5611
Development 126, 5611-5620 (1999)
Printed in Great Britain © The Company of Biologists Limited 1999
DEV4184
Fgfr1 and Fgfr2 have distinct differentiation- and proliferation-related roles in
the developing mouse skull vault
S. Iseki1,2, A. O. M. Wilkie3 and G. M. Morriss-Kay1,*
1Department
2Department
of Human Anatomy and Genetics, South Parks Road, Oxford OX1 3QX, UK
of Molecular Craniofacial Embryology, Tokyo Medical and Dental University, 1-5-45, Yushima, Bunkyo-ku, Tokyo 113-
8549, Japan
3Institute of Molecular Medicine, John Radcliffe Hospital, Headington, Oxford OX3 9DS, UK
*Author for correspondence (e-mail: [email protected])
Accepted 22 September; published on WWW 24 November 1999
SUMMARY
Fibroblast growth factor receptors (FGFRs) play major
roles in skeletogenesis, and activating mutations of the
human FGFR1, FGFR2 and FGFR3 genes cause
premature fusion of the skull bones (craniosynostosis). We
have investigated the patterns of expression of Fgfr1, Fgfr2
and Fgfr3 in the fetal mouse head, with specific reference
to their relationship to cell proliferation and differentiation
in the frontal and parietal bones and in the coronal suture.
Fgfr2 is expressed only in proliferating osteoprogenitor
cells; the onset of differentiation is preceded by downregulation of Fgfr2 and up-regulation of Fgfr1. Following
up-regulation of the differentiation marker osteopontin,
Fgfr1, osteonectin and alkaline phosphatase are downregulated, suggesting that they are involved in the
osteogenic differentiation process but not in maintaining
the differentiated state. Fgfr3 is expressed in the cranial
cartilage, including a plate of cartilage underlying the
coronal suture, as well as in osteogenic cells, suggesting a
dual role in skull development. Subcutaneous insertion of
FGF2-soaked beads onto the coronal suture on E15
resulted in up-regulation of osteopontin and Fgfr1 in the
sutural mesenchyme, down-regulation of Fgfr2, and
inhibition of cell proliferation. This pattern was observed
at 6 and 24 hours after bead insertion, corresponding to the
timing and duration of FGF2 diffusion from the beads. We
suggest (a) that a gradient of FGF ligand, from high levels
in the differentiated region to low levels in the environment
of the osteogenic stem cells, modulates differential
expression of Fgfr1 and Fgfr2, and (b) that signalling
through FGFR2 regulates stem cell proliferation whereas
signalling
through
FGFR1
regulates
osteogenic
differentiation.
INTRODUCTION
experimental and clinical evidence that calvarial growth is
regulated by a complex interaction of signalling systems and
transcription factors including MSX2, SHH, BMPs, and
TGFβs (Kim et al., 1998; Liu et al., 1999; Dodig et al., 1999;
Opperman et al., 1997).
Each skull suture is a growth centre, and it is clear from both
gene expression and genetic studies in mouse and human that
the different sutures share some signalling processes but also
show some specificity. These differences are reflected in the
different incidences of synostosis of single sutures, of which
sagittal synostosis is the most common (50-60%), followed by
synostosis of one or both coronal sutures (17-30%) (Wall,
1997).
The importance of FGF/FGFR signalling in human skull
development has been revealed by the identification of
mutations in FGFR genes in several craniosynostosis
syndromes, some of which involve multiple or all sutures
(reviewed by Wilkie, 1997). Mutations affecting the IgIIIc
region of FGFR2 (expressed in the BEK isoform of the protein)
Development of the normal skull vault (calvaria) requires
mechanisms to ensure that both its morphology and its rate of
growth are precisely matched to those of the developing brain.
This precise relationship suggests that there are important
tissue interactions between the brain and skeletogenic
membrane, also involving the mesenchymal layers between
them (the developing meninges). Very little is known about
these interactions, although there is evidence that the
membranous tissue underlying the sutures is specifically
required for the maintenance of sutural growth (Opperman et
al., 1993; Kim et al., 1998). However, the signalling pathways
acting within the skeletogenic tissues to regulate calvarial
growth are beginning to be elucidated, and it may be that some
of these are also involved in the mechanisms that integrate
growth of the skull and brain. In addition to the roles of FGFR
signalling that are the subject of this study, and the possibly
related role of the transcription factor TWIST, there is good
Key words: FGFR1, FGFR2, FGFR3, Osteogenesis, Proliferation,
Differentiation, Skull, Suture, Mouse fetus, Human
5612 S. Iseki, A. O. M. Wilkie and G. M. Morriss-Kay
cause Crouzon, Pfeiffer, and other multi-suture syndromes. In
addition, mutations of equivalent amino acids in the IgII/IgIII
linker region of FGFR1-3 paralogues are associated with
different syndromes, suggesting that all three genes play
essential but specific roles in skull development: the Pro252Arg
mutation of FGFR1 causes a milder form of Pfeiffer syndrome,
the Pro253Arg mutation of FGFR2 causes Apert syndrome, in
which there is bony syndactyly of the hands and feet as well as
a severe form of craniosynostosis, and the Pro250Arg mutation
of FGFR3 is associated with unicoronal or bicoronal synostosis
(Muenke et al., 1994; Wilkie et al., 1995; Bellus et al., 1996).
The phenotype of the Pro250Arg FGFR3 mutation is similar to
that of Saethre-Chotzen syndrome, which is genetically
characterised by mutation of TWIST (El Ghouzzi et al., 1997;
Howard et al., 1997; Paznekas et al., 1998). TWIST encodes a
basic helix-loop-helix transcription factor whose Drosophila
orthologue, twist, regulates transcription of heartless (htl,
formerly DFR1 or DFGF-R2), a homologue of vertebrate
FGFRs that is expressed in mesodermal tissues (Shishido et al.,
1993; Casal and Leptin, 1996; Beiman et al., 1996).
These observations from clinical molecular genetics,
together with the genetic information from Drosophila, provide
important insights into the molecular mechanisms governing
normal and abnormal development of the mammalian skull. In
this study, we use this information to gain a better
understanding of the developmental biology of the processes
involved. We have chosen to investigate the coronal suture in
detail because it is the only suture that is formed from the
outset as a close apposition of two adjacent bone domains. It
is also an ideal suture for investigating the different roles of
FGFR1, FGFR2 and FGFR3 in calvarial development, since it
is involved (sometimes specifically) in even the mildest forms
of craniosynostosis resulting from mutations of all three
receptors (Bellus et al., 1996).
