FGF Signaling Regulates the Number of Posterior Taste
Papillae by Controlling Progenitor Field Size
Camille I. Petersen1., Andrew H. Jheon1., Pasha Mostowfi1, Cyril Charles1,2, Saunders Ching1, Shoba
Thirumangalathu3, Linda A. Barlow3, Ophir D. Klein1,4*
1Department of Orofacial Sciences and Program in Craniofacial and Mesenchymal Biology, University of California San Francisco, San Francisco, California, United States of
America, 2Team ‘‘Evo-Devo of Vertebrate Dentition,’’ Institut de Ge ´nomique Fonctionnelle de Lyon, Universite ´ de Lyon, CNRS, Ecole Normale Supe ´rieure de Lyon, Lyon,
France, 3Department of Cell and Developmental Biology and the Rocky Mountain Taste and Smell Center, University of Colorado School of Medicine, Denver, Colorado,
United States of America, 4Department of Pediatrics and Institute for Human Genetics, University of California San Francisco, San Francisco, California, United States of
The sense of taste is fundamental to our ability to ingest nutritious substances and to detect and avoid potentially toxic
ones. Sensory taste buds are housed in papillae that develop from epithelial placodes. Three distinct types of gustatory
papillae reside on the rodent tongue: small fungiform papillae are found in the anterior tongue, whereas the posterior
tongue contains the larger foliate papillae and a single midline circumvallate papilla (CVP). Despite the great variation in the
number of CVPs in mammals, its importance in taste function, and its status as the largest of the taste papillae, very little is
known about the development of this structure. Here, we report that a balance between Sprouty (Spry) genes and Fgf10,
which respectively antagonize and activate receptor tyrosine kinase (RTK) signaling, regulates the number of CVPs. Deletion
of Spry2 alone resulted in duplication of the CVP as a result of an increase in the size of the placode progenitor field, and
Spry12/2;Spry22/2embryos had multiple CVPs, demonstrating the redundancy of Sprouty genes in regulating the
progenitor field size. By contrast, deletion of Fgf10 led to absence of the CVP, identifying FGF10 as the first inductive,
mesenchyme-derived factor for taste papillae. Our results provide the first demonstration of the role of epithelial-
mesenchymal FGF signaling in taste papilla development, indicate that regulation of the progenitor field size by FGF
signaling is a critical determinant of papilla number, and suggest that the great variation in CVP number among mammalian
species may be linked to levels of signaling by the FGF pathway.
Citation: Petersen CI, Jheon AH, Mostowfi P, Charles C, Ching S, et al. (2011) FGF Signaling Regulates the Number of Posterior Taste Papillae by Controlling
Progenitor Field Size. PLoS Genet 7(6): e1002098. doi:10.1371/journal.pgen.1002098
Editor: Irma Thesleff, University of Helsinki, Finland
Received December 17, 2010; Accepted April 8, 2011; Published June 2, 2011
Copyright: ? 2011 Petersen et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was funded by NIH R21-DC011108 and NIH DP2-OD007191 to ODK. The funders had no role in study design, data collection and analysis,
decision to publish, or preparation of the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
* E-mail: firstname.lastname@example.org
. These authors contributed equally to this work.
Taste sensory capability is mediated by aggregates of receptor
cells, called taste buds, which reside within the oral and
pharyngeal cavities. The majority of taste buds in mammals
reside on the tongue in epithelial-mesenchymal specializations
termed gustatory papillae. In the rodent tongue, the smaller
fungiform papillae, each of which possesses a single taste bud, are
found in a distributed array on the anterior tongue. By contrast,
the larger bilateral foliate papillae and a single midline
circumvallate papilla (CVP) each house hundreds of buds and
reside on the posterior tongue (Figure 1A). Recently, there has
been increasing recognition that anterior fungiform taste buds
differ from those of the posterior CVP in terms of both gene
expression and taste function [1–4].
In recent years, significant progress has been made in defining
the molecular regulation of fungiform development. Fungiform
papillae initially form as placodes that subsequently undergo
epithelial morphogenesis and acquire a mesenchymal core  in a
process that is similar to morphogenesis of other vertebrate
epithelial specializations, such as hair, teeth, and mammary glands
[6,7]. The development of these other organs requires signaling
between epithelium and mesenchyme, suggesting that such
epithelial-mesenchymal interactions are also involved in patterning
and morphogenesis of taste placodes. However, expression of all of
the key signaling factors implicated in taste placode development –
including Sonic Hedgehog (SHH), Bone Morphogenetic Proteins
(BMPs), Epidermal Growth Factor (EGF), and WNTs – is
restricted to the epithelium [8–15]. Thus, to date, no inductive,
mesenchyme-derived factor involved in taste development has
Despite its importance in taste function and its status as the
largest of the taste papillae, very little is known about the genes
involved in the development of the CVP. Like fungiform placodes,
the CVP forms as an initial epithelial thickening that undergoes
complex morphogenesis to form a large papilla. However, it
appears that genes known to regulate fungiform development do
not function similarly in development of the CVP. For example,
inhibition of SHH results in more and larger fungiform placodes,
but has no effect on the CVP . In addition, BMP7 and its
PLoS Genetics | www.plosgenetics.org1 June 2011 | Volume 7 | Issue 6 | e1002098
antagonist follistatin have significant functions in fungiform
development, but the CVP appears unaffected by inactivation of
either gene . These differences may be ascribed to the distinct
embryonic origins of the anterior tongue, which is thought to be
derived from ectoderm, whereas the posterior tongue likely has
endodermal origins .
Expression of several Fibroblast Growth Factors (FGFs) and
their receptors has previously been detected in the developing
tongue . Therefore, we hypothesized that Sprouty (Spry)
genes, which antagonize several receptor-tyrosine kinase (RTK)
signaling pathways including those triggered by FGFs [18–20],
may play a role in the development of taste papillae. Originally,
spry was identified as a regulator of tracheal branching in Drosophila
, and it was later found that three of the four mouse Sprouty
genes (Spry1, Spry2, and Spry4) are expressed during embryonic
Here, we used mouse genetic models to show that FGF signaling
is required for CVP formation. We found that Spry1 and Spry2
antagonize signaling by FGF10 to restrict the size of the progenitor
field of the circumvallate (CV) placode, such that loss of Sprouty
function results in a dramatic expansion of the CV placode as it
first forms. In adult Spry2 mutants, a striking and complete
duplication of the CVP emerges, whereas embryos lacking both
Spry1 and Spry2 have multiple CVPs. Our findings thus represent
the first example of molecular genetic regulation of taste organs in
the posterior tongue. We found that Fgf10, which is expressed
exclusively in the mesenchyme underlying the CV placode, is
required for formation of the CVP, and thus we provide the first
Figure 1. Deletion of Spry2 leads to a duplication of the CVP in the posterior tongue. (A) Cartoon showing location of gustatory papillae in
the rodent tongue. (B,C) In situ hybridization staining of Shh. Wild-type mice possess a single CVP in the posterior tongue, whereas the CVP is
duplicated in Spry22/2mice (arrowheads). (D,E) SEM images of the tongue at E14.5. Scale bar, 500 mm. (F,G) DAPI fluorescence staining shows that the
two CVPs (arrowheads) persist into adulthood in Spry22/2mice. (D’-G’) Higher magnification images of boxed areas. (D’,E’) Scale bar, 25 mm.
