Endocrinology Vol. 141, No. 12 4671-4680
Copyright © 2000 by The Endocrine Society
Parathyroid Hormone-Related Protein Down-Regulates Bone Sialoprotein Gene Expression in Cementoblasts: Role of the Protein Kinase A Pathway1
Hongjiao Ouyang,
Renny T. Franceschi,
Laurie K. McCauley,
Dian Wang and
Martha J. Somerman
Departments of Periodontics/Prevention/Geriatrics
(H.O., R.T.F., L.K.M., D.W., M.J.S.), and Restorative
Sciences/Cardiology/Endodontics (H.O.), School of
Dentistry, Departments of Biochemistry (R.T.F.), and
Pharmacology (M.J.S.), School of Medicine, The University
of Michigan, Ann Arbor, Michigan 48109-1078
Address all correspondence and requests for reprints to: Hongjiao Ouyang, D.D.S., Ph.D., Departments of Periodontics/Prevention/Geriatrics and Restorative Sciences/Cardiology/Endodontics, The University of Michigan, 1011 North University Avenue, Ann Arbor, Michigan 48109-1078. E-mail:
hongjiao{at}umich.edu
 |
Abstract
|
|---|
PTH-related protein (PTHrP) acts as a paracrine and/or autocrine
regulator of cell proliferation, apoptosis, and differentiation and is
implicated in tooth development. The current studies employed
cementoblasts to determine the role(s) and mechanisms of PTHrP in
regulating cementum formation. Results demonstrated that PTHrP
repressed gene expression and protein synthesis of bone sialoprotein
(BSP) and abolished cementoblast-mediated biomineralization in
vitro. The BSP gene inhibition required protein synthesis. The
PTHrP analog (131) and other activators of the PKA pathway
(3-isobutyl-1-methylxathine (IBMX), forskolin (FSK) and
Sp-Adenosine-3', 5'-cyclic monophosphorothioate (Sp-cAMPss) also
down-regulated BSP gene expression and blocked cementoblast-mediated
biomineralization. In contrast, the PTHrP analog (734), a PTHrP
antagonist, and the activators of the PKC pathway [phorbol
12-myristate 13-acetate (PMA) and phorbol 12, 13-dibutyrate (PDBu)]
promoted BSP gene expression. In addition, the PKA pathway inhibitor
(9-(2-tetrahydrofuryl) adenine (THFA) partially, but significantly
reversed the PTHrP-mediated down-regulation of BSP gene expression.
Furthermore, THFA alone significantly increased BSP messenger RNA
(mRNA) expression in cementoblasts. In contrast, the inhibitor of the
PKC pathway (GF109203X) did not reverse the PTHrP inhibitory effect on
BSP gene expression. Furthermore, GF109203X alone dramatically reduced
the BSP transcript levels. These data indicate that the cAMP/PKA
pathway mediates the PTHrP-mediated down-regulation of BSP mRNA
expression in cementoblasts; and furthermore, this pathway may, through
an intrinsic inhibition mechanism, regulate the basal level of BSP mRNA
expression. In contrast, the activation of PKC promotes BSP gene
expression. These data provide new insights into the molecular
mechanisms involved in PTHrP regulation of cementogenesis.
 |
Introduction
|
|---|
PTH-RELATED PROTEIN (PTHrP) was initially
isolated and cloned from tumors due to its PTH-like properties and
resultant humoral hypercalcemia of malignancy (1, 2). Subsequently, it was recognized that PTHrP is
associated with a variety of normal tissues/organs and acts as an
important autocrine and/or paracrine factor of cell proliferation,
apoptosis, and differentiation (3). Overexpression of
PTHrP in chondrocytes in vivo causes a delay in their
differentiation and in mineralization of cartilage (4). In
contrast, but consistent with these findings, disruption of either
PTHrP (5) or PTH-1 receptor (6) genes leads
to an accelerated differentiation of chondrocytes. These studies
demonstrated that PTHrP plays a pivotal role in regulating formation of
the skeletal tissues. Although the regulatory functions of PTHrP during
the development of the skeletal system have been well established, the
role of PTHrP in regulating other mineralized tissues, such as those
associated with teeth, is poorly understood. Tooth-related mineralized
tissues differ from bone in many aspects, such as histological features
and physiological function (7). Therefore, understanding
the role of PTHrP in controlling formation and maintenance of these
tooth-related mineralized tissues will expand the knowledge of
regulatory mechanisms involved in PTHrP effects on mineralized tissue
formation.
Cementum is a thin mineralized tissue covering the tooth root surface.
It differs from bone in that it lacks innervation and vascularization
(7). Furthermore, cementum has limited remodeling
potential when compared with bone (7). The unique
functions of cementum are related to its role as an important component
of the periodontal attachment apparatus (the periodontium), anchoring
teeth to the surrounding alveolar bone, and thereby contributing to the
maintenance of structural stability and physiologic function of the
dentition. Although cementum has these important functions, the
molecular mechanisms involved in regulating the formation of cementum
remain largely unknown. Accumulating evidence indicates that PTHrP may
serve as a critical paracrine and/or autocrine regulator of
cementogenesis. These findings include 1) during cementogenesis, PTHrP
is produced by cells associated with the periodontium, such as
Hertwigs epithelial cells, osteoblasts lining the surrounding
alveolar bone, and cementoblasts themselves (8); 2)
cementoblasts possess PTH/PTHrP receptors (9, 10); and 3)
PTHrP deficiency of certain stains of PTHrP knock-out mice results in
the failure of tooth eruption and in fusion of cementum with
surrounding alveolar bone (11). To further elucidate the
function of PTHrP in cementogenesis, it is important to determine how
PTHrP regulates cementoblast activities, such as their gene expression
and biomineralization ability. The current studies focus on identifying
the signal transduction pathways underlying PTHrP regulation of
cementoblast gene expression in vitro.
