Você está na página 1de 10

JOURNAL OF BONE AND MINERAL RESEARCH

Volume 21, Number 5, 2006


Published online on February 20, 2006; doi: 10.1359/JBMR.020603
2006 American Society for Bone and Mineral Research

Enhanced Expression of the Inorganic Phosphate Transporter Pit-1 Is


Involved in BMP-2Induced Matrix Mineralization in
Osteoblast-Like Cells
Atsushi Suzuki,1 Chafik Ghayor,2 Jrome Guicheux,3 David Magne,3 Sophie Quillard,3 Ayako Kakita,4 Yasunaga
Ono,1 Yoshitaka Miura,4 Yutaka Oiso,4 Mitsuyasu Itoh,1 and Joseph Caverzasio2

ABSTRACT: Pi handling by osteogenic cells is important for bone mineralization. The role of Pi transport in
BMP-2induced matrix calcification was studied. BMP-2 enhances Pit-1 Pi transporters in osteogenic cells.
Experimental analysis suggest that this response is required for bone matrix calcification.
Introduction: Bone morphogenetic proteins (BMPs) are produced by osteogenic cells and play an important
role in bone formation. Inorganic phosphate (Pi) is a fundamental constituent of hydroxyapatite, and its
transport by osteogenic cells is an important function for primary calcification of the bone matrix. In this study,
we investigated the role of Pi transport in BMP-2induced matrix mineralization.
Materials and Methods: Confluent MC3T3-E1 osteoblast-like cells were exposed to BMP-2 for various time
periods. Pi and alanine transport was determined using radiolabeled substrate, Pit-1 and Pit-2 expression by
Northern blot analysis, cell differentiation by alkaline phosphatase activity, matrix mineralization by alizarin
red staining, and the characteristics of mineral deposited in the matrix by transmission electron microscopy,
electron diffraction analysis, and Fourier transformed infrared resolution (FTIR).
Results: BMP-2 time- and dose-dependently stimulated Na-dependent Pi transport in MC3T3-E1 cells by
increasing the Vmax of the transport system. This effect was preceded by an increase in mRNA encoding Pit-1
but not Pit-2. BMP-2 also dose-dependently enhanced extracellular matrix mineralization, an effect blunted by
either phosphonoformic acid or expression of antisense Pit-1. Enhanced Pi transport and matrix mineralization induced by BMP-2 were blunted by a specific inhibitor of the c-Jun-N-terminal kinase (JNK) pathway.
Conclusions: Results presented in this study indicate that, in addition to its well-known effect on several
markers of the differentiation of osteoblastic cells, BMP-2 also stimulates Pi transport activity through a
selective increase in expression of type III Pi transporters Pit-1. In MC3T3-E1 cells, this effect is mediated by
the JNK pathway and plays an essential role in bone matrix calcification induced by BMP-2.
J Bone Miner Res 2006;21:674683. Published online on February 20, 2006; doi: 10.1359/JBMR.020603
Key words: Pit-1 transporter, BMP-2, c-Jun-N-terminal kinase, osteoblast, mineralization

INTRODUCTION

NORGANIC PHOSPHATE (Pi) transport represents a particular function of bone-forming cells for extracellular matrix
mineralization.(1,2) As in all eukaryotic cells, osteogenic
cells (i.e., osteoblasts and chondrocytes) are endowed with
Na-dependent Pi transport systems driven by an inwardly
directed Na gradient.(3) The type III Pi transport systems,
namely Pit-1 and Pit-2, are involved in Pi handling by osteogenic cells.(4,5) In situ hybridization analysis revealed expression of the type III transporter Pit-1 (Glvr-1) in a subpopulation of hypertrophic chondrocytes during
endochondral bone formation, suggesting a potential role

The authors state that they have no conflicts of interest.

for this Pi carrier in primary events of extracellular matrix


calcification.(6) Enhanced expression and a role of this Pi
transporter in pathological calcification have also been recently suggested.(7,8) Pi transport in osteogenic cells not
only provides sufficient Pi for metabolic processes but also
serves a specialized function in matrix vesicles that are microstructures playing a critical role in initial events of bone
matrix calcification.(9) Studies have shown that Pi transport
activity is low in matrix vesicles released from chondrocytes
and osteoblasts during the proliferative phase and increases
during their differentiation.(4,10) The increase in Pi transport activity in matrix vesicles released during the formation of the collagenous matrix is likely explained by an
enrichment of Pi carriers during the differentiation process.
In line with this concept, several in vitro studies have al-

1
Division of Endocrinology, Department of Internal Medicine, Fujita Health University School of Medicine, Toyoake, Japan; 2Service
of Bone Diseases, Department of Rehabilitation and Geriatrics, University Hospital of Geneva, Geneva, Switzerland; 3INSERM EM
99-03, School of Dental Surgery, Nantes, France; 4Department of Metabolic Diseases, Nagoya University Graduate School of Medicine,
Nagoya, Japan.

