| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Division of Endocrinology and Diabetes, The Children’s Hospital of Philadelphia, University of Pennsylvania, Philadelphia, Pennsylvania 19104
Address all correspondence and requests for reprints to: Pinchas Cohen, M.D., Division of Endocrinology, Department of Pediatrics, University of Pennsylvania, The Children’s Hospital of Philadelphia, 3400 Civic Center Boulevard, Philadelphia, Pennsylvania 19104. E-mail: cohenp{at}email.chop.edu
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
We investigated the binding of IGFBP-3 to additional association proteins in human serum (IGFBP-3 APs). Ligand blots, column chromatography, and affinity cross-linking experiments revealed the specific binding of IGFBP-3 to at least three novel serum proteins. These techniques demonstrated the presence of proteins with molecular masses of 70, 100, and 150 kDa that bind IGFBP-3 with high affinity. Serum ALS migrated separately (at 88 kDa) from the novel IGFBP-3 APs (as evident by Western immunoblot), and bound IGFBP-3 weakly (by reverse ligand blots). We also demonstrated that large amounts of one of the IGFBP-3 APs and small amounts of ALS were coimmunoprecipitated with IGFBP-3 from human serum. Similar to ALS, these IGFBP-3 APs are acid labile and lose their IGFBP-3 binding capacity after exposure to low pH.
We conclude that there are several serum proteins in addition to ALS and IGFs that bind IGFBP-3 with high affinity. These IGFBP-3 APs may serve as an additional reservoir of IGFBP-3 or modulate its functions.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Of the six IGFBPs, IGFBP-3 is the most abundant in serum, where it binds to IGFs and the acid-labile subunit (ALS) and forms the 150-kDa ternary complex (3, 4). IGFBP synthesis occurs in the liver as well as in other tissues, with each tissue having specific levels of several IGFBPs (5). Similar to IGF-I and ALS, IGFBP-3 synthesis is under GH regulation (1, 3, 6). IGFBP levels and function are modulated not only by the rate of expression and synthesis, but also by posttranslational molecular modifications and proteolysis (1, 7, 8, 9).
The binding of IGFs to their binding proteins prolongs the half-lives of the IGFs. Similarly, the ternary complex prolongs the half-life of IGFBP-3 and decreases the passage of IGFBP-3 from the intravascular to the extravascular space. The half-life of unbound IGF-I is less then 10 min, and the half-life of unbound IGFBP-3 is approximately 60 min. The formation of the ternary complex allows IGF-I and IGFBP-3 to remain intact in circulation for close to 12 h (3, 10).
In serum, IGFBP-3 is currently thought to bind only to the IGFs and ALS. At the tissue level, IGFBP-3 binds to other proteins; intact IGFBP-3 binds to a putative cell surface receptor (11, 12, 13) as well as to heparin and matrix proteoglycans (14, 15, 16, 17). The binding of IGFBP-3 to the cell surface may facilitate the presentation of IGFs to the IGF receptor and consequently enhance IGF actions (1, 2, 9). The binding of IGFBP-3 to its receptor is reported to decrease cell growth and induce apoptosis (11, 12, 13, 18).
As IGFBP-3 is able to bind multiple proteins, we hypothesized that human serum contains additional proteins, other than ALS, that may bind IGFBPs to regulate the availability of IGFBP. To investigate this hypothesis, we used several different methods to assess the binding of IGFBP-3 to serum proteins.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Samples were collected from healthy, young adult, nonpregnant (three women and five men) and pregnant (n = 5) volunteers after informed consent was obtained. Samples were collected in serum separator tubes, centrifuged, and frozen at -70 C until analysis.
