University of Southern California Dissertations and Theses

Lineage boundaries and cell migration in the patterning of the mammalian skull

(USC Thesis Other)

Lineage boundaries and cell migration in the patterning of the mammalian skull

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content LINEAGE BOUNDARIES AND CELL MIGRATION IN THE
PATTERNING OF THE MAMMALIAN SKULL

by

Man-Chun Ting

_____________________________________________

A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(BIOCHEMISTRY AND MOLECULAR BIOLOGY)

August 2008

Copyright 2008 Man-Chun Ting
ii

Dedication
I dedicate this dissertation to my parents, Ho-Ching Ting and Mon-Lin Hsieh. Your
unconditional love is the strength that keeps me walking towards my dream.

iii
Acknowledgments

Completing a PhD takes a lot of efforts and time, I thank these who made this
dissertation possible. First, my sincerest appreciation goes to my professor, Dr.
Robert Maxson. Thank you for your time and effort in teaching me to be independent
in my research.
I thank my committee members, Dr. Michael Stallcup and Dr. Henry Sucov for
being inspiring and for giving insightful advices.
I thank members of the Maxson lab: Nancy, Youzhen, Linda, Mamoru, Jingjing,
Helen, Paul, Chris and John, thank you all for making this a wonderful experience
for me.
Finally, I thank my soul mate, Chun-Peng. Thank you for holding onto me
through the hard times and giving me the strength and courage to keep moving.
Without your understanding and supports, it would have been impossible for me to
finish this work.

iv
Table of Contents
Dedication………………………………………………………………………... ii
Acknowledgements………………………………………………………………. iii
List of Figures………………………………………………………..................... v
Abstract…………………………………………………………………………... vii
Chapter 1: Eph-ephrin signaling functions downstream of Twist in neural
crest-mesoderm boundary formation and coronal craniosynostosis……………... 1
Abstract………………………………………………………………………. 1
Background…………………………………………………………………....2
Materials and Methods……………………………………………………….. 8
Results………………………………………………………………………... 11
Loss of EphA4 function causes coronal synostosis and defects in the
neural crest-mesoderm boundary at the coronal suture………………... 11
EphA4 is a downstream effector of Twist in the coronal suture……...... 14
Twist and EphA4 cooperatively control P-Erk and P-smad1/5/8
activity in the coronal suture…………………………………………… 24
Twist and EphA4 cooperatively control the neural crest-mesoderm
boundary……………………………………………………………….. 28
Discussion……………………………………………………………………. 34
Chapter 2: Identification of murine ephrin-A4 cis-regulatory elements that
control expression in the developing skull………………………………………. 42
Abstract………………………………………………………………………. 42
Background…………………………………………………………………... 43
Materials and Methods……………………………………………………….. 46
Results………………………………………………………………………... 50
Phylogenetic footprinting reveals conserved Twist and Msx2 binding
sequence in ephrinA4-flanking genomic region……………………...... 50
Twist and Msx2 are capable of interacting with conserved sequences
flanking ephrin-A4 gene……………………………………………... 55
1.6 kb segment contains elements sufficient for driving tissue-specific
expression in the developing skull…………………………………… 55
Expression level of 1.6 kb-hsp68-pksLacZ was reduced in the Twist
heterozygous mutant…………………………………………………. 59
Discussion…………………………………………………………………….. 63
Bibliography……………………………………………………………………... 68

v
List of Figures
Figure 1-1: Craniosynostosis in EphA4 homozygous and EphA4
-/-
; EphrinA2
-/-

double mutant mice………………………………………………………………. 13

Figure 1-2: Increased number of ALP-expressing cells in the EphA4
-/-
and
Twist
+/-
mutant…………………………………………………………….......... 15

Figure 1-3: Disrupted neural crest-paraxial mesoderm boundary in coronal
suture of Twist
+/-
and EphA4
-/-
mutants………………………………………... 16

Figure 1-4: EphA4 expression was dramatically altered in Twist mutants………. 17

Figure 1-5: Twist interacts directly with conserved regions of EphA4 upstream
segment…………………………………………………………………………... 19

Figure 1-6: Increased severity of craniosynostosis in Twist
+/–
; EphA4
+/–
double
heterozygous mice……………………………………………………………….. 22

Figure 1-7: Increased number of osteogenic cells within the coronal sutures of
compound heterozygotes compared with individual heterozygotes……………... 23

Figure 1-8: Twist and EphA4 cooperate in the regulation of P-Erk levels………. 26

Figure 1-9: Twist and EphA4 cooperate in the regulation of P-smad levels…….. 27

Figure 1-10: Noggin expression expanded to the sutural mesenchyme in the
Twsit mutant……………………………………………………………………… 29

Figure 1-11: Neural crest cells crossed the boundary into the undifferentiated
mesoderm in Twist
+/-
and Twist
+/–
EphA4
+/–
mutants…………………………….. 30

Figure 1-12: Mesoderm cells crossed the boundary into the neural crest derived
frontal bone region in Twist
+/-
and Twist
+/–
EphA4
+/–
mutants…………………… 31

Figure 1-13: Abnormal migratory behavior of osteogenic cells in Twist
+/-
and
Twist
+/-
; EphA4
+/-
mutant………………………………………………………… 33

Figure 2-1: Twist and Msx2 interact directly with conserved sequences of
ephrin-A4 flanking genomic region……………………………………………… 52

Figure 2-2: Generation of DNA constructs for microinjection…………………... 57

Figure 2-3: Section images through the coronal suture of E14.5 transient
transgenic embryos………………………………………………………………. 58
vi

Figure 2-4: Whole mount images demonstrate β-galactosidase activity of 1.6
kb-hsp68-pksLacZ transgenic embryos at serial developmental stages…………. 60

Figure 2-5: Histochemical analysis of β-galactosidase activity in the region of
coronal suture of 1.6 kb-hsp68-pksLacZ transgenic embryos…………………… 62

Figure 2-6: Expression level of 1.6 kb-hsp68-pksLacZ was reduced in the Twist
heterozygous mutant……………………………………………………………... 63
vii
Abstract
Eph-ephrin signaling has been implicated in craniosynostosis in humans:
Mutations in the ephrin ligands, EFNA4 and EFNB1, are known to cause
craniosynostosis. Our present work shows that Twist and Msx2, also craniosynostosis
genes, interact functionally with the Eph-ephrin signaling pathway. Reduced
expression of ephrinA2, ephrinA4 and EphA4 in a layer of cells located along the
neural-crest mesoderm boundary was previously described associating with the
upregulated expression of Msx2 in osteogenic mesenchyme and expanded into the
non-osteogenic mesoderm domain in Twist mutants. And the expression of
Eph-ephrin genes, as well as the boundary defect, can be rescued by genetic
inactivation of Msx2 in the context of the Twist
+/-
genotype. Here we demonstrate
that EphA4 mutant mice phenocopy Twist mutants: Wnt1-Cre and Mesp1-Cre
marking of neural crest and mesoderm show that EphA4 mutant embryos exhibit
defects in the neural crest-mesoderm boundary within the coronal suture. Osteogenic
neural crest-derived cells mix with non-osteogenic mesodermal cells fated to form
the suture. Associated with this mixing is the development of ectopic bone in the
suture. Further, genetic inactivation of a Twist allele in the context of EphA4
exacerbated both the boundary and synostosis phenotypes. By labeling migratory
osteogenic cells that contribute to the frontal and parietal bones, we showed that
Twist and EphA4 are required for the exclusion of osteogenic cells from the coronal
suture. Finally, to elucidate the mechanism of spatial and temporal control of
Eph-ephrin expressions, we characterized murine EphA4 and ephrin-A4
cis-regulatory elements by using phylogenetic footprinting approaches. Comparative
viii
genomic analysis revealed conserved blocks surrounding EphA4 and ephrinA4
genomic regions with putative Twist and Msx2 binding sequences. ChIP analysis
provided evidence Twist and/or Msx2 interact directly with conserved genomic
regions of EphA4 and ephein-A4. Together these data suggest that Twist and
Eph-ephrin signaling in a network controls the frontal-parietal boundary and coronal
suture development.

1
Chapter 1: Eph-ephrin signaling functions downstream of Twist in
neural crest-mesoderm boundary formation and coronal
craniosynostosis

Abstract
Heterozygous loss of Twist function causes coronal synostosis in both mice and
humans. We showed previously that in mice, this phenotype is associated with a
defect in the neural crest-mesoderm boundary within the coronal suture, as well as
the reduction expression of ephrinA2, A4 and EphA4 in the coronal suture. Further,
we demonstrated that mutations in human EPHRIN-A4 are a cause of non-syndromic
coronal synostosis. Here we investigate the cellular mechanisms by which Twist
acting through Eph-ephrin signaling, regulate coronal suture development. We show
that EphA4 mutant mice exhibit defects in the coronal suture and neural
crest-mesoderm boundary that phenocopy those of Twist
+/-
mice. Further, we
demonstrate that Twist and EphA4 interact genetically: compound heterozytoes have
suture defects of greater severity than those of individual heterozygotes.
Phylogenetic footprinting together with ChIP provided evidence that Twist interacts
directly with conserved genomic regions of EphA4. Finally, by labeling migratory
osteogenic cells that contribute to the frontal and parietal bones, we showed that
Twist and EphA4 are required for the exclusion of osteogenic cells from the coronal
suture. Together these data bring together Twist and Eph-ephrin signaling in a
network that controls coronal suture development, and they identify cellular
2
mechanisms through which perturbations in this network can lead to coronal
synostosis.

Background
The mammalian skull vault is a composite structure, consisting of membrane
bones with distinct lineage origins from cranial neural crest or paraxial mesoderm
(Jiang et al., 2002). Migratory neural crest cells give rise to the paired frontal bones,
whereas paraxial mesoderm cells contribute to parietal bones. The interparietal bone
is composed both neural crest-derived and mesoderm-derived cells. The bones of the
skull vault are separated by fibrous joints, termed sutures. Sutures form relatively
late in the craniofacial morphogenesis. They develop as a result of interactions
between two opposing osteogenic fronts. In the cases of sagittal, metopic and
lambdoid sutures, the bones are in direct opposition, formed by the narrowing of
membranous gaps between bones that are initially widely separate. In the case of
coronal suture, however, the bones overlap, with the parietal bone overlapping the
frontal bone.

Expansion of the skull by appositional growth at the cranial sutures
accommodates the growth of the brain during postnatal development. In humans, the
bony fusion of cranial sutures does not normally occur until early adulthood. The
only exception is the metopic suture, which begins to fuse at approximately 18
3
months of age. Equivalent to the posterior part of the human metopic suture, the
murine posterior frontal suture is the only one that undergoes fusion; other sutures
remain patent throughout life, allowing for normal expansion of the brain (Bradley et
al., 1996). The growth of sutures involves maintaining a balance of proliferating
osteogenic progenitor cells at the margins of the cranial bones and differentiating
osteoblasts that form new bones. Abnormal growth and morphogenesis of the cranial
vault result in cranial deformities. Signaling mechanisms involving maintenance of a
balance of this process include growth factors such as FGFs and members of the
TGF superfamily as well as transcription factors such as Twist and Msx2 (Jabs et al.,
1993; Karsenty, 1998; Kim et al., 1998; Twigg et al., 2004; Wilkie, 1997). Mutations
in these genes are known to cause several human syndromes with clinical features of
abnormalities in skull shape (Wilkie, 1997).

Craniosynostosis, the premature fusion of calvarial bones at the cranial sutures,
occurs as frequently

as 1 per 2500 live births. It results in an abnormal skull shape,
and may also cause impaired vision and hearing and mental retardation (Cohen and
MacLean, 1999; Wilkie and Morriss-Kay, 2001). Most of the syndromes for which
the molecular defect has been identified affect the coronal suture (Cunningham et al.,
2007). The coronal suture forms between bones of distinct embryonic tissues, the
neural crest-derived frontal and paraxial mesoderm-derived parietal bones. Based on
findings that boundaries often serve as centers of signaling and growth for
subsequent tissue patterning, it has been proposed that the coronal suture not only
4
serves as a tissue boundary, but also functions as the growth center (Blair, 1995;
Dahmann and Basler, 1999).

Coronal synostosis in patients with craniofrontonasal syndrome (CFNS) was
found to be associated with loss-of-function mutations in EFNB1, whose expression
domain coincided with the frontonasal neural crest–mesoderm boundary, suggesting
the cause of defect was through disturbing tissue boundary formation at the
developing coronal suture (Twigg et al., 2004). More recently, by using a genetic
lineage marker for neural crest, we have shown that the invasion of neural crest cells
into the mesoderm-derived parietal bone and mid-suture mesenchyme was associated
with craniosynostosis in Twist mutant mice. In addition, we demonstrated that
reduced expression of the boundary genes, ephrinA2, ephrinA4 and EphA4 in a layer
of cells located along the neural-crest mesoderm boundary was associated with the
boundary defect in Twist
+/-
mutants (Merrill et al., 2006).

Twist, a basic helix–loop–helix (bHLH) gene, was originally identified in
Drosophila as a gene required for mesoderm formation (Nusslein-V olhard et al.,
1984). Mutant embryos become abnormal during the early stages of development:
the torso becomes “twisted'. In contrast, homozygous knockout mouse embryos
survive through early development and die in mid-gestation of exsanguination due to
deficient vascular development. In addition, their neural tubes fail to close (Chen and
Behringer, 1995). Twist is expressed in the sutural mesenchyme between the
opposing proliferating osteogenic fronts in the developing skull (Johnson et al.,
5
2000). Increased activities of early osteoblast marker, alkaline phosphatase, in Twist
anitisense-treated osteoblastic cell lines and declined cell maturation when Twist is
overexpressed have suggested its role in regulating osteoblast differentiation
(Komaki et al., 2007; Lee et al., 1999; Murray et al., 1992; Rice et al., 2000). In
humans, heterozygous loss of TWIST function causes Saethre-Chotzen syndrome,
whose clinical features include synostosis of the coronal suture (el Ghouzzi et al.,
1997; Howard et al., 1997). Twist
+/-
mutant mice also exhibit coronal synostosis
consistent with the Saethre-Chotzen phenotype (el Ghouzzi et al., 1997).