674

ROLE OF PI TRANSPORT IN BMP-2INDUCED MINERALIZATION


ready shown a stimulation of Pi transport by well-known
osteotropic factors such as PTH and IGF-I in either osteoblast-like cells(5,11,12) or chondrocytes,(10) as well as basic
fibroblast growth factor in osteoblastic cells.(13) Bone morphogenetic proteins (BMPs) are members of the TGF-
superfamily and exert a wide range of biological effects in
different tissues. In particular, they contribute to the formation of bone and connective tissues(14) by inducing the
differentiation of mesenchymal cells into bone-forming
cells.(15) This factor is one of the most potent stimulators of
osteoblastic cell differentiation mainly characterized by expression of alkaline phosphatase (ALP), type I collagen,
the production of osteocalcin, and bone matrix mineralization.(1618) Whether BMP-2 influences Pi transport in osteogenic cells for inducing matrix mineralization is not
known and was investigated in this study. Results obtained
indicate that BMP-2 selectively enhances Pi transport activity in MC3T3-E1 osteoblast-like cells. This effect is
mainly caused by enhanced expression of Pit-1 by a transcriptional process that probably involves the c-Jun-Nterminal kinase (JNK) pathway for its activation. Biochemical and molecular experimental evidence suggests that
enhanced expression of this Pi transport system contributes
to matrix mineralization induced by BMP-2 in osteoblastic
cells.

MATERIALS AND METHODS


Chemicals
FCS, glutamine, antibiotics, and trypsin/EDTA were obtained from Gibco (Life Technologies, Basel, Switzerland).
-MEM was purchased from Amimed (Bioconcept, Allschwill, Switzerland). Phosphonoformic acid (PFA), actinomycin D, and cycloheximide were obtained from Sigma (St
Louis, MO, USA). SB203580, SP600125, and U0126 compounds were obtained from Calbiochem-Novabiochem
(San Diego, CA, USA). Recombinant human BMP-2 was
kindly provided by Genetics Institute (Cambridge, MA,
USA). H3[32P]O4 and L-[3H]alanine were from DuPont de
Nemours (Brussel, Belgium). All other chemicals were
from standard laboratory suppliers and were of the highest
purity available.

Cell culture and transfection


Murine calvaria-derived MC3T3-E1 osteoblast-like cells
(ATCC, LGC Promochem, France) were grown in -MEM
medium containing 10% FCS, 1% (vol/vol) nonessential
amino acids, 100 IU/ml penicillin, and 100 g/ml streptomycin. Subcultures were obtained once a week by removing
the cells from the dish using 0.1% collagenase and 0.25%
trypsin in a Ca- and Mg-free Earles salt solution containing
0.02% EDTA. Cultures were maintained at 37C in a humidified atmosphere of 5% CO2-95% air and medium was
changed every 23 days. Cells were seeded into 24-well
plates (40 103 cells/well) for transport and 6-well plates
(150 103 cells/well) for matrix mineralization studies.
The Pit-1 antisense oligonucleotide (AS-Pit-1; ODN sequence: TATTATTTAAAGAACCACTACACTAGAGAATGGAATCTACTGTGGCAACGATTACTAGTACCCTAGCTGC) was synthesized by MWG-Biotech

675

GmbH (Ebersberg, Germany) and inserted into pcDNA3


vector. MC3T3-E1 cells were transfected with either the
AS-Pit-1 or empty vector (pcDNA3) using superfect reagent following manufacturers instructions (QIAGEN,
Venlo, The Netherlands). Transfected cells were selected
with G418 sulphate (Calbiochem-Novabiochem) at 0.8, 0.4,
and 0.2 mg/ml during the first, second, and third week, respectively, after plasmid transfection. Selected cell lines
were cultured in the presence of 0.02 mg/ml G418 sulphate.
G418 is an aminoglycoside similar in structure to gentamicin B1. G418 blocks polypeptide synthesis by inhibiting the
elongation step in both prokaryotic and eukaryotic cells.
Resistance to G418 is conferred by the neo gene from Tn5
encoding an aminoglycoside 3-phosphotransferase. G418 is
thus used for selection of mammalian cells expressing a neo
cassette in a vector like pcDNA3.

Pi and alanine transport analysis


BMP-2 effect on Pi and alanine transport was analyzed in
confluent MC3T3-E1 cells cultured in -MEM medium
containing 2% FCS for 24 h. In some experiments, cells
were pretreated either with inhibitors or their vehicle for 1
h before adding BMP-2 for 6 h. After treatment with
agents, Pi and alanine transport activity was determined in
Earles buffered salt solution (EBSS) containing either
0.052.5 mM labeled H3[32P]O4 or 0.1 mM [3H]alanine, respectively, as previously described.(13) Transport experiments started by rinsing three times the cell layer with
EBSS. Then, 0.3 ml of EBSS containing labeled substrate (1
Ci/ml) was added during 6 minutes for Pi and 2 minutes
for alanine transport analysis (corresponding to respective
time-points of initial rate of transport activities(11)). Finally,
the uptake solution was aspirated, and the cell layer was
rinsed three times with 0.3 ml of ice-cold EBSS. The cell
layer was solubilized with 0.25 ml of 0.2N sodium hydroxide, and the radioactivity of a 200-l aliquot was counted by
standard liquid scintillation technique. As previously documented, preliminary experiments indicated that the sodium-independent component of Pi transport determined
in presence of 143 mM choline chloride was <10% of the
total uptake of Pi(19) and that this component was not influenced by BMP-2 (data not shown). Thus, this Naindependent component was neglected in this study.

