University of Southern California Dissertations and Theses

Integrin expression and signaling during palatal fusion

(USC Thesis Other)

Integrin expression and signaling during palatal fusion

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content INTEGRIN EXPRESSION AND SIGNALING DURING PALATAL FUSION

by

Daniela Schmid

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
(CRANIOFACIAL BIOLOGY)

August 2008

Copyright 2008 Daniela Schmid

ii

TABLE OF CONTENTS
LIST OF FIGURES iv
ABBREVIATIONS v
ABSTRACT vii
CHAPTER 1: INTRODUCTION 1
Palate Development 1
Histology 2
Fate of the MEE 3
Molecular control of palatal fusion – Apoptosis vs. EMT 4
EMT 9
Cell Adhesion 10
E-cadherin 10
Integrins 12
Integrin-Linked Kinase 14
Glycogen synthase kinase-3β 17
PKB/Akt 18
CHAPTER 2: HYPOTHESIS 20
CHAPTER 3 21
Immunolocalization of endogenous E-cadherin in palate tissue in vivo. 21
CHAPTER 4 24
Immunolocalization of endogenous β1 integrin in palate tissue in vivo. 24
β1 integrin siRNA reduces β1 integrin in 3T3-Swiss Cells and MEE. 27
β1 integrin silencing inhibits palatal fusion. 29
CHAPTER 5 32
Immunolocalization of endogenous ILK in palate tissue in vivo. 32
ILK is expressed in palate epithelium and disappears with MEE involution
in vivo and in vitro. 34
ILK siRNA specifically reduces ILK in 3T3-Swiss cells and MEE cells. 35
ILK silencing inhibits palatal fusion. 38
ILK signals through GSK-3β and Akt in cultured palates. 41
CHAPTER 6: DISCUSSION 43
Summary of possible signaling cascades activated in MEE during EMT 53
iii

CHAPTER 7: MATERIALS AND METHODS 58
Animals 58
Transfection of siRNA 59
Cell culture 60
Assessment of palatal fusion 60
Immunoblotting 61
Immunohistochemistry 62
REFERENCES 63

iv

LIST OF FIGURES
Figure 1: Schematic of ILK primary structure 15
Figure 2: Immunolocalization of endogenous E-cadherin in palate tissue in vivo. 23
Figure 3: Immunohistochemical detection of β1 integrin in the developing palate
in vivo. 25
Figure 4: Immunolocalization of β1 integrin in palate tissues in vitro. 27
Figure 5: β1 integrin expression is reduced in siRNA transfected 3T3-Swiss cells
and palate tissue. 28
Figure 6: Effects of β1 integrin siRNA treatment on the disappearance of the
epithelial midline seam. 31
Figure 7: Immunolocalization of ILK in palate tissues in vivo. 33
Figure 8: Immunolocalization of ILK in palate tissues in vitro. 35
Figure 9: ILK expression is reduced by siRNA. 37
Figure 10: Immunofluorescence of DY-547 labeled control siRNA. 37
Figure 11: Effects of ILK siRNA treatment on involution of the medial edge
epithelium. 40
Figure 12: ILK siRNA transfection reduces ILK signaling. 42
Figure 13: Schematic overview for the roles of E-cadherin, β1 integrin and ILK
during palate development and possible activation of signaling pathways. 53

v

ABBREVIATIONS
BM Basement membrane
β-gal β-galactosidase
CCFSE Carboxydichlorofluorescein diacetate succinimidyl ester
DiI 1,1-dioctadecyl-3,3,3’,3’-tetramethylindocarbocyanine
ECM Extracellular matrix
EGF Epidermal growth factor
EMT Epithelial-mesenchymal transformation
Erk Extracellular signal-regulated kinase
GAPDH Glyceraldehyde-3-phosphate dehydrogenase
GSK-3β Glycogen synthase 3 beta
ILK Integrin-linked kinase
MEE Medial edge epithelia
MMP Matrix metalloproteinase
PCD Programmed cell death
PI3K Phosphatidylinositol-3 kinase
PINCH Particularly interesting Cys-His-rich protein
PIP2 Phosphatidylinositol 4,5-bisphosphate
PIP3 Phosphatidylinositol 3,4,5-trisphosphate
vi

PKB Protein kinase B
siRNA Small interfering RNA
TCF/LEF T-cell factor/lymphoid enhancer factor
TEM Transmission electron microscopy
TGF-β Transforming growth factor β

vii

ABSTRACT
The fusion of the secondary palate is a complex event requiring epithelial
and mesenchymal cell differentiation. Palatal fusion and mesenchymal
confluence entail adhesion of the opposing medial edge epithelia (MEE),
extracellular matrix (ECM) remodeling and MEE disappearance. In this process
the MEE are believed to undergo epithelial-mesenchymal transformation (EMT).
While E-cadherin mediates cell-cell adhesion, integrins are the major cell surface
receptors. Integrins transduce ECM signals into the cell by associating with
adaptor proteins such as integrin-linked kinase (ILK). Integrin and growth factor
receptor signaling has been implicated

in the regulation of EMT. We
hypothesized that integrin signaling is necessary for palatal fusion. To assess this,
we examined E-cadherin, β1 integrin and ILK expression. Immunohistochemistry
demonstrated that E-cadherin expression was downregulated in the MEE prior to
EMT. During palatal fusion, β1 integrin strongly immunolocalized to the MEE and
oral epithelia while ILK expression was restricted to the MEE. Silencing of either
β1 integrin or ILK by siRNA transfection into palate organ culture resulted in the
persistence of MEE cells in the midline. Western blot confirmed the protein
reduction. Finally, epidermal growth factor-induced ILK signaling was assessed in
palate cultures treated with either ILK siRNA or non-silencing control siRNA. The
resulting phosphorylation of the known ILK downstream mediators GSK-3β and
viii

Akt was then assessed. ILK silencing significantly reduced the phosphorylation of
both proteins. These results suggest that palatal fusion and the onset of
mesenchymal transdifferentiation of the MEE is correlated with the
downregulation of E-cadherin, β1 integrin and integrin-linked kinase. Integrin
signaling through integrin-linked kinase and its downstream partners GSK-3β and
Akt is crucial for EMT as reduction of β1 integrin and integrin-linked kinase
expression prevented palatal fusion resulting in a cleft palate.

1

CHAPTER 1: INTRODUCTION
Palate Development
Development of the mammalian secondary palate is a complex process that is
frequently disturbed resulting in a cleft palate (Ferguson, 1988; Ferguson, 1984). A fused
palate is necessary for proper suction in newborns, mastication, speech development
and articulation. Morphogenetic events involved in the normal process of palatogenesis
include ECM synthesis and degradation (Brinkley and Morris-Wiman, 1984),
neurotransmitter synthesis (Zimmerman and Wee, 1984), cell adhesion (Greene and
Pratt, 1977), epithelial-mesenchymal interactions (Ferguson, 1988; Shuler et al., 1991;
Shuler et al., 1992) and regional patterning. Disruption at any stage in this complex
process can result in a cleft palate. Embryologically, the secondary palate arises as
bilateral projections of the maxillary processes, which are derivates of the first branchial
arch (Ferguson, 1988; Ferguson, 1994; Greene, 1989; Shuler, 1995; Shuler et al., 1991;
Tyler and Koch, 1975). Formation of the secondary palate begins about the sixth week of
gestation in humans and about the twelfth day twelfth day post-coitus (E12) in mice.
The two palatal processes initially grow vertically down the lateral sides of the tongue
and then at around the seventh and eighth week of human development (E14.5 in
mice), reorient to a horizontal position above the dorsum of the tongue. After elevation
the palatal shelves approximate and contact each other in the region of the second ruga
and fusion then spreads both posteriorly and anteriorly. Epithelial cells of the medial
2

edge of the approximating palatal shelves adhere to form a midline epithelial seam
(Ferguson, 1988; Sharpe and Ferguson, 1988).

Histology
The two palatal shelves consist of neural crest derived mesenchymal extensions
covered by ectodermal cells (Ferguson, 1988). The medial edge epithelium (MEE) is a
subpopulation of the palatal surface epithelium and is two-cells thick prior to elevation.
The basal layer is cuboidal in shape and the overlying periderm layer consists of flat
cells. The basement membrane (BM) is continuous and is contacted by numerous
mesenchymal cells in the underlying connective tissue. Shortly after the shelves assume
their horizontal position, the previously flat cells of the periderm become irregular in
shape, protrude from the surface, and are sloughed from the outer surface (Fitchett and
Hay, 1989). The MEE initiates the contact and adhesion of the opposing palatal shelves.
MEE adherence is specific, and does not normally fuse with other epithelia (Ferguson,
1988). At the time of shelf fusion, the midline seam consists of the two adjoined basal
cell layers. With growth of the palate, the midline seam thins to one layer of cells.
Subsequently, the epithelial seam and BM involute, establishing a continuous
population of mesenchymal cells that separate the oral and nasal cavities (Fitchett and
Hay, 1989). The ectodermal cells on the nasal side of the palate differentiate into a
pseudostratified epithelium while cells on the oral side develop into a squamous
3

epithelium with stratified and keratinized cells (Ferguson, 1988). The fusion with the
nasal septum initiates anteriorly around the ninth week and is completed by the twelfth
week (E15 in mice).

Fate of the MEE
The fate of the MEE during palatogenesis has been controversial. The
disappearance of the MEE from the midline during palatal fusion was for many years
ascribed to programmed cell death (PCD). The PCD hypothesis for the fate of the MEE
was based on three lines of evidence: cessation of DNA synthesis (Hudson and Shapiro,
1973; Pratt and Martin, 1975); increase in lysosomal enzymes in the MEE (Greene and
Pratt, 1976); and the presence of cells with ultrastructural features consistent with cell
death (Chaudhry and Shah, 1973; Farbman, 1968). Alternatively, MEE cells may migrate
away from the medial edge to populate the nasal and oral epithelia (Carette and
Ferguson, 1992). Finally, MEE cells may transform into mesenchymal cells in a process
termed epithelial-mesenchymal transdifferentiation (EMT). In 1989, Fitchett and Hay
completed an ultrastructural characterization of the MEE morphology during palatal
fusion (Fitchett and Hay, 1989). This study was followed by cell lineage analysis of the
MEE using vital cell dyes (Carette and Ferguson, 1992; Shuler, 1995; Shuler et al., 1991;
Shuler et al., 1992). The cell lineage analyses permitted the observation of the
populations of cells throughout the different stages of palatogenesis. The results of
4

these studies documented that the MEE undergo phenotypic transformation coincident
with the mesenchymal confluence between the two opposing palatal shelves. The
phenotypic transformation requires several specific events to occur both in the MEE and
in the surrounding ECM; the cytoskeleton of the MEE changes from keratin to vimentin
intermediate filaments (Fitchett and Hay, 1989; Griffith and Hay, 1992; Shuler et al.,
1991; Shuler et al., 1992); the BM underlying the MEE is degraded such that the MEE
come into contact with a new ECM (Carette and Ferguson, 1992; Fitchett and Hay, 1989;
Griffith and Hay, 1992; Shuler et al., 1991; Shuler et al., 1992); and the MEE stop
expressing epithelial specific cell adhesion molecules such as desmosomes and E-
cadherin (Luning et al., 1994; Mogass et al., 2000; Sun et al., 1998a). The process of
epithelial-mesenchymal transformation thus requires marked changes in the pattern of
gene expression and a molecular mechanism that is unique to the MEE.

Molecular control of palatal fusion – Apoptosis vs. EMT
To this date, the fate of the MEE has been controversial. Apoptosis, EMT and
migration towards the periphery of the midline have been proposed to explain how the
opposing palatal shelves fuse into one continuous structure. Some consistent

features
have been observed in various morphological studies. The sequence of events includes
removal of the superficial flat periderm cells, adhesion of the basal cells, BM

breakdown,
and disappearance of the midline seam.
5

The apoptosis model has long been proposed for MEE seam degeneration. Early
studies have provided ultrastructural and molecular evidence for the occurrence of
apoptosis in the disappearing MEE cells (Chaudhry and Shah, 1973; DeAngelis and
Nalbandian, 1968; Farbman, 1968; Greene and Pratt, 1976; Saunders, 1966; Smiley and
Dixon, 1968). However,

only recently have relevant experimental data been obtained

regarding this hypothesis. Cuervo at al. observed apoptosis

in the MEE seam during
palatal fusion in vivo and in vitro using

TUNEL assay (Cuervo and Covarrubias, 2004;
Cuervo et al., 2002). As to the functional requirement

of apoptosis in mediating seam
degeneration, contradictory results

were reported by different investigators. Cuervo and
Covarrubias

reported that the addition of cell death inhibitors, e.g. z-VAD,

in palate
organ culture prevented seam and BM

degeneration (Cuervo and Covarrubias, 2004). In
contrast,

Takahara et al. stated that palatal shelves treated

with caspase inhibitors,
YVAD-CHO and DEVD-CHO, still underwent

normal fusion (Takahara et al., 2004). Since
both reports are based on in vitro culture experiments and since cell death is highly
sensitive

to the environment and the condition of the cell, one has to be cautious in
applying in vitro cell death data to the in vivo

situation. In addition, specificity is often a
concern with

the use of chemical inhibitors. Very recently, convincing results, using a
Cre-Loxp-based genetic labeling system, where the expression of Cre recombinase was
driven by a cytokeratin 14 (K14) promoter (Vasioukhin and Fuchs, 2001) and where
R26R reporter locus was specifically activated and irreversibly labeled in the MEE,
6

showed apoptosis as the only mechanism of midline seam disintegration (Vaziri Sani et
al., 2005). However, the β-galactosidase (β-gal)-labeling level in the MEE

prior to fusion
was rather low in that reported study, which could

have meant that the signal was not
sufficient to be detected. Using a similar technique, there have been contradictory
results. β-gal staining with K14-Cre; R26R embryos revealed that a significant number of
mesenchymal cells were β-gal positive during and after palate fusion (Jin and Ding,
2006). To find out whether apoptosis is functionally required for seam degeneration in
vivo, Jin and Ding showed that Apaf1 deficient mice, in which caspases 9 and 3 are
inactive, exhibit normal palatal fusion (Jin and Ding, 2006). DNA fragmentation-
mediated apoptosis

was completely abolished in the palates of Apaf1 mutant embryos,

as judged by the TUNEL assay, however, palate fusion

and seam degeneration did occur
in the Apaf1 mutant embryos (Jin and Ding, 2006). These results contradict an earlier
finding with a different line of Apaf1 mutant mice showing the lack of palatal shelf
adhesion (Honarpour et al., 2000) and absence of midline seam disintegration (Cecconi
et al., 1998). However,

the analysis in that study stopped at E14.5, at which time

the
seam in a wild-type embryo is still evident. Knockouts for genes of the apoptotic
pathways have also been problematic. It is, however, noteworthy that some of the
knockout models can demonstrate a surprisingly high degree of phenotypic variability
among individual mouse lines and penetrance of the phenotype in a mixed-background
colony could well be due to the presence of additional modifier loci (Doetschman,
7

