Nfl

AYa

b

a

ARRAA

KTBMBLB

1

gtaopitHcuFF

eAf

0d

Toxicology Letters 189 (2009) 78–83

Contents lists available at ScienceDirect

Toxicology Letters

journa l homepage: www.e lsev ier .com/ locate / tox le t

eurobehavioral effects of tetrabromobisphenol A, a brominatedame retardant, in mice

kira Nakajimaa,1, Daisuke Saigusaa,b,1, Naomi Tetsua, Tohru Yamakunia,oshihisa Tomiokab,∗, Takanori Hishinumaa,b

Laboratory of Pharmacotherapy, Graduate School of Pharmaceutical Sciences, Tohoku University, Aoba, Aramaki, Aoba-ku, Sendai 980-8578, JapanLaboratory of Oncology, Pharmacy Practice and Sciences, Graduate School of Pharmaceutical Sciences, Tohoku University, Aoba, Aramaki, Aoba-ku, Sendai 980-8578, Japan

r t i c l e i n f o

rticle history:eceived 7 February 2009eceived in revised form 10 May 2009ccepted 12 May 2009vailable online 20 May 2009

eywords:etrabromobisphenol A

a b s t r a c t

Tetrabromobisphenol A (TBBPA) is widely used as a flame retardant and is suspected to be stable inthe environment with possible widespread human exposures. In the present study, we investigated thebehavioral effects of TBBPA and measured the levels of TBBPA in the brain after oral administrationin mice. Acute treatment with TBBPA (5 mg/kg body weight) 3 h before the open-field test induced anincrease in the horizontal movement activities. In contextual fear conditioning paradigm, mice treatedwith TBBPA (0.1 mg/kg or 5 mg/kg body weight) showed more freezing behavior than vehicle-treatedmice. In addition, TBBPA (0.1 mg/kg body weight) significantly increased the spontaneous alternationbehavior in the Y-maze test. The levels of TBBPA in the brain following TBBPA treatment were determined

ehavior

icerainC/MS/MSrominated flame retardant

by using LC/ESI-MS/MS system. In the brain regions examined, high amounts of TBBPA were detectedin the striatum after treatment with 0.1 mg/kg or 5 mg/kg body weight TBBPA, whereas non-specificaccumulation of TBBPA in the brain was found after treatment with 250 mg/kg body weight TBBPA. Theseresults suggest that TBBPA accumulates in brain regions including the striatum and induces the behavioralalterations. Together, the possibility of widespread human exposure to TBBPA warrants further studies tocharacterize its neurotoxicity.

. Introduction

Tetrabromobisphenol A (TBBPA) is a highly lipophilic halo-enated aromatic molecule and currently the most widely usedype of brominated flame retardant (BFR) (de Wit, 2002). TBBPAnd other BFRs are employed as additives in the manufacturing offfice and home electronic equipment, such as computer boards,rinters, mobile phones, televisions, and washing machines. TBBPA

s primarily used as a chemical bound flame retardant (∼90%) and isherefore not expected to reach the environment in larger amounts.owever, Sellström and Jansson (1995) showed that this compound

ould leak from treated products and Osako et al. (2004) reportedp to 620 ng/L TBBPA in leachate samples from landfills in Japan.urther, TBBPA is found in significant amounts in human samples.or example, TBBPA in concentrations of 0.3–1.8 ng/g of lipid was

∗ Corresponding author at: Laboratory of Oncology, Pharmacy Practice and Sci-nces, Graduate School of Pharmaceutical Sciences, Tohoku University, 6-3 Aoba,ramaki, Aoba-ku, Sendai 980-8578, Japan. Tel.: +81 22 795 6851;

ax: +81 22 795 6850.E-mail address: ytomioka@mail.pharm.tohoku.ac.jp (Y. Tomioka).

1 A.N. and D.S. contributed equally to this work.

378-4274/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reserved.oi:10.1016/j.toxlet.2009.05.003

© 2009 Elsevier Ireland Ltd. All rights reserved.

found in plasma samples from individuals who dismantle electroniccomponents (Thomsen et al., 2001). Investigation of concentra-tions in serum taken from hospital patients in Norway revealedslight increases from 1985 to 1999 with the highest values (average0.71 ng/g lipids) detected in young children up to 4 years (Thomsenet al., 2002). Since the half-life of TBBPA is reported to be about 2days in humans, these results indicate continuous uptake of thiscompound (Sjödin et al., 2003).

