Abstract
Background: An investigation of matrix metalloproteinase-9 (MMP-9)
and its influence on vascular invasion in the secondary ossification center at
the chondroepiphysis of developing long bones was undertaken. The effect of
MMP-9 was compared with that of basic fibroblast growth factor (b-FGF), a
potent angiogenic factor, and we assessed the chorioallantoic membrane (CAM)
culture as a model for angiogenesis in osteochondral tissue.
Methods: Seventy-two femoral and seventy-two humeral heads of
thirty-six four-day postnatal rabbits were dissected immediately after each
animal was killed. Solutions of MMP-9, b-FGF, and phosphate-buffered saline
solution were applied, and the femoral and humeral chondroepiphyseal explants
were incubated for ten days in CAM culture. This was used as an in vivo model
to investigate the growth of blood vessels into the femoral and humeral heads
of the neonatal rabbit. The explants were harvested from the CAM culture and
analyzed histologically. A three-day incubation was also performed to look for
early signs of vascular ingrowth into the cartilage matrix.
Results: One hundred and twenty epiphyses from thirty rabbits were
placed onto CAM culture successfully; of these, two were harvested at three
days to assess early changes and 118 were harvested at ten days. Forty of the
118 cultures were still viable when harvested after ten days, giving a 33%
yield. Both MMP-9 and b-FGF caused an increased vascular invasion into the
chondroepiphysis. New blood vessels derived from the chorioallantoic membrane
within cartilage canals were more numerous in MMP-9 treated epiphyses, and
larger canals were more commonly seen when compared with a control group.
Conclusions: These findings confirmed that b-FGF is angiogenic at
the chondroepiphysis. Matrix metalloproteinase-9 appears to be implicated in
vascular invasion and induces the formation of new cartilage canals at the
chondroepiphysis. The CAM culture model was a useful model for investigating
angiogenesis in osteochondral tissue.
Clinical Relevance: This study adds to the understanding of the
complex biochemical interaction that occurs in cartilage when the advancing
vasculature begins growing into the chondroepiphysis. A better knowledge of
this angiogenic process will enable a better understanding of the pathological
failure or disturbance of vasculogenesis, which results in dysplastic growth
disorders and osteonecrosis.
Within the chondroepiphysis, there is a direct interaction between the
cartilage and the developing vasculature. This interaction must initiate
vascular invasion and matrix degradation, control chondrocyte differentiation
and cartilage mineralization, and sustain the ossification process.
Cartilage is usually resistant to vascular
invasion1,2
and is known to contain a number of antiangiogenic
factors2-4.
At the time of ossification, something must happen to allow in-growth of
vessels into the matrix. It is this transformation of cartilage from an
angioresistant status that remains an area of controversy and intense research
interest. The commonly held view is that angiogenic factors are released by
chondrocytes and stored within the cartilage and the surrounding tissues until
they reach a concentration, at which point their angiogenic effect outweighs
the influence of the antiangiogenic
factors2,5,6.
This balance was first proposed by Hanahan and
Folkman7 as the
mechanism for the "angiogenic switch" and has been found to be
responsible for neoangiogenesis in tumor growth. Whether the activation of the
switch is dependent on an increase in the inductive signals or a decrease in
the inhibitory signals remains unclear.
It is known that inducers of angiogenesis, including vascular endothelial
growth factor (VEGF), transforming growth factor-b (TGF-b) and basic
fibroblast growth factor
(b-FGF)8-10
exist within cartilage and are likely to play a role in triggering the
angiogenic "switch." These inducers are mitogenic and chemotactic
for endothelial cells, resulting in vessel-wall cell proliferation and
migration toward the stimulus.
Basic fibroblast growth factor is known to exist within cartilage, and its
role as a powerful inducer of angiogenesis is well
documented11,12.
During vascular invasion at the growth plate, intense staining for b-FGF, both
inside the osteoblasts and in the matrix, has been
shown13. The early
chondrocytes of the growth plate (in the proliferative zone) show expression
of b-FGF, but, in the region of hypertrophic chondrocytes, the expression is
entirely in the
matrix14. In this
way, the concentration of b-FGF available in the matrix increases as the
vascular front is approached. This supports the notion that b-FGF is released
by chondrocytes in the growth plate and plays a role in regulating vascular
invasion. It is not known whether b-FGF exerts any effect over vascular
ingrowth into the chondroepiphysis.
