University of Southern California Dissertations and Theses

Elucidating the role of neural crest specific Stat3 signaling in maintaining coronal suture patency during embryonic development

(USC Thesis Other)

Elucidating the role of neural crest specific Stat3 signaling in maintaining coronal suture patency during embryonic development

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

ELUCIDATING THE ROLE OF NEURAL CREST SPECIFIC STAT3
SIGNALING IN MAINTAINING CORONAL SUTURE PATENCY DURING
EMBRYONIC DEVELOPMENT

By

Krishnakali Dasgupta

A Dissertation to
THE FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirement of the degree
DOCTOR OF PHILOSOPHY
(GENETIC, MOLECULAR, CELLULAR BIOLOGY)

DECEMBER 2014
ii

DEDICATION

I dedicate my thesis firstly to the 2 the people who I know would have been happiest if
they had been able to see this,

My Uncle the Late Nabendu Narayan Dasgupta who was more than a father to me

And

My dearest friend the Late Reema Ray who was more than a sister.

iii

ACKNOWLEDGEMENTS

I wish to thank Dr. Maxson for his immense guidance and support during my years as a
graduate student in his lab. Without your encouragement this would not have been possible.

I wish to thank every single member of the Maxson lab Jingjing, Mandy, Helen Mamoru,
Chris, Mia, Nancy, Youzhen, Camilla, Jessica, Dhvani and Athena for their immense help
and enthusiasm through the years.

I wish to thank my committee members Dr. Baruch Frenkel, Dr. Wange Lu and Dr.
Samantha Butler for your continued support and immeasurable help in my work.

My deepest gratitude to,

My Father and Mother, Debabrata and Ujjala Dasgupta whom I wish to mention separately
and not just as my parents, the 2 most unconventional people on earth, for showing me that
every moment in life was to be taken as an adventure and there was nothing on earth I
needed to be afraid of.

Especially my aunts Ratna Dasgupta and Ellora Roy Chowdhury and Uncles Biplab
Sengupta, Jadulal Bhowmick, and Late Ajay Roy Chowdhury for giving their all and being
my surrogate parents in every sense of the word.

All my Uncles, Aunts and Grandparents who are here or in my heart and memories for
supporting me to the best of their ability in every possible way. I don’t space to mention
all your names but I am counting every single person who has ever sent a blessing, a helping
hand or even a smile my way.

My beautiful family of in laws especially my parents in law Dr.Kumardeb Bose and
Bhaswati Bose and who not only welcomed me whole heartedly into their fold but also
supported and cheered me along all the way.

My brothers and sisters especially Jishnu, Arup, Shatarupa, Shuvalaxmi, Debadrita, Pratik,
Shubhra, Shantonob, Shushanta, Sudeshna, Milan Suchita, Soumik, Sourav, Mahua,
Paramita, Sangram, Rupasree, Dyuti, Shimana and all others, as well as my all my nephews
and nieces who have been my strength and took this journey with me. I cannot bring myself
to call you just my cousins. You are my life.

My friends from school especially Rilina, Madhumita, Sayantani, Sreeparna, Diya,
Debasreeta, Moumita, Priyanka, Bonojytosna, Tulika, Shibapriya, Ujjaini, Urmi, and
Debarati di from GMGS and Mary, Mandrita, Priyanka, Rituparna, Poushali, Sudarshana,
Tilottama and Sohini from LMG and all others whom I have not mentioned . I have grown
iv

up idolizing every one of you and each of your stories inspire me every day. Every success
story of your lives is my success and this success is also yours.

My friends from University Simantini, Pradipta, Ayan, Dipanwita, Budhaditya, Mandovi,
Rashmi, Sunetra, Amrita, Deblina, Joyita, Sayantani, Atreyee, Shubhadeep, Satyaki,
Paromita, Anindita and Nilotpal. This journey might not have happened if I had not met
you all. We dreamt this together and I draw strength from every one of you every single
day. I hope we continue this journey as colleagues and see this till as far as we can.

All my teachers and professors, especially the uncomparable Dr. Maitrayee Dasgupta,
Dr.Chandan Mitra and Dr.Amrita Bannerjee. I am what you have made me. I hope I can
live up to your expectations.

Aarti, Ankita, Minal, Shikhar, Rahul, Divya, Srividya Vicky, Dipankarda , Pavinder
Tumul bhaia, Leena, Mallika and Joyita, Rumi, Adarsh, Sushmita, Anustup for being my
family at USC. I shall never be able to give you back the tremendous amount of support I
received from you all. You have indulged me and tried to smoothen every difficulty in my
way. I bow to you.

All my friends at USC, especially my friends from Vidushak ( Kimish, Animesh, Manoj,
Jon, Vikram, Pankaj, Mansi, Yamini, Megha, Mallika ) and my ex-labmates (Jason, Han
and Benny) who kept my spirit going for all these years.

My dance teachers Dr. Sohini Roy and Poushali Chatterjee and my colleagues Debanjana
Roy and Debanjali Biswas. You made sure I could achieve this and yet follow my passion,
adjusting all your schedules and even life events. I am so lucky and proud to have you in
my life. Without your support one half of my world would have been empty.

Even though I have mentioned these names before, specially calling out to Aarti, Simantini,
Sushmita, Adarsh and Sudeshna (didi). Your contribution to this is immeasurable. You
truly make up my pillars of strength.

And of course,

Sayantan, friend of my heart and husband of my soul. You are my silent super-hero.
You are – therefore I am.
v

TABLE OF CONTENTS

Dedication
Acknowledgements
List of Tables
List of Figures
Abstract
INTRODUCTION
1.1 The Human Skull
1.2 Calvarial Sutures: Definition and development
1.3 Calvarial Sutures: Histology and formation (how bone growth actually occurs)
1.4: Calvarial Sutures: Tissue origin
1.5 : Calvarial Sutures : Craniosynostosis
1.6: Calvarial Sutures: Genetics of Craniosynostosis and Signaling in the sutures.
1.7: Craniosynostosis : Syndromes and Molecular mechanisms
1.7.1: FGFR [Apert, Crouzon, Pfeiffer, Muenke]
1.7.2: MSX2 [Boston Type]
1.7.3: TWIST1 [Saethre-Chotzen]
1.7.4: EPHNB1 [Craniofrontonasal syndrome]
1.7.5: FBN1 & TGFB [Loeys Deitz Syndrome, Marfan syndrome]
1.8: Craniosynostosis: Signaling mechanisms and cranial suture development
speculations
1.8.1: FGF signaling in mice
1.8.2: Twist1/MSX2/ Ephrin-Eph/ Jagged-Notch signaling
1.8.2a: Twist1 and Msx2
ii
iii
ix
x
13
14
16
19
24
27
28
31
35
35
38
38
40
41
42
42
43
43
vi

1.8.2b: Twist1 and Eph-Ephrin
1.8.2c: Twist1 and Jagged-Notch
1.8.2d: Twist1 and Fgfr
1.8.3: BMP signaling
1.8.4: HH /Gli3 /RAB23 signaling – Metopic suture
1.9 : Craniosynostosis : Cellular & Tissue induction and interaction
1.9.1: Cell mixing and Craniosynostosis
1.9.2: Osteoblast differentiation and Craniosynostosis
1.9.3: Dura mater signaling – Tissue induction theory
2.0: Conclusion : Summarized introduction to the project
Chapter 1 :
Neural crest specific conditional loss of Stat3 leads to partial bicoronal suture
synostosis in mice via crest specific loss of Notch2 receptor expression and
concurrent activation of ectopic osteogenic activity in the suture area
Introduction
Results
1.1 Neural crest specific loss of Stat3 signaling leads to Coronal suture synostosis
1.2A. Deficiency in the neural crest-mesoderm boundary within the coronal suture
during suture genesis in neural crest-specific Stat3-/- mice
1.2B Partial and varied forms of fusion of newly formed frontal and parietal bone
fronts in horizontal sections of E14.5 Neural crest specific homozygous
inactivation of Stat3 mutants
1.3.A : No gross difference in cellular distribution or tissue morphology is
observed in Wnt1cre population of cells in whole mounted samples stained for
LacZ
1.3B The distribution of Wnt1-Cre/R26R-marked cells shows individual cells
from the neural crest population beyond the neural crest boundary and in the
mesodermal zone.
1.4 Investigation of molecular markers expressed in the developing coronal suture
region early during suture development and projected to play an active role in the
establishment and maintenance of the latter, shows no significant change in
45
45
47
49
50
52
53
53
55
56

58
58
62
62
68

71

74

77

vii

expression of suture cell specific markers (Twist1, Jagged1) or neural crest-
mesoderm boundary establishment marker (Ephrin-Eph), but a reduction in the
expression of the membrane bound receptor Notch2 specifically in the growing
frontal bone front.
1.4.A. Investigation of Eph-Ephrin expression in the ectocranial layer of
developing coronal suture region during early suture development
1.4.B. Investigation of Twist1 expression in the developing coronal suture region
during early coronal suture development
1.4.C. Investigation of Jagged1 expression in the developing coronal suture
region during early coronal suture development
1.4.D. Investigation of Notch2 expression in the developing coronal suture region
and growing bone fronts during early coronal suture development
1.5. Investigation of expression of early osteogenic marker Runx2 in the
developing coronal suture region early during suture development, shows ectopic
expression in the suture region in neural crest specific homozygous loss of Stat3.
1.6. Investigation of expression of PSmad1/5/8 (BMP signaling marker) in the
developing coronal suture region early during suture development, shows no
ectopic expression in the suture region in neural crest specific homozygous loss
of Stat3.
Discussion
Materials and Methods
Chapter 2:
Notch2 signaling in cranial neural crest tissue is directly involved in
maintaining coronal suture patency and acts downstream of Stat3 in
production of the coronal suture synostosis phenotype observed in neural
crest specific Stat3-/- mice.
Introduction
Results
2.1 Neural crest-specific loss of Notch2 leads to coronal suture synostosis of a
greater severity that in case of neural crest specific loss of Stat3
2.2 Investigation of Jagged1 expression in the developing coronal suture region
during in neural crest specific Notch2 floxed mice

80
83

85

87

91

94

97
98
110

116
116
119
119

125
viii

2.3 Investigation of expression of early osteogenic marker Runx2 in the
developing coronal suture region early during suture development, shows ectopic
expression in the suture region in neural crest specific homozygous loss of Notch2
2.4. Investigation of LacZ staining Wnt1cre population of cells of neural crest
origin, shows presence individual cells from the neural crest population beyond
the neural crest boundary & in the mesodermal zone in Wnt1cre; Stat3
cko/cko
mutants
2.5 Neural crest-specific expression of constitutively active Notch signaling
rescues coronal synostosis in mutants with neural specific loss of Stat3
Discussion
Materials and Methods
Chapter 3: [Preliminary Data of Parallel Project ]
Stat3 signaling is active in a cluster of the migratory neural crest population
in the ectocranial region
Introduction
Preliminary Results
3.1 P-Stat3 positive cells cluster in the ectocranial layer adjacent to the frontal
bone
3.2 Assessment of P-Stat3 positive cells in coronal section of mice heads at
successive embryonic stages show cluster of PStat3 positive cells in the
ectocranial layer are absent in the neural crest specific Stat3 mutants
3.3 Analysis of PH3 positive cells in the growing frontal bone primordia in E13.5
mouse heads during early the embryonic stage of coronal suture formation shows
no significant reduction in the number of proliferative cells in neural crest-specific
Stat3 mutants.
Discussion
Materials and Methods
Summary and future directions
Supplementary data
Bibliography
References

128

131

134
136
141

147
147
149
149

155

158
159
160
164
168
177
178
ix

List of Tables
Table A Genetics of Craniosynostoses 32
Table 1 Craniosynostosis Index [ Neural crest specific loss of
Stat3 ]
63
Table 2 Craniosynostosis Index [ Neural crest specific loss of
Notch2 ]
121
Table 3 Craniosynostosis Index [ Neural crest specific rescue of
Stat3 mutants by constitutively active form of NICD ]
133

x

List of Figures
Figure 1.1

Figure 1.1a
Figure 1.1b
Neural crest specific conditional loss of Stat3 leads to partial
bicoronal suture synostosis
P21 stage mice skulls stained with Alizarin Red (calcified tissue)
uCT imaging of P21 stage mouse skulls

63
64
Figure 1.2

Figure 1.2A

Figure 1.2B
Neural crest specific conditional loss of Stat3 leads to partial
disruption of coronal suture early during embryonic development
Whole Mounted mouse heads stained for detection of ALP
(developing bone tissue)
ALP stained Horizontal sections of E14.5 stage mouse embryo
heads

67

70

Figure 1.3

Figure 1.3A
Figure 1.3B

LacZ marking of neural crest cells in Neural crest specific
conditional loss of Stat3 displays minor loss of crest-mesoderm
border recognition
Whole Mounted E14.5 mouse embryo heads stained for LacZ.
Investigation of localization of LacZ stained cells in Horizontal
sections of mice heads at successive embryonic stages

73

76
Figure 1.4A Investigation of Ephrin-Eph expression in Horizontal sections of
mice heads during early embryonic stages of coronal suture
formation

82
Figure 1.4B Investigation of Twist1 expression in Horizontal sections of mice
heads during early embryonic stages of coronal suture formation

84
Figure 1.4C Investigation of Jagged1 expression in Horizontal sections of mice
heads during early embryonic stages of coronal suture formation

86
Figure 1.4D Investigation of Notch2 expression in Horizontal sections of mice
heads during early embryonic stages of coronal suture formation

90
xi

Figure 1.5 Investigation of early osteoblast marker Runx2 expression in the
developing coronal suture area in mice with Neural crest specific
conditional loss of Stat3 shows ectopic expression of Runx2 in the
suture area

93
Figure 1.6 Investigation of PSmad1/5/8 expression in the developing coronal
suture area in mice with Neural crest specific conditional loss of
Stat3

96
Figure 2.1

Figure 2.1a
Neural crest specific conditional loss of Notch2 leads to partial
bicoronal suture synostosis
P21 stage mice skulls stained with Alizarin Red (calcified tissue)
and ALP stained Horizontal sections of E14.5 stage mouse embryo
heads.

120
Figure 2.2 Investigation of Jagged1 expression in Horizontal sections of mice
heads during early embryonic stages of coronal suture formation

124
Figure 2.3 Investigation of Runx2 expression in Horizontal sections of mice
heads during early embryonic stages of coronal suture formation

127
Figure 2.4 LacZ marking of neural crest cells in Neural crest specific
conditional loss of Notch2 and Stat3 display similar minor loss of
crest-mesoderm border recognition

130
Figure 2.5 Neural crest specific conditional activation of Notch signaling
with constitutively active Notch-Intracellular domain (NICD)

xii

expression leads to rescue of the partial bicoronal suture synostosis
caused by Neural crest specific conditional loss of Stat3

133
Figure 3.1 Assessment of PStat3 positive cells in horizontal section of E14.5
mice show group of PStat3 positive cells in frontal area of
Ectocranial layer (ECL)

150
Figure 3.2A Assessment of PStat3 positive cells in Coronal (Dorso-ventral)
section of E13.5 and E14.5 mice show group of PStat3 positive
cells in frontal area of Ectocranial (ECL) layer close to the
growing frontal bone front

152
Figure 3.2B Assessment of PStat3 positive cells in Coronal section of E13.5
and E14.5 mice showing group of PStat3 positive cells in
Ectocranial (ECL) layer close to the growing frontal bone front
only in Wildtype mice

153
Figure 3.3 Assessment of proliferative cells (PH3 positive) in Coronal
section of E13.5 mice in the growing neural crest derived frontal
bone primordium area

157

13

ABSTRACT
Craniosynostosis refers to the genetic condition of premature fusion of calvarial sutures
which occurs in roughly in 1 out of 2000 -2500 live births. It occurs as both syndromic
and nonsyndromic disorder, where premature fusion of the skull bones hinders the proper
expansion of the brain causing increase in intracranial pressure. This leads to improperly
or distortedly formed brain and the skull accounting for mild to severe mental retardation
and other life threatening craniofacial defects. Over the last two decades distinct cases of
syndromic and non syndromic cases of Craniosynostoses have been shown to be caused by
the inactivation or hyper-activation of individual genes e.g. Twist1, Jagged1, FGF, Msx2,
Ephrin-Eph. Recently it was shown that Hyper Immunoglobulin syndrome (HIES) a multi-
system immunological disorder characterized by increased serum IgE levels and displaying
craniosynostosis as one of its clinical manifestations, is caused due to a loss of function
mutation of the Stat3 gene .
My work focusses on elucidating the role of Stat3 signaling in the neural crest cell derived
frontal bone in maintenance of coronal suture patency. I show here that neural crest specific
loss of Stat3 signaling leads to loss of Notch2 expression in the crest derived frontal bone
and causes synostosis of the coronal suture. Subsequently it is shown that neural crest
specific loss of Notch2 signaling also causes fusion of the coronal suture. These studies
help in further understanding of etiology of the craniosynostosis disorders and the defects
in molecular mechanisms that cause the phenotype.

14

INTRODUCTION
The idea of biological inheritance of organismal traits and the process of formation of fully
functional organisms from the earliest observable form of a new individual has fascinated
biologists for several centuries, but it was in the early decades of the twentieth century that
several of these diverse fields of study were begun to be collectively grouped under the
common school of studies on ‘Developmental Biology’. The main reason for this initiative
was the realization that several areas of study carried out in the field of biological sciences:
(i) Embryology (mostly descriptive studies of the process of gastrulation and formation of
the germ layers and the subsequent emergence of the different tissues and organs in various
vertebrate and invertebrate organisms); (ii) Cell biology (morphology and characteristics
and changes thereof, cell division, proliferation, differentiation and death, cellular
migration); (iii) Evolution ( interrelation of species and emergence of new tissues and
organs); (iv) Genetic inheritance (of physical and physiological traits as well as diseases);
(v) Metamorphosis (or cyclical transformation of organisms from form to the other during
their lifetime), and (vi) Regeneration (of body parts, organs or tissues), are broadly related
to each other to a large extent and it is much more fruitful to study one idea in the context
of the others to be able to draw a more logical and hence more accurate picture of the
phenomenon called life.
As discussed above, one the primary foci of the study of Developmental Biology
encompasses Embryology, i.e., the study of the process of formation of the embryo from
the maternal and paternal gametes and its subsequent development into a fully formed
individual organism. It is one of the most astoundingly synchronized and spatio-temporally
15

coordinated cellular and tissue organizing process, by virtue of which, after originating as
a single celled zygote, every organism subsequently passes through a series of
characteristically defined multicellular embryonic stages to finally grow into a fully
functional biological entity capable of self-sustenance and reproduction. During this
process of transformation of a single celled entity into a fully functional biological
organism with distinctive morphological features and well-defined organ systems with
assigned functions, (the total time assigned for which actually encompasses a very short
segment in the life span of the multicellular organism), on close observation, an astonishing
amount of dynamic cellular processes, including proliferation, migration, differentiation,
pattern formation (not a cellular process itself , but a result thereof) and apoptosis are
revealed to be at play. These processes manifest as large-scale morphogenetic events such
as gastrulation, germ layer formation, fronto-parietal vs dorso-lateral axis formation, tissue
morphogenesis and organogenesis. Interestingly the order and logic of cellular and
molecular processes underlying tissue morphogenesis during embryogenesis vs adult tissue
repair, regeneration or metamorphosis are more often than not similar than singular, no
matter how disparate in location, function, cellular composition or time of development
may the tissues be. This congruency in cellular and molecular mechanisms used to achieve
similar tissue morphogenesis is of great value to developmental biologists making it
feasible to intensively study the developmental dynamics of single tissues or portions
thereof in greater detail than any other parts of the organism and yet be able to extrapolate
data gathered from those studies to the understanding of developmental processes in the
organism as a whole. Besides, the formation and patterning of any organ or tissue structure
16

is also used by developmental biologists as a model of investigation not only to gain
insights about the underlying cellular and molecular mechanisms in order to understand
patterning as a whole ,but also to investigate the pathophysiology of heritable or non-
heritable defects in the tissue.
1.1 The Human Skull
One of the most important organs that humans possess that has had an insurmountable
impact on our development into a species of advanced intelligence that we are today, is not
just the brain but the bony skull or cranium that we possess that encases and protects that
brain throughout life. Evolutionary studies suggest that the emergence of the skulled
species of chordates mark the emergence of active predators from the existing pool of filter
feeders among the primitive chordates [Northcutt 2005]. Consequently we find Linneaus
alternatively referring to vertebrates as belonging to the Clade ‘Craniata’ which roughly
includes all organisms that possess a head like structure with a bony cranium encapsulating
a brain and a jaw like structure assisting in predatory feeding. The purpose of this bony
cranium is also proposed to be manifold, since other than encapsulating and protecting the
soft brain tissue, the predesigned morphological shape of the skull helps to fix the position
of the eyeballs helping in binocular vision as well as determining the position of the
earflaps thus helping in an enhanced auditory perception of the vertebrate predators
[Heinen & Vinken 2011]. In addition to this, the emergence of the head with the ossified
skull is potentially explained by the “New Head Hypothesis” as proposed by Northcutt and
Gans (’83) - which argues that vertebrate cranial tissue originating specifically from the
17

cranial neural crest, a transiently arising migratory population identified only in
vertebrates. Neural crest cells arise early during neural tube formation from the cells at the
junction of the neural and non-neural ectoderm and contribute heavily to craniofacial tissue
as well as to a few other specific tissues including skin keratinocytes, enteric nervous
system and smooth muscles in different parts of the body. Many of these cell types are
crucial for an active predatory feeding strategy characteristic of vertebrates [Northcutt
2005].
The vertebrate skull or cranium consists dorso-ventrally of two major components, broadly
referred to as the Neurocranium, which encases the brain and the Viscerocranium, which
supports the pharyngeal structures. The Viscerocranium (in humans consisting of 14 bones
namely the paired nasal, maxilla, palatine, zygoma, lacrimal and inferior nasal conchae and
the singular vomer and mandible) is the vertebrate equivalent of the most primitive
‘Splanchnocranium’ which in earlier vertebrates supported the gills and respiratory
muscles and in higher vertebrates like mammals forms the jaws (mandible and maxilla),
the hyoid bone, as well as the entire fronto-nasal and facial structure. Conversely, the
Neurocranium or the dorsal portion of the skull also referred to as the skull case, is the
portion that actually encases and protects the brain. The Neurocranium is made of the of
the more primitive ‘chondrocranium’ or the skull base and the ‘cranium proper’ or the skull
cap at the top made of the skull bones which give the skull its characteristic hollow
spherical shape and is considered evolutionarily to be arising from the dermal plates
making the ‘Dermatocranium’ in primitive jawless fishes. In humans the Neurocranium is
made up of 8 bones namely the parietal and temporal bones which are paired and the
18

frontal, sphenoid, ethmoid and occipital bones which are single elements. [Chai & Maxson
2006, Helms et al 2005, Kuratani 2005, Morris-Kay & Wilkie 2005]
From an evolutionary perspective, the skull is considered a ‘neomorphic’ organ (newly
formed for a specific purpose) that has been produced as a result of encephalisation
(formation of a brain) with a cranium portion and a jaw like portion. But overall the skull
is the product of several small bony pieces attached together with fibrous tissue giving rise
to a hollow box like structure with several flattened surfaces encircling a cavity. Originally
it was theorized to be a rostral extension of the segmented vertebrae made of an uneven set
of vertebral pieces. [Kuratani 2005]. This idea was later modified into the theory that the
head is actually a completely neomorphic organ (rather than just an extension of the
vertebral column) but one that showed a certain segmentation pattern in the development
of the cavities of the viscerocranium that arise systematically from anterior to posterior as
premandibular, mandibular and hyoid cavities. The idea of a preformed segmentation
pattern similar to the more primitive vertebral column pattern stems from the pairing of a
specific set of cranial nerves with each pair of developing cranial cavities of the
viscerocranium. These include the oculomotor, trochlear and abducens respectively with
premandibular, mandibular and hyoid cavities, which are coincidentally also associated
with 3 pairs of eye musculature. A similar segmentation for the neurocranium is currently
an argument of discussion especially since though the viscerocranium has been
demonstrated to be composed entirely of Neural Crest Cells, the segments or cavities have
gone on to develop several class and species specific modifications (e.g. middle ear
osscicles in vertebrates) , besides the composition of the neurocranium itself being shown
19

to originate from both neural crest and mesodermal, with the relative contribution of the
two types of cells for the different parts making up the neurocranium being different for
different vertebrate species. [Kuratani 2005, Chai & Maxson 2006, Hanken & Hall (
book)].
Currently the skull is thought to develop as a result of bidirectional inductive interactions
between multiple tissue (epithelial and mesenchymal) and cell types (neural crest and
mesodermal), with the help of the surrounding nonskeletal tissue. A significant effort is
under way to understand the inductive effect of the surrounding epithelial, mesodermal and
growing brain tissue on the proliferation, migration and differentiation of the neural crest
derived craniofacial structural elements as well as the rostral most part of the neurocranium
encapsulating the prefrontal and frontal cortex of the brain.
Notable anatomical differences among skulls across species within the same genus or
classes is also regularly used landmarks to assign evolutionary relationships of newly
discovered species with previously known ones .
1.2 Calvarial Sutures: Definition and development
As we study the anatomy of the adult skulls of humans and other vertebrates, it is evident
that the skull is composed of a number of ossified elements (of varied shape and size)
attached to each other through unossified dense fibrous connective tissue. These unossified,
movable fibrous joints between the flat bones of the neurocranium are known as sutures.
Sutures are thus the location of apposition of two bones. Such bones include those of the
neurocranium namely the frontal, parietal, temporal, occipital are flat bones
20

From anterior to posterior the sutures of the mammalian skull vault are namely Metopic
(midline suture between 2 frontal bones), Coronal (horizontal and perpendicular to the
midline between frontal and parietal bones), Sagittal (along the midline between the 2
parietal bones) and Lambdoid (perpendicular to the midline between parietal and occipital
bones), along with the paired Squamosal sutures (between the temporal and parietal bones
on either lateral surface of the head. Among the sutures, only the metopic suture is fused
early during childhood (by 18 months) as is its counterpart the posterior frontal suture in
mice (at birth). The rest of the sutures remain patent for the greater part of the life of
humans. The functionally similar growth centers along the border of cartilaginous pieces
of the chondrocranium or skull base are known as Synchondroses [Morris-Kay & Wilkie
2005, M. Michael Cohen Jr, 1993, M. Michael Cohen Jr, 2005].
Flat bones and Intramembranous ossification
Flat bones or intramembranous bones are formed by direct deposition of osteoid material
by osteogenic cells (osteoblasts) within compact condensate of neural crest or mesodermal
cells, while the bones of the skull base are made by endochondral ossification where a
previously formed cartilaginous framework is replaced later by osteoid material from the
osteoblasts. In contrast the cartilage of the skull base is formed by direct deposition of
cartilage material (mainly collagen and other extracellular matrix material) by the cartilage
forming cells (Chondrocyte). [David, Poswillo, Simpson 1982 (Book), Hanken & Hall,
1993 (Book)].
21

In humans at around 28 days after conception after the formation of the brain, eyes and
cranial nerves primordia, the first evidence of a mesenchymal coalescence becomes evident
below the brain shelf, Later by post-conception day 40 this mesenchyme begins
differentiating into cartilage condensates setting the location and time of the establishment
of the chondrocranium or skull base. After the cartilaginous chondrocranium is laid down,
intramembranous ossification begins at the caudal end and proceeds rostrally and laterally
to produce the four basicranium bones – the occipital, temporal, sphenoid and ethmoid.
The timing of embryonic development of this skull vault, follows that of the
chondrocranium very closely. Eventually, the skull vault is composed of the paired frontal
and parietal bones and the unpaired interparietal bone making up the Desmocranium, which
is later called the Neurocranium.
It has been shown [Yoshida et al 2008, Morris-Kay & Wilkie 2005] that in vertebrates the
intramembranous development of the flat bones begin at the cranial base and extend
anteriorly and towards the edges. The sutures thus become obvious after the main structure
of the bones is laid out and the edges begin to form and come close to each other, though
the sutural precursors are present in the region previously.
Significance of the sutures
One of the consequences of bipedal locomotion developed in later apes is restriction of the
hip width within a certain range for easier maintenance of balance, which in turn restricts
the birth canal to be relatively narrow as compared to the eventual size of the cranium.
Thus human babies are born much earlier than the full development of the eventual shape,
22

size or bone density of the skull vault as well as the development of the full extent of the
brain. Postpartum, the human brain takes almost a decade and a half to grow to its full size
and consequently the skull also has to remain pliable to allow continued growth.
Consequently, throughout the period of brain expansion, growth and development, the
skull also remains a dynamically expanding and growing structure increasing in size, in
tandem with the growth of the brain proper in order to perfectly accommodate the organ at
every stage of its development. Concurrently, calcification of the already formed bone
tissue continues to take place to give the brain the protection and support , as is expected
of the skull vault . This apparent flexibility to accommodate the expanding and rapidly
developing brain is achieved by the growing skull with the help of the fibrous suture.
[Morris-Kay & Wilkie 2005, M. Michael Cohen Jr, 1993, M. Michael Cohen Jr, 2005]
Each of the members making up the individual bony pieces of the neurocranium develop
separately , and as the underlying brain expands in volume, the subsequent surface
expansion of the bones occurs at the boundaries which also happens to be the outer border
of the sutures. Growth of the skull bone edges hence occur along the suture zone in a
direction perpendicular to the sutures. Thus, throughout the period of brain and skull
development (which in humans continues through puberty) the bone edges which
consequently form the outer boundaries of the sutures are proposed to be zones of dynamic
tissue development where there is continuous cell proliferation, migration, differentiation,
deposition and turnover. [Morris-Kay & Wilkie 2005, M. Michael Cohen Jr, 1993, M.
Michael Cohen Jr, 2005]
23

It is interesting to note that the midline sutures (sagittal and metopic) are end-to end sutures
i.e. ,the level of the adjacent bone fronts perfectly align with each other, while in the case
of the sutures perpendicular to the midline i.e., the coronal and lambdoid either of the bone
edges ride above the other creating an overlapping of bone edges. This interlocking of the
sutures while allowing a comparative amount of movement among the bones thus is also
predicted to help in protecting the bones from becoming separated or displaced from
external mechanical injuries and trauma during early childhood and also efficiently
spreading out any incoming external force or trauma along the movable softer fibrous
connective tissue of the sutures acting as shock absorbers [D P Rice 2008 (book)].
Besides acting as the location of dynamic growth of the calvarial bones and the tissue
helping in absorbing and dissipating mechanical trauma, the developing suture tissue is
also a major signaling center affecting the proliferation and differentiation and hence the
eventual identity and patterning of the surrounding tissue [D P Rice 2008 (book)].
Thus, rather than being a functionally passive mechanical movable joint between the bones
of the skull, the suture is an active participant in the development and patterning of the
skull vault from very early on during the period of neurocranium development.
Consequently, over the past few decades cranial sutures have been closely studied by
developmental biologists to understand fundamentals of tissue patterning in a general
context as well as the specific context of the head as well as to for better understanding of
the etiology of clinical syndromes and the cause-effect relationship between the genes
affected and phenotypes produced.
24

1.3 Calvarial Sutures: Histology and formation (how bone growth actually occurs)
As mentioned before, the sutures found in the craniofacial system are mainly of two types,
those made up of fibrous connective tissue (sometimes referred to as syndesmoses) which
are generally annotated by the word ‘craniofacial sutures’ and those which are cartilaginous
joints between bones referred to as synchondroses, mainly found in the chondrocranium or
skull base. [David, Poswillo, Simpson 1982 (Book) , Morris-Kay & Wilkie 2005, M.
Michael Cohen Jr, 1993, M. Michael Cohen Jr, 2005].
Regarding the actual development of the sutures, the definitions and descriptions are
slightly ambiguous, especially since effectively the sutures have two zones, the area lining
the growing bone front and the expanse of the undifferentiated layer of cells between them.
Initially at the time of the appearance of the small condensations at the ventral edges of the
skull vault which are the calvarial bone primordia, a layer of soft, undifferentiated tissue
(a few cell layers thick) is observable to extend between across the top of the skull between
the primordia. [D P Rice 2008 (book)]
Histologically the calvarial bones lie between the outermost epithelial covering of the head
on the dorsal side and the dura mater (the outermost of the brain meninges) lying ventral
to them. These being flat bones they form by intramembranous ossification which begins
by the appearance of an ossification center in the existing mesenchymal cell layer. For this
a minimum number of Mesenchymal Stem Cells of wither neural crest or mesodermal
origin cluster together which leads to the differentiation of a few of them into the basic
osteoprogenitor cells called Osteoblasts. The osteoblasts secrete osteogenic material like
25

