University of Southern California Dissertations and Theses

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

(USC Thesis Other)

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

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content LEUCINE-RICH AMELOGENIN PEPTIDE INDUCES OSTEOGENESIS
IN MOUSE EMBRYONIC STEM CELLS

by

Rungnapa Warotayanont

__________________________________________________________________________

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)

December 2008

Copyright 2008 RungnapaWarotayanont

ii
DEDICATION

This dissertation is dedicated to my beloved parents, Wirachai Warotayanont and Ruengnettra
Warotayanont, who have given their very best to me to achieve what I have today. I am forever
thankful for your love and support. Thank you and I love you, Papa and Mama.

iii
ACKNOWLEDGEMENTS

I would like to express my gratitude and appreciation to my graduate advisor and Dissertation
Chair, Prof. Malcolm Snead, who has persevered with me throughout the time it takes to
complete my research and to write this dissertation. His critical advice and invaluable support has
shaped and inspired me to become a better scientist not only during the time of my graduate
training, but also in the many years to come in my academic career.

My Graduate Committee members, Dr Yan Zhou, Dr Michael Paine, Dr Yang Chai, and
Dr Baruch Frenkel, have generously given their time, expertise, and advice to help me
accomplish this research and dissertation. I would like to express my gratitude to them for their
contribution and their support.

My thanks also go to Dr Songtao Shi, and his postdoctoral research fellow, Dr Takayoshi
Yamaza, who have given me the opportunity to work with human mesenchymal stem cells. I am
grateful for their advice and support, as well as their trust and belief in my ability.

My time as a Ph.D. student at the University of Southern California has been one of the most
important and formative experiences in my life. And this experience would never have happened
without the existence of many friends, colleagues, faculties, students, and administrative staffs
both at the Center for Craniofacial Molecular Biology, and at the USC School of Dentistry, who
have assisted, advised, and supported my research over the years.

iv
TABLE OF CONTENTS

Dedication ii
Acknowledgements iii
List of Tables vii
List of Figures viii
Abstract x
Introduction 1
1. The structure and function of amelogenins 1
2. Leucine-rich amelogenin peptide 4
3. Regulation of osteogenesis 5
4. Embryonic stem cells and bone marrow mesenchymal stem cells 9
Chapter 1: The function of leucine-rich amelogenin peptide 12
in osteogenic induced mouse embryonic stem cells
Materials and Methods 14
1. Mouse embryonic stem cell culture 14
2. Generation of mouse amelogenin-null ES cells 14
3. Induction of cell differentiation 17
4. RNA extraction, cDNA synthesis, 19
and quantitative real-time PCR analysis
5. Analysis of mineral deposition 19
6. Statistical analysis 20
Results 21
1. Different amelogenin isoforms are expressed during 21
osteogenic differentiation of ES cells
2. LRAP enhanced mineral deposition in 24
osteogenic-induced mouse ES cells
3. LRAP induces the expression of bone marker genes 28
in osteogenic-induced ES cells
Discussion 32
Chapter 2: The function of leucine-rich amelogenin peptide during 36
osteogenic induction and adipogenic induction of
human bone marrow mesenchymal stem cells
Materials and Methods 40
1. Human bone marrow mesenchymal stem cell culture 40
2. Induction of cell differentiation 40
3. Bone marrow mesenchymal stem cell transplantation 41
4. Western immunoblot analysis 42

v
5. Analysis of mineral deposition and fat deposition 43
6. Cell proliferation assay 43
7. Fluorescence-activated cell sorting analysis 44

Results 46
1. LRAP increases Runx2 and OCN expression 46
in human BMMSCs and OF-MSCs
2. LRAP increases bone formation and hematopoiesis in vivo 50
3. LRAP inhibits adipogenesis of human BMMSCs 54
4. The effect of LRAP on cell proliferation 56
5. The effect of LRAP on cell surface antigens of human BMMSCs 57
Discussion 61
1. LRAP induces bone-tissue formation along with 62
the formation of bone-marrow like structure
2. The role of LRAP on maintenance of homeostasis 64
3. The role of LRAP in adipogenesis and the possible linkage 67
to Wnt signaling pathway
Chapter 3: Analysis of signaling pathways responsible for the effects of 68
leucine-rich amelogenin peptide
Materials and Methods 77
1. Cell culture and differentiation 77
2. Western immunoblot analysis 78
3. Detection of Wnt reporter activity 79
4. RNA extraction, cDNA synthesis, 80
and quantitative real-time PCR analysis
5. PCR Array analysis 80
6. Statistical analysis 81
Results 82
1. LRAP enhanced -catenin protein accumulation and 82
Wnt reporter activity during osteogenic differentiation
of mouse ES cells
2. The expression of Wnt7b and Wnt10b is upregulated 87
in LRAP-treated mouse ES cell
3. Genes associated with Wnt signaling pathway 90
are upregulated by LRAP
4. LRAP rescued the decreased Wnt reporter activity 99
in the presence of Wnt antagonist
5. LRAP moderately activates BMP signaling pathway 103
at the later stage of osteogenic differentiation
6. Increased LRAP concentration does not increase mineral formation 107
7. The effect of exogenous canonical Wnt and scrambled peptides 110
on osteogenic induced ES cells

vi
Discussion 115
1. LRAP activates canonical Wnt signaling pathway 116
to induce osteoblastogenesis
2. LRAP changes the expression of genes associated with 118
canonical and noncanonical Wnt signaling pathway
3. Exogenous Wnt3a fails to mimic the effect of LRAP 123
on osteogenic differentiation of ES cells
4. The implication of BMP signaling pathway 124
5. The specificity of LRAP peptide sequence, peptide concentration, 125
and culture time

Conclusions 127
References 131

vii
LIST OF TABLES

Table 1: Real-time PCR primer sequences and their annealing temperatures 22
Table 2. Calcium concentration analysis for embryoid body (EB) culture at day-20 27
Table 3. Genes associated with mouse Wnt signaling pathway 70
for polymerase chain reaction (PCR) array analysis

viii
LIST OF FIGURES

Figure 1: A schematic diagram showing the mouse X-chromosomal amelogenin 3
gene containing 9 exons.
Figure 2: Regulation of osteoblast specification and differentiation 6
Figure 3: Generation of amelogenin deficient murine embryonic stem cell lines 15
Figure 4: A schematic diagram showing stages of differentiation 18
of mouse embryonic stem (ES) cells
Figure 5: Identification of amelogenin splicing isoforms during osteogenic 23
differentiation of mouse embryonic stem cells
Figure 6: ES cell differentiation and mineral nodule formation 25
Figure 7: LRAP induced bone sialoprotein (BSP) and osterix (Osx) expression 28
in embryoid bodies
Figure 8: LRAP increases Runt-related transcription factor 2 (Runx2) 47
and osteocalcin (OCN) protein expression in human bone marrow
mesenchymal stem cells (hBMMSCs) and human mesenchymal stem cells
from the orofacial origin (hOF-MSCs).
Figure 9: LRAP increases bone formation and bone marrow formation in vivo 50
Figure 10: LRAP inhibits adipogenesis of human bone marrow mesenchymal 55
stem cells (BMMSCs)
Figure 11: The effect of LRAP on cell proliferation 56
Figure 12: Characterization of surface antigens of BMMSCs following LRAP treatment 58
Figure 13: LRAP increases -catenin protein expression 83
in mouse embryonic stem (ES) cells

ix
Figure 14: LRAP increases Wnt reporter activity 86
Figure 15: The expression of Wnt7b and Wnt10b is upregulated 88
in LRAP-treated mouse ES cell
Figure 16: Changes in Wnt signal-associated gene expression 91
in mouse embryoid bodies (EBs) treated with LRAP for 1 hour
Figure 17: Changes in Wnt signal-associated gene expression 95
in mouse embryoid bodies (EBs) treated with LRAP for 4 hours
Figure 18: Quantitative real-time PCR analysis for mouse embryoid bodies (EBs) 98
treated with LRAP for 4 hours.
Figure 19: LRAP rescued the diminished Wnt reporter activity 101
by the addition of secreted Wnt antagonist
Figure 20: The effect of LRAP on BMP signaling pathway 105
Figure 21: The effect of LRAP concentration on mineral formation 108
in MC3T3 cells
Figure 22: The effect of exogenous Wnt and scrambled LRAP peptide 112
on osteogenic-induced mouse ES cells
x
ABSTRACT

The extracellular matrix proteins in the developing enamel are constituted primarily by a group of
highly conserved proteins called amelogenins. A number of amelogenin isoforms are expressed
at distinctive stages during enamel development, but the function(s) for each isoform remain
unclear. Emdogain, a mixture of porcine amelogenins has been used in clinical dentistry to
promote cementogenesis and osteogenesis. A protein constituent in Emdogain called leucine-rich
amelogenin peptide (LRAP) has been shown to possess a signaling property shown to induce
osteogenic differentiation. However, the pathway(s) involved in the LRAP-mediated osteogenic
effect is still unclear, and the function(s) exerted by LRAP to determine stem cell differentiation
remains unknown. The purpose of the current study is to explore the effect of LRAP on
osteogenesis in different cell types, including mouse embryonic stem (ES) cells and human bone
marrow mesenchymal stem cells (hBMMSCs). In addition, the signaling pathway(s) responsible
for the osteogenic effect of LRAP was examined. The LRAP-induced phenotype was determined
using osteogenic-induced mouse ES cells, and by using osteogenic-induced hBMMSCs. The
effect of LRAP on hBMMSCs in vivo was assessed by implantation of hBMMSCs, LRAP and
hydroxyapatite/calcium phosphate carrier, into immunocompromised mice. To explore the
molecular pathway(s) responsible for the inductive effects of LRAP, I measured -catenin and
pSmad1/5/8 protein level in LRAP-treated ES cells by Western immunoblotting. The BMP-
Smad activity and Wnt promoter activity induced by LRAP was detected using a luciferase
reporter containing BMP-Smad binding elements, or a reporter containing the TCF/LEF binding
elements, respectively. The results suggest that ES cells, in response to endogenous an
exogenous LRAP, showed an increase in bone marker gene expression, with increased calcium
deposition in the mineralized matrix. Human BMMSCs, after exposed to LRAP, showed
enhanced osteoblastogenesis both in vitro and in vivo. On the other hand, adipogenesis was
xi
observed to be inhibited in LRAP-treated hBMMSCs primed for adipogenesis. The -catenin
protein level and the Wnt-reporter activity was upregulated within 24 hours after LRAP
treatment. The pSmad1/5/8 protein level and BMP-Smad-reporter activity was moderately
increased at 4 days after LRAP treatment. These results suggest that the function of LRAP is to
induce osteogenesis in stem cells via the activation of Wnt signaling pathway.

1
INTRODUCTION

1. The structure and function of amelogenins

Amelogenins are a group of highly conserved proteins secreted by ameloblasts during the
development of enamel (Simmer and Fincham, 1995). Generally, an amelogenin gene is located
on the X chromosome but in some species, including human, bovine, and porcine, there is an
amelogenin gene on the Y chromosome (Gibson et al., 1991; Veis, 2003). In other vertebrates,
such as mice, the amelogenin gene is restricted to only the X chromosome. The function of full-
length amelogenin is to control mineral formation in enamel and this function has been
extensively demonstrated by many in vitro and in vivo studies. Mutations of the amelogenin
gene in humans display X-linked amelogenesis imperfecta, the developmental defect affecting the
formation of the enamel (Aldred et al., 1992; Lagerstrom et al., 1991; Lench et al., 1994).
Disruption of the amelogenin gene in mice shows signs of enamel hypoplasia similar to
amelogenesis imperfecta phenotype (Gibson et al., 2001). In agreement with the results from
human and mouse models, antisense and ribozyme-mediated inhibition of amelogenin expression
leads to enamel mineral defects (Diekwisch et al., 1993; Lyngstadaas et al., 1995). These results
support the requirement for amelogenin protein during the formation of enamel.

Apart from providing structural organization of enamel, a different aspect of amelogenin function
has been recently identified in relation to cell signaling. Enamel matrix derivative (EMD), a
mixture of enamel matrix proteins, consisting of 90% amelogenins and 10% other non-
amelogenins enamel matrix proteins, has been used in clinical dentistry to promote periodontal
regeneration (Jepsen et al., 2004; Sculean et al., 2004). EMD as a commercial product known as

2
Emdogain, a mixture that has the ability to induce tissue repair by promoting periodontal tissue
regeneration by enhancing proliferation and differentiation of osteoblastic precursor cells (Jiang
et al., 2006), along with the upregulation of bone marker genes (Carinci et al., 2006;
Hammarstrom, 1997; Veis et al., 2000).

The signaling property of amelogenin protein isoforms was demonstrated by Nebgen and
colleagues, who showed that small amelogenin proteins detected in rat and bovine dentin matrix
were able to induce chondrogenesis and osteogenesis from muscle fibroblasts (Nebgen et al.,
1999), indicating that amelogenin functions as a signaling molecule to change gene expression
and cell differentiation. These small amelogenin proteins were later identified as the product of
amelogenin exon 2, 3, 5, 6D and 7, with a translated protein mass of 6.9 kDa (leucine-rich
amelogenin peptide, LRAP, or [A-4]), and the product of amelogenin exon 2, 3, 4, 5, 6D and 7,
with a translated protein mass of 8.1 kDa ([A+4]) (Fig 1) (Veis, 2003). Although closely related
in size, the two proteins were found to have distinguishable signaling effects on immature
mesenchymal cells (Veis, 2003; Veis et al., 2000). In embryonic muscle fibroblast culture, [A+4]
increased the chondrogenic marker, Sox-9, whereas LRAP also known as[A-4] increased the
early bone marker, Runx2 (Veis et al., 2000). In bead implantation experiment in the pulp of rat
molars, [A+4] beads induced formation of dentin bridge in 21 days, whereas LRAP/[A-4] beads
induced mineralization of the pulp tissue (Goldberg et al., 2003). Further evidence supports the
notion that amelogenin protein isoforms possess signaling property, although the mechanisms of
how these signals are initiated are still unclear (Boabaid et al., 2004; Veis et al., 2000).

3
Fig 1.

A schematic diagram showing the mouse X-chromosomal amelogenin gene containing 9 exons.
Pre-mRNAs transcribed from all 9 exons undergo extensive alternative splicing, giving rise to 11
mRNA and protein isoforms identified to date. Illustrated here are mRNA products of exons 2,
3, 5, 6D and 7 encoding the leucine-rich amelogenin peptide (LRAP, also known as [A-4]), and
the mRNA product of exons 2, 3, 4, 5, 6D and 7 encoding a peptide known as [A+4] for its
retention of exon 4.

4
2. Leucine-rich Amelogenin Peptide (LRAP)

Emdogain is composed in its majority by the alternatively spliced full-length amelogenin,
amelogenin splice products and LRAP (Fincham et al., 1983), suggesting that LRAP in Emdogain
may have the potential to induce osteogenesis. In addition, failure of LRAP to rescue the
amelogenin-null enamel phenotype indicates that LRAP may have distinctive function(s) outside
guiding mineral crystallite formation, which is the function of the full-length amelogenins (Chen
et al., 2003). In fact, Veis and colleagues have found that LRAP was able to induce osteogenesis
in rat muscle fibroblasts (Veis et al., 2000). Subsequent studies on LRAP have suggested that
LRAP was also able induce osteogenesis in mouse cementoblasts (Boabaid et al., 2004) and
mouse oral mucosal cells (Lacerda-Pinheiro et al., 2006a). For example, embryonic muscle
fibroblasts induced by LRAP increased expression of Runx2 (Veis et al., 2000). On the other
hand, for murine cementoblasts, LRAP induction increased the expression of OPN and OPG but
had no effect on Runx2 expression (Boabaid et al., 2004).

Despite the evidence that supports the osteoinductive property of LRAP, studies to identify the
mechanism of action of LRAP-induced bone regeneration have been limited. LAMP-1, a 95 kDa
lysosomal associate membrane protein receptor for receptor-mediated endocytosis, has been
identified as a cell surface receptor for LRAP (Tompkins et al., 2006). However, it remains
unclear how LAMP-1 relays the signal from LRAP after the internalization of the protein since
the lysosomal pathway generally leads to protein degradation. One study using an experimental
strategy of inhibition assay for signaling pathways has shown that the addition of MAPK
inhibitor inhibited osteopontin mRNA expression in LRAP-induced mouse cementoblasts
(Boabaid et al., 2004), suggesting MAPK may provide some of the inductive signal.

5
Using the yeast two-hybrid system, Wang and colleagues have identified three intracellular
proteins bound to LRAP, including Eef2, Fez1 and Lsm (Wang et al., 2006). However, despite
progress from these studies, it is still unclear whether LRAP-mediated osteogenic induction
operates by binding to a cell surface receptor, binding to an intracellular protein(s), or activating
another signaling cascade, as suggested by the physical interaction of LRAP to Eef2, Fez1 and
Lsm1 identified in the yeast two hybrid assay (Wang et al., 2006).

3. Regulation of osteogenesis

Osteoblasts are derived from mesenchymal stem cells, the multipotent cells that give rise to a
variety of cellular components including osteoblasts, chondrocytes, adipocytes, myoblasts, and
tendon cells while maintaining undifferentiated reservoirs of cells for their maintenance and
regeneration (Aubin, 2001). Complex networks of signaling pathways, signaling molecules,
growth factors, and transcription factors are coordinated to induce the osteoblastic phenotype
(Fig 2). Two transcription factors, Runx2 and Osx, are required for osteoblast differentiation
(Schroeder et al., 2005). Deletion of one Runx2 allele leads to cleidocranial dysplasia (Otto et al.,
2002), and Runx2 knockout (KO) mice lack both intramembranous and endochondral bone
formation due to a profound defect in osteoblast differentiation (Komori et al., 1997). Similar to
the phenotypes observed in Runx2 KO mice, the knockout of Osx shows defects in bone
formation, but Runx2 expression was maintained suggesting that Osx acts downstream of Runx2
(Nakashima et al., 2002).

6
Fig 2.

Regulation of osteoblast specification and differentiation. Signals from secreted proteins in the
Wnt and BMP signaling pathways regulate the two transcription factors, Runx2 and Osx, required
for osteoblast differentiation and maturation. Runx2 plays a role during osteoblast proliferation
and specification resulting in increased expression of collagen type I, alkaline phosphatase, and
osteonectin. Osx plays a role during osteoblast maturation and mineralization resulting in an
increase in the expression of osteopontin, bone sialoprotein and osteocalcin. The signal from Wnt
and BMP signaling pathway coordinately activate the transcription of Msx-2, which in turn acts
as a transcriptional repressor to inhibit the expression of -catenin and Runx2.

7
Signals from secreted proteins in the BMP and Wnt signaling pathways are known to play
important roles in the regulation of osteoblast differentiation. The BMP family of proteins has
been identified as a potent stimulator for osteogenesis (Chen et al., 2004a; Rosen et al., 1996).
BMP-2 has been found to upregulate the expression of a group of homeodomain (HD) proteins,
including the Msx and Dlx families of transcription factors (Dodig et al., 1999; Hassan et al.,
2004; Hassan et al., 2006; Lee et al., 2003a; Lee et al., 2003b; Miyama et al., 1999). Msx-2 has
been reported to act as a transcriptional repressor during osteoblast differentiation (Ichida et al.,
2004; Yoshizawa et al., 2004), whereas Dlx-3 and Dlx-5 are positive regulators during the
process of osteoblast differentiation (Hassan et al., 2004; Lee et al., 2005). Lee and colleagues
found that antisense inhibition of Dlx-5 results in complete abrogation of Runx2 and Osx
expression, and overexpression of Dlx-5 was sufficient to induce Runx2 and Osx expression, in
the absence of BMP-2 (Lee et al., 2003b).

The canonical Wnt signaling pathway has been implicated in a large array of developmental
events during embryogenesis (Wodarz and Nusse, 1998). -catenin, an essential mediator in this
pathway, is detected in the nucleus when stimulated by Wnt and binds to Tcf/LEF family of
transcription factors to affect target gene expression (Westendorf et al., 2004). Inactivation of -
catenin in osteoprogenitor cells inhibits osteoblast differentiation and shifts the preference of cell
differentiation from osteoblast to chondrocyte in calvarial and long bones (Day et al., 2005; Hill
et al., 2005; Hu et al., 2005). In the absence of -catenin, the development of osteoprogenitor
cells is arrested, and only type I collagen and alkaline phosphatase were expressed, suggesting
that -catenin is required for an early step of osteoblast differentiation (Hu et al., 2005).
Activation of the -catenin/Tcf-1 complex also results in increase of Runx2 expression (Gaur et
al., 2005).

8
Interestingly, although -catenin stabilization in the mesenchymal cells result in suppression of
chondrogenesis, it does not increase osteoblast differentiation (Hill et al., 2005). This
observation may suggest the presence of a negative feedback mechanism to regulate the
amplitude of Wnt signals.

The co-receptor of the canonical Wnt signaling pathway, LRP5/6 has been shown to be involved
in the regulation of bone mass and osteoblast proliferation (Johnson et al., 2004; Kato et al.,
2002). Inactivation of LRP5 results in osteopenia due to impaired osteoblast development
(Fujino et al., 2003; Kato et al., 2002). In addition, a mutation to the extracellular domain of the
LRP5 protein at the G171 residue, converting it to valine within a domain that facilitates the
interaction between LRP5/6 and the Wnt antagonist DKK, causes the formation of irregular high
bone mass and osteoporosis pseudoglioma (Fujino et al., 2003). A similar high bone mass
phenotype has also been observed in transgenic mice expressing mutated LRP5 (Fujino et al.,
2003). Constitutively active LRP-5 in MC3T3-E1 cells facilitates osteogenic differentiation, an
outcome evidenced by increasing BSP, OCN, and alkaline phosphatase expression level (Guo and
Cooper, 2007).

The canonical Wnt signaling pathway can be regulated at both the extracellular and intracellular
levels. Intracellularly, the GSK3-Axin-APC complex controls the level of -catenin by regulating
phosphorylation and subsequent degradation of -catenin (Westendorf et al., 2004).
Extracellularly, Wnt receptors may be bound by a secreted protein like Dickkopf1 (Dkk1), which
acts as a Wnt antagonist by binding to LRP5/6 and interfering with the relay of Wnt signal via its
receptor (Glinka et al., 1998). In mutant mice with inactive Dkk-1 mutation, a high bone mass
phenotype is observed (MacDonald et al., 2007).

9
Interestingly, Dkk1 is also a direct transcriptional target for canonical Wnt signaling pathway
(Chamorro et al., 2005). Alternative to the binding of Wnt at LRP5/6 receptor, frizzled related
proteins (FRPs), a protein whose structure resemble that of authentic frizzled proteins, but lacking
the transmembrane segments, can bind to Wnts. In the this capacity, FRPs can act as a decoy
receptor to interfere with Wnt signal activity (Kawano and Kypta, 2003). In addition, Wif-1
(Hsieh et al., 1999), Sclerostin and WISE (Ellies et al., 2006), and Chibby have also been
reported to act as antagonists to Wnt signaling pathway.

4. Embryonic stem cells and bone marrow mesenchymal stem cells

Although the osteogenic potential of LRAP has been identified to act on several cell types
(Boabaid et al., 2004; Lacerda-Pinheiro et al., 2006b; Veis et al., 2000), so far, none of the studies
have been performed on stem cells. Unlike the more commonly used lineage-committed cells
whose behavior has undergone cellular and molecular modifications, ES cells are naïve and
pluripotent. Therefore, the results from LRAP induced stem cell studies are predicted to more
directly denote molecular mechanisms involving the signaling function of LRAP. From the
concept of tissue engineering and bone repair therapy, the ability of ES cells to be maintained at
the pluripotent state indefinitely in vitro suggests that the ES cells can be particularly useful in
generating a large population of reserve osteoblastic cells without compromising their
differentiation potential. However, concerns regarding the use of ES cells in clinical application
remain. These concerns include the removal of animal-derived components used during the
maintenance of ES cells in vitro, the potential for the ES cells to form a teratoma upon injection
into the body (Gertow et al., 2004), and the controversial ethical issues (Scott and Reijo Pera,
2008).

10
Alternative sources of stem cells include adult sources that can be obtained primarily from bone
marrow stroma, and possibly from other tissues including adipose tissue, synovium, periostium,
and skeletal muscle. These adult stem cells, known as mesenchymal stem cells (MSCs), possess
multi-lineage differentiation potential to form several cell types in the connective tissue cell
population including fibroblast, osteoblast, chondrocyte, adipocyte, and myoblast (Pittenger et al.,
1999). Due to their ease of isolation and differentiation potential, MSCs have been used in
numerous clinical trials for example for the treatment of large bone defects (Pittenger et al.,
1999), cartilage defect (Xian and Foster, 2006), osteogenic imperfecta (Horwitz et al., 1999), and
for the treatment of skin wounds (Wu et al., 2007).

Although MSCs can be readily isolated, the defining characteristics of MSCs remain
controversial among investigators. In particular, the lack of definite surface markers to
distinguish MSCs from their differentiating progeny has impeded the isolation of a homogeneous
MSC population required for large-scale cell therapy (Abdallah and Kassem, 2008).
Furthermore, despite the establishment of some surface markers for MSCs, these markers are not
always predictive of their differentiation potential (Phinney and Prockop, 2007). In 2006, the
International Society for Cellular Therapy (ISCT) proposed a universal minimal criteria for
characterization of human MSCs, which includes the ability of these cells to adhere to plastic
surface, their lack of hematopoietic cell surface markers, their expression of cell surface antigens
CD105, CD73, and CD90, and their capacity for differentiation in vitro into at least three lineages
(Dominici et al., 2006). However, problems still persist for investigators wishing to acquire a
large numbers of MSCs with stable phenotypes and differentiation potential. Reports for long-
term in vitro culture of MSCs suggested that MSCs exhibit altered and limited differentiation
potential, reduced cell proliferation rate, and cell growth arrest over the time (Bonyadi et al.,
2003; D'Ippolito et al., 1999; Stenderup et al., 2003).

11
In the following chapters, I present the results from my studies of the effect of LRAP across two
species and in two different stem cell types. Mouse ES cells are used as a model to study the
effect of LRAP on osteogenic induction of ES cells, as well as to explore the underlying
molecular mechanisms for LRAP to direct stem cells into the osteogenic phenotype. Human
BMMSCs are used as a model to study the effect of LRAP on osteogenic and adipogenic
induction in vitro, and to study the effect of LRAP on BMMSCs in vivo. The results from these
studies are predicted to provide a better understanding of the function of LRAP as a signaling
molecule, as well as to provide a preliminary data for potential use of LRAP for clinical
application, to facilitate the treatment of human osseous defects.

