (23, 24). BPA had no effect alone, whereas in the presence of T3, BPA (g10-6M) significantly inhibited, by between 20 and 30%, T3 signaling (Figure 5A). Similarly, TBBPA (Figure 5B) at the same concentrations, inhibited by 20% the T3-dependent signal (p < 0.05). These results compare favorably, as regards their sensitivity and significance, with results obtained in longer-term metamorphic assays. For instance, when BPA and TBBPA were tested against T3 induction of tail regression in Rana rugosa (25), significative antithyroidian effects of 10-6M BPA VOL. 41, NO. 16, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

5911

FIGURE 4. Stage 45 TH reporter-tadpoles show significant responses to different categories of thyroid inhibitors. Inhibitors of TH action or thyroid gland function itself can be detected within 72 h using NF 45 F1 generation transgenic TH/bZIP-eGFP tadpoles. Each substance was tested alone or in the presence of T3 5nM or T4 10 nM. (A) Transcriptional responses in tadpoles treated with 2 µM NH-3. (B) Transcriptional responses in tadpoles treated with 3.56 µM (500 µg/L) sodium perchlorate. (C) Transcriptional responses in tadpoles treated with 1 mM methimazole. (D) Transcriptional responses in tadpoles treated with 10 µM iopanoic Acid. TH/bZIP-eGFP transcription was measured using Ultra Evolution multiple well plate reader set at gain 60, z: 5850 µm. Values shown are means (SE (n ) 20/group). Each experiment was done at least three times providing the same result. Statistical differences were assessed using the Mann-Whitney nonparametric test. *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 5. Significant effects of Bisphenol A (BPA) and Tetrabromobisphenol A (TBBPA) can be detected within 72 h on NF 45 transgenic TH/bZIP-eGFP tadpoles. (A) Transcriptional responses in tadpoles treated for 72 h with 10-6 M to 10-5 M of BPA, with or without T3 5.10-1 M. (B) Transcriptional responses in tadpoles treated for 72 h with 10-7 M to 10-6 M of TBBPA, with or without T3 5.10-9 M. TH/bZIP-eGFP transcription was measured using an Ultra Evolution multiple well plate reader set at gain 60, z: 5980 µm. Values shown are means (SE (n = 20/group). One way ANOVA, two tailed, r = 5% followed by Dunnet’s test was performed for substances alone or against a T3 challenge. Each experiment was done four times. Typical experiments are shown. *p < 0.05, **p < 0.01. were seen after 9 days. Similarly, using the long term X. laevis metamorphosis assay (21 days) (12) TBBPA inhibited larval 5912

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 16, 2007

development at 0.9 10-6M. We detected the same effect at similar concentrations within 72 h. Possible pathways

implicated in the effects of TBBPA and BPA include binding to TRs (23, 26). Bisphenol solubility in aqueous incubation media and their possible adherence to plastic material can cause experimental problems impairing the interpretation of results. Experiments carried out with radio-labeled compounds demonstrated that the solubility of [14C]-TBBPA using our preparation protocol was 90% ( 2%, while the solubility of [3H]-BPA was close to 100%. Further experiments will be necessary to determine the exact proportion of bisphenols absorbed by tadpoles and the possible biotransformation of these molecules. Disruptors of thyroid function are of major concern for public health issues, as TH homeostasis and signaling is essential to vertebrate brain development, from early embryogenesis (8) to later foetal and perinatal development in mammals (27, 28). Two lines of reasoning underline the need to assess potential effects of EDCs in early development. First, current thinking emphasizes the detrimental effects in later life of early disruption of endocrine signaling (29). This concept is particularly important for TH as a number of recent reports have appeared showing that TH signaling is present during early vertebrate neurogenesis (8). These experimental findings corroborate clinical data showing free T4 (FT4) levels in the first trimester of pregnancy to be more significantly correlated with infant psychomotor development than at later stages of pregnancy (30). Second, there is increasing accumulation of potential or confirmed EDCs in the environment and in human tissues. For instance in 2006, worldwide production of BPA reached 2 214 000 million metric tons (31) and mean concentrations of BPA of 8.3 ng/ mL (3.6 10-8M) have been found in samples of human amniotic fluid values reaching 17 ng/mL (7.4 10-8M). Furthermore, a potential source of human exposure is tap water (for review, see ref 32). Indeed, almost 95% of urine samples taken from U.S. citizens contained detectable levels of BPA (between 0.4 µg/L (1.7 10-9M) and 8 µg/L (3.5 10-8M) (33)), results confirmed by studies measuring blood levels in European citizens (34). Moreover, as certain EDCs affect multiple pathways, there is a need to test for potentially broader actions. An example is BPA, initially described as a mild estrogenic agonist that is also a TR antagonist (23, 35, 36). High levels of BPA have been suggested as being associated with changes in psychoneural development in human infants (37). The possibility that the combinatorial effects of EDCs may be affecting early embryogenesis and neurogenesis in humans underlines the critical need for rapid, reproducible screening methods. Also, there is a need to detect EDCs in food and environmental samples. It should be possible to use the methodology we describe to assess thyroid disrupting effects of food extracts added to aquarium water. In conclusion, this is the first report to use a tetrapod vertebrate embryo in a multiwell assay to detect both TH agonists and antagonists. This versatile screening method could be applied, using different enhancers coupled with distinct fluorochromes, to study in vivo effects on multiple signaling systems induced by a single chemical.