RNA extraction and Northern blotting


Total cellular RNA was extracted using Tripure Reagent
(Roche Molecular Biochemicals, Rotkreuz, Switzerland)
according to the manufacturers instructions and as previously described.(20) For Northern blotting analysis, 10 g of
total RNA was denatured and separated by electrophoresis
on a 1.2% agarose gel. RNA was transferred to GeneScreen membranes (DuPont de Nemours, Brussels, Belgium) by capillary transfer. Membranes were UV crosslinked and stained with methylene blue to control for equal
sample loading and RNA integrity. The membranes were
hybridized to a PCR-amplified fragment (12081920 bp) of
either the human Pit-1 (Glvr-1) cDNA (GenBank accession
no. L20859)(21) or Pit-2 (Ram-1) cDNA (GenBank acces-

676
sion no. L20852).(22) The differences in sample loading
were corrected by the expression of a housekeeping gene,
either cyclophilin or GAPDH. Probes were labeled with 50
Ci [-32P] dCTP by random priming. Pre-hybridization
(30 minutes) and hybridization (2 h) were performed in
Quickhyb buffer (Stratagene, La Jolla, CA, USA), which
was supplemented for the hybridization with 100 g/ml
salmon sperm DNA. Membranes were washed twice for 15
minutes in 2 SSC, 0.1% SDS at 25C and for 1030 minutes at 60C in 0.1 SSC, 0.1% SDS before exposition to
Kodak X-OMAT AR films for 1872 h at 80C.

Alizarin red staining


Confluent MC3T3-E1 cells (day 5) were treated every
23 days with BMP-2 or its vehicle in -MEM culture medium containing 10% FCS and 10 mM -glycerophosphate.
When indicated, cells were pretreated with agents for 1 h
and stimulated by 1030 ng/ml of BMP-2 or its vehicle.
After 6 weeks, mineralized nodules was determined by
alizarin red staining. Calcified area was quantified using
NIH image analysis.

Mineral deposited in the organic matrix of


MC3T3-E1 cells
Cells were prepared as described above for alizarin red
staining. After 6-week treatment with either BMP-2 or vehicle, cells were fixed in cacodylate/glutaraldehyde (4%)
buffer for 90 minutes at 4C and postfixed in cacodylate
buffer with 2% osmium tetroxide for 1 h at 4C. Samples
were dehydrated with successive dilutions in ethanol and
embedded in Epon overnight at 37C and during 2 additional days at 55C. Sections (80 nm thick) were cut with an
Ultracut E ultramicrotome (Reichert-Jung, Heidelberg,
Germany), mounted on copper grids, stained according to
Reynolds lead citrate stain method, and analyzed with a
Jeol 1010 electron microscope at 80 kV. The crystal structure was determined by selected area diffraction patterns at
100 kV and camera length of 50 cm.(23) The infrared spectra
were registered with a Magna IR 550 spectrometer (Nicolet, Trappes, France) coupled to an IR Plan Advantage
microscope (15 reflachromat lens; spectra Tech, Shelton,
CT, USA), equipped with a mercury-cadmium-telluride detector. The measurements were made on thin films (2 m
thick) in the transmission configuration at 4 cm1 resolution. Samples (with and without BMP-2 treatment) were
embedded in Epon, and sections were cut with an ultra
microtome and deposited on BaF2 window for analysis. The
spectra were acquired with a spatial aperture of 40 40 m2
and corrected for residual H2O and CO2 absorptions. Epon
peaks were also subtracted. For further detailed mineral
analysis, deconvolution of the n1n3 (PO4) was made with
the following typical parameters: k 2.3 and s 24.5
cm1.

ALP activity
ALP activity was determined as previously described.(24)
Dose-dependent effect of BMP-2 on ALP activity was determined in culture medium containing 1% FCS after 48-h
incubation. Cells were harvested in 0.2% Nonidet P-40 and

SUZUKI ET AL.
disrupted by sonication. The homogenate was centrifuged
at 1500g for 5 minutes, and ALP activity was determined in
the supernatant by the method of Lowry et al.(25)

Statistical analysis
Results are expressed as mean SE. A two-sided unpaired Students t-test or ANOVA for multiple comparison
was used for statistical analysis. A difference between experimental groups was considered to be significant when
p < 0.05.

RESULTS
Characteristics of Pi transport stimulation by
BMP-2 in MC3T3-E1 cells
The effect of recombinant human BMP-2 was studied in
MC3T3-E1 cells that responded quite well to BMP-2 for
inducing expression of ALP (Fig. 1B). In confluent cells, we
found that this BMP dose-dependently and selectively increased Pi transport (Fig. 1A). A stimulatory effect was
detected at the dose of 1 ng/ml of BMP-2 with maximal
stimulation at 30 ng/ml (Fig. 1A). Interestingly, the dose
response effect on Pi uptake stimulation was different from
that of ALP induced by BMP-2 (Figs. 1A and 1B). The
stimulation of Pi uptake induced by BMP-2 was detected
after 3 h, further enhanced after 6 h, and reached a maximal
value after 24-h incubation (Fig. 2). Alanine uptake was not
significantly affected in the same conditions (Table 1). Because alanine is transported into mammalian cells mainly
through an Na co-transport system, this observation indicates that the effect of BMP-2 on Pi uptake is mediated by
a selective cellular mechanism. Furthermore, the stimulatory effect of BMP-2 on Pi uptake seemed to be dependent
on RNA and protein synthesis, because it was completely
blocked in cells pretreated with either 2.5 g/ml actinomycin D or 5 M cycloheximide respectively (Table 2). Finally, kinetic analysis indicated that the stimulatory effect
of BMP-2 on Pi uptake is related to an increase in the
maximal rate (Vmax) of the transport system, without a
change in its apparent affinity constant (Km) for phosphate
(Table 1).