Materials
Recombinant Chinese hamster ovary-derived IGFBP-3 was a gift from Genentech (Sount San Francisco, CA). Recombinant human IGF-I was donated by Pharmacia (Sweden). Iodinated IGFBP-3, affinity-purified anti-human IGFBP-3 antibody, and anti-human ALS antibody were purchased from Diagnostic Systems Laboratories (Webster, TX). The anti-IGFBP-3 antibody was affinity purified on a column of recombinant human IGFBP-3. SDS-PAGE reagents, Tween, and fat-free milk for blocking were purchased from Bio-Rad (Richmond, CA). Tris (crystallized free base) was purchased from Fisher Chemical Co. (Fairlawn, NJ). Disuccinimidyl suberate (DSS) was purchased from Pierce (Rockford, IL). Dimethylsulfoxide (DMSO) and Igepal CA-630 were purchased from Sigma Chemical Co. (St. Louis, MO). Gel filtration was performed using a HiPrep 26/60 Sephacryl 200 column purchased from Pharmacia (Uppsala, Sweden).
Reverse Western ligand blot
Reverse Western ligan blots were used to assess the presence of serum IGFBP-3 association proteins (APs). Two microliters of serum were electrophoresed on SDS-PAGE (8%) at constant voltage overnight and transferred onto a nitrocellulose membrane (4 h). The nitrocellulose membranes were buffered and permeabilized by immersion in Tris-buffered saline (TBS) with 3% Igepal CA-630 for 30 min. After this, the membranes were blocked in TBS with 1% BSA for 3 h and then incubated overnight with [125I]IGFBP-3 (106 cpm) in 0.1% Tween in TBS and 1% BSA. Visualization was accomplished with autoradiography after four washes in TBS-0.1% Tween and two washes in TBS.
Western immunoblots (WIBs)
WIBs identified IGFBP-3 and the ALS. Two microliters of serum were electrophoresed on SDS-PAGE (8%) at constant voltage, then transferred onto a nitrocellulose membrane (4 h). The nitrocellulose membranes were immersed in blocking solution (5% fat-free milk in TBS) for 45 min, washed with 0.1% Tween in TBS, and incubated with primary (anti-IGFBP-3 or anti-ALS) antibodies (1:4,000) for 2.5 h. After unbound antibodies were washed off, nitrocellulose membranes were incubated with secondary antibodies (1:10,000) for 1 h. After washing excessive unbound antibodies with four washes in 0.1% Tween in TBS, the peroxidase-linked enhanced chemiluminescence detection system (ECL) from Amersham (Arlington Heights, IL) was used for visualization.
Affinity cross-linking
Serum samples were cross-linked to [125I]IGFBP-3. One milliliter of serum was incubated with [125I]IGFBP-3 (10,000 cpm) in the presence or absence of unlabeled ligand for 2.5 h in binding buffer (50 mmol/L Tris, pH 7.4), then cross-linked using DSS in DMSO (0.33 mg/mL) for 15 min. Recombinant human IGFBP-3 was used to competitively displace the radiolabeled ligands. Samples were electrophoresed on 8% SDS-PAGE overnight at constant voltage. Gels were dried and then visualized by autoradiography.
Gel filtration
Serum was filtered using a 0.2-mm Whatman filter (Clifton, NJ) and centrifuged to remove debris that may clog the column. Control, nonpregnant serum (3 mL) was loaded onto the gel filtration column. The column was equilibrated and run with 20 mmol/L HEPES (pH 7.2)-150 mmol/L NaCl buffer. A total of 90 sample fractions (3.2 mL each) were collected. The majority of protein eluted in fractions 19–40. Proteins in these fractions ranged from approximately 250 to 20 kDa, as confirmed by gel electrophoresis and Coomassie staining.
Immunoprecipitation of the IGFBP-3 or ALS-bound proteins
Serum samples were immunoprecipitated with the use of specific antibodies. Affinity-purified anti-IGFBP-3 antibody or anti-ALS antibody was incubated with protein A for 24 h at 4 C. The protein A/antibody complex was incubated with 4 µL control serum for 48 h at 4 C and then precipitated. The pellet was resuspended and washed three times with staph-A antibody coupling immunoprecipitation (SACI) buffer, suspended in 1 x sample buffer, vortexed, and boiled for 5 min to release the antigen-antibody complex. The supernatant was electrophoresed on 8% SDS-PAGE and visualized by WIBs and IGFBP-3 Western ligand blots.
