| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Unité Interactions entre Systèmes Protéiques et Différenciation dans la Cellule Tumorale, UPRES-A CNRS 6032, Université d’Aix-Marseille I et II, Faculté de Pharmacie, 13385 Marseille; and INSERM U-387 (J.-C.L.), 13009 Marseille, France
Address all correspondence and requests for reprints to: Dr. Jacques Marvaldi, Laboratoire de Biochimie Cellulaire, UPRES-A CNRS 6032, Faculté de Pharmacie, 27 boulevard Jean Moulin, 13385 Marseille Cedex 05, France. E-mail: marvaldi{at}newsup.univ-mrs.fr
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-IR3) showed
that LoVo-C5 cells expressed a major high molecular mass receptor (200
kDa) corresponding to the unprocessed /ß pro-receptor. A small
amount of successfully cleaved /ß heterodimers was also produced,
indicating a residual endoproteolytic cleavage activity in these cells.
In vitro, a soluble form of recombinant furin was able
to cleave the pro-IGF-IR (200 kDa) into -subunit (130 kDa) and
ß-subunit (97 kDa). Measurement of IGF binding parameters in LoVo-C5
cells indicated a low number of typical type I IGF-binding sites
(binding capacity, 5 x 103 sites/cell;
Kd, 1.9 nM for IGF-I and 7.0 nM for
IGF-II). These findings in LoVo-C5 contrast with those in HT29-D4
cells, which have active furin, and where IGF-IR (2.8 x
104 sites/cell) was fully processed. Moreover, the 200-kDa
pro-IGF-IR of LoVo-C5 was unable to induce intracellular signaling,
such as ß-subunit tyrosine autophosphorylation and insulin-related
substrate-1 tyrosine phosphorylation. Flow immunocytometry analysis
using -IR3 antibody indicated that LoVo-C5 cells expressed 40% more
receptors than HT29-D4 cells, suggesting that in LoVo-C5 cells only the
small amount of mature type I IGF-IR binds IGFs with high affinity. To
provide evidence for this idea, we showed that mild trypsin treatment
of living LoVo-C5 cells partially restored /ß cleavage of IGF-IR,
and greatly enhanced (6-fold) the IGF-I binding capacity of LoVo-C5
cells, but did not restore IGF-IR signaling activity. Moreover, LoVo-C5
cells were totally unresponsive to IGF-I in terms of cell migration, in
contrast to fully processed IGF-IR-HT29-D4 cells. Our data indicate
that furin is involved in the endoproteolytic processing of the IGF-IR
and suggest that this posttranslational event might be crucial for its
ligand binding and signaling activities. However, our data do not
exclude that other proprotein convertases could participate to IGF-IR
maturation. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
2/ß2 heterotetramer structurally related to the insulin receptor.
The -subunits bear IGF-I- and IGF-II-binding sites, whereas
ß-subunits possess intrinsic tyrosine kinase activity. IGF-IR is
synthesized as a high molecular weight /ß pro-receptor, which is
proteolytically processed into mature - and ß-chains, linked by
disulfide bonds. The Arg-Lys-Arg-Arg pro-receptor cleavage site (3)
belongs to a family of sites comprising pairs of basic amino acids
involved in the posttranslational processing of many proteins (4, 5).
The pro-protein convertases, a family of subtilisin-related
calcium-dependent serine proteases, are well known to be involved in
processing of such sites (for a review, see Refs. 6, 7, 8). One such
protease, furin, contains a transmembrane domain and is widely
distributed among cells. Furin is mainly located intracellularly in the
trans-Golgi network, but is also expressed in the plasma
membrane (9) or secreted after truncation (10). Furin has been involved
in the processing of various membrane receptors, including the insulin
receptor (11, 12), the met protooncogene (13), and the
integrin -subunits (14) and in the cleavage of viral glycoproteins
(15). The human colonic carcinoma cell line LoVo has been reported to
be deficient in furin activity due to a mutation in both alleles of the
furin gene (16, 17). LoVo cells, therefore, are useful to examine the
function of furin-dependent processing for protein activity. In the
present report, we compare the structural characteristics and the
functional activity of the IGF-IR from two colonic cancer cell lines:
HT29-D4 cells, which exhibit a classical receptor and LoVo-C5 cells, a
clone of LoVo cells (18). We demonstrate that in LoVo-C5 cells, IGF-IR
was expressed as unprocessed polypeptide chains at the plasma membrane
and that the pro-receptor was unable to bind IGF-I and to transduce
intracellular signals and mediate biological responses. |
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
2000 Ci/mmol), horseradish
peroxidase-streptavidin complex (HRP-streptavidin), enhanced
chemiluminescence detection reagents (ECL), and Hybond C+
nitrocellulose sheet were purchased from Amersham (Les Ullis, France).
