Cell fate specification during calvarial bone and suture development
Eva Lana-Elolaa,⁎, Ritva Ricea, Agamemnon E. Grigoriadisa, David P.C. Ricea,b
aDepartments of Craniofacial Development and Orthodontics, Floor 27 Guy’s Tower, King’s College, London, SE1 9RT, UK
bInstitute of Dentistry, PO Box 63, 00014 University of Helsinki and Helsinki University Central Hospital, Helsinki, Finland
Received for publication 10 April 2007; revised 27 July 2007; accepted 13 August 2007
Available online 14 September 2007
In this study we have addressed the fundamental question of what cellular mechanisms control the growth of the calvarial bones and
conversely, what is the fate of the sutural mesenchymal cells when calvarial bones approximate to form a suture. There is evidence that the
size of the osteoprogenitor cell population determines the rate of calvarial bone growth. In calvarial cultures we reduced osteoprogenitor cell
proliferation; however, we did not observe a reduction in the growth of parietal bone to the same degree. This discrepancy prompted us to
study whether suture mesenchymal cells participate in the growth of the parietal bones. We found that mesenchymal cells adjacent to the
osteogenic fronts of the parietal bones could differentiate towards the osteoblastic lineage and could become incorporated into the growing
bone. Conversely, mid-suture mesenchymal cells did not become incorporated into the bone and remained undifferentiated. Thus
mesenchymal cells have different fate depending on their position within the suture. In this study we show that continued proliferation of
osteoprogenitors in the osteogenic fronts is the main mechanism for calvarial bone growth, but importantly, we show that suture mesenchyme
cells can contribute to calvarial bone growth. These findings help us understand the mechanisms of intramembranous ossification in general,
which occurs not only during cranial and facial bone development but also in the surface periosteum of most bones during modeling and
© 2007 Elsevier Inc. All rights reserved.
Keywords: Calvaria; Suture; Intramembranous ossification; Osteoblast; Cell fate
At a molecular level we have a good understanding of what
regulates the commitment of undifferentiated mesenchymal
cells into osteoblasts (reviewed by Cohen, 2006; Karsenty,
2003). We also know a great deal about endochondral
ossification and what controls long bone growth which is a
complex process involving the formation of a cartilaginous
template that is later replaced by bone. Intramembranous bone
ossification accounts for most of the bone growth of the face
and calvaria and in the periosteum of long bones (Kronenberg,
2003; Ornitz, 2005). In contrast to endochondral ossification,
intramembranous bone ossification is a relatively simple pro-
cess with no cartilage anlagen and osteoblasts differentiating
directly from mesenchymal cells. It is therefore surprising that
we know much less about the cellular processes that control
intramembranous bone growth compared to the processes
controlling endochondral bone growth.
In the craniofacial region, intramembranous bone develop-
ment starts with the aggregation of mesenchymal cells into
condensation centers (Hall and Miyake, 2000). These tightly
packed collections of cells expand and once they reach a critical
size the cells at the center of these masses differentiate into
osteoblasts. Growth of immature bones occurs at the leading
edges or osteogenic fronts and when an osteogenic front
confronts its neighbor, the two fronts either merge to create a
single bone or form a suture. Thus a suture is a joint composed
of two osteogenic fronts and the interposed mesenchyme. Once
the basic pattern of bones and sutures has been established
further craniofacial growth primarily occurs in the osteogenic
fronts of the sutures. The processes of osteogenic condensation
formation and function during embryogenesis, to form the first
bony elements, are directly comparable to the processes of cell
aggregation, proliferation, differentiation and function that
Available online at www.sciencedirect.com
Developmental Biology 311 (2007) 335–346
⁎Corresponding author. Fax: +44 2071881674.
E-mail address: Eva.Lana_Elola@kcl.ac.uk (E. Lana-Elola).
0012-1606/$ - see front matter © 2007 Elsevier Inc. All rights reserved.
occur in the osteogenic fronts of established craniofacial
The rate of calvarial bone growth is determined by the size of
the osteoprogenitor cell population in the perimeter of the
osteogenic condensations and then later by the size of the
osteoprogenitor population in the osteogenic fronts. We have
previously shown that the transcription factor Foxc1 regulates
bone morphogenetic protein (BMP)-mediated osteoprogenitor
proliferation, specifically at the leading edge of the developing
calvarial bones thus regulating bony expansion. Reduced
osteoprogenitor proliferation in Foxc1−/−mice results in small
calvarial bones that do not grow beyond a rudimentary size and
remain at the sites of the initial osteogenic condensations. Foxc1
controls the osteoprogenitor population by regulating the BMP
targets Msx2 and Alx4 (Rice et al., 2005). In both humans and
mice, loss-of-function mutations in Msx2 and Alx4 result in
similar ‘hole in the head’ phenotypes to those exhibited by
Foxc1−/−mutant mice (Antonopoulou et al., 2004; Satokata et
al., 2000; Wilkie et al., 2000). Conversely, mice which
overexpress Msx2 exhibit increased osteoprogenitor prolifera-
tion in osteogenic fronts of the calvarial bones, which results in
enhanced bone growth (Liu et al., 1999b). Consistent with this,
a gain-of-function mutation in MSX2 causes Boston-type
craniosynostosis in humans which is characterized by excessive
calvarial bone growth and an obliteration of the calvarial sutures
(Liu et al., 1995).
In addition to BMP signaling, the size of the calvarial
osteoprogenitor population is regulated by fibroblast growth
factor (FGF) signaling. Several Fgf ligands and receptors are
expressed in the developing calvarial mesenchyme and bones
(Hajihosseini and Heath, 2002; Hajihosseini et al., 2001; Rice
et al., 2003a), and targeted disruption of Fgfr signaling in
mice results in altered calvarial osteoprogenitor proliferation
and abnormal bone growth (Eswarakumar et al., 2004;
Eswarakumar et al., 2002; Wang et al., 2005). It has been
proposed that signaling through Fgfr2 primarily regulates
osteoprogenitor proliferation while signaling through Fgfr1
primarily regulates osteoblast differentiation (Iseki et al.,
1999). In humans, mutations in FGFR1, 2 and 3 cause
several forms of syndromic and non-syndromic craniosynos-
tosis including Apert, Crouzon and Pfeiffer syndromes and
Muenke craniosynostosis. All these conditions are character-
ized by premature craniofacial suture fusion as well as other
skeletal anomalies (Rice, 2005).
