bisphenol et hyperactivité (1) PDF

Nom original: bisphenol et hyperactivité (1).pdf

Ce fichier au format PDF 1.4 a été généré par Arbortext Advanced Print Publisher 9.1.510/W Unicode / Acrobat Distiller 10.1.8 (Windows), et a été envoyé sur Fichier-PDF le 17/01/2015 à 19:02, depuis l'adresse IP 89.225.x.x. La présente page de téléchargement du fichier a été vue 700 fois.
Taille du document: 1.1 Mo (6 pages).
Confidentialité: fichier public

Aperçu du document

Low-dose exposure to bisphenol A and replacement
bisphenol S induces precocious hypothalamic
neurogenesis in embryonic zebrafish
Cassandra D. Kincha,b,c, Kingsley Ibhazehiebob,c, Joo-Hyun Jeongb,c, Hamid R. Habibia, and Deborah M. Kurraschb,c,1
Departments of aBiological Sciences and bMedical Genetics and cAlberta Children’s Hospital Research Institute, University of Calgary, Calgary, AB, Canada
T2N 4N1

Bisphenol A (BPA), a ubiquitous endocrine disruptor that is present
in many household products, has been linked to obesity, cancer,
and, most relevant here, childhood neurological disorders such as
anxiety and hyperactivity. However, how BPA exposure translates
into these neurodevelopmental disorders remains poorly understood. Here, we used zebrafish to link BPA mechanistically to
disease etiology. Strikingly, treatment of embryonic zebrafish
with very low-dose BPA (0.0068 μM, 1,000-fold lower than the
accepted human daily exposure) and bisphenol S (BPS), a common
analog used in BPA-free products, resulted in 180% and 240%
increases, respectively, in neuronal birth (neurogenesis) within
the hypothalamus, a highly conserved brain region involved in
hyperactivity. Furthermore, restricted BPA/BPS exposure specifically during the neurogenic window caused later hyperactive
behaviors in zebrafish larvae. Unexpectedly, we show that BPAmediated precocious neurogenesis and the concomitant behavioral phenotype were not dependent on predicted estrogen receptors but relied on androgen receptor-mediated up-regulation of
aromatase. Although human epidemiological results are still
emerging, an association between high maternal urinary BPA during gestation and hyperactivity and other behavioral disturbances in
the child has been suggested. Our studies here provide mechanistic
support that the neurogenic period indeed may be a window of
vulnerability and uncovers previously unexplored avenues of research
into how endocrine disruptors might perturb early brain development. Furthermore, our results show that BPA-free products are not
necessarily safer and support the removal of all bisphenols from
consumer merchandise.
endocrine disruption

Indeed, a strong negative correlation between BPA and BPS
levels exists, whereby thermal paper that contained high quantities of BPA (milligrams per gram) demonstrated low quantities of BPS (nanograms per gram), and vice versa, suggesting
that BPS is the primary replacement for BPA in thermal receipts
(20). A recent examination of urine samples in the United States
and Asia confirmed previous work showing that 93% of people had
detectable levels of BPA but surprisingly showed that 81% had
detectable levels of BPS, illustrating the wide-spread use of this
poorly known bisphenol analog in consumer products (21).
The physiological effects of BPA on adults are well documented, but the mode of BPA action on the developing brain has
yet to be defined clearly, especially in the hypothalamus, which
plays a known role in neuroendocrine disorders that are on the
rise, including obesity and precocious puberty, as well as anxiety
and hyperactivity (1, 2, 4, 10, 22). BPA is commonly thought to
exert its effects by acting as a weak estrogen receptor (ER) agonist (8), although antagonism at androgen receptors (ARs) and
thyroid receptors (ThRs) has been shown also (23). Proliferating
cells in the developing hypothalamus express ERs and ARs (24,
25), and a role for sex steroids (e.g., estrogen, testosterone) in
regulating neurogenesis is emerging (26–29). Although results
are mixed (30, 31), neurobehavioral studies in humans suggest
that the prenatal period could be a window of BPA vulnerability
(1–4, 7). Given that BPA is a known endocrine disruptor and
that steroid hormones increasingly are being shown to play a role
in cell differentiation, our objective here was to determine whether
BPA-mediated behavioral phenotypes were the consequence
Significance

| androgen receptor | aromatase | hyperactivity

Here we demonstrate that bisphenol A (BPA) exposure during
a time point analogous to the second trimester in humans has
real and measurable effects on brain development and behavior. Furthermore, our study is the first, to our knowledge, to
show that bisphenol S, a replacement used in BPA-free products, equally affects neurodevelopment. These findings suggest that BPA-free products are not necessarily safe and
support a societal push to remove all structurally similar bisphenol analogues and other compounds with endocrine-disruptive
activity from consumer goods. Our data here, combined with
over a dozen physiological and behavioral human studies that
begin to point to the prenatal period as a BPA window of vulnerability, suggest that pregnant mothers limit exposure to plastics and receipts.

I

n humans and rodent models, gestational exposure to bisphenol A (BPA) has been associated with increased risk of developing social (e.g., aggression), psychiatric (e.g., depression),
and behavioral (e.g., hyperactivity) challenges later in life (1–7).
BPA is a compound used in the production of diverse consumer
products, ranging from baby bottles to thermal paper used for
credit card receipts (8–10). Even though adults can experience
adverse health following continued exposure to BPA, the fetal
brain is especially vulnerable because of an immature xenobioticmetabolizing system and blood–brain barrier (11, 12). Exactly
how BPA exposure in utero translates into neurodevelopmental
disorders later in life is only beginning to be explored (13–15).
Despite a wide body of research illustrating adverse effects of
BPA, controversy exists around the true effects of low-dose exposure, as is most often the case in humans. In accordance with
standardized toxicological testing procedures, government agencies in the United States (the US Environmental Protection
Agency, USEPA), Canada (Health Canada), and Europe (the
European Food Safety Authority, EFSA) have established tolerable daily intake levels, ranging from 25–50 μg BPA·kg body
weight−1·d−1 (16–18). Given these restrictions and societal pressure, manufacturers seeking BPA alternatives have turned to
primarily bisphenol S (BPS) to produce “BPA-free” products (19).
www.pnas.org/cgi/doi/10.1073/pnas.1417731112

Author contributions: C.D.K., H.R.H., and D.M.K. designed research; C.D.K., K.I., and J.-H.J.
performed research; C.D.K. analyzed data; C.D.K. and D.M.K. wrote the paper; and H.R.H.
provided intellectual input on experimental design and data analysis.
The authors declare no conflict of interest.
*This Direct Submission article had a prearranged editor.
Freely available online through the PNAS open access option.
1

To whom correspondence should be addressed. Email: kurrasch@ucalgary.ca.

