| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Kolling Institute of Medical Research (L.J.S., T.N., R.C.B.), University of Sydney, Royal North Shore Hospital, Sydney, New South Wales 2065, Australia; John Curtin School of Medical Research (A.P.J., D.A.J.), Australian National University, Canberra, Australian Capital Territory 2601, Australia; and Department of Biochemistry and Molecular Biology (D.A.J.), Monash University, Clayton, Victoria 3168, Australia
Address all correspondence and requests for reprints to: Dr. Lyn Schedlich, Kolling Institute of Medical Research, Royal North Shore Hospital, St. Leonards, New South Wales 2065, Australia. E-mail: lyns{at}med.usyd.edu.au.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The mechanism(s) for the IGF-independent effects of IGFBP-3 are currently unknown but may involve the regulation of gene expression either indirectly by activation of a cell-surface receptor or directly following uptake of IGFBP-3 into the nucleus. Nuclear localization of IGFBP-3 has now been reported in a number of cell lines (9, 10, 11, 12), and more recently in human colon tumor samples (13). We have shown that nuclear import of IGFBP-3 (and the highly homologous IGFBP-5) is a nuclear localization signal (NLS)-dependent process and that it is mediated by the importin-ß nuclear transport factor (14). Following nuclear entry, IGFBP-3 actively accumulates through binding to insoluble nuclear components. The fact that exogenous IGFBP-3 is internalized and transported to the nucleus makes it a good candidate for exerting direct effects on gene expression. Indeed, IGFBP-3 modulates the transcriptional activity of the nuclear retinoid X receptor-
as part of the IGFBP-3-mediated induction of apoptosis (15). Interestingly, this cannot be the only pathway by which IGFBP-3 induces apoptosis because IGFBP-3 with a mutated NLS is still apoptotic in breast cancer cells (16). IGFBP-3 has also been shown to act as a carrier for IGF-I nuclear import (10). Thus, nuclear interactions between IGFs, IGFBP-3, and other nuclear proteins may contribute an additional level of control to cell growth, differentiation, and apoptosis that is independent of signaling through either the type I IGF receptor or the putative IGFBP-3 receptor.
Although some proteins appear to be constitutively nuclear, others enter the nucleus only in response to certain stimuli (17). This important mechanism for regulating cellular responses is often associated with protein phosphorylation. IGFBP-3 is secreted as a phosphoprotein, and its major phosphorylation sites are contained within consensus protein kinase CK2 (CK2) sites (18). We have shown that IGFBP-3 can be phosphorylated in vitro by CK2 and cAMP-dependent protein kinase (PKA; Refs.19 and20). Interestingly, there is a cluster of three potential phosphorylation sites for the double-stranded DNA-dependent protein kinase (DNA-PK) in the nonconserved domain of IGFBP-3 (21). DNA-PK is a nuclear serine/threonine protein kinase, also present in the cytosol (22), which is activated by DNA damage and is thought to have a role in the repair of DNA strand breaks, DNA replication, gene transcription, and recombination of Ig genes (23). DNA-PK is capable of phosphorylating many transcription factors in vitro, including the tumor suppressor protein p53.
As part of our ongoing investigation into the role of nuclear IGFBP-3, the present study examines the modulation of IGFBP-3 function by phosphorylation with a number of serine/threonine kinases, including DNA-PK. We report that nuclear accumulation of IGFBP-3 is enhanced by its phosphorylation with DNA-PK. In addition, this phosphoform has significantly reduced affinity for IGFs and appears to be unable to cotransport IGF-I into the nucleus. The potential exists for nuclear cotransport of IGFBP-3 and IGF-I to be followed by phosphorylation of IGFBP-3 with DNA-PK and subsequent dissociation of both polypeptides in the nucleus. Thus, phosphorylation of IGFBP-3 with DNA-PK has the potential to modulate directly the effects of nuclear IGFBP-3 as well as its indirect effects as a carrier of nuclear IGFs.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
70 kDa) were purchased from Molecular Probes, Inc. (Eugene, OR). Antiserum against IGFBP-3 was prepared in this laboratory following immunization of rabbits with purified antigen. The kinases CK2, PKA, p34cdc2/cyclin B (cyclin-dependent protein kinase; cdc2), and p42 MAPK were purchased from New England Biolabs, Inc. (Beverly, MA), and DNA-PK and DNA-PK peptide substrate were from Promega Corp. (Madison, WI). Creatine phosphokinase, creatine phosphate, ATP, leupeptin, 3-[(3-cholamidopropyl)-dimethylammonio]-1-propane-sulfonate, and RIA grade BSA were obtained from Sigma (St. Louis, MO).
Cell culture
HeLa cells were grown in a monolayer and maintained in DMEM supplemented with 5% fetal calf serum. Chinese hamster ovary (CHO) cells were maintained in -MEM supplemented with 10% fetal calf serum.