We have previously proposed the developing mouse
sagittal suture as an uncomplicated model of osteoblastic
differentiation and intramembranous bone growth (Rice et al.,
2003b). By analyzing the distribution of markers that are
expressed at different stages of osteoblastic differentiation, we
can build up a picture of the suture so that we can see all
stages of differentiation within a single tissue section. In this
study we have expanded and refined this analysis and used it
to study calvarial bone growth. We have previously shown
that embryonic day 15.5 (E15.5) mouse parietal bones grow
towards each other and after 2 days of culture form a normal
patent sagittal suture (Kim et al., 1998). Using this system we
now ask the fundamental questions: what are the cellular
mechanisms controlling calvarial bone growth; and conver-
sely what is the fate of the calvarial mesenchymal cells when
parietal bones approximate and a suture is formed?
In this study we show that proliferation and subsequent
differentiation of osteoprogenitors at the osteogenic fronts,
although important, is not the only cellular mechanism that
contributes to calvarial bone growth. Sutural mesenchymal cells
can differentiate into osteoblasts and become incorporated into
the growing parietal bones, but only if adjacent to the
osteogenic fronts. Thus the fate of calvarial mesenchymal
cells varies depending on their position within the suture. Under
normal conditions we demonstrate that a small percentage of
sutural mesenchymal cells are incorporated into the growing
bones. We can hypothesize that during pathological conditions
such as craniosynostosis the relative contribution of recruitment
from the mesenchyme into the calvarial bones may be altered.
Materials and methods
Calvaria were dissected free from skin and brain from E15.5 mouse
embryos. Explants were placed in Nuclepore polycarbonate filters (Whatman)
andculturedin a Trowell-typeorganculture system for 2 or 5 days dependingon
the experiment. DMEM (Sigma) was supplemented with 20 IU/ml penicillin/
streptomycin (Sigma), 10% fetal bovine serum (Life Technologies), and 100 μg/
ml of ascorbic acid. Media were replaced every 2 days. The antimitotic drug
nocodazole (Sigma-Aldrich) was added to the culture media in different
concentrations (0.1 μg/ml, 0.25 μg/ml, and 1.0 μg/ml) from a stock of 0.5 mg
nocodazole in 2 ml DMSO. Explants treated with 1 μl of DMSO per 1 ml of
media were used as control. At least two explants treated with DMSO as control
and two of each nocodazole dilutions were sectioned and at least 4 sections of
each explant were used to count BrdU positive cells.
Cells were enzymatically isolated from the calvaria of CD1 mice postnatal
day 1 by sequential digestion with trypsin (1 mg/ml), dispase II (2 mg/ml), and
two digestions with collagenase A (2 mg/ml) (Roche). The last two of the four
digestion steps (populations III–IV) were pooled and plated in T75 flasks
containing α-MEM (Sigma), penicillin–streptomycin, glutamine and 15% of
heat activated fetal bovine serum (FBS). After 24 h of incubation attached cells
were then collected by trypsinization. Aliquots were counted and remaining
cells were resuspended in the standard medium described above supplemented
with ascorbic acid (50 mg/ml) and 10 mM of β-glycerol phosphate. The
resuspended cells were plated in tissue culture dishes at approximately 4×103
cells/cm2. The medium was changed every 2–3 days. Cells were incubated at
37 °C in humidified atmosphere at 95% air 5% CO2 incubator. DMSO
(1:50,000) or nocodazole (0.1 μg/ml or 1.0 μg/ml) was added to the media on
day 1 on cell culture (experiment 1), when cells had reached confluence
(experiment 2) or just when nodule formation was staring to occur (experiment
3). Cells were washed in PBS and then stained for alkaline phosphatase using
naphtol-AS-MX phosphate and fast red (Sigma) with 100 mM of Tris pH 8.3.
Vital dye injections
DiI C18 (1-1′-dioctadecyl-3,3,3′,3′-tetramethilindocarbocyanine perchlo-
rate, Molecular Probes, Inc) was dissolved at 5% in absolute ethanol and then
diluted 1:10 in 0.3 M sucrose. CellTracker™CM-DiI (Molecular Probes, Inc)
was used for labeling in the same manner. Using microcapillars filled with DiI,
microinjections were performed under a stereo microscope on the day 0 of
culture. DiI-labeled cells were tracked using an Olympus SZX12 microscope
with a fluorescent light by photographing at day 0, day 1 and day 2 of culture.
336 E. Lana-Elola et al. / Developmental Biology 311 (2007) 335–346
Explants were fixed after 2 days of culture in 4% paraformaldehyde (PFA)
overnight at 4 °C. Explants labeled with DiI C18were embedded with O.C.T
compound (BDH, VWR), stored in −70 °C and cryosectioned (7 μm). Explants
in which CellTracker™CM-DiI was used were dehydrated in gradient ethanol
series, paraffin embedded and sectioned (7 μm). Images of both cryosections
and paraffin sections were taken with a Zeiss Axioskop 2 plus coupled with
fluorescence light before proceeding with any staining. Hematoxylin and eosin
staining or alkaline phosphatase (Roche) (Liu et al., 1999a) staining was
performed and images were taken and then superimposed using Adobe
Photoshop 6.0 software in order to establish the location and the identity of the
Green fluorescence protein tissue transplantations
Two types of transplantations were performed: sagittal suture mesenchyme
transplantations and osteogenic front transplantations. The tissue was dissected
from green fluorescence protein expressing mice (chicken beta-actin/CMV
enhancer construct) and grafted into CD1 calvarial explants in the exact same
position as it was in the donor tissue. The recipient’s sagittal suture mesenchyme
was removed prior the grafting. The operation was performed in the day 0 of
Olympus SZX12 microscope. Explants were fixed in 4% PFA and processed.
Paraffin sections were photographed under fluorescence light before any staining.
H&E staining or Bone sialoprotein detection by35S in situ hybridization was
performed and then superimposed with the florescence images.
BrdU incorporation and TUNEL analysis
Pregnant females were injected i.p. with 2 ml/100 g body weight of 5-
bromo-2′-deouxyridine (BrdU) solution (Zymed). After 2 h, embryos were
collected, fixed in 4% PFA, dehydrated and paraffin embedded for sectioning.
