help button home button Endocrine Society Endocrinology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Endocrinology, doi:10.1210/en.2007-0959
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Iosef, C.
Right arrow Articles by Han, V. K. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Iosef, C.
Right arrow Articles by Han, V. K. M.
Right arrowPubmed/NCBI databases
*Gene*GEO Profiles
*HomoloGene*UniGene
*Compound via MeSH
*Substance via MeSH
Endocrinology Vol. 149, No. 3 1214-1226
Copyright © 2008 by The Endocrine Society

A Functional Nuclear Localization Signal in Insulin-Like Growth Factor Binding Protein-6 Mediates Its Nuclear Import

Cristiana Iosef, Theofanis Gkourasas, Christina Y. H. Jia, Shawn S.-C. Li and Victor K. M. Han

Departments of Biochemistry and Pediatrics (C.I., T.G., C.Y.H.J., S.S.-C.L., V.K.M.H.), Schulich School of Medicine and Dentistry, The University of Western Ontario, London, Ontario, Canada N6A 5C1; and Children’s Health Research Institute and Lawson Health Research Institute (C.I., S.S.-C.L., V.K.M.H.), London, Ontario, Canada N6C 2V5

Address all correspondence and requests for reprints to: Victor Han, M.D., Children’s Health Research Institute, A5-107 Victoria Research Laboratories, 800 Commissioner’s Road East, London, Ontario, Canada N6C 2V5. E-mail: victor.han{at}sjhc.london.on.ca.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
IGF binding protein (IGFBP)-6 is a member of the IGFBP family that regulates the actions of IGFs. Although IGFBPs exert their functions extracellularly in an autocrine/paracrine manner, several members of the family, such as IGFBP-3 and -5, possess nuclear localization signals (NLS). To date, no NLS has been described for IGFBP-6, an IGFBP that binds preferentially to IGF-II. We report here that both exogenous and endogenous IGFBP-6 could be imported into the nuclei of rhabdomyosarcoma and HEK-293 cells. Nuclear import of IGFBP-6 was mediated by a NLS sequence that bears limited homology to those found in IGFBP-3 and -5. IGFBP-6 nuclear translocation was an active process that required importins. A peptide corresponding to the IGFBP-6 NLS bound preferentially to importin-{alpha}. A comprehensive peptide array study revealed that, in addition to positively charged residues such as Arg and Lys, amino acids, notably Gly and Pro, within the NLS, played an important part in binding to importins. Overexpression of wild-type IGFBP-6 increased apoptosis, and the addition of IGF-II did not negate this effect. Only the deletion of the NLS segment abolished the apoptosis effect. Taken together, these results suggest that IGFBP-6 is translocated to the nucleus with functional consequences and that different members of the IGFBP family have specific nuclear import mechanisms.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
THE BIOAVAILABILITY AND the activity of IGFs, IGF-I and IGF-II, are regulated by a family of six IGF-binding proteins (IGFBPs) (1, 2, 3). The majority of IGFs in the circulation exists in the form of a ternary complex with one of the binding proteins, mainly IGFBP-3, and another protein termed acid-labile subunit (3, 4). By associating with IGFs, an IGFBP plays several roles, which include transporting of IGFs, protecting them from degradation, and regulating the interaction between an IGF and the IGF-I receptor. Cell surface-associated IGFBPs can enhance IGF actions by sequestering a pool of IGFs to the vicinity of the receptor and thereby facilitate their interaction (4). Depending on whether it is in extracellular fluids or associated with the cell membrane, a binding protein can either inhibit or promote the action of an IGF (2, 5, 6). These effects are believed to be critical in regulating IGF-dependent biological effects on many cells, including cancer cells (7).

Apart from IGF-binding activities, most IGFBPs have intrinsic cellular functions, such as growth inhibition (8, 9, 10), apoptosis (11, 12, 13, 14, 15, 16, 17, 18), and regulation of gene expression (19), that are independent of binding to IGFs. Although IGFBPs are secreted proteins, some members of the family, such as IGFBP-3 and IGFBP-5, have been detected intracellularly (20, 21, 22). The biological functions of cytosolic IGFBPs, however, are not completely understood. It has been shown that IGFBP-3 binds specifically to uncharacterized proteins present in cellular membranes (23) or cell lysates (24). Expression of IGFBP-3 in several cell lines promotes apoptosis by modulating the expression of apoptosis-specific genes such as Bax and Bcl-2 (17, 18). Overexpression of IGFBP-6 in cells has either a growth-promoting or a growth-inhibitory effect, depending on the cell type (25, 26).

IGFBP-3 and IGFBP-5 are also found in the nuclei of cells. Nuclear translocation of IGFBP-3 or -5 was shown to be an active process that requires the nuclear transport machinery (27, 28). Regulated import of molecules into the nucleus through the nuclear pores is a vital event in eukaryotic cells (29, 30). Whereas small molecules may enter or leave the nucleus through passive diffusion through nuclear pore complexes, trafficking of large macromolecules such as proteins (>45 kDa) between the cytosol and the nucleus is mediated by specific motifs called nuclear localization sequences (NLSs). A typical NLS is composed of one or more clusters of basic residues (31). Monopartite NLS, such as that found in the Simian virus 40 large antigen (126PKKKRKV132), contains a single cluster (32). Bipartite NLS, as typified by that in nucleoplasmin (150KRPAAIKKAGQAKKKKLDK173), contains two clusters of basic amino acids (33). These NLSs are recognized by importin-{alpha}, which forms a dimer with importin-β, and docks the NLS-bearing protein to the nuclear pore complex for translocation through the pore (35, 36, 37). The two basic clusters of a bipartite NLS are separated by a spacer sequence that varies from 10 to 33 residues in length and plays a role in target specificity (38, 39).

The C-terminal domains of IGFBP-3 and -5 contain clusters of basic residues that resemble bipartite nuclear localization sequences. These sequences have been shown to be necessary and sufficient for the nuclear uptake and accumulation of the two IGFBPs (27). IGFBP-3 and, to a lesser extent, IGFBP-5 bound preferentially to importin-β, indicating that their nuclear translocation is mediated primarily by importin-β. To investigate whether this mechanism of nuclear import is also used by other members of the IGFBP family, we investigated extensively in rhabdomyosarcoma (RD) cells, and less in HEK-293 cells, the subcellular localization of IGFBP-6, a relatively unexplored member. RD cells were chosen because they are of skeletal muscle in origin, and our previous studies have shown the abundance of IGF-II in developing skeletal muscle and overexpression of IGFBP-6 in skeletal muscle in conditions of fetal growth restriction (Woode, J.C., C. Iosef, C. Wu, T. Gkourasas, S. S.-C. Li, and V.K.M. Han, unpublished data). We report here the identification of a functional NLS in IGFBP-6 and provide evidence demonstrating that the NLS mediates the nuclear uptake and accumulation of IGFBP-6 in RD cells by selectively binding to importin-{alpha}. Therefore, although all three IGFBPs are capable of translocating to the nucleus, the mechanisms of entry into the nucleus may differ.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Materials
The following materials were used in this study: human recombinant IGF-II (Bachem Co., Torrance, CA); geneticin (Life Technologies, Grand Island, NY); annexin-V-Fluos staining kit (Roche Molecular Biochemicals, Ingelheim, Germany); antibody to IGFBP-6 (Austral Biological, San Ramon, CA); monoclonal antibody to IGF-II (Chemicon International Inc., Temecula, CA); DMEM, fetal bovine serum, and importin-{alpha} and -β (Sigma Chemical Co., St. Louis, MO); antibody to phosphorylated casein kinase-II (CK2)-b(Ser209) (ProSci Inc., Poway, CA); pRC-cytomegalovirus (CMV) and pFLAG vectors (Invitrogen Life Technologies, Carlsbad, CA); phosphorylated enhanced green fluorescent protein (EGFP)-C2 vector (CLONTECH, Palo Alto, CA); placental total RNA (Ambion Inc., Austin, TX); amino-PEG membrane for SPOT peptide synthesis (INTAVIS AG, Koeln, Germany); amino acids (Peptides International, Louisville, KY); and Transfast (Promega, Mississauga, Ontario, Canada).

Plasmids expressing intact human IGFBP-6 and its mutants
A 722-bp cDNA fragment containing the complete coding sequence of human IGFBP-6 (GI:11321593) was obtained by RT-PCR from placental total RNA and inserted between HindIII and BamHI sites in a pEGFP-C2 vector. In addition, an IGFBP-6 C-terminal fragment (IGFBP-6: R163-G240) was also amplified by PCR and inserted between the same sites in pEGFP-C2 vector. After transfection in RD cells, the pEGFP-C2 (IGFBP-6) and pEGFP-C2 (C-IGFBP-6) constructs generated green fluorescent protein (GFP)-fusion proteins (as identified by immunoblot with antibodies against IGFBP-6 and to GFP): an approximately 56-kDa fusion protein (GFP-IGFBP-6) and respectively an approximately 34-kDa (GFP-C-IGFBP-6). The reason for not fusing the NLS fragment alone to the GFP gene was because the increase in size might reduce the capacity of the recombinant protein to passively cross the nuclear barrier. It is known that peptides with a molecular mass closer to 40 kDa have a lower probability to passively locate into the nucleus. The full-length IGFBP-6 cDNA was also cloned into a pRC/CMV vector using HindIII and XbaI sites and expressed subsequently in RD cells, in which it generated an approximately 29-kDa recombinant IGFBP-6 protein with an intact, functional signal peptide. The pEGFP-C2 and pRC/CMV vectors alone were used as controls. A third variant of expression construct, FLAG/IGFBP-6, containing a FLAG tag at the amino terminus of IGFBP-6 was obtained by inserting the cDNA between HindIII and EcoRV sites of pFLAG/CMV-II vector. To generate arginine-specific mutations within the putative NLS sequence, pFLAG/IGFBP-6 and pRc/CMV-IGFBP-6 were subjected to site-directed mutagenesis using the QuickChange kit (Stratagene, Santa Clara, CA). The following oligonucleotides were used: 197R>Q, 5'-CGAGGCTTCTACCAGAAGCGGCAGTGCC-3'; 197/199R>Q, 5'-GGCTTCTACCAGAAGCAGCAGTGCCGCTCCTCC-3'; 208R>H, 5'-GGCTTCTACCAGAAGCAGCGGTGCCGCTCCTCC-3'; and 209R>Q, 5'-CTCTACCAGCGCCAAGGTC-CCTGCTGG-3'.

