| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
From The Centre for Hormone Research (V.C.R., N.L.B., G.A.W.), Royal Children’s Hospital, Parkville 3052, Victoria, Australia; University of Melbourne Department of Medicine (L.A.B.), Austin and Repatriation Medical Centre, Heidelberg 3084 and Department of Paediatrics (A.J.F.), Orthopaedic Molecular Biology Research Unit, Royal Children’s Hospital, Parkville 3052, Victoria, Australia
Address all correspondence and requests for reprints to: Assoc. Prof. George A. Werther, Centre for Hormone Research, Royal Children’s Hospital, Flemington Road, Parkville, Victoria 3052, Australia. E-mail: Werther{at}cryptic.rch.unimelb.edu.au
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Recent studies have demonstrated association of some IGFBPs with the ECM or cell surface via glycoproteins, collagens, integrins (3, 4), and glycosaminoglycans (5, 6, 7, 8, 9). These studies showed that IGFBP-3 when cell-associated (7) or IGFBP-5 when ECM-bound (3) had lowered binding affinity for IGF-I. Additionally, the formation of IGF-I/IGFBP-5 complexes can be inhibited by heparin by altering the binding affinity of IGFBP-5 for IGF-I (5). More recently, it has been shown that IGFBP-2/IGF complexes can bind to heparin and extracellular matrix (10). These in vitro findings support the hypothesis that IGFBPs, in addition to stabilizing and regulating levels of diffusible IGFs, might regulate IGF-I cellular responses by facilitating receptor targeting of IGF-I or modulating IGF-I bioavilability in the pericellular space in vivo (2, 9).
The sites of expression of messenger RNA for IGFs and IGFBPs have been precisely located in rat brain (11, 12, 13, 14). The olfactory bulb, a brain region undergoing postnatal differentiation and remodeling, is rich in locally expressed components of the IGF system including IGF-I, IGF-I receptor, and IGFBPs, most abundantly IGFBP-2 (15, 16, 14), which we have further characterized in vitro (14, 17).
Although brain-derived IGFBPs have also been characterized (14, 18), very little is known about their functional role in the nervous system. For example, it is not known whether interactions between IGFs and IGFBPs occurring in the extracellular space are further modulated by other biomolecules in the extracellular environment. We therefore explored whether any of the IGFBPs were associated with cell membrane or extracellular matrix components in the rat brain.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
2000 Ci/mmol) and 14C protein
molecular weight markers were bought from Amersham (North Ryde, New
South Wales, Australia). Disuccinimidyl Suberate (DSS) was obtained
from Pierce (Rockford, IL). Chemical reagents (Analar and HPLC grade)
were purchased from BDH-Merck Pty Ltd (Kilsyth, Victoria, Australia).
ECL immunodetection kit, phenylmethylsulfonylfluoride (PMSF), Triton
X-100, proteinase inhibitors E-64 and pepstatin were purchased from
Boehringer (Mannheim, Germany). Heparin, aminoalkilsilane, aprotinin,
RIA grade BSA, and protein A-Sepharose CL-4B were purchased from Sigma
Chemical Co. (St. Louis, MO). GRGDSP and GRGESP peptides were purchased
from Auspep (Parkville, Victoria, Australia). Nitrocellulose membranes
were obtained from Schleicher and Schuell (Dassel, Germany). X-Omat AR
and Biomax films were obtained from Kodak (Eastman Kodak Company, NY).
Microtiter immunoassay plates (Immunolon-4) were purchased from
Dynatech Laboratories Inc. (Sullyfield Circle, Chantilly, VA).
[4-(2-aminoethyl)benzene]sulfonyl-fluoride (AEBSF) was from
Calbiochem-Novabiochem Pty Ltd (Alexandria, New South Wales,
Australia). PBS was from Histo Labs (New South Wales, Australia) and
Ilford K5 nuclear emulsion was from Ilford Limited (Mobberley, UK).
