THE ROLE OF MATRIX METALLOPROTEINASES IN MURINE
ADRIANA COSTA DA SILVEIRA
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE
UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE
REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
I would like to thank many people that helped me along the way during these five
years of graduate school. Thanks go to my mentors, Dr. Michael Humphreys-Beher and
Dr. Linda Brinkley, who encouraged me to think on my own. I value their advice and
guidance and hope that we will continue our friendship for years to come.
To all members of my committee, Dr. Nancy Denslow, Dr. Greg Schultz, Dr. David
Muir and Dr. William Williams, thanks go for the support and belief that I could finish
this project. Their expertise and advice helped me tremendously.
A few people were essential for my acceptance to graduate school, Dr. Mary Jo
Koroly, Dr. Sheldon Schuster and Dr. Arnold Bleiweis. My first year tuition was covered
by the ICBR while I worked in the Protein Core Facility. My gratitude goes to all of
them. I can not forget to thank everybody in the Protein Core Facility where I worked for
almost 2 years and where I learned my baby steps in molecular biology. Special thanks
go to Dr. Denslow, Ms. Marjorie Chow and Mr. Hung Nguyen. My research and stipend
were covered in part by my mentors' NIH grants and by the University of Florida. Other
institutions also responsible were the Florida/Brazil Institute and the Hispanic Dental
I would like to thank past and present oral biology students that gave me their
support and made these years in graduate school enjoyable, Dr. Heather Allison, Mr.
Jason Brayer, Dr. Lina Bueno, Dr. Dong Mei Gao, Dr. James Kohler, Dr. Shawn
Macauley, Dr. Greg Oxford, Dr. Chris Robinson, Ms. Lori Wojciechowski. I can not
forget to thank the MHB lab where working is a fun process. Special thanks go to Mr.
Micah Kerr for assistance with the PCR experiments, Mrs. Amy Shawley for assistance
with the culture experiments and Ms. Jennifer Lowry for her assistance. In the Brinkley
lab, special thanks go to Dr. Yan Du for her assistance with the zymography experiments.
My eternal gratitude goes to Drs. David and Nancy Denslow for all the support they
gave us through these years. Their friendship and caring are beyond words. To all my
family at home, my parents Antonio and Antonia Tavares da Costa, my sister, Ana Maria,
and my brother, Antonio Jose, and family, to Dr. Jose Dantas and Valniza da Silveira, Dr.
Polina da Silveira and Dr. Romulo da Silveira and family, thanks for the support and
This dissertation is dedicated to my husband, Dimitrio Sergio da Silveira, who never
let me give up and always encouraged me with unconditional love and support. This
could not be accomplished without him and the endless hours of his help.
TABLE OF CONTENTS
A CKN OW LED GM EN TS ........................................................................................... ii
ABSTRA CT....................................................................................................................... vi
IN TROD U CTION ........................................................................................................
LITERA TU RE REV IEW .............................................................................................4
Description of Facial Form ation.......................................................................................... 6
Models for Studying Facial Morphogenesis: Chicken and Mouse...................................... 8
Mechanisms of Face Morphogenesis: Cell Proliferation, ECM Synthesis and
D egradation.................................................................................................................. 10
Experim ental A lterations of ECM in Facial Processes...................................................... 18
Proteinase D egradation of the Extracellular M atrix .......................................................... 21
M atrix M etalloproteinases in Other Em bryonic System s.................................................. 27
Other M orphogenetic Factors ............................................................................................ 29
The Proposed Role of M M Ps in Facial Form ation............................................................ 30
Cleft Lip and Palate ........................................................................................................... 32
Significance of This Study................................................................................................. 33
M A TERIA LS AN D M ETH OD S....................................................................................... 35
M materials ............................................................................................................................ 35
Anim als and Tissue Preparation ........................................................................................ 36
Zym ography .......................................................................................................................37
Reverse Zym ography ......................................................................................................... 38
Im m unoblotting ................................................................................................................. 38
Reverse Transcriptase-Polymerase Chain Reaction (RT-PCR)......................................... 40
Southern Blot Analysis ...................................................................................................... 41
Com petition-based Quantitative RT-PCR ......................................................................... 42
Analysis of RN A Results................................................................................................... 44
Statistical Analysis of Q-RT-PCR D ata............................................................................. 44
Em bryo H eads Culture....................................................................................................... 44
Sectioning of Tissues ......................................................................................................... 45
R E S U L T S ................................................................. ........................................................ 4 6
Extracellular Matrix Molecules Are Being Synthesized and
Degraded at the Time of Facial Formation In Vivo ...................................................... 46
Matrix Metalloproteinases and Tissue Inhibitors Expression Do Not
Show Major Temporal Regulation During Facial Morphogenesis
In V ivo .............................................................................................................. 5 3
Inhibition of Matrix Metalloproteinases Alters Facial Development
In V itro ............................................................................................................ 6 6
D IS C U S S IO N ....................................................................................................................72
Expression and Degradation of ECM Components by MMPs in
D developing Facial Processes......................................................................... 73
Matrix Metalloproteinase Degradation of the Basement Membrane
Is Necessary for the Fusion of the Facial Processes .................................... 76
C O N C L U SIO N S ................................................................................................................ 80
LITE R A TU R E C IT ED ...................................................................................................... 81
BIOGRAPHICAL SKETCH ......................................................................................... 94
Abstract of Dissertation Presented to the Graduate School of the University of Florida in
Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy
THE ROLE OF MATRIX METALLOPROTEINASES IN MURINE FACIAL
Adriana Costa da Silveira
Chairman: Michael Humphreys-Beher
Major Department: Oral Biology
Facial morphogenesis is dependent on extracellular matrix (ECM) synthesis and
accumulation. It is tightly regulated by the interaction of the proteins that degrade the
extracellular matrix in a given time and space. Current literature suggests that matrix-
degrading enzymes play a role in many developmental processes. We suggest that facial
development is one of these processes. Molecules known or expected to be involved in
various aspects of normal development in animal models have been tested in humans
with defects and have been shown to be involved in a number of common malformations.
This study allows determination of whether matrix proteinases play a role in primary
palate and mid-face morphogenesis.
The objectives of this study were to detect changes in the expression of ECM and
matrix metalloproteinases (MMPs), to measure the amount of ECM message being
produced at the time of facial formation and to determine the role of MMPs in facial
development by perturbing their function in vitro. Immunoblots demonstrated the
presence of basement membrane ECM components and MMP-2, MMP-3 and MMP-9 in
the facial processes. Quantitative RT-PCR was used in order to measure the production of
ECM molecules' mRNA and indicated an increase in the number of these molecules as
development progresses. Specific MMP inhibitors were used to block proteinase function
in cultured embryo heads. After 7 days in culture, controls demonstrated complete
closure whereas experiments showed a cleft resultant from the lack of fusion between the
lateral and medial nasal processes. These data suggest that production of ECM mRNA
and degradation of ECM by MMPs are occurring in the facial processes at the time of
facial development in the mouse and that the MMPs are involved in the formation of the
Our study may provide insights into normal and abnormal facial development
particularly relevant to the occurrence of cleft lip. By identifying the players and
providing indications on how these players alter development, we can have a better
understanding of the whole process. This may lead to strategies for preventing the action
of perturbing agents and compensating for defective genes.
The goal of this research is to achieve a better understanding of the role of the
extracellular matrix (ECM) and matrix-degrading enzymes in normal mid-face and
primary palate morphogenesis. Many morphogenetic processes require balancing the
synthesis and destruction of ECM. These changes can be mediated by regulating the
synthesis of ECM components, matrix-degrading enzymes, and their inhibitors, by
differential proteinase-mediated degradation of ECM, or both. A vast array of factors,
including growth factors, cytokines, hormones, steroids, vitamins, protooncogenes and
even the molecules themselves are known to regulate the transcription of ECM,
proteinase and proteinase inhibitor genes (Pan et al., 1995; Gutman and Wasylyk, 1990;
Wasylyk et al., 1991; Campbell et al., 1991; Lin, Georgescu and Evans, 1993). Less work
has been done on the interaction of matrix-degrading enzymes and their substrates as it
relates to specific morphogenetic events. The present proposal will focus on just such
activities in the development of the mid-face and primary palate.
Several studies have demonstrated the presence of several ECM molecules,
matrix metalloproteinases (MMPs), plasminogen activators (PAs), tissue inhibitors of
matrix metalloproteinases (TIMPs) and their messages in the facial processes over the
course of murine facial development (Chin and Werb, 1997; lamaroon and Diewert,
1996; lamaroon et al., 1996). The combined activation of various MMPs is essential for
the efficient turnover of the structure, and the expression of MMP genes is regulated for
this purpose. This regulation is complex and can occur at different levels: intracellularly
by transcriptional and translational regulators, extracellularly by proteolytic activation of
the proenzyme and binding of inhibitors.
The following hypothesis was proposed for this study: Formation of the normal
primary palate and the mid-face is dependent on the morphogenetic movements of the
facial processes caused by specific changes in tissue architecture. These changes involve
temporo-spatially localized alterations in the distribution of the ECM molecules.
Proteinase mediated degradation of the ECM molecules plays a role in these changes.
Also, correct facial and upper lip formation depends on the degradation of the
basement membrane of the facial processes by proteases.
From the above hypothesis, we can predict that
1) The presence of matrix metalloproteinases in the facial processes will be
associated with the changes in the distribution of their target ECM molecules during
2) Disrupting the normal temporal and spatial sequence of appearance of matrix
degrading proteinases will result in abnormal facial morphogenesis.
Aim 1 is to describe the temporal appearance of MMPs, their target ECM molecules and
their messages during mid-face and primary palate morphogenesis in vivo.
Aim 2 is to determine the effects of altering the expression of MMPs by inhibiting the
matrix-degrading proteinases on primary palate and mid-face morphogenesis in vitro.
One of the major determinants of the development and maintenance of three-
dimensional form in animals is the assemblage of the extracellular matrix (ECM) (Hay,
1981). In the embryo, the ECM is the scaffolding that helps determine tissue patterns. In
the adult, it serves to stabilize these same patterns. Extracellular matrix can be found as
both interstitial matrix or organized as a basement membrane interposed between tissue
layers. In early development, interstitial matrix is found between cells in the mesoderm
and a basement membrane is found between ectoderm or endoderm and the underlying
mesoderm. As development proceeds and differentiation takes place, interstitial matrix is
found in connective tissues and as a basement membrane underlying epithelial or
endothelial layers, separating them from the underlying connective tissue, as well as
around muscle and nerve cells.
Collagens, proteoglycans, glycosaminoglycans (GAGs) and glycoproteins
compose the extracellular matrix. There are at least sixteen different types of collagen
described to date. Their composition differs in the amount and type of the basic unit of
collagens, the triple helix (comprised of three polypeptide ca chains) and also in their
configuration and tissue location.
The proteoglycans are a diverse family of molecules distinguished by a core
protein attached by one or more GAG side chains. Recent review of the nomenclature of
the proteoglycans classifies this class of glycoproteins according to their function or their
protein core whereas in the past, it reflected the chemical composition of the GAG chains
(Ayad et al., 1994). The glycosaminoglycans are a group of carbohydrates that include
chondroitin sulfate (CS), dermatan sulfate (DS), heparan sulfate (HS), keratan sulfate
(KS) and hyaluronan (HA).
Fibronectin, vitronectin and laminin are examples of glycoproteins found in the
extracellular matrix. Similar structure and function are attributed to this class of proteins.
Many hold RGD sequence motifs that mediate cellular adhesion. They have been
implicated to influence cell behavior by allowing attachment and migration and to have
the ability to concentrate ions and growth factors.
The interstitial ECM is composed of various collagens, glycosaminoglycans,
glycoproteins and elastin bathed by a tissue fluid. In the adult, it is the substrate on which
fibroblasts and macrophages normally reside. The basement membrane, defined
originally by light microscopy, is composed of substructures visible only by electron
microscopy, the principal one being the basal lamina. The basal lamina is a thin sheet-
like structure composed mostly of glycosaminoglycans, collagen and proteoglycans.
During morphogenesis tissues must change size, shape and character. This
requires spatial and temporal changes in the distribution and composition of both the
interstitial and the basement membrane ECM. Such changes have been shown to take
place during the development of several embryonic structures (Fitch, Mayne and
Linsenmayer, 1983; Bernfield et al., 1984; Thesleffet al., 1981; Sahlberg et al., 1992).
The synthesis and degradation of ECM is integral to morphogenesis (Nakanishi et al.,
1986). Tissue development and structure is controlled by dynamic and interactive
relationships between cells and the ECM they secrete. The development of the face and
primary palate is no exception. Yet little work has been done on the temporal and spatial
changes in the ECM during the development of these important structures.
Description of Facial Formation
The development of the face may be considered to begin at the time of neural tube
formation with the production of a cell population crucial to facial development, the
neural crest cells. These cells originate from the neural tube and migrate extensively in
the subectodermal region of the loose mesenchyme to their final destinations throughout
the embryo. There they give rise to a number of differentiated cell types. The neural crest
cells will ultimately give rise to most of the differentiated structures of the face and neck
since they form the mesenchyme of the swellings that will give rise to these structures,
the frontonasal process and the paired branchial arches. The facial skeleton, connective
tissues, parts of the teeth and nerve fibers are all of neural crest origin.
The formation of the five primordia that will give rise to the structures of the face
and neck begins about the fourth week of development in humans or about gestational
day eight in mice. These primordia are the frontonasal process, and the paired maxillary
and mandibular processes. The frontonasal process is initially identifiable as a single
swelling overlying the forebrain. At about the same time, paired swellings develop on
either side of the developing pharynx, the first branchial. Subsequently four other paired
swellings will develop branchiall or pharyngeal arches 2-5) although branchial arches 4
and 5 are rudimentary and very small. The first or mandibular arches rapidly subdivide
into superiorly located maxillary processes and an inferiorly located mandibular process.
These branchial arches are composed of neural crest-derived mesenchyme, sometimes
called ectomesenchyme, encased by a layer of ectoderm (Richman, Rowe and Brickwell,
1991). Extracellular matrix is present between these mesenchymal cells and in the
basement membrane between the mesenchyme and the overlying ectoderm.
The frontonasal process will also subdivide into a medial nasal and a pair of
lateral nasal processes. A critical stage in this morphogenetic event is the elevation and
outgrowth of the nasal processes around the nasal (olfactory) placodes. The nasal region
is first identified by two thickened epithelial regions (nasal placodes) that curl to form a
nasal groove that delineates the lateral from the medial nasal process on each side of the
future nose. The enlargement and fusion of the medial nasal and lateral processes
(derived from the frontonasal process) with each other and with the maxillary process
(derived from the first branchial arch) will determine the following structures: the nose,
the upper lip and the primary palate, which will ultimately constitute the middle third of
the face (Johnston and Bronsky, 1995).
Facial development occurs over about an eight-day period in humans and over a
four-day period in mice. Major changes in spatial relations take place during this time:
the brain enlarges, the distance between the nasal placodes narrows, the frontonasal
process elongates vertically and narrows, and the maxillary processes from both sides
grow forward and enlarge (Diewert and Shiota, 1990; Diewert and Lozanoff, 1993a,
1993b; Diewert, Lozanoff and Choy, 1993; Rude et al., 1994). It has been demonstrated
that the facial processes have different growth patterns and directions in different regions
of the face that may contribute to changes in morphology (Diewert, Wang and Tai, 1993).
As facial development continues, the lateral and medial nasal processes on both sides of
the developing nose make contact with each other and with their respective maxillary
process (fig. 1). Epithelial adhesion and later fusion occur at the inferior part of the nasal
groove. Subsequently this nasal fin disappears as adhered epithelial cells of the apposed
lateral and medial nasal processes die or transdifferentiate into mesenchymal cells, basal
lamina is degraded and a mesenchymal bridge between the processes that allows tissue
continuity is formed (Diewert and Shiota, 1990).
Models for Studying Facial Morphogenesis: Chicken and Mouse
Studies ofmorphogenesis require the use of experimental models. Early studies of
facial development were confined to chick embryos. Chick development occurs in a large
egg that develops outside the uterus that can be windowed to allow direct surgical
manipulation of the large embryo. Although some differences were observed between
avian and mammalian craniofacial development, researchers in the field often chose to
continue their work using the chicken model, because of its accessibility. With the
discovery that quail cells were easily histologically distinguished from chick cells, tissue
recombinations and chimera studies in culture were exploited to study craniofacial
development. The advent of new techniques such as antibodies and whole embryo culture
enabled the re-introduction of the mammalian embryo as a model for studies of
Fig 1. Mouse facial development from gestational days 9 through 11. At day 9, the nasal placodes and the
first branchial arch are easily recognized. At day 10, the frontonasal process divides into medial nasal process
and lateral process. The first branchial arch divides into mandibular and maxillary process. At day 10.5, the
facial process grow towards the mid-line, and at day 11, just prior to fusion, the lateral, maxillary process from
each side of the developing nose make contact with each other and with the medial nasal process (modified
from Mayer and Swanker, 1958).