1999). Finally, the findings by microarray analysis show a steady downregulation of
major apoptotic genes and an upregulation of anti-apoptotic genes from the time of
shelf adherence to seam disintegration (LaGamba et al., 2005).
EMT has also been extensively studied. However, the experimental data remains
controversial. The existence of EMT was first suggested based on transmission electron
microscopy (TEM) examination and vimentin immunostaining (Fitchett and Hay, 1989)
as well as on in vitro and in utero epithelial cell tracking with the lipophilic molecule
carboxyfluorescein (CCFSE) (Griffith and Hay, 1992). Similarly, in

vivo intraperitoneal DiI
(1,1-dioctadecyl-3,3,3’,3’-tetramethylindocarbocyanine) injection labeled cells in the
mouse palatal mesenchymal area at E15.5 (Shuler et al., 1992). The avian palate does
not fuse in vivo but can be induced to fuse in vitro organ cultures by the addition of
transforming growth factor (TGF)-β3. These embryonic avian palatal shelves were
labeled with CCFSE and highly labeled cells persisted in the mesenchyme which has
replaced the midline epithelial seam (Sun et al., 1998b). Cuervo and Covarrubias using
CCFSE labeling in mouse in vitro found no labeled cells in the mesenchymal region after
fusion (Cuervo and Covarrubias, 2004). The same investigators infected

the palatal
shelves with a lacZ-expressing adenoviral vector

and cultured them in serum-free
medium, and found no β-gal-positive

cells in the mesenchymal region after in vitro

palatal fusion (Cuervo and Covarrubias, 2004). In contrast, Martinez-Alvarez et al.
infected the palate shelves

with a retroviral vector that constitutively expressed the
8

bacterial

lacZ gene, and cultured the infected palatal pairs in vitro

in complete medium
with serum. Following in vitro palate fusion

and β-gal staining, the

investigators
observed blue cells in the mesenchymal region (Martinez-Alvarez et al., 2000). Fitchett
and Hay claimed that cell death occurs only in the superficial peridermal cells that are
sloughing off or trapped in the midline but not in the basal MEE cells (Fitchett and Hay,
1989). If all of the periderm sloughs, no dying cells appear among the basal epithelial
cells forming the midline palatal seam; these cells were negative in TUNEL assay
(Nawshad et al., 2004). In vitro, inappropriate culture conditions can be attributed to
much of the reported basal cell death (Takigawa and Shiota, 2004).
It is likely that apoptotic cell death contributes to efficiently eliminating some
MEE cells that fail to migrate or transform properly. In this sense, the cell death
observed in the midline epithelial seam of the fusing palate may not be genetically
“programmed” but may be induced by some environmental cues or by the state of
differentiation of the cell. It is likely that some environmental and/or intracellular signals
trigger the caspase pathway and initiate the apoptotic process. But given the timeline
and rapid fusion of the palate it is unlikely that apoptosis is the sole mechanism for MEE
removal as only moderate numbers of TUNEL positive cells have been shown in the
MEE. One would expect to be able to detect a majority of cells undergoing apoptosis in
one section if this was the only cell fate. It is reasonable to assume that apoptotic cell
9

death, EMT, and cell migration are all involved in palatal fusion and their occurrence
may be regulated in a spatially and temporally coordinated manner.

EMT
The defining functional changes occurring in cells undergoing EMT are
detachment from neighboring cells and migration into the adjacent tissue. Molecular
hallmarks of EMT include downregulation of E-cadherin leading to the loss of cell–cell
adhesion; upregulation of matrix-degrading proteases; actin cytoskeleton
reorganization; upregulation and/or nuclear translocation of transcription factors
underlying the specific gene program of EMT (reviewed in Thiery and Sleeman, 2006).
Since EMT is a highly complex process, its molecular regulation is similarly
complex. Indeed, various mechanisms leading to specific gene repression and activation,
signal transduction pathways and a multitude of mediator molecules seem to cooperate
in controlling EMT. The molecular analysis of EMT-associated events has revealed
extensive cross-talk between signaling pathways leading to very complex biochemical
circuits. In most studied systems, EMT is regulated by growth factor activation of
tyrosine or serine–threonine kinase receptors, ECM stimulation of integrins and their
downstream signaling pathways that lead to E-cadherin downregulation and/or to
dynamic changes in the cytoskeleton (Huber et al., 2005; Thiery and Sleeman, 2006). E-
cadherin is regarded as a master regulator of the epithelial/mesenchymal phenotype
10

switch. Repression of E-cadherin can both induce and complete EMT and, on the other
hand, its re-activation can result in the reverse process, mesenchymal-epithelial
transition. Indeed, most of the known signal transduction pathways involved in EMT
ultimately converge on the control of E-cadherin expression (reviewed in Thiery and
Sleeman, 2006).

Cell Adhesion
The interactions of cells with neighboring cells and the surrounding ECM are
mediated by different classes of cell adhesion receptors, including integrins, cadherins,
and members of the immunoglobulin and selectin families.

E-cadherin
E-cadherin is a type I cadherin transmembrane glycoprotein that is specifically
expressed on the surface of epithelial cells, where it mediates the formation of cell to
cell adherens junctions, via homophilic, calcium-dependent interactions (Perl et al.,
1998; Shiozaki et al., 1996; Vleminckx et al., 1994). E-cadherin provides a physical link
between adjacent cells and is crucial for the establishment and maintenance of polarity
and the structural integrity of epithelia and, in addition, it is a component of the cellular
signaling network. The extracellular domain interacts with E-cadherin molecules on
11

adjacent cells, and the intracellular domain is associated with a multiprotein complex
comprising α-, β-, and p120-catenin (Kemler, 1993; Takeichi, 1995). β-catenin binds
tightly to the cytoplasmic domain of E-cadherin and, through α-catenin, to the actin
microfilament network of the cytoskeleton, thereby connecting cell-cell adhesion
complexes with the intracellular machinery involved in cell shape and regulation of cell
migration. The binding of E-cadherin with catenins and actin cytoskeleton is essential for
the formation of strong cell-cell adhesion and any event that perturbs the
cadherin/catenin/cytoskeleton complex leads to destabilization of cell-cell adhesion and
reorganization of the actin cytoskeleton (Cavallaro and Christofori, 2004; Christofori and
Semb, 1999; Perez-Moreno et al., 2003). By linking the cells together, E-cadherin based
junctional complexes keep epithelial cells in a stationary, non-motile state. The
downregulation of E-cadherin, during both normal and pathological EMT, leads to
destabilization of the epithelial architecture and allows dissociation of single cells from
their neighbors. Therefore, an understanding of the mechanism which regulate the
function and expression of E-cadherin is critical for the understanding of invasion,
metastasis and EMT.

12

Integrins
In all cell types, the primary group of receptors for ECM proteins are integrins.
Integrins are a family of heterodimeric transmembrane glycoproteins consisting of 8 β
and 18 α subunits, so far known to assemble into 24 distinct integrins. Each subunit is
composed of a large extracellular, a single transmembrane and a short cytoplasmic
domain. An exception to this is the β4 subunit, which has an extended cytoplasmic
domain of over 1000 amino acids (Curley et al., 1999). Integrin expression is cell type
specific and the combination of the α and β subunit determines the ligand specificity of
the integrin. While some integrins selectively recognize primarily a single ECM protein
ligand, others can bind several ligands and it is the combination of the integrin
expression/activation pattern and the availability of ligand that specifies the interactions
in vivo. Integrin ligands are usually members of either the cell surface immunoglobulin
superfamily or are large modular ECM molecules, such as fibronectin, collagen and
laminin (Humphries and Newham, 1998). Genes for the β subunits and all but four of
the α subunits have been knocked out and each phenotype is distinct, reflecting the
different roles of the various integrins (Hynes, 2002). In epithelial cells, most integrins
are β1-containing heterodimers.
Integrins not only mediate cell adhesion to ECM, they also transduce intracellular
signals that promote cell migration as well as cell survival (Meredith et al., 1993;
Schwartz and Shattil, 2000). The short cytoplasmic tails of integrins are devoid of
13

enzymatic features. They transduce signals by associating with adapter proteins that
connect the integrin to the cytoskeleton, cytoplasmic kinases and transmembrane
growth factor kinases. Ligand binding to integrins leads to integrin clustering and
recruitment of actin filaments and signaling proteins to the cytoplasmic domain of
integrins (Hynes, 2002). In the inactive state, it is believed that the α subunit
cytoplasmic tail sterically shields the β subunit tail, and that this association renders the
extracellular parts of the integrin molecule unable to bind to the ligand. Activation is
achieved when intracellular proteins such as talin bind to the β subunit, releasing it from
the α subunit interaction. These conformational changes are then communicated across
the membrane to move the extracellular segments into a conformation able to bind
ligand (Liddington and Ginsberg, 2002; Matlin et al., 2003).
The β1 integrin subunit is able to form functional receptors with the largest
diversity of known α integrins leading to the ability of cells to detect the composition of
diverse ECM environments (Hynes, 2002). It has been reported that β1 integrin is
essential for the maintenance of the lens epithelium (Simirskii et al., 2007). Similarly,
upregulation of β1 integrin is associated with increased invasion and metastasis in
cancer cells (Arboleda et al., 2003; Takenaka et al., 2000). Complete deletion of the β1
integrin gene from mice leads to lethality at the blastocyst stage demonstrating its
importance for development (Fassler and Meyer, 1995; Stephens et al., 1995).
Conditional inactivation of the β1 integrin gene in embryonic skin keratinocytes leads to
14

defective epithelial proliferation in the absence of increased apoptosis coincident with
obvious defects in BM assembly (Brakebusch et al., 2000; Raghavan et al., 2000). These
data demonstrate that β1 integrin is important for the maintenance of the epithelial
phenotype.

Integrin-Linked Kinase
Integrin-linked kinase (ILK) is an intracellular serine/threonine

protein kinase that
interacts with the cytoplasmic domains of

β1 and β3 integrins and numerous
cytoskeleton-associated proteins (Dedhar, 2000; Dedhar et al., 1999; Wu, 2001; Wu and
Dedhar, 2001b). ILK was first identified based on its interaction with the cytoplasmic
domain of the β1 integrin subunit using a yeast two-hybrid screen (Hannigan et al.,
1996). Subsequent co-immunoprecipitation studies also demonstrated an interaction
with the cytoplasmic tail of β3 integrin subunit (Hannigan et al., 1996). ILK is
ubiquitously expressed in mammalian cells and is highly conserved evolutionarily, with
homologues identified in human, mouse, rat, Drosophila, and Caenorhabditis elegans.
The gene encoding human ILK has been localized to human chromosome 11p15.5-p15.4
(Hannigan et al., 1997).

15

Figure 1: Schematic of ILK primary structure.
The three conserved functional domains of integrin-linked kinase (ILK) are represented. Four amino-
terminal ankyrin repeats (blue) mediate protein interactions that serve to localize ILK to focal adhesions
and also regulate ILK signaling. The pleckstrin homology domain is indicated in yellow. β-integrins (β1 and
β3) bind within the extreme carboxyl terminus of the catalytic domain.

As schematically depicted in Figure 1, ILK is made up of three structurally and
functionally distinct domains (Delcommenne et al., 1998; Hannigan et al., 1996; Li et al.,
1997). In addition to β1 integrin and β3 binding sites, the C-terminal domain of ILK also
contains a protein kinase catalytic site that exhibits significant homology to other
protein kinase catalytic domains. N-terminal to the kinase catalytic domain is a
pleckstrin homology (PH)-like domain that binds phosphatidylinositol triphosphate
(PIP3) and is involved in the regulation of the endogenous kinase activity of ILK
(Delcommenne et al., 1998; Li et al., 1997). The N-terminus contains four ankyrin
repeats (Dedhar, 2000), which are involved in mediating protein-protein interactions
and have been shown to interact with an adapter protein called PINCH (particularly
interesting Cys-His-rich protein) (Tu et al., 1999). PINCH links ILK to growth factor
16

receptors and phosphatidylinositol-3 kinase (PI3K) via its interaction with Nck-2, another
adapter protein (Li et al., 1999).
Since its identification, ILK has been found to be crucial in many different
biological processes such as proliferation, migration, angiogenesis, survival and EMT.
The kinase activity of ILK is rapidly and transiently induced by the engagement of
integrins to the ECM (Dedhar, 2000; Wu et al., 1998), ILK is also activated through
growth factor receptor tyrosine kinases (Delcommenne et al., 1998; Wu et al., 1998).
This stimulation appears to be PI3K-dependent, and involves binding of PIP3 to the PH-
like domain (Wu and Dedhar, 2001b). ILK is capable of autophosphorylation, as well as
phosphorylating the cytoplasmic domain of β1 integrin subunit and downstream targets
such as PKB/Akt on serine 473 and glycogen synthase

kinase 3β (GSK-3β) on serine 9
(Delcommenne et al., 1998; Hannigan et al., 1996).
Loss of ILK expression results in embryonic lethality. ILK
-/-
embryos arrest at the
peri-implantation stage at E5.5-E6.5 because they fail

to polarize their epiblast and to
cavitate. The impaired epiblast

polarization is associated with abnormal F-actin
accumulation

at sites of integrin attachments to the BM

zone (Sakai et al., 2003b). Mice
that carry conditional deletions of ILK in chondrocytes show skeletal defects

and develop
chondrodysplasia. In culture, ILK-mutant chondrocytes had defective spreading,
abnormal F-actin distribution, and impaired adhesion (Grashoff et al., 2003; Terpstra et
al., 2003). Conditional deletion of ILK in endothelial cells results in impaired vascular
17

development and embryonic lethality (Friedrich et al., 2004). Immortalized ILK
-/-

macrophages also show decreased AKT/PKB phosphorylation on Ser473 (Troussard et
al., 2003).
Accumulating evidences from many cell biological studies indicate that ILK is a
critical mediator for induction of EMT (reviewed in Oloumi et al., 2004). EMT induced by
overexpression of ILK in epithelial cells is likely mediated by inhibition of E-cadherin
transcription through the transcriptional repressor Snail-1 (Tan et al., 2001). ILK
phosphorylation of Akt activates the protein which promotes cell survival through
prevention of apoptosis and promotion of EMT. ILK has been shown to directly
phosphorylate GSK-3β which results in its inhibition, and this effect is strengthened by
the activation of Akt, which also phosphorylates and inactivates GSK-3β. This
downregulation of GSK-3β activity, in turn, stimulates Wnt/β-catenin signaling and leads
to an increase of Snail and therefore downregulation of E-cadherin (Novak et al., 1998;
Somasiri et al., 2001; Wu, 1999).

Glycogen synthase kinase-3β
GSK-3β is a serine-threonine kinase that was initially identified to phosphorylate
and inactivate glycogen synthase. Two isoforms, α and β, show a high degree of amino
acid homology (Stambolic and Woodgett, 1994). GSK-3β is involved in the control of
different cellular processes, including development, apoptosis and metabolism. GSK-3β
18

binds to and phosphorylates several proteins in the Wnt pathway including β-catenin.
Phosphorylation of β-catenin leads to ubiquitination and degradation, preventing it
from entering the nucleus. The consequence of GSK-3β phosphorylation is usually
inhibition of the substrate. Both ILK and Akt phosphorylate and inhibit the activity of
GSK-3β, which results in the stabilization of β-catenin allowing it to enter the nucleus
and interact with TCF/LEF (T-cell factor/lymphoid enhancer factor) family transcription
factors to promote specific gene expression such as cyclin D1 (D'Amico et al., 2000;
Dedhar, 2000; Delcommenne et al., 1998; Tan et al., 2001; Troussard et al., 2000;
Troussard et al., 1999). Recent findings support a model wherein GSK-3β activity also
controls Snail phosphorylation. GSK-3β inhibition induces stability and increased nuclear
levels of Snail protein (Bachelder et al., 2005; Yook et al., 2006; Zhou et al., 2004).

PKB/Akt
Akt is a serine-threonine kinase that regulates a diverse array of cellular
functions, including apoptosis, cellular proliferation, differentiation, and intermediary
metabolism (Chan et al., 1999).The Akt family consists of three isoforms: Akt1 (PKBα),
Akt2 (PKBβ) and Akt3 (PKBγ). They are expressed ubiquitously, share similar activation
mechanisms and typically initiate signal relay to secondary signaling cascades by
phosphorylating an overlapping subset of proteins. However, it is also clear that
isoform-specific functions for Akt must exist, since isoform-specific knockout mice reveal
19

distinct phenotypes, whereby loss of Akt1 in mice leads to growth defects whereas loss
of Akt2 primarily affects glucose homeostasis (Garofalo et al., 2003). The role of Akt3 is
less clear, though it appears to be predominantly expressed in brain. It has been
reported that mice lacking Akt3 have small brains (Yang et al., 2004). Akt is activated by
numerous growth factors and immune receptors through lipid products of PI3K. Akt
possesses a PH domain which binds either PIP3 or phosphatidylinositol (3,4)-
biphosphate (PIP2). ILK directly phosphorylates Akt on serine-473, which activates Akt,
leading to suppression of apoptosis (Yoganathan et al., 2002).

20

CHAPTER 2: HYPOTHESIS
Palatal fusion and the onset of mesenchymal transdifferentiation of the MEE is
correlated with the downregulation of E-cadherin, β1 integrin and integrin-linked
kinase. Integrin signaling through integrin-linked kinase and its downstream partners
GSK-3β and Akt is crucial for EMT as inhibition of β1 integrin and integrin-linked kinase
would prevent palatal fusion resulting in a cleft palate.

Specific Aims
1. To define temporal and spatial expression of E-cadherin, β1 integrin and ILK during
palatal development and to analyze the changes in expression from the initial
adherence to the final migration of the transformed MEE.
2. To examine the functional effects of specific β1 integrin and ILK silencing on the
phenotypic transformation of the medial edge epithelial cells.
3. To assess ILK signaling and to determine whether ILK activation phosphorylates GSK-
3β and Akt in cultured murine palates.

The analysis of the expression pattern of these molecules and the study of the
regulatory mechanisms involved in controlling their expression will allow for a better
understanding of the etiology of palate malformations resulting in clefting.

21

CHAPTER 3
Immunolocalization of endogenous E-cadherin in palate tissue in vivo.
Immunohistochemical analysis was used to assess the temporal and spatial
expression of E-cadherin in palatal tissue during the period when MEE cells lose their
epithelial phenotype and transform into mesenchyme. At embryonic stage E13 the
palatal shelves were vertical along the side of the tongue. E-cadherin could be detected
throughout the epithelial layer covering the palatal shelves. At the tip of the shelf, in the
epithelial cells that later give rise to the MEE, stronger staining for E-cadherin was
observed (Fig. 2B). At E14, the palatal shelves reoriented to a horizontal position above
the dorsum of the tongue. E-cadherin was present in the MEE of the elevated shelves
(Fig. 2C). The palatal shelves subsequently touched and adhered to form a two-cell layer
thick epithelial midline seam. Some MEE in the midline had converted to an E-cadherin
negative phenotype at this stage (Fig. 2D, arrow). As the fusion process progressed the
midline seam became discontinuous to allow mesenchymal confluence (Fig. 2E). The
discontinuous midline epithelial seam showed different E-cadherin expression. MEE
cells isolated in islands disconnected from the oral and nasal epithelia were all E-
cadherin negative (Fig. 2E and 1F, asterisks). MEE that remained in contact with oral and
nasal epithelia were E-cadherin positive. To prove that the epithelial cells lose E-
cadherin signal prior to their transformation into mesenchymal cells, serial sections
were stained with either E-cadherin or keratin antibodies. Fig. 2G shows the midline
22

seam with the E-cadherin expression, the arrow points to cells where E-cadherin is
already downregulated (arrow in Fig. 2G). Yet the E-cadherin negative cells were keratin
positive (arrow in Fig. 2H). The epithelial islands and epithelial triangle facing the oral
side remained E-cadherin positive (Fig. 2E-H). After the palatal fusion by E15 (Fig. 2I),
only mesenchymal cells were observed in the midline of the palate. E-cadherin
expression was restricted to oral and nasal epithelia and no E-cadherin could be
detected in the midline mesenchyme (Fig. 2J). The immunohistochemical analysis of E-
cadherin showed that the MEE clearly downregulate E-cadherin expression prior to
epithelial-mesenchymal transformation.
23

Figure 2: Immunolocalization of endogenous E-cadherin in palate tissue in vivo.
Negative control on a vertical palatal shelf at stage E13 (A). The palatal shelf is still in a vertical position
along the side of the tongue. E-cadherin is strongly expressed at the tip of the shelf (B). E14 the palatal
shelves are elevated to a horizontal position. E-cadherin can be detected in the epithelial cells, but not in
the mesenchyme (C). E14 the palatal shelves are touching. The two-cell layer thick MEE contains cells that
are already E-cadherin negative (arrow) (D). E14.5 the palatal shelves are fusing and the MEE transform
into mesenchyme (E). F shows an enlargement of E, E-cadherin is clearly downregulated in the epithelial
cells of the seam, arrow points to still E-cadherin expressing cells, asterisks to cells that lost E-cadherin
expression (F). The arrow points to the cells of the epithelial seam that lost E-cadherin expression (G).
Keratin staining showing that the midline cells are still epithelial (H). E15. After complete fusion of the
palate and mesenchymal confluence, distribution of E-cadherin is localized to the epithelium (I).
Magnification of the region where the shelves fused, no E-cadherin can be detected in the mesenchyme
(J).

24

CHAPTER 4
Immunolocalization of endogenous β1 integrin in palate tissue in vivo.
Immunohistochemistry was used to assess the temporal and spatial expression
of palatal β1 integrins during the transformation of the medial edge epithelium to a
mesenchymal phenotype between E12 and E15. At E12, when the palatal shelves have
initially extended from the maxillary processes into the oral cavity, β1 integrins were
detectable in the palatal epithelium (Fig. 3A). At E13, when the palatal shelves were
oriented vertically along the lateral sides of the tongue, β1 integrins could be detected
throughout the epithelial layer covering the palatal shelves but not in the underlying
mesenchyme (Fig. 3B). At E14, the palatal shelves reoriented to a horizontal position
above the dorsum of the tongue, and β1 integrin expression was examined in serial
sections anteriorly to posteriorly (Fig. 3C-E). At this stage, β1 integrins were present in
the two-cell layer thick epithelial midline seam (Fig. 3C). β1 integrins continued to be
strongly expressed in the MEE cells at the time of shelf contact (Fig. 3D) and persisted in
the medial edge and oronasal epithelium through palatal shelf elevation (Fig. 3E). By
E15, the midline seam had become discontinuous as a result of continued palatal fusion.
The distribution of β1 integrins became restricted to the oral triangle area and the oral
epithelia (Fig. 3F). The temporospatial distribution of β1 integrins in the MEE suggests
that it regulates MEE involution during palatal fusion.
25

Figure 3: Immunohistochemical detection of β1 integrin in the developing palate in vivo.
At E12, the palatal shelf has grown into the oral cavity. β1 integrin is detectable in the palatal epithelium
but not mesenchyme (A). At E13, the palatal shelf is still in a vertical position along the side of the tongue
and β1 integrin can be detected in the palatal and nasal epithelium (B). At E14 (C-E), the palatal shelves
reoriented to a horizontal position above the dorsum of the tongue and as palatal fusion progresses from
anterior to posterior, β1 integrin expression can be tracked on serial sections (C-E). When the palatal
shelves are touching and the MEE is two-cell layers thick β1 integrin is strongly expressed in the touching
seam (C). β1 integrin staining is intense at the tip of the palatal shelves before fusion (D). MEE cells have a
strong β1 integrin expression in palatal shelves that have assumed a horizontal position but have not
fused yet (E). At E15, after complete fusion of the palate and mesenchymal confluence, β1 integrin is
localized to the oral triangle and epithelium (F).

26

Immunolocalization of endogenous β1 integrins in cultured palate tissue.
To assess the functional requirement for β1 integrins during palate fusion, we
cultured palatal shelves from embryonic mice just prior to fusion in vivo and evaluated
their fusion in vitro at time periods that match in vivo stages of palate development (Cui
and Shuler, 2000; Shuler et al., 1991). After 12 h in organ culture, the medial edges of
opposing palatal shelves remained separate, and β1 integrin expression was confined to
the epithelium (Fig. 4A). By 24 h, the shelves had become adherent and a two-cell layer
thick epithelial seam had formed in the midline. β1 integrin expression continued to be
restricted to the epithelium (Fig. 4B). After the midline MEE seam became
discontinuous, the distribution of β1 integrins became similarly discontinuous (Fig. 4C).
By 48 h in culture, the midline seam had involuted, resulting in mesenchymal
confluence. β1 integrin was observed only in the oral epithelium and in a single
epithelial island (Fig. 4D). The expression of β1 integrins in cultured palatal tissues
replicated the expression observed in vivo. β1 integrins are initially expressed
throughout the epithelium but subsequently disappear from the midline seam
coincident with the disappearance of the MEE. These findings are consistent with the
participation of β1 integrins in the phenotypic transformation of MEE cells.
27

Figure 4: Immunolocalization of β1 integrin in palate tissues in vitro.
E13 + 12 h, the medial edges from opposing palatal shelves remain apart. β1 integrin is expressed in the
epithelium but not in the mesenchyme (A). E13 +24 h, a two-layer MEE seam is present in the midline.
The distribution of β1 integrin is restricted to the MEE and the epithelium covering the palate tissue (B).
E13 + 36 h, the midline seam becomes fragmented. β1 integrin expression loses its continuity at the
midline but persists in the oral epithelium (C). E13 + 48 h, the palate shelves have fused completely. β1
integrin is present only in an epithelial island remnant and in the oral epithelium and cannot be detected
in the mesenchyme (D).

β1 integrin siRNA reduces β1 integrin in 3T3-Swiss Cells and MEE.
To examine the role of β1 integrin during palate development, we specifically
silenced β1 integrin expression in palate cultures. A mixture of four β1 integrin siRNAs
(Dharmacon, SMARTselection) and a pooled non-targeting control siRNA were used in
this study. First, the efficacy of these constructs was confirmed in 3T3-Swiss cells.
Transfection with pooled β1 integrin siRNA markedly reduced β1 integrin expression,
relative to identical cells transfected with a non-targeting siRNA or to non-transfected
cells (Fig. 5A). We next examined the efficacy of β1 integrin silencing in palate cultures.
28

E13 mouse palatal tissues were transfected with 500 nM pooled β1 integrin or non-
silencing siRNA (Nakajima et al., 2007; Shiomi et al., 2006), harvested after 48 h, and the
MEE midline region dissected to extract the proteins (Saito et al., 2005; Yamamoto et
al., 2003). Transfection of β1 integrin siRNA greatly reduced β1 integrin expression
relative to palates transfected with the non-silencing control siRNA. In contrast,
glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was not affected in either the
experimental or control siRNA treatment groups (Fig. 5B), confirming specific β1 integrin
silencing.

Figure 5: β1 integrin expression is reduced in siRNA transfected 3T3-Swiss cells and palate tissue.
3T3-Swiss cells were transfected with either non-targeting control siRNA or β1 integrin siRNA. β1 integrin
protein expression was compared between siRNA transfected cells and non-transfected control cells. β1
integrin Western blot demonstrates reduced β1 integrin expression in β1 integrin siRNA transfected cells
(A). E13 palate cultures treated with 500 nM of either non-targeting control siRNA or β1 integrin siRNA
were evaluated after 48 h of culture. Western blotting of β1 integrin isolated from MEE cells
demonstrated reduced β1 integrin expression in cultures treated with 500 nM β1 integrin siRNA (B).
GAPDH expression levels remained unaltered.
29

β1 integrin silencing inhibits palatal fusion.
To determine whether β1 integrin silencing inhibits palatal fusion, E13 palatal
shelves were transfected with 500 nM of β1 integrin siRNA or non-silencing control
siRNA. Since palatal fusion is normally complete after 72 h of organ culture with only
limited numbers of MEE cells remaining, palate cultures were maintained for 72 h after
siRNA transfection. Palatal fusion was characterized by the persistence of midline MEE
cells from defined regions of the palate. Treatment with β1 integrin siRNA resulted in
greater numbers of midline MEE cells relative to palates transfected with the control
siRNA (Fig. 6 A-D). Differences in residual midline MEE between the anterior, middle and
posterior thirds of the palate were also observed. The posterior third exhibited the most
pronounced MEE retention with an almost complete cleft and a two-cell layer thick MEE
(Fig. 6D). In contrast, the middle third retained a partial seam (Fig. 6C) whereas the
anterior third retained epithelial islands (Fig. 6B). Immunostaining provided additional
evidence of β1 integrin silencing. After 72 h, β1 integrin was strongly expressed in the
oral epithelium of palate cultures transfected with non-silencing control siRNA (Fig. 6E).
In comparison, very low levels of β1 integrin were observed in palates treated with β1
integrin siRNA (Fig. 6F). To quantify palatal fusion, the presence or absence of persistent
midline MEE cells was assessed in serial sections from the anterior border of the
secondary palate to the end of the first molar region. MEE persistence was evaluated at
E13 + 72 h for non-transfected controls, palates transfected with non-silencing control
30

siRNA, and palates transfected with β1 integrin siRNA. 94% of the sections from either
non-transfected or non-targeting siRNA control treated palates had fused completely
with complete mesenchymal confluence and no midline MEE cells. In contrast, cultures
treated with β1 integrin siRNA retained 61% of MEE in the anterior region, 70% in the
middle, and 73% in the posterior region (Fig. 6G).

31

Figure 6: Effects of β1 integrin siRNA treatment on the disappearance of the epithelial midline seam.
E13 palate cultures were transfected with either β1 integrin siRNA or a non-targeting control siRNA and
collected after 72 h. Hematoxylin and eosin histology showed that palates transfected with control siRNA
fused completely with no MEE remaining (A). In contrast, β1 integrin siRNA-treated palates retained MEE
in the midline with differences from anterior to posterior. Sections in the anterior third of the palate
exhibited the least severe phenotype with epithelial island remnants in the midline (B), sections in the
middle third showed a partial MEE seam (C) and sections in the posterior third were found with a two-cell
layer thick MEE seam (D). Immunohistochemical analysis E13 + 72 h palate cultures confirmed that β1
integrin was strongly expressed in cultures treated with control siRNA (E) but greatly reduced in cultures
treated with β1 integrin siRNA (F). When the presence or absence of midline MEE was quantitated for all
sections from all trials (n=7), 6% of sections from palates of non-transfected control cultures and non-
targeting siRNA transfected cultures had persistent MEE, while 61% of the β1 integrin siRNA transfected
in the anterior third, 70% in the middle third and 73% in the posterior third demonstrated persistent
midline MEE (G). These results provide evidence that treatment of palatal organ cultures with β1 integrin
siRNA resulted in a persistence of the MEE in the midline position, consistent with an inhibition of palatal
fusion.

32

CHAPTER 5
Immunolocalization of endogenous ILK in palate tissue in vivo.
ILK expression was assessed in palatal tissues during the transformation of the
medial edge epithelium to a mesenchymal phenotype between E12 and E15. At E12,
when the palatal shelves initially extend from the maxillary processes into the oral
cavity, ILK was not detectable in the palatal epithelium (Fig. 7A). ILK first appeared at
E13, when the palatal shelves were oriented vertically along the lateral sides of the
tongue, and was limited to the tip and distal portion of the palatal nasal epithelium (Fig.
7B). By E14, the palatal shelves had assumed a horizontal position above the dorsum of
the tongue. At this stage, ILK was present in the MEE of the elevated shelves (Fig. 7C).
The palatal shelves subsequently contacted and adhered to one another, forming a two-
cell layer thick epithelial midline seam that strongly expressed ILK (Fig. 7D). As the
palatal fusion progressed, the midline seam became discontinuous and distribution of
ILK was restricted to the oral triangle area (Fig. 7E). By late E15, the MEE had
disappeared from the midline and the mesenchyme had become confluent. At this
point, ILK expression was restricted to the oral epithelia (Fig. 7F). The temporospatial
distribution of ILK in the MEE suggests its participation in the disappearance of the MEE
during palatal fusion.
33

Figure 7: Immunolocalization of ILK in palate tissues in vivo.
At E12, the palatal shelf extends into the oral cavity. No ILK is detectable in palatal epithelium or
mesenchyme (A). At E13, the palatal shelf is oriented vertically along the lateral borders of the tongue. ILK
is detected in the palatal epithelium (B). At E14, the palatal shelves have assumed a horizontal position
dorsal to the tongue. ILK is present in the epithelium of the approaching shelves (C) and in the two-cell
layer thick midline MEE seam (D). At E14.5, the midline seam is disrupted as palatal fusion progresses. ILK
is restricted to discontinuous MEE islands in the midline and to the oral epithelium (E). After complete
palate fusion at E15, ILK is expressed only in the oral epithelium (F).

34

ILK is expressed in palate epithelium and disappears with MEE involution in vivo and
in vitro.
We then examined ILK expression in an organ culture model of palatal fusion
(Cui and Shuler, 2000; Shuler et al., 1991). Through the first 12 h of culture, ILK was
expressed throughout the epithelium of the separate palatal shelves (Fig. 8A). By 24 h,
the shelves had become adherent and ILK was expressed in a two-cell layer thick
epithelial seam that formed in the midline (Fig. 8B). After 36 h, the midline MEE seam
became discontinuous, and midline ILK expression was restricted to isolated islands of
MEE cells (Fig. 8C). By 72 h in culture, MEE had disappeared from the midline resulting
in mesenchymal confluence. ILK was observed only in the oral epithelium (Fig. 8D).
These results recapitulate ILK expression in vivo, wherein ILK is initially expressed
throughout the oral epithelium but subsequently disappears from the midline seam
coincident with the involution of the MEE. These findings suggest that ILK might
regulate the phenotypic transformation of MEE cells.

35

Figure 8: Immunolocalization of ILK in palate tissues in vitro.
E13 + 12 h, the opposing palatal shelves are not in contact. ILK is expressed in the epithelium but not in
the mesenchyme (A). E13 +24 h, a two-layer MEE seam has formed in the palate midline. ILK is expressed
in the medial edge and palatal epithelium, but not in the mesenchyme (B). E13 + 36 h, the midline seam
becomes fragmented. ILK expression loses its continuity at the midline but persists in the oral epithelium
(C). E13 + 72 h, the palate shelves have fused completely. ILK is present in the oral epithelium and cannot
be detected in the mesenchyme.

ILK siRNA specifically reduces ILK in 3T3-Swiss cells and MEE cells.
To evaluate the role of ILK during palate development, we specifically silenced
ILK expression in palate cultures. Three different ILK siRNAs (termed ILK siRNA 1, ILK
siRNA 2, and ILK siRNA 3) and a non-silencing DY-547 labeled siRNA were developed.
These were validated in 3T3-Swiss cells. Transfection with the ILK siRNAs, but not with
the control siRNA markedly reduced ILK expression as assessed by Western analysis (Fig.
9A). We next examined the efficacy of ILK silencing in palate cultures. The palatal tissues
were collected at E13 + 48 h, and the MEE midline region was dissected to extract the
proteins (Saito et al., 2005; Yamamoto et al., 2003). The fluorophore-tagged control
36

siRNA served a as transfection control, and fluorescence was detected in the midline
and rugae of palate cultures for up to 72 h (Fig. 10A and 10B). DY-547 labeled control
siRNA cultures and non-transfected BGJb control cultures were collected after 24 h,
embedded in OCT, sectioned and then subjected to immunofluorescence. BGJb control
sections showed no fluorescent signal (Fig. 10C) whereas control siRNA transfected
culture sections showed a clear signal in the epithelial layer (Fig. 10D) confirming
successful siRNA transfection. Palate cultures were transfected with one of the three ILK
siRNAs or the non-silencing control DY-547 labeled siRNA at 500 nM (Nakajima et al.,
2007; Shiomi et al., 2006). Transfection of ILK siRNA greatly reduced ILK expression
compared to the control siRNA. In contrast, GAPDH was not affected in either the
experimental or control siRNA treatment groups (Fig. 9B), confirming specific ILK
silencing.

37

Figure 9: ILK expression is reduced by siRNA.
3T3-Swiss cells were transfected with either non-silencing control siRNA or three different ILK siRNAs (ILK
siRNA 1, ILK siRNA 2, and ILK siRNA 3). ILK Western blot demonstrates reduced ILK expression in ILK siRNA
transfected cells (A). E13 palate cultures treated with 500 nM of either non-targeting control siRNA or ILK
siRNA were evaluated after 48 h of culture by Western blotting. ILK protein expression was reduced in
cultures treated with 500 nM ILK siRNA (B).

Figure 10: Immunofluorescence of DY-547 labeled control siRNA.
Transfection of 500 nM control siRNA on cultured palates was demonstrated by immunofluorescence. The
intensity of the fluorescent signal remained the same in cultures after 24 h (A) and 72 h (B). Frozen
sections of non-transfected control palates (C) or fluorescent labeled control siRNA transfected (D) palate
cultures after 24 h. A clear fluorescent signal was detectable in the epithelial layer (D) when compared to
the controls (C).
38

ILK silencing inhibits palatal fusion.
To determine whether ILK silencing inhibits palatal fusion, E13 palatal shelves
were transfected with 500 nM of ILK siRNA 1, ILK siRNA 2, ILK siRNA 3 or control siRNA.
Since palatal fusion is substantially complete after 72 h of organ culture, palate cultures
were maintained for 72 h after siRNA transfection. Treatment with ILK siRNA resulted in
greater numbers of persistent midline MEE cells relative to palates transfected with the
control siRNA. ILK siRNA 1 treated palate cultures retained a partial seam or epithelial
islands (Fig. 11B), the ILK siRNA 2 treated palates showed a one cell layer of MEE (Fig.
11C) and ILK siRNA3 transfected palates maintained a two cell layer thick MEE (Fig. 11D).
Palates treated with the non-silencing control siRNA showed a small number of
epithelial islands in very few sections and mostly demonstrated complete fusion with an
intact oral epithelium and complete mesenchymal confluence (Fig. 11A). ILK
immunostaining confirmed the efficacy of ILK silencing. At E13 + 72 h, ILK was strongly
expressed in the oral epithelium of control siRNA treated palate cultures (Fig. 11E);
whereas cultures treated with ILK siRNA exhibited a markedly reduced ILK expression
(Fig. 11F-H).
To quantify palatal fusion, the presence or absence of persistent midline MEE
was assessed in serial sections from the anterior border of the secondary palate to the
end of the first molar region. Ten cultured palates were evaluated for each of the three
ILK siRNA constructs and the non-silencing control siRNA. 93% of the sections from
39

control siRNA treated palates had fused completely with total mesenchymal confluence
and no detectable midline MEE. In contrast, MEE persisted in 72% of the sections
treated with ILK siRNA 1, 75% of sections treated with ILK siRNA 2, and 82% of sections
treated with ILK siRNA 3 (Fig. 11I).

40

Figure 11: Effects of ILK siRNA treatment on involution of the medial edge epithelium.
The epithelium of E13 palates were transfected with either one of three different ILK siRNAs or a non-
silencing control siRNA, then cultured an additional 72 h. Hematoxylin and eosin histology showed that
palates transfected with control siRNA fused completely with no MEE remaining (A). In contrast, midline
MEE persisted in palates transfected with ILK siRNA oligonucleotides (B-D). Immunohistochemical analysis
of ILK siRNA treated E13 + 72 h palate cultures confirmed that ILK was strongly expressed in cultures
treated with control siRNA (E) but greatly reduced in cultures treated with ILK siRNA (F-H). When the
presence or absence of midline MEE was quantitated for all sections from all trials (n=10 for each siRNA),
7% of sections from palates treated with non-silencing control siRNA had persistent MEE while 72-83% of
the ILK siRNA transfected sections demonstrated persistent midline MEE (I).

41

ILK signals through GSK-3 β and Akt in cultured palates.
Epidermal growth factor (EGF) activates ILK by means of the PI3-kinase pathway
(Wennstrom and Downward, 1999). ILK, in turn, induces Akt phosphorylation

at Ser473
and GSK-3β phosphorylation at Ser9 (Persad et al., 2000). The functional effects of ILK
silencing may therefore be confirmed by evaluating EGF-induced GSK-3β and Akt
phosphorylation. E13 + 48h palate cultures transfected with ILK siRNA 1, ILK siRNA 1, ILK
siRNA 4 or control siRNA as previously described were stimulated with 50 ng/ml EGF for
30 min, a dose that induces peak Extracellular signal-regulated kinase (Erk) 1/2 activity
(Yamamoto et al., 2003). Phosphorylation of GSK-3β and Akt was assessed by Western
blot. In cultures treated with the control siRNA, GSK-3β and Akt were strongly
phosphorylated by EGF. In contrast, EGF-induced GSK-3β and Akt phosphorylation were
markedly reduced in cultures treated with ILK siRNA (Fig. 12, data only shown for ILK
siRNA 1). Total GSK-3β did not differ between palate tissues transfected with control
siRNA or ILK siRNA 1. These data confirm that ILK activates GSK-3β and Akt in the
developing palate and that this activation is attenuated by ILK silencing. To determine if
EGF-induced Erk phosphorylation is affected by ILK silencing, the blot in Figure 11 was
probed for phosphorylated Erk. EGF-induced Erk1/2 phosphorylation was confirmed in
palates transfected with the non-silencing control siRNA. However, EGF induced Erk1/2
phosphorylation was noticeably reduced in palate cultures treated with ILK siRNA. The
mechanism of this unexpected interaction is unclear.
42

Figure 12: ILK siRNA transfection reduces ILK signaling.
Cultures were transfected with control siRNA (lanes 1-6) or ILK siRNA 1 (lanes 7-12). At E13 + 48 h,
cultures were collected without EGF (lanes 1-3, 7-9) or with EGF treatment (lanes 4-6, 10-12).
Immunoblots for P-GSK-3β (Ser9), P-Akt (Ser 473) and P-Erk demonstrate increased phosphorylation in
both control and targeting siRNA transfected cultures. ILK siRNA treated samples show reduced
phosphorylation when treated with EGF demonstrating that during palate development ILK signals
through GSK-3β and Akt.
43

CHAPTER 6: DISCUSSION
Regulation of palatal fusion is a complex process controlled by several different
molecular mechanisms. The inhibition of any of these mechanisms can disrupt the
normal process of craniofacial development and result in the formation of a cleft palate.
The successful union to form the secondary palate initially depends upon epithelial
adherence between the opposing medial edge epithelia. MEE express desmosomes and
E-cadherin to initiate the fusion process (Luning et al., 1994; Mogass et al., 2000); as the
epithelial seam breaks up, the BM loses its density and is degraded (Fitchett and Hay,
1989). At the same time, the basal MEE cells lose keratin to acquire a vimentin rich
cytoskeleton, then extend filopodia into the adjacent connective tissue and move into
the ECM. In late stages only small islands of epithelial cells still surrounded by a BM and
two epithelial triangles at the oral and nasal ends remain with an epithelial phenotype
(Carette and Ferguson, 1992; Sun et al., 1998a).
It has been shown that palatal fusion requires the epithelial-mesenchymal
transdifferentiation of the MEE surfacing the palatal shelves. Recent insights into the
network of signals that regulate EMT and their role and integration during
embryological processes indicated that several mechanisms are involved in the initiation
and execution of EMT in development, and that the molecular mechanisms that
regulate EMT overlap with those that control cell adhesion, motility invasion, survival
and differentiation. A number of studies show that several extracellular activators can
44

trigger EMT, that extensive cross-talk exists between the signaling pathways that
activate and repress EMT, and that EMT-inducing signaling pathways have many
common endpoints, including downregulation of E-cadherin expression and expression
of EMT-associated genes. The precise spectrum of changes that occur during EMT is
probably context dependent and determined by the integration of extracellular signals
that the cell receives. EMT has an important role in the development of many tissues
during embryogenesis. The palatal shelf MEE cells are only one example of an epithelial
cell type that undergoes EMT (Hay, 1995). EMT has also been shown for Mullerian duct
cells (Trelstad et al., 1982), mammary gland epithelia (Miettinen et al., 1994) and
cardiac cushion cells (Potts et al., 1991). Loss of E-cadherin expression seems to be
heavily involved in EMT, and E-cadherin is therefore emerging as one of the caretakers
of the epithelial phenotype. E-cadherin has been shown to have a role in EMT both in
palatal cells and in other systems (Lochter et al., 1997; Sun et al., 1998a; Sun et al.,
1998b). As the expression of E-cadherin is specific for epithelial cells it would be
expected that the phenotypic transdifferentiation of the MEE would be associated with
a change in expression. Immunohistochemistry was used to assess the temporal and
spatial expression of E-cadherin in palatal tissue during the time period when the MEE
undergo EMT. The analysis of E-cadherin expression showed that the MEE clearly
downregulate E-cadherin expression prior to EMT. This data is substantiated by the
microarray analysis of MEE cells that were obtained before and during critical stages of
45

palate development (LaGamba et al., 2005). An increase in E-cadherin within the first 12
hr in culture was observed to allow for palatal seam adherence and a decrease in E-
cadherin expression was observed as early as 12 to 24 hr (equivalent to E14), when
palatal shelves become adherent and an epithelial seam forms in the midline (LaGamba
et al., 2005).

In addition to dissociating cell–cell adhesions, the cells that undergo EMT have to
regulate the integrin-mediated contacts with the ECM. β1 integrin family members are
important mediators linking the ECM to the cytoskeleton by means of a scaffold
containing cytoskeletal and signaling proteins assembled at their cytoplasmic face
(Giancotti and Ruoslahti, 1999; Miyamoto et al., 1995). Although outside-in signals
transmitted by ligand-activated integrin receptors have been extensively described and
are known to control cell migration, survival, and proliferation in many cell types (for
reviews see (Guo and Giancotti, 2004; Janes and Watt, 2006; Miranti and Brugge, 2002),
little is known about integrin expression and function during palate development. To
investigate the role of β1 integrins in MEE transformation, we analyzed the expression
pattern of β1 integrin in the developing palate in vivo and in vitro by
immunohistochemistry. β1 integrin expression was strong throughout the palatal
epithelium that includes oral epithelium, nasal epithelium and the midline epithelial
seam, but was absent from the underlying mesenchyme. Expression in the midline was
46

lost when the MEE disappeared. The consistent expression pattern of in vivo and in vitro
studies allowed us to further investigate the functional role of β1 integrin in vitro.
It has previously been shown that impairing β1 integrin function using
neutralizing antibodies inhibits TGF-β induced

EMT in mouse mammary epithelial cells
(Bhowmick et al., 2001). Lens epithelia also undergo EMT when they give rise to
mesenchymal cells upon suspension in collagen gels; α5β1 integrin is greatly
upregulated during this process and neutralizing antibodies to β1 integrin significantly
inhibit fibroblastic transformation, again suggesting that integrin/ECM interaction
promotes EMT (Zuk and Hay, 1994). Based on the similarity between the β1 integrin
expression patterns in vivo and in vitro, we evaluated the effect of silencing β1 integrin
using a siRNA-based approach. Silencing of target proteins by siRNA has been used for
both cell culture and organ culture model system (Davies et al., 2004; Elbashir et al.,
2001; Harborth et al., 2001; Nakajima et al., 2007; Sakai et al., 2003a; Shiomi et al.,
2006). The efficiency and specificity of β1 integrin siRNA inhibition was confirmed by
Western blot in cells. In palate organ cultures both Western blot and
immunohistochemistry were used. The reduction of β1 integrin resulted in delayed
palatal fusion in vitro with persistence of the MEE in the midline seam as shown by
hematoxylin and eosin staining. Interestingly we observed an anterior to posterior
gradient in MEE retention with the least number of MEE cells maintained in the anterior
palate and the most in the posterior. However, immunostaining for β1 integrin did not
47

demonstrate an expression difference along the anterior-posterior axis in vivo more
than likely due to the high basal level of protein expression in this tissue. Differential
gene expression along the anterior–posterior and medial–lateral axes of the developing
palate has been observed for a variety of molecules as reviewed in Hilliard et al., 2005.
Elevation, maturation and fusion of the palatal shelves follow an anterior to posterior
sequence (Dudas et al., 2004; Taya et al., 1999). In humans and rodents, closure of the
palate begins medially at the earliest sites of contact, proceeding both anterior and
posterior from it until fusion is complete along its entire length (Francis-West et al.,
2003). This sequence of palatal closure could be the result of regionally distinct signaling
pathways being activated along the anterior–posterior axis (Hilliard et al., 2005).
TGF-β3 has been established to have an essential role in palate development.
The expression of TGF-β3 has been shown to be specific for MEE cells and TGF-β3 null
mutations result in cleft palate (Brunet et al., 1995; Kaartinen et al., 1997; Kaartinen et
al., 1995; Nawshad and Hay, 2003; Proetzel et al., 1995). As an alternative downstream
TGF-β signaling effecter, PI3K has been identified in TGF-β-mediated EMT (Bakin et al.,
2000; Metzner et al., 1996). As one of the activators of PI3K, β1 integrin expression is
significantly increased before and during EMT in lens epithelial cells (Zuk and Hay, 1994).
Another study demonstrated that during tumor cell invasion, cytoskeleton
rearrangement and matrix metalloproteinase (MMP)-2 production upon α3β1 integrin
stimulation were dependent on PI3K activity (Sugiura and Berditchevski, 1999). The
48

signaling capability of β1 integrins emanates from the recruitment of cytoplasmic
protein kinases and adaptor elements such as focal adhesion kinase (Schaller, 2001) and
ILK which is stimulated by PI3K (Hannigan et al., 1996). Taken together, our results
demonstrate that EMT during palatal fusion in vitro is dependent on β1 integrin possibly
activating the downstream PI3K pathway.
Overall though, the function of β1 integrin in the developing palate is likely to be
complex. Integrins serve as both structural linkers between the ECM and the
cytoskeleton as well as cell signaling molecules that regulate cell behavior and identity
(Belkin and Stepp, 2000; Heino, 2000; Wiesner et al., 2005). The loss of β1 integrin is
expected to result in the alteration of many cell signaling cascades including those
utilizing FAK, Rho GTPases, ERK and ILK (Cohen and Guan, 2005; Grashoff et al., 2004;
Longhurst and Jennings, 1998).
Since the cytoplasmic tails of integrin molecules lack any enzymatic activity, their
signaling capability is mediated by integrin associated molecules. Currently more than
10 molecules are known to bind the cytoplasmic tail of β1 integrin, and there are 5
which interact with the intracellular domain of β1 associated α subunits (Liu et al.,
2000). The serine/threonine protein kinase ILK is one of the proteins that not only link
integrins with the actin cytoskeleton, but also functions as a binding platform for
additional cytoskeletal and signaling molecules (Brakebusch and Fassler, 2003). ILK
49

localizes at the focal adhesion allowing it not only to interact with different structural
proteins, but also to mediate many different signaling pathways.
Microarray analysis of isolated MEE cells transforming into mesenchyme not
only revealed the importance of E-cadherin downregulation and TGF-β pathway
associated molecules but also demonstrated that other molecular mechanisms such as
integrin signaling are involved (LaGamba et al., 2005). At the time point between the
loss of adherent morphology of the MEE seam and the acquisition of the mesenchymal
phenotype, a significant upregulation of ILK and Akt has been observed (LaGamba et al.,
2005). During palatal fusion in vitro, it was established that EMT was dependent on PI3K
(Kang and Svoboda, 2002). Using a highly selective PI3K inhibitor cultured palates only
partially fused and MEE persisted in the midline (Kang and Svoboda, 2002).ILK
stimulation by integrins or growth factors is PI3K-dependent, and involves binding of
PIP3 to the PH-like domain (Dedhar, 2000; Wu and Dedhar, 2001a). ILK is therefore
considered an important component of the PI3K signaling pathway. Adding a specific
PI3K inhibitor to palate cultures delayed but not completely blocked EMT of the MEE
indicating that the PI3K plays a role during palate development (Kang and Svoboda,
2002). The potential roles of ILK as a signaling molecule downstream of PI3K have not
been evaluated during EMT of the MEE.
In the present study, the temporal and spatial distribution of ILK in the palate
during the most critical stages of secondary palatal fusion (E12 to E15 of mouse
50

development) was characterized by immunohistochemistry. ILK expression was not
observed until embryonic day E13 and was restricted to the MEE cells at the tip of the
vertical shelf and then, at E14, expression was specific in the midline seam. ILK was
downregulated when the midline seam was breaking down. Immunohistochemistry was
repeated in palate organ cultures and ILK expression was also epithelial specific at all
stages examined. The functional role of ILK during palate development was therefore
further investigated in vitro. Accumulating evidence from many cell biological studies
indicates that ILK is a critical mediator for induction of EMT. Overexpression

of ILK in two
epithelial cell lines (intestinal epithelial cells

(IEC)-18 intestinal epithelial cells and scp2
mammary epithelial

cells) with a wild-type ILK cDNA vector has been observed to induce

EMT (Somasiri et al., 2001; Wu et al., 1998). To determine whether ILK is essential for
regulation of palate development and the disappearance of the MEE, ILK was silenced
using a siRNA-based approach. To confirm efficiency and specificity of ILK siRNA
inhibition, three different siRNAs were used and reduced ILK protein expression was
validated by Western blot. The reduction in ILK resulted in a delay in the process of
palatal fusion with 72% to 82% retention of MEE cells in the midline. This failure of
palatal fusion in organ culture supported the hypothesis that ILK has an important role
during palatogenesis. In order to gain further knowledge of the molecular mechanism
underlying the normal and abnormal palate fusion ILK mediated downstream signaling
was analyzed in more detail.
51

The phenotype in the ILK siRNA treated palate organ cultures was more severe
compared to β1 integrin transfected cultures. The more restrictive ILK expression
pattern may have allowed for better silencing. It is noteworthy that endogenous β1
integrin expression was very high, therefore silencing could have been less efficient
which would also explain the phenotypic differences.
Previous studies have found that ILK kinase activity is stimulated by the growth

factor EGF (Ahmed et al., 2006; Driver and Veale, 2006). We found that EGF induced ILK
expression in 3T3-Swiss cells and in palate cultures. EGF

also activated the downstream

ILK pathway with enhanced phosphorylation of GSK-3β and Akt. On the contrary,
reduction of ILK protein by siRNA inhibited Akt and GSK-3β phosphorylation in EGF
treated palate cultures, as measured by Western blot using phosphospecific antibodies.
Expression levels of GAPDH and GSK-3β were unchanged

upon ILK silencing This effect
was seen independently with three different ILK-specific siRNAs. Collectively, these data
provide strong support for

previous findings that inhibition of ILK activity by dominant-
negative

ILK or highly selective small molecule ILK inhibitors inhibit

Akt Ser-473
phosphorylation (Troussard et al., 2003)
Cross-talk between ILK and MAP kinase has been reported (Wang et al., 1998). It
was previously shown that EGF-induced inhibition of palatal fusion was dependent on
Erk1/2 (Yamamoto et al., 2003). There are conflicting reports as to whether Erk1/2 acts
downstream of ILK or independently. Studies on dendritogenesis indicate that ILK
52

signaling via GSK-3β is mediated partially through Erk (Naska et al., 2006). Similarly, a
study on hepatic fibrosis demonstrated that ILK is involved in the phosphorylation of
Erk1/2 and Akt and that selective inhibition of ILK expression by siRNA results in a
significant decrease in their phosphorylation (Zhang et al., 2006). In contrast, ILK siRNA
inhibited Akt phosphorylation but had no effect on Erk activation

in EGF-induced EMT in
human ovarian surface epithelium (Zhang et al., 2006). In this study, in EGF stimulated
palate cultures, selective inhibition of ILK expression by siRNA resulted not only in a
significant decrease of Akt and GSK-3β phosphorylation but also Erk1/2
phosphorylation.
In summary, the studies described provide evidence for a critical role of ILK
signaling during palatal fusion. The findings of this study clearly establish that ILK is
present in the MEE in vivo and in palate organ cultures during MEE disappearance.
Inhibition of ILK subsequently leads to retention of MEE in the midline and failure of
palate shelves to fuse properly. The data also demonstrate an essential role for ILK in
the EGF induced phosphorylation of the known downstream signaling proteins, Akt and
GSK-3β, as well as the phosphorylation of Erk1/2 during palate development. We
propose that ILK signaling acting via Akt, GSK-3β and ERK1/2 is required for MEE
disappearance and normal palatal fusion and that disruption contributes to the etiology
of cleft palate.

