University of ConnecticutOpenCommons@UConn
SoDM Masters Theses School of Dental Medicine
June 2004
Effect of Enamel Matrix Derivatives on MC3T3OsteoblastsGita Safaian
Follow this and additional works at: https://opencommons.uconn.edu/sodm_masters
Recommended CitationSafaian, Gita, "Effect of Enamel Matrix Derivatives on MC3T3 Osteoblasts" (2004). SoDM Masters Theses. 118.https://opencommons.uconn.edu/sodm_masters/118
EFFECT OF ENAMEL MATRIX DERIVATIVES ON MC3T3OSTEOBLASTS
Gita Safaian, D.M.D.
B.S., Queen's University, 1995
D.M.D., University of Connecticut, 1999
A Thesis
Submitted in Partial Fulfillment of the
Requirements for the Degree of
Master of Dental Science
At the
University of Connecticut
2004
APPROVAL PAGE
Master of Dental Science Thesis
EFFECT OF ENAMEL MATRIX DERIVATIVES ON MC3T3OSTEOBLASTS
Presented by
Gita Safaian, B.S., D.M.D.
Qu:~Major Advisor _
Qiang Zhu, D.D.S., Ph.D.
~~~,s\>b_-Lar~:Jg7~~;.~.s.,Ph.D.
Kamran E. Safavi, D.M.D., M.Ed.
S., Ph.D.Associate Advisor
-----+-+------r----=---.;~-~-----
Associate Advisor--------------
Associate Advisor
University of Connecticut2004
11
ACKNOWLEDGEMENTS
I would like to acknowledge the dedication and guidance of Dr. Qiang Zhu.
rlis approach to research is admirable and stress-free. I hope many future
endodontic residents take the opportunity to learn from his impressive ability to
turn a concept into a meaningful research project.
Dr. Larz Spangberg has been fundamental in helping me combine the
technical and the intellectual aspects of endodontics. It was a great honor and
privilege to learn from him. I thank him for all his advice.
I am forever grateful to Dr. Kamran Safavi for teaching me the history of
endodontology. I thank him for encouraging me to try different techniques while
emphasizing proper patient care. I am grateful for his guidance and his
emphasis on the ethics of being a Doctor.
I would like to thank Dr. Wang for his assistance with my thesis and Dr.
Ashraf Fouad for my undergraduate training.
Doctor Lisa Sanchez was invaluable in helping me prepare my thesis and
oral defense.
Finally, I thank my husband Dr. Tarik Kardestuncer for all his help and
support.
iii
TABLE OF CONTENTS
Page
lrllrL.E: ~~c;E:--------------------------------------------------------------------i
~~~~()"~L. ~~c;E:-----------------------------------------------------------ii
~CKN()WL.E:c;ME: Nlr--------------------------------------------------------i ii
lr~B L. E: ()F C()NlrE:NlrS-----------------------------------------------------iv
L.ISlr ()F lr~BL.E:S-------------------------------------------------------------v
L.ISlr ()F FIe;U~E:S-----------------------------------------------------------v i
~~~1r~~c:lr--------------------------------------------------------------------vii
~E:"IE:W ()F L.llrE:~~lrU~E:-------------------------------------------------1
~IM ()F SlrU[)~----------------------------------------------------------------15
M~lrE:~I~L. ~N[) ME:lrH()[)S-----------------------------------------------16
~E:SUL. lrS------------------------------------------------------------------------23
[)ISCLJSSI()N-------------------------------------------------------------------25
C()NCL.US1()N-----------------------------------------------------------------37
lr~BL.E:S-------------------------------------------------------------------------38
Flc;U.~E:S------------------------------------------------------------------------~2
~E:F~E:NCE:S-------------------------------------------------------------------51
IV
LIST OF TABLES
Table 1. Number of genes in response to EMD or TGF-J31---38
Table 2. Full DNA array results---------------------------------------39
v
LIST OF FIGURES
Figure1. Adhesion and spreading of MC3T3 cells onFibronectin, EMDand BSA--------------------------------------------42
Figure 2. EMD promotes cell attachment of MC3T3 cells---------------43
Figure 3. EMD promotes proliferation of MC3T3 cells-------------------44
Figure 4. Increase in alkaline phosphotase activity incultured MC3T3 cells by EMD---------------------------------------45
Figure 5. Northern blot analysis of EMD on collagena1 mRNA------46
Figure 6. Northern blot analysis of EMD on bone sialoprotein(BSP) mRNA---------------------------------------------------------------47
Figure 7. Northern blot analysis of EMD on osteocalcin«()C) mRNA-----------------------------------------------------------------48
Figure 8. Northern blot analysis of EMD on insulin-like growthfactor (IGF-1) mRNA----------------------------------------------------49
Figure 9. DNA micro array images in control, TGF-B1 andEMD treated groups-----------------------------------------------------------------50
vi
ABSTRACT
During tooth development, cells of Hertwig's epithelial root sheath secrete
enamel matrix proteins. A purified enamel matrix product (Enamel Matrix
Derivative, EMD) from developing porcine teeth became commercially available
and has been successfully employed to restore fully functional periodontal
ligament, cementum and alveolar bone. The regenerative capacity of EMD has
been demonstrated in clinical and animal studies. However the biological
mechanisms are not known. We hypothesize that EMD stimulates growth and
differentiation, and regulates certain gene expression of osteoblasts. In this
project we will determine the effect of EMD on cell adhesion, proliferation and
differentiation of preosteoblastic cell line MC3T3-E1 cells. A DNA array will also
be used to identify the cell cycle gene changes regulated by EMD and compare
these genes to those regulated by TGF-~1. This study will identify the role of
EMD on osteoblasts to understand the mechanisms of EMD induced bone
regeneration. By comparing EMD regulated genes to that regulated by TGF-~1,
it could be determined whether EMD has a similar function to TGF-~1, which has
been shown to induce bone regeneration.
VII
INTRODUCTION
During tooth development, cells of Hertwig's epithelial root sheath (HERS)
secrete an extracellular matrix similar to enamel matrix proteins secreted by
mature ameloblasts (Lindskog S 1982; Schendel et al. 1997; Slavkin et al.
1989b). Immunologic investigation suggest that cellular cementum contains
proteins similar to enamel matrix proteins (Slavkin et al. 1989a). In fact, enamel
matrix proteins have been recently detected at the apical portion of the forming
root by immunoblot analysis (Fukae 1996) . Hence j new advances j resulting in
the development of more sensitive methodology, have shown that many of the
so-called 'enamel proteins' have other functions besides that of enamel formation
(Fincham et al. 2000). To fully comprehend the function of these proteins it is
essential to review them.
What are enamel proteins?
During the development of the crown of the tooth, two mineralizing matrixes are
formed next to each other by two different layers of cells, odontoblasts and
ameloblasts. Dentin is a collagen-based matrix. Enamel on the other hand
contains no collagen and is mainly composed of hydrophobic proteins known as
amelogenins and non-amelogenin proteins (anionic enamel proteins;
ameloblastin; enamelin; tuftelin; tuft proteins; sulfated proteins and enamel
proteases such as enamelysin. The ameloblasts control the synthesis and
secretion of the organic extracellular matrix which is deposited along the dentino-
1
2
enamel junction (DEJ) which in turn controls enamel biomineralization. It is
s~ggested that enamel proteins function as nucleators and modulators for
calcium hydroxyl apatite crystal formation (Fincham 2000; Robinson et al. 1998;
Simmer et al. 1995; Zeichner-David et al. 1995).
Amelogenins are the most abundant enamel proteins comprising
approximately 900/0 of all proteins secreted by ameloblast cells. They are
hydrophobic proteins. Multiple amelogenins present in the enamel extracellular
matrix are the product of alternative splicing of the amelogenin gene and
processing of parent molecules (Gibson 1991). Recent studies shows that
hydroxyapatite crystals grown in the presence of amelogenins develop into
characterically long crystals associated with enamel formation (Wen 2000).
Tuftelin is a non-amelogenin, anionic enamel protein. This protein is
expressed prior to the onset of mineralization. The function of tuftelin in tooth
development remains unknown although recent studies suggest that it might
function at the level of ameloblast differentiation and/or extracellular matrix
secretion (Paine et al. 2000).
Ameloblastin represents 50/0 of the non-amelogenin mRNAs. This
molecule is localized to chromosome 4q21, the same region where a family with
autosomal dominant Amelogenesis Imperfecta has been linked, suggesting that
this protein is important in enamel formation. This protein is present in the
secretory stage of enamel formation and plays a role in enamel biomineralization.
Ameloblastin is also detected in pulpal mesenchymal cells and Hertwig's Root
3
sheath (Fong et al. 1998; Fong et al. 2000). The function of ameloblastin in
these tissues is unknown.
Enamelin is the largest enamel protein. Based on its presence in forming
enamel and the fact that enamelin gene is localized to human chromosome
4q21, the region localized to Amelogenesis Imperfecta, it is believed that it has a
role in enamel biomineralization (Dong et al. 2000).
Enamel proteases are important for processing the secreted amelogenins,
ameloblastin and enamelin in the extracellular matrix and also for their
degradation and removal from mineralizing matrix during the maturation of
amelogenin. A matrix metalloproteinase, enamelysin was recently cloned from
tissues and was later localized within the most recently formed enamel. Another
class of proteins present in enamel extracellular matrix are the 'sulfated enamel
proteins' (Smith 1995). The role of these proteins is unknown however.