Many transplantation and extirpation experiments in mammalian embryos have
demonstrated that cells derived from the neural crest participate in the formation of the
midfacial and visceral arch skeletal structures (Morris-Kay and Tam, 1987). Migratory
pathways have been well defined by the application of antibodies and have shown that
cells derived from different facial processes have different origins (Noden, 1988). Cells
destined for the frontonasal and maxillary processes remain in close proximity to the
prosencephalon while others move away from the neural epithelium (Johnston, 1966;
Mechanisms of Face Morphogenesis: Cell Proliferation, ECM Synthesis and
It has been proposed that growth of the facial processes is due to the proliferation
of the mesenchymal cells (Andersen and Matthiessen, 1967). Various studies have
demonstrated that cells are proliferating at high rates in the mesenchyme of facial
processes during facial development (Wilson and Hendrickx, 1977; Minkoff and Kuntz,
1977, 1978; Figueroa and Pratt, 1979; Gaare and Langman, 1980; Igawa et al., 1986;
Minkoff, 1980, 1991). Additional research has found differences in mitotic activity
within different facial regions. Growth rates were higher in the subepithelial mesenchyme
of the maxillary processes than they were more interiorly (Bailey, Minkoff and Koch,
1988). Tissue recombination studies have demonstrated that the mesenchyme is
dependent on the epithelium for its growth and survival (Saber and Minkoff, 1991;
Richman and Tickle, 1989). Other studies have found similar increased cell proliferation
when different regions within the lateral and medial nasal mesenchyme were examined
(Gui, Osumi-Yamashita and Eto, 1993). When the epithelium was analyzed, lower
synthesis was found in fusing epithelia compared to non fusing areas as development
progresses, supporting the idea that these cells are undergoing changes in preparation for
contact and cell death (Kosaka and Eto, 1986).
Failure of fusion of the facial processes has often been attributed to defective
outgrowth of the facial processes. In some instances, the processes are too small or are
not properly positioned to permit contact and fusion to occur. Defective process
formation has typically been attributed to an insufficient number of mesenchymal cells in
the processes, a condition that might result from interference with neural crest cell
migration to the area, decreased mesenchymal cell proliferation, or aberrant cell death
(Noden, 1975; Minkoff and Kuntz, 1978). The excision of the frontonasal process in the
beginning of facial morphogenesis in chick embryos results in arrested development of
the upper beak and inhibited growth of the maxillary processes (McCann, Owens and
Wilson, 1991). Similar operations involving removal of each facial process in cultured rat
embryos demonstrated that removal of the medial nasal process results in a large number
of malformations of the upper lip (Ohbayashi and Eto, 1986). The excision of maxillary
and lateral processes, however, only resulted in a small number of malformations. The
results can be interpreted in two ways. First, it has been demonstrated that the growth of
the medial nasal process is greater than of the lateral process (Patterson and Minkoff,
1985) thus, the medial nasal process could extend to the maxillary process and substitute
for the tissue lost. Second, since the medial nasal process is the biggest facial process, it
is likely the embryo cannot regenerate such a big tissue volume and malformations can
It has also been suggested that synthesis of ECM plays a role in the facial process
outgrowth (Sadler, Langman and Burk, 1980). Various studies have reported the
incorporation of radiolabeled glucosamine into new glycosaminoglycan (GAG)
molecules synthesized at the time of facial processes formation (Burk, 1983). The results
demonstrate that over half of the GAGs synthesized during the period of mouse facial
process formation and outgrowth are of hyaluronan (HA) form. Lesser amounts of
chondroitin sulfate (CS), heparan sulfate and others are also synthesized.
The composition of the extracellular matrix of the facial processes has been
studied by histochemical techniques, specific for visualization of ECM molecules (for
example, tannic acid) and more recently by employing antibodies. Antibodies permit a
better identification of the ECM components whereas specific staining procedures only
facilitate speculation on the true identity of these molecules. For example, tannic acid
(TA) may not preserve all the ECM. Collagen like fibers are seen without special
preparation and are most prevalent in subepithelial spaces associated with basal lamina
(Hall and MacSween, 1984), whereas fixation with TA enhances the retention of some
other matrix components. Chondroitin sulfate, chondroitinase ABC and HA were all
identified by different staining techniques and/or digestion with a specific enzymes as
being components of ECM in the facial processes.
An immunofluorescent study of the composition of murine day 11 mandibular
process mesenchyme described the presence of fibronectin, type II collagen and cartilage
specific proteoglycan (Richman and Diewert, 1987). Type I collagen was absent on early
days 10 and 11, and its synthesis was correlated with chondrogenesis. The distribution of
type IV collagen, laminin and fibronectin and their alteration has also been examined in
the maxillary and medial nasal processes of chick embryos (Xu, Parker and Minkoff,
1990a). Differences in concentration of type IV collagen were noted within the basement
membrane of the maxillary process and its localization was more prominent in the medial
nasal basement membrane than in any other region. In the medial nasal process, type IV
collagen appeared to be predominantly located in the basement membrane in the lateral
regions, whereas in the central region, it appeared to be located in the epithelial cells.
These differences were confirmed using two other monoclonal antibodies to type IV
collagen. Laminin was found to be more evenly distributed in the basement membrane of
the facial processes as was fibronectin within the mesenchyme. The results suggest that,
during facial morphogenesis, primordia such as the maxillary processes, which undergo
rapid changes in form, have alterations in collagen distribution to a much greater extent
than laminin or fibronectin, which could serve as structural support for the basement
membrane and the mesenchyme, respectively.
Synthesis of different ECM molecules in the basement membrane has also been
implicated as influential in the epithelial-mesenchymal interactions of the maxillary
processes (Xu, Parker and Minkoff, 1990b). In order to investigate whether the presence
of specific basal lamina components was a requirement for epithelial-mesenchymal
interactions, a study was conducted in which fluorescent monoclonal antibodies to
laminin (Ln) and type IV collagen were employed to detect the presence of these
components during tissue isolation procedures. Each isolated chick embryo epithelial and
mesenchymal portions of maxillary process produced different basement membrane
components in tissue culture. Only laminin was found in isolated epithelia after 24 hrs in
culture. When recombined, however, not only was immediate production of laminin
observed, but the basement membrane was reconstituted by 24 hours in vitro, including
type IV collagen, laminin and fibronectin. Therefore, the recombination of the epithelium
with the mesenchyme altered the patterns of synthesis of ECM molecules. Since the
examiners only looked for laminin and collagen IV, it is not known whether GAG
components of basement membrane were present or being synthesized. They suggested
that the alteration of synthesis of ECM molecules after tissue recombination can provide
an environment for the generation of signals between the epithelium and the mesenchyme
which are necessary for mesenchymal growth and survival, for example the passage of
growth factors. Alterations of ECM composition of basal lamina can also be obtained by
specific proteinase degradation.
In summary, the composition of the extracellular matrix of the facial processes has
not been completely defined. Two different experimental models have been used, the
chicken and the mouse and it is not known if their matrix compositions are the same.
Overall, the mesenchyme of the facial processes appears to be composed primarily of
glycosaminoglycan molecules, especially hyaluronan, but fibronectin and collagens are
also present. The basement membrane contains type II mandibularr process) collagen,
type IV collagen and laminin in different amounts, dependent on the region or process
The morphogenesis of tissue and organs during development requires dynamic
changes in the extracellular matrix composition. Extracellular matrix is present between
mesenchymal cells and in the basal lamina underneath the epithelium layer. This
morphogenesis can be achieved by differential synthesis or by degradation or both. The
rapid changes in tissue form that occur during morphogenesis suggest that both the
interstitial matrix and the basement membrane must be remodeling. The basement
membrane also acts as a medium by which inductive epithelial-mesenchymal interactions
are mediated and is also associated with directed cell migration. Both of these processes
require the ability to locally, quickly and specifically alter the composition of the
The difference in staining of collagen IV observed in previous studies (Xu, Parker
and Minkoff, 1990a) suggests that basement membrane degradation occurs in regions of
outgrowth of the maxillary process, more specifically away from the fusion regions.
These patterns may be associated with the morphogenetic events of primary palate
formation. It is known that basement membranes show altered temporo-spatial patterns of
distribution of components such as laminin, fibronectin, and collagens IV and I during
organogenesis. Such changes have been reported during ocular development (Fitch,
Mayne and Linsenmayer, 1983), salivary gland morphogenesis (Bernfield et al., 1984)
and tooth development (Thesleff et al., 1981; Sahlberg et al., 1992). In facial
morphogenesis, degradation of the basement membrane is a necessary step in
mesenchymal bridge formation and process fusion. The distribution of basement
membrane in the mesenchymal bridge of mouse primary palate was described by using
fluorescent antibodies to laminin, type IV collagen and fibronectin (lamaroon, Tse and
More recently, lamaroon and Diewert (1996) examined the distribution of
basement membrane components in the mouse primary palate by immunofluorescence.
They report the presence of laminin, fibronectin and collagen IV in the basement
membrane that becomes fragmented along the epithelial seam prior to fusion. This is
suggestive of a rapid disruption of the basement membrane in association with formation
of a mesenchymal bridge. Fibronectin was also found in the mesenchyme of the facial
processes. After fusion or closure, basement membrane components are found intact
along the margins of the facial processes. The continuance of the existence of the
epithelial seam and the basement membrane may be associated with cleft lip formation
(Diewert and Wang, 1992). Similar breakdown of basement membrane and regression of
the epithelial components can be found in other embryonic systems, such as during the
development of the tooth, involution of mammary glands and development of secondary
palate (Thesleff, Partanen and Vainio, 1991; Dickson and Warburton, 1992; Diewert and
Wang, 1992; Morris-Wiman and Brinkley, 1993).
Matrix degradation appears to play a role in facial morphogenesis in two ways:
interstitial and basement membrane extracellular matrix composition is altered for the
necessary shape changes of the lateral process and subsequent curling of the nasal
groove; and turnover of the basement membrane is associated with expansion and growth
of the facial processes and at the time of fusion, its degradation allows mesenchymal
The morphogenic movements of the lateral process that lead to the curling of the
nasal groove require alterations in tissue architecture. The alterations result from
temporo-spatial changes in the distribution of the ECM in the interstitial matrix and in the
basement membrane between the mesenchyme and the epithelium of the processes. Also,
by specific temporo-spatial changes in the interstitial matrix and basement membrane
composition, differential communication could take place between the epithelium and the
mesenchyme of the facial processes. This could direct the growth of the facial processes,
such as the changes in ECM composition in reorientation of palatal shelves in culture
(Morris-Wiman and Brinkley, 1993), and act as a stimulus for cell differentiation or
death. At the same time, the basement membrane must be degraded in regions of the
facial processes where growth is constant or intensified or where processes fuse to allow
mesenchymal continuity to take place.
Little work on the role of matrix-degrading enzymes in facial morphogenesis has
been done to date. Some matrix metalloproteinases and their inhibitors have been
localized in the craniofacial complex (Kinoh et al., 1996; lamaroon et al., 1996, Chin and
Werb, 1997). Among them, MMP-2 seems to be constitutively expressed during
embryogenesis. Matrix metalloproteinase-2 has been localized by immunohistochemistry
in the tips and the periphery of the nasal processes, lateral and medial nasal, at the time of
growth and just prior to fusion (lamaroon et al., 1996). After fusion, staining was
significantly reduced with some remaining at the peripheral regions of the formed nostril.
Chin and Werb (1997) also demonstrated expression of MMPs and TIMPs during
mandibular development in culture. When these MMPs activities were blocked by
specific inhibitors of matrix metalloproteinases, fusion of mandibular processes and
tongue formation were arrested or delayed in a dose dependent manner (Chin and Werb,
1997). This study also showed abundant expression of MMP-2 by gestational day 9 in the
first branchial arch in the stroma underlying the epithelium, making MMP-2 the most
likely candidate for the major player in mandibular development. Although inhibition
with synthetic broad-spectrum inhibitors of matrix metalloproteinases was successful,
attempts of blocking specific MMP activity by antisense oligonucleotides failed.
Therefore, it is understood that identification of the temporal and spatial
distribution of the major matrix molecules and their proteinases is an important step in
understanding facial morphogenesis, and an integral part of this study.
Experimental Alterations of ECM in Facial Processes
It has previously been shown that perturbation of some ECM molecules
(glycosaminoglycans and collagen) results in the production of cleft palate (Brinkley and
Vickerman, 1982). Proper temporo-spatial distribution of some ECM molecules in
mammalian embryonic palatal tissue is required for normal development of the secondary
palate. The alterations of these patterns of molecular distribution are not attributable to
specific synthesis of the molecules (Morris-Wiman and Brinkley, 1992) and likely
involve specific, local ECM degradation. It is hypothesized that the same could be true
for facial development.
Several studies have utilized agents that degrade ECM or interfere with their
synthesis. Hyaluronic acid has been shown to be a major constituent of the facial
processes which will give rise to the primary palate (Burk, 1983). Streptomyces
hyaluronidase specifically degrades HA and was injected into mouse embryos that were
cultured up to 24 hrs (Burk, 1985). The results of these studies were inconclusive since
after 8 hrs of treatment, circulation in cultured embryos ceased and cellular degeneration
Other studies have used various inhibitors. Median facial clefts were induced in
mice by using DON (Diazo-Oxo-Norleucine), a glutamine antagonist that affects
glycosylation, mesenchymal proliferation and synthesis of extracellular matrix
components (Burk and Sadler, 1983), specifically glycosaminoglycans and
proteoglycans. A decrease in cell density was observed in all areas of facial mesenchyme
after treatment with DON as measured by 3H thymidine incorporation. However, these
results were inconclusive because of the dose of the compound and its potential toxicity.
The observed decrease in cell density was probably associated with cell death. Other
experiments reported a reduction in glycoprotein and GAG synthesis when DON was
administered in cultured mouse secondary palatal shelves (Greene and Pratt, 1980).
However, the appropriate control, which would have been the addition of glutamine or
glucosamine to the culture medium to override the inhibitory effect, was not performed.
A fair number of teratological studies have indicated the medial nasal process as
being the main site of action of various teratogens during facial development. Retinoic
acid (RA), a known teratogen, causes a number of craniofacial defects. Retinoic acid is
normally present in the embryo in low amounts but defects can result when excess or
deficiency of RA occurs. Retinoic acid acts through its nuclear receptors that in turn bind
to different DNA regions directly affecting the expression of some genes, such as
homeobox genes, growth factors, oncogenes and ECM molecule genes. In facial
morphogenesis, medial nasal mesenchyme exposed to retinoic acid do not develop in vivo
(Wedden, 1987; Tamarin et al., 1987), exhibiting an effect resembling that seen after
medial nasal process excision. Recent studies involving injection of retinoic acid in
pregnant mice at the time of facial development revealed that the RA effect was time
dependent. When injected at gestational day 8.25 of mouse embryos, defects of both
branchial arches and frontonasal process occurred, while injection at gestational day 10
results in minor branchial arch malformations (Webster et al., 1986; Grant et al., 1997).
The same results were recapitulated using rat embryos in vitro (Webster et al., 1986).
Cell division might be one mechanism of facial growth that is affected, although one
should not rule out that teratogens may also affect cell migration (Patterson, Minkoff and
Johnston, 1979), ECM synthesis, ECM degradation or a combination of all. It is known
that RA suppresses the expression of the collagenase (MMP-1) gene (Pan, Eckhoff and
Brinckerhoff, 1995), therefore possibly inhibiting matrix degradation. In chicks, the
specific affects of RA on matrix production and cell proliferation of mandibular and
medial nasal processes were examined by analysis of 35S sulfate, 3H thymidine and 3H
proline incorporation (Sakai and Langille, 1992). Low levels of RA stimulated cartilage
matrix production in mandibular but not the medial nasal process as determined by the
patterns of sulfate incorporation. At higher levels, sulfated proteoglycans were reduced in
both processes. At the same time, RA induced incorporation of proline in the mandibular
but not the medial nasal process. The differences observed might reflect the different
origins of these two groups of cells within the neural crest cells. It is also known that RA
plays a role in homeobox gene expression in the developing head (Studer et al., 1994;
Helms et al., 1997). These reports show that high doses of RA can affect the growth of
frontonasal and maxillary processes by inhibiting the expression of shh and patched
homeobox genes disrupting the epithelial-mesenchymal interactions. In addition,
methotrexate and aminopterin, have produced similar effects in mouse embryos in vivo
(Burk and Sadler, 1983; Darab et al., 1987).
Most studies of alterations of ECM in facial development have been confined to
known teratogens such as DON and RA. Although they seem effective in producing
clefts, their precise mechanisms of action have not been elucidated, complicating the
analysis of the results obtained. Despite this major shortcoming, results from these
teratogen studies suggest that normal temporo-spatial distribution and production of
matrix molecules are required for normal morphogenesis.