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.
1073/pnas.1417731112/-/DCSupplemental.

PNAS Early Edition | 1 of 6

ENVIRONMENTAL
SCIENCES

Edited* by Joan V. Ruderman, Harvard Medical School, Boston, MA, and approved November 26, 2014 (received for review September 16, 2014)

of altered neurogenesis, a developmental process that occurs
during the second trimester of gestation. Further, we also
studied whether BPS, a common replacement for BPA, likewise
causes precocious neurogenesis and concomitant hyperactive
behaviors.
Results
BPA Exposure During Hypothalamic Neurogenesis Induces Hyperactivity.

First, we sought to recapitulate the human data using the neurodevelopmentally similar zebrafish as a model and asked whether
acute BPA exposure during a time point analogous to gestation
would cause behavioral changes in larvae. Specifically, zebrafish
embryos were exposed to BPA just before (10–16 h post fertilization; hpf), at the onset of (16–24 hpf), and at the peak of (24–36
hpf) hypothalamic neurogenesis. BPA was washed out after each
restricted time point, and zebrafish were assayed at the larval stage,
5 d post fertilization (dpf), for changes in locomotor activity (Fig.
1A; see SI Materials and Methods for further dosing information).
Low-dose BPA exposure at 0.1 μM just before neurogenesis (10–16
hpf) had no effect on locomotion but resulted in 2.8- and 2.9-fold
increases in locomotor activity when exposure occurred during the
early (16–24 hpf) and late (24–36 hpf) neurogenic periods, respectively (Fig. 1 B and C). Precedents in zebrafish demonstrate
that this type of locomotor activity (i.e., hyperactivity burst) can be
an indicator of anxiety-like behavior (32, 33). We also found that
chronic BPA exposure (across 0–5 dpf) resulted in an inverted
U-shaped hyperactivity dose–response curve (Fig. S1A), consistent
with previous findings (32). Given that physiological responses to
endocrine disruptors are known to be biphasic (10), with elevated
responses at nanomolar concentrations, we next asked whether
a lower dose of BPA also would cause hyperactive behavior. Additionally, because BPA is thought to act via ERs, we sought to

Fig. 1. BPA exposure induces hyperactive behaviors in larval zebrafish during
the window of hypothalamic neurogenesis. (A) Treatment paradigm for BPA
exposure. (B) Representative locomotor activity scribes with fish (green) and
movement (red) shown. Locomotion in 5-dpf controls and groups exposed to
0.1 μM BPA from 10–16 hpf is compared with locomotion in groups exposed to
0.1 μM BPA from 16–24 hpf and 24–36 hpf. (C) Quantified locomotor activity
for 5-dpf zebrafish treated with 0.1 μM BPA from 10–16 hpf, 16–24 hpf, and
24–36 hpf. (D) Locomotor activity in 5-dpf zebrafish coexposed from 0–5 dpf to
0.0068 μM BPA + 1 μM ICI. (E) Locomotor activity in 5-dpf AroB morphants
exposed from 0–5 dpf to 0.0068 μM BPA. (F) Locomotor activity in 5-dpf
zebrafish coexposed from 24–48 hpf to 1 μM BPA + FAD. (D and E) BPA exposure in AroB morphants and BPA+ICI treatment were run in the same experiment. Results have been separated for simplicity, and controls are shown
twice. Data in C–F are shown as mean ± SEM; *P < 0.05, **P < 0.01, ***P <
0.001, ****P < 0.0001 [one-way ANOVA, Tukey’s Honestly Significant Difference test (Tukey’s HSD)]; n = 5–10 fish per group.

2 of 6 | www.pnas.org/cgi/doi/10.1073/pnas.1417731112

determine whether ER inhibition alleviated BPA effects on
hyperactivity. Indeed, the same BPA dose found in a local water
body (Oldman River, Alberta, Canada) (34) (0.0068 μM, considered a very low dose) (Fig. 1D) and a more moderate BPA dose
(1 μM) (Fig. S1B) caused hyperactivity that, unexpectedly, was not
blocked by the broad ER ligand ICI 182,780 (hereafter, “ICI”)
(Figs. 1D and 2C). This failure to antagonize BPA effects indicated
that BPA might not act via classic nuclear (i.e., ERα, ERβ) or
membrane-bound (i.e., mER, GPR30) ERs. This same ICI dose
was sufficient to antagonize ER activity, because it blocked vtg1
expression in gonad and liver tissues (Fig. S2).
These data were puzzling, because BPA generally is considered to act via ERs. We therefore looked to other components of
the estrogen synthesis pathway and asked whether aromatase B
(AroB), the key enzyme for local estradiol synthesis, which is
expressed in hypothalamic progenitor cells, mediates the effects
of BPA. First, we relied on transient knockdown of AroB with
targeted morpholinos (MOs) (Fig. S3) and observed a complete
block of the BPA-mediated increase in locomotor activity at both
very low (0.0068 μM) (Fig. 1E) and moderate (1 μM) (Fig. S1C)
BPA doses. Second, we used the selective aromatase inhibitor
fadrozole (FAD) (1 μM) to determine whether AroB catalytic
activity was required for BPA-mediated behavioral changes. Indeed, coexposure to very-low-dose BPA (0.0068 μM) + FAD
(1 μM) (Fig. 1F) and moderate-dose BPA (1 μM) + FAD (1 μM)
(Fig. S1D) during the hypothalamic neurogenic window (24–48
hpf) lowered BPA-induced hyperactivity nearly to control levels,
suggesting that AroB enzymatic activity indeed is required.
BPA Causes Precocious Neurogenesis in the Hypothalamus. Next, we
sought to test our hypothesis that these behavioral changes are
the result of altered neurogenesis in the hypothalamus. Zebrafish hypothalamic progenitors undergo neuronal differentiation
between 18–36 hpf (35), and neural development is complete by
48 hpf. By 5 dpf, zebrafish emerge as prey-seeking larva. To
assess quantitatively the timing of neuronal birth in embryonic
zebrafish exposed from 0–5 dpf to very-low-dose (0.0068 μM)
BPA or vehicle control, we performed birthdating experiments
by pulse-labeling embryos with 5-ethynyl-2′-deoxyuridine (EdU)
at specific time points before (9 and 12 hpf), during (24 and 36
hpf), and after (48 hpf) the neurogenic window (Figs. 2A and 3).
We then assessed 5-dpf zebrafish for the number of neurons
born (i.e., that exited the cell cycle) at each time point of EdU
exposure (which therefore retained high levels of the thymidine
analog) and that also were colabeled with the early neuronal
marker HuC/D (Fig. 3). The number of newly born neurons in
the hypothalamus of BPA-treated embryos increased significantly, to 180% relative to controls (set at 100%), at 24 hpf (Figs.
2B and 3 and Fig. S4A) and decreased to 60% of control levels by
36 hpf (Figs. 2 and 3 and Fig. S4B), consistent with the notion
that precocious neurogenesis at 24 hpf exhausted a portion of the
progenitor pool normally reserved for 36 hpf. In contrast, hypothalamic neurogenesis was not significantly altered at 9, 12, or
48 hpf—time points outside the window of neurogenesis (Fig.
2B). Moreover, neuronal birth in several other embryonic brain
regions was examined, including the thalamus, tectum, and hindbrain, and no significant changes in neurogenesis were detected;
thalamic data are shown in Fig. 4. In addition, a moderate (1 μM)
BPA dose did not increase neurogenesis statistically in zebrafish
pulsed with EdU at 24 hpf (Fig. S4A) or decrease neurogenesis in
zebrafish pulsed with EdU at 36 hpf (Fig. S4B). Taken together
these data suggest that the effect of BPA exposure is brain regionspecific and nonlinear, with nanomolar concentrations perhaps
having greater potency in inducing neurogenesis than micromolar
concentrations.
Given that BPA is widely considered to function as a weak
estrogen agonist (23) [the BPA EC50 is 10,000-fold lower than that
of endogenous estrogen (36)] and that estrogen has been shown to
regulate neurogenesis (26, 28), we examined whether the same
low dose (0.0068 μM) of estrogen (17β-estradiol) likewise caused
precocious hypothalamic neurogenesis. Strikingly, treatment with
Kinch et al.

