HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 22, 19673-19678, May 31, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/22/19673    most recent M112112200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Braem, C. V. Articles by Voz, M. L. Search for Related Content PubMed PubMed Citation Articles by Braem, C. V. Articles by Voz, M. L. Social Bookmarking                      What's this?

Identification of a Karyopherin 2 Recognition Site in PLAG1, Which Functions As a Nuclear Localization Signal^*

Caroline V. Braem, Koen Kas^§, Eva Meyen, Maria Debiec-Rychter, Wim J. M. Van de Ven^¶, and Marianne L. Voz

From the Laboratory for Molecular Oncology, Department of Human Genetics, University of Leuven and Flanders Interuniversity Institute for Biotechnology, Herestraat 49, B-3000 Leuven, Belgium

Received for publication, December 19, 2001, and in revised form, February 22, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The activation of the pleomorphic adenoma gene 1 (PLAG1) is the most frequent gain-of-function mutation found in pleomorphic adenomas of the salivary glands. To gain more insight into the regulation of PLAG1 function, we searched for PLAG1-interacting proteins. Using the yeast two-hybrid system, we identified karyopherin 2 as a PLAG1-interacting protein. Physical interaction between PLAG1 and karyopherin 2 was confirmed by an in vitro glutathione S-transferase pull-down assay. Karyopherin 2 escorts proteins into the nucleus via interaction with a nuclear localization sequence (NLS) composed of short stretches of basic amino acids. Two putative NLSs were identified in PLAG1. The predicted NLS1 (KRKR) was essential for physical interaction with karyopherin 2 in glutathione S-transferase pull-down assay, and its mutation resulted in decreased nuclear import of PLAG1. Moreover, NLS1 was able to drive the nuclear import of the cytoplasmic protein -galactosidase. In contrast, predicted NLS2 of PLAG1 (KPRK) was not involved in karyopherin 2 binding nor in its nuclear import. The residual nuclear import of PLAG1 after mutation of the NLS1 was assigned to the zinc finger domain of PLAG1. These observations indicate that the nuclear import of PLAG1 is governed by its zinc finger domain and by NLS1, a karyopherin 2 recognition site.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The activation of the pleomorphic adenoma gene 1 (PLAG1)¹ is the most frequent gain-of-function mutation found in pleomorphic adenomas of the salivary glands. Indeed, >50% of these benign tumors carry aberrations in chromosome 8q12 in the region were PLAG1 is localized. These aberrations result in the replacement of the PLAG1 promoter, which is inactive in the normal adult salivary gland, with a strong promoter (1-3). As such, PLAG1 was identified as a candidate oncogene in the tumorigenesis of pleomorphic adenomas of the salivary glands.

The PLAG1 protein consists of a C-terminal trans-activation domain preceded by seven C₂H₂ zinc fingers, which are responsible for DNA binding (4). The consensus PLAG1 binding site comprises a core sequence (GRGGC) and a G-cluster (RGGK), separated by seven random nucleotides (5). DNA binding is mediated mainly via three of the seven zinc fingers with fingers 6 and 7 interacting with the core and with finger 3 interacting with the G-cluster. In transient trans-activation assays, PLAG1 specifically activates transcription from its consensus DNA binding site, indicating that PLAG1 is a genuine transcription factor. Potential PLAG1 binding sites were found in the promoter of many genes and notably in the promoter 3 of the human insulin-like growth factor II (IGF-II) gene for which we have proved that it is a bona fide PLAG1 target gene (5).²

PLAG1 is a member of the highly conserved PLAG subfamily of zinc finger proteins comprising two other members, PLAGL1 (also called LOT1 or ZAC1) and PLAGL2 (4, 6-8). Structurally and functionally, PLAG1 and PLAGL2 are very similar. PLAGL2 is able to bind to the consensus binding site of PLAG1, and IGF-II is a target of both PLAG1 and PLAGL2.² PLAG1 and PLAGL2 both are able to transform NIH-3T3 cells in vitro, indicating they are genuine proto-oncogenes.² In contrast, PLAGL1 inhibits tumor cell growth through the induction of apoptotic cell death and G₁ arrest and recognizes a different consensus binding site (9).

