Download Evidence for Functional Platelet-derived Growth Factor Receptors

[CANCER RESEARCH 44, 2966-2970,
July 1984]
Evidence for Functional Platelet-derived
MG-63 Human Osteosarcoma Cells1
Growth Factor Receptors on
Dana T. Graves,2 Harry N. Antoniades,3 Steven R. Williams, and Albert J. Owen
Department of Periodontology, Harvard School of Dental Medicine [D. T. G.], Department of Nutrition, Harvard School of Public Health [H. N. A , S R W AJO]
Center for Blood Research [H. N. A.], Boston, Massachusetts 02115
and
ABSTRACT
cells involves activation of sis-gene transcription resulting in the
The specific interaction of platelet-derived growth factor
(PDGF) with the human osteosarcoma cell line MG-63 was
studied. Scatchard analysis of 125I-PDGFbinding to MG-63 cells
indicated there were 32,000 specific PDGF-binding sites per cell
with a K<jof 2.4 x 10~11M. Unlabeled PDGF blocked the specific
production of biologically active PDGF. This process however,
may not be universal in the development of osteogenic sarcomas.
In this paper we present evidence that proliferation of MG-63
osteosarcoma cells does not involve endogenous secretion of
PDGF-like factors. It was found that MG-63 cells produced
mitogenic activity which was not PDGF-like, had membrane
binding of labeled PDGF to MG-63 cells at concentrations greater
than 1 ng/ml. When assayed for phosphorylation of MG-63
membrane vesicles, PDGF was shown to stimulate a dosedependent phosphorylation of a protein (phosphoprotein with a
molecular weight of 185,000) which was stable in 1 M NaOH. In
the absence of PDGF, a prominent alkali-stable phosphoprotein
with a molecular weight of 116,000 was noted. PDGF also
stimulated a dose-dependent increase in [3H]aminoisobutyric
acid uptake, [3H]thymidine incorporation, and cell proliferation.
When tested for secretion of PDGF-like factors, the mitogenic
activity of MG-63-conditioned
serum-free medium was not
blocked by anti-PDGF antiserum. Concentrated MG-63-conditioned medium did not compete with 125I-PDGF for specific
receptor sites on diploid fibroblasts. Therefore, MG-63 osteosar
coma cells have functional PDGF receptors and do not secrete
PDGF-like mitogens.
INTRODUCTION
It has previously been reported that the U-2 OS (formerly 2T)
osteosarcoma cell line secretes a PDGF4-like mitogen (11, 12),
and does not respond to exogenous PDGF (11). Nontransformed
mesenchymal cells require both PDGF and factors present in
PPP for growth (19), while transformed cells, such as the U-2
OS, may overcome the PDGF requirement. The U-2 OS osteo
sarcoma cells apparently satisfy their requirement for PDGF by
endogenous secretion, which is consistent with the autocrine
hypothesis (22). The importance of these findings was demon
strated by recent reports that the oncogene in the simian sar
coma virus, v-s/s, is capable of coding for PDGF, suggesting that
PDGF production plays an important role in the transformation
process (9, 23).
In collaboration with researchers at the California Institute of
Technology, we have recently established that mRNA from the
U-2 OS cells hybridizes with cDNA probes based on the v-s/s
probe.5 It appears, therefore, that transformation of the U-2 OS
' Supported by USPHS Grant HL27607, HL29583, and CA30101.
2 Recipient of Training Grant 5T 32-DE 07010.
3 To whom requests for reprints should be addressed.
4The abbreviations used are: PDGF, platelet-derived growth factor; PPP, plate
let-poor plasma; MEM, modified Eagle's medium; FBS, fetal bovine serum; CM-,
carboxymethyl-; DPBS, Dulbecco's phosphate-buffered saline; EBSS, Earie's bal
receptors to PDGF, and responded to exogenous sources of
PDGF. When assayed for phosphorylation of membrane pro
teins, a high degree of apparent tyrosine kinase activity was
noted without the addition of PDGF.
