Full text of DOI:10.1359/jbmr.2001.16.2.221

JOURNAL OF BONE AND MINERAL RESEARCH
Volume 16, Number 2, 2001
©
2001 American Society for Bone and Mineral Research
Transcriptional Induction of Matrix Metalloproteinase-13
(
Collagenase-3) by 1␣,25-Dihydroxyvitamin D3
in Mouse Osteoblastic MC3T3-E1 Cells
1
2
3
1
1
MOTOYUKI UCHIDA, MASAAKI SHIMA, DAICHI CHIKAZU, AYAKO FUJIEDA, KAZUMI OBARA,
1
1
1
3
HIROYUKI SUZUKI, YUMIKO NAGAI, HIDEYUKI YAMATO, and HIROSHI KAWAGUCHI
ABSTRACT
The removal of unmineralized matrix from the bone surface is essential for the initiation of osteoclastic bone
resorption because osteoclasts cannot attach to the unmineralized osteoid. Matrix metalloproteinases (MMPs)
are known to digest bone matrix. We recently reported that among the MMPs expressed in mouse osteoblastic
cells, MMP-13 (collagenase-3) was the one most predominantly up-regulated by bone resorbing factors
including 1␣,25-dihydroxyvitamin D [1␣,25(OH) D ]. In this study, we examined the mechanism of regula-
3
2 3
tion of MMP-13 expression by 1␣,25(OH) D in mouse osteoblastic MC3T3-E1 cells. 1␣,25(OH) D increased
2
3
2 3
steady-state messenger RNA (mRNA) and protein levels of MMP-13. De novo protein synthesis was essential
for the induction because cycloheximide (CHX) decreased the effect of 1␣,25(OH) D on the MMP-13 mRNA
2
3
level. 1␣,25(OH) D did not alter the decay of MMP-13 mRNA in transcriptionally arrested MC3T3-E1 cells;
2
3
however, it increased the MMP-13 heterogeneous nuclear RNA (hnRNA) level and MMP-13 transcriptional
rate. The binding activity of nuclear extracts to the AP-1 binding site, but not to the Cbfa1 binding site, in the
MMP-13 promoter region was up-regulated by 1␣,25(OH) D , suggesting the mediation of AP-1 in this
2
3
transcriptional induction. To determine the contribution of MMPs to bone resorption by 1␣,25(OH) D , the
2
3
inhibitory effect of BB94, an MMP inhibitor, on resorbed pit formation by mouse crude osteoclastic cells was
examined on either an uncoated or collagen-coated dentine slice. BB94 did not prevent resorbed pit formation
on uncoated dentine whereas it did on collagen-coated dentine. We therefore propose that the transcriptional
induction of MMP-13 in osteoblastic cells may contribute to the degradation of unmineralized matrix on the
bone surface as an early step of bone resorption by 1␣,25(OH) D . (J Bone Miner Res 2001;16:221–230)
2
3
Key words: matrix metalloproteinase, collagenase, 1␣,25-dihydroxyvitamin D , osteoblast, bone
3
(
4)
INTRODUCTION
bone resorbing agents and is known to affect osteoclasts
indirectly through osteoblastic cells in which its receptor
(5)
␣
,25-DIHYDROXYVITAMIN D3 [1␣,25(OH) D ], the bio- (vitamin D receptor [VDR]) exists.
This action of
2
3
1
logically active metabolite of vitamin D , not only plays 1␣,25(OH) D was reported recently to be mediated by the
3
2 3
an essential role in calcium homeostasis in bone, kidney, induction of the osteoclast differentiation factor (ODF/
(
1,2)
and intestine,
but also regulates cell growth and differ- receptor activator of nuclear factor ␬␤ ligand [RANKL]) in
(
3)
(6)
entiation. 1␣,25(OH) D is also one of the most potent osteoblastic cells. However, because osteoclasts are not
2
3
1
2
3
Biomedical Research Laboratories, Kureha Chemical Industry Co., Ltd., Shinjuku, Tokyo, Japan.
Department of Pediatrics, Faculty of Medicine, Osaka University, Suita, Osaka, Japan.
Department of Orthopedic Surgery, Graduate School of Medicine, University of Tokyo, Bunkyo, Tokyo, Japan.
2
21

2
22
UCHIDA ET AL.
capable of attaching to the surface of unmineralized bone, sulfate were purchased from Sigma (St. Louis, MO, USA).
digestion of unmineralized bone by proteinases is necessary Polydeoxyinosine-deoxycytidilic acid [poly(dI-dC)] and
(7,8)
for osteoclastic bone resorption to proceed.
Thus, Complete were purchased from Boehringer Mannheim
1
␣,25(OH) D may possibly degrade matrix proteins by (Mannheim, Germany)
2
3
regulating proteinases in osteoblastic cells as one of its bone
resorbing activities. In fact, previous studies indicated that
Cell culture
1
␣,25(OH) D induced collagenase production in osteo-
2 3
(9–11)
blastic cells.
The MC3T3-E1 clonal cell line was kindly provided by
(19)
Matrix metalloproteinases (MMPs) are a family of struc- Dr. Kumegawa of Meikai University (Saitama, Japan).
turally and functionally related enzymes responsible for the MC3T3-E1 cells were plated at a density of 20,000 cells/
2
proteolytic degradation of extracellular matrix components cm in 100-mm dishes and cultured in ␣-MEM containing
such as collagens, elastin, glycoproteins, proteoglycans, and 10% FBS. After 3 days of culture, cells were precultured in
(12,13)
glucosaminoglycans.
To date, at least 16 distinct mem- serum-free medium for 24 h and exposed to experimental
bers of the MMP family have been identified and classified medium.
(14,15)
into subgroups based on substrate specificity.
Among
For the study on MMP-13 stability, MC3T3-E1 cells
them, MMP-1 (interstitial collagenase), MMP-8 (neutrophil were precultured in serum-free medium with or without
Ϫ8
collagenase), and MMP-13 (collagenase-3) are known to 1␣,25(OH) D (10 M) for 16 h and exposed to the ex-
2
3
Ϫ4
have collagenase activity and to be able to cleave the native perimental medium containing DRB (10 M) with or with-
helix of type I collagen, the major protein of bone ma- out 1␣,25(OH) D by completely changing the medium. At
2
3
(16,17)
trix.
Recently, we have reported that MMP-2, -9, -11, 0, 4, 8, 16, and 24 h after the medium change, total RNA
-
13, and -14 were expressed both in mouse osteoblastic from control or 1␣,25(OH) D -treated cells was prepared as
2
3
MC3T3-E1 cells and in mouse primary osteoblastic cells, described in the following section.
making MMP-13 the only collagenase expressed in mouse
(18)
osteoblastic cells.
