(Received for publication, May 28, 1997)
From the Department of Biochemistry and Molecular Biology, St. Louis University School of Medicine, St. Louis, Missouri 63104
Insulin-like growth factor-binding protein
3 (IGFBP-3) has been shown to inhibit cell growth by
IGF-dependent and -independent mechanisms. The
putative cell-surface IGFBP-3 receptor that mediates the
IGF-independent growth inhibition has not been identified. Here we show
that recombinant human IGFBP-3 inhibits
125I-transforming growth factor
(TGF)-1 binding to the type V TGF-
receptor
(Mr 400,000) in mink lung epithelial
cells. We also demonstrate that the ~400-kDa
125I-IGFBP-3 affinity-labeled putative IGFBP-3
receptor is immunoprecipitated by specific antiserum to the type V
TGF-
receptor. The 125I-IGFBP-3 affinity labeling of the
putative receptor and IGFBP-3-induced growth inhibition as measured by
DNA synthesis in these cells is blocked by a TGF-
1
peptide antagonist. The 125I-IGFBP-3 affinity-labeled
putative receptor can only be detected in cells expressing the type V
TGF-
receptor, but not in cells lacking the type V TGF-
receptor.
These results indicate that the type V TGF-
receptor is the putative
IGFBP-3 receptor and that IGFBP-3 is a functional ligand for the
type V TGF-
receptor.
The type V transforming growth factor (TGF-
)1 receptor is a
400-kDa non-proteoglycan membrane glycoprotein that co-expresses with
the type I, type II, and type III TGF-
receptors in most cell types
(1-5). The type V TGF-
receptor as well as the type I and type II
TGF-
receptors are Ser/Thr-specific protein kinases and belong to
the new class of membrane receptors associated with a Ser/Thr-specific
protein kinase activity (1-6). The type I and type II TGF-
receptors have been shown to be important in TGF-
-induced cellular
responses (1-6), but the role of the type V TGF-
receptor in these
responses has not been defined (3-5). Recently, we have demonstrated
that the type V TGF-
receptor mediates TGF-
-induced growth
inhibition and that both type I and type II TGF-
receptors are
required for mediating maximal growth inhibition (7).
Insulin-like growth factor-binding protein 3 (IGFBP-3) is the most abundant insulin-like growth factor-binding protein in the circulation (8-11). In human plasma, IGFBP-3 forms an ~140-kDa ternary complex with IGFs and an acid-labile subunit (12). This complex serves as a reservoir for IGFs (12). IGFBP-3 is produced by a variety of cell types (12) and appears to inhibit cell growth by IGF-dependent and -independent mechanisms (13-15). Although several small cell membrane-associated IGFBP-3 binding proteins have recently been reported (16-18), the putative IGFBP-3 receptor that mediates the IGF-independent growth inhibition has not been identified.
IGFBP-3 has been implicated as a mediator of the actions of TGF-,
retinoic acid, and the tumor suppressor gene p53 (19-21). Since the
type V TGF-
receptor appears to play an important role in
TGF-
-induced growth inhibition (7), we tested the hypothesis that
IGF-independent actions of IGFBP-3 are mediated by the type V TGF-
receptor. In this communication, we demonstrate that IGFBP-3 inhibits
125I-labeled TGF-
1
(125I-TGF-
1) binding to the type V TGF-
receptor in mink lung epithelial cells. We also show that
125I-labeled IGFBP-3 (125I-IGFBP-3)
affinity-labeled putative cell-surface IGFBP-3 receptor is
immunoprecipitated by specific antiserum to the type V TGF-
receptor
and that the 125I-IGFBP-3-putative receptor complex is
detected only in cells expressing the type V TGF-
receptor. Finally,
we show that 125I-IGFBP-3 affinity labeling of the putative
IGFBP-3 receptor and IGFBP-3-induced growth inhibition can be blocked
by a TGF-
1 peptide antagonist.
Na125I (17 Ci/mg) and
[methyl-3H]Thymidine (67 Ci/mmol) were
purchased from ICN Biochemicals Inc. (Costa Mesa, CA). High molecular mass protein standards (myosin, 205 kDa; -galactosidase, 116 kDa;
phosphorylase, 97 kDa; bovine serum albumin, 66 kDa) and other chemical
reagents were obtained from Sigma. Disuccinimidyl suberate (DSS) was
obtained from Pierce. TGF-
1 was purchased from Austral
Biologicals (San Ramon, CA). Recombinant nonglycosylated human IGFBP-3
(expressed in Escherichia coli) was provided by Celtrix
Pharmaceutical Inc. (Santa Clara, CA).
