Neuroendocrine and Behavioral Effects of Maternal Exposure to Oral Bisphenol a in Female Mice

L NAULE´ and others Gestational and lactational BPA 220:3 375–388 Research in female mice

Neuroendocrine and behavioral effects of maternal exposure to oral bisphenol A in female mice

Lydie Naule´1,2,3, Marie Picot1,2,3, Mariangela Martini4,5,6, Caroline Parmentier1,2,3, He´le`ne Hardin-Pouzet1,2,3, Matthieu Keller4,5,6, Isabelle Franceschini4,5,6 and Sakina Mhaouty-Kodja1,2,3

1Sorbonne Universite´ s, UPMC University Paris 06, UMR 7224, 2Institut National de la Sante´ et de la Recherche Me´ dicale (INSERM) UMR_S 952 and 3Centre National de la Recherche Scientiﬁque (CNRS) UMR 7224, Physiopathologie des Maladies du Syste` me Nerveux Central (PMSNC), Universite´ Pierre et Marie Curie, Correspondence 9 Quai St Bernard Baˆ tB2e` me E´ tage, F75005 Paris, France should be addressed 4Institut National de la Recherche Agronomique (INRA) UMR85, Physiologie de la Reproduction et des to S Mhaouty-Kodja Comportements, F-37380 Nouzilly, France Email 5CNRS UMR 7247, F-37380 Nouzilly, France sakina.mhaouty-kodja@ 6Universite´ Franc¸ois Rabelais, F-37000 Tours, France snv.jussieu.fr

Abstract

Bisphenol A (BPA) is a widespread estrogenic compound. We investigated the effects of Key Words maternal exposure to BPA at reference doses on sexual behavior and neuroendocrine " bisphenol A functions offemale offspring in C57BL/6J mice. The dams were orally exposed to vehicle alone " female mice or vehicle-containing BPA at doses equivalent to the no observed adverse effect level (5 mg/kg " nervous system " Journal of Endocrinology body weight per day) and tolerable daily intake (TDI, 0.05 mg/kg body weight per day) level reproduction from gestational day 15 until weaning. Developmental exposure to BPA increased the lordosis " behaviors quotient in naive females exposed to BPA at the TDI dose only. BPA exposure had no effect on " kisspeptin olfactory preference, ability to express masculine behaviors or number of calbindin-positive cells, a sexually dimorphic population of the preoptic area. BPA at both doses selectively increased kisspeptin cell number in the preoptic periventricular nucleus of the rostral periventricular area of the third ventricle in adult females. It did not affect the number of GNRH-positive cells or percentage of kisspeptin appositions on GNRH neurons in the preoptic

area. These changes were associated with higher levels of estradiol (E2) at the TDI dose while levels of LH, estrus cyclicity, ovarian and uterine weights, and fertility remained unaffected. Delay in the time of vaginal opening was observed during the postnatal period at TDI dose, without any alteration in body growth. This shows that developmental exposure to BPA at reference doses did not masculinize and defeminize the neural circuitry underlying sexual behavior in female mice. The TDI dose speciﬁcally exacerbated responses normally induced by

ovarian E2, through estrogen receptor a, during the postnatal/prepubertal period. Journal of Endocrinology (2014) 220, 375–388

Introduction

Bisphenol A (BPA) is one of the most abundant endocrine beverages. Environmental and health concerns about disrupters. It is used in epoxy resins and a wide variety of the extensive use of polycarbonate plastic have been raised, polycarbonate containers for the storage of foods and in particular, concerning BPA leaching from plastic.

http://joe.endocrinology-journals.org Ñ 2014 Society for Endocrinology Published by Bioscientiﬁca Ltd. DOI: 10.1530/JOE-13-0607 Printed in Great Britain Downloaded from Bioscientifica.com at 09/27/2021 09:45:02PM via free access Research L NAULE´ and others Gestational and lactational BPA 220:3 376 in female mice

Most individuals in industrialized nations are exposed to sexual behavior in female mice. The studies investigating BPA contamination (Calafat et al. 2005, Vandenberg et al. the effects of BPA on reproductive physiology report 2007). BPA remains widely used in industry despite some various effects on vaginal opening, fertility, and the estrus countries recently banning the use of polycarbonate in cycle. Exposure to BPA during prenatal or postnatal life the production of baby bottles. Known for its estrogenic resulted in advanced- (Honma et al. 2002, Nikaido et al. activity, BPA binds to estrogen receptors (ERs) with a lower 2004, Nah et al. 2011), delayed- (Nikaido et al. 2005), or

afﬁnity than that of estradiol (E2)(Dodds & Lawson 1936) unaffected- (Howdeshell et al. 1999, Tyl et al. 2008) vaginal and acts as a partial agonist (Delfosse et al. 2012). opening. Fertility was either reduced or unaltered in The sexual differentiation of reproductive behavior females exposed to BPA (Honma et al. 2002, Tyl et al. and neuroendocrine function in rodents occurs during 2008, Cabaton et al. 2011). Irregular (Honma et al. 2002, the perinatal (late gestational and early neonatal) and Nikaido et al. 2004) or normal estrus cycles (Tyl et al. 2008) postnatal periods. This process is tightly regulated by have been described for BPA-exposed females. This gonadal hormones and is therefore highly sensitive to diversity in effects could be related to factors including hormonal changes or exposure to exogenous factors differences in mouse strains, BPA dose, and time of exhibiting hormone-mimetic activities. exposure. At the neuroanatomical level, only one study In males, testosterone liberated from fetal and reported increased levels of Kiss1 and Gnrh mRNAs after neonatal testes permanently potentiates male (masculini- exposure to high doses of oral BPA (Xi et al. 2011). These zation) and inhibits female (defeminization) behavioral genes code for kisspeptin and GNRH, two important and neuroanatomical characteristics (Phoenix et al. 1959, neuropeptides for the central regulation of reproductive Morris et al. 2004). The organizational effects of testoster- function throughout life and known targets of a variety of one result in the adult expression of male-typical behaviors estrogeno-mimetic endocrine disruptors (Gore 2010, (such as preference for a receptive female, mounts and Tena-Sempere 2010, Franceschini & Desroziers 2013). thrusts) and an inability to adopt the female posture Therefore, this study aimed to assess the effects of BPA (lordosis), which is triggered by male mounts. Testosterone exposure on sexual differentiation of copulatory behavior also acts perinatally to suppress the neuroanatomical and reproductive physiology, as well as to investigate the characteristics required for sex steroids to induce the neuronal populations involved in the regulation of these ovulatory surge of luteinizing hormone (LH) during responses in female mice. Indeed, documenting oral BPA adulthood (Corbier 1985, Homma et al. 2009). Sex steroids exposure in mice is of great interest because transgenic Journal of Endocrinology regulate gonadotropin release through an inhibitory or mice could be used to unravel the molecular mechanisms stimulatory feedback effect. Both male and female rodents underlying BPA effects. Female C57BL/6J strain mice, respond to negative feedback, but only females respond to widely used in reproductive and transgenic studies, were positive feedback. Testosterone effects are mainly induced thus studied following maternal exposure during the

by E2, which is derived from neural aromatization of critical developmental periods for brain sexual differen- testosterone by aromatase cytochrome P450. tiation. BPA was orally administered to dams at the two In females, the ovaries are initially inactive, and selective dose levels set by the European Food Safety Agency and US binding of a-fetoprotein to maternal- and male sibling- Environmental Protection Agency as protective measures:

derived E2 protects the perinatal brain from its masculinizing the ‘no observed adverse effect level’ (NOAEL) and the ‘100-

effects (Bakker et al.2006). Signiﬁcant levels of E2 are liberated fold lower tolerable daily intake’ (TDI) dose. This mimicked from the ovaries around postnatal day 7. Ovarian liberation the major route of contamination for animals and humans.

