Transient retinoic acid signaling confers anterior-
posterior polarity to the inner ear
Jinwoong Boka,1, Steven Raftb, Kyoung-Ah Konga, Soo Kyung Kooc, Ursula C. Drägerd, and Doris K. Wub,1
aDepartment of Anatomy, BK21 Project for Medical Science, Yonsei University College of Medicine, Seoul 120-752, South Korea;bNational Institute on
Deafness and Other Communication Disorders, Rockville, MD 20850;cDepartment of Biomedical Sciences, National Institute of Health, Osong 363-951,
South Korea; anddDepartment of Psychiatry, University of Massachusetts Medical School, Worcester, MA 01655
Edited* by Brigid L.M. Hogan, Duke University Medical Center, Durham, NC, and approved November 11, 2010 (received for review July 19, 2010)
Vertebrate hearing and balance are based in complex asymmetries
of inner ear structure. Here, we identify retinoic acid (RA) as an
extrinsic signal that acts directly on the ear rudiment to affect its
compartmentalization along the anterior-posterior axis. A rostro-
caudal wave of RA activity, generated by tissues surrounding the
nascent ear, induces distinct responses from anterior and posterior
halves of the inner ear rudiment. Prolonged response to RA by
posterior otic tissue correlates with Tbx1 transcription and forma-
tion of mostly nonsensory inner ear structures. By contrast, ante-
rior otic tissue displays only a brief response to RA and forms
neuronal elements and most sensory structures of the inner ear.
axial specification|developmental compartments|morphogen
precisely positioned within the asymmetric membranous labyrinth
of the inner ear (Fig. 1A). Five vestibular sensory patches are
present in all vertebrate inner ears: the three cristae (anterior,
lateral, and posterior) that detect angular head movements and
two maculae (utricle and saccule) that detect linear acceleration.
The specialized organ for detecting sound in chickens and
mammals is the basilar papilla and organ of Corti, respectively.
The entire membranous labyrinth and its innervating neurons
are derived from an ectodermal thickening adjacent to the
hindbrain known as the otic placode. As the placode deepens to
form a cup and then pinches off to form the otocyst, some cells of
the otic epithelium delaminate to form neuroblasts of the
cochleovestibular ganglion (CVG). Inner ear sensory organs, and
the neurons that innervate them, are thought to arise from
a neural-sensory competent domain (NSD), most of which is
located in the anterior region of the otic cup (1). By contrast,
posterior otic epithelium forms nonsensory tissues and only one
sensory organ, the posterior crista. This basic organization of
functional elements in the ear is thought to be governed by
signals emanating from adjacent tissues (2, 3); however, molec-
ular mechanisms that establish the initial anterior-posterior (A-
P) asymmetry of the ear primordium are poorly defined. Here,
we show that a rostrocaudal wave of retinoic acid activity pro-
vides signals to the ear rudiment and establishes structural
asymmetries required for normal hearing and balance.
ormal hearing and balance require that discrete patches of
mechanosensory hair cells, each with a distinct function, be
Ectoderm Adjacent to the Otic Cup Confers A-P Polarity to the
Otocyst. A clear manifestation of A-P asymmetry in developing
amniote ears is the anterior expression of transcripts associated
with cochleovestibular ganglion neurogenesis. We performed tis-
sue transplantations in ovo to identify source(s) of signals that
specify the otic A-P axis in the chicken. Transplantations were
carried out at the otic cup stage (11–15 somite stages), before the
otic A-P axis is specified (4). As expected, reversing the A-P ori-
entation of the otic cup alone resulted in a high occurrence of
otocysts with the axial plan of the host (Fig. 1 C, D, and G and
Fig. S1A). However, a small percentage of transplants had either
a posterior duplication of the NSD (double anterior) (Fig. S1 F
and I) or a single posterior NSD, suggestive of an A-P inversion
We hypothesized that A-P polarity inversion was due to an
unintended transfer of the donor’s A-P inductive signal into the
host along with responding otic tissue. Because changing the A-P
axis of the hindbrain has no apparent effect on A-P patterning of
the inner ear (4), we modified our transplantation protocol to
include ectoderm and underlying mesoderm adjacent to the otic
cup (Fig. S1C). This modification increased the occurrence of
A-P inversion (Fig. 1G and Fig. S1 E and H). Similarly, an in-
creased occurrence of A-P inversion was obtained when ecto-
derm but not mesoderm was included in the otic cup transplant
(Fig. 1 B and E–G), indicating that an activity within the periotic
ectoderm influences A-P patterning of the inner ear.