125I-TGF-
1 and 125I-IGFBP-3 were
prepared as described previously (3, 5, 26) except 0.2 M
sodium phosphate buffer, pH 7.4, was used as the solvent for Sephadex
G-25 column chromatography to separate 125I-IGFBP-3 from
free 125I. The specific radioactivity of
125I-TGF-
1 and 125I-IGFBP-3 was
1-4 × 105 cpm/ng. The antigen used to prepare
specific rabbit antiserum to the type V TGF-
receptor was
thyroglobulin-conjugated to a hexadecapeptide whose amino acid sequence
was derived from the partial amino acid sequence of bovine type V
TGF-
receptor (7). The antiserum specifically reacted with the type
V TGF-
receptor from different species, including mink, rat, mouse,
cow, and human (7). This antiserum did not react with the type I, type
II, and type III TGF-
receptors on Western blot analysis and in
immunoprecipitation (7). TGF-
1 and TGF-
3
peptide antagonists were synthetic pentacosapeptides whose amino acid
sequences were derived from those of TGF-
1 and TGF-
3,
respectively.2 The
IC50 values of TGF-
1 and
TGF-
3 peptide antagonists for inhibiting
125I-TGF-
1 (0.1 nM) binding to
TGF-
receptors in mink lung epithelial cells are ~1-2 and
~20-30 µM, respectively.2 Human colorectal
carcinoma cells transfected with neo vector only and with
vector expressing type II TGF-
receptor cDNA (HCT-116 and RII-37
cells) were provided by Dr. Michael G. Brattain. (Department of
Biochemistry and Molecular Biology, Medical College of Ohio, Toledo,
OH)The type I and type II TGF-
receptor-defective mutant mink lung
epithelial cells (R1-B and DR 26 cells) were provided by Dr. Joan
Massagué (Sloan-Kettering Cancer Center, New York). Wild-type and
mutant mink lung epithelial cells and other cell types were maintained
in Dulbecco's modified Eagle medium containing 10% fetal calf
serum.
The 125I-TGF- binding and affinity labeling
were carried out as described previously (3, 26). The specific binding
of 125I-TGF-
1 was calculated by subtracting
the total binding from the nonspecific binding obtained in the presence
of 100-fold excess of unlabeled TGF-
1 or 10 µM TGF-
1 peptide antagonist. The
125I-TGF-
1 affinity labeling of cell-surface
TGF-
receptors was carried out using DSS as the cross-linking agent
(3, 26).
Cells grown on 60-mm Petri dishes were incubated with 5 nM 125I-IGFBP-3 (specific radioactivity:
~1 × 105 cpm/ng) in the presence of 100-fold excess
of unlabeled IGFBP-3 or inhibitors, including heparin, IGF-I,
TGF-1, and TGF-
3 peptide antagonists.
After 125I-IGFBP-3 affinity labeling in the presence of DSS
(3, 26), the 125I-IGFBP-3-putative receptor complex was
analyzed by 5% SDS-polyacrylamide gel electrophoresis under reducing
conditions and autoradiography.
After 125I-IGFBP-3 affinity labeling, the
cells were detached and lysed in 100 µl of 1% Triton X-100 in 10 mM Tris-HCl, pH 7.0, 125 mM NaCl, and 1 mM EDTA. After centrifugation, the Triton X-100 extracts
were then diluted 10-fold with Triton X-100-free buffer and incubated
with antiserum or non-immune serum (1:100 dilution) at 4 °C
overnight. The immunocomplexes were precipitated with 20 µl of
protein A-Sepharose (50%, v/v). After washing with 20 mM
Tris-HCl, pH 7.4, 0.2% Triton X-100, the immunoprecipitates were
analyzed by 5% SDS-polyacrylamide gel electrophoresis under reducing
conditions and autoradiography. The relative intensity of
125I-IGFBP-3-type V TGF- receptor complex on the
autoradiogram was quantitated by a PhosphorImager.
Cells were plated on 24-well clustered dishes at near
confluence and incubated with various concentrations of IGFBP-3 or 10 pM of TGF-1 ± 10 µM
TGF-
1 peptide antagonist in Dulbecco's modified Eagle
medium containing 0.1% fetal calf serum. After incubation at 37 °C
for 16 h, the cells were pulse-labeled with 1 µCi/ml of
[methyl-3H]thymidine at 37 °C for 4 h.