In a previous study of prenatal development of the coronal
suture in the mouse (Iseki et al., 1997), we showed that Fgfr2 is
expressed in proliferating osteogenic cells, and is mutually
exclusive with cells expressing osteopontin, a gene involved in
osteogenic differentiation. Here we develop this theme by
describing the expression patterns of all four mouse Fgfr genes,
and analysing the relationship between the expression of Fgfr1,
Fgfr2 and Fgfr3, cell proliferation and cell differentiation. We also
test the effects of increasing FGFR signalling by local application
of FGF2-soaked beads. The results confirm our previous
observation that Fgfr2 is expressed only in proliferating
osteogenic stem cells, and reveal that Fgfr2 expression and cell
proliferation are both lost when FGF2 levels are increased. In
contrast, Fgfr1 expression is associated with the differentiation
process in both normal and experimental conditions. Fgfr3 is
expressed in both the osteogenic and chondrogenic regions of the
skeletogenic membrane, including a thin plate of cartilage
underlying part of the coronal suture, suggesting that its role in
sutural development may involve tissue interactions. For
completeness, the expression of Fgfr4 was also detected; as
expected, it is confined to the cranial musculature. Comparison of
the timing and extent of diffusion of FGF2 from the beads supports
the interpretation that a concentration gradient of ligand from the
differentiated bone domains to the suture plays an essential role
in maintaining the proliferation-differentiation balance in normal
skull growth. The effects of increased levels of ligand are
consistent with the idea that signalling through both FGFR1 and
FGFR2 has a regulatory function in relation to the behaviour of
osteogenic cells, FGFR2 being involved in regulating cell
proliferation, and FGFR1 in regulating differentiation.
MATERIALS AND METHODS
Animals
C57BL/6 mouse fetuses (plug day = E0) were used for all
experiments.
Probes for in situ hybridisation
The following cDNAs were used: (1) a 3.8 kb fragment of mouse
Fgfr1 (flg/IIIc splice variant) cloned into pBluescript KS(+)
(Stratagene); (2) a 1.96 kb fragment of mouse Fgfr2 (Bek/IIIc splice
variant, including the whole extracellular and transmembrane
domains) cloned into pBluescript KS(+) (Stratagene); (3) a full length
mouse Fgfr3 (IIIc splice variant) cloned into pBluescript SK(+)
(Stratagene); (4) a 583 bp 3′ untranslated region fragment of mouse
Fgfr4 cloned into pGEM 4 (Promega); (5) a 984 bp HindIII fragment
of mouse osteopontin cloned into pGEM1 and pGEM2 (Promega); (6)
a 1.5 kb fragment of mouse osteonectin cloned into pBluescript I
KS(−) (Stratagene); (7) a 784 bp rat alkaline phosphatase cDNA
cloned into pBluescript KS(−) (Stratagene). To generate antisense and
sense transcripts the plasmids were linearised and transcribed as
described previously (Iseki et al., 1997) using T3, T7 or Sp6 RNA
polymerase with digoxigenin or fluorescein RNA mixture (Roche); all
transcripts except Fgfr4 were reduced to an average size of 200 bases
by limited alkaline hydrolysis.
In situ hybridisation
Single whole-mount in situ hybridisation was carried out as previously
described (Iseki et al., 1997). In situ hybridisation on sections were
carried out on fresh frozen horizontal sections of fetal heads. The
sections were fixed in 4% paraformaldehyde for 10 minutes followed
by acetylation, and hybridised in hybridisation buffer with probes
overnight. For double-labelled in situ hybridisation, a digoxigeninlabelled probe and a fluorescein-labelled probe were used and
detected by anti-digoxigenin-rhodamine (Roche) and anti-fluoresceinalkaline phosphatase (Roche), respectively.
Histochemical detection of alkaline phosphatase
Histochemical detection of alkaline phosphatase was carried out on
fresh frozen sections (Drury and Wallington, 1967). The sections were
fixed in calcium formalin for 1.5-2.0 hours. After being washed in
PBS, the sections were incubated in reaction solution at 37°C for 30
minutes followed by PBS wash and 2% aqueous cobalt nitrate. Colour
was developed in 1% ammonium sulphate solution for 1 minute.
BrdU immunohistochemical detection after in situ
hybridisation
BrdU solution (100 mg/kg) was injected intraperitoneally into
pregnant mice 2 hours before fixation of fetuses used for double
detection of BrdU and Fgfr mRNA. After in situ hybridisation the
sections were post-fixed in 4% paraformaldehyde solution for 20
minutes followed by microwave treatment in citrate buffer. After
blocking, the sections were incubated with anti-BrdU antibody
(Roche), biotinylated anti-mouse IgG antibody (Vector), and
Vectastain ABC (Vector) followed by the diaminobenzidine (DAB,
Sigma) reaction.
Subcutaneous insertion of FGF2 beads in fetal heads
Heparin acrylic beads (Sigma), 125-150 µm diameter, were soaked in
PBS or in human FGF2 (400 µg/ml in PBS) or in digoxigenin-labelled
FGF2 (400 µg/ml) (gifts from John Heath) for at least 60 minutes at
room temperature and rinsed twice with PBS prior to implantation.
Fgfr1 and Fgfr2 functions in skull development 5613
was carried out in citrate buffer to unmask the epitope. After the
blocking, PC10 antibody was applied and subsequently detected by
biotinylated anti-mouse IgG antibody (Vector) and Vectastain ABC
(Vector) followed by the DAB (Sigma) reaction.
RESULTS
Fig. 1. Schematic representation of an E16 mouse head. (A) Side
view showing section position and orientation in relation to the
frontal (f) and parietal (p) bone domains (green), and the coronal
suture between them; cartilage (c) that is included in sections near to
the base of the suture is shown in brown. (B) The major features
shown in the sections: black, skin and brain; grey, skeletogenic
membrane; green, osteogenic domains. The rectangle indicates the
area shown in the sections.
Surgery was carried out on E15 fetuses as described previously (Iseki
et al., 1997). The fetuses were collected from the dam after 2, 4, 6 or
24 hours and directly embedded into CRYO-M-BED (Bright
Instrument Company) or fixed for whole-mount in situ hybridisation.
Digoxigenin was detected as described above, after heat-inactivation
of endogenous alkaline phosphatase. Skeletal staining was carried out
as described previously (Iseki et al., 1997).
Detection of proliferating cells by anti-proliferating cell
nuclear antigen (PCNA)
Anti-PCNA antibody (PC10; Santa-Cruz) detects PCNA p36 protein
in cells in G1 as well as S phase. Fresh frozen sections were fixed in
4% fresh paraformaldehyde for 30 minutes and microwave treatment
Fig. 2. Expression of Fgfrs (A-H) and
osteogenic markers (I-L), as indicated, in the
right coronal suture region at E16; skin (s) is
at the top, brain (b) at the bottom. A-G:
Fgfr1-3 transcripts are all located in both the
frontal (f) and parietal (p) bone domains;
Fgfr3 is expressed at very low levels in these
sites, requiring a long reaction time for
detection; it is expressed at higher levels in a
thin layer of cartilage underlying the
osteogenic layer of the skeletogenic
membrane in the lower part of the suture, i.e.
close to the skull base (G). (H) Fgfr4 is
expressed only in differentiating muscle (m).