Proteinase Degradation of the Extracellular Matrix
The matrix metalloproteinases are a family of enzymes that can digest at least one
ECM component (for a review, Parks and Mecham, 1998). Catalytic activity of these
enzymes requires a zinc ion at the active site. There is a second zinc ion in some MMPs
and a calcium ion that helps stabilize the tertiary structure of the enzymes (Lovejoy et al.,
1994). The combined activation of various matrix metalloproteinases (MMPs),
plasminogen activators (PAs) and tissue inhibitors of matrix metalloproteinases (TIMPs)
is essential for the efficient turnover of the ECM (Matrisian, 1992; Toumir et al., 1994).
They are inhibited by chelating agents, natural and synthetic inhibitors, and are secreted
as proenzymes that require proteolytic activation by other proteases in vivo or by trypsin
and 4-aminophenylmercuric acetate (APMA) in vitro. Four-aminophenylmercuric acetate
(APMA) works by inducing a conformational change in the zymogen, which allows
intramolecular self-cleavage (Nagase et al., 1990).
There are four subclasses of MMPs:
(1) interstitial collagenases (MMP-1 and MMP-8), which degrade collagens I, II
(2) gelatinases (MMP-2 and MMP-9), which degrade collagens IV, V, VII, X,
fibronectin (Fn) and gelatins;
(3) stromelysins (MMP-3, MMP-7, MMP-10, MMP-11 and MMP-12), which act
on proteoglycans, laminin (Ln), fibronectin, collagens III, IV and V, and
(4) membrane-type matrix metalloproteinases (MT1-MMP and MT2-MMP),
recently identified, which are bound to the cellular membrane and can also
mediate the activation ofpro-MMP-2 in the cell surface (Takino et al., 1995).
Most recently, a novel matrix metalloproteinase, MMP-19 has been characterized
(Pendds et al., 1997). It is thought to be a unique MMP due to different structural
characteristics, but its activity is blocked by TIMP-2 and EDTA, and it is expressed in
human placenta, lung, pancreas, ovary, spleen and intestine.
Matrix metalloproteinase-2, also called gelatinase A or 72 kDa type IV collagenase,
degrades gelatin, collagens IV, V, fibronectin, collagens VII, and X (basement
membrane), in order of preference (Birkedal-Hansen, 1995). It is usually found as a 72
kDa protein (68 kDa intermediate and 66 kDa as proteolytic end product). In rat tracheal
tissue 68 kDa and 60 kDa isoforms have been isolated (Lim et al., 1995). In tooth
development, the isoform of the MMP-2 is primarily isolated at 72 kDa (Heikinheimo
and Salo, 1995). Matrix metalloproteinase-2 is activated at the cell surface and retained
there through interaction with a receptor like molecule (Monsky et al., 1993). It is usually
found in complex with TIMP-2, resistant to activation by plasma proteins, but their
interaction is a controversial subject. Some authors have demonstrated that TIMP-2
inhibits the activation of pro-MMP-2 while others suggest that it may facilitate it through
the formation of a complex with MT 1-MMP (MMP-14) on the cell surface or to another
TIMP-2 acting as a receptor. It is thought that MMP-2/TIMP-2 complex may still be
active but gelatinase activity is only 10% of the free active MMP-2 (Kleiner et al., 1992;
Yu et al., 1996). The transmembrane (TM) domain of MMP-14 is essential for MT1-
MMP-MMP2 interaction (Cao et al., 1995). The C-terminal domain of MMP-14 binds to
the C-terminal domain of the pro-MMP-2 (Werb, 1997). It can also act as a tissue
inhibitor of metalloproteinase 2 (TIMP-2) receptor in the cell surface by its N-terminal
domain. MT1-MMP makes an excellent substrate for serine proteases, the same sequence
is found in collagenases and stromelysins that can be activated by plasmin (Strogin et al.,
1995). It has been demonstrated that MT1-MMP is activated in the Golgi by a furin
family protease or on the cell surface by proteinases such as plasmin (Sato et al., 1994;
Okumura et al., 1997).
Pro-matrix metalloproteinase-2 appears to be constitutively expressed by many
cell types (unlike other MMPs) and its message RNA is not easily induced by 12-0-
tetradecanoyl-phorbol-13-acetate (TPA) or interleukin-lcL (IL-1C), which has been
shown to increase the expression of other MMPs. The promoter lacks activating protein-1
(AP-1) or polyomavirus enhancer-binding protein-3 (PEA-3) sites and also transforming
growth factor-P3 (TGF-3) inhibitory element (Yu et al., 1996).
Matrix metalloproteinase-9, also called gelatinase B or 92 kDa type IV
collagenase, degrades collagens IV, V and XI, laminin, elastin, entactin, aggrecan core
protein and cartilage link protein (Birkedal-Hansen et al., 1993). Matrix
metalloproteinase-9 is 92 kDa in humans or 105 kDa in rodents (Tanaka et al., 1993).
Matrix metalloproteinase-9 forms a complex with tissue inhibitor of metalloproteinase-1
(TIMP-1). When the pro-MMP-9/TIMP-l complex is treated with APMA or trypsin, no
gelatinolytic activity is detected, although pro-MMP-9 is converted to lower molecular
weight forms, which correspond to activated forms (Ogata, Itoh and Nagase, 1995). A
complex with pro-MMP-1 can be formed which does not allow MMP-9 to bind to TIMP-
1. The complex MMP-1/MMP-9 is readily activated by MMP-3. Matrix
metalloproteinase-2 also seems to activate pro-MMP-9, resulting in 86 kDa
(intermediate) and 67 kDa (active) isoforms (Fridman et al., 1995). At the same time,
activated MMP-7 can also activate pro-MMP-9 in human rectal carcinoma cells (Imai et
al., 1995). In human neutrophils, MMP-9 is detected at 57.5 kDa (unglycosylated
proenzyme), 49 kDa and 41.5 kDa as active forms (Bu and Pourmotabbed, 1995).
The stromelysins are MMP-3, MMP-7, MMP-10, MMP-11 and MMP-12. They
degrade fibronectin, elastin, laminin and proteoglycans and have limited activity on
nonhelical regions of collagens IV, V, VIII, IX and the amino terminal of collagen I
(Birkedal-Hansen, 1995). Matrix metalloproteinase-3 can activate MMP-9 to a greater
extent than APMA (Ogata, Enghild and Nagase, 1992). Matrix metalloproteinase-3
(human stromelysin 1) is a pro-enzyme of 57 kDa and 60 kDa (unglycosylated and
glycosylated) and has a molecular weight of 45 kDa as the active form. It has been
detected in some mesenchymal tissues (Okada et al., 1992).
Matrix metalloproteinase-7, also called matrilysin or PUMP-1, is distinct in that it
contains only the catalytic domain required for activity, in contrast to the other members
of the family, which contain additional carboxyl-terminal domains (Gaire et al., 1994). It
is 28 kDa in the latent form, 21 kDa and 19 kDa as active forms. It degrades a wide range
of substrates such as fibronectin, laminin, proteoglycans, elastin, gelatin, collagens IV
and IX, aggrecan, entactin and small tenascin (Birkedal-Hansen, 1995). Human and
mouse MMP-7 share 75% of protein sequence similarity (Wilson et al., 1995). MMP-7
can be activated by APMA, trypsin and also by MMP-3. In turn, MMP-7 can activate
pro-MMP-1 and pro-MMP-9 (Fridman et al., 1995). The expression of MMP-7 is almost
exclusively found in epithelial cells (Wilson et al., 1995; Saarialho-Kere, Crouch and
The collagenases are MMP-1, MMP-8 and MMP-12. They cleave native triple
helical collagen in each ca chain, generating two fragments. These fragments denature at
37C to form gelatin, and although the collagenases can degrade gelatins to a certain
extent, the gelatinases are the ones to do so more often and faster. Matrix
metalloproteinase-1, also known as interstitial collagenase or fibroblast collagenase,
degrades collagens I, II, III, VII and X (Welgus et al., 1990), proteoglycan link protein
and gelatins. Many studies have shown that MMP-1 is not expressed until gestational day
14.5 or 15 in mouse embryos (Gack et al., 1995; Mattot et al., 1995). Its expression can
be correlated with the onset of bone or cartilage formation.
The Matrix metalloproteinases can be inhibited by their natural inhibitors, TIMPs,
by chelating agents, such as EDTA, and by specifically synthesized inhibitors such as
GM6001 (galardin) (Boghaert et al., 1994; Odake et al., 1994; Shams, Hanninen and
Kenyon, 1994), a hydroxamate or by Batimastat (Sledge et al., 1995), a new class of
agents specifically designed to inhibit MMP activity. There are four members of the class
of TIMPs: TIMP-1, TIMP-2, TIMP-3 (Leco et al., 1994) and the recently isolated TIMP-
4 (Leco et al., 1997). All TIMPs are glycoproteins, each with a different molecular
weight: TIMP-1 is 28 kDa (glycosylated); TIMP-2, 20 kDa (unglycosylated); TIMP-3, 24
kDa; and TIMP-4, presumably 23 kDa. They form a non-covelantly linked 1:1 complex
with the activated form of MMPs and inhibit their proteolytic activity through a
mechanism that is not entirely clear. It has been shown that TIMP-1 occupies the entire
length of the active site of MMP-3 therefore blocking its activity (Gomis-Ruth et al.,
1997). Several lines of evidence suggest TIMPs play a role in growth and development
(Brenner et al., 1989). Messanger RNA transcripts for collagenase, stromelysin and
TIMPs were detected in mouse embryos suggesting their function during growth,
development and implantation of mammalian embryos (Behrendtson, Alexander and
Regulation of the MMPs is complex and can occur at different sites:
intracellularly, at the transcriptional or translational level, extracellularly, by activation of
the proenzyme and binding of the inhibitors. Since it is likely that little or no active
matrix proteinase is free in the tissue in physiologic remodeling, MMPs are either
released in a latent form to be activated by another protease, or have been blocked by
Matrix Metalloproteinases in Other Embryonic Systems
A large and growing body of evidence suggests that MMPs play an important role
in the remodeling of ECM in various embryonic developing systems. The presence of
MMPs, TIMPs and their mRNAs was described in early blastocyst stage of mouse
embryos using zymograms and RT-PCR (Brenner et al., 1987). The addition of TIMPs to
the culture medium of early mouse embryos slowed development but did not inhibit it
completely. Matrix metalloproteinase-9 has also been detected in murine and human
osteoclasts where it is required for removal of collagen during resorption remodeling of
bone (Reponen et al., 1994). Matrix metalloproteinase-9 mRNA has also been found
during mouse neurogenesis (Canete-Soler et al., 1995). At day 11, MMP-9 message was
localized in the epithelium of nasal pit and distributed within the mandibular process
among other regions of the mouse head. Elevated expression of MMP-9 in
cytotrophoblasts of developing embryos has been detected as well (Librach et al., 1991),
suggesting that this enzyme may play an important role in cellular migration, invasion
and tissue remodeling.
Matrix metalloproteinase-2 has also been found in numerous developing systems.
The occurrence and distribution of MMP-2 in human embryonic tissues also suggests that
ECM remodeling and degradation by proteinases is a physiological event in association
with ECM deposition during development (Casasco et al., 1995). Tracheal tissue
morphogenesis involves the penetration of epithelial cells into the submucosa, a process
that requires digestion of the basal lamina and the surrounding ECM. Bovine tracheal
tissue cells produce and secrete MMP-2 (Toumrnier et al., 1994). Its localization in tracheal
tissue and gland acini was demonstrated using antibodies. Matrix metalloproteinase-2
was upregulated in epithelial and stromal cells suggesting that it probably plays a role
during rat tracheal gland morphogenesis (Lim et al., 1995). The message for MMP-2 is
expressed during murine embryogenesis. Using in situ hybridization, intense expression
was found in day 10-15 embryos in mesenchymal tissue of kidney, lung and epithelium
of the submandibular gland (Reponen et al., 1992). The overall results of MMP-2
localization studies showed that MMP-2 is primarily expressed in mesenchymal cells
with the exception of gland morphogenesis (salivary, tracheal).
There is limited knowledge about the physiological role of MMP-7. However,
there is evidence for it having a role in fetal development. The synthesis of MMP-7 in
developing human mononuclear phagocytes was reported (Busiek et al., 1992). Matrix
metalloproteinase-7 was also found in germinal basal cells during human fetal skin
development (Karelina et al., 1994). In the mouse, MMP-7 appears to be primarily
expressed by epithelial cells or epithelial derived cells and in tumors, by parenchymal
cells rather than stromal cells (Wilson et al., 1995; Wilson and Matrisian, 1996, 1998).
Due to its localization, MMP-7 has been associated with basement membrane
degradation in neoplastic lesions (Wilson et al., 1995) and in many embryonic systems
(Busiek et al., 1992; Karelina et al., 1994; Saarialho-Kere, Crouch and Parks, 1995).
Formation and degradation of dental basement membrane are important for tooth
development. Expression of MMP-2 and MMP-9 was observed in human fetal tooth
development (Heikinheimo and Salo, 1995). During later stages, high levels of MMP-2
mRNA were confined to differentiating and secretary odontoblasts. This indicates that
MMP-2 might participate in both remodeling of the enamel organ-dental papilla
basement membrane and its final degradation.
Other Morphogenetic Factors
Besides MMPs, another family of proteinases, the plasminogen activator (PA)
family, may influence ECM degradation through activation of the MMPs or by directly
acting on the matrix. Two different enzymes have been identified in mammals as
members of the PA family: urokinase-type plasminogen activator (uPA) and tissue-type
plasminogen activator (tPA). Like PAs, plasmin has also been implicated to have a role in
ECM degradation (Vassalli, Sappino and Belin, 1991). Plasminogen is the preferred
substrate for PAs whereas fibrin is the preferred substrate for plasmin but most matrix
proteins can also be cleaved by these enzymes. Plasminogen activators convert
plasminogen into plasmin and, in addition, plasmin is one of the activators of
metalloproteinase precursors (He et al., 1989). Plasminogen activator catalyzed activation
of plasminogen results in a dramatic amplification of the proteolysis of the matrix
components. The expression and activity of uPA has been demonstrated in
preimplantation rat embryos (Zhang et al., 1994), suggesting that embryonic uPA may be
involved in early embryo development and implantation.
Recent studies involving genetic knockouts of uPA, tPA and plaminogen activator
inhibitor-1 (PAI-1) (Carmeliet et al., 1995) produced mild phenotypes, suggesting that
these proteases have a limited role in regulatory development. It might be possible that
their functions have been substituted by other similar proteins since there seems to be a
high level of gene redundancy that still remains to be elucidated.
The Proposed Role of MMPs in Facial Formation
In normal primary palate and face morphogenesis, closure is dependent on the
lateral nasal processes making contact with the medial nasal process. Besides the rapid
growth and high rates of cell proliferation, this contact is enhanced by the morphogenetic
movements of the lateral process, initiated as a curling forward of the lateral portion to
form the nasal wing. These morphogenetic movements result from specific changes in
tissue architecture involving temporo-spatially localized alterations in the distribution of
the ECM molecules. Proteinase mediated degradation of the ECM molecules likely plays
a role in these changes. This suggests that matrix proteinases that degrade extracellular
matrix are necessary for these events to occur. The role of MMPs, their activators and
inhibitors in modulating the ECM environment during this morphogenetic event has not
The MMPs and PAs have also been shown to play a role in the metabolism of the
ECM in development. We suggest that they are probably involved in facial
morphogenesis. Several MMPs, TIMPs and their messages are present in the facial
processes, including MMP-2, MMT1-MMP and MMP-9 (Kinoh et al., 1996; lamaroon et
al., 1996, Chin and Werb, 1997), and seem to be temporally regulated over the course of
marine facial development. At this time, the processes are a collection of undifferentiated
cells that will only begin to differentiate at late day 11, forming specialized organs such
as teeth and later on, specialized tissues such as cartilage and bone. Thus, the MMPs
detected should be produced only by the mesenchymal and epithelial cells of the facial
It has been demonstrated that many tumor types express MMPs. It is important to
note that expression within a tumor type does not recapitulate its fetal expression. For
example squamous cell carcinomas of the skin (Pyke et al., 1993) and lung (Canete-Soler,
Litzky and Muschel, 1994) express MMP-9 but only the lung demonstrates expression of
MMP-9 during embryogenesis. Therefore, the fetal expression pattern does not predict
which tumors in adults will express MMPs and vice-versa.
It is interesting to discuss why two different types of matrix metalloproteinases
(MMP-2 and MMP-9) with similar substrate specificities are expressed at the same time
during morphogenesis. Results of transgenic mouse experiments with knock-outs for
MMP-3, MMP-7, MMP-9 and MMP-12 and overexpression of MMP-1 and MMP-3 were
disappointing (Shapiro, 1997) since no dramatic phenotypes were produced. These results
seem to give an idea the redundancy of MMP function. One or more enzymes can take
over and compensate after one MMP is knocked out. The characterization of their genes
have demonstrated that each enzyme may have a different control of expression (Huhtala,
Chow and Tryggvason, 1990; Huhtala et al., 1991). By responding to different elements
(for example, growth factors or the presence of inhibitors), the cell can regulate the
expression of the different enzymes and change its surrounding environment.