Sample acidification
To assess the effects of low pH on the IGFBP-3 APs, control serum samples were diluted 1:10, and the pH was lowered by the addition of increasing amounts of HCl (1 mol/L). Samples were incubated at room temperature for 4 h, after which the pH was elevated to 7.0. Samples were electrophoresed on 8% SDS-PAGE and transferred to a nitrocellulose membrane. The nitrocellulose membrane was then incubated with [125I]IGFBP-3 overnight (IGFBP-3 Western ligand blot), dried, and visualized by autoradiography.
ELISA assays
Samples from the gel filtration column fractions were assayed for IGFBP-3 and IGF-I by ELISA kits (Diagnostic Systems Laboratories).
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
To assess the binding of IGFBP-3 to serum proteins, two different
techniques were used. Figure 1
is an
IGFBP-3 reverse Western ligand blot, and Figs. 2 and 3
represent an [125I]IGFBP-3 affinity cross-linking with
the corresponding densitometrical analysis of the displacement
curve.
|
|
|
. Samples from six healthy, nonpregnant (C1–C6) subjects and two
pregnant (P1 and P2) subjects were electrophoresed on an 8% SDS-PAGE
and visualized by autoradiography after incubating the nitrocellulose
membrane with [125I]IGFBP-3. The specificity of the
binding of these serum proteins to IGFBP-3 were demonstrated by
incubating the membranes with or without unlabeled IGFBP-3 (1000 ng/mL)
in addition to the [125I]IGFBP-3 trace, as shown in Fig. 1B. The specificities of the binding of the serum proteins to IGFBP-3
were also demonstrated by affinity cross-linking of
[125I]IGFBP-3, as shown in Fig. 2. Sera from a control
subject and from a pregnant woman were incubated with
[125I]IGFBP-3 in duplicate. Unlabeled recombinant IGFBP-3
was used to displace [125I]IGFBP-3, demonstrating the
specificity of binding. After cross-linking, samples were
electrophoresed on an 8% SDS-PAGE. The binding was visualized by
autoradiography. At 44 kDa, unbound [125I]IGFBP-3 trace
is evident. The binding of IGFs (with a molecular mass of 7 kDa) to
[125I]IGFBP-3 and the displacement by unlabeled IGFBP-3
are represented by a 51-kDa band that is present on the samples
incubated with labeled IGFBP-3. There is a progressive decrement in its
signal as more unlabeled IGFBP-3 is added to the reaction. When 100
ng/mL IGFBP-3 were added to both control and pregnancy serum samples,
the signal at 51 kDa starts to decrease and is completely absent in the
presence of 1000 ng/mL IGFBP-3. Also shown in the right
panel is 20 ng/mL recombinant IGF-I cross-linked to
[125I]IGFBP-3, demonstrating size identity with serum
IGFs. There is also an 88-kDa band that increases in intensity as the
concentration of unlabeled IGFBP-3 increases. This band demonstrates
the likely formation of dimers consisting of labeled and unlabeled
IGFBP-3. Of particular note is that there are two bands at 150 and 200
kDa that represent proteins with molecular masses of 100 and 150 kDa
bound to [125I]IGFBP-3. The intensity of these bands
decreases as the concentration of unlabeled IGFBP-3 increases, similar
to the 51-kDa band ([125I]IGFBP-3-bound IGFs). These two
bands are presumably the same proteins seen at 100 and 150 kDa on the
reverse ligand blot, again demonstrating that these proteins can
specifically bind IGFBP-3 with high affinity. The results of the
densitometrical analysis of the previous mentioned bands are shown in
Fig. 3. It is clear that with the addition of increasing amounts of
unlabeled IGFBP-3, the signal of the band at 88 kDa (dimers) increases,
whereas the signal of the 51-kDa band as well as those of the 150- and
200-kDa bands decrease. The displacement of the labeled ligand using
unlabeled ligand demonstrates the specificity of the binding and shows
that the IGFBP-3 APs have an affinity to IGFBP-3 that may be similar to
that of IGFs.