The [methyl-3H]thymidine incorporation into
cellular DNA was determined by a liquid scintillation counter. For RNA
analysis, cells grown on 12-well cluster dishes were treated with
various concentrations (0, 0.2, 0.4, 0.8, and 5 µg/ml) of IGFBP-3 or
with 0.1 nM TGF-
(as a positive control) for 2.5 h
at 37 °C in 0.1% fetal calf serum. Total cellular RNA was extracted
with RNAzol B (Tel-Test, Inc.) according to the manufacturer's
protocol. RNA was electrophoresed in 1.2% formaldehyde-agarose gel and
transferred to Duralon UV membrane using 10 × SCC. The Northern
blot was probed at 42 °C with a random-primed, radiolabeled
1-kilobase fragment of HindIII and NcoI digests
of plasminogen activator inhibitor 1 (PAI-1) cDNA. The blots were
washed with 0.1 × SCC containing 0.1% SDS at room
temperature.
TGF- is the most potent known polypeptide growth inhibitor for
epithelial cells and other cell types (23-25). Our recent studies have
indicated that the type V TGF-
receptor, a 400-kDa membrane glycoprotein which co-expresses with the type I, type II, and type III
TGF-
receptors in most cell types (2-4, 26), plays an important
role in mediating TGF-
-induced growth inhibition in mink lung
epithelial cells (7). To see if the IGF-independent growth inhibitory
action of IGFBP-3 is mediated by the type V TGF-
receptor or other
TGF-
receptor types, we investigated the effect of IGFBP-3 on the
binding of 125I-TGF-
1 to mink lung
epithelial cells, for which IGFBP-3 is also a growth inhibitor. As
shown in Fig. 1A, IGFBP-3
inhibited the specific binding of 125I-TGF-
1
in a concentration-dependent manner. At 16 µg/ml (~500 nM) or higher of IGFBP-3, a maximal ~50% inhibition was
observed. This partial inhibition implies that IGFBP-3 competes with
125I-TGF-
1 for binding to specific TGF-
receptor types. To identify which TGF-
receptor types are
responsible for IGFBP-3 binding, we performed
125I-TGF-
1 affinity labeling of cell-surface
TGF-
receptors after incubation of the cells with
125I-TGF-
1 in the presence of 16 µg/ml
unlabeled IGFBP-3 or 10 µM of TGF-
1
peptide antagonist. The TGF-
1 peptide antagonist is a
synthetic pentacosapeptide whose amino acid sequence was derived from
that of TGF-
1.2 As shown in Fig.
1B, the type I, type II, type III, and type V TGF-
receptors were all affinity-labeled with
125I-TGF-
1 in the presence of the
cross-linking agent DSS (lane 2). Unlabeled IGFBP-3 (~500
nM) appeared to completely block
125I-TGF-
1 affinity labeling of the type V
TGF-
receptor, and to a much lesser extent (30-40% inhibition),
the type III TGF-
receptor (Fig. 1B, lane 3). In the
control experiment, the TGF-
1 peptide antagonist
completely blocked 125I-TGF-
1 affinity
labeling of all TGF-
receptor types (Fig. 1B, lane 1).
These results suggest that IGFBP-3 strongly competes with
125I-TGF-
1 for binding to the type V TGF-
receptor.
To further confirm that IGFBP-3 binds to the type V TGF- receptor
with high affinity or that the type V TGF-
receptor is the putative
receptor for IGFBP-3, we performed the binding and cross-linking of
125I-labeled recombinant nonglycosylated human IGFBP-3
(125I-IGFBP-3, 5 nM) to its putative
cell-surface receptor, followed by immunoprecipitation with specific
antiserum to the type V TGF-
receptor (7). At 5 nM,
125I-IGFBP-3 was found to bind to the type V TGF-
receptor but not other TGF-
receptor types. As shown in Fig.