Collagen type I leading to the production of the earliest stage of a bony matrix around the
osteoblasts called the Osteoid. At this point and in the following period the osteoblasts are
found lining the newly formed matrix with the osteoid lying at the center of the cluster.
[Brighton & Hunt 1991, D P Rice 2008 (book)]
Subsequently as the osteoblasts continue to remain in the periphery of the growing
ossification center which steadily grows in size radially outward, the central osteoid
becomes further mineralized and forms a trabecular structure with the help of bone
spicules. At this point the presence of a clear periosteum (membranous connective tissue
covering the surface of the bones) is observable. The periosteum contains the osteoblasts
that continue to secrete bony material. With time some pf these osteoblasts get trapped in
the bony matrix and become hardened into osteocytes while fresh population of osteoblasts
are seen to be supplied by the respective mesodermal or neural crest source or later by
proliferation of the existing population of MSCs . This periosteum, in the case of the
calvarial bones, is often in contact with the outer dermal layer of the skin as well as the
underlying dural layer, which gave rise to the speculation that signaling from either of these
membranes may have a role to play in the formation and development of the calvarial bones
and associated sutures. [Brighton & Hunt 1991, D P Rice 2008 ( book)]
As the bones begin to expand dorso - laterally, mesenchymal cells have been shown to
migrate along the mesodermal ectocranial layer (lying between the bones and the skin) in
order to reach the growing bone edges, where they give rise to osetogenic precursors.
[Roybal et al 2010, Ting et al 2009 ] As the bone edges grow closer to each other, the non-
26

bony suture areas become more defined and distinguishable until they become established
as narrow fibrous connective tissue joints between the moderately complete bones around
the time of birth. At this point the general shape and structure of the bones is established
which is followed by the period of expansion of the borders radially. Further growth of the
intramembranous calvarial bones thus occurs at the suture borders which is the periosteal
location of the differentiating osteoblasts. Though after this during the childhood years
there is a continuous episode of expansion of the suture zone followed by extension of the
calvarial bone edges over this newly opened up space to keep the suture effectively within
a narrow width range till puberty after which though we do find a continued turnover of
the cellular makeup of the bony edges throughout the lifetime of the individual due to
typical bone turnover activity of the skeletal tissue. [M. Michael Cohen Jr, 1993, M.
Michael Cohen Jr, 2005]
The exact location the coronal and lambdoid sutures have also been under heavy scrutiny
since the midline sutures clearly get aligned at the apposition points of the equidistantly
separated growing fronts of the parietal and frontal bone pairs, but the exact location of
development of the Coronal and Lambdoidal sutures seem ambiguous and unpredictable.
It has been argued that the location of the lambdoid suture is actually determined by the
relative growth of the adjacent bone fronts while the position of coronal suture has a
patterning determinant, but no specific correlation with any underlying brain structure has
gained momentum. [Morris Kay & Wilkie 2005, D P Rice 2008 (book)] Besides, it is also
proposed that the underlying dura mater can act as a source of inductive signal in
27

determining the location of the sutures in response to its own changes in tension due to
cephalic expansion [Opperman et al 1993, Opperman 2000].
1.4: Calvarial Sutures: Tissue origin
It has been known since the beginning of the twentieth century that the craniofacial skeletal
and associated structural and connective tissue system is of mixed origin with varying
degrees of contribution by the Mesodermal and Neural crest cells in different vertebrate
species. Extensive studies have been carried out since the early 1980s to determine the
actual contribution of the two cell types for each facial element. From the beginning it was
proven beyond doubt that the viscerocranium forming the facial structures is almost solely
of neural crest origin , but the elements of the neurocranium , specifically the different
bones of the skull vault, the intervening sutures, the outer dermal covering and the ventral
dura mater layer has a species specific distribution of mesodermal vs neural crest cells.
Initially there were contrasting reports from studies done with quail –chick chimeric system
by Couly [Couly et al 1992] and his colleagues vs Noden [Noden 1978] that the full set of
calvarial bones and associated sutures along with the underlying dura were entirely of
cranial neural crest origin vs of paraxial mesoderm origin with contribution of the cranial
neural crest cell being restricted only to the rostral portion of the skull respectively.
[Opperman 2000, Chai & Maxson 2006]. In 2002 it was shown by Jiang et al 2002 [Jiang
et al 2002] that in mouse that the entire fronto-nasal part of the cranium as well as the
frontal bones and the dura mater below the frontal and parietal bones are of neural crest
origin with contribution of this cell type also restricted to the rostral part of the interparietal
28

bone, while the parietal bones and the rest of the posterior elements of the neurocranium
are of mesodermal origin. Among the sutures, the interfrontal (equivalent of the metopic
suture) is entirely of neural crest origin while the coronal, sagittal and lambdoid sutures are
of mesodermal origin. [Jiang et al 2002, Chai & Maxson 2006] It appears thus that there
exists a very clear neural crest-mesoderm boundary among the neurocranium elements in
mammals ( which lies at the junction of the frontal and parietal bones as well as the rostral
boundary of the interparietal bones), careful maintenance of whose location with the help
of specific signaling pathways from the surrounding ectocranial tissue appears to be
significant in maintaining the border between different bony elements of the cranium
during their development period. [Jiang et al 2002, Merrill et al 2006, Ting et al 2009 ,
Chai & Maxson 2006].
1.5 : Calvarial Sutures : Craniosynostosis
The development of the calvarial bones is different from other bones in that instead of a
single periosteum, the calvarial bones develop two, one dorsally or bone and the other
ventrally or endocranially. These periostea are referred to as the ectocranium and
endocranium respectively, the endocranium being often in contact with the dermal layer
beneath. The calvarial bones grow by intramembranous ossification as a developing bony
layer that is continuously growing out radially in the dorsolateral direction from the ventral
ridge, between these two periosteal membranes. Consequentially since the bones grow
along the edges, at any given point of time, a significant zone along the suture borders
contain an actively proliferating population of osteogenic cells. These are major centers of
29

bone growth and active signaling and remain so throughout the period of calvarial
development beginning first around 7 weeks post conception up to puberty. Throughout
the period of calvarial growth as well as for years beyond, the major sutures (coronal,
sagittal, lambdoid) remain patent or unossified. This provides a flexible tissue covering to
accommodate the expanding brain and later allows the growth and regeneration centers at
the bone edges to function effectively. Only the metopic suture fuses early in live (18
months) while the others stay patent until decades later when they ossify completely and
the bones fuse.
Congenital defects in which this balance of bone growth and maintenance of suture tissue
is disrupted result in the premature ossification of sutures and consequent fusion of the
bones, a phenotype clinically referred to as ‘Craniosynostosis’. ‘Cranio’ stands for the
cranial bones, ‘syn’ is from Greek for together and ‘ostosis’ refers to ossification. [Daniel
E. Lieberman 2011(book), Wilkie 1997, Morris-Kay & Wilkie 2005]
Craniosynostosis has an incidence of 1/2500 live births, about 20-30% are syndromic in
which the synostosis occurs as a part of repertoire of other observable craniofacial, limb or
other non-skeletal tissue-specific phenotypes (this includes more than 100 different
annotated syndromes, mostly with an autosomal dominant inheritance pattern and a few
with autosomal recessive or X linked inheritance), while 70-80 % are nonsyndromic, with
synostosis occurring as the only observable defect. Though synostosis is caused by both
environmental factors (predisposition due to unbalanced mechanical forces due to the
growing cephalic tissue underneath) as well as genetic causes, 40-55 % of all cases show
30

fusion of the sagittal suture, making it the most common region of synostosis, followed by
the coronal (20-25%), metopic or inter-frontal (5–15 %) and lambdoid suture synostosis
(0-5%). [Wilkie 1997, Senarath-Yapa et al 2012, D P Rice 2008 (book)].
The immediate and most far-reaching effect of premature suture closure is on the
expanding brain tissue below. The brain tissue expands radially outward from its center
which is thus perpendicular to the orientation of the sutures and the suture has to remain
patent to allow the accommodative expansion of the skull bones to accommodate this
growth .Premature ossification and fusion of the associated bones thus causes a sealing
effect, effectively preventing any possible expansion of the brain tissue directly beneath it.
Bone fusion leads to compensatory expansion of brain tissue in a direction parallel to the
suture, leading to increase in intracranial pressure and resulting asymmetric growth of the
underlying brain tissue. Such increase in intracranial pressure may cause trauma to the
brain as well as craniofacial defects, resulting in a significantly reduced quality of life.
[Daniel E. Lieberman 2011(book), Wilkie 1997, Morris-Kay & Wilkie 2005]
Originally craniosynostosis was believed to be caused solely by mechanical effects i.e.,
uneven pressure application and distribution during the gestation period This view
changed in 1993, when the first study to show a genetic correlation between a specific type
of craniosynostosis syndrome called the Boston type with a mutation in a specific gene
MSX2 was published [Jabs et al 1993]. Msx2 is a member of the homeobox gene family
which encodes transcription factors that bind DNA and play fundamental roles in the
control of cell specification. The identification of the mutation in Msx2 first led to the
31

acceptance of the idea that craniosynostosis could be the manifestation of regulatory defect
associated with an inheritable genetic mutation that influences the molecular mechanisms
at play during the growth and patterning of the calvarial bones and sutures. Since then,
several of the well-known syndromic versions of craniosynostosis disease have been shown
to be caused by genetic mutations in regulatory genes, revealing the role of such genes in
cranial development to be of utmost significance in both development and disease. Such
studies have thus opened up the field of suture development to be a major field of
developmental patterning studies. [Daniel E. Lieberman , 2011(book), Wilkie 1997,
Morris-Kay & Wilkie 2005]
1.6: Calvarial Sutures: Genetics of Craniosynostosis and Signaling in the sutures.
Over the last 20 years, several genetic mutations (inherited or spontaneously arising) have been
shown to cause synostosis of various calvarial sutures. The current list of such genes as of 2014
is given as a compilation in the table below :

32

Table A. Genetics of Craniosynostoses
A. Syndromic Craniosynostosis diseases with known genetic etiology :
Human
Gene
Syndrome Gene function OMIM/Ref
MSX2 Boston Type transcription factor 123101
FGFR
(1-3)
Crouzon
Pfeiffer
Apert
Muenke
FGF signaling 123500
101600
101200
134394
TWIST1,
TCF12*
Seathre-Chotzen

transcription factor 101400
EFNB1 Craniofrontonasal
syndrome
Ephrin-Eph signaling 304110
FBN1 Shprintzen-Goldberg,
Marfan Syndrome
Cytoskeleton 604308

JAGGED1 Alagille Jagged-Notch signaling 118450
TGFBR1,
TGFBR2
Loeys Deitz Syndrome1
Loeys Deitz
Syndrome2
TGF signaling 609192
610168
RAB23
MEGF8
Carpenter Syndrome 1
(CRPT2)
Hedgehog signaling 201000
614976
GLI3 Greig Syndrome Hedgehog signaling 175700
GLI3
IHH
Acrocallosal syndrome Hedgehog signaling 175700
33

ASXL1 Bohring Opitz Putative Polycomb Group
Protein
605039
WDR35 Sensenbrenner WD40 domain-containing
protein-intraflagellar
transport
613610
SKI Shprintzen Goldberg Repressor of SMAD3 182212
MASP1 Michels complement-activating
component
257920
POR
cytoP450
Antley Bixler CytochromeP450
oxidoreductase
201750
GPC3 Simpson–Golabi
Behmel
heparan sulfate
proteoglycans
312870
ESCO2 Roberts acetyltransferases 268300
KRAS Noonan RAS protein ( oncogene) 609942
RECQL4 Baller Gerold DNA helicase 218600
IL11(Il11) Autosomal recessive
Crouzon like
interleukin 11 Nieminen et al
2011
Keupp et al 2013
ERF ets2 repressor factor 600775
MCPH1 CMCB regulator of chromosome
condensation
251200
RUNX2 runt-related transcription
factor 2
Greives 2013
B. Nonsyndromic isolated single suture Craniosynostoses with known genetic
etiology :
Gene Syndrome Gene function Reference
EFNA4 NSC ephrin-eph signaling Merrill et al 2006
34

ALX4 ” homeodomain transcription factor Yagnik 2012
IGF1R growth factor receptor Cunningham et al
2011
IRS1 insulin receptor substrate Stamper et al 2012
FREM1 Isolated
metopic
fras1-related ECM protein 1 Vissers 2011
C. Genes found to be associated with Craniosynostoses from Mouse studies
Nell1
Gdf6
Axin2
Dusp6
Epha4
Pdgfr 

Ptch1(DL)
Protein kinase C-binding protein
TGF beta superfamily
Wnt signaling pathway
interact with MAPK3
Ephrin-Eph signaling
Growth factor receptor

Shh receptor
Zhang et al, 2002
Settle et al, 2003
Yu et al , 2005
Li et al, 2007
Ting et al, 2009
Moenning et al
2009
Feng 2013

The genes above generally belong to signaling pathways, from surface bound ligand–receptor
sets (FGFR, JAGGED1, EFNA4, EFNB1, Epha4, TGFBR1, TGFBR2, Pdgfr alpha) to intracellular
signaling molecules (Nell1, Gdf6, Axin2, Dusp6) and transcription factors (MSX2, TWIST1,
TCF12, RAB23). FBN1 or Fibrillin1 though a microfibril protein involved in maintaining
structural integrity in various types of connective tissue is also known to be able to control the
amount of TGFb signaling in the adjacent tissue by sequestering the TGFb ligand molecules.
Thus taken together, the main signaling pathways that are connected to craniosynostosis
syndromes, are the BMP (Tgfbr1and consequently FBN1, Msx2, Gdf6, partially Twist1 and
Tcf12), RTK (FGFR1-3, EphrinA4, EphrinB1, EphA4, Dusp6, Pdgfr alpha), Notch (Jagged1),
35

Hedgehog (RAB23) and Wnt (Axin2) pathway. Alternatively some of the identified genes could
also be classified under a set of genes involved in osteoblast differentiation [Yen et al 2010,
Wilkie et al 2011]. Not only does their identification aid in the correct clinical assessment of
the disease subtypes but they also are invaluable in understanding the underlying mechanisms
of suture development.
The most significant syndromes among these are discussed below.
1.7: Craniosynostosis : Syndromes and Molecular mechanisms
The main genes or gene groups which have been shown to be associated with the most major
known causes of craniosynostosis (syndromic or non-syndromic) are FGR (1-3), MSX2,
TWIST1, EFNB1 and more recently TGFbR1, TGFbR2, FBN1, RAB23, Jagged1 and GLI3.
I will discuss the first three in detail and the others according to the signaling mechanism they
are a part of. [ Wilkie 1997, Connerney & Spicer 2011 ( book) , Morris-Kay & Wilkie 2005,
Coussens et al 2007, Johnson & Wilkie 2011, Senarath-Yapa et al 2012]
1.7.1: FGFR [Apert, Crouzon, Pfeiffer, Muenke]
Most syndromic cases of craniosynostosis are associated with mutations in the various FGFR
(fibroblast growth factor receptor) genes. The FGF ligand –receptor set comprises of 22 known
FGF ligands and corresponding 5 different receptors (FGFR1-5), of which FGFR1, 2 and 3
have been shown to be involved in craniosynostosis. Fgf receptors belong to the class of RTK
or receptor tyrosine kinases, with each of the receptors composed of 3 extracellular ligand
binding domains, single hydrophobic membrane spanning domain and intracellular tyrosine
kinase domains (except the recently discovered FGFR5, which lacks the intracellular Tyr-
kinase domain). FGFRs are cell surface receptors capable of homo- or hetero-dimerizing. The
36

binding of ligand to the receptor activates the ERK signaling pathway. FGF signaling via the
FGF receptors is involved in various cellular mechanisms, including cell proliferation,
differentiation and migration, as well complex developmental tissue organization processes
such as somitogenesis. [Connerney & Spicer 2011(book)]
Approximately 100 missense, insertion or deletion mutations in FGFR1, FGFR2 and FGFR3
together count for about 20% of all syndromic or nonsyndromic cases of craniosynostosis.
Mutations in FGFR2 alone are responsible for 90% of all syndromic craniosynostoses. The
main syndromes associated with FGFR mutation are Pfeiffer, Apert, Crouzon, Muenke,
Jackson-Weiss, Beare Stevenson of which Muenke syndrome, linked to FGFR3, is the most
commonly occurring form of syndromic craniosynostosis. Apart from that, FGFR1 is strongly
associated with Pfeiffer syndrome while FGFR2 with Apert and Crouzon syndromes, though
in several cases such as Crouzon, Jackson-Weiss etc. more than one FGFR is causal.
Generally most of these syndromes display bicoronal synostosis along with limb abnormalities
and marked facial deformities. Taken with the fact that FGFR mutations have been implicated
in a few non-syndromic isolated cases of craniosynostosis as well, these findings support the
view that there may be a common signaling mechanism involved in suture formation and
maintenance. [ Wilkie 1997, Connerney & Spicer 2011 ( book) , Morris-Kay & Wilkie 2005,
Coussens et al 2007, Johnson & Wilkie 2011, Senarath-Yapa et al 2012]
All FGFR mutations that lead to craniosynostosis are autosomal dominant, gain-of-function
mutations affecting various functional domains of the receptors. Normally two FGF ligand
molecules interact with FGFR dimers to initiate signaling, an interaction which is initially a
weak one stabilized by heparan sulphate proteoglycans (HSPs). All the known mutations lead
to an elevated level of FGF signaling by causing either a strengthened ligand-receptor
37

interaction (increased ligand affinity, decreased specificity, ectopically expressed truncated
receptor forms) or dimerization and activation independent of the ligand. Most of the
functionally affecting mutations cause amino acid substitutions, or insertion or loss of cysteine
residues thus disrupting formation of disulphide bonds which are significant in protein folding.
The unbound free cysteine residues are able to cause intermolecular bonding leading to
receptor dimerization without ligand involvement. On the other hand mutations leading to
alternative splicing may cause ectopic expression of receptor forms leading to activation of
FGF signaling in these cells in the presence of ligands (especially in cells normally not
expressing that particular FGFR type). [ Wilkie 1997, Connerney & Spicer 2011 ( book) ,
Morris-Kay & Wilkie 2005, Coussens et al 2007, Johnson & Wilkie 2011, Senarath-Yapa et
al 2012]
There is also evidence that the same mutation can cause different phenotypes (even different
syndromic features). It is also the case that different mutations can cause the same phenotypes,
suggesting that genetic modifiers may influence FGFR function. Some mutations are also
observed to cause more severe phenotypes than others (Ser252Trp of FGFR2 – around 66%
of Apert syndrome cases). e.g. Pro252Arg (FGFR1), Pro253Arg (FGFR2), Pro250Arg
(FGFR3), one of them (Pro253Arg -FGFR2) is observed to produce a stronger phenotype than
the others. [ Wilkie 1997, Connerney & Spicer 2011 ( book) , Morris-Kay & Wilkie 2005,
Coussens et al 2007, Johnson & Wilkie 2011, Senarath-Yapa et al 2012]
Dusp6 is a dual specificity phosphatase-coding gene involved in the MAPK pathway known
to be a downstream target of FGF signaling. Recently, loss of Dusp6 function in mouse models
has been shown to cause cranial suture synostosis [A. Goldbeter, O. Pourquié. 2008].

38

1.7.2: MSX2 [Boston Type]
MSX2 is a homeobox-containing transcription factor. A missense mutation in Msx2 is closely
associated with bicoronal synostosis observed in a family from the Northeastern US. [Jabs et
al 1993, Warman et al 1993, Muller et al 1993 ]. The syndrome subsequently named Boston
type craniosynostosis is an autosomal dominant disorder. The mutation is Pro148His, which
affects the DNA binding domain of the MSX2 protein, increasing the DNA binding avidity,
thus making it a gain-of-function mutation[Ma et al 1996]. Functionally the Msx2 gene has
been shown to play a significant role in cellular proliferation and differentiation during murine
embryonic development [Hill 1989, Robert 1989- from Ishii 2007], craniofacial and calvarial
bone development [Ishii 2003, 2005, 2007] and to direct anabolism in long bones via the Wnt
pathway. [Cheng et a 2008]. As a signaling molecule that directs osteogenic cell proliferation,
osteoblast differentiation and positively regulates formation of skeletogenic tissue, MSX2 is an
upstream regulator of both Wnt and Hedgehog pathways. Though Boston type
craniosynostosis is extremely rare, recently a second family from Bosnia with several members
affected by this syndrome was reported to carry a second mutation in MSX2 (Pro148Leu).
This mutation was at the same location (Pro148) as the first mutation described in 1993,
making the case that a mutation in MSX2 is the cause of the Boston type syndrome. [Florisson
et al 2013][ Wilkie 1997, Connerney & Spicer 2011 ( book) , Morris-Kay & Wilkie 2005,
Coussens et al 2007, Johnson & Wilkie 2011, Senarath-Yapa et al 2012]
1.7.3: TWIST1 [Saethre-Chotzen]
The most common autosomal dominant form of craniosynostosis, Saethre Chotzen syndrome
(SCS), is caused by a heterozygous loss of function of the TWIST1 gene.[el Ghouzzi et al
1997, Howard et al 1997 ] . Twist1 is a bHLH transcription factor identified in Drosophila as
39

an important factor required for mesoderm induction [Morris-Kay & Wilkie 2005, Thisse et
al 1987] . Haplosufficiency of Twist1 recapitulates the coronal suture synostosis phenotype in
mice (Twist1
+/-
) as well [Merrill et al 2006]. More than 100 different mutations have been
identified in the TWIST1 and shown to cause SCS, thus confirming that it is indeed
haplosufficiency of the gene that is responsible for the syndrome. Twist1 has been shown to
be a positive regulator of epithelial-mesenchymal transitions in several developmental and
cancer settings and seems to be central in connecting several different signaling in suture
formation (BMP, FGF, MSX2, Eph-Ephrin, Runx2)] [ Wilkie 1997, Connerney & Spicer
2011 ( book) , Morris-Kay & Wilkie 2005, Coussens 2007, Johnson & Wilkie 2011, Senarath-
Yapa 2012, Merrill et al 2006, Ting et al 2009]
Besides forming homodimers, Twist1 is also known to heterodimerize with other bHLH
factors. The identity of the factors in such dimer pairs may have an effect on the functional
role of Twist1 in different cell types. [Connerney et al 2005] Recently, mutations in the TCF12
gene, a known bHLH heterodimeric partner of TWIST1 have been reported in individuals
with Saethre-Chotzen syndrome [Sharma et al 2013]. The study encompassed close to 350
individuals with the synostosis phenotype. All of these patients displayed coronal synostosis
as is common in SCS patients with the TWIST1
+/-
genotype. Subsequently in mouse studies
carried out with Tcf12 and Twist1 heterozygous mice, the double heterozygous had a stronger
fusion phenotype than even the known Twist1 heterozygotes , thus supporting the idea of a
dosage dependent Twist1 function loss mechanism in the synostosis of SCS and a synergistic
role of Twist1 and Tcf12 in the process. [Sharma et al 2013]

40

1.7.4: EPHNB1 [Craniofrontonasal syndrome]
The last of the most common craniosynostosis syndromes is Craniofrontonasal Syndrome, the
only known X-linked, dominantly inherited disorder more severely affecting females. It is
caused by a loss-of-function mutations in the EFNB1 gene. EFNB1 encodes a ligand of the
Eph-Ephrin type of receptor –ligand signaling sets. They belong to the special category of trans
signaling membrane bound receptor-ligand pairs, in which the ligand and the receptor are
expressed by membranes of adjacent cells and via their interaction intracellular signaling is
initiated in both the adjacent cells. Largest family of RTKs, Ephrin-Eph signaling, is common
in developing embryos where it has multiple roles, including tissue patterning, the control of
cell proliferation, cell migration, and boundary formation. The Ephrin receptors fall into two
subclasses: A (EphrinA1- A5) anchored to the cell membrane by a GPI
(glycosylphosphatidylinositol)–linkage and B (EphrinB1-B3) consisting of a transmembrane
and a short cytoplasmic domain. The EFNB1 gene mutation in craniofrontonasal syndrome
causes coronal suture synostosis along with other facial deformities. [Connerney & Spicer
2011 ( book) , Morris-Kay & Wilkie 2005, Coussens 2007, Johnson & Wilkie 2011, Senarath-
Yapa 2012 ]
Besides EphrinB1, EFNA4 (EphrinA4) mutations have been detected in several patients with
nonsyndromic craniosynostosis. EFNA4 was for a long time the only known genetic cause
of the nonsyndromic type. [Connerney & Spicer 2011 (book), Morris-Kay & Wilkie 2005,
Coussens 2007, Johnson & Wilkie 2011]

41

1.7.5: FBN1 & TGFB [Loeys Deitz Syndrome, Marfan syndrome]
Mutations in the TGFBR1 and TGFBR2 have been shown to be responsible for the
craniosynostosis in Loeys Deitz Syndrome and similar to that in Marfan syndrome. [ Singh
et al 2006].
The main cause of Marfan syndrome (OMIM# 154700), a connective tissue disorder, is
mutations in the FBN1 (Fibrillin1) gene [Ramirez et al 1993, Wilkie 1997]. Fibrillin1 is a
microfibril protein that makes up the connective tissue of the ECM. Structurally, Fibrillins are
also known to bind and sequester TGF  molecules from binding to TGF  receptors in
surrounding cell membranes and consequently keeping the TGF signal within a certain
threshold. The mutated form of FBN1 leads to misfolding and functional loss of the protein,
preventing sequestration of the TGF molecules and thus excessive TGFR1 and TGFBR2
signaling in surrounding tissues. Subsequently, a portion of Marfan syndrome patients were
also reported to be genetically associated with gain-of-function mutations of the TGFBR1 and
TGFBR2 genes [Singh et al 2006, Akutsu K et al 2007].
Recently Shprintzen-Goldberg syndrome, a rare autosomal dominant disorder considered
another member of the Marfanoid set of syndromes, was found to be caused in several patients
by mutations in the SKI gene, which is an inhibitor of the TGFsignaling. The mutations
render the protein unable to bind to TGFdownstream signaling molecules (the SMAD
proteins) and inhibit their signaling, consequently resulting in abnormally raised TGF
signaling. [Doyle et al 2012, Carmignac et al 2012, Au et al 2014].
I will discuss the remaining genes reported to be associated with craniosynostosis according
to the signaling cascades to which they belong.
42

1.8: Craniosynostosis: Signaling mechanisms and cranial suture development
speculations
Sutures can be visualised as a zone of unossified, fibrous connective tissue flanked by
developing bony tissue. The borders of these two bony tissues are constantly being remodeled
and differentiated into a fresh bony tissue later while the intervening suture tissue region acts
the zone of proliferation, migration and activation of fresh waves of osteoblast population of
cells, yet maintaining their undifferentiated status. In this context very commonly the sutures
are referred to as “growth centers” that retain their apparently undifferentiated status. The
major question I am trying to address here is how the suture is able to achieve this. In particular
I wish to determine how specific gain –of –function and loss-of –function mutations in the
different genes are able to bring about specific suture fusions, and how the different mutated
genes appear to perturb the expression or function of the other genes. Our main understanding
of the prospective roles of the signaling mechanisms in suture formation comes from studies
on ubiquitous and tissue-specific perturbation of specific genes in mice.
1.8.1: FGF signaling in mice
The maximum number of known syndromic cases of craniosynostosis is caused by mutations
in different members of the Fgf signaling family leading to a gain-of-function effect in the
signaling cascade. Mouse studies have shown that increase in expression of Fgf3 and Fgf4,
and gain-of function mutations in Fgf9 all lead to synostosis of the cranial sutures. Mouse
mutants for deletion of exon9 of Fgfr2, display ectopic expression of Fgfr2IIIb and coronal
synostosis . All these indicate that synostosis occurs in case of gain-of-function of different
Fgf signaling components. On the other hand decreased Fgf signaling can also lead to
synostosis of sutures , as observed in Fgfr2IIIc isoform mutants. Thus it is proposed that
43

actually a controlled level of Fgf signaling is required to keep the sutures patent. . [Connerney
& Spicer 2011( Book)]
Fgfr1, Fgfr2 and Fgfr3 are all expressed in the suture zone , though Fgfr2 and Fgfr3 are
expressed in similar zones not overlapping with Fgfr1 expression zone. While Fgfr1 is
associated with differentiating osteoblast cells at the osteoid edge, the other 2 Fgf receptors are
expressed by the highly proliferative osteogenic cells beyond the osteoid zone. Among the
ligands, though all Fgf ligands except Fgf4 and Fgf8 are expressed in the suture, Fgf2 and Fgf9
show the highest expression levels [Connerney & Spicer 2011(Book)]. Craniosynostosis has
been shown to be caused by both enhanced affinity of FGFR (eg.FGFR3 (Pro250Arg) in
Muenke syndrome) and ectopic expression of Fgf ligands ( as in Eks mice with a mutant Fgf9
with higher diffusion capacity ) [Harada et al 2009 ]. But overall, the mechanism of synostosis
brought about due to hyper activation of the Fgf signaling system has been shown to be due
increase in the osteogenic activity in the suture area by either increased proliferation or faster
differentiation of the osteoblast population, or by ectopic differentiation of the suture cells by
direct increase of osteogenic marker Runx2 expression. [Connerney & Spicer 2011(Book), D
P Rice 2008 (book)]
1.8.2: Twist1/MSX2/ Ephrin-Eph/ Jagged-Notch signaling
1.8.2a: Twist1 and Msx2
It was shown early on during suture studies that Msx2 [Jabs et al 1993], Fgfr2 [Iseki et al 1997]
and Tgf β (1−3) [Opperman 1997, Roth 1997], that 3 of the earliest known factors genetically
linked to the largest group of syndromic craniosynostoses, are expressed in the sutures of mice
and rat models. [Connerney & Spicer 2011( Book) .]
44

In 2000, it was shown that Twist1 is expressed in the mouse coronal suture tissue early during
suture development but how these factors interact with each other remained to be understood.[
Johnson et al 2000]. Haplosufficiency of Twist1 (TWIST1
+/-
) causes synostosis of the coronal
suture in Saethre-Chotzen syndrome, while a similar effect is brought about in Boston type
synostosis by a gain-of-function mutation in Msx2 (MSX2) . Thus a reduction in Twist1
signaling and an elevation in Msx2 signaling causes coronal suture fusion. In 2006, Merrill et
al. showed that in heterozygous knockout Twist1
+/-
mice, there is an increase in the expression
of Msx2 transcripts in the coronal suture and adjacent tissue early during suture formation,
which indicates a common pathway affected by the two genes. The coronal suture also very
interestingly presents a tripartite tissue organization system, in which the frontal bone is of
neural crest origin while the parietal bone and the unossified suture tissue cells are of
mesodermal origin. Labeling studies using LacZ to stain the neural crest show a loss of
boundary integrity in both the neural crest and mesodermal cells, which suggests one probable
mechanistic cause for suture loss and fusion of bones. It was further shown that the synostosis
in Twist1
+/-
mice was rescued in Twist1+/-, Msx2+/- double heterozygotes, further
supporting the theory that Msx2 functions downstream of Twist1 in coronal suture
development. [Merrill et al 2006].
Mouse studies done from two different laboratories showed slightly differing evidence in the
case of transgenic manipulation of Msx2 dosage in mice. In one case which achieved a 2-fold
increase in the Msx2 protein expression premature suture fusion did not occur [Liu et al 1995],
whereas in the second case an integration of 13-22 copies of the MSX2 transgene though
caused severe craniofacial defects in the pups leading to premature death (P0), but did not
display a fusion phenotype [Winograd et al 1997]. These contrasting studies actually point
further towards the requirement of a specific dosage of the MSX2 gene and subsequently the
45

affinity of DNA binding by its homeobox domain in normal versus synostosed sutures and in
other craniofacial tissues. [ Wilkie 1997]
1.8.2b: Twist1 and Eph-Ephrin
Merrill and colleagues [Merrill et al 2006] showed that EphrinA2/A4 and EphA4 display a
specific pattern of expression with respect to the coronal suture. Horizontal sections show a
single layer of cells expressing EphrinA2/A4 in the ectocranial zone between the frontal bone
and outer dermal layer, extending anteriorly from the facial area to the site of the coronal
suture. The expression of ephrinA2 and A4 retracts anteriorly in the Twist1
+/-
mice. EphA4
is expressed in a double layer of cells above and below the EphrinA2/A4 layer up to the
coronal suture, whereupon it continues as a single layer posteriorly above the parietal bone.
In the Twist1
+/-
mice, upon the loss of the EphrinA2/A4 close to the suture area, the double
EphA4 lines meet up much more anteriorly and are present as a single layer above the coronal
suture zone. In a further study Ting et al., (2009) showed that EphA4
-/-
mice exhibit a similar
phenotype of suture fusion as well as the mislocalisation of neural crest cells beyond the frontal
bone boundary as well as the mesodermal suture cells beyond the suture boundary. The loss
of the boundary between osteogenic and non-osteogenic (sutural) cells during early suture
development in both Twist1
+/-
and EphA4
-/-
mice suggests that this mechanism is a strong
contender for the process by which bone fusion may occur. Consistent with this hypothesis,
Ting et al (2009) also showed that EphA4
+/-
, Twist1
+/-
have a more severe fusion phenotype.
1.8.2c: Twist1 and Jagged-Notch
A comparative study of coronal synostosis by Ting et al. (2009) showed that the level of
synostosis produced in the coronal suture in EphA4
+/-
mice was less than that in Twist1
+/-
46

mice. In addition, EphA4 expression was controlled of Twist1 expression [Ting et al 2009]
This indicated other downstream players in Twist1’s control of suture formation.
Studies by Yen et al 2010 showed that the Notch2 ligand, Jagged1, is also a downstream
effector of Twist1. JAGGED1 loss in humans is known to cause Alagille syndrome, a multi-
organ syndrome that besides other craniofacial defects, is also known to cause synostosis in a
subset of patients [Alagille et al 1975, Kamath et al 2002].
The study by Yen et al 2010 showed that in mice Jagged1 is expressed in undifferentiated cells
in the central zone of the suture as well as in a specific layer of ectocranial cells immediately
adjacent and dorsal to the calvarial bones (frontal and parietal), and that mesoderm-specific
conditional loss of Jagged1 (Mesp1cre Jagged1
cko/cko
and Dermo1cre Jagged1
cko/cko
) leads to
loss of Jagged1 from the suture zone and fusion of the coronal suture. Interestingly, loss of
Jagged1 from the coronal suture using the conventional heterozygous mutant Jagged1
+/-
mice
or the conditional mesoderm loss versions Mesp1cre Jagged1
cko/cko
and Dermo1cre
Jagged1
cko/cko,
lead to ectopic expression of Notch2 protein in the suture area. Notch2 is a
membrane bound receptor of the Jagged1 ligand [Gridley 2003]. In wildtype animals, Notch2
is detected only along the growing frontal and parietal bone osteogenic fronts, but not in the
sutures. Further, ectopic expression of Notch2 in the sutures was observed in the conventional
heterozygous Twist1 mutant mice (Twist1
+/-
) and Twist1
+/-
; Jagged1
+/-
double heterozygous
mice developed a more severe bicoronal suture synostosis phenotype that of Twist1
+/-
alone.
Jagged1
+/-
adult mice do not display coronal synostosis. This argues that Jagged1 signaling
acts downstream of Twist1 in the maintenance of coronal suture patency [Yen et al 2010].