Furthermore, among the MMPs ex-
Northern blot analysis
pressed in osteoblastic cells, MMP-13 was the one most
predominantly up-regulated by systemic bone resorbing fac-
DNA probes for MMP-13, -2, and glyceraldehyde-3-
tors such as parathyroid hormone, prostaglandin E , and phosphate dehydrogenase (GAPDH) were generated by RT-
2
1
␣,25(OH) D and was the major MMP whose steady-state polymerase chain reaction (PCR) with the total RNA from
2 3
messenger RNA (mRNA) level was increased by MC3T3-E1 cells. For MMP-13 complementary DNA
␣,25(OH) D . Hence, the present study was undertaken to (cDNA), the PCR primers were 5Ј-CATTCAGCTATCC-
1
2
3
determine the mechanism whereby 1␣,25(OH) D regulates TGGCCACCTTC-3Ј and 5Ј-CATCCACATGGTTGGGA-
2
3
MMP-13 expression in mouse osteoblastic MC3T3-E1 AGTTCTG-3Ј, yielding a 1016-base pair (bp) fragment. For
cells.
MMP-2 cDNA, the PCR primers were 5Ј-AAGGATG-
GACTCCTGGCACATGCCTTT-3Ј and 5Ј-ACCTGTGG-
GCTTGTCACGTGGTGT-3Ј, yielding a 963-bp fragment.
For GAPDH cDNA, the PCR primers were 5Ј-TGAAG-
GTCGGTGTGAACGGATTTGGC-3Ј and 5Ј-CATGTAG-
GCCATGAGGTCCACCAC-3Ј, yielding a 983-bp frag-
MATERIALS AND METHODS
Materials
1
␣,25(OH) D , Denhardt’s solution, dithiothreitol (DTT), ment. Each PCR product was subcloned in a TA-cloning
2 3
DNAse I, and bacterial collagenase were purchased from vector pCRII vector (Invitrogen, Carlsbad, CA, USA), as
Wako Pure Chemical (Osaka, Japan). 24R,25-dihydroxy- confirmed by sequencing. The cDNA fragments were radio-
vitamin D [24R,25(OH) D ] was synthesized at Kureha labeled with the Rediprime DNA labeling system and
3
2 3
3
2
Chemical Industry Co., Ltd. (Tokyo, Japan). BB94 was the [␣- P]dCTP.
generous gift of Dr. M. Nakajima (Novaritis Pharma K. K.,
Total RNA was isolated using ISOGEN following the
Hyogo, Japan). ␣-Modification of Eagle’s minimal essential manufacturer’s instructions (Wako Pure Chemical). Sam-
medium (␣-MEM), fetal bovine serum (FBS), and Moloney ples of 20 ␮g of total RNA were separated by electrophore-
murine leukemia virus (MMLV) reverse transcriptase (RT) sis in 2.2 M formaldehyde–1.0% agarose gels and trans-
ϩ
were purchased from Life Technologies, Inc. (Grand Island, ferred to a Hybond N nylon membrane. Hybridization was
3
2
NY, USA). Rediprime labeling kit, [␣- P]deoxycytidine performed in 4ϫ SSCP (0.48 M NaCl, 60 mM Na citrate,
3
triphosphate (dCTP, specific activity of 3000 Ci/mmol), 60 mM Na HPO , and 20 mM NaH PO ), 10% dextran
2
4
2
4
3
2
[
3
␣- P]uridine 5Ј-triphosphate (UTP, specific activity of sulfate, 1ϫ Denhardt’s solution, 1% sodium dodecyl sulfate
000 Ci/mmol), [␥- P]deoxyadenosine 5Ј-triphosphate (SDS), and 100 ␮g/ml sheared salmon sperm DNA at 65°C
3
2
(
dATP, specific activity of 3000 Ci/mmol), Sephadex G-25 overnight. The filter was washed in 1ϫ SSC (0.15 M NaCl
ϩ
column, Hybond N Nylon membrane, and Hyperfilm were and 15 mM Na citrate, pH 7.0)–0.1% SDS four times for
3
purchased from Amersham Pharmacia Biotech (Tokyo, Ja- 15 minutes at 65°C and then once for 15 minutes in 0.1ϫ
pan). Recombinant Taq polymerase, T4 polynucleotide ki- SSC–0.1% SDS at 65°C. Blottings were analyzed by auto-
nase, deoxynucleotide triphosphate mixture solution (dNTP, radiography using Hyperfilm and the bands were quanti-
2
.5 mM), and SeaKem GTG agarose were obtained from tated using a Molecular Imager (Bio-Rad, Hercules, CA,
Takara Shuzo (Shiga, Japan). Cycloheximide (CHX), 5,6- USA). Quantitative mRNA levels of MMPs were expressed
dichlorobenzimidazole riboside (DRB), NP-40, and dextran by the density of each blotting normalized to GAPDH. Each

INDUCTION OF MMP-13 BY 1␣,25(OH) D IN OSTEOBLASTS
223
2
3
Northern blot analysis experiment was performed using two on a 5–20% polyacrylamide gel containing 8 M urea, 100 mM
or three independent cultures.
Tris-borate, and 1 mM EDTA and then visualized by autora-
diography. The amplification protocol yielded products that
were within the linear range for both the MMP-13 hnRNA and
the GAPDH mRNA.
Western blot analysis
Culture supernatants were collected from MC3T3-E1 cells
cultured for 48 h in the presence and absence of
Nuclear extract preparation
1
␣,25(OH) D . The supernatants were heat-denatured in boil-
2 3
ing water and subjected to 5–20% SDS-polyacrylamide gel
electrophoresis (PAGE) under reducing conditions. Proteins treated with or without 1␣,25(OH) D (10 M) for 2 h
Nuclear extracts were prepared from MC3T3-E1 cells
Ϫ8
2
3
were transferred onto Immobilon P membranes (Millipore, using a modified version of the protocol described by Dig-
(21)
7
Bedford, MA, USA). Western blot analysis for detection of nam et al. In brief, MC3T3-E1 cells (about 2 ϫ 10 cells)
MMP-13 was performed using specific anti-mouse MMP-13 were harvested in 5 ml of ice-cold phosphate-buffered sa-
(20)
primary antibody
and secondary anti-rabbit alkaline line by centrifugation at 150g for 5 minutes. The pellets
phosphatase–conjugated immunoglobulin G (IgG). The bound were resuspended into 500 ␮l of Dignam buffer A contain-
antibodies were detected with Western blotting detection re- ing 0.2% NP-40, 1 mM DTT, and 1ϫ Complete (Boehr-
agents using as the alkaline phosphatase substrate CSPD inger Mannheim). The lysate was centrifuged at 2200g for
(Clontech, Palo Alto, CA, USA). Western blot analysis was 1 minute to pellet the NP-40–insoluble material, and the
performed using two independent cultures.
supernatant was removed. The pellet was resuspended in 50
l of Dignam buffer C, and incubated on ice for 10 minutes
␮
with occasional vortexing to disrupt the nuclear membranes.