To delete the putative NLS from IGFBP-6, we used a procedure that involved two rounds of PCR. First, we amplified separately the amino-terminal fragment before NLS as well as the carboxyl-terminal fragment after NLS, and then the two fragments obtained from the first round of PCR were joined. To amplify the entire cDNA, wild-type or mutant, the following specific primers were used: sense, 5'TACGTGCCCTGGTGTGTGGATCG-GATGGGCAAG-3' and antisense, 5'CACACACCAGGGCACGTAGAGTGTTTGAGCCC-3.

The NLS-truncated IGFBP-6 cDNA was subsequently inserted into a pFLAG vector between HindIII and EcoRV sites.

Cell culture
RD cells (American Type Culture Collection, Manassas, VA) were cultured at 37 C in DMEM supplemented with 10% fetal calf serum, penicillin (5 U/ml), streptomycin (5 µg/ml), and gentamicin (25 µg/ml). Cells were passaged once every 3 d. Transfast was used to transiently transfect RD cells with pEGFP-IGFBP-6 or pRC/CMV2-IGFBP-6. Stable transfectants (RD-GFP-IGFBP-6 or RD-pRC-IGFBP-6) were selected by maintaining the culture in 800 µg/ml geneticin (G418) for 21 d. For cell synchronization, nocodazole was used as a G2/M arresting reagent (0.4 µg/ml). Cells were exposed to nocodazole for 20 h and then release at different time points during mitosis and analyzed for IGFBP-6 expression as well as mitosis or apoptosis markers.

Nuclear import assay
Digitonin-permeabilized RD cells were prepared as described originally by Adam et al. (40). Cells grown on glass coverslips were permeabilized with 55 µg/ml digitonin (Sigma) in transport buffer containing 20 mM HEPES (pH 7.3), 110 mM potassium acetate, 1 mM EGTA, 2 mM dithiothrietol, and proteases inhibitors cocktail (Sigma). Nuclear import assays were performed in the transfer buffer containing an energy regenerating system (1 mM ATP, 10 mM creatine phosphate, and 0.4 U/ml creatine phosphokinase) and 40 µg rabbit reticulocyte lysate (Promega) as a cytosol replacement source. To assay for nuclear import of IGFBP-6, purified GFP-IGFBP-6 or fluorescein isothiocyanate (FITC)-labeled IGFBP-6 (0.1 µM) was added to the above mixture. GFP-IGFBP-6 was purified on an immunoaffinity column. IGFBP-6-FITC was purified by size exclusion FPLC on a Sephadex G75. Nuclear import was terminated by washing the cells in cold PBS and fixation in 3.6% paraformaldehyde for 30 min. To compete IGFBP-6 effects, we used a physiological dose of IGF-II of 100 ng/ml (Bachem). The IGF-II peptide was identified by immunocytochemistry using a primary anti-IGF-II antibody (1:30 dilution) from Novozymes GroPep Ltd. (Adelaide, Australia) and a secondary antirabbit IgG tetramethylrhodamineithio-cyanate (TRITC)-labeled antibody (1:50 dilution) (Jackson ImmunoResearch Laboratories, Hornby, Ontario, Canada). The incubation with the primary anti-IGF-II antibody was for 3 h at 37 C, whereas the secondary antibody was incubated for 45 min at room temperature. The immunochemistry reactions were performed in humidified chambers.

Confocal microscopy analysis
Confocal analysis was performed on a Zeiss-410 Ar/Kr laser scanning confocal microscope with dual- and triple-channel detection optics (Carl Zeiss, Gottingen, Germany; purchased through Carl Zeiss Canada Ltd., Toronto, Ontario, Canada), interfaced to a Zeiss 400 Axiovert microscope with Zeiss Plan-Apo x63 oil immersion objective or Zeiss x40 objective. The samples were analyzed with appropriate filters using single or simultaneous excitation at {lambda} = 488 nm (green, GFP, or FITC), {lambda} = 568 nm (red, TRITC, or propidium iodide), or {lambda} = 320 nm [blue; Hoechst (Molecular Probes, Eugene, OR)]. Images were collected under identical nonsaturated conditions after multiple scans averaging of eight to 12 sections per cell. For a better display of the nuclear staining, some of the images were magnified x200. The intensity of fluorescence was measured within cytoplasm or nuclei as well as outside the cells by analyzing the pixel intensity using a Fluor-S MultiImager (Bio-Rad Laboratories, Hercules, CA; purchased through Bio-Rad Ltd., Mississauga, Ontario, Canada). The pixel intensity was averaged for at least 100 cells in triplicate samples (~300 transfectants) for each subcellular region; all values were corrected for background fluorescence. The results were expressed as the ratio of nuclear to cytoplasmic fluorescence (Fn/c).

Immunocytochemistry
RD wild-type and RD-IGFBP-6 cells were grown for 24 h on glass chamber slides as described above and fixed with 3% formaldehyde/PBS solution for 15 min. Fixed cells were treated with lysis buffer (containing 0.1% sodium citrate, 0.1% Triton X-100) for 10 min. For IGFBP-6 detection, cells were incubated for 2 h at 37 C with a rabbit antihuman-IGFBP-6 antibody (1:500; Austral Biologicals) in PBS-1% BSA. After washing (3 x 10 min) in PBS, the cells were incubated for 1 h at 37 C with a goat antirabbit antibody labeled with TRITC or FITC (Jackson ImmunoResearch Laboratories) or conjugated to horseradish peroxidase (HRP). In the latter case, diaminobenzidine was used as a substrate for HRP and staining agent for the immunocomplexes. After incubation with the secondary antibody, cells were washed three times (3 x 10 min) and subjected to chromatin staining by propidium iodide (2 µg/ml) or Hoechst reagent (1 µg/ml). Fluorescent imaging was performed using on a confocal-laser scanning microscopy (krypton/argon laser scanning microscopy; Zeiss); for printing purposes images were submitted to Photoshop (Adobe, San Jose, CA) processing.

Cell fractionation
Scraped cells were washed in cold Tris-buffered saline [TBS; 10 mM Tris base (pH 8), 150 mM NaCl, 1 mM MgCl2] and resuspended in hypotonic buffer containing 10 mM Tris-HCl (pH 8.0), 0.5 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride, and the Sigma protease inhibitor cocktail. After incubation for 10 min on ice, the cell lysate was vortexed and homogenized. Nuclei were removed from the homogenates by centrifugation at 3000 x g for 1 min in an Eppendorf tube (39). Nuclear pellets were washed in the hypotonic buffer and resuspended in the extraction buffer supplemented with protease inhibitor cocktail and incubated on ice for 20 min. The supernatant was collected by centrifugation at 3000 rpm for 10 min and assayed for content of the nuclear fraction by Western blotting using an anti-poly-ADP-ribose polymerase (PARP) antibody. The exnuclear cell lysate containing the cytoplasmic and membrane fractions was centrifuged at 50,000 rpm for 60 min in a SW60 rotor (Beckman, Coulter Canada Inc., Mississauga, Ontario, Canada), and the resulting pellet (which should contain the membrane fraction) and the supernatant (which should correspond to the cytoplasmic fraction) were collected separately.

Western blotting
Protein fractions from the cell lysates were separated by 12.5–15% SDS-PAGE and transferred onto nitrocellulose or polyvinyl difluoride membranes (Millipore, Bedford, MA) (41). The membranes were blocked with 4% BSA-TTBS (0.5% Tween 20 in TBS) for 1 h at room temperature. The membranes were washed in TTBS (3 x 10 min) followed by incubation at 4 C overnight with rabbit anti IGFBP-6 polyclonal IgG (1:1000) in 1% BSA-TTBS. The membranes were then incubated with HRP-labeled goat antirabbit IgG (1:4000) in 1% BSA-TTBS for 1 h at room temperature. Identification of the tagged proteins was similar and was performed with antitag antibodies (anti-6xHis or -FLAG) following manufacturer protocols. The immunocomplexes were detected by enhanced chemiluminescence according to the manufacturer’s protocol (Amersham Pharmacia Biotech, Quebec, Canada) and documented on Biomax films (Kodak Laboratories, Rochester, NY).