Preparation of HeLa cell extracts
Nuclear and cytoplasmic extracts from HeLa cells were prepared by a modification of the original procedure of Dignam et al. (25). These cells have been used in many of the defining studies on DNA-PK (26, 27). HeLa cells were resuspended in a hypotonic buffer and allowed to swell on ice for 10 min and lysed in a Dounce homogenizer. Centrifugation of the lysate separated the soluble cytoplasmic proteins (supernatant) and the nuclei (pellet). Soluble nuclear proteins were extract by the addition of a high salt buffer and the extracted nuclei were pelleted by centrifugation. Both the nuclear and cytoplasmic fractions were dialyzed against 20 mM HEPES, pH 7.9; 20% glycerol; 100 mM KCl; 0.2 mM EDTA; 0.5 mM dithiothreitol; and a protease inhibitor cocktail (Roche, Mannheim, Germany). Protein concentrations were determined using the Bradford assay. To test for DNA-PK activity in the soluble cellular extracts, the DNA-PK peptide substrate (50 µg) was phosphorylated with purified DNA-PK (60 U) or 230 µg of either cytoplasmic or nuclear extract in the presence of 
32P[ATP] according to the manufacturers’ instructions. As controls, the reactions were also carried out in the absence of DNA-PK peptide substrate. The reaction mixtures were spotted onto P-81 phospho-cellulose paper (Whatman, Maidstone, UK) and the unincorporated radioactivity removed by washing with 15% acetic acid. Phosphorylation of the DNA-PK peptide substrate was detected by scintillation counting. The level of phosphorylation of the peptide substrate by a known amount of purified enzyme was used to determine the level of DNA-PK activity in the nuclear and cytoplasmic extracts. After correction for background phosphorylation, a significant level of DNA-PK-specific activity was found in HeLa cytosol (0.33 U/µg cytosol), equivalent to 20% of that detected in the nuclear extracts (1.6 U/µg nuclear extract).
Phosphorylation of IGFBP-3
IGFBP-3 was phosphorylated in vitro using the kinases CK2, PKA, DNA-PK, p34cdc2/cyclin B and p42 MAPK, in the presence of 50 µM ATP and 50 µCi/ml 32P[ATP] according to the manufacturers’ instructions. Similarly, 275 µg of HeLa cytosolic or nuclear extract was used to phosphorylate IGFBP-3 in the absence or presence of a DNA-PK-specific peptide substrate (10 µg). IGFBP-3 was phosphorylated alone (1 µg) or in complex with a 10-fold molar excess of IGF-I (IGFBP-3, 1 µg; and IGF-I, 1.8 µg). To allow for complex formation IGFBP-3 and IGF-I were coincubated at 22 C for 90 min. Following phosphorylation, samples were separated on 10% SDS-PAGE, the gel was dried and phosphorylated IGFBP-3 detected by autoradiography.
Ligand binding
The ability of IGFs to bind the different phosphoforms of IGFBP-3 was determined by ligand binding assay and Western ligand blotting as previously described (28). IGFBP-3 was phosphorylated in vitro and purified by reverse-phase HPLC. Reactions for the ligand binding assay contained 35 pg of 125I-IGF-I and amounts of IGFBP-3 ranging from 0–25 ng. Following a 2-h incubation at 22 C, the complex was immunoprecipitated with IGFBP-3 antiserum, goat antirabbit immunoglobulin and polyethylene glycol. Results are expressed as the percent of total counts bound to IGFBP-3. For Western ligand blotting, 100 ng of non-phospho- or phospho-IGFBP-3 were separated on 10% SDS-PAGE before membrane transfer. The membranes were hybridized at 4 C for 16 h in buffer containing 125I-IGF-I or 125I-IGF-II (2 x 104 cpm/ml). Binding of IGFs to IGFBP-3 was detected by autoradiography.
Phosphorylation of fluorescently labeled IGFBP-3
To prepare fluorescently labeled phosphoforms of IGFBP-3, two approaches were used. In the first, IGFBP-3 (600 µg) was fluorescently labeled with BODIPY FL as previously described (11). We used a derivative of this dye that lacked an ionic charge making it ideal for studies where the addition of negatively charged moieties would be undesirable. All conjugation reactions were carried out at pH 7.0 to favor modification of the amino terminus of IGFBP-3 rather than its lysine residues. Aliquots (100 µg) of fluorescently labeled IGFBP-3 were phosphorylated with CK2, PKA, DNA-PK, p34cdc2/cyclin B, or p42 MAPK in the presence of 200 µM ATP and a trace amount of 32P[ATP] (5 µCi/ml). A control, nonphosphorylated sample was prepared by incubating another 100-µg aliquot from the same batch of BODIPY-labeled IGFBP-3 in the absence of enzyme. The kinases were destroyed by acidification to pH 4.0 with 1 M acetic acid and IGFBP-3 purified by gel filtration on a Sephadex G-100 column and concentrated by diafiltration. The immunoreactivity of IGFBP-3 was not altered by phosphorylation with CK2, PKA, cdc2, or MAPK. However, the DNA-PK phosphoform did display reduced immunoreactivity (data not shown). Therefore, the concentration of IGFBP-3 was determined by RIA for all phosphoforms except the DNA-PK phosphoform, which was quantitated using amino acid analysis. The level of fluorescence was determined by spectrofluorophotometry and the integrity of IGFBP-3 by Western immunoblotting (data not shown). The success of phosphorylation was confirmed by detecting ß-emissions by scintillation counting. In the second approach, 100 µg aliquots of recombinantly derived and purified IGFBP-3 were phosphorylated with CK2, PKA, DNA-PK, p34cdc2/cyclin B, or p42 MAPK as described above. Following removal of the kinases by reverse-phase HPLC, IGFBP-3 was fluorescently labeled with BODIPY or DTAF and the unincorporated dye removed by gel filtration as previously described (11).
