| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Medical Research Council Group in Fetal and Neonatal Health and Development (S.L.B., V.K.M.H.), Departments of Biochemistry (S.L.B., V.K.M.H.) and Pediatrics (V.K.M.H.), University of Western Ontario, The Lawson Research Institute, London, Ontario, Canada N6A 4V2; and the Department of Pediatrics, University of North Carolina (A.J.D.), Chapel Hill, North Carolina 27514
Address all correspondence and requests for reprints to: Victor K. M. Han, M.D., Room H308, The Lawson Research Institute, 268 Grosvenor Street, London, Ontario, Canada N6A 4V2. E-mail: vhan{at}julian.uwo.ca
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
There is now substantial evidence that supports the role of IGFs in the growth and development of the brain. IGF messenger RNAs (mRNAs) have been localized in the rat brain during pre- and postnatal development (13, 14, 15, 16, 17). Both IGF-I and IGF-II receptor mRNAs have been localized to regions of the developing rat brain (18, 19), and specific binding of IGFs has been demonstrated in vivo (20, 21, 22). IGFs stimulate mitogenesis in astroglial and oligodendroglial cells and promote the survival and stimulate the outgrowth of neurons in vitro (23, 24, 25). Transgenic mice overexpressing IGF-I have brains larger than those of wild-type mice (26, 27), whereas mice lacking IGF-I have smaller brains (28). These differences in brain size can be attributed partly to hyper- or hypomyelination, respectively, providing additional evidence for the growth-promoting effects of IGF-I on glial cell types. Additionally, mice that overexpress IGFBP-1 showed brain growth retardation from the second week of postnatal life, suggesting that IGFBP-1 may inhibit the growth-promoting effects of IGF-I during brain development at this time (29).
IGFBP mRNAs have also been localized to developing regions of the rodent brain (13, 16, 30). In particular, IGFBP-2 is expressed in the developing rat brain in vivo, primarily in glial cell types (16, 17, 31). In addition, we and others (32, 33, 34) have demonstrated that cultured astroglial cells synthesize IGFBP-2 in vitro. IGFBP-2 as a modulator of IGF action may, therefore, have a significant role in the growth and differentiation of glial cells.
We have shown previously that the exogenous addition of IGFBP-2 purified from BRL-3A cells can inhibit the mitogenic action of IGFs on primary astroglial cultures, and that this inhibition is a result of competition for IGF binding to the receptor (32). However, exogenous addition of IGFBPs may not accurately reflect the true autocrine/paracrine nature of the IGF system, as there is increasing evidence demonstrating the interaction of IGFBPs with either extracellular matrix (8) or cellular components at the membrane level (6, 9). We therefore sought to create a system by which the endogenous expression of IGFBP-2 in glial cells is altered and to address the question of whether the overexpression of IGFBP-2 leads to an altered growth phenotype. The rat C6 glioma cell line was chosen as a model system because it synthesizes IGF-I and expresses IGF receptors (35), and its growth is regulated by IGF-I (36). Primary astroglial cells, which demonstrate regulated growth, synthesize high levels of IGFBP-2, whereas C6 glioma cells, with tumorigenic growth, express very low levels of IGFBP-2 (37). We have therefore transfected an ovine IGFBP-2 cDNA constructed in a constitutive expression vector, into the rat C6 glioma cell line and have demonstrated that expression of components of the IGF system were variably altered and that cell growth was conditionally altered.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Cell culture
Rat C6 glioma cells, obtained from the American Type Culture
Collection (a gift from Dr. C. C. G. Naus, University of
Western Ontario), were grown in DMEM supplemented with 10% FBS,
penicillin (5 U/ml), streptomycin (5 µg/ml), and gentamicin (25
µg/ml; Life Technologies) in 5% CO2-95%
air in humidified tissue culture incubators at 37 C. The medium was
changed at cell passage every 3–4 days.
