© 2024. Published by The Company of Biologists Ltd | Development (2024) 151, dev202116. doi:10.1242/dev.

202116

STEM CELLS AND REGENERATION RESEARCH REPORT

Cranial suture lineage and contributions to repair of the

mouse skull
Daniel Doro1,2, *, Annie Liu1, Jia Shang Lau1, Arun Kumar Rajendran2, Christopher Healy1, Marko Krstic1,
Agamemnon E. Grigoriadis1, Sachiko Iseki2, * and Karen J. Liu1,*

ABSTRACT intermediate (Nakashima and de Crombrugghe, 2003). Growth of

these skull bones is organised at fibrous joints called sutures and
The cranial sutures are proposed to be a stem cell niche, harbouring
occurs at the leading edges of the separate bones (Doro et al., 2017).
skeletal stem cells that are directly involved in development,
Throughout embryonic and postnatal development, the sutures
homeostasis and healing. Like the craniofacial bones, the sutures
remain as active sites of bone formation. By early adulthood, when
are formed from both mesoderm and neural crest. During cranial bone
growth of the skull is complete, sutures become quiescent and
repair, neural crest cells have been proposed to be key players;
gradually fuse. Nevertheless, cells residing in the sutures or adjacent
however, neural crest contributions to adult sutures are not well
periosteum retain the potential to heal calvarial bone in adults,
defined, and the relative importance of suture proximity is unclear.
following injury or disease (Doro et al., 2017). The healing capacity
Here, we use genetic approaches to re-examine the neural crest–
in the skull appears to rely on the sutural mesenchyme, which has
mesoderm boundaries in the adult mouse skull. These are combined
recently been proposed to act as a stem cell niche (Zhao et al., 2015).
with calvarial wounding experiments suggesting that suture proximity
In response to wounding, the skeletal mesenchyme rapidly
improves the efficiency of cranial repair. Furthermore, we
undergoes proliferation, with sutural cells expanding toward the
demonstrate that Gli1 + and Axin2 + skeletal stem cells are present
wound. Several groups have demonstrated that ablation of these
in all calvarial sutures examined. We propose that the position of the
resident skeletal stem cells blocks the healing capacity of the skull
defect determines the availability of neural crest-derived progenitors,
(Zhao et al., 2015; Maruyama et al., 2016; Wilk et al., 2017).
which appear to be a key element in the repair of calvarial defects.
Here, we make use of a crucial and underappreciated observation:
KEY WORDS: Sutures, Stem cells, Cranial repair, Neural crest, that adult frontal bones in the mouse heal more efficiently than
Mesoderm, Bone parietal bones (Quarto et al., 2010; Doro et al., 2019). Studies from
mouse models have demonstrated that the embryonic frontal bones
INTRODUCTION are of neural crest origin, whereas the parietal bones are mesodermally
Craniofacial bone repair, which can be necessary as a result of derived (Jiang et al., 2002), raising the possibility that developmental
accidents, congenital anomalies, or diseases such as cancer, is a history influences osteogenic potential. Indeed, we have demonstrated
daunting clinical challenge and a significant biomedical burden. that adult neural crest-derived osteoblasts have an increased
Current treatment strategies include replacements; however, these do osteogenic capacity when cultured in vitro (Doro et al., 2019). In
not truly mimic bone and can lead to difficulties in integration with contrast, parietal osteoblasts rarely generate osteogenic nodules (Doro
other tissues, such as muscle. Poor healing can have severe impact on et al., 2019); however, osteogenesis could be restored by co-culturing
function and aesthetics, and multi-fragment breaks can be difficult to with neural crest-derived cells, either from frontal bones, or from the
reconstruct (Lanza et al., 2020). Therefore, there is a need to improve dura mater (Doro et al., 2019). Anecdotal evidence suggests a
our understanding of the endogenous craniofacial repair process, similarly improved healing capacity in human frontal bones (Skogh
including a clearer definition of the osteogenic stem cell niche. et al., 2013). These observations raise the possibility that there are
In craniofacial structures, the calvarial and facial bones form via increased numbers (or activity) of calvarial stem cells in sutures
intramembranous ossification, in which osteoblasts coalesce and adjacent to the frontal bones. Most of the original studies comparing
differentiate directly within a membrane, without a cartilaginous frontal and parietal healing efficiency do not account for the role of
suture-derived osteoprogenitors, or the relevance of defect position in
relation to the sutures. More recently, it has been demonstrated that
1
Centre for Craniofacial and Regenerative Biology, Faculty of Dentistry, Oral &
Craniofacial Sciences, King’s College London, London SE1 9RT, UK. 2Department
bone healing is not an evenly distributed event across the parietal bone
of Molecular Craniofacial Embryology and Oral Histology, Tokyo Medical and surface (Park et al., 2016). Although these studies show a correlation
Dental University, 1-5-45 Yushima, Bunkyo-ku, Tokyo 113-8549, Japan. between repair efficiency and suture proximity, they disregard the
*Authors for correspondence ([email protected];
dual embryonic origin of the cranial sutures and the possibility that
[email protected]; [email protected]) distinct lineage contributions may influence repair. Furthermore,
DEVELOPMENT

several studies suggest that neural crest-derived cells, and not the
D.D., 0000-0001-5058-273X; A.L., 0009-0002-8362-6983; J.S.L., 0009-0003-
1889-5439; A.K.R., 0000-0001-6443-6063; C.H., 0009-0006-9720-5185; M.K., mesoderm, play key roles in the pathogenesis of craniofacial
0009-0008-7806-0210; A.E.G., 0000-0002-5941-8132; S.I., 0000-0001-8448-9410; malformations (Hari et al., 2002; Brault et al., 2001; Wu et al.,
K.J.L., 0000-0002-2483-2165 2017), but also that neural crest may account for differences in cranial
This is an Open Access article distributed under the terms of the Creative Commons Attribution healing efficiency upon injuries of every sort (Quarto et al., 2010;
License (https://creativecommons.org/licenses/by/4.0), which permits unrestricted use, Li et al., 2010; Senarath-Yapa et al., 2013).
distribution and reproduction in any medium provided that the original work is properly attributed.
Based on historical lineage-tracing assays using Wnt1::cre-driven
Handling Editor: Patrick Tam ROSA26-lacZ reporters (Jiang et al., 2002; Wu et al., 2017;
Received 27 June 2023; Accepted 8 January 2024 Yoshida et al., 2008), both interfrontal and sagittal sutures have