Brain tissue samples
One day postnatal Sprague-Dawley rats were killed by
decapitation and whole brains or olfactory bulbs (OB) were obtained as
previously described (24). Dissected whole brains were immediately snap
frozen in dry ice/isopentane and stored at -70 C until tissue
sectioning by cryostat. All procedures involving dissection of animals
were approved by the Royal Children’s Hospital Animal Experimentation
Ethics Committee.
125I-IGF-I in vitro autoradiography
The in vitro autoradiography protocol was adapted
from that previously described by Werther et al. (15).
Briefly, 20 µm parasaggittal whole brain and coronal OB frozen
sections were cut by cryostat, allowed to dry, and stored at -70 C
until used. Slides were rehydrated at RT for 1 h with 2 ml of 10
mM Tris/HCl, pH 7.4, 0.1% BSA, 2 mM
MgCl2, 2 mM CaCl2, changing the
buffer every 10 min. Sections were then incubated at 4 C for 16 h
with the same buffer containing 125I IGF-I ( 30000 cpm)
in the presence or absence of 1 µg/ml of IGF-I or des(1, 2, 3)IGF-I.
Sections were washed four times for 5 min with 2 ml of ice-cold buffer.
Slides were air dried and exposed to x-ray film for 1–5 days and then
dipped in Ilford K5 nuclear emulsion and stored with dessicant at 4 C.
Emulsion was developed in Kodak D19, fixed in Ilford Hypam, and the
sections stained with hematoxylin.
OB membrane preparation
Olfactory bulb membranes were obtained by modification of a
previously described method (25). Olfactory bulbs, fresh or from a
frozen stock stored at -70 C, were resuspended in ice-cold buffer (10
mM Tris/HCl, 2 mM PMSF, 1 TIU/ml of Aprotinin)
and mechanically disaggregated through 19 gauge and 23 gauge needle
syringes. The tissue suspension was centrifuged at 500 rpm for 5 min at
4 C, and the supernatant was centrifuged at 14,000 rpm for 1 h at
4 C. The pellet, called the membrane fraction (MF), was resuspended in
ice-cold 10 mM Tris-HCl, pH 7.4. Aliquots of MF were
adjusted to a total protein concentration of 100 µg/80 µl, followed
by addition of 0.1% BSA and storage at -20 C until used.
125I-IGF-I binding and cross-linking to OB
membranes
OB membranes (100 µg/80 µl) were incubated in a final volume
of 100 µl with 125I- IGF-I (40,000 cpm/tube) in the
presence or absence of the following competitors: 1 µg/ml IGF-I,
IGF-II, des(1, 2, 3)-IGF-I or 10 µg/ml of insulin for 2 h at 37 C
with rotation. Samples were then incubated on ice for 10 min followed
by addition of DSS to a final concentration of 1 mM and
further incubated for 15 min at 4 C. The cross-linking reaction was
quenched by the addition of 45 µl of ice-cold 100 mM
Tris, pH 7.4/10 mM EDTA. Samples were analyzed by
denaturing SDS-PAGE (3–16% gradient gel) under reducing conditions
where indicated. Gels were fixed (50% methanol/10% acetic acid),
stained with Coomassie blue and dried before exposure to x-ray film for
3–10 days.
Western ligand blot (WLB) analysis
Olfactory bulb membranes were solubilized in nonreducing sample
buffer and electrophoresed by SDS-PAGE. Adult rat serum (5 µl) and 5
µl of nonreduced 14C protein MW markers were run in
parallel lanes as necessary. Separated proteins were transferred to
nitrocellulose filters and WLB of the transferred protein was carried
out according to the method of Hossenlopp et al. (26) using
125I-IGF-I (1.5 x 106 cpm/50 ml). Dried
filters were exposed to x-ray film for 3–10 days.
Immunoprecipitation
Immunoprecipitation of IGFBP-2 and IGFBP–5 from fresh OB
membranes or OB membranes cross-linked with 125I-IGF-I was
performed as previously described (14). Anti-IGFBP-2, anti-IGFBP-5, or
normal rabbit serum were used at a final dilution of 1:100. The
immunoprecipitated samples derived from fresh OB membranes were
dissolved in Laemmli sample buffer, boiled for 5 min, run on 12%
SDS-PAGE or 3–16% gradient gels, and WLB performed as described
above. The immunoprecipitated samples derived from cross-linked OB
membranes were processed as described for the cross-linking analysis.