53

Summary of possible signaling cascades activated in MEE during EMT

Figure 13: Schematic overview for the roles of E-cadherin, β1 integrin and ILK during palate
development and possible activation of signaling pathways.
See text for details.
54

Here is a brief overview of the findings of this study and the possible the signaling
pathways involved (Fig. 13).
Epithelial to mesenchymal phenotype transition is a common phenomenon not
only

during embryonic development but also tumor metastasis. In addition to
dissociating cell–cell adhesions, the cells that undergo EMT have to regulate the
integrin-mediated contacts with the ECM. In most studied systems, EMT takes place by
growth factor stimulation, such as TGF-β or EGF, by ECM stimulation of integrins or their
downstream signaling pathways. Loss of E-cadherin protein and/or transcriptional
repression of its mRNA are hallmarks of EMT. Central in E-cadherin repression is the zinc
finger transcription factor Snail. The triggering of EMT by Snail is mediated by the direct
repression of E-cadherin transcription (Batlle et al., 2000; Cano et al., 2000). In situ
hybridization and immunohistochemistry in the developing palate showed that Snail
was not detected in MEE cells before adhesion but was upregulated in the midline seam
at the time of fusion (Martinez-Alvarez et al., 2004). These findings correlate with our
observations of early E-cadherin downregulation in the MEE midline seam.
Stimulation of integrins by ECM molecules results in their clustering at adhesion
sites and in activation of a number of signaling molecules including ILK. A role for ILK in
the regulation of E-cadherin expression was first identified in the IEC-18 rat intestinal
and Scp2 mouse mammary epithelial cell lines. Overexpression of ILK, but not kinase-
deficient or antisense ILK, in IEC-18, Scp2 and human renal tubular epithelial cells has
55

been shown to result in the loss of E-cadherin expression, and the acquisition of a
fibroblastic morphology (Guaita et al., 2002; Novak et al., 1998; Somasiri et al., 2001;
Tan et al., 2001; Wu et al., 1998). This role was further supported by the observation
that the inhibition of ILK activity in human cancer cell lines results in the upregulation of
E-cadherin expression. ILK activity is PI3K-dependent; activated PI3K phosphorylates
phosphatidylinositol 4,5-bisphosphate (PIP2)

to generate PIP3,

which binds to ILK.
Activated ILK directly phosphorylates its downstream targets Akt and GSK-3β.
Phosphorylation of GSK-3β results in its inhibition, and this effect is strengthened by the
activation of Akt, which also phosphorylates GSK-3β. GSK-3β, which phosphorylates β-
catenin and thus targets it for proteolytic destruction, was found to downregulate Snail
by two independent mechanisms: transcriptional repression (Bachelder et al., 2005),
and phosphorylation by GSK-3β leading to export from the nucleus and destruction via
the ubiquitin–proteasome pathway (Zhou et al., 2004). The final result of GSK-3β
downregulation is an increase in β-catenin and Snail. The increase in the nuclear amount
of β-catenin is another characteristic of EMT. In addition to its pivotal role in cadherin-
based cell adhesion, β-catenin can act as a transcriptional activator through its
interaction with TCF/LEF. Activity of β-catenin/TCF complex is essential for the
transcription of a variety of target genes including cyclin D1 (Savagner, 2001). However,
it was recently demonstrated that β-catenin

does not translocate to the nucleus in
palatal MEE (Nawshad and Hay, 2003). It has been shown that LEF1 can be activated by
56

a Smad2/4 complex instead of β-catenin in the MEE when stimulated by TGF-β3 leading
to transcriptional repression of E-cadherin (Nawshad et al., 2007). Why β-catenin
remains in the cytoplasm during palatal EMT

remains to be determined. Based on these
results the effects of ILK inhibition of GSK-3β are unclear and need to be addressed
further.
The EGF signaling pathway is expressed in many organs whose development is
dependent on epithelial–mesenchymal interactions, and EGF has a crucial role in the
regulation of cell proliferation and differentiation (Fisher and Lakshmanan, 1990; Pratt,
1987; Thesleff et al., 1995). The binding of EGF to its receptor initially triggers
recruitment of the Grb2/Sos complex to the plasma membrane leading to sequential
activation of Ras, Raf, MEK and Erk, which translocates to the nucleus. Integrins can
intersect the MAPK pathway at multiple points. Because integrins and growth factors
share many common elements in their signaling pathways, it is clear that there are
many opportunities for integrin signals to modulate growth-factor signals and vice versa.
In addition, crosstalk between growth factor receptors and integrin receptors modulate
cell adhesion and migration via PI3K. Activated MAPK can directly suppress the activity
of GSK-3β, thus upregulating Snail functions (Ding et al., 2005).
In summary, this study shows an early downregulation of E-cadherin, a strong β1
integrin and a specific ILK expression in the MEE. Downregulation of either β1 integrin
or ILK results in retention of MEE in the midline and an inhibition of palatal fusion. ILK
57

activation by EGF demonstrated an upregulation in GSK-3β and Akt phosphorylation and
this effect was proven to be direct as ILK silencing by ILK siRNA reduced phosphorylation
of GSK-3β and Akt. It has been shown that that EGF is an inducer of Erk1/2
phosphorylation in the MEE (Yamamoto et al., 2003). This study also demonstrated ILK
dependent Erk1/2 phosphorylation, as Erk1/2 phosphorylation in ILK silenced palates
was also reduced. The downstream activation/silencing of GSK-3β and Akt signaling
pathways in an ILK dependent fashion still remain to be elucidated.
TGF-β, another extensively researched pathway in palate development, has also
been demonstrated to activate the ILK signaling cascade. ILK is known to be a signaling
intermediate in TGF-β mediated EMT (Lee et al., 2004; Li et al., 2003). In this study we
decided to use EGF to stimulate ILK activity. However, TGF-β induced ILK activation
should also be affected. Examining ILK expression and signaling in the TGF-β3 knock-out
mice would give further insight not only into the ILK activation but also the downstream
TGF-β3 effecters.
However, it is clear from the results that the process of EMT

is quite complex,
involving not only structural and cytoskeletal

reorganization but also extensive genetic
changes.

58

CHAPTER 7: MATERIALS AND METHODS
Animals
Female Swiss Webster mice were mated overnight and the presence of a vaginal
plug used to determine day 0. The mice were sacrificed between embryonic days E12
and E15 by cervical dislocation and the fetuses removed from the amniotic sacs and
immediately decapitated. The heads were immediately cultured, fixed or frozen.

Palatal Shelf Organ Culture
Palatal shelves were cultured according to the methods previously described (Cui
and Shuler, 2000; Shuler et al., 1991). Briefly, palatal shelves were dissected from fetal
Swiss-Webster mice on embryonic day 13 (E13) under sterile conditions and placed in
pairs on Millipore filters (mixed cellulose esters, 0.8 µm) in the correct anterior–
posterior orientation and with the medial edges in contact. The palatal shelves were
cultured in Grobstein organ culture dishes using BGJb medium (GIBCO, Grand Island, NY)
at 37°C in 5% CO
2
. Analysis of the organ cultures was performed at three key time points
that are equivalent to critical stages in palatogenesis in vivo: (1) 24 h of organ culture
(equivalent to E14), when palatal shelves become adherent and an epithelial seam
forms in the midline; (2) 48 h of organ culture (equivalent to E15), when palatal fusion
begins and the midline epithelial seam begins to breakdown; and (3) 72 h of organ
59

culture (equivalent to E16), when palatal fusion is completed with disappearance of the
MEE and continuity of the mesenchyme. All experiments were replicated in triplicate.

Transfection of siRNA
Previous studies have demonstrated that small interfering RNA (siRNA) can
reduce the expression of specific proteins in palate organ cultures (Nakajima et al.,
2007; Shiomi et al., 2006). RNA interference was performed as previously described
(Davies et al., 2004; Sakai et al., 2003a).
For the β1 integrin siRNA, palate cohorts were transfected with Dharmacon
SMARTpool siRNA reagent (Thermo Fisher Scientific, Lafayette, CO) targeted against β1
integrin at final concentrations of 500 nM siRNA in BGJb with 0.2% Oligofectamine
(Invitrogen, Carlsbad, CA). The pooled sequences for the β1 integrin siRNA were: 5’-
GAA CGG AUU UGA UGA AUG Auu-3’, 5’- CCA CAG AAG UUU ACA UUA Auu-3’, 5’- GCA
CAG AUC CCA AGU UUC Auu-3’ and 5’- CAA GAG GGC UGA AGA UUA Cuu-3’. Control
palates were identically transfected with Dharmacon siCONTROL non-targeting siRNA
Pool (Thermo Fisher Scientific, Lafayette, CO). After 48 h, the culture medium was
changed and palatal fusion was evaluated at E13 + 72 h as previously described (Shiomi
et al., 2006).
For ILK siRNA , experimental groups were transfected with one of three siRNA
constructs targeted against ILK (Ambion, Austin, TX) at final concentrations of 500 nM
60

siRNA in BGJb with 0.2% Oligofectamine (Invitrogen, Carlsbad, CA). The sequences were
ILK 1: 5’-GGG CAA UGA UAU UGU UGU Gtt-3’ and 5’-CAC AAC AAU AUC AUU GCC Ctg-3’;
ILK 2: 5’-CCG UAU UCC AUA CAA GGACtt-3’ and 5’-GUC CUU GUA UGG AAU ACG Gtt-3’;
and ILK 3: 5’-GCA CGG AUU AAU GUG AUG Att-3’ and 5’-UCAUCACAUUAAUCCGUGCtc-
3’. DY-547-labeled non-silencing control siRNA was purchased from Dharmacon
(Lafayette, CO). After 48 h, the culture medium was changed (Shiomi et al., 2006).
Selected cultures were stimulated with 50 ng/ml recombinant human EGF (R&D
Systems, Minneapolis, MN) for 30 min (Yamamoto et al., 2003). Palatal fusion was
evaluated at E13 + 72 h (Shiomi et al., 2006).

Cell culture
3T3-Swiss (ATCC, Manassas, VA) were cultured in DMEM with 10% fetal calf
serum to 60% confluence in six-well plates prior to treatment with 2 l/well of 50 M
siRNA suspended in 4 l Oligofectamine (Invitrogen, Carlsbad, CA). Cells were harvested
after 48 h for immunoblotting.

Assessment of palatal fusion
Assessment of palatal fusion was performed as previously described (Saito et al.,
2005). Briefly, cultured palates were harvested at E13 + 72 h, fixed in 4%
paraformaldehyde in PBS, embedded in paraffin, and sectioned. Coronal serial sections
61

(6 m) between the 3rd ruga and the posterior boundary of 1st molar tooth organ were
stained with hematoxylin and eosin (Lerner Laboratories, Pittsburgh, PA). Every section
was examined for persistence of MEE cells in the midline epithelial seam.

Immunoblotting
At the end of each culture period, the middle region of each pair of palatal
epithelium was isolated by microdissection and lysed in RIPA buffer (150 mM NaCl, 1%
NP-40, 0.25% deoxycholate, 0.1% SDS, 50 mM Tris (pH 8.0), 1 mM NaVO
4
, 10 mM NaF,
and Complete protease inhibitor (Roche, Indianapolis, IN)). Lysate protein
concentrations were quantified and equalized using Micro BCA (Pierce, Rockford, IL). For
each sample, 10 μg of protein was resolved by 10% SDS-PAGE and electroblotted onto
polyvinylidene difluoride membrane (Bio-Rad, Hercules, CA). The blots were then
blocked for 1 h at room temperature with supplied 1% blocking solution (Roche), then
sequentially incubated with a 1:2000 dilution of primary antibody (overnight at 4°C) and
with horseradish peroxidase (HRP)-conjugated goat anti-rabbit IgG (30 min at room
temperature). Proteins were visualized with chemiluminescence using X-ray films.
The primary antibodies were anti-β1 integrin from R&D Systems (Minneapolis,
MN); anti-ILK from Upstate (Charlottesville, VA); anti-GSK-3β, anti-phosphoGSK-3β
(Ser9), anti-Akt, anti-phosphoAkt (Ser 473), anti-phosphoErk1/2 from Cell Signaling
62

Technologies (Beverly, MA); and anti-GAPDH from Chemicon International Inc.
(Temecula, CA).

Immunohistochemistry
Tissues were fixed in 4% paraformaldehyde-PBS at 4°C, embedded in paraffin,
and serially sectioned. Mouse monoclonal anti-E-cadherin IgG (BD Transduction
Laboratories, San Diego, CA ), goat polyclonal anti β1 integrin IgG (R&D Systems,
Minneapolis, MN) or rabbit polyclonal anti-ILK IgG (Upstate, Charlottesville, VA) was
applied overnight at room temperature and visualized by dye deposition (Histostain,
Zymed Laboratories, South San Francisco, CA) as previously described (Cui et al., 1998).
Normal serum was used as a negative control. Each of the experiments was repeated at
least five times.

63

REFERENCES
Ahmed, N., et al., 2006. Molecular pathways regulating EGF-induced epithelio-
mesenchymal transition in human ovarian surface epithelium. Am J Physiol Cell Physiol.
290, C1532-42.
Arboleda, M. J., et al., 2003. Overexpression of AKT2/protein kinase Bbeta leads to up-
regulation of beta1 integrins, increased invasion, and metastasis of human breast and
ovarian cancer cells. Cancer Res. 63, 196-206.
Bachelder, R. E., et al., 2005. Glycogen synthase kinase-3 is an endogenous inhibitor of
Snail transcription: implications for the epithelial-mesenchymal transition. J Cell Biol.
168, 29-33.
Bakin, A. V., et al., 2000. Phosphatidylinositol 3-kinase function is required for
transforming growth factor beta-mediated epithelial to mesenchymal transition and cell
migration. J Biol Chem. 275, 36803-10.
Batlle, E., et al., 2000. The transcription factor snail is a repressor of E-cadherin gene
expression in epithelial tumour cells. Nature Cell Biology. 2, 84-9.
Belkin, A. M., Stepp, M. A., 2000. Integrins as receptors for laminins. Microsc Res Tech.
51, 280-301.
Bhowmick, N. A., et al., 2001. Integrin beta 1 signaling is necessary for transforming
growth factor-beta activation of p38MAPK and epithelial plasticity. J Biol Chem. 276,
46707-13.
Brakebusch, C., Fassler, R., 2003. The integrin-actin connection, an eternal love affair.
EMBO Journal. 22, 2324-33.
Brakebusch, C., et al., 2000. Skin and hair follicle integrity is crucially dependent on beta
1 integrin expression on keratinocytes. EMBO J. 19, 3990-4003.
Brinkley, L. L., Morris-Wiman, J., 1984. The role of extracellular matrices in palatal shelf
closure. Current Topics in Developmental Biology. 19, 17-36.
Brunet, C. L., et al., 1995. Inhibition of TGF-beta 3 (but not TGF-beta 1 or TGF-beta 2)
activity prevents normal mouse embryonic palate fusion. International Journal of
Developmental Biology. 39, 345-55.
64