Sulfosuccinimidyl-6-(biotinamido) hexanoate (NHS-LC-biotin) was
obtained from Pierce (Rockford, IL). Anti-IGF-IR monoclonal antibody
(mAb; clone -IR3) and protein G-agarose were obtained from Oncogene
Science (Uniondale, NY). Antiinsulin receptor mAb (B6) was purchased
from Immunotech (Marseille, France). Antiphosphotyrosine mAb (PY 20)
was obtained from ICN Biomedical (Aurora, OH). All other reagents were
of analytical grade.
Cell culture
The human colon adenocarcinoma cell lines HT29-D4 and LoVo-C5
were cloned by limiting dilution of corresponding parental cell lines
and were cultured as previously described (18, 19) in DMEM and McCoy’s
5A modified medium, respectively, containing 10% FCS.
Cell toxicity of pseudomonas exotoxin A
Cells were seeded at 10,000 cells/well in 48-well cell culture
plates. After a 24-h growing phase, cells were incubated in triplicate
for 3 days in culture medium with or without PEA at the indicated
concentration. Cells were then fixed in 2% glutaraldehyde solution and
stained with crystal violet (1% in water). After extensive washing,
cells were lysed with 1% SDS, and the intensity of incorporated
staining was measured using a microplate reader at 600 nm.
Cell surface labeling
Cell surface proteins were labeled with NHS-LC biotin as
previously described (20). Labeled cells were lysed in RIPA buffer [50
mM Tris-HCl (pH 8.0), 150 mM NaCl, 1% Triton
X-100, 0.5% BSA, and 1 mM EDTA] containing a mixture of
protease inhibitors [leupeptin, iodoacetamide, pepstatin,
ß2-microglobulin (1 µg of each), and 1 mM
phenylmethylsulfonylfluoride]. Extracted proteins were then
immunoprecipitated with mAb -IR3 as previously described (21).
Biotinylated antigens were analyzed by SDS-PAGE (7.5% acrylamide)
under reducing conditions and were blotted onto Hybond C+
nitrocellulose sheets. Membranes were then blocked with 50
mM Tris-HCl (pH 8.0), 10% (wt/vol) glycerol, 1
M glucose, 0.5% BSA, and 0.1% Tween-20 for 1 h at
room temperature. Biotinylated proteins were labeled with
HRP-streptavidin in the same buffer for 1 h at room temperature
and were detected by the ECL system according to supplier
instructions.
Furin digestion of pro-IGF-IR
Recombinant furin (a gift of Dr. N. Seidah, Institut de
Recherche Clinique de Montréal, Montreal, Canada) was obtained
from the medium of sf 9 cells infected with a recombinant bacullovirus.
The cells express a soluble human furin with a stop codon introduced in
place of the codon for furin amino acid 586 located before the
cysteine-rich domain (BCRD-furin). BCRD-furin’s specific activity was
150 nM pERTKR-AMC cleavage activity/ml·h. IGF-IR was
immunopurified from the lysate of cell surface labeled LoVo-C5 cells
using -IR3 mAb as described above. After washing, beads were
resuspended in 62 mM Tris-HCl, pH 6.8, containing 1%
N-octylglucoside. Beads were incubated with furin at the
indicated amount in the same buffer in presence of either 2
mM CaCl2 or 10 mM EDTA for 4 h
at room temperature under constant shaking. The reaction was stopped by
the addition of 5 x electrophoresis buffer sample buffer
containing 1% dithiothreitol. Biotinylated IGF-IR was analyzed by
7.5% SDS-PAGE, transferred onto a nitrocellulose sheet, and detected
with the HRP-streptavidin/ECL system.