MATERIALS AND METHODS
Human osteosarcoma
cells (MG-63) and BALB/C-3T3
clone A31
mouse fibroblasts were purchased from the American Type Culture
Collection, Rockville, MD. The MG-63 osteosarcoma cell line was origi
nally isolated and characterized by Heremans et al. (14). Human fetal
fibroblasts (GM-10) were purchased from the Genetic Mutant Repository,
Camden, NJ. Human fibroblasts and osteosarcoma cells were grown in
MEM (Grand Island Biological Co., Grand Island, NY) supplemented with
extra glucose (0.09 g/liter) and pyruvate (0.616 g/liter). BALB/C-3T3 cells
were grown in Dulbecco's MEM (Biofluids, Inc., Rockville, MD). FBS was
purchased from Biofluids. [3H]Thymidine (16 Ci/mmol), carrier-free Na125l,
[3H]aminoisobutyric acid, and [-y-^PJATP were obtained from New Eng
land Nuclear, Boston MA. PDGF was prepared as previously described
(1). PDGF I and PDGF II (3000 units/^g protein) were isolated and
separately labeled with 125Iby lodo-Beads (Pierce Chemical Co., Rockford, IL). Labeled PDGF II was used in binding experiments. Partially
purified PDGF was used to study effects on biological activity (1000
units/Mg protein) and to compete with labeled PDGF in receptor assays
(500 un¡ts/¿Ã-g
protein). Human PPP and CM-Sephadex-treated PPP were
prepared as described in Ref. 11.
DNA Synthesis. BALB/C-3T3 fibroblasts were plated on microtiter
plates (Linbro) in MEM containing FBS (10%), and were allowed to
deplete medium for 8- to 10 days. Medium was then removed and
changed to assay medium containing PPP (2%), sample, and [3H]thymidine (5 MCi/ml). Cells were incubated 24 hr under appropriate conditions,
fixed with cold 5% trichloroacetic acid, dissolved in 1% SDS, and counted
on a Beckman LS-250 liquid scintillation counter. MG-63 cells were
plated on microtiter plates in MEM supplemented with FBS (10%).
Twenty-four hr later they were rinsed once with EBSS, transferred into
0.05% FBS, and were allowed to deplete medium for 3 days. Medium
was removed and assay medium containing MEM, PDGF, and [3H]thymidine (5 ^Ci/ml) was added. The cells were incubated, fixed, and
counted as above.
Collection and Assay of Conditioned Medium. Serum-free MEM (20
ml) was conditioned for 24 hr with confluent MG-63 cells in 150-sq cm
flasks (Coming Glass Works, Corning, NY) as previously described (11).
Biological activity was assayed by stimulation of [3H]thymidine incorpo
ration in BALB/C-3T3 cells after combining conditioned medium with 2
anced salt solution; SDS, sodium dodecyl sulfate; HSA, human serum albumin.
5 D. J. Graves, A. J. Owen, R. K. Barth, P. Jempst, A. Winoto, L. Fors, L. E.
volumes of fresh MEM supplemented with PPP (3%). To study the effect
of anti-PDGF antiserum on MG-63-conditioned medium, serum-free con
Hood, and H. N. Antoniades, manuscript in preparation.
Received October 24, 1983; accepted April 11, 1984
ditioned medium was collected and added to 2 volumes fresh MEM
supplemented with PPP (3%). Anti-PDGF antiserum or normal rabbit
2966
CANCER
RESEARCH
VOL. 44
Downloaded from cancerres.aacrjournals.org on April 20, 2017. © 1984 American Association for Cancer Research.
Functional PDGF Receptors on MG-63 Osteosarcoma Cells
serum was added and allowed to incubate overnight at 4°.Biological
activity was tested by stimulation of DMA synthesis in BALB/C-3T3 cells.
Serum-free conditioned MEM was concentrated by exhaustive dialysis
against 1 M acetic acid, lyophilization, and reconstitution in 0.9% NaCI
solution (saline) containing 1% HSA. This material was tested for stimu
lation of DNA synthesis in BALB/C-3T3 cells, and was assayed for
competition with 125I-PDGF receptor sites on human fibroblasts, as
described below.
Radioiodination
of PDGF. PDGF I and PDGF II were isolated as
described (1). For radioiodination, 5 to 10 /ig of PDGF I or PDGF II were
dissolved in 100 n\ of 0.1 M sodium phosphate buffer, pH 7.0, in a
borosilicate test tube (10 x 75 mm). One to 2 mCi of 125I(about 5 ¡i\in
0.1 M NaOH) were added, followed by the addition of a single lodo-Bead
(Pierce), a chloramine-T-derivatized
polystyrene bead. After 40 min, the
radioiodinated PDGF was removed and the test tube with the lodo-Bead
applied to SDS polyacrylamide gel electrophoresis (10%). In some cases,
gels were gently shaken in 1 M NaOH (1 hr, 50°).All gels were fixed and
stained with Coomassie brilliant blue. The gels were then dried, and
labeled proteins were detected by autoradiography.