Eph receptors comprise the largest family

of mammalian receptor tyrosine
kinases (RTKs), with at least 14 members. Ephrins are membrane-bound ligands for
Eph receptor tyrosine kinases. Ephrins are grouped into class A and class B based on
the mechanism of membrane attachment. Ephrin-As (A1~A6) are anchored in the
membrane through a glycosyl- phosphatidyl inositol (GPI) linkage, whereas
ephrin-Bs (B1~B3) have a transmembrane domain and a cytoplasmic region. Eph
receptors are divided into A- and B-subclasses as well. Each EphA (A1~A10)
receptor is able to bind several Ephrin-A ligands; and EphB (B1~B6) receptors
predominantly bind to Ephrin B ligands (Kullander and Klein, 2002; Wilkinson,
2001). There are several examples of interactions between members of different
subclasses (Gale et al., 1996; Himanen et al., 2004; Xu et al., 1999). Homologues of
Eph and ephrins have been identified in vertebrate and invertebrate species,
including mice, Xenopus, zebrafish and Caenorabhditis elegans and are widely
expressed in tissues of developing embryos (Drescher, 2002).
6

Interactions between Eph receptors and the ephrins usually occur at the
interface of complementary expression domains or within regions of overlapping
gradients. Eph-ephrin signaling is bidirectional, through both the receptor and the
ligand. Engagement of Eph receptors

by membrane-bound ephrin ligands induces
dimerization and subsequent trans-phosphorylation of the receptors and regulates a
variety of developmental processes

including embryonic vascular and neuronal
development and the establishment of developmental boundaries in tissue patterning
(Klein, 2004; Kullander and Klein, 2002; Martinez and Soriano, 2005; Palmer and
Klein, 2003; Pasquale, 2005; Poliakov et al., 2004; Surawska et al., 2004).
Compartment boundaries are lineage borders maintained by cell segregation. They
serve to maintain the position and shape of organizers during growth of a tissue. The
formation of boundaries between fields of cells is important in patterning
multi-cellular organisms. Eph and ephrins act at boundaries or in gradients to
regulate cell migration via repulsion and/or attraction responses. Genes controlled by
Eph-ephrin signalings include major regulators of cytoskeletal function and cell
adhesion as well as signaling molecules such as FGFR (fibroblast growth factor
receptor) (Arvanitis and Davy, 2008). Signaling effectors such as the
mitogen-activated protein kinases ERK, c-Jun N-terminal kinase, Src family kinases
and Ras/Rho family GTPases were shown to function downstream of the Eph-ephrin
signaling pathway. These effectors are important for the organization of the
cytoskeleton and cell adhesion (Elowe et al., 2001; Miao et al., 2001; Pasquale, 2008;
Poliakov et al., 2004; Pratt and Kinch, 2002; Schmucker and Zipursky, 2001; Stein et
7
al., 1998).

EphA4, which interacts with both ephrinAs and ephrinBs, is expressed in a
defined spatiotemporal pattern within the developing forebrain, hindbrain, and
mesoderm during early development (Mori et al., 1995; Nieto et al., 1992). EphA4
has been implicated in the formation of segment boundaries in the hindbrain and
somites via a repulsive

interaction with ephrin molecules (Barrios et al., 2003; Cooke
et al., 2005; Durbin et al., 1998). Zebrafish embryos treated with EphA4 antisense
morpholinos (EphA4MO) have defective rhombomere boundaries, consistent with a
requirement for Eph-ephrin signaling in boundary formation (Cooke et al., 2005).
EphA4 null mice are viable but have a gross motor dysfunction involving aberrant
spinal neuronal connections that result in the production of a characteristic hopping

gait (Dottori et al., 1998). EphA4 is a direct transcriptional target of several early
patterning genes including the zinc finger protein krox20, Pax3/FKHR and the
bHLH transcription factor Mesp2 (Begum et al., 2005; Nakajima et al., 2006; Theil
et al., 1998). In the developing skull, EphA4 is expressed on the ectocranial side of
the prospective frontal bone beginning at E14.5, when a discrete boundary of neural
crest- and mesoderm-derived cell populations is formed at the prospective coronal
suture (Merrill et al., 2006).

I have investigated the role of Eph-ephrin during coronal suture development. I
demonstrate that EphA4 mutant mice phenocopy Twist mutants: Wnt1-Cre and
Mesp1-Cre marking of neural crest and mesoderm show that EphA4 mutant embryos
8
exhibit defects in the neural crest-mesoderm boundary within the coronal suture.
Osteogenic neural crest-derived cells mix with non-osteogenic mesodermal cells
fated to form the suture. Associated with this mixing is the development of ectopic
bone in the suture. Genetic inactivation of a Twist allele in the context of EphA4
exacerbated both the boundary and synostosis phenotypes. In addition, Chromatin
immunoprecipitation (ChIP) analysis identified the Twist protein on the promoter
region of EphA4 gene suggesting that Twist regulated EphA4 directly. These results
suggest that Twist controls the frontal-parietal boundary through Eph-ephrin
signaling pathway.

Materials and Methods

Mouse mutants and genotyping
The EphA4 mutant was a kind gift of Dr. Elena Pasquale; the Ephrin-A2 mutant, of
Dr. David Feldheim. Both mutant lines were maintained in a C57Bl/6 background.
The Twist (Chen and Behringer, 1995), R26R (Soriano, 1999), Wnt1-cre (Danielian
et al., 1998) and Mesp1-cre (Saga et al., 1999) alleles have been described. I
genotyped EphA4, Ephrin-A2, Twist, R26R, Wnt1-cre and Mesp1-cre alleles by
PCR as described (Chen and Behringer, 1995; Dottori et al., 1998; Feldheim et al.,
2000; Jiang et al., 2002; Saga et al., 1999).

Histology, immunostaining, and in situ hybridization
9
Heads of embryos were embedded in OCT medium (Histoprep, Fisher Scientific)
before sectioning. Frozen sections were cut at 10 μm. Analysis of ß-galactosidase
activity of Wnt1-Cre/R26R and Mesp1-Cre/R26R reporter gene expression was
carried out as described (Ishii et al., 2003). Immunostaining of frozen sections was
largely carried out as previously reported (Ishii et al, 2003). Immunohistochemistry
was performed using rabbit anti-ephrinA2 (Zymed, 10ug/ml), rabbit anti-ephrinA4
(Zymed, 10ug/ml), goat anti-EphA4 (R and D, 5u), rabbit anti-Runx2 (Sigma, #9403,
1/600), rabbit phosphor-Erk1/2 (Cell Signaling, #4376), rabbit Erk1/2 (Cell
Signaling, #4695) and rabbit anti-phospho-smad1/5/8 (Cell Signaling, #9511L)
diluted in 1%BSA/PBS and incubated overnight at 4℃. Detection of goat primary
antibody was performed by incubating rabbit anti-goat-HRP (Zymed, 1/250) for 1
hour at room temperature and visualized with DAB substrate. Non-radioactive
section in situ hybridization using TSA (Tyramide Signal Amplification) method
was performed as described (Adams, 1992; Paratore et al., 1999; Yang et al., 1999).
Briefly, to analyze mRNA expression by TSA, DIG-labeled or FL-labeled
riboprobes was hybridized to the section and later detected with anti-DIG or anti-FL
antibodies conjugated to horseradish peroxidase (POD). Indirect Tyramide Signal
Amplification (TSA) fluorescence system (TSA-biotin/avidin-FITC) was used to
detect the POD-conjugated antibody (Perkin Elmer). RNA probes were generated as
reported: Epha4 (Nieto et al., 1992), Twist (Rice et al., 2000).

Whole mount skull alizarin red S staining
Skulls from 21 day post-natal mice were stained for bone with 2% alizarin red S in
10
1% KOH for 1 to 2 days. The specimens were then cleared and stored in 100%
glycerol.

Chromatin immunoprecipitation (ChIP) assays
MLB13MYC-clone-14 (C-14) cells (Rosen et al., 1994) were cultured

in Dulbecco’s
modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum.
ChIP assays were performed

30 minutes after 60 ng/ml BMP4 (R&D) treatment as
described (Ma et al., 2003) using an anti Twist (Santa Cruz; sc-15393) and an anti
Msx2 antibody

(Hybridoma Bank; 4G1). PCR amplification was performed using
primers that

flank Twist consensus biding sites of the EphA4 promoter (35

cycles).
Primer sequences were: EphA4-30 kb upstream (forward) 5'-ACA AAA CCC AAG
AGC CAA GTT G-3' and (reverse) 5'-TGT TCT GGT GAA TTC TGC CCT T-3';
EphA4-5 kb upstream (forward) 5'-TAT AAA TAG GCA GGT TGA AGC GGC-3'
and (reverse) 5'-GCC GGA TGG TGA GGA TGT TTA AAA-3'; EphA4-10 kb
upstream (forward) 5'-GTG AGC TGA CGT TGC CCT TGA CTA-3' and (reverse)
5'-CGC AGG CAT TTG TTT CTT CTT G-3'; 27 kb upstream control (forward)
5'-TTG CTT GTG ACA TCT TCT ATG CCC-3' and (reverse) 5'-GGG CTT AAG
CCT GTC TGC TAA ATT-3'.

Exo utero DiI (1,1-dioctadecyl-3,3,3,3–tetramethylindocarbocyanine perchlorate)
labeling
Details of the exo utero manipulation have been described (Muneoka et al., 1986;
Serbedzija et al., 1992). Briefly, E13.5 embryos with embryonic membranes were
11
carefully exposed out of the uterus by incising the uterine wall. Two embryos from
each side of

the uterine horns were designated as the experimental groups, and all
others were removed. DiI was injected into the area of the calvarial bone rudiments
under a microscope. After injection, the embryos were returned to the peritoneal
cavity of dams and allowed to continue development exo utero. After 3 days of
additional development, the embryos were removed for analysis. The survival rate of
the embryos after DiI injection was greater than 70%.

Results
Loss of EphA4 function causes coronal synostosis and defects in the neural
crest-mesoderm boundary at the coronal suture

We showed previously that fusion of the frontal and parietal bones (synostosis)
in Twist
+/-
mutants is associated with reduced expression of ephrinA4, ephrinA2 and
EphA4 in layers of mesoderm-derived cells distributed along the neural-crest
mesoderm boundary. This boundary is located at the coronal suture, at the posterior
edge of the frontal bone and the anterior edge of the parietal bone. To test whether
these changes in ephrinA ligand and receptor expression are functionally important in
the development of synostosis, we examined the morphology of coronal sutures in
individual EphA4
-/–
and EphrinA2
-/-
mutants (An ephrinA4 mutant is not available.).
In addition, because ephrins can function redundantly with their receptors (Wang et
12
al., 1999), we assessed compound EphA4
-/-
EphrinA2
+/–
mutant mice.

Analysis of Alizarin-red stained skulls of EphA4
-/-
mice at P21 revealed that the
coronal sutures were partially fused, a phenotype that closely resembles that of
Twist
+/-
mice (Figure 1-1). EphA4
+/-
and EphrinA2
-/-
mutants exhibited normal
coronal suture development. EphA4
-/-
; EphrinA2
+/–
double mutant skulls exhibited at
most a slight increase in the severity of the synostosis phenotype, indicating that
EphA4 plays a more prominent role than ephrinA2 in coronal synostosis, at least in
the mouse. We therefore concentrated on EphA4 mutants in further efforts to
understand the extent to which the Twist mutant phenotype is caused by deficient
ephrinA-EphA signaling.

To determine whether the developmental basis of the synostosis phenotype in
EphA4 mutants is similar to that of Twist mutants, we examined the activity of
alkaline phosphatase (ALP), an early osteoblast marker in EphA4
-/-
embryos at
E14.5. A neural crest-mesoderm boundary defect is detectable in Twist mutant mice
at this stage (Merrill et al., 2006). In wild type embryos, a layer of non-ALP
expressing cells was evident between the

ALP positive cells of the prospective
frontal and parietal bones. This layer is the prospective coronal suture. However, in
EphA4
-/-
mutants

this layer exhibited a disorganized appearance and was filled with
ALP-expressing

cells, as is the case in Twist
+/–
mutants (Merrill et al., 2006). In
addition, a layer of non-osteogenic mesenchyme located ectocranial to the
osteogenic layer expressed ALP ectopically, similar to a phenotype observed in the
13
Figure 1-1: Craniosynostosis in EphA4 homozygous and EphA4
-/-
;EphrinA2
double mutant mice.
Craniosynostosis in EphA4 homozygous and EphA4
-/-
;EphrinA2
-/-
double
mutant mice. (A-F) Alizarin-red stained p21 skulls show unilateral or bilateral
fusion of the coronal suture in the Twist
+/-
(D), EphA4
-/-
(E) and EphA4;
EphrinA2 (F) mutants. (G) The percentage of P21 animals with coronal suture
fusion was calculated for each genotype: wild-type (0%; 0/10), EphA4+/- (0%;
0/16), EphrinA2 -/- (0%; 0/9), Twist+/- (64%; 9/14), EphA4 -/- (40%; 8/20)
and EphA4 -/-; EphrinA2 +/- (47%; 8/17). fb, frontal bone; pb, parietal bone;
cs, coronal suture; ss, sagittal suture.
G G G G
14
Twist mutant (Figure 1-2).

We assessed the status of the neural crest-mesoderm boundary directly by
means of the Wnt1-Cre/ R26R neural crest lineage marking system (Jiang et al.,
2002). We carried out intercrosses between EphA4 mutants and embryos carrying
Wnt1-Cre and R26R and examined embryos at a series

of developmental stages
(Figure 1-3). In heterozygous mutants, no boundary defect was detected.
EphA4
-/-
mutants in contrast, lacZ positive cells could be seen outside the neural crest
domain, suggesting that the dosage of EphA4 controls the neural crest-mesoderm
boundary (Figure 1-3 C, F, I). The mixing phenotype was first found at E14.5 and
became more evident at E16.5. These results show that reduced EphA4 function
results in a set of phenotypes in the coronal suture that resemble those of Twist
mutants.