Extracts were centrifuged for 1 minute at 20,000g, the
Nuclear run-on assay
MC3T3-E1 cells were treated with 1␣,25(OH) D for supernatants were removed, and the pellets were discarded.
2
3
1
8
6 h, placed in a hypotonic lysis buffer (10 mM HEPES, pH The protein content of the nuclear extracts was determined
.0, and 1.5 mM MgCl ),l and kept on ice for 15 minutes. using the Bio-Rad DC assay kit (Bio-Rad).
2
The cell membrane was disrupted by homogenization in a
Dounce homogenizer and released nuclei were harvested by
centrifugation. Nuclei were resuspended with a buffer com-
prised of 40% glycerol, 50 mM Tris-HCl, pH 8.3, 5 mM
Electrophoretic mobility shift assay
Mobility shift assay was performed using a 22-oligomer
MgCl , and 0.1 mM EDTA. Run-on transcription was per- ATAAGTGATGACTCACCATTGC and a 27-oligomer
2
formed by mixing 100 ␮l of nuclei with 100 ␮l of 2ϫ GATTCTGCCACAAACCACACTTAGGAA containing the
reaction buffer (10 mM Tris-HCl; pH 8.0; 5 mM MgCl ; 0.3 putative AP-1 and Cbfa1 binding sites, respectively, in the
2
M KCl; 1 mM DTT; 0.2 mM EDTA; 4 mM ATP, GTP, and 5Ј-upstream region of the mouse MMP-13 gene. Oligonu-
3
2
32
CTP; and 100 ␮Ci [␣- P]UTP; 3000 Ci/mmol) and incu- cleotides were annealed, end-labeled with [␥- P]dATP us-
bated at 30°C for 30 minutes. Radiolabeled RNA from the ing T4 polynucleotide kinase, and further purified by Seph-
reaction mixture was extracted with ISOGEN. Approxi- adex G-25 column chromatography. Five micrograms of
mately 500,000 cpm was hybridized with equal amounts nuclear extracts was incubated with binding buffer contain-
of heat-denatured and slot-blotted plasmids containing ing 20 mM Tris-HCl (pH 8.0), 10 mM NaCl, 3 mM EDTA,
MMP-13 and GAPDH cDNAs. The pCRII vector served as 0.05% NP-40, 5 mM DTT, 5% glycerol, and 1 ␮g of
a negative control.
poly(dI-dC) for 10 minutes. Labeled probe (40 fmol) was
then added and incubated for an additional 20 minutes. The
samples were subjected to electrophoresis at room temper-
ature on 5% polyacrylamide gel in 89 mM Tris, 89 mM
RT-PCR
MMP-13 heterogenous nuclear RNA (hnRNA) was ana- boric acid, and 2 mM EDTA buffer at 80 V for 1 h. In
lyzed by RT-PCR using sense primer 5Ј-GTGTTCTG- competitive experiments, competitor oligos (2 pmol) were
CTGCATATACAGCCAC-3Ј (hnMMP13S), which spans incubated with nuclear extracts during the first 10 minutes
intron 2 of the mouse MMP-13 gene, and antisense primer before addition of the labeled probe. After electrophoresis
5Ј-CCTTCTCCACTTCAGAATGGGAC-3Ј (hnMMP13A), the gel was dried and exposed for autoradiography.
which spans exon 3, yielding a 280-bp product. GAPDH
mRNA was analyzed by RT-PCR using sense primer 5Ј-
GTCTTCACCACCATGGAGAAG-3Ј (position 341–361 of
Resorbed pit formation assay
mouse GAPDH) and antisense primer 5Ј-CATGTAGGC-
CATGAGGTCCACCAC-3Ј (position 1010–1033 of mouse cocultured as described previously.
Mouse osteoblastic cells and bone marrow cells were
(22)
Briefly, primary os-
GAPDH). RNA samples, extracted as described for Northern teoblastic cells were prepared from calvariae of neonatal
analysis, were treated with DNAse I. RNA (1 ␮g) was copied ddY mice (Shizuoka Laboratories Animal Center, Shizuoka,
6
into cDNA using the antisense primer and MMLV RT. The Japan). Osteoblastic cells (1 ϫ 10 cells/well) and bone
7
cDNA was amplified by PCR with 25 cycles of 94°C for 30 s, marrow cells (2 ϫ 10 cells/well) prepared from 8-week-old
60°C for 30 s, and 72°C for 1 minute in the presence of ddY mice were cocultured on a 0.24% collagen gel (Nitta
recombinant Taq polymerase, 0.2 ␮M sense and antisense Gelatin, Tokyo, Japan) coated on 100-mm tissue culture
32
Ϫ8
primers, and 5 ␮Ci [␣- P]dCTP. PCR products were resolved dishes with ␣-MEM containing 10% FBS and 10
M

2
24
␣,25(OH) D . The medium was changed every 2 days.
UCHIDA ET AL.
1
2
3
After culture for 6 days, each dish was treated with 4 ml of
.2% collagenase solution for 20 minutes at 37°C. The cells
0
released from the dish were collected by centrifugation
twice at 1000 rpm for 5 minutes and suspended in 5 ml of
␣
-MEM containing 10% FBS. Dentine slices (diameter, 6
mm; thickness, 0.15 mm) were distributed in a 96-well
plate. A collagen solution prepared as described previously
was diluted to 1.5 mg/ml with ␣-MEM and used to coat
dentine slices (10 ␮l/slice). The collagen was allowed to
(23)
polymerize by incubating it for 30 minutes at 37°C.
An
aliquot of the crude osteoclast preparation (0.1 ml) was then
seeded onto either uncoated or collagen-coated dentine
slices and cultured with ␣-MEM containing 10% FBS in the
Ϫ8
Ϫ5
presence and absence of 10 M 1␣,25(OH) D , 10
M
2
3
Ϫ8
BB94, or 10 M salmon calcitonin (CT; Peninsula Labo-
ratories, Inc., Belmont, CA, USA). After 48 h of culture,
cells on dentine slices were removed in 1N NH OH solution
4
and stained with 0.5% toluidine blue for 1 minute. The total
area of pits on the dentine slice was estimated under a light
microscope with a micrometer using an image analyzer
(
System Supply Co., Nagano, Japan) and expressed as a
FIG. 1. (A) Time course and (B) dose response of steady-state mRNA
levels of MMP-13 and MMP-2 stimulated by 1␣,25(OH) D in
percentage of the whole area of dentine.
2
3
Ϫ8
MC3T3-E1 cells. (A) MC3T3-E1 cells were treated with 10
M
Statistical analyses
1␣,25(OH) D for the indicated periods. (B) MC3T3-E1 cells were treated
2 3
with 1␣,25(OH) D or 24R,25(OH) D at the indicated concentrations for
2
3
2 3
Means of groups were compared by analysis of variance
16 h. Northern blot analysis was performed using 20 ␮g of total RNA to
(ANOVA) and significance of differences was determined determine MMP-13 and MMP-2 mRNA levels with GAPDH as a control.
by post hoc testing using Bonferroni’s method.
