Developmental Biology 287 (2005) 390 402

www.elsevier.com/locate/ydbio

Functions and regulations of fibroblast growth factor signaling during

embryonic development
Bernard Thisse, Christine Thisse *
Institut de Genetique et de Biologie Moleculaire et Cellulaire, UMR 7104, CNRS/INSERM/ULP, 1 rue Laurent Fries, BP 10142, CU de Strasbourg,
67404 ILLKIRCH cedex, France
Received for publication 16 June 2005, revised 29 July 2005, accepted 5 September 2005
Available online 10 October 2005

Abstract
Fibroblast growth factors (FGF) are secreted molecules which function through the activation of specific tyrosine kinases receptors, the FGF
receptors that transduce the signal by activating different pathways including the Ras/MAP kinase and the phospholipase-C gamma pathways.
FGFs are involved in the regulation of many developmental processes including patterning, morphogenesis, differentiation, cell proliferation or
migration. Such a diverse set of activities requires a tight control of the transduction signal which is achieved through the induction of different
feedback inhibitors such as the Sproutys, Sef and MAP kinase phosphatase 3 which are responsible for the attenuation of FGF signals, limiting
FGF activities in time and space.
D 2005 Elsevier Inc. All rights reserved.
Keywords: FGF; FGF receptors; Ras/MAP kinase; HSPG; Sprouty; Sef; MKP3

Introduction
Fibroblast growth factors represent a large family of secreted
molecules. Upon binding to their cognate receptors, FGFs
activate signal transduction pathways which are required for
multiple developmental processes. The wide-ranging biological
roles of FGFs, the variety of signaling pathways activated and
the complex and dynamic expressions of FGF ligands and
receptors imply that FGF signaling must be tightly regulated.
The scope of the present review is to provide readers with
updated information about FGF signaling. After a short
introduction on FGFs evolutionary history and structural
characteristics and on FGF receptors (FGFRs) particularities,
various functions of the FGF signaling will be presented,
focusing on their implication in development. The last part of
the review will describe the attenuation of FGF signaling mainly
dependent on the Ras/MAP kinase signaling pathway, through
the action of different factors, such as the Sproutys, Sef or MAP
kinase phosphatase 3. Recent publication of various studies has
led to controversial results and we will present the raising
debate on how FGF can be regulated and on how multiple
* Corresponding author.
E-mail address: thisse@igbmc.u-strasbg.fr (C. Thisse).
0012-1606/$ - see front matter D 2005 Elsevier Inc. All rights reserved.
doi:10.1016/j.ydbio.2005.09.011

feedback regulators of the FGF pathway act at different levels of

the signal transduction cascades to attenuate FGF signaling.
Key structural properties of FGFs and FGF Receptors
FGFs represent an extended family of polypeptide growth
factors found through evolution, ranging from nematode to
human whereas they have not been identified in unicellular
organisms (for review, Itoh and Ornitz, 2004). In invertebrates,
only three Drosophila genes (branchless, pyramus and thisbe)
have been found and two (egl-17 and let-756) in Caenorhabditis elegans. In contrast, in vertebrates, a large number of FGF
genes has been identified: 10 FGFs in zebrafish (FGF2 4, 6,
8, 10, 17a, 17b, 18, 24), 6 in Xenopus (FGF2 4, 8 10), 13 in
chicken (FGF1 4, 8 10, 12, 13, 16, 18 20), 22 genes in
mouse (FGF1 18, 20 23) and human (FGF1 14, 16 23).
FGF genes are scattered throughout the genome, indicating
that the FGF gene family was generated both by gene and
chromosomal duplication and translocation during evolution
(for review Ornitz and Itoh, 2001).
All FGFs share an internal core of similarity as well as a
high affinity for heparin. FGFs act through binding and
activation of FGFRs which are tyrosine kinase receptors that
contain three immunoglobulin-like domains and a heparin-

B. Thisse, C. Thisse / Developmental Biology 287 (2005) 390 402

binding sequence (Lee et al., 1989; Johnson et al., 1990, Fig.

1). FGFR forms are expressed in two possible ways: by the
expression of splice variants of a given FGFR gene or by the
expression of different FGFR genes (Johnson and Williams,
1993). Alternative splicing specifies the sequence of the

391

carboxy-terminal half of the Ig domain III, resulting in either

IIIb or IIIc isoforms (Miki et al., 1992; Chellaiah et al., 1994).
This alternative splicing event is regulated in a tissue-specific
manner and dramatically affects ligand receptor binding
specificity (see Powers et al., 2000).

Fig. 1. FGF receptors and FGF signal transduction. FGFRs are modular proteins comprising 3 immunoglobulin domains (IgI, IgII and IgIII). IgI and IgII are
separated by an acidic box (AD). IgII contains a heparin binding domain (HBD). The IgIII domain is followed by a unique transmembrane (TM), a juxtamembrane
(JM) and a kinase domain (KD) interrupted by an interkinase domain (IKD). FGF ligands linked to heparin sulfate proteoglycan (HSPG) bind to IgII and IgIII of
FGFR. This results in the dimerization and the subsequent transactivation by phosphorylation of specific tyrosine residues. The main two transduction pathways
involve the phospholipase C-g (PLCg) and the Ras/MAP kinase. The SH2 domain of the PLCg interacts with the phosphorylated Y766 of the activated receptor. The
activated PLCg hydrolyzes the phosphatidyl-inositol-4,5-diphosphate (PIP2) to inositol-1,4,5-triphophate (IP3) and the diacylglycerol (DAG). IP3 releases Ca2+
while DAG activates the protein kinase C-y (PKCy). Activated PKCy activates Raf by phosphorylating its S338 and stimulates the downstream pathway in a Ras
independent manner. The main pathway involves the interaction of the docking protein FRS2a with the amino-acid residues 407 433 (Xu et al., 1998). This protein
is activated by phosphorylation on multiple tyrosine residues and subsequently interacts and activates Grb2 linked to Sos, a nucleotide exchange factor involved in
the activation of Ras. Activated Ras then activates Raf which stimulates MEK which in turn phosphorylates the MAP kinase ERK. This last activated component
translocates to the nucleus and phosphorylates specific transcription factors of the Ets family which in turn activate expression of specific FGF target genes. P:
phosphorylation.

392

B. Thisse, C. Thisse / Developmental Biology 287 (2005) 390 402

Role of Heparin Sulfate Proteoglycans in FGF ligand

binding
In addition to high affinity for their receptors, members of
the FGF family have strong affinity for heparin (Burgess and
Maciag, 1989). FGF FGFR interaction and signaling are
further regulated by the spatial and temporal expression of
endogenous heparan sulfate proteoglycan (HSPG). HSPGs are
cell-surface and extracellular matrix macromolecules that
comprise a core protein to which heparan sulfate (HS)
glycosaminoglycan (GAG) chains are attached. HSPGs play
crucial roles in regulating key developmental pathways such as
the FGF signaling (for review, Lin, 2004). The genetic
evidences of such a role for HSPGs come from the isolation
of mutants, both in invertebrates and vertebrates that show
defective FGF signaling (Lin et al., 1999; Nybakken and
Perrimon, 2002). Biochemical (Nakato and Kimata, 2002) and
cristallographic studies (Schlessinger et al., 2000; Pellegrini et
al., 2000) have revealed the importance of 6 0 sulfation of HS
for FGF signaling which appears to be required for FGF1
FGFR2 and FGF2 FGFR1 interactions. HSPGs were found to
directly interact with FGFs and their receptors in a ternary
complex on the cell surface (Ornitz, 2000; Pellegrini, 2001).
One possibility could be that Heparin/HSPGs within the FGF
signaling complex facilitate the FGF FGFR interaction. The
second possibility could be that Heparin/HSPGs enhance the
stability of a FGF FGFR complex. Examination of the
spatiotemporal changes in HS to promote complexes formation
with FGFRs and FGFs led to the finding that these changes
depend on glycosyltransferases and sugar sulfotransferases.
Furthermore, for each FGF FGFR pair examined, unique
developmental HS binding patterns were identified that
correlate with different HS binding requirements, as well as
variations in FGF signaling. These results suggested that each
FGF FGFR combination seeks distinct HS domains that are
spatially and temporally regulated during development. These
domains are unique to HS and distinguish it from heparin,
which is highly sulfated and lacks any domain structure,
explaining the ability of these FGF FGFR pairs to interact
equally with heparin in vitro. Importantly, the HS activity
necessary for ternary complex assembly does not represent the
additive binding requirements of individual FGFs or FGFRs;
rather, it represents requirements dictated by the synergistic
interaction of the FGFs, HS and FGFRs. Finally, given that HS
specifically mediates each FGF FGFR interaction examined,
these results suggest that developmental changes in HS may
also specifically modulate the signaling of other families of
morphogens and growth factors (Allen and Rapraeger, 2003).
In addition, these interactions limit FGF diffusion and release
into interstitial spaces, regulating FGF activity in a given tissue
(Moscatelli, 1987; Flauennhaft et al., 1990).
FGF transduction pathway
FGFR transmits extracellular signals to various cytoplasmic
signal transduction pathways through tyrosine phosphorylation
(Fig. 1). FGFRs exist as inactivated monomers, which are

activated when two FGF molecules connected by a heparan

sulfate proteoglycan bind to the extracellular IgII and IgIII
domains of the receptor leading to its homodimerization
(Schlessinger et al., 2000). This dimerization brings together
the intracellular domain of the receptor leading to transautophosphorylation of several critical tyrosine residues. Activation
of the receptor allows proteins containing Src homology (SH2)
or phosphotyrosine binding (PTB) domains to bind to sequence
recognition motifs in FGFR1, resulting in phosphorylation and
activation of these proteins (Pawson, 1995; Forman-Kay and
Pawson, 1999; Dhalluin et al., 2000). For example, the
Phospholipase C-g (PLCg) was identified to be associated
with FGFR following ligand-dependent activation. This association is due to binding between the SH2 domain of PLCg and
Tyr 766 of FGFR1. Activated PLCg can hydrolyze phosphatidylinositol-4,5-diphosphate (PIP2) to inositol-1, 4, 5-triphosphate (IP3) and diacylglycerol (DAG). IP3 induces Ca2+
release from intracellular stores, whereas DAG is a protein
kinase C activator. The activated protein kinase C-y (PKCy) is
able to phosphorylate and to stimulate Raf therefore leading to
activate the MAP kinase pathway in a Ras-independent manner
(Ueda et al., 1996).
The main signaling activated through the stimulation of
FGFRs is the Ras/MAP kinase pathway. Links between
elements of this pathway and the activated FGFR involve the
interaction of the juxtamembrane domain of FGFR with the
FGFR substrate 2a (FRS2a) which is a membrane-anchored
docking protein. FRS2 lacks an SH2 domain, but contains a
phosphotyrosine binding domain (Kouhara et al., 1997), which
was shown to mediate phosphotyrosine-independent interaction with amino-acid residues 407 433 in FGFR1. Activation
of FGFR1 leads to tyrosine phosphorylation of FRS2 at several
sites. This allows the binding of Grb2, a small adaptor
molecule, which exists in complex with the nucleotide
exchange factor Sos and is involved in the activation of the
GTP-binding protein Ras. In this complex, Sos catalyzes the
exchange of GDP for GTP on Ras, promoting Ras activation as
well as downstream elements of the pathway, comprising Raf,
MEK and MAP kinases (ERK1, 2), the last member of which
finally enters the nucleus and phosphorylates target transcription factors (Sternberg and Alberola-Ila, 1998). Among these
are the Ets domain containing factors ERM and PEA3
(Wasylyk et al., 1998; Sharrocks, 2001). They belong to a
family of transcription factors, which share a winged-helix
loop helix domain with which they bind to DNA as
monomers. They are able to form heterocomplexes with
transcription factors and are activated most prominently by
the MAP kinase (Wasylyk et al., 1998).
A study performed during the course of somite development
revealed the contribution of these Ets-domain containing
proteins activated by FGFs. The FGF signal secreted from
the myotome induces formation of a scleraxis (Scx)-expressing
tendon progenitor population in the sclerotome. Brent and
Tabin (2004) showed that transcriptional activation by Pea3
and Erm in response to FGF signaling is both necessary and
sufficient for Scx expression in the somite. They proposed that
the domain of the somitic tendon progenitors is regulated both

B. Thisse, C. Thisse / Developmental Biology 287 (2005) 390 402

by the restricted expression of Pea3 and Erm, and by the

precise spatial relationship between these Ets transcription
factors and the FGF signal originating in the myotome.
Implication of FGF signaling during development
FGFs were described to be implicated in multiple roles such
as patterning, morphogenesis, differentiation, regulation of cell
proliferation or migration (for review see Goldfarb, 1996;
Powers et al., 2000; Coumoul and Deng, 2003). In the next
paragraphs, we will illustrate major functions of FGF signaling
during development.
Early patterning and dorso-ventral axis formation
Induction and patterning of mesoderm is one of the earliest
events in which FGF signaling was revealed to be essential.
The first mesodermal inducer identified was basic FGF (Slack
et al., 1987), for which its role is evolutionarily conserved.
The effect of FGF on mesoderm formation was initially
studied in Xenopus in which not only FGF was described to
have a mesoderm inducing capacity (Slack et al., 1987;
Kimelman and Kirschner, 1987; Kimelman et al., 1988) but
components of the FGF signaling pathway such as FGFRs,
HSPGs, FRS2, Grb2 and the Ras downstream cascade, when
inhibited, were shown to block mesoderm formation and to
induce both gastrulation and posterior defects (Whitman and
Melton, 1992; MacNicol et al., 1993; Umbhauer et al., 1995;
LaBonne, 1995; Gotoh, 1995).
Involvement of FGFs in the establishment of the dorsoventral (D/V) axis has been extensively studied using the
zebrafish model. Genetic analysis revealed that BMP2b/BMP7
heterodimers (Schmid et al., 2000) act as morphogens secreted
ventrally and are responsible for the specification of ventral cell
fates. The expression of BMP genes, initially activated in the
whole blastula becomes progressively restricted ventrally. This
ventral restriction coincides with the spreading of FGF activity
from the dorsal side of the embryo, suggesting an implication
of FGF signaling in the dorsal downregulation of BMP gene
expression. Consistently, general activation of the FGF/Ras/
MAP kinase signaling pathway inhibits BMP gene expression
in the whole blastula. Conversely, inhibition of FGF signaling
causes BMP gene expression to enlarge dorsally, leading to an
expansion of ventral cell fates. Altogether, this shows that FGF
signaling is essential to delimit the expression domain of BMPs
in blastula stage embryos. Therefore, FGFs act upstream of the
ventral morphogens and function as initial signals for the
establishment of the D/V patterning (Furthauer et al., 2004).
FGF signaling was also identified to be implicated in
ventro-posterior mesodermal tissue maintenance. Consistently,
blocking FGF signaling by dominant-negative FGFR results in
deletions of tail and trunk structures (Amaya et al., 1991;
Griffin et al., 1995). As well, targeted disruption of mouse
FGFR1 leads to the disorganization of axial mesoderm and to a
defective development of the posterior structures. FGFs control
the specification and maintenance of mesoderm by regulation
of T box genes (Griffin et al., 1995; Smith et al., 1991; Strong

393

et al., 2000; Sun et al., 1999; Ciruna and Rossant, 2001; Griffin
et al., 1998; Zhao et al., 2003). One FGF mesodermal target is
brachyury, required for posterior mesoderm and axis formation
in mouse, zebrafish and Xenopus (Smith et al., 1991; Herrmann
et al., 1990; Halpern et al., 1993; Conlon et al., 1996). In
Xenopus, Brachyury induces eFGF, which is required for
myogenic cell lineage (Stanley et al., 2001; Fisher et al., 2002)
and vice versa (Isaacs et al., 1994; Schulte-Merker and Smith,
1995).
Role of FGFs on cell movements
Evidence has been provided that FGF signaling plays a
direct role in cell movements during convergence extension
(Nutt et al., 2001). In the mouse, FGF signaling may affect
gastrulation movements via the transcription factor snail
(Ciruna and Rossant, 2001). FGFR1 mutant embryos die at
late gastrulation, showing defects in cell migration, cell fate
specification and patterning. FGFR1 mutant cells may fail to
migrate properly because they maintain high levels of Ecadherin resulting from a snail downregulation. An additional
study performed in the mouse revealed that, in the absence of
both FGF8 and FGF4, epiblast cells move into the streak and
undergo an epithelial-to-mesenchymal transition; however,
most cells then fail to move away from the streak. As a
consequence, no embryonic mesoderm- or endoderm-derived
tissues develop demonstrating that signaling via FGF8 and/or
FGF4 is required for cell migration away from the primitive
streak (Sun et al., 1999).
Moreover, there is an increasing evidence for a direct
chemotactic response of migrating cells in response to FGF
signals. As such, FGF2 and FGF8 are potent chemoattractants
in migration of mesencephalic neural crest cells (Kubota and
Itoh, 2000) and FGF2 and FGF4 attract mesenchymal cells
during limb bud development in the mouse (Webb et al., 1997;
Li and Muneoka, 1999). Also, FGF10 exerts a potent
chemoattractive effect to direct lung distal epithelial buds to
their destination (Park et al., 1998). Furthermore, FGFs can act
as chemotactic signals to coordinate cell movements during
gastrulation. In the chick, cells can be guided by environmental
signals and beads coated with FGFs alter their trajectories. The
observed movement patterns of anterior streak cells can be
explained by an FGF8-mediated chemorepulsion of cells away
from the streak followed by chemoattraction towards an FGF4
signal produced by the forming notochord (Yang et al., 2002).
Neural induction and FGF signaling
Several studies on neural induction have led to the ?default
modelX which proposes that ectodermal cells are fated to
become neural by default. In this model, cells are normally
inhibited from a neural fate by BMPs that are expressed in the
ectoderm, from which they must be released for neural
induction to occur (for reviews, Hemmati-Brivanlou and
Melton, 1997; Weinstein and Hemmati-Brivanlou, 1999).
Additional data have suggested that FGF signaling plays a
crucial role for an early step of neural induction in the chick. In