Cultured tissues were BrdU pulsed diluting the labeling reagent 1:100 with
tissue culture medium. BrdU incorporation was immunodetected by using BrdU
staining kit (Zymed). Terminal deoxynucleotidyl transferase-mediated dUTP
nick-end labeling (TUNEL) assay was performed by using the DeadEnd
Colorimetric TUNEL system (Promega).
Preparation of probes and in situ hybridization
35S in situ hybridization on paraffin sections was performed as previously
described (Vainio et al., 1993). The preparation of the following RNA probes
has been described: Bsp, Msx2 Runx2 and Twist1 (Rice et al., 2003b), Osteo-
calcin cDNA was inserted in pBluescipt KS+ vector and was digested with
EcoRI for sense and Pst1 for antisense riboprobes, Osx (Yunker et al., 2004).
Both bright and dark field images were taken from hybridized sections. Silver
grains were selected from the dark field image, colored red and then
superimposed into the identical bright field image using Adobe Photoshop 6.0
software. Whole mount in situ hybridization was performed using digoxigenin-
UTP-labeled riboprobes as previously described (Rice et al., 1997).
Cell counting and statistical analysis
One way ANOVA and independent samples t-test were used for the
statistical analysis of normal distributed samples. Non-parametric test Kruskal–
Wallis and Mann–Whitney test were the chosen tests for non-normal samples. A
P value of less than 0.05 was considered statistically significant. SPSS 14.0 was
used for statistical analysis of the data.
The developing mouse sagittal suture as a model of
intramembranous bone growth
E15.5 sagittal suture presents an excellent model to study
osteogenesis as it is possible to see all stages of differentiating
osteoblasts in a single tissue section. Expression profiles of
different genes that indicate different stages in osteogenesis and
suture patency were analyzed (Fig. 1). Genes encoding a bone
matrix glycoprotein Bone sialoprotein (Bsp) and a bone matrix
gla protein Osteocalcin (Oc) were expressed by mature
osteoblasts in the parietal bones (Figs. 1A, B). Runx2 and Os-
terix (Osx) are transcription factors that are essential for
osteoblast development, and their expression is known to
precede that of Bsp and Oc (Ducy et al., 1997). Osx acts
downstream of Runx2, it has an important role in the
commitment of bipotential (chondrocyte/osteoblast) mesench-
ymal cells into the osteoblastic lineage (Nakashima et al., 2002).
Both Osx and Runx2 were expressed by early osteoblasts, and
their expression was detected in the parietal bones and in the
bone margins or osteogenic fronts. However, some Runx2
expression was detected in the sutural mesenchyme, suggesting
that these cells, previously thought to be uncommitted, may
have osteoblastic potential (Figs. 1C, D). The expression of
Msx2 was restricted to the sutural mesenchyme and around
the osteogenic fronts (Fig. 1E). Twist1 was expressed by a
defined strip of mesenchymal cells in the suture with higher
intensity near the osteogenic fronts which is consistent with its
proposed role as a negative regulator of osteoblastic differentia-
tion thus keeping the suture unossified. Twist1 was also
detected in the dermis and epidermis (Fig. 1F).
Based on these gene expression patterns and in our previous
work (Rice et al., 2000, 2003a) it is possible to determine the
cellular identity and degree of differentiation of cells forming a
suture based on their location in the developing suture. This
makes the mouse E15.5 sagittal suture an uncomplicated and
valuable model for intramembranous ossification and bone
Proliferation is important for calvarial bone growth
We and others have previously shown that proliferation
mainly takes place in the osteogenic fronts (Figs. 1G, G′) (Iseki
et al., 1997; Liu et al., 1999b; Rice et al., 2003c). In vivo
calvarial bone growth can be mimicked in organ culture (Kim et
al., 1998). We cultured E15.5 calvarial explants in different
concentrations of nocodazole (or DMSO as control) for 2 days.
As nocodazole is an antimitotic drug that disrupts microtubules
by binding to β-tubulin and arrests the cell cycle in G2/M phase
(Jordan et al., 1992; Luduena and Roach, 1991). Nocodazole is
known to induce apoptosis in several normal and tumor cell
lines (Wang et al., 1998), we examined cell death in paraffin
sections of the cultured calvarial explant and observed no
differences in the number of apoptotic cells between nocoda-
zole-treated and non-treated calvarial explants (data not shown).
Whole mount in situ hybridization for Bsp was performed in
order to help visualize the developing parietal bones. We
observed patchy expression of Bsp, which is expressed by
mature osteoblasts, in all the explants treated with the
antimitotic drug. This prompted us to investigate whether
nocodazole has an effect on osteoblast differentiation or
whether patchy appearance of mature osteoblasts within the
growing parietal bones was simply caused by reduced
337E. Lana-Elola et al. / Developmental Biology 311 (2007) 335–346
proliferation. We isolated mouse postnatal day 1 calvarial
osteoblasts for primary cell culture. The cells were continuously
treated with media containing nocodazole (0.1 μg/ml or 1.0 μg/
ml) (or DMSO in control samples) starting at three different
time points: 1. After 1 day of culture: while control cells reached
confluence and formed osteoblastic nodules, the nocodazole-
treated cells did not reach confluence and did not form any
nodules. 2. Cultured cells were treated with nocodazole only
after cells had reached confluence. With low concentration of
nocodazole osteoblastic nodules were formed but high
Fig. 1. Calvarial development is tightly controlled and proliferation is important for calvarial bone growth. Diagram shows an apical view of a calvarial explant. Dotted
red line indicates the plane of section. Sections of sagittal suture at E15.5 were hybridized with35S-UTP-labeled riboprobes Bsp, Oc, Runx2, Osx, Msx2 and Twist1.
Boxes show the areas of enlargements, named with a prima. Intense expression of Bsp and expression of Oc is detected in the parietal bones (A, A′, B, B′). Runx2 and
Osterix are expressed along the parietal bones and in the osteogenic fronts (C, C′, D, D′). In contrast, the expression of Msx2 is restricted to the sagittal suture
mesenchyme and in the osteogenic fronts (E, E′). Twist1 is expressed by a strip of cells in the sutural mesenchyme as well as to the lower layers of the skin (F, F′).
Arrows indicate the osteogenicfront. BrdU positive cellsare detectedin more number in the osteogenicfront (double arrow) (G′). p, parietal bone; s, sagittal suture; of,
osteogenic front. Scale bars 200 μm.
