University of ConnecticutOpenCommons@UConn

SoDM Masters Theses School of Dental Medicine

6-1-2007

The Role of Dlx5 in Murine Cranial Suture FusionAna T. Ortiz

Follow this and additional works at: https://opencommons.uconn.edu/sodm_masters

Recommended CitationOrtiz, Ana T., "The Role of Dlx5 in Murine Cranial Suture Fusion" (2007). SoDM Masters Theses. 151.https://opencommons.uconn.edu/sodm_masters/151

The Role of Dlx5 in

Murine Cranial Suture Fusion

Ana Teresa Ortiz

D.D.S., University of California at Los Angeles, 2004

A Thesis

Submitted in the Partial FulfIllment of the

Requirements for the Degree of

Master of Dental Science

at the

University of Connecticut

2007

APPROVAL PAGE

Master of Dental Science Thesis

The Role of Dlx5 • In

Murine Cranial Suture Fusion

Presented by

Ana Ortiz, D.D.S.

Major Advisor: __ --S!..2A-"t=---"-__ W __ {,._~ ______ ·_-·_· _____ _ Dr. Sunil Wadhwa

7 _1 " J.JJ}" Associate Advisor: ___ a_·._k_/_t~_tL_.-'_'~'-,,-j,C,,-1<:;...' _-.... dk;;....;:..:·~=--__ fl _______ _

Dr. Alex Lichtler

Associate Advisor: ____ ~_-_~....::' ;;......0::;.-=---.;"'--. --<-~_L_"-==--~_''''''''''=;;.;;:a..'''''_===-----­Dr. Carol Pilbeam

University of Connecticut

2007

ii

Table of Contents

Title Page ............................................................................................ .i Approval Page ...................................................................................... .ii Table of Contents ...................................................................................... .iii Abstract ............................................................................................ .iv

Chapter 1: Introduction ........................................................................ 1 Rationale ............................................................................ 2 Hypothesis ......................................................................... 3 Specific Aims/Objectives/Predictions .......................................... .3 Background ........................................................................ 4

A. Cranial Sutures ....................................................... 4 B. Dura ................................................................... 6 C. Disturbances in Suture Fusion ..................................... 8

1. Craniosynostosis ............................................. 9 2. Enlarged Parietal Foramina ............................... 11

D. Members ofthe TGF-beta superfamily and the regulation of cranial sutural fusion ............................................... 12

1. BMP .......................................................... 13 E. Transcription Factors Associated with BMP- signaling ....... 14

1. Dlx Genes .................................................... 14

Chapter 2: Wadhwa S, Bi Y, Ortiz A, Embree M, Kilts T, Iozzo R, Opperman L,

Young M. Impaired posterior frontal suture fusion in the biglycan/decorin double deficient mice. Bone 2007; 40:861-866 .................................................... 17-23

Chapter 3: Cranial Sutural Growth in Heterozygous Dlx5 (+/-) Mice ................ 24 A. Introduction ................................................. ~ ................ 25 B. Materials and Methods ................................................... .26 C. Results ....................................................................... 31 D. Discussion ................................................................... 33

Chapter 4: Discussion ................................................................................. 37

A. Future Studies ...................................................... 39 B. Significance ................................. ; ...................... 40

References .......................................................................................... 41 Figure Legends ..................................................................................... 48 Figures ........................................................................................... 49-57

111

ABSTRACT

Objective: The genetic etiology for the majority of sutural growth disturbances,

in humans, is unknown. Therefore, the development of our knowledge of the genes

involved in suture fusion is essential for progress in the treatment of individuals with

cranial suture fusion disorders. In this study, we have examined Distal-Iess-5 (Dlx5), a

homeobox gene, which is required for Bone Morphogenic Protein (BMP)-mediated

osteoblast differentiation and is thought to play an important role in the development of

mineralized tissues. The significance ofDlx5 in these biologic processes has developed

interest in the role of this gene in cranial suture fusion. Methods: We examined the

expression of Dlx5 in the fusing and non-fusing cranial sutures of wild-type (wt) mice

by RT-PCR and immunohistochemistry. In order to determine the role ofDlx5 in the

cranial sutures, the Sagittal (S) and posterior frontal suture (PFS) were analyzed from

Dlx5 +/- and wild-type mice at postnatal day 10. To determine ifDlx5 +/- mice had a

decrease in osteoblast differentiation in the cranial sutures, Dlx5 +/- and wt mice were

bred with mice transgenic for the 2.3 kb fragment of the collagen Type I promoter fused

to a Green Fluorescent Protein (2.3 OFP). The 2.3 GFP promoter has been shown to be

an accurate marker of differentiated osteoblastic cells (Kalajzic et al., 2002). Results:

Dlx5 was abundantly expressed in the Sand PFS during sutural maturation (postnatal

day (10-45) and there was an increase in the expression ofDlx5 in the posterior frontal

suture (PFS) compared to the sagittal suture (S) at day 35. At day 10, there was an

increase in space (p< 0.05) between the parietal bone fronts in the sagittal suture from

Dlx5 compared to wt mice. Conclusiou: Genes involved, or thought to be involved,

in cranial suture fusion are of great interest among those studying both normal and

IV

abnormal calvarial development. Knowledge of the intricate processes and genes

responsible for proper growth of the cranium is essential to fmding therapies for

patients who are affected by disorders involving abnormal sutural fusion, namely

craniosynostosis. Evidence that a Dlx5 signaling pathway is involved in suture fusion,

may make available a new candidate gene for analysis in patients with these sutural

growth disturbances.

v

INTRODUCTION

Cranial sutures function as the major sites of bone expansion during postnatal

craniofacial growth and allow development of the skull to accommodate rapid growth

ofthe brain. For sutures to function as growth sites, they need to remain unossified, and

simultaneously allow formation of new bone at the edges of the overlapping bone

fronts. Too little or delayed bone growth at these sutures results in wide-open fontanels,

whereas too much or accelerated bone growth results in osseous obliteration of the

sutures (Opperman 2000).

Despite the obvious differences in human and murine craniofacial

characteristics, there is a marked similarity and conservation in the molecular pathways

for the assembly and growth ofthe cranial vault between the two species (Warren,

Greenwald et aL 2001). In mice, all the sutures of the cranial vault remain patent for the

life of the animal; except for the posterior frontal suture (PFS), which fuses in a

predictable manner between 21-45 days postnatal (Opperman 2000). In humans, the

equivalent metopic suture fuses within the first three years postnatal (Vu, Panchal et al.

2001; Weinzweig, Kirschner et aL 2003). Therefore, the murine PFS serves as an

excellent model of post-natal human sutural development (Warren, Greenwald et al.

2001).

BMP signaling has been linked to murine PFS fusion, and is expressed in

developing (Kim, Rice et al. 1998) and in adult murine cranial sutures (Warren, Brunet

et al. 2003). In fact, Warren et al. showed that inhibition ofthe BMP signaling

pathway, by adenovirus mis-expression ofthe BMP antagonist Noggin, blocks murine

1

PFS fusion in vivo (Warren, Brunet et al. 2003). However, BMP added to murine

calvarial organ culture sutures did not cause sutural fusion (Kim, Rice et al. 1998).

Taken together, these results suggest that BMP is necessary but not sufficient for

murine calvarial suture fusion.

Many extracellular and intracellular proteins, such as Dlx5, regulate the BMP

signaling pathway. In an avian calvarial explant culture system, Dlx5 transcription was

upregulated by BMP and inhibited by the BMP antagonist Noggin (Holleville, Quilhac

et al. 2003) which proved that Dlx5 acts downstream from BMP signaling. In

subsequent investigations of osteoblastic cell cultures, it was shown that BMP causes an

increase in the expression of Dlx5 (Miyama, Yamada et al. 1999; Takagi, Kamiya et al.

2004), and that regulation of osteoblast differentiation by BMP is dependent on Dlx5.

Others showed, that in osteoblast-like cells, BMP can induce markers of osteoblast

differentiation including alkaline phosphatase (Kim, Lee et al. 2004), Runx2 (Lee, Kim

et al. 2003), and Osterix (Lee, Kwon et al. 2003) all of which have been shown to be

dependent upon Dlx5.

RATIONALE

The incidence of craniosynostosis in the United States is about .04-.1 %.

Although this developmental disorder is rare, it may result in severe malformations of

the cranial vault and in the more harshly affected, increased pressure on the brain. The

most detrimental consequences described are mental impairment, loss of hearing, and

poor vision. The severity of these complications requires that infants, sometimes at a

2

young age, receive several surgeries to avoid further damage and to correct the

deformity.

The genetic etiology for the majority of sutural growth disturbances in humans

is unknown. Providing evidence that a Dlx5 signaling pathway is involved in sutural

fusion, may make available a new candidate gene for analysis in patients with sutural

growth disturbances.

HYPOTHESIS

In this study, we plan to examine the role of the transcription factor, Dlx5, in

controlling murine cranial suture development and growth. We hypothesize that Dlx5

expression will be increased in the murine PFS during the time of posterior frontal

suture fusion and that there will be a decrease in suture fusion in the PFS in

heterozygous Dlx5 (+1-) mice.

SPECIFIC AIMS/OBJECTIVES/PREDICTIONS

Aim 1. Examine the expression and localization of Db5 in the posterior

frontal and the sagittal cranial sutures.

The posterior frontal suture in mice fuses between days 21 and 45 postnatally.

Our working hypothesis is that Dlx5 expression will increase in the PFS during the

period of expected suture fusion (days 21-45 postnatal) relative to that in the non-fusing

sagittal suture. We plan on testing this hypothesis by examining expression levels of

3

DlxS by RT-PCR and its localization by immunohistochemistry in the posterior frontal

and sagittal sutures from CD-lmice postnatal day 21, 2S, 3S, and 4S.

Aim 2. Examine the posterior frontal and sagittal sutures of 10 day-old

wild-type and heterozygous Dlx5 (+1-) mice.

Previous studies have shown that mice deficient in both alleles (DlxS -/-), die in

utero, therefore we plan on examining heterozygous DlxS mice. In order for PF

sutural fusion to occur, the bone fronts ofthe frontal bone must come in contact with

each other, which occurs prior to PFS sutural fusion (postnatal day 21-4S). Therefore,

our objective is to determine the extent of bone front spacing in the posterior frontal and

sagittal sutures of 10 day-old wild-type and heterozygous DlxS (+/-) mice. Our working

hypothesis is that there will be an increase in the space between bone fronts in the PFS

in heterozygous DlxS (+/-) mice.

BACKGROUND

A. Cranial Sutures

The skull vault is composed of paired frontal, parietal, squamosal portions of the

temporal bones, and the anterior portion of the occipital bone. (Couly et al., 1993; Jiang

et al., 2002; Noden, 1978, 1988). These bones form the calvaria and develop through

intramembranous ossification; whereas, the bones that form the base of the skull

develop through endochondral ossification (Morriss-Kay and Wilkie 200S).

4

In vertebrates, the skull develops from either neural crest, or mesodermal

tissues. Chai et al. and Jiang et al., incorporated markers in mice that enabled

visualization of the migration of cranial neural crest cells during development (Chai et

aI., 2000 and Jiang et al. 2000, 2002). These studies showed that neural crest cells

emigrate as three separate populations: the trigeminal crest, the hyoid crest, and the

vagal crest. The trigeminal crest migrates to the frontonasal and first branchial arch

regions and provides the neural crest component of the trigeminal ganglion; this is the

only popUlation of neural crest cells that contributes to the skull.

Cranial sutures separate the bones of the calvariae and form as the bony edges

meet. Growth at these sutures occurs perpendicular to the orientation of the suture;

normally, it coincides with the expansion ofthe brain (Morriss-Kay and Wilkie 2005).

At birth, the cranial sutures are patent and the flexibility that results allows the bones of

the cranium to overlap as the baby passes through the birth canal. Another significant

task of the cranial sutures is their role as the primary sites of bone growth during

craniofacial development. This is most notable during the period when the brain is

rapidly expanding (Baer, 1954).

The human metopic suture undergoes fusion during the first three years and all

of the other cranial sutures remain patent well into adulthood (Weinzweig et al., 2003).

In the mouse, the PFS, equivalent to the metopic or interfrontal suture in humans, fuses

in the first 45 days oflife. However, all of the other sutures, including the sagittal and

coronal, remain patent (Warren et at. 2001). Similarities between mice and humans in

craniofacial development, the specific molecules involved, as well as the structure of

those molecules make the murine posterior frontal suture an excellent model for the

5

study of human suture development and pathology (Richtsmeier et ai., 2000; Schneider

et al., 1999; Waterston et al., 2002; Wilkie and Morriss-Kay, 2001).

B.Dura

The presence of the dura mater has been found to be essential for normal

development and maintenance of cranial vault sutures (Opperman et aI., 2005). In fact,

the ability of neonatal dura mater to regenerate bone has been observed clinically and

demonstrated in several animal models (Lenton et ai., 2005). For example, it has been

shown that the presence of intact dura is required for regeneration of the skull vault

after removal of a small area of the calvaria in infants and young animals (Babler et at.,

1982; Mabbutt and Kokich, 1979; Mabbutt et al., 1979; Mossaz and Kokich, 1981).

These findings and the intimate association between the dura mater and crania] sutures,

initiated investigations by others, to determine a role for dura mater in suture

development and fate (Lenton et al. ,2005).

Opperman et al. transplanted embryonic and postnatal corona] sutures with and

without dura mater to the midparietal bones of syngeneic adult rats and showed that

both embryonic and postnatal suture explants fused in the absence of their original,

suture-associated dura mater. On the other hand, explants of both sutures transplanted

with their dura mater intact remained patent (Opperman et aI., 1993). Similar results

were obtained when the embryonic sutures were placed in an in vitro organ culture

system, with intact sutures remaining patent and those lacking their dura mater fusing

(Opperman et at., 1995). These results suggested that factors and/or physical signals

from the dura mater associated with the coronal suture are responsible in some way for

6

maintaining suture patency in the late embryonic and early postnatal stages of

development.