EphA4 is a downstream effector of Twist in the coronal suture

We showed previously that the expression of EphA4 protein is downregulated
in Twist mutants (Merrill et al, 2006). To determine whether EphA4 mRNA is
similarly downregulated, we performed in situ hybridization on control and mutant
samples. As is evident in Figure 1-4, EphA4 mRNA was localized in the periosteal
layers above and below the developing frontal and parietal bones. It was also
expressed in a layer of cells above (ectocranial to) the bone layer, and broadly within
the suture mesenchyme. EphA4 expression was dramatically altered in Twist
15
Figure 1-2: Increased number of ALP-expressing cells in the EphA4
-/-
and
Twist
+/-
mutant.
Examination the activity of ALP in EphA4
-/-
and Twist
+/-
embryos at E14.5 (C,
D) and E16.5 (E, F) revealed increased number of ALP-expressing cells in the
sutural mesenchyme and non-osteogenic mesenchyme surrounding the
osteogenic layer when compared with wildtype (A, B; curly bracket). (G) G is
a schematic showing difference of ALP activities between wildtype and
mutants. fb, frontal bone; pb, parietal bone; cs, coronal suture.
G G G G
16
Figure 1-3: Disrupted neural crest-paraxial mesoderm boundary in
coronal suture of Twist
+/-
and EphA4
-/-
mutants.
Wnt1-Cre; R26R analysis of neural crest distribution in wild type (A,D,G),
Twist
+/-
(B,E,H) and EphA4
-/-
(C,F,I) coronal suture at different stages. Invasion
of neural crest cells to the parietal bone and mid-suture mesenchyme were
found in Twist
+/-
heterozygous and EphA4
-/-
homozygous at E14.5 (B-D,
arrowheads), E16.5 (J-L, arrowheads) and P0 (R-T, arrows). fb: frontal bone;
pb: parietal bone; cs: coronal suture.

17
Figure 1-4: EphA4 expression was dramatically altered in Twist mutants.

EphA4 expression was dramatically altered in Twist mutants at E14.5, whereas
no detectable change of Twist expression was found in EphA4 mutants. (A, B)
EphA4 in situ hybridization on E14.5 head sections through the site of the
coronal suture of wildtype (A) and Twist
+/-
(B). EphA4 mRNA was localized in
the periosteal layers above and below the developing frontal and parietal
bones. It was also expressed in a layer of cells above (ectocranial to) the bone
layer (arrows), and broadly within the suture mesenchyme (A, ∗). In Twist
+/-

mutant, EphA4 mRNA was reduced substantially in the periosteal, ectocranial
layers and in sutural mesenchyme (B). (E, F) Twist in situ hybridization on
E14.5 head sections through the site of the coronal suture of wildtype (E) and
EphA4
-/-
(F). Twist mRNA was localized in the developing frontal and parietal
bones and sutural mesenchyme. It was also expressed in a layer of cells above
the bone layer (E, F, arrows). Adjacent sections (C, D, G, H) were stained for
Alkaline Phosphatase (ALP) to visualize the osteogenic mesenchyme. fb:
frontal bone; pb: parietal bone.

18
mutants. It was reduced substantially in the periosteal and ectocranial layers. It was
reduced to a single cell layer in the ALP territory, and was reduced in sutural
mesenchyme (Figure 1-4 B). These data corroborate our previous finding that Twist
controls the levels of EphA4 expression in calvarial tissues. Examination of
ephrinA4 mRNA in control and mutant embryos by in situ hybridization also
confirmed our previous results with ephrinA4 protein in wild type and Twist mutants
(Merrill et al., 2006).

Twist could control EphA4 directly or indirectly. If Twist is a direct regulator,
then Twist should interact with the EphA4 promoter, in vivo. We used a ChIP
approach to test this hypothesis. An in silico search for phylogenetically conserved
sequences in the region of the EphA4 gene identified several blocks. Two sites with
optimum CATATG sites were evident, one at ~5 kb and one at ~30 kb upstream of
EphA4. A third site had multiple E-box consensus sites, which although not optimal
for Twist, will bind E-boxproteins that can form heterodimers with Twist. We
prepared primer pairs to test for interactions between Twist and these three sites.
Control primers examined Twist occupancy at a site that lacked consensus Twist
sites. An antibody specific for Twist was used for immunoprecipitation. Extracts
were prepared from MLB13MYC-clone-14 (C14) cells, a mesenchymal cell line in
which Twist and EphA4 are expressed (data not shown). As can be seen in Figure
1-5, Twist protein was present on a PCR fragment that included the Twist site at ~5
kb region, as well as a fragment with the E box sites at ~10 kb upstream of EphA4.
No interaction was detectable at the control site. These data suggest that Twist is
19
Figure 1-5: Twist interacts directly with conserved regions of EphA4
upstream segment.

Twist binding site (CATATG) 77517264 Twist binding site (CATATG) 77517264 Twist binding site (CATATG) 77517264 Twist binding site (CATATG) 77517264 – – – – 77517335 77517335 77517335 77517335
Mouse tggaccagcatatgccctgccattttaagcta--ttctcaagaaccttcttacccatgga
Rat tggaccagcatatgccctaccattttaagctgctttctccacagctcctttacccataca
Human tagacaaagatatgtcatacaacttaacaatg--------------------taaagtga

E E E E- - - -boxes (CANNTG) 77521783 boxes (CANNTG) 77521783 boxes (CANNTG) 77521783 boxes (CANNTG) 77521783 - - - - 77521912 77521912 77521912 77521912
Mouse tttggacaaatgtgcccacctctgtaaacagactctctgatctcatttcccattgcctgg
Rat attggacaaatgtgcccacttctgtaaacaggatatctcattacatttcctattgcctgg
Human attggacaaatgtgcccacttctgtaaacaggatatctcattacatttcctattgtctct

Mouse actgcaaacttgtgatttcgggagcttgccaactgctgcctcacaccgcaggcatttgtt
Rat acagcaaacttgtgatttcaggagcttgccggttgctgcctcacaccgcagacatttgtt
Human gcagtcaactttgggtttcaggagcttgaggactgttttccaaagctgcagacatttgct

Control (no Twist binding sites) Control (no Twist binding sites) Control (no Twist binding sites) Control (no Twist binding sites)

Twist binding site (CATATG) 77541989 Twist binding site (CATATG) 77541989 Twist binding site (CATATG) 77541989 Twist binding site (CATATG) 77541989 - - - - 77542042 77542042 77542042 77542042
Mouse ttactttatccatatgacattgtcttttgaggcatatggataaatgagaccgtcttta
Rat ttactttatccatatgacattgtcttttgaggcatatggataaaggagaccgtcttta
Human ttagtttatccatatgacattgtctttttaggcatatggataaattagactgtcttta

A A A A
B B B B
20
C C C C

Twist interacts directly with conserved regions of EphA4 upstream segment.
(A) Schematic map of murine EphA4 locus on Chromosome 1. Arrows
indicate locations of primers used for ChIP assays. (B) Phylogenetic
footprinting revealed conserved E-box (; CANNTG) and Twist binding sites
(, ; CATATG) on non-coding region upstream of EphA4. (C) ChIP analysis
with Twist antibody revealed a direct interaction of Twist to regions and
of EphA4 promoter. No Twist binding was detected in region and the
control region , which lacks of Twist protein binding sequences.

21
capable of interacting with sequences upstream of the EphA4 gene.

We next asked if Twist interacts functionally with ephrinA-EphA signaling. If
EphA4 functions as an effector of Twist, then combination Twist
+/-
; EphA4
+/-

heterozygotes should exhibit phenotypes of greater severity than individual
heterogygous mutants. We crossed Twist
+/–
heterozygous mice with EphA4 mutant mice
and examined skulls at P21 by staining with Alizarin Red S. The penetrance of
craniosynostosis increased from 70% to 94% in Twist
+/-
; EphA4
+/-
compound mutants
(n=31) compared with Twist
+/-
mutants (n= 20); also, a larger portion of the suture was
fused (45% vs 21%) in the compound mutants (Figure 1-6).

Further evidence for an interaction between Twist and EphA4 came from
analysis of embryos at earlier stages of development. Whole mount ALP stains of
heads showed that while at E12.5 there was no difference in the ALP expression
domain between mutants and control embryos (data not shown), by E13.5, there was
a substantial change. In combination heterozygotes, the ALP domain expanded into
the coronal suture and also extended dorsally (Figure 1-7 A-C). Expression of the
osteoblast markers ALP and Runx2 in sections through the coronal suture revealed a
significant increase in the number of osteogenic cells within the sutures of
compound heterozygotes compared with individual heterozygotes. Mutants also had
ectopic Runx2-positive cells in the non-osteogenic layer ectocranial to the
osteogenic layer (Figure 1-7 E, F). These data suggest that Twist and EphA4
cooperatively control the distribution of osteogenic cells in coronal suture and
22

Figure 1-6: Increased severity of craniosynostosis in Twist
+/–
; EphA4
+/–

double heterozygous mice.
Increased severity of craniosynostosis in Twist
+/–
; EphA4
+/–
double
heterozygous mice. At P21, alizarin-red S stained skulls show unilateral or
bilateral fusion of the coronal suture in the Twist+/- (B) and Twist
+/-
; EphA4
+/-

(C) mutants. (D) The percentage of P21 animals with coronal suture fusion was
calculated for each genotype: wild-type (0%; 0/10), Twist
+/-
(70%; 14/20),
Twist
+/-
; EphA4
-/-
(94%; 29/31). fb, frontal bone; pb, parietal bone; cs, coronal
suture; ss, sagittal suture.

D D D D
23
Figure 1-7: Increased number of osteogenic cells within the coronal
sutures of compound heterozygotes compared with individual
heterozygotes.
Presence of alkaline phosphatase (ALP)- and Runx2- expressing cells in the
prospective coronal suture (ALP) in the Twist
+/-
;EphA4
+/-
mutant embryos.
(A-C) Whole mount ALP staining of wild-type Twist
+/-
and Twist
+/-
; EphA4
+/-

mutants at E13.5 revealed disorganized ALP-expressing cells filled the
prospective coronal suture in the mutants. (D-F) An antibody against Runx2
was used to stain frozen transverse sections of E14.5 wild type, Twist
+/-
, and
Twist
+/-
; EphA4
+/-
mutants. Runx2 (D, arrows) is expressed in the developing
prospective frontal and parietal bones. (E-H) Adjacent sections were stained
for ALP to visualize the osteogenic mesenchyme. Note that the expression of
ALP and Runx2 were increased in the number of osteogenic cells within the
sutures of mutants compared with wildtypes. Ectopic Runx2-positive cells in
the non-osteogenic layer ectocranial to the osteogenic layer were also found in
mutants. fb, frontal bone; pb, parietal bone; cs, coronal suture; ss, sagittal
suture.

24
ectocranial mesenchyme.

Twist and EphA4 cooperatively control P-Erk and P-smad1/5/8 activity in the
coronal suture.

We next asked whether Twist and EphA4 cooperate in the control of signaling
pathways known to have important roles in suture development. We examined both
the RTK and Bmp pathways. Twist and ephrin-Eph are known to function through
the RTK pathway (Guenou et al., 2005; Pratt and Kinch, 2002; Vindis et al., 2003).
FGF/FGFR signaling has a well documented role in craniosynostosis and normal
suture development (Johnson et al., 2000; Marie et al., 2005; Rice et al., 2000;
Yamaguchi and Rossant, 1995). Twist has been shown to control levels of FGFR
expression in sutures, and our previous and present results establish ephrin-Eph
signaling as a critical element in the maintenance of the neural crest mesoderm
boundary and the development of the coronal suture (Guenou et al., 2005; Merrill et
al., 2006; Twigg et al., 2004). Finally, the Bmp pathway is known to be involved in
the specification and differentiation of calvarial osteogenic cells (Kim et al., 1998;
Ryoo et al., 2006); forced expression of the Bmp antagonist noggin can prevent
fusion of the sagittal suture (Warren et al., 2003).

We examined the expression of the RTK effector, P-Erk, and the BMP effector,
P-smad1, 5, 8. Immunostaining of tissue sections derived from E14.5 embryos
showed that P-Erk was expressed in the ectocranial non-osteogenic cell layer as well
25
as in the underlying osteogenic layer. As the dosages of Twist and EphA4 were
reduced, the number of P-erk expressing cells decreased progressively in both layers
(Figure 1-8). These data show that Twist and EphA4 cooperate in the regulation of
P-Erk levels.

The distribution of p-smad 1/5/8 was also strongly influenced by Twist and
EphA4 (Figure 1-9). Control embryos expressed p-smad at a high level in the
leading edges of the growing frontal and parietal bones and in the outer
non-osteogenic cell layer. Lower levels were evident in more mature osteoblasts,
distal to the leading edges. Sutural cells, located between the osteogenic fronts, also
expressed low levels (Figure 1-9 A, B). There was a clear boundary between
domains of high and low psmad expression in the osteogenic fronts and suture. In
both Twist and EphA4 mutants, psmad activity was reduced significantly in cells of
the leading edges of the frontal and parietal bones, blurring or eliminating the
boundary between these cells and the prospective sutural cells (1-9 E, F, I, J).
Combination mutants exhibited an even more dramatic reduction in psmad staining
in the suture and outer layer of non-osteogenic mesenchyme (Figure 1-9 M, N).
These results, together with the finding of ectopic Runx2 expression in these same
cell types (figure 1-7), suggest that the cells in which phospho-smad activity was
downregulated took on characteristics of osteogenic cells. While this may seem at
first surprising, we note that in control embryos, cells located within the developing
bone, distal to the leading edges also exhibit reduced smad activity.
26

Figure 1-8: Twist and EphA4 cooperate in the regulation of P-Erk levels.

p-ERK1/2 expression, not ERK1/2, was reduced progressively as the dosage
of Twist and EphA4 was reduced. (A, C, E, G) An antibody against
phospho-ERK1/2 was used to stain frozen transverse sections of E14.5 wild
type (A), EphA4
-/-
(C)

, Twist
+/-
(E) and Twist
+/-
; EphA4
+/-
(G) mutant skulls.
The expression of p-ERK1/2 can be detected in the developing prospective
frontal and parietal bones, and also in layers above the prospective bones
(arrowhead). Reduced signals were found in individual and compound
EphA4 and Twist mutants (C, E, G) when compared with wild type (A). In
contrast to reduced expression of p-ERK in the mutant (K) when compared
with wildtype (I), no noticeable change of unphosphorylated ERK expression
was detected in the wildtype (J) and mutants (L). (B, D, F, H) Adjacent
sections were stained for Alkaline Phosphatase (ALP) to visualize the
osteogenic mesenchyme. fb, frontal bone; pb, parietal bone; cs, coronal
suture.