Peptide spot arrays
Peptides were synthesized as arrayed spots on an amino-PEG membrane using an ASP222 autospot robot (Abimed GmbH, Langenfeld, Germany) following the protocol described previously (42). Peptides corresponding to the putative NLSs in IGFBP-6, IGFBP-3, and IGFBP-5 were synthesized. Arrays of Ala-scanning peptides were also generated for each NLS peptide. Binding of the peptide spots to either importin-{alpha} or -β was assayed by far-Western blotting using 0.5 µg/ml importin protein (containing a 6xHis tag) diluted in TBS. An anti-6 x His monoclonal antibody (QIAGEN Canada, Mississauga, Ontario, Canada; diluted at 1:500 in 2% BSA/TBS) was used to detect the NLS-importin complexes. The membrane was subsequently incubated for 45 min with an alkaline phosphatase-conjugated antimouse antibody (Bio-Rad) and developed using an ECF kit (Amersham). Fluorescent signals were captured and analyzed on a Fluor-S MultiImager (Bio-Rad).

Prediction of IGFBP-6 NLS target sequence
IGFBP-6 putative NLS sequence was analyzed by the PSORT programs currently available in the European Molecular Biology Laboratory database (http://psort.ims.u-tokyo.ac.jp/cgi-bin/runpsort.pl). Briefly, to detect a potential NLS, these programs use the following rules: four residue pattern composed of basic amino acids (K or R) or composed of three basic amino acids (K or R) and H or P; a pattern starting with P and followed within three residues by a basic segment containing three K or R residues of four residues. PSORT programs also use a heuristic that nuclear proteins are generally rich in basic residues: if the sum of K and R is higher than 20%, then the protein is considered to have a higher possibility of being nuclear than cytoplasmic. The reliability of the prediction results is estimated by the authors of the current versions of PSORT programs with a score of 66–83% accuracy when applied to known sequences.

Assessment of apoptosis onset
RD-IGFBP-6 cells and controls (vector transfected cells and IGFBP-6{Delta}NLS expressing cells) were cultivated in serum-free conditions for 72 h. At 24, 48, and 72 h after seeding, cells were analyzed for the presence of phosphatidylserine on the membrane, as an early effect of apoptosis. For this purpose the annexin-V-Fluos kit was used following manufacturer’s protocol (Roche Molecular Biochemicals). Nuclear chromatin was stained with propidium iodide. Annexin V-FITC-positive cells were identified and analyzed both by flow cytometry and confocal microscopy.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Both exogenous and endogenous IGFBP-6 accumulate in the nuclei of RD cells
Embryonic RD cells are known to express IGF-II endogenously, an autocrine growth factor that is regulated by IGFBP-6 (26). We chose RD cells as a model system of skeletal muscle to examine the intracellular trafficking of IGFBP-6 because of its potential relevance to IGF-II function in skeletal muscle cell growth. RD cells were stably transfected with the expression construct pRc/CMV containing full-length IGFBP-6 cDNA and selected via G418 restriction. IGFBP-6 proteins present in the cell lysate or secreted into the culture medium were assayed by Western blotting. In comparison with the original RD cells, the stable clones expressed more IGFBP-6, which was reflected by elevated IGFBP-6 levels in both cell lysate and culture medium (Fig. 1AGo).


Figure 1
View larger version (25K):
[in this window]
[in a new window]

  		FIG. 1. IGFBP-6 is found both in the culture medium and nuclei of RD cells. A, RD cells stably express IGFBP-6 (RD-IGFBP-6). The levels of IGFBP-6 in the medium and cell lysate were assayed by Western blots for both the untransfected parental line and the IGFBP-6 cell clone using a rabbit antihuman-IGFBP-6 antibody. An anti-β-actin blot is shown to demonstrate equal loading of samples. B, Confocal microscopic images of RD cells overexpressing GFP or GFP-IGFBP-6(NLS) (Zeiss-410, objective x63, oil immersion). C, Immunostaining of endogenous IGFBP-6 in RD cells. IGFBP-6 was detected using a rabbit anti-IGFBP-6 antibody and visualized using an HRP-conjugated, antirabbit secondary antibody. Diaminobenzidine (DAB, brown) was used to stain the immunocomplexes. Cells were visualized by standard microscopy (objective x63, oil immersion). IHC, Immunohistochemistry. D, Schematic of IGFBP-6 showing N terminus (N), linker domain (L), and carboxyl terminus (C). The NLS motif is located within the C domain inclusive of residues 164–210.

 
Because IGFBP-6 was present at a high level in the cell lysates, we examined its subcellular localization. To verify the functionality of the nuclear localization signal, an IGFBP-6 carboxyl-terminal fragment of 48 amino acids including the assumed NLS was fused to the EGFP gene and expressed in RD cells. Confocal microscopic images of transfected cells showed that the green fluorescence was concentrated mostly in the nuclei (64.4 ± 3%), whereas the control cells expressing EGFP showed green fluorescence distributed uniformly, mainly in the cytosol (86.3 ± 4%) (Fig. 1BGo). Because GFP alone was seen in the cytoplasm of 86.3% of cells and the nucleus of only 13.7% of cells, the observed nuclear accumulation of GFP-IGFBP-6 and the restriction of the EGFP distribution within the cells up to 50.7% could be attributed to the IGFBP-6 fragment that contained the NLS. Figure 1BGo represents RD cells transiently transfected with EGFP or EGFP-IGFBP-6(NLS), and photomicrographed at 48 h after transfection; 10 ± 2% of cells expressed EGFP or EGFP-IGFBP-6(NLS). When full-length IGFBP-6-EGFP construct was transfected, the recombinant protein was distributed mainly in the cytoplasm (59 ± 4% of the cells), but 30.1 ± 3% was distributed in the perinuclear region and 10.6 ± 1.2% in the nucleus. Proportions were calculated from the total number of cells counted in triplicate samples in 100 microscopic fields (x20 magnification). It was unlikely that the nuclear translocation was due to passive diffusion because the size of the GFP-IGFBP-6 fusion protein (56 kDa) exceeded the upper limit (~45 kDa) allowed by the nuclear pores. To ascertain that this phenomenon was not caused by GFP-IGFBP-6 overexpression, we also immunostained endogenous IGFBP-6 in RD cells and found that it was present in both the cytosol and nucleus (Fig. 1CGo). Taken together, these data suggest that IGFBP-6 is not only a secreted protein but it is also present in the cytoplasm and inside the nucleus.

Identification of a putative NLS of IGFBP-6
To date, two IGFBPs, namely IGFBP-3 and IGFBP-5, have been shown to possess functional NLSs. Sequence alignment reveals a major and a minor cluster of basic residues conserved between the two proteins (27). A basic motif, KKK for IGFBP-3 and KRK for IGFBP-5, is embedded within an 11-residue spacer that separates the minor and major basic clusters (Fig. 2AGo). These characteristics of the IGFBP-3 and -5 NLSs are partially conserved in IGFBP-6. However, despite having a minor basic cluster (192H193R) and a 197RKR199 basic motif, the IGFBP-6 sequence lacks the major basic cluster found in IGFBP-3 and IGFBP-5. The corresponding region in IGFBP-6, which may be described more appropriately as a pseudomajor basic cluster, contains only two basic residues, compared with four each for IGFBP-3 and IGFBP-5 (Fig. 2AGo). This is probably why IGFFBP-6 has not been considered a nuclear protein to date. In addition, we detected a pair of arginine residues (165RR166) upstream of the minor cluster that are conserved among all three proteins. These dual-Arg residues may constitute another minor basic cluster for the IGFBP proteins.


Figure 2
View larger version (39K):
[in this window]
[in a new window]

  		FIG. 2. IGFBP-6 contains a functional nuclear localization signal. A, Alignment of a putative NLS in IGFBP-6 against the NLS sequences identified in IGFBP-5 and IGFBP-3. Positive charged Arg, Lys, and His residues are shown in bold. Residues mutated in subsequent studies are identified in red. The major basic cluster, the two minor basic clusters, and a triamino acid, basic cluster within the spacer region of IGFBP-3 are underlined. B, Deletion of the predicted NLS led to loss of IGFBP-6 nuclear translocation. Wild-type IGFBP-6 or IGFBP-6{Delta}NLS, a mutant that contains a deletion of the NLS (residues 192–210, specifically), were tagged with FLAG and transiently expressed in RD cells. The distribution of the proteins in the cells was revealed using a monoclonal anti-FLAG antibody and an antimouse secondary antibody labeled with TRITC (red). Chromatin was stained in blue using Hoechst dye. Cells were imaged on a Zeiss-410 confocal laser-scanning microscope. C, IGFBP-6 and IGFBP-6{Delta}NLS fused to GFP were expressed transiently in RD cells. Confocal microscopy was used to monitor the localization of the green fluorescent fusion proteins. DNA was stained by propidium iodide. Left panels, Green; right panels, green and red. D, Transfection of IGFBP-6 C-GFP and wtIGFBP-6-GFP into HEK293 cells reveals that the nuclear localization signal of IGFBP-6 is also functional in human embryonic kidney cells. Images of fixed cells show the distribution of IGFBP-6 in both the cytoplasm and nucleus, as confirmed by immunostaining with an anti-IGFBP-6 antibody. Immunocomplexes were revealed by using a TRITC secondary antibody. Images from the left panel, representing the distribution of IGFBP6-C-GFP proteins in live cells have been captured with a spin-disc confocal microscope at a magnitude of x40.