In subsequent studies, Opperman and colleagues added conditioned media from

dura mater cultures to coronal suture explants placed in organ culture without their

suture-associated dura mater. They showed that this conditioned media was capable of

preventing suture fusion. When the media was divided into heparin and non-heparin

binding fractions, only the heparin-binding was capable of maintaining coronal suture

patency (Opperman et ai., 1996). These results suggested that heparin-binding growth

factors, such as FGFs and Transforming Growth Factor-fJ (TGF-fJ) superfamily

members, were responsible for differences in coronal suture fate (Opperman, et al.,

1996).

The role of the dura mater in maintenance of suture patency has also been

demonstrated in the sagittal suture (Kim et al., 1998). Embryonic (EI6) murine sagittal

suture explants that were cultured without their dura mater, fused within three days; this

contrasted with control explants, cultured with their dura intact, that remained patent.

However, no difference was demonstrated in the patency of sagittal suture explants

from postnatal (PI) mice cultured with and without their associated dura mater (Kim et

al., 1998). This prompted investigators such as Kim et al. to study the expression of

genes known to act in conserved signaling pathways throughout development. Different

patterns of expression of these genes were found at the end of embryonic development

in the osteogenic fronts of the sagittal suture. This suggested that different molecular

mechanisms are responsible for suture morphogenesis during the embryonic and

postnatal stages of development and lead to the hypothesis that in the murine sagittal

7

suture, the dura mater is responsible for maintaining suture patency prior to birth, while

molecular signals from the osteogenic bone fronts are the principal regulators of suture

patency during the neonatal stages of development (Kim et al., 1998).

Other studies have focused on the effects of removal or disruption of the dura on

PFS fusion. Separating the dura mater from the overlying PFS either temporarily or by

placement of an impermeable silicone membrane, resulted in a significant delay in

fusion of the PFS and suggested that there are interactions between the PFS and the

dura during the process ofPFS fusion (Roth et al., 1996).

Verification that molecules originating in the dura mater were responsible for

coronal and sagittal suture patency, as well as PFS fusion, led to the hypothesis that the

dura mater has a specific "make -up", based upon it's location, and is therefore capable

of coordinating the differing fate of each cranial suture (Lenton et al., 2005). Levine et

al. (1998) tested this hypothesis by removing a portion of the calvariae, in rats,

containing both the posterior frontal and sagittal sutures without the dura mater and

rotated it 1800• This placed the dura that was previously associated with the PFS in

contact with the sagittal suture and vice versa. The new position resulted in obliteration

of the sagittal suture which normally remains patent and sustained patency of the

normally fused PFS in vivo.

The results of these experiments clearly indicate a role of the dura mater in

determining cranial suture fate.

C. Disturbances in Suture Fusion

For sutures to function as growth sites, they need to remain patent, and allow new

8

bone to be formed at the edges of the overlapping bone fronts. Too much or accelerated

bone growth results in osseous obliteration of the sutures; whereas too little or delayed

bone growth results in wide-open fontanels (Opperman 2000).

1. Craniosynostosis, or premature fusion of the cranial sutures, results in an

abnormal skull shape, and may cause severe deformities such as blindness and mental

retardation (McCarthy et at., 1989). Although the majority of craniosynostoses are

nonsyndromic, more than 100 different syndromes have been identified that include

craniosynostosis as one of their many characteristics (Wilkie, 1997). Because

craniosynostosis is such a common finding among syndromes and may result in severe

deformities, it is important to study the mechanisms and roles of the molecules involved

in cranial suture fusion in an effort to develop therapies to treat the disorder.

In the last ten years, some of the more common syndromes of these have been

linked to dominant mutations in genes encoding Fibroblast Growth Factor Receptors

(FGFR1, FGFR2, and FGFR3), and the transcription factors TWIST and Msx-2 (Bellus

et at., 1996; EI Ghouzzi et at., 1997; Howard et at., 1997; Jabs et at., 1994, 1993;

Oldridge et at., 1995; Reardon and Winter, 1994). Mutations ofFGFRs and Msx-2

genes are usually described as gain-of-function mutations whereas TWIST mutations

are known as loss-of-function mutations. (Opperman et at.,2000). Mutations in these

genes and transcription factors account for about 25% of all cases of craniosynostosis

(Twigg et at., 2004).

Msx-2 (muscle segment homeobox 2) is a highly conserved homeobox gene that

is involved in the regulation of inductive tissue interactions during embryogenesis (Jabs

et at., 1993). A mutation ofthe Msx-2 gene was the first to be identified among gene

9

mutations associated with craniosynostosis (Jabs et al., 1993). Boston-type

craniosynostosis is the only syndrome with craniosynostosis that is associated with

mutations in the gene for Msx-2 (Jabs et al., 1993). The specific mutation associated

with Boston-type craniosynostosis is a cysteine to adenosine transversion on the gene

for Msx-2 that leads to an amino acid substitution. (Jabs et al., 1993), this results in an

increase in DNA-binding ofthe Msx-2 protein and augmentation of the normal function

ofMsx-2 (Ma et al., 1996).

TWIST is a transcription factor that was originally identified as a principal

stimulator of mesoderm formation in Drosophila (Lenton et al., 2005). Mutations in the

gene encoding TWIST are also associated with craniosynostosis among many other

abnormalities. Heterozygous TWIST-l mutations have been identified in 80% of

patients with Saethre-Chotzen syndrome (Chun et al., 2002; Johnson et al., 1998).

Some of the many abnormalities found in patients diagnosed with this disorder are,

craniosynostosis involving one or both coronal sutures, midface hypoplasia, and facial

asymmetry (El Ghouzzi et al., 1997; Johnson et al., 1998; Reardon and Winter, 1994).

In contrast to the Msx-2 gain-of-function or overexpression mutation, the majority of

TWIST -1 mutations resulting in craniosynostosis, produce haploinsufficiency and result

in a lack of functional protein production (Paznekas et al., 1998; El Ghouzzi et al.,

1997).

Heterozygous FGFR mutations are the most common genetic abnormality

associated with syndromes including craniosynostosis in their phenotype (Lenton et al.,

2005). The most frequently occurring mutations are missense, or small in-frame

insertions or deletions that affect the ligand-binding domain (Wilkie, 1997). Other

10

mutations involve the tyrosine kinase domain of the isofonn FGFR2 and are thought to

have an impact on downstream signaling pathways (Kan et a/., 2002). However, most

mutations of this gene are dominant and are believed to result in a gain of function by

either increasing ligand binding affinity, decreasing the specificity of receptor-ligand

interactions, or increased receptor dimerization and stabilization (Lenton et a/., 2005).

Most of the FGFR mutations that cause craniosynostotic syndromes are

mutations of the FGFR2 isoform. In fact, two mutations that result in an increase in

function account for almost all of the cases of Apert syndrome (Lenton et a/., 2005). In

addition, other mutations of the FGFR2 gene have been associated with Pfeiffer and

Crouzon syndromes. An analogous mutation, to one found in the gene encoding FGFR2

that causes Apert syndrome, has been identified in the FGFRI gene, and is responsible

for some cases of Pfeiffer syndrome (Robin et a/., 1994; Wilkie et a/., 1995). Two

mutations in the FGFR3 gene have been linked to Muenke syndrome and type I

thanatophoric dysplasia (Bellus et aI., 1996; Muenke et a/., 1997, 1994; Tavonnina et

a/., 1995).

2. Enlarged Parietal Foramina

Too little sutural growth results in a disease known as Enlarged Parietal

Foramina. Loss-of-function mutations in the transcription factors Alx-4 and Msx-2 have

been associated with this disorder among many others in development.

Alx-4 and Msx-2 encode homeodomain transcription factors known to have

regulatory roles in the development of various parts of the body in vertebrates (Alappat

et a/., 1999; Meijlink et a/., 1999). Studies of Alx-4 and Msx-2 knock out mice have

displayed the importance of their roles in skeletal patterning, differentiation, and

11

growth. These mice had craniofacial, axial, and appendicular abnormalities among their

many defects (Antonopoulou et al., 2004; Qu et al., 1997; Satokata et al., 2000).

In humans, heterozygous loss-of-function mutations of either of these genes may

result in defects of the skull vault in the form of enlarged parietal foramina (PFM)

(Mavrogiannis et al., 2006). Typical PFM are bilateral oval or round openings in the

parietal bones on either side ofthe posterior frontal suture (O'Rhailly et al., 1952).

They are easily distinguished from the normally present minute foramina that blood

vessels run through (Tubbs et al., 2003), and are a result of delayed ossification during

rapid calvarial expansion (Antonopoulou et al., 2004; Satokata et al., 2000). Frequently,

PFM can be traced back to a defect in infancy that extends between the anterior and

posterior fontanelles; when this persists, it is termed cranium bifidum (CB)

(Mavrogiannis et al., 2006). A bridge of bone usually develops along the midline, that

creates two separate foramina, which usually decrease in size with age through

circumferential bone growth Eventually, complete closure may occur, leading to clinical

nonpenetrance (Mavrogiannis et al., 2006).

D. Members of the TGF-beta superfamily and the regulation of cranial sutural

fusion

The TGF-~ superfamily includes several molecules that are involved in

signaling and have vital roles in various processes throughout development, such as,

embryogenesis, tissue formation, and wound repair (Gold et al., 1997; Kingsley, 1994).

Some members of this family that have been found to be involved in cranial suture

12

development are the TGF-~ isoforms (1, 2 and 3), and Bone Morphogenetic Proteins

(BMPs).

l.BMP

Previous studies have shown, that BMP4 is expressed in the sagittal suture

mesenchyme and dura mater of embryonic mice (Kim et at., 1998). Further study of

postnatal sutures showed that BMP4 protein is localized to the suture mesenchyme and

osteogenic fronts in fusing posterior frontal and patent sagittal and coronal sutures

(Warren et at., 2003).

Since BMPs have been found in patent and fusing cranial sutures, Warren et al.

screened for BMP antagonist mRNAs before, during, and after the period of expected

suture fusion. They discovered the presence of Noggin in the patent sutures in all three

of these periods and almost no Noggin expression in the fusing PFS as early as the time

period prior to suture fusion (15d) (Warren et at., 2003).

Takao et aI., was the first to demonstrate the function of Noggin as a BMP

antagonist in cultured murine bone marrow stromal cells (Takao et aI., 1996). Noggin

protein was found to bind BMP4 with high affinity and inhibit BMP4 activity by

blocking its ability to bind to cognate cell-surface receptors (Zimmerman et at., 1996).

Noggin was originally described in the Xenopus and was found to playa role in

normal dorsal development (Smith and Harland, 1992). Further studies by Brunet et at.,

found that Noggin induces lateral mesodermal tissues to form more dorsal elements

including muscle, heart, and the pronephros. In addition to these roles, they also found

that Noggin was able to rescue dorsal patterning in ventralized embryos (Brunet et at.,

1998).

13

Many of the roles identified for Noggin, oppose those known for BMPs, for

example, the role of BMP to rival ventralizing potential in a developing embryo. As a

result of these studies, recent work has looked further into the roles of Noggin and

suggested that it might play an important part in maintaining suture patency (Lenton et

al., 2005).

Recently, the BMP antagonist Noggin has been linked with FGF2 activity and

determination of suture fate (Warren et al., 2003). Only the dura of the PFS expresses

FGF2 mRNA and protein in vivo (Greenwald et al., 2000) and produces high levels of

FGF2 in culture. FGF2 disrupts Noggin induction in a dose-dependent fashion,

suggesting that environments with a high FGF2 concentration, such as the PFS, reduce

Noggin expression and enable suture fusion (Warren et al., 2003).

Further evidence which suggests that Noggin may be responsible in determining

suture fate is the 3 day old CD-I mice injected with a noggin adenovirus that had

widely patent posterior frontal sutures after 50 days (Warren et al., 2003).

E. Transcription factors associated with BMP-Signaling

1. DIx Genes

Dlx homeobox containing genes are mammalian homo logs of the Drosophila

Distal-less (Dll) gene. They are a group of six genes that are organized into pairs,

specifically, Dlx 112, Dlx 5/6, and Dlx 317 (Robledo et al., 2002). Mammalian Dlx

genes are expressed in the primitive craniofacial region, the developing brain, and

limbs, as well as, the apical ectodermal ridge (Robledo et aI., 2002). Dlx gene products

have a conserved 60 amino acid homeodomain and are members of a large family of

14

DNA-binding transcription factors. They are believed to control the transcription of

genes that affect organogenesis for skeletal fonnation (Miyama et al. 1999).

Dlx5 has been coined by some to be a new candidate responsible for skull

fonnation (Depew et al., 1999; Acampora et al., 1999). Compared with their other

family members, Dlx5 and Dlx6 are uniquely localized to all skeletal tissues (Holleville

et aI., 2003). In fact, Acampora et al., found Dlx5 and Dlx6 in immature chondrocytes

and developing skull bones of embryonic mice and rats and in the periosteum of more

mature bones (Acampora et al., 1999).

Results from several in vivo studies have provided evidence that indicates the

involvement of the Dix family of hom eo domain proteins in the regulation of osteoblast

differentiation and bone tissue-specific gene expression (Miyama et ai., 1999). Ryoo et

al., showed that among the Dlx genes, Dlx5 is a bone-inducing transcription factor that

is expressed in fetal rat calvarial cells in the later stages of osteoblast differentiation

(Ryoo et al., 1997). In addition, it has been shown, that forced expression of Dlx5 in

cultured cells leads to osteocalcin expression and a fully mineralized matrix (Lee et al.,

2003, Miyama et al., 1999, and Tadic et al., 2002). These results strongly suggest that

Dlx5 plays an important role in the development of mineralized tissues.