of E2 between postnatal days 15 and 25 is important for a full The levels of environmental exposure to BPA could expression of female sexual behavior during adulthood actually be similar to the TDI dose (Vom Saal et al. 2007). (Bakker et al. 2002, Brock et al.2011), suggesting that this prepubertal period is also an important time window for the differentiation of female reproductive behaviors. Materials and methods BPA does not bind to a-fetoprotein (Milligan et al. Animals 1998) and could therefore act in the female brain to masculinize reproductive behaviors and functions during C57BL/6J mice, purchased from Janvier (Le Genest,

the perinatal period. It could also interfere with E2 effects France), were group housed under a controlled photo- during the prepubertal period. No studies, to our knowl- period (12 h light:12 h darkness cycle – lights on at edge, have explored the effects of BPA exposure on the 0700 h), maintained at 22 8C, and fed with a standard

diet with free access to food and water. Estrus females were Body length (nose to base of tail) and anogenital caged with males overnight, and successful mating was distance were measured. The ovaries and uterine horns of determined by the presence of a vaginal plug (day 1 of each animal were removed and weighed. Ovaries were fixed pregnancy). Animals were handled in accordance with the in Bouin’s buffer overnight at 4 8C, washed with ethanol, European guidelines for use of experimental animals paraffin-embedded, and sliced into 5 mm sections to carry (Decree 87-848, 86/609/ECC). Experiments were carried out histological analyses. These sections were mounted and out accordingly, to minimize animal number and dis- deparaffinized, stained with hematoxylin–eosin, covered comfort and were approved by the local Department of with a coverslip, and qualitatively examined. Animal Protection and Health. To evaluate the capacity of females to produce offspring, females from the three treatment groups were individually housed with age-matched males for 4 months Treatments and examined regularly for the presence of litters. Litter Studied females, exposed to BPA in utero and during the size and number of litters were recorded. lactational period, were obtained from a second pregnancy due to the low survival rates of pups from Behavioral analyses primiparous mothers, who have poor maternal behavior. Survival of pups from multiparous females was higher Tests were conducted under red light illumination for 2 h than that of those from primiparous dams (87 vs 32%; after lights off and videotaped for analysis. nZ38). Second-pregnancy dams were administered, by oral gavage, sesame oil (vehicle group) or sesame oil Lordosis Females were ovariectomized under general containing BPA (Sigma–Aldrich) at 5 mg/kg body weight anesthesia and received Silastic implants containing 50 mg per day (BPA–NOAEL group) or BPA at 0.050 mg/kg body of estradiol benzoate (Sigma–Aldrich) in 30 ml sesame oil weight per day (BPA–TDI group). Treatments were carried (Raskin et al. 2009, 2012). Four to five hours before the test, out each day from gestational day 15 until postnatal day females were given a s.c. injection of 1 mg progesterone 21 (Picot et al. 2014). After weaning, female offspring were (Sigma–Aldrich) in 100 ml sesame oil. The animals were allowed to grow until analyses. Analyses were carried out tested twice, with a time interval of 1 week between tests. on adult females (from 8 weeks of age), with the exception Sexually experienced males were used as partners. Tests of body weight measurements and vaginal opening ended when females had received 20 mounts from males. Journal of Endocrinology evaluation, which started earlier. The subjects displayed a lordosis posture when the four paws were grounded: the hind region lifted and the back region arched. We calculated the lordosis quotient Reproductive physiology (lordosis number/number of mounts) for each subject in Females from the three treatment groups were weighed response to male mounting. weekly from postnatal day 15 until 90 days of age. They were examined daily for evidence of vaginal opening from Olfactory preference Mice were placed into an postnatal day 25. enclosed Plexiglas Y-maze without any stimuli, for 5 min Vaginal smears flushed with physiological saline were on two consecutive days, to allow them to adapt to the taken daily from adult females for 3 weeks and colored apparatus. Females were tested for mate preference on the with hematoxylin–eosin to identify the estrus cycle phase third day by placing an anesthetized receptive female and under a microscope. gonadally intact male in boxes, one at the distal end of Females were killed at the diestrus stage for measure- each arm. The time spent sniffing each goal box was scored

ment of E2 and LH and morphometric analysis of the over the 5-minute test. Results are expressed as a genital tract, in order to avoid variability in these percentage of total time spent snifﬁng male or female

parameters due to the increasing levels of E2 in the cues. The maze was cleaned with 10% ethanol between proestrus/estrus phase. Blood samples were collected and trials (Keller et al. 2006). the sera were extracted as described previously (Raskin et al.

2009). The circulating levels of E2 and LH were measured by Mounting and thrusting Females were ovari- RIA (ultra-sensitive estradiol kit; Beckman Coulter, Ville- ectomized under general anesthesia and received Silastic pinte, France) and immunoassay (ultrasensitive Elisa kit; implants ﬁlled with 10 mg testosterone. Tests were Endocrine Technologies, Newark, CA, USA) respectively. conducted 2 weeks after testosterone implantation.

Females of each treatment group were individually placed kisspeptin immunoreactive cell bodies on anatomically in a cage with fresh bedding and allowed to adapt for 2 h. matched sections identiﬁed using the Mouse Brain Atlas of