Pure IGFBP-2 (10 ng) or cross-linked 125I-IGF-I/IGFBP-2
complex were also immunoprecipitated with both anti-IGFBP-2 and IGFBP-5
antisera and used as controls.
Immunohistochemistry
Olfactory bulbs, obtained as above, were fixed in 4%
paraformaldehyde in PBS then processed through graded ethanol and
xylene and infiltrated with paraffin wax. Sections (10 µm) were cut
and mounted on aminoalkylsilane-coated slides. Dewaxed and PBS
equilibrated sections were incubated at 4 C overnight with a 1:400
dilution of the rabbit anti-IGFBP-2 antiserum or with normal rabbit
serum (1:400) as a control. IGFBP-2 immunoreactivity was detected with
a Vectastain ABC kit according to the manufacturer’s instruction.
Sections were counter stained with hematoxylin and coverslipped.
Dissociation of endogenous IGFBP-2 from OB membranes by RGD/RGE
peptides or ionic strength
OB membranes (100 µg/80 µl) were incubated in 120 µl of 10
mM Tris-HCl, pH 7.4, 0.1% BSA with or without: GRGDSP
(10–100 µg/ml), GRGESP (10–100 µg/ml) or NaCl (150–500
mM). Samples were incubated for 16 h at 4 C with
gentle rotation. Tubes were centrifuged at 15,000 rpm for 30 min at 4 C
and the pellet was washed, spun, and resuspended in 80 µl of 10
mM Tris-HCl, pH 7.4, 0.1% BSA before incubation with
125I-IGF-I in the presence or absence of 1 µg/ml of IGF-I
(for 120 min at 37 C). Samples were cross-linked and electrophoresed as
described above. Gels were fixed, stained, dried, and exposed to x-ray
film for 3–10 days.
Binding of IGFBP-2 to proteoglycan and glycosaminoglycans
The procedure for binding of IGFBP-2 to proteoglycan and
glycos-aminoglycans was adapted from Bonaldo et al.
(27). Immulon-4 96 well-plates were coated with 500 ng of aggrecan (a
proteoglycan containing both chondroitin sulfate and keratan sulfate)
or with the glycosaminoglycans chondroitin-4-sulfate,
chondroitin-6-sulfate, keratan sulfate or heparin in 200 µl of PBS
(16 h, 37 C). To remove excess salt and unbound reagents, wells were
washed 4 times for 5 min with 250 µl of binding buffer (BB) [16
mM Tris-HCl, pH 7.2, 50 mM NaCl, 2
mM CaCl2, 2 mM MgCl2, 1
mg/ml BSA, 0.02% Tween 20]. Wells were then blocked with BSA (250
µl of 1% BSA in 16 mM Tris-HCl, pH 7.2, 50
mM NaCl) for 1 h at 37 C, washed four times with BB,
then incubated at 37 C for 60 min with 200 µl of BB in the presence
or absence of 10 ng of purified IGFBP-2. IGFBP-2 was also bound to
aggrecan previously digested with chondroitinase ABC (0.25 U/mg) and
keratanase (0.03 U/mg) in the presence of protease inhibitors (10
mM EDTA, 20 µg/ml E-64, 1 mM AEBSF and 2
µM pepstatin) at 37 C for 3 h. Unbound IGFBP-2 was
removed by washing, and wells were incubated at 4 C for 16 h with
125I-IGF-I (3 x 104 cpm/200 µl or
6 x 104 cpm/200 µl as indicated) in the presence
or absence of unlabeled IGF-I (0.5 µg/ml). Wells were then washed
four times with BB. BSA coated wells were used as a control. Bound cpm
were solubilized with 250 µl of 200 mM NaOH/0.1% Triton
X-100 and radioactivity was determined by 
-counting. Experiments
were performed three times, and each point was measured in triplicate
or quadruplicate as indicated.
