University of Southern California Dissertations and Theses

Role of FGFR2b signaling pathway in the development of ectodermal derivatives

(USC Thesis Other)

Role of FGFR2b signaling pathway in the development of ectodermal derivatives

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

ROLE OF FGFR2B SIGNALING PATHWAY IN THE DEVELOPMENT OF
ECTODERMAL DERIVATIVES

by

Sara Parsa

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

December 2010

Copyright 2010 Sara Parsa
ii
Dedication
I would like to dedicate this dissertation to:
My beloved father, Mahmoud Parsa,
whom I lost during my second year at USC and who always believed in me and
encouraged me to pursue my passions.
My adorable mother, Soheila Abolhoda,
whose strength, perseverance and patience, has been always a safe haven for me and my
brothers.
My wonderful husband, Reza Tabatabai,
who has been my biggest supporter during my PhD.

iii
Acknowledgements
I am heartily thankful to my mentor, Dr. Saverio Bellusci for his encouragement,
guidance and support from the initial to the final step. This dissertation would have not
been completed without his expert advise and unfailing patience.
I owe my deepest gratitude to my dear friend, Denise Al Alam for all her help during
my PhD. She has made available her support in a number of ways, by helping me with
the research and proofreading the manuscript. I could have never imagined a more caring
person as a friend and I am sure that this thesis would not have been possible with out her
help. Not to mention Andrew and Brian, her lovely twins, who put a smile on my face
since their “embryonic stage”.
It is an honor for me to thank my dissertation committee, Dr. David Warburton, Dr.
Florence Hofman, Dr. Randall Widelitz and Dr. Louis Dubeau who have patiently and
kindly given their advice during my PhD.
A special thanks to my husband, Reza Tabatabai for keeping up with all the pressure
and stress he had to bear for having a PhD student as a wife. Not only he was helping me
at home and tolerating the fast food life style during these long years, he was helping me
with the projects in the lab during the week and sometimes during the weekend. I am
extremely grateful for all his love, tolerance and unconditional support at all moment.
Indeed, he showed me what for “better and worse” really means…
I am indebted to many of my colleagues for their support during five years of
research in the Bellusci’s “empire”: Caterina Tizzo, Jonathan Branch, Fredric Sala (our

iv
FGF), Suresh Ramasamy, Varsha Gupte, Maria Lavarreda-Pearce, Clarence Wigfall and
Soula Danopoulos (my adopted daughter).
Also, I express my regards to all those who supported me in any respect during the
completion of these projects including all the scientists who have worked at the
developmental and regenerative biology program, FACS core facility personnel and
animal facility personnel at CHLA, also to our collaborators at USC and UCSF.
Last but not least, I would like to express a special word of thanks to my family and
friends who tirelessly listened to my ideas and offered encouragement when it was most
needed. Specially, I am grateful for all the support and love I have received from my dear
mother and brothers, Arash and Shahab. For all those long calls and all words of
assurance which were desperately needed at the time.

v
Table of Contents
Dedication
Acknowledgements
List of Tables
List of Figures
Abstract
Introduction
Chapter 1, Part 1: Terminal end bud maintenance in
mammary gland is dependent upon FGFR2b signaling.
Chapter 1, Part 2: FGFR2b signaling pathway regulates
progenitor lineage formation during postnatal
development of mammary gland.
Chapter 2
FGFR2b signaling controls the regenerative capacity of
the adult mouse incisors during homeostasis.
Chapter 3
FGF10/FGFR2b signaling controls the progressive
formation of the skeletal elements in the limb autopod.
Conclusion
Bibliography

ii
iii
vi
vii
x
1

59

92

116

148
175
181

vi
List of Tables
Table 1: Specificity of FGFRs for different FGFs
Table 2: Phenotypes in Fgf knockout mice
Table 3: FGF Mutations in Human Diseases
and the Corresponding Mouse Models

10
12

22
vii
List of Figures
Fig. 1. Functional evolutionary history of ancestors of the
mouse Fgf gene family
Fig. 2. The phylogenic analysis of the mouse Fibroblast
growth factor (Fgf) gene family
Fig. 3. Gene location analysis of mouse Fgf gene family
Fig. 4. Three-dimentional Structure of FGFs
Fig. 5. Structure of FGF and FGFR
Fig. 6. FGF signaling pathway acts through three main
pathways in different cells
Fig. 7. Activation of class IA phosphatidylinositol 3-kinase
(PI3Ks) through stimulation of receptor tyrosine kinases
(RTKs)
Fig. 8. Model of mammary line specification and placode
formation
Fig. 9. Embryonic mammary development
Fig. 10. Induction of mammary bud outgrowth
Fig. 11. Adult mammary gland development
Fig. 12. Lineage comitment in mammary progenitor cells
Fig. 13. An overview of tooth development illustrating the
reciprocal interactions between epithelium and
mesenchyme during the early stages of tooth development.
Fig. 14. Anatomy of early limb bud and the skeletal
elements of human arm
Fig. 15. The progress zone (PZ) model was formulated to
explain the limb skeletal phenotypes that result from the
manipulation of chicken limb buds.
Fig. 16. The two-signal model.
Fig. 17. The differentiation front model
Fig. 18. The French-flag model

4

5
6
7
8

15

17

30
32
33
35
39

45

51

52
54
55
56
viii
Fig. 19. The interlinked signaling feedback loops
Fig. 20. Expression in WT Mammary gland of FGFR1,
FGFR2 and FGF10.
Fig. 21. Validation of the double transgenic (rtTA;
tet(O)soluble Fgfr2b) embryos
Fig. 22. FGFR2b attenuation leads to the formation of a
rudimentary mammary epithelial tree.
Fig. 23. FGFR2b pathway in postnatal MG is not critical
for the maintenance of the regenerative capacities of the
epithelial MG progenitor cells.
Fig. 24. FGFR2b signaling attenuation leads to decreased
proliferation of the mammary epithelium.
Fig. 25. Evidence that FGFR2b signaling in post-natal
mammary gland development is epithelial cell autonomous.
Fig. 26. FGFR2b signaling attenuation in late puberty leads
to decreased proliferation of the mammary epithelium
Fig. 27. Role of FGFR2b signaling in the mammary gland
during gestation and lactation
Fig. 28. Attenuation of FGFR2b signaling leads to
disappearance of TEBs after 48 hours
Fig. 29. Differential expression of FGFR2 in TEBs and
mature ducts
Fig. 30. Attenuation of the FGFR2b signaling leads to
maturation of the tips of the mammary tree.
Fig. 31. Stem/progenitor cell population of mammary gland
changes after downregulation of Fgfr2b.
Fig. 32. Overexpression of Fgf10 for 6 days leads to
enlargement of the TEBs in DTG-Fgf10 females.
Fig. 33. Overexpression of Fgf10 leads to the increase in
the luminal progenitor cells and decrease in the bipotent
stem cells.
Fig. 34. CL is still present in embryonic incisors in spite of
long-term FGFR2b signaling attenuation

58

68

70

73

78

81

84

90

91

102

104

106

109

111

112

125

ix
Fig. 35. Defective maintenance of epithelial cells in DTG
embryonic mandibular incisors exposed to doxycycline
from E16.5 to E18
Fig. 36. Reduction in the expression of Shh and amelogenin
in DTG mandibular incisors
Fig. 37. Maxillary incisors disappear in the DTG as a result
of long-term attenuation of FGFR2b signaling.
Fig. 38. Long-term postnatal attenuation of FGFR2b
signaling leads to amelogenesis defects in maxillary
incisors.
Fig. 39. Ameloblast formation and ameloblast progenitor
cell proliferation in cervical loop of mandibular incisors
decreases after 4 weeks of treatment in DTG mice.
Fig. 40. Re-growth of defective incisors upon removal of
doxycycline.
Fig. 41. Role of FGFR2b in incisor homeostasis
Fig. 42. Validation of the Rosa26
rtTA
; tet(O)sFgfr2b double
transgenic system during embryonic development.
Fig. 43. Impact of down-regulation of FGFR2b first
appears at the posterior end of the maxillary incisors.
Fig. 44. FGFR2b signaling is critical to control progressive
limb growth along the proximal-distal axis
Fig. 45. The loss of functional AER as a result of
downregulation of FGFR2b signaling
Fig. 46. Rapid loss of AER upon FGFR2b attenuation
Fig. 47. The reversibility of the loss of the AER after
doxycycline IP injection
Fig. 48. Expression pattern of Fgf10 during limb
development using the Fgf10
LacZ
reporter line.
Fig. 49. Fgf10 hypomorph embryos display hindlimb
abnormalities.
Fig. 50. Classical stem cell hierarchy

128

130

132

134

136

139
142

146

147

159

161
164

166

169

171
176
x
Abstract
The Fibroblast growth factor (FGF) family consists of 23 members, which play
important roles during development, homeostasis and pathogenesis by controlling
proliferation, migration and differentiation of cells in multiple organs. Among FGFs, we
are interested in the role of FGF10 and its receptor, FGFR2b in development of
ectodermal derivatives such as mammary gland, limbs and incisors. In this study we
mainly used rtTA transactivator/tetracycline promoter approach allowing inducible and
reversible attenuation of the FGFR2b pathway throughout the embryonic and adult
mouse upon addition of doxycycline. Our study in mammary gland demonstrates the
importance of FGFR2b signaling pathway for maintenance of the terminal end buds
(TEBs) at the tip of the adult mammary gland. TEBs consist of transit amplifying cells
(TACs), which are developed from adult mammary stem cells. We also report that while
FGFR2b signaling is not crucial for the regenerative potential of the mammary epithelial
stem cells, it has a critical role in the regulation of luminal epithelial lineage commitment
of mammary stem cells in the adult mouse. In the second study, it is shown that FGFR2b
signaling is critical for the regenerative capacity of adult incisors by controlling the
proliferation of ameloblast progenitor cells. In the last study, we show that Apical
Ectodermal Ridge (AER), the key structure for limb bud formation, requires FGFR2b
signaling to maintain its structure. At last, in the following dissertation we discuss the
crucial role of FGFR2b signaling pathway in controlling the behavior of stem/progenitor
cells of different ectodermal-derived organs in both embryonic and adult mouse.
1
Introduction
1. 1 Overview of Fibroblast Growth Factors (FGFs) Signaling:
The Fibroblast growth factor family consists of 23 members, which play important
roles during development, homeostasis and pathogenesis by controlling proliferation,
migration and differentiation of cells in multiple organs. Their biological activity is
mediated both by conserved high affinity receptors called "Fibroblast Growth Factor
Receptors" (FGFRs) and "low affinity" receptors, which include heparin sulfates of the
membrane and extracellular glycoproteins. These low affinity receptors are essential for
the activity of FGFs and their binding to the FGFRs.
The first two members of the fibroblast growth factor family were initially purified
from extract of pituitary gland almost forty years ago (Armelin 1973; Gospodarowicz and
Moran 1974). In this extract, two different peptides of FGF were characterized as acidic
and basic FGFs or FGF1 (aFGF) and FGF2 (bFGF) (Gospodarowicz and Bialecki 1978;
Gospodarowicz, Bialecki et al. 1978; Gospodarowicz, Mescher et al. 1978) They were
called fibroblast growth factors based on their ability to induce the proliferation of
embryonic fibroblasts such as mouse 3T3 fibroblasts in vitro. Later, their role in
proliferation, migration and differentiation of other cell types such as endothelial and
epithelial cells was identified (Gospodarowicz, Brown et al. 1978; Gospodarowicz,
Greenburg et al. 1978). The high affinity of these proteins to heparin helped purifying
them from extract of tissues using heparin-sepharose columns (Maciag, Mehlman et al.
1984; Thomas 1984). Sequencing the N-terminal of these purified peptides allowed
cloning of these proteins consequently.
2
The function of FGF signaling is evolutionarily conserved among multi-cellular
organisms. The simpler the organism, the lower the number of FGFs. In Drosophila, two
different FGF receptors have been characterized. One receptor which is called heartless
(htl) is expressed in the mesoderm and is crucial for the development of mesodermal-
derived organs such as the heart (Shishido, Higashijima et al. 1993; Beiman, Shilo et al.
1996; Gisselbrecht, Skeath et al. 1996). Another receptor, called breathless, is important
for the ramification of the tracheal placodes in the fly (Klambt, Glazer et al. 1992;
Reichman-Fried, Dickson et al. 1994; Jarecki, Johnson et al. 1999). The ligand for the
heartless receptor has been elusive for a long time. However, in 2004 two different
groups of scientists reported the discovery of two FGF8-like ligands with redundant
functions which are important for mesoderm migration (Gryzik and Muller 2004;
Stathopoulos, Tam et al. 2004). Comparing the function of these two Drosophila FGF8-
like ligands to FGF8 in Xenopus and mice, the conserved evolutionary role is evident.
Also the ligand for the breathless receptor called branchless is similar to FGF10 in
mouse which has a pivotal role in lung development (Sutherland, Samakovlis et al. 1996;
Bellusci, Grindley et al. 1997). This comparison demonstrates the importance of FGF
signaling during development in an organism as simple as Drosophila to organisms as
complex as mice and Humans.
So far, 23 members of the fibroblast growth factor family have been characterized
based on their structural homology in mice and Human. They were named based on the
order of their identifications. FGF1 and FGF2 were first identified in brain and pituitary
gland. FGF3, FGF4 and FGF5 were identified as oncogenes in the eighties (Sakamoto,
Yoshida et al. 1988; Brookes, Smith et al. 1989; Brookes, Smith et al. 1989; Zhang,
3
Kharbanda et al. 1998) while FGF7 and FGF9 were discovered from conditioned medium
of cultured cells (Finch, Rubin et al. 1989; Miyamoto, Naruo et al. 1993). The cDNA of
Fgf10 was isolated from rat embryos for the first time and based on 120 conserved
central amino acids, it was categorized in the FGF family (Yamasaki, Miyake et al.
1996). Later, more homology was found between this new FGF and FGF3, FGF7 and
FGF22. Completion of human and murine genome project led to the discovery of the
remaining members of the FGF family.

1.1.1 FGFs, an Evolutionary and Developmental Analysis
Single cell organisms do not have FGF genes. However, different numbers of this
gene can be found from nematode Caenorhabditis elegans to homo sapiens (Satou, Imai
et al. 2002). During evolution, FGF genes have duplicated in two different phases. The
first duplication occurred during early metazoan evolution and the second happened
during evolution of early vertebrate (Horton, Mahadevan et al. 2003). During the second
phase, FGF receptors and specific ligand-receptor interactions were generated. The
diversity and specificity of the interactions are increased when the receptors gained the
ability to undergo alternative splicing in the second phase (Mistry, Harrington et al.
2003).
The mouse FGF family divides to 3 different subfamilies based on their functions.
First subfamily is called intracellular FGFs or iFGFs (FGF11,12,13,14) which do not
have receptors and act as intracellular factors (Goldfarb 2005). The second subfamily
consists of hormone-like FGFs or hFGFs (FGF15/19,21,23) which perform as long
distance factors (Ornitz, Xu et al. 1996; Zhang, Ibrahimi et al. 2006). The third and most
4
broad subfamily of FGFs belong to the canonical FGFs subfamily that consists of the rest
of the members of FGF family. The canonical FGFs act as local mediators which bind to
the FGF receptors and initiate the intracellular signaling (Eswarakumar, Lax et al. 2005).
Therefore, based on functional evolution of these genes, intracellular molecules evolved
to become paracrine factors and later they evolved to endocrine factors (Fig. 1).

Fig. 1. Functional evolutionary history of ancestors of the mouse Fgf gene family
(Itoh and Ornitz 2008).

The mouse Fgf gene family is classified based on two different evolutionary
relationships. Based on phylogenic analyses, it is divided into seven different subfamily:
Fgf1(1,2), Fgf4(4,5,6), Fgf7(7,10,22), Fgf8(8,17,18), Fgf9(9,16,20), iFgfs(11,12,13,14),
and hFgfs(15,21,23) (Fig. 2) (Itoh and Ornitz 2008). However, gene location analysis
divides the FGF family into only six subfamilies: Fgf1/2/5, Fgf3/4/6/15/21/23,
Fgf7/10/22, Fgf8/17/18, Fgf9/16/20, and Fgf11/12/13/14 (Fig. 3) (Satou, Imai et al.
2002).
5

Fig. 2. The phylogenic analysis of the mouse Fibroblast growth factor (Fgf) gene
family. This analysis of 22 members of mouse Fgf genes demonstrates seven subfamilies,
each containing two to four members. The evolutionary distance between each gene is
illustrated by the length of each branch (Itoh and Ornitz 2008).

6

Fig. 3. Gene location analysis of mouse Fgf gene family. Based on their gene location,
mouse Fgf gene family divides to six subgroups (Itoh and Ornitz 2008).

7
1.1.2 Fibroblast Growth Factors
Fibroblast growth factors (FGFs) are categorized based on their structural similarities
and their ability to signal through tyrosine kinase receptors. The structural similarity is a
result of homology in their 120 central amino acid domain. This sequence assists their
binding and signaling through the receptors. FGFs are divided into more subgroups based
on their homologies, chemical properties and their expression pattern. They act as a
chemo attractant and their concentration has impact on the response of the cells.

Fig. 4. Three-dimentional Structure of FGFs. (A) Common primary structure of FGF
polypeptide. Shaded area at amino terminus contains a signal sequence found in some
FGF. All FGFs contain a core region that contains conserved amino-acid residues and
conserved structural motifs. The locations of β strands within the core region are
numbered and shown as black boxes. The heparin-binding region (pink) includes residues
in the loop between β strands 1 and 2 and in β strands 10 and 11. Residues that contact
the FGFR are shown in green (the region contacting Ig-domain 2 of the receptor), blue
(contacting Ig-domain 3). (B) Three-dimensional structure of FGF2, a prototypical
member of the FGF family. A ribbon diagram of FGF2 is shown; β strands are labeled 1-
12 and regions of contact with the FGFR and heparin are color-coded as in (a) (Eriksson
et al., 1991; Plotnikov et al., 1999). (C) Three-dimensional superposition of the structures
of FGF7 (purple) and FGF10 (orange).

8
1.1.3 FGF Receptors
There are two different types of receptors for FGFs; one type is the group of high
affinity receptors which belongs to the receptor tyrosine kinase (RTK) family. So far,
four of them have been identified. These tyrosine-kinase receptors are called FGFR1,
FGFR2, FGFR3 and FGFR4. Fgfr1, Fgfr2 and Fgfr3 have 2 different splicing isoforms,
expressed spatially and temporally in different organisms. FGF receptors consist of three
domains: the extracellular domain, transmembrane domain and intracellular domain (Fig.
5).

Fig. 5. Structure of FGF and FGFR. (A) Primary structure of FGFs. The core region
which is conserved among FGFs mediates the binding to FGFR and HSPG. (B) Primary
structure of FGFR. Extracellular domain consists of three Ig domain, an acidic box for
binding of bivalent cations which optimizes the interaction between FGF receptors and
HSPG, a heparin binding domain and CAM-homology domain (CHD) for interaction
with extracellular matrix (ECM) and cell adhesion molecules (CAM). Intracellular
domain consists of juxtamembrane domain and tyrosine kinase domain. Juxtamembrane
domain binds to protein kinase C (PKC) and FRS-2, an adaptor protein (Bottcher and
Niehrs 2005).

9
The extracellular domain is an immunoglobulin-like structure with three
immunoglobulin (Ig) homology loops (I, II, III) which recognize the ligands. Between
Ig-domain I and II, there are two different sequences: one is an acidic box which
facilitates the interaction between FGFR and a low affinity receptor (heparin sulfate
proteoglycan). The other sequence is a heparin-binding domain which interacts with
extracellular matrix and cell adhesion molecules (CAM). The alternative splicing of Fgfrs
occurs in Ig-domain III. While all three Ig-domains are critical for FGF binding, Ig-
domain III indicates ligand specificity. The transmembrane domain of FGFR anchors the
receptor into the cell membrane. The intracellular domain of the receptor displays
tyrosin-kinase activity, which phosphorylates downstream targets and initiates the
intracellular signaling. Each FGFR recognizes several ligands (Table 1).
FGFs also interact with heparin sulfate proteoglycans or HSPG with low affinity.
This interaction stabilizes FGFs by preventing their thermal denaturation and proteolysis.
It also optimizes the activation of FGFR by FGFs (Ornitz 2000; Ornitz and Itoh 2001).
This intrinsic capacity of FGFs to bind to the heparin in vitro and to heparane sulphate
proteoglycans in vivo, gave them also the name Heparin Binding Growth Factors or
HBGFs (Maciag, Mehlman et al. 1984; Thomas 1984; Esch, Baird et al. 1985; Esch,
Ueno et al. 1985).

10
Table 1: Specificity of FGFRs for different FGFs.

"+" quantifies the mitogenic effect of FGF1-9 on BaF3 cells transfected with the respective FGFR isoform.
"-" indicates an absence of interaction. For the other FGFs, "+", "-" and "? " respectively indicate a proven
interaction, an absence of interaction or not studied. (According to Ornitz et al, 1996; Konishi et al, 2000;
Xu et al, 1999; Xie et al, 1999; Beer et al, 2000; Ellsworth et al, 2002 and Ohmachi et al, 2003).

FGFR1b FGFR1c FGFR2b FGFR2c FGFR3b FGFR3c FGFR4
FGF1 ++++ ++++ ++++ ++++ ++++ ++++ ++++
FGF2 ++ ++++ - ++ - ++++ ++++
FGF3 + - ++ - - -
FGF4 - ++++ - ++++ - +++
FGF5 - ++ - + - -
FGF6 - ++ - ++ - -
FGF7 - - +++ - - -
FGF8 - - - - - ++
FGF9 - + ? ++++ ++ ++++
FGF10 +++ - +++ - ? ?
FGF11-
14
- - - - - -
FGF16 ? - ? - ? ?
FGF17 ? +++ ? +++ ? ?
FGF18 - - - +++ - +++
FGF19 - - - - - -
FGF20 ? +++ ? ? ? ? ?
11
1.1.4 Mouse Models to Study FGF Signaling
To study the role of FGFs in vivo, different mice models have been designed. Some
of the traditional Fgf knockout mice models die during the embryonic stages as a result
of gastrulation or cardiac defects. Some such as Fgf10 null embryos die at birth due to
pulmonary defects. Many of these knockout mice survive but they have various
phenotypes in different organs. Table 2 shows the list of Fgf knockout mice models and
their phenotypes (Itoh and Ornitz 2004). Since some of these mice die at birth due to
developmental defects, conditional knockout mouse models, in which a tissue-specific
cre mouse is crossed with the genetically modified mouse that carries Fgf gene flanked
by lox sequence, are used by many scientists.

12
Table 2: Phenotypes in Fgf knockout mice (Itoh and Ornitz 2004).
Gene Phenotype
Fgf1
Viable Not identified
Fgf2
Viable Mild cardiovascular, skeletal, and neuronal defects
Fgf3
Viable Mild inner ear, tail, and CNS defects
Fgf4
Lethal, E4-5 Inner cell mass proliferation
Fgf5
Viable Long hair, angora mutation
Fgf6
Viable Subtle, muscle regeneration
Fgf7
Viable Hair follicle growth, ureteric bud growth
Fgf8
Lethal, E8 Gastrulation defect, CNS and limb development
Fgf9
Lethal, PD0 Lung mesenchyme, heart, gastrointestinal tract, skeleton,
testes development
Fgf10
Lethal, PD0 Multiple organ development (limb, lung, adipose tissues,
etc)
Fgf11
----- -----
Fgf12
Viable Neuromuscular phenotype
Fgf13
----- -----
Fgf14
Viable Neurological phenotype
Fgf15
Lethal, E13.5-
PD21
Cardiac outflow tract development
Fgf16
Viable Mild heart development
Fgf17
Viable Cerebellar development
Fgf18
Lethal, PD0 Skeletal and lung development

13
(Table 2 continued)
Fgf20
----- -----
Fgf21
Viable Not clear
Fgf22
----- -----
Fgf23
Lethal, PW4-13 Growth retardation, phosphate and
vitamin D metabolism

1.1.5 FGF Signaling Pathway
FGF signaling initiates by the binding of FGFs to FGF receptors. Heparin in
extracellular matrix or proteoglycans, which are on the surface of the cells facilitate this
binding. Binding of FGFs to receptors induces receptor dimerization and triggers
autophosphorylation of the tyrosine kinase domains. Phosphorylation of the intracellular
domain, in addition to the induction of the kinase activity, provides docking positions for
recruitment of other kinases such as phospholipase Cγ (PLCγ) and Src-Kinase or
assembly of other docking proteins such as fibroblast growth factor substrate 2 (FRS2).
Through these docking proteins and kinases, three different signaling pathways are
activated: Ras/MAP kinase pathway, PLCγ/Ca
2+
pathway and PI3 Kinase/Akt pathway
(Fig. 6).

1.1.5a Ras/MAP Kinase pathway
One of the docking molecules in FGF signaling is FRS2, a lipid-anchored protein,
that constitutively binds to FGFR1 (Kouhara, Hadari et al. 1997). FRS2 accumulates
proteins for both mitogen-activated protein (MAP) kinase pathway and
14
phosphoinositide3 (PI3) kinase pathway. In Drosophila, another protein called Dof was
identified which is equivalent to FRS2 (Wang, Xu et al. 1996; Vincent, Wilson et al.
1998). After activation of FGFR, the tyrosine kinase domain on FGFR phosphorylates
tyrosines on FRS2 which become the docking sites for adaptor proteins such as growth
factor receptor-bound protein 2 (GRB2) which contains the SRC homology 3 domain
(SH3) or protein tyrosine phosphatase (PTP) SHP2 – SH2 domain containing tyrosine
phosphatase (Kouhara, Hadari et al. 1997; Hadari, Kouhara et al. 1998). Son of sevenless
(SOS) binds to GRB2 through its SH3 domain. SOS is a guanine nucleotide exchange
factor, which forms a complex with GRB2, activates Ras and recruits it into the
membrane. Activation of Ras occurs via exchange of GDP with GTP by SOS. Activated
Ras interacts with effector proteins such as Raf to activate MAP kinase signaling cascade.
Downstream of MAP kinase is transcription factors such as c-myc, AP1 and members of
the Ets family of transcription factors (Fig. 6) (Lee and McCubrey 2002). Ras/ MAP
kinase controls cell cycle and its activation leads to the survival, proliferation, apoptosis
and migration of the cells.
15

Fig. 6. FGF signaling pathway acts through three main pathways in different cells.
All these pathways are not active in all cells that are induced by FGF signaling (Bottcher
and Niehrs 2005).

1.5.b PLCγ/Ca
+2
pathway
After phosphorylation of intracellular domain of FGFR, phospholipase Cγ binds to
the phosphorylated tyrosine 766 on FGFR1 and is phosphorylated by tyrosine kinase
domain of the receptor. The role of FGF signaling in the activation of PLCγ /Ca
2+

pathway was demonstrated by specific mutation in the FGF receptors. Point mutations in
16
tyrosine kinase domain of FGFRs demonstrate the absence of phosphatidylinositol
hydrolysis without any impacts on mitogenesis. This data shows the existence of other
pathways important for mitogenesis besides PLCγ/Ca
2+
pathway downstream of FGF
signaling (Mohammadi, Dionne et al. 1992; Peters, Marie et al. 1992). Activated PLCγ
hydrolyzes the membrane-bound phosphatidylinositol-4,5-diphosphate to diacylglycerol
(DAG) and inositol-1,4,5-triphosphate (IP3). Both of these secondary messengers play
roles in regulation of Ca
+2
concentration inside the cells. DAG activates protein kinase C
and PI3 regulates the intracellular Ca
2+
. Among different types of cells, this pathway is
well-studied in neural tissue in Xenopus as a downstream pathway of FGF2/FGFR4
signaling (Doherty and Walsh 1996; Umbhauer, Penzo-Mendez et al. 2000).

1.5.c PI3 Kinase/Akt pathway
The PI3 kinase/ Akt pathway can be activated by three mechanisms downstream of
FGFR. It can be activated directly by the receptor or indirectly through Gab1 or Ras. Akt
or protein kinase B is activated by PI3 kinase (Fig. 7). In most stages of development
such as mesoderm formation in Xenopus, PI3 kinase/ Akt pathway and Ras/MAP kinase
pathway act parallel to each other (Carballada, Yasuo et al. 2001). The biological
activity of this pathway is illustrated in Fig. 7.

17

Fig. 7. Activation of class IA phosphoinositide 3-kinases (PI3Ks) through
stimulation of receptor tyrosine kinases (RTKs) (Vivanco and Sawyers 2002).

18
1.1.6 Regulation of FGF signaling pathway
The FGF signaling pathway is regulated both by negative and positive signals. Many
of these regulatory signals are common for all tyrosine kinase receptors and many are
specific for FGF signaling (Dikic and Giordano 2003). Also, FGF regulates the
expression of some of its regulators and, as a consequence, they are co-expressed with
FGFs (Niehrs and Meinhardt 2002). These regulations occur at different cellular levels;
extracellular level, transmembrane level and intracellular level.
At the extracellular level, heparin sulfate proteoglycans or HSPGs play an important
role in regulating the interaction between FGF and FGFR. HSPGs, which consist of
sulfated glycosaminoglycans covalently bond to a core protein, form a very diverse
structure based on the sequence of its glycosaminoglycans and core protein (Bernfield,
Gotte et al. 1999). To form the optimal structure for interaction between FGF and FGFR,
glycosyltransferases and sugar sulfotransferases change the spatial and temporal structure
of the HSPGs. Studies show the specific change in the structure of HSPGs throughout
embryonic development and specific FGF-FGFR interaction based on specific HSPGs
structure (Allen and Rapraeger 2003). HSPGs also protect FGFs from thermal
denaturation and proteolysis and prevents diffusion of FGF to interstitial spaces
(Gospodarowicz and Cheng 1986; Moscatelli 1987; Flaumenhaft, Moscatelli et al. 1990).
Not many proteins have been identified at the transmembrane level. The
transmembrane protein Sef, which expression pattern is similar to FGF8 has been
characterized in zebrafish and mice (Furthauer, Lin et al. 2002; Lin, Furthauer et al. 2002;
Tsang, Friesel et al. 2002). Two different regulatory mechanisms have been proposed for
Sef; in one possible mechanism Sef inhibits FGF signaling by interaction with FGFR1,
19
FRS2 and inhibition of MAP kinase and Akt pathways (Kovalenko, Yang et al. 2003). In
other studies, Sef is reported to be downstream or at the level of MEK. Sef prevents
nuclear translocation of ERK without inhibition of its cytoplasmic activity (Torii,
Kusakabe et al. 2004). Another transmembrane regulator has been identified in Xenopus
called XFLRT3, which is a positive regulator of FGF signaling (Bottcher, Pollet et al.
2004). Also, cell adhesion molecules of CAMs regulate FGF signaling through
PLCγ/Ca
2+
pathway by their extracellular domains. These interactions are well studied in
neurons (Hall, Williams et al. 1996; Saffell, Williams et al. 1997; Williams, Walsh et al.
2003).
At the intracellular level, FGF signaling is regulated by two groups of proteins; First,
PTPs which are positive and negative regulators of RTK signaling in general. However,
some of their members such as PTP MAPK phosphatase 3 (MPK3, Pyst1) have the same
expression as the FGF family and their inhibition in chick gives the same phenotype as
some Fgf inhibition (Eblaghie, Lunn et al. 2003; Kawakami, Rodriguez-Leon et al.
2003). The second group of intracellular regulators consists of SPROUTY proteins.
Sprouty was first identified in Drosophila. Subsequently, four members of this family of
regulators were found in mammals. All members of SPROUTY proteins have a highly
conserved cysteine-rich domain at their C-terminal which helps their translocation to the
plasma membrane and negative regulation of the FGF signaling and MAP kinase
pathway. Sprouty proteins act as negative regulators of the FGF signaling. Sprouty 2 and
4 are regulated by FGF signaling and belong to the FGF8 synexpression group- genetic
modules in which expression of different genes are closely correlated and whose
20
members function in a common pathway (Hacohen, Kramer et al. 1998; de Maximy,
Nakatake et al. 1999; Minowada, Jarvis et al. 1999; Lim, Yusoff et al. 2002).

1.2 FGF Signaling and Pathogenesis
There are many different developmental syndromes which are caused by lack of
proper FGF signaling in Human. Mutations in FGF3, FGF8, FGF9, FGF10, FGF14 and
FGF23 lead to embryonic or adult conditions such as deafness, cancer or Parkinson’s
disease (Krejci, Prochazkova et al. 2009). Table 3 illustrates diseases caused by
mutations in the FGFs.
FGF10/FGFFR2b Signaling Pathogenesis:
Apert syndrome (AS) is a congenital disease caused by gain of function in FGFR2b
signaling which can be both ligand-dependent or ligand independent conditions. AS is
characterized by craniofacial, limb, visceral and neural abnormalities (Cohen and
Kreiborg 1993; Cohen and Kreiborg 1993; Cohen and Kreiborg 1993; Cohen and
Kreiborg 1993). In a ligand-dependent AS mouse model, it is demonstrated that FGF10 is
the main ligand which overexpression recovers the phenotype (Hajihosseini, Duarte et al.
2009).
On the other hand, loss of FGF10/FGFR2b signaling leads to the disease called
Lacrimo-Auriculo-Dento-Digital Syndrome (LADD) in which glands, such as parotid,
lacrimal and salivary glands, are affected. In addition to the glands, defects are detected
in the respiratory system, tooth, ear, distal limb and genitalia (Hollister, Klein et al. 1973;
Francannet, Vanlieferinghen et al. 1994; Rohmann, Brunner et al. 2006). In LADD, the
21
FGF10/FGFR2b signaling is completely lost (Shams, Rohmann et al. 2007). This loss of
function is a result of mutations in either FGFR2/3 or FGF10 which leads to the rapid
degradation of the ligand, reduction in ligand/receptor affinity and loss of tyrosine kinase
activity of the receptors (Rohmann, Brunner et al. 2006; Lew, Bae et al. 2007; Shams,
Rohmann et al. 2007). LADD shares some of its symptoms with a milder condition called
Autosomal dominant aplasia of Lacrimal and Salivary Glands (ALSG) (Wiedemann
1997) in which FGF10 is partially lost (Min, Danilenko et al. 1998; Sekine, Ohuchi et al.
1999; Entesarian, Matsson et al. 2005).

22

Table 3: FGF Mutations in Human Diseases and the Corresponding Mouse Models
ψ
(Krejci, Prochazkova et al. 2009).

FGF mutation
coding DNA
protein
Human conditions Corresponding mouse
model/phenotype
FGF3
Deafness associated with complete
absence of inner ear structure (Michel
aplasia), microtia, and microdontia.

Impaired inner ear
development in Fgf3 knockout
mice and in Fgf3/Fgf10
knockout
FGF8
Nonsyndromic cleft lip and palate

Hypogon adotropic hypogonadism and
Kallmann syndrome

Cleft palate in Fgf8
hypomorphic mice

Lack of GnRH neuronsin Fgf8
hypomorphic mice
FGF9
Colorectal, endometrial and ovarian
carcinoma
None
FGF10
Autosomal-dominant aplasia of
lacrimal and salivary glands

Autosomal-dominant lacrimo-auriculo-
dento-digital syndrome

Nonsyndrome cleft lip and palate

Aplasia/hypoplasia of lacrimal
and salivary glands in Fgf10
+/-
mice

None

Cleft palate in Fgf10
+/-
FGF14
Autosomal-dominant spinocerebellar
ataxia 27

Impaired cognitive abilities,
ataxia and paroxysmal
dyskinesia in Fgf14
-/-
and
Fgf11
-/-
/Fgf14
-/-
FGF23
Autosomal-dominant
hypophosphatemic rickets

Familial tumoral cakinosis
Hypophosphatemia, rickets,
and osteomalacia in Fgf23-
Arg176Gln transgenic mice
Hypercakemia and vascular
calcifications in Fgf23
-/-
23
1.3 FGF Signaling and Stem Cells
FGFs are very important growth factors during normal development and homeostasis
of organisms. They play vital roles in regulation of stem cell population and stem cell
niche. The requirement for different type of FGFs is completely different for embryonic
and adult stem cells between mouse and human.
Human Embryonic Stem Cells (hES): hES cells require FGF2 and Activin/Nodal
under serum-free condition which is vital for the maintenance and self-renewal of hES
(Rossant 2008). These cultures usually contain two types of cells, Pluripotent ES cells
and Fibroblast-like cells (Reynolds and Weiss 1992). The later group of cells establishes
the stem cell niche by the direct response to FGF2 and release of insulin-like growth
factor (IGF). IGF supports the self-renewal ability of the stem cells (Bendall, Stewart et
al. 2007).
Mouse Embryonic Stem Cells (mES): Unlike hES, mES do not require FGF for self-
renewal activity but FGF10 is required for differentiation. Studies have shown that loss of
FGF4 or ERK2 results in the failure of the mES to differentiate while retaining their self-
renewal ability (Kunath, Saba-El-Leil et al. 2007).
Neural Stem Cell (NSC): Found in both embryonic and adult mammalian central
nervous system, NSCs are able to differentiate into astrocytes, oligodendrocytes or
neurons. In embryonic neural stem cell culture, 2 different stem cell populations are
found; FGF2-dependent cells and EGF-dependent cells. The former appear during earlier
stages of development and retain their self-renewal ability for longer time (Reynolds and
24
Weiss 1992; Rao 1999) Under physiological conditions, in the adult brain, stem cells are
dormant in the adult brain. While after brain injury, NSCs requires FGF2 to differentiate
and initiate the repair process (Yoshimura, Takagi et al. 2001).
Hematopoietic Stem Cells (HSCs): FGFs are among proteins found in the stem cell
niche that regulate proliferation, differentiation and survival of HSCs (Zhang and Lodish
2008).
Mesenchymal Stem Cells (MSC): FGFs play an important role in differentiation and
growth of MSCs into the connective tissue lineage (Ng, Boucher et al. 2008).

1.4 FGF Signaling and Development
FGF signaling occurs temporally and spatially during development of multicellular
organisms. Most of studies of early development have been done in Xenopus, zebrafish,
mouse and chick.

1.4.1 FGF Signaling and Early Development
1.4.1a Gastrulation
bFGF is the earliest FGF required for mesoderm formation in multicellular organisms
(Slack, Darlington et al. 1987). FGF signaling regulates specification of mesoderm
through T box transcription factor and movement of mesendodermal precursors during
early gastrulation by activation of epithelial-mesenchymal transition (EMT) (Ciruna and
Rossant 2001). EMT is induced when E-cadherin is downregulated by overexpression of
25
Snail which is a transcriptional repressor downstream of FGF signaling (Carver, Jiang et
al. 2001). During early gastrulation in chick, FGF8 acts as a chemorepellent and directs
mesodermal cells away from the blastopore. On the other hand, FGF4 attracts the lateral
mesoderm towards the extending axial mesoderm (Yang, Dormann et al. 2002). Studies
in different organisms such as mice, medaka fish and Xenopus have shown that higher
concentration of FGF is required for induction of migration in the comparison to the
amount needed for the specification of mesoderm (Nutt, Dingwell et al. 2001; Guo and Li
2007; Shimada, Yabusaki et al. 2008). Loss of any components of FGF signaling such as
FGFR, HSPG, or MAP kinase pathway results in gastrulation defects (Amaya, Musci et
al. 1991; Whitman and Melton 1992; MacNicol, Muslin et al. 1993; Itoh and Sokol 1994;
Neilson and Friesel 1996). Also, overexpression of any of these components mimics the
FGF signaling overexpression phenotype. It has been shown that FGF signaling induces
T box transcription factor which is a key molecule in posterior mesoderm and axis
formation in Xenopus (Standley, Zorn et al. 2001; Fisher, Isaacs et al. 2002).
1.4.1b Neural Induction
Neural induction is caused by a “default model” based on which all cells are destined
to be neural cells unless they receive other signals (Hemmati-Brivanlou and Melton 1997;
Hemmati-Brivanlou and Melton 1997) (Weinstein and Hemmati-Brivanlou 1999). In
chick, it has been shown that the interaction between BMP and FGF can determine the
fate of the ectoderm. BMP inhibits the neural fate for the ectoderm while FGF prevents
BMP signaling to interfere with the neural fate of the ectoderm (Alvarez, Araujo et al.
1998; Storey, Goriely et al. 1998; Streit, Berliner et al. 2000; Wilson, Graziano et al.
26
2000). Other studies on Xenopus demonstrated that FGF has a direct role in neural
induction, however, in many of these studies, BMP signaling was partially downregulated
(Kengaku and Okamoto 1995; Lamb and Harland 1995; Ribisi, Mariani et al. 2000).
1.4.1c Anterior-posterior Patterning of the Neural Plate
FGFs are also crucial for Anterior-posterior patterning of the neural plate (Doniach
1995; Johnston, Cox et al. 1995; Lamb and Harland 1995; Partanen, Schwartz et al. 1998;
Holowacz and Sokol 1999). These growth factors favor the posteriorizing of the neural
plates through Wnt signaling or Hox gene (Kudoh, Wilson et al. 2002); (Kolm and Sive
1995; Pownall, Tucker et al. 1996; McGrew, Hoppler et al. 1997; Holowacz and Sokol
1999; Kazanskaya, Glinka et al. 2000; Domingos, Itasaki et al. 2001; Bel-Vialar, Itasaki
et al. 2002).

27
1.4.1d Endoderm Formation
Absence of endoderm in mouse embryos lacking Fgfr1, Fgfr2 or Fgf4 expression
demonstrates the importance of FGF signaling in endoderm formation (Feldman,
Poueymirou et al. 1995; Arman, Haffner-Krausz et al. 1998; Ciruna and Rossant 2001;
Goldin and Papaioannou 2003). However, further studies have shown that in the absence
of FGF signaling, while a defined and distinguishable endoderm is not visible, endoderm
markers (α-fetoprotein, HNF4, and Evx1) are still expressed in early embryo (Chen,
Spencer et al. 2000). Therefore, FGF signaling is vital for pattering and specification
rather than induction of the endoderm (Wells and Melton 2000). Also, since endoderm
formation is affected by mesoderm and mesoderm is extremely defected in the absence of
FGF signaling, it is hard to distinguish the direct or indirect impact of loss of FGF
signaling on endoderm formation (Horb 2000; Kimelman and Griffin 2000).

28
1.4.2 FGF Signaling and Mammary Gland Development
Development of mammary gland initiates around E10 by formation of a defined line,
dorsal to the limbs at two flanks of the body along the anterior-posterior axis in rabbit,
Human and rat. This line called the mammary line is an elevated ectodermal ridge. In
mouse, however, this line can only be detected by expression of Wnt6 as a wide line and
Wnt10b as a thin line which later becomes more restricted to the five inguinal and
thoractic placodes at precise locations along the mammary line (Veltmaat, Van Veelen et
al. 2004). Studies have shown significant role of canonical WNT signaling in the
specification of the mammary line (Eblaghie, Lunn et al. 2003; Chu, Hens et al. 2004;
Veltmaat, Van Veelen et al. 2004). TBX3 is another important molecule in mammary line
formation; its absence leads to the loss of both WNT signaling and placode development
(Davenport, Jerome-Majewska et al. 2003). The expression of Tbx3 is induced by both
WNT and FGF signaling in the mammary gland (Eblaghie, Lunn et al. 2003). The dorso-
ventral border where the mammary line forms, is the result of interaction between BMP4
and TBX3 (Cho, Kim et al. 2006) (Fig. 8).
At E11.5, the mammary line is distinguished as a pseudostratified epithelium in the
ectoderm that connects the mammary placodes (Fig. 8). These placodes are developed
between E11-E12 by migration of epidermal cells along the line and thickening of the
ectoderm (Veltmaat, Van Veelen et al. 2004). The shape of early placodes changes from
comet to disc-like structures. Placode 3 is the first one to appear, followed by placodes 4
and 1; after which placodes 5 and 2 are formed. The temporal difference of placode
appearance demonstrates the different molecular requirement for each placode. The
29
migration of the ectodermal cells is the result of the presence of chemoattractants such as
FGF10. The role of FGF10 as a morphogen in lung bud formation in the lung
development has been previously reported (Bellusci, Grindley et al. 1997; Park, Miranda
et al. 1998; Weaver, Dunn et al. 2000). Fgf10 is expressed in the most ventral-lateral
reaches of the dermatomyotome of the somites close to the mammary line transiently
between E10.5-11 and induces the loss of intercellular contacts and formation of the
pseudopodia. Fgfr2b, the main receptor for FGF10, is expressed in the placodes since
E11.5. The other ligand for FGFR2b which is expressed in the mammary gland is FGF7.
The expression of Fgf7 is decreased by E15.5 and diffuse to the adjacent fat pad
precursors. At this stage, the expression of Fgf10 reappears in the mesenchyme. Fgfr1b,
Fgf4, Fgf8, Fgf9 and Fgf17 are also expressed in the placodes. Among these members of
FGF family, only inhibition of FGFR1 has been studied in the mammary gland which
results in the inhibition of Tbx3 gene and prevention of placode formation (Eblaghie,
Lunn et al. 2003; Chu, Hens et al. 2004).

30

Fig. 8. Model of mammary line specification and placode formation. Fibroblast
growth factor (Fgf)10 from the ventrolateral portion of the somites acts in concert with
Wnts expressed within the epidermis to specify the mammary line. This induces TBX3
expression, which in turn induces the expression of specific Wnts and Fgfs within the
mammary line. These molecules act in an autocrine fashion and cooperate with other
signaling pathways to form 5 pairs of placodes at specific locations along the original
mammary line (by Hens and Wysolmerski 2005).

In addition to FGF signaling, many different signaling pathways are involved in the
development of placodes (Fig. 9). LEF1 is one of the earliest markers identified which
inactivation leads to the formation of only one inguinal placode (van Genderen, Okamura
et al. 1994; Veltmaat, Van Veelen et al. 2004). However, inhibition of WNT signaling
does not affect the expression of Fgf10 and Fgfr2b which indicates that these two
signaling pathways are independent (Mailleux, Spencer-Dene et al. 2002). Ectodysplasin
(Eda) and its receptor (Edar) are also expressed in the mesenchyme surrounding the
placode and inside the placode itself (Mustonen, Pispa et al. 2003). Overexpression of
Eda leads to appearance of an ectopic mammary placode in the mouse embryo but
31
Eda/Edar are not required for the specification of the mammary placodes. Parathyroid
hormone-related protein (PTHrP; also known as PTHLH – Mouse Genome Informatics)
and bone morphogenetic protein 4 (BMP4) are also expressed around E11 in the
mammary placodes. Bmp4 is expressed in both mesenchyme and epithelium during
E11.5-14.5. In Fgfr2b
-/-
mouse model, at E12.5, only one mammary bud is detected in
inguinal region by Lef1 and Bmp4 in situ hybridization. (This mammary bud is either bud
number 4 or a single placode that fails to separate into the placodes number 5 and 4). At
E13, no mammary bud is detected in this mouse. Therefore, FGFR2b signaling is crucial
for the formation and maintenance of 4 placodes and maintenance of the inguinal bud
(Mailleux, Spencer-Dene et al. 2002). Also, as Mailleux et al., (2002) reported, there is a
defect in early stages of mammary gland development in Fgf10 null mouse. Among 5
placodes which form around E11.5 in wild type mice, only a single inguinal bud is
observed and maintained up to E13.5. This primary bud, however, is not able to form
primary and secondary branches. Ability of the same bud to grow branches into the
normal fat pad of the NOD/SCID mice demonstrated the importance of Fgf10 expression
in the mesenchyme of the mammary gland during branching stage. Due to mammary
phenotype similarity between lef1 null mice and Fgf10 null mice, it was suggested that
FGF10 induces expression of Lef1 in mammary placodes. However, implantation of
FGF10 under the flank between E10.5-11.5 demonstrated no induction of Lef1 in the
placodes (Positive control in this experiment was the induction of Sprouty2 by FGF10 in
the lung endoderm).

32
Fig. 9. Embryonic mammary gland development. (A,B) Diagram of an E10.5 mouse
embryo (A) showing the position of the milk line (dashed line between limbs), and of an
E12.5 mouse embryo (B) showing the positions of the five pairs of mammary placodes,
which become mammary buds (MB1-5) along the anteroposterior axis (MB1 and MB5
are hidden by the limb buds and only one flank is shown). (C) Overview of mouse
embryonic mammary gland development. Placodes, which are visible at E11.5, transform
into bulbs of epithelial cells, which sink into the underlying mesenchyme at E13.5 to
become the mammary buds. The mesenchymal cells (orange) that surround the buds
condense to become the mammary mesenchyme (grey). By E15.5, these buds elongate to
form sprouts, which develop a lumen with an opening to the skin, marked by the
formation of the nipple sheath. As the end of pregnancy approaches, at E18.5, the sprouts
become small arborized glands that invade what has now become the fat pad.
Development is essentially arrested at this stage until puberty (by Watson and Khaled
2008).

Around E12.5, the ectodermal cells residing in placodes start to move downwards and
invade the underlying mesenchymal tissue to form a bud. At E12.5, bud number 3 is the
most elevated bud. During the next 2-3 days (E12-16), while the proliferation rate is very
low in the mammary bud and it is in a resting phase, the nipple forms. Also, mammary
placodes disappear around E15 in male mouse as a result of androgen-induced apoptosis.
At the same time the mesenchymal condensation occurs, which depends on the PTHrP in
the mesenchyme and its epithelial receptor, PTHR1. Therefore, bud elongation and
androgen –mediated apoptosis in male, do not occur in the absence of either PTHrP or
33
PTHR1 (Fig. 10) (Wysolmerski, Philbrick et al. 1998; Wysolmerski and Stewart 1998;
Dunbar and Wysolmerski 1999; Foley, Dann et al. 2001) .
Fig. 10. Induction of
mammary bud outgrowth.
Cross-section of an embryonic
mouse mammary bud at E13.5-
15.5. PTHrP and BMP
signaling interact to initiate
mammary bud outgrowth and
nipple formation. PTHrP,
which is secreted from
mammary epithelial cells of the
mammary bud, increases
BMPR1A expression in the
mammary mesenchymal cells
(purple shading), which can
now respond to BMP4. This
triggers epithelial outgrowth,
elevates MSX2 expression, and
inhibits hair follicle formation within the nipple sheath. Modified with permission from
(Hens and Wysolmerski 2005) (by Watson and Khaled 2008).

At E15.5, the mammary bud continues its growth into the mesenchyme to form
primary and secondary branches. This growth is completely independent of hormones
and not all the factors involved in this process are fully identified. PTHrP and PTH1R
play an important role during this stage Wysolmerski, Philbrick et al. 1998; Wysolmerski
and Stewart 1998; Dunbar and Wysolmerski 1999; Foley, Dann et al. 2001).
At birth, the mammary gland consists of only primary and secondary branches.
During the first three weeks after birth, the growth of the mammary tree is quite slow. At
around postnatal day 21, bulb-like structures called terminal end buds (TEBs) appear at
the end of the epithelial ducts which are the leading edge for growth of the epithelial tree
into the mammary fat pad (Fig. 11-A, B). TEBs are very dynamic structures in which
34
cells demonstrate highly proliferative and apoptotic capability. The high rate of
proliferation leads to an invasion of the epithelial cells leading to a new duct formation.
The lumen of this duct forms after apoptosis of the central fraction of the invading cells
(Humphreys, Krajewska et al. 1996). Around 8-10 weeks after birth, hormonal changes
result in the disappearance of TEBs and cessation of the growth of the epithelial tree in
the adult mammary gland. In adult mammary gland, the fat pad is filled with a network of
mammary epithelial trees and each mature duct consists of two epithelial cell types;
luminal epithelial cells and myoepithelial cells. The former cell type surrounds the central
lumen of each duct and during pregnancy differentiates into the milk-producing cells.
Myoepithelial cells cover the mature ducts and their contraction assist the release of milk
into the lumen. Dynamic populations of cells found in TEBs are transit amplifying cells
that can differentiate to either luminal or myoepithelial cells or both. These transit
amplifying cells share some markers (Sca1) with terminally differentiated epithelial cells
(Li and Rosen 2005).

35

Fig. 11. Adult mouse mammary gland development. A schematic of the stages (A-F)
of mammary gland development in the adult mouse, from pre-puberty to pregnancy,
lactation and involution. LN, lymph node; TEB, terminal end bud (by Watson and Khaled
2008).

After puberty, the mammary tree undergoes subtle changes during each estrus cycle
by formation of small branches at the beginning of each cycle and disappearance of these
branches by apoptosis at the end of the cycle caused by the hormonal changes (Watson
and Khaled 2008). During early gestation, the mammary tree forms tertiary branches
which are terminated into the alveolar buds where at late gestation, the milk is produced.
After delivery, proliferation of Luminal epithelial cells and their differentiation and
commitment to the secretory alveolar lineage occurs as a result of a molecular switch.
Milk Whey Acidic Protein (WAP) is one of the molecular indicators of this switch. After
weaning, the mammary alveolar lineage undergoes apoptosis and 80% of its cells
disappear during involution. After involution, the shape and structure of the mammary
tree goes back to its shape before pregnancy (Fig. 11).

36
Adult Progenitor Cells in Mammary Gland
The first time scientists noticed the presence of stem cells in adult mammary gland
was 50 years ago. DeOme et al. reported that pieces from different parts of adult
mammary gland can develop into a mature duct in a cleared mammary fat pad (Deome,
Faulkin et al. 1959). Since then, different approaches have been used to identify the adult
mammary stem cells. First, Bickenbach et al. tried to perform a Long-term
bromodeoxyuridine (BrdU) labeling approach to mark label-retaining cells or LRCs
(1981). These cells were introduced as putative adult mammary gland progenitor cells at
the time (Morris and Slotkin 1985; Cotsarelis, Sun et al. 1990) but later it was found that
these cells are not completely quiescent and they possess the characteristics of transit
amplifying cells.
Later, using the Hoechst 33342 dye efflux, the side population (SP) of mouse
mammary gland and human breast was identified. Although SP of hematopoetic system
is enriched with stem cells, the same number of stem cells is found in SP fraction of
mammary gland versus non-SP fraction (Spangrude and Johnson 1990; Welm, Tepera et
al. 2002). SP of human breast is enriched in cells with high expression of Musashi, Ck19,
p21 and Estrogen receptor. SP of mouse mammary gland, on the other hand, is enriched
with Sca1
+
cells and LRCs. However, none of these side populations possess stem cell
activity. Percentage of stem cells in SP of mammary gland is 0.5%, while the total
percentage of mammary stem cells is 2%. Therefore, side population (SP) cannot be
considered as a candidate subpopulation of adult stem cells in the breast and mammary
gland (Alvi, Clayton et al. 2003; Clayton, Titley et al. 2004; Smith 2005; Woodward,
Chen et al. 2005).
37
Another approach to identify the stem cells in adult mammary gland is to use specific
markers. One of the candidate markers is stem cell antigen 1 (Sca1) which is found in
immortalized mammary cell lines such as COMMA-D (Woodward, Chen et al. 2005).
However, Sca1 is highly expressed in cap cells which have a high rate of proliferation in
pre-puberty mammary gland. This suggests that SCA1 is more likely a marker for transit
amplifying cells rather than stem cells (Woodward, Chen et al. 2005).
In 2006, identification of new markers, CD29 and CD49f (Shackleton, Vaillant et al.
2006; Stingl, Eirew et al. 2006), provided a new approach to sort the subpopulation of
stem/progenitor cells by Fluorescent Activated Cell Sorting (FACS) analysis. Three types
of mammary progenitor cells have been identified in mouse and Human; bipotent stem
cells, myoepithelial progenitor cells and luminal progenitor cells (Stingl, Raouf et al.
2006). Bipotent stem cells are able to differentiate to both luminal progenitor cells and
myoepithelial progenitor cells. In this study, using FACS analysis, it has been reported
that the bipotent stem cells and myoepithelial progenitor cells express high levels of
CD29 and CD49f therefore called CD29
hi
and CD49f
hi
. These cells are called Mammary
Reforming Units (MRU) which are capable to form a complete mammary gland in vivo
and in vitro. On the other hand, luminal progenitor cells are enriched in CD29
med
and
CD49f
med
epithelial cells, expressing lower level of these markers. This population of
cells is called Mammary Colony Forming cells (Ma-CFC) and forms uniform colonies in
vitro, Ma-CFCs are progenies of MRUs.

38
Fig. 12. Lineage commitment in mammary progenitor cells (previous page). (A) A
mammary stem cell (blue) in its niche (purple). (B) Stem cells self-renew in their niche in
a process that might require GATA3. Asymmetrically dividing stem cells produce transit-
amplifying (light blue) daughter cells that commit to either the basal or luminal
progenitor lineages. A common progenitor for ductal and alveolar cells is depicted,
although there might be two different progenitors. Induction of STAT6 in response to IL4
and IL13 induces further expression of GATA3 and possibly of c-MAF, which are
required for differentiation to alveolar cells. RBPJκ is required to maintain the alveolar
phenotype and suppresses differentiation along the basal lineage (by Watson and Khaled
2008).

39
(Fig. 12 continued)

40
1.4.3 FGF Signaling and Tooth Development
Mouse display one generation of three molars and one incisor in each quadrant. While
pattern and shape of teeth is different between mice strains, the pattern of molecular
expression is very identical during tooth development and homeostasis. Incisors grow
continuously during the life-time of the animal due to the presence of adult stem cells, a
well-defined structure, called cervical loop.
Early stages of tooth development are very similar to the early stages of mammary
gland and hair follicle development in mice. During the initiation stage, the oral
epithelium thickens around E11.5 at regions with high expression of Sonic hedgehog, Shh
(Hardcastle, Mo et al. 1998). At E12, the oral epithelium evaginates into the surrounding
mesenchyme, which is derived from neural crest to form a bud at E13.5. Recombination
experiments have shown that the instruction for tooth bud formation before E12 is
provided by the epithelium and after E12, by the mesenchyme (Mina and Kollar 1987;
Lumsden 1988; Tucker and Sharpe 1999).
At E13.5, the condensation of the tooth mesenchyme and the expression of key
molecules in this mesenchymal compartment such as Bmp4, Msx1 and Wnt are hallmarks
of tooth initiation. Bmp4 expression in the mesenchyme induces tooth epithelium to form
the enamel knot which is visible at the tip of the bud by induction of p21 in the
epithelium. Later at the cap stage, the enamel knot can be detected as a bulge inside the
inner enamel epithelium. Conditional knockout of Bmp4 in the epithelium, however,
results in the arrest of tooth development at the bud stage, which demonstrates the
importance of BMP4 for cap formation (Andl, Ahn et al. 2004).
41
Bmp4 can be induced positively by Msx1 and Wnt or negatively by BMP4
antagonists. Bmp4 and Msx1 control each other expression by positive feedback loop.
Loss of Msx1 and Pax9 results in loss of Bmp4 and eventually, arrest of tooth
development at the bud stage (Satokata and Maas 1994). Addition of recombinant BMP4
can rescue this phenotype in vitro (Chen, Bei et al. 1996; Bei, Kratochwil et al. 2000;
Zhao, Zhang et al. 2000). Loss of Wnt signaling also results in loss of Bmp4 and arrest of
tooth development (Liu, Chu et al. 2008). Loss of ectodin, an antagonist of BMP4, results
in overexpression of p21, enlarged enamel knot and cuspal defects(Laurikkala, Kassai et
al. 2003); (Kassai, Munne et al. 2005)
The cells inside the enamel knot do not proliferate or differentiate but they control the
movement and the differentiation of other cells by the expression of molecules such as
Shh, Bmp4, Fgf4 and Wnt10b (Thesleff, Vaahtokari et al. 1996; Vaahtokari, Aberg et al.
1996; Vaahtokari, Aberg et al. 1996); (Sarkar and Sharpe 1999); (Jernvall, Kettunen et al.
1994). Therefore, the enamel knot could be compared to an organizer. The cap-shaped
structure forms around E14.5 as a result of high proliferation outside the knot and low
proliferation inside the knot. The epithelium is divided into inner and outer enamel
epithelium by folding. Dental papilla forms from mesenchymal cells adjacent to the inner
enamel epithelium and dental follicles are formed from mesenchymal cells outside the
dental papilla.
Molars have two enamel knots and incisors have one enamel knot. In the molars, the
first enamel knot disappears after apoptosis at the cap stage and reappears at E16.5 (Bell
stage) and induce formation of more folding and therefore, a complex multi-cuspid tooth.
Bmp4 has an active role in induction of apoptosis in the enamel knot. After formation of a
42
multi-cuspid tooth, hard tissue formation such as enamel and dentin is initiated by
ameloblasts and odontoblasts to shape the tooth. A tertiary enamel knot also forms in the
molars and appears as a cluster of Slit1-expressing epithelial cells next to the enamel-free
areas at the cusp tips (Luukko, Loes et al. 2003). The cuspal pattern is determined by
number of enamel knots (Vaahtokari, Aberg et al. 1996).
Shape and size of the enamel knot determines the curvature of the teeth. EDA and
EDAR are important markers which determine the form of the curvature in the tooth. .
The loss of Eda affects the folding of the epithelium and location of the secondary
enamel knot and results in a flattened cusp of the tooth (Tucker, Headon et al. 2000;
Ohazama, Courtney et al. 2004). The loss of Eda receptor or Edar also does not affect the
size of the tooth but has an impact on the normal shape of the teeth (Tucker, Headon et al.
2000). Loss of Eda does not have any significant impact on the incisors. Absence of Wnt
signaling during the early bell stage leads to downregulation of Eda and appearance of
the same phenotype as the one observed in Eda knockout mice (Laurikkala, Mikkola et
al. 2001; Liu, Guo et al. 2008).
The location of molars is determined by the specific expression pattern of homeobox-
containing genes. The ‘Homeobox code’ is the expression used to describe the
overlapping pattern of homeobox genes in the mesenchyme across the jaw (Sharpe 1995).
Based on the Homeobox code, the mandible is divided to oral (Lhx6, 7), aboral (Gsc),
distal (presumptive incisors) (Msx1.2) and proximal (presumptive molar) (Dlx1,2, Barx1,
Pitx1) domains (Thomas, Tucker et al. 1997; Tucker, Al Khamis et al. 1998; Tucker,
Matthews et al. 1998; Tucker and Sharpe 1999; Mitsiadis and Drouin 2008) Change in
the expression pattern of any of these genes can result in the development of molars in
43
the place of the incisors or vice versa (Tucker, Al Khamis et al. 1998; Miletich, Cobourne
et al. 2005). Interestingly, the loss of homeobox genes expression has a different impact
on mandibular vs. maxillary molars. Studies have shown that while the loss of homeobox
genes results in loss of maxillary molars, the mandibular molars are still unaffected
(Thomas, Tucker et al. 1997)
The expression of homeobox genes in the mesenchyme is induced by positive and
negative signals from the oral epithelial cells. Shh from the pharyngeal endoderm during
pharyngeal arch formation induces the expression of Bmp4 and Fgf8 in the distal
epithelium (Haworth, Smith et al. 2007; Brito, Escalona et al. 2008). While Bmp4 in the
distal epithelium induces expression of Msx1 in the underlying mesenchyme, Fgf8
induces the expression of Barx1 and Pitx1 in the molar epithelium (Tucker, Al Khamis et
al. 1998). Bmp4 and Fgf8 which are expressed adjacent to each others in the epithelium
negatively regulate each other expression (Wilson and Tucker 2004) and define the
boundaries between incisors- and molars- forming regions. On the other hand, Bmp4 and
Fgf8 are induced by other homeobox genes such as Pitx2 (Otlx2) and Islet-1 (Mucchielli,
Mitsiadis et al. 1997; Lu, Pressman et al. 1999; Mitsiadis, Angeli et al. 2003).

44
Fig. 13. An overview of tooth development illustrating the reciprocal interactions
between epithelium and mesenchyme during the early stages of tooth development.
In the boxes the various signaling molecules (in bold) involved are shown and the
transcription factors expressed in the tissue compartment (in italics). This set of
regulatory molecules depicted here is incomplete owing to space constraints and lack of
knowledge, but should give an indication of the complexity of regulatory interactions
involved. This figure is also lacking the inhibitors of the major signaling pathways, which
increase the regulatory complexity. This complex set of interaction leads up to the
formation of a specific cusp pattern. This cusp pattern varies a lot between species, and
even between tooth types within a species. The differences in cusp patterns are set up by
the tinkering of the complex reciprocal interactions between epithelium and
mesenchyme. In the lower part of the figure regulatory alternatives are given for late
tooth development leading to the formation of radically different tooth types, such as
low-crowned or brachydont teeth, high-crowned or hypsodont teeth, and continuously
growing or hypselodont teeth, such as the rodent incisor and also continuously growing
molars found in many species. The different qualities of these tooth types are modified by
means of tinkering with specific regulatory pathways. For instance the epithelial stem cell
niche needs the mesenchymal signaling factor Fgf10 or the niche disappears, root
formation is initiated and a low-crowned molar is formed. Continuously growing teeth
also produce root tissue, but they maintain their stem cell niche as can be seen by the
upkeep of the Fgf10 signal. Instead of removing the stem cell niche they modify the fate
of the progeny of the stem cell niche by modifying BMP signaling, resulting in the
production of ameloblasts in the crown analog, or fragmented root epithelium in the root
45
(Fig. 13 continued) analog, as can be seen in the incisor (by Thesleff et al., 2008).

46
Cells constituting the tooth differentiate terminally after disappearance of the
secondary enamel knot. At this stage, inner enamel epithelium (IEE) differentiates to
ameloblasts and dental pulp mesenchymal cells differentiate into the odontoblasts (Ruch,
Lesot et al. 1982; Begue-Kirn, Smith et al. 1994). Epithelial-mesenchymal interaction has
a pivotal role in this stage of tooth development. IEE through TGFβ1, FGFs and BMP2
induces the formation of the dental mesenchyme, while signals from odontoblasts allow
the differentiation of the pre-ameloblasts to ameloblasts in IEE. Odontoblasts produce
dentin which is a bone-like component. Dentin consists of both Collagen and non-
collagenous proteins (NCPs). Ameloblasts produce enamel which is the only epithelial-
derived calcified tissue in mammals. Amelogenin is the major component of the enamel;
it is crucial for root formation, periodontium regeneration and it acts as a growth factor
(Zeichner-David 2001). Both dentin and enamel give to the tooth its rigid and firm
properties.
Incisor development is slightly different from molar development in mouse. While
molars achieve terminal differentiation neonatally, incisors grow continuously during the
lifetime of the animal. This continuous growth is the result of the presence of adult stem
cells in the cervical loop. Incisors never form the secondary enamel knot and they only
have enamel on the labial side. This asymmetric deposition of the enamels helps
sharpening the incisors by abrasion. Studies have shown that the dental epithelium in the
lingual side of the incisors is not able to respond to the signals from the underlying
mesenchyme due to the presence of Follistatin, an inhibitor of ameloblasts differentiation
(Wang, Suomalainen et al. 2004). This inhibition occurs directly and indirectly through
the Tgfβ superfamily and Fgf3. Fgf3 is inhibited indirectly by Follistatin. In normal
47
conditions, expression of Fgf3 is inhibited by Bmp4. However, Activin, prevents the
inhibition of Fgf3 expression by repressing Bmp4. The presence of Follistatin at the
lingual side of the incisors leads to activation of Bmp4 and consequently inhibition of
Fgf3 (Wang, Suomalainen et al. 2007). Other inhibitors on the lingual side of the incisors
are the Sprouty genes. They are expressed in the dental mesenchyme and prevent the FGF
signaling on the lingual side. Studies on mutant mouse model in which both alleles of
Sprouty4 and one allele of Sprouty2 were eliminated, demonstrated the presence of the
enamel on both lingual and labial sides of the incisors (Klein, Lyons et al. 2008).
Incisors have two distinct continuously growing structures: Crown analog located at
the labial side of the incisor consists of enamel and dentin and root analog, which is on
the lingual side of the incisor and consists of dentin, cementum and periodontal
ligaments. During tooth development, Fgf10 is expressed in the mesenchyme adjacent to
the epithelium of the molars. FGF10 and NOTCH1 are active in molars during
development of the crown and their expressions disappear after initiation of root
formation. Therefore, it has been proposed that FGF10 and NOTCH1 are required for
maintenance of the stem cell niche in tooth (Tummers and Thesleff 2003). In incisors,
Fgf10 is expressed around the cervical loop in the mesenchyme (Harada, Kettunen et al.
1999) and maintains the cervical loop. The cervical loop is where both epithelial and
mesenchymal stem cells are located. The exact location of stem cells in cervical loop is
believed to be at the border of the basal epithelium with high expression of Lunatic
Fringe and the stellate reticulum with Notch1 expression (Harada, Kettunen et al. 1999).
It has been shown previously that the NOTCH signaling pathway is important for the
maintenance of the undifferentiated state of the stem cells (Fortini and Artavanis-
48
Tsakonas 1993; Fortini, Rebay et al. 1993; Bray 1998; Carlesso, Aster et al. 1999).
Further studies reported the importance of FGF10 in maintenance of the continuous
growth in the incisors (Yokohama-Tamaki, Ohshima et al. 2006). FGF10 regulates the
stem cell niche at the cervical loop of the incisors by direct and indirect mechanisms.
First FGF10 induces the expression of Lunatic Fringe in the basal epithelium and Lunatic
Fringe, on the other hand, regulates Notch1 expression in the dental epithelium. Second,
FGF10 acts as a mitogen and induces the proliferation in the epithelium (Harada,
Kettunen et al. 1999). Therefore, FGF10 induces proliferation and differentiation of
epithelial stem cells in the incisors. However, the absence of FGF10 results in apoptosis
in the cervical loop and the reduction of this structure (Harada, Toyono et al. 2002).
In addition to Fgf10, Fgf3 is also expressed in the mesenchyme and binds to FGFR2b
on the epithelial cells. However, unlike Fgf10, Fgf3 is not expressed adjacent to the
cervical loop and it is detected only next to the pre-ameloblasts and ameloblasts in the
enamel epithelium. Although the cervical loop is reduced in Fgf10
-/-
mice, the enamel
epithelial layer is still visible. This observation proves the importance of FGF3 in
maintenance of the enamel epithelial layer (Harada, Toyono et al. 2002).

49
1.4.4 FGF Signaling and Limb Bud Formation
Limb development has been one of the most interesting topics in developmental
biology since 1948 (Saunders 1948). It has also been one of the most prominent models
used to study epithelial-mesenchymal interactions. Limb develops from interaction
between mesoderm and its overlying ectoderm. Skeleton and connective tissue of the
limb are developed from mesenchymal elements in the limb bud while the muscle of the
limb is the result of cell migration from lateral edges of the somites to the limb. Limb
consists of thee anatomical parts; stylopod (proximal part), zeugopod (middle part) and
autopod (distal part) (Fig. 14-b).
The primary location of forelimb formation is in the lateral plate mesoderm and next
to the somites 16-20 and presumptive hindlimb location is adjacent to the paraxial
mesoderm, which is not segmented. Although the early process of development for
forelimb and hindlimb is similar, the temporal expression of critical molecules is
different which leads to later initiation of hindlimb (E10) in comparison to the forelimb
(E9.5). Basic developmental biology techniques such as foil barrier and extirpation
studies have defined different regions and molecules involved in different stages of the
limb development. Most of the information available on limb development comes from
studies on chick and rodents (Summerbell, Lewis et al. 1973).
Limb bud which has a very simple anatomical structure, mesenchyme covered by a
layer of epithelial cells (Fig. 14- a), is divided into different molecular zones that
determine three different axis of the limb. The first molecular zone is a specialized
epithelial region called the Apical Ectodermal Ridge or AER, which was first identified
50
in the chick (Saunders 1948; Fallon and Kelley 1977). AER is induced by transient
expression of FGF8 in a thin layer of mesenchymal cells underlying the ectoderm at
presumptive limb locations on the anterior-posterior axis of the body. Transient
expression of FGF8 in few cells in the mesenchyme is induced by FGF10 which is
initially expressed in the segmental plate and eventually becomes confined to the lateral
plate of the mesoderm. To study the importance of epithelial-mesenchymal interaction in
limb bud formation, a foil barrier was placed between intermediate mesoderm and
ectoderm. The earlier the disruption between AER and underlying mesenchyme called
the ‘progress zone’ during limb development, the more the defect in the limb is striking.
This indicates the importance of the instructive signals from the AER for induction of the
skeletal elements (Stephens and McNulty 1981; Strecker and Stephens 1983; Geduspan
and Solursh 1992).

51

Fig. 14. Anatomy of early limb bud and the skeletal elements of a human arm. a.
Scanning electron microscopy image of a mouse embryo at gestational day 10.5. The
forelimb bud forms at the level of the heart at E9.5; hindlimb develops 12 hours later at
E10 at the level of the kidneys. The enlarged inset shows the forelimb bud with the two
main limb bud axes indicated. The apical ectodermal ridge (AER) is indicated in green.
b. The stylopod gives rise to the most proximal limb skeletal element, the humerus. The
zeugopod forms the radius (anterior) and ulna (posterior). The distal autopod forms the
wrist bones (carpals), palm bones (metacarpals) and digit bones (phalanges). The scapula
and clavicle do not derive from the limb bud (by Zeller, Lopez-Rios et al. 2009).
Loss of AER in early stages of limb development (≅ E10) results in loss of all parts of
the limb, while its loss in the later stages (after E11) leads to the absence of only autopod.
Swapping AER between early and late stages of the development does not have any
impact on the normal development of the limbs which demonstrates the permissive but
not instructive role of AER. Specific expression of FGFs in AER conducts the outgrowth
of the bud and early development of limb in the vertebrates (Zeller, Lopez-Rios et al.
2009).

52

Fig. 15. The progress zone (PZ) model was formulated to explain the limb skeletal
phenotypes that result from the manipulation of chicken limb buds. It was proposed
that the mesenchyme underlying the AER contains unspecified progenitors (the PZ is
indicated by black stripes), the fates of which are controlled by AER signals. As limb bud
outgrowth progresses distally, proximal cells no longer receive AER signals. The time of
their 'exit' from the PZ determines their proximodistal (PD) identity (that is, there is
'clock-type' specification). Mesenchymal cells that exit early generate proximal elements,
whereas cells that remain in the PZ for longer form more distal structures (by Zeller,
Lopez-Rios et al. 2009)
To identify the importance of FGFs in the development of limb, AER was replaced
by beads covered with FGFs (Niswander, Tickle et al. 1993; Fallon, Lopez et al. 1994).
These beads are able to induce a complete limb formation in the absence of the AER
itself. Among different Fgfs, Fgf9, 17 and 4 are expressed in posterior AER and Fgf8 is
expressed along the AER. Among these FGFs, only Fgf9 and Fgf17 null mice are viable.
Due to early embryonic lethality, conditional knock out of Fgf4 and Fgf8 is required to
study the role of these two FGFs in the later stages of limb development. Therefore, using
Msx2-cre, expressed specifically in the ectoderm, the inactivation of Fgf8 has shown a
more severe phenotype in the limb than the inactivation of Fgf4 (Moon, Boulet et al.
2000; Boulet, Moon et al. 2004). Also, in the Fgf8 conditional knockout, while the size
53
of the limb is drastically reduced, the limbs are not completely lost suggesting a reduction
in the progenitor pool of the limb skeleton (Boulet, Moon et al. 2004).
Three different models were proposed for proximal-distal axis formation in the limb.
The first model was called the “progress-zone model” (Fig. 15). Based on this model,
AER induces the underlying mesenchyme to proliferate and form the proximal-distal axis
of the limb (Saunders 1948; Summerbell, Lewis et al. 1973). In the progress zone model,
while AER maintains undifferentiated and proliferative condition in the progress zone by
expressing FGFs, molecules such as FGF10 and BMP4 from the progress zone preserve
the active status of the AER. This interaction results in the growth promoting an
epithelial-mesenchymal feedback loop which main players are FGF8 from AER and
FGF10 from mesenchyme (Ohuchi, Nakagawa et al. 1997). While this model explains
some of classical experiments explained earlier, it does not describe the formation of
proximal parts of limb in the absence of AER at the later stage of the limb development.
The second model proposed is called the “two-signal” model in which two signals,
one from the AER and one from the proximal mesenchyme induce limb formation (Fig.
16). In this model, the signal from the AER instructs the distal mesenchyme of the limb
bud to become distal limb such as autopod and zeugopod. On the other hand, retinoic
acid (RA) from proximal mesenchyme induces proximal cell fate in the limb bud
(Mercader, Leonardo et al. 2000); (Capdevila, Tsukui et al. 1999). Although primary
molecular studies confirmed this model, later it has been shown that RA inhibits Fgf8
expression in intermediate mesenchyme which raises some doubts about the role of RA
in the induction of proximal limb (Zhao, Sirbu et al. 2009).
54
Fig. 16. The two-signal model.
Based on molecular analysis of
chicken limb bud development.
During the onset of limb bud
development, the proximal region
(blue) is probably specified by
retinoic acid (RA) signaling from
the flank, and the distal region
(orange) is specified by AER-
derived fibroblast growth factor
(AER-FGF) signaling. The
zeugopod arises from the more
proximal distal cells, and the
autopod primordia is formed by the
most distal mesenchymal cells (by
Zeller, Lopez-Rios et al. 2009).
The next proposed model is called the “differentiation front” model (Fig. 17). In this
model, it is suggested that early limb mesenchyme is fated to become proximal limb by
default, however, later AER gives instruction to the mesenchyme to become distal (Tabin
and Wolpert 2007). Also, during early stages of limb development, FGF from AER
interacts with ectodermal Wnt to maintain the distal mesenchyme in undifferentiated and
proliferative states (ten Berge, Brugmann et al. 2008). This model also fails to explicate
the molecular data obtained from in situ hybridization and microarray data.
55

Fig. 17. The differentiation-front model postulates that PD identities are determined as
the proliferating mesenchyme leaves the undifferentiated zone — that is, when the
mesenchyme is no longer under the influence of AER-FGF signaling. After cells have
crossed the differentiation front (blue wavy line) they only express genes that mark the
identity of a particular segment (for example, Meis1 expression in the stylopod territory,
homeobox A11 (Hoxa11) expression in the zeugopod territory and Hoxa13 expression in
the autopod territory) (by Zeller, Lopez-Rios et al. 2009)

Another critical area in posterior mesenchyme of the limb bud which determines the
anterior-posterior axis is called Zone of Polarizing Activity (ZPA). The main player in
this zone is sonic hedgehog (SHH). Grafting of ZPA into the anterior mesenchyme leads
to mirror image digit formation (Saunders 1948; Tickle 1981). The model that explains
the role of ZPA in AP axis formation is called the “French-flag” model (Wolpert 1969).
In this model, which was first proposed by Wolpert forty years ago, SHH as a morphogen
establishes a gradient with three different thresholds on the AP axis and leads to the
formation of different size digits (Riddle, Johnson et al. 1993). It has been shown that
GLI3, RA, TBX from flanks and FGF8 from AER determine the pattern of Shh
expression in the limb (te Welscher, Fernandez-Teran et al. 2002).
56
Fig. 18. The French-flag model. a. the French-flag model can explain the results of
manipulating the chicken limb bud organizer (zone of polarizing activity (ZPA), which is
located in the posterior limb bud mesenchyme). This model proposes that the ZPA
secretes a morphogen that diffuses across the limb bud to generate a spatial gradient. The
identities of the three digits (likened to the colors of the French flag) are specified by
threshold levels of the morphogen. b. The gene network that restricts the activation and
maintenance of Sonic hedgehog (Shh) expression to the posterior limb bud mesenchyme.
c. Specification of anteroposterior identities by a spatial and temporal gradient of SHH
signalling46, 47. As cells cease to express Shh, they exit the ZPA. The expanding
population of cells derived from Shh-expressing cells (Shh descendants) displaces non-
ZPA cells (which are specified by long-range SHH signaling) towards the anterior. Shh
descendants give rise to the ulna, digit 4 and digit 5, and contribute to digit 3. Cells that
give rise to digit 2 and parts of digit 3 are specified by long-range SHH signaling. The
humerus, radius and digit 1 are specified in a SHH-independent manner. 5′-HOXD, 5′-
located homeobox D; FGF8, fibroblast growth factor 8; GLI3R, repressor form of GLI3;
HAND2, heart and neural crest derivatives 2; RA, retinoic acid; TBX, T-box (by Zeller,
Lopez-Rios et al. 2009)
Although the models discussed above have explained the primary observations to
some extent, each of them is unable to explain the molecular data. Therefore, Tabin and
Wolpert proposed a new model which is a combination of previously known models
(Tabin and Wolpert 2007). In this model, ‘progress zone’ is called ‘undifferentiated zone’
which is a 200 µm- wide zone located under the AER. As explained in the two-signal
model, this model confirms the existence of two signaling centers; one at the AER which
57
controls the proliferation and maintenance of the undifferentiated cells and another signal
from flanks which induces the specification of progenitor cells to the proximal cells or
stylopod. Therefore, the undifferentiated zone on one side is restricted to the AER and on
the other border touches the proximal specified cells. As the undifferentiated zone
proliferates under the influence of AER, cells leave the undifferentiated zone and enter
the area which is under the control of proximal signals and become differentiated. This
process continues until the cells completely exit the zone where the signals from the
flanks can be received. At this point, ZPA acts as another signaling center that helps AER
in development of distal part of the limb, zeugopod and autopod by regulation of many
important molecules at this stage. All together, the latest model introduced by Tabin and
Wolpert seems to be the most sophisticated model that can explain to some extent both
classical and molecular data (2007).
Digit formation is the result of very stringent molecular signaling that changes
gradually during limb development. In general, although AER is critical for PD axis and
ZPA is required for AP axis, ZPA and AER maintain each other and the interaction
between these two molecular zones is important for both axis formation. As shown in
figure 1.17 while Shh is regulated by FGFs from AER, SHH regulates expression of
Gremlin1 in the intermediate mesenchyme which inhibits BMP4 and consequently
activates FGFs in the AER. This loop, which is called the ‘interlinked feedback loop’, is
required for the maintenance of the proliferating progenitor cells in the mesenchyme and
the growth of the limb bud (Tickle 1981) (Fig. 19).
58

Fig. 19. The interlinked signaling feedback loops that operate at each stage are shown
as solid lines. Broken lines indicate inactive loops. a. Initiation phase: BMP4 upregulates
Grem1 expression in a fast initiator loop (~2 h loop time). Shh expression and signaling
are activated independently of GREM1 and AER-FGFs. b. Propagation phase: the
establishment of loops that control the distal progression of limb bud development. SHH
predominantly upregulates Grem1 expression. GREM1 reinforces AER-FGF and zone of
polarizing activity-derived SHH (ZPA-SHH) signaling by an epithelial–mesenchymal
feedback loop (with a loop time of ~12 h). The activity of the fast BMP4–Grem1 initiator
module is low. However, this low BMP activity controls the length of the AER (not
shown). c. Termination phase: the widening gap between ZPA-SHH signaling and the
Grem1 expression domain, together with the onset of AER-FGF-mediated inhibition of
Grem1, terminates the signaling system. As a consequence, BMP4 activity is likely to
increase again. The mesenchymal bone morphogenetic protein (BMP4; light blue), sonic
hedgehog (SHH; red), gremlin 1 (GREM1; purple) and apical ectodermal ridge-derived
fibroblast growth factor (AER-FGF; green) expression domains during mouse limb
organogenesis are indicated schematically (by Zeller, Lopez-Rios et al. 2009)

59
Chapter 1, Part 1
Terminal End Bud Maintenance in Mammary Gland is
dependent upon FGFR2b Signaling
*

Chapter 1, Part 1 Abstract
We previously demonstrated that Fibroblast Growth Factor 10 (FGF10) and its
receptor FGFR2b play a key role in controlling the very early steps of mammary gland
development during embryogenesis (Mailleux et al., 2002; Veltmaat et al., 2006).
However, the role of FGFR2b signaling in postnatal mammary gland development is still
elusive. In the mammary gland of young virgin female mice, we show that FGF10 is
expressed at high level throughout the adipose tissue whereas its main receptor FGFR2, is
found mostly in the epithelium. Using a rtTA transactivator/tetracycline promoter
approach allowing inducible and reversible attenuation of the FGFR2b pathway
throughout the adult mouse upon addition of doxycycline, we are now reporting that
FGFR2b signaling is also critical during postnatal mammary gland development.

*
This paper was published in Developmental Biology 317(2008) 121-131. Sara Parsa
1,2,#
,
Suresh K Ramasamy
1,#
, Stijn De Langhe
1
, Varsha V. Gupte
1
, Jody J. Haigh
3
, Daniel
Medina
4
, Saverio Bellusci
1,2
(
1
Developmental Biology Program, Saban Research Institute
of Children’s Hospital Los Angeles, Los Angeles, CA 90027,USA,
2
Department of
Pathology, Keck School of Medicine, University of Southern California, Los Angeles,
CA 90089, USA,
3
Department of Molecular Biomedical Research, UHent, Ghent,
Belgium,
4
Department of Cellular and Molecular Biology, Baylor College of Medicine,
Houston, TX 77030.
#:
These authors contributed equally to this work. Correspondence:
sbellusci@chla.usc.edu)

60
Ubiquitous attenuation of FGFR2b signaling in the postnatal mouse for 6 weeks starting
immediately after birth is not lethal and leads to minor defects in the animal. Upon
dissection of the mammary glands, a 40% reduction in size compared to the WT control
is observed. Further examination shows a rudimentary mammary epithelial tree with
completely absent terminal end buds (TEBs), compared to a well-branched structure
observed in wild type. Transplantation of mammary gland explants into cleared fat pad of
wild type mouse recipients indicates that the observed abnormal branching results from
defective FGFR2b signaling in the epithelium. We also demonstrate that this rudimentary
tree reforms TEBs and resumes branching upon removal of doxycycline suggesting that
the regenerative capacities of the mammary epithelial progenitor cells were still
functional in spite of long-term inactivation of the FGFR2b pathway. At the cellular
level, upon FGFR2b attenuation, we show a marked increase in apoptosis associated with
a decrease in the proliferation of the mammary luminal epithelium. We conclude that in
young virgin female mice, there is a differential requirement for FGFR2b signaling in
ductal vs. TEBs epithelium. FGFR2b signaling is crucial for the proliferation of the
mammary luminal epithelial cells, but does not affect the regenerative potential of the
mammary epithelial progenitor cells.

61

Chapter 1, Part 1 Introduction
Mammogenesis in mouse starts at about embryonic day 10.5 (E10.5) with the
formation of two mammary lines running in an antero-posterior direction ventrally
between fore- and hind limbs, one line along each flank of the embryo (Veltmaat et al.,
2003; 2004). At around E11.5, lens-shaped multilayered ectodermal structures (called
placodes) can be detected slightly elevated above the surrounding ectoderm presumably
along each mammary line at five reproducibly precise positions. These mammary
placodes subsequently transform into bulbs of epithelial cells that are morphologically
distinct from the surrounding epidermis. At around E15.5, each bud elongates to form a
sprout, invading the underlying fat pad precursor. Each sprout forms a lumen, which
opens on the surface of the skin, where the nipple forms concurrently by epidermal
invagination. At about E16, the first ramifications of the sprouts occur, and by E18.5 the
sprouts have developed into small, arborized glands. After birth, the gland grows
isometrically with body growth until puberty, when hormonal influences induce a
differential growth spurt in the mammary gland (for review see Veltmaat et al., 2003).
We have previously reported the crucial role played by Fibroblast Growth Factor 10/
Fibroblast Growth factor Receptor 2b (FGF10/FGFR2b) in the initial steps leading to the
formation of the mammary placodes. FGF10, produced by the somites, functions
upstream of Wnt signaling to control the formation of the mammary line from which four
of the five mammary placodes will form (Mailleux et al., 2002; Veltmaat et al., 2004;
Veltmaat et al., 2006). We have also reported that in Fgfr2b null embryos, the remaining
62
mammary buds (mammary bud 4) regressed due to decreased proliferation and increased
apoptosis in the mammary gland epithelium. This study therefore suggested that the
FGFR2b pathway might be crucial for survival and proliferation of the mammary
epithelial progenitors during postnatal development.
Indeed, an important role for FGFR2 signaling during postnatal mammary gland
(MG) development is suggested by the fact that Fgfr2 expression is maximal in mature
virgin mice, declines during pregnancy and lactation, but rises after weaning (Pedchenko
and Imagawa, 2000). The rise in Fgfr2 mRNA in the virgin animal corresponds to a
significant increase in the branching of the mammary epithelial tree. During ductal
development, the genes encoding the two main FGFR2b ligands, FGF10 and FGF7, are
expressed at a ratio of 15 to 1, respectively (Pedchenko and Imagawa, 2000). The
expression of Fgfr2 and its associated ligands in the MG during ductal development
suggests an important role for the FGFR2b pathway in post-natal growth of the mammary
epithelium. Such a positive role for FGFR2b signaling in epithelial growth is re-enforced
by the fact that FGF10, which mostly binds to FGFR2b, has been described as an
oncogene in breast cancer cells (Theodorou et al., 2004). In addition, FGFR2 is amplified
and overexpressed in breast cancer (Grose and Dickson, 2005; Moffa and Ethier, 2007).
Mutations in FGFR2 have recently been strongly associated with a higher risk of breast
cancer in postmenopausal women with no previous family history of breast cancer
(Hunter et al., 2007).
The role of FGF10/FGFR2b signaling in the epithelial/mesenchymal interactions that
characterize postnatal MG development is demonstrated through the analysis of the MG
phenotype of transgenic mice allowing inducible and reversible attenuation of the
63
FGFR2b pathway throughout the whole adult mouse upon addition of doxycycline. We
are now reporting that FGFR2b signaling in mammogenesis is not only critical during
embryogenesis but also during postnatal development.
Chapter 1, Part 1 Material and Methods
Analysis of LacZ expression
Fgf10/LacZ expression was monitored in Fgf10
LacZ
mice (Kelly et al., 2001) by b-
galactosidase activity using whole-mount and histological revelation as described by
Kelly et al. (1995). Mammary glands (MGs) from 3 weeks old virgin females were fixed
2 hours in 4% PFA. Following the fixation, the MGs were washed in 1x PBS and stained
in X-gal solution. A similar protocol was carried out for MG from Rosa26 mice
(Zambrowicz et al., 1997).
Generation of rtTA; tet(O)sFgfr2b animals
CMV-Cre mice (Schwenk et al., 1995) were crossed with rtTA
flox
mice (Belteki et al.,
2005) to generate rtTA mice expressing rtTA from the Rosa26 promoter in every single
cell of the body. rtTA mice are now crossed with tet(O)soluble Fgfr2b mice (Hokuto et
al., 2003) to generate double transgenic rtTA; tet(O)soluble Fgfr2b mice. These mice are
on the CD1 mixed genetic background and allow inducible and reversible attenuation of
the FGFR2b pathway in the embryo or mouse simply by feeding the mice with
doxycycline containing food (Rodent diet with 0.0625 % Doxycycline, Harlan Teklad
TD01306). The reversibility of the phenotype is studied by putting the treated mice on
normal food. Genotyping of each allele was done as previously described (Schwenk et
al., 1995; Belteki et al., 2005; Hokuto et al., 2003).
64

Antibodies
MG from 3 weeks old female mice were fixed in 4% PFA overnight and stored in
70% ethanol. The tissues were embedded in paraffin and 5 µm longitudinal sections were
made. IHC was performed with the Envision kit from Dako cytomation. FGFR1 antibody
(Flg, 1:200, Santa Cruz Inc.) and FGFR2 (Bek, 1:200, Santa Cruz, Inc) were incubated
overnight at 4ºC as previously described (Sala et al., 2006). MG paraffin sections were
treated with Cy3 conjugated-monoclonal anti a-SMA Ab (1:200, Sigma, clone 1A4) as
previously described (De Langhe et al., 2005). Photomicrographs were taken using a
Leica DMRA fluorescence microscope with a Hamamatsu Digital CCD Camera.
Proliferation
One of the two MGs number 4 from P35 WT and double transgenic animals (n=3 for
each) non-induced with dox (control group) were surgically removed and the operated
mice were put on doxycycline food for 1 week to induce gene expression. Similar
experiments were also carried out with P60 double transgenic animals (n=3) to assess the
role of FGFR2b signaling in late puberty. After a week, the animals were sacrificed and
the second mammary glands 4 were removed (experimental group). The mammary
glands were fixed overnight in paraformaldehyde, rinsed in PBS twice for five minutes,
transferred to 70% EtOH overnight and stored in 100% EtOH. The samples were then
embedded in paraffin and sections (5 µm) were cut. The PCNA staining kit (Zymed Labs
Inc. 93-1143) was used for immunostaining. The total number of cells in the epithelium
as well as the number of PCNA positive cells in the epithelium were scored in six
65
photomicrographs (64x magnification), in random portions of a section of three
independent mammary glands in the experimental and control group. A total number of
600 cells were counted per sample. The significance in proliferation in mammary glands
between control and experimental group was evaluated by one-tailed paired t-test. P
values less than 0.05 were considered to be statistically significant. To assess the effect of
inactivating FGFR2b signaling on the overall ramification process, mammary glands
from the control and experimental group were also stained with Carmin Red.
Analysis of cell death
Apoptotic cells on 5 mm paraffin section of P35 and P38 (days on dox) MG were
detected by the incorporation of terminal deoxynucleotidyltransferase mediated dUTP
nick-end labeling (TUNEL) using the “In Situ Cell Death Detection, Fluorescein” kit
(Roche Applied Science) as recommended by the manufacturer. The total number of cells
in the epithelium as well as the number of TUNEL positive cells in the epithelium were
scored in five photomicrographs (64x magnification), in random portions of a section of
three independent mammary glands in the experimental and control group. A total
number of 500 cells were counted per sample.
Whole mount Carmin Red staining
The mammary gland epithelium was stained using Carmin Red as previously
described (Faraldo et al., 1998).

66

Quantification of the mammary gland size
Mammary glands from WT and mutant animals (treated with dox from P0 to P42,
n=3) were spread out flat on a glass slide and photographed at the same magnification
(0.6 x). The surface occupied by the MG (in arbitrary units) in the total field was
measured using Metamorph. The significance in the size of the mammary glands between
WT and mutant was evaluated by one-tailed paired t-test. P values less than 0.05 were
considered to be statistically significant.
Grafting of wild type and mutant mammary gland in cleared mammary fat pad of
NOD/SCID mice
The mammary gland 4 from WT (n=2) and Mutant (n=2) at P21 were freshly
dissected and individually cut in 1mm
3
explants (8-10 pieces/MG). Explants were then
transplanted into cleared mammary fat pads of NOD/SCID mice (Medina, 1996). In these
experiments, 19 days old females were used as transplant recipients. The endogenous
mammary epithelium was surgically removed from the fourth inguinal glands to provide
a cleared mammary fat pad. Mutant and wild-type mammary gland explants were
transplanted separately into contralateral glands of each recipient to ensure an identical
host environment. 1 week after surgery, the animals were put either on doxycycline food
(n=10 animals) or normal food (n=8 animals). The mice were sacrificed 9 weeks after
surgery and the fourth inguinal mammary glands dissected. The epithelium was stained
using Carmin Red as previously described (Faraldo et al., 1998). The number of cleared
fat pad filled with mammary gland epithelium was quantified.
67
Chapter 1, Part 1 Results
Expression of FGF10 and associated receptors FGFR1 and FGFR2 in the mammary
gland of young virgin females.
Immunohistochemistry in mammary glands of three weeks-old virgin females was
carried out to determine the expression of FGFR1 and FGFR2. The antibodies used
recognize both the “b” isoform, usually expressed in epithelia and the “c” isoform
generally expressed in mesenchyme. FGF10 binds to the “b” isoform of both receptors.
Fig. 20-A shows that FGFR1 is detected at high level in the adipose tissue but is present
only at very low level in the epithelium. By contrast, FGFR2 is strongly detected in the
ductal epithelium and expressed at lower levels in the adipose tissue. FGFR2 is also
expressed in the epithelium of the terminal end buds located at the tips of the ducts in 5
weeks-old MG (Fig. 20-C).
We also investigated the expression of Fgf10 in mammary glands of three weeks old
virgin females using Fgf10
LacZ
mice (Kelly et al., 2001). Previous studies have shown that
nuclear b-galactosidase expression in this mouse strain is a bona fide reporter for Fgf10
expression (Kelly et al., 2001; Mailleux et al., 2005; Sala et al., 2006; Veltmaat et al.,
2006). We found that Fgf10/b-galactosidase was expressed at high level in the adipocytes
(Fig. 20-D-F) confirming our previous studies showing embryonic Fgf10 expression in
the fat pad (Mailleux et al., 2002), an embryonic tissue which will give rise to the white
adipose tissue.

68
These data therefore suggest that FGF10 produced in large amount by the adipocytes
acts in a paracrine fashion on the epithelium expressing FGFR2.

Fig. 20. Expression of FGFR1, FGFR2 and FGF10 in WT Mammary gland.
Expression of FGFR1 (A) and FGFR2 (B) in 3 weeks-old MG by Immunohistochemistry.
(C) Expression of FGFR2 is also detected in the terminal end bud epithelium in 5 week-
old female mice. (D) X-gal staining of 3 weeks old MG from Fgf10
LacZ
animals. (E)
Higher magnification of (D). Note the blue dots in the adipocytes corresponding to
nuclear ß-galactosidase staining. Scale bar: (A, B, C, F) 120 mm; (D) 1000 mm; (E) 200
mm.

Attenuation of FGFR2b during embryonic and postnatal stages demonstrate a
differential requirement for FGFR2b signaling
In order to determine the role of FGFR2b signaling during postnatal stages, we have
generated transgenic mice allowing doxycyclin-based inducible and reversible
attenuation of FGFR2b signaling. CMV-Cre mice, a Cre-transgenic mouse strain allowing
ubiquitous deletion of loxP-flanked gene segments including deletion in germ cells
69
(Schwenk et al., 1995) have been crossed with rtTA
flox
mice (Belteki et al., 2005). After
recombination of the rtTA
flox
allele, the transactivator rtTA is constitutively expressed
from the Rosa 26 locus in every cell of the mouse including the germ cells. This
constitutive rtTA mouse line is then crossed with tet(O)sFgfr2b responder line (Hokuto et
al., 2003) to generate double transgenic animals allowing ubiquitous expression of
soluble FGFR2b, acting as a dominant negative, from every cell of the mouse using the
rtTA transactivator/tetracycline promoter system (Gossen and Bujard, 1992). We have
validated that our double transgenic system is functional by inducing the expression of
soluble FGFR2b from embryonic day 0.5 (E0.5) through E13.5. Our results indicate that
double transgenic embryos exposed to doxycycline exhibit the phenotype of the
previously reported Fgfr2b
-/-
embryos (De Moerlooze et al., 2000). The rtTA;
tet(O)sFgfr2b double transgenic embryos not exposed to doxycycline as well as the
single transgenic embryos exposed to doxycycline were phenotypically identical to wild
type embryos (data not shown), indicating that there is no leakiness in the expression of
soluble FGFR2b in our double transgenic embryos in absence of doxyclycline and that
there is no toxic effect of doxycycline on embryonic development.
A Comparison of Fig. 21-A (showing a single transgenic embryo identical to wild
type) with Fig. 21-B indicates that double transgenic embryos exposed to doxycycline
fail to form limbs and lungs (compare insets a and b). In addition, we observed a curly
tail, which is a phenotype present in Fgfr2b
-/-
embryos (de Moerlooze et al., 2000) but not
in Fgf10
-/-
embryos (Sekine et al., 1999). These results confirm that FGFR2b signaling is
crucial during organogenesis. By contrast, expression of a dominant negative FGFR2b
during the postnatal period (P0 through P60) leads only to a few defects such as
70
Fig. 21. Validation of the
double transgenic (rtTA;
tet(O)soluble Fgfr2b)
embryos (A, B) and mice
(C-H). (A, B) Pregnant
females carrying single or
double transgenic
embryos have been put on
doxycycline-containing
food. Embryos at
embryonic day 13.5 from
doxycycline treated
pregnant females were
collected and genotyped to
confirm the presence of
both rtTA and
tet(O)sFgfr2b transgene.
Single transgenic embryos
were considered as wild
type as they had the same
phenotype. (A) Single
transgenic/WT embryo.
(a) H&E staining of a
frontal section going
through the lung. Note the
presence of a well-
developed lung. (B) By
contrast double transgenic
embryos exposed to
doxycycline had no limbs
(arrows) and a curly tail
(arrowhead). (b) H&E
staining of a frontal
section going through the
lung. Note the abnormal
lung comprised only of
the trachea and the first
two bronchi. (C) Double
transgenic adult female mice were put on normal food or on doxycycline-containing food
from birth (P0). 2 months old female mice are shown in (C). Note that the treated mouse
appears slimmer. (D) Weight quantification in 2 months-old wild type and mutant mice
show reduction in weight in both males (n=2 for each genotype) and females (n=2 for
each genotype) mutant mice. (E-H) Obvious macroscopic defects in 3 month-old
doxycycline-treated female mice included the absence of upper incisors and the degraded
lower incisors (compare insets in E and F) as well as the longer nails (compare G and H).
71
abnormally shaped incisors and elongated claws (Fig. 21-E-H). As during embryonic
stages, adult double transgenic mice not exposed to doxycycline are phenotypically
equivalent to wild type mice (data not shown) indicating that there is no leakiness in the
expression of soluble FGFR2b in post-natal stages. The mutant mice also appear
significantly slimmer compared to wild type animals (Fig. 21-C). Quantification of the
weight shows indeed a significant reduction in mutant (n=4) vs. wild type (n=4) animals
regardless of the gender (Fig. 21-D. P<0.03). Upon dissection of the mammary glands
from control (wild type) and double transgenic animals, we observed a 40% reduction in
the total size (measured in arbitrary units) of the mutant mammary glands (7451 ± 881 vs.
12477 ± 1123 in mutant vs. control, P<0.02) as well as reduced visceral fat (data not
shown). Our double transgenic system relies on the proper activity of the Rosa26
promoter in different tissues allowing expression of the transactivator rtTA. As a control
for our experiment, we tested whether the Rosa26 promoter was active by examining ß-
galactosidase activity in skin, limbs, gastrointestinal tract, heart, kidneys, lungs and
mammary glands of adult Rosa26 mice. In all organs examined (data not shown),
including the mammary gland (Fig. 22-A-C), we observed a robust and homogenous
expression of ß-galactosidase. We therefore validated the use of the Rosa26 promoter to
drive significant gene expression in major organs in adult mice.
Altogether, our data suggest that FGFR2b signaling is not essential for the
function/homeostasis of organs important for sustaining the vital functions of the mice
such as the lung, heart, digestive track or kidneys. However, discrete defects unravel a
function postnatally for FGFR2b in specific organs such as the incisors, the claws or the
mammary gland. In this work, we examined the potential mammary gland defects in
72
rtTA; tet(O)sFgfr2b double transgenic animals. In order to obtain consistent results, we
have focused our analysis on mammary gland 4 even though similar results were
obtained with the other pairs of mammary gland (data not shown). The choice of this
mammary gland was dictated by the easy accessibility and dissection of this gland.
Inactivation of FGFR2b signaling immediately after birth leads to absence of
branching of the mammary epithelium
LacZ expression in 8 weeks-old MG from Rosa26 mice shows that the Rosa26
promoter is active throughout the mammary epithelium both in proximal (ductal) and
distal (tip) position. LacZ expression was also detected throughout the white adipose
tissue, but at lower level. The expression of soluble Fgfr2b indirectly from the Rosa26
promoter is therefore likely to affect homogeneously the mammary epithelium. We
carried out Carmin Red staining of wild type and double transgenic animals exposed to
doxycycline postnatally for various period of time. Fig. 22 shows the degree of
ramification in wild type (Fig. 22-D-F) and rtTA; tet(O)sFgfr2b (Fig. 22-G-I) mammary
glands after 42 days of treatment with doxycycline (from P0 to P42). In the wild type MG
(n=3), a well-developed mammary tree extending beyond the lymph node used as a
landmark, is observed with prominent terminal end buds (TEBs). By contrast, the
epithelial tree of the mutant MG (n=5) did not reach the first lymph node but rather
stayed close to the nipple region. Further examination indicated a poorly branched
structure with rudimentary, thin ducts and no visible TEBs. Histological analysis revealed
a simplified luminal epithelium (compare insets in Fig. 22-I and F). Normal SMA
expression in mutant MG indicates that the myoepithelial layer is not significantly
affected (inset in Fig. 22-I).
73

Fig. 22. FGFR2b attenuation leads to the formation of a rudimentary mammary
epithelial tree. (A-C) Rosa26 MG from P120 female stained with X-gal shows strong ß-
galactosidase acivity throughout the mammary epithelium and at lower level in the white
adipose tissue. (D-I) Carmin Red staining of P42 MG in WT and mutant MG exposed to
dox from P0 through P42. (D) WT MG at P42 showing extensive branching of the
epithelial tree. (E) High magnification of (D) showing the TEBs. (F) Higher
magnification of the TEBs shown in (E). Insets show H&E staining of the ductal
epithelium as well as SMA expression by IF to label the myoepithelial cells. (G) MG
from rtTA; tet(O)sFgfr2b females exposed to doxycycline for 42 days since birth. (H)
Higher magnification of the rudimentary tree shown in (G). Note the absence of TEBs at
the extremity of the ducts. (I) Higher magnification of the ducts shown in (H). Insets
show H&E staining of the mammary duct with a simplified epithelium. SMA is normally
in the mutant MG. Scale bar: (A, D, G) 1000 mm; (B, E, H) 400 mm; (C, F, I) 80 mm.
Insets in F and I, 300 mm.
74
We also examined the role of FGFR2b signaling in the mammary gland during
gestation and lactation. We compared pregnant double transgenic (n=2) and WT (n=4)
females (which were crossed with wild type males). Doxycycline treatment of pregnant
double transgenic mice (from E7 through E14) led to a slight reduction in the formation
of the alveoli compared to the WT MG (Supplementary Fig. 27-A-F). However, when
double transgenic mice are treated during the early lactation period (from P0 through P7),
no major defects can be detected compared to WT MG (Fig. 27-G-L). Supporting the
conclusion that lactation is not affected, the difference in the weight of the P7 pups in the
WT and double transgenic group was not significant (6.1g ± 0.2g vs. 6.3g ± 0.4g, for
pups in WT females vs. double transgenic, P=0.14).
Altogether, our results indicate that FGFR2b signaling is crucial for mammary
epithelial branching during early-mid puberty. It also suggests that ductal epithelium,
even though affected, is less dependent on the loss of FGFR2b signaling that the
epithelium in the terminal end buds.
The rudimentary mammary tree resumes branching upon removal of doxycycline
Upon removal of doxycycline, the rtTA/Tet on system allows reversibility in the
expression of the gene under the control of the tetracycline promoter (Gossen and Bujard,
1992). We have taken advantage of this feature to further evaluate the regenerative
capacities of the epithelial progenitor cells in mutant mammary glands of female mice
exposed to doxycycline for 42 days (from P14 to P56). In this experiment, we have
chosen to induce expression of doxycycline at P14 instead of P0, as we wanted to
determine also the effect of expressing soluble FGFR2b on the mammary epithelium:
75
namely, growth arrest vs. regression of the mammary epithelial tree. Another advantage
linked to mouse mammary glands is that they come in pairs. One of the two MG of a
given pair can therefore be ressected for evaluation without sacrificing the animal. At a
later time, the second MG can be isolated and compared to the first MG originating from
the same pair and from the same animal, therefore limiting sample-to-sample variability.
Fig. 4 summarizes our experimental plan. In wild type mice (n=3), we isolated the
first MG at P14 and the second MG at P84. In double transgenic mice (n=3), we started
the treatment with doxycycline at P14 and isolated the first MG at P56. We also isolated
independently MGs from P56 WT animals as age-matched controls. The decision to
isolate P14 WT MG was justified, as this is the stage when the mutant mice were put on
doxycycline. This will therefore allows us to determine whether the development of the
rudimentary tree was simply arrested immediately at the time of doxycycline treatment or
if further regression occurred. We therefore compared mutant P56 MG (Fig. 23-D-F)
with wild type P14 and P56 MG (Fig. 23-A-C and G-I, respectively). Our results indicate
that, in spite of a well-developed white adipose tissue, the mutant mammary gland (at
P56) exhibits a significant decrease in lateral ramifications compared to the control MGs
at P14 and P56. Interestingly, the main ducts are still maintained in the mutant MG
suggesting again a differential requirement for FGFR2b signaling in the ductal vs.
terminal end bud epithelium. This result suggests that attenuation of FGFR2b signaling in
already formed post-natal mammary gland not only arrests the growth of the epithelial
tree but also leads to its partial regression.
In order to determine whether the regenerative capacities of the mammary epithelial
progenitors were still functional in the rudimentary epithelial tree observed in the mutant
76
MG, the operated mutant animals (P56) were put on normal food (with no doxycycline)
for 28 days (until P84). At the end of this period (P84), the second MG4 was surgically
removed from wild type and mutant animals. Fig. 23M-O shows that upon doxycycline
removal, the mutant MG epithelium can indeed reform TEBs and resume branching
beyond the lymph node. However, close comparison with the corresponding wild type
MG (Fig. 23-J-L) indicates that the degree of lateral branching is reduced in mutant MG.
The P84 mutant MG resembles to a MG from a younger animal, such as the one
presented in Fig. 22-D (P42 WT virgin female). This result therefore indicates the
FGFR2b pathway in postnatal MG is not critical for the maintenance of the regenerative
capacity of the epithelial MG progenitors cells.

77
Fig. 23. FGFR2b pathway in postnatal MG is not critical for the maintenance of the
regenerative capacities of the epithelial MG progenitor cells. Analysis of MG
development by Carmin Red staining. (A-C) P14 Wt MG showing that the mammary tree
did not reach the lymph node (L). (D-F) MG from rtTA; tet(O)sFgfr2b female mouse
treated with dox from P14 through P56 showing that the growth of the mammary tree
was arrested. In comparison, the white adipose tissue was not significantly affected. Note
the marked reduction in lateral branches and the absence of TEBs. (G-I) Corresponding
P56 age matched WT control MG. (J-L) P84 Wt MG showing that the mammary tree
grew beyond the lymph node. Note also the abundant lateral ramifications in the
mammary tree. (M-O) MG from a rtTA; tet(O)sFgfr2b female mouse treated with
doxycycline from P14 through P56 and then put on normal food up to P84. Note that the
TEBs are present and that the growth of the mammary tree has resumed. Scale bar: (A, D,
G, J, M) 1000 mm; (B, E, H, K, N) 400 mm; (C, F, I, L, O) 80 mm.

78
(Fig. 23 continued)

79
Attenuation of FGFR2b signaling leads to a drastic decrease in the proliferation of
the mammary luminal epithelium
FGFR2 expression in the mammary epithelium (Fig. 20) as well as the resulting
rudimentary epithelial mammary tree upon FGFR2b attenuation (Fig. 22 and 23) suggest
that FGF10 acts mostly on the epithelium to control its survival and proliferation. To
determine that this in indeed the case, we have surgically removed one of the two MG4 in
P35, mid–puberty, wild type and mutant females (n=3) which were never fed with
doxycycline and exposed the operated animal to doxycycline food for one week. At this
time, the second MG4 (at P42) was surgically removed. We therefore obtained the
control and experimental mammary glands from the same animals. This allowed us to
reduce the number of animals needed for this study as well as reduce the variability due
to the mixed genetic background. Comparison of P35 (no dox, Fig. 24-A-C) vs. P42 (1
week on dox, Fig. 24-D-E) wild type MG indicates no changes in the appearance of the
mammary epithelial tree indicating that doxycycline treatment does not affect MG
development. In addition, mutant MG at P35 shows well-developed TEBs, a feature of
WT MG, therefore supporting our previous conclusion that there is no leakage in soluble
Fgfr2b expression in absence of doxycycline. By contrast, mutant MG at P42 (one week
on dox) displays a specific regression of the TEBs (Fig. 24-J-L).
Our results also indicate similar expression of SMA in P35 MG compared to P42 MG
(data not shown), suggesting that the myoepithelial layer is not directly affected by
FGFR2b attenuation. We then examined the proliferation in the luminal epithelium of
mutant MG with or without doxycycline by PCNA. We observed a 65% decrease in
proliferation in P42 (Fig. 24-N) vs. P35 MG (Fig. 24-M) (16.1 ± 3.8 vs. 48.4 ± 2.1 % for
80
P42 vs. P35, respectively. P<0.001). Similar results were observed in late puberty (P60-
P67, Fig. S1). We observed a 83% decrease in proliferation in P67 (7 days on dox) vs.
P60 (no dox) mutant MG (Fig. 26) (5.9% ±1.6% vs. 34.0% ± 6.2% for P67 vs. P60,
respectively. P<0.001).
Our initial results indicated no difference in apoptosis after week treatment with
doxycycline either at mid or late puberty (data not shown). However, examination of
apoptosis in mutant MG after 3 days treatment with doxycycline (between P35 and P38)
indicated significant apoptosis (Fig. 24-O,P) (8.6% ± 2.8% vs. 2.4% ± 0.6% at P38 vs.
P35, P<0.01). Altogether our results demonstrate that FGFR2b signaling controls luminal
epithelial cell proliferation as well as survival.

81

Fig. 24. FGFR2b signaling attenuation leads to decreased proliferation of the
mammary epithelium. Analysis of the MG phenotype isolated from the WT or rtTA;
tet(O)sFgfr2b female non treated with dox (P35) and then treated with dox for 1 week
(P42). Analysis was done by Carmin Red staining, (A-C) Low and high magnification of
P35 MG from WT female non-treated with doxycyline showing a normally ramified
mammary tree. (D-F) P42 WT female treated with doxycyline for one week showing a
normally ramified mammary tree. (G-H) Low and high magnification of P35 MG from
rtTA; tet(O)sFgfr2b female non-treated with doxycyline showing a normally ramified
mammary tree. (J-L) P42 MG from rtTA; tet(O)sFgfr2b female treated for 1 week (from
P35) with doxycyline showing the main ducts with absent TEBs. (M-P) Corresponding
PCNA staining indicating decrease proliferation in the luminal epithelium in mutant MG
exposed to dox (P) vs. mutant MG not exposed to dox (O) or wild type controls (M,N).
Scale bars: (A, D, G, J) 1000 mm; (B, E, H, K) 400 mm; (C, F, I, L) 80 mm; (M, N) 60
mm; (O, P) 120 mm.
82
Branching of mutant mammary explants is severely impaired in cleared mammary
fat pad of recipient mice exposed to doxycycline
In order to confirm that the observed defects in the mutant mammary glands where
directly linked to the perturbation of the FGFR2b signaling in the mammary epithelium,
we carried out transplantation of explants from wild type and mutant 3 weeks-old MG
into 19 days-old cleared fat pads of NOD/SCID female mice. The recipient mice
containing the mutant or wild type explants were put on normal food for one week and
then put either on dox containing food or on normal food. The animals were sacrificed 9
weeks after transplantation and the development of the explants were examined. The fate
of the explants into the fat pads was divided into three possible outcomes; ramified
explants, explants present with no ramification and no explant visible. In the later case,
the lack of visible explants in the fat pad is likely the consequence of failure to implant
due to technical reasons.
For WT explants (no doxycycline), a total of 8 transplanted fat pads (TFP) were
considered. 87.5% of TFP show ramified explants, 0% of the TFP show explants with no
ramification and 12.5% of the TFP show no explants indicating failure of the explants to
engraft. Similar results were observed for WT explants exposed to doxycycline. For the
mutant explants without doxycycline, a total of 8 TFP were considered. 75% of the TFP
show ramified explants, 12.5% of the TFP show explants with no ramification and 12.5%
of the TFP show no explants. This confirms that there is no leakage in our soluble
transgenic system. Finally, for the mutant explants on doxycycline, a total of 10 TFP
were considered. 20% of the TFP show ramified explants, 60% of the TFP show explants
with no ramification such as the one shown in Fig. 25 and 20% of the TFP show no
83
explants. Altogether our results indicate a drastic decrease in the TFP with mutant
explants showing ramification in doxycycline vs. normal food (20% vs. 75%,
respectively) indicating that the effect of FGFR2b signaling on the mammary epithelium
is self-autonomous and that the impaired epithelial tree branching is not the consequence
of alterations in the white adipose tissue and/or in levels of hormones known to control
mammary branching morphogenesis such as estrogen, progesterone and prolactin.
84

Fig. 25. Evidence that FGFR2b signaling in post-natal mammary gland
development is epithelial cell autonomous. MG explants from WT and mutant animals
were transplanted in cleared fat pad of recipient animals. Animals were put on dox or
normal food one week after surgery and sacrificed 9 weeks after surgery. Fat pad were
stained with Carmin Red. (A,B) Recipient mouse is not exposed to doxycycline. Fat pad
from with WT explant shows presence of epithelial ramification. (C,D) Recipient mouse
is exposed to doxycycline. No difference with (A,B). (E,F) Recipient mouse is not
exposed to doxycycline. Fat pad from with mutant explant shows presence of epithelial
ramification. (G,H) Recipient mouse is exposed to doxycycline. Fat pad from with
mutant explant does not ramify. Scale bar: (A, C, E, G) 1000 mm; (B, D, F, H) 400 mm.

85
Chapter 1, Part 1 Discussion
An inducible and reversible expression system to attenuate FGFR2b signaling in the
postnatal stages
We are the first group to report the use of the rtTA/Tet system to attenuate the
FGFR2b pathway in the entire postnatal mouse. In contrast to the severe phenotype
observed when FGFR2b is knocked down during the embryonic stage, we could detect
only minor defects when FGFR2b is inactivated in the postnatal period. These defects
include defective incisors, longer claws and reduced white adipose tissue. Interestingly,
the incisors unlike the molars are constantly renewed in the postnatal mouse. Indeed, it
was previously suggested by Harada and colleagues that FGF10 maintains the dental
epithelial stem cell compartment in the incisors, not only during the embryonic stages,
but also postnatally (Harada et al., 2002; Yokohama-Tamaki et al., 2006). Our results are
directly confirming the important role for FGFR2b signaling during postnatal incisor
development. The longer claw phenotype observed upon attenuation of the FGFR2b
pathway is more surprising as usually loss of FGFR2b signaling leads to loss of tissue.
However, this phenotype is similar to the one displayed by mice where soluble FGFR2b
is expressed in the hair cortex using the Foxn1 promoter (Schlake, 2005). More work will
be needed to determine the mechanisms leading to the claw phenotype. Finally, the mice
under long-term treatment with doxycycline exhibit a significant decrease in the white
adipose tissue, illustrated by the reduced size of the mammary gland and visceral fat in
mutant animals (data not shown). Such decrease is not observed when mice are treated
for a week (data not shown). The role of FGFR2b in the formation of the white adipose
86
tissue during development was previously reported (Yamasaki et al., 1999). It has been
proposed that FGF10 controls adipogenesis by inducing the differentiation of the
progenitors for the adipocytes (called pre-adipocytes) into adipocytes (Sakaue et al.,
2002). More recently, it has been shown that FGF10 plays also a key role in the
proliferation of the pre-adipocytes (Asaki et al., 2004). Collectively, our results indicate
that long-term attenuation of FGFR2b signaling in postnatal mice leads to defects
consistent with an important role of FGFR2b in survival and/or proliferation of
progenitor and/or partially differentiated cells in the incisors and adipose tissue.
FGFR2b signaling plays a major role during postnatal mammary gland
development
The mammary gland is mostly composed of ducts that contain two differentiated cell
types, the luminal epithelial cells, which secrete milk protein, and the myoepithelial cells,
which are located at the basal surface of the luminal cells. Before the start of puberty, the
MG forms a rudimentary network of ductal epithelium. At the onset of puberty, under
the action of circulating hormones, unique structures called terminal end buds (TEBs)
form at the tips of the mammary ducts. These TEBs proliferate, ramify and actively
invade the adipose tissue to allow the formation of a complex branching structure. This
process takes place up to 10-12 weeks. After this developmental stage, the TEBs regress
(Sternlicht et al., 2006). Little is known about the signaling pathways controlling in vivo
the proliferation of the luminal cells, which are the cells mostly involved in breast cancer.
Using an inducible and reversible FGFR2b attenuation approach, we demonstrate a
differential requirement for FGFR2b signaling in the ductal vs. terminal end bud
epithelium. In particular, we show that the proliferation of the luminal epithelium is
87
dependent upon FGFR2b signaling. Such a positive role for FGFR2b signaling in
mammary epithelial growth is reinforced by the fact that FGF10 has been described as an
oncogene in breast cancer cells (Theodorou et al., 2004).
Could estrogen activity on the mammary gland epithelium be mediated by FGFR2b
signaling?
During puberty, the mammary epithelial tree develops rapidly in response to changes
in circulating hormones, including estrogen. The role of estrogen receptor a (Era) in
mammary gland development was previously established by Kourach and colleagues in
their genomic Era KO mouse model (Bocchinfuso et al., 2000; Korach et al., 1996).
These mice, which turned out to be hypomorphic for Era, displayed lack of MG
development beyond the pre-pubertal stage with absence of TEBs and a rudumentory
epithelial tree, However, later studies demonstrated that this phenotype could be rescued
by prolactin, estradiol or progesterone demonstrating that the phenotype was due to
abnormal pituitary and ovarian hormones in the mutant animals whereas Era function was
intact in the mammary gland. To address the function of estrogen signaling in the
mammary epithelium, recent studies described the mammary epithelial specific
inactivation of Era (Feng et al., 2007). The resulting mutant mammary gland displayed
an arrest in MG development beyond the pre-pubertal stage with absence of TEBs. Given
the phenotypic similarities between our mutant MG and the Era MG described
previously, it is tempting to speculate that estrogen signaling via Era in the epithelium
could regulate the expression of FGFR2b in this compartment. Interestingly, FGFR2
expression in the prostate epithelium, which is also under hormonal control for its
postnatal development, appears to be required for androgen-mediated tissue homeostasis
88
(Lin et al., 2007). Therefore by keeping the level of FGFR2b expression under control,
estrogen signaling would prevents mammary epithelial cells to respond excessively to the
otherwise high levels of FGF10 arising from the surrounding stroma. Further studies
using the available mutant mice and in vitro cell modeling will allow determining the
possible relationship between FGFR2b signaling acting locally during mammary gland
morphogenesis and the circulating hormones.
In conclusion, we report herein that FGFR2b signaling in mammogenesis is not only
critical during embryogenesis to control the induction of the mammary placodes (for
placodes 1, 2, 3 and 5), as well as to maintain the survival and proliferation of the
mammary epithelial progenitor cells in placode 4 (Mailleux et al., 2002) but also during
postnatal development to allow the formation of the TEBs as well as the control of
proliferation of the luminal epithelial progenitor cells.
89
Chapter 1, Part 1 Supplementary data
Fig. 26: FGFR2b signaling attenuation in late puberty leads to decreased
proliferation of the mammary epithelium. Analysis of the MG phenotype isolated from
the same rtTA; tet(O)sFgfr2b female non treated with dox (P60) and then treated with
dox for 1 week (P67). Analysis was done by Carmin Red staining, IF for SMA
expression and PCNA for proliferation. (A) P60 MG from rtTA; tet(O)sFgfr2b female
non treated with doxycyline showing a normally ramified mammary tree. (B) Higher
magnification of the mammary tree shown in (A). Inset shows expression of SMA in the
myoepithelial cells. (C) P67 MG from rtTA; tet(O)sFgfr2b female treated for 1 week
(from P60) with doxycyline no major changes in the overall branching pattern. (D)
Higher magnification of the mammary tree shown in (C). Note the regression of the
primary and secondary ducts. Inset shows normal expression of SMA in the
myoepithelial cells. (E) PCNA staining on P60 MG showing many labeled cells in the
epithelium of a main duct. (F) Higher magnification of the ductal epithelium shown in
(E). (G) PCNA staining on P67 MG showing a drastic reduction in the number of PCNA
positive cells in the ductal epithelium. (H) Higher magnification of the ductal epithelium
shown in (G). (A, C) 1000 mm; (B, D) 80 mm; (E, G) 300 mm; (F, H) 60 mm. Insets in
B and D. 300 mm.

90
(Fig. 26 continued)

91

Fig. 27: Role of FGFR2b signaling in the mammary gland during gestation and
lactation. Carmin Red staining. (A-C) WT and (D-F) rtTA; tet(O)sFgfr2b mutant MG
from pregnant females treated with doxycycline between E7 and E14. Note the slight
decrease in alveolar formation. (G-I) WT and (J-L) rtTA; tet(O)sFgfr2b MG from
lactating females treated with doxycycline from P0 through P7. No significant differences
are observed. Scale bar: A, D, G, H) 1000 mm; (B, E, H, K) 400 mm; (C, F, I, L) 80 mm.
92
Chapter 1, Part 2
FGFR2b Signaling regulates Luminal Epithelial Lineage
Formation during Postnatal Mammary Gland Development
*

Chapter 1, Part 2 Abstract
We previously reported that FGFR2b signaling pathway plays an important role in the
maintenance of terminal end buds (TEBs), which are dynamic structures at the tip of the
developing mammary tree. TEBs consist of transit amplifying cells (TACs), which are
developed from adult mammary stem cells. We demonstrated the ability of postnatal
mammary stem cell to survive in the absence of FGFR2b signaling. Here, we report the
critical role played by FGFR2b signaling in the regulation of luminal epithelial lineage
commitment of mammary stem cells in adult mouse. In this study, we used both inducible
and reversible gain (DTG-Fgf10) and loss (DTG-sFgfr2b) of function mouse models for
FGFR2b signaling. Histological analyses of the mammary glands of short time
doxycycline treated DTG-sFgfrr2b females display reduction in proliferation and
terminal differentiation of end ducts of mammary tree. On the other hand, overexpression
of Fgf10 leads to proliferation and increase of TAC layer surrounding the TEBs. The

*
The Manuscript for this study is under preparation. Sara Parsa
1,2
, Reza Tabatabai
2
,
Denise Al Alam
2
, and Saverio Bellusci
1,2,3
. (
1
Department of Pathology, Keck School of
Medicine, University of Southern California, Los Angeles, CA 90089, USA,
2
Developmental Biology and Regenerative Medicine Program, Saban Research Institute
of Childrens Hospital Los Angeles, Los Angeles, CA 90027, USA,
3
University of
Giessen Lung Center, Department of Internal Medicine II, Klinik strasse 36, 35392
Giessen Germany. Correspondence: sbellusci@chla.usc.edu)
93
fluorescent activated cell sorting (FACS) analysis of mammary epithelial stem/progenitor
cells isolated from mammary glands of 36-hour doxycycline treated DTG-Fgfr2b females
demonstrates a significant increase in bipotent stem cell population and decrease in
luminal progenitor cell population. On the other hand, FACS results on 7-day
doxycycline treated DTG-Fgf10 mammary epithelial cells show decrease in bipotent stem
cells and increase of luminal progenitor cells. Based on FACS results obtained from both
gain and loss of function mouse models, we conclude that FGFR2b signaling pathway is
crucial for luminal progenitor lineage commitment of bipotent stem cells in adult
mammary gland.
Chapter 1, Part 2 Introduction
Mammary gland (MG) is one of the rare organs in the body which development
occurs mostly postnatally. At birth, the MG primitive tree is composed of primary and
secondary branches (Watson and Khaled 2008). During the next 7 to 8 weeks, this
rudimentary mammary tree invades the mammary fat pad and undergoes extensive
growth. The leading edges of the epithelial tree at this invading phase are transit bulb-like
structures called terminal end buds (TEBs). TEBs appear around forth week after birth
and disappear as a result of hormonal changes at puberty (8 to 10 weeks) (Sternlicht,
Kouros-Mehr et al. 2006). TEBs are dynamic structures with a high rate of proliferation
and differentiation in the body cells. The layer of cells that cover each TEB consists of
transit amplifying cells (TACs). After puberty, this population of cells disappears,
however, mammary gland continues to undergo subtle changes during estrous cycle
(every 3-4 days) and drastic changes during pregnancy, lactation and involution which
94
indicates the existence of adult stem cells in mature ducts. Each mature duct has two
layers of epithelial cells. The epithelial layer that surrounds the central lumen of the duct
is called luminal epithelial cells. These cells differentiate during late pregnancy into milk-
producing cells. The second layer of cells is called myoepithelial cells which contract and
cause release of milk from milk-producing cells.
In recent years, many studies have led to the identification of specific markers of
mammary gland epithelial stem cells. So far, three different mammary epithelial
progenitor cell subpopulations have been characterized in the mammary gland; First,
bipotent stem cells which differentiate to both luminal epithelial cells and myoepithelial
cells; Second, luminal epithelial progenitor cells which differentiate only to luminal
epithelial cells; Third, myoepithelial progenitor cells which differentiate only to
myoepithelial cells (Stingl, Emerman et al. 2005). Different techniques have been carried
out to isolate these candidate cells. In 2006, Stingl et al. successfully sorted by stem cell
population of mammary gland into two distinct subpopulations. First, Mammary
Reforming Units (MRUs) are sorted on the basis of Lin
-
CD24
med
CD49f
hi
expression.
MRUs are able to develop into a mature mammary gland in vivo and in vitro. These cells
are enriched with both bipotent stem cells and myoepithelial progenitor cells. However,
the second population, Mammary Colony Forming Cells (Ma-CFC) only differentiate
into luminal epithelial cells. Ma-CFCs are sorted on the basis of Lin
-
CD24
med
CD49f
med

expression (Stingl, Eirew et al. 2006). Ma-CFC represents luminal epithelial progenitor
cells.
FGFR2b signaling pathway plays a vital role during prenatal and postnatal
development of the mammary gland. FGFR2b is expressed by the mammary epithelium
95
(Dillon, Spencer-Dene et al. 2004) and among its ligands only FGF10 and FGF7 are
expressed by adipocytes surrounding the mammary epithelial tree (Asaki, Konishi et al.
2004). FGFR2b signaling is vital in the maintenance and proliferation of TEBs.
Downregulation of the FGFR2b signaling pathway leads to disappearance of TEBs which
is a reversible phenomenon. This demonstrates that FGFR2b signaling is not important
for the survival of the adult epithelial stem cells in the mammary gland but instead it
determines their fate (Parsa, Ramasamy et al. 2008).
To further study the consequences of loss and gain of function of FGFR2b signaling on
the population of the adult mammary gland progenitor cells we used two different inducible
and reversible mouse models. In one mouse model the expression of a soluble form of
FGFR2b acting as a dominant negative attenuates the endogenous activation of the
FGFR2b signaling pathway. In our study, short time treatment of the DTG females with
doxycycline results in the disappearance of the TEBs. Our Fluorescent Activated Cell
Sorting (FACS) data displays increase in the population of bipotent stem cells associated
with corresponding decrease in the number of luminal progenitor cells. In rtTa;tet(o)Fgf10
mouse model, the gain of function of FGFR2b signaling occurs through overexpression of
Fgf10, the main ligand for FGFR2b. In this model, FACS data demonstrates an increase in
the number of luminal progenitor cells and a decrease in the bipotent stem cell population.
These results indicate that FGFR2b signaling is vital for either the commitment of bipotent
stem cells to the luminal progenitor cell lineage or the self-renewal of the bipotent stem
cells per se.

96

Chapter 1, Part 2 Materials and Methods:
Mice
CMV-Cre mice (Schwenk, Baron et al. 1995) were crossed with rtTa
f/f
; tet(o)sFgfr2b
(Gossen and Bujard 1992; Hokuto, Perl et al. 2003; Belteki, Haigh et al. 2005) and
rtTa
f/f
;tet(o)Fgf10 (Belteki, Haigh et al. 2005); (Clark, Tichelaar et al. 2001) mice. The
rtTa;tet(o)sFgfr2b mice were previously validated (Parsa, Ramasamy et al. 2008). In this
study, rtTa;tet(o)sFgfr2b females are referred to as double transgenic-r2b or DTG-Fgfr2b
and rtTa;tet(o)Fgf10 females are referred to as double transgenic-Fgf10 or DTG-Fgf10.
For histological analysis, the control mammary gland is one of the two inguinal (number
4) mammary glands removed by survival surgery from the DTG mice before doxycycline
treatment. For cell sorting assay controls are single transgenic siblings of the DTG
females.
Immunostaining
MGs from 5 weeks old female mice were fixed in 4% PFA at 4°C overnight and
stored in 70% ethanol. The tissues were embedded in paraffin and 5 µm longitudinal
sections were perfomed. Sections were incubated with Cytokeratin 8/18 (5D3,
Novocastra), Cytokeratin 14 (AF 64, 1:200, Covance), FGFR2 (C-17, 1:50, Santa Cruz,
Inc) and α-smooth muscle actin (1A4, 1:200, Sigma) antibodies overnight at 4ºC. Cy3-
conjugated goat anti-rabbit (Jackson Immunoresearch) and Alexa Fluor 488 goat anti-
97
mouse (Invitrogen) were used as the secondary antibodies. Photomicrographs were taken
using a Leica DM4000B fluorescence microscope with a Leica DFC 290 Camera.

Proliferation Assay
Sections of paraffin embedded mammary gland were stained for PCNA using the
PCNA staining kit (Zymed Labs Inc., San Francisco, CA, 93-1143). The total number of
cells in the epithelium as well as the number of PCNA positive cells in the epithelium
were scored in six photomicrographs (64x magnification) in random portions of a section
of five independent mammary glands in the experimental and control group. A total
number of 1000 cells were counted per sample. The significance of the proliferation in
the mammary glands between control and experimental group was evaluated by one-
tailed paired t-test. P values less than 0.05 were considered statistically significant.
Analysis of cell death:
“In Situ Cell Death Detection, Fluorescein” kit (Roche Applied Science) was used, as
recommended by the manufacturer, to detect the apoptotic cells on 5 mm paraffin section
of 36-hour treated P35 DTG and wild type. The total number of epithelial cells, as well as
the number of apoptotic positive cells in the epithelium, were scored in five
photomicrographs (64x magnification), in TEBs of three independent mammary glands in
the experimental and control group.
Whole Mount Carmin Red Staining
98
The mammary gland epithelium was stained using Carmin Red as previously
described (Faraldo, Deugnier et al. 1998).

Single cell preparation
5 weeks old control (wild type or single transgenic) and DTG females from the same
litter were fed on doxycycline-containing diet for 36 hours. Mammary gland number
three and four were dissected and after removal of the lymph nodes tissues were cut to
smaller pieces and incubated in digestion mix containing 300 unit/ml Collagenase
(C5894-50mg, Sigma-Aldrich, Inc., St. Louis, MO) and 100 unit/ml Hyaluronidase
(H4272-30mg, Sigma-Aldrich, Inc., St. Louis, MO) overnight at 37°C. Red blood cells
were lysed by incubation in Red Blood Cell Lysing Buffer (R7775, Sigma-Aldrich, Inc.,
St. Louis, MO) at 37°C for 10 minutes. To obtain single cell suspension mamospheres
were further digested in 0.25% Trypsin-EDTA solution (T4049, Sigma-Aldrich, Inc., St.
Louis, MO), Dispase II (04942078001, Roche Diagnostics, Indianapolis, IN) and DNase I
(DN25-100mg, Sigma-Aldrich, Inc., St. Louis, MO) and passed through a 40 µm cell
strainer (BD Biosciences, Bedford, MA, 352340).
FACS Assay
Cells were blocked with mouse BD Fc Block (2.4G2, BD Pharmingen) for 10
minutes. Then, they were incubated with primary antibodies - biotinylated CD45 (30-
F11, BD Pharmingen), CD31 (MEC13.3, BD Pharmingen) and TER119 (Erythroid cells,
BD Pharmingen), PE-Cy5-CD49f (BD Pharmingen), PE-CD24 (M1/69, BD Pharmingen)
99
for 15 minutes on ice. After one wash, PE-Texas-Red streptavidin (BD Pharmingen) was
added to the cells and incubated for 10 minutes on ice. The Hematopoietic and
endoplasmic cells (Lin
+
cells) were excluded and Lin
-
cells were sorted based on
expression level of CD49f and CD24 using FACSAria cell sorter (BD Bioscience). FACS
analysis has done using BD FACSDIVA software.
Matrigel Culture
Sorted epithelial cells were placed in equal number in the 100% matrigel (BD
biosciences) in Epicult-B medium as previously described by Stingl et al. (2006). The
serum was removed after 24 hour. The colonies were counted 5 days later.
In Vitro Colony Formation
Wild type Lin
-
CD24
+
CD49f
med
cells were placed in the cell culture plates covered by
the feeder cells (irradiated NIH3T3 cells). Recombinant FGF10 (R&D, 345-FG-025) was
added to experimental wells and 7 days later colonies were counted. Also, the size of the
colonies was compared between treated and non-treated colonies.
100
Chapter 1, Part 2 Results
Attenuation of the FGFR2b signaling pathway reduces the proliferation of the
transient amplifying cells.
We previously reported the loss of terminal end buds (TEBs) after down-regulation of
FGFR2b signaling pathway for 7 days (Parsa, Ramasamy et al. 2008). To better
understand the mechanism of disappearance of the entire structure shorter treatments
were executed. Therefore, DTG-Fgfr2b females were fed on doxycycline-containing diet
for 24 hours, 36 hours and 48 hours at P35. The whole-mount carmin red staining of
these mammary glands (MGs) shows a gradual disappearance of the TEBs within the first
48 hours of treatment. The control mammary gland in this study is taken from the same
mice prior to the initiation of the treatment by survival surgery (Fig. 28-A, A’).
Treatment of DTG-Fgfr2b females for 24 hours does not affect the shape and size of the
TEBs (data not shown). However, the size of TEBs is reduced after 36-hour of treatment
(Fig. 28-B,B’) and TEBs of the DTG-Fgfr2b females disappear completely after 48-hour
treatment (Fig. 28-C,C’). Therefore, disappearance of the TEBs occurs during the first
48 hours of treatment with doxycycline containing food in the DTG-Fgfr2b females.
Disappearance of the TEBs can be the result of the reduction in the cell proliferation
or the increase in the apoptosis. PCNA staining on paraffin-embedded sections of
mammary glands from wild type, 36-hour and 48-hour doxycycline treated females
demonstrates a significant decrease in the number of proliferating cells surrounding the
TEBs after 36 hours of treatment. In the DTG-Fgfr2b MGs before initiation of the
treatment proliferation occurs in the cell bodies of the TEBs and high levels of
101
proliferation are observed in the transit amplifying cells surrounding the TEBs in a very
organized cell layer (Fig. 28-D). After 36-hour of treatment, this organized layer of
transient amplifying cells is no longer detectable in the DTG-Fgfr2b mammary glands
(Fig. 28-E) and longer treatment up to 48 hours leads to the 22 % decrease in the
proliferation of total cells in the TEBs (Fig. 28-F) (p-value≅ 0.022<0.05, SD(non-
treated)≅ 4% , SD(treated)≅ 7%).
Unlike proliferation, apoptosis is not affected by reduction of the FGFR2b signaling
pathway in the TEBs. In non-treated mammary gland few apoptotic cells are observed
(Fig. 28-G). TUNEL staining on the 12 hours and 24 hours treated DTG-Fgfr2b females
display no significant change in the number of TUNEL positive cells in comparison to
the non-treated MG (data not shown). However, 36 hours after initiation of the
doxycycline treatment cell death increases at the level of the cells surrounding the lumen
of the duct (Fig. 28-H). The cell death around the lumen leads to the formation of a
mature lumen, a process that occurs during normal maturation of the lumen.
Nevertheless, no cell death was detected in the cell layer surrounding the remaining
TEBs. Also, 48 hours after initiation of the treatment no significant cell death was
observed in the mammary glands of DTG-Fgfr2b females (Fig.28-I). Therefore, while
there is a significant decrease in the proliferation of transit amplifying cells in the TEBs,
there is not a significant increase in apoptosis.

102

Fig. 28. Attenuation of FGFR2b signaling leads to disappearance of TEBs after 48
hours of doxycycline treatment. Whole-mount carmin red staining of the MG from
DTG-Fgfr2b females before treatment (A, A’). Whole-mount staining of the MG from
doxycycline treated females for 36 (B, B’) and 48 hours (C, C’). Whole mount staining
display disappearance of the bulb-like structure or TEBs at the tip of the mammary tree.
Proliferation assay (PCNA) demonstrates reduction in the proliferation of the body cells
in 48-hour treated TEBs (20X, D-F). Also, layer of highly proliferative cells surrounding
the TEB in the control MG disappeared after 48-hour treatment with doxycycline (inset
in D-F, 40X). Apoptosis assay (TUNEL) does not show any significant rate of apoptosis
in the doxycycline treated TEBs and especially in the layer of transit amplifying cells
(TACs) around the TEBs (20X, G-I). High rate of apoptosis is only detected after 36-
hour treatment in the cells surrounding the central lumen of the duct.
103
Transit amplifying cells in the TEBs express high levels of FGFR2
So far, no specific marker has been identified for transit amplifying cells (TACs)
around the TEBs. TACs share some markers such as Sca 1 or α-Smooth Muscle Actin
(α-SMA) with differentiated myoepithelilal cells or FGFR2 with body cell and luminal
epithelial cells. Surprisingly, staining for FGFR2 demonstrates different localization
between TACs and body cells of the TEB. FGFR2 expression is localized in the nucleus
of the TACs around the TEBs, while in the body cells of the TEBs, FGFR2 is localized to
the plasma membrane of the cells (Fig. 29-A,A’). In addition, lower level of FGFR2 is
found in the committed myoepithelial cells in comparison to the mature luminal epithelial
cells (Fig. 29-D,F). We propose that FGFR2 distribution could facilitate the distinction
between committed luminal epithelial and myoepithelial cells and TACs. On the other
hand, Cytokeratin 8/18 is only expressed in the body cells of the TEBs (Fig. 29-B,B’) and
mature luminal epithelial cells in the mature lumen (Fig. 29-E). Therefore, TACs can be
best recognized as FGFR2
+
CK8/18
-
CK14
-
cells in immunostaining of the TEBs. To
become a mature duct, the nuclear expression of Fgfr2 in transit amplifying cells is lost
and mature myoepithelial cells with low levels of Fgfr2 expression surround the terminal
end of the ducts.

104

Fig. 29. Differential expression of FGFR2 in TEBs and mature ducts.
Immunostaining of the TEBs shows the localization of FGFR2 in the nuclei of the TACs
(A-C, A’-C’) while it is localized in the cytoplasm and on the membrane of the body
cells. To distinguish the body cells and TACs, immunostaining for Cytokeratin 8/18
(CK8/18) was performed (B-B’). The immunostaining for FGFR2 on mature ducts
indicates the higher presence of FGFR2 in the luminal epithelial cells than myoepithelial
cells (F, D). CK8/18 is only detected in the luminal epithelial cells (F, E)

Attenuation of FGFR2b signaling pathway does not prevent differentiation of the
transit amplifying cells.
Since loss of TEBs in the absence of FGFR2b signaling have not been caused by
massive cell death of TACs we suggest that TACs become most likely terminally
differentiated. Therefore, immunostainings for FGFR2, α-SMA, Cytokeratin14 and
105
Cytokeratin 8/18 were performed on sections of MG from different time points after
initiation of doxycyclin treatment. Non-treated TEBs and 24-hour treated TEBs display
nuclear localization of FGFR2b in the cell layer surrounding the TEB (Fig. 30-A,B) in
comparison to the cell body of the TEBs. However, the high levels of FGFR2b are lost
after 36-hour treatment in the TEBs. Cells around the TEBs demonstrate lower levels of
FGFR2 in comparison to the cell body or mature luminal epithelial cells (Fig. 30-C). The
same pattern of expression is also observed after 48-hour treatment (Fig. 30-D). In
addition, while no change was detected in the expression pattern of α-Smooth muscle
actin (α-SMA), after 48-hour treatment, the shape of these cells becomes more elongated.
Furthermore, the expression pattern of CK14 can be used to distinguish between the
basal layer in the mature ducts and TACs in the TEBs. In wild type MG or 24-hour
treated MG, the cell layer surrounding the TEBs is negative for CK14 (Fig. 30-E, F).
However, maturation of the ducts after 36-hour treatment results in the appearance of the
CK14 expression around the terminal ducts (Fig. 30-G). The change in the expression
pattern of CK14 is another evidence for maturation of the duct in the absence of FGFR2b
signaling. Cytokeratin 8/18, on the other hand, is expressed specifically in the body cells
and luminal epithelial cells (Fig. 30-H). Downregulation of FGFR2b signaling does not
affect the expression of the Ck18 (Fig. 30-E-H).

106

Fig. 30. Attenuation of the
FGFR2b signaling leads to the
maturation of the tips of the
mammary tree. Co-
immunostaining for FGFR2 and α-
smooth muscle actin (α-SMA) on
MGs isolated from DTG-Fgfr2b
females. 0 and 24-hour doxycycline
treatment of DTG-Fgfr2b females
display nuclear expression of
FGFR2 in the TACs and
cytoplasmic expression of FGFR2
in the body cells of the TEBs (A,
B). After 36 (C) and 48 (D)-hour
doxycycline treatment, FGFR2 is
reduced in the cell layer
surrounding the tip of the TEBs in
DTG-Fgfr2b females. Co-
immunostaining for cytokeratin 14,
a marker for myoepithelial cells and
cytokeratin 18, a marker for luminal
epithelial cells. Cytokeratin 14 is
absent around the TEBs in 0 and
24-hour doxycycline treated DTG-
Fgfr2b MGs (E, F). After 36 and
48-hour doxycycline treatment, the
ends of ducts are covered by CK14
positive cells (G, H).

Commitment of the bipotent stem cells to the luminal progenitor cells is inhibited by
downregulation of the FGFR2b signaling pathway
Stingl et al. (2006) reported the presence of bipotent stem cells in the Lin
-
CD24
med
CD49f
hi
cell population. We chose 36-hour treated females for our experiment
because at this stage, while TEBs are reduced, they are still visible. First, Lin
+
cells
which are hematopoietic (CD45, TER119) and endothelial (CD31) cells were excluded
107
from the cell pool. Mammary progenitor cells are enriched in both CD29
hi
(β1-integrin)
and CD49f
hi
(α6–integrin) cell population. Here, we used CD49f as our marker of choice.
Cells were sorted based on the intensity of CD49f and CD24 markers on their surface.
Fluorescent Activated Cell Sorting (FACS) analysis demonstrate 50% increase (p value ≅
0.02, SD(treated)≅1.65%, SD(non-treaeted)≅0.087%) in the Lin
-
CD24
med
CD49f
hi
cell
population (MRUs- enriched with bipotent and myoepithelial progenitor cells- Fig.31-F)
and 30% decrease (p velue≅0.003, SD(treated)≅0.28%, SD(non-treated)≅0.32%) in Lin
-
CD24
hi
CD49f
lo
cell population in the 36-hour treated DTG-Fgfr2b mammary gland in
comparison to the non-treated DTG-Fgfr2b females (Ma-CFCs- enriched with luminal
progenitor cells; Fig. 31-E). FACS analysis based on Sca1 marker shows 19% increase in
the treated females vs. non-treated ones (Fig. 31-D). This result indicates the increase in
the stem cell markers in the absence of the FGFR2b signaling pathway.
Our FACS data also suggests that the attenuation of FGFR2b signaling leads to the
lack of commitment for bipotent stem cells to luminal progenitor lineage. To test this
possibility an equal number of sorted cells from control and mutant mammary glands
from each cell population (MRUs and Ma-CFCs) were cultured in matrigel. Serum was
removed 24 hours after. 5 days later, colonies in matrigel were counted in both Lin
-
CD24
med
CD49f
hi
/MRUs and Lin
-
CD24
hi
CD49f
lo
/Ma-CFCs cell populations from treated
and non-treated DTG-Fgfr2b females (data not shown). The results display a decrease in
the number of colonies in Lin
-
CD24
hi
CD49f
lo
/Ma-CFCs cell population and increase in
the number of colonies in Lin
-
CD24
med
CD49f
hi
/MRUs from 36-hour treated DTG-Fgfr2b
females. This data indicates that FGFR2b signaling is crucial for the determination of the
fate of bipotent progenitor cells. Therefore, in the absence of the FGFR2b signaling
108
pathway bipotent progenitor cells will not commit to the luminal epithelial cell lineage
and accumulate in the Lin
-
CD24
med
CD49f
hi
/MRUs population.
To confirm that both of these cell populations express Fgfr2b, Real-time PCR was
performed using the RNA from both treated and non-treated subpopulations. The results
not only show the significant expression of Fgfr2b in both Ma-CFCs and MRUs but also
higher expression in the cells from treated DTG females. This higher expression of
Fgfr2b is another evidence of overexpression of the soluble form of Fgfr2b in the DTG
females after doxycycline treatment (study in progress).
109

Fig. 31. Stem/progenitor cell population of mammary gland changes after
downregulation of Fgfr2b (previous page). Cells obtained from mammary gland of 36-
hour treated and non-treated DTG-Fgfi2b females were sorted using Flourescent Activated
Cell Sorting (FACS) assay based on CD49f and CD24 markers (A-B) and Sca1 marker (D).
Hematopoietic and endoplasmic cells were excluded. Plot illustrates change in the cell
populations of the mammary gland (C). Sca1 is increased in the treated DTG-Fgfr2b
mammary gland vs. control (D). The histogram E illustrates the decrease in the percentage
of Lin
-
CD24
hi
CD49f
lo
/Ma-CFCs while histogram F illustrates the increase in the
percentage Lin
-
CD24
med
CD49f
hi
/MRUs.

110
Overexpression of Fgf10 leads to increase in luminal progenitor population
To study the impacts of overexpression of Fgf10 on adult mammary epithelial cells,
DTG-Fgf10 females were treated with doxycycline for 7 days at P35. The controls for
this experiment were taken from both wild type females after treatment and from the
same DTG-Fgf10 female before treatment. The results demonstrate that while the
mammary tree shape and number TEBs at the of mammary tree has not changed in DTG-
Fgf10, the TEBs are larger in size in comparison to the control TEBs (Fig. 32-
A,A’,B,B’). Also, immunostaining for α-Smooth muscle actin (α-SMA) and FGFR2
demonstrates appearance of multi-layered α-SMA
+
cells at the tip of DTG-Fgf10 TEBs
(Fig. 32-C,C’,D,D’). Therefore, we concluded that overexpression of Fgf10 increases the
number TACs at the tip of the TEBs.
To further investigate the impacts of Fgf10 overexpression on mammary epithelial
progenitor cells, FACS analysis was carried out using CD24 and CD49f markers (Fig.
33-A, B). Our results demonstrate an increase in Lin
-
CD24
hi
CD49f
lo
or luminal
progenitor cells (P6, Fig. 33) and concommitent decrease in Lin
-
CD24
med
CD49f
hi
or
bipotent and myoepithelial progenitor cells in 7-day doxycycline treated DTG-Fgf10
females (P8, Fig. 33). Together, with the results from the DTG-Fgfr2b mice, we conclude
that the FGFR2b signaling patway controls the commitment of the bipotent stem cells in
the luminal epithelial cell lineage.

111

Fig. 32. Overexpression of Fgf10 for 6 days leads to enlargement of the TEBs in
DTG-Fgf10 females. Wholemount carmin red staining of the 6-day treated mammary
gland from DTG-Fgf10 female demonstrates the increase in the size of the TEBs (A-A’,
B-B’). Histological analysis shows the increase in the number of α-SMA positive cells at
the tip of doxycycline treated DTG-Fgf10 TEBs (D-D’) where one layer of α-SMA
positive cells are in the control TEBs (C- C’).

We next asked whether this increase in the cell population in DTG-Fgf10 MG, which
is enriched with luminal progenitor cells, is also the result of proliferation of luminal
progenitor cells. Therefore, we carried out an in vitro colony formation experiment to
observe the direct impact of FGF10 on luminal progenitor cells. Wild type sorted Lin
-
CD24
hi
CD49f
lo
cells were placed in culture dishes covered by feeder cells in the presence
and absence of recombinant FGF10. The results demonstrate increase in size of colonies
in the presence of recombinant FGF10 in comparison of the colonies grown in absence of
recombinant FGF10 (Data not shown). Using both gain and loss of function of FGFR2b
mouse models, we conclude that FGFR2b signaling, in addition to the regulation of
bipotent stem cells commitment to the luminal progenitor cell lineage, plays also an
important role in the proliferation of the luminal progenitor cells.
112

Fig. 33. Overexpression of Fgf10 leads to the increase in the luminal progenitor cells
and decrease in the bipotent stem cells. FACS analysis, using CD49f and CD24 markers
demonstrates the increase of P6/Ma-CFC/Lin
-
CD49f
lo
CD24
hi
population in 6-day
doxycyline treated rtTa;tet(o)Fgf10 MGs. Also, P8/MRU/Lin
-
CD49f
hi
CD24
med
population
is decreased in 6-day doxycycline treated rtTa;tet(o)Fgf10 MGs.

113
Chapter 1, Part 2 Discussion
Breast cancer is usually the result of uncontrolled proliferation of the epithelial cells
in the mammary epithelial tree under the influence of internal or external signals. Fgf10
and Fgfr2b are two of well-known factors identified to be overexpressed in breast cancer
(Theodorou, Boer et al. 2004; Hunter, Kraft et al. 2007). Treatment for breast cancer has
been so far difficult due to the presence of breast cancer stem cells in the tumor. It is
assumed that local endogenous mediators controlling the maintenance of the normal stem
cell niche during development and homeostasis of MG are high jacked by the cancer stem
cells to escape the different treatments currently used against cancer. Since the
identification of adult mammary stem cell (Deome, Faulkin et al. 1959; Medina 1996),
the search to find the exact location of the mammary stem cells has been unsuccessful.
However, it is known that adult mammary stem cells reside in basal layer of mature
ducts. We previously reported the importance of FGFR2b in the maintenance of the
transit amplifying cell population. In addition, our data demonstrated that adult mammary
stem cells survive in the absence of FGFR2b signaling (Parsa, Ramasamy et al. 2008).
The two inducible and reversible mouse models (rtTa;tet(o)sFgfr2b and
rtTa;tet(o)Fgf10) gave us the advantage of temporal control of the attenuation or
overexpression of FGFR2b signaling. Since TEBs are very well formed around P35, we
started treatment at this stage and we observed changes in the TEBs after less than 48
hours of treatment with doxycycline. TEBs consist of transit amplifying cells and their
disappearance is an evidence that downregulation of FGFR2b signaling pathway is
important for the regulation of the pool of transit amplifying cells.
114
A relatively novel assay, Fluorescent Activated Cell Sorting (FACS), based on
specific markers has been instrumental in the identification of many important factors
controlling the maintenance of different progenitor cell pools in mammary gland.
Bipotent stem cells and myoepithelial progenitor cells, which are located in the basal
layer of mature ducts, belong to the same sorted cell population. However, in addition to
myoepithelial progenitor cells, bipotent stem cells can differentiate into luminal
progenitor cells. Our FACS results demonstrate decrease in luminal progenitor cells
which are transit amplifying cells and increase in bipotent stem cells which are stem
cells. The decrease in bipotent stem cells is two times the increase in luminal progenitor
cells which indicates that FGFR2b signaling pathway is instrumental for the commitment
of bipotent stem cells to the luminal progenitor cells.
Furthermore, in our gain of function DTG-Fgf10 model, short term overexpression of
Fgf10 results in increase of number of transit amplifying cells around the TEBs. This
specific change which occurs only in TEBs after a short time of treatment shows that
these cells are the first responders to the increase of FGF10 in MG. In addition, loss of
the majority of bipotent stem cell population is another indicator of the commitment of
these cells to the luminal epithelial lineage. As bipotent stem cells and myoepithelial
progenitor cells belong to the same sorted cell population (Lin
-
CD24
med
CD49f
hi
) we
would expect either no change or an increase in the number of Lin
-
CD24
med
CD49f
hi
cell
population between control and DTG-Fgf10 MGs if FGF10 led to increased commitment
of the bipotent cells to the myoepithelial lineage. As, to the contrary, we observe a
decrease in the Lin
-
CD24
med
CD49f
hi
cell population, we can exclude a role for FGF10 in
the commitment of the bipotent cells to the myoepithelial lineage.
115
During early stages of development, FGF10, produced by the somites, acts upstream
of Wnt signaling to control the formation of the mammary line (Mailleux, Spencer-Dene
et al. 2002; Veltmaat, Van Veelen et al. 2004; Veltmaat, Relaix et al. 2006). Fgfr2b is
expressed by the epithelial cells starting at E11, while Fgf10 is expressed at E11 in
dermamyotome and later around E14.5 in the underlying mesenchyme. After birth, Fgf10
is expressed by adipocytes and preserves their differentiated status while it also controls
the invasion and branching of the epithelial tree into the fat pad (Konishi, Asaki et al.
2006). FGF7, the other ligand for FGFR2b in MG, is also expressed in underlying
mesenchyme from E11.5. However, Fgf10 is expressed 15-fold more than Fgf7 in the
mammary fat pad (Pedchenko and Imagawa 2000). Furthermore, Fgf10 null embryos
phenocopy Fgfr2b null embryos (Sekine, Ohuchi et al. 1999; De Moerlooze, Spencer-
Dene et al. 2000). In our rtTa;tet(o)sFgfr2b mice model, both endogenous FGF7 and
FGF10 in MG are depleted by binding to the soluble form of FGFR2b and results in
reduction of endogenous FGFR2b signaling in the epithelial cells of MG.
In conclusion, we propose a model in which FGFR2b signaling controls the
commitment of bipotent stem cells to the luminal epithelial lineage. Also, FGFR2b
signaling is vital for proliferation of luminal progenitor cells. On the other hand FGFR2b
signaling pathway does not control the commitment of bipotent stem cells to the
myoepithelial cell lineage. This study identified one of the important signaling pathways
in the maintenance and regulation of the stem cell niche which if de-regulated can lead to
cancer.

116
Chapter 2
FGFR2b Signaling controls the Regenerative Capacity of the
Adult Mouse Incisors during Homeostasis
∗

Chapter 2 Abstract
Rodent incisors regenerate throughout the animal lifetime as a result of the
continuous deposition of enamel, the hardest component of the tooth, by ameloblasts.
Putative ameloblast stem cells (ASC) reside at the base of the incisors in a structure
called the cervical loop (CL) and have the capacity for self-renewal. Previous studies on
incisor development in Fibroblast growth factor 10 (Fgf10) null embryos and inhibition
of FGF10 signaling on developing incisors in in vitro organ cultures suggested that
FGF10, acting mainly via Fibroblast Growth Factor Receptor 2b (FGFR2b), is critical for
the maintenance of the ASC population in developing mouse incisors. To further explore

∗
Manuscript is under revision. Parsa, S.
#,1,2
, Kuremoto, K.
#,3
, Tabatabai, R.
1
, MacKenzie,
B.
4
, Yamaza, T.
5
, Akiyama, K.
6
, Koh, C. J.
1
, Al Alam, D.
1
, Bellusci, S
1,2,4
.
(
1
Developmental Biology and Regenerative Medicine Program, Saban Research
Institute of Childrens Hospital Los Angeles, Los Angeles, CA 90027, USA,
2
Department of Pathology, Keck School of Medicine, University of Southern
California, Los Angeles, CA 90033, USA,
3
Department of Removable Prothodontics
and Occlusion, Osaka Dental University, Osaka 540‐0008, Japan,
4
Excellence Cluster
in Cardio-Pulmonary Systems, University of Giessen Lung Center, Department of
Internal Medicine II, Klinikstrasse 36, 35392 Giessen Germany,
5
Department of Oral
Anatomy and Cell Biology, Graduate School of Dental Science, Kyushu University,
Fukuoka 812-8582, Japan,
6
Center for Craniofacial Molecular Biology, School of
Dentistry, University of Southern California, Los Angeles, CA 90033, USA,
#
Both
authors contributed equally to this work. Correspondence: sbellusci@chla.usc.edu

117
the role of FGFR2b signaling at different stages during development and homeostasis, we
used an rtTA transactivator/tetracycline promoter approach allowing inducible and
reversible attenuation of FGFR2b signaling. Down regulation of FGR2b signaling during
the embryonic stages leads to the formation of a rudimentary labial CL and loss of inner
enamel epithelial (IEE) layer. In addition, post-natal attenuation of this signaling pathway
results in abnormal incisor regeneration characterized by improper enamel formation
leading to the precocious degradation of the incisors. Histological analyses of adult
incisors indicate absence of the IEE layer in the region directly adjacent to the CL
associated with decreased proliferation of the ameloblast progenitor cells (APCs) in the
CL. Upon removal of doxycycline after 4 weeks of treatment (from P14 through P42),
the incisors resume growth and reform an enamel layer demonstrating that the integrity of
the ASC niche is not compromised by transient postnatal attenuation of FGFR2b
signaling. Taken together, our results demonstrate that FGFR2b signaling plays a critical
role in the regenerative capacity of adult incisors by controlling the proliferation of
APCs, but not by regulating the survival of ASCs.
118
Chapter 2 Introduction
The initial growth and continuous regeneration of rodent incisors during the animal
lifetime occurs as a result of the activation of stem cells located at the posterior end of the
incisors in a distinct anatomical structure called the cervical loop (CL). The CL is a bud-
like epithelial structure comprised of a central core of stellate reticulum cells surrounded
by basal epithelial cells. It is proposed that epithelial stem cells are located at the border
between basal epithelium and stellate reticulum at the apex of the CL (Harada, Kettunen
et al. 1999). Each incisor exhibits two CLs; one located on the labial aspect and the other
on the lingual aspect. Ameloblast stem cells (ASCs) are epithelial cells that reside
exclusively in the labial CL. These cells have the capacity for self renewal and give rise
to ameloblast progenitor cells (APCs), which proliferate and differentiate into
ameloblasts as they migrate distally. Ameloblasts produce enamel, a protective coat of
calcium phosphate, covering only the labial surface of the incisors. The lingual CL is
smaller and has the ability to differentiate into root-like structures that strengthen the
connection of the incisors to the jaws.
Incisor development is one of the most commonly studied models to determine the
role of mesenchymal-epithelial interaction during organogenesis. Incisor development
during embryogenesis starts with the formation of an incisor placode from the oral
ectoderm at embryonic day 12 (E12). During the next 2 days, the placode expands to
form a bud invading the underlying mesenchyme. By E15, the developing incisor shows
the presence of a CL in the labial and lingual side as well as the inner and outer enamel
epithelium. During the next three days, the incisor will continue to expand and will form
119
on the labial side, mature ameloblasts, enamel, dentin and pre-dentin (Kerley 1975).
Factors such as Notch, Fibroblast Growth Factors (FGFs) and Bone Morphogenetic
Proteins (BMPs) play important roles during tooth development (Wang, Suomalainen et
al. 2007). In particular, FGFs are critical growth factors for incisor and molar
development. In molars, mesenchymal FGF3 and FGF10 signal through epithelial
Fibroblast Growth Factor Receptor 2-b (FGFR2b). FGF10 also maintains Sonic
Hedgehog (Shh) expression, which marks cells that are differentiating along the
ameloblast lineage (Bitgood and McMahon 1995; Klein, Lyons et al. 2008).
In the incisors, Fgf10 is expressed specifically in the mesenchymal cells located
around the CL and under the inner enamel epithelium. Fgf3 also shares the later location.
FGF10 binds to (FGFR2b), which is expressed in the basal epithelium and the stratum
intermediate of the CL (Harada, Kettunen et al. 1999). Due to perinatal lethality of
Fgfr2b null embryos, it has been impossible to study the role of FGFR2b ablation in
incisors post-natally (De Moerlooze, Spencer-Dene et al. 2000). Analysis of the incisor
phenotype in Fgf10 null embryos and in vitro culture of developing incisors in the
presence of FGF10 blocking antibodies have suggested that FGF10 acts as a survival
factor for ASCs in the CL (Harada, Toyono et al. 2002). It has been reported that in the
incisors of Fgf10
-/-
mice, a root analog forms in the labial side as a result of cessation of
the proliferation in the inner enamel epithelial (IEE) and higher proliferation rate in the
outer enamel epithelial (OEE) (Yokohama-Tamaki, Ohshima et al. 2006). This transition
from crown to root is therefore triggered by cessation of Fgf10 signaling.

120
More recently, epithelial specific deletion of Fgfr2 in the CL using the Nkx3.1
Cre

driver line suggested that FGFR2 signaling is required for the development and
maintenance of the maxillary CL (Lin, Cheng et al. 2009). However, this experiment is
not informative about the role of FGFR2b during later stages of embryonic development
and adult homeostasis of the incisor as Nkx3.1
Cre
is active already at embryonic day 11.5
(E11.5) in the epithelium of the developing incisor. To circumvent the perinatal lethality
of Fgfr2b mutants, we developed a mouse model allowing inducible and reversible
attenuation of FGFR2b signaling (Parsa, Ramasamy et al. 2008). Using this model, we
analyzed the specific role of FGFR2b signaling during different stages of embryonic
development and homeostasis of the incisors. Attenuation of FGFR2b signaling during
embryonic incisor development leads to the formation of a rudimentary CL with a
reduced or absent pool of APCs. Blockade of FGFR2b signaling from postnatal day 14
onwards leads to an almost complete loss of maxillary incisor and deficient enamel
deposition in the mandibular incisor. However, release of inhibition of FGFR2b signaling
allows incisor growth to resume normally, suggesting that in our experimental conditions,
FGFR2b signaling is not critical for the maintenance of the ASCs in the adult mice. Our
model allows us to determine the cellular mechanisms controlled by FGFR2b signaling
during different stages of development and homeostasis of the incisors in adult mice.

121
Chapter 2 Materials and Methods
Generation of rtTA; tet(O)sFgfr2b animals
CMV-Cre mice (Schwenk, Baron et al. 1995) were crossed with rtTA
flox
mice
(Belteki, Haigh et al. 2005) to generate rtTA mice expressing rtTA under the ubiquitous
Rosa26 promoter in all somatic and germ cells. This constitutive rtTA mouse line was
then crossed with tet(O)sFgfr2b responder line (kind gift of Dr. Jeffrey Whitsett-
(Hokuto, Perl et al. 2003) to generate double transgenic (DTG) heterozygous animals
allowing ubiquitous expression of soluble FGFR2b, acting as a dominant negative in
every cell using the rtTA transactivator/tetracycline promoter system (Gossen and Bujard
1992). All mice were generated on a CD1 mixed background and allowed inducible and
reversible attenuation of the FGFR2b pathway in the embryo or postnatal mouse by
doxycycline containing food ingestion (Rodent diet with 0.0625 % doxycycline, Harlan
Teklad TD01306). Mice were genotyped as previously described (Schwenk, Baron et al.
1995; Hokuto, Perl et al. 2003; Belteki, Haigh et al. 2005). 5 animals were used for each
control and experimental group. Adult DTG animals not exposed to doxycycline were
phenotypically undistinguishable from control mice indicating that in our double
transgenic system, soluble FGFR2b expression is strictly doxycycline dependent (Parsa,
Ramasamy et al. 2008). Due to the low number of DTGs generated for this study, all
DTG mice were used in the experimental group. Control group was wild type or single
transgenic animals from the same litter. Animal experiments were performed under the
research protocol approved by the Animal Research Committee at Childrens Hospital Los
Angeles and at Osaka Dental University (protocol No. 08-04001 and 09-02034). All
122
animals were maintained in a temperature-controlled room with a 12-hour alternating
light-dark cycle and fed sufficient diet and water ad libitum throughout the experimental
period.
Tissue Preparation and Histology
Embryonic heads were fixed in 4% paraformaldehyde at 4°C overnight. Postnatal
wild type and double transgenic jaws were collected and fixed in 4% paraformaldehyde
and 0.2% picric acid in PBS overnight at 4°C and then decalcified in 5 %
ethylenediamine tetra acetic acid containing 4% sucrose in PBS for 2 weeks at 4°C.
Samples were dehydrated in a graded ethanol series, paraffin embedded, and sectioned to
5µm. Sections were deparaffinized and stained with hematoxylin and eosin (H&E) or
used for immunohistochemistry.
Immunohistochemistry
Deparaffinized sections of both prenatal and postnatal samples were washed in 3%
H
2
O
2
in methanol for 10 minutes at room temperature. Antigen retrieval was performed
in citrate buffer (pH 6) (Invitrogen, Carlsbad, CA) at 95
o
C for 15 minutes. Sections were
incubated with primary antibodies at 4
o
C, overnight. The primary antibodies used in this
study were against Amelogenin (Santa Cruz Inc., Santa Cruz, CA), proliferating nuclear
antigen (PCNA, Santa Cruz Inc., Santa Cruz, CA), Vimentin (Santa Cruz Inc., Santa
Cruz, CA), FGFR1 (Flg, Santa Cruz Inc., Santa Cruz, CA,), FGFR2 (Bek, Santa Cruz
Inc., Santa Cruz, CA,), Keratin-14 (Thermo Fisher Scientific, Waltham, MA,) and E-
cadherin (BD Bioscience, San Jose, CA). Immunohistochemistry was performed with
SuperPicture (Invitrogen, Carlsbad, CA) or Dako EnVision kit (Dako North America,
123
Carpinteria, CA) followed by counterstaining with hematoxylin. Photomicrographs were
taken using a Leica DM4000B microscope with a Leica color Camera.
Cell Proliferation Assays
5-bromo-2'-deoxyuridine (BrdU, Invitrogen, Carlsbad, CA) was injected
intraperitoneally (1ml /100g of body weight) 2 hours before sacrifice. Incorporated BrdU
was detected by ZYMED BrdU Staining Kit (Invitrogen, Carlsbad, CA). Section from
maxillary and mandibullary incisors from 3 independent control and DTG animals were
analyzed. The area of the CL was first quantified for each section. The number of
proliferating cells within the CL area was quantified. For each section, the results were
presented as ratios between the number of proliferating cells versus the area of the CL.
Significance of the difference between control and DTG was evaluated by a one-tailed
paired t-test. P values less than 0.05 were considered to be statistically significant.
In Situ Hybridization
In situ hybridization for Shh was carried out as previously described (Bellusci,
Henderson et al. 1996).

124
Chapter 2 Results
Ubiquitous inducible expression of soluble FGFR2b phenocopies the inactivation of
Fgfr2b expression
We validated our double transgenic system by inducing the expression of soluble
FGFR2b from embryonic day 9.5 (E9.5) through E18.5 (Fig. 42). Our results indicate that
DTG embryos exposed to doxycycline phenocopy Fgfr2b
-/-
embryos (De Moerlooze,
Spencer-Dene et al. 2000). DTG embryos not exposed to doxycycline, as well as single
transgenic embryos exposed to doxycycline, were identical to wild type embryos (data
not shown). This demonstrates that there is no leakiness in the expression of soluble
FGFR2b in our double transgenic embryos in the absence of doxyclycline and that there
is no toxic effect of doxycycline on embryonic development. A Comparison of wild type
and DTG embryos exposed to doxycycline from E9.5 to E18.5 indicates that mutant
embryos exhibit limb agenesis, absence of eyelid closure (Fig. 42-A,B) and cleft palate
(Fig. 42-B,C and E,F). In addition, we observed a curly tail (arrow in Fig. 42-D), which is
a phenotype observed in Fgfr2b
-/-
embryos (De Moerlooze, Spencer-Dene et al. 2000) but
not in Fgf10
-/-
embryos (Sekine, Ohuchi et al. 1999). These results support our conclusion
that our mouse model allows specific and robust inhibition of FGFR2b signaling in an
inducible and non-leaky fashion.

125

Fig. 34. CL is still present in embryonic incisors in spite of long-term FGFR2b
signaling attenuation. Longitudinal section of maxillary incisors of (A-C) Control and
(D-F) DTG embryos exposed to doxycycline from E12.5 to E18.5. Note the presence of a
rudimentary CL in (E) DTG incisors. Note the completely abnormal ameloblast layer in
DTG incisors (F). CL: Cervical loop, OD: odotonblast; IEE: inner epithelium enamel;
OEE: outer epithelium enamel.

126
Attenuation of FGFR2b signaling during development leads to the formation of a
rudimentary cervical loop.
In order to gain insight on the role of FGFR2b signaling in mandibular and maxillary
incisors during early embryonic development, pregnant females carrying DTG and
control embryos were exposed to doxycycline from E12.5 to E18.5 and the incisors were
analyzed. The control embryos exhibited well-formed mandibular and maxillary incisors
with clearly distinguishable cervical loop (CL) in the labial and lingual aspects (Fig. 34-
A,B- please note that Fig. 34 focuses on the mandibular incisor). Close-up examination
shows presence of well-organized inner (IEE) and outer (OEE) enamel epithelial cells
(Fig. 34-C). By contrast, in the DTG embryos, both maxillary and mandibular incisors
are affected. Interestingly, the mandibular incisors were less affected than the maxillary
(data not shown). DTG mandibular incisors exhibit a rudimentary labial cervical loop
(Fig. 34-E) similar in shape to the one observed in control lingual cervical loop (data not
shown). We also noticed defective ameloblast layer in DTG incisors (Fig. 34-F). It
appears that the IEE is more disrupted than the OEE. Indeed, our results indicate high
rate of proliferation in the OEE and loss of proliferation in the IEE (data not shown). This
phenotype is very similar to the one previously reported by Yokohama-Tamaki et al.
(2006).

127
Fig. 35. Defective maintenance of epithelial cells in DTG embryonic mandibular
incisors exposed to doxycycline from E16.5 to E18. H&E staining (A-D) and analysis
of BrdU incorporation (E-H) of control (A,B,E,F) and mutant (C,D,G,H) incisors of
embryos exposed to doxycycline from E16.5 to 18. (A,B) control incisor showing the
well formed CL and ameloblast layer. (C,D) DTG incisor displaying absence of well-
identified epithelial cells in the CL and defective enamel epithelial layer. (E,F) High rate
of proliferation in both the CL and enamel epithelial layer of control incisors. (G,H) Lack
of proliferation in the CL and enamel epithelial layer of the DTG incisors. CL: Cervical
loop, OD: odotonblast; IEE: inner epithelium enamel; OEE: outer epithelium enamel.

128
(Fig. 35 continued)

129
FGFR2b Signaling plays a critical role in the maintenance of epithelial cells in the
CL and IEE of the mandibular incisors
In order to investigate the importance of FGFR2b signaling in the maintenance of
epithelial cells during early incisor development, pregnant females with DTG and control
embryos were treated post-incisor formation, from E16.5 to E18. Histological analysis of
longitudinal sections of the mandibular incisors demonstrate the total loss of epithelial
cells in the CL of the DTG embryos (Fig. 35-C) in comparison to control (Fig. 35-A)
littermates. In more mature regions, IEE is still visible in DTG incisors but it is very
disorganized compared to the control counterpart (Fig. 35-D vs. B). In addition, cell
proliferation is completely lost in both mandibular (Fig. 35-G,H) and maxillary (data not
shown) incisors of the DTG embryos, while active proliferation continues in control
incisors (Fig. 35-E,F). Interestingly, proliferation is still detected in the mesenchymal
cells of the DTG incisors indicating the specific role of FGFR2b signaling in the
regulation of the cell proliferation in the epithelial cells (read arrows in Fig. 35-G).
Sonic hedgehog (Shh) expression has been shown to be a marker of transit amplifying
cells/Ameloblast Progenitor cells (APCs) in the CL (Bitgood and McMahon, 1995; Klein
et al., 2008). Our results demonstrate significant decrease in the expression level of Shh
in DTG incisors (Fig. 36-B) in comparison to control incisors (Fig. 36-A). This result
suggests loss of transit amplifying cells as a result of FGFR2b signaling downregulation.
Also to verify the status of mature ameloblasts in the IEE of the DTG incisors,
immunohistochemistry for amelogenin protein was performed. Our results indicate a
significant decrease in amelogenin expression in DTG vs. control incisors (Fig. 36-C,D),
130
demonstrating that transient attenuation of FGFR2b signaling leads to abnormal
amelogenesis.

Fig. 36. Reduction in the expression of Shh and amelogenin in DTG mandibular
incisors. Control and DTG incisors of embryos exposed to doxycycline from E16.5 to 18.
(A) Shh expression by in situ hybridization showing the presence of Ameloblast
Progenitor Cells (APCs) in control incisors. (B) Reduction in the intensity of Shh
expression in DTG incisors indicating defective amelogenesis. (C) Amelogenin
expression by IHC demonstrating the presence of mature ameloblasts (Amelob) in the
control incisors. (D) Decrease in Amelogenin expression confirming defective
amelogenesis in DTG incisors. CL: Cervical loop.

131
Post-natal attenuation of FGFR2b signaling leads to the loss of amelogenesis in
maxillary incisors.
To study the impact of FGFR2b down-regulation on incisor homeostasis, postnatal
DTG and control mice (n=3 each) were fed doxycycline-containing food for different
lengths of time starting at P14 (Fig. 37). Hereafter control mice are referred to either as
wild type or single transgenic animals from the same litter. At P14, mandibular and
maxillary incisors in the control and DTG animals were normally developed (data not
shown). After 14 or 28 days of treatment with doxycycline (at P28 and P42,
respectively), both maxillary and mandibular incisors of DTG mice (Fig. 37-F,G) were
undistinguishable from the corresponding control incisors (Fig. 37-A,B). However, at
P70 (56 days on doxycycline), the maxillary incisors had almost disappeared and
mandibular incisors had grown excessively (Fig. 37-H) compared to wild type mice (Fig.
37-C). The increased length of mandibular incisors is most likely due to the absence of
abrasion between the upper and lower incisors. At P90, the mandibular incisors in the
mutants were also degraded compared to the controls (Fig. 37-I,J vs. D,E, respectively)
indicating severe enamel deposition defects.
To understand the early impact of the downregulation of FGFR2b on maxillary
incisors homeostasis, histological analyses were performed on incisors harvested and
sectioned from mice treated for 2 weeks (P28) and 4 weeks (P42) with doxycycline
(supplementary Fig. 43). At P28, while no visible abnormalities could be detected in the
external appearance of the maxillary incisors, serial sagittal sections indicated a failure to
develop new ameloblasts in posterior regions close to the CL (cartoon in Fig. 43 indicates
132
position of sections shown in A-C. A is anterior and C is posterior). This observation
suggests that FGFR2b signaling is not critical for already formed ameloblasts, such as
those found in more distal regions (Fig. 43-A’). However, FGFR2b signaling appears to
be critical for the formation of new ameloblasts indicated by defective stem cell
homeostasis in the labial CL.

Fig. 37. Maxillary incisors disappear in the DTG as a result of long-term attenuation
of FGFR2b signaling. (A-E) control incisors at different post-natal time points. (F-G)
DTG Incisors from mice continuously exposed to doxycycline from P14. No obvious
difference between control and experimental groups is observed until P42. (H) At P70,
the maxillary incisors are no longer visible in DTG mice. (I and J) At P90, the
mandibular incisors are also damaged in DTG mice. (J) Dissected incisors displaying
defective enamel deposition.

133
Longer exposure to doxycycline treatment (28 days, P42) leads to more extensive and
severe defects in the maxillary incisors of DTG mice. At P42, in the posterior region
adjacent to the CL, the enamel space of DTG mice was filled by connective tissue and
blood vessels (Fig. 38-B). Also, the dentin tissue was abnormally formed and attached to
defective enamel (Fig. 38-B). Corresponding maxillary incisor sections of doxycycline
fed control mice demonstrated intact structures related to amelogenesis (Fig. 38-A).
Longitudinal sections indicated the presence of a rudimentary CL in DTG compared to
control (Fig. 38-A’,B’). Cell proliferation analysis in the CL of 4 weeks treated DTG
(n=3) and control (n=3) incisors demonstrated significant decrease in cell proliferation in
the mutant CL. (5.7 ± 2.0 vs. 22.6 ± 6.9, number of BrdU positive cells/arbitrary unit of
CL area, P=0.028, in DTG vs. control, respectively).
To further study the impact of FGFR2b downregulation on the adult maxillary
incisor, immunohistochemistry for different marker of the epithelial cells in the incisors
was carried out. Immunohistological staining in control mice demonstrated the specific
expression of FGFR1 (Fig, 38-C), FGFR2 (Fig. 38-G), and Amelogenin (Fig. 38-D) in
the inner enamel epithelium (IEE) and expression of cytokeratin-14 in both the IEE and
outer enamel epithelium (OEE) (Fig. 38-H). None of these markers were expressed in the
DTG mice treated for 28 days (Fig. 38-E,I,F,J respectively). This result shows that
although down regulation of FGFR2b signaling does not lead to a macroscopic phenotype
at early stages of treatment, drastic changes are occurring at the cellular level and are
observed in the region of the maxillary incisor close to the CL.

134

Fig. 38. Long-term postnatal attenuation of FGFR2b signaling leads to amelogenesis
defects in maxillary incisors. H&E staining of (A) control and (B) DTG incisors after
28 days of doxycycline treatment. Please note the absence of inner and outer enamel
epithelium. (A’B’) Longitudinal section through the CL. Please note the hypoplastic
labial CL in the DTG (B’) compared to control (A’) incisors. Expression of FGFR1
(C,E), FGFR2 (G,I), Amelogenin (D,F) and cytokeratin 14 (H,J) were lost in DTG
incisors after 4 weeks of doxycycline treatment. IEE: inner enamel epithelium.

135
Post-natal downregulation of FGFR2b signaling leads to loss of amelogenesis and
decrease in proliferation in the mandibular incisors
Next, our analysis focused on the role of FGFR2b in mandibular incisor homeostasis.
Nkx3.1
Cre
mediated inactivation of Fgfr2b resulted in defective maxillary incisors and
apparently normal mandibular incisors (Lin, Cheng et al. 2009). In agreement with these
results, we also observed defective maxillary incisor. However, our data also show that
enamel deposition in mandibular incisors is impaired after long-term treatment with
doxycycline, albeit to a lesser degree compared to the phenotype displayed by the
maxillary incisors (Fig. 37-I,J). Histological analyses of control and mutant incisors were
carried out (Fig. 39-A-F) at P42, after 28 days of doxycycline treatment. Control
mandibular incisors display amelogenic structures, such as the enamel space covered by
inner and outer enamel epithelia, CL, as well as odontogenic structures such as the
odontoblast layer lining dentin and dental papilla (Fig. 39-A-C). In contrast, the only
recognizable amelogenic structure in the corresponding DTG mandibular incisors was a
rudimentory CL, while other odontogenic structures persisted (Fig. 39-D-F).
To confirm the histological phenotype, immunostaining for amelogenic markers were
carried out. Amelogenin, which is localized in the IEE, especially in fully developed
ameloblasts, is absent from the CL in mutant mice (data not shown). Similarly, keratin-14
and E-cadherin, normally expressed in the IEE, OEE, and CL of wild type incisors was
absent from the IEE and OEE of DTG incisors (data not shown). These results confirm
that ameloblast function is impaired in the mandibular incisors as a result of the
downregulation of the FGFR2b signaling pathway.
136

Fig. 39. Ameloblast formation and ameloblast progenitor cell proliferation in
cervical loop of mandibular incisors decreases after 4 weeks of treatment in DTG
mice. H&E staining of longitudinal sections of control (A-C) and DTG (D-F) incisors.
Note the defective inner and outer enamel epithelium in DTG incisors. In addition, the
CL is still present in DTG incisors (E) in spite of significant enamel defects (F). (G,H)
BrdU staining for control (G) and DTG (H) incisors showing decreased proliferation in
the cervical loop of adult DTG incisors.

137
To study the impact of FGFR2b signaling on cell proliferation in the maxillary
incisors, proliferation assays using BrdU label incorporation were carried out in vivo and
longitudinal sections of 4-week treated DTG and control mandibular incisors (n=3 for
each) were examined. Analysis of BrdU incorporation indicated a drastic reduction in the
epithelial proliferation in the CL of DTG incisors (Fig. 39-H) compared to controls (Fig.
6-G). Quantification of the BrdU signal confirmed the reduction in proliferation (22.5 ±
5.0 vs. 58.2 ± 10.7, P=0.012, number of BrdU positive cells/arbitrary unit of CL area,
DTG vs. control, respectively). Our results demonstrate that FGFR2b is critical for the
formation of the IEE and OEE layer in mandibular incisors. Attenuation of FGFR2b
signaling results in a significant reduction of the proliferation rate of transit amplifiying
ameloblast progenitor cells in the CL. Reduced proliferation of APCs is likely the
underlying cause of enamel defects observed in the mandibular incisors. However, the
question remains whether or not ASCs are also affected in mandibular and maxillary
incisors.
Defective maxillary and mandibular incisors resume normal growth upon cessation
of doxycycline treatment
To elucidate the role of FGFR2b signaling in survival and/or proliferation of ASCs in
the incisors, the reversibility of the incisor phenotype after long-term treatment of the
DTG mice with doxycycline was tested. Beginning at P14, DTG and control mice
(n=4/group) were fed doxycycline containing food for 4 weeks. At P42, 28 days after
doxycycline treatment, no difference in gross incisor phenotype was observed (Fig. 40-
A,H). At P70, 28 days post treatment, the tips of the maxillary incisors of DTG mice
138
were broken (Fig. 40-I). Newly formed maxillary incisors began to grow out of the
broken site by P99 (57 days post treatment) (Fig. 40-K). At this stage, the mandibular
incisors lost their transparency. Between P99 and P109, the mandibular incisors broke as
a result of contact with the newly formed maxillary incisors. By P120, the maxillary and
mandibular incisors of DTG mice developed to a length that allowed for a perfect contact
between maxillary and mandibular incisors (Fig. 40-N). This experiment demonstrates
the reversible incisor phenotype after reactivation of FGFR2b signaling.
To show the de novo formation of the enamel epithelial layer after reversal of the
phenotype macroscopically, histological sections were prepared from both control and
DTG mandibular incisors at P120, 78 days post doxycycline treatment. Our results show
the de novo formation of a defined enamel epithelial layer in longitudinal sections of the
mandibular incisors as well as a clearly visible CL (Fig. 40-R-T) similar to the control
(Fig. 40-O-Q). Therefore, normalization of FGFR2b signaling in the DTG animal allows
for proper formation of ameloblasts to resume. This is associated with the corresponding
development of normal enamel epithelial layer.
In conclusion, adult DTG mice treated for four weeks (P14 through P42) lost the
visible part of the maxillary incisors at P70, indicative of progressive ameloblast defects
occurring at the level of the labial CL upon attenuation of FGFR2b signaling. However,
re-growth of the maxillary incisors of DTG mice after the animals resumed a normal diet,
demonstrates that attenuation of FGFR2b signaling does not compromise the survival of
the adult ASCs as they still retain the ability to give rise to transit amplifying APCs
which can properly differentiate to form enamel producing ameloblasts.
139

Fig. 40. Re-growth of defective incisors upon removal of doxycycline. After 4 weeks
of treatment with doxycycline (P42), control (A-G) and DTG (H-N) mice were fed on
normal diet and changes in the appearance of maxillary incisors were monitored at
different time points from P42 to P120. At P70, defective upper and lower incisors are
observed in DTG. From P91 through P120, both pairs of incisors are re-growing. H&E
staining of control (O-Q) and DTG (R-T) maxillary incisors at P120 demonstrating
normal CL and inner and outer enamel epithelium in DTG compared to control animals.

140
Chapter 2 Discussion
The ultimate goal in modern regenerative medicine is to harness the power of
endogenous adult stem cells to allow for the recovery of affected organs after injury.
Adult stem cells replace damaged cells by asymmetrical division, contributing to the
proliferation of transit amplifying cells, and differentiation into functional cells. Mouse
incisors constantly regenerate during adult life and are therefore an ideal model to use for
determining the signaling pathways that are essential to tooth regeneration. Previously, it
was reported that FGFR2b signaling is important for the maintenance and survival of the
ASCs and APCs in the CL during development (Harada, Toyono et al. 2002). More
recently, it has been reported using the Nkx3.1
Cre
driver line that FGFR2b signaling
controls the formation of the CL in the maxillary but not in the mandibular incisors
during embryonic development (Lin, Cheng et al. 2009). These results are different from
our own which indicate that indeed both the mandibular and maxillary incisors are
affected during development, albeit with different severity. A possible explanation for
this discrepancy is that the driver line used in the previous study did not allow effective
Cre expression in the developing mandibular incisors.
Contrary to the studies mentioned above on the role of FGFR2b during incisor
development, our results during homeostasis using a reversible and inducible system
demonstrate that attenuation of FGFR2b signaling does not compromise the survival of
the ASCs located in the CL but instead leads to decreased proliferation of transit
amplifying APCs (Fig. 41). This, in turn, decreases de novo ameloblast formation, a
defect visualized over time by insufficient enamel deposition in both the maxillary and
141
mandibular incisors, with the upper incisors more severely affected. Interestingly, we also
found that inactivation of FGFR2b signaling during embryonic development (from E12.5
to E18.5) leads to the formation of a distinct rudimentary CL in both the maxillary and
mandibular incisors suggesting that embryonic ASCs, like the post-natal counterpart, are
also not affected by FGFR2b attenuation during embryonic development. The molecular
and cellular bases for the differences observed between maxillary and mandibular
incisors during development and homeostasis are still elusive but our results suggest that
attenuation of FGFR2b signaling is partially compensated in mandibular CL by an
alternative signaling pathway that remains to be identified.

142

Fig. 41. Role of FGFR2b in incisor homeostasis. Schematic of the regenerative incisor.
(A) General organization of the incisor. Growth occurs along the proximal-apical axis.
The Ameloblast Stem Cells (ASC) are contained in the cervical loop (CL) located on the
lingual and labial aspects. (B) Close up of the CL indicating the putative position of the
ASCs as well as the Ameloblast Progenitor Cells (APCs). (C) FGFR2b signaling controls
the maintenance of the APCs but not the survival of the ASCs. Model adapted from
Wang et al., 2007.

In the CL of 2 day-old mice, the main ligands for FGFR2b are FGF3 and FGF10.
While Fgf10 is strongly expressed by the dental mesenchyme adjacent to the CL and IEE,
both structures express FGF receptors Fgfr1b and Fgfr2b. Fgf3 is mostly expressed in the
mesenchyme adjacent to the IEE (Harada, Kettunen et al. 1999). Such difference
143
between Fgf3 and Fgf10 expression domains is also observed during embryonic
development at E16, when the CL is forming (Harada, Toyono et al. 2002). The more
restricted pattern of Fgf3 expression suggests that Fgf3 and Fgf10 may play redundant
and unique functions during incisor development and homeostasis. These two ligands
establish a dynamic interaction with critical signaling pathways controlling the fate of
APCs during incisor development and homeostasis. Using CLs cultured in vitro harvested
from P2 incisors, it has been shown that FGF10 regulates the Notch signaling pathway in
epithelial progenitor cells via the stimulation of lunatic fringe expression (Harada,
Kettunen et al. 1999). The border where lunatic fringe (expressed in basal epithelial cells)
and Notch 2 (expressed in stellate reticulum) meet is considered to be the niche for
incisor epithelial stem cells in the CL. During development, the induction of lunatic
fringe by FGF10 in the basal epithelium of the CL is therefore thought to be essential for
the initial formation of the early pool of ASCs. In support of this finding, Fgf10 null
embryo incisors display a rudimentary CL (Harada, Toyono et al. 2002). Fgf3

null mice
are viable and demonstrate only pigmentation defects in their incisors. However,
additional inactivation of one Fgf10 allele in Fgf3 null mice (Fgf3
-/-
; Fgf10
+/-
) leads to
smaller incisors with hypoplastic labial CLs and an absent enamel layer suggesting that
Fgf10 compensates for the loss of Fgf3 during incisor development (Wang, Suomalainen
et al. 2007). Our results with DTG incisors treated from E12.5 to E18 indicate a
rudimentary CL similar to the one observed on the lingual aspect. This CL phenotype is
more severe than the one previously published for Fgf3
-/-
; Fgf10
+/-
P1 incisors (Wang,
Suomalainen et al. 2007).

144
Interestingly, none of the previous studies addressed the simultaneous role of both
FGFR2b ligands during incisor homeostasis. Such studies could not be conducted as
Fgf10 null animals die at birth from many defects including lung agenesis (Sekine,
Ohuchi et al. 1999). Using our in vivo model of inducible expression of the dominant
negative FGFR2b, we achieved attenuation of both FGF3 and FGF10 signaling. Our data
from postnatal treated incisors show a more drastic phenotype than previously published
studies with Fgf3
-/-
; Fgf10
+/-
adult incisors (Wang, Suomalainen et al. 2007).
Recently, FGF3 and 10 in the mesenchyme have been shown to regulate Fgf9
expression in the epithelium, which in turn regulates the expression of FGF3 and 10 in
the mesenchyme (Klein, Lyons et al. 2008). This FGF epithelial-mesenchymal signaling
loop is finely regulated by members of the Sprouty family, which act as intracellular
inhibitors of the FGF signaling pathway (Mason, Morrison et al. 2006). In the developing
incisor, Sprouty genes are dynamically expressed according to the embryonic stage, and
are expressed in the epithelium and/or mesenchyme. The function of Sprouty members is
to limit the formation of enamel producing ameloblasts on the labial aspect of the incisors
by inhibiting the establishment of a lingual FGF epithelial-mesenchymal signaling loop
(Klein, Lyons et al. 2008). The inactivation of Sprouty genes corresponds to a gain of
function of FGF signaling. These results support our own conclusions that FGFR2b
signaling positively controls ameloblast lineage formation during homeostasis
We are the first group to demonstrate that, FGFR2b signaling does not control the
survival of ASCs (see our model in Fig. 8). This is supported by the regeneration of
maxillary and mandibular incisors after long-term attenuation of FGFR2b signaling.
Interestingly, a similar conclusion was reached for FGFR2b signaling in epithelial stem
145
cells in the adult mammary gland (Parsa, Ramasamy et al. 2008). Inhibition of FGFR2b
signaling from birth leads to the formation of rudimentary mammary epithelial trees,
which are capable of resuming growth when the inhibition of FGFR2b signaling is
released. Inhibition of the FGFR2b signaling during puberty, when trees are forming,
leads to loss of the terminal end buds, a putative niche for adult mammary
stem/progenitor cells. In the future, it will be critical to better characterize the ASCs. This
aim will be investigated in our model system where the transit amplifying population of
APCs is reduced. In addition, it is interesting to ponder whether or not the regenerative
capability of human teeth could be linked to FGFR2b signaling. Indeed a genetic
deficiency in the FGFR2b signaling pathway can severely affect the production of enamel
in humans. Amelogenesis imperfecta is part of a relatively rare group of inherited
disorders characterized by abnormal enamel formation. Lacrimo-auriculo-dento-digital
(LADD) syndrome has been linked to a mutation in FGF10/FGFR2b (Shams, Rohmann
et al. 2007). Patients with LADD exhibit tooth dysplasia due to a amelogenesis
imperfecta-like defect. Our mouse model is therefore in agreement with a critical role of
FGFR2b signaling in ameloblast formation in human.
Our study demonstrates that FGFR2b signaling is critical for incisor homeostasis and
that activation of FGFR2b signaling, mediated by FGF10 or FGF3, could be instrumental
in promoting tooth regeneration in adults.

146
Chapter 2 Supplementary Data

Fig. 42. Validation of the Rosa26
rtTA
; tet(O)sFgfr2b double transgenic system during
embryonic development. Control (A-C) or DTG (D-F) embryos exposed to doxycycline
from E9.5 to E18.5. Note the absence of eyelid and limb and the curly tail (arrow). High
magnification of the palate of control (B) and DTG (E) embryos showing the presence of
a cleft palate. Sagital sections of (B) and (E) stained with H&E illustrating the defective
closure of the palatal shelves in the DTG embryo (arrows).

147

Fig. 43. Impact of down-regulation of FGFR2b first appears at the posterior end of
the maxillary incisors. Histological staining of control (A-C) and DTG (A’-C’)
maxillary incisors after 2 weeks of treatment with doxycycline. Sagital sections were
made in the anterior (A), median (B) and posterior (C) part of the incisor. Please note the
loss of ameloblast layer in DTG incisors as sections move from anterior (A) to posterior
(C) towards the cervical loop (CL). IEE: inner enamel epithelium. CL: Cervical loop.

148
Chapter 3
FGF10/FGFR2b Signaling controls the Progressive Formation
of the Skeletal Elements in the Limb Autopod
∗

Chapter 3 Abstract
Epithelial-mesenchymal interaction controls both growth and patterning in the
developing limb. Fibroblast Growth Factor 10/Fibroblast Growth Factor Receptor 2b
(FGF10/FGFR2b) signaling has been shown to control limb bud initiation but the role of
this signaling pathway beyond this early phase is still unclear. Using genetically modified
mouse allowing inducible and reversible expression of soluble FGFR2b acting as a
dominant negative to attenuate FGFR2b signaling, we demonstrated that FGFR2b
signaling is critical for the progressive formation of the skeletal elements in the autopod.
In our mouse model, attenuation of FGFR2b signaling for 4 hours leads to a defect in
AER. This defect was detected using both Fgf8 in situ hybridization and Topgal reporter
line. This phenotype is reversible after E8.5, however, the morphology of the limb buds

∗
The manuscript for this study is under preparation. Sara Parsa
1,2
, Suresh Ramasamy
2
,
Reza Tabatabai
2
, Denise Al Alam
2
, Robert Kelly
4
, Mohammad Hajihosseini
5
, Saverio
Bellusci
1,2,3
. (
1
Department of Pathology, Keck School of Medicine, University of
Southern California, Los Angeles, CA 90089, USA,
2
Developmental Biology and
Regenerative Medicine Program, Saban Research Institute of Childrens Hospital Los
Angeles, Los Angeles, CA 90027, USA,
3
University of Giessen Lung Center, Department
of Internal Medicine II, Klinikstrasse 36, 35392 Giessen Germany.
4
Avenir Research
Group Developmental Biology Institute of Marseille- Luminy IBDML- UMR 6216
CNRS, Marseilles, France,
5
University of East Anglia. Correspondence:
sbellusci@chla.usc.edu)

149
are altered. FGF10 is the principal ligand binding to FGFR2b during limb development.
Using a Fgf10
LacZ
reporter line to follow Fgf10 expression we demonstrate that at E10.5
Fgf10/lacZ is expressed in the limb mesenchyme at higher levels in the forelimb
compared to the hindlimb. At E13.5 Fgf10/LacZ is expressed at high levels at the tip of
the future digits. Generation of Fgf10 hypomorph embryos was carried out to determine
the impact of decreasing Fgf10 expression on limb formation. Most of Fgf10 hypomorph
embryos displayed hindlimb defects with either absent or reduced skeletal elements in the
autopod while forelimbs are overall normal. Altogether, our results highlight the
differential Fgf10 gene dosage requirements on hindlimbs vs. forelimbs and establish the
critical role played by FGF10/FGFR2b signaling throughout the outgrowth phase of limb
development to allow progressive formation of skeletal elements in the autopod.
Chapter 3 Introduction
Epithelial-mesenchymal interaction controls both growth and patterning in the
developing limb. Limb development initiates by formation of four buds along the body
flank at the presumptive limb locations. At the cellular level, the primary limb bud has
three topologically distinct signaling centers: the Apical Ectodermal Ridge (AER) at the
distal tip, the zone of polarizing activity (ZPA) at the posterior limb bud and the
Undifferentiated Zone (UZ) in the mesenchyme underlying the AER (Saunders 1948;
Summerbell, Lewis et al. 1973). After induction and outgrowth, the limb is patterned
along the proximal-distal, antero-posterior and dorsal-ventral axes. Along the proximal-
distal axis the limb can be divided into three domains: the stylopod (humerus), the
zygopod (radius and ulna) and the autopod (carpal/tarsal, metacarpal/metatarsal,
150
phalanges). The stylopod forms when progenitor cells located in the UZ leave this zone
and under the impact of signals from the flank contribute to the development of this
proximal part of the limb. Then the reciprocal interactions between the AER and PZ
command the growth and patterning along the proximal-distal axis to form zygopod and
autopod (Tabin and Wolpert 2007). Fibroblast Growth Factors (FGFs) play important
roles in different stages of the limb development. For example, FGF10 expressed by the
lateral plate mesoderm acts via its main receptor Fibroblast Growth Factor Receptor 2b
(FGFR2b) expressed in the ectoderm to control initial limb bud induction (Sekine,
Ohuchi et al. 1999). This induction of the ectoderm results in expression of Fgf8 in the
ectoderm and specification of ectoderm to the AER. However, Fgf10 is also expressed
during the subsequent phases of limb development in the UZ (de Maximy, Nakatake et al.
1999), therefore suggesting a role for FGF10/FGFR2b signaling beyond the induction
phase.
Indeed, critical role for FGFR2b signaling beyond the induction phase has been
recently demonstrated. Inactivation of Fgfr2 specifically in the AER leads to the absence
of the hindlimbs and autopods in the forelimbs (Lu, Yu et al. 2008; Yu and Ornitz 2008).
These reports using the Msx2-Cre driver line to target the limb ectoderm, demonstrate
that inactivation of Fgfr2 in the epithelium of the limb leads to the ablation of the AER.
Consistent with AER removal, decreased AER-Fgf8 and complete loss of autopod
skeletal elements were described. The difference between the forelimb and hindlimb
phenotype in these mutant embryos can be explained by the fact that Cre expression in
the Msx2-Cre transgenic line occurs after forelimb initiation (E9.5) but before hindlimb
bud initiation (E10) (Lewandoski, Sun et al. 2000). This approach therefore allowed
151
determining the role of FGFR2b signaling post limb bud induction. The impact of the
genetic ablation of Fgfr2b in the AER on the forelimb mesenchyme is still controversial.
Lu et al (2008) reported normal cell proliferation and apoptosis in the distal mesenchyme
of the Fgfr2
Msx2-Cre
forelimb mutants. Using the same mouse model, Yu and Ornitz (2008)
described normal cell proliferation at E10 and reduced proliferation at E10.5 and E11.5 in
the forelimb of the mutant embryos, while no cell death was detected in the mutant
mesenchyme. Interestingly, using Hoxa13 expression to follow the formation of autopod
progenitors of the distal limb, Lu et al reported that there is a delay in the generation of
autopod progenitors in Fgfr2
Msx2-Cre
forelimbs suggesting that reduction in the pool of
autopod progenitors is the leading cause of the observed impaired autopod formation.
In this paper, we address the role of epithelial FGFR2 signaling post autopod
induction. To accomplish this goal, double transgenic mice allowing inducible and
reversible attenuation of FGFR2b signaling by doxycycline treatment were generated as
previously described (Parsa, Ramasamy et al. 2008). This genetic system is a powerful
tool to explore the role of FGFR2b signaling at different times during organogenesis. We
induced the expression of soluble FGFR2b at different stages of limb development and
identified the time window where expression of soluble FGFR2b is critical for the
formation of the different skeletal elements found in the autopod. Using
immunohistological approaches we have detected decrease in apoptosis of the
mesenchyme as a result of abolishing epithelial FGFR2b in the AER. Also, the
reversibility of the phenotype was tested by a single dose injection of doxycycline which
results in the loss of proper morphology in the limb bud. Using Fgf10 hypomorphic
embryos (Mailleux, Kelly et al. 2005; Ramasamy, Mailleux et al. 2007), we have also
152
determined the function of FGF10, as a main ligand for FGFR2b, during limb
development. Altogether, this study demonstrates the critical role of FGF10/FGFR2b
signaling during the post-induction phases of limb development, especially autopod
formation.
Chapter 3 Material and Methods
Animals
To generate the inducible mouse model which expresses soluble FGFR2b
(Rosa26
rtTA/+
;tet(O)sFgfr2b
+/-
), CMV-Cre mice were first crossed with rtTA
flox
mice and
(Belteki, Haigh et al. 2005) then, the Rosa26
rtTA
mice were crossed with tet(O)sFgfr2b
mice (Hokuto, Perl et al. 2003). The double transgenic mice will be called thereafter
DTG mice. These mice allowed inducible expression of soluble FGFR2b in the embryo at
different stages following doxycycline treatment. The Topgal reporter cassette (DasGupta
and Fuchs 1999) was also introduced in the DTG transgenic mice. Animals were crossed
and time pregnant females were either put on doxycycline food (Rodent diet with
0.0625% doxycycline, Harlan Teklad TD01306) for several days or injected
intraperitoneally (IP) with a single dose of doxycycline (1.5 mg/kg of mouse in PBS) at a
specific developmental stage. Pregnant females were euthanized to collect embryos at
specific stages. All samples were fixed in 4% PFA and dehydrated in successive bathes
of increasing Ethanol concentration for further studies. All the animal experiments
described were done following IACUC regulations.

153

Bone and Cartilage Staining
Embryos at various stages were collected. After peeling the skin, the embryos were
eviscerated. The samples were fixed in 95% EtOH for three days, changing ethanol
every day and subsequently, stained for cartilage in 0.3% alcian blue 8GS in freshly
prepared staining solution at room temperature for two days. Embryos were then washed
in ddH2O for 2 hours and stained in 0.2% alizarin red S in 2.5% KOH for two days at
room temperature. Samples were cleared through 20%, 50%, and 80% glycerol /2.5%
KOH for 1 to 3 weeks depending on the size of the specimen. All stained samples were
stored in glycerol and documented.
LacZ staining
Fgf10
LacZ
(Mailleux, Kelly et al. 2005) or Topgal (DasGupta and Fuchs 1999)
embryos were fixed with 4% PFA and stained with the X-gal substrate from one hour to
overnight depending on the sample size. After sufficient staining was obtained the
samples were washed in PBS, fixed again with 4% PFA and documented.
Immunohistochemistry
Limb buds were embedded in paraffin sectioned and stained for various analyses.
Apoptotic cells were detected by incorporation of terminal deoxynucleotidyl transferase
mediated UTP nick-end labeling (TUNEL) using the “In Situ Cell Death Detection
Fluorescein kit” (Roche Applied Science) as recommended by the manufacturer. The
total number of cells was scored in five photomicrographs (64x magnification) of three
154
independent control and mutant limbs. A total number of 500 cells were counted per
sample. Cell proliferation studies were performed using the PCNA staining kit from
Invitrogen. Antibodies against mouse FGFR2 (Bek, 1:200, Santa Cruz, Inc) were used.
In Situ Whole Mount Hybridization
Embryos were fixed in 4%PFA, washed and dehydrated. In situ hybridization was
performed with digoxigenin labeled UTP RNA probes according to a modified protocol
from Bellusci et al. (Bellusci, Henderson et al. 1996).
Detection of luciferase activity
[Rosa26
rtTA/rtTA
] males were crossed with [tet(O)luciferase
+/-
] females and WT
females to generate experimental pregnant females carrying
[Rosa26
rtTA/+
;tet(O)luciferase
+/-
] or control pregnant females carrying [Rosa26
rtTA/+
]
embryos. When the embryos reached E11.5, the control and experimental pregnant
mothers were injected IP with doxycycline (1.5 mg/kg of mouse in PBS). A single dose
of Luciferine (1 mg of D-luciferine potassium salt in 0.2 ml of 1xPBS) was injected IP 20
minutes before imaging in vivo luciferase activity using the Xenogen detection system.
Imaging was carried out at 0, 5, 113, 242 minutes after doxycycline injection. The mice
were kept under anesthesia (isoflurane) during the image acquisition of luciferase
activity.

155

Chapter 3 Results
FGFR2b signaling is critical to control progressive limb growth along the proximal-
distal axis
Genetically modified mice allowing ubiquitous induction of soluble Fgfr2b
expression by doxycycline treatment, acting as a dominant negative receptor (Parsa,
Ramasamy et al. 2008), were used to inhibit FGFR2b signaling pathway at specific
developmental stages and study the outcome on limb development. Doxycycline
treatment of the [Rosa26
rtTA/+
;tet(O)sFgfr2b
+/-
] (DTG) embryos leads to the activation of
the transactivator rtTA expressed constitutively from the ubiquitous Rosa26 locus. This
in turn allows the transcription of a fusion transcript where the second and third
extracellular immunoglobulin domains of FGFR2b are fused with the mouse lg heavy
chain and Fc domain. The resulting fusion protein acts as a soluble antagonist of FGFR2b
by trapping FGFR2b ligands and preventing them from acting on their endogenous
receptors.
Pregnant females carrying DTG and control embryos were put on doxycycline food
between E8.5 and E13.5. DTG embryos exposed to doxycycline (Dox) food from E8.5 to
E13.5 display total limb agenesis (Fig. 44-A,B) as previously reported for the classical
inactivation of Fgfr2b (De Moerlooze, Spencer-Dene et al. 2000). By contrast, DTG
embryos exposed to dox food from E10.5, at a time where forelimb induction but not
hindlimb induction occurred, display shorter forelimb and absence of hindlimb (Fig. 44-
C,D). Dox treatment from E11.5 leads to absence of autopod formation in both forelimb
156
and hindlimb in the DTG embryos (Fig. 44-E-G). In order to better visualize the
formation of cartilage condensation at later developmental stages, the Topgal cassette
used as a reporter for canonical Wnt signaling (DasGupta and Fuchs 1999), was inserted
into the DTG experimental or single transgenic control embryos. Dox treatment from
E13.0 to E16.5 leads to generally shortened digits with digits 1 and 5 being more affected
(Fig. 44-H-K). Dox treatment from E13.5 to E16.5 shows a normal number of digits but
all of them are truncated distally, similar to a human pathological condition known as
brachydactyly. Interestingly, the forelimbs are slightly less affected than the hindlimbs,
indicating that at E13.5, the development of the forelimbs and hindlimbs is not
synchronized.
Cartilage staining of the mutant and wild type limbs at this stage was carried out. In
the control forelimb, cartilage primordial of carpal and metacarpal bones are observed. In
addition, three phalanges (p1, p2, p3) with normal absence of ossification for this stage
are detected (Fig. 44-P). In the control hindlimb, the same number of tarsal, metatarsal
and phalanges are detected (Fig. 44-R). Fig. 1S shows that the third phalange (p3) is
missing at the digit tip of the mutant hindlimbs. As previously described, a distal
rudimentary phalange is observed at the tip of digits 3 and 4 in the less affected mutant
forelimb (Fig. 44-Q). No obvious limb defects were detected when double transgenic
embryos where exposed to dox food at E14. However, it is difficult to analyze the
functionality of these limbs as these DTG embryos die at birth from abnormal lung
development (Al Alam and Bellusci, unpublished observation). We cannot therefore
exclude a role for FGFR2b signaling in limb development beyond E14. Overall, these
data indicate that FGFR2b signaling from E9.5 to E14 is required not only for limb
157
induction (De Moerlooze, Spencer-Dene et al. 2000) and autopod initiation (Lu, Yu et al.
2008; Yu and Ornitz 2008) as previously described but throughout all the developmental
phases subsequent to autopod initiation.

158
Fig. 44. FGFR2b signaling is critical to control progressive limb growth along the
proximal-distal axis. This figure shows the defective development of the limb in the
DTG embryos treated by doxycycline at various developmental stages. Initiation of
doxycycline treatment of the DTG and control embryos before induction of the limb bud
at E8.5 until E13.5 leads to loss of both hindlimbs and forelimbs in DTG embryos (A-B).
Doxycycline treatment of DTG and control embryos after the bud induction in the
forelimbs at E10.5 demonstrates absence of autopod in DTG embryos. At this stage,
hindlimbs are completely absent due to the time difference in bud induction between
forelimb and hind limb (C-D). Induction of DTG embryos at E11.5 after the initiation of
both forelimb and hindlimb bud formation demonstrates great defects in hindlimb and
loss of autopod in the forelimb (E-G). Also, induction of DTG embryos at E13.0 leads to
the loss of distal part of autopod in both forelimb and hindlimb (H-K). The reduction in
lacZ expression, which indicates the activation of the Wnt signaling, is also evident in the
treated DTG embryos (H-K). The bone staining for DTG and control embryos treated
from E13.5 to E16 demonstrates the reduction in the size of the last phalange in the
forelimb and complete loss of last phalange in the hindlimb (L-S).

159
(Fig. 44 continued)

160
Attenuation of FGFR2b signaling leads to the loss of functional AER
Previously published reports indicate that deletion of Fgfr2 expression in the AER
using the Msx2-Cre driver line leads to the reduction of AER-FGF signaling and
complete loss of autopod skeletal elements (Lu, Yu et al. 2008; Yu and Ornitz 2008). In
order to visualize AER defects in our DTG embryos, we crossed them with the Topgal
transgenic reporter, which allows visualizing the activation of canonical Wnt signaling
reported to be expressed specifically in the AER (Kengaku, Capdevila et al. 1998).
Females carrying either control or DTG embryos with the Topgal allele were fed with
doxycycline-containing food from E10.5 to E12.5. Fig. 2A-C shows the robust and
specific expression of LacZ in the AER of the forelimb and hindlimb. By contrast, DTG
embryos show a strong reduction of LacZ expression in the AER. Note that LacZ
expression is more severely affected in the hindlimb compared to the forelimb (Fig. 45-
D-F). We also examined Topgal expression in the control (Fig. 45-K-M) and DTG (Fig.
45-N-P) embryos exposed to doxycycline from E12.5 to E13.5. At this time point, Topgal
expression in the hindlimb and forelimb of the DTG embryos is similarly reduced. These
results indicate that Wnt signaling is affected in the AER upon FGFR2b attenuation and
suggest that the AER is no longer functional.
To test this possibility, we directly checked the expression of Fgf8 expression by ISH
in the AER of control (Fig. 45-G,H) and DTG (Fig. 45-I,J) embryos exposed to
doxycycline from E10.5 to E11.5. Our results indicate a significant decrease in the
expression of AER-Fgf8 in DTG embryos. Please note that similar to the changes in
Topgal expression, Fgf8 expression is more severely affected in the hindlimb compared
161
to the forelimb (Fig. 45-J). Therefore, our results demonstrate the loss of functional AER
after attenuation of the FGFR2b signaling.

Fig. 45. The loss of functional AER as a result of downregulation of FGFR2b
signaling. LacZ staining of the DTG and control embryos carrying the Topgal reporter
transgene. In the control embryos, AER appears as a thin layer of the lacZ positive cells
at the very tip of the limb bud at E12.5 (A-C). In the treated DTG embryos, however, the
limb bud is smaller and the lacZ staining is less and diffused along the distal part of the
limb bud (D-F). In situ hybridization for Fgf8 (specific marker for AER) demonstrates
the loss of Fgf8 expression in the AER of the DTG embryos treated with Dox food from
E10.5-E11.5 (G-J).

Extremely rapid loss of AER upon FGFR2b attenuation
The critical developmental window (E8.5 through E10.5) where major decisions in
terms of limb development occur mostly through the amplification of mesenchymal
progenitors. In our system, the efficiency of FGFR2b expression depends on the intake of
doxycycline-containing food by the pregnant female which is difficult to control
162
temporally. Therefore, we decided to refine our experimental approach by injecting a
single dose of doxycycline intraperitoneally (IP). We first aimed to define the timing of
tet(o) promoter activation in vivo using luciferase as a reporter gene, after doxycycline
injection as well as the reversibility of this activity upon doxycycline removal. To
perform this experiment, we used pregnant females carrying either
Rosa26
rtTA/+
;tet(O)Luciferase
+/-
or wild type embryos and imaged the luciferase activity
at different time points after doxycycline IP injections. While doxycycline was injected
once at T(0 min), luciferine which is the substrate for luciferase was injected 15 minutes
before imaging at 0, 5, 113 and 242 minutes after doxycycline IP injection. While no
significant expression was detected in the female carrying wild type embryos (data not
shown), our results indicates that in the female carrying Rosa26
rtTA/+
;tet(O)Luciferase
+/-

embryos, luciferase activity was detected as early as 5 minutes after dox injection with a
peak at 113 minutes and a return to base line at 242 minutes (Fig. 46-A-D). Hence, our
results indicate that the primary analysis of developmental changes has to be done within
4 hours following single IP doxycycline injection.
We also injected pregnant females carrying DTG and control embryos with a single
dose of doxycycline at E11 and examined 4 hours later the impact on the forelimb
development (please note that similar results were obtained in hind limb- data not
shown). While a clear AER was observed in forelimb controls (Fig. 46-E,F), such
structure is not observed in the forelimb of DTG embryos (Fig. 46-G, H). Therefore, in
addition to rapid activation of the promoter in the response to the IP injection of
doxycycline, the loss of the AER occurs within the first four hours after IP injection. This
result demonstrates that our system acts quickly and efficiently.
163
We also analyzed the cell death in these samples using TUNEL assay. While TUNEL
positive cells where observed in AER of control limbs (Fig. 46-I,J), no TUNEL positive
cells were observed in the DTG limb either in the epithelium or in the mesenchyme (Fig.
46-K, L). Decrease in the rate of apoptosis in the AER of the DTG embryos indicates that
the AER does not disappear as a result of cell death in the absence of FGFR2b signaling.
Also, the loss of apoptosis in the DTG limb indicates that apoptosis play a vital role in
proper development of limb as early as bud stage.

164

Fig. 46. Rapid loss of AER upon FGFR2b attenuation. Imaging of a wild type
pregnant female and a single transgenic female carrying the Rosa26
rtTa/+
; tet(o)luciferase
embryos. Images were taken at 5 (A), 113 (B) and 242 (C) min after IP injection of the
doxycycline. As illustrated on the graph (D), the promoter was activated 5 min after
injection and remained active for four hours. Also, as histological analyses demonstrate,
the AER disappears in the DTG embryos within the first four hours after IP injection of
doxycycline at E11 (E-H). TUNEL analyses show reduction in apoptosis in the DTG
embryos within the four hours after IP injection of the doxycycline in comparison to wild
type embryos (I-L).

165
Functional ablation of the AER is reversible after doxycycline IP injection after
E8.5
To examine the eventual reversibility of the observed developmental defects after a
transient inhibition of the FGFR2b signaling at different critical time points, we injected
pregnant females carrying DTG and control embryos with a single IP dose of
doxycycline at E7.5, E8.5, E9.5 and E10.5 and analyzed the resulting phenotype at E12.5.
DTG embryos exposed to doxycycline at E7.5 (2 days before the forelimb induction and
2.5 days before hindlimb induction) did not show any limbs at E12.5 (Fig. 47-E,F). This
suggests that FGFR2b signaling plays a completely unsuspected role two days before
limb induction. DTG embryos injected at E8.5 (1 and 1.5 days before forelimb and
hindlimb induction respectively) display rudimentary limbs at E12.5 (Fig. 47-H,I). As
shown previously, this is the stage when the continuous exposure to doxycycline leads to
the limb agenesis (Fig. 44-B). This result, therefore, demonstrates the reversibility of the
phenotype upon transient inactivation of the FGFR2b signaling at E8.5. DTG embryos
injected at E9.5 (corresponding to the induction time for the forelimb and the pre-
induction time for the hindlimb) also show rudimentary limbs (Fig. 47-K,L) indicating
partial recovery upon doxycycline removal. DTG embryos injected at E10.5 (post limb
induction) show delayed limb formation (Fig. 47-N,O) compared to the E12.5 control
limbs (Fig. 47-B,C). Altogether, our data demonstrates the reversibility of the phenotype
after doxycycline treatment at these stages. However, the shape and size of the limb bud
is not similar to the wild type limb bud which indicates the important role FGFR2b
signaling plays in the patterning of the limb bud at stages as early as E8.5.
166

Fig. 47. The reversibility of the loss of the AER after doxycycline IP injection.
The panel depicts the limb phenotypes in DTG embryos after transient treatment with
doxycycline at various stages of development. E12.5 wild type embryo showing normal
forelimb and hindlimb development used as control (A-C). One time shot of IP
doxycycline to the DTG embryo at E7.5. Embryos were collected at E12.5. The embryo
demonstrates limb agenesis in addition to the various developmental defects (D-F). One
time shot of IP doxycycline at E8.5 (G-I), E9.5 (J-L) shows abnormally developed
forelimb (H, K) and hindlimb (I, L) of the embryo. Also, one time shot of IP doxycycline
to the DTG embryo at E10.5 leads to the lack of proper development in the limbs (M-O).
This DTG embryo was collected at E14.5.

167
An Fgf10
LacZ
reporter line can be used to monitor Fgf10 expression in the limbs
FGF10 is the primary ligand for FGFR2b during the limb development as
demonstrated by the remarkable similarity of phenotypes exhibited by embryos where
these genes have been inactivated (De Moerlooze, Spencer-Dene et al. 2000; Ohuchi,
Hori et al. 2000; Mailleux, Spencer-Dene et al. 2002). However, the exact expression
pattern of Fgf10 beyond the induction phase has not been reported so far. We have
previously reported the use of the Mlc1v-nlacZ-24 (also called Fgf10
LacZ
) transgenic line
to follow Fgf10 expression in the lung (Mailleux, Kelly et al. 2005), gut (Tai, Tu et al.
2008) and liver (Berg, Rountree et al. 2007). We used this previously published LacZ
reporter line to follow Fgf10 expression.
Using the Fgf10
LacZ/+
reporter line we monitored precisely Fgf10 expression in the
developing limbs at E8.5, E10.5, E11.5 and E13.5. As previously reported for Fgf10
expression pre limb bud induction (Kawakami, Capdevila et al. 2001) at E8.5, Fgf10
expression is detected in the lateral plate mesoderm (Fig. 48-A). At E10.5, Fgf10
expression is detected at high level in the forelimb mesenchyme (Fig. 48-B, C) and at
lower level in the hindlimb mesenchyme (Fig. 48-B, D). At E11.5, Fgf10 expression is
detected only in the posterior half of the limb buds and is still higher in the forelimb
compared to the hindlimb (Fig. 48-E-G). At E13.5, the expression of Fgf10 is found at
high level at the tip of the digits (Fig. 48-H). Again at this stage, the expression level of
Fgf10 in the hindlimb is significantly lower compared to the one observed in the forelimb
(Fig. 48-I,J). These data indicate that Fgf10 is expressed post bud induction until at least
E14.5 (data not shown).
168
To confirm the difference in Fgf10 expression of the forelimb vs. hindlimb, real time-
PCR was performed on the hindlimbs and forelimbs of the wild type embryos at E11.5
(data not shown). Our results indicate 1.9X increase in the expression of Fgf10 in the
forelimbs vs. hindlimbs which confirms our previous results on Fgf10
LacZ
reporter line.
Higher expression of Fgf10 in the forelimb compared to the hindlimb of the wild type
embryos explains the dissimilarity of the defects between the hindlimb and forelimb of
our DTG embryos induced at post bud induction.

169

Fig. 48. Expression pattern of Fgf10 during limb development using the Fgf10
LacZ

reporter line. Expression pattern of lacZ reporter in Fgf10
nlacZ/+
embryos. Expression
pattern of Fgf10 at E8.5 (A). The lacZ staining demonstrates higher expression of Fgf10
in the presumptive limb location for the forelimbs in the comparison to the presumptive
location of the hindlimbs. LacZ staining at E10.5 (B), E12.5 (E) and E13.5 (H) shows
higher expression of Fgf10 in the mesenchyme of the forelimb (C, F, I) than the hindlimb
(D, G, J). At E13.5 (H), the expression of Fgf10 is more localized to the mesenchyme
underlying the tips of the digits (I-J). In all of the above stages, the expression of Fgf10
is higher in the posterior (p) side of the limb than the anterior (a).

Hindlimbs of Fgf10 hypomorphic embryos display impaired proximal-distal growth
due to the defective of the AER.
We have previously reported that the insertion of the Mlvc1v-nLacZ cassette 120 kb
upstream of the transcriptional start site of the Fgf10 gene reduces Fgf10 expression.
Whole-mount in situ experiments showed a general decrease in Fgf10 expression in
170
Fgf10
LacZ/LacZ
hemizygous embryos when compared with the wild-type littermates (Fig.
49-A,B). In particular, the expression of Fgf10 in the hindlimb of the mutant embryos is
significantly reduced (Fig. 48-C,D). This confirms that the insertion of the transgenic
cassette creates a hypomorphic Fgf10 allele. We have therefore generated allelic series by
crossing hemizygous or homozygous Fgf10
LacZ
mice with heterozygous Fgf10 null mice.
Forty-eight Fgf10
LacZ/-
compound heterozygous embryos were obtained at Mendelian
ratio between E12.5 and P0. At E12.5, Fgf10
LacZ/-
display normal forelimb but severely
truncated hindlimb compared with wild type (Fig. 49-E,F). A similar phenotype is
observed at E14.5 (data not shown). Altered expression of Fgf8 using in situ
hybridization demonstrates the presence of the defective AER of the hindlimbs in the
Fgf10
LacZ/-
embryos at E11.5 (Fig. 49-G,G’, H,H’). At E18.5, a stump instead of well-
developed hindlimb is observed (Fig. 49-I,J). Mutant embryos display only hindlimb
defects regardless of their location on the right or left side of the body. A few of the
Fgf10
LacZ/-
newborn pups survived up to 48 hours postnatal and displayed fully developed
limbs. As during the embryonic stages, these mutant mice have phenotypes only in the
hindlimbs, with normally developed forelimbs. The hindlimb phenotype varies from
mild malformation of digits to the loss of whole limb. In rare instances where hindlimb
development could proceed (3 out of the 16 mutants examined), the formation of
proximal and distal skeletal elements was examined by alcian blue/alizarin red staining.
In the mutant hindlimb, our results indicate the presence of normal proximal elements
(humerus, radius and ulna- compare Fig. 49-M, K) but missing ossification centers in
distal phalangeal bones arising from the distal phalange 3 (arrows in Fig. 49-L, N) as well
as defective autopod with missing digits.
171

Fig. 49: Fgf10 hypomorphic embryos display hindlimb abnormalities. Panel shows
the hindlimb phenotypes in Fgf10
lacZ/neo
at different stages of development. In situ
hybridization showing the Fgf10 expression in a wild type (A) and Fgf10
nlacZ/nlacZ
(B)
embryos at E10.5. The expression of Fgf10 in the hindlimb of Fgf10
nlacZ/nlacZ
embryo (D)
is less than its expression in the wild type hindlimb bud (C). Fgf10
lacZ/-
embryos at E12.5
(F) and E18.5 (J) showing hindlimb malformation in comparison to the wild type
embryos at E12.5 (E) and E18.5 (I). Deficient expression of Fgf8 in the hindlimb of the
Fgf10
lacZ/-
embryo using in situ hybridization (G-G’, H-H’). Note the presence of Fgf8
expression in the AER of the forelimb of the Fgf10
lacZ/-
embryo. Bone and cartilage
staining of the hindlimb of the wild type at E18.5 shows clear distal phalange (K-L).
Bone and cartilage staining of the Fgf10
lacZ/-
hindlimb demonstrates loss of digits and loss
of the distal phalanges in the digits present (M-N).

172
Chapter 3 Discussion
The Fibroblast Growth Factors (FGF) play important roles in vertebrate development
(Ornitz and Itoh 2001). FGFR2b is the main receptor for FGF10 during the embryonic
development as evidenced by the remarkable similarity of phenotypes displayed by Fgf10
and Fgfr2b null embryos (De Moerlooze, Spencer-Dene et al. 2000; Ohuchi, Hori et al.
2000; Mailleux, Spencer-Dene et al. 2002). Mice deficient for Fgf10 show multiple organ
defects including limb agenesis (Min, Danilenko et al. 1998; Sekine, Ohuchi et al. 1999;
Ohuchi, Hori et al. 2000). In addition, inactivation of Fgfr2b in the embryo also to limb
agenesis (De Moerlooze, Spencer-Dene et al. 2000). Previous studies, using the Msx2-
cre;Fgfr2
fl/+
mouse model showed a loss of the distal part of the limb (autopod) in the
absence of FGFR2b in the AER (Lu, Yu et al. 2008; Yu and Ornitz 2008). However, they
failed to describe the role of FGFR2b signaling in post limb bud induction (after E10.5)
since Msx2 is expressed around E9.5-E10 in the AER.
Our goal in this study is to determine the role of FGFR2b signaling at different post-
limb bud induction stages using an inducible and reversible mouse model in which the
endogenous FGFR2b signaling is reduced after treatment of the mouse with doxycycline.
Since the transactivator rtTA is expressed constitutively from the ubiquitous Rosa26
locus, the transcription of the soluble form of Fgfr2b occurs universally. The soluble
form of FGFR2b traps all FGFR2b ligands and prevents their binding to the membrane-
bound FGFR2b. We previously reported that induction of soluble FGFR2b from E8.5
(Parsa, Ramasamy et al. 2008) leads to a similar limb and lung agenesis phenotype as the
one observed in Fgf10 or Fgfr2b null mutant embryos (Sekine, Ohuchi et al. 1999; De
173
Moerlooze, Spencer-Dene et al. 2000). Strikingly, expression of dominant negative
FGFR2b also leads to the formation of a curly tail, a phenotype reported in the Fgfr2b but
not in Fgf10 null embryos (Parsa, Ramasamy et al. 2008) indicating the efficiency of this
approach to abolish completely and specifically FGFR2b signaling. Finally, the
inactivation of FGFR2b signaling in the post-natal stages leads to multiple defects
including defective maintenance of the terminal end buds in mammary gland
development (Parsa, Ramasamy et al. 2008). A similar phenotype has been observed
following Fgfr2 specific deletion in the mammary epithelium (Lu, Yu et al. 2008). These
results validated therefore our mice as an in vivo tool to conditionally inactivate FGFR2b
signaling at different stages of embryonic or post-natal development.
The rapid disappearance of the AER upon FGFR2b signaling attenuation using our
DTG system allows unraveling new mechanisms of action for FGFR2b signaling in AER
maintenance and opens the possibility to analyze the primary versus secondary defects
resulting from loss of functional FGFR2b on the AER itself and also on the limb
mesenchyme. The timing of FGFR2b inactivation in our system is fundamentally
different from the one aiming to genetically delete Fgfr2 expression using the Cre/Lox
technology (Lu, Yu et al. 2008; Yu and Ornitz 2008). This later system does not affect
the function of the already existing receptor at the time of the genetic deletion.
Interestingly, it takes at least 24 hours, from the time of Cre expression in the AER using
the Msx2-Cre driver line, to get the complete disappearance of the AER (from stage 29
through 45 somites roughly corresponding to E10.5 through E11.5- (Lu, Yu et al. 2008).
The appearance of the phenotype may therefore depend not only on the time that Cre is
expressed but also on the status of FGFR2b.
174
The results gathered so far with these mice in the post-natal stages in other organs
than limb indicate that FGFR2b signaling controls the amplification of transient
amplifying cells but not the survival of the early epithelial progenitors. For example, our
work on the role of FGFR2b signaling on post-natal mammary gland development shows
that FGFR2b downregulation leads to the disappearance of the terminal end buds and
simplification of the epithelial tree. This phenotype is completely reversible upon
resuming a doxycycline-free diet. Similar results have been obtained for the incisors,
which are constantly growing in rodents (Parsa, Ramasamy et al. 2008 and Kuremoto and
Bellusci, unpublished data). Our data on the early limb development also demonstrate the
reversibility of the phenotype and formation of the rudimentary limb bud after one time
IP injection of the doxycycline into the DTG embryos. However, this growth lack the
same patterning observed in the wild type limb bud which indicates the importance of the
temporal expression of Fgfr2b in the maintenance of the morphology of the limb.
In summary, FGFR2b signaling plays an important role in the formation and
maintenance of the AER in early limb formation. The later the Fgfr2b is downregulated
in the embryos, the fewer defects are observed in the limb formation. The attenuation of
Fgfr2b before the induction of the limb bud leads to the absence of the bud formation in
both forelimb and hindlimb. Also, the absence of FGFR2b signaling after induction
results in distal defects in the limb. The defect is reversible which indicates that FGFR2b
signaling is not crucial for the survival of the progenitor cells. Altogether, in this study
we report that FGFR2b signaling is important for the patterning of the distal part of the
limb.

175
Conclusion
The ‘stem cell niche’ was first proposed by R. Schofield after the observation that
hematopoietic stem cells demonstrate different behavior in different molecular
environments such as the spleen or the bone marrow (Schofield 1978). The niche
provides the molecular information for either the maintenance or survival of the stem
cells in a dormant state or the induction of proliferation and differentiation of the stem
cells. Active stem cell undergoes asymmetric cell division, which results in formation of
a stem cell and a more committed progenitor cell or symmetric cell division which gives
rise to two identical stem cells (Voog and Jones). The committed progenitor cells receive
signals to become more and more differentiated. Many of the signals controling the
maintenance/survival of stem cells and the proliferation and differentiation of transit
amplifying cells are known. However, many factors still remain to be identified (Fig. 50).
In our mouse model, we successfully reduced ubiquitously the activity of FGFR2b
ligands such as FGF10, FGF3 and FGF7. These ligands are present in the extracellular
matrix of the cells including in the stem cell niche of all organs of the mouse. While the
stem cell niche in most of the organs during homeostasis preserves the dormant condition
of the adult stem cells, the niche in the hematopoietic system, hair, skin and incisors
constitutively induces the growth and differentiation of the stem/progenitor cells. Our
data on mammary gland and incisors demonstrate that while FGFR2b signaling is not
required for the maintenance and survival of the adult stem cells, it plays an important

176

Fig. 50. Classical stem cell hierarchy. “Model of the `classical' hierarchy of
undifferentiated epithelial stem cell, transit amplifying (TA) progenitor cells and mature
postmitotic differentiated cells. Cell fate choices are indicated by red arrows. In this
model, the stem cell in its `niche' and different TA cell subclasses can self-renew (curved
arrows). Stem cells self-renew infrequently and TA cells more rapidly. Early TA cells
may be able to replace stem cells if the niche is depleted (dashed arrow 1). The niche
probably consists of several cell types and associated molecules, including blood vessels
and nerves. `Transdifferentiation' of one well-defined differentiated cell type into another
could occur directly, without cell division (dashed arrow 2) or might also involve
reversion or de-differentiation between distinct TA progenitor populations (dashed
arrows 3). Rarely, stem cells switch from one tissue-specific lineage to another (dashed
arrow 4) in a process called metaplasia or transdetermination (see Box 1). Adapted, with
permission, from Watt and Hogan” (Watt and Hogan 2000)

177
role in regulation of adult stem cell behavior. Based on our studies, we suggest that
FGFR2b signaling regulates the transit amplifying cell population in both mammary
gland and incisor. However, further studies have to be done to elucidate the exact role of
FGFR2b signaling in either regulation of transit amplifying cell proliferation or
asymmetric division of the adult stem cells. Therefore, understanding the complete role
of FGFR2b signaling leads to a better understanding of the adult stem cell niche in some
of the ectodermal derivatives such as mammary gland and incisors.
Besides development and homeostasis, FGFR2b signaling is also important in the
pathogenesis of other organs such as the breast. In 2008, a three-stage genome-wide
genome association study demonstrated the association of single nucleotide
polymorphism (SNPs) in FGFR2 with both ER-positive and ER-negative breast cancers
(Garcia-Closas, Hall et al. 2008). Later studies on breast cancer patients from different
origins show significant presence of FGFR2 mutation in these patients. Although most
types of breast cancers are from epithelial origin, stromal signals are also critical for the
prevention or the induction of cancer.
One of the complications in breast cancer treatment is the dormancy of the cancer
stem cells which results in recurrence of cancer after several years. Although it is not
clear whether cancer stem cells originate from normal adult cells or from terminally
differentiated stem cells, they demonstrate the same biological properties as the adult
breast stem cells (Morrison, Schmidt et al. 2008). Our studies demonstrate the importance
of FGFR2b signaling in the regulation of adult mammary stem cells. Also, we reported

178
that adult stem cells do not require FGFR2b signaling for their survival (Parsa,
Ramasamy et al. 2008). Therefore, we speculate that FGFR2b signaling induces breast
cancer by either increasing the proliferation of the luminal progenitor cells or by inducing
the differentiation of bipotent stem cells. Also, upregulation of Fgfr2b can be the result of
dedifferentiation of terminally differentiated cells into the transit amplifying cells and,
therefore, increase in the population of luminal progenitor cells. Either way, we conclude
that FGFR2b signaling plays a critical role in regulation of adult mammary stem cells
behavior. Based on this conclusion, we speculate that FGFR2 can play the same role in
regulation of cancer stem cell behavior. However, further studies need to be performed
specifically on breast cancer models to a better understanding of this subject.
Another significant study presented in this thesis is the role of FGFR2b signaling in
limb development. The data shown in chapter 3, can be applied to a better understanding
of tissue engineering and regenerative medicine. This branch of science seeks further
understanding of the molecular requirement of stem cell during normal development to
propose new biological treatments for organ replacement therapy. Due to many
complication of the organ transplantation, scientists from different disciplines of science
cooperate to make the dream of a functional engineered tissue closer to the reality. While
the progress in this field has been quick and findings have been tremendous, a lot still
needs to be done.
Three different groups of tetrapods have been identified in respect to their limb
regenerative capacity. The first group is able to regenerate their limbs during their

179
lifetimes. Urodele amphibians such as salamanders belong to the first group. The second
group is capable of limb regeneration before metamorphosis while unable to regenerate
after metamorphosis. Xenopus belongs to this group. Other tetrapods such as birds are
completely unable to regenerate new limbs at all the developmental stages considered.
Human like other mammals can only regenerate the digit tips(Borgens 1982; Reginelli,
Wang et al. 1995).
Limb regeneration after amputation consists of three steps: Wound healing;
dedifferentiation; and de novo induction of a developmental process. During the first
step, quick wound healing occurs by migration of epithelial cells which results in
covering the wound site allowing prevention of the inflammation (Repesh and Oberpriller
1978; Carlson, Bryant et al. 1998). This epithelium that covers the wound is called apical
epithelial cap or AEC which has a vey similar structure to AER during normal limb
development (Muneoka and Sassoon 1992). The next step is the dedifferentiation of the
underlying mesenchyme. This leads to blastema formation under the control of the AEC.
Therefore, the regeneration process tends to reproduce the same structures as the one
found in the early limb bud (AER≅AEC; Progress zone≅Blastema). The blastema, which
consists of undifferentiated mesenchymal cells, acts as a progress zone in the embryonic
limb bud. During the third step, the limb bud develops into functional limb. Loss of the
regenerative capacity in some tetrapods is the result of inflammation, lack of accelerated
wound healing and also, lack of signals that induce AEC and subsequent blastema
formation (Yokoyama 2008).

180
Our studies on embryonic limb development indicate the importance of FGFR2b
signaling pathway in coherence and formation of the AER in the limb bud. Based on this
study, we propose that FGFR2b signaling is required for the regeneration of the limb.
However, rebuilding an intact organ is not only dependent on differentiation but also
proper morphogenesis of the organ itself. Therefore, if one intends to rebuild a limb, in
addition to the presence of different important signaling pathway, FGF, BMP and WNT,
the proper interaction between these pathways is also required. Consequently, in the
future, further studies on the interaction between the major signaling pathways in limb
development have to be carried out in order to achieve limb regeneration.

181
Bibliography
Allen, B. L. and A. C. Rapraeger (2003). "Spatial and temporal expression of heparan
sulfate in mouse development regulates FGF and FGF receptor assembly." J Cell Biol
163(3): 637-48.
Alvarez, I. S., M. Araujo, et al. (1998). "Neural induction in whole chick embryo cultures
by FGF." Dev Biol 199(1): 42-54.
Alvi, A. J., H. Clayton, et al. (2003). "Functional and molecular characterisation of
mammary side population cells." Breast Cancer Res 5(1): R1-8.
Amaya, E., T. J. Musci, et al. (1991). "Expression of a dominant negative mutant of the
FGF receptor disrupts mesoderm formation in Xenopus embryos." Cell 66(2): 257-70.
Andl, T., K. Ahn, et al. (2004). "Epithelial Bmpr1a regulates differentiation and
proliferation in postnatal hair follicles and is essential for tooth development."
Development 131(10): 2257-68.
Arman, E., R. Haffner-Krausz, et al. (1998). "Targeted disruption of fibroblast growth
factor (FGF) receptor 2 suggests a role for FGF signaling in pregastrulation mammalian
development." Proc Natl Acad Sci U S A 95(9): 5082-7.
Armelin, H. A. (1973). "Pituitary extracts and steroid hormones in the control of 3T3 cell
growth." Proc Natl Acad Sci U S A 70(9): 2702-6.
Asaki, T., Konishi, M., Miyake, A., Kato, S., Tomizawa, M., and Itoh, N. (2004). "Roles
of fibroblast growth factor 10 (Fgf10) in adipogenesis in vivo." Mol Cell Endocrinol 218,
119-28.
Begue-Kirn, C., A. J. Smith, et al. (1994). "Comparative analysis of TGF beta s, BMPs,
IGF1, msxs, fibronectin, osteonectin and bone sialoprotein gene expression during
normal and in vitro-induced odontoblast differentiation." Int J Dev Biol 38(3): 405-20.
Bei, M., K. Kratochwil, et al. (2000). "BMP4 rescues a non-cell-autonomous function of
Msx1 in tooth development." Development 127(21): 4711-8.
Beiman, M., B. Z. Shilo, et al. (1996). "Heartless, a Drosophila FGF receptor homolog, is
essential for cell migration and establishment of several mesodermal lineages." Genes
Dev 10(23): 2993-3002.
Bel-Vialar, S., N. Itasaki, et al. (2002). "Initiating Hox gene expression: in the early chick
neural tube differential sensitivity to FGF and RA signaling subdivides the HoxB genes
in two distinct groups." Development 129(22): 5103-15.
182
Bellusci, S., J. Grindley, et al. (1997). "Fibroblast growth factor 10 (FGF10) and
branching morphogenesis in the embryonic mouse lung." Development 124(23): 4867-
78.
Bellusci, S., R. Henderson, et al. (1996). "Evidence from normal expression and targeted
misexpression that bone morphogenetic protein (Bmp-4) plays a role in mouse embryonic
lung morphogenesis." Development 122(6): 1693-702.
Belteki, G., Haigh, J., Kabacs, N., Haigh, K., Sison, K., Costantini, F., Whitsett, J.,
Quaggin, S. E., and Nagy, A. (2005). "Conditional and inducible transgene expression in
mice through the combinatorial use of Cre-mediated recombination and tetracycline
induction." Nucleic Acids Res 33, e51.
Bendall, S. C., M. H. Stewart, et al. (2007). "IGF and FGF cooperatively establish the
regulatory stem cell niche of pluripotent human cells in vitro." Nature 448(7157): 1015-
21.
Berg, T., C. B. Rountree, et al. (2007). "Fibroblast growth factor 10 is critical for liver
growth during embryogenesis and controls hepatoblast survival via beta-catenin
activation." Hepatology 46(4): 1187-97.
Bernfield, M., M. Gotte, et al. (1999). "Functions of cell surface heparan sulfate
proteoglycans." Annu Rev Biochem 68: 729-77.
Bitgood, M.J. and McMahon, A.P. (1995) "Hedgehog and Bmp genes are coexpressed at
many diverse sites of cell-cell interaction in the mouse embryo." Dev Biol. 172, 126-38.
Bocchinfuso, W.P., Lindzey, J.K., Hewitt, S.C., Clark, J.A., Myers, P.H., Cooper, R. and
Korach, K.S. (2000) "Induction of Mammary Gland Development in Estrogen Receptor-
Knockout Mice." Endocrinology 141, 2982–2994
Borgens, R. B. (1982). "Mice regrow the tips of their foretoes." Science 217(4561): 747-
50.
Bottcher, R. T. and C. Niehrs (2005). "Fibroblast growth factor signaling during early
vertebrate development." Endocr Rev 26(1): 63-77.
Bottcher, R. T., N. Pollet, et al. (2004). "The transmembrane protein XFLRT3 forms a
complex with FGF receptors and promotes FGF signalling." Nat Cell Biol 6(1): 38-44.
Boulet, A. M., A. M. Moon, et al. (2004). "The roles of Fgf4 and Fgf8 in limb bud
initiation and outgrowth." Dev Biol 273(2): 361-72.
Bray, S. (1998). "Notch signalling in Drosophila: three ways to use a pathway." Semin
Cell Dev Biol 9(6): 591-7.
183
Brito, A., L. A. Escalona, et al. (2008). "Periodontal conditions and distribution of
Prevotella intermedia, Porphyromonas gingivalis and Aggregatibacter
actinomycetemcomitans in HIV-infected patients undergoing anti-retroviral therapy and
in an HIV-seronegative group of the Venezuelan population." Acta Odontol Latinoam
21(1): 89-96.
Brookes, S., R. Smith, et al. (1989). "Sequence organization of the human int-2 gene and
its expression in teratocarcinoma cells." Oncogene 4(4): 429-36.
Brookes, S., R. Smith, et al. (1989). "The mouse homologue of hst/k-FGF: sequence,
genome organization and location relative to int-2." Nucleic Acids Res 17(11): 4037-45.
Capdevila, J., T. Tsukui, et al. (1999). "Control of vertebrate limb outgrowth by the
proximal factor Meis2 and distal antagonism of BMPs by Gremlin." Mol Cell 4(5): 839-
49.
Carballada, R., H. Yasuo, et al. (2001). "Phosphatidylinositol-3 kinase acts in parallel to
the ERK MAP kinase in the FGF pathway during Xenopus mesoderm induction."
Development 128(1): 35-44.
Carlesso, N., J. C. Aster, et al. (1999). "Notch1-induced delay of human hematopoietic
progenitor cell differentiation is associated with altered cell cycle kinetics." Blood 93(3):
838-48.
Carlson, M. R., S. V. Bryant, et al. (1998). "Expression of Msx-2 during development,
regeneration, and wound healing in axolotl limbs." J Exp Zool 282(6): 715-23.
Carver, E. A., R. Jiang, et al. (2001). "The mouse snail gene encodes a key regulator of
the epithelial-mesenchymal transition." Mol Cell Biol 21(23): 8184-8.
Chen, C., T. E. Spencer, et al. (2000). "Fibroblast growth factor-10: a stromal mediator of
epithelial function in the ovine uterus." Biol Reprod 63(3): 959-66.
Chen, Y., M. Bei, et al. (1996). "Msx1 controls inductive signaling in mammalian tooth
morphogenesis." Development 122(10): 3035-44.
Cho, K. W., J. Y. Kim, et al. (2006). "Molecular interactions between Tbx3 and Bmp4
and a model for dorsoventral positioning of mammary gland development." Proc Natl
Acad Sci U S A 103(45): 16788-93.
Chu, E. Y., J. Hens, et al. (2004). "Canonical WNT signaling promotes mammary
placode development and is essential for initiation of mammary gland morphogenesis."
Development 131(19): 4819-29.
Ciruna, B. and J. Rossant (2001). "FGF signaling regulates mesoderm cell fate
specification and morphogenetic movement at the primitive streak." Dev Cell 1(1): 37-
49.
184
Clark, J. C., J. W. Tichelaar, et al. (2001). "FGF-10 disrupts lung morphogenesis and
causes pulmonary adenomas in vivo." Am J Physiol Lung Cell Mol Physiol 280(4):
L705-15.
Clayton, H., I. Titley, et al. (2004). "Growth and differentiation of progenitor/stem cells
derived from the human mammary gland." Exp Cell Res 297(2): 444-60.
Cohen, M. M., Jr. and S. Kreiborg (1993). "An updated pediatric perspective on the Apert
syndrome." Am J Dis Child 147(9): 989-93.
Cohen, M. M., Jr. and S. Kreiborg (1993). "Growth pattern in the Apert syndrome." Am J
Med Genet 47(5): 617-23.
Cohen, M. M., Jr. and S. Kreiborg (1993). "Skeletal abnormalities in the Apert
syndrome." Am J Med Genet 47(5): 624-32.
Cohen, M. M., Jr. and S. Kreiborg (1993). "Visceral anomalies in the Apert syndrome."
Am J Med Genet 45(6): 758-60.
Cotsarelis, G., T. T. Sun, et al. (1990). "Label-retaining cells reside in the bulge area of
pilosebaceous unit: implications for follicular stem cells, hair cycle, and skin
carcinogenesis." Cell 61(7): 1329-37.
DasGupta, R. and E. Fuchs (1999). "Multiple roles for activated LEF/TCF transcription
complexes during hair follicle development and differentiation." Development 126(20):
4557-68.
Davenport, T. G., L. A. Jerome-Majewska, et al. (2003). "Mammary gland, limb and yolk
sac defects in mice lacking Tbx3, the gene mutated in human ulnar mammary syndrome."
Development 130(10): 2263-73.
De Langhe, S. P., Sala, F. G., Del Moral, P. M., Fairbanks, T. J., Yamada, K. M.,
Warburton, D., Burns, R. C., and Bellusci, S. (2005). "Dickkopf-1 (DKK1) reveals that
fibronectin is a major target of Wnt signaling in branching morphogenesis of the mouse
embryonic lung." Dev Biol 277, 316-31.
de Maximy, A. A., Y. Nakatake, et al. (1999). "Cloning and expression pattern of a
mouse homologue of drosophila sprouty in the mouse embryo." Mech Dev 81(1-2): 213-
6.
De Moerlooze, L., B. Spencer-Dene, et al. (2000). "An important role for the IIIb isoform
of fibroblast growth factor receptor 2 (FGFR2) in mesenchymal-epithelial signalling
during mouse organogenesis." Development 127(3): 483-92.
Deome, K. B., L. J. Faulkin, Jr., et al. (1959). "Development of mammary tumors from
hyperplastic alveolar nodules transplanted into gland-free mammary fat pads of female
C3H mice." Cancer Res 19(5): 515-20.
185
Dikic, I. and S. Giordano (2003). "Negative receptor signalling." Curr Opin Cell Biol
15(2): 128-35.
Dillon, C., B. Spencer-Dene, et al. (2004). "A crucial role for fibroblast growth factor
signaling in embryonic mammary gland development." J Mammary Gland Biol
Neoplasia 9(2): 207-15.
Doherty, P. and F. S. Walsh (1996). "CAM-FGF Receptor Interactions: A Model for
Axonal Growth." Mol Cell Neurosci 8(2/3): 99-111.
Domingos, P. M., N. Itasaki, et al. (2001). "The Wnt/beta-catenin pathway posteriorizes
neural tissue in Xenopus by an indirect mechanism requiring FGF signalling." Dev Biol
239(1): 148-60.
Doniach, T. (1995). "Basic FGF as an inducer of anteroposterior neural pattern." Cell
83(7): 1067-70.
Dunbar, M. E. and J. J. Wysolmerski (1999). "Parathyroid hormone-related protein: a
developmental regulatory molecule necessary for mammary gland development." J
Mammary Gland Biol Neoplasia 4(1): 21-34.
Eblaghie, M. C., J. S. Lunn, et al. (2003). "Negative feedback regulation of FGF
signaling levels by Pyst1/MKP3 in chick embryos." Curr Biol 13(12): 1009-18.
Entesarian, M., H. Matsson, et al. (2005). "Mutations in the gene encoding fibroblast
growth factor 10 are associated with aplasia of lacrimal and salivary glands." Nat Genet
37(2): 125-7.
Esch, F., A. Baird, et al. (1985). "Primary structure of bovine pituitary basic fibroblast
growth factor (FGF) and comparison with the amino-terminal sequence of bovine brain
acidic FGF." Proc Natl Acad Sci U S A 82(19): 6507-11.
Esch, F., N. Ueno, et al. (1985). "Primary structure of bovine brain acidic fibroblast
growth factor (FGF)." Biochem Biophys Res Commun 133(2): 554-62.
Eswarakumar, V. P., I. Lax, et al. (2005). "Cellular signaling by fibroblast growth factor
receptors." Cytokine Growth Factor Rev 16(2): 139-49.
Fallon, J. F. and R. O. Kelley (1977). "Ultrastruct analysis of the apical ectodermal ridge
duri;g vertebrate limb morphogenesis. II. Gap junctions as distinctive ridge structures
common to birds and mammals." J Embryol Exp Morphol 41: 223-32.
Fallon, J. F., A. Lopez, et al. (1994). "FGF-2: apical ectodermal ridge growth signal for
chick limb development." Science 264(5155): 104-7.
186
Faraldo, M. M., Deugnier, M. A., Lukashev, M., Thiery, J. P., and Glukhova, M. A.
(1998). "Perturbation of beta1-integrin function alters the development of murine
mammary gland." Embo J 17, 2139-47.
Feldman, B., W. Poueymirou, et al. (1995). "Requirement of FGF-4 for postimplantation
mouse development." Science 267(5195): 246-9.
Feng, Y., Manka, D., Wagner, K.U. and Khan, S.A. (2007) "Estrogen receptor-alpha
expression in the mammary epithelium is required for ductal and alveolar morphogenesis
in mice." Proc Natl Acad Sci USA 104(37):14718-23.
Finch, P. W., J. S. Rubin, et al. (1989). "Human KGF is FGF-related with properties of a
paracrine effector of epithelial cell growth." Science 245(4919): 752-5.
Fisher, M. E., H. V. Isaacs, et al. (2002). "eFGF is required for activation of XmyoD
expression in the myogenic cell lineage of Xenopus laevis." Development 129(6): 1307-
15.
Flaumenhaft, R., D. Moscatelli, et al. (1990). "Heparin and heparan sulfate increase the
radius of diffusion and action of basic fibroblast growth factor." J Cell Biol 111(4): 1651-
9.
Foley, J., P. Dann, et al. (2001). "Parathyroid hormone-related protein maintains
mammary epithelial fate and triggers nipple skin differentiation during embryonic breast
development." Development 128(4): 513-25.
Fortini, M. E. and S. Artavanis-Tsakonas (1993). "Notch: neurogenesis is only part of the
picture." Cell 75(7): 1245-7.
Fortini, M. E., I. Rebay, et al. (1993). "An activated Notch receptor blocks cell-fate
commitment in the developing Drosophila eye." Nature 365(6446): 555-7.
Francannet, C., P. Vanlieferinghen, et al. (1994). "LADD syndrome in five members of a
three-generation family and prenatal diagnosis." Genet Couns 5(1): 85-91.
Furthauer, M., W. Lin, et al. (2002). "Sef is a feedback-induced antagonist of
Ras/MAPK-mediated FGF signalling." Nat Cell Biol 4(2): 170-4.
Garcia-Closas, M., P. Hall, et al. (2008). "Heterogeneity of breast cancer associations
with five susceptibility loci by clinical and pathological characteristics." PLoS Genet
4(4): e1000054.
Geduspan, J. S. and M. Solursh (1992). "Cellular contribution of the different regions of
the somatopleure to the developing limb." Dev Dyn 195(3): 177-87.
187
Gisselbrecht, S., J. B. Skeath, et al. (1996). "heartless encodes a fibroblast growth factor
receptor (DFR1/DFGF-R2) involved in the directional migration of early mesodermal
cells in the Drosophila embryo." Genes Dev 10(23): 3003-17.
Goldfarb, M. (2005). "Fibroblast growth factor homologous factors: evolution, structure,
and function." Cytokine Growth Factor Rev 16(2): 215-20.
Goldin, S. N. and V. E. Papaioannou (2003). "Paracrine action of FGF4 during
periimplantation development maintains trophectoderm and primitive endoderm."
Genesis 36(1): 40-7.
Gospodarowicz, D. and H. Bialecki (1978). "The effects of the epidermal and fibroblast
growth factors on the replicative lifespan of cultured bovine granulosa cells."
Endocrinology 103(3): 854-65.
Gospodarowicz, D., H. Bialecki, et al. (1978). "Purification of the fibroblast growth
factor activity from bovine brain." J Biol Chem 253(10): 3736-43.
Gospodarowicz, D., K. D. Brown, et al. (1978). "Control of proliferation of human
vascular endothelial cells. Characterization of the response of human umbilical vein
endothelial cells to fibroblast growth factor, epidermal growth factor, and thrombin." J
Cell Biol 77(3): 774-88.
Gospodarowicz, D. and J. Cheng (1986). "Heparin protects basic and acidic FGF from
inactivation." J Cell Physiol 128(3): 475-84.
Gospodarowicz, D., G. Greenburg, et al. (1978). "Factors involved in the modulation of
cell proliferation in vivo and in vitro: the role of fibroblast and epidermal growth factors
in the proliferative response of mammalian cells." In Vitro 14(1): 85-118.
Gospodarowicz, D., A. L. Mescher, et al. (1978). "Control of cellular proliferation by the
fibroblast and epidermal growth factors." Natl Cancer Inst Monogr(48): 109-30.
Gospodarowicz, D. and J. Moran (1974). "Effect of a fibroblast growth factor, insulin,
dexamethasone, and serum on the morphology of BALB/c 3T3 cells." Proc Natl Acad Sci
U S A 71(12): 4648-52.
Gossen, M. and Bujard, H. (1992). "Tight control of gene expression in mammalian cells
by tetracycline-responsive promoters." Proc Natl Acad Sci U S A 89, 5547-51.
Grose, R., and Dickson, C. (2005). "Fibroblast growth factor signaling in tumorigenesis."
Cytokine Growth Factor Rev 16, 179-86.
Gryzik, T. and H. A. Muller (2004). "FGF8-like1 and FGF8-like2 encode putative
ligands of the FGF receptor Htl and are required for mesoderm migration in the
Drosophila gastrula." Curr Biol 14(8): 659-67.
188
Guo, Q. and J. Y. Li (2007). "Distinct functions of the major Fgf8 spliceform, Fgf8b,
before and during mouse gastrulation." Development 134(12): 2251-60.
Hacohen, N., S. Kramer, et al. (1998). "sprouty encodes a novel antagonist of FGF
signaling that patterns apical branching of the Drosophila airways." Cell 92(2): 253-63.
Hadari, Y. R., H. Kouhara, et al. (1998). "Binding of Shp2 tyrosine phosphatase to FRS2
is essential for fibroblast growth factor-induced PC12 cell differentiation." Mol Cell Biol
18(7): 3966-73.
Hajihosseini, M. K., R. Duarte, et al. (2009). "Evidence that Fgf10 contributes to the
skeletal and visceral defects of an Apert syndrome mouse model." Dev Dyn 238(2): 376-
85.
Hall, H., E. J. Williams, et al. (1996). "Inhibition of FGF-stimulated phosphatidylinositol
hydrolysis and neurite outgrowth by a cell-membrane permeable phosphopeptide." Curr
Biol 6(5): 580-7.
Harada, H., P. Kettunen, et al. (1999). "Localization of putative stem cells in dental
epithelium and their association with Notch and FGF signaling." J Cell Biol 147(1): 105-
20.
Harada, H., T. Toyono, et al. (2002). "FGF10 maintains stem cell compartment in
developing mouse incisors." Development 129(6): 1533-41.
Hardcastle, Z., R. Mo, et al. (1998). "The Shh signalling pathway in tooth development:
defects in Gli2 and Gli3 mutants." Development 125(15): 2803-11.
Haworth, K., F. Smith, et al. (2007). "Expression of the Scube3 epidermal growth factor-
related gene during early embryonic development in the mouse." Gene Expr Patterns
7(5): 630-4.
Hemmati-Brivanlou, A. and D. Melton (1997). "Vertebrate embryonic cells will become
nerve cells unless told otherwise." Cell 88(1): 13-7.
Hemmati-Brivanlou, A. and D. Melton (1997). "Vertebrate neural induction." Annu Rev
Neurosci 20: 43-60.
Hens, J. R. and J. J. Wysolmerski (2005). "Key stages of mammary gland development:
molecular mechanisms involved in the formation of the embryonic mammary gland."
Breast Cancer Res 7(5): 220-4.
Hokuto, I., A. K. Perl, et al. (2003). "Prenatal, but not postnatal, inhibition of fibroblast
growth factor receptor signaling causes emphysema." J Biol Chem 278(1): 415-21.
Hollister, D. W., S. H. Klein, et al. (1973). "The lacrimo-auriculo-dento-digital
syndrome." J Pediatr 83(3): 438-44.
189
Holowacz, T. and S. Sokol (1999). "FGF is required for posterior neural patterning but
not for neural induction." Dev Biol 205(2): 296-308.
Horb, M. E. (2000). "Patterning the endoderm: the importance of neighbours." Bioessays
22(7): 599-602.
Horton, A. C., N. R. Mahadevan, et al. (2003). "Phylogenetic analyses alone are
insufficient to determine whether genome duplication(s) occurred during early vertebrate
evolution." J Exp Zool B Mol Dev Evol 299(1): 41-53.
Humphreys, R. C., M. Krajewska, et al. (1996). "Apoptosis in the terminal endbud of the
murine mammary gland: a mechanism of ductal morphogenesis." Development 122(12):
4013-22.
Hunter, D. J., Kraft, P., Jacobs, K. B., Cox, D. G., Yeager, M., Hankinson, S. E.,
Wacholder, S., Wang, Z., Welch, R., Hutchinson, A., Wang, J., Yu, K., Chatterjee, N.,
Orr, N., Willett, W. C., Colditz, G. A., Ziegler, R. G., Berg, C. D., Buys, S. S., McCarty,
C. A., Feigelson, H. S., Calle, E. E., Thun, M. J., Hayes, R. B., Tucker, M., Gerhard, D.
S., Fraumeni, J. F., Jr., Hoover, R. N., Thomas, G., and Chanock, S. J. (2007). "A
genome-wide association study identifies alleles in FGFR2 associated with risk of
sporadic postmenopausal breast cancer". Nat Genet.
Hunter, D. J., P. Kraft, et al. (2007). "A genome-wide association study identifies alleles
in FGFR2 associated with risk of sporadic postmenopausal breast cancer." Nat Genet
39(7): 870-4.
Itoh, K. and S. Y. Sokol (1994). "Heparan sulfate proteoglycans are required for
mesoderm formation in Xenopus embryos." Development 120(9): 2703-11.
Itoh, N. and D. M. Ornitz (2004). "Evolution of the Fgf and Fgfr gene families." Trends
Genet 20(11): 563-9.
Itoh, N. and D. M. Ornitz (2008). "Functional evolutionary history of the mouse Fgf gene
family." Dev Dyn 237(1): 18-27.
Jarecki, J., E. Johnson, et al. (1999). "Oxygen regulation of airway branching in
Drosophila is mediated by branchless FGF." Cell 99(2): 211-20.
Jernvall, J., P. Kettunen, et al. (1994). "Evidence for the role of the enamel knot as a
control center in mammalian tooth cusp formation: non-dividing cells express growth
stimulating Fgf-4 gene." Int J Dev Biol 38(3): 463-9.
Johnston, C. L., H. C. Cox, et al. (1995). "bFGF and aFGF induce membrane ruffling in
breast cancer cells but not in normal breast epithelial cells: FGFR-4 involvement."
Biochem J 306 ( Pt 2): 609-16.
190
Kassai, Y., P. Munne, et al. (2005). "Regulation of mammalian tooth cusp patterning by
ectodin." Science 309(5743): 2067-70.
Kawakami, Y., J. Capdevila, et al. (2001). "WNT signals control FGF-dependent limb
initiation and AER induction in the chick embryo." Cell 104(6): 891-900.
Kawakami, Y., J. Rodriguez-Leon, et al. (2003). "MKP3 mediates the cellular response to
FGF8 signalling in the vertebrate limb." Nat Cell Biol 5(6): 513-9.
Kazanskaya, O., A. Glinka, et al. (2000). "The role of Xenopus dickkopf1 in prechordal
plate specification and neural patterning." Development 127(22): 4981-92.
Kelly, R., Alonso, S., Tajbakhsh, S., Cossu, G., and Buckingham, M. (1995). "Myosin
light chain 3F regulatory sequences confer regionalized cardiac and skeletal muscle
expression in transgenic mice." J Cell Biol 129, 383-96.
Kelly, R. G., Brown, N. A., and Buckingham, M. E. (2001). "The arterial pole of the
mouse heart forms from Fgf10-expressing cells in pharyngeal mesoderm." Dev Cell 1,
435-40.
Kengaku, M. and H. Okamoto (1995). "bFGF as a possible morphogen for the
anteroposterior axis of the central nervous system in Xenopus." Development 121(9):
3121-30.
Kengaku, M., J. Capdevila, et al. (1998). "Distinct WNT pathways regulating AER
formation and dorsoventral polarity in the chick limb bud." Science 280(5367): 1274-7.
Kerley, M.A. (1975) "Pre-natal development of the mouse incisor." Proc. Okla. Acad.
Sci. 55, 6-10.
Kimelman, D. and K. J. Griffin (2000). "Vertebrate mesendoderm induction and
patterning." Curr Opin Genet Dev 10(4): 350-6.
Klambt, C., L. Glazer, et al. (1992). "breathless, a Drosophila FGF receptor homolog, is
essential for migration of tracheal and specific midline glial cells." Genes Dev 6(9):
1668-78.
Klein, O. D., D. B. Lyons, et al. (2008). "An FGF signaling loop sustains the generation
of differentiated progeny from stem cells in mouse incisors." Development 135(2): 377-
85.
Kolm, P. J. and H. L. Sive (1995). "Regulation of the Xenopus labial homeodomain
genes, HoxA1 and HoxD1: activation by retinoids and peptide growth factors." Dev Biol
167(1): 34-49.
Konishi, M., T. Asaki, et al. (2006). "Role of Fgf10 in cell proliferation in white adipose
tissue." Mol Cell Endocrinol 249(1-2): 71-7.
191
Korach, K.S., Couse, J.F., Curtis, S.W., Washburn, T.F., Lindzey, J., Kimbro, K.S.,
Eddy, E.M., Migliaccio, S., Snedeker, S.M., Lubahn, D.B., Schomberg D.W. and Smith,
E.P. (1996) "Estrogen receptor gene disruption: molecular characterization and
experimental and clinical phenotypes." Recent Prog Horm Res 51, 159–188
Kouhara, H., Y. R. Hadari, et al. (1997). "A lipid-anchored Grb2-binding protein that
links FGF-receptor activation to the Ras/MAPK signaling pathway." Cell 89(5): 693-702.
Kovalenko, D., X. Yang, et al. (2003). "Sef inhibits fibroblast growth factor signaling by
inhibiting FGFR1 tyrosine phosphorylation and subsequent ERK activation." J Biol
Chem 278(16): 14087-91.
Krejci, P., J. Prochazkova, et al. (2009). "Molecular pathology of the fibroblast growth
factor family." Hum Mutat 30(9): 1245-55.
Kudoh, T., S. W. Wilson, et al. (2002). "Distinct roles for Fgf, Wnt and retinoic acid in
posteriorizing the neural ectoderm." Development 129(18): 4335-46.
Kunath, T., M. K. Saba-El-Leil, et al. (2007). "FGF stimulation of the Erk1/2 signalling
cascade triggers transition of pluripotent embryonic stem cells from self-renewal to
lineage commitment." Development 134(16): 2895-902.
Lamb, T. M. and R. M. Harland (1995). "Fibroblast growth factor is a direct neural
inducer, which combined with noggin generates anterior-posterior neural pattern."
Development 121(11): 3627-36.
Laurikkala, J., Y. Kassai, et al. (2003). "Identification of a secreted BMP antagonist,
ectodin, integrating BMP, FGF, and SHH signals from the tooth enamel knot." Dev Biol
264(1): 91-105.
Laurikkala, J., M. Mikkola, et al. (2001). "TNF signaling via the ligand-receptor pair
ectodysplasin and edar controls the function of epithelial signaling centers and is
regulated by Wnt and activin during tooth organogenesis." Dev Biol 229(2): 443-55.
Lee, J. T., Jr. and J. A. McCubrey (2002). "The Raf/MEK/ERK signal transduction
cascade as a target for chemotherapeutic intervention in leukemia." Leukemia 16(4): 486-
507.
Lew, E. D., J. H. Bae, et al. (2007). "Structural basis for reduced FGFR2 activity in
LADD syndrome: Implications for FGFR autoinhibition and activation." Proc Natl Acad
Sci U S A 104(50): 19802-7.
Lewandoski, M., X. Sun, et al. (2000). "Fgf8 signalling from the AER is essential for
normal limb development." Nat Genet 26(4): 460-3.
Li, Y. and J. M. Rosen (2005). "Stem/progenitor cells in mouse mammary gland
development and breast cancer." J Mammary Gland Biol Neoplasia 10(1): 17-24.
192
Lim, J., P. Yusoff, et al. (2002). "The cysteine-rich sprouty translocation domain targets
mitogen-activated protein kinase inhibitory proteins to phosphatidylinositol 4,5-
bisphosphate in plasma membranes." Mol Cell Biol 22(22): 7953-66.
Lin, Y., Liu, G., Zhang, Y., Hu, Y.P., Yu, K., Lin, C., McKeehan, K., Xuan, J.W., Ornitz,
D.M., Shen, M.M., Greenberg, N., McKeehan, W.L. and Wang, F. (2007) "Fibroblast
growth factor receptor 2 tyrosine kinase is required for prostatic morphogenesis and the
acquisition of strict androgen dependency for adult tissue homeostasis." Development
134, 723-34
Lin, W., M. Furthauer, et al. (2002). "Cloning of the mouse Sef gene and comparative
analysis of its expression with Fgf8 and Spry2 during embryogenesis." Mech Dev 113(2):
163-8.
Lin, Y., Cheng, Y.S., Qin, C., Lin, C., D'Souza, R. and Wang, F. (2009) "FGFR2 in the
dental epithelium is essential for development and maintenance of the maxillary cervical
loop, a stem cell niche in mouse incisors." Dev Dyn. 238, 324-30.
Liu, F., E. Y. Chu, et al. (2008). "Wnt/beta-catenin signaling directs multiple stages of
tooth morphogenesis." Dev Biol 313(1): 210-24.
Liu, H., J. Guo, et al. (2008). "Fabrication and functionalization of dendritic
poly(amidoamine)-immobilized magnetic polymer composite microspheres." J Phys
Chem B 112(11): 3315-21.
Lu, M. F., C. Pressman, et al. (1999). "Function of Rieger syndrome gene in left-right
asymmetry and craniofacial development." Nature 401(6750): 276-8.
Lu, P., Y. Yu, et al. (2008). "The apical ectodermal ridge is a timer for generating distal
limb progenitors." Development 135(8): 1395-405.
Lumsden, A. G. (1988). "Spatial organization of the epithelium and the role of neural
crest cells in the initiation of the mammalian tooth germ." Development 103 Suppl: 155-
69.
Luukko, K., S. Loes, et al. (2003). "Identification of a novel putative signaling center, the
tertiary enamel knot in the postnatal mouse molar tooth." Mech Dev 120(3): 270-6.
Maciag, T., T. Mehlman, et al. (1984). "Heparin binds endothelial cell growth factor, the
principal endothelial cell mitogen in bovine brain." Science 225(4665): 932-5.
MacNicol, A. M., A. J. Muslin, et al. (1993). "Raf-1 kinase is essential for early Xenopus
development and mediates the induction of mesoderm by FGF." Cell 73(3): 571-83.

193
Mailleux, A. A., Kelly, R., Veltmaat, J. M., De Langhe, S. P., Zaffran, S., Thiery, J. P.,
and Bellusci, S. (2005). "Fgf10 expression identifies parabronchial smooth muscle cell
progenitors and is required for their entry into the smooth muscle cell lineage."
Development 132, 2157-66.
Mailleux, A. A., Spencer-Dene, B., Dillon, C., Ndiaye, D., Savona-Baron, C., Itoh, N.,
Kato, S., Dickson, C., Thiery, J. P., and Bellusci, S. (2002). "Role of FGF10/FGFR2b
signaling during mammary gland development in the mouse embryo." Development 129,
53-60.
Mason, J.M., Morrison, D.J., Basson, M.A. and Licht, J.D. (2006) "Sprouty proteins:
multifaceted negative-feedback regulators of receptor tyrosine kinase signaling." Trends
Cell Biol. 16, 45-54.
McGrew, L. L., S. Hoppler, et al. (1997). "Wnt and FGF pathways cooperatively pattern
anteroposterior neural ectoderm in Xenopus." Mech Dev 69(1-2): 105-14.
Medina, D. (1996). "The mammary gland: a unique organ for the study of development
and tumorigenesis." J Mammary Gland Biol Neoplasia 1(1): 5-19.
Mercader, N., E. Leonardo, et al. (2000). "Opposing RA and FGF signals control
proximodistal vertebrate limb development through regulation of Meis genes."
Development 127(18): 3961-70.
Miletich, I., M. T. Cobourne, et al. (2005). "Expression of the Hedgehog antagonists
Rab23 and Slimb/betaTrCP during mouse tooth development." Arch Oral Biol 50(2):
147-51.
Min, H., D. M. Danilenko, et al. (1998). "Fgf-10 is required for both limb and lung
development and exhibits striking functional similarity to Drosophila branchless." Genes
Dev 12(20): 3156-61.
Mina, M. and E. J. Kollar (1987). "The induction of odontogenesis in non-dental
mesenchyme combined with early murine mandibular arch epithelium." Arch Oral Biol
32(2): 123-7.
Minowada, G., L. A. Jarvis, et al. (1999). "Vertebrate Sprouty genes are induced by FGF
signaling and can cause chondrodysplasia when overexpressed." Development 126(20):
4465-75.
Mistry, N., W. Harrington, et al. (2003). "Of urchins and men: evolution of an alternative
splicing unit in fibroblast growth factor receptor genes." RNA 9(2): 209-17.
Mitsiadis, T. A., I. Angeli, et al. (2003). "Role of Islet1 in the patterning of murine
dentition." Development 130(18): 4451-60.
194
Mitsiadis, T. A. and J. Drouin (2008). "Deletion of the Pitx1 genomic locus affects
mandibular tooth morphogenesis and expression of the Barx1 and Tbx1 genes." Dev Biol
313(2): 887-96.
Miyamoto, M., K. Naruo, et al. (1993). "Molecular cloning of a novel cytokine cDNA
encoding the ninth member of the fibroblast growth factor family, which has a unique
secretion property." Mol Cell Biol 13(7): 4251-9.
Moffa, A. B., and Ethier, S. P. (2007). "Differential signal transduction of alternatively
spliced FGFR2 variants expressed in human mammary epithelial cells." J Cell Physiol
210, 720-31.
Mohammadi, M., C. A. Dionne, et al. (1992). "Point mutation in FGF receptor eliminates
phosphatidylinositol hydrolysis without affecting mitogenesis." Nature 358(6388): 681-4.
Moon, A. M., A. M. Boulet, et al. (2000). "Normal limb development in conditional
mutants of Fgf4." Development 127(5): 989-96.
Morris, G. and T. A. Slotkin (1985). "Beta-2 adrenergic control of ornithine
decarboxylase activity in brain regions of the developing rat." J Pharmacol Exp Ther
233(1): 141-7.
Morrison, B. J., C. W. Schmidt, et al. (2008). "Breast cancer stem cells: implications for
therapy of breast cancer." Breast Cancer Res 10(4): 210.
Moscatelli, D. (1987). "High and low affinity binding sites for basic fibroblast growth
factor on cultured cells: absence of a role for low affinity binding in the stimulation of
plasminogen activator production by bovine capillary endothelial cells." J Cell Physiol
131(1): 123-30.
Mucchielli, M. L., T. A. Mitsiadis, et al. (1997). "Mouse Otlx2/RIEG expression in the
odontogenic epithelium precedes tooth initiation and requires mesenchyme-derived
signals for its maintenance." Dev Biol 189(2): 275-84.
Muneoka, K. and D. Sassoon (1992). "Molecular aspects of regeneration in developing
vertebrate limbs." Dev Biol 152(1): 37-49.
Mustonen, T., J. Pispa, et al. (2003). "Stimulation of ectodermal organ development by
Ectodysplasin-A1." Dev Biol 259(1): 123-36.
Neilson, K. M. and R. Friesel (1996). "Ligand-independent activation of fibroblast
growth factor receptors by point mutations in the extracellular, transmembrane, and
kinase domains." J Biol Chem 271(40): 25049-57.

195
Ng, F., S. Boucher, et al. (2008). "PDGF, TGF-beta, and FGF signaling is important for
differentiation and growth of mesenchymal stem cells (MSCs): transcriptional profiling
can identify markers and signaling pathways important in differentiation of MSCs into
adipogenic, chondrogenic, and osteogenic lineages." Blood 112(2): 295-307.
Niehrs, C. and H. Meinhardt (2002). "Modular feedback." Nature 417(6884): 35-6.
Niswander, L., C. Tickle, et al. (1993). "FGF-4 replaces the apical ectodermal ridge and
directs outgrowth and patterning of the limb." Cell 75(3): 579-87.
Nutt, S. L., K. S. Dingwell, et al. (2001). "Xenopus Sprouty2 inhibits FGF-mediated
gastrulation movements but does not affect mesoderm induction and patterning." Genes
Dev 15(9): 1152-66.
Ohazama, A., J. M. Courtney, et al. (2004). "Traf6 is essential for murine tooth cusp
morphogenesis." Dev Dyn 229(1): 131-5.
Ohuchi, H., T. Nakagawa, et al. (1997). "The mesenchymal factor, FGF10, initiates and
maintains the outgrowth of the chick limb bud through interaction with FGF8, an apical
ectodermal factor." Development 124(11): 2235-44.
Ohuchi, H., Y. Hori, et al. (2000). "FGF10 acts as a major ligand for FGF receptor 2 IIIb
in mouse multi-organ development." Biochem Biophys Res Commun 277(3): 643-9.
Ornitz, D. M. (2000). "FGFs, heparan sulfate and FGFRs: complex interactions essential
for development." Bioessays 22(2): 108-12.
Ornitz, D. M. and N. Itoh (2001). "Fibroblast growth factors." Genome Biol 2(3):
REVIEWS3005.
Ornitz, D. M., J. Xu, et al. (1996). "Receptor specificity of the fibroblast growth factor
family." J Biol Chem 271(25): 15292-7.
Park, W. Y., B. Miranda, et al. (1998). "FGF-10 is a chemotactic factor for distal
epithelial buds during lung development." Dev Biol 201(2): 125-34.
Parsa, S., S. K. Ramasamy, et al. (2008). "Terminal end bud maintenance in mammary
gland is dependent upon FGFR2b signaling." Dev Biol 317(1): 121-31.
Pedchenko, V. K. and W. Imagawa (2000). "Pattern of expression of the KGF receptor
and its ligands KGF and FGF-10 during postnatal mouse mammary gland development."
Mol Reprod Dev 56(4): 441-7.
Partanen, J., L. Schwartz, et al. (1998). "Opposite phenotypes of hypomorphic and Y766
phosphorylation site mutations reveal a function for Fgfr1 in anteroposterior patterning of
mouse embryos." Genes Dev 12(15): 2332-44.
196
Pedchenko, V. K., and Imagawa, W. (2000). "Pattern of expression of the KGF receptor
and its ligands KGF and FGF-10 during postnatal mouse mammary gland development."
Mol Reprod Dev 56, 441-7.
Peters, K. G., J. Marie, et al. (1992). "Point mutation of an FGF receptor abolishes
phosphatidylinositol turnover and Ca2+ flux but not mitogenesis." Nature 358(6388):
678-81.
Pownall, M. E., A. S. Tucker, et al. (1996). "eFGF, Xcad3 and Hox genes form a
molecular pathway that establishes the anteroposterior axis in Xenopus." Development
122(12): 3881-92.
Ramasamy, S. K., A. A. Mailleux, et al. (2007). "Fgf10 dosage is critical for the
amplification of epithelial cell progenitors and for the formation of multiple
mesenchymal lineages during lung development." Dev Biol 307(2): 237-47.
Rao, M. S. (1999). "Multipotent and restricted precursors in the central nervous system."
Anat Rec 257(4): 137-48.
Reichman-Fried, M., B. Dickson, et al. (1994). "Elucidation of the role of breathless, a
Drosophila FGF receptor homolog, in tracheal cell migration." Genes Dev 8(4): 428-39.
Reginelli, A. D., Y. Q. Wang, et al. (1995). "Digit tip regeneration correlates with regions
of Msx1 (Hox 7) expression in fetal and newborn mice." Development 121(4): 1065-76.
Repesh, L. A. and J. C. Oberpriller (1978). "Scanning electron microscopy of epidermal
cell migration in wound healing during limb regeneration in the adult newt,
Notophthalmus viridescens." Am J Anat 151(4): 539-55.Reynolds, B. A. and S. Weiss
(1992). "Generation of neurons and astrocytes from isolated cells of the adult mammalian
central nervous system." Science 255(5052): 1707-10.
Ribisi, S., Jr., F. V. Mariani, et al. (2000). "Ras-mediated FGF signaling is required for
the formation of posterior but not anterior neural tissue in Xenopus laevis." Dev Biol
227(1): 183-96.
Riddle, R. D., R. L. Johnson, et al. (1993). "Sonic hedgehog mediates the polarizing
activity of the ZPA." Cell 75(7): 1401-16.
Rohmann, E., H. G. Brunner, et al. (2006). "Mutations in different components of FGF
signaling in LADD syndrome." Nat Genet 38(4): 414-7.
Rossant, J. (2008). "Stem cells and early lineage development." Cell 132(4): 527-31.
Ruch, J. V., H. Lesot, et al. (1982). "Facts and hypotheses concerning the control of
odontoblast differentiation." Differentiation 21(1): 7-12.
197
Saffell, J. L., E. J. Williams, et al. (1997). "Expression of a dominant negative FGF
receptor inhibits axonal growth and FGF receptor phosphorylation stimulated by CAMs."
Neuron 18(2): 231-42.
Sakamoto, H., T. Yoshida, et al. (1988). "Cloned hst gene from normal human leukocyte
DNA transforms NIH3T3 cells." Biochem Biophys Res Commun 151(3): 965-72.
Sakaue, H., Konishi, M., Ogawa, W., Asaki, T., Mori, T., Yamasaki, M., Takata, M.,
Ueno, H., Kato, S., Kasuga, M., and Itoh, N. (2002). "Requirement of fibroblast growth
factor 10 in development of white adipose tissue." Genes Dev 16, 908-12.
Sala, F. G., Curtis, J. L., Veltmaat, J. M., Del Moral, P. M., Le, L. T., Fairbanks, T. J.,
Warburton, D., Ford, H., Wang, K., Burns, R. C., and Bellusci, S. (2006). "Fibroblast
growth factor 10 is required for survival and proliferation but not differentiation of
intestinal epithelial progenitor cells during murine colon development." Dev Biol 299,
373-85.
Sarkar, L. and P. T. Sharpe (1999). "Expression of Wnt signalling pathway genes during
tooth development." Mech Dev 85(1-2): 197-200.
Satokata, I. and R. Maas (1994). "Msx1 deficient mice exhibit cleft palate and
abnormalities of craniofacial and tooth development." Nat Genet 6(4): 348-56.
Satou, Y., K. S. Imai, et al. (2002). "Fgf genes in the basal chordate Ciona intestinalis."
Dev Genes Evol 212(9): 432-8.
Saunders, J. W., Jr. (1948). "The proximo-distal sequence of origin of the parts of the
chick wing and the role of the ectoderm." J Exp Zool 108(3): 363-403.
Schlake, T. (2005). "FGF signals specifically regulate the structure of hair shaft medulla
via IGF-binding protein 5". Development 132, 2981-90.
Schlessinger, J. and Lax, I. (2007) "Lacrimo-auriculo-dento-digital syndrome is caused
by reduced activity of the fibroblast growth factor 10 (FGF10)-FGF receptor 2 signaling
pathway." Mol. Cell Biol. 27, 6903-12.
Schwenk, F., U. Baron, et al. (1995). "A cre-transgenic mouse strain for the ubiquitous
deletion of loxP-flanked gene segments including deletion in germ cells." Nucleic Acids
Res 23(24): 5080-1.
Schofield, R. (1978). "The relationship between the spleen colony-forming cell and the
haemopoietic stem cell." Blood Cells 4(1-2): 7-25.
Sekine, K., H. Ohuchi, et al. (1999). "Fgf10 is essential for limb and lung formation." Nat
Genet 21(1): 138-41.
198
Shackleton, M., F. Vaillant, et al. (2006). "Generation of a functional mammary gland
from a single stem cell." Nature 439(7072): 84-8.
Shams, I., E. Rohmann, et al. (2007). "Lacrimo-auriculo-dento-digital syndrome is
caused by reduced activity of the fibroblast growth factor 10 (FGF10)-FGF receptor 2
signaling pathway." Mol Cell Biol 27(19): 6903-12.
Sharpe, P. T. (1995). "Homeobox genes and orofacial development." Connect Tissue Res
32(1-4): 17-25.
Shimada, A., M. Yabusaki, et al. (2008). "Maternal-zygotic medaka mutants for fgfr1
reveal its essential role in the migration of the axial mesoderm but not the lateral
mesoderm." Development 135(2): 281-90.
Shishido, E., S. Higashijima, et al. (1993). "Two FGF-receptor homologues of
Drosophila: one is expressed in mesodermal primordium in early embryos." Development
117(2): 751-61.
Slack, J. M., B. G. Darlington, et al. (1987). "Mesoderm induction in early Xenopus
embryos by heparin-binding growth factors." Nature 326(6109): 197-200.
Smith, G. H. (2005). "Label-retaining epithelial cells in mouse mammary gland divide
asymmetrically and retain their template DNA strands." Development 132(4): 681-7.
Spangrude, G. J. and G. R. Johnson (1990). "Resting and activated subsets of mouse
multipotent hematopoietic stem cells." Proc Natl Acad Sci U S A 87(19): 7433-7.
Standley, H. J., A. M. Zorn, et al. (2001). "eFGF and its mode of action in the community
effect during Xenopus myogenesis." Development 128(8): 1347-57.
Stathopoulos, A., B. Tam, et al. (2004). "pyramus and thisbe: FGF genes that pattern the
mesoderm of Drosophila embryos." Genes Dev 18(6): 687-99.
Stephens, T. D. and T. R. McNulty (1981). "Evidence for a metameric pattern in the
development of the chick humerus." J Embryol Exp Morphol 61: 191-205.
Sternlicht, M. D., Kouros-Mehr, H., Lu, P., and Werb, Z. (2006). "Hormonal and local
control of mammary branching morphogenesis." Differentiation 74, 365-81.
Stingl, J., P. Eirew, et al. (2006). "Purification and unique properties of mammary
epithelial stem cells." Nature 439(7079): 993-7.
Stingl, J., A. Raouf, et al. (2006). "Deciphering the mammary epithelial cell hierarchy."
Cell Cycle 5(14): 1519-22.
Stingl, J., J. T. Emerman, et al. (2005). "Enzymatic dissociation and culture of normal
human mammary tissue to detect progenitor activity." Methods Mol Biol 290: 249-63.
199
Storey, K. G., A. Goriely, et al. (1998). "Early posterior neural tissue is induced by FGF
in the chick embryo." Development 125(3): 473-84.
Strecker, T. R. and T. D. Stephens (1983). "Peripheral nerves do not play a trophic role in
limb skeletal morphogenesis." Teratology 27(2): 159-67.
Streit, A., A. J. Berliner, et al. (2000). "Initiation of neural induction by FGF signalling
before gastrulation." Nature 406(6791): 74-8.
Summerbell, D., J. H. Lewis, et al. (1973). "Positional information in chick limb
morphogenesis." Nature 244(5417): 492-6.
Sutherland, D., C. Samakovlis, et al. (1996). "branchless encodes a Drosophila FGF
homolog that controls tracheal cell migration and the pattern of branching." Cell 87(6):
1091-101.
Tabin, C. and L. Wolpert (2007). "Rethinking the proximodistal axis of the vertebrate
limb in the molecular era." Genes Dev 21(12): 1433-42.
te Welscher, P., M. Fernandez-Teran, et al. (2002). "Mutual genetic antagonism
involving GLI3 and dHAND prepatterns the vertebrate limb bud mesenchyme prior to
SHH signaling." Genes Dev 16(4): 421-6.
ten Berge, D., S. A. Brugmann, et al. (2008). "Wnt and FGF signals interact to coordinate
growth with cell fate specification during limb development." Development 135(19):
3247-57.
Theodorou, V., M. Boer, et al. (2004). "Fgf10 is an oncogene activated by MMTV
insertional mutagenesis in mouse mammary tumors and overexpressed in a subset of
human breast carcinomas." Oncogene 23(36): 6047-55.
Thesleff, I., A. Vaahtokari, et al. (1996). "Molecular mechanisms of cell and tissue
interactions during early tooth development." Anat Rec 245(2): 151-61.
Thomas, B. L., A. S. Tucker, et al. (1997). "Role of Dlx-1 and Dlx-2 genes in patterning
of the murine dentition." Development 124(23): 4811-8.
Thomas, D. P. (1984). "Heparin, low molecular weight heparin, and heparin analogues."
Br J Haematol 58(3): 385-90.
Tickle, C. (1981). "The number of polarizing region cells required to specify additional
digits in the developing chick wing." Nature 289(5795): 295-8.
Torii, S., M. Kusakabe, et al. (2004). "Sef is a spatial regulator for Ras/MAP kinase
signaling." Dev Cell 7(1): 33-44.
200
Tsang, M., R. Friesel, et al. (2002). "Identification of Sef, a novel modulator of FGF
signalling." Nat Cell Biol 4(2): 165-9.
Tucker, A. S., A. Al Khamis, et al. (1998). "Interactions between Bmp-4 and Msx-1 act
to restrict gene expression to odontogenic mesenchyme." Dev Dyn 212(4): 533-9.
Tucker, A. S., D. J. Headon, et al. (2000). "Edar/Eda interactions regulate enamel knot
formation in tooth morphogenesis." Development 127(21): 4691-700.
Tucker, A. S., K. L. Matthews, et al. (1998). "Transformation of tooth type induced by
inhibition of BMP signaling." Science 282(5391): 1136-8.
Tucker, A. S. and P. T. Sharpe (1999). "Molecular genetics of tooth morphogenesis and
patterning: the right shape in the right place." J Dent Res 78(4): 826-34.
Tummers, M. and I. Thesleff (2003). "Root or crown: a developmental choice
orchestrated by the differential regulation of the epithelial stem cell niche in the tooth of
two rodent species." Development 130(6): 1049-57.
Umbhauer, M., A. Penzo-Mendez, et al. (2000). "Signaling specificities of fibroblast
growth factor receptors in early Xenopus embryo." J Cell Sci 113 ( Pt 16): 2865-75.
Vaahtokari, A., T. Aberg, et al. (1996). "The enamel knot as a signaling center in the
developing mouse tooth." Mech Dev 54(1): 39-43.
Vaahtokari, A., T. Aberg, et al. (1996). "Apoptosis in the developing tooth: association
with an embryonic signaling center and suppression by EGF and FGF-4." Development
122(1): 121-9.
van Genderen, C., R. M. Okamura, et al. (1994). "Development of several organs that
require inductive epithelial-mesenchymal interactions is impaired in LEF-1-deficient
mice." Genes Dev 8(22): 2691-703.
Veltmaat, J. M., Mailleux, A. A., Thiery, J. P., and Bellusci, S. (2003). "Mouse
embryonic mammogenesis as a model for the molecular regulation of pattern formation."
Differentiation 71, 1-17.
Veltmaat, J. M., Relaix, F., Le, L. T., Kratochwil, K., Sala, F. G., van Veelen, W., Rice,
R., Spencer-Dene, B., Mailleux, A. A., Rice, D. P., Thiery, J. P., and Bellusci, S. (2006).
"Gli3-mediated somitic Fgf10 expression gradients are required for the induction and
patterning of mammary epithelium along the embryonic axes." Development 133, 2325-
35.
Veltmaat, J. M., Van Veelen, W., Thiery, J. P., and Bellusci, S. (2004). "Identification of
the mammary line in mouse by Wnt10b expression." Dev Dyn 229, 349-56.
201
Vincent, S., R. Wilson, et al. (1998). "The Drosophila protein Dof is specifically required
for FGF signaling." Mol Cell 2(4): 515-25.
Vivanco, I. and C. L. Sawyers (2002). "The phosphatidylinositol 3-Kinase AKT pathway
in human cancer." Nat Rev Cancer 2(7): 489-501.
Voog, J. and D. L. Jones "Stem cells and the niche: a dynamic duo." Cell Stem Cell 6(2):
103-15.
Wang, J. K., H. Xu, et al. (1996). "Broadly expressed SNT-like proteins link FGF
receptor stimulation to activators of Ras." Oncogene 13(4): 721-9.
Wang, X. P., M. Suomalainen, et al. (2007). "An integrated gene regulatory network
controls stem cell proliferation in teeth." PLoS Biol 5(6): e159.
Wang, X. P., M. Suomalainen, et al. (2004). "Follistatin regulates enamel patterning in
mouse incisors by asymmetrically inhibiting BMP signaling and ameloblast
differentiation." Dev Cell 7(5): 719-30.
Watson, C. J. and W. T. Khaled (2008). "Mammary development in the embryo and
adult: a journey of morphogenesis and commitment." Development 135(6): 995-1003.
Watt, F. M. and B. L. Hogan (2000). "Out of Eden: stem cells and their niches." Science
287(5457): 1427-30.
Weaver, M., N. R. Dunn, et al. (2000). "Bmp4 and Fgf10 play opposing roles during lung
bud morphogenesis." Development 127(12): 2695-704.
Weinstein, D. C. and A. Hemmati-Brivanlou (1999). "Neural induction." Annu Rev Cell
Dev Biol 15: 411-33.
Wells, J. M. and D. A. Melton (2000). "Early mouse endoderm is patterned by soluble
factors from adjacent germ layers." Development 127(8): 1563-72.
Welm, B. E., S. B. Tepera, et al. (2002). "Sca-1(pos) cells in the mouse mammary gland
represent an enriched progenitor cell population." Dev Biol 245(1): 42-56.
Whitman, M. and D. A. Melton (1992). "Involvement of p21ras in Xenopus mesoderm
induction." Nature 357(6375): 252-4.
Wiedemann, H. R. (1997). "Salivary gland disorders and heredity." Am J Med Genet
68(2): 222-4.
Williams, E. J., F. S. Walsh, et al. (2003). "The FGF receptor uses the endocannabinoid
signaling system to couple to an axonal growth response." J Cell Biol 160(4): 481-6.
202
Wilson, J. and A. S. Tucker (2004). "Fgf and Bmp signals repress the expression of
Bapx1 in the mandibular mesenchyme and control the position of the developing jaw
joint." Dev Biol 266(1): 138-50.
Wilson, S. I., E. Graziano, et al. (2000). "An early requirement for FGF signalling in the
acquisition of neural cell fate in the chick embryo." Curr Biol 10(8): 421-9.
Wolpert, L. (1969). "Positional information and the spatial pattern of cellular
differentiation." J Theor Biol 25(1): 1-47.
Woodward, W. A., M. S. Chen, et al. (2005). "On mammary stem cells." J Cell Sci
118(Pt 16): 3585-94.
Wysolmerski, J. J., W. M. Philbrick, et al. (1998). "Rescue of the parathyroid hormone-
related protein knockout mouse demonstrates that parathyroid hormone-related protein is
essential for mammary gland development." Development 125(7): 1285-94.
Wysolmerski, J. J. and A. F. Stewart (1998). "The physiology of parathyroid hormone-
related protein: an emerging role as a developmental factor." Annu Rev Physiol 60: 431-
60.
Yamasaki, M., A. Miyake, et al. (1996). "Structure and expression of the rat mRNA
encoding a novel member of the fibroblast growth factor family." J Biol Chem 271(27):
15918-21.
Yamasaki, M., Emoto, H., Konishi, M., Mikami, T., Ohuchi, H., Nakao, K., and Itoh, N.
(1999). "FGF-10 is a growth factor for preadipocytes in white adipose tissue." Biochem
Biophys Res Commun 258, 109-12.
Yang, X., D. Dormann, et al. (2002). "Cell movement patterns during gastrulation in the
chick are controlled by positive and negative chemotaxis mediated by FGF4 and FGF8."
Dev Cell 3(3): 425-37.
Yokohama-Tamaki, T., H. Ohshima, et al. (2006). "Cessation of Fgf10 signaling,
resulting in a defective dental epithelial stem cell compartment, leads to the transition
from crown to root formation." Development 133(7): 1359-66.
Yokoyama, H. (2008). "Initiation of limb regeneration: the critical steps for regenerative
capacity." Dev Growth Differ 50(1): 13-22.
Yoshimura, S., Y. Takagi, et al. (2001). "FGF-2 regulation of neurogenesis in adult
hippocampus after brain injury." Proc Natl Acad Sci U S A 98(10): 5874-9.
Yu, K. and D. M. Ornitz (2008). "FGF signaling regulates mesenchymal differentiation
and skeletal patterning along the limb bud proximodistal axis." Development 135(3):
483-91.
203
Zambrowicz BP, Imamoto A, Fiering S, Herzenberg LA, Kerr WG, Soriano P. (1997).
"Disruption of overlapping transcripts in the ROSA beta geo 26 gene trap strain leads to
widespread expression of beta-galactosidase in mouse embryos and hematopoietic cells."
Proc Natl Acad Sci USA 94,3789-94.
Zeichner-David, M. (2001). "Is there more to enamel matrix proteins than
biomineralization?" Matrix Biol 20(5-6): 307-16.
Zeller, R., J. Lopez-Rios, et al. (2009). "Vertebrate limb bud development: moving
towards integrative analysis of organogenesis." Nat Rev Genet 10(12): 845-58.
Zhang, C. C. and H. F. Lodish (2008). "Cytokines regulating hematopoietic stem cell
function." Curr Opin Hematol 15(4): 307-11.
Zhang, L., S. Kharbanda, et al. (1998). "Both autocrine and paracrine effects of
transfected acidic fibroblast growth factor are involved in the estrogen-independent and
antiestrogen-resistant growth of MCF-7 breast cancer cells." Cancer Res 58(2): 352-61.
Zhang, X., O. A. Ibrahimi, et al. (2006). "Receptor specificity of the fibroblast growth
factor family. The complete mammalian FGF family." J Biol Chem 281(23): 15694-700.
Zhao, X., I. O. Sirbu, et al. (2009). "Retinoic acid promotes limb induction through
effects on body axis extension but is unnecessary for limb patterning." Curr Biol 19(12):
1050-7.
Zhao, X., Z. Zhang, et al. (2000). "Transgenically ectopic expression of Bmp4 to the
Msx1 mutant dental mesenchyme restores downstream gene expression but represses Shh
and Bmp2 in the enamel knot of wild type tooth germ." Mech Dev 99(1-2): 29-38.

Abstract (if available)

Abstract The Fibroblast growth factor (FGF) family consists of 23 members, which play important roles during development, homeostasis and pathogenesis by controlling proliferation, migration and differentiation of cells in multiple organs. Among FGFs, we are interested in the role of FGF10 and its receptor, FGFR2b in development of ectodermal derivatives such as mammary gland, limbs and incisors. In this study we mainly used rtTA transactivator/tetracycline promoter approach allowing inducible and reversible attenuation of the FGFR2b pathway throughout the embryonic and adult mouse upon addition of doxycycline. Our study in mammary gland demonstrates the importance of FGFR2b signaling pathway for maintenance of the terminal end buds (TEBs) at the tip of the adult mammary gland. TEBs consist of transit amplifying cells (TACs), which are developed from adult mammary stem cells. We also report that while FGFR2b signaling is not crucial for the regenerative potential of the mammary epithelial stem cells, it has a critical role in the regulation of luminal epithelial lineage commitment of mammary stem cells in the adult mouse. In the second study, it is shown that FGFR2b signaling is critical for the regenerative capacity of adult incisors by controlling the proliferation of ameloblast progenitor cells. In the last study, we show that Apical Ectodermal Ridge (AER), the key structure for limb bud formation, requires FGFR2b signaling to maintain its structure. At last, in the following dissertation we discuss the crucial role of FGFR2b signaling pathway in controlling the behavior of stem/progenitor cells of different ectodermal-derived organs in both embryonic and adult mouse.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

The role of Wnt signaling in organogenesis: limb and lung

PDF

The role of fibroblast growth factor signaling on postnatal hepatic progenitor cell expansion

PDF

Constitutively active canonical BMP signaling in the apical ectodermal ridge disrupts dorsal-ventral and anterior-posterior axes resulting in impaired development of the distal (autopod) limb domain

PDF

The role of endoplasmic reticulum chaperones in adipogenesis, liver cancer and mammary gland development

PDF

FGFR2 Regulates Connective Tissue Development in the Skull

PDF

β-catenin couples self-renewal, induction and epithelial morphogenesis and patterning at the outset of mammalian nephrogenesis

PDF

Molecular programs in epithelial morphogenesis in mammalian organ systems

PDF

Proliferation and maturation events in second heart field cells during cardiovascular development activated by the Delta like ligand-4 mediated notch signaling

PDF

Harnessing the power of stem cell self-renewal pathways in cancer: dissecting the role of BMI-1 in Ewing’s sarcoma initiation and maintenance

PDF

Cross-species comparison of non-canonical Wnt signaling in the developing retina

PDF

Role of endoplasmic reticulum chaperones GRP78 and GRP94 in regulating unfolded protein response and cell survival

PDF

Roles of epithelial-mesenchymal transition and niche in tumorigenesis of tumor-initiating cells

PDF

The role of neuregulin receptors in cell differentiation and the response to inflammatory cytokines in the intestinal epithelium

PDF

Evaluation of the role of the basic Helix-Loop-Helix transcription factor Twist-2 in lymphocyte development and activation

PDF

Role of the bone marrow niche components in B cell malignancies

PDF

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

PDF

Understanding the role of BRCA1 and Aurora A kinase in polyploidy development in ovarian carcinoma precursor cells

PDF

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

PDF

The effects of hepatitis C virus infection on host immune response and signaling pathways

PDF

Nr5a2 balances lineage decisions to promote diverse connective tissue fates in the jaw

Asset Metadata

Creator Parsa, Sara (author)

Core Title Role of FGFR2b signaling pathway in the development of ectodermal derivatives

School Keck School of Medicine

Degree Doctor of Philosophy

Degree Program Pathobiology

Publication Date 12/14/2010

Defense Date 05/05/2010

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

Tag apical ectodermal ridge,bipotent progenitor cells,cervical loop,FGF signaling,fgf10,FGFr2b,incisor,limb,luminal progenitor cells,mammary gland,OAI-PMH Harvest,terminal end buds

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Bellusci, Saverio (committee chair), Dubeau, Louis (committee member), Hofman, Florence M. (committee member), Warburton, David (committee member), Widelitz, Randall B. (committee member)

Creator Email memolus@yahoo.com,sarapars@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m3595

Unique identifier UC152037

Identifier etd-Parsa-3818 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-437776 (legacy record id),usctheses-m3595 (legacy record id)

Legacy Identifier etd-Parsa-3818.pdf

Dmrecord 437776

Document Type Dissertation

Rights Parsa, Sara

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