Acknowledgments This work was supported by grants from CNRS, MNHN, and EU contracts n° 506319 CASCADE, n° 512065 Xomics, n° FP6018652 CRESCENDO, and the French Ministry of Environment (MEDD, PNR). We thank Gerard Benisti, Jean-Paul Chaumeil, and Etienne LeGoff for excellent animal care. We thank Watchfrog and Genopole (Evry) for technical support.

Supporting Information Available Supporting Information Figure 1 illustrates differences between stage NF45 and pre-metamorphic (stage NF51-55)

tadpoles. The small size of stage 45 Xenopus laevis tadpoles (overall length, 8 mm) allows them to be easily placed individually in wells of a 96 well plate, suitable for robotized reading. This material is available free of charge via the Internet at http://pubs.acs.org.

Glossary BPA

Bisphenol A

CNS

central nervous system

EDCs

endocrine disrupting chemicals

eGFP

enhanced green fluroescent protein

MFA

Multiwell Fluorescence Assay

NF

Nieuwkoop-Faber stages

NIS

sodium-iodine symporter

OECD

Organisation for Economic Cooperation and Development

PCR

polymerase chain reaction

REACH

Registration Evaluation Authorization of Chemicals

RFU

relative fluorescence unit

T3

3,5,3′-tri-iodo-L-thyronine

T4

3,5,3′,5′-tetra-iodo-L-thyronine

TBBPA

tetrabromobisphenol A

TH

thyroid hormones

TPO

thyroid peroxydase

TR

thyroid hormone receptor

TRE

thyroid hormone responsive element

TRIAC

3,5,3′-triiodothyroacetic acid

TRβ

thyroid hormone receptor isoform beta

XEMA

Xenopus metamorphosis assay

Literature Cited (1) REACH, The New EU Chemicals Legislation, 2006. http:// eur-lex.europa.eu/LexUriServ/site/en/com/2006/ com2006_0375en01.pdf. (2) Brucker-Davis, F. Effects of environmental synthetic chemicals on thyroid function. Thyroid. 1998, 8, 827-856. (3) Tabb, M. M.; Blumberg, B. New modes of action for endocrinedisrupting chemicals. Mol. Endocrinol. 2006, 20, 475-482. (4) Leloup, J.; Buscaglia, M. [article in french] La triiodothyronine : hormone de la me´tamorphose des amphibiens. C. R. Acad. Sci. 1977, 284, 2261-2263. (5) Normal Table of Xenopus laevis; Nieuwkoop, P. D., Faber, J., Eds.; Garland Publishing: New York, 1994. (6) Honda, J.; Ogawa, K.; Taniguchi, K. [Immunohistochemical and morphometric studies on the development of the thyroid, parathyroid and ultimobranchial body in Xenopus laevis Daudin]. Jikken Dobutsu. 1993, 42, 23-32. (7) Tindall, A. J.; Morris, I. D.; Pownall, M. E.; Isaacs, H. V. Expression of enzymes involved in thyroid hormone metabolism during the early development of Xenopus tropicalis. Biol. Cell 2006, 99, 151-163. (8) Morvan Dubois, G.; Sebillot, A.; Kuiper, G. G.; Verhoelst, C. H.; Darras, V. M.; Visser, T. J.; Demeneix, B. A. Deiodinase activity is present in Xenopus laevis during early embryogenesis. Endocrinology 2006, 147, 4941-4949. (9) Havis, E.; Le Mevel, S.; Morvan Dubois, G.; Shi, D. L.; Scanlan, T. S.; Demeneix, B. A.; Sachs, L. M. Unliganded thyroid hormone receptor is essential for Xenopus laevis eye development. EMBO J. 2006, 25, 4943-4951. (10) Turque, N.; Palmier, K.; Le Mevel, S.; Alliot, C.; Demeneix, B. A. A rapid, physiologic protocol for testing transcriptional VOL. 41, NO. 16, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