IGF competitive binding to cell surface
IGF competitive binding experiments were carried out essentially
as previously described (22). Briefly, HT29-D4 and LoVo-C5 cells
cultured in 16-mm wells were washed three times in serum-free DMEM,
then incubated in a total volume of 0.2 ml binding medium (RPMI-HEPES
containing 0.1% BSA) for 2 h at 4 C (equilibrium binding
conditions) with 0.15 nM [125I]IGF-I with or
without various concentrations of unlabeled IGF-I or IGF-II. Cells were
washed three times with cold PBS containing 0.1% BSA, then lysed with
1 ml 0.1 M NaOH. Cell-associated radioactivity was counted
in a 
-spectrometer. Nonspecific binding, determined as the
radioactivity bound to cells in the presence of 1 µM
unlabeled peptide, was subtracted from total binding to obtain specific
binding. Nonspecific binding never exceed 2.5% of the total binding of
[125I]IGF-I. Binding data were analyzed using the
EBDA/Ligand software (23).
IGF-stimulated tyrosine phosphorylation
Tyrosine phosphorylation of cellular proteins was examined by
immunoblotting with PY20 antiphosphotyrosine mAb essentially as
described by Kato et al. (24). Confluent HT29-D4 and LoVo-C5
cells were washed three times with RPMI-HEPES containing 1% BSA and
starved in this medium for 16 h at 37 C. Cells were washed and
then stimulated with IGF-I, IGF-II, or insulin (100 nM) in
the same buffer for 1, 5, or 15 min at 37 C; washed with cold PBS on
ice; frozen on liquid nitrogen; and thawed on ice. Cells were lysed
(250 µl/well) with 50 mM HEPES containing 150
mM NaCl, 1% Triton X-100, 1 µg/ml bacitracin, 1
mM phenylmethylsulfonylfluoride, 5 mM sodium
orthovanadate, 100 mM NaF, and 10 mM EDTA.
Lysates were clarified by centrifugation (12,000 x g
for 3 min). In some experiments, IGF-IR were immunopurified from cell
lysates with -IR3 mAb as described above. Equal amounts of proteins
were subjected to 7.5% acrylamide gel and transferred onto
nitrocellulose sheet. Tyrosine-phosphorylated proteins were probed with
PY20 mAb and detected by chemiluminescence.
Flow cytometric analysis of IGF-IR surface expression
Cell surface expression of IGF-IR in HT29-D4 and LoVo-C5 cells
was determined by flow cytometry using the -IR3 antibody on a Becton
Dickinson flow cytometry device as previously described (25). Gating
parameters were set so that 90% of the injected cells were
analyzed.
Trypsin digestion of cell surface IGF-IR in living cells
Subconfluent HT29-D4 or LoVo-C5 cells were washed in serum-free
DMEM and incubated for 10 min at 37 C in the presence of various
amounts of trypsin in serum-free medium. After adding an excess of
soybean trypsin inhibitor, cells were washed three times with DMEM
containing 10% FCS, then with ice-cold PBS. Cell surface NHC-LC biotin
labeling, protein extraction, immunoprecipitation with -IR3
antibody, and analysis by Western blotting were carried out as
described above. In the same set of experiments, trypsin-treated cells
were also tested for their IGF-I binding capacity as described above.
In addition, internalization of membrane-bound
[125I]IGF-I was determined as previously described (22).
Briefly, cells were washed three times with cold PBS containing 1% BSA
and incubated for 5 min at 4 C with 1 ml 0.2 M acetic acid
and 0.5 M NaCl, pH 2.5, solution. After one rinsing step,
acidic extracts were collected, and radioactivity corresponding to cell
surface-bound [125I]IGF-I was evaluated in a -counter.
Cells were then lysed in 0.1 M NaOH, and the amount of
internalized [125I]IGF-IR complexes was counted.