(I,L) Histochemical detection of alkaline
phosphatase shows that this extracellular
protein has a similar distribution to
osteonectin transcripts (K) but also diffuses
into the sutural mesenchyme (arrow in L, in
a section prepared using a longer reaction
time than that shown in I). Osteopontin
expression (J) is restricted to sites in which
osteogenesis is more advanced compared
with sites of expression of osteonectin and
localisation of alkaline phosphatase. b, brain;
c, cartilage; f, frontal bone domain; m,
muscle; p, parietal bone domain; s, skin.
Scale bars represent 150 µm in A-C and GK, 100 µm in D-F, and 50 µm in L.
Fgfr expression and osteogenic differentiation in the
skeletogenic membrane
Gene expression patterns of Fgfr1, Fgfr2, Fgfr3 and Fgfr4
were analysed by in situ hybridisation on sections of E16
mouse heads as indicated in Fig. 1. The coronal suture is the
only suture to show a close relationship between its component
bones ab initio, having a small overlap between the frontal and
parietal bone-forming tissue domains, the parietal bone always
being on the outer side of the overlap. The other skull vault
sutures are represented by broad areas of undifferentiated
skeletogenic membrane in E16 embryos. In all of the sections
illustrated, skin is at the top and brain at the bottom, and the
right coronal suture is shown (except for Fig. 4A-F, which
show part of the right frontal bone).
The expression of Fgfr1, Fgfr2 and Fgfr3 in the coronal
suture region is illustrated in Fig. 2, and compared with
expression patterns of the osteogenic differentiation-related
genes osteopontin and osteonectin, and with histochemical
detection of alkaline phosphatase. Fgfr1 is expressed in the
developing bone domains in cells that appear to be close to or
within the osteoid (Fig. 2A,D). Cells expressing Fgfr2 (Fig.
2B,E) are more distant from the osteoid, both above and below
each plate of developing bone, and extending further into the
suture. Fgfr3 (Fig. 2C,F) is expressed in osteogenic cells at the
5614 S. Iseki, A. O. M. Wilkie and G. M. Morriss-Kay
Fig. 3. Comparison of the patterns of cell proliferation and Fgfr1
(A) or Fgfr2 (B) expression in the E16 coronal suture region, as
indicated by histochemical detection of BrdU uptake (brown) and in
situ hybridisation to detect Fgfr transcripts (purple). The BrdU
uptake pattern coincides with Fgfr2 expression but is external to the
frontal (f) and parietal (p) domains of Fgfr1 expression. b, brain; f,
frontal domain; p, parietal domain; arrow, position of the coronal
suture. Scale bar, 100 µm.
periphery of the osteoid of the future frontal and parietal bones,
particularly on the outer (skin side) surface of the developing
bones, with extension around the sutural edge of the osteoid
plates; this osteogenic expression of Fgfr3 was only detected
after a prolonged period of colour development. In contrast,
high levels of Fgfr3 are expressed in the thin layer of cartilage
underlying the lower part of the coronal suture (Fig. 2G, colour
development time equal to that for Fig. 2A,B,D and E); this
cartilaginous plate extends from the skull base to about halfway up the suture (Fig. 2A-F show sections above this level).
Fgfr4 is not expressed in the skeletogenic membrane, being
specific to the developing muscles of the scalp and occipital
region (Fig. 2H), in accordance with previous observations
(Stark et al., 1991).
The differentiation-related genes osteopontin and
osteonectin are, as expected, expressed only in the developing
bone domains, but their expression patterns are not identical.
Osteopontin expression is confined to the region of osteoid
deposition (Fig. 2J); osteonectin (Fig. 2K) is preferentially
expressed in cells on the outer (skin side) surface of the
osteoid, and in preosteoblasts at the sutural margin of the
osteoid plate (i.e. further into the suture than cells expressing
osteopontin). Alkaline phosphatase protein is histochemically
detectable in roughly the same sites as osteonectin transcripts
in relation to the coronal suture, but throughout the thickness
of the osteoid plate (Fig. 2I). In order to reveal low levels of
alkaline phosphatase activity, some specimens were allowed
to develop for longer; this revealed that mid-sutural
mesenchyme cells, in which osteogenic differentiation is not
taking place, are also exposed to this osteogenesis-related
protein (Fig. 2L). Alkaline phosphatase gene expression was
also detected by in situ hybridisation (not shown); it is
identical to that of osteonectin, indicating that the midsutural
Fig. 4. The relationship between Fgfr1 expression and osteogenic
differentiation in E16 skull bones; the outer (skin) side is uppermost
in all panels. (A-C) In situ hybridisation of Fgfr1 (A), osteopontin
(B), and osteonectin (C) in the mineralising trabecular area of the
frontal bone. Fgfr1 and osteonectin transcripts are present in the
outer layer of newly differentiating cells (asterisks); osteopontin
transcripts are absent from these cells (arrow in B). (DF) Relationship between expression of Fgfr1 (D) and osteopontin (E)
in the trabecular part of the frontal bone: double-labelled in situ
hybridization. Superimposition of the images (F) shows that the
outermost cells show only Fgfr1 expression (arrows) and the
innermost cells show only osteopontin expression (arrowheads); cells
intermediate in position show expression of both genes. (GI) Double-labelled in situ hybridization of Fgfr1 (G) and osteonectin
(H) in the coronal suture region. Superimposition of the images (I)
reveals that cells expressing only Fgfr1 (arrows) extend further into
the sutural mesenchyme than those expressing both genes. Scale bar,
100 µm in A-C, 50 µm in D-I.
alkaline phosphatase activity is due to diffusion from the
differentiated region.
Fgfr expression and cell proliferation in the
skeletogenic membrane
BrdU was immunohistochemically detected in cells outlining
the parietal and frontal bone domains of the E16 skull; these
two overlapping domains are separated by a thin area of
mid-sutural mesenchyme in which there were very few
stained nuclei (Fig. 3A,B). BrdU immunohistochemistry was
combined on the same sections with in situ hybridisation for
either Fgfr1 or Fgfr2. Fgfr1 expression was seen to be internal
to the BrdU-positive cells (Fig. 3A), whereas Fgfr2 expression
and BrdU staining were observed in the same cell layer (Fig.
3B). This result indicates that (a) Fgfr2 expression within the
skeletogenic membrane is characteristic of proliferating cells
at the sutural, inner and outer margins of the calvarial bone
domains, confirming our previous observations (Iseki et al.,
Fgfr1 and Fgfr2 functions in skull development 5615
Fig. 5. Summary of Fgfr and osteopontin expression patterns in the
coronal suture region at E16 from least differentiated (outer) to most
differentiated (inner) positions.
1997), and (b) that Fgfr1 expression is characteristic of cells
that are not actively proliferating.