Role of enamel proteins in Root formation
The involvement of enamel proteins in root formation was first suggested
in 1974 (Slavkin 1974). Several different hypothesis have been postulated to
explain the function of enamel proteins in root formation. First, they serve in
attachment of cementum to root dentine (Ten Cate 1996). Second, they help
initiate cementogenesis (Heritier 1982). Third, they serve as an inducer of dental
follicle cells to differentiate into cementoblasts (Hammarstrom 1997b;
Hammarstrom 1996).
4
Function of EMD in periodontal regeneration
Recently, with the avent of tissue bioengineering, it has been desirable to
take advantage of biological principles in organ development to induce tissue
regeneration. In 1997 a series of studies were published in the Journal of
Clinical Periodontology showing that when pig enamel matrix was placed in
experimental cavities it could initiate the formation of a tissue identical to
acellular, extrinsic fiber cementum (Hammarstrom 1997a). Extracted enamel
proteins or purified enamel matrix derivative (EMO) resulted in 60-800/0 formation
of new cementum and bone in experimental periodontal pockets in monkeys
(Hammarstrom et al. 1997). There were other studies showing that EMO
suspended in propylene glycol alginate (PGA) adsorbs to Hydroxyapatite and
collagen forming an insoluble complex which can remain on the root surface for
two weeks, sufficient time to promote re-colonization by periodontal ligament
cells (Gestrelius et al. 1997b).
In human studies, an experimental buccal dehiscence defect received
EMD and after four months the tissue was examined histologically. New
periodontal ligaments with functionally oriented collagen fibers and associated
alveolar bone were present (Heijl et al. 1997). All these studies suggested that
enamel proteins have the ability to promote periodontal regeneration by inducing
cementum, periodontal ligament (POL) and bone formation. This product was
marketed as EMDOGAIN by Siora, Inc. The active material in this product is
hydrophobic enamel matrix proteins extracted from porcine embryonic enamel.
The possibility of inducing immunologic reaction has been analyzed in vivo
5
(Heard et al. 2000; Heijl et al. 1997; Zetterstrom et al. 1997) and in vitro (Petinaki
et al. 1998). It has been shown that EMD exposed to blood lymphocytes caused
a slight increase of the proliferation of lymphocytes restricted to CD25, fraction of
DC4 positive cells and decrease of DC19 positive B lymphocytes (Petinaki et al.
1998). There have been several clinical trials where it has been shown that
individuals treated with EMDOGAIN have increased periodontium attachment
and increased levels of bone formation (Froum et al. 2001; Heijl et al. 1997;
Mellonig 1999; Okuda et al. 2000; Pontoriero et al. 1999; Rasperini et al. 1999;
Sculean et al. 2000; Sculean et al. 1999b). However in many of these cases the
results were not significantly different than other treatment modalities such as
barriers. Moreover some studies suggested that EMDOGAIN results were
unpredictable and other barrier techniques actually have better results (Heden et
al. 1999; Parashis et al. 2000; Pontoriero et al. 1999; Sculean et al. 1999b;
Silvestri et al. 2000).
It has been suggested that although the use of EMD for the treatment of
periodontal defects appears to be promising, the call for larger and controlled
clinical trials is considered necessary (Greenstein 2000; Heard et al. 2000;
Mellonig 1999). Furthermore most of the EMD studies are based on clinical
evaluation; such as pocket reduction, periodontium attachment and radiographic
images. Few of these studies have a small number of samples and some of their
observations are subjective and most importantly there is not a specific marker
for cementum, so it is difficult to determine it is cementum being formed or some
type of bone-like tissue.
6
Regenerative capabilities of EMD
In order to investigate if enamel proteins really have regenerative
properties and how they induce regeneration, in vitro models have been used in
several studies. These studies analyze the ability of EMO to influence migration,
attachment, proliferation, biosynthetic activity and mineral nodule formation in
POL cells in culture (Gestrelius et al. 1997b). Results indicate that EMO had no
effect on migration, and attachment but it enhanced POL cell proliferation. Also
EMO increased total protein synthesis and promoted nodule formation by POL
cells. To determine if the effect of EMO could be due to growth factors,
immunoassays were used to screen for molecules such as: Calbindin D, GM
CSF, fibronectin, b-FGF, gamma-Interferon, IL-1 beta, 2,3,6; IGF-1,2, NGF,
PDGF, TNF and TGF-beta. None of these molecules were detected but
contamination with growth factors can not be ruled out since growth factors
produced by ameloblasts such as bone morphogenic proteins (BMPs) were not
analyzed.
Several other studies have examined the effect of EMD on different cells.
Hoang (2000) investigated the effect of EMO on wound healing using an in vitro
wound healing model; measuring the wound -fill rates of POL, gingival and MG63
osteoblast-like cells in culture where a "wound" was created on a surface of the
cell (Hoang et al. 2000). EMD enhanced the wound-fill rate of all cells tested
with the POL cells being significantly higher at the early days of treatment. Also
the POL cells showed a greater response to EMD than to plasminogen derived
growth factor. These studies support the hypothesis that EMO increases cell
7
proliferation. In an experiment on human POL cells, (Lyngstadaas et al. 2001)
found that EMD significantly increased the attachment rate, growth and
metabolism of these cells. Moreover cells exposed to EMO showed increased
intracellular c-AMP signaling and production of TGF-beta1, interleukin 6 (IL-6)
and platelet derived growth factor (PDGF-AB).
Conversely with epithelial cells ( HeLa cells ), EMD had no effect on the
rate of attachment. Proliferation and growth were inhibited even though CAMP,
PDGF-AB had increased due to EMD. These studies indicate that EMD favors
mesenchymal cell growth while it inhibits epithelium growth. Moreover the
autocrine growth factors released by the PDL cells exposed to EMD might
contribute to periodontal healing and regeneration mimicking natural root
development.
In a study by Schwartz (2000) the effect of EMD on osteoblast cells were
analyzed and it was found that EMD regulates proliferation and differentiation
(Schwartz et al. 2000). This regulation is dependent on the 'stage' of maturation
of cells. In early stages (Pre-osteoblasts), it stimulates proliferation while in the
later stages in enhances differentiation, in dose dependent manner, as measured
by alkaline phosphatase activity and osteocalcin production. These studies
suggest that EMO might work as a growth factor (in these studies EMD
stimulated expression of TGF-~1 and MG63 osteoblast-like cells). On the other
hand EMD did not influence the production of collagen and proteoglycans
sulfation, while they both increased in the presence of other growth factors such
as BMP of TGF-~. It has been shown that EMD prolongs the growth of primary
8
cultures of mouse calvaria osteoblasts (Jiang et al. 2001). In this study, EMD
enhanced collagen I, Interleukin-6 and Prostaglandin G/H sythase2 (PGHS-2)
mRNA expression, but it did not stimulate the expression of osteocalcin and IGF
I (Jiang et al. 2001). Differences between the results of Jiang et al. and Schwartz
et al could be due to differences in the type of cells and\or the developmental
stage of cells used. To further understand the effect of EMD in bone formation,
ability of EMD to induce de novo bone formation in mouse calf muscle was
studied (Boyan et al. 2000). The results indicated that EMD was osteopromotive
and not osteoinductive.
The effect of EMD on epithelial cells in vitro was studied using oral
epithelial (SCC25) cells (Kawase et al. 2001). After three days of treatment,
EMD inhibited cell division and in tandem arrested cell cycle at the G1 phase.
Prior to this inhibition EMD up regulated p21WAF1/Cip1, a cyclin dependent
kinase inhibitor, induced G1- arrest, and inhibited DNA synthesis. In addition,
EMD down regulated expression of cytokeratin-18 (CK18) protein. These studies
suggest that EMD acts as a cytostatic agent on epithelial cells and its anti
proliferative action is probably due to cyclin-dependent kinase inhibitor,
P21WAF1/Cip1-mediated G1 arrest. This data supports Lyngstadaas et al.
(2000) data and can explain the observation that EMD suppresses the down
growth of junctional epithelium onto dental root surfaces, a process that
interferes with formation of new connective tissue attachment (Hammarstrom
1997b).
9
EMO promotes periodontal proliferation, enhancement of mineralization in
the POL cells and stimulation of collagen synthesis (Gestrelius et al. 1997a;
Gestrelius et al. 1997b). However, EMO inhibits mineralization in vitro when
immortalized cementoblast cells are used (Tokiyasu et al. 2000). All these
studies demonstrate that EMD promotes proliferation of cementoblast cells while
decreasing osteocalcin gene expression, though the expression of osteopathic is
slightly increased. It also suggests that increasing concentrations of EMD inhibit
mineralization in vitro but do not interfere with mineral formation in vivo when
using EMO-treated cementoblast soaked sponges implanted in
immunocompromised mice. These implants were kept in place for six weeks and
it is unknown how long after EMD treatment of cells they are still under the
influence of its activity (Zeichner-David 2001). Therefore, considering all these
experiments, both in vivo and in vitro suggests that EMD can act as a
multipurpose growth factor that stimulates proliferation of mesenchymal cells
while it inhibits cell division in epithelial cells or, it can stimulate cell differentiation
in some cells while inhibiting matrix production in others.
Given all these studies, it has been important to identify the actual protein
responsible for its function. EMD (EMDOGAIN), is a mixture of acidic extracted
enamel proteins. Despite HPLC purification, one cannot assume that it contains
only amelogenin. Ameloblast and enamelin are processed into smaller
fragments, some with hydrophobic properties similar to amelogenins.
Furthermore the possibility of contamination with growth factors has not been
completely ruled out.
10
Are enamel proteins growth factors?