Cell migration
Wounding assays were performed on HT29-D4 and LoVo C5 cell
monolayers. Briefly, cells were seeded into 35-mm diameter 6-well
plates at 1.5 x 105 cells/cm2 and allowed
to grow to confluence during 4 days. Confluent cell monolayers were
wounded by pressing a sterilized razor blade down onto the plate to cut
the cell sheet and to mark on the plate a sharp visible demarcation at
the wound edge. The blade was then gently moved to one side to remove
part of the cell monolayer sheet. Two approximately 15- to 20-mm wounds
separated by about 10 mm were made in the same well. The wounded
monolayers were then washed 4 times in DMEM to remove cell debris and
incubated for 24 h at 37 C in serum-free DMEM containing 0.1% BSA
with or without test substances. At the end of incubation, cells were
rinsed twice with PBS, fixed in 4% formaldehyde in PBS for 20 min at 4
C, and stained using the May-Grünwald/Giemsa method. Migration of
HT29-D4 and LoVo C5 cells was assessed using an inverted Zeiss MI
microscope (Zeiss, New York, NY) by numbering the cell nuclei observed
across the wound borders. To avoid observer bias, 10 microscopic fields
were analyzed for each wound in a blinded fashion. Migration results
are expressed as the average ± SD of the number of
migrating cells per microscopic field (magnification, x320). All data
presented in Results are from three independent experiments
performed in duplicate.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
).
|
-subunit mAb
(-IR3). As previously reported (27), we coimmunoprecipitated the
-subunit (130 kDa) and two isoforms of the ß-subunits (97 and 102
kDa) from our control HT29-D4 cell line (Fig. 2A). Two isoforms of IGF-IR ß-subunits
have been reported to exist in several other cell types (27, 28). When
similar experiments were performed on LoVo-C5 cells, we observed a
major high molecular mass protein (200 kDa); however, faint bands
corresponding to the processed - and two ß- subunits isoforms were
also visible (Fig. 2, A and B). This result suggests that in LoVo-C5
cells, the majority of IGF-IR was expressed on the plasma membrane as
an immature 200-kDa form, and that only a small amount of the receptor
was successfully cleaved into - and ß-subunits. This lack of
IGF-IR endoproteolytic processing might be related to the absence of
furin activity in LoVo-C5 cells. To test this hypothesis, we incubated
the -IR3-immunopurified 200-kDa protein in the presence of a
recombinant soluble form of furin (BCRD-furin). As illustrated in Fig. 2B, in vitro BCRD-furin treatment led to a massive
conversion of the 200-kDa protein into two peptides (130 and 97 kDa)
corresponding to the IGF-IR - and ß-subunits. The processing of
IGF-IR was calcium dependent because no cleavage was observed in the
absence of calcium and the presence of 10 mM EDTA. It
should be noted that in contrast with HT29-D4 cells, furin-processed
IGF-IR in LoVo-C5 cells appeared to contain only one ß-subunit
isoform (compare Fig. 2, A and B).
|
shows representative results of a
competition experiment with unlabeled IGF-I or IGF-II and
[125I]IGF-I. [125I]IGF-I binding to HT29-D4
cells was inhibited by both IGF-I and IGF-II, with IC50
values of 3 and 7 nM, respectively. The IC50
for binding inhibition to LoVo-C5 cells was similar, 4 and 7
nM for IGF-I and IGF-II, respectively (Fig. 3A). Analysis
of these [125I]IGF-I binding inhibition curves, using the
Ligand program, gave linear Scatchard plots, consistent with a single
class of binding sites for both cell lines (Fig. 3B). The
Kd values obtained for IGF-I were 1.5 and 1.9
nM for HT29-D4 and LoVo-C5 cells, respectively. In
addition, Ligand analysis of the inhibition of
[125I]IGF-I binding by unlabeled IGF-II, yielded a
Kd of 8.0 and 7.0 nM in HT29-D4 and LoVo C5
cells, respectively. The calculated binding capacity revealed that
LoVo-C5 cells expressed a low number of IGF-I-binding sites (5.1
x 103 sites/cell), whereas HT29-D4 cells expressed
2.8 x 104 binding sites/cell. Therefore, we observed
that LoVo-C5 expressed 5.4-fold less typical IGF-IR than HT29-D4
cells.