47

1.8.2d: Twist1 and Fgfr
Several lines of evidence indicate a functional molecular interaction between the Twist1 and
FGFR signaling pathways in the formation and maintenance of cranial sutures especially the
coronal suture. However, the exact form of their functional interaction is not as yet clear.
One of the main pieces of evidence that suggest that the TWIST1 and FGFR may work in the
same pathway is the report that mutations in FGFR2 and FGFR3 are also found in some
patients with the TWIST1 associated Saethre-Chotzen syndrome [Jabs et al 2001]. The gene
twist was discovered in Drosophila as a mesodermal inducer and was also shown to be required
for the expression of fibroblast growth factor, DFR1. [Shishido et al 1993, Casal et al 1996].
But in vitro studies with osteoblasts derived from human patients suffering from Saethre-
Chotzen syndrome with heterozygous loss of the TWIST1 gene, displayed a decrease in
FGFR2 mRNA levels as well as a decrease in the expression of RUNX2 and other markers of
bone differentiation. It was also shown that TWIST1 protein is able to bind to a specific
sequence in the FGFR2 promoter. From these studies it may seem that Twist1 acts upstream
of FGFR signaling, but there also exists evidence of the reverse [Rice et al 2000]. Similarly
RUNX2 is able to bind to the FGFR2 promoter, and Runx2 expression is reduced in cells with
heterozygous loss of TWIST1[Guenou et al 2005]
In mice, Fgfr2 has is expressed in sutures [Iseki et al in 1997]. Fgfr2 mRNA expression coincides
with the area of actively proliferating cells in the suture but remains separate from the layer of
osteopontin (early bone differentiation marker) expression. This is an indication that Fgfr2 is a
marker of proliferative, undifferentiated suture cells [Wilkie 1997] Twist1 is also expressed in
the early developing suture [Rice et al., 2000] but mainly in the non-Fgfr2 expressing mid-
sutural cells bordering with and coinciding with a layer of the Fgfr2-expressing osteoprogenitor
48

cells [Johnson et al 2000]. This observation appears to support the existence of a boundary
between the osteogenic and non-osteogenic population in the suture–bone front border area.
The significance of this boundary is indicated by the phenotype of the Twist1
+/-
mice in which
the boundary is disrupted. Interestingly Twist1
+/-
also shows elevated expression of Fgfr2 in the
suture mesenchyme [Connerney & Spicer 2011(book)] whereas FGF supplied exogenously
upregulates Twist1 expression in mouse tissues. [Rice et al 2000].
In summary, the evidence shows that FGF signaling can act upstream of Twist1, while Twist1
is able to affect FGF signaling as well. But in a cellular context, Twist1 has a negative influence
on osteoblast differentiation, while FGF signaling seems to have both negative and positive
influences on osteoblast differentiation, depending on the specific ligand-receptor type in
action.[Rice et al 2000]
Recently it has been suggested by various studies that the effect of Twist1 on FGF signaling is
dependent on the dimerization status of Twist1. Twist1 is known to form homo- as well as
heterodimers. Connerney and colleagues argued that homodimers have a positive effect on
Fgfr2 expression while heterodimers with other ubiquitously expressed bHLH proteins
negatively affect Fgfr2 expression in the sutures. [Connerney et al 2006]. They propose further
that this bipartite regulation helps in establishment of a low Fgfr2-expressing non-osteogenic
vs high Fgfr2-expressing osteogenic zone at the suture borders. The BMP-downstream
signaling molecule Id1 appears to be involved in maintaining the ratio among the Twist1 dimer
versions. It is also argued here that heterozygous loss of Twist1 ( Twist
+/-
) disbalances this
dimer ratio and thus also leads to upregulation of Fgfr2 and BMP effectors in the suture.

49

1.8.3: BMP signaling
The role of BMP signaling in suture formation and patency continues to be debated, especially
since no form of syndromic or nonsyndromic craniosynostosis has been yet reported to be
associated with the members of the main BMP signaling pathway. BMP signaling is
particularly known to promote osteogenic differentiation. But the effect on suture formation
and calvarial development appears to act more in more than one indirect ways which I discuss
below.
The coronal, sagittal and lambdoid sutures remain patent at birth and for years afterward in
humans as well as in mice, whereas the metopic suture (interfrontal in mice) fuses before birth.
BMP4 has been detected in the suture mesenchyme of both patent as well as fused sutures and
in osteogenic bone fronts early in during calvarial development. But surprisingly, the BMP
inhibitor noggin (whose expression is dependent on high BMP dosage) was detected only in
patent sutures with no expression in the fused posterior frontal suture. This suggests a noggin-
dependent control of the BMP regulation of osteoblasts lining the sutures, and thus suture
patency [Warren et al 2003].
This idea is supported by the finding that ectopic expression of noggin in the sutures lead to
prevention of suture fusion [Wan et al 2008]. Warren et al 2003 showed that FGF2 can disrupt
this noggin dependent control of BMP4 in a dose-dependent fashion. Hence the patency of the
sutures seems to be dependent on the level of FGF2 present in the suture area. Fgf2 is
produced by the underlying dura which may be disrupted during excessive syndromic
expression of FGF ligands, leading to down regulation of noggin and suture fusion.
Interestingly, BMP-soaked beads implanted in the suture do not lead to suture fusion but FGF-
soaked beads do [Kim et al 1998 ].
50

BMP signaling is also known to induce Msx2 expression which is normally expressed in the
sutures. It also induces expression of Id1( bHLH inhibitor ) which is expressed in the
osteogenic fronts. This shows a differential influence of BMP signaling in the two tissues.
[Connerney & Spicer 2011(book)]. Additionally, Sun et al [Sun et al 2013] found that Msx2
regulation by BMP is dependent on an intervention by the transcription factor FoxC1 ( whose
zone of expression lies adjacent to the growing osteogenic fonts, corresponding with the non-
osteogenic zone) , while no such intervention is necessary for Id1 expression. Expression of
FoxC1 is also regulated by FGF signaling [Rice et al 2005]. The maintenance of the distinction
between these two zones (osteogenic and non osteogenic) is thus of significance in suture
patency, as supported by experiments with Twist1+/- mice, in which loss of boundary
between these two zones leads to the fusion phenotype.[ Merrill et al 2006 ]
Further support for this pathway comes from the report that Gdf6
-/-
mice with homozygous
loss of the BMP family ligand molecule Gdf6 (Growth Differentiation factor 6), display
synostosis of the coronal suture [Settle et al 2003]. A follow up study shows that these mice
develop synostosis due to premature ossification in the suture area, but do not show a loss of
boundary integrity. [Clendenning & Mortlock 2012].
1.8.4: HH /Gli3 /RAB23 signaling – Metopic suture
The Hedgehog signaling pathway, which has three known ligands SHH, IHH and DHH (
Sonic, Indian and Desert hedgehog respectively), is extremely important in the early
development and morphogenesis of many tissues in bilaterally symmetrical animals. In
particular it is involved in the patterning of the spinal cord and the establishment the of anterio-
posterior axis in vertebrate limb development. In mouse, transcripts of Shh and its receptor
51

Ptc ( Patched) are expressed in the osteogenic bone fronts of the sagittal and metopic suture
[Kim et al 1998 ].
More recent work has shown that heterozygous mutations in the human GLI3 gene cause
Greig cepalosyndactyly syndrome (GPCS), a disorder with craniofacial and limb defects as
well as infrequent synostosis of the metopic suture [Vortkamp et al 1991, McDoanld-McGinn
et al 2010, Hurst et al 2011]. Gli3 is a transcription factor acting in the Hedgehog signaling
pathway. In the absence of active Hedgehog signaling, Gli3 exists in its repressor form and
prevents transcription of Hh target genes, while in the presence of signaling by Hh ligand
molecules, the Gli3 exists in its non-truncated activator form and promotes active transcription
of Hh target genes [Wang et al 2000] . In mice, the experimental model currently used for
GPCS is the Gli3 extra toe mouse (Gli3-
Xt-J/Xt-J
) which has a 51.5 kb intragenic deletion that
truncates the first zinc finger domain of Gli3, thus reducing its DNA binding specificity [
Maynard et al 2002 ] . Studies with Gli3 mutant mice show the presence of heterotrophic bone
in the metopic suture in mice as well as synostosis in both the metopic and lambdoid sutures
[ Viestinen et al 2012]. In addition, loss of Gli3 leads to an upregulation of Hh signaling in
suture cells, leading to a down regulation of Twist1, upregulation of Runx2 and premature
suture fusion, likely resulting from an increased proliferation of osteoblasts [Rice et al 2010].
Recently, a nonsense mutation of RAB 23 (an ATPase and negative regulator of Hh signaling)
has been associated with Carpenter syndrome, a rare autosomal recessive disorder showing
craniosynostosis. Duplication of Indian hedgehog (IHH), a ligand of the Hh signaling
pathway, has also been reported to cause syndactyly and synostosis in humans. These studies
together show that the previously neglected Hh signaling may also be involved in suture
development and calvarial patterning [ Klopocki 2011]
52

A new development in the study of the role of Hedgehog signaling in craniosynostosis is the
isolation of the Dogface-like (DL) mouse strain, which has mutations in the Hedgehog
receptor Ptch1 gene, leading to abnormally elevated Hh signaling. These mice show synostosis
of the lambdoid suture besides displaying various craniofacial and limb abnormalities similar
to the Basal cell nevus syndrome (BCNS) in humans which is also associated with mutated
PTCH1 gene. [ Feng 2013]
Metopic suture
The metopic suture is the only suture that fuses early during development in humans as well
as in mice. The fact that it is completely of neural crest origin, compared to the mesodermal
origin of the other sutures raises the possibility that the basic molecular identity of this suture
might be different from the others [Jiang et al 2002]. Microarray analysis of gene expression
in all fused, fusing and unfused sutures by Coussens et al in 2007 showed that the gene
expression profile of the developing yet unfused metopic suture resembled that of fused
versions of the other sutures, while all other unfused sutures had gene expression profiles that
correlated amongst themselves to a great extent. [Coussens et al 2007]. These reports reinforce
that idea that development of the interfrontal or metopic suture should be considered
separately from that of the other sutures , though information from the normal development
of the metopic suture may shed light on the cause of the fusion in the mesodermal sutures.
1.9 : Craniosynostosis : Cellular & Tissue induction and interaction
As per the studies discussed above, there appears to be several probable theories being put
forth to explain the cellular mechanism controlling the formation and maintenance of patent
sutures as well those that may cause the fusion of the same. In summary we find two
53

arguments coming forth about the mechanistic cause of the synostosis. These are discussed in
short below.
1.9.1: Cell mixing and Craniosynostosis
Molecular, genetic and in vivo staining studies identify a boundary between osteogenic cells
of the calvarial bones and nonosteogenic cells of the growing sutures, especially in the coronal
suture for which the anterior border of the suture with the frontal bone coincides with the
boundary between mesoderm-derived and neural crest-derived tissue [Jiang et al 2002]. LacZ
staining studies of neural crest and mesodermal cells in the Twist1
+/-
, EphA4
-/-
and Mesp1cre
Jagged1
cko/cko
and Dermo1cre Jagged1
cko/cko
mice show that synostosis in coronal sutures in
these mice is preceded by a crossing of this boundary by the cells of the two different tissues
(neural crest and mesoderm) and significant cell mixing in the suture zone. [Merrill 2006, Ting
2009, Yen 2010]. In Twist1
+/-
and EphA4
-/-
mutant mice , neural crest cells were found to be
present in the suture mesenchyme and adjacent ectocranial layer, while in mesoderm specific
Jagged1 knockout mice (Mesp1cre Jagged1
cko/cko
), mesoderm cells were found to have
infiltrated the neural crest tissue. This loss of boundary between the osteogenic and non
osteogenic cell types has been proposed to be crucial to the development of synostosis since in
Msx2
+/-
;Twist1
+/-
in which the cell mixing phenotype of Twist1
+/-
was not observable , the
synostosis was rescued as well [Merrill 2006, Ting 2009, Yen 2010].
1.9.2: Osteoblast differentiation and Craniosynostosis
A second hypothesis is the specification state of cells in the suture itself, and the possible
abnormal differentiation of the normally non-osteogenic suture cells into osteoblasts due to
ectopic upregulation of osteogenic signals. Several lines of experimentation support this view.
54

For example, the expression of noggin, the BMP inhibitor molecule only in the patent (unfused)
sutures as reported by Warren et al in 2003 reportedly to keep the effect of the BMP expressed
in the sutures below a certain threshold [Warren et al 2003].
In the Saethre-Chotzen model Twist1
+/-
mice, there is an increase in the expression of the
transcription factor Msx2, a well-known positive regulator of craniofacial bone development
and mediator of long bone anabolism, as well as the early osteoblast marker Runx2 in the
suture zone [Liu et al 1999, Satokata et al 2000, Cheng et al 2008, Merrill et al 2006, Ting et
al 2009]. Further, the coronal suture synostosis phenotype of Twist1
+/-
mice is rescued by
reducing the dosage of Msx2 (Twist1
+/-
; Msx2
+/-
mice) [Merrill et al 2006] as well as that of
the early osteoblast differentiation marker Runx2 (Twist1
+/-
; Runx2
+/-
mice) [ Bialek et al
2004]. The study by Bialek et al 2004 also shows that Twist1can functionally inhibit Runx2
via a C-terminal Twist box domain of the Twist1 gene. In addition, Twist-EphA4 single or
combination mutant mice and Dermo1cre-Jagged1
cko/cko
also show ectopic staining of the
BMP effector molecule P-Smad1/5/8 in the suture zone. [Merrill 2006, Ting 2009, Yen 2010]
The master regulator of osteoblast differentiation, Runx2, is also regulated by FGFR signaling.
Mutations in FGFR genes are a leading cause of syndromic craniosynostosis in Apert
syndrome, one of the most common syndromes of craniosynostosis. Consistently, the gain-of-
function mutants FGFR1(P250R) and FGFR2 (C342Y) that developed craniosynostosis
showed enhanced expression of Runx2[ Chen et al 2003, Eswarakumar et al 2004]. Calvarial
osteoblasts isolated from Apert syndrome patients with P253R and S252W mutations in the
FGFR2 gene also showed increased expression of Runx2 as well as increased alkaline
phosphatase activity [ Tanimoto et al 2004, Baroni et al 2005]. These studies appear to indicate
55

an effect on the osteoblast differentiation in FGFR mutants as a possible mechanism of
craniosynostosis.
Additional factors involved in osteoblast differentiation are also associated with
craniosynostosis. Early onset of Runx2 expression under a mesenchyme promoter (Prrx1)
causes early differentiation of osteoblast and synostosis of several sutures, as well as ectopic
bone formation and limb defects[Maeno 2011]. Axin forms a part of the -catenin destruction
complex that deactivates beta catenin in the absence of active Wnt signaling, which plays a
significant role in osteoblast differentiation. Axin2
-/-
mice develop craniosynostosis by a
proposed mechanism of proliferation of calvarial osteoblast [Yu et al 2005]. Mutations in the
homeobox transcription factor ALX4 have been associated with nonsyndromic
craniosynostosis in multiple patients [Yagnik et al 2012].
In mice, overexpression of Nell1, causes synostosis of the sagittal and metopic sutures[ Zhang
et al 2002]. Nell1 encodes a secreted molecule with six EGF domains. It has been shown to
accelerate osteoblast differentiation in cultured osteoblasts.
Together these studies indicate the possible role of ectopic osteogenic activity in the suture
cells ( most arguably causing a change in the specification state of the suture cells from non-
osteogenic to osteogenic ) to be affecting the patency of developing sutures.
1.9.3: Dura mater signaling – Tissue induction theory
The dura mater underlying the calvarial bones may also have a role in the formation and
maintenance of sutures. Early experiments by Opperman et al. 1995 show that in absence of
underlying fetal dura mater tissue calvarial bones form in an overlapping fashion but in later
stages of development, the sutures are obliterated and the bones fuse. [Opperman et al 1995].
56

This indicates that the dura has a permissive role in suture formation and that instructive
signals may arise from the dura. Opperman and colleagues hypothesized further that there
might be two separate signals from the dura, one functioning to initiate the development of
the suture and a second signal that inhibits osteogenesis and functions to stabilize the suture
after the bones are formed. Failure of the osteoinhibitory signal could then lead to suture
closure [Opperman 2000]. The signaling molecules involved in this dura-suture signaling are
proposed to be TGF- 1 and TGF- 2 as well as FGF molecules secreted by the dura.
Opperman and colleagues have suggested that TGF 2 is responsible for bone growth along
the edges of the bones, while TGF- 1and TGF- 3 play a role in induction or stabilization of
the suture [Rawlins & Opperman 2008].
Prefrontal suture is known to fuse earlier in mammals. Opperman and colleagues have
suggested that FGF or TGFarising from the growing suture is able to instruct the underlying
dura to send appropriate levels of osteogenic and non-osteogenic signals that make sutures
patent (coronal and sagittal ) vs fused ( prefrontal) [Opperman 2000]. More studies along these
lines may reveal more about the role of the dura mater in cranial suture development.
2.0: Conclusion : Summarized introduction to the project
Craniofacial sutures present a valuable model to understand the concurrent development of
neighboring tissues to form a 3-dimensional structure. Various signaling pathways play an
active role in the process. The functionality of each is restricted in a spatio-temporal manner
and the cellular processes such as proliferation, migration, differentiation and apoptosis are
also regulated at different levels in a time and space dependent manner. Perhaps because of
the involvement of such complex signaling, the incidence of errors in signaling pathways,
leading to malformation of the sutures, is fairly high. Suture development is thus studied not
57

only to understand the pathways governing the process, but also to understand the etiology of
the genetic disorders and thus develop more effective treatments for patients. There are several
syndromic and a very great number of non-syndromic cases of craniosynostosis. The etiology
and molecular mechanisms of such syndromic and non-syndromic craniosynostosis is poorly
understood. One such syndrome is the Hyper-IgE syndrome (HIES), an autoimmune disorder
in which patients also display synostosis of sutures. Recently it was reported that HIES is
caused by loss-of-function mutations in the STAT3 gene , thus associating this signaling
mechanism with development or synostosis of sutures, the details of which are discussed in
the following chapters.
In my thesis, I present for the first time, a study aiming to elucidate the mechanism by which
a tissue specific loss of the signaling molecule STAT3 may cause coronal suture synostosis in
mice.
58

CHAPTER 1
Neural crest specific conditional loss of Stat3 leads to partial bicoronal suture
synostosis in mice via crest specific loss of Notch2 receptor expression and concurrent
activation of ectopic osteogenic activity in the suture area
INTRODUCTION
Hyper Immunoglobulin E (Hyper IgE) syndrome (HIES) ( OMIM #147060) as the name
suggests is a immunological disorder most characteristically defined by the observation of
elevated levels of IgE in the serum (>2000UI/ml) form early neonatal stage. It is one the
rare primary immunodeficiencies and is almost always accompanied with polycystic
pneumonia and skin abscesses due to Staphylococcus infection, which led Davis et al in
1966 to rename the disease as Job’s syndrome in reference to Job in the Bible who was
‘smote with sore boils ‘from his foot-soles to his crown. Though most cases arise
sporadically, both Autosomal Dominant (AD) and Autosomal Recessive (AR) forms of
HIES have been reported. Interestingly, though the AR form clinically manifests itself in
mainly viral infections and complications of the CNS, the AD form (arguably a much more
multi-symptomatic form) also displays several non-immunological clinical manifestations
in the skeletal, connective and lung tissue among others. [Freeman & Holland, 2008]
Notably, Craniosynostosis (premature fusion of calvarial bones ~ a genetic birth defect
occurring in 1/2000 live births) of coronal, sagittal and lambdoid sutures (of varying
degrees) were reported by more than one clinician to be a part of the clinical manifestations
of AD HIES. [Hoger et al 1985, Smithwick et al 1978]. Further, besides other
59

musculoskeletal and connective tissue defects, certain markedly consistent facial feature
anomalies like asymmetry of the face, broadened nose, raised forehead, high arched palate
and deep set eyes etc. were also found to be characteristic of AD HIES, which collectively
led Borges et al to claim in 1998 -the existence of a ‘recognisable face of Job’s syndrome’
[Freeman & Holland, 2008].
After being treated for more than 5 decades as a disease whose underlying molecular basis
was unexplained, recently a single case of AR-HIES was reported to have resulted from a
homozygous mutation of the TYK2 (Tyrosine kinase2 receptor signaling molecule)
[Minegishi et al 2006]. Two follow up studies to this led to the discovery of spontaneously
arising dominant negative point mutations in the DNA binding domain of the Stat3 gene
(downstream signaling executer of TYK2 and other Jak Tyrosine kinase receptors activated
by several chemokines e.g. IL6, IL10 etc.) in > 50% of unrelated AD HIES patients,
conclusively proving non-functional STAT3 protein to be the major causative agent of this
disease [Minegishi et al 2007, Holland et al 2007].
Stat3 is an intracellular signaling molecule involved in the Jak-Stat signaling cascade
pathway originally identified to be activated by the IFN  (Interferon gamma) and IL-6
(interleukin 6 family members). They involve the Janus tyrosine kinase family of proteins
( JAKs) consisting of JAK1, JAK2, JAK3 AND Tyk2 which are found to be associated
with the intracellular domain of cytokine receptors of mainly IL-6 or gp130 family,
consisting of IL-6, IL-11, OSM (oncostatin M) , LIF ( leukemia inhibitory factor), CT-1
(cardiotropin -1) etc. In presence of cytokine signaling, members of the IL-6 receptor may
60

homo or hetero-dimerize leading to autophosphorylation and activation of the intracellular
JAK domains (at the Tyrosine residue of the YXXQ motif). The STAT proteins on the
other hand are mainly of 7 known types in mammals, STAT 1, 2, 3, 4, 5A, 5B and 6.
Discovered initially as latent cytoplasmic transcription factors, the STAT proteins are
activated by phosphorylation on specific residues on their SH2 domain by the activated
Janus Tyrosine kinases, whereupon they dimerize and are transported to the nucleus to
activate the transcription of several genes involved in cell proliferation, differentiation and
motility in a cell-specific manner [Li 2013].
The Jak-Stat pathway has thus been shown to be actively involved in the proliferation,
differentiation and fate determination of a variety of cell types and are activated by a
several classes of signaling molecules including ligands for G-protein coupled receptors
and several classes of growth factors ( EGF, PDGF etc.), but direct involvement of the
pathway in development of craniofacial tissue has only been indicated by Massimo et al,
who recently showed that Stat3 signaling is actively involved in the proliferation and
specification of early neural crest population in Xenopus embryos [Li 2013, Massimo et
al , 2010].
These combined reports of subtle but consistent malformations in craniofacial structures
and craniosynostosis resulting from a nonfunctional STAT3 protein as well as the
concurrent report of its involvement in proliferation and differentiation of early neural crest
population led us to speculate about the necessity of an active Stat3 signaling cascade early
during development of cranial sutures. In order to investigate this in the mammalian
61

system, we used transgenic mice as models and conditionally knocked down the Stat3
gene individually in the 2 tissues i.e. neural crest (using the Wnt1 Cre system) and
mesoderm ( using the Mesp1cre system ), making up the entire craniofacial tissue to
exclusively investigate the effect of loss of the functional Stat3 signaling in in the 2 cell
types in on the various craniofacial sutures.

62

Results :
1.1 Neural crest specific loss of Stat3 signaling leads to Coronal suture synostosis
It has been consistently reported that patients suffering from HIES (Hyper IgE syndrome)
exhibit craniosynostosis [Hoger et al 1985, Smithwick et al 1978]. Following recent
reports that HIES is caused by a functional loss of Stat3 [Minegishi et al 2007, Holland et
al 2007 ] we wished to investigate the effect of loss of Stat3 signaling on the maintenance
of suture patency. Conventional knockout Stat3 mouse embryos do not survive beyond
E8.5 [ Takeda et al 1997 ]. Consequently we decided to conditionally knockout Stat3
signaling separately in the neural crest and mesodermal population of cells, since these two
population make up the calvarial bones and coronal suture. For this purpose Stat3
flox/flox

allele mice were crossed with either Wnt1cre carrying mice to inactivate Stat3 signaling in
the neural crest-derived neurocranial tissue that includes the entire frontal and nasal bones,
the frontal portion of the occipital bone and with contribution to the sagittal suture [ Jiang
et al 2002 ], or Mesp1cre carrying mice to inactivate Stat3 in mesoderm-derived cranial
tissue including the parietal, temporal and a major portion of the occipital bone as well as
the coronal and lambdoid sutures [Saga et al 1999].

63

Wildtype
Fig1.1A. P21 stage
mice skulls stained
with Alizarin Red
(calcified tissue)
A-E : Dorsal view of
the Alizarin red
stained skulls of
respective genotypes
showing the Frontal
and Parietal bones
juxtaposed to each
other and the coronal
suture demarcating
their boundary.
A’,B’,D’: Magnified
view of the coronal
suture in figures A,B &
D respectively.
pb: parietal bone; fb:
frontal bone; cs:
coronal suture
Arrowheads mark the
coronal suture tissue
Wnt1Cre; Stat3
cko/wt

Mesp1Cre;Stat3
cko/cko

Wildtype
Wnt1Cre;Stat3
cko/cko
Mesp1Cre;Stat3
cko/cko

fb
fb
fb
fb fb
pb
pb pb
pb
pb
pb
cs
cs
cs
cs
cs
Mesp1Cre;Stat3
cko/wt

fb
pb
cs
Wnt1Cre;Stat3
cko/cko

fb
cs
fb
pb
cs
Fig1.1: Neural crest specific conditional loss of Stat3 leads to partial bicoronal
suture synostosis
A
E’ D’ A’
E D
B
C
Table 1 : Craniosynostosis Index [ Neural crest specific loss of Stat3 ]
P21
Table 1 : Craniosynostosis Index
Modified from Oram and Gridley , 2005, CI is a synostosis scoring method whereby the
relative length of each suture synostosed is given a numerical score and the Mean score
is tabulated along with + SD which gives the idea about relative severity of the synostosis
phenotype.
Scoring : 0-unfused, 0.5- >12.5% fused , 1- >25% fused , 2- >50% fused , 3- >75% fused ,
4- >100% fused
64

P21
Fig1.1B. : uCT imaging of P21 stage
mouse skulls
A,B : Dorsal view of the mouse skulls at
stage P21 showing the left frontal and
parietal bone segments with the coronal
suture between them. The arrowhead
points the location of fused regions along
the coronal suture in the mutant. The line
marks the location of the sagittal section
image depicted below.
C,D: Sagittal section images depicting
the relative position of the frontal bone
edge in wildtype (C) and mutant (D)
sample.
Black arrowheads mark location of fusion
in mutant coronal suture zone
White arrowhead marks deformed frontal
bone

Wildtype
Wnt1Cre;Stat3
cko/cko

Wildtype
Wnt1Cre;Stat3
cko/cko

fb fb
fb
fb
pb
pb
pb
pb
C
A
D
B
Fig1.1: Neural crest specific conditional loss of Stat3 leads to partial bicoronal
suture synostosis
65

In order to ascertain the patency of the sutures we looked at postnatal day 21 (3 week) old
mouse skulls stained with Alizarin Red to identify ossified tissue. Only mice with
homozygous conditional knockout of Stat3 in neural crest derived tissues i.e., mice with
the genotype Wnt1cre; Stat3
cko/cko
, display partial bilateral coronal suture synostosis with a
100% penetrance (Fig1a). The other sutures, namely the frontal, squamosal, lambdoidal
and interparietal, appear unaffected.
To quantify the extent of synostosis occurring in the sutures we used a modified version of
the Craniosynostosis Index (CI) measurement, first described by Oram and Gridley [Oram
and Gridely 2005]. The Index, which gives an indication of the relative length of the suture
fused, is calculated using a scoring method attributing a 0 to fully patent suture, followed
by a 1 for fusion of equal to or less than ¼ of the suture, 2 for equal to or less than ½ , 3
for equal to or less than ¾ and a 4 indicating a fully fused one. Modifying it to add a score
of 0.5 to a suture fused equal to or less than 1/8 of its length the average CI for the
homozygous mutants (average from 5 mutants) is 1.00 ( Table 1).
All sutures appear patent in mice with heterozygous conditional loss of Stat3 allele in the
Neural crest tissue i.e. Wnt1cre;Stat3
cko/wt
as well as those in which the Stat3 allele is
conditionally knocked out homozygously or heterozygously in mesenchyme derived tissue
(Mesp1cre;Stat3
cko/cko
and Mesp1cre;Stat3
cko/wt
respectively).
A dorsal view of the uCT analysis of the skull of a P21 homozygous Wnt1cre; Stat3
cko/cko

mutant (Fig1b.) also shows that the coronal suture is fused at several points along its
length. An image of a horizontal section of the same shows the frontal bone front to be
66

growing above the parietal bone and touching the latter at certain marked locations, in
contrast to wildtype cases in which the frontal bone extends below the parietal bone with a
clear non-osteogenic zone of the coronal suture present between the two bones.
I also note from the uCT studies that the frontal bones of the Wnt1cre; Stat3
cko/cko
mutants
are significantly less dense (almost having a perforated appearance) than the
corresponding wildtype samples of the same age.