The size of the hybridizing signals was 2.9 kb and 3.1 kb with MMP-13
and MMP-2 probe, respectively. Representative autoradiograms from
duplicate experiments that yielded similar results are shown.
RESULTS
Effect of 1␣,25(OH) D on steady-state mRNA and
protein levels of MMP-13
was investigated by Northern blot analysis in the presence
and absence of CHX (10 ␮M; Fig. 3). CHX inhibited not
only basal MMP-13 mRNA expression, but also
2
3
Figures 1A and 1B show the time course and dose re-
1
␣,25(OH) D -induced MMP-13 mRNA expression to a
2 3
sponse of the effects of 1␣,25(OH) D , respectively, on the
2
3
level similar to that of the control. These inhibitions by
CHX were greater than those seen for MMP-2 mRNA.
Thus, de novo protein synthesis was assumed to be requisite
for the induction of MMP-13 by 1␣,25(OH) D .
steady-state MMP-13 mRNA level in cultured MC3T3-E1
cells as determined by Northern blot analysis.
1
␣,25(OH) D caused a time-dependent increase in this
2 3
2
3
level: 6.0-fold at 16 h and approximately 10-fold at 24 and
Ϫ11
4
1
8 h (Fig. 1A). At 16 h of culture, 1␣,25(OH) D (10
–
2
3
Ϫ8
Ϫ8
Ϫ6
Effect of 1␣,25(OH) D on MMP-13 mRNA stability
2 3
0
M) and 24R,25(OH) D (10 –10 M), a less active
2 3
vitamin D3 metabolite, stimulated the MMP-13 mRNA
level dose dependently with maximal effects of 16-fold at
The effect of 1␣,25(OH) D on the stability of MMP-13
2
3
mRNA in MC3T3-E1 cells was determined using DRB
Ϫ8
Ϫ6
Ϫ4
1
0
M and 15-fold at 10 M, respectively (Fig. 1B).
(
10 M), an RNA polymerase II inhibitor, to arrest tran-
Neither 1␣,25(OH) D nor 24R,25(OH) D regulated the
2
3
2
3
scription, and the decay of MMP-13 mRNA was monitored
by Northern blot analysis. In the presence and absence of
MMP-2 mRNA level.
To determine whether the MMP-13 mRNA stimulated by
␣,25(OH) D was translated into protein, Western blot
1
␣,25(OH) D , MC3T3-E1 cells were cultured with DRB
2 3
1
2
3
for up to 24 h after 16 h of preculture without DRB (Fig. 4).
No difference of MMP-13 mRNA degradation was ob-
served between control and 1␣,25(OH) D -treated cultures,
analysis was conducted at 48 h of culture (Fig. 2). Immu-
noreactive MMP-13 protein could not be detected in the
2
3
control culture medium; however, it was induced by
indicating that 1␣,25(OH) D does not affect MMP-13
2
3
Ϫ8
1
␣,25(OH) D (10 M).
2 3
mRNA stability.
Involvement of de novo protein synthesis in MMP-13
Effect of 1␣,25(OH) D on transcriptional activation
2
3
induction by 1␣,25(OH) D
of MMP-13
2
3
The involvement of de novo protein synthesis in
␣,25(OH) D stimulation of the MMP-13 mRNA level MMP-13 by 1␣,25(OH) D using RT-PCR for hnRNA (Fig.
We further investigated the transcriptional activation of
1
2
3
2 3

INDUCTION OF MMP-13 BY 1␣,25(OH) D IN OSTEOBLASTS
225
2
3
FIG. 2. Immunoreactive MMP-13 protein level in the presence and
absence of 1␣,25(OH) D in the culture media of MC3T3-E1 cells. The
2
3
FIG. 4. Effect of 1␣,25(OH) D on MMP-13 mRNA stability in
2
3
media in which MC3T3-E1 cells were cultured with or without
MC3T3-E1 cell culture. After the preculture in serum-free medium
Ϫ8
1
␣,25(OH) D (10 M) for 48 h were collected. Western blot analysis
Ϫ8
2
3
with or without 1␣,25(OH) D (10 M) for 16 h, cells were treated
2
3
was performed using a polyclonal antibody originally raised against
mouse MMP-13. A representative picture from duplicate experiments
that yielded similar results is shown.
Ϫ4
with DRB (10 M, open circles for each time point and the straight
line for regression) or DRB ϩ 1␣,25(OH) D (closed circles for each
2
3
time point and the dotted line for regression) by completely changing
the medium. Cells were harvested at 0, 4, 8, 16, and 24 h after
treatment, and MMP-13 mRNA was analyzed by Northern blotting.
Data are expressed as means (symbols) Ϯ SEMs (error bars) of three
independent experiments. The values in each experiment were calcu-
lated as the percentage of the intensity of each band normalized to that
of GAPDH measured by densitometry as compared with that of time 0
(100%). The autoradiogram in this figure is representative of three
experiments that yielded similar results.
for MMP-2. Thus, it was concluded that 1␣,25(OH) D
2
3
up-regulated MMP-13 expression at the transcriptional
level, but not the posttranscriptional level.
Effect of 1␣,25(OH) D on the binding activity of
2
3
nuclear extracts to AP-1 and Cbfa1 binding
sites of the MMP-13 promoter region
FIG. 3. Effect of CHX on mRNA levels of MMP-13 and MMP-2 in
the presence and absence of 1␣,25(OH) D in MC3T3-E1 cell culture.
2
3
After 1␣,25(OH) D binds to the VDR, the ligand-
2
3
Ϫ8
MC3T3-E1 cells were cultured with or without 1␣,25(OH) D (10
2
3
receptor complex is known to form a heterodimer with a
retinoid X receptor and to interact directly with a specific
Ϫ4
M) for 16 h in the presence or absence of CHX (10 M). CHX was
added 1 h before and during the treatment with 1␣,25(OH) D . Total
2
3
DNA element, the vitamin
D
responsive element
RNA (20 ␮g/lane) was analyzed by Northern blotting. A representative
picture from three independent experiments that yielded similar results
is shown.
(24)
(VDRE).
However, the VDRE-like sequence has not
been identified in the mouse MMP-13 promoter region.
Instead, binding domains of several transcription factors
including AP-1 and Cbfa1 have been detected in this
(25–27)
region.
We therefore examined the binding activity of
5
A). 1␣,25(OH) D induced MMP-13 hnRNA expression nuclear extracts from MC3T3-E1 cells treated with or with-
2
3
after 8 h of culture. This time course of the effect of out 1␣,25(OH) D to the AP-1 and Cbfa1 binding sites in
2
3
1
␣,25(OH) D on the hnRNA level corresponded well with the mouse MMP-13 promoter region (Fig. 6). Binding to the
2 3
that on mRNA levels shown in Fig. 1A. To confirm the AP-1 site was already detectable in nuclear extracts from
transcriptional induction of MMP-13 by 1␣,25(OH) D , a untreated cells and was enhanced on stimulation by
2
3
nuclear run-on assay was performed at 16 h of culture (Fig. 1␣,25(OH) D . The specificity of proteins binding to the
2
3
5
B). 1␣,25(OH) D potently stimulated the transcription AP-1 site was confirmed by the competition with unlabeled
2 3
rate for MMP-13, and the effect was much stronger than that homologous AP-1 site probes. Although specific binding to

2
26
UCHIDA ET AL.