Selective stimulation of Pit-1 by BMP-2 in


MC3T3-E1 cells
Northern blotting analysis showed that BMP-2 increased
Pit-1 but not Pit-2 mRNA expression in MC3T3-E1 cells
with a time-course response similar to the time stimulation
of Pi uptake. An increase in Pit-1 mRNA was indeed detectable after 2 h and was nearly maximal after 24-h incubation (Fig. 3A). Expression of Pit-2 and of the housekeeping gene cyclophilin was not significantly affected by BMP2. Preincubation of the cells with actinomycin D, which
markedly reduced the abundance of Pit-1 mRNA in vehicle-treated cells, completely blocked the induction of
Pit-1 by BMP-2, further suggesting that this response is
dependent on transcription (Fig. 3B). Inhibition of protein
synthesis by preincubation with cycloheximide slightly reduced the stimulation of Pit-1 mRNA expression induced
by BMP-2 without, however, blocking this response, sug-

ROLE OF PI TRANSPORT IN BMP-2INDUCED MINERALIZATION

677

FIG. 2. Time-dependent effect of BMP-2 on Pi transport in


MC3T3-E1 cells. Cells cultured for 4 days were switched to
-MEM containing 2% FCS for 24 h. Then, they were incubated
with either 10 ng/ml of BMP-2 () or its vehicle () for various
time periods. Pi uptake was determined in EBSS containing 0.1
mM 32PO4. Each value represents mean SE of five to six determinations from a representative experiment. *p < 0.01, compared
with vehicle.

TABLE 1. KINETIC ANALYSIS OF BMP-2-INDUCED PI TRANSPORT


STIMULATION AND ANALYSIS OF THE SELECTIVITY OF THIS
EFFECT BY MEASURING ALANINE TRANSPORT
Pi transport
Kinetic
analysis

Vmax
(nmol/6 minutes/well)

Km (mM)

Veh
BMP-2
Selectivity

0.398 0.02
0.705 0.03*

0.491 0.04
0.508 0.04

Veh
BMP-2

FIG. 1. Dose-dependent effect of BMP-2 on Pi transport and


ALP activity in MC3T3-E1 osteoblast-like cells (A) For Pi transport (Pi TRANSP) analysis, cells were cultured for 4 days and
switched to -MEM containing 2% FCS for 24 h. Then, they were
incubated for 6 h with various concentrations of BMP-2 or its
vehicle. Pi uptake was determined in EBSS containing 0.1 mM
32
PO4. (B) For ALP analysis, cells were cultured for 5 days and
switched to culture medium containing 2% FCS for 24 h. Then,
they were exposed to various concentrations of BMP-2 or its vehicle for 24 h. ALP activity and the amount of proteins were
determined. Each value represents mean SE of four to six determinations from a representative experiment. *p < 0.05 compared with vehicle.

Alanine transport
(pmol/2 minutes/well)
91.1 6.3
76.7 5.3

Confluent MC3T3-E1 cells were stimulated by either BMP-2 (10 ng/ml)


or its vehicle for 6 h. Then Pi and alanine uptake were determined. Values
are the mean SE of five to six determinations.
* p < 0.001 compared with vehicle.

gesting that this effect of BMP-2 is, at least in part, independent of new protein synthesis (Fig. 3B).

Expression of Pit-1 and matrix mineralization


induced by BMP-2
Matrix mineralization of MC3T3-E1 cells was determined by alizarin red staining and quantitative computer

678

SUZUKI ET AL.

TABLE 2. EFFECT OF ACTINOMYCIN D OR CYCLOHEXIMIDE


BMP-2INDUCED PI TRANSPORT

ON

Pi transport (pmol/6 minutes/well)


Vehicle
BMP-2
Actinomycin D
Actinomycin D + BMP-2
Vehicle
BMP-2
Cycloheximide
Cycloheximide + BMP-2

89.0 5.6
148.1 7.6*
47.9 3.9
58.5 4.8
108.2 6.3
167.6 6.1*
57.7 5.2
59.0 2.7

Confluent MC3T3-E1 cells were pretreated with either 2.5 g/ml of actinomycin D, 5 M of cycloheximide, or their vehicle for 3 h and exposed
to BMP-2 (10 ng/ml) or its vehicle for 6 h. Then, Pi uptake was determined.
Values are the mean SE of six determinations.
* p < 0.001 compared with vehicle.

FIG. 3. Effect of BMP-2 on mRNA expression of type III Na-Pi


transporters Pit-1 and Pit-2 in MC3T3-E1 cells. (A) Cells were
prepared as described in Fig 2 for Pi transport analysis. Then, they
were exposed to BMP-2 (10 ng/ml) or its vehicle for various time
periods before analysis of Pit-1, Pit-2, and cyclophilin (CYCLO)
mRNA expression by Northern blotting analysis. (B) In another
set of experiments, cells were pretreated with either 5 M of
cycloheximide (CYCLOHEX) or 2.5 g/ml actinomycin D
(ACTINO) as well as their vehicles for 3 h before exposure to
BMP-2 (10 ng/ml) or its vehicle for 8 h and determination of Pit-1
and cyclophilin mRNA expression by Northern blot analysis.