Cano, A., et al., 2000. The transcription factor snail controls epithelial-mesenchymal
transitions by repressing E-cadherin expression. Nature Cell Biology. 2, 76-83.
Carette, M. J., Ferguson, M. W., 1992. The fate of medial edge epithelial cells during
palatal fusion in vitro: an analysis by DiI labelling and confocal microscopy.
Development. 114, 379-88.
Cavallaro, U., Christofori, G., 2004. Cell adhesion and signalling by cadherins and Ig-
CAMs in cancer. Nature Reviews. Cancer. 4, 118-32.
Cecconi, F., et al., 1998. Apaf1 (CED-4 homolog) regulates programmed cell death in
mammalian development. Cell. 94, 727-37.
Chan, T. O., et al., 1999. AKT/PKB and other D3 phosphoinositide-regulated kinases:
kinase activation by phosphoinositide-dependent phosphorylation. Annu Rev Biochem.
68, 965-1014.
Chaudhry, A. P., Shah, R. M., 1973. Palatogenesis in hamster. II. Ultrastructural
observations on the closure of palate. J Morphol. 139, 329-50.
Christofori, G., Semb, H., 1999. The role of the cell-adhesion molecule E-cadherin as a
tumour-suppressor gene. Trends Biochem Sci. 24, 73-6.
Cohen, L. A., Guan, J. L., 2005. Mechanisms of focal adhesion kinase regulation. Curr
Cancer Drug Targets. 5, 629-43.
Cuervo, R., Covarrubias, L., 2004. Death is the major fate of medial edge epithelial cells
and the cause of basal lamina degradation during palatogenesis. Development. 131, 15-
24.
Cuervo, R., et al., 2002. Programmed cell death is required for palate shelf fusion and is
regulated by retinoic acid. Developmental Biology. 245, 145-56.
Cui, X. M., Shuler, C. F., 2000. The TGF-beta type III receptor is localized to the medial
edge epithelium during palatal fusion. International Journal of Developmental Biology.
44, 397-402.
Cui, X. M., et al., 1998. Immunohistochemical localization of TGF-beta type II receptor
and TGF-beta3 during palatogenesis in vivo and in vitro. International Journal of
Developmental Biology. 42, 817-20.
65

Curley, G. P., et al., 1999. Integrin antagonists. Cellular & Molecular Life Sciences. 56,
427-41.
D'Amico, M., et al., 2000. The integrin-linked kinase regulates the cyclin D1 gene
through glycogen synthase kinase 3beta and cAMP-responsive element-binding protein-
dependent pathways. Journal of Biological Chemistry. 275, 32649-57.
Davies, J. A., et al., 2004. Development of an siRNA-based method for repressing specific
genes in renal organ culture and its use to show that the Wt1 tumour suppressor is
required for nephron differentiation. Hum Mol Genet. 13, 235-46.
DeAngelis, V., Nalbandian, J., 1968. Ultrastructure of mouse and rat palatal processes
prior to and during secondary palate formation. Arch Oral Biol. 13, 601-8.
Dedhar, S., 2000. Cell-substrate interactions and signaling through ILK. Current Opinion
in Cell Biology. 12, 250-6.
Dedhar, S., et al., 1999. Integrin-linked kinase (ILK): a regulator of integrin and growth-
factor signalling. Trends in Cell Biology. 9, 319-23.
Delcommenne, M., et al., 1998. Phosphoinositide-3-OH kinase-dependent regulation of
glycogen synthase kinase 3 and protein kinase B/AKT by the integrin-linked kinase.
Proceedings of the National Academy of Sciences of the United States of America. 95,
11211-6.
Ding, Q., et al., 2005. Erk associates with and primes GSK-3beta for its inactivation
resulting in upregulation of beta-catenin. Mol Cell. 19, 159-70.
Doetschman, T., 1999. Interpretation of phenotype in genetically engineered mice. Lab
Anim Sci. 49, 137-43.
Driver, G. A., Veale, R. B., 2006. Modulation of integrin-linked kinase (ILK) expression in
human oesophageal squamous cell carcinoma cell lines by the EGF and TGFbeta1
growth factors. Cancer Cell Int. 6, 12.
Dudas, M., et al., 2004. Tgf-beta3-induced palatal fusion is mediated by Alk-5/Smad
pathway. Dev Biol. 266, 96-108.
Elbashir, S. M., et al., 2001. Duplexes of 21-nucleotide RNAs mediate RNA interference
in cultured mammalian cells. Nature. 411, 494-8.
66

Farbman, A. I., 1968. Electron microscope study of palate fusion in mouse embryos.
Developmental Biology. 18, 93-116.
Fassler, R., Meyer, M., 1995. Consequences of lack of beta 1 integrin gene expression in
mice. Genes Dev. 9, 1896-908.
Ferguson, M. W., 1988. Palate development: mechanisms and malformations.
Development. 103 Suppl, 41-60.
Ferguson, M. W., 1994. Craniofacial malformations: towards a molecular understanding.
Nat Genet. 6, 329-30.
Ferguson, W. J., 1984. Epithelial-mesenchymal interactions during vertebrate
palatogenesis. Current Topics in Developmental Biology. 19, 137-64.
Fisher, D. A., Lakshmanan, J., 1990. Metabolism and effects of epidermal growth factor
and related growth factors in mammals. Endocr Rev. 11, 418-42.
Fitchett, J. E., Hay, E. D., 1989. Medial edge epithelium transforms to mesenchyme after
embryonic palatal shelves fuse. Developmental Biology. 131, 455-74.
Francis-West, P. H., et al., 2003. Craniofacial development: the tissue and molecular
interactions that control development of the head. Adv Anat Embryol Cell Biol. 169, III-
VI, 1-138.
Friedrich, E. B., et al., 2004. Integrin-linked kinase regulates endothelial cell survival and
vascular development. Mol Cell Biol. 24, 8134-44.
Garofalo, R. S., et al., 2003. Severe diabetes, age-dependent loss of adipose tissue, and
mild growth deficiency in mice lacking Akt2/PKB beta. J Clin Invest. 112, 197-208.
Giancotti, F. G., Ruoslahti, E., 1999. Integrin signaling. Science. 285, 1028-32.
Grashoff, C., et al., 2003. Integrin-linked kinase regulates chondrocyte shape and
proliferation. EMBO Rep. 4, 432-8.
Grashoff, C., et al., 2004. Integrin-linked kinase: integrin's mysterious partner. Curr Opin
Cell Biol. 16, 565-71.
Greene, R. M., 1989. Signal transduction during craniofacial development. Critical
Reviews in Toxicology. 20, 137-52.
67

Greene, R. M., Pratt, R. M., 1976. Developmental aspects of secondary palate formation.
Journal of Embryology & Experimental Morphology. 36, 225-45.
Greene, R. M., Pratt, R. M., 1977. Inhibition by diazo-oxo-norleucine (DON) of rat palatal
glycoproteins synthesis and epithelial cell adhesion in vitro. Experimental Cell Research.
105, 27-37.
Griffith, C. M., Hay, E. D., 1992. Epithelial-mesenchymal transformation during palatal
fusion: carboxyfluorescein traces cells at light and electron microscopic levels.
Development. 116, 1087-99.
Guaita, S., et al., 2002. Snail induction of epithelial to mesenchymal transition in tumor
cells is accompanied by MUC1 repression and ZEB1 expression. J Biol Chem. 277, 39209-
16.
Guo, W., Giancotti, F. G., 2004. Integrin signalling during tumour progression. Nat Rev
Mol Cell Biol. 5, 816-26.
Hannigan, G. E., et al., 1997. Mapping of the gene encoding the integrin-linked kinase,
ILK, to human chromosome 11p15.5-p15.4
Protein kinase mediators of integrin signal transduction. Genomics. 42, 177-9.
Hannigan, G. E., et al., 1996. Regulation of cell adhesion and anchorage-dependent
growth by a new beta 1-integrin-linked protein kinase. Nature. 379, 91-6.
Harborth, J., et al., 2001. Identification of essential genes in cultured mammalian cells
using small interfering RNAs. J Cell Sci. 114, 4557-65.
Hay, E. D., 1995. An overview of epithelio-mesenchymal transformation. Acta
Anatomica. 154, 8-20.
Heino, J., 2000. The collagen receptor integrins have distinct ligand recognition and
signaling functions. Matrix Biol. 19, 319-23.
Hilliard, S. A., et al., 2005. Regional regulation of palatal growth and patterning along the
anterior-posterior axis in mice. J Anat. 207, 655-67.
Honarpour, N., et al., 2000. Adult Apaf-1-deficient mice exhibit male infertility. Dev Biol.
218, 248-58.
68

Huber, M. A., et al., 2005. Molecular requirements for epithelial-mesenchymal transition
during tumor progression. Curr Opin Cell Biol. 17, 548-58.
Hudson, C. D., Shapiro, B. L., 1973. A radioautographic study of deoxyribonucleic acid
synthesis in embryonic rat palatal shelf epithelium with reference to the concept of
programmed cell death. Archives of Oral Biology. 18, 77-84.
Humphries, M. J., Newham, P., 1998. The structure of cell-adhesion molecules. Trends in
Cell Biology. 8, 78-83.
Hynes, R. O., 2002. Integrins: bidirectional, allosteric signaling machines. Cell. 110, 673-
87.
Janes, S. M., Watt, F. M., 2006. New roles for integrins in squamous-cell carcinoma. Nat
Rev Cancer. 6, 175-83.
Jin, J. Z., Ding, J., 2006. Analysis of cell migration, transdifferentiation and apoptosis
during mouse secondary palate fusion. Development. 133, 3341-7.
Kaartinen, V., et al., 1997. Transforming growth factor-beta3 regulates
transdifferentiation of medial edge epithelium during palatal fusion and associated
degradation of the basement membrane. Developmental Dynamics. 209, 255-60.
Kaartinen, V., et al., 1995. Abnormal lung development and cleft palate in mice lacking
TGF-beta 3 indicates defects of epithelial-mesenchymal interaction. Nature Genetics. 11,
415-21.
Kang, P., Svoboda, K. K., 2002. PI-3 kinase activity is required for epithelial-mesenchymal
transformation during palate fusion. Dev Dyn. 225, 316-21.
Kemler, R., 1993. From cadherins to catenins: cytoplasmic protein interactions and
regulation of cell adhesion. Trends in Genetics. 9, 317-21.
LaGamba, D., et al., 2005. Microarray analysis of gene expression during epithelial-
mesenchymal transformation. Dev Dyn. 234, 132-42.
Lee, Y. I., et al., 2004. Integrin-linked kinase function is required for transforming growth
factor beta-mediated epithelial to mesenchymal transition. Biochemical & Biophysical
Research Communications. 316, 997-1001.
69

Li, F., et al., 1997. Identification and characterization of a mouse protein kinase that is
highly homologous to human integrin-linked kinase. Biochimica et Biophysica Acta.
1358, 215-20.
Li, F., et al., 1999. Integrin-linked kinase is localized to cell-matrix focal adhesions but
not cell-cell adhesion sites and the focal adhesion localization of integrin-linked kinase is
regulated by the PINCH-binding ANK repeats. Journal of Cell Science. 112, 4589-99.
Li, Y., et al., 2003. Role for integrin-linked kinase in mediating tubular epithelial to
mesenchymal transition and renal interstitial fibrogenesis.[erratum appears in J Clin
Invest. 2004 Feb;113(3):491]. Journal of Clinical Investigation. 112, 503-16.
Liddington, R. C., Ginsberg, M. H., 2002. Integrin activation takes shape. Journal of Cell
Biology. 158, 833-9.
Liu, S., et al., 2000. Integrin cytoplasmic domain-binding proteins. J Cell Sci. 113 ( Pt 20),
3563-71.
Lochter, A., et al., 1997. Matrix metalloproteinase stromelysin-1 triggers a cascade of
molecular alterations that leads to stable epithelial-to-mesenchymal conversion and a
premalignant phenotype in mammary epithelial cells. Journal of Cell Biology. 139, 1861-
72.
Longhurst, C. M., Jennings, L. K., 1998. Integrin-mediated signal transduction. Cell Mol
Life Sci. 54, 514-26.
Luning, C., et al., 1994. Expression of E-cadherin during craniofacial development.
Journal of Craniofacial Genetics & Developmental Biology. 14, 207-16.
Martinez-Alvarez, C., et al., 2004. Snail family members and cell survival in physiological
and pathological cleft palates. Dev Biol. 265, 207-18.
Martinez-Alvarez, C., et al., 2000. Medial edge epithelial cell fate during palatal fusion.
Developmental Biology. 220, 343-57.
Matlin, K. S., et al., 2003. Integrins in epithelial cell polarity: using antibodies to analyze
adhesive function and morphogenesis. Methods. 30, 235-46.
Meredith, J. E., Jr., et al., 1993. The extracellular matrix as a cell survival factor.
Molecular Biology of the Cell. 4, 953-61.
70

Metzner, B., et al., 1996. Evidence of the involvement of phosphatidylinositol 3-kinase in
the migration, actin stress fiber formation, and alpha v beta 3-integrin-mediated
adherence of human melanoma cells. J Invest Dermatol. 107, 597-602.
Miettinen, P. J., et al., 1994. TGF-beta induced transdifferentiation of mammary
epithelial cells to mesenchymal cells: involvement of type I receptors. J Cell Biol. 127,
2021-36.
Miranti, C. K., Brugge, J. S., 2002. Sensing the environment: a historical perspective on
integrin signal transduction. Nat Cell Biol. 4, E83-90.
Miyamoto, S., et al., 1995. Integrin function: molecular hierarchies of cytoskeletal and
signaling molecules. J Cell Biol. 131, 791-805.
Mogass, M., et al., 2000. Characterization of desmosomal component expression during
palatogenesis. International Journal of Developmental Biology. 44, 317-22.
Nakajima, A., et al., 2007. Functional role of transforming growth factor-beta type III
receptor during palatal fusion. Dev Dyn. 236, 791-801.
Naska, S., et al., 2006. An essential role for the integrin-linked kinase-glycogen synthase
kinase-3 beta pathway during dendrite initiation and growth. J Neurosci. 26, 13344-56.
Nawshad, A., Hay, E. D., 2003. TGFbeta3 signaling activates transcription of the LEF1
gene to induce epithelial mesenchymal transformation during mouse palate
development. J Cell Biol. 163, 1291-301.
Nawshad, A., et al., 2004. Transforming growth factor beta (TGFbeta) signalling in
palatal growth, apoptosis and epithelial mesenchymal transformation (EMT). Arch Oral
Biol. 49, 675-89.
Nawshad, A., et al., 2007. TGFbeta3 inhibits E-cadherin gene expression in palate
medial-edge epithelial cells through a Smad2-Smad4-LEF1 transcription complex. J Cell
Sci. 120, 1646-53.
Novak, A., et al., 1998. Cell adhesion and the integrin-linked kinase regulate the LEF-1
and beta-catenin signaling pathways. Proceedings of the National Academy of Sciences
of the United States of America. 95, 4374-9.
Oloumi, A., et al., 2004. Regulation of E-cadherin expression and beta-catenin/Tcf
transcriptional activity by the integrin-linked kinase. Biochimica et Biophysica Acta.
1691, 1-15.
71