67

E15.5
E16.5
E12.5
E14.5
Wildtype
Wnt1cre;Stat3
cko/cko

fb
fb
fb
fb
fb
fb
fb
fb
fb
fb
fb
fb
fb
pb
pb
pb
pb
pb
pb
pb
pb
pb
cs
cs
cs
cs
pb
pb
B
A
fb
pb
C D
C ’
D ’
F ’
F
E
’
E
H ’ G ’
H G
Fig1.2A: Whole Mounted
mouse heads stained for
detection of ALP
(developing bone tissue ) at
stages E12.5(A,B),
E14.5(C,C’,D,D’),
E15.5(E,E’,F,F’),
E16.5(G,G’,H,H’) . ALP stains
the developing frontal and
parietal bone fronts in the
samples leaving an un-
ossified (hence unstained) line
of coronal suture forming cells
in between forming a
boundary between the 2 bone
fronts. A loss of defined
Coronal suture boundary is
observable from stage E14.5
onwards.
C’,D’,E’,F’: magnified view of
the suture zone of figures
C,D,E,F.
At E16.5 a dorsal view (G’, H’)
shows the delayed bone
growth in the mutants.
pb: parietal bone, fb: frontal
bone, cs: coronal suture
Arrowheads mark location of
coronal suture tissue
Fig1.2: Neural crest specific conditional loss of Stat3 leads to partial
disruption of coronal suture early during embryonic development
68

1.2A. Deficiency in the neural crest-mesoderm boundary within the coronal suture
during suture genesis in neural crest-specific Stat3-/- mice
In order to ascertain whether any disturbance in the morphology of the suture occurs early
during its development, I looked at whole mount Wnt1cre; Stat3
cko/cko
heads stained with
alkaline phosphatase (ALP) to identify locations of osteogenic tissues at successive stages
of coronal suture development. (Fig2a)
At E12.5 the frontal and parietal bone primordia are beginning to grow both in the wildtype
and homozygous Stat3 mutant; no significant difference was evident in the ALP stained
tissues of either sample (Fig 2a.A,B). In contrast, at E14.5 (Fig 2a.C, D), whole mount
ALP stained samples of wildtype mice clearly display the early osteogenic tissue borders
of the frontal and parietal bones as well as the clear unstained zone of the future coronal
suture between them, while the homozygous mutants fail to display the demarcation
between the frontal and parietal bone zones due to the absence of a clear unstained non-
osteogenic zone. The clear non-osteogenic zone becomes increasingly less distinct with
successive stages of E15.5 and E16.5 in the homozygous Wnt1cre specific Stat3 floxed
mutants (Fig2a.E-H). Taken together, these data suggest that the morphology of the suture
in homozygous neural crest specific Stat3 knockout mice is abnormal early on in the period
of suture development, later manifesting as the synostosis observed at 3 weeks of age.
Further, by E16.5, the overall area covered by the osteogenic tissue in the homozygous
Wnt1cre Stat3 floxed mutants is significantly less than the corresponding wildtype. This
69

indicates that ossification is delayed, which may be cause of the eventual reduction in dense
bony tissue in the adult mice. (Fig 2a.G’,H’).

70

Fig1.2B: ALP stained
Horizontal sections of
E14.5 stage mouse
embryo heads.
A: wildtype sample
B-E: 4 separate mutant
samples displaying the
diverse ways in which the
fusion of bones is observed
in different mutant samples.
E and E’ are sections from
the same sample, with E
being a more dorsal section
and E’ more ventral (closer
to the eye)
pb: parietal bone, fb: frontal
bone, cs : coronal suture
Arrowheads mark the
coronal suture tissue
E14.5
Wildtype
Wnt1cre;Stat3
cko/cko

pb
cs
fb
fb
fb
fb
fb
fb
pb
pb
pb
pb
pb
cs
cs
cs
cs
cs
E
D
C
B
A
E ’
Sample1
Sample2
Sample3
Sample4
Fig1.2: Neural crest specific conditional loss of Stat3 leads to partial disruption
of coronal suture early during embryonic development
71

1.2B Partial and varied forms of fusion of newly formed frontal and parietal bone
fronts in horizontal sections of E14.5 Neural crest specific homozygous inactivation
of Stat3 mutants
ALP staining of whole mounted heads of Wnt1cre; Stat3
cko/cko
samples reveals a disruption
in the formation of a clear unstained non-osteogenic layer of cells fated to give rise to the
coronal suture between the developing frontal and parietal bone front tips. Hence I decided
to look at horizontal sections stained with ALP at this stage to ascertain whether any fusion
of the bones is evident.
At this stage in wildtype embryos, the frontal and parietal bone fronts are visible as two
ALP-stained lines positioned between the outer skin above and the dura mater (the outer
most of the meninges covering the brain ) below (Fig 2b.A). At the point of their junction,
the frontal bone tip extends below the parietal bone tip, which lies above the former. A
clear zone of non-ALP stained cells is present between the two bone tips. This zone marks
the location of the future coronal suture.
ALP stained horizontal sections of several homozygous mutants show intermittent ALP
staining in the suture region at single or multiple points along the suture (Fig2b.B-E). The
suture is not completely obliterated since I do not see a continuous and homogeneous ALP
stain in the suture region joining the frontal and parietal bone tips. Rather, only a few cells
within the unstained suture tissue is stained in any given location, such that there appears
to be a single or multiple individual lines of discrete ALP positive cells forming a
connection between the two bones. Interestingly the phenotype found consistently in all
72

mutants and all sections is a discrepancy in the morphology of the frontal bone tip. The
frontal bone tip does not consistently extend below the parietal bone and the intervening
suture cell region. Rather the bone tip appears to extend in distinctly varied directions in
every mutant, ending in some embryos above (Fig2b.B) and in other embryos directly in
the line of the parietal bone tip (Fig2b.C,D). For the few cases in which the frontal bone
front extends below the parietal bone front, the growing edge of the frontal bone front does
not continue its growth in the ventral direction away from the parietal bone and towards
the dura mater, but makes a turn and grows dorsally towards the parietal bone even in some
cases invading the suture cell region in the process (Fig2b.E, E’).

73

Wnt1Cre;Stat3
wt/wt

Wnt1Cre;Stat3
cko/wt
Wnt1Cre;Stat3
cko/cko
Wnt1Cre;Stat3
wt/wt

Wnt1Cre;Stat3
cko/cko

Fig1.3A: Whole Mounted E14.5 mouse embryo heads stained for LacZ.
A-C : Lateral view of wildtype, heterozygous and homozygous conditional Stat3
knockdown mutants stained with LacZ to identify Wnt1 positive tissue ( of
neural crest origin).
D, E : Dorsal view of the frontal bone front marking the border between the
LacZ stained frontal
D’,E’: magnified view of figures D and F.
e:eye fb: frontal bone
Arrowheads mark location of neural crest-mesoderm boundary
E ’
A
E
D ’
D
C
B
E14.5
Fig1.3: LacZ marking of neural crest cells in Neural crest specific conditional
loss of Stat3 displays minor loss of crest-mesoderm border recognition
e
e
e
e
e
e
e
fb
fb
fb
fb
fb
fb
fb
74

1.3.A : No gross difference in cellular distribution or tissue morphology is observed
in Wnt1cre population of cells in whole mounted samples stained for LacZ
The observation of partial fusion of frontal and parietal bone fronts in ALP stained
horizontal sections of E14.5 homozygous Wnt1cre; Stat3
cko/cko
mutants as well as a
disruption in the non-osteogenic layer of suture cells in the whole mount ALP stained
homozygous mutants, indicate a possible defect in the distribution of the Wnt1cre
population of cells. Such a defect occurs in Twist1
+/-
and EphA4
-/-
mutant mice, both of
which display severe coronal suture synostosis[Merrill et al 2006 , Ting et al 2009 ]. In
order to mark neural crest cells in the homozygous Wnt1cre; Stat3
cko/cko
mutants, we used
the previously generated Wnt1cre;R26RlacZ
ki/ki
[Jiang et al 2000 ] to create Stat3 floxed
mutants with the b-galactosidase gene in the R26R locus i.e., Wnt1cre; Stat3
cko/wt
;
R26RlacZ
ki/ki
and Wnt1cre; Stat3
cko/cko
; R26RlacZ
ki/ki..

Since ALP-stained whole mount homozygous mutants have an obvious disruption in the
clear non-ALP staining (consequently non-osteogenic) cells constituting the suture layer
which separates the frontal and parietal bones, we first looked at the gross morphology and
distribution of LacZ stained tissue in Wnt1cre; Stat3
cko/cko
;R26RlacZ
ki/ki
and Wnt1cre;
Stat3
cko/wt
;R26RlacZ
ki/ki
homozygous and heterozygous mutants. (Fig 3a.A,E).
In E14.5 wildtype samples, as described previosuly [Jiang et al 2000], the entire fronto-
nasal region of the developing skull, including the mandible, the maxilla and the periocular
mesenchyme of the developing eye, are positive for beta galactosidase (Fig 3a.A-C). Of
the calvarial bones, the area covering the developing frontal bones and the dorsal part of
75

the interparietal bone are also LacZ positive, as is the rostral protion of the underlying dura
mater. As shown previously by Jiang et al, the caudal-most margin of the LacZ stained
frontal bone tissue markes the rostral margin and subsequently the location of the future
coronal suture. The beta gal stained portion of the underlying dura mater extends below in
an arched morphology below the parietal bone. In the heterozygous as well as the
homozygous mutants all the above mentioned tissues (fronto-nasal structures, frontal bone
area, anterior portion of interparietal bone and the underlying dura mater ) appear to be
stained with LacZ, such that no gross morphological defect in the tissues stained with LacZ
is evident. Neither is there any significant ectopic staining.
On closer inspection (Fig 3a.D,E). I found that in homozygous mutants, the caudal-most
boundary of the forntal bone population of cells stained darker and appeared significantly
more uneven than the wildtype, indicating a lack of a smooth boundary between the LacZ
stained neural crest and the unstained mesodermal suture tissue. In order to further
ascertain the distribution of the LacZ stained cells in the homozygous and heterozygous
mutants, I further studied horizontal sections of three mutants at several developmental
stages. I discuss the results in the following section.

76

fb
fb
fb
fb
pb
pb
E14.5
pb
pb
c
s
c
s
c
s
c
s
D
’
D
C’
C
fb
fb
pb
pb
c
s
c
s
E
E
’
pb
pb
fb
fb
F’
F
c
s
c
s
E13.5
Wnt1Cre;Stat3
cko/cko

Control
fb
fb
pb
c
s
c
s
pb
LacZ
B
B’
fb
fb
pb
c
s
c
s
pb
A
A’
LacZ ALP
E16.5
ALP ALP LacZ
Fig1.3: LacZ marking of neural crest cells in Neural crest specific conditional
loss of Stat3 displays minor loss of crest-mesoderm border recognition
Fig1.3B: Investigation of localization of LacZ stained cells in Horizontal
sections of mice heads at successive embryonic stages .
A-F : Horizontal sections stained with Wnt1cre tissue stained for LacZ , focusing on
the developing coronal suture region at stages E13.5(A,B), E14.5(C,D), E16.5 (E,F).
A’-F’ : Sections corresponding to A-F stained for ALP to mark the relative locations
of the frontal and parietal bone tips and the position of the coronal suture.
pb: parietal bone, fb: frontal bone, cs : coronal suture
Arrowheads mark individual cells in the coronal suture tissue
77

1.3B The distribution of Wnt1-Cre/R26R-marked cells shows individual cells from
the neural crest population beyond the neural crest boundary and in the mesodermal
zone.
Previous studies carried out in the lab sought to elucidate the molecular mechanism
underlying the development of the coronal suture, as well the abnormal cellular behaviors
that may lead to a loss in the suture tissue and premature fusion of the bones. Approaches
included cell labeling studies using the LacZ transgene expression under the control of
Wnt1Cre for tissues of neural crest origin (frontal bone) and Mesp1 Cre for those of
mesodermal origin (parietal bone and sutre cells ). Previously published data show that in
Twist1
+/-
and EphA4
-/-
mice that are known to develop severe coronal synostosis, there is
significant mismigration of the frontal bone cells (of Wnt1-cre origin) into the developing
coronal suture area. This phenotype may be a prelude to the fusion of the two bones.
[Merrill et al 2006 , Ting et al 2009]. The observation of the partial fusion of the bones in
mutant mice with homozygous loss of Stat3 in the neural crest led me to question whether
similar mismigration of frontal bone cells and corresponding cell mixing is also responsible
for the fusion phenotype in such mutants.
For this purpose I looked at beta gal-stained horizontal sections of heterozygous and
homozygous mutants carrying the LacZ gene, i.e., Wnt1cre;Stat3
cko/cko
;R26RlacZ
ki/ki and

Wnt1cre; Stat3
cko/wt
;R26RlacZ
ki/ki
mice, at several stages of development [E13.5, E14.5,
E16.5 and E17.5]. I assessed these mice for cell mixing. (Fig 3b.A-F). Beginning at E13.5
I observed individual beta gal stained cells beyond the boundary of the Wnt1Cre-marked
78

neural crest tissue and in and around the zone of the developing prospective suture in the
homozygous mutants. (Fig 3b.A, B). This is most prominent at E14.5 (Fig 3b.C, D) and is
discernible clearly up to E16.5 (Fig 3b.E, F). Such mismigration of cells is not observable
in every successive section stained with LacZ , but in about 30-40% of sections per slide
and clustered in a region coinciding closely with corresponding ALP stained sections
displaying a fusion phenotype. I noted that theer appears to be a correlation between the
relative position of the sections with beta gal stained sections with mismigrated cells and
the corresponding location of ALP stained sections with a fusion phenotype [ Supplemental
Data 1-4] . Thus the fusion phenotype coincides physically with boundary-crossing cell
behavior.
From close observation, it appears that there is a correlation between the presence of LacZ
positive cells of neural crest origin in the suture region and an extension of the ALP stain
beyond the frontal bone region into the developing coronal suture region, especially around
E14.5. But beta gal stained cells do not appear to be compose the ectopically stained ALP
stained population in its entirety. At E16.5, I notice a zone of ectopic ALP stain specifically
above the frontal bone tip, which appears to be in apposition with the parietal bone tip,
This positioning may be the cause of the subsequent fusion observed in three week old
pups, described in Fig1. Hence I speculate that mixing of cells of disparate origins occurs
in our homozygous mutants as observed in previously published observations from
Twist1
+/-
and EphA4
-/-
mice. However the number of boundary-crossing cells is small.
Thus it seems unlikely that the fusion of the bones is entirely a result of boundary-crossing
79

cells , but a non autonomous signaling mechanism affecting the suture tissue appear to be
active as well.

80

1.4 Investigation of molecular markers expressed in the developing coronal suture
region early during suture development and projected to play an active role in the
establishment and maintenance of the latter, shows no significant change in
expression of suture cell specific markers (Twist1, Jagged1) or neural crest-
mesoderm boundary establishment marker (Ephrin/Eph), but a reduction in the
expression of the membrane bound receptor Notch2 specifically in the growing
frontal bone front.
2 of the most well known syndromic forms of craniosynostosis , Saethre-Chotzen
syndrome and Craniofrontonasal syndrome are caused by mutations in the TWIST1 and
EPHNB1 , whereas EPHNA4 mutation has been long associated with non-syndromic
synostosis [ Morris Kay & Wilkie 2005, Wilkie 1997]. This gives a strong indication of
the apparent involvement of these signaling mechanisms, ie Twist1 (bHLH transcription
factor) and Eph-Ephrin (cell surface bound membrane receptor-ligand complex) in the
normal developemnt of the coronal suture.
Previous studies in the lab with mouse models for Twist heterozygous knockout (Twist1
+/-
) and EphA4 homozygous knockout ( EphA4
-/-
) mice have shown that EphrinA2 and
EphrinA4 are exptessed in a single layer of cells in the ectocranial layer (ECL) above the
frontal bone primordium beginning at the anterior most end of the head and continuing
upto the position of the coronal sutre , while the corresponding receptor EphA4 is expressed
in 2 layers above and below the Ephrin layer anterior to the coronal suture which later join
and become a single layer in the ectocranial zone posterior the coronal suture and above
81

the parietal bone [Merrill et al 2006]. In Twist1
+/-
mice, there is marked reduction in the
expression of Ephrin in the ECL, with the double EphA4 layers merging together [Merrill
et al 2006]. Further, it was also shown that EphA4 signaling acts downstream of Twist1
since Twist1
+/-
; EphA4
+/-
mice were shown to develop a stronger synostosis phenotype as
well as with higher penetrance than Twist1
+/-
(which has higher severity and penetrance
of synostosis than EphA4
-/-
mice) [Ting et al 2009]. Mechanistically it was shown that in
both Twist1
+/-
and

EphA4
-/-
neural crest cells

labelled with LacZ (Wnt1cre;R26RLacZ)
show a disregard for the neural crest- mesoderm (frontal bone –coronal suture tissue )
boundary , with neural crest cells from the frontal bone escaping into the coronal suture
area as well as even into the parietal bone. These observations led to the currently proposed
hypothesis that loss of Twist1 or Eph-Ephrin signaling ledas to a disruption of boundary
maintenance between these 2 tissues causing the eventual synostosis of the frontal and
parietal bones in these mutant mice. Since I observe limited but definite number of LacZ
stained cells in the Wnt1cre;Stat3
cko/cko
;R26RLacZ , mutant samples early during coronal
suture development (Fig1.3B) I wanted to investigate whether this defect in neural crest-
mesoderm boundary was a result of similar reduction in Eph-Ephrin and Twist1 expression
marking the location of the coronal suture.

82

EphrinA2
E15.5
EphA4
Control
fb fb
cs cs
cs
cs
cs
cs
pb
pb
Wnt1Cre;Stat3
cko/cko

D’
D
C’
C
B
A
Fig1.4A: Investigation of Ephrin/Eph expression in Horizontal sections of mice
heads during early embryonic stages of coronal suture formation
A,B, C,D : Horizontal sections of Control and Wnt1Cre;Stat3
cko/cko
E15.5 mouse
head sections immunostained for detection of EphrinA2 (A,B) and EphA4 (C,D)
respectively , in the ECL ( Ectocranial layer).
C’,D’ : Higher magnification views of annotated areas of C and D
pb: parietal bone, fb: frontal bone, cs : coronal suture
Arrowheads mark comparative staining among 2 tissues

Fig1.4A: Investigation of Ephrin/Eph in the developing coronal suture area
affecting production and maintenance of the crest-mesoderm boundary in mice
with Neural crest specific conditional loss of Stat3 shows no specific
difference in expression pattern
83

1.4.A. Investigation of Eph-Ephrin expression in the ectocranial layer of developing
coronal suture region during early suture development
In order to investigate for any changes in Eph-Ephrin expression in the ectocranial layer
(ECL) , horizontal sections of Control and Wnt1cre;Stat3
cko/cko
mice heads were made via
cryosectioning post sucrose gradient and OCT embedding at embryonic day 15.5 (E15.5)
stage and immunohistochemistry was performed for detection of EphrinA2 and EphA4 in
the ECL. Sections adjacent to the immunostained slides were stained with the oseoblast
marker ALP to detect the location of the coronal suture and the morphology of the frontal
and parietal bone fronts realtive to the suture ( data not shown).
As shown, in the Control samples a single row of EphrinA2 staining cells are observable
in the ECL extending from the anterior end upto the location of the coronal suture , while
a double layer expressing EphA4 is also detected meeting at above the location of the
coronal suture and then continuing posteriorly above the bone as a single layer (Fig1.4A.
A). Interestingly, no significant difference was observed in the expression of Epha4-
EphrinA2 in the neural crest specific homozygous Stat3 mutant sections
[Wnt1cre;Stat3
cko/cko
]. (Fig1.4A. B)

84

Twist1 ALP
Control
fb
fb
cs cs
c
s
cs
pb
pb
Wnt1Cre;Stat3
cko/cko

E13.5
B’ A’
A B
Fig1.4B: Investigation of Twist1 expression in Horizontal sections of mice
heads during early embryonic stages of coronal suture formation
A,B : Horizontal sections of Control and Wnt1Cre;Stat3
cko/cko
E13.5 mouse head
sections immunostained for detection of Twist1 in the coronal suture area .
A”, B”: Higher magnification of annotated area from A and B
A’,B’, A”’ and B””: Sections corresponding to A,B , A” and B’ respectively , stained
for ALP to mark the relative locations of the frontal and parietal bone tips and the
position of the coronal suture.
pb: parietal bone, fb: frontal bone, cs : coronal suture
Arrowheads mark comparative staining among 2 tissues
Fig1.4B: Investigation of Twist1 in the developing coronal suture area affecting
production and maintenance of the crest-mesoderm boundary in mice with
Neural crest specific conditional loss of Stat3 shows no specific difference in
expression pattern
fb
pb
fb
ALP
cs
A”
fb
pb
cs
B”
fb
pb
Twist1
fb fb
A”
’
B” ’
cs
cs
pb
pb
85

1.4.B. Investigation of Twist1 expression in the developing coronal suture region
during early coronal suture development
In order to study the relative expression pattern of Twist1 protein , horizontal sections of
E13.5 Control and Wnt1cre;Stat3
cko/cko
mice heads were made via cryosectioning post
sucrose gradient and OCT embedding and immunohistochemistry was performed for
detection of Twist1 in the suture and adjacent frontal and parietal bones. Sectioning was
carried out in sets of multiple slides and sections adjacent to the immunostained slides were
stained with the early oseoblast marker ALP to detect the location of the coronal suture and
the morphology of the frontal and parietal bone fronts realtive to the suture.
As shown, in Control samples strong Twist1 expression is observable in the coronal suture
area , while the adjacent frontal and parietal bones too show presence of Twist1 protein
expression (Fig1.4B. A). This is in accordance with previously published data that Twist1
has a role in the development of the osteogenic tissue of the frontal bone as well [ Ishii et
al 2003]. On investigation of the neural crest specific homozygous Stat3 mutant sections
[Wnt1cre;Stat3
cko/cko
] it was observed that there was no significant reduction in the
expression of Twist1 protein in the coronal suture area. (Fig1.4B. B) On closer inspection
a slight reduction in the expression of Twist1 in the frontal bone is detectable (Fig1.4B.
B”).

86

fb
cs
pb
cs
cs
cs
fb
cs
pb
cs
Jagged 1
ALP
Control
Wnt1Cre;Stat3
cko/cko

pb
pb
pb
cs
cs
cs
fb
fb
fb
ALP
pb cs fb
Jagged 1
E13.5 E14.5
D
’
D
C
’
C
B
B’
A
A’
A’’
B ’’
Fig 1.4C: Investigation of Jagged1 expression in Horizontal sections of
mice heads during early embryonic stages of coronal suture formation
A,B ,C,D: Horizontal sections of Control and Wnt1Cre;Stat3
cko/cko
E13.5 [A,B]
and E14.5 [C,D] mouse head sections immunostained for detection of Jagged1
in the coronal suture
A’,B’ : Higher magnification of annotated area from A and B.
A”,B”,C’,D’: Sections corresponding to A,B stained for ALP to mark the relative
locations of the frontal and parietal bone tips and the position of the coronal
suture.
pb: parietal bone, fb: frontal bone, cs : coronal suture
Arrowheads mark comparative staining among 2 tissues
Fig1.4C: Investigation of different markers (Jagged1) in the developing
coronal suture area affecting production and maintenance of the crest-
mesoderm boundary in mice with Neural crest specific conditional loss of
Stat3shows no loss in expression in the suture region
87

1.4.C. Investigation of Jagged1 expression in the developing coronal suture region
during early coronal suture development
Heterozygous loss of JAGGED1 (membrane bound ligand of Notch2 receptor) leads to the
development of Alagille syndrome in human patients who display coronal suture
synostosis as one of the identifiable phenotypic features of the syndrome. [Alagille et al
1987, Kamath et al 2004] . Recent studies in the lab have shown that Jagged1 protein is
expressed in the developing coronal suture region of mice as well as in a layer above
adjacent to the frontal and parietal bones dorsal to them and in a single layer ventral to the
parietal bone. All these areas are made up of cells of mesodermal origin and consistently
transgenic mice with mesoderm specific loss of Jagged1 (Dermo1cre;Jagged1
cko/cko
)
expression display a reduction in Jagged1 expression in the above mentioned locations and
a correlated loss of coronal suture and fusion of the frontal and parietal bones leading to
significant coronal suture synostosis [Yen et al 2010]. On the other hand neural crest
specific knockdown of Jagged1(Wnt1cre;Jagged1
cko/cko
) did not affect the expression
pattern of Jagged1 , thus proving that Jagged1 is not expressed in neural crest derived tissue
in the neural crest-mesoderm boundary region. Further, it was also shown that in
developing coronal suture of mice Twist1 and Jagged1 have overlapping zones of
expression in the coronal suture and mice with heterozygous loss of Twist1 (Twist1
+/-
)
show reduced expression of Jagged1 in the coronal suture area by the stage E14.5, though
Jagged1 gene dosage has no effect on Twist1 expression. Finally, a combined heterozygous
loss of Twist1 and Jagged1 gene ( Twist1
+/-
;Jagged1
+/-
) leads to a greater severity of
coronal suture synostosis than Twist1
+/-
alone , the

latter having a strnger synostosis than
88

in Jagged1
+/-
mice (synostosis severity :

Twist1
+/-
;Jagged1
+/-
> Twist1
+/-
> Jagged1
+/-
)
All these results indicate a possible genetic interaction between Twist1 and Jagged1 in
normal development of the coronal suture tissue cells and that Jagged1 acts downstream of
Twist1 in the process. Besides , mesoderm specific loss of Jagged1
(Mesp1cre;Jagged1
cko/cko
) leads to a defect in neural crest- mesoderm boundary
maintenance and cell mixing in the suture and growing bone fronts, detected by labeling
mesoderom cells with LacZ (Mesp1cre;Jagged1
cko/cko
;

R26R LacZ) [Yen et al 2010].

Hence, I decided to look for any possible changes in Jagged1 expression which may be the
cause of the cell mixing and bone fusion observable in our Wnt1cre;Stat3
cko/cko
mutants.
For this purpose immunohistochemistry was performed on horizontal sections of Control
and Wnt1cre;Stat3
cko/cko
mice heads from stages embryonic day 13.5 and 14.5 (E13.5 and
E14.5) for detection of Jagged1 in the suture and ECL. Sectioning was carried out in sets
of multiple slides and sections adjacent to the immunostained slides were stained with the
early oseoblast marker ALP to detect the location of the coronal suture and the morphology
of the frontal and parietal bone fronts realtive to the suture.
As shown, in the Control samples a single row of Jagged1 staining cells can be observed
above the Frontal and Parietal bones while the coronal suture also stains deeply for Jagged1
along with a layer of Jagged1 staining cells below the parietal bone. (Fig1.4C. A, C). In
the mutant ( neural crest specific loss of Stat3 signaling ) no significant loss of Jagged1
staining is observable in the coronal suture area, though there might be a slight reduction
in the staining in the ECL layer above the bones. (Fig1.4C: B, D). It can be noted that at
89

E14.5 in the ALP staining of the the section adjacent to the Jagged1 immostained section
, fusion of the frontal and parietal bones is clearly observable (Fig1.4C:D’) , but as noted
from the adjacent Jagged1 immostained section Jagged1 staining in maintained in the
suture cells despite the clear fusion phenotype.

90

Notch2
ALP
Control
Wnt1Cre;Stat3
cko/cko

ALP
Notch2
B
B’
A
A’
A’’
B’’
Fig1.4D: Investigation of Notch2 expression in Horizontal sections of mice
heads during early embryonic stages of coronal suture formation
A,B: Horizontal sections of Control and Wnt1Cre;Stat3
cko/cko
E14.5 mouse head
sections immunostained for detection of Notch2 in the coronal suture area .
A”,B” : Greater magnification of annotated area from A and B.
A’,B’,A”’, B”’ : Sections corresponding to A,B,A” and B” stained for ALP to mark the
relative locations of the frontal and parietal bone tips and the position of the coronal
suture.
pb: parietal bone, fb: frontal bone, cs : coronal suture
Arrowheads mark comparative staining among 2 tissues
A’’
’
B’’
’
fb
pb
cs
fb
fb
fb
fb
fb
fb
fb
cs
cs
cs
cs
cs
cs
cs
pb
pb
pb
pb
pb
pb
pb
E13.5
Fig1.4D: Investigation of different markers (Notch2) in the developing coronal
suture area affecting production and maintenance of the crest-mesoderm
boundary in mice with Neural crest specific conditional loss of Stat3 shows
loss of expression the frontal bone area
91

1.4.D. Investigation of Notch2 expression in the developing coronal suture region and
growing bone fronts during early coronal suture development
Jagged1 protein is a transmembrane ligand of the type 2 Notch receptor ( Notch2 ) known
to be involved in several developmental and cell specification processes [ Bolos et al 2007,
Lai et al 2004] . The previously published work with mesoderm specific loss of Jagged1
showed that in E14.5 wildtype mice Notch2 is expressed at the growing frontal and parietal
bone fronts and is absent from the coronal suture area. In transgenic mice with mesoderm
specific loss of Jagged1 (Dermo1cre;Jagged1
cko/cko
and

Mesp1cre;Jagged1
cko/cko
)
expression of Notch2 is extended into the coronal suture area. Similar expansion of Notch2
protein expression in to coronal suture area is also observable in Twist1
+/-
as well as in
Twist1
+/-
;Jagged1
+ -
double hetreozygous mice. These observations indicated that Twist1
is reponsible for excluding the expression of Notch2 from the coronal suture area and that
an increase in ectopic Notch2 expression in the suture may be involved in raising
ossification signals in the suture tissue causing misdifferentiation of the suture tissue into
bony tissue thus causing fusion of the bone fronts and consequent synostosis [Yen et al
2010].
Hence, I decided to look for any possible changes in Notch2 expression which may be the
cause of the cell mixing and bone fusion observable in our Wnt1cre;Stat3
cko/cko
mutants.
For this purpose immunohistochemistry was performed on horizontal sections of Control
and Wnt1cre;Stat3
cko/cko
mice heads from stages embryonic day 13.5 (E13.5) for detection
of Notch2 in the developing bone fronts. Sections adjacent to the immunostained slides
92

were stained with the early oseoblast marker ALP to detect the location of the coronal
suture and the morphology of the frontal and parietal bone fronts realtive to the suture.
As published before , in the Control samples Notch2 staining was detected to be
overlapping with the developing frontal and parietal bone fronts and no staining in the
suture area. (Fig1.4D. A, A”).
Interestingly, in the mutant ( neural crest specific loss of Stat3 signaling ) a significant loss
of Notch2 protein expression was noted only in the frontal bone front while the expression
in the parietal bone region is maintained. In contrast, no ectopic expression of Notch2 was
observable , in the suture area or otherwise. (Fig1.4D. B, B”).
93

fb
pb
cs
cs
Runx2 ALP
Control
Wnt1Cre;Stat3
cko/cko

ALP
Runx2
pb
pb
pb
pb
pb
pb
pb
fb
fb
fb
fb
fb
fb
fb
cs
cs
cs
cs
cs
cs
Fig 1.5: Investigation of Runx2 expression in Horizontal sections of mice heads
during early embryonic stages of coronal suture formation
A,B: Horizontal sections of Control and Wnt1Cre;Stat3
cko/cko
E13.5 mouse head
sections immunostained for detection of Runx2 in the coronal suture area .
A”, B” : Higher magnification of annotated area from A and B
A’,B’, A”’, B”’: Sections corresponding to A,B, A” and B” respectively, stained for ALP
to mark the relative locations of the frontal and parietal bone tips and the position of
the coronal suture.
pb: parietal bone, fb: frontal bone, cs : coronal suture
Arrowheads mark comparative staining among 2 tissues

B
B’
A
A’
A’’
B’’
A’’’
B’’
’
E13.5
Fig1.5: Investigation of early osteoblast marker Runx2 expression in the
developing coronal suture area in mice with Neural crest specific conditional
loss of Stat3 shows ectopic expression of Runx2 in the suture area
94

1.5. Investigation of expression of early osteogenic marker Runx2 in the developing
coronal suture region early during suture development, shows ectopic expression in
the suture region in neural crest specific homozygous loss of Stat3.
As discussed in Fig 1.2B, from ALP staining studies of horizontal sections of different
samples of Wnt1cre;Stat3
cko/cko
homozygous mutants it appears that the fusion phenotype
is different in every individual case, but from LacZ staining of neural crest cells in
homozygous mutants (Wnt1cre;Stat3
cko/cko
;R26RLacZ ) as described in Fig1.3B though I
observed a limited number of individual neural crest cells disregarding the neural crest-
mesoderm boundary and infiltrating the suture area, they do not seem to be able to account
for the total number of ALP positive cells bridging the frontal and parietal bone tips. Hence
I wanted to investigate the presence of any ectopic or increased osteogenic activity other
than in the bone fronts, in the developing coronal suture region in Wnt1cre;Stat3
cko/cko
mice
samples.
For this purpose, immunohistochemistry was performed on horizontal sections of Control
and Wnt1cre;Stat3
cko/cko
mice heads from stages embryonic day 13.5 (E13.5) before the
actual fusion is observable for detection of the early osteogenic marker Runx2 in the
developing bone fronts and intervening suture area. Sections adjacent to the
immunostained slides were stained with the oseoblast marker ALP to detect the location of
the coronal suture and the morphology of the frontal and parietal bone fronts realtive to the
suture.
95

In the Control samples Runx2 staining is observed in a single layer of cells lining the frontal
and parietal bone fronts and no staining in the suture area. (Fig1.5A. A, A”).
Significantly, in the neural crest specific loss of Stat3 mutants (Wnt1cre;Stat3
cko/cko
) a
significant ectopic expression of Runx2 protein is noted in the developing suture area most
predictably in the non osteogenic suture cells. (Fig1.5A. B, B”). The corresponding ALP
stained slides also show initiation of abnormal ALP staining in the suture region towards
the dorsal side.