Amino Acid Transport. Amino acid transport was measured, following
the method described by Owen ef al. (17). MG-63 cells were plated in
FBS (10%) on 24-well plates and were allowed to reach confluence.
Medium was changed to FBS (0.5%) and allowed to deplete for 72 hr.
Cells were rinsed and exposed to PDGF in MEM for 3 hr. This medium
was removed and cells were incubated for 1 hr with DPBS. Cells were
then changed to assay medium (0.5 mw [3H]aminoisobutyric acid, 8000
dpm/nmol: DPBS) and were allowed to incubate for 10 min. Cells were
then rinsed, solubilized in SDS (0.1%), and the [3H]aminoisobutyric acid
was counted.
Colony Growth. Cells (3.5 x 106) were trypsinized and resuspended
in 1 ml of FBS (10%). This suspension was diluted 1:1000 in CMwas rinsed twice with 200 n\ of buffer, and once with 200 n\ of 0.02%
Sephadex-treated PPP (0.5,1, and 2%), or in FBS (10%) to give 3.5 x
SDS. The radioiodinated PDGF and washes were dialyzed against 0.05%
103 cells/60-mm dish. Cells were allowed to grow for 2 weeks, with
SDS for 6 hr at room temperature, with frequent changes of dialysis
fluid. After dialysis, the solution (about 1 ml) was placed in an equal
medium changed after 7 days. Cells were rinsed with saline and were
volume of 10% HSA. Aliquots (100 n\) were kept frozen. The radioiodi
fixed with 0.1% méthylèneblue in methanol. Colonies were counted
nated PDGF I and PDGF II contained 15,000 to 20,000 cpm/ng protein.
manually.
Cell Proliferation. MG-63 cells suspended in FBS (10%) were plated
Over 80% of the radioactivity was immunoprecipitable with the PDGF
at 60,000 cells/well on 24-well plates, and were allowed to attach for 6
antisera. Radioiodinated PDGF I and PDGF II prepared with the proce
dure described above retained full biological activity, as judged by its
hr. Cells were then rinsed with EBSS and transferred into MEM supple
ability to stimulate incorporation of [3H]thymidine. For these assays, 125I- mented with PDGF. Medium was changed after 3 days, and cells were
PDGF was washed out prior to measuring [3H]thymidine uptake, and the
counted on Day 6 as previously described (11).
tritium channel on the scintillation counter was adjusted so that the small
amounts of 125Ipresent in the cells did not significantly interfere with the
RESULTS
measurement of [3H]thymidine uptake.
Binding of 125I-PDGF. Binding assays were performed, following mod
We previously reported that U-2 OS osteosarcoma cells prolif
ification of the protocol established by Bowen-Pope and Ross (4). MGerate in PPP in the absence of PDGF (11). The ability of MG-63
63 cells were plated in FBS (10%) on 24-well plates (Flow Laboratories,
cells to form colonies in PPP was tested (Table 1). Plating
efficiency in 2% PPP was equivalent to that in 10% FBS, and
decreased appreciably with lower concentrations of PPP. Colony
or concentrated conditioned medium was added to cells in cold binding
medium (DPBS: 1% HSA, 4°).Incubation was carried out for 3 hr at 4°, size was proportional to PPP concentration, indicating that MGMcLean, VA), grown to confluence, changed to depletion medium (0.5%
FBS), and incubated for 72 hr. Labeled 125I-PDGFwith unlabeled PDGF
either with the culture plates at rest, or while they were being gently
shaken. Binding medium was recovered and cells were rinsed 3 times
with DPBS: 0.1% HSA, followed by addition of SDS (1%). Triton X-100
was originally used to extract bound PDGF, but this yielded incomplete
recovery of cell-bound 125I-PDGF. 125I-PDGFwas extracted with 1% SDS
instead. Nonspecific binding was less than 10% of total PDGF binding,
under these conditions. 125I-PDGF was counted on a Packard Model
6230 gamma scintillation spectrometer.
Membrane Phosphorylation. MG-63 cells were grown to confluence
in FBS (10%), changed to depletion medium (0.5% FBS), and left for 48
hr. Cells were collected with harvest medium (0.1% EDTA, 0.2 mM
phenylmethylsulfonyl fluoride in EBSS) and then pelleted by centrifugation. All subsequent steps were done at 4°.Cells were resuspended in
homogenization buffer (25 mw sucrose: 4 mM Tris: 0.2 mw phenylmeth
ylsulfonyl fluoride, pH 8.4) and homogenized with a Dounce homogenizer.