394

B. Thisse, C. Thisse / Developmental Biology 287 (2005) 390 402

this specie, induction of neural tissue precedes the formation of

the Hensens node. FGFs are sufficient to induce preneural
markers and inhibition of FGF signaling blocks neural
induction by Hensens node (Streit et al., 2000; Wilson et al.,
2000; Alvarez et al., 1998; Storey et al., 1998). FGF signals
repress BMP expression which leads to induction of neural
tissue. On the contrary, when FGFs are inhibited, BMP
expression is maintained and neural fate is blocked (Wilson
et al., 2000). FGFs may block BMP signaling during neural
induction of the chick by inducing Churchill which is required
for the expression of Smad-interacting protein 1 (Sip1) in the
neural plate. Sip1 binds and represses Smad1/5 blocking
therefore BMP signaling (Sheng et al., 2003).
In Xenopus, the role of FGF signaling in neural induction
has also been intensively studied. However, its implication in
neural induction has led to conflicting results. Several
investigators suggested that FGF signaling is required directly
for neural induction (Launay et al., 1996; Hongo et al., 1999;
Streit et al., 2000; Bertrand et al., 2003). In most cases,
however, experiments were performed in a background where
the BMP signaling was partially attenuated (Kengaku and
Okamoto, 1995; Lamb and Harland, 1995; Ribisi et al., 2000).
In addition, several studies using dominant negative FGFR1
have led to different results showing that neural tissue forms
even when FGF signaling is inhibited (Kroll and Amaya, 1996;
Ribisi et al., 2000; Barnett et al., 1998; McGrew et al., 1997;
Holowacz and Sokol, 1999). It appears that FGF signaling
required for anterior neural development may be mediated
through FGFRs other than FGFR1, such as through FGFR4a
(Hongo et al., 1999).
In contrast, other studies report a requirement for FGFs in
the antero-posterior patterning rather than the initial induction
of the neural plate (Holowacz and Sokol, 1999). FGFs are able
to change anterior neural fate to more posterior neural cell
types (Umbhauer et al., 2000; Lamb and Harland, 1995; Ribisi
et al., 2000; Holowacz and Sokol, 1999; Cox and HemmatiBrivanlou, 1995). This may thus involve interactions with the
Wnt signaling pathway (McGrew et al., 1997; Kazanskaya et
al., 2000; Domingos et al., 2001; Kudoh et al., 2002).
Finally, a controversial study provides evidence that BMP
inhibition is required in the chick, but only as a relatively late
step. BMP inhibition would not be sufficient to cause
competent cells to acquire neural fates either in chick or in
Xenopus. In the frog, FGF synergizes with BMP inhibition to
induce neural markers. In chick, inhibition of BMP signaling
with Wnt antagonists and/or FGF is not sufficient for neural
induction. Neural induction would not occur by default but
would rather involve a succession of signaling events with
some players which would remain to be identified (Linker and
Stern, 2004).
Midbrain hindbrain patterning
Early patterning of the vertebrate presumptive midbrain and
rhombomere 1 (rh1) is regulated by a local organizer located at
the mid/hindbrain junction (Liu and Joyner, 2001), a territory
expressing FGF8. This factor has organizer properties and is

required for midbrain and cerebellum development (Meyers et

al., 1998, Reifers et al., 1998). FGF17 and FGF18 are also
expressed in the mid/hindbrain junction in broader domains
than FGF8, including posterior midbrain (Maruoka et al., 1998;
Xu et al., 1998). Loss-of-function of FGF17 in the mouse
results in truncation of the posterior midbrain and reduced
proliferation of the anterior cerebellum. Removal of one copy
of fgf8 in a fgf17 mutant background leads to exaggerated
cerebellum phenotype (Xu et al., 2000). To complicate matters,
fgf8 is differentially spliced to generate FGF8a and FGF8b
isoforms. FGF8a misexpression causes expansion of the
midbrain and FGF8b transforms the midbrain into cerebellum.
The initial effect of FGF8b is to reduce growth of the midbrain.
Therefore, FGF8a and FGF8b have distinct activities, both on
growth and regulation of gene expression (reviewed in Liu and
Joyner, 2001; Sato et al., 2001).
Interactions of FGF8a and FGF8b with FGF17 and FGF18
have been studied. It appears that FGF8b after being induced in
the presumptive rh1 territory induces FGF18 in surrounding
tissue, further extending the gradient of FGF RNA expression.
FGF8b maintains two negative feedback loops by inducing the
expression of the negative feedback FGF inhibitors Spry1 and
Spry2 and repressing FGFR2 and FGFR3 (Liu et al., 2003).
FGF17, FGF18 proteins and possibly FGF8a as well as a low
level of FGF8b then regulate proliferation of the midbrain and
cerebellum (Liu et al., 2003; Chi et al., 2003).
Limb induction and morphogenesis
The role of FGFs has been established in induction,
initiation and maintenance of limb development (reviewed in
Martin, 1998). Limb bud initiation is triggered by FGF8
inducing the expression of FGF10. FGF10 then in turn induces
FGF8 in ectodermal cells resulting in the formation of the
apical ectodermal ridge (AER). FGFs from the AER maintain
cell proliferation in the progress zone (a population of
undifferentiated mesenchymal cells located near the distal tip
of the limb bud, adjacent to the AER). FGF2 (Fallon et al.,
1994), FGF4 (Laufer et al., 1994) and FGF8 (Crossley et al.,
1996) induce sonic hedgehog (shh) expression in the zone of
polarizing activity (ZPA). Outgrowth and patterning of the limb
result from the combined effects of FGF and Shh and
regulation of various genes, including members of the Hox
family (reviewed in Martin, 1998).
A first model proposed for explaining the patterning of the
limb development along the proximal distal (PD) axis is the
progress zone model (Summerbell et al., 1973) which explains
that as the limb grows, cells are pushed out of the progress zone
and become fixed in the positional value that they acquired.
Cells that leave the proximal zone early will adopt a proximal
fate and cells that leave the progress zone late will form distal
elements (Sauders, reviewed in Niswander, 2003). The
conclusion from this model is that the PD fate is specified
progressively. Recent studies have provided an alternative
model on how the AER may influence the PD pattern. This
second model proposes that all PD fates are already specified
within the early limb bud rather than progressively (Dudley et

B. Thisse, C. Thisse / Developmental Biology 287 (2005) 390 402

al., 2002). In the absence of FGF8 in mice, the proximal

element of the limb is reduced or even lost. However, the distal
elements are formed quite normally because other FGFs are
activated (Lewandoski et al., 2000; Moon and Capecchi, 2000).
These results of loss of proximal but not of distal elements
cannot be interpreted within the progress zone model. In
addition, new experimental data have proposed that FGFs
function is first to influence the initial size of the limb (Sun et
al., 2002). In FGF8 mutant mice, early limb bud is smaller.
This phenotype is detected right after FGF8 expression is
detectable, excluding the possibility that FGF8 alters cell
proliferation or cell death. Instead, FGF8 would (by affecting
morphogenetic movements or cell adhesion) influence the
number of cells in the limb bud by preventing their death. In
mutant limbs, death of proximal but not distal mesenchyme can
be observed. The PD precartilage condensations are formed but
smaller. Sun et al. (2002) proposed that the AER generates an
adequate number of mesenchymal cells for correctly sized
condensations to form. When the AER is disrupted, the number
of skeletal elements is decreased resulting in smaller or in
absence of condensations arguing that the FGFs function is to
control mesenchymal cell number but not the PD patterning.

395

osteoblasts and thereby enhance the osteoblast population,

whereas continuous signaling may promote apoptosis in more
mature osteoblasts, therefore limit the early increase in the
osteoblasts pool. FGFs interact with other growth factors
signaling pathways to regulate osteoblast function. For
example, FGF2 and FGFR2 inhibit the expression of the
BMP antagonist Noggin in the cranial sutures, resulting in
increased BMP4 activity and suture fusion (Warren et al.,
2003). Therefore, FGF signaling can control cranial suture
fusion indirectly through BMP signaling. FGFs regulate also
genes involved in matrix degradation (Kawaguchi et al., 1995)
and induce the expression of tissue inhibitors of metalloproteinases (Delany and Canalis, 1998; Varghese et al., 1995)
indicating that FGFs (mainly FGF2) may modulate bone matrix
proteolysis by regulating collagenase activity. Overall, activation of FGF signaling is able to regulate genes involved at all
steps of osteogenesis (Fig. 2). This array of genes may
participate in the regulation of cell progression from osteoprogenitor cells to the end of the osteoblast life (for review, Marie,
2003). Further studies have led to the identification of signaling
pathways that are involved in FGF signaling in osteoblasts,
mainly MAP kinase signal transduction cascade (Mansukhani
et al., 2000; Debiais et al., 2001).

Bone formation
Attenuation of FGF signaling
The importance of FGF signaling in skeletal development
was first revealed with the discovery that a point mutation in
the transmembrane domain of FGFR3 is the etiology of
achondroplasia, the most common genetic form of dwarfism
in human (Rousseau et al., 1994; Shiang et al., 1994). A major
role of FGFs and FGFRs has then been scored by finding
missense mutations resulting in more than 15 human disorders,
ranging from skeletal dysplasias, craniosysnostosis syndromes
or short stature (for review Coumoul and Deng, 2003). In the
bone, FGF2 and FGF9 transcripts are found in mesenchymal
cells and osteoblats (Gonzalez et al., 1996; Kim et al., 1998).
FGF18 is expressed in mesenchymal cells, in differentiating
osteoblasts during calvarial bone development and in the
perichondrium of long bones (Ohbayashi et al., 2002). The
action of FGFs is dependent on the spatiotemporal pattern of
expression of FGFRs like FGFR1 and FGFR2 which are
located in mesenchymal cells during condensation of mesenchyme prior to deposition of bone matrix at early stages of long
bone development and in the cranial sutures (Orr-Urtreger et
al., 1991; Delezoide et al., 1998). Later, they are found in
preosteoblasts and osteoblasts together with FGFR3 (Molteni
et al., 1999). Overexpression of FGF2 in mouse induces
abnormal bone formation (Coffin et al., 1995) whereas its lossof-function leads to inhibition of bone formation (Montero et
al., 2000). Mice lacking FGF18 display ossification and cranial
sutures closure delay (Liu et al., 2002; Ohbayashi et al., 2002).
Altogether, FGF signaling appears to regulate cell proliferation
and differentiation positively in membranous and long bone
osteogenesis. FGF2 was shown also to act in vivo on osteoblast
precursor cells replication to increase osteoblast number. FGF
signaling also controls genes involved in osteoblast apoptosis.
High FGF signaling may reduce apoptosis of immature

The multiple range of biological effects of FGFs and the

variety of signaling pathways activated by this family of
ligands imply that FGF signaling must be tightly regulated
regarding timing, duration and spread of the signal. This can be
achieved by both positive and negative feedback loops. FGF
signaling can be regulated at many levels in the extracelular
space and within the cell. A growing number of proteins have
been identified that specifically regulate the activity of FGF
signaling pathways through FGFRs. Many of these regulatory
factors belong to the FGF synexpression group (Furthauer et
al., 2002; Niehrs and Meinhardt, 2002; Tsang et al., 2002). Not
only are these factors coexpressed with FGFs but they are
themselves regulated by FGF signaling and inhibit FGF
signaling by establishing a negative feedback loop.
The Sprouty proteins
The first feedback regulator of the FGF pathway isolated is
Sprouty (Spry). Spry was originally identified as an antagonist
of FGF signaling that patterns apical branching in Drosophila
tracheae (Hacohen et al., 1998). In spry mutants, ectopic
airway branches are induced due to excess of FGF signaling.
Spry was then described to be a general inhibitor of receptor
tyrosine kinase (RTK) signaling (for review, Kim and Bar-Sagi,
2004). Spry proteins are evolutionarily conserved with four
members (Spry1, 2, 3, 4) identified in mammals (Dikic and
Giordano, 2003). These proteins share a conserved carboxyterminal cystein-rich domain, necessary for their specific
localization and function. Their amino-terminal part is divergent except that they possess an invariant tyrosine phosphorylation site. It is thought that this sequence divergence dictates

396

B. Thisse, C. Thisse / Developmental Biology 287 (2005) 390 402

Fig. 2. FGF signaling regulates osteogenesis. During intramembranous bone development, bone formation is characterized by the proliferation of mesenchymal cells
followed by their commitment to become osteoprogenitor cells. They differentiate into preosteoblasts then into bone matrix-forming mature osteoblasts. FGFs are
required for each step of bone formation through FGFRs to control various genes involved in osteoblast commitment, differentiation and apoptosis. ALP: alkaline
phosphatase, BSP: bone sialoprotein, COL I: collagen type I, OC: osteocalcin, OP: osteospondin, ON: osteonectin (adapted from P. J. Marie, 2003).

various functions by mediating different protein protein

interactions. Gain- and/or loss-of-functions studies in the
mouse, Xenopus and zebrafish (Minowada et al., 1999;
Mailleux et al., 2001; Nutt et al., 2001; Furthauer et al.,
2001, 2004) have demonstrated that Sprys antagonize FGF
signaling. Spry expression is induced by the FGF signaling and
has various biological consequences, including regulation of
cell proliferation and differentiation, endocytic sorting of
activated RTKs, control of cytoplasmic calcium concentration,
migratory behavior of cells, retro-control of D/V patterning and
lung and bone development. Spry genes contribute also to
reciprocal epithelial mesenchymal and stromal signaling
controlling ureteric branching, which involves the coordination
of Fgf/Wnt11/Gdnf pathways during mammalian kidney
development (Chi et al., 2004).
Study of Sprys function has led to intense investigation and
controversy (Fig. 3). Works from many groups have identified
Sprys as negative regulators of the MAP kinase ERK.
However, contradictory reports have identified Spry inhibition
upstream of Ras (Casci et al., 1999; Hanafusa et al., 2002) or at
the level of Raf (Reich et al., 1999; Sasaki et al., 2003; Yusoff
et al., 2002). Moreover, Spry family members have varying
interacting partners, such as c-Cbl, Grb2, Raf1 and FRS2 (for
review, Dikic and Giordano, 2003).
The mechanisms by which Sprys interfere with ERK
activation have been extensively studied on EGF signaling
pathway and have come from the finding that Spry proteins are
regulated by tyrosine phosphorylation (Tyr55 in spry2, Tyr53
in Spry1/4) in response to stimuli. Phosphorylated Tyr55 serves
as a docking site for the SH2 domain of c-Cbl (Egan et al.,
2002; Wong et al., 2002; Rubin et al., 2003; Hall et al., 2003).
Usually, c-Cbl binds to phosphorylated tyrosines on the

activated EGFR, which in turn results in its ubiquination,

endocytosis and degradation through the proteasomal/lysosomal pathways (Dikic and Giordano, 2003). As Spry proteins
phosphorylated on Tyr 55 can also interact with c-Cbl, they are
in direct competition with activated EGFRs for binding to this
protein. As a result, the EGFR is not ubiquinated, internalized
or degraded in the presence of Spry, leading to prolonged cellsurface exposure of the receptor and sustained signaling
activity. Therefore, in this context, Sprys function as positive
regulators of the EGF signaling resulting in an upregulation of
the MAP kinase ERK activity. A similar degradation of Spry2
by c-Cbl may also be true for FGFR signaling (Hall et al.,
2003).
However, most of the studies performed on FGF signaling
pathway show that Sprys associate with Grb2. For example,
mouse Spry2 and Xspry1 can interact with the SH2 domain of
Grb2 after FGFR-induced phosphorylation of Tyr55 (Hanafusa
et al., 2002). As a result, the interaction of Grb2 with FRS2 is
blocked and FGF-induced signal transduction is repressed. In
addition, Spry2 and 4 have been shown to bind directly to Raf
through a Raf-binding motif in their C-terminal domain and
suppress phosphorylation of Raf on Ser338 by the Raf kinase
PKCy (Sasaki et al., 2003).
Altogether, Sprys function through different pathways,
which determine the overall effect on RTK signaling. First,
binding of c-Cbl to tyrosine-phosphorylated Spry results in the
stabilization of EGFR and in sustained ERK activity. Second,
binding of Grb2 to tyrosine-phosphorylated Spry can interrupt
the FGFR/Ras/Raf/MAPK pathway upstream of Ras. Finally,
binding of the C-terminal domain of Spry by its Raf-binding
motif inhibits Raf phosphorylation by PLCg-activated PKCy.
However, the criteria to decide between Spry different

B. Thisse, C. Thisse / Developmental Biology 287 (2005) 390 402

Fig. 3. Models for the attenuation of FGF signaling by Sprys. (I) Spry acts downstream of the FGFR and upstream of Ras by binding to the adaptor protein Grb2 at the level of its phosphorylated Tyr55. This prevents
the activation of Ras and the subsequent stimulation of the downstream elements of the pathway. (II) Spry also acts downstream of Ras by directly binding Raf. This is achieved by interaction of its Raf-binding domain
located in its C-terminal part and this prevents both activation by Ras and phosphorylation of Ser338 by the PKCy. This explains how spry can inhibit the transduction of FGF signaling by the PLCg.

397

398

B. Thisse, C. Thisse / Developmental Biology 287 (2005) 390 402

functions are not yet known and will have to be defined in the
near future.
Other FGF synexpression group members
In addition to sprys, several other genes involved in the
regulation of the FGF pathway have been isolated through the
course of large-scale in situ hybridization screens (Gawantka et
al., 1998; Thisse et al., 2001; Kudoh et al., 2001) designed to
identify genes showing expression pattern similar or overlapping with FGFs. Various genes were isolated such as
XFLRT3 (Bottcher et al., 2004), Sef (Furthauer et al., 2002;
Tsang et al., 2002), the MAP kinase phosphatase 3 (MKP3,
also called Pyst1, Eblaghie et al., 2003; Tsang et al., 2004),
PEA3 (for polyoma virus enhancer activator 3) and ERM

(Munchberg et al., 1999). These last two factors are the

downstream players of the FGF signaling pathway while the
others, also induced by FGF stimulation are involved in the
attenuation of the signal (for review, Tsang and Dawid, 2004).
XFLRT3
This gene encodes a transmembrane protein characterized
by a cluster of leucine-rich repeats and one FNIII domain
within the extracellular region. Its expression is induced after
activation of FGF signaling and downregulated after inhibition
of this pathway. XFLRT3 signaling results in phosphorylation
of the MAP kinase ERK and is blocked by the MAP kinase
phosphatase 1. In gain- and loss-of-function experiments
performed in Xenopus, FLRT3 was shown to phenocopy the

Fig. 4. FGF signaling and its regulation. The stimulation of FGFR by FGF ligands results in the activation of specific target genes. Among them are several feedback
inhibitors which are involved in the attenuation of FGF signaling. Spry acts at the level of Grb2 and/or at the level of Raf. Sef and XFLRT3 are both located at the
membrane and interact with FGFR. Sef functions as a negative regulator while XFLRT3 enhances the FGF activity resulting in the phosphorylation of the MAP
kinase ERK. Sef was shown also to affect the phosphorylation of the MAP kinase, either directly at the level of ERK or by preventing the phosphorylation of ERK by
MEK. Finally, MKP3 negatively regulates the FGF signaling pathway by dephosphorylation of activated MAP kinase.

B. Thisse, C. Thisse / Developmental Biology 287 (2005) 390 402

399

FGF signaling (Bottcher et al., 2004). Surprisingly, in zebrafish, although FLRT3 is expressed in well known centers of
FGF signaling, neither gain nor loss of FGF signaling appears
to directly affect FLRT3 expression. Inactivating FLRT3
through injection of antisense morpholinos impairs convergence-extension movements that direct axial elongation but
fails to mimick phenotypes observed after either loss or gain
function of FGF activity. These observations suggest that, in
zebrafish, FLRT3 may be involved in directing cell movements
but does not appear to be related to the FGF pathway
(Furthauer M., Thisse, B. and Thisse, C., unpublished data).

regulates the MAP kinase cascade through dephosphorylation

of activated MAPK proteins. Upon binding of phosphorylated
MAPKs to the MAPK-binding domain in MKP3, a conformational activation of the C-terminal phosphatase domain is
achieved, leading to the inactivation of MAPKs (Camps et al.,
2000; Fjeld et al., 2000; Zhao and Zhang, 2001). In chicken,
ectopic expression of MKP3 results in the disruption of limb
outgrowth, a phenotype characteristic of FGF signaling
blockage (Kawakami et al., 2003; Eblaghie et al., 2003). In
zebrafish, MKP3 limits the extent of FGF activity of the
gastrula embryo (Tsang et al., 2004).

Sef

Concluding remarks

The Sef protein is conserved across zebrafish, mouse and

human but not in invertebrates. Homology searches with Sef
revealed that it encodes a putative transmembrane protein, with
weak similarities to the Interleukin-17 receptor and a putative
tyrosine phosphorylation site into the intracellular region
(Furthauer et al., 2002; Tsang et al., 2002). In zebrafish, lossor gain-of-function studies revealed Sef to function as an
antagonist of FGF signaling and to interfere with FGF signal
transduction by acting dowstream or at the level of MEK and
upstream (or at the level) of the MAP kinase (Furthauer et al.,
2002). Further studies performed using human Sef (hSef)
showed that it binds to activated forms of MEK, inhibits the
dissociation of the MEK ERK complex and blocks nuclear
translocation of activated ERK. Consequently, hSef inhibits
phosphorylation and activation of the nuclear ERK substrate
Elk-1, while it does not affect phosphorylation of the
cytoplasmic ERK substrate RSK2. Downregulation of endogenous hSef by hSef siRNA enhances the stimulus-induced
ERK nuclear translocation and the activity of Elk-1. These
results favor the hypothesis that hSef acts as a spatial regulator
for ERK signaling by targeting ERK to the cytoplasm (Torii et
al., 2004). Other reports showed that the intracellular domain of
Sef is required for its inhibitory function and its interaction
with FGFR1 and FGFR2 (Tsang et al., 2002; Kovalenko et al.,
2003, Xiong et al., 2003). Ectopic expression of Sef blocks the
phosphorylation of FRS2. The conclusion of these studies was
that Sef acts at the level of FGFRs blocking the Ras and PI3K
pathways downstream. To complicate the further understanding
of Sef function, alternative spliced isoforms of Sef were found
in human (Preger et al., 2004). Sef-b which lacks the signal
peptide is localized in the cytoplasm and can suppress FGF
signaling downstream of MEK. Because FGFs deliver multiple
biological responses, it is possible that like Spry, Sef can inhibit
the FGF pathway at various points depending on the context, in
order to allow fine-tuning of signal regulation.

A substantial amount of important work has been done to

establish the role of FGF signaling during development showing
that FGFs regulate cell fate and specification of germ layers.
FGF downstream elements have been characterized and it
becomes evident that a growing number of these factors are
themselves tightly regulated by FGFs. It is clear from various
studies that multiple feedback inhibitors exist and that they act at
different levels of the signal transduction cascades to attenuate
FGF signal (Fig. 4). Regulation can occur at different levels of
the signal transduction pathways including the membrane down
to the level of phosphorylation of the MAP kinase. Some of the
feedback inhibitors are involved in more than one process (Sef,
Spry) and negatively regulate other RTK signaling pathways.
The next challenge will be to integrate the intricate
interactions between these different regulators within the
FGF pathway and between different RTK pathways. In the
future, it will be important to assess the pathways employed by
FGFs in different cells and define the determinants for
signaling specificity. This can include feedback regulation,
threshold-type responses and posttranslational modifications.
This will be complicated by the fact that these modulators
combine a wide range of activity. The complete understanding
of these mechanisms will have diverse implications in biology
as well as in therapeutic approaches.

MAP kinase phosphatase 3 (MKP3)

Two distinct domains are characteristic of MKP3 proteins,
the N-terminal region, which contains a high-affinity ERK/
MAPK binding domain and the C-terminal domain, which
constitute a dual specificity phosphatase (Stewart et al., 1999;
Zhang et al., 2003; Zhao and Zhang, 2001). MKP3 negatively

Acknowledgment
We thank the ?FGF communityX for help and apologize for
not having being able to cite as many papers as we would have
wished because of limited space.
References
Allen, B.L., Rapraeger, A.C., 2003. Spatial and temporal expression of heparan
sulfate in mouse development regulated FGF and FGF receptor assembly.
J. Cell Biol. 163, 637 648.
Alvarez, I.S., Araujo, M., Nieto, M.A., 1998. Neural induction in whole chick
embryo cultures by FGF. Dev. Biol. 199 (1), 42 54.
Amaya, E., Musci, T.J., Kirschner, M.W., 1991. Expression of a dominant
negative mutant of the FGF receptor disrupts mesoderm formation in
Xenopus embryos. Cell 66 (2), 257 270.
Barnett, M.W., Old, R.W., Jones, E.A., 1998. Neural induction and patterning
by fibroblast growth factor, notochord and somite tissue in Xenopus. Dev.
Growth Differ. 40 (1), 47 57.

400

B. Thisse, C. Thisse / Developmental Biology 287 (2005) 390 402

Bertrand, V., et al., 2003. Neural tissue in ascidian embryos is induced by

FGF9/16/20, acting via a combination of maternal GATA and Ets
transcription factors. Cell 115 (5), p. 615.
Bottcher, R.T., et al., 2004. The transmembrane protein XFLRT3 forms a
complex with FGF receptors and promotes FGF signalling. Nat. Cell Biol.
1, 38 44.
Brent, A.E., Tabin, C.J., 2004. FGF acts directly on the somitic tendon
progenitors through the Ets transcription factors Pea3 and Erm to regulate
scleraxis expression. Development 131 (16), 3885 3896.
Burgess, W.H., Maciag, T., 1989. The heparin-binding (fibroblast) growth
factor family of proteins. Annu. Rev. Biochem. 58, 575 606.
Camps, M., Nichols, A., Arkinstall, S., 2000. Dual specificity phosphatases: a
gene family for control of MAP kinase function. FASEB J. 14, 6 16.
Casci, T., Vinos, J., Freeman, M., 1999. Sprouty, an intracellular inhibitor of
Ras signaling. Cell 96 (5), 655 665.
Chellaiah, A.T., et al., 1994. Fibroblast growth factor receptor (FGFR): 3.
Alternative splicing in immunoglobulin-like domain III creates a
receptor highly specific for acidic FGF/FGF-1. J. Biol. Chem. 269 (15),
11620 11627.
Chi, C.L., et al., 2003. The isthmic organizer signal FGF8 is required for cell
survival in the prospective midbrain and cerebellum. Development 130
(12), 2633 2644.
Chi, L., et al., 2004. Sprouty proteins regulate ureteric branching by
coordinating reciprocal epithelial Wnt11, mesenchymal Gdnf and stromal
Fgf7 signalling during kidney development. Development 131 (14),
3345 3356.
Ciruna, B., Rossant, J., 2001. FGF signaling regulates mesoderm cell fate
specification and morphogenetic movement at the primitive streak. Dev.
Cell 1 (1), 37 49.
Coffin, J.D., et al., 1995. Abnormal bone growth and selective translational
regulation in basic fibroblast growth factor (FGF-2) transgenic mice. Mol.
Biol. Cell 12, 1861 1873.
Conlon, F.L., et al., 1996. Inhibition of Xbra transcription activation causes
defects in mesodermal patterning and reveals autoregulation of Xbra in
dorsal mesoderm. Development 122 (8), 2427 2435.
Coumoul, X., Deng, C.X., 2003. Roles of FGF receptors in mammalian
development and congenital diseases. Birth Defects Res. Part C Embryo
Today 69 (4), 286 304.
Cox, W.G., Hemmati-Brivanlou, A., 1995. Caudalization of neural fate by
tissue recombination and bFGF. Development 121 (12), 4349 4358.
Crossley, P.H., Martinez, S., Martin, G.R., 1996. Midbrain development
induced by FGF8 in the chick embryo. Nature 380 (6569), 66 68.
Debiais, F., et al., 2001. Fibroblast growth factor-2 (FGF-2) increases Ncadherin expression through protein kinase C and Src-kinase pathways in
human calvaria osteoblasts. J. Cell. Biochem. 81 (1), 68 81.
Delezoide, A.L., et al., 1998. Spatio-temporal expression of FGFR 1, 2 and 3
genes during human embryo-fetal ossification. Mech. Dev. 77 (1), 19 30.
Delany, A.M., Canalis, E., 1998. Dual regulation of stromelysin-3 by
fibroblast growth factor-2 in murine osteoblasts. J. Biol. Chem. 273 (26),
16595 16600.
Dhalluin, C., et al., 2000. 1H, 13C and 15N resonance assignments of the SNT
PTB domain in complex with FGFR1 peptide. J. Biomol. NMR 18 (4),
371 372.
Dikic, I., Giordano, S., 2003. Negative receptor signalling. Curr. Opin. Cell
Biol. 15 (2), 128 135.
Domingos, P.M., et al., 2001. The Wnt/beta-catenin pathway posteriorizes
neural tissue in Xenopus by an indirect mechanism requiring FGF
signalling. Dev. Biol. 239 (1), 148 160.
Dudley, A.T., Ros, M.A., Tabin, C.J., 2002. A re-examination of
proximodistal patterning during vertebrate limb development. Nature
418 (6897), 539 544.
Eblaghie, M.C., et al., 2003. Negative feedback regulation of FGF signaling
levels by Pyst1/MKP3 in chick embryos. Curr. Biol. 13 (12), 1009 1018.
Egan, J.E., et al., 2002. The bimodal regulation of epidermal growth factor
signaling by human Sprouty proteins. Proc. Natl. Acad. Sci. U. S. A. 99 (9),
6041 6046.
Fallon, J.F., et al., 1994. FGF-2: apical ectodermal ridge growth signal for chick
limb development. Science 264 (5155), 104 107.

Fisher, M.E., Isaacs, H.V., Pownall, M.E., 2002. eFGF is required for activation
of XmyoD expression in the myogenic cell lineage of Xenopus laevis.
Development 129 (6), 1307 1315.
Fjeld, C., et al., 2000. Mechanistic basis for catalytic activation of mitogenactivated protein kinase phosphatase 3 by extracellular signal-regulated
kinase. J. Biol. Chem. 275, 6749 6757.
Flauennhaft, R., Moscatelli, D., Rifkin, D.B., 1990. Heparin and heparan
sulfate increase the radius of diffusion and action of basic fibroblast growth
factor. J. Cell Biol. 111, 1651 1659.
Forman-Kay, J.D., Pawson, T., 1999. Diversity in protein recognition by PTB
domains. Curr. Opin. Struct. Biol. 9, 690 695.
Furthauer, M., et al., 2001. Sprouty4 acts in vivo as a feedback-induced antagonist of FGF signaling in zebrafish. Development 128
(12), 2175 2186.
Furthauer, M., et al., 2002. Sef is a feedback-induced antagonist of Ras/MAPKmediated FGF signalling. Nat. Cell Biol. 2, 170 174.
Furthauer, M., et al., 2004. FGF signalling controls the dorsoventral patterning
of the zebrafish embryo. Development 131 (12), 2853 2864.
Gawantka, V., et al., 1998. Gene expression screening in Xenopus identifies
molecular pathways, predicts gene function and provides a global view of
embryonic patterning. Mech. Dev. 77 (2), 95 141.
Goldfarb, M., 1996. Functions of fibroblast growth factors in vertebrate
development. Cytokine Growth Factor Rev. 7 (4), 311 325.
Gonzalez, A.M., et al., 1996. Distribution of fibroblast growth factor (FGF)-2
and FGF receptor-1 messenger RNA expression and protein presence in the
mid-trimester human fetus. Pediatr. Res. 39 (3), 375 385.
Gotoh, Y., 1995. Involvement of the MAP kinase cascade in Xenopus
mesoderm induction. EMBO J. 14 (11), 2491 2498.
Griffin, K., Patient, R., Holder, N., 1995. Analysis of FGF function in normal
and no tail zebrafish embryos reveals separate mechanisms for formation of
the trunk and the tail. Development 121 (9), 2983 2994.
Griffin, K.J., et al., 1998. Molecular identification of spadetail: regulation of
zebrafish trunk and tail mesoderm formation by Tbox genes. Development
125 (17), 3379 3388.
Hacohen, N., et al., 1998. Sprouty encodes a novel antagonist of FGF
signaling that patterns apical branching of the Drosophila airways. Cell
92 (2), 253 263.
Hall, A.B., et al., 2003. hSpry2 is targeted to the ubiquitin-dependent
proteasome pathway by c-Cbl. Curr. Biol. 13 (4), 308 314.
Halpern, M.E., et al., 1993. Induction of muscle pioneers and floor plate is
distinguished by the zebrafish no tail mutation. Cell 75 (1), 99 111.
Hanafusa, H., et al., 2002. Sprouty1 and Sprouty2 provide a control
mechanism for the Ras/MAPK signalling pathway. Nat. Cell Biol. 11,
850 858.
Hemmati-Brivanlou, A., Melton, D., 1997. Vertebrate neural induction. Annu.
Rev. Neurosci. 20, 43 60.
Herrmann, B.G., et al., 1990. Cloning of the T gene required in mesoderm
formation in the mouse. Nature 343 (6259), 617 622.
Holowacz, T., Sokol, S., 1999. FGF is required for posterior neural patterning
but not for neural induction. Dev. Biol. 205 (2), 296 308.
Hongo, I., Kengaku, M., Okamoto, H., 1999. FGF signaling and the anterior
neural induction in Xenopus. Dev. Biol. 216 (2), 561 581.
Isaacs, H.V., Pownall, M.E., Slack, J.M., 1994. eFGF regulates Xbra expression
during Xenopus gastrulation. EMBO J. 13 (19), 4469 4481.
Itoh, N., Ornitz, D.M., 2004. Evolution of the Fgf and Fgfr gene families.
Trends Genet. 20 (n-11), 563 569.
Johnson, D.E., Williams, L.T., 1993. Structural and functional diversity in the
FGF receptor multigene family. Adv. Cancer Res. 60, 1 41.
Johnson, D.E., et al., 1990. Diverse forms of a receptor for acidic and basic
fibroblast growth factors. Mol. Cell. Biol. (9), 4728 4736.
Kawaguchi, H., et al., 1995. Transcriptional induction of prostaglandin
G/H synthase-2 by basic fibroblast growth factor. J. Clin. Invest. 96 (2),
923 930.
Kawakami, Y., et al., 2003. MKP3 mediates the cellular response to FGF8
signalling in the vertebrate limb. Nat. Cell Biol. 5, 513 519.
Kazanskaya, O., Glinka, A., Niehrs, C., 2000. The role of Xenopus dickkopf1
in prechordal plate specification and neural patterning. Development 127
(22), 4892 4981.