To identify this activity, we sought candidate genes that are
signaling has been implicated in patterning of the hindbrain and
other embryonic structures (5–7). The location of the otic placode
within a gap between domains of retinaldehyde dehydrogenase2
(Raldh2), the earliest and most widely expressed gene encoding
a RA-synthesizing enzyme, and Cyp26 genes (Cyp26A1, Cyp26B1,
and Cyp26C1), which encode P450-associated RA-catabolizing
specifying the otic A-P axis (Fig. 2 A, B, D, and E). Our expres-
sion analyses confirmed previous reports of Raldh2 and Cyp26s
in tissues adjacent to the otic epithelium and showed Cyp26C1
to be expressed in rostral but not caudal periotic ectoderm
(Fig. 2C) (8, 9).
Responsiveness of the Otic Epithelium to RA Changes over Time. To
determine whether otic tissue responds to RA, we used the
transgenic mouse strain RARE-lacZ—in which lacZ is driven by
a RA responsive element (10)—to assay for reporter expression
within the otic epithelium. At embryonic day (E) 8.25, the an-
terior border of β-gal activity lies at the border of rhombomeres 4
and 5, rostral to the location of the otic placode (Fig. 3 A and B;
ref. 11). One-half day later, β-gal staining is detectable only in
the posterior half of the otic cup (Fig. 3 C–F), and by E9.5, β-gal
is absent from the otocyst (Fig. 3 G and H). This gradual with-
drawal of RA responsiveness, first from anterior otic tissue and
then from posterior otic tissue, is a likely consequence of cau-
dally shifting boundaries of Raldh2 and Cyp26 expression sur-
rounding the ear (12).
Author contributions: J.B. and D.K.W. designed research; J.B., S.R., K.-A.K., S.K.K., and
D.K.W. performed research; U.C.D. contributed new reagents/analytic tools; J.B., S.R.,
K.-A.K., S.K.K., and D.K.W. analyzed data; and J.B., S.R., U.C.D., and D.K.W. wrote the
The authors declare no conflict of interest.
*This Direct Submission article had a prearranged editor.
1To whom correspondence may be addressed. E-mail: firstname.lastname@example.org or email@example.com.
This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.
| January 4, 2011
| vol. 108
| no. 1
RA Confers Posterior Identity to the Inner Ear. Administering RA to
timed-pregnant RARE-lacZ mice at E7.75 (before otic placode
formation) induced widespread lacZ activity throughout the
embryo, including the entire otocyst, within 1 d (10). Similar
administrations of RA to wild-type mice at E7.75, E8.25, and
E8.5 down-regulated anterior expression of Lunatic fringe (Lfng)
and NeuroD1 in the otocyst (Fig. 4 A, B, D, E, G, and H) and
caused ectopic anterior expression of Tbx1, which is normally
expressed selectively in the posterior otic region (Fig. 4 C, F, and
I). RA administered at the later gestational times caused less
severe dysmorphology (Fig. 4 G–I) and fewer ears with altered
gene expression patterns (Table S1). These results support the
idea of a temporal window during which the ear rudiment is most
sensitive to RA.
We next assayed for posteriorizing activity of RA on the ear
rudiment of chicken embryos by implanting RA-soaked beads
into the mesenchyme anterior to the otic cup at E1.5. An ante-
rior source of exogenous RA reduced or abolished Lfng and
NeuroD1 expression in the otocyst (Fig. 4 J, K, M, and N) and
induced ectopic expression of posterior otic genes Tbx1 (Fig. 4 L
and O) and SOHo1 (n = 11) in the anterior otocyst. Implanting
an RA-soaked bead posterior to the otic cup yielded similar
results, indicating that otic patterning is highly sensitive to dif-
ferences in effective RA concentration at a distance from the RA
source. Interestingly, posterior implantation of RA-soaked beads
12–15 h later in development (E2) did not cause an expansion of
the Tbx1 expression domain, once again indicating that the
competence of otic tissue to respond to RA is developmentally
regulated. These results in both mouse and chicken strongly
suggest that the changing patterns of RA responsiveness we
characterized in the RARE-lacZ mouse (Fig. 3) reflect func-
tionally relevant events in ear development.
