| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
,1
* Department of Periodontology, Academic Centre for Dentistry, and
Department of Cell Biology and Histology, Academic Medical Centre, University of Amsterdam, Amsterdam, The Netherlands; and
Celltech, Slough, United Kingdom
1Correspondence: Academic Medical Centre, Department Cell Biology and Histology, P.O. Box 22700, 1100 DE Amsterdam, The Netherlands. E-mail: v.everts{at}amc.uva.nl
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: MMP activity • cathepsin K • proteinase inhibitor • cysteine proteinase
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2
3)
. Is resorption of
all calcified tissues the same? Various bone diseases in fact suggest
differences in resorption patterns. Bone abnormalities due to increased
or decreased bone resorption are often systemic in nature, such as in
osteopetrosis (4)
, but resorptive disturbances can be restricted to
certain parts of the skeleton such as in craniometaphyseal dysplasia
(5)
. This phenomenon in itself suggests differences in activity of the
resorbing cells or in regulation of their activity, but a more direct
clue provides yet unexplained contradictory results obtained from
in vitro studies. In these studies, attempts were made to
unravel the enzyme machinery involved in osteoclastic digestion of
matrix components of mineralized tissues. One class of enzymes, the
cysteine proteinases (CPs;
6
7
8
9
),2
particularly cathepsin K (10
11
12
13)
, is clearly essential for digestion.
However, data are controversial with respect to another class of
proteinases, the matrix metalloproteinases (MMPs). It has been
convincingly demonstrated that MMPs are involved in osteoclastic bone
digestion in calvarial tissue explants (9
, 14
, 15)
. Yet when isolated
osteoclasts were seeded on bone slices, participation of MMPs in the
process of resorption could not be shown (16
17
18)
. A possible
explanation for this discrepancy may be that the source of the
osteoclasts was different. Studies with isolated osteoclasts always
use cells from long bones, whereas involvement of MMPs in osteoclastic
bone digestion was established in calvarial bone. Thus, the question
arose whether osteoclasts from different sites of the body would
degrade bone matrix in different ways, and if so, whether these
differences would be intrinsically related to the osteoclast or to the
substratum they resorb. In the present study, we investigated whether
osteoclasts of calvariae and long bones differ in their resorptive
activity. We analyzed involvement of two classes of proteinases, the
CPs and MMPs, in this process. Long bone and calvarial explants were
obtained from mice, cultured in the presence of low molecular weight
inhibitors selective for these proteinases, and bone digestion was
morphometrically analyzed at the ultrastructural level (9
, 14)
. In
addition, enzyme activities were determined biochemically and enzyme
histochemically in rabbit bones. Finally, we isolated osteoclasts from
rabbit calvariae and long bones, seeded onto bone slices obtained from
bovine skull and cortical bone, and analyzed for their resorptive
activity in the presence of selective proteinase inhibitors. |
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-MEM, fetal calf serum (FCS), and
antibiotics (penicillin, streptomycin, amphotericin) were obtained
from Life Technologies, Inc. (Grand Island, N.Y.). The proteinase
inhibitors E-64 and PMSF were purchased from Sigma Chemical Co. (St.
Louis, Mo.). The synthetic MMP inhibitors CT1166
(N1-[1-(S)-(morpholinosulfonylaminoethylaminocarbonyl)-2-cyclohexylethyl]-N4-hydroxy-2-(R)-[3-(4-methylphenyl)-propyl]succinamide),
CT1399
(N1-(S-(morpholinosulfonyl-aminoethylaminocarbonyl)-2-cyclohexylethyl]-N4-hydroxy-2-(R)-(4-chlorophenylpropyl)succinamide),
CT1746
(N1-[2-(S)-(3,3-dimethylbutanamideyl)]-N4-hydroxy-2-(R)-[3-(4-chlorophenyl)propyl]succinamide),
and CT1847
(N1-[(1-(S)-methylaminocarbonyl-2-methyl-2-methylthiopropyl)]-N4-hydroxy-2(R)-(2-methylpropyl)succinanamide)
(19
, 20)
were kindly provided by Drs. T. Crabbe and R. Morphy of the
Departments of Biology and Chemistry, respectively (Celltech
Therapeutics Ltd., Slough, U.K.). The mouse anti-human
vß3 antibody, a mouse
IgG1 control antibody, and FITC-labeled
anti-mouse IgG were purchased from Instruchemie (Hilversum, The
Netherlands). The proteinase substrates
Z-Arg-Arg-Arg-4-methoxynaphtylamide(MNA) (cathepsin B
substrate; 21
) and Z-Gly-Pro-Arg-4-MNA (cathepsin K substrate; 22
) were
from Bachem (Bubendorf, Switzerland). The MMP-substrate TNO211
(Dabcyl-Gaba-Pro-Gln-Gly-Leu-Glu(EDANS)-Ala-Lys-NH2;
23
) was kindly donated by Drs. B. Beekman and J. M. TeKoppele
(TNO, Leiden, The Netherlands). All other reagents used were of
analytical grade.