A receptive female, ovariectomized and primed with E2 Paxinos & Franklin (2001) as a guide to define the and progesterone, was then introduced in the cage. We anteroventral periventricular nucleus (plates 28–29; two recorded the number of mounts and pelvic thrusting sections per animal) and the preoptic periventricular movements of testosterone-implanted females during nucleus, divided into the rostral region (plate 30; four 30 min (Martel & Baum 2009). sections per animal) and caudal region (plates 31–32, four sections per animal) (Clarkson & Herbison 2006). The quantification of GNRH immunoreactive cell bodies was Immunohistochemistry carried out on four anatomically matched section selected at Females were killed at the diestrus stage and transcardial the level of the rostral preoptic area. In addition, each GNRH perfusion was performed with a solution of 4% paraformal- neuron was evaluated for close apposition to a kisspeptin dehyde (PFA) in phosphate buffer (PB). Brains were fiber using a confocal microscope (Zeiss, LSM 700). The postfixed overnight in 4% PFA–PB, cryoprotected in kisspeptin fiber was considered to be in close apposition if it sucrose, and stored until analyses. Brains were sliced into was directly adjacent to the GNRH neuron cell body and/or coronal sections of 30 mm using a cryotome (Leica CM 300). the proximal dendrite was in the same plane of focus. The sections were blocked for 2 h with 5% normal goat Anatomically matched sections were also selected at the serum (Sigma–Aldrich) in PBS–Triton 0.1%, then incubated level of the middle arcuate nucleus for the quantification of with anti-ERa antibody (1:400; Santa Cruz Biotechnology) kisspeptin fiber density (plate 45; one section per animal) by for 48 h, monoclonal anti-calbindin antibody (1:1000; voxel counts on a set of ten serial image planes (z step Sigma–Aldrich) for 24 h, or rabbit anti-GNRH (1:10 000; distanceZ1 mm) as described by Losa et al.(2011). Caldani et al.1988) and sheep anti-kisspeptin antibody To assess calbindin and ERa immunoreactivity, sections (1:2000; Franceschini et al.2013) for 72 h. were scanned using a high-resolution Nanozoomer Hama- ERa and calbindin immunostaining was carried out matsu scanner (Hamamatsu Corporation). The Mouse Brain over a 2 h period using biotinylated goat anti-rabbit Atlas was also used to define the sexually dimorphic cluster of (1:500) or anti-mouse (1:250) antibodies (Vector Labora- calbindin immunoreactive neurons located in the medial tories, Les Ulis, France) respectively. Bound antibodies preoptic area (plate 31; one section per animal) and ERa were visualized by 1 h incubation with the streptavidin immunoreactive neurons in the medial preoptic area (plate Journal of Endocrinology complex reagent (Vector Laboratories), followed by color 29), the ventromedial and arcuate nuclei (plate 44; one development with the chromogenic substrate 3,3-diami- section per animal). They were compared across animal nobenzidine tetrahydrochloride (Sigma–Aldrich). groups by tallying the number of labeled cells in each region. Kisspeptin and GNRH dual-label immunofluorescence was carried out using an Alexa Fluor 488-conjugated Statistical analyses donkey anti-sheep secondary antibody (1:500; Life Tech- nologies–Invitrogen) and a CY3 donkey anti-rabbit second- Data were expressed as meansGS.E.M. Two-way ANOVA ary antibody (1:500; Jackson Immunoresearch, Montlucon, was used to analyze the main effects of treatment and France) for kisspeptin and GNRH immunostaining, respect- experience, with Tukey tests for post-hoc comparisons. ively, for 2.5 h at room temperature. After several rinses in One-way ANOVA or Student’s t-tests were used for PBS, the sections were immersed in a Hoechst stain bath unpaired data. A c2-test was used to analyze the (2 mg/ml in water; Invitrogen) for 2 min, rinsed in water percentage of females displaying the different behaviors (two baths, 5 min each), dried, mounted in Fluoromount-G studied. P values !0.05 were considered to be significant. (Southern Biotechnology, Birmingham, AL, USA) under a coverslip, and stored at 4 8C in the dark. Results

Analysis and quantiﬁcation of immunohistochemical Developmental exposure to BPA at the TDI dose delayed studies the time of vaginal opening and increased levels of circulating E Kisspeptin cell numbers within each of the three sub- 2 divisions of the rostral periventricular area of the third The average age of vaginal opening was signiﬁcantly later in ventricle (RP3V) were analyzed by counting the number of the female group exposed to BPA at the TDI dose (BPA–TDI)

A 25

15 Vehicle

Body weight (g) Body weight BPA–TDI 10 BPA–NOAEL

5 15 22 29 36 43 50 5764 71 78 84 90 Age (days)

B C Vehicle 5 2 BPA–TDI BPA–NOAEL 4

3 1 2 Average length of Average

estrus cycle (days) 1 Length of stage (days) 0 0 Vehicle BPA– BPA– Proestrus Estrus Diestrus I Diestrus II TDI NOAEL

# §

Journal of Endocrinology § #

§ #

Vehicle BPA–TDI BPA–NOAEL

Figure 1 Effect of developmental exposure to BPA on body weight, estrus cycle at TDI or NOAEL dose are expressed as meansGS.E.M.(nZ19–24 females per length, and ovarian histology. (A) The growth curve of female mice treatment group). (D) Representative histology of the ovary in females exposed to the vehicle or BPA at TDI and NOAEL doses was measured from exposed to vehicle or BPA at TDI or NOAEL dose. Primary follicles (black postnatal day 15 until 90 days of age. Data were expressed as meansGS.E.M. arrows), atretic follicles (#), and corpora lutea (§) were observed in each (nZ27–38 females per treatment group). The average cycle length (B) and treatment group. Scale barZ100 mm. average length of each estrus stage (C) in females exposed to vehicle or BPA

than in females exposed to vehicle or BPA at the NOAEL distance (5.61G0.10, 5.81G0.10, and 5.76G0.12% of dose (BPA–NOAEL) (31.1G0.4 vs 29.4G0.3 for vehicle and body length for vehicle, BPA–TDI, and BPA–NOAEL

29.6G0.3 for BPA–NOAEL; F(2, 140)Z7.48, P!0.05). The groups respectively) were also similar for the three growth curves for body weight between 15 and 90 days treatment groups. of age were, however, similar between vehicle and Adult females had a similar average length of estrus BPA-exposed groups (Fig. 1A). Body length (85.43G0.42, cycle (Fig. 1B) and total length of each estrus cycle stage 85.30G0.44, and 84.60G0.69 mm for vehicle, BPA–TDI, (Fig. 1C), regardless of BPA dose. Ovarian and uterine and BPA–NOAEL groups respectively) and anogenital weights at the diestrus stage were also unchanged by BPA

Table 1 Effects of developmental exposure to BPA on ovary and uterine weights, expressed as percentage of body weight (BW), as

well as circulating levels of estradiol and LH in adult females at the diestrus stage. Values are given as meansGS.E.M.(nZ10–21 females per treatment group)

Treatments Ovaries weight (% BW) Uterine weight (% BW) Estradiol levels (pg/ml) LH levels (ng/ml)

Vehicle 0.0147G0.0005 0.104G0.0042 14.83G2.054 0.78G0.141 BPA–TDI 0.0144G0.0004 0.100G0.0040 24.09G3.082* 1.37G0.304 BPA–NOAEL 0.0127G0.0007 0.107G0.0076 17.00G2.425 1.09G0.302

*P!0.05 compared with the vehicle group.

exposure (Table 1). Histological analysis of ovary sections to BPA–TDI exhibited higher sexual behavior than both showed the presence of follicles, at different stages of vehicle- and NOAEL–BPA-exposed females, but only when folliculogenesis, and corpora lutea, indicating that they were at their ﬁrst sexual experience. ovulation occurred for all treatment groups (Fig. 1D). In olfactory preference tests using intact males and

The levels of circulating E2 were significantly increased in receptive females, two-way ANOVA showed a significant Z Z the BPA–TDI group by comparison with females exposed effect of stimulus (F(1, 29) 9.33, P 0.005) with females to the vehicle (Table 1). Measurement of circulating levels displaying a preference for intact males over receptive of LH showed no significant differences in the means females in all three groups (Fig. 2C). The total time between the three treatment groups (Table 1), although devoted to chemoinvestigation did not differ significantly slightly increased levels (2–2.86 ng/ml) were found in between groups (around 100 s). four of the ten females exposed developmentally to the BPA–TDI dose. Developmental exposure to BPA did not masculinize the Analysis of female fertility showed that females female brain exposed to vehicle or BPA during development were able to produce offspring as a result of mating tests lasting We investigated if exposure to BPA, particularly during the for 4 months. There was no significant difference in the perinatal period, masculinized the behavior and neuro- number of litters (3.67G0.33, 3.67G0.67, and 3G0.00 anatomy of female mice. for vehicle, BPA–TDI, and BPA–NOAEL groups respec- As testosterone is also needed during adulthood to Journal of Endocrinology tively) and litter size (7.19G0.96, 8.20G0.32, and 6.89G activate male sexual behavior, females exposed to BPA or 0.44 for vehicle, BPA–TDI, and BPA–NOAEL groups vehicle during development were ovariectomized during respectively). adulthood and received implants that released testosterone before testing their ability to exhibit a masculine behavior. Similar percentages of females from each group Developmental exposure to BPA–TDI increased lordosis displayed mounting and thrusting behaviors toward quotient in naive females receptive females (Fig. 3A). The numbers of mounts and In rodents, mating is induced by olfactory cues. We thrusts were also similar between the three treatment therefore analyzed whether developmental exposure to groups, indicating no increase in the number of events in BPA affected female sexual behavior and olfactory females exposed to BPA (Fig. 3B and C). preference. We analyzed a cluster of calbindin-D28 immuno- Lordosis behavior was measured in adult females, reactive cells located in the medial preoptic area and