In an investigation of potential functional roles for Dlx5, Miyama et aI., isolated

Dlx5 as a BMP-inducible gene by differential display technique. Subsequently, they

demonstrated that mDlx5-transfected cells show a more mature osteoblast phenotype

when compared to parental MC3T3-El cells, suggesting that mDlx5 plays a role in

osteogenesis (Miyama et al., 1999). In the study, mDlx5-transfected cells showed higher

alkaline phosphatase activities, osteocalcin production, and mineralization of

15

extracellular matrix, suggesting that mDlx5 is a positive regulator of osteoblast

differentiation.

In addition, when the time course ofmDlx5 mRNA induction ofMC3T3-El

cells was examined, transcripts of mDlx5 continuously occurred 48h past the time of

BMP treatment. It was proposed from these findings, that the mDlx5 gene product plays

an important role in the later stage of the differentiated osteoblast since osteoblast fate

has been clearly determined at this time (Miyama et at, 1999).

Recent investigations, showed that Runx2, a transcription factor that plays an

essential role in osteoblast differentiation and bone mineralization (Lee et al., 2005), as

well as alkaline phosphatase (ALP) expression, was induced, in the absence ofBMP-2

with forced expression ofDlx5 (Kim et al., 2004). Another fmding was that blocking of

Dlx5 completely inhibited BMP2-induced stimulation ofRunx2 and ALP expression.

These results proved, that Dlx5, is required as an upstream regulator of the expression

of Runx2 that is generated by BMP signaling and suggested that Dlx5 enhances ALP

expression by increasing Runx2 expression (Kim et al., 2004, and Holleville et

al., 2007).

Results of these studies have shown that Dlx5 is essential for BMP-mediated

osteoblast differentiation and have suggested that Dlx5 plays a role in the mature stages

of mineralization in skeletal tissue. These, and previous findings linking BMP signaling

to murine PFS fusion, in addition to its expression in developing (Kim et al., 1998) and

adult cranial sutures (Warren et al., 2003), has lead us to believe that Dlx5 may play an

important role in cranial suture fusion.

16

Chapter 2

Wadhwa S, Bi Y, Ortiz A, Embree M, Kilts T, 10220 R, Opperman L, Young M.

Impaired posterior frontal suture fusion in the biglycan/decorin double deficient mice.

Bone 2007; 40:861-866

17

BONE ELSEVIER Bone 40 (2007) 861- 866

www.elsevier.cof1ll1ocaterbone

Impaired posterior frontal sutural fusion in the biglycanldecorin double deficient mice

Sunil Wadhwa a .b ,*, Yanming Bi b, Ana T. Ortiz a, Mildred C. Embree b, Tina Kilts b,

Renato 10zzo c, Lynne A. Opperman d, Marian F. Young b

a Division o/Orthodontics, School qfDellfCl[ Medicine. Departmenl ojCl'lJl1iojacial Sciellces. 'University qfCOnlleClicul Health Cenlel; FarmillglOll , CT 06030. USA b Molecular Biology of BOlles and Teelh Section. O"(1I1itlfaciai and Skeletal Diseose., Brallch. NIDeR. NIH. DHHS Belhesda. MD 20891. USA

C Departments of Pal/wiog" ami A,ulIomy and Cell Biology, Jefferson M&licaJ College, Thomas Jefferson UlliversjtJ~ Phi/mlelphia, PA /9107, USA d Center/or Cra1l;o}acia/ Research (lnd Diagnosis. BaylOf' College of Dentist,)!. Texas A&Jt.f University System Health Science Cell tel: Dallas. TX 75246. USA

Received I Febru.11)' 2006; revised J3 October 2006; accepted 3 November 2006 Available online 26 December 2006

Abstract

Biglyc.n (Bgn) and decorin (Ocn) arc highly expressed in numerous tissues in the craniofacial complex. However, their expression and function in the cranial sutures are unknown. In order to study this, we first examined Ihe expression ofbiglycan and deeOl;n in the posterior irontal suture (PFS). whicb predictably fuses between 21 and 45 days post-natal and in the non-fusing sagi1tal (S) suture from wild-type (WI) mice. Our data showed tbal Bgn and Den were expressed in bolh cranial sutures. We Lhen characterized the cranial suture phenotype in Bgn deficient, DCIl

deficient, BgllIDcl1 double deficient, and Wt mice. At embryonic day 18.5, alizarin red/ale ian blue stain ing showed !bat !be Bgn/DclI double deficient mice bad hypomineralization of the trontal and patietal cmniofacial bOlles. Histological analysis of adult mice (45- 60 days post,natal) sbowed that the Bgn or Den deficient mice had no cranial sutUl'e abnormalities and immunohistochemistry staining showed incrc.1scd production of Den in the PFS from Bgll deficient mice. To test possible compensation of Den in the Bgn de1ieient sutures, lVe examined the BgllIDcll double deficient mice and found that they had impaired fusion of the PFS. Semi-quantitative RT-PCR analysis of RNA from 35 day-old mice revealed increased expression of Bmp-4 and Dlx-5 in the PFS compared to their non-fusing S suture in Wt tissues and decreased expression of Dlx-5 in both PF and S sutures in the Bgll/DclI double deficient mice compared to the \Vt mice. Fai lure of PFS fusion and hypomineralization of the calvaria in the BgnlDcn double deficient mice demonstrates that these extracellular matrix proteoglyeans could have a role in controlling the formation and growth of the cranial vault. ~";l 2006 Elsevier Inc. All rights reserved.

Keyword.,: Small proteoglycans: Biglycan; Decorin; Crania l suture; Mouse

Introductiou

Biglycan (Bgn) and decorin (Dcn) are members of the small leucine repeat proteog1ycan family (SLRP). Members of this family are characterized by a small protein core, which consists predominantly of leucine,rich repeaL~. There arc 13 known members of this family that are divided into three classes depending on their genomic organization and the similarity of their amino acid sequences. Bgn and Den belong to the class I

• Corresponding author. Division of Orthodontics, School of Dent. I Medicine, UCHC, Fannington, CT 06030, USA.

E'Jnuil address: Wadhwa@uchc.edu (S. Wadhwa).

8756-3282/$ , see front matter to 2006 El.evier Inc. All rights reserved. doi: I 0.1 OJ 6ij.bone.2006.11 .003

18

type SLRP and are highly expressed in skeletal and connective tissues [1 , 12, 14, 15.31]. The exact function of both Bgn and Den is unknown; however, both can bind and modulate members of the TGF-beta superfamily [II].

Bgll deficient mice have impaired post,nalal bone fonllation which may be due to the ability of Bgn to modulate the actions of Bone morphogenic proteins,214 (B1l1p,2f4) in osteoblastic cells, potentially contributing to the failure to achieve peak bone mass and early onset of osteoporosis [4,36]. Den deficient mice do !lot have obvious skeletal defects, but instead, have skin fragility which is hypothesized to be due to tile ability ofDcn to modulate collagen fibril structure and intcgrity [7]. Because of their similarity ill structure and tlleir overlapping expression in skeletal and connective tissues, it is probable that Bgll and Den

862 S, Wadhwa et ai, i Bone 40 (2007) 861-866

fimction redundantly. Support of this concept comes from the analysis of Bgl/IDcl/ double deficient mice that have a more severe phenotype in both long bone and skin [25] compared to WI or singly deficient SLRP mice.

The purpose of this study was to examine the role orBgu and Den in the craniofacial complex. In the craniofacial region, Bgn and Den are present in teeth [22,28], periodontal tissues [21]. nasal cartilage [29], eye [8]. temporomandibular joint [18], the developing mandible [16,35]. and tbe developing calvaria [35]. Despite their prevalence in numerous tissues of the craniofacial complex, little is knowl] about their function in these tissues. In order to examine this furlher and to eliminate potential compensation of Den with Bgn, we examined whether a craniofacial phenotype could be unmasked by creating Bgl// Den double deficient mice and comparing thera to the single Bgn deficient, single Dell deficient, and control Wt miee.

There is similarity and conservation in the molecular pathways thai control the assembly and growth of the cranial vault between humans and mice [33]. In mice. all the cranial vault sutures remain opcn throughout the life span ofthe animal, except for the posterior frontal suture (PFS), which fuses in a predictable fashion between 21 and 45 days post-natal (25]. In hltnlanS, the equivalent metopic suture fuses within the first 3 years post-natal [30,34]. Therefore. the mouse PFS serves as a model of post-natal human sutuml growth [33]. Bmp-2f4 signaling has been linked to murine PFS fusion by Warren el al.. who showed that inhibition of the BnIP-2!4 signaling pathway by adenovirus mis-expression of the Bmp antagonist Noggill blocks mllfine PFS fusion in vivo [32]. Since both biglycan and deeorin can modulate Bmp-2!4 signaling, we hypothesized that. in their absence, there would be defective Bmp-2/4 signaling and impaired PF sutlU1l1 fusion.

In this study, we found that Bgn and Den were expressed in the fusing PFS and in the lion-fusing S suture in Wt mice. The PFS in hoth Den deficient and Bgn deficient mice revealed no defects. However, further examination of the PFS in the Bgn/ Dell double deficient mice showed a lack of PFS fusion ill a mechanism that is potentially related to a decreased expression of Dlx-5.

Experimental procedures

GeJ1(ffation of Bgn and Dell single and double deficient mice

All experiments were performed under an instilulionaUy approved protoC{)] for the use of animals in research (IiNIDCR-lRf'·98-0S8 and 01-151). Mice deficient in Bgn and Den were generated by gene targeting in embryonic stem cells as de=ibed previously [7.36]. Heterozygous 8gn!Dcn deficient mice were produced by breeding a homozygous Bgn deficient female (B15Il-I-IDcn+I+) with an Den deficient male (Bgll+I/iIDcIl+/-); deficient males are since the Bgn gene is located chromo-some and absent from the Y chromosome, F2 BgnlDen double deficient mice were obtained by interbreeding Fl heterozygnus BgnJDeI/ deficient animals, In order to maximize the number ofrlouble deficient mice obtained, the following breeding scheme was tilled: Bgll O/-IDw+/- males were bred to BI,,'ll +/-IDcn+/­females, The crauial vault "fthe mice was analyzed at embryonic day 1&.5. and post-natal days 35, 42, and 60, We chose to examine 42 days and 60 days post­natal (Ill. posterior frontal SUIllre normally fuses hetween 21 and 4S days post­natal) hecause they represent time points during and well past normal PFS closure which allowed us to evaluate ifthere was a delay or failure "fPFS fu,im>

19

in the transgenic mice. We selected post-natal day 35 for our RT-l'CR analysis because we wanted to evaluate whether there would be differences in gene ""pres.sion during the time of normal posterior frontal sutural fusion.

Gellotyping

All mice were genotyped for Bgn and De" alleles by PCR aualysis as described [5], PeR products were resolved by electrophoresis through 1.8"1. agarose gels. yielding bands of 212 bp for the WI Bgll allele, 310 bp for the disrupted B15I1 allele. J61 bp for the WI Dell allele, and 2.~8 bp for the disrupted Dell allele,

Alizarin red and alcUm bille staining of mice littermate.,

In our experimeJ1lal "I'Pmach, we used a breeding scheme that would maximize the number of BgnJDcn double deficient mice, and lOr this reason, we were ullable to obtain a humo:zygous WI mouse and BgllIDcn double dell· cient mouse in the same liner. Therefore, hI this first part of the study. We used (Bgn+!-iDcn+/+) mice as our controls. 18.5 d.r.c. linenn.les obtained from 5 litters were euthanized, and then their skin, muscle, and fat were =ved. The animals were fixed in 100% ethanol for 4 days and then placed in _cerone for 3 days. Mice were tben stained with alizarin red (0,09%) and aleian blue (0.05%) in a solution com.ioing ethanol, glacial acetic acid, and water (67:5:28) for 3 days. After staining, mouse samples were transferred to 1% KOH until their soft ,issues were dissolved and then preserved in a 100% gtycerol [23].

FuxiJroll

Sixty day.old mice were divided inlo four groups. WI. Bgll defICient, Den deficient and Bgn/Dc'11 double deficient, "ith 3n 11 ~ 3 for each group, The mice were sacrificed and the soft tl",ues on their heads were removed. The calvaria were radiographed using a Faxitron MX-20 Specimen Radiography System (Faxitron X"ay ('.orp., Wheeling. II... USA) al an energy of90 kV for 20 s. The images were captured with Eastman Kodak ('.0. X-OMAT TL (Eastroan Kodak ('0., Rochester, NY, USA).

In order to detennine the optimal length of a rectangular box that could be used to quantitate the amount of radio-opacity in the PFS. the length of the PFS from 60 day-old WI, Bg/l deficient, Den deficient, and Bgn!De/T double deficient mice wu, firsl measured. The mean length±the standard elTor of the PFS was I 1.0",0,6 mm forWt, IOJ±O.3 mm from BglI deficient. JO.7±1.2 mm fromD"/t deficient, and 11.1±0.3 mm foc Bb~liDc'11 double deficient mice. Evaluation of diffurences among the memlS by analysis of variance (ANOVA) reve-aled that there were no significant differences in the len gth of the PFS between the groups. Therefore. a 10 mm by 3 mm rectangular box was used to measure the amount of radio-opacity in the PFS from the different gmups, The amount of radio-upacity in the PFS was quantitated by using Image J NIH IMAGE.

Histology alld immunohistochemistry

WI. Dell deficient, Bgn deficient, and Bg/riD!" douhle deficient ntice were sacrificed by CO2 inhalation at 35-45 days al\er birth, Calvaria were removed and fixed for 2 weeks at room temperature in 10% IOmllllin. After being wasbed in water for 5 min, they were decalcified in ronnie acid bone decalcifier solution ([mmunocal from Decal Corporation, Tallman. NY, USA) for 4 weeks. Specimens "' ..... II,en washed in water for 5 min and fixed for 3 days in butTered zinc fonn.lin (Z-fiJ( from Anatech Ltd., Battle Creek. MI, USA) before being classically processed for histology and sectioned coronally. Section. were stained withH&E.