0.0068 μM 17β-estradiol had no effect on hypothalamic neuronal
birth at 24 hpf (Fig. 2D) but significantly induced neurogenesis at
36 hpf (Fig. 2E), perhaps unmasking an unappreciated temporal
competence window in hypothalamic development for BPA- and
17β-estradiol–sensitive periods. Consistent with our behavioral
data, coexposure with 0.0068 μM BPA + 1 μM ICI failed to reduce
significantly the number of neurons born at 24 hpf (Fig. 2D),
suggesting that BPA functions in a nonestrogenic manner.
Mechanistic Evaluation of BPA-Induced Precocious Neurogenesis. To
determine whether AroB also plays a role in BPA-mediated
changes in hypothalamic neurogenesis, we coexposed developing
embryos to very-low-dose (0.0068 μM) BPA + 1 μM FAD.
Strikingly, coexposure to BPA + FAD significantly reduced the
number of neurons born at 24 hpf, suggesting that BPA-induced
precocious hypothalamic neurogenesis also is dependent on
AroB (Fig. 5 A and E). Moreover, increased hypothalamic
neurogenesis was attenuated in AroB morphants exposed to
very-low-dose BPA (Fig. 5 B and E). BPA has been shown
previously to bind a wide range of other receptors, including
ThRs, ARs, and estrogen-related receptors (ERRs) (23, 37). To
identify the receptor upstream of AroB activation, we exposed
zebrafish to the following treatment paradigms: (i) BPA (0.0068
μM) and the AR antagonist flutamide (6.17 μM) (38); (ii) BPA
(0.0068 μM) and the ThRα/ThRβ transcriptional repressor amiodarone (50 nM) (39); or (iii) the ERR GSK4716 agonist alone (0.1
μM; no broad antagonist is commercially available) (32). Coexposure to BPA + flutamide significantly attenuated neurogenesis
by 40% at 24 hpf, suggesting that AR activation is at least partially
required to induce precocious neurogenesis (Fig. 5 B and E). In
contrast, cotreatment with BPA + amiodarone did not reduce
significantly the number of neurons born, and, conversely, treatment with GSK4716 did not increase neurogenesis (Fig. 5B).
Because BPA is commonly thought to act as an AR antagonist
and not an AR agonist (23), we reasoned that if BPA indeed
activates ARs either directly or indirectly, exposure to the endogenous AR agonist should phenocopy BPA-mediated precocious neurogenesis. Therefore we treated zebrafish embryos
with dihydrotestosterone (DHT), a nonaromatizable androgen
that is unable to be converted by AroB to estradiol and which
Kinch et al.

BPS Exposure Alters Brain Development and Behavior. Because of
the societal push to rid consumer products of BPA, manufacturers have replaced BPA with structurally similar monomer
compounds, particularly BPS (Fig. 6A) (19). Because BPS also
displays endocrine-disruptive activity (46), we examined whether
BPS had effects similar to those of BPA on precocious hypothalamic neurogenesis and pursuant hyperactive behaviors. We

Fig. 3. BPA exposure induces precocious neuronal birth in the hypothalamus.
Representative immunohistochemistry images of rostral hypothalamus in larval zebrafish exposed to 0.0068 μM BPA at 0–5 dpf. EdU (red), α-HuC (green),
and merged hypothalamic sections are shown. The red box in the DAPI (blue)stained image marks the hypothalamus. Mth, mouth. (Scale bar: 50 μm.)