27
Figure 1-9: Twist and EphA4 cooperate in the regulation of P-smad levels.
15 15 15 150 0 0 0X X X X 60 60 60 600X 0X 0X 0X
An antibody against phospho-smad was used to stain frozen transverse
sections of E14.5 wild type (A, B), EphA4
-/-
(E, F), Twist
+/-
(I, J) and Twist
+/-
;
EphA4
+/-
(M, N) mutant skulls. Control embryos expressed p-smad at a high
level in the leading edges of the growing frontal and parietal bones (A, B).
Reduced signals were found in individual and compound EphA4 and Twist
mutants (E, F, I, J, M, N) when compared with wild type (A, B). (C, D, G, H,
K, L, O, P) Adjacent sections were stained for Alkaline Phosphatase (ALP)
to visualize the osteogenic mesenchyme. fb, frontal bone; pb, parietal bone.

28
We surveyed the expression of components of the Bmp pathway in order to identify
potential causes of the change in p-smad activity. Of the several molecules we
assessed, which included the ligands Bmp2, 4 and 7, and the Bmp inhibitor noggin,
noggin exhibited a dramatic change in expression. In wild type embryos at E14.5,
noggin was expressed in cells of the periosteal layer, the ectocranial non-osteogenic
layer, and the osteogenic front of the frontal bone. In mutants, noggin expression
expanded to the sutural mesenchyme (Figure 1-10).

Twist and EphA4 cooperatively control the neural crest-mesoderm boundary

Wnt1-Cre/R26R analysis of the neural crest-mesoderm boundary demonstrated
a larger number of neural crest cells that had crossed the boundary into the
undifferentiated mesoderm in a series of developmental stages from E14.5 to P0 in
Twist
+/–
EphA4
+/–
double heterozygous mice compared with individual heterozygotes
(Figure 1-11). Complementary results came from an assessment of the mesoderm
lineage by means of the pan mesoderm marker, Mesp1-Cre/R26R (Figure 1-12). We
produced mice with the genotype Twist; Mesp1-Cre and crossed the EphA4; R26R
mutant allele into these mice. In control embryos, the Mesp1-Cre-directed lacZ
territory was the inverse of the Wnt1-Cre lacZ domain: lacZ-positive cells were
evident in the parietal bone, and in the layer of cells ectocranial to the parietal and
frontal bones. There was strong labeling of the non-osteogenic cells interposed
between the frontal and parietal bones, i.e., the mesenchyme of the coronal suture. In
embryos with the Twist
+/-
; EphA4
+/-
genotype, a substantial number of lacZ positive
29
Figure 1-10: Noggin expression expanded to the sutural mesenchyme in
Twsit mutant.

The influence of Twist on noggin expression by using a Noggin
lacZ
knock-in
mouse. (A, E) In wild type embryos, noggin was expressed in cells of the
periosteal layer, the ectocranial non-osteogenic layer, and the osteogenic
front of the frontal bone. (B, F) In mutants, noggin expression expanded to
the sutural mesenchyme (arrows). fb: frontal bone; pb: parietal bone; cs:
coronal suture.

30
Figure 1-11: Neural crest cells crossed the boundary into the
undifferentiated mesoderm in Twist
+/-
and Twist
+/–
EphA4
+/–
mutants.
Wnt1-Cre R26R analysis of the neural crest-mesoderm boundary. (A-C, G-I
and M-O) Invasion of neural crest cells to the parietal bone and mid-suture
mesenchyme were found in both Twist
+/-
and Twist
+/-
; EphA4
+/-
mutants
at E14.5 (B and C, arrowheads), E16.5 (H and I, arrowheads) and P0 (N and
O, arrows). fb, frontal bone; pb, parietal bone; cs, coronal suture; ss, sagittal
suture; d, dura.

31
Figure 1-12: Mesoderm cells crossed the boundary into the neural crest
derived frontal bone region in Twist
+/-
and Twist
+/–
EphA4
+/–
mutants.

Mesp1-Cre; R26R analysis of mesodermal cells distribution in wild type (A,
B), Twist
+/-
(E, F) and Twist
+/-
; EphA4
+/-
(I, J) coronal suture at E16.5.
Invasion of mesodermal cells to the neural crest-derived frontal bone area
were found in Twist
+/-
(E, F; arrowhead) and Twist
+/-
; EphA4
+/-
mutants (I, J;
arrowhead). (C, D, G, H, K, L) Adjacent sections were stained for Alkaline
Phosphatase (ALP) to visualize the osteogenic mesenchyme. fb, frontal bone;
pb, parietal bone.

32
cells were ectopically located in the neural crest territory. Thus cells derived from
mesoderm, as well as neural crest, failed to respect the neural crest-mesoderm
boundary between the frontal and parietal bones.

These data raised the possibility that synostosis is caused at least in part by the
replacement or dilution of non-osteogenic sutural cells with osteogenic cells from
adjacent territories. To test this hypothesis further, we used diI labeling in
conjunction with exo utero culture of embryos. This allowed us to examine directly
the behavior of populations of migratory osteogenic cells that contribute to the
frontal and parietal bones. At E12.5, the frontal and parietal bone rudiments consist
of patches of osteogenic precursor cells located above the eye. The frontal bone
rudiment is anterior to the eye, the parietal bone rudiment posterior. Between the
rudiments is the prospective coronal suture, identifiable by the absence of osteogenic
precursors. The prospective frontal and parietal osteogenic cells can be labeled by
injecting diI into the area of the rudiments (supraorbital ridge) at E13.5 (Yoshida,
2005)(Iseki; Paul, unpublished). During subsequent development of the injected
embryos exo utero, labeled cells migrate dorsally, adding to the leading edge of the
frontal and parietal bones.

We asked whether such migratory osteogenic cells exhibit abnormal behavior in
Twist and EphA4 mutants. We injected diI into E13.5 embryos and allowed them to
develop exo utero until E16.5. We then examined the embryos by epifluourescence
microscopy. As is evident in Figure 1-13, labeled cells were excluded from the area
33
Figure 1-13: Abnormal migratory behavior of osteogenic cells in Twist
+/-

and Twist
+/-
; EphA4
+/-
mutant.
Abnormal migratory behavior of osteogenic cells in Twist
+/-
and Twist
+/-
;
EphA4
+/-
mutant. (A) DiI-labeled E13.5 embryos were later examined by
epifluourescence microscopy at E16.5. (B-E) Transverse frozen section of
E16.5 DiI-labeled heads. Note that DiI-labeled cells were excluded from the
area of the prospective suture in wild type mice (B; arrowhead), but not in
Twist
+/-
and Twist
+/-
; EphA4
+/-
mutant (D, F; arrow). (C, E, G) Adjacent
sections were stained for Alkaline Phosphatase (ALP) to visualize the
osteogenic mesenchyme. fb: frontal bone; pb: parietal bone; cs: coronal suture;
e, eye.

34
of the prospective suture in wild type mice. However in Twist-Eph mutants, labeled
cells were present in the sutural space. In addition, labeled cells were distributed
non-uniformly in the parietal bone territory of Twist-Eph mutants (data not shown).
These data suggest that Twist and EphA4 not only control the behavior of osteogenic
(migratory calvarial cells) cells at the crest mesoderm boundary, but also regulate the
large scale migratory behavior of such cells.

Together, results with neural crest and mesoderm lineage markers suggest that
Twist and EphA4 cooperate in the maintenance of the neural crest mesoderm
boundary within the coronal suture. DiI experiments suggest further that Twist and
Eph-ephrin signaling control the migration and sorting of osteogenic cells.

Discussion

Craniosynostosis represents a significant medical problem, occurring in ~1 in
2500 individuals. There is still very little known about the molecular and cellular
mechanisms underlying this defect. Prevalent views of the cause of the defect are
unbalanced proliferation, differentiation and death of cells in the cranial sutures
(Bialek et al., 2004; Lee et al., 1999; Yousfi et al., 2002; Yousfi et al., 2001). Our
work suggests an alternative possibility: that the premature suture fusion in Twist
mutant mice is a patterning defect involving change expression of boundary genes-
Eph and ephrins (Merrill et al., 2006). Here we show that Eph receptor mutations
35
cause a spectrum of phenotypes that overlap with those of Twist
+/-
mice. We show
further that EphA4 is an effector of Twist, and likely a direct target. Finally, we
demonstrate that Twist and EphA4 control the targeting of cells to the coronal suture,
and are required for the exclusion of osteogenic cells from the coronal suture.

EphA4 functions downstream of Twist in the coronal suture

Eph-ephrin signaling is bidirectional, through both the receptor and the ligand
(Pasquale, 2008). Reduced expression of the receptor EphA4, and the ligand
ephrin-A2 and ephrin-A4 was previously described in Twist
+/-
mutants (Merrill et al.,
2006). The similarity in the expression pattern of ephrin-A2 and ephrin-A4 in the
developing skull raises the possibility of functional redundancy. Our finding that
there was no increase in the severity of the suture defects in EphA4
-/-
; ephrin-A2
+/-

mutants may indicate that all signaling goes through the receptor (as opposed to
through the ligand)--i.e., if it is bidirectional signaling, then signaling going through
ephrinA2 is not important. Alternatively, there may be redundancy in ligand
function.

EphA4 is known to function downstream of the bHLH transcription factor
Mesp2, acting through an E-box containing region to control segment border
formation in the presomitic mesoderm (PSM)(Nakajima et al., 2006). Our finding of
a reduction in the mRNA level of EphA4 in Twist mutants together with ChIP results
placing the Twist protein on the EphA4 promoter in an area containing several
36
E-boxes, suggest, first, EphA4 is a direct target of Twist, and second that Twist may
regulate EphA4 expression through the conserved E-box region. However, our data
do not speak directly to the issue of whether the E-box containing region plays a
functional role in regulating EphA4 expression during coronal suture development.
Generating reporter transgenic lines and functional tests of the function of the
E-boxes in vivo need to be carried out to address the question.

Both the penetrance and severity of craniosynostosis were increased
significantly in Twist+/-; EphA4+/- mutants compared with individual mutants,
suggesting an interaction between Twist and EphA4. Further evidence for an
interaction came from analysis of the osteoblast markers ALP and Runx2 in embryos
at earlier stages of development. Runx2 was shown to be a key regulator of
osteoblast differentiation: analysis of Runx2 deficient mice revealed arrested
osteoblast differentiation (Komori et al., 1997; Otto et al., 1997). Moreover, genetic
inactivation of Runx2 in the Twist mutant rescued craniosynostosis (Bialek et al.,
2004). Inhibition of Twist synthesis using morpholino antisense oligonucleotides in
calvarial organ culture also showed upregulated expression of Runx2 and Fgfr2 in
the developing bone domain (Yoshida et al., 2005). Consistent with these results, our
data showed a dorsal expansion of ALP and Runx2 into ecto-cranial mesenchyme at
E14.5 and an increased number of ALP- and Runx2-expressing cells within the
coronal suture. These data suggest that Twist and EphA4 cooperatively control both
the differentiation of the osteogenic mesenchyme and the distribution of osteogenic
cells.
37

Twist and EphA4 cooperatively control RTK and BMP signaling in the coronal
suture

Receptor tyrosine kinases (RTKs) are a diverse group of transmembrane
proteins involved in signal transduction. Signaling through RTKs is a major
mechanism for intercellular communication during development. The RTK pathway
functions in many cell types to drive a wide variety of cellular responses by
transducing growth factor signals from the external milieu to intracellular processes.
The Ras-mitogen-activated protein kinase (MAPK) pathway is activated by RTKs,
of which Eph receptors are members. P-erk is a downstream marker of RTK
signaling. Twist is known to function through the RTK pathway by regulating FGFR
expression (Guenou et al., 2005). We therefore expected that compound Twist-Eph
mutants might exhibit cooperative downregulation of RTK signaling. To test this
hypothesis, we examined P-erk expression in individual and combination Twist and
Eph mutants. As expected, we observed progressive downregulation of p-erk
expression as the dosage of Twist and EphA4 was reduced, suggesting that Twist and
EphA4 cooperatively control RTK signaling. The result provides a potential
explanation of genetic interaction between Twist and EphA4 in the control of suture
development.

BMPs have been implicated in a variety of functions, including induction of
both cartilage and bone. The expression of BMPs in the developing skull was found
38
in presumptive osteogenic fronts and weakly in the sutural mesenchyme (Rice et al.,
1999). The BMP inhibitor, noggin, is also expressed in cranial sutures, and, from the
work of Longaker and colleagues, may influence suture closure (Warren et al., 2003).
Given the role of the Bmp pathway in osteogenesis, and the role of noggin in cranial
suture development, we examined the influence of Twist on noggin expression by
using a Noggin
lacZ
knock-in mouse. We crossed this noggin-LacZ allele into controls
and Twist mutant mice (Kim et al., 1998; McMahon et al., 1998; Rice et al., 1999;
Warren et al., 2003). Our finding of ectopic expression of Noggin
lacZ
in sutural
mesenchyme suggested that the BMP pathway is involved in sutural defect in the
Twist mutant. To test this idea definitively, we carried out staining with an antibody
against the key effector of the BMP pathway, p-smad 1/5/8. We found reduction of
p-smad1/5/8 activity in cells of the leading edges of the frontal and parietal bones
and the prospective suture in Twsit-EphA4 mutants. Together with a finding of
upregulated expression of Runx2, which is also a target of BMP signaling, these data
suggest that Twist controls the distribution of osteogenic cells within the suture
through the Eph-ephrin and Bmp pathways.

By using genetic lineage markers for neural crest- and mesoderm- derived cells,
we have shown that cells derived from different embryonic origins failed to respect
the neural crest-mesoderm boundary between the frontal and parietal bones. The
mixing phenotype seemed worse in embryos with the Twist
+/-
; EphA4
+/-
genotype
than in individual mutants. These results are consistent with the view that Twist
maintains the frontal-parietal boundary at least in part through Eph-ephrin signaling
39
pathway.