1
STEM CELLS AND REGENERATION Development (2024) 151, dev202116. doi:10.1242/dev.202116

been shown to be neural crest-derived, whereas the coronal suture is only mesoderm (Fig. 1B,L-N). Altogether, this shows that the
mesodermal. These studies focused on late embryonic/early post- sagittal suture has dual embryonic origin with a well-defined neural
natal stages, when cranial sutures are not yet clearly defined. To crest–mesoderm boundary that extends past the anterior edge of the
date, no definitive description of the neural crest–mesoderm parietal bones, recapitulating the pattern observed in early
boundary in the murine cranial vault has been established. Our development. This separates the sutures into two domains: the
overall hypothesis is that, in addition to suture proximity, the neural crest-derived sutures (interfrontal, coronal, squamous and
embryonic origins of the cranial sutures are also relevant when anterior sagittal) and the mesodermal sutures (lambdoid and
assessing repair. We hypothesise that calvarial sutures, notably posterior sagittal).
those with majority neural crest contributions, may harbour more We then performed the parallel experiment with Mesp1::cre
substantial skeletal stem cell populations. Thus, for non-neural- males bred with females carrying a Cre-responsive Rosa26RmTmG
crest-derived bone, such as the parietal bones, proximity to specific reporter (Saga et al., 1999; Muzumdar et al., 2007). The Mesp1::cre
sutures is crucial for homeostasis as well as cranial repair. transgenic animals carry Cre recombinase under the control of the
In this study, we use genetic labelling to compare Wnt1::cre+ and endogenous Mesp1 promoter, which activates initially a
Mesp1::cre+ lineages, which define the proposed neural crest- gastrulation, and later cranial mesenchyme (Yoshida et al., 2008).
mesoderm domains of the calvarium. We confirm that murine Yoshida and colleagues used Mesp1::cre animals to demonstrate
frontal bones heal more efficiently than parietal bones and note that the mesodermal contributions to the calvarial skeleton. We were
the position of the defect in relation to the sutures is paramount for able to confirm the putative neural crest–mesoderm boundary in
the outcome. We then demonstrate that Wnt1::cre+ cells are found mouse skulls at P40 (Fig. 1). Mesp1::cre; Rosa26RmTmG mice
in every subcritical defect regardless of the embryonic origin of the clearly showed boundaries at the midpoint of the sagittal suture
wounded bone, or the proximity to a neural crest suture. We find the (Fig. 1B′,F′-K′). We observed no positive labelling within the most
presence of Axin2+ and Gli1+ stem cell populations therein, noting rostral domains (Fig. 1C′-E′) adjacent to the interfrontal suture,
the contributions of Gli1-CreERT2-positive cells in the healing except the minimal marrow cavity. The posterior sections at the
wounds. This suggests that localised sources of neural crest cells and lamboid suture (Fig. 1L′-N′) were entirely positive for Mesp1::cre.
skeletal stem cells play a crucial role in healing of mesodermal skull Of note, using this approach we did see both Mesp1::cre-positive
bones. and -negative cells in the rostral part of the sagittal suture (‘SAG1’;
Fig. 1F′-H′), consistent with this domain having dual contributions
RESULTS AND DISCUSSION from neural crest and mesoderm.
Neural crest-mesoderm boundary separates cranial sutures
into distinct domains Suture proximity determines the outcome of a subcritical
In mouse, Wnt1::cre-dependent lineage labelling has demonstrated cranial defect
the neural crest-origin of the frontal bones whereas Mesp1::cre- Although regional differences in repair efficiency have previously
labelled mesoderm contributes to the parietal bones (Jiang et al., been reported, the dura mater and periosteum were thought to be the
2002; Wu et al., 2017; Yoshida et al., 2008). However, the main sources contributing osteoprogenitors for cranial repair
embryonic origins of the cranial sutures are debatable, as they rely (Ochareon and Herring, 2011; Greenwald et al., 2000). Here, we
on early observations that lacked resolution. Nevertheless, we and set out to determine the role of proximity to the sutural niche.
others have repeatedly confirmed a higher osteogenic potential of Subcritical defects (termed midfrontal and midparietal) of 1 mm
neural crest osteoblasts compared with mesoderm osteoblasts width were made in the frontal bone of adult P40 mice, equidistant
(Quarto et al., 2010; Doro et al., 2019; Behr et al., 2010a; Li to the interfrontal and coronal sutures, as well as in the parietal bone,
et al., 2013). We sought to determine the neural crest-mesoderm equidistant to the lambdoid, squamous, sagittal and coronal sutures
domain and the presence of stem cell lineages in calvarial sutures (Fig. 2A, unfilled, dotted outlines; 2B). After 4 weeks, we observed
using high-resolution lineage tracing. a striking difference between midfrontal and midparietal repair
Transgenic Wnt1::cre mice were crossed with mice carrying a (Fig. 2D), which accords with previous observations (Quarto et al.,
Cre-responsive Rosa26RmTmG reporter. In these mice, Wnt1::cre 2010; Li et al., 2010; Senarath-Yapa et al., 2013; Behr et al., 2010b).
expression begins approximately at embryonic day (E) 8.5 in the When the defects were made proximally to the sutures ( peri-coronal
dorsal neural tube, just as cranial neural crest cells are induced frontal, peri-coronal parietal and peri-lambdoid) (Fig. 2A, filled,
(Danielian et al., 1998). Soon after, these cells migrate into the dotted outlines; Fig. 2C), no significant difference in repair was seen
cranial vault and can be tracked with membrane green fluorescent between frontal and parietal bones (Fig. 2D), whereas the peri-
protein (mGFP). Cells lacking Wnt1::cre are labelled with coronal parietal defect healing was comparable to that of the frontal
membrane tomato (mTom) and are presumptive mesoderm. As bones (Fig. 2D). Our findings corroborate recent studies that show
expected, at postnatal day (P) 40, frontal bone shows exclusive that subcritical parietal defects are more likely to heal the closer they
neural crest contribution (mGFP+) as opposed to the parietal bones, are to the cranial sutures (Park et al., 2016). However, we did not see
which are of mesodermal origin, lacking mGFP (Fig. 1A). When we any specific increase in repair in the vicinity of a specific suture in
examined the midline sutures, we observed a clearly defined neural relation to the other. This shows that, even though the sutures may
DEVELOPMENT