FIG. 5. Transcriptional activation of MMP-13 by 1␣,25(OH) D in
2
3
MC3T3-E1 cells. (A) RT-PCR for hnRNA. Total RNA extracted from
Ϫ8
MC3T3-E1 cells treated with 1␣,25(OH) D (10 M) for the indicated
2
3
periods was reverse-transcribed, and signals specific for GAPDH
mRNA and MMP-13 hnRNA were amplified by PCR in the presence
3
2
of 5 ␮Ci of [␣- P]dCTP. PCR products were visualized by autora-
diography. (B) Transcriptional rate of MMP-13. Nuclei isolated from
subconfluent MC3T3-E1 cultures treated with or without
Ϫ8
1
␣,25(OH) D (10 M) for 16 h were analyzed by nuclear run-on
2 3
FIG. 6. Effect of 1␣,25(OH) D on the binding activity of nuclear
2
3
assay for MMP-13, MMP-2, and GAPDH gene transcription. Tran-
scripts were labeled by an in vitro elongation reaction in the presence
of [␣- P]UTP. Labeled RNAs (5 ϫ 10 cpm) were hybridized to
cDNAs immobilized on nylon filters. These cDNAs were those for
MMP-13, MMP-2, and GAPDH cloned into the pCRII vector and the
pCRII vector plasmid itself.
extracts from cultured MC3T3-E1 cells to AP-1 and Cbfa1 binding
sites in MMP-13 promoter region. Electrophoretic mobility shift assay
was performed using a 22-oligomer ATAAGTGATGACTCACCAT-
TGC and a 27-oligomer GATTCTGCCACAAACCACACTTAGGAA
containing the putative AP-1 and Cbfa1 binding sites in the 5Ј-
upstream region of the mouse MMP-13 gene. Nuclear extracts from
3
2
5
Ϫ8
MC3T3-E1 cells treated with or without 1␣,25(OH) D (10 M, 2 h)
2
3
were incubated with these labeled probes. To show the specificity of
these bindings, 50-fold molar excesses of unlabeled oligonucleotide
probes were used as competitors.
the Cbfa1 binding element also was detectable in untreated
cells, it was not altered by 1␣,25(OH) D . These results
2
3
imply that 1␣,25(OH) D might possibly regulate the tran-
2
3
scriptional activation of MMP-13 through enhancement of
DNA binding activity of AP-1 subunits.
tures whereas CT inhibited it in both cultures. This result
indicates that the stimulation of osteoclast bone resorptive
function on mineralized matrix by 1␣,25(OH) D is not
2
3
Functional relevance of MMPs to bone resorptive
mediated by MMPs. However, when cultured on collagen-
coated dentine, BB94, as well as CT, inhibited resorbed pit
formation induced by 1␣,25(OH) D . Hence, 1␣,25(OH) D
appears to stimulate bone resorption at least in part through the
degradation of unmineralized matrix on bone surface by
MMPs and thus initiate osteoclastic bone resorption.
effect of 1␣,25(OH) D
2
3
2
3
2 3
The contribution of MMPs to bone resorption by
1
␣,25(OH) D was determined by examining the inhibitory
2
3
effects of BB94, an MMP inhibitor, using cultures of crude
osteoclastic cells formed by the coculture of osteoblastic
and bone marrow cells (Fig. 7). Resorbed pit formation by
crude osteoclastic cells was compared between cultures on
uncoated and collagen-coated dentine slices. When cultured
on uncoated slices, BB94 did not decrease resorbed pit
DISCUSSION
This study showed that 1␣,25(OH) D time and dose
2
3
formation either in control or in 1␣,25(OH) D -treated cul- dependently stimulated MMP-13 production by increasing
2
3

INDUCTION OF MMP-13 BY 1␣,25(OH) D IN OSTEOBLASTS
227
2
3
Three MMPs, MMP-1, MMP-8, and MMP-13, currently
(16,17)
are known to be able to digest type I collagen.
The
expression of MMP-8, which has been cloned recently,
could not be detected by RT-PCR in MC3T3-E1 cells or
mouse primary osteoblastic cells (M. Uchida, unpublished
observation, 1998). Because the mouse MMP-1 gene has
not been identified yet, we examined human MMP-1 and
-13 expressions in cultured human primary osteoblasts by
RT-PCR and found that only MMP-13 mRNA could be
detected (M. Uchida, unpublished observation, 1998). Jo-
hansson et al. also reported that MMP-13, but not MMP-1,
was expressed in periosteal cells and osteoblasts as well as
hypertrophic chondrocytes during human fetal bone devel-
(28)
opment. Furthermore, in mice, MMP-13 has been shown
to be expressed in osteoblasts localized along the newly
(29)
(30)
formed bone trabeculae
and adjacent to osteoclasts.
These results suggest that among these three collagenases
MMP-13 is the main MMP expressed in osteoblasts.
1
␣,25(OH) D is known to exert its pleiotropic effects by
2 3
binding to its specific receptor (VDR), a member of the
(31)
steroid hormone receptor superfamily.
To confirm the
involvement of VDR in the stimulation of MMP-13 expres-
sion, we examined the effect of 24R,25(OH) D on the
2
3
steady-state mRNA level of MMP-13 because we previ-
ously reported that this vitamin D analogue also bound to
(32,33)
VDR (Fig. 1B).
24R,25(OH) D is known to exhibit an
2 3
approximately 1000-fold lower affinity for VDR and a
1
00-fold higher affinity for vitamin D binding protein in
(34)
serum as compared with 1␣,25(OH) D .
We also re-
2
3
ported that 24R,25(OH) D induced an activation of human
2
3
osteocalcin promoter through a VDR-VDRE–dependent
mechanism in medium containing 0.1% FBS, but did not
(32)
induce it in medium containing 5% FBS.
In the present
study using serum-free medium, 24R,25(OH) D stimulated
the steady-state mRNA expression of MMP-13 at 1000-fold
2
3
FIG. 7. Effect of BB94, an MMP inhibitor, on resorbed pit formation
by crude osteoclastic cells on uncoated and collagen-coated dentine
slices. Osteoblastic cells from neonatal mouse calvariae and bone
marrow cells from 8-week-old mice were cocultured on a collagen gel
for 6 days to form osteoclastic cells. Crude osteoclastic cells were then
released from the dish and further cultured either on (A) uncoated or
higher concentrations of 1␣,25(OH) D . Thus, the effects of
2
3
both vitamin D metabolites on the MMP-13 expression in
3
osteoblastic cells may be mediated by a VDR-dependent
mechanism.
(
1
B) collagen-coated dentine slices for 48
h
with or without
A number of VDREs have been identified in the promoter
Ϫ8
Ϫ5
Ϫ8
␣,25(OH) D (10 M), BB94 (10 M), and CT (10 M). The total region of several genes: osteocalcin, osteopontin, 25-
2
3
(35)
area of pits on the dentine slice was estimated under a light microscope hydroxyvitamin D3 24-hydroxylase, and ␤3-integrin.
and expressed as a percentage of the whole area of dentine. No resorbed
pit was seen without 1␣,25(OH) D on collagen-coated dentine (B).
Data are expressed as means (bars) Ϯ SEMs (error bars) for 8 cultures/
group. *p Ͻ 0.01, significant inhibition versus cultures without inhib-
itors.
These VDREs consist of a direct repeat structure of two
hexanucleotides with a spacer of three nucleotides: AG-
GTCA NNN AGGTCA. Grumbles et al. reported that
2
3
1
␣,25(OH) D increased 2.1-kilobase (kb) rat MMP-13 pro-
2 3
moter activity in rat chondrocytes; however, they did not
identify the VDRE-like sequence in this promoter re-
(36)
gion.
We also could not find a VDRE-like sequence in
the hnRNA level and transcriptional rate without changing mouse MMP-13 promoter region approximately 2.2 kb up-
mRNA stability in MC3T3-E1 cells. Resorbed pit formation stream from the transcriptional starting site. Our functional
assay disclosed an essential role of MMPs in the bone study using this 2.2-kb mouse promoter construct ligated to
resorptive effect of 1␣,25(OH) D to degrade unmineral- the luciferase reporter gene failed to show the stimulation of
2
3
ized matrix on bone surface for the initiation of osteoclastic the reporter activity of transfected MC3T3-E1 cells by
bone resorption. Because our previous study revealed that 1␣,25(OH) D (data not shown). Because the increase in the
2
3
MMP-13 is the MMP whose expression was most predom- MMP-13 mRNA level induced by 1␣,25(OH) D was
2
3
inantly up-regulated by 1␣,25(OH) D in osteoblastic shown to be dependent on de novo protein synthesis,
2
3
(18)
cells,
MMP-13 in osteoblastic cells may play an important role in transcription by inducing the synthesis of one or more
the initial step of bone resorption by 1␣,25(OH) D . transcription factors. This might explain why the increase in
it is proposed that this transcriptional induction of 1␣,25(OH) D may act indirectly on mouse MMP-13 gene
2 3
2
3

2
28
UCHIDA ET AL.
the hnRNA as well as steady-state mRNA level were seen at that 1␣,25(OH) D degrades unmineralized matrix through
2
3
later time points (after 8 h of culture). It is speculated that the induction of MMPs in cells of osteoclastic lineage
the expression of the transcription factor(s) was sufficient cannot be denied. The establishment of specific inhibitors
for the activation of the endogenous MMP-13 promoter but may further clarify the contribution of each MMP to bone
might be insufficient for that of the exogenous promoter- resorption.
reporter construct in MC3T3-E1 cells.
1␣,25(OH) D is known to exhibit bone resorptive activ-
2 3
1
␣,25(OH) D increased the binding activity of nuclear ity indirectly through the induction of ODF/RANKL in
2
3
(6)
extract to AP-1 binding site but not to Cbfa1 binding site in stromal/osteoblastic cells. The stimulation of resorbed pit
MC3T3-E1 cells. In fact, 1␣,25(OH) D has been reported formation on uncoated dentine by 1␣,25(OH) D shown in
2
3
2 3
to increase mRNA levels of the AP-1 family: c-fos, c-jun, Fig. 7A is assumed to be mediated by this action. In addition
(37)
and jun-B in MC3T3-E1 cells.
Among them, c-fos has to this well-known function, we hereby propose that the
been suggested to play an important role in cell growth, transcriptional induction of MMP-13 by 1␣,25(OH) D in
2
3
differentiation, transformation, and regulation of specific osteoblastic cells may contribute to the degradation of un-
(38–40)
gene expression in osteoblastic cells.
The c-fos over- mineralized matrix on bone surface as a step before oste-
expressing transgenic mice that exhibit impaired develop- oclastic bone resorption.
(40–42)
ment of long bones and developed osteosarcoma
have
been shown to be associated with the increase in MMP-13
(
43)
expression in bone.
that up-regulates the promoter activity of MMP-13.
However, it is reported that 1␣,25(OH) D down-regulates
Cbfa1 is another transcription factor
ACKNOWLEDGMENTS
(26)
2
3
This work was supported by a Grant-in-Aid for Scientific
(44)
cbfa1 mRNA level in mouse primary osteoblasts.
This Research from the Japanese Ministry of Education, Science,
result and our finding in this study indicate that it is unlikely Sports and Culture (12470303 and 12877221; H.K.) and the
that 1␣,25(OH) D stimulates MMP-13 expression through Uehara Memorial Foundation (H.K.)
2
3
Cbfa1.
Grumbles et al. reported that 1␣,25(OH) D up-regulated
2
3
the MMP-13 mRNA level in primary chondrocytes derived
(36)
REFERENCES
from rachitic rats. In their study 1␣,25(OH) D increased
2
3
the transcriptional rate of MMP-13, and this stimulation was
mediated by de novo protein synthesis, as we observed in
this study. However, the concentration of 1␣,25(OH) D
1. Reichel H, Koeffler HP, Norman AW 1989 The role of the
vitamin D endocrine system in health and disease. N Engl
J Med 320:980–991.
2
3
2
. Walter MR 1992 Newly identified actions of the vitamin D
endocrine system. Endocr Rev 13:719–764.
affecting MMP-13 was much higher in rat chondrocytes as
Ϫ8
compared with MC3T3-E1 cells in our study (10 M vs.
3
. Tanaka H, Abe E, Miyaura C, Suda T 1982 1␣,25-
Dihydroxycholecalciferol and a human myeloid leukaemia cell
line (HL-60). Biochem J 204:713–719.
Ϫ11
1
0
M). Because vitamin D supplementation is known to
3
(45)
restore cartilage growth function in rachitis,
the pharma-
cologic action of 1␣,25(OH) D at high concentrations may
4. Reynolds JJ, Holick MF, Deluca HF 1974 The effects of
vitamin D analogues on bone resorption. Calcif Tissue Res
2
3
stimulate the resorption of the growth plate cartilage by
inducing MMP-13. Although the physiological and patho-
logical roles of 1␣,25(OH) D on bone metabolism have not
1
5:333–339.
5
. Partridge NC, Frampton RJ, Eisman JA, Michelangeli VP,
Elms E, Bradley TR, Martin TJ 1980 Receptors for 1,25(OH)2-
vitamin D3 enriched in cloned osteoblast-like rat osteogenic
sarcoma cells. FEBS Lett 115:139–142.
2
3
been elucidated yet, this study infers a possible role for
endogenous 1␣,25(OH) D on matrix degradation during
2
3
bone remodeling. In addition, Grumbles et al. also indicated
the involvement of a protein kinase C pathway in
6. Suda T, Takahashi N, Udagawa N, Jimi E, Gillespie MT,
Martin TJ 1999 Modulation of osteoclast differentiation and
function by the new members of the tumor necrosis factor
receptor and ligand families. Endocr Rev 20:345–357.
(36)
1
␣,25(OH) D action on MMP-13.
Because the media-
2
3
tion of this pathway also has been shown in 1␣,25(OH) D
2 3
(46)
7
. Kahn AJ, Partridge NC 1987 New concepts in bone remodel-
ing: An expanding role for the osteoblast. Am J Otolaryngol
stimulation on 25-hydroxyvitamin D 24-hydroxylase,
3
such a nongenomic pathway might possibly contribute to
MMP-13 induction in osteoblastic cells, although this study
did not investigate that aspect.
8
:258–264.
8. Holliday LS, Welgus HG, Fliszar CJ, Veith GM, Jeffrey JJ,
Gluck SL 1997 Initiation of osteoclast bone resorption by
interstitial collagenase. J Biol Chem 272:22053–22058.
Although BB94, a nonspecific MMP inhibitor, negatively
affected the bone resorptive activity of 1␣,25(OH) D on
9
. Shen V, Kohler G, Jeffrey JJ, Peck WA 1988 Bone-resorbing
agents promote and interferon-␥ inhibits bone collagenase
production. J Bone Miner Res 3:657–666.
2
3
collagen-coated dentine, the origin of MMPs mediating this
activity cannot be ascribed only to osteoblastic cells in this
study. It is believed that osteoblastic cells are the main
source for MMPs in bone; however, MMPs such as MMP-9,
1
1
0. Thomson BM, Atkinson SJ, Reynolds JJ, Meikel MC 1987
Degradation of type I collagen films by mouse osteoblasts is
stimulated by 1,25-dihydroxyvitamin D and inhibited by hu-
3
man recombinant TIMP. Biochem Biophys Res Commun 148:
596–602.
-
12, and -14 have been reported to be expressed in oste-
(47–52)
oclastic cells as well.
MMP-9, in fact, is known to be
1. Thomson BM, Atkinson SJ, McGarrity AM, Hembry RM, Reyn-
olds JJ, Meikel MC 1989 Type I collagen degradation by mouse
calvarial osteoblasts stimulated with 1,25-dihydroxyvitamin D3:
Evidence for a plasminogen-plasmin-metalloproteinase activation
cascade. Biochim Biophys Acta 1014:125–132.
produced by osteoclastic cells in higher amounts than by
(47,48)
osteoblastic cells.
Although our previous report
showed that MMP-13 is the major MMP that is up-regulated
(18)
by 1␣,25(OH) D in osteoblastic cells,
the possibility
2
3

INDUCTION OF MMP-13 BY 1␣,25(OH) D IN OSTEOBLASTS
229
2
3
1
1
1
2. Woessner JF Jr 1991 Matrix metalloproteinases and their in-
mouse is restricted to osteoblast-like cells and hypertrophic
chondrocytes. Cell Growth Differ 6:759–767.
hibitors in connective tissue remodeling. FASEB J 5:2145–
2
154.
30. Holliday LS, Welgus HG, Fliszar CJ, Veith GM, Jeffrey JJ,
Gluck SL 1997 Initiation of osteoclast bone resorption by
interstitial collagenase. J Biol Chem 272:22053–22058.
31. Fuller PJ 1991 The receptor superfamily: Mechanism of di-
versity. FASEB J 5:3092–3099.
3. Massova I, Kotra LP, Fridman R, Mobashery S 1998 Matrix
metalloproteinases: Structures, evolution, and diversification.
FASEB J 12:1075–1095.
4. Nagase H, Suzuki K, Itoh Y, Kan CC, Gehring MR, Huang W,
Brew K 1996 Involvement of tissue inhibitors of metallopro- 32. Uchida M, Ozono K, Pike JW 1994 Activation of the human
teinases (TIMPs) during matrix metalloproteinase activation.
Adv Exp Med Biol 389:23–31.
5. Kahari VM, Saarialho-Kere U 1999 Matrix metalloproteinases
osteocalcin gene by 24R,25-dihydroxyvitamin D3 occurs
through the vitamin D receptor and vitamin D-responsive
element. J Bone Miner Res 9:1981–1987.
1
1
1
and their inhibitors in tumour growth and invasion. Ann Med 33. Uchida M, Ozono K, Pike JW 1997 In vitro binding of vitamin
3
1:34–45.
6. Kn a´ uper V, L o¨ pez-Otin C, Smith B, Knight G, Murphy G
996 Biochemical characterization of human collagenase-3.
D receptor occupied by 24R,25-dihydroxyvitamin D to vita-
min D responsive element of human osteocalcin gene. J Ste-
roid Biochem 60:181–187.
3
1
J Biol Chem 271:1544–1550.
7. Krane SM, Byrne MH, Lemaitre V, Henriet P, Jeffrey JJ,
Witter JP, Liu X, Hong W, Jaenish R, Eeckhout Y 1996
34. Lawson DEM 1978 Binding protein in tissues for cholecalcif-
erol and metabolites. In: Lawson DEM (ed.) Vitamin D. Ac-
ademic Press, London, UK, pp. 172–186.
Different collagenase gene products have different roles in 35. Ohyama Y, Ozono K, Uchida M, Shink T, Kato S, Suda T,
degradation of type I collagen. J Biol Chem 271:28509–
8515.
Yamamoto O, Noshiro M, Kato Y 1994 Identification of a
vitamin D-responsive element in the 5Ј-flanking region of the
2
1
8. Uchida M, Shima M, Shimoaka T, Fujieda A, Obara K, Suzuki
H, Nagai Y, Ikeda T, Yamato H, Kawaguchi H 2000 Regula-
rat 25-hydroxyvitamin D 24-hydroxylase gene. J Biol Chem
269:10545–10550.
3
tion of matrix metalloproteinases (MMPs) and tissue inhibitors 36. Grumbles RM, Shao L, Jeffrey JJ, Howell DS 1996 Regulation
of metalloproteinases (TIMPs) by bone resorptive factors in
osteoblastic cells. J Cell Physiol 185:207–214.
of rat interstitial collagenase gene expression in growth carti-
lage and chondrocytes by vitamin D , interleukin-1␤, and
3
1
2
2
2
2
9. Sudo H, Kodama H, Amagai Y, Yamato S, Kasai S 1983 In
okadaic acid. J Cell Biochem 63:395–409.
vitro differentiation and calcification in a new clonal osteo- 37. Candeliere GA, Prud’homme J, St-Arnaud R 1991 Differential
genic cell line derived from newborn mouse calvaria. J Cell
Biol 97:191–198.
stimulation of Fos and Jun family members by calcitriol in
osteoblastic cells. Mol Endocrinol 5:1780–1788.