Perez-Moreno, M., et al., 2003. Sticky business: orchestrating cellular signals at
adherens junctions. Cell. 112, 535-48.
Perl, A. K., et al., 1998. A causal role for E-cadherin in the transition from adenoma to
carcinoma. Nature. 392, 190-3.
Persad, S., et al., 2000. Inhibition of integrin-linked kinase (ILK) suppresses activation of
protein kinase B/Akt and induces cell cycle arrest and apoptosis of PTEN-mutant
prostate cancer cells. Proceedings of the National Academy of Sciences of the United
States of America. 97, 3207-12.
Potts, J. D., et al., 1991. Epithelial-mesenchymal transformation of embryonic cardiac
endothelial cells is inhibited by a modified antisense oligodeoxynucleotide to
transforming growth factor beta 3. Proc Natl Acad Sci U S A. 88, 1516-20.
Pratt, R. M., 1987. Role of epidermal growth factor in embryonic development. Curr Top
Dev Biol. 22, 175-93.
Pratt, R. M., Martin, G. R., 1975. Epithelial cell death and cyclic AMP increase during
palatal development. Proceedings of the National Academy of Sciences of the United
States of America. 72, 874-7.
Proetzel, G., et al., 1995. Transforming growth factor-beta 3 is required for secondary
palate fusion. Nature Genetics. 11, 409-14.
Raghavan, S., et al., 2000. Conditional ablation of beta1 integrin in skin. Severe defects
in epidermal proliferation, basement membrane formation, and hair follicle
invagination. J Cell Biol. 150, 1149-60.
Saito, T., et al., 2005. Effect of N'-nitrosonornicotine (NNN) on murine palatal fusion in
vitro. Toxicology. 207, 475-85.
Sakai, T., et al., 2003a. Fibronectin requirement in branching morphogenesis. Nature.
423, 876-81.
Sakai, T., et al., 2003b. Integrin-linked kinase (ILK) is required for polarizing the epiblast,
cell adhesion, and controlling actin accumulation. Genes Dev. 17, 926-40.
Saunders, J. W., Jr., 1966. Death in embryonic systems. Science. 154, 604-12.
Savagner, P., 2001. Leaving the neighborhood: molecular mechanisms involved during
epithelial-mesenchymal transition
72

Familial medullary thyroid carcinoma with noncysteine ret mutations: phenotype-
genotype relationship in a large series of patients. Bioessays. 23, 912-23.
Schaller, M. D., 2001. Biochemical signals and biological responses elicited by the focal
adhesion kinase. Biochim Biophys Acta. 1540, 1-21.
Schwartz, M. A., Shattil, S. J., 2000. Signaling networks linking integrins and rho family
GTPases. Trends in Biochemical Sciences. 25, 388-91.
Sharpe, P. M., Ferguson, M. W., 1988. Mesenchymal influences on epithelial
differentiation in developing systems. Journal of Cell Science - Supplement. 10, 195-230.
Shiomi, N., et al., 2006. Inhibition of SMAD2 expression prevents murine palatal fusion.
Dev Dyn. 235, 1785-93.
Shiozaki, H., et al., 1996. E-cadherin mediated adhesion system in cancer cells. Cancer.
77, 1605-13.
Shuler, C. F., 1995. Programmed cell death and cell transformation in craniofacial
development. Critical Reviews in Oral Biology & Medicine. 6, 202-17.
Shuler, C. F., et al., 1991. Molecular and morphologic changes during the epithelial-
mesenchymal transformation of palatal shelf medial edge epithelium in vitro.
International Journal of Developmental Biology. 35, 463-72.
Shuler, S. E., et al., 1992. Medial edge epithelium fate traced by cell lineage analysis
during epithelial-mesenchymal transformation in vivo. Clinical Chemistry. 38, 887-94.
Simirskii, V. N., et al., 2007. Conditional deletion of beta1-integrin from the developing
lens leads to loss of the lens epithelial phenotype. Dev Biol. 306, 658-68.
Smiley, G. R., Dixon, A. D., 1968. Fine structure of midline epithelium in the developing
palate of the mouse. Anat Rec. 161, 293-310.
Somasiri, A., et al., 2001. Overexpression of the integrin-linked kinase mesenchymally
transforms mammary epithelial cells. Journal of Cell Science. 114, 1125-36.
Stambolic, V., Woodgett, J. R., 1994. Mitogen inactivation of glycogen synthase kinase-3
beta in intact cells via serine 9 phosphorylation. Biochem J. 303 ( Pt 3), 701-4.
Stephens, L. E., et al., 1995. Deletion of beta 1 integrins in mice results in inner cell mass
failure and peri-implantation lethality. Genes Dev. 9, 1883-95.
73

Sugiura, T., Berditchevski, F., 1999. Function of alpha3beta1-tetraspanin protein
complexes in tumor cell invasion. Evidence for the role of the complexes in production
of matrix metalloproteinase 2 (MMP-2). J Cell Biol. 146, 1375-89.
Sun, D., et al., 1998a. Simultaneous loss of expression of syndecan-1 and E-cadherin in
the embryonic palate during epithelial-mesenchymal transformation. International
Journal of Developmental Biology. 42, 733-6.
Sun, D., et al., 1998b. TGFbeta3 promotes transformation of chicken palate medial edge
epithelium to mesenchyme in vitro. Development. 125, 95-105.
Takahara, S., et al., 2004. Programmed cell death is not a necessary prerequisite for
fusion of the fetal mouse palate. Int J Dev Biol. 48, 39-46.
Takeichi, M., 1995. Morphogenetic roles of classic cadherins. Curr Opin Cell Biol. 7, 619-
27.
Takenaka, K., et al., 2000. Altered expression and function of beta1 integrins in a highly
metastatic human lung adenocarcinoma cell line. Int J Oncol. 17, 1187-94.
Takigawa, T., Shiota, K., 2004. Terminal differentiation of palatal medial edge epithelial
cells in vitro is not necessarily dependent on palatal shelf contact and midline epithelial
seam formation. Int J Dev Biol. 48, 307-17.
Tan, C., et al., 2001. Inhibition of integrin linked kinase (ILK) suppresses beta-catenin-
Lef/Tcf-dependent transcription and expression of the E-cadherin repressor, snail, in
APC-/- human colon carcinoma cells. Oncogene. 20, 133-40.
Taya, Y., et al., 1999. Pathogenesis of cleft palate in TGF-beta3 knockout mice.
Development. 126, 3869-79.
Terpstra, L., et al., 2003. Reduced chondrocyte proliferation and chondrodysplasia in
mice lacking the integrin-linked kinase in chondrocytes. J Cell Biol. 162, 139-48.
Thesleff, I., et al., 1995. Regulation of organogenesis. Common molecular mechanisms
regulating the development of teeth and other organs. Int J Dev Biol. 39, 35-50.
Thiery, J. P., Sleeman, J. P., 2006. Complex networks orchestrate epithelial-mesenchymal
transitions. Nat Rev Mol Cell Biol. 7, 131-42.
74

Trelstad, R. L., et al., 1982. The epithelial-mesenchymal interface of the male rate
Mullerian duct: loss of basement membrane integrity and ductal regression. Dev Biol.
92, 27-40.
Troussard, A. A., et al., 2000. The integrin linked kinase (ILK) induces an invasive
phenotype via AP-1 transcription factor-dependent upregulation of matrix
metalloproteinase 9 (MMP-9). Oncogene. 19, 5444-52.
Troussard, A. A., et al., 2003. Conditional knock-out of integrin-linked kinase
demonstrates an essential role in protein kinase B/Akt activation. Journal of Biological
Chemistry. 278, 22374-8.
Troussard, A. A., et al., 1999. Cell-extracellular matrix interactions stimulate the AP-1
transcription factor in an integrin-linked kinase- and glycogen synthase kinase 3-
dependent manner. Mol Cell Biol. 19, 7420-7.
Tu, Y., et al., 1999. The LIM-only protein PINCH directly interacts with integrin-linked
kinase and is recruited to integrin-rich sites in spreading cells. Molecular & Cellular
Biology. 19, 2425-34.
Tyler, M. S., Koch, W. E., 1975. In vitro development of palatal tissues from embryonic
mice. I. Differentiation of the secondary palate from 12-day mouse embryos. Anat Rec.
182, 297-301.
Vasioukhin, V., Fuchs, E., 2001. Actin dynamics and cell-cell adhesion in epithelia. Curr
Opin Cell Biol. 13, 76-84.
Vaziri Sani, F., et al., 2005. Fate-mapping of the epithelial seam during palatal fusion
rules out epithelial-mesenchymal transformation. Dev Biol. 285, 490-5.
Vleminckx, K. L., et al., 1994. Enlarged cell-associated proteoglycans abolish E-cadherin
functionality in invasive tumor cells. Cancer Res. 54, 873-7.
Wang, F., et al., 1998. Reciprocal interactions between beta1-integrin and epidermal
growth factor receptor in three-dimensional basement membrane breast cultures: a
different perspective in epithelial biology. Proc Natl Acad Sci U S A. 95, 14821-6.
Wennstrom, S., Downward, J., 1999. Role of phosphoinositide 3-kinase in activation of
ras and mitogen-activated protein kinase by epidermal growth factor. Mol Cell Biol. 19,
4279-88.
Wiesner, S., et al., 2005. Integrin-actin interactions. Cell Mol Life Sci. 62, 1081-99.
75

Wu, C., 1999. Integrin-linked kinase and PINCH: partners in regulation of cell-
extracellular matrix interaction and signal transduction. J Cell Sci. 112 ( Pt 24), 4485-9.
Wu, C., 2001. ILK interactions. Journal of Cell Science. 114, 2549-50.
Wu, C., Dedhar, S., 2001a. Integrin-linked kinase (ILK) and its interactors: a new
paradigm for the coupling of extracellular matrix to actin cytoskeleton and signaling
complexes. Journal of Cell Biology. 155, 505-10.
Wu, C., Dedhar, S., 2001b. Integrin-linked kinase (ILK) and its interactors: a new
paradigm for the coupling of extracellular matrix to actin cytoskeleton and signaling
complexes
Integrin-linked protein kinase regulates fibronectin matrix assembly, E-cadherin
expression, and tumorigenicity. Journal of Cell Biology. 155, 505-10.
Wu, C., et al., 1998. Integrin-linked protein kinase regulates fibronectin matrix assembly,
E-cadherin expression, and tumorigenicity. Journal of Biological Chemistry. 273, 528-36.
Yamamoto, T., et al., 2003. Role of ERK1/2 signaling during EGF-induced inhibition of
palatal fusion. Developmental Biology. 260, 512-21.
Yang, Z. Z., et al., 2004. Physiological functions of protein kinase B/Akt. Biochem Soc
Trans. 32, 350-4.
Yoganathan, N., et al., 2002. Integrin-linked kinase, a promising cancer therapeutic
target: biochemical and biological properties. Pharmacology & Therapeutics. 93, 233-42.
Yook, J. I., et al., 2006. A Wnt-Axin2-GSK3beta cascade regulates Snail1 activity in breast
cancer cells. Nat Cell Biol. 8, 1398-406.
Zhang, Y., et al., 2006. Involvement of integrin-linked kinase in carbon tetrachloride-
induced hepatic fibrosis in rats. Hepatology. 44, 612-22.
Zhou, B. P., et al., 2004. Dual regulation of Snail by GSK-3beta-mediated
phosphorylation in control of epithelial-mesenchymal transition. Nat Cell Biol. 6, 931-40.
Zimmerman, E. F., Wee, E. L., 1984. Role of neurotransmitters in palate development.
Current Topics in Developmental Biology. 19, 37-63.
76

Zuk, A., Hay, E. D., 1994. Expression of beta 1 integrins changes during transformation of
avian lens epithelium to mesenchyme in collagen gels. Developmental Dynamics. 201,
378-93.

Abstract (if available)

Abstract The fusion of the secondary palate is a complex event requiring epithelial and mesenchymal cell differentiation. Palatal fusion and mesenchymal confluence entail adhesion of the opposing medial edge epithelia (MEE), extracellular matrix (ECM) remodeling and MEE disappearance. In this process the MEE are believed to undergo epithelial-mesenchymal transformation (EMT). While E-cadherin mediates cell-cell adhesion, integrins are the major cell surface receptors. Integrins transduce ECM signals into the cell by associating with adaptor proteins such as integrin-linked kinase (ILK). Integrin and growth factor receptor signaling has been implicated in the regulation of EMT. We hypothesized that integrin signaling is necessary for palatal fusion. To assess this, we examined E-cadherin, beta1 integrin and ILK expression. Immunohistochemistry demonstrated that E-cadherin expression was downregulated in the MEE prior to EMT. During palatal fusion, beta1 integrin strongly immunolocalized to the MEE and oral epithelia while ILK expression was restricted to the MEE. Silencing of either beta1 integrin or ILK by siRNA transfection into palate organ culture resulted in the persistence of MEE cells in the midline. Western blot confirmed the protein reduction. Finally, epidermal growth factor-induced ILK signaling was assessed in palate cultures treated with either ILK siRNA or non-silencing control siRNA.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

The function of TGF-beta signaling in palatal and dental epithelium during embryogenesis

PDF

Non-canonical Tgf-beta signaling in craniofacial development

PDF

Genetic review of mouse and human cleft lip and palate

PDF

Leucine-rich amelogenin peptide induces osteogenesis in mouse embryonic stem cells

PDF

Interaction of epigenetics and SMAD signaling in stem cells and diseases

PDF

Alk5 mediated TGF-β signaling acts upstream of FGF10 to regulate the proliferation and maintenance of dental epithelial stem cells

PDF

The role of Wnt signaling in organogenesis: limb and lung

PDF

Transcellular calcium transport in amelogenesis

PDF

TGF-β signaling regulates gingival epithelial wound healing and barrier function

PDF

Biomaterial (Col-Tgel)-induced multinucleated cells formation, polarization, and functions

PDF

Zebrafish as a blueprint for cardiac regeneration

PDF

A Western blot analysis and immunohistochemical study of 4-day old mice teeth over-expressing amelotin during amelogenesis

PDF

Tight junction protein CLDN18.1 attenuates malignant properties and related signaling pathways of human lung adenocarcinoma in vivo and in vitro

Asset Metadata

Creator Schmid, Daniela (author)

Core Title Integrin expression and signaling during palatal fusion

School School of Dentistry

Degree Doctor of Philosophy

Degree Program Craniofacial Biology

Publication Date 07/29/2008

Defense Date 06/10/2008

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

Tag beta1 integrin,integrin-linked kinase,OAI-PMH Harvest,palate development

Language English

Advisor Shuler, Charles (committee chair), Chai, Yang (committee member), Frenkel, Baruch (committee member), Lee, Matthew (committee member), Paine, Michael (committee member)

Creator Email dschmid@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m1438

Unique identifier UC1146876

Identifier etd-Schmid-20080729 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-90105 (legacy record id),usctheses-m1438 (legacy record id)

Legacy Identifier etd-Schmid-20080729.pdf

Dmrecord 90105

Document Type Dissertation

Rights Schmid, Daniela

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