The homogenate was centrifuged (4,000 x g, 10 min) and the superna
tant was collected. The pellet was resuspended, homogenized, and
centrifuged again. This supernatant was combined with the first super
natant and then centrifuged (30,000 x g for 60 min). The resulting
supernatant was discarded and the pellet was resuspended in 4 mw
Tris, pH 7.2, which was layered over 2 volumes of 35% sucrose: 4 mw
Tris buffer, pH 7.2, and centrifuged (100,000 x g for 60 min.). Plasma
membranes were then collected at the resulting interface. Phosphoryla
tion was carried out following a modification of methods previously
described (10). Briefly, 20 /ig of membrane protein, 30 n\ of cold 2x
incubation buffer [40 mw 4-(2-hydroxyethyl)-1-piperazineethanesulfonic
63 cells proliferate in low concentrations of PPP. One possible
explanation for the altered growth requirement of these cells
compared to nontransformed mesenchymal cells, which require
both PDGF and PPP for maximal growth (2), is endogenous
secretion of mitogens. Serum-free medium conditioned by MG63 cells was found to be mitogenic (Table 2A). Since the U-2 OS
osteosarcoma cell line secretes PDGF-like factors, MG-63-conditioned medium was tested for PDGF-like activity. Exposure to
rabbit anti-PDGF antiserum did not reduce mitogenicity of MG63-conditioned medium, indicating that PDGF is not a major
mitogen produced by these cells (Table 2A). Higher concentra
tions of anti-PDGF antiserum were also unable to reduce the
mitogenic effect of MG-63-conditioned medium (data not shown).
In contrast, anti-PDGF antiserum reduced the stimulatory effect
of PDGF.
DiCorleto and Bowen-Pope (8) reported that anti-PDGF antiserum is not capable of blocking mitogenic activity produced by
endothelial cells in culture, even though medium conditioned by
Table 1
Colony growth of MG-63 osteosarcoma cells
MG-63 cells were plated in MEM supplemented with PPP or FBS and were
allowed to grow as described in "Materials and Methods." Each value represents
the mean of duplicate dishes.
acid: 20 mM MgCI2: 0.2 mw MnCI2 (pH 7.4)], and PDGF were combined
and allowed to incubate for 30 min at 4°. The kinase reaction was
initiated by the addition of 0.6 nmol [T-^PJATP (6 to 10 Ci/mmol), and
was continued for 20 min at 4°.The reaction was stopped by addition
% PPP
of SDS sample buffer (4x) followed by boiling 3 min. Samples were then
10 (FBS)
0.5
1
2
No. of colonies/dish
Colony size (sq mm)
15
42
88
70
JULY 1984
Downloaded from cancerres.aacrjournals.org on April 20, 2017. © 1984 American Association for Cancer Research.
2967
D. T. Graves et al.
Table 2
Assay for PDGF-like activity in MG-63-conditioned
A. Stimulation of [3H)thymidine incorporation in
BALB/C-3T3 cells8
AdditionCM"
(33%)
CM (33%) + AS9 (1.5^1)
médium
B. Competition with 12SIPDGF for PDGF-binding
sites on human fibrcblasts"
samples cpm back
ground038,927
CCM' (100%)
242
CCM (25%)
234236
71,628
CCM
(7%)
44,519
PDGF (5 ng/ml)
CCM
(2%)
16,854AdditionBuffer
233
PDGF (5 ng/ml) + AS (1 .5 ^l)cpm
PDGF (350 ng)ricpm230
65
* Stimulation of acid-insoluble [3H]thymidine incorporation in BALB/c-313 cells
is described in "Materials and Methods."
6 Competition for PDGF-binding sites on GM-10 fibroblasts was carried out as
described in "Materials and Methods."
c Each value represents the mean of triplicate wells. S.E. < 0.15 of the mean.
" Each value represents the mean of duplicate wells.
8 MG-63-conditioned medium is described in "Materials and Methods."
' Concentrated MG-63-conditioned medium is described in "Materials and Meth
ods."
9 AS, anti-PDGF antisera, as described in "Materials and Methods."
in 1 M NaOH (Fig. 1, B and D), suggesting that tyrosine is the
amino acid residue phosphorylated (5,15). However, the phos
phorylation of MG-63 membrane proteins did differ significantly
from phosphorylation in membrane proteins from human diploid
fibroblasts. A prominent alkali-stable phosphoprotein with M, of
116,000 was evident in MG-63 membranes in the absence of
PDGF(Fig. 1S). In the human diploid fibroblasts, prominent alkalistable phosphoproteins were present only after the addition of
PDGF (Fig. 10).