|
-IR3 antibody was significantly higher in LoVo-C5 than in HT29-D4
cells (Fig. 4). A quantification derived
from the cytofluorometric analysis indicated that the amount of IGF-IR
was 40% higher in LoVo-C5 cells than in HT29-D4 cells. Therefore,
LoVo-C5 cells expressed on their surface more IGF-IR (mature and
unprocessed forms) than HT29-D4 cells. Because LoVo-C5 expressed mainly
unprocessed IGF-IR (Fig. 2
), and we measured only very few high
affinity binding sites (Fig. 3), the majority of IGF-IR was expressed
in LoVo-C5 cells as the pro form, that is biologically inactive. Thus,
a correct endoproteolytic processing appears to be a crucial event for
the binding activity of the IGF-IR.
|
clearly
shows a 6-fold rise in the ligand binding capacity of trypsin-treated
LoVo-C5 cells (trypsin at 100 µg/ml). We also immunoprecipitated
IGF-IR after cell surface biotinylation of trypsin-treated LoVo-C5
cells. Under these conditions, single chain receptors were mainly
cleaved into one -subunit (130 kDa) and two ß-subunit (97 and 102
kDa) isoforms by the enzyme (Fig. 5B). These results directly
demonstrate that the lack of endoproteolytic processing of IGF-IR
inhibits its binding activity. We did not observed significant
endocytosis of [125I]IGF-I in LoVo-C5 cells, treated or
not with trypsin, as we did in HT29-D4 cells (Fig. 5C). Therefore,
trypsin treatment, did not change the endocytic activity of
IGF-I-IGF-IR complexes in LoVo C5 cells.
|
).
Figure 6B shows that treatment of HT29-D4 cells with either IGF-I,
IGF-II, or insulin (100 nM) enhanced tyrosine
phosphorylation of three proteins (185 kDa and a doublet of 97/102
kDa). These three polypeptides probably correspond to the IRS-1 protein
(185 kDa) and to the IGF-IR ß-subunits isoforms (97/102 kDa). As
expected for the type I IGF-IR, IGF-I was more potent than IGF-II and
even more potent than insulin in inducing tyrosine phosphorylation.
However, we failed to detect any significant change in the amount of
phosphorylated tyrosine residues in LoVo-C5 cells regardless of the
stimulating peptide used or the time of stimulation, i.e.
1–10 min (Fig. 6A). Because trypsin restored the binding activity of
IGF-IR, we have looked for IGF-I-induced tyrosine phosphorylation
signaling in LoVo-C5 after a mild trypsin treatment. Figure 6A
(LoVo-C5-Ti) shows that neither IRS-1 nor the ß-subunits of the
IGF-IR was phosphorylated after trypsin treatment. These data indicated
that the cleavage of the IGF-IR by trypsin is sufficient to restore the
binding activity of the receptor but not its signaling activity. In
LoVo-C5, a polypeptide that migrated just above the IRS-1 protein was
found to be constitutively phosphorylated. It is unlikely that this
polypeptide corresponded to the pro-IGF-IR because we have performed
preclearing experiments of lysates from LoVo-C5 cells in the presence
of -IR3 antibodies without altering the amount of this polypeptide
under experimental conditions where 90% of cell surface IGF-IR was
depleted (Fig. 6, C and D). To confirm the absence of IGF-I-induced
tyrosine phosphorylation of IGF-IR in LoVo-C5 cells, HT29-D4 and
LoVo-C5 cells were stimulated by IGF-I, then IGF-IR were immunopurified
by -IR3 mAb before analysis by Western blotting with an
anti-phosphotyrosine antibody. Figure 7
shows that the two ß-subunits isoforms of HT29-D4 cells were
phosphorylated after IGF-I stimulation, whereas in LoVo-C5 cells, we
failed to observe any phosphorylation in either the ß-subunits or in
the proreceptor regions of the gel. Taken together, these data confirm
that in LoVo-C5 cells no detectable amount of IGF-IR was expressed as a
fully active processed form and that the proreceptor was under an
inactive form.
|
|
show that both types of cell were able to migrate in the presence of
PMA. LoVo-C5 cells were totally refractory to IGF-I-induced migration.