So far, few effect studies have been carried out on TBBPA. Somehave been found to elicit thyroidogenic, estrogenic and dioxin-likeactivity at relatively high concentrations (Darnerud et al., 2001;Birnbaum and Staskal, 2004). Information about neurotoxicityis very limited. An in vitro study indicated that TBBPA inhib-ited plasma membrane uptake of the neurotransmitters includingdopamine, glutamate and �-amino-n-butyric acid (GABA) at aconcentration level similar to what previously found for polychlo-rinated biphenyls (PCBs) and even for a drug of abuse, ecstasy(Mariussen and Fonnum, 2003). TBBPA also inhibited the vesicu-

lar uptake of dopamine with IC50 value of 3 �M (Mariussen andFonnum, 2003). In contrast, TBBPA did not result in behavioraleffects in adult mice following a single neonatal exposure (Erikssonet al., 2001). Lilienthal et al. (2008) reported that an average dailyintake of TBBPA throughout life did not induce significant effects

logy L

oTha

3itodttib

2

2

dCUhACuA

2

2

(Ttspalo(orfi

2

sN(mclmaa

2

qrmEi2ga2am

2

ao2

A. Nakajima et al. / Toxico

n fear memory in adult rats. However, maternal oral exposure toBBPA from gestational 7 to postnatal day 18 resulted in impairedabituation of motor activity and spatial learning in adult rats (Hassnd Wamberg, 2002).

Since it is reported that TBBPA concentrations in plasma peakh after oral administration in rodents (Schauer et al., 2006), we

nvestigated the effects of TBBPA on the open-field activity, spa-ial working memory and contextual fear memory 3 h after acuteral administration in mice in the present study. In addition, weeveloped an analytical method for TBBPA using LC/ESI-MS/MS sys-em and examined the levels of TBBPA in the brain following TBBPAreatment. We show here that TBBPA is detected in brain regionsncluding the striatum following TBBPA treatment and inducesehavioral alterations in mice.

. Materials and methods

.1. Chemicals

Chemical structure of TBBPA is shown in Supplementary Fig. 1. TBBPA stan-ard (99%) and 13C12-TBBPA (99%, 50 mL/mL in methanol) were obtained from Tokyohemical Industry (Tokyo, Japan) and Cambridge Isotope Laboratories (Andover, MA,SA), respectively. Methanol (MeOH) of LC/MS-grade and acetonitrile (CH3CN), n-exane and ethyl acetate (AcOEt) of LC-grade were obtained from Kanto chemical.mmonium acetate (CH3COONH4) of LC/MS-grade was obtained from Wako Purehemical industries. Water was of ultra pure-grade from Organo. Abselut Nexus col-mn (polystyrene divinylbenzene, 60 mg, 3 mL) were purchased from Varian (Palolto, CA, USA).

.2. Development of determination method for TBBPA

.2.1. Preparation methodThe tissue sample was placed into a 2 mL sample tube. Then 1 mL of solution

CH3CN/H2O, 70/30, v/v) and 40 �L (100 pg/�L) of an IS (13C12-TBBPA) were added.his mixture was homogenized for 1 min by the ultrasonic bath. After centrifuga-ion at 16,400 × g for 20 min, the supernatant was removed and transferred to otherample tube with 1 mL of n-hexane. After centrifugation at 16,400 × g for 20 min, thehase of n-hexane was removed and 2 mL of H2O (liquid–liquid extraction; LLE) wasdded to the phase of CH3CN. Then, the CH3CN/H2O fraction was loaded on an Abse-ut Nexus column that was conditioned with 2 mL of AcOEt, 2 mL of CH3CN and 2 mLf H2O. The cartridge was washed with 3 mL of H2O, 3 mL of the organic solutionCH3CN/H2O, 40/60, v/v), and 3 mL of n-hexane. Then, TBBPA was eluted with 3 mLf AcOEt and the eluted fraction was dried under nitrogen stream. Finally the driedesidue was dissolved with 40 �L of MeOH, mixed for 30 s and passed through theler (pore size 0.2 �m). The filtered solution was retained for further LC/ESI-MS/MS.

.2.2. Liquid chromatography systemThe column for the present study was a CAPCELL PAK C18 MGII (Shi-

eido, 1.5 mm × 150 mm, 5 �m) maintained at 40 ◦C. The LC system consisted ofANOSPACE SI-2 was used, comprising LC pump, auto-sampler and on-line degasser

Shiseido). Gradient elution was performed using 10 mM CH3COONH4 (eluent A) andethanol/10 mM CH3COONH4 = 90/10, v/v, (eluent B). The gradient elution initial

onditions were 35% B with a linear gradient to 90% from 4 to 5 min, followed by ainear gradient to 100% B at 6 min, this proportion being maintained for 5 min. The

obile phase was then returned to the initial conditions immediately after 11 minnd maintained until the end of the run at 15 min. The flow rate was 200 �L/minnd the injection volume was 10 �L.

.2.3. Mass spectrometry systemThe MS system was TSQ quantum ultra (Thermo Fisher Scientific) triple

uadrupole mass spectrometer with ESI source with operating conditions summa-ized in Supplementary Table 1. Mass spectrometer was operated in negative ionode and tuned using the built-in auto-tuning system. Operating parameters of the

SI interface of MS/MS were optimized in full scan mode (m/z 300–600) using annfusion system of TBBPA at 5 �L/min by means of syringe pump. The spray voltage.5 kV, heated capillary temperature 300 ◦C, sheath gas pressure 40 psi, auxiliaryas setting 5 psi and heated vaporizer temperature 300 ◦C. Both the sheath gas anduxiliary gas used were nitrogen gas. The collision gas was argon at a pressure of.7 mTorr. For the MS/MS analysis, the optimized tube lens offset was 97 V and rel-tive collision energies for collision-induced dissociation (CID) was 35 eV (TBBPA,/z 543 → 291) or 33 eV (13C12-TBBPA, m/z 555 → 297).