Bone explant cultures
Five-day-old Swiss random mice were decapitated, and the
calvarial and metacarpal bones were isolated as described previously
(7
, 9)
. The bone explants were cultured for 24 h in M-199 with
antibiotics and 2.5% fetal calf serum. The following proteinase
inhibitors were added to the explants: the cysteine proteinase
inhibitor E-64 at a final concentration of 40 µM; the MMP inhibitors
CT1166, CT1399, CT1746, or CT1847, each at a final concentration of 10
µM. These concentrations of inhibitors are sufficient to inhibit the
proteinases of the respective class of proteolytic enzymes (6
, 20
, 24)
.
The inhibitors do not interfere with the activity of other proteinases.
DMSO used to dissolve the MMP inhibitors was added in equal amounts
(0.1%) to explants cultured in the absence of inhibitors (controls) or
in the presence of the CP inhibitor.
After the culture period, explants were fixed for 48 h at room temperature in 4% paraformaldehyde and 1% glutaraldehyde in 0.1 M sodium cacodylate buffer (pH 7.4). The bones were then washed, postfixed in 1% OsO4, washed in buffer, dehydrated through a graded series of ethanol, and embedded in epoxy resin (LX-112).
Semi-thin sections were made with glass knives perpendicular to the surface of the calvarial bone explants and in the mid-sagittal plane of the long bones. The sections were stained with methylene blue, and areas with osteoclasts were selected. Ultrathin sections were cut of these areas, stained with lead and uranyl, and examined in a Zeiss EM10 electron microscope. For morphometric analysis, each bone explant was represented by one randomly chosen ultrathin section.
Morphometric analysis
Morphometric analysis of the resorption lacunae was performed as
described previously (9
, 14)
. All osteoclasts attached to bone as well
as their resorption zones were micrographed at a low magnification of
x950. In addition, all areas of demineralized bone not occupied by
osteoclasts were micrographed. The micrographs were printed at a final
magnification of x5700, coded, randomized, and the surface area of
demineralized bone matrix was quantitatively assessed using a
computerized X-Y tablet (MOP-Videoplan, Kontron, Muenchen, Germany).
The data were expressed as mean square micrometer of demineralized area
per osteoclast (DA/OC) ± SE of five or six explants.
Enzyme histochemistry
Five-day-old Chinchilla rabbits were decapitated; calvarial,
metacarpal, and metatarsal bones were isolated, periosteal tissue was
removed, and the bones were frozen directly in liquid nitrogen for
biochemical analysis or immersed in 8% gelatin and subsequently frozen
in liquid nitrogen for enzyme histochemical detection of CPs.
Sections were made on a motor-driven cryostat, collected on adhesive
tape (21)
, and used to detect enzyme activity. The sections were
incubated for up to 180 min at 37°C in an aqueous solution of
Z-Arg-Arg-Arg-4-MNA (1 mg/ml cathepsin B substrate) or
Z-Gly-Pro-Arg-4-MNA (1 mg/ml cathepsin K substrate) in 100 mM phosphate
buffer (pH 6.0; in the presence of dithiothreitol, EDTA, and cysteine)
according to the method described by Van Noorden and Frederiks (21)
.
Immediately after incubation, the sections were washed in 10 mM
N-ethylmaleimide in 0.1 M phosphate buffer (pH 8.0) to stop the
reaction. To visualize liberated MNA, nitrosalicyladehyde or Fast Blue
BB was added to the incubation medium. Control sections were incubated
in the presence of substrate and E-64. The insoluble fluorescent (NSA)
or colored (FBB) end products were examined in a Leica microscope with
or without epifluorescence and micrographs were made with Kodak 400 ASA
film.
In consecutive sections, osteoclasts were localized by staining the
sections for the activity of tartrate-resistant acid phosphatase (TRAP)
according to the method described by Andersson and Marks (25)
.
Biochemical analysis
Periosteum-free rabbit long bones (metacarpals and metatarsals)
and calvariae were cut into small pieces and immersed in sodium acetate
buffer (0.1 M, pH 5.3; 0.2% Triton X-100) or sodium cacodylate buffer
(pH 7.4; 0.2% Triton X-100), sonicated (3 x5 s) on ice,
extracted overnight in the same buffer at 4°C, and again sonicated
and centrifuged at 15,000 rpm (5 min). The supernatant was collected
and frozen at -20°C. The pellet was used for analysis of protein and
DNA content. Protein content was assessed with the BCA method (Pierce,
Rockford, Ill.) and DNA with the method described by Janakidevi et al.
(26)
.
Cathepsin B and K activity
Activity of the CPs cathepsins B and K was analyzed using the
substrates mentioned above (see `Enzyme histochemistry'). Acetate
buffer-extracted tissue samples were equalized with respect to protein
or DNA content, incubated with the respective substrates for up to 60
min, and analyzed fluorometrically (Jasco, Tokyo, Japan). Controls
included incubations in the presence of substrate but without extract
and in the presence of substrate with extract, including a proteinase
inhibitor (E-64, CT1166, EDTA, or PMSF).
MMP activity
Activity of MMPs was assessed in cacodylate buffer-extracted
tissue samples according to the method described by Beekman et al. (23)
by using the fluorogenic substrate TNO211. Controls included
incubations in the presence of substrate without extract and in the
presence of substrate with extracts to which different enzyme
inhibitors were added (see above). In addition, tissue extracts were
electrophoresed and subjected to gelatin zymography as described
previously by Creemers et al. (24)
.
Studies with isolated osteoclasts
Long bones and calvariae were dissected from 5-day-old
Chinchilla rabbits, cleaned of their periosteum, and minced in -MEM
and 5% FCS. The minced bone fragments were sedimented and the
supernatant was centrifuged for 10 min at 500 rpm. Cell numbers were
established and 2 x 107/ml cells were added
to 1 mm thick bone slices (diameter 1.5 cm) obtained from bovine
cortical and calvarial bone, which were placed in 24-well culture
plates. After a 3 h incubation, the slices were thoroughly washed
with culture medium and subsequently incubated for 48 h with or
without the CP inhibitor E-64 (40 µM) or the MMP inhibitor CT1166 (10
µM). After the culture period, slices were cleaned in 0.25 M
NH4OH, washed in distilled water, stained with
Coomassie brilliant blue, and the number of resorption pits was
determined at the light microscopic level. A series of slices was
dehydrated with ethanol, immersed in hexamethyldisilazone, air-dried,
evaporated with gold, and the resorption pits were analyzed in a
scanning electron microscope (SE 525, Philips, Eindhoven,
The Netherlands).
In addition, isolated cells were seeded on plastic culture dishes.
After an incubation period of 3 h, osteoclasts were attached to
the bottom of the dishes. The dishes were washed extensively and
incubated to determine cathepsin B or K activity (27)
, TRAP activity,
or immunolocalization of the vitronectin receptor
vß3.
Immunolocalization of the integrin
vß3 was performed on
isolated cells by fixing the cells with 4% paraformaldehyde, washing
with phosphate-buffered saline (PBS), and incubation with the primary
antibody (anti-vß3)
for 60 min. Cells were washed extensively with PBS and incubated with
FITC-labeled anti-mouse IgG, washed again, and visualized in a Leica
microscope with epifluorescence. Control cells were incubated with an
irrelevant mouse IgG antibody.
Mice with collagenase-resistant type I collagen
Fully grown mice with or without mutated type I collagen
rendering them resistant to collagenase (28)
were kindly donated by Dr.
R. Jaenisch (Massachusetts Institute of Technology, Whitehead
Institute, Cambridge, Mass.). The mice were decapitated; long bones
(tibiae and ulnae) and calvariae were dissected, fixed (see above), and
embedded in epoxy resin. Light microscopic sections were cut at the
mid-sagittal plane of the long bones and perpendicular to the surface
of the calvaria. Thickness and length of the bones were established at
the light microscopic level with a calibrated eyepiece.
Statistical analysis
Statistically significance of differences was determined
by using the Kruskal-Wallis nonparametric analysis of variance test,
followed by Tukey-Kramer's multicomparison test. Student's
t test was used when single comparisons were made. Effects
were considered statistically significant when P < 0.05
(two-tailed).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Such areas were virtually absent in control explants. The effects of
E-64 were found in both long bone and calvarial explants (see also refs
7
, 9
), and appeared to be comparable in either type of bone (Fig. 1B, E
).
|
When MMPs were inhibited with the selective inhibitor CT1166,
subosteoclastic demineralized areas were abundant in calvarial explants
(Fig. 1F
) and were comparable to those caused by the CP
inhibitor (9)
. However, inhibition of MMPs proved to have hardly an
effect in long bones, i.e., demineralized regions were virtually absent
(Fig. 1C
). These observations were confirmed by morphometric
analysis of the explants (Fig. 2
), showing that the volume density of demineralized bone matrix in long
bone explants cultured in the presence of the MMP inhibitor did not
differ from control explants.
|
We then analyzed the effects of a series of MMP inhibitors and
confirmed that CT1166 MMP inhibitors did not cause demineralized areas
(Fig. 3
). On the other hand, the MMP inhibitors induced large demineralized
areas in calvarial explants.
|
Proteinase activity
Proteinase histochemistry in tissues
Cryostat sections of long bones and calvariae were
incubated for localization of the activity of the CPs cathepsins B and
K. Activity was found in either type of bone, but remarkable
differences in the amounts of reaction product were found. In long
bones, relatively high activity of these enzymes was apparent, whereas
considerably lower levels were present in calvarial bone (Fig. 4
).
|
Activity of cathepsin B was not restricted to osteoclasts, but was also found in other cells such as periosteal fibroblasts and chondrocytes. Cathepsin K activity was almost exclusively detected in osteoclasts in close spatial relationship to the bone surface. Localization of TRAP activity in consecutive sections confirmed that the cathepsin K-positive cells were osteoclasts.
Proteinase activity in isolated cells
Cathepsin K
Isolated cells from both long bones and
calvariae that had been seeded on plastic dishes showed cathepsin K
activity, which was monitored as a fluorescent end product for up to
3 h after addition of the substrate. During the first hour of
incubation, activity was found only in multinucleated osteoclasts
(Fig. 5
).After prolonged incubation, some precipitate also appeared in
mononucleated cells, although the level in these cells was low.
Incubations performed in the presence of the CP inhibitor E-64
completely prevented formation of fluorescent final reaction product.
|
By comparing long bone osteoclasts and calvarial osteoclasts, we found
a striking difference in enzyme activity: long bone osteoclasts
contained a high level whereas only small amounts of reaction product
were found in osteoclasts obtained from calvariae (Fig. 5)
. Assessment
of the number of positive and negative osteoclasts after a 2 h
incubation revealed that ~30% of the calvarial osteoclasts and over
95% of the long bone osteoclasts contained final reaction product.
Cathepsin B
Comparable data were obtained with cathepsin B:
high activity was found in osteoclasts of long bone and a low activity
in calvarial osteoclasts (data not shown).
TRAP
Localization of the activity of TRAP in isolated
osteoclasts demonstrated a comparable level of enzyme activity in the
two osteoclast populations (data not shown).
Biochemical analysis of proteinase activity
Cathepsins B and K
Long bone and calvarial bone extracts
were analyzed for activity of cathepsin B and K. Extracts from either
type of bone contained enzyme activity. The activity of both enzymes
was significantly higher in extracts of long bones than in those of
calvarial bones (Fig. 6
). Cathepsin B activity was ~10-fold higher than activity of cathepsin
K. No activity was found when incubations were carried out in the
presence of E-64. Other inhibitors, PMSF for serine proteinases and
EDTA or CT1166 for MMPs, had no effect on the activity.
|
TRAP
TRAP activity was similar in the two types of bone
(long bones: 3.54 ±1.87 mU/mg protein; calvariae: 3.28±0.71mU/mg
protein; n=5 bone extracts of each bone type,
mean ±SD; not significant).
MMPs
Determination of MMP activity in the bone extracts by
using a synthetic MMP-substrate revealed no differences in enzyme
activity between long bones and calvariae (long bones, 0.21±0.04 A/mg
protein; calvariae 0.19±0.02 A/mg protein; n=7 bone
extracts of each bone type, mean absorbance units
±SD; not significant).
Gelatin zymography of bone extracts indicated the presence of various
bands with gelatinolytic activity (92, 72, and 66 kDa; Fig. 7
). The overall pattern was comparable for the two types of bone, and no
major difference in activity was found.
|
Bone resorption by isolated osteoclasts
Osteoclasts were isolated from rabbit long bones and calvariae and
seeded on slices from bovine cortical bone or skull. The two osteoclast
populations proved to resorb these matrices. Scanning electron
microscopic analysis revealed large numbers of resorption pits. No
differences were observed between pits formed by both types of
osteoclasts. Resorption pits formed in skull slices were less regular
in shape than in cortical bone slices; the bottom of the pits in the
skull slices had a more fibrillar appearance.
In the presence of the CP inhibitor E-64, resorption pits were more
shallow and smaller. This was found for both types of osteoclasts.
Inhibition of the activity of MMPs had no effect on the resorption by
long bone osteoclasts. Resorption by calvarial osteoclasts,
however, was clearly disturbed. The pits were shallow and comparable to
those found after incubation in the presence of the CP inhibitor
(Fig. 8
).
|
The number of pits was decreased in the presence of the CP inhibitor
for both osteoclast populations (Fig. 9
). Inhibition of MMPs had no effect on the number of pits produced by
long bone osteoclasts, but significantly reduced the number of pits
formed by calvarial osteoclasts.
|
Immunolocalization of vß3
Since the vitronectin receptor,
vß3, is highly
expressed in osteoclasts and relatively selective for this type of cell
(29)
, it was immunolocalized in isolated osteoclasts attached to
plastic dishes. Both cell populations were positive for this receptor
(not shown). In addition to the osteoclasts, some mononuclear cells
with a round morphology proved to be positive, too. Fibroblasts were
negative.
Collagenase-resistant collagen type I mouse
If, indeed, MMPs are important for osteoclastic digestion of
calvarial bone and less essential for resorption of long bone, it may
be hypothesized that differences exist in the resorption of bones if
the major constituent of the matrix, type I collagen, is resistant to
MMP-1. Therefore, we analyzed bones of a MMP-1-resistant collagen type
I mouse and compared this with a control. Measurements of the width,
thickness, and length of long bones of these mice revealed no
differences. The calvarial bone of the mutant mouse, however, was much
thicker (by approximately fourfold) than that of its normal littermate
(Fig. 10
).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
E-64 and other CP inhibitors (14)
induce in either type of bone large
demineralized areas of undigested matrix adjacent to actively resorbing
osteoclasts, indicating that demineralization continues whereas
digestion of the matrix is inhibited (7
, 9)
. These findings demonstrate
that CPs are essential for the degradation of bone matrix constituents
in the subosteoclastic resorption zone. Recent data indicate that the
CP cathepsin K is one of the most important enzymes involved in
osteoclastic bone digestion (see beginning of text). In this respect,
it is of interest that a rare osteopetrosis-like bone disease,
pyknodysostosis, is characterized by the presence of large
demineralized areas adjacent to osteoclasts (30)
. Since this phenomenon
was mimicked in an in vitro model system by inhibiting CP
activity (7)
, we hypothesized that pyknodysostosis was caused by a
decreased activity of these enzymes. Recently, several studies have
provided additional evidence for this hypothesis by showing a mutation
in the cathepsin K gene in pyknodysostosis patients (31
, 32)
. Saftig
and co-workers (33)
confirmed the significance of the enzyme in
resorption in their study of cathepsin K-deficient mice; however, it is
not yet known whether this enzyme is equally important in all parts of
the skeleton. The present study strongly suggests otherwise because of
the distinct differences in activity of cathepsin K between calvarial
and long bone osteoclasts. Whereas the high levels found in long bone
osteoclasts is in line with its alleged importance in resorption (33)
,
the low levels in calvarial osteoclasts suggest that cathepsin K is
less important for the digestion of this type of bone. In line with
this assumption are data presented by Saftig and co-workers (33)
, who
were unable to find anomalies in calvarial bones of cathepsin
K-deficient mice, whereas resorption of the long bones was clearly
affected. Since inhibitors that are more selective for cathepsins B and
L have profound inhibitory effects on calvarial bone resorption (8
, 14
, 34
35
36)
, we suggest that, in addition to cathepsin K, an important role
is played by other CPs as well.