To gain more insight into the regulation of PLAG1 function, a mouse embryonic cDNA library was screened for PLAG1-interacting proteins in the yeast two-hybrid assay system. We used an embryonic library, because PLAG1 is expressed mainly during embryonic development. We opted to screen with the N-terminal tail of PLAG1 for two reasons. First, this region has a high surface probability as predicted by the Protean program (DNAstar), indicating its availability for interaction with other proteins. Second, in contrast to the DNA-binding zinc finger domain and the trans-activating C-terminal domain, no function had been assigned to the N-terminal part of PLAG1. Proteins interacting with this region might reveal new aspects of PLAG1 function.

The screen revealed interaction between PLAG1 and karyopherin 2, a protein involved in active nuclear import. This interaction provided the basis for the identification of a karyopherin 2 recognition site in PLAG1, which functions as a nuclear localization signal.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Plasmid Constructs-- GST-PLAG1-(2-244) and GST-PLAG1-(84-244) were described previously (5, 10). GST-PLAGL1-(2-215) and GST-PLAGL2-(2-250) deletion mutants were constructed by PCR cloning and subsequent ligation into the EcoRI/XhoI sites (as underlined in the primers) of the pGEX-5X-2 vector and verified by sequencing. To clone PLAGL1-(2-215), the following PCR primers were used: P2N2, 5'-CCCGAATTCTGGCCACGTTCCCCTGCC-3', and P2C215, 5'-GGGCTCGAGCTAGCTCTCTTTCATCAGCTCC-3'. To clone PLAGL2-(2-250), the following PCR primers were used: P3N2, 5'-CCCGAATTCTGACCACATTTTTCACCAGCG-3', and P3C250, 5'-GGGCTCGAGCTACTTGATCTTGAGCAGCTCCT-3'. All sequences are 5' to 3', and the direction forward (N) or reverse (C) is indicated. To obtain the PLAG1 bait construct used in the yeast two-hybrid assay, we cloned the PLAG1 fragment of the GST-PLAG1-(2-244) construct into the BamHI/PstI sites of pGBT9 (CLONTECH). This plasmid was again digested with StyI (cutting PLAG1 at 209 bp in the open reading frame) and PstI, blunt-ended, and religated, generating pGBT9-PLAG1-(2-70). -Galactosidase-PLAG1-GFP fusion constructs for intracellular fluorescence studies were prepared by cloning PCR fragments amplified from the pCDNA3-PLAG1 expression construct (5) into the SacII/XbaI sites of pHM829 (11). The following primers were used: SacII-Plag1N2, 5'-AGTCCCGCGGGCCACTGTCATTCCTGGTG-3'; SacII-Plag1N41, 5'-AGTCCCGCGGAAGGCCTTTAACAGTGTTGAG-3'; SacII-Plag1N242, 5'-AGTCCCGCGGAAGGTCAAAACAGAACCAGTG-3'; XbaI-Plag1C500, 5'-ACTGTCTAGACTGAAAAGCTTGATGGAAAC-3'; XbaI-Plag1C41, 5'-ACAGTCTAGACTTGTCACACAGTTGGCAAG-3'; and XbaI-Plag1C241, 5'-ACTGTCTAGACAGAAGCTCTTGATTGTGAC-3'. To clone the putative NLS1 (SGKRKRGE) into the SacII/XbaI site of pHM829 vector, the following oligonucleotides were used: oligoNLSup, 5'-GGTCAGGGAAACGTAAGCGTGGTGAAT-3', and oligoNLSlow, 5'-CTAGATTCACCACGCTTACGTTTCCCTGACCGC-3'. Mutations of the NLS motifs within GST-PLAG1-(N2-C244), -galactosidase-PLAG1-(2-500)-GFP, and -galactosidase-PLAG1-(2-42)-GFP constructs were performed as described in the Stratagene protocol for PCR-based site-directed mutagenesis. To mutate the first candidate NLS motif ²²KRKR²⁵ into ²²KAAR²⁵, the following primers were used: mutaNLS1-up, 5'-CCTTCAGGGAAAGCTGCGCGTGGTGAAACC-3', and mutaNLS1-low, 5'-GGTTTCACCACGCGCAGCTTTCCCTGAAGG-3'. To mutate the second candidate NLS motif ²⁹KPKR³² into ²⁹KPAA³², we used the following primers: mutaNLS2-up, 5'-GGTGAAACCAAACCAGCAGCTAACTTTCCTTGCCAAC-3', and mutaNLS2-low, 5'-GTTGGCAAGGAAAGTTAGCTGCTTGGTTTGGTTTCACC-3'. The frame and mutations in all of the newly synthesized constructs were verified by sequencing.