which were ovariectomized and primed with E2 and corresponding to the rat sexually dimorphic nucleus

progesterone, in response to the mounting of a stud male. known to be involved in male sexual behavior. E2, derived Lordosis behavior was displayed by similar percentages of from the neural aromatization of testosterone, acts naive (Test 1) and experienced females (Test 2) who had perinatally to masculinize this neuronal population been exposed during development to vehicle or BPA at resulting in more cells in males than in females NOAEL or TDI dose (Fig. 2A). Two-way ANOVA showed no (Gilmore et al. 2012). Developmental exposure to BPA

effect of exposure (F(2, 41)Z1.39, PZ0.26), but a signiﬁcant had no effect on this cluster in the medial preoptic area as

effect of experience (F(1, 41)Z34.86, P!0.0001) (Fig. 2B). no difference in the number of calbindin immunoreactive There was an interaction between exposure and experience cells was found between vehicle and BPA-exposed females

(F(2, 41)Z3.33, PZ0.046), indicating that females exposed (Fig. 4A and B).

Vehicle BPA–TDI BPA–NOAEL targets of ER signaling and major gatekeepers of pubertal onset and adult female reproduction (Franceschini & A 100 Desroziers 2013). Kisspeptin neurons are localized in two 80 hypothalamic regions: i) the RP3V including the anteroventral periventricular and the preoptic periventricular 60 nuclei and ii) the arcuate nucleus. The kisspeptin

20 A 100 Percentage of females Percentage showing lordosis behavior showing 0 80 Test 1 Test 2 B *** 60 80

40 60

Percentage of animals Percentage 20 40 male behaviors showing 0 20 Vehicle BPA– BPA– TDI NOAEL Lordosis quotient (%)

0 B 30 Test 1 Test 2 C ## 60 ## ## 20 50

40 10 30 Number of mounts

Journal of Endocrinology 20 Time spent in 10 0 chemoinvesigation (%) chemoinvesigation Vehicle BPA– BPA– 0 TDI NOAEL

Vehicle BPA– BPA– C 500 TDI NOAEL 400

Figure 2 Sexual behavior and olfactory preference in adult female mice exposed to 300 BPA during development. (A) Percentage of adult ovariectomized and hormone-primed females exhibiting lordosis behavior in response to 200 mounting of stud males in Tests 1 and 2. (B) Lordosis quotient measured GS E M for each female in Tests 1 and 2. Data were expressed as means . . . Number of thrusts (nZ18–21 females per treatment group). ***P!0.001 compared with 100 Test 1. (C) Time spent chemoinvestigating male vs female stimuli expressed as percentage of total time spent in chemoinvestigation. ##P!0.01 0 compared with male stimulus in the same treatment group. Vehicle BPA– BPA– TDI NOAEL Selective increase in kisspeptin immunoreactive cells in the rostral preoptic periventricular nucleus by BPA Figure 3 exposure Mounting and thrusting behaviors displayed by adult female mice exposed to vehicle or BPA during development. (A) Percentage of adult We examined whether exposure to BPA had an effect on ovariectomized females with testosterone-releasing implants exhibiting masculine behavior toward receptive females. Number of mounts (B) and the kisspeptin/GNRH system. GNRH neurons receive thrusts (C) displayed toward receptive females. Data are expressed as projections from kisspeptin neurons, which are direct meansGS.E.M.(nZ20–22 females per treatment group).

ABVehicle BPA–TDI BPA–NOAEL 180

120

60 Number of calbindin immunoreactive cells immunoreactive 0 Vehicle BPA– BPA– TDI NOAEL

Figure 4 Characterization of a cluster of calbindin immunoreactive cells in the or NOAEL dose. Scale barZ100 mm. (B) Quantiﬁcation of the number of medial preoptic area. (A) Representative immunostaining of a sexually calbindin immunoreactive cells. Data are expressed as meansGS.E.M. dimorphic calbindin neuronal population in adult female mice exposed (nZ4 females per treatment group). during the gestational and lactational periods to vehicle or BPA at the TDI

population located in the RP3V contains more cells in periventricular and the caudal preoptic periventricular females than in males (Clarkson & Herbison 2006, 2009). nuclei, we found no significant difference in the number The number of kisspeptin immunoreactive cells in the of kisspeptin-immunoreactive cells between treatment rostral preoptic periventricular nucleus was significantly groups (Fig. 5B). In the arcuate nucleus, there was no higher in females exposed to BPA at TDI or NOAEL dose difference in the density of kisspeptin fibers between the

(F(2, 15)Z5.56, P!0.05; Fig. 5A and B). In the anteroventral three treatment groups (Fig. 6A).

A Vehicle BPA–TDI BPA–NOAEL Journal of Endocrinology

B 100 ** Vehicle ** BPA–TDI 80 BPA–NOAEL

20 immunoreactive cells immunoreactive Number of kisspeptin

0 AVPV rPen cPen

Figure 5 Characterization of kisspeptin system in adult female mice exposed to BPA anteroventral periventricular area (AVPV), rPen, and caudal periventricular during development. (A) Representative kisspeptin immunoreactivity in nucleus (cPen). Data are expressed as meansGS.E.M.(nZ6 females per the rostral periventricular nucleus (rPen) in females exposed to vehicle or treatment group). **P!0.01 compared with the vehicle group of the BPA at the TDI or NOAEL dose. Scale barZ100 mm. (B) Quantiﬁcation of corresponding hypothalamic area. the number of kisspeptin immunoreactive cells per slice in the

A the forebrain or in kisspeptin neurons induced female 5000 infertility and suppressed lordosis behavior (Ogawa 4000 et al. 1998, Wintermantel et al. 2006, Mayer et al. 2010). 3000 We investigated whether increased lordosis, kisspeptin

per slice 2000 immunoreactivity in the RP3V, and E2 levels were 1000 associated with changes in the hypothalamic expression Average kisspeptin Average

immunoreactive voxels immunoreactive a 0 of ER in BPA-exposed females. Vehicle BPA– BPA– We quantiﬁed the number of Era immunoreactive TDI NOAEL B cells in the three main regions involved in the regulation 80 of the hypothalamus–pituitary–gonad (HPG) axis and 60 expression of female sexual behavior (the ventromedial

40 hypothalamus nucleus, medial preoptic area, and arcuate nucleus). The three treatment groups showed no signi- 20

Number of GNRH ﬁcant differences in ERa immunoreactivity (Fig. 7A and B). immunoreactive cells immunoreactive 0 Vehicle BPA– BPA– C TDI NOAEL Discussion 100