It is well established that PTHrP and PTH interact with a G
protein-coupled receptor, the PTH-1 receptor. The PTH-1 receptor-G
protein complex activates two well-defined signal transduction pathways
(12). Activation of adenylyl cyclase and phospholipase C
leads to production of cAMP, inositol phosphates (IPs) and
diacylglycerol (DAG), respectively. These intracellular mediators
ultimately exert their functions through activation of various protein
kinase complexes, notably, protein kinase A (PKA) by cAMP and protein
kinase C (PKC) by DAG. PKA and PKC complexes can subsequently
phosphorylate specific transcription factors and consequently affect
transcription of specific genes. PKA mainly targets members of the cAMP
response element binding protein families (CREBs and ATF members),
whereas PKC mainly acts on members of the AP-1 family. The selective
stimulation of PKA and/or PKC pathways and the resulting specific
activation of transcription factors generates the diversity of actions
by which different genes, i.e. extracellular matrix
molecules, may be regulated in a tissue/cell type-specific manner in
response to PTH/PTHrP.
The immortalized cementoblast cell lines used in the current study were
obtained from tooth root surfaces of OC-TAg transgenic mice, which
carry an osteocalcin promoter driving the SV-40 large T antigen gene to
ensure the selective immortalization of OCN expressing cells (13, 14). We have shown that these cementoblast cell lines closely
resemble their in vivo counterparts in many properties, such
as gene expression and ability to mineralize. Relevant to the current
study is that these cementoblast cell lines express PTH-1 receptor
messenger RNA (mRNA), and exhibit a PTHrP-induced cAMP elevation and
c-fos gene expression, thereby providing an in
vitro model to study the effects of PTHrP on cementoblast activity
(15). Moreover, PTHrP (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34) represses the expression of
cementum-associated gene markers in these cells, such as BSP, OCN, and
type I collagen (15). BSP was selected for the current
studies because 1) the temporo-spatial gene expression pattern of BSP
closely correlates with the initiation of cementum and bone formation
(16, 17); and 2) BSP protein is thought to play a critical
role in promoting biomineralization (18, 19, 20). The current
investigation focused on determining the molecular mechanisms
underlying PTHrPs effect on BSP gene expression in cementoblasts
in vitro.
 |
Materials and Methods
|
|---|
Cell culture and experimental design
The cementoblast cell line, OC-CM 30 (15), was
cultured in DMEM (Life Technologies, Inc., Gaithersburg,
MD) supplemented with 10% FBS (Life Technologies, Inc.),
100 U/ml penicillin and 100 g/ml streptomycin (BioWhittaker, Inc., Walkersville, MD) in a humidified atmosphere of 5%
CO2 at 37 C. This clonal population was selected
for the current studies based on its robust expression of PTH-1
receptor mRNA and responses to PTH/PTHrP stimulation, in
vitro (15).
Dose-dependent effect of PTHrP
Cementoblasts were plated as described above, and upon
confluence, were treated with PTHrP (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34) (Bachem California, Inc., Torrance, CA) at concentrations ranging from
10-13 M to
10-6 M for
24 h. At these doses, cell viability was maintained as monitored
by morphological observation and by cell proliferation assay (data not
shown). To determine steady-state BSP mRNA levels, total RNA was
isolated from PTHrP- and vehicle- (1 mg/ml bovine albumin/4
mM HCl) treated cultures and analyzed by Northern blotting.
At the doses of PTHrP used for these studies, cell viability was
maintained.
Time-course experiment
To determine the temporal effect of PTHrP on BSP mRNA
expression, cementoblasts were plated at 5 x
104 cells/cm2 in 100-mm
tissue culture plates and grown in DMEM media as described above. At
confluence (designated as 0 h), cells were treated with human
PTHrP (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34) (10-7
M) (hPTHrP, Bachem California, Inc.) and total
RNA extracted at 0, 2, 6, 12, and 24 h for determination of the
steady-state BSP mRNA levels by Northern blot analysis. The specific
dose of PTHrP was selected due to its prominent inhibition of BSP gene
expression.
Effect of cycloheximide
The effect of cycloheximide (Sigma, St. Louis, MO)
(21) on PTHrP-mediated inhibition of BSP gene expression
was also determined. Confluent cementoblasts were exposed to either
vehicle, or 10-7
M PTHrP (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34), or
10-7 M PTHrP
(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34) plus cycloheximide (10 µg/ml), or cycloheximide (10 µg/ml)
alone for 9 h, followed by total RNA extraction and Northern blot
analysis to measure steady-state BSP mRNA levels.