Determination of the binding affinities of IGF-I for
glycosaminoglycan associated and soluble IGFBP-2
Glycosaminoglycan associated.
Microtiter 96-well plates were
coated with 500 ng/200 µl of chondroitin-6-sulfate(C-6S), saturated
with 1% BSA, and incubated with IGFBP-2 (10 ng/well) as described
above. Following removal of unbound IGFBP-2, wells were incubated for
16 h at 4 C with 125I-IGF-I (3 x 104
cpm) in the presence or absence of increasing concentrations (0.006–6
nM) of unlabeled IGF-I. Nonspecific binding was measured in
wells coated with C-6S without added IGFBP-2, and wells coated with BSA
with (10 ng/well) or without IGFBP-2. Nonspecific binding was less than
1.5% of the added radioactivity for all these conditions. All wells
were then washed four times with BB and bound radioactivity was
measured in a -counter. Each point was measured in quadruplicates in
each of three experiments. Data were analyzed by the LIGAND program
(28).
Solution binding assays (29).
Rat IGFBP-2 (2 ng) was
incubated with 125I-IGF-I (13 000 cpm) with and without
increasing concentrations of unlabeled IGF-I (0.006–6 nM)
for 16 h at 4 C in binding buffer as described above (final volume
0.4 ml). Bound and free 125I-IGF-I were separated by
incubation with 0.5 ml ice-cold 5% charcoal/2% fatty-acid free BSA in
Dulbecco’s PBS for 10 min on ice and centrifugation at 1300 x g
(4 C for 30 min). Bound radioactivity in supernatants was quantitated
in a -counter. Each point was measured in triplicate in each of two
experiments. Binding affinities were calculated using the LIGAND
program (28).
Western immunoblot analysis with antichondroitin sulfate
antibodies
The monoclonal antibodies 3-B-3 and 2-B-6 recognize chondroitin
sulfate chains containing a nonreducing terminal saturated or
-4–5-unsaturated glucuronic acid residue adjacent to
N-acetylgalactosamine-6-sulfate or
N-acetylgalactosamine-4-sulfate, respectively (21, 30).
Because very few chondroitin sulfate chains terminate with these
structures, pretreatment of the membrane with chondroitinase-ABC (0.01
U/ml for 3 h at 23 C in 50 mM sodium acetate pH 7.4)
is required to remove the terminal saccharides and expose internal
epitopes. Monoclonal antibodies 3-B-3 and 2-B-6 together, each diluted
1:2,000 in 1% skim milk powder in PBS were incubated on the membranes
overnight at 4 C. Antimouse IgM/HRP was used at a dilution of 1:7,500.
Immunoreactivity was detected using the ECL kit according to the
manufacturer’s instructions.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
).
In contrast, des(1, 2, 3)IGF-I, an IGF-I analog which has conserved
binding affinity for the IGF-I receptor (29) but has greatly reduced
affinity for all the IGFBPs (31) displaced binding throughout the brain
except in the OB, the cerebellum and possibly hippocampus (Fig. 1b). In
the OB, this noncompetible binding was localized to the mitral cell
layer (MI), which is a layer of high density binding for
125I-IGF-I (15). Because IGFBPs, particularly IGFBP-2
and IGFBP-5, are also synthesized in this region (11, 12, 13, 14), we
hypothesized that cell-associated IGFBPs could account for the
nonreceptor binding of IGF-I in this brain region.
|
110
and >200 kDa), autoradiography also revealed a band with apparent
Mr of 38 kDa (Fig. 2, lane
a). Although des(1, 2, 3)-IGF-I was equipotent to IGF-I and IGF-II in
competing for 125I-IGF-I binding to the IGF-I receptor
(bands at 100 kDa and >200 kDa) (lanes b, c, e), the 38 kDa band
was abolished by coincubation with an excess of cold IGF-I or II (lanes
b and c, respectively) but not by des(1, 2, 3)IGF-I (lanes e) or insulin
(lanes d). These findings are consistent with the presence of a
membrane-associated IGFBP/IGF-I complex.
|
, lane 2). This band was
barely detectable by WLB on membranes obtained from whole brain (data
not shown). The size of this band is consistent with either IGFBP-2 or
IGFBP-5, both of which are expressed in this region of the brain (14).