.2.4. Quantification and validationPeak areas were used for the quantification of TBBPA. All peaks were integrated

utomatically. The TBBPA amounts were calculated on the calibration curve by ratiof the peak area of TBBPA to that of IS. The ranges used for calibration curve were from5 to 5000 pg. To indicate accuracy and precision for determination of TBBPA, data

etters 189 (2009) 78–83 79

were validated from three concentrations: 50, 500 and 5000 pg. The detection limitwas defined as the absolute amount of an analyte that yields a signal to noise ratio of3. The data validated on the bases of Bioanalytical Method Validation from Guidancefor Industry and standard errors (mean ± SEM) were calculated using Microsoft Excelspreadsheets.

2.3. Behavioral analysis

We examined the neurobehavioral effects of TBBPA in the open-field test, contex-tual fear conditioning paradigm and Y-maze test. The reasons why we chose thesebehavioral tests were as follows. Assessment of vertical and/or horizontal move-ments (e.g., motor activity) in the open-field test is non-invasive and has beenused to evaluate the effects of acute and repeated exposure to many chemicals(Maurissen and Mattsson, 1989). This analysis is incorporated in the guidelines forthe assessment of neurotoxicity (OECD TG424). Contextual fear conditioning is aform of Pavlovian learning in which an association is made between stimuli andtheir aversive consequences. Clinically, it has relevance to human behavior sincefear conditioning can be produced in humans (LeDoux, 2000). It is a form of one-trial learning, and engages the hippocampus and amygdala (Phillips and LeDoux,1992). Hippocampus- and amygdala-dependent one-trial learning tasks are used toassess chemical effects on cognitive functions (OECD guidance document number20 for neurotoxicity testing). Spatial working memory was assessed by recordingspontaneous alternation behavior in a Y-maze. Sustained alternating performancein a Y-maze task is based upon the assumption that mice will explore their environ-ment in a repeating and alternating fashion in order to optimize the amount of areainvestigated. Normal alternation performance in this task is consistent with an intactworking memory ability. The task is sensitive to drugs or toxins that either enhanceor impair spatial memory (Sarter et al., 1988; Maurice et al., 1994). Spatial mazes arecommonly used to assess chemical effects on cognitive functions (OECD guidancedocument number 20 for neurotoxicity testing). The open-field test, contextual fearconditioning paradigm and Y-maze test were previously used in our laboratory toexamine the neurobehavioral effects of a citrus flavonoid nobiletin (Nagase et al.,2005; Nakajima et al., 2007; Onozuka et al., 2008).

2.3.1. AnimalsMale closed colony ddY mice (12 ± 2.0 g, 3 weeks old) were obtained from Nip-

pon SLC (Hamamatsu, Japan). Animals were housed in cages with free access tofood and water under the condition of constant temperature (23 ± 1 ◦C) and humid-ity (55 ± 5%) and adapted to a standard 12-h light/12-h dark cycle (light cycle:9:00–21:00). The procedures used in this study were approved by the Committeeon the Care and Use of Experimental Animals, Tohoku University in accordance withthe Guide for the Care and Use of Laboratory Animals published by the US NationalInstitute of Health.

2.3.2. TreatmentTBBPA (99%, Tokyo Chemical Industry, Tokyo, Japan) was completely dissolved

in corn oil and administered orally at a volume of 20 mL/kg body weight. TBBPA(0.1 mg/kg, 5 mg/kg or 250 mg/kg body weight) was administered 3 h before theopen-field test, Y-maze test or training of contextual fear conditioning paradigm.Control animals were treated with the same volume of vehicle (corn oil). Sixty-four mice were used for the open-field test. These consisted of 17 control mice, 16TBBPA (0.1 mg/kg)-exposed mice, 17 TBBPA (5 mg/kg)-exposed mice and 14 TBBPA(250 mg/kg)-exposed mice. Fifty-nine mice were used for the contextual fear condi-tioning paradigm. These consisted of 15 control mice, 15 TBBPA (0.1 mg/kg)-exposedmice, 15 TBBPA (5 mg/kg)-exposed mice and 14 TBBPA (250 mg/kg)-exposed mice.Forty-eight mice were used for the Y-maze test. These consisted of 12 controlmice, 12 TBBPA (0.1 mg/kg)-exposed mice, 12 TBBPA (5 mg/kg)-exposed mice and12 TBBPA (250 mg/kg)-exposed mice. The doses of TBBPA were chosen on the basisof a previous report showing that maternal oral exposure to 50 mg/kg or 250 mg/kgbody weight TBBPA resulted in neurobehavioral effects in rats (Hass and Wamberg,2002). We did not have a positive control, since a test chemical showed a positiveeffect.