The question arises as to whether or not enamel proteins are indeed
growth factors. The only way to determine if enamel proteins (and which enamel
proteins) can function as growth factors is to use recombinant proteins. The
advantage is due to the fact that enamel recombinant proteins are pure and don't
have the problem of contamination with growth factors and each protein can be
tested separately or in different combinations. In preliminary studies done by
Zeichner-David et al. (2000), the effect of amelogenin and recombinant
ameloblastin was analyzed on cell proliferation of different types of
Immortomouse-derived cell lines. Periodontal ligament and HERS cells were
grown in vitro in the presence of different concentrations of recombinant
amelogenin and ameloblastin. The rate of proliferation of these cells was
determined. The result of this study showed that recombinant ameloblastin had
no apparent effect on POL proliferation. However recombinant amelogenin
appeared to increase the rate of PDL proliferation. In the case of HERS cells
ameloblastin increased the rate of proliferation while amelogenin had a slight
effect in higher concentrations. These studies suggest that amelogenin has an
effect on PDL proliferation. These results are surprising since amelogenin is not
present during root formation. Studies using specific probes for amelogenin
mRNA and in situ hybridization shows that HERS cells do not synthesize
amelogenins (Luo et al. 1991). In studies using amelogenin and ameloblastin
antibodies, epitopes of these proteins were detected in the cervical portion of rat
molars but none was evident in the HERS cells or cementum layer (Bosshardt et
11
al. 1998; Nanci et al. 1998). Fong & Hammarstrom reported the presence of
epitopes cross-reactive with an amelogenin antibody in the area where
cementogenisis is inhibited in human teeth, but not in HERS (Fong &
Hammarstrom 2000). Hamamoto et al. reported the capacity of the epithelial rest
cells of Malassez, assumed to be derived from HERS cells, to express
amelogenin in response to pulp inflammation (Hamamoto et al. 1996). The
amelogenin production only occurred under pathologic conditions in the rest of
Malassez. The mechanism by which amelogenin may act like a "growth factor"
and influence PDL proliferation and differentiation remains to be determined.
The expression of ameloblastin by HERS cells has been demonstrated in
developing mouse root using in situ hybridization (Fong et al. 1998; Fong &
Hammarstrom 2000; Fong 1996). Ameloblastin is synthesized at the early
stages of HERS formation when HERS are actively proliferating to form the roots
(Fong 1996). Recently in vitro studies using the HERS cells, RT-PCR and
Westernblot has confirmed that these cells do not synthesize amelogenin and
enamelin, but do synthesize tuftelin and ameloblastin (Ohishi 2003). There has
been a suggestion that ameloblastin might have something to do with HERS
proliferation during apical migration of these cells to form the roots during tooth
development. The mechanisms of amelogenin and ameloblastin actions on
proliferation, and whether they have any effect on cell differentiation, remains to
be determined.
The 'growth factor'-like activity of amelogenin has been demonstrated
(Veis et al. 2000). They have used a 73 amino acid protein termed A+4
12
(containing exons 2, 3, 4, 5, 6d and 7) representing splice-products of the rat
amelogenin. Both peptides were able to increase corporation of 35[S]-804 into
proteoglycan in embryonic rat muscle fibroblasts in vitro and in vivo. These
proteins also induced the expression of transcription factors SOX9 and Cbfa 1 as
well as collagen type II and bone associated protein BAG-75 and BSP followed
by mineralization. A+4 appears to induce the osteoblast phenotype. These
experiments support previous data by Wang which suggested that amelogenins
and soluble dentin proteins could induce the differentiation of muscle
mesenchymal cells into chondrocytes and osteocytes (Wang 1993). Also in
studies by Nebgen they isolated and characterized a dentin matrix protein
exhibiting chondrogenic activity similar to amelogenin (Nebgen et al. 1999). This
protein, originally believed to be ameloblast contaminant, represents an actual
translation product of odontoblast cells and not the degradation product of
enamel matrix proteins.
It is also claimed that amelogenin splice products are actually transcribed
by odontoblasts cells (Veis et al. 2000). The presence of amelogenin epitopes
in odontoblast cells has been demonstrated in a study, which postulated that
these proteins were translocated from pre-ameloblast to odontoblasts to serve
as yet unknown biological function (Nakamura et al. 1994). The expression of
amelogenins by odontoblast cells has been reported , however the possibility of
contamination of tissue used to obtain mRNA by ameloblast cells has not been
ruled out (Dey et al. 2001). Transient expression of specific gene products by
other cells also has been shown for dentin sialophosphoproteins using in situ
13
hybridization. However similar studies using amelogenin probes have shown
expression exclusively by ameloblast cells. The possibility that amelogenin
splice products from root odontoblast cells mediate the signal for POL
proliferation and bone formation during root development should be considered.
Clinical evidence of EMD induced regeneration
In 1997, the concept of EMD guided regeneration was introduced to the
periodontal community in a series of scientific reports. The clinical efficacy of
EMD in terms of probing measurements and radiographic bone gain was first
demonstrated in 2 multicenter studies conducted by a group of Scandinavian
investigators (Hejil et al. 1997; Zetterstrom et al. 1997) . Case reports published
by several clinicians and researchers confirmed the initial findings and
established that EMD application into deep bony lesions promotes significant
attachment level gain, probing depth reduction and bone regeneration (Heard et
al. 2000; Heden 2000; Heden et al. 1999; Parashis &Tsiklakis 2000; Parodi et al.
2000; Sculean et al. 2001; Sculean et al. 1999a).
In a review by Kalpidis , clinical data from EMD controlled studies were
pooled for meta analysis and weighed according to the number of treated defects
(Kalpidis et al. 2002). Clinical attachment gain amounted to 33% of the original
attachment level and probing reduction averaged to SOak of the baseline probing
depth. Improvements in the clinical parameters achieved with EMO were
statistically significant in reference to preoperative measurements. Preliminary
histologic investigation with surgically created defects and experimental
14
periodontal lesions demonstrated the ability of EMD guided regeneration to
induce formation of acellular cementum and promote significant analysis of the
supporting periodontal tissues. The potential of EMD to encourage periodontal
regeneration was also confirmed in human infrabony defects. However this
review points at the recent human histologic studies where consistency of the
histologic outcomes and ability of EMD Guided Regeneration to predict stimulate
formation of acellular cementum have been questioned. Moreover, identifying
the exact cellular interactions is essential for development of methodologies and
enhances predictability of the EMD usage in dentistry.
SPECIFIC AIMS
Specific Aim 1:
To study the effect of EMD on adhesion, proliferation, and
differentiation of osteoblasts.
Specific Aim 2:
To identify the cell cycle genes regulated by EMD using DNA array, and to
compare these genes to those regulated by TGFp1.
15
MATERIALS AND METHODS
Cell culture
Mouse pre-osteoblasts MC3T3 cells were obtained from American Type
Culture Collection (ATCC, Manassas, Va). Cells were cultured in Dulbecco's
modified eagle medium (DMEM) (Gibco BRL, Gaithersburg, MD) supplemented
with 100/0 fetal bovine serum (FBS, Hyclone Laboratories, Inc, Logan, UT) and
1% of an antibiotic/antimycotic cocktail (300 U/ml penicillin, 300 Jlg/ml
streptomycin, and 5 J.lg/ml amphotericin B) under standard cell culture conditions
(37°C, 100% humidity, 95% air and 50/0 CO2).
Cell adhesion
The method for cell adhesion assay was similar as previously described.
Bacterial plastic Petri dishes (100 mm) were coated with 4 ml fibronectin (50 Jlg/ml)
or EMD (50 Jlg/ml) in serum free DMEM for 1.5 hrs. Control plates were only
incubated with DMEM. The plates were then blocked with 8 ml 0.1 % bovine serum
albumin (BSA) in serum free DMEM for 2 hrs. Cells (1x105/dish) were placed into
the plate and incubated for 1 hr and 18 hrs. At the end of each incubation period,
the total cell number per well was electronically counted, and cell morphology
was examined under microscope and photographed.
16
17
Cell proliferation
Cells were seeded onto 6-well or 12-well cell culture plates (Costar,
Corning, NY) at an initial density of 5,000/cm2, and allowed to attach for 24
hours. Medium was then changed into serum-free medium. 100 J.lg/ml EMD
(BIORA AB, Malmo, Sweden) was added to the treatment groups for 1, 2, 3, 5
and 7 days. Cells cultured without EMD served as negative control. At the end
of each incubation period, cells were released with 0.25% trypsin and 1 mM
EDTA in Hanks' balanced salt solution (Gibco BRL, Gaithersburg, MD), and the
total cell number per well was electronically counted (Coulter Corporation, Miami,
FL). Data were statistically analyzed using one-way ANOVA by comparing total
cell numbers per well.
Cell differentiation
Cell were cultured in DMEM with 2% FBS. Osteoblastic differentiation
supplements (50 J.lg/ml ascorbic acid, 8mM ~-glycerophosphotate) were added to
the culture media on day 7 when cells reached confluence. Cells were then
further incubated for 3 weeks. Media was changed every 2 days. Fresh EMD
and supplements were added during each media change. Control group was
cultured under the same conditions without EMD. Osteoblastic differentiation
was evaluated by ALP activity assay and the expression of bone differentiation
markers including collagen a1 (I), SSP, and osteocalcin (OC).