The relationship between Fgfr1 expression and
differentiation
In order to gain a more detailed picture of the relationship
between different gene expression patterns and the osteogenic
differentiation process, the expression of Fgfr1, osteopontin
and osteonectin was observed at high magnification and in
double-hybridisation preparations (Fig. 4). The area in which
differentiation is most advanced in E16 fetal heads is the
frontal bone, which is at this stage already thickened and
trabecular, with gaps between the areas of osteoid, parts of
which are mineralised (Iseki et al., 1997). Fgfr1 is expressed
mainly in osteoblasts attached to the osteoid, and in only a
small number of the osteocytes (cells completely embedded
within the matrix) (Fig. 4A). Expression of Fgfr1 is stronger
on the outer (skin side) surface of the developing bony plate
than on the inner (brain side) surface. Osteopontin is not
expressed in the outermost cells, but transcripts are present in
the layer of cells internal to them (Fig. 4B). Some osteocytes
show osteopontin expression, but others are negative for all
three RNA species. The pattern of osteonectin expression
resembles that of Fgfr1 more closely than that of osteopontin,
being strongest in the cells of the outer (skin side) layer, and
low or undetectable in the internal cells (Fig. 4C).
Double-hybridisation preparations for Fgfr1 and osteopontin
transcripts (Fig. 4D-F) confirm that the outermost (least mature)
cells express Fgfr1 but not osteopontin (arrowed in Fig. 4F),
whereas the cells deep within the matrix (most mature)
express only osteopontin (arrowheads), with cells intermediate
in position expressing both genes. Double-hybridisation
preparations for Fgfr1 and osteonectin (Fig. 4G-I) confirm that
the expression patterns of these two genes are largely
coincidental, except in the suture where the least mature
osteogenic cells at the sutural margin show only Fgfr1
expression (arrowed).
The results illustrated in Figs 2-4, simplified to show only
one differentiation marker (osteopontin), are summarised
diagrammatically in Fig. 5.
Fig. 6. The coronal suture region of E15
heads 6 hours after implantation of beads
soaked in FGF2 in PBS or PBS alone, as
indicated. (A,D) FGF2-induced up-regulation
of Fgfr1 at the edge of the bone domains and
in the sutural mesenchyme (arrows in A).
(B,E) FGF2-induced down-regulation of
Fgfr2 in the region of the suture (arrowheads
in B). (C,F) FGF2-induced down-regulation
of Fgfr3 in the region of the suture
(arrowheads in C); (G,J) FGF2-induced upregulation of osteopontin in the region of the
suture (arrows in G) and within the bone
domains adjacent to the beads. (H) Alkaline
phosphatase localization and (K) osteonectin
expression patterns are unaltered after 6
hours exposure to exogenous FGF2.
(I) Immunodetection of digoxigenin-labelled
FGF2 indicates the extent of protein diffusion
from the beads 6 hours after implantation. b,
brain; f, frontal domain; p, parietal domain; s,
skin. Scale bar, 150 µm.
5616 S. Iseki, A. O. M. Wilkie and G. M. Morriss-Kay
The effect of exogenous FGF2 on Fgfr expression,
proliferation and differentiation in the region of the
coronal suture
Heparin acrylic beads soaked in a solution of either FGF2 in PBS
or PBS alone were inserted between the skin and the coronal
suture of E15.0 fetuses and the effects on gene expression and
proliferation observed after 2, 4, 6 and 24 hours. The effects on
Fgfr2 and osteopontin expression at 48 hours have been described
previously (Iseki et al., 1997). Fgfr1, Fgfr2, Fgfr3, osteopontin
and osteonectin transcripts were detected by in situ hybridisation
and alkaline phosphatase was detected histochemically. FGF2
diffusion from the bead into the surrounding tissues was detected
by immunochemical localisation of digoxigenin-labelled FGF2.
No effects of FGF2 were detected after 2 hours, at which time
digoxigenin-FGF2 had only diffused about one cell diameter from
the bead, and had not yet reached the skeletogenic membrane (not
shown). By 4 hours, digoxigenin-FGF2 was detected at the
skeletogenic membrane and Fgfr2 showed slight down-regulation
in all specimens; Fgfr1 and osteopontin expression were variable,
being up-regulated in some, but not all, specimens (not shown).
By 6 hours (Fig. 6A-K), the following pattern was observed in all
specimens exposed to FGF2 beads: Fgfr1 expression extended
across the suture, though at slightly lower transcript levels than in
the regions of osteoid deposition (Fig. 6A,D); in contrast,
expression of Fgfr2 and Fgfr3 was down-regulated within the
sutural region adjacent to the beads (Fig. 6B,C,E,F). Osteopontin
transcripts were ectopically detected at low levels within the
sutural mesenchyme and at high levels at the sutural margins of
the frontal and parietal bones beyond the limit of the osteoid (Fig.
6G,J). The other differentiation markers, osteonectin and alkaline
phosphatase, were unaltered at 6 hours (Fig. 6H,K; control
sections not shown). The spread of FGF2 from the beads was the
same as at 4 hours, extending to the skeletogenic membrane and
the skin (Fig. 6I). PBS beads had no effect on gene expression at
any time point (Fig. 6D,E,F,J and data not shown).
Following bead insertion, cell proliferation was assessed by
means of PCNA immunohistochemistry, following fixation
with 4% paraformaldehyde, which detects all cells that are
actively cycling (Bravo and MacDonald-Bravo, 1987). BrdU
immunohistochemistry was not used because of the risk in
handling the animals and in giving an intraperitoneal injection
so soon after major abdominal surgery. Six hours after bead
insertion, cells in the skeletogenic membrane close to PBS
beads (Fig. 7B) or in unoperated specimens (not shown)
showed a similar PCNA staining pattern to that of BrdU
immunohistochemistry at E16 (Fig. 3), except that there is less
osteoid and a larger number of proliferating cells in these
younger (E15 + 6 hours) specimens. In contrast, very few cells
in the skeletogenic membrane adjacent to FGF2 beads were
positive for PCNA (Fig. 7A). This result indicates that most
osteogenic stem cells ceased proliferating within 6 hours of
exposure to the bead-derived FGF2.
24 hours after bead insertion on E15, gene expression in E16
heads was detected by in situ hybridisation in both whole heads
and in sections (Fig. 8). PBS beads had no effect on gene
expression (not illustrated; see Fig. 2 for the normal E16
patterns). In the whole heads (Fig. 8A-C) the area around the
beads showed up-regulation of Fgfr1 and osteopontin, and
down-regulation of Fgfr2 and Fgfr3 (Fgfr3 not shown). In the
sections (Fig. 8D-F) Fgfr1 was ectopically expressed in
mesenchymal cells on both sides of the osteoid (Fig. 8D,
Fig. 7. Immunodetection of proliferating cell nuclear antigen
(PCNA) in the coronal suture region 6 hours after insertion of
beads soaked in FGF2 in PBS (A) and PBS alone (B). PCNA levels
are high at the sutural edges of the normal frontal (f) and parietal
(p) bone domains (arrows in B) but are low throughout the
skeletogenic membrane adjacent to the FGF2 beads (A). b, brain.