Twist and Eph-ephrin signaling cooperatively control targeting of osteogenic
cells

We showed previously that the increase in osteogenic cells within the coronal
sutures of Twist
+/-
mice is associated with a defect in the neural crest-mesoderm
boundary, which is located at the coronal suture (Merrill et al., 2006). By using
Wnt1-Cre/R26R as a genetic lineage marker for neural crest, we showed that in Twist
mutant mice, neural crest-derived osteogenic cells become distributed ectopically
within the undifferentiated mesoderm of the coronal suture, and we suggested that
this resulted in the ectopic differentiation of sutural cells (Merrill et al., 2006).
However, we did not address the mechanism by which osteogenic cells are
distributed ectopically. Our work here raises the possibility that is the result of
pathfinding defect of migratory osteogenic cells.

Lineage tracing experiments by diI of labeling head mesenchyme showed that
DiI labeled cells migrate apically from calvarial bone rudiments to the edge of the
developing bone. This suggests that the growth of skull occurs not by proliferation at
the osteogenic front, but by the addition of migrating cells at the edge of bone
(Yoshida, 2005)(Paul; unpublished data). Furthermore, Eph-ephrin signaling has
been shown to be involved in the distal pathfinding of motor axons and the
40
projection of axons from the retina to the optic tectum (Eberhart et al., 2000;
Huberman et al., 2005). Changes in the expression of EphAs were sufficient to cause
misprojection of axons within the lateral geniculate nucleus (LGN) (Huberman et al.,
2005). In the developing skull, the expression of EphA4 not only was found in
mid-suture, at the neural-crest boundary, but also in ectocranial mesenchyme,
suggesting a dual role in both boundary formation and the guidance of cell
migration. Theses data suggest that the effect of Twist-ephrin signaling on the neural
crest-mesoderm boundary may be at the level of the migration or pathfinding of
migratory ostoegenic cells.

To test this possibility we labeled cells of the frontal and parietal osteogenic
rudiments by diI injection and allowed embryos to develop ex utero. The exo utero
approach has been used effectively to explore molecular and cellular aspects of
mammalian development. It allows for direct experimental access to the embryo and
can be performed for extended periods to study developmental mechanisms in
postimplantation mammalian embryos (Hatta et al., 2004; Muneoka et al., 1986).
Our results show that diI labeled cells were excluded from the area of the
prospective suture in wild type mice. However in Twist-Eph mutants, labeled cells
were present in the sutural space. We suggest that this results in an altered
population of sutural cells with characteristics that allow these cells to mix (i.e., they
contain osteogenic cells). The primary defect, in this view, is in
migration/pathfinding of migratory osteogenic cells. Our results showing diI labeled
cells crossing into the coronal suture are consistent with the view that process that is
41
disrupted is pathfinding behavior.

In summary, our results suggest that Twist controls the neural crest-mesoderm
boundary and the distribution of osteogenic cells within the suture through the
Eph-ephrin and Bmp pathways. They also suggest that there is crosstalk between the
Eph-ephrin and Bmp pathways.
42
Chapter 2: Identification of murine ephrin-A4 cis-regulatory
elements that control expression in the developing skull

Abstract
Eph-ephrin signaling pathway acts at boundaries or in gradients to regulate a
variety of developmental processes

including embryonic vascular and neuronal
development and establishing developmental boundaries in tissue patterning.
Ephrin-A4 is a member of the ephrin ligand family, contains 4 exons and is located
in a cluster with ephrin-A3 and ephrin-A1 on human chromosome 1 and mouse
chromosome 3. Identification of loss-of-function mutations in the human EFNA4 in
patients with non-syndromic coronal synostosis has suggested its role in
craniosynostosis. To elucidate the mechanism of spatial and temporal control of
restricted pattern of expressions, given the role of Eph-ephrin as cell guidance and
boundary signals during development, we characterized murine ephrin-A4
cis-regulatory elements by using phylogenetic footprinting approaches in
combination with transient and stable transgenic analyses. Comparative genomic
analysis revealed four conserved blocks surrounding ephrinA4 genomic regions with
putative Twist and Msx2 binding sequences. ChIP analysis provided evidence of
direct interaction of Twist and Msx2 to these conserved fragments. Further,
transgenic analysis reveals that a 1.6 kb region, a non-coding genomic region
between ephrin-A4 and upstream Adam15 gene, contains elements that are sufficient
to direct tissue-specific expression of lacZ reporter in the developing skull and
43
genetically interact with Twist.

Background
The development of multi-cellular organisms is a coordinated process involving
many signaling pathways and billions of cells destined to differentiate in numerous
different ways. The inhibition of cell mixing between tissues or between distinct
tissue domains is essential in the patterning of multi-cellular organisms during
development. The mechanisms that accomplish cell segregation at boundaries are
proposed to be related to differential cell affinities (Dahmann and Basler, 2000;
Irvine and Rauskolb, 2001). Experimental support for this hypothesis came from cell
mixing assays, in vitro, which demonstrated that transfected cells expressing high
levels of a cell adhesion molecule sorted out from cells that expressed the same
molecule at lower levels (Steinberg and Takeichi, 1994). Eph-ephrin can regulate
cell adhesive interactions during embryonic development, particularly in the process
of boundary formation during vertebrate hindbrain segmentation (Dahmann and
Basler, 1999).

Ephs and ephrins are major downstream target genes of homeobox-containing
transcription factors (HOX) (Bruhl et al., 2004; Chen and Ruley, 1998; Cobb and
Duboule, 2005; Salsi and Zappavigna, 2006; Stadler et al., 2001). For example,
Eprhin A1 expression is differently regulated by Hox A5 and Hox B3 genes in
angiogenesis (Myers et al., 2000; Rhoads et al., 2005); and EphA7 expression is
44
downregulated in the Hoxd13 and Hoxa13 mutants in developing limbs (Salsi and
Zappavigna, 2006). Retroviral misexpression of homeobox-containing genes SOHo1
or GH6 completely and specifically repressed EphA3 expression in the neural retina
(Schulte and Cepko, 2000). Promoter analysis of Eph and ephrins in conjuction with
chromatin immunoprecipitation or electrophoretic mobility-shift (EMSA) assay
futher supports direct regulation of Eph/ephrins by Hox transcription factors (Bruhl
et al., 2004; Chen and Ruley, 1998). Our previous work also shows that reduced
expression of ephrinA4, ephrinA2 and EphA4 is associated with the expanded
expression of the homeobox gene, Msx2, in the non-osteogenic mesoderm domain in
Twist mutants, which suggests a genetic interaction between Twist, Msx2, and the
Eph-ephrin signaling pathway. The expression of Eph-ephrin genes can be rescued
by genetic inactivation of Msx2 in the context of the Twist
+/-
genotype (Merrill et al,
2006). In addition to homeobox-containing transcription factors, several early
patterning genes such as the zinc finger protein krox20 and the bHLH transcription
factor Mesp2 have been implicated in regulation of Eph-ephrin expression (Begum et
al., 2005; Nakajima et al., 2006; Theil et al., 1998).

Ephrin-A4 is a member of the ephrin family of signaling ligands. It is highly
conserved in the mouse and human, sharing 82% amino acid identity (Flenniken et
al., 1996). The ephrin-A4 gene contains 4 exons and is located in a cluster with
ephrin-A3 and ephrin-A1 on human chromosome 1 and mouse chromosome 3. The
expression of ephrin-A4 can be detected in the ectoderm of the branchial arches, the
ventral regions of the head, somites and the limb bud of E9.5 mouse embryos
45
(Flenniken et al., 1996). Later in the development, ephrin-A4 positive cells are found
in a single cell layer in the mesenchyme outside the osteogenic layer of the
developing skull (Merrill et al, 2006). The human orthologue of ephrin-A4, EFNA4,
was originally identified in a T-lymphoma cell line. It is expressed in a number of
human tissues, including pancreas, heart and fetal kidney (Kozlosky et al., 1995). A
splice variant of the EFNA4, EFNA4(s), was later found to encode a secreted form of
this ligand. Soluble EFNA4 is produced by mature B cells in the tonsil and the blood
(Aasheim et al., 2000).

Ephrin ligands are capable of interacting with multiple Eph receptors. The
binding of ephrin-A4 to EphA2, EphA5, and EphA6 results in autophosphorylation
of the receptor reflecting high-affinity, physiologically relevant ligand receptor
interactions (Gale et al., 1996). Ephrin-A4 can inhibit the outgrowth of sensory
neurites (Moss et al., 2005), and exert a positive influence on the formation of the
coronal suture (Merrill et al, 2006). Heterozygous mutations in the human EFNA4
gene were identified in patients with non-syndromic coronal synostosis. In addition,
reduced expression of ephrinA4 and its receptor EphA4 in E14.5 Twist mutant
mouse embryos was associated with the coronal suture defect in Twist
+/-
mutant mice,
suggesting that these molecules function in maintaining organized tissue patterns
during coronal suture development (Merrill et al, 2006).

Given the role of Eph-ephrin as cell guidance and boundary signals during
development, it is important to understand the mechanism of spatial and temporal
46
control of this restricted pattern of expression. However, limited information was
provided concerning the transcriptional regulation required for the expression of
ephrinA4. A 227 bp fragment upstream of the human EFNA4 gene (-366~ -140
relative to the translational start site) was previously described as the region required
for maximal promoter activity in human haemopoietic cell lines. Nuclear factor-Y
(NF-Y) can interact with this region (Munthe and Aasheim, 2002). Beyond this,
nothing is known about upstream factors that directly regulate the expression of
ephrin-A4.

To localize and identify ephrin-A4 cis-acting regulatory elements, we first make
use of phylogenetic footprinting approaches by performing multispecies sequence
comparisons. We identified several conserved blocks in genomic regions flanking
the ephrinA4 gene. Within these sequences were several conserved Twist and Msx2
consensus binding sequences. ChIP analysis provided evidence that Twist and Msx2
interact directly with conserved regions of ephrin-A4 promoter. Further, transgenic
analysis revealed that elements sufficient to direct tissue specific expression of a lacZ
reporter in the developing skull are located in a 1.6 kb region between ephrinA4 and
Adam15. Moreover, this region also contains sequences sufficient to respond to
Twist.

Materials and Methods

47
Phylogenetic Footprinting (Comparative Genomic Analysis of Conserved
Sequences)
Multi-species sequence alignments of genomic sequence surrounding mouse
ephrinA4 gene were generated using multiz and other tools in the UCSC/Penn State
Bioinformatics comparative genomics alignment pipeline at UCSC genome browser
(http://genome.ucsc.edu/cgi-bin/hgGateway). The conservation measurements were
created using the phastCons package from Adam Siepel at Cornell University.
Putative Twist and Msx2 -binding elements were identified by searching short match
of consensus biding sequences CATATG and TAATTG, respectively, using UCSC
genome browser (http://genome.ucsc.edu/cgi-bin/hgGateway) as well.

Chromatin immunoprecipitation (ChIP) assays
MLB13MYC-clone-14 (C-14) cells (Rosen et al., 1994) were cultured

in Dulbecco’s
modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum.
ChIP assays were performed

30 minutes after 60 ng/ml BMP4 (R&D) treatment as
described (Ma et al., 2003) using an anti Twist (Santa Cruz) and an anti Msx2
antibody

(Hybridoma Bank; 4G1). PCR amplification was performed using primers
that

flank Twist and Msx2 consensus biding sites of the ephrin-A4 promoter (35

cycles). Primer sequences were: Msx2 binding site 1 (forward) 5'-CCC ACC ACA
GCT AGA GAC AAA T-3' and (reverse) 5'-TCT AGA GCA TCG TGC TGG GT-3';
Msx2 binding site 2 (forward) 5'-TAA GGG AAG GGG TGT TGT CC-3' and
(reverse) 5'-CCT TCC CCC CAA TTA AAC ACT A-3'; Msx2 binding site 3 (forward)
5'-TGG CAT AGT GTG AGG TCC CTT-3' and (reverse) 5'-TC TCC CTT AGC TTG
48
GCC AGT-3'; Twist binding site 1 (forward) 5'-AAG ACA CTT GTG TGT GGA
AGC G-3' and (reverse) 5'-CAC ACG CAA ACT ACC AGC ACT-3'.

Generation of DNA constructs for microinjection
The 1.6 kb-ephrinA4-hsplacZ construct

contained a 1696 bp PCR fragment located
between –1696

and –1 upstream

of the ephrinA4 translation start site. The PCR
primers introduced a PstI site at

both the 5' and the 3' end. The 2.8
kb-ephrinA4-hsplacZ construct

contained a 2844 bp PCR fragment located
approximately 3 kb downstream

of the ephrinA4. The PCR primers introduced a
SmaI site at

both the 5' and the 3' end. The 2.9 kb-ephrinA4-hsplacZ construct

contained a 2919 bp PCR fragment located approximately 27 kb downstream

of the
ephrinA4. The PCR primers introduced a XhoI site and a SmaI site at the 5' and the 3'
end, respectively. The 3.3 kb-ephrinA4-hsplacZ construct

contained a 3287 bp PCR
fragment located approximately 52 kb downstream

of the ephrinA4. The PCR
primers introduced a KpnI site at

both the 5' and the 3' end. These fragments were
cloned

into designed restriction enzyme sites, those are PstI, SmaI, XhoI-SmaI, and
KpnI sites immediately upstream of the hsp68-lacZ-SV40

cassette (Kothary et al.,
1989). Primer sequences were: 1.6 kb-ephrinA4-hsplacZ (forward) 5'-CTG CAG
GAA TGT GTC TGC TCC TGG-3' and (reverse) 5'-CTG CAG CGC CCC TGG
GGT CCA GTC-3'; 2.8 kb-ephrinA4-hsplacZ (forward) 5'-CCC GGG AAA GCC
TTT TCT TAG AG-3' and (reverse) 5'-CCC GGG AAA CTT TGA TCT TTA GC-3';
2.9 kb-ephrinA4-hsplacZ (forward) 5'-CTC GAG TGG GGA TCT GAG CTA GTT-3'
and (reverse) 5'-CCC GGG TAG AAT TAA GCA GCA CAG-3'; 3.3
49
kb-ephrinA4-hsplacZ (forward) 5'-GGT ACC CTT GAT GGC TAC GTT GGA-3'
and (reverse) 5'-GGT ACC GCC AGG GCT ACA TAT TA-3'.