PNAS Early Edition | 3 of 6

ENVIRONMENTAL
SCIENCES

Fig. 2. Precocious neurogenesis in the hypothalamus following BPA exposure is independent of ER signaling. (A) Cartoon of the experimental design
and sample neurogenic curves for BPA-exposed (blue) and control (black)
groups. (B) Quantification of neurons born at each time point in BPA (0.0068
μM) relative to vehicle treatment (set to 100%). Neuronal birth was identified by EdU and α-HuC colabeling in BPA-treated and control coronal sections through the hypothalamus. Data are shown as mean ± SEM; *P < 0.05
(Student’s t test); n = 3–11 fish per group. (C) Representation of ICI interaction with ERs. *ICI 182,780 is an antagonist to genomic ERs and membrane-bound ER (mER) but is an agonist toward GPR30. (D) Neuronal birth in
5-dpf zebrafish exposed to 0.0068 μM estradiol or 0.0068 μM BPA or coexposed to 0.0068 μM BPA + 1 μM ICI and pulsed with EdU at 24 hpf. (E)
Neuronal birth in 5-dpf zebrafish exposed to 0.0068 μM 17β-Estradiol or
0.0068 μM BPA and pulsed with EdU at late neurogenesis (36 hpf). Data in
D and E are shown as mean ± SEM; *P < 0.05, **P < 0.01, ****P < 0.0001
(ANOVA, Tukey’s HSD); n = 3–11.

binds AR with greater affinity than testosterone (40). Zebrafish
exposed to 1 μM DHT (41) displayed a significant increase in the
number of early hypothalamic neurons born (160%), whereas
cotreatment of embryos with DHT + flutamide (6.17 μM) significantly attenuated neuronal birth (Fig. 5C). In addition,
coexposure of DHT (1 μM) + ICI (1 μM) failed to block DHTmediated increased neurogenesis, further suggesting that verylow-dose BPA at this time point acts through ARs and not ERs
to promote hypothalamic neurogenesis.
This unexpected finding led us to question how AR activation
links to AroB catalytic activity. Evidence supporting AR-mediated regulation of AroB activity in the hypothalamus was shown
previously using androgen-insensitive testicular feminized male
rats (42). Moreover, studies in rat and nonhuman primates show
that androgen treatment up-regulates aromatase expression and
catalytic activity within the nonhuman primate and rat hypothalamus (43). In zebrafish, androgens have been shown to increase AroB (cyp19a1b) expression, although this up-regulation
was attributed to the conversion of androgens to estrogenic
metabolites, which in turn activate ERs that bind to AroB regulatory elements (41). However, the AroB promoter does contain
androgen response elements (AREs) (44). Thus, to characterize
the link between BPA-mediated precocious neurogenesis and ARs
and AroB catalytic activity, we confirmed that AroB (cyp19a1b)
expression was induced in larval zebrafish exposed to 0.0068 μM
BPA (45). In addition, we explored whether cotreatment with
BPA (0.0068 μM) and the AR antagonist flutamide (6.17 μM)
affected AroB (cyp19a1b) expression. Using quantitative reverse
transcriptase PCR (qRT-PCR), we found that AroB (cyp19a1b)
transcripts at 48 hpf were elevated in zebrafish treated from 8–48
hpf with both BPA (0.0068 μM) and DHT (1 μM) (Fig. 5D).
Significantly, coexposure to BPA + flutamide (6.17 μM) as well as
DHT (1 μM) + flutamide (6.17 μM) reduced cyp19a1b expression
when these compounds were administered across the neurogenic
window (8–48 hpf), showing that BPA can induce transcription of
AroB (cyp19a1b) via ARs specifically during this developmental
period. Interestingly, coexposure of BPA (0.0068 μM) and ICI
(1 μM) as well as DHT (1 μM) and ICI (1 μM) also significantly
decreased AroB (cyp19a1b) transcript levels (Fig. 5D). Thus, because antagonism of ERs had no effect on BPA- and DHTmediated precocious neurogenesis (Figs. 2D and 5C), we propose
that ER-driven AroB transcription is uncoupled from the neurogenic phenotype of BPA-exposed zebrafish. Combined, these data
support the hypothesis that BPA might act via agonism of ARs to
increase AroB transcription to drive precocious neurogenesis.

Fig. 4. BPA exposure does not cause precocious neurogenesis in other brain
regions. (A) Quantification of neurons born at 24 hpf in the thalamus of
larvae exposed to 0.0068 μM BPA at 0–5 dpf. Data are shown as mean ± SEM;
P > 0.05 (Student’s t test); n = 3 or 4 fish. (B) Representative immunohistochemistry images of the thalamus of a larval zebrafish exposed to 0.0068 μM
BPA at 0–5 dpf and pulsed with EdU at 24 hpf. Mth, mouth; ot, optic tract.
(Scale bar: 50 μm.)

exposed embryonic zebrafish to BPS at the same very low BPA
dose used previously, 0.0068 μM. Strikingly, this BPS exposure
resulted in a 240% increase in neuronal birth in the rostral hypothalamus at 24 hpf (Fig. 6B) and a significant (160%) increase
in the same locomotor bursting activity, which was reduced by
transient knockdown of AroB (Fig. 6D) but not by treatment
with 1 μM ICI (Fig. 6C). Together, these data imply that both
BPS and BPA influence hypothalamic development and may act
through a similar AroB-mediated mechanism.
Discussion
Our study is the first, to our knowledge, to show that BPA/BPS
can alter the normal developmental timing of this critical neuroendocrine center, the consequences of which can lead to early
synaptogenesis and improper fine-tuning of the brain later in
development. In addition, we are the first, to our knowledge, to
link BPA to AR and AroB signaling during hypothalamic development. Finally, our study provides further evidence for the
hypothesis that the prenatal period is indeed a window of vulnerability to BPA/BPS in humans and suggests that pregnant
woman should be particularly mindful of their use of plastics and
the handling of thermal receipt papers throughout gestation,
especially during the second trimester.
Region-specific perturbations in neurogenesis have been
linked to altered synaptic connectivity (47), which can lead to
downstream effects on proper pruning and manifest in behavioral deficits later in life. Exactly how the timing of neurogenesis
becomes altered is an active area of research. For example,
proliferating cells in the developing cortex, hippocampus, olfactory bulb, and hypothalamus all express ERs (48) and ARs
(25, 49), and a clear role for sex steroids (e.g., estrogen, testosterone) in regulating neurogenesis in vivo (26, 28) and in vitro
(50, 51) is emerging, especially in the adult brain (27, 29, 52). In
embryonic and adult zebrafish, ERs and AroB are coexpressed in
hypothalamic radial glia and are thought to control differentiation and migrational changes that occur in response to estrogen
exposure (29, 53). These findings have led to the notion that
AroB-expressing radial glia are plastic and initiate adult neurogenesis in response to estrogenic challenges as a way to maintain
homeostatic physiologies (52). In the developing mammalian
brain, AroB is broadly expressed in radial glia throughout the
CNS and becomes restricted to only a small subset of neurogenic
niches by adulthood, the most studied of which is the hippocampus (26, 43). Indeed, blocking ERs halfway through the
mouse neurogenic window (at embryonic day15) results in decreased proliferation in the developing neocortex (26), illustrating the sex steroid sensitivity of progenitors in mammals.
Within the adult zebrafish hypothalamus, antagonism of ERs
also decreases cell proliferation (29), demonstrating conserved
roles of ER signaling among teleosts and mammalian progenitors. Interestingly, AroB inhibition in the adult zebrafish forebrain
4 of 6 | www.pnas.org/cgi/doi/10.1073/pnas.1417731112