338E. Lana-Elola et al. / Developmental Biology 311 (2007) 335–346
concentration of nocodazole was toxic to the cells. This shows
that nocodazole does not block osteoblastic differentiation. 3.
Cultured cells were treated with nocodazole just when nodule
formation was starting: nocodazole-treated cells formed small
nodules (data not shown). These cell culture experiments
showed that nocodazole does not have a direct effect on
osteoblastic differentiation, rather it delays it only by reducing
the level of cell proliferation, for instance not enough
osteoblastic cells are produced with the window of time given
for osteoblastic cell mass production before differentiation
process is started during calvarial bone development and
growth. The smaller size of nocodazole-treated nodules could
also be explained by the fact that even in differentiated nodules,
there is proliferation and it seems to be important for nodule
formation (Malaval et al., 1999). A reduction in parietal bone
growth was also observed (Figs. 2A–D). We measured the
distance between the parietal bones and found that nocodazole-
treated explants had significantly wider sagittal sutures when
compared to the controls, indicating reduced bone growth
(P=0.005). When comparing the growth during culture period
(distance between parietal bones on day 0 minus distance
between parietal bones after 2 days in culture) we observed that
there had been a 19% reduction in parietal bone growth in the
nocodazole-treated calvarial explants. Nocodazole significantly
reduced cell proliferation compared to DMSO controls
(P=0.001). There were no statistically significant differences
in cell proliferation between the different concentrations of
nocodazole (P=0.165, P=1.000, P=0.670) (Figs. 2E–H).
Pooling data, proliferation was reduced by 44% in nocoda-
zole-treated cultures in comparison to controls.
These results showed that nocodazole caused a reduction in
cell proliferation and this correlated with reduced growth of the
parietal bones. However, there was a discrepancy in these
findings in that bone growth was not reduced to the same degree
Fig. 2. Inhibition of cell proliferation affects parietal bone growth. Bsp whole mount in situ hybridization for calvarial explants treated with DMSO (A) or nocodazole
dilutions: 0.1 μg/ml (B), 0.25 μg/ml (C) and 1.0 μg/ml (D). Parietal bones are further apart in nocodazole-treated tissues, and the expression of Bsp in those explants is
patchy. Scale bar 500 μm. BrdU detection in paraffin sections of calvarial explants treated with DMSO (E) or nocodazole dilutions: 0.1 μg/ml (F), 0.25 μg/ml (G) and
1.0 μg/ml (H). Scale bar 100 μm. BrdU positive cells in calvarial explants in nocodazole-treated and DMSO control explants (I). A minimum of two explants were
analyzed in each group. For each calvarial explant, a minimum of four serial sections were analyzed. The number of BrdU positive cells was significantly decreased in
nocodazole-treated tissues compared to DMSO controls. The distance between the parietal bones was measured on the calvarial explants where Bsp whole mount in
situhadbeenperformed.Thegraphrepresentsthe resultsinthe lengthofthe sagittalsuture.Eightexplants wereusedon averageoneachgroupto calculatethe distance
between the parietal bones. The length of the sagittal suture is significantly increased in nocodazole-treated tissues compared to DMSO controls (J). p, parietal bone.
339 E. Lana-Elola et al. / Developmental Biology 311 (2007) 335–346
Fig. 3. Sagittal suture mesenchymal cells can differentiate to form parietal bone. Mesenchymal tissue from the sagittal suture of a GFP expressing embryo at E15.5 was
transplanted into the sagittal suture of an E15.5 calvaria recipient. Bright field images show the bone growth after the transplant during 5 days of culture (A, B, C).
Solid line indicates the width of the sagittal suture. Images under fluorescent light showing the GFP positive tissue transplanted (A′, B′, C′). Scale bar 500 μm.
Hematoxylin and eosin stained tissue section of a transplanted tissue on day 0 of culture (D). Same tissue section under fluoresce light showing the GFP positive
mesenchyme grafted (D′). Superimposition of images D and D′ (D″). Adjacent section probed for Bsp as a demonstration that only mesenchyme was transplanted in
the host calvarial explant (E). Scale bar 100 μm. Calvarial explant that was transplanted with a GFP mesenchyme, taken at day 0 as control (F–F″). Alkaline
the GFP positive transplanted mesenchyme (F′). Superimposition of images F and F′ demonstrating that mesenchymal tissue transplanted was undifferentiated and did
not express the early osteoblast marker alkaline phosphatase (F″). Black dotted lines demarcate the parietal bones and red dotted circle demarcates the transplanted
mesenchyme. H&E stained section of the sagittal suture after 5 days of culture (G). Same tissue section under fluorescent light showing the position of GFP positive
cells after 5 days of culture (G′). Superimposition of F and F′ images showing three GFP positive cells integrated in the parietal bone (G″ arrows). Bsp detection by in
situ hybridization in the sagittal suture from a different transplanted calvarial explant (H). Same tissue section as in G under fluorescent light, GFP positive cells after 5
days of culture of the operated explant (H′). Superimposition of images H and H′ showing a GFP positive cell is also a Bsp expressing cell (H″, arrow). Scale bar
100 μm. p, parietal bone; s, sagittal suture.
340 E. Lana-Elola et al. / Developmental Biology 311 (2007) 335–346
as the reduction observed in cell proliferation. This suggests that
proliferation (and subsequent osteoblast differentiation) is an
important mechanism whereby the calvarial bones grow but not
the only one.
Sutural mesenchymal cells can differentiate into osteoblasts
and become incorporated into the growing parietal bones
We performed sagittal suture mesenchyme grafting from
green fluorescent protein (GFP) expressing mice into wild type
sagittal suture calvarial explants. This experiment enabled us to
follow the fate of the cells transplanted in the cultured calvaria
(Fig. 3). Time lapse pictures of the tissues in culture (Figs. 3A–
C, representative explant) showed the GFP positive mesench-
yme transplanted into the calvaria (Figs. 3A′–C′). To locate the
GFP-positive transplanted cells the cultured calvarial explants
were sectioned. The tissue grafted was undifferentiated
mesenchyme as explants harvested on day 0 showed no
expression of the late osteoblast marker Bsp (Fig. 3E) or early
osteoblast marker alkaline phosphatase (Figs. 3F–F″). Hema-
toxylin and eosin staining in Fig. 3G shows a cross-section of a
cultured sagittal suture. When this section was studied under
Fig. 4. Cells in the osteogenic front can remain in the bony front or form part of the body of the parietal bone. Osteogenic front tissue from a GFP expressing embryo at
E15.5 was transplanted into the osteogenic front region of an E15.5 calvaria recipient. Bright field images show the bone growth after the transplant during 5 days of
culture demarcated with a dotted line (A, B, C). Images of the same calvarial explant under fluorescent light showing the GFP positive tissue transplanted (A′, B′, C′).