B. Thisse, C. Thisse / Developmental Biology 287 (2005) 390 402

Kengaku, M., Okamoto, H., 1995. bFGF as a possible morphogen for the
anteroposterior axis of the central nervous system in Xenopus. Development 121 (9), 3121 3130.
Kim, H.J., Bar-Sagi, D., 2004. Modulation of signalling by Sprouty: a
developing story. Nat. Rev., Mol. Cell Biol. 5 (6), 441 450.
Kim, H.J., et al., 1998. FGF-, BMP- and Shh-mediated signalling pathways in
the regulation of cranial suture morphogenesis and calvarial bone
development. Development 125 (7), 1241 1251.
Kimelman, D., Kirschner, M., 1987. Synergistic induction of mesoderm by
FGF and TGF-beta and the identification of an mRNA coding for FGF in
the early Xenopus embryo. Cell 51 (5), 869 877.
Kimelman, D., et al., 1988. The presence of fibroblast growth factor in the
frog egg: its role as a natural mesoderm inducer. Science 242 (4881),
1053 1056.
Kovalenko, D., et al., 2003. Sef inhibits fibroblast growth factor signaling by
inhibiting FGFR1 tyrosine phosphorylation and subsequent ERK activation. J. Biol. Chem. 278 (16), 14087 14091.
Kouhara, H., et al., 1997. A lipid-anchored Grb2-binding protein that links
FGF-receptor activation to the Ras/MAPK signaling pathway. Cell 89 (5),
693 702.
Kroll, K.L., Amaya, E., 1996. Transgenic Xenopus embryos from sperm
nuclear transplantations reveal FGF signaling requirements during gastrulation. Development 122 (10), 3173 3183.
Kubota, Y., Itoh, K., 2000. Chemotactic migration of mesencephalic neural
crest cells in the mouse. Dev. Dyn. 217, 170 179.
Kudoh, T., et al., 2001. A gene expression screen in zebrafish embryogenesis.
Genome Res. 12, 1979 1987.
Kudoh, T., Wilson, S.W., Dawid, I.B., 2002. Distinct roles for Fgf, Wnt and
retinoic acid in posteriorizing the neural ectoderm. Development 129 (18),
4335 4346.
LaBonne, C., 1995. Role of MAP kinase in mesoderm induction and
axial patterning during Xenopus development. Development 121 (5),
1475 1486.
Lamb, T.M., Harland, R.M., 1995. Fibroblast growth factor is a direct neural
inducer, which combined with noggin generates anterior posterior neural
pattern. Development 121 (11), 3627 3636.
Laufer, E., et al., 1994. Sonic hedgehog and FGF-4 act through a signaling
cascade and feedback loop to integrate growth and patterning of the
developing limb bud. Cell 79 (6), 993 1003.
Launay, C., et al., 1996. A truncated FGF receptor blocks neural induction by
endogenous Xenopus inducers. Development 122 (3), 869 880.
Lee, P.L., et al., 1989. Purification and complementary DNA cloning of a
receptor for basic fibroblast growth factor. Science 245 (4913), 57 60.
Lewandoski, M., Sun, X., Martin, G.R., 2000. Fgf8 signalling from the AER is
essential for normal limb development. Nat. Genet. 26 (4), 460 463.
Li, S., Muneoka, K., 1999. Cell migration and chick limb development
chemotactic action of FGF4 and the AER. Dev. Biol. 211, 335 347.
Lin, X., 2004. Functions of heparan sulfate proteoglycans in cell signaling
during development. Development 131 (24), 6009 6021.
Lin, X., et al., 1999. Heparan sulfate proteoglycans are essential for FGF
receptor signaling during Drosophila embryonic development. Development 126 (17), 3715 3723.
Linker, C., Stern, C., 2004. Neural induction requires BMP inhibition only as a
late step, and involves signals other than FGF and Wnt antagonists.
Development 131 (22), 5671 5681.
Liu, A., Joyner, A.L., 2001. Early anterior/posterior patterning of the midbrain
and cerebellum. Ann. Rev. Neurosci. 24, 869 896.
Liu, Z., et al., 2002. Coordination of chondrogenesis and osteogenesis by
fibroblast growth factor 18. Genes Dev. 16 (7), 859 869.
Liu, A., et al., 2003. FGF17b and FGF18 have different midbrain regulatory
properties from FGF8b or activated FGF receptors. Development 130 (25),
6175 6185.
MacNicol, A.M., et al., 1993. Raf-1 kinase is essential for early Xenopus
development and mediates the induction of mesoderm by FGF. Cell 73 (3),
571 583.
Mailleux, A.A., et al., 2001. Evidence that SPROUTY2 functions as an
inhibitor of mouse embryonic lung growth and morphogenesis. Mech. Dev.
102 (1 2), 81 94.

401

Mansukhani, A., et al., 2000. Signaling by fibroblast growth factors (FGF) and
fibroblast growth factor receptor 2 (FGFR2)-activating mutations blocks
mineralization and induces apoptosis in osteoblasts. J. Cell Biol. 149 (6),
1297 1308.
Marie, P.J., 2003. Fibroblast growth factor signaling controlling osteoblast
differentiation. Gene 316, 23 32.
Martin, G.R., 1998. The roles of FGFs in the early development of vertebrate
limbs. Genes Dev. 12 (11), 1571 1586.
Maruoka, Y., et al., 1998. Comparison of the expression of three highly related
genes, FGF8, FGF17 and FGF18, in the mouse embryo. Mech. Dev. 74
(1 2), 175 177.
McGrew, L.L., Hoppler, S., Moon, R.T., 1997. Wnt and FGF pathways
cooperatively pattern anteroposterior neural ectoderm in Xenopus. Mech.
Dev. 69 (1 2), 105 114.
Meyers, E.N., Lewandoski, M., Martin, G.R., 1998. An FGF8 mutant allelic
series generated by Cre- and Flp-mediated recombination. Nat. Genet. 2,
136 141.
Miki, T., et al., 1992. Determination of ligand-binding specificity by alternative
splicing: two distinct growth factor receptors encoded by a single gene.
Proc. Natl. Acad. Sci. U. S. A. 89 (1), 246 250.
Minowada, G., et al., 1999. Vertebrate Sprouty genes are induced by FGF
signaling and can cause chondrodysplasia when overexpressed. Development 126 (20), 4465 4475.
Molteni, A., et al., 1999. Differential expression of fibroblast growth factor
receptor-1, -2, and -3 and syndecan-1, -2, and -4 in neonatal rat mandibular
condyle and calvaria during osteogenic differentiation in vitro. Bone 4,
337 347.
Montero, A., et al., 2000. Disruption of the fibroblast growth factor-2 gene
results in decreased bone mass and bone formation. J. Clin. Invest. 105 (8),
1085 1093.
Moon, A.M., Capecchi, M.R., 2000. Fgf8 is required for outgrowth and
patterning of the limbs. Nat. Genet. 26 (4), 455 459.
Moscatelli, D., 1987. High and low affinity binding sites for basic fibroblast
growth factor on cultured cells: absence of a role for low affinity binding in
the stimulation of plasminogen activator production by bovine capillary
endothelial cells. J. Cell. Physiol. 131 (1), 123 130.
Munchberg, S.R., Ober, E.A., Steinbeisser, H., 1999. Expression of the Ets
transcription factors erm and pea3 in early zebrafish development. Mech.
Dev. 88 (2), 233 236.
Nakato, H., Kimata, K., 2002. Heparan sulfate fine structure and
specificity of proteoglycan functions. Biochim. Biophys. Acta 21573
(3), 312 318.
Niehrs, C., Meinhardt, H., 2002. Modular feedback. Nature 417 (6884), 35 36.
Niswander, L., 2003. Pattern formation: old models out on a limb. Nat. Rev. 4,
133 143.
Nutt, S.L., et al., 2001. Xenopus Sprouty2 inhibits FGF-mediated gastrulation
movements but does not affect mesoderm induction and patterning. Genes
Dev. 15 (9), 1152 1166.
Nybakken, K., Perrimon, N., 2002. Heparan sulfate proteoglycan modulation of
developmental signaling in Drosophila. Biochim. Biophys. Acta 1573 (3),
280 291.
Ohbayashi, N., et al., 2002. FGF18 is required for normal cell proliferation and
differentiation during osteogenesis and chondrogenesis. Genes Dev. 16 (7),
870 879.
Ornitz, D.M., 2000. FGFs, heparan sulfate and FGFRs: complex interactions
essential for development. BioEssays 2, 108 112.
Ornitz, D.M., Itoh, N., 2001. Fibroblast growth factors. Genome Biol. 2 (3),
1 12.
Orr-Urtreger, A., et al., 1991. Developmental expression of two murine
fibroblast growth factor receptors, flg and bek. Development 113 (4),
1419 1434.
Park, W.Y., Miranda, B., Lebeche, D., Hashimoto, G., Cardoso, W.V., 1998.
FGF10 is a chemotactic factor for distal epithelial buds during lung
development. Dev. Biol. 201, 125 134.
Pawson, T., 1995. Protein modules and signalling networks. Nature 373 (6515),
573 580.
Pellegrini, L., 2001. Role of heparan sulfate in fibroblast growth factor
signalling: a structural view. Curr. Opin. Struct. Biol. 11 (5), 629 634.