RA bead implantations also caused posteriorization of the
hindbrain, indicated by anterior expansion of genes normally
expressed in the posterior hindbrain and down-regulation of
whether the RA-induced inner ear phenotypes are indirect phe-
nomena resulting from RA-induced changes in the hindbrain
(Fig. S2 A–F; ref. 13), we surgically altered rhombomere patterns
to mimic the molecular changes brought about by RA bead im-
plantation. This alteration was accomplished by replacing the
segment of hindbrain adjacent to the ear (r4–r6) with a block of
caudal neural tissues containing r7 and spinal cord (Fig. S3). Such
including adjacent ectoderm (blue arrowheads). (C and D) Transplanting an otic cup alone results in normal patterning of anterior inner ear neurosensory
markers, Lfng and NeuroD1 (arrows). (E and F) Transplanting an otic cup plus adjacent ectoderm causes an inversion of Lfng and NeuroD1 patterning (red
arrows). (G) Percentages of samples from various surgical conditions with a normal or inverted A-P axis, or duplicated anterior domains. aa, anterior ampulla;
asc, anterior semicircular canal; bp, basilar papilla; cd, cochlear duct; ed, endolymphatic duct; es, endolymphatic sac; la, lateral ampulla; lsc, lateral semicircular
canal; pa, posterior ampulla; psc, posterior semicircular canal; s, saccule; u, utricle.
(A) Paint-filled inner ears from mouse and chicken embryos. (B) The replacement of a host’s right-sided otic cup with a donor’s left-sided otic cup,
Cyp26C1 is expressed in the ectoderm anterior to the otic placode/cup region
(A–C, arrowheads), whereas Raldh2 is expressed in the mesodermal tissues
caudal to the otic tissues such as somites and the lateral mesoderm (D–F,
arrows). Brackets indicate the location of the otic placode/cup. Weak
Cyp26C1 expression in the otic region in B is associated with the mesoderm
beneath the otic cup.
Expression patterns of Raldh2 and Cyp26C1 in chicken embryos.
| www.pnas.org/cgi/doi/10.1073/pnas.1010547108 Bok et al.
hindbrain operations failed to generate gene expression changes
in ears similar to those with RA implantations (Fig. S3), sug-
gesting that the effects of localized exogenous RA on inner ear
development are independent of hindbrain-inner ear signaling.
If exogenous RA posteriorizes otic tissue, then blocking en-
dogenous RA should have the opposite effect: namely, to ante-
riorize posterior otic tissue. Endogenous RA signaling can be
blocked by using citral, an inhibitor of Raldh activity (14). Imp-
lanting a citral-soaked bead posterior to the otic cup down-
regulated expression of posterior otic genes Tbx1 and SOHo1
(Fig. 5 A–E, brackets). In contrast, expression domains of the
anterior genes Lfng and NeuroD1 were duplicated or expanded
into the posterior otic region (Fig. 5 F–K, red arrows). In-
terestingly, citral-bead implantations did not cause observable
not mediated by the hindbrain.
The longer-term effects of early exogenous RA on inner ear
development were determined at E7, when all gross structures
are normally distinguishable (ref. 16 and Fig. 6A). Embryos in
which an RA bead was implanted anterior to the otic cup had
inner ears resembling a mirror image duplication of two poste-
rior halves, each half consisting of a posterior-like canal and
ampulla (Fig. 6 B and C). Anterior structures such as the anterior
ampulla/canal, utricle, and saccule were missing (Fig. 6A, red
labels). The lateral ampulla and canal, both of which are con-
sidered anterior structures (17), were also absent. The cochlear
duct, composed of both A-P and medial-lateral components (1),
was misshaped (Fig. 6 B and C). Consistent with these results,
similarly treated embryos left to develop until E9 and immu-
nostained for hair cells had inner ears with only two sensory
patches resembling posterior cristae (Fig. 6 E and F).
RA Induction of Otic Tbx1 Transcription Occurs Rapidly and
Independently of Protein Synthesis. The T-box transcription fac-
tor gene Tbx1 is implicated in the establishment of posterior otic
identity (18, 19), making it a likely mediator of RA’s effect on
otic tissue. We therefore tested whether RA directly activates
Tbx1 transcription in otic epithelial cells, which would require
that the response to exogenous RA be rapid and independent of
protein synthesis. Indeed, Tbx1 was up-regulated within 3 h of
RA bead implantation (Fig. 7 B and C). Pretreating chicken
embryos to inhibit protein synthesis (20) did not block this rapid
RA-induced up-regulation of Tbx1 in otic tissue (Fig. 7 E and F).
In contrast, RA-induced down-regulation of mesodermal Tbx1 in
these same embryos, which has been shown to require protein
synthesis (21), was blocked (Fig. 6 C and F, bracket), verifying
that protein synthesis was inhibited in our experimental system.
These results suggest that exposure to RA posteriorizes the ru-
dimentary ear at least, in part, through direct transcriptional
activation of Tbx1.