In other measures of biological activity, PDGF induced similar
responses seen in nontransformed cells (2). A dose-dependent
increase in [3H]aminoisobutyric acid uptake was observed, as
was incorporation of [3H]thymidine and cell proliferation in re
sponse to PDGF (Chart 3).
DISCUSSION
In experiments presented here, MG-63 osteosarcoma cells did
not produce PDGF-like mitogens, in contrast to previous findings
with the U-2 OS osteosarcoma cell line. MG-63 cells produced
mitogenic activity that was not neutralized by anti-PDGF antiserum, nor was capable of competing with 125I-PDGF.Further-
these cells can compete with PDGF for membrane receptor
sites. Therefore, MG-63-conditioned medium was tested for
ability to compete with 125I-PDGFfor receptor sites on human
1.6-,
1.4-
diploid fibroblasts (Table 2B). Conditioned medium was exhaus
tively dialyzed against 1 M acetic acid, lyophilized and reconsti
tuted with HSA (1%) in EBSS. A1:20 dilution of this concentrated
material stimulated acid-insoluble [3H]thymidine incorporation
£ 1.2u.
1 1.0-
into 3T3 cells (data not shown). The same material could not
block 125I-PDGFbinding over the concentration range tested. In
o
CD 0.8-
contrast, unlabeled PDGF significantly reduced specific binding
of labeled PDGF. Thus, MG-63 cells do not secrete PDGF-like
factors, as measured by competition with labeled PDGF for
receptor sites.
Cells of mesenchymal origin have receptors to PDGF (4, 24).
Since MG-63 cells do not produce PDGF-like factors and are of
mesenchymal origin, it was expected that they would have
specific PDGF-binding sites. This was confirmed in binding ex
periments with 125I-PDGF.In the experiment described in Chart
1, the specific activity of 125I-PDGFwas held constant. Scatchard
analysis indicated there were 32,000 PDGF receptors/MG-63
cell, with a dissociation constant of 2.4 x 10~11M. In the
§ 0.6a.
Receptors/Cell = 32,000
0.23.5
7.0
10.5
PDGF Bound (fmol)
2968
14.0
Chart 1. Scatchard analysis of PDGF binding to MG-63 cells. 1AI-PDGF (0.2 to
3.5 ng/ml; 6 x 10"4 ng/cpm) was added to each well, as described in "Materials
and Methods." Points, mean of duplicate wells. The slope and intercepts were
determined by least-squares analysis.
experiment described in Chart 2, the concentration of labeled
PDGF (1.5 ng/ml) was held constant, and various amounts of
unlabeled PDGF (0.05 ng/ml to 0.9 /tg/ml) were added. Displace
ment of labeled PDGF occurred when greater than 1.0 ng of
unlabeled PDGF/ml was added. The addition of 0.9 /¿g
of unla
beled PDGF/ml resulted in virtually no binding of 125I-PDGF,
indicating that the nonspecific binding of labeled PDGF is low in
our system, probably due to the purity of the labeled PDGF.
Scatchard analysis of these data gave a receptor number and
«oin close agreement with the experiment described in Chart 1
(not shown).
Among the earliest known events modulated by PDGF is the
phosphorylation of membrane proteins. PDGF was able to en
hance a dose-dependent phosphorylation of a M, 185,000 protein
in plasma membrane vesicles prepared from MG-63 cells (Fig. 1,
A and B). PDGF induced phosphorylation of a membrane protein
in human diploid fibroblasts with a similar molecular weight (Fig.
1, C and D). The phosphoprotein from both cell types was stable
kd = 2.4«10'11M
2.42.0O
5
0.
1.6-
0.80.4-
0.5
5.0
50
Unlabeled PDGF (ng/ml)
500
Chart 2. Competition of unlabeled PDGF with labeled PDGF for specific binding
sites on MG-63 cells. 12SI-PDGF(1.5 ng/ml; 1.6 x 1Q-1 ng/cpm) with unlabeled
PDGF (0, 0.25, 0.9, 3.7, 15, 60, 235, and 900 ng/ml) were added to each well, as
described in "Materials and Methods." Pointu, mean of duplicate wells.
CANCER
RESEARCH
Downloaded from cancerres.aacrjournals.org on April 20, 2017. © 1984 American Association for Cancer Research.
VOL.