Tissue sections were deparaff'mized with xylene. Following rehydration with decreasing ooncenlrolionsof ethanol. the sections wefI'lreated with 3% peroxide in methanol fur20min loblockendogenous peroxidase activity. Iuordertoexposethe antigen, section. were predigested in chondroitina", ABC (cat# KEOI502, SeikagakuCorp, Tokyo, J."an) ata concentration of 0.015 unilslmlrorl bOI37"C. Incubating the sections in 10% gnat serum fur 30 min reduced non-specific binding, Sections were then incubated overnight at 4 0(' with primary antibody at a 1:200 dilution. Den and Bgn antibodies (respectively LF 113 and LF 106) were kind gift. from Larry Filiher(NIOCR. NIH~ Biotinylated rabbil anti-goat secondary antibody

S. Wadlnva e/ at. / BOlle 40 (2007i 86J~66 863

was used and visualized by a slrept.vidin- peroxidase solution io !he presence of DAB chromagen. The negative control consisted of the above mentioned procedures except for the substitution of the primary antibody with rabbitlgE.

The data presented in this paper were reproduced in at least 6 different sections from two separate animals for each group. Serial sections through the whole suture were obtained for each animal and observations were confirmed in different interspaced serial sections chosen to coverlhe whole suture. In this way. it was ensured that the reported observations were genuine and not local random abnonnalities. ln each case, data from a single representative experiment are shown.

isolalioll oflhe sulu,.e complexfor mRNA extraction

Posterior frontal (PF) and sagittal (S) suture complexes (including the associated dura mater, suture mesenchyme~ and osteogeJJic fronl<;) were harvested from 35 day-old male Wt (11 = 4) and BglI/Dc/l (11 = 2) double deficient mice. Mice were eutb3Dized with carbon dioxide, and the posterior frontal and sagittal suture complexes were isolated under a dissecting microscope as previously described [241.

Semi-qual1litalive RT-PCR

RNA was extracted from the cranial suture material using Trizo.l·Reagent (Molecular Research Center. Cincinnati, OH) following the manufacture's instructions. One microgram total RNA from the sample preparation was reverse transcribed with 50 units of SuperScript II RT usillg random hexamer primers (lnVitrogeo Life Technology, Carlsbad, CAl following the manufacturer' s instructions. The primers used for amplification were designed with Primer 3 software (http://www-genome.wi.mit.eclu/cgi-binlprimeriprimer3.cgi). The pri­mers used for RT-PCR were for Gapdh 5'gagaggccetatcccaact3' (forward) 5' gtgggtgcagcgaacUtat3'(reverse), for Bmp-4 5'ggtgggtttgcttccactta3' (forward) and 5'atgggagcgagttgtgtacc3' (reverse), for DL<-5 5'ccgggacgctttaltagatg3' (forward) and 5'tggacactatcaatggtgcc3' (reverse), for biglycan 5'acctgtcccct­tecatctct3' (fonvard) and 5 'ccgtgtgtgtgtgtgtgtgt3' (reverse), for decorin 5 ' ccoac· ataactgcgatccct3'(forward) and 5'tgtccaagtggagucccte3'(reverse), and for l8s 5' catgtggtgtgaggaaa3'(fOlward) and 5'gcccagagactcatttctt3'(reverse). J'CR was performed using a hot start (with Taq Gold™, Applied Biosystems) at 95 'c for 10 min followed by 45 cycles (biglycan. decorin, 18s. and GAPDH) or 50 cycles (Dlx-5 and Bmp-4) of I on," at 94 "C,20sat 57 'Caod 30s at 72 'C followed by a 7 min extension at 70 "C. The reaction was chilled to 4 'C afterwards until it was analyzed. Sixteen microliters of the PeR product and a I OO-bp DNA ladder (Gibco BRL) were run in a 10% acrylamide gel (THE) buffer at 100 V. The products separated by electrophoresis were visualized after ethidiuon bromide staining under a UV light aod photographed in a single field of view with a digital camera. IndiVIdual bands on digital micrographs of RT-PCR products were analyzed by densitometer using ImageJ NlH IMAGE Intensity for Dl,·.5, Bgn. Den. and Bmp-4 was calculated as a percentage of densitometry values for control bands of GapdJr or 18s. Each primer set was tested previously to assure that amplification was in the linear range by analyzing amplification af several different cycle numbers .

Statistical analysi"

Statistical significance of differences among means was detennined by analysis of vanance (ANOVA) wid! post hoc comparison of more than two means by the Bonferroni method using GraphPad Prism (San Diego, CAl.

Results

Alizarin red- alciall blue staining oj E 18.5 /itlermates

Alizarin red-alcian blue staining ofE 18.5 control (Bgn t / !

DCI/+ it) was used as the control due to the inability to generate a Wt and a BgnlDcn double deficient in the same litter) and BgnlDcn double deficient littcnnales revealed that the Bgnl

20

DCII double deficient embryos had severe hypominera1ization of the Ji'ontal and parietal bones (Figs. I A-D). Alizarin red and alcian blue staining were also performed on two additional BglI/ Dell double deficient embryos along with two littermate controls revealing similar results (data not shown).

Expression of Bgn alld Den ill the PFS aJ/d sagiJta/ su/ure /iom WI mice

A time course for the expression of Bgn and Den was determined in the PF and the S suhtre in WI mice at day 21, day 25, and day 35 post-natally by semi-quantitative RT-PCR. The data indicated that mRNA encoding both biglycan and decorin is abundantly expressed in both the posterior frontal and sagittal suture during maruration (Fig. 2A). Since there appeared to be increased expression of BglI mRNA in the PF compan:d to the S sutnre during the late phases of PFS fusion. we repeated Ihe cxpelimCllts two more times at day 35 post-natal. In the three independent experiments from 35 day-old Wt mice, we found that the ratio of Bgn mRNA expression in the PFS/S suture was l.l ±O.3.

[n order to dctennine the precise localization of Bgn and Den protein expression in the PFS and to evaluate possible compensation at the protein level, immtUlohistochemistry for Bgn and Den was pelfollUed in 42 day-old Wt and Bgo deficient mice (Figs. 2B- E). [n the PFS from the Wt and Bgn deficient mice, Den was found in the underlying dura (Figs. 2e, E). [n the Wt mice, Dcn appeared to have a more restricted

A B Control BgnlDcn DKO

c D f,ontal

~ fr~ parietal

&I'

Control 8gnlDcn DKO

Fig. I. Alizarin red/a leian blue 'taining of E18.5 littermates. (A. C) Control BgII+I-/DcII+/+ mice. (8, D) Bb~';Dm double deficient mice.

864 S. Wadlnva e/ al. I BOlle 40 (J()()7) 861-866

A

Bgn

Den

188

21 d 25d 35 d

PF S PF S PF S

-- .......- --

~ .. --.. -

Fig. 2. Expression ofBgn and Den in cranial sutures. (A) Semi-quanlilative RT­PCR analysis perfonned [or DCIl and Bgn on RNA extracted from the sagittal (S) suture and Ihe posterior frontal suture (PFS) from 21,25. and 35 d.y-old wild-type (Wt) mice. (B-E) Immunohistochemistry (brown) for biglycanlB. D) and decorin (C, E) on coronal sections of the posterior frontal suture from 42 day-old Wt (B, C) and Bgll deficient mice (0. E). Sections were counterstained with hematoxylin (blue).

expression pattem in the dura compared to BglI deficient mice where it was found at a higher level and Ihroughout the structure. In contrast, the PFS from the Dcn deficient mice had no change in the localization of 8gn in the underlying dura compared to Wt mice (data not shown). Negative controls consisting of the substitution of the primary antibody with rabbit IgE revealed no visible staining (data not shown).

Ana(vsis of the PFS jivm Bgn dejicient. Dcn deficient, BgIIIDcf/ Double deficient and WI mice

Tn order to detennine what role 8gn and Den deficiency had on latcr developmental stages. 60 day-old Wt and Bgnl

Dcn double deficient mice were examined. Faxitron analysis showed that the Bgn/Dcn double deficient mice had a clear absence of radio-opacity in the PFS. On the other hand. calvaria obtained from 60 day-old Wt, Bgn deficient. and Dcn deficieut mice, all had PFS that was radio-opaque (data not sho\>,'I\). Quantification of the amount of radio-opacity in the PFS by X-ray scanning and NIH-image analysis revealed that it was significantly decreased in BgIIIDcll double deficient mice compared to the other groups (p< 0.001) (Fig. 3A). Besides a decrease in mineralization of the PFS. the decrease in radio-opacity in the PFS from tlle BgIIIDclI double deficient could also be due to a difference ill the structure and curvature of the calvaria. To test this possibility. coronal histological sections of the PFS from 45 day-old Bgn deficient (Fig. 3C), DCIl deficieut (Fig. 3D), Bgn/Dcn double deficient (Fig. 3E), and Wt mice (Fig. JB) were examined. This analysis revealed PFS patency only in the BgnlDcn double deficient samples compared to the other groups.

A 35000

! 30000

j 25000

" ~ 211000

! 15000

I! 10000

c! 6000

wt ego KO Dcn KO BgnlDcn DKO

B c Suture

/

E

Bone

Fig. 3. Analysis of radio-opacity and histology of the PFS from Wt. BglI

deficien~ DCII deficien~ and Bgl//Dcl/ double deficient mice. (A) Quantifica­tion of intensity of radio-opacity in lbe PFS from faxitron images of the calvaria from 60 day·old WI. Bgl/ deficient. Dw deficient, and Bgll/ Dcl/ double deficient mice. Each point is the mean and SEM for 11 = 3 for each group. 'Significant decrease in the intensity of radi<Hlpacity in the PFS from the BgnlDcl/ double deficient mice compared to the other groups. p > O.OOI. (B-E) H&E staining of corona l sections of the posterior frontal suture from 45 day-old Wt IB), BgJl deficient (C). Dw deficient (0), and Bgl/lDcl/ double deficient (E) mice.

21

S. WctdhwlI e/ al. / BUlle 40 (2007) 861-866 865

Bmp-4

DIx-5

Gapdh

WT DKO

s PF s PF

.......... - ... - - .

.02 .55

.53 1.15

.19 o

.9 BmplGapdh .15 DlxlGapdh

Fig. 4. Semi·quantitative RT-PCR perfonned for Bmp-4 and Dlx-5 on RNA extracted from the sagiu.1 (S) suture and the posterior frontal suture (PFS) limn 35 day-old wild-type (Wt) and BgJlIDclI double deficient mice. RT-PCR of Gapdh served as a Ilonnalizing control.

£>;pressioll ofBmp-4 alld Dlx-5 mRNA in PFS alld sagittal suture from 35 day-old WI alld BgnlDcn double deficient mice

Because Bmp-2/4 signaling has been previollsly implicated in controlling PFS fusion [32], we examined the relative expression levels ofBmp-4 mRNA, and the Bmp-4 target gene, Dlx-5 mRNA [10], in our mouse model. Messenger RNA was extracted from the fusing PFS and non-fusing S sutures of 35 day-old Wt and in BgnlDcl1 double deficient and analyzed by semi-quantitative RT-PCR These analyses showed that Bmp-4 mRNA levels were higher in the (fusing) PFS compared to S suture in both genotypcs. Intcrestingly, whcn Dlx-5 was measured in the same samples, it was significantly highcr in the Wt PFS compared to the S suturc but absent in both Sand PF sutures in BgnlDcn deficient mice (Fig. 4). Similar results were obtained in a second independently repeated experiment (data not shown).

Discussion

This report shows that both biglycan (Bgn) and decorin (Dcn) are abundantly expressed in the cranial sutures of mice, with overlapping yet distinct pattems of expression in the fusing suture and the dura maIer. We found that BglllDc1/ double deficient mice have open sutures and severe hypomineralization of both frontal and parietal bones as early as E 18.5, while mice singly deficient in Bgll or DCII have no suture fonnation defects.

Murine PFS fusion is controlled by a complex interaction between many different types of cells, tissue types, and growth factors. For example, Warren et a!. [32] has proposed that Fgf-2 secreted by the underlying dura causes a downregulation of Noggin within the posterior frontal sutural cells that allows Bmp-214 to interact with the bone fronts to cause fil~ion of the PFS. For such orchestration of events to occur. therc must be control of the localization and activity of the secreted growth factors. Proteoglycans that could modulate the activity of growth factors within thc extracellular matrix of the posterior frontal suture may providc this control [26].

Scverallines of evidence link both Bgn and DCll to Bmp-2/4 activity. III a recent study, we found that Bgll deficient calvarial osteoblast-like cells had less ccll surface Bmp-4 binding, which led to reduced Bmp4 mediated osteoblast differentiation in Bgl/ deficient calvarial osteoblast-like cells compared to Wt

controls [4]. Similar results have been noted in MC3T3-EI cell­derived clones expressing higher or lower levels of biglYC3n [27]. Den has been shown to modulate the actions ofBmp-2/4. Specifically Bmp-2 induction of alkaline phosphatase was diminishcd in DCII deficicnt myoblast cells compared to Wt controls [9].

Bgo and Dcn are similar in structure and function and it is possible that they function redundantly. Immunohistochemistry revealed tllat, in the absence of Bgn, Den is upregulatcd throughout the PFS. Dcn expression was restricted to the interior portion of the dura in Wt mice but more evenly distributed throughout thc underlying dura of the PFS in Bgl/ deficient mice. We have also previously reported increased expression of Den in Bgn deficient osteoblastic cell cultures [4]. Furthermore we found that the ability of Bmp-2 to induce mineralization, as measured by alizarin red accumulation, was decreased 52% in osteoblast-like cells from BgnlDcll double deficient mice [2] compared to 30% in Bgl/ only deficient osteoblast-like cells [4].