Yeast Two-hybrid Assay-- The Matchmaker Two-hybrid System 2 was purchased from CLONTECH (Palo Alto, CA). All experiments were performed in the yeast reporter strain CG-1945 (Trp and Leu). The "bait" construct consisted of the N-terminal part of human PLAG1- (N2-C70) cloned into the yeast vector pGBT9 (CLONTECH). This vector allows the fusion of the protein of interest to the C-terminal end of the GAL4 DNA-binding domain and contains TRP1 reporter gene for selection of transformants. The PLAG1 bait construct did not show autonomous transcriptional activation and hence was a good candidate for the detection of protein interactions in the yeast two-hybrid transcriptional activation assay. An oligo(dT) and randomly primed "prey" cDNA library from 12.5-day-old embryonic mice cloned into the pACT2 vector was kindly provided by Drs. K. Verschueren and D. Huylebroeck (University of Leuven and Flanders Interuniversity Institute for Biotechnology, Belgium). The pACT2 vector allows the fusion of proteins to the C-terminal end of the major GAL4 activation domain and contains LEU2 for selection of transformants. 1 × 10⁹ CG-1945 yeast were transformed with 66 µg of bait-DNA and 33 µg of prey-library-DNA using a LiAc high efficiency transformation protocol (12). This yeast strain contains the HIS3 and lacZ reporter genes under the control of promoters containing GAL4 binding sites. Transformants were grown for 10 days at 30 °C on triple selective (lacking Trp, Leu, and His) synthetic dropout (SD) agar plates containing 5 mM 3-aminotriazol (Sigma). Double transformed His⁺ yeast colonies were restreaked on new SD agar plates and grown for another 24-48 h. For the qualitative measurement of -galactosidase activity, colony lift filter assays were performed according to standard protocols. Plasmid DNA was isolated from positive (blue) colonies by glass bead lysis, extraction with phenol/chloroform, and ethanol precipitation and subsequently used to transform the Escherichia coli strain HB101 (Leu) by electroporation. pACT2 plasmids containing different inserts as analyzed by PCR amplification and BglII digestion were reassayed by cotransformation into yeast-competent cells with either the PLAG1-pGBT9 construct, the empty pGBT9 vector, or pGBT9 containing an unrelated cDNA insert (human lamin C) shown to interact in a two-hybrid assay with an independent protein. Plasmids that generated colonies on SD agar plates and were only positive in the X-gal filter assay with the PLAG1 bait construct were considered for further analysis.

GST Pull-down Assays-- Full-length murine karyopherin 2 cDNA in the pET15b vector (generously provided by Dr. M. Waterman, Department of Microbiology and Molecular Genetics, College of Medicine, University of California, Irvine, CA) was used to prepare in vitro synthesized [³⁵S]methionine-labeled karyopherin 2 protein. The in vitro translation reaction was carried out using the TNT T7 Quick-coupled transcription/translation system (Promega) following the manufacturer's instructions. Bacterial expression constructs were made using the pGEX-5X-2 vector, directing the synthesis of GST fused to different parts of PLAG1. The fusion proteins were purified according to supplier's instructions and verified by SDS-PAGE. The different PLAG1 fusion proteins or GST alone as negative control were bound to glutathione-agarose beads. Equal amounts of these GST fusion proteins (~20 µg) were incubated with 10 µl of in vitro synthesized [³⁵S]methionine-labeled full-length karyopherin 2 protein in 500 µl of NENT₅₀₀ buffer (500 mM NaCl, 20 mM Tris-HCl, pH 7.6, 1 mM EDTA, 0.5% Nonidet P-40). This mixture was tumbled for 1 h at 4 °C. Subsequently, the beads were washed four times in 500 µl of NENT₅₀₀ buffer, resuspended in 30 µl of SDS-PAGE sample buffer, and incubated at 95 °C for 2 min. Proteins were size-separated on SDS-PAGE, and interacting karyopherin 2 was detected by autoradiography.