This was observed even at very high concentrations (500 ng/ml) of the
growth factor (data not shown), whereas HT29-D4 cells migration was
strongly stimulated under the same experimental conditions.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The clonal population (C5) of LoVo cells we used to examine the IGF-IR
processing was PEA resistant (26), thus indicating that they were furin
deficient. These cells expressed a 200-kDa protein immunologically
related to IGF-IR. We demonstrated that the 200-kDa protein
corresponded to the single chain pro-receptor, and we were able to
cleave it with a soluble recombinant furin to obtain the two processed
-subunits (130 kDa) and ß-subunits (97 kDa). These data indicate
that in LoVo-C5, furin deficiency resulted in an impaired
endoproteolytic processing of the IGF-IR. In contrast, trypsin added to
intact LoVo-C5 cells induced cleavage of pro-IGF-IR into one
-subunit (130 kDa) and two distinct ß-subunit isoforms (97 and 102
kDa). Two isoforms of ß-subunits were also observed for endogenous
IGF-IR in HT29-D4 cells (27). The discrepancy between furin- and
trypsin-induced cleavage of pro-IGF-IR could be related to either
experimental conditions (cleavage of purified pro-IGF-IR vs.
cleavage of native pro-IGF-IR at the cell surface, respectively) or the
broad specificity of trypsin compared with furin for basic residues in
the processing of recognition motifs. Several IGF-IR variants
containing different ß-subunit isoforms that can be distinguished by
their apparent molecular mass have been widely reported (1, 3, 27, 28).
It is unlikely that the different ß-subunits come from alternative
splicing of a single IGF-IR transcript or from an unidentified gene.
Moreover, it is also unclear whether they arise from posttranslational
alterations, e.g. glycosylation (3). Thus, we suggest that
IGF-IR variants exhibiting distinct ß-subunits might be the result of
pro-IGF-IR differential processing by several acting convertases. The
low signal of processed /ß-chains in LoVo-C5 cells would reflect
such a residual processing activity of convertases other than furin,
such as PACE-4 or the recently cloned PC7 convertase (30, 31). This
point is now under investigation.
IGF-IR is not the first receptor known to be expressed as a pro form.
Mondino and collaborators reported that hepatocyte growth factor and
insulin receptors were also expressed as an uncleaved form at the
plasma membrane of the parental LoVo cell line (13). We were unable to
detect any insulin receptor at the surface of C5 clone by ligand
binding, immunoprecipitation, or flow immunocytometry experiments (data
not shown). More recently, we have shown that only uncleaved forms of
the 6, 3, and v integrin subunits are expressed at the surface
of LoVo-C5 cells (14). Therefore, it appears that endoproteolytic
processing is not required for cell surface targeting of receptors,
including, as we report here, the IGF-IR.
We compared the biological activity of uncleaved and cleaved IGF-IR by
competitive binding experiments between radiolabeled and unlabeled
IGF-I or -II. LoVo-C5 cells exhibited only 17% of the binding capacity
of HT29-D4 cells. Scatchard analysis indicated that this discrepancy
was a consequence of a lower number of binding sites at the cell
surface (5.1 x 103 for LoVo-C5 vs.
