ORIGINAL RESEARCH ARTICLE

published: 12 September 2014

doi: 10.3389/fphys.2014.00345

Sox2 acts as a rheostat of epithelial to mesenchymal

transition during neural crest development
Nikolaos Mandalos 1 , Muriel Rhinn 2 , Zoraide Granchi 3 , Ioannis Karampelas 1,4 , Thimios Mitsiadis 3 ,
Aris N. Economides 5 , Pascal Dollé 2 and Eumorphia Remboutsika 1*
1
Stem Cell Biology Laboratory, Division of Molecular Biology and Genetics, Biomedical Sciences Research Centre “Alexander Fleming,” Vari-Attica, Greece
2
Institut de Génétique et de Biologie Moléculaire et Cellulaire, INSERM, U964, CNRS, UMR7104, Université de Strasbourg, Illkirch, France
3
Orofacial Development and Regeneration Unit, Faculty of Medicine, Institute of Oral Biology, University of Zurich, ZZM, Zurich, Switzerland
4
Department of Neurosurgery, University Hospitals Case Medical Center, Cleveland, OH, USA
5
Regeneron Pharmaceuticals, Tarrytown, New York, NY, USA

Edited by: Precise control of self-renewal and differentiation of progenitor cells into the cranial neural
Gianpaolo Papaccio, Second crest (CNC) pool ensures proper head development, guided by signaling pathways such
University of Naples, Italy
as BMPs, FGFs, Shh and Notch. Here, we show that murine Sox2 plays an essential
Reviewed by:
role in controlling progenitor cell behavior during craniofacial development. A “Conditional
Petros Papagerakis, University of
Michigan, USA by Inversion” Sox2 allele (Sox2COIN ) has been employed to generate an epiblast
Virginia Tirino, Department of ablation of Sox2 function (Sox2 EpINV ). Sox2 EpINV /+(H) haploinsufficient and conditional
Experimental Medicine, Italy (Sox2 EpINV /mosaic ) mutant embryos proceed beyond gastrulation and die around E11. These
*Correspondence: mutant embryos exhibit severe anterior malformations, with hydrocephaly and frontonasal
Eumorphia Remboutsika, Stem Cell
truncations, which could be attributed to the deregulation of CNC progenitor cells during
Biology Laboratory, Division of
Molecular Biology and Genetics, their epithelial to mesenchymal transition. This irregularity results in an exacerbated and
Biomedical Sciences Research aberrant migration of Sox10+ NCC in the branchial arches and frontonasal process of
Centre “Alexander Fleming,” the Sox2 mutant embryos. These results suggest a novel role for Sox2 as a regulator
34 Fleming Str., 16672 Vari, Attica,
Greece
of the epithelial to mesenchymal transitions (EMT) that are important for the cell flow in
e-mail: [email protected] the developing head.
www.eumorphiaremboutsika.com
Keywords: Sox genes, SoxB, neurocristopathies, craniofacial development, stem cells, neural progenitors,
heterochrony

INTRODUCTION Neural crest (NC) development depends on the activation

The head develops from anteriorly located cells of the epi- of NCC-specific genes at the neural plate border (Gammill and
blast. These cells form the neuroectoderm that gives rise to Bronner-Fraser, 2003; Heeg-Truesdell and Labonne, 2004; Huang
the brain and craniofacial structures stemming via epithelial and Saint-Jeannet, 2004). A number of these genes belong to
to mesenchymal transitions (EMT). Balanced control between the Sox (Sry HMG-box) family of transcription factors (sub-
self-renewal of neural progenitors and their differentiation into divided into A-H groups) harboring an HMG-box as a DNA
cranial neural crest cells (NCCs) ensures proper head devel- binding domain (Pevny and Lovell-Badge, 1997; Wegner, 1999;
opment. Cranial NCCs (CNCCs) arise from a NCC pool Wilson and Koopman, 2002; Bernard and Harley, 2010; Kamachi
derived from the neural ectoderm, and give rise to most of et al., 2013; Karnavas et al., 2013). Whereas all SoxE genes show
the peripheral nervous system and craniofacial structures. NCCs expression in NCC progenitors at some point following NC
are induced by interactions between the neuroectoderm and induction, differences exist in the onset and sequence of events.
adjacent non-neural ectoderm (Dickinson et al., 1995; Selleck Induction of NCC formation is triggered by the expression of
and Bronner-Fraser, 1995). These interactions are orchestrated SoxE genes (Gammill and Bronner-Fraser, 2002), such as Sox8,
by a combination of signaling molecules such as Wnt pro- Sox9, and Sox10 (Bowles et al., 2000). These genes are already
teins, bone morphogenetic proteins (BMPs) (Liem et al., 1995), expressed in all premigratory NCCs, while later their expression
fibroblast growth factors (FGFs), retinoic acid and proteins of becomes restricted to distinct NCC-derived subpopulations (Stolt
the Notch pathway (Labonne and Bronner-Fraser, 1998; Aybar et al., 2004; Betancur et al., 2011). Sox8 expression occurs before
et al., 2002; Aybar and Mayor, 2002; Christiansen et al., 2002; NCC migration from the neural tube, followed by Sox9, and
Endo et al., 2002; Garcia-Castro et al., 2002; Villanueva et al., shortly after Sox10 (Cheng et al., 2000). Sox9 expression over-
2002; Wu et al., 2003). NCCs delaminate from the dorsal laps with that of a number of NCC determinant genes, such
neural tube, migrate along defined territories of the cranio- as FoxD3, Bmp4, cadherin 6b, Slug, and RhoB (Liu and Jessell,
facial complex and finally differentiate into many cell types, 1998; Briscoe et al., 2003). BMP signaling drives the induction,
including neurons, glial cells, Schwann cells, melanocytes, and formation, determination and migration of CNCCs (Nie et al.,
cells of the connective tissue (Ayer-Le Lievre and Le Douarin, 2006). Cadherin 6b establishes the premigratory NCC domain
1982). in the neural tube (Taneyhill, 2008). Slug induces premigratory

www.frontiersin.org September 2014 | Volume 5 | Article 345 | 1

Mandalos et al. Head development is fine tuned by Sox2

and migratory CNCC behavior (Del Barrio and Nieto, 2002). GENOTYPING
Foxd3 induces the segregation of NCC from the neural tube (Kos Tail, yolk sack or embryonic tissues were isolated and pro-
et al., 2001). Rho is involved in delamination of NCC from the cessed according to previously described methodology (Mandalos
dorsal neural tube (Rutishauser and Jessell, 1988; Cordero et al., et al., 2012). PCR amplification conditions and primers used are
2011). As migration starts (Nakagawa and Takeichi, 1995; Jessel described elsewhere (Mandalos et al., 2012).
and Weiss, 1998), Slug, RhoB, N-cadherin, and cadherin 6b are
down-regulated at the trunk level (Akitaya and Bronner-Fraser, EMBRYO PROCESSING AND HISTOLOGICAL ANALYSIS
1992; Monier, 1995), while FoxD3 expression persists in all migra- For staging of the embryos, midday of the vaginal plug was
tory NCCs (Cheng et al., 2000; Dottori et al., 2001). On the other considered as embryonic day 0.5 (E0.5). All embryos were har-
hand, Sox10 persists only in the trunk NCC populations (Cheng vested in cold 0.12 M phosphate-buffered saline (PBS, pH 7.5).
et al., 2000; Remboutsika et al., 2011). For histological analysis, embryos were fixed with 10% forma-
Amongst SoxB genes (Sox1,2,3,14, and 21), Sox2 has been lin for 24 h at room temperature and then washed several times
reported to play a cell-autonomous role in NCC development with PBS, placed in embedding cassettes and sectioned in a Leica
(Pan and Schultz, 2011). Sox2 is one of the early-activated genes RM2125RT microtome. Paraffin sections (10 µm) were stained
in the developing neural plate (Graham et al., 2003; Wen et al., with hematoxylin and eosin and mounted with xylene based
2008; Hutton and Pevny, 2011), and its expression is reduced in mounting medium, according to standard procedures (Fischer
the dorsal neural tube as NCCs segregate and migrate. Thereafter, et al., 2008; Cardiff et al., 2014).
Sox2 expression is upregulated in a subset of cells that arrived Embryos were harvested and fixed with 4% paraformalde-
at their final destination, and gradually becomes restricted to hyde (PFA) in PBS overnight at 4◦ C and thoroughly washed
the glial sublineages (Aquino et al., 2006). Sox2 prevents ter- with PBS. Fixed embryos were incubated with 20% sucrose
minal differentiation of Schwann cells (Wakamatsu et al., 2004; overnight at 4◦ C for cryoprotection before they were embed-
Le et al., 2005) and represses the melanocyte fate (Laga et al., ded with O.C.T. compound (VWR International), snap-frozen
2010). Ectopic expression of Sox2 in embryonic ectoderm and in dry ice and stored at −80◦ C. Sagittal sections were pre-
neural plate explants reveals that Sox2 is sufficient to inhibit pared using a Leica cryostat. Cryosections (10–12 µm) were col-
NCC formation both in chick and mouse embryos (Papanayotou lected on Superfrost Plus microscope slides (VWR International)
et al., 2008; Remboutsika et al., 2011). It has been postulated and stored at −20◦ C before analysis. For in situ hybridization,
that Sox2 counteracts NC development (Scherson et al., 1993; embryos were fixed overnight in PFA 4% in PBS then rinsed
Placzek and Briscoe, 2005; Remboutsika et al., 2011), as the NCC three times in PBS/Tween (0.1%) followed by three times wash in
marker Slug is only expressed at regions of low Sox2 expression methanol before storage at −20◦ . Conventional bright field and
in premigratory and migratory NCCs (Wakamatsu et al., 2004). fluorescence microscopy was performed under a Leica MZ16FA
Despite evidence that points to an involvement for Sox2 in mul- stereoscope.
tiple steps of NCC development in mouse embryo, its role is yet
elusive. RNA IN SITU HYBRIDIZATION
Homozygous Sox2-null mouse embryos die around implan- RNA probes for in situ hybridization reactions were prepared
tation (Avilion et al., 2003b; Mandalos et al., 2012). We have by in vitro transcription as previously described (Knuchel et al.,
developed a conditional ablation strategy, using a “Conditional by 2000; Chotteau-Lelievre et al., 2006; Rhinn et al., 2011) The
Inversion” Sox2 allele (Sox2COIN ) (Mandalos et al., 2012), in order probes used were: Hoxa2 and Hoxb1, kindly provided by Robb
to study the role of Sox2 in epiblast-derived multipotent lineages. Krumlauf; Sox2, kindly provided by Robin Lovell-Badge; Sox10,
Here, we show that Sox2 plays an essential role in controlling the kindly provided by Benoît de Crombrugghe. Whole-mount in situ
behavior of the progenitor cells during head development. EMT hybridization (ISH) was performed using an Intavis InSituPro
is affected in mutant embryos during CNCC development, result- robot (detailed protocol available at http://empress.har.mrc.ac.
ing in hydrocephaly and frontonasal defects. These results suggest uk/, gene expression section).
a novel role for Sox2 as a rheostat of the EMTs that influence head
development in mice. IMAGE ANALYSIS
Embryo dissections, conventional bright field and fluorescence
MATERIALS AND METHODS microscopy were performed under a Leica MZ16FA stereo-
EXPERIMENTAL ANIMALS scope, equipped with a Leica 350 camera and Leica Software.
Generation of Sox2COIN mice was described elsewhere (Mandalos Sections were photographed under a Leica M420 macroscope
et al., 2012). All animals were handled in strict accordance and DMLB/DM4000B microscopes equipped with Photometrics
with good animal practice as defined by the Animals Act digital cameras and the CoolSnap imaging software (Roger
160/03.05.1991 applicable in Greece, revised according to the Scientific).
86/609/EEC/24.11.1986 EU directive regarding the proper care
and use of laboratory animals and in accordance to the Hellenic RESULTS
License for Animal Experimentation at the BSRC “Alexander EPIBLAST DELETION OF Sox2 RESULTS INTO LETHALITY AROUND E11
Fleming” (Prot. No. 767/28.02.07) issued after protocol approval The epiblast is destined to derive all multipotent lineages in
by the Animal Research Committee of the BSRC “Alexander the mouse embryo. Previously generated, null alleles of Sox2
Fleming” (Prot. No. 2762/03.08.05). (Sox2βgeo/βgeo and Sox2βgeo2/βgeo2 ) are responsible for an early