We hypothesize that the impaired PFS in the BgnlDcl1 double deficient mice is due to altered Bmp-4 signaling sincc we found equivalent mRNA expression of Bmp-4 but marked reduction of Dlx-5 expression in the BgnlDcII double deficient cranial sutures. Many extracellular and intracellular proteins. such as Dlx-5, regu1ale the Bmp-2/4 signaling pathway. Specifically. Bmp-2/4 has been shown to increa~e Dlx-5 cxpression in avian calvarial explant cultures [13] and in osteoblastic cell cultures [10]. In addition, the Bmp-2!4 regulation of the osteoblast differentiation markers, RUIlX-2, Osterix, and Alkaline phosphatase, appears to be dependent on Dlx-5 [17,19.20]. Impaired PFS fusion in the BgnlDcn donble deficient mice may be due to failure of appropriate cxpression of signaling molecules involved in promoting bone formation and mineralization, which is consistent with previous findings of severe osteopenia [2,5] and decreased expression ofDlx-5 in the cranial sutures. However, as has been suggested by others, it is important to consider the possibility that additional factors besides increased or decreased bone formation could control abnonnal cranial suture fusion [3,6].

Table ! Summary of craniofacial anomalies in BglI. DCII, and BglI/DclI deficient mice compared to Wt

Genotype E 18.5 Post-natal Suture gene/protein calvaria PF fusion expressjon

Wt Nonna! Nonnal DIx-5 and Bmp-4 mRNA higher in PF vs. S

BgII deficient ND Nonnal Den protein higher in Bgn deficient

D ell deficient NO Nonnal Bgll protein unaffected BglIIDcn deficient Hypomineralized Patent suture Dlx-5 mRNA absent in

PF and S sutures

Nonna! WI mice and mice deficient in Bgn, Den, or both (genotype, column I) were analyzed at embryonic day 18.5 (E 18.5) by alizarin red/alcian blue staining (column 2) or post-natally (60 days after birth) by faxitron X-ray or histology (column 3). Specific mRNA or protein .xpression in the normal and mutant mice wa' examined in PF (posterior frontal) or sagittal (S) sutures 35-45 days after birth (colwun4). ND~not detennined.

22

866 S Wadhwa e/ aI, / Bmw 40 ()(j07) 8til~866

In conclusion, we show that the alteration in the extraeellular matrix of the posterior frontal suture causes impaired fusion (Table I). These new data provide the fOlmdation for the identification of novel candidate genes for patients who have sutuml growth disturbances,

Acknowledgments

This research was supported by the National Institute of Dental and Craniofacial Research in the Intramural Program of the NIH.

References

! J I Arneye L. Young ME Mice deficient in small leucine-rich proteoglycans; novel in vivo models for OsteOllorosis, osteoarthritis, EhIers-Danlos syndrome, muscular dystrophy, and comeal di,eases. Glycobiology 2002; 12:107R-16R.

PI Bi Y, &'tuelten CH. Kilts T, Wadhwa S, lozzo RV, Robey PO, el at Extracellular matrix proteoglycans control the fate of bone mruTOw stromal cells. J Bioi Chern 2005.

[3] Chen L,U D, LiC. Engel A, DengCx. A Ser250Trp substitotionin mouse fibroblast growth Illctor receptor 2 (F gfr 2) results in craniosyoostosis. Bone 2003;33;169~ 78,

[4] Chen XD, Fisher LW, Robey PG, Young IVIF. The small leucine-rich proteoglycan biglycan modulates BMP-4-indllced Gsteob!.st differentia­tion. FASEB J 2004;18:948··58.

[5] Corsi A, Xu T, Chen XD. Boyde A_ Liang J. Mankani M, et aL Phenotypic eflects ofb;glycan deficiency are linked to collagen fibril abnonnalities, are synergized by decorin deficiency. and lnimic EhIers-Danlos-like changes in bone and other connective tissues. J Bone Miner Res 2002;17:1180-9,

[6) Dabovic B, Chen Y. Col.rossi C. Zambuto L, Obata H. Rifkin DB. Bone defects in latent TGF-beta binding prolein (Ltbp)-3 nun mice; a role for Ltbp in TGY-beta presentation. J EndocrinoI2oo2;175:l29~4L

[7] Danielson KG, Earlham! H, Holmes DF, Graham H, Kadler KE, 1(221) RV. TalgCted disruption of decorin leads to abnotmal collagen fibril morphology and skin fragility. J Cell Bioi 1997;136:729-43.

[8] Flll'Iderbnrgh JI-. Hevelone NO, Rolb M~ Funderbnrgh ML, Rodrigues MR, Nirankari VS, et aI. Decorin and big!ycan of normal and palbologic human cnmeas. Invest Ophthalmol Visual Sci 1998;39:1951~64.

[9] Gutierrez J, Osses N, Brnndan E. Changes in secreted aod cell associated proteoglycan synthesis during conversion of myoblast< to osteoblasts in response to bOlle morphogenetic protein.2: role of decorin in cell response to BMP-2. J Cell Physiol 200S.

[IOJ Harris SE, Guo D, Harris MA, Krislmaswamy A. Lichtler A Transcrip­tional regulation of BMl'-2 activated genes in osteoblasts using gene expression microarray analysis: role of Dlx2 and Dlx5 transcription factors. Front Biosel 2003;8:.1249-65.

[II] Hildebrand A, Ronlaris M, Rasmussen LM, Heinegare D, Twrudzik DR, Border WA, et .1. lnree.ellen of the small interstitial proteoglycans biglycan, decor;n and fibromodulin wilb transfonning growth factor beta, Biochen> J 1994;302(1'1 2):527~34.

[12] Hocking AM, ShillOmura T, McQuillan OJ. Leucine-rich repeat glyco­proteins of the exlracellularmalnx. Matrix BioI 1998;17:1-19.

[13J Holleville N, Quilhac A, Bontoux M, Monsoro-Burq AH. BMP signals regulate Dlx5 during eady avian skull development. Dev Bioi 2003; 257: J 77-89.

[l4] Iouo RV. The biology of the small leucine-rich proteoglycans. Function.1 network of interactive proteins. J BioI Chern 1999;274:1:\843-6.

[15] louo RV. The family of the small leucine-rich proteoglycans; key regulators of matrix assembly and cellular growth. Crit Rev Biochem Mol Bioi 1997;32:141-74.

[16] Karolya N. Shigemasa K, Takagi M. Gene expression and immunohislo-

chemical localization of decorin and biglycan in association with early bone formation in Ibe developing mandible. J Oral Sci 2001 ;43;179-88.

[17] Kim YJ, Lee MH, Wozney 1M, Cha JY. Ryoo HM. Bone morplmgenetic protein-2-induced alkaline phosphatase expres.ion is stimulated by Dlx5 and repressed by Msx2. J Bioi Chern 2004;279:50173~80.

[I8J Kuwahara M, Takuma T, Scott PG. Dodd CM, Mizoguchi 1. Biochemical and inununohistnchen>ical stodies of Ule protein expression and Inc.liz.­lion of decorin and biglycan in the temporomandibular joint disc of growing rats. Arch Oral BioI 2002;47;473-80.

[19] Lee MH, Kim YJ, Kim HI. Park HD, Kang AR. KYlH1g HM, BMP-2-induced Runx2 expression is mediated by D1x5. and TGF-beta 1 opposes the BMP-l-induced osteoblast differentiation by suppressiou of DLx5 expression. J Bioi Chen> 2003;278;34387~94.

[20] Lee MH, Kwon TG, Park: HS, W(Jl;l1fJ)I JM, Ryoo HM, BMP-2-induced OSterix e.xpression is mediated by DIx5 but is independent of Runx2. Biochem Biophys Res Commun 2003;309;689~94.

[21] Matheson S, Lmjava H, Hakkinen L Distinctive localization and function for lumican, fibromodulin and decorin 10 regulale collagen fibril organization in periodontallissues. J Periodontal Res 2005;40:312-24.

[22] Matsuura T, Duarte WR, Cheng H, U2lIwa K. Yamauchi M, el al. Differential expression of decorin and biglycan genes during monse tooth development. Matrix Biol2001;20:367~73_

[23J Miura M, Chen XD. Allen M~ Bi Y. Gronthos S, Seo BM. et al. A crucial role of caspase-3 in osteogenic differentiation of bone marrow stromal stem cells. J Clin Invest 2004;114;1704~13.

[24] Nacamuli RP, Song HM, Fang TD, Fong lCD, Mathy ,lA, Shi YY. et a1. Quantitative transcriptional analysis of fusing and nonfusing cranial sumre complexes in mice. Plast Reconstr Surg 2004;114:1818-25.

[251 Oppermall L..-\. Cranial sutures as intrrunembranous bone growth sites. Dev Dyn 2000;219;472~85.

[26} Opperman LA, Rawlillli JT. The extracellular matrix environmenl in suture morphogenesis and growth. ('-ells Tissues Organs 2005:181;127-35,

[27] Parisnlhimao D. Mochida y, Duarte WR. Yamauchi M. Biglycan modulates osteoblast differentiation and matrix minernlization. J Bone Miner Res 2005;20:1878-86.

[28] Tenorio DM, Santos MF, Zorn TM. Distribution of big lye an and decorin in rat dental tissue. Braz J Med BioI Res 2003:36:1061-5.

[29] Thench.ris AD, KarrunrulOs NK.. Papageorgakopoulou N, Tsiganos CP, Theocharis DA. Isnlation and cbaracterization of matrix proteoglycalls from hnman na.,al cartilage. Compositi()"'~l and structural compariwn between normal and scoliotic tissues, Biochim Biophys Acta 2002; I 569:11 7-26.

[30J Vu HL, Panehal J, Parker EE, Levine NS, Francel P. The timing of physiologic closure of Ibe metopic suture: a review of 159 patients using reconstructed 3D CT scans of the craniofacial region. J Craniofac Surg 2001;12:527-32.

[31] Wadhwa S, Embree Me, Bi Y. Young MF. Regulation, regulatory activities, and function of biglycan. Crtt Rev Eukaryotic Gene Expression 2004;14:301-15.

[32] Warren SM. Brunet Ll, Harland RM, Economides AN. Longaker MT. The BMP antagonist noggin regulates cranial suture fusion. Nature 2003; 422:625~9.

[33] Warren SM. Greenwald JA, Spector JA, Bouletreall P. Mehrara BJ, Longaker MT. New developments in cranial suture research. PI .. I Reconslr Surg 2001;107:523-40.

[34] Weinzweig J, Kirschner RE. Farley A, Reiss P, Hunter J, Whitaker LA, et .1. Metopic synostosis: defining the temporal sequence of normal suture fusion and differentiating it from synostosis on Ibe basis of computed tomography images, PlaS! Reconstr Surg 2003;112:12118.

[35] Wilda M, Bachner D, Just W, Geerkens C. Krans p, Vogel W, et a!. A comparison of the expression patlern of five genes of the family ofsmall letlCine-rich proleoglycans during moose developroent J Bone Miner Res 2000:15:2187-96.

[36] Xu T, Bianco P. Fisher LW, Lnngenecker G, Smith E, Goldsrein S, 01 al.

23

Targeted disruption of the biglycan gene leads to an osteoporosis-like phenotype in mice. Nat Genet 1998:20;78~82.

Chapter 3

CRANIAL SUTURAL GROWTH IN HETEROZYGOUS Dlx5 +1- MICE

Ana Ortiz DDS 1, Jing Chen DDS, PhD1, Alex Lichtler PhD2

, Sunil Wadhwa DDS,

PhD!

I Division of Orthodontics, Department of Craniofacial Sciences, School of Dental

Medicine, University of Connecticut Health Center, Farmington, CT

2 Department of Reconstructive Sciences, School of Dental Medicine, University of

Connecticut Health Center, Farmington, CT

24

Introduction

There are two main types of cranial suture growth disturbances. Too little or

delayed bone growth at these sutures results in wide-open fontanels and a human

disease known as enlarged parietal formania, whereas too much or accelerated bone

growth results in osseous obliteration of the sutures, and the human disease

craniosynostosis (Opperman 2000).

We have previously reported that in mice deficient in both biglycan and decorin,

there is lack of PFS fusion at 60 days. Analysis of gene expression in the PFS from 35

day-old wt and double deficient biglycanJdecorin mice, revealed equivalent expression

ofBMP-4, but a marked decrease in the expression of Dlx5 in the double deficient

mice. Therefore, we hypothesized that Dlx5 plays a role in PFS fusion. In order to test

this hypothesis, we examined the expression of Dlx5 in the cranial sutures from CD-l

mice and examined if a cranial suture phenotype was present in Dlx5 mIce.

In this study, we found that Dlx5 is abundantly expressed in both the sagittal and

PFS between 10 and 35 days postnatal. There was an increase in the expression of Dlx5

in the PFS compared to the sagittal suture at day 35. Examination of the cranial sutures

from Dlx5 +1- mice revealed no defects in PFS fusion, but instead an increase in the

distance between the parietal bone fronts in the sagittal suture. Taken together, these

results suggest that Dlx5 plays a role in postnatal growth of the sagittal suture in CD-l

mIce.

25

Materials and Methods

To develop the mouse model for mice used in our experiments, Dlx5 transgenic

mice (Obtained from Dr Alex Lichler, UCllC, Farmington, CT) were bred with mice

transgenic for the 2.3 kb fragment of the collagen Type I promoter fused to a Green

Fluorescent Protein (2.3 GFP) (Obtained from Dr. David Rowe, UCHC, Farmington,

CT). The specific breeding pairs were Col I 2.3 GFP (+1+) Dlx5 (+1+) with

heterozygous Dlx5 (+1-), and Col 12.3 (+1+) heterozygous Dlx5 (+1-) with Col 12.3

GFP (+1+) Dlx5 (+1+). Heterozygous Dlx5 (+1-) 2.3 GFP (+1+) and homozygous Dlx5

(+1+) 2.3 GFP (+1+) littermates were analyzed.