 
To investigate whether IGFBP-6 possesses a functional NLS, we generated a mutant of IGFBP-6 that contained a deletion of residues 188–211 that spans the predicted NLS. The mutant and wild-type (WT) IGFBP-6 was expressed in RD cells as either FLAG-tagged proteins or GFP-fusion proteins. Whereas 12.6 ± 2% of the cells expressing the FLAG-tagged WT protein showed the protein concentrated in the nuclei of RD cells, the NLS-deleted mutant, IGFBP-6{Delta}NLS, was detected only in the cytoplasm of transfected cells (Fig. 2BGo). Similarly, EGFP-IGFBP-6 was found in the nucleus of 10.9 ± 1.6% transfected cells, whereas the NLS-deleted mutant was seen exclusively in the cytosol (Fig. 2CGo). These results demonstrate that the fragment spanning residues 170–188 constitutes a functional NLS for IGFBP-6. These cells were examined at a low rate of transfection of about 10–13%. Higher rates of transfection for IGFBP-6 induced higher expression of this protein in the RD cells but also apoptosis. To avoid programmed cell death, the rate of transfection was kept low but similar between the two constructs including the WT-IGFBP-6 or IGFBP-6{Delta}NLS. The NLS-truncated IGFBP-6 did not induce apoptosis at higher rates of transfection than 10%.

To determine the role of each of the amino acids of the putative NLS, a series of mutants, each of which contained one or more Arg residues within the putative NLS (highlighted in red in Fig. 2AGo) mutated to Gln, were constructed. Mutants R197Q, R199Q, and R(197/9)Q were generated to explore the role of positively charged residues in the spacer. Mutations of corresponding residues in IGFBP-3 led to approximately 60% reduction in its nuclear accumulation (27). Mutant R209Q was created to investigate the importance of positive charge in the pseudomajor basic cluster. Replacement of the major basic cluster in IGFBP-3 by an unrelated sequence resulted in the abolition of nuclear accumulation of the protein (27). A mutant bearing all three mutations, R(197/9209)Q, was also constructed to test whether the effects of these mutations were additive. Because the second minor basic cluster was shown not to play a role in nuclear translocation of IGFBP-3 (27), we investigated the role of the first minor basic cluster using the double mutant R(164/5)Q and quadruple mutant R(165/6, 197/9)Q.

The WT and mutant proteins were each tagged with the FLAG epitope to distinguish them from endogenous IGFBP-6. The subcellular localization of transiently expressed FLAG-IGFBP-6 or a mutant was monitored by immunocytochemistry using an anti-FLAG monoclonal antibody and a TRITC-conjugated secondary antibody. For a quantitative evaluation of protein nuclear localization, the number of cells displaying fluorescent signal in the nucleus greater than in the cytoplasm was counted, and the percentage of these cells of a total of 300 transfectants was calculated. As shown in Fig. 3AGo, mutation of either Arg197 or Arg199 to Gln caused a significant reduction in the percentage of cells with nuclear accumulation of IGFBP-6, compared with that for the WT protein. A mutant bearing both mutations had a more pronounced reduction in nuclear IGFBP-6, as expected. Nuclear accumulation of mutant R209Q was diminished to approximately 10% of that of the WT protein, suggesting that the positive charge in the pseudomajor basic cluster of IGFBP-6 plays a pivotal role in its nuclear uptake and accumulation. As expected, the loss of positive charges in both the pseudobasic cluster and the spacer region as in the triple mutant R(197/9,209)Q completely eliminated nuclear accumulation of the mutant protein, a phenotype that was shared by the {Delta}NLS mutant (Fig. 3AGo). In contrast, nuclear accumulation was only slightly reduced for mutant R(165/6)Q, suggesting that the first minor basic cluster is not essential for the nuclear translocation of IGFBP-6. This observation was also supported by the finding that the quadruple mutant R(164/5, 197/9)Q did not exhibit further attenuation in protein nuclear accumulation, compared with mutant R(197/9)Q (Fig. 3AGo).


Figure 3
View larger version (21K):
[in this window]
[in a new window]

  		FIG. 3. The role of Arg residues in the nuclear localization of IGFBP-6. A, IGFBP-6 mutants bearing Arg to Gln mutations within the putative NLS display attenuated nuclear accumulation. RD cells transfected with FLAG-tagged IGFBP-6 or a mutant were stained with an anti-FLAG monoclonal primary antibody and then a secondary antibody labeled with TRITC. Nuclei were stained with Hoechst dye. Cells showing nuclear red fluorescence were counted and plotted as percentage of all cells displaying red fluorescence in both the nucleus and cytoplasm. B, Distribution of IGFBP-6 and mutants in RD cells. Cells transfected with pFLAG-IGFBP-6 or a mutants were lysed and fractionated into nuclear and exnuclear fractions by centrifugation. The levels of IGFBP-6 protein in these fractions were determined by Western blotting using an anti-FLAG antibody. An anti-PARP immunoreactivity was used to identify the nuclear fraction, whereas an anti-β-importin immunoreactivity was used to show equal loading of the exnuclear fractions. {Delta}NLS, NLS mutant; IB, immunoblotting.

 
To confirm the mutant phenotypes described above, we fractionated the corresponding cell lysates and assayed for the amount of proteins contained in the nuclear vs. the cytoplasmic/membranous fractions for each mutant by Western blotting. The nuclear fraction was identified by the presence of the nuclear protein, PARP, whereas the exnuclear fraction was confirmed by β-actin immunoreactivity. As shown in Fig. 3BGo, a significant amount of WT IGFBP-6 was detected in the nuclear fraction. In contrast, the relative amounts of protein detected in the nucleus for mutants R197, R199, or R197/9Q were significantly reduced. No IGFBP-6 immunoreactive band was detected in the nuclear fraction of mutants R209Q, R197/9,209Q, or {Delta}NLS. Reduction in nuclear IGFBP-6 was relatively small for mutant R164/5Q or R164/5, 197/9Q, compared with that for the WT protein (Fig. 3BGo). These results are in agreement with those obtained from immunocytochemical analysis.

IGFBP-6 structure was additionally scanned for a nuclear export signal (NES) using NetNES 1.1 server (Technical University of Denmark, Lyngby, Denmark), and we identified two putative NES motifs: one of high confidence, located in the signal peptide (sequence:10L-21S) and another of low confidence (sequence: 88K-105L), in which the most powerful residues are: 99L100G and 104C105L. The first putative NES is located in the signal peptide, which would be expected to be cleaved off because IGFBP-6 is a secreted peptide. The second putative NES could be functional; however, the prediction analysis showed that it might not be highly active.

Nuclear import of IGFBP-6 is a saturable and an IGF-II-independent event related to apoptosis
To further establish that nuclear import of IGFBP-6 is an active process, IGFBP-6 was expressed by recombinant methodology, purified, and labeled with FITC. The fluorescent protein was allowed to incubate with digitonin-permeabilized RD cells in the presence of a fully reconstituted transport system that included rabbit reticulocyte lysate and an ATP-regenerating system (27, 40). In parallel experiments, unlabeled IGFBP-6 was added in 10- or 20-fold excess. The distribution of green fluorescence in cells was monitored by confocal microscopy, and the cell images obtained were subsequently analyzed on a Fluoro-S MultiImager, which allowed for the quantification of green fluorescent signal in the nucleus vs. that in the cytosol. The Fn/c ratio is calculated for each import assay (Fig. 4AGo). When the FITC-IGFBP-6 was applied alone, this ratio was 5.7, indicating that the majority of the protein was imported to the nucleus. However, the ratio decreased to approximately 3.0 and 1.5, respectively, in the presence of 10- and 20-fold competing IGFBP-6. These results show that the nuclear import of IGFBP-6 is a saturable process, similar to IGFBP-3 or IGFBP-5 (27).


Figure 4
View larger version (29K):
[in this window]
[in a new window]

  		FIG. 4. Nuclear import of IGFBP-6 in RD cells is a saturable process, apparently independent of IGF-II actions. A, RD cells were permeabilized by digitonin and incubated with a fully reconstituted transport system and fluorescein-labeled IGFBP-6 in the absence or presence of 10- to 20-fold excess of unlabeled IGFBP-6. Images were collected on a confocal microscope and analyzed on a Fluoro-S Imager (Bio-Rad). The results were plotted as Fn/c ratios with SEs. B, Digitonin-permeabilized RD cells were incubated with fully reconstituted transport system and IGFBP-6-FITC in the presence or absence of 300 nM IGF-II. Images were generated in a Zeiss-410 laser-scanning confocal microscope and are representative of two independent experiments. In green, it is IGFBP-6-FITC; in red, is IGF-II, which was immunostained with a rabbit polyclonal antibody to IGF-II and a TRITC-red labeled secondary antibody; in blue, chromatin stained with Hoechst dye. C, Diagram represents the percentages of annexin V-positive cells, as demonstrated by flow cytometry analysis. RD-IGFBP-6 cells underwent apoptosis in serum-free conditions. RD-IGFBP-6{Delta}NLS cells showed a similar profile to the control cells. The apoptosis effect could not be reversed by supplementing the medium with IGF-II, which is the main ligand for IGFBP-6. IGF-II peptide was used at a concentration of 100 ng/ml.