Transfection
A 1.1-kb ovine (o) IGFBP-2 cDNA consisting of the entire coding
region (39) was cloned into the HindIII site of the
mammalian expression vector pRc/cytomegalovirus (CMV) (40). The
oIGFBP-2 cDNA in the antisense orientation and the pRc/CMV vector alone
were used as controls. The plasmid was transfected into C6 cells using
the polybrene/dimethylsulfoxide (DMSO) technique (41). Cells were
plated on 60-mm Falcon tissue culture grade petri dishes (Becton Dickinson and Co., Franklin Lakes, NJ) at a density of 5 x
106 cells/dish in complete medium (CM) and allowed
to attach overnight. The medium was aspirated, and 20 µg plasmid DNA
were equilibrated with 10 µg/ml polybrene in CM at 37 C and added to
the cells with periodic swirling for 16 h. The DNA mixture was
aspirated, and an equilibrated mixture of 15% DMSO in CM was added to
the cells for 4.5 min at 37 C. The DMSO mixture was removed, and the
cells were washed with CM (twice) and allowed to recover for 24 h.
Each 60-mm dish was subplated into 3 x 100-mm tissue grade petri
dishes (Falcon) and covered with selection medium (1.5 mg/ml G418 in
CM). Selection medium was changed every 4–5 days for a total selection
period of 14 days. After selection, the cells were grown in CM, and
discrete colonies were picked after 4 days.
CM were collected from confluent clones in T-75 flasks (Falcon) after 24 h of incubation in serum-free medium (SFM; medium supplemented with 0.1 mg/ml BSA; Sigma Chemical Co.) and centrifuged at 3000 x g to remove cellular debris, and the supernatant was stored at -20 C until analyzed. Cells from the same cultures were used for RNA extraction.
Ligand and immunoblot analyses
Ligand blot analysis was performed as previously described (42).
The same membrane was processed for immunoblotting. The membrane was
blocked with 4% BSA-TTBS (0.5% Tween in Tris-buffered saline) for
1 h at room temperature. The membrane was washed in TTBS (three
times, 10 min each time) followed by incubation in bovine IGFBP-2
antiserum (1:2000 dilution) in 1% BSA-TTBS overnight at 4 C. The
membrane was washed in TTBS (three times, 10 min each time) and
incubated with secondary antibody, biotinylated goat antirabbit IgG
(1:1000 dilution) in 1% BSA-TTBS, for 1 h at room temperature.
After washing in TTBS (three times, 10 min each times), the membrane
was incubated in ExtrAvidin-peroxidase (1:1000) in 1% BSA-TTBS for
1 h at room temperature. The membrane was washed in TTBS (twice,
10 min each time) and TBS (once, 10 min) and developed with the
chromagen diaminobenzadine (Sigma Chemical Co.) in 0.05
M Tris, pH 7.6. The membrane was washed in 0.05
M Tris, pH 7.6, for 1 h at room temperature,
air-dried, and exposed to x-ray film (Biomax, Eastman Kodak Co., Rochester, NY) with intensifying screens at -70 C for 3–7
days. Total protein concentrations of the CM were determined by the
Bio-Rad protein assay (Bio-Rad Laboratories, Inc.,
Richmond, CA).
Northern blot analysis
Total RNA was prepared from cell cultures by the guanidine
thiocyanate-cesium chloride method and subjected to Northern analysis
as previously described (42). The resulting blots on Zeta-Probe nylon
membranes (Bio-Rad Laboratories, Inc.) were probed in
sequence with 32P-labeled cDNAs encoding ovine IGFBP-2
(39); rat IGFBP-1 (a gift from Dr. M. Rechler, NIH, Bethesda, MD); rat
IGFBP-3 (a gift from Dr. A. Herington, Melbourne, Australia); rat
IGFBP-4, -5, and -6 (gifts from Dr. S. Shimasaki, San Diego, CA); rat
IGF-I (a gift from Dr. L. Murphy, Winnipeg, Canada); and mouse IGF-II
(a gift from Dr. G. Bell, Chicago, IL). The cDNA inserts were labeled
with [32P]deoxy-CTP (ICN Biomedicals, Inc.
Canada, St. Laurent, Canada) to specific activities of 1–2 x
109 cpm/µg by the random priming technique using the
Oligo-labeling kit (Pharmacia Canada, Inc., Baie d’Urfe, Canada). The
blots were stripped in between hybridizations by washing in 0.01
x SSC (standard saline citrate) and 0.5% SDS at 80 C for 30 min
twice. Consistency in the relative amounts of total RNAs being loaded
into each lane was checked by probing the blots with a radiolabeled
cDNA for 18S ribosomal RNA (a gift from Dr. D. Denhardt, Piscataway,
NJ).