Production and genotyping of transgenic mice
Transgenic mouse embryos and lines were generated by pronuclear injections as
described by Liu et al. (Liu et al., 1994). DNA prepared from mouse tails or embryo
yolk sacs was used for genotyping of LacZ allele by PCR as described (Kwang et al.,
2002).

ß-galactosidase staining, histology, and in situ hybridization
Embryos were fixed in 4% paraformaldehyde for 30-45 minutes at 4°C prior to
staining (Liu et al., 1994). Staining for ß-galactosidase activity in whole embryos and
on frozen sections was carried out as previously reported. Heads of embryos were
embedded in OCT medium (Histoprep, Fisher Scientific). Frozen sections were cut
at 10 μm. Non-radioactive section in situ hybridization using TSA (Tyramide Signal
Amplification) method was performed as previously described (Adams, 1992;
Paratore et al., 1999; Yang et al., 1999). Briefly, to analyze mRNA expression by
TSA, DIG-labeled or FL-labeled riboprobe was hybridized to the section and later
detected with anti-DIG or anti-FL antibodies conjugated to horseradish peroxidase
(POD). Indirect Tyramide Signal Amplification (TSA) fluorescence system
(TSA-biotin/avidin-FITC) was used to detect the POD-conjugated antibody (Perkin
Elmer).

50
Results

Phylogenetic footprinting reveals conserved Twist and Msx2 binding sequence
in ephrinA4-flanking genomic region

To identify ephrinA4 candidate cis-acting regulatory elements, we used
phylogenetic footprinting by performing multispecies sequence comparisons
(Lenhard et al., 2003; Thomas and Green, 2003; Wasserman and Sandelin, 2004;
Zhang and Gerstein, 2003). We focused on genomic regions flanking the mouse
ephrinA4 gene. The mouse ephrin-A4 gene is located on mouse chromosome 3,
consisting of 4 exons spanning approximately 4.6 kb. Downstream of the ephrin-A4
gene are ephrin-A3 and ephrin-A1, located ~10.5 kb and ~53.7 kb away, respectively.
Approximately 1.6 kb upstream of ephrin-A4 is Adam15, a member of the
membrane-anchored glycoproteins family.

A previous study of the human EFNA4 promoter identified a proximal 227 bp
fragment required for maximal activity in human haemopoietic cell lines. This
fragment contained a nuclear factor-Y (NF-Y) binding site (Munthe and Aasheim,
2002). The possibility that the promoter activity may extend to more distal parts of
the locus cannot be excluded. We therefore performed cross-species sequence
comparisons by using alignment program Multiz on the UCSC genome browser. We
focused on 45.5 kb downstream and 1.6 kb upstream of ephrin-A4. Sequence
analysis of these non-coding genomic regions revealed several blocks of
51
ephrin-A4-flanking sequences that are highly conserved among vertebrates
(figure2-1A).

The presence of conserved transcription factor binding sites within the blocks
suggests that these evolutionarily conserved motifs likely encode cis-regulatory
regions relevant for the regulation of the gene. We searched for potential
transcription factor binding sequences within these conserved blocks to confine the
number of blocks for following studies. Twist and Msx2 were chosen based on their
expression patterns and our previous findings to the effect that they both interact
functionally with the Eph-ephrin signaling pathway to control the frontal-parietal
boundary (Merrill et al., 2006).

Of more than ten Msx2- (TAATTG) and Twist- (CATATG) consensus binding
sites identified, only two Msx2 sites are highly conserved in human and mouse, and
are located within conserved blocks that span at least 100 bp in length (figure 2-1B
and 2-1C). These two putative Msx2 binding sites are located approximately 3 kb
and 27 kb downstream of ephrin-A4. None of the Twist consensus binding sites is
highly conserved among species; however, three highly conserved bHLH protein
binding E-boxes (CANNTG) were found clustered approximately 52.8 kb
downstream of ephrin-A4. In contrast to the downstream regions, the 1.6 kb region
upstream of ephrin-A4 contained neither Msx2- (TAATTG) nor Twist- (CATATG)
consensus binding sequences. Instead, a TAAT homeodomain binding core sequence,
which could also serve as a low affinity binding site for Msx2, was present at
52
Figure 2-1: Twist and Msx2 interact directly with conserved sequences
of ephrin-A4 flanking genomic region.
B
A
53

C
Msx2 binding site 1 Msx2 binding site 1 Msx2 binding site 1 Msx2 binding site 1

Mouse gattacagccacggggttcatgcagcaccacctaagagggccggttcggtttatgtttag
Human ==========================================================ag
Rat aattacacccacgcgctt-atgcaccaccagctaagaaggctggtttggttt-tgtttag

Msx2 binding site 2 Msx2 binding site 2 Msx2 binding site 2 Msx2 binding site 2

Mouse cacctcattagtgtttaattggggggaaggaagctgc-cctggcaccagacaggtg
Rat ccctccattagtgt-taattgggggaggga-agctgc-cctggcaccagacaggtg
Human cacctcattagtgtttaattggaggaagga-agctgc-cttggcaccagacaggtg
TreeShrew cacctcattagtgtttaattggaggaaggattgctgcgcctggcaccagacaggtg
Shrew cacctcattagtgtttaattggaggaagga-agctgc-cctggcaccagacaggtg
Dog cacctcattagtatttaattggaggaagga-agctgc-cctggcaccagacaggtg
Horse cacctcattagtgtttaattggaggaagga-agctgc-cctggcaccagacaggtg
Cow cacatcattagtgcttaattggagaaagga-agctgc-cctggcaccagacaggtg

Msx2 binding site 3 Msx2 binding site 3 Msx2 binding site 3 Msx2 binding site 3

Mouse ta---caattaccaggggccagatgctgct----aagaaggcctagagattggtat
Rat ta---caattactgggggccagatgctggctgtgatgaaggcctagagattggtat
Human ccctgtaattacctggggccagatgctgtg----gcgaaggccaagagtttggtgc
Dog tcctataattgccccaagccggacactctg----gtgaaagccaggagcttggcac
Horse tcctattgttaccccgggccagacggtgtg----gtggatgctaagaatttgctgc
Cow ccccctaattacccgtggccagacgctgtg----gtgaaagccaagagtgttgtac

1 Twist Twist Twist Twist binding site 1 (E binding site 1 (E binding site 1 (E binding site 1 (E- - - -Box: CANNTG) Box: CANNTG) Box: CANNTG) Box: CANNTG)

Mouse catatggcgtgtgtaaatgcgtgtgatggcgggtaggtgtaaagacacttgtgtgt
Rat catatggcgtgtgtaaatgcgtgtgatggcgggtaggtgtaaagacacttgtgtgt
Human --------------aaatacgtgtgatggcgcgctagcatgaaagcatttgtgtgt
TreeShrew catctggcgtgtgtaaatacgtgtgatggcgcgcaagtgtgaaagcatttgtg--
Dog catctggcatgtgtaaatacgtgtgctggcacgcgagcgtgaaggcatgtgtgtgg
Cow catctggcatgtgtaaatacgtgtgatggcgcgcaagcgtgaaagcgtttgtgtgg

Mouse ggaagcgtgtgaaac-gcctgacatacatatgctaatgtttgtgaacagtgtgtga
Rat ggaagcgtgtgaaac-gcctgacatacatatgctaatgtttgtgaacagtgtgtga
Human ggaagcgtgtgaaactctttgacagacgtatgctaatgtttgtgaacagcgtgtga
TreeShrew --------------------------------------------------------
Dog ggaagcgtgtgcagc--tctggcgtacgcgtgctaatgttagtgaacggcgtgtaa
Cow ggaagcgtgtgaaac--tttgacgaacgtatgctaatgtctgtgaacagtgtgtga

54
D D D D
Figure 2-1: (A) Phylogenetically conserved sequences of human and mouse
surrounding the ephrin-A4 gene. Pairwise alignment of human to the mouse
genome was generated using BLASTZ and displayed the conservation histogram
as a grayscale density plot that indicates alignment quality. Relative positions of
genes surrounding ephrin-A4 are as shown. Primers for ChIP analysis are shown
as black arrows flanking conserved Twist (red bar) or Msx2 (green bar) binding
sequences. (B) Schematic map of ephrin-A4 and flanking genes on mouse
chromosome 3. Relative positions of conserved Msx2 and Twist binding sites to
ephrin-A4 gene are as shown. (C) Phylogenetic footprinting revealed conserved
Msx2 (TAATTG; site and ) and Twist (CATATG; site 1) binding sites
downstream of ephrin-A4. Upstream of ephrin-A4, one TAAT homeodomain
binding core sequence (site ) found at position -728 related to translational start
site, could also be potential Msx2 binding sequence. (D) ChIP analysis with
anti-Msx2 and anti-Twist antibody revealed a direct interaction of Msx2 and Twist
with conserved region of ephrin-A4 promoter.

55
position -728 relative to the translational start site; however, this sequence is only
conserved in rodents (mouse and rat) (figure2-1C).

Twist and Msx2 are capable of interacting with conserved sequences flanking
ephrin-A4 gene

We showed previously that ephrinA4 interacts genetically with Twistand Msx2
to maintain the frontal-parietal boundary (Merrill et al, 2006). Whether this
interaction is direct or indirect is still not known. We used chromatin
immunoprecipitation (ChIP) to test the four genomic regions mentioned above for a
direct interaction with Twist or Msx2. A skeletal progenitor cell line
MLB13MYC-clone-14 (C14) was used as the source of chromatin. We tested this
cell line with and without added BMP. Association of Twist and Msx2 with
endogenous genomic segments was detected by PCR assays using primers flanking
the conserved Twist and Msx2 biding sites. Strong binding of Twist and Msx2 was
found in the E-box-containing region and the region 3 kb downstream of ephrin-A4.
Much less association of Msx2 was detected in the 27 kb downstream of ephrin-A4
(Figure 2-1 D). Interestingly, although the TAAT sequence found within the 1.6 kb
region upstream of ephrin-A4 was not conserved, recruitment of Msx2 to this region
was also detected, suggesting this fragment could also serve as a potential
cis-regulator of ephrin-A4.

1.6 kb segment contains elements sufficient for driving tissue-specific expression
56
in the developing skull

To further test the regulatory activity of these segments, we amplified
approximately 1.6 ~3.3 kb long DNA fragments containing conserved Msx2 or Twist
consensus binding sites by PCR on genomic DNA. We fused these fragments with
the hsp68-LacZ reporter and used them to generate transient transgenic mouse
embryos. These constructs are shown in figure (2-2 A, B). At least 3 independent
transient transgenic embryos were generated for each construct. The regulatory
activity of each fragment was monitored by staining for beta-galactosidase activity at
embryonic day E14.5.

Of four constructs injected, two showed regulatory activity in E14.5 embryos.
One is 1.6 kb-hsp68-pksLacZ, a non-coding genomic region located between
ephrin-A4 and Adam15, and the other is 2.9 kb-hsp68-pksLacZ, which is located
approximately 27 kb downstream of ephrin-A4 and 8.5 downstream of ephrin-A3.
Two transient transgenic embryos carrying 1.6 kb-hsp68-pksLacZ showed LacZ
activity with similar pattern of expression at E14.5 (Figure 2-3 C and E). The
expression pattern of a highly localized line on the ecto-cranial mesenchyme along
the prospective frontal bone recapitulated the expression of the endogenous
ephrin-A4 gene, suggesting 1.6 kb region contains elements sufficient for driving
tissue-specific expression in the developing skull (figure 2-3 A and Merrill el al,
2006). Two transient transgenic embryos carrying 2.9 kb-hsp68-pksLacZ showed
slightly different patterns of expression at E14.5 (Figure 2-3 G and I). The LacZ
57
Figure 2-2: Generation of DNA constructs for microinjection.
A
B
Generation of DNA constructs for microinjection (A)Schematic map of
murine ephrin-A4 genomic locus showing four regions containing conserved
Msx2 or Twist binding sequence. (B) Schematic diagram of ephrin-A4-lacZ
transgene constructs. Specified ephrin-A4 promoter fragments (, , and 1)
are fused to an hsp68 core minimal promoter, which is fused to a lacZ reporter
gene, followed by a SV40 poly adenylation signal.

58
Figure 2-3: Section images through the coronal suture of E14.5 transient
transgenic embryos.

At E14.5, transverse sections through the head region reveals a similar LacZ
expression pattern driven by 1.6kb and 2.9 kb fragments to that observed in the
endogenous ephrin-A4. (A) Ephrin-A4 in situ hybridization on E14.5 section of
wildtype embryo. Ephrin-A4 transcripts are localized the ecto-cranial mesenchyme
along the prospective frontal bone and periosteal layer. (C, E) The expression of
1.6 kb-hsp68-pksLacZ transgene was found on the ecto-cranial mesenchyme
(arrowhead). (G, I) The LacZ activities of 2.9 kb-hsp68-pksLacZ transient lines
were present on the periosteal layer along the frontal bone (arrow); the expression
on the ecto-cranial mesenchyml layer was either absent or with low activity
(arrowhead). (B, D, F, H, J) Adjacent sections were stained for ALP to visualize
the osteogenic mesenchyme. fb, frontal bone; pb, parietal bone.

59
expression of both 2.9 kb-hsp68-pksLacZ transient lines was in the periosteal layer
along the frontal bone; however, the expression in the ectocranial mesenchyml layer
was either absent or low.

Expression level of 1.6 kb-hsp68-pksLacZ was reduced in the Twist heterozygous
mutant

Previously we found that ephrinA4 expression was reduced in Twist mutants.
To further test the possibility that Twist elicits its effect through the 1.6 kb fragment,
we made stable transgenic lines carrying 1.6 kb-hsp68-pksLacZ and later crossed the
line with Twist heterozygous mutant mice to generate F1 embryos. We first checked
the expression of 1.6 kb-hsp68-pksLacZ stable transgenic line at different
developmental stages.