96

PSmad1/5/8
Control
Wnt1Cre;Stat3
cko/cko

Fig 1.6: Investigation of PSmad 1/5/8 expression in Horizontal sections of
mice heads during early embryonic stages of coronal suture formation
A,B : Horizontal sections of Control and Wnt1Cre;Stat3
cko/cko
E14.5 mouse
head sections immunostained for detection of PSmad1/5/8 in the coronal
suture area .
A’,B’ : Higher magnification of annotated area from A and B
pb: parietal bone, fb: frontal bone, cs : coronal suture
Arrowheads mark comparative staining in the coronal suture among the 2
genotypes
Fig1.6: Investigation of PSmad expression in the developing coronal suture
area in mice with Neural crest specific conditional loss of Stat3
E14.5
A
B
’
A
’
B
pb
fb
cs pb
pb
pb fb
fb
fb
cs
cs
cs
PSmad1/5/8
PSmad1/5/8
PSmad1/5/8
97

1.6. Investigation of expression of PSmad1/5/8 (BMP signaling marker) in the
developing coronal suture region early during suture development, shows no ectopic
expression in the suture region in neural crest specific homozygous loss of Stat3.
Previously published studies with mesoderm specific loss of Jagged1
(Dermo1cre;Jagged1
cko/cko
) showed that suture specific loss of Jagged1 expression also
leads to a significant increase in BMP signaling in the coronal suture tissue , as observable
by increased PSmad1/5/8 expression detected by immunofluorescence. [Yen et al 2010]
In order to test any similar changes in BMP signaling in the Wnt1cre;Stat3
cko/cko
mutants ,
immunofluorescence studies were performed on horizontal sections of Control and
Wnt1cre;Stat3
cko/cko
mice heads at the stage embryonic day 14.5 (E14.5) for detection of
activated BMP signaling molecule PSmad1/5/8 in the developing bone fronts and
intervening suture area. Sections adjacent to the immunostained slides were stained with
the oseoblast marker ALP to detect the location of the coronal suture and the morphology
of the frontal and parietal bone fronts realtive to the suture ( data not shown).
As published before , in the Control samples PSmad1/5/8 staining is observed to be
overlapping with the developing frontal and parietal bone fronts and no staining is noted
in the suture area. (Fig1.4E: A,A’). No significant difference was observed in the
expression of PSmad1/5/8 in the neural crest specific homozygous Stat3 mutant sections
[Wnt1cre;Stat3
cko/cko
]. (Fig1.4E: B,B’), especially with no staining notable in the suture
area.

98

Discussion
Neural crest specific inactivation of Stat3 causes coronal suture synostosis.
We show here for the first time that neural crest-specific and not mesoderm-specific
inactivation of Stat3 leads to incomplete bilateral coronal suture synostosis. (Fig1.1A)
Recently [ Minegishi et al 2007] showed that the immunologic disorder Hyper-IgE
syndrome (HIES) which has craniosynostosis as one of its clinical manifestations, is caused
by a mutation in the DNA binding domain of Stat3, causing functional loss of the protein.
This was the first time the Jak-Stat signaling pathway was hinted to be involved in the
development of cranial sutures. Notably, our mouse studies showed that Stat3 is required
only for the development of the neural crest-derived frontal bone and not in the mesoderm-
derived parietal bone or coronal suture tissue itself. Significantly, this is the only case in
which neural crest-specific inactivation of a gene has led to synostosis of the coronal
suture. In all cases to date, genes involved in synostosis function in mesoderm or in
mesoderm and neural crest together. Previously Wnt1cre was used to inactivate the neural
crest-specification gene, AP2  This led to neural tube closure defects and excencephaly
as well craniofacial defects such as shortening of the snout, cleft palate, middle ear ossicles
[Brewer et al 2004]. Wnt1cre mediated knockout of the Wnt intracellular signaling
molecule -catenin or the Hedgehog signaling peptide, Smoothened (Smo), both led to
severe craniofacial tissue loss but not to synsostosis [Brault et al 2001, Jeong et al 2004].
I also show that the development of the synostosis occurs only in the homozygous loss of
Stat3 which indicates a dose-dependent effect of the gene in the synostosis mechanism
99

(Fig1.1A) and also that the loss of suture tissue takes place early in coronal suture
development (Fig1.2A) which suggests that Stat3 is active in the early stages rather than
the later stages of suture establishment.
The coronal synostosis caused by neural crest-specific inactivation of Stat3 is different
phenotypically from other cases of synostosis.
It is very important to note that the form of synostosis we observe in our neural crest-
specific homozygous Stat3 knockdown mutants (Wnt1cre; Stat3
cko/cko
) is phenotypically
distinct from the type observed in other well-known and regularly studied mutants. For
example, in the previously studied constitutive Twist1
+/-
and EphA4
-/-
mice or the
conditional Dermo1cre;Jagged1
cko/cko
and

Mesp1cre;Jagged1
cko/cko
mutants, even as early
as E14.5 bone fusion is complete, with the frontal and parietal bones fused to each otheer
and the suture completely obliterated. In contrast, from the final morphology of the fusion
detected in Alizarin red-stained skulls and uCT studies of 3 week old mice (Fig1.1, Fig1.2)
, as well as the initial defect in morphology observed early during suture development (
E14.5) , it becomes clear that the fusion observed has several typical characteristics. These
are,
i) In wildtype E14.5 horizontal sections, a normal suture is characterized by the
parietal bone tip extending above the suture zone and the frontal bone extending
below the same, an arrangement that is maintained in the ossified adult bones
too. But in the homozygous neural crest-specific Stat3 knockout mutants
(Wnt1cre; Stat3
cko/cko
) the frontal bone tip appears to be positioned randomly
100

above, below or at the same level as the parietal bone tip in individual mutant
samples from different experimental sets. In several cases, the frontal bone tip
appears to be in direct apposition to the parietal bone , rather than below it as
occurs in wildtype cases.
ii) Beside the defect in directional specificity there also appears to be a defect in
the integrity of the bone front morphology as well. As it becomes clear from
the ALP studies of horizontal sections at E14.5 stage (Fig1.1, Fig1.2B), the
frontal tip does not show an obvious tapering morphology similar to the parietal
bone tip. Rather the shape differs from a sharper but more extended version
(Fig1.2B. B,D) to a broader end with undefined morphology (Fig1.2B. C,E’).
iii) Rather than a clear fusion of the 2 bones, I observe individual lines (single or
multiple) of ALP staining cells bridging the 2 bone tips in E14.5 sections
(Fig1.2B.B-E’), as well as later stages, as observed from E16.5 sections
(Fig1.3B: F’) and uCT studies at P21 stage (Fig1.1B. D)
iv) Fusion occurs at disjointed individual points along the length of the whole
suture rather than entire sections of the suture being fused (Fig1.1A.D, D’;
Fig1.1B.B) and the phenotype at individual points (visualized in the successive
sections >10um apart) may appear varied even within a single suture sample
(Fig1.2B.E, E’)
All these observations indicate that the actual mechanism for bone fusion and synostosis
in this case is different than the mechanisms known so far and that the fusions observed
101

are probably the collective result of multiple influences of Stat3 inactivation in neural crest
on surrounding tissue.
LacZ marking of neural crest cells in neural crest-specific Stat3 mutant mice indicates
a loss of neural crest-mesoderm border integrity
As stated before, one of the best demonstrated mechanisms for fusion of frontal and parietal
bones is loss of a clear boundary between neural crest (frontal bone) and mesoderm (suture
proper) and the intermixing of the two population of cells leading to ectopic osteogenesis
in the suture and thus to bone fusion [Merrill et al 2006, Ting et al 2009, Yen et al 2010].
Significantly, on labeling the neural crest population of cells in our homozygous mutants
with the R26RLacZ transgene expressed under the Wnt1Cre, it becomes clear that starting
from E14.5 onwards, individual neural crest cells are present in the suture area. Thus
inactivation of Stat3 in the neural crest leads to the ectopic location of a portion of the
neural crest cell population.
Stat3 is activated by several cytokine factors, including IL-6, PDGF, and G-CSF. Stat3 is
involved in cell migration and invasion via a Rac1-mediated process in normal as well as
cancerous cells [Teng et al 2009] . Thus it is reasonable to expect that a highly proliferative
and actively migrating population of cells such the neural crest cells forming the
craniofacial bones, would depend on Stat3 for migration.
But it also seems from the distribution of ALP-expressing and lacZ expressing cells that
the total number of ectopically located neural crest cells in the suture is insufficient to
102

account for the total number ectopic ALP stained cells in the suture region.( Fig1.3B. D,
D’)
Hence we conclude that although the mismigration of neural crest cells into the suture area
in Stat3 mutants may be one of the causes of the synostosis, it may not be the only cause.
If the neural crest cells are not solely accountable for all osteogenic cells causing the fusion
of the bones, the alternative cause would be involvement of non-osteogenic suture or
surrounding ectocranial cells , which may nonautonomously differentiate into osteogenic
tissue and contribute to the fusion of the suture. If the latter case is true we would expect
to detect some ectopic osteogenic activity early during suture development in the Stat3
knockout mutants.
No change in expression of the suture cell markers Twist1 and Jagged1 or the neural
crest-mesoderm boundary establishment markers (EphrinA2-EphA4), but a
reduction in the expression of Notch2 in the frontal bone.
Previous studies published from the Maxson laboratory have greatly increased our
understanding of the normal development of the coronal suture. The basic molecular cues
involved in the process and the changes in them in other cases of synostosis can be
summarized as follows:
i) EphrinA2 and A4 are expressed in a single layer of cells in the dermis up to the
anterior-posterior axial position of the coronal suture. The corresponding
receptor EphA4 is expressed in a pair of layers adjacent to the ligand. These
two layers become a single layer at the axial position posterior to the coronal
103

suture. Analysis of EphA4 mutants shows that EphA4 is required for the correct
migration of the neural crest cells [Merrill et al 2006] . We show that there is
no change in the expression of the EphirnA2-EphA4 ligand –receptor pair in
neural crest-specific Stat3 mutants. This is expected, since the cells expressing
this ligand-receptor pair are of mesodermal origin. However, this finding shows
that cell mixing can occur in the developing coronal suture despite the correct
organization of the Eph-Ephrin layer.
ii) The suture cells exclusively express the membrane bound Notch ligand,
Jagged1. The loss of Jagged1 expression leads to cell mixing, as well as ectopic
osteogenic differentiation in the suture region. [Yen et al 2010]. However, we
show that there is no loss of Jagged1 expression in the suture of neural crest
specific homozygous Stat3 mutants, even in regions with clear fusion of the
suture (Fig 1.4C.D,D’). This indicates that the fusion in Stat3 mutants occurs
by ectopic osteogenic activity despite the normal differentiation of suture cells,
thus supporting the view that synostosis occurs by a mechanism different from
the ones described to date.
iii) Thirdly, Twist1, is present in the suture, the ectocranial layer as well as the bone
fronts. It appears to have differential activity in each of these zones. In the
ectocranial layer it controls the expression of EphA4 and EphrinsA2 and A4. In
the suture, it appears to control the expression of Jagged1. The function of
Twist1 in the suture also appears to be responsible for inhibiting expression of
the receptor of the Jagged1 ligand, Notch2, in the suture. In summary, Twist1
104

controls Jagged1 expression thus maintaining suture integrity. Reduction of
Twist1 leads to loss of Jagged1 and upregulation of Notch2 expression in the
suture and consequently to the onset of osteogenic activity in the suture. We
show that similar to Jagged1, there is no change in Twist1 expression in the
suture of neural crest specific homozygous Stat3 mutants.
iv)
Finally, the fourth molecular marker that appears to be involved directly in the
formation of the coronal suture is the receptor for the Jagged1 ligand, Notch2.
Among all the markers described above, Notch2 is the one to be exclusively
expressed only in the growing frontal and parietal bone fronts and is absent
from the suture area. Ectopic expression of Notch2 in the suture occurs on
mesoderm specific inactivation of Jagged1, thus linking an increase in the
expression of Notch2 in the suture with suture fusion. In humans TWIST1
+/-
,
JAGGED1
+/-
as well as EPHA4
+/-
cause craniosynostosis, but there is no report
of mutations in the Notch2 gene to be associated with any synostosis phenotype.
[McDaniell et al 2006] In contrast, in the neural crest-specific Stat3 knockdown
mutants (Wnt1cre; Stat3
cko/cko
), we show that there is extensive loss of Notch2
protein expression from the neural crest-derived frontal bone. This observation
gives rise to two conclusions. Firstly that in contrast to the case for mesoderm-
specific loss of Jagged1, which led to ectopic expression of Notch2 in the
suture, loss of Notch2 protein from the growing bone fronts may also lead to
fusion of the coronal suture, which naturally would be by a distinct mechanism
from that caused by loss of Notch2 from the suture. Secondly, it appears that
105

Notch2 expression is downstream of Stat3 activity in the neural crest cells and
that the synostosis observed in the homozygous knockout of Stat3 in neural
crest (Wnt1cre; Stat3
cko/cko
) could be brought about by loss of Notch2 in the
neural crest-derived frontal bone.

Thus, in summary, it appears that in Wnt1cre; Stat3
cko/cko
mutants, the expression and
relative distribution of the mesodermal markers EphA4-EphrinA2/4 (ectocranial layer),
Twist1 and Jagged1 (suture) are unchanged, while there is significant loss is the expression
of Notch2 specifically in the neural crest-derived frontal bone. This is in contrast to
previously observed ectopic expression of Notch2 in the suture area accompanying other
cases of coronal suture fusion.
Neural crest specific conditional loss of Stat3 mice show ectopic osteogenic activity in
the suture area but no change in change in the BMP signaling
Mechanistically, the development of the synostosis of the coronal suture appears to follow
a different mechanism than the previously established ones, since there is no disruption in
the differentiation markers for the suture cells itself but a reduction in the crest specific
marker expression. I show here that in the neural crest specific Stat3 knockdown mutants
there is increased osteogenic activity in and around the tip of the growing frontal bone early
during suture development, as well as in the suture area adjacent to it and arguably in a
layer of cells overlapping with but dorsal to the suture cells extending up to the parietal
bone tip (Fig 1.5.B). This region of increased osteogenic activity may well be responsible
106

for the expression ectopic of ALP positive cells that eventually create a bridge between the
2 bone tips and cause the fusion of the suture.
Notably, increase in BMP signaling which is also regularly associated with fusion of
sutures [Yen et al 2010] , was not observed in our homozygous Stat3 mutants (Wnt1cre;
Stat3
cko/cko
) , thus confirming that knocking down Stat3 signaling in the neural crest
population does not affect the BMP signaling in the suture area.
Proposed mechanism of coronal suture fusion in mice with neural crest-specific
conditional loss of Stat3
The observation that only neural crest-specific and not mesoderm-specific homozygous
loss of Stat3 leads to coronal synostosis has manifold significance in the understanding of
suture biology. Firstly, it provides strong evidence for active Stat3 signaling in neural crest
cells making up the frontal bones. Secondly, this is the first report of fusion of the bones
due to loss of gene activity specifically in the Wnt1-Cre population, thus opening up the
idea that synostosis is not solely dependent on the suture cells, but that all three tissues
present at the tripartite junction (neural crest-derived frontal bone tip, mesoderm derived-
non osteogenic suture cells and mesoderm derived parietal bone tip) are in a constant
dynamic state of proliferation, differentiation, morphogenetic rearrangement and a
disruption in any of the tissues in any manner during its formative stages, may lead to a
loss of balance between the established boundaries and differentiation states of all the three
tissues involved.
107

Since suture cells appear to maintain their location and differentiation identity in the Stat3
mutant, I would like to propose the alternative mechanism, hinted at by our data, by which
the fusion process proceeds in neural crest-specific Stat3 homozygous loss mutants
(Wnt1cre; Stat3
cko/cko
).
My data suggest that there is a migration defect in the LacZ-stained neural crest cells in the
Stat3 mutant, which can be attributed to reduction in Stat3 function in the migrating cells.
But the observation of a reduction in Notch2 activity in the neural crest-derived frontal
bone makes me propose an alternative hypothesis for the mechanism. As I observed in
wildtype samples, the Jagged1 –Notch2 transmembrane ligand –receptor pair is expressed
by adjacent tissues in the developing coronal sutures. Jagged1 is expressed by the
nonosteogenic suture tissue, while Notch2 is expressed by the osteogenic bone fronts. It
has been long known that Notch receptors carry out bidirectional signaling with their
transmembrane ligands and are involved in cell-cell border recognition and tissue
patterning. [Artvanis Tsakonas et al 1995] Thus it can be extrapolated that the expression
of Jagged-and Notch in this patterned manner reflects the recognition of a border by the
two subset of cells (osteogenic vs non osteogenic) and consequently in the anterior part of
the coronal suture, the boundary between the neural crest vs mesoderm derived tissue as
well. It has been also previously proposed that Twist1 expression in the suture is able to
exclude the Notch2 expressing cells from the suture region [Yen et al 2010]

108

Model : Proposed mechanism for coronal suture fusion in Wnt1cre; Stat3
cko/cko
mutants

In summary, I propose a new model mechanism of synostosis for the Wnt1cre; Stat3
cko/cko
mutants. I propose that it is due to the loss of Notch2 expression in the migrating neural
crest cells allocated to the frontal bone in the mutants , that the cells on reaching the
growing frontal bone tip are unable to read the Jagged1 signal in the adjacent suture tissue
and consequently infiltrate the suture area, (as observed by the presence of individual LacZ
stained cells) being unable to identify the border specification signal. Once in the suture,
these potentially osteogenic cells cause non-autonomous ectopic expression of osteogenic
FB
Jag1
PB
FB
CS
PB
FB
CS
PB
FB
Jag1
CS
PB
FB CS
Stat3
Stat3
Wnt1cre; Stat3
cko/cko

Wildtype
[Wnt1cre;LacZ]
[Wnt1cre;LacZ]
PB
FB: frontal bone, PB: parietal bone, CS: coronal suture
Jag1: Jagged1 expression zone, Notch2: Notch2 expression along the bone fronts ,
Stat3: Tissue in which Stat3 signaling is active or deactivated
109

signals (Runx2) in a limited number of their surrounding cells thus creating the Runx2-
induced bridging effect in limited regions in the suture. Since the localization of the
infiltrating cells varies for different locations along the suture, hence the fusion type
(observable by ALP staining) is also varied between different mutant samples as well as
different location within the same sample. We also propose from our studies that Stat3
actually acts upstream of Notch2 signaling and is responsible for the faithful expression of
the Notch2 receptor, the expression of which is lost in the absence of Stat3, causing in the
ensuing suture disruption.

110

MATERIALS AND METHODS
A. Mice & Genotyping
All transgenic mice used for this work has been described previously. They are as
follows,
Wnt1cre [Danielian et al 1998]
Mesp1cre [Saga et al 1999]
Stat3
flox/flox
[Takeda et al 1998]
Wnt1cre/R26R(LacZ) [Jiang et al 2000]

Genotyping Primers:
Cre F: TGC TGT TTC ACT GGT TAT GCG G
R: TTG CCC CTG TTT CAC TAT CCA G
Mesp1cre Mesp1cre : GCC ATA GGT GCC TGA CTT ACT
Comm : GTT ATT CAA CTT GCA CCA TGC
Stat3 flox F (wt) : CCT GAA GAC CAA GTT CAT CTG TGT GAC
F (flox) : TTT GGA AAG TAC TGT AGC CCC GAG AGC
R : CAC ACA AGC CAT CAA ACT CTG GTC TCC
R26R LacZ LacZ WT : GGA GCG GGA GAA ATG GAT ATG
LacZ KO : GCG AAG AGT TTG TCC TCA ACC
LacZ comm : AAA GTC GCT CTG AGT TGT TAT

111

B. Skull preparation and Alizarin Red staining
All pups were gathered at P21. The heads were separated, skinned and kept in 2%KOH
solution for 2 overnights before staining in dilute Alizarin Red solution in 2% KOH to stain
calcified tissue. Stained skulls were then washed, stored and photographed in 100%
glycerol.
C. Whole mount preparation of embryonic heads and ALP ( alkaline
phosphatase) detection staining
Embryos at required embryonic stages were dissected out in cold PBS and the outer skin
is carefully removed before fixing in 4%PFA overnight, following which they were washed
with PBST. For staining, the samples were submerged and rocked in ice cold TBST ( 10
mins ) , NTMT ( 10 mins ) and 0.01%BCIP-0.025%NBT in NTMT solution ( till adequate
staining in achieved) in succession. The staining was then stopped with 1mMEDTA/PBST
solution overnight. The samples were then washed with and photographed in dH20 and
finally stored in 70% EtOH.
D. Whole mount preparation of embryonic heads and LacZ staining
Embryos at required embryonic stages were dissected out in cold PBS and the outer skin
is carefully removed before fixing in 4%PFA for 20 mins, following which they were
washed with PBS. For staining, the samples were submerged in LacZ staining solution (1M
MgCl2/ %%DOC / 2%NP-40 / 2% X-Gal soln / 1.05%FeII/ 0.825%FeIII in PBS) and
rocked in 37C overnight. Following the staining, the samples were washed with PBS until
112

all excess staining were removed. The samples were then postfixed in 4%PFA overnight,
washed with and photographed in PBS and stored in 70% EtOH.
E. Embedding and Cryosectioning of embryonic heads
Embryos at required embryonic stages were dissected out in cold PBS, fixed in 4%PFA
for 30 mins, and washed in PBS followed by sucrose gradient ( 15%, 30%) and OCT
embedding . Sectioning was done at -20
o
C to produce sections 10um apart in multiple
corresponding slides and stored in -80
o
C until used for staining.
F. ALP staining of sections
Slides stored at -80
o
C, were dried in room temp for 10 mins followed by 10 mins of
Acetone fixation at 4
o
C. For staining, the sections were treated with ice cold TBST (10
mins ) , NTMT (10 mins ) and 0.01%BCIP-0.025%NBT in NTMT solution ( 30 secs-
1min) in succession. Excess stained was drained and slides were washed with dH20
followed by counterstaining with dilute NFR (nuclear fast red soln) for about 30 secs.
Excess NFR was then drained off and the slides were washed with dH20 before
dehydrating in a gradient of increasing % of Ethanol solution and mounted. Post mounting,
the slides were dried for at least 1 overnight before being photographed.
G. LacZ staining of sections
Slides stored at -80
o
C, were dried in room temp for 10 mins followed by 10 mins of fixing
Glutaraldehyde soln (/ PBS/2mMMgCl2/0.02% Glutaraldehyde). After fixation, the slides
are washed with ice cold PBS for 10 mins. For staining, the slides are then treated with (i)
113

ice cold PBS/2mMMgCl2 solution for 10 mins , followed by (ii) ice cold PBS/2mMMgCl2/
0.01%DOC/0.02%NP40 for 10 mins and eventually submerged in LacZ staining solution
(1M MgCl2/ %%DOC / 2%NP-40 / 2% X-Gal soln / 1.05%FeII/ 0.825%FeIII in PBS) and
rocked in 37C overnight. The next day the slides are first allowed to come room
temperature , following which they are removed from the staining solution and washed
with fresh PBS 3 times in succession , till all excess stain is removed. The slides are then
post fixed with 4%PFA for about 2 hrs followed by PBS wash, counterstain with NFR ( 30
secs) , dH20 wash in succession. Finally the slides were dehydrated in a gradient of
increasing % of Ethanol solution and mounted. Post mounting, the slides were dried for at
least 1 overnight before being photographed.
H. Immunohistochemistry & Immunofluorescence
Immunostaining for detection of various proteins was carried out as previously described
by Ishii et al , 2003. Briefly :
Slides stored at -80
o
C, were dried in room temp for 20 mins followed by 10 mins of fixing
at 4
o
C. Following fixation, slides meant for Immunohistochemistry were treated with
Peroxo block for 1 min and then washed with PBST before. Blocking was done for 1 hr at
Room Temp, 1
o
Antibody exposure was done for 1 overnight and 2
o
Antibody exposure for
1 hr in Room Temp.
All slides meant for Rabbit primary antibody were blocked with 10%Goat serum and the
antibodies were diluted in the same. All slides meant for Goat primary antibody were
blocked with 1%BSA serum and the antibodies were diluted in the same. Slides for
114

immunohistochemistry were visualized with DAB chromogen and slides for
immunofluorescence were visualized by epifluorescence microscopy. For
immunochemistry, after detection of antibody with chromogen, slides are washed and
counterstained with Hematoxylin ( 30 sec), dH20 was, PBS ( 30 secs ) in succession ,
before a last dH20 wash, dehydration and mounting in mounting medium) For
immunofluorescence the slides are counterstained with DAPI and after washing the excess
DAPI with dH20, the slides are mounted with ProLong Gold Antifade reagent.

I. Measurements : Coronal suture index (CI)
Modified from Oram and Gridley ( 2005) , each suture in the 2 hemispheres are
individually scored for the relative length of suture fused . The scoring is as follows :
Primary
Antibody
Company Source Dilution Fixing Secondary Counterstain
EphrinA2 Zymed Rabbit 1:250 4%PFA From Kit Hematoxylin
EphA4 R&D Rabbit 1:200 4%PFA From Kit -
Twist1 SIGMA Rabbit 1:200 Acetone Biotinylated
AntiGoat
Hematoxylin
Jagged1 Santa
Cruz
Goat 1:300 4%PFA From Kit Hematoxylin
Notch2 Santa
Cruz
Goat 1:150 4%PFA Biotinylated
AntiGoat
-
Runx2 Sigma Rabbit 1:200 Acetone From Kit Hematoxylin
PSmad1/5/8 Cell
signaling
Rabbit 1:50 Acetone FITC labelled
Anti-Rabbit
DAPI
115

0 to fully patent suture, 0.5 for fusion equal to or less than 1/8 of the suture 1 for fusion
of equal to or less than ¼ of the suture, 2 for equal to or less than ½ , 3 for equal to or less
than ¾ and a 4 indicating a fully fused one.
The scores for each type of suture is added for the individual genotype sets and the Mean
score and SD are calculated which are then tabulated.

116

CHAPTER 2
Notch2 signaling in cranial neural crest tissue is directly involved in maintaining
coronal suture patency and acts downstream of Stat3 in production of the coronal
suture synostosis phenotype observed in neural crest specific Stat3-/- mice.
INTRODUCTION
Notch signaling is a highly conserved signaling mechanism present in all currently known
vertebrate as well invertebrate organisms known to be ubiquitously involved in regulating
cell fate determining and consequently tissue patterning mechanisms. [Artavanis-Tsakonas
et al 1999]. Identified in fruitflies, the Notch gene codes for a transmembrane receptor
which consists of (i) a large extracellular domain consisting of 36 EGF (epidermal growth
factor) like repeats and 3 cysteine rich Notch/LIN12 repeats, (ii) single pass transmembrane
domain and (iii) a characteristic intracellular domain containing 6 ankyrin repeats, a
glutamine rich domain and a PEST sequence making up the Notch Intracellular Domain
(NICD) [Wharton et al 1985, Kidd et al 1986]. The 4 Notch receptors (Notch 1-4) are
activated by 5 known transmembrane ligands ( Delta1, 3, 4 and Jagged 1,2 ) and on
activation, the NICD is cleaved and translocated to the nucleus. Within the nucleus NICD
requires its binding partner RBP-J (Recombination signal Binding Protein for
immunoglobulin kappa J) in order to bind to DNA sequences and activate transcription of
its downstream target genes. [Gridley et al 2003, Kovall 2008, Kopan & Ilagan 2009]. The
signaling between the Notch receptor and its corresponding transmembrane ligands appear
to be bidirectional and occurs between 2 different types of cells expressing the receptor or
117

the ligand respectively come in contact with each other. This is able to activate separate
signaling cascades in the adjacent cells and thus induce differential tissue patterning in the
2 population of cell types. [Artavanis-Tsakonas et al 1999, Kopan & Ilagan 2009]
From previous studies carried out in the lab it has been elucidated that Notch2 protein is
exclusively expressed in growing frontal and parietal bone fronts during early coronal
suture development while its ligand Jagged1 is concurrently expressed in the
nonosteogenic suture. [Yen et al 2010]
Alagille syndrome (AGS), a multi-organ system disorder, is associated with heterozygous
loss of the Notch2 ligand JAGGED1. Humans express 5 Notch ligands (Jag1, Jag2, Dll1,
Dll3, Dll4) and 3 receptor subtypes (Notch1-3), but among them, AGS is associated with
mutation in the JAGGED1 gene in at least 94% of reported cases. To date, there is no report
of any syndromic or nonsyndromic craniosynostosis genetically linked to a defect in the
NOTCH2 gene or any other NOTCH homologue McDaniell et al [McDaniell et al 2006]
detected mutations in exon8 of NOTCH2 gene (substitution mutation affecting the EGF
like ankyrin repeats) in 2 out of 11 cases of non-JAGGED1 associated AGS patients. But
these patients, despite displaying several distinguishing phenotypes of AGS, were not
reported to have craniosynostosis. In 2002, McCright et al reported the development of
Jag1
dDSL
/+Notch2
del
/+ double heterozygous mice as a model for AGS [McCright et al
2002]. Thus combination mutant recapitulated the defects in the bile duct, heart, eye and
growth typical of AGS, but not craniosynostosis. Conversely, we showed previously that
in conditional Jagged1 mutants (Dermo1cre;Jagged1
cko/cko
) the mesoderm-specific loss of
118

Jagged1 that causes synostosis also leads to increased ectopic expression of the Notch2 by
the suture cells [Yen et al 2010].
In this context it was significant to note from my studies with neural crest specific Stat3
mutant mice that develop coronal suture synostosis, that these mice display almost
complete loss of Notch2 protein in the neural crest derived frontal bone tissue.
This led to the investigation of the hypothesis that Notch2 acts downstream of Stat3
signaling in maintenance of neural crest derived tissue integrity during the development of
coronal suture and a crest specific loss of Notch2 may lead to synostosis of the suture.

119

Results
2.1 Neural crest-specific loss of Notch2 leads to coronal suture synostosis of a greater
severity that in case of neural crest specific loss of Stat3
From Notch 2 immunohistochemistry studies described previously (Fig1.4D) it was
noted that neural crest specific inactivation of Stat3 (Wnt1cre; Stat3
cko/cko
) leads to a
dramatic reduction in the expression of Notch2 receptor specifically in the growing neural
crest derived frontal bone tissue. This observation led me to investigate whether this neural
crest specific loss of Notch2 could be responsible for bone fusion and eventual synostosis.
For this purpose I generated mice with neural crest-specific (Wnt1cre) as well as
mesoderm-specific (Mesp1cre) inactivation of the Notch2 gene (by crossing
Wnt1cre;Notch2
flox/wt
or Mesp1cre; Notch2
flox/wt
male mice with Notch2
flox/flox
homozygous
females) and assessed craniosynostosis in different sutures in Alizarin red stained skulls
(to distinguish ossified vs non-ossified tissue) of P21 mice.