However, an antiserum against IGFBP-2 (Fig. 4A, lane b) but not IGFBP-5 (Fig. 4A, lane c) immunoprecipitated a 32-kDa band from solubilized OB membranes.
Similarly, a 38-kDa complex was immunoprecipitated from cross-linked OB
membrane (Fig. 4B, lane b) or cross-linked purified IGFBP-2 (Fig. 4B, lane d) by the IGFBP-2 antiserum but not by anti-IGFBP-5 antiserum
(Fig. 4B, lane c).
|
|
), WLB (Fig. 3), and immunoprecipitation (Fig. 4) show
that IGFBP-2 is cell membrane associated in the OB and supports the
hypothesis that it may account for the nonreceptor-IGF-I binding shown
by in vitro autoradiography (Fig. 1B). We therefore
performed immunohistochemistry using the same anti-IGFBP-2 antiserum on
OB paraffin sections and localized IGFBP-2 to cells in the mitral cell
layer (MI) of the OB (Fig. 5B). When
emulsion autoradiography was performed on the section shown in Fig. 1b following coincubation with
des(1, 2, 3)IGF-I and 125I-IGF-I, residual binding indicated
by silver grains was similarly located in the mitral cell layer (Fig. 5A). These findings strongly suggest that the membrane-associated IGFBP
in the OB in vivo is IGFBP-2.
|
5ß1 integrin (4). A potential heparin binding motif (PKKLRP)
is also present in the intercysteine rich domain of IGFBP-2 (aa
160–165), and a similar glycosaminoglycan binding motif is present in
IGFBP-3 and -5 (33), which may mediate their association with
heparin.
We therefore first investigated whether the RGD sequence of IGFBP-2 was
involved in membrane binding. A GRGDSP hexa-peptide and a GRGESP
hexa-peptide (used as control), both at 10–100 µg/ml, failed to
dissociate IGFBP-2 from OB membranes (Fig. 6A), suggesting that the RGD sequence was
not involved in cell membrane association of IGFBP-2 in this system. In
contrast, NaCl dissociated IGFBP-2 from the OB membrane in a dose
dependent manner (Fig. 6B), and dissociation was complete with 500
mM NaCl. These findings suggested that IGFBP-2 may bind to
cell membranes via ionic interactions and raise the possibility that
proteoglycans may be involved.
|
).
125I-IGF-I in the absence of IGFBP-2 bound minimally to
glycosaminoglycans (Fig. 7A) and IGFBP-2 did not bind to wells coated
with BSA. IGFBP-2 was also able to bind to the proteoglycan aggrecan
(Fig. 7, A and B), which contains both chondroitin sulfate and keratan
sulfate chains and shares homologies and functional domains with some
of the brain chondroitin sulfate proteoglycans (CSPGs) (35). IGFBP-2
binding to aggrecan was reduced following digestion with chondroitinase
ABC or keratanase (Fig. 7B), suggesting specific interaction of IGFBP-2
with the glycosaminoglycan side chains of this proteoglycan. Thus,
IGFBP-2 binds glycosaminoglycans and, further, IGF-I efficiently binds
to IGFBP-2 associated with glycosaminoglycans, suggesting that IGF-I
might bind to both receptors and glycosaminoglycan-associated IGFBP-2
in vivo.
|
, open
squares) was 5.8 ± 1.1 x 109
M–1 (mean ± SEM), whereas
that for soluble IGFBP-2 (Fig. 8, filled squares) was
15 ± 2.0 x 109 M-1,
indicating that association of IGFBP-2 with glycosaminoglycans
decreased the binding affinity of IGFBP-2 for IGF-I by approximately
3-fold.