2A, 125I-IGFBP-3
was cross-linked to an ~400-kDa putative receptor on the cell surface
of mink lung epithelial cells (lane 1). This 125I-IGFBP-3 binding and subsequent cross-linking was
blocked by 100-fold excess of unlabeled IGFBP-3 or 10 µM
TGF-
1 peptide antagonist but not by 10 µM
TGF-
3 peptide antagonist (Fig. 2A, lanes 2, 3, and 4, respectively). The TGF-
3
peptide antagonist, a pentacosapeptide whose amino acid sequence was
derived from TGF-
3, has a lower affinity to the type V
TGF-
receptor.2 The antiserum to the type V TGF-
receptor specifically immunoprecipitated the ~400-kDa
125I-IGFBP-3-putative receptor complex (Fig. 2A,
lanes 5 and 7). Two, ~70-kDa and ~64 kDa,
125I-IGFBP-3 complexes were also found in the cell lysates
and in the immunoprecipitates (Fig. 2A, lanes 1, 4, 5, and
7). Since the preparation of 125I-IGFBP-3
(apparent Mr ~35,000 on SDS-polyacrylamide gel
electrophoresis) used in the experiments was found to contain
proteolytic products (apparent Mr
32,000),
and since 125I-IGFBP-3 has been shown to form a dimer in
solution (17),3 these
125IGFBP-3 complexes may be cross-linked dimers of
125I-IGFBP-3 and its proteolytic products. In the control
experiments, no 125I-IGFBP-3-putative receptor complex was
found in the immunoprecipitates when the cells were incubated with
125I-IGFBP-3 in the presence of 10 µM
TGF-
1 peptide antagonist prior to cross-linking and
immunoprecipitation (Fig. 2A, lane 6). Non-immune serum did
not immunoprecipitate the 125I-IGFBP-3-putative receptor
complex (Fig. 2A, lane 8). These results suggest that
125I-IGFBP-3 specifically binds to the type V TGF-
receptor in mink lung epithelial cells. To further characterize the
binding of 125I-IGFBP-3 to the type V TGF-
receptor, we
determined the specific binding of various concentrations of
125I-IGFBP-3 to the type V TGF-
receptor in mink lung
epithelial cells. As shown in Fig. 2B, 125 I-IGFBP-3 bound to the type V TGF-
receptor in these cells in a
concentration dependent manner (lanes 1-5). The
Scatchard plot analysis of the binding revealed that the apparent
Kd for 125I-IGFBP-3 binding to the type
V TGF-
receptor was 6 ± 2 nM (data not shown).
Since IGFBP-3 is known to bind IGFs with high affinity, and since it
contains a heparin-binding site near its C-terminal end (8-11), we
determined the effects of the IGF-I complex and heparin on the binding
of 125I-IGFBP-3 to type V TGF-
receptor in mink lung
epithelial cells. As shown in Fig. 2C, at 1 mol:1 mol
stoichiometry of IGF-I and 125I-IGFBP-3, approximately 80%
of the 125I-IGFBP-3 specific binding to the type V TGF-
receptor was inhibited. Heparin at 100 µg/ml inhibited ~80% of the
125I-IGFBP-3 binding to the type V TGF-
receptor (Fig.
2D, lane 9 versus lane 1). The control (Fig. 2D, lane
1) was overexposed to show the 125I-IGFBP-3-type V
TGF-
receptor complex in lanes 2, 3, and 8. These results suggest that both the 125I-IGFBP-3-IGF-I
complex and the 125I-IGFBP-3-heparin complex are not
capable of binding to the type V TGF-
receptor. As a control,
TGF-
1 peptide antagonist (3 µM) strongly
inhibited >95% of the 125I-IGFBP-3 binding to the type V
TGF-
receptor (Fig. 2D, lane 2 versus lane 1). To further
demonstrate that the type V TGF-
receptor is the putative IGFBP-3
receptor, we performed the 125I-IGFBP-3 affinity labeling
of its putative cell-surface receptor in cells expressing and lacking
the type V TGF-
receptor. As shown in Fig.
3, human colorectal carcinoma cells
(RII-37 cells and HCT 116 Neo cells), which lack the type V TGF-
receptor (7, 27), did not show the ~400-kDa
125I-IGFBP-3-putative receptor complex (Fig. 3, lanes
3-6). In contrast, NIH 3T3 cells, which are known to express the
type V TGF-
receptor (26), showed the ~400-kDa
125I-IGFBP-3-putative receptor complex (Fig. 3, lanes
1 and 2). The formation of the ~400-kDa
125I-IGFBP-3-putative receptor complex in NIH 3T3 cells was
blocked in the presence of 100-fold excess of unlabeled IGFBP-3 or 10 µM TGF-
1 peptide antagonist (data not
shown).
Since the type V TGF- receptor has been shown to mediate the growth
inhibitory response in mink lung epithelial cells (7), we examined the
effect of IGFBP-3 on the proliferation of wild-type and type I and type
II TGF-
receptor-defective mutant mink lung epithelial cells (Mv1Lu,
R-1B, and DR26 cells, respectively) (22, 28-30). All Mv1Lu, R-1B, and
DR26 cells have been shown to express the type V TGF-
receptor (7).