12
CHAPTER 1: THE FUNCTION OF LEUCINE-RICH AMELOGENIN PEPTIDE ON MOUSE
EMBRYONIC STEM CELLS INDUCED TO FORM BONE

The formation of hard tissues such as bone, cartilage and tooth occurs by a process of matrix-
mediated biomineralization in which intracellular and extracellular organic proteins regulate the
initiation, growth and deposition of mineral crystals (Weiner et al., 2005). In developing enamel
matrix, the majority of the organic extracellular matrix proteins are comprised by a group of
highly conserved, structural proteins called amelogenins (Snead, 2003). Amelogenins have been
identified to function as cell adhesion molecules and to facilitate nucleation and growth of
hydroxyapatite crystals during the mineralization phase of amelogenesis (Tarasevich et al., 2007).
As a result of amelogenin mRNA alternative splicing, different amelogenin isoforms are detected
in the developing enamel matrix (Iacob and Veis, 2006). Although the function of amelogenin
has been proven indispensable for proper development of enamel (Gibson et al., 2001), the
function of various amelogenin isoforms in enamel biomineralization, and else where, is not
known.

Amelogenin null mice develop hypomineralized enamel lacking normal prism structure, but are
healthy and fertile (Gibson et al., 2001). However, these mice are smaller than their wild-type
littermates prior to weaning, with the wild-type mice having a greater average weight each day
within the 3-week period. Using reverse transcription-polymerase chain reaction (RT-PCR),
LRAP expression is detected in wild-type teeth, brain, eye, and calvariae, but not in any of the
samples from the amelogenin null mice. The smaller size of the amelogenin null mice could
potentially be due to their lack of LRAP expression, leading to a delay in development (Li et al.,
2006).

13
In the experiment described in this chapter, I examine the effect of exogenous LRAP on ES cell
differentiation to an osteogenic lineage by monitoring changes in gene expression and changes in
mineral deposition in the presence of exogenous LRAP. In addition, I explore the effect of LRAP
to rescue the phenotype observed in osteogenic differentiation of amelogenin-null ES cells. I
hypothesize that LRAP functions as a moonlighting protein (Jeffery, 2003) working outside its
original location in enamel matrix and that LRAP can serve as a signaling molecule in other
tissue. The data presented here are predicted to lay the groundwork for understanding the
molecular mechanism(s) underlying LRAP-mediated ES cell commitment to a tissue regeneration
paradigm.

14
MATERIALS AND METHODS

1. Mouse embryonic stem cell culture

Mouse ES cells (RW4) were maintained on an irradiated MEF feeder layer in knockout D-MEM
(Invitrogen) containing 15% FBS (Hyclone) supplemented with murine leukemia inhibitory
factor (LIF; 1000 U/mL; Chemicon), 10 mM HEPES buffer solution, 0.1 mM MEM non-
essential amino acids solution, 0.05 mM 2-mercaptoethanol, 2 mM L-glutamine, penicillin (50
U/mL) and streptomycin (50 μg/mL). Cells were grown at 37°C in 7.5% CO
2
with daily media
changes. Cell passage was achieved by treatment of the cells with 0.05% trypsin and reseeded
the recovered cells onto a fresh MEF feeder layer. Amelogenin-null (KO) ES cells were
maintained in the same condition as the wild-type ES cells.

2. Generation of mouse amelogenin-null ES cells

Amelogenin-null (KO) ES cells were generated as previously described (Zhu et al., 2006). The
mouse amelogenin gene is located on the X chromosome and consists of 9 exons (Li et al., 1998;
Papagerakis et al., 2005). The open reading frame for translation initiation is located on exon 2
(Gibson et al., 1998). Deletion of exon 2 will result in the removal of signal peptide and the first
two amino acids of the secreted amelogenin protein. In this experiment, deletion of exon 2 was
achieved by replacing exon 2 with a neomycin resistant cassette in the reverse direction (Fig 3A).
After electroporation, ES cell clones surviving G418 selection were isolated and their genotypes
were confirmed by Southern blot analysis, which revealed 2 amelogenin-null ES cell clones and 2
wild-type ES cell clones (Fig 3B).

15
Fig 3.

A.

B.

Generation of amelogenin deficient murine embryonic stem cell lines. (A) Comparison of
genomic organization at the amelogenin loci in wild-type (WT) and knockout (KO) mouse ES
cells. The second exon of the murine amelogenin gene was replaced by the neomycin cassette
inserted in the opposite orientation to transcription. The 3'-flanking probe hybridizes to a 6.0-kb
fragment in the wild-type gene and a 3.6-kb band in the knock-out allele. The open box
represents exons and the solid line represents introns.

WT

KO

16
The numeral denotes exon number. Neo: neomycin resistant gene, E: EcoRI site, H: HindIII site,
F: amelogenin forward primer in exon 1 (5 -ATCAAGCATCCCTGAGCTTCAGAC-3 ),
R: amelogenin reverse primer in exon 6D (5 -GCTCAGGAAGAATGGGGGACAG-3 ).
(B) Southern blot of genomic DNA from ES cell clones. The genomic DNA was digested with
EcoRI, and the blots were probed with the 276-bp HindIII-EcoRI fragment not included in the
targeting vector, which revealed two knock-out and two wild-type ES clones.

17
3. Induction of cell differentiation

Mouse ES cells were induced to form embryoid bodies (EBs) in a hanging drop culture according
to a standard protocol (Phillips et al., 2001). After 2 days, EBs were collected and resuspended in
100 mm petri dishes in knockout D-MEM supplemented with 10
-7
M all-trans retinoic acids for
additional 3 days with the medium changed every day.

After 3 days, the EBs were collected, and transferred to gelatinized 6-well tissue culture plates for
osteogenic induction experiments using one of the three media each with different supplements.
The “basal media” was prepared by using MEM-alpha media, supplemented with 15% batch-
tested FBS, 10 mM HEPES buffer, 0.1 mM MEM non-essential amino acids solution, 0.05 mM
2-mercaptoethanol, 2 mM L-glutamine, penicillin (50 U/mL) and streptomycin (50 μg/mL). The
“control media” was prepared by supplementing the basal media with ascorbic acid (50 μg/mL)
and -glycerophosphate (10mM). The “LRAP media” was prepared by adding chemically
synthesized and purified murine LRAP protein (10ng/mL) to the control media. The EBs were
cultured in one of these media at 37°C in 5.0% CO
2
for 15 days with the media changed every
other day. A schematic diagram for different stages of EB differentiation is shown in Fig 4.

18
Figure 4.

A schematic diagram showing stages of differentiation of mouse embryonic stem (ES) cells. In
the stage of embryoid body (EB) propagation, mouse ES cells were induced to differentiate by the
formation of EB (EB d0) in the absence of mouse embryonic feeder layer and leukemia inhibitory
factor (LIF). The EBs were induced to form for two days prior to EB collection and suspension
in 10
-7
M all-trans retinoic acid (EB d2) for 3 days. At day-5 (EB d5), EBs were collected and
transferred to a cell culture dish to begin the stage of LRAP induction. In the stage of LRAP
induction, EBs at day-5 (EBd5) were grown in the mineralization medium containing LRAP
(10ng/mL) with medium change every other day. Five days after LRAP induction (EB d10), the
EBs were collected for analysis of osteogenic gene expression. The mineral formation, as well as
additional osteogenic gene expression, was assessed in EBs at day-20 (EB d20), which marked
the ending of the stage of LRAP induction.

19
4. RNA extraction, cDNA synthesis, and quantitative real-time PCR analysis

RNA was isolated from EB cultures at day-10, and at day-20 (day 0=EB formation) by using
RNAqueous®-4PCR Kit (Ambion) following the manufacturer’s instructions. Synthesis of
cDNA was performed by using RETROscript® Kit (Ambion). For cDNA template preparation, 2
μg of total RNA from each EB sample was used in a 20 μL reaction.

For quantitative real-time PCR analysis, a 25 μL reaction was prepared for each sample.
Included in this reaction volume was 1 μL of the resulting cDNA, iQ SYBR

green supermix (Bio-
Rad) containing dNTP and iTaq DNA polymerase, and the appropriate primers (Table 1). The
resulting threshold cycle (C
T
) value from each primer pair was normalized with the C
T
value for
Gapdh, which served as an internal control.

5. Analysis of mineral deposition

EBs at day-20 were stained with Alizarin Red. Quantification of calcium concentration was
measured by means of the intensity of absorbed light at 612 nm (QuantiChrom Calcium Assay
kit; BioAssay Systems). The total amount of protein in each sample was used as a standard to
normalize the calcium data.

20
6. Statistical analysis

For gene expression analysis and calcium concentration analysis, independent measurements
were performed in triplicate and averaged for control of internal error. Each experiment was
repeated at least three times. Mean and standard deviation from each experiment were used to
statistically analyze the difference between each pair of samples. A P-value less than 0.05 was
considered significant.

21
RESULTS

1. Different amelogenin isoforms are expressed during osteogenic differentiation of ES cells

To understand the function of LRAP during the differentiation of mouse ES cells, we first
determined the expression level of LRAP during differentiation for both wild-type ES cells and
amelogenin-null ES cells. Mouse ES cells were induced to form EBs (day-0), followed by
subsequent osteogenic induction for 20 days according to an established protocol (Phillips et al.,
2001). At selected time points, RNA samples were extracted for PCR analyses using a forward
primer in exon 1 and a reverse primer in exon 6D (Table 1) of the amelogenin gene to detect all
possible splicing variants. The results showed that amelogenin transcripts were detected only in
wild-type ES cells but not in amelogenin-null ES cells (Fig 5). Only the full-length M180
isoform (598 bp) was expressed in EBs prior to osteogenic differentiation (Fig 5 lane 3). LRAP
(aka M59; 235 bp) was detected in differentiating cells (Fig 5 lanes 5, 7, 9), suggesting its
functional implication during the osteogenic differentiation process. An additional isoform,
M156 (520 bp) was also detected in differentiating cells (Fig 5 lanes 5, 7, 9).

22
Table 1. Real-time PCR primer sequences and their annealing temperatures

Gene Primer sequence (5 to 3 ) Annealing
temperature (°C)
Bsp Forward - GAGACGGCGATAGTTCC 55.7
Reverse - AGTGCCCGTAACTCAA
Osx Forward - CCCTTCTCAAGCACCAATGG 55.5
Reverse - AAGGGTGGGTAGTCATTTGCATA
Runx2 Forward - CCGTGGCCTTCAAGGTTGT 55.0
Reverse - TTCATAACAGCGGAGGCATTT
Amelogenin Forward - ATCAAGCATCCCTGAGCTTCAGAC 55.0
Reverse - GCTCAGGAAGAATGGGGGACAG
Wnt8b Forward - TTGGGACCGTTGGAATTGCC 55.0
Reverse - AGTCATCACAGCCACAGTTGTC
Sfrp4 Forward - GTGGCGTTCAAGGATGATGCTTC 55.0
Reverse - TTACTGCGACTGGTGCGACTG
Dkk1 Forward - ATGAGGCACGCTATGTGCT 55.0
Reverse - CGTTGTGGTCATTACCAAGG
Dkk2 Forward - CAGTCACTGAGAGCATCCTCA 55.0
Reverse - CCTGATGGAGCACTGGTTTGC
Gapdh Forward - CATGGCCTTCCGTGTTCCTA 55.0
Reverse - GCGGCACGTCAGATCCA

23
Fig 5.

Identification of amelogenin splicing isoforms during osteogenic differentiation of mouse
embryonic stem cells. Mouse wild-type RW4 (WT) and amelogenin knock-out -ES cells (KO)
were induced to form embryoid bodies (EBs), and subsequently induced to osteogenic
differentiation (Phillips et al., 2001). RNA samples were extracted from EB-derived cells at day-
5, day-10, day-15, and day-20 (day-0=EB formation). First strand complementary DNAs
(cDNAs) were subject to PCR analyses using a forward primer in exon 1
(5 -ATCAAGCATCCCTGAGCTTCAGAC-3 ) and a reverse primer in exon 6D
(5 -GCTCAGGAAGAATGGGGGACAG-3 ) of the amelogenin gene to detect all possible
splicing variants. GAPDH was used as an internal control. Top panel: amelogenin; Bottom panel:
GAPDH. Lane 1, negative control; Lane 2, positive control; Lane 3, RW4 day-5; Lane 4, KO
day-5; Lane 5, RW4 day-10; Lane 6, KO day-10; Lane 7, RW4 day-15; Lane 8, KO day-15; Lane
9, RW4 day-20; Lane 10, KO day-20; M, marker.

24
2. LRAP enhanced mineral deposition in osteogenic-induced mouse ES cells

To analyze the terminal phenotype, EBs at day-20 were stained with Alizarin red to detect
mineral deposits in the matrix. In both RW4 and KO ES cell differentiation, increased dye-
stained area was evidenced in control group, suggesting that both RW4 and KO ES cells
underwent proper osteogenic cell differentiation (Fig 6A). The addition of exogenous LRAP to
the control media resulted in a marked increase in mineral deposition, when compared to
outcomes from control media alone, suggesting enhanced mineral formation with LRAP
treatment (Fig 6A).

Quantification of the calcium accumulated in the matrix was achieved using the QuantiChrom
calcium assay kit to measure the amount of free calcium. In accordance with the visual record
from Alizarin red staining, the calcium concentration in the control group was significantly higher
(2-fold) than that observed for the basal group. The calcium accumulation in the LRAP-treated
group was significantly higher than the control group for both RW4 (4-fold) and amelogenin KO
ES cells (1.5 fold) (Fig 6B). Comparing RW4 and KO ES cells, a significant reduced level of
calcium accumulation was observed in the basal, control, and LRAP-treated groups for KO cell
types. Noticeably, LRAP could partially rescue the reduced level of calcium deposited in the
matrix created by KO ES cells (Fig 6B, Table 2).

25
Fig 6.

A.

B.

26
ES cell differentiation and mineral nodule formation. (A) Alizarin red staining. RW4 ES cells, or
amelogenin-KO ES cells were induced to osteogenic differentiation in basal media, control
media, or LRAP media. At day-20, the cells were analyzed for mineral nodule formation by
Alizarin red staining. Both RW4- and KO-ES cells underwent osteogenic differentiation as
indicated by the dye accumulation in the synthesized matrix. A marked increase in mineral
formation, as indicated by the increased number of nodules and the increased intensity of Alizarin
dye-stained areas (seen as the color red), was observed in LRAP-treated culture for both wild-
type and amelogenin KO embryonic stem cells. KO = amelogenin null ES cells, RW4 = wild-
type ES cells, Basal = basal media, Control = control media, LRAP = LRAP media.
(B) Quantification of calcium concentration in osteogenic-induced ES cells. A statistically
significant increase for calcium concentration was observed in the LRAP-treated group from both
wild-type RW4 and amelogenin KO EBs. The calcium concentration for KO EBs (patterned
bars) was significantly lower than the calcium concentration for wild-type EBs (filled bars).
LRAP partially rescued the reduced calcium accumulation phenotype observed in KO ES cells.
WT = wild-type ES cells. KO = amelogenin-null ES cells. Basal = basal media, Control = control
media, LRAP= LRAP media. Relative calcium concentration was calculated by normalization of
the calcium concentration determined by dye binding to the total protein concentration for each
sample. **p<0.05, , which indicates a statistically significant difference.

27
Table 2. Calcium concentration analysis for embryoid body (EB) culture at day-20

Samples

RW4 EBs
(mg/dL)

KO EBs
(mg/dL)
Basal media 0.67±0.51 0.20±0.09
Control media 2.35±0.38 0.51±0.29
LRAP media
(LRAP 10ng/ml)
6.28±2.49 1.11±0.43
LRAP media
(LRAP 100ng/ml)
5.73±0.40 1.52±0.09

28
3. LRAP induces the expression of bone marker genes in osteogenic-induced ES cells

To explore the effect of LRAP on osteogenic gene expression, RNA from osteogenic-induced
RW4 EB culture at day-10 and day-20, was collected to represent the EBs at early and late stages
of differentiation, respectively. Gene expression was analyzed by quantitative real-time RT-
PCR. At day-10, the expression of two osteogenic marker genes, bone sialoprotein (BSP) and
osterix (Osx), in the control group was markedly increased. Analyses of the effect of LRAP on
gene expression revealed significant enhancement in the expression of BSP (Fig 7A) and Osx
(Fig 7C). At day-20, the EBs in the control group expressed increased level of BSP, but not Osx.
The addition of LRAP in the media resulted in marked increase in both BSP (Fig 7B) and Osx
expression (Fig 7D).

29
Fig 7.

A. BSP expression in EBs at day-10

B. BSP expression in EBs at day-20

30
Fig 7, Continued.

C. Osx expression in EBs at day-10

D. Osx expression in EBs at day-20

31
LRAP induced bone sialoprotein (BSP) and osterix (Osx) expression in embryoid bodies. Wild-
typed ES cells were induced to osteogenic differentiation. EBs at day-10 and day-20 of culture
duration were selected for RT-PCR analysis to represent early and late stages of osteogenesis,
respectively. Marked increase in the expression of BSP (A) and Osx (C) was observed in LRAP-
treated EB as early as at day-10 of culture, when compared to the values from the mineralization
control group. A more pronounced increase of BSP expression was observed in LRAP-treated
EB at day-20 of culture (B) compared to mineralization control. Osx expression was significantly
increased in LRAP-treated EB at day-20, but no change in Osx expression was observed in EB at
day-20 of culture in control mineralization media (D). Values on Y-axis represent fold change of
real-time PCR C
T
value compared to GAPDH that was used as an internal control. Basal = basal
media, Control = control media, LRAP = LRAP media. **p< 0.05, which indicates a statistically
significant difference compared to basal group,
##
p<0.05, which indicates a statistically
significant difference compared to control group.

32
DISCUSSION

In this chapter, I provide data supporting the function of LRAP as a signaling molecule that
enhances osteoblastic cell differentiation in mouse ES cells. My conclusion is supported by the
findings that LRAP is expressed by cells during osteogenic differentiation and that exogenous
LRAP increases mineral matrix formation and increases calcium accumulation in the matrix from
both wild-type ES cells and amelogenin-null ES cells. In addition, I show that the osteogenic
effect of LRAP results in increased expression of BSP and Osx at both early- and late-stage of
osteogenic differentiation. These results suggest a unique signaling role for LRAP during
osteogenesis of mouse embryonic stem cell, and support the previously reported osteogenic
functions of LRAP in other cell types (Boabaid et al., 2004; Nebgen et al., 1999; Veis et al.,
2000).

A previous study of the effects of BMP-2 on osteogenic differentiation of mouse ES cells
suggested that embryoid body cultured for 20 days, BMP-2 enhances mineral formation and
increases osteocalcin expression by approximately 4-fold as analyzed by semi-quantitative RT-
PCR (Phillips et al., 2001). Here, I demonstrate that LRAP increases as much as 4000-fold for
BSP expression, 5-fold for Osx expression, and 6-fold for calcium accumulation in LRAP-treated
EBs at the 20
th
day of culture. This dramatic increase in gene expression and calcium
accumulation suggests that LRAP exerts an important role as an osteo-inductive molecule that is
equal to or more potent than BMP-2 (Cheng et al., 2003) during osteogenic differentiation of
mouse ES cells.

33
However, other investigators suggest that the ability of LRAP to induce osteogenesis is dependent
upon the cell type used, their stage of differentiation and the local environment at the site of
action (Lacerda-Pinheiro et al., 2006b). Previous in vitro studies on the effect of LRAP on
expression of bone marker genes in several progenitor cell types demonstrated that LRAP
enhances osteogenesis. In mouse muscle fibroblasts, LRAP treatment caused an immediate
upregulation of Runx2, but the expression of Runx2 was diminished after 48 hours (Veis et al.,
2000). In mouse cementoblasts, LRAP treatment increased expression of osteopontin at 72 hours,
but had no effect on Runx2 or BSP expression (Boabaid et al., 2004). Runx2 expression is
believed to be critical for early osteoblast differentiation and osteoblastic cell lineage
commitment (Otto et al., 1997). Based on these observations, one may predict that Runx2
expression is more susceptible to modulation in less differentiated cells, such as fibroblasts, when
compared to more differentiated cells, such as cementoblasts. However, in the case of
pluripotent ES cells, we observed little change in Runx2 expression for embryoid bodies at day-
10 and at day-20 of culture with LRAP treatment. On the other hand, we detected a substantial
increase in expression of BSP and Osx, two markers for bone development detected in more fully
differentiated osteoblasts (Ganss et al., 1999; Nakashima et al., 2002). In our study, we selected
two time points of EB culture that represent the early (EB day-10) and the late stage of ES cell
differentiation (EB day-20). Therefore, it is possible that by day-10 of culture, differentiation has
progressed sufficiently, as indicated by expression of the late markers of differentiation, BSP and
Osx, that further changes in Runx2 expression are not required. I hypothesize that Runx2
expression may be upregulated at an earlier time point than the EBs at day-10 that have been
tested to date.

34
In the amelogenin-null (KO) ES cell experiment, I explored the ability of LRAP to rescue the
reduced osteogenesis observed in the KO ES cells as a consequence of the loss of endogenous
amelogenin expression including the loss of endogenous LRAP which is observed in WT cells. I
have shown that although LRAP significantly increases mineral deposition and calcium
accumulation when compared to the control media used in KO ES cells, LRAP can only partially
rescue the amelogenin null phenotypes, even when the concentration of LRAP is increased to
from 10 ng/ml to 100 ng/ml (Fig 5B). The failure of LRAP to fully rescue osteogenesis suggests
that other amelogenin isoforms may be required to work coordinately with LRAP to fully induce
the osteogenic gene expression. One of the candidate amelogenin isoforms is the mouse
amelogenin splicing product expressed predominately by mouse ameloblasts called M180. The
M180 protein is encoded by exon 2, 3, 5, 6, and 7, consisting of 180 amino acids and is believed
to be involved principally in controlling enamel crystal habit (Fincham and Moradian-Oldak,
1995; Zhu et al., 2006). A recent study in cementoblasts suggested that LRAP and p172, the
porcine ortholog of mouse M180, work together to promote cell proliferation and migration in
cementoblasts and in the periodontal ligament cells from the amelogenin-knockout mouse
(Hatakeyama et al., 2006).

For tissue engineering applications, there are limitations to the use of conventional grafting
therapies primarily due to immuno-rejection and post-operative complications. In addition,
mesenchymal stem cells or lineage-committed cells have been reported to lose their proliferative
capacity and identity during ex vivo expansion (Mauney et al., 2004). Therefore, the result from
this study of LRAP-directed differentiation of ES cells is predicted to provide another option for
the potential therapeutic use of LRAP as an alternative pharmacologic agent in bone tissue
engineering and for regenerative application in the repair of craniofacial and axial skeletal
defects.

35
The concept of moonlighting proteins suggests that one protein can exert two or more unique
functions in different tissues (Jeffery, 2003). I propose that LRAP, in addition to its original
function in enamel formation, also moonlights as a signaling molecule to induce bone formation
at other anatomical sites during development. The study to identify the signaling pathway(s)
responsible for osteogenic effect of LRAP will be explored and discussed in Chapter 3.

The study of amelogenin protein and its splicing isoforms, LRAP, to function as a signaling
molecule active during bone regeneration began only in the past decade (Hammarstrom, 1997)
(Veis et al., 2000). Several in vitro and in vivo studies have identified the role of LRAP as a
potential signaling molecule implicated in bone formation in different cell types (Boabaid et al.,
2004; Nebgen et al., 1999; Veis et al., 2000). In the present study, I chose to use embryonic stem
cells in these assays for amelogenin isoform signaling, since these cells are pluripotent and
display no previous overt differentiation pattern. Thus, their response to exogenous LRAP is as
unconstrained as is biologically possible today. I show that during their differentiation as
embryoid bodies, amelogenin isoforms, M56/LRAP and M156, are expressed and therefore are
available to provide instructive signals in the absence of a preexistent bias to cell commitment. I
identify that exogenous LRAP functions as a signaling molecule in the differentiation of mouse
ES cells, based on the ability of LRAP to upregulate bone marker gene expression and enhance
mineral deposition in these ES cells.

36
CHAPTER 2: THE FUNCTION OF LEUCINE-RICH AMELOGENIN PEPTIDE (LRAP)
DURING OSTEOGENIC INDUCTION AND ADIPOGENIC SUPPRESSION OF HUMAN
BONE MARROW MESENCHYMAL STEM CELLS

Extracellular signals participate in regulating cell fate decision of mesenchymal stem cells
(MSCs) to differentiate to various cell types including cells of mesodermal lineages (Bianco et
al., 2001) and non-mesodermal lineages (Dezawa et al., 2004; Luk et al., 2005). For mesodermal
lineage cell differentiation, the canonical Wnt signaling pathway is known to determine cell fate
between osteoblastogenesis and adipogenesis of MSCs. The inter-relationship between the
formation of fat and bone has also been demonstrated in osteoporosis patients where low bone
mass occurs in association with high fat mass deposition. Genetic analyses for patient suffering
from juvenile osteoporosis reveal mutations in low-density lipoprotein receptor related protein 5
(Lrp5), an essential co-receptor in the canonical Wnt signaling pathway (Boyden et al., 2002;
Kwee et al., 2005). Previous work from MacDougald and colleagues suggested that activation of
the Wnt signaling pathway leads not only to increased osteoblastogenesis but also to decreased
adipogenesis (Bennett et al., 2005) by inhibiting the adipogenic transcription factors C/EBP and
PPAR (Ross et al., 2000). Additionally, Wnt10b has been purposed to act as a key factor to
regulate bone mass as evidenced by the increased trabecular bone formed in transgenic mice
overexpressing Wnt10b, and contrasted by the finding of decreased bone mass and decreased
serum osteocalcin in Wnt10b
-/-
mice (Bennett et al., 2005). According to the results presented in
Chapter 1, LRAP acts as a signaling molecule to enhance bone formation of ES cells. Thus, in
the current chapter, I will explore the signaling function of LRAP to regulate cell fate decision of
human MSCs by identifying the signaling axis used by LRAP. I hypothesize that exogenous
LRAP stimulates osteogenesis and inhibits adipogenesis of human BMMSCs, and I will test this

37
hypothesis using both in vitro MSC culture and in vivo MSC implantation in
immunocompromised mice, following LRAP treatment.

MSCs can be isolated from different sites including the traditional sites of axial bone from the
iliac crest, femur, tibia, and spine, as well as non-traditional sites such as alveolar bone from the
orofacial region (Matsubara et al., 2005). Because alveolar bones are easily accessible, and the
derivation of MSCs is less invasive, MSCs obtained from the orofacial region can serve as an
alternative source for osseous defect repair (Akintoye et al., 2006). However, the difference
between the embryonic origin for orofacial bones and axial bones suggest that different
characteristic of MSCs isolated from these two sites may influence their fate (Helms and
Schneider, 2003). MSCs from orofacial origin (OF-MSCs) have been reported to possess a
greater proliferation rate but also demonstrated a decreased response to chondrogenic and
adipogenic induction when compared to MSCs from axial bones (Akintoye et al., 2006). In
addition, the amount of bone and bone marrow formed from OF-MSCs implantation in vivo has
been reported to be different from that of BMMSCs with decreased amount of bone and marrow
elements formed by OF-MSCs (Akintoye et al., 2006). In this chapter, I will explore the effect
of LRAP on BMMSCs and OF-MSCs both in vitro and in vitro. I hypothesize that LRAP
induces bone formation in MSCs obtained from both orofacial and long bone sources, but the
outcome for bone formation is unique based upon the embryologic region of the stem cells.

In addition to a function for LRAP in cell differentiation, there is little known about the capacity
of LRAP to influence cell proliferation. It is well-appreciated that a balance between cell
differentiation and cell proliferation is essential for the maintenance of cellular homeostasis, and
past evidence suggests that when the cells become differentiated, their ability to proliferate

38
subsequently decreased. For example, human corneal epithelial cells induced to increase
proliferation by Notch signal express increased level of Ki67 (a marker for cell proliferation), but
decreased level of cytokeratin 3 (CK3), a marker for epithelial cell differentiation (Ma et al.,
2007). On the other hand, when the Notch signal is inhibited and Ki67 expression is decreased,
the expression of CK3 is decreased (Ma et al., 2007). Additionally, for the control of bone
homeostasis, increased immature osteoblast cell proliferation is observed as osteoblast
differentiation is repressed (Engin et al., 2008). These observations suggest an inverse
relationship between cell proliferation and cell differentiation. In contrast, other evidence
suggests that the control of cell proliferation and cell differentiation may be independent of one
another. For example, LRAP has no effect on the differentiation marker alkaline phosphatase and
dentine sialoprotein in dental pulp cell culture, yet LRAP increases the rate of cell proliferation in
these cells (Ye et al., 2006). Furthermore, although LRAP is found to induce enamel epithelial
cell differentiation, it has no effect on cell proliferation (Le et al., 2007). Because of the
unpredictable results for cell proliferation in different experimental systems, in the current study I
will explore the effect of LRAP on human BMMSC cell proliferation. I hypothesize that LRAP-
treated human BMMSCs may exhibit decreased cell proliferation as a result of increased cell
differentiation.

The characteristics of MSCs are defined by several factors including their differentiation
potential, their ability to adhere to the plastic culture dish, and their expression of cell-surface
markers (Dominici et al., 2006). Despite the heterogeneity of MSCs and the controversies on
defining MSC surface markers, MSCs are generally considered to express a variety of select
surface antigens including Stro-1, CD29, CD73, CD90, CD105, CD166, and CD44, while lacking

39
the expression of hematopoietic surface antigens including CD34, CD45 and CD14 (Abdallah
and Kassem, 2008; Dominici et al., 2006). Changes in the distribution of the cell surface
markers expressed are believed to be associated with the niche location and the differentiation
potential of these cells. For example, a recent study has reported that the expression of CD106, a
cell surface antigen for MSCs, is downregulated upon the cell entering either the osteogenic,
chondrogenic or adipogenic differentiation of MSCs (Liu et al., 2008). Another study suggested
that MSCs cultured in adipogenic factors express decreased expression of CD90, along with
increased capacity for cell differentiation to osteogenic and adipogenic lineages (Campioni et al.,
2008). Therefore, based on the ability of LRAP to induce cells to osteogenic lineages, I
hypothesize that LRAP may also change the expression of the select surface antigen(s) on human
BMMSCs, and alter their proliferation. The experimental strategy outlined in this chapter will
address the issue of LRAP influences on surface marker expression and on cell proliferation.

40
MATERIALS AND METHODS

1. Human bone marrow mesenchymal stem cell culture

Human bone marrow mesenchymal stem cells (BMMSCs; AllCells LLC, Berkeley, CA) and
human orofacial human BMMSCs (OF-MSCs) were a gift from Dr Songtao Shi. Cell surface
markers including STRO-1, MUC18, CD18, CD34, CD45, CD44, CD16, CD105, CD106, and
CD90 were used to isolate the BMMSCs and OF-MSCs free from possible hematopoietic cell
contamination by fluorescence-activated cell sorting (FACS). Human BMMSCs were cultured
in MEM- medium, supplemented with 15% batch-tested fetal bovine serum, 100 μM ascorbic
acid 2-phosphate, 2 mM L-glutamine, penicillin (50 U/mL) and streptomycin (50 μg/mL). Cells
were grown at 37°C in 5.0 % CO
2
with medium changed every 3 days. Human OF-MSCs were
cultured and maintained under the same condition as human BMMSCs.

2. Induction of cell differentiation

Human BMMSCs were cultured until over-confluent. Induction for osteogenic differentiation
was initiated by the culture of BMMSCs in osteogenic medium consisting of -MEM
supplemented with 15% batch-tested FBS, 2 mM L-glutamine, 100 μM L-ascorbic acid 2-
phosphate, 1.8 mM KH
2
PO
4
, 10 nM dexamethasone, 50 U/mL penicillin and 50 μg/mL
streptomycin in the presence or absence of LRAP (10 ng/mL). Osteogenic induction for human
OF-MSCs was achieved following the same protocol.

41
Induction for adipogenic differentiation was initiated by the culture of BMMSCs in adipogenic
medium consisting of -MEM supplemented with 15% batch-tested FBS, 2 mM L-glutamine,
100 μM L-ascorbic acid 2-phosphate, 50 mM isobutyl-methylxanthine (Sigma), 0.5 mM
hydrocortisone (Sigma), 6 mM indomethacin, 10 mg/mL insulin (Sigma), 50 U/mL penicillin and
50 μg/mL streptomycin in the presence or absence of LRAP (10 ng/mL).

3. Bone marrow mesenchymal stem cell transplantation

Human BMMSCs or OF-MSCs were cultured until over-confluent and held in this state for 3
days. At day-3, human BMMSCs or OF-MSCs (4 10
6
cells) were collected and amalgamated
with 40 mg hydroxyapatite/tricalcium phosphate (HA/TCP) ceramic particles (Zimmer Inc.) and
LRAP (10ng/mL). The amalgamated mix of cells, particles and LRAP were incubated at 37°C
with 90-minute rotation and subsequently transplanted into the dorsal subcutaneous region of 8-
week-old immunocompromised beige mice (NIH-bg-nu/nu-xid, Harlan Sprague Dawley). The
procedure was performed under institutional approved animal protocols (NIDCR #04-317 and
USC#10874). After 8 weeks, the transplants were collected, fixed with 4% formalin, decalcified
with buffered 10% EDTA (pH 8.0), and embedded in paraffin. Tissue sections were achieved
after de-paraffinization and subsequent staining by hematoxylin and eosin.

42
4. Western immunoblot analysis

Osteogenic induced human BMMSCs or OF-MSCs at 1 week post-induction were collected from
their implant site and processed for Western immunoblotting. Cell lysate was prepared by
washing the cells with PBS 2 times, followed by the addition of M-PER Mammalian Extraction
Reagent (Pierce). The cells were collected in 2x SDS loading buffer consisting of 8% v/v SDS,
400 mM dithiothreitol, 200 mM Tris pH 6.8, 40% v/v glycerol, 0.4% w/v bromphenol blue.
Protein concentration was measured by Bio-Rad protein assay using known amounts of bovine
serum albumin to establish a standard curve for extrapolation of protein mass in the experimental
unknowns. Approximately 10 μg of proteins from each experimental sample group was loaded to
a 4%-20% Tris-glycine SDS-polyacrylamide gel electropheoresis (PAGE) gel and resolved by
size. The size-separated proteins were transferred to Immobilon-P membranes (Millipore) for 1
hour at 100 mA. The membrane was blocked with 5% non-fat milk in TBST (1xTBS, 0.1%
Tween-20) for 1 hour at room temperature. Rabbit anti-Runx2 antibody (1:500; Oncogene),
rabbit anti-Osteocalcin antibody (1:500; LF32, NIDCR), or mouse anti-PPAR antibody (1:200;
Santa Cruz) was added to the membrane and incubated at 4°C overnight. HRP-conjugated anti-
rabbit antibody (1:4000; Amersham Biosciences) or HRP-conjugated anti-mouse antibody
(1:2000; Amersham Biosciences) was used as a secondary antibody and incubated with the
membrane for 1h. The signal from the antibody-antigen complex was detected using an ECL
detection system and normalized to the amount of GAPDH from the same sample. Quantification
of the target antigen was achieved by scanning the resulting X-ray film on a scanner (Kodak
Image Station 1000; PerkinElmer Life Sciences). ImageJ software version 1.37v (National
Institutes of Health, USA) was used to compare dentistry signals among the samples.

43
5. Analysis of mineral deposition and fat deposition

Human BMMSCs 4-week post-osteogenic-induction were stained with Alizarin Red.
Quantification of calcium concentration was measured as described in chapter 1, briefly relying
on dye uptake to measure calcium deposited in the cell matrix. The total amount of protein in
each sample was used as a standard with which to normalize calcium concentration. Human
BMMSCs 4-week post-adipogenic-induction were stained with Oil Red O (0.5 g Oil Red O in
100 ml isopropanolol) for analysis of triglyceride and lipid deposition.

6. Cell proliferation assay

Human BMMSCs were maintained in culture medium consisting of MEM- medium,
supplemented with 15% batch-tested fetal bovine serum, 100 μM ascorbic acid 2-phosphate, 2
mM L-glutamine, 50 U/mL penicillin and 50 μg/mL streptomycin, until the cells reached 30-40%
confluence. For BrdU labeling, the culture medium was removed and replaced with BrdU
labeling regent (labeling reagent 1 ml in 100 ml culture medium; Zymed) with or without LRAP
(10 ng/mL) at 37°C overnight. The next day, cells were washed with PBS 2 times, removed from
the culture dish by 0.05% trypsin and dispersed into single cells prior to collection and staining
by an immunohistochemical procedure modified from the manufacturer’s protocol (BrdU
Staining Kit; Zymed). In brief, peroxidase quenching solution (10 mL 30% H
2
O
2
in 90 mL
methanol) was mixed with the cells for 10 minute followed by PBS washes (2 min, 3 times).
Cells were permeabilized in the diluted denaturing solution for 30 min, washed with PBS, and
non-specific sites modified with blocking solution for 10 minutes.

44
Biotinylated mouse anti-BrdU reagent, serving as the primary antibody, was added to the cells for
60 minutes followed by PBS washes (2 min, 3 times). Streptavidin Alexa Fluor 488 conjugate (1
μg/mL; Invitrogen) was used a secondary antibody to label the cells with fluorescent conjugated
streptavidin for 10 min followed by PBS washes (2 min, 3 times). The cells were resuspended in
FACS medium (3% FBS and 0.1% NaN
3
in PBS) and processed for flow cytometry analysis in
flow cytometer (Epics XL-MCL; Beckman Coulter).

7. Fluorescence-activated cell sorting analysis

Human BMMSCs were maintained in culture medium consisting of MEM- medium,
supplemented with 15% batch-tested fetal bovine serum, 100 μM ascorbic acid 2-phosphate, 2
mM L-glutamine, 50 U/mL penicillin and 50 μg/mL streptomycin. At 40% confluence, LRAP
(10ng/mL) was added to the medium and incubated at 37°C in 5% CO
2
for 24 hours. The next
day, cells were removed from the plate by 0.05% trypsin and reconstituted in FACS buffer (3%
head-inactivated fetal bovine serum and 0.1% NaN
3
in PBS). Approximately 1 X 10
5
cells/100
μL FACS buffer were transferred to each polypropylene tube (BD Bioscience), and the primary
antibody (1 μg) was added to the cells, mixed and incubated for 45 minutes at 4°C. Primary
antibodies used were phycoerythrin (PE) conjugated mouse IgG
1
anti-human CD90/CD73 (BD
Bioscience), mouse IgG
1
anti-human CD34/CD45 (BD Bioscience), and mouse IgM anti-human
STRO-1 ( a gift from Dr. Stan Gronthos, Institute of Medical and Veterinary Science, Australia).
Subclass-matched antibodies including non-immune mouse IgM (Southern Biotechnology
Associates, Inc.) and IgG
1
(BD Bioscience) were used as controls. After incubation, cells were
washed twice with FACS buffer and reconstituted in 100 μl FACS buffer.

45
PE-conjugated goat F(ab )
2
anti-mouse IgM ( μ chain specific; Biosource/Invitrogen) was used as
a secondary antibody to mouse IgM anti-human STRO-1. PE-conjugated anti-mouse IgG (H+L)
(Southern Biotechnology Associates, Inc.) was used as a secondary antibody to mouse IgG
1
anti-
human CD34/CD45. The cells were incubated with secondary antibodies for 30 minutes at 4°C
followed by washing and resuspension in 1 ml FACS buffer. The cells were sorted using a flow
cytometer (Epics XL-MCL; Beckman Coulter) by collecting 10,000 events. The percent of
positive cells was normalized by the percent of cells positive to the subclass-matched antibodies.

46
RESULTS

1. LRAP increases Runx2 and OCN expression in human BMMSCs and OF-MSCs

Human BMMSCs and human OF-MSCs were cultured to overconfluent density prior to
osteogenic induction. The cells were induced to osteogenic differentiation for a week and cells
were extracted and protein samples were analyzed for osteogenic markers. With the presence of
LRAP in the osteogenic medium, Runx2 protein signal was upregulated in both BMMSCs (1.24-
fold) and OF-MSCs (2.57-fold) (Fig 8A, 8B). Osteocalcin protein signal was upregulated 2-fold
in LRAP-treated BMMSCs and 2.18-fold in LRAP-treated OF-MSCs (Fig 8A, 8B). When
comparing BMMSCs with OF-MSCs, OF-MSCs exhibited a lesser degree (50% less) of Runx2
protein level, but greater OCN protein level (700% more) at 1-week post osteogenic induction.

47
Fig 8.

A.

48
Fig 8, Continued.

B.
Runx2

OCN

49
LRAP increases Runt-related transcription factor 2 (Runx2) and osteocalcin (OCN) protein
expression in human bone marrow mesenchymal stem cells (hBMMSCs) and human
mesenchymal stem cells from the orofacial origin (hOF-MSCs). Human bone marrow
mesenchymal stem cells (hBMMSCs) and human mesenchymal stem cells from the orofacial
origin (hOF-MSCs) were subjected to osteogenic induction in the presence of LRAP (10ng/mL).
One week after osteogenic induction, BMMSCs or OF-MSCs were collected for analysis of bone
marker protein signals. (A) Western immunoblotting shows that LRAP (L) increased Runx2
protein level to 1.24-fold in BMMSCs, and Runx2 protein level to 2.57-fold in OF-MSCs, when
compared to Runx2 protein level for those cells grown in mineralization medium alone (M).
Control culture medium (C) was used as the base line control. Protein expression for OCN was
increased 2-fold in LRAP-treated BMMSCs and 2.18 fold in LRAP-treated OF-MSCs. GAPDH
was used as an internal control, and band intensity values were normalized to it to control for
loading errors. (B) Quantification for Runx2 and OCN protein expression levels from Western
blot analysis for human BMMSCs or OF-MSCs cultured in growth medium (C), mineralization
medium (M) or mineralization medium containing LRAP (L). Band intensity ratio was
calculated by normalizing the band intensity for each sample with the band intensity of GAPDH
using ImageJ software (National Institute of Health, USA). *P<0.05, which indicates a
statistically significant difference.

50
2. LRAP increases bone formation and hematopoiesis in vivo

To validate the function of LRAP in vivo, ex vivo expanded human BMMSCs or OF-MSCs were
subcutaneously transplanted with hydroxyapatite/tricalcium phosphate (HA/TCP) carrier in the
presence of LRAP into immunocompromised host mice. Eight weeks after transplantation, the
transplants were collected and analyzed for bone tissue formation by hematoxylin and eosin
staining. Histologic assessment of for bone tissue formation suggested that LRAP significantly
increased bone formation in both BMMSC (2-fold increase) and OF-MSC transplants (2-fold
increase) (Fig 9A, 9C). In addition, marked increases in hematopoietic cell population were
observed in the marrow space of the bone formed from BMMSCs and OF-MSCs in the presence
of LRAP (Fig 9A, 9B). When compared between the transplants from BMMSCs and OF-MSCs
in the absence of LRAP, the bone formed by OF-MSCs was thicker but contained much fewer
marrow spaces and fewer hematopoietic cells (Fig 9B). On the other hand, in the presence of
LRAP, bones formed from both BMMSCs and OF-MSCs were more uniform in thickness and
contained more marrow spaces with higher number of hematopoietic cells (Fig 9B).

51
Fig 9.

A.

B.

52
Fig 9, Continued.

C.

LRAP increases bone formation and bone marrow formation in vivo. Human bone marrow
mesenchymal stem cells (BMMSCs) or human mesenchymal stem cells from orofacial origin
(OF-MSCs) were mixed with hydroxyapatite/tricalcium phosphate (HA/TCP) carriers and LRAP
(10ng/mL). The mixture was subcutaneously transplanted into immunocompromised mice for
eight weeks. The tissue was recovered and processed for histologic examination and tissue
sections were stained with hematoxylin/eosin for analyses of bone formation. (A) At 4
magnification, an increase in bone formation was observed in both BMMSC or OF-MSCs treated
with LRAP at the time of transplantation, when compared to the BMMSCs or OF-MSCs
transplanted in HA/TCP carrier alone. The dashed boxes indicated areas illustrated in 20X
magnification shown in panel B. (B) At 20X magnification, an increase for the hematopoietic
cell population along with the bone-marrow like structures were observed both in BMMSC and
OF-MSC transplants in the presence of LRAP.

53
(C) Measurement of the bone area formed by the transplants show a 2-fold increase in bone area
formation in both BMMSC and OF-MSC transplants when these cells were prepared for
transplantation the presence of LRAP. *P<0.05, which indicates a statistically significant
difference.

54
3. LRAP inhibits adipogenesis of human BMMSCs

Mesenchymal stem cells have the capacity to differentiate into tissues derived from both
mesodermal and non-mesodermal origin. Among the tissues in mesodermal lineages, MSCs
have the plasticity to differentiate into multiple cell types including bone, cartilage, fat, muscle,
and hematopoietic cells. To examine whether LRAP alters the adipogenic differentiation
potential of BMMSCs, BMMSCs were cultured until overconfluent and induced to adipogenic
cell fate in the presence of LRAP (10ng/mL). Approximately one week after adipogenic cell fate,
protein samples were collected for analyses of adipogenic markers, and the cells were stained for
Oil red O staining. The results showed statistically marked decrease in lipid droplets in
adipogenic-induced human BMMSCs in the presence of LRAP when compared to BMMSCs
cultured in adipogenic medium alone (Fig 10A). Analysis for the number of lipid droplets
formed in BMMSC culture in the presence of LRAP showed statistically significant decrease (3-
fold) in formation of fat cells (P<0.05) (Fig 10B).

55
Fig 10.

A.

B.

LRAP inhibits adipogenesis of human bone marrow mesenchymal stem cells (BMMSCs).
Human BMMSCs were induced to adipogenesis and treated with LRAP (10ng/mL). (A) Oil red
O staining result for adipogenic induced BMMSCs shows lower numbers of lipid droplets formed
in BMMSCs cultured in adipogenic medium containing LRAP when compared to the number of
lipid droplets formed in BMMSCs cultured in adipogenic medium alone. (B) Quantification of
the lipid droplets formed in BMMSC culture in the presence of LRAP shows a significant
decrease (3-fold) in formation of adipogenic cells. *P<0.05, which indicates a statistically
significant difference.

56
4. The effect of LRAP on cell proliferation

To address the effect of LRAP on hBMMSC proliferation, cells were cultured in the presence of
LRAP (10ng/mL) and labeled with BrdU for 24 hours prior to cell sorting analyses for the percent
of BrdU positive cells. In comparison to the control culture, LRAP treatment had only a small
effect on cell proliferation of hBMMSCs showing a slight increase in the percent of BrdU
positive cells, from 13.05±1.05% in growth medium to 19.10±2.00% in the presence of LRAP
(Fig 11).

Fig 11.

The effect of LRAP on cell proliferation. The number of BrdU positive cells in bone marrow
mesenchymal stem cells (BMMSCs) cultured in LRAP containing medium (10 ng/mL) showed a
relatively small increase in the number of BrdU positive cells in BMMSC culture with LRAP.

57
5. The effect of LRAP on cell surface antigens of human BMMSCs

To characterize the response of human BMMSCs upon the addition of LRAP, hBMMSCs were
cultured in the presence of LRAP for 24 hours followed by cell surface marker analyses by flow
cytometry. After 24 hours, both BMMSCs and OF-MSCs in the control group expressed high
levels of mesenchymal stem cell markers, CD90 and CD 73, and low levels of hematopoietic
stem cells markers, CD34 and CD45 (Fig 12A). These markers remained relatively unchanged in
the LRAP-treated hBMMSCs (Fig 12B). However, the expression of STRO-1, an early marker
defining mesenchymal progenitor cells (Shi and Gronthos, 2003), was decreased in the LRAP-
treated human BMMSC culture (Fig 12B), suggesting that LRAP may have the effect to decrease
the population of the progenitor cells while increasing the population of the differentiated cells.

58
Fig 12.

A.

59
Fig 12, Continued.

B.

Characterization of surface antigens presented by BMMSCs following LRAP treatment.
(A) Flow cytometric analysis for bone marrow mesenchymal stem cells (BMMSCs) maintained
in control culture medium without LRAP for 24 hours. The majority of BMMSCs in the control
medium expressed the mesenchymal stem cells markers, CD90 (99.6%) and CD73 (99.8%), and
STRO-1 was expressed at moderate levels (44.8%). The hematopoietic cell markers, CD34 and
CD45, were expressed at low level (0.3%).

60
Non-immune mouse IgG
1
was used as a subclass isotype control for calculation of the number of
BMMSCs positive for CD90, CD73, CD34, and CD45, whereas mouse IgM was used as a
subclass isotype control for calculation of the number of BMMSCs positive control for STRO-1.
(B) Flow cytometric analysis for BMMSCs maintained in control culture medium with LRAP
(10ng/mL) for 24 hours. The numbers of STRO-1 positive cells was decreased to 32.7% in the
presence of LRAP, whereas the numbers of BMMSCs positive for CD90, CD73, CD34, and
CD45 remained relatively unchanged from those expressed in BMMSC culture without LRAP.
Non-immune mouse IgG
1
was used as a subclass isotype control for calculation of the number of
BMMSCs positive for CD90, CD73, CD34, and CD45, whereas mouse IgM was used as a
subclass isotype control for calculation of the number of BMMSCs positive control for STRO-1.

61
DISCUSSION

In this chapter, I have provided data that supports the hypothesis that LRAP functions as a
signaling molecule, serving to induce osteoblastogenesis in adult stem cells from human bone
marrow mesenchymal cell origins (BMMSCs) and orofacial bone marrow mesenchymal stem
cells (OF-MSCs). This conclusion is supported by (1) the in vivo transplantation experiment,
which demonstrates that LRAP enhances bone formation and bone marrow formation in
subcutaneously transplanted BMMSCs/OF-MSCs in immunocompromised mice, and (2) the in
vitro osteogenic differentiation experiment, which shows that LRAP enhances osteogenesis in
BMMSCs and OF-MSCs by upregulating two markers for bone formation, Runx2 and OCN.
Furthermore, in addition to its role in enhancing bone formation of MSCs, the result from in vitro
differentiation also suggests that LRAP inhibits adipogenic differentiation potential during
adipogenic differentiation of human BMMSCs. Additionally, I have provided further insight
into the effect of LRAP on cell proliferation. I have found that although LRAP has the capability
to induce cell differentiation, the cell proliferation rate in the treated cells remains largely
unchanged by LRAP. Lastly, my experiment has suggested that LRAP may be capable of
shifting human BMMSCs out of mesenchymal progenitor population by decreasing the
population of STRO-1 positive cells.

62
1. LRAP induces bone-tissue formation along with the formation of bone-marrow like structure

The ability of LRAP to induce bone formation is in agreement with previous in vitro osteogenic
differentiation for other cell types (Boabaid et al., 2004; Lacerda-Pinheiro et al., 2006a; Veis et
al., 2000; Warotayanont et al., 2007), also supporting such a role for LRAP is my experiments on
mouse ES cells described in chapter 1. These data together suggest that LRAP has the capability
to function as a signaling molecule for stem cells from both human and mouse species. This
observation is confirmed by not only the in vitro culture, but also the in vivo animal model. In
human MSC transplants, where LRAP induced increased bone tissue formation, high numbers of
hematopoietic cells resembling a marrow-like structure was also observed. This phenomenon of
marrow formation in the formed bone is novel to the function of LRAP reported in the literature
to date.

The capability of LRAP to induce the formation of a bone marrow-like structure is more striking
in the osteoblastogenesis of implanted OF-MSCs. In the control group, where HA/TCP carriers
were placed with OF-MSCs in the absence of LRAP, bone tissue is formed, but very few marrow-
like structures are observed. This observation is in accord with the published data, which refer to
the ability of OF-MSCs to form bone, but with less hematopoietic tissue compared to BMMSCs
(Akintoye et al., 2006; Matsubara et al., 2005). In the current experiment, I have shown that in
the presence of LRAP, bone tissue formed by OF-MSCs exhibits not only increased bone mass,
but also increased areas of bone-marrow like structures, an outcome that closely resembles the
results obtained from human BMMSCs. In addition, in the presence of LRAP, OF-MSC-derived
bone tissue exhibits more regular bone thickness, comparable to the thickness observed for
BMMSC-derived bone tissue.

63
Analyses of bone markers in the tissue formed by both BMMSCs and OF-MSCs show marked
increase in Runx2 and OCN in the presence of LRAP (Fig 8A, 8B). Although this result
suggests that LRAP induces bone markers in MSCs from both sources, the intensity of Runx2
and OCN protein expression is different between BMMSCs and OF-MSCs. In contrast to LRAP-
treated BMMSCs, LRAP-treated OF-MSCs exhibit weaker Runx2 expression along with more
pronounced OCN expression (Fig 8A, 8B). Runx2 is a known as the marker for “early” bone
formation, whereas OCN is known as the marker for “late” bone formation (Aubin, 2001).
Therefore, the data from the current experiment may suggest that in comparison to BMMSCs,
OF-MSCs may have more rapidly differentiated to the latter stages of osteoblastogenesis, where
the cells are known to express higher levels of OCN and lower levels of Runx2.

Taken together, this unique and improved understanding of the effect of LRAP on BMMSCs and
OF-MSCs may provide a promising application for LRAP in further translational applications,
aiming to improve not only the quantity, but also the quality of the bone tissue, especially the
bone tissue derived from OF-MSCs, which are readily accessible as compared to the more
difficult procedure needed to harvest BMMSCs.

64
2. The role of LRAP on maintenance of homeostasis

The anatomical proximity of MSCs to hematopoietic stem cells (HSCs) and the HSC niche
suggests that MSCs may play a role in regulating hematopoietic homeostasis. Several lines of
evidence suggest that MSCs co-exist with HSCs and help regulate cell renewal and differentiation
of hematopoietic cells in the bone marrow stroma (Caplan, 1991; Dexter and Shadduck, 1980; Shi
and Gronthos, 2003; Weiss and Sakai, 1984), as well as the establishment of HSC niche (Calvi et
al., 2003; Moore and Lemischka, 2006). However, the hematopoietic differentiation potential of
MSCs varies by the cell types defined by cell surface antigens expressed by MSCs. A recent
study suggests that MSCs expressing CD146

(von Willebrand factor) on the surface are capable
of establishing a new hematopoietic environment in vivo, whereas the CD146

negative MSCs
with the same osteogenic differentiation potential are unable to establish the hematopoietic
microenvironment (Sacchetti et al., 2007). Given that LRAP is able to establish a bone marrow-
like environment in vivo, LRAP may be involved in maintaining hematopoietic homeostasis by
determining the population of the MSCs to those that favor hematopoiesis, such as by increasing
CD146 positive cells. In the current experiment, I observed the change of STRO-1 cell surface
marker upon the treatment of LRAP. STRO-1 is the mesenchymal stem cell markers normally
expressed on the surface of mesenchymal progenitor cells and on the outer wall of bone marrow
blood vessels (Shi and Gronthos, 2003). Treatment of the BMMSCs with LRAP results in a
decrease in STRO-1 population, suggesting that LRAP may have the function to switch the cell
population to that of a less pluripotent and a more differentiated phenotype. This observation is
supported by the ability of LRAP to induce cell differentiation as previously reported in this
chapter.

65
The results in this chapter suggest a function for LRAP during the induction of cell
differentiation; however, the implication for LRAP in cell proliferation remains less clear.
Previously reported data suggested that LRAP increases dental pulp cell proliferation (Ye et al.,
2006), but has no effect on enamel epithelial cell proliferation (Le et al., 2007). My current
finding for the effect of LRAP on cell proliferation suggests that LRAP has no effect on cell
proliferation of hBMMSCs following LRAP culture for 24 hours. This observation is in
agreement with the previous report that has shown no effect of LRAP on enamel epithelial cell
proliferation (Le et al., 2007), but is opposed by the reported proliferative effect of LRAP on
dental pulp cells (Ye et al., 2006). The paradox may be explained by variation in cell context
and/or the experimental setting used in each reported assay. In my current experimental strategy,
the concentration of LRAP used to alter cell proliferation is 10 ng/mL. This 10 ng/mL
concentration is used in all ES cell differentiation assays, as well as in the human BMMSC
differentiation assays, and the LRAP implantation experiments. According to Ye and
colleagues’ findings, LRAP enhanced dental pulp cell proliferation at much greater
concentrations, between 100 and 500 ng/mL (Ye et al., 2006) than the 10 ng/mL used herein.
Therefore, it may be possible that cell proliferation rate is controlled by LRAP concentration.
However, increasing the concentration failed to enhance cell proliferation in enamel epithelial
cells (Le et al., 2007), suggesting that there are other factors in these experimental settings that
also participating in regulating cell proliferation.

66
The ability of LRAP to organize bone formation and increase bone mass in BMMSC and OF-
MSC cell implantation suggests that LRAP may have a function in bone homeostasis. Previous
evidence has suggested that in addition to its function in osteoblastogenesis, LRAP also plays a
role in the osteoclastogenesis, by regulating the expression of receptor activator of NF- B ligand
(RANKL)(Hatakeyama et al., 2006). RANKL, expressed on the surface of preosteoblastic cells
and bone marrow stromal cells, binds to its receptor RANK on the osteoclastic precursor cells to
subsequently activate osteoclastic activity (Boyce and Xing, 2008). In amelogenin-null mice,
mice lacking the LRAP isoforms, as well as all other amelogenin isoforms, cementum defects are
observed, along with increased osteoclastogenesis activity and increased expression of RANKL
(Hatakeyama et al., 2003). When LRAP (0.1 μg/mL) was added back to the amelogenin-KO
cementoblast, RANKL expression was reduced (Hatakeyama et al., 2003), suggesting that LRAP
has a role in inhibition of osteoclastogenesis. Therefore, the ability of LRAP to induce bone
tissue formation may result from reduced osteoclastic activity. Additionally, unlike osteoblasts
whose origin is from mesenchymal stem cells, osteoclasts derive from hematopoietic stem cells
(Teitelbaum, 2000), suggesting another possible regulatory step for osteoclastic activity through
LRAP-inducing hematopoietic stem cell niche.

67
3. The role of LRAP in adipogenesis and the possible linkage to Wnt signaling pathway

Previous studies indicate that Wnt signaling has a role in promoting osteoblastogenesis but
suppressing adipogenesis (Bennett et al., 2005). Specifically, Wnt10b was identified as a positive
regulator for osteoblastogenesis by increasing bone mass, Runx2, Dlx5, and Osterix expression
(Bennett et al., 2007), and a negative regulator for adipogenesis by suppressing C/EBP and
PPAR (Kang et al., 2007). In agreement with the previously reported inverse relationship
between osteoblastogenesis and adipogenesis, I have observed that LRAP exerts an inhibitory
effect on adipogenesis of BMMSCs. This observation suggests that LRAP may act as a
molecular switch in determination of cell fate decision for BMMSCs. In addition, given the
similarity for the function of LRAP and Wnt signaling pathway for osteogenesis, LRAP may
function through the Wnt signaling pathway to regulate osteoblastogenesis. This hypothesis will
be tested and discussed in chapter three.

In summary, in this chapter I have shown data to support a remarkable function for LRAP to
regulate MSC differentiation to 3 different cell lineages, osteogenesis, adipogenesis and
hematopoiesis. In addition, I have proposed that LRAP is also implicated in the maintenance of
mesenchymal stem cell homeostasis, by shifting the cell population from progenitor cells to
differentiated cells, as well as organizing hematopoietic stem cell niche and facilitating bone
tissue remodeling. Furthermore, because the canonical Wnt signaling pathway is observed to
operate during osteogenesis (Kim et al., 2007), hematopoiesis (Nemeth and Bodine, 2007), and
adipogenesis (Ross et al., 2000) and the similarity between the function of LRAP and canonical
Wnt signaling, I hypothesize that LRAP may exert its osteogenic function through the activation
of the Wnt signaling pathway.

68
CHAPTER 3: ANALYSES OF SIGNALING PATHWAYS RESPONSIBLE FOR THE
EFFECTS OF LEUCINE-RICH AMELOGENIN PEPTIDE

Embryonic stem (ES) cells can be used as a model to study cellular differentiation. ES cells
grown in the presence of LRAP have shown that LRAP enhances osteogenic induction in mouse
ES cells (Warotayanont et al., 2007), an outcome that supports the previously reported function of
LRAP as a signaling molecule in other cell types (Boabaid et al., 2004; Lacerda-Pinheiro et al.,
2006a; Veis et al., 2000). Despite the evidence to support the osteoinductive property of LRAP,
studies to identify the mechanism of action of LRAP to induce bone formation have been limited.
LAMP-1, the lysosomal associate membrane protein receptor for receptor-mediated endocytosis,
has been identified as a cell surface receptor for LRAP (Tompkins et al., 2006). It remains
unclear how LAMP-1 relays the signal from LRAP occupancy of the receptor after the
internalization of the protein since the lysosomal pathway generally leads to protein degradation.
One study using the experimental strategy of inhibiting specific signaling pathways has shown
the addition of MAPK inhibitor resulted in the reduction of osteopontin mRNA expression in
osteogenic-induced mouse cementoblasts (Boabaid et al., 2004). Using the yeast two-hybrid
system, Wang and colleagues have identified three intracellular proteins bound to LRAP,
including Eef2, Fez1 and Lsm1(Wang et al., 2006), suggesting the possibility that the activity
from these other pathways may also influence cell fate decisions. However, these additional
pathways remain largely uncharacterized. Despite progress from these prior studies, in the
context of osteogenesis, it is still unclear whether LRAP mediates its effect through binding to a
cell surface receptor, binding to an intracellular protein(s), or activating another signaling
cascade.

69
The involvement of canonical Wnt signaling pathway in the determination of naïve cells to
commit to the osteogenic lineage (Baron et al., 2006) and adipogenic lineage (Ross and Golub,
1988) suggest that LRAP may exert its signaling property through activation of the Wnt signaling
pathway. The activation of the canonical Wnt pathway results in stabilization and nuclear
localization of -catenin, the downstream effector for the Wnt signal (Schohl and Fagotto, 2002).
In the nucleus, -catenin binds to its nuclear binding partners, T-cell factor/lymphoid enhancer-
binding factor (TCF/LEF) proteins, to mediate the transcription of Wnt/ -catenin target genes
(Brantjes et al., 2002), such as cyclin D1, p21, E-cadherin and BMP4 (Vlad et al., 2008). In the
absence of Wnt ligands, the -catenin is targeted for degradation by the -catenin destruction
complex containing Axin, APC, GSK3, and CK1 proteins (Huang and He, 2008).

In this chapter, I hypothesize that LRAP exerts its signaling function through activation of the
Wnt signaling pathway. This hypothesis is tested by the measurement of the Wnt-induced protein
stabilization of -catenin and the activation of TCF/LEF reporter activity. In addition, I will also
explore the change(s) in expression of Wnt-associated genes including the expression of
canonical Wnts such as Wnt7b and Wnt3a, the expression of Wnt inhibitors such as Sfrp-1 and
Dkk-1, and other selected genes associated with Wnt biological functions and regulations
(Table 3).

70
Table 3. Genes associated with mouse Wnt signaling pathway for polymerase chain reaction
(PCR) array analysis

Unigene GeneBank Symbol Description Gene Name
Mm.180
013
NM_010347 Aes Amino-terminal
enhancer of split
AL024115/Grg
Mm.384
171
NM_007462 Apc Adenomatosis
polyposis coli
AI047805/AU020952
Mm.236
84
NM_009733 Axin1 Axin 1 AI316800/Axin
Mm.226
175
NM_029933 Bcl9 B-cell
CLL/lymphoma 9
2610202E01Rik/803047
5K17Rik
Mm.119
717
NM_009771 Btrc Beta-transducin
repeat containing
protein
Beta-Trcp1/FWD1
Mm.299
735
NM_023465 Ctnnbip1 Catenin beta
interacting protein
1
1110008O09Rik/231000
1I19Rik
Mm.273
049
NM_007631 Ccnd1 Cyclin D1 AI327039/Cyl-1
Mm.333
406
NM_009829 Ccnd2 Cyclin D2 2600016F06Rik/AI2568
17
Mm.246
520
NM_007632 Ccnd3 Cyclin D3 9230106B05Rik/AA682
053
Mm.269
08
NM_146087 Csnk1a1 Casein kinase 1,
alpha 1
2610208K14Rik/463240
4G05Rik
Mm.216
227
NM_139059 Csnk1d Casein kinase 1,
delta
1200006A05Rik/AA409
348
Mm.236
92
NM_007788 Csnk2a1 Casein kinase 2,
alpha 1
polypeptide
Csnk2a1-rs4
Mm.728
6
NM_013502 Ctbp1 C-terminal
binding protein 1
BARS/CtBP1-L
Mm.246
240
NM_009980 Ctbp2 C-terminal
binding protein 2
AA407280/D7Ertd45e
Mm.291
928
NM_007614 Ctnnb1 Catenin (cadherin
associated
protein), beta 1
Catnb/Mesc
Mm.471
127
NM_172464 Daam1 Dishevelled
associated
activator of
morphogenesis 1
1700066F09Rik/2310028
E21Rik
Mm.825
98
NM_178118 Dixdc1 DIX domain
containing 1
4930563F16Rik/BC0481
82

71
Table 3, Continued.

Mm.214717 NM_010051 Dkk1 Dickkopf homolog
1 (Xenopus laevis)
mdkk-1

Mm.5114 NM_007888 Dvl2 Dishevelled 2, dsh
homolog
(Drosophila)
DVL2
Mm.258397 NM_177821 Ep300 E1A binding
protein p300
A430090G16/A730
011L11
Mm.28017 NM_134015 Fbxw11 F-box and WD-40
domain protein 11
2310065A07Rik/AA
536858
Mm.4465 NM_013890 Fbxw2 F-box and WD-40
domain protein 2
2700071L08Rik/FB
W2
Mm.254739 NM_013907 Fbxw4 F-box and WD-40
domain protein 4
Dac/Fbw4
Mm.4956 NM_010202 Fgf4 Fibroblast growth
factor 4
Fgf-4/Fgfk
Mm.6215 NM_010235 Fosl1 Fos-like antigen 1 AW538199/Fra1
Mm.4496 NM_008238 Foxn1 Forkhead box N1 D11Bhm185e/Hfh1
1
Mm.4573 NM_008043 Frat1 Frequently
rearranged in
advanced T-cell
lymphomas
AW060382
Mm.427436 NM_011356 Frzb Frizzled-related
protein
Frp/Sfrp3
Mm.249525 NM_008045 Fshb Follicle
stimulating
hormone beta
Fshbeta
Mm.246003 NM_021457 Fzd1 Frizzled homolog
1 (Drosophila)
AW227548/FZ-1
Mm.36416 NM_020510 Fzd2 Frizzled homolog
2 (Drosophila)
AL033370/AW4568
35
Mm.243722 NM_021458 Fzd3 Frizzled homolog
3 (Drosophila)
AU020229/Fz3
Mm.387968 NM_008055 Fzd4 Frizzled homolog
4 (Drosophila)
Fz4
Mm.150813 NM_022721 Fzd5 Frizzled homolog
5 (Drosophila)
5330434N09Rik/AI
427138
Mm.4769 NM_008056 Fzd6 Frizzled homolog
6 (Drosophila)
Fz6
Mm.297906 NM_008057 Fzd7 Frizzled homolog
7 (Drosophila)
Fz7

72
Table 3, Continued.

Mm.184289 NM_008058 Fzd8 Frizzled homolog
8 (Drosophila)
Fz8/mFZ8
Mm.394930 NM_019827 Gsk3b Glycogen synthase
kinase 3 beta
7330414F15Rik/843
0431H08Rik
Mm.275071 NM_010591 Jun Jun oncogene AP-1/Junc
Mm.255219 NM_010703 Lef1 Lymphoid
enhancer binding
factor 1
3000002B05/AI451
430
Mm.274581 NM_008513 Lrp5 Low density
lipoprotein
receptor-related
protein 5
BMND1/HBM
Mm.321990 NM_008514 Lrp6 Low density
lipoprotein
receptor-related
protein 6
Cd
Mm.2444 NM_010849 Myc Myelocytomatosis
oncogene
AU016757/Myc2
Mm.30219 NM_027280 Nkd1 Naked cuticle 1
homolog
(Drosophila)
2810434J10Rik/903
0215G15Rik
Mm.9001 NM_008702 Nlk Nemo like kinase AI194375
Mm.246804 NM_011098 Pitx2 Paired-like
homeodomain
transcription
factor 2
9430085M16Rik/Br
x1
Mm.443425 NM_023638 Porcn Porcupine
homolog
(Drosophila)
2410004O13Rik/A
W045557
Mm.260288 NM_019411 Ppp2ca Protein
phosphatase 2
(formerly 2A),
catalytic subunit,
alpha isoform
PP2A/R75353
Mm.294138 NM_016891 Ppp2r1a Protein
phosphatase 2
(formerly 2A),
regulatory subunit
A (PR 65), alpha
isoform
6330556D22Rik/PP
2A

73
Table 3, Continued.

Mm.295009 NM_009358 Ppp2r5d Protein
phosphatase 2,
regulatory subunit
B (B56), delta
isoform
TEG-271/Tex271
Mm.273605 NM_028116 Pygo1 Pygopus 1 2600014C22Rik
Mm.168257 NM_133955 Rhou Ras homolog gene
family, member U
2310026M05Rik/AI
182090
Mm.297431 NM_029457 Senp2 SUMO/sentrin
specific peptidase
2
2310007L05Rik/493
0538C18Rik
Mm.19155 NM_009144 Sfrp2 Secreted frizzled-
related protein 2
AI851596/Sdf5
Mm.42095 NM_016687 Sfrp4 Secreted frizzled-
related protein 4
SFRP4
Mm.27842 NM_012030 Slc9a3r
1
Solute carrier
family 9
(sodium/hydrogen
exchanger),
member 3
regulator 1
EBP-50/NHE-RF
Mm.279103 NM_011441 Sox17 SRY-box
containing gene 17
Sox
Mm.913 NM_009309 T Brachyury Bra/D17Mit170
Mm.440067 NM_009332 Tcf3 Transcription
factor 3
Tcf-3/Tcf7l1
Mm.31630 NM_009331 Tcf7 Transcription
factor 7, T-cell
specific
AI465550/TCF-1
Mm.278444 NM_011599 Tle1 Transducin-like
enhancer of split
1, homolog of
Drosophila E(spl)
C230057C06Rik/Gr
g1
Mm.38608 NM_019725 Tle2 Transducin-like
enhancer of split
2, homolog of
Drosophila E(spl)
Grg2/mKIAA4188

74
Table 3, Continued.

Mm.32831 NM_011915 Wif1 Wnt inhibitory
factor 1
AW107799
Mm.10222 NM_018865 Wisp1 WNT1 inducible
signaling pathway
protein 1
AW146261/Elm1
Mm.1123 NM_021279 Wnt1 Wingless-related
MMTV
integration site 1
Int-1/Wnt-1
Mm.5130 NM_009518 Wnt10a Wingless related
MMTV
integration site
10a
WNT10A
Mm.22182 NM_009519 Wnt11 Wingless-related
MMTV
integration site 11
WNT11
Mm.137403 NM_053116 Wnt16 Wingless-related
MMTV
integration site 16
E130309I19Rik
Mm.33653 NM_023653 Wnt2 Wingless-related
MMTV
integration site 2
2610510E18Rik/Int
1l1

Mm.159091 NM_009521 Wnt3 Wingless-related
MMTV
integration site 3
Int-4/Wnt-3
Mm.1367 NM_009522 Wnt3a Wingless-related
MMTV
integration site 3A
Wnt-3a/vt
Mm.20355 NM_009523 Wnt4 Wingless-related
MMTV
integration site 4
Wnt-4
Mm.287544 NM_009524 Wnt5a Wingless-related
MMTV
integration site 5A
8030457G12Rik/Wn
t-5a
Mm.321818 NM_009525 Wnt5b Wingless-related
MMTV
integration site 5B
AW545702/Wnt-5b
Mm.268282 NM_009526 Wnt6 Wingless-related
MMTV
integration site 6
AA409270/Wnt-6
Mm.56964 NM_009527 Wnt7a Wingless-related
MMTV
integration site 7A
AI849442/Wnt-7a

75
Table 3, Continued.

Mm.306946 NM_009528 Wnt7b Wingless-related
MMTV
integration site 7B
Wnt-7b
Mm.558 NM_009290 Wnt8a Wingless-related
MMTV
integration site 8A
Stra11/Wnt-8A
Mm.88365 NM_011720 Wnt8b Wingless related
MMTV
integration site 8b
WNT8B
Mm.218794 NM_139298 Wnt9a Wingless-type
MMTV
integration site 9A
Wnt14
Mm.3317 NM_010368 Gusb Glucuronidase,
beta
AI747421/Gur
Mm.299381 NM_013556 Hprt1 Hypoxanthine
guanine
phosphoribosyl
transferase 1
C81579/HPGRT
Mm.2180 NM_008302 Hsp90a
b1
Heat shock protein
90kDa alpha
(cytosolic), class
B member 1
90kDa/AL022974
Mm.343110 NM_008084 Gapdh Glyceraldehyde-3-
phosphate
dehydrogenase
Gapd
Mm.328431 NM_007393 Actb Actin, beta,
cytoplasmic
Actx/E430023M04R
ik
N/A SA_00104 RTC Reverse
Transcription
Control
RTC
N/A SA_00104 RTC Reverse
Transcription
Control
RTC
N/A SA_00104 RTC Reverse
Transcription
Control
RTC
N/A SA_00103 PPC Positive PCR
Control
PPC
N/A SA_00103 PPC Positive PCR
Control
PPC
N/A SA_00103 PPC Positive PCR
Control
PPC

76
In addition to the known implication of BMP pathway in osteoblastogenesis (Wang et al., 1990),
recent studies suggest that the amelogenin protein isoforms present in Emdogain can activate
BMP and TGF- signaling pathways (Kawase et al., 2001; Lee et al., 2008; Suzuki et al., 2005;
Takayama et al., 2005). Furthermore, amelogenin has been reported to bind to BMP-2 and
suppress Noggin, two activities that serve to promote bone formation (Saito et al., 2008). BMP
signals are mediated by the binding of BMPs to the BMP type I and type II serine-threonine
kinase receptors on the cell surface, resulting in phosphorylation of its downstream effector Smad
1, 5 and 8 in the cytosol (Massague, 1996). Phosphorylated Smad1, 5 and 8 proteins then form a
complex with Smad4 and the complex translocates into the nucleus where the N-terminal domain
of Smad4 binds DNA via a Smad binding element (SBE) to effect the transcription of the target
genes (Xiao et al., 2007) (Chen et al., 2004b; Shi et al., 1998).

Based on the previous evidence for the activation of BMP signaling by amelogenin, in the current
experiment, I hypothesize that the signaling function of LRAP may be related to the activation of
BMP signaling pathway. This hypothesis is tested by the measurement of phosphorylated Smad
protein and by using the Smad reporter activity based upon a construct containing concatamerized
BMP-Smad binding element (GCCG12) when the cells are treated by LRAP in the medium.

77
MATERIALS AND METHODS

1. Cell culture and differentiation

Mouse embryonic stem (ES) cells (cell line RW4; Genome Systems) were cultured and
maintained in an undifferentiated stage as described previously (Warotayanont et al., 2007). In
brief, the ES cells were grown on irradiated mouse embryonic fibroblasts (MEFs) in knockout D-
MEM (Invitrogen) containing 15% fetal bovine serum, supplemented with murine leukemia
inhibitory factor (LIF; 1000 U/mL; Chemicon), 10 mM HEPES buffer solution (Invitrogen). 0.1
mM MEM non-essential amino acids solution (Invitrogen), 0.05 mM 2-mercaptoethanol (Gibco),
2 mM L-glutamine (Invitrogen), penicillin (50 U/mL) and streptomycin (50 μg/mL) (Invitrogen)
in a humidified 7.5% CO
2
incubator at 37°C.

The induction of embryonic stem cell differentiation was performed according to a standard
protocol (Phillips et al., 2001). Embryoid bodies (EBs) were induced to form from the culture of
mouse ES cells (5x10
5
cells/ml) in rotary suspension culture, at 30 rpm, according to a published
protocol (Carpenedo et al., 2007)(Day 0=EB formation). EBs at day-2 were collected and
resuspended in knockout D-MEM supplemented with 10
-7
M all-trans retinoic acids for 3 days.
The resulting EBs (EB day-5) were collected and cultured in the mineralization medium
consisting of -MEM, 15% GIBCO knockout serum replacement, 10 mM HEPES buffer, 0.1
mM MEM non-essential amino acids solution, 0.05 mM 2-mercaptoethanol, 2 mM L-glutamine,
penicillin (50 U/mL) and streptomycin (50 μg/mL), ascorbic acid (50 μg/mL) and -
glycerophosphate (10mM)) with or without LRAP (10 ng/mL) to initiate osteogenic
differentiation.

78
Recombinant human sFRP-1 (20ng/mL; R&D Systems) was added to EB culture at day-5 in the
mineralization medium to perform the Wnt inhibition assay. Recombinant mouse Wnt3a
(100ng/mL; R&D Systems) was added to the mineralization medium when indicated to provide
Wnt signal. Purified scrambled LRAP peptide, containing the amino acid sequence
PPHMPLPGSPL SYEGYINVLT WEYQPLKSMR IRSPIKLQPP LPELAWPPLE ATDKEVD
(GenWayBiotech Inc.) was added to the medium as a control to the bioactive LRAP peptide.

For the culture of the MC3T3 osteoblastic cell line, MCT3T cells were maintained in -MEM
supplemented with 10% fetal bovine serum (FBS) (Invitrogen). Mineralization medium was
added to the cells to induce osteogenic differentiation when the cells reached approximately 90%
confluency.

2. Western immunoblot analysis

Embryoid bodies (EBs) were cultured until day-5 when osteogenic medium and LRAP was added
to the cells for 5-10 days. The EBs was collected by washing with PBS 2 times followed by the
addition of M-PER mammalian extraction reagent (Pierce). An aliquot of the protein was added
to 2X SDS loading buffer consisting of 8% v/v SDS, 400 mM dithiothreitol, 200 mM Tris pH 6.8,
40% v/v glycerol, 0.4% w/v bromphenol blue for polyacrylamide gel electrophoresis (PAGE).
Protein concentration was measured using the Bio-Rad protein assay on the lysate samples, or a
known amount of bovine serum albumin, to establish a standard curve for extrapolation of the
protein mass for unknown samples. Approximately 10 μg of proteins from each experimental
sample group was loaded to a 4%-20% Tris-glycine SDS-polyacrylamide gel electropheoresis
(PAGE) gel. The size-resolved proteins were transferred to Immobilon-P membranes (Millipore)
for 1 hour at 100 mA.

79
The membrane was blocked with 5% non-fat milk in TBST (1xTBS, 0.1% Tween-20) for 1 hour
at room temperature. Mouse anti -catenin antibody (1:2000; BD bioscience) or Phospho-
Smad1/Smad5/Smad8 antibody (1:2000; Cell Signaling) was added to the TBST and the
membrane was incubated at 4°C overnight. HRP-conjugated anti-mouse antibody (1:2000;
Amersham Biosciences) was used as a secondary antibody and incubated with the membrane for
1h. The antigen-antibody signal was detected by ECL detection system and normalized to the
amount of -actin from the same sample. Quantification of the signal was achieved by using
ImageJ software version 1.37v (National Institutes of Health, USA) on the scanned image as
previously described.

3. Detection of Wnt and BMP reporter activity

For detection of Wnt reporter activity, MC3T3 cells grown in a 12-well culture dish were
transiently transfected with TOPFLASH or FOPFLASH, both luciferase reporter constructs
responsive to Wnt proteins (1.6 μg/well), and co-transfected with CMV-lacZ (0.16 μg/well) or
CMV-GFP (0.16 μg/well) overnight to determine transfection efficiency. The conditioned
medium from the EB culture groups was transferred to the MC3T3 culture, replacing the
transfection medium. Luciferase activity was detected using the Dual-Light reporter gene assay
system (Applied Biosystems). Relative luciferase activity was calculated by normalization of the
average luciferase activity to the -galactosidase activity at 24h post-transfection.

For detection of BMP-Smad reporter activity, the conditioned medium from the EB culture
groups was transferred to an MC3T3 cell line stably transfected with BMP-Smad binding
elements (GCCG12-luciferase; SBE-luc; a generous gift from Dr Baruch Frenkel), and the
luciferase activity was measured by the method described for the Wnt reporter.

80
4. RNA extraction, cDNA synthesis, and quantitative real-time PCR analysis

RNA was isolated from EB culture groups at 4h and 6h by using RNAqueous Kit (Ambion)
following the manufacturer’s instructions. Synthesis of first strand cDNA was achieved using the
RETROscript kit (Ambion). For cDNA template preparation, 2 μg of total RNA from each EB
sample was used in a 20- μL reaction. For quantitative real-time PCR analysis, a 25- μL reaction
was prepared for each sample. Included in this reaction volume was 1 μL of the resulting cDNA,
iQTM SYBR green supermix (Bio-Rad) containing dNTP and iTaq DNA polymerase, and the
appropriate primers (Table 1). The resulting threshold cycle (C
T
) value from each primer pair
was normalized with the C
T
value for Gapdh, which served as an internal control.

5. PCR array analysis

RNA samples from EB cultured in either mineralization medium or cultured in LRAP containing
medium were converted to cDNA by using RT First Strand Kit (SABiosciences). For each RNA
sample, 5 μg of total RNA was mixed with genomic DNA elimination buffer (SABiosciences) to
make a final volume of 10 μl. This mixture was then combined and mixed with RT cocktail (RT
buffer, primer and control mix, RT enzyme; SABiosciences) to a final volume of 20 μl. The
resulting first strand cDNA was diluted with ddH
2
O (91 μl of ddH
2
O to each 20 μl of cDNA
synthesis reaction). The cDNA reaction was mixed with RT
2
SYBR green qPCR master mix
(SABiosciences) and 25 μl of the experimental cocktail was transferred into 96-well RT Profiler
PCR Array for mouse Wnt signaling pathway (SABiosciences) containing appropriate primers for
genes associated with the Wnt signaling pathway and primers for house keeping genes (Table 3).

81
A two-step real-time PCR reaction was performed starting at 95°C (10 minutes) for 1 cycle and
followed by 95°C (15 s) and 60°C (1 min) for 40 cycles. Normalized threshold cycle data from
real-time instrument was calculated and interpreted using the PCR array data analysis web portal
(SABiosciences).

6. Statistical analysis

Mean and standard deviation of relative luciferase reading and C
T
value from each sample were
retrieved from 3 independent experiments. An independent t-test was used to assess the
difference between two experimental pairs. A P-value 0.05 was considered significant.

82
RESULTS

1. LRAP enhances -catenin protein accumulation and Wnt reporter activity during osteogenic
differentiation of mouse ES cells

In the canonical Wnt signaling pathway, Wnt ligands bind to Frizzled and LRP5/6 co-receptors
resulting in cytosolic -catenin protein stabilization and nuclear translocation of -catenin to bind
to TCF/LEF transcription proteins, leading to the activation of target genes (Miller, 2002). To
explore the involvement of canonical Wnt signaling pathway for the osteogenic effect of LRAP, I
measured the change in -catenin protein in EBs cultured in LRAP containing medium for
different periods of time. I found that the -catenin protein level increased after the EBs were
cultured in LRAP for 4 hours (Fig 13A, 13B), suggesting the activation of the canonical Wnt
signaling pathway as early as 4 hours after LRAP treatment. However, after 6 hours of EBs
cultured in LRAP containing medium, the -catenin protein levels were measured to be the same
when compared to the -catenin levels in EBs cultured in mineralization medium (Fig 13A, 13B).

83
Fig 13.

A.
4h 6h
______ ______
M L M L

-catenin

-actin

B.

LRAP increases -catenin steady-state protein level in mouse embryonic stem (ES) cells. (A)
Western immunoblot analysis shows increased steady-state protein levels of -catenin in mouse
ES cells cultured in the mineralization medium containing LRAP (L) for 4 hours when compared
to the expression level of -catenin the mouse ES cells cultured in mineralization medium (M)
alone. The expression of -catenin of mouse ES cells cultured with LRAP for 6 hours was
unchanged when compared to that for mouse ES cells cultured in mineralization medium alone.

84
(B) Quantification of band intensity from the Western immunoblot result in (A) shows 2.5-fold
increase in -catenin protein expression in mouse ES cells cultured in mineralization medium
containing LRAP (10 ng/mL) for 4 hours when compared to the -catenin protein expression in
EBs cultured in mineralization medium alone.

85
To further confirm the involvement of -catenin in transcriptional activation by Wnt released by
EBs grown in the presence of LRAP, the TCF/LEF reporter activity was measured using a
luciferase reporter that contains 16 TCF/LEF binding sites for -catenin (TOPFLASH). Another
construct containing the mutated TCF/LEF binding sites (FOPFLASH) upstream of the luciferase
reporter was used as a control. Both TOPFLASH and FOPFLASH plasmids were transiently
transfected into MC3T3 cells the day before the experiment. A transfection efficiency of
approximately 40%, a level sufficient for this reporter assay, was achieved. EBs at day-5 were
exposed to mineralization medium with the presence or absence of 10ng/mL LRAP and then
sampled after 1 hour, 4 hours or 6 hours. The conditioned medium from each time point was
separately transferred to the transiently transfected MC3T3 cells. Luciferase activity was
measured 24 hours after the EB-conditioned medium transfer. As shown in Fig 14, TCF/LEF
reporter activity was found to be increased after the cells were cultured in the medium
conditioned by EBs whose media contained LRAP for 1 hour, 4 hours or 6 hours, with the
greatest increase of Wnt reporter activity observed at 4 hours. Taken with the result from -
catenin protein quantification, these results gained with conditioned media, suggest that LRAP
activated the canonical Wnt signaling pathway during the first hours of the osteogenic
differentiation of EBs.

86
Fig 14.

LRAP increases Wnt reporter activity. The conditioned medium from mouse embryonic stem
(ES) cell culture in mineralization medium with or without LRAP for 1 hour, 4 hours or 6 hours
was transferred to MC3T3 cells transiently transfected with a luciferase reporter containing 16
TCF/LEF binding sites for -catenin (TOPFLASH) or a luciferase reporter containing the
mutated TCF/LEF binding sites for -catenin (FOPFLASH). The luciferase activity was
measured 24 hours after medium transfer. The result shows statistically significant increase in
TOPFLASH reporter activity in the culture medium derived from EBs treated with LRAP for all
time points (1 hour, 4 hours, and 6 hours) when compared to the TOPFLASH reporter activity in
the culture medium derived from EBs treated with mineralization medium alone. *P<0.01,
**P<0.05, indicating a statistically significant difference.

87
2. The expression of Wnt7b and Wnt10b is upregulated in LRAP-treated mouse ES cell

The observation that the expression of Wnt10b is upregulated during osteoblastogenesis (Bennett
et al., 2007; Kang et al., 2007), but downregulated during adipogenesis (Ross et al., 2000) along
with the observation that LRAP enhances osteogenesis but inhibits adipogenesis, suggest that the
LRAP signaling pathway involves the Wnt10b isoform. Furthermore, the reported function of
Wnt7b to serve as an osteogenic signal (Hu et al., 2005) to promote bone formation (Tu et al.,
2007) has suggested that the LRAP treatment might also involve the induced expression of
Wnt7b. Therefore, in the current study, I measured the expression of two Wnt isoforms targets,
Wnt10b and Wnt7b, during osteogenic differentiation of EBs in media containing LRAP. I
found a significant increase for Wnt10b and Wnt7b mRNAs in the LRAP-treated EBs at 6 hours
following LRAP exposure, but I found no significant increase of Wnt10b and Wnt7b mRNA
expression in LRAP treated-EBs at 4 hours post-LRAP exposure (Fig 15). These results
suggested that LRAP activates Wnt signaling pathway during the early stage of osteogenic
differentiation, but that such induction requires more than 6 hours to be observed at the level of
mRNA steady-state accumulation.

88
Fig 15.

Wnt7b expression

Wnt10b expression

89
The expression of Wnt7b and Wnt10b is upregulated in LRAP-treated mouse ES cell. Mouse
embryonic stem (ES) cells induced to osteogenic differentiation in the presence of LRAP for 6
hours expressed increased Wnt7b or Wnt10b gene expression when compared to mouse ES cells
not exposed to LRAP and cultured in mineralization medium alone. In contrast, no significant
change in Wnt7b and Wnt10b expression was observed during the 4-hour period of LRAP
treatment to EBs. *P<0.01, indicating a statistically significant difference.

90
3. Genes associated with Wnt signaling pathway are upregulated by LRAP

The Wnt gene family is constituted by 19 Wnt proteins, each sending a signal through the Wnt
receptor to regulate a variety of cell functions including cell growth and proliferation, control of
cell fate decision and regulation of the cell cycle (Kikuchi et al., 2007). Wnt signals are regulated
by both intracellular proteins, including transcription factors, protein kinases and phosphatases,
and extracellular proteins, including Wnt binding antagonists and regulators of the Wnt receptors
(Gordon and Nusse, 2006). To analyze the effect of LRAP on the components and regulators of
the complex Wnt signaling pathway, a commercially available Wnt gene array containing primers
for Wnt-associated genes (Table 3) was used to probe the expression of genes associated with
Wnt signaling pathway by quantitative real-time PCR. Mouse EBs were induced to osteogenic
differentiation as previously described in chapter 1, with EBs at day-5 were cultured in
mineralization medium containing LRAP, or control EBs were cultured in mineralization medium
in the absence of LRAP. After EB culture for the designated time, RNA samples were extracted
and used as templates for cDNA synthesis and quantitative RT-PCR, following the array
manufacturer’s instructions. The relative expression levels for each target gene from the LRAP
group or the control group were plotted against each other, in a scatter plot (Fig 16A, 17A) and in
a clustergram (Fig 16B, 17B).

After LRAP was added to the EBs at day-5 for 1 hour, the increased expression of several genes
associated with Wnt signal was observed. These upregulated targets included genes encoding
Wnt proteins themselves, such as Wnt3, Wnt3a, Wnt6, Wnt7b and Wnt8b; genes encoding the
secreted frizzled related proteins such as Sfrp1 and Sfrp4; and genes controlling the regulation of
growth and proliferation such as Axin1 (Fig 16A, 16B).

91
In contrast, genes encoding Frizzled homolog 2 (Fzd2) and solute carrier family 9
(sodium/hydrogen exchanger), member 3 regulator 1 (Slc9a3r1) were observed to be
downregulated by exposure to LRAP (Fig 16A, 16B).

92
Fig 16.

A.

93
Fig 16, Continued.

B.

Changes in Wnt signal-associated gene expression in mouse embryoid bodies (EBs) treated with
LRAP for 1 hour. (A) A scattered plot shows the expression of genes in LRAP-treated EBs for 1
hour, normalized to the expression of genes in EBs cultured in mineralization medium alone.
Marked increase in gene expression was observed in the expression of Wnt6 (1.92-fold), Wnt7b
(1.45-fold), Wnt8b (2.53-fold), Sfrp4 (2.71-fold), and Wnt3 (1.56-fold). On the other hand, the
expression of two genes, including Fzd2 (0.48-fold) and Slc9a3r1 (0.48-fold), were observed to
be downregulated. (B) A clustergram showing the magnitude of gene expression in LRAP-
treated EBs at 1 hour (Min+L1h), comparing to the gene expression in EBs cultured in
mineralization medium (Min 1h) and EBs cultured in the control medium (Control 1h). The
expression of Wnt3, Wnt3a, Wnt6, Wnt7b, Wnt8b, and Sfrp4 was observed to be upregulated,
whereas the expression of Fzd2, Slc9a3r1 and Sfrp1 was observed to be downregulated.

94
When LRAP was added to the EBs at day-5 for 4 hours, the expression levels of Wnt3a, Wnt6,
Wnt8b and Sfrp4 were observed to be upregulated, whereas the expression levels of Wnt3, and
Wnt7b was observed to be downregulated (Fig 17A, 17B). On the other hand, the expression of
the Wnt antagonists including Dkk1, Sfrp1, and Sfrp2 was observed to be downregulated after 4
hours of LRAP treatment of day-5 EBs. In addition, the expression level for Fzd2 and Slc9a3r1
were observed to also be downregulated (Fig 17A, 17B).

95
Fig 17.

A.

96
Fig 17, Continued.

B.

Changes in Wnt signal-associated gene expression in mouse embryoid bodies (EBs) treated with
LRAP for 4 hours. (A) A scattered plot for the expression of selected genes in LRAP-treated EBs
for 4 hours normalized to the expression of genes in EBs cultured in mineralization medium
alone. Marked increase in gene expression was observed in Wnt3a (1.15-fold), Srp4 (1.23-fold)
and Wnt8b (1.52-fold). In contrast, the expression of Sfrp1, Sfrp2, Fzd2, Slc9a3r1, Wnt7b,
Wnt3, and Dkk1 in LRAP-treated EBs for 4 hours was observed to be downregulated. (B) A
clustergram showing the magnitude of gene expression in LRAP-treated EBs for 4 hours (LRAP
4h), compared to the gene expression in EBs cultured in mineralization medium (Min 4h). The
expression of Wnt3a, Wnt6, Wnt8b, and Sfrp4 was observed to be upregulated, whereas the
expression of Dkk1, Fzd2, Sfrp1, Sfrp2, Slc9a3r1, Wnt3 and Wnt7b was observed to be
decreased.

97
In addition to the Wnt gene array analyses of genes associated with LRAP treatment inducing the
Wnt signaling pathway, I also performed independent quantitative RT-PCR on identically treated
EBs using select primers for Wnt8b, Sfrp4, Dkk1 and Dkk2 (Table1). In accordance with the
results from Wnt gene array analyses, the expression for Wnt8b and Sfpr4 in EBs cultured in
LRAP for 4 hours was increased, whereas the expression for Dkk1 and Dkk2 was decreased (Fig
18). The independent result obtained from quantitative RT-PCR supports the validity of the data
obtained from the Wnt gene array analysis.

98
Fig 18.

Quantitative real-time PCR analysis for mouse embryoid bodies (EBs) treated with LRAP for 4
hours. A significant increase in Wnt8b and Sfrp4 expression, and a significant decrease in Dkk1
and Dkk2 expression was observed in LRAP-treated EBs for 4 hours. *P<0.05, indicating a
statistically significant difference.

99
4. LRAP rescued the decreased Wnt reporter activity in the presence of Wnt antagonist

In the extracellular matrix, Wnt signals are known to be modulated by a number of secreted Wnt
antagonists that react with Wnt proteins or Wnt receptors and thereby alter the Wnt signaling
activity (Kawano and Kypta, 2003). Some Wnt antagonists, including secreted Frizzled-related
proteins (sFRPs), Wnt inhibitory factor-1 (WIF-1) and Cerberus antagonize Wnt activity by direct
interaction with Wnt proteins (Hsieh et al., 1999) (Piccolo et al., 1999) (Moon et al., 1997).
Others Wnt antagonists, including the members of the Dickkopf family and SOST/Sclerostin,
antagonize Wnt activity by interacting with the Wnt receptors, LRP5 and LRP6 (Itasaki et al.,
2003; Nusse, 2001).

To analyze the effect of LRAP induction of the Wnt signaling pathway, I used the Wnt
antagonist, secreted frizzled-related protein-1 (sFRP-1) as a Wnt binding competitor to inhibit
Wnt signals in the current experiment design. The sFRP-1 belongs to the family of Frizzled-
related proteins (FRP), which antagonize both canonical and non-canonical Wnt signaling
pathways by acting as decoy receptors for Wnt protein (Bodine et al., 2005). In mouse calvarial
cells, s-FRP-1 has been reported to block osteoblastic lineage commitment by blocking Wnt
signals (Zhou 2007).

EBs from day-5 were cultured in the mineralization medium in the presence or absence of LRAP
for 24 hours with or without the soluble sFRP-1 (20ng/mL). Medium conditioned by the EBs
was taken from the culture and transferred to the MC3T3 transiently transfected with Wnt
reporter TOPFLASH or FOPFLASH construct. Luciferase activity by the MC3T3 cells was
measured 24 hours after the medium transfer.

100
The results suggested that during the osteogenic differentiation of EBs at day-5, sFRP-1
decreased Wnt reporter activity in the EB cultured in mineralization medium containing LRAP
(Fig 19A), indicating that sFRP-1 inhibited the Wnt signal observed in LRAP-treated EBs
without sFRP-1. However, the antagonism mediated by sFRP-1 was overcome by the addition
of a 5-fold increase in LRAP concentration (50 ng/mL) as evidence by the resumption of Wnt
reporter activity (Fig 19A), suggesting that excess LRAP can induce Wnt protein from the EBs
even in the presence of the Wnt antagonist. LiCl and KCl was used as a positive control and
negative control for Wnt reporter activity, respectively (Fig 19B).

101
Fig 19.

A.

B.

102
LRAP rescued the diminished Wnt reporter activity in when challenged by the addition of
secreted Wnt antagonist.
(A) Analysis for Wnt reporter activity in the presence of LRAP and an exogenously added Wnt
antagonist. Embryoid bodies (EBs) at day-5 were cultured in the mineralization medium, with or
without LRAP (10ng/mL), in the presence of exogenous secreted frizzled related protein-1
(sFRP-1) (20 ng/mL), for 24 hours. The medium conditioned by EBs were transferred to MC3T3
cells transiently transfected with a luciferase reporter containing 16 TCF/LEF binding sites for -
catenin (TOPFLASH) or with a luciferase reporter containing the mutated TCF/LEF binding sites
for -catenin (FOPFLASH). Luciferase signal was measured at 24 hours post medium transfer.
The presence of sFRP-1 resulted in a significant decrease (approximately 30%) for the Wnt
reporter activity in the presence of LRAP (10ng/mL) when compared to the Wnt reporter activity
measured in EBs cultured in LRAP containing medium in the absence of sFRP-1. In the presence
of sFRP-1, LRAP at 50 ng/mL concentration in EB culture resulted in a 2-fold increase in Wnt
reporter activity when compared to Wnt reporter activity measured in EBs cultured in LRAP (10
ng/mL), in the presence of sFRP-1.
(B) Luciferase reporter activity for MC3T3 cells transiently transfected with TOPFLASH
construct was measured after 24-hours of the MCT3T cell exposure to alpha-MEM (growth
medium), alpha-MEM containing 40 ng/mL lithium chloride (LiCl), or alpha-MEM containing 40
ng/mL potassium chloride (KCl).

103
5. LRAP moderately activates BMP signaling pathway at the late stage of osteogenic
differentiation

Previous studies have suggested that the amelogenin component in Emdogain possess BMP-like
activity that serves to activate BMP signaling pathways (Kawase et al., 2001; Lee et al., 2008;
Suzuki et al., 2005; Takayama et al., 2005). For example, during the osteoblastic differentiation
of C2C12 cells, Emdogain was found to increase Runx2 expression, as well as the
phosphorylation of Smad1 protein, and these effects were diminished in the presence of Noggin,
an inhibitor of the BMP signaling pathway (Takayama et al., 2005). In addition, the downstream
effector of BMP signaling pathway, Smad4, has been reported to be a co-activator working with
the transcription factor Lef1, to activate the expression of homeobox trancription factor Msx2
(Hussein et al., 2003), suggesting cross-talk between the BMP and the Wnt signaling pathways.
Based on these observations, I hypothesize that LRAP may also signal cell fate determination
through the activation of the BMP-signaling pathway during osteogenic differentiation of mouse
ES cells.

Operating in the BMP-2 signaling pathway, Smad1 and Smad5 have been shown to be
downstream components that induce osteogenic gene expression (Yamamoto et al., 1997).
Phosphorylation of the intracellular signaling component, Smad1, 5, and 8, is required for signal
transduction of TGF- and BMPs (Kretzschmar et al., 1997). To determine the activation of
BMP signaling pathway, I measured the steady-state protein level of phosphorylated Smad1,
Smad5 and Smad8 (pSmad1/5/8) in osteogenic-induced EBs at day-5 and day-9, in the presence
or absence of LRAP (10ng/mL).

104
The results for EBs at day-5, when cultured in LRAP containing medium for 24 hours, show that
the protein signal for Smad1/5/8 remained relatively the same when compared to EB cultured in
mineralization medium without LRAP (Fig 20A, 20B). The continuous exposure of LRAP to the
EBs until they reached day-9 of culture, resulted in moderately increased levels of pSmad1/5/8
protein signal being measured (EB day-9) (Fig 20A, 20B), suggesting a possible activation of
BMP signaling pathway at the later stage (day-9) of osteogenic differentiation.

The activation of the BMP-Smad signaling pathway was independently assessed by measuring
BMP reporter activity using BMP-Smad reporter construct containing Smad binding element tied
to a luciferase reporter (GCCG12-luciferase; SBE-luc; a gift from Dr Baruch Frenkel). The
conditioned medium, derived from osteogenic-induced EBs were cultured in the presence of
LRAP (10ng/mL) from day-5 to day-10, was transferred to the MC3T3 cells stably transfected
with SBE-luc and the luciferase activity of MC3T3 cells cultured in conditioned medium was
assessed after 24 hours.

Here I found the BMP reporter activity was not activated by LRAP during the culture period of
EB at day-5 to day-9. At day-10, EB cultured in the presence of LRAP showed moderately
increased BMP reporter activity when compared to the EB cultured in the mineralization medium
in the absence of LRAP (Fig 20C). The results from pSmad1/5/8 protein detection and the BMP
reporter activity assay suggests the possible involvement of the BMP signaling pathway when it
is induced by LRAP treatment during the later stage of EB differentiation (day-9 to day-10)
although such BMP expression may be primed by LRAP induced Wnt signals at a earlier stage of
EB development (day-5).

105
Fig 20.

A.

B.

106
Fig 20, Continued.

C.

The effect of LRAP on the BMP signaling pathway. (A) Western immunoblot result for
phosphorylated Smad1/5/8 (pSmad1/5/8) protein expression in embryoid bodies (EBs) cultured in
the LRAP-containing medium at day-5 (d5) and day-9 (d9) shows increased pSmad1/5/8 protein
expression in the presence of LRAP (10ng/mL) at day-9 when compared to the pSmad1/5/8
protein expression in EBs grown in mineralization medium alone. (B) Quantification of
pSmad1/5/8 protein expression shows a moderate increase (25%) in pSmad1/5/8 protein level in
the presence of LRAP at day-9. (C) Luciferase reporter analysis for BMP reporter activity of EB
cultured in mineralization medium, with or without LRAP, shows a moderate increase (30%) in
BMP reporter activity in the presence of LRAP at day-10. Recombinant human BMP-2 (rhBMP-
2) was used as a positive control for the measurement of BMP reporter activity.

107
6. Increased LRAP concentration does not increase mineral formation

My previous data from the Wnt antagonist assay has shown that increasing the concentration of
LRAP (50 ng/mL) resulted in enhanced Wnt reporter activity sufficient to overcome the
inhibitory effect of sFRP-1. This conclusion was reached based on the observations of the early
effect of LRAP on the Wnt signaling on EBs at day-5. To address whether an enhancement in
the terminal phenotype results from increased LRAP concentration, LRAP at the high
concentration (50ng/mL) was left in the osteogenic medium throughout the course of osteogenic
differentiation. The MC3T3 cells were induced to osteogenic differentiation in the presence of
LRAP (10ng/mL, 50 ng/mL) or in the absence of LRAP. Alizarin red staining and relative
calcium concentration were measured at the end of the culture period, 21 days later. Alizarin red
S staining of the extracellular matrix and measurement of the calcium deposition into the matrix
from the treated cells showed that mineral formation in the presence of low concentration of
LRAP (10ng/mL) was greater (2-fold) than that observed for the mineralization control.
However, no enhancement was observed in mineral formation in the matrix created by MC3T3
cells when LRAP concentration was increased to 50 ng/mL (Fig 21A, 21B). The level of calcium
deposition formed in the presence of increased LRAP (50ng/mL) was observed to be higher than
that for the control group, but lower than that observed for the cells grown in LRAP at the
concentration of 10ng/mL. These results suggested that increasing concentration of LRAP had
no additive effect on the development of terminal phenotype during osteogenic differentiation for
MC3T3 cells. It also suggests that induction by LRAP is at a maximal response, indicating that
the pathway is saturated.

108
Fig 21.

A.

B.

109
The effect of LRAP concentration on mineral formation in MC3T3 cells. MC3T3 cells were
induced to osteogenic differentiation in two different concentrations of LRAP, 10 ng/mL and 50
ng/ml, and assessed for mineral formation. (A) Alizarin red S staining analysis for mineral
formation shows increased mineral formation in MC3T3 cells cultured in mineralization medium
in the presence of LRAP at both concentrations, 10 ng/mL and 50 ng/mL, when compared to the
MC3T3 cells cultured in mineralization medium alone. However, LRAP at 50 ng/mL
concentration did not enhance mineral formation when compared to the mineral formation
observed in MC3T3 cultured in LRAP at 10ng/mL concentration. (B) Quantification of calcium
deposition in extracellular matrix made by MC3T3 cells cultured as described above shows a 2-
fold increase in calcium formation in MC3T3 cultured in LRAP 10 ng/mL, when compared to
MC3T3 cultured in mineralization medium alone. In contrast, only a 0.5-fold increase in
calcium formation was observed in MC3T3 cultured in LRAP 50 ng/mL, when compared to
MC3T3 cultured in mineralization medium alone.

110
7. The effect of exogenous canonical Wnt and scrambled LRAP peptide on osteogenic induced
ES cells

The observations that LRAP activates the signaling components in the canonical Wnt pathway,
and the close resemblance between the role of Wnt proteins and LRAP to enhance
osteoblastogenesis and inhibit adipogenesis suggested that LRAP may behave in the same way as
did exogenous Wnt proteins to activate Wnt signal and mediate osteogenic cell differentiation. In
the current experiment, I hypothesized that exogenous Wnt mimics the effect of LRAP in
osteogenic induction of mouse ES cells. A scrambled LRAP peptide was used as a control to
eliminate any non-specific signaling function(s) induced by bioactive LRAP.

To test exogenous Wnt, I used EBs cultured for 5 days, as before. Osteogenic induction was
performed by adding mineralization medium with or without LRAP (10 ng/mL), Wnt3a
(100ng/mL) or scrambled LRAP peptide (10 ng/mL). The medium was left overnight and
replaced with mineralization medium for all samples on the next day. The EB culture was
maintained in mineralization medium throughout the remainder of the osteogenic differentiation
period. Samples were collected after 10 or 20 days of culture for assessment of osteogenic
markers and mineral formation in the extracellular matrix formed by the EBs.

Gene expression analysis of EBs at day-10 suggested that exogenous Wnt3a increased Osx
expression to levels comparable to Osx expression previously observed for EBs induced by
LRAP alone, but failed to mimic the increased BSP expression observed in LRAP-treated EBs
(Fig 22A). As expected, the scrambled LRAP peptide failed to modulate the expression of BSP
and Osx in these EB cultures. Noticeably, the fold increase for Osx and BSP mRNAs was less
dramatic when compared to data obtained in other experiment and as shown in chapter 1.

111
Differences between the exogenous Wnt experiment described here and previous experiment
described in chapter 1 may be linked to the limited LRAP exposure time (24 hours for EB day-5),
which may explain the modest fold change in gene expression for exogenous Wnt. This data also
suggests that LRAP treatment might be required throughout the course of osteogenic
differentiation to produce the dramatic fold increase in both Osx and BSP expression.

Analyses of the terminal phenotypes of EBs after 20 days in culture suggested that Wnt3a
treatment alone could not replicate the enhanced mineral deposition and calcium accumulation in
the extracellular matrix produced by the EBs treated with LRAP. Instead, the level of
mineralization and calcium accumulation was decreased, approximately 4-fold, when compared
to control EBs cultured in mineralization medium (Fig 22C). EBs cultured in the scrambled
LRAP peptide containing medium for the same culture period also exhibited no enhanced mineral
deposition or calcium accumulation (Fig 22C). Analyses for gene expression of EBs at day-20
cultured in the presence of the scrambled peptide or Wnt3a showed that the Osx expression was
significantly lower than the Osx expression of EB at day-20 cultured in medium containing
LRAP (Fig 22B).

112
Fig 22.

A.

B.

113
Fig 22, Continued.

C.

D.

114
The effect of exogenous Wnt and scrambled LRAP peptide on osteogenic-induced mouse ES
cells. Mouse embryoid bodies (EBs) were induced to osteogenic differentiation and maintained
in mineralization medium, or in mineralization medium containing LRAP (10 ng/mL), or in
medium containing Wnt3a (100ng/mL) or in medium containing scrambled LRAP peptide (10
ng/mL) for 24 hours. After 24 hours, the medium for all culture groups was replaced with
mineralization medium alone and used throughout the course of the experiment. (A) Quantitative
real-time PCR analysis for EBs at day-10 shows the expression of Osx and BSP. EBs cultured in
mineralization medium containing Wnt3a (M+Wnt3a) expressed decreased BSP and Osx
expression when compared to the expression of both genes in EBs cultured with LRAP (M+L).
EB cultured in mineralization medium containing scrambled LRAP peptide (Lsc) also failed to
upregulate the expression of BSP and Osx, when compared to the expression of both genes in
EBs cultured with LRAP (M+L). *P<0.01 compared to M+L group.
(B) Quantitative real-time PCR analysis for EBs at day-20 shows the expression of Osx was
decreased in Wnt3a-treated EBs (M+Wnt3a) as well as in scrambled LRAP peptide-treated EBs
(M+Lsc), when compared to the Osx expression in LRAP-treated EBs (M+L). **P<0.05
compared to M+L group.
(C) Mineral and calcium concentration analysis for EBs at day-20. Alizarin red S staining shows
a decrease in mineral formation (red stained areas) in Wnt3a-treated EBs and in scrambled LRAP
peptide-treated EBs, compared to the mineral formation in LRAP-treated EBs.
(D) Analysis for calcium concentration shows the decreased amount of calcium formed by
Wnt3a-treated EBs and scrambled LRAP peptide-treated EBs, when compared to the amount of
calcium formed by LRAP-treated EBs.

115
DISCUSSION

Previous studies have suggested a function for the amelogenin isoforms of LRAP to serve as a
signaling molecule that is capable of inducing osteogenic differentiation of cells in different
cellular contexts (Boabaid et al., 2004; Hammarstrom, 1997; Lacerda-Pinheiro et al., 2006b; Veis
et al., 2000; Warotayanont et al., 2007). However, little evidence has accumulated that addresses
a mechanistic function for LRAP for such induced differentiation (Boabaid et al., 2004;
Tompkins et al., 2006). In this chapter, I have provided data to identify and partially explain the
possible mechanism(s) for the signaling function of LRAP. First, I have shown that LRAP
activates canonical Wnt signaling pathway as evidenced by the increased -catenin protein level,
the increased Wnt reporter activity, and the ability of LRAP to rescue the Wnt inhibitory effect of
sFRP-1. In addition, the expression of Wnt10b, Wnt7b and several other genes associated with
Wnt signaling, including Wnt3, Wnt3a, Wnt6, Wnt7b, and Wnt8b, were upregulated following
LRAP treatment, whereas genes associated with Wnt antagonistic activity, including Dkk1,
Dkk2, Sfrp1 and Sfrp2 were downregulated following LRAP treatment. Second, I have shown
that BMP signaling pathway is modestly activated during the late stage of osteogenic
differentiation. The activation of the BMP pathway is based on the observation that following
LRAP treatment, pSmad1/5/8 protein level is increased in EB at day-9, and the BMP-Smad
reporter activity is increased for EB at day-10 of culture. Third, I have shown that increasing the
concentration of LRAP in the osteogenic medium does not have an additive effect on the terminal
cell phenotype, such as mineral deposition in MC3T3 cells. Fourth, LRAP specificity during
osteogenesis has been demonstrated by the failure of a scrambled LRAP peptide to mimic the
osteogenic terminal phenotypes that are observed in the presence of canonical LRAP. And fifth, I
have demonstrated that an exogenous Wnt3a alone lacks the capacity to induce osteogenic
phenotypes that are observed in the presence of LRAP.

116

This conclusion is made based on the observation that Wnt3a alone fails to fully induce BSP and
Osx expression and mineral formation of osteogenic induced EBs at day-10 and day-20 to the
extent that LRAP does induce such change. Taken together, the current experimental results
have provided additional information to explain the mechanistic function(s) for LRAP and its
signaling property during cell differentiation to the osteogenic lineage.

1. LRAP activates canonical Wnt signaling pathway to induce osteoblastogenesis

The function of the canonical Wnt signaling pathway during osteoblast differentiation has been
extensively studied (Wodarz and Nusse, 1998). Inactivation of the downstream effector of the
canonical Wnt, -catenin, in mouse calvaria osteoprogenitor cells inhibits osteoblast
differentiation (Hill et al., 2005). On the other hand, activation of -catenin/Tcf-1 complex
results in increased Runx2 expression and osteoblast phenotypes (Gaur et al., 2005). Mutation of
the Wnt co-receptor low-density lipoprotein receptor-related protein 5 (LRP5) results in defects in
eye development and reduced bone mass in humans (Gong et al., 2001), and defects in osteoblast
proliferation in mice (Kato et al., 2002).

During adipogenesis, Wnt signal inhibits the expression of the adipogenic transcription factors,
C/EBP and PPAR , thereby reducing cell differentiation to the adipogenic lineage (Ross et al.,
2000). Two Wnt proteins, including Wnt10b and Wnt1, have been reported to inhibit adipocyte
differentiation of MC3T3 preosteoblastic cell line (Ross et al., 2000). The similarity between the
function of Wnt protein and LRAP to induce osteoblastogenesis and inhibit adipogenesis has
suggested that LRAP stimulates Wnt release and activation of the canonical pathway.

117

Hence, the strategy for the current study is focused on finding the relationship between the
activation of canonical Wnt pathway by the signaling effect of LRAP. This idea has been tested
using the osteogenic-induced mouse ES cells in the presence of LRAP. The observations that -
catenin protein level was increased and the Wnt reporter was increased following LRAP
treatment for 1 hours and 4 hours suggests that the canonical Wnt pathway is activated during the
early events leading to osteogenic differentiation.

Wnt signal propagation can be modulated by its secreted antagonists, which act to interfere with
the binding of Wnt proteins to the Wnt co-receptors and thus alter Wnt signal transduction (Jones
and Jomary, 2002). Like Cerberus (CER) and Wnt inhibitory factor-1 (WIF-1), secreted
Frizzled-related proteins (sFRPs) antagonizes Wnt activity by directly binding to Wnt through its
cysteine-rich domain, and preventing Wnt from binding to its receptor, Frizzled, resulting in the
inactivation of both the canonical and the noncanonical Wnt pathways (Kawano and Kypta,
2003). In the current study, I show that exogenous sFRP-1 successfully inhibits the increasing
Wnt reporter activity induced by ES cells when LRAP (10 ng/mL) is present in the medium,
suggesting that physical inhibition of the Wnt signaling pathway by sFRP-1 diminishes the Wnt
signal induced by LRAP treatment. The capability of LRAP at the highest concentration (50
ng/mL) to rescue the inhibitory effect of sFRP-1 on Wnt reporter activity suggests the possibility
that excess LRAP can also increase Wnt signal to the level that overcomes the inhibition from the
concentration of sFRP-1 used experimentally. Of course, there may be other pathway(s) that
involve the activation of Tcf/LEF reporter activity when Wnt signal is attenuated. One example
is based on the finding that co-activators of Wnt and BMP signaling pathways have been shown
to be required for the expression of Msx2 during cell fate determination of stem cells in the
ectoderm (Hussein et al., 2003).

118

Therefore, it may be possible that, in the context of mouse ES cells used in the current
experiments, there are other co-activators induced by LRAP that serve to independently stimulate
Wnt reporter activity in the presence of a soluble Wnt antagonist.

2. LRAP alters the expression of genes associated with canonical and non-canonical Wnt
signaling pathways

In addition to the enhanced -catenin protein levels and Tcf/LEF reporter activity, I have also
observed changes in the expression of genes associated with Wnt signaling pathway. Within the
first few hours following LRAP treatment of osteogenic-induced cells from EBs, the upregulation
of genes encoding Wnt proteins such as Wnt3, Wnt3a, Wnt6, Wnt7b and Wnt8b has been
observed. In vertebrates, the 19 Wnt proteins have been broadly classified based in their
biological and functional activity to induce axis formation in embryo or transform mouse
mammary epithelial cell line (Kuhl et al., 2000b; Wong et al., 1994). The family of Wnt-1,
including Wnt -1, -3a, -7, -7b, -8 and -8b, is capable of inducing a secondary axis in embryo and
transforming mammalian epithelial cells (Du et al., 1995; Wong et al., 1994). On the other hand,
the family of Wnt5a, including Wnt-4, -5a, -6 and -11, cannot induce axis formation and have no
effect on mammary epithelial transformation (Du et al., 1995; Wong et al., 1994). It has been
suggested that this second class of Wnts transmit their signal through different Wnt signaling
pathways. While members of the Wnt1 class are known to be activators of the canonical Wnt
pathway, the members of the Wnt5a class are known to act through the non-canonical Wnt
pathway (Kikuchi et al., 2007).

119
Other lines of evidence suggest that despite the classification of Wnt proteins, each Wnt protein is
not necessarily designated to function exclusively through the canonical or non-canonical Wnt
pathways. For example, the canonical Wnt3a has been reported to activate Rho and Rho-kinase,
which are the downstream effectors in Wnt planar cell polarity (PCP) pathway (Kishida et al.,
2004). On the contrary, a non-canonical Wnt, namely Wnt5a, has been shown to bind to Fzd5
receptor and to activate -catenin/canonical Wnt pathway (He et al., 1997). Furthermore, Wnt7b
can induce osteoblast differentiation via the PKC delta-mediated pathway (Tu et al., 2007).
Therefore, in my current experiment, although the upregulation of Wnts, Wnt3, Wnt3a, Wnt7b
and Wnt8b may reflect an increase in the activity of Wnt proteins implicated in the activation of
canonical Wnt pathway, it is still possible that the non-canonical Wnt pathways are also activated
upon the addition of LRAP to the EB culture.

In addition to activating the canonical or non-canonical Wnt pathways, additional insights may be
gained from analysis of the upregulation of specific Wnt genes. For example, the observed
upregulation of Wnt10b by LRAP treatment supports the previous observations for the functional
similarity between the function of Wnt10b and LRAP to inhibit adipocyte differentiation and
promote osteoblast differentiation. The upregulation of Wnt10b and Wnt7b expression may
reflect an increase in osteoblastogenesis as a result of LRAP treatment to ES cells based on
previous studies suggesting that for transgenic mice characterized by disruption of glucocorticoid
signaling in osteoblastic lineage, it is observed that the expression of Wnt10b and Wnt7b is
downregulated, and the osteoblast differentiation is inhibited (Zhou et al., 2008). In addition,
targeted disruption of Wnt3 expression resulted not only in the defects in axis formation and
gastrulation (Liu et al., 1999), but also created defects in hair growth and hair structure
(Kishimoto et al., 2000).

120
Therefore, the observed increase in Wnt3 expression along with the Wnt target gene, Axin1
(Zhou et al., 2008), may suggest that increased cell growth and proliferation is promoted by
LRAP treatment.
Several genes are downregulated in their expression levels during LRAP treatment.
Downregulation of Wnt antagonists, including Sfrp1, Sfrp2, Dkk1 and Dkk2, may explain the
increased activation of the Wnt pathway with these soluble inhibitors diminished resulting in
decreased inhibition of Wnt pathway as an effect of LRAP treatment. The sFRPs proteins are
extracellular proteins with a physical structure similar to Frizzled but lacking a transmembrane
domain for signal propagation, but the sFRP proteins retain the ability to bind to Wnts and alter
their signaling activity (Kawano and Kypta, 2003). Dkk1, on the other hand, antagonizes Wnt
activity by directly interacting with the Wnt receptor LRP5/6 (Bafico et al., 2001), and the single-
pass transmembrane proteins Kremen1 (Krm1) and Kremen2 (Krm2) (Mao et al., 2002).
Therefore, the decreased expression of Dkk1 may suggest that diminished Wnt signaling activity
exists after LRAP stimulation. Decreased expression of Dkk1 may also explain in part how
LRAP increases bone mass in vivo, based on the finding that the deletion of a single allele of
Dkk1 leads to increased bone mass in mice (Morvan et al., 2006).

In contrast to Sfrp1 and Sfrp2 expression, the expression of Sfrp4 is markedly upregulated
following LRAP treatment. This observation may be explained by the possible feedback
mechanism in response to the increase Wnt signal mediated by LRAP. Another possible
explanation for the increase in Sfrp4 expression is the variation of sFRP function in different
cellular context. For example, Sfrp4 has been shown to inhibit osteoblast proliferation and
decrease bone formation in mice (Nakanishi et al., 2008).

121
Although sFRPs are traditionally known as Wnt antagonists, recent findings suggest the sFRPs
have other functions during the development and disease progression (Rubin et al., 2006) as well
as the regulation of other signaling pathways (Yabe et al., 2003). When there is a low
concentration of sFRP, it has been reported that sFRP interact with Wg (Wingless) to stabilize
Arm (armadillo) in Drosophila, and hence to activate the canonical Wnt pathway (Uren et al.,
2000). This finding has been explained by the difference in the binding sites on sFRPs and their
different affinity to Wg. It has been suggested that a high affinity in sFRP sites leads to Wg
activation instead of inhibition (Kawano and Kypta, 2003).

In the experiment described in this chapter, the expression of the solute carrier family 9
(sodium/hydrogen exchanger), member 3 regulator 1 (Slc9a3r1) was found to be markedly
decreased following LRAP treatment of EBs. This finding may be explained by the implication
of Slc9a3r1 functioning at the late stage of bone formation, as opposed to the immediate early
stage of the differentiation of ES cells. The solute carrier family 9 (sodium/hydrogen
exchanger), member 3 regulator 1 (Slc9a3r1), also known as sodium/proton exchanger regulatory
factor type 1 (NHERF1), or known as ezrin binding protein 50 (EBP50), is a membrane protein
that has a role in the organization of the actin cytoskeleton (Bretscher et al., 2000). Slc9a3r1
interacts with a number of intracellular and extracellular signaling molecules including the Wnt
effector protein, -catenin (Shibata et al., 2003), as well as the receptor for parathyroid hormone,
type 1 PTH/PTHrP receptor (PTH1R) (Sneddon et al., 2003). The function of Slc9a3r1 has been
reported to implicate in the control of osteoblast maturation, bone mineralization and bone
density (Shenolikar et al., 2002). In NHERF1
-/-
mice, severe bone demineralization with multiple
fractures is observed along with decreased blood phosphate level and increased urinary phosphate
excretion (Prie et al., 2004).

122
During osteoblast differentiation induced by a histone deacetylase inhibitor, Slc9a3r1 is observed
to be upregulated (Schroeder et al., 2007). However, overexpression of Slc9a3r1 alone in
osteoblasts is not sufficient to promote osteoblast maturation (Schroeder et al., 2007). Taken
together, these lines of evidence suggest that Slc9a3r1 may play a role during osteoblast
maturation and mineralization. In my current experiment in which the RNA is extracted as early
as 1 hour or 4 hours after LRAP treatment of EBs, the expression of Slc9a3r1 may not yet be
upregulated at this time but may be upregulated when the culture time for osteogenic
differentiation is extended. The decreased expression of Slc9a3r1 in my current experiment may
be explained by negative feedback regulation as a result of increased -catenin-mediated Wnt
signal transduction upon the addition of LRAP. Such an interpretation is supported by the
observation that Slc9a3r1 activates the Wnt signaling pathway by interacting with -catenin to
promote Tcf/LEF transactivation in hepatocellular carcinoma cell lines (Shibata et al., 2003).

Frizzled homolog 2 (Fzd2), a 7-transmembrane receptor for the Wnt signal transduction cascade,
has been reported to play a role during signal transduction of the non-canonical Wnt pathways,
including G-protein signaling and activation of calcium/calmodulin-dependent protein kinase II
(CamKII) in Wnt/Ca
2+
pathway (Ahumada et al., 2002; Kuhl et al., 2000a). In addition, for lung
epithelial cells, inhibition of the Fzd2 expression leads to an increase in TOPFLASH Wnt
reporter activity, suggesting that Fzd2 antagonizes the canonical Wnt pathway (Zhang et al.,
2008). The results from these previous studies suggest that the reduced expression of Fzd2
observed in the current experiment facilitates the increased Wnt reporter activity and Wnt
signaling transduction as a result of LRAP treatment to the EBs at 1-hour and 4-hour.

123
3. Exogenous Wnt3a fails to mimic the effect of LRAP on osteogenic differentiation of ES cells

Despite the above findings that support the conclusion that LRAP activates the canonical Wnt
pathway, it remains unclear how LRAP transmits the signaling which results in Wnt release and
Wnt pathway activation. A cell surface receptor for LRAP called LAMP-1 has been identified in
C2C12 mouse fetal myoblasts (Tompkins et al., 2006), and upregulation of LAMP-1 gene
expression was found associated with human enamel organ epithelial cell differentiation (Le et
al., 2007). Even though existing evidence suggests an association between endocytosis and cell
signaling for LAMP-1 (Donaldson et al., 2008; Liberali et al., 2008), a detailed understanding of
the association between LAMP-1 and the activation of Wnt signaling is still missing. Because
the accumulated data has directed my attention to the mechanism(s) of LRAP stimulation through
the activation of the canonical Wnt pathway, in the current study, Wnt3a, the prototype of the
canonical Wnt, is used as an exogenous Wnt signal to treat the ES cells during their
differentiation in the osteogenic lineage. I hypothesize that if the terminal osteogenic phenotype
induced by Wnt3a resembles the osteogenic phenotypes induced by LRAP, the molecular
property of LRAP induction may be comparable to that achieved solely by Wnt3a signals.

Analyses of the terminal phenotypes following Wnt3a treatment suggest that Wnt3a alone failed
to mimic the terminal phenotypes observed when the same cells are treated with LRAP alone, as
shown by the low level of Osx and BSP expression at day-10 and at day-20, and approximately 3-
fold decrease in mineral and calcium formation. Since Wnt3a sends its signal through the
canonical Wnt signaling pathway (Qiang et al., 2003; Qiang et al., 2005), the findings from my
experimental strategy suggest that other pathways, including the non-canonical Wnt pathway or

124
BMP signaling pathway are also involved, in the formation of LRAP-induced osteogenic
phenotypes.

4. The implication of BMP signaling pathway

Several lines of evidence clearly define the bone morphogenetic protein (BMP) family of
proteins, especially BMP-2, BMP-4 and BMP-7 as potent stimulators for osteogenesis (Chen et
al., 2004a; Rosen et al., 1996). BMP-2 has been reported to induce the expression of osteogenic
markers including Runx2 (Lee et al., 2003a) and Osx (Lee et al., 2003b; Nakashima et al., 2002)
via phosphorylation of Smad proteins, suggesting that the increased osteogenic genes expression
in my current experiment may be the consequence of BMP signal activation. In addition, several
other lines of evidence have shown that amelogenin has the capability to activate the BMP
signaling pathways (Kawase et al., 2001; Lee et al., 2008; Suzuki et al., 2005; Takayama et al.,
2005), resulting in activation of Smad1 phosphorylation and an increase in Runx2 expression
(Takayama et al., 2005). Furthermore, co-activation of the components of the Wnt (Lef1) and
BMP (Smad4) signaling pathways has been reported (Hussein et al., 2003). Therefore, the data
from these studies suggest that BMP-2/Smad signaling pathway may also be activated upon
LRAP addition concurrent with Wnt signaling pathway activation.

Smad1, Smad5 and Smad8 are downstream components of the BMP-2 signaling pathway
(Yamamoto et al., 1997) (Ross and Hill, 2008). The phosphorylation at the C-terminus of Smad
proteins is required for the signal transduction through the BMP signaling pathway (Kretzschmar
et al., 1997). In addition to the BMPs, Smad1, Smad5 and Smad8 can also be phosphorylated in
response TGF- ligands in the presence of the expression of TGF- type I receptor, activin
receptor-like kinase (ALK1) (Goumans et al., 2002). The unchanged level of phosphorylated

125
Smad1/5/8 protein during LRAP treatment at day-5 in the current experiment suggests the
absence of BMP signal, and this observation is supported by the unchanged BMP reporter activity
at day-5.
However, at the later stages of osteogenic differentiation (EB day-9 to day-10), the level of
phosphorylated Smad1/5/8 protein increases along with the BMP reporter activity, suggesting the
presence of BMP or TGF- ligands. Although my data supports the activation of BMP signal, its
fold increase is moderate when compared to the previously reported data for the function of
amelogenin to activate BMP signaling pathway pathways (Kawase et al., 2001; Lee et al., 2008;
Suzuki et al., 2005; Takayama et al., 2005). This inconsistency may be due to the different cell
types and the different concentrations of LRAP used in each experimental protocol. The previous
study by Takayama and colleagues suggests that the effect of Emdogain on Smad1
phosphorylation is dose-dependent with the optimal concentration at 50-100 μg/mL(Takayama et
al., 2005). In addition, the experiment by Takayama and colleagues was performed on a
mesenchymal precursor cell line, as opposed to my experiment that used ES cells. Therefore,
changes in phosphorylation for Smad1/5/8 may be dependent on the concentration of LRAP and
upon the cell type used in the experimental design.

5. The specificity of LRAP peptide sequence, peptide concentration, and culture time

The specificity of the LRAP peptide sequence, as well as the LRAP concentration, and the
required LRAP exposure time to induce osteogenesis have also been demonstrated in my current
study. A scrambled LRAP peptide, presumably without biological activity, failed to induce the
osteogenic phenotype, an outcome that suggests that the signaling function induced by LRAP is
specific to its amino acid sequence and/or conformation.

126
The reduced mineral and calcium formation observed in MC3T3 cells treated with increased
concentration of LRAP suggests that the effect of LRAP is not concentration dependent for these
cells, and that LRAP concentration (10ng/mL) used in the current experiment gives the optimal
outcomes in regard to mineral formation. In addition, the results for the current experiment has
shown that the osteogenic phenotypes, including BSP and Osx expression, mineral formation and
calcium accumulation, induced by the EB cultured in LRAP for 24 hours (Fig 21A, B, C) are
decreased when compared to the osteogenic phenotypes induced by the EB cultured with
continuous exposure to LRAP throughout the course of osteogenic differentiation (Fig 5, 6).
Although further studies to determine the effect of LRAP at different time points are required, the
observation from the current study suggests that the extended LRAP exposure time has an
accumulative effect on osteogenic phenotypes

In this chapter I have provided insights into the mechanisms controlling the osteogenic inductive
effect provided by LRAP. I have shown that the mechanism of action of LRAP may be explained
by means of the activation of canonical Wnt signaling pathway at the early stage of osteogenic
differentiation, and possibly the activation of BMP signaling pathway at the later stage of
osteogenic differentiation. In addition, I have demonstrated the functional specificity of LRAP
as a signaling molecule to induce the cells along the osteogenic lineage. This function is shown
to be concentration-independent, time-dependent, and cannot be mimicked by a single Wnt
protein activating the canonical Wnt pathway. Further studies will be required to address the
questions as to how LRAP regulates osteogenic differentiation at the latter stages of
differentiation, and how LRAP signal(s) relates to other signaling pathway(s).

127
CONCLUSIONS

Complex extracellular and intracellular signaling networks and their regulatory components play
critical roles in determination of cell fate decision. Intracellular components including
transcription factors, transcriptional activators and transcriptional inhibitors regulate the
expression of targeted genes. Extracellular matrix components bind to cell surface receptor(s)
and other extracellular protein(s) resulting in changes to cell behavior in response to these signals.
The cumulative response from other signals is further complicated by the existence of
“moonlighting” proteins, whereby the function of a known protein is distinctively different,
changing the known function of the protein to an entirely different purpose (Jeffery, 2003). One
such example is the enamel proteins serving as structural proteins during late developmental
stages, but moonlighting as cell signaling proteins during earlier developmental stages. The
understanding of how cells respond to these different signals, differentiate to a specific cell type
and eventually commit to a specific cell lineage is fundamental to the concept of tissue
engineering and cell-based therapy, which may provide potential solutions for today’s uncured
diseases.

The focus on my studies has been on the function(s) of a candidate signaling molecule and a
moonlight protein called leucine-rich amelogenin peptide (LRAP). Although LRAP, a splice
product of amelogenin, is found in enamel matrix during enamel development, several lines of
evidence indicate that LRAP has a unique function apart from the structural role of the full-length
amelogenin (Boabaid et al., 2004; Chen et al., 2003; Lacerda-Pinheiro et al., 2006a; Veis et al.,
2000). In the current study, I have focused on the use of stem cells as a model to study cell
differentiation in the presence of LRAP.

128
Based on previous findings from other investigators, I hypothesized that LRAP has an
osteogenic-inducing capacity that serves to guide stem cells to form bone cells and tissue.

In my experiments, I found that LRAP is able to guide the naïve stem cells towards the bone cell
fate. The guidance to osteoblastic phenotypes is true not only for embryonic stem (ES) cells, but
also for human bone marrow mesenchymal stem cells (BMMSCs), human orofacial mesenchymal
stem cells (OF-MSCs) and protodifferentiated osteoblast-like cells MC3T3. These results
suggest that LRAP has the potential to signal a variety of cells from different species and of
different competencies to induce cell determination and differentiation to an osteogenic lineage.

In addition to the function of LRAP during osteogenic induction, I have shown that LRAP also
has the capacity to inhibit adipogenesis in vitro and to induce hematopoiesis while modulating
bone remodeling of human mesenchymal stem cells in vivo. These observations suggest the
unique properties of LRAP to serve as a molecule maintaining homeostasis among various cell
types. Provided that the experiments are performed with adult human mesenchymal stem cells,
the practical use of LRAP in the cell-based therapeutic application, for not only the correction of
bone defects, but also for a wider rage of defects including hematopoiesis and adipogenesis is
promising.

Although the signaling property of LRAP has been known for some time and has been used in
different clinical settings (Popowics et al., 2005), little was known about the mechanism
controlling the effect of LRAP on cell fate determination. The similarity between the effect of
canonical Wnt signaling and the effect of LRAP on osteogenesis and adipogenesis gave me the
clue that LRAP exerts its function through activation of the Wnt signaling pathway.

129
The hypothesis that LRAP stimulates the Wnt signaling cascade has been verified in my study by
the detection of Wnt reporter activity, the increased -catenin protein signal, and the observed
upregulation of several Wnt associated genes upon the addition of LRAP to the target cells. The
activation of the canonical Wnt pathway alone, however, is not sufficient to explain the complete
effects of LRAP on cell fate decisions. For example, the failure of the canonical Wnt3a to mimic
the effect of LRAP, along with the increased expression of genes associated with non-canonical
Wnt pathways following LRAP treatment, suggest the involvement of other signaling pathway(s).
I have explored this possibility by hypothesizing the involvement of BMP signaling pathway,
based on previous observation of the BMP-like properties observed for the enamel matrix
derivative (EMD) (Kawase et al., 2001; Lee et al., 2008; Suzuki et al., 2005; Takayama et al.,
2005), and based on the possible cross-talk between Wnt and BMP signaling pathways (Hussein
et al., 2003). However, I have found that BMP signaling pathway is only moderately activated
upon addition of LRAP to the ES cell culture for 4-5 days (embryoid bodies at day-9 and day-10).
This outcome from the BMP signaling cascade suggests that LRAP activates BMP signal at the
latter stage of osteogenic differentiation, and it does so only after the activation of Wnt signal that
must occur at the earlier stage of osteogenic differentiation. In addition, it is likely that other
signaling pathway(s) may also be operating. The implications from some of the other candidate
signaling pathways, including the non-canonical Wnt signaling (Takada et al., 2007), TGF-
signaling (Goumans et al., 2002), and Notch signaling (Canalis, 2008), remain to be investigated.

The study of the signaling function(s) of LRAP has begun only in the past decade. Although
enamel matrix derivative (EMD), known commercially as Emdogain, has been used with
documental success in clinical dentistry to promote hard tissue regeneration (Casarin et al., 2008;
Hammarstrom, 1997), the study of its molecular function has been limited, partly due to the fact

130
that Emdogain is a complex mixture of amelogenin isoform proteins and their degradation
products. Hence, the long-term effect for Emdogain has been called into question (Fridstrom et
al., 2008; Schjott and Andreasen, 2005). The osteogenic inducing property of LRAP created by
chemical synthesis is the case of my experiments, suggests a potential application for further
clinical translational studies. Here, my data suggests that LRAP could be used to enhance bone
formation and correct bone defects in combination with the use of embryonic stem cells or
mesenchymal stem cells. Furthermore, the mechanistic details revealed by my studies regarding
the effect(s) of LRAP not only provide for a better understanding of the molecular basis for
LRAP function, but also the identity of the pathway(s) that could be targeted by other
pharmacologic agents in order to activate or inhibit the relevant signaling pathways. Finally, the
signaling property of LRAP operating during hematopoiesis and adipogenesis has opened the
door for the additional research to explore the function of LRAP in directing or ablating cells to
these lineages, in addition to the better-known and now better-understood function of LRAP to
stimulate the osteogenic pathway.

131
REFERENCES

Abdallah, B.M., and Kassem, M. (2008). Human mesenchymal stem cells: from basic biology to
clinical applications. Gene Ther 15, 109-116.
Ahumada, A., Slusarski, D.C., Liu, X., Moon, R.T., Malbon, C.C., and Wang, H.Y. (2002).
Signaling of rat Frizzled-2 through phosphodiesterase and cyclic GMP. Science 298, 2006-2010.
Akintoye, S.O., Lam, T., Shi, S., Brahim, J., Collins, M.T., and Robey, P.G. (2006). Skeletal site-
specific characterization of orofacial and iliac crest human bone marrow stromal cells in same
individuals. Bone 38, 758-768.
Aldred, M.J., Crawford, P.J., Roberts, E., and Thomas, N.S. (1992). Identification of a nonsense
mutation in the amelogenin gene (AMELX) in a family with X-linked amelogenesis imperfecta
(AIH1). Hum Genet 90, 413-416.
Aubin, J.E. (2001). Regulation of osteoblast formation and function. Rev Endocr Metab Disord 2,
81-94.
Bafico, A., Liu, G., Yaniv, A., Gazit, A., and Aaronson, S.A. (2001). Novel mechanism of Wnt
signalling inhibition mediated by Dickkopf-1 interaction with LRP6/Arrow. Nat Cell Biol 3, 683-
686.
Baron, R., Rawadi, G., and Roman-Roman, S. (2006). Wnt signaling: a key regulator of bone
mass. Curr Top Dev Biol 76, 103-127.
Bennett, C.N., Longo, K.A., Wright, W.S., Suva, L.J., Lane, T.F., Hankenson, K.D., and
MacDougald, O.A. (2005). Regulation of osteoblastogenesis and bone mass by Wnt10b. Proc
Natl Acad Sci U S A 102, 3324-3329.
Bennett, C.N., Ouyang, H., Ma, Y.L., Zeng, Q., Gerin, I., Sousa, K.M., Lane, T.F., Krishnan, V.,
Hankenson, K.D., and MacDougald, O.A. (2007). Wnt10b increases postnatal bone formation by
enhancing osteoblast differentiation. J Bone Miner Res 22, 1924-1932.
Bianco, P., Riminucci, M., Gronthos, S., and Robey, P.G. (2001). Bone marrow stromal stem
cells: nature, biology, and potential applications. Stem Cells 19, 180-192.
Boabaid, F., Gibson, C.W., Kuehl, M.A., Berry, J.E., Snead, M.L., Nociti, F.H., Jr., Katchburian,
E., and Somerman, M.J. (2004). Leucine-rich amelogenin peptide: a candidate signaling molecule
during cementogenesis. J Periodontol 75, 1126-1136.
Bodine, P.V., Billiard, J., Moran, R.A., Ponce-de-Leon, H., McLarney, S., Mangine, A., Scrimo,
M.J., Bhat, R.A., Stauffer, B., Green, J., et al. (2005). The Wnt antagonist secreted frizzled-
related protein-1 controls osteoblast and osteocyte apoptosis. J Cell Biochem 96, 1212-1230.

132
Bonyadi, M., Waldman, S.D., Liu, D., Aubin, J.E., Grynpas, M.D., and Stanford, W.L. (2003).
Mesenchymal progenitor self-renewal deficiency leads to age-dependent osteoporosis in Sca-
1/Ly-6A null mice. Proc Natl Acad Sci U S A 100, 5840-5845.
Boyce, B.F., and Xing, L. (2008). Functions of RANKL/RANK/OPG in bone modeling and
remodeling. Arch Biochem Biophys 473, 139-146.
Boyden, L.M., Mao, J., Belsky, J., Mitzner, L., Farhi, A., Mitnick, M.A., Wu, D., Insogna, K.,
and Lifton, R.P. (2002). High bone density due to a mutation in LDL-receptor-related protein 5.
N Engl J Med 346, 1513-1521.
Brantjes, H., Barker, N., van Es, J., and Clevers, H. (2002). TCF: Lady Justice casting the final
verdict on the outcome of Wnt signalling. Biol Chem 383, 255-261.
Bretscher, A., Chambers, D., Nguyen, R., and Reczek, D. (2000). ERM-Merlin and EBP50
protein families in plasma membrane organization and function. Annu Rev Cell Dev Biol 16,
113-143.
Calvi, L.M., Adams, G.B., Weibrecht, K.W., Weber, J.M., Olson, D.P., Knight, M.C., Martin,
R.P., Schipani, E., Divieti, P., Bringhurst, F.R., et al. (2003). Osteoblastic cells regulate the
haematopoietic stem cell niche. Nature 425, 841-846.
Campioni, D., Lanza, F., Moretti, S., Ferrari, L., and Cuneo, A. (2008). Loss of Thy-1 (CD90)
antigen expression on mesenchymal stromal cells from hematologic malignancies is induced by
in vitro angiogenic stimuli and is associated with peculiar functional and phenotypic
characteristics. Cytotherapy 10, 69-82.
Canalis, E. (2008). Notch signaling in osteoblasts. Sci Signal 1, pe17.
Caplan, A.I. (1991). Mesenchymal stem cells. J Orthop Res 9, 641-650.
Carinci, F., Piattelli, A., Guida, L., Perrotti, V., Laino, G., Oliva, A., Annunziata, M., Palmieri,
A., and Pezzetti, F. (2006). Effects of Emdogain on osteoblast gene expression. Oral Dis 12, 329-
342.
Carpenedo, R.L., Sargent, C.Y., and McDevitt, T.C. (2007). Rotary suspension culture enhances
the efficiency, yield, and homogeneity of embryoid body differentiation. Stem Cells 25, 2224-
2234.
Casarin, R.C., Del Peloso Ribeiro, E., Nociti, F.H., Jr., Sallum, A.W., Sallum, E.A., Ambrosano,
G.M., and Casati, M.Z. (2008). A double-blind randomized clinical evaluation of enamel matrix
derivative proteins for the treatment of proximal class-II furcation involvements. J Clin
Periodontol 35, 429-437.

133
Chamorro, M.N., Schwartz, D.R., Vonica, A., Brivanlou, A.H., Cho, K.R., and Varmus, H.E.
(2005). FGF-20 and DKK1 are transcriptional targets of beta-catenin and FGF-20 is implicated in
cancer and development. EMBO J 24, 73-84.
Chen, D., Zhao, M., Harris, S.E., and Mi, Z. (2004a). Signal transduction and biological functions
of bone morphogenetic proteins. Front Biosci 9, 349-358.
Chen, D., Zhao, M., and Mundy, G.R. (2004b). Bone morphogenetic proteins. Growth Factors 22,
233-241.
Chen, E., Yuan, Z.A., Wright, J.T., Hong, S.P., Li, Y., Collier, P.M., Hall, B., D'Angelo, M.,
Decker, S., Piddington, R., et al. (2003). The small bovine amelogenin LRAP fails to rescue the
amelogenin null phenotype. Calcif Tissue Int 73, 487-495.
Cheng, H., Jiang, W., Phillips, F.M., Haydon, R.C., Peng, Y., Zhou, L., Luu, H.H., An, N.,
Breyer, B., Vanichakarn, P., et al. (2003). Osteogenic activity of the fourteen types of human
bone morphogenetic proteins (BMPs). J Bone Joint Surg Am 85-A, 1544-1552.
D'Ippolito, G., Schiller, P.C., Ricordi, C., Roos, B.A., and Howard, G.A. (1999). Age-related
osteogenic potential of mesenchymal stromal stem cells from human vertebral bone marrow. J
Bone Miner Res 14, 1115-1122.
Day, T.F., Guo, X., Garrett-Beal, L., and Yang, Y. (2005). Wnt/beta-catenin signaling in
mesenchymal progenitors controls osteoblast and chondrocyte differentiation during vertebrate
skeletogenesis. Dev Cell 8, 739-750.
Dexter, T.M., and Shadduck, R.K. (1980). The regulation of haemopoiesis in long-term bone
marrow cultures: I. role of L-cell CSF. J Cell Physiol 102, 279-286.
Dezawa, M., Kanno, H., Hoshino, M., Cho, H., Matsumoto, N., Itokazu, Y., Tajima, N., Yamada,
H., Sawada, H., Ishikawa, H., et al. (2004). Specific induction of neuronal cells from bone
marrow stromal cells and application for autologous transplantation. J Clin Invest 113, 1701-
1710.
Diekwisch, T., David, S., Bringas, P., Jr., Santos, V., and Slavkin, H.C. (1993). Antisense
inhibition of AMEL translation demonstrates supramolecular controls for enamel HAP crystal
growth during embryonic mouse molar development. Development 117, 471-482.
Dodig, M., Tadic, T., Kronenberg, M.S., Dacic, S., Liu, Y.H., Maxson, R., Rowe, D.W., and
Lichtler, A.C. (1999). Ectopic Msx2 overexpression inhibits and Msx2 antisense stimulates
calvarial osteoblast differentiation. Dev Biol 209, 298-307.
Dominici, M., Le Blanc, K., Mueller, I., Slaper-Cortenbach, I., Marini, F., Krause, D., Deans, R.,
Keating, A., Prockop, D., and Horwitz, E. (2006). Minimal criteria for defining multipotent
mesenchymal stromal cells. The International Society for Cellular Therapy position statement.
Cytotherapy 8, 315-317.

134
Donaldson, J.G., Porat-Shliom, N., and Cohen, L.A. (2008). Clathrin-independent endocytosis: A
unique platform for cell signaling and PM remodeling. Cell Signal.
Du, S.J., Purcell, S.M., Christian, J.L., McGrew, L.L., and Moon, R.T. (1995). Identification of
distinct classes and functional domains of Wnts through expression of wild-type and chimeric
proteins in Xenopus embryos. Mol Cell Biol 15, 2625-2634.
Ellies, D.L., Viviano, B., McCarthy, J., Rey, J.P., Itasaki, N., Saunders, S., and Krumlauf, R.
(2006). Bone density ligand, Sclerostin, directly interacts with LRP5 but not LRP5G171V to
modulate Wnt activity. J Bone Miner Res 21, 1738-1749.
Engin, F., Yao, Z., Yang, T., Zhou, G., Bertin, T., Jiang, M.M., Chen, Y., Wang, L., Zheng, H.,
Sutton, R.E., et al. (2008). Dimorphic effects of Notch signaling in bone homeostasis. Nat Med
14, 299-305.
Fincham, A.G., Belcourt, A.B., Termine, J.D., Butler, W.T., and Cothran, W.C. (1983).
Amelogenins. Sequence homologies in enamel-matrix proteins from three mammalian species.
Biochem J 211, 149-154.
Fincham, A.G., and Moradian-Oldak, J. (1995). Recent advances in amelogenin biochemistry.
Connect Tissue Res 32, 119-124.
Fridstrom, M., Schollin, J., and Crossner, C.G. (2008). Evaluating Emdogain and healing of
replanted teeth using an intra-individual experimental-control study design. Dent Traumatol 24,
299-304.
Fujino, T., Asaba, H., Kang, M.J., Ikeda, Y., Sone, H., Takada, S., Kim, D.H., Ioka, R.X., Ono,
M., Tomoyori, H., et al. (2003). Low-density lipoprotein receptor-related protein 5 (LRP5) is
essential for normal cholesterol metabolism and glucose-induced insulin secretion. Proc Natl
Acad Sci U S A 100, 229-234.
Ganss, B., Kim, R.H., and Sodek, J. (1999). Bone sialoprotein. Crit Rev Oral Biol Med 10, 79-98.
Gaur, T., Lengner, C.J., Hovhannisyan, H., Bhat, R.A., Bodine, P.V., Komm, B.S., Javed, A., van
Wijnen, A.J., Stein, J.L., Stein, G.S., et al. (2005). Canonical WNT signaling promotes
osteogenesis by directly stimulating Runx2 gene expression. J Biol Chem 280, 33132-33140.
Gertow, K., Wolbank, S., Rozell, B., Sugars, R., Andang, M., Parish, C.L., Imreh, M.P., Wendel,
M., and Ahrlund-Richter, L. (2004). Organized development from human embryonic stem cells
after injection into immunodeficient mice. Stem Cells Dev 13, 421-435.
Gibson, C., Golub, E., Herold, R., Risser, M., Ding, W., Shimokawa, H., Young, M., Termine, J.,
and Rosenbloom, J. (1991). Structure and expression of the bovine amelogenin gene.
Biochemistry 30, 1075-1079.

135
Gibson, C.W., Collier, P.M., Yuan, Z.A., and Chen, E. (1998). DNA sequences of amelogenin
genes provide clues to regulation of expression. Eur J Oral Sci 106 Suppl 1, 292-298.
Gibson, C.W., Yuan, Z.A., Hall, B., Longenecker, G., Chen, E., Thyagarajan, T., Sreenath, T.,
Wright, J.T., Decker, S., Piddington, R., et al. (2001). Amelogenin-deficient mice display an
amelogenesis imperfecta phenotype. J Biol Chem 276, 31871-31875.
Glinka, A., Wu, W., Delius, H., Monaghan, A.P., Blumenstock, C., and Niehrs, C. (1998).
Dickkopf-1 is a member of a new family of secreted proteins and functions in head induction.
Nature 391, 357-362.
Goldberg, M., Six, N., Decup, F., Lasfargues, J.J., Salih, E., Tompkins, K., and Veis, A. (2003).
Bioactive molecules and the future of pulp therapy. Am J Dent 16, 66-76.
Gong, Y., Slee, R.B., Fukai, N., Rawadi, G., Roman-Roman, S., Reginato, A.M., Wang, H.,
Cundy, T., Glorieux, F.H., Lev, D., et al. (2001). LDL receptor-related protein 5 (LRP5) affects
bone accrual and eye development. Cell 107, 513-523.
Gordon, M.D., and Nusse, R. (2006). Wnt signaling: multiple pathways, multiple receptors, and
multiple transcription factors. J Biol Chem 281, 22429-22433.
Goumans, M.J., Valdimarsdottir, G., Itoh, S., Rosendahl, A., Sideras, P., and ten Dijke, P. (2002).
Balancing the activation state of the endothelium via two distinct TGF-beta type I receptors.
EMBO J 21, 1743-1753.
Guo, J., and Cooper, L.F. (2007). Influence of an LRP5 cytoplasmic SNP on Wnt signaling and
osteoblastic differentiation. Bone 40, 57-67.
Hammarstrom, L. (1997). The role of enamel matrix proteins in the development of cementum
and periodontal tissues. Ciba Found Symp 205, 246-255; discussion 255-260.
Hassan, M.Q., Javed, A., Morasso, M.I., Karlin, J., Montecino, M., van Wijnen, A.J., Stein, G.S.,
Stein, J.L., and Lian, J.B. (2004). Dlx3 transcriptional regulation of osteoblast differentiation:
temporal recruitment of Msx2, Dlx3, and Dlx5 homeodomain proteins to chromatin of the
osteocalcin gene. Mol Cell Biol 24, 9248-9261.
Hassan, M.Q., Tare, R.S., Lee, S.H., Mandeville, M., Morasso, M.I., Javed, A., van Wijnen, A.J.,
Stein, J.L., Stein, G.S., and Lian, J.B. (2006). BMP2 commitment to the osteogenic lineage
involves activation of Runx2 by DLX3 and a homeodomain transcriptional network. J Biol Chem
281, 40515-40526.
Hatakeyama, J., Philp, D., Hatakeyama, Y., Haruyama, N., Shum, L., Aragon, M.A., Yuan, Z.,
Gibson, C.W., Sreenath, T., Kleinman, H.K., et al. (2006). Amelogenin-mediated regulation of
osteoclastogenesis, and periodontal cell proliferation and migration. J Dent Res 85, 144-149.
Hatakeyama, J., Sreenath, T., Hatakeyama, Y., Thyagarajan, T., Shum, L., Gibson, C.W., Wright,
J.T., and Kulkarni, A.B. (2003). The receptor activator of nuclear factor-kappa B ligand-mediated
osteoclastogenic pathway is elevated in amelogenin-null mice. J Biol Chem 278, 35743-35748.

136
He, X., Saint-Jeannet, J.P., Wang, Y., Nathans, J., Dawid, I., and Varmus, H. (1997). A member
of the Frizzled protein family mediating axis induction by Wnt-5A. Science 275, 1652-1654.
Helms, J.A., and Schneider, R.A. (2003). Cranial skeletal biology. Nature 423, 326-331.
Hill, T.P., Spater, D., Taketo, M.M., Birchmeier, W., and Hartmann, C. (2005). Canonical
Wnt/beta-catenin signaling prevents osteoblasts from differentiating into chondrocytes. Dev Cell
8, 727-738.
Horwitz, E.M., Prockop, D.J., Fitzpatrick, L.A., Koo, W.W., Gordon, P.L., Neel, M., Sussman,
M., Orchard, P., Marx, J.C., Pyeritz, R.E., et al. (1999). Transplantability and therapeutic effects
of bone marrow-derived mesenchymal cells in children with osteogenesis imperfecta. Nat Med 5,
309-313.
Hsieh, J.C., Kodjabachian, L., Rebbert, M.L., Rattner, A., Smallwood, P.M., Samos, C.H., Nusse,
R., Dawid, I.B., and Nathans, J. (1999). A new secreted protein that binds to Wnt proteins and
inhibits their activities. Nature 398, 431-436.
Hu, H., Hilton, M.J., Tu, X., Yu, K., Ornitz, D.M., and Long, F. (2005). Sequential roles of
Hedgehog and Wnt signaling in osteoblast development. Development 132, 49-60.
Huang, H., and He, X. (2008). Wnt/beta-catenin signaling: new (and old) players and new
insights. Curr Opin Cell Biol 20, 119-125.
Hussein, S.M., Duff, E.K., and Sirard, C. (2003). Smad4 and beta-catenin co-activators
functionally interact with lymphoid-enhancing factor to regulate graded expression of Msx2. J
Biol Chem 278, 48805-48814.
Iacob, S., and Veis, A. (2006). Identification of temporal and spatial expression patterns of
amelogenin isoforms during mouse molar development. Eur J Oral Sci 114 Suppl 1, 194-200;
discussion 201-192, 381.
Ichida, F., Nishimura, R., Hata, K., Matsubara, T., Ikeda, F., Hisada, K., Yatani, H., Cao, X.,
Komori, T., Yamaguchi, A., et al. (2004). Reciprocal roles of MSX2 in regulation of osteoblast
and adipocyte differentiation. J Biol Chem 279, 34015-34022.
Itasaki, N., Jones, C.M., Mercurio, S., Rowe, A., Domingos, P.M., Smith, J.C., and Krumlauf, R.
(2003). Wise, a context-dependent activator and inhibitor of Wnt signalling. Development 130,
4295-4305.
Jeffery, C.J. (2003). Moonlighting proteins: old proteins learning new tricks. Trends Genet 19,
415-417.

137

Jepsen, S., Heinz, B., Jepsen, K., Arjomand, M., Hoffmann, T., Richter, S., Reich, E., Sculean,
A., Gonzales, J.R., Bodeker, R.H., et al. (2004). A randomized clinical trial comparing enamel
matrix derivative and membrane treatment of buccal Class II furcation involvement in mandibular
molars. Part I: Study design and results for primary outcomes. J Periodontol 75, 1150-1160.
Jiang, J., Goodarzi, G., He, J., Li, H., Safavi, K.E., Spangberg, L.S., and Zhu, Q. (2006).
Emdogain-gel stimulates proliferation of odontoblasts and osteoblasts. Oral Surg Oral Med Oral
Pathol Oral Radiol Endod 102, 698-702.
Johnson, M.L., Harnish, K., Nusse, R., and Van Hul, W. (2004). LRP5 and Wnt signaling: a
union made for bone. J Bone Miner Res 19, 1749-1757.
Jones, S.E., and Jomary, C. (2002). Secreted Frizzled-related proteins: searching for relationships
and patterns. Bioessays 24, 811-820.
Kang, S., Bennett, C.N., Gerin, I., Rapp, L.A., Hankenson, K.D., and Macdougald, O.A. (2007).
Wnt signaling stimulates osteoblastogenesis of mesenchymal precursors by suppressing
CCAAT/enhancer-binding protein alpha and peroxisome proliferator-activated receptor gamma. J
Biol Chem 282, 14515-14524.
Kato, M., Patel, M.S., Levasseur, R., Lobov, I., Chang, B.H., Glass, D.A., 2nd, Hartmann, C., Li,
L., Hwang, T.H., Brayton, C.F., et al. (2002). Cbfa1-independent decrease in osteoblast
proliferation, osteopenia, and persistent embryonic eye vascularization in mice deficient in Lrp5,
a Wnt coreceptor. J Cell Biol 157, 303-314.
Kawano, Y., and Kypta, R. (2003). Secreted antagonists of the Wnt signalling pathway. J Cell Sci
116, 2627-2634.
Kawase, T., Okuda, K., Momose, M., Kato, Y., Yoshie, H., and Burns, D.M. (2001). Enamel
matrix derivative (EMDOGAIN) rapidly stimulates phosphorylation of the MAP kinase family
and nuclear accumulation of smad2 in both oral epithelial and fibroblastic human cells. J
Periodontal Res 36, 367-376.
Kikuchi, A., Yamamoto, H., and Kishida, S. (2007). Multiplicity of the interactions of Wnt
proteins and their receptors. Cell Signal 19, 659-671.
Kim, J.B., Leucht, P., Lam, K., Luppen, C., Ten Berge, D., Nusse, R., and Helms, J.A. (2007).
Bone regeneration is regulated by wnt signaling. J Bone Miner Res 22, 1913-1923.
Kishida, S., Yamamoto, H., and Kikuchi, A. (2004). Wnt-3a and Dvl induce neurite retraction by
activating Rho-associated kinase. Mol Cell Biol 24, 4487-4501.
Kishimoto, J., Burgeson, R.E., and Morgan, B.A. (2000). Wnt signaling maintains the hair-
inducing activity of the dermal papilla. Genes Dev 14, 1181-1185.

138
Komori, T., Yagi, H., Nomura, S., Yamaguchi, A., Sasaki, K., Deguchi, K., Shimizu, Y.,
Bronson, R.T., Gao, Y.H., Inada, M., et al. (1997). Targeted disruption of Cbfa1 results in a
complete lack of bone formation owing to maturational arrest of osteoblasts. Cell 89, 755-764.
Kretzschmar, M., Liu, F., Hata, A., Doody, J., and Massague, J. (1997). The TGF-beta family
mediator Smad1 is phosphorylated directly and activated functionally by the BMP receptor
kinase. Genes Dev 11, 984-995.
Kuhl, M., Sheldahl, L.C., Malbon, C.C., and Moon, R.T. (2000a). Ca(2+)/calmodulin-dependent
protein kinase II is stimulated by Wnt and Frizzled homologs and promotes ventral cell fates in
Xenopus. J Biol Chem 275, 12701-12711.
Kuhl, M., Sheldahl, L.C., Park, M., Miller, J.R., and Moon, R.T. (2000b). The Wnt/Ca2+
pathway: a new vertebrate Wnt signaling pathway takes shape. Trends Genet 16, 279-283.
Kwee, M.L., Balemans, W., Cleiren, E., Gille, J.J., Van Der Blij, F., Sepers, J.M., and Van Hul,
W. (2005). An autosomal dominant high bone mass phenotype in association with
craniosynostosis in an extended family is caused by an LRP5 missense mutation. J Bone Miner
Res 20, 1254-1260.
Lacerda-Pinheiro, S., Jegat, N., Septier, D., Priam, F., Bonnefoix, M., Bitard, J., Kellermann, O.,
Tompkins, K., Veis, A., Goldberg, M., et al. (2006a). Early in vivo and in vitro effects of
amelogenin gene splice products on pulp cells. Eur J Oral Sci 114 Suppl 1, 232-238; discussion
254-236, 381-232.
Lacerda-Pinheiro, S., Septier, D., Tompkins, K., Veis, A., Goldberg, M., and Chardin, H.
(2006b). Amelogenin gene splice products A+4 and A-4 implanted in soft tissue determine the
reorientation of CD45-positive cells to an osteo-chondrogenic lineage. Journal of biomedical
materials research 79, 1015-1022.
Lagerstrom, M., Dahl, N., Nakahori, Y., Nakagome, Y., Backman, B., Landegren, U., and
Pettersson, U. (1991). A deletion in the amelogenin gene (AMG) causes X-linked amelogenesis
imperfecta (AIH1). Genomics 10, 971-975.
Le, T.Q., Zhang, Y., Li, W., and Denbesten, P.K. (2007). The effect of LRAP on enamel organ
epithelial cell differentiation. J Dent Res 86, 1095-1099.
Lee, A.Z., Jiang, J., He, J., Safavi, K.E., Spangberg, L.S., and Zhu, Q. (2008). Stimulation of
cytokines in osteoblasts cultured on enamel matrix derivative. Oral Surg Oral Med Oral Pathol
Oral Radiol Endod 106, 957-962.

139
Lee, M.H., Kim, Y.J., Kim, H.J., Park, H.D., Kang, A.R., Kyung, H.M., Sung, J.H., Wozney,
J.M., Kim, H.J., and Ryoo, H.M. (2003a). BMP-2-induced Runx2 expression is mediated by
Dlx5, and TGF-beta 1 opposes the BMP-2-induced osteoblast differentiation by suppression of
Dlx5 expression. J Biol Chem 278, 34387-34394.
Lee, M.H., Kim, Y.J., Yoon, W.J., Kim, J.I., Kim, B.G., Hwang, Y.S., Wozney, J.M., Chi, X.Z.,
Bae, S.C., Choi, K.Y., et al. (2005). Dlx5 specifically regulates Runx2 type II expression by
binding to homeodomain-response elements in the Runx2 distal promoter. J Biol Chem 280,
35579-35587.
Lee, M.H., Kwon, T.G., Park, H.S., Wozney, J.M., and Ryoo, H.M. (2003b). BMP-2-induced
Osterix expression is mediated by Dlx5 but is independent of Runx2. Biochem Biophys Res
Commun 309, 689-694.
Lench, N.J., Brook, A.H., and Winter, G.B. (1994). SSCP detection of a nonsense mutation in
exon 5 of the amelogenin gene (AMGX) causing X-linked amelogenesis imperfecta (AIH1). Hum
Mol Genet 3, 827-828.
Li, W., Mathews, C., Gao, C., and DenBesten, P.K. (1998). Identification of two additional exons
at the 3' end of the amelogenin gene. Arch Oral Biol 43, 497-504.
Li, Y., Yuan, Z.A., Aragon, M.A., Kulkarni, A.B., and Gibson, C.W. (2006). Comparison of body
weight and gene expression in amelogenin null and wild-type mice. Eur J Oral Sci 114 Suppl 1,
190-193; discussion 201-192, 381.
Liberali, P., Ramo, P., and Pelkmans, L. (2008). Protein Kinases: Starting a Molecular-Systems
View of Endocytosis. Annu Rev Cell Dev Biol.
Liu, F., Akiyama, Y., Tai, S., Maruyama, K., Kawaguchi, Y., Muramatsu, K., and Yamaguchi, K.
(2008). Changes in the expression of CD106, osteogenic genes, and transcription factors involved
in the osteogenic differentiation of human bone marrow mesenchymal stem cells. J Bone Miner
Metab 26, 312-320.
Liu, P., Wakamiya, M., Shea, M.J., Albrecht, U., Behringer, R.R., and Bradley, A. (1999).
Requirement for Wnt3 in vertebrate axis formation. Nat Genet 22, 361-365.
Luk, J.M., Wang, P.P., Lee, C.K., Wang, J.H., and Fan, S.T. (2005). Hepatic potential of bone
marrow stromal cells: development of in vitro co-culture and intra-portal transplantation models.
J Immunol Methods 305, 39-47.

140
Lyngstadaas, S.P., Risnes, S., Sproat, B.S., Thrane, P.S., and Prydz, H.P. (1995). A synthetic,
chemically modified ribozyme eliminates amelogenin, the major translation product in
developing mouse enamel in vivo. Embo J 14, 5224-5229.
Ma, A., Boulton, M., Zhao, B., Connon, C., Cai, J., and Albon, J. (2007). A role for notch
signaling in human corneal epithelial cell differentiation and proliferation. Invest Ophthalmol Vis
Sci 48, 3576-3585.
MacDonald, B.T., Joiner, D.M., Oyserman, S.M., Sharma, P., Goldstein, S.A., He, X., and
Hauschka, P.V. (2007). Bone mass is inversely proportional to Dkk1 levels in mice. Bone 41,
331-339.
Mao, B., Wu, W., Davidson, G., Marhold, J., Li, M., Mechler, B.M., Delius, H., Hoppe, D.,
Stannek, P., Walter, C., et al. (2002). Kremen proteins are Dickkopf receptors that regulate
Wnt/beta-catenin signalling. Nature 417, 664-667.
Massague, J. (1996). TGFbeta signaling: receptors, transducers, and Mad proteins. Cell 85, 947-
950.
Matsubara, T., Suardita, K., Ishii, M., Sugiyama, M., Igarashi, A., Oda, R., Nishimura, M., Saito,
M., Nakagawa, K., Yamanaka, K., et al. (2005). Alveolar bone marrow as a cell source for
regenerative medicine: differences between alveolar and iliac bone marrow stromal cells. J Bone
Miner Res 20, 399-409.
Mauney, J.R., Kaplan, D.L., and Volloch, V. (2004). Matrix-mediated retention of osteogenic
differentiation potential by human adult bone marrow stromal cells during ex vivo expansion.
Biomaterials 25, 3233-3243.
Miller, J.R. (2002). The Wnts. Genome Biol 3, REVIEWS3001.
Miyama, K., Yamada, G., Yamamoto, T.S., Takagi, C., Miyado, K., Sakai, M., Ueno, N., and
Shibuya, H. (1999). A BMP-inducible gene, dlx5, regulates osteoblast differentiation and
mesoderm induction. Dev Biol 208, 123-133.
Moon, R.T., Brown, J.D., Yang-Snyder, J.A., and Miller, J.R. (1997). Structurally related
receptors and antagonists compete for secreted Wnt ligands. Cell 88, 725-728.
Moore, K.A., and Lemischka, I.R. (2006). Stem cells and their niches. Science 311, 1880-1885.
Morvan, F., Boulukos, K., Clement-Lacroix, P., Roman Roman, S., Suc-Royer, I., Vayssiere, B.,
Ammann, P., Martin, P., Pinho, S., Pognonec, P., et al. (2006). Deletion of a single allele of the
Dkk1 gene leads to an increase in bone formation and bone mass. J Bone Miner Res 21, 934-945.
Nakanishi, R., Akiyama, H., Kimura, H., Otsuki, B., Shimizu, M., Tsuboyama, T., and
Nakamura, T. (2008). Osteoblast-targeted expression of Sfrp4 in mice results in low bone mass. J
Bone Miner Res 23, 271-277.
Nakashima, K., Zhou, X., Kunkel, G., Zhang, Z., Deng, J.M., Behringer, R.R., and de
Crombrugghe, B. (2002). The novel zinc finger-containing transcription factor osterix is required
for osteoblast differentiation and bone formation. Cell 108, 17-29.

141
Nebgen, D.R., Inoue, H., Sabsay, B., Wei, K., Ho, C.S., and Veis, A. (1999). Identification of the
chondrogenic-inducing activity from bovine dentin (bCIA) as a low-molecular-mass amelogenin
polypeptide. J Dent Res 78, 1484-1494.
Nemeth, M.J., and Bodine, D.M. (2007). Regulation of hematopoiesis and the hematopoietic stem
cell niche by Wnt signaling pathways. Cell Res 17, 746-758.
Nusse, R. (2001). Developmental biology. Making head or tail of Dickkopf. Nature 411, 255-256.
Otto, F., Kanegane, H., and Mundlos, S. (2002). Mutations in the RUNX2 gene in patients with
cleidocranial dysplasia. Hum Mutat 19, 209-216.
Otto, F., Thornell, A.P., Crompton, T., Denzel, A., Gilmour, K.C., Rosewell, I.R., Stamp, G.W.,
Beddington, R.S., Mundlos, S., Olsen, B.R., et al. (1997). Cbfa1, a candidate gene for
cleidocranial dysplasia syndrome, is essential for osteoblast differentiation and bone
development. Cell 89, 765-771.
Papagerakis, P., Ibarra, J.M., Inozentseva, N., DenBesten, P., and MacDougall, M. (2005). Mouse
amelogenin exons 8 and 9: sequence analysis and protein distribution. J Dent Res 84, 613-617.
Phillips, B.W., Belmonte, N., Vernochet, C., Ailhaud, G., and Dani, C. (2001). Compactin
enhances osteogenesis in murine embryonic stem cells. Biochem Biophys Res Commun 284,
478-484.
Phinney, D.G., and Prockop, D.J. (2007). Concise review: mesenchymal stem/multipotent stromal
cells: the state of transdifferentiation and modes of tissue repair--current views. Stem Cells 25,
2896-2902.
Piccolo, S., Agius, E., Leyns, L., Bhattacharyya, S., Grunz, H., Bouwmeester, T., and De
Robertis, E.M. (1999). The head inducer Cerberus is a multifunctional antagonist of Nodal, BMP
and Wnt signals. Nature 397, 707-710.
Pittenger, M.F., Mackay, A.M., Beck, S.C., Jaiswal, R.K., Douglas, R., Mosca, J.D., Moorman,
M.A., Simonetti, D.W., Craig, S., and Marshak, D.R. (1999). Multilineage potential of adult
human mesenchymal stem cells. Science 284, 143-147.
Popowics, T., Foster, B.L., Swanson, E.C., Fong, H., and Somerman, M.J. (2005). Defining the
roots of cementum formation. Cells Tissues Organs 181, 248-257.
Prie, D., Beck, L., Friedlander, G., and Silve, C. (2004). Sodium-phosphate cotransporters,
nephrolithiasis and bone demineralization. Curr Opin Nephrol Hypertens 13, 675-681.
Qiang, Y.W., Endo, Y., Rubin, J.S., and Rudikoff, S. (2003). Wnt signaling in B-cell neoplasia.
Oncogene 22, 1536-1545.
Qiang, Y.W., Walsh, K., Yao, L., Kedei, N., Blumberg, P.M., Rubin, J.S., Shaughnessy, J., Jr.,
and Rudikoff, S. (2005). Wnts induce migration and invasion of myeloma plasma cells. Blood
106, 1786-1793.

142
Rosen, V., Thies, R.S., and Lyons, K. (1996). Signaling pathways in skeletal formation: a role for
BMP receptors. Ann N Y Acad Sci 785, 59-69.
Ross, A.M., and Golub, E.E. (1988). A computer graphics program system for protein structure
representation. Nucleic Acids Res 16, 1801-1812.
Ross, S., and Hill, C.S. (2008). How the Smads regulate transcription. Int J Biochem Cell Biol 40,
383-408.
Ross, S.E., Hemati, N., Longo, K.A., Bennett, C.N., Lucas, P.C., Erickson, R.L., and
MacDougald, O.A. (2000). Inhibition of adipogenesis by Wnt signaling. Science 289, 950-953.
Rubin, J.S., Barshishat-Kupper, M., Feroze-Merzoug, F., and Xi, Z.F. (2006). Secreted WNT
antagonists as tumor suppressors: pro and con. Front Biosci 11, 2093-2105.
Sacchetti, B., Funari, A., Michienzi, S., Di Cesare, S., Piersanti, S., Saggio, I., Tagliafico, E.,
Ferrari, S., Robey, P.G., Riminucci, M., et al. (2007). Self-renewing osteoprogenitors in bone
marrow sinusoids can organize a hematopoietic microenvironment. Cell 131, 324-336.
Saito, K., Konishi, I., Nishiguchi, M., Hoshino, T., and Fujiwara, T. (2008). Amelogenin binds to
both heparan sulfate and bone morphogenetic protein 2 and pharmacologically suppresses the
effect of noggin. Bone 43, 371-376.
Schjott, M., and Andreasen, J.O. (2005). Emdogain does not prevent progressive root resorption
after replantation of avulsed teeth: a clinical study. Dent Traumatol 21, 46-50.
Schohl, A., and Fagotto, F. (2002). Beta-catenin, MAPK and Smad signaling during early
Xenopus development. Development 129, 37-52.
Schroeder, T.M., Jensen, E.D., and Westendorf, J.J. (2005). Runx2: a master organizer of gene
transcription in developing and maturing osteoblasts. Birth Defects Res C Embryo Today 75,
213-225.
Schroeder, T.M., Nair, A.K., Staggs, R., Lamblin, A.F., and Westendorf, J.J. (2007). Gene profile
analysis of osteoblast genes differentially regulated by histone deacetylase inhibitors. BMC
Genomics 8, 362.
Scott, C.T., and Reijo Pera, R.A. (2008). The road to pluripotence: the research response to the
embryonic stem cell debate. Hum Mol Genet 17, R3-9.
Sculean, A., Windisch, P., and Chiantella, G.C. (2004). Human histologic evaluation of an
intrabony defect treated with enamel matrix derivative, xenograft, and GTR. Int J Periodontics
Restorative Dent 24, 326-333.
Shenolikar, S., Voltz, J.W., Minkoff, C.M., Wade, J.B., and Weinman, E.J. (2002). Targeted
disruption of the mouse NHERF-1 gene promotes internalization of proximal tubule sodium-
phosphate cotransporter type IIa and renal phosphate wasting. Proc Natl Acad Sci U S A 99,
11470-11475.

143
Shi, S., and Gronthos, S. (2003). Perivascular niche of postnatal mesenchymal stem cells in
human bone marrow and dental pulp. J Bone Miner Res 18, 696-704.
Shi, Y., Wang, Y.F., Jayaraman, L., Yang, H., Massague, J., and Pavletich, N.P. (1998). Crystal
structure of a Smad MH1 domain bound to DNA: insights on DNA binding in TGF-beta
signaling. Cell 94, 585-594.
Shibata, T., Chuma, M., Kokubu, A., Sakamoto, M., and Hirohashi, S. (2003). EBP50, a beta-
catenin-associating protein, enhances Wnt signaling and is over-expressed in hepatocellular
carcinoma. Hepatology 38, 178-186.
Simmer, J.P., and Fincham, A.G. (1995). Molecular mechanisms of dental enamel formation. Crit
Rev Oral Biol Med 6, 84-108.
Snead, M.L. (2003). Amelogenin protein exhibits a modular design: implications for form and
function. Connect Tissue Res 44 Suppl 1, 47-51.
Sneddon, W.B., Syme, C.A., Bisello, A., Magyar, C.E., Rochdi, M.D., Parent, J.L., Weinman,
E.J., Abou-Samra, A.B., and Friedman, P.A. (2003). Activation-independent parathyroid
hormone receptor internalization is regulated by NHERF1 (EBP50). J Biol Chem 278, 43787-
43796.
Stenderup, K., Justesen, J., Clausen, C., and Kassem, M. (2003). Aging is associated with
decreased maximal life span and accelerated senescence of bone marrow stromal cells. Bone 33,
919-926.
Suzuki, S., Nagano, T., Yamakoshi, Y., Gomi, K., Arai, T., Fukae, M., Katagiri, T., and Oida, S.
(2005). Enamel matrix derivative gel stimulates signal transduction of BMP and TGF-{beta}. J
Dent Res 84, 510-514.
Takada, I., Mihara, M., Suzawa, M., Ohtake, F., Kobayashi, S., Igarashi, M., Youn, M.Y.,
Takeyama, K., Nakamura, T., Mezaki, Y., et al. (2007). A histone lysine methyltransferase
activated by non-canonical Wnt signalling suppresses PPAR-gamma transactivation. Nat Cell
Biol 9, 1273-1285.
Takayama, T., Suzuki, N., Narukawa, M., Tokunaga, T., Otsuka, K., and Ito, K. (2005). Enamel
matrix derivative stimulates core binding factor alpha1/Runt-related transcription factor-2
expression via activation of Smad1 in C2C12 cells. J Periodontol 76, 244-249.
Tarasevich, B.J., Howard, C.J., Larson, J.L., Snead, M.L., Simmer, J.P., Paine, M., and Shaw,
W.J. (2007). The nucleation and growth of calcium phosphate by amelogenin. Journal of Crystal
Growth 304, 407-415.

144
Teitelbaum, S.L. (2000). Bone resorption by osteoclasts. Science 289, 1504-1508.
Tompkins, K., George, A., and Veis, A. (2006). Characterization of a mouse amelogenin [A-
4]/M59 cell surface receptor. Bone 38, 172-180.
Tu, X., Joeng, K.S., Nakayama, K.I., Nakayama, K., Rajagopal, J., Carroll, T.J., McMahon, A.P.,
and Long, F. (2007). Noncanonical Wnt signaling through G protein-linked PKCdelta activation
promotes bone formation. Dev Cell 12, 113-127.
Uren, A., Reichsman, F., Anest, V., Taylor, W.G., Muraiso, K., Bottaro, D.P., Cumberledge, S.,
and Rubin, J.S. (2000). Secreted frizzled-related protein-1 binds directly to Wingless and is a
biphasic modulator of Wnt signaling. J Biol Chem 275, 4374-4382.
Veis, A. (2003). Amelogenin gene splice products: potential signaling molecules. Cell Mol Life
Sci 60, 38-55.
Veis, A., Tompkins, K., Alvares, K., Wei, K., Wang, L., Wang, X.S., Brownell, A.G., Jengh,
S.M., and Healy, K.E. (2000). Specific amelogenin gene splice products have signaling effects on
cells in culture and in implants in vivo. J Biol Chem 275, 41263-41272.
Vlad, A., Rohrs, S., Klein-Hitpass, L., and Muller, O. (2008). The first five years of the Wnt
targetome. Cell Signal 20, 795-802.
Wang, E.A., Rosen, V., D'Alessandro, J.S., Bauduy, M., Cordes, P., Harada, T., Israel, D.I.,
Hewick, R.M., Kerns, K.M., LaPan, P., et al. (1990). Recombinant human bone morphogenetic
protein induces bone formation. Proc Natl Acad Sci U S A 87, 2220-2224.
Wang, H.J., Tannukit, S., Wen, X., Shapiro, J.L., Snead, M.L., and Paine, M.L. (2006). Using the
yeast two-hybrid assay to discover protein partners for the leucine-rich amelogenin peptide and
for tuftelin-interacting protein 11. Eur J Oral Sci 114 Suppl 1, 276-279; discussion 285-276, 382.
Warotayanont, R., Zhu, D., Snead, M.L., and Zhou, Y. (2007). Leucine-rich amelogenin peptide
induces osteogenesis in mouse embryonic stem cells. Biochem Biophys Res Commun.
Weiner, S., Sagi, I., and Addadi, L. (2005). Structural biology. Choosing the crystallization path
less traveled. Science 309, 1027-1028.
Weiss, L., and Sakai, H. (1984). The hematopoietic stroma. Am J Anat 170, 447-463.
Westendorf, J.J., Kahler, R.A., and Schroeder, T.M. (2004). Wnt signaling in osteoblasts and
bone diseases. Gene 341, 19-39.
Wodarz, A., and Nusse, R. (1998). Mechanisms of Wnt signaling in development. Annu Rev Cell
Dev Biol 14, 59-88.
Wong, G.T., Gavin, B.J., and McMahon, A.P. (1994). Differential transformation of mammary
epithelial cells by Wnt genes. Mol Cell Biol 14, 6278-6286.

145
Wu, Y., Chen, L., Scott, P.G., and Tredget, E.E. (2007). Mesenchymal stem cells enhance wound
healing through differentiation and angiogenesis. Stem Cells 25, 2648-2659.
Xian, C.J., and Foster, B.K. (2006). Repair of injured articular and growth plate cartilage using
mesenchymal stem cells and chondrogenic gene therapy. Curr Stem Cell Res Ther 1, 213-229.
Xiao, Y.T., Xiang, L.X., and Shao, J.Z. (2007). Bone morphogenetic protein. Biochem Biophys
Res Commun 362, 550-553.
Yabe, T., Shimizu, T., Muraoka, O., Bae, Y.K., Hirata, T., Nojima, H., Kawakami, A., Hirano, T.,
and Hibi, M. (2003). Ogon/Secreted Frizzled functions as a negative feedback regulator of Bmp
signaling. Development 130, 2705-2716.
Yamamoto, N., Akiyama, S., Katagiri, T., Namiki, M., Kurokawa, T., and Suda, T. (1997).
Smad1 and smad5 act downstream of intracellular signalings of BMP-2 that inhibits myogenic
differentiation and induces osteoblast differentiation in C2C12 myoblasts. Biochem Biophys Res
Commun 238, 574-580.
Ye, L., Le, T.Q., Zhu, L., Butcher, K., Schneider, R.A., Li, W., and Besten, P.K. (2006).
Amelogenins in human developing and mature dental pulp. J Dent Res 85, 814-818.
Yoshizawa, T., Takizawa, F., Iizawa, F., Ishibashi, O., Kawashima, H., Matsuda, A., Endo, N.,
and Kawashima, H. (2004). Homeobox protein MSX2 acts as a molecular defense mechanism for
preventing ossification in ligament fibroblasts. Mol Cell Biol 24, 3460-3472.
Zhang, Y., Goss, A.M., Cohen, E.D., Kadzik, R., Lepore, J.J., Muthukumaraswamy, K., Yang, J.,
DeMayo, F.J., Whitsett, J.A., Parmacek, M.S., et al. (2008). A Gata6-Wnt pathway required for
epithelial stem cell development and airway regeneration. Nat Genet 40, 862-870.
Zhou, H., Mak, W., Zheng, Y., Dunstan, C.R., and Seibel, M.J. (2008). Osteoblasts directly
control lineage commitment of mesenchymal progenitor cells through Wnt signaling. J Biol
Chem 283, 1936-1945.
Zhu, D., Paine, M.L., Luo, W., Bringas, P., Jr., and Snead, M.L. (2006). Altering
biomineralization by protein design. J Biol Chem 281, 21173-21182.

Abstract (if available)

Abstract The extracellular matrix proteins in the developing enamel are constituted primarily by a group of highly conserved proteins called amelogenins. A number of amelogenin isoforms are expressed at distinctive stages during enamel development, but the function(s) for each isoform remain unclear. Emdogain, a mixture of porcine amelogenins has been used in clinical dentistry to promote cementogenesis and osteogenesis. A protein constituent in Emdogain called leucine-rich amelogenin peptide (LRAP) has been shown to possess a signaling property shown to induce osteogenic differentiation. However, the pathway(s) involved in the LRAP-mediated osteogenic effect is still unclear, and the function(s) exerted by LRAP to determine stem cell differentiation remains unknown. The purpose of the current study is to explore the effect of LRAP on osteogenesis in different cell types, including mouse embryonic stem (ES) cells and human bone marrow mesenchymal stem cells (hBMMSCs). In addition, the signaling pathway(s) responsible for the osteogenic effect of LRAP was examined. The LRAP-induced phenotype was determined using osteogenic-induced mouse ES cells, and by using osteogenic-induced hBMMSCs. The effect of LRAP on hBMMSCs in vivo was assessed by implantation of hBMMSCs, LRAP and hydroxyapatite/calcium phosphate carrier, into immunocompromised mice. To explore the molecular pathway(s) responsible for the inductive effects of LRAP, I measured beta-catenin and pSmad1/5/8 protein level in LRAP-treated ES cells by Western immunoblotting. The BMPSmad activity and Wnt promoter activity induced by LRAP was detected using a luciferase reporter containing BMP-Smad binding elements, or a reporter containing the TCF/LEF binding elements, respectively. The results suggest that ES cells, in response to endogenous an exogenous LRAP, showed an increase in bone marker gene expression, with increased calcium deposition in the mineralized matrix.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

The effect of leucine-rich amelogenin peptide on mouse bone density by histological analysis

PDF

Modulation of C/EBP alpha in the regulation of mouse amelogenin transcription during tooth formation

PDF

From mesenchymal stem cell therapy to discovery of drug therapy for systemic sclerosis

PDF

Antibody mediated osseous regeneration

PDF

Amelogenin-ameloblastin protein interaction and function in dental enamel formation

PDF

Amelogenin peptides for biomimetic remineralization of enamel and dentin

PDF

Transcellular calcium transport in amelogenesis

PDF

Role of beta-catenin in mouse epiblast stem cell, embryonic stem cell self-renewal and differentiation

PDF

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

PDF

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

PDF

Role of MMP-20 in preventing protein occlusion in enamel apatite crystals: relevance in enamel biomineralization and biomimetics

PDF

Hepatic differentiation in human naïve stem cells compared to human embryonic stem cells

PDF

Non-canonical Tgf-beta signaling in craniofacial development

PDF

Integrin expression and signaling during palatal fusion

PDF

Interaction of epigenetics and SMAD signaling in stem cells and diseases

PDF

Cytomegalovirus induced amelogenesis imperfecta

PDF

The role of ERK1/2 in mouse embryonic stem cell fate control

PDF

Ameloblastin-protein interactions pattern enamel matrix

PDF

Mesenchymal stem cells from human gingiva ameliorate murine alimentary mucositis

PDF

Derivation and characterization of human embryonic stem (hES) cells and human induced pluripotent stem (hiPS) cells in clinical grade conditions

Asset Metadata

Creator Warotayanont, Rungnapa (author)

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

School School of Dentistry

Degree Doctor of Philosophy

Degree Program Craniofacial Biology

Publication Date 11/18/2008

Defense Date 10/20/2008

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

Tag amelogenin,LRAP,OAI-PMH Harvest,Osteogenesis,Wnt signaling

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Snead, Malcolm L. (committee chair), Chai, Yang (committee member), Frenkel, Baruch (committee member), Paine, Michael (committee member), Zhou, Yan (committee member)

Creator Email dorisan@gmail.com,warotaya@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m1813

Unique identifier UC188967

Identifier etd-Warotayanont-2470 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-133554 (legacy record id),usctheses-m1813 (legacy record id)

Legacy Identifier etd-Warotayanont-2470.pdf

Dmrecord 133554

Document Type Dissertation

Rights Warotayanont, Rungnapa

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