In vitro nuclear transport assay
In vitro nuclear transport was carried out as described previously (14). CHO cells were cultured on glass coverslips, washed with ice-cold transport buffer and permeabilized with 50 µg/ml digitonin (Calbiochem, La Jolla, CA), 5 min on ice. The cells were incubated with transport buffer containing fluorescently labeled IGFBP-3 (5 ng/µl) with or without fluorescently labeled IGF-I (0.9 ng/µl) in the presence of 45 mg/ml rabbit reticulocyte lysate (Promega Corp.), an ATP-regenerating system and 200 µg/ml Texas Red-dextran. To test for nuclear cotransport of IGFBP-3 and IGF-I, the fluorescently labeled proteins were coincubated for 1 h at 22 C and Texas Red-dextran was omitted from the reaction. The protease inhibitors, leupeptin, Trasylol (Bayer Corp. AG, Leverkusen, Germany) and bestatin (Roche) and a cocktail of serine/threonine protein phosphatase inhibitors (Sigma, catalog no. P2850) were added to the transport assay. In experiments designed to assess the contribution to nuclear accumulation of binding to nuclear components, the nuclear envelope was permeabilized by addition of 0.025% 3- [(3-cholamidopropyl)-dimethylammonio]-1-propane-sulfonate to the transport solution.
The cells were incubated in a humidified environment at 22 C. At the indicated time points, cells were fixed and examined using a confocal laser scanning microscopic (CLSM) system (Optiscan F900e Personal Confocal System, Melbourne, Australia) fitted with a krypton-argon laser and dual channel detection optics. Individual cells were optically sectioned in the X-Y plane with multiple scan averaging. All images were collected under identical, nonsaturating conditions. The intensity of fluorescent labeling within cells was analyzed using the program NIH Image version 1.61. Pixel intensity, as a measure of fluorescence intensity, was measured within specific regions of the cell (cytoplasmic and nuclear) as well as in regions outside the cell (background). The pixel intensity from each subcellular region was averaged over at least 100 cells at each time point. After correction for background fluorescence, the results were expressed as the ratio of nuclear to cytoplasmic fluorescence.
Statistical analysis
Data were analyzed by an unpaired t test or by ANOVA followed by Fisher’s protected least significant difference test using Statview 4.02 (Abacus Concepts, Inc., Berkeley, CA).
Western ligand binding assay
Binding of importin ß to IGFBP-3 was examined by Western ligand binding assay as previously described (29). The ELISA-based binding assay (29) is not appropriate for use with IGFBP-3 because it requires an extended incubation at high pH, which is detrimental to IGFBP-3 structure/function. The mouse importin ß-subunit was expressed as a glutathione-S-transferase (GST) fusion protein. Non-phospho- and phospho-IGFBP-3 samples and control proteins (1.25 µg) were separated on 10% SDS-PAGE before membrane transfer. The membranes were hybridized at 4 C for 16 h in intracellular buffer containing mouse importin ß fused to GST at a final concentration of 150 nM. Binding of importin ß to IGFBP-3 was detected using a GST-specific antibody (Amersham Biosciences, Little Chalfont, Buckinghamshire, UK), followed by a horseradish peroxidase-conjugated secondary antibody (Amersham Biosciences) and ECL (Amersham Biosciences).