IGFBP-3 should be a specific ligand to test the function of the type V
TGF-
receptor, because it does not bind to the type I, type II, or
type III TGF-
receptor with high affinity. As shown in Fig.
4, IGFBP-3 (0.6 µg/ml or ~20
nM) induced a similar growth inhibitory response as
measured by DNA synthesis (~60% inhibition) in either wild-type
(Mv1Lu cells) or type II TGF-
receptor-defective mutant mink lung
epithelial cells (DR26 cells), but to a lesser extent (~20%
inhibition) in type I TGF-
receptor-defective mutant cells (R-1B
cells). The growth inhibitory response induced by IGFBP-3 in these
cells could be blocked in the presence of TGF-
1 peptide
antagonist (Fig. 4). These results indicate that IGFBP-3 induces a
growth inhibitory response in cells expressing the type V TGF-
receptor. These results also support the hypothesis that the type V
TGF-
receptor can mediate the growth inhibitory response (7).
In a previous study (26), we reported that many types of carcinoma
cells lacked the type V TGF- receptor and that such cells do not
respond to TGF-
1 stimulation, as measured by growth inhibition (7). Recently, hereditary human colorectal carcinoma cells
(HCT 116 cells) were shown to be deficient in the type II TGF-
receptor (27). Stable transfection of these carcinoma cells with the
type II TGF-
receptor cDNA was found to rescue the
transcriptional response but failed to restore the growth inhibitory
response to exogenous TGF-
stimulation (27). This appears to be due
to the lack of the type V TGF-
receptor expression in cells stably
transfected with the neo vector only (HCT 116 Neo cells) or
with vector expressing the type II TGF-
receptor cDNA (RII-37
cells) (7, 27). As would be expected, IGFBP-3 also failed to inhibit
growth in these HCT 116 Neo and RII-37 cells that do not express the
type V TGF-
receptor (data not shown).
TGF- elicits a variety of biological activities in different cell
types (23-25). In addition to growth inhibitory activity, the other
prominent activity of TGF-
is transcriptional activation of
fibronectin, collagen, and PAI-1 genes (23-25). To see if IGFBP-3 and
TGF-
share similar activities, we determined the effect of IGFBP-3
on the transcriptional expression of PAI-1 in mink lung epithelial
cells. IGFBP-3 showed little if any effect on the transcription of
PAI-1 in these epithelial cells (data not shown). TGF-
has been
shown to be a bifunctional growth regulator: a growth inhibitor for
epithelial cells, endothelial cells, and other cell types, and a
mitogenic factor for mesenchymal cells (23-25). We therefore determined the effect of IGFBP-3 on DNA synthesis in NIH 3T3 cells, for
which TGF-
is a mitogen. IGFBP-3 did not stimulate DNA synthesis of
NIH 3T3 cells at concentrations of 0.1-100 nM, suggesting
that IGFBP-3 is a partial agonist of TGF-
. These results also
support the hypothesis that the type V TGF-
receptor preferentially
mediates the growth inhibitory response in responsive cells (7).
IGFBP-3 has been implicated as a mediator of the actions of TGF-,
retinoic acid, and p53 (19-21). Antisense deoxyoligonucleotide to
IGFBP-3 has been shown to diminish the growth inhibitory response induced by TGF-
and retinoic acid in human mammary carcinoma cells
(19-21). The functional role of IGFBP-3 in TGF-
-induced growth
inhibition in other cell types is unknown. We speculate that the
IGFBP-3 expression induced by TGF-
and retinoic acid in these
mammary carcinoma cells may cause the growth inhibition by sequestering
IGFs from binding to IGF-I receptor in IGF-responsive cells and by
propagating the growth inhibitory response mediated by the type V
TGF-
receptor in IGF-unresponsive cells. It is important to note
that upon ligand activation, the type V TGF-
receptor may also
decrease IGF-II concentration in the extracellular compartment by
increasing the internalization and recycling of the cell surface
mannose 6-phosphate/IGF-II
receptor.4 This effect on
IGF-II may also contribute to the growth inhibitory response mediated
by the type V TGF-
receptor.4
We thank Celtrix Pharmaceutical Inc. for
providing recombinant nonglycosylated human IGFBP-3; Drs. Joan
Massagué and Michael G. Brattain for providing TGF-
receptor-defective mutant mink lung epithelial cells (R-1B and DR26
cells) and human colorectal carcinoma cells (HCT 116 Neo and RII-37
cells), respectively; Drs. William S. Sly and Frank E. Johnson for
critical comments and review of the manuscript; and Maggie Klevorn for
editorial assistance in the preparation of this manuscript.