|
). The apparent molecular size of these
bands, before and after treatment with chondroitinase ABC, was greater
than 200 kDa (Fig. 9). This finding is consistent with the studies of
Gonzalez et al., who showed that proteoglycans
immunoreactive with 3B3 were prominent in extracts from OB tissue
(34).
|
). No
immunoreactivity was present in the control lane which was
immunoprecipitated with nonimmune serum. WLB analysis of the same
samples confirmed that the anti-IGFBP-2 antiserum, but not the
nonimmune antiserum, was able to specifically immunoprecipitate IGFBP-2
(Fig. 10B). This experiment shows unequivocally that IGFBP-2 binds to
cell surface chondroitin sulfate proteoglycans in rat OB and so
suggests a role for proteoglycans in modulation of IGF-I bioactivity
in vivo.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Like IGFBP-1, the amino acid sequence of IGFBP-2 contains an
Arg-Gly-Asp sequence, which is implicated in binding of extracellular
matrix components to integrins, near its carboxyl-terminus (1). IGFBP-1
binds to the 5ß1 integrin through this
motif, and this interaction mediates IGF-independent migration of CHO
cells (4). However, an Arg-Gly-Asp-containing peptide did not compete
for binding of IGFBP-2 to membranes in the present study, making
binding to integrins an unlikely mechanism for the membrane association
of this IGFBP.
The findings in the present study indicate that IGFBP-2 binds to the glycosaminoglycan component of membrane proteoglycans. Binding of proteins to the glycosaminoglycan component of proteoglycans is often charge-dependent, although more specific, higher affinity interactions may also occur (37). Because charge-dependent binding to glycosaminoglycans is inhibited by increasing ionic strength, the inhibition of cell-association of IGFBP-2 in a dose-dependent manner by NaCl indirectly supports the hypothesis that IGFBP-2 was binding to glycosaminoglycans. In vitro, IGFBP-2 bound to chondroitin-4-sulfate, chondroitin-6-sulfate, keratan sulfate, heparin and the proteoglycan aggrecan, the latter of which was blunted by specific digestion of component glycosaminoglycans. Although the structural basis of the binding to glycosaminoglycans was not addressed in this study, it is possible that the putative heparin-binding motif (Pro-Lys-Lys-Leu-Arg-Pro) at positions 160–165 in the mid-region of IGFBP-2 (33) may be involved.
Three proteoglycan core proteins each containing chondroitin-6-sulfate glycosaminoglycan have been extracted and partially characterized from extracellular matrix of the rat olfactory bulb, and immunologically localized to the mitral and plexiform layers (34). In the present study, we show, for the first time, that a large proteoglycan (>200 kDa) containing chondroitin sulfate chains, is able to bind IGFBP-2. We provided definitive evidence of IGFBP-2 binding to a glycosaminoglycan-containing proteoglycan in vivo in the rat olfactory bulb by showing that a 3B3/2B6 immunoreactive proteoglycan of similar size (>200 kDa) was coimmunoprecipitated with IGFBP-2 by the anti-IGFBP-2 antiserum. Whether this is a matrix proteoglycan as described above by Gonzales et al. (34) or a distinct membrane proteoglycan such as the 280 kDa chondroitin sulfate proteoglycan identified in chicken brain neurons (38), is not known. However, the precise composition of proteoglycans in this region may underlie the specific localization of membrane-associated IGFBP-2 in rat brain.