In addition to angiogenic inducers, there are other factors involved in
aiding neovascularization. Connective tissues synthesize and secrete
metalloproteinases (MMPs), which are capable of degrading the components of
the extracellular
matrix15 and are
likely to play a role in bone turnover and
ossification16. It
is probable that MMPs help to degrade the matrix and thereby make way for the
advancing vasculature. The family of MMPs include collagenases (MMPs 1, 8, and
13), which degrade interstitial collagens types I, II, and III; the
gelatinases (MMPs 2 and 9), which degrade type-IV collagen and gelatin; and
the stromelysins (MMPs 3 and 10), which cleave a wider range of matrix
components, including proteoglycans, laminins, fibronectin, gelatin, and
casein. The regulation of the degradative function of the MMPs is governed by
another balance mechanism or "switch." Tissue inhibitors of matrix
metalloproteinases have been found to be widely distributed in cartilage, and
these control the expression of free MMPs such that their function is
restricted17-19.
The collagenases have been investigated in the context of vascular ingrowth
into the chondroepiphysis, and Vu et
al.20 found some
interesting occurrences in homozygous mice with a null mutation in the MMP-9,
gelatinase-B gene. Matrix metalloproteinase-9 shows high specificity for
denatured collagens (gelatin). It can cleave collagen types IV, V, and XI but
not native type-I collagen, proteoglycans, or laminins. It is frequently
expressed at sites of active tissue remodeling and neovascularization and has
been found to be highly expressed by osteoclastic
cells21. The MMP-9
null mice exhibited an abnormal pattern of skeletal growth-plate
vascularization and ossification. They showed normal maturation of
chondrocytes through to hypertrophy, but apoptosis, vascularization, and
ossification were all delayed, indicating the importance of MMP-9 in these
processes.
The aim of this study was to investigate whether MMP-9 plays a role in the
secondary ossification at the chondroepiphysis and to compare that effect to
the known angiogenic factor, b-FGF. The effects of MMP-9 and b-FGF were
studied in the rabbit chondroepiphysis due to the fact that, in that model,
extensive cartilage canals exist prior to the ossification
process22 in much
the same way as they do in humans. Smaller animals, such as mice, rats, and
guinea pigs, do not exhibit this feature.
Tissue Samples
The femoral and humeral heads from four-day postnatal rabbits were
dislocated and then divided at the femoral or humeral necks immediately after
each animal was killed. The animals were killed with use of terminal
anesthesia by trained staff of the university animal house, as per the Home
Office license. Rabbits of this age were used because, at birth, there is no
secondary ossification center, but by eight days, a large osseous epiphysis
has developed. A total of one hundred and forty-four epiphyses from thirty-six
rabbits was used, of which 120 epiphyses from thirty rabbits were placed onto
the culture medium successfully. The tissue explants therefore consisted of
cartilaginous epiphyses (chondroepiphyses) with a physeal growth plate at the
cut edge and, in most cases, some spongiosa. There was one center of
ossification (or occasionally two in the humeral heads) contained within the
explants. The epiphysis gradually merged into the physeal growth plate at the
lower cut edge of the explant. Usually, most of the spongiosa had been
removed, but on some explants, a variable amount of endochondral bone remained
attached to the ex-plant. In comparison with the femoral heads, the humeral
heads were generally larger and the secondary ossification center took up a
greater proportion of the chondroepiphysis.
Chorioallantoic Membrane Culture
The chorioallantoic membrane (CAM) experimental model was utilized to
investigate the influence that b-FGF and MMP-9 exerted over vascular invasion
into the chondroepiphysis. The CAM of growing chick embryos was first used to
test resistance to vascular invasion by Eisenstein et
al.23 and Sorgente
et al.4 back in the
early 1970s. Since then, it has been used extensively as an assay system for
angiogenesis by placing aga-rose disks, methylcellulose disks, or collagen
sponges onto the
CAM24-26.