Effect of actinomycin D
To determine whether PTHrP affects the stability of BSP mRNA,
the decay rates of BSP mRNA in vehicle- and PTHrP-pretreated groups
were measured. Confluent cementoblasts were pretreated with either
vehicle or PTHrP (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34)
(10-7 M) for
6 h before transcription inhibition. Then, media were removed and
cells were stimulated with vehicle or PTHrP
(10-7 M),
either in the presence or in the absence of actinomycin D
(Sigma, 1 µg/ml), an inhibitor of transcription
(22), for designated time intervals (0, 3, 6, 9, and
12 h). Total RNA was harvested and Northern blot analysis
performed to follow the decay course of BSP mRNA associated with
different treatment conditions. Data are presented as logarithmic
values of mean ± SD of cpm of BSP mRNA after being
normalized to 18S rRNA.
Dissection of signal transduction pathway(s)
Three strategies were employed to identify the signal
transduction pathway(s) involved in PTHrP regulation.
First, the effect of PTHrP analogs on BSP mRNA expression was examined.
Confluent cementoblasts were treated with human PTHrP (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31), or
(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34), or (7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34) (Bachem California, Inc.)
(23) at a concentration of
10-7 M for
24 h and total RNA was extracted for Northern blot analysis.
In a parallel approach, the effects of various activators of PKA/PKC
pathways on BSP mRNA expression in cementoblasts were analyzed. At
confluence, cementoblasts were treated with various activators for
24 h, and total RNA was extracted for Northern blot analysis.
Activators used included 3-isobutyl-1-methylxathine (IBMX, 10
µM) (Sigma) (24),
Sp-Adenosine-3', 5'-cyclic monophosphorothioate (Sp-cAMPss) (10
µM) (BIOMOL Research Laboratories, Inc.,
Plymouth, PA) (25) or forskolin (FSK, 10 µM)
(Sigma) (26) for activating the PKA pathway,
and phorbol 12, 13-dibutyrate (PDBu 10 ng/ml) (BIOMOL Research Laboratories, Inc.) (27, 28, 29) and phorbol
12-myristate 13-acetate (PMA 10 µM) (BIOMOL Research Laboratories, Inc.) (27, 28, 29) for
stimulating the PKC pathway.
Additionally, the effects of inhibitors of PKA/PKC pathways on BSP mRNA
expression in the presence and absence of PTHrP were determined.
Confluent cementoblasts were pretreated with an inhibitor for 12 h
before treatment with either PTHrP (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34)
(10-7 M) or
vehicle in the presence of the inhibitor for an additional 12 h.
Total RNA was extracted and Northern blot analysis performed to
determine the level of BSP mRNA. A cell-permeable adenylyl cyclase
inhibitor, 9-(2-tetrahydrofuryl) adenine (THFA) (100 µM)
(Sigma) (24, 30), was used to inhibit the
activity of cAMP/PKA pathway, whereas GF109203X (3 µM)
(BIOMOL Research Laboratories, Inc.) (31) was
used to repress/block the PKC pathway.
Northern blot analysis
Total RNA was isolated from cells and quantitated by
spectrophotometry as described (32). Ten micrograms of
total RNA was electrophoresed on 1.2% agarose-formaldehyde gels,
blotted to Duralon-UV membranes Stratagene, La Jolla, CA),
and immobilized by UV light, using Stratagenes Stratalinker. The
membranes were hybridized with a labeled cDNA probe. The probes were
labeled with
-32P-dCTP (Amersham Pharmacia Biotech, Arlington Heights, IL) using a Rediprime
labeling kit (Amersham Pharmacia Biotech). After
hybridization and washing, blots were exposed to Kodak
X-OMAT film at -70 C with intensifying screens for 2448 h. Blots
were subsequently stripped and reprobed with either GAPDH and/or 18S
rRNA cDNA probe to standardize RNA loading and hybridization
efficiency. Blots were quantitatively scanned by a phosphoImager,
Packard A2024 InstantImager (Downers Grove, IL); and the signals were
normalized by calculating the ratio of signal intensities of the
assayed markers vs. either the internal control GAPDH mRNA
or 18S rRNA. For most of the studies, GAPDH and/or 18S rRNA were used
for normalization and to rule out possible nonspecific effects of
agents used. No differences were noted for using GAPDH vs.
18S rRNA. Therefore, in some experiments, GAPDH data are presented,
whereas in other situations 18S rRNA was presented. Probes used
included: 1) a mouse BSP cDNA probe cloned into PCR II vector
(33); 2) a 1100 bp chicken GAPDH cDNA cloned into pGEM3
(34); and 3) an 18S rRNA cDNA in PBR 322
(35).
Western immunoblot analysis
The procedures were described previously (36).
Briefly, cells were plated at a density of 5 x
104 cells/cm2 in 60-mm
culture plates and incubated in DMEM containing 10% FBS. On reaching
confluence, media were removed and cells were exposed to either vehicle
or 10-7 M
hPTHrP (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34) for 24 h in serum-free DMEM. Conditioned media were
collected, lyophilized, and subsequently dissolved in 1x SDS-PAGE
loading buffer. A 10 µg aliquot of protein was fractionated by
SDS-PAGE. LF-6 anti-BSP antibody (37), diluted at a ratio
of 1:1000, was used to detect BSP protein. A secondary antibody
(horseradish peroxidase-conjugated goat antirabbit immunoglobin G) was
used at a dilution of 1: 10,000. Hybridization with antibodies was
performed as previously reported (38). Immunoreactivity
was determined by an enhanced chemiluminescence reaction assay
(Amersham Pharmacia Biotech).