IGF-I RIA
Conditioned media (5 ml) were lyophilized and solubilized in 1
M acetic acid at 4 C overnight, and then IGFs were
separated from IGFBPs using gel filtration chromatography on a Sephadex
G-50 (Pharmacia) column. Serial dilutions of extracted media were
assayed for IGF-I as previously reported (43).
Growth curves
Selected clones were plated at a density of 1 x
105 cells in T-25 tissue culture flasks (Falcon) in CM.
Cell number was determined in triplicate on the following day (time
zero) and every 24 h thereafter for a period of 72 h on a
Coulter counter (model Zf, Coulter Electronics, Hialeah,
FL). Growth curves were constructed by plotting mean cell number
against time. Each experiment was performed three times, and the
mean ± SEM doubling time was calculated. Statistical
analysis for comparing growth rates among different clones was
performed using ANOVA.
Affinity cross-linking
Selected clones were plated at a density of 3 x
105 cells in poly-L-lysine (0.05 mg/ml)-coated
six-well tissue culture plates in CM. Upon reaching confluence at
48 h, the cell monolayers were washed in HEPES binding buffer, pH
7.4 (0.1 M HEPES, 7.75 mM sodium phosphate
dibasic, 2.25 mM sodium phosphate monobasic, and 0.1%
BSA), and subsequently incubated with 5 x 105 cpm
[125I]IGF-I or -IGF-II with or without competitors
(IGF-I, IGF-II, insulin, heparin, GRGDSP, and GRGESP) in HEPES binding
buffer for 6 h at 4 C. The incubating solution was aspirated, the
cells were washed thoroughly with cold (4 C) PBS, and then the proteins
were cross-linked by incubation in 0.1 mM dissuccinimidyl
suberate in cross-linking buffer (0.1 M HEPES, 7.75
mM sodium phosphate dibasic, and 2.25 mM sodium
phosphate monobasic) for 30 min at room temperature. The cross-linking
solution was aspirated, and the proteins were solubilized by the
addition of 1 x Laemmli buffer. The samples were denatured by
boiling and were run on a 6–14% gradient SDS-PAGE. The gels were
fixed in 25% methanol-10% acetic acid in water, dried, and exposed to
Biomax (Eastman Kodak Co.) film for autoradiography.
Immunoprecipitation
Conditioned medium was incubated with 2 x 105
cpm [125I]IGF-II for 1 h at 4 C and subsequently
cross-linked with the addition of 10 mM dissuccinimidyl
suberate. Cell monolayers were cross-linked with
[125I]IGF-II as described above, solubilized in membrane
homogenization buffer (20 mM Tris, 0.33 M sucrose, 2
mM EDTA, 0.5 mM EGTA, 1% Nonidet P-40, 2
mM phenylmethylsulfonylfluoride, and 0.3% aprotinin, pH
7.5), and sonicated for three 10-sec bursts with a membrane sonicator.
The resulting cell suspensions and CM were incubated with antisera
against bovine IGFBP-2 (Upstate Biotechnology, Inc.) at a
1:50 dilution at 4 C overnight. Immune complexes were precipitated with
the addition of protein A-Sepharose (Pharmacia) and centrifuged, and
the pelleted immune complexes were washed twice with membrane
homogenization buffer. Protein complexes were boiled in 1 x
Laemmli buffer, boiled, and run on 6–14% gradient SDS-PAGE. The gels
were dried and exposed to Biomax film (Eastman Kodak Co.)
for autoradiography.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). The 40-
to 45-kDa IGFBP has been identified immunologically as IGFBP-3 (33, 34). The 22-kDa IGFBP, although not identified immunologically because
of the lack of a specific antiserum, was deduced to be IGFBP-4 based on
its molecular size and the detection of IGFBP-4 mRNA in Northern blots.