Scale bar, 100 µm.
arrowheads) and in sub-dermal connective tissue fibroblasts
(arrowed), but the midsutural cells (that do not normally show
Fgfr1 expression) were less affected, explaining the faint
sutural line in the whole specimens. Fgfr2 (Fig. 8E) was
ectopically expressed at low levels in cells deep to the
skeletogenic membrane (possibly representing developing
dura mater fibroblasts), but was undetectable in the sutural
region. Osteopontin was ectopically expressed both in the
skeletogenic membrane layer, bridging the suture, and in cells
on both sides of the skeletogenic membrane that are not
normally osteogenic (Fig. 8F). In contrast to osteopontin, the
other two differentiation markers, osteonectin gene expression
and alkaline phosphatase activity, were reduced (not shown).
Expression of Fgfr3 in the cartilage layer that underlies part
of the coronal suture was not affected at 24 hours or at any earlier
time point (not shown); this may be because FGF2, as indicated
by the digoxigenin labelling, did not diffuse deeper than the
osteogenic layer of the skeletogenic membrane (Fig. 6I).
The effects of FGF2 on Fgfr2 and osteopontin expression
after 48 hours have been described previously (Iseki et al.,
1997). In comparison with the effects detected in this study at
24 hours, the 48-hour pattern shows some down-regulation of
osteopontin expression and up-regulation of Fgfr2 expression
in a ‘halo’ around the area of induced gene expression. We
therefore examined the spread of digoxigenin-labelled FGF2
from beads 48 hours after their insertion. The spread of
digoxigenin-FGF2 was much reduced compared with the
maximal distance of diffusion seen at 4, 6 and 24 hours, and
no longer reached as far as the skeletogenic membrane (Fig.
8G). Mineralisation in the region of the suture was not
enhanced 48 hours after bead insertion, as might be expected,
but on the contrary was absent from an area corresponding to
the domain of altered gene expression around the beads (Fig.
8H).
Fgfr1 and Fgfr2 functions in skull development 5617
DISCUSSION
This study has described the relative expression patterns of
Fgfr1, Fgfr2 and Fgfr3 in the developing mouse skull vault,
particularly the coronal suture, and compared them with the
localisation of markers for cell proliferation and osteogenic
differentiation. The comparisons indicate functional
correlations for Fgfr1 with osteogenic cell differentiation, and
for Fgfr2 with proliferation of osteogenic stem cells.
Expression of Fgfr3 in the osteogenic membrane was too
variable to draw functional conclusions from, but our results
agree with those of studies in human fetuses at 22 weeks
(Delezoide et al., 1998) in showing that this receptor is
expressed
in
the
osteogenic
membrane
during
intramembranous ossification. Fgfr3 expression was more
strongly detected in the cranial cartilage, which later undergoes
endochondral ossification. Although the endochondral and
intramembranous ossification regions of the calvaria show very
little overlap, it is clear where they do coincide that the
chondrogenic membrane is internal to the osteogenic
membrane. This is consistent with the evolutionary origins of
the vertebrate skull as a cartilaginous braincase strengthened
by bony plates formed in the lower layer of the dermis
(Goodrich, 1958). Increasing FGFR signalling by means of
insertion of beads soaked in FGF2 altered the expression
patterns of Fgfrs 1 and 2, increased expression of osteopontin
(but not the other differentiation
markers examined), and decreased cell
proliferation.
Expression and function of
Fgfr3 in cranial cartilage
The expression of Fgfr3 in a plate of
cartilage underlying the lower part of
the developing coronal suture suggests
that it may be involved in tissue
interactions in this region. This
cartilage is morphologically separate
from the suture itself, but its position
subjacent to the proliferating and
differentiating margins of the frontal
and parietal bones suggests that it
could play a role in maintaining the
coronal suture as a growth centre. It
could also mediate tissue interactions
between the sutural cells and the brain,
and/or it could play a role in the initial
positioning of the coronal suture, since
it is present by E13.5 (our unpublished
observations), whereas the earliest
osteogenesis-related gene expression
is observable at E14.0. It lies just
caudal to the eye, and later forms the
greater wing of the sphenoid bone,
which, like the occipital region of the
skull, is mesoderm-derived in contrast
to the neural crest origin of the major
part of the skull vault (Couly et al.,
1993). The expression of Fgfr3 in the
cartilage of the occipital region, in
which endochondral ossification takes
place, suggests that it may play a role in growth of this region
of the skull. Achondroplasia is due to a mutation affecting the
transmembrane region of FGFR3 (Gly380Arg) (Rousseau et
al., 1994; Shiang et al., 1994). The phenotype is characterised
by decreased proliferation of epiphyseal chondrocytes, causing
diminished growth of the long bones. In contrast, growth of the
skull is excessive (macrocephaly), suggesting that the
functional response to FGFR3 signalling in the skull and long
bones is different.
A prolonged colour reaction time was required for in situ
hybridisation detection of Fgfr3 transcripts in the osteogenic
membrane compared with the conditions used for its detection
in cartilage, which were equivalent to those used for the
detection of Fgfr1 and Fgfr2. Like Fgfr2, Fgfr3 was downregulated in the skeletogenic membrane following exposure to
exogenous FGF2, consistent with its similar localisation to that
of Fgfr2, and suggesting a co-operative role between FGFR2
and FGFR3 signalling in osteogenic cell proliferation. The
observation that the Pro250Arg FGFR3 mutation specifically
causes synostosis of the coronal suture (Bellus et al., 1996)
suggests that there may be a coronal suture-specific functional
role for FGFR3 signalling; alternatively, it could be that the
mutation has a minimal effect on the proliferationdifferentiation balance but the tissue structure of the coronal
suture makes it more vulnerable than the other sutures to this
functional change.
Fig. 8. FGF2 bead insertion on E15: in situ hybridisation on whole E16 fetal heads (A-C) and on
sections (D-F, skin side uppermost); Fgfr1 (A,D), Fgfr2 (B,E) and osteopontin (C,F). The
expression patterns at 24 hours are similar to those observed after 6 hours, with further spread of
Fgfr1 (D) and osteopontin (F) up-regulation to the mesenchyme peripheral to the bone plates
(arrowheads in D) and the mesenchyme between the skeletogenic membrane and the skin
(arrow); a low level of ectopic Fgfr1 expression is detectable in the midsutural mesenchyme
(ms). (E) Fgfr2 is down-regulated to undetectable levels in the skeletogenic membrane in the
region of the suture, but a low level of ectopic Fgfr2 expression is detectable in the subsutural
mesenchyme (arrow). After 48 hours (G,H) digoxigenin-labelled FGF2 no longer reaches the
skeletogenic membrane (G); alizarin staining shows that mineralisation is inhibited close to the
beads (asterisk in H). b, brain; e, eye; f, frontal bone; ms, midsutural mesenchyme; p, parietal
bone. Scale bars, A-C, 1 mm; D-F, 150 µm.