 
Because IGFBP-6 binds preferentially to IGF-II, we determined whether IGF-II played a role in the nuclear import of IGFBP-6. To this end, digitonin-permeabilized RD cells were incubated with FITC-IGFBP-6 and a complete transport system in the presence or absence of IGF-II (100 ng/ml). As shown in Fig. 4BGo, nuclear accumulation of fluorescent IGFBP-6 was not affected by the presence of IGF-II, which was found to locate primarily in the cytosol as revealed by immunostaining using an IGF-II-specific antibody. Endogenous IGF-II was also found to reside in either the cytoplasmic or perinuclear space (Fig. 4BGo, lower panel). These data indicate that nuclear import of IGFBP-6 is apparently an IGF-II-independent event. We further investigated the additional functions that may be attributable to nuclear IGFBP-6. RD-IGFBP-6WT cells underwent apoptosis in serum-free conditions after 48 h with a 2-fold higher rate than the vector control cells (Fig. 4CGo) as demonstrated by annexin-V staining, followed by flow cytometry. However, RD cells overexpressing IGFBP-6{Delta}NLS showed a similar profile to the control cells. The peak of apoptosis was at 48 h after treatment, and this effect could not be reversed by supplementing the medium with IGF-II (at a physiological dose), which is the IGFBP-6 main ligand. On the other hand, at 72 h the number of the apoptotic cells was still higher in the IGFBP-6WT-overexpressing cells, compared with controls, regardless of the presence of IGF-II, showing that depletion in growth factors enhanced the apoptosis, but this process remained associated only with the IGFBP-6WT expression and not with the IGFBP-6{Delta}NLS mutant. These data indirectly support the fact that NLS is, at least in part, responsible for the apoptosis process. Overall, these results suggest that the apoptosis events were not related to the IGFBP-6 secreted into the medium or the sequestering of IGF-II by IGFBP-6. Because the deletion of the NLS segment negated the effects of overexpressed IGFBP-6; therefore, we concluded that the nuclear localization of this protein was related to apoptosis and significantly contributed to this effect.

Nuclear import of IGFBP-6 is dependent on importin-{alpha} and energy
Nuclear import assays were carried out with selective exclusion of components, such as ATP and soluble cytosolic factors, from the transport system to examine their respective roles in IGFBP-6 nuclear import. The subcellular localization of IGFBP-6 under these conditions was monitored by laser-scanning confocal microscopy. As shown in Fig. 5Go, the absence of ATP led to the disappearance of fluorescence from the nucleus, suggesting that nuclear import of IGFBP-6 was an energy-dependent process. Similarly, omitting ATP from the transport system led to exclusion of IL-5, a protein that contains a classic NLS, from the nucleus. Exclusion of soluble cytosolic factors from the system, however, resulted in only a moderate decrease in nuclear uptake of IGFBP-6. This was in contrast to IL-5, whose nuclear localization was essentially abolished under the same conditions (Fig. 5Go). A significant reduction in nuclear accumulation of IGFBP-6 was observed when an antibody specific to importin-{alpha} was added to the transport system (Fig. 5Go), indicating that importin-{alpha} and, by inference, the importin-{alpha}/β heterodimer played an important role in the nuclear import of IGFBP-6. Because it is known that nuclear import of IL-5 is mediated by importin-{alpha}, the same antibody was also used to block the nuclear uptake of IL-5. As expected, this treatment led to a complete inhibition of nuclear import of IL-5 (Fig. 5Go). These data indicate that nuclear import of IGFBP-6 and IL-5 do not follow the same mechanism. In control experiments, the IGFBP-6 {Delta}NLS mutant was found in the cytosol under all conditions examined, suggesting that the NLS was necessary for its translocation into the nucleus and that passive diffusion of the protein into the nucleus had a negligible effect in nuclear accumulation of the protein. Collectively, these results suggest that nuclear import of IGFBP-6 is an active process that is partially dependent on importin-{alpha} but independent of soluble factors.


Figure 5
View larger version (47K):
[in this window]
[in a new window]

  		FIG. 5. Nuclear import of IGFBP-6 is dependent on energy and importin-{alpha}. After plasma membrane permeabilization with digitonin, RD cells were incubated with FITC-labeled IGFBP-6 in a fully reconstituted nuclear transport system in a system that either omitted the ATP-regenerating system or the cytosol or in a transport system that included anti-importin-{alpha} antibodies. Parallel experiments were conducted on IL-5 and the IGFBP-6 {Delta}NLS mutant. Images shown are representative of three sets of independent experiments. Cells were visualized by laser-scanning confocal microscopy (Zeiss-410 microscope, objective x63, oil immersion).

 
IGFBP-6 interacts directly with importin-{alpha}
To dissect the molecular mechanism of IGFBP-6 nuclear import, we investigated a potential interaction between IGFBP-6 and importin-{alpha}. His-tagged importin-{alpha} was added to the lysate of RD cells expressing FLAG-IGFBP-6. The mixture was subjected to immunoprecipitation using either an anti-FLAG or anti-6 x His antibody. Immunoblots using the same antibodies showed that IGFBP-6 was present in the anti-His 6x-importin-{alpha} precipitates and vice versa (Fig. 6AGo), indicating that IGFBP-6 forms a complex with importin-{alpha} in vitro. In contrast, the IGFBP-6{Delta}NLS truncated form failed to coimmunoprecipitate with importin-{alpha} under the same conditions, suggesting that the NLS directly mediated the interaction between the two proteins. Far-Western-blot analysis using the antibody against WT-IGFBP-6 (Austral Biologicals) recognized protein complexes (both monomers and dimers) among importin-{alpha} immunoprecipitation products (data not shown). However, this antibody could not be used for IGFBP-6 immunoprecipitation reactions. The coimmunoprecipitation reactions were possible only if using IGFBP-6 and importin-{alpha}-enriched cell lysates.


Figure 6
View larger version (30K):
[in this window]
[in a new window]

  		FIG. 6. IGFBP-6 binds selectively to importin-{alpha}. A, Purified importin-{alpha} (tagged with 6xHis) was mixed with lysate of RD cells that overexpress FLAG-tagged IGFBP-6. The mixture was subjected to immunoprecipitation by either an anti-FLAG or an anti-6 x His antibody. The resulting blots were developed using the same antibodies as indicated on the figure. IB, Immunoblotting. B, Binding of purified importin-{alpha} to peptides corresponding, respectively, to the NLS sequences of IGFBP-3, -5, -6, and -1. Graph shown is based on quantification of the actual binding data (shown on the left) using the Bio-Rad Fluoro MultiImager. C, Binding of purified importin-β to the NLS peptides derived from IGFBP-3, -5, and -6 and to a control peptide from IGFBP-1. Data processing and graphing followed the same procedures as used in B.

 
Differential binding of NLSs from IGFBP-6, -5, and -3 to importin-{alpha} and -β
To validate the above concept, we examined whether the IGFBP-6 NLS alone could be recognized by importin-{alpha}. A peptide corresponding to residues 175–188 of IGFBP-6 (Fig. 2AGo) was synthesized. For the purpose of comparison, two peptides representing, respectively, the IGFBP-3 and -5 NLS and a peptide derived from the homologous region in IGFBP-1 were synthesized. These peptides were assembled on a functionalized nitrocellulose membrane using the SPOT method of multiple peptide synthesis (42). The peptide spots were subsequently probed for binding to purified 6xHis-importin-{alpha}, and the bound proteins were detected using an anti-6 x His antibody in a far-Western blot. As seen in Fig. 6BGo, NLS peptides from IGFBP-3, -5, and -6 were all capable of binding to importin-{alpha}. The IGFBP-6 NLS bound most strongly to importin-{alpha}, followed by the IGFBP-5 NLS, whereas the IGFBP-3 NLS peptide was much weaker in binding in comparison with the other IGFBPs. The control peptide from IGFBP-1, which does not contain a functional NLS, was completely inactive in binding.

Because it was reported that IGFBP-3 interacts preferentially with importin-β (27), we next investigated binding of the NLS peptides to purified importin-β. As shown in Fig. 6CGo, the IGFBP-3 NLS bound most strongly to importin-β in the group. Significant binding was also observed for the IGFBP-5 peptide. In contrast, the IGFBP-6 NLS bound only weakly to importin-β. Taken together, these data demonstrate that the IGFBP-3 NLS interacts preferentially with importin-β, whereas the IGFBP-6 NLS binds more favorably to importin-{alpha}. The IGFBP-5 NLS, on the other hand, is capable of binding both importins but does not display a preference for either, unlike IGFBP-3 or -6.

Identification of key residues in an NLS for binding to importin-{alpha} or -β by Ala-scanning peptide arrays
To further define residues in the NLS region that are critical for binding to importin-{alpha} or -β, a series of peptide analogs were synthesized, each of which contained the same NLS sequence except that one residue was replaced by an Ala. The resulting array of Ala-scanning peptides was then probed for binding to either importin-{alpha} or importin-β. The same procedures were carried out for each NLS peptide (from IGFBP-3, -5, or -6), and the results obtained are displayed in Fig. 7Go. For binding to importin-{alpha}, none of the residues in IGFBP-3 NLS appeared to play an essential role because greater than 50% of affinity was retained for every Ala-substituted peptide. This is consistent with our earlier observation (Fig. 6Go, B and C) that the IGFBP-3 NLS selectively bound to importin-β rather than importin-{alpha}. In contrast, higher losses of binding affinity were evident for the IGFBP-6 NLS peptide when one of the basic residues, such as R197, K198, R199, R208, and R209, was replaced by an Ala. The greatest reduction in affinity was associated with replacement of R209, reconfirming our observation that a form of IGFBP-6 bearing the same mutation was the most incompetent in nuclear translocation (Fig. 3Go). Interestingly, Gly206A, which is located within the boundaries of the pseudomajor basic cluster of the IGFBP-6 NLS, was also found to be important for binding because its substitution by an Ala resulted in an approximately 50% reduction in affinity. In contrast, substitution of Cys201A, a residue conserved among all NLS examined, led to a significant increase in affinity (Fig. 7Go). The binding profile of the IGFBP-5 NLS peptides to importin-{alpha} was similar to that for the IGFBP-3 peptides, although the former NLS exhibited a stronger dependence on positively charged residues, especially those in the major basic cluster.