5913

(11)

(12)

(13) (14)

(15)

(16)

(17)

(18) (19) (20) (21)

(22)

(23)

(24)

(25)

effects of thyroid-disrupting agents in premetamorphic Xenopus tadpoles. Environ. Health Perspect. 2005, 113, 15881593. Furlow, J. D.; Brown, D. D. In vitro and in vivo analysis of the regulation of a transcription factor gene by thyroid hormone during Xenopus laevis metamorphosis. Mol. Endocrinol. 1999, 13, 2076-2089. Jagnytsch, O.; Opitz, R.; Lutz, I.; Kloas, W. Effects of tetrabromobisphenol A on larval development and thyroid hormoneregulated biomarkers of the amphibian Xenopus laevis. Environ. Res. 2006, 101, 340-348. Kloas, W. Amphibians as a model for the study of endocrine disruptors. Int. Rev. Cytol. 2002, 216, 1-57. OECD. Detailed Review Paper on Amphibian Metamorphosis Assay for the Detection of Thyroid Active Substances; Series on testing and assessment N°46; Organisation for Economic Cooperation and Development: Paris, 2006. Zalko, D.; Prouillac, C.; Riu, A.; Perdu, E.; Dolo, L.; Jouanin, I.; Canlet, C.; Debrauwer, L.; Cravedi, J. P. Biotransformation of the flame retardant tetrabromo-bisphenol A by human and rat sub-cellular liver fractions. Chemosphere. 2006, 64, 318-327. Chiellini, G.; Apriletti, J. W.; Yoshihara, H. A.; Baxter, J. D.; Ribeiro, R. C.; Scanlan, T. S. A high-affinity subtype-selective agonist ligand for the thyroid hormone receptor. Chem. Biol. 1998, 5, 299-306. Eliceiri, B. P.; Brown, D. D. Quantitation of endogenous thyroid hormone receptors R and β during embryogenesis and metamorphosis in Xenopus laevis. J. Biol. Chem. 1994, 269, 2445924465. Lim, W.; Nguyen, N. H.; Yang, H. Y.; Scanlan, T. S.; Furlow, J. D. A thyroid hormone antagonist that inhibits thyroid hormone action in vivo. J. Biol. Chem. 2002, 277, 35664-35670. Cooper, D. S. Antithyroid drugs. N. Engl. J. Med. 2005, 352, 905-917. Wolff, J. Perchlorate and the thyroid gland. Pharmacol. Rev. 1998, 50, 89-105. Tietge, J. E.; Holcombe, G. W.; Flynn, K. M.; Kosian, P. A.; Korte, J. J.; Anderson, L. E.; Wolf, D. C.; Degitz, S. J. Metamorphic inhibition of Xenopus laevis by sodium perchlorate: effects on development and thyroid histology. Environ. Toxicol. Chem. 2005, 24, 926-933. Schmutzler, C.; Schmitt, T. L.; Glaser, F.; Loos, U.; Kohrle, J. The promoter of the human sodium/iodide-symporter gene responds to retinoic acid. Mol. Cell. Endocrinol. 2002, 189, 145155. Moriyama, K.; Tagami, T.; Akamizu, T.; Usui, T.; Saijo, M.; Kanamoto, N.; Hataya, Y.; Shimatsu, A.; Kuzuya, H.; Nakao, K. Thyroid hormone action is disrupted by bisphenol A as an antagonist. J. Clin. Endocrinol. Metab. 2002, 87, 5185-5190. Kitamura, S.; Jinno, N.; Ohta, S.; Kuroki, H.; Fujimoto, N. Thyroid hormonal activity of the flame retardants tetrabromobisphenol A and tetrachlorobisphenol A. Biochem. Biophys. Res. Commun. 2002, 293, 554-559. Goto, Y.; Kitamura, S.; Kashiwagi, K.; Oofusa, K.; Tooi, O.; Yoshizato, K.; Sato, J.; Ohta, S.; Kashiwagi, A. Suppression of