18
Alkaline phosphotase (ALP) activity assay
Cells cultured in differentiating media in the presence or absence of EMD
for 3 weeks were analyzed for ALP activity by ALP staining and ALP activity
assay. For ALP staining, cells were rinsed twice with phosphate-buffered saline
(PBS), fixed with 20/0 paraformaldehyde, ALP substrate mixture (ALP staining kit,
Sigma Diagnostic) was then added and incubated for 15 minutes for color
development. For ALP activity assay, cell layers were scraped off culture plates
in scrapping buffer. Cell pellets were collected after a quick spin. Cells were
then lysed in ALP lysis buffer, and subjected to 3 freeze-thaw cycles. After
centrifuge at 14,000 rpm for 5 minutes, supernatant was collected. Twenty JlI
supernatant from each sample was added to each well in duplicates in a 96-well
plate (Costar, Corning, NY), and incubated with an assay mixture of p-nitrophenyl
phosphate. Plates were then scanned for spectrophometric analysis using a
plate reader (Quant plate reader, Bio-Tek, Winooski, VT). Absorbance was
measured at 405 nm every 5 minutes for 30 minutes. Activity was calculated
using KC junior software (Bio-Tek, Winooski, VT). Protein content in each
sample was determined by BeA assay. ALP activity is expressed as
nMol/min/mg. Group difference of ALP activity was determined using student's t
test.
19
Northern blot analysis
Total mRNA from control and EMD-treated cells was extracted using
TRlzol reagent and further purified with phenol-chloroform method after each
incubation period. After quantification, 20 J.lg of mRNA from each group was
loaded onto 1% agarose gel and separated by electrophoresis. RNA was then
transferred onto Gene-Screen Plus Hybridization Transfer membrane by capillary
pressure. RNA was cross-linked to each membrane by UV irradiation. The
membrane was hybridized with 32P-labeled cDNA probes at 42°C for 20h with 5-6
million counts per minute (cpm) per ml for each probe. The following mouse
cDNA probes were labeled with 32p-dGTP using random primer oligonucleotides:
collagen a1 (I), SSP, ac. p-actin probe served as loading control. After
hybridization, membranes were washed and exposed to a phosphoimager, the
intensity of radioactive label was quantified using ImageQuant™. A photographic
film was then placed over the membrane for exposure, and was developed using
a Kodak developer. The intensity of the bands was normalized to p-actin levels.
20
eDNA micro array
Cell treatment
MC3T3-E1 cells were seeded to the cell culture dishes (Sigma, St Louis
MO) at an initial density of 5000 cells/cm2. Cells were cultured in Dulbecco'a
Modified Eagle Medium (DMEM) (gibco BRL, Gaithersburg, MD) supplemented
with 10% fetal bovine serum (Hyclone Laboratories Inc., Logan, UT) under
standard cell culture conditions (37 degrees Celicius, 1000/0 humidity, 95% air
and 5% carbon dioxide). After cells grew to 80% confluence, the medium was
divided into three different groups: Group 1 DMEM (control group) only, Group 2
DMEM plus TGF-J31 (10 ng/ml), and Group 3 DMEM plus EMD (100 IJg/ml)
(Emdogain, BIORA AB, and Malmo, Sweden). Cells were incubated for 24hrs.
RNA extraction
Total cellular RNA was isolated with guanidine thiocyanate followed by
phenol chloroform extraction and ethanol precipitation (Chomczynski and Sacchi,
1987). The RNA was digested with 10 units of RNase-free DNase for 30 minutes
at 37 degrees Celicius, extracted with Sodium acetate and ethanol. The
recovered RNA was quantitated by spectrophotometry. A twenty microgram
aliquot of RNA from each group was fractionated by electrophoresis through a
1% agarose/2.2M formaldehyde gel. The gel was stained with ethidium bromide
to visualize the 18S and 28S RNA bands and to assess the RNA integrity. The
RNA samples were stored at -700 C for further use.
21
cDNA probe preparation and array hybridization
Labeled probes were prepared by reverse transcriptase of 3 micrograms
of RNA in the presence of Biotin labeled dUTP (Super Array, Bioscience
Corporation). GE arrayTM cDNA expression array membranes (Super Array,
Bioscience Corporation) containing human cell cycle gene array series, which
was developed to profile the expression of 96 genes involved in cell cycle
regulation, were used. Cell cycle progression is precisely controlled by cyclin
dependent kinases (CDKs) and proteins that regulate COKs. These CDKs and
CDK-modifying proteins, including cyclins, COK inhibitors, COK phosphatases,
and COK kinases, are included in the human cell cycle GE Array series. Genes
essential for ONA damage and mitotic spindle checkpoints, as well as genes in
the SCF and APC ubiquitin-conjugation complexes, are also represented. These
membranes were prehybridized and hybridized with biotinylated cDNA probes
and washed twice with two wash solutions (2X SSC, 1%SOS and 0.1XSSC,
05% S0S) as specified in the instruction manual (Super Array, Bioscience
corporation) .
Chemiluminescent signal detection
The hybridized filter membranes were incubated with alkaline
phosphatase-conjugated Streptavidin (AP). Filters were washed and 0.1 ml
COP-star chemiluminescent substrate was added. The membranes were then
exposed to x-ray film. The image from x-ray film was scanned and analyzed
using GE ArrayTM Analyzer software. GE Array Analyzer matches the raw data
22
table with the gene list for the particular array_ Genes with two fold induction or
repression were selected and further analyzed.
RESULTS
MC3T3 -E1 cells adhere and spread on EMD
After 1-hr incubation cells attach and spread on fibronectin and EMD (Fig.
1). However cells spread less on EMD than fibronectin (Fig. 1). Cells in the BSA
group are rounded and most of the cells can be washed out (Fig. 1). After 18-hr
incubation cells spread extensively both on fibronectin and EMD (Fig. 1).
However cell morphology looks different. Cells did not attach on BSA coated
plates after 18-hr incubation (Fig. 1).
The cell number in the EMD group was significantly more than that in the
BSA group (P < 0.05) after 1-hr incubation (Fig. 2). The cell number in the
fibronectin group is greater (P < 0.05) than that in the BSA and EMD groups (Fig.
2). After 18-hr incubation, there is no cell adhesion in the BSA group, and there
is no cell number difference between the EMD group and fibronectin group.
EMD promotes proliferation of MC3T3-E1 cells.
Total cell counts in control and treatment groups after each incubation
period are shown in figure 3. Cell morphology in both groups appeared normal.
No significant difference was seen between control and EMD-treated groups on
day 1. Starting from day 2, EMD treatment resulted in a marked increase in cell
number (p<0.01). This increase was persistent into day 7. Cell numbers in both
groups increased with extended incubation time, and reached plateau around
day 3.
23
24
EMD promotes differentiation of MC3T3-E1 cells.
ALP staining showed more and bigger ALP-positive nodules in EMD
showed no significant difference between control and treatment groups
determined by SeA assay (data not shown). ALP activity was up-regulated in
cells exposed to EMD after normalized to total protein levels (Fig. 4).
Densitometry of Northern blot with 32P-labeled collagen a1 (I) probes
revealed that the levels of collagen a1 (I) mRNA expression were increased after
3-week culture in differentiating media treated wells (Fig. 5). SSP and OC were
also enhanced by EMD treatment (Fig. 6, 7). Increased expression of growth
factor IGF-I was also observed after 3 weeks in differentiation medium in EMD
treated group (Fig. 8).
Both EMD and TGF-p1 upregulate large number of common genes in the
cell cycle array
The DNA Array results showed that thirty two genes in the cell cycle are
upregulated by both EMD and TGF-~1 (see table 1). There are also fifty six
genes which are not affected by either EMD or TGF-~1. There are only two
genes which are upregulated by EMD alone and not TGF-~1. Moreover two
genes are down regulated only by EMD and not by TGF-~1. Fig. 9 shows the
DNA micro array images. Table 2 shows the complete list of cell cycle genes
from the DNA array.
DISCUSSION
I. Effect of EMD cell adhesion, proliferation, and differentiationof osteoblasts.
A BACKGROUND ON EFFECTS OF EMD ON OSTEOBLASTS
EMD had been used in promoting periodontal tissue regeneration. Its
osteogenic activities have been shown clinically in the repair of periodontal
infrabony defects in alveolar bone (Cardaropoli et al. 2002; Hammarstrom 1997a;
Heden et al. 1999; Heijl et al. 1997; Rasperini et al. 1999; Sculean et al. 1999a;
Trombelli et al. 2002). This study investigates the effects of EMD in osteoblast
adhesion, proliferation, differentiation and gene expression using DNA array
technology. To accomplish this purpose, our study was divided into two parts.
First, a thorough understanding of the adhesion, proliferation and differentiation
of MC3T3 mouse pre-osteoblastic cells in the presence of EMD was studied.
With a thorough understanding of the effect of EMD on this pre-osteoblastic cell
line we can then focus our attention on the specific aim of this study. The main
objective of this project was, therefore, to identify cell cycle genes regulated by
EMD and compare these genes to those regulated by TGFJ3-1 using DNA array
technology. By using this powerful molecular tool, DNA Array was applied in the
identification of up regulated and down regulated genes in EMD treated cells
compared with untreated cells.
25
26
Other studies have shown that during tooth development, cells of
Hertwig's epithelial root sheath secrete enamel matrix proteins (Lindskog S 1982;
Schonfeld et al. 1977; Slavkin et al. 1989b). A purified enamel matrix protein
called EMD, is a protein mixture with no detectable growth factors (BIORA AB,
Malmo, Sweden). EMD is 90°A, amelogenin plus other proteins including
enamlein, ameloblastin, tuftelin and dentine sialophosphoprotein (Brookes et al.
1995 & Robinson et al. 1998). It is well known that these enamel proteins play
an important role in the formation of hydroxyapetite crystals in mature enamel
(Robinson et al. 1998). Recently. EMD was also shown to be associated with
cementogenesis, periodontal tissue development and regeneration
(Hammarstrom 1997b; Lindskog S 1982; Slavkin et al. 1989a). Now, purified
EMD from porcine teeth is commercially available and has been used for its
regenerative capacity in preclinical studies and in clinical treatment. However, the
biological mechanisms of EMD in periodontal regeneration and in alveolar bone
are not clearly elucidated.
What is known, however, is the effect of EMD on bone regeneration in
vivo. EMD adsorbs to denuded root dentin surfaces and induces the formation of
a new layer of acellular cementum and alveolar bone in humans and monkeys
(Gestrelius et al. 1997a; Hammarstrom 1997a; Heijl et al. 1997; Safavi et al.
1999). Formation of new bone, cementum and alveolar bone was also
demonstrated histologically in cases with periodontal defects. Clinical studies in
the treatment of intrabony periodontal defects with EMD found bone regeneration
and a significant gain in clinical attachment (Heden et al. 1999; Heijl et al. 1997;
27
Rasperini et al. 1999; Sculean et al. 1999b). Controlled clinical trials comparing
the treatment of infrasonic defects with EMD and guided tissue regeneration fund
that there was no difference between the two treatment modalities. Both resulted
in significant improvement of probing depth, clinical attachment level and other
investigated clinical parameters (Pontoriero et al. 1999; Sculean et al. 1999a) .
In tissue implanted with EMD, there is enhanced new bone formation induced by
active demineralized freeze-dried bone allograft which may indicate that EMD is
an osteopromotive agent (Boyan et al. 2000).
Previous studies have shown that EMD enhances cell attachment
spreading the proliferation of cultured periodontal ligament cells (Gestrelius et al.
1997b; Lyngstadaas et al. 2001; Van der Pauw et al. 2000). It also stimulates
TGF p1, platelet derived growth factor and a variety of other growth factors in
PDL cells (Lyngstadaas et al. 2001; Van der Pauw et al. 2000). Alkaline
phosphatase activity has also been enhanced by the presence of EMD.
The role of EMD on cell specific osteoblastic proliferation and
differentiation is less well understood. Others have shown that it stimulates
proliferation but not differentiation of preosteoblastic 2T9 cells and it inhibits
proliferation and stimulates differentiation of osteoblastic like MG63 cells and
increases proliferation and differentiation of normal human osteoblasts
(Schwartz et al. 2000). Yet the effect of EMD on the well understood
preosteoblastic MC3T3 cell line was part of the focus of this study.
28
EMD PRMOTES ADHESION OF MC3T3-E1 CELLS
Our adhesion data is quite interesting. We have shown that after one hour
of incubation of MC3T3 cells on either fibronectin, BSA or EMD shows that cells
spread less on EMD than on fibronectin. Cells in the BSA group were rounded
and were washed out. However, after 18 hours of incubation cells spread
extensively both on fibronectin and EMD. However the cell morphology looked
quite different. Cells did not attach on the BSA coated plates after 18 hours. A
statistical significant increase in cell number was found after only 1 hour of
incubation between the cell numbers in the EMD group compared with the BSA
group. In the one hour incubation period the cell number in the fibronectin group
was greater that in either of the two other groups. After 18 hours there was no
cell adhesion in the BSA group and there was no difference in cell number
between EMD and fibronectin. These results clearly show that adhesion of these
MC3T3 is enhanced in the presence of EMD.
EMD PROMOTES PROLIFERATION OF MC3T3-E1 CELLS
With regards to proliferation of these cells, EMD also showed interesting
results. First, cell morphology in both groups appeared normal. On day one
there was no difference between the control group and the EMD group. On day
two, EMD treatments resulted in a statistically significant increase in cell number.
This increase was persistent into day seven with a peak at day three. Both
groups reached a plateau by day three and then slowly declined. Therefore EMD
treatment in MC3T3-E1 cells resulted in increased cell numbers compared to
29
control which suggest cell proliferation. This effect is consistent with results
obtained in some studies (He et al. 2004) but inconsistent with other reports
using the same culture system (Tokiyasu et al. 2000). This robust proliferation
of MC3T3 cells in the presence of EMD compared with control leads one to
conclude that part of the molecular action of EMD perhaps results from this
significant proliferation of these pre-osteoblastic cell line.
EMD PROMOTES DIFFERENTIATION OF MC3T3-E1 CELLS
With regards to differentiation, EMD also showed a stunning effect on the
MC3T3 cell line. Alkaline phosphatase staining showed much larger and greater
number of positive wells after three weeks in culture. However, total soluble
protein levels showed no difference between control and EMD groups as
determined by PCA assay. Alkaline phosphatase activity was up regulated in
cells exposed to EMD after being normalized to total protein levels.
Densitometry of northern blot with collagen alpha-1 probes revealed that
the amount of collagen mRNA expression was increased after three weeks of
culture in EMD treated groups. SSP and OC were also enhanced by EMD
treatment.
Type I collagen, SSP and OC are important extracellular matrix proteins
and contribute to bone matrix formation and mineralization (Stein et al. 1996).
The expression levels of theses genes represents different stages of cell cycle.
30
Levels of collagen alpha 1 m RNA was increased under both the short and long
term differentiating conditions. SSP and OC are relatively later markers for
osteoblast differentiation and are not expressed in early culture period (Stein et
al. 1996). However the expression levels of both SSP and DC were significantly
increased in cells treated with EMD when sells were cultured under differentiating
conditions. Up-regulation of these genes indicates that EMD enhance
osteoblastic differentiation at both early and late stages.
In summary, this portion of this study shows that, in the presence of EMD,
the MC3T3 cell line shows a significant increase in the amount of adhesion,
proliferation and differentiation.
31
II. Both EMD and TGF-131 upregulate large number of common genes in the
cell cycle
The objective of the second part of the study is to identify cell cycle genes
regulated by EMD and compare these genes to those regulated by TGF:-~1. The
results from the DNA Array shows a remarkable similarity between the genes
regulated by EMD compared to TGF-~1. Table 1 summarizes the number of
overlapping genes between EMD and TGF-~1. There are 32 cell cycle genes that
are unregulated both by TGF-~1 and EMD. EMD unregulated two cell cycle
genes which are not affected by TGF-~1 and, TGF-~1 up regulates another two
genes which are not influenced by EMD. There are 56 genes in the cell cycle
which are not affected either EMD or TGF-~.1. Also it is important to mention that
EMD and TGF- ~.1 down regulate multiple genes but these genes are not
common to both EMD and TGF-~.1. These results require further analysis.
What is TGF:- 131?
Transforming growth factor-~1 (TGF-~.1) is a 25 kD homodimetric
polypeptide, belonging to a super family of multifunctional cytokines, which
participates in a broad array of biologic activities such as normal development
and wound repair, as well as pathologic processes (Attisano et al. 1996; Choi et
al. 1997). TGF:- ~1 regulates multiple cellular functions including inhibition and
stimulation of cell growth, cell death or apoptosis, and cellular differentiation.
TGF:-~1 is also a potent inducer of extra cellular matrix (ECM) protein synthesis
32
and has been implicated as the key mediator of fibro genesis in various tissues
(Border et al. 1994).
TGF-~.1, a basic protein with high hydrophobicity, regulates proliferation
and differentiation of fibroblasts, keratinocytes and osteoblasts, extracellular
matrix metabolism in the bone or the connective tissue, and immunological
system. TGF:-~1 molecules bind to latency associated peptide (LAP), and latent
TGF-~.1 binding protein (LTBP) through non covalent and covalent bonds,
respectively. This binding enables in vivo unstable TGF:-~1 molecule to be stored
in the extracellular matrix thereby regulating its cellular functions. TGF-~.1
molecules are released in an active form from the extracellular matrix by disulfide
bond cleavage.
TGF-~.1 exerts its multiple biologic actions by the interaction with two
transmembrane serine/threonine kinase receptors, types I and II, which are co
expressed by most cells. Initiation of signaling requires the binding of TGF- ~.1 to
TBR-II, a constitutively active serine! threonine kinase, resulting in the
recruitment and phosphorylation of TBR-I to produce a heterometric signaling
complex that in turn activates downstream signaling pathways (Vivien et al. 1995;
Wrana et al. 1994).
33
Mechanism of TGF- 131 signaling
To understand the mechanisms involved in TGF- ~1 signaling from the cell
membrane to the nucleus, it is crucial to unravel the intracellular events that
provide a link between the activated cell surface TGF- ~.1 receptors and the
downstream effects of TGF:-r31, such as cell growth inhibition and matrix
induction. With the identification of the Smad family of signal-transducing
proteins involved in mediating TGF-r3.1 signals downstream of the
transmembrane serine/threonine kinase receptors, the focus of many
investigators has centered in studies of the Smad proteins. In particular, it is now
known that Smad2 and Smad3 both act as signaling proteins and transcription
factors and are serine phosphorylated in a TGF-~.1 dependent fashion and
associated with Smad4 to translocate to the nucleus where they bind to target
sites on specific gene promoters (Massague 1996).
Other researchers have demonstrated the critical involvement of the
mitogen-activated protein kinase (MAPK) pathways in TGF:-~1 signaling (Atfi et
al. 1997; Chin et al. 1999; Hartsough et al. 1996). MAPK is a major signaling
system used by eukaryotic cells to transducer extracellular signals to intracellular
responses (Seger et al. 1995; Vojtek et al. 1995). The signal transduction
cascades involved in the activation of MAPKs require a well-coordinated cascade
of tree protein kinase reactions that transducer signals by sequential
phosphorylation and activation of the next kinase in their respective pathways.
The MAPKs require dual phosphorylation at the threonine and tyrosine sites by
MAPK kinases, which are in turn activated by MAPK kinase kinases through
34
phosphorylation (Vojtek & Cooper 1995). The MAPK cascades are considered to
play essential roles in the signal transduction of many biological events, such as
the regulation of the cell growth, differentiation, and apoptosis and cellular
responses to environmental stresses. TGF- J3.1 has been demonstrated in
various cell types to be capable of activating each of the three major MAPK
members.
Potential functions of amelogenins which are the main constituent of EMD
To better understand the exact mechanism of action of EMD and
compared it to TGF-J31, it may be worthwhile to investigate the postulated
functions of amelogenins since EMD is essentially a mixture of amelogenins
(Gestrelius et al. 1997a; Hammarstrom 1997b). In the studies by Maycock et al
(2002) whereby EMD was compared to porcine enamel matrix derivative, it was
shown that EMD was heterogeneous but all the amelogenin components such as
degradation products and specific splice products were concentrated in EMD.
In addition to the direct role that amelogenins are postulated to have in the
organization and mineralization of the enamel matrix (Moradian-Oldak et al.
1995), they have been proposed to have signaling effects. As Lyngstadaas et al.
(2001) have suggested for DL cells , EMD interactions with mesenchymal cells
could initiate signaling pathways leading to secretion of transforming growth
factor-B1, interleukin-6 or platelet-derived growth factor or similar growth factors
which then permit tissue regeneration, depending on the cells involved.
35
Moreover it has been suggested t different peptide fractions which are
produced as a result of gene splicing in amelogenins may be essential in
induction of tissue generation (Gibson 1991; Viswanathan et al. 2002). In a
study by Veis et al.2000 it has been shown that two peptide fractions GST [A-4]
and [A+4] which were produced in the process of amelogenin splicing had
activities comparable to recombinant human BMP-2. These peptides
incorporated statistically significant levels of sulfate compared to controls. Same
group have also shown that [A+4] and [A-4] induce the appearance of type II
collagen message. In vivo implants into muscle, using a polylactide-polyglycolide
carrier confirmed that [A+4] and [A-4] were active in inducing cellular in growth
into the implants, followed by extracellular matrix production, vascularization and
mineralization after 4 weeks (Veis et al. 2000).
Therefore it has been suggested that smaller proteins produced by minor
(Fincham et al. 2000) spliced forms may have very specific functions. These
functions may relate to epithelial-mesenchymal signaling that is a prominent part
of tooth morphogenesis. It has been suggested that [A+4] may promote
odontoblast development and [A-4] may inhibit ameloblast maturation until a
layer of mineralized dentin is formed (Veis et al. 2000) .
Recently the preliminary studies on the use of [A+4] and [A-4] on tooth
repair via dentin bridge formation or pulp capping and differentiation effects on
fibroblasts and stem cells , presents a possibility for cartilage and bone tissue
regeneration or repair based on initiating factors other than the BMPs.
36
Our data agrees with other investigation whereby it appears that EMD
have, besides their activity in regulating hydroxyapatite crystal growth and
enamel formation, other functions as growth factors such as TGF- ~1. EMD may
act as a signaling molecule such as TGF:- ~1 in exerting its functions or it may
simply be involved in induction of growth factors such as TGF:- ~1. It is still not
conclusive that EMD can eliminate the damage of periodontal or bone disease by
the regeneration of new tissue. Furthermore it is unclear whether they could be
some other growth factor in the EMD preparation in addition to enamel proteins.
It remains to be determined if any of the other 'enamel' or 'dentin' proteins also
have functions in enamel or dentin mineralization. What are the signaling
mechanisms? Are they receptors? It is still so much to understand and all the
studies thus far are just the beginning of a new era of discovery.
CONCLUSION
1. EMD promotes cell adhesion of MC3T3-E1 cells at levels comparable to
fibronectin.
2. Compared to serum free medium, EMD enhances cell proliferation of
MC3T3-E1 cells.
3. EMD increases expression of osteoblastic makers including CoI1-a1,
ALP, BSP and OC.
4. EMD and TGF-(31 have similar effects on cell cycle gene expression.
37
Table 1* Number of genes in response to EMD or TGF-~1.
38
Unregulated
Down regulated
No Effect
EMD only
2
2
4
3
1
4
EMD and TGF-~1
32a56
* Please see Table 2 for a complete list of the genes which are summarizedabove=
Table 2: Full DNA array results
Gene expression increased by both EMD and TGF-131:
Abelson murine leukemia oncogeneataxia telangiectasia gene mutated in human beingsBcl2-associated X proteincyclin A2cyclin BMus musculus cyclin 82 (Ccnb2)Cyclin CCyclin D1(Ccnd1)Cyclin ECyclin E2cyclin GMus musculus cyclin H (Ccnh) mRNACDC16 (cell division cycle 16, S. cerevisiae, homolog)Mus musculus mmCdc20 mRNAMus musculus cell division cycle 6 homolog (S. cerevisiae) (Cdc6)Mus musculus M015-associated kinase (M015) mRNAcyclin-dependent kinase inhibitor p21Waf1Mus musculus checkpoint kinase Chk1 (Chk1) mRNAMus musculus cullin 3 (CuI3) mRNACullin 4A, RIKEN cDNA 2810470J21 geneSimilar to human E2F2, EST cloneMus musculus transcriptional activator (E2F3) mRNAE2F transcription factor 4, p107/p130-bindingE2F transcription factor 5Mus musculus mitotic checkpoint component Mad2 mRNAM. musculus mRNA for P1 protein (P1.m)Mus musculus mini chromosome maintenance deficient 2Mouse mRNA for mcdc21 proteinmeiotic recombination 11 homolog A (S. cerevisiae)Mus musculus nibrin (Nbn)Mus musculus Rad51 homolog (S. cerevisiae) (Rad51)Mus musculus RAD9 homolog (S. pombe) (Rad9)
Gene expression increased by EMD (2):
Mus musculus Cdc25a (cdc25a) mRNAcyclin-dependent kinase 6
39
40
Gene expression decreased by EMD (2):
Transformed mouse 3T3 cell double minute 2Mus musculus DNA repair protein RAD50 (RAD50) mRNA
Gene expression increased by TGF-J31 (3):
apoptotic protease activating factor 1Cyclin-dependent kinase inhibitor p57Kip2Mouse CDK4 and CDK6 inhibitor p16ink4a
Gene expression decreased by TGF 131 (1):
Retinoblastoma protein (Rb)
Gene expression no changes (56):
B-cell leukemia/lymphoma 2breast cancer 1Mus musculus cyclin A1 (Ccna1)Cyclin D2 (Ccnd2)Cyclin D3 (Ccnd3)M.musculus mRNA for cyclin FMus musculus cyclin G2 mRNAMus musculus cell division cycle 25 homolog 8 (S. cerevisiae)Cyclin-dependent kinase 1(Cdc2)Mus musculus cdc37 homolog mRNAMus musculus testes CDC45L (Cdc451) mRNAMus musculus muCdc7 mRNA, complete cdscyclin dependent kinase 2Lcyclin-dependent kinase 4Mus musculus cyclin-dependent kinase 5Homolog of Human Cyclin-dependent kinase 8, image cloneMouse cyclin-dependent kinase inhibitor p27Kip1cyclin-dependent kinase inhibitor p151NK4bCdk4 and Cdk6 inhibitor p18cyclin-dependent kinase inhibitor p19Mus musculus CDC28 protein kinase 1 (Cks1)Mus musculus SCF complex protein cul-1 mRNACullin 2,Mus musculus adult male liver cDNA, RIKEN full-length enrichedlibrary, clone: 1300003018, fCullin 48, RIKEN cDNA 2700050M05 geneE2F-related transcription factor (DP-1)
41
protein regulating cell cycle transcription factor DRTF1/E2FE2F1Mus musculus E2F-like transcriptional repressor protein mRNA, completecdsMus musculus forkhead box M1 (Foxm1)DNA-damage inducible transcript 1Mus musculus Hus1 homolog (S. pombe) (Hus1)Mouse mRNA for mCDC46 proteinmini chromosome maintenance deficient 6 (S. cerevisiae)Mus musculus mRNA for mCDC47antigen identified by monoclonal antibody Ki 67neural precursor cell expressed, developmentally down-regulated gene 8Mus musculus I-kappa B alpha chainProliferating cell nuclear antigenprotein regulator of cytokinesis 1Mus musculus RAD17 homolog (S. pombe) (Rad17)Mus musculus protein kinase Chk2 (Rad53-pending)Retinoblastoma-like protein 1 (p107)Retinoblastoma-like protein 2 (p130)Mus musculus ring-box protein 1 (Rbx1) mRNAGTP-binding protein ROC2ESTs, Highly similar to REPLICATION PROTEIN A 14 KD SUBUNIT[Homo sapiens]SCF complex protein Skp2 (Skp2)SCF complex protein Skp1 (Skp1)Tissue inhibitor of metalloproteinase 3Transformation related protein 53, Tumor antigeneTransformation related protein 63Ubiquitin CMus musculus ubiquitin-activating enzyme E1, Chr X (Ube1x)Mus musculus E6-AP ubiquitin protein ligase mRNAMus musculus ubiquitin-homology domain protein (UbI1) mRNATyrosine 3-monooxygenase/tryptophan 5-monooxygenase activationprotein epsilon polypeptide
Fibronectin EMD BSA
42
1 hr incubation before washing
1 hr incubation after washing 3 times with DMEM.
18 hrs incubation and washing 3 times with DMEM
Fig 1. Adhesion and spreading of MC3T3-E1 cells on fibronectin (A),
EMD (B), and BSA (C).
BSA.EMDD Fibronectin
43
Fig 2. EMD promotes cell attachment of MC3T3-E1 cells.
750000
...G).Q 500000E::::JC-G)0 250000
* * *
---Control- ... - EMD
44
O-+--------,..-------r--............---,...-------.
1 2 3 5 7
Time (day)
Fig 3. EMD promotes proliferation of MC3T3-E1 cells.
45
A.
B.
Control EMD
Control EMD
Fig. 4. Increase in alkaline phosphatase (ALP) activity in cultured MC3T3E1 cells by EMD.
46
A.
~:w= '<A ,"%Ym, ., "% "
~ ,
if ,'"" '
'N.". ~~,;:&, _,_
B.
c 50;i
40(.)ca........
'-e- 30ca~- 2000 10
0
Con EMD
Fig. 5. Northern blot analysis of EMD on collagen a1 (I) messenger RNA
expression. Results are shown by autoradiography (A) and quantitatively
analyzed by ImageQuant (B).
A.
47
'C~ wiffi ' 'g.(o~ "" ... :,: oX' ~ Yo ... "'" ~ "'-
• x
'" '\'
- .,,- ,
...~~ »;;::0;:- ......... ~ ,,'" ~ :-z: :jh
B.
c; 5u-!!Q.U)
m 2
Q
Control EMD
Fig. 6. Northern blot analysis of EMD on bone sialoprotein (BSP)
messenger RNA expression. Results are shown by autoradiography (A)
and quantitatively analyzed by ImageQuant (B).
48
A.
'"' ~ ;.: :=::::;'"'"'" v '" ...~ '... ;.: ...;}-::~::; "' ... \: ....
~ ,~"
1
",&. 01h~'l'W:'"""" v'1m" ,
B.
c~(,)CV(3o
Control EMD
Fig. 7. Northern blot analysis of EMD on osteocalcin (OC) messenger RNA
expression. Results are shown by autoradiography (A) and quantitatively
analyzed by ImageQuant (B).
49
A.
:i- ~ ...,:-; "'..m: ~~ ,
z ~% :-:: .... 7'( ~
> ,
, 1': <
% '
~ mid, "'~~","~ ,
B.
Control EMD
Fig. 8. Northern blot analysis of EMD on insulin-like growth factor I (IGF-I)
messenger RNA expression. Results are shown by autoradiography (A)
and quantitatively analyzed by ImageQuant (B). .
50
Fig. 9. DNA micro array images in control, TGF-p1, and EMD treated
groups.
REFERENCES
Atfi A, Djelloul S, Chastre E et al. Evidence for a role of rho-like GTPases andstress-activated protein kinase/c-Jun N-terminal kinase in transforming growthfactor B-mediated signaling. J Bioi Chern 1997; 272(3):1429-32.
Attisano L, Wrana J. Signal transduction by members of the tranforming growthfactor-B superfamily. Cytokine Growth Factor Rev 1996; 7(4):327-39.
Border W, Noble N. Tranforming growth factor B in tissue fibrosis. N Eng J Med1994; 331 (19): 1286-92.
Bosshardt DO, Nanci A. Immunolocalization of epithelial and mesenchymalmatrix constituents in association with inner enamel epithelial cells. J I-listochemCytochem 1998; 46(2): 135-42.
Boyan BD, Weesner TC, Lohmann CH et al. Porcine fetal enamel matrixderivative enhances bone formation induced by demineralized freeze dried boneallograft in vivo. J Periodontol 2000; 71 (8): 1278-86.
Cardaropoli G, Leonhardt AS. Enamel matrix proteins in the treatment of deepintrabony defects. J Periodontol 2002; 73(5):501-4.
Chin B, Petrache I, Choi AMK et al. Transforming growth factor-B1 rescuesserum deprivation-induced apoptosis via the mitogen-activated protein kinase(MAPK) pathway in macrophages. J Bioi Chern 1999; 274(16):11362-8.
Choi M, Liu A, Ballerman B. Differential expression of transforming growth factorB receptors in rat kidney development. Am J Physiol 1997; 273(3):F386-F95.
Dey R, Son HH, Cho MI. Isolation and partial sequencing of potentiallyodontoblast-specific/enriched rat cDNA clones obtained by suppressionsubtractive hybridization. Arch Oral Bioi 2001; 46(3):249-60.
Dong J, Gu TT, Simmons D et al. Enamelin maps to human chromosome 4q21within the autosomal dominant amelogenesis imperfecta locus. Eur J Oral Sci2000; 108(5):353-8.
51
52
Fincham A, Luo, W., Moradian-Oldak J., Paine M.L., Snead,M.L., zeichner-DavidM. Enamel Biomineralization: the assembly and disassembly of the proteinextracellular organic matrix. In: Ferguson, M.W.J. et al (Ed.), Development,Function and Evolution of teeth 2000.Cambridge University Press, UK.
Fincham VJ, James M, Frame MC et al. Active ERKIMAP kinase is targeted tonewly forming cell-matrix adhesions by integrin engagement and v-Src. Embo J2000; 19(12):2911-23.
Fong CD, Cerny R, Hammarstrom L et al. Sequential expression of an amelingene in mesenchymal and epithelial cells during odontogenesis in rats. Eur JOral Sci 1998; 106(Suppl 1):324-30.
Fong CD, Hammarstrom L. Expression of amelin and amelogenin in epithelialroot sheath remnants of fully formed rat molars. Oral Surgery, Oral Medicine,Oral Pathology, Oral Radiology, & Endodontics. 2000; 90(2):218-23.
Fong CD, Slaby,I.,Hammarstrom, L. Amelin: an enamel related protein,transcribed in the cells of epithelial root sheath. J. Bone Mineral Res 1996;11 (7):892-8.
Froum SJ, Weinberg MA, Rosenberg E et al. A comparative study utilizing openflap debridement with and without enamel matrix derivative in the treatment ofperiodontal intrabony defects: a 12-month re-entry study. J Periodontol 2001;72(1 ):25-34.
Fukae M, Tanabe, T., Murakami, C., Dohi, N.,Uchida,T., Shimizu,M., . primarystructure of the porcrine 89-kDa enamelin. Adv Dent Res 1996; 10(2):111-118.
Gestrelius S, Andersson C, Johansson AC et al. Formulation of enamel matrixderivative for surface coating. Kinetics and cell colonization. J Clin Periodontol1997a; 24(9 Pt 2):678-84.
Gestrelius S, Andersson C, Lidstrom 0 et al. In vitro studies on periodontalligament cells and enamel matrix derivative. J Clin Periodontol 1997b; 24(9 Pt2):685-92.
Gibson CW, Golub, E., Ding, W. et al. Identification of the leucine-richamelogenin peptide (LRAP) as the translation product of the alternatively splicedtranscript. Biochem Biophys Res Commun 1991; 174(3):1306-12.
53
Greenstein G. Emdogain: evidence of efficacy. Compend Contin Educ Dent2000; 21(4):299-305, 8,10 passim; quiz 14.
Hamamoto Y, Nakajima T, Ozawa H et al. Production of amelogenin by enamelepithelium of Hertwig's root sheath. Oral Surg Oral Med Oral Pathol Oral RadiolEndod 1996; 81 (6):703-9.
Hammarstrom L. Enamel matrix, cementum development and regeneration. JClin Periodontol 1997a; 24(9 Pt 2):658-68.
Hammarstrom L. The role of enamel matrix proteins in the development ofcementum and periodontal tissues. Ciba Found Symp 1997b; 205:246-55;discussion 55-60.
Hammarstrom L, Alatli, I., Fang, C.D. Origins of cementum. Orai Dis 1996;2(1 ):63-9.
Hammarstrom L, Heijl L, Gestrelius S. Periodontal regeneration in a buccaldehiscence model in monkeys after application of enamel matrix proteins. J ClinPeriodontol 1997; 24(9 Pt 2):669-77.
Hartsough M, Frey R, Zipfel P et al. Altered transforming growth factor Bsignaling in epithelial cells when ras activation in blocked. J Bioi Chem 1996;271 (37):22368-75.
He J, Jiang J, Safavi K et al. Emdogain promotes osteoblast proliferation anddifferentiation and stimulates osteoprotegerin expression. Oral Surg Oral MedOral Pathol 2004; 97(2):239-45.
Heard RH, Mellonig JT, Brunsvold MA et al. Clinical evaluation of wound healingfollowing multiple exposures to enamel matrix protein derivative in the treatmentof intrabony periodontal defects. J Periodontol 2000; 71 (11 ):1715-21.
Heden G. A case report study of 72 consecutive Emdogain-treated intrabonyperiodontal defects: clinical and radiographic findings after 1 year. Int JPeriodontics Restorative Dent 2000; 20(2): 127-39.
54
Heden G, Wennstrom J, Lindhe J. Periodontal tissue alterations followingEmdogain treatment of periodontal sites with angular bone defects. A series ofcase reports. J Clin Periodontol 1999; 26(12):855-60.
Heijl L, Heden G, Svardstrom G et al. Enamel matrix derivative (EMDOGAIN) inthe treatment of intrabony periodontal defects. J Clin Periodontol 1997; 24(9 Pt2):705-14.
Heritier M. Experimental induction of cementogenesis on the enamel oftransplanted mouse tooth germs. Arch Oral Bioi 1982; 27(2):87-97.
Hoang AM, Oates TW, Cochran DL. In vitro wound healing responses to enamelmatrix derivative. J Periodontol2000; 71(8):1270-7.
Jiang J, Safavi KE, Spangberg LS et al. Enamel matrix derivative prolongsprimary osteoblast growth. J Endod 2001; 27(2): 110-2.
Kalpidis CD, Ruben MP. Treatment of intrabony periodontal defects with enamelmatrix derivative: a literature review. J Periodontol2002; 73(11):1360-76.
Kawase T, Okuda K, Momose M et al. Enamel matrix derivative (EMDOGAIN)rapidly stimulates phosphorylation of the MAP kinase family and nuclearaccumulation of smad2 in both oral epithelial and fibroblastic human cells. JPeriodontal Res 2001; 36(6):367-76.
Lindskog S H. Formation of intermediate cementum.lIl: H-tryptophane and Hproline uptake into the epithelial root sheath of Hertwig in vitro. J Craniofac GenetDev Bioi 1982; 2(2):170-7.
Luo W, Slavkin HC, Snead ML. Cells from Hertwig's epithelial root sheath doesnot transcribe amelogenin. J Periodontal Res 1991; 26(1 ):42-7.
Lyngstadaas SP, Lundberg E, Ekdahl H et al. Autocrine growth factors in humanperiodontal ligament cells cultured on enamel matrix derivative. J ClinPeriodontol 2001; 28(2):181-8.
Massague J. TGFB signaling receptors, transducers, and Mad proteins. Cell1996; 85(7):947-50.
55
Mellonig JT. Enamel matrix derivative for periodontal reconstructive surgery:technique and clinical and histologic case report. Int J Periodontics RestorativeDent 1999; 19(1):8-19.
Moradian-Oldak J, Simmer JP, Lau EC et al. A review of the aggregationproperties of a recombinant amelogenin. Connect Tissue Res 1995; 32(1-4):12530.
Nakamura M, Bringas P, Jr., Nanci A et al. Translocation of enamel proteins frominner enamel epithelia to odontoblasts during mouse tooth development. AnatRec 1994; 238(3):383-96.
Nanci A, Zalzal S, Lavoie P et al. Comparative immunochemical analyses of thedevelopmental expression and distribution of ameloblastin and amelogenin in ratincisors. J Histochem Cytochem 1998; 46(8):911-34.
Nebgen DR, Inoue H, Sabsay B et al. Identification of the chondrogenic-inducingactivity from bovine dentin (bCIA) as a low-molecular-mass amelogeninpolypeptide. J Dent Res 1999; 78(9):1484-94.
Ohishi K, Gonzales E.,Arzate, H., Simjee, R., Bringas, P., Zeichner-David, M.Characterization of Hertwig's Epithelial Root Sheath cells maintained in cellculture. manuscript in preparation 2003.
Okuda K, Momose M, Miyazaki A et al. Enamel matrix derivative in the treatmentof human intrabony osseous defects. J Periodontol 2000; 71 (12): 1821-8.
Paine CT, Paine ML, Luo W et al. A tuftelin-interacting protein (TIP39) localizesto the apical secretory pole of mouse ameloblasts. J Bioi Chem 2000;275(29):22284-92.
Parashis A, Tsiklakis K. Clinical and radiographic findings following application ofenamel matrix derivative in the treatment of intrabony defects. A series of casereports. J Clin Periodontol 2000; 27(9):705-13.
Parodi R, Liuzzo G, Patrucco P et al. Use of Emdogain in the treatment of deepintrabony defects: 12-month clinical results. Histologic and radiographicevaluation. Int J Periodontics Restorative Dent 2000; 20(6):584-95.
56
Petinaki E, Nikolopoulos S, Castanas E. Low stimulation of peripherallymphocytes, following in vitro application of Emdogain. J Clin Periodontal 1998;25(9):715-20.
Pontoriero R, Wennstrom J, Lindhe J. The use of barrier membranes and enamelmatrix proteins in the treatment of angular bone defects. A prospective controlledclinical study. J Clin Periodontol 1999; 26(12):833-40.
Rasperini G, Ricci G, Silvestri M. Surgical technique for treatment of infrabonydefects with enamel matrix derivative (Emdogain): 3 case reports. Int JPeriodontics Restorative Dent 1999; 19(6):578-87.
Robinson C, Brookes SJ, Shore RC et al. The developing enamel matrix: natureand function. Eur J Oral Sci 1998; 106(SuppI1):282-91.
Safavi K, Kazemi R, Watkins D. Adherence of enamel matrix derivatives on rootend filling materials. J Endod 1999; 25(11 ):71 0-2.
Schendel SL, Xie Z, Montal MO et al. Channel formation by antiapoptotic proteinBcl-2. Proc Natl Acad Sci USA 1997; 94(10):5113-8.
Schonfeld SE, Slavkin HC. Demonstration of enamel matrix proteins on rootanalogue surfaces of rabbit permanent incisor teeth. Calcif Tissue Res 1977;24(3):223-9.
Schwartz Z, Carnes DL, Jr., Pulliam R et al. Porcine fetal enamel matrixderivative stimulates proliferation but not differentiation of pre-osteoblastic 2T9cells, inhibits proliferation and stimulates differentiation of osteoblast-like MG63cells, and increases proliferation and differentiation of normal human osteoblastNHOst cells. J Periodontol 2000; 71(8):1287-96.
Sculean A, Chiantella GC, Windisch P et al. Clinical and histologic evaluation ofhuman intrabony defects treated with an enamel matrix protein derivative(Emdogain). Int J Periodontics Restorative Dent 2000; 20(4):374-81.
Sculean A, Donos N, Miliauskaite A et al. Treatment of intrabony defects withenamel matrix proteins or bioabsorbable membranes. A 4-year follow-up splitmouth study. J Periodontal 2001; 72(12):1695-701.
57
Sculean A, Donos N, Windisch P et al. Healing of human intrabony defectsfollowing treatment with enamel matrix proteins or guided tissue regeneration. JPeriodontal Res 1999a; 34(6):310-22.
Sculean A, Reich E, Chiantella GC et al. Treatment of intrabony periodontaldefects with an enamel matrix protein derivative (Emdogain): a report of 32cases. Int J Periodontics Restorative Dent 1999b; 19(2):157-63.
Seger R, Krebs EG. The MAPK signaling cascade. Faseb J 1995; 9(9):726-35.
Silvestri M, Ricci G, Rasperini G et al. Comparison of treatments of infrabonydefects with enamel matrix derivative, guided tissue regeneration with anonresorbable membrane and Widman modified flap. A pilot study. J ClinPeriodontol 2000; 27(8):603-10.
Simmer JP, Fincham AG. Molecular mechanisms of dental enamel formation.Crit Rev Oral Bioi Med 1995; 6(2):84-108.
Slavkin He, Boyd A. Cementum: an epithelial secretory product? J. Dent. Res.1974; 53(2):157.
Slavkin HC, Bessem C, Fincham AG et al. Human and mouse cementumproteins immunologically related to enamel proteins. Biochim Biophys Acta1989a; 991 (1): 12-8.
Slavkin HC, Bringas P, Jr., Bessem C et al. Hertwig's epithelial root sheathdifferentiation and initial cementum and bone formation during long-term organculture of mouse mandibular first molars using serumless, chemically-definedmedium. J Periodontal Res 1989b; 24(1 ):28-40.
Smith CE, Chen, W.Y., Issid, M.,Fazel, A. Enamel matrix protein-turnover duringamelogenesis-basic biochemical properties of short-lived sulfated enamelproteins. Calcif Tissue Int 1995; 57(2): 133-44.
Stein G, Lian J, Stein J et al. Transcriptional control of osteoblast growth anddifferentiation. physiol Rev 1996; 76(2):593-629.
Ten Cate AR. The role of epithelium in the development, structure and function ofthe tissues of tooth support. Oral Dis 1996; 2(1 ):55-62.
58
Tokiyasu V, Takata T, Saygin E et al. Enamel factors regulate expression ofgenes associated with cementoblasts. J Periodontol 2000; 71 (12): 1829-39.
Trombelli L, Heitz-Mayfield LJ, Needleman I et al. A systematic review of graftmaterials and biological agents for periodontal intraosseous defects. J ClinPeriodontol 2002; 29(Suppl 3): 117-35; discussion 60-2.
Van der Pauw MT, Van den Bos T, Everts V et al. Enamel matrix-derived proteinstimulates attachment of periodontal ligament fibroblasts and enhances alkalinephosphatase activity and transforming growth factor beta1 release of periodontalligament and gingival fibroblasts. J Periodontol 2000; 71 (1 ):31-43.
Veis A, Tompkins K, Alvares K et al. Specific amelogenin gene splice productshave signaling effects on cells in culture and in implants in vivo. J Bioi Chern2000; 275(52):41263-72.
Viswanathan H, Berry JE, Tokiyasu V et al. Amelogenin:a signaling protein forperiodontal development. J Dent Res 2002; 81 (abstract 2497.
Vivien D, Attisano L, Wrana J et al. Signaling activity of homologous andheterologous transforming growth factor-B receptor kinase complexes. J BioiChem 1995; 270(13):7134-41.
Vojtek AB, Cooper JA. Rho family members: activators of MAP kinase cascades.Cell. 1995; 82(4):527-9.
Wang W. Ectopic bone induction by human fetal enamel proteins. Zhonghua KouQiang Vi Xue Za Zhi 1993; 28(6):362-4.
Wen HB, Moradian-Oldak, J., Fincham, A.G. Dose-dependent modulation ofoctacalcium phosphate crystal habit by amelogenins. J. Dent. Res. 2000;79(11):1902-6.
Wrana J, Attisano L, Wieser R et al. Mechanism of activation of the TGF-Breceptor. Nature 1994; 370(6488):341-7.
Zeichner-David M. Is there more to enamel matrix proteins thanbiomineralization? Matrix Bioi 2001; 20(5-6):307-16.
Zeichner-David M, Diekwisch T, Fincham A et al. Control of ameloblastdifferentiation. Int J Dev Bioi 1995; 39(1 ):69-92.
Zetterstrom 0, Andersson C, Eriksson L et al. Clinical safety of enamel matrixderivative (EMDOGAIN) in the treatment of periodontal defects. J ClinPeriodontol1997; 24(9 Pt 2):697-704.
59