Scale bar 500 μm. Histological section of a transplanted tissue on day 0 of culture probed for Bsp (D). Same tissue section under fluoresce light showing the GFP
positive tissue grafted (D′). Superimposition of images D and D′ (D″). Scale bar 100 μm. Expression of Bsp on the sagittal suture after 5 days of culture (E). Same
tissue section under fluorescent light showing GFP positive cells after 5 days of culture (E′). Superimposition of E and E′ images showing the positioning of the GFP
positive cells after 5 days of culture (E″). Arrows indicate GFP positive cells expressing Bsp incorporated in the parietal bone. Arrowhead pointing at a cell that
remains within the osteogenic front. p, parietal bone; of, osteogenic front.
341 E. Lana-Elola et al. / Developmental Biology 311 (2007) 335–346
342 E. Lana-Elola et al. / Developmental Biology 311 (2007) 335–346
fluorescent light (Fig. 3G′) we can identify the GFP-positive
transplanted cells. A superimposition of Figs. 3G and G′
showed that most of the GFP-positive cells were pushed out of
the suture and remained mesenchymal, but interestingly some
of the GFP-positive cells were integrated into the parietal bone
(Fig. 3G″, arrows). To confirm that the GFP-positive cells,
which had been incorporated into the parietal bone, had
differentiated into osteoblasts, sections from grafted explants
were hybridized with Bsp (Fig. 3H, representative section).
When the section shown in Fig. 3H was observed under
fluorescent light (Fig. 3H′), just as in Fig. 3H′, GFP-positive
cells were identifiable within the parietal bone. Thus GFP-
positive cells, which were previously Bsp-negative and were
located in the suture mesenchyme, were now located within the
parietal bone and were Bsp-positive (Fig. 3H″).
From this experiment we can conclude that firstly, calvarial
explants healed after the grafting and the suture continued
normal development, as shown by the parietal bone growth and
a narrowing of the sagittal suture (Figs. 3A–C). Secondly,
mesenchymal cells were excluded from the sutures as bones
continued their growth (Figs. 3F″ and G″). Thirdly, most of the
transplanted mesenchymal cells remained in the mesenchyme,
but some cells became located in the parietal bones (Figs. 3F′
and G′). Taken together, a few sutural mesenchymal cells
became incorporated into the parietal bone (6 out of 11
explants) (Figs. 3F, G). The osteoblastic identity of these cells
was confirmed by their expression of the osteoblast marker Bsp
(Fig. 3G″, arrow).
Cells within the osteogenic front can either remain in the
advancing osteogenic front or become incorporated into the
As only a small portion of mesenchymal cells from the suture
contributed to the parietal bones, the driving force for the
expansion of these bones must lie within the proliferative
osteogenic fronts. We hypothesized that cells in the osteogenic
fronts undergo cell division in which some daughter cells
remain at that location and differentiate into osteoblasts, while
other daughter cells remain in the osteogenic front and are
carried forward by this proliferation/differentiation process into
the sutural space. To test this, we performed recombination
experiments, transplanting GFP-positive tissue from an osteo-
genic front into the osteogenic front of the recipient calvarial
explant to see if the GFP-positive cells remained within the
osteogenic front or if they were integrated into the parietal bone.
After 5 days of culture the transplant was integrated, and the two
parietal bones had approximated normally (Figs. 4A–C). The
transplanted osteogenic front had moved with the developing
parietal bone (Figs. 4A′–C′). Explants were taken on day 0 of
culture in order to assess the histological positioning of the graft
(Figs. 4D–D″). In Figs. 4E–E″, a representative 5-day culture
shows GFP-positive osteogenic front cells that had stayed in the
osteogenic front (Fig. 4E″, arrowhead) and also GFP-positive
osteogenic front cells that had become incorporated into the
parietal bone and were expressing the osteoblast marker Bsp
(Fig. 4E″, arrow). The small number of GFP-positive cells
detected in Fig. 3E′ is due to the plane of section, this section
being taken at the edge of the GFP positive graft. These results
were confirmed by DiI-labeled cells in the osteogenic front (see
Sutural cells have different fate depending on their position
within the suture
Vital dye injections were performed in order to make a fate
map of the mesenchymal cells in the developing suture.
Injections were performed in labeling cells in three different
regions: (1) middle of the sagittal suture mesenchyme, (2)
mesenchyme in the proximity of an osteogenic front, or (3)
osteogenic front. Explants were cultured for 2 days and
photographed daily. Interestingly, DiI-labeled cells did not
move considerably. It appeared as if the DiI-labeled cells stayed
in their original position and the developing bones approached
them as their growth progressed (Figs. 5A–C″).
Labeled cells’ fate was determined by analyzing tissue
sections stained either for alkaline phosphatase (osteoblastic
cells) or hematoxylin and eosin to aid localization.
The analysis of the tissues in which mid-sutural
mesenchymal cells were labeled with DiI showed that all
labeled cells remained undifferentiated. The osteogenic fronts
grew into the area of the injected mesenchyme but none of
these cells became incorporated into the bone (11 out of 11
explants) (Figs. 5D–D″). DiI-labeled mesenchymal cells
located next to an osteogenic front were either localized in
the mesenchyme or in the osteogenic front (7 out of 16
explants) (Figs. 5E–E″). Taken together, suture mesenchymal
DiI injections indicated that mid-sutural cells were not
destined to become osteoblasts during normal calvarial bone
development and that only sutural cells next to the advancing
Fig. 5. Differential fate of mesenchymal cells within the suture depending on their positioning. Vital dye labeling with an explant by DiI injection on E15.5 calvarial
explants cultured for 48 h. Bright field images (A, B, C), images under fluorescent light (A′, B′, C′), superimposition of bright and fluorescent images (A″, B″, C″).