44
Functional PDGF Receptors on MG-63 Osteosarcoma Cells
PHOSPHORYLATION OF
• AMWO ACID TRANSPORT
2.4 w
if
0
yr
-2.2°
-2.0 I
o
-1.8 ¿
O
CELL PROLIFERATION
•1.3
-1.1
î
0.9 Z
I
-0.7 r
o
oÃ0.5
0-
1000 2
5 10 20
125
5
10 20
PDGF (ng/ml)
PDGF (ng/ml)
Chart 3. PDGF-induced biological activity in MG-63 osteosarcoma cells. Phosphorylation, amino acid transport, DNA synthesis, and cell proliferation were
measured as described in "Materials and Methods." For DNA synthesis, points
02
represent the mean of triplicate wells; oars, S.E. For phosphorylation, amino acid
transport, and cell proliferation, points represent the mean of duplicate determina
tions, pp 185k, phosphoprotein with a molecular weight of 185,000; AIB, aminoisobutyric acid.
more, in collaboration with researchers at the California Institute
of Technology, we have recently found that polyadenylic acidcontaining mRNA prepared from MG-63 cells does not contain
sequences capable of hybridizing with v-s/s probes, substanti
ating that MG-63 cells do not synthesize PDGF-like polypeptides.5 It has been suggested that PDGF synthesis or independ
ence of exogenous PDGF may play an important role in the
transformation of mesenchymal cells (7,11, 20, 21 ). The recent
finding that the oncogene of an acute transforming retrovirus
code for a PDGF-like protein (9, 23) supports this contention.
However, our findings with the MG-63 osteosarcoma cell line
indicate that secretion of PDGF-like factors is not a universal
feature of osteosarcoma-derived
cells and hence, is not an
absolute requirement for transformation of osteogenic cells.
The binding of labeled PDGF to MG-63 cells is similar to the
binding of PDGF to nontransformed cells in published accounts.
Scatchard analysis indicated that MG-63 cells had 32,000 PDGFbinding sites per cell, with a dissociation constant of 2.4 x 10~11
M. Williams ef al. (24) reported 50,000 PDGF receptors with a
dissociation constant of 10~11 M on vascular smooth muscle
cells. Bowen-Pope and Ross (4) reported a dissociation constant
of 10~11M, with an average of 4 x 10" receptors/cell for human
fibroblasts and arterial smooth muscle cells.
The number of PDGF receptors on MG-63 cells similar to
those found on other cell types is consistent with lack of PDGF
secretion noted above. If MG-63 cells secreted PDGF-like fac
tors, it would be expected that the number of PDGF receptor
sites would be significantly lower than that observed (18, 22).
The presence of specific PDGF-binding sites on a human osteo
sarcoma cell line was previously reported by Heldin ef al. (13).
They found that the U-393 OS cell line had PDGF receptors,
while the U-2 OS human osteosarcoma cell line did not.
The ability of PDGF to stimulate biological activity in MG-63
cells was tested. PDGF induces a dose-dependent increase in
amino acid transport, DNA synthesis, and proliferation. Binding
of PDGF to its receptor elicits the same events associated with
PDGF-receptor interaction in nontransformed cells (2). Although
PDGF may be capable of inducing biological activity in MG-63
cells in the conditions tested, under appropriate conditions, the
need for PDGF might be overcome by the production of other
growth factors. This is supported by evidence that MG-63 cells
secrete mitogens and grow, although not maximally, in PPP
treated with CM-Sephadex to remove residual amounts of PDGF.
Diploid fibroblasts do not proliferate in low concentrations of
CM-Sephadex-treated PPP (11, 19). It was also noted that the
addition of anti-PDGF antiserum to MG-63-conditioned medium
resulted in greater mitogenic activity than was found in condi
tioned medium alone. This is probably due to synergy between
factors secreted by MG-63 cells and factors present in rabbit
serum.