2.3.3. Open-field testThe open-field test was performed as described previously (Onozuka et al.,

2008). Mice were placed into the corner of a wooden box (50 cm × 50 cm × 40 cm)and allowed to freely explore for 5 min. The floor of the field was divided into 25identical squares so that the ambulation of animals could be measured. The ambu-lation of the mice was measured by manually counting the number of times that theanimals crossed from one square to another. The number of rearing and groomingevents was also manually recorded.

2.3.4. Contextual fear conditioningContextual fear conditioning paradigm was performed as described previously

(Nakajima et al., 2007). Animals were placed in the training chamber and allowedto explore for 2 min, after which they received an electric shock (2 s, 0.7 mA).The 2 min/2 s shock paradigm were repeated for a total of three shocks. After thelast shock, animals were allowed to explore the context for an additional 1 minprior to removal from the training chamber. During training, freezing behavior wasmeasured during 1 min after each shock. To assess contextual fear memory, the

80 A. Nakajima et al. / Toxicology Letters 189 (2009) 78–83

F ivities (B), and grooming behaviors (C) in the open-field test. TBBPA (0.1 mg/kg, 5 mg/kg or2 mber of passed squares and rearings was evaluated as horizontal and vertical movementa 16; 5 mg/kg TBBPA, n = 17; 250 mg/kg TBBPA, n = 14). *p < 0.05 vs control.

afe

2

b23oTbtwita

2

tciaL

2

imAs

3

3

t

Fig. 2. Effects of acute treatment with TBBPA on contextual fear memory. Animals

ig. 1. Effects of acute treatment with TBBPA on horizontal activities (A), vertical act50 mg/kg body weight, p.o.) was administered 3 h before the open-field test. The nuctivities, respectively. Values are mean ± SEM (control, n = 17; 0.1 mg/kg TBBPA, n =

nimals were placed back into the training context 24 h after training and scoredor freezing for 5 min. Freezing behavior was measured by observing the animalsvery 5 s.

.3.5. Y-maze testSpatial working memory was assessed by recording spontaneous alternation

ehavior during an 8-min session in a Y-maze, as described previously (Nagase et al.,005). The maze was made of black Plexiglas. Each arm was 40 cm long, 12 cm high,cm wide at the bottom, and 10 cm wide at the top. Each mouse was placed at the endf one arm and allowed to move freely through the maze during an 8-min session.he sequence of arm entries was recorded manually. Arm entry was considered toe completed when the hind paws of the mouse had been completely placed inhe arm. The alternation behavior defined as consecutive entries into all three armsithout repeated entries was expressed in percent of the total arm entries. If, for

nstance, the arms are called A, B, C, and the animal performs: ABCBABCBACBAACB, theotal alternation opportunities would be 13 (total entries minus 2) and the percentlternation behavior would be 61.5% (8 out of 13).

.4. Determination of TBBPA levels in brain samples

Mice used in a Y-maze test were sacrificed immediately after completion of theest, and their brains were rapidly removed. The hippocampus, striatum, cortex,erebellum, midbrain, thalamus and medulla-oblongata were dissected out on ance-cold plate. Each brain sample was quickly frozen and stored in a deep freezert −80 ◦C until assayed. TBBPA levels in brain samples were determined by usingC/MS/MS system as described above.

.5. Statistical analyses

The results are expressed as mean ± SEM. The data for the freezing during train-ng of contextual fear conditioning paradigm were analyzed by two-way repeated

easures analysis of variance (ANOVA). Other data were analyzed by one-wayNOVA, followed by the Student–Neumann–Keuls test. A level of p < 0.05 was con-idered statistically significant.

. Results

.1. Behavioral effects of TBBPA

Visual inspection did not reveal gross abnormalities in TBBPA-reated mice. We examined the effects of TBBPA on general

were exposed to the contextual fear conditioning paradigm, and freezing behaviorwas measured either during training (A) or test (B). TBBPA (0.1 mg/kg, 5 mg/kg or250 mg/kg body weight, p.o.) was administered 3 h before training of the contextualfear conditioning paradigm. The test session was performed 24 h after training. Val-ues are mean ± SEM (control, n = 15; 0.1 mg/kg TBBPA, n = 15; 5 mg/kg TBBPA, n = 15;250 mg/kg TBBPA, n = 14). *p < 0.05 vs control.

A. Nakajima et al. / Toxicology L

Fig. 3. Effects of acute treatment with TBBPA on spatial working memory in a Y-mazetest. (A) Alternation performance. (B) Number of arm entries. TBBPA (0.1 mg/kg,5tn

bstash

negative ion mode by infusing a solution of the compound, directly

FTm

mg/kg or 250 mg/kg body weight, p.o.) was administered 3 h before the Y-mazeest. Values are mean ± SEM (control, n = 12; 0.1 mg/kg TBBPA, n = 12; 5 mg/kg TBBPA,= 12; 250 mg/kg TBBPA, n = 12). *p < 0.05 vs control.

ehaviors in the open-field test. We evaluated the number of passedquares and rearings as horizontal and vertical movement activi-

ies. The curve for the effects of TBBPA on horizontal movementctivity was bell-shaped, and the dose of 5 mg/kg body weight hadignificant effect (p = 0.0388 by one-way ANOVA; p < 0.05 by postoc) (Fig. 1A). There were no significant differences in the number

ig. 4. Levels of TBBPA in various brain regions after acute treatment with 0.1 mg/kg boBBPA levels in brain samples were determined by using LC/MS/MS system. Values areedulla-oblongata. #p < 0.05 vs hippocampus, cortex, cerebellum, midbrain, thalamus or

etters 189 (2009) 78–83 81

of rearing and grooming behaviors between groups (p = 0.3514 and0.7389 by one-way ANOVA, respectively) (Fig. 1B and C).