was shown to have no significant effect on progenitor cell proliferation or neuronal differentiation (29), implying differences
in progenitor biology in embryonic zebrafish versus adults. Furthermore, we propose that we have uncovered a temporal
competency window for BPA effects on hypothalamic neurogenesis that is separate from estrogen signaling. This phenotype
could be caused by changes in the expression of all three
zebrafish ERs (ERα, ERβ1, and ERβ2), which increases dramatically between 24 and 48 hpf (53). Given that the peak of
BPA susceptibility occurs at 24 hpf, and we show a role for ARs
in BPA-mediated precocious neurogenesis, it will be interesting
to see whether AR expression increases earlier than ER expression in the developing hypothalamus. Considered together,
the known influence of sex steroids on adult neurogenesis and
the evolving role for these hormones in regulating neural development point to a need to understand fully how endocrinedisrupting chemicals can influence this process in utero.
Evidence for androgen pathway-mediated increases in AroB
transcription is conflicting in both mammalian and lower vertebrates. For example, up-regulation of AroB transcripts following
androgen treatment has been shown in mammals, ricefield eel,
and in teleosts such as zebrafish and channel catfish; direct AR
regulation of AroB has been demonstrated in ricefield eel, but in
teleosts increased AroB expression is attributed to signaling by
estrogenic metabolites and not to ARs directly (41, 43, 54–56).
In rodents (mice and rats), however, results of AR-mediated
changes in AroB expression are mixed, with reports that
androgens increase, decrease, or have no effect on AroB transcription (57–60). Interestingly, AREs have been identified in
the promoter of AroB (cyp19a1b) in teleosts and humans, suggesting that AR may modulate AroB directly; however, AREs
have not been identified in murine animals, perhaps suggesting
that there is an indirect mechanism of AroB modulation in these
species (43, 44). Here, through the use of an AR antagonist,
we show that AR pathway activation also can increase AroB

Fig. 5. BPA-induced precocious neurogenesis is mediated via ARs and aromatase. (A) Quantification of neuronal birth in 5-dpf zebrafish coexposed from
0–5 dpf to 1 μM FAD + 0.0068 μM BPA and pulsed with EdU at 24 hpf. (B)
Neuronal birth in 5-dpf zebrafish exposed from 0–5 dpf to 0.1 μM GSK4716
alone or coexposed to BPA + 50 nM amiodarone (AMIO) or to BPA + 6.17 μM
flutamide (FLU). The red arrow indicates exposure at 8–48 hpf. The hash mark
(#) indicates the AroB morphant (AroB-MO) exposed to BPA. (C) Neuronal birth
in 5-dpf zebrafish exposed from 0–5 dpf to 0.0068 μM BPA or 1 μM DHT or
coexposed to 1 μM DHT from 0–5 dpf and to 6.17 μM FLU (red arrow) for 8–48
hpf or to 1 μM ICI from 0–5 dpf. (D) Log-transformed relative AroB (cyp19a1b)
expression at 48 hpf in zebrafish exposed to 0.0068 μM BPA or 1 μM DHT or
coexposed to BPA or 1 μM DHT + 6.17 μM FLU or 1 μM ICI at 8–48 hpf. Data in
A–D are shown as mean ± SEM; *P < 0.05, **P < 0.01, ***P < 0.001, ****P <
0.0001 (ANOVA, Tukey’s HSD); n = 3–13. (E) Diagram illustrating targets of
various pharmacological agents and AroB MO.

Kinch et al.

expression in teleosts and starts to build a connection between
BPA and AR activation, which then modulates AroB. Because
BPA conventionally is thought to act as an AR antagonist, our
finding that BPA indeed might activate AR was unexpected;
however, in vitro studies previously have shown BPA agonism of
AR at similar concentrations (1–10 nM) (23). In mammals, hypothalamic AR and AroB transcript expression is higher in
prenatal males than in prenatal females, and thus our results
linking BPA and AR-mediated AroB up-regulation with altered
brain development might provide insight into why certain neurodevelopmental etiologies are male-specific (60, 61). Nevertheless, future studies are needed to determine the exact
mechanism by which BPA modulates AR signaling and AroB
up-regulation, leading to precocious neurogenesis.
In our model (Fig. S5), BPA activation of ARs results in the upregulation of AroB transcription, presumably leading to increased
estradiol production in the smooth endoplasmic reticulum and
diffusion into the cytoplasm. The exact downstream targets of
locally produced estradiol remain uncertain, because here we
show that ERs, as likely targets of either BPA or locally produced
estradiol, are not involved in BPA-mediated precocious neurogenesis. Alternatively, AroB also converts androstenedione to the
estradiol intermediate estrone, which may play an unappreciated
role in hypothalamic progenitors. Continued examination of the
biological effects of local estrogen synthesis on hypothalamic
neural progenitor cells may yield important insights into novel
mechanisms of endocrine-disrupting chemicals.
Controversy regarding the true effects of low-dose BPA exposure on human health remains, probably reflecting the variability
that accompanies different assays and model systems. Our study
shows that low-dose BPA and BPS exposure has physiological
effects. We purposefully chose the 0.0068-μM dose because this is
the exact concentration of BPA measured in the Oldman River,
a major Albertan waterway that serves as a life source to two
major urban centers (34). For comparison, waterborne exposure
of 1 μM (12 μg/kg) BPA is comparable to the BPA concentration
normally found in human placental tissue (12.7 μg/kg) (32, 62);
thus the 0.0068-μM doses used herein are magnitudes lower than
levels found in human placenta and 100-fold lower than circulating levels measured in fetal serum (0.0101 μM) (62). When determining tolerable daily intake levels, government organizations
such as USEPA, Health Canada, and the EFSA rely on linear
dose–response relationships and so begin compound testing at
high doses, then lowering the dose to the level at which no
physiological effect is observed. However, many endocrine-disrupting compounds follow alternative U-shaped dose–response
Kinch et al.