120

Wildtype
Wnt1Cre;Notch2
cko/wt

Mesp1Cre;Notch2
cko/cko

Wnt1Cre;Notch2
cko/cko

fb
pb
cs
fb
pb
cs
fb
pb
cs
A B C
Wildtype
Wnt1Cre;Notch2
cko/cko

ALP
fb fb
cs
cs
pb
pb
pb
cs
fb
Mesp1Cre;Notch2
cko/wt

Fig 2.1: Neural crest specific conditional loss of Notch2 leads to partial
bicoronal suture synostosis
Fig2.1 P21 stage mice skulls stained with Alizarin Red (calcified tissue) and
ALP stained Horizontal sections of E14.5 stage mouse embryo heads.
A-E : Dorsal view of the Alizarin red stained skulls of respective genotypes
showing the Frontal and Parietal bones juxtaposed to each other and the coronal
suture demarcating their boundary.
A’,C’: Magnified view of the coronal suture in figures A,& C respectively.
F,G: Horizontal sections of mice head at E14.5 , stained with ALP to mark the
relative locations of the frontal and parietal bone tips and the position of the
coronal suture.
pb: parietal bone, fb: frontal bone, cs : coronal suture
Arrowheads mark the coronal suture tissue
E D
A’ C’
AR
G F
AR
P21 E14.5
pb
cs
fb
121

Table 2 : Craniosynostosis Index [ Neural crest specific loss of Notch2 ]
Modified from Oram and Gridley , 2005. Mean score is tabulated along with + SD which
gives the idea about relative severity of the synostosis phenotype.
Scoring : 0-unfused, 0.5- >12.5% fused , 1- >25% fused , 2- >50% fused , 3- >75% fused
, 4- >100% fused
122

I did not observe craniosynostosis in any of the sutures in the mice with mesoderm-specific
loss of Notch2 signaling (Mesp1cre; Notch2
flox/wt
or

Mesp1cre; Notch2
flox/flox
) which knocks
down Notch2 signaling in the parietal bone and the coronal suture tissue (Fig2.1D,E). In
contrast, mice with homozygous conditional knockout of Stat3 in neural crest derived
tissues (Wnt1cre;Notch2
cko/cko
) displayed partial bilateral coronal suture synostosis with
100% penetrance (Fig2.1C). The synostosis of the coronal suture in the
Wnt1cre;Notch2
cko/cko
mice appeared to be of a more severe type than the previously
observed Wnt1cre;Stat3
cko/cko
mice, supported by the fact that the calculated
Craniosynostosis Index (CI) (Oram and Gridley [ 2005]; Materials and Methods) for fusion
in the Wnt1cre;Notch2
cko/cko
was1.83 as opposed to 1.00 for Wnt1cre;Stat3
cko/cko
(Table2 ).
Interestingly the sutures on both sides appeared to be fused at the dorsal-most surface of
the skull in the zone of the most recently formed section of the suture. The remaining
sutures, including the frontal, squamosal, lambdoid and interparietal, appeared unaffected.
Mice with heterozygous loss of Notch2 in neural crest (Wnt1cre;Notch2
cko/wt
) did not
develop synostosis in the coronal suture or any other. (Fig2.1B).
This observation is in accordance with the previous observation that neural crest-specific
knockout of Stat3, which led to incomplete synostosis of the coronal suture, also caused a
reduction of Notch2 signaling specifically in the frontal bone.
In addition to synostosis, I also observed a foramen in the interfrontal region. This suggests
that Notch2 might be involved in the proliferation and differentiation of the osteoblast
123

population forming the frontal bone. I did not follow up on this phenotype since it did not
seem to be relevant to suture development.
To confirm the severity of the synostosis phenotype, I stained horizontal sections of E14.5
Control and Wnt1cre;Notch2
cko/cko
mouse heads for Alkaline

Phosphatase (ALP) to
visualize the morphology of the growing frontal and parietal bones. As I observed
previously in the control section, the frontal bone extended below the parietal bone with a
clear non-osteogenic zone of the coronal suture between the two bones. In the case of the
mutant, I observed complete fusion of the two bones (Fig2.1F,G). This was in contrast to
varied degree and morphology of fusion in the case of Wnt1cre;Stat3
cko/cko samples

(Fig1.2B)).

124

Control
Wnt1Cre;Notch2
cko/cko

cs
fb
cs
pb
cs Jagged1
ALP
fb
cs
pb
Fig 2.2: Investigation of Jagged1 expression in Horizontal sections of mice heads
during early embryonic stages of coronal suture formation
A,B: Horizontal sections of Control and Wnt1Cre;Notch2
cko/cko
E14.5 mouse head
sections immunostained for detection of Jagged1 in the coronal suture area .
A’,B’: Sections corresponding to A and B respectively, stained for ALP to mark the
relative locations of the frontal and parietal bone tips and the position of the coronal
suture.
pb: parietal bone, fb: frontal bone, cs : coronal suture
Arrowheads mark comparative location of coronal suture
A
B
A’ B’
E14.5
Fig2.2: Investigation of Jagged1 expression in the developing coronal suture
area in mice with Neural crest specific conditional loss of Notch2 displays
no relative change in expression pattern
fb
pb
fb
pb
125

2.2 Investigation of Jagged1 expression in the developing coronal suture region
during in neural crest specific Notch2 floxed mice
As mentioned before, Alagille syndrome is caused by heterozygous loss of JAGGED1.
Affected individuals develop craniosynostosis as one of the prominent phenotypes of this
multi-organ disorder [Alagille et al 1987, Kamath et al 2004]. Previously published data
from the Maxson lab show that mesoderm specific knockout of Jagged1 causes loss of
Jagged1 protein expression from the coronal suture region and subsequent fusion of the
frontal and parietal bones [Yen et al 2010]. This phenotype is accompanied by ectopic
expression of Notch2 in the coronal suture region. This suggests that fusion of coronal
suture is associated with the ectopic expression of Notch2. However, neural crest-specific
Stat3 knockout mice display a loss of Notch2 expression specifically in the neural crest
tissue itself (Fig1.4D: B, D). In addition, neural crest-specific loss of Notch2 alone results
in a severe synostosis phenotype (Fig2.1: B, D).
Hence, I decided to look for any possible changes in Jagged1 expression in the neural crest
specific Notch2 knockout mutants (Wnt1cre;Notch2
cko/cko
) that may result from loss of
Notch2 from the neural crest tissue and that may be involved in the bone fusion observable
in these mutants.
For this purpose I performed immunohistochemistry on horizontal sections of control and
Wnt1cre;Notch2
cko/cko
mice heads from stages embryonic day 14.5 (E14.5) for detection of
Jagged1 in the suture and ECL ( ectocranial layer ) I also stained ections adjacent to the
126

immunostained slides for ALP to detect the location of the coronal suture and the
morphology of the frontal and parietal bone fronts.
As shown, in the control samples I observed strong Jagged1 staining in the coronal suture
with significant staining in a layer above the bone fronts, especially the frontal bone
(Fig2.2: A). In the mutant (neural crest specific loss of Notch2) I detected no significant
loss of Jagged1 staining in the coronal suture area, or in the ECL above the bones (Fig2.2:
B). From the adjacent ALP-stained sections it is evident that Jagged1 staining in the suture
zone persists despite complete fusion of the frontal and parietal bones in the
Wnt1cre;Notch2
cko/cko
mice.

127

Control
Wnt1Cre;Notch2
cko/cko

cs
fb
cs
pb
cs
Runx2
ALP
fb
cs
pb
E14.5
Fig 2.3: Investigation of Runx2 expression in Horizontal sections of mice heads
during early embryonic stages of coronal suture formation
A,B: Horizontal sections of Control and Wnt1Cre;Notch2
cko/cko
E14.5 mouse head
sections immunostained for detection of Runx2 around the coronal suture area .
A’,B’: Sections corresponding to A and B respectively, stained for ALP to mark the
relative locations of the frontal and parietal bone tips and the position of the coronal
suture.
pb: parietal bone, fb: frontal bone, cs : coronal suture
Arrowheads mark comparative staining at the coronal suture area

Fig2.3: Investigation of early osteoblast marker Runx2 expression in the
developing coronal suture area in mice with Neural crest specific conditional
loss of Stat3 shows increased expression of Runx2 in the suture area
A
B
A’
B’
pb
pb
fb
fb
128

2.3 Investigation of expression of early osteogenic marker Runx2 in the developing
coronal suture region early during suture development, shows ectopic expression in
the suture region in neural crest specific homozygous loss of Notch2
I found that mutants with neural crest-specific loss of Stat3 (Wnt1cre;Stat3
cko/cko
)
developed fusion of the suture. These mutants also showed ectopic Runx2 staining in the
suture region at E13.5 while the suture is still developing before the fusion becomes
detectable by ALP staining (Fig2.5A: B, B”). This gives the indication that ectopic
osteogenic signals are active in the suture region in Wnt1cre;Stat3
cko/cko
mutants.
Since Wnt1cre;Notch2
cko/cko
mice also develop fusion of the bones, I wished to investigate
the presence of any increased osteogenic activity other than in the bone fronts, or ectopic
osteogenesis in the developing coronal suture region in Wnt1cre; Notch2
cko/cko
.mutants.
For this purpose, I performed immunohistochemistry on horizontal sections of control and
Wnt1cre;Stat3
cko/cko
mouse heads from E14.5 for the detection of the early osteogenic
marker, Runx2 in the developing bone fronts and intervening suture area. I stained sections
adjacent to the immunostained slides with the oseoblast marker ALP to detect the location
of the coronal suture and the morphology of the frontal and parietal bone fronts realtive to
the suture.
In the control samples I observed Runx2 staining in the frontal and parietal bone fronts and
no staining in the suture area. (Fig2.3: A).
129

In the neural crest specific loss of Notch2 mutants (Wnt1cre; Notch2
cko/cko
) I observed
significant ectopic expression of Runx2 in the developing suture area (Fig2.3: B). The
corresponding ALP stained slides show initiation of the fusion of the bones in the mutant.

130

Control
Wnt1Cre;Notch2
cko/wt

Wnt1Cre;Stat3
cko/wt

LacZ
ALP
LacZ
LacZ
ALP
ALP
Fig2.4: LacZ marking of neural crest cells in Neural crest specific conditional
loss of Notch2 and Stat3 display similar minor loss of crest-mesoderm border
recognition
E 16.5
cs
A
A’
fb pb
cs
fb
pb
cs
fb
pb
cs
fb
pb
cs
fb
pb
cs
fb
pb
C
B
C’
B’
A’, B’,C’: Sections corresponding to A-C with Wnt1cre tissue stained for LacZ for
detection of Neural crest cells in the coronal suture area.
pb: parietal bone, fb: frontal bone, cs : coronal suture
Arrowheads mark the coronal suture area and individual neural crest cells in the
coronal suture tissue or ectocranial zone
Fig 2.4: Investigation of localization
of LacZ stained cells in Horizontal
sections of mice heads
A,B,C : Horizontal sections from E16.5
Control (A), Wnt1Cre;Stat3
cko/wt
(B) and
Wnt1Cre;Notch2
cko/wt
(C) mice stained
for ALP to mark the relative locations of
the frontal and parietal bone tips and
the position of the coronal suture

131

2.4. Investigation of LacZ staining Wnt1cre population of cells of neural crest origin,
shows presence individual cells from the neural crest population beyond the neural
crest boundary & in the mesodermal zone in Wnt1cre; Stat3
cko/cko
mutants
As mentioned in Chapter 2, previously published data from Twist1
+/-
and EphA4
-/-
mutant
mice show that disruption of neural crest-mesoderm boundary and subsequent mixing of
neural crest and mesoderm appear to be a significant part of the mechanim of suture fusion
in these mutants. I tested this idea in the neural crest-specific Stat3 homozygous mutant
mice in which I found that individual LacZ stained cells marking the neural crest
population in Wnt1cre; Stat3
cko/cko
;R26RlacZ
ki/ki
mice were located beyond the boundary
of the Wnt1Cre marked neural crest tissue. These cells were in the zone of the developing
suture from E13.5 onwards and most prominently in E16.5 and beyond (Fig1.3B. A, B).
In order to

determine whether the neural crest specific reduction of Notch2 expression is
a true downstream effect of the loss of Stat3 function, I asked whether I could detect the
mismigration of the frontal bone cells and corresponding cell mixing in the Notch2 mutants
as well.
For this purpose, I generated triple mutants with homozygous and heterozygous knockout
of Notch2 and LacZ marking of neural crest-derived cells (Wnt1cre; Notch2
cko/cko

;R26RlacZ
ki/wt
and Wnt1cre;Notch2
cko/wt
;R26RlacZ
ki/wt
). I cut horizontal sections of E16.5
mouse heads and stained these for LacZ to assess any cell mixing.
At E16.5, in control samples only, the neural crest-derived frontal bone is stained with
LacZ and the coronal suture area is not (Fig2.4.A). On the other hand I found that in
132

mutants with heterozygous loss of Notch2 in the neural crest (Wnt1cre;Notch2
cko/wt
), at
E16.5, individual LacZ stained cells werelocated beyond the boundary of the Wnt1Cre-
marked neural crest tissue and in the zone of the developing suture (Fig2.4.C)
I also observed an increase in the number of LacZ-positive cells in the ectocranial zone in
case of the neural crest specific Notch2 heterozygous mutant compared with control or
heterozygous Stat3 mutants. It appears from these data that the neural crest-specific
heterozygous Notch2 mutants display a slightly more severe lacZ phenotype than the
neural crest-specific heterozygous Stat3 conditional mutants (both in case of number of
sutures affected and distribution of LacZ staining cells). I will carry our further work to
distinguish whether the phenotype is more severe in the case of homozygous mutants (i.e.,
whether the mutation is truly dose-dependent) and how such mutants appear in comparison
to the neural crest-specific homozygous Stat3 mutants.

133

Table 3 : Craniosynostosis Index [ Neural crest specific rescue of Stat3 mutants by
constitutively active form of NICD ]
Fig 2.5 Neural crest specific conditional activation of Notch signaling with
constitutively active Notch-Intracellular domain (NICD) expression leads to
rescue of the partial bicoronal suture synostosis caused by Neural crest
specific conditional loss of Stat3
Modified from Oram and Gridley , 2005. Mean score is tabulated along with + SD which
gives the idea about relative severity of the synostosis phenotype.
Scoring : 0-unfused, 0.5- >12.5% fused , 1- >25% fused , 2- >50% fused , 3- >75% fused
, 4- >100% fused
Wildtype
Wnt1Cre;Stat3
cko/cko
;
ROSA
NICD/wt

Wnt1Cre;Notch2
cko/cko

Wnt1Cre;Stat3
cko/wt
;
ROSA
NICD/wt

Wnt1Cre;Stat3
cko/cko

fb
pb
cs
pb
pb
pb pb
cs cs
cs
cs
fb
fb
fb
fb
Fig2.5. P21 stage mice skulls stained with Alizarin Red (calcified tissue)
A-E : Dorsal view of the Alizarin red stained skulls of respective genotypes showing
the Frontal and Parietal bones juxtaposed to each other and the coronal suture
demarcating their boundary.
pb: parietal bone, fb: frontal bone, cs : coronal suture
Arrowheads mark the coronal suture tissue
P21
A
B C
D
E
134

2.5 Neural crest-specific expression of constitutively active Notch signaling rescues
coronal synostosis in mutants with neural specific loss of Stat3
I have stated that the neural crest-specific loss of Stat3 leads to a characteristic type of
coronal suture fusion. It appears that in such cases the previously reported mechanisms of
loss of suture cell identity as observable by loss of Eph-Ephrin signaling, Twist1 or Jagged1
expression from the suture and consequent expansion of Notch2 expression to the sutures
is not at play here. Rather I have observed a concurrent loss of Notch2 expression from the
neural crest domain as well. Subsequently, I showed that neural crest specific inactivation
of the Notch2 gene causes synostosis of the coronal suture. This phenotype is more severe
than in case of the homozygous loss of Stat3 alone. These findings point to the idea that
Notch2 acts downstream of Stat3 in neural crest cells that make the frontal bone and in
fact, the activation of Stat3-dependent Notch2 expression is imperative for maintaining
suture patency.
To confirm this idea I decided to use the transgenic mouse strain with a Cre-dependent
constitutively active Notch intracellular domain (NICD) sequence knocked into the
ROSA26 allele ( Materials & Methods ). The NICD as mentioned previously is the
intracellular signaling peptide cleaved from the Notch receptor when it is activated. Using
Wnt1cre;Stat3
cko/wt
males and Stat3
cko/wt
; R26R
NICD/wt
females , I generated triple mutant
mice of genotype Wnt1cre;Stat3
cko/flox
R26R
NICD/wt,
in which the Stat3 signaling is knocked
out and Notch signaling constitutively activated exclusively in the neural crest tissue (
135

Wnt1cre). I assessed sutures for synostosis by Alizarin red staining of skulls (to distinguish
ossified vs non-ossified tissue) of P21 mice.
As shown in Fig2.5, we know that while in wildtype samples the coronal suture remains
fully patent at this stage, in neural crest specific homozygous loss of Stat3
(Wnt1cre;Stat3
ckocko
), partial bicoronal suture synostosis is observable throughout the
length of the suture. In the case of neural crest-specific homozygous loss of Notch2
(Wnt1cre;Notch2
cko/cko
) there is also partial bicoronal suture synostosis (Fig2.5.A, B, C).
However, in the case of the triple mutants (Wnt1cre;Stat3
cko/wt
; R26R
NICD/wt
) I found that
the synostosis phenotype caused by homozygous inactivation of Stat3cis completely
rescued, with the sutures completely patent (Fig2.5.E), such that the Craniosynostosis
Index (CI) for the triple mutant is 0 (Table 3).

136

Discussion
Notch2 signaling acts downstream of Stat3 in formation and maintenance of frontal
bone identity and morphology during coronal suture development.
One of the novel ideas that came forth form our studies with the neural crest specific Stat3
knockout mice is that inactivation of Stat3 in neural crest cells leads to a significant
reduction in the expression of the transmembrane receptor, Notch2, in the neural crest. The
expression of Notch2 is not even through the growing frontal bone primordia, but is
concentrated in the dynamic growth zone at the frontal bone tip. The function this receptor
is not yet understood. As we did not find any significant change in any of the other markers
known to be involved in suture fusion (both by human genetic studies as well as mouse
model data) , the loss of Notch2 in Stat3 neural crest specific mutants becomes more
significant, leading to the idea that Notch2 acts directly downstream of Stat3 in the process
of suture development. Under such circumstances, neural crest specific inactivation of
Notch2 should be able to recapitulate the fusion phenotype of the coronal suture. We show
here for the first time that that is indeed the case. Using knockout mice with the Notch2
gene knocked down in the neural crest tissue with the help if Cre/loxP system under Wnt-
Cre , we show that at three weeks after birth homozygous mutant mice develop synsostosis
of the coronal suture which is phenotypically more severe than in case of the homozygous
loss of Stat3 in the same tissue ( Fig 2.1) . We also show here that already during the early
formative stages of the suture there is much a more severe fusion in the case of the Notch2
mutant despite normal expression of Jagged1 by the suture cells. We also observed an
137

associated increase of osteogenic activity (Runx2 immunostaining) in the suture area and
the presence of individual LacZ stained neural crest cells infiltrating the suture (Fig 2.1,
2.2, 2.3, 2.4).
Finally we show that the synostosis phenotype produced by homozygous loss of Stat3 is
rescued by constitutively active Notch signaling in the neural crest tissue, produced by a
Cre dependent heterozygous allele encoding a constitutively active Notch intracellular
domain (NICD). Though further work needs to be done to confirm my findings, I propose
that Notch2 activity in the neural crest cells is acts downstream of Stat3 and that coronal
synostosis that results from neural crest dependent homozygous loss of Stat3 occurs via a
Notch2 dependent process.
Adding these observations to the previously proposed model I show:

138

As displayed in the top panel, neural crest specific loss of Notch2 (even in presence of a
functional Stat3 signaling) leads to loss of boundary maintenance by neural crest cells and
ectopic osteogenic activity leading to synostosis of the coronal suture , to a degree higher
than that in case of neural crest specific loss of Stat3 signaling alone.
The lower panel shows that on constitutive activation of the Notch signaling in the neural
crest tissue in presence of loss of Stat3 signaling in the latter, leads to a rescue of the
synostosis phenotype.

139

The significance of this finding is manifold. Firstly this is the first report of suture
synostosis developing due to an imbalance in Notch receptor expression and signaling.
There are two known genetic diseases associated with mutations in NOTCH2. A small
percentage of Alagille syndrome patients who normally display craniosynostosis have
substitution mutations in the Notch2 gene [ McDaniell et al 2006]. Although displaying
several other AGS phenotypes, these patients did not exhibit synostosis. Frameshift
mutations in the last exon of the NOTCH2 gene have been identified by at least two groups
in patients with the rare skeletal disorder Hajdu-Cheney syndrome (characterized by
progressive focal bone loss, craniofacial anomalies and renal cysts), which appears to cause
a gain-of-function in Notch2 protein activity [ Isidor et al . 2011, Simpson et al 2011]. No
craniosynostosis has been reported in these patients either. Recently a study involving
transgenic activation and deactivation of Notch signaling in neural crest cells was
published [ Mead &Yutzey 2012] . These authors used Wnt1cre; ROSA
Notch
mice to study
Notch gain of function and the Wnt1cre;RBP-J
flox/flox
mice to study Notch loss- of function.
The ROSA
Notch
allele is the same one used in our Notch gain-of –function studies, encoding
the constitutively active NICD. Interestingly mice from neither genotype (Wnt1cre;
ROSA
Notch
or

Wnt1cre;RBP-J
flox/flox
) were found to survive after birth, but no synostosis
was reported from studies up to E18.5.
In our experiments we have used specifically Wnt1cre;Notch2
flox/flox
mice which apparently
have a greater survival rate (all homozygous mutants survived at least up to 3 weeks after
birth), indicating other Notch pathways are active in neural crest cells. The use of
Wnt1cre;RBP-J
flox/flox
mice which completely abolishes all Notch signaling , by the

above
140

mentioned authors appears to be is fatal for the mutants and may interfere with the true
suture phenotype caused by specific loss of Notch2, accounting for them failing to note
any synostosis in their E18.5 mice. They also did not carry out sectioning of the suture to
investigate any synostosis.
Notably, in this report [Mead &Yutzey 2012] it is shown that Wnt1cre;RBP-J
flox/flox
mice
show reduced

ossification in craniofacial elements resulting in a foramen between the
frontal bones, This phenotype is also observed in my neural crest-specific Notch2 loss of
function mutants (Wnt1cre;Notch2
flox/flox
) indicating there is a separate Notch dependent
pathway affecting ossification of the craniofacial skeleton in which Notch2 may be heavily
involved.
These studies showing a completely separate fusion mechanism involving Stat3 and
Notch2 opens up the opportunities to study other upstream and downstream genes affected
by these pathways to further increase the known etiology of craniosynostosis.

141

MATERIALS AND METHODS
A. Mice & Genotyping
All transgenic mice used for this work has been described previously. They are as
follows,
Wnt1cre [Danielian et al 1998]
Mesp1cre [Saga et al 1999]
Stat3
flox/flox
[Takeda et al 1998]
Notch2
flox/flox

[McCright et al 2006]
Notch gof : ROSA
NICD/-
[Jackson Lab :
Gt(ROSA)26Sor
tm1(Notch1)Dam
/J]
Wnt1cre/R26R(LacZ) [Jiang et al 2000]

Genotyping Primers:
Cre F: TGC TGT TTC ACT GGT TAT GCG G
R: TTG CCC CTG TTT CAC TAT CCA G
Mesp1cre Mesp1cre : GCC ATA GGT GCC TGA CTT ACT
Comm : GTT ATT CAA CTT GCA CCA TGC
Stat3 flox F (wt) : CCT GAA GAC CAA GTT CAT CTG TGT GAC
F (flox) : TTT GGA AAG TAC TGT AGC CCC GAG AGC
R : CAC ACA AGC CAT CAA ACT CTG GTC TCC
Notch2 flox F: TAG GAA GCA GCT CAG CTC ACA G
142

R: ATA ACG CTA AAC GTG CAC TGG AG
Notch GOF Comm: AAA GTC GCT CTG AGT TGT TAT
R( mut) : GAA AGA CCG CGA AGA GTT TG
R ( wt) : TAA GCC TGC CCA GAA GAC TC
R26R LacZ LacZ WT : GGA GCG GGA GAA ATG GAT ATG
LacZ KO : GCG AAG AGT TTG TCC TCA ACC
LacZ comm : AAA GTC GCT CTG AGT TGT TAT

B. Skull preparation and Alizarin Red staining
All pups were gathered at P21. The heads were separated, skinned and kept in 2%KOH
solution for 2 overnights before staining in dilute Alizarin Red solution in 2% KOH to stain
calcified tissue. Stained skulls were then washed, stored and photographed in 100%
glycerol.
C. Whole mount preparation of embryonic heads and ALP ( alkaline
phosphatase) detection staining
Embryos at required embryonic stages were dissected out in cold PBS and the outer skin
is carefully removed before fixing in 4%PFA overnight, following which they were washed
with PBST. For staining, the samples were submerged and rocked in ice cold TBST ( 10
mins ) , NTMT ( 10 mins ) and 0.01%BCIP-0.025%NBT in NTMT solution ( till adequate
staining in achieved) in succession. The staining was then stopped with 1mMEDTA/PBST
143

solution overnight. The samples were then washed with and photographed in dH20 and
finally stored in 70% EtOH.
D. Whole mount preparation of embryonic heads and LacZ staining
Embryos at required embryonic stages were dissected out in cold PBS and the outer skin
is carefully removed before fixing in 4%PFA for 20 mins, following which they were
washed with PBS. For staining, the samples were submerged in LacZ staining solution (1M
MgCl2/ %%DOC / 2%NP-40 / 2% X-Gal soln / 1.05%FeII/ 0.825%FeIII in PBS) and
rocked in 37C overnight. Following the staining, the samples were washed with PBS until
all excess staining were removed. The samples were then postfixed in 4%PFA overnight,
washed with and photographed in PBS and stored in 70% EtOH.
E. Embedding and Cryosectioning of embryonic heads
Embryos at required embryonic stages were dissected out in cold PBS, fixed in 4%PFA
for 30 mins, and washed in PBS followed by sucrose gradient ( 15%, 30%) and OCT
embedding . Sectioning was done at -20
o
C to produce sections 10um apart in multiple
corresponding slides and stored in -80
o
C until used for staining.
F. ALP staining of sections
Slides stored at -80
o
C, were dried in room temp for 10 mins followed by 10 mins of
Acetone fixation at 4
o
C. For staining, the sections were treated with ice cold TBST (10
mins ) , NTMT (10 mins ) and 0.01%BCIP-0.025%NBT in NTMT solution ( 30 secs-
1min) in succession. Excess stained was drained and slides were washed with dH20
144

followed by counterstaining with dilute NFR (nuclear fast red soln) for about 30 secs.
Excess NFR was then drained off and the slides were washed with dH20 before
dehydrating in a gradient of increasing % of Ethanol solution and mounted. Post mounting,
the slides were dried for at least 1 overnight before being photographed.
G. LacZ staining of sections
Slides stored at -80
o
C, were dried in room temp for 10 mins followed by 10 mins of fixing
Glutaraldehyde soln (/ PBS/2mMMgCl2/0.02% Glutaraldehyde). After fixation, the slides
are washed with ice cold PBS for 10 mins. For staining, the slides are then treated with (i)
ice cold PBS/2mMMgCl2 solution for 10 mins , followed by (ii) ice cold PBS/2mMMgCl2/
0.01%DOC/0.02%NP40 for 10 mins and eventually submerged in LacZ staining solution
(1M MgCl2/ %%DOC / 2%NP-40 / 2% X-Gal soln / 1.05%FeII/ 0.825%FeIII in PBS) and
rocked in 37C overnight. The next day the slides are first allowed to come room
temperature , following which they are removed from the staining solution and washed
with fresh PBS 3 times in succession , till all excess stain is removed. The slides are then
post fixed with 4%PFA for about 2 hrs followed by PBS wash, counterstain with NFR ( 30
secs) , dH20 wash in succession. Finally the slides were dehydrated in a gradient of
increasing % of Ethanol solution and mounted. Post mounting, the slides were dried for at
least 1 overnight before being photographed.
H. Immunohistochemistry & immunofluorescence
Immunostaining for detection of various proteins was carried out as previously described
by Ishii et al , 2003. Briefly : Slides stored at -80
o
C, were dried in room temp for 20
145

mins followed by 10 mins of fixing at 4
o
C. Following fixation, slides meant for
Immunohistochemistry were treated with Peroxo block for 1 min and then washed with
PBST before. Blocking was done for 1 hr at Room Temp, 1
o
Antibody exposure was done
for 1 overnight and 2
o
Antibody exposure for 1 hr in Room Temp.
All slides meant for Rabbit primary antibody were blocked with 10%Goat serum and the
antibodies were diluted in the same. All slides meant for Goat primary antibody were
blocked with 1%BSA serum and the antibodies were diluted in the same.
Slides for immunohistochemistry were visualized with DAB chromogen and slides for
immunofluorescence were visualized by epifluorescence microscopy.
For immunochemistry, after detection of antibody with chromogen, slides are washed and
counterstained with Hematoxylin ( 30 sec), dH20 was, PBS ( 30 secs ) in succession ,
before a last dh20 wash, dehydration and mounting in mounting medium). For
immunofluorescence the slides are counterstained with DAPI and after washing the excess
DAPI with dH20, the slides are mounted with ProLong Gold Antifade reagent.

I. Measurements : Coronal suture index (CI)
In the modified version from Oram and Gridley (2005), Measurements: Coronal
suture index (CI)
Primary
Antibody
Company Source Dilution Fixing Secondary Counterstain
Jagged1 Santa Cruz Goat 1:300 4%PFA From Kit Hematoxylin
Runx2 Sigma Rabbit 1:200 Acetone From Kit Hematoxylin
146

Modified from Oram and Gridley ( 2005) , each suture in the 2 hemispheres are
individually scored for the relative length of suture fused . The scoring is as follows :
0 to fully patent suture, 0.5 for fusion equal to or less than 1/8 of the suture 1 for fusion
of equal to or less than ¼ of the suture, 2 for equal to or less than ½ , 3 for equal to or less
than ¾ and a 4 indicating a fully fused one. The scores for each type of suture is added
for the individual genotype sets and the Mean score and SD are calculated which are then
tabulated.

147

Chapter 3 [Preliminary data]
Stat3 signaling is active in a cluster of the migratory neural crest population in the
ectocranial region
My finding that craniosynostosis of the coronal suture is observed in neural crest specific
Stat3 mutants is significant since there has been no prior report of synostosis produced due
to neural crest specific inactivation of any other gene . But Stat3 signaling has not been
previously studied at this stage of embryonic development in context of coronal suture
previously. Only one report has shown that Stat3 is actively involved in the proliferation
and specification of the early neural crest population [ Nichane et al , 2010] . Since the
coronal suture begins to develop after neural crest cells have already undergone EMT and
are migrating, proliferating and differentiating to form the different aspects of the
craniofacial structures, I decided to investigate whether at this relatively late stage of
development neural crest, but early suture formation. Stat3 signaling was active in the
entire population of neural crest cells making up the frontal bone or not. Also since Notch2
which I have established to be downstream of Stat3 is exclusively expressed at the frontal
bone tip ( and not in its entirety) , I wanted to establish the location of cells with active
Stat3 in context of the suture while the latter gets established.
Activation of Stat3 involves phosphorylation at 2 residues; the more significant Tyr705
phosphorylation in the SH2 domain by JAK peptides and the lesser common Ser727
phosphorylation by protein kinase C. Upon Tyr705 phosphorylation STAT3 proteins are
able to homodimerize with the help of interaction between mirrored phosphotyrosine
148

containing SH2 domain and enter the nucleus. Active Stat3 signaling in cells is thus readily
identifiable by the presence of PTyr705Stat3 in the cells. [ Rebe et al 2013] .

149

Results
3.1 P-Stat3 positive cells cluster in the ectocranial layer adjacent to the frontal bone
In order to locate cells with active Stat3 signaling in the coronal suture area during early
suture formation, I performed immunofluorescence on horizontal sections of control and
Wnt1cre;Stat3
cko/cko
mouse heads from E14.5 embryos for detection of PTyr705Stat3 in and
around the developing frontal bone and coronal suture region. Sections adjacent to the
immunostained slides were stained for ALP to detect the location of the coronal suture and
the morphology of the frontal and parietal bone fronts relative to the suture.