Both IGFBP-3 and IGFBP-5 bind to membranes and/or ECM from different cells, and salt and heparin displacement suggest that binding to glycosaminoglycan may be involved (3, 5, 6, 39, 40). Whether these IGFBPs bind directly to glycosaminoglycans, however, is controversial because heparinase (which digests the carbohydrate component) or chlorate (which inhibits glycosaminoglycan sulfation) variably affect IGFBP-3 and IGFBP-5 binding (39, 41). Glycosaminoglycans compete for binding of IGFBP-3 and IGFBP-5 to cell monolayers (6, 42) and decrease the binding affinity of IGFBP-5 for IGF-I (5). The amino acid sequences of IGFBP-3 and IGFBP-5 both contain putative heparin binding sites, and peptides based on these binding sites are effective competitors of membrane binding by these IGFBPs (5, 39). The specific basic amino acids that are necessary for heparin binding by IGFBP-5 have recently been identified by site-directed mutagenesis (41). It would therefore appear that IGFBP-3 and IGFBP-5 interact both with proteins and proteoglycans on the cell surface and in ECM. Whether binding to glycosaminoglycans is direct is unclear; this may depend upon cell type and specific conditions of study. Similar to these IGFBPs, IGFBP-2 binds to heparin-agarose (2), heparin sepharose, and extracellular matrix proteins (10) in vitro, but lower salt concentrations are needed to displace IGFBP-2 from heparin and ECM (2) and the presence of IGF-I or IGF-II is required to stabilize binding to heparin-sepharose and ECM proteins (10). There are differences in the interactions reported for IGFBP–2, -3, and -5 and those which we describe for IGFBP-2. In the present study, IGFBP-2 bound specifically to chondroitin 4-sulfate, chondroitin 6-sulfate, and keratan sulfate, glycosaminoglycans that interact minimally with IGFBP-5 or IGFBP-3 (39, 41). Furthermore, these glycosaminoglycans do not dissociate IGFBP-2/IGF-I complexes from heparin-sepharose (10). In addition, we found that pure rat IGFBP-2 binds in vitro to heparin and these glycosaminoglycans in the absence of IGF-I which was only added subsequently in radiolabeled form to detect glycosaminoglycan-bound IGFBP-2. These differences in our findings compared with those of Arai et al. (10) might reflect the inherent differences in our solid phase binding, compared with solution binding assay, or may be due to the different species of IGFBP-2 used. However, because the rat OB is a brain region rich in IGF-I (11, 14, 16), we cannot exclude the possibility that association of IGFBP-2 to an OB membrane chondroitin sulfate proteoglycan might require the presence of IGF-I.
Cell-association of IGFBPs is of particular interest because it may influence the way in which IGF action is modulated by IGFBPs (1, 2, 43). In many situations, IGFBPs inhibit IGF activity, but they may also enhance IGF activity. The mechanism of potentiation of IGF activity by IGFBPs is not completely understood, but much attention has focused recently on cell-association as a contributing factor. Some growth factors, notably basic fibroblast growth factor, bind to cell membrane and ECM heparan sulfate proteoglycans, and this interaction is important for subsequent binding to specific high affinity receptors (43). Additionally, basic FGF is protected from proteolysis while bound to proteoglycans. In contrast, IGFs do not substantially bind to proteoglycans or other ECM components. Cell-associated IGFBPs may therefore act as linker molecules allowing pericellular sequestration of IGFs.
The binding affinities of cell-associated and heparin-associated
IGFBP-3 (5, 39) and IGFBP-5 (5, 44) for IGFs are lower than those of
soluble IGFBPs. It has been postulated that this decrease in binding
affinity facilitates release of the IGFs from IGFBPs for binding by
receptors. Similarly, in the present study, glycosaminoglycan-bound
IGFBP-2 had reduced (3-fold) binding affinity for IGF-I compared
with soluble IGFBP-2. In the case of IGFBP-3, cell association leads to
a 10-fold decrease in affinity for IGF-I, related to processing of the
IGFBP to lower molecular weight forms (7). Whether a similar process
may occur with cell association of IGFBP-2 in vivo, further
reducing its affinity for IGF-I, was not examined in the current
study.