Genotyping

All mice were genotyped for Dlx5 alleles by polymerase chain reaction (PCR)

analysis as described previously (Depew et a1. 1999). PCR products were resolved by

electrophoresis through 1 % agarose gels, yielding a 368 bp band for the wt Dlx5 allele

and a 459 bp band for the disrupted Dlx5 allele. PCR products were visualized by

ethidium bromide staining under an ultraviolet light

26

Histology and Immunohistochemistry

Frozen Embedding

Wild-type and heterozygous Dlx5 (+/-) mice were sacrificed by CO2 inhalation

at 10 days after birth. Calvaria were removed and fixed for 2 weeks at room temperature

in 10% formalin. Next, they were separated into posterior frontal and sagittal regions

and placed in separate cassettes. After a 5 minute water wash, they were decalcified in

formic acid bone decalcifier solution (Immunocal from Decal Corporation, Tallman,

NY, USA) for 4 weeks. Specimens were then washed in water for 5 minutes and fixed

for 3 days in buffered zinc formalin (Z-fix from Anatech Ltd, Battle Creek, MI, USA).

Prior to embedding, a 200mI beaker containing 2-methylbutane was pre-chilled

over dry ice under a hood. Disposable base molds (Thermo Shandon) were filled with

frozen embedding medium (Thermo Shandon), care was taken to avoid the introduction

of bubbles. Posterior frontal and sagittal suture regions of the calvariae were immersed

in individual molds containing the embedding medium. The embedding media was flash

frozen by holding the mold with forceps in a solution of 2-methylbutane keeping the

embedding mold on a horizontal level. Once the medium was frozen, the mold was

allowed to sink to the bottom of the beaker until it was completely frozen. The molds

were removed from the methyl butane solution and wrapped in a square of aluminum

foil, placed in a plastic container, and stored at -20°C.

27

CryoJane Frozen Section

Cryosectioning was perfonned on a Leica CMl900 Cryostat (D-69226; Leica,

Inc., Nussloch, Gennany) equipped with CryoJane Frozen Sectioning Kit (Instrumedics

Inc.; Hackensack, NJ). The CryoJane system is designed to capture a frozen cryostat

section on special cold adhesive tape to assist transferring the section to a cold glass

microscope slide coated with an ultraviolet light-curable, pressure-sensitive adhesive.

The section is pennanently bonded to the adhesive on the slide with a flash of

ultraviolet light.

The block containing either the decalcified posterior frontal or sagittal suture

was oriented in the block holder to obtain a 5-~ coronal section. After capturing the

frozen section on a piece of cold tape, the section was transferred to the cold slide, and

the transfer tape was removed, leaving the frozen section behind on the microscope

slide. The slides were air dried in the dark and kept in a slide box at 4°C before

process mg.

Fluorescent Imaging

Slides holding the frozen sections were soaked in IX PBS for 20 minutes then

mounted with cover slips using 50% glycerol in PBS. Col I 2.3 GFP was examined with

a microscope under ultraviolet light and pictures were taken with a color digital camera

at lOX magnification.

28

Immunohistochemistry

Tissue sections were deparaffinized with xylene. Following rehydration with

graded ethanol, sections were treated in 3 % peroxide in methanol for 20 minutes to

block endogenous peroxidase activity. Next, sections were incubated in 10% goat

serum for 30 minutes to reduce non-specific binding and incubated overnight at 4° C

with primary antibody at a 1: 200 dilution. Dlx5 antibody was purchased from

Chemicon (Temecula, CA) cat #AB5728 and Biotinylated rabbit anti-goat was used as a

secondary antibody. Immunostaining was visualized by a streptavidin-peroxidase

solution in the presence of DAB chromagen. The negative control consisted of the

above mentioned procedures except for the substitution of the primary antibody with

rabbit IgE.

Semiquantitative RT-PCR

Posterior frontal and sagittal suture complexes, including the associated dura

mater, suture mesenchyme, and osteogenic fronts, were harvested from male wt mice.

Mice were euthanized with carbon dioxide, and the posterior frontal and sagittal suture

complexes will be isolated under a dissecting microscope as described by (Nacamuli,

Song et al. 2004). RNA was extracted from the calvarial sutures using Trizol-Reagent

(Molecular Research Center, Cincinnati, OH) following the manufacture's instructions.

One microgram total RNA from the sample preparation was reverse transcribed with 50

units of SuperScript II RT using random hexamer primers (InVitrogen Life Technology,

Carlsbad, CA) following the manufacturer's instructions. The primers used for

amplification were designed with Primer 3 software (http://www-

29

genome.wi.mit.edulcgi-binlprimer/primer3.cgi). The primers used for RT-PCR were

5' gagaggccctatcccaact3 ' (forward) 5' gtgggtgcagcgaactttat3 ' (reverse) for 18S,

5' catgtggtgtgaggaaa3' (forward) and 5' gcccagagactcatttctt3 ' (reverse) for BMP-4, and

5'ccgggacgctttattagatg3' (forward) and 5'tggacactatcaatggtgcc3' (reverse) for Dlx5. The

reaction was chilled to 4°C until it was analyzed. Sixteen microlitres of the PCR

product and a 100-bp DNA ladder (Gibco BRL) were run in a 10% acrylamide gel

(TBE) buffer at 100 V. The products were separated by electrophoresis and visualized

after ethidium bromide staining under a UV light then photographed with a digital

camera. The R T -PCR procedure was repeated three times. Individual bands on digital

micrographs of RT -PCR products were analyzed by densitometer using Image J NIH

IMAGE. Intensity for Dlx5 and Bmp-4 was calculated as a percentage of densitometry

values for control bands of 18S.

Measuring the Width of the Sagittal and Posterior Frontal Sutures

H & E stained coronal sections of the posterior frontal and sagittal sutures from 10

day-old wt and heterozygous Dlx5 (+1-) mice were captured with a digital camera. In

Adobe Photoshop the distance between the bone fronts was measured in arbitrary units.

A micro-ruler was captured using the same digital camera, at the same magnification,

and arbitrary units were converted to Ilm by measurement against a known distance.

Statistical Analysis

Statistical significance of differences among means was determined by analysis

of variance with post-hoc comparison of more than two means by the Bonferroni

30

method or the Mann-Whitney rank sum test for nonparametric populations using

SigmaStat (Jandel Scientific, San Rafael, CA).

Results

Dlx5 expression in cranial sutures

Immunohistochemistry for Dlx5 was performed in 10 day-old CD-l mice. In the

PFS from 10 day-old mice, Dlx5 expression was in the underlying dura and overlying

periosteum (Figure lA). A similar expression ofDlx5 was noted in the S compared to

the PFS (Figure IB).

A time course for the expression of Dlx5 and BMP-4 was determined in the

posterior frontal and sagittal sutures from CD-I mice at day 21, day 25, day 35, and day

45 postnatally by semi-quantitative RT -PCR. The data indicated that mRNA encoding

Dlx5 and BMP-4 were abundantly expressed in both the posterior frontal and sagittal

sutures during maturation (Figure 2). Since there appeared to be an increase in

expression ofDlx5 and BMP-4 mRNA in the PF compared to the S suture during the

late phases ofPFS fusion, we repeated the experiments two more times at day 35

postnatal. In three independent experiments from 35 day-old wt mice, the mean

decrease and standard error of Bmp-4 mRNA expression in the sagittal suture compared

to the PFS was 83 +/- 15%. Additionally, in three independent experiments from 35

day-old wt mice, the mean decrease and standard error of Dlx5 mRNA expression in the

sagittal compared to the PFS was 59 14 % (Figure 2 b).

31

Analysis of the Posterior frontal and Sagittal sutures from 10 day-old heterozygous

Dlx5+/- andwt mice

Mice deficient in both Dlx5 alleles die shortly after birth and have almost a

complete absence of their calvaraie. In contrast, Dlx5 +/- heterozygous mice are viable.

In order for PF sutural fusion to occur, the bone fronts of the frontal bone must come in

contact with each other, which is expected to occur between 21 and 45 days postnatal.

However, we have observed this process of suture fusion beginning at a time point prior

to 21 days postnatal. Therefore, the cranial sutures from 10 day-old wt and Dlx5 +/­

were examined. H & E coronal sections of the Posterior frontal (Figure 3) and Sagittal

sutures (Figure 5) revealed that there appeared to be greater separation of the bony

fronts in the sagittal suture ofDlx5 +/- mice compared to wt mice. Measurement of the

distance between the bone fronts of the Posterior frontal and Sagittal sutures from 10

day-old wt (n=3) and Dlx5 mice (n=3), revealed a significant increase (p<.05) in the

distance between the parietal bone fronts in the sagittal suture from Dlx5 +/- compared

to wild-type controls (Figure 6). No significant differences were noted in the PFS

suture (Figure 4),

Osteoblast differentiation in the cranial sutures from 10 day- old heterozygous Dlx5 +/­

andwtmice

Dlx5 has been show to mediate osteoblast differentiation (Ryoo et ai., 1997),

therefore we hypothesized that there would be a decrease in osteoblast differentiation in

the cranial sutures from Dlx5 mice compared to wt mice. In order to examine this,

we bred Dlx5 +/- mice with mice transgenic for the 2.3 kb fragment of the collagen

32

Type I promoter fused to a Green Fluorescent Protein (2.3 GFP). Previous investigators

have shown that the 2.3 GFP promoter is an accurate marker of differentiated

osteoblastic cells (Kalajzc et al.,2002). We have initial histological sections of the PFS

(Figure 7) and S suture (Figure 8) from 10 day-old Dlx5 +1+ 2.3 GFP and Dlx5 2.3

GFP mice. However, we have not analyzed the samples.

Discussion

Our experiments show that Dlx5 is expressed both in the fusing posterior frontal

and non-fusing sagittal sutures of mice during the time period of expected suture fusion.

Holleville et al. found, in chick calvariae, that Dlx5 is highly expressed at the

osteogenic fronts and at the edges of the suture mesenchyme, but not in the suture itself

(Holleville et aI., 2003). Data from Acampora et aI., who studied mice, also indicated

that the expression of Dlx5 was more pronounced in periosteal bone and was also

observed in endosteal cells (Acampora et al., 1999). These findings are similar although

some were found in chick embryos and others were found in mice. Our results also

showed that Dlx5 was expressed in the osteogenic fronts as well as the periosteum and

underying dura but not in the suture space.

In this study, we chose to use heterozygous Dlx5 (+1-) mice to investigate the

role ofDlx5 in cranial suture fusion due to previous findings that Dlx5 (-1-) mice are

affected by a severe phenotype. Depew et al. found that Dlx5 (-1-) mutants die shortly

after birth and approximately 25% of them are born exencepha1ic. The mutant mice

born without exencephaly have hypomineralized parietals and interparietals and all of

33

the mutants had the phenotype of regional defects in their nasal and otic capsules

(Depew et al., 2005). Our results lacked a distinct phenotype in the PFS of Dlx5 (+1-)

mice. Previous reports from Depew et al. also recorded that no morphologic changes

were noted with the loss of a single Dlx5 allele. Therefore, it is very possible that the

phenotype of our heterozygous Dlx5 (+1-) mice was too mild for changes in fusion of

the PFS to be visible. In addition, several studies have shown evidence which might

suggest that other Dlx genes compensate when their family member is deficient. Dlx

homeobox genes are organized into pairs as follows: Dlx1l2, Dlx5/6, and Dlx

317(Robledo et al., 2002). Compared with their other family members, Dlx5 and Dlx6

are uniquely localized to all skeletal tissues (Holleville et al., 2003). In fact, Acampora

et al., found Dlx5 and Dlx6 in immature chondrocytes and developing skull bones of

embryonic mice and rats and in the periosteum of more mature bones (Acampora et al.,

1999). Robledo et aI., 2002 showed that there are redundant roles for Dlx5 and Dlx6

during osteogenesis in his study ofDlx5/6 (-1-) mice whose phenotype included a

complete lack of calvarial bones.

Developmental defects found in Dlx5 mutants that support our findings have

been described by Depew et al. in 1999. Scanning electron microscope analysis ofDlx5

homozygous mutant embryos revealed that a neural tube closure defect is centered in

the midbrain but also includes the forebrain and hindbrain in the more severe cases.

Examination ofthe calvarial roofs of non-exencephalic Dlx5 mutants revealed small,

hypomineralized parietal and interparietals (Depew et al., 1999). This developmental

phenotype, both in the specific areas of the brain, and in the parietal and interparietals

34

of the Dlx5 mutants supports our findings in the sagittal sutures ofDlx5 heterozygous

(+1-) mice.

Another possible reason for the lack of phenotype that was seen in the PFS but

was seen in the sagittal suture from heterozygous Dlx5 (+1-) mice could be that PFS

fusion is dependent in part on sox-9 expression and a cartilage template (Sahar et aT.,

2005). The role of Dlx5 in chondrocyte differentiation is different than that in bone, and

this is why we propose that an obvious phenotype was not seen in the PFS. However,

we did appreciate a phenotype in the sagittal suture ofDlx5 heterozygous (+1-) mice;

this may be attributed to the lack of the necessity for a cartilage template in the sagittal

suture and the role ofDlx5 in regulating osteoblast differentiation (Sahar et aT., 2005 &

Ryoo et aT., 1997).