In Fig. 4A, two control serum samples
were electrophoresed on 8% SDS-PAGE. Distinct bands with molecular
masses of 70, 88, 100 and 150 kDa that bind [125I]IGFBP-3
were seen in all subjects. These bands represent proteins that were
able to bind specifically to IGFBP-3. To assess the possibility that
ALS may be one of the binding proteins shown by the IGFBP-3 Western
ligand blot, ALS WIB was performed on the same membrane (Fig. 4B). When
comparing the molecular weights of the high molecular mass binding
proteins and ALS, a protein doublet with a molecular mass of 88 kDa was
observed that may represent ALS.
|
Figure 5 represents an IGFBP-3
Western ligand blot of human serum after neutral column chromatography
separation. This experiment was performed to assess the distribution of
the IGFBP-3 APs before electrophoresis on SDS-PAGE. Three bands, one
(which sometimes appears as a doublet) at 70, 100, and 150 kDa were
observed in multiple fractions, demonstrating that those proteins were
present in serum at their respective molecular masses before SDS-PAGE.
Also shown is the distribution in the same fractions of IGFBP-3 and
IGF-I (by ELISA) and ALS (by WIB). These results demonstrate that
IGFBP-3 APs (primarily IGFBP-3 APb) are found in the same fractions as
IGFBP-3, possibly bound to it. Of note is that IGFBP-3 APb is found in
the same fractions as IGFBP-3, whereas IGFBP-3 APa and IGFBP-3 APc are
found both in the same fractions with IGFBP-3 and in fractions without
IGFBP-3.
|
Figure 6 compares identical
nitrocellulose membranes processed by WIB with IGFBP-3 (A) or ALS (B)
antibodies or Western ligand blotting with [125I]IGFBP-3
(C). Two microliters of nonpregnant (lane 1) or pregnant (lane 2) serum
were electrophoresed on an 8% SDS-PAGE. Lanes 3 and 4 contain samples
of immunoprecipitated proteins from similar sera using anti-IGFBP-3
antibody. ALS and its associated serum proteins were also
immunoprecipitated from these sera and are shown in lanes 5 and 6.
|
. IGFBP-3 and its fragments were present in the
control samples (Fig. 6A, lanes 1 and 2) and, as expected, in the
samples immunoprecipitated with the anti-IGFBP-3 antibody (Fig. 6A, lanes 3 and 4). Only a small amount of IGFBP-3 was coimmunoprecipitated
with ALS (Fig. 6A, lanes 5 and 6).
Figure 6B shows an ALS WIB. ALS is found in the control serum (Fig. 6B, lanes 1 and 2) and in the samples immunoprecipitated with the anti-ALS
antibody (Fig. 6B, lanes 5 and 6). Similar to Fig. 6A (lanes 5 and 6),
a small amount of ALS was coimmunoprecipitated with the anti-IGFBP-3
antibody (Fig. 6B, lanes 3 and 4).
In Fig. 6C, the nitrocellulose membrane was incubated with
[125I]IGFBP-3. Two forms of IGFBP-3 APs can be seen in
control lanes 1 and 2 (Fig. 6C). The IGFBP-3 AP with a molecular mass
of 150 kDa was coimmunoprecipitated with IGFBP-3, as shown in Fig. 6C, lanes 3 and 4, whereas no IGFBP-3 APs were coimmunoprecipitated with
ALS (Fig. 6C, lanes 5 and 6). These results indicate that IGFBP-3 in
serum is found bound to both ALS and another IGFBP-3 AP.
Effects of sample acidification on the IGFBP-3 APs
Figure 7 demonstrates the
densitometrical analysis of the 100-kDa IGFBP-3 APs after incubation in
different degrees of acidification. It is evident that IGFBP-3 APs
(like ALS) are acid labile.
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Fielder et al. demonstrated that in hypophysectomized rats, treatment with rhIGF-I induced the formation of an IGFBP-3-containing complex with a molecular mass above 200 kDa. The formation of this complex was increased in the absence of ALS, suggesting that with increased availability of free IGFBP-3, the formation of complexes of IGFBP-3 to other association proteins was facilitated. In their report, the composition or regulation of this complex was not documented. The authors also proposed that the larger the IGF-bound complexes, the longer the IGF half-life (20).
We demonstrated that human serum contains proteins that bind IGFBP-3 specifically. Most of the IGFBP-3 in serum is assumed to be in the ternary complex form with ALS and IGF. In this complex, ALS serves as an IGFBP-3 (or IGF AP). To investigate the possibility that ALS is one of these novel high molecular mass serum proteins, ALS immunoblots were performed and clearly demonstrated that three of the high molecular mass proteins that bind IGFBP-3 are distinct from ALS.