Materials and Methods
For further information on materials and methods, see SI Materials
and Methods.
Zebrafish Husbandry and Contaminant Preparation. All protocols and procedures
were approved by the Health Science Animal Care Committee (protocol
#M10079) at University of Calgary in compliance with the Guidelines of the
Canadian Council of Animal Care. Wild-type zebrafish embryos were maintained at 28 °C in embryo medium (E3) as described by Westerfield, et al. (64).
BPA (239658; Sigma), BPS (103039; Sigma), and 17β-estradiol (E8875; Sigma)
contaminants as well as the pharmacological agents ICI 182,780 (I4409; Sigma),
fadrozole (F3806; Sigma), flutamide (F9397; Sigma), GSK4716 (G6173; Sigma),
and amiodarone (A8423; Sigma) were prepared in a 1:3 ratio of 0.002% (vol/vol)
1 M NaOH to 95% EtOH (vehicle) and were diluted to the final concentration in
E3. DHT (D-073; Cerilliant) treatments and vehicle controls were prepared in
0.08% (vol/vol ) MeOH in E3. Zebrafish embryos were immersed in treatment
within 3 hpf except where otherwise indicated. Chemical structures were
designed using ChemBioDraw 13.0 (PerkinElmer) software.
MO Analyses. Mixtures of ATG and splice-blocking MOs (AroB-MOs) were
engineered by Gene-Tools (Philomath) (5′-AGGCTTCCATCATCCCCAACTTCAT-3′), (5′-CGAGCCTGAGAGGACAACAAAGACA-3′) and were injected (2.6–
4.6 nL) at the one-cell stage.
Locomotor Behavior Assays. Larval zebrafish were maintained in 96-well
plates immersed in respective treatments from 0–5 dpf, and the duration of
hyperactivity bursts (33) was assayed by methods similar to those described
by Saili, et al. (32).
Neurogenesis Assessments. Contaminant-exposed embryos were pulsedlabeled with 0.01% (vol/vol) EdU (C10338; Molecular Probes) at 9, 12, 24, 36, and
48 hpf, corresponding to the window of neurogenesis. EdU was washed off
after 30 min, and embryos were replaced in their respective treatments until
they were killed at 5 dpf by overnight immersion in 4% paraformaldehyde
(PFA) at 4 °C. Processed slides were imaged, and EdU+/α-HuC+ cells were
counted by three independent persons to ensure accuracy. No differences in
cell morphology or total number of cells were noted between treatments.
Sectioning and Immunohistochemistry. After overnight fixation, larvae were
cryoprotected, embedded in optimum cutting temperature (OCT) compound
(Clear Frozen Section Compound; CA95057-838; VWR Scientific), snap frozen,
and kept at −80 °C. Embedded larvae were cryosectioned coronally through
the hypothalamus. After 20 min of antigen retrieval, slides were blocked in
5% normal goat serum (S1000; Vector Labs) for 1 h. Slides were incubated
overnight with α-HuC (A21271; Molecular Probes) and then were incubated
with Alexa Fluor 488 IgG (A11001; Molecular Probes) for 2 h. DAPI (D1308;
Molecular Probes) was applied, and slides were treated with Click-iT EdU Kit
(C10338; Molecular Probes).
Transcript Measurement Assays. qRT-PCR and PCR reactions were performed
on treated larvae at 48 hpf following the protocol in Kurrasch, et al. (65).
Primer sequences are given in Table S1.
Statistical Analyses. Assumptions of normality and equality of variance were
met, and ANOVAs with Tukey’s honestly significant difference (HSD) and

PNAS Early Edition | 5 of 6

ENVIRONMENTAL
SCIENCES

Fig. 6. BPS exposure increases neurogenesis and hyperactive behavior via
a mechanism similar to that of BPA. (A) Chemical structures of BPA and BPS.
(B) Neuronal birth in 5-dpf zebrafish exposed from 0–5 dpf to 0.0068 μM BPS.
Data are shown as mean ± SEM; *P < 0.05 (Student’s t test); n = 6. (C) Locomotor activity in fish coexposed from 0–5 dpf to 0.0068 μM BPS + 1 μM ICI.
(D) Locomotor activity in 5-dpf controls and AroB morphants exposed from
0–5 dpf to 0.0068 μM BPS. Data in C and D are shown as mean ± SEM; *P <
0.05 (ANOVA, Tukey’s HSD); n = 3–9. (C and D) BPS exposure in AroB morphants and BPS+ICI treatment were run in same experiment. Results have
been separated for simplicity, and controls are shown twice.

curves, whereby exposure to midrange concentrations activates
physiological defense mechanisms against the compound, but at
low-range concentrations, the compound mimics endogenous
hormones (10). Thus, our finding that BPA at a very low dose
(0.0068 μM) alters neurogenesis and that a moderate BPA (1 μM)
dose did not affect neurogenesis significantly calls for change in
government-sanctioned methods of assessing human tolerable
daily intake levels.
Recently, manufacturers have turned to BPS with little proper
toxicology testing to produce the “BPA-free” products (19)
demanded by society. Indeed, there is a strong negative correlation
between BPA and BPS levels in thermal receipts (20), showing
that BPS is the primary replacement for BPA. Because there are
many structurally similar bisphenol analogs [e.g., BPB, BPE, BPF,
BPS, and 4-cumylphenol] (63) that are potential candidates for
manufacturing, and given our data herein, a societal push to
remove all bisphenols from our consumer goods is justified.

Student’s t test were performed where indicated using Prism 6 (GraphPad
Software). qRT-PCR data were log transformed before analyses.
ACKNOWLEDGMENTS. We thank Gaurav Kaushik and Natalia Klenin for
technical assistance on this project. The research described in these

studies was supported by National Sciences and Engineering Research
Council of Canada (NSERC) Grant DG386445 (to D.M.K.), NSERC Discovery
Grants Program–Individual Grant 156910 (to H.R.H.), and NSERC Postgraduate Scholarships–Doctoral (2-year) Program Grant 459881-2014
(to C.D.K.).