The identity of the 28-kDa IGFBP is presently unknown, but may be a
glycosylated variant of IGFBP-4 based on its molecular size and the
lack of detectable mRNAs for IGFBP-1, -5, or -6.
|
). Clones C6BP2-8, -10, -11, and -12 secreted
relatively high levels of this protein, whereas clones C6BP2-1, -4, -6,
-7, and -9 secreted lower levels. The identity of this protein was
confirmed by immunoblotting with a specific antiserum against bovine
IGFBP-2 (Fig. 1B). The levels of immunoreactive IGFBP-2 corresponded to
the levels of IGFBP-2 as observed by ligand blotting. The
vector-transfected control clone (C6CMV-3) did not express IGFBP-2 and
was chosen as an appropriate control over other vector control clones
because the IGFBP and IGF-I expression profile was most similar to that
of the wild-type C6 cells. The levels of the other IGFBPs in the various clones were also variably altered. IGFBP-3 was down-regulated, to a variable degree, in all clones. Clones C6BP2-7 and -9 had just detectable levels of IGFBP-3 protein, whereas clones C6BP2-8, -10, and -11 had no detectable levels. IGFBP-4 was also down-regulated in all clones except the vector control C6CMV-3. The 28-kDa IGFBP was detected in only clones C6CMV-3 and C6BP2-10. An additional IGFBP of 29 kDa was detected in clones C6BP2-1, -4, -7, and -9.
Expression of IGFBP genes
Expression of the transfected ovine IGFBP-2 was analyzed by
Northern blotting to determine steady-state mRNA levels (Fig. 2). IGFBP-2 mRNA was not detected in the
wild-type C6 glioma cells or the vector control clone, C6CMV-3. A
1.6-kb band was detected, at variable levels, in the IGFBP-2-secreting
clones. The relative levels of IGFBP-2 steady-state mRNAs
corresponded to the relative amount of secreted IGFBP-2 protein for
each clone. Clones C6BP2-8, -10, -11, and -12 expressed relatively high
levels of IGFBP-2 mRNA, whereas clones C6BP2-1, -4, -6, -7, and -9
expressed lower levels of IGFBP-2 mRNA.
|
). The relative amount of steady-state IGFBP-3
mRNA did not necessarily correspond to the amount of secreted IGFBP-3
protein, suggesting that the regulation of protein production occurred
at the translational or posttranslational level. IGFBP-3-specific
protease activity was not observed in any of the conditioned media,
indicating that this posttranslational processing was not responsible.
The relative levels of IGFBP-4 steady-state mRNA for each clone
corresponded to the levels of secreted IGFBP-4 protein. IGFBP-1 and -6
mRNAs were not detected by Northern analysis using 20 µg total
RNA/lane (data not shown). However, IGFBP-5 mRNA was detected only in
clones C6BP2-1 and C6BP2-4, but not in others, by Northern analysis
using 40 µg total RNA, indicating that in these clones, IGFBP-5 mRNA
was expressed in low abundance (Fig. 8).
|
). A major 7.5-kb transcript was
detected in addition to minor transcripts of 4.0, 2.0, and 1.2 kb. The
relative level of IGF-I steady-state mRNA was low in the wild-type C6
cells. Clones C6BP2-1, -4, and -9 and C6CMV-3 had similarly low levels
of IGF-I mRNA compared with the wild-type C6 cells. IGF-I steady-state
mRNA was up-regulated in clones C6BP2-6, -7, -8, -10, and -12 compared
with that in the wild-type C6 cells. The 7.5- and 1.2-kb IGF-I
transcripts were preferentially up-regulated in these clones. IGF-II
mRNA was not detected (not shown).
|
). The total IGF-I mRNA
levels in various clones correlated with IGF-I peptide concentrations
in the conditioned media (r2 = 0.86). In most clones,
except clones C6BP2-1 and -4, the concentration of secreted IGF-I
corresponded to the relative level of secreted IGFBP-2. Clones C6BP2-8
and -12, which secreted the highest levels of IGFBP-2, also
secreted the highest levels of IGF-I.
|
. Clones C6BP2-12 and -10, which highly
overexpressed IGFBP-2, exhibited a similar increase in cell number
over 72 h, compared with the wild-type C6 cells. In contrast, the
moderate expresser of IGFBP-2, clones C6BP2-1, -4, and -9, had
significant reductions in cell number at 48 and 72 h, compared
with that in the wild-type C6 cells.
|
). Clones C6CMV-3, C6BP2-10, and C6BP-12 had similar doubling times
as the wild-type C6 cells. In contrast, the moderate expressers of
IGFBP-2, clones C6BP2-1, -4, and -9, had significantly greater doubling
times than the C6 cells. Of significance, clones C6BP2-1, -4, and -9
also expressed very low levels of IGF-I mRNA and peptide, whereas
clones C6BP2-10 and -12 expressed high levels of IGF-I (Fig. 3 and
Table 1).