DiI-labeled cells appear to extend radially. Scale bar 500 μm. Diagrams indicate where the DiI injection was performed. Cryosection of the sagittal suture from an
explant cultured for 2 days after DiI was injected in the middle of the sutural mesenchyme stained with alkaline phosphatase (D). Same tissue section under fluorescent
light (D′). Superimpositionof images DandD′ showingDiI-labeled cellsremain in the suturalmesenchyme(D″). Hematoxylinandeosinstained paraffin section from
an explant cultured for 2 days where DiI injection was performed in the mesenchyme near the osteogenic front (E). Same section under fluorescent light (E′).
Superimposition of images E and E′ (E″). Dotted outline denotes the parietal bones and their osteogenic fronts. Superimposition of images E and E′ showing labeled
cells in the osteogenic front (E″). Tissue section of an explant that was DiI injected in the osteogenic front, fluorescence light picture (F). Exact same tissue section
stained for H&E (F′). Superimposition of F and F′ showing the location of the DiI positive cells within the parietal bone (F″, arrow). Image taken under fluorescence
light of a cryosection from a calvarial explant in which DiI was injected in the osteogenic front (G). Alkaline phosphatase staining of the same tissue section (G′).
Superimposition of images G and G′ showing DiI-labeled cells in the osteogenic front (G″, arrow) and in the parietal bone (G″ arrowhead). Scale bar 100 μm. p,
343E. Lana-Elola et al. / Developmental Biology 311 (2007) 335–346
osteogenic fronts could be incorporated into the growing
To elucidate if the cells in the osteogenic front remained
within the osteogenic front or if they became part of the parietal
bone, we injected DiI directly into the osteogenic front. Tissue
sections of these explants showed that some DiI-labeled
cells had moved with the osteogenic front as it advanced
(Figs. 5F–F″ and G″ arrow), while other labeled cells were in
the parietal bone after 2 days of culture (5 out of 13 explants)
(Figs. 5G–G″ arrowhead). This supports the findings obtained
with GFP tissue transplantations shown in Fig. 4.
Is intramembranous bone growth autonomous so that the
initial condensed mesenchymal cell population keeps prolifer-
ating and differentiating into mature functioning osteoblasts
without any assistance from the surrounding mesenchyme? Or
do intramembranous bones grow by a combination of
proliferation and recruitment of neighboring mesenchymal
cells into the expanding bone?
Using mutant mouse lines we and others have shown
previously that proliferation in the growing bones is essential
for normal intramembranous growth and development (Rice et
al., 2003c; Satokata et al., 2000). Here we show that reducing
cell proliferation by 44 percent in wild-type calvarial cultures
results in a reduction in bone growth of only 19% (Fig. 2). This
discrepancy led us to study the involvement of the sutural
mesenchyme in calvarial bone growth. We used tissue grafting
to demonstrate that the majority of the surrounding mesench-
yme was not incorporated into the calvarial bones. However a
small proportion of mesenchymal cells were taken into the
bones where they differentiated into functioning osteoblasts
(Figs. 3 and 4). This incorporation of mesenchymal cells into
the developing calvarial bones cannot fully account for the
remaining growth that occurs when proliferation is blocked. In
our tissue culture experiments there will be a time-lag before the
antimitotic drug is effective. During this time osteoprogenitors
will continue to proliferate and then differentiate. Together with
these cells, the explants will contain postmitotic osteoblasts
which can lay down bone matrix and contribute to the
lengthening of the bones. Another mechanism which may
contribute to bone growth is the morphological changes that
occur in the sagittal suture. The osteogenic fronts are folded
endocranially, due in part to their location in the sulcus between
cerebral hemispheres. Both in our tissue culture and in vivo the
osteogenic fronts flatten leading to an extension of the bone
margins and a narrowing of the sagittal suture.
To further study whether any mesenchymal cells from the
suture can contribute to the bone, we performed vital dye
injections into different sites in the suture. We found that, firstly,
there does not appear to be any active, directional cell
movement within the suture, even when close to the osteogenic
fronts which express several potential chemoattractants.
Labeled sutural cells radiated from the injection site to give
the effect that some cells migrated away. However, this may be
due to less and less dye being taken up by subsequent
generations of cells in a proliferative environment. Secondly,
calvarial bone growth appeared to be autonomous so that
proliferation in the osteogenic fronts supplied sufficient cells
needed for bone expansion and that this cell population
‘pushed’ forward into the sutural mesenchyme (Fig. 5G″).
Thirdly, growing parietal bones can use sutural mesenchymal
cells as building blocks but only if adjacent to the osteogenic
fronts. The use of mesenchymal cells from the suture could be
regarded as opportunistic: the sutural cells being in the right
place at the right time.
By combining data from tissue grafting (Figs. 3 and 4) and
vital dye experiments (Fig. 5) with the analysis of molecular
marker expression patterns in the developing suture (Fig. 1), we
observed that the majority of the sutural mesenchyme does not
contribute to the parietal bones. This is a little surprising as a
distinct band of cells stretching across the suture expresses the
osteoblasticmarkers Runx2 and Msx2.Also, we have previously
shown that undifferentiated mesenchymal cells possess plur-
ipotency; being able, under specific in vitro conditions, to
differentiate into several different cell types including osteo-
blasts (Grigoriadis et al., 1988). One reason that these sutural
cells do not more readily differentiate into osteoblasts may be
because this population of cells also expresses Twist1. Twist1
binds to and negatively regulates Runx2 thereby inhibiting
osteoblastic differentiation, and it has been proposed that Twist1
may have a role in keeping sutures patent (Bialek et al., 2004;
Rice et al., 2000). As Twist1 is not expressed by mature
into the developing bones must stop expressing Twist1. Loss of
TWIST1 function in humans causes the craniosynostosis
syndrome Saethre–Chotzen (El Ghouzzi et al., 2001). Runx2
determines the osteoblastic lineage from multipotent mesench-
ymal cells, enhancing osteoblast differentiation at an early stage
and inhibiting osteoblast differentiation at a late stage (Komori,
2003). Runx2 also regulates osteoblastic proliferation and loss-
of-function mutations in RUNX2 cause cleidocranial dysplasia
which is characterized by disrupted skeletal development, most
notably in the skull (Mundlos, 1999; Pratap et al., 2003).