1. Braun JM, et al. (2009) Prenatal bisphenol A exposure and early childhood behavior.
Environ Health Perspect 117(12):1945–1952.
2. Braun JM, et al. (2011) Impact of early-life bisphenol A exposure on behavior and
executive function in children. Pediatrics 128(5):873–882.
3. Perera F, et al. (2012) Prenatal bisphenol a exposure and child behavior in an innercity cohort. Environ Health Perspect 120(8):1190–1194.
4. Harley KG, et al. (2013) Prenatal and early childhood bisphenol A concentrations and
behavior in school-aged children. Environ Res 126:43–50.
5. Anderson OS, et al. (2013) Perinatal bisphenol A exposure promotes hyperactivity,
lean body composition, and hormonal responses across the murine life course. FASEB
J 27(4):1784–1792.
6. Tian YH, Baek JH, Lee SY, Jang CG (2010) Prenatal and postnatal exposure to bisphenol
a induces anxiolytic behaviors and cognitive deficits in mice. Synapse 64(6):432–439.
7. Evans SF, et al. (2014) Prenatal bisphenol A exposure and maternally reported behavior in boys and girls. Neurotoxicology 45C:91–99.
8. Krishnan AV, Stathis P, Permuth SF, Tokes L, Feldman D (1993) Bisphenol-A: An estrogenic substance is released from polycarbonate flasks during autoclaving. Endocrinology 132(6):2279–2286.
9. Liao C, Kannan K (2011) Widespread occurrence of bisphenol A in paper and paper
products: Implications for human exposure. Environ Sci Technol 45(21):9372–9379.
10. Vandenberg LN, Maffini MV, Sonnenschein C, Rubin BS, Soto AM (2009) Bisphenol-A
and the great divide: A review of controversies in the field of endocrine disruption.
Endocr Rev 30(1):75–95.
11. Nishikawa M, et al. (2010) Placental transfer of conjugated bisphenol A and subsequent reactivation in the rat fetus. Environ Health Perspect 118(9):1196–1203.
12. Adinolfi M (1985) The development of the human blood-CSF-brain barrier. Dev Med
Child Neurol 27(4):532–537.
13. Kim K, et al. (2009) Potencies of bisphenol A on the neuronal differentiation and
hippocampal neurogenesis. J Toxicol Environ Health A 72(21-22):1343–1351.
14. Komada M, et al. (2012) Maternal bisphenol A oral dosing relates to the acceleration of
neurogenesis in the developing neocortex of mouse fetuses. Toxicology 295(1-3):31–38.
15. Itoh K, Yaoi T, Fushiki S (2012) Bisphenol A, an endocrine-disrupting chemical, and
brain development. Neuropathology 32(4):447–457.
16. US Environmental Protection Agency (2008) Child-specific exposure factors handbook, EPA/600/R-06/096F, September 2008 (National Center for Environmental Assessment, Office of Research and Development, Washington, DC). Available at www.
epa.gov/ncea. Accessed Dec. 13, 2013.
17. Health Canada (2009) Survey of bisphenol A in canned drink products. Available at
www.hc-sc.gc.ca/fn-an/securit/packagemball/bpa/bpa_survey-enquete-can-eng.php.
Accessed December 13, 2013.
18. European Food Safety Authority (2007) EFSA re-evaluates safety of bisphenol A and sets
tolerable daily intake. Available at www.efsa.europa.eu/en/press/news/afc070129.htm. Accessed December 13, 2013.
19. Grignard E, Lapenna S, Bremer S (2012) Weak estrogenic transcriptional activities of
Bisphenol A and Bisphenol S. Toxicol In Vitro 26(5):727–731.
20. Liao C, Liu F, Kannan K (2012) Bisphenol s, a new bisphenol analogue, in paper
products and currency bills and its association with bisphenol a residues. Environ Sci
Technol 46(12):6515–6522.
21. Liao C, et al. (2012) Bisphenol S in urine from the United States and seven Asian
countries: Occurrence and human exposures. Environ Sci Technol 46(12):6860–6866.
22. Bourguignon JP, et al. (2010) Neuroendocrine disruption of pubertal timing and interactions between homeostasis of reproduction and energy balance. Mol Cell Endocrinol 324(1-2):110–120.
23. Wetherill YB, et al. (2007) In vitro molecular mechanisms of bisphenol A action. Reprod Toxicol 24(2):178–198.
24. MacLusky NJ, Lieberburg I, McEwen BS (1979) The development of estrogen receptor
systems in the rat brain: Perinatal development. Brain Res 178(1):129–142.
25. Gorelick DA, Watson W, Halpern ME (2008) Androgen receptor gene expression in
the developing and adult zebrafish brain. Dev Dyn 237(10):2987–2995.
26. Martínez-Cerdeño V, Noctor SC, Kriegstein AR (2006) Estradiol stimulates progenitor
cell division in the ventricular and subventricular zones of the embryonic neocortex.
Eur J Neurosci 24(12):3475–3488.
27. Fowler CD, Liu Y, Wang Z (2008) Estrogen and adult neurogenesis in the amygdala
and hypothalamus. Brain Res Brain Res Rev 57(2):342–351.
28. Vosges M, et al. (2012) 17α-Ethinylestradiol and nonylphenol affect the development
of forebrain GnRH neurons through an estrogen receptors-dependent pathway. Reprod Toxicol 33(2):198–204.
29. Diotel N, et al. (2013) Effects of estradiol in adult neurogenesis and brain repair in
zebrafish. Horm Behav 63(2):193–207.
30. Yolton K, et al. (2011) Prenatal exposure to bisphenol A and phthalates and infant
neurobehavior. Neurotoxicol Teratol 33(5):558–566.
31. Miodovnik A, et al. (2011) Endocrine disruptors and childhood social impairment.
Neurotoxicology 32(2):261–267.
32. Saili KS, et al. (2012) Neurodevelopmental low-dose bisphenol A exposure leads to early
life-stage hyperactivity and learning deficits in adult zebrafish. Toxicology 291(1-3):83–92.
33. Kalueff AV, et al.; Zebrafish Neuroscience Research Consortium (2013) Towards a comprehensive catalog of zebrafish behavior 1.0 and beyond. Zebrafish 10(1):70–86.