2.8 x 104 for HT29-D4 cells), rather than a
difference in the receptor affinity (Kd of IGF-I = 1.9
nM and 1.5 nM for LoVo-C5 and HT29-D4 cells,
respectively). On the contrary, results from immunocytofluorescence
experiments indicate that LoVo-C5 cells expressed 40% more IGF-IR than
HT29-D4 cells. As most of the cell surface IGF-IR were in the pro form
at the LoVo-C5 cell surface, these pro-IGF-IR should be biologically
inactive. Indeed, we demonstrated that the pro-IGF-IR was not tryrosine
phosphorylated in response to IGF-I stimulation. Due to the very small
number of properly processed IGF-IR, it is not surprising that we were
unable to detect either IGF-IR tyrosine autophosphorylation and
intracellular phosphorylation signaling or ligand endocytosis after IGF
stimulation of LoVo-C5 cells. Endoproteolytic processing of the IGF-IR
might control the ligand binding capacity, and subsequently, the
intracellular signaling activities of the receptor. This was further
supported by a method successfully used by Yoshimasa and co-workers in
their study of the insulin receptor in the furin-deficient CHO-RPE 40
cell line. This group showed that trypsin mimicked furin activity by
cleaving the pro-insulin receptor at the surface of CHO-RPE 40 cells
(32). We obtained similar results, showing that trypsin and recombinant
furin cleaved the pro-IGF-IR at the LoVo-C5 plasma membrane. Moreover,
trypsin treatment greatly enhanced (6-fold) the binding capacity of
LoVo-C5 cells, without modifying that of HT29-D4 cells. Thus, recovery
of - and ß-subunit cleavage could restore IGF-IR-binding activity
in LoVo cells. In contrast, the trypsin-induced /ß cleavage of
pro-IGF-IR did not restore its ability to induce ß-subunit and IRS-1
tyrosine phosphorylation. This suggests that this cleavage was not
enough accurate to allow the necessary spatial relationship between
-subunit ligand occupancy and ß-subunit autophosphorylation within
the mature heterotetrameric 2ß2 IGF-IR. This observation also
explains why ligand-induced IGF-IR endocytosis that requires IGF-IR to
be functional was not restored by trypsin treatment of LoVo-C5
cells.
What could be the effect of type I IGF-IR cleavage deficiency in vivo? Although a deficient processing of the type I IGF-IR in vivo have not yet been reported, a disease phenotype is associated with aberrant processing of a closely related receptor, the insulin receptor. Patients with leprechaunism display mutations in the insulin receptor gene within a region outside the tetrabasic cleavage site. These mutations lead to an alteration of the cell surface targeting of the insulin receptor, which probably does not traffic through the intracellular compartment where functional furin is concentrated (33, 34). One case of type A insulin resistance is interesting, because a point mutation within the tetrabasic processing site of the insulin receptor dramatically reduces insulin binding without affecting cell surface exportation of the receptor (35). Surprisingly, the same group reports that a mutation of the arginine residue in position P4 within the cleavage site does not affect either the exportation of the insulin receptor or the ligand binding to the uncleaved pro-receptor. Moreover, mutations on arginine in P2 or lysine in P3 position do not impair the endoproteolytic processing of the insulin receptor. The last data are in favor of a highly specific control of the processing involving endoproteases.
Several recent studies have reported the essential role of IGFs (essentially IGF-II) autocrine loops in the regulation of growth, differentiation, and survival of colorectal cell lines (36, 37, 38, 39). Because HT29-D4 and LoVo-C5 cells were found unresponsive to IGF-I or des-(1–3)-IGF-I in term of cell proliferation (36, 39), we have investigated the effects of these growth factors on cell migration. The accumulation of evidence indeed indicates that IGFs are strong promoters of cell mobility for a variety of normal and malignant cell types (2). Our results show that LoVo-C5 cells were unable to migrate in the presence of des-(1–3)-IGF-I, whereas HT29-D4 cell migration was greatly stimulated by this polypeptide growth factor. The defect in LoVo-C5 cell migration was not an intrinsic property of these cells because PMA was a potent migration stimulator as it was for HT29-D4 cells. We believe that the refractoriness to IGF-I-induced LoVo-C5 cell migration was due to defects in the maturation of the IGF-IR and subsequent signaling events.