We hypothesize that the phenotype observed in the sagittal sutures ofD1x5 (+1-)

mice is due to a decrease in osteoblast differentiation compared to wt controls. In order

to test this hypothesis, we bred Dlx5 (+1-) mice with mice transgenic for the 2.3 kb

fragment of the collagen Type I promoter fused to a Green Fluorescent Protein (2.3

GFP). The 2.3 GFP promoter has been shown to be an accurate marker of differentiated

osteoblastic cells (Kalajzc et al., 2002). However, we have not analyzed the samples.

Although craniosynostosis is a rare finding, it may cause severe malformations

and developmental disorders in those that are affected. Since the genetic etiology is

unknown in the majority of human patients with sutural growth disturbances, evidence

that a Dlx5 signaling pathway is involved in cranial suture fusion, may provide a new

candidate gene for analysis in patients with sutural growth disturbances.

35

The development of our knowledge of the genes involved in suture fusion is

essential for progress in the medical and genetic treatment of individuals with cranial

suture fusion disorders.

36

Chapter 4

Discussion

Previously, our lab studied PFS fusion in biglycanldecorin double deficient

mice. In the study, we found that biglycan and decorin are highly expressed in fusing

cranial sutures as well as the dura in overlapping but distinct patterns. The double

deficient mice that were studied had patent sutures and severe hypomineralization of

frontal and parietal bones. However, mice deficient with either biglycan or dec orin

separately were absent of suture formation defects.

Similar to Dlx5, both Biglycan and Decorin have been linked to BMP2/4

activity. Recently, it was found that Biglycan deficient calvarial osteoblast-like cells

have less cell surface BMP4 binding. This led to a decrease in BMP4 mediated

osteoblast differentiation in the Biglycan deficient calvarial osteoblast-like cells

compared to the wild-type controls (Chen et al., 2004). Similarly, Parisuthiman et al.

found that MC3T3-E1 clonal cells expressing higher or lower levels ofbiglycan show

less BMP4 mediated osteoblast differentiation (Parisuthiman et at., 2005). Decorin has

also been shown to alter BMP2/4 activity. For example, Gutierrez et al showed a

decrease in BMP2 induction of alkaline phosphatase in decorin deficient myoblast cells

compared to wild-type controls (Gutierrez e aI., 2006).

In this previous study, we hypothesized that the deficiency in PFS fusion that

was observed in biglycanl dec orin double knock out mice was due to altered BMP4

signaling because we found equal amounts of mRNA expression of BMP4 but markedly

reduced expression of Dlx5 in the cranial sutures ofbiglycanldecorin double deficient

37

mice. This finding drove our interest in studying the role ofDlx5 in cranial suture

fusion.

Our experiments show that Dlx5 is expressed both in the fusing posterior frontal

and non-fusing sagittal sutures of mice during the time period of expected suture fusion.

Holleville et al. found that Dlx5 is highly expressed at the osteogenic fronts and at the

edges of the suture mesenchyme, but not in the suture itself in chick calvariae

(Holleville et al., 2003). Although we studied murine sutures, we also found that Dlx5

was expressed in the osteogenic fronts as well as the periosteum and underying dura but

not in the suture space. We also found that 10 day-old heterozygous Dlx5 (+/ -) mice had

increased space between their bony fronts in the sagittal suture compared to wt controls,

and that there was no apparent phenotype in the PFS from the Dlx5 (+/-) heterozygous

mIce.

Different models have been proposed in regards to the factors responsible for

Dlx5 expression in cranial sutures (Figure 9). Choi et al. found that the addition of

FGF2 to developing bone fronts in murine sutures stimulated BMP2 expression but that

the addition of BMP2 could not induce FGF2 expression. Another finding, was that

disruption ofRunx2 completely eliminated expression ofBMP2 and its downstream

genes Dlx5 and Msx2. These results suggested that Runx2 is required in addition to

FGF2 to work upstream ofBMP2 for the expression ofDlx5 (Choi et al.,2005).

Alternatively, Holleville et al. used a dominant-negative construct (DlxHD-EnR) that

interfered with chick Dlx5 transactivating activity in E6 calvaria cells transfected with a

constitutively active type I BMP receptor (AcBMPR). Results of their studies showed

that DlxHD-EnR prevents the induction ofRunx2 activation by BMP signals. They

38

concluded that Dlx5 or related family members such as Dlx6 are necessary for the

initiation of Runx2 in immature chick calvaria mesenchyme triggered by BMP

signaling (Holleville et al., 2007). These results contrast with the findings ofChoi et aL

who found that Runx2 was involved in this cascade prior to activation ofBMP2/4.

Differences in the results obtained by these authors may be explained by their use of

animals from different species (Figure 9).

The increase in spacing between parietal bone fronts in the sagittal sutures from

Dlx5 (+1-) mice could be due to a decrease in osteoblast differentiation, changes in

apoptosis, and/or changes in proliferation in the sutural complex. Holleville et at.

recently showed that Dlx5 does not effect proliferation in immature calvarial cells

(Holleville et al.,2007). In addition, Acampora et al. (1999) showed no difference in

the apoptoic cells in a variety of tissues from the Dlx5 (-1-) compared to wt controls.

Therefore taken together, we hypothesize that increased spacing in the sagittal suture

seen in the Dlx5 (+1-) mice is due to a defect in osteoblast differentiation. In our study,

we started to examine differentiation by the use of mice transgenic for Col I 2.3 GFP,

however we have not finished analyzing the samples.

A. Future studies

Future studies may be useful in furthering our knowledge ofDlx5 and its role in

cranial suture fusion. Since a phenotype was observed only in the sagittal suture of

mutant mice, these studies may involve examining the proliferation, differentiation and

apoptosis of cells in the sagittal suture ofDlx5 mutant and wild-type mice.

39

Additional studies that may enhance our knowledge of Dlx5 may be to knock

out the gene postnatally. Because the homozygous mutant has such a severe phenotype,

resulting sometimes in death at birth, we propose that the gene be knocked out

conditionally by flox Dlx5 and giving a tamoxifen cre postnatally. This would avoid the

problem of the lethal phenotype and still allow examination of the role of Dlx5 during

suture fusion.

B. Significance

Although craniosynostosis is a rare finding, it may cause severe malformations

and developmental disorders in those that are affected. Since the genetic etiology is

unknown in the vast majority of human patients with sutural growth disturbances,

evidence that a Dlx5 signaling pathway is involved in cranial suture fusion, may

provide a new candidate gene for analysis in patients with sutural growth disturbances.

The development of our knowledge of the genes involved in suture fusion is

essential for progress in the medical and genetic treatment of individuals with cranial

suture fusion disorders.

40

References

Acampora, D., G. R. Merlo, et al. (1999). "Craniofacial, vestibular and bone defects in mice lacking the Distal-less-related gene Dlx5." Development 126(17): 3795-809.

Alappat, S., Z. Y. Zhang, et al. (2003). "Msx homeobox gene family and craniofacial development." Cell Res 13(6): 429-42.

Antonopoulou, 1., L. A. Mavrogiannis, et al. (2004). "Alx4 and Msx2 play phenotypically similar and additive roles in skull vault differentiation." J Anat 204(6): 487-99.

Babler, W. J., J. A. Persing, et al. (1982). "Skull growth after coronal suturectomy, periostectomy, and dural transection." J Neurosurg 56(4): 529-35.

Baer, M. J. (1954). "Patterns of growth of the skull as revealed by vital staining." Hum BioI 26: 80-126.

Bellus, G. A., K. Gaudenz, et aI. (1996). "Identical mutations in three different fibroblast growth factor receptor genes in autosomal dominant craniosynostosis syndromes." Nat Genet 14(2): 174-6.

Brunet, L. J., J. A. McMahon, et aL (1998). "Noggin, cartilage morphogenesis, and joint formation in the mammalian skeleton." Science 280(5368): 1455-7.

Chen, X. D., L. W. Fisher, et al. (2004). "The small leucine-rich proteoglycan biglycan modulates BMP-4-induced osteoblast differentiation." Faseb J 18(9): 948-58.

Choi, K. Y., H. 1. Kim, et al. (2005). ItRunx2 regulates FGF2-induced Bmp2 expression during cranial bone development." Dev Dyn 233(1): 115-21.

Chun, K., A. S. Teebi, et al. (2002). "Genetic analysis of patients with the Saethre­Chotzen phenotype." Am J Med Genet 110(2): 136-43.

Couly, G. F., P. M. Coltey, et al. (1993). "The triple origin of skull in higher vertebrates: a study in quail-chick chimeras." Development 117(2): 409-29.

Depew, M. J., J. K. Liu, et al. (1999). "Dlx5 regulates regional development of the branchial arches and sensory capsules." Development 126(17): 3831-46.

Depew, M. J., C. A. Simpson, et al. (2005). "Reassessing the Dlx code: the genetic regulation of branchial arch skeletal pattern and development." J Anat 207(5): 501-61.

41

El Ghouzzi, V., M. Le Merrer, et al. (1997). "Mutations of the TWIST gene in Saethre­Chotzen Syndrome. II Nat Genet 15: 42-46.

Gold, L. I., J. 1. Sung, et al. (1997). "Type I (RI) and type II (RII) receptors for transforming growth factor-beta isoforms are expressed subsequent to transforming growth factor-beta ligands during excisional wound repair." Am J PatholI50(1): 209-22.

Greenwald,1. A., B. J. Mehrara, et al. (2000). "Regional differentiation of cranial suture-associated dura mater in vivo and in vitro: implications for suture fusion and patency." J Bone Miner Res 15(12): 2413-30.

Gutierrez, J., N. Osses, et al. (2006). "Changes in secreted and cell associated proteoglycan synthesis during conversion of myoblasts to osteoblasts in response to bone morphogenetic protein-2: role of dec orin in cell response to BMP-2." J Cell Physio1206(1): 58-67.

Henderson, 1. H., M. T. Longaker, et al. (2004). "Sutural bone deposition rate and strain magnitude during cranial development." Bone 34(2): 271-80.

Holleville, N., S. Mateos, et aL (2007). "Dlx5 drives Runx2 expression and osteogenic differentiation in developing cranial suture mesenchyme." Dev BioI 304(2): 860-874.

Holleville, N., A. Quilhac, et al. (2003). "BMP signals regulate Dlx5 during early avian skull development." Dev Bio1257(1): 177-89.

Howard, T. D., W. A. Paznekas, et a1. (1997). "Mutations in TWIST, a basic helix-Ioop­helix transcription factor, in Saethre-Chotzen syndrome." Nat Genet 15(1): 36-41.

Iseki, S., A. O. Wilkie, et al. (1997). "Fgfr2 and osteopontin domains in the developing skull vault are mutually exclusive and can be altered by locally applied FGF2.11 Development 124(17): 3375-84.

Jabs, E. W., X. Li, et al. (1994). "Jackson-Weiss and Crouzon syndromes are allelic with mutations in fibroblast growth factor receptor 2." Nat Genet 8(3): 275-9.

Jabs, E. W., U. Muller, et al. (1993). "A mutation in the homeodomain of the human MSX2 gene in a family affected with autosomal dominant craniosynostosis." Cell 75(3): 443-50.

Jiang, x., S. Iseki, et al. (2002). "Tissue origins and interactions in the mammalian skull vault." Dev BioI 241(1): 106-16.

Johnson, D., S. W. Horsley, et al. (1998). "A comprehensive screen for TWIST mutations in patients with craniosynostosis identifies a new microdeletion syndrome of chromosome band 7p21.1." Am J Hum Genet 63(5): 1282-93.

42

Kalajzic, I., Z. Kalajzic, et al. (2002). "Use of type I collagen green fluorescent protein transgenes to identify subpopulations of cells at different stages of the osteoblast lineage." J Bone Miner Res 17(1): 15-25.

Kan, S. H., N. Elanko, et aL (2002). "Genomic screening of fibroblast growth-factor receptor 2 reveals a wide spectrum of mutations in patients with syndromic craniosynostosis." Am J Hum Genet 70(2): 472-86.

Kim, H. J., D. P. Rice, et al. (1998). "FGF-, BMP- and Shh-mediated signalling pathways in the regulation of cranial suture morphogenesis and calvarial bone development." Development 125(7): 1241-51.

Kim, Y. J., M. H. Lee, et al. (2004). "Bone morphogenetic protein-2-induced alkaline phosphatase expression is stimulated by Dlx5 and repressed by Msx2." J BioI Chern 279(49): 50773-80.

Kingsley, D. M. (1994). "The TGF-beta superfamily: new members, new receptors, and new genetic tests of function indifferent organisms." Genes Dev 8(2): 133-46.

Lee, M. H., Y. J. Kim, et al. (2003). "BMP-2-induced Runx2 expression is mediated by Dlx5, and TGF-beta 1 opposes the BMP-2-induced osteoblast differentiation by suppression ofDlx5 expression." J BioI Chern 278(36): 34387-94.

Lee, M. H., Y. J. Kim, et aL (2005). "Dlx5 specifically regulates Runx2 type II expression by binding to homeodomain-response elements in the Runx2 distal promoter." J BioI Chern 280(42): 35579-87.

Lee, M. H., T. G. Kwon, et al. (2003). "BMP-2-induced Osterix expression is mediated by Dlx5 but is independent of Runx2." Biochem Biophys Res Commun 309(3): 689-94.

Lenton, K. A., R. P. Nacamuli, et al. (2005). "Cranial suture biology." CUff Top Dev BioI 66: 287-328.

Levine, J. P., J. P. Bradley, et al. (1998). "Studies in cranial suture biology: regional dura mater determines overlying suture biology." Plast Reconstr Surg 101(6): 1441-7.

Longaker, M. (2001). "Role ofTGF-beta signaling in the regulation of programmed cranial suture fusion." J Craniofac Surg 12(4): 389-90.

Ma, L, S. Golden, et aL (1996). "The molecular basis of Boston-type craniosynostosis: the Pro148-->His mutation in the N-terminal arm of the MSX2 homeodomain stabilizes DNA binding without altering nucleotide sequence preferences." Hum Mol Genet 5(12): 1915-20.