Figure 7
View larger version (23K):
[in this window]
[in a new window]

  		FIG. 7. Mapping of key residues in the NLS sequences for binding to importin-{alpha} (Imp-{alpha}) or -β (Imp-β) by Ala-scanning peptide arrays. Each NLS peptide from IGFBP-3, -5, or -6 was subjected to Ala-scanning substitutions, and the resulting peptides were synthesized as spot arrays. The binding profile of each NLS peptide array to either importin-{alpha} or -β was shown both in raw data form and as a plot of relative affinities to that of the wild-type (wt) sequence. The original sequence of an NLS is shown underneath the corresponding graph with amino acid numbering listed between the peptide sequence and the peptide spots. Hashed bars indicate residues whose substitution by Ala resulted in greater than 50% loss of binding.

 
The respective binding profiles of the IGFBP-3 and IGFBP-6 NLS peptide arrays for importin-β were dramatically different from those for importin-{alpha}. Whereas the former NLS displayed a strong sequence dependence for binding to importin-β, the latter NLS was largely insensitive to Ala substitutions. Significant differences were also detected between the IGFBP-3 and -5 NLS peptides. These data are again in agreement with those from previous studies demonstrating that IGFBP-3 bound more strongly to importin-β than did IGFBP-5 (27). Close inspection of the binding profiles revealed that the nine residues in the 16-mer NLS peptide of IGFBP-3, which were scattered in both the major cluster and the spacer, were indispensable for binding. These included not only the basic Lys and Arg residues but also residues that are not charged, such as Y219, Q223, P226, and G229. In contrast, only four residues, namely Q209, R214, R218, and G219, appeared to be crucial in binding of IGFBP-5 NLS to importin-β. Except for Q209, all of these residues reside in the major basic cluster (Fig. 7Go, lower panel). The roles of corresponding amino acids in the two NLS peptides also differed. Whereas substitution of either one of residues Y219, K220, C224, R224, P226, G229, R230, or K231 by an Ala led to 50% or greater reduction in affinity, equivalent substitutions in IGFBP-5 NLS had either a negligible or an opposite effect (Fig. 7Go). Moreover, replacement of either K228 or G233 decreased the affinity of the corresponding IGFBP-3 NLS peptides only slightly, whereas the equivalent peptides R214A and G219A of the IGFBP-5 NLS were essentially inactive in binding importin-β. Therefore, despite the high sequence homology between the NLSs of IGFBP-3 and IGFBP-5, the structural basis for their respective recognition by importin-β may differ.

Possible cellular functions of nuclear IGFBP-6
RD cells overexpressing WT IGFBP-6 were synchronized using nocodazole, a G2/M-arresting reagent. After nocodazole release, cells were fixed and examined by confocal microscopy for IGFBP-6 localization. In Fig. 8AGo, cells transfected with vector shows normal mitotic segregation of the cells, whereas in cells transfected with WT IGFBP-6 construct, asymmetric cell division was noted (Fig. 8AGo). Cells that appear to divide asymmetrically also stained positive for nuclear IGFBP-6. Based on their morphology, both mitotic and premitotic cells expressed IGFBP-6 abundantly. To confirm this observation, cells were fractionated and assessed for IGFBP-6 before and after nocodazole release. Because the mammalian cell cycle takes approximately 24 h to complete [G1 (1 h), S (8 h), G2 (4 h), and M (1 h)], we fractionated the cells before and after nocodazole release, at 10, 20, and 40 min (during mitosis). This procedure should allow the collection of both nuclear and cytoplasmic proteins. We investigated the activation of pCK2b protein as a marker for mitosis, together with IGFBP-6 expression and the activation of p-Akt, as a marker of defense against apoptosis (Fig. 8BGo). We demonstrated that IGFBP-6 was slightly increased during mitosis and the activation of Akt was minimal in the RD-IGFBP-6 cells, whereas the vector control cells still had basal Akt activation. These findings suggest that although IGFBP-6 may not be a constitutive protein of the nucleus, it may still associate or regulate different mitotic events that may finally lead to apoptosis.


Figure 8
View larger version (38K):
[in this window]
[in a new window]

  		FIG. 8. G2/M synchronized RD cells show that nuclear IGFBP-6 participates in mitosis. A, Careful analysis of nocodazole-arrested cells (at G2/M check point) indicates that IGFBP-6 is present in the nucleus at this time point. Cells were also analyzed after nocodazole release at 5, 15, and 30 min when they underwent mitosis. RD-IGFBP-6 cells show an increase in IGFBP-6 synthesis during mitosis and also randomly nuclear localization of this protein, especially in cells that seem to segregate dysfunctional or asymmetrically (fifth image from the RD-IGFBP-6 panel). This effect was not observed in vector control cells. Images show confocal photomicrographs of cells fixed and immunostained with an antibody against IGFBP-6, followed by a secondary antibody labeled with TRITC. Chromatin was counterstained with Hoechst reagent. Images were cropped magnified using Photoshop software. The original images were captured by Zeiss-410 confocal microscope at a magnitude of x20. B, Synchronized RD-IGFBP-6 cells and controls were fractionated at 10, 20, and 40 min, after nocodazole release, to assess by Western blot the presence or the activation of several cellular markers such as pCK2b(Ser209) for mitosis and pAkt as a defense marker for apoptosis. Whereas vector control cells show basal Akt activation during mitosis, RD-IGFBP-6 cells did not activate this marker.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
The structure of IGFBP-6-NLS and its functional implications
After the identification of functional NLSs in IGFFBP-3 and -5 (27), our studies uncovered yet another functional NLS in IGFBP-6. The NLSs of IGFBP-3 and -5 resemble that found in IL-5 (43) and have the characteristics of a bipartite NLS, i.e. they each contain a major and a minor basic cluster separated by an 11-residue spacer, as follows: IGFBP-3, RR(11aa)KK-GFYKKKQCRPS-KGRKRG; and IGFBP-5, RR(11aa)RK-GFYKRKQCRPS-RGRKRG.

Schedlich et al. (27) showed that the major cluster is indispensable for nuclear accumulation of either IGFBP-3 or IGFBP-5 and that the minor basic cluster and the KKK/KRK basic motif in the spacer also play an important part in the nuclear uptake of the two proteins (27). These characteristics are largely conserved in the putative NLS of IGFBP-6 [RR(11aa)HRGFYRKRQCRSS-QGQRRG] but not the corresponding regions of IGFBP-1, -2, or -4. A significant difference is seen in the amino acid composition of the IGFBP-6 NLS clusters, compared with the residues present in either the IGFBP-3 or -5 NLS. However, in reviewing the composition in Arg or Lys in the NLS, we observed that IGFBP-3 and -5 have 12 basic residues, whereas IGFBP-6 has nine residues, which represents 75% of the total number of Arg and Lys residues, characteristic for the other two nuclear IGFBPs. Nevertheless, a database search using the pseudomajor basic cluster sequence of IGFBP-6 NLS (i.e. the GQRRG motif) retrieved a group of diverse proteins, including the serine/arginine-rich (RNA binding) proteins (44), the NKx-homeodomain proteins (45), and the Ser/Thr protein kinase LKB1/STK11 (46) that may use the same motif for nuclear translocation. Moreover, the structure of IGFBP-6 NLS is very similar to the NLS of another well-established nuclear protein Ku70 (537EGKVTKRKHDNEGSGSKRPKV557), involved in DNA repair processes and apoptosis (47). Therefore, a reduction in positive charge based on the presence of 12 basic residues (as seen in IGFBP-3 or -5) to nine residues (IGFBP-6) up to seven residues in Ku70 does not necessarily reduce the function of nuclear translocation of a protein.

The minor cluster in the IGFBP-6 NLS is also incomplete because its basic charge at physiological pH is partly reduced with the substitution of a His for an Arg or a Lys residue. Despite these differences, the IGFBP-6 NLS was found to be functional in RD cells. Nuclear uptake and accumulation of IGFBP-6 was observed to rely on intact NLS, especially on the few Arg residues present in the spacer and the pseudomajor basic cluster, because their mutation, either singularly or in combination, rendered the NLS largely nonfunctional. The positive charge in the pseudomajor basic cluster, e.g. R209, appeared to play a most important role in nuclear import because its mutation led to a complete loss of binding to importin-{alpha}, and consequently, the exclusion of IGFBP-6 from the nucleus. Thus, the mutation of this single basic residue produced the same phenotype as that resulting from the deletion of the complete NLS. In comparison, mutations of other basic amino acids, such as the 165Arg166Arg pair, had only a moderate effect. However, because this minor cluster of basic residues is conserved in IGFBP-3, -5, and -6 but not in other members of the IGFBP family that lack a functional NLS, it is likely these two Arg residues participate in mediating the nuclear import of IGFBP-6.

Nuclear import of IGFBP-6 is an active process that involves energy (ATP) and importin-{alpha}
We have shown that both the intact protein and an NLS peptide derived from IGFBP-6 were capable of binding importin-{alpha}. Moreover, blocking importin-{alpha} using a specific antibody resulted in a significant reduction in nuclear translocation and accumulation of IGFBP-6, in accordance with a central role for by importin-{alpha}. The dependence of IGFBP-6 on importin-{alpha} is unique among IGFBPs that are capable of nuclear translocation. Previous studies have demonstrated that nuclear import of IGFBP-3 and -5 was more dependent on importin-β than importin-{alpha} (27). Our data indicate that importin-{alpha} is the preferred binding partner for IGFBP-6. Despite these discrepancies, it should be noted that the NLSs of all three proteins are capable of binding to both types of importins, suggesting that the importin-{alpha}/β dimer, rather than a particular importin alone, may be the functional unit that recognizes these NLS motifs in vivo. The difference in relative binding affinity to a given importin may provide a mechanism by which the nuclear translocation of the IGFBP proteins can be differentially regulated.