Frontiers in Physiology | Craniofacial Biology September 2014 | Volume 5 | Article 345 | 2

Mandalos et al. Head development is fine tuned by Sox2

embryonic lethal phenotype (Avilion et al., 2003a; Ekonomou We generated epiblast inverted Sox2EpINV/+ mice, making
et al., 2005), masking any subsequent role of Sox2 in the gen- use of the Tg(Sox2-CRE) mouse line that exerts efficient Cre-
eration of epiblast-derived multipotent lineages during develop- mediated recombination in the epiblast, but not in extraembry-
ment. Taking into account the challenges of the Sox2 locus, in that onic tissues (Hayashi et al., 2002a,b, 2003; Vincent and Robertson,
its proximal promoter and coding region are entirely contained 2003). Excision of the floxed sequences by Tg(Sox2-CRE) mice
within a CpG island, and are also spanned by an overlapping efficiently results in the visualization of eGFP in the epiblast at
transcript, Sox2Ot, which contains mmu-miR1897 (Amaral et al., E6.5 (Figures 1A–D).
2009; Shahryari et al., 2014), we developed a novel conditional As Sox2 is known to function as a cell fate determinant (Kondo
by inversion Sox2COIN allele (Mandalos et al., 2012). The inverted and Raff, 2004; Yamaguchi et al., 2011; Karnavas et al., 2013),
COIN Sox2 allele (Sox2INV ) is functionally null (Mandalos et al., we harvested embryos from various intercrosses at the onset
2012), as Sox2INV/INV mutants recapitulate the phenotype of of organogenesis (E11.5). Initially, we performed Sox2EpINV/+
Sox2βgeo/βgeo (Avilion et al., 2003a), Sox2βgeo2/βgeo2 (Ekonomou x Sox2+/+ intercrosses and found that heterozygous E11.5
et al., 2005) and Sox2EpINV/βgeo2 (Mandalos et al., 2012) embryos, Sox2EpINV/+ embryos do not show any obvious abnormalities and
which die around implantation. are indistinguishable from control Sox2COIN/COIN and Sox2+/+

FIGURE 1 | Sox2 inversion in the epiblast leads to embryonic lethality at Sox2 COIN/COIN and Sox2 EpINV /+ embryos are indistinguishable from Sox2+/+
E11.5. Generation of the epiblast-inverted Sox2EpINV allele. Sox2 EpINV /+ littermates (E–J). Sox2 EpINV /+ intercrosses generate normal (Sox2EpINV /+ )
embryos at E6.5 show normal morphology compared with Sox2+/+ control (K,L,O,P) and haploinsufficient (Sox2EpINV /+(H) ) embryos (M,N,Q,R). 50% of
embryos. eGFP expression indicates that Sox2 is highly expressed in the the haploinsufficient Sox2 EpINV /+(H) embryos have normal size (M,Q), whereas
epiblast, but not in the extraembryonic tissue (A–D). Sox2EpINV /+ embryos the remaining ones have a smaller size (N,R) at E11.5. Both types of
obtained from Sox2 EpINV /+ × Sox2 +/+ intercrosses are normal. At E11.5, Sox2 EpINV /+(H) mutant embryos show evident defects in the head region (M,N).