34. Sosiak A, Hebben T (2005) A Preliminary Survey of Pharmaceuticals and Endocrine
Disrupting Compounds in Treated Municipal Wastewaters and Receiving Rivers of
Alberta. Technical Report AE T/773 (Alberta Environment, Edmonton, AB, Canada).
35. Nesan D, Vijayan MM (2013) Role of glucocorticoid in developmental programming:
Evidence from zebrafish. Gen Comp Endocrinol 181:35–44.
36. Matthews JB, Twomey K, Zacharewski TR (2001) In vitro and in vivo interactions of
bisphenol A and its metabolite, bisphenol A glucuronide, with estrogen receptors
alpha and beta. Chem Res Toxicol 14(2):149–157.
37. Teng C, et al. (2013) Bisphenol A affects androgen receptor function via multiple
mechanisms. Chem Biol Interact 203(3):556–564.
38. Schiller V, et al. (2013) Transcriptome alterations in zebrafish embryos after exposure
to environmental estrogens and anti-androgens can reveal endocrine disruption.
Reprod Toxicol 42:210–223.
39. Liu YW, Chan WK (2002) Thyroid hormones are important for embryonic to larval
transitory phase in zebrafish. Differentiation 70(1):36–45.
40. Grino PB, Griffin JE, Wilson JD (1990) Testosterone at high concentrations interacts with the
human androgen receptor similarly to dihydrotestosterone. Endocrinology 126(2):1165–1172.
41. Mouriec K, et al. (2009) Androgens upregulate cyp19a1b (aromatase B) gene expression in the brain of zebrafish (Danio rerio) through estrogen receptors. Biol Reprod 80(5):889–896.
42. Roselli CE, Salisbury RL, Resko JA (1987) Genetic evidence for androgen-dependent and
independent control of aromatase activity in the rat brain. Endocrinology 121(6):2205–2210.
43. Roselli CE, Liu M, Hurn PD (2009) Brain aromatization: Classic roles and new perspectives. Semin Reprod Med 27(3):207–217.
44. Tong SK, Chung BC (2003) Analysis of zebrafish cyp19 promoters. J Steroid Biochem
Mol Biol 86(3-5):381–386.
45. Chung E, Genco MC, Megrelis L, Ruderman JV (2011) Effects of bisphenol A and triclocarban on brain-specific expression of aromatase in early zebrafish embryos. Proc
Natl Acad Sci USA 108(43):17732–17737.
46. Ji K, Hong S, Kho Y, Choi K (2013) Effects of bisphenol s exposure on endocrine
functions and reproduction of zebrafish. Environ Sci Technol 47(15):8793–8800.
47. Itoh K, Yaoi T, Fushiki S (2012) Bisphenol A, an endocrine-disrupting chemical, and
brain development. Neuropathology 32(4):447–457.
48. MacLusky NJ, Chaptal C, McEwen BS (1979) The development of estrogen receptor systems
in the rat brain and pituitary: Postnatal development. Brain Res 178(1):143–160.
49. Young WJ, Chang C (1998) Ontogeny and autoregulation of androgen receptor
mRNA expression in the nervous system. Endocrine 9(1):79–88.
50. Toran-Allerand CD (1976) Sex steroids and the development of the newborn mouse
hypothalamus and preoptic area in vitro: Implications for sexual differentiation. Brain
Res 106(2):407–412.
51. Murashov AK, Pak ES, Hendricks WA, Tatko LM (2004) 17beta-Estradiol enhances
neuronal differentiation of mouse embryonic stem cells. FEBS Lett 569(1-3):165–168.
52. Diotel N, et al. (2010) Aromatase in the brain of teleost fish: Expression, regulation
and putative functions. Front Neuroendocrinol 31(2):172–192.
53. Mouriec K, et al. (2009) Early regulation of brain aromatase (cyp19a1b) by estrogen
receptors during zebrafish development. Dev Dyn 238(10):2641–2651.
54. Zhang Y, et al. (2012) Androgen rather than estrogen up-regulates brain-type cytochrome
P450 aromatase (cyp19a1b) gene via tissue-specific promoters in the hermaphrodite teleost
ricefield eel Monopterus albus. Mol Cell Endocrinol 350(1):125–135.
55. Lassiter CS, Linney E (2007) Embryonic expression and steroid regulation of brain
aromatase cyp19a1b in zebrafish (Danio rerio). Zebrafish 4(1):49–57.
56. Kazeto Y, Trant JM (2005) Molecular biology of channel catfish brain cytochrome P450
aromatase (CYP19A2): Cloning, preovulatory induction of gene expression, hormonal gene
regulation and analysis of promoter region. J Mol Endocrinol 35(3):571–583.
57. Karolczak M, Küppers E, Beyer C (1998) Developmental expression and regulation of
aromatase- and 5alpha-reductase type I mRNA in the male and female mouse hypothalamus. J Neuroendocrinol 10(4):267–274.
58. Abe-Dohmae S, Tanaka R, Harada N (1994) Cell type- and region-specific expression of
aromatase mRNA in cultured brain cells. Brain Res Mol Brain Res 24(1-4):153–158.
59. Lephart ED, Simpson ER, Ojeda SR (1992) Effects of Cyclic AMP and Andre-gens on in
vitro Brain Aromatase Enzyme Activity During Prenatal Development in the Rat.
J Neuroendocrinol 4(1):29–36.
60. Negri-Cesi P, Colciago A, Motta M, Martini L, Celotti F (2001) Aromatase expression
and activity in male and female cultured rat hypothalamic neurons: Effect of androgens. Mol Cell Endocrinol 178(1-2):1–10.
61. Beyer C, Hutchison JB (1997) Androgens stimulate the morphological maturation of
embryonic hypothalamic aromatase-immunoreactive neurons in the mouse. Brain Res
Dev Brain Res 98(1):74–81.
62. Schönfelder G, et al. (2002) Parent bisphenol A accumulation in the human maternalfetal-placental unit. Environ Health Perspect 110(11):A703–A707.
63. Rosenmai AK, et al. (2014) Are structural analogues to bisphenol a safe alternatives?
Toxicol Sci 139(1):35–47.
64. Westerfield M (2000) The Zebrafish Book. A Guide for the Laboratory Use of Zebrafish
(Danio rerio) (Univ of Oregon Press, Eugene, OR), 4th Ed.
65. Kurrasch DM, Nevin LM, Wong JS, Baier H, Ingraham HA (2009) Neuroendocrine
transcriptional programs adapt dynamically to the supply and demand for neuropeptides as revealed in NSF mutant zebrafish. Neural Dev 4(22):1–16.

6 of 6 | www.pnas.org/cgi/doi/10.1073/pnas.1417731112

Kinch et al.