150

fb
Wildtype
PStat
3
c
s
cs
fb
PStat
3
fb
fb
cs
cs
Wnt1Cre;Stat3
cko/cko

ALP ALP
ALP ALP
Fig3.1 : Assessment of PStat3 positive cells in horizontal section of E14.5 mice
show group of PStat3 positive cells in frontal area of Ectocranial layer (ECL) .
Fig 3.1: Assessment of PStat3 expression in Horizontal sections of mice heads
during early embryonic stages of coronal suture formation
A,B: Horizontal sections of Control and Wnt1Cre;Stat3
cko/cko
E13.5 mouse head
sections immunostained for detection of PStat3 in the coronal suture area .
A’,B’: Sections corresponding to A and B respectively, stained for ALP
A”,B”: Lower magnification pictures of A’ and B’ ( ALP) to show the relative locations of
the frontal and parietal bone tips and the position of the coronal suture.
pb: parietal bone, fb: frontal bone, cs : coronal suture
A : Anterior , P: Posterior
Arrowheads mark the anterio-posterior boundary of the cluster of PStat3 staining cells
A B
A
’
B
’
A” B”
pb
pb
E13.5
P A
151

Interestingly in the control (without cre) samples, no signal for P-Stat3 was detected in the
coronal suture region (mesoderm derived tissue). Nor did I detect any significant PStat3 in
the frontal bone primordia (neural crest tissue) either. On the other hand I detected a cluster
of heavily P-Stat3 positive cells in the ectocranial area close to the anterior tip of the
growingfrontal bone front (Fig3.1A). The layer in which I see these cells overlaps with the
zone where cranial neural crest cells forming the forntal bone migrate dorso-laterally as
previously established from DiI lalebling studies carried out in the lab. [Roybal et al 2010].
Besides, these clustered cells were only observed in a subset of the horizontal sections,
beginning in the region where the frontal bone tip is identifiable in adjacent sections via
ALP staining and stopping significantly above the eye level. This gives an indication that
a cluster of cells with active Stat3 signaling is present in the ecotcranial zone in the vicinity
of the of dorsal most region of the anterior part of the growing frontal bone primordium.
Notably such a cluster of cells is not identifiable in the neural crest-specific Stat3
(Wnt1cre;Stat3
cko/cko
) mutants. (Fig3.1B).
152

E13.5
PStat3
ALP
E14.5
PStat3
Wildtype Wildtype
Fig 3.2A: Assessment of PStat3 expression in Coronal (Dorso-ventral) sections
of mice heads during early embryonic stages of coronal suture formation
A,B: Coronal sections of Wildtype E13.5 (A) and E14.5 (B) mouse head sections
immunostained for detection of PStat3 in the ECL ( Ectocranial layer )
A’,B’: Sections corresponding to A and B respectively, stained for ALP to mark the
relative locations of the growing frontal bone tips
pb: parietal bone, fb: frontal bone, cs : coronal suture
D: dorsal end V: ventral end
Arrowheads mark the dorso-ventral boundary of the cluster of PStat3 staining cells
A
B A’ B’
ALP
Fig3.2A: Assessment of PStat3 positive cells in Coronal (Dorso-ventral)
section of E13.5 and E14.5 mice show group of PStat3 positive cells in frontal
area of Ectocranial (ECL) layer close to the growing frontal bone front

fb
fb
D
V
153

Wnt1Cre; Stat3
cko/cko

Wildtype
E
Fig3.3B: Assessment of PStat3 positive cells in Coronal section of E13.5 and
E14.5 mice showing group of PStat3 positive cells in Ectocranial (ECL) layer
close to the growing frontal bone front only in Wildtype mice
PStat3
PStat3
ALP PStat3
ALP
B
D B
’
B” D
’
fb
fb
PStat3
PStat3
ALP PStat3 ALP
A C A
’
A” C
’
fb
fb
E13.5 E14.5
D
V
154

Fig 3.3B: Assessment of PStat3 expression in Coronal (Dorso-ventral) sections of
mice heads during early embryonic stages of coronal suture formation
A,B, C,D :Coronal sections of Wildtype (A,B) and Wnt1Cre;Stat3
cko/cko
(C,D) E13.5 and
E14.5 mouse head sections immunostained for detection of PStat3 in the ECL (
Ectocranial layer )
A’,B’: Higher magnification pictures of annotated area of A and B
A”,B’,C”,D’: Sections corresponding to A,B,C and D respectively, stained for ALP to
mark the relative locations of the growing frontal bone tips
pb: parietal bone, fb: frontal bone, cs : coronal suture ; D: dorsal end V: ventral end
E: Comparative study of average number of PStat3 cells observable per section in
Wildtype and Wnt1Cre;Stat3
cko/cko,
E13.5 and E14.5 mouse head coronal sections
155

3.2 Assessment of P-Stat3 positive cells in coronal section of mice heads at successive
embryonic stages show cluster of PStat3 positive cells in the ectocranial layer are
absent in the neural crest specific Stat3 mutants
As discussed in the previous section, immunofluorescence studies to locate PStat3
expressing tissue in E14.5 mouse heads led to the identification of a cluster of cells in the
ectocranial layer close to the anterior tip of the growing frontal bone primordia.
In order to confirm these data and correctly identify the location of these cells E13.5 and
E14.5 mouse heads were sectioned coronally and immunofluorescence studies were carried
out for identification of PStat3 with respect to the dorso-ventral aspect of the frontal and
parietal bones. Sections adjacent to the immunostained slides were stained with ALP to
correctly detect the location and morphology of the frontal and parietal bone fronts realtive
to the ectocranial layer, the base of the bone primordia and the eye.
As in the previous case, in the wildtype samples of both E13.5 and E14.5 stage mouse
coronal sections, I was able to identify a cluster of P-Stat3 staining cells in the ectocranial
layer between the skin and frontal bone primordia (as marked by the ALP stained tissue in
the corresponding slides) (Fig3.2A, A’, B, B’), (Fig3.3A, A’, A”, B, B’, B”).
It was also noted that the cluster of cells do not extend throughout the ectocranial layer but
appear to have distinct dorso-ventral boundaries coinciding closely with the dorsal-most
growth zone in the corresponding frontal tip at each successive stage.
156

These clustered cells were not present in the neural crest-specific Stat3 knockdown
(Wnt1cre;Stat3
cko/cko
) samples of E13.5, while an extremely low number of individual cells
showing possible P-Stat3 signaling can be observed at the E14.5 stage (Fig3.3:C,D).
To get a quantitative idea of the PStat3 positive cell population, the total number of cells
in the clusters of each section were counted for more than 5 random sections in each
wildtype and mutant sample and the average number of such cells per section was
computed . A comparison of these numbers at successive stages of embryonic development
is displayed in the corresponding bar graph (Fig3.3: E).
A Student’s T distribution of the number of cells/ section yields p values of 0.011 (E13.5)
and 0.012 (E14.5) for a single pair of wildtype and mutant samples compared at each stage.

157

Wildtype
Wnt1Cre; Stat3
cko/cko

PH3 PH3
Fig3.3: Assessment of proliferative cells (PH3 positive) in Coronal section of
E13.5 mice in the growing neural crest derived frontal bone primordium area
A A’ B B’
C
fb
fb
fb
fb
Fig 3.3: Assessment of PH3 positive cells in the growing frontal bone primordia
in E13.5 mice heads during early embryonic stage of coronal suture formation
A,B: Coronal sections of Wildtype and Wnt1Cre;Stat3
cko/cko,
E13.5 mouse head
sections immunostained for detection of PH3 ( proliferative cell marker) in the frontal
one primordia
A’,B’: Higher magnification pictures of annotated area of A and B
C: Comparative study of average number of PH3 positive cells observable per section
in the growing frontal bone primordia of Wildtype, Wnt1Cre; Stat3
cko/wt
and
Wnt1Cre;Stat3
cko/cko,
E13.5 mouse head coronal sections
pb: parietal bone, fb: frontal bone, cs : coronal suture
D: dorsal end V: ventral end
Arrows mark the zone of proliferative cells
D
V
158

3.3 Analysis of PH3 positive cells in the growing frontal bone primordia in E13.5
mouse heads during early the embryonic stage of coronal suture formation shows no
significant reduction in the number of proliferative cells in neural crest-specific Stat3
mutants.
The previous observation of a cluster of P-Stat3 labelled cells in the ectocranial layer of
wildtype two week old mouse embryo heads, which is absent in the Wnt1cre;Stat3
cko/cko

mutants led me to investigate whether this loss resulted from a reduced proliferation rate
of the neural crest cells in the mutants.
For this purpose, E13.5 staged coronal sections corresponding to the P-Stat3
immunofluorescently labeled sections used in the previous experiment were
immunostained for the detection of the proliferative cell marker PH3 (Phospho-Histone3),
which marks mitotic cells (Fig3.4: A, A’, B, B’).
The total number of cells in the frontal bone primordia in all stained sections of wildtype,
Wnt1cre;Stat3
cko/
and Wnt1cre;Stat3
cko/cko
samples were counted and the average number
of cells per section was computed, the comparison of which is displayed in the
corresponding bar graph (Fig3.4: C). I did not find any significant difference in the total
number of proliferative cells in the neural crest derived tissue between the three different
genotypes.

159

Discussion
Stat3 signaling is active in the migratory neural crest population in the ectocranial
region
Here for the first time I show that Stat3 signaling (identified by immunostaining for
PTyr705Stat3) is active in a cluster of cells in the ectocranial layer. These cells are located
close to the dorsal most tip of the frontal bone. This area may be an active growth zone
where a new set of cells emigrate from the ectocranial layer to the tip of the growing frontal
bone edge before expressing their osteogenic fate. These cells appear to be positioned in a
region significantly anterior to the coronal suture and just anterior to the line of the eye.
These cells do not appear to be actively proliferating since I do not find Phospho-Histone3
(PH3) activity in the vicinity of these cells. The major set of proliferative cells appears to
be in the basal region of the frontal bone. I found no significant difference in the average
number of proliferative cells between wildtype and homozygous or heterozygous neural
crest specific Stat3 mutant samples, As expected, the cluster of cells staining for active
Stat3 signaling is absent in homozygous (not heterozygous ) Stat3 conditional loss mutants,
confirming the specificity of the P-Stat3 antibody
These data suggest that PStat3 may be a marker for a subset of migratory neural crest
population emigrating in the dorso-lateral direction from the frontal bone rudiment.
I intend to carry out DiI labeling studies in order to confirm this idea. I also plan to examine
wildtype mouse heads at preceding and later embryonic stages to identify the time point
when these cells are first detectable as well as their ultimate fate.
160

MATERIALS AND METHODS
A. Mice & Genotyping
All transgenic mice used for this work has been described previously. They are as
follows,
Wnt1cre [Danielian et al 1998]
Stat3
flox/flox
[Takeda et al 1998]

Genotyping Primers:
Cre F: TGC TGT TTC ACT GGT TAT GCG G
R: TTG CCC CTG TTT CAC TAT CCA G
Stat3 flox F (wt) : CCT GAA GAC CAA GTT CAT CTG TGT GAC
F (flox) : TTT GGA AAG TAC TGT AGC CCC GAG AGC
R : CAC ACA AGC CAT CAA ACT CTG GTC TCC

B. Embedding and Cryosectioning of embryonic heads
Embryos at required embryonic stages were dissected out in cold PBS, fixed in 4%PFA
for 30 mins, and washed in PBS followed by sucrose gradient ( 15%, 30%) and OCT
embedding . Sectioning was done at -20
o
C to produce sections 10um apart in multiple
corresponding slides and stored in -80
o
C until used for staining.

161

C. ALP staining of sections
Slides stored at -80
o
C, were dried in room temp for 10 mins followed by 10 mins of
Acetone fixation at 4
o
C. For staining, the sections were treated with ice cold TBST (10
mins ) , NTMT (10 mins ) and 0.01%BCIP-0.025%NBT in NTMT solution ( 30 secs-
1min) in succession. Excess stained was drained and slides were washed with dH20
followed by counterstaining with dilute NFR (nuclear fast red soln) for about 30 secs.
Excess NFR was then drained off and the slides were washed with dH20 before
dehydrating in a gradient of increasing % of Ethanol solution and mounted. Post mounting,
the slides were dried for at least 1 overnight before being photographed.
D. LacZ staining of sections
Slides stored at -80
o
C, were dried in room temp for 10 mins followed by 10 mins of fixing
Glutaraldehyde soln (/ PBS/2mMMgCl2/0.02% Glutaraldehyde). After fixation, the slides
are washed with ice cold PBS for 10 mins. For staining, the slides are then treated with (i)
ice cold PBS/2mMMgCl2 solution for 10 mins , followed by (ii) ice cold PBS/2mMMgCl2/
0.01%DOC/0.02%NP40 for 10 mins and eventually submerged in LacZ staining solution
(1M MgCl2/ %%DOC / 2%NP-40 / 2% X-Gal soln / 1.05%FeII/ 0.825%FeIII in PBS) and
rocked in 37C overnight. The next day the slides are first allowed to come room
temperature , following which they are removed from the staining solution and washed
with fresh PBS 3 times in succession , till all excess stain is removed. The slides are then
post fixed with 4%PFA for about 2 hrs followed by PBS wash, counterstain with NFR ( 30
secs) , dH20 wash in succession. Finally the slides were dehydrated in a gradient of
162

increasing % of Ethanol solution and mounted. Post mounting, the slides were dried for at
least 1 overnight before being photographed.
E. Immunohistochemistry & immunofluorescence
Immunostaining for detection of various proteins was carried out as previously described
by Ishii et al , 2003. Briefly :
Slides stored at -80
o
C, were dried in room temp for 20 mins followed by 10 mins of fixing
at 4
o
C. Following fixation, slides meant for Immunohistochemistry were treated with
Peroxo block for 1 min and then washed with PBST before. Blocking was done for 1 hr at
Room Temp, 1
o
Antibody exposure was done for 1 overnight and 2
o
Antibody exposure for
1 hr in Room Temp.
All slides meant for Rabbit primary antibody were blocked with 10%Goat serum and the
antibodies were diluted in the same. All slides meant for Goat primary antibody were
blocked with 1%BSA serum and the antibodies were diluted in the same.
Slides for immunohistochemistry were visualized with DAB chromogen and slides for
immunofluorescence were visualized by epifluorescence microscopy.
For immunochemistry, after detection of antibody with chromogen, slides are washed and
counterstained with Hematoxylin ( 30 sec), dH20 was, PBS ( 30 secs ) in succession ,
before a last dh20 wash, dehydration and mounting in mounting medium)
For immunofluorescence the slides are counterstained with DAPI and after washing the
excess DAPI with dH20, the slides are mounted with ProLong Gold Antifade reagent.
163

F. Measurements : Cell counting

1) PStat3 : Total number of cells staining for PStat3 were counted in Wildtype and
Mutant samples for equal to or greater than 5 sections per sample . The Mean
number of cells per section and SD of the sample distribution was the computed
and displayed in the graphical form . A paired , two-tailed Student’s T test was
conducted on the raw data to calculate the P value.
2) PH3 : Total number of cells staining for PH3 in the frontal bone primordia zone
were counted in Wildtype and Mutant samples for equal to or greater than 10
sections. The Mean number of cells per section and SD of the sample distribution
was the computed and displayed in the graphical form .

Primary
Antibody
Company Source Dilution Fixing Secondary Counterstain
PTyr705 Cell
signaling
Rabbit 1:50 Acetone FITC
labelled
Anti-
Rabbit
DAPI
PH3 Upstate Rabbit 1:200 Acetone From Kit Hematoxylin
164

SUMMARY DISCUSSION AND FUTURE DIRECTIONS
In summary, I show here for the first time that neural crest specific loss of Stat3 signaling
leads to incomplete bilateral fusion of the coronal suture.
This is significant since this is the first time only neural crest specific loss of any signaling
molecule has been reported to display synostosis of the coronal suture. the coronal suture
not only defines a boundary between 2 flat bones , but also signifies a boundary between
neural crest ( frontal bone ) and mesodermal ( coronal suture) tissue, and till date all cases
of fusion in coronal suture has been attributed to disruption of signaling tin the suture tissue
and not the adjacent neural crest tissue.
As discussed in the introduction of Chapter1, the HIES (Hyper IgE syndrome) which
displays synostosis as one of its clinical features, has been shown to be caused due to loss
of functional Stat3. [Minegishi et al 2006]. It is notable here that the main clinical reports
describing the craniosynostosis phenotype in HIES patients have described the patients to
show synostosis in different combinations of the coronal, sagittal and lambdoid sutures to
varying degrees. [Hoger et al 1985, Smithwick et al 1978]. From Wnt1cre labeling done
by Maxson lab earlier, it has been shown that all 3 of these sutures represent a neural crest-
mesoderm boundary. While in the coronal suture the frontal bone is entirely made up of
neural crest tissue and the adjacent coronal suture tissue is mesodermal, in the lambdoid
suture, the adjacent interparietal bone has significant contribution from the neural crest
while the suture itself is of mesodermal origin. On the other hand the sagittal suture itself
has contribution from the neural crest [Jiang et al 2002]. In my mice studies I observe
165

fusion in only the coronal suture which actually represents the most significant crest-
mesoderm boundary with the frontal bone being made entirely out of crest cells.
I also show that in the neural crest specific Stat3-/- mutants, bone fusion is observable early
during suture formation and that during this stage, the fusion phenotype is of a distinct type
not previously observed for any other mutant. There is loss of directional specificity of the
frontal bone tip as well as an ectopic bridging between frontal and parietal bone tips.
I also show that a neural crest display a migration defect with a small number of distinctive
cells infiltrating into the suture mesenchyme, disobeying the crest-mesoderm boundary.
But also that these cells do not account for the entire set of osteogenic cells involved in
bridging the frontal and parietal bone tips, thus indicating that the fusion involves non
autonomous ectopic osteogenic activity of the cells surrounding the infiltrating cells.
Supporting this idea I have shown that there significant increase in Runx2 (early osteogenic
marker) activity in the suture region surrounding the growing frontal bone and extending
above the suture zone up to the parietal bone tip.
I also show that neural crest specific Stat3-/- have a dramatic reduction of Notch2 receptor
in the frontal bone tissue , with no changes observable in any of the mesodermal markers .
From these studies I conclude that neural crest specific loss of Stat3 leads to loss of Notch2
signaling in the neural crest derived cells, leading to their mismigration into the coronal
suture area and ensuing nonautonomous ectopic osteogenic activity in a section of the
surrounding suture cells and eventual bone fusion.
166

In Chapter 2 I have shown that neural crest specific loss of Notch2 is indeed able to cause
coronal suture fusion, which is of a greater severity that in the case of Stat3-/- mutants and
that Notch mutants show mismigration of neural crest cells and increased Runx2 activity
in the suture as well. This further supports the idea that Notch2 acts downstream of Stat3
in the neural crest cells. Finally I was able to show that constitutively active Notch signaling
was able to rescue the suture fusion phenotype in the Stat3 mutants, thus proving that
indeed Notch2 signaling acts downstream of Stat3 in neural crest cells making the frontal
bone and that the fusion observed in Stat3 mutants is caused due to the ensuing loss of
Notch2 in the tissue.
This is also the first time that it has been shown that loss of Notch2 signaling leads to
synostosis of the coronal suture. I specifically show here that it is the neural crest specific
loss of Notch2 and not mesoderm specific loss of the same that is able to cause the fusion
phenotype.
In the final section I have shown that the strongest active Stat3 signaling is observable in a
cluster of the cells in the ectocranial layer ( ECL) , located close to the anterior side of the
growing frontal bone tip at each successive stage and that this population is distinctly
absent in the Stat3 mutants. In future I wish to follow up these studies to locate the earliest
time point at which this population rises and their location at that point. In order to see
whether this population overlaps with the migrating neural crest cells I plan to perform DiI
labeling studies to concurrently label the neural crest cells as they begin to migrate
dorsolaterally along the ectocranial layer to contribute to the growing frontal bone.
167

In conclusion, my studies open up a new field of contribution of the non sutural tissue in
causing suture synostosis, especially the role of proper development of the neural crest
derived frontal bone in the establishment and maintenance of patency of the coronal suture.
It also establishes a newly found role of Stat3-Notch2 signaling mechanism in the normal
development of the frontal bone and the adjacent coronal suture, adding to our knowledge
of etiology and probable mechanism of various types of synostosis.

168

SUPPLEMENTARY DATA

Fig S1: Visual Correlation between locations of LacZ stained section with individual
cells infiltrating the suture and closest ALP sections displaying bone fusion
Samples studied :

No. Suture details Sample Details Sample No.
1a. E14.5 –Suture 1 E14.5 – Mutant1 –Suture1 (Right) 19B.7(1)
1b. E14.5 –Suture 2 E14.5 – Mutant1 –Suture2 (Left) 19B.7(1)
2a. E14.5 –Suture 3 E14.5 – Mutant2 –Suture1 (Right) 19B.7(6)
2b. E14.5 –Suture 1 E14.5 – Mutant2 –Suture2 (Left) 19B.7(6)
3a. E16.5 –Suture 1 E16.5 – Mutant1 –Suture1 (Left) 19B.21(4)
3b. E16.5 –Suture 2 E16.5 – Mutant1 –Suture2 (Right) 19B.21(4)
4a. E17.5 –Suture 1 E17.5 – Mutant1 –Suture1 (Right) 19B.15(7)
4b. E17.5 –Suture 2 E17.5 – Mutant1 –Suture2 (Left) 19B.15(7)
169

170

171

172

173

174

175

176

177

BIBLIOGRAPHY
Books & Book Chapters
Daniel E. Lieberman ; The Evolution of the Human Head; Belknap Press ;2011. ISBN-
13: 978-0674046368
David P. Rice (Editor); Craniofacial Sutures: Development, Disease and Treatment
(Frontiers of Oral Biology); Karger, 2008. ISBN-13: 978-3805583268

David John; David Poswillo; and Donald Simpson ; The Craniosynostoses: Causes,
Natural History, and Management. Springer-Verlag,1982. ISBN-13: 978-1447113256

Hanken, J. and B.K. Hall, eds. 1993. [Introduction] The Skull: Vol. 3, Functional and
Evolutionary Mechanisms. University of Chicago Press, Chicago. ISBN-13: 978-
0226315737

Connerney J.J., Spicer D.B. Signal Transduction Pathways and Their Impairment in
Syndromic Craniosynostosis [Chap 4: Muenke M, Kress W, Collmann H, Solomon BD
(eds): Craniosynostoses: Molecular Genetics, Principles of Diagnosis, and Treatment.
Monogr .Hum Genet. Basel, Karger, 2011, vol 19, pp 28–44] ISBN-13: 978-3805595940

178

REFERENCES

Akutsu K, Morisaki H, Takeshita S, Sakamoto S, Tamori Y, Yoshimuta T, Yokoyama N,
Nonogi H, Ogino H, Morisaki T. Phenotypic heterogeneity of Marfan-like connective
tissue disorders associated with mutations in the transforming growth factor-beta
receptor genes. Circ J. 2007 Aug;71(8):1305-9, PMID: 17652900

Alagille D, Odièvre M, Gautier M, Dommergues JP. Hepatic ductular hypoplasia
associated with characteristic facies, vertebral malformations, retarded
physical, mental, and sexual development, and cardiac murmur. J Pediatr. 1975
Jan;86(1):63-71.PMID: 803282

Artavanis-Tsakonas S, Matsuno K, Fortini ME. Notch signaling. Science. 1995 Apr
14;268(5208):225-32. Review. PMID: 7716513

Artavanis-Tsakonas S, Rand MD, Lake RJ. Notch signaling: cell fate control and
signal integration in development. Science. 1999 Apr 30;284(5415):770-6.
Review. PMID: 10221902

Au PY, Racher HE, Graham JM Jr, Kramer N, Lowry RB, Parboosingh JS, Innes AM;
FORGE Canada Consortium. De novo exon 1 missense mutations of SKI and
Shprintzen-Goldberg syndrome: two new cases and a clinical review. Am J Med Genet
2014 Mar;164A(3):676-84. doi: 10.1002/ajmg.a.36340. Epub 2013 Dec 19.PMID:
24357594
179

Baroni T, Carinci P, Lilli C, Bellucci C, Aisa MC, Scapoli L, Volinia S, Carinci
F, Pezzetti F, Calvitti M, Farina A, Conte C, Bodo M. P253R fibroblast growth
factor receptor-2 mutation induces RUNX2 transcript variants and calvarial
osteoblast differentiation. J Cell Physiol. 2005 Feb;202(2):524-35.PubMed
[citation] PMID: 15389579

Bialek P, Kern B, Yang X, Schrock M, Sosic D, Hong N, Wu H, Yu K, Ornitz DM,
Olson EN, Justice MJ, Karsenty G. A twist code determines the onset of osteoblast
differentiation. Dev Cell. 2004 Mar;6(3):423-35. PMID: 15030764

Bolós V, Grego-Bessa J, de la Pompa JL. Notch signaling in development and
cancer. Endocr Rev. 2007 May;28(3):339-63. Epub 2007 Apr 4. Review. PMID:
17409286

Brault V, Moore R, Kutsch S, Ishibashi M, Rowitch DH, McMahon AP, Sommer L,
Boussadia O, Kemler R. Inactivation of the beta-catenin gene by Wnt1-Cre-mediated
deletion results in dramatic brain malformation and failure of craniofacial
development. Development. 2001 Apr;128(8):1253-64. PMID: 11262227

Brewer S, Feng W, Huang J, Sullivan S, Williams T.Wnt1-Cre-mediated deletion of
AP-2alpha causes multiple neural crest-related defects. Dev Biol. 2004 Mar 1;267(1):135
52. PMID: 14975722

180

Brighton CT, Hunt RM. Early histological and ultrastructural changes in medullary
fracture callus. J Bone Joint Surg Am. 1991 Jul;73(6):832-47. PMID: 2071617

Carmignac V, Thevenon J, Adès L, Callewaert B, Julia S, Thauvin-Robinet C,
Gueneau L, Courcet JB, Lopez E, Holman K, Renard M, Plauchu H, Plessis G, De
Backer J, Child A, Arno G, Duplomb L, Callier P, Aral B, Vabres P, Gigot N,
Arbustini E, et al. In-frame mutations in exon 1 of SKI cause dominant
Shprintzen-Goldberg syndrome. Am J Hum Genet. 2012 Nov 2;91(5):950-7. doi:
10.1016/j.ajhg.2012.10.002. Epub 2012 Oct 25. PMID: 23103230, PMCID:
PMC3487125

Casal J, Leptin M.Identification of novel genes in Drosophila reveals the complex
regulation of early gene activity in the mesoderm.Proc Natl Acad Sci U S A. 1996
Sep 17;93(19):10327-32. PMID: 8816799, PMCID: PMC38383

Chai Y, Maxson RE Jr. Recent advances in craniofacial morphogenesis. Dev Dyn. 2006
Sep;235(9):2353-75. Review. PMID: 16680722

Chen L, Li D, Li C, Engel A, Deng CX.A Ser252Trp [corrected] substitution in
mouse fibroblast growth factor receptor 2 (Fgfr2) results in craniosynostosis. Bone. 2003
Aug;33(2):169-78. Erratum in: Bone. 2005,Dec;37(6):876. PubMed [citation] PMID:
14499350

Cheng SL, Shao JS, Cai J, Sierra OL, Towler DA. Msx2 exerts bone anabolism via
181

canonical Wnt signaling. J Biol Chem. 2008 Jul 18;283(29):20505-22. doi:
10.1074/jbc.M800851200. Epub 2008 May 15. PMID: 18487199, PMCID: PMC2459292

Clendenning DE, Mortlock DP. The BMP ligand Gdf6 prevents differentiation of
coronal suture mesenchyme in early cranial development. PLoS One.
2012;7(5):e36789. doi: 10.1371/journal.pone.0036789. Epub 2012 May 31. PMID:
22693558, PMCID: PMC3365063

Cohen MM Jr. Sutural biology and the correlates of craniosynostosis. Am J Med
Genet. 1993 Oct 1;47(5):581-616. PMID: 8266985

Cohen MM Jr. Editorial: perspectives on craniosynostosis. Am J Med Genet A. 2005
Aug 1;136A(4):313-26. Review. PMID: 15999346

Connerney J, Andreeva V, Leshem Y, Muentener C, Mercado MA, Spicer DB. Twist1
dimer selection regulates cranial suture patterning and fusion. Dev Dyn. 2006
May;235(5):1345-57. Erratum in: Dev Dyn. 2012 Feb;241(2):433. PMID: 16502419

Connerney J, Andreeva V, Leshem Y, Mercado MA, Dowell K, Yang X, Lindner V,
Friesel RE, Spicer DB. Twist1 homodimers enhance FGF responsiveness of the cranial
sutures and promote suture closure. Dev Biol. 2008 Jun 15;318(2):323-34. doi:
10.1016/j.ydbio.2008.03.037. Epub 2008 Apr 8. Erratum in: Dev Biol. 2012 Jun
15;366(2):434. PMID: 18471809, PMCID: PMC2605972

182

Couly GF, Coltey PM, Le Douarin NM. The developmental fate of the cephalic
mesoderm in quail-chick chimeras. Development. 1992 Jan;114(1):1-15. PMID:
1576952

Coussens AK, Wilkinson CR, Hughes IP, Morris CP, van Daal A, Anderson PJ, Powell
BC. Unravelling the molecular control of calvarial suture fusion in children with
craniosynostosis. BMC Genomics. 2007 Dec 12;8:458. PMID:18076769, PMCID:
PMC2222648

Cunningham ML, Horst JA, Rieder MJ, Hing AV, Stanaway IB, Park SS, Samudrala R,
Speltz ML.IGF1R variants associated with isolated single suture craniosynostosis. Am J
Med Genet A. 2011 Jan;155A(1):91-7. doi: 10.1002/ajmg.a.33781.PubMed [citation]
PMID: 21204214, PMCID: PMC3059230

Doyle AJ, Doyle JJ, Bessling SL, Maragh S, Lindsay ME, Schepers D, Gillis E,
Mortier G, Homfray T, Sauls K, Norris RA, Huso ND, Leahy D, Mohr DW, Caulfield
MJ, Scott AF, Destrée A, Hennekam RC, Arn PH, Curry CJ, Van Laer L, McCallion AS,
et al. Mutations in the TGF-β repressor SKI cause Shprintzen-Goldberg syndrome
with aortic aneurysm. Nat Genet. 2012 Nov;44(11):1249-54. doi: 10.1038/ng.2421.
Epub 2012 Sep 30. PMID: 23023332, PMCID: PMC3545695

el Ghouzzi V, Le Merrer M, Perrin-Schmitt F, Lajeunie E, Benit P, Renier D,
Bourgeois P, Bolcato-Bellemin AL, Munnich A, Bonaventure J. Mutations of the TWIST
gene in the Saethre-Chotzen syndrome. Nat Genet. 1997 Jan;15(1):42-6. PMID: 8988167
183

Eswarakumar VP, Monsonego-Ornan E, Pines M, Antonopoulou I, Morriss-Kay GM,
Lonai P. The IIIc alternative of Fgfr2 is a positive regulator of bone formation.
Development. 2002 Aug;129(16):3783-93. PMID: 12135917

Feng W, Choi I, Clouthier DE, Niswander L, Williams T. The Ptch1(DL) mouse: a new
model to study lambdoid craniosynostosis and basal cell nevus syndrome-associated
skeletal defects. Genesis. 2013 Oct;51(10):677-89. doi: 10.1002/dvg.22416. Epub
2013 Aug 30.PubMed [citation] PMID: 23897749, PMCID: PMC3918964

Florisson JM, Verkerk AJ, Huigh D, Hoogeboom AJ, Swagemakers S, Kremer A,
Heijsman D, Lequin MH, Mathijssen IM, van der Spek PJ. Boston type craniosynostosis:
report of a second mutation in MSX2.Am J Med Genet A. 2013 Oct;161(10):2626-33.
doi: 10.1002/ajmg.a.36126. Epub 2013 Aug 15. PMID: 23949913

Freeman AF, Holland SM. The hyper-IgE syndromes. Immunol Allergy Clin North Am.
2008 May;28(2):277-91, viii. doi: 10.1016/j.iac.2008.01.005. Review. PMID: 18424333,
PMCID: PMC2683262

Gans C, Northcutt RG. Neural crest and the origin of vertebrates: a new head. Science.
1983 Apr 15;220(4594):268-73. PMID: 17732898

Goldbeter A, Pourquié O. Modeling the segmentation clock as a network of coupled
oscillations in the Notch, Wnt and FGF signaling pathways. J Theor Biol. 2008 Jun
184

7;252(3):574-85. doi: 10.1016/j.jtbi.2008.01.006. Epub 2008 Jan 18. PMID: 18308339

Gridley T. Notch signaling and inherited disease syndromes. Hum Mol Genet. 2003 Apr
1;12 Spec No 1:R9-13. Review. PMID: 12668592

Guenou H, Kaabeche K, Mée SL, Marie PJ.A role for fibroblast growth factor receptor-2
in the altered osteoblast phenotype induced by Twist haploinsufficiency in the Saethre-
Chotzen syndrome. Hum Mol Genet. 2005 Jun1;14(11):1429-39. Epub 2005 Apr 13.
PMID: 15829502

Harada M, Murakami H, Okawa A, Okimoto N, Hiraoka S, Nakahara T, Akasaka R,
Shiraishi Y, Futatsugi N, Mizutani-Koseki Y, Kuroiwa A, Shirouzu M, Yokoyama S,
Taiji M, Iseki S, Ornitz DM, Koseki H. FGF9 monomer-dimer equilibrium regulates
extracellular matrix affinity and tissue diffusion. Nat Genet. 2009 Mar;41(3):289-98. doi:
10.1038/ng.316. Epub 2009 Feb 15. PMID: 19219044, PMCID: PMC2676118

Heinen T, M Vinken P. Monocular and binocular vision in the performance of a
complex skill. J Sports Sci Med. 2011;10(3):520-7. PMID: 24150627, PMCID:
PMC3737810

Helms JA, Cordero D, Tapadia MD. New insights into craniofacial morphogenesis.
Development. 2005 Mar;132(5):851-61. Review. Erratum in: Development. 2005
Jun;132(12):2929. PMID: 15705856

185

Höger PH, Boltshauser E, Hitzig WH. Craniosynostosis in hyper-IgE-syndrome. Eur J
Pediatr. 1985 Nov;144(4):414-7. PMID: 4076261

Holland SM, DeLeo FR, Elloumi HZ, Hsu AP, Uzel G, Brodsky N, Freeman AF,
Demidowich A, Davis J, Turner ML, Anderson VL, Darnell DN, Welch PA, Kuhns DB,
Frucht DM, Malech HL, Gallin JI, Kobayashi SD, Whitney AR, Voyich JM, Musser JM,
Woellner C, et al. STAT3 mutations in the hyper-IgE syndrome. N Engl J Med. 2007
Oct 18;357(16):1608-19. Epub 2007 Sep 19. PMID: 17881745