image analysis after 6-week culture in the presence of 10


mM -glycerophosphate. Under this experimental condition, BMP-2 induced a dose-dependent increase in matrix

mineralization. Enhanced accumulation of mineral was detected at the concentration of 3 ng/ml with a gradual increase and near maximal effect between 30 and 100 ng/ml
BMP-2 (Fig. 4A), a dose relationship very close to that of
changes in Pi uptake induced by BMP-2 (Fig. 1A) and quite
different from changes in ALP activity (Fig. 1B). The calcium-phosphate crystals formed in the extracellular organic
matrix of MC3T3-E1 cells cultured for 6 weeks with 30
ng/ml BMP-2 were examined by transmission electron microscopy (TEM), electron diffraction (EM), and resolution
enhanced Fourier transformed infrared (FTIR) spectroscopy. As depicted in Fig. 4B, TEM analysis revealed that
thin and relatively long crystals are deposited in association
with collagen fibrils, an observation that is consistent with
physiological mineralization.(26) EM analysis indicated that
the pattern of ring diffraction, with a D-spacing of the first
and second ring of 0.35 and 0.28 nm, respectively (Fig. 4C),
corresponds to an apatite crystal. Resolution enhanced
FTIR analysis showed a typical IR spectrum in the region
of 8001800 cm1, similar to spectra obtained from mouse
trabecular bone (data not shown). Increased matrix mineralization induced by BMP-2 was completely blocked by 0.1
mM phosphonoformic acid (Fig. 5A), a competitive inhibitor of Na-Pi cotransport.(27) This effect was observed in
presence of a slightly reduced BMP-2induced stimulation
of ALP in cells exposed to PFA compared with vehicletreated cells (Fig. 5B), suggesting that part of the blunting
effect of BMP-2 on matrix mineralization observed in PFAtreated cells was probably caused by other factors than
merely an inhibition of Pi transport. Another mechanism by
which this analog of inorganic phosphate might have reduced matrix mineralization is through its reduction of the
rate of matrix vesicleinduced mineral formation, which
plays an important role in initiating the process of mineralization.(28) Thus, to validate above data suggesting an important role of Pit-1 in matrix mineralization induced by
BMP-2, we constructed MC3T3-E1 cells expressing an antisense Pit-1 oligonucleotide (AS-Pit-1). Of several clones
stably transfected with AS-Pit-1, two expressed lower levels
of Pit-1 mRNA compared with other Pit-1 (Fig. 6, clones 2
and 3) and empty vector transfected cells (data not shown).
Clone 2 was used for further analysis because, in addition to
expressing a lower baseline and a blunted response to
BMP-2 on Pi uptake (Fig. 6B and similar effect obtained
with clone 3, data not shown), this clone had a near normal
stimulation of ALP by BMP-2 (data not shown). The matrix
mineralization induced by BMP-2 in this clone was markedly decreased compared with cells stably transfected with
the empty vector (Fig. 6C), strongly suggesting that Pit-1
plays a critical role in bone matrix mineralization induced
by BMP-2.

Role of MAP kinase pathways in the stimulation of


Pi transport and matrix calcification induced
by BMP-2
Previous studies have shown that both Smad and mitogen-activated protein kinase (MAPK) pathways are essen-

ROLE OF PI TRANSPORT IN BMP-2INDUCED MINERALIZATION

679

FIG. 4. Dose-dependent effect of BMP-2


on MC3T3-E1 cell matrix calcification. (A)
Cells cultured for 5 days were treated with
various concentrations of BMP-2 or its vehicle every other day for 6 weeks in -MEM
medium containing 10% FCS and 10 mM
-glycerophosphate. Matrix mineralization
was determined by alizarin red staining and
by NIH image analysis. (B and C) Cells exposed to 30 ng/ml BMP-2 for 6 weeks as described in A were fixed in cacodylate/
glutaraldehyde buffer and embedded in
Epon, and 80-nm-thick sections and crystals
were analyzed by transmission electron microscopy and electron diffraction. (B) Transmission electron microscopy: crystal localization (7500) associated with collagen fibrils
(#) is shown by arrows. *Cellular compartment. (C) Electron diffraction analysis: arrows indicate various diffraction rings with
their corresponding D-spacing values.

tial components of the TGF- superfamily signaling machinery.(24,2932) Inhibition of the p38 and ERK pathways
by 10 M of SB203580 and U0126, respectively, had no
effects on Pi transport stimulation induced by BMP-2 (data
not shown). In contrast, the JNK inhibitor, SP600125 (SP),
dose-dependently reduced this effect without influencing
basal Pi transport activity (Fig. 7A). This effect was associated with a marked decrease in BMP-2induced matrix
mineralization by 10 M SP600125 and complete inhibition
at 20 M (Fig. 7B). As previously observed,(30) SP600125
significantly enhanced basal (1.5 0.04-fold after 11 days,
p < 0.01) and BMP-2stimulated ALP activity (data not
shown). Enhanced basal ALP activity was associated with
increased baseline matrix mineralization (data not shown),
indicating that the blunting effect of SP600125 on matrix
mineralization induced by BMP-2 was not caused by alteration in ALP or a cytotoxic effect.

DISCUSSION
Results presented in this study indicate that BMP-2 induces a selective, dose- and time-dependent increase in Pi
transport in MC3T3-E1 osteoblasts and that this effect is
involved in the mineralization of the bone matrix induced
by this BMP. Pi transport stimulation by a member of the

TGF- family has already been previously reported for


TGF- in chondrocytes,(20) indicating that members of this
family regulate Pi transport in several type of osteogenic
cells. Pi transport stimulation by BMP-2 is dependent on
RNA and protein synthesis and reflects a change in the
Vmax of the transport system. These characteristics are similar to those previously reported for the regulation of Pi
uptake by several hormones or growth factors in different
types of bone-forming cells and suggest that the observed
increase in Pi uptake may require the synthesis of new Pi
carriers and their insertion into the plasma membrane.(10,11,13,20) Consistently, Northern blotting analysis
showed a time-dependent increase in the expression of
transcripts encoding the type III Pi transporter Pit-1 in response to BMP-2 in MC3T3-E1 cells. The induction of Pit-1
preceded expression of Pi transport stimulation and this
effect is dependent on gene transcription. Interestingly, although the related type III Pi transporter Pit-2 was also
present in MC3T3-E1 cells, its expression level was not
affected by BMP-2. A similar observation has been reported in ATDC5 chondrocytic cells with TGF-,(20) suggesting distinct cellular mechanisms for the regulation of
these two transport systems in osteogenic cells previously
characterized as housekeeping Pi transport systems.(33)
Recent in vitro and in vivo studies have suggested an
important role of Pi transport for the mineralization of the

680

SUZUKI ET AL.