crest–mesoderm boundary, which lies right at the middle of the contribute differently to cranial repair, the proximity to any suture is
sagittal suture, delimiting a neural crest domain (interfrontal, enough to provide subcritical healing, whereas the more distant
coronal and anterior-sagittal suture) and a mesoderm domain parietal defect (here termed midparietal) seems to exceed the
( posterior-sagittal and lambdoid suture) (Fig. 1A,B). A coronal maximum critical distance from a suture.
section at a more anterior part of the sagittal suture shows complete
absence of mesoderm (Fig. 1B,C-E), contrasting a more posterior Neural crest lineage contributes progenitors to any
section, which shows very few neural crest cells with abundant cranial defect
mesoderm (Fig. 1F-K). Likewise, the interfrontal suture exhibited We then assessed the extent to which Wnt1::cre-expressing neural
only neural crest (Fig. 1C-E), whereas the lambdoid suture showed crest cells infiltrate the repair site after injury. Subcritical defects

2
STEM CELLS AND REGENERATION Development (2024) 151, dev202116. doi:10.1242/dev.202116

Fig. 1. Embryonic origins of the cranial sutures. (A) Top view of P40 Wnt1Cre; Rosa26R mTmGmouse. Grey, brightfield; green, neural crest; red, non-
neural-crest tissue. Dashed box shows area of the midline explant in B. Scale bar: 2 mm. Schematic depicts the neural crest-mesoderm domain in the cranial
vault based on the linage tracing shown in A. (B) Confocal scan of the midline explant confirming the neural crest–mesoderm boundary at the middle of the
sagittal suture. Scale bar: 1 mm. (C-N) Confocal scans of coronal sections at different regions of the midline showing non-neural-crest tissue (C,F,I,L; red),
neural-crest tissues (D,G,J,M; green) and merged channels (E,H,K,N). Nuclear staining is shown in blue. Dashed lines in B show estimated planes of section
in C-N. IF, interfrontal suture (C-E), SAG1, sagittal suture region 1 (F-H), SAG2, sagittal suture region 2 (I-K), LAMB, lambdoid suture (L-N). Scale bar:
DEVELOPMENT

50 µm. n=3. (A′) Top view of P40 Mesp1::cre; Rosa26R mTmGmouse. Green, mesoderm; red, non-mesodermal tissue. Dashed box shows area of the midline
explant in B′. Scale bar: 2 mm. Schematic depicts the neural crest-mesoderm domain in the cranial vault based on the linage tracing shown in A′.
(B′) Confocal scan of the midline explant confirming the neural crest–mesoderm boundary at the middle of the sagittal suture. Scale bar: 1 mm. (C′-N′)
Confocal scans of coronal sections at different regions of the midline showing non-mesodermal tissue (C′,F′,I′,L′; red), mesodermal tissues (D′,G′,J′,M′;
green) and merged channels (E′,H′,K′,N′). Nuclear staining is shown in blue. Dashed lines in B′ show estimated planes of section in C′-N′ (abbreviations as
in B-N). Scale bar: 50 µm. n=3. Dashed lines in C-N,C′-N′ outline bone.

were made at six distinct locations in Wnt1::cre; Rosa26R +/mTmG interfrontal and coronal sutures, to lambdoid-parietal (O), which
mice (Fig. 2E). Defects varied from midfrontal (J) and coronal- is only surrounded by mesoderm-derived sutures. One week after
frontal (K), which are surrounded by neural crest-derived craniotomy, Wnt1::cre+ cells were detected around and/or within all

3
STEM CELLS AND REGENERATION Development (2024) 151, dev202116. doi:10.1242/dev.202116

DEVELOPMENT

Fig. 2. See next page for legend.

defects (Fig. 2F-I). This was clearly observed in coronal sections expected to be GFP+ (white arrows) as these bones are of neural
(Fig. 2J-O′). Even the defects in bones that are mesodermal in origin crest origin (Fig. 2J′,L′). In contrast, within parietal bone outlines
were filled with a large proportion of mGFP+ cells, suggesting the we observed a mix of GFP− (yellow arrows) and GFP+ (white
infiltration of neural crest-derived progenitors (Fig. 2M-O′). arrows) cells (Fig. 2L′,M′,O′), suggesting that new bone was
Importantly, within the frontal bone outlines, every cell was formed from neural crest-derived progenitors, given that parietal

4
STEM CELLS AND REGENERATION Development (2024) 151, dev202116. doi:10.1242/dev.202116

Fig. 2. Subcritical repair efficiency correlates with suture proximity Axin2+ cell content in the cranial sutures examined. These findings
irrespective of the calvarial bone, and neural crest cells are recruited to are consistent with prior proposals that the Axin2+ cells could be a
every defect. (A) Schematic showing the position of cranial defects in
subset of Gli1+ progenitors (Maruyama et al., 2016; Wilk
relation to adjacent sutures. Green outline, midfrontal; solid green, peri-
coronal frontal; solid red, peri-coronal parietal; red outline, midparietal; solid et al., 2017). Further assessment of Axin2 mRNA expression on
pink, peri-lambdoid. (B,C) Top view microCT scan of CD1 mouse heads the Gli1+-derived cells would be necessary to define definitively
4 weeks after 1 mm subcritical defect surgery. (B) Defects equidistant from these subpopulations of skeletal stem cells.
surrounding sutures in right frontal and left parietal bones. (C) Defects To investigate whether Gli1+ osteoprogenitors could populate all
proximal to coronal and interfrontal suture (right), coronal and sagittal suture healing defects, we labelled Gli1::creERT2; Rosa26 Tomato mice as
(top left), and sagittal and lambdoid suture (bottom left). (D) Bone volume
described. Subcritical defects were made at five sites of Gli1-
fraction (BVF) analysis over a 4-week period post-subcritical defect surgery.
Lower and upper box limits indicate the lower and upper quartile of BVF
CreERT2; Rosa26 Tomato/+ mice as indicated (Fig. 4A) and Gli1+
values, respectively. Horizontal line indicates the median. Upper and lower cells were examined in healing wounds 1-week post-surgery.
whiskers indicate minimum and maximum BVF values. n=6. *P<0.05, Coronal sections of the wounds shown in Fig. 4B,C reveal that
**P<0.001 (two-tailed, unpaired t-test). (E) Schematic showing the position Gli1+ cells (in red) were present in all wounds (Fig. 4D-H). Cells
of six subcritical defects (1 mm). Sutures are indicated in the picture as were present regardless of whether wounds were made in neural
follows: COR, coronal suture; IF, interfrontal suture; SQ, squamous suture; crest-derived frontal bone (Fig. 4D,E) or parietal (Fig. 4G,H). In the
SAG, sagittal suture; LAMB, lambdoid suture. (F-I) Top view of 40-day-old
Wnt1-Cre; Rosa26RGFP mice 1 week after the surgical procedure. Green
parietal bones, Gli1+ cells were also seen regardless of proximity to
staining indicates neural crest-derived tissue. (J-O′) Coronal sections at the the sagittal suture (Fig. 4G, compared with Fig. 4F,H). This suggests
centre of each defect described in E. Neural crest-derived cells are shown in that, although midparietal defects (Fig. 4G) were substantially more
green, nuclear staining in blue. Dashed lines outline the bones, and the distant to the sutures than the other defects were, they were still
boxed area is shown at high magnification in J′-O′. Yellow arrowheads supplied with suture-derived osteoprogenitors, revealing the
indicate non-neural crest-derived cells, white arrowheads neural crest- reparative capability of the suture mesenchyme in relation to
derived cells. n=4.
distant defects. In the future, a quantitative assessment of the
signalling events and cell content will be important reveal the
mechanistic links between distance, cell identity and healing.
bones are expected to be mesodermal. One exception is the
midparietal defect (Fig. 2N′), which maintained its original flat Conclusion
edges, as obtained by the cylindrical nature of the drill bit, indicating Our data suggest that the embryonic origins of the cranial bone may
minimal repair. This wound had an absence of green cells within the predict the outcome of defect repair based on the neural crest
bone outline. This is consistent with previous observations of poor composition of the intervening sutures and the presence of suture-
healing capacity of the mid-parietal bone region. residing osteoprogenitors relative to the calvarial wound site.
Although the osteogenic potential of neural crest versus However, it is important to note that our observations are based on
mesoderm sutures has not yet been defined, it is intriguing that several crucial assumptions. First, the genetic approaches used in
Wnt1::cre + neural crest cells seem to contribute to the repair of our studies, although powerful, cannot exclude the possibility that
every defect, including those in a mesoderm-dominant area these transgenic lines are not reactivated later in development, or
(Fig. 2). Although we propose that the suture proximity is a key subject to unknown transcriptional influences, or are simply not Cre
factor in repair capacity, it is also worth noting that the underlying responsive. Second, it is important to note that meninges, notably
dura mater is also neural crest derived, and a recent study has the dura mater, are also neural crest derived (Yoshida et al., 2008;
shown contributions of dura mater cells to suture regeneration Gagan et al., 2007; Dasgupta and Jeong, 2019). Indeed, we and
(Yu et al., 2021). Regardless, neural crest-derived cells appear to others have demonstrated that the dura mater has osteogenic
be required for efficient repair, supporting previous observations capacity (Petrie et al., 2008; Peptan et al., 2007), and when
that neural crest osteoprogenitors have higher osteogenic co-cultured with parietal cells are capable of nucleating
capabilities than other mesodermally derived calvarial cell osteogenesis (Doro et al., 2019). Finally, the composition of the
populations. sutural mesenchyme is surely dynamic, as the biological
requirements change from embryogenesis to adult aging. Cell
Stem cell populations are present in every calvarial suture migration and mixing in adult life has been reported to occur in the
and contribute to calvarial repair sutures and in the meninges (Gagan et al., 2007; Deckelbaum et al.,
Several putative calvarial stem cell markers have been identified: 2012). Furthermore, the meninges, which are comprised of neural
the Hedgehog-pathway transcription factor Gli1 (Zhao et al., 2015), crest cells during development, seem later to be invaded by
the Wnt-responsive gene Axin2 (Maruyama et al., 2016) and mesodermal derivatives, leaving the remaining neural crest cells to
the transcription factor Prx1 (Prrx1) (Wilk et al., 2017). Here, we act as resident stem cells in later life.
investigate the presence of Gli1+ and Axin2+ stem cell populations In the future, it will be important to track more localised cell
using Gli1::creERT2 or Axin2::creER drivers in combination with contributions, such as dura mater-derived cells or the adjacent
Rosa26 Tomato and Rosa26R mTmG reporters, respectively. To label periosteal cells. It will also be important to assess the environment
DEVELOPMENT

cells derived from Gli1 + and Axin2 + populations, 38-day-old mice unique to the frontal versus parietal bones: for example, Marghoub
were given tamoxifen and the heads were collected 2 days after and colleagues have demonstrated distinct mechanical forces in
injection. Top view and coronal sections of Gli1-CreERT2; the anterior versus the posterior sutures, owing to differences in
Rosa26 Tomato/+ heads revealed the abundant presence of Gli1+ bone size, anatomy and underlying brain morphology (Marghoub
cells in all calvarial sutures (Fig. 3A-F). The analogous experiment et al., 2018). Altogether, although we cannot definitively state
with Axin2-CreERT2; Rosa26R+/mTmG animals revealed the that availability of neural crest cells is the key element, we
presence of Axin2+-derived cells also in every suture; however, can nevertheless conclude that lineage identity and spatial
they were sparse in comparison with Gli1+ (compare Fig. 3G-L with positioning are both important in relation to the healing of
3A-F). Overall, we observed no apparent difference in Gli1+ and calvarial defects.

5
STEM CELLS AND REGENERATION Development (2024) 151, dev202116. doi:10.1242/dev.202116

Fig. 3. Gli1+-derived and Axin2+-derived

progenitors are found in all cranial sutures.
(A) Top view of 40-day-old Gli1CreERT/+;
Rosa26R TdTomato mouse 48 h after tamoxifen
induction. Grey, brightfield; red, Gli1+ domain. Red
staining on the right side is autofluorescence of
opaque tissue remaining in the sample. COR,
coronal suture; IF, interfrontal suture; SQ,
squamous suture; SAG, sagittal suture; LAMB,
lambdoid suture. Scale bar: 2 mm. Schematics
show the Gli1+ domain in the transgenic mouse
calvarium. (B-F) Coronal sections of the calvarial
sutures at the regions marked in A. Dashed lines
show the bone outline. White arrows show red
Gli1+ cells. Scale bar: 250 µm. n=3. (G) Top view
of 40-day-old Axin2CreERT/+;
Rosa26RmTmGmouse 48 h after tamoxifen
induction (red channel not shown). Grey,
brightfield; green, Axin2+ domain. Abbreviations as
in A-F. Scale bar: 2 mm. Schematics show the
Axin2+ domain in the transgenic mouse calvarium.
(H-L) Coronal sections of the calvarial sutures at
the regions marked in G. Dashed lines show the
bone outline. Axin2+ cells are in green. White
arrows show green Axin2+ cells. bone marrow
(bm), muscle (m), dura mater (dm) and periosteum
( p) are notoriously autofluorescent tissues.
Scale bar: 250 µm.

DEVELOPMENT

MATERIALS AND METHODS and Use Committee of Tokyo Medical and Dental University [A2019-
Animal procedures 060C3 (S.I.)].
All procedures were approved by King’s College London ethical review Mouse lines used were CD-1 mice (obtained from Charles River
process and performed in accordance with UK Home Office guidelines Laboratories), and Rosa26RmTmG (MGI ID 3716464), Rosa26RTdTomato
Project Licence P8D5E2773 (K.J.L.) or by the Institutional Animal Care (MGI ID 3809523), Rosa26R-eGFP (MGI ID 2136519) mouse

6
STEM CELLS AND REGENERATION Development (2024) 151, dev202116. doi:10.1242/dev.202116

Fig. 4. Gli1+-derived cells are found in every subcritical defect after 1 week. (A) Schematics show sites of 1 mm subcritical defects in P40
Gli1CreERT/+; Rosa26R TdTomatomice. Circular outlines mark the positions of the wounds in relation to the different bones and sutures. (B,C) Top-view CT
scans of heads after 1 week from the surgical procedure. Red dashed lines correspond to the approximate plane of sections in D-H. (D-H) Confocal scans of
coronal sections at different wound sites 1-week post-surgery. Red, Gli1+ cells. Dashed lines outline bone. if, interfrontal suture; c, coronal suture; s, sagittal
suture; l, lambdoid suture. Scale bar: 250 µm. n=3.
DEVELOPMENT

reporter lines (all described previously: Muzumdar et al., 2007; Subcritical defects
Madisen et al., 2010; Mao et al., 2001). The following Cre drivers P40 mice were weighed and anaesthetised with an appropriate dose (10 μl/g of
were used: Wnt1::cre (MGI ID 2386570), Axin2::CreERT2 (MGI ID body weight) of a 10 mg/ml ketamine/2 mg/ml xylazine cocktail (Vetalar®,
5433373), Gli1::CreERT2 (MGI ID 3053957), Mesp1::cre (MGI ID Zoetis; Rompun®, Dechra). The state of deep anaesthesia was confirmed
2176467) (Danielian et al., 1998; Saga et al., 1999; van Amerongen through tail flick test and hind paw withdrawal response. Once the animals
et al., 2012; Ahn and Joyner, 2004). were heavily sedated, the fur on the top of the head was shaved using a hair

7
STEM CELLS AND REGENERATION Development (2024) 151, dev202116. doi:10.1242/dev.202116

trimmer. A sagittal incision was performed along the midline with a scalpel. Confocal microscopy was performed on a Leica Microsystems CMS TCS
The periosteal layer was then removed with the help of a cotton bud and 1- SP5 DM16000. Image sequences were reconstructed using Fiji (ImageJ)
mm-width defects were drilled at top speed (50,000 rpm) into frontal and analysis software.
parietal bones of the mouse skull using a dental hand drill (Handy-ECO 1000,
Marathon®). The defects were carefully performed to avoid injuring the dura Acknowledgements
mater. The bone surface was then rinsed with sterile PBS to remove any debris We thank members of the Liu and Iseki labs and colleagues at the Centre for
and the sagittal incision was sutured with 6-0 ETHILON® nylon absorbable Craniofacial and Regenerative Biology for support with experiments and scientific
suture. Finally, the mice were moved to a 28°C incubator until full recovery. discussions. We thank William Barrell and Jade Desjardins support with mouse
work. We also thank the Biological Services Unit at Guy’s Hospital for all the support
For the CD1 mice shown in Fig. 2, six mice received two wounds distant from
with the mouse work.
the sutures and six mice received three wounds proximal to the sutures. Wnt1-
Cre; Rosa 26RGFP mice (n=4) all received six wounds each, according to the
Competing interests
schematics. The Gli1CreERT; Rosa26RtdTomato shown in Fig. 4 (n=3) received The authors declare no competing or financial interests.
separate wounds similarly to CD1 in Fig. 2, i.e. three animals with two
wounds and the other three with three wounds. Author contributions
Conceptualization: D.D., A.E.G., S.I., K.J.L.; Methodology: D.D., A.L., J.S.L., C.H.,
Tamoxifen injection M.K.; Validation: D.D.; Formal analysis: D.D., K.J.L.; Investigation: D.D., A.L., J.S.L.,
Cre induction in CreERT mice was performed by peritoneal injection of a A.K.R., C.H., M.K.; Resources: S.I., K.J.L.; Data curation: D.D., K.J.L.; Writing -
10 mg/ml tamoxifen solution (Sigma-Aldrich) in the adult mouse at the original draft: D.D., A.E.G., K.J.L.; Writing - review & editing: D.D., A.L., J.S.L.,
desired stage. The solution was previously prepared by dissolving 10 mg of A.K.R., C.H., M.K., A.E.G., S.I., K.J.L.; Visualization: D.D.; Supervision: A.E.G., S.I.,
tamoxifen into 100 μl of absolute ethanol and 900 μl of corn oil. The dosage K.J.L.; Project administration: D.D., S.I., K.J.L.; Funding acquisition: S.I., K.J.L.
was determined according to the following: 1.5 mM/g of body weight in a
volume of 7.5 μl/g of body weight. Funding
This work was supported by King’s College London Dental Institute Seed Funding
(K.J.L.), the Biotechnology and Biological Sciences Research Council (BB/I021922/
Micro-CT scanning 1 and BB/R015953/1 to K.J.L.), the Medical Research Council (PC21044 to K.J.L.),
All head samples were fixed for 48 h at room temperature in 4% funding from Brazil CAPES (Coordenaçao ̃ de Aperfeiçoamento de Pessoal de Nı́vel
paraformaldehyde and scanned using a Scanco Medical µCT50® with the Superior) (D.D.), a Japan Society for the Promotion of Science Short-term
following settings: energy 70 kV, intensity 114 µA, resolution 10 µm/voxel. Postdoctoral Fellowship (D.D.) and grants-in-aid from the Ministry of Education,
The images were then reconstructed on Parallax Microview® with isosurface Culture, Sports, Science and Technology of Japan (MEXT) 21H03098 (S.I.). Open
image threshold set to 6000 and surface quality factor set to 40% with Access funding provided by Biotechnology and Biological Sciences Research
decimation factor of 0%. Council and the Medical Research Council. Deposited in PMC for immediate
release.
To determine the bone volume fraction of each subcritical defect, a
cylindrical region of interest calibrated to the volume of a 1 mm defect was
used as total volume (mm3) and bone volume (mm3) was then measured Data availability
All relevant data can be found within the article.
using the Bone Analysis Tool on MicroView®.
The people behind the papers
Sample fixation and sectioning
This article has an associated ‘The people behind the papers’ interview with some of
Samples were fixed in 4% paraformaldehyde for 48 h at 4°C. After three the authors.
PBS washes, the skull cap was dissected and decalcified in 10% formic
acid. After three more PBS washes, the samples were moved to a 30% Peer review history
sucrose solution in PBS until they sunk. The embedding solution was then The peer review history is available online at https://journals.biologists.com/dev/
replaced with a 30% sucrose solution mixed (1:1) with OCT compound lookup/doi/10.1242/dev.202116.reviewer-comments.pdf
(CellPath®). The samples were incubated at 4°C for another 48 h. Finally,
the samples were moved and oriented in a plastic Tissue-Tek® Cryomold® References
filled with OCT compound. The mould was quickly moved into a dry-ice bath Ahn, S. and Joyner, A. L. (2004). Dynamic changes in the response of cells to
with absolute ethanol until the OCT block was fully solidified. Cryosections positive hedgehog signaling during mouse limb patterning. Cell 118, 505-516.
of 15 μm thickness were taken using OFT5000® cryostat microtome (Bright doi:10.1016/j.cell.2004.07.023
Behr, B., Longaker, M. T. and Quarto, N. (2010a). Differential activation of
Instruments®) and mounted onto Superfrost Ultra Plus® slides (Thermo Fisher
canonical Wnt signaling determines cranial sutures fate: a novel mechanism for
Scientific®), which were then stored at −80°C for future use. sagittal suture craniosynostosis. Dev. Biol. 344, 922-940. doi:10.1016/j.ydbio.
2010.06.009
Immunostaining Behr, B., Panetta, N. J., Longaker, M. T. and Quarto, N. (2010b). Different
GFP and RFP (Tomato) fluorescence in all experiments was assessed endogenous threshold levels of Fibroblast Growth Factor-ligands determine the
by immunofluorescence staining. Primary antibodies were: anti-GFP healing potential of frontal and parietal bones. Bone 47, 281-294. doi:10.1016/
j.bone.2010.05.008
(ab13970, Abcam®), anti-RFP (5F*, ChromoTek®). Secondary antibodies
Brault, V., Moore, R., Kutsch, S., Ishibashi, M., Rowitch, D. H., McMahon, A. P.,
were: Alexa Fluor™ 488 goat anti-chicken (A-11039, Invitrogen®), Alexa Sommer, L., Boussadia, O. and Kemler, R. (2001). Inactivation of the β-catenin
Fluor™ 568 goat anti-rat (A-11077, Invitrogen®). After dissolution of the gene by Wnt1-Cre-mediated deletion results in dramatic brain malformation and
OCT in PBS, slides were blocked with 1% bovine serum albumin (Sigma- failure of craniofacial development. Development 128, 1253-1264. doi:10.1242/
Aldrich®), 10% goat serum (Gibco®), 0.05% Tween 20 (Sigma-Aldrich®) dev.128.8.1253
in PBS. Primary antibodies were diluted to 1:100 in blocking buffer and Danielian, P. S., Muccino, D., Rowitch, D. H., Michael, S. K. and McMahon, A. P.
(1998). Modification of gene activity in mouse embryos in utero by a tamoxifen-
DEVELOPMENT

incubated overnight at 4°C. Secondary antibodies were diluted to 1:500 in

inducible form of Cre recombinase. Curr. Biol. 8, 1323-1326. doi:10.1016/S0960-
the same buffer and incubated for 1 h at room temperature. After both 9822(07)00562-3
primary and secondary antibody incubation steps, slides were washed three Dasgupta, K. and Jeong, J. (2019). Developmental biology of the meninges.
times for 10 min each wash in PBS with 0.05% Tween 20 (Sigma-Aldrich). Genesis 57, 1-12. doi:10.1002/dvg.23288
After the immunostaining procedure, sections were mounted with Deckelbaum, R. A., Holmes, G., Zhao, Z., Tong, C., Basilico, C. and Loomis,
Fluoroshield™ mounting medium with DAPI (F6057, Sigma-Aldrich®). C. A. (2012). Regulation of cranial morphogenesis and cell fate at the neural crest-
mesoderm boundary by engrailed 1. Development 139, 1346-1358. doi:10.1242/
dev.076729
Microscopy Doro, D. H., Grigoriadis, A. E. and Liu, K. J. (2017). Calvarial suture-derived stem
A stereoscope (Nikon SMZ1500) with an attached camera (Nikon digital cells and their contribution to cranial bone repair. Front. Physiol. 8, 956. doi:10.
sight DS-Fi1) was used to take top view fluorescent images of whole heads. 3389/fphys.2017.00956