Low Concentrations of RA Are Required for Proper Anterior Gene
Expression. Our finding that some critical concentration of RA is
necessary and sufficient for posteriorizing the otic epithelium is
consistent with gene expression analyses showing a close prox-
imity of the otic posterior pole to mesodermal Raldh2 expression
(Fig. 2). However, developmental analyses of RARE-lacZ
staining revealed an early, albeit brief, responsiveness of the
anterior otic placode to endogenous RA (Fig. 3A). Furthermore,
the inversion of A-P polarity achieved by rotating otic cup plus
surrounding periotic ectoderm (Fig. 1 B and G) could be due to
a posterior translocation of the rostral Cyp26C1-positive ecto-
derm, which might reduce the effective local RA concentration
to a level suitable for stabilizing the anterior neural fate. We
therefore asked whether a low concentration of RA (relative to
that present posteriorly) promotes anterior otic identity. Pre-
sentation of low RA concentrations to the anterior otic epithe-
lium during normal development could be due to distance from
the mesodermal Raldh2 source, proximity to the catabolizing
activity of rostral Cyp26 gene products, or both. The activity of
a catabolic “sink” may be of particular importance in controlling
RA activity for ear development, given that RA synthesis un-
related to known sites of Raldh1-3 expression has been reported
in the neural tube anterior to the otic placode (22). We therefore
sought to reduce the effective concentration of endogenous RA
near the anterior otic cup by rostral implantation of a citral bead.
This implantation resulted in a near complete loss of the anterior
neurosensory marker Lfng and down-regulation of NeuroD1
(Fig. 5 L and M), indicating that some effective concentration of
RA activates or potentiates gene expression associated with
anterior otic identity.
Retinoic Acid Specifies the A-P Axis of the Inner Ear. In invertebrates,
compartments and boundaries are thought to drive pattern for-
mation. Cells within the embryo and primordial larval structures
such as the wing imaginal disk of Drosophila are organized into
compartments based on positional information in the form of
morphogen gradients. Each compartment is then stabilized by the
H). Arrow in A indicates the anterior border of β-gal staining at E8.25, which is comparable to the r4/r5 boundary indicated in B (arrow) as assessed by Hoxb1
expression. Arrows in C, E, and G indicate the anterior-most extent of β-gal staining, which lies at the r5/r6 boundary at E8.75 (C Inset and D) and posterior to
r6 by E9.5 (G); arrowheads in C, D, and F indicate the anterior-most extent of otic epithelial β-gal staining. Otocyst: E Inset, dashed line, and G, bracket.
β-Gal histochemical staining in RARE-lacZ embryos. (A and C–H) β-Gal histochemical staining at E8.25 (A), E8.75 (C and D), E9 (E and F), and E9.5 (G and
Bok et al.PNAS
| January 4, 2011
| vol. 108
| no. 1
activities of its intrinsic selector genes, which instruct cells as to
their fate and how to interact with cells in adjacent compartments
(23). A similar hypothesis of compartments and boundaries has
been proposed for pattern formation of the inner ear (24).
In the early 1900s, mirror image “twinned” ears of either
double anterior or double posterior identity were described as
resulting from surgical rotations of the presumptive ear ecto-
derm in salamanders (25). These mirror image duplications
suggest that the inner ear rudiment is at first equipotential along
the A-P axis and later compartmentalized about its A-P midline.
In recent years, similar mirror image duplications of inner ears in
zebrafish and frogs have also been reported from perturbing
hedgehog (hh) signaling (26, 27). Paradoxically, hedgehog sig-
naling does not appear to be a primary determinant for A-P
patterning of the inner ear in amniotes, even though it is es-
sential for the dorsal-ventral (D-V) patterning (4, 28, 29). Al-
though it remains unclear how a continuous source of hedgehog
emanating from the ventral midline can impart A-P character-
istics to inner ears of anamniotes, more recent data indicate that
hh is also required for D-V patterning of the zebrafish inner ear,
which underscores the similarities of inner ear patterning among
Our use of a localized source of exogenous RA to elicit
a double posterior ear strongly supports the notion that RA is
a key morphogen for patterning A-P compartments of the inner
ear in amniotes. Perturbing RA signaling during a critical period
of A-P specification affected A-P identity of the inner ear.