FIG. 5. Effect of PFA on BMP-2induced MC3T3-E1 matrix


calcification and ALP activity. Cells cultured for 5 days were
treated with phosphonoformic acid (PFA, 0.1 mM), an inhibitor
of Pi transport, for 1 h before exposure to BMP-2 (30 ng/ml) or its
vehicle every other day for 6 weeks for analysis of matrix mineralization (A) and 24 h for ALP activity (B). Matrix mineralization
was determined by alizarin red staining and by NIH image analysis and ALP activity by the method of Lowry et al.(25) Each value
represents mean SE of five to six determinations from a representative experiment. *p < 0.01, compared with vehicle.

FIG. 6. Effect of a Pit-1 antisense oligonucleotide on Pit-1 expression, Pi transport, and matrix calcification. (A) MC3T3-E1
cells were stably transfected with either the empty vector
pcDNA3 or a plasmid expressing a Pit-1 antisense mRNA (ASPit-1). Pit-1 mRNA expression was determined by Northern blotting in various clones. (B and C) Confluent pcDNA3 or AS-Pit1
(clone 2) transfected cells were switched to culture medium containing 2% FCS for 24 h. Then they were incubated with either
BMP-2 (30 ng/ml) or its vehicle during 6 h for Pi transport (Pi
TRANSP) analysis (B) and 6 weeks for matrix calcification
(CALCIFIED AREA) (C). Pi transport and matrix calcification
were determined as described in Figs. 1 and 4. Each value represents mean SE of four to five determinations from a representative experiment. *p < 0.01 compared with vehicle.

bone matrix. Essentially, it has been documented that Pi


transport in osteogenic cells not only provide sufficient Pi
for metabolic processes but also plays a specialized function
in matrix vesicles, microstructures that likely serve as nucleation sites for mineralization of the extracellular matrix of
certain tissues.(9,34) This hypothesis, mainly supported by in

vitro studies, was reinforced by the in vivo observation of


Pit-1 expression in a subset of hypertrophic chondrocytes
during endochondral bone formation.(6) In this study, enhanced Pi uptake by BMP-2 was associated with a corresponding increase in physiological matrix mineralization.
Interestingly, the dose effect of BMP-2 on Pi uptake corre-

ROLE OF PI TRANSPORT IN BMP-2INDUCED MINERALIZATION

FIG. 7. Effect of the JNK inhibitor, SP600125, on BMP-2


induced changes in Pi transport and matrix calcification. MC3T3E1 cells were prepared for either (A) Pi transport (Pi TRANS) or
(B) matrix calcification (CALCIFIED AREA) analysis as described in Figs. 1 and 4. Cells were preincubated for 1 h with or
without various doses of the JNK inhibitor SP600125. Then, they
were exposed to 10 ng/ml BMP-2 in presence of the inhibitor or its
vehicle for 6 h or 4 weeks (every 23 days) before the determination of Pi transport and matrix calcification, respectively. Pi
transport and matrix mineralization were determined as described
in Figs. 1 and 4, respectively. Each value represents mean SE of
four to five determinations from a representative experiment.
*p < 0.01, compared with vehicle.

lated quite well with changes in matrix mineralization but


poorly with alteration in ALP expression, suggesting that
the rate-limiting step of this process is probably not the
amount of extracellular Pi available in the extracellular
compartment but rather the capacity of the cells to transport Pi, caused by an increased number of Pit-1 carriers