Affinity cross-linking of IGFs
Cross-linking of [125I]IGF-I or
[125I]IGF-II to monolayers of selected clones in the
absence or presence of variable concentrations of unlabeled IGF-I,
IGF-II, and insulin was used to examine relative IGF receptor binding
affinity. When IGF-I was used as the radioligand, all clones examined
displayed a 135-kDa band consistent in size with the IGF-I receptor
-subunit in addition to a 270-kDa band consistent in size with
-subunit dimers (Fig. 5, A–C).
[125I]IGF-I binding was competed with unlabeled IGF-I
(lanes 2–4), to a lesser degree with IGF-II (lanes 5–7), and with
very high concentrations of insulin (lanes 8 and 9). This
characteristic binding pattern of the IGF-I receptor was observed with
wild-type C6 cells (Fig. 5A) and all clones examined, C6BP2-1 (Fig. 5B)
and C6BP2-12 (Fig. 5C), C6CMV-3, and C6BP2-4, -9, and -10 (data not
shown). An additional band of 41 kDa was observed in C6BP2-1 (Fig. 5B)
and C6BP2-4 and C6BP2-9 (not shown). This band was competed with excess
IGF-I and IGF-II, but not with insulin, suggesting that it was a
membrane-associated IGFBP. IGFs competed more effectively for binding
to the 41-kDa membrane-associated IGFBP than for the IGF-I receptor,
suggesting that the membrane-associated IGFBP had a greater affinity
for IGF-I than the IGF-I receptor.
|
, D–F). This band was competed most effectively by
unlabeled IGF-II, to a lesser degree by IGF-I, and not at all by
insulin, consistent with the competition profile of the IGF-II
receptor. The binding competition profile of [125I]IGF-II
was similar in the wild-type C6 cells and in all clones examined
(C6CMV-3 and C6BP2-4, -9, and -10; not shown). The 41-kDa band was also
observed in C6BP2-1 (Fig. 5E) and C6BP2-4 and -9 (data not shown) when
[125I]IGF-II was used as radioligand. Its binding was
competed by unlabeled IGF-I and IGF-II, but not by insulin, again
suggesting its identity as a membrane-associated IGFBP. IGFs competed
more effectively for binding to the 41-kDa membrane-associated IGFBP
than for the IGF-II receptor, again suggesting that it had a greater
affinity for IGF-II than the IGF-II receptor.
Characterization of the cell-associated IGFBP
An antiserum against bovine IGFBP-2 was used to immunoprecipitate
[125I]IGF-II cross-linked proteins from conditioned media
and cell monolayers of selected clones (Fig. 6). In the conditioned media of C6BP2-12,
as expected, a large amount of [125I]IGF-II-IGFBP-2
cross-linked complexes were immunoprecipitated, whereas a lesser
amount from C6BP2-1 and -4 and a very small amount from C6 were
immunoprecipitated. The size of the [125I]IGF-II-IGFBP-2
immunoprecipitated complex from the conditioned media was slightly
larger than that of the [125I]IGF-II-IGFBP cross-linked
complex associated with the C6BP2-1 cells.
|
Ligand blot analysis demonstrated that the 29-kDa IGFBP found in
the conditioned media of clones C6BP2-1, -4, and -9, but not -12,
migrated at the same relative molecular size as IGFBP-5 produced by
FRTL-5 cells (Fig. 7). Total RNA from the
different clones was reexamined by Northern blotting for the expression
of IGFBP-1, -5, and -6 mRNAs. A 6.0-kb IGFBP-5 transcript
was just detectable in C6BP2-1 and -4 (Fig. 8). These data suggest that the 29-kDa
IGFBP found in the conditioned media and associated with the cell of
the slow growing clones C6BP2-1, -4, and -9 may be IGFBP-5.