PDGF has been shown to enhance phosphorylation of tyrosine
in membrane proteins (10, 16). Of particular interest is the
phosphorylation of a M, 165,000 to 185,000 membrane protein
which is thought to be the PDGF receptor. The effect of PDGF
on the phosphorylation of MG-63 membranes was tested. A
dose-dependent increase in M, 185,000 phosphoprotein oc
curred in MG-63 membranes, which was similar to the enhanced
M, 185,000 phosphoprotein levels observed in membranes pre
pared from human diploid fibroblasts. In order to distinguish the
phosphorylation of tyrosine from serine or threonine, phosphorylated proteins were tested for alkali stability (5,15). When alkali
stability was determined, a striking difference was noted between
the diploid fibroblasts and the MG-63 cells. In fibroblast mem
branes, the only alkali-stable phosphoprotein occurred in the
presence of PDGF and had a molecular weight of 185,000. In
the MG-63 cell membranes, an alkali-stable M, 116,000 phos
phoprotein was observed without the addition of PDGF. This
suggests that there is a high level of endogenous tyrosine kinase
activity in MG-63 cells. These findings may reflect intrinsic differ
ences between transformed and nontransformed cells. It has
been noted that some of the acute transforming retroviruses
have oncogene products with membrane-associated tyrosine
kinase activity (3). Future investigation will attempt to determine
if a retroviral oncogene-related mRNA, capable of coding for a
tyrosine kinase, is transcribed in these cells. Since peptide
growth factors induce tyrosine phosphorylation (2, 6), it is also
possible that MG-63 cells secrete a factor unrelated to PDGF
which subsequently binds to a membrane receptor and stimu
lates tyrosine phosphorylation in the M, 116,000 membrane
protein. Still another possibility is that membrane kinase activity
in the MG-63 osteosarcoma cells is not closely linked to receptor
occupancy, resulting in high levels of M, 116,000 phosphopro
tein. Although the possibilities are intriguing, there is no concrete
evidence that tyrosine phosphorylation is required for cell growth.
ACKNOWLEDGMENTS
The authors would like to thank Ronald Siraco, Kaoru Momomura, Kathleen
Burke, and Anna Antoniades for expert technical assistance, and Jeanne MacLaren,
Christine Kelter-McGandy, and Gertrude Easterly for help in preparing this manu
script.
JULY 1984
Downloaded from cancerres.aacrjournals.org on April 20, 2017. © 1984 American Association for Cancer Research.
2969
D. T. Graves et al.
REFERENCES
1. Antoniades, H. N. Human platelet-derived growth factor (PDGF): purification
of PDGF (I) and PDGF (II) and separation of their reduced subunits. Proc. Nati.
Acad. Sci. USA, 78: 7314-7317, 1981.
2. Antoniades, H. N., and Owen, A. J. Growth factors and regulation of cell
growth. Annu. Rev. Med., 33. 445-446, 1982.
3. Bishop, J. M. Cellular oncogenes and retroviruses. Annu. Rev. Brachem., 52.
301-354, 1983.
4. Bowen-Pope, D. F., and Ross, R. Platelet-derived growth factor: specific
binding to cultured cells. J. Biol. Chem., 257: 5161-5171,1982.
5. Cheng, Y-S. E., and Chen, L. B. Detection of phosphotyrosine-containing
34,000 dalton protein in the framework of cells transformed with Rous sarcoma
virus. Proc. Nati. Acad. Sci. USA, 78: 2388-2392, 1981.
6. Cohen, S., Carpenter, G., and King, L. Epidermal growth factor-receptorprotein kinase interactions. J. Biol. Chem., 255. 4834-4842, 1980.
7. Curie, G. A. Platelet-derived growth factor requirements for in vitro proliferation
of normal and malignant mesenchymal cells. Br. J. Cancer, 43: 336-343,1981.
8. DiCorleto, P. E., and Bowen-Pope, D. F. Cultured endothelial cells produce a
platelet-derived growth factor-like protein. Proc. Nati. Acad. Sci. USA. 80:
1919-1923, 1983.
9. Doolittle. R. F., Hunkapiller, M. W., Hood, L. E., Devare, S. G., Robbins, K. C.,
Aaronson, S. A., and Antoniades, H. N. Simian sarcoma oncogene, v-sis, is
derived from the gene (or genes) encoding a platelet-derived growth factor.
Science (Wash. DC), 22J: 275-277, 1983.
10. Ek, B., Westermark, B., Wasteson, A., and Heldin, C. H. Stimulation of tyrosinespecific phosphorylation by platelet-derived growth factor. Nature (Lond.), 295:
419-420, 1982.
11. Graves, D. T., Owen, A. J., and Antoniades, H. N. Evidence that a human
osteosarcoma cell line which secretes a mitogen similar to platelet-derived
growth factor requires growth factors present in platelet-poor plasma. Cancer
Res., 43: 83-87, 1983.
12. Heldin. C. H., Westermark, B., and Wasteson, A. Chemical and biological
properties of a growth factor from human cultured osteosarcoma cells, resem
blance with PDGF. J. Cell. Physiol., 705: 235-246,1980.
13. Heldin, C. H., Westermark. B.. and Wasteson, A. Specific receptors for platelet-
14.