In conclusion, we have shown that IGFBP-2 is membrane-associated in the mitral cell layer of rat olfactory bulb, at least in part by interacting with the glycosaminoglycan component of membrane proteoglycans. The specificity of the binding is unique amongst IGFBPs so far described in that IGFBP-2 binds to chondroitin sulfate and keratan sulfate, glycosaminoglycans that are abundant in proteoglycans from that part of the rat brain. In addition, glycosaminoglycan association lowered the binding affinity of IGFBP-2 for IGF-I. Although the structural basis of these effects remains to be determined, this may represent an additional mechanism for spatial localization of IGFBPs and targeting of IGF actions to specific sites in the developing brain.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received August 28, 1997.
|
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:
 |
|
 |
|
  K. W Frommer, K. Reichenmiller, B. S Schutt, A. Hoeflich, M. B Ranke, G. Dodt, and M. W Elmlinger IGF-independent effects of IGFBP-2 on the human breast cancer cell line Hs578T. J. Mol. Endocrinol., August 1, 2006; 37(1): 13 - 23. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Xu, C. Liu, J. C. Clark, and J. A. Whitsett Functional Genomic Responses to Cystic Fibrosis Transmembrane Conductance Regulator (CFTR) and CFTR{Delta}508 in the Lung J. Biol. Chem., April 21, 2006; 281(16): 11279 - 11291. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. C. Russo, P. D. Gluckman, E. L. Feldman, and G. A. Werther The Insulin-Like Growth Factor System and Its Pleiotropic Functions in Brain Endocr. Rev., December 1, 2005; 26(7): 916 - 943. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. C. Russo, B. S. Schutt, E. Andaloro, S. I. Ymer, A. Hoeflich, M. B. Ranke, L. A. Bach, and G. A. Werther Insulin-Like Growth Factor Binding Protein-2 Binding to Extracellular Matrix Plays a Critical Role in Neuroblastoma Cell Proliferation, Migration, and Invasion Endocrinology, October 1, 2005; 146(10): 4445 - 4455. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. B. Bergman, C. J. Moravski, S. R. Edmondson, V. C. Russo, L. A. Bach, J. L. Wilkinson-Berka, and G. A. Werther Expression of the IGF System in Normal and Diabetic Transgenic (mRen-2)27 Rat Eye Invest. Ophthalmol. Vis. Sci., August 1, 2005; 46(8): 2708 - 2715. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Vorwerk, K. Mohnike, H. Wex, F.-W. Rohl, M. Zimmermann, W. F. Blum, and U. Mittler Insulin-Like Growth Factor Binding Protein-2 at Diagnosis of Childhood Acute Lymphoblastic Leukemia and the Prediction of Relapse Risk J. Clin. Endocrinol. Metab., May 1, 2005; 90(5): 3022 - 3027. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. M. Firth and R. C. Baxter Cellular Actions of the Insulin-Like Growth Factor Binding Proteins Endocr. Rev., December 1, 2002; 23(6): 824 - 854. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Kita, K. Nagao, N. Taneda, Y. Inagaki, K. Hirano, T. Shibata, M. A. Yaman, M. A. Conlon, and J.-i. Okumura Insulin-Like Growth Factor Binding Protein-2 Gene Expression Can Be Regulated by Diet Manipulation in Several Tissues of Young Chickens J. Nutr., February 1, 2002; 132(2): 145 - 151. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. R. Clemmons Use of Mutagenesis to Probe IGF-Binding Protein Structure/Function Relationships Endocr. Rev., December 1, 2001; 22(6): 800 - 817. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Mazerbourg, J. Zapf, R. S. Bar, D. R. Brigstock, and P. Monget Insulin-Like Growth Factor (IGF)-Binding Protein-4 Proteolytic Degradation in Bovine, Equine, and Porcine Preovulatory Follicles: Regulation by IGFs and Heparin-Binding Domain-Containing Peptides Biol Reprod, August 1, 2000; 63(2): 390 - 400. [Abstract] [Full Text]
|
|
|
|
 |
|
 R. C. Baxter Insulin-like growth factor (IGF)-binding proteins: interactions with IGFs and intrinsic bioactivities Am J Physiol Endocrinol Metab, June 1, 2000; 278(6): E967 - E976. [Abstract] [Full Text] [PDF]
|
|
), or absence (
), of an excess of cold IGF-I.
125I-IGF-I alone was also incubated in the presence (
), of cold IGF-I. Wells coated with BSA were incubated
in the same way and used as control. Results are shown as mean
±SD of three experiments, and each point was measured in
quadruplicate.