Superose 12 gel chromatography
To determine whether DNA-PK phosphorylation of the IGF-I: IGFBP-3 complex caused the complex to dissociate, radioiodinated IGF-I (1 x 105 or 5 x 105 cpm) and IGFBP-3 (20 or 80 ng) were incubated in 50 mM sodium phosphate, 0.1% BSA (pH 6.5) for 2 h at 22 C to allow binary complex formation. The reaction mixtures were then incubated in DNA-PK reaction buffer with or without DNA-PK (20 U) or with 135 µg HeLa cell nuclear extract in the absence or presence of DNA-PK peptide substrate (50 µg) to inhibit DNA-PK phosphorylation of IGFBP-3. After incubation at 30 C for 2 h, the samples were cross-linked with disuccinimidyl suberate (250 µM; Pierce Chemical Co., Rockford, IL) for 30 min at 22 C, quenched by the addition of Tris.HCl (pH 7; 50 mM) and applied to a Superose 12 gel permeation column (Amersham Biosciences). Fractions of 0.5 ml were eluted at 1 ml/min in 50 mM sodium phosphate, 100 mM NaCl, 0.05% BSA (pH 6.5) and the radioactivity of each fraction determined. Under these conditions, IGF-I tracer (7.5 kDa) elutes between fractions 32 and 34 and IGF-I tracer cross-linked to IGFBP-3 (50 kDa) elutes between fractions 26 and 28 (30).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
32P[ATP], separated the phosphoforms on SDS-PAGE and detected them by autoradiography. Figure 1
shows the level of phosphorylation achieved after 2-h incubation of IGFBP-3 with the kinases CK2 (lane 1), PKA (lane 2), DNA-PK (lane 3), cdc2 (lane 4), and MAPK (lane 5). The 30-kDa degradation product of IGFBP-3 also appears to be phosphorylated by PKA and DNA-PK. The 26-kDa protein phosphorylated by CK2 is likely to be the product of autophosphorylation of the 26-kDa ß-subunit of CK2.
|
shows that in the presence of the peptide substrate (lane 2), phosphorylation of IGFBP-3 with purified DNA-PK (lane 1) was reduced by 88.8% ± 5.5% (n = 3). Similarly, the level of IGFBP-3 phosphorylation in nuclear (lane 3) and cytosolic (lane 5) extracts was reduced in the presence of the DNA-PK inhibitor (lanes 4 and 6, respectively), suggesting that IGFBP-3 acted as a substrate for DNA-PK-specific phosphorylation in cell extracts. Quantitation of the data obtained from five similar experiments (Fig. 2B) showed that there was a significant amount of DNA-PK-specific phosphorylation of IGFBP-3 by both nuclear (samples 3 vs. 4: P < 0.0001) and cytoplasmic (samples 5 vs. 6: P = 0.0165) extracts.
|
) showed that, apart from DNA-PK phosphorylated IGFBP-3, all the other phosphoforms displayed similar binding profiles to the nonphosphorylated form. We also examined IGF binding using Western ligand blotting. Equivalent amount of IGFBP-3 (100 ng) were separated on SDS-PAGE, incubated with 125I-IGF-I (Fig. 3B) or 125I-IGF-II (Fig. 3C) and detected by autoradiography. The results show that compared with nonphosphorylated IGFBP-3 (lane 1), there was a significant reduction in the binding of IGF-I and IGF-II to DNA-PK phosphorylated IGFBP-3 (lane 4). Phosphorylation of IGFBP-3 with CK2 (lane 2), PKA (lane 3), cdc2 (lane 5), and MAPK (lane 6) had no effect on ligand binding, confirming the results of the ligand binding assay. The inability of IGFBP-3 phosphorylated with DNA-PK to complex with the IGFs may have important consequences for the putative role of IGFBP-3 as a cotransporter of IGFs to the nucleus, and possibly of other proteins as well.
|
). The subcellular distribution of the fluorescent signal at different time points was determined using CLSM. Nonphosphorylated IGFBP-3 had a maximum nuclear to cytoplasmic fluorescence ratio (Fn/c) of 3.27 with a t1/2 of 4.57 min (Fig. 4A and Table 1). Phosphorylation of IGFBP-3 with CK2, cdc2 and MAPK did not alter their nuclear import compared with the nonphosphorylated form of IGFBP-3. However, there was a reproducible and significant increase in the maximal level of nuclear import of IGFBP-3 following its phosphorylation with DNA-PK (Fn/c = 5.05: P < 0.0001) and to a lesser extent with PKA (Fn/c = 4.03: P = 0.0007) compared with nonphosphorylated IGFBP-3. This was associated with a significant increase in the time at which half-maximal Fn/c was achieved (Table 1). However, this may not reflect any real change in the actual rate of nuclear import but rather be related to the elevated Fn/c. Interestingly, a consensus phospho-acceptor site for PKA at 181T is located adjacent to the cluster of three potential phosphorylation site for DNA-PK.
|
|
). Under these circumstances soluble proteins are able to pass freely between the cytoplasm and nucleus, and nuclear accumulation occurs only if the protein is able to bind insoluble nuclear components (34). Quantitation showed that maximal nuclear:cytoplasmic ratio for the accumulation of nonphosphorylated IGFBP-3 was 3.49 ± 0.09 (t1/2 = 0.64 ± 0.13 min), and 5.13 ± 0.06 (t1/2 = 0.83 ± 0.06 min) for IGFBP-3 phosphorylated by DNA-PK. Thus, the maximal level of nuclear accumulation of IGFBP-3 was significantly enhanced (P = 0.0001) following phosphorylation of IGFBP-3 with DNA-PK compared with the nonphosphoform, suggesting that this phosphoform of IGFBP-3 may display enhanced binding to nuclear components such as chromatin.