15.
16.
17.
18
19
20.
21.
22.
23.
24.
derived growth factor on cells derived from connective tissue and glia. Proc.
Nati. Acad. Sci. USA, 78: 3665-3668. 1981.
Heremans, H., Billiau, A., Cassiman, J. J., Muller, J. C., and DeSomer, P. In
vitro cultivation of human tumor tissues: morphological and virological char
acterization of three cell lines. Oncology, 35: 246-252, 1980.
Hunter, T., and Sefton, B. Transforming gene product of Rous sarcoma virus
phosphorylates tyrosine. Proc. Nati. Acad. Sci. USA, 77: 1311-1315, 1980.
Nishimura, J., Huang, J. S.. and Deuel, T. F. Platelet-derived growth factor
stimulates tyrosine specific protein kinase activity in Swiss mouse 3T3 cell
membranes. Proc. Nati. Acad. Sci. USA. 79: 4303-4307, 1982.
Owen, A. J., Geyer, R. P., and Antoniades, H. N. Human platelet-derived
growth factor stimulates amino acid transport and protein synthesis by human
diploid fibroblasts in plasma-free media. Proc. Nati. Acad. Sci. USA. 79: 32033207,1982.
Ozanne, B., Fulton, J. R., and Kaplan, P. Kirsten murine sarcoma virus
transformed cell lines and a spontaneously transformed rat cell line produce
transforming factors. J. Cell. Physiol., 705: 153-180, 1980.
Pledger, W. J.. Stiles, C. D., Antoniades, H. N., and Scher. C. D. Induction of
DNA synthesis in Balb/c-3T3 cells by serum components: réévaluation
of the
commitment process. Proc. Nati. Acad. Sci. USA, 74: 4481-4485,1977.
Scher, C. D., Pledger, W. J., Martin, P.. Antoniades, H. N.. and Stiles, C. D.
Transforming viruses directly reduce the cellular growth requirements of a
platelet-derived growth factor. J. Cell. Physiol.. 97: 371-380, 1978.
Singh, J. P., Gudas, L. J., and Stiles, C. D. Production of platelet-derived
growth factor-like activity in association with the transformed state. Fed. Proc.,
42: 1829 (Abstract), 1983.
Sporn, M. D., and Todaro, G. J. Autocrine secretion and malignant transfor
mation of cells. N. Engl. J. Med.. 303: 878-880, 1980.
Waterfield, M. D., Scrace, G. T., Whittle, N., Stroobant. P., Johnsson, A.,
Wasteson, A., Westermark, B., Heldin, C. H., Huang, J. S., and Deuel, T. F.
Platelet-derived growth factor is structurally related to the putative transform
ing protein p28"" of simian sarcoma virus. Nature (Lond.), 304: 35-39, 1983.
Williams, L. T., Tremble, P., and Antoniades, H. N. Platelet-derived growth
factor binds specifically to receptors on vascular smooth muscle cells and the
binding becomes nondissociable. Proc. Nati. Acad. Sci. USA, 79: 5867-5870,
1982.
B
,-200
til—
200-,
116
-116
-92
92-
-68
68
45
-45
i-
IO
CM
OÃ-
PDGF (ng)
t-
IO
O 24 O 24
PDGF (ng/ml)
Fig. 1. PDGF-modulated phosphorylation of MG-63 human osteosarcoma and GM-10 human fibroblast cell membranes. Membranes were prepared and phosphorylated
as described in "Materials and Methods." A, phosphorylation of MG-63 membranes exposed to PDGF (0, 1, and 25 ng); 8, phosphorylation of MG-63 membranes
exposed to PDGF (0 to 50 ng). and subsequent alkali treatment of SDS polyacrylamide gel; C, phosphorylation of GM-10 membranes exposed to PDGF (0 and 24 ng/
ml); D, phosphorylation of GM-10 membranes exposed to PDGF (0 and 24 ng/ml), and subsequent alkali treatment of SDS polyacrylamide gel.
2970
CANCER
RESEARCH
Downloaded from cancerres.aacrjournals.org on April 20, 2017. © 1984 American Association for Cancer Research.
VOL. 44
Evidence for Functional Platelet-derived Growth Factor
Receptors on MG-63 Human Osteosarcoma Cells
Dana T. Graves, Harry N. Antoniades, Steven R. Williams, et al.
Cancer Res 1984;44:2966-2970.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/44/7/2966
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on April 20, 2017. © 1984 American Association for Cancer Research.