We next examined the effects of TBBPA on learning andmemory abilities in the contextual fear conditioning paradigm,which is characterized by the association of a normally innocu-ous context with an aversive stimulus (Phillips and LeDoux, 1992;Atkins et al., 1998). Treatment with TBBPA did not affect freez-ing behavior during training (treatment × shock event interaction,p = 0.6011) (Fig. 2A). In the test session performed 24 h after train-ing, mice injected with TBBPA (0.1 mg/kg or 5 mg/kg body weight)showed more freezing behavior than vehicle-treated control mice(p = 0.0135 by one-way ANOVA; p < 0.05 by post hoc) (Fig. 2B). Incontrast, TBBPA (250 mg/kg body weight) had no effects on freezingbehavior in the test session (p > 0.05 by post hoc) (Fig. 2B).

We next investigated the effects of TBBPA on spatial workingmemory in a Y-maze test. During an 8-min session, vehicle-treated control mice showed 48.2 ± 10.1% spontaneous alternationand 18 ± 2 arm entries. Treatment with TBBPA (0.1 mg/kg bodyweight) significantly increased the spontaneous alternation behav-ior (p = 0.0211 by one-way ANOVA; p < 0.05 by post hoc) withoutaffecting the number of arm entries (p = 0.1436 by one-way ANOVA)(Fig. 3A and B). In contrast, treatment with higher doses of TBBPA(50 mg/kg or 250 mg/kg body weight) did not result in significanteffects on the spontaneous alternation behavior as well as numberof arm entries (p > 0.05 by post hoc, p = 0.1436 by one-way ANOVA,respectively) (Fig. 3A and B).

3.2. Mass spectrum of TBBPA

We next examined the levels of TBBPA in the brain follow-ing TBBPA treatment. To determine the levels of TBBPA in variousbrain regions, we first developed an analytical method involvingLC/ESI-MS/MS with detection of the precursor ion of TBBPA andits fragmentation pattern. The first step in the development of theanalytical method was to establish the mass spectrum of TBBPA in

into the mass spectrometer. The mass spectrum acquired in fullscan (mass range of m/z 300–600) is illustrated in SupplementaryFig. 2A. The precursor ion dominates the mass spectrum at m/z 543with an isotopic distribution in accordance with the presence of the

dy weight (A), 5 mg/kg body weight (B) or 250 mg/kg body weight (C) of TBBPA.mean ± SEM (n = 3–6). *p < 0.05 vs hippocampus, cortex, cerebellum, thalamus ormedulla-oblongata.

8 logy L

fisaasb

3

astatodctcti

3

ccbaorsltSpwtao

4

swait

toalmpwaipntas

2 A. Nakajima et al. / Toxico

our bromine atoms on the ion. Next we acquired a mass spectrumn MS/MS mode with the parent ion m/z 543 and a normalized colli-ion energy of 35 eV. Four product ions at m/z 293, m/z 291, m/z 289nd m/z 81 were found. The intensity of m/z 291 was the highestmong those four product ions (Supplementary Fig. 2B). Thus, theelected reaction monitoring (SRM) scan type is a valuable strategyy monitoring the MS/MS reaction 543 → 291.

.3. Chromatography and method validation

Chromatograms of a whole brain sample from naïve mice andwhole brain sample from naïve mice spiked with TBBPA are

hown in Supplementary Fig. 3A and B, respectively. A representa-ive chromatogram of a striatum sample from TBBPA-treated micere shown in Supplementary Fig. 3C. Supplementary Table 2 showshe intra- and inter-day accuracy and precision data for the assayf TBBPA in brain tissue samples. They were accurate and repro-ucible as shown in Supplementary Table 2. The calibration graphs,onstructed from peak area ratio of TBBPA to IS vs TBBPA concen-rations, were linear with a correlation coefficient of 0.999 in theoncentration range of 25–5000 pg. The detection limit of TBBPA inhe brain samples was 0.112 pg at a signal to noise ratio of 3. Thenter-day precision data for the assay of TBBPA at 50 pg was 6.96%.