Immunological characterization of the cell-associated IGFBP was
not possible due to the lack of reactivity of the commercially or
collaboratively available IGFBP-5 antiserum with various sources of rat
IGFBP-5 in Western blotting and immunoprecipitation.
|
). The expected 260-kDa type 2 receptor
was observed in both C6 cells and clone C6BP2-1. Without preincubation
with IGFBP-5, the 41-kDa cell-associated IGFBP was observed in clone
C6BP2-1, but not C6 cells. When IGFBP-5 was preincubated
for 24 h before cross-linking, a 41-kDa band was observed in C6
cells. Also, a less intense, smaller band of 38 kDa was observed on
both the C6 cells and clone C6BP2-1, with preincubation of
IGFBP-5, which may be a proteolytic fragment of IGFBP-5. These results
suggest that C6 glioma cells have the capacity to associate with
IGFBP-5.
|
). C6 and C6BP2-12 grew at similar
rates and were faster than C6BP2-1. Cell number at 72 h was
significantly greater for both C6 and C6BP2-12 than for C6BP2-1.
Addition of IGF-I to both C6 and C6BP2-12 resulted in a significant
increase in cell number at 72 h compared with their cell number in
1% FBS alone. In contrast, the addition of IGF-I to clone C6BP2-1 (or
clones C6BP2-4 or -9) did not result in any change in cell number at
72 h compared with 1% FBS alone. These observations suggest that
in these clones, factors other than the balance between IGF-I and
IGFBP-2, such as the cell-associated IGFBP-5, are the dominant
regulators of cell growth.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Not only was the level of IGFBP-2 synthesis by the various clones different, but other endogenous C6 IGFBPs were variably altered. The amounts of IGFBP-3 and IGFBP-4 detected in the conditioned medium were less than those observed with the wild-type C6 cells. Differences between steady-state mRNA levels and the amount of protein detected in the conditioned medium suggest that posttranscriptional and/or posttranslational processing may have occurred, such as proteolytic processing. Specific protease activity has been described for IGFBP-3 (44, 45) and for IGFBP-4 (46). However, in the cells studied, IGFBP proteases did not appear to play a role in the changes in IGFBP-3 and -4. An increase in the levels of serum IGFBP-3 and IGFBP-4 has been observed in mice lacking the IGF-II receptor gene (47). Additionally, mice lacking the IGFBP-2 gene also have increased serum levels of IGFBP-3 and IGFBP-4 (48). The changes in IGFBP expression in response to a perturbation in IGF system components may be a compensatory response designed to maintain an appropriate balance of growth-promoting and growth-inhibiting factors.
Based on the growth rates of selected clones, we divided the clones into two separate groups: those that grew at rates similar to the wild-type cells and those that grew about 30% slower than the wild-type cells. Interestingly, the clones that had very high levels of expression of IGFBP-2 grew at the same rate as the wild-type cells, whereas those that moderately overexpressed IGFBP-2 grew at slower rates. Those clones that highly overexpressed IGFBP-2 had compensatory up-regulation of IGF-I, which may account for the maintenance of a wild-type growth rate. This inference was supported by the fact that in those clones (e.g. C6BP2-12) that did not express the cell-associated IGFBP-5, the addition of exogenous IGF-I further increased the growth rates. The mechanism for this up-regulation of IGF-I expression is presently unknown. Regardless of the reason for up-regulation of IGF-I gene expression, it is important to note that this compensatory affect is associated with the maintenance of a normal (wild-type) growth rate, whereas the failure to do so results in impaired growth.
IGF-I has been shown to be an important regulator of C6 cell growth. Reduction of endogenous IGF-I production in C6 cells by transfection of an antisense cDNA, resulted in the loss of tumorigenicity of C6 cells in vivo (36). A similar effect was observed when expression of the IGF-I receptor was blocked by antisense oligonucleotides or cDNA in C6 cells (49). Monolayer growth was inhibited, and tumorigenicity was lost both in vitro and in vivo. Similar studies in other cell types have demonstrated that loss of expression of IGF-I or the IGF-I receptor leads to reduced growth and tumorigenicity (50, 51, 52). We have previously shown that when C6 cells are transfected with the gap junction cDNA connexin 43, the reduction in growth rate of the clones overexpressing the connexin 43 cDNA is also associated with a reduction in the level of expression of IGF-I (37). Our present results support the hypothesis that the interaction of IGF-I with its receptor is an important regulator of C6 cell growth.