43

Mabbutt, L. W. and V. G. Kokich (1979). "Calvarial and sutural re-development following craniectomy in the neonatal rabbit." J Anat 129(Pt 2): 4l3-22.

Mabbutt, L. W., V. G. Kokich, et aL (1979). "Subtotal neonatal calvariectomy. A radiographic and histological evaluation of calvarial and sutural redevelopment in rabbits." J Neurosurg 51(5): 691-6.

Mavrogiannis, L. A., I. B. Taylor, et aL (2006). "Enlarged parietal foramina caused by mutations in the homeobox genes ALX4 and MSX2: from genotype to phenotype." Eur J Hum Genet 14(2): 151-8.

McCarthy, T. L., M. Centrella, et aL (1989). "Insulin-like growth factor (IGF) and bone." Connect Tissue Res 20(1-4): 277-82.

Mehrara, B. J., J. Greenwald, et aL (1999). "Regional differentiation of rat cranial suture-derived dural cells is dependent on association with fusing and patent cranial sutures. 1I Plast Reconstr Surg 104(4): 1003-13.

Meijlink, F., A. Beverdam, et al. (1999). "Vertebrate aristaless-related genes." =-=-=:..=...:.. BioI 43(7): 651-63.

Miyama, K., G. Yamada, et al. (1999). "A BMP-inducible gene, dlx5, regulates osteoblast differentiation and mesoderm induction." Dev Bioi 208(1): 123-33.

Mossaz, C. F. and V. G. Kokich (1981). "Redevelopment of the calvaria after partial craniectomy in growing rabbits: the effect of altering dural continuity. II Acta Anat (Basel) 109(4): 321-31.

Muenke, M., K. W. Gripp, et al. (1997). IIA unique point mutation in the fibroblast growth factor receptor 3 gene (FGFR3) defmes a new craniosynostosis syndrome." Am J Hum Genet 60(3): 555-64.

Muenke, M., U. Schell, et al. (1994). "A common mutation in the fibroblast growth factor receptor 1 gene in Pfeiffer syndrome." Nat Genet 8(3): 269-74.

Noden, D. M. (1978). "The control of avian cephalic neural crest cytodifferentiation in skeletal and connective tissues." Dev BioI 67: 296-312.

Noden, D. M. (1988). "Interactions and fates of avian craniofacial mesenchyme." Development 103: 121-140.

Oldridge, M., A. O. Wilkie, et aL (1995). "Mutations in the third immunoglobulin domain of the fibroblast growth factor receptor-2 gene in Crouzon syndrome." Hum Mol Genet 4(6): 1077-82.

44

Opperman, L. A. (2000). "Cranial sutures as intramembranous bone growth sites." Dev Dyn 219(4): 472-85.

Opperman, L. A., P. T. Gakunga, et aL (2005). "Genetic factors influencing morphogenesis and growth of sutures and synchondroses in the craniofacial complex." Semin Orthod 11: 199-208.

Opperman, L. A., R. W. Passarelli, et al. (1995). "Cranial sutures require tissue interactions with dura mater to resist osseous obliteration in vitro." J Bone Miner Res 10(12): 1978-87.

O'Rahilly, R. and M. J. Twohig (1952). "Foramina parietalia permagna." Am J Roentgenol Radium Ther Nucl Med 67(4): 551-61.

Parisuthiman, D., Y. Mochida, et al. (2005). "Biglycan modulates osteoblast differentiation and matrix mineralization." J Bone Miner Res 20(10): 1878-86.

Paznekas, W. A., M. L. Cunningham, et al. (1998). "Genetic heterogeneity ofSaethre­Chotzen syndrome, due to TWIST and FGFR mutations." Am J Hum Genet 62(6): 1370-80.

Qu, S., K. D. Niswender, et al. (1997). "Polydactyly and ectopic ZPA formation in Alx-4 mutant mice." Development 124(20): 3999-4008.

Reardon, W. and R. M. Winter (1994). "Saethre-Chotzen syndrome." J Med Genet 31(5): 393-6.

Rice, D. P., T. Aberg, et al. (2000). "Integration ofFGF and TWIST in calvarial bone and suture development." Development 127(9): 1845-55.

Richtsmeier, J. T., L. L. Baxter, et al. (2000). "Parallels of craniofacial maldevelopment in Down syndrome and Ts65Dn mice." Dev Dyn 217(2): 137-45.

Robin, N. H., G. J. Feldman, et al. (1994). "Linkage of Pfeiffer syndrome to chromosome 8 centromere and evidence for genetic heterogeneity." Hum Mol Genet 3(12): 2153-8.

Robledo, R. F., L. Rajan, et al. (2002). "The Dlx5 and Dlx6 homeobox genes are essential for craniofacial, axial, and appendicular skeletal development." Genes Dev 16(9): 1089-101.

Roth, D. A., J. P. Bradley, et al. (1996). "Studies in cranial suture biology: part II. Role of the dura in cranial suture fusion." Plast Reconstr Surg 97(4): 693-9.

45

Ryoo, H. M., H. M. Hoffmann, et al. (1997). "Stage-specific expression of Dlx-5 during osteoblast differentiation: involvement in regulation of osteocalcin gene expression." Mol Endocrinolll(ll): 1681-94.

Sahar, D. E., M. T. Longaker, et al. (2005). "Sox9 neural crest determinant gene controls patterning and closure of the posterior frontal cranial suture." Dev BioI 280(2): 344-61.

Satokata, I., L. Ma, et a1. (2000). "Msx2 deficiency in mice causes pleiotropic defects in bone growth and ectodermal organ formation." Nat Genet 24(4): 391-5.

Schneider, R. A., D. Hu, et aL (1999). "From head to toe: conservation of molecular signals regulating limb and craniofacial morphogenesis." Cell Tissue Res 296(1): 103-9.

Sheth, R. and B. Iskandar (2005). Craniosynostosis.

Smith, W. C. and R. M. Harland (1992). "Expression cloning of noggin, a new dorsalizing factor localized to the Spemann organizer in Xenopus embryos." Cell 70(5): 829-40.

Tadic, T., M. Dodig, et aL (2002). "Overexpression ofDlx5 in chicken calvarial cells accelerates osteoblastic differentiation." J Bone Miner Res 17(6): 1008-14.

Takagi, M., N. Kamiya, et al. (2004). "Effects of bone morphogenetic protein-2 and transforming growth factor betal on gene expression of transcription factors, AJ18 and Runx2 in cultured osteoblastic cells." J Mol Histo135(1): 81-90.

Takao, M., J. Hino, et al. (1996). "Identification of rat bone morphogenetic protein-3b (BMP-3b), a new member ofBMP-3.!! Biochem Biophys Res Commun 219(2): 656-62.

Tavormina, P. L., R. Shiang, et aL (1995). "Thanatophoric dysplasia (types I and II) caused by distinct mutations in fibroblast growth factor receptor 3." Nat Genet 9(3): 321-8.

Tubbs, R. S., M. D. Smyth, et al. (2003). "Parietal foramina are not synonymous with giant parietal foramina." Pediatr N eurosurg 39: 216-217.

Vu, H. L., J. Panchal, et aL (2001). "The timing of physiologic closure of the metopic suture: a review of 159 patients using reconstructed 3D CT scans of the craniofacial region." J Craniofac Surg 12(6): 527-32.

Warren, S. M., L. 1. Brunet, et al. (2003). "The BMP antagonist noggin regulates cranial suture fusion." Nature 422(6932): 625-9.

Warren, S. M., J. A. Greenwald, et al. (2003). "Regional dura mater differentially regulates osteoblast gene expression." J Craniofac Surg 14(3): 363-70.

46

Warren, S. M., J. A. Greenwald, et al. (2001). IINew developments in cranial suture research." Plast Reconstr Surg 107(2): 523-40.

Waterston, R. H., K. Lindblad-Toh, et al. (2002). IIInitial sequencing and comparative analysis of the mouse genome." Nature 420(6915): 520-62.

Weinzweig, 1., R. E. Kirschner, et a1. (2003). "Metopic synostosis: Defining the temporal sequence of normal suture fusion and differentiating it from synostosis on the basis of computed tomography images." Plast Reconstr Surg 112(5): 1211-8.

Wilkie, A. O. (1997). "Craniosynostosis: genes and mechanisms." Hum Mol Genet 6(10): 1647-56.

Wilkie, A. O. and G. M. Morriss-Kay (2001). "Genetics of craniofacial development and malformation." Nat Rev Genet 2(6): 458-68.

Wilkie, A. 0., S. F. Slaney, et al. (1995). "Apert syndrome results from localized mutations ofFGFR2 and is allelic with Crouzon syndrome." Nat Genet 9(2): 165-72.

Zhou, Y. X., X. Xu, et al. (2000). "A Pro250Arg substitution in mouse Fgfri causes increased expression of Cbfal and premature fusion of calvarial sutures." Hum Mol Genet 9(13): 2001-8.

Zimmerman, L. B., J. M. De Jesus-Escobar, et aL (1996). "The Spemann organizer signal noggin binds and inactivates bone morphogenetic protein 4." Cell 86(4): 599-606.

47

Figure Legends

Figure 1. Expression of Dlx5 in the cranial sutures. Immunohistochemistry (brown) for Dlx5 in the Posterior frontal suture (A) and Sagittal suture (B) from 10 day-old CD-l wild-type mice. Sections were counterstained with Hemotoxylin (Blue).

Figure 2. Expression ofBMP4 and Dlx5 in the cranial sutures from 21 to 45 day old wild-type mice. Semiquantitative RT-PCR analysis performed for BMP4~ Dlx5~ and the house keeping gene 18S on RNA extracted from the posterior frontal and sagittal sutures (A). Relative expression ofBMP4 and Dlx5 in the Sagittal/Posterior frontal suture from 35 day-old wild-type mice. RT-PCR products separated by electrophoresis were visualized after ethidium bromide staining under a UV light and photographed in a single field of view with a digital camera. Individual bands on digital micrographs of RT -PCR products were analyzed by densitometer usng Image J NIH IMAGE. Intensity for Dlx5 and BMP4 was calculated as a percentage of densitometry values for control bands of 18S. Each primer set was tested previously to assure that amplification was in the linear range by analyzing amplification at several different cycle numbers (B).

Figure 3. H&E stained coronal sections of the posterior frontal suture from 10 day-old wild-type(A) and heterozygous Dlx5 (+1-) (B) mice.

Figure 4. The mean distance in Ilm and standard error between osteogenic fronts of the posterior frontal suture in 10 day-old Dlx5 (+1+) (mean=36.5; SE=21; n=3) and Dlx5 (+1-) (mean=78.1; SE=27.4;n=3) mice.

Figure S. H&E stained coronal sections of the sagittal suture from 10 day-old wild-type (A) and heterozygous Dlx5 (+1-) (B) mice.

Figure 6. The mean distance in Ilm and standard error between parietal bone fronts of the sagittal suture in 10 day-old Dlx5 (+1+) (mean=-45.9; SE=15.5; n=3) and Dlx5 (+1-) (mean=69; SE=27.9;n=3) mce.

Figure 7. Transgenic Dlx5 (+1-) mice were bred with mice transgenic for the 2.3kb fragment of the Collagen Type I alpha 1 promoter fused to a Green Fluorescent Protein (2.3 GFP). GFP expression in the posterior frontal suture of 10 day-old Dlx5 (+1+)/2.3 GFP (+1-) (A) and Dlx5 (+1-)12.3 GFP (+1-) (B) mice.

Figure 8. Transgenic Dlx5 (+1-) mice were bred with mice transgenic for the 2.3kb fragment of the Collagen Type I alpha 1 promoter fused to a Green Fluorescent Protein (2.3 GFP). GFP expression in the sagittal suture of 10 day-old Dlx5 (+1+)/2.3 GFP (+1-) (A) and Dlx5 (+1-)/2.3 GFP (+1-) (B) mice.

Figure 9. Proposed model of Dlx5 expression in cranial sutures

48

Figure 1

49

BMP4

Dlx5

18S

A

~ ~ l.L 0... en co

Cf)

B

21d PF Sag

25d PF Sag

35d 45d PF Sag PF Sag

Expression of Genes in Sag/PF

60

50

40

30

20

10

0 BMP4 Dlx5

Figure 2

50

Dlx5 +/+ PFS

100 ~m A

Dlx5 (+/-) PFS

100 ~m B

Figure 3

51

i _ _ _

Distance between Osteogenic Bone Fronts in the PFS of 10 day-old Dlx5 (+1+) and Dlx5 (+1-) Heterozygous Mice

120 ,--------------------------------------------------------------------------------,

E 100 .= J!l c: £ 80 .. c:

d5 60 c: ., i m 40 .. " c: .. ... 20 o

o Dlx5 (+/+) Dlx5 (+/-)

Figure 4

52

Dlx5 +/+ Sagittal suture

100 /-!m A

Dlx5 +/- Sagittal Suture

B

Figure 5

53

e--= J!l c:: e u..

'" c::

dl c::

'" '" .! '" m

'" " c:: ~ .. is

120

100

80

60

40

20

0

-20

-40

-60

-80

Distance Between Parietal Bone Fronts in 10 day-old DlxS (+1+) and DlxS (+I-)Heterozygous Mice

*

Dlx5 (+1-)

Figure 6

54

2.3 GFP Dlx5 +/+ PFS

2.3 GFP Dlx5 +/- PFS

Figure 7

55

2.3 GFP Dlx5 +/+ Sagittal suture

2.3 GFP Dlx5 +/- Sagittal Suture

Figure 8

56

Models of Factors Involved in the expression of Dlx5 in the Cranial Sutures

I

• •

r Choi et al. r Holleville et al. ! • •

BMP FGF2

Added to Bone Fronts

Runx2 Dlx5

- Runx2 BMP2 -

Dlx5 and Msx2

Figure 9

57