.4. Determination of TBBPA in brain samples

Mice used in a Y-maze test were sacrificed immediately afterompletion of the test, and their brains were dissected and pro-essed for quantification of TBBPA. The levels of TBBPA in variousrain regions determined by LC/ESI-MS/MS system as describedbove are presented in Fig. 4. There was a dose-dependent increasef TBBPA concentrations in brain tissues. Interestingly, in the brainegions examined, high amounts of TBBPA were detected in thetriatum compared with the hippocampus, cortex, cerebellum, tha-amus or medulla-oblongata after 0.1 mg/kg body weight TBBPAreatment (p = 0.0080 by one-way ANOVA; p < 0.05 by post hoc).imilarly, high amounts of TBBPA were found in the striatum com-ared with all other brain regions at the dose of 5 mg/kg bodyeight (p = 0.0287 by one-way ANOVA; p < 0.05 by post hoc). In con-

rast, non-specific accumulation of TBBPA in the brain was foundfter treatment with 250 mg/kg body weight TBBPA (p = 0.0791 byne-way ANOVA).

. Discussion

The present study showed that acute TBBPA exposure produceseveral behavioral alterations in mice. It was also shown that TBBPAas detected in the brain regions including the striatum after oral

dministration. To our knowledge, this is the first study to exam-ne the levels of TBBPA in various brain regions following TBBPAreatment.

In the series of behavioral analyses, we found that acutereatment with 0.1 mg/kg body weight TBBPA resulted in mem-ry enhancement in the contextual fear conditioning paradigmnd Y-maze test. Also, 5 mg/kg body weight TBBPA induced theocomotor-stimulating effect in the open-field test as well as

emory-enhancing effect in the contextual fear conditioningaradigm. In contrast, a higher dose of TBBPA (250 mg/kg bodyeight) did not induce any effects on motor activity and learning

nd memory-related behaviors. TBBPA concentrations in the brainncreased with increasing exposure levels, thus ruling out simple

harmacokinetic explanations of these findings. A possible expla-ation for the lack of a dose–response relationship is that exposureo a high TBBPA dose could trigger a compensation phenomenon,s in the case of exposure to high doses of other compounds,uch as lead (Gilbert et al., 1999). A similar biphasic response has

etters 189 (2009) 78–83

been reported after exposure to a polybrominated diphenyl ether(PBDE99). Mice treated with low and medium doses of PBDE99 (0.6and 6 mg/kg) showed hyperactivity in the open-field test, whereas ahigh dose of PBDE99 (30 mg/kg) did not induce any effects (Branchiet al., 2002). Alternatively, the lack of a dose–response relationshipin the present study may be related to the distribution of TBBPAin the brain after oral administration. We detected high amountsof TBBPA in the striatum compared with other brain regions afterlow and medium doses of TBBPA exposure, whereas non-specificaccumulation was found after a high dose of TBBPA exposure.Although an in vitro study indicates that TBBPA inhibits plasmamembrane uptake of the neurotransmitters including dopamine,glutamate and GABA with IC50 value of 9, 6, and 16 �M, respec-tively (Mariussen and Fonnum, 2003), TBBPA concentrations foundin the present study were considerably lower than those used inMariussen and Fonnum (2003). Therefore, it is not likely that TBBPAinduced behavioral alterations by inhibiting the uptake of the neu-rotransmitters. Further studies are needed to clarify the underlyingmechanisms of the behavioral alterations induced by TBBPA expo-sure. In addition, it remains to be determined if TBBPA inducesneurochemical changes in vivo.

Inconsistent findings on behavioral effects of TBBPA have beenobserved. TBBPA did not result in effects on motor activity, habit-uation and spatial learning and memory in 2- and 4-month-oldmice following neonatal exposure to a single dose of 0.75 mg/kgor 11.5 mg/kg body weight (Eriksson et al., 2001). Lilienthal et al.(2008) reported that an average daily intake of TBBPA ranging from0 to 3000 mg/kg body weight throughout life did not induce signif-icant effects on context- or tone-dependent fear memory at 50–110days of age in rats. In contrast, spatial learning in the Morris watermaze was found to be affected by developmental exposure to amaternal dose of 250 mg/kg body weight TBBPA in rats (Hass andWamberg, 2002). Changes in motor activity were also found in ratsexposed to a maternal dose of 50 mg/kg or 250 mg/kg body weightTBBPA (Hass and Wamberg, 2002). Our results showed behavioralalterations 3 h after acute treatment with 0.1 mg/kg or 5 mg/kg bodyweight TBBPA in mice. Although the reason for discrepancy as tothe behavioral effects of TBBPA is unknown, it may be caused bythe differences in animal species, genetic background of animalstrains and the schedule of TBBPA treatment. Notably, we exam-ined the behavioral effects of TBBPA 3 h after oral administration,the time point at which maximal plasma concentrations of TBBPAwere observed (Schauer et al., 2006).

Previous reports suggest absorption of TBBPA from gastroin-testinal tract and rapid metabolism of the absorbed TBBPA byconjugation resulting in a low systemic bioavailability of TBBPAafter oral administration (Schauer et al., 2006; Kuester et al., 2007).In the present study, only small amounts of TBBPA were detectedin brain tissue. This result confirms the low systemic bioavailabilityof TBBPA after oral administration. Although human exposure lev-els of TBBPA are scarcely reported, limited information indicatesconcentration of <1 ng/g lipids in serum samples from the gen-eral population (Thomsen et al., 2002). According to the EU RiskAssessment Report for TBBPA, daily human intake is approximately0.19 mg/kg body weight via different routes (Munn et al., 2006).Future research should focus on exposure, body burden and fateof TBBPA in the tissues including the brain, and possible effects ofmetabolites to build up a broader database for the assessment ofrisks related to TBBPA exposure.