Although the variable expression of IGF-I among the different clones could account for the differences in growth, we have observed that different growth patterns are associated with changes in the levels of IGF-I and IGFBPs and IGFBP cellular localization. Consistent with the observation of concomitant increased expression of IGF-I and IGFBP-2, is the hypothesis that IGFBP-2 is acting as a competitive inhibitor for IGF-I binding to its receptor. When changes in the balance between IGFBP-2 to IGF-I occurred, as in those clones with IGFBP-2 overexpression and IGF-I underexpression, growth rates were reduced significantly. However, the slow growing clones exhibited a cell-associated IGFBP in addition to the reduced levels of IGF-I. The cell-associated IGFBP displayed preferential affinity for IGFs compared with the receptors and therefore may be the primary determinant for their slow growth.
Immunoprecipitation of the cytosolic fraction with the IGFBP-2 antiserum yielded faint bands of a smaller size compared with those obtained using IGFBP-2 immunoprecipitated from the conditioned medium. As the antiserum is known to have some cross-reactivity with other IGFBPs, it is possible that the immunoprecipitated cell-associated IGFBP may be another IGFBP. The slow growing clones were found to express IGFBP-5 mRNA and have a secreted IGFBP that corresponded in size to a known IGFBP-5 standard. Preincubation with IGFBP-5 before cross-linking resulted in increased levels of the cell-associated IGFBP in the slow growing clone and the wild-type C6 cells. In addition, coincubation with heparin, but not RGD peptides, reduced cell association of the IGFBP, which strongly indicates that the cell-associated IGFBP is IGFBP-5 (8). This was confirmed by the demonstration of a cell-associated IGFBP after incubation with C6 cells. The association can occur either with the cell surface or the extracellular matrix. The appearance of a second smaller cross-linked species may be due to the proteolytic processing of IGFBP-5. These collective results provide evidence to suggest that the cell-associated IGFBP is IGFBP-5. Whether the IGFBP-5 exerts an IGF-dependent or -independent biological action (4) remains to be delineated.
The observation that IGF-I further stimulated the growth of C6 and C6BP2-12, but not C6BP2-1, supports our hypothesis that the cell-associated IGFBP is acting to inhibit the growth of these clones by preventing the interaction of IGF-I with its receptor. Similar effects have been observed with IGFs and cell-associated IGFBPs in different cell systems. Fibroblasts transformed by simian virus 40, which have tumorigenic growth, displayed an increase in expression of IGF-I and loss of cell-associated IGFBP-5. In untransformed cells, which have normal growth, IGF-I binding was principally to cell-associated IGFBP-5, which inhibited its interaction with the type 1 receptor (53). Additionally, IGFs were found to bind predominantly to a cell-associated IGFBP-2, which was identified on a small cell lung tumor cell line (54). These cells did not respond to IGFs in a DNA synthesis assay, suggesting that the cell-associated IGFBP-2 inhibited IGF action by competing with IGF receptors for IGF binding.
A decrease in IGF-I receptors or a reduction in IGF binding affinity may also explain the reduction in growth of C6BP2-1, -4, and -9. However, this did not appear to be the case in the cells studied because IGF receptor binding affinity was similar in all of the clones, but the affinity cross-linking studies indicate semiquantitatively that there was not an alteration in IGF-I receptor binding. An accurate quantification of IGF receptor number was not possible due to the presence of the cell-associated IGFBP on some clones. A reduction in IGF receptor number could also account for the lack of growth response of C6BP2-1 to exogenous IGF-I. It is possible that expression of other endogenous growth factors was altered by the transfection process, and this may also account for the observed changes in growth; however, this was not examined.
