Mechanical Tensile Stress Effects on the Expression of Bone

Sialoprotein in Bovine Cementoblasts

Angle Orthodontist, Vol 79, No 2, 2009

 Introduction
External root resorption is a common phenomenon that is an adverse
effect of orthodontic treatment. Researchers have found that the
magnitude of applied force is strongly related to the volume of
resorption craters. The cementogenesis procedure is very important in
resorption reparation.
As a unique avascular mineralized tissue synthesized by cementoblasts,
cementum plays an important role in anchoring the teeth through the
periodontal ligament (PDL) and protects the integrity of the root
surface. Cementum has a biological composition similar to that of bone
but shows a lack of remodeling. Recent studies have shown that new
cementum was formed at the tension area during orthodontic
intervention, suggesting that the cementum may undergo slow
Cementum is formed by cementoblasts, which are unique phenotypic
cells on the outer layer of the cementum. Cementoblasts share
properties with osteoblast-like cells, such as the ability to form a
mineralized matrix and to express the genes of bone sialoprotein
(BSP), osteopontin (OPN), and osteocalcin (OCN).
Additionally, cementoblasts specifically express cementum-derived
attachment protein (CAP) and cementum-derived growth factor (CGF),
the unique biological markers for cementoblasts.
BSP, one of the major proteins in the cementum, is a highly sulphated,
phosphorylated, and glycosylated protein that plays an essential role in
cementogenesis.
A relationship has been noted between the speed of cementogenesis
and the amount of BSP present in cementum.
Expression of BSP in the cementoblast is regulated by amelogenins in
vitro, basic fibroblast growth factors, and structure or compositional
changes in root surface matrix components.
Mechanical stimuli and up-regulated BSP expression in osteoblasts
have been previously demonstrated.
However, no data are yet available on the effects of mechanical tensile
stress on the expression of mineralization of related genes in
cementoblasts.
Therefore, the aim of the present study was to investigate the effects of
mechanical tensile stress on in vitro BSP expression in cementoblasts.
Our hypothesis was that mechanical stimuli up-regulate BSP mRNA
expression, and that the level of expression is related to the magnitude
of tensile strength.
 Material & method
Cell Cultures -
Healthy teeth and alveolar bone were obtained from six newborn
bovines.
PDL cells and cementoblasts were harvested from the teeth and
osteoblasts from the alveolar bone.
Cementoblasts were loaded with mechanical tensile stress; PDL cells
and osteoblasts were used as controls for cementoblast identification.
After the mandibular teeth had been extracted, PDL tissue and
cementum were separated according to the methods of D’Errico et al.
Separated tissues were digested with 0.1% collagenase I and 0.25%
trypsin (Sigma) with rotation at a speed of 40 rpm at 37C.
PDL tissue was digested for 1.5 hours and was centrifuged at a speed of
1000 rpm for 15 minutes so PDL cells could be collected.
The cementum was digested completely within 4 hours.
After each hour, new digestion solution was added.
Subsequently, the cementum samples were cut into 0.5 x 0.5 mm pieces
for culturing of cementoblasts.
The alveolar bone was minced after soft tissues were carefully removed
for culturing of osteoblasts.
 Cementum and alveolar bone specimens were placed separately in
DMEM/F12 (Gibco, Carlsbad, Calif) containing 100 U/mL of
penicillin G and 100 g/mL of streptomycin (Gibco) and 10% fetal
bovine serum and were incubated in a 25 cm2 flask in the incubator
with 5% carbon dioxide (CO2) at 37C. The medium was changed
every 3 days. When cells grew in up to 80% of flasks, they were
passed with trypsin–ethylenediaminetetraacetic acid (EDTA) at a ratio
of 1:2. The third or fourth passage was used for the experiments.
 Cell Proliferation & Mineralization
assay
Cell proliferation was evaluated by colorimetric MTT [3-(4,5-
dimethylthazol-2-yl)-2,5-diphenyl tetrazolium bromide] assay. Briefly,
cells were plated at 1 104 density at 200 L volume in 96-well plates
(BD Falcon) and were cultured for 1, 2, 3, 4, and 5 days. At the end of
each time point, 100 L (5 mg/mL) of MTT solution (Boehringer
Mannheim, Ingleheim, Germany) was added to the wells and incubated
at 37C for 4 hours. Then 100 L dimethyl sulfoxide (DMSO) (Merck,
Nottingham, UK) was added. The resultant solution was read in a
microplate reader at 570 nm.
To assess cell mineralization, cementoblast, osteoblast, and PDL cells
were grown in differentiated medium (DMEM/F12 supplemented with
1M dexamethazone and 50 g/mL ascorbic acid) in 6-well plates for 5
days, and von Kossa staining was performed.
Briefly, the plates were fixed in phosphated buffered formalin for 10
minutes and were serially dehydrated and rehydrated.
Then 2% silver nitrate was added with sunlight exposure for 20
minutes, after which 5% sodium thiosulfate was added for 3 minutes
and acid fuchsin counterstain for 5 minutes.
Samples were dried for image analysis. Software Quantity One
(BioRad, Hercules) was used to quantify the density of calcium
nodules.
 Immunochemistry
CAP monoclonal murine antibovine antibody was used as the marker
for cementoblasts. Prepared fresh cell slides were fixed in 4%
paraformaldehyde. Endogenous peroxidase activity was inhibited by
30-second exposure to 0.3% hydrogen peroxide (H2O2).
The sections were incubated to block nonspecific binding for 40
minutes in phosphate-buffered saline solution (PBS) with 20% normal
goat serum.
All procedures were carried out at room temperature, and the slides
were washed for 5 minutes with PBS three times after each step was
completed.
The three-step immunoperoxidase method was used to detect CAP. The
sections were blocked with 5% bovine serum albumin (BSA); after this
was added, incubation with the primary antibody occurred at 4C
overnight, then with the secondary horseradish peroxidase (HRP)-
conjugated goat antimurine antibody (BD Falcon) at 37C for 30
minutes; subsequently, a tertiary HRP-conjugated antibody (swine
antigoat) containing 1% BSA and 10% normal human serum (NHS)
was added.
HRP activity was detected with H2O2 used as the substrate and
aminoethylcarbazole as the dye.
Counterstaining was performed briefly with Mayer’s hematoxylin
solution. The sections were mounted for light microscopic
examination, and photomicrographs were taken with a digital camera.
 Application of Mechanical Tensile Stress