Signaling from the dura mater adjacent to a suture has been
shown to be important in the control of suture patency (Kim et
al., 1998; Opperman et al., 1995). Tissue separation and
recombination experiments have demonstrated that the orienta-
tion of the dura in relation to the suture is also important, with
that regional differences regulating whether a suture closes at
the appropriate time (Levine et al., 1998). In our experiments,
we dissected out the sutural mesenchyme together with its
underlying dura mater and we maintained the orientation of the
mesenchymal tissue transplanted in the host tissue.
What is the fate of the mesenchymal cells within the suture?
Although some cells become incorporated into the calvarial
bones, this would not account for the majority of the sutural
mesenchyme. We have previously shown that there are clusters
of apoptotic cells in the sutural mesenchyme but that this is a
relatively uncommon event (Rice et al., 1999). Using GFP-
positive sutural tissue transplantation, we show that as bones
grew towards each other most of the sagittal suture
mesenchymal cells are expelled from the narrowing suture.
344 E. Lana-Elola et al. / Developmental Biology 311 (2007) 335–346
Also, the formation of a sutural blastema has been described
(Markens, 1975). This is a mesenchymal thickening that
develops as intramembranous bones approximate (mouse
E15.5–E17.5). This may, in part, be due to high levels of
cell proliferation in the mesenchyme, concomitant with intense
osteogenesis. However, it may also, in part, be explained by
growth of the adjacent bones firstly compressing the
mesenchyme between the osteogenic fronts and secondly
expelling mesenchymal cells from their path, as we have
observed in our explants (Fig. 3).
In this study we have analyzed calvarial bone development
under normal conditions. The next challenge will be to apply
our findings to abnormal developmental situations such as the
formation of sutural bones, craniosynostosis and delayed
suture closure typified in cleidocranial dysplasia. Within
developing sutures additional ossification centers can arise
which eventually develop into small bones, called sutural or
Wormian bones. Sutural bones most commonly occur in the
lambdoid suture and are seen in a number of conditions
including cleidocranial dysplasia, osteogenesis imperfecta and
hypothyroidism (Gorlin et al., 2001; Hinton et al., 1984).
Sutural bones are also seen in Apert syndrome. Patients with
Apert syndrome are characterized by several abnormalities
including premature fusion of the coronal sutures. In addition,
they present with a wide calvarial defect between the frontal
and parietal bones. Within this region bony islands form in
the calvarial mesenchyme, these later coalesce with each other
and with the neighboring calvarial bones, resulting in the
fusion across the midline and no further calvarial growth in
this plane (Cohen and Kreiborg, 1996). Under normal
conditions we have shown that a small percentage of sutural
mesenchymal cells are incorporated into the growing bones.
However, during craniosynostosis or the formation of sutural
bones the relative contribution of recruitment from the
mesenchyme into the calvarial bones may be abnormally
During normal development the prime mechanism of
calvarial bone expansion is by continued proliferation in the
osteogenic fronts (Fig. 6). As well as studying how intramem-
branous bones grow, we have also examined the fate of the
sutural mesenchyme and found that cells have different fates
depending on their location within the suture. Most mesench-
ymal cells remain undifferentiated in the suture, however a
small proportion are recruited into the developing bones. We
hope that these findings may help us understand the mechan-
isms of intramembranous ossification in general, which occurs
not just in craniofacial bones but also in the periosteum of most
bones during modeling and remodeling.
We thank A Coudert, S Dietrich, T Mitsiadis and A Streit for
their help with the study. We thank S. Pudaruth for her technical
assistance. This work was funded by EU Marie Curie-EST
fellowship, EC contract number MEST-CT-2004-504025 (ELE)
and MRC (DPCR).
Antonopoulou, I., et al., 2004. Alx4 and Msx2 play phenotypically similar and
additive roles in skull vault differentiation. J. Anat. 204, 487–499.
Bialek, P., et al., 2004. A twist code determines the onset of osteoblast
differentiation. Dev. Cell 6, 423–435.
Cohen Jr., M.M., 2006. The new bone biology: pathologic, molecular, and
clinical correlates. Am. J. Med. Genet., A 140, 2646–2706.
CohenJr., M.M.,Kreiborg, S.,1996.Suture formation,prematuresutural fusion,
and suture default zones in Apert syndrome. Am. J. Med. Genet. 62.,
Ducy, P., et al., 1997. Osf2/Cbfa1: a transcriptional activator of osteoblast
differentiation. Cell 89, 747–754.
El Ghouzzi, V., et al., 2001. Mutations in the basic domain and the loop–helix II
junction of TWIST abolish DNA binding in Saethre–Chotzen syndrome.
FEBS Lett. 492, 112–118.
Eswarakumar, V.P., et al., 2002. The IIIc alternative of Fgfr2 is a positive
regulator of bone formation. Development 129, 3783–3793.
Eswarakumar, V.P., et al., 2004. A gain-of-function mutation of Fgfr2c
demonstrates the roles of this receptor variant in osteogenesis. Proc. Natl.
Acad. Sci. U. S. A. 101, 12555–12560.
Gorlin, R., et al., 2001. Syndromes of the Head and Neck. Oxford Univ. Press,
Grigoriadis, A.E., et al., 1988. Differentiation of muscle, fat, cartilage, and bone
from progenitor cells present in a bone-derived clonal cell population: effect
of dexamethasone. J. Cell Biol. 106, 2139–2151.
Fig. 6. Model for parietal bone growth and suture development. The primary
mechanism for parietal bone growth is proliferation and subsequent differentia-
tion of osteoprogenitors at the osteogenic fronts. In addition, sutural
mesenchymal cells can differentiate into osteoblasts and become incorporated
into the growing parietal bones, but only if adjacent to the osteogenic fronts.
Thus the fate of calvarial mesenchymal cells varies depending on their position
within the suture. Yellow figures represent the process of proliferation within the
osteogenic front, in which osteoprogenitors in the osteogenic front undergo
mitosis and remain within the osteogenic front. These cells move with the
osteogenic front into the sutural mesenchyme as the two parietal bones
approximate. Red figures represent the fate of preosteoblasts in the osteogenic
front, which can either advance with the developing bone at its front or can
terminally differentiate into mature osteoblasts and become incorporated in the
parietal bone. Green figures represent the different fate of cells positioned in the
sagittal suture: mesenchymal cells can either remain undifferentiated, getting
expulsed from the suture as parietal bones grow towards each other, or they can
become incorporated by the adjacent parietal bone and consequently express
osteoblast marker genes as Bsp and AP. Neither of these two alternatives
implicates cell migration, the basic movement is the parietal bone growth. of,
345 E. Lana-Elola et al. / Developmental Biology 311 (2007) 335–346
Hajihosseini, M.K., Heath, J.K., 2002. Expression patterns of fibroblast growth
factors-18 and -20 in mouse embryos is suggestive of novel roles in calvarial
and limb development. Mech. Dev. 113, 79–83.