Using column chromatography followed by fraction electrophoresis or ELISA, we showed that the IGFBP-3 APs with molecular masses of 150 and 70 kDa are unlikely to be major carriers of IGFBP-3 in normal or pregnancy serum. The IGFBP-3 AP form with a molecular mass of 100 kDa was separated into the fractions containing high concentrations of IGFBP-3 as well as ALS and IGFs. With the techniques used, it is difficult to distinguish the IGFBP-3-carrying capacity of ALS vs. that of the 100-kDa IGFBP-3 AP form. The differentiation of the relative binding of IGFBP-3 to ALS and to the novel IGFBP-3 AP was analyzed using immunoprecipitation techniques. Immunoprecipitation of IGFBP-3 resulted in the coprecipitation of some forms of IGFBP-3 APs as well as ALS.
We also demonstrated that some of the IGFBP-3 APs are acid labile, like the ALS of the 150-kDa complex. This may explain why serum acidification results in a complete separation of IGFs, IGFBP-3, and the high molecular mass acid-labile proteins (both ALS and IGFBP-3 APs).
Gargosky et al. demonstrated the presence of a high molecular mass form of IGFBP-3 in baboons using size-exclusion chromatography (21). The researchers postulated that these complexes could be formed from altered forms of IGFBP-3 that are able to bind ALS but not IGFs. They even suggested that an alternative high molecular mass form of IGFBP-3 may be present. The recent preliminary report of ALS gene-deleted mice that have no apparent phenotypic abnormalities suggest that IGFBP-3 APs might serve as back-up proteins to ALS (22).
We have demonstrated the presence of novel proteins that possibly function as IGFBP-3 carriers or modulators in serum. We speculate that the formation of IGFBP-3/IGFBP-3 APs complexes is a normal phenomena in vivo that may function to protect IGFBP-3 from proteolysis or to modulate its actions. Further characterization of these novel proteins, including purification and sequencing followed by in-depth analysis of their physiological functions, is clearly warranted.
|
Footnotes |
|---|
Received December 1, 1997.
Revised April 24, 1998.
Accepted May 12, 1998.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  J. V. Silha, Y. Gui, S. Mishra, A. Leckstrom, P. Cohen, and L. J. Murphy Overexpression of Gly56/Gly80/Gly81-Mutant Insulin-Like Growth Factor-Binding Protein-3 in Transgenic Mice Endocrinology, March 1, 2005; 146(3): 1523 - 1531. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Gui and L. J. Murphy Insulin-Like Growth Factor (IGF)-Binding Protein-3 (IGFBP-3) Binds to Fibronectin (FN): Demonstration of IGF-I/IGFBP-3/FN Ternary Complexes in Human Plasma J. Clin. Endocrinol. Metab., May 1, 2001; 86(5): 2104 - 2110. [Abstract] [Full Text]
|
|
|
|
|
|
S. A. Weinzimer, T. B. Gibson, P. F. Collett-Solberg, A. Khare, B. Liu, and P. Cohen Transferrin Is an Insulin-Like Growth Factor-Binding Protein-3 Binding Protein J. Clin. Endocrinol. Metab., April 1, 2001; 86(4): 1806 - 1813. [Abstract] [Full Text]
|
|
|
|
 |
|
 V. Hwa, Y. Oh, and R. G. Rosenfeld The Insulin-Like Growth Factor-Binding Protein (IGFBP) Superfamily Endocr. Rev., December 1, 1999; 20(6): 761 - 787. [Abstract] [Full Text]
|
|
|
|
 |
|
 B. Liu, H.-Y. Lee, S. A. Weinzimer, D. R. Powell, J. L. Clifford, J. M. Kurie, and P. Cohen Direct Functional Interactions between Insulin-like Growth Factor-binding Protein-3 and Retinoid X Receptor-alpha Regulate Transcriptional Signaling and Apoptosis J. Biol. Chem., October 20, 2000; 275(43): 33607 - 33613. [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 |