The sense of taste is important for an animal’s ability to
survive and thrive, because it enables discrimination
between nutritious substances and toxins. Taste buds are
housed largely on the tongue in structures called papillae;
of the three types of gustatory papillae, the circumvallate
papilla (CVP) is the largest. In rodents, a single CVP is
located in the posterior midline of the tongue housing
hundreds of taste buds, whereas in other mammals up to
dozens of CVPs can be found. However, despite the great
variation in the number of CVPs in mammals, its status as
the largest of the taste papillae, and its importance in taste
function, very little is known about its development. We
identified members of the FGF signaling pathway as
determinants of CVP number. We propose that perturba-
tions to the FGF signaling pathway may have been
involved in the dramatic differences in CVP number that
arose during mammalian evolution.
FGF Signaling Regulates CVP Development
PLoS Genetics | www.plosgenetics.org2 June 2011 | Volume 7 | Issue 6 | e1002098
example of an inductive, mesenchymal signal involved in
specifying the epithelial domain of a developing taste papilla.
Further, while FGF10 signaling drives CVP development and
Spry1 and Spry2 repress this process, fungiform taste papillae are
oppositely affected by the loss of these genes: Fgf102/2tongues
appear to possess more and larger fungiform papillae, and the loss
of Sprouty genes results in fewer fungiform papillae. Thus, these
results demonstrate that molecular mechanisms regulating devel-
opment of anterior and posterior taste organs differ considerably.
Finally, we postulate that the role of FGF signaling in defining the
size of the CV progenitor field in mice may underlie the large
variation in CVP number across mammalian taxa, and that
changes in FGF signaling during evolution may have caused
expansion of the initial progenitor field to allow formation of
multiple CVPs in some species.
Deletion of Spry2 leads to a duplication of the CVP
Wild-type mice possess a single CVP in the midline of the
posterior tongue; foliate papillae reside on the lateral tongue and
(Figure 1A, 1B). In Spry22/2mice, we observed a duplication of
the CVP in the anterior-posterior orientation at embryonic day (E)
14.5 using Shh expression as a marker (Figure 1C). By contrast,
fungiform placodes in the anterior tongue, which also express Shh,
were significantly reduced in number (6164.37 in wild-type versus
3362.02 in Spry22/2embryos; p=0.0002; n=6 embryos per
genotype). Because alterations in fungiform papilla development
had been reported in other mutants [9–15], we chose to pursue the
novel CVP duplication phenotype. The presence of two CVPs in
the embryonic tongue was confirmed using scanning electron
microscopy (SEM; Figure 1D-1E’). To test whether the duplicated
CVP persisted into adulthood, we used DAPI staining in adult
tongues and again identified two discrete papillae (Figure 1F-1G’).
Notably, both CVPs in Spry22/2mice house fully differentiated
taste buds containing the three types of taste receptor cells, and
both showed positive staining for b3-tubulin, indicating that taste
buds and the papillae are innervated (Figure S1). Taken together,
these data indicate that the duplication in Spry22/2embryos leads
to two functional adult CVPs.
To reconstruct the development of the CVP, closely staged
specimens between E11.5 to E14.5 were stained with E-
cadherin and imaged in whole-mount (Figure 2A-2N), as well
as analyzed by H&E staining of sections (Figure 2O-2X). In
wild-type embryos, CVP development was initiated shortly after
the tongue rudiment formed by fusion of the lateral lingual
swellings, and it was detected as a more compact collection of
epithelial cells at E11.5 (Figure 2A) . At E12.5, a placodal
condensation was readily observable on coronal sections
(Figure 2O). Between E13 and E13.5, the wild-type placode
underwent major morphological changes as the epithelial
trenches characteristic of the CVP began to form (Figure 2P,
2Q). The growth of the trenches was coincident with the
initiation of innervation at approximately E13 (Figure S2), and
the trenches continued to grow into the mesenchyme at E14.5
(Figure 2G, 2S) [24,25].
In contrast to the wild-type, the CV placode in Spry22/2
embryos was larger at E11.5 (compare Figure 2A, 2H). Relative to
the wild-type, the mutant placode continued to grow larger both
laterally and along the anterior-posterior axis between E11.5 and
E13.5 (Figure 2B-2E, 2I-2L). By E14, two distinct structures were
observed in Spry22/2embryos, with the anterior of the two
papillae usually appearing smaller than the posterior at E14 and
E14.5 (Figure 2M, 2N). In addition, CVPs in Spry22/2embryos
showed a raised dome shape compared to the flatter shape of the
wild-type CVP (Figure 2R, 2S, 2W, 2X).
Several possible mechanisms could account for the duplication
of the CVP at ,E14, including a timer that causes the organ to
split at a certain point after its development starts or a threshold
that causes splitting after a critical size limit is achieved. To
address this issue, we quantified the size of the CV placode and
CVP in wild-type and Spry22/2tongues between E13 to E14,
corresponding to the time points before, during and after CVP
duplication (Figure S2J). Relative to wild-type mice, the mutant
CV placode was already significantly larger at E13 and E13.5, but
when CVP duplication was observed at E14, there was a further
and dramatic increase in the total area of the papillae. Thus, our
data suggest that there is a critical threshold size for the CV
placode, beyond which the placode destabilizes, leading to CVP
The CV progenitor field is enlarged in Spry22/2mice
Sox2 is one of the earliest markers of taste placodes and is
required for taste bud development . Sox2 expression was
expanded in the placode domain in both anterior-posterior and
lateral directions in Spry22/2embryos compared to wild-type
littermates at both E11.5 (data not shown) and E12.5
(Figure 3A-3B’). Next, we analyzed the expression of three
additional factors that mark the early placode: Shh, Bmp7, and
Wnt10b (Figure 3C-3H’). Shh has previously been shown to be
expressed in the developing CVP [9,27], and in fungiform
placodes, Shh is expressed specifically by taste bud progenitors
development of fungiform papillae [15,16]. The expression
domains of all these markers were expanded in the CV placode
of Spry22/2embryos relative to wild-type (Figure 3C-3H’). We
quantified the differences in expression levels by qPCR, and
interestingly, whereas Sox2 and Bmp7 levels were strongly
upregulated in the mutants, consistent with the expanded
expression domains seen in whole mount, Shh and Wnt10b
expression levels were not dramatically different in the mutants
(Figure S3A). This is consistent with the decreased intensity of
Shh and Wnt10b staining despite an increased CV placodal
domain size (Figure 3D, 3D’, 3H, 3H’).