5914

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 16, 2007

(26)

(27)

(28) (29) (30)

(31) (32)

(33)

(34)

(35)

(36)

(37)

amphibian metamorphosis by bisphenol A and related chemical substances. J. Health Sci. 2006, 52, 160. Kitamura, S.; Suzuki, T.; Sanoh, S.; Kohta, R.; Jinno, N.; Sugihara, K.; Yoshihara, S.; Fujimoto, N.; Watanabe, H.; Ohta, S. Comparative study of the endocrine-disrupting activity of bisphenol A and 19 related compounds. Toxicol. Sci. 2005, 84, 249-259. Zoeller, T. R.; Dowling, A. L.; Herzig, C. T.; Iannacone, E. A.; Gauger, K. J.; Bansal, R. Thyroid hormone, brain development, and the environment. Environ. Health Perspect. 2002, 110 (Suppl. 3), 355-361. Bernal, J. Thyroid hormones and brain development. Vitam. Horm. 2005, 71, 95-122. Gore, A. C.; Heindel, J. J.; Zoeller, R. T. Endocrine disruption for endocrinologists (and others). Endocrinology 2006, 147, S1-3. Pop, V. J.; Kuijpens, J. L.; van Baar, A. L.; Verkerk, G.; van Son, M. M.; de Vijlder, J. J.; Vulsma, T.; Wiersinga, W. M.; Drexhage, H. A.; Vader, H. L. Low maternal free thyroxine concentrations during early pregnancy are associated with impaired psychomotor development in infancy. Clin Endocrinol (Oxf). 1999, 50, 149-155. Burridge, E. Bisphenol A: product profile. Eur. Chem. News 2003, April, 14-20. vom Saal, F. S.; Hughes, C. An extensive new literature concerning low-dose effects of bisphenol A shows the need for a new risk assessment. Environ. Health Perspect. 2005, 113, 926933. Calafat, A. M.; Kuklenyik, Z.; Reidy, J. A.; Caudill, S. P.; Ekong, J.; Needham, L. L. Urinary concentrations of bisphenol A and 4-nonylphenol in a human reference population. Environ. Health Perspect. 2005, 113, 391-395. WWF. Chemical check up An analysis of chemicals in the blood of members of the European Parliament. DETOX campaign 2004. World Wildlife Fund. www.panda.org/news_facts/newsroom/ press_releases/news.cfm?uNewsID)12622. Zoeller, R. T.; Bansal, R.; Parris, C. Bisphenol-A, an environmental contaminant that acts as a thyroid hormone receptor antagonist in vitro, increases serum thyroxine, and alters RC3/neurogranin expression in the developing rat brain. Endocrinology 2005, 146, 607-612. Iwamuro, S.; Yamada, M.; Kato, M.; Kikuyama, S. Effects of bisphenol A on thyroid hormone-dependent up-regulation of thyroid hormone receptor alpha and beta and down-regulation of retinoid X receptor gamma in Xenopus tail culture. Life Sci. 2006. Welshons, W. V.; Nagel, S. C.; vom Saal, F. S. Large effects from small exposures. III. Endocrine mechanisms mediating effects of bisphenol A at levels of human exposure. Endocrinology 2006, 147, S56-69.

Received for review February 17, 2007. Revised manuscript received April 30, 2007. Accepted May 3, 2007. ES0704129