Finally, it should be kept in mind that TBBPA induces cell death,reactive oxygen species formation, calcium influx and an eleva-

tion of extracellular glutamate in cerebellar granule cells in vitroat concentrations comparable to those of PCB (Reistad et al., 2007).In addition, BFRs such as TBBPA and PBDE, PCBs and halogenatedpesticides are lipophilic compounds with similar neurochemicalproperties in vitro (Mariussen and Fonnum, 2003). Further, it is

logy L

rna2dbsadeigi

C

A

Ra

A

t

R

A

B

B

D

d

E

E

E

G

H

A. Nakajima et al. / Toxico

eported that neonatal exposure of mice to some BFRs produceeurobehavioral aberrations similar to those observed in PCB-nd DDT-exposed mice (Eriksson et al., 2001, 2002; Viberg et al.,003a,b). Therefore, a neurotoxic effect is not necessarily depen-ent on the level of a single BFR in the brain, but it may be causedy a mixture of organic halides. An interesting study showed thatimultaneous administration of PCB52 (0.4 mg/kg body weight)nd PBDE99 (0.8 mg/kg body weight) at postnatal day 10 enhancesevelopmental neurotoxic effects in mice, indicating a synergisticffect of the two compounds (Eriksson et al., 2006). Collectively, it ismportant to investigate whether TBBPA induces additive or syner-istic neurobehavioral effects in combination with other toxicantsn the future study.

onflict of interest

None.

cknowledgment

This work was supported in part by a Grant-in-Aid for Scientificesearch from the Ministry of Education, Culture, Sports, Science,nd Technology of Japan.

ppendix A. Supplementary data

Supplementary data associated with this article can be found, inhe online version, at doi:10.1016/j.toxlet.2009.05.003.

eferences

tkins, C.M., Selcher, J.C., Petraitis, J.J., Trzaskos, J.M., Sweatt, J.D., 1998. The MAPK cas-cade is required for mammalian associative learning. Nat. Neurosci. 1, 602–609.

irnbaum, L.S., Staskal, D.F., 2004. Brominated flame retardants: cause for concern?Environ. Health Perspect. 112, 9–17.

ranchi, I., Alleva, E., Costa, L.G., 2002. Effects of perinatal exposure to a poly-brominated diphenyl ether (PBDE 99) on mouse neurobehavioural development.Neurotoxicology 23, 375–384.

arnerud, P.O., Eriksen, G.S., Jóhannesson, T., Larsen, P.B., Viluksela, M., 2001. Poly-brominated diphenyl ethers: occurrence, dietary exposure, and toxicology.Environ. Health Perspect. 109 (suppl. 1), 49–68.

e Wit, C.A., 2002. An overview of brominated flame retardants in the environment.Chemosphere 46, 583–624.

riksson, P., Fischer, C., Fredriksson, A., 2006. Polybrominated diphenyl ethers,a group of brominated flame retardants, can interact with polychlorinatedbiphenyls in enhancing developmental neurobehavioral defects. Toxicol. Sci. 94,302–309.

riksson, P., Jakobsson, E., Fredriksson, A., 2001. Brominated flame retardants: a novelclass of developmental neurotoxicants in our environment? Environ. HealthPerspect. 109, 903–908.

riksson, P., Viberg, H., Fisher, M., Fredriksson, A., 2002. A comparison ondevelopmental neurotoxic effects of hexabromocyclododecan 2,2′ ,4,4′ ,5,5′-hexabromodiphenyl ether (PBDE 153) and 2,2′ ,4,4′ ,5,5′-hexachlorobiphenyl(PBC 153). Organohalogen Compd. 57, 390.

ilbert, M.E., Mack, C.M., Lasley, S.M., 1999. Chronic developmental lead exposureand hippocampal long-term potentiation: biphasic dose–response relationship.Neurotoxicology 20, 71–82.

ass, U., Wamberg, C., 2002. Developmental neurotoxicity study of the bromi-nated flame retardant tetrabromobisphenol A in rats. Reprod. Toxicol. 16, 412(Abstract).

etters 189 (2009) 78–83 83

Kuester, R.K., Sólyom, A.M., Rodriguez, V.P., Sipes, I.G., 2007. The effects of dose, route,and repeated dosing on the disposition and kinetics of tetrabromobisphenol Ain male F-344 rats. Toxicol. Sci. 96, 237–245.

LeDoux, J.E., 2000. Emotion circuits in the brain. Annu. Rev. Neurosci. 23, 155–184.

Lilienthal, H., Verwer, C.M., van der Ven, L.T., Piersma, A.H., Vos, J.G., 2008.Exposure to tetrabromobisphenol A (TBBPA) in Wistar rats: neurobehavioraleffects in offspring from a one-generation reproduction study. Toxicology 246,45–54.