www.frontiersin.org September 2014 | Volume 5 | Article 345 | 3

Mandalos et al. Head development is fine tuned by Sox2

littermates (Figures 1E–G). E11.5 Sox2EpINV/+ embryos express these mutants, similarly to Sox2EpINV/+(H) mutants, die around
eGFP precisely in the areas where Sox2 is expressed (Avilion et al., E11.5. Sox2EpINV/mosaic embryos exhibited similar, albeit more
2003a; Mandalos et al., 2012) (Figures 1H–J). Sox2EpINV/+ adult severe, abnormalities compared to Sox2EpINV/+(H) embryos
mice are fertile, feed normally, have normal body weight and (Figures 2A–F).
normal lifespan, whilst Sox2INV/+ adult male exhibit no infertil- To examine whether a regional loss of Sox2 expression was
ity problems, as some Sox2βgeo/+ mice reportedly have (Avilion responsible for those defects, we analyzed Sox2 expression by RNA
et al., 2003b). It is tempting to suggest that infertility problems in situ hybridization (Figures 2D–F). Sox2 was absent through-
in Sox2βgeo2/+ and Sox2βgeo2/+ mice could arise from the removal out the spinal cord (sc) in E10.5 mutants, with an exception of
of regulatory regions due to the design of the mutants, while in the tail tip of both Sox2EpINV/+ and Sox2EpINV/mosaic mutants
Sox2EpINV/+ mice the whole sequence of the locus remains intact (Figures 2E,F). However, there were no obvious morphological
after inversion (Mandalos et al., 2012). defects in the SC at E10.5, suggesting that down-regulation of
Sox2 could be rescued due to the functional redundancy of Sox2
EPIBLAST DELETION OF Sox2 RESULTS INTO HYDROCEPHALY AND with other SoxB genes, namely Sox1 and Sox3 (Uwanogho et al.,
CRANIOFACIAL DEFECTS 1995; Rex et al., 1997; Wood and Episkopou, 1999; Archer et al.,
We then proceeded to Sox2EpINV/+ intercrosses and har- 2011; Elkouris et al., 2011). We observed a down-regulation of
vested the embryos again at E11.5 (Table 1). Normally, if Sox2 in the frontonasal process (fnp), the forebrain region (fb),
all Sox2EpINV/+ and Sox2EpINV/EpINV embryos would survive, the midbrain (mb) and the hindbrain (hb), the eye (Hever et al.,
one would expect that 75% of the embryos would be eGFP- 2006) and the otocyst (ot) (Kiernan et al., 2005; Hume et al., 2007;
positive (eGFP+ ). However, we observed that only 48.8% of Pan et al., 2013) of Sox2EpINV/+(H) and Sox2EpINV/mosaic mutant
the embryos harvested were eGFP+ , suggesting that as previ- embryos. Morphologically, there was a distortion of the eye, an
ously observed (Mandalos et al., 2012), the Sox2EpINV/EpINV die enlargement of the brain ventricles, and marked translucency in
in the deciduas at an early stage (Table 1). Half of the har- the hindbrain region.
vested eGFP+ heterozygote embryos had normal phenotypes Several studies have shown that regulation of EMT during
Sox2EpINV/+ (Figures 1L,P), while the remaining ones repre- NCC development plays a crucial role for the normal develop-
sented haploinsufficient Sox2EpINV/+ (Sox2EpINV/+(H) ) mutants ment of the frontonasal region, including the palate and nasal
(Figures 1Q,R). The Sox2EpINV/+(H) observed embryonic phe- cavities (Kang and Svoboda, 2005). We observed that Sox2 protein
notypes fall in two categories: (a) Sox2INV/+(H) embryos with a marks specific brain and craniofacial regions during their devel-
similar size to Sox2+/+ and Sox2INV/+ embryos (Figures 1M,Q) opment (Figure 3). In E11.5 embryos, expression is seen in the
and (b) Sox2INV/+(H) embryos with a significantly reduced hindbrain and forebrain neuroepithelium, and in the oral epithe-
size (Figures 1N,R) compared to Sox2+/+ (Figures 1K,O) and lium (Figures 3A–D). At E15.5, Sox2 is expressed in the dermis
Sox2EpINV/+ littermates (Figures 1L,P). Sox2EpINV/+(H) embryos surrounding developing hair follicles and whiskers, including in
exhibit heart defects, hemorrhage, bulging fourth ventricu- their dermal papilla (Figures 3H–J). Sox2 is also detected in the
lar roof, and severe craniofacial defects (Figure 1M). We har- epithelium of the retina (Figure 3K), in developing bone and
vested litters at E12.5 and E15.5 derived from Sox2EpINV/+ cartilage (Figure 3L), in muscle fibers (Figures 3M,N) and the
intercrosses and found only Sox2+/+ and Sox2EpINV/+ normal acinar structures of the salivary glands (Figure 3O). Furthermore,
embryos, suggesting that Sox2EpINV/+(H) embryos die at around Sox2 asymmetrically marks the epithelium that connects the
E11. Phenotypic differences among heterozygote Sox2EpINV/+ developing molar teeth with the oral epithelium (Figure 3P).
and Sox2EpINV/+(H) littermates, derived from Sox2EpINV/+ inter- At later post-natal stages Sox2 is expressed in the alveolar bone
crosses confirm that there is a Sox2 expression threshold, below (ab) (Figure 3T) and in the incisor labial cervical loop (cl,
which phenotypic abnormalities appear during embryogene- Figures 3R,S). In the hair follicles the expression is confined to
sis. To generate conditional epiblast-inverted Sox2 mutants, some cells of the inner and outer sheath (Figures 3Q,U).
we performed Sox2COIN/COIN to Sox2EpINV/+ ; Tg(Sox2CRE) To further analyze the craniofacial defects observed in
intercrosses and named these conditional mutant embryos Sox2EpINV/+(H) embryos, we harvested E11.5 Sox2EpINV/+(H)
Sox2EpINV/mosaic . We could harvest Sox2EpINV/mosaic embryos mutants and sectioned them for hematoxylin and eosin
from E8.5 to E11.5, but not beyond E11.5, indicating that histological analysis (Figures 4A–D). Compared to Sox2+/+

Table 1 | Analysis of progeny from Sox2 EpINV /+ × Sox2 EpINV /+ intercrosses# .

Genotypic distribution obtained at E11.5*

Total Dead Live Sox2 EpINV /+ Sox2 EpINV/+(H) Sox2 EpINV/EpINV Sox2 +/+

43 11 (25.6%) 32 (74.4%) 11 (25.6%) 10 (23.2)** 0 (0%) 11 (25.6%)

# Data
collected from mice in C57BL6 background.
*Genotypes were assessed by PCR either from tail biopsies or from embryonic yolk sac or whole embryos.
**50% of Sox2INV /+(H) (11.6% of the total number of embryos) from Sox2INV /+ × Sox2INV /+ intercrosses have a smaller size (Figures 1N,R).

Frontiers in Physiology | Craniofacial Biology September 2014 | Volume 5 | Article 345 | 4

Mandalos et al. Head development is fine tuned by Sox2

FIGURE 2 | Sox2 loss leads to multiple developmental defects. eGFP (ba1, ba2), and the spinal cord (sc) but not the tail tip, of E10.5 Sox2EpINV /+(H)
expression recapitulates the expression of Sox2 in Sox2EpINV /mosaic mutant and Sox2 EpINV /mosaic embryos. These embryos show increased translucency
(A–C). Whole mount in situ hybridization shows down-regulation of Sox2 in mostly visible at the level of the hindbrain (hb); forebrain (fb) and midbrain
the hindbrain (hb), midbrain (mb) and forebrain (fb) regions, in the frontonasal (mb). Ventricles are also enlarged and frontonasal truncations are evident
process (fnp), the eye (ey), the surface ectoderm of branchial arches 1 and 2 (D–F).

normal embryos, we observed a reduction of the thickness of the Sox2 FINE-TUNES THE FLOW OF MIGRATING Sox10 + NCCs
neuroepithelial wall lining the telencephalic (fb), mesencephalic To investigate the fate of NCCs that may under to craniofa-
(mb) and rhombencephalic (hb) ventricles, defective frontonasal cial abnormalities observed in Sox2 mutants, we analyzed Sox10
proccess and oral cavity formation, and dilated lumen of the optic expression in Sox2+/+ and Sox2EpINV/+(H) embryos at E9.5, at
stalk (Figures 4A–D). Thus, Sox2 loss leads to severe brain and an embryonic stage in which Sox10+ NCCs are migrating along
craniofacial defects. the lateral surface of the neural tube in wild-type embryos. We
found that Sox10+ cells expressing high levels of Sox10 are het-
SOX2 LOSS LEADS TO DOWN-REGULATION OF Hoxa2 IN erotopically present in the frontonasal region and in the branchial
RHOMBOMERE 3, BUT NOT Hoxb1 IN RHOMBOMERE 4 arches (ba) 1-2 of Sox2EpINV/+(H) mutants (Figure 5A). Thus,
In order to find out whether Sox2 loss of function disrupts the down-regulation of Sox2 causes a dramatic up-regulation of
Hox code in hindbrain, we analyzed the expression of Hoxa2 Sox10 expression and an outflow of Sox10+ cells in the hind-
and Hoxb1 at E8.5 in Sox2+/+ and Sox2EpINV/mosaic embryos, brain (hb) and branchial areas of mutant embryos. Furthermore,
using whole mount in situ hybridization (Figures 5A,B). We we observed Sox10+ cells in Sox2EpINV/+(H) mutant embryos in
observed that Hoxa2 is down-regulated in rhombomere (r)3, but frontonasal areas, where Sox10 is normally not expressed, while
not in r5 of Sox2EpINV/mosaic mutant embryos, underscoring a migrating CNCC do not express Sox10 at this embryonic stage
specific role for Sox2 in the regulation of Hoxa2 in r3 at E8.5 in Sox2+/+ control embryos (Figures 6A–D). Thus, Sox2 loss
(Figures 5A,B). To find out whether Sox2 loss of function affects disrupts severely the CNCC development.
the facial innervation programme, we analyzed the expression
of Hoxb1 in Sox2+/+ and Sox2EpINV/mosaic embryos, and found DISCUSSION
that Hoxb1 expression was unaffected both in r4 (Figures 5A,B) EMT plays a crucial role in the development of the embryonic
and in the spinal cord (data not shown) of E8.5 Sox2EpINV/mosaic head (Mitsiadis, 2011). CNCCs undergo EMT and individual cells
embryos. Thus, our results do not support previous reports on delaminate from the lateral ridges of the dorsal neural tube and
Sox2 involvement in the regulation of Hoxb1 in vitro, at least with migrate to the craniofacial area (Kouskoura et al., 2011) to form
regard to E8.5 hindbrain and spinal cord regions. As predicted the frontonasal, maxillary and mandibular processes (Bronner-
from these observations, when cranial nerves were stained for Fraser, 2002). Amongst genes of the SoxB1 group (Sox1-3), which
Sox10 expression at E11.5 in mutant embryos (Figures 5C,D), we are predominantly expressed in the developing central nervous
found that Sox2 loss of function does not affect the formation of system (CNS) (Collignon et al., 1996; Wood and Episkopou,
Sox10+ ganglia of the spinal accessory, vagus (n10), glossopha- 1999), Sox3 activity is required for pharyngeal segmentation and
ryngeal (n9), and branches of facial (n7) nerves (Figures 5C,D). for the pharyngeal epithelium to proceed toward craniofacial

www.frontiersin.org September 2014 | Volume 5 | Article 345 | 5

Mandalos et al. Head development is fine tuned by Sox2

FIGURE 3 | Immunohistochemical analysis of Sox2 expression in (oe) along the lingual side (P). At postnatal day 1 (PN1), Sox2 expression is
developing craniofacial structures. At E11.5 Sox2 is detected in hindbrain limited to some cells of the inner and outer root sheath of the whisker
(hb) and forebrain (fb), as well as in the oral epithelium (oe) (A–D). At epithelium (we, the outer root sheath being delimited by red dots, Q)
E13.5, expression appears in the nasal epithelium (ne) (E–G). At E15.5 (Q–U). Sox2 is furthermore expressed in the labial cervical loop (cl) of the
Sox2 is expressed in the dermis surrounding the developing whiskers (w), incisor (inc) (R,S), in the alveolar bone (ab) (T,S) and the vasculature (U).
and in their dermal papilla (wp) (H–P). Expression is also detected in the de, dental epithelium; dm, dental mesenchyme; dp, dental papilla; fnp,
retina (r, K), in developing bone (b) and cartilage (c, L), in developing frontonasal process; hl, hindlimb; m, mesenchyme; md, mandible; mx,
muscle fibers (mf, M,N), and in salivary glands (sg, O). Noteworthy, the maxilla; n, nose; np, nasal process; e, epithelium; g, glomerulae; k,
expression of Sox2 in the developing molar teeth is asymmetrical, being keratinized part of the whisker; v, vessel; sm, smooth muscle; sg, salivary
restricted to the connection between the tooth and the oral epithelium gland; c, cartilage; r, retina; f, follicle; mf, muscular fibers.