Of course, it will be crucial in the future to carefully examine receptor structure and function in LoVo-C5 cells transfected with genes encoding active furin or other convertases. This will allow us to better understand the nature of the convertases involved in the maturation process of the IGF-IR and the precise role of the cleavage in the function of this receptor.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received December 29, 1997.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-chain endoproteolytic cleavage in
furin-deficient human colon adenocarcinoma cell LoVo. Biochem J 317:803–809
vß5 and vß6 integrin glycosylation in the
adhesion of a colonic adenocarcinoma cell line (HT29–D4). J Cell
Biochem 61:266–277[CrossRef][Medline]
. Br J Cancer 65:341–346[Medline]
This article has been cited by other articles:
 |
|
 |
|
  C. Gyamera-Acheampong and M. Mbikay Proprotein convertase subtilisin/kexin type 4 in mammalian fertility: a review Hum. Reprod. Update, March 1, 2009; 15(2): 237 - 247. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. K. McColl, K. Paavonen, T. Karnezis, N. C. Harris, N. Davydova, J. Rothacker, E. C. Nice, K. W. Harder, S. Roufail, M. L. Hibbs, et al. Proprotein convertases promote processing of VEGF-D, a critical step for binding the angiogenic receptor VEGFR-2 FASEB J, April 1, 2007; 21(4): 1088 - 1098. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. A. Samani, S. Yakar, D. LeRoith, and P. Brodt The Role of the IGF System in Cancer Growth and Metastasis: Overview and Recent Insights Endocr. Rev., February 1, 2007; 28(1): 20 - 47. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. E. Bassi, R. Lopez De Cicco, J. Cenna, S. Litwin, E. Cukierman, and A. J.P. Klein-Szanto PACE4 Expression in Mouse Basal Keratinocytes Results in Basement Membrane Disruption and Acceleration of Tumor Progression Cancer Res., August 15, 2005; 65(16): 7310 - 7319. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Q. Qiu, A. Basak, M. Mbikay, B. K. Tsang, and A. Gruslin Role of pro-IGF-II processing by proprotein convertase 4 in human placental development PNAS, August 2, 2005; 102(31): 11047 - 11052. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Kawashima, S. Kanzaki, F. Yang, T. Kinoshita, K. Hanaki, J.-i. Nagaishi, Y. Ohtsuka, I. Hisatome, H. Ninomoya, E. Nanba, et al. Mutation at Cleavage Site of Insulin-Like Growth Factor Receptor in a Short-Stature Child Born with Intrauterine Growth Retardation J. Clin. Endocrinol. Metab., August 1, 2005; 90(8): 4679 - 4687. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. L. de Cicco, J. C. Watson, D. E. Bassi, S. Litwin, and A. J. Klein-Szanto Simultaneous Expression of Furin and Vascular Endothelial Growth Factor in Human Oral Tongue Squamous Cell Carcinoma Progression Clin. Cancer Res., July 1, 2004; 10(13): 4480 - 4488. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Duckert, S. Brunak, and N. Blom Prediction of proprotein convertase cleavage sites Protein Eng. Des. Sel., January 1, 2004; 17(1): 107 - 112. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. E. Bassi, H. Mahloogi, R. Lopez De Cicco, and A. Klein-Szanto Increased Furin Activity Enhances the Malignant Phenotype of Human Head and Neck Cancer Cells Am. J. Pathol., February 1, 2003; 162(2): 439 - 447. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Croissandeau, A. Basak, N. G. Seidah, M. Chretien, and M. Mbikay Proprotein convertases are important mediators of the adipocyte differentiation of mouse 3T3-L1 cells J. Cell Sci., March 15, 2002; 115(6): 1203 - 1211. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A.-M. Khatib, G. Siegfried, A. Prat, J. Luis, M. Chretien, P. Metrakos, and N. G. Seidah Inhibition of Proprotein Convertases Is Associated with Loss of Growth and Tumorigenicity of HT-29 Human Colon Carcinoma Cells. IMPORTANCE OF INSULIN-LIKE GROWTH FACTOR-1 (IGF-1) RECEPTOR PROCESSING IN IGF-1-MEDIATED FUNCTIONS J. Biol. Chem., August 10, 2001; 276(33): 30686 - 30693. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. E. Bassi, R. Lopez De Cicco, H. Mahloogi, S. Zucker, G. Thomas, and A. J. P. Klein-Szanto Furin inhibition results in absent or decreased invasiveness and tumorigenicity of human cancer cells PNAS, August 28, 2001; 98(18): 10326 - 10331. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
) and LoVo-C5 cells (•) were treated by increasing
concentrations of PEA. To measure cell survival, cells were fixed,
stained with crystal violet, and the intensity of staining was
monitored at 600 nm.
). After a 2-h
incubation period at 4 C, specific binding was determined as described
in Materials and Methods. The results were expressed as
a percentage of binding in the absence of competitor. Each
point represents a mean of three duplicate experiments.
SD was always lower than 5%. B, Scatchard plots of IGF-I
binding to HT29-D4 (
) cells analyzed by the Ligand
program. The two solid lines are the computer-generated
best fit for one-site IGF binding model.