5618 S. Iseki, A. O. M. Wilkie and G. M. Morriss-Kay
Functional correlates of Fgfr1 and Fgfr2 expression
The results suggest that in the skull vault, FGFR2-mediated
FGF signalling within the osteoprogenitor cells of the coronal
suture is mitogenic at physiological levels of FGF ligand. In
our previous study (Iseki et al., 1997) we showed that levels of
FGF2 are lower in the suture than in the differentiated region.
As osteogenic cells differentiate at the periphery of the boneforming domains, adjacent proliferating cells in the region of
the suture are exposed to higher levels of FGF2, resulting in
down-regulation of Fgfr2, up-regulation of Fgfr1, and the
cessation of cell division.
FGFR1-mediated signalling appears to be associated with
the process of osteogenic differentiation, but once
differentiation is well established, Fgfr1 is down-regulated.
This result is analogous to the data of Itoh et al. (1996),
showing that down-regulation of Fgfr1 is essential for
completion of the process of muscle differentiation. Mice
lacking one allele of Fgfr1 show increased osteogenesis in the
coronal suture (F. Perrin-Schmitt, personal communication).
Taken together, these observations are consistent with the
interpretation that Fgfr1 negatively regulates the rate at which
preosteoblast cells progress through the stages of
differentiation that precede terminal differentiation to mature
osteoblasts and osteocytes. Osteonectin is co-expressed with
Fgfr1, suggesting that it, too, may be involved in the
differentiation process rather than in the function of
differentiated osteoblasts. It is not essential for osteogenesis,
however, since osteonectin-null mutant mice show normal
skeletal development (Gilmour et al., 1998). Comparisons
between the position of cells expressing Fgfr1 and osteopontin
suggest that osteopontin is not expressed until after the
osteogenic cells have begun to secrete bone matrix; like
osteonectin, it is not essential for bone differentiation, since
osteopontin-null mutant mice show normal bone structure
(Rittling et al., 1998).
These two osteogenic genes also differed in their response
to exogenous FGF2, osteonectin being down-regulated and
osteopontin up-regulated. This difference is consistent with the
expression data if the bead-derived FGF2 acts to speed up the
normal process of differentiation, since osteonectin is normally
down-regulated before differentiation is complete. However,
this interpretation fails to account for the uncoupling of Fgfr1
and osteonectin coexpression in the bead experiments.
Alkaline phosphatase synthesis was affected by FGF2 in a
similar manner to osteonectin expression; these two osteogenic
factors may have similar functions.
Ligand-dependent receptor expression and the
outcome of signalling
Ligands known to activate the IIIc isoform of FGFR2 include
FGF1, FGF2, FGF4, FGF6, FGF8 and FGF9 (Ornitz et al.,
1996; Xu et al., 1998). Some of these ligands (FGF1, FGF2,
FGF4 and FGF6) also activate the IIIc isoform of FGFR1. In
a preliminary study (unpublished) we have found that the
Fgfr2 expressed in the skeletogenic membrane stem cells is
the IIIc (bek) isoform, but it is not known which Fgfr1
isoform is expressed in the preosteoblasts (the probe used is
likely to detect both isoforms). FGFs 1 and 2 activate both
isoforms of FGFR1 (Ornitz et al., 1996), so FGF2, which is
abundant in the mouse fetal skeletogenic membrane (Iseki et
al., 1997), is well placed to be the major ligand involved in
the proliferation-differentiation mechanism. Clearly, other
FGF ligands can compensate for the absence of FGF2
in Fgf2-null mice, since these mice have no skeletal
abnormalities (Zhou et al., 1998), but this does not rule out
Fig. 9. Model for the role of FGFRs in the proliferationdifferentiation transition of osteogenic stem cells in the early fetal
coronal suture and the effect of increasing FGFR signalling, based on
the results of this study combined with those of Iseki et al. (1997).
(A) Normal sutural growth: FGF2 is secreted by osteoblasts; it is
adsorbed onto the unmineralised bone matrix, and lower levels
diffuse into the extracellular environment of the sutural stem cells.
Osteogenic stem cells proliferate only where FGF2 levels are low;
these proliferating cells express Fgfr2. As new matrix is secreted by
the differentiating cells, FGF2 levels rise in the environment of the
osteogenic stem cells closest to the new matrix, and they differentiate
into preosteoblasts. The differentiation process involves downregulation of Fgfr2 and exit from the cell cycle, followed by upregulation of Fgfr1 and (slightly later) up-regulation of the
osteogenesis-related genes, including osteopontin. The preosteoblasts
begin to secrete matrix and are then defined as osteoblasts. Fgfr1 is
down-regulated when differentiation is complete. (B) Addition of
ectopic FGF2 to the environment of the osteoprogenitor stem cells
accelerates the differentiation process (down-regulation of Fgfr2, upregulation of Fgfr1 and osteopontin) so that the proliferating cell
population is lost for the duration of the increased signal. This
experimental model mimics in the short-term some of the effects of
the activating mutations of FGFR1, FGFR2 and FGFR3 that cause
craniosynostosis, although mineralisation is inhibited.
Fgfr1 and Fgfr2 functions in skull development 5619
the possibility that in wild-type mice (and humans) FGF2 is
the major physiological ligand. Neither we nor Kim et al.
(1998) were able to detect FGF4 in the developing calvaria,
but that group did detect FGF9, which activates FGFR2-IIIc;
the presence or absence of the other candidate ligands has not
been reported.
FGF-FGFR activation was measured in terms of cell
proliferation, either in tissue culture (Ornitz et al., 1996) or in
embryonic limb bud outgrowth (Xu et al., 1998). It is clear
from other observations, including our own, that FGF ligands
are not merely mitogens, but play essential roles in the
differentiation of certain tissues, e.g. mice lacking FGF2 show
defects of neuronal differentiation but not proliferation (Dono
et al., 1998). It is particularly interesting that in the developing
skull, signalling through FGFR2 is associated with
osteoprogenitor cell proliferation, whereas signalling through
FGFR1 is associated with osteogenic differentiation. The
concentration of ligand appears to be crucial to the outcome of
signalling, through a mechanism that involves control over
which receptor is expressed. Down-regulation of FGFR2
(through a post-transcriptional mechanism) has also been
observed in response to increased levels of FGF2 (and FGF4)
in embryonal carcinoma cells, although the expression of
FGFR1 was unaffected (Ali et al., 1995).
The time-related response to FGF2 beads
In our previous study (Iseki et al., 1997), ectopic osteopontin
expression in the region of the beads was surrounded by a
‘halo’ of Fgfr2 expression. This observation was made 48
hours after bead insertion; in contrast, the results shown here
at 6 and 24 hours do not show the halo effect. We now report
that mineralisation is actually delayed around the site of bead
implantation. The digoxigenin-labelled FGF2-soaked bead
implants clearly indicate that FGF2 takes more than 2 hours to
diffuse as far as the skeletogenic membrane, and that between
24 and 48 hours after bead insertion the radius of diffusion
decreases. During this second 24 hour period, FGF2 levels are
clearly falling at the periphery of the area in which osteopontin
and Fgfr1 are up-regulated, resulting in their down-regulation,
and up-regulation of Fgfr2. This suggests that the effects of the
exogenous FGF2 on gene expression (and presumably also on
proliferation) are still reversible after 24 hours, possibly related
to the fact that the induced changes did not progress to the
osteogenic maturational stage when Fgfr1 is down-regulated,
and/or that Fgfr1 up-regulation in the midsutural cells was not
maintained for long enough for the osteoblasts to begin osteoid
secretion before exogenous FGF2 levels began to fall again.