Dynamic RA Signaling May Pattern Multiple Cranial Structures in
Parallel. The “source and sink” configuration of RA synthesis
(caudal mesoderm) and RA degradation (rostrally in the neural
tube and ectoderm) is an excellent model for explaining how
signals that establish anterior and posterior compartments of the
inner ear are generated (Fig. 7G). A critical feature of this model
is its dynamism, with both synthetic and catabolic activity shifting
caudally along the early embryo (9, 12). We describe here two
results to suggest that A-P otic compartmentalization is effected
within a limited time window by this dynamic process. First, we
have used the RARE-lacZ reporter mouse to show a devel-
opmentally regulated withdrawal of RA responsiveness from the
anterior and later from posterior otic epithelium. Second, we
find in both chicken and mouse that the potency of exogenous
RA to alter otic gene expression diminishes with advancing
gestational age. Similar RA signaling dynamics are proposed to
M) to the otic cup. (A) Posterior citral-bead (red) implantation diagram. (B–
E) Posterior citral-bead implantation causes down-regulation of Tbx1 (D,
bracket; n = 5/10) and SOHo1 (E, bracket; n = 6), compared with controls (B
and C). No detectable change is seen in Tbx1 expression in the branchial
mesoderm (arrowheads). (F–K) Posterior citral-bead implantation causes
ectopic expression of Lfng (I; n = 5/6) and NeuroD1 (J and K; n = 11/18) in the
posterior otocyst (red arrows). (L and M) Anterior citral-bead implantation
causes down-regulation of Lfng (L, arrow; n = 16/18) and NeuroD1 (M, ar-
row; n = 14/17).
Effects of implanting a citral bead posterior (A–K) or anterior (L and
ministering RA to pregnant mice at E7.75 affects the size of the otocyst,
down-regulates anterior neurosensory markers Lfng (D, asterisk; n = 5) and
NeuroD1 (E, asterisk; n = 6), and up-regulates Tbx1 anteriorly (F, bracket; n =
3), compared with controls at E9.5 (A–C). Similar gene expression changes
are observed less frequently and with less severe dysmorphology when RA is
administered at E8.25 (G–I). (J–O) Implantation of an RA bead in mesoderm
anterior to the otic cup in chicken causes down-regulation of Lfng (M, as-
terisk; n = 16) and NeuroD1 (N, asterisk; n = 15), and up-regulation of Tbx1
anteriorly (O, bracket; n = 6) in comparison with controls wherein beads are
soaked with DMSO alone (J–L). Tbx1 expression in the branchial mesoderm is
down-regulated in response to exogenous RA (L and O, asterisks).
RA signaling confers posterior identity to the inner ear. (A–F) Ad-
| www.pnas.org/cgi/doi/10.1073/pnas.1010547108Bok et al.
specify the identity of anterior rhombomeres (12). More recent
data in zebrafish suggest that the effects of RA signaling along
the A-P axis of the body may not depend on simple diffusion of
RA molecules from a posterior source alone, but rather on the
complex regulation of genes involved in RA metabolism, in
particular the Cyp26s (31). Regardless of the mechanisms in-
volved in regulating RA signaling, our demonstration of the di-
rect effects of RA on Tbx1 expression in the otic epithelium and
analyses of surgical hindbrain alterations (Fig. S3; ref. 4) strongly
suggest that—at least in chicken—RA signaling patterns the
hindbrain and inner ear independently of each other. It is pos-
sible that a wave of RA signaling coordinates the formation of
multiple organs at these axial levels, and other structures such as
the branchial arches and epibranchial placodes may rely on this
dynamic regulation (32).
Perturbation of RA signaling at stages earlier than those used
in the experiments reported here, as in vitamin-A–deficient
quails and Raldh2 knockout mice, affects the axial location of the
otic placode (33, 34). These studies, taken together with results
presented here, suggest that a dynamic source and sink config-
uration of RA regulation first specifies where the otic placode
will be positioned along an animal’s A-P axis and later specifies
the A-P axis of the ear rudiment itself.
Molecular Mechanisms by Which RA Patterns the Otic A-P Axis. The
stepwise and developmentally regulated withdrawal of RA re-
sponsiveness we have observed indicates that direct effects of
RA on otic transcription are maintained for a longer period in
the posterior than in the anterior region, a difference that could
underlie the divergence of these otic regions into functionally
distinct inner ear structures. Thus, a low RA concentration or
short exposure should induce an anterior, neurosensory fate,
whereas a high concentration or longer exposure of RA induces
a posterior, largely nonsensory fate. Neurosensory genes directly
regulated by RA in the anterior otic region remain to be de-
termined. Recent data suggest that the anterior neurogenic fate
in chicken depends on Fgf8 and Sox3 (35).