681

inserted in the plasma membrane. Because matrix vesicles


are derived from the plasma membrane of osteogenic cells,
it is likely that matrix vesicles derived from BMP-2treated
cells also have a higher capacity to accumulate Pi and mineralize, which could partly explain the enhanced matrix
mineralization observed in MC3T3-E1 cells treated with
BMP-2. Increased matrix mineralization by BMP-2 was
completely blocked by PFA, a competitive inhibitor of NaPi cotransport,(27) suggesting a role of Pi transport in this
process. However, whether this blunting effect of PFA on
matrix mineralization was caused by alteration in Pi transport remained unclear because this analog of inorganic Pi
has been recently shown to interact with the rate of mineral
formation in matrix vesicles.(28) However, analysis of matrix mineralization in MC3T3-E1 cells expressing a Pit-1
antisense oligonucleotide strongly suggests that Pi transport
plays a critical role in matrix mineralization induced by
BMP-2. Indeed, in these cells having a lower expression of
Pit-1 but a nearly normal response of ALP induced by
BMP-2, enhanced matrix mineralization by this BMP was
markedly impaired. Data from this experiment also indicate
that Pi transport is not the only factor involved in the calcification of the organic matrix (Fig. 6). Indeed, in cells
expressing the Pit-1 antisense oligonucleotide and having a
lower capacity to transport Pi, matrix mineralization was
not lower but slightly increased compared with cells expressing the empty vector. This apparent discrepancy is best
explained by enrichment of cells having a higher capacity to
mineralize during clonal selection. Among factors likely to
influence this process, one could mention the amount of
matrix vesicles released or their capacity to transport calcium or other molecules involved in accumulation of mineral inside the vesicles as well as many other parameters
involved in the control of the deposition, maturation, and
calcification of the bone matrix. Obviously, the formation
of a collagenous matrix and its mineralization is a complex
process that is not yet very well understood.
In MC3T3-E1 cells, we recently documented that, in addition to activation of the Smad-signaling pathway, BMP-2
also activates the p38 and JNK pathways with a selective
role of each pathway in the differentiation of these cells.(30)
In this study, we found that JNK is probably involved in
mediating the stimulation of Pi transport by BMP-2 in osteoblastic cells. Whereas the ERK and p38 inhibitors had
no effect on Pi transport stimulation induced by BMP-2, the
JNK inhibitor, SP600125, was quite effective in blunting this
response in MC3T3-E1 cells. A dose of 20 M that selectively inhibits activation of JNK by BMP-2 in these cells(30)
completely blocked this effect without influencing basal Pi
transport activity. Associated with this blunting effect on Pi
transport, inhibition of JNK completely prevented BMP-2
induced matrix mineralization. Interestingly, this effect was
observed despite preservation of the increase in ALP activity induced by BMP-2 in presence of this inhibitor, suggesting distinct roles of Pi transport and ALP in initial
events of primary calcification of the bone matrix. To our
knowledge, this is the first observation for an implication of
this pathway in the regulation of a Pi transport system in
mammalian cells. Further studies are required to determine
downstream regulatory molecules activated by this pathway

682

SUZUKI ET AL.

that mediate the stimulation of Pi transport by BMP-2 in


osteoblastic cells.
In conclusion, data presented in this study indicate that
BMP-2 stimulates both Pi uptake and Pit-1 expression in
MC3T3-E1 cells. This effect is probably mediated by activation of the JNK pathway. They also provide strong evidence that this effect is involved in bone matrix mineralization induced by BMP-2, further supporting the concept that
Pi handling by osteogenic cells is an important parameter
for the primary calcification of the bone matrix.

14.

15.

16.

17.

ACKNOWLEDGMENTS
18.

The authors thank P Apostolides for expert technical


assistance. We also acknowledge the contribution of Paul
Pilet, Thierry Bouillon, and Pierre Weiss for the biomineralization analysis by transmission electron microscopy and
electron diffraction. This study was supported by the Swiss
National Science Foundation (3100AO-100607) and a
Grant-in-Aid for Scientific Research from the Ministry of
Education, Culture, Sports, Science and Technology of Japan (15590987).

REFERENCES
1. Caverzasio J, Bonjour JP 1996 Characteristics and regulation
of Pi transport in osteogenic cells for bone metabolism. Kidney
Int 49:975980.
2. Caverzasio J, Montessuit C, Bonjour JP 1996 Functional role of
Pi transport in osteogenic cells. News Physiol Sci 11:119125.
3. Caverzasio J, Selz T, Bonjour JP 1988 Characteristics of phosphate transport in osteoblastlike cells. Calcif Tissue Int 43:8387.
4. Montessuit C, Caverzasio J, Bonjour JP 1991 Characterization
of a Pi transport system in cartilage matrix vesicles. Potential
role in the calcification process. J Biol Chem 266:1779117797.
5. Palmer G, Bonjour JP, Caverzasio J 1997 Expression of a
newly identified phosphate transporter/retrovirus receptor in
human SaOS-2 osteoblast-like cells and its regulation by insulin-like growth factor I. Endocrinology 138:52025209.
6. Palmer G, Zhao J, Bonjour J, Hofstetter W, Caverzasio J 1999
In vivo expression of transcripts encoding the Glvr-1 phosphate transporter/retrovirus receptor during bone development. Bone 24:17.
7. Giachelli CM, Speer MY, Li X, Rajachar RM, Yang H 2005
Regulation of vascular calcification: Roles of phosphate and
osteopontin. Circ Res 96:717722.
8. Cecil DL, Rose DM, Terkeltaub R, Liu-Bryan R 2005 Role of
interleukin-8 in PiT-1 expression and CXCR1-mediated inorganic phosphate uptake in chondrocytes. Arthritis Rheum
52:144154.
9. Caverzasio J, Murer H, Tenenhouse HS 2003 Phosphate Homeostasis, Regulatory Mechanisms. Pediatric Bone. Biology &
Diseases. Elsevier Science, San Diego, CA, USA.
10. Montessuit C, Bonjour JP, Caverzasio J 1994 Pi transport regulation by chicken growth plate chondrocytes. Am J Physiol
267:E24E31.
11. Selz T, Caverzasio J, Bonjour JP 1989 Regulation of Nadependent Pi transport by parathyroid hormone in osteoblastlike cells. Am J Physiol 256:E93100.
12. Zoidis E, Ghirlanda-Keller C, Gosteli-Peter M, Zapf J, Schmid
C 2004 Regulation of phosphate (Pi) transport and NaPi-III
transporter (Pit-1) mRNA in rat osteoblasts. J Endocrinol
181:531540.
13. Suzuki A, Palmer G, Bonjour JP, Caverzasio J 2000 Stimulation of sodium-dependent phosphate transport and signaling

19.

20.

21.

22.

23.
24.

25.

26.

27.

28.

29.

30.