The inhibition of mineralisation in the bead insertion area after
48 hours may be due to the fact that osteopontin expression is
induced at a very high level: osteopontin protein has been
shown to inhibit the growth of hydroxyapatite crystals (Hunter
et al., 1996).
48 hours after bead implantation, ectopic cell proliferation
is widespread within the cranial mesenchyme on the beadimplanted side of the head, and not restricted to the region of
the beads (our unpublished observations). This result is likely
to be due to a chain of signalling events spreading out from the
region of the beads. The reaction of the mesenchymal and
osteogenic membrane cells to exogenous FGF2 is clearly
different in kind, proliferation being stimulated in the
mesenchyme and differentiation in the osteogenic cells.
Interpretation of the roles of Fgfr1 and Fgfr2 in
normal sutural growth and in conditions of
increased signalling
A model for the roles of FGFR signalling in the normal coronal
suture, and how ectopic FGF2 induces the same processes, is
presented in Fig. 9. The craniosynostosis phenotypes resulting
from activating FGFR2 mutations affecting the IgIIIc region of
the protein suggest that this model is also valid for other
sutures, although there are regional differences in the
involvement of other genes such as Msx2 (Liu et al., 1995).
Our suggestion that increasing levels of FGF2 are functionally
correlated with the differentiation process (Iseki et al., 1997)
is supported by observations that this ligand is increased within
the sutures at the time of fusion (Mehrara et al., 1998). We
suggest (1) that the role of FGFR1 signalling in normal skull
development is to regulate the rate of osteogenic differentiation
in an analogous manner to the regulation of cell proliferation
by FGFR3 in epiphyseal plate chondrocytes (Colvin et al.,
1996; Deng et al., 1996) and by FGFR2 in cranial vault sutures
(Iseki et al., 1997); (2) that the function of FGFR signalling in
maintaining the proliferation-differentiation balance in the
coronal suture principally involves FGFR1 and FGFR2,
FGFR3 playing only a minor role.
We are grateful to Action Research and Grants-in-Aid for Scientific
Research from the Japanese Ministry of Education, Science and
Culture (No. 10770998) for their support of this project. A. O. M. W.
is a Wellcome Trust Senior Clinical Fellow. We thank John Heath for
donation of digoxigenin-labelled and unlabelled FGF2 for the bead
experiments, and for many helpful discussions. Probes for in situ
hybridisation were kindly donated by John Heath, Brigid Hogan,
Shintarou Nomura, and David Ornitz. We thank Youichirou Ninomiya
for help with figures 1 and 4.
REFERENCES
Ali, J., Mansukhani, A. and Basilico, C. (1995). Fibroblast growth factor
receptors 1 and 2 are differentially regulated in murine embryonal carcinoma
cells and in response to fibroblast growth factor-4. J. Cell Physiol. 165, 438448.
Beiman, M., Shilo, B. Z. and Volk, T. (1996). Heartless, a Drosophila FGF
receptor homolog, is essential for cell migration and establishment of several
mesodermal lineages. Genes Dev. 10, 2993-3002.
Bellus, G. A., Gaudenz, K., Zackai, E. H., Clarke, L. A., Szabo, J.,
Francomano, C. A. and Muenke, M. (1996). Identical mutations in three
different fibroblast growth factor receptor genes in autosomal dominant
craniosynostosis syndromes. Nat. Genet. 14, 174-176.
Bravo, R. and MacDonald-Bravo, H. (1987). Existence of two populations
of cyclin/proliferating cell nuclear antigen during the cell cycle: association
with DNA replication sites. J. Cell Biol. 105, 1549-1554.
Casal, J. and Leptin, M. (1996). Identification of novel genes in Drosophila
reveals the complex regulation of early gene activity in the mesoderm. Proc.
Natl. Acad. Sci. USA 93, 10327-10332.
Colvin, J. S., Bohne, B. A., Harding, G. W., McEwen, D. G. and Ornitz,
D. M. (1996). Skeletal overgrowth and deafness in mice lacking fibroblast
growth factor receptor 3. Nat. Genet. 12, 390-397.
Couly, G. F., Coltey, P. M. and Le Douarin, N. M. (1993). The triple origin
of skull in higher vertebrates: a study in quail-chick chimeras. Development
117, 409-429.
Delezoide, A. L., Benoist-Lasselin, C., Legeai-Mallet, L., Le Merrer, M.,
Munnich, A., Vekemans, M. and Bonaventure, J. (1998). Spatio-temporal
expression of FGFR 1, 2 and 3 genes during human embryo-fetal
ossification. Mech. Dev. 77, 19-30
Deng, C., Wynshaw-Boris, A., Zhou, F., Kuo, A. and Leder, P. (1996).
Fibroblast growth factor receptor 3 is a negative regulator of bone growth.
Cell 84, 911-921.
Dodig, M., Tadic, T., Kronenberg, M. S., Dacic, S., Liu, Y. H., Maxson, R.,
5620 S. Iseki, A. O. M. Wilkie and G. M. Morriss-Kay
Rowe, D. W. and Lichtler, A. C. (1999). Ectopic msx2 overexpression
inhibits and msx2 antisense stimulates calvarial osteoblast differentiation.
Dev. Biol. 209, 298-307.
Dono, R., Texido, G., Dussel, R., Ehmke, H. and Zeller, R. (1998). Impaired
cerebral cortex development and blood pressure regulation in FGF-2deficient mice. EMBO J. 17, 4213-25.
Drury, R. A. B. and Wallington, E. A. (1967). Carlton’s Histological
Technique, pp.235-254. New York/Toronto: Oxford University Press.
El Ghouzzi, V., Le Merrer, M., Perrin Schmitt, F., Lajeunie, E., Benit, P.,
Renier, D., Bourgeois, P., Bolcato Bellemin, A. L., Munnich, A. and
Bonaventure, J. (1997). Mutations of the TWIST gene in the SaethreChotzen syndrome. Nature Genet. 15, 42-46.
Gilmour, D. T., Lyon, G. J., Carlton, M. B., Sanes, J. R., Cunningham, J.
M., Anderson, J. R., Hogan, B. L., Evans, M. J. and Colledge, W. H.
(1998). Mice deficient for the secreted glycoprotein SPARC/osteonectin/
BM40 develop normally but show severe age-onset cataract formation and
disruption of the lens. EMBO J. 17, 1860-1870.