morphogenesis (Rizzoti and Lovell-Badge, 2007). On the other generation of NCC progeny, as observed in differentiation exper-
hand, Sox2 activity has been implicated in processes that coun- iments of human ES cells (ESC)-derived NCCs into sensory
teract NCC development (Hutton and Pevny, 2011; Remboutsika neurons in vitro (Cimadamore et al., 2012). Thus, the severity
et al., 2011; Cimadamore et al., 2012) and could affect the of defects observed in the developing brain and facial structures

Frontiers in Physiology | Craniofacial Biology September 2014 | Volume 5 | Article 345 | 6

Mandalos et al. Head development is fine tuned by Sox2

FIGURE 4 | Sox2 INV /+(H) E11.5 embryos exhibit brain and craniofacial
malformations. Histological analysis of Sox2+/+ and Sox2 EpINV /+(H)
embryos head structures (A–D). Serial sagittal sections of E11.5 embryos
stained with hematoxylin and eosin. Sox2EpINV /+(H) embryos exhibit
thinness of the ventricular wall and abnormal oral cavity formation enlarged
forebrain and midbrain ventricles, while the frontonasal process (fnp) is FIGURE 5 | Hoxa2 and Hoxb1 expression in rhombomeres, and Sox10
severely reduced with an abnormal oral cavity (B,D) when compared expression in developing cranial nerves and ganglia. In situ
Sox2 +/+ littermates (A,C). hybridization for Hoxa2 and Hoxb1 at E8.5 reveals that Hoxa2 is
down-regulated in rhombomere 3 of a Sox2INV /mosaic embryo, but not in
rhombomere 5 (A,B). Hoxb1 is normally expressed in r4 in Sox2INV /mosaic
mutants (A,B). Sox10 + nerves form normally in Sox2 EpINV /+(H) embryos.
of Sox2EpINV/+(H) and Sox2EpINV/mosaic mutants underline a Sox2 In situ hybridization reveals that Sox10 expression is not affected in
branchial arches 1 and 2 (ba1, ba2) in developing cranial nerves and ganglia
dosage-dependent role in the development of both the head and
(n10, n9, n5mx, n5md, g7-8, g5, n5o) of Sox2EpINV /mosaic mutants
craniofacial areas. compared with Sox2 +/+ control embryos (C,D). ey, eye.
Sox2EpINV/+(H) and Sox2EpINV/mosaic embryos suffer from
ventriculomegaly, which in turn leads to the accumulation of
increased amounts of cerebrospinal fluid (CSF) in the brain. that could contribute to early pathoanatomical events resulting
Hydrocephaly involves both dilated ventricular system and in hydrocephalus and craniofacial defects in humans (Panetta
increased intracranial pressure. Every reduction in the thick- et al., 2008). Thus, disruption of Sox2 function in the embryonic
ness of the ventricular wall invevitably will result in dilation of head region could be an additional cause for the development of
the ventricular system (hydrocephalus ex vacuo). It is difficult hydrocephalus later on in life.
to establish cause and effect relationship in this reciprocal set- Hox genes play an essential role in the development of cranio-
ting, however abnormal brain wall development will definitely facial structures (Trainor and Krumlauf, 2000; Narita and Rijli,
lead to morphologically enlarged ventricular system. The indi- 2009; Tumpel et al., 2009; Di Bonito et al., 2013). At E8.5,
cation that the ventricular system is enlarged could not in itself Hoxb1 is expressed in r4 and throughout the spinal cord region
be considered proof of hydrocephaly, because there is thickness (Gavalas et al., 2003), where it is required for the specification of
of the wall, thus pointing to ventriculomegaly only (Deveale facial branchiomotor neuron progenitors that are programmed
et al., 2013). Likewise, reduction in the diameter of the aque- to innervate the facial muscles (Arenkiel et al., 2004). Despite the
duct and abnormal periaqueductal region development does not fact that Sox2 has been shown to regulate Hoxb1 in vitro (Di Rocco
prove etiological relationship toward hydrocephalus causation et al., 2001; Williams et al., 2004; Lian et al., 2010), the expression
(Lee et al., 2012). In adult animals, hydrocephaly involves also of Hoxb1 appeared to be unaffected both in r4 and in the spinal
defects in choroid plexus, but as this is in very early stages in cord of E8.5 Sox2INV/mosaic embryos (Figure 4). Our results do
development at the embryonic period we examined, the contri- not support previous reports on Sox2 involvement in the regu-
bution of CSF overproduction in hydrocephalus causation is hard lation of Hoxb1 in vitro, at least with regard to E8.5 hindbrain
to assess (Mizusawa, 1997). Sox2EpINV/+(H) and Sox2EpINV/mosaic and spinal cord regions. On the other hand, Hoxa2 appears to
mutants display developmental defects both in ventricular sys- be down-regulated in our mutants. At E8.5, Hoxa2 has a limit of
tem/wall formation, as well as in the oral cavity morphology expression in the rhombencephalic neural tube corresponding to

www.frontiersin.org September 2014 | Volume 5 | Article 345 | 7

Mandalos et al. Head development is fine tuned by Sox2

FIGURE 6 | Sox2 regulates the flow of Sox10 + NCC. NCC formation and and Sox2 +/+ embryos. Enlargement of the fourth ventricle and the
migration is exacerbated in E9.5 Sox2INV /+(H) mouse embryos, as observed mandibular component of the ba1 is observed. Exacerbated numbers of
by the Sox10 in situ hybridization pattern in branchial arches 1 and 2 (ba1, Sox10+ cells are observed in cranial and trunk regions, in the frontonasal
ba2), and in the frontonasal (fnp) area of Sox2INV /+(H) embryos when region and throughout ba1 (C,D). fb, forebrain; fnp, frontonasal process; mb,
compared to Sox2 +/+ embryos (A,B). Coronal sections of E9.5 Sox2INV /+(H) midbrain; sc, spinal cord; h, heart.

of Hoxa2 expression in Sox2EpINV mutants indicates that Sox2

controls an integral component of NCC morphogenetic program,
which requires Hoxa2 at discrete time points to pattern distinct
derivatives in craniofacial structures (Santagati et al., 2005).
As neural progenitor cells differentiate into NCCs, a switch in
expression from SoxB to SoxE genes becomes evident, with Sox2
inactivated in the NCC progenitors, whereas Sox9 and Sox10 are
activated in newly migrating trunk NCCs (Melton et al., 2004;
Remboutsika et al., 2011). This is a necessary switch for the activa-
tion of the complex mechanism that generates NCCs (Wakamatsu
et al., 2004). Amongst the SoxE genes, Sox10 is required for the
formation, maintenance of multipotency, specification and dif-
ferentiation of NCCs (Kelsh, 2006). Sox10 is the only SoxE gene
that maintains its expression during migration of NCCs along
the lateral surface of the neural tube (McKeown et al., 2005),
except in the cranial region. Sox10 mutations lead to several
craniofacial abnormalities in humans, called neurocristopathies,
FIGURE 7 | Sox2: a rheostat of EMT transition during neural crest including Waardenburg-Hirschsprung syndrome and peripheral
development. Precise timing ensured by an extremely accurate neuropathies (Hoke, 2012). Sox2 over-expression and Sox2+ neu-
developmental clock regulates the dynamics of the decisions to generate ral stem cell transplantation experiments in avian and murine
NCC from neural progenitors. Sox2 controls the flow of the EMT transition,
cranial neural tubes have demonstrated that Sox2 restricts neu-
leading to NCC migration in appropriate numbers at the appropriate regions
in the head. This way, sequence (from neural progenitor to NCC) and
roepithelial differentiation into CNCCs (Cheung and Briscoe,
genetic heterochrony (spatially and temporally controlled Sox10 2003; Remboutsika et al., 2011; Wahlbuhl et al., 2012). Thus, the
expression), resulting into craniofacial malformations could be averted. exacerbation of Sox10+ migrating cells in the Sox2EpINV mutants
may not be surprising. These observations point out that Sox2
could act to repress Sox10 expression. However, any genetic inter-
r3 and r5 (Prince and Lumsden, 1994). It is not surprising that no action between Sox2 and Sox10 in neural progenitor or NCC
effect was observed in Hoxa2 expression in r5, as Sox2 is expressed progenitor cells is far from evident in vivo and in vitro. Whether
along the hindbrain in all rhombomeres, but not in r5 (Wood and Sox2 could influence the expression of Sox10 directly or indi-
Episkopou, 1999). Hoxa2-null mutant embryos lack craniofacial rectly by affecting levels of other SoxE genes such as Sox8 and
and cartilage elements derived from the first and second branchial Sox9 that contribute to the induction of Sox10 in NCC progen-
arch and die perinatally due to cleft palate (Vieille-Grosjean et al., itors, once NC-inducing signals are set (Taylor and Labonne,
1997; Rijli et al., 1998; Trainor and Krumlauf, 2001; Santagati 2005; McCauley and Bronner-Fraser, 2006; Haldin and Labonne,
et al., 2005). Sox2 has been shown to interact in vitro with 2010; Stolt and Wegner, 2010; Wahlbuhl et al., 2012), remains
a SoxB DNA binding element (ACAAT motif) present in the to be investigated. In the Sox2EpINV mutants ,Sox10 levels appear
enhancer of the Hoxa2 gene and mutation of this motif reduces dramatically increased both in the branchial arches area and in
the expression of a Hoxa2 reporter in electroporation experiments the frontonasal area. Sox10 over-expression has been shown to
in chick embryo hindbrains (Tumpel et al., 2008). The reduction arrest the neuroepithelial and cranial mesenchymal cells in an