It has been used for the specific purpose of investigating angiogenic factors
by studying their effect on CAM vascularity. Fertilized eggs were incubated at
37° C for ten days. A 1-cm-square hole was sawed into the shell of the egg
with a hacksaw blade. The humeral or femoral head explants, with their
membrane attached (see next paragraph), were placed onto the CAM
(Fig.1); the cut shell squares
were then replaced and the eggs were sealed with adhesive tape and incubated
for a further ten days in a humid atmosphere, again at 37° C. The explants
were then harvested from the CAM, and the chick embryos were killed by
decapitation.
Application of Exogenous Factors
A method of administering exogenous factors to the chondroepiphyses was
devised to investigate the influence of MMP-9 and b-FGF. A tangential cut,
measuring approximately 2 mm in length, was made on the articular surface of
the epiphysis with a scalpel (Fig.
2). Two-mm-square pieces of nylon membrane were soaked with
phosphate-buffered saline (control) solutions of external factors and applied
to the cut surface of the chondroepiphysis. Prior to application of the
membranes, they were dipped onto tissue paper to remove any excess liquid. The
ex-plants, complete with the factor-soaked membrane, were then placed into
culture on the CAM. The following factors were used: murine gelatinase B
(MMP-9, CC069; Chemicon International, Hampshire, United Kingdom), which was
predominantly the inactive proenzyme diluted to 0.4 µL, and murine b-FGF
(Chemicon International).
Tissue Processing and Staining
Bone tissues were harvested from the CAM and were fixed in 4%
phosphate-buffered paraformaldehyde (pH 7.4). After fixation, the explants
were processed in graded alcohol and chloroform and embedded in paraffin wax.
Between twenty and forty 7-µm-thick sections from each explant were cut.
The sections were mounted on slides that were coated with poly-L-lysine.
Weigert's hematoxylin, alcian blue, and sirius red staining methods (after
Lison staining)27
were applied to distinguish between bone matrix (which stains red) and
cartilage (which stains blue), and the Tunel method was used for the detection
of in situ DNA fragmentation. Following staining, the results were analyzed
with use of light microscopy.
New cartilage canals that had grown in from the CAM were distinguished from
those that were present before dissection by way of a useful morphological
variance. Chick erythrocytes are nucleated, whereas rabbit erythrocytes are
not. That differentiation, as seen with use of high-power microscopy,
indicated which blood vessels in the cartilage canals were derived from the
CAM. Figure 3, A shows
nucleated chick erythrocytes in a canal (arrow). Canals with no nucleated
erythrocytes can be assumed to predate the dissection and CAM culture of that
epiphysis. With use of this method, the number of new, CAM-derived vessels was
defined.
One hundred and twenty epiphyses from thirty rabbits were placed into the
CAM culture successfully; of these, two epiphyses were harvested at three days
to assess early changes and the remaining 118 epiphyses were harvested at ten
days. Forty of the 118 cultures were viable after ten days, giving a 33%
yield. This yield is lower than previously found in studies of rabbit
chondroepiphyses placed on CAM, in which the yield was as high as
70%6. This was
probably due to a variety of factors, including (1) our initial inexperience
with the CAM culture technique, (2) the complexity of the model used, and (3)
the addition of factors (phosphate-buffered saline solution, b-FGF, and MMP-9)
as well as the osteochondral explants onto the CAM.
Figure 3, B shows
the results, after ten days of incubation, of a humeral head to which a
membrane soaked in phosphate-buffered saline solution (a control substance)
had been applied. The alcian blue stain has stained the cartilage blue, and
the sirius red stain has stained the areas of bone red. The outline of the
humeral head, the cartilage matrix, and a developing central ossification
center can be seen. The cartilage matrix is continuous and is interrupted only
by a single cartilage canal adjacent to the ossification center (arrow).
Figure 3, C shows
the results, after ten days of incubation, of a humeral head to which a
membrane soaked with b-FGF had been applied. This slice through the humeral
head shows similar components, including two ossification centers (a normal
variant in the humeral head), but the notable feature is the greatly increased
number of cartilage canals in the matrix, as shown by the areas in the
cartilage matrix in which there is no staining, thus representing a cross
section through a canal. This increase in the number of CAM-derived cartilage
canals indicates that vascular invasion is taking place more widely than was
seen in controls.
Figure 3, D shows
the results, after ten days of incubation, of a humeral head to which a
membrane soaked with MMP-9 had been applied. It depicts many cartilage canals
throughout the cartilage matrix. In particular, canals can be seen breaching
the external surface of the epiphysis, which is a feature that is rarely seen
in control epiphyses. This increased level of vascular ingrowth was
substantially greater than that seen in the control specimens and was
considerably greater than that seen in the b-FGF membranes. Even at only three
days of incubation, epiphyses with an MMP-9-treated membrane showed a massive
vascular invasion, which was greater than that seen in control epiphyses from
previous studies6.
The canal that breached the growth plate in
Figure 3, E (arrow)
was quite large, and no canals of this size were seen in control or b-FGF
epiphyses.
An unexpected observation was made for one particular epiphysis that had
been incubated with an MMP-9 membrane.
Figure 3, F shows a
femoral head with a central secondary ossification center. Toward the right of
the epiphysis, it can be seen where a slice was made across the articular
surface and where an MMP-9 membrane had been placed. There has been in situ
development of another "pseudo-ossification" center beneath the
membrane (arrow).
After immunocytochemical staining for b-FGF receptors, the distribution of
receptors within the matrix was analyzed. In general, receptor expression of
b-FGF was located in semicircular patterns between the developing ossification
center and the articular surface. The expression was not greatly altered in
the various experimental groups (control, b-FGF, or MMP-9 membranes).
Another feature seen was a clear association between the developing canals
and the chondrocytes that stained for b-FGF receptor. Only in chondrocytes
that were proximal to the central canal was receptor expression found. Again,
this pattern was not altered in the various experimental groups.
Eisenstein et
al.23 and Sorgente
et al.4 were the
first to report that vascular tissues are invaded by host blood vessels when
placed onto CAM but that tissues which are normally avascular resist invasion.
Roach et al.6
confirmed that the cartilage of the chondroepiphysis was generally resistant
to vascular invasion but questioned whether that resistance could be
overridden locally by appropriate stimuli. In this study, we investigated
whether b-FGF, an angiogenic cytokine, and MMP-9, a matrix metalloproteinase
with the ability to cleave components of the extracellular matrix, could
override the suppression of angio-genesis in the epiphysis and cause vascular
ingrowth.
Our findings support the observation that the epiphysis is resistant to
vascular invasion, but we also found that both b-FGF and MMP-9 altered that
resistance and caused a greater level of neovascularization, as measured by
the number of new cartilage canals that had penetrated the matrix of the
chondroepiphysis.
Cartilage canals develop as infoldings of the
perichondrium28,29
into the cartilage matrix and are the mechanism of vascular ingrowth that is
seen in the "quiescent" mode of
angio-genesis22. In
vivo, the balance of angiogenic and anti-angiogenic factors must be tipped in
favor of angiogenesis to create a localized angiogenic environment in the
epiphyseal cartilage and permit cartilage canal ingrowth. In this in vivo
model, we found that the concentrations of b-FGF and MMP-9 were enough to
override angiogenic suppression in localized areas and stimulate increased
vascular ingrowth through the formation of cartilage canals. This suggests
that the concentration in vivo is equal to or less than that which we
used.
The localization of b-FGF receptors that we saw was unexpected. Since
chondrocytes have been shown to synthesize and secrete
b-FGF14, it seemed
reasonable that they should act chemotactically and mitogenically on receptors
in the endothelial cells of the approaching blood vessels. We expected to find
receptor expression in these endothelial cells but did not find this to be the
case in this study. The distribution was more generalized in the cartilage
matrix and occasionally in localized chondrocytes in association with a canal.
Why should we see cartilage canal-associated expression within chondrocytes?
It is possible that b-FGF, as it diffuses from the blood within the
approaching canal, acts on the chondrocytes as part of the invasive process.
Whether it causes the up-regulation of MMP synthesis, allowing further matrix
degradation and propagation of the canal, is unclear. The release of b-FGF
from the vasculature could be involved in the initiation of chondrocyte
hypertrophy and apoptosis. It is possible that b-FGF acts on the chondrocytes
to down-regulate the synthesis and secretion of anti-angiogenic factors. The
results from a previous
work6 have shown
that diffusible factors from vessels can cause hypertrophic changes in
chondrocytes and can even initiate ossification. It is feasible that b-FGF
could diffuse from vessels within the canals and cause chondrocyte changes,
and it may even be involved in the causation of hypertrophic differentiation.
To date, experimental data in support of these hypotheses are lacking, and
further research is required.
Another important point requiring discussion is the issue of why a
gelatinase (one that cleaves collagenase types IV, V, and X) should be able to
cause neovascularization through a cartilage matrix made up largely of
collagen type II, which requires a collagenase for its initial degradation to
gelatin. It is known that collagenases do exist in the cartilage matrix. It is
possible, therefore, that collagenase degradation of type-II collagen is free
to occur and that it is only when a gelatinase degrades the resulting gelatin
that vascular invasion can take place. This would imply that the arrest of
suppression of MMP-9 by its related tissue inhibitor of matrix
metalloproteinase (from which MMP-9 must be cleaved and dissociated in order
to be
activated30,31)
is the rate-determining step for invasion and would thus play a key role in
the initiation of the cartilage canals.
The pseudo-ossification that we saw develop in an epiphysis with an MMP-9
membrane laid on it was of interest. Ossification is thought to occur when
converging canals that contain blood vessels release diffusible factors that
lead to hypertrophy and apoptosis of chondrocytes. A process of calcification
and chondroclastic resorption of the cartilage matrix and osteoblastic bone
deposition follows. The observation that MMP-9 was able to initiate this whole
process is novel. It implies that not only can MMP-9 degrade the matrix in
resting zone cartilage, as seen with cartilage canals in
"quiescent" angiogenesis, but also that it can initiate
chondrocyte hypertrophy similar to that which occurs in
"reactionary" angiogenesis at the growth plate and, later on, at
the secondary ossification center.
Chondroepiphyseal cartilage matrix is usually resistant to vascular
invasion. The concentration of anti-angiogenic factors outweighs that of
angiogenic factors at this stage. At a given point in time, in localized areas
of cartilage, the balance is altered in favor of the angiogenic factors, and
the angiogenic switch is activated. This allows the ingrowth of blood vessels
through perichondrial infoldings that develop into tubular structures, or
cartilage canals, growing into the cartilage matrix. A whole host of factors,
including the balance of angiogenic and/or anti-angiogenic components, the
matrix degradation by MMPs, and chondroclastic activity as well as the
hypertrophy and eventual apoptosis of chondrocytes, are involved in the
propagation of the vasculature into the matrix and the initiation of the
secondary ossification center. This study has shown that b-FGF and MMP-9 may
be important components of vascular invasion, with MMP-9 acting both in its
capacity of matrix degradation and in its ability to ameliorate endothelial
cell morphogenesis. This study demonstrated that each of these substances can
initiate an increased vascular invasion into the chondroepiphysis in a model
of "quiescent" angiogenesis. The pseudo-ossification seen with the
MMP-9 membrane also implicates MMP-9 in the process of
"reactionary" angiogenesis.
Future work could include identification of threshold concentrations of
both b-FGF and MMP-9, below which the level of vascular invasion is not
increased above and beyond that seen in controls. This will provide
information regarding the angiogenic switch and the more intricate detail of
its activation. Furthermore, the pattern of b-FGF-receptor expression could be
investigated. Other MMPs, particularly MMP-2 (a lighter-weight gelatinase) and
MMPs 1, 8, and 13 (collagenases that degrade type-2 collagen) may be involved
as well and should be investigated. The activation of the angiogenic switch
must be under external control, whether that control be hormonal, genetic, or
through other factors, and this very important facet of the ossification
process remains unknown. It is only with further analysis of this subject area
that we will further develop our understanding of the complicated interaction
of the cartilage matrix, the factors within it, and the developing vasculature
that will eventually lead to the initiation of the secondary center of
ossification. ?
Carlevaro MF, Albini A, Ribatti D,
Gentili C, Benelli R, Cermelli S, Cancedda R, Cancedda FD. Transferrin
promotes endothelial cell migration and invasion: implication in cartilage
neovascularization. J Cell Biol.
1997;136:
1375-84.1361375
1997
[PubMed][CrossRef]
Hiraki Y, Inoue H, Iyama K, Kamizono A,
Ochiai M, Shukunami C, Iijima S, Suzuki F, Kondo J. Identification of
chondromodulin I as a novel endothelial cell growth inhibitor. Purification
and its localization in the avascular zone of epiphy-seal cartilage. J
Biol Chem. 1997;272:
32419-26.27232419
1997
[CrossRef]
Moses MA, Sudhalter J, Langer R.
Isolation and characterization of an inhibitor of neovascularization from
scapular chondrocytes. J Cell Biol.
1992;119:
475-82.119475
1992
[PubMed][CrossRef]
Sorgente N, Kuettner KE, Soble LW,
Eisenstein R. The resistance of certain tissues to invasion. II. Evidence for
extractable factors in cartilage which inhibit invasion by vascularized
mesenchyme. Lab Invest.
1975;32:
217-22.32217
1975
[PubMed]
Bergers G, Brekken R, McMahon G, Vu TH,
Itoh T, Tamaki K, Tanzawa K, Thorpe P, Itohara S, Werb Z, Hanahan D. Matrix
metalloproteinase-9 triggers the angiogenic switch during carcinogenesis.
Nat Cell Biol. 2000;2:
737-44.2737
2000
[PubMed][CrossRef]
Roach HI, Baker JE, Clarke NM.
Initiation of the bony epiphysis in long bones: chronology of interactions
between the vascular system and the chondrocytes. J Bone Miner
Res. 1998;13:
950-61.13950
1998
[CrossRef]
Hanahan D, Folkman J. Patterns and
emerging mechanisms of the angiogenic switch during tumorigenesis.
Cell. 1996;86:
353-64.86353
1996
[PubMed][CrossRef]
Harada S, Nagy JA, Sullivan KA, Thomas
KA, Endo N, Rodan GA, Rodan SB. Induction of vascular endothelial growth
factor expression by prostaglandin E2 and E1 in osteoblasts. J Clin
Invest. 1994;93:
2490-6.932490
1994
[CrossRef]
Gelb DE, Rosier RN, Puzas JE. The
production of transforming growth factor-beta by chick growth plate
chondrocytes in short term monolayer culture. Endocrinology.
1990;127:
1941-7.1271941
1990
[PubMed][CrossRef]
Gonzalez AM, Buscaglia M, Ong M, Baird
A. Distribution of basic fibroblast growth factor in the 18-day rat fetus:
localization in the basement membranes of diverse tissues. J Cell
Biol. 1990;110:
753-65.110753
1990
[CrossRef]
Wilting J, Weich HA, Christ B. Effects
of vascular endothelial growth factor and basic fibroblast growth factor:
application with corneal grafts on the chorioallantoic membrane. Acta
Anat (Basel). 1993;147:
207-15.147207
1993
[CrossRef]
Basilico C, Moscatelli D. The FGF family
of growth factors and oncogenes. Adv Cancer Res.
1992;59:
115-65.59115
1992
[PubMed]
Yu K, Xu J, Liu Z, Sosic D, Shao J,
Olson EN, Towler DA, Ornitz DM. Conditional inactivation of FGF receptor 2
reveals an essential role for FGF signaling in the regulation of osteoblast
function and bone growth. Development.
2003;130:
3063-74.1303063
2003
[PubMed][CrossRef]
Friedlander M, Brooks PC, Shaffer RW,
Kincaid CM, Varner JA, Cheresh DA. Definition of two angiogenic pathways by
distinct alpha v integrins. Science.
1995;270:
1500-2.2701500
1995
[PubMed][CrossRef]
Murphy G, Stanton H, Cowell S, Butler G,
Knauper V, Atkinson S, Gavrilovic J. Mechanisms for pro matrix
metalloproteinase activation. APMIS.
1999;107:
38-44.10738
1999
[PubMed][CrossRef]
Breckon JJ, Hembry RM, Reynolds JJ,
Meikle MC. Matrix metalloproteinases and TIMP-1 localization at sites of
osteogenesis in the craniofacial region of the rabbit embryo. Anat
Rec. 1995;242:
177-87.242177
1995
[CrossRef]
Bord S, Horner A, Hembry RM, Reynolds
JJ, Compston JE. Distribution of matrix metalloproteinases and their
inhibitor, TIMP-1, in developing human osteophytic bone. J
Anat. 1997;191:
39-48.19139
1997
[CrossRef]
Schnaper HW, Grant DS, Stetler-Stevenson
WG, Fridman R, D'Orazi G, Murphy AN, Bird RE, Hoythya M, Fuerst TR, French DL,
Quigley JP, Kleinman HK. Type IV collagenase(s) and TIMPs modulate endothelial
cell morphogenesis in vitro. J Cell Physiol.
1993;156:
235-46.156235
1993
[PubMed][CrossRef]
Murphy AN, Unsworth EJ,
Stetler-Stevenson WG. Tissue inhibitor of metalloproteinases-2 inhibits
bFGF-induced human microvascular endothelial cell proliferation. J Cell
Physiol. 1993;157:
351-8.157351
1993
[CrossRef]
Vu TH, Shipley JM, Bergers G, Berger JE,
Helms JA, Hanahan D, Shapiro SD, Senior RM, Werb Z. MMP-9/gelatinase B is a
key regulator of growth plate angiogenesis and apoptosis of hypertrophic
chondrocytes. Cell.
1998;93:
411-22.93411
1998
[PubMed][CrossRef]
Alexander CM, Hansell EJ, Behrendtsen O,
Flannery ML, Kishnani NS, Hawkes SP, Werb Z. Expression and function of matrix
metalloproteinases and their inhibitors at the maternal-embryonic boundary
during mouse embryo implantation. Development.
1996;122:
1723-36.1221723
1996
[PubMed]
Ganey TM, Love SM, Ogden JA. Development
of vascularization in the chondroepiphysis of the rabbit. J Orthop
Res. 1992;10:
496-510.10496
1992
[CrossRef]
Eisenstein R, Sorgente N, Soble LW,
Miller A, Kuettner KE. The resistance of certain tissues to invasion:
penetrability of explanted tissues by vascularized mesenchyme. Am J
Pathol. 1973;73:
765-74.73765
1973
Burkus JK, Ganey TM, Ogden JA.
Development of the cartilage canals and the secondary center of ossification
in the distal chondroepiphysis of the prenatal human femur. Yale J Biol
Med. 1993;66:
193-202.66193
1993
Yang EY, Moses HL. Transforming growth
factor beta 1-induced changes in cell migration, proliferation, and
angiogenesis in the chicken chorioallantoic membrane. J Cell
Biol. 1990;111:
731-41.111731
1990
[CrossRef]
Danesi R, Del Bianchi S, Soldani P,
Campagni A, La Rocca RV, Myers CE, Paparelli A, Del Tacca M. Suramin inhibits
bFGF-induced endothelial cell proliferation and angiogenesis in the chick
chorioallantoic membrane. Br J Cancer.
1993;68:
932-8.68932
1993
[PubMed][CrossRef]
Lison L. Alcian blue 8 G with
chlorantine fast red 5 B. A technic for selective staining of
mucopolysaccharides. Stain Technol.
1954;29:
131-8.29131
1954
[PubMed]
Agrawal P, Kulkarni DS, Atre PR. The
participation of cartilage canals in the ossification of the human fetal
calcaneum. J Anat.
1986;147:
135-42.147135
1986
[PubMed]
Stockwell RA. The ultrastructure of
cartilage canals and the surrounding cartilage in the sheep fetus. J
Anat. 1971;109:
397-410.109397
1971
Nagase H. Activation mechanisms of
matrix metalloproteinases. Biol Chem.
1997;378:
151-60.378151
1997
[PubMed]
Gibson G. Active role of chondrocyte
apoptosis in endochondral ossification. Microsc Res Tech.
1998;43:
191-204.43191
1998
[PubMed][CrossRef]