Goodrich, E. S. (1958). Studies on the Structure and Development of
Vertebrates, pp. 230-395. New York: Dover Publications Inc.
Howard, T. D., Paznekas, W. A., Green, E. D., Chiang, L. C., Ma, N.,
Ortiz de Luna, R. I., Garcia Delgado, C., Gonzalez Ramos, M., Kline,
A. D. and Jabs, E. W. (1997). Mutations in TWIST, a basic helix-loophelix transcription factor, in Saethre-Chotzen syndrome. Nat. Genet. 15,
36-41.
Hunter, G. K., Hauschka, P. V., Poole, A. R., Rosenberg, L. C. and
Goldberg, H. A. (1996). Nucleation and inhibition of hydroxyapatite
formation by mineralized tissue proteins. Biochem. J. 317, 59-64.
Iseki, S., Wilkie, A. O. M., Heath, J. K., Ishimaru, T., Eto, K. and MorrissKay, G. M. (1997). Fgfr2 and osteopontin domains in the developing skull
vault are mutually exclusive and can be altered by locally applied FGF2.
Development 124, 3375-3384.
Itoh, N., Mima, T. and Mikawa, T. (1996). Loss of fibroblast growth factor
receptors is necessary for terminal differentiation of embryonic limb muscle.
Development 122, 291-300.
Kim, H. J., Rice, D. P., Kettunen, P. J. and Thesleff, I. (1998). FGF-, BMPand Shh-mediated signalling pathways in the regulation of cranial suture
morphogenesis and calvarial bone development. Development 125, 12411251.
Liu, Y. H., Kundu, R., Wu, L., Luo, W., Ignelzi, M. A., Jr., Snead, M. L.
and Maxson, R. E., Jr. (1995). Premature suture closure and ectopic cranial
bone in mice expressing Msx2 transgenes in the developing skull. Proc.
Natl. Acad. Sci. USA 92, 6137-6141.
Liu, Y. H., Tang, Z., Kundu, R. K., Wu, L., Luo, W., Zhu, D., Sangiorgi,
F., Snead, M. L. and Maxson, R. E. (1999). Msx2 gene dosage influences
the number of proliferative osteogenic cells in growth centers of the
developing murine skull: a possible mechanism for MSX2-mediated
craniosynostosis in humans. Dev. Biol. 205, 260-274.
Mehrara, B. J., Mackool, R. J., McCarthy, J. G., Gittes, G. K. and
Longaker, M. T. (1998). Immunolocalization of basic fibroblast growth
factor and fibroblast growth factor receptor-1 and receptor-2 in rat cranial
sutures. Plast. Reconstr. Surg. 102, 1805-1817.
Muenke, M., Schell, U., Hehr, A., Robin, N. H., Losken, H. W., Schinzel,
A., Pulleyn, L. J., Rutland, P., Reardon, W., Malcolm, S. et al. (1994).
A common mutation in the fibroblast growth factor receptor 1 gene in
Pfeiffer syndrome. Nat. Genet. 8, 269-274.
Opperman, L. A., Sweeney, T. M., Redmon, J., Persing, J. A. and Ogle, R.
C. (1993). Tissue interactions with underlying dura mater inhibit osseous
obliteration of developing cranial sutures. Dev. Dyn. 198, 312-322.
Opperman, L. A., Nolen, A. A. and Ogle, R. C. (1997). TGFβ1, TGFβ2, and
TGFβ3 exhibit distinct patterns of expression during craniofacial suture
formation and obliteration in vivo and in vitro. J. Bone Miner. Res. 12, 322331.
Ornitz, D. M., Xu, J., Colvin, J. S., McEwen, D. G., MacArthur, C. A.,
Coulier, F., Gao, G. and Goldfarb, M. (1996). Receptor specificity of the
fibroblast growth factor family. J. Biol. Chem. 271, 15292-15297.
Paznekas, W. A., Cunningham, M. L., Howard, T. D., Korf, B. R., Lipson,
M. H., Grix, A. W., Feingold, M., Goldberg, R., Borochowitz, Z., Aleck,
K., et al. (1998). Genetic heterogeneity of Saethre-Chotzen syndrome, due
to TWIST and FGFR mutations. Am. J. Hum. Genet. 62, 1370-1380.
Rittling, S. R., Matsumoto, H. N., McKee, M. D., Nanci, A., An, X. R.,
Novick, K. E., Kowalski, A. J., Noda, M. and Denhardt, D. T. (1998).
Mice lacking osteopontin show normal development and bone structure but
display altered osteoclast formation in vitro. J. Bone Miner. Res. 13, 11011111.
Rousseau, F., Bonaventure, J., Legeai Mallet, L., Pelet, A., Rozet, J. M.,
Maroteaux, P., Le Merrer, M. and Munnich, A. (1994). Mutations in the
gene encoding fibroblast growth factor receptor-3 in achondroplasia. Nature
371, 252-254.
Shiang, R., Thompson, L.M., Zhu, Y.Z., Church, D.M., Fielder, T.J.,
Bocian, M., Winokur, S.T., Wasmuth, J.J. (1994). Mutations in the
transmembrane domain of FGFR3 cause the most common form of
dwarfism, achondroplasia. Cell 78, 335-342.
Shishido, E., Higashijima, S-i., Emori, Y. and Saigo, K. (1993). Two FGFreceptor homologues of Drosophila: one is expressed in mesodermal
primordium in early embryos. Development 117, 751-761.
Stark, K. L., McMahon, J. A. and McMahon, A. P. (1991). FGFR-4, a new
member of the fibroblast growth factor receptor family, expressed in the
definitive endoderm and skeletal muscle lineages of the mouse.
Development 113, 641-651.
Wall, S. A. (1997). Diagnostic features of the major non-syndromic
craniosynostosis and the common deformational conditions which may be
confused with them. Curr. Paediat. 7, 8-17.
Wilkie, A. O. M., Slaney, S. F., Oldridge, M., Poole, M. D., Ashworth, G.
J., Hockley, A. D., Hayward, R. D., David, D. J., Pulleyn, L. J., Rutland,
P. et al. (1995). Apert syndrome results from localized mutations of FGFR2
and is allelic with Crouzon syndrome. Nat. Genet. 9, 165-172.
Wilkie, A. O. M. (1997). Craniosynostosis: genes and mechanisms. Hum. Mol.
Genet. 6, 1647-1656.
Xu, X., Weinstein, M., Li, C., Naski, M., Cohen, R. I., Ornitz, D. M., Leder,
P. and Deng, C. (1998). Fibroblast growth factor receptor 2 (FGFR2)mediated reciprocal regulation loop between FGF8 and FGF10 is essential
for limb induction. Development 125, 753-765.
Zhou, M., Sutliff, R. L., Paul, R. J., Lorenz, J. N., Hoying, J. B.,
Haudenschild, C. C., Yin, M., Coffin, J. D., Kong, L., Kranias, E. G., et
al. (1998). Fibroblast growth factor 2 control of vascular tone. Nat. Med. 4,
201-207.