We have provided evidence for a direct effect of RA signaling
on transcription of Tbx1, a gene that has been implicated in
promoting posterior otic identity. In mice, lack of Tbx1 causes
a posterior expansion of the anterior neurosensory domain and
duplication of the CVG, whereas transgenic overexpression of
human TBX1 results in a neurogenic domain and ganglion of
reduced size (19). Results presented here indicate that at least
some of RA’s effects on A-P patterning are mediated by the
activities of Tbx1.
In summary, RA, which is a well-known morphogen for
somitogenesis, heart morphogenesis, and branchial arch pat-
terning (7, 32, 36, 37), is also an essential determinant of A-P
patterning for the amniote inner ear. The detailed mechanism by
which RA confers A-P identity and promotes diverse otic fates at
different concentrations/exposure durations demands further
investigation and may prove to be of value to emerging techni-
ques involving the use of pluripotent stem cells as a therapeutic
approach to alleviate sensorineural hearing loss (38, 39).
Materials and Methods
Microsurgical Manipulations of Chicken Embryos. Fertilized eggs (CBT Farms)
were incubated in a humidified chamber at 37 °C. Chicken embryos between
E1.5 and E2 [equivalent to 11–22 somite stages (ss) or Hamburger Hamilton
stage 11–14 (HH 11–14)] were used (40).
Otic tissue transplantation. Otic tissue transplantation procedures were per-
formed as described with minor modifications (4). A right otic cup or otic cup
with adjacent mesoderm and/or ectoderm (Fig. 1B and Fig. S1 A–C) from
a host embryo was replaced with comparable left otic tissues of an age-
matched donor. The otic cup was rotated such that the AP axis was reversed
(A–C) Ectopic Tbx1 transcription in the anterior otic cup is observed within 3 h
(B, arrowheads; n = 9/9) and 6 h (C, arrowheads; n = 12/12) of RA bead im-
plantation. Tbx1 expression is down-regulated in mesoderm at 6 h after RA
bead implantation (C, bracket). (D–F) Ectopic Tbx1 transcripts in the anterior
otic cup within 3 h (E, arrowheads; n = 12/15) and 6 h (F, arrowheads; n = 17/
19)of RAbead implantation inembryos pretreated with the protein synthesis
blocker cycloheximide. Cycloheximide blocks the protein synthesis-de-
pendent down-regulation of mesodermal Tbx1 by RA (n = 17/19; compare
the source and sink configuration of RA synthesis and degradation enzymes,
which is a key component in establishing the A-P axis of the inner ear.
RA induces otic Tbx1 transcription in the presence of cycloheximide.
to develop in RA-treated ears, and only two posterior canal-like structures
are evident (B, n = 5/9; C, n = 4/9). Box in A highlights the normal posterior
canal’s more ventral point of insertion into the common crus relative to the
anterior canal’s insertion and is comparable to regions highlighted by arrows
in B and C. (D) Whole-mount anti-HCA staining of chicken inner ear at E9
showing locations within the membranous labyrinth of the auditory sensory
organ, the basilar papilla (bp) and six vestibular sensory organs: anterior
crista (ac), lagena (lg), lateral crista (lc), utriculi (mu), sacculi (ms), and pos-
terior crista (pc). (E and F) RA-treated chicken ears lack all sensory organs
except for two cristae, each associated with a posterior-like canal (n = 5).
Paint-filled chicken inner ears. Anterior structures (A, red labels) fail
Bok et al. PNAS
| January 4, 2011
| vol. 108
| no. 1
relative to the host, but the other axes (D-V and medio-lateral) were un- Download full-text
changed. Before transplantation, 0.05% CM-DiI (Molecular Probes) in 300
mM sucrose solution was injected into the anterior region of the otic cup of
the donor for orientation and tracking. Only embryos with appropriately
transplanted tissues were used for further analyses.
Bead implantation and cycloheximide pretreatment. Bead implantation was car-
ried out as described with minor modifications (41, 42). For delivery of RA
(Sigma), AG1-X2 beads (Bio-Rad) were soaked in 0.5 mg/mL RA. For delivery
of Citral (an inhibitor of retinaldehyde dehydrogenases; Sigma), SM2 beads
(Bio-Rad) were soaked in 0.4 g/mL Citral solution diluted in DMSO. Anterior
bead implantations were conducted by making a slit in the ectoderm rostral
to the right otic cup, at the level of rhombomere 3/4 boundary, whereas
posterior bead implantations were performed by making an incision in the
ectoderm between the posterior otic cup and the first somite. A single bead
soaked with specific reagents was pressed down into the slit by using the tip
of a forcep, and implanted embryos were further incubated and harvested
for whole mount in situ hybridization at E2.5–E3, paint fill analysis at E7, or
anti-HCA (hair cell antigen) staining at E9 (17).