Howard TD, Paznekas WA, Green ED, Chiang LC, Ma N, Ortiz de Luna RI, Garcia
Delgado C, Gonzalez-Ramos M, Kline AD, Jabs EW. Mutations in TWIST, a basic
helix-loop-helix transcription factor, in Saethre-Chotzen syndrome. Nat Genet. 1997
Jan;15(1):36-41.PubMed [citation] PMID: 8988166

Hurst JA, Jenkins D, Vasudevan PC, Kirchhoff M, Skovby F, Rieubland C, Gallati S,
Rittinger O, Kroisel PM, Johnson D, Biesecker LG, Wilkie AO. Metopic and sagittal
synostosis in Greig cephalopolysyndactyly syndrome: five cases with intragenic
mutations or complete deletions of GLI3. Eur J Hum Genet. 2011 Jul;19(7):757-62.
doi: 10.1038/ejhg.2011.13. Epub 2011 Feb 16. PMID: 21326280, PMCID: PMC3128494

Iseki S, Wilkie AO, Heath JK, Ishimaru T, Eto K, Morriss-Kay GM. Fgfr2 and
osteopontin domains in the developing skull vault are mutually exclusive and can
be altered by locally applied FGF2. Development. 1997 Sep;124(17):3375-84. PMID:
9310332
186

Ishii M, Merrill AE, Chan YS, Gitelman I, Rice DP, Sucov HM, Maxson RE Jr. Msx2
and Twist cooperatively control the development of the neural crest-derived
skeletogenic mesenchyme of the murine skull vault. Development. 2003
Dec;130(24):6131-42. PMID: 14597577

Ishii M, Han J, Yen HY, Sucov HM, Chai Y, Maxson RE Jr. Combined deficiencies of
Msx1 and Msx2 cause impaired patterning and survival of the cranial neural
crest. Development. 2005 Nov;132(22):4937-50. Epub 2005 Oct 12. PMID: 16221730

Isidor B, Lindenbaum P, Pichon O, Bézieau S, Dina C, Jacquemont S,
Martin-Coignard D, Thauvin-Robinet C, Le Merrer M, Mandel JL, David A, Faivre L,
Cormier-Daire V, Redon R, Le Caignec C. Truncating mutations in the last exon of
NOTCH2 cause a rare skeletal disorder with osteoporosis. Nat Genet. 2011 Mar
6;43(4):306-8. doi: 10.1038/ng.778. PMID: 21378989

Jabs EW, Müller U, Li X, Ma L, Luo W, Haworth IS, Klisak I, Sparkes R, Warman ML,
Mulliken JB, et al. A mutation in the homeodomain of the human MSX2 gene in a
family affected with autosomal dominant craniosynostosis. Cell. 1993 Nov 5;75(3):443-
50. PMID: 8106171

Jeong J, Mao J, Tenzen T, Kottmann AH, McMahon AP. Hedgehog signaling in the
neural crest cells regulates the patterning and growth of facial primordia. Genes
Dev. 2004 Apr 15;18(8):937-51. PMID: 15107405, PMCID: PMC395852
187

Jiang X, Iseki S, Maxson RE, Sucov HM, Morriss-Kay GM. Tissue origins and
interactions in the mammalian skull vault. Dev Biol. 2002 Jan 1;241(1):106-16. PMID:
11784098

Jiang X, Rowitch DH, Soriano P, McMahon AP, Sucov HM. Fate of the mammalian
cardiac neural crest. Development. 2000 Apr;127(8):1607-16.PubMed [citation] PMID:
10725237

Johnson D, Wilkie AO. Craniosynostosis. Eur J Hum Genet. 2011 Apr;19(4):369-76.
doi: 10.1038/ejhg.2010.235. Epub 2011 Jan 19. PMID: 21248745, PMCID:
PMC3060331

Johnson D, Iseki S, Wilkie AO, Morriss-Kay GM. Expression patterns of Twist and
Fgfr1, -2 and -3 in the developing mouse coronal suture suggest a key role for twist in
suture initiation and biogenesis. Mech Dev. 2000 Mar 1;91(1-2):341-5. PMID: 10704861

Kamath BM, Stolle C, Bason L, Colliton RP, Piccoli DA, Spinner NB, Krantz
ID. Craniosynostosis in Alagille syndrome. Am J Med Genet. 2002 Oct
1;112(2):176-80. PMID: 12244552

Keupp K, Li Y, Vargel I, Hoischen A, Richardson R, Neveling K, Alanay Y, Uz E,
Elcioğlu N, Rachwalski M, Kamaci S, Tunçbilek G, Akin B, Grötzinger J, Konas E,
Mavili E, Müller-Newen G, Collmann H, Roscioli T, Buckley MF, Yigit G, Gilissen
188

C, et al. Mutations in the interleukin receptor IL11RA cause autosomal recessive
Crouzon-like craniosynostosis. Mol Genet Genomic Med. 2013 Nov;1(4):223-37. doi:
10.1002/mgg3.28. Epub 2013 Aug 19. PMID: 24498618, PMCID: PMC3865590

Kidd S, Kelley MR, Young MW. Sequence of the notch locus of Drosophila
melanogaster: relationship of the encoded protein to mammalian clotting and growth
factors. Mol Cell Biol. 1986 Sep;6(9):3094-108. PMID: 3097517, PMCID: PMC367044

Kim HJ, Rice DP, Kettunen PJ, Thesleff I.FGF-, BMP- and Shh-mediated signalling
pathways in the regulation of cranial suture morphogenesis and calvarial bone
development. Development. 1998 Apr;125(7):1241-51. PMID: 9477322

Klopocki E, Lohan S, Brancati F, Koll R, Brehm A, Seemann P, Dathe K, Stricker S,
Hecht J, Bosse K, Betz RC, Garaci FG, Dallapiccola B, Jain M, Muenke M, Ng VC,
Chan W, Chan D, Mundlos S. Copy-number variations involving the IHH locus are
associated with syndactyly and craniosynostosis. Am J Hum Genet. 2011 Jan
7;88(1):70-5. doi: 10.1016/j.ajhg.2010.11.006. Epub 2010 Dec 17. PMID: 21167467,
PMCID: PMC3014361

Kopan R, Ilagan MX. The canonical Notch signaling pathway: unfolding the activation
mechanism. Cell. 2009 Apr 17;137(2):216-33. doi: 10.1016/j.cell.2009.03.045.
Review. PMID: 19379690, PMCID: PMC2827930

Kovall RA. More complicated than it looks: assembly of Notch pathway transcription
189

complexes. Oncogene. 2008 Sep 1;27(38):5099-109. doi: 10.1038/onc.2008.223.
Review. PMID: 18758478

Kuratani S. Craniofacial development and the evolution of the vertebrates: the old
problems on a new background. Zoolog Sci. 2005 Jan;22(1):1-19. Review. PMID:
15684579

Li C, Scott DA, Hatch E, Tian X, Mansour SL. Dusp6 (Mkp3) is a negative feedback
regulator of FGF-stimulated ERK signaling during mouse development. Development.
2007 Jan;134(1):167-76. PMID: 17164422, PMCID: PMC2424197

Li J.JAK-STAT and bone metabolism. JAKSTAT. 2013 Jul 1;2(3):e23930. doi:
10.4161/jkst.23930. Epub 2013 Aug 15. Review. PMID: 24069548, PMCID:
PMC3772100

Liu YH, Kundu R, Wu L, Luo W, Ignelzi MA Jr, Snead ML, Maxson RE Jr. Premature
suture closure and ectopic cranial bone in mice expressing Msx2 transgenes in the
developing skull. Proc Natl Acad Sci U S A. 1995 Jun 20;92(13):6137-41. PMID:
7597092, PMCID: PMC41657

Ma L, Golden S, Wu L, Maxson R. The molecular basis of Boston-type craniosynostosis:
the Pro148-->His mutation in the N-terminal arm of the MSX2 homeodomain stabilizes
DNA binding without altering nucleotide sequence preferences. Hum Mol Genet. 1996
Dec;5(12):1915-20. PMID: 8968743
190

Maeno T, Moriishi T, Yoshida CA, Komori H, Kanatani N, Izumi S, Takaoka K, Komori
T.Early onset of Runx2 expression caused craniosynostosis, ectopic bone formation, and
limb defects. Bone. 2011 Oct;49(4):673-82. doi: 10.1016/j.bone.2011.07.023. Epub 2011
Jul 23. PMID: 21807129

Maxson R, Ishii M. The Bmp pathway in skull vault development. Front Oral Biol.
2008;12:197-208. doi: 10.1159/0000115042. Review. PMID: 18391502

Maynard TM, Jain MD, Balmer CW, LaMantia AS. High-resolution mapping of the Gli3
mutation extra-toes reveals a 51.5-kb deletion. Mamm Genome. 2002 Jan;13(1):58-61.
No abstract available. PMID: 11773971

McCright B, Lozier J, Gridley T. Generation of new Notch2 mutant alleles. Genesis.
2006 Jan;44(1):29-33. PMID: 16397869

McCright B, Lozier J, Gridley T. A mouse model of Alagille syndrome: Notch2 as a
genetic modifier of Jag1 haploinsufficiency. Development. 2002 Feb;129(4):1075-82.
PMID: 11861489

McDaniell R, Warthen DM, Sanchez-Lara PA, Pai A, Krantz ID, Piccoli DA, Spinner
NB. NOTCH2 Mutations Cause Alagille Syndrome, a Heterogeneous Disorder of the
Notch Signaling Pathway. American Journal of Human Genetics. 2006 Jul; 79(1): 169-
173 PMC [article] PMCID: PMC1474136, PMID: 16773578, DOI: 10.1086/505332
191

McDonald-McGinn DM, Feret H, Nah HD, Bartlett SP, Whitaker LA, Zackai EH.
Metopic craniosynostosis due to mutations in GLI3: A novel association. Am J Med
Genet A. 2010 Jul;152A(7):1654-60. doi: 10.1002/ajmg.a.33495. PMID: 20583172

Mead TJ, Yutzey KE. Notch pathway regulation of neural crest cell development in
vivo. Dev Dyn. 2012 Feb;241(2):376-89. doi: 10.1002/dvdy.23717. Epub 2012 Jan
3. PMID: 22275227, PMCID: PMC3266628

Merrill AE, Bochukova EG, Brugger SM, Ishii M, Pilz DT, Wall SA, Lyons KM, Wilkie
AO, Maxson RE Jr. Cell mixing at a neural crest-mesoderm boundary and deficient
ephrin-Eph signaling in the pathogenesis of craniosynostosis. Hum Mol Genet. 2006
Apr 15;15(8):1319-28. Epub 2006 Mar 15. PMID: 16540516

Minegishi Y, Saito M, Tsuchiya S, Tsuge I, Takada H, Hara T, Kawamura N, Ariga T,
Pasic S, Stojkovic O, Metin A, Karasuyama H. Dominant-negative mutations in the
DNA-binding domain of STAT3 cause hyper-IgE syndrome. Nature. 2007 Aug
30;448(7157):1058-62. Epub 2007 Aug 5. PMID: 17676033

Moenning A, Jäger R, Egert A, Kress W, Wardelmann E, Schorle H. Sustained
platelet-derived growth factor receptor alpha signaling in osteoblasts results in
craniosynostosis by overactivating the phospholipase C-gamma pathway. Mol Cell
Biol. 2009 Feb;29(3):881-91. doi: 10.1128/MCB.00885-08. Epub 2008 Dec 1.
PMID: 19047372, PMCID: PMC2630675
192

Morriss-Kay GM. Derivation of the mammalian skull vault.J Anat. 2001 Jul-Aug;199 (Pt
1-2):143-51. Review. PMID: 11523816, PMCID: PMC1594961

Morriss-Kay GM, Wilkie AO. Growth of the normal skull vault and its alteration in
craniosynostosis: insights from human genetics and experimental studies. J Anat.
2005 Nov;207(5):637-53. Review. PMID: 16313397, PMCID:
PMC1571561

Müller U, Warman ML, Mulliken JB, Weber JL. Assignment of a gene locus involved in
craniosynostosis to chromosome 5qter. Hum Mol Genet. 1993 Feb;2(2):119-22.
PMID: 8499900

Nichane M, Ren X, Bellefroid EJ. Self-regulation of Stat3 activity coordinates
cell-cycle progression and neural crest specification. EMBO J. 2010 Jan 6;29(1):55-67.
doi: 10.1038/emboj.2009.313. Epub 2009 Oct 22. PMID: 19851287, PMCID:
PMC2808363

Nieminen P, Morgan NV, Fenwick AL, Parmanen S, Veistinen L, Mikkola ML, van der
Spek PJ, Giraud A, Judd L, Arte S, Brueton LA, Wall SA, Mathijssen IM, Maher ER,
Wilkie AO, Kreiborg S, Thesleff I. Inactivation of IL11 signaling causes
craniosynostosis, delayed tooth eruption, and supernumerary teeth. Am J Hum Genet.
2011 Jul 15;89(1):67-81. doi: 10.1016/j.ajhg.2011.05.024. PMID: 21741611, PMCID:
PMC3135804
193

Noden DM. The control of avian cephalic neural crest cytodifferentiation. I.
Skeletal and connective tissues. Dev Biol. 1978 Dec;67(2):296-312. No abstract
available. PMID: 738529

Northcutt R. The new head hypothesis revisited. J Exp Zool B Mol Dev Evol.
2005 Jul 15;304(4):274-97. Review. PMID: 16003768

Opperman LA. Cranial sutures as intramembranous bone growth sites. Dev Dyn. 2000
Dec;219(4):472-85. Review. PMID: 11084647

Opperman LA, Nolen AA, Ogle RC.TGF-beta 1, TGF-beta 2, and TGF-beta 3 exhibit
distinct patterns of expression during cranial suture formation and obliteration
in vivo and in vitro. J Bone Miner Res. 1997 Mar;12(3):301-10. PMID: 9076572

Opperman LA, Sweeney TM, Redmon J, Persing JA, Ogle RC. Tissue interactions with
underlying dura mater inhibit osseous obliteration of developing cranial sutures. Dev
Dyn. 1993 Dec;198(4):312-22. PMID: 8130378

Opperman LA, Passarelli RW, Morgan EP, Reintjes M, Ogle RC. Cranial sutures
require tissue interactions with dura mater to resist osseous obliteration in
vitro. J Bone Miner Res. 1995 Dec;10(12):1978-87. PMID: 8619379

Oram KF, Gridley T. Mutations in snail family genes enhance craniosynostosis of
194

Twist1 haplo-insufficient mice: implications for Saethre-Chotzen Syndrome. Genetics.
2005 Jun;170(2):971-4. Epub 2005 Mar 31. PMID: 15802514, PMCID: PMC1450388

Paznekas WA, Cunningham ML, Howard TD, Korf BR, Lipson MH, Grix AW, Feingold
M,Goldberg R, Borochowitz Z, Aleck K, Mulliken J, Yin M, Jabs EW. Genetic
heterogeneity of Saethre-Chotzen syndrome, due to TWIST and FGFR mutations. Am J
Hum Genet. 1998 Jun;62(6):1370-80. PMID: 9585583, PMCID: PMC1377134

Ramirez F, Pereira L, Zhang H, Lee B. The fibrillin-Marfan syndrome
connection. Bioessays. 1993 Sep;15(9):589-94. Review. PMID: 8240311

Rawlins JT, Opperman LA. Tgf-beta regulation of suture morphogenesis and growth.
Front Oral Biol. 2008;12:178-96. doi: 10.1159/0000115038. Review. PMID: 18391501

Rébé C, Végran F, Berger H, Ghiringhelli F.STAT3 activation: A key factor in
tumor immunoescape. JAKSTAT. 2013 Jan 1;2(1):e23010. doi: 10.4161/jkst.23010.
Review. PMID: 24058791, PMCID: PMC3670267

Rice DP, Aberg T, Chan Y, Tang Z, Kettunen PJ, Pakarinen L, Maxson RE, Thesleff
I. Integration of FGF and TWIST in calvarial bone and suture development.
Development. 2000 May;127(9):1845-55. PMID: 10751173

Rice DP, Connor EC, Veltmaat JM, Lana-Elola E, Veistinen L, Tanimoto Y, Bellusci
S, Rice R. Gli3Xt-J/Xt-J mice exhibit lambdoid suture craniosynostosis which
195

results from altered osteoprogenitor proliferation and differentiation. Hum Mol
Genet. 2010 Sep 1;19(17):3457-67. doi: 10.1093/hmg/ddq258. Epub 2010 Jun
22. PMID: 20570969, PMCID: PMC2916710

Roth DA, Longaker MT, McCarthy JG, Rosen DM, McMullen HF, Levine JP, Sung J,
GoldLI. Studies in cranial suture biology: Part I. Increased immunoreactivity for
TGF-beta isoforms (beta 1, beta 2, and beta 3) during rat cranial suture fusion. J
Bone Miner Res. 1997 Mar;12(3):311-21. PMID: 9076573

Roybal PG, Wu NL, Sun J, Ting MC, Schafer CA, Maxson RE. Inactivation of Msx1 and
Msx2 in neural crest reveals an unexpected role in suppressing heterotopic bone
formation in the head. Dev Biol. 2010 Jul 1;343(1-2):28-39. doi:
10.1016/j.ydbio.2010.04.007. Epub 2010 Apr 14. PMID: 20398647, PMCID:
PMC3279331

Saga Y, Miyagawa-Tomita S, Takagi A, Kitajima S, Miyazaki Ji, Inoue T. MesP1 is
expressed in the heart precursor cells and required for the formation of a single
heart tube. Development. 1999 Aug;126(15):3437-47. PMID: 10393122

Senarath-Yapa K, Chung MT, McArdle A, Wong VW, Quarto N, Longaker MT, Wan
DC. Craniosynostosis: molecular pathways and future pharmacologic therapy.
Organogenesis. 2012 Oct-Dec;8(4):103-13. doi: 10.4161/org.23307. Epub
2012 Oct 1. Review. PMID: 23249483, PMCID: PMC3562251

196

Settle SH Jr, Rountree RB, Sinha A, Thacker A, Higgins K, Kingsley DM. Multiple
joint and skeletal patterning defects caused by single and double mutations in
the mouse Gdf6 and Gdf5 genes. Dev Biol. 2003 Feb 1;254(1):116-30.
PMID: 12606286

Sharma VP, Fenwick AL, Brockop MS, McGowan SJ, Goos JA, Hoogeboom AJ, Brady
AF, Jeelani NO, Lynch SA, Mulliken JB, Murray DJ, Phipps JM, Sweeney E, Tomkins
SE, Wilson LC, Bennett S, Cornall RJ, Broxholme J, Kanapin A; 500 Whole-Genome
Sequences (WGS500) Consortium, Johnson D, Wall SA, et al. Mutations in TCF12,
encoding a basic helix-loop-helix partner of TWIST1, are a frequent cause of
coronal craniosynostosis. Nat Genet. 2013 Mar;45(3):304-7. doi: 10.1038/ng.2531.
Epub 2013 Jan 27. Erratum in: Nat Genet. 2013 Oct;45(10):1261. PMID: 23354436,
PMCID: PMC3647333

Shishido E, Higashijima S, Emori Y, Saigo K.Two FGF-receptor homologues of
Drosophila: one is expressed in mesodermal primordium in early
embryos. Development. 1993 Feb;117(2):751-61. PMID: 8330538

Simpson MA, Irving MD, Asilmaz E, Gray MJ, Dafou D, Elmslie FV, Mansour S,
Holder SE, Brain CE, Burton BK, Kim KH, Pauli RM, Aftimos S, Stewart H, Kim CA,
Holder-Espinasse M, Robertson SP, Drake WM, Trembath RC. Mutations in NOTCH2
cause Hajdu-Cheney syndrome, a disorder of severe and progressive bone loss. Nat
Genet. 2011 Mar 6;43(4):303-5. doi: 10.1038/ng.779. PMID: 21378985

197

Singh KK, Rommel K, Mishra A, Karck M, Haverich A, Schmidtke J, Arslan-Kirchner
M. TGFBR1 and TGFBR2 mutations in patients with features of Marfan syndrome and
Loeys-Dietz syndrome. Hum Mutat. 2006 Aug;27(8):770-7. PMID:
16799921

Smithwick EM, Finelt M, Pahwa S, Good RA, Naspitz CK, Mendes NF, Kopersztyk S,
Spira TJ, Nahmias AJ. Cranial synostosis in Job's syndrome. Lancet. 1978 Apr
15;1(8068):826. No abstract available. PMID: 85845

Stamper BD, Mecham B, Park SS, Wilkerson H, Farin FM, Beyer RP, Bammler TK,
Mangravite LM, Cunningham ML. Transcriptome correlation analysis identifies two
unique craniosynostosis subtypes associated with IRS1 activation. Physiol
Genomics. 2012 Dec 1;44(23):1154-63. doi: 10.1152/physiolgenomics.00085.2012.
Epub 2012 Oct 16. PMID: 23073384, PMCID: PMC3544483

Takeda K, Noguchi K, Shi W, Tanaka T, Matsumoto M, Yoshida N, Kishimoto T, Akira
S. Targeted disruption of the mouse Stat3 gene leads to early embryonic lethality. Proc
Natl Acad Sci U S A. 1997 Apr 15;94(8):3801-4. PMID: 9108058, PMCID: PMC20521

Takeda K, Kaisho T, Yoshida N, Takeda J, Kishimoto T, Akira S. Stat3 activation is
responsible for IL-6-dependent T cell proliferation through preventing apoptosis:
generation and characterization of T cell-specific Stat3-deficient mice. J Immunol. 1998
Nov 1;161(9):4652-60. PMID: 9794394

198

Tanimoto Y, Yokozeki M, Hiura K, Matsumoto K, Nakanishi H, Matsumoto T, Marie PJ,
Moriyama K. A soluble form of fibroblast growth factor receptor 2 (FGFR2) with
S252W mutation acts as an efficient inhibitor for the enhanced osteoblastic
differentiation caused by FGFR2 activation in Apert syndrome. J Biol Chem. 2004
Oct 29;279(44):45926-34. Epub 2004 Aug 13.PubMed [citation] PMID: 15310757

Teng TS, Lin B, Manser E, Ng DC, Cao X. Stat3 promotes directional cell migration
by regulating Rac1 activity via its activator betaPIX. J Cell Sci. 2009 Nov
15;122(Pt 22):4150-9. doi: 10.1242/jcs.057109. Epub 2009 Oct 27.
PMID: 19861492

Thisse B, el Messal M, Perrin-Schmitt F. The twist gene: isolation of a Drosophila
zygotic gene necessary for the establishment of dorsoventral pattern. Nucleic
Acids Res. 1987 Apr 24;15(8):3439-53. PMID: 3106932, PMCID:
PMC340740

Ting MC, Wu NL, Roybal PG, Sun J, Liu L, Yen Y, Maxson RE Jr. EphA4 as an effector
of Twist1 in the guidance of osteogenic precursor cells during calvarial bone
growth and in craniosynostosis. Development. 2009 Mar;136(5):855-64. doi:
10.1242/dev.028605. PMID: 19201948, PMCID: PMC2685950

Veistinen L, Takatalo M, Tanimoto Y, Kesper DA, Vortkamp A, Rice
DP. Loss-of-Function of Gli3 in Mice Causes Abnormal Frontal Bone Morphology and
Premature Synostosis of the Interfrontal Suture. Front Physiol. 2012;3:121. doi:
199

10.3389/fphys.2012.00121. PMID: 22563320, PMCID: PMC3342524

Vissers LE, Cox TC, Maga AM, Short KM, Wiradjaja F, Janssen IM, Jehee F, Bertola
D, Liu J, Yagnik G, Sekiguchi K, Kiyozumi D, van Bokhoven H, Marcelis C,
Cunningham ML, Anderson PJ, Boyadjiev SA, Passos-Bueno MR, Veltman JA, Smyth I,
Buckley MF, Roscioli T. Heterozygous mutations of FREM1 are associated with an
increased risk of isolated metopic craniosynostosis in humans and mice. PLoS
Genet. 2011 Sep;7(9):e1002278. doi: 10.1371/journal.pgen.1002278. Epub 2011 Sep
8. PMID: 21931569, PMCID: PMC3169541

Vortkamp A, Gessler M, Grzeschik KH. GLI3 zinc-finger gene interrupted by
translocations in Greig syndrome families. Nature. 1991 Aug
8;352(6335):539-40. PMID: 1650914

Wan DC, Kwan MD, Lorenz HP, Longaker MT. Current treatment of craniosynostosis
and future therapeutic directions. Front Oral Biol. 2008;12:209-30. doi:
10.1159/0000115043. Review. PMID: 18391503

Wang B, Fallon JF, Beachy PA. Hedgehog-regulated processing of Gli3 produces an
anterior/posterior repressor gradient in the developing vertebrate limb. Cell. 2000 Feb
18;100(4):423-34. PMID: 10693759

Warman ML, Mulliken JB, Hayward PG, Müller U. Newly recognized autosomal
dominant disorder with craniosynostosis. Am J Med Genet. 1993 Jun 1;46(4):444-9.
200

PMID: 8357019

Warren SM, Brunet LJ, Harland RM, Economides AN, Longaker MT. The BMP
antagonist noggin regulates cranial suture fusion. Nature. 2003 Apr 10;422(6932):625-9.
PMID: 12687003

Wharton KA, Johansen KM, Xu T, Artavanis-Tsakonas S. Nucleotide sequence from the
neurogenic locus notch implies a gene product that shares homology with proteins
containing EGF-like repeats. Cell. 1985 Dec;43(3 Pt 2):567-81. PMID: 3935325

Wilkie AO, Oldridge M, Tang Z, Maxson RE Jr. Craniosynostosis and related limb
anomalies. Novartis Found Symp. 2001;232:122-33; discussion 133-43. Review.
PMID: 11277076

Wilkie AO, Morriss-Kay GM. Genetics of craniofacial development and malformation.
Nat Rev Genet. 2001 Jun;2(6):458-68. Review. PMID: 11389462

Wilkie AO. Craniosynostosis: genes and mechanisms. Hum Mol Genet.
1997;6(10):1647-56. Review. PMID: 9300656

Winograd J, Reilly MP, Roe R, Lutz J, Laughner E, Xu X, Hu L, Asakura T, vander
Kolk C, Strandberg JD, Semenza GL. Perinatal lethality and multiple craniofacial
malformations in MSX2 transgenic mice. Hum Mol Genet. 1997 Mar;6(3):369-79.
PMID: 9147639
201

Yagnik G, Ghuman A, Kim S, Stevens CG, Kimonis V, Stoler J, Sanchez-Lara PA,
Bernstein JA, Naydenov C, Drissi H, Cunningham ML, Kim J, Boyadjiev SA. ALX4
gain-of-function mutations in nonsyndromic craniosynostosis. Hum Mutat. 2012
Dec;33(12):1626-9. doi: 10.1002/humu.22166. Epub 2012 Aug 13. PMID: 22829454,
PMCID: PMC3495992

Yen HY, Ting MC, Maxson RE. Jagged1 functions downstream of Twist1 in the
specification of the coronal suture and the formation of a boundary between
osteogenic and non-osteogenic cells. Dev Biol. 2010 Nov 15;347(2):258-70. doi:
10.1016/j.ydbio.2010.08.010. Epub 2010 Aug 19. PMID: 20727876,
PMCID: PMC3210079

Yoshida T, Vivatbutsiri P, Morriss-Kay G, Saga Y, Iseki S. Cell lineage in
mammalian craniofacial mesenchyme. Mech Dev. 2008 Sep-Oct;125(9-10):797-808. doi:
10.1016/j.mod.2008.06.007. Epub 2008 Jun 20. PMID: 18617001

Yu HM, Jerchow B, Sheu TJ, Liu B, Costantini F, Puzas JE, Birchmeier W, Hsu W. The
role of Axin2 in calvarial morphogenesis and craniosynostosis. Development. 2005
Apr;132(8):1995-2005. PMID: 15790973, PMCID: PMC1828115

Zhang X, Kuroda S, Carpenter D, Nishimura I, Soo C, Moats R, Iida K, Wisner E, Hu
FY, Miao S, Beanes S, Dang C, Vastardis H, Longaker M, Tanizawa K, Kanayama N,
Saito N, Ting K. Craniosynostosis in transgenic mice overexpressing Nell-1.J Clin
202

Invest. 2002 Sep;110(6):861-70. Erratum in: J Clin Invest 2002 Nov;110(10):1573.
PMID: 12235118, PMCID: PMC151127

Abstract (if available)

Abstract Craniosynostosis refers to the genetic condition of premature fusion of calvarial sutures which occurs in roughly in 1 out of 2000‐2500 live births. It occurs as both syndromic and nonsyndromic disorder, where premature fusion of the skull bones hinders the proper expansion of the brain causing increase in intracranial pressure. This leads to improperly or distortedly formed brain and the skull accounting for mild to severe mental retardation and other life threatening craniofacial defects. Over the last two decades distinct cases of syndromic and non syndromic cases of Craniosynostoses have been shown to be caused by the inactivation or hyper-activation of individual genes e.g. Twist1, Jagged1, FGF, Msx2, Ephrin-Eph. Recently it was shown that Hyper Immunoglobulin syndrome (HIES) a multi-system immunological disorder characterized by increased serum IgE levels and displaying craniosynostosis as one of its clinical manifestations, is caused due to a loss of function mutation of the Stat3 gene. ❧ My work focuses on elucidating the role of Stat3 signaling in the neural crest cell derived frontal bone in maintenance of coronal suture patency. I show here that neural crest specific loss of Stat3 signaling leads to loss of Notch2 expression in the crest derived frontal bone and causes synostosis of the coronal suture. Subsequently it is shown that neural crest specific loss of Notch2 signaling also causes fusion of the coronal suture. These studies help in further understanding of etiology of the craniosynostosis disorders and the defects in molecular mechanisms that cause the phenotype.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

TWIST1 functions in both mesoderm and neural crest derived cranial tissues to establish and maintain coronal suture patency

PDF

Role of STAT3 phosphorylation in mouse embryonic stem cell self-renewal and differentiation

PDF

Utilizing zebrafish and mouse models to uncover the underlying genetics of human craniofacial anomalies

PDF

Pleotropic potential of Stat3 in determining self-renewal, apoptosis, and differentiation in mouse embryonic stem cells

PDF

Tissue interactions & molecular pathways in specification of the ectomesenchyme from cranial neural crest

PDF

Tissue-specific action of Msx genes in the regulation of skull vault development

PDF

Investigating the role of STAT3 in mouse and rat embryonic stem cell self-renewal and differentiation

PDF

Jagged1 functions downstream of Twist1 in the specification of the coronal suture and the formation of a boundary between osteogenic and non-osteogenic cells

PDF

Epigenetic checks and balances: PHF6 activity restricts neural crest migration

PDF

Jagged-notch signaling: patterning the vertebrate face

PDF

FGFR2 Regulates Connective Tissue Development in the Skull

PDF

Treatment of HPV driven tumors with SH5-07, a small molecule inhibitor targeting STAT3

PDF

The function of BS69 in mouse embryogenesis and embryonic stem cell differentiation

PDF

Novel roles for Maf1 in embryonic stem cell differentiation and adipogenesis

PDF

The role of HES/HEY transcriptional repressors in specification and maintenance of cell fate in the mouse organ of Corti

PDF

The role of Prkci in stem cell maintenance and cell polarity using a 3-D culture system

PDF

Elucidating the functional role of CHD7 associated nuclear PDH complex and other associated proteins on neural crest development

PDF

Fgfr2 regulates cell fate at the interface between tendon and bone

PDF

Protein arginine methyltransferases in murine skull development

PDF

Characterization of human embryonic stem cell derived retinal pigment epithelial cells for age-related macular degeneration

Asset Metadata

Creator Dasgupta, Krishnakali (author)

Core Title Elucidating the role of neural crest specific Stat3 signaling in maintaining coronal suture patency during embryonic development

School Keck School of Medicine

Degree Doctor of Philosophy

Degree Program Genetic, Molecular and Cellular Biology

Publication Date 09/22/2015

Defense Date 07/30/2014

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag coronal suture synostosis,Notch2,OAI-PMH Harvest,Stat3

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Maxson, Robert E., Jr. (committee chair), Frenkel, Baruch (committee member), Lu, Wange (committee member)

Creator Email aroekta@gmail.com,kdasgupt@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-482928

Unique identifier UC11286826

Identifier etd-DasguptaKr-2974.pdf (filename),usctheses-c3-482928 (legacy record id)

Legacy Identifier etd-DasguptaKr-2974.pdf

Dmrecord 482928

Document Type Dissertation

Format application/pdf (imt)

Rights Dasgupta, Krishnakali

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA