| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
(Received for publication, May 16, 1994; and in revised form, October 27, 1994) From the
The contraction of floating collagen gels is suggested to mimic
the reorganization of collagenous matrix during development and tissue
healing. Here, we have studied two osteogenic cell lines, namely MG-63
and HOS, and a chemically transformed subclone of HOS cells, HOS-MNNG.
Transforming growth factor- Hydrated collagen lattices were first described as a method to
mimic soft-tissue matrices (Elsdale and Bard, 1972). Watery milieu
within fibrous collagen net has since been used as a substrate for
growth and differentation for many types of cells. When fibroblasts are
embedded in collagen they are able to reduce the areas of the collagen
gels and form a tissue-like structure. The ultrastructure of
fibroblasts during contraction differ from the same cells plated on a
plastic surface. Differences in the prominence of cell coat, changes of
the prevalence of cell processes and the state of aggregation of their
associated microfilamentous material are seen (Bellows et al.,
1981). Collagen gels can be used to study the contraction of either
floating matrix or anchored matrix (Grinnell, 1994). In the two
variations of in vitro collagen matrix reorganization model,
the morphology and the behavior of fibroblasts differ. The first one is
suggested to mimic dermis or scars and the latter granulation tissue
(Grinnell, 1994). The exact molecular mechanism of collagen gel
contraction is unknown. Previous studies have suggested an essential
role for cell surface collagen receptors (Gullberg et al.,
1990; Shiro et al., 1991; Klein et al., 1991). Also
cellular fibronectin (FN) ( Transforming growth factor- We have determined the effects
of TGF-
The same 4.5-kb cDNA was
linked to pAW vector also in antisense orientation. Cells were
transfected and selected as above. Among the first 28 cell clones
analyzed, two showed markedly reduced expression of
Figure 1:
The effects of TGF-
Figure 2:
The
effects of 10 µg/ml plasma fibronectin (pFN) and cellular
fibronectin (cFN) on collagen gel contraction by MG-63 cells.
Cells were added together with pFN or cFN into neutralized Vitrogen-100
collagen solution and were then plated into 24-well culture plates.
Collagen polymerization was initiated by incubating the plates at 37
°C for 45 min. After the polymerization was complete, DMEM
supplemented with 10% FCS was added on gels, and the sides of the gels
were gently removed from the sides of the wells. MG-63 cells treated
with 200 pM TGF-
Figure 3:
The blocking of TGF-
Figure 4:
Effect of TGF-
Figure 5:
Integrin expression in and collagen gel
contraction by HOS cells. A, immunoprecipitation of
collagen-binding integrins from HOS cells. Cells were labeled with
[
Chemical transformation of HOS cells turns on the
expression of
Figure 6:
Integrin expression in pAW
Figure 7:
Adhesion to different substrates and
collagen gel contraction by pAW
Figure 8:
Effects of the expression of sense and
antisense
TGFs- We have tested two osteogenic cell lines, MG-63 and HOS, for their
ability to contract collagen gels and studied the effect of TGF- In addition to altered integrin
expression, TGF- Cell behavior can be regulated by
both growth factors and extracellular matrix. The fact that a growth
factor, TGF- To conclude, we have shown that TGF-
Volume 270,
Number 1,
Issue of January 6, 1995 pp. 376-382
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
Regulates Collagen Gel Contraction by Increasing
2
1 Integrin Expression in Osteogenic Cells (*)
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
(TGF-), a putative regulator of
bone fracture healing, increased collagen gel contraction by MG-63 and
HOS-MNNG, but not by HOS cells. Our data show that TGF--induced
fibronectin synthesis is not sufficient for the process. Instead,
anti-1 integrin antibodies could prevent the contraction. There
are three different integrin heterodimers that are known to mediate the
cell-collagen interaction, namely 
11, 21, and
31. In MG-63 cells TGF- increased the expression of
21 integrin and decreased the expression of 31
integrin, whereas 11 integrin is not expressed. HOS cells had
no 21 integrin, neither did TGF- induce its expression.
However, HOS-MNNG cells expressed more 21 integrin when
treated with TGF-. Thus, we suggest that the mechanism of the
enhanced collagen gel contraction by TGF- is the increased
expression of 21 integrin heterodimer. To further test this
hypothesis, we expressed a full-length 2 integrin cDNA in HOS
cells and in MG-63 cells. We obtained HOS cell clones that expressed
21 heterodimer, and the ability of these cells to contract
collagen gels was greatly enhanced. Furthermore, the contraction by
MG-63 cells transfected with 2 integrin cDNA was enhanced, and the
contraction by cells transfected with antisense oriented 2
integrin cDNA was decreased. Thus, both in MG-63 and HOS cells the
increased 2 integrin expression alone was sufficient for the
enhanced contraction of collagen gels. Furthermore, the amount of
2 integrin is critical for the process, and its decrease leads to
diminished ability to contract gels.
)has been suggested to be
required in the collagen matrix contraction by fibroblasts (Asaga et al., 1991). (TGF-)
has been discovered to enhance collagen gel contraction by skin
fibroblasts (Montesano and Orci, 1988). Later, a similar effect has
been described with other growth factors and cytokines (Clark et
al., 1989; Gullberg et al., 1990). Some others, like
interferon-, can inhibit the process (Dans and Isseroff, 1994).
TGF-
is a multifunctional regulator of cells which acts in the
process of wound healing (for review, see
Massagué, 1990). Many cell types can in vitro and in vivo produce TGF-, and it has been found
abundant in platelets and in bone. It stimulates the accumulation of
extracellular matrix (Ignotz and Massagué, 1986)
and regulates cell adhesion apparatus (Ignotz and
Massagué, 1987). TGF- is supposed to have a
central role in tissue healing. The fact that it is present in large
amounts in bone and that it increases bone formation in vivo in test animals (Noda and Camilliere, 1989; Joyce et al.,
1990) indicates the possible role of TGF- in the healing of bone
fractures. Here, we have shown that TGF- can increase collagen gel
contraction by osteogenic cells. Since TGF- is known to regulate
the integrin type collagen receptors in different cell lines (Heino and
Massagué, 1989; Heino et al., 1989) and
also to increase production of cellular FN (Ignotz and
Massagué, 1986; Roberts et al., 1987),
these are the candidate mechanisms to be responsible for the
phenomenon. Integrins are a family of cell surface receptors consisting
of two subunits, and (for review, see Hynes, 1992;
Ruoslahti, 1991). Three different integrin heterodimers are known to
mediate cell-type I collagen interaction, namely 11,
21, and 31. In addition to 1, 2, and
3 subunits, 1 subunit can form heterodimers with
4-9 subunits and v subunit. Several of these 1
integrins recognize FN as their ligand, namely 31,
41, 51, and v1. Laminin is the ligand for
11, 21, 31, 61, and
71 integrin heterodimers. on collagen gel contraction by osteogenic cell lines, in
which the pattern of collagen-binding integrins differs from each
other, and identified the molecular mechanism by which TGF-
enhances collagen gel contraction. Our data suggest that up-regulation
of 21 integrin is required and alone sufficient to increase
contraction.
Cell Cultures
Human osteosarcoma cells used were
MG-63, HOS, and HOS-MNNG (HOS cells transformed with N-methyl-N`-nitro-N-nitrosoguanidine,
tumorigenic) all from American Type Culture Collection. Cells were
maintained in Dulbecco's modification of Eagle's medium
(DMEM, Flow Laboratories, Irvine, United Kingdom) supplemented with 10%
fetal calf serum (FCS, Flow Laboratories).Cytokines and Growth Factors
Purified bovine bone
transforming growth factor-1 (TGF-1) was kindly provided by
Dr. Joan Massagué (Sloan-Kettering Institute, New
York). Human recombinant interleukin-1 (IL-1) was purchased
from Boehringer Mannheim. Concentrations used were 200 pM (TGF-) and 10 units/ml (IL-1).Antibodies
Polyclonal rabbit antiserum against
human 1 (Heino et al., 1989), 2, and 3 integrin
subunits (Santala et al., 1994) were used in
immunoprecipitation assays. Functional monoclonal antibodies used to
block the collagen gel contraction were against 1 (mab13; Akiyama et al., 1989), 2 (Gi9; Immunotech S.A, Marseille,
France), 1 (SR-84; Rettig et. al., 1984), and 3
(B1B5; Wayner and Carter, 1987) subunits.Immunoprecipitations
Cell cultures were
metabolically labeled with 50 or 100 µCi/ml of
[S]methionine (Tran
S-label, ICN
Biochemicals) for 24 h in methionine-free minimum essential medium.
When cytokines and growth factors were used, cells were preincubated
with them in serum-free medium overnight, and they were also used
during the labeling. Cell monolayers were rinsed on ice with a solution
containing 150 mM NaCl, 1 mM CaCl
, 1
mM MgCl, and 25 mM Tris-HCl (pH 7.4), and
then detached by scraping. Cell pellets obtained by centrifugation at
500 
g for 5 min were solubilized in 200 µl of the
same buffer containing 100 mMn-octyl--D-glucopyranoside (Sigma) on ice
with occasional vortexing. Insoluble material was removed by
centrifugation at 10,000 g for 5 min at 4 °C.
Radioactivity in cell lysates was counted, and an equal amount of
radioactivity was used in immunoprecipitation assays. Cytokines did not
cause any systematic changes in the incorporation of
[S]methionine into cellular proteins as
determined by trichloroacetic acid precipitation. Triton X-100 (0.5%
v/v) and bovine serum albumin (0.5 mg/ml) were added to the
supernatants, which were then precleaned by incubation with 50 µl
of packed protein A-sepharose (Pharmacia LKB Biotechnology Inc.).
Supernatants were immunoprecipitated with anti-integrin antibodies for
12 h at 4 °C. Immunocomplexes were recovered by binding to protein
A-sepharose and washing the beads four times with 25 mM Tris-buffered isotonic saline (pH 7.4) containing 0.5% Triton
X-100 and 1 mg/ml bovine serum albumin and twice with 0.5 M NaCl and 25 mM Tris-HCl (pH 7.4). The immunoprecipitates
were analyzed by electrophoresis on sodium dodecyl sulfate-containing
6% polyacrylamide gels under nonreducing conditions followed by
fluorography. Integrin bands were quantified from fluorograms by the
Microcomputer Imaging Device version M4 (Imaging Research Inc). FN
immunoprecipitations were done from
[
S]methionine-labeled methionine-free cell
culture mediums. An equal volume of medium and phosphate-buffered
isotonic saline containing 100 mMn-octyl-
-D-glucopyranoside were precleaned
with packed protein A-sepharose and incubated with polyclonal antiserum
against human plasma fibronectin (Chen et al., 1986) at 4
°C for 12 h. Immunocomplexes were recovered and washed as mentioned
previously, and reduced immunoprecipitants were analyzed by
electrophoresis and quantified from fluorograms.Northern Blot Hybridizations
Total cellular RNA
was isolated by using the guanidium thiocyanate-CsCl method (Chirgwin et al., 1979). RNAs were separated in 1%
formaldehyde-containing agarose gels, transferred to nylon membranes
(ZETA-probe, Bio-Rad), and hybridized with
[P]-labeled (Amersham) human
2 integrin
cDNA probe (Takada and Hemler, 1989).Collagen Gel Contractions
Collagen gels were
prepared by using Vitrogen-100 collagen (Celtrix). Eight volumes of
Vitrogen were mixed with two volumes of a 1:1 mixture of 10-fold
concentrated medium and 0.1 N NaOH and kept on ice. Cells were
detached, counted, and mixed gently into neutralized Vitrogen solution
before transferred into 24-well plates (50,000 cells/well). In some
experiments 96-well plates were used. Collagen polymerization was
initiated by incubating the plates at 37 °C for 45 min. DMEM
supplemented with 10% FCS and appropriate cytokines was then added on
gels before detaching the edges of the gels from the sides of the
wells. Contraction process was observed daily. After 2-6 days the
cell culture wells were photographed and the surface areas of the gels
were measured from prints. Data are presented as relative values
(experiment/control). In one set of experiments 1 or 10 µg/ml human
plasma fibronectin (pFN, Boehringer Mannheim) or cell-derived
fibronectin (cFN, Fibrogenex, Mortons Grove, IL) were added into
neutralized Vitrogen solution before adding the cells. In antibody
blocking experiments monoclonal antibodies were added similarly.Cell Adhesion Assays
The coating of a 96-well
immunoplate (Maxi Sorp, Nunc, Denmark) was done by exposure to 0.2 ml
of phosphate-buffered saline (pH 7.4) containing 4-5
µg/cm laminin (purified from basement membranes of the
Engelbreth-Holm-Swarm mouse tumor, Collaborative Research), fibronectin
(human plasma fibronectin, Boehringer Mannheim), type I collagen (from
lathyric rat skin, Boehringer Mannheim), or type IV collagen (Sigma)
for 12 h at 4 °C. Bovine serum albumin was used to measure the
nonspecific binding. Residual protein absorption sites on all wells
were blocked with 1% bovine serum albumin in phosphate-buffered saline
for 1 h at 37 °C. Confluent cell cultures were detached by using
0.01% trypsin and 0.02% EDTA. Trypsin activity was inhibited by washing
the cells with 1 mg/ml of soybean trypsin inhibitor (Sigma). Cells were
suspended in MEM (Life Technologies, Inc.). 10,000 cells were
transferred into each well and incubated at 37 °C for 45 min.
Non-adherent cells were removed by rinsing the wells with medium.
Adherent cells were fixed with 2% paraformaldehyde, stained with 0.5%
crystal violet in 20% ethanol, and washed with distilled water. The
immunoplates were allowed to air-dry, and the stain bound to cells was
dissolved into 10% acetic acid and spectrophotometrically measured at
600 nm with Multiscan Plus (Lab-Systems).Transfections
A 4.5-kb fragment of human 2
integrin cDNA (nucleotides 1-4559 in the published sequence),
containing the full translated sequence (kindly provided by Dr. Martin
Hemler; Takada and Hemler, 1989) was linked into pAWneo2 expression
vector (kindly provided by Dr. Arthur Weiss; Ohashi et al.,
1985) which carries the neomycin resistance gene. Transfections were
done by using Lipofectin reagent (Life Technologies, Inc.) according to
the manufacturer's recommendations. The neomycin analog G418
(Life Technologies, Inc.) was added to the culture medium in the
concentration of 400 µg/ml. G418-resistant cell clones were
selected for 2-3 weeks, isolated, and analyzed for their
expression of 2 integrin mRNA as well as 2 protein. Three out
of 24 clones (HOS cells), and 3 out of 19 clones (MG-63 cells) showed
high expression levels for 2 integrin. Control cells used were
transfected with pAWneo2 plasmid only.2 integrin and
the presence of antisense 2 integrin mRNA.Statistical Analysis
The statistical significance
of differences in cell adhesion and in collagen gel contraction assays
was assessed with general mixed model analysis of variance
(BMDP/Dynamic, 7.0, BMDP Statistical Software Inc., Cork, Ireland),
when multiple independent experiments were analyzed (random factor
being the experiment and fixed factor being the treatment). One way
analysis of variance with Dunnett's method (SAS statistical
program package; SAS Institute, Cary, NC) was used when one experiment
contained several differentially treated groups. Experiments with two
groups were analyzed with Student's t test for
independent samples.
TGF-
Collagen gel contraction
assays cell-type I collagen interaction. It is suggested to simulate
matrix reorganization during tissue healing. Type I collagen is the
major component of bone matrix, proposing its importance for osteogenic
cells. TGF- Increases Collagen Gel Contraction by MG-63
and HOS-MNNG Cells But Not by HOS Cells is a growth factor involved in the regulation of
fracture healing process. In this study we wanted to elucidate the
effects of TGF- on collagen gel contraction by osteogenic cells
and to find out the mechanism by which it makes other cells more
capable of contracting the gels than others. We have used in our
experiments two different osteogenic cell lines, namely MG-63 and HOS,
and a chemically transformed variant of HOS cells (HOS-MNNG). Cells
were embedded in neutralized type I collagen solution, and growth
medium supplemented with 10% of FCS was added on gels after
polymerization was complete. Without serum no contraction could be seen
(not shown). TGF- was used in the concentration of 200 pM and it constantly enhanced collagen gel contraction by MG-63 (p <0.001; mixed model analysis of variance) and HOS-MNNG
cells (p <0.05; mixed model analysis of variance; Fig. 1). In HOS cells, however, TGF- had no effect (Fig. 1). In MG-63 cells also IL-1 could increase the
contraction, but it seemed to be less potent than TGF- (Fig. 1).
and IL-1
on collagen gel contraction by MG-63 (A), HOS (B),
and HOS-MNNG (B) cells. Collagen gels were prepared by using
Vitrogen-100 collagen. 50,000 cells were added in neutralized collagen
and poured into 24-well plates. Gels were let to polymerize at 37
°C for 45 min after which the edges of the gels were gently removed
from the sides of the wells, and DMEM supplemented with 10% FCS was
added on them. Concentrations of TGF- and IL-1 used were 200
pM and 10 units/ml, respectively. After 2-4 days of
contraction, the surface areas of the collagen gels were measured. Mean
± S.D. of four parallel measurements is
shown.
Cellular Fibronectin Is Not Involved in TGF-
The importance of cellular FN in the
process of collagen gel contraction by fibroblasts has been proposed
(Asaga et al., 1991). Knowing the fact that in many cell lines
TGF--induced
Collagen Gel Contraction can increase the production of FN, this is a putative
mechanism for TGF--induced contraction. We added into the gels
both plasma- and cell-derived FN (1-10 µg/ml), but this did
not have any effect on the contraction (Fig. 2). We also
measured FN synthesis by labeling cells with
[S]methionine and immunoprecipitating FN from
the growth media. In HOS cells TGF-
increased the production of FN
10-fold (not shown). The fact that the contraction by HOS cells did not
increase although their production of FN increased when treated with
TGF- is also contradictory to the involvement of FN.
were used as a contraction control.
After 4 days the areas of the gels were measured. Mean ± S.D. of
four parallel measurements is shown.
Expression of
Previous studies have suggested the
role of collagen-receptor integrins (Gullberg et al., 1990),
especially 21 Integrin Correlates with
Collagen Gel Contraction21 (Shiro et al., 1991; Klein et al. 1991), but also 11 in collagen gel contraction by
fibroblasts. Here, the first evidence of the involvement of
1-integrins came from the experiments showing that anti-1
integrin antibody can almost completely block the TGF--induced
contraction by MG-63 cells (p <0.05; one way analysis of variance; Fig. 3). Previous studies have shown that anti- integrin
antibodies are often alone incapable of inhibiting the contraction
(Klein et al., 1991). Additionally, here the blocking effect
of anti-2 integrin antibody was much weaker than that of
anti-1 integrin antibody (inhibition was not statistically
significant; Fig. 3). Anti-1 antibody had no effect and
anti-3 antibody constantly increased contraction (p <0.05; one way analysis of variance; Fig. 3). We have
previously analyzed the integrin pattern in MG-63 cells and shown that
these cells express small amounts of 2 integrin subunit, whereas
3 integrin is the major partner of 1 subunit (Heino and
Massagué, 1989). TGF- increases the
expression of 2 subunit to about 6-8-fold (Heino and
Massagué, 1989; also shown here in Fig. 4)
and concomitantly down-regulates the expression of 3 subunit about
60-80% (Heino and Massagué, 1989). HOS
cells express both 1 and 3 subunits, but no 2 subunit
(Santala et al., 1994). Here, we show that in spite of the
lack of 21 integrin HOS cells can slightly contract the
collagen gels (Fig. 5). Anti-1 antibody can block this
phenomenon (Fig. 5). This suggests that other 1 integrins,
in addition to 21, can, at least to some extent, mediate
collagen gel contraction. More importantly, TGF- could not turn on
the expression of 2 integrin in this cell line (not shown) as it
could not increase their ability to contract collagen gels (Fig. 1).
-induced collagen
gel contraction by MG-63 cells with functional antibodies against
different collagen-binding integrin subunits. Antibodies were added
into neutralized Vitrogen-100 collagen together with the cells. After
incubation at 37 °C for 45 min, the edges of the gels were detached
from the sides of the wells, and DMEM supplemented with 10% FCS and 200
pM TGF- was added on them. The areas of the gels were
measured after 4 days of contraction. Antibodies against 1,
1, 2, and 3 subunits were used. Mean ± S.D. of
four parallel measurements is shown.
on 2 integrin
subunit biosynthesis. Confluent cultures of MG-63 and HOS-MNNG cells
were incubated in serum-free medium with TGF- for 12 h. Cells were
transferred to methionine-free medium containing 50 µCi/ml
[S]methionine and 200 pM TGF-
. An
equal amount of radioactivity from the cell lysates was precipitated
with anti-2 subunit antibody. The immunoprecipitants were analyzed
by electrophoresis followed by
fluorography.
S]methionine in methionine-free medium for 24
h. An equal amount of radioactivity from the cell lysates was incubated
with antibodies against
1, 2, and 3 integrin subunits.
Immunoprecipitants were analyzed by SDS-PAGE under nonreducing
conditions followed by fluorography. B, blocking of the
collagen gel contraction of HOS cells by functional antibody against
1 integrin subunit. 10,000 cells were added together with mAb13
into neutralized Vitrogen-100 collagen solution. Gels were let to
polymerize in 96-well plates before detaching the gels from the sides
of the wells and adding DMEM supplemented with 10% FCS on them. The
areas of the gels were measured after 6 days of incubation. The figure
shows four parallel control and anti-1 antibody-containing
wells.
2 integrin subunit (Santala et al., 1994).
Here, TGF- could slightly (about 2.5-fold) increase the expression
of 2 integrin in HOS-MNNG cells (Fig. 4). TGF- also
enhanced the contraction by these cells (Fig. 1). The data
suggest that the increased expression of 21 integrin
correlates with the enhanced collagen gel contraction by TGF-.Forced Expression of
To further examine the
involvement of 2 Integrin in HOS Cells
Increases Collagen Gel Contraction2 integrin to the phenomenon, we linked the
full-length 2 integrin cDNA (Takada and Hemler, 1989) to pAWneo2
expression vector (Ohashi et al., 1985) under a constitutive
Friend's spleen focus forming virus long terminal repeat
promoter. After transfection with the plasmid, G418-resistant HOS cell
clones were selected and analyzed by immunoprecipitation. Polyclonal
rabbit anti-2 integrin antibody immunoprecipitated two
polypeptides of 140 and 130 kDa (non-reduced) from the transfected HOS
cells, the 140-kDa polypeptide representing the 2 integrin subunit
and the 130-kDa representing the 1 subunit while the wild type
cells and cells transfected with only the vector gave no signal (Fig. 6). Importantly, the expressed 2 subunit was capable
of forming dimers with 1 subunit, which is essential for the
integrin heterodimer to be expressed on the cell surface (Heino et
al., 1989). Furthermore, the fact that these cell clones showed
increased cell adhesion to type I collagen (p <0.001; t test; Fig. 7) suggests that the collagen receptor was
functional. mRNAs encoding 2 integrin subunit were analyzed by
Northern blot hybridization. HOS wild type cells give no signal for
2 subunit mRNA (Santala et al., 1994), and MG-63 cells
were used as a positive control. A 8-kb 2 integrin mRNA was seen
in wild type MG-63 cells and a smaller about 4-kb transcript in cells
transfected with the construct containing 2 integrin cDNA (Fig. 6). This suggests that 2 integrin seen in these cells
is translated from cDNA-derived mRNA and is not due to artificial
activation of the wild type gene. HOS cells forced to express
21 integrin heterodimer were more capable of contracting
collagen gels than wild type cells or cells transfected with only the
vector carrying the neomycin resistance gene (p <0.001; t test; Fig. 7).
2
transfected HOS cells. A, immunoprecipitation of 2
integrin in HOS cells transfected with full-length 2 integrin
cDNA. Human full-length 2 integrin cDNA was linked into pAWneo2
expression vector, which carries the neomycin resistance gene. HOS
cells were transfected by using Lipofectin reagent, and stable cell
clones were selected by using G418. G418-resistant clones were tested
for their expression of 2 integrin protein in immunoprecipitation
assays (in the figure clone 16 is shown). Control cells
(marked as C) were transfected with pAWneo2 expression vector
only. Wild type HOS cells are marked as Wt. Cells were
metabolically labeled with [S]methionine, and an
equal amount of radioactivity from each cell lysate was precipitated
with
2 integrin antibody. Immunoprecipitants were analyzed by gel
electrophoresis and fluorography. B, Northern blot
hybridization of total RNA from 2 integrin-transfected HOS cells (clones 11 and 16) and MG-63 wild type cells (MG). cDNA probe specific to 2 integrin was
used.
2 transfected HOS cells. A, adhesion of HOS cells expressing 2 integrin (clones 16 and 7) to type I collagen (T I
COL), laminin (LM), type IV collagen (T IV COL),
and cellular fibronectin (cFN). 96-well immunoplates were
coated with different matrix molecules, and bovine serum albumin (BSA) was used to measure the nonspecific binding. Residual
protein absorption sites were blocked with 1% bovine serum albumin.
10,000 cells were let to adhere for 45 min after which the adherent
cells were fixed with paraformaldehyde, stained with crystal violet,
and air-dried. The cell bound stain was dissolved in acetic acid and
measured spectrophotometrically at 600 nm. B, collagen gel
contraction by HOS cells expressing 2 integrin (clone
16). HOS cells transfected with pAWneo2 vector and wild type HOS
cells were used as a control for a cell clone to be tested. 50,000
cells were added in neutralized type I collagen solution and
immediately transferred into 24-well plates. Collagen gel
polymerization was initiated by incubating the plates at 37 °C for
45 min. DMEM supplemented with 10% FCS was added on gels, and the edges
of the gels were gently removed from the sides of the wells. After 2
days of incubation at 37 °C, the areas of the gels were measured.
Two parallel experiments are shown.
The Transfection of MG-63 Cells with Antisense
The
transfection of MG-63 cells with 2
Integrin cDNA Decreases Collagen Gel Contraction2 integrin cDNA increased their
expression of 21 integrin and concomitantly their ability to
contract collagen gels (p <0.001; t test; Fig. 8). Transfection of MG-63 cells with the same expression
vector containing antisense-oriented 2 integrin cDNA produced cell
clones with decreased 21 expression (Fig. 8). Two cell
clones were further analyzed by Northern blot hybridization revealing
that 2 integrin mRNA levels had decreased to 46 and 56% of that in
control cells (not shown). Furthermore, both cell clones expressed an
approximately 4-kb transcript which was supposed to represent the
antisense mRNA. The ratios of antisense/2 integrin mRNA were one
and six (not shown). Thus, we suggest, that there is a pool of 2
integrin mRNA not reached by antisense mRNA, even if a larger amount of
antisense mRNA is synthesized. However, it is not probable that all
this mRNA is translated because the decrease in 2 integrin protein
synthesis seems to be larger than in the corresponding mRNA levels.
Both cell clones showed decreased ability to contract collagen gels (p <0.001; t test; Fig. 8). Thus, both in
MG-63 and in HOS cells the increased 2 integrin expression alone
is sufficient for the increased contraction of collagen gels.
Furthermore, the amount of 2 integrin is critical for the process,
and its decrease leads to diminished ability to contract gels.
2 integrin mRNA in MG-63-cells. Immunoprecipitation of
MG-63 cells transfected with sense (A, clone 4) or
antisense (B, clones 1-4)-oriented 2
integrin cDNA. MG-63 cells were transfected with pAWneo2 expression
vector containing the 2 integrin cDNA. Transfections were done by
using Lipofectin reagent, and G418-resistant cell clones were tested
for their expression of 2 integrin by immunoprecipitation.
Immunoprecipitants were analyzed by SDS-PAGE followed by fluorography.
Contraction of collagen gels by MG-63 cells transfected with sense (C, clone 4) or antisense (D, clones 1 and 4)-oriented 2 integrin cDNA. MG-63 cells
transfected with pAWneo2 vector only (pAW1) were used as
control cells for sense-transfection clones, and wild type MG-63 cells
were used as controls for antisense-transfected cell clones. 50,000
cells were added into neutralized Vitrogen solution and poured into
24-well plates. Collagen polymerization was initiated by incubating the
plates at 37 °C. After polymerization was complete DMEM
supplemented with 10% FCS was added on gels. After 4 days the areas of
the collagen gels were measured. Mean ± S.D. of four parallel
measurements is shown.
are a family of growth and differentiation
factors. Among other biological functions, TGFs- are supposed to
be essential for wound healing and tissue repair processes
(Massagué, 1990). In general, they increase the
accumulation of connective tissue macromolecules and angiogenesis
(Massagué, 1990). Both phenomena are important in
the formation of scars. A major source of TGF-1, and also
TGF-2, is bone matrix. In the healing of bone fractures, recently
discovered members of TGF- superfamily called bone morphogenetic
proteins (BMP 2-7) are suggested to be involved in the initiation
of the healing process (Reddi, 1992). TGF- is more probably
involved in the other stages of the healing process, e.g. in
the synthesis of new matrix (Reddi, 1992). The important role of
extracellular matrix and growth factors for bone forming cells has also
been emphasized in the recent review by Robey et al.(1993).
on the process. Both cell lines could contract collagen gels, and
anti-1 integrin antibodies could inhibit this contraction. We and
others (Takada et al., 1987; Heino and
Massagué, 1989; Dedhar and Saulnier, 1990;
Santala et al., 1994) have previously described the 1
integrin pattern in both cell lines. There are three putative
integrin-type collagen receptors, namely 11, 21,
and 31. In MG-63 cells 21 and 31 are
expressed, whereas no 11 is present. HOS cells express
distinct integrin pattern: 11 and 31 are present,
but 21 is missing. Both MG-63 cells and HOS cells express
51 fibronectin receptor, whereas 61 laminin
receptor is expressed only in HOS cells (Santala et al.,
1994). Previous reports have suggested the involvement of 1
integrins in the collagen gel contraction, especially by skin
fibroblasts (Klein et al., 1991; Shiro et al., 1991).
Here, we show that osteogenic cells can still induce contraction, even
if the 11 or 21 heterodimers are missing. Thus, the
data suggest that these heterodimers can replace each other or that
they both can be replaced by a third 1 integrin containing
heterodimer. TGF- could increase collagen gel contraction by MG-63
cells but not by HOS cells. In MG-63 cells TGF- strongly induces
the synthesis of 2 integrin and concomitantly decreases the
expression of 3 subunit, whereas it cannot turn on the expression
of 1 subunit (Heino and Massagué, 1989).
This suggests that the effect of TGF- is due to increased
expression of 2 integrin. Here, the increased contraction could be
inhibited totally by anti-1 integrin antibody and partially by
anti-2 integrin antibody, whereas anti-1 integrin antibody
had no effect. Previous studies have shown that anti- integrin
antibodies can not alone inhibit collagen gel contraction by
fibroblasts, but only enhance the effect of anti-1 antibody (Klein et al., 1991). Interestingly, anti-3 antibody constantly,
in three out of three experiments, stimulated contraction. The
mechanism of this phenomenon stays unknown. However, we have recently
shown that in keratinocytes 31 heterodimer is connected to a
signal transduction pathway regulating the expression of gelatinases
(Larjava et al., 1993a). The role of 2 subunit was also
suggested by the fact that in HOS cells TGF- cannot induce its
expression (Santala et al., 1994). We have previously shown
that transformation of HOS cells with MNNG induces the cells to express
2 integrin subunit (Santala et al., 1994). Here, we show
that TGF- increases both 2 integrin expression and collagen
gel contraction by HOS-MNNG cells. induces several other changes in cell metabolism
that could have explained the increased collagen gel contraction. Here,
we have excluded the involvement of TGF--induced increase in
fibronectin synthesis. More importantly, we have shown that increased
expression of 2 integrin subunit alone is sufficient to increase
contraction by both MG-63 and HOS cells. This was done by expressing
2 integrin cDNA in these cells. We have previously shown that both
cell lines express a large intracellular pool of excess precursor
1 integrin subunit (Heino and Massagué,
1989; Santala et al., 1994). Therefore, it is possible to get
functional integrin heterodimers by forced expression of an
subunit. Furthermore, transfection of MG-63 cells with
antisense-oriented 2 integrin cDNA generated cell clones with
reduced ability to contract gels., can regulate the synthesis of matrix molecules and
their cellular receptors connects the two mechanisms together.