We next asked whether there was a difference in the number of
cells in the placode between wild-type and mutant. We quantified
the number of cells in the CV placode of Spry22/2embryos
relative to wild-type littermates at E12.5 from three-dimensional
confocal images of E-cadherin-positive stained cells (Figure 3I-3K).
We counted cells in 10 mm intervals beginning at the deepest tip
of invaginated cells using ventral views; the borders of the placode
were defined by the invagination. This analysis showed that
already by the mid-placode stage, the mutant placode had almost
twice as many cells as the wild-type. Because differences in
proliferation or cell death could account for the larger placode and
the morphogenetic abnormalities in the CVPs of Spry2 mutants,
we assayed proliferation by PCNA staining (Figure 3L-3N), and
cell death by TUNEL staining (Figure S3B-S3D). We detected no
differences between wild-type and mutant CV placode at E12.5
for either assay, suggesting that a larger number of cells was
recruited into the placode at the earliest stages of placodogenesis.
Thus, our data indicate that inactivation of Spry2 leads to a larger
progenitor field, as detected by multiple molecular markers, such
that the duplication in the Spry2 mutant results from recruitment of
more cells into the CV placode at the earliest stages of its
also been implicatedin
FGF Signaling Regulates CVP Development
PLoS Genetics | www.plosgenetics.org3June 2011 | Volume 7 | Issue 6 | e1002098
Spry1 and Spry2 are co-expressed in the CV placode, and
hypersensitivity to FGF signaling is detected in Spry2
Because Sprouty genes are often co-expressed during develop-
ment [23,29], we assayed for expression of Spry1, Spry2, and Spry4
in CV placodes at E12.5 (Figure 4A-4C); Spry3 was not analyzed
due to its lack of expression in embryonic craniofacial tissues .
We found that Spry1 and Spry2 were both expressed in the
epithelium of the CV placode at E12.5, whereas no expression was
observed for Spry4 (Figure 4A-4C).
The expression of Etv5 (Erm), which is a target of FGF signaling
, was upregulated in Spry22/2embryos (compare Figure 4D,
4E), consistent with increased FGF signaling observed with
Sprouty loss-of-function in other developmental contexts [31–
34]. To quantify expression levels, RNA was extracted from the
area containing the CV placode at E12.5 and analyzed by qPCR,
and a three-fold increase in Etv5 was observed (Figure 4F). As
expected, essentially no Spry2 expression was detected in the Spry2
Spry1 expression was dramatically
increased in Spry22/2mice (Figure 4F). This upregulation is
consistent with the known role of Sprouty genes as FGF targets
[21,23], and it suggested that increased expression of Spry1 may
serve a compensatory role in Spry22/2mice. qPCR showed that
Spry4 was expressed at very low levels, with no detectable
difference between wild-type and Spry22/2littermates (data not
Fgf10 is required for CVP development and is
antagonized by Spry2
The epithelial expression of Spry1, Spry2, and Etv5, all of which
are expressed in response to RTK activity [21,23,30], strongly
suggested the presence of a mesenchymal RTK ligand that signals
to the epithelium. To identify such a ligand, we surveyed the
expression of a number of candidate RTKs and their cognate
ligands by ISH, including Fgfr1-4, Egfr, Fgf7, Fgf8, Fgf9, and Fgf10
(data not shown). Of the ligands we examined, only Fgf10 was
expressed in the mesenchyme subjacent to the epithelium along
the midline of the posterior tongue at E11.5 and E12.5 (Figure 5A-
5B’). Two FGF receptors, Fgfr2 and Fgfr3, were also expressed in
the CV placode, whereas no expression of Fgfr1 or Fgfr4 was
detected (Figure 5C-5F).
To further test the role of FGF signaling in CVP development,
we isolated tongues from wild-type embryos at E11.5 and E12.5
and grew them in vitro either in the absence or presence of SU5402
, a potent FGF receptor inhibitor (Figure S4). In the absence
of SU5402, we found the formation of a single CVP. The
inhibition of FGF signaling by SU5402 led to the absence of the
CVP in tongues cultured at E11.5 and to either an absent or a
malformed CVP in tongues cultured at E12.5. These observations
demonstrate the requirement of FGF signaling during CVP
formation at E11.5 and to a lesser extent at E12.5.
developing CV placode, we hypothesized that the ligand
encoded by this gene may play a role in CV development,
and therefore we examined the CVP in Fgf102/2embryos.
Strikingly, the deletion of Fgf10 led to the complete absence of
the CVP (Figure 5H-5H’’), indicating that this gene plays a
critical inductive role in development of this structure. Next, we
reasoned that the combined deletion of Fgf10 and Spry2 may
balance each other and rescue CVP development, and indeed,
we found a single CVP in Fgf102/2;Spry22/2mice (Figure 5I-
5I’’). This result also implies the presence of a yet-unknown
secondary ligand that can fulfill the function of FGF10 if the
epithelium is rendered hypersensitive to RTK signaling by
inactivation of Spry2. Taken together, our data indicate that the
FGF signaling pathway is critical for proper CVP development
and that it determines the final number of CVPs.
Figure 2. Development of the CVP in wild-type and Spry22/2mice. (Inset) Cartoon depicting the planes of observation. (A–N) E-cadherin
(E-CAD) immunofluorescence staining of CV placodes and papillae in wild-type (A–G) and Spry22/2(H-N) mice. (O–X) H&E stains of coronal sections of
the single CV placode and papilla in wild-type and the proximal CV placode and papilla in Spry22/2mice. The size of the papilla was quantified
between E13 to E14 (D–F, K–M; 61023um). Arrowheads indicate developing trenches. Scale bars, 50 mm.