The cementoblasts were seeded onto 10 cm2 (5 2 cm) cell culture

plates at a density of 4 104 cells/ cm2 and were cultured in an
incubator with 5% CO2 at 37C.
When the cells had grown up to 80% of the plates, they were subjected
to 2000 or 4000 microstrains with the use of a uniaxial four-point
bending system (Figure 1) in the incubator at a frequency of 0.5 Hz for
3, 6, 12, 24, or 36 hours.
Loading procedures were repeated three times for each time point. In
the control groups, cementoblasts were cultured in the same conditions
but without mechanical stress loading.
 Real-Time Quantitative RT-PCR (qRT-PCR)

Total RNA was extracted with phenol-chloroform solution according to

the manufacturer’s protocol. First-strand cDNA synthesis was carried
out with the use of Superscript RT III. The qRT-PCR was performed in
25 L of a solution containing SYBR green master buffer, 1 M specific
sense and antisense primers, and 100ng cDNA. The primer pairs and
accession numbers were listed. PCR was carried out in a MyIQ
thermocycler (Bio-Rad), and data were analyzed with the use of MyIQ
software (Bio-Rad).
The PCR conditions were 94C for 30 seconds, 62C for 30 seconds, and
72C for 30 seconds with 30 cycles; measurements were taken at the
end of the annealing step at 80C for 20 seconds during each cycle. The
qRT-PCR products were quantified with gylceralde-3-phosphate
 Statistics
Two-way analysis of variance (ANOVA) and Tukey’s post hoc tests
were used for comparisons across time within the same group and
between different groups at the same time point. Differences were
considered significant when P < 0.05.
 Cell Proliferation and Mineralization

The cementoblasts showed a tendency toward slower growth compared

with the osteoblasts and PDL cells (Figure 2A).
The density of calcium nodules formed in cementoblasts was found to
be between that of osteoblasts and PDL cells (Figure 2B,C,D).
 BSP mRNA Expressions

BSP mRNA expressions did not change with time in the control
groups. In both 2000 and 4000 microstrain groups, BSP mRNA
showed significant up-regulation with time (P .01) and peak expression
at 24 hours.
However, with 2000 and 4000 microstrains, tenfold and sixfold
increases were noted, respectively, compared with the level that was
present before mechanical loading; at 36 hours, the BSP levels in both
groups decreased significantly (P <0.05).
At all time points, BSP expressions at the mechanical loading groups
were significantly higher than at the non-loading groups (P<.01 at 3, 6,
and 12 hours; P< 0.05 at 36 hours). Significant differences were noted
between the two loading strains at 6, 12, and 24 hours (P <0.01).
 Discussion
Cementoblasts are known for the difficulty associated with obtaining
and culturing them. Several methods, such as the use of osteocalcin
TAg (OC-TAg) transgenic mice, transfection of the thermolabile SV40
T-antigen into root lining cell subpopulations, and isolation and clone
expansion in human cementum-derived cells, have been introduced in
the literature.
However, each method has significant drawbacks; for instance, it
might alter the nature of the cementoblast, or it may make it difficult
for the cementoblast to proliferate, and viability is often insufficient.
In addition, SV40 Tag-transfected cementoblasts must be cultured at
33C with interferon-(INF), which down-regulates osteocalcin
expression and subsequently affects its signaling pathway.
The present study used newborn bovine teeth for cementoblast culture.
The advantage is that the roots were still forming, and therefore, many
functional and active cementoblasts were embedded in the cementum.
Electron microscopy showed that the cementum surface contained only
embedded cementoblasts after 4 hours predigestion. Compared with
the above-mentioned methods, the cementoblasts obtained by our
method were relatively easy to cause to proliferate, and they kept their
in vivo cellular characteristics. In the present study, the cementoblast
was positive to CAP antibody, and PDL cells and osteoblasts were
negative. This confirmed the purity of the cementoblasts.
Our results showed that mechanical tensile stress up-regulated BSP
mRNA expression in cementoblasts in vitro. Expression of BSP
reached a peak level and then decreased toward baseline level,
irrespective of stress magnitudes.
Previous studies also showed that mechanical stress up-regulated BSP
expression in osteoblasts. However, after the peak level was reached,
BSP expression stayed at a plateau without an obvious decrease in
osteoblasts.
The variable intensity of BSP expression in the present study and in
previous studies on osteoblasts under mechanical tensile stress stimuli
might be related to variations in cell phenotype.
Moderate mechanical stress is the key factor in maintaining bone mass.
Cell surface receptors such as integrins, cell adhesion molecules, and
ion-activated channels trigger mechanically induced signal
transductions within the cell. To study mechanical stimulation of cells
in vitro, several in vitro models have been developed.
A uniaxial four-point bending system used in the present study is a
longitudinal stretch system. It is easy to apply the force, and the
technique has the capability for longer mechanical loading.
The present study is the first to show that mechanical tensile stress up-
regulated BSP mRNA expression in cementoblasts, and that a high
magnitude of tensile stress induced less BSP up-regulation compared
with a low magnitude.
The up-regulated BSP mRNA might have been induced by
prostaglandin E2 (PGE2) in cementoblasts. Previous studies reported
up-regulation of BSP expression in osteoblasts resulting from
compressive forces induced by autocrine actions of PGE2. Because
cementoblasts share many common properties with osteoblasts, one
would expect to observe similar mechanisms in cementoblasts in
response to mechanical stress; this, however, needs further
investigation.