A remarkable finding of the present study was the lack of an effect of
MMP inhibitors on osteoclastic digestion of long bone, whereas
calvarial digestion was significantly affected (9
, 14
, 15)
. We also
showed that calvarial bones of a collagenase (MMP-1) -resistant
collagen type I mouse (28)
were considerably thicker than those of a
control animal, whereas long bones were normal. Although the thickened
calvarial bone may have been caused by increased bone apposition, a
more likely explanation is that resorption was decreased. After all,
the major fraction of bone collagen is type I collagen, which is
resistant to MMP-1, and our findings indicate that calvarial bone
resorption by osteoclasts depends on MMP activity, whereas resorption
of long bones appears to be independent of MMP activity. Several data
indicate, however, that MMPs do play a role in the processes that
eventually result in osteoclastic resorption of long bones. The enzymes
are not only involved in resorption of nonmineralized collagen from the
bone surface prior to osteoclastic attack (18
, 37)
, but also in
migration of osteoclasts (38
, 39)
. The latter finding may explain the
presence of a number of MMPs in long bone osteoclasts (40
41
42)
. Our
inability to detect differences in the level of MMP activity in the two
types of bone extracts is likely to be due to the involvement of MMPs
in these processes, but also to their participation in events not
directly related to osteoclastic bone degradation.
Differences in resorptive activities between the osteoclast
(sub)populations described here may either be caused by differences in
the substrate they resorb (intramembranous vs. endochondral bone) or
due to phenotypical heterogeneity of osteoclasts. The experiments with
the isolated osteoclasts seeded on either type of bone indicate that
the substrate itself has no obvious modulating effect on the activity
of osteoclasts or the enzymes used for digestion. Inhibition of MMPs
did not affect resorption of bone slices obtained from either type of
bone by long bone osteoclasts, whereas resorption by calvarial
osteoclasts was inhibited. Although we cannot exclude the possibility
that mononucleated cells in the direct vicinity of the osteoclasts
somehow modulated the activity of these cells (e.g., ref 43
), our data
strongly suggest phenotypic differences between the two osteoclast
populations. If, indeed, such differences do exist, the important
question arises as to whether the cells at the different sites
originate from different progenitors. It has been well established that
hematopoietic cells of the monocyte lineage are progenitors of
osteoclasts (e.g., refs 44
, 45
). Could it be that different subsets of
circulating monocytes give rise to different subsets of osteoclasts or
are osteoclasts generated by fusion of specific local cells with
blood-born progenitors common to all osteoclast populations?
Further research is under way to elucidate this intriguing
problem.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Received for publication January 27, 1999. Accepted for publication February 15, 1999.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. P. M. Lutgens, K. B. J. M. Cleutjens, M. J. A. P. Daemen, and S. Heeneman Cathepsin cysteine proteases in cardiovascular disease FASEB J, October 1, 2007; 21(12): 3029 - 3041. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. Chen, S. Yang, Y. Abe, M. Li, Y. Wang, J. Shao, E. Li, and Y.-P. Li Novel pycnodysostosis mouse model uncovers cathepsin K function as a potential regulator of osteoclast apoptosis and senescence Hum. Mol. Genet., February 15, 2007; 16(4): 410 - 423. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Thomas, P. Moffatt, P. Salois, M.-H. Gaumond, R. Gingras, E. Godin, D. Miao, D. Goltzman, and C. Lanctot Osteocrin, a Novel Bone-specific Secreted Protein That Modulates the Osteoblast Phenotype J. Biol. Chem., December 12, 2003; 278(50): 50563 - 50571. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Spessotto, F. M. Rossi, M. Degan, R. Di Francia, R. Perris, A. Colombatti, and V. Gattei Hyaluronan-CD44 interaction hampers migration of osteoclast-like cells by down-regulating MMP-9 J. Cell Biol., September 16, 2002; 158(6): 1133 - 1144. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. T. Engsig, Q.-J. Chen, T. H. Vu, A.-C. Pedersen, B. Therkidsen, L. R. Lund, K. Henriksen, T. Lenhard, N. T. Foged, Z. Werb, et al. Matrix Metalloproteinase 9 and Vascular Endothelial Growth Factor Are Essential for Osteoclast Recruitment into Developing Long Bones J. Cell Biol., November 13, 2000; 151(4): 879 - 890. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Miyaura, M. Inada, T. Suzawa, Y. Sugimoto, F. Ushikubi, A. Ichikawa, S. Narumiya, and T. Suda Impaired Bone Resorption to Prostaglandin E2 in Prostaglandin E Receptor EP4-knockout Mice J. Biol. Chem., June 23, 2000; 275(26): 19819 - 19823. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