The interactions of the ECM molecules, MMPs, PAs and their inhibitors must be
intrinsically regulated in normal morphogenesis. These factors may be controlled by
differential spatial and temporal localization within a given tissue. The initial step in
elucidating the role of these molecules in mid-facial and primary palate morphogenesis is
to determine the temporal appearance and distribution of matrix metalloproteinases, their
target ECM molecules, and their messages. These data will permit the identification of
specific times in facial morphogenesis for optimal targeting of perturbation experiments
that will further define the role of matrix metalloproteinases in the process.
Cleft Lip and Palate
Although there is considerable knowledge of the development of various
craniofacial malformations, less progress has been made in research related to the
mechanisms involved in development and fusion of the upper facial processes.
Alterations of the developmental sequence of these processes lead to the most common
major craniofacial malformations, cleft lip and/or palate. It is estimated that one of every
700 live births in the United States has a cleft lip or palate or both. Since embryological
studies demonstrated that the lip and the palate close at different times, it is hypothesized
that cleft lip and cleft palate may arise through different mechanisms or processes.
Clefts have different etiologies: genetic, with transforming growth factor-a
(TGFa) being the candidate gene in certain families with predisposition to cleft palate
(Ardinger et al., 1989), and environmental, which is thought to account for the majority
of the occurrences. Knock-out experiments of transforming growth factor P33 (TGF33)
showed that null homozygous mutant animals suffer from cleft palate (Kaartinen et al.,
1995). Recent studies have suggested that a combination of genetic background of the
affected person and of the mother and environmental agents have to be present for the
occurrence of nonsyndromic oral clefts (Maestri et al., 1997). These are complex birth
defects characterized by an uncertain mode of inheritance, incomplete penetrance, and
heterogeneity within and among populations. Environmental factors can be divided into
five groups: (1) infectious agents, (2) irradiation, (3) drugs, (4) hormones, and (5)
nutritional deficiencies (Ten Cate, 1989).
The use of inbred mice A/J, A/WySn and C57BL6/J strains (susceptible to clefts
after use of teratogens) is an excellent tool for the identification of genes or chromosomal
locations involved in clefting. A number of candidate genes have been found by
performing a genome wide search for loci in mouse (Diehl and Erickson, 1997). Among
eight collagen genes and several extracellular matrix components;
twenty oncogenes or tumor suppressor genes;
sixteen genes related to detoxification ofteratogens such as glutathione S-transferase;
homeobox genes such as Msxl, Msx2 and Pax9; five genes related to retinoic acid,
including P3 and y (cx, involved in spontaneous cleft, lies outside);
* growth factors and receptors, including epidermal growth factor (EGF), epidermal
growth factor receptor (EGFR), fibroblast growth factor receptor (FGFR) and TGF[3.
The authors suggested that multiple loci contribute to facial clefts and some of
these loci can cause susceptibility to both cleft lip and cleft palate but with different
parental strain alleles contributing to two different forms.
Significance of This Study
Facial morphogenesis is dependent on ECM synthesis and accumulation. It is
tightly regulated by the interaction of the proteins that degrade the ECM in a given time
and space. Current literature suggests that matrix-degrading enzymes play a role in many
developmental processes. We suggest that facial development is one of these processes.
Molecules known or expected to be involved in various aspects of normal development in
animal models have been tested in humans with defects and have been shown to be
involved in a number of common malformations. This study allows determination of
whether matrix proteinases play a role in primary palate and mid-face morphogenesis.
Our study may provide insights into normal and abnormal facial development particularly
relevant to the occurrence of cleft lip. By identifying the molecules that may play a role
and providing indications on how they may alter development, we can have a better
understanding of the whole process. This may lead to strategies for preventing the action
of perturbing agents and compensating for defective genes.
MATERIALS AND METHODS
CD-1 mice were purchased from Charles River Laboratory. Microfast kits for
RNA extraction were purchased from Invitrogen (Carlsbad, CA). Reverse transcriptase-
polymerase chain reaction (RT-PCR) Gene-Amp RNA kits were obtained from Perkin-
Elmer-Cetus (Emeryville, CA). Antibodies to laminin (A/B) were purchased from
Chemicon (Temecula, CA), as were positive controls for gelatinases A/B. Monoclonal
antibodies to fibronectin were purchased from NeoMarkers (Fremont, CA). Another set
of antibodies to laminin (B1/B2) was obtained from UBI (Lake Placid, NY). Antibodies
to MMP-2, MMP-3 and MMP-9 were gracious gifts from Dr. David Muir (University of
Florida)(Muir, 1994; 1995). Secondary Alkaline Phosphatase-conjugated goat anti-rabbit
IgG and goat anti-mouse IgG were purchased from American Qualex (San Clemente,
CA). Anti-sheep IgG (whole molecule) alkaline phosphatase-conjugated second antibody
was purchased from Sigma (St. Louis, MO). Culture media and supplements were
obtained from Gibco BRL/Life Technologies (Gaithersburg, MD). Culture dishes and
Transwell-Clear supports were acquired from Costar (Cambridge, MA). Galardin
(GM6001), a dipeptide analogue with the structure of N-[2(R)-2-(hydroxamido
carbonylmethyl)-4-methylpentanoyl]-L-tryptophanemethylamide synthesized as
described (Grobelny D, Poncz L, Galardy R, 1992 and unpublished data) was a gift from
Dr. Greg Schultz (University of Florida). Tissue inhibitor of metalloproteinase-2 (TIMP-
2) for culture experiments was acquired from Boehringer Mannheim (Indianapolis, IN).
Animals and Tissue Preparation
The presence of a vaginal plug following overnight mating was taken to indicate
pregnancy and designated gestational day 0 (gd 0). At days 9, 10, 10.5 and/or 11, the
pregnant mice were killed by cervical dislocation. Embryos and facial processes were
dissected under the light microscope using microinstruments. Besides date of pregnancy,
total appearance, crown-rump length, facial appearance and limb development were used
as physical features in the developmental classification of the embryos in each litter
through their comparison to anatomical charts (Kaufman, 1992; Theiler, 1989). Maxillary
and mandibular processes were obtained from gestational days 10 and 11. Lateral and
medial nasal processes were obtained from day 11. Total RNA was extracted by using
Micro-Fast Track kit. Tissues were homogenized for immunoblots and zymography in
non denaturing buffer (200 mm NaCl, 50 mM Tris, 5 mM CaC12, pH 7.6 and 0.1% Triton
X-100) (Carol Brenner, personal communication) and DNA was enzymatically digested.
Samples were maintained in -80C freezer. Protein concentration was determined by Bio
Rad Protein Assay.
Facial processes samples containing 2 to 10 gg of total protein were mixed with
Laemmli sample buffer and were separated under non-reducing conditions on 10%
polyacrylamide gels containing gelatin, a-casein or a-casein with 0.04 U/ml plasminogen
as substrates (Adler, Brenner and Werb, 1990; Chin, Murphy and Werb, 1985). The gels
were incubated at 37C overnight in buffer (50 mM Tris, 5 mM CaC12, 150 mM NaCl,
0.02% NaN3, pH7.8) with or without the inhibitors; 4mM ethylenediaminetetraacetic acid
(EDTA), 10 mM phenanthroline, 20mM phenylmethylsulfonyl fluoride (PMSF),
aprotinin (ltg/al). Following removal of SDS by washing in buffer containing 0.1%
Triton-X 100, the gels were stained with Coomassie blue. After succeeding destaining,
enzyme activity appears as a clear band against a blue background of undigested protein.
Gels were photographed and then dried. Dry gels and their pictures were scanned
and analyzed using Gel Pro Analyzer program. This program gives the approximate
molecular weight for each band corresponding to regions of gelatinolytic activity and
compares to the known molecular weight standards run on the same gel. It also performs
a densitometric analysis that gives the ratio of intensity of each active band in the gel
compared to the others. Band sizes and activity intensities were estimated using this
One mM of 4-aminophenylmercuric acetate (APMA) was used for activation of
gelatinases, by incubating the samples overnight at 37C. Four-aminophenylmercuric
acetate (APMA) activates matrix metalloproteinases by cleaving the propeptide and
subsequently exposing the active site.
Reverse zymography is a useful technique that allows visualization of the
presence of natural inhibitors. To detect TIMP inhibitory activity, gels were performed
using collected conditioned media at 23.5% (vol) in 10% SDS-PAGE gelatin gels. The
gels were incubated at 37C overnight in buffer (50 mM Tris, 5 mM CaCl2, 150 mM
NaC1, 0.02% NaN3, pH7.8). Following removal of SDS by washing in buffer containing
0.1% Triton-X 100, the gels were subsequently stained with Coomassie blue. Since the
whole gel carries media that contains proteinases, a blue band of non-degradation appears
in the place where the inhibitor is located, demonstrating the inhibitor's presence and
Forty |tg of each facial process sample was dissolved in a sample buffer
containing 5% P3-mercaptoethanol, separated on 10% SDS-PAGE gels and then
transferred to a nylon membrane (Millipore, Bedford, Mass). The membrane was
incubated with blocking solution (5% dry milk/ 3% bovine serum albumin (BSA)) in
TTBS buffer (10 mM Tris-HCl pH 7.4, 200 mM NaCl, 0.02% NaN3, Tween-20) for 3
hours at room temperature prior to overnight incubation with primary antibody diluted in
the blocking solution. The primary antibodies used were: rabbit IgG anti-human MMP-2
(made against peptide 475-490 residues, gift from Dr. David Muir), sheep anti-human
MMP-9 (gift from Dr. David Muir), sheep anti-human MMP-3 (gift from Dr. David
Muir) (Muir, 1994; 1995), mouse monoclonal anti-fibronectin (NeoMarkers, Fremont,
CA), rabbit polyclonal anti-mouse laminin B1/B2 (Upstate Biotechnology Incorporated),
rabbit polyclonal anti-mouse laminin A/B and rabbit anti-mouse collagen IV (Chemicon,
Temecula, CA). Antibodies against MMP-2 were raised against the sequence
MGPLLVATFWPELPEK corresponding to amino acid residues 475-490 near the
carboxy terminus of MMP-2 (Muir, 1994). This antibody binds to both the proform and
the activated form of the MMP-2 enzyme but demonstrates a variable detection of the
proform in immunoblots (Collier et al., 1988; Muir, 1994). The antibody against MMP-3
was raised against the sequence DPNAGKVTHILKSN, corresponding to the residues
457-470 of the rat MMP-3 at the carboxy terminal (Muir, 1994). Blots were washed three
times with TTBS for 45 min. The membranes were incubated with the respective
secondary antibodies overnight, followed by washing step described above. Bands were
detected by a visualization protocol coupling alkaline phosphatase to BCIP and NBT as
substrates, a fast procedure that allows visualization within 30 min (average) with
sensitivity as low as 100 pg of protein. In addition to tissue samples from the different
facial processes, controls for antibody specificity were comprised of no incubation with
the primary antibody. Negative control samples used tissues homogenates that did not
contain the target protein while positive controls included tissue preparations that
contained known amounts of target antigen, such as for MMP-2 and MMP-9 positive
controls from Chemicon (Temecula, CA) or isolated human placenta (Shimamori,
Watanabe, Fujimoto, 1995).
Reverse Transcriptase-Polymerase Chain Reaction (RT-PCR)
Reverse Transcriptase-Polymerase Chain Reaction reactions were performed to
detect the presence of specific messages. The nucleotide sequences for mouse genes were
retrieved from Genbank and primers were specifically designed using MacVector
software program version 6.0.1 (Oxford Molecular Ltd., San Jose, CA) and synthesized
in the University of Florida DNA Synthesis Core Facility. We designed specific primers
for MMP genes, MMP-1, MMP-2, MMP-3, MMP-7, MMP-9, TIMP-1, TIMP-2, tPA and
uPA (table 1). CopyDNA was synthesized in 20 pl reactions containing 10 mM Tris-HC1
(pH 8.3), 50 mM KC1 buffer; 5mM MgCl2; 1 mM of each dNTP; 1 U of Rnase inhibitor;
2.5 U of Reverse Transcriptase; 2.5 uiM of random hexamers primers, and < 1 tg of total
RNA. The reaction was carried out at 42C for 15 min, 99C for 5 min and 5C for 5
min. DNA amplification was performed in 100 j1 reactions containing 20 tl of initial RT
reaction in 10 mM Tris-HCl (pH 8.3), 50 mM KC1 buffer; 2 mM of MgC12; 2.5 U of Taq
polymerase and 30 mM of each 5'and 3'primers. The PCR reaction was carried out at
94C for 4 min for the initial hot start, 94C for 1 min, 60C for 1 min and 72C for 1
min for 35 to 40 times and 72C for the final 7 min. The annealing temperature of 60C
was adjusted according to the conditions set for each reaction and each targeted message.
Glyceraldehyde-3-phosphate dehydrogenase (G3PDH) or P3-actin were used as
positive controls for reactions. Each product was further analyzed by southern blot
technique to confirm its identity using an internal probe. The primers used are shown
Table 1: MMP, PA and TIMP Primers.
Gene 5' 3' bp
MMP-1 GCCATTACCAGTCACCGAGGA GGAATTTGTTGGCATCAC 467
MMP-2 GGCCATGCCATGGGGCTGGA CCAGTCGGATTTGATGCT 762
MMP-3 CTGGAGGTTTGATGAGAAGA AACCAGCTGTTGCTCTT 207
MMP-7 GTGAGGACGCAGGAGTGAAC ACAGGTGCAGCTCAGGAA 309
MMP-9 AGGCCTCTACAGTCTTTG CGTCCTTTCTTGTTGGAC 825
uPA GTGGAGAACCAGCCCTGGT GGCAGGCAGATGGTCTGT 348
tPA TCCACCTGCGGCCTGAGGCAA CACACTCTGTCCAGTCAG 445
TIMP-1 TGCCTCATCCCATCCCCACAA AGGGGTGTCAGAGGGTGA 530
TIMP-2 GAGCCAAAGCAGTGAGCGAG GGTACCACGGCGAAGAAC 400
P-ACTIN TCTACAATGAGCTGCGTGTG GAAGTACTCCATCAGGCA 304
G3PDH TGAAGGTCGGTGTGAACGGAT CATGTAGGCCATGAGGTC 983
_______ TTGGC CACCAC
Southern Blot Analysis
To confirm the identity of the amplicons generated by PCR, the products were
separated in a 1.5% agarose gel and transferred to a nylon membrane (Boehringer
Mannheim, Indianapolis, IN). Internal probes were specifically designed from gene
sequences retrieved from Genbank and digoxigen labeled as per manufacturers
instructions (Boehringer Mannheim, Indianapolis, IN) (Table 2). After hybridization,
alkaline-phosphatase anti-digoxigen antibody was applied to the membrane, followed by
a chromogenic substrate. The probes used are shown in Table 2.
Table 2: MMP, PA and TIMP Probes.
Competition-Based Quantitative RT-PCR
Competition-based Q-RT-PCR allows correct calculation of the number of
mRNA molecules in a sample. This new method is based on the fact that up and
downstream oligonucleotide primers will compete equally for authentic cellular cDNA
molecules and synthetic template cDNA from the reverse transcription reaction granted
that the synthetic template contains the identical complimentary sequences found in the
original cellular mRNA. Therefore, a competition will be established. The log of the ratio
of band intensities within each lane is plotted against the log of the copy number of
template added per reaction. When the ratio of template and target intensities is equal to
one, the quantity of target messages is determined. A correct standard curve can be
considered a strong argument for equal rates of amplification (Raeymaekers, 1994;
The ECM supertemplate (Tamrnuzzer et al., 1996) was diluted in known
concentrations of 2.16 x 106 2.16 x 1010 copies and used in reverse transcriptase
reactions to compete with the gene of interest. For the initial RT step, 1 il of the diluted
supertemplate and l ug of total RNA were added to each reaction and cDNA was
synthesized in a series of standard 20utl reactions each containing 1.5 mM MgC2, 200
iM dNTPs, 200 U of Reverse Transcriptase, 50 U/ml of RNase inhibitor, 2.5 jIM of
oligo d(t) primer, 1 gl of the diluted supertemplate and 1 gg of total RNA. The reaction
was carried out at 42C for 15 min, 99C for 5 min and 5C for 5 min. Polymerase chain
reaction was performed for each ECM gene for DNA amplification in 100 p1 of total
reaction containing 10 igl the initial RT reaction, 50 pmol of each primer 5'and 3', 1.5
mM MgCl2, 200 [tM dNTPs and 2.5 U of Taq polymerase. This reaction was carried out
at 94C for 4 min for initial hot start, 94C for 1 min, 58C for 1 min and 72C for 1 min
for 35 to 50 cycles and 72C for the final 7 min of extension in a Biometra thermocycler.
The following table shows the expected product sizes for each gene.
Table 3: ECM Supertemplate, expected product sizes.