80 We investigated whether developmental exposure to BPA altered the sexual differentiation of mating behavior and 60 brain circuits in C57BL/6J female mice. Analysis of sexual 40 behavior showed that the lordosis behavior was not 20 abolished and was even signiﬁcantly higher at the TDI

with kisspeptin apposition 0

Percentage of GNRH neurons Percentage Vehicle BPA– BPA– dose in naive females. Administration of testosterone TDI NOAEL to gonadectomized adult females did not trigger more mounting or thrusting behaviors in BPA-exposed females Figure 6 than in the vehicle group. Similarly, analysis of olfactory Characterization of kisspeptin and GNRH systems in adult female mice preference, another sexually dimorphic behavior, did not exposed to BPA during development. (A) Representative kisspeptin reveal any masculinizing or defeminizing effects of BPA Journal of Endocrinology immunostaining in the middle arcuate nucleus of vehicle-treated females (left) and quantitative density for the vehicle, BPA–TDI, and BPA–NOAEL in female mice because BPA-exposed females exhibited groups (right). Scale barZ50 mm. (B) Representative immunostaining of normal interest in chemoinvestigating males rather than GNRH cells (red labeling) in the rostral preoptic area of vehicle-treated female mice (left) and quantification of the average cell number per slice in females. These results contrast with known perinatal the vehicle, BPA–TDI, and BPA–NOAEL groups (right). The white arrow effects of E2. Administration of E2 to female neonates points to a GNRH body cell. Scale barZ100 mm. (C) Representative GNRH abolishes receptivity and elicits the male pattern of sexual neuron (red labeling) with kisspeptin appositions (green labeling) in vehicle-treated females (left) and quantification of the percentage of behavior in adulthood (Levine & Mullins 1964). Similarly, appositions in the vehicle, BPA–TDI, and BPA–NOAEL groups (right). The females lacking a-fetoprotein, and thus exposed to the cell nuclei are revealed by Hoechst blue staining. Scale barZ15 mm. Values masculinizing effects of maternal estrogens, exhibit low expressed as meanGS.E.M.(nZ6 females per treatment group). lordosis behavior and increased mounting behavior during adulthood (Bakker et al. 2006). The number of GNRH-positive cells within the The behavioral observations are supported by the preoptic area showed no significant differences between results of the neuroanatomical analyses. We observed no the three treatment groups (Fig. 6B). Furthermore, the increase in the number of calbindin immunoreactive percentage of kisspeptin appositions on GNRH neurons in neurons, a sexually dimorphic population highly sensitive

the medial preoptic area was not inﬂuenced by dose type to the masculinizing effects of perinatal E2. The number (Fig. 6C). of kisspeptin cells in the RP3V, a neuronal population

normally reduced by perinatal E2 (Kauffman et al. 2007, Gonza´lez-Martı´nez et al. 2008), was even higher in BPA- BPA exposure did not affect the number of ERa exposed individuals. These data indicate that during the immunoreactive cells perinatal period, when endogenous ovarian estrogens are Neural ERa plays a key role in female reproduction undetectable and exogenous estrogens have limited brain because ubiquitous or conditional inactivation of ERa in access (due to their binding to a-fetoprotein), BPA at

A Vehicle BPA–TDI BPA–NOAEL B

1600 α 1200

800 MPOA 400 Number of ER

immunoreactive cells immunoreactive 0 Vehicle BPA– BPA– TDI NOAEL 800 α 600

400 VMH 200 Number of ER

immunoreactive cells immunoreactive 0 VehicleBPA– BPA– TDI NOAEL 1000 α 800 600

ARC 400 200 Number of ER

immunoreactive cells immunoreactive 0 VehicleBPA– BPA– TDI NOAEL

Figure 7 Quantiﬁcation of the number of ERa immunoreactive cells in hypothalamic BPA at the TDI or NOAEL dose. Scale barZ50 mm. (B) Number of ERa areas. (A) Representative ERa immunolabeling in the medial preoptic immunoreactive cells. Values are expressed as meansGS.E.M. area (MPOA), ventromedial hypothalamus (VMH), and arcuate nucleus (nZ3–5 females per treatment group). (ARC) of adult female mice exposed during development to vehicle or Journal of Endocrinology reference doses was unable to mimic the masculinizing subdivisions (anteroventral periventricular and caudal

and defeminizing effects of perinatal E2 on the reproduc- preoptic periventricular nuclei). The studies administering tive functions and behaviors in female mice. In accord- BPA at the TDI dose to neonatal rats did not detect any ance with this hypothesis, previous studies in rats reported signiﬁcant effect on kisspeptin immunoreactivity in the unchanged (Kwon et al. 2000, Adewale et al. 2009, anteroventral periventricular nucleus (Patisaul et al. 2009, Monje et al. 2009, Ryan et al. 2010, Jones et al. 2011)or Losa-Ward et al. 2012). The observed increase in kisspeptin increased (Farabollini et al. 2002) female sexual behavior immunoreactive cells could be due to BPA-induced and unaltered volume of the rat sexually dimorphic transcriptional regulation of Kiss1 because increased nucleus following early exposure to BPA at doses hypothalamic levels of Kiss mRNAs were observed equivalent or lower than the TDI dose (Takagi et al. following oral exposure to high doses (25–50 mg/kg) of 2004, He et al. 2012). However, as behaviors and brain gestational and lactational BPA in CD1 mice (Xi et al. 2011). circuits may exhibit different sensitivities to estrogen There is now accumulating evidence indicating the signaling, it is possible that our BPA exposure treatment involvement of kisspeptin neurons located in the RP3V

may have masculinized/defeminized other behaviors and nucleus in the positive feedback exerted by E2 to trigger the circuits not investigated in this study as reported for preovulatory surge of LH during proestrus, while those of

tyrosine-hydroxylase-expressing cells in the preoptic area the arcuate nucleus integrate the negative signal of E2.BPA and open-ﬁeld activity in CD1 mice (Rubin et al. 2006). treatment had no effect on kisspeptin immunoreactive ﬁber At the neuroendocrine level, more RP3V kisspeptin- density in the arcuate nucleus or on the number of GNRH immunoreactive cells were observed in adult females neurons in the preoptic area. It is therefore possible that

exposed developmentally to BPA, regardless of the BPA increased E2 levels result, at least partly, from a change in dose. A signiﬁcant increase was found in the preoptic the balance between negative- and positive-feedback periventricular nucleus and not in the two other RP3V responses at the hypothalamic portion of the HPG axis,

although a direct effect of BPA on follicular steroidogenesis opening inﬂuences the age at ﬁrst estrus and pubertal (Xi et al.2011) cannot be excluded. onset in BPA-exposed females. The observations that females developmentally BPA induced a long-term increase in lordosis behavior exposed to BPA exhibited normal estrus cyclicity, ovarian and kisspeptin immunoreactivity in the RP3V. Interestingly, and uterine weights, and were able to ovulate and produce these two parameters are known to increase around puberty

offspring indicate that BPA-induced alterations did not in a manner dependent on the postnatal rise in E2 release interfere with the normal functioning of the HPG axis. from the ovaries (Clarkson et al.2009, Brock et al. 2011). As A normal length of estrus cycle was also described in BPA exposure lasted until postnatal day 21, it is thus likely studies using ICR and CD1 mice exposed to BPA during that BPA exerted its predominant effect during the post- the postnatal period (Nikaido et al. 2005, Nah et al. 2011) natal/prepubertal period to reinforce the action of endogen-

or during both gestation and lactation (Tyl et al. 2008) ous E2, most probably on the ERa signaling pathway. The whereas BPA exposure during the prenatal period has importance of this pathway in the regulation of female been shown to increase estrus cycle length (Honma et al. reproduction and sexual differentiation of the brain and of 2002, Nikaido et al. 2004). This indicates again that the behaviors is demonstrated by the infertile phenotype and time window of BPA exposure could result in different reduced RP3V kisspeptin neuronal population and lordosis effects. It is, however, important to note that our fertility behavior in female mice lacking aromatase or neuronal studies were carried out on young adult females and ERa (Bakker et al. 2002, Wintermantel et al.2006). The deleterious effects of developmental exposure to BPA on delayed vaginal opening observed in females exposed to BPA reproductive capacity may be observable at advanced ages also favors the suggestion that BPA acts during the