At E9.5, the expression of 1.6 kb-hsp68-pksLacZ replicated a subset of the
endogenous ephrin-A4 expression pattern (figure 2-4 A-C)(Flenniken et al., 1996).
LacZ staining was detected predominantly in the ventral regions of the head. Weak
staining was found in the hindbrain and somites. At E13.5, transgene expression
became evident in somites and limbs (figure2-4 E). At E14.5, transverse sections
through the head region revealed an expression pattern similar to the endogenous
pattern as well the pattern observed in the transient transgenic embryos. However,
60
Figure 2-4: Whole mount images demonstrate β β β β-galactosidase activity of 1.6
kb-hsp68-pksLacZ transgenic embryos at serial developmental stages.

Whole mount images demonstrate β-galactosidase activity of 1.6
kb-hsp68-pksLacZ transgenic embryos at serial developmental stages. (A-C) At
E9.5, the expression was detected predominantly in the ventral regions of the head.
Weak staining was found in hindbrain and somites (C, arrowhead). (D, E) At
E11.5 and E13.5, the transgene expression in somites and limbs became evident.
(F) At P0, strong expression of 1.6 kb-hsp68-pksLacZ was evident in the sutures.

61
we noted expanded expression of beta-galactosidase activity in the suture
mesenchymal layer and periosteum, unlike the endogenous pattern (figure 2-5). At
P0, strong expression was evident in the suture mesenchymal layer (figure 2-4F).

We next crossed this transgenic line with Twist heterozygous mutant mice,
generating F1 embryos. These were analyzed for lacZ activity at E14.5 to test the
possibility of genetic interaction between 1.6 kb fragment of ephrin-A4 promoter and
Twist. We found that the highly localized line of expression in the ectocranial
mesenchyme site retracted anteriorly in the lacZ
+/-
; Twist
+/-
embryos when compared
to wild type embryos. This change in expression was very similar to that observed
previously when the endogenous ephrinA4 protein was examined with an
anti-ephrinA4 antibody (figure 2-6 and Merrill et al, 2006), suggesting that Twist
regulates Ephrin-A4 expression through elements located within the 1.6 kb promoter
region.

Discussion
Significant progress has been made in recent years in elucidating the
developmental roles of the largest family

of mammalian receptor tyrosine kinases,
Eph, and their membrane-bound ephrin ligands. With a wide range of expression in
the developing embryos, the Eph-ephrin signaling triggers a variety of cellular
responses, such as cell adhesion/repulsion, boundary formation and the coordination
of growth, differentiation, and patterning of almost every organ and tissue
62
Figure 2-5: Histochemical analysis of β β β β-galactosidase activity in the region of
coronal suture of 1.6 kb-hsp68-pksLacZ transgenic embryos.

(A, B) At E14.5, transverse sections through the head region of 1.6
kb-hsp68-pksLacZ transgenic embryos reveals a similar expression pattern to that
observed in the transient transgenic lines as well as the endogenous ephrin-A4.
Note that expanded expression of β-galactosidase activity in the suture
mesenchymal layer and periosteum were found in the stable transgenic line. (C, D)
Adjacent sections were stained for ALP to visualize the osteogenic mesenchyme.
fb, frontal bone; pb, parietal bone; cs, coronal suture.

63
Figure 2-6: Expression level of 1.6 kb-hsp68-pksLacZ was reduced in the Twist
heterozygous mutant.

The expression of 1.6 kb-hsp68-pksLacZ transgene was found on the ecto-cranial
mesenchyme in wild type (A, arrowhead); while in Twist, expression is restricted
anteriorly (B, arrowhead). (C, D) Adjacent sections were stained for ALP to
visualize the osteogenic mesenchyme. fb, frontal bone; pb, parietal bone; cs,
coronal suture.

64
(Palmer and Klein, 2003). Our work on cis-regulatory sequences of ephrinA4 has
expanded the understanding of the developmental regulation of this important class
of genes.

A previous study of the human EFNA4 promoter identified a fragment required
for maximal promoter activity in human haemopoietic cell lines However, this
activity was found by means of biochemical studies and transfection assays in tissue
culture cells (Munthe and Aasheim, 2002). Proper protein-DNA interactions that
occur on an endogenous promoter may not take place due to abnormal chromatin
structure in transiently transfectged DNA. Here we have used a phylogenetic
footprinting approach combined with transgenic analysis to identify regulatory
elements for the ephrin-A4 gene. Cross-species DNA sequence comparisons are a
valuable approach for identifying putative regulatory regions of genes. Phylogenetic
footprinting relies on the hypothesis that regulatory elements tend to evolve at a
slower rate than surrounding non-coding sequences due to selective pressure. Having
been maintained throughout evolution, these conserved elements likely play an
important role in gene regulation (Dickmeis and Muller, 2005; Duret and Bucher,
1997; Maeder et al., 2007; Zhang and Gerstein, 2003).

The Ephrin-A4 gene is clustered with ephrin-A3 and ephrin-A1 on human
chromosome 1q21–q22 and mouse chromosome 3. These three genes appear to have
arisen by gene duplication in the vertebrate lineage. The evolutionary relatedness of
ephrin-A4, -A3 and -A1 is supported not only by their close physical localization on
65
chromosomes but also by their degree of protein-coding sequence homology
(38~40% amino acid identity) and their functional properties (Cerretti et al., 1996;
Kozlosky et al., 1995). By using multi-species comparison of the clustered regions of
ephrin-A4, -A3 and -A1 on chromosome 3, we also show conservation of the
surrounding genomic regions of these 3 genes.

We have identified a regulatory fragment in the ephrin-A4 locus that controls
expression in the embryonic head, somites, and limb. These sites of expression
represent a subset of endogenous ephrin-A4 expression patterns. Thus in contrast to
the expression of the endogenous gene, lacZ was not expressed in the ectoderm of
the branchial arches or the limb bud at E9.5 (Flenniken et al., 1996). Taken together,
our results suggest that ephrin-A4 expression is controlled by many distinct, modular
elements dispersed across the locus. This needs a bit more development.

Based on the expression patterns of Twist and Mx2, as well as our previous
findings that heterozygous mutations in Twist caused expanded expression of Msx2
in the osteogenic mesenchyme and down-regulated expression of ephrin-A4 within
the Msx2 ectopic-expressed region of Twist
+/-
embryos (Merrill et al., 2006), we
hypothesized that Msx2 and Twist act to regulate ephrinA4 gene expression. In this
study, we began with an analysis of transcription factor binding sites. This revealed
putative Msx2 and Twist binding sequences, consistent with our hypothesis that Msx2
and Twist coregulate ephrinA4 gene expression. We also used ChIP to show that
Twist and Msx2 interact directly with conserved genomic regions of ephrin-A4.
66
Interestingly, we also detected recruitment of Msx2 to a segment containing the
non-conserved TAAT sequence within the 16 kb region, raising the possibilities of
direct interaction or indirect association with this region through other transcription
factors. In vitro binding assays such as EMSA will help to confirm whether Twist
and Msx2 have a strong preference to these putative binding sites.

We analyzed the conserved genomic regions flanking the mouse ephrin-A4
gene for their transcription activity by generating transient transgenic embryos
carrying these fragments fused with an hsp68-LacZ reporter. It is striking that among
four conserved blocks selected for analysis, only the least conserved of these—the
1.6 kb fragment—possesses consistent regulatory activity in transgenic lines. While
comparative genomics may help identify areas which may be functionally significant,
there are still some limitations of in silico approaches. Regions not conserved
between species may still have important functions.

Twist has been shown physically interact with homeobox proteins including
Msx2 and Hoxa5 (Merrill, Dissertation 2006; (Stasinopoulos et al., 2005).
Downregulated expression of the 1.6kb-hsp68-LacZ transgene in the Twist
+/-
mutant
mice raises the possibility that Twist regulates Ephrin-A4 expression through
interacting with Msx2 protein on a TAAT homeodomain binding core sequence
located within the 1.6 kb promoter region. However, our ChIP results showed only
presence of Msx2 but lack of direct association of Twist to the 1.6 kb region,
suggesting, first, ephrin-A4 is not an immediately downstream target of Twist;
67
second, Twist regulates ephrin-A4 expression through regulating expression of Msx2,
which subsequently down regulates ephrin-A4 expression through binding to
elements within the 1.6 kb fragment (data not shown).

In summary, we have identified a 1.6 kb fragment upstream of the ephrin-A4
gene that contains elements sufficient for cell-type specific gene expression in the
skull. We also showed that the pattern of 1.6 kb-hsp68-LacZ expression in the
developing skull changed in the Twist mutant in a manner similar to the endogenous
gene, suggesting that Twist regulates Ephrin-A4 expression through elements located
within the 1.6 kb promoter region. Mutagenesis of these elements may provide a
more detailed understanding of the molecular mechanisms of spatio-temporal
expression control of the ephrin gene and its role in skull development.
68
Bibliography

Aasheim, H. C., et al., 2000. A splice variant of human ephrin-A4 encodes a soluble
molecule that is secreted by activated human B lymphocytes. Blood. 95, 221-30.

Adams, J. C., 1992. Biotin amplification of biotin and horseradish peroxidase signals
in histochemical stains. J Histochem Cytochem. 40, 1457-63.

Arvanitis, D., Davy, A., 2008. Eph/ephrin signaling: networks. Genes Dev. 22,
416-29.

Barrios, A., et al., 2003. Eph/Ephrin signaling regulates the
mesenchymal-to-epithelial transition of the paraxial mesoderm during somite
morphogenesis. Curr Biol. 13, 1571-82.

Begum, S., et al., 2005. Cell-type-specific regulation of distinct sets of gene targets
by Pax3 and Pax3/FKHR. Oncogene. 24, 1860-72.

Bialek, P., et al., 2004. A twist code determines the onset of osteoblast differentiation.
Dev Cell. 6, 423-35.

Blair, S. S., 1995. Compartments and appendage development in Drosophila.
Bioessays. 17, 299-309.

Bradley, J. P., et al., 1996. Studies in cranial suture biology: in vitro cranial suture
fusion. Cleft Palate Craniofac J. 33, 150-6.

Bruhl, T., et al., 2004. Homeobox A9 transcriptionally regulates the EphB4 receptor
to modulate endothelial cell migration and tube formation. Circ Res. 94, 743-51.

Cerretti, D. P., et al., 1996. The genes encoding the eph-related receptor tyrosine
kinase ligands LERK-1 (EPLG1, Epl1), LERK-3 (EPLG3, Epl3), and LERK-4
(EPLG4, Epl4) are clustered on human chromosome 1 and mouse chromosome 3.
Genomics. 33, 277-82.
69

Chen, J., Ruley, H. E., 1998. An enhancer element in the EphA2 (Eck) gene
sufficient for rhombomere-specific expression is activated by HOXA1 and HOXB1
homeobox proteins. J Biol Chem. 273, 24670-5.

Chen, Z. F., Behringer, R. R., 1995. twist is required in head mesenchyme for cranial
neural tube morphogenesis. Genes Dev. 9, 686-99.

Cobb, J., Duboule, D., 2005. Comparative analysis of genes downstream of the Hoxd
cluster in developing digits and external genitalia. Development. 132, 3055-67.

Cohen, M. M., Jr., MacLean, R. E., 1999. Should syndromes be defined
phenotypically or molecularly? Resolution of the dilemma. Am J Med Genet. 86,
203-4.

Cooke, J. E., et al., 2005. EphA4 is required for cell adhesion and
rhombomere-boundary formation in the zebrafish. Curr Biol. 15, 536-42.

Cunningham, M. L., et al., 2007. Syndromic craniosynostosis: from history to
hydrogen bonds. Orthod Craniofac Res. 10, 67-81.

Dahmann, C., Basler, K., 1999. Compartment boundaries: at the edge of
development. Trends Genet. 15, 320-6.

Dahmann, C., Basler, K., 2000. Opposing transcriptional outputs of Hedgehog
signaling and engrailed control compartmental cell sorting at the Drosophila A/P
boundary. Cell. 100, 411-22.

Danielian, P. S., et al., 1998. Modification of gene activity in mouse embryos in utero
by a tamoxifen-inducible form of Cre recombinase. Curr Biol. 8, 1323-6.

Dickmeis, T., Muller, F., 2005. The identification and functional characterisation of
conserved regulatory elements in developmental genes. Brief Funct Genomic
Proteomic. 3, 332-50.

70
Dottori, M., et al., 1998. EphA4 (Sek1) receptor tyrosine kinase is required for the
development of the corticospinal tract. Proc Natl Acad Sci U S A. 95, 13248-53.

Drescher, U., 2002. Eph family functions from an evolutionary perspective. Curr
Opin Genet Dev. 12, 397-402.

Durbin, L., et al., 1998. Eph signaling is required for segmentation and
differentiation of the somites. Genes Dev. 12, 3096-109.

Duret, L., Bucher, P., 1997. Searching for regulatory elements in human noncoding
sequences. Curr Opin Struct Biol. 7, 399-406.

Eberhart, J., et al., 2000. Expression of EphA4, ephrin-A2 and ephrin-A5 during
axon outgrowth to the hindlimb indicates potential roles in pathfinding. Dev
Neurosci. 22, 237-50.

el Ghouzzi, V ., et al., 1997. Mutations of the TWIST gene in the Saethre-Chotzen
syndrome. Nat Genet. 15, 42-6.

Feldheim, D. A., et al., 2000. Genetic analysis of ephrin-A2 and ephrin-A5 shows
their requirement in multiple aspects of retinocollicular mapping. Neuron. 25,
563-74.

Flenniken, A. M., et al., 1996. Distinct and overlapping expression patterns of
ligands for Eph-related receptor tyrosine kinases during mouse embryogenesis. Dev
Biol. 179, 382-401.

Gale, N. W., et al., 1996. Eph receptors and ligands comprise two major specificity
subclasses and are reciprocally compartmentalized during embryogenesis. Neuron.
17, 9-19.

Guenou, H., et al., 2005. 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. 14, 1429-39.

71
Hatta, T., et al., 2004. Application of the mouse exo utero development system in the
study of developmental biology and teratology. Congenit Anom (Kyoto). 44, 2-8.

Himanen, J. P., et al., 2004. Repelling class discrimination: ephrin-A5 binds to and
activates EphB2 receptor signaling. Nat Neurosci. 7, 501-9.

Howard, T. D., et al., 1997. Mutations in TWIST, a basic helix-loop-helix
transcription factor, in Saethre-Chotzen syndrome. Nat Genet. 15, 36-41.

Huberman, A. D., et al., 2005. Ephrin-As mediate targeting of eye-specific
projections to the lateral geniculate nucleus. Nat Neurosci. 8, 1013-21.