0. Yamagiwa H, Tokunaga K, Hayami T, Hatano H, Uchida M, 38. Shmidt J, Livne E, Erfle V, Gossner W, Silbermann M 1986
Endo N, Takahashi HE 1999 Expression of metalloproteinase-13
collagenase-3) is induced during fracture healing in mice. Bone
5:197–203.
Morphology and in vivo growth characteristics of an atypical
murine proliferative osseous lesion induced in vitro. Cancer
Res 46:3090–3098.
(
2
1. Dignam JD, Lebovitz RM, Roeder RG 1983 Accurate tran- 39. Wang Z-Q, Ovitt C, Grigoriadis AE, Mohle-Steinlein U,
scription initiation by RNA polymerase II in a soluble extract
from isolated mammalian nuclei. Nucleic Acids Res 11:1475–
Ruther UC, Wagner EF 1992 Bone and haematopoietic defects
in mice lacking c-fos. Nature 360:741–745.
40. Ruther U, Komitowski D, Schubert FR, Wagner EF 1989 c-fos
expression induces bone tumors in transgenic mice. Oncogene
4:861–865.
1
489.
2. Akatsu T, Takahashi N, Debaki K, Suda T 1989 Prostaglan-
dins promote osteoclast-like cell formation by a mechanism
involving cyclic adenosine 3Ј,5Ј-monophosphate in mouse 41. Grigoriadis AE, Schellander K, Wang Z-Q, Wagner EF 1993
bone marrow cell cultures. J Bone Miner Res 4:29–35.
3. Sato T, Foged NT, Delaisse JM 1998 The migration of purified
Osteoblasts are target cells for transformation in c-fos trans-
genic mice. J Cell Biol 122:685–701.
osteoclasts through collagen is inhibited by matrix metallopro- 42. Wang Z-Q, Liang J, Schellander K, Wagner EF, Grigoriadis
teinase inhibitors. J Bone Miner Res 13:59–66.
4. Ozono K, Sone T, Pike JW 1991 The genomic action of
AE 1995 c-fos-induced osteosarcoma formation in transgenic
mice: Cooperativity with c-jun and the role of endogenous
c-fos. Cancer Res 55:6244–6251.
2
2
1
,25-dihydroxyvitamin D . J Bone Miner Res 6:1021–1027.
3
5. Crawford HC, Matrisian LM 1996 Mechanism controlling the 43. Gack S, Vallon R, Schaper J, Ruther U, Peter A 1994 Pheno-
transcription of matrix metalloproteinase genes in normal and
neoplastic cells. Enzyme Protein 49:20–37.
6. Porte D, Tuckermann J, Becker M, Baumann B, Teurich S,
typic alterations in fos-transgenic mice correlate with change
in Fos/Jun-dependent collagenase type I expression. J Biol
Chem 269:10363–10369.
2
2
2
2
Higgins T, Owen MJ, Schorpp-Kistner M, Angel P 1999 Both 44. Ducy P, Zhang R, Geoffroy V, Ridall AL, Karsenty G 1997
AP-1 and cbfa1-like factors are required for induction of
interstitial collagenase by parathyroid hormone. Oncogene 18:
Osf2/Cbfa1: A transcriptional activator of osteoblast differen-
tiation. Cell 89:747–754.
6
67–678.
45. Atkin I, Pita JC, Ornoy A, Agundez A, Castiglione G, Howell
DS 1985 Effects of vitamin D metabolites on healing of low
phosphate, vitamin D-deficient induced rickets in rats. Bone
6:113–123.
7. Selvamurugan N, Chou WY, Pearman AT, Pulumati MR,
Partridge NC 1998 Parathyroid hormone regulates the rat
collagenase-3 promoter in osteoblastic cells through the coop-
erative interaction of the activator protein-1 site and the runt 46. Koyama H, Inaba M, Nishizawa Y, Ohno S, Morii H 1994
domain binding sequence. J Biol Chem 273:10647–57.
8. Johansson N, Saarialho-Kere U, Airola K, Herva R, Nissinen
L, Westermarck J, Vuorio E, Heino J, Kahari VM 1997
Protein kinase C is involved in 24-hydroxylase gene expres-
sion induced by 1,25(OH) D in rat intestinal epithelial cells.
J Cell Biochem 55:230–240.
2
3
Collagenase-3 (MMP-13) is expressed by hypertrophic chon- 47. Reponen P, Sahlberg C, Munaut C, Thesleff I, Tryggvason K
drocytes, periosteal cells, and osteoblasts during human fetal
bone development. Dev Dyn 208:387–397.
9. Gack S, Vallon R, Schmidt J, Grigoriadis A, Tuckermann J,
1994 High expression of 92-kD type IV collagenase (gelati-
nase B) in the osteoclast lineage during mouse development.
J Cell Biol 124:1091–1102.
Schenkel J, Weiher H, Wagner EF, Angel P 1995 Expression 48. Tezuka K, Nemoto K, Tezuka Y, Sato T, Ikeda Y, Kobori M,
of interstitial collagenase during skeletal development of the Kawashima H, Eguchi H, Hakeda Y, Kumegawa M 1994

2
30
UCHIDA ET AL.
Identification of matrix metalloproteinase 9 in rabbit oste-
type matrix metalloproteinase MT1-MMP in osteoclasts. J Cell
oclasts. J Biol Chem 269:15006–15009.
Sci 110:589–596.
4
9. Okada Y, Naka K, Kawamura K, Matsumoto T, Nakanishi I,
Fujimoto N, Sato H, Seiki M 1995 Localization of matrix
metalloproteinase 9 (92-kilodalton gelatinase/type IV collage-
nase ϭ gelatinase B) in osteoclasts: Implications for bone
resorption. Lab Invest 72:311–322.
Address reprint requests to:
Motoyuki Uchida, Ph.D.
Biomedical Research Laboratories
Kureha Chemical Industry Co., Ltd.
3–26-2 Hyakunin-cho
5
0. Sakai D, Tong HS, Minkin C 1995 Osteoclast molecular
phenotyping by random cDNA sequencing. Bone 17:111–119.
1. Rice DP, Kim HJ, Thesleff I 1997 Detection of gelatinase B
expression reveals osteoclastic bone resorption as a feature of
early calvarial bone development. Bone 21:479–486.
5
Shinjuku, Tokyo 169-0073, Japan
5
2. Sato T, Ovejero MC, Hou P, Heegaard AM, Kumegawa M, Received in original form January 31, 2000; in revised form July 3,
Foged NT, Delaiss e´ JM 1997 Identification of the membrane- 2000; accepted September 13, 2000.