(Cabaton et al. 2011). postnatal/prepubertal period as E2 can affect this process Exposure to oral BPA at the TDI dose from gestational through the postnatal action of ERa in kisspeptin cells day 15 until weaning induced delayed vaginal opening. (Mayer et al. 2010). Although we did not detect any Other studies using mouse strains with various exposure long-term effect of BPA exposure on ERa expression in the routes and BPA doses have reported advanced vaginal brain, it remainspossible thatBPAtransiently inﬂuenced ERa opening or ﬁrst vaginal estrus associated with BPA exposure expression or its signaling pathway during the postnatal/ during the prenatal period (Nikaido et al.2004)orunaltered prepubertal period, reinforcing the feminizing develop-

vaginal opening when mice were exposed to BPA from pre- mental effects of endogenous E2. breeding until weaning (Tyl et al.2008). Although advanced The majority of BPA effects described here were found Journal of Endocrinology vaginal opening is described as the most common effect of at the TDI and not the 100-fold higher NOAEL dose. Except early exposure to endocrine disrupters, it is important to for the increased number of kisspeptin cells in the RP3V, note that delayed vaginal opening was observed with which was induced by both doses, increased sexual

varying exposure periods or BPA doses. For instance, the behavior and circulating levels of E2 and delayed vaginal same dose of BPA induced delayed vaginal opening in ICR opening were all observed only at the TDI dose. This adds mice exposed between postnatal days 15 and 18 (Nikaido arguments in favor of the emerging idea in the ﬁeld of et al.2005) and early vaginal opening in CD1 females endocrine disruption that more effects of BPA are observed exposed during the prenatal period (Nikaido et al.2004). at low but relevant doses than at high doses (Welshons Bourguignon et al.(2013)described early vaginal opening et al. 2003, Jenkins et al. 2011). Such non-monotonic

after neonatal exposure to diethylstilbestrol (DES) at responses, known for endogenous hormones such as E2, 10 mg/kg per day and delayed vaginal opening at 1 mg/kg follow the dynamics of hormone receptor occupancy and

per day, indicating that precocity could be attributed to both saturation. Speciﬁc responses for E2 are observed in peripheral and central effects of DES effects while delayed response to low physiological ranges below the concen- vaginal opening could be more related to neuroendocrine tration at which the receptor-binding site is half occupied, effects. Therefore, the discrepancy in BPA effects on the age while no or only nonspeciﬁc effects not mediated by ER are at vaginal opening could be due to differences in the time of observed at high ranges. It is possible that BPA at the lower

BPA exposure and/or the dose used. reference dose exacerbated E2 effects by acting directly The delayed time of vaginal opening in C57BL/6J through ERa or indirectly through other receptors, which females was not due to delayed body growth. The vaginal in turn regulate the ERa-signaling pathway. opening might represent one of the ﬁrst processes In conclusion, gestational and lactational exposure to

triggered by increasing levels of postnatal E2. Further BPA enhanced lordosis behavior at the TDI dose and studies should investigate whether delayed vaginal augmented kisspeptin immunoreactivity in the RP3V

nucleus at both the NOAEL and TDI doses in C57BL/6J Bakker J, De Mees C, Douhard Q, Balthazart J, Gabant P, Szpirer J & female mice. BPA at the TDI dose also delayed the time of Szpirer C 2006 a-Fetoprotein protects the developing female mouse brain from masculinization and defeminization by estrogens. Nature vaginal opening and increased diestrus levels of E2. The Neuroscience 9 220–226. (doi:10.1038/nn1624) comparison with known effects of perinatal E2 indicates Bourguignon J-P, Franssen D, Ge´rard A, Janssen S, Pinson A, Naveau E & that BPA at reference doses, in the absence of endogenous Parent A-S 2013 Early neuroendocrine disruption in hypothalamus and hippocampus: developmental effects including female sexual matu- E2, did not affect sexual brain development during the ration and implications for endocrine disrupting chemical screening. perinatal period. BPA may act, in particular at the TDI Journal of Neuroendocrinology 25 1079–1087. (doi:10.1111/jne.12107) dose, during the postnatal/prepubertal period to increase Brock O, Baum MJ & Bakker J 2011 The development of female sexual behavior requires prepubertal estradiol. Journal of Neuroscience 31 a the action of ovarian E2 on the ER signaling pathway. 5574–5578. (doi:10.1523/JNEUROSCI.0209-11.2011) The role of sex steroids in sexual differentiation of the Cabaton NJ, Wadia PR, Rubin BS, Zalko D, Schaeberle CM, Askenase MH, nervous system has been well documented in many Gadbois JL, Tharp AP, Whitt GS, Sonnenschein C et al. 2011 Perinatal exposure to environmentally relevant levels of bisphenol A decreases mammalian species. In addition, BPA pharmacokinetics fertility and fecundity in CD-1 mice. Environmental Health Perspectives in women, female monkeys, and mice seem to be similar 119 547–552. (doi:10.1289/ehp.1002559) (Taylor et al. 2011), although the characterization of BPA Calafat AM, Kuklenyik Z, Reidy JA, Caudill SP, Ekong J & Needham LL 2005 bioavailability still needs to be improved in humans. It is Urinary concentrations of bisphenol A and 4-nonylphenol in a human reference population. Environmental Health Perspectives 113 391–395. therefore possible that BPA triggers responses similar to (doi:10.1289/ehp.7534) those described in this study in females of other Caldani M, Batailler M, Thie´ry JC & Dubois MP 1988 LHRH-immuno- mammalian species. reactive structures in the sheep brain. Histochemistry 89 129–139. (doi:10.1007/BF00489916) Clarkson J & Herbison AE 2006 Postnatal development of kisspeptin neurons in mouse hypothalamus; sexual dimorphism and projections Declaration of interest to gonadotropin-releasing hormone neurons. Endocrinology 147 The authors declare that there is no conﬂict of interest that could be 5817–5825. (doi:10.1210/en.2006-0787) perceived as prejudicing the impartiality of the research reported. Clarkson J & Herbison AE 2009 Oestrogen, kisspeptin, GPR54 and the preovulatory luteinising hormone surge. Journal of Neuroendocrinology 21 305–311. (doi:10.1111/j.1365-2826.2009.01835.x) Clarkson J, Boon WC, Simpson ER & Herbison AE 2009 Postnatal Funding development of an estradiol–kisspeptin positive feedback mechanism This work was supported by the ‘Contaminants-Ecosyste` mes-Sante´ ’ 2009 implicated in puberty onset. Endocrinology 150 3214–3220. program of the ‘Agence Nationale de la Recherche’, the ‘Agence Nationale (doi:10.1210/en.2008-1733) de Se´ curite´ Sanitaire’, and ‘Re´ seau Sante´ Environnement Toxicologie’ of Corbier P 1985 Sexual differentiation of positive feedback: effect of hour of Journal of Endocrinology the ‘Re´ gion Ile de France’. castration at birth on estradiol-induced luteinizing hormone secretion in immature male rats. Endocrinology 116 142–147. (doi:10.1210/ endo-116-1-142) Delfosse V, Grimaldi M, Pons J-L, Boulahtouf A, le Maire A, Cavailles V, Author contribution statement Labesse G, Bourguet W & Balaguer P 2012 Structural and mechanistic M K, I F, and S M-K designed the experiments. L N, M P, M M, and C P insights into bisphenols action provide guidelines for risk assessment performed the experiments. L N, M P, M M, H H-P, M K, I F, and S M-K and discovery of bisphenol A substitutes. PNAS 109 14930–14935. analyzed the experiments. L N and S M-K wrote the paper. (doi:10.1073/pnas.1203574109) Dodds E & Lawson W 1936 Synthetic oestrogenic agents without the phenanthrene nucleus. Nature 137 996. (doi:10.1038/137996a0) Farabollini F, Porrini S, Della Seta D, Bianchi F & Dessı`-Fulgheri F 2002 Acknowledgements Effects of perinatal exposure to bisphenol A on sociosexual behavior of The authors wish to thank Dr Alain Caraty (INRA UMR85, CNRS UMR 7247, female and male rats. Environmental Health Perspectives 110(Suppl 3) ´ Universite de Tours, Nouzilly, France) for providing anti-kisspeptin antibody 409–414. (doi:10.1289/ehp.02110s3409) and the platform of the UPMC University Paris 06 for taking care of the Franceschini I & Desroziers E 2013 Development and aging of the animals. kisspeptin–GPR54 system in the mammalian brain: what are the impacts on female reproductive function? Frontiers in Endocrinology 4 22. (doi:10.3389/fendo.2013.00022) Franceschini I, Yeo SH, Beltramo M, Desroziers E, Okamura H, Herbison AE References & Caraty A 2013 Immunohistochemical evidence for the presence of Adewale HB, Jefferson WN, Newbold RR & Patisaul HB 2009 Neonatal various kisspeptin isoforms in the mammalian brain. Journal of bisphenol-A exposure alters rat reproductive development and ovarian Neuroendocrinology 25 839–851. (doi:10.1111/jne.12069) morphology without impairing activation of gonadotropin-releasing Gilmore RF, Varnum MM & Forger NG 2012 Effects of blocking hormone neurons. Biology of Reproduction 81 690–699. (doi:10.1095/ developmental cell death on sexually dimorphic calbindin cell groups biolreprod.109.078261) in the preoptic area and bed nucleus of the stria terminalis. Biology of Sex Bakker J, Honda S-I, Harada N & Balthazart J 2002 The aromatase knock-out Differences 3 5. (doi:10.1186/2042-6410-3-5) mouse provides new evidence that estradiol is required during Gonza´lez-Martı´nez D, De Mees C, Douhard Q, Szpirer C & Bakker J 2008 development in the female for the expression of sociosexual behaviors Absence of gonadotropin-releasing hormone 1 and Kiss1 activation in in adulthood. Journal of Neuroscience 22 9104–9112. a-fetoprotein knockout mice: prenatal estrogens defeminize the

potential to show preovulatory luteinizing hormone surges. Monje L, Varayoud J, Munõz-de-Toro M, Luque EH & Ramos JG 2009 Endocrinology 149 2333–2340. (doi:10.1210/en.2007-1422) Neonatal exposure to bisphenol A alters estrogen-dependent Gore AC 2010 Neuroendocrine targets of endocrine disruptors. Hormones 9 mechanisms governing sexual behavior in the adult female rat. 16–27. Reproductive Toxicology 28 435–442. (doi:10.1016/j.reprotox.2009.06.012) He Z, Paule MG & Ferguson SA 2012 Low oral doses of bisphenol A increase Morris JA, Jordan CL & Breedlove SM 2004 Sexual differentiation of the volume of the sexually dimorphic nucleus of the preoptic area in male, vertebrate nervous system. Nature Neuroscience 7 1034–1039. but not female, rats at postnatal day 21. Neurotoxicology and Teratology (doi:10.1038/nn1325) 34 331–337. (doi:10.1016/j.ntt.2012.03.004) Nah WH, Park MJ & Gye MC 2011 Effects of early prepubertal exposure to Homma T, Sakakibara M, Yamada S, Kinoshita M, Iwata K, Tomikawa J, bisphenol A on the onset of puberty, ovarian weights, and estrous cycle Kanazawa T, Matsui H, Takatsu Y, Ohtaki T et al. 2009 Significance of in female mice. Clinical and Experimental Reproductive Medicine 38 neonatal testicular sex steroids to defeminize anteroventral periventricular 75–81. (doi:10.5653/cerm.2011.38.2.75) kisspeptin neurons and the GnRH/LH surge system in male rats. Biology of Nikaido Y, Yoshizawa K, Danbara N, Tsujita-Kyutoku M, Yuri T, Uehara N & Reproduction 81 1216–1225. (doi:10.1095/biolreprod.109.078311) Tsubura A 2004 Effects of maternal xenoestrogen exposure on Honma S, Suzuki A, Buchanan DL, Katsu Y, Watanabe H & Iguchi T 2002 development of the reproductive tract and mammary gland in Low dose effect of in utero exposure to bisphenol A and diethylstilbes- female CD-1 mouse offspring. Reproductive Toxicology 18 803–811. trol on female mouse reproduction. Reproductive Toxicology 16 117–122. (doi:10.1016/j.reprotox.2004.05.002) (doi:10.1016/S0890-6238(02)00006-0) Nikaido Y, Danbara N, Tsujita-Kyutoku M, Yuri T, Uehara N & Tsubura A Howdeshell KL, Hotchkiss AK, Thayer KA, Vandenbergh JG & vom Saal FS 2005 Effects of prepubertal exposure to xenoestrogen on development 1999 Exposure to bisphenol A advances puberty. Nature 401 763–764. of estrogen target organs in female CD-1 mice. In Vivo 19 487–494. (doi:10.1038/44517) Ogawa S, Eng V, Taylor J, Lubahn DB, Korach KS & Pfaff DW 1998 Roles of Jenkins S, Wang J, Eltoum I, Desmond R & Lamartiniere CA 2011 Chronic estrogen receptor-a gene expression in reproduction-related behaviors in oral exposure to bisphenol A results in a nonmonotonic dose response female mice. Endocrinology 139 5070–5081. (doi:10.1210/endo.139.12.6357) in mammary carcinogenesis and metastasis in MMTV–erbB2 mice. Patisaul HB, Todd KL, Mickens JA & Adewale HB 2009 Impact of neonatal Environmental Health Perspectives 119 1604–1609. (doi:10.1289/ exposure to the ERa agonist PPT, bisphenol-A or phytoestrogens on ehp.1103850) hypothalamic kisspeptin fiber density in male and female rats. Jones BA, Shimell JJ & Watson NV 2011 Pre- and postnatal bisphenol A Neurotoxicology 30 350–357. (doi:10.1016/j.neuro.2009.02.010) treatment results in persistent deficits in the sexual behavior of male Paxinos G & Franklin KBJ 2001 In The Mouse Brain in Stereotaxic Coordinates, rats, but not female rats, in adulthood. Hormones and Behavior 59 246– 2nd edn. San Diego: Academic Press. 251. (doi:10.1016/j.yhbeh.2010.12.006) Phoenix CH, Goy RW, Gerall AA & Young WC 1959 Organizing action of Kauffman AS, Gottsch ML, Roa J, Byquist AC, Crown A, Clifton DK, prenatally administered testosterone propionate on the tissues med- Hoffman GE, Steiner RA & Tena-Sempere M 2007 Sexual differentiation iating mating behavior in the female guinea pig. Endocrinology 65 of Kiss1 gene expression in the brain of the rat. Endocrinology 148 1774– 369–382. (doi:10.1210/endo-65-3-369) 1783. (doi:10.1210/en.2006-1540) Picot M, Naule´ L, Marie-Luce C, Martini M, Raskin K, Grange-Messent V, Keller M, Pierman S, Douhard Q, Baum MJ & Bakker J 2006 The Franceschini I, Keller M & Mhaouty-Kodja S 2014 Vulnerability of the vomeronasal organ is required for the expression of lordosis behaviour, neural circuitry underlying sexual behavior to chronic adult exposure but not sex discrimination in female mice. European Journal of to oral bisphenol A in male mice. Endocrinology 155 502–512. Neuroscience 23 521–530. (doi:10.1111/j.1460-9568.2005.04589.x) (doi:10.1210/en.2013-1639) Journal of Endocrinology Kwon S, Stedman DB, Elswick BA, Cattley RC & Welsch F 2000 Pubertal Raskin K, de Gendt K, Duittoz A, Liere P, Verhoeven G, Tronche F & development and reproductive functions of Crl:CD BR Sprague–Dawley Mhaouty-Kodja S 2009 Conditional inactivation of androgen receptor rats exposed to bisphenol A during prenatal and postnatal develop- gene in the nervous system: effects on male behavioral and neuro- ment. Toxicological Sciences 55 399–406. (doi:10.1093/toxsci/55.2.399) endocrine responses. Journal of Neuroscience 29 4461–4470. Levine S & Mullins R Jr 1964 Estrogen administered neonatally affects adult (doi:10.1523/JNEUROSCI.0296-09.2009) sexual behavior in male and female rats. Science 144 185–187. Raskin K, Marie-Luce C, Picot M, Bernard V, Mailly P, Hardin-Pouzet H, (doi:10.1126/science.144.3615.185) Tronche F & Mhaouty-Kodja S 2012 Characterization of the spinal Losa SM, Todd KL, Sullivan AW, Cao J, Mickens JA & Patisaul HB 2011 nucleus of the bulbocavernosus neuromuscular system in male mice Neonatal exposure to genistein adversely impacts the ontogeny of lacking androgen receptor in the nervous system. Endocrinology 153 hypothalamic kisspeptin signaling pathways and ovarian development 3376–3385. (doi:10.1210/en.2012-1001) in the peripubertal female rat. Reproductive Toxicology 31 280–289. Rubin BS, Lenkowski JR, Schaeberle CM, Vandenberg LN, Ronsheim PM & (doi:10.1016/j.reprotox.2010.10.002) Soto AM 2006 Evidence of altered brain sexual differentiation in mice Losal-Ward SM, Todd KL, McCaffrey KA, Tsutsui K & Patisaul HB 2012 exposed perinatally to low, environmentally relevant levels of bisphenol Disrupted organization of RFamide pathways in the hypothalamus is A. Endocrinology 147 3681–3691. (doi:10.1210/en.2006-0189) associated with advanced puberty in female rats neonatally exposed to Ryan BC, Hotchkiss AK, Crofton KM & Gray LE Jr 2010 In utero and bisphenol A. Biology of Reproduction 87 28. (doi:10.1095/biolreprod.112. lactational exposure to bisphenol A, in contrast to ethinyl estradiol, 100826) does not alter sexually dimorphic behavior, puberty, fertility, and Martel KL & Baum MJ 2009 Adult testosterone treatment but not surgical anatomy of female LE rats. Toxicological Sciences 114 133–148. disruption of vomeronasal function augments male-typical sexual (doi:10.1093/toxsci/kfp266) behavior in female mice. Journal of Neuroscience 29 7658–7666. Takagi H, Shibutani M, Masutomi N, Uneyama C, Takahashi N, Mitsumori K (doi:10.1523/JNEUROSCI.1311-09.2009) & Hirose M 2004 Lack of maternal dietary exposure effects of Mayer C, Acosta-Martinez M, Dubois SL, Wolfe A, Radovick S, Boehm U & bisphenol A and nonylphenol during the critical period for brain sexual Levine JE 2010 Timing and completion of puberty in female mice differentiation on the reproductive/endocrine systems in later life. depend on estrogen receptor a-signaling in kisspeptin neurons. PNAS Archives of Toxicology 78 97–105. (doi:10.1007/s00204-003-0517-0) 107 22693–22698. (doi:10.1073/pnas.1012406108) Taylor JA, Vom Saal FS, Welshons WV, Drury B, Rottinghaus G, Hunt PA, Milligan SR, Khan O & Nash M 1998 Competitive binding of xenobiotic Toutain P-L, Laffont CM & VandeVoort CA 2011 Similarity of bisphenol oestrogens to rat a-fetoprotein and to sex steroid binding proteins in A pharmacokinetics in rhesus monkeys and mice: relevance for human human and rainbow trout (Oncorhynchus mykiss) plasma. General and exposure. Environmental Health Perspectives 119 422–430. (doi:10.1289/ Comparative Endocrinology 112 89–95. (doi:10.1006/gcen.1998.7146) ehp.1002514)