Frontiers in Physiology | Craniofacial Biology September 2014 | Volume 5 | Article 345 | 8

Mandalos et al. Head development is fine tuned by Sox2

undifferentiated state, causing a range of cell fate specification Akitaya, T., and Bronner-Fraser, M. (1992). Expression of cell adhesion molecules
defects (Ahlstrom and Erickson, 2009). Neural progenitor cells, during initiation and cessation of neural crest cell migration. Dev. Dyn. 194,
12–20. doi: 10.1002/aja.1001940103
which over-express Sox10 remain undifferentiated and fail to
Amaral, P. P., Neyt, C., Wilkins, S. J., Askarian-Amiri, M. E., Sunkin, S. M.,
form neuronal, Schwann, or melanocyte cells (Stolt et al., 2008). Perkins, A. C., et al. (2009). Complex architecture and regulated expression
Thus, it is tempting to suggest that the failure of the embryos to of the Sox2ot locus during vertebrate development. RNA 15, 2013–2027. doi:
form the craniofacial region could, in part, be due to the failure 10.1261/rna.1705309
of the cranial mesenchyme to proceed through development due Aquino, J. B., Hjerling-Leffler, J., Koltzenburg, M., Edlund, T., Villar, M. J., and
to an aberrant and exacerbated population of Sox10+ cells in the Ernfors, P. (2006). In vitro and in vivo differentiation of boundary cap neu-
ral crest stem cells into mature Schwann cells. Exp. Neurol. 198, 438–449. doi:
frontonasal region. 10.1016/j.expneurol.2005.12.015
In recent years, the importance of NCCs as inducers of periph- Aramaki, M., Kimura, T., Udaka, T., Kosaki, R., Mitsuhashi, T., Okada, Y., et al.
eral neural structures, craniofacial tissues and other peripheral (2007). Embryonic expression profile of chicken CHD7, the ortholog of the
mesodermal-derived structures has become evident (Trainor and causative gene for CHARGE syndrome. Birth Defects Res. A Clin. Mol. Teratol.
79, 50–57. doi: 10.1002/bdra.20330
Tam, 1995; Trainor et al., 2003; Hong and Saint-Jeannet, 2005; Archer, T. C., Jin, J., and Casey, E. S. (2011). Interaction of Sox1, Sox2,
Cordero et al., 2011; Hagiwara et al., 2014). Defects in their Sox3 and Oct4 during primary neurogenesis. Dev. Biol. 350, 429–440. doi:
development has been attributed to a failure and/or abnormal 10.1016/j.ydbio.2010.12.013
NCCs migration and differentiation (Bronner, 2012), resulting Arenkiel, B. R., Tvrdik, P., Gaufo, G. O., and Capecchi, M. R. (2004). Hoxb1
into the generation of neurocristopathies in humans (Etchevers functions in both motoneurons and in tissues of the periphery to establish
and maintain the proper neuronal circuitry. Genes Dev. 18, 1539–1552. doi:
et al., 2006) and expanding the most recent classification of 10.1101/gad.1207204
neurocristopathies to an entire category of abnormal induction Avilion, A. A., Nicolis, S. K., Pevny, L. H., Perez, L., Vivian, N., and Lovell-Badge, R.
of non-neural NCC-derived peripheral structures of the body (2003a). Multipotent cell lineages in early mouse development depend on SOX2
(Cossais et al., 2010). Recent evidence has shown that Sox2 has function. Genes Dev. 17, 126–140. doi: 10.1101/gad.224503
Avilion, A. A., Nicolis, S. K., Pevny, L. H., Perez, L., Vivian, N., and Lovell-Badge,
been indirectly associated with defects that are characteristic of
R. (2003b). Multipotent cell lineages in early mouse development depend on
the CHARGE syndrome, a human neurocristopathy (Aramaki SOX2 function. Genes Dev. 17, 126–140. doi: 10.1101/gad.224503
et al., 2007). CHARGE syndrome patients exhibited mutations in Aybar, M. J., Glavic, A., and Mayor, R. (2002). Extracellular signals, cell interactions
the Chd7 gene (Vallaster et al., 2012), the product of which acts and transcription factors involved in the induction of the neural crest cells. Biol.
as a Sox2 transcriptional cofactor (Engelen et al., 2011; Puc and Res. 35, 267–275. doi: 10.4067/S0716-97602002000200018
Aybar, M. J., and Mayor, R. (2002). Early induction of neural crest cells: lessons
Rosenfeld, 2011). Our results suggest a Sox2 dosage-dependent
learned from frog, fish and chick. Curr. Opin. Genet. Dev. 12, 452–458. doi:
mechanism acting during head development, with a specific role 10.1016/S0959-437X(02)00325-8
for Sox2 in the prevention of these CNCC-related pathologies. We Ayer-Le Lievre, C. S., and Le Douarin, N. M. (1982). The early development
propose that Sox2 acts as a rheostat of EMT during CNCC devel- of cranial sensory ganglia and the potentialities of their component cells
opment that influences cell fates involved in head development studied in quail-chick chimeras. Dev. Biol. 94, 291–310. doi: 10.1016/0012-
1606(82)90349-9
(Figure 7). These findings open novel avenues to target Sox2 in a Bernard, P., and Harley, V. R. (2010). Acquisition of SOX transcription factor
number of craniofacial malformations in humans. specificity through protein-protein interaction, modulation of Wnt signalling
and post-translational modification. Int. J. Biochem. Cell Biol. 42, 400–410. doi:
ACKNOWLEDGMENTS 10.1016/j.biocel.2009.10.017
The authors thank Brigitte Schuhbaur for excellent technical Betancur, P., Sauka-Spengler, T., and Bronner, M. (2011). A Sox10 enhancer ele-
assistance and Maria-Angeliki Gavala for the design of the ment common to the otic placode and neural crest is activated by tissue-specific
paralogs. Development 138, 3689–3698. doi: 10.1242/dev.057836
model. Eumorphia Remboutsika and Nikolaos Mandalos were
Bowles, J., Schepers, G., and Koopman, P. (2000). Phylogeny of the SOX fam-
supported by the “Bilateral cooperation grant” (NON-EU82) ily of developmental transcription factors based on sequence and structural
funded by National grants from the Ministry of Development, co- indicators. Dev. Biol. 227, 239–255. doi: 10.1006/dbio.2000.9883
financed by the European Union and Greek operational program Briscoe, C. P., Tadayyon, M., Andrews, J. L., Benson, W. G., Chambers, J. K., Eilert,
“Competitiveness and Enterpreneurship” European Regional M. M., et al. (2003). The orphan G protein-coupled receptor GPR40 is activated
by medium and long chain fatty acids. J. Biol. Chem. 278, 11303–11311. doi:
Development Fund (NSFR 2007-2013), the “Synergasia grant” 10.1074/jbc.M211495200
(09-SYN66-12), the ADiSC - Thalis Grant funded by Framework Bronner, M. E. (2012). Formation and migration of neural crest cells in the
Program “Education and Lifelong Learning,” co-financed by the vertebrate embryo. Histochem. Cell Biol. 138, 179–186. doi: 10.1007/s00418-
European Commission (European Social Fund) and National 012-0999-z
funds from the Ministry of Education. This work was performed Bronner-Fraser, M. (2002). Molecular analysis of neural crest formation. J. Physiol.
Paris 96, 3–8. doi: 10.1016/S0928-4257(01)00074-2
under the auspices of a research collaboration agreement with Cardiff, R. D., Miller, C. H., and Munn, R. J. (2014). Manual hematoxylin and eosin
REGENERON Pharmaceuticals Inc., New York, USA. Thimios staining of mouse tissue sections. Cold Spring Harb. Protoc. 2014, 655–658. doi:
Mitsiadis and Zoraide Granchi were supported by a SNSF grant 10.1101/pdb.prot073411
(09-SYN66-12) and from UZH. Muriel Rhinn and Pascal Dollé Cheng, Y., Cheung, M., Abu-Elmagd, M. M., Orme, A., and Scotting, P. J. (2000).
Chick sox10, a transcription factor expressed in both early neural crest cells
were supported by an Agence Nationale de la Recherche grant
and central nervous system. Brain Res. Dev. Brain Res. 121, 233–241. doi:
(ReSiNeS-SVSE2-2011). 10.1016/S0165-3806(00)00049-3
Cheung, M., and Briscoe, J. (2003). Neural crest development is regulated by
REFERENCES the transcription factor Sox9. Development 130, 5681–5693. doi: 10.1242/dev.
Ahlstrom, J. D., and Erickson, C. A. (2009). New views on the neural crest 00808
epithelial-mesenchymal transition and neuroepithelial interkinetic nuclear Chotteau-Lelievre, A., Dolle, P., and Gofflot, F. (2006). Expression analysis of
migration. Commun. Integr. Biol. 2, 489–493. doi: 10.4161/cib.2.6.9406 murine genes using in situ hybridization with radioactive and nonradioactively