Irvine, K. D., Rauskolb, C., 2001. Boundaries in development: formation and
function. Annu Rev Cell Dev Biol. 17, 189-214.

Jabs, E. W., et al., 1993. A mutation in the homeodomain of the human MSX2 gene
in a family affected with autosomal dominant craniosynostosis. Cell. 75, 443-50.

Jiang, X., et al., 2002. Normal fate and altered function of the cardiac neural crest
cell lineage in retinoic acid receptor mutant embryos. Mech Dev. 117, 115-22.

Johnson, D., et al., 2000. 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. 91, 341-5.

Karsenty, G., 1998. Genetics of skeletogenesis. Dev Genet. 22, 301-13.

Kim, H. J., et al., 1998. FGF-, BMP- and Shh-mediated signalling pathways in the
regulation of cranial suture morphogenesis and calvarial bone development.
Development. 125, 1241-51.

Klein, R., 2004. Eph/ephrin signaling in morphogenesis, neural development and
plasticity. Curr Opin Cell Biol. 16, 580-9.

72
Komaki, M., et al., 2007. Twist negatively regulates osteoblastic differentiation in
human periodontal ligament cells. J Cell Biochem. 100, 303-14.

Komori, T., et al., 1997. Targeted disruption of Cbfa1 results in a complete lack of
bone formation owing to maturational arrest of osteoblasts. Cell. 89, 755-64.

Kothary, R., et al., 1989. Inducible expression of an hsp68-lacZ hybrid gene in
transgenic mice. Development. 105, 707-14.

Kozlosky, C. J., et al., 1995. Ligands for the receptor tyrosine kinases hek and elk:
isolation of cDNAs encoding a family of proteins. Oncogene. 10, 299-306.

Kullander, K., Klein, R., 2002. Mechanisms and functions of Eph and ephrin
signalling. Nat Rev Mol Cell Biol. 3, 475-86.

Kwang, S. J., et al., 2002. Msx2 is an immediate downstream effector of Pax3 in the
development of the murine cardiac neural crest. Development. 129, 527-38.

Lee, M. S., et al., 1999. TWIST, a basic helix-loop-helix transcription factor, can
regulate the human osteogenic lineage. J Cell Biochem. 75, 566-77.

Lenhard, B., et al., 2003. Identification of conserved regulatory elements by
comparative genome analysis. J Biol. 2, 13.

Liu, Y . H., et al., 1994. Regulation of the Msx2 homeobox gene during mouse
embryogenesis: a transgene with 439 bp of 5' flanking sequence is expressed
exclusively in the apical ectodermal ridge of the developing limb. Mech Dev. 48,
187-97.

Ma, H., et al., 2003. Study of nuclear receptor-induced transcription complex
assembly and histone modification by chromatin immunoprecipitation assays.
Methods Enzymol. 364, 284-96.

Maeder, M. L., et al., 2007. Phylogenetic footprinting analysis in the upstream
regulatory regions of the Drosophila enhancer of split genes. Genetics. 177, 1377-94.
73

Martinez, A., Soriano, E., 2005. Functions of ephrin/Eph interactions in the
development of the nervous system: emphasis on the hippocampal system. Brain Res
Brain Res Rev. 49, 211-26.

McMahon, J. A., et al., 1998. Noggin-mediated antagonism of BMP signaling is
required for growth and patterning of the neural tube and somite. Genes Dev. 12,
1438-52.

Merrill, A. E., et al., 2006. Cell mixing at a neural crest-mesoderm boundary and
deficient ephrin-Eph signaling in the pathogenesis of craniosynostosis. Hum Mol
Genet. 15, 1319-28.

Mori, T., et al., 1995. Differential expressions of the eph family of receptor tyrosine
kinase genes (sek, elk, eck) in the developing nervous system of the mouse. Brain
Res Mol Brain Res. 29, 325-35.

Moss, A., et al., 2005. Ephrin-A4 inhibits sensory neurite outgrowth and is regulated
by neonatal skin wounding. Eur J Neurosci. 22, 2413-21.

Muneoka, K., et al., 1986. Mouse embryos develop normally exo utero. J Exp Zool.
239, 289-93.

Munthe, E., Aasheim, H. C., 2002. Characterization of the human ephrin-A4
promoter. Biochem J. 366, 447-58.

Murray, S. S., et al., 1992. Expression of helix-loop-helix regulatory genes during
differentiation of mouse osteoblastic cells. J Bone Miner Res. 7, 1131-8.

Myers, C., et al., 2000. Homeobox B3 promotes capillary morphogenesis and
angiogenesis. J Cell Biol. 148, 343-51.

Nakajima, Y ., et al., 2006. Identification of Epha4 enhancer required for segmental
expression and the regulation by Mesp2. Development. 133, 2517-25.

74
Nieto, M. A., et al., 1992. A receptor protein tyrosine kinase implicated in the
segmental patterning of the hindbrain and mesoderm. Development. 116, 1137-50.

Otto, F., et al., 1997. Cbfa1, a candidate gene for cleidocranial dysplasia syndrome,
is essential for osteoblast differentiation and bone development. Cell. 89, 765-71.

Palmer, A., Klein, R., 2003. Multiple roles of ephrins in morphogenesis, neuronal
networking, and brain function. Genes Dev. 17, 1429-50.

Paratore, C., et al., 1999. Embryonic gene expression resolved at the cellular level by
fluorescence in situ hybridization. Histochem Cell Biol. 111, 435-43.

Pasquale, E. B., 2005. Eph receptor signalling casts a wide net on cell behaviour. Nat
Rev Mol Cell Biol. 6, 462-75.

Pasquale, E. B., 2008. Eph-ephrin bidirectional signaling in physiology and disease.
Cell. 133, 38-52.

Poliakov, A., et al., 2004. Diverse roles of eph receptors and ephrins in the regulation
of cell migration and tissue assembly. Dev Cell. 7, 465-80.

Pratt, R. L., Kinch, M. S., 2002. Activation of the EphA2 tyrosine kinase stimulates
the MAP/ERK kinase signaling cascade. Oncogene. 21, 7690-9.

Rhoads, K., et al., 2005. A role for Hox A5 in regulating angiogenesis and vascular
patterning. Lymphat Res Biol. 3, 240-52.

Rice, D. P., et al., 2000. Integration of FGF and TWIST in calvarial bone and suture
development. Development. 127, 1845-55.

Rice, D. P., et al., 1999. Apoptosis in murine calvarial bone and suture development.
Eur J Oral Sci. 107, 265-75.

Rosen, V ., et al., 1994. Responsiveness of clonal limb bud cell lines to bone
morphogenetic protein 2 reveals a sequential relationship between cartilage and bone
75
cell phenotypes. J Bone Miner Res. 9, 1759-68.

Ryoo, H. M., et al., 2006. Critical molecular switches involved in BMP-2-induced
osteogenic differentiation of mesenchymal cells. Gene. 366, 51-7.

Saga, Y ., et al., 1999. MesP1 is expressed in the heart precursor cells and required for
the formation of a single heart tube. Development. 126, 3437-47.

Salsi, V ., Zappavigna, V ., 2006. Hoxd13 and Hoxa13 directly control the expression
of the EphA7 Ephrin tyrosine kinase receptor in developing limbs. J Biol Chem. 281,
1992-9.

Schulte, D., Cepko, C. L., 2000. Two homeobox genes define the domain of EphA3
expression in the developing chick retina. Development. 127, 5033-45.

Serbedzija, G. N., et al., 1992. Vital dye analysis of cranial neural crest cell migration
in the mouse embryo. Development. 116, 297-307.

Soriano, P., 1999. Generalized lacZ expression with the ROSA26 Cre reporter strain.
Nat Genet. 21, 70-1.

Stadler, H. S., et al., 2001. Loss of Eph-receptor expression correlates with loss of
cell adhesion and chondrogenic capacity in Hoxa13 mutant limbs. Development. 128,
4177-88.

Stasinopoulos, I. A., et al., 2005. HOXA5-twist interaction alters p53 homeostasis in
breast cancer cells. J Biol Chem. 280, 2294-9.

Steinberg, M. S., Takeichi, M., 1994. Experimental specification of cell sorting,
tissue spreading, and specific spatial patterning by quantitative differences in
cadherin expression. Proc Natl Acad Sci U S A. 91, 206-9.

Surawska, H., et al., 2004. The role of ephrins and Eph receptors in cancer. Cytokine
Growth Factor Rev. 15, 419-33.

76
Theil, T., et al., 1998. Segmental expression of the EphA4 (Sek-1) receptor tyrosine
kinase in the hindbrain is under direct transcriptional control of Krox-20.
Development. 125, 443-52.

Thomas, J. W., Green, E. D., 2003. Comparative sequence analysis of a single-gene
conserved segment in mouse and human. Mamm Genome. 14, 673-8.

Twigg, S. R., et al., 2004. Mutations of ephrin-B1 (EFNB1), a marker of tissue
boundary formation, cause craniofrontonasal syndrome. Proc Natl Acad Sci U S A.
101, 8652-7.

Vindis, C., et al., 2003. EphB1 recruits c-Src and p52Shc to activate MAPK/ERK
and promote chemotaxis. J Cell Biol. 162, 661-71.

Warren, S. M., et al., 2003. The BMP antagonist noggin regulates cranial suture
fusion. Nature. 422, 625-9.

Wasserman, W. W., Sandelin, A., 2004. Applied bioinformatics for the identification
of regulatory elements. Nat Rev Genet. 5, 276-87.

Wilkie, A. O., 1997. Craniosynostosis: genes and mechanisms. Hum Mol Genet. 6,
1647-56.

Wilkie, A. O., Morriss-Kay, G. M., 2001. Genetics of craniofacial development and
malformation. Nat Rev Genet. 2, 458-68.

Xu, Q., et al., 1999. In vivo cell sorting in complementary segmental domains
mediated by Eph receptors and ephrins. Nature. 399, 267-71.

Yang, H., et al., 1999. An optimized method for in situ hybridization with signal
amplification that allows the detection of rare mRNAs. J Histochem Cytochem. 47,
431-46.

Yoshida, T., 2005. [Growth pattern of the frontal bone primordium and involvement
of Bmps in this process]. Kokubyo Gakkai Zasshi. 72, 19-27.
77

Yoshida, T., et al., 2005. Twist is required for establishment of the mouse coronal
suture. J Anat. 206, 437-44.

Yousfi, M., et al., 2002. Twist haploinsufficiency in Saethre-Chotzen syndrome
induces calvarial osteoblast apoptosis due to increased TNFalpha expression and
caspase-2 activation. Hum Mol Genet. 11, 359-69.

Yousfi, M., et al., 2001. Increased bone formation and decreased osteocalcin
expression induced by reduced Twist dosage in Saethre-Chotzen syndrome. J Clin
Invest. 107, 1153-61.

Zhang, Z., Gerstein, M., 2003. Of mice and men: phylogenetic footprinting aids the
discovery of regulatory elements. J Biol. 2, 11.

Abstract (if available)

Abstract Eph-ephrin signaling has been implicated in craniosynostosis in humans: Mutations in the ephrin ligands, EFNA4 and EFNB1, are known to cause craniosynostosis. Our present work shows that Twist and Msx2, also craniosynostosis genes, interact functionally with the Eph-ephrin signaling pathway. Reduced expression of ephrinA2, ephrinA4 and EphA4 in a layer of cells located along the neural-crest mesoderm boundary was previously described associating with the upregulated expression of Msx2 in osteogenic mesenchyme and expanded into the non-osteogenic mesoderm domain in Twist mutants. And the expression of Eph-ephrin genes, as well as the boundary defect, can be rescued by genetic inactivation of Msx2 in the context of the Twist+/- genotype. Here we demonstrate that EphA4 mutant mice phenocopy Twist mutants: Wnt1-Cre and Mesp1-Cre marking of neural crest and mesoderm show that EphA4 mutant embryos exhibit defects in the neural crest-mesoderm boundary within the coronal suture. Osteogenic neural crest-derived cells mix with non-osteogenic mesodermal cells fated to form the suture. Associated with this mixing is the development of ectopic bone in the suture.

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

Functions of Msx2 and Msx1 during mouse skull and hair follicle development

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

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

PDF

Novel roles for Maf1 in embryonic stem cell differentiation and adipogenesis

PDF

Studies of murine prostate cancer stem / progenitor cells

PDF

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

PDF

Retinoic acid and TGFβ signaling regulate cardiovascular development

PDF

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

PDF

Construction of plasmids for constitutive and induced expression of secreted alkaline phosphatase in mammalian ovarian carcinoma cells

PDF

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

PDF

Epigenetic checks and balances: PHF6 activity restricts neural crest migration

PDF

Pten deletion in adult pancreatic beta-cells induces cell proliferation and G1/S cell cycle progression

PDF

Identification and characterization of adult stem cells in the oral cavity

PDF

Protein arginine methyltransferases in murine skull development

PDF

Enhanced Burkitt 's lymphoma cell killing by the combination treatments of bortezomib with celecoxib and 2,5-dimethyl-celecoxib (DMC)

PDF

Genomic stability of transcriptionally targeted replication competent retroviral vectors

PDF

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

PDF

The mechanism of mammalian immunoglobulin class switch recombination: R-loop structures and activation-induced deaminase site preferences

PDF

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

Asset Metadata

Creator Ting, Man-Chun (author)

Core Title Lineage boundaries and cell migration in the patterning of the mammalian skull

School Keck School of Medicine

Degree Doctor of Philosophy

Degree Program Biochemistry and Molecular Biology

Degree Conferral Date 2008-08

Publication Date 07/24/2008

Defense Date 06/18/2008

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

Tag craniosynostosis,EphA4,OAI-PMH Harvest,Twist

Language English

Advisor Maxson, Robert E. (committee chair), Stallcup, Michael R. (committee member), Sucov, Henry (committee member)

Creator Email manchunt@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m1387

Unique identifier UC187326

Identifier etd-Ting-20080724 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-196690 (legacy record id),usctheses-m1387 (legacy record id)

Legacy Identifier etd-Ting-20080724.pdf

Dmrecord 196690

Document Type Dissertation

Rights Ting, Man-Chun

Type texts

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

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu