Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Functional analysis of MSX2 and its role in skull patterning
(USC Thesis Other)
Functional analysis of MSX2 and its role in skull patterning
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI films
the text directly from the original or copy submitted. Thus, some thesis and
dissertation copies are in typewriter face, while others may be from any type of
computer printer.
The quality of this reproduction is dependent upon the quality of the
copy submitted. Broken or indistinct print, colored or poor quality illustrations
and photographs, print bleedthrough, substandard margins, and improper
alignment can adversely affect reproduction.
In the unlikely event that the author did not send UMI a complete manuscript
and there are missing pages, these will be noted. Also, if unauthorized
copyright material had to be removed, a note will indicate the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by
sectioning the original, beginning at the upper left-hand comer and continuing
from left to right in equal sections with small overlaps.
Photographs included in the original manuscript have been reproduced
xerographically in this copy. Higher quality 6” x 9” black and white
photographic prints are available for any photographs or illustrations appearing
in this copy for an additional charge. Contact UMI directly to order.
ProQuest Information and Learning
300 North Zeeb Road, Ann Arbor, Ml 48106-1346 USA
800-521-0600
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Functional Analysis Of MSX2 And Its Role In
Skull Patterning
by
Zequn Tang
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requiremants for the Degree
DOCTOR OF PHILOSOPHY
(Microbiology and Immunology)
May 2000
Copyright 2000 Zequn Tang
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UMI Number: 3018038
___ ®
UMI
UMI Microform 3018038
Copyright 2001 by Bell & Howell Information and Learning Company.
All rights reserved. This microform edition is protected against
unauthorized copying under Title 17, United States Code.
Bell & Howell Information and Learning Company
300 North Zeeb Road
P.O. Box 1346
Ann Arbor, Ml 48106-1346
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UNIVERSITY OF SOUTHERN CALIFORNIA
THE GRADUATE SCHOOL
UNIVERSITY PARK
LOS ANGELES, CALIFORNIA 90007
This dissertation, written by
under the direction of h.J.£t Dissertation
Committee, and approved by all its members,
bias been presented to and accepted by The
Graduate School in partial fulfillment of re
quirements for the degree of
DOCTOR OF PHILOSOPHY
Dean of Graduate Studies
£ )a tg ............Mar ch _ .1.0»_ ^ ,2000 _ _
DISSERTATION COMMITTEE
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ACK NO W LEDG EM ENTS
The reason I chose to study biology was by no means only out of
"intellectual curiosity", it also served as a way to get into the United States,
allowing me to work toward an advanced degree, and hopefully to pursue a better
life.
I understand that the meaning of the science is pursuing the truth, but I
never dreamed how difficult the unveiling the truth could be. Sometimes it seems
that all of the mysterious elements integrated into a formidable complex pattern are
so intimidating as to thwart any efforts to reveal it. Without Rob's encouragement, I
would not have been able to finish my dissertation work. He has tremendous
patience in training students, and he encourages independent thinking. One of his
favorite stories is the very different ways Russians and the Americans train pilot.
The Russians control every step of their pilots through a ground control center,
while the Americans let their pilots choose the best way to deal with specific
situations. I truly believe that the experience in Rob's lab will benefit me for the
rest o f my life; I learned to be a good observer, and to be able to see the facts
instead of what I wanted to see.
Over the past 4 to 5 years, I spent most o f my time staring at the
(osteogenic) "tip" of the skull, but only recently have I truly been able to appreciate
its beauty. I believe that both the spatial and temporal developmental pattern of the
ii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
tip reveals one o f the important developmental processes: membrane bone
formation, which is a complex process integrating cell migration, proliferation and
differentiation. The question in my mind is related to the homeodomain-containing
gene family, genes that have been shown to be expressed in both early body pattern
determination and later organogenesis stages. The single homeobox gene can also
be involved in the developmental procedure o f multiple different tissues. Whether
or not these processes share the same downstream genes is very intriguing. From
the pure aesthetic sense, I tend to believe they share the same down-stream targets,
but since evolution is essentially building upon "mistakes", it also very possible
there are different networks between these processes. In other words, exactly how
is the patterning process really associated with the cell proliferation and
differentiation processes?
I started my work trying to solve something, but the result has been just the
opposite. While the original questions remain, more mind-boggling questions have
arisen. I had no idea at the beginning that I would have to cross the millennium to
finish my thesis. Without my family's support, it would be impossible for me to be
where I am today.
iii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE OF CONTENTS
INTRODUCTION
Chapter 1
MSX2 gene dosage Influences the number of proliferative
osteogenic cells in the growth centers of the developing
murine skull
Chapter 2
MSX2 lost-of-function mutations have reduced target
DNA binding ability
Chapter 3
Functional study of MSX2 in FGF and BMP pathways
References
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF FIGURES
CHAPTER I
Figure 1.1. Alkaline phosphatase staining o f 4 Day postnatal MSX2
transgenic mouse skull
Figure 1.2. Spatial expression pattern o f ALP and BSP in osteogenic fronts
Figure 1.3. Influence of MSX2 transgene expression on osteogenic
lineage in the sagittal suture
Figure 1.4. Cell proliferation in sutural cell populations of MSX2
transgenic mice
Figure 1.5. Numbers of alkaline phosphatase-expressing and BrdU-labeled
cells in calvarial osteogenic fronts o f MSX2 transgenic mice
Figure 1.6. Alkaline phosphatase staining o f 4 Day postnatal MSX2
knockout mouse skull
Figure 1.7. Alkaline phosphatase staining o f 4 Day postnatal MSX2
Homozygous knockout calvarial bone
Figure 1.8. BrdU staining of 4 Day postnatal MSX2 homozygous
knockout mouse calvaria
Figure 1.9. Alkaline phosphatase staining o f NB and P4 stage
osteogenic fronts
Figure 1.10. BrdU staining of NB and P4 stage osteogenic fronts
Figure 1.11. Numbers o f alkaline phosphatase-expressing and
BrdU-labeled cells in calvarial osteogenic fronts of MSX2 homozygous
knockout mice
Figure 1.12. Whole mount skull stained with alizarin red
Figure 1.13. TUNEL assay on 4 Day postnatal mouse skull
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 1.14. Bone sialoprotein staining o f P4 stage postnatal MSX2
Homozygous knockout skull calvarial bones 49
Figure 1.15. N-CAM staining of P4 stage mouse skull 58
Figure 1.16. Model for MSX gene function 60
Figure 1.17. Cell affinity gradient model o f MSX2 function in
tissue regeneration 65
CHAPTER H
Figure 2.1. Alignment o f MSX gene family 70
Figure 2.2. MSX2 mutations 72
Figure2.3. In vitro translation o f MSX2 proteins 74
Figure 2.4. Mutant MSX2 proteins bind to target DNA sequence with
low affinity 76
Figure 2.5. Quantitation o f mutant MSX2 DNA binding 78
CHAPTER HI
Figure 3.1. Spatial expression pattern o f FGFR3 in osteogenic fronts 92
Figure 3.2. Spatial expression pattern o f FGFR2 in osteogenic fronts 94
Figure 3.3. Co-localization of FGFR3 and BrdU staining 97
Figure 3.4. The effect of BMP2 bead implant on osteogenic front 99
Figure 3.5. The effect o f FGF2 bead implant on osteogenic front 102
vi
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Abstract
Vertebrate skull formation is a complex developmental process, involving
extensive cell-cell and cell-matrix interaction. Human genetic studies have shown
that signaling molecules such as FGFs, as well as transcription factors like MSX2
and TWIST, are closely associated with the skull patterning process. A Prol48His
gain-of— function point mutation in MSX2 causes Boston-type craniosynostosis,
which is characterized by enhanced bone growth and premature fusion of the skull
suture. Mutations in FGFRs and TWIST also have been shown to cause varieties of
craniosynostosis. Recently, a new type of skull defect which affects bone formation
and mineralization, parietal foramina, has been linked to the MSX2. Human genetic
data suggests that MSX2 haploinsufficiency might cause defects in skull bone
formation. Using both MSX2 transgenic mice and MSX2 knockout mice as models,
we find that MSX2 gene dosage influences the number o f proliferating osteogenic
cells in the growth centers o f the developing skull. MSX2 transgenic mice show
enhanced growth of osteogenic fronts in the skull, with bone overgrowth invading
the suture space, mimicking the phenotype of craniosynostosis. MSX2 knockout
mice show decreased numbers of proliferating osteogenic cells in osteogenic fronts,
and part of the skull bone is missing, reflecting the characters of parietal foramina.
Interestingly, our results also suggested that MSX2 might also play an important
role in the osteoblast cell differentiation process. In MSX2 homozygous knockout
vii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
mice, a significant number o f osteoblastic cells failed to reach terminal
differentiation. I propose a simple model for the mechanism by which MSX2
regulates osteoblast proliferation and differentiation. In addition, I characterized the
two new mutations, R172H and D159-160, found in parietal foramina patients. My
results showed that these two mutations have significantly reduced affinity to
MSX2 target sequences, consistent with the model that a loss-of-function mutation
in MSX2 caused parietal foramina. Finally, my studies showed that MSX2
knockout mice have a defect in the FGF signaling pathway, suggesting that the
FGF signaling pathway and MSX2 are involved in the same developmental
process.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Introduction
Skull morphorgenesis is a complex developmental process. The vertebrate
skull is composed of bony plates with fibrous sutures, which serve as a protective
barrier. Since the historical discovery o f neural crest cells by His (His 1868), several
subsequent studies have shown that these neural crest cells contribute to the cranial
mesenchyme in sharks, teleosts and birds (Kastschenko 1888, Goronowitsch 1892).
Further morphological studies in the urodele by Landacre (Landacre 1921) and
experimental extirpation of parts o f amphibian neural crest cells by Stone (Stone-
1929) showed that skull tissues originate from cephalic neural crest cells. The more
detailed mapping of the origin of vertebrate cranial skeletons using quail-chick marker
system has been reviewed by Le Douarin (Le Douarin 1973). The neural crest-derived
cephalic skeleton includes parts of the membrane bones of the skull and face: frontal,
parietal, squamosal, columella, nasal, maxillar, vomer, palatine, quardratojugal and
mandibular bones. The neural crest-derived cells also form part o f the cartilaginous
bones in the cranial region: interorital septum, basipresphenoid, sclerotic ossicles,
ethmoid, pterygoid, meckel, quadrato-articular, hyoid, and part o f the pars cochlearis
of otic capsule. Somatic-originated skull bones includes basi- and exo-occipital, part
of pars canalicularis o f otic capsule. Other parts of skull bones are derived from
cephalic mesoderm origin: supra occipital, sphenoid (basipost, orbito-), and part of the
l
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
pars canalicularis and cochlearis o f otic capsule (Couly et al. 1993). Several methods,
such as vital dye analysis and labeling cells with replication defective retroviruses or
specific antibodies, have been used for neural crest cell lineage studies (review by
Fraser 1995). Using in vitro clonal analysis, Baroffio (1991) has shown that some
cranial neural crest cells are multipotent and can give rise to different types o f
descendents, such as neurons and cartilage. The neural crest cell precursors are located
at the junction site between the neural plate and epidermis. The inductive interaction
between these two tissues plays an important role in the genesis of neural crest cells,
and whenever these two tissue are in proximity, the neural crest cells will be
generated. Several neural crest markers have been identified, including SLUG, PAX3,
MSX 1/2, and DSL1. It has been shown that BMP4 and BMP7 can induce these neural
crest marks in neural plate explants, and the combination o f noggin and bFGF can
induce the expression of XSLU (reviewed by Selleck 1996). The cranial neural crest
cells originate at the neural tube-ectodermal junction, and proceed through epithelial-
mesenchymal transition. The prospective neural crest cells have a rounded
pleiomorphic shape, which is distinct from the regular, elongated shape and radial
orientation o f the neural tube cells. Changes in cell-cell and cell-matrix interactions
are also associated with neural crest cell epithelial mesenchyme transition and
migration. N-cadherin and N-CAM are down-regulated, and the disruption o f the basal
lamina is closely linked to the onset o f crest cell migration. The matrix components on
the surface o f the neural tube are substantially modified upon the initiation of
2
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
migration. Fibronectin and collagens IV, VI and I are detected, and believed to
promote neural crest cell migration. The a4(3l integrin (receptor for fibronectin) is
expressed in neural crest cells as soon as they migrate out o f the neural tube.
Antibodies against laminin-heparan sulfate proteoglycan, tenascin, p i subunit of
integrin, and RGD-containing peptides can perturb the neural crest cell migration
processes (Duband et al 1995). Once the neural crest cells escape from the epithelial
through the basal lamina, and after massive cell migration, these neural crest cells
move to the rostral and ventral regions o f the head (Hanken and Hall 1993). These
cells then undergo a series of inductive events with underlying neural tissue (Benoit
and Schowing, 1970), then these neural crest cell-derived mesenchymal precursors
will condense and synthesize bone matrix (Cohen 1993; Couly et al. 1993), forming
the unique protective skull vault. Type I collagen, cadherins, pi intergrins, RGD-
containing proteins, fibronetin splice variants and several other cell matrix proteins
and cell surface molecules have been implicated in osteoblast development (Bilezikian
et al., 1996), although their specific functions in calvarial bone formation is not clear.
Subsequent skull growth will occur at the osteogenic fronts of the skull to
accommodate the growth of the brain. Though skull bones originate differently from
bones found in other parts of the body, they have similar composition (Hanken and
Hall 1993).
While much is known about the origin and migration pathways o f skull
precursor cells, little is known about the molecular mechanisms controlling these
3
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
processes. Several signaling molecules have been detected at the developing skull.
FGF2 (bek) is expressed in the osteogenic front o f developing skull, and FGF9 is
expressed at the suture dural layer, calvarial mesenchyme and overlapping dermis.
BMP2 and BMP4 are expressed in the osteogenic front, and BMP4 expressed in the
suture mesenchyme and dura matter. MSX1, 2 are also expressed in the suture
mesenchyme of the developing calvarial skull (Kim et al., 1998). Interestingly, the
very similar sets of signaling molecules are expressed over and over again in neural
crest cell ontogeny, from the initial neural crest cell genesis and migration, to the final
skull bone genesis and further bone growth.
In recent years, human genetic studies have made a great contribution to the
understanding of the molecular mechanisms of skull formation. One of the common
human skull defects is craniosynostosis, the premature ossification and closure of
sutures between the skull vault bones. Approximately 1 in 3000 human infants is bom
with craniosynostosis (Lammer et al., 1987). Several mutations related to
craniosynostosis have been identified. One o f the interesting targets is a
homeodomain-containing protein MSX2. A mutation in the homeodomain of human
MSX2 causes autosomal dominant Boston-type craniosynostosis (Jabs et al. 1993).
The constitutive activation of FGF receptors is also associated with many human
craniosynostosis syndromes. FGFR1 mutations cause Pfeiffer syndrome, mutations in
FGFR2 cause Apert, Crouzon and Pfeiffer syndrome, while FGFR3 mutations cause
Crouzon with acanthosis (Webster and Donoghue 1997). Mutations in Twist, a basic
4
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
helix-loop-helix transcription factor, cause another type of craniosynostosis, Saethre-
Chotzen syndrome (Jabs et at. 1997). Recently, a new type of dominant autosomal
disease parietal foramina has been found to map to the MSX2 region (personal
communication). Human genetic studies have revealed the genes related to normal
skull development, but little is know about the mechanisms by which the candidate
genes cause the pathological cellular changes. Animal models are indispensable for
bridging the gap between molecular data from human genetics and the cellular level
disorders o f the corresponding tissue. Since all craniosynostosis have similar
pathological changes in suture, it will be very interesting to see whether these
candidate genes are functioning in the same pathway. Animal models also provide the
possibility to using genetic methods to determine the epigenetic relationship between
these genes.
The first part of my study focuses on the cellular mechanisms o f the MSX2
gene in skull development. Using both MSX2-overexpressing mice (created by Dr. Yi-
Hsin Liu, who also contributed parts of the analysis and wrote up the final results for
this particular mouse) and knockout mice (the MSX2 knockout mice were kindly
provided by Dr. Richard J. Mass), our study confirms that the MSX2 gene has a
dosage effect on the growth of osteogenic cells at osteogenic fronts. Overexpression of
MSX2 in mice promotes the growth of osteogenic cells at the osteogenic fronts,
consistent with the model that MSX2 gain-of-function mutation in human results in
enhanced growth o f osteogenic fronts and the premature fusion o f the developing
5
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
suture. On the other hand, MSX2 homozygous knockout mice have reduced number of
osteogenic cells at the osteogenic fronts. My results also confirmed that MSX2 knockout
mice have a gross morphological character very similar to human parietal foramina,
consistent with the model that MSX2 loss-of-fimction mutations have decreased bone
formation in humans, and have caused parietal foramina. The second part of my study
focuses on the structure-fimction relationship of MSX2 protein. In collaboration with
Dr. Wilkie, I found that two MSX2 homeodomain mutations, D159-160 and R172H,
which cause human parietal foramina, have dramatically reduced DNA binding
ability. In contrast to previous studies (Newberry et al., 1997), my results suggest that
intact homeodomain structure, thus the DNA-binding ability o f MSX2 proteins, is
crucial for its biological function. The third part of my study shows that MSX2 and
FGF pathways are functionally related. I have shown that in MSX2 homozygous
knockout mice there is a reduced level, as well as changes of expression domain, of
FGF receptors. Using an FGF2-soaked bead implant experiment, we have
demonstrated that MSX2 knockout mice have a reduced response to FGF signaling
compared to their wild-type littermates.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Chapter 1
MSX2 gene dosage influences the number of proliferative osteogenic
cells in the growth centers of the developing murine skull
Introduction
MSX genes have been isolated from a variety of animal species. They contain
a highly conserved homeodomain and less conserved N-terminal and C-terminal
domains. Drosophila and other invertebrates have a single MSX-related gene;
vertebrates possess several MSX genes. Mice, for example, have three such genes,
designated MSX1, 2 and 3. Within the homeodomain, each of these genes is
approximately 90% similar to the Drosophila MSX gene, msh, suggesting an ancient
and fundamental function.
The Drosophila msh has been implicated in the dorsal-ventral patterning of
neuroectoderm and mesoderm, and is expressed in the delaminating neuroblast cells.
A loss-of-fimction mutation in msh leads to failed neuroblast cell division and
migration (Isshiki et al., 1997). In the ascidian, MSX-a is expressed in presumptive
neural plate and notochord cells, which migrate over the anterior lip, and it is also
expressed in the presumptive muscles cells which migrate over the posterior lip o f the
blastopore. The expression pattern of MSX-a suggests that it may function in the
process o f mesoderm involution (Ma et al., 1996). A sea urchin MSX gene, Spmsx, is
expressed in invaginating archenteron and migrating secondary mesenchyme cells
7
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
(Dobias 1997). Over-expression o f Spmsx retards gastrulation and causes abnormal
spiculogenesis in sea urchin embryos (Tan et al., 1998).
Vertebrate MSX genes are expressed in locations where inductive epithelial-
mesenchymal interaction occurs. They are also expressed in association with a variety
of morphorgenic processes, including the formation o f the primitive streak,
neuralulation, and the migration of cranial neural crest cells. In addition, MSX1 and
MSX2 are expressed in branching alveoli in mammary gland development (Davidson
1995; Friedmann 1996). MSX1 and MSX2 have also been found to be expressed
during tissue regeneration, which involves massive amounts o f cell migration and cell
proliferation (Reginelli 1995; Muneoka 1992). MSX2 is expressed in tumor cell-lines
of epithelial and gonadal origin, such as adenocarcinoma, gastric cancer, renal cell
carcinoma, ovarian yolk sac tumor, choriocarcinoma and testicular seminoma, but it is
not detected in hematopoietic tumor cell-lines (Suzuki et al. 1993). There are some
evidences suggesting that MSX2 may be involved in the RAS-induced cell
transformation pathway (Takahashi et al. 1996). MSX2 has also been associated with
BMP4-induced apoptosis, and shown to function downstream of BMP4 (Ferrari et al.
1998; Marazzi et at. 1997; Rice 1999).
Jabs et al. (1993) demonstrated that a P148H mutation in human MSX2 gene
causes autosomal dominant Boston-type craniosynostosis. Our laboratory group (Liu
et at. 1995) created transgenic mice using CMV (Cytomegalovirus) or TIMP1 (tissue
inhibitor o f metalloprotease) promotor to drive the expression of wild type MSX2, as
8
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
well as P148H mutant MSX2 minigene. We showed that both MSX2 wild type and
P148H mutant transgenic mice have enhanced bone growth, suggesting that P148H is
a gain-of-function mutation. More recently, our group created another MSX2
transgenic mouse line, using 5.2-kb o f MSX2 5'-flanking sequence fused with wild
type MSX2 minigenes. This 5.2-kb promoter is expressed similarly to the endogenous
MSX2, and thus will better model how overexpresson/misexpression of MSX2 causes
craniosynostosis than the CMV and TIMP transgenes.
Recently, A. O. Wilkie found three patient families with new mutations in the
MSX2 gene (personal communication). One family has a two-amino-acid deletion
(D 159-160), while the other has a point mutation (R172H). The third family has a
large deletion in the MSX2 genome. All three families show a new type of autosomal
dominant skull defect called parietal foramina. Unlike craniosynostosis, which is
characterized by ossification and bony fusion of the suture, parietal foramina patients
have congenital defects of the parietal bone caused by deficient ossification around the
parietal notch. In the normal condition, the parietal notch is ossified during the fifth
fetal month; however, these patients have enlarged bone defects up to their adult life.
The mutation is believed to be caused by haploinsufficiency of MSX2 protein
(personal communication). However, nothing is known about the molecular
mechanisms of the disease.
R. Mass’s lab at Harvard University kindly provided us with MSX2 knockout
mice, created by ES cell homologous recombination. These two mice provided
9
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
excellent animal models for the two human genetic diseases with apparently opposite
phenotypes: craniosynostosis and parietal foramina. In our study, we used both 5.2kb
promoter MSX2 transgenic mice and MSX2 knockout mice to investigate the cellular
and molecular mechanisms of craniosynostosis and parietal foramina. We found that
over-expression of wild-type MSX2 causes enhanced early-stage osteoblast cell
proliferation at the osteogenic fronts, resulting in enhanced calvarial bone formation.
In MSX2 homozygous knockout mice, there is reduced osteoblast cell proliferation at
the osteogenic fronts, and thus reduced calvarial bone formation. My results are
consistent with the model that parietal foramina and craniosynostosis result from
opposite perturbation in the developmental pathway o f cranial osteoblast cells.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Materials and Methods
BrdU labeling
Conditions for BrdU labeling and detection were adapted from (Silvestrini et
al., 1994). Mice were injected intraperitoneally with lOOug o f BrdU per gram of body
weight. After two hours, the animals were sacrificed and their heads were fixed and
decalcified in 4% paraformaldehyde + 10% EDTA for 24 to 48 hours, depending on
the age of the animal (older animals required longer times.). The heads were
dehydrated through graded ethanol and embedded in paraffin. Sections were cut (5
pm), deparaffinized, and soaked in 3% hydrogen peroxide in methanol for 10 minutes.
They were washed three times in PBS and treated with proteinase K (lOOug/ml in
50mM Tris-HCl, 5mM EDTA, pH 8.0) for 20 minutes at 37°C. Subsequently, to
depurinate the DNA, sections were incubated in freshly prepared 2N HC1 for 45
minutes at room temperature and neutralized in 0.1 M sodium borate (pH 8.5) for 10
minutes. The sections were then rinsed three times in IX PBST (IX PBS, 0.1%
tween-20). Immunodetection of BrdU was then performed with a Zymed HistoMouse
Immunostaining kit according to the manufacturer's instructions.
Histochemical detection of alkaline phosphatase
The deparaffinized sections were washed three times in IX TBST (10X TBS: 8
g NaCl, 0.2 g KC1, 3g Tris pH 7.6 in 100 ml; 0.1% Tween-20). This was followed by
three washes in IX NMTT (lOOmM NaCl, 50mM MgC12, lOOmM Tris-HCl pH9.5,
0.1% Tween-20). NBT-BCIP (0.34 mg/ml nitroblue tetrazolium salt, 0.175 mg/ml 5-
11
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
bromo-4-chloro-3-indolyl-phosphate) staining solution in IX NMT was applied to the
sections until a dark purple color appeared. The reaction was stopped by washing the
sections with IX PBS, and then the sections were counterstained with Nuclear Fast
Red. Simultaneous detection of lacZ and alkaline phosphatase was carried out as
follows. The calvariae, stripped of skin and brain tissue, were stained in X-gal
solution lacking NP-40 and sodium deoxychoiate at 37°C for 4 to 16 hours. The
stained calvariae were then fixed for 1 to 2 hours in 4% paraformaldehyde, 10%
EDTA and subsequently embedded in paraffin and sectioned. The sections were
stained for alkaline phosphatase in the presence o f NBT-BCIP as described above.
Immunohistochemistry
The bone sialoprotein rabbit polyclonal antibody was a generous gift from Dr.
Larry Fisher of the NIH (NIDR). The FGFR2 antibody was purchased from Santa
Cruz Biotechnology. The primary antibody was allowed to react with tissue sections
o f mouse calvariae. A biotinylated, affinity-purified secondary antibody was then
bound to the primary antibody (Zymed Laboratories, San Francisco). An HRP-
conjugated strepadvidin was added as a signal amplifier. The immunoreactivity was
visualized with an AEC chromagen/substrate system that creates a red deposit. The
sections were counter-stained with hematoxylin.
Ceil counts and statistical analysis
Histological sections o f the midsagittal suture were visualized under Nomarsky
optics. Osteoblastic (alkaline phosphatase positive) cells in the osteogenic front,
12
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
which were clearly distinct from other sutural cells, were counted independently by
three different observers in at least three different sections. These cell counts were
averaged to obtain an overall average for each section and for each individual mouse.
A minimum o f three mice were analyzed in each experimental group. Our test o f the
statistical significance of the cell counts took into account the variability between
sections and between mice. An analysis of variance was used to test the effect of
genotype on cell number. We used a standard F-test, the denominator of which was a
linear combination of the mean square of the individual and the mean square o f the
tissue section, as determined by the expected mean squares (with the individual and
section taken as random effects and treatment taken as a fixed effect). Estimates of
the mean effect were based on mouse averages; standard errors were based on the
variance components for the individual and section variability. An identical approach
was used to test differences in cell numbers between developmental stages.
Alizarin red stain of adult skull
The Alizarin red stain protocol was adopted from Hogan et al. (1994). Briefly,
24-day-old mice were sacrificed. The skin, muscle, eyes, salivary gland, and as much
fat and loose tissue as possible was removed. The mice were placed in 2% KOH for 3
to 5 days, then stained in Alizarin Red (50mg Alizarin Red, Sigma A 5533, per liter of
2% KOH) for 3 to 5 days, until the bone was red. The stained bone tissue was stored
in glycerol. O f the 24 MSX2 homozygous knockout mice that were stained with
13
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Alizarin Red, 23 (96%) of the mice showed skull defects, while in 12 control wild-
type mouse, none showed skull defects.
Apoptosis
An ApopTag kit system (Intergen) was used to identify the number of cells
exhibiting apoptotic DNA fragmentation in both MSX2 homozygous knockout mice
and litter mate wild-type mouse paraffin slides. Deparaffinized 4-day-old mouse
sections were washed three times in PBS, and then the slides were treated with
proteinase K (lOOug/ml in 50mM Tris-HCl, 5mM EDTA, pH 8.0) for 20 minutes at
37°C. The slides were quenched with 3% hydrogen peroxide in methanol for 10
minutes. 3'-hydroxyl-DNA strand breaks in permeabilized tissue sections were
enzymatically labeled with digoxigenin-nucleotide by terminal deoxynucleotidyl
transferase and subsequently exposed to horseradish peroxidase-conjugated
antidigoxigenin antibody. Staining was developed in diaminobenzidine and sections
were counterstained with hematoxylin. Cells with typical apoptotic bodies were
considered as undergoing apoptosis and were counted.
Genotyping the MSX2 Knockout Mouse
PCR method was used for genotyping MSX2 knockout, and three primers were
used:
Sxg-4 (5’ CCCTCTCTGTCCTCTAGGAC 3’)
5’ end oligo anneals to wild-type allele
Sxg-2 (5 ’ TCTGGACGAAGAGCATCAGG 3 ’)
14
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5’ end oligo anneals to knockout mutant allele
Sxg-5 (5’ GCCTGAGGGCAGCATAGGCT 3’)
3 ’ end oligo anneals to both wild type and mutant alleles
PCR oligoes are synthesized from USC chemical core facility, PCR reagents were
obtained from Quiagen PCR kit (cat. #201203) and Life Technologies lOOmM dNTP
set (cat.# 10297-018)
Each PCR reaction requires the following reagents:
Qiagen buffer lOx 5pi
Q solution 5x 10pl
dNTP mix (12.5mM each) lpl
Sxg-4 or Sxg-2 primer lpl
Sxg-5 primer lpl
ddH20 30.75pl
Taq DNA polymerase 0.25pl
PCR program for MSX2 genotyping
94°C 5min
1 cycle
94°C 1 min
62°C 2 min
15
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
72°C 2 min
35 cycles
72°C 10 min
1 cycle
4°C hold
Run 20 (il of PCR reaction alongside DNA molecular weight marker VIII (Boehringer
Mannheim, cat.# 1336045) on a 2% agarose gel.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Results
Overexpression of MSX2 causes an increase in the proliferating
osteoblast cells in the osteogenic fronts of the calvarial bone
To address whether or not overexpression of MSX2 causes enhanced bone
growth, we first observed the general morphology of calvaria and sutures of transgenic
mice and their wild-type littermates. We focused on the sagittal suture, since this is
where craniosynostosis typically occurs in individuals affected with Boston type
craniosynostosis. Examining mice at different developmental stages, we found that in
the first week after birth, there were more cells at the growing edges (osteogenic
fronts) of MSX2-overexpressing transgenic mice than their wild-type littermates.
Osteogenic fronts o f these transgenic mice protruded into the sutural space; in some
cases, there was overlap of the calvarial bones (Fig. 1.1). To identify the
differentiation state of the cells at the osteogenic front, we used alkaline phosphatase
as a marker for early osteoblasts (preosteoblasts), and bone sialoprotein and
osteocalcin for late-stage (mature) osteoblasts (Bilezikian 1996). Interestingly, there
was a gradient o f differentiation at the osteogenic front: the more medially located
osteoblastic cells were in the early stage o f differentiation, while the lateral located
cells were in the later stage of differentiation (Fig. 1.2).
There was no significant difference in spatial distribution of osteoblast cells
between transgenic mice and their wild-type littermates. We detected alkaline
17
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fig.1.1. Alkaline phosphatase staining of 4 day postnatal mouse skull
Images of cross-sections of the midsagittal suture of MSX2 transgenic and
control mice showing alkaline phosphatase activity (ALP), blue color. Panel A shows
a P4 (postnatal day 4) stage wild-type mouse. Panel B shows a P4 MSX2 transgenic
mouse. Sections were visualized under differential interference contrast optics at 100X
magnification. Notice that in the transgenic mouse, the skull bone invades into the
suture space, and there is an overlap in the osteogenic fronts (arrows). Calvarial tissue
was harvested and processed as described in Materials and Methods.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
20
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fig. 1.2. Spatial expression pattern of ALP and BSP in osteogenic fronts
Images o f cross-sections of the midsagittal suture o f a wild-type mouse
showing alkaline phosphatase activity (blue in color) and BSP (brown in color)
antibody staining. The panel A shows the BSP staining o f a P4 stage wild-type mouse.
The panel B shows alkaline phosphatase staining of the same stage mouse. Notice that
BSP mainly stains at the bone matrix region, while alkaline phosphatase stains at the
osteogenic front medial to BSP staining sites as well as in the periosterium. Calvarial
tissue was harvested and processed as described in Materials and Methods. Primary
antibody was allowed to react with tissue sections of mouse calvariae. A biotinylated,
affinity-purified secondary antibody was then bound to the primary antibody (Zymed
Laboratories, San Francisco). An HRP-conjugated strepadvidin was added as a signal
amplifier. The immunoreactivity was visualized with an AEC chromagen/substrate
system that creates a red/brown deposit. Sections were counter-stained with
hematoxylin.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
phosphatase-positive cells in the osteogenic fronts as well as in the periosteum above
and below the calvarial bones. In both transgenic and wild-type mice, these alkaline
phosphatase-positive cells had a rounded, pleiomorphic shape (Fig. 1.3). There was no
premature expression of alkaline phosphotase in the suture mesenchyme cells of the
suture. Bone sialoprotein and osteocalcin protein was detected at the site of bone
matrix deposition and in the cells of the osteogenic front lateral to the tip (Fig. 1.3).
No ectopic expression of these later osteogenic markers was detected in either
transgenic or wild-type mice.
BrdU incorporation was employed to measure the number o f proliferative cells
at the osteogenic fronts. In both transgenic and wild-type mice, BrdU-positive cells
were located at the osteogenic front and in the periosteum cells, as well as in some of
the suture mesenchyme (Fig. 1.4.).
To quantitate the difference in the number of alkaline phosphatase-positive
cells and BrdU positive cells between transgenic and wild type mice, we counted cells
on the serial sagittal suture sections of postnatal day 0 to 4 mice. At each stage, more
than three pairs o f transgenic and wild-type mice have been counted. Our results
showed a significant difference at PO and P4 stages. At PO, transgenic mice had an
average of 3.1 ±1.1 BrdU-labeled cells, while controls had an average o f 0.6±0.4
labeled cells (P<0.05). By P4, there was an average of 11.5±2.3 BrdU-labeled cells in
the osteogenic front of transgenic mice compared to an average of 5.4±1.7 in controls
(P<0.05). Also, at PO, control mice had an average of 16.6 ±3.4 ALP-positive cells per
22
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
23
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fig. 1.3. Influence of MSX2 transgene expression on osteogenic lineage in the
sagittal suture
Images of cross-sections of the sagittal suture o f Msx2 transgenic and control
mice showing alkaline phosphatase activity (ALP, blue in color), panels (A-H) and
immunostaining with antibodies against FGFR2 (red in color, panels D, I) and bone
sialoprotein (BSP, red in color, panels E, J). Panels A through G are images of PO
(newborn) animals; panels C through J are images o f P4 mice. The left panels (A
through E) show controls, the right panels (F through J) MSX2 transgenics. B and G
are enlargements of the boxed areas in A and F. Sections were visualized under
differential interference contrast optics at 100 x magnification. Note that at both the
PO and P4 stages, there are more alkaline phosphatase-expressing cells in the
osteogenic fronts (arrows) of transgenic mice than in their littermate controls. FFGR2
staining is evident in cells of the osteogenic front, as well as in osteoblastic cells
adjacent to the bone (D, I). At P4, bone sialoprotein expression is present in cells of
the matrix of the advancing bone (E, J).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
WT TG
PO
B
A"
.7
/ W ; ■ 3* - ,
• -
' ~ ' / f - V / / /
r'-' / ; ^ f/ J / ;.
if"
' - A r
€ /
* '.w 3 ^ ' ✓ * ’ !L • - • * * '
- / , -■ /
■ • ■ n ■,/ V
t * ■ ' * - i
* -'
25
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fig. 1.4. Ceil proliferation in sutural cell populations of MSX2 transgenic
mice.
BrdU was injected intraperitoneally into MSX2 transgenic and littermate
control mice at PO (newborn) and P4. Two hours after injection, calvarial tissue was
harvested and processed as described in Materials and Methods. Sections through the
midsagittal suture were stained with an antibody against BrdU and counterstained with
hematoxylin. Stained sections were viewed under NOMARSKY optics. Images of
control sutures are shown in left panels, transgenic sutures in right panels. Panels B,
F, D, and H are higher magnification images of the boxed regions in the panels above
them. Note that at both PO and P4, the number of BrdU-labeled cells (arrow) in the
osteogenic front is greater in MSX2 transgenic mice than in their littermate controls.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
osteogenic front (95% confidence interval), while MSX2-transgeic mice had an
average o f 28 ± 4.3 ALP-positive cells per osteogenic front. The difference between
these mean numbers is statistically significant (PO.05). By P4, control mice had an
average o f 17.6 ± 3.7 ALP-positive cells per osteogenic front, while MSX2-transgenic
mice had a mean o f 26.6 ± 4.6 ALP-positive cells per osteogenic front (P<0.05)
(Fig. 1.5). I conclude that overexpression o f MSX2 caused an increase in the number
of BrdU-positive and alkaline phosphotase-positive cells at the osteogenic fronts of PO
and P4 mice (Fig. 1.5) (Dr. Susan Groshen o f USC/Norris Biostatistics Core performed
the statistical analysis).
MSX2 knockout decreases the number of proliferating osteoblast
cells in the osteogenic fronts of the calvarial bone
To assess the effect of loss of MSX2 function on osteogenic cell differentiation
and proliferation in osteogenic fronts, I performed BrdU and ALP labeling studies on
calvaria o f MSX2 knockout mice. I examined both the homozygous and heterozygous
MSX2 knockout mice. I found that while heterozygous mice had no phenotype,
homozygous exhibited significant changes in calvarial osteogenesis.
I examined MSX2 knockouts at different stages, from E l6.5 to postnatal day 5.
I found no phenotype in embryonic stages, but significant changes in postnatal stages.
In postnatal skulls, serial sections of PO to P5 sagittal sutures showed a retardation of
bone growth (measured by the distance between the osteogenic fronts). Calvarial
27
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
w
■ 5 5
o
< D
0)
40
P 4 PO
30
20
10
WT TG
W
©
o
< D
*-•
5
o
a
3
■ D
hm
C Q
15
10
PO P 4
WT TG W T TG
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fig. 1.5. Numbers of alkaline phosphatase-expressing and BrdU-labeled cells
in calvarial osteogenic fronts of MSX2 transgenic mice
Using light microscopy, as illustrated in Fig.3 and 4, we counted the number of
alkaline phosphatase-expressing and BrdU-positive cells in the osteogenic fronts of
the midsagittal sutures of MSX2-transgenic and their littermate control mice. A
minimum of three mice were analyzed in each group. For each mouse, approximately
10 serial sections were examined. Panel A shows the mean numbers of ALP-positive
cells per osteogenic front + the 95% confidence intervals. Panel B shows the mean
numbers of BrdU-positive cells per osteogenic front + the 95% confidence intervals.
In both panels, PO and P4 stages, the statistical analysis were described in Materials
and Methods. WT, wild type (littermate control); TG, MSX2-overexpressing
transgenic.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
bones of the MSX2 homozygous knockout mice were thinner and the osteogenic tips
smaller (Fig. 1.6). In the anterior sagittal suture, the difference between MSX2
knockout mice and wild-type mice are very dramatic, with substantial thinning of the
bone and a gross retardation of the osteogenic front. In some mice, part of frontal bone
was missing, and only several layers of ALP-positive cells were found at what would
normally be the bone region (Fig. 1.7). BrdU labeling studies revealed that these ALP-
positive cells were highly proliferative (Fig. 1.8).
Interestingly, in MSX2 homozygous knockout mice, osteoblast cells at the
osteogenic front have a more elongated morphology, which is different from the
rounded pleiomorphic-shaped osteoblast cells in wild-type mice (Fig. 1.9.). These
knockout cells express the early osteoblast marker ALP at a nearly normal level. It is
thus possible that the production of molecular marker is separable from the
morphological changes that occur during osteoblast differentiation.
I quantitated the difference in the number of alkaline phosphatase-positive
cells and BrdU-positive cells between MSX2 homozygous knockout mice and its
wild-type littermates (Fig. 1.10). I did this by counting cells in serial sections through
the sagittal sutures o f mice at postnatal day 0 and day 4. At each stage, I counted more
than three pairs of MSX2 homozygous knockout mice and wild-type mice. At PO,
control mice had an average of 18.0 ±5.5 ALP-positive cells per osteogenic front,
while MSX2 homozygous knockout mice had an average of 9.5 ±4.1 ALP-positive
cells per osteogenic front. The difference between these means is statistically
30
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fig. 1.6. Alkaline phosphatase staining of 4 day postnatal mouse calvaria
Images of cross-sections o f the frontal sagittal sutures o f MSX2 homozygous
knockout mouse and control mouse showing alkaline phosphatase activity (ALP) (blue
in color). Panel A is the skull o f P4-stage wild-type mouse. Notice that the two
osteogenic fronts (arrows) are close to each other, and the calvarial bone is thicker
relative to the MSX2 homozygous knockout mouse. Panels B and C show skulls from
P4-stage MSX2 homozygous knockout mice. Notice that the osteogenic fronts
(arrows) are far apart from each other, and the calvarial bone is much thinner than the
wild-type mouse. Calvarial tissue was harvested and processed as described in
Materials and Methods.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fig. 1.7. Alkaline phosphatase staining of 4 day postnatal MSX2 homozygous
knockout skull flat bones
Images of cross-sections o f the frontal sagittal sutures of MSX2 homozygous
knockout mice showing alkaline phosphatase activity (ALP) (blue in color). Panel A
shows a lower power view of P4-stage MSX2 homozygous knockout mouse. Notice
that no bone matrix is formed, instead there are only several layers of ALP (alkaline
phosphatase)-positive cells. Panel B is the enlargement of the boxed area in A. Panels
C is a high power view of ALP staining from a different P4 stage MSX2 homozygous
knockout mouse. Notice that these alkaline phosphatase-positive cells are rounded in
shape, and no mineralized bone can be found in the corresponding regions.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
35
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fig. 1.8. BrdU-staining of 4 day postnatal MSX2 homozygous knockout mouse
skull
BrdU was injected intraperitoneally into MSX2 homozygous knockout mice at
P4 stage. Two hours after injection, calvarial tissue was harvested and processed as
described in Materials and Methods. Sections through the frontal sagittal suture were
stained with an antibody against BrdU and counterstained with hematoxylin. Stained
sections were viewed under NOMARSKY optics. The sectioned regions are adjacent
to the ALP-positive region in Figl.7. Panel A is a lower power view of BrdU-positive
cells in the region. Panel B is the enlargement o f the boxed area in panel A. Notice
that these regions are highly proliferative. The red staining cells are BrdU-positive.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A B
37
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fig. 1.9. Alkaline phosphatase staining of osteogenic fronts of NB and P4
calvaria
Images o f cross-sections o f the frontal sagittal sutures of MSX2 homozygous
knockout mice and control mice showing alkaline phosphatase activity (ALP) (blue in
color). Panel A shows the lower power view of ALP staining o f P4 stage wild-type
mice, panel B is the enlargement of the boxed area in A. Panel C shows the lower
power view of ALP staining of P4 stage MSX2 homozygous knockout mice, panel D
is the enlargement of the boxed area in C. Notice that cells at the osteogenic fronts of
MSX2 homozygous knockout mice are elongated but still express a relatively normal
ALP level, and the calvarial bone is thinner. The osteoblastic cells at the osteogenic
front of wild-type mice are round, and the calvarial bone is relatively thicker than
MSX2 homozygous knockout mice.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
39
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fig.1.10. BrdU staining of NB and P4 stage osteogenic fronts
Images of cross-sections of the frontal sagittal sutures o f the MSX2
homozygous knockout mice and the control mouse showing BrdU staining (red in
color). Panel A is the lower power view of BrdU staining of P4 stage wild-type mice,
and panel B shows the enlargement of the boxed area in A. Panel C is the lower
power view of BrdU staining of the P4 stage MSX2 homozygous knockout mice
osteogenic front. Panel D shows the enlargement of the boxed area in C. Notice wild-
type mice have more BrdU-positive cells at the osteogenic fronts compared with
corresponding MSX2 homozygous knockout mice. BrdU was injected
intraperitoneally into MSX2 homozygous knockout mice at the P4 stage. Two hours
after injection, calvarial tissue was harvested and processed as described in Materials
and Methods. Sections through the frontal sagittal suture were stained with an
antibody against BrdU and counterstained with hematoxylin. Stained sections were
viewed under NOMARSKY optics.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
significant (P<0.05). By P4, control mice had an average o f 20.0 ALP-positive cells
per osteogenic front, while MSX2 homozygous knockout mice had a mean of 17.0
ALP-positive cells per osteogenic front (P<0.05) (Fig. 1.11). For the BrdU-labeling
study, atPO, MSX2 homozygous knockout mice had an average o f 1.3 ± 1.0 BrdU-
Iabeled cells, while controls had an average o f 1.9 ± 1.5 labeled cells (P<0.05). By P4,
there was an average of 2.9 ± 0.9 BrdU-labeled cells in the osteogenic front o f MSX2
homozygous knockout mice compared to an average of 5.5 ± 1.9 in the controls
(P<0.05) (Fig.1.11.).
I performed whole mount stains of skulls o f 3 week-old mice. A total of 24
MSX2 homozygous knockout mice were analyzed: 23 out of 24 (96%) had defects in
their frontal bones. The bone defect varied in shape and size; in some mice there was a
single foramen, others had bilateral foramina. O f the 14 wild-type littermates
examined, none of had bone defects (Fig. 1.12).
A TUNEL assay showed no obvious cell death at the suture and osteogenic
fronts of wild-type or MSX2 homozygous knockout mice (Fig. 1.13), suggesting that
the bone defect was not caused by apoptosis of the osteoblast cells.
Loss of MSX2 function affects the differentiation of calvarial
osteoblastic cells
In some anterior parts of MSX2 homozygous knockout mice skulls, the normal flat
bony tissue was missing. It was replaced by several layers o f ALP-positive
41
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
(0
K O W T KO W T
M 151“ PO P4
fl)
O
0)
>
5 10
C O
0
a
D
2 5
00
K O W T KO W T
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fig. 1.11. Numbers of alkaline phosphatase-expressing and BrdU-labeled cells
in calvarial osteogenic fronts of MSX2 homozygous knockout mice
Using light microscopy, as illustrated in Figures 9 and 10, we counted the
number of alkaline phosphatase-expressing and BrdU-positive cells in the osteogenic
fronts of the frontal sagittal sutures of MSX2 homozygous knockout and littermate
control mice. A minimum o f three mice was analyzed in each group. For each mouse,
approximately 25 serial sections were examined. Panel A shows the mean numbers of
ALP-positive cells per osteogenic front + the 95% confidence intervals. Panel B
shows the mean numbers o f BrdU-positive cells per osteogenic front + the 95%
confidence intervals. In both panels, P0 and P4 stages, the statistical analysis are
described in Materials and Methods. WT, wild-type (littermate control); KO, MSX2
homozygous knockout mice.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
44
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fig.1.12. Whole mount skull stained with alizarin red
The panel A shows a whole mount skull of a stage P24 wild type mouse
stained with alizarin red. Panel B shows the alizarin red staining o f the P24 MSX2
homozygous knockout mouse skull. Notice that there is a bony defect in the frontal
region o f MSX2 homozygous knockout mouse. O f the total 24 MSX2 homozygous
knockout mice analyzed, 23 of them had skull bone defects. O f the 14 wild-type mice
examined, none was found to have a skull bone defect. Conditions for alizarin red
staining was adapted from (Hogan, et al. 1994). Briefly, 24-day-old mice were
sacrificed. Skin and muscle were removed. The skulls were soaked in 2% KOH for 4
to 5 days. Then the skulls were stained in Alizarin red (2% KOH) for 3 to 5 days,
excess stain were removed with 2% KOH. The skull bones were then transferred into
glycerol and stored at 4°C.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fig. 1.13. TUNEL assay on 4 day postnatal mouse skull
Images of cross-sections of the frontal sagittal suture of P4 stage MSX2
homozygous knockout and control mouse showing apoptotic staining (red in color).
P4 stage calvarial tissues were harvested and processed as described in Materials and
Methods. The TUNEL assays were performed on the P4 wild-type, as well as P4
MSX2 homozygous knockout sections. The panel A shows DNAsel-treated wild-type
sample, which served as a positive control. Notice that almost every cell on the slides
is positive. Panel B and C show TUNEL assays on P4-stage wild-type and the MSX2
homozygous knockout mouse correspondingly. N otice that in both MSX2
homozygous knockout and wild-type mice, apoptotic reactions were positive in the
skin. None of them showed apoptotic reactions in the suture and osteogenic fronts.
The ApopTag was purchased from Intergen.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
osteoblast cells (Fig. 1.8). BrdU staining o f the corresponding regions showed that
these cells were in a proliferating state (Fig. 1.8), while the BSP staining o f the
corresponding region showed that they did not express the late differentiation marker,
BSP (Fig. 1.14.). In some other regions, BSP expression was in a punctuated pattern
and did not form a continuous flat skull bone (Fig. 1.14). These results suggested that
in MSX2 homozygous knockout mice, bone developmental processes were blocked in
the early ALP-positive osteoblast stage; i.e., these cells were in an active state of
proliferation and fail to reach terminal differentiation.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
49
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fig. 1.14. Bone sialoprotein staining of P4 stage postnatal MSX2 homozygous
knockout calvaria
Images of cross-sections of the frontal sagittal sutures of MSX2 homozygous
knockout mice showing bone sialoprotein (BSP) expression (red in color). Panel A
shows the lower power view of P4 stage wild-type mice. Panel B shows the lower
power view of P4 stage MSX2 homozygous knockout mice. Notice that the bone
sialoprotein staining are punctuated and does not form a continuous flat bone. Panel C
is the enlargement of the boxed area in B. Panel D is the higher power view of BSP
staining from sections adjacent to Fig. 1.7, which shows multiple ALP-positive cells
but no bone matrix formation. These ALP-positive regions are also highly
proliferative, as demonstrated in Fig. 1.8. Notice that these regions fail to express the
late osteoblast differentiation marker BSP. Only a small region showed BSP
expression (arrows).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Discussion
Craniosynostosis and parietal foramina are both congenital autosomal-
dominant diseases, and up to now little was known about the developmental
mechanismsunderlying them. Two models may explain the abnormal ossification of
suture space in Boston type craniosynostosis. One model states that P148H mutation
in MSX2 induces premature differentiation in the suture mesenchymal cells. Under
this model, we would expect to see prematured expression o f osteoblast differentiation
markers such as ALP and BSP in the sutural space. In our study with MSX2-
overexpressing transgenic mice, we did not observed any such phenomenon. The other
possible mechanism is that the P148H mutation in MSX2 can enhance bone growth at
the osteogenic center of skull bones, and delay the onset of the differentiation of
osteoblast cells in the growth center, keeping the osteogenic cells in a proliferating
state for a longer period of time. Further differentiation o f these cells would result in
more bone formation and ultimately craniosynostosis. Under this hypothesis, we
would expect to see more proliferating cells at the osteogrowth center, which in turn
would result in more early-stage osteoblast cells at the osteogenic front.
In the previous study, transgenic mice with either wild-type or P148H mutant
MSX2 minigene displayed a similar enhanced/overlapping skull bone growth
phenotype (Liu et al, 1995). This suggests that P148H is a gain-of-function mutation,
possibly enhancing bone growth and delaying the onset of differentiation, which
causes craniosynostosis syndrome. Using retroviruses expressing sense and antisense
51
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
MSX2 gene, Dodig et al. (1998) showed that continued high expression of MSX2
prevented osteoblast differentiation and mineralization of the extracellular matrix. In
contrast, expression o f antisense MSX2 RNA decreased proliferation and accelerated
differentiation. Cohen (1993) reported that in patients with several types of
craniosynostosis, there was enhanced growth o f bones into suture space and enhanced
osteoblast activity adjacent to the site of fusion. In patients with Apert’s Syndrome,
craniosynostosis caused by a mutation in the FGF receptors, there are an increased
number of alkaline phosphatase-positive cells in the suture tissue (Lomri et al. 1998).
Our transgenic mice analysis is consistent with the human disease data, and supports
the model that craniosynostosis is caused by enhanced bone growth at the osteogenic
fronts. However, our results did not address the nature o f P148H mutation in human
patients. The conclusion that this mutation is of a dominant active nature originates
from Liu’s previous P148H transgenic study, and is supported by former graduate
student Ma's (Ma et al. 1996) gel shifting data showing that P148H mutant protein can
bind MSX2 target DNA with higher affinity than wild-type.
In Wilkie’s human genetic study, five affected individuals from a four-
generation family had one MSX2 allele deleted (personal communication). In two
other sporadic cases, both cases exhibited heterozygous m utations in the
homeodomain of MSX2. One had an in-frame six nucleotide deletion and the other
had a point mutation. Human genetic studies suggest that these parietal foramina are
caused by haploinsufficiency of MSX2 protein, which results in parietal bone defects.
52
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Though there is a large number o f reports on the clinical manifestation of parietal
foramina, little is known about the pathological changes of this disease. The striking
similarity of the skull bony defect between human patients and MSX2 homozygous
knockout mice suggests that the MSX2 homozygous knockout mice give us an
excellent model to study the pathological process of this disease.
Two interesting points should be mentioned. One is that in the MSX2
knockout study, I did not find any significant skull defect in heterozygous MSX2
knockout mice. In fact, all of the heterozygous mice appear normal, except some
minor hair phenotype in a small fraction o f mice. This phenomenon could be caused
by the difference in sensitivity to the dosage o f MSX2 in mice and humans. Since both
of the two MSX gene families (MSX1 and MSX2) have been found expressed in the
skull, it is also possible that these two genes weigh differently in the process of skull
development in humans and mice. However, no similar bone defect has been observed
in MSX1 homozygous knockout mice (personal communication). I should mention
that there is no MSX2 expression data in patients o f human genetic study. So it is still
possible that in human patients both alleles o f MSX2 are affected, or that the
mutations found in patients may be dominant negative in nature.
The other interesting phenomenon is the position of the foramina. In humans,
the bone defect is in the parietal bone, while in mice it is in the frontal bone. This
could have resulted from the different expression pattern of human and mice MSX2 in
53
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
different parts o f the calvarial bone. Intriguingly, Boston type craniosynostosis occurs
in the frontal suture, at a similar location as in MSX2 homozygous knockout mice.
Histological findings in MSX2 homozygous knockout mice are very
interesting. In general, there is less bone formation, the osteogenic front is smaller, the
flat bone is thinner, and there are reduced numbers of both total and proliferative
(BrdU-positive) cells at the osteogenic fronts. The alkaline phosphatase and BrdU data
from the osteogenic fronts are consistent with the model that this autosomal-dominant
disease could have resulted from a reduced level of proliferation in osteoblast
precursor cells. The osteoblast cells at the osteogenic fronts o f MSX2 homozygous
knockout mice express ALP as they do in wild-type mice, although these cells have an
elongated shape instead of a rounded pleiomorphic one. This suggests that cell shape
change and differentiation marker expression are separable events.
In some regions of flat bone, instead of normal osteoid deposition there are
only several layers o f proliferative ALP-positive cells, suggesting that there is a defect
in the condensation o f osteoblast cells and osteiod deposition, thus affecting the
terminal differentiation and mineralization of the osteoblast cells. This result is
consistent with the finding that MSX2 may be involved in up regulation of the late
osteoblast differentiation marker osteocalcin gene (Hoffmann et at. 1994). However,
there are contradictory reports as to the function of MSX2 on the osteocalcin promoter
(Towler et al. 1994). It is possible that the late differentiation defect observed in
MSX2 homozygous knockout mice are secondary to proper ceil condensation and
54
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
proper ECM deposition. Several lines o f evidences suggest that ECMs play an
essential role in bone cell differentiation (Lanza et al. 1996).
My results suggest that parietal foramina are caused by reduced bone
formation. MSX2 promotes the proliferation of the early osteoblast cells, and it may
also affect the condensation, osteiod deposition and late stage differentiation
processes. TUNEL assay experiments also show that there is no massive osteoblast
cell apoptosis in MSX2 homozygous knockout mice compared with wild-type
littermate, supporting the view that reduced bone formation is caused at the level of
osteoblast cell formation rather than cell death. Combined with MSX2 transgenic data,
our conclusion is that MSX2 regulates the number o f early osteoblast cells by
promoting the proliferation o f early osteoblast cells. Interestingly, in MSX2
homozygous/ MSX1 heterozygous double knockout mice, the suture mesenchyme
cells express ALP prematurely, suggesting that MSX2 and MSX1 may function
similarly, inhibiting the premature differentiation of suture mesenchyme cells (data not
shown). Interestingly, in MSX2 homozygous knockout mice the proliferation defects
are localized in the osteogenic fronts. In other parts of the flat bones where osteoid
deposits are absent, massive amounts of proliferative cells are observed, suggesting
that MSX2 promotes the proliferation o f osteoblast cells, depending on the cellular
environment rather than pro-proliferation per se. In fact, MSX2 homozygous knockout
mice showed a defect in the differentiation of osteoblast cells, and these cells failed to
express the late differentiation marker BSP. Bone sialoprotein (BSP) has been shown
55
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
to be capable o f mediating cell attachment probably through the av|33 receptor, and
BSP expression is tightly associated with mineralization phenomena (Bilezikian et al.
1996). Thus there appear to be conflicting phenomena in the MSX2 homozygous
knockout mice. At the osteogenic front, the osteoblast cells have low proliferation
rates, while in some other areas, osteoblast cells are blocked in an ALP-positive
proliferation active stage, failing to further differentiation.
Having described the cellular changes in MSX2 homozygous knockout mice,
the next inevitable question is how MSX2 is linked to these changes. However, little is
known about the direct downstream targets o f MSX2, and MSX//nsA-related genes are
often expressed in highly proliferative and migrating cells, mutation in this gene
family may result in defects in cell migration and cell proliferation (Isshiki et al.,
1997, Reginelli et al., 1995; Muneoka 1992). Cell migration and proliferation involve
multiple cellular components, including cell surface molecules, ECMs, GTPases,
TIMPs and MMPs, et al. (Lauffenburger et al., 1996; Chant et al., 1995; Shapiro 1998;
Slack et al. 1996). Several lines of evidence suggest that MSX-related genes may
regulate some of these molecules. Lincecum etal. (1998) showed that MSX1 genes
regulate cadherin-mediated cell adhesion and cell-cell sorting. Takahashi et al. (1996)
showed that MSX2 may be involved in the Ras pathway. One interesting finding is
that in the ontogeny of neural crest-derived osteoblast cells, several cell surface
molecules and matrix proteins are closely associated with the MSX2 expression
profile, from the initial neural crest migration to the final osteoblast differentiation.
56
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
These molecules include type I collagen, cadherins, |31 intergrins, RGD- containing
proteins, N-CAM, fibronetin splice variants and several other cell matrix proteins and
cell surface molecules (Bilezikian et al., 1996). Interestingly, immunohistochemistry
data shows that in MSX2 homozygous knockout mice, N-CAM is up regulated
(Fig. 1.15).
I propose a simple model to address the function of MSX2 (Fig. 1.16) in the
osteoblast pathway. MSX2 may regulate specific sets of cell surface molecules and
determine the relative cellular affinity with the surrounding cells or ECM. At the
beginning of neural crest cell genesis, the expression of MSX2 will change the
expression pattern of cell surface molecules as well as certain ECM molecules, which
in turn will change the relative cell surface affinity with surrounding cells and
extracellular matrix. Thus MSX2-expressing cells will segregate from the surrounding
tissue. These cells will be released from cell contact inhibition, and cell proliferation
will ensue. In later stages, after extensive proliferation, large numbers of newly-
generated MSX2-expressing cells will have similar relative cell surface affinity, cell
condensation will be the dominant event, and cell contact inhibition will be restored.
This model is consistent with the finding that MSX2 regulates the expression of
certain cell surface molecules and ECMs such as N-CAM, and BSP. It may also
explain the fact that the MSX2 pro-proliferation effects depend on cell environment,
and in some cases ectopic expression of MSX2 may inhibit cell proliferation and
promote cell apoptosis (Ferrari et al. 1998). Since the proper regulation o f cell surface
57
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
a ? ■ - •"- *
$ \ -
. t :
©
9
’' ^ Siail
9 : ‘
I> -
« Z * m .
> JW I #
58
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fig.1.15. N-CAM staining of P4 stage mouse calvaria
Images o f cross-sections o f the frontal sagittal sutures of wild-type and a
MSX2 homozygous knockout mouse showing N-CAM antibody staining (red in
color). Sections were visualized under differential interference contrast optics. Panel A
shows a lower power view of N-CAM staining in 4 day postnatal wild-type mice.
Panel B is the enlargement of the boxed area in A. Panel C shows a lower power view
of N-CAM staining in 4 day postnatal MSX2 homozygous knockout mice. Panel D is
the enlargement of the boxed area in C. Notice that N-CAM is up-regulated in the
MSX2 homozygous knockout mouse, in the osteogenic fronts, suture mesenchyme,
and periostrium region. Calvarial tissue was harvested and processed as described in
Materials and Methods. Rabbit anti-mouse polyclonal anti-N-CAM antibody was
kindly provided by Dr. Chung.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
60
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fig. 1.16. Model for MSX gene function
This is a simplified model to illustrate the mechanism of MSX genes in
regulating the cell proliferation and differentiation processes. Shaded cells express
MSX-related genes. Panel A shows that in the early stage of neural crest cell genesis,
MSX-related genes regulate the expression o f sets of cell surface molecules and
ECMs. The expression of MSX-related genes will change the cell surface affinity of
these cells with surrounding cells and ECMs, and these cells will delaminate and
migrate out of the neural tube. The loss of MSX-related genes will result in defects in
cell delamination and cell migration. In Drososphila, which only has one msh gene,
deletion of msh results in a defect of neuroblast delamilation and migration. Once the
cells are delaminated, they lose cell contact inhibition and the cells start to proliferate.
Later on, newly generated cell share similar relative surface affinity, accumulated to
certain cell mass, and they start to condense, which in turn restores the cell contact
inhibition. These cells will then stop proliferation and begin differentiation. MSX2
loss-of-function may effect the process of cell condensation, these cells will continue
to proliferate. One o f the most important things in this model is that the cell surface
affinity is governed by the effects of total surface molecules and their relative
interaction with the extra-cellular matrix.
Panel B shows MSX-related gene in skull development. The shaded cells
express MSX2 and are proliferative. This early proliferation can amplify the number
of osteogenic precursors. Later on these cells condense to form bone. If the MSX2
61
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
defect mainly affects the early stage, we will expect to see a reduced number of
proliferating cells at the osteogenic fronts. If MSX2 defect mainly affects the late
stage, these cells will fail to condense and will remain in a proliferative state.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
molecules and ECMs is essential for a varieties o f cell functions, this model can also
explain the interesting fact that MSX2 homozygous transgenic mice (personal
communication) and MSX2/MSX1 double homozygous knockout mice have similar
open brain phenotypes (data not shown). A similar mechanism may be used by the
homeodomain-containing Hox proteins, which have been shown to be involved in
m ultiple developmental processes such as early body axial determination,
organogenesis, and may also play a role in cancer invasiveness (Krumlauf 1994; Cillo
1994).
According to this model, MSX2 may have a self-limiting process in regulating
cell behavior; the expression of MSX2 will result in cell segregation and cell
proliferation, which in turn will result in cell condensation and cell differentiation.
One o f the interesting findings in CMV-MSX2 transgenic mice is that some mice line
showed ectopic skull bone formation (personal communication). Though the MSX2
mini gene was driven under the CMV promoter, these cells were still able to turn off
the mini-gene expression and condense to form the bone tissue. This model directly
links the morphogenetic process with the cell proliferation and differentiation
processes, as well as to the expression of the cell surface molecules.
MSX2 has been shown to be closely associated with the regeneration process
(Akimenko et al. 1995). This relative cell affinity-cell contact inhibition-cell
proliferation model may also be used to explain one of the puzzling phenomena in
regeneration biology—that regenerative response is proportional to the amount of
63
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
tissue removed (Wolpert 1998). The cells at the proximal and distal ends of removed
tissue have differential cell surface affinity, so the cells at the both edges will
proliferate. The newly synthesized cells must bridge the relative surface affinity
gradient gap between the two ends, finally restoring homeostasis of cell surface
affinity (Fig. 1.17). This is one o f the possible ways for the remaining tissue to
“measure” the amount of tissue lost, and to control the timing of proliferation and
differentiation of newly generated tissue, thus regenerating the body form. Another
finding related to MSX genes function is by Tan et al. (1998), who showed that sea
urchin SpMSX regulates the arrangement of primary mesenchyme cells in a non-cell
autonomous manner, demonstrating that complex cell-cell or cell-matrix interactions
are under the control of this amazing gene family.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
r r x ...................
4 . 4
f t
f t
/ «
/ 4/
65
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fig. 1.17. Cell affinity gradient model of MSX2 function in tissue
regeneration
This model suggests that cells at the proximal and distal ends of regenerating
tissue have different overall cell surface characteristics and forms a “relative cell
surface affinity gap”. The cells at the edges of both ends are released from cell contact
inhibition and start to proliferate. The newly-generated cells fill up the “affinity gap”.
The amount of tissue generated is directly related to the corresponding affinity gap
between the two ends. This is one o f the possible mechanisms to “measure” the
volume o f the lost tissue. Only when the cells reach a certain affinity threshold, and to
restore the cell contact inhibition mechanism, they will stop proliferate and cell
differentiation will be the dominant event. This is my oversimplified view on one of
the most fascinating questions in regenerative biology. More detailed processes need
to be fill in, but several predictions may be made based on this model. P stands for
proximal, D stands for distal cell types.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Chapter 2
MSX2 loss-of-function mutations have reduced target DNA-binding
ability
Human genetic data and MSX2 homozygous knockout mice analysis support
the hypothesis that parietal foramina result from MSX2 loss-of-function mutations.
However, nothing is known about the biochemical nature of the two mutations in
sporadic parietal foramina patients. Both of the mutations are located in the
homeodomain of MSX2. Homeodomains are composed of 60 amino acids, with a
highly conserved DNA-binding motif. Homeodomain-containing genes play an
important role in body patterning and morphogenesis (Graba et al., 1996),
organogenesis (Vieille-Grosjean et al, 1997), as well as in human disease (Boncinelli
1997). Three-dimensional structure studies have shown that all homeodomains share a
very similar structure consisting of a flexible N-terminal arm and three a-helix
structures. The N-terminal arm contacts the minor groove o f the DNA, and is
important for high-affmity DNA binding. Three a-helix structures immediately follow
the N-terminal arm. Helix three, also called recognition helix, which contacts the
major groove of DNA, is essential for high affinity and specificity of DNA binding
(Sharkey et al. 1997). One of the two mutations involved in an in-frame six nucleotide
deletion (475-480delCGCAAG), corresponds to the deletion of amino acid Argl59
and Lys 160 at position 18 and 19 of the helix I of the homeodomain. Another
67
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
mutation is a 515G— >A transition (occurring at a CG dinucleotide), corresponding to
an Argl72His point mutation at position 31 of the helix II of the homeodomain. These
two sites are highly conserved in the MSX-related gene family, suggesting a
functional importance of these residues (Fig.2.1). By homology modelling with the
MSX1 homeodomain, residue 19 (Lysl60) of the homeodomain (helix I) forms a salt
bridge with the glutamate at position 30 of the homeobox. Similarly, residue 31 (Arg
172) of the homeodomain (helix II) is predicted to form a salt bridge with glutamate at
position 42 o f the homeodomain (Fig.2.2) (personal communication). A similar
mutation in MSX1 helix II homeodomain position 31 (Arg31Pro) causes reduced
thermostability and loss o f DNA binding. Patients with this MSX1 mutation have
agenesis of specific premolar and molar teeth (Hu et al. 1998).
To determine the nature of these mutations, I constructed the mouse MSX2
cDNA with three corresponding mutation sites. Since mice and humans are virtually
identical in their homeodomain, these mutations will be a very good approximation of
human MSX2 mutations. Using a TNT coupled transcription-translation system to
produce both wild-type, dell59-169, R172H and P148H MSX2 proteins (Fig.2.3). A
gel-shifting assay was performed to test the affinity o f these proteins to bind to target
DNA oligo sequences. The target DNA sequences were discovered by Ma (Ma et al.
1996) based on target selection methods. My results show that both the R K 159-160
deletion and the R172H substitution mutations bound the MSX2 target sequence with
68
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
reduced avidity (<10% of wild type) (Fig.2.4, Fig.2.5). My results support the model
that parietal foramina are caused by the loss-of-function mutations in the MSX2 allele.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
70
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fig.2.1. Alignment of MSX gene family
The alignment was carried out using the GeneJockey II program (Biosoft,
Cambridge UK.)- The boxed amino acids show where the human mutations are
located. P— P148H mutation site, which causes Boston-type craniosynostosis.
RK—Del 159-160 site, R—P172H mutation site. These two mutations cause parietal
foramina.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
HELIX 1 HELIX 2 HELIX 3
TLRKHKTNRKPRTPFrjrSQLLALERKFRqKQYLS lAEHAEFSSSL^L T jETQVKIWFQNRSAKAKRLC EAELEKLKMAAK
H XX H
Craniosynostosis (Bonston type) P148H
Parietal Foramina D159-60
R1 72H
72
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
F ig.2.2. MSX2 mutations
Schematic map showing the location of MSX2 mutation in the homeodomain.
The three boxed regions represent the three a-helix structures in the homeodomain.
Shaded amino acids represent the mutation sites, and the letters below the shaded
amino acid represent the mutated sequences. XX- represents the two amino acids
deletion mutation at this site.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 2 3 4 5 6 7 8
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
F ig.2.3. In vitro translation of MSX2 proteins
The in vitro translation was conducted using TnT-coupled reticulocyte lysate
systems. For 50pl reaction, the translation was carried out using 0.5 pg of DNA, 5 pi
o f 35S-methionine (Amersham), and reaction mixture. The protein then was loaded
into 10% SDS-page gel. The translated proteins showed expected size on the SDS gel.
Lane 1 and 2 are lpl, 5pi of wild-type MSX2 protein. Lane 3 and 4 are 5pl, lpl of
R172 mutant MSX2 proteins. Lane 5 and 6 are 5pl, lpl of D159-160 mutant MSX2
proteins. Lane 7 and 8 are 5 pi, lp l of P148H mutant MSX2 proteins. Notice that the
translation efficiency is very close (phospha-image data not shown).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
R172H Del 159-160
76
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission
Fig.2.4. The mutant MSX2 proteins bind to target DNA sequence with low
affinity
The EMSA were done with in vitro translated wild-type, RK159-160del and
R172H mutant MSX2 proteins. The double-stranded oligonucleotides were 5' end
labeled, using y32P-ATP (3,000 Ci/mmol) and T4 polynucleotide kinase, ethanol
precipitated and used as probes for EMSA. The shafting assay was performed as
described in Materials and Methods. BK, blank control (using mock translated lysate).
WT, wild type MSX2 protein (each lane with different amount of proteins: 0.5pl, lpl,
2pl and 4pl). R-172H, MSX2 protein with R172H point mutation (similar amount of
proteins is added to corresponding lanes: 0.5pl, lpl, 2pl and 4pl). Del 159-16, MSX2
protein with 159-160 deletion. Notice that both two mutations bind DNA with much
lower affinity compared with similar amounts of wild-type MSX2 protein.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Radioactivity i n Complex (arbitrary units)
4 0 -
[protein] (arbitrary units)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
F ig.2.5. Quantitation of MSX2 EMSA data
To quantify the relative affinity of MSX2 wild-type and mutant protein to the
target DNA sequence, EMSA was performed and gel was dried on to the Whatman
paper. The Whatman paper was then exposed on the Phosphorlmager (Molecular
Dynamics, CA) plates for 30 minutes, and the intensity o f each lane was recorded. The
resulting data was plotted. Open circles represent wild-type MSX2 protein. Filled
circles represent D 159-160 deletion mutation. Open squares represent R172H point
mutation.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Materials and Methods
In vitro mutagenesis
In vitro mutagenesis was conducted using the QuickChange Site-Directed
Mutagenesis Kit (Stratagene, La Jolla, CA) according to the manufacturer's
instruction.
The primers used were:
CC T G T C C A T A G C A G A G C A T G C T G A G T T C T C C A G C T C and
GAGCTGGAGAACTCAGCATGCTCTGCTATGGACAGG for RI72H mutation,
CACCACATCCCAGCTTCTAGCCTTGGAGTTCCGCCAGAAACAGTACCTGTCCATAG and
CTATGGACAGGTACTGTTTCTGGCGGAACTCCAAGGCTAGAAGCTGGGATGTGGTG f O r
Del 159-160 mutation,
CCGGAAGCCACGCACACACTTCACCACATCCCAG
and
CTGGGATGTGGTGAAGTGTGTGCGTGGCTTCCGG for P148H mutation.
The mutated nucleotide was confirmed by DNA sequencing.
In Vitro Translation
Full-length wild-type, R172H, D 159-160 and P148H MSX2 cDNAs were
cloned into the pCDNA3 vector (Invitrogen). The in vitro translation was conducted
using TnT-coupled reticulocyte lysate systems (Promega), according to the
manufacturer's instructions. Briefly, for 50 p .1 reaction, the translation was carried out
80
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
using 0.5 (ig o f DNA, 5 pi of 35S-methionine (Amersham), along with reaction
mixture. After incubation at 30°C for 2 hours, the reaction products were aliquoted.
Electrophoretic mobility shift assay (EMSA)
The MSX2 wild-type and mutant proteins were obtained through the coupled
in vitro translation system described above. The double-stranded oligonucleotides
containing consensus binding sequences for w ild-type MSX2 (5'-
CTGGGTAATTGAATGGGATC-31 and 5'-GATCCCATTCAATTACCCAG-3') were
obtained from the Sequence Core Facility of USC. The double-stranded
oligonucleotides were prepared by annealing synthesized complementary
oligonucleotides. These oligonucleotides were 5' end-labeled, using y32P-ATP (3,000
Ci/mmol) and T4 polynucleotide kinase, ethanol precipitated and used as probes for
EMSA. Binding reactions for MSX2 proteins were performed with various amounts of
MSX2 wild-type and mutant proteins in lOmM Tris-HCl, pH 7.5, 50 mM NaCl, 7.5
mM MgCb, 1 mM EDTA, 5% glycerol, 12.5 jig of BSA (amount varies to equalize
the total amount o f proteins in different reactions), 5 mM DTT, 0.5 mg poly (dl-
dC)-(dl-dC) (Pharmacia, Uppsala, Sweden), 0.1% NP-40, with an end-labeled probe
(approximately 35 fmol) in total volume of 15 pi at room temperature for 20 minutes.
Samples were electrophoresed in 5% nondenaturing acrylamide gel prepared in
lXTris-glycine buffer pH8.3. The gel was run at 100V at 4°C for 3 to 4 hrs.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Results And Discussion
RK159-160 deletion and the R172H have reduced DNA binding
affinity
Human parietal foramina patients showed a dramatic resemblance to the
MSX2 homozygous knockout mice phenotype, and human genetic study also suggests
that autosomal dominance of the disease is caused by MSX2 haploinsufficiency.
However, it is still possible that RK159-160 deletion and the R172H substitution are
either loss-of-function mutations, or dominant-negative mutations. Sequence analogy
with the MSX1 homeodomain family shows that these two mutations will likely
disrupt the homeodomain structure (personal communication). However, nothing is
known about whether these mutations will prevent MSX2 from binding to its target
DNA sequences. Carton et al. (1995) reported that transcriptional repression by MSX-
1 does not require homeodomain DNA-binding sites. Subsequently, Newberry et al.
(1997) showed that MSX2-mediated suppression of transcription also is independent
of three homeodomain helices. Our gel mobility shift results showed that in both
RK.159-160 deletion and the R172H substitution, there is a significant reduction in
target DNA affinity (<10% of the wild-type MSX2 protein). Hu et al. (1998) showed
that MSX1 R31P mutation does not interfere with the functions of wild-type MSX1
protein in transcription repression assay system, nor in the chicken wing biological
activity assay system. Their results suggest that the R31P mutation in MSXl
homeodomain (corresponding to MSX2 R172H mutation) is a loss-of-function
82
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
mutation, and does not have dominant negative activity. My results, in conjunction
with Hu et al. (1998), are consistent with the idea that homeodomains are
indispensable for the biological function of MSX-related gene family, and that human
parietal foramina are caused by haploinsufficiency of MSX2 protein.
Craniosynostosis and Parietal Foramina with opposite or similar
mechanisms?
From current data, we believe that craniosynostosis and parietal foramina
likely result from opposite perturbations in the osteogenic pathways o f the developing
skull. However, several remaining problems need to be addressed. First, to confirm
that parietal foramina is due to MSX2 haploinsuficiency, we need to have MSX2
expression data from the patients with one of the MSX2 alleles deleted, showing that
these patients do express the other allele. We also need to further investigate the nature
of the two mutations with a better biological assay system to address whether they
have dominant negative activity. Second, we also need to further investigate the
biochemical nature o f P148H mutation. Detailed analysis of P148H transgenic mice
will be very important in understanding this particular mutation. Until these problems
are solved, it is still possible that craniosynostosis and parietal foramina may result
from the same mechanism due to the loss-of-function mutation in MSX2: premature
differentiation with reduced osteoblast production. If the premature differentiation
occurs mainly in the late stage of the skull development, when most o f the skull bones
are already formed, then premature differentiation of suture mesenchyme could result
in craniosynostosis. If the premature differentiation occurs in the early stage of skull
83
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
development, then osteoblast proliferation will be severely affected, and there will be
a significant reduction of osteogenic cells, resulting in parietal foramina. Interestingly,
several reports showed that craniosynostosis and parietal foramina have been found in
the same patient (Thompson et al. 1984).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Chapter 3
Functional study of MSX2 in FGF and BMP pathways
Introduction
Fibroblast growth factors are single polypeptides that function in paracrine or
autocrine ways. It has been shown that FGFs are involved in a variety of cellular
activities, including promoting cell growth, stimulating cell migration, angiogenesis,
vasculogenesis, wound healing and tissue repair, differentiation and transformation
(Bilezikian 1996). To date, more than 13 FGFs have been isolated. FGF1 and FGF2
are widely expressed, while other FGFs are expressed in a restricted spatial-temporal
way (Smallwood et at. 1996). The functions o f FGFs are mediated by FGF receptors.
Currently four human FGF receptor genes have been isolated; these receptors have
distinct but overlapping expression patterns during the developmental stages of
embryos (Webster 1997). FGF pathways are involved in many developmental and
regenerative processes, including axial organization, mesoderm patterning,
organogenesis and morphogenesis. FGF pathways play a very important role in neural
crest cell ontogenesis. Mayors et al. (1995) showed that when Xenopus animal cap
explants from blastulae and gastrulae were treated with bFGF and noggin, they
expressed Xslu, an early neural crest cell marker. LaBonne et al. (1998) showed that
FGFs are important in Xenopus Neural crest induction. Recently, it has been shown
that FGF-mediated signaling pathways are important for cranial suture morphogenesis
85
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
and the neural crest-derived calvarial bone development. FGF9 transcripts are detected
in dural layers, calvarial mesenchyme and the overlaying epidermis o f the developing
skull, FGFR2 transcripts are also detected in osteogenic fronts of parietal bones (Kim
et al. 1998). Human genetic studies also point out the important role o f FGF pathways
in calvarial bone development. Activating mutations in FGF receptors cause a number
of craniosynostosis and dwarfism syndromes, FGFR1 mutations cause Pfeiffer
Syndrome, FGFR2 mutations cause Apert, Crouzon, and Pfeiffer Syndromes, FGFR3
mutations cause Crouzon Syndrome (Webster et al., 1997).
BMPs, or bone morphogenetic proteins, are members of the transforming
growth factors-P family. BMPs are synthesized as larger precursor polypeptides that
are processed to carboxyl-terminal mature protein dimers before their secretion from
the cell. Up to now, more than 20 members o f the BMP family have been isolated
from a variety of animals (Hogan 1996). The biological functions o f BMPs are
transduced through BMP receptors, which belong to the family of trans-membering
serine-threonine kinases. Two types of BMP receptors are known, and both type I
receptors and type II receptors are capable of binding to BMPs 2, 4, and 7. However,
the physical association o f types I and II receptors (heterodimer formation) may have a
unique biological function (Bilezikian etal., 1996; Hogan 1996). BMPs are expressed
in virtually all developing tissues. BMPs have been implicated in mesoderm
formation, and they also play a role in the patterning o f ectoderm and endoderm (Graff
1997). BMPs are also involved in multiple organogenesis process, such as kidney,
86
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
tooth, lung, gut, skin and hair, limb, skeleton tissues et al. (Hogan 1996). Like FGFs,
BMPs also play an important role in patterning the cranial neural crest (Graham et al.
1996), and BMPs play a special role in skeleton development. It has been shown that
BMP5 is expressed in the precartiliaginous condensations that give rise to the skeleton
elements. BMP2 and BMP4 are expressed in the mesenchyme surrounding the areas
where cartilaginous condensation takes place. BMP7 has also been found in the
perichondrial layer surrounding the rib, vertebrate, limb bone, and cranial-facial
skeletons (King etal. 1994).
BMP and FGF signaling pathways have been shown to associate with MSX-
related genes in multiple organogenesis process. Keranen et al. (1998) showed that
BMP2, BMP4, FGF4, MSX1 and MSX2 are co-localized in the different stages of
tooth development. Bei et al. (1998) showed that FGFs and BMP4 can induce MSX1
in early tooth development. Barlow, et al. (1997) showed that ectopic application of
BMP2 and BMP4 can activate MSX1 and MSX2 gene expression in the developing
facial primordia. Chuong et al. (1996) showed that in skin appendage regeneration,
there is an ordered expression of BMP2, BMP4, and FGF4, WNT 7a, SHH , MSX1
and MSX2. Vogel et al. (1995) demonstrated that local application of FGF-soaked
beads to posterior limb mesenchyme following ridge removal or addition of FGF to
cultured cells maintains expression of MSX1. Kim et al. (1998) showed that BMP4,
FGF9, MSX1, MSX2 were expressed in the sutural mesenchyme and dura mater of the
developing calvarial skull, and BMP2 and BMP4 were intensely expressed in the
87
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
osteogenic fronts of the developing skull. BMP4 bead implants induce expression of
both MSX1 and MSX2, while FGF4 bead implants induce only MSX1 (Kim et al.
1998).
To understand how signaling molecules like FGFs and BMPs regulate calvarial
osteoblast cell proliferation and differentiation processes, and the role o f MSX2 in
these biological processes, we implanted BMP2- and FGF2-soaked beads onto the
osteogenic front, and suture mesenchyme o f wild-type and MSX2 homozygous
knockout mice. My results suggested that the FGF2 and BMP2 signal can affect the
osteoblastic cells of the calvarial bone in a distinctive manner: BMP2 promotes the
osteoblast to express the early differentiation marker alkaline phosphatase, with little
effect on the proliferation of these cells, while FGF2 stimulates the osteoblast to
proliferate and remain in an undifferentiated state. MSX2 homozygous knockout mice
respond to BMP2 in a manner similar to their wild-type littermates. However, MSX2
homozygous knockout mice have a significantly reduced response to the FGF2 signal,
suggesting that MSX2 and FGF2 are functioning in the same pathway and MSX2 is
possibly a downstream target of FGFR2.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Materials and methods
Calvaria culture
The calvaria culture method was adapted from Kim et al. (1998). Briefly,
calvaria were dissected from mice of PO stages for bead implant experiments. The skin
and dura matter were removed from the skull before the start o f the organ culture.
Explants were placed on 0.8|il pore size filters (Minipore, type AA), supported by
metal grids and cultured in Dulbecco’s minimal essential medium supplemented with
10% fetal bovine serum and penicillin/streptomycin in a humidified cell culture
incubator with 5% € 0 2 at 37°C. Explants were cultured for 24 to 120 hours. In either
BMP2 or FGF2 bead implant experiments, a total of 6 mice were used for each group
(wild-type, MSX2 homozygous knockout mice and BSA control).
Treatment of beads
Affi-gel blue beads (Bio-Rad, cat. 153-7302) were used for the BMP2 implant,
and Heparin beads (Sigma, H-5263) were used for the FGF2 bead implant. Beads were
incubated in recombinant human BMP2 protein (100 ug/pl), FGF2 protein (500 ug/pl)
or bovine serum albumin (BSA) (same concentration as for BMP2 or FGF2) for at
least one hour. Before use, beads were rapidly washed in D ulbecco’s minimal
essential, and 6 to 10 beads were placed onto the osteogenic fronts or the mesenchyme
o f frontal sagittal suture of freshly dissected calvaria with a mouth-controlled capillary
pipette.
8 9
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Preparation of Tissues
After several days of culture, the bead-implanted calvaria were fixed with 4%
paraformaldehyde in PBS for 2 days, and then decalcified with 10% EDTA pH7.0,
changed every day for 2 days at 4°C. Then the tissue was dehydrated and embedded in
paraffin along with the supporting filters. Sections of 5 pm were mounted onto
TESPA- (3-aminopropyltriethoxysiIane) treated slides, dried for 2 hours at 45 °C and
stored in 4°C.
Cell proliferation assay
The bead-implanted calvaria tissues were labelled with bromodeoxyuridine
(BrdU, Sigma, lOOpg/ml) for 2 hours before harvesting for further tissue preparation
(see above). BrdU-positive cells were detected by immunoperoxidase staining using a
Histomouse SP kit (Zymed, cat.95-9541).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Results
Expression of FGFR2, 3 in sagittal suture of wild-type and MSX2
homozygous knockout mice
To understand the role o f FGF pathways in calvarial bone and suture
development, and the functional relationship between FGFs and MSX2, we examined
the expression pattern of FGFR1, 2 and FGFR3 on both newborn and postnatal day 4
mice. FGFR2 and FGFR3 are expressed in the osteogenic fronts and suture
mesenchyme of the postnatal sagittal sutures (Fig.3.1). However, we were unable to
detect FGFR1 expression in the osteogenic front or suture mesenchyme in postnatal
mice (data not shown). MSX1 and MSX2 are also expressed in the osteogenic fronts
and suture mesenchyme, with overlapping expression regions with FGFR2 and
FGFR3 (data not shown). Interestingly, both FGFR2 and FGFR3 have higher
expression levels at P4 than at the newborn stage. In the serial section of sagittal
sutures, both types of receptors are more intensely expressed in the places where two
sides o f the osteogenic fronts are closer to each other. Both FGFR2 and FGFR3
showed nuclear localization, and specific blocking peptides can completely block the
antibody signal, suggesting that the FGFR antibody staining is authentic (Fig.3.2).
These data are consistent with the finding that FGFR2 and FGFR3 may translocate
into nuclear matrix in vitro (Maher 1996; Johnston et al. 1995). In MSX2 homozygous
knockout mice, there is a significantly reduction in expression of both FGFR2 and
91
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fig.3.1.Spatial expression pattern of FGFR3 in osteogenic fronts
Images of cross-sections of the frontal sagittal suture o f P4 stage wild-type
mice as well as MSX2 homozygous knockout mice showing FGFR3 (brown in color)
antibody staining. Panel A shows the FGFR3 staining o f a P4 stage wild-type mouse,
while panel B shows FGFR3 staining o f the same stage in MSX2 homozygous
knockout mice. Notice that FGFR3 is expressed in the suture and osteogenic fronts of
wild-type as well as MSX2 homozygous knockout mice. But in MSX2 homozygous
knockout mice, there is an obvious reduction in the number of FGFR3-positive cells.
In wild-type mice, FGFR3 is located in the nuclear, while in the MSX2 homozygous
knockout mice, FGFR3 is more diffusely expressed in the cytoplasm. Calvarial tissue
was harvested and processed as described in Materials and Methods. The FGFR3
rabbit polyclonal antibody was purchased from Santa Cruz Biotechnology, Inc. The
primary antibody was allowed to react with tissue sections of mouse calvariae
overnight. A biotinylated, affinity-purified secondary antibody was then bound to the
primary antibody (Zymed Laboratories, San Francisco). An HRP-conjugated
strepadvidin was added as a signal amplifier. The immunoreactivity was visualized
with an AEC chromagen/substrate system that creates a red/brown deposit. Sections
were counter-stained with hematoxylin.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fig.3.2 . Spatial expression pattern of FGFR2 in osteogenic fronts
Images of cross-section of the frontal sagittal suture of P4 stage wild-type mice
as well as MSX2 homozygous knockout mice showing FGFR2 (brown in color)
antibody staining. Panel A shows the FGFR2 staining of a P4 stage wild type mouse,
while panel B shows FGFR2 staining o f the same stage in MSX2 homozygous
knockout mice. Notice that the FGFR2 are expressed in the suture and osteogenic
fronts of wild-type as well as MSX2 homozygous knockout mice. But in MSX2
homozygous knockout mice, there is an obvious reduction in the number of FGFR2-
positive cells. Similar to FGFR3 antibody staining, in wild type mice, FGFR2 is
located in the nucleus, while in the MSX2 homozygous knockout mice, FGFR2 is
more diffusely expressed in the cytoplasm. Calvarial tissue was harvested and
processed as described in Materials and Methods. The FGFR2 rabbit polyclonal
antibody was purchased from Santa Cruz Biotechnology, Inc. Primary antibody was
allowed to react with tissue sections of mouse calvariae overnight. A biotinylated,
affinity-purified secondary antibody was then bound to the primary antibody (Zymed
Laboratories, San Francisco). An HRP-conjugated strepadvidin was added as a signal
amplifier. The immunoreactivity was visualized with an AEC chromagen/substrate
system that creates a red/brown deposit. Sections were counter-stained with
hematoxylin.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
FGFR3 levels. Also, the signal is expressed more diffuse in the cytoplasm of the
osteogenic fronts as well as in the suture mesenchyme cells (Fig.3.2).
Expression of FGFR2 is co-localized with BrdU-positive cells
To investigate the role o f FGFR2 with osteoblast proliferation and
differentiation, I performed serial staining of FGFR2 and BrdU at the same section of
P4 sagittal sutures. It has been found that a small fraction o f the cells in the suture
mesenchyme that express FGFR2 are also BrdU-positive (Fig.3.3). In MSX2
homozygous knockout mice, no co-expression o f FGFR2 and BrdU uptake were
detected. These results suggest that the pro-proliferate effect of FGF on osteoblastic
and suture mesenchyme cells may be transduced through FGF receptors.
Local BMP2 beads on suture promote differentiation of osteogenic
cells
To study the role of BMP2 in calvarial bone development, I implanted BMP2
beads on the osteogenic front and suture mesenchyme o f the newborn wild-type as
well as MSX2 homozygous knockout mice. I did not observe significant amounts of
tissue around the BMP2 beads compared with the BSA control beads. However, in
both wild-type as well as MSX2 homozygous knockout skull cultures, the tissue
surrounding the BMP2 beads expressed the ALP marker. No ALP expression was
observed in the BSA control beads (Fig.3.4).
96
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
9 7
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fig.3.3. Co-localization of FGFR3 and BrdU staining
Images of cross-section of the frontal sagittal suture o f P4 stage wild-type mice
showing co-localization of FGFR3 and BrdU (brown in color) antibody staining.
Calvarial tissue was harvested and processed as described in Materials and Methods.
FGFR3 (brown in color) antibody staining was performed as described before.
Pictures were taken (as shown in panel A). The same sets o f slides were then used for
BrdU staining. Briefly, these slides were dehydrated through a series of alcohol, and
rinsing three times in xylene. After being rehydrated, BrdU staining was performed as
described before. Pictures were taken and compared with FGFR3 staining. In control
slides, anti-BrdU antibody was omitted, and no signals were detected.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fig.3.4. The effect o f BMP2 bead implantation on calvarial osteogenic fronts
Calvarial newborn explants with BMP2 beads placed on the osteogenic fronts
and cultured for 4 days. Panel A shows the BSA control bead on the newborn wild-
type calvarial osteogenic front. Notice that there was no cellular response around the
bead. Panel B shows BMP2 bead on the wild-type new bom mice calvarial osteogenic
front. Panel C shows BMP2 bead on the MSX2 homozygous knockout ostogenic
front. Notice that in both wild-type and MSX2 homozygous knockout mice, BMP2
promotes the osteogenic cells to express alkaline phosphatase
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Local FGF2 beads on suture development
A similar bead implantation method was employed to investigate the effects of
FGF2 on calvarial bone development. Both wild-type and MSX2 homozygous
knockout mice skull cultures were used. In wild-type mice with FGF2 bead
implantation, there were large numbers of cells surrounding the beads 48 hours after
the implant. The cells covered the surface of the FGF2 beads (Fig.3.5). In MSX2
homozygous knockout mice, the amount of tissue surrounding the FGF2 beads was
significantly reduced. Most o f the tissue covered only the basal parts of the beads. In
calvarial implanted with BSA control beads, no cellular responses were observed. The
cells surrounding the FGF2 beads, both in wild-type tissue and in MSX2 homozygous
knockout tissue, did not express ALP, suggesting that FGF2 has a function distinct
from BMP2.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A
102
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
F ig .3 .5 . The effect of FGF2 bead implantation on calvarial osteogenic fronts
Calvarial newborn explants with FGF2 beads placed on the osteogenic fronts
and cultured for 4 days. Panel A shows the BSA control bead on the newborn wild-
type calvarial osteogenic front. Notice that there is no cellular response around the
bead. Panel B shows FGF2 bead on the wild-type newborn mice calvarial osteogenic
front. Panel C shows FGF2 bead on the MSX2 homozygous knockout ostogenic front.
Notice that in both wild-type and MSX2 homozygous knockout mice, the cells
surrounding the FGF2 bead do not express alkaline phosphatase. In the wild-type
mouse, there is extensive cellular response surrounding the bead, while in the MSX2
homozygous knockout the cellular response is dramatically reduced.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Discussion
Substantial evidence suggested that FGFs play an important role in calvarial
bone development. Using cultured fetal rat calvarial, Canalis et al. (1980) showed that
purified FGF stimulates fetal rat calvarial cell proliferation. Subsequent reports and
recent findings (Roden 1987; Shen et al. 1989; Iseki et al, 1997; Kim et al. 1998)
confirmed this result. However, little is known about the downstream targets of the
FGF pathways in skull development. Though MSX1 and MSX2 have been shown to
be closely associated with FGF and FGFR in the developing skull (Iseki et al, 1997;
Kim et al, 1998), no functional relationship has been established. To understand the
functional importance o f MSX2 in the FGF2-mediated bone development, we first
examined the expression pattern of FGFR1, 2, 3 in postnatal skulls o f both wild-type
and MSX2 homozygous knockout mice. In wild-type mice, we found that FGFR2 and
3 are expressed in the nucleus of osteoblast cells and suture mesenchyme cells of the
postnatal skull. FGFR2 expression was co-localized with BrdU intake up to a single
cell resolution, supporting previous reports that the FGF pathway has a positive effect
on osteoblast cell proliferation. One interesting finding is that FGFR2 and 3 are more
intensely expressed in the suture where two sides of the osteogenic fronts are closer to
each other, suggesting that FGFR 2 and 3 may play a role in keeping the suture cells in
undifferentiated states to prevent premature differentiation of suture mesenchyme.
In MSX2 homozygous knockout mice, there was significant reduction in the
number of FGFR2- and FGFR3-positive cells, and they were expressed more diffusely
104
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
in cytoplasm. FGF2-soaked bead im plant experiments showed that MSX2
homozygous knockout mice have significantly reduced cellular response compared
with their wild-type littermates. In both cases, the cells surrounding the FGF2 beads
were alkaline phosphatase-negative, consistent with the previous reports that FGFs
inhibit ALP expression (Shen 1989). Previously, we have shown that in the osteogenic
fronts of MSX2 transgenic mice, FGFR2 was up-regulated. These data, in
combination with current data, lead us to propose a model in which MSX2 and FGFRs
form a regulatory loop, wherein MSX2 is indispensable to FGF2’s pro-proliferation
function in the calvarial skull, and MSX2 may also play a positive role in regulating
the expression o f FGF receptors. Based on this model, overexpression of MSX2 or
FGF will result in the excessive bone growth that causes craniosynostosis, while loss
o f MSX2 function will result in less bone formation causing parietal foramina.
The function of BMP in bone development has been the subject of extensive
study. Several models have been proposed for BMP function (reviewed by Hogan
1996). Using a cell line that behaves like a mesenchymal stem cell, Yamaguchi et al.
(1991) showed that BMPs stimulate multipotent stem cells to osteogenic
differentiation pathways and inhibit myogenic pathways. Thies et al. (1992) showed
that rhBMP-2 (recombinant human BMP-2) increased alkaline phosphatase activity in
W-20-17 stromal cells in a dose-responsive manner in the absence of an effect on
proliferation. BMP natural mutants and knockout mice also provided evidence of the
important role o f BMP in skeleton development. Se/se mice (BMP5 mutant) exhibit a
105
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
variety o f skeletal defects, including reduced size o f lumber vertebra processes,
deletion o f ribs, and small ears (Kingsley et al. 1992). BMP2 and BMP4 knockout
mice fail to form mesoderm and die at early embryonic stages (Winnier et al 1995;
Zhang 1996). BMP7 knockout mice die shortly after birth with severe renal
dysfunction and eye problems, but exhibit only a mild skeleton phenotype with
variable penetrance. It is believed that overlapping expression domains o f bone
morphogenetic protein family members may account for limited tissue defects in
BMP7-deficient embryos (Dudley 1997).
To investigate the functional role of MSX2 in BMP2-mediated pathways, we
implanted BMP2-soaked beads on both MSX2 homozygous knockout mice and their
wild-type littermates. Our results showed that postnatal osteogenic fronts and sutural
mesenchyme do not have a significant proliferative response to BMP2, and there is no
significant difference between wild-type and MSX2 homozygous knockout mice in
response to BMP2. Our results are consistent with the results of Thies et al. (1992),
but different from those of Kim et al. (1998). The difference could be explained by the
fact that we used different stages of skull from Kim’s group. In both wild-type and
MSX2 homozygous knockout mice, BMP promotes the expression of the early
osteoblast marker, alkaline phosphatase, while in BSA control mice no ALP
expression has been observed. Our results demonstrate that in the postnatal skull,
FGF2 and BMP play distinctive roles. FGF2 promotes proliferation of the postnatal
osteoblast cells and BMP2 promote differentiation.
106
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Sahni et al. (1999) showed that FGF signaling could inhibit chondrocyte
proliferation through the STAT-1 pathway. Dono et al. (1998) showed that FGF-2-
deficient mice had impaired cerebral cortex development. Still, the proliferation of
neuronal progenitors was normal, although a fraction of them failed to migrate to their
target layers in the cerebral cortex. Wang et al. (1999) showed that FGFR3
homozygous knockout mice exhibited bone overgrowth, suggesting that FGFR3 might
inhibit bone growth. Hebert (1994) showed that FGF5 homozygous knockout mice
have abnormally long hair. All these data suggest that the effect of FGF on cell
proliferation is highly dependent on cell stage and cell type, and in certain conditions,
FGF may have inhibitory functions. It is very possible that FGF-like signaling
molecules may regulate the expression o f transcription factors, such as the MSX
genes, which regulate the cell surface affinity o f the responding cells. The overall
consequences of cell-cell and cell-ECM interaction will determine the fate o f the cell:
to proliferate, to inhibitory, or to die (apoptosis). The original idea o f selective
adhesion of embryonic cells (Townes et al. 1955) and cell contact inhibition (Erickson
et al. 1980) will certainly play an important role in understanding the process of
morphorgenesis.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
References
Akimenko, M. A., Johnson, S. L., Monte Westerfield. Ekker, M. (1995).
Differential induction o f four msx homeobox genes during fin development and
regeneration in zebrafish. Development 121, 347-357.
Barlow, A.J., Francis-West, P.H. (1997). Ectopic application of recombinant BMP-
2 and BMP-4 can change patterning o f developing chick facial primordia.
Development 124, 391-398.
Baroffio, A., Dupin, E., and LeDouarin, N. M. (1991). Common precursors for
neural and mes ectodermal derivatives in the cephalic neural crest Development
112,301-305.
Bei, M, Maas, R. (1998). FGFs and BMP4 induce both Msxl-independent and
Msx 1-dependent signaling pathways in early tooth development. Development 125,
4325-33.
Benoit, J., and Schowing, J. (1970). Morphogenesis of the neural cranium. In
"Tissue Interaction during Organogenesis" (E. Wolff, Ed.) Gordon & Breach, New
York.
Bilezikian, J. P., Raisz, L.G. and Rodan, G. A. (1996). Principles o f bone biology
Academic press, pp. 155-230.
Boncinelli, E. (1997). Homeobox genes and human disease. Curr. Opion. Gene.
Dev. 7, 331-337.
Canalis, E., Raisz, L.G. (1980). Effect of fibroblast growth factor on cultured fetal
rat calvaria. Metabolism 29, 108-14.
Catron, K.M., Zhang, H., Marshall, S.C., Inostroza, J.A., Wilson, J.M., Abate, C.
(1997). Transcriptional repression by M sx-1 does not require homeodomain DNA-
binding sites. Mol. Cell. Biol. 15, 861-871.
Chuong, C. M., Widelitz, R.B., Ting-Berreth, S., Jiang, T.X. (1996). Early events
during avian skin appendage regeneration: dependence on epithelial-mesenchymal
interaction and order of molecular reappearance. J. Invest. Dermatol. 107,639-46.
Cillo, C. (1994). Hox genes in human cancers. Invasion Metastasis 14,38-49.
108
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Cohen, M. M., Jr. (1993). Sutural biology and the correlates o f craniosynostosis.
Am. J. Med. Genet. 47,581-616.
Couly, G. F., Coltey, P. M., and Le Douarin, N. M. (1993). The triple origin of
skull in higher vertebrates-A study in quail-chick chimeras. Development 117,409-
429.
Davidson, D. (1995). The function and evolution msx genes: pointers and
paradoxes Trends Genet. 11,405-411.
Dodig, M., Tadic, T., Kronenberg, M.S., Dacic, S., Liu, Y.H., Maxson, R., Rowe,
D.W., Lichtler, A.C. (1999). Ectopic Msx2 overexpression inhibits and Msx2
antisense stimulates calvarial osteoblast differentiation. Dev. Biol. 209, 298-307.
Dono, R., Texido, G., Dussel, R., Ehmke, H., Zeller, R. (1998). Impaired cerebral
cortex development and blood pressure regulation in FGF-2-deficient mice. EMBO
J. 17,4213-25.
Duband, J.L., Monier, F., Delannet, M. and Newgreen, D. (1995). Epithelium-
Mesenchyme Transition during Neural Crest Development Acta. A nat 154, 63-78.
Dudley, A.T., Robertson, E.J. (1997). Overlapping expression domains of bone
morphogenetic protein family members potentially account for limited tissue
defects in BMP7 deficient embryos. Dev. Dyn. 208,349-62.
Erickson, C. A., Tosney, K.W., Weston, J.A. (1980). Analysis of migratory
behavior of neural crest and fibroblastic cells in embryonic tissues. Dev. Biol. 77,
142-156.
Ferrai, D., Lichtler, A.C., Pan, Z.Z., Dealy, C.N., Upholt, W.B., Kosher, R.A.
(1998). Ectopic expression of Msx-2 in posterior limb bud mesoderm impairs limb
morphogenesis while inducing bmp-4 expression, inhibiting cell proliferation, and
promoting apoptosis. Dev. Biol. 197, 12-24.
Fraser, M.B. (1995). Origins and developmental potential of the neural crest. Exp.
Cell. Res. 218,405-417.
Fraser, S. E. (1992). Vital dye analysis o f cranial neural crest cell migration in the
mouse embryo. Development 116,297-307.
109
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Friedmann, Y., Daniel, C. (1996). Regulated expression of homeobox gene msx-1
and msx-2 in mouse mammary gland development suggests a role in hormone
action and epithelial-stromal interactioa Dev. Biol. 177,347-355.
Goronowitsch, N. (1892). Weiters uber die ektodermal Enstehung von
skeletanlangen im Kopfe der Wirbeltiere. Morphologisches Jahrbuch 20,425-428.
Graba, Y. et al. (1997). Drosophila hox complex downstream targets and the
function o f homeotic genes. BioEssays 19, 379-388.
Graff, J. M. (1997). Embryonic patterning: to BMP or not to BMP, that is the
question. Cell 89, 171-174.
Graham, A., Lumsden, A (1996). Patterning the cranial neural crest. Bioch. Soci.
Symp. 62,77-83.
Gruschus, J.M., Tsao, D.H., Wang, L.H., Nirenberg, M., Ferretti, J.A. (1999). The
three-dimensional structure o f the vnd/NK-2 homeodomain-DNA complex by
NMR spectroscopy. J. Mol. Biol. 289, 529-45.
Gruschus, J.M., Tsao, D.H., Wang, L.H., Nirenberg, M., Ferretti, J.A. (1997).
Interactions of the vnd/NK-2 homeodomain with DNA by nuclear magnetic
resonance spectroscopy: basis o f binding specificity. Biochemistry 36, 5372-80.
Gumbiner, B. M. (1996). Cell adhesion: the molecular basis of tissue architecture
and morphogenesis. Cell 84, 345-357.
Hanken, J. and Hall, B.K. (1993). The Skull vol. 1, ppl 12-145. The University of
Chicago Press.
Hebert, J.M., Rosenquist, T., Gotz, J., Martin, G.R. (1994). FGF5 as a regulator of
the hair growth cycle: evidence from targeted and spontaneous mutations. Cell 78,
1017-25.
His, W. (1868). Understanding uber die erste anlage des wirbeltierleibes. Die
erste Entwicklung des Huhnchen im Ei. Leipzig: F.C. Vogel.
Hoffmann, H.M., Catron, K.M., van Wijnen, A.J., McCabe, L.R., Lian, J.B., Stein,
G.S., Stein, J.L. (1994). Transcriptional control of the tissue-specific,
developmentally regulated osteocalcin gene requires a binding m otif for the Msx
family o f homeodomain proteins. Proc. Natl. Acad. Sci. USA 91, 12887-12891.
no
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Hogan, B.L.M. (1996). Bone Morphogenetic proteins : multifunctional regulator of
vertebrate development. Gene. Dev. 10,1580-1594.
Hogan, B.L.M., Beddington, R., Costantini, F., Lacy, E. (1994). Manipulating the
mouse embryo. Second edition. Cold Spring Harbor Laboratory Press pp381-383.
Horstadius, S. and Sellman, S. (1946). Experimentelle untersuchungen uber die
Determination des Knorpeligen bei Urodelen. Nova acta Regiae societatis
scientiarum Upsaliensis, ser. 4,1-170.
Howard, T.D., Paznekas, W.A., Green, E.D., Chiang, L.C., Ma, N., Ortiz de Luna,
R.I., Garcia Delgado, C, Gonzalez-Ramos, M., Kline, A.D., Jabs, E.W. (1997).
Mutations in Twist, a basic helix-loop-helix transcription factor, in Saethre-
Chotzon syndrome. Nature genetics 15,36-46.
Hu, G., Vastardis, H., Bendall, A.J., Wang, Z., Logan, M., Zhang, H., Nelson, C.,
Stein, S., Greenfield, N., Seidman, C.E., Seidman, J.G., Abate-Shen, C. (1998).
Haploinsufficiency o f MSX1: a mechanism for selective tooth agenesis. Mol. Cell.
Bio. 18,6044-51.
Isshiki, T., Takeichi, M , and Nose, A. (1997). The role o f the msh homeobox gene
during Drosophila neurogenesis: implication for the dorsoventral specification of
the neuroectoderm. Development 124, 3099-3109.
Jabs, E.W., Muller, U., Li, X., Ma, L., Luo, W., Haworth, I.S., Klisak, I., Sparkes,
R., Warman, M.L., Mulliken, J.B. et al. (1993). A mutation in the homeodomain of
the human Msx2 gene in a family affected with autosomal dominant
craniosynostosis. Cell 75, 1-20.
Johnston, C.L., Cox, H.C., Gomm, J.J., Coombes, R.C. (1995). Fibroblast growth
factor receptors (FGFRs) localize in different cellular compartments. A splice
variant of FGFR-3 localizes to the nucleus. J. Biol. Chem. 270, 30643-50.
Kastschenko, N. (1888). Zur Entwicklungsgeschichte der Selachier-embryos.
Anatomische Anziger 3,445-467.
Keranen, S.V., Aberg, T., Kettunen, P., Thesleff, I., Jemvall, J. (1998). Association
of developmental regulatory genes with the development o f different molar tooth
shapes in two species o f rodents. Dev. Gene. Evol. 208,477-486.
i l l
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Kim, H.J., Rice, D. P. C., Kettunen, P.J., and Thesleff, I. (1998). FGF-, BMP- and
Shh-mediated signaling pathways in the regulation o f cranial suture
morphogenesis and calvarial bone development. Development 125,1241-1251.
King, J.A., Marker, P.C., Seung, K.J., Kingsley, D.M. (1994). BMP5 and
molecular, skeletal and soft-tissue alterations in short-ear mice. Dev. Biol. 166,
112- 122.
Kingsley, D.M., Bland, A.E., Grubber, J.M., Marker, P.C., Russell, L.B., Copeland,
N.G., Jenkins, N.A. (1992). The mouse short ear skeletal morphogenesis locus is
associated with defects in a bone morphogenetic member o f the TGF beta
superfamily. Cell 71, 399-410.
Krumlauf, R. (1994). Hox genesin vertebrate development. Cell I, 191-201.
LaBonne, C., Bronner-Fraser, M. (1998). Neural crest induction in Xenopus:
evidence for a two-signal model. Development 125,2403-14.
Lammer, E. J., Cordero, J. F., Wilson, M. J., Oimette, D., and Ferguson, S. (1987).
Investigation o f a suspected increased prevalence o f craniosynostosis. Proc.
Greenwood Genet. Ctr. 6,126-127.
Landacre, F.L. (1921). The fate o f the neural crest in the head of the Urodeles. J.
Comparative Neurology 33,1-43.
Lanza, R. P., Langer, R. and Chick, W.L. (1996). Principles o f tissue engineering.
pp89-96 Landes and Academic Press.
Lauffenburger, D.A. and Horwitz, A.F. (1996). Cell migration: a physical
integrated molecularprocess. Cell 84, 359-369.
Le Douarin, N. M. (1973). The neural crest, Cambridge: Cambridge University
Press.
Lincecum, J.M., Fannon, A., Song, K., Wang, Y., Sassoon, D.A. (1998). Msh
homeobox genes regulate cadherin-mediated cell adhesion and cell-cell sorting. J.
Cell Biochem. 70, 22-28.
Liu, Y.H., Kundu, R., Wu, L., Luo, W., Ignelzi, M.A. Jr, Snead, M.L., Maxson,
R.E. Jr (1995). Premature suture closure and ectopic cranial bone in mice
expressing Msx2 transgenes in developing skull. Proc. Natl. Acad. Sci. USA 92,
6137-6141.
112
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Ma, L., Golden, S. and Maxson, R_ (1996). The molecular basis of Boston-type
craniosynostosis: the Prol48->H is mutation in the N-terminal arm of the MSX2
homeodomain stabilizes DNA binding without altering nucleotide sequence
prefemces. Hum. Mol. Genet. 5, 1915-20.
Ma, L., Swalla, B.J., Zhou, J., Dobias, S.L., Bell, J.R., Chen, J., Maxson, R.E.,
Jeffery, W.R. (1996). Expression o f an Msx Homeobox gene in ascidians: insights
into the archetypal chordate expression pattern. Dev. Dyn.205, 308-318.
Maher, P.A. (1996). Nuclear Translocation of fibroblast growth factor (FGF)
receptors in response to FGF-2. J. Cell. Biol. 134, 529-36.
Marazzi, G., Wang, Y. and Sassoon, D. (1997). Msx2is a transcriptional regulator
in the bmp-4 mediated programmed cell death pathway. Dev. Biol. 186,127-138.
Mayor, R.; Morgan, R. and Sargent, M.G. (1995). Induction of the prospective
neural crest of Xenopus. Development 121,767-77.
Muneoka, K. and David Sasson (1992). Molecular aspects o f regeneration in
developing vertebrate limbs. Dev. Biol. 152, 37-49.
Newberry, E.P., Latifi, T., Battaile, J.T., Towler, D.A. (1997). Structure-function
analysis o f Msx2-mediated transcriptional suppression. Biochemistry 36, 10451-
10462.
Noden, D.M. (1991). Vertebrate Craniofacial Development: The relation between
Ontogenic Process and Morphological Outcome. Brain Behav. Evol. 38, 190-225.
Reginelli, A.D., Wang, Y.Q., Sassoon, D., Muneoka, K. (1995). Digit tip
regeneration correlates with regions o f msxl (Hox7) expression in fetal and
newborn mice. Development 121,1065-1076.
Rice, D.P., Kim, H.J., Thesleff, L, (1999). Apoptosis in murine calvarial bone and
suture development. Eur. J. Oral. Sci. 107, 265-75.
Rodan, S.B., Wesolowski, G., Thomas, K., Rodan, G.A. (1987). Growth
stimulation of rat calvarial osteoblastic cell by acidic fibroblast growth factor.
Endocrinology 121, 1917-1923.
113
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Sahni, M., Ambrosetti, D.C., Mansukhani, A., Gertner, R., Levy, D., Basilico, C.
(1999). FGF signaling inhibits chondrocyte proliferation and regulates bone
development through the STAT-1 pathway. Genes Dev. 13,1361-6.
Selleck, M. A. (1996). The genesis of avian neural crest cells: A classical
embryonic induction. Proc. Natl. Acad. Sci. USA 93,9352-9357.
Shapiro, D.S. (1998). Matrix metalloproteinase degradation o f extracellular matrix:
biological consequences. Curr. Opion. Cell. Biol. 10,602-608.
Sharkey, M., Graba, Y. and Scott, M.P. (1997). Hox genes in evolution: protein
surfaces and paralog groups. TIG 13, 145-151.
Shen, V., Kohler, G., Huang, J., Huang, S.S., Peck, W.A. (1989). An acidic
fibroblast growth factor stimulates DNA synthesis, inhibits collagen and alkaline
phosphatase synthesis and induces resorption in bone. Bone Miner. 7, 205:219.
Slack, J.K., Higgins, P.J. (1996). Cytoarchitecture and cell growth control. Cell
Motility and the Cytoskeleton 33, 83-87.
Smallwood, P.M., Munoz-Sanjuan, I., Tong, P., Macke, J.P., Hendry, S.H., Gilbert,
D.J., Copeland, N.G., Jenkins, N.A., Nathans, J. (1996). Fibroblast growth factor
(FGF) homologous factors: new members o f the FGF family implicated in nervous
system development. Proc.Natl. Acad. Sci. USA.93,9850-9857.
Stone, L.S. (1929). Experiments showing the role of migrating neural crest
(mesectoderm) in the formation of head skeleton and loose connective tissue in
Rana palnstri. Wilhelm Roux' Archiv fur Entwicklungsmechanik der Organismen
118,40-77.
Stowers, J.C. (1995). GTPase cascades choreographing cellular behavior:
movement, morphogenesis, and more. Cell 81,1-4.
Suzuki, M., Tanaka, M., Iwase, T., Naito, Y., Sugimura, H., Kino, I. (1993). Over
expression of Hox-8, the human homologue of the mouse hox-8 homeobox gene, in
human tumors. Bioch. Biophys. Res. Commun. 194,187-193.
Takahashi, C., Akiyama, N., Matsuzaki, T., Takai, S., Kitayama, H., Noda, M.
(1996). Characterization of a human msx-2 cDNA and its fragment isolated as a
transformation suppressor gene against v-Ki-ras oncogene. Oncogene 12, 2137-
2146.
114
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Takahashi, C., Akiyama, N., Kitayama, H., Takai, S., Noda, M. (1997). Possible
involvement o f MSX-2 homeoprotein in v-Ras-induced transformation. Leukemia
11 Suppl 3, 340-3.
Tan, H., Ransick, A., Wu, H., Dobias, S., Liu, Y.H., Maxson, R. (1998). Disruption
of primary mesenchyme cell patterning by misregulated ectodermal expression of
Spmsx in sea urchin embryos. Development 201,230-46.
Thies RS, Bauduy M, Ashton BA, Kurtzberg L, Wozney JM, Rosen V (1992).
Recombinant human bone morphogenetic protein-2 induces osteoblastic
differentiation in W-20-17 stromal cells. Endocrinology 130,1318-24.
Thompson,M.; Baraister, M. and Hayward, R.D. (1984). Parietal foramina in
Saethre-Chotzen syndrome. J. Med. Gene. 21, 369-372.
Towler, D.A. (1994). Msx-2/Hox8.1: a transcriptional regulator o f the rat
osteocalcin promotor. Molecular Endocrinology 8, 1484-1493.
Townes, P. L. et al. (1955). Directed movements and selective adhesion of
embryonic amphibian cells. J. Exp. Zool. 128, 53-120.
Vieille-Grosjean, I., Hunt, P., Gulisano, M., Boncinelli, E., Thorogood, P. (1997).
Branchial hox gene expression and human craniofacial development. Dev. Biol.
183,49-60.
Vogel, A., Roberts-Clarke, D., Niswander, L. (1995). Effect of FGF on gene
expression in chick limb bud cells in vivo and in vitro. Dev Biol 171, 507-20.
Wang, Y., Spatz, M.K., Kannan, K., Hayk, H., Avivi, A., Gorivodsky, M., Pines,
M., Yayon, A., Lonai, P., Givol, D. (1999). A mouse model for achondroplasia
produced by targeting fibroblast growth factor receptor 3. Proc Natl Acad Sci U S
A. 96, 4455-60.
Webster, M. K. and Donoghue, D. J. (1997). FGFR activation in skeletal disorders:
too much o f a good thing. Trends Genet 13, 178-182.
Winnier, G., Blessing, M., Labosky, P.A., Hogan, B.L. (1995). Bone
morphogenetic protein-4 is required for mesoderm formation and patterning in the
mouse. Genes Dev. 9,2105-16.
Wolpert, L. (1998). Principles of development. Current Biology Ltd. pp.399-415.
115
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Yamaguchi, A., Katagiri, T., Ikeda, T., Wozney, J.M., Rosen, V., Wang, E.A.,
Kahn, A.J., Suda, T., Yoshiki, S. (1991). Recombinant human bone morphogenetic
protein-2 stimulates osteoblastic maturation and inhibits myogenic differentiation
in vitro. J.Cell. Biol. 113, 681-7.
Zhang, H., Bradley, A. (1996). Mice deficient for BMP2 are nonviable and have
defects in amnion/chorion and cardiac development Development 122,2977-86.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Cellular factors involved in mouse hepatitis virus replication
PDF
Identification of oncogenes cooperating in murine mammary tumorigenesis and transgenic mouse models of breast cancer
PDF
A new model for hepatitis delta virus transcription and replication
PDF
Expression of the RGR opsin and its function in the photic visual cycle
PDF
Biochemical characterization of hydrogen,potassium-ATPase-rich membranes from the gastric parietal cell
PDF
Cloning, characterization, anti -apoptotic molecular mechanism, transgenic and protein binding studies of mouse TOSO gene
PDF
Cloning and characterization of Grb4: A novel adapter protein interacting with BCR -Abl
PDF
A Study Of The Mechanism Of Lead Chromate Induced Neoplastic Transformation Of C3H/10T1/2 Mouse Embryo Cells
PDF
Effect of genetic factors in the development of childhood lymphocytic leukemia (ALL)
PDF
Immunogenicity of acute lymphoblastic leukemia: In vivo and in vitro studies of Bcr -Abl specific immune responses
PDF
Association between body mass and benign prostatic hyperplasia in Hispanics: Role of steroid 5-alpha reductase type 2 (SRD5A2) gene
PDF
Different roles of p160 coregulators in myogenesis
PDF
CYP17 polymorphism and risk for colorectal adenomas
PDF
Dietary fats, fat metabolizing genes, and the risk of breast cancer
PDF
Colorectal cancer risks in Singapore Chinese: Polymorphisms in the insulin-like growth factor-1 and the vitamin D receptor
PDF
Assessment of fatigue as a late effect of therapy among survivors of childhood leukemia
PDF
Biopsychosocial factors in major depressive disorder
PDF
Extent, prevalence and progression of coronary calcium in four ethnic groups
PDF
Characterization of the physiological ligand and function of a novel opsin RGR from the retinal pigment epithelium
PDF
BRCA1 mutations and polymorphisms in African American women with a family history of breast cancer identified through high throughput sequencing
Asset Metadata
Creator
Tang, Zequn
(author)
Core Title
Functional analysis of MSX2 and its role in skull patterning
School
Graduate School
Degree
Doctor of Philosophy
Degree Program
Microbiology and Immunology
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
biology, genetics,biology, molecular,OAI-PMH Harvest
Language
English
Contributor
Digitized by ProQuest
(provenance)
Advisor
Lai, Michael M.C. (
committee chair
), [illegible] (
committee member
), Ou, Jing-Hsiung James (
committee member
)
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c16-75771
Unique identifier
UC11327417
Identifier
3018038.pdf (filename),usctheses-c16-75771 (legacy record id)
Legacy Identifier
3018038.pdf
Dmrecord
75771
Document Type
Dissertation
Rights
Tang, Zequn
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA
Tags
biology, genetics
biology, molecular