www.frontiersin.org September 2014 | Volume 5 | Article 345 | 9

Mandalos et al. Head development is fine tuned by Sox2

labeled RNA probes. Methods Mol. Biol. 326, 61–87. doi: 10.1385/1-59745- regulatory interactions among these Hox genes. Development 130, 5663–5679.
007-3:61 doi: 10.1242/dev.00802
Christiansen, A. E., Keisman, E. L., Ahmad, S. M., and Baker, B. S. (2002). Sex Graham, V., Khudyakov, J., Ellis, P., and Pevny, L. (2003). SOX2 functions to
comes in from the cold: the integration of sex and pattern. Trends Genet. 18, maintain neural progenitor identity. Neuron 39, 749–765. doi: 10.1016/S0896-
510–516. doi: 10.1016/S0168-9525(02)02769-5 6273(03)00497-5
Cimadamore, F., Shah, M., Amador-Arjona, A., Peran, E. M., Chen, C., and Hagiwara, K., Obayashi, T., Sakayori, N., Yamanishi, E., Hayashi, R., Osumi, N.,
Terskikh, A. V. (2012). SOX2 Modulates levels of MITF in human embryonic et al. (2014). Molecular and cellular features of murine craniofacial and trunk
stem cell - derived neural crest, normal human melanocytes, and melanoma neural crest cells as stem cell-like cells. PLoS ONE 9:e84072. doi: 10.1371/jour-
lines in vitro. Pigment Cell Melanoma Res. 25, 533–536. doi: 10.1111/j.1755- nal.pone.0084072
148X.2012.01012.x Haldin, C. E., and Labonne, C. (2010). SoxE factors as multifunctional neu-
Collignon, J., Sockanathan, S., Hacker, A., Cohen-Tannoudji, M., Norris, D., ral crest regulatory factors. Int. J. Biochem. Cell Biol. 42, 441–444. doi:
Rastan, S., et al. (1996). A comparison of the properties of Sox-3 with Sry and 10.1016/j.biocel.2009.11.014
two related genes, Sox-1 and Sox-2. Development 122, 509–520. Hayashi, S., Lewis, P., Pevny, L., and McMahon, A. P. (2002a). Efficient gene mod-
Cordero, D. R., Brugmann, S., Chu, Y., Bajpai, R., Jame, M., and Helms, J. A. (2011). ulation in mouse epiblast using a Sox2Cre transgenic mouse strain. Gene Expr.
Cranial neural crest cells on the move: their roles in craniofacial development. Patterns 2, 93–97. doi: 10.1016/S0925-4773(02)00292-7
Am. J. Med. Genet. A 155A, 270–279. doi: 10.1002/ajmg.a.33702 Hayashi, S., Lewis, P., Pevny, L., and McMahon, A. P. (2002b). Efficient gene mod-
Cossais, F., Wahlbuhl, M., Kriesch, J., and Wegner, M. (2010). SOX10 structure- ulation in mouse epiblast using a Sox2Cre transgenic mouse strain. Mech. Dev.
function analysis in the chicken neural tube reveals important insights into 119(Suppl. 1), S97–S101. doi: 10.1016/S0925-4773(03)00099-6
its role in human neurocristopathies. Hum. Mol. Genet. 19, 2409–2420. doi: Hayashi, S., Tenzen, T., and McMahon, A. P. (2003). Maternal inheritance of Cre
10.1093/hmg/ddq124 activity in a Sox2Cre deleter strain. Genesis 37, 51–53. doi: 10.1002/gene.10225
Del Barrio, M. G., and Nieto, M. A. (2002). Overexpression of Snail family members Heeg-Truesdell, E., and Labonne, C. (2004). A slug, a fox, a pair of sox: transcrip-
highlights their ability to promote chick neural crest formation. Development tional responses to neural crest inducing signals. Birth Defects Res. C Embryo
129, 1583–1593. Today 72, 124–139. doi: 10.1002/bdrc.20011
Deveale, B., Brokhman, I., Mohseni, P., Babak, T., Yoon, C., Lin, A., et al. (2013). Hever, A. M., Williamson, K. A., and Van Heyningen, V. (2006). Developmental
Oct4 is required ∼E7.5 for proliferation in the primitive streak. PLoS Genet. malformations of the eye: the role of PAX6, SOX2 and OTX2. Clin. Genet. 69,
9:e1003957. doi: 10.1371/journal.pgen.1003957 459–470. doi: 10.1111/j.1399-0004.2006.00619.x
Di Bonito, M., Glover, J. C., and Studer, M. (2013). Hox genes and region-specific Hoke, A. (2012). Animal models of peripheral neuropathies. Neurotherapeutics 9,
sensorimotor circuit formation in the hindbrain and spinal cord. Dev. Dyn. 242, 262–269. doi: 10.1007/s13311-012-0116-y
1348–1368. doi: 10.1002/dvdy.24055 Hong, C. S., and Saint-Jeannet, J. P. (2005). Sox proteins and neural crest develop-
Dickinson, M. E., Selleck, M. A., McMahon, A. P., and Bronner-Fraser, M. (1995). ment. Semin Cell Dev. Biol. 16, 694–703. doi: 10.1016/j.semcdb.2005.06.005
Dorsalization of the neural tube by the non-neural ectoderm. Development 121, Huang, X., and Saint-Jeannet, J. P. (2004). Induction of the neural crest
2099–2106. and the opportunities of life on the edge. Dev. Biol. 275, 1–11. doi:
Di Rocco, G., Gavalas, A., Popperl, H., Krumlauf, R., Mavilio, F., and Zappavigna, V. 10.1016/j.ydbio.2004.07.033
(2001). The recruitment of SOX/OCT complexes and the differential activity of Hume, C. R., Bratt, D. L., and Oesterle, E. C. (2007). Expression of LHX3 and
HOXA1 and HOXB1 modulate the Hoxb1 auto-regulatory enhancer function. SOX2 during mouse inner ear development. Gene Expr. Patterns 7, 798–807.
J. Biol. Chem. 276, 20506–20515. doi: 10.1074/jbc.M011175200 doi: 10.1016/j.modgep.2007.05.002
Dottori, M., Gross, M. K., Labosky, P., and Goulding, M. (2001). The winged-helix Hutton, S. R., and Pevny, L. H. (2011). SOX2 expression levels distinguish between
transcription factor Foxd3 suppresses interneuron differentiation and promotes neural progenitor populations of the developing dorsal telencephalon. Dev. Biol.
neural crest cell fate. Development 128, 4127–4138. 352, 40–47. doi: 10.1016/j.ydbio.2011.01.015
Ekonomou, A., Kazanis, I., Malas, S., Wood, H., Alifragis, P., Denaxa, M., Jessel, A. S., and Weiss, P. L. (1998). Orthodox practices clarified. Am. J. Occup.
et al. (2005). Neuronal migration and ventral subtype identity in the telen- Ther. 52, 71–72. doi: 10.5014/ajot.52.1.71d
cephalon depend on SOX1. PLoS Biol. 3:e186. doi: 10.1371/journal.pbio. Kamachi, M., Bruce, V., Mukaida, S., Gyoba, J., Yoshikawa, S., and Akamatsu,
0030186 S. (2013). Dynamic properties influence the perception of facial expressions.
Elkouris, M., Balaskas, N., Poulou, M., Politis, P. K., Panayiotou, E., Malas, S., et al. Perception 42, 1266–1278. doi: 10.1068/p3131n
(2011). Sox1 maintains the undifferentiated state of cortical neural progenitor Kang, P., and Svoboda, K. K. (2005). Epithelial-mesenchymal transforma-
cells via the suppression of Prox1-mediated cell cycle exit and neurogenesis. tion during craniofacial development. J. Dent. Res. 84, 678–690. doi:
Stem Cells 29, 89–98. doi: 10.1002/stem.554 10.1177/154405910508400801
Endo, T., Nakagawa, T., Lee, J. E., Dong, Y., Kim, T. S., Iguchi, F., et al. (2002). Karnavas, T., Mandalos, N., Malas, S., and Remboutsika, E. (2013). SoxB, cell cycle
Alteration in expression of p27 in auditory epithelia and neurons of mice dur- and neurogenesis. Front. Physiol. 4:298. doi: 10.3389/fphys.2013.00298
ing degeneration. Neurosci. Lett. 334, 173–176. doi: 10.1016/S0304-3940(02) Kelsh, R. N. (2006). Sorting out Sox10 functions in neural crest development.
01115-1 Bioessays 28, 788–798. doi: 10.1002/bies.20445
Engelen, E., Akinci, U., Bryne, J. C., Hou, J., Gontan, C., Moen, M., et al. (2011). Kiernan, A. E., Pelling, A. L., Leung, K. K., Tang, A. S., Bell, D. M., Tease, C., et al.
Sox2 cooperates with Chd7 to regulate genes that are mutated in human (2005). Sox2 is required for sensory organ development in the mammalian inner
syndromes. Nat. Genet. 43, 607–611. doi: 10.1038/ng.825 ear. Nature 434, 1031–1035. doi: 10.1038/nature03487
Etchevers, H. C., Amiel, J., and Lyonnet, S. (2006). Molecular bases of human Knuchel, M. C., Graf, B., Schlaepfer, E., Kuster, H., Fischer, M., Weber, R., et al.
neurocristopathies. Adv. Exp. Med. Biol. 589, 213–234. doi: 10.1007/978-0-387- (2000). PCR-derived ssDNA probes for fluorescent in situ hybridization to HIV-
46954-6_14 1 RNA. J. Histochem. Cytochem. 48, 285–294. doi: 10.1177/002215540004800214
Fischer, A. H., Jacobson, K. A., Rose, J., and Zeller, R. (2008). Hematoxylin and Kondo, T., and Raff, M. (2004). Chromatin remodeling and histone modification
eosin staining of tissue and cell sections. CSH Protoc. 2008, pdb prot4986. doi: in the conversion of oligodendrocyte precursors to neural stem cells. Genes Dev.
10.1101/pdb.prot4986 18, 2963–2972. doi: 10.1101/gad.309404
Gammill, L. S., and Bronner-Fraser, M. (2002). Genomic analysis of neural crest Kos, R., Reedy, M. V., Johnson, R. L., and Erickson, C. A. (2001). The winged-
induction. Development 129, 5731–5741. doi: 10.1242/dev.00175 helix transcription factor FoxD3 is important for establishing the neural crest
Gammill, L. S., and Bronner-Fraser, M. (2003). Neural crest specification: migrat- lineage and repressing melanogenesis in avian embryos. Development 128,
ing into genomics. Nat. Rev. Neurosci. 4, 795–805. doi: 10.1038/nrn1219 1467–1479.
Garcia-Castro, M. I., Marcelle, C., and Bronner-Fraser, M. (2002). Ectodermal Kouskoura, T., Fragou, N., Alexiou, M., John, N., Sommer, L., Graf, D., et al. (2011).
Wnt function as a neural crest inducer. Science 297, 848–851. doi: 10.1126/sci- The genetic basis of craniofacial and dental abnormalities. Schweiz. Monatsschr.
ence.107824 Zahnmed. 121, 636–646.
Gavalas, A., Ruhrberg, C., Livet, J., Henderson, C. E., and Krumlauf, R. (2003). Labonne, C., and Bronner-Fraser, M. (1998). Induction and patterning of the
Neuronal defects in the hindbrain of Hoxa1, Hoxb1 and Hoxb2 mutants reflect neural crest, a stem cell-like precursor population. J. Neurobiol. 36, 175–189.