Peptide array analysis showed the key residues mediating the binding of an NLS to importin-{alpha} or -β
An interesting finding from these studies is that, apart from the predictable involvement of basic residues in binding (29, 30, 31), neutral residues such as Gly and Pro were also seen to be important in the binding of an NLS to importins. Pro and/or Gly, which frequently flank basic residues in an NLS (48), may play unique structural roles in importin binding because they are known breakers of regular secondary structures (48). The participation of specific, noncharged residues in importin binding may have general relevance because a number of NLS sequences are enriched in glycine rather than basic residues (49, 50).

The IGFBP-6 NLS differs from a classic NLS such as that in IL-5. Whereas nuclear import of IL-5 was strictly dependent on soluble cytosolic factors in our nuclear transport assay (43), IGFBP-6 was capable of entering the nuclei of RD cells without the addition of these factors. This may be due to the presence of residual amounts of importin-{alpha} in digitonin-treated cells. In fact, anti-importin-{alpha} staining could be detected on the nuclear envelope (data not shown). Because importin-{alpha} can migrate into the nucleus in an importin-β-independent fashion and in the absence of cytosol (51), it is likely that IGFBP-6 enters the nucleus via direct binding to the residual importin-{alpha} present in the permeabilized cells that is associated with the nuclear membrane.

Nuclear localization of IGFBP-6 is related to apoptosis, and it acts in an IGF-II-independent manner
Although this report does not provide a complete answer to the biological function of nuclear IGFBP-6, it demonstrates that it is related to apoptosis and it does not involve its preferred ligand IGF-II (Fig. 4Go, B and C). Even if IGFBP-6 binds with high affinity to IGF-II, its nuclear localization appears be uncoupled from this physiological interaction. IGFBP-6 was detected in both the cytosol and nucleus, whereas IGF-II was exclusively found in the cytosol. Further investigation is needed to identify IGFBP-6 partners for nuclear activity. Our cellular system allowed a quantification of IGFBP-6-induced apoptosis events because in serum-free conditions the presence of this protein increased the rate of apoptosis, up to 2-fold. Whether nuclear IGFBP-6 simply acts in changing the cell fate toward apoptosis or it also induces differentiation in muscle cells remained to be answered.

Truncation of the NLS segment from the IGFBP-6 structure decreased the apoptosis process, typically induced by WT IGFBP-6 overexpression. Although this effect was due to the truncation of the protein at the NLS motif, it can be either NLS dependent, by inhibiting the interaction of IGFBP-6 with partner proteins, which may bind straightly to the NLS region (linear interaction), or NLS independent, by changing the conformation of the entire IGFBP-6 molecule, which may actually impair interactions with either cytoplasmic or nuclear proteins, unrelated to NLS. Thus, it would be inappropriate to consider the NLS sequence, by itself, being responsible for the IGFBP-6-induced apoptosis. Both site-directed mutagenesis of the NLS importin binding residues and the truncation of the entire NLS fragment could alter the structure of IGFBP-6. Based on the current data, we suggest that modification of the IGFBP-6 carboxyl terminus as a whole impairs the apoptosis process and not specifically the linear sequence of the NLS. As shown in Fig. 8AGo, in RD cells, IGFBP-6 was demonstrated to concentrate in the nuclei of mitotic cells on synchronization (after nocodazole release), when morphologic aspects of asymmetric cell division were observed. Both IGFBP-3 and IGFBP-5 have been suggested to be potential substrates of Ser/Thr kinases such as CK2 (52). Interestingly, sequence analysis revealed that IGFBP-6 also possesses candidate phosphorylation sites for CK2, as well as DNA-PK, a DNA repair protein involved in apoptosis. The putative functions of nuclear IGFBP-6 may be therefore related to DNA stability and repair during mitosis. Further studies are needed to confirm whether IGFBP-6 directly interacts with DNA repair proteins as a potential mechanism of action of nuclear IGFBP-6.


   Acknowledgments
 
We thank Dr. Kem Rogers (Department of Anatomy and Cell Biology, University of Western Ontario) for advice and assistance in laser-scanning confocal analyses.


   Footnotes
 
This work was supported by grants from the Canadian Institutes of Health Research and the Canada Research Chairs Program (to V.K.M.H. and S.S.-C.L.). S.S.-C.L. is a scientist at the National Cancer Institute of Canada with funds made available from the Canadian Cancer Society. V.K.M.H. holds a Canada Research Chair in Fetal and Maternal Health, and S.S.-C.L. holds a Canada Research Chair in Functional Genomics and Cellular Proteomics.

Disclosure Statement: The authors of this manuscript have nothing to declare.

First Published Online November 26, 2007

Abbreviations: CK2, Casein kinase-II; CMV, cytomegalovirus; EGFP, enhanced GFP; FITC, fluorescein isothiocyanate; Fn/c, nuclear to cytoplasmic fluorescence; GFP, green fluorescent protein; HRP, horseradish peroxidase; IGFBP, IGF binding protein; NES, nuclear export signal; NLS, nuclear localization signal; PARP, poly-ADP-ribose polymerase; RD, rhabdomyosarcoma; TBS, Tris-buffered saline; TRITC, tetramethylrhodamineithio-cyanate; TTBS, Tween 20 in TBS; WT, wild type.

Received July 13, 2007.