Phosphorylation of IGFBP-3 with DNA-PK does not alter its ability to interact with the importin ß-subunit
The importin - and ß-subunits constitute the high affinity NLS receptor used by many proteins to effect their nuclear import (35). However, we have shown that nuclear import of IGFBP-3 is mediated solely by the importin ß-subunit (14). The ability of the different phosphoforms of IGFBP-3 to be recognized by importin ß was examined using a Western ligand binding analysis (Fig. 5). Green fluorescent protein, which does not interact with importin ß, was used as a negative control (lane 8) and the cAMP-response element binding protein, which is known to interact with importin ß (36), was used as a positive control (lane 1). Nonphosphorylated IGFBP-3 (1.25 µg; lane 2) showed strong binding by importin ß and there appeared to be no difference in the ability of importin ß to bind IGFBP-3 following its phosphorylation with DNA-PK (lane 3), CK2 (lane 4), PKA (lane 5), cdc2 (lane 6), or MAPK (lane 7).
|
). Nonphosphorylated and phosphorylated IGFBP-3 were labeled with the fluorophore DTAF, and IGF-I and [LR3]IGF-I were labeled with Texas Red. After 20 min the cells were fixed, and the subcellular localization of IGFBP-3 and IGF-I was visualized using dual channel confocal microscopy with the IGFBP-3 channel shown in green and the IGF-I channel shown in red. Colocalization of IGF-I and IGFBP-3 appears as yellow. Panels F, I, and L are combined images from panels D and E, G and H, and J and K, respectively.
|
) and IGFBP-3 phosphorylated with DNA-PK (Fig. 6B) were rapidly imported into the nucleus, with the phosphoform displaying a higher maximal level of nuclear accumulation. When cells were incubated with IGF-I alone, the peptide did not associate with any cellular structures but dissipated during washing of the sample leaving no discernible signal (Fig. 6C). However, when cells were incubated with IGF-I in the presence of nonphosphorylated IGFBP-3, IGF-I was imported into the cell nucleus (Fig. 6, D–F). When the IGF-I analog, [LR3]IGF-I, was added to cells in the presence of nonphosphorylated IGFBP-3, only low levels of [LR3]IGF-I were detectable in the nucleus (Fig. 6, G–I). These results support previous findings that IGF-I, but not its analogs which have reduced affinity for IGFBP-3, are cotransported into the nucleus by IGFBP-3. When cells were incubated with IGF-I in the presence of IGFBP-3 phosphorylated with DNA-PK, IGF-I was not efficiently imported into the nucleus (Fig. 6, J–L). Thus, a binary complex of IGF-I and IGFBP-3 may not be imported into the nucleus when IGFBP-3 has been previously phosphorylated with DNA-PK.
IGFBP-3 retains its ability to be phosphorylated by DNA-PK when complexed to IGF-I
DNA-PK acts principally in the nucleus and in a physiological context the phosphorylation of IGFBP-3 by this kinase would be expected to follow its nuclear import. In situations where IGFBP-3 is cotransporting IGFs into the nucleus, the reduction in affinity of IGFs for IGFBP-3 following its phosphorylation with this kinase may result in a rapid increase in free nuclear IGF-I. For this scenario to occur, IGFBP-3 would have to be capable of acting as a substrate for DNA-PK when complexed to IGFs. To investigate this, nonphosphorylated IGFBP-3 (1 µg) was incubated with or without a 10-fold molar excess of IGF-I (1.8 µg) to allow binary complex formation. The samples were phosphorylated (Fig. 7) with CK2 (lanes 1 and 2), PKA (lanes 3 and 4), DNA-PK (lanes 5 and 6), cdc2 (lanes 7 and 8), and MAPK (lanes 9 and 10) for 2 h in the presence of 32P[ATP], the proteins were separated on SDS-PAGE and detected by autoradiography. Lanes 1, 3, 5, 7, and 9 contain phospho-IGFBP-3 only (controls), whereas lanes 2, 4, 6, 8, and 10 contain IGFBP-3 phosphorylated in the presence of IGF-I. When IGFBP-3 was complexed to IGF-I, there was no difference in the phosphorylation of IGFBP-3 with CK2, PKA, DNA-PK or MAPK compared with phosphorylation of the IGFBP-3 monomer. However, we observed a marked reduction in the ability of cdc2 to phosphorylate IGFBP-3 when complexed to IGF-I. These results suggest that binding of IGF-I to IGFBP-3 blocks access of the p34cdc2/cyclin B kinase to its recognition site on IGFBP-3. Interestingly, the same effect was not seen when IGFBP-3 was phosphorylated with MAPK. Although cdc2 and MAPK phosphorylate Ser/Thr residues located within similar proline-directed consensus protein kinase sites, these results suggest that they are phosphorylating different sites in IGFBP-3.