Frontiers in Physiology | Craniofacial Biology September 2014 | Volume 5 | Article 345 | 10

Mandalos et al. Head development is fine tuned by Sox2

Laga, A. C., Lai, C. Y., Zhan, Q., Huang, S. J., Velazquez, E. F., Yang, Q., et al. (2010). Remboutsika, E., Elkouris, M., Iulianella, A., Andoniadou, C. L., Poulou, M.,
Expression of the embryonic stem cell transcription factor SOX2 in human skin: Mitsiadis, T. A., et al. (2011). Flexibility of neural stem cells. Front. Physiol. 2:16.
relevance to melanocyte and merkel cell biology. Am. J. Pathol. 176, 903–913. doi: 10.3389/fphys.2011.00016
doi: 10.2353/ajpath.2010.090495 Rex, M., Orme, A., Uwanogho, D., Tointon, K., Wigmore, P. M., Sharpe, P. T.,
Le, N., Nagarajan, R., Wang, J. Y., Araki, T., Schmidt, R. E., and Milbrandt, J. (2005). et al. (1997). Dynamic expression of chicken Sox2 and Sox3 genes in ectoderm
Analysis of congenital hypomyelinating Egr2Lo/Lo nerves identifies Sox2 as an induced to form neural tissue. Dev. Dyn. 209, 323–332.
inhibitor of Schwann cell differentiation and myelination. Proc. Natl. Acad. Sci. Rhinn, M., Schuhbaur, B., Niederreither, K., and Dolle, P. (2011). Involvement
U.S.A. 102, 2596–2601. doi: 10.1073/pnas.0407836102 of retinol dehydrogenase 10 in embryonic patterning and rescue of its loss of
Lee, K., Tan, J., Morris, M. B., Rizzoti, K., Hughes, J., Cheah, P. S., et al. function by maternal retinaldehyde treatment. Proc. Natl. Acad. Sci. U.S.A. 108,
(2012). Congenital hydrocephalus and abnormal subcommissural organ devel- 16687–16692. doi: 10.1073/pnas.1103877108
opment in Sox3 transgenic mice. PLoS ONE 7:e29041. doi: 10.1371/journal. Rijli, F. M., Gavalas, A., and Chambon, P. (1998). Segmentation and specification
pone.0029041 in the branchial region of the head: the role of the Hox selector genes. Int. J. Dev.
Lian, P., Liu, L. A., Shi, Y., Bu, Y., and Wei, D. (2010). Tethered-hopping model Biol. 42, 393–401.
for protein-DNA binding and unbinding based on Sox2-Oct1-Hoxb1 ternary Rizzoti, K., and Lovell-Badge, R. (2007). SOX3 activity during pharyngeal segmen-
complex simulations. Biophys. J. 98, 1285–1293. doi: 10.1016/j.bpj.2009.12.4274 tation is required for craniofacial morphogenesis. Development 134, 3437–3448.
Liem, K. F. Jr., Tremml, G., Roelink, H., and Jessell, T. M. (1995). Dorsal differen- doi: 10.1242/dev.007906
tiation of neural plate cells induced by BMP-mediated signals from epidermal Rutishauser, U., and Jessell, T. M. (1988). Cell adhesion molecules in vertebrate
ectoderm. Cell 82, 969–979. doi: 10.1016/0092-8674(95)90276-7 neural development. Physiol. Rev. 68, 819–857.
Liu, J. P., and Jessell, T. M. (1998). A role for rhoB in the delamination of neural Santagati, F., Minoux, M., Ren, S. Y., and Rijli, F. M. (2005). Temporal require-
crest cells from the dorsal neural tube. Development 125, 5055–5067. ment of Hoxa2 in cranial neural crest skeletal morphogenesis. Development 132,
Mandalos, N., Saridaki, M., Harper, J. L., Kotsoni, A., Yang, P., Economides, A. 4927–4936. doi: 10.1242/dev.02078
N., et al. (2012). Application of a novel strategy of engineering conditional Scherson, T., Serbedzija, G., Fraser, S., and Bronner-Fraser, M. (1993). Regulative
alleles to a single exon gene, Sox2. PLoS ONE 7:e45768. doi: 10.1371/jour- capacity of the cranial neural tube to form neural crest. Development 118,
nal.pone.0045768 1049–1062.
McCauley, D. W., and Bronner-Fraser, M. (2006). Importance of SoxE in neural Selleck, M. A., and Bronner-Fraser, M. (1995). Origins of the avian neural crest: the
crest development and the evolution of the pharynx. Nature 441, 750–752. doi: role of neural plate-epidermal interactions. Development 121, 525–538.
10.1038/nature04691 Shahryari, A., Ra Ee, M. R., Fouani, Y., Alipour, N., Samaei, N. M., Shafiee, M.,
McKeown, S. J., Lee, V. M., Bronner-Fraser, M., Newgreen, D. F., and Farlie, P. G. et al. (2014). Two novel splice variants of SOX2OT, SOX2OT-S1 and SOX2OT-
(2005). Sox10 overexpression induces neural crest-like cells from all dorsoven- S2, Are Co-upregulated with SOX2 and OCT4 in Esophageal Squamous Cell
tral levels of the neural tube but inhibits differentiation. Dev. Dyn. 233, 430–444. Carcinoma. Stem Cells 32, 126–134. doi: 10.1002/stem.1542
doi: 10.1002/dvdy.20341 Stolt, C. C., Lommes, P., Friedrich, R. P., and Wegner, M. (2004). Transcription
Melton, K. R., Iulianella, A., and Trainor, P. A. (2004). Gene expression and regu- factors Sox8 and Sox10 perform non-equivalent roles during oligodendrocyte
lation of hindbrain and spinal cord development. Front. Biosci. 9, 117–138. doi: development despite functional redundancy. Development 131, 2349–2358. doi:
10.2741/1202 10.1242/dev.01114
Mitsiadis, T. A. (2011). Everything is on the head. Front. Physiol. 2:2. doi: Stolt, C. C., Lommes, P., Hillgartner, S., and Wegner, M. (2008). The transcription
10.3389/fphys.2011.00002 factor Sox5 modulates Sox10 function during melanocyte development. Nucleic
Mizusawa, H. (1997). [Corticobasal degeneration]. Rinsho Shinkeigaku 37, Acids Res. 36, 5427–5440. doi: 10.1093/nar/gkn527
1131–1133. Stolt, C. C., and Wegner, M. (2010). SoxE function in vertebrate ner-
Monier, R. (1995). [Cell transformation]. Rev. Prat. 45, 1867–1872. vous system development. Int. J. Biochem. Cell Biol. 42, 437–440. doi:
Nakagawa, S., and Takeichi, M. (1995). Neural crest cell-cell adhesion con- 10.1016/j.biocel.2009.07.014
trolled by sequential and subpopulation-specific expression of novel cadherins. Taneyhill, L. A. (2008). To adhere or not to adhere: the role of Cadherins in
Development 121, 1321–1332. neural crest development. Cell Adh. Migr. 2, 223–230. doi: 10.4161/cam.2.
Narita, Y., and Rijli, F. M. (2009). Hox genes in neural patterning and circuit 4.6835
formation in the mouse hindbrain. Curr. Top. Dev. Biol. 88, 139–167. doi: Taylor, K. M., and Labonne, C. (2005). SoxE factors function equivalently
10.1016/S0070-2153(09)88005-8 during neural crest and inner ear development and their activity is reg-
Nie, X., Luukko, K., and Kettunen, P. (2006). BMP signalling in craniofacial ulated by SUMOylation. Dev. Cell 9, 593–603. doi: 10.1016/j.devcel.2005.
development. Int. J. Dev. Biol. 50, 511–521. doi: 10.1387/ijdb.052101xn 09.016
Pan, H., and Schultz, R. M. (2011). Sox2 modulates reprogramming of gene expres- Trainor, P. A., and Krumlauf, R. (2000). Patterning the cranial neural crest: hind-
sion in two-cell mouse embryos. Biol. Reprod. 85, 409–416. doi: 10.1095/biolre- brain segmentation and Hox gene plasticity. Nat. Rev. Neurosci. 1, 116–124. doi:
prod.111.090886 10.1038/35039056
Pan, W., Jin, Y., Chen, J., Rottier, R. J., Steel, K. P., and Kiernan, A. E. (2013). Trainor, P. A., and Krumlauf, R. (2001). Hox genes, neural crest cells and branchial
Ectopic expression of activated notch or SOX2 reveals similar and unique roles arch patterning. Curr. Opin. Cell Biol. 13, 698–705. doi: 10.1016/S0955-
in the development of the sensory cell progenitors in the mammalian inner ear. 0674(00)00273-8
J. Neurosci. 33, 16146–16157. doi: 10.1523/JNEUROSCI.3150-12.2013 Trainor, P. A., Melton, K. R., and Manzanares, M. (2003). Origins and plasticity of
Panetta, N. J., Gupta, D. M., Slater, B. J., Kwan, M. D., Liu, K. J., and Longaker, M. T. neural crest cells and their roles in jaw and craniofacial evolution. Int. J. Dev.
(2008). Tissue engineering in cleft palate and other congenital malformations. Biol. 47, 541–553.
Pediatr. Res. 63, 545–551. doi: 10.1203/PDR.0b013e31816a743e Trainor, P. A., and Tam, P. P. (1995). Cranial paraxial mesoderm and neu-
Papanayotou, C., Mey, A., Birot, A. M., Saka, Y., Boast, S., Smith, J. C., et al. (2008). ral crest cells of the mouse embryo: co-distribution in the craniofacial
A mechanism regulating the onset of Sox2 expression in the embryonic neural mesenchyme but distinct segregation in branchial arches. Development 121,
plate. PLoS Biol. 6:e2. doi: 10.1371/journal.pbio.0060002 2569–2582.
Pevny, L. H., and Lovell-Badge, R. (1997). Sox genes find their feet. Curr. Opin. Tumpel, S., Cambronero, F., Sims, C., Krumlauf, R., and Wiedemann, L. M. (2008).
Genet. Dev. 7, 338–344. doi: 10.1016/S0959-437X(97)80147-5 A regulatory module embedded in the coding region of Hoxa2 controls expres-
Placzek, M., and Briscoe, J. (2005). The floor plate: multiple cells, multiple signals. sion in rhombomere 2. Proc. Natl. Acad. Sci. U.S.A. 105, 20077–20082. doi:
Nat. Rev. Neurosci. 6, 230–240. doi: 10.1038/nrn1628 10.1073/pnas.0806360105
Prince, V., and Lumsden, A. (1994). Hoxa-2 expression in normal and transposed Tumpel, S., Wiedemann, L. M., and Krumlauf, R. (2009). Hox genes and seg-
rhombomeres: independent regulation in the neural tube and neural crest. mentation of the vertebrate hindbrain. Curr. Top. Dev. Biol. 88, 103–137. doi:
Development 120, 911–923. 10.1016/S0070-2153(09)88004-6
Puc, J., and Rosenfeld, M. G. (2011). SOX2 and CHD7 cooperatively regulate Uwanogho, D., Rex, M., Cartwright, E. J., Pearl, G., Healy, C., Scotting, P. J.,
human disease genes. Nat. Genet. 43, 505–506. doi: 10.1038/ng.843 et al. (1995). Embryonic expression of the chicken Sox2, Sox3 and Sox11 genes