The exact mechanism(s) controlling the growth of the various C6BP2 clones remains to be determined; however, it is clear that the interactions between IGF-I and IGFBPs remain an important determinant of their growth. This study suggests that glial cells may have a limited capacity to modulate various components of the IGF system to maintain normal growth, as evidenced by the concomitant up-regulation of IGF-I and down-regulation of endogenous IGFBPs among various clones. It also highlights the importance of examining all components of the IGF system before formulating conclusions on their role in cellular growth. The growth of astroglial cells, like that of many other cells, is regulated by a balance between growth-promoting and growth-inhibiting factors. Therefore, a decrease in a growth factor, IGF-I, or an increase in a growth inhibitory factor, IGFBP-2, should result in decreased growth. However, this study suggests that the regulation of growth of glial cells is complex, and a coordinated interaction between IGFs and IGFBPs, and specific properties of the latter (i.e. cell association), are required to determine the growth of these cells. Such a mechanism of regulation of glial cell growth most likely exists in the developing brain.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Recipient of a Medical Research Council Studentship.
Received July 23, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
5ß1
integrin by means of its Arg-Gly-Asp sequence. Proc Natl Acad Sci USA 90:10553–10557This article has been cited by other articles:
 |
|
 |
|
  C. Iosef, T. Gkourasas, C. Y. H. Jia, S. S.-C. Li, and V. K. M. Han A Functional Nuclear Localization Signal in Insulin-Like Growth Factor Binding Protein-6 Mediates Its Nuclear Import Endocrinology, March 1, 2008; 149(3): 1214 - 1226. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. L. Martin and R. C. Baxter Expression of Insulin-Like Growth Factor Binding Protein-2 by MCF-7 Breast Cancer Cells Is Regulated through the Phosphatidylinositol 3-Kinase/AKT/Mammalian Target of Rapamycin Pathway Endocrinology, May 1, 2007; 148(5): 2532 - 2541. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. S. Watson, P. Bialek, M. Anzo, J. Khosravi, S.-P. Yee, and V. K. M. Han Elevated Circulating Insulin-Like Growth Factor Binding Protein-1 Is Sufficient to Cause Fetal Growth Restriction Endocrinology, March 1, 2006; 147(3): 1175 - 1186. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Hoeflich, R. Reisinger, H. Lahm, W. Kiess, W. F. Blum, H. J. Kolb, M. M. Weber, and E. Wolf Insulin-like Growth Factor-binding Protein 2 in Tumorigenesis: Protector or Promoter? Cancer Res., December 1, 2001; 61(24): 8601 - 8610. [Full Text] [PDF]
|
|
|
|
|
|
L. Wang, X. Ma, L.-C. C. Yeh, and M. L. Adamo Differential Regulation of IGF-Binding Protein Gene Expression by cAMP in Rat C6 Glioma Cells Endocrinology, September 1, 2001; 142(9): 3917 - 3925. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Hoeflich, S. Nedbal, W. F. Blum, M. Erhard, H. Lahm, G. Brem, H. J. Kolb, R. Wanke, and E. Wolf Growth Inhibition in Giant Growth Hormone Transgenic Mice by Overexpression of Insulin-Like Growth Factor-Binding Protein-2 Endocrinology, May 1, 2001; 142(5): 1889 - 1898. [Abstract] [Full Text]
|
|
|
|
|
|
M. W. Elmlinger, M. H. Deininger, B. S. Schuett, R. Meyermann, F. Duffner, E. H. Grote, and M. B. Ranke In Vivo Expression of Insulin-Like Growth Factor-Binding Protein-2 in Human Gliomas Increases with the Tumor Grade Endocrinology, April 1, 2001; 142(4): 1652 - 1658. [Abstract] [Full Text]
|
|
|
|
 |
|
 C. Duan, J. Ding, Q. Li, W. Tsai, and K. Pozios Insulin-like growth factor binding protein 2 is a growth inhibitory protein conserved in zebrafish PNAS, December 21, 1999; 96(26): 15274 - 15279. [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 |
), C6BP2-1 (
), and
C6BP2-12 (
) in medium containing 1% FBS with (filled
symbols) or without (open symbols) the daily
addition of IGF-I (200 ng/ml). Statistically significant increases in
cell number at 72 h with the addition of IGF-I compared with those
without IGF-I for the same clones are designated by * and 
(P < 0.05).