Previously, some of the effects of TGF- on cell growth and
differentiation have been partially explained by the altered structure
of extracellular matrix in TGF--treated cell cultures (Heino and
Massagué, 1990; Nugent and Newman, 1989). The
data presented here show that TGF- regulates the cell behavior
also by altering their integrin pattern. We and others have shown, that
in addition to TGF- also other cytokines, including interleukin-1,
tumor necrosis factor-, and interferon-, can regulate
integrin expression in numerous tissue-cultured cell types (Santala and
Heino, 1991; Defilippi et al., 1991). In in vivo conditions, including wound healing (Larjava et al.,
1993b) and chronic inflammation (Nikkari et al., 1993), where
cytokines are present in large amounts, it is possible to detect
dramatic changes in integrin expression. The role of osteoblast
integrins in the healing of bone fractures is unknown. Recent studies
have shown, that osteoblastic cells in bone and in culture express
several integrin heterodimers. However,
21 is expressed only
in small quantities, if at all (Hughes et al., 1993; Brighton
and Albelda, 1992; Clover et al., 1992; Saito et al.,
1994). Our data suggest that TGF- might be involved in the
reorganization of collagenous matrix also by osteogenic cells; however,
only in the case that these cells already express 21
integrin. We have previously suggested the presence of an inhibitory
element, other than DNA methylation, in HOS cells, which prevents the
expression of 2 integrin gene (Santala et al., 1994).
Transformation of cells with both MNNG and Kirsten murine sarcoma virus
can induce the expression of 2 integrin in HOS cells (Santala et al., 1994), but we have not yet found a physiological
inducer. induces collagen
gel contraction by osteogenic cells. The process is due to increased
expression of 21 integrin-type collagen receptor. In cells
which do not express 2 subunit, TGF- can not turn on its gene
expression or induce collagen gel contraction. The data show that
TGF- can regulate cellular functions by altering integrin pattern.
Furthermore, we propose that TGF- might be one of the factors
involved in reorganization of bone matrix during the healing of
fractures.
)
We thank Dr. M. Hemler for 2 integrin cDNA, Dr.
A. Weiss for pAWneo2 expression vector, Dr. J.
Massagué for TGF-, and Drs. K. Yamada, E.
Wayner, and W. Rettig for antibodies. We are grateful to H. Helenius
and J. Tuominen for their help in statistical analysis. Expert
technical assistance of M. Potila is gratefully acknowledged.
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  H.-H. Chang, C.-B. Wu, Y.-J. Chen, C.-Y. Weng, W.-P. Wong, Y.-J. Chen, B.-E. Chang, M.-H. Chen, and C.-C.J. Yao MMP-3 Response to Compressive Forces in vitro and in vivo J. Dent. Res., July 1, 2008; 87(7): 692 - 696. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Dufour, X. Holy, and P. J. Marie Transforming growth factor-{beta} prevents osteoblast apoptosis induced by skeletal unloading via PI3K/Akt, Bcl-2, and phospho-Bad signaling Am J Physiol Endocrinol Metab, April 1, 2008; 294(4): E794 - E801. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z. Ferdous, V. M. Wei, R. Iozzo, M. Hook, and K. J. Grande-Allen Decorin-transforming Growth Factor- Interaction Regulates Matrix Organization and Mechanical Characteristics of Three-dimensional Collagen Matrices J. Biol. Chem., December 7, 2007; 282(49): 35887 - 35898. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. A. McBrien, R. Metlapally, A. I. Jobling, and A. Gentle Expression of Collagen-Binding Integrin Receptors in the Mammalian Sclera and Their Regulation during the Development of Myopia Invest. Ophthalmol. Vis. Sci., November 1, 2006; 47(11): 4674 - 4682. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Chen, R. Roberts, M. Pohl, S. Nigam, J. Kreidberg, Z. Wang, J. Heino, J. Ivaska, S. Coffa, R. C. Harris, et al. Differential expression of collagen- and laminin-binding integrins mediates ureteric bud and inner medullary collecting duct cell tubulogenesis Am J Physiol Renal Physiol, October 1, 2004; 287(4): F602 - F611. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Jokinen, E. Dadu, P. Nykvist, J. Kapyla, D. J. White, J. Ivaska, P. Vehvilainen, H. Reunanen, H. Larjava, L. Hakkinen, et al. Integrin-mediated Cell Adhesion to Type I Collagen Fibrils J. Biol. Chem., July 23, 2004; 279(30): 31956 - 31963. [Abstract] [Full Text] [PDF]
|
|
|
|
</ |