31.

mechanisms induced by basic fibroblast growth factor in


MC3T3-E1 osteoblast-like cells. J Bone Miner Res 15:95102.
Schmitt JM, Hwang K, Winn SR, Hollinger JO 1999 Bone
morphogenetic proteins: An update on basic biology and clinical relevance. J Orthoped Res 17:269278.
Yamaguchi A 1995 Regulation of differentiation pathway of
skeletal mesenchymal cells in cell lines by transforming growth
factor-beta superfamily. Semin Cell Biol 6:165173.
Hughes FJ, Collyer J, Stanfield M, Goodman SA 1995 The
effects of bone morphogenetic protein-2, -4, and -6 on differentiation of rat osteoblast cells in vitro. Endocrinology
136:26712677.
Hay E, Hott M, Graulet AM, Lomri A, Marie PJ 1999 Effects
of bone morphogenetic protein-2 on human neonatal calvaria
cell differentiation. J Cell Biochem 72:8193.
Cheng SL, Lou J, Wright NM, Lai CF, Avioli LV, Riew KD
2001 In vitro and in vivo induction of bone formation using a
recombinant adenoviral vector carrying the human BMP-2
gene. Calcif Tissue Int 68:8794.
Zhen X, Bonjour JP, Caverzasio J 1997 Platelet-derived
growth factor stimulates sodium-dependent Pi transport in osteoblastic cells via phospholipase Cgamma and phosphatidylinositol 3 -kinase. J Bone Miner Res 12:3644.
Palmer G, Guicheux J, Bonjour JP, Caverzasio J 2000 Transforming growth factor-beta stimulates inorganic phosphate
transport and expression of the type III phosphate transporter
Glvr-1 in chondrogenic ATDC5 cells. Endocrinology
141:22362243.
OHara B, Johann SV, Klinger HP, Blair DG, Rubinson H,
Dunn KJ, Sass P, Vitek SM, Robins T 1990 Characterization of
a human gene conferring sensitivity to infection by gibbon ape
leukemia virus. Cell Growth Differ 1:119127.
Miller DG, Edwards RH, Miller AD 1994 Cloning of the cellular receptor for amphotropic murine retroviruses reveals homology to that for gibbon ape leukemia virus. Proc Natl Acad
Sci USA 91:7882.
Misell DL, Brown EB 1987 Electron Diffraction: An Introduction for Biologists. Elsevier, New York, NY, USA.
Suzuki A, Guicheux J, Palmer G, Miura Y, Oiso Y, Bonjour
JP, Caverzasio J 2002 Evidence for a role of p38 MAP kinase
in expression of alkaline phosphatase during osteoblastic cell
differentiation. Bone 30:9198.
Lowry OH, Roberts NR, Wu ML, Hixon WS, Crawford EJ
1954 The quantitative histochemistry of brain. II. Enzyme measurements. J Biol Chem 207:1937.
Rey C, Kim HM, Gerstenfeld L, Glimcher MJ 1995 Structural
and chemical characteristics and maturation of the calciumphosphate crystals formed during the calcification of the organic matrix synthesized by chicken osteoblasts in cell culture.
J Bone Miner Res 10:15771588.
Loghman-Adham M, Szczepanska-Konkel M, Yusufi AN, Van
Scoy M, Dousa TP 1987 Inhibition of Na+-Pi cotransporter in
small gut brush border by phosphonocarboxylic acids. Am J
Physiol 252:G244G249.
Wu LN, Sauer GR, Genge BR, Valhmu WB, Wuthier RE 2003
Effects of analogues of inorganic phosphate and sodium ion on
mineralization of matrix vesicles isolated from growth plate
cartilage of normal rapidly growing chickens. J Inorg Biochem
94:221235.
Gallea S, Lallemand F, Atfi A, Rawadi G, Ramez V, SpinellaJaegle S, Kawai S, Faucheu C, Huet L, Baron R, RomanRoman S 2001 Activation of mitogen-activated protein kinase
cascades is involved in regulation of bone morphogenetic protein-2-induced osteoblast differentiation in pluripotent C2C12
cells. Bone 28:491498.
Guicheux J, Lemonnier J, Ghayor C, Suzuki A, Palmer G,
Caverzasio J 2003 Activation of p38 mitogen-activated protein
kinase and c-Jun-NH2-terminal kinase by BMP-2 and their
implication in the stimulation of osteoblastic cell differentiation. J Bone Miner Res 18:20602068.
Lemonnier J, Ghayor C, Guicheux J, Caverzasio J 2004 Protein
kinase C-independent activation of protein kinase D is in-

ROLE OF PI TRANSPORT IN BMP-2INDUCED MINERALIZATION


volved in BMP-2-induced activation of stress mitogenactivated protein kinases JNK and p38 and osteoblastic cell
differentiation. J Biol Chem 279:259264.
32. Hu Y, Chan E, Wang SX, Li B 2003 Activation of p38 mitogenactivated protein kinase is required for osteoblast differentiation. Endocrinology 144:20682074.
33. Werner A, Dehmelt L, Nalbant P 1998 Na+-dependent phosphate cotransporters: The NaPi protein families. J Exp Biol
201:31353142.
34. Anderson HC, Garimella R, Tague SE 2005 The role of matrix
vesicles in growth plate development and biomineralization.
Front Biosci 10:822837.

683
Address reprint requests to:
Joseph Caverzasio, PhD
Service of Bone Diseases
Department of Rehabilitation and Geriatrics
University Hospital of Geneva
CH-1211 Geneva 14, Switzerland
E-mail: Joseph.Caverzasio@medecine.unige.ch

Received in original form December 6, 2005; revised form January


30, 2006; accepted February 13, 2006.