|
). Samples were chromatographed on a Superose 12 column and the radioactivity measured in the eluted fractions. In the nonphosphorylated sample, the major peak of radioactivity corresponded to IGF-I tracer bound to IGFBP-3 (fractions 26–28), with some evidence of unbound IGF-I tracer (fractions 32 and 33; Ref.30). In contrast, when the IGF-I:IGFBP-3 complex was phosphorylated with DNA-PK, the major peak of radioactivity corresponded to free IGF-I tracer (fractions 32–34) with a residual amount of binary complex evident (fractions 26–30). We carried out a similar study using HeLa cell nuclear extract (135 µg) to phosphorylate IGFBP-3 complexed to radioiodinated IGF-I in the absence or presence of a DNA-PK peptide inhibitor (Fig. 8B). The results show that phosphorylation of IGFBP-3 by nuclear extract released the majority of IGF-I complexed to IGFBP-3 (fractions 32 and 33), whereas phosphorylation by nuclear extract in the presence of the DNA-PK inhibitor did not cause a significant disassociation of the binary complex. Together, these results suggest that phosphorylation of the IGF-I:IGFBP-3 complex by purified DNA-PK or HeLa nuclear extract increases the amount of free IGF-I compared with the nonphosphorylated state. The ability of DNA-PK to efficiently phosphorylate IGFBP-3 when complexed to IGF-I, and the potential for this to lead to dissociation of the complex in the nucleus, may be an important first step in the role of these bioactive peptides in the nuclear environment.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The ability of a cell to regulate the nuclear import of molecules such as transcription factors and kinases in response to the activation of a particular signaling pathway provides a powerful mechanism for controlling nuclear protein activity. An important mechanism used for this regulation is protein phosphorylation, which can enhance or inhibit nuclear import (17). Although IGFBP-3 can be phosphorylated in vitro by a number of kinases, in vivo phosphorylation has only been described at serine residues that are contained within consensus CK2 sites (18). The kinase(s) responsible for this phosphorylation remains unknown. IGFBP-3 appears to be constitutively phosphorylated at these sites, so their differential phosphorylation is unlikely to be a major determinant of nuclear import. We have now shown that phosphorylation of IGFBP-3 by CK2 has little effect on the nuclear import of IGFBP-3. In contrast, phosphorylation of IGFBP-3 by PKA and DNA-PK increases the nuclear import of IGFBP-3 (in cells with an intact nuclear envelope) and the DNA-PK phosphoform displays enhanced nuclear accumulation of IGFBP-3 (in cells with a perforated nuclear envelope).
IGFBP-3 is a secreted protein which is internalized into the cell by an unknown mechanism. As all the kinases used in this study are found in either the cytoplasmic or nuclear compartments (17, 22), our hypothesis is that phosphorylation of IGFBP-3 by these kinases occurs after uptake into the cell. Although the majority of DNA-PK activity is nuclear, low levels have been detected in cytoplasmic extracts from HeLa cells (this study; and Ref.39) and cytosol derived from HTC rat hepatoma cells (22). It is therefore possible that IGFBP-3 may be phosphorylated by DNA-PK in the cytosol, leading to enhanced nuclear import of IGFBP-3. Alternatively, following cellular uptake and nuclear import of IGFBP-3, phosphorylation by DNA-PK in the nucleus may enhance the ability of IGFBP-3 to interact with nuclear component(s), thus enhancing nuclear accumulation.
Several mechanisms have been proposed to explain why protein phosphorylation may enhance nuclear import/accumulation (17). These include increased affinity of the NLS for the importin /ß nuclear transport factors, release of the NLS containing protein from cytoplasmic retention factor(s) and enhanced binding of the nuclear protein to structures within the nucleus. The increase in nuclear import of a fusion protein containing the NLS of the Drosophila transcription factor Dorsal phosphorylated with PKA was shown to involve enhanced affinity of the Dorsal NLS for importin /ß (31). Similarly, phosphorylation at the CK2 site in the simian virus 40 large tumor antigen increased the affinity of its NLS for importin /ß (29). This effect was also observed when a negative charge was engineered at the DNA-PK phosphorylation site adjacent to the simian virus 40 large tumor antigen NLS (22). In addition, the presence of this negative charge increased phosphorylation at the nearby CK2 site that further enhanced NLS binding to importin /ß. However, we have shown that the binding of importin ß to IGFBP-3 was not significantly affected by its phosphorylation with DNA-PK, suggesting any effect of phosphorylation must be at another point in the import pathway.