8
STEM CELLS AND REGENERATION Development (2024) 151, dev202116. doi:10.1242/dev.202116

Doro, D., Liu, A., Grigoriadis, A. E. and Liu, K. J. (2019). The osteogenic potential Ochareon, P. and Herring, S. W. (2011). Cell replication in craniofacial periosteum:
of the neural crest lineage may contribute to craniosynostosis. Mol. Syndromol. appositional vs. resorptive sites. J. Anat. 218, 285-297. doi:10.1111/j.1469-7580.
10, 48-57. doi:10.1159/000493106 2010.01336.x
Gagan, J. R., Tholpady, S. S. and Ogle, R. C. (2007). Cellular dynamics and tissue Park, S., Zhao, H., Urata, M. and Chai, Y. (2016). Sutures possess strong
interactions of the dura mater during head development. Birth Defects regenerative capacity for calvarial bone injury. Stem Cells Dev. 25, 1801-1807.
Res. C. Embryo Today 81, 297-304. doi:10.1002/bdrc.20104 doi:10.1089/scd.2016.0211
Greenwald, J. A., Mehrara, B. J., Spector, J. A., Warren, S. M., Crisera, F. E., Peptan, I. A., Hong, L. and Evans, C. A. (2007). Multiple differentiation potentials of
Fagenholz, P. J., Bouletreau, P. J. and Longaker, M. T. (2000). Regional neonatal dura mater-derived cells. Neurosurgery 60, 346-352. doi:10.1227/01.
differentiation of cranial suture-associated dura mater in vivo and in vitro: implications NEU.0000249278.72063.59
Petrie, C., Tholpady, S., Ogle, R. and Botchwey, E. (2008). Proliferative capacity
for suture fusion and patency. J. Bone Miner. Res. 15, 2413-2430. doi:10.1359/jbmr.
and osteogenic potential of novel dura mater stem cells on poly-lactic-co-glycolic
2000.15.12.2413
acid. J. Biomed. Mater. Res. A 85, 61-71. doi:10.1002/jbm.a.31367
Hari, L., Brault, V., KleÌber, M., Lee, H.-Y., Ille, F., Leimeroth, R., Paratore, C.,
Quarto, N., Wan, D. C., Kwan, M. D., Panetta, N. J., Li, S. and Longaker, M. T.
Suter, U., Kemler, R. and Sommer, L. (2002). Lineage-specific requirements of
(2010). Origin matters: differences in embryonic tissue origin and Wnt signaling
beta-catenin in neural crest development. J. Cell Biol. 159, 867-880. doi:10.1083/ determine the osteogenic potential and healing capacity of frontal and parietal
jcb.200209039 calvarial bones. J. Bone Miner. Res. 25, 1680-1694. doi:10.1359/jbmr.091116
Jiang, X., Iseki, S., Maxson, R. E., Sucov, H. M. and Morriss-Kay, G. M. (2002). Saga, Y., Miyagawa-tomita, S., Takagi, A., Kitajima, S. and Miyazaki, J. (1999).
Tissue origins and interactions in the mammalian skull vault. Dev. Biol. 241, MesP1 is expressed in the heart precursor cells and required for the formation of a
106-116. doi:10.1006/dbio.2001.0487 single heart tube. Development 127, 3437-3447. doi:10.1242/dev.126.15.3437
Lanza, R., Langer, R. and Vacanti, J. (2020). Principles of Tissue Engineering. Senarath-Yapa, K., Li, S., Meyer, N. P., Longaker, M. T. and Quarto, N. (2013).
Principles of Tissue Engineering. Elsevier – Academic Press. doi:10.1016/B978- Integration of multiple signaling pathways determines differences in the
0-12-818422-6.00089-7 osteogenic potential and tissue regeneration of neural crest-derived and
Li, S., Quarto, N. and Longaker, M. T. (2010). Activation of FGF signaling mediates mesoderm-derived calvarial bones. Int. J. Mol. Sci. 14, 5978-5997. doi:10.3390/
proliferative and osteogenic differences between neural crest derived frontal and ijms14035978
mesoderm parietal derived bone. PLoS One 5, e14033. doi:10.1371/journal. Skogh, A. C. D., Kihlströ m, L., Neovius, E., Persson, C., Beckman, M. O. and
pone.0014033 Engstrand, T. (2013). Variation in calvarial bone healing capacity: a clinical study
Li, S., Meyer, N. P., Quarto, N. and Longaker, M. T. (2013). Integration of multiple on the effects of BMP-2-hydrogel or bone autograft treatments at different cranial
signaling regulates through apoptosis the differential osteogenic potential of locations. J. Craniofac. Surg. 24, 339-343. doi:10.1097/SCS.0b013e31827ff2b6
neural crest-derived and mesoderm-derived osteoblasts. PLoS One 8, e58610. van Amerongen, R., Bowman, A. N. and Nusse, R. (2012). Developmental stage
doi:10.1371/journal.pone.005861 and time dictate the fate of Wnt/β-catenin-responsive stem cells in the mammary
Madisen, L., Zwingman, T. A., Sunkin, S. M., Oh, S. W., Zariwala, H. A., Gu, H., gland. Cell Stem Cell 11, 387-400. doi:10.1016/j.stem.2012.05.023
Ng, L. L., Palmiter, R. D., Hawrylycz, M. J., Jones, A. R. et al. (2010). A robust Wilk, K., Yeh, S.-C. A., Mortensen, L. J., Ghaffarigarakani, S., Lombardo, C. M.,
and high-throughput Cre reporting and characterization system for the whole Bassir, S. H., Aldawood, Z. A., Lin, C. P. and Intini, G. (2017). Postnatal
calvarial skeletal stem cells expressing PRX1 reside exclusively in the calvarial
mouse brain. Nat. Neurosci. 13, 133-140. doi:10.1038/nn.2467
sutures and are required for bone regeneration. Stem Cell Rep. 8, 933-946. doi:10.
Mao, X., Fujiwara, Y., Chapdelaine, A., Yang, H. and Orkin, S. H. (2001).
1016/j.stemcr.2017.03.002
Activation of EGFP expression by Cre-mediated excision in a new ROSA26
Wu, T., Chen, G., Tian, F. and Liu, H.-X. (2017). Contribution of cranial neural crest
reporter mouse strain. Blood 97, 324-326. doi:10.1182/blood.V97.1.324
cells to mouse skull development. Int. J. Dev. Biol. 61, 495-503. doi:10.1387/ijdb.
Marghoub, A., Libby, J., Babbs, C., Pauws, E., Fagan, M. J. and Moazen, M.
170051gc
(2018). Predicting calvarial growth in normal and craniosynostotic mice using a Yoshida, T., Vivatbutsiri, P., Morriss-Kay, G., Saga, Y. and Iseki, S. (2008). Cell
computational approach. J. Anat. 232, 440-448. doi:10.1111/joa.12764 lineage in mammalian craniofacial mesenchyme. Mech. Dev. 125, 797-808.
Maruyama, T., Jeong, J., Sheu, T.-J. and Hsu, W. (2016). Stem cells of the suture doi:10.1016/j.mod.2008.06.007
mesenchyme in craniofacial bone development, repair and regeneration. Nat. Yu, M., Ma, L., Yuan, Y., Ye, X., Montagne, A., He, J., Ho, T.-V., Wu, Y., Zhao, Z.,
Commun. 7, 10526. doi:10.1038/ncomms10526 Sta Maria, N. et al. (2021). Cranial suture regeneration mitigates skull and
Muzumdar, M. D., Tasic, B., Miyamichi, K., Li, L. and Luo, L. (2007). A global double- neurocognitive defects in craniosynostosis. Cell 184, 243-256.e18. doi:10.1016/
fluorescent Cre reporter mouse. Genesis 45, 593-605. doi:10.1002/dvg.20335 j.cell.2020.11.037
Nakashima, K. and de Crombrugghe, B. (2003). Transcriptional mechanisms in Zhao, H., Feng, J., Ho, T.-V., Grimes, W., Urata, M. and Chai, Y. (2015). The suture
osteoblast differentiation and bone formation. Trends Genet. 19, 458-466. doi:10. provides a niche for mesenchymal stem cells of craniofacial bones. Nat. Cell Biol.
1016/S0168-9525(03)00176-8 17, 386-396. doi:10.1038/ncb3139

DEVELOPMENT