Hajihosseini, M.K., et al., 2001. A splicing switch and gain-of-function
mutation in FgfR2-IIIc hemizygotes causes Apert/Pfeiffer-syndrome-like
phenotypes. Proc. Natl. Acad. Sci. U. S. A. 98, 3855–3860.
Hall, B.K., Miyake, T., 2000. All for one and one for all: condensations and the
initiation of skeletal development. BioEssays 22, 138–147.
Hinton, D.R., et al., 1984. Lambdoid synostosis. Part 1. The lambdoid suture:
normal development and pathology of “synostosis”. J. Neurosurg. 61,
Iseki, S., et al., 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.
Iseki, S., et al., 1999. Fgfr1 and Fgfr2 have distinct differentiation- and
proliferation-related roles in the developing mouse skull vault. Development
Jordan, M.A., et al., 1992. Effects of vinblastine, podophyllotoxin and
nocodazole on mitotic spindles. Implications for the role of microtubule
dynamics in mitosis. J. Cell Sci. 102 (Pt 3), 401–416.
Karsenty, G., 2003. The complexities of skeletal biology. Nature 423, 316–318.
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, 1241–1251.
Komori, T., 2003. Requisite roles of Runx2 and Cbfb in skeletal development.
J. Bone Miner. Metab. 21, 193–197.
Kronenberg, H.M., 2003. Developmental regulation of the growth plate. Nature
Levine, J.P., et al., 1998. Studies in cranial suture biology: regional dura
mater determines overlying suture biology. Plast. Reconstr. Surg. 101,
Liu, Y.H., et al., 1995. Premature suture closure and ectopic cranial bone in mice
expressing Msx2 transgenes in the developing skull. Proc. Natl. Acad. Sci.
U. S. A. 92, 6137–6141.
Liu, P., et al., 1999a. Regulation of osteogenic differentiation of human bone
marrow stromal cells: interaction between transforming growth factor-beta
and 1,25(OH)(2) vitamin D(3) in vitro. Calcif. Tissue Int. 65, 173–180.
Liu, Y.H., et al., 1999b. 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.
Luduena, R.F., Roach, M.C., 1991. Tubulin sulfhydryl groups as probes and
targets for antimitotic and antimicrotubule agents. Pharmacol. Ther. 49,
Malaval, L., et al., 1999. Kinetics of osteoprogenitor proliferation and osteoblast
differentiation in vitro. J. Cell. Biochem. 74, 616–627.
Markens, I.S., 1975. Embryonic development of the coronal suture in man and
rat. Acta Anat. (Basel) 93, 257–273.
Mundlos, S., 1999. Cleidocranial dysplasia: clinical and molecular genetics.
J. Med. Genet. 36, 177–182.
Nakashima, K., et al., 2002. The novel zinc finger-containing transcription
factor osterix is required for osteoblast differentiation and bone formation.
Cell 108, 17–29.
Opperman, L.A., et al., 1995. Cranial sutures require tissue interactions with
dura mater to resist osseous obliteration in vitro. J. Bone Miner. Res. 10,
Ornitz, D.M., 2005. FGF signaling in the developing endochondral skeleton.
Cytokine Growth Factor Rev. 16, 205–213.
Pratap, J., et al., 2003. Cell growth regulatory role of Runx2 during proliferative
expansion of preosteoblasts. Cancer Res. 63, 5357–5362.
Rice, D.P., 2005. Craniofacial anomalies: from development to molecular
pathogenesis. Curr. Mol. Med. 5, 699–722.
Rice, D.P., et al., 1997. Detection of gelatinase B expression reveals osteoclastic
bone resorption as a feature of early calvarial bone development. Bone 21,
Rice, D.P., et al., 1999. Apoptosis in murine calvarial bone and suture
development. Eur. J. Oral Sci. 107, 265–275.
Rice, D.P., et al., 2000. Integration of FGF and TWIST in calvarial bone and
suture development. Development 127, 1845–1855.
Rice,D.P., et al., 2003a.Fgfr mRNAisoforms incraniofacialbone development.
Bone 33, 14–27.
Rice, D.P., et al., 2003b. Molecular mechanisms in calvarial bone and suture
development, and their relation to craniosynostosis. Eur. J. Orthod. 25,
Rice, R., et al., 2003c. Progression of calvarial bone development requires
Foxc1 regulation of Msx2 and Alx4. Dev. Biol. 262, 75–87.
Rice, R., et al., 2005. Foxc1 integrates Fgf and Bmp signalling independently of
twist or noggin during calvarial bone development. Dev. Dyn. 233,
Satokata, I., et al., 2000. Msx2 deficiency in mice causes pleiotropic defects in
bone growth and ectodermal organ formation. Nat. Genet. 24, 391–395.
Vainio, S., et al., 1993. Identification of BMP-4 as a signal mediating secondary
induction between epithelial and mesenchymal tissues during early tooth
development. Cell 75, 45–58.
Wang, T.H., et al., 1998. Microtubule-interfering agents activate c-Jun N-
terminal kinase/stress-activated protein kinase through both Ras and
apoptosis signal-regulating kinase pathways. J. Biol. Chem. 273,
Wang, Y., et al., 2005. Abnormalities in cartilage and bone development in the
Apert syndrome FGFR2(+/S252W) mouse. Development 132, 3537–3548.
Wilkie, A.O., et al., 2000. Functional haploinsufficiency of the human
homeobox gene MSX2 causes defects in skull ossification. Nat. Genet.
Yunker, L.A., et al., 2004. The tyrosine phosphatase, OST-PTP, is expressed
in mesenchymal progenitor cells early during skeletogenesis in the mouse.
J. Cell. Biochem. 93, 761–773.
346 E. Lana-Elola et al. / Developmental Biology 311 (2007) 335–346