To investigate the possibility that the increase in nuclear import of the DNA-PK phosphoform of IGFBP-3 was related to the enhanced binding of IGFBP-3 within the nucleus, we examined the nuclear accumulation of IGFBP-3 in cells where both the plasma membrane and nuclear envelope had been perforated. In these studies, we observed a significant increase in the retention of the DNA-PK phosphoform of IGFBP-3 in the nucleus. Because DNA-PK acts principally in the nucleus, these findings suggest that DNA-PK phosphorylation of nuclear IGFBP-3 may result in enhanced binding of IGFBP-3 to nuclear structures such as lamin and chromatin (34). Similar findings have been described for the interferon-inducible nuclear factor, IFI 16, where the presence of an intact CK2 site enhanced binding of IFI 16 to nuclear components (40).
DNA-PK belongs to a family of protein kinases that includes the protein that is defective in ataxia telangiectasia (ataxia telangiectasia mutated; Ref.41) and the ataxia telangiectasia-related protein (42). All three kinases belong to the phosphatidylinositol 3-kinase-like family and recognize similar motifs in their substrates (21). They are all activated by DNA damage and can phosphorylate p53 in vitro, although a role for DNA-PK as an upstream regulator of p53 remains controversial (43, 44, 45). Interestingly, a recent study found that the radiation-induced p53-dependent induction of Bax and apoptosis was deficient in the DNA-PK null mice (44). The effect of in vivo phosphorylation by DNA-PK on the activity of IGFBP-3 is unknown. However, if IGFBP-3 is a physiological substrate for DNA-PK and if IGFBP-3 is capable of directly regulating gene expression, then its phosphorylation by DNA-PK and the associated increase in binding of the phosphoform to nuclear structures may be related to the regulation of gene expression.
A number of growth factors with classical roles as extracellular signaling molecules, including IGF-I, are specifically transported into the nucleus where they may act directly to induce mitogenic signaling. IGF-I has been detected in the nuclei of epithelial cells from chicken embryonic lens (37) and its nuclear uptake has been described in opossum kidney cells (10). In these cells, IGFBP-3 was shown to cotransport IGF-I, but not IGF-I analogs with reduced affinity for IGFBP-3, into the nucleus. We have described similar findings in T47D human breast cancer cells that had been engineered to express recombinant human IGFBP-3 (38). In addition, in T47D cells that expressed a recombinant form of IGFBP-3 that was unable to translocate to the nucleus, IGF-I was also not transported to the nucleus. Because the ability of IGFs to bind IGFBP-3 was severely attenuated following its phosphorylation with DNA-PK, we speculate that this may affect its ability to act as a carrier for IGF nuclear transport. Of greater interest in this context is the possibility that following cotransport of IGF-I to the nucleus by IGFBP-3, its phosphorylation by DNA-PK may cause the release of bound IGFs, thereby providing a concentrated pool of nuclear IGF-I not seen when the peptide passively diffuses into the nucleus. In support of this, we have shown that IGFBP-3 can be phosphorylated in vitro with DNA-PK when complexed to IGF-I and that phosphorylation leads to the release IGF-I. The nuclear phosphorylation of IGFBP-3 by DNA-PK may also modulate its interactions with other signaling molecules such as nuclear retinoid X receptor- (15).
DNA-PK is activated by DNA strand breaks, such as occur following exposure to DNA-damaging agents. Although the role(s) of IGFs and IGFBP-3 in the nucleus remains unknown, the fact that phosphorylation of IGFBP-3 by this kinase enhances its binding in the nucleus may suggest a role for nuclear IGFBP-3 in DNA damage repair pathways. Our results also suggest a mechanism whereby IGF-I may actively accumulate in the nucleus.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Abbreviations: cdc2, Cyclin-dependent protein kinase; CHO, Chinese hamster ovary; CK2, protein kinase CK2; CLSM, confocal laser scanning microscopy; DNA-PK, double-stranded DNA-dependent protein kinase; DTAF, dichlorotriazinylaminofluorescein I HCl; GST, glutathione-S-transferase; IGFBP, IGF binding protein; [LR3]IGF-I, [long Arg3] IGF-I; BODIPY FL, 6-((4,4-difluoro-5,7-dimethyl-4-bora-3a,4a-diaza-s-indacene-3-propionyl)amino)hexanoic acid; NLS, nuclear localization signal; PKA, cAMP-dependent protein kinase.
Received October 25, 2002.