Accepted for publication November 12, 2007.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Shimasaki S, Ling N 1991 Identification and molecular characterization of insulin-like growth factor binding proteins (IGFBP-1, -2, -3, -4, -5 and -6). Prog Growth Factor Res 3:243–266[CrossRef][Medline]
  2. Jones JI, Clemmons DR 1995 Insulin-like growth factors and their binding proteins. Endocr Rev 16:3–34[Abstract/Free Full Text]
  3. Han VK, Carter AM 2000 Spatial and temporal patterns of expression of messenger RNA for insulin-like growth factors and their binding proteins in the placenta of man and laboratory animals. Placenta 21:289–305[CrossRef][Medline]
  4. Janosi JB, Ramsland PA, Mott MR, Firth SM, Baxter RC, Delhanty PJ 1999 The acid-labile subunit of the serum insulin-like growth factor-binding protein complexes. Structural determination by molecular modeling and electron microscopy. J Biol Chem 274:23328–23332[Abstract/Free Full Text]
  5. Kelley KM, Oh Y, Garcosky SE, Gucev Z, Matsumoto T, Hwa V, Ng L, Simpson DM, Rosenfeld RG 1996 Insulin-like growth factor-binding proteins (IGFBPs) and their regulatory dynamics. Int J Biochem Cell Biol 28:619–637.[CrossRef][Medline]
  6. Clemmons DR 1999 Insulin-like growth factor binding proteins and their role in controlling IGF actions. Cytokine Growth Factor Rev 8:45–62[Medline]
  7. Martin JL, Coverly JL, Baxter RC 1994 Regulation of Immunoreactive insulin like growth factor binding protein-6 in normal and transformed fibroblast. J Biol Chem 269:11470–11477[Abstract/Free Full Text]
  8. Yan T, Wergedal J, Zhou Y, Mohan S, Baylink DJ, Strong DD 2001 Inhibition of human osteoblast marker gene expression by retinoids is mediated in part by insulin-like growth factor binding protein-6. Growth Horm IGF Res 11:368–377[CrossRef][Medline]
  9. Kim EJ, Kang YH, Schaffer BS, Bach LA, MacDonald RG, Park JH 2002 Inhibition of Caco-2 cell proliferation by all-trans retinoic acid: role of insulin-like growth factor binding protein-6. J Cell Physiol 190:92–100[CrossRef][Medline]
  10. Sueoka N, Lee HY, Walsh GL, Fang B, Ji L, Roth JA, LaPushin R, Hong WK, Cohen P, Kurie JM 2002. Am J Respir Cell Mol Biol 23:297–303
  11. Grellier P, Berrebi D, Peuchmaur M, Babajko S 2002 The IGF system in neuroblastoma xenografts: focus on IGF-binding protein 6. J Endocrinol 172:467–476[Abstract]
  12. Sueoka N, Lee HY, Wiehle S, Cristiano RJ, Fang B, Ji L, Roth JA, Hong WK, Cohen P, Kurie JM 2000 Insulin-like growth factor binding protein-6 activates programmed cell death in non-small cell lung cancer cells. Oncogene 19:4432–4436[CrossRef][Medline]
  13. Babajko S, Leneuve P, Loret C, Binoux M 1997 IGF-binding protein-6 is involved in growth inhibition in SH-SY5Y human neuroblastoma cells: its production is both IGF- and cell density-dependent. J Endocrinol 152:221–227[Abstract/Free Full Text]
  14. Hale K, Murray AW, Cosgrove LJ, Bach LA, Hartfield PJ 2000 Prevention of apoptosis by insulin-like growth factor (IGF)-I and IGF-II is differentially attenuated by IGF-binding proteins in PC12 cells. Neurosci Res Commun 27:75–83[CrossRef]
  15. Grellier P, de Galle B, Babajko S 1998 Expression of insulin-like growth factor (IGF) binding protein-6 complementary DNA alters neuroblastoma cell growth. Cancer Res 58:1670–1676[Abstract/Free Full Text]
  16. Sueoka N, Lee HY, Walsh GL, Fang B, Ji L, Roth JA, LaPushin R, Hong WK, Cohen P, Rosenfeld RG 1996 Insulin-like growth factor-binding proteins (IGFBPs) and their regulatory dynamics. Int J Biochem Cell Biol 28:619–637[CrossRef][Medline]
  17. Butt AJ, Firth SM, Baxter RC 1999 The IGF axis and programmed cell death. Cell Biol 77:256–262
  18. Butt AJ, Firth SM, King MA, Baxter RC 2000 Insulin-like growth factor-binding protein-3 modulates expression of Bax and Bcl-2 and potentiates p53-independent radiation-induced apoptosis in human breast cancer cells. J Biol Chem 275:39174–39181[Abstract/Free Full Text]
  19. Spagnoli A, Torello M, Nagalla SR, Horton WA, Pattee P, Hwa V, Chiarelli F, Roberts Jr CT, Rosenfeld RG 2002 Identification of STAT-1 as a molecular target of IGFBP-3 in the process of chondrogenesis. J Biol Chem 277:18860–18867[Abstract/Free Full Text]
  20. Taylor VL, Spencer EM 2001 Characterisation of insulin-like growth factor-binding protein-3 binding to a novel receptor on human platelet membranes. J Endocrinol 168:307–315[Abstract]
  21. Liu B, Lee HY, Weinzimer SA, Powell DR, Clifford JL, Kurie JM, Cohen P 2000 Direct functional interactions between insulin-like growth factor-binding protein-3 and retinoid X receptor-{alpha} regulate transcriptional signaling and apoptosis. J Biol Chem 275:33607–33613[Abstract/Free Full Text]
  22. Baxter RC 2001 Inhibition of the insulin-like growth factor (IGF)-IGF-binding protein interaction. Horm Res 2:68–72
  23. Rajah R, Valentinis B, Cohen P 1997 Insulin-like growth factor (IGF)-binding protein-3 induces apoptosis and mediates the effects of transforming growth factor-β1 on programmed cell death through a p53- and IGF-independent mechanism. J Biol Chem 272:12181–12188[Abstract/Free Full Text]
  24. Oh Y, Muller HL, Lamson G, Rosenfield RG 1993 Insulin-like growth factor (IGF)-independent action of IGF-binding protein-3 in Hs578T human breast cancer cells. J Biol Chem 268:14964–14971[Abstract/Free Full Text]
  25. Schmid C, Keller C, Gosteli-Peter M, Zapf J 1999 Mitogenic and antiapoptotic effects of insulin-like growth factor binding protein-6 in the human osteoblastic osteosarcoma cell line Saos-2/B-10. Biochem Biophys Res Commun 263:786–789[CrossRef][Medline]
  26. Gallicchio MA, Kneen M, Hall C, Scott AM, Bach LA 2001 Overexpression of insulin-like growth factor binding protein-6 inhibits rhabdomyosarcoma growth in vivo. Int J Cancer 94:645–651[CrossRef][Medline]
  27. Schedlich L, Le Page S, Firth SM, Briggs LJ, Jans DA, Baxter RC 2000 Nuclear import of Insulin-like growth factor binding protein-3 and -5 is mediated by importin-β subunit. J Biol Chem 275:23462–23470[Abstract/Free Full Text]
  28. Gorlich D, Vogel F, Mills AD, Hartmann E, Laskey RA 1995 Distinct functions for the two importin subunits in nuclear protein import. Nature 377:246–249[CrossRef][Medline]
  29. Gorlich D, Mattaj IW 1996 Nucleocytoplasmic transport. Science 271:1513–1518[Abstract]
  30. Mattaj IW, Englmeier L 1998 Nucleocytoplasmic transport: the soluble phase. Ann Rev Biochem 67:265–306[CrossRef][Medline]
  31. Dingwall C, Laskey RA 1991 Nuclear targeting sequences—a consensus? Trends Biochem Sci 16:478–481[CrossRef][Medline]
  32. Kalderon D, Roberts BL, Richardson WD, Smith AE 1984 A short amino acid sequence able to specify nuclear location. Cell 39:499–509[CrossRef][Medline]
  33. Robbins J, Dilworth SM, Laskey RA, Dingwall C 1991 Two interdependent basic domains in nucleoplasmin nuclear targeting sequence: identification of a class of bipartite nuclear targeting sequence. Cell 64:615–623[CrossRef][Medline]
  34. Kohler M., Speck C, Christianensen M, Bischoff R, Prehn S, Haller H, Gorlich D, Hartmann E 2000 Evidence for distinct substrate specificities of importin-{alpha} family members in nuclear protein import. Mol Cell Biol 19:7782–7791
  35. Cingolani G, Bednenko J, Gillespie MT, Gerace L 2002 Molecular basis for the recognition of a nonclassical nuclear localization signal by importin β. Mol Cell 10:1345–1353[CrossRef][Medline]
  36. Chook YM, Blobel G 2001 Karyopherins and nuclear import. Curr Opin Struct Biol 11:703–715[CrossRef][Medline]
  37. Catimel B, Teh T, Fontes MR, Jennings IG, Jans DA, Howlett GJ, Nice EC, Kobe B 2001 Biophysical characterization of interactions involving importin-{alpha} during nuclear import. J Biol Chem 276:34189–34198[Abstract/Free Full Text]
  38. Romanelli MG, Morandi C 2002 Importin-{alpha} binds to unusual nuclear localization signal in the heterogeneous ribonucleoprotein type I. Eur J Biochem 269:2727–2734[Medline]
  39. Tkizawa G C, Weiss K, Morgan D O 1999 Ran-independent nuclear import of cycline B1-cdc2 by importin-β. Proc Natl Acad Sci USA 96:7938–7943[Abstract/Free Full Text]
  40. Adam SA, Marr RS, Gerace L 1990 Nuclear protein import in permeabilized mammalian cells requires soluble cytoplasmic factors. J Cell Biol 111:807–816[Abstract/Free Full Text]
  41. Bradshaw SL, D’Ercole SJ, Han VKM 1999 Overexpression of insulin-like growth factor binding protein-2 in C6 glioma cells result in conditional alteration of cellular growth. Endocrinology 140:575–584[Abstract/Free Full Text]
  42. Li C, Iosef C, Jia, CY, Han VK, Li SS 2003 Dual functional roles for the X-linked lymphoproliferative syndrome gene product SAP/SH2D1A in signaling through the signaling lymphocyte activation molecule (SLAM) family of immune receptors. J Biol Chem 278:3852–3859[Abstract/Free Full Text]
  43. Jans DA, Briggs LJ, Gustin SE, Jans P, Ford S, Young IG 1997 The cytokine interleukin-5 (IL-5) effects cotransport of its receptor subunits to the nucleus in vitro. FEBS Lett 406:315–320[CrossRef][Medline]
  44. Kawano T, Fujita M, Sakamoto H 2000 Unique and redundant functions of SR proteins, a conserved family of splicing factors, in Caenorhabditis elegans development. Mech Dev 95:67–76[CrossRef][Medline]
  45. Buchberger A, Pabst O, Brand T, Seidl K, Arnold HH 1996 Chick NKx-2.3 represents a novel family member of vertebrate homologues to the Drosophila homeobox gene tinman: differential expression of cNKx-2.3 and cNKx-2.5 during heart and gut development. Mech Dev 56:151–163[CrossRef][Medline]
  46. Boudeau J, Kieloch A, Alessi DR, Stella A, Guanti G, Resta N 2003 Functional analysis of LKB1/STK11 mutants and two aberrant isoforms found in Peutz-Jeghers syndrome patients. Hum Mutat 21:172–178[CrossRef][Medline]
  47. Bertinato J Schild-Poulter C, Hache RJK 2000 Nuclear localization of Ku antigen is promoted independently by basic motifs in the Ku70 an Ku80 subunits. J of Cell Sci 114:89–99
  48. Boulikas T 1994 Transcription factor binding sites in the matrix attachment region (MAR) of the chicken {alpha}-globin gene. J Cell Biochem 55:32–58[CrossRef][Medline]
  49. Bogerd HP, Benson RE, Truant R, Herold A, Phingbodhipakkiya M, Cullen BR 1999 Definition of a consensus transportin-specific nucleocytoplasmic transport signal. J Biol Chem 274:9771–9777[Abstract/Free Full Text]
  50. Siomi H, Dreyfuss G 1995 A nuclear localization domain in the hnRNP A1 protein. J Cell Biol 129:551–560[Abstract/Free Full Text]
  51. Miyamo Y, Hieda M, Harreman MT, Fukumoto F, Saiwaki T, Hodel AE, Corbett AH, Yoneda Y 2002 Importin-{alpha} can migrate into the nucleus of an importin-β and Ran-independent manner. EMBO J 21:5833–5842[CrossRef][Medline]
  52. Coverley JA, Martin JL, Baxter RC 2000 The effect of phosphorylation by casein kinase 2 on the activity of insulin-like growth factor-binding protein-3. Endocrinology 141:564–570[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Am. J. Physiol. Regul. Integr. Comp. Physiol.Home page
X. Wang, L. Lu, Y. Li, M. Li, C. Chen, Q. Feng, C. Zhang, and C. Duan
Molecular and functional characterization of two distinct IGF binding protein-6 genes in zebrafish
Am J Physiol Regulatory Integrative Comp Physiol, May 1, 2009; 296(5): R1348 - R1357.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Iosef, C.
Right arrow Articles by Han, V. K. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Iosef, C.
Right arrow Articles by Han, V. K. M.
Right arrowPubmed/NCBI databases
*Gene*GEO Profiles
*HomoloGene*UniGene
*Compound via MeSH
*Substance via MeSH

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