Mariussen, E., Fonnum, F., 2003. The effect of brominated flame retardants on neu-rotransmitter uptake into rat brain synaptosomes and vesicles. Neurochem. Int.43, 533–542.

Maurice, T., Hiramatsu, M., Itoh, J., Kameyama, T., Hasegawa, T., Nabeshima, T., 1994.Behavioral evidence for a modulating role of sigma ligands in memory processes.I. Attenuation of dizocilpine (MK-801)-induced amnesia. Brain Res. 647, 44–56.

Maurissen, J.P., Mattsson, J.L., 1989. Critical assessment of motor activity as a screenfor neurotoxicity. Toxicol. Ind. Health 5, 195–202.

Munn, S.J., Allanou, R., Aschberger, K., Cosgrove, O., Pakalin, S., Paya-Perez, A.,Pellegrini, G., Schwarz-Schulz, B., Vegro, S., 2006. European Union RiskAssessment Report: 2,2′ ,6,6′-Tetrabromo-4,4′-isopropylidenediphenol(tetrabromobisphenol-A or TBBP-A), Vol. 63, Part II, 4th priority list.Human Health Office for Official Publications of the European Communities,Luxembourg.

Nagase, H., Omae, N., Omori, A., Nakagawasai, O., Tadano, T., Yokosuka, A., Sashida,Y., Mimaki, Y., Yamakuni, T., Ohizumi, Y., 2005. Nobiletin and its relatedflavonoids with CRE-mediated transcription-stimulating and neuritegenic activ-ities. Biochem. Biophys. Res. Commun. 337, 1330–1336.

Nakajima, A., Yamakuni, T., Matsuzaki, K., Nakata, N., Onozuka, H., Yokosuka,A., Sashida, Y., Mimaki, Y., Ohizumi, Y., 2007. Nobiletin, a citrus flavonoid,reverses learning impairment associated with N-methyl-d-aspartate receptorantagonism by activation of extracellular signal-regulated kinase signaling. J.Pharmacol. Exp. Ther. 321, 784–790.

Onozuka, H., Nakajima, A., Matsuzaki, K., Shin, R.W., Ogino, K., Saigusa, D., Tetsu, N.,Yokosuka, A., Sashida, Y., Mimaki, Y., Yamakuni, T., Ohizumi, Y., 2008. Nobiletin,a citrus flavonoid, improves memory impairment and Abeta pathology in atransgenic mouse model of Alzheimer’s disease. J. Pharmacol. Exp. Ther. 326,739–744.

Osako, M., Kim, Y.J., Sakai, S., 2004. Leaching of brominated flame retardants inleachate from landfills in Japan. Chemosphere 57, 1571–1579.

Phillips, R.G., LeDoux, J.E., 1992. Differential contribution of amygdala and hip-pocampus to cued and contextual fear conditioning. Behav. Neurosci. 106, 274–285.

Reistad, T., Mariussen, E., Ring, A., Fonnum, F., 2007. In vitro toxicity oftetrabromobisphenol-A on cerebellar granule cells: cell death, free radical for-mation, calcium influx and extracellular glutamate. Toxicol. Sci. 96, 268–278.

Sarter, M., Bodewitz, G., Stephens, D.N., 1988. Attenuation of scopolamine-inducedimpairment of spontaneous alteration behaviour by antagonist but not inverseagonist and agonist beta-carbolines. Psychopharmacology (Berl.) 94, 491–495.

Schauer, U.M., Völkel, W., Dekant, W., 2006. Toxicokinetics of tetrabromobisphenola in humans and rats after oral administration. Toxicol. Sci. 91, 49–58.

Sellström, U., Jansson, B., 1995. Analysis of tetrabromobisphenol A in a product andenvironmental samples. Chemosphere 31, 3085–3092.

Sjödin, A., Patterson Jr., D.G., Bergman, A., 2003. A review on human exposureto brominated flame retardants—particularly polybrominated diphenyl ethers.Environ. Int. 29, 829–839.

Thomsen, C., Lundanes, E., Becher, G., 2002. Brominated flame retardants in archivedserum samples from Norway: a study on temporal trends and the role of age.Environ. Sci. Technol. 36, 1414–1418.

Thomsen, C., Lundanes, E., Becher, G., 2001. Brominated flame retardants in plasmasamples from three different occupational groups in Norway. J. Environ. Monit.3, 366–370.

Viberg, H., Fredriksson, A., Eriksson, P., 2003a. Neonatal exposure to polybrominateddiphenyl ether (PBDE 153) disrupts spontaneous behaviour, impairs learning

and memory, and decreases hippocampal cholinergic receptors in adult mice.Toxicol. Appl. Pharmacol. 192, 95–106.

Viberg, H., Fredriksson, A., Jakobsson, E., Orn, U., Eriksson, P., 2003b. Neurobehavioralderangements in adult mice receiving decabrominated diphenyl ether (PBDE209) during a defined period of neonatal brain development. Toxicol. Sci. 76,112–120.