ECM Template Cellular mRNA
_______ (bp) (bp)
Fibronectin 347 768
Laminin B2 344 677
Collagen III 338 522
Collagen I 335 769
Collagen IV 335 743
Laminin BI 335 594
Elastin 338 704
Analysis of RNA Results
Gels containing the PCR products were photographed and scanned. The intensity
of staining was determined using NIH Image program version 1.54. These intensity
values were normalized for their molecular weight. The log of the ratio of the band within
each lane was plotted against the log of the copy number of template added per reaction.
Quantity of target messages was determined where the ratio of template and target band
intensities equals one (Tarnuzzer et al., 1996).
Statistical Analysis of Q-RT-PCR Data
All data were statistically analysed by both Kruskall-Wallis one-sided ANOVA.
The null hypothesis stated that all means were equal while the alternative hypothesis
stated that at least two of the means were different. Data were considered statistically
significant atp < 0.05.
Embryonic Heads Culture
Embryonic heads from gestational days 9, 10 and 10.5 were cultured in 24-well
dishes using Transwell-Clear permeable supports (6.5mm, 3.0 [am) with BGJb medium
supplemented with 0.1 mg/ml ascorbic acid, 10 mM Hepes, penicillin-streptomycin, pH
7.4 (Slavkin et al., 1989; Shum et al., 1993; Chin and Werb, 1997) and 20% fetal bovine
serum. Cultures were maintained from 9 to 12 days at 37C and 5% carbon dioxide with
media changed every 48 hs. Working dilutions of all inhibitors and comparable controls
(dimethyl sulfoxide (DMSO), methylcellulose in phosphate buffer (PBS), H20, TIMP-2
buffer and medium alone) were made in culture medium. The following inhibitors were
added to the culture: 1,10 phenanthroline (150 p1 of solutions of 400mM); galardin (15
.tg/ml, 150 [g/ml and lmg/ml) (Boghaert et al, 1994; Odake et al, 1994); EDTA (ImM
and 400 mM); TIMP-2 (0.02 [tg/ml, 0.05 ug/ml, 0.1 ug/ml, 0.2 utg/ml, 0.5 tg/ml and 1
tg/ml) (Albini et al, 1991); and aprotinin (1.5 tg/ml and 3.0 tg/ml).
All experiments were performed at least three times on separate occasions with
five embryonic heads of gestational ages 10 or 10.5 for each treatment. The progress
towards fusion of facial processes was documented using a Nikon camera attached to an
Sectioning of Tissues
Tissues from culture experiments were fixed in vials with 10% buffered formalin
for up to four hours at 4C. Using an automated preparation device, tissues were
dehydrated in graded series of alcohol, cleared in xylene, infiltrated and embedded in
paraffin. The paraffin molds were left overnight in the -20C freezer to harden. Paraffin
blocks were serial-sectioned at 5-8 Jtm of thickness. Sections were transferred to subbed
slides and placed on a slide warmer at 45C to stretch the sections for at least 1 hour.
Slides were stained with haematoxylin/eosin or stored for subsequent use in
Extracellular Matrix Molecules Are Being Synthesized and Degraded at the Time of
Facial Formation In Vivo
Immunoblots were performed to identify the extracellular matrix composition of
the facial processes (Fig. 2). These demonstrated the presence of basement membrane
molecules such as collagen IV, fibronectin and laminin in all samples for gestational days
10 and 11. For collagen IV, densitometric analysis of western blots showed an increase of
2.5 fold in protein expression for mandibular process from day 10 to day 11. In contrast,
a decrease of 1.2 to 1.5 fold of collagen IV in maxillary, lateral and medial nasal
processes from day 10 to day 11 was observed. However, immunoblots for fibronectin
and laminin demonstrated a slight increase in the expression of these proteins from day
To further correlate the presence of specific protein with message RNA synthesis,
competition-based quantitative RT-PCR (Tarnuzzer et al., 1996) was performed. Using
the calculations described in the Materials and Methods section, the number of steady-
state copies per cell of each ECM gene was calculated. A high number of ECM mRNA
copies per cell were detected for maxillary, lateral and medial nasal processes of
gestational day 11. Collagen I message was absent in all samples for both days. The null
hypothesis which states the means are equal on all days analyzed can be rejected, as p<
0.05 for collagens III and IV, for laminin B1 and for fibronectin. For collagen III, there
was a steady state increase of 2.8 fold in message copies/cell for maxillary process from
day 10 to 11. For laminin B 1, the increase was even higher, 5.7 fold. However, only a
slight increase in production of mRNA of ECM components from samples of day 10 to
day 11 maxillaryy and mandibular processes). For collagen III, there was a steady
production of message for fibronectin and collagen IV was observed from gestational day
10 to day 11. The number of copies of mRNA per cell detected in this study was similar
to other studies for different tissues (Shim et al., 1997; Macauley et al., 1997). These
results are summarized in Figs 3, 4, 5, 6 and 7.
10 T- GD
* -. p
Cn (n v
a- E ?2
I, ~ ~250
E- _. E
GD 10 -I--
V co .
c' x E
in E E E
E E E
Fig 2. Identification of ECM composition of the facial processes, demonstrating the
presence of basement membrane molecules such as collagen IV, fibronectin and laminin
in all samples for gestational days 10 and 11. For fibronectin, a band at 250 kDa was
detected and co-migrated with human purified plasma fibronectin. Laminin blots
demonstrated presence of bands at 220 and 400 kDa, presumably A and B chains of
laminin protein. Mouse EHS tumor cells purified laminin was used as positive control.
For collagen IV, immunoblots revealed bands at 68 and 85 kDa. Densitometric analysis
of western blots showed an increase in collagen IV protein expression for mandibular
process from day 10 to day 11 and a decrease in maxillary, lateral and medial nasal
processes from day 10 to day 11. Immunoblots for fibronectin and laminin demonstrated
a slight increase in the expression of these proteins from day 10 to 11. The molecular
weight (kDa) is indicated in the side of the panel. Molecular weight standards are: 111
kDa, Phosphorylase B; 77 kDa, Bovine serum albumin; 48.2 kDa, Ovalbumin; 33.8 kDa,
Collagen IV Fibronectin Collagen III Laminin B1
max 11 max 11
Fig.3. Representative composite data for quantitative RT-PCR. The products size are the
same as listed in table 3. The lanes of each agarose gel contain RT-PCR products
generated with decreasing numbers of pMATRIX template moving left to right where the
upper bands represent the gene and the lower bands represent the template. All reactions
were made on three separate preparations of total RNA.
R E I^^^
Fig.4. Steady state levels of mRNA copies per cell of facial processes for ECM genes
measured by Q-RT-PCR. For laminin B1, there was an increase of 5.7 fold in message
copies/cell for maxillary process from day 10 to 11. Similar numbers were obtained using
the same technique to investigate pre-implantation rat uterus (Shim et al., 1997). All
reactions were made on three separate preparations of total RNA. Values are expressed as
means of three determinations + Standard Error Mean (SEM). SEM values for mand 10,
5 x 106; for mand 11, 2 x 106; for max 10, +1 x 106; for max 11, 4 x 106; for lateral,
2 x 106; for med nasal, 3 x 106.
-- 20 0 11 \0
10 e .........
E mand mand max 10 max 11 lateral med
10 11 nasal
mand 10 mand 11 max 10 max 11 lateral med nasal
Fig.5. Steady state levels of mRNA copies per cell of facial processes for ECM genes
measured by Q-RT-PCR for Collagen Ul. An increase in the steady state levels of 2.8
fold in message copies/cell was observed for maxillary process from day 10 to 11. All
reactions were made on three separate preparations of total RNA. Values are expressed as
means of three determinations SEM. SEM values for mand 10, 3 x 106 ; for mand 11,
8 x 106; for max 10, +1 x 106; for max 11, +5 x 105; for lateral, +5 x 106; for med nasal,
5 x 106.
i 0 ......^,.,.^^ . ^ ^. ____
mand mand max 10 max 11 lateral med
10 11 nasal
Fig.6. Steady state levels of mRNA copies per cell of facial processes for ECM genes
measured by Q-RT-PCR for Collagen IV. Only a small increase in the steady state levels
of collagen IV was observed. All reactions were made on three separate preparations of
total RNA. Values are expressed as means of three determinations SEM. SEM values
for mand 10, +2.1 x 105; for mand 11, +1 x 105; for max 10, 2 x 106; for max 11, 3.1 x
106; for lateral, +1.1 x 106; for med nasal, +1 x 106.
Fig.7. Steady state levels of mRNA copies per cell of facial processes for ECM genes
measured by Q-RT-PCR for Fibronectin. Only a small increase in the steady state levels
of fibronectin was observed. All reactions were made on three separate preparations of
total RNA. Values are expressed as means of three determinations SEM. SEM values
for mand 10, 2 x 107; for mand 11, 2 x 107; for max 10, 2.7 x 107; for max 11, +8 x
107; for lateral, 2.7 x 107; for med nasal, 6.5 x 107.
Matrix Metalloproteinases and Tissue Inhibitors Expression Do Not Show
Major Temporal Regulation During Facial Morphogenesis In Vivo
Zymograms of gelatin gels showed that all facial processes isolated from the three
different gestational ages (10, 10.5 and 11 days) demonstrated a major 58 kDa band and
minor bands at 50, 60, 90 and 100 kDa. These appear to be-constitutively expressed. An
additional minor band at 55 kDa was not seen in gestational day 10 facial processes, but
appeared in all tissues from day 11. Bands at molecular weights of 50, 58 and 60 kDa had
increased protease activity by 1.6 fold as development progressed (from day 10 to day
11), as indicated by densitometric analysis. Other minor bands appeared in tissues of day
11 at molecular weights of 28, 55/52, 72 and 120 kDa (fig. 8).
All gelatinases were inhibited by incubating samples with 4 mM EDTA or 1,10-
phenanthroline prior to electrophoresis, demonstrating that gelatinolytic activity was due
to matrix metalloproteinase activity. Twenty mM of serine protease inhibitor,
phenylmethylsulfonyl fluoride (PMSF), did not affect gelatinolytic activity when
incubated with samples prior to electrophoresis.
c ,s ?' c ~5u
.. fl .I o
- -o -= C3 .^ 'o*^
7- T >- -
E E- E E
Fig.8. Matrix metalloproteinases in facial processes of gestational days 10 and 11.
Zymography of gelatin gels with facial processes revealed proteolytic bands with major
activity at 58 kDa and minor bands at 50, 60, 90 and 100 kDa. These appear to be
constitutively expressed. An additional minor band at 55 kDa was not seen in gestational
day 10 facial processes, but appeared in all tissues from day 11. Bands at molecular
weights of 50, 58 and 60 kDa had increased protease activity by 1.6 fold as development
progressed (from day 10 to day 11), as indicated by densitometric analysis. Other minor
bands appeared in tissues of day 11 at molecular weights of 28, 55/52, 72 and 120 kDa
Purified MMP-2 and MMP-9 (Chemicon) were used as positive controls and co-migrated
with the samples. A 28 kDa band was only seen on day 11 samples. The molecular
weight (kDa) is indicated in the side of the panel.
The zymographic results for gelatin gels are summarized in Tables 4 and 5.
Table 4: Summary of zymographic results for gelatin gels.
Day 10 Day 10.5 Day 11
T0 S. 0 >B -O S T a T ^ T 0 pi T3 h 0 PL
o ... 0 = o _
bands X0 0 |
CD C-DT C 'D M a"S C& Cr r CD CD 0 CD 1 Ct *y
(kDa) | _
120 + + + +
100 + + + + + + + + +
90 + + + + + ++ + + +
72 + -
60 ++ + ++ + + ++ ++ ++ ++
58 +++ ++ +++ +++ +++ ++++ ++++ ++++ ++++
55/52 -+ + + +
50 + + + + + ++ ++ ++ ++
28 + + + +
v aiyiig udgIees of autiuvity: u- iuI udnu, -- visioib aCuIviLy, -r aCuViVLy, -'---r-- -
very strong activity.
Four-aminophenylmercuric acetate (APMA) was used for activation of gelatinases.
On gelatin gels, an active band at 120 kDa that appeared in all processes of day 11 could
not be detected after activation with APMA. Also, a 55 kDa band had the most
gelatinolytic activity with minor bands at approximately 58 kDa. Gelatinases activated
with APMA (Fig. 8) had a greater increase in gelatinolytic activity than the non-activated
bands as detected by densitometric analysis, indicating organomercuric activation.
Fig.9. Matrix metalloproteinases in facial processes of gestational days 10 and 11.
Zymography of gelatin gels with facial processes activated with APMA revealed
proteolytic bands with major activity at 55 to 58 kDa range. The molecular weight (kDa)
is indicated in the side of the panel.
Table 5: Summary of zymographic results for APMA activated samples on gelatin gels.
Day 10 Day 11
bands c '
CD CD~ CD CD 0Z-0
100 ++ + +
90 ++ + + +
60 + + + +
58 + + ++ ++ ++ ++
55 ++ ++ +++ +++ +++ +++
28 + + + +
Varying degrees of activity: + = faint band, ++ = visible activity, +++ = activity, ++++ =
very strong activity.
Zymography on casein gels did not demonstrate any bands using the procedures
described above, but revealed a faint 28 kDa active band when using the Novex system
(San Diego, CA) which utilizes a 4-16% Tris-Glycine gel with blue-stained beta casein
incorporated as substrate. However, casein-plasminogen gels revealed proteinase bands
with molecular weights at 54 kDa and at 40/38 kDa (Fig. 10). No difference in activity
was observed between different ages. All plasminogen dependent caseinases were
inhibited by 1 p.g/ml of aprotinin, a specific inhibitor of serine proteases, added to
samples and incubated at 37C for 30 min prior to incubation.
Reverse zymograms were performed using collected conditioned media at 23.5%
(vol) in 10% SDS-PAGE gels. All tissues demonstrated a 28 kDa band compatible with
TIMP-1 with multiple high molecular weight bands suggesting inhibitor bound to
proteases (Fig 11).
To further identify the MMPs present in the facial processes, Western blots were
performed (Fig 12). With antibody to MMP-2 (IgG-antipeptide antibody), a major band
at 58 kDa was detected. The molecular weight of this band corresponds with the most
active degradation bands from the zymograms. Although this band is immunostained
with MMP-2 antibody, its identity remains to be elucidated. The results here indicate it to
be a differentially glycosylated form of the activated enzyme, a degraded form of MMP-2
or it might be a distinct enzyme altogether. Other bands are seen in the blot at different
molecular weights (28 kDa, 80 kDa and 100 kDa) representing active degradation
products at lower molecular weights and differentially glycosylated forms at higher
molecular weights. The 58 kDa band was not stained by any other antibodies (i.e. MMP-
3 and MMP-9) tested by Western immunoblotting. Similarly, controls for antibody
specificity comprised of no incubation with the primary antibody demonstrated no bands.
Western blots for MMP-9 demonstrated multiple bands with the major reactivity
at 90 kDa. Additional bands represent variants in glycosylation of the enzyme. Human
placenta and purified human MMP-2 and MMP-9 (Chemicon, Temecula, CA) were used
as positive controls for each blot. Negative controls were performed with secondary
antibody only. Immunoblots probed with antibody to MMP-3 demonstrated the presence
of a major band at 50 kDa.
Reverse transcriptase-polymerase chain reaction (RT-PCR) results indicated the
presence of MMP-2, MMP-3, MMP-7 and MMP-9 mRNA but not MMP-1. All primers
have been tested against other tissues such as salivary glands, which demonstrate their
specificity, and through DNA sequence analysis of amplicon products. The identity of the
PCR products was confirmed by using the Southern blot technique, which demonstrated
the presence of the authentic amplicons for matrix metalloproteinases, MMP-2, MMP-3,
MMP-7, MMP-9, the natural inhibitors, TIMP-1 and TIMP-2, and the their activators,
uPA and tPA. These appear to be constitutively expressed since no major differences in
band intensities were found between the different gestational days relative to G3PDH or
P3-actin expression using densitometric analysis (fig.13). An example of a RT-PCR
reaction in an agarose gel and its corresponding southern blot is shown in Fig. 14.
Fig. 10. Presence of Serine proteases in facial processes of gestational days 10 and 11.
Zymography of casein-plasminogen gels with facial processes revealed proteolytic bands
with major activity at 54 and 40 kDa. These bands appeared to have same level of
activity. The molecular weight (kDa) is indicated in the side of the panel.
F"GDl I.0 GD11 ---
.2 cO 3 2 CO
4 = -= 1 3 = ct
C x C X
-0 co 3 (z a-
SE E E E c" E
48-- : ..t.
Fig. 1. Reverse zymography was performed using collected conditioned media at 23.5%
(vol) in 10% SDS-PAGE gels. All tissues demonstrated a 28 kDa band compatible with
TIMP-1. The molecular weight (kDa) is indicated in the side of the panel.
- GD 10----
-u co c'3 c3
n E E E
48.2 58 A
- E) C
GD10 --O-- GD11
Fig.12. Presence of MMPs in the facial processes at the time of facial development, a)
immunoblots for MMP-2 revealed a major band at 58 kDa. This corresponds with the
most active degradation bands from the zymograms. Additional bands represent active
degraded forms of the enzyme at lower molecular weights and differentially glycosylated
forms at higher molecular weights. Purified MMP-2 (Chemicon) and human placenta
were used as positive controls, and bacterial collagenase IV as negative control, b)
western blots for MMP-9 demonstrated multiple bands with the major reactivity at 90
kDa. Additional bands represent variants in glycosylation of the enzyme. Human placenta
and purified human MMP-9 (Chemicon, Temecula, CA) were used as positive controls.
Molecular weight standards are: 111 kDa, Phosphorylase B; 77 kDa, Bovine serum
albumin; 48.2 kDa, Ovalbumin; 33.8 kDa, Carbonic anhydrase.
Fig. 12cont. c) immunoblots probed with antibody to MMP-3 demonstrated the presence
of a major band at 50 kDa. The molecular weight (kDa) is indicated in the side of the
panel. Negative controls were performed with secondary antibody only (data not shown).
GD10 -T GD11
Fig. 13. Expression of MMPs, TIMPs and PAs in the facial processes at the time of facial
formation. Example of RT-PCR reaction with corresponding southern blot.
Fig. 14. Representative composite data for RT-PCR. Results demonstrate the presence of
MMP-2, MMP-3, MMP-7, MMP-9, TIMP-1, TIMP-2, uPA and tPA but not MMP-1
message. These appear to be constitutively expressed since no differences between the
different gestational days to internal controls (3-actin or G3PDH) were found after
no 4M t
Inhibition of Matrix Metalloproteinases Alters Facial Development In Vitro
In order to determine the role of MMPs in facial morphogenesis, a number of
known MMP inhibitors were used to block proteinase function during in vitro
development of mouse embryonic heads.
In normal facial development, the lateral and medial nasal processes on both sides
of the developing nose make contact with each other and with their respective maxillary
process at gestational day 11. Epithelial adhesion and later fusion occur at the inferior
part of the nasal groove. Subsequently, this nasal fin disappears as adhered epithelial cells
of the apposed lateral and medial nasal processes die or transdifferentiate into
mesenchymal cells, basal lamina is degraded and a mesenchymal bridge between the
processes that allows tissue continuity is formed (Diewert and Shiota, 1990). At mouse
gestational day 13, the nose and upper lip are completely formed.
In vitro culture embryos treated with galardin, exhibited susbtantial deficiencies
in morphogenesis. Generally, the embryonic heads had the same size, suggesting normal
growth with defective or delayed differentiation. Inhibition was apparent at a range of
concentrations beginning with 15 4g/ml of galardin, the lowest dose, as well as at 1
mg/ml, the highest dose. Galardin has been used in different systems as an inhibitor of
MMPs and has been demonstrated to have minimal toxic effects (Schultz et al., 1992;
Broverman et al., 1998).
Effects were also detected when a single dose of 150 tg/ml of galardin was
administered to the cultured media for the first three days of culture only. Galardin or
phenanthroline (10mM) specifically inhibited the fusion of lateral and medial nasal
processes, although the maxillary process appears to fuse to the lateral process. At
gestational days 10+7 (7 days in culture), controls demonstrated complete facial
formation and closure whereas explants cultured with galardin or phenanthroline showed
a noticeable cleft resulting from the non fused lateral and medial nasal processes (Fig. 15).
Nonetheless, mandibular development did not seem affected, although it was delayed.
Histological examination revealed that lateral and medial processes remained
close but did not fuse maintaining their structure with mesenchymal tissue encased by a
layer of epithelial cells (Fig. 16).
In contrast, embryo heads cultured with EDTA (lmM and 400 mM) exhibited
gross morphological changes. The tissues displayed a gelatinous and dispersed
organization that worsened as the time in culture progressed, culminating in total disarray
of the head form.
Furthermore, embryo heads cultured in the presence of aprotinin (1.5[tg/ml and
3.0jtg/ml), a specific serine protease inhibitor, did not exhibit any morphological
To further explore the role of MMPs activity and to identify the major proteinases
of facial morphogenesis, a series of different concentrations of TIMP-2 (0.02 utg/ml, 0.05
utg/ml, 0.1 ptg/ml, 0.2 [ig/ml, 0.5 ug/ml and 1 u.g/ml) (Albini et al, 1991), a natural
inhibitor of MMP-2, were added to the media culture as described before. Although the
embryo heads displayed features corresponding to arrested development, there were no
apparent effects on nose development. Overall, the heads were small, approximately 50%
of the controls size, and had a ball like shape, unlike the controls. The brain and the
forehead were particularly smaller in size compared to controls and to other inhibitor
experiments, as were the lateral and medial nasal processes (Fig. 17).
GD 10+5 GD 10+7
Fig. 15. Use of MMP inhibitors in mouse embryonic head culture affected facial morphonesis. Galardin (15 ig/ml, 150 tg/ml and
I mg/ml) or phenanthroline (150 pil of solutions of 400mM) specifically inhibited the fusion of lateral and medial nasal processes. At
gestational days 10+7 (7 days in culture), controls demonstrated complete facial formation and closure whereas experiments with
galardin or phenanthroline showed a noticeable cleft resulting from the non fused lateral and medial nasal processes. All experiments
were performed at least three times on separate occasions with five embryonic heads of gestational ages 10 or 10.5 for each
treatment. Mand mandibularr process), max maxillaryy process), lat (lateral process), med nasal (medial nasal process).
k med ^^
A B C
med cleft i
Fig.16. Hematoxylin-eosin stained frontal sections of mouse head embryos demonstrating the effect of MMP inhibitors on facial
development: A) growth of lateral and medial nasal processes prior to fusion; B) controls demonstrated complete fusion after 9 days
of culture whereas explants using galardin showed a cleft with defined separation of the processes. Epithelial layer and mesenchyme
are noticeable at both sides of the cleft; C) lower magnification of fusion sites. Mand, mandibular process; max, maxillary process;
lat, lateral process; med nasal, medial nasal process.
Fig. 17. Effects of excess of TIMP-2 to culture of mouse embryo heads. The heads displayed features corresponding to arrested
development but there were no apparent effects on nose development. Compared to the controls, experiment mouse embryo heads
with TIMP-2 excess were small and had a ball like shape. A reduction of approximately 50% in size was observed. The brain and the
forehead were particularly affected compared to controls and to other inhibitor experiments, as were the lateral and medial nasal
processes. Dosage used was 0.02 utg/ml, 0.05 jg/ml, 0.1 pg/ml, 0.2 pg/ml, 0.5 pg/ml and 1 pg/ml. Each dose was used for at least 5
different embryo heads and repeated 3 times at different occasions.
This study shows that a number of matrix metalloproteinases, their inhibitors and
activators are expressed and present at the time of murine facial morphogenesis. Matrix
metalloproteinases promote the degradation of extracellular matrix necessary for fusion
of the facial processes. Normal growth and development of facial processes into mid-face
and upper lip require cell proliferation and migration influenced by cell-extracellular
matrix interactions and by epithelial-mesenchymal interactions regulated by signals that
cross the basement membrane. Facial development also requires ECM synthesis and
degradation performed by MMPs and mediated by the presence of inhibitors and by
The basement membranes of the lateral and medial nasal processes are degraded
at the sites of fusion and replaced by a continuous mesenchymal tissue. Alterations of cell
proliferation, cell migration or death, expression of certain genes such as homeobox
genes, growth factors or ECM components may contribute to the appearance of cleft lip
and palate. The results of this study suggest that facial morphogenesis is, in part,
dependent on MMPs activity.
Expression and Degradation of ECM Components by MMPs in Developing Facial
Development and growth of the facial processes involve expression of
extracellular matrix molecules. The distribution of basement membrane components in
the mouse primary palate at the time of fusion has been reported recently (lamaroon and
Diewert, 1996). It has been demonstrated that a rapid degradation of the facial processes
basement membrane and its components, laminin, collagen type IV and fibronectin,
occurs at the time of completion of facial morphogenesis (lamaroon and Diewert, 1996).
Our investigation shows the presence of the same ECM components in the developing
murine facial processes. The detectable amounts of collagen IV are decreased in
maxillary, lateral and medial nasal processes of gestational day 11 compared to
gestational day 10, suggesting degradation of basement membrane of these processes
prior to fusion.
Quantitative RT-PCR data demonstrated that an increase in steady state mRNA
levels of ECM genes occurs during facial development. The number of mRNA copies per
cell obtained within the facial processes are similar to those of other studies using the
same technique, Q-RT-PCR, but in different tissues (Macauley et al., 1997; Shim et al.,
1997). Facial morphogenesis in mice occurs over the course of 3 days, therefore, the need
for an increased production of structural proteins such as collagen III from gestational
day 10 to 11 is not unexpected. At the same time, the increased steady state levels of
mRNA copies per cell for collagen IV and fibronectin was much smaller. These two
genes, along with laminin, have been found primarily in the basement membrane of facial
processes (lamaroon and Diewert, 1996), while fibronectin can also be found in the
mesenchyme. Since basement membrane is being degraded prior to fusion, we can
speculate that there is little need for more of these molecules. However, laminin BI is
being produced in high numbers and an increase of its expression would be expected
during highly proliferative stages due to its suggested roles in development. Laminin has
additionally been implicated in cell migration, in regulating gene expression (Streuli et
al., 1991) and has growth-like properties (Adams and Watts, 1993), therefore a
modulation of its steady state concentration of mRNA potentially implies laminin as
having a regulatory role in facial development.
Although the levels of ECM mRNA increased, MMP message steady state
concentrations appear to remain constant during facial formation. Expression of MMPs
and TIMPs have been investigated in many developing processes, including the
craniofacial complex (Chin and Werb, 1997; lamaroon et al., 1996; Werb, 1997). Diffuse
labelling of matrix metalloproteinase-9 message was found by in situ hybridization in
mandibular processes from gestational days 10 to 13 when it becomes more intensified in
the mesenchyme surrounding the tooth buds (Chin and Werb, 1997). The same study
demonstrated that MMP-3 was diffusely expressed in gestational days 9 to 15 mandibular
processes. They also reported increased amounts of mRNA for MMP-2 as detected by
RT-PCR. This study used RT-PCR methods where reactions were normalized to equal
amounts of input total RNA and did not have an internal control. We report otherwise,
according to our results, the message levels remained virtually constant as detected by
RT-PCR where reactions were performed with same amounts of input total mRNA and
were compared to positive controls (G3PDH or P3 actin) in the same reaction. Other
studies have demonstrated the presence of MMP-2 in regions of the nasal and facial
prominences, with higher intensity in the zones of fusion of the lateral, medial nasal and
maxillary processes (Iamaroon et al., 1996). In our studies, we not only detected mRNA
for MMP-2, but also for MMP-3, MMP-7 and MMP-9. Zymographic evidence and
Western blotting demonstrated that the proteolytic activity was possibly due to MMP-2,
MMP-3 and MMP-9. Although the 58 kDa band was only immunostained with MMP-2
antibody, our results were inconclusive and we are uncertain of its identity. We can
speculate of it being a differentially glycosylated form of the activated MMP-2 enzyme, a
degraded form of MMP-2 or it might be a distinct enzyme altogether. Furthermore, the
identity of the protease band relative to MMP-7 (28kDa) could not be confirmed due to
the lack of antibodies.
Although some temporal differences in the MMPs appearance were found in the
zymographic study, these differences were not reproduced in the Western blotting or the
RT-PCR studies. It is likely that these differences reflect differential activation of MMPs,
and differential activation and/or inhibition may be the modes of regulation of ECM
degradation in facial morphogenesis. Although MMP-2 appears to be constitutively
expressed during embryogenesis, there is accumulating evidence that other MMPs are
expressed only when their activity is required. Our RT-PCR results demonstrate
otherwise. Further work, probably using in situ hybridization, may elucidate these
Matrix metalloproteinase-1 message was not found to be expressed during facial
development. At the same time, collagen I mRNA was not detected in the facial
processes by Q-RT-PCR. The major ECM substrates for MMP-1 are the interstitial
collagens, among them, collagen type I. Considering the relation between enzyme and
substrate, we can speculate that with the absence of collagen I, there is no need for the
expression of MMP-1. Many studies have shown that MMP-1 is not expressed until
gestational day 14.5 or 15 in mouse embryos (Gack et al., 1995; Mattot et al., 1995). Its
expression can be correlated with the onset of bone or cartilage formation.
Matrix Metalloproteinase Degradation of the Basement Membrane Is Necessary for
the Fusion of the Facial Processes
Since we did not detect major temporal changes in the MMPs activity of facial
processes, the question became how essential are these enzymes for proper craniofacial
development. Correct facial morphogenesis relies on sufficient growth of the facial
processes with subsequent fusion in a time dependent manner. The effects of MMP
inhibitors on cultured mouse embryo heads indicate that, with inhibition of enzyme
activity, fusion of lateral and medial nasal processes was retarded. However, overall
cranial growth remained unaffected. The normal fusion of these processes requires
degradation of basement membrane ECM molecules by allowing tissue continuity after
cell migration. These migrations consequently result in upper lip and mid-face formation.
Galardin specifically inhibited the fusion of lateral and medial nasal processes, whereas
inhibitors for serine proteases did not exhibited the same effect, suggesting that MMPs
are involved directly in this occurrence. The cleft resulting from the blocked fusion
between the two processes has the basic characteristics of cleft lip (unilateral or bilateral),
a common congenital defect. This data suggests that degradation of ECM by MMPs is a
continuous process in facial development of the mouse. The observations here also
indicate that MMP activity is necessary for proper formation of the mid-face.
Furthermore, genetic or environmental perturbations of these enzymes may contribute to
the observed birth defect, cleft lip/palate.
Inhibition of facial morphogenesis, however, was not achieved by blocking MMP
activity with TIMP-2. Since we detected the presence of MMP-2 and because of its
suggested developmental role in other studies and its ability to degrade many ECM
substrates, we initially hypothesized its having a major role in facial development.
Because of TIMP-2's known inhibitory properties against MMP-2, by comparing both
phenotypic results we could correlate whether the inhibition effects seen with galardin
were also due to inhibition of MMP-2 activity. The results obtained here were
inconclusive. Although TIMP-2 excess clearly affected overall cranial development
reducing the size of the explants by approximately 50%, the nose seemed to develop
normally. Some studies have suggested that it is necessary to achieve >90% inhibition of
MMP activity to see morphological effect (Chin and Werb, 1997). At this gestational
time point, the amount of MMP-2 in the facial processes is not known. Furthermore, the
role of TIMP-2 and its interaction with MMP-2 are still being investigated. Recent
reports have demonstrated that TIMP-2 inhibits the activation of pro-MMP-2 while
Others suggest that it may facilitate its activation through the formation of a compl.N with
MT1-MMP (MMP-14) on the cell surface or to another TIMP-2 acting as a receptor.
Recent reports of a TIMP-2 targeted mutation mouse suggest that TIMP-2 is necessary
for the activation of MMP-2 (Caterina et al., 1998). Homozygous TIMP-2 knock-out
mice only produced pro-MMP-2, whereas heterozygous produced both forms of the
enzyme, pro and activated. It is also thought that MMP-2/TIMP-2 complex may still be
active but the level of gelatinase activity detected is only 10% of the free active MMP-2
(Kleiner et al., 1992; Yu et al., 1996).
It is interesting to speculate as to why many different types of matrix
metalloproteinases (MMP-2, MMP-3, MMP-7 and MMP-9) with similar substrate
specificities are expressed at the same time during morphogenesis. Results of transgenic
mouse experiments with knock-outs for MMP-3, MMP-7, MMP-9, MMP-12 and TIMP-2
and overexpression of MMP-1 and MMP-3 were disappointing (Shapiro, 1997; Caterina
et al., 1998). No dramatic phenotypes were produced in the analysis of embryonic
developmental patterns. These results seem to give an impression of the redundancy of
MMPs function. One or more enzymes may take over and compensate for each MMP
knocked out. Potentially this is the case when TIMP-2 excess is used as an inhibitor of
facial morphogenesis. However, when a broad spectrum inhibitor is used such as
galardin, a variety MMP activity may potentially be blocked to some extent. The
characterization of their genes have demonstrated that each enzyme may have a different
control of expression (Huhtala, Chow and Tryggvason, 1990; Huhtala et al., 1991). By
responding to different elements (for example, growth factors or the presence of
inhibitors), the cell can regulate the expression of the different enzymes and change its
It has been proposed that fusion of the hard palate is time-critical (Smiley, 1972).
If the palatal shelves meet after the critical period for fusion, fusion will not take place.
The same might also be true for facial morphogenesis, if the facial processes are not
degraded in the critical time window, fusion will not occur. Therefore, the production of
a high number of enzymes at the time of fusion is justified where one class of enzymes
can be substituted by another similar class but the absence or inhibition of a great number
of different metalloproteinases can not be overcome. Perhaps, if degradation does not
occur at the right time, other mechanisms such as apoptosis and cell migration may also
be affected or prevented from occurring. This might explain why such a localized effect
such as cleft lip and palate can occur in an otherwise healthy individual.
To date, detailed knowledge of how MMPs bind to their inhibitors and how the
cell regulates the activation of MMPs remain to be clarified. Although a lot of work has
been done in this field, MMPs are still being "discovered" (Pendas et al., 1997) which
demonstrates how exciting and interesting it has been and yet has to come.
This study indicates that MMPs are involved in facial morphogenesis by
promoting migration of undifferentiated mesenchyme through degradation of the
basement membrane prior to fusion of facial processes. A potential mechanism by which
the mesenchymal bridge was prevented from occurring was by either inhibition of
epithelial cell migration or apoptosis at the sites of fusion. The widespread expression
and activity of MMPs suggest that other regulatory events may control local proteolysis
therefore making the extracellular environment a dynamic one. We have developed a new
approach that will be useful for investigating similar remodeling events in vivo or in
vitro. The ability of in vitro culture experiments to generate a cleft lip phenotype
identifies the matrix metalloproteinases as potentially a key component of proper facial
morphogenesis. Fusion of the facial processes may also be time-critical. If degradation of
extracellular matrix at the sites of fusion is inhibited for a period of time, fusion may not
occur even at later stages. Such a critical event may require the activity of a large number
of enzymes that may compensate for others if necessary.
Adams J, Watts F. Regulation of development and differentiation by the extracellular
matrix. Development 1993 117: 1183-98.
Adler R, Brenner C, Werb Z. Expression of extracellular matrix-degrading
metalloproteinases and metalloproteinase inhibitors is developmentally regulated during
endoderm differentiation of embryonal carcinoma cells. Development 1990 110: 211-20.
Albini A, Melchiori A, Santi L, Liotta L, Brown P, Stetler-Stevenson W. Tumor Cell
Invasion Inhibited by TIMP-2. J Natl Cancer Inst 1991 83: 775-79.
Andersen H, Matthiessen H. Histochemistry of the early development of the human
central face and nasal cavity with special reference to the movements and fusion of the
palatine processes. Acta Anat 1967 68:473-508.
Ardinger H, Buetow K, Bell G, Bardach J, VanDemark D, Murray J. Association of
genetic variation of the transforming growth factor alpha gene with cleft lip and palate.
Am J Hum Genet 1989 45: 348-353.
Ayad S, Boot-Handford R, Humphries M, Kadler K, Shuttleworth C. The Extracellular
Matrix. FactsBook. 1994, Harcourt, Brace, and Jovanovich, Academic Press, New York.
Bailey L, Minkoff R, Koch W. Relative Growth Rates of Maxillary Mesenchyme in the
Chick Embryo. J Craniofac Gen Dev Biol 1988 8:167-77.
Behrendtsen 0, Alexander C, Werb Z. Metalloproteinases mediate extracellular matrix
degradation by cells from blastocyst outgrowths. Development 1992 114:447-56.
Bernfield M, Banerjee S, Koda J, Rapraeger A, Remodeling of basement membrane as a
mechanism of morphogenetic tissue interaction. In: Trelstad R (eds): The Role of
Extracellular Matrix in Development. 1984, Liss, New York, 545-72.
Birkedal-Hansen H, Moore W, Bodeden M, Windsor L, Birkedal-Hansen B, DeCarlo A,
Engler J. Matrix metalloproteinases: A review. Crit Rev Oral Biol Med 1993 4(2):197-
Birkedal-Hansen H. Proteolytic remodeling of extracellular matrix. Curr Op Cell Biol
Boghaert E, Chan S, Zimmer C, Grobelny D, Galardy R, Vanaman T, Zimmer S.
Inhibition of collagenolytic activity relates to quantitative reduction of invasion in vitro in
a c-Ha-ras transfected glial cell line. J Neurooncol 1994 21(2): 141-150.
Brenner CA, Adler RR, Rappolee DA, Pedersen RA, Werb Z. Genes for extracellular
matrix-degrading metalloproteinases and their inhibitor, TIMP, are expressed during
early mammalian development. Genes Dev 1989 3:848-859.
Brinkley L, Vickerman M. The effects of chlorcyclizine-induced alterations of
glycosaminoglycans on mouse palatal shelf elevation in vivo and in vitro. JEEM 1982
Broverman R, Nguyen K, da Silveira A, Brinkley L, Macauley S, Zeng T, Yamamoto H,
Tamuzzer R, Schultz G, Kerr M, Humphreys-Beher M. Changes in the expression of
extracellular matrix (ECM) and matrix metalloproteinases (MMP) of proliferating rat
parotid acinar cells. J Dental Res 1998 77(7): 1504-14.
Bu C, Pourmotabbed T. Mechanism of activation of human neutrophil gelatinase B. J
Biol Chem 1995 270(31):18563-69.
Burk D. Distribution of glycosaminoglycans in the developing facial region of mouse
embryos. J Craniofac Genet Dev Biol 1983 3:339-349.
Burk D. Morphological effects of streptomyces hyaluronidase treatment on the outgrowth
of the nasal processes in mouse embryos. J Craniofac Genet Dev Biol 1985 5:385-98.
Burk D, Sadler T. Morphogenesis of median facial clefts in mice treated with diazo-oxo-
norleucine (DON). Teratology 1983 27:385-94.
Busiek D, Ross F, McDonnell S, Murphy G, Matrisian L, Welgus H. The matrix
metalloproteinase matrilysin (PUMP) is expressed in developing human mononuclear
phagocytes. J Biol Chem 1992 267(13):9087-92.
Campbell CE, Flenniken AM, Skup D, Williams BRG. Identification of a serum and
phorbol ester-responsive element in the murine tissue inhibitor of metalloproteinase gene.
J Biol Chem 1991 266:7199-7206.
Canete-Soler R, Gui Y-H, Linask K, Muschel R. MMP-9 (gelatinase B) mRNA is
expressed during mouse neurogenesis and may be associated with vascularization. Dev
Brain Res 1995 88:37-52.
Canete-Soler R, Litzky I, Muschel R. Localization of the 92 kd gelatinase mRNA in
squamous cell and adenocarcinomas of the lung using in situ hybridization. Am J Pathol
Cao J, Sato H, Takino T, Seiki M. The C-terminal region of membrane type matrix
metalloproteinase is a functional transmembrane domain required for pro-gelatinase A
activation. J Biol Chem 1995 270(2):801-5.
Carmeliet P, Bouch6 A, De Clercq C, Janssen S, Pollefeyt S, Wyns S, Mulligan R, Collen
D. Biological effects of disruption of the tissue-type plasminogen activator, urokinase-
type plasminogen activator, and plasminogen activator inhibitor-1 genes in mice. Ann
NYAcad Sci 1995 748: 367-81.
Casasco A, Casasco M, Reguzzoni M, Caligaro A, Tateo S, Stetler-Stevenson WG, Liotta
LA. Occurrence and distribution of matrix metalloproteinase-2-immunoreactivity in
human embryonic tissues. Eur J Histochem 1995 39(1):31-38.
Caterina N, Caterina J, Yamada S, Longenecker G, Yermovski A, Shi J, Netzel-Arnett S,
Windsor J, Holmback K, Novas N, Mann M, Engler J, Birkedal-Hansen H. Targeted
Mutation of Murine TIMP-2 Gene. J Dent Res 1998 77: 995.
Chin J, Murphy G, Werb Z. Stromelysin, a connective tissue-degrading
metalloendopeptidase secreted by stimulated rabbit synovial fibroblasts in parallel with
collagenase. 1985 J Biol Chem 260: 12367-76.
Chin J, Werb Z. Matrix metalloproteinases regulate morphogenesis, migration and
remodeling of epithelium, tongue skeletal muscle and cartilage in the mandibular arch.
Development 1997 124: 1519-30.
Collier I, Wilhelm S, Eisen A, Mariner B, Grant G, Seltzer J, Kronberger A, He C, Bauer
E and Goldberg G. H-ras oncogene-transformed human bronchial epithelial cells(TBE- 1)
secrete a single metalloprotease capable of degrading basement membrane collagen. J
Biol Chem 1988 263: 6579-87.
Darab D, Sciote J, Minkoff, Sulik K. Pathogenesis of medial facial clefts in mice treated
with methotrexate. Teratology 1987 36:77-86.
Dickson S, Warburton M. Enhanced synthesis of gelatinase and stromelysin by
myoepithelial cells during involution of the rat mammary gland. J Histochem Cytochem
1992 40: 697-703.
Diehl S, Erickson R. Genome scan for teratogen-induced clefting susceptibility loci in the
mouse: Evidence of both allelic and locus heterogeneity distinguishing cleft lip and cleft
palate. Proc Natl Acad Sci USA 1997 94: 5231-36.
Diewert V, Lozanoff S. A morphometric analysis of human embryonic craniofacial
growth in the median plane during primary palate formation. J Craniofac Genet Dev Biol
Diewert V, Lozanoff S. Growth and morphogenesis of the human embryonic midface
during primary palate formation analyzed in frontal sections. J Craniofac Genet Dev Biol
Diewert V, Lozanoff S, Choy V. Computer reconstructions of human embryonic
craniofacial morphology showing changes in relations between the face and brain during
primary palate formation. J Craniofac Genet Dev Biol 1993 13:193-201.
Diewert V, Shiota K. Morphological observations in normal primary palate and cleft lip
embryos in the Kyoto collection. Teratology 1990 41:663-677.
Diewert V, Wang K. Recent advances in primary palate and midface morphogenesis
research. Crit Rev Oral Biol Med 1992 4:111-30.
Diewert V, Wang K-Y, Tai B. A morphometric analysis of cell densities in facial
prominences of the rhesus monkey embryo during primary palate formation. J Craniofac
Genet Dev Biol 1993 13:236-249.
Figueroa A, Pratt R. Autoradiographic study of macromolecular synthesis in the fusion
epithelium of the developing rat primary palate in vitro. J Embryol Exp Morph 1979
Fitch J, Mayne R, Linsenmayer T. Developmental acquisition of basement membrane
heterogeneity: Type IV collagen in the avian lens capsule. J Cell Biol 1983 97:940-43.
Fridman R, Toth M, Pena D, Mbashery S. Activation of Progelatinase B (MMP-9) by
gelatinase A (MMP-2). Cancer Res 1995 55:2548-55.
Gaare J, Langman J. Fusion of nasal swelling in the mouse embryo: DNA synthesis and
histological features. Anat Embryol 1980 159:85-99.
Gack S, Vallon V, Schmidt J, Grigoriads A, Tuckermann J, Schenkel J, Weither H,
Wagner E, Angel P. Expression of interstitial collagenase during skeletal developmental
of the mouse is restricted to osteoblast-like cells and hypertrophic chondrocytes. Cell
Growth Differ 1995 6: 759-767.
Gaire M, Magbanua Z, McDonnell S, McNeil L, Lovett D, Matrisian L. Structure and
expression of the human gene for the matrix metalloproteinase matrilysin. J Biol Chem
1994 Jan 269: 2032-2040.
Gomis-Ruth F, Maskos K, Betz M, Bergner A, Huber R, Suzuki K, Yoshida N, Nagase
H, Brew K, Bourenkov G, Bartunik H, Bode W. Mechanism of inhibition of the human
matrix metalloproteinase stromelysin- 1 by TIMP- 1. Nature 1997 389: 77-81.
Grant J, Maggio-Price L, Reutebuch J, Cunningham M. Retinoic acid exposure of the
mouse on embryonic day 9 selectively spares derivatives of the frontonasal neural crest. J
Craniofac Genet Dev Biol 1997 17: 1-8.
Greene R, Pratt R. Immunohistochemical localization of cyclic AMP in developing
rodent secondary palate. J Embryol Exp Morph 1980 60:271-81.
Grobelny D, Poncz L, Galardy R. Inhibition of human skin fibroblast collagenase,
thermolysin, and Pseudomonas aeruginosa elastase by peptide hydroxamic acids.
Biochem 1992 31: 7152.
Gui T, Osumi-Yamashita N, Eto K. Proliferation of nasal epithelial and mesenchymal
cells during primary palate formation. J Craniofac Genet Dev Biol 1993 13:250-258.
Gutman A, Wasylyk B. The collagenase promoter contains a TPA and oncogene-
responsive unit encompassing the PEA3 and AP-1 binding sites. EMBO J 1990 7:2241-
Hall B, MacSween M. A SEM analysis of the epithelial-mesenchymal interface in the
mandible of the embryonic chick. J Craniofac Genet Dev Biol 1984 4:59-76.
Hay E. Cell Biology of Extracelluar Matrix. 1981, Plenum Press, New York.
He CS, Wilhelm SM, Pentland AP, Marmer BL, Grant GA, Eisen AZ, Goldeberg GI.
Tissue cooperation in a proteolytic cascade activating human interstitial collagenase. Proc
Natl Acad Sci USA 1989 86:2632-36.
Heikinheimo K, Salo T. Expression of basement membrane type IV collagenases (MMP-
2 and MMP-9) in human fetal teeth. J Dent Res 1995 74(5): 1226-34.
Helms J, Kim C, Hu D, Minkoff R, Thaller C, Eichele G. Sonic hedgehog participates in
craniofacial morphogenesis and is down-regulated by teratogenic doses of retinoic acid.
Dev Biol 1997 187: 25-35.
Huhtala P, Chow L, Tryggvason K. Structure of the human type IV collagenase gene. J
Biol Chem 1990 265:11077-82.
Huhtala P, Tuuttila A, Chow L, Lohi J, Keski-Oja J, Tryggvason K. Complete structure
of the human gene for 92-kDa type IV collagenase. Divergent regulation of expression
for the 92- and 72-kilodalton enzyme genes in HT-1080 cells. J Biol Chem 1991
lamaroon A, Diewert V. Distribution of basement membrane components in the mouse
primary palate. J Craniofac Genet Dev Biol 1996 16: 48-51.
lamaroon A, Tse C, Diewert V. Distribution of basement membrane components in the
mouse primary palate. J Denr Res 1996 75 (IADR Abstracts) 179.
Iamaroon A, Wallon M, Overall C, Diewert V. Expression of 72-kDa gelatinase (matrix
metalloproteinase-2) in the developing mouse craniofacial complex. Archs Oral Biol
1996 41(12): 1109-19.
Igawa H, Yasuda M, Nakamura H, Okura T. Changes in the subepithelial mesenchymal
cell process meshwork in developing facial prominences in mouse embryos. J Craniofac
Genet Dev Biol 1986 6: 27-39.
Imai K, Yokohama Y, Nakanishi I, Ohuchi E, Fujii Y, Nakai N, Okada Y. Matrix
Metalloproteinase 7 (Matrilysin) from human rectal carcinoma cells. J Biol Chem 1995
Johnston M. A radioautographic study of the migration and fate of cranial neural crest
cells in the chick embryo. Anat Rec 1966 156:143-56.
Johnston MC, Bronsky PT. Prenatal craniofacial development: New insights on normal
and abnormal mechanisms. Crit Rev Oral Bio Med 1995 6(1): 25-79.
Kaartinen V, Voncken J, Shuler C, Warburton D, Bu D, Heisterkamp N, Groffen J.
Abnormal lung development and cleft palate in mice lacking TGF-P33 indicates defects of
epithelial-mesenchymal interaction. Nature Gen 1995 11: 415-21.
Karelina T, Goldberg G, Eisen A. Matrilysin (PUMP) correlates with dermal invasion
during appendageal development and cutaneous neoplasia. J Invest Derm 1994103:482-
Kaufman MH. The Atlas of Mouse Development. 1992, Academic Press, San Diego.
Kinoh H, Sato H, Tsunezuka Y, Takino T, Kawashima A, Okada Y, Seiki M. MT-MMP,
the cell surface activator of proMMP-2 (pro-gelatinase A), is expressed with its substrate
in mouse tissue during embryogenesis. J Cell Science 1996 109: 953-59.
Kleiner D, Unsworth E, Krutzsch H, Stetler-Stevenson W. Higher-order complex
formation between the 72-kilodalton type IV collagenase and tissue inhibitor of
metalloproteinases-2. Biochem 1992 31: 1665-72.
Kosaka K, Eto K. Appearance of a unique cell type in the fusion sites of facial processes.
J Craniofac Genet Dev Biol 1986 Suppl 2:45-52.
Leco K, Apte S, Taniguchi G, Hawkes S, Khokha R, Schultz G, Edwards D. Murine
tissue inhibitor of metallloproteinases-4 (Timp-4): cDNA isolation and expression in
adult mouse tissues. FEBS Letters 1997 401: 213-17.
Leco K, Khokha R, Pavloff N, Hawkes SP, Edwards DR. Tissue inhibitor of
metalloproteinases-3 (TIMP3) is an extracellular matrix-associated protein with a
distinctive pattern of expression in mouse cells and tissues. J Biol Chem 1994 Mar;
Librach C, Werb Z, Fitzgerald M, Chin K, Corwin N, Esteves R, Grbelny D, Galardy R,
Fisher S. 92-kD Type IV collagenase mediates invasion of human cytotrophoblasts. J
Cell Biol 1991 113:437-49.
Lim M, Elfman F, Dohrman A, Cunha G, Basbaum C. Upregulation of the 72-kDa type
IV collagenase in epithelial and stromal cells during rat tracheal gland morphogenesis.
Dev Biol 1995 171:521-30.
Lin CW, Georgescu HI, Evans CH. The role of AP-1 in matrix metalloproteinase gene
expression. Agents Act 1993 (Spec Conference issue) 39:C215-18.
Lovejoy B, Cleasby A, Hassell A, Longley K, Luther M, Weigl D, McGeehan G,
McElroy A, Drewry D, Lambert M, Jordan S. Structure of the Catalytic Domain of
Fibroblast Collagenase Complexed with an Inhibitor. Science 1994 Jan, 263:375-377.
Macauley S, Tarnuzzer R, Schultz G, Chegini N, Oxford G, Humphreys-Beher M.
Extracellular-matrix gene expression during mouse submandibular gland development.
Archs Oral Biol 1997 42: 443-54.
Maestri N, Beaty T, Hetmanski J, Smith E, McIntosh I, Wyszynski D, Liang K, Duffy D,
VanderKolk C. Application of transmission disequilibrium tests to nonsyndromic oral
clefts: Including candidate genes and environmental exposures in the models. Am J Med
Genet 1997 73: 337-44.
Matrisian L. The Matrix-Degrading Metalloproteinases. BioEssays 1992 14(7):455-463.
Mattot V, Raes M, Henriet P, Eeckhout Y, Stehelin D, Vandenbunder B, Desbiens X.
Expression of interstitial collagenase is restricted to skeletal tissue during mouse
embryogenesis. J Cell Sci 1995 106: 1071-82.
Mayer D and Swanker W. Anomalies of infants and Children. New York, McGraw-Hill,
McCann J, Owens P, Wilson D. Chick frontonasal process excision significantly affects
mid-facial development. Anat Embryol 1991 184:171-78.
Minkoff R. Regional variation of cell proliferation within the facial processes of the chick
embryo: a study of the role of merging during development. J Embryol Exp Morph 1980
Minkoff R. Cell proliferation during formation of the embryonic facial primordia. J
Craniofac Genet Dev Biol 1991 11: 251-61.
Minkoff R, Kuntz A. Cell proliferation during morphogenetic change; analysis of
frontonasal morphogenesis in the chick embryo employing DNA labeling indices. J
Embryol Exp Morph 1977 40:101-113.
Minkoff R, Kuntz A. Cell proliferation and cell density of mesenchyme in the maxillary
process and adjacent regions during facial development in the chick embryo. J Embryol
Exp Morph 1978 46:65-74.
Monsky W, Kelly T, Lin C, Yeh Y, Stetler-Stevenson W, Mueller S, Chen W. Binding
and localization of M(r) 72,000 matrix metalloproteinase at cell surface invadopodia.
Cancer Res 1993 53:3159-64.
Morris-Kay G, Tam P. Mapping cranial neural crest cell migration pathways in
mammalian enbryos. Trends in Genet 1987 3:257-62.
Morris-Wiman J, Brinkley L. An extracellular matrix infrastructure provides support for
murine secondary palatal shelf remodelling. Anat Rec 1992 234:575-586.
Morris-Wiman J, Brinkley L. Rapid changes in the extracellular matrix accompany in
vitro palatal shelf remodelling. Anat Embryol 1993 188:75-85.
Muir D. Metalloproteinase-dependent neurite outgrowth within a synthetic extracellular
matrix is induced by nerve growth factor. Exp Cell Res 1994 210: 243-52.
Muir D. Differences in proliferation and invasion by normal, transformed and NF1
Schwann cell cultures are influenced by matrix metalloproteinase expression. Clin Exp
Metastasis 1995 13: 303-14.
Nagase H, Enghild J, Suzuki K, Salvesen G. Stepwise activation mechanisms of the
precursor of matrix metalloproteinase 3 (Stromelysin) by proteinases and (4-
aminophenyl) mercuric acetate. Biochem 1990 29: 5783-89.
Nakanishi Y, Sugimra F, Kishi JI, Hayakama J. Collagenase inhibitor stimulates cleft
formation during early morphogenesis of mouse salivary gland. Dev Biol 1986 113:201-
Noden D. An analysis of the migratiory behavior of avian cephalic neural crest cells. Dev
Biol 1975 42:106-130.
Noden D. Interactions and fates of avian craniofacial mesenchyme. Development 1988
103 Suppl: 121-140.
Odake S, Morita Y, Morikawa T, Yoshida N, Hori H, Nagai Y. Inhibition of matrix
metalloproteinases by peptidyl hydroxamic acids. Bioch Biophy Res Comm 1994 199(3):
Ogata Y, Enghild J, Nagase H. Matrix metalloproteinase 3 (stromelysin) activates the
precursor for the human matrix metalloproteinase 9. J Biol Chem 1992 267:3581-84.73.
Ogata Y, Itoh Y, Nagase H. Steps involved in activation of the pro-matrix
metalloproteinase 9 (Progelatinase B)-tissue inhibitor of metalloproteinases-1 complex by
4-aminophenylmercuric acetate and proteinases. J Biol Chem 1995 270(31): 18506-11.
Ohbayashi N, Eto K. Relative contributions of the facial processes to facial development:
A microsurgical assay. J Craniofac Genet Dev Biol Suppl 2 1986 41-44.
Okada Y, Shimnei M, Tanaka 0, Naka K, Kimura A, Nakanishi I, Bayliss M, Iwata K,
Nagase H. Localization of matrix metalloproteinase 3 (stromelysin) in osteoarthritic
cartilage and synovium. Lab Invest 1992 66: 680-90.
Okumura Y, Sato H, Seiki M, Kido H. Proteolytic activation of the precursor of
membrane type 1 matrix metalloproteinase by human plasmin. FEBS Letters 1997 402:
Pan L, Eckhoff C, Brinckerhoff C. Suppression of collagenase gene expression by all-
trans and 9-cis retinoic acid is ligand dependent and requires both RARs and RXRs.
Biochem 1995 57:575-589.
Patterson S, Minkoff R. Morphometric and autoradiographic analysis of frontonasal
development in chick embryo. Anat Rec 1985 212: 90-9.
Patterson S, Minkoff R, Johnston M. Autoradiographic studies of cell migration during
primary palate morphogenesis. J Dent Res 1979 58:113.
Pendds A, Knauper V, Puente X, Llano E, Mattei M, Apte S, Murphy G, L6pez-Otin C.
Identification and characterization of a novel human matrix metalloproteinase with
unique structural characteristics, chromossomal location, and tissue distribution. J Biol
Chem 1997 272(7): 4281-86.
Pyke C, Ralfkiaer E, Tryggvason K, Dano K. Messenger RNA for two type IV
collagenases is located in stromal cells in human colon cancer. Am J Pathol 1993
Reponen P, Sahlberg C, Huhtala P, Hurskainen T, Thesleff I, Tryggvason K. Molecular
cloning of murine 72-kDa type IV collagenase and its expression during mouse
development. J Biol Chem 1992 267(11): 7856-62.
Reponen P, Sahlberg C, Munant C, Thessleff I, Tryggvason K. High expression of 92-kD
type IV collagenase (gelatinase B) in the osteoclast lineage during mouse development. J
Cell Biol 1994 124:1091-1102.
Richman J, Diewert V. An immunofluorescence study of chonrogenesis in murine
mandibular ectomesenchyme. Cell Differentiation 1987 21:161-73.
Richman J, Rowe A, Brickell P. Signals involved in patterning and morphogenesis of the
embryonic face. In: The Morphogenesis of Down Syndrome. 1991, Wiley-Liss, New
Richman J, Tickle C. Epithelia are interchangeable between facial primordia of chick
embryos and morphogenesis is controlled by the mesenchyme. Dev Biol 1989 136:201-
Riedy M, Timm E, Stewart C. Quantitative RT-PCR for measuring gene expression.
Biotechniques 1995 18(1): 70-4, 76.
Rude F, Anderson L, Conley D, Gasser R. Three-Dimensional Reconstructions of the
Primary Palate Region in Normal Human Embryos. Anat Rec 1994 238: 108-113.
Saarialho-Kere U, Crouch E, Parks W. Matrix metalloproteinase matrilysin is
constitutively expressed in adult human exocrine epithelium. J Invest Dermatol 1995
Saber G, Minkoff R. Mesenchymal interaction on growth of facial primordia. J Dent Res
1991 Abstr 165.
Sadler T, Langman J, Burk D. Mammalian embryos in culture: A unique approach to
craniofacial research. In: Pratt R, Christiansen R (eds): Current Research Trends in
Prenatal Craniofacial Development. 1980, Elsevier, New York, 133-147.
Sahlberg C, Reponen P, Tryggvason K, Thesleff I. Association between the expression of
murine 72 kDa type IV collagenase by odontoblasts and basement membrane degradation
during mouse tooth development. Archs Oral Biol 1992 37(12):1021-30.
Sakai A, Langille R. Differential and stage dependent effects of retinoic acid on
chondrogenesis and synthesis of extracellular matrix molecules in chick craniofacial
mesenchyme in vitro. Differentiation 1992 52:19-32.
Sato H, Takino T, Okada Y, Cao J, Shinagawa A, Yamamoto E, Seiki M. A matrix
metalloproteinase expressed on the surface of invasive tumour cells. Nature 1994 370:
Schultz G, Strelow S, Stern G, Chegini N, Grant M, Galardy R, Grobelny D, Rowsey J,
Stonecipher K, Parmley V, Khaw P. Treatment of alkali-injured rabbit corneas with a
synthetic inhibitor of matrix metalloproteinases. Invest Opht Vis Science 1992 33: 3325-
Shams N, Hanninen L, Kenyon K. Increased gelatinolytic activity in the thermally
injured, nutritionally compromised rat cornea: Detection of a 27-kDa lymphoreticular
cell-associated caseinase. Curr Eye Res 1994 13(1):11-19.
Shapiro S. Mighty mice: Transgenic technology "knock outs" questions of matrix
metalloproteinase function. Matrix Biol 1997 15: 527-33.
Shim C, Choi D, Kwon H, Kim K. Expression of laminin chain-specific gene transcripts
in mouse uterine tissues during peri-implantation period. Mol Repr Dev 1997 48: 176-84.
Shimamori Y, Watanabe Y, Fujimoto Y. Characterization of gelatinases in human
placenta. Biochem Mol Med 1995 56: 84-86.
Shum L, Sakakura Y, Bringas Jr P, Luo W, Snead M, Mayo M, Crohin C, Millar S, Werb
Z, Buckley S, Hall F, Warburton D, Slavkin H. EGF abrogation-induced fusilli-form
dysmorphogenesis of Meckel's cartilage during embryonic mouse mandibular
morphogenesis in vitro. Development 1993 118: 903-17.
Sledge G, Quali M, Goulet R, Bone E, Fife R. Effect of matrix metalloproteinase
inhibitor batimastat on breast cancer regrowth and metastasis in athymic mice. J NIH
1995 87: 1546-1550.
Smiley, G.: A possible genesis for cleft palate formation. Plast Reconst Surg 1972 50:390
Streuli C, Baily N, Bissell M. Control of mammary epithelial differentiation: Basement
membrane induces tissue-specific gene expression in the absence of cell-cell interaction
and morphological polarity. J Cell Biol 1991 115: 1383-95.
Strogin A, Collier I, Bannikov G, Marminer B, Grant G, Goldeberg G. Mechanism of cell
surface activation of 72-kDa type IV collagenase. J Biol Chem 1995 270:5331-38.
Studer M, Popperl H, Marshall H, Kuroiwa A, Krumlauf R. Role of a conserved retinoic
acid response element in rhombomere restriction of Hoxb-1. Science 1994 265: 1728-32.
Sympson C, Talhouk R, Alexander C, Chin J, Clift S, Bissell M, Werb Z. Targeted
expression of stromelysin-1 in mammary gland provides evidence for a role of
proteinases in branching morphogenesis and the requirement for an intact basement
membrane for tissue-specific gene expression. J Cell Biol 1994 125:681-93.
Takino T, Sato H, Shinagawa A, Seiki M. Identification of the second membrane-type
matrix metalloproteinase (MT-MMP2) gene from a human placenta cDNA library. J Biol
Chem 1995 270(39): 23013-23020.
Tamarin A, Crawley A, Lee J, Tickle C. Analysis of upper beak defects in chick embryos
following treatment with retinoic acid. J Embryol Exp Morph 1987 84:105-123.
Tanaka H, Hojo K, Yoshida H, Yoshioka, Sugita K. Molecular cloning and expression of
the mouse 105-kDa gelatinase cDNA. Biochem Biophys Res Commun 1993 190(3): 732-
Tamrnuzzer R, Macauley S, Farmerie W, Caballero S, Ghassemifar M, Anderson J,
Robinson C, Grant M, Humphreys-Beher M, Franzen L, Peck A, Schultz G. Competitive
RNA templates for detection and quantitation of growth factors, cytokines, extracellular
matrix components and matrix metalloproteinases by RT-PCR. BioTechniques 1996 20:
Ten Cate A. Oral Histology. Development, Structure, and Function. 1989, Mosby,
Theiler K. The House Mouse. Atlas of Embryonic Development. 1989, Springer-Verlag,
Thesleff I, Barrach H, Foidart J, Vaheri A, Pratt R, Martin G. Changes in the distribution
of type IV collagen, laminin, proteoglycan and fibronectin during mouse tooth
development. Dev Biol 1981 81:182-92.
Thesleff I, Partanen A, Vainio S. Epithelial-mesenchymal interactions in tooth
morphogenesis: the roles of extracellular matrix, growth factors and cell surface
receptors. J Craniofac Gen Dev Biol 1991 11: 229-37.
Toumir JM, Polette M, Hinnrasky J, Beck J, Werb Z, Basbaum C. Expression of
gelatinase A, a mediator of extracellular matrix remodeling, by tracheal gland serious cells
in culture and in vivo. J Biol Chem 1994 269(41): 25454-64.
Vanden-Heuvel J, Tyson F, Bell D. Construction of recombinant RNA templates for use
as internal standards in quantitative RT-PCR. Biotechniques 1993 14(3): 395-8.
Vassalli JD, Sappino AP, Belin D. The Plasminogen Activator/Plasmin System. J Clin
Invest 1991 88: 1067-72.
Wasylyk C, Gutman A, Nicholson R, Wasylyk B. The c-Ets oncoprotein activates the
stromelysin promoter through the same elements as several non-nuclear oncoproteins.
EMBO J 1991 10:1127-1134.
Webster W, Johnston M, Lammer E, Sulik K. Isotretinoin embryopathy and the cranial
neural crest: An in vivo and in vitro study. J Craniofac Genet Dev Biol 1986 6: 211-22.
Wedden S. Epithelial-mesenchymal interactions in the development of chick facial
primordia and the target ofretinoid action. Development 1987 99:341-51.
Welgus H, Fliszar C, Seltzer J, Schmid T, Jeffrey J. Differential susceptibility of type X
collagen to cleavage by two mammalian interstitial collagenases and 72-kDa type IV
collagenase. J Biol Chem 1990 265: 13521-27.
Werb Z. ECM and cell surface proteolysis: Regulating cellular ecology. Cell 1997 91:
Wilson D, Hendrickx A. Quantitative aspects of proliferation in the nasal epithelium of
the rhesus monkey embryo. J Embryol Exp Morph 1977 38:217-26.
Wilson C, Heppner K, Rudolph L, Matrisian L. The metalloproteinase matrilysin is
preferentially expressed by epithelial cells in a tissue-restricted pattern in the mouse. Mol
Biol Cell 1995 6:851-69.
Wilson C, Matrisian L. Matrilysin: An epithelial matrix metalloproteinase with
potentially novel functions. Int J Biochem Cell Biol 1996 28: 123-36.
Wilson C, Matrisian L. Matrilysin. In: Parks W, Mecham R. (eds) Matrix
Metalloproteinases. 1998, Academic Press, New York, 149-184.
Xu Z, Parker S, Minkoff R. Distribution of type IV collagen, laminin, and fibronectin
during maxillary process formation in the chick embryo. Am J Anat 1990 187:232-46.
Xu Z, Parker S, Minkoff R. Influence of epithelial-mesenchymal interaction on the
viability of facial mesnchyme II: Synthesis of basement-membrane components during
tissue recombination. Anat Rec 1990 228:58-68.
Yu A, Hewitt R, Kleiner D, Stetler-Stevenson W. Molecular regulation of cellular
invasion-role of gelatinase A and TIMP-2. Biochem Cell Biol 1996 74: 823-31.
Zhang X, Kidder GM, Zhang C, Khamsi F, Armstrong DT. Expression of plasminogen
activator genes and enzymatic activities in rat preimplantation embryos. J Repr Fer 1994