Accepted for publication January 9, 2003.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
regulate transcriptional signaling and apoptosis. J Biol Chem 275:33607–33613 confers binding to importin-ß and hence transit into the nucleus. EMBO J 15:1810–1817[Medline]
. Biochemistry 40:5208–5217[Medline]
This article has been cited by other articles:
 |
|
 |
|
  L. J. Cobb, H. Mehta, and P. Cohen Enhancing the Apoptotic Potential of Insulin-Like Growth Factor-Binding Protein-3 in Prostate Cancer by Modulation of CK2 Phosphorylation Mol. Endocrinol., October 1, 2009; 23(10): 1624 - 1633. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Yasuoka, E. Hsu, X. D. Ruiz, R. A. Steinman, A. M.K. Choi, and C. A. Feghali-Bostwick The Fibrotic Phenotype Induced by IGFBP-5 Is Regulated by MAPK Activation and Egr-1-Dependent and -Independent Mechanisms Am. J. Pathol., August 1, 2009; 175(2): 605 - 615. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. M. Yamada and K.-W. Lee Perspectives in mammalian IGFBP-3 biology: local vs. systemic action Am J Physiol Cell Physiol, May 1, 2009; 296(5): C954 - C976. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. E. Graham, D. M. Kilby, S. M. Firth, P. J. Robinson, and R. C. Baxter The in Vivo Phosphorylation and Glycosylation of Human Insulin-like Growth Factor-binding Protein-5 Mol. Cell. Proteomics, August 1, 2007; 6(8): 1392 - 1405. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K.-W. Lee, L. J. Cobb, V. Paharkova-Vatchkova, B. Liu, J. Milbrandt, and P. Cohen Contribution of the orphan nuclear receptor Nur77 to the apoptotic action of IGFBP-3 Carcinogenesis, August 1, 2007; 28(8): 1653 - 1658. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R.-A. Toillon, C. Lagadec, A. Page, V. Chopin, P.-E. Sautiere, J.-M. Ricort, J. Lemoine, M. Zhang, H. Hondermarck, and X. Le Bourhis Proteomics Demonstration That Normal Breast Epithelial Cells Can Induce Apoptosis of Breast Cancer Cells through Insulin-like Growth Factor-binding Protein-3 and Maspin Mol. Cell. Proteomics, July 1, 2007; 6(7): 1239 - 1247. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. J. Cobb, B. Liu, K.-W. Lee, and P. Cohen Phosphorylation by DNA-Dependent Protein Kinase Is Critical for Apoptosis Induction by Insulin-Like Growth Factor Binding Protein-3. Cancer Res., November 15, 2006; 66(22): 10878 - 10884. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Zhao, P. Yin, L. A. Bach, and C. Duan Several Acidic Amino Acids in the N-domain of Insulin-like Growth Factor-binding Protein-5 Are Important for Its Transactivation Activity J. Biol. Chem., May 19, 2006; 281(20): 14184 - 14191. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Cohen Insulin-like growth factor binding protein-3: insulin-like growth factor independence comes of age. Endocrinology, May 1, 2006; 147(5): 2109 - 2111. [Full Text] [PDF]
|
|
|
|
 |
|
 B. Liu, K.-W. Lee, H. Li, L. Ma, G. L. Lin, R. A.S. Chandraratna, and P. Cohen Combination Therapy of Insulin-Like Growth Factor Binding Protein-3 and Retinoid X Receptor Ligands Synergize on Prostate Cancer Cell Apoptosis In vitro and In vivo Clin. Cancer Res., July 1, 2005; 11(13): 4851 - 4856. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H.-Y. Lee, H. Moon, K.-H. Chun, Y.-S. Chang, K. Hassan, L. Ji, R. Lotan, F. R. Khuri, and W. K. Hong Effects of Insulin-like Growth Factor Binding Protein-3 and Farnesyltransferase Inhibitor SCH66336 on Akt Expression and Apoptosis in Non-Small-Cell Lung Cancer Cells J Natl Cancer Inst, October 20, 2004; 96(20): 1536 - 1548. [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 |
), PKA (
), MAPK (
), CK2 (
), and cdc2 (
) and purified by reverse-phase HPLC. The ability of the different phosphoforms of IGFBP-3 to bind IGF-I was compared with nonphosphorylated IGFBP-3 (
) using a ligand binding assay (A). The ability of IGF-I and IGF-II to bind different phosphoforms of IGFBP-3 was also examined using Western ligand blotting. One hundred nanograms of nonphosphorylated IGFBP-3 (lane 1) and IGFBP-3 phosphorylated with CK2 (lane 2), PKA (lane 3), DNA-PK (lane 4), cdc2 (lane 5), and MAPK (lane 6) were separated on 10% SDS-PAGE before membrane transfer and incubation with 125I-IGF-I (B) or 125I-IGF-II (C).
), PKA (
), CK2 (
). Following permeabilization of the plasma membrane without (A) [nuclear import] or with permeabilization of the nuclear envelope (B) [nuclear accumulation], CHO cells were incubated in transport solution containing cytosol, an ATP-regenerating system and the different phosphoforms of IGFBP-3 fluorescently labeled with BODIPY. Images were collected at different time points using CLSM and quantitated using NIH Image. The results are expressed as the ratio of nuclear to cytoplasmic fluorescence (Fn/c) as a function of time. Graphs are representative of at least two independent experiments, where the SEM was not greater than 5% of the value of the mean between experiments.