FGF Signaling Regulates CVP Development
PLoS Genetics | www.plosgenetics.org4 June 2011 | Volume 7 | Issue 6 | e1002098
Combined deletion of Spry1 and Spry2 leads to multiple
Based on the overlapping expression of Spry1 and Spry2 in the
CV placode and the upregulation of Spry1 in Spry22/2embryos
(Figure 4F), we hypothesized that these two family members may
be partially redundant. We therefore generated and analyzed
Spry12/2as well as Spry12/2;Spry22/2(DKO) embryos. No
difference in the number or morphology of the CVP was detected
in Spry12/2embryos (data not shown), indicating that Spry1 is not
required for CVP development if Spry2 is present. In contrast,
CVPs from DKO embryos were dramatically abnormal relative to
control Spry1+/2;Spry2+/2(DHet) littermates, whose tongues were
indistinguishable from those of wild-types both as embryos and as
adults (data not shown). At E12.5, the expression domains of Etv5
and Sox2 were expanded in the CV placodes of DKO mice relative
to DHet littermates (Figure 6A, 6B, 6F, 6G). The expanded
expression domains in the DKOs were greater than those seen in
Spry22/2mice (compare to Figure 4D, 4E and Figure 3A-3B’). At
later stages, in contrast to the single duplication seen in Spry22/2
mice, the DKO embryos had multiple CVPs (Figure 6C, 6E, 6H,
6J). The multiple CVPs in DKO embryos were smaller than the
single CVP in controls and showed pronounced raised domes
(Figure 6D, 6I) similar to Spry22/2embryos (Figure 2S, 2X).
Immunofluorescence staining of b3-tubulin revealed innervation
of the multiple CVPs (Figure S5).
Of the numerous gustatory papillae, very little is known about
the development of the CVP, despite the great variation in CVP
number in mammals, its importance in taste function, and its
status as the largest of the taste papillae. Most of the studies on
gustatory organ and taste development have focused on fungiform
papillae, and it is probably for this reason that significant genetic
abnormalities of the CVP have not yet been reported in mice.
Here, we report several findings: first, FGF signaling plays a key
role in the regulation of taste papilla development; second, FGF10
is an inductive, mesenchyme-derived factor required for CVP
development; third, the regulation of the progenitor field size by
Figure 3. The CV progenitor field at E12.5 is expanded in
Spry22/2mice. (A–H) Dorsal view of the expression of Sox2, Shh, Bmp7,
and Wnt10b in wild-type and Spry22/2littermates. Scale bar, 50 mm.
(A’–H’) Coronal sections of A–H. Scale bar, 50 mm. (I,J) Sagittal views of
the CV placodes stained with E-cadherin show an expansion of the
placode in the anterior(A)-posterior(P) orientation in Spry22/2mice.
Scale bar, 20 mm. (K) Quantification of the total number of cells in the
CV placode. (L,M) Proliferating cells were detected by PCNA staining in
coronal view. Scale bar, 20 mm. (N) The percentage of PCNA-positive
cells in the CVplacode at E12.5.* p,0.01.WT,wild-type; S22/2, Spry22/2.
Figure 4. Spry1 and Spry2 are co-expressed in the CV placode and increased FGF signaling is detected in Spry2 mutants. (A–C)
Expression of Spry1, Spry2, and Spry4 in the CV placode of wild-type mice. (D,E) Expression of Etv5 in the CV placode of wild-type and Spry22/2mice.
(F) qPCR analysis of the expression profiles of Etv5 and Sprouty genes in the posterior tongue of wild-type and Spry22/2mice at E12.5. Scale bar,
20 mm. **, p,0.001.
FGF Signaling Regulates CVP Development
PLoS Genetics | www.plosgenetics.org5June 2011 | Volume 7 | Issue 6 | e1002098
Sprouty genes is a critical determinant of CVP number. Finally,
we postulate that the great variation in CVP number among
mammalian species may be linked to alterations in the genetic
dosage of agonists and antagonists in the FGF signaling pathway.
Antagonism between mesenchymal FGF10 and epithelial
Sprouty genes controls CVP number
Antagonism of FGF signaling by Sprouty genes occurs during
development of a number of organs [31–34]. Because FGFs and
their receptors had previously been detected in the developing
tongue , we hypothesized that Sprouty genes, which modulate
several RTK signaling pathways including those triggered by
FGFs [18–20], may play a role in taste papillae development.
Whereas deletion of Spry2led to a duplication of the CVP,Fgf102/2
mice showed a complete loss of the CVP, and importantly, CVP
development was rescued in compound Spry22/2; Fgf102/2
mutants. These results confirm the specific antagonism of FGF10
signaling by SPRY2 and establish the importance of the FGF
signaling pathway in determining CVP number. Additionally, the
results with the Spry22/2;Fgf102/2mutants indicate the presence of
an unknown, secondary factor; such a factor could induce CVP
formation in lieu of FGF10 when the epithelium is hypersensitive to
RTK signaling due to the absence of Spry2.
Because of the overlapping expression profiles of Spry1 and
Spry2, and the upregulation of Spry1 in the Spry22/2CV placode,
we generated Spry12/2; Spry22/2embryos. We observed multiple
Figure 5. Fgf10 is required for CVP development and is antagonized by Spry2. (A–B’) Expression of Fgf10 in the CV placode of wild-type
mice (arrowhead). Dashed line depicts the coronal plane of section for A’ and B’. (C–F) Expression of FGF receptors (Fgfr1-4) in coronal sections. (G–I’’)
Genetic rescue of the loss of CVP phenotype in Fgf102/2mice (H) by deletion of Spry2 (I) as visualized by E-cadherin immunofluorescence staining.
(G’–I’) Higher magnification views of boxed areas. (G’’–I’’) H&E staining of coronal sections through the CVP. Scale bars, 25 mm.
Figure 6. Formation of multiple CVPs in Spry12/2;Spry22/2(DKO) mice. (A,B,F,G) Dorsal views of the expression patterns of Etv5 and Sox2 in
Spry1+/2;Spry2+/2(DHet) and DKO mice. (C,H) Multiple CVPs in the anterio-posterior direction were observed in DKO mice visualized by E-cadherin
immunofluorescence staining (arrowheads). Scale bar, 50 mm. (D,I) H&E staining of coronal sections through the CVP as indicated by dashed lines in C
and H. Scale bar, 25 mm. (E,J) Multiple, presumptive CVPs visualized by SEM are indicated by dashed circles. Scale bar, 50 mm.
FGF Signaling Regulates CVP Development
PLoS Genetics | www.plosgenetics.org6June 2011 | Volume 7 | Issue 6 | e1002098
CVPs along the A-P axis in the compound mutant embryos,
demonstrating the redundancy of Sprouty genes, for which there is
precedent in other tissues , in regulating CVP development.
Although a slightly dysplastic CVP was recently reported in
ectodysplasin mutant mice , the phenotypes in Fgf102/2,
Spry22/2, and Spry12/2;Spry22/2mice are the first genetic
abnormalities reported involving the regulation of CVP number.
Despite indications that mesenchyme-derived factors are
involved in lingual papillae development [5,37], the only
mesenchymal factor implicated in fungiform papillae formation
to date is follistatin, an inhibitor of BMP signaling . Therefore,
FGF10 is the first inductive, mesenchyme-derived factor to be
identified in taste papilla development.
Expansion of the CV progenitor field
The duplication of the CVP at E14 in Spry22/2embryos was
preceded by an increase in placode size. The CV placode in
mutant mice was significantly larger at E13 and E13.5 relative to
wild-type, and at E14 there was a dramatic increase in total CVP
size as duplication occurred (Figure S2J). Thus, there appears to be
a critical size threshold for the CV placode, beyond which the
placode destabilizes, resulting in splitting and duplication.
Recently, Munne et al.  reported that, during incisor tooth
development, the impairment of BMP4 signaling or an increase in
Activin concentration led to the destabilization of the large, single
placode and the formation of multiple incisors. Future experiments
will enable comparison of the relative roles of different signaling
pathways in maintenance of epithelial placode integrity.
The increase in placode size at E13 and E13.5 was preceded at
E11.5 and E12.5 by the expansion of the expression domains for
various genes, such as Etv5, Sox2, Shh, Bmp7, and Wnt10b. By E12.5,
the mutant CV placode already had more cells than the control, and
we did not detect any apoptosis in the CV placode or any differences
the Spry2 mutant appears to result at least in part from an increase in
the recruitment of CV placodal cells. At the earliest stages of placode
formation, this could potentially occur either through specification of
more placodal precursors or through migration of more cells into the
these possibilities, but interestingly, several reports have pointed to a
role for Sprouty genes in cell migration [39–41].
In the CV placode of Spry12/2;Spry22/2tongues, there were
further increases in the expression domains of Etv5 and Sox2 beyond
what was seen in the Spry22/2alone, demonstrating the redundancy
of Spry1 and Spry2 in this organ (Figure 6A, 6B, 6F, 6G). The
compound deletion of Spry1 and Spry2 led to an even larger placode
(Figure 6A, 6B, 6F, 6G), which resulted in more than two CVPs.
Comparision of fungiform and CV papillae
Whereas the deletion of Spry2 led to CVP duplication, it also
resulted in a marked decrease in the number of fungiform papillae
(Figure 1B, 1C). In contrast, in Fgf102/2mice, there was an
absence of the CVP, whereas the fungiform papillae appeared to
be larger and more closely spaced (Figure 4G, 4H). Thus, Fgf10
and Sprouty genes differentially affect the anterior versus posterior
taste fields. Together with the previous studies showing effects of
SHH and BMP7 on anterior but not posterior taste fields, these
results provide further evidence for important developmental
differences between these fields [9,10,12,16]. This observation is
consistent with the notion that fungiform papillae in the anterior
tongue are derived from ectoderm, whereas the CVP is likely
derived from endoderm . The diverse responses of fungiform
versus CV papillae to developmental factors have only recently
been appreciated and will require further efforts to tease apart.
Evolution of CVP number in mammals
Although most rodents, including the mouse, possess a single
midline CVP in the posterior tongue, there is great variation in
mammalian CVP number (Figure S6B, Figure S7). For example,
humans possess anywhere from 3 to 14 CVPs in an inverted V or
Y orientation . Other mammals such as the hyrax [43,44] and
hippopotamus  possess no CVP, whereas siamang, chimpan-
zee, gorilla  and lemur  possess at least two CVPs along
the midline, in addition to varying numbers of lateral CVPs
(Figure S6A). We have shown that modifications in FGF signaling
can lead to increased or decreased numbers of CVPs. Thus, we
speculate that changes in levels of signaling through this pathway
provide an attractive candidate for producing variation in CVP
number, and in particular, for generating CVPs that are multiplied
along the midline or anterior-posterior axis. It is interesting that
both Fgf10 and Etv5, an indicator of FGF signaling, were expressed
along the midline, which correlates with anterior-posterior CVP
multiplication in Spry22/2and Spry12/2;Spry22/2mice.
The effect of CVP number on taste is currently unclear.
Correlation between taste sensitivity and the number of taste buds
within the CVP has been previously reported . However,
because there is variation in the number of taste buds per CVP, as
well as in the number of taste cells within each taste bud ,
there is no clear indication of what the number of CVPs reveals
about taste preferences. Whether the variation in mammalian
CVP number provides evolutionary advantages in terms of
identification of nutritious substances and detection and avoidance
of potentially toxic ones remains to be elucidated.
In 1950, Spuhler  postulated that the variation in CVP
number observed in humans was likely due to genetic factors
involving at least 5 multiple alleles. Our studies indicate, for the
first time, that the perturbation of a single gene such as Fgf10 or
Spry2 may be sufficient to confer the vast genetic variation in
mammalian CVP number.
Materials and Methods
This study was carried out in strict accordance with the
recommendations in the Guide for the Care and Use of
Laboratory Animals of the National Institutes of Health. The
protocol was approved by the UCSF Institutional Animal Care
and Use Committee (Protocol Number: AN084146). All efforts
were made to minimize animal suffering.
Mouse lines carrying null or floxed alleles of Spry1 , Spry2
, and Fgf10  or b-actin Cre transgenes  were
maintained and genotyped as described. Spry12/2;Spry22/2mutant
embryos were generated by crossing b-actin Cre;Spry1+/2;Spry2+/2
males with Spry1flox/flox;Spry2flox/floxfemales (double mutants pro-
duced at an expected Mendelian frequency of 1:4). Tongues and
taste papillae from double heterozygous embryos (Spry1+/2;Spry2+/
2) were indistinguishable from wild-type CD-1 embryos and served
as controls. Mice were mated overnight, and the presence of a
vaginal plug indicated embryonic day (E) 0.5.
Embryonic and adult tongues were fixed overnight in 4%
paraformaldehyde at 4uC. For sections, tongues were dehydrated,
embedded in paraffin wax, and serially sectioned at 7 mm.
Histological sections were stained with haematoxylin and eosin
FGF Signaling Regulates CVP Development
PLoS Genetics | www.plosgenetics.org7June 2011 | Volume 7 | Issue 6 | e1002098
In situ hybridization (ISH)
Whole-mount ISH using digoxigenin-labeled RNA probes was
performed according to standard protocols. RNA probes were
generated from plasmids containing fragments of Shh, Spry1, Spry2,
Spry4, Fgf10, and Sox2 or from PCR-amplified fragments of
Wnt10b and Bmp7. A 10 minute 2% H2O2 treatment was
performed on tissues E14.5 and older. Stained specimens were
incubated overnight in 30% sucrose, embedded in tissue freezing
media (Triangle Biomedical Sciences, Durham, N.C.), and
cryosectioned using a Leica CM1900 at 18 mm intervals.
Whole-mount IHC was performed according to a published
protocol . Rat anti-E-cadherin (1:1000; Invitrogen Cat. 1300)
or mouse anti-b3 tubulin (1:250; R&D Systems) were applied
followed by incubation in goat anti-rat AlexaFluor 555 secondary
antibody (1:250, Invitrogen). For b3-tubulin staining, the Mouse
on Mouse kit (Vectastain) was used. Specimens were counter-
stained with DAPI. For IHC on sectioned embryonic specimens
the same procedure was used after a rehydration step. For IHC on
adult tongues, tissue was processed according to a published
protocol  for markers of three types of taste receptor cells: type
1, rabbit anti-NTPdase2 (Nucleoside triphosphate diphosphohy-
drolase-2, 1:1000) is the ecto-ATPase of type I cells in taste buds
; type II, monoclonal anti-IP3R3 (Receptor for inositol 1,4,5-
trisphosphate, 1:1000; BD Transduction) is a second messenger
that mediates the release of intracellular calcium ; and type III,
rabbit anti-NCAM (Neural cell adhesion molecule, 1:1000) .
Secondary antibody was goat anti-rabbit Alexa-488 (1:500;
Invitrogen); this was applied for 2 hours and counterstained with
Analyses of placode size, apoptosis, proliferation, and
total number of cells
For quantification of placode size, wild-type and Spry22/2
tongues between E13 and E14 were stained with anti-E-Cadherin
antibody and the area measured using ImageJ software. Apoptosis
in the CV placode at E12.5 was measured using the In situ Cell
Death Detection kit (Roche) following manufacturer’s protocol.
Proliferating cells were identified by anti-PCNA immunofluores-
cence staining or injection of 1 mg BrdU for 2 hours followed by
staining with anti-BrdU antibody (Invitrogen). Anti-PCNA stained
and total (i.e. DAPI-stained) number of cells were counted using
ImageJ software and presented as a percentage of proliferating
cells. The total number of cells in the CV placode was quantified
from confocal images of E-cadherin stained sections using
Volocity5 software (Improvision).
qPCR reactions were performed using the GoTaq qPCR
Master Mix (Promega) in a Mastercycler Realplex (Eppendorf). All
primers were designed using PerlPrimer3 software ; sequences
are available upon request. qPCR conditions were as follows:
95uC, 2 minutes; 40 cycles at 95uC,15 seconds; 58uC,15 seconds;
68uC, 20 seconds; followed by a melting curve gradient.
Expression levels for the genes of interest were normalized to
levels of L19 and are presented as levels relative to wild-type.
Developing tongues at E11.5 and E12.5 were isolated and
cultured on 0.45 mm Millicell-HA membranes (Millipore) in F12/
DMEM medium (GIBCO/Invitrogen) containing 1% FBS, 2%
B27 culture supplement (GIBCO/Invitrogen), and antibiotics.
FGF signaling was inhibited by addition of 25 mM SU5402
suspended in DMSO (Calbiochem), and an equivalent volume of
DMSO was added to control wells. After 3 days in culture, the
tongues were fixed in 4% paraformaldehyde for 2 h and analyzed.
Fluorescent and bright field images were taken using a Leica
DM5000B with a Leica DFC500 camera. For confocal images, a
Leica SP5 Confocal was used.
Sample size, penetrance, and statistical analysis
Unless otherwise noted, all experiments were performed
independently in triplicate on at least three different specimens
(N$3), and when applicable, presented as an average 6 standard
deviation. Unpaired Student t-test was used to determine p-values
and p,0.01 was deemed to be significant. The CVP phenotype
observed in Spry2 null mice was 100% penetrant (n.12); the loss of
CVP in Fgf10 null mice was observed in 100% of the mice (n=6),
however, there were indications of small trenches or invaginations
(although absence of innervation) in 66% of the embryos; the
rescue of the single CVP in the double Fgf10;Spry2 null mice was
100% penetrant (n=4); the presence of multiple (i.e. $3) CVPs in
Spry1;Spry2 null mice was 86% penetrant (n=6). The Fisher exact
probability test was used to determine the p-values of the tongue
three types of taste receptor cells. Cartoons depict the sagittal
views of figure. (A-F) Immunofluorescence staining (green) in wild-
type and Spry22/2mice show the presence of the three taste
receptor cell types in the CVPs. Arrowheads point to the two
CVPs in Spry22/2mice. Scale bar, 100 mm. (G,H) b3-tubulin
immunofluorescence staining shows the innervation of both CVPs
in sagittal sections in tongues of Spry22/2mice. Scale bar, 10 mm.
CVP taste buds are innervated and are comprised of
staining of coronal sections of the developing CVP between E12.5
and E13.5. Scale bar, 50 mm. (J) Quantification of the placode size
between E13 to E14. **, p,0.001.
Innervation and placode size of the CVP. (A-I) H&E
of cell death in the CV placode at E12.5. (A) Total RNA was
extracted from the presumptive CVPs in wild-type and Spry22/2
littermates. The expression levels of various genes of interest were
assayed by qPCR. *, p,0.01; **, p,0.001. (B-D) TUNEL assay
was performed to quantify apoptotic cells. DNase-treated sections
from wild-type served as positive controls. Scale bar, 20 mm.
Expression levels of various genes and quantification
CVP development in vitro. (A-E’) Tongues were isolated from mice
at E11.5 (A-B’) and E12.5 (C-E’), grown in the absence and
presence of an inhibitor of FGF signaling (SU5402) for 3 days in
vitro, and immunostained using anti-E-cadherin. In tongues from
E11.5, there was an absence of CVP with SU5402 treatment
(A-B’). In E12.5 tongues, there was either an absence of CVP
(D,D’) or the presence of a malformed CVP (E,E’) relative to
controls (Co; C,C’). Scale bars, 50 mm. (F) A summary of
observations is presented with p-values calculated using the Fisher
exact probability test.
Inhibition of FGF signaling leads to the absence of
FGF Signaling Regulates CVP Development
PLoS Genetics | www.plosgenetics.org8June 2011 | Volume 7 | Issue 6 | e1002098
strates the innervation of multiple CVPs in tongues of DKO
mice. Scale bar, 20 mm.
b3-tubulin immunofluorescence staining demon-
mammalian taxa. (A) Perturbation of FGF signaling leads to effects
on CVP number in mice ranging from zero to three CVPs. The
speculative levels of FGF signaling are correlated with the number
of CVPs in the anterior-posterior orientation observed in various
mammalian species including the siamang, gorilla, chimpanzee,
and lemur. (B) Variation in mammalian CVP number. Common
species are listed for each clade. The red numbers represent the
typical number of CVPs with bracketed numbers showing the
range of numbers observed.
Summary of the variation in CVP number in the
number. Numbers represent the number of CVPs. The clades
are color-coded: Brown, Monotremata; Orange, Metatheria; Dark
blue, Afrotheria; Yellow, Xenarthra; Purple, Glires; Red, Arch-
onta; Green, Lipotyphla; Pink, Chiroptera; Black, Ferae; Light
Analysis of the variation in mammalian CVP
Blue, Euungulata. The number of CVPs is color-coded: Red, ,3;
Black, =3; Blue, 4–10; Green, .10.
We thank Gail Martin for providing mice and for insight and mentorship;
Jon Licht and Albert Basson for kindly providing mice; Uta Grieshammer,
Markus Bussen, and the members of the Martin and Klein laboratories for
their technical assistance and helpful discussions; Diana Laird and
members of her laboratory for help with Volocity; the Eli and Edythe
Broad Center of Regeneration Medicine and Stem Cell Research
microscopy facility; and the Robert D. Ogg Electron Microscope
Laboratory at UC Berkeley.
Conceived and designed the experiments: CIP AHJ PM LAB ODK.
Performed the experiments: CIP AHJ PM CC SC ST ODK. Analyzed the
data: CIP AHJ PM CC SC ST LAB ODK. Contributed reagents/
materials/analysis tools: CIP AHJ PM CC SC ST LAB ODK. Wrote the
paper: CIP AHJ CC ST LAB ODK.
1. Nguyen HM, Barlow LA (2010) Differential expression of a BMP4 reporter
allele in anterior fungiform versus posterior circumvallate taste buds of mice.
BMC Neurosci 11: 129.
2. Tizzano M, Dvoryanchikov G, Barrows JK, Kim S, Chaudhari N, et al. (2008)
Expression of Galpha14 in sweet-transducing taste cells of the posterior tongue.
BMC Neurosci 9: 110.
3. Kim MR, Kusakabe Y, Miura H, Shindo Y, Ninomiya Y, et al. (2003) Regional
expression patterns of taste receptors and gustducin in the mouse tongue.
Biochem Biophys Res Commun 312: 500–506.
4. Zhang C, Oakley B (1996) The distribution and origin of keratin 20-containing
taste buds in rat and human. Differentiation 61: 121–127.
5. Mistretta CM, Liu HX (2006) Development of fungiform papillae: patterned
lingual gustatory organs. Arch Histol Cytol 69: 199–208.
6. Chuong CM, Chodankar R, Widelitz RB, Jiang TX (2000) Evo-devo of feathers
and scales: building complex epithelial appendages. Curr Opin Genet Dev 10:
7. Chuong CM, Patel N, Lin J, Jung HS, Widelitz RB (2000) Sonic hedgehog
signaling pathway in vertebrate epithelial appendage morphogenesis: perspec-
tives in development and evolution. Cell Mol Life Sci 57: 1672–1681.
8. Jung HS, Oropeza V, Thesleff I (1999) Shh, Bmp-2, Bmp-4 and Fgf-8 are
associated with initiation and patterning of mouse tongue papillae. Mech Dev
9. Hall JM, Bell ML, Finger TE (2003) Disruption of sonic hedgehog signaling
alters growth and patterning of lingual taste papillae. Dev Biol 255: 263–277.
10. Liu HX, Maccallum DK, Edwards C, Gaffield W, Mistretta CM (2004) Sonic
hedgehog exerts distinct, stage-specific effects on tongue and taste papilla
development. Dev Biol 276: 280–300.
11. Mistretta CM, Liu HX, Gaffield W, MacCallum DK (2003) Cyclopamine and
jervine in embryonic rat tongue cultures demonstrate a role for Shh signaling in
taste papilla development and patterning: fungiform papillae double in number
and form in novel locations in dorsal lingual epithelium. Dev Biol 254: 1–18.
12. Zhou Y, Liu HX, Mistretta CM (2006) Bone morphogenetic proteins and
noggin: inhibiting and inducing fungiform taste papilla development. Dev Biol
13. Liu HX, Henson BS, Zhou Y, D’Silva NJ, Mistretta CM (2008) Fungiform
papilla pattern: EGF regulates inter-papilla lingual epithelium and decreases
papilla number by means of PI3K/Akt, MEK/ERK, and p38 MAPK signaling.
Dev Dyn 237: 2378–2393.
14. Iwatsuki K, Liu HX, Gronder A, Singer MA, Lane TF, et al. (2007) Wnt
signaling interacts with Shh to regulate taste papilla development. Proc Natl
Acad Sci U S A 104: 2253–2258.
15. Liu F, Thirumangalathu S, Gallant NM, Yang SH, Stoick-Cooper CL, et al.
(2007) Wnt-beta-catenin signaling initiates taste papilla development. Nat Genet
16. Beites CL, Hollenbeck PL, Kim J, Lovell-Badge R, Lander AD, et al. (2009)
Follistatin modulates aBMP autoregulatoryloopto controlthesizeandpatterning
of sensory domains in the developing tongue. Development 136: 2187–2197.
17. Nie X (2005) Apoptosis, proliferation and gene expression patterns in mouse
developing tongue. Anat Embryol (Berl) 210: 125–132.
18. Dikic I, Giordano S (2003) Negative receptor signalling. Curr Opin Cell Biol 15:
19. Guy GR, Wong ES, Yusoff P, Chandramouli S, Lo TL, et al. (2003) Sprouty:
how does the branch manager work? J Cell Sci 116: 3061–3068.
20. Kim HJ, Bar-Sagi D (2004) Modulation of signalling by Sprouty: a developing
story. Nat Rev Mol Cell Biol 5: 441–450.
21. Hacohen N, Kramer S, Sutherland D, Hiromi Y, Krasnow MA (1998) sprouty
encodes a novel antagonist of FGF signaling that patterns apical branching of the
Drosophila airways. Cell 92: 253–263.
22. de Maximy AA, Nakatake Y, Moncada S, Itoh N, Thiery JP, et al. (1999)
Cloning and expression pattern of a mouse homologue of drosophila sprouty in
the mouse embryo. Mech Dev 81: 213–216.
23. Minowada G, Jarvis LA, Chi CL, Neubuser A, Sun X, et al. (1999) Vertebrate
Sprouty genes are induced by FGF signaling and can cause chondrodysplasia
when overexpressed. Development 126: 4465–4475.
24. Jitpukdeebodintra S, Chai Y, Snead ML (2002) Developmental patterning of the
circumvallate papilla. Int J Dev Biol 46: 755–763.
25. Ahn SK, Chung J, Lee SH, Lee WS (1996) Prominent pigmented fungiform
papillae of the tongue. Cutis 58: 410–412.
26. Okubo T, Pevny LH, Hogan BL (2006) Sox2 is required for development of
taste bud sensory cells. Genes Dev 20: 2654–2659.
27. Kim JY, Lee MJ, Cho KW, Lee JM, Kim YJ, et al. (2009) Shh and ROCK1
modulate the dynamic epithelial morphogenesis in circumvallate papilla
development. Dev Biol 325: 273–280.
28. Thirumangalathu S, Harlow DE, Driskell AL, Krimm RF, Barlow LA (2009)
Fate mapping of mammalian embryonic taste bud progenitors. Development
29. Zhang S, Lin Y, Itaranta P, Yagi A, Vainio S (2001) Expression of Sprouty genes
1, 2 and 4 during mouse organogenesis. Mech Dev 109: 367–370.
30. Roehl H, Nusslein-Volhard C (2001) Zebrafish pea3 and erm are general targets
of FGF8 signaling. Curr Biol 11: 503–507.
31. Basson MA, Akbulut S, Watson-Johnson J, Simon R, Carroll TJ, et al. (2005)
Sprouty1 is a critical regulator of GDNF/RET-mediated kidney induction. Dev
Cell 8: 229–239.
32. Klein OD, Lyons DB, Balooch G, Marshall GW, Basson MA, et al. (2008) An
FGF signaling loop sustains the generation of differentiated progeny from stem
cells in mouse incisors. Development 135: 377–385.
33. Klein OD, Minowada G, Peterkova R, Kangas A, Yu BD, et al. (2006) Sprouty
genes control diastema tooth development via bidirectional antagonism of
epithelial-mesenchymal FGF signaling. Dev Cell 11: 181–190.
34. Shim K, Minowada G, Coling DE, Martin GR (2005) Sprouty2, a mouse
deafness gene, regulates cell fate decisions in the auditory sensory epithelium by
antagonizing FGF signaling. Dev Cell 8: 553–564.
35. Mohammadi M, McMahon G, Sun L, Tang C, Hirth P, et al. (1997) Structures
of the tyrosine kinase domain of fibroblast growth factor receptor in complex
with inhibitors. Science 276: 955–960.
36. Wells KL, Mou C, Headon DJ, Tucker AS (2011) Defects and rescue of the
minor salivary glands in Eda pathway mutants. Dev Biol 349: 137–146.
37. Barlow LA (2003) Toward a unified model of vertebrate taste bud development.
J Comp Neurol 457: 107–110.
38. Munne PM, Felszeghy S, Jussila M, Suomalainen M, Thesleff I, et al. (2010)
Splitting placodes: effects of bone morphogenetic protein and Activin on the
patterning and identity of mouse incisors. Evol Dev 12: 383–392.
39. Lee CC, Putnam AJ, Miranti CK, Gustafson M, Wang LM, et al. (2004)
Overexpression of sprouty 2 inhibits HGF/SF-mediated cell growth, invasion,
migration, and cytokinesis. Oncogene 23: 5193–5202.
FGF Signaling Regulates CVP Development
PLoS Genetics | www.plosgenetics.org9June 2011 | Volume 7 | Issue 6 | e1002098
40. Poppleton HM, Edwin F, Jaggar L, Ray R, Johnson LR, et al. (2004) Sprouty
regulates cell migration by inhibiting the activation of Rac1 GTPase. Biochem
Biophys Res Commun 323: 98–103.
41. Yigzaw Y, Cartin L, Pierre S, Scholich K, Patel TB (2001) The C terminus of
sprouty is important for modulation of cellular migration and proliferation. J Biol
Chem 276: 22742–22747.
42. Spuhler JN (1950) Genetics of three normal morphological variations: Patterns
of superficial veins of the anterior thorax, peroneus tertius muscle, and number
of vallate papillae. Cold Spring Harb Symp Quant Biol 15: 175–189.
43. Emura S, Okumura T, Chen H (2008) Morphology of the lingual papillae and
their connective tissue cores in the cape hyrax. pp 29–34.
44. Yoshimura K, Hama N, Shindo J, Kobayashi K, Kageyama I (2008) Light and
scanning electron microscopic study on the lingual papillae and their connective
tissue cores of the Cape hyrax Procavia capensis. Journal of Anatomy. pp
45. Sharma R, Vidyadaran M, Zulkifli I, Azlan J, Sumita S, et al. (1999)
Ecomorphological implications of the microstructures on the tongue of the fawn
roundleaf bat, Hipposideros cervinus (Chiroptera: Hipposideridae). Australian
Journal of Zoology. pp 405–409.
46. Sonntag CF (1921) The comparative anatomy of the tongues of the Mammalia.
II. Family 1. Simiidae. Proc Zool Soc 1: 1–29.
47. Sonntag CF (1921) The comparative anatomy of the tongues of the Mammalia.
V. Lemuroidea and Tarsioidea. Proc Zool Soc 1: 741–755.
48. Miller IJ, Jr., Whitney G (1989) Sucrose octaacetate-taster mice have more
vallate taste buds than non-tasters. Neurosci Lett 100: 271–275.
49. Hosley MA, Oakley B (1987) Postnatal development of the vallate papilla and
taste buds in rats. Anat Rec 218: 216–222.
50. Min H, Danilenko DM, Scully SA, Bolon B, Ring BD, et al. (1998) Fgf-10 is
required for both limb and lung development and exhibits striking functional
similarity to Drosophila branchless. Genes Dev 12: 3156–3161.
51. Lewandoski M, Martin GR (1997) Cre-mediated chromosome loss in mice. Nat
Genet 17: 223–225.
52. Metzger RJ, Klein OD, Martin GR, Krasnow MA (2008) The branching
programme of mouse lung development. Nature 453: 745–750.
53. Bartel DL, Sullivan SL, Lavoie EG, Sevigny J, Finger TE (2006) Nucleoside
triphosphate diphosphohydrolase-2 is the ecto-ATPase of type I cells in taste
buds. J Comp Neurol 497: 1–12.
54. Miura H, Nakayama A, Shindo Y, Kusakabe Y, Tomonari H, et al. (2007)
Expression of gustducin overlaps with that of type III IP3 receptor in taste buds
of the rat soft palate. Chem Senses 32: 689–696.
55. Yee CL, Yang R, Bottger B, Finger TE, Kinnamon JC (2001) "Type III" cells of
rat taste buds: immunohistochemical and ultrastructural studies of neuron-
specific enolase, protein gene product 9.5, and serotonin. J Comp Neurol 440:
56. Rozen S, Skaletsky H (2000) Primer3 on the WWW for general users and for
biologist programmers. In: Krawetz S, Misener S, eds. Bioinformatics Methods
and Protocols: Methods in Molecular Biology. TotowaNJ: Humana Press. pp
FGF Signaling Regulates CVP Development
PLoS Genetics | www.plosgenetics.org10 June 2011 | Volume 7 | Issue 6 | e1002098