www.frontiersin.org September 2014 | Volume 5 | Article 345 | 11

Mandalos et al. Head development is fine tuned by Sox2

suggests an interactive role in neuronal development. Mech. Dev. 49, 23–36. doi: Wilson, M., and Koopman, P. (2002). Matching SOX: partner proteins and co-
10.1016/0925-4773(94)00299-3 factors of the SOX family of transcriptional regulators. Curr. Opin. Genet. Dev.
Vallaster, M., Vallaster, C. D., and Wu, S. M. (2012). Epigenetic mechanisms in 12, 441–446. doi: 10.1016/S0959-437X(02)00323-4
cardiac development and disease. Acta Biochim. Biophys. Sin. 44, 92–102. doi: Wood, H. B., and Episkopou, V. (1999). Comparative expression of the mouse Sox1,
10.1093/abbs/gmr090 Sox2 and Sox3 genes from pre-gastrulation to early somite stages. Mech. Dev. 86,
Vieille-Grosjean, I., Hunt, P., Gulisano, M., Boncinelli, E., and Thorogood, P. 197–201. doi: 10.1016/S0925-4773(99)00116-1
(1997). Branchial HOX gene expression and human craniofacial development. Wu, J., Saint-Jeannet, J. P., and Klein, P. S. (2003). Wnt-frizzled signaling
Dev. Biol. 183, 49–60. doi: 10.1006/dbio.1996.8450 in neural crest formation. Trends Neurosci. 26, 40–45. doi: 10.1016/S0166-
Villanueva, S., Glavic, A., Ruiz, P., and Mayor, R. (2002). Posteriorization by FGF, 2236(02)00011-5
Wnt, and retinoic acid is required for neural crest induction. Dev. Biol. 241, Yamaguchi, S., Hirano, K., Nagata, S., and Tada, T. (2011). Sox2 expression effects
289–301. doi: 10.1006/dbio.2001.0485 on direct reprogramming efficiency as determined by alternative somatic cell
Vincent, S. D., and Robertson, E. J. (2003). Highly efficient transgene-independent fate. Stem Cell Res. 6, 177–186. doi: 10.1016/j.scr.2010.09.004
recombination directed by a maternally derived SOX2CRE transgene. Genesis
37, 54–56. doi: 10.1002/gene.10226 Conflict of Interest Statement: The authors declare that the research was con-
Wahlbuhl, M., Reiprich, S., Vogl, M. R., Bosl, M. R., and Wegner, M. (2012). ducted in the absence of any commercial or financial relationships that could be
Transcription factor Sox10 orchestrates activity of a neural crest-specific construed as a potential conflict of interest.
enhancer in the vicinity of its gene. Nucleic Acids Res. 40, 88–101. doi:
10.1093/nar/gkr734 Received: 18 July 2014; accepted: 22 August 2014; published online: 12 September 2014.
Wakamatsu, Y., Endo, Y., Osumi, N., and Weston, J. A. (2004). Multiple roles of Citation: Mandalos N, Rhinn M, Granchi Z, Karampelas I, Mitsiadis T, Economides
Sox2, an HMG-box transcription factor in avian neural crest development. Dev. AN, Dollé P and Remboutsika E (2014) Sox2 acts as a rheostat of epithelial to
Dyn. 229, 74–86. doi: 10.1002/dvdy.10498 mesenchymal transition during neural crest development. Front. Physiol. 5:345. doi:
Wegner, M. (1999). From head to toes: the multiple facets of Sox proteins. Nucleic 10.3389/fphys.2014.00345
Acids Res. 27, 1409–1420. doi: 10.1093/nar/27.6.1409 This article was submitted to Craniofacial Biology, a section of the journal Frontiers in
Wen, J., Hu, Q., Li, M., Wang, S., Zhang, L., Chen, Y., et al. (2008). Pax6 Physiology.
directly modulate Sox2 expression in the neural progenitor cells. Neuroreport Copyright © 2014 Mandalos, Rhinn, Granchi, Karampelas, Mitsiadis, Economides,
19, 413–417. doi: 10.1097/WNR.0b013e3282f64377 Dollé and Remboutsika. This is an open-access article distributed under the terms of
Williams, D. C. Jr., Cai, M., and Clore, G. M. (2004). Molecular basis for the Creative Commons Attribution License (CC BY). The use, distribution or repro-
synergistic transcriptional activation by Oct1 and Sox2 revealed from the duction in other forums is permitted, provided the original author(s) or licensor are
solution structure of the 42-kDa Oct1.Sox2.Hoxb1-DNA ternary transcrip- credited and that the original publication in this journal is cited, in accordance with
tion factor complex. J. Biol. Chem. 279, 1449–1457. doi: 10.1074/jbc.M3097 accepted academic practice. No use, distribution or reproduction is permitted which
90200 does not comply with these terms.

Frontiers in Physiology | Craniofacial Biology September 2014 | Volume 5 | Article 345 | 12