Tena-Sempere M 2010 Kisspeptin/GPR54 system as potential target for health at current levels of exposure. Reproductive Toxicology 24 131–138. endocrine disruption of reproductive development and function. (doi:10.1016/j.reprotox.2007.07.005) International Journal of Andrology 33 360–368. (doi:10.1111/j.1365- Welshons WV, Thayer KA, Judy BM, Taylor JA, Curran EM & vom Saal FS 2605.2009.01012.x) 2003 Large effects from small exposures. I. Mechanisms for endocrine- Tyl RW, Myers CB, Marr MC, Sloan CS, Castillo NP, Veselica MM, disrupting chemicals with estrogenic activity. Environmental Health Seely JC, Dimond SS, Van Miller JP, Shiotsuka RN et al. 2008 Perspectives 111 994–1006. (doi:10.1289/ehp.5494) Two-generation reproductive toxicity study of dietary bisphenol A Wintermantel TM, Campbell RE, Porteous R, Bock D, Gro¨ne H-J, Todman in CD-1 (Swiss) mice. Toxicological Sciences 104 362–384. (doi:10.1093/ MG, Korach KS, Greiner E, Pe´rez CA, Schu¨tz G et al. 2006 Deﬁnition of toxsci/kfn084) estrogen receptor pathway critical for estrogen positive feedback to Vandenberg LN, Hauser R, Marcus M, Olea N & Welshons WV 2007 Human gonadotropin-releasing hormone neurons and fertility. Neuron 52 exposure to bisphenol A (BPA). Reproductive Toxicology 24 139–177. 271–280. (doi:10.1016/j.neuron.2006.07.023) (doi:10.1016/j.reprotox.2007.07.010) Xi W, Lee CKF, Yeung WSB, Giesy JP, Wong MH, Zhang X, Hecker M & Vom Saal FS, Akingbemi BT, Belcher SM, Birnbaum LS, Crain DA, Eriksen M, Wong CK 2011 Effect of perinatal and postnatal bisphenol A exposure Farabollini F, Guillette LJ Jr, Hauser R, Heindel JJ et al. 2007 Chapel to the regulatory circuits at the hypothalamus–pituitary–gonadal axis Hill bisphenol A expert panel consensus statement: integration of of CD-1 mice. Reproductive Toxicology 31 409–417. (doi:10.1016/ mechanisms, effects in animals and potential to impact human j.reprotox.2010.12.002)

Received in ﬁnal form 18 December 2013 Accepted 6 January 2014 Accepted Preprint published online 8 January 2014 Journal of Endocrinology