Adenylyl cyclase stimulation assay
Adenylyl cyclase stimulation assays were performed as described
previously (39). Briefly, the OC-CM 30 cells were plated
in DMEM media in 24-well tissue culture plates in triplicate, at 5
x 104 cells/cm2. At
confluence, cells were stimulated with either vehicle, or PTHrP
(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31), or (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34) or (7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34) (Bachem California, Inc.) for 15 min at 37 C in calcium- and magnesium-free HBSS
(Life Technologies, Inc.) containing 0.1% BSA and 1
mM IBMX (Sigma), a phosphodiesterase
inhibitor. Intracellular cAMP was extracted and measured, using a cAMP
binding protein assay (39). The cAMP levels were
calculated by a log-logit method (Graph Pad Prism, normalized to cell
number, and expressed as pmol/105 cells.
In vitro mineralization assay
The in vitro mineralization assay was performed as
reported previously (40). Cells were plated at 5 x
104 cells/cm2 in 24-well
plates with DMEM containing 10% FBS and supplemented with 50 µg/ml
ascorbic acid. When cell reached confluence, designated day 0, medium
was removed and cells were incubated in mineralization media (ascorbic
acid-free
-MEM) (Life Technologies, Inc.) containing
10% FBS and supplemented with 50 µg/ml ascorbic acid) for 7 days,
with inorganic phosphate (3 mM) added on day 6,
to induce mineral nodule formation. Samples were processed on day 7 and
von Kossa staining used to detect mineral nodule formation. To
determine the effect of various agents on cementoblast-mediated
mineralization, either vehicle or test agent was added to cultures on
day 6. On day 7, samples were assayed for mineral deposition.
Statistical analysis
All experiments were performed a minimum of two times with
comparable findings obtained. For cAMP stimulation assays, a Students
t test was conducted to analyze the results (Instat 2.0,
GraphPad Software, Inc., San Diego, CA). For mRNA decay
rate measurements, the data were analyzed by either ANOVA followed by a
Tukey-Kramer multiple comparison test and a Students t
test as needed. For Northern blot analyses, representative
autoradiographs are shown with a graphical analysis demonstrating
statistical evaluation of multiple assays.
 |
Results
|
|---|
Dose-dependent inhibitory effect of PTHrP on BSP gene
expression
Northern blot analysis demonstrated that PTHrP (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34) exhibited a
dose-dependent inhibition of BSP mRNA expression in cementoblasts; a
significant inhibitory effect was first detected at
10-9 M
(P < 0.001) with a greater magnitude of inhibition
associated with higher concentrations
(10-8,
10-7, and
10-6
M) of PTHrP (Fig. 1
, A and B). Given that BSP protein has
been implicated in initiating biomineralization (18, 20, 41), the ability of PTHrP at these same concentrations to
regulate cementoblast-mediated biomineralization also was determined.
As revealed by a von Kossa assay, PTHrP (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34) abolished
cementoblast-induced formation of mineralized nodules in a
dose-dependent fashion with the first detectable inhibition of
mineralization noted at
10-9
M (Fig. 1C
). Furthermore, the dose-dependent
profile of PTHrP effects on both BSP gene expression and
biomineralization were consistent with the dose-dependent effect of
PTHrP (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34) on elevating cAMP synthesis in cementoblasts (Fig. 1D
and
Ref. 15).

View larger version (39K):
[in this window]
[in a new window]
|
Figure 1. Dose-dependent effects of PTHrP on cementoblast
activities. A, A representative Northern blot autoradiograph of BSP
mRNA and 18S rRNA levels associated with 24 h PTHrP treatment at
concentrations ranging from 10-13
M to 10-6 M. BSP mRNA
of murine cementoblasts runs slightly above 18S rRNA with sizes
approximating 2.0 kb. B, Graphical analysis of cpm (mean values
expressed as BSP mRNA normalized to 18S rRNA ± SD).
PTHrP (1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 ) exhibited a dose-dependent inhibitory effect on BSP gene
expression by cementoblasts; and compared with vehicle (V), PTHrP at
doses from 10-9 M to
10-6 M, significantly
down-regulated BSP mRNA levels (* P < 0.001
vs. vehicle). The experiments were performed twice in
duplicate with similar results noted. C, Dose-dependent effect of PTHrP
on cementoblast-mediated biomineralization, in vitro.
When cultured for 7 days in the presence of ascorbic acid (50 µg/ml)
and incubated for the last 24 h with 3.0 mM inorganic
phosphate (Pi), vehicle-treated cementoblasts promoted mineral nodule
formation. However, when present for the last 24 h, PTHrP (1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 )
blocked mineralization in a dose-dependent manner with a dose as low as
10-9 M having an inhibitory
effect. This experiment was performed twice in duplicate with
comparable results obtained. D, PTHrP dose-dependent cAMP response:
cementoblasts were exposed to PTHrP (1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 ) at increasing
concentrations ranging from 10-11
M to 10-6 M for 15
min. cAMP levels were measured by a cAMP stimulation assay and the
values were expressed as the mean ± SEM pmol
cAMP/105 cells. PTHrP significantly stimulated cAMP
elevation in cementoblasts in a dose-dependent manner (*
P < 0.001 significance vs. basal
level of cAMP). Experiments were performed three times with similar
results noted.
|
|
Time course of PTHrP-induced inhibition of BSP mRNA
The temporal effect of PTHrP on BSP gene expression in
cementoblasts was determined. Cementoblasts were exposed to either
vehicle or 10-7
M PTHrP for the designated time intervals up to 24 h
and BSP mRNA levels were determined by Northern blot analysis. As
demonstrated in Fig. 2
, a detectable
inhibition of BSP mRNA (P < 0.001) was first
noted at 6 h, with more prominent inhibition noted at 12 and
24 h.

View larger version (61K):
[in this window]
[in a new window]
|
Figure 2. Time-dependent effect of PTHrP on BSP mRNA
expression by cementoblasts. Cementoblasts were plated at an initial
density of 5 x 104 cells/cm2. At
confluence, designated as 0 h, either vehicle (V) or PTHrP (1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 )
(10-7 M) (P) was added followed by
isolation of total RNA at designated time points (0, 2, 6, 12, 24
h). A, A representative Northern blot autoradiograph of BSP mRNA and
18S rRNA levels. 18S rRNA hybridization is shown to demonstrate
relative loading and blotting efficiency. B, Graphical analysis of BSP
mRNA levels. Plot of cpm (mean values expressed as BSP mRNA normalized
to 18S rRNA ± SD). Compared with vehicle, PTHrP
significantly down regulated BSP mRNA levels at 6, 12, and 24 h (*
P < 0.001). The experiments were performed four
times with similar results observed.
|
|
Effect of PTHrP on BSP protein synthesis
To determine whether alterations in BSP protein levels in
PTHrP-treated cementoblasts correlated with changes in mRNA levels, the
effect of PTHrP (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34) on BSP protein synthesis was examined.
Cementoblasts were treated with
10-7 M PTHrP
(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34) for 24 h. Western blot analysis revealed that, in
comparison to vehicle, PTHrP (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34) dramatically reduced BSP protein
levels, correlating with the inhibitory effect of PTHrP on BSP mRNA
expression (Fig. 3
).

View larger version (44K):
[in this window]
[in a new window]
|
Figure 3. Effect of PTHrP on BSP protein level in
cementoblasts. Confluent cementoblasts were treated with either vehicle
or 10-7 M PTHrP (1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 ) for
24 h in serum-free DMEM media. Media were collected and BSP
protein levels were measured by Western blot analysis as described in
Materials and Methods section. MC3T3E1 cell conditioned
medium served as a positive control. The size of the BSP protein is
approximately 70 kDa. Experiments were done twice with one performed in
duplicate.
|
|
Effect of cycloheximide on PTHrP-mediated inhibition of BSP mRNA
expression
To determine whether the effect of PTHrP on BSP mRNA expression
required protein synthesis, cycloheximide (10 µg/ml)
(21), a protein synthesis inhibitor, was used. Confluent
cementoblasts were treated with vehicle, PTHrP, PTHrP plus
cycloheximide, or cycloheximide alone for 9 h. As illustrated in
Fig. 4
, PTHrP significantly decreased BSP
mRNA expression (1 P < 0.001 vs. vehicle).
In contrast, cycloheximide partially (** P < 0.05
vs. vehicle) but significantly reversed the PTHrP-mediated
inhibition of BSP gene expression (*** P < 0.01
vs. PTHrP only). Cells exposed to cycloheximide for 9 h
did not exhibit any visible morphological changes.

View larger version (44K):
[in this window]
[in a new window]
|
Figure 4. Effect of cycloheximide on PTHrP-mediated BSP mRNA
down-regulation. Confluent cementoblasts were treated for 9 h with
either vehicle (V), or PTHrP (1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 ) (10-7
M), or PTHrP (1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 ) (10-7
M) plus cycloheximide (cyclo. 10 µg/ml), or cycloheximide
(10 µg/ml) alone, followed by total RNA extraction and Northern blot
analysis. A, A representative Northern blot autoradiograph of BSP mRNA
and 18S rRNA levels. B, Graphic analysis of normalized BSP mRNA levels.
Values are expressed as mean values of the normalized BSP mRNA
vs. 18S rRNA ± SD. The BSP mRNA levels
were significantly reduced in PTHrP- and PTHrP plus
cycloheximide-treated groups (* P < 0.001 for
PTHrP vs. vehicle control and ** P
< 0.05 for PTHrP plus cycloheximide vs. vehicle
control). Nevertheless, cycloheximide significantly reversed the
PTHrP-mediated repression of BSP gene expression (***
P < 0.01 for PTHrP plus cycloheximide
vs. PTHrP). Experiments were performed in duplicate on
two separate occasions with similar results observed.
|
|
Effect of actinomycin D on PTHrP-mediated inhibition of BSP
mRNA
Inhibition of BSP gene expression can result from inhibition of
transcription and/or an increase in mRNA degradation. To investigate
whether reduction in BSP mRNA level resulted from a decrease in BSP
mRNA stability, the decay rates of BSP mRNA associated with vehicle and
PTHrP treatments were determined. As illustrated in Fig. 5
, in untreated cells, BSP mRNA remained
stable over a 12-h time period. In the presence of the transcriptional
inhibitor actinomycin (Act D), the BSP mRNA decay curves for the
vehicle- and PTHrP-treated samples revealed the half-life of BSP mRNA
was approximately 6.5 and 7.2 h for vehicle- and PTHrP-treated
groups, respectively. Statistical analysis revealed no difference in
BSP mRNA decay rate between these two groups (P >
0.5).

View larger version (23K):
[in this window]
[in a new window]
|
Figure 5. Effect of actinomycin D on PTHrP-mediated
inhibition of BSP gene expression. BSP mRNA stability was analyzed in
the presence and absence of PTHrP. Cementoblasts were pretreated with
either vehicle or PTHrP (10-7 M)
for 6 h. Media were then changed (designated as 0 h) and the
cementoblasts were treated with either vehicle or PTHrP
(10-7 M) in either the presence or
the absence of actinomycin D for designated time intervals. Total RNA
was extracted at 0, 3, 6, 9, and 12 h and Northern analyses
performed to detect BSP gene expression. y-axis, Log of
mean values of BSP mRNA vs. 18S rRNA ±
SD. Line V, The basal level BSP gene expression in the
absence of transcription inhibitor actinomycin D; line V+Act. D, the
decay curve of BSP mRNA in vehicle-treated group with a half-life
(t1/2) of 6.5 h in the presence of actinomycin D; line
PTHrP, PTHrP-mediated down-regulation of BSP gene expression in the
absence of actinomycin D; line PTHrP+Act. D, the decay curve of BSP
mRNA in PTHrP-treated group with a half-life (t1/2) of
7.2 h in the presence of actinomycin D. Experiments were performed
in duplicate on two separate occasions with similar results observed.
Statistical analyses revealed no significant difference in BSP mRNA
decay rate between vehicle- and PTHrP-treated group
(P > 0.5).
|
|
Effect of PTHrP analogs on BSP gene expression
PTHrP (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34) activates both the cAMP/protein kinase A (PKA) and
protein kinase C (PKC) pathways (12). Based on the finding
that PTHrP exhibited a similar dose-dependent profile in inhibiting BSP
gene expression and in stimulating cAMP levels in cementoblasts (Fig. 1
), it was hypothesized that the cAMP-dependent PKA pathway played a
dominant role in mediating the inhibitory effect of PTHrP (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34) on
BSP gene expression. As an initial strategy to test this hypothesis,
the effect of PTHrP analogs on steady-state BSP mRNA levels in
cementoblasts was investigated. Results indicated that PTHrP (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31),
reported to activate the PKA cascade only (23), mimicked
PTHrP (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34) in inhibiting BSP gene expression (Fig. 6
, A and B) (1 P <
0.001), in stimulating cAMP levels (Fig. 6C
) (1 P <
0.001), and in blocking cementoblast-mediated mineralization (Fig. 6C
).
In contrast, PTHrP (7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34), an antagonist for PTHrP (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34)
(23), failed to stimulate cAMP levels (Fig. 6C
), repress
BSP mRNA expression (Fig. 6A
), and block biomineralization (Fig. 6D
).
Instead, PTHrP (7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34) promoted BSP gene expression when compared with
vehicle (Fig. 6
, A and B) (** P < 0.01).

View larger version (29K):
[in this window]
[in a new window]
|
Figure 6. Effects of PTHrP analogs on cementoblast
activities. A and B, Role of PTHrP analogs in regulating BSP gene
expression. Confluent cementoblasts were treated with either vehicle
(V) or different PTHrP analogs at 10-7
M for 24 h and total RNA was extracted to measure the
steady-state BSP mRNA levels using Northern blot analysis. A, A
representative Northern blot analysis autoradiograph of BSP mRNA and
GAPDH mRNA levels. B, Plot of cpm (mean values expressed as BSP mRNA
vs. GAPDH mRNA ± SD). PTHrP (1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 )
and (1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 ) significantly reduced BSP mRNA level (*
P < 0.001 vs. vehicle), whereas
PTHrP (7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 ) promoted BSP mRNA expression (** P <
0.01 vs. vehicle). C, Effect of PTHrP analogs on cAMP
synthesis: Cementoblasts were stimulated with either
10-7 M of either PTHrP (1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 ),
or (1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 ), or (7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 ) for 15 min. cAMP levels were
measured as described in the Materials and Methods
section. PTHrP (1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 ) and (1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 ) significantly stimulated cAMP levels
in cementoblasts (* P < 0.001 vs.
vehicle), whereas PTHrP (7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 ) did not increase cAMP levels.
Experiments were performed three times with similar results noted. D,
Effect of PTHrP analogs on cementoblast-mediated biomineralization,
in vitro. When cultured for 7 days in the presence of
ascorbic acid and incubated for the last 24 h with inorganic
phosphate (Pi; 3 mM), cementoblasts promoted mineral nodule
formation. PTHrP (1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 ) and (1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 ) at 10-7
M, when present for the last day, blocked mineralization.
In contrast, PTHrP (7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 ) at 10-7
M did not prevent mineral nodule formation under similar
treatment conditions. The experiment was performed three times. C,
negative control without A. A, Induction; V, vehicle plus A. A; 131,
PTHrP (1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 ); 134, PTHrP (1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 ); 734, PTHrP (7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 ).
|
|
Effects of second message modulators of PKA/PKC pathways on BSP
mRNA
As an additional strategy to elucidate the signal transduction
pathway(s) involved, the effects of the PKA/PKC pathway activators on
BSP gene expression were examined. IBMX (a phosphodiesterase inhibitor)
(24), forskolin (a cAMP synthesis activator)
(26), and Sp-cAMPss (a cAMP analog) (25),
were used to activate the PKA pathway. The phorbol 12, 13-dibutyrate
(PDBu) and phorbol 12-myristate 13-acetate (PMA), two phorbol esters
that dramatically increase the affinity of PKC for
Ca2+ resulting in its full activation
(27, 28, 29), were employed to stimulate the PKC pathway. As
illustrated in Fig. 7
, A and B, mimicking
the effect of PTHrP (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34) on cementoblasts, IBMX (1 mM),
Sp-cAMPss (10 µM), and FSK (10 µM) reduced
BSP mRNA levels to approximately 35%, 23%, and 2% of their basal
levels, respectively (1 P < 0.001). In contrast, PMA
and PDBu were ineffective in inhibiting steady-state BSP mRNA; instead,
they up-regulated BSP mRNA by approximately 1.8- and 1.9-fold,
respectively (** P < 0.01). In addition, the
effect of PKA/PKC activators on BSP mRNA levels correlated with their
action on cementoblast-mediated biomineralization, in vitro.
FSK not only reduced BSP gene expression but also abolished formation
of mineral nodules in cementoblast cultures (Fig. 7C
), whereas PDBu
failed to inhibit biomineralization (Fig. 7C
) as it failed to decrease
BSP gene expression (Fig. 7
, A and B).

View larger version (45K):
[in this window]
[in a new window]
|
Figure 7. Effect of PKA/PKC pathway activators on
cementoblast activities. A and B, Roles of PKA/PKC pathway activators
in regulating steady-state BSP mRNA levels. A, Representative
autoradiographs of Northern blot analysis: GAPDH was used to
standardize RNA loading. B, A plot of mean values of
relative expression of BSP mRNA vs. GAPDH mRNA ±
SD. The activators of the PKA pathway, IBMX (10
mM), Sp-cAMPss (10 µM)
and forskolin (10 µM) significantly decreased
BSP mRNA expression. In contrast, the activators of the PKC pathway,
PDBu (100 ng/ml) and PMA (100 nM), promoted BSP
gene expression by approximately 1.8- and 1.9-fold, respectively (**
P < 0.01 vs. vehicle). IBMX,
3-isobutyl-1-methylxanthine; PDBu, phorbol 12, 13-dibutyrate; PMA,
phorbol 12-myristate 13-acetate. C, Effects of activators on
cementoblast-mediated biomineralization, in vitro.
Twenty-four-hour treatment with forskolin
(10-5
M) abolished formation of mineral nodules in
cementoblast cultures. In contrast, PDBu (100 ng/ml) did not inhibit
mineralization under the similar treatment condition. V, Vehicle.
Experiments were performed three times with similar results observed.
|
|
To confirm the involvement of PKA/PKC pathway(s), the ability of
specific inhibitors of the PKA/PKC pathway to affect PTHrP action on
BSP gene expression was evaluated. Consistent with the inhibitory role
of cAMP/PKA stimulators, such as IBMX, Sp-cAMPss and forskolin, on BSP
mRNA expression, 12 h pretreatment with a cell-permeable adenylate
cyclase inhibitor THFA (24, 30) partially but
significantly rescued PTHrP-induced down-regulation of the BSP
transcripts (Fig. 8
) (**
P < 0.01). Furthermore, THFA alone, compared with
vehicle, elevated the steady-state mRNA levels by 1.7-fold (***
P < 0.01) (Fig. 8
, A and B). In marked contrast, 12
h-pretreatment with GF109203X, an aminoalkyl bisindolylmaleimide that
potently and selectively inhibits PKC activity (31),
failed to prevent PTHrP (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34)-mediated down-regulation of BSP gene
expression (1 P < 0.001) (Fig. 8
, C and D).
Furthermore, 24 h-treatment with GF109203X alone dramatically reduced
the basal level of BSP mRNA by approximately 80% (1 P
< 0.001) (Fig. 8
, C and D).

View larger version (30K):
[in this window]
[in a new window]
|
Figure 8. Effect of the PKA/PKC pathway inhibitors on
PTHrP-mediated repression of BSP mRNA expression. Cementoblasts were
pretreated with THFA, an adenylate cyclase inhibitor, or GF109203X, a
PKC pathway inhibitor, for 12 h in before the stimulation with
PTHrP (1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 ) (10-7 M) plus the
inhibitors for an additional 12 h. A and C, Representative
Northern blot autoradiographs of BSP and 18S rRNA of the experiments
using inhibitors of PKA and PKC pathways, respectively. B and D,
Graphic analysis of cpm (mean values expressed as BSP mRNA normalized
to 18S rRNA level ± SD). PTHrP significantly
inhibited BSP mRNA gene expression (* P < 0.001
vs. vehicle). THFA, partially but significantly, rescued
PTHrP-induced down-regulation of BSP mRNA level (**
P < 0.01 for PTHrP plus THFA vs.
PTHrP only). Twenty-four-hour treatment with THFA alone significantly
stimulated BSP approximately 1.7-fold (*** P <
0.01 vs. vehicle). In contrast, GF109203X failed to
prevent PTHrP-induced BSP down-regulation. Furthermore, the inhibitor
alone significantly reduced BSP mRNA level (* P <
0.001 vs. vehicle).
|
|
 |
Discussion
|
|---|
The role of PTHrP in regulating cells associated with cartilage
and bone has been studied extensively both in vivo and
in vitro (42, 43). In contrast, the function of
PTHrP in controlling activities of cells associated with mineralized
tissues of the tooth, such as cementum, has not been investigated
in-depth due to lack of suitable model systems, such as in
vitro cell lines. Understanding PTHrPs function in cells
associated with teeth is important for broadening our knowledge of the
regulatory role of PTHrP during formation and regeneration of all
mineralized tissues. The current studies, using well-defined murine
cementoblast cell lines, explored the molecular mechanisms underlying
the inhibitory effect of PTHrP on cementoblast activities.
The time course experiment revealed that a minimum of 6 h exposure
of cells to PTHrP was required to observe an inhibitory effect on BSP
mRNA expression. This pattern indicates that inhibition of BSP mRNA
expression in cementoblasts is a secondary response requiring protein
synthesis, as confirmed by the experiments using cycloheximide.
Effects of PTH/PTHrP on gene markers associated with mineralized
tissues have been linked to transcriptional (24, 26, 44),
and/or posttranscriptional (45) events. Data from the
current studies suggest that the inhibitory effect of PTHrP on BSP mRNA
expression in cementoblasts is at the transcription level. The mRNA
decay curves revealed that when the transcription inhibitor actinomycin
D was included, BSP mRNA in the vehicle-treated group decayed with a
half-life of 6.5 h approximately, and the transcripts in
PTHrP-pretreated cells decayed with a half-life of 7.2 h. In
three different experiments, a slightly upward trend was noted for the
BSP mRNA decay curve of PTHrP-treated group, compared with that of
vehicle-treated group; however, no statistical difference in BSP mRNA
decay rate was detected between the groups, suggesting that PTHrP did
not dramatically affect the stability of BSP mRNA. Thus, the
PTHrP-mediated down-regulation of BSP mRNA must result from
transcriptional inhibition.
PTHrP stimulates both cAMP/PKA and PKC pathways (12, 46).
The data presented in the current study suggested that the inhibitory
effect of PTHrP on BSP mRNA expression in cementoblasts was mediated
through the cAMP/PKA pathway, but not through the PKC pathway. First,
the dose-dependent profile of PTHrP-mediated inhibition of BSP gene
expression and biomineralization resembles the dose-dependent effect of
PTHrP on promotion of cAMP levels in cementoblasts. Second, only the
PTHrP analog capable of stimulating a cAMP response [PTHrP (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31)]
(47) inhibited BSP gene expression and biomineralization.
In contrast, PTHrP (7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34), an antagonist for PTHrP (23, 48) failed to inhibit gene expression and biomineralization in
cementoblast cultures. Third, other activators of the cAMP/PKA pathway,
IBMX (a cAMP phosphodiesterase inhibitor) (24), forskolin
(an adenylyl cyclase stimulator) (26), and Sp-cAMPs (a
potent and specific activator for cAMP-dependent protein kinase A with
excellent cell permeability and resistance to cyclic nucleotide
phosphodiesterase) (25), mimicked the inhibitory effects
of PTHrP (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34) on BSP gene expression and biomineralization, whereas
two PKC activators, PMA and PDBU, which stimulate PKC function by
binding to PKC and increasing its affinity to calcium
(27, 28, 29), failed to do so. Finally, THFA, an adenylate
cyclase inhibitor (24), partially but significantly
blocked the effect of PTHrP on BSP gene expression, and THFA alone
stimulated basal level BSP mRNA expression. In contrast, the PKC
inhibitor, GF109203X (31), did not reverse PTHrP-mediated
down-regulation of BSP mRNA expression, but, in fact, dramatically
lowered the basal BSP mRNA level. Collectively, these results suggest
that the cAMP/PKA pathway, but not the PKC pathway, mediates the
inhibitory effect of PTHrP on BSP mRNA expression. Furthermore, it
appears that this pathway may also play an inhibitory role in
regulating the basal BSP gene expression. In support to this
conclusion, Wang et al. reported that treatment of
osteoblasts with forskolin blocked apatite deposition and decreased BSP
mRNA expression (49).
In contrast to the inhibitory effects of the
cAMP/PKA, the activation of the PKC pathway resulted in an increase of
BSP gene expression in cementoblasts. PTHrP (7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34) is widely accepted
as an effective antagonist for PTH/PTHrP functions (23, 30). Thus, the up-regulation of BSP mRNA expression might result
form the PTHrP (7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34)-mediated inhibition of endogenous PTHrP.
However, endogenous PTHrP mRNA and protein were not detected in the
cementoblasts used for these studies (data not shown). Moreover, PTH
(7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34) did not further decrease the basal cAMP level in cementoblasts
(Fig. 6C
). Therefore, this is not a likely explanation. Instead,
stimulation of the PKC pathway may directly target a PKC-responsive
element in the murine cementoblast BSP gene. This hypothesis is
supported by the finding that the avian BSP gene contains a
PKC-responsive cis-element in the promoter region
(26) and the strong homology between the murine and avian
promoter. Although this PKC-responsive element did not mediate the
up-regulation of avian BSP mRNA expression associated with short-term
exposure to PTH (26), it provides a potential mechanism by
which cells regulate BSP gene expre