To inhibit protein synthesis, cycloheximide solution (2 mg/50 mL Tyrode’s
solution) was applied onto the chorioallantoic membrane of chicken em-
bryos 2 h before bead implantations. Control embryos received 50 mL of
Tyrode’s solution alone.
RARE-LacZ Mice and RA Administration. RARE-lacZ mouse strain was gener-
ated by J. Rossant (10). RA solution emulsified in corn oil was administered
to mice by gavage (50 mg/kg of body weight) between E7.75 and E8.5.
Embryos were harvested at E9.5 and analyzed by whole-mount in situ hy-
bridization or β-galactosidase histochemical staining. All animal procedures
were approved and conducted according to the National Institutes of Health
Animal Use and Care Committee guidelines.
Whole-Mount in Situ Hybridization and β-Galactosidase Staining. Whole-mount
in situ hybridization and β-galactosidase histochemical staining were carried
out as described (10, 43). Details of probes used are available upon request.
ACKNOWLEDGMENTS. We thank Drs. Robert Morell and Andrew Griffith for
critical reading of the manuscript. This work was supported by the National
Institute on Deafness and Other Communication Disorders Intramural Pro-
gram, faculty research Grant 6-2008-0102 from Yonsei University College of
Medicine (to J.B.), and Basic Science Research Program Grant 2010-0015354
through the National Research Foundation of Korea (to J.B.).
1. Fekete DM, Wu DK (2002) Revisiting cell fate specification in the inner ear. Curr Opin
2. Bok J, Chang W, Wu DK (2007) Patterning and morphogenesis of the vertebrate inner
ear. Int J Dev Biol 51:521–533.
3. Whitfield TT, Hammond KL (2007) Axial patterning in the developing vertebrate
inner ear. Int J Dev Biol 51:507–520.
4. Bok J, Bronner-Fraser M, Wu DK (2005) Role of the hindbrain in dorsoventral but not
anteroposterior axial specification of the inner ear. Development 132:2115–2124.
5. Gavalas A (2002) ArRAnging the hindbrain. Trends Neurosci 25:61–64.
6. Hochgreb T, et al. (2003) A caudorostral wave of RALDH2 conveys anteroposterior
information to the cardiac field. Development 130:5363–5374.
7. Vermot J, Pourquié O (2005) Retinoic acid coordinates somitogenesis and left-right
patterning in vertebrate embryos. Nature 435:215–220.
8. Blentic A, Gale E, Maden M (2003) Retinoic acid signalling centres in the avian embryo
identified by sites of expression of synthesising and catabolising enzymes. Dev Dyn
9. Reijntjes S, Gale E, Maden M (2004) Generating gradients of retinoic acid in the chick
embryo: Cyp26C1 expression and a comparative analysis of the Cyp26 enzymes. Dev
10. Rossant J, Zirngibl R, Cado D, Shago M, Giguère V (1991) Expression of a retinoic acid
response element-hsplacZ transgene defines specific domains of transcriptional
activity during mouse embryogenesis. Genes Dev 5:1333–1344.
11. Uehara M, et al. (2007) CYP26A1 and CYP26C1 cooperatively regulate anterior-
posterior patterning of the developing brain and the production of migratory cranial
neural crest cells in the mouse. Dev Biol 302:399–411.
12. Sirbu IO, Gresh L, Barra J, Duester G (2005) Shifting boundaries of retinoic acid activity
control hindbrain segmental gene expression. Development 132:2611–2622.
13. Sundin OH, Eichele G (1990) A homeo domain protein reveals the metameric nature
of the developing chick hindbrain. Genes Dev 4:1267–1276.
14. Kikonyogo A, Abriola DP, Dryjanski M, Pietruszko R (1999) Mechanism of inhibition of
aldehyde dehydrogenase by citral, a retinoid antagonist. Eur J Biochem 262:704–712.
15. Dupé V, Lumsden A (2001) Hindbrain patterning involves graded responses to retinoic
acid signalling. Development 128:2199–2208.
16. Bissonnette JP, Fekete DM (1996) Standard atlas of the gross anatomy of the
developing inner ear of the chicken. J Comp Neurol 368:620–630.
17. Wu DK, Nunes FD, Choo D (1998) Axial specification for sensory organs versus non-
sensory structures of the chicken inner ear. Development 125:11–20.
18. Funke B, et al. (2001) Mice overexpressing genes from the 22q11 region deleted in
velo-cardio-facial syndrome/DiGeorge syndrome have middle and inner ear defects.
Hum Mol Genet 10:2549–2556.
19. Raft S, Nowotschin S, Liao J, Morrow BE (2004) Suppression of neural fate and control
of inner ear morphogenesis by Tbx1. Development 131:1801–1812.
20. Oppenheim RW, Prevette D, Tytell M, Homma S (1990) Naturally occurring and
induced neuronal death in the chick embryo in vivo requires protein and RNA
synthesis: Evidence for the role of cell death genes. Dev Biol 138:104–113.
21. Roberts C, Ivins SM, James CT, Scambler PJ (2005) Retinoic acid down-regulates Tbx1
expression in vivo and in vitro. Dev Dyn 232:928–938.
22. Mic FA, Haselbeck RJ, Cuenca AE, Duester G (2002) Novel retinoic acid generating
activities in the neural tube and heart identified by conditional rescue of Raldh2 null
mutant mice. Development 129:2271–2282.
23. Lawrence PA, Struhl G (1996) Morphogens, compartments, and pattern: Lessons from
drosophila? Cell 85:951–961.
24. Fekete DM (1996) Cell fate specification in the inner ear. Curr Opin Neurobiol 6:
25. Harrison RG (1936) Relations of symmetry in the developing ear of amblystoma
punctatum. Proc Natl Acad Sci USA 22:238–247.
26. Hammond KL, Loynes HE, Folarin AA, Smith J, Whitfield TT (2003) Hedgehog
signalling is required for correct anteroposterior patterning of the zebrafish otic
vesicle. Development 130:1403–1417.
27. Waldman EH, Castillo A, Collazo A (2007) Ablation studies on the developing inner
ear reveal a propensity for mirror duplications. Dev Dyn 236:1237–1248.
28. Bok J, et al. (2007) Opposing gradients of Gli repressor and activators mediate Shh
signaling along the dorsoventral axis of the inner ear. Development 134:1713–1722.
29. Riccomagno MM, Martinu L, Mulheisen M, Wu DK, Epstein DJ (2002) Specification of
the mammalian cochlea is dependent on Sonic hedgehog. Genes Dev 16:2365–2378.
30. Hammond KL, van Eeden FJ, Whitfield TT (2010) Repression of Hedgehog signalling is
required for the acquisition of dorsolateral cell fates in the zebrafish otic vesicle.
31. White RJ, Schilling TF (2008) How degrading: Cyp26s in hindbrain development. Dev
32. Mark M, Ghyselinck NB, Chambon P (2004) Retinoic acid signalling in the devel-
opment of branchial arches. Curr Opin Genet Dev 14:591–598.
33. Kil SH, et al. (2005) Distinct roles for hindbrain and paraxial mesoderm in the
induction and patterning of the inner ear revealed by a study of vitamin-A-deficient
quail. Dev Biol 285:252–271.
34. Niederreither K, Vermot J, Schuhbaur B, Chambon P, Dollé P (2000) Retinoic acid
synthesis and hindbrain patterning in the mouse embryo. Development 127:75–85.
35. Abelló G, et al. (2010) Independent regulation of Sox3 and Lmx1b by FGF and BMP
signaling influences the neurogenic and non-neurogenic domains in the chick otic
placode. Dev Biol 339:166–178.
36. Dubrulle J, Pourquié O (2004) fgf8 mRNA decay establishes a gradient that couples
axial elongation to patterning in the vertebrate embryo. Nature 427:419–422.
37. Xavier-Neto J, et al. (2001) Retinoid signaling and cardiac anteroposterior segmen-
tation. Genesis 31:97–104.
38. Li H, Roblin G, Liu H, Heller S (2003) Generation of hair cells by stepwise
differentiation of embryonic stem cells. Proc Natl Acad Sci USA 100:13495–13500.
39. Oshima K, et al. (2010) Mechanosensitive hair cell-like cells from embryonic and
induced pluripotent stem cells. Cell 141:704–716.
40. Hamburger V, Hamilton HL (1951) A series of normal stages in the development of
the chick embryo. J Morphol 88:173–192.
41. Chang W, Brigande JV, Fekete DM, Wu DK (2004) The development of semicircular
canals in the inner ear: Role of FGFs in sensory cristae. Development 131:4201–4211.
42. Song Y, Hui JN, Fu KK, Richman JM (2004) Control of retinoic acid synthesis and FGF
expression in the nasal pit is required to pattern the craniofacial skeleton. Dev Biol
43. Wu DK, Oh SH (1996) Sensory organ generation in the chick inner ear. J Neurosci 16:
| www.pnas.org/cgi/doi/10.1073/pnas.1010547108Bok et al.