402

B. Thisse, C. Thisse / Developmental Biology 287 (2005) 390 402

Pellegrini, L., et al., 2000. Crystal structure of fibroblast growth factor receptor
ectodomain bound to ligand and heparin. Nature 407 (6807), 1029 1034.
Powers, C.J., McLeskey, S.W., Wellstein, A., 2000. Fibroblast growth factors,
their receptors and signaling. Endocr. Relat. Cancer 3, 165 197.
Preger, E., 2004. Alternative splicing generates an isoform of the human Sef
gene with altered subcellular localization and specificity. Proc. Natl. Acad.
Sci. U. S. A. 101 (5), 1229 1234.
Reich, A., Sapir, A., Shilo, B., 1999. Sprouty is a general inhibitor of receptor
tyrosine kinase signaling. Development 126 (18), 4139 4147.
Reifers, F., 1998. FGF8 is mutated in zebrafish acerebellar (ace) mutants and is
required for maintenance of midbrain hindbrain boundary development
and somitogenesis. Development 125 (13), 2381 2395.
Ribisi Jr., S., et al., 2000. Ras-mediated FGF signaling is required for the
formation of posterior but not anterior neural tissue in Xenopus laevis. Dev.
Biol. 227 (1), 183 196.
Rousseau, F., et al., 1994. Mutations in the gene encoding fibroblast growth
factor receptor-3 in achondroplasia. Nature 371 (6494), 252 254.
Rubin, C., Litvak, V., Medvedovsky, H., Zwang, Y., Lev, S., Yarden, Y., 2003.
Sprouty fine-tunes EGF signaling through interlinked positive and negative
feedback loops. Curr. Biol. 13 (4), 297 307.
Sasaki, A., et al., 2003. Mammalian Sprouty4 suppresses Ras-independent
ERK activation by binding to Raf1. Nat. Cell Biol. 5, 427 432.
Sato, T., Araki, I., Nakamura, H., 2001. Inductive signal and tissue
responsiveness defining the tectum and the cerebellum. Development 128,
2461 2469.
Sharrocks, A.D., 2001. The ETS-domain transcription factor family. Nat. Rev.,
Mol. Cell Biol. 2 (11), 827 837.
Schlessinger, J., et al., 2000. Crystal structure of a ternary FGF FGFR
heparin complex reveals a dual role for heparin in FGFR binding and
dimerization. Mol. Cell 3, 743 750.
Schmid, B., et al., 2000. Equivalent genetic roles for bmp7/snailhouse and
bmp2b/swirl in dorsoventral pattern formation. Development 127 (5),
957 967.
Schulte-Merker, S., Smith, J.C., 1995. Mesoderm formation in response to
Brachyury requires FGF signalling. Curr. Biol. 5 (1), 62 67.
Sheng, G., Dos Reis, M., Stern, C.D., 2003. Churchill, a zinc finger
transcriptional activator, regulates the transition between gastrulation and
neurulation. Cell 115 (5), 603 613.
Shiang, R., et al., 1994. Mutations in the transmembrane domain of FGFR3
cause the most common genetic form of dwarfism, achondroplasia. Cell 78
(2), 335 342.
Slack, J.M., et al., 1987. Mesoderm induction in early Xenopus embryos by
heparin-binding growth factors. Nature 326 (6109), 197 200.
Smith, J.C., et al., 1991. Expression of a Xenopus homolog of Brachyury
(T) is an immediate-early response to mesoderm induction. Cell 67 (1),
79 87.
Stanley, H.J., Zorn, A.M., Gurdon, J.B., 2001. eFGF and its mode of action in
the community effect during Xenopus myogenesis. Development 128 (8),
1347 1357.
Sternberg, P.W., Alberola-Ila, J., 1998. Conspiracy theory: RAS and RAF do
not act alone. Cell 95 (4), 447 450.
Stewart, A.E., et al., 1999. Crystal structure of the MAPK phosphatase Pyst1
catalytic domain and implications for regulated activation. Nat. Struct. Biol.
6, 174 181.
Storey, K.G., et al., 1998. Early posterior neural tissue is induced by FGF in the
chick embryo. Development 125 (3), 473 484.
Streit, A., et al., 2000. Initiation of neural induction by FGF signalling before
gastrulation. Nature 406 (6791), 74 78.
Strong, C.F., et al., 2000. Xbra3 induces mesoderm and neural tissue in
Xenopus laevis. Dev. Biol. 222 (2), 405 419.
Summerbell, D., Lewis, J.H., Wolpert, L., 1973. Positional information in chick
limb morphogenesis. Nature 244 (5417), 492 496.
Sun, X., et al., 1999. Targeted disruption of Fgf8 causes failure of cell migration
in the gastrulating mouse embryo. Genes Dev. 13 (14), 1834 1846.

Sun, X., Mariani, F.V., Martin, G., 2002. Functions of FGF signaling
from the apical ectodermal ridge in limb development. Nature 418,
501 508.
Thisse, B., et al., 2001, Expression of the zebrafish genome during embryogenesis, ZFIN. http://zfin.org/cgibin/webdriver?MIval=aapubview2.apg_
and_OID=ZDB-PUB-010810-1.
Torii, S., et al., 2004. Sef is a spatial regulator for Ras/MAP kinase signaling.
Dev. Cell 7 (1), 33 44.
Tsang, M., Dawid, I.B., 2004. Promotion and attenuation of FGF signaling
through the Ras-MAPK pathway. Sci. STKE 228, e17.
Tsang, M., et al., 2002. Identification of Sef, a novel modulator of FGF
signalling. Nat. Cell Biol. 4 (2), 165 169.
Tsang, M., et al., 2004. A role for MKP3 in axial patterning of the zebrafish
embryo. Development 131 (12), 2769 2779.
Ueda, Y., et al., 1996. Protein kinase C activates the MEK-ERK pathway in
a manner independent of Ras and dependent on Raf. J. Biol. Chem. 271
(38), 23512 23519.
Umbhauer, M., et al., 1995. Mesoderm induction in Xenopus caused by
activation of MAP kinase. Nature 376 (6535), 58 62.
Umbhauer, M., et al., 2000. Signaling specificities of fibroblast growth factor
receptors in early Xenopus embryo. J. Cell Sci. 113 (16), 2865 2875.
Varghese, S., et al., 1995. Basic fibroblast growth factor stimulates expression
of interstitial collagenase and inhibitors of metalloproteinases in rat bone
cells. Endocrinology 136 (5), 2156 2162.
Warren, S.M., et al., 2003. The BMP antagonist noggin regulates cranial suture
fusion. Nature 422 (6932), 625 629.
Wasylyk, B., Hagman, J., Gutierrez-Hartmann, A., 1998. Ets transcription
factors: nuclear effectors of the Ras-MAP-kinase signaling pathway. Trends
Biochem. Sci. 6, 213 216.
Webb, S.E., et al., 1997. Fibroblast growth factors 2 and 4 stimulate
migration of mouse embryonic limb myogenic cells. Dev. Dyn. 209,
206 216.
Weinstein, D.C., Hemmati-Brivanlou, A., 1999. Neural induction. Annu. Rev.
Cell Dev. Biol. 15, 411 433.
Whitman, M., Melton, D.A., 1992. Involvement of p21ras in Xenopus
mesoderm induction. Nature 357 (6375), 252 254.
Wilson, S.I., et al., 2000. An early requirement for FGF signalling in the
acquisition of neural cell fate in the chick embryo. Curr. Biol. 10 (8),
421 429.
Wong, E.S., et al., 2002. Sprouty2 attenuates epidermal growth factor receptor
ubiquitylation and endocytosis, and consequently enhances Ras/ERK
signalling. EMBO J. 21 (18), 4796 4808.
Xiong, S., et al., 2003. hSef inhibits PC-12 cell differentiation by interfering
with Ras-mitogen-activated protein kinase MAPK signaling. J. Biol. Chem.
278 (50), 50273 50282.
Xu, X., et al., 1998. Fibroblast growth factor receptor 2 (FGFR2)-mediated
reciprocal regulation loop between FGF8 and FGF10 is essential for limb
induction. Development 125 (4), 753 765.
Xu, J., Liu, Z., Ornitz, D.M., 2000. Temporal and spatial gradients of FGF8 and
FGF17 regulate proliferation and differentiation of midline cerebellar
structures. Development 9, 1833 1843.
Yang, X., et al., 2002. Cell movement patterns during gastrulation in the chick
are controlled by positive and negative chemotaxis mediated by FGF4 and
FGF8. Dev. Cell 3 (3), 425 437.
Yusoff, P., et al., 2002. Sprouty2 inhibits the Ras/MAP kinase pathway by
inhibiting the activation of Raf. J. Biol. Chem. 277 (5), 3195 3201.
Zhang, J., et al., 2003. A bipartite mechanism for ERK2 recognition by its
cognate regulators and substrates. J. Biol. Chem. 278, 29901 29912.
Zhao, Y., Zhang, Z.Y., 2001. The mechanism of dephosphorylation of
extracellular signal-regulated kinase 2 by mitogen-activated protein kinase
phosphatase 3. J. Biol. Chem. 276, 32382 32391.
Zhao, J., Cao, Y., Zhao, C., Postlethwait, J., Meng, A., 2003. An SP1-like
transcription factor Spr2 acts downstream of Fgf signaling to mediate
mesoderm induction. EMBO J. 22 (22), 6078 6088.