Immunofluorescence on Cultured Tumor Cells-- A pleomorphic adenoma with a translocation involving chromosome 8q12 (t(3;8)(p21;q12)), the reciprocal translocations with breakpoints at chromosome 3p21 and 8q12 was retrieved from the tumor bank of the Center for Human Genetics (University of Leuven, Leuven, Belgium). It was a primary tumor originating from the salivary gland of a previously untreated patient. Primary in situ cultures were obtained from the original single cell suspension of tumor cells and cultured in Dulbecco's modified Eagle's medium/F12 (1:1) (Invitrogen) supplemented with 10% fetal calf serum at 37 °C in a humidified 5% CO₂ atmosphere. These cells were subsequently grown to 40-60% confluency on chamber slides (Lab-Tek^®), and immunofluorescence was performed as described previously (10). In short, the slides were fixed in cold acetone followed by methanol (5 min each) and air-dried. Cells were then permeabilized and blocked in phosphate-buffered saline containing 0.2% Triton X-100 and 0.5% blocking reagent (Roche Molecular Biochemicals) for 30 min at room temperature. Subsequently, the slides were incubated with polyclonal rabbit anti-PLAG1 antibody (PEM190) (10) followed by washes in phosphate-buffered saline containing 0.2% Triton X-100 (3 × 10 min), incubation with Texas Red-conjugated donkey anti-rabbit secondary antibody (Amersham Biosciences), and again washes (three times). Finally, the slides were mounted in Vectashield (Vector Laboratories, Inc.) supplemented with 0.4 µg/ml 4'-6-diamine-2-phenylinidole-dihydrochloride (DAPI, Roche Molecular Biochemicals) and analyzed on a Zeiss Axiophot fluorescence microscope equipped with a cooled digital CCD camera system (Photometrics) using SmartCapture^TM software.

Cell Lines, Transfection, and Fluorescence-- 293T human embryonic kidney epithelial cells were used to examine the expression and subcellular localization of -galactosidase-PLAG1-GFP fusion proteins. Cell lines were grown in Dulbecco's modified Eagle's medium/F12 (1:1) (Invitrogen) supplemented with 10% fetal calf serum and cultured at 37 °C in a humidified 5% CO₂ atmosphere.

Transient transfections were performed using FuGENE 6 transfection reagent (Roche Molecular Biochemicals) according to the supplier's instructions. The cells were grown to 70-80% confluency on coverslips in 24-well plates. For each transfection 0.75 µl of FuGENE 6 transfection reagent in 25 µl of serum-free Dulbecco's modified Eagle's medium (Invitrogen) was added to 0.5 µg of DNA and incubated at room temperature for 15 min after which the mixture was applied directly to the growth medium of the cells. Cells were incubated further at 37 °C for 18-24 h followed by fixation in 4% formaldehyde in phosphate-buffered saline for 10 min at room temperature and three subsequent wash steps in phosphate-buffered saline. Finally, the slides were mounted in Vectashield (Vector Laboratories, Inc.) supplemented with 1/1000 DAPI and analyzed under the fluorescence microscope.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Identification of Karyopherin 2 As a PLAG1-interacting Protein via a Yeast Two-hybrid Assay-- The yeast two-hybrid assay was used to identify proteins interacting with the N-terminal part of PLAG1-(2-70) (Fig. 1). Approximately 160,000 colonies were screened on triple selective (lacking Trp, Leu, and His) agar and subsequently on X-gal filters. Thirty colonies were retained as candidates for specifying PLAG1-interacting proteins. Different prey plasmids were isolated from these yeast colonies. The plasmids were subsequently reassayed by cotransformation into yeast-competent cells with either the PLAG1 pGBT9 construct, the empty pGBT9, or pGBT9 containing an unrelated cDNA insert. Three independent prey plasmids generated yeasts that were able to grow on SD agar plates and that were only positive in the X-gal filter assay with the PLAG1 bait construct. All three prey plasmids contained an insert of ~2 kb, which upon nucleotide sequence analysis appeared to encode mouse karyopherin 2 (GenBank^TM accession number AAH06720) starting at amino acid 38. This N-terminal truncated protein lacks a part of the importin -binding domain but still contains the armadillo repeats responsible for the binding to the NLS.

View larger version (10K): [in this window] [in a new window]   Fig. 1.   Overview of human PLAG1 domain structure. The diagram indicates the location of the trans-activation domain and the seven zinc fingers (F). The fingers shown in bold (fingers 3, 6, and 7) are essential for DNA binding. Different isoforms of the protein are predicted starting either at methionine 83 or methionine 100. The N-terminal fragment of PLAG1-(2-70) was used as a bait in the yeast two-hybrid screen. The two NLSs as predicted by Psort are depicted.

Karyopherin 2 Binds to PLAG1 and not to PLAGL1 or PLAGL2 in a GST Pull-down Assay-- To confirm the physical interaction between karyopherin 2 and PLAG1, we carried out a GST pull-down assay using in vitro translated full-length mouse [³⁵S]karyopherin 2 and PLAG-(2-244) fused to GST. As shown in Fig. 2, karyopherin 2 bound strongly to GST-PLAG1-(2-244), while it did not bind to GST used as a control. Moreover, GST-PLAG1-(84-244), lacking the N-terminal PLAG1 segment used in the two-hybrid screen, failed to interact with karyopherin 2. Together, these results confirm the interaction between karyopherin 2 and PLAG1 identified in the two-hybrid assay and moreover show that this interaction is specific for the N-terminal region PLAG1-(2-84). As PLAG1 is a member of the highly conserved PLAG subfamily of zinc finger proteins, which includes PLAGL1 and PLAGL2, we were interested in whether these proteins could interact with karyopherin 2. To determine this possibility, we studied the interaction of karyopherin 2 with PLAGL1 and PLAGL2 in a GST pull-down experiment. As shown in Fig. 2, GST-PLAGL1-(2-215) and GST-PLAGL2-(2-250) did not interact with karyopherin 2 in contrast with the strong interaction between GST-PLAG1-(2-244) and karyopherin 2.

View larger version (33K): [in this window] [in a new window]   Fig. 2.   Karyopherin 2 binds to GST-PLAG1-(N2-C244) in a GST pulldown assay. A, autoradiograph of 35S-labeled full-length karyopherin 2 recovered after interaction with GST alone, GST-PLAG1-(84-244), GST-PLAG1-(2-244), GST-PLAGL1-(2-215), and GST-PLAGL2-(2-250). 10% of the input amount of 35S-labeled karyopherin 2 used in the GST pull-down assay is shown in the input lane. B, 5% of the input amount of GST fusion proteins used in the GST pull-down assay was put on SDS-PAGE and visualized by Coomassie Blue staining. The upper band of GST-PLAG1 corresponds to the full-length fusion, and the lower band is the result of site-specific cleavage of PLAG1 fusion proteins.

Identification of Two Putative NLSs in PLAG1-- Karyopherin 2 is implicated in active nuclear import of various proteins via interaction with a NLS (13), consisting of short stretches of basic amino acids. The identification of karyopherin 2 as a PLAG1-interacting protein suggests that the interaction is mediated via a nuclear localization signal in PLAG1. An analysis of PLAG1 on the Psort server (psort.nibb.ac.jp/) identified two candidate NLS consensus signals, KRKR and KPKR, positioned at amino acids 22-25 and 29-32, respectively (Fig. 1).

PLAG1 Putative NLS1 Is Essential for Interaction with Karyopherin 2-- Previous studies have shown that mutagenesis of the basic residues within an NLS motif eliminates karyopherin 2 recognition. Therefore, we mutated the candidate NLS motifs in GST-PLAG1-(2-244) by replacing two basic residues with alanines and analyzed the PLAG1-karyopherin 2 interaction in a GST pull-down assay (Fig. 3). The mutation of putative NLS1 of PLAG1 eliminated the interaction with karyopherin 2 (mNLS1). In contrast, the mutation of the predicted NLS2 did not affect the binding of karyopherin 2 to GST-PLAG1 (mNLS2). As expected, the mutation of both predicted NLSs eliminated the interaction between PLAG1 and karyopherin 2 (mNLS1+2). These data show that NLS1 of PLAG1 is involved in karyopherin 2 interaction, whereas NLS2 is not.

View larger version (28K): [in this window] [in a new window]   Fig. 3.   The predicted NLS1 of GST-PLAG1-(N2-C244) is essential for karyopherin 2 binding. A, autoradiograph of 35S-labeled full-length karyopherin 2 recovered after interaction with GST-PLAG1-(2-244) (WT), GST-PLAG1-(2-244) with mutated NLS1 (mNLS1), GST-PLAG1-(2-244) with mutated NLS2 (mNLS2), and GST-PLAG1-(2-244) with mutated NLS1 and mutated NLS2 (mNLS1+2). 10% of the input amount of 35S-labeled karyopherin 2 used in the GST pull-down assay is shown in the input lane. B, 5% of the input amount of GST fusion proteins used in the GST pull-down assay was loaded on SDS-PAGE and visualized by Coomassie Blue staining. The upper band of GST-PLAG1 corresponds to the full-length fusion, and the lower band is the result of site-specific cleavage of PLAG1 fusion proteins.

Putative NLS1 of PLAG1 Plays a Role in Its Nuclear Localization-- Because karyopherin 2 is involved in active nuclear import via physical interaction with its cargo, PLAG1 is likely to be such a cargo protein and therefore is expected to localize to the nucleus. We recently showed that exogenous PLAG1 transfected in COS-1 kidney fibroblast cells is localized in the nucleus (5). In addition, immunofluorescence on cultured pleomorphic adenoma cells of the salivary gland with PLAG1-specific antibodies shows nuclear localization of the endogenous PLAG1 protein (Fig. 4). To assess the role of the predicted NLSs in vivo, we monitored the intracellular localization in 293T cells of various -galactosidase-PLAG1-GFP fusion proteins by fluorescence microscopy. Whereas wild-type PLAG1 fusion protein was found exclusively in the nucleus of all transfected cells (Fig. 5, panel A), the mutation of putative NLS1 resulted in a heterogenic picture with 56% of the cells showing only nuclear fluorescence, 28% only cytoplasmic, and 16% both nuclear and cytoplasmic fluorescence (Fig. 5, panel B). The mutation of predicted NLS2 had no effect on the nuclear localization of the fusion protein (Fig. 5, panel C). The effect of the mutation in NLS1 was even more pronounced when we fused only the N-terminal segment of PLAG1-(2-41) to -galactosidase and GFP. A complete shift from exclusively nuclear localization for the wild-type fusion (Fig. 5, panel E) to exclusively cytoplasmic localization for the NLS1 mutant (Fig. 5, panel F) was observed. Again, the mutation of predicted NLS2 resulted in a picture indistinguishable from the wild-type fusion (Fig. 5, panel G). In addition, the fusion of only eight amino acids including the predicted NLS1 (SGKRKRGE) to the -galactosidase-GFP converted this exclusively cytoplasmic protein to an exclusively nuclear one (Fig. 5, panel I). Taken together, these results show the functional importance of NLS1 for active nuclear import of PLAG1, whereas we could not demonstrate a role for NLS2.

View larger version (23K): [in this window] [in a new window]   Fig. 4.   Endogenous PLAG1 is localized in the nucleus of tumor cells. Immunofluorescence of cultured pleomorphic adenoma cells from the salivary gland with PLAG1-specific antibodies (PEM190) and Texas Red-conjugated donkey anti-rabbit antibodies. Nuclei are stained with DAPI.

View larger version (35K): [in this window] [in a new window]   Fig. 5.   The intracellular localization of various -galactosidase-PLAG1-GFP fusion proteins in transiently transfected 293T cells. Fluorescence of 293T cells transfected with -galactosidase-PLAG1-(full-length, wild-type)-GFP (A); -galactosidase-PLAG1-(full-length, NLS1-mutated)-GFP (B); -galactosidase-PLAG1-(full-length, NLS2-mutated)-GFP (C); -galactosidase-PLAG1-(full-length, NLS1- and NLS2-mutated)-GFP (D); -galactosidase-PLAG1-(2-41, wild-type)-GFP (E); -galactosidase-PLAG1-(2-41, NLS1-mutated)-GFP (F); -galactosidase-PLAG1-(2-41, NLS2-mutated)-GFP (G); -galactosidase-PLAG1-(2-41, NLS1- and NLS2-mutated)-GFP (H); -galactosidase-(SGKRKRGE)-GFP (I); -galactosidase-PLAG1-(244-500)-GFP (J); and -galactosidase-PLAG1-(42-242)-GFP (K). All nuclei showed blue fluorescence after DAPI staining. The intracellular localization of the corresponding -galactosidase-PLAG1-GFP fusion proteins in terms of percentage is shown in the table. For each different fusion protein around 100 cells was evaluated by fluorescent microscopy and classified as having either green fluorescence exclusively in the nucleus (N) or exclusively in the cytoplasm (C) or both in nucleus and cytoplasm (N + C).

Assay of Other Segments of PLAG1 for Nuclear Import-- The mutation of the predicted NLS1 in the context of full-length PLAG1 protein fused to -galactosidase and GFP did not completely inhibit nuclear import as shown above. Indeed, 72% of the transfected cells still showed nuclear localization of the fusion proteins (Fig. 5). In contrast, the same NLS1 mutation in the N-terminal segment of PLAG1-(2-41) fully inhibited nuclear import. These results indicate the presence of another motif in PLAG1 that can determine active nuclear localization independently of NLS1. To pinpoint this motif, several PLAG1 deletion constructs were assessed for nuclear import. A -galactosidase-PLAG1-GFP fusion protein containing the C-terminal activation domain PLAG1-(244-500) failed to enter the nucleus (Fig. 5, panel J). On the other hand, the fusion of the zinc finger domain without the N-terminal part PLAG1-(41-242) gave a heterogenic picture (Fig. 5, panel K) with 47% of the transfected cells showing exclusive nuclear fluorescence, 33% exclusive cytoplasmic, and 20% both nuclear and cytoplasmic fluorescence (Fig. 5). As such, these data assign a functional role in nuclear import to this region despite its apparent lack of an NLS motif and its failure to interact with karyopherin 2 in the GST pull-down assay (Fig. 2, GST-PLAG1-(84-244)).

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Karyopherin 2 or Kpna2 (also called pendulin) (Genbank^TM accession number AAH06720), a member of the importin family, was identified as a PLAG1-interacting protein in a yeast two-hybrid screen. Karyopherin 2 is implicated in active nuclear import of various proteins by interaction with a NLS (13). A conventional nuclear localization sequence consists of short stretches of basic amino acids, either a single stretch (i.e. monopartite NLS) or two stretches separated by a spacer region (i.e. bipartite). Importin  proteins deliver their NLS-tagged cargo into the nucleus through interaction with importin . Importin  docks the karyopherin-cargo complex with the nuclear pore followed by translocation of the complex through the pore via an energy-dependent process. In the nucleus the karyopherin/cargo complex dissociates, after which the karyopherins recycle to the cytoplasm. Karyopherin 2 consists of an N-terminal hydrophilic importin -binding domain (14-16) and a hydrophobic central region composed of armadillo repeats, which bind to the NLS motif (17-19).

The results presented here reveal the presence of two regions in PLAG1 that independently determine nuclear localization. The first one is the predicted NLS1 in the N-terminal part of PLAG1 that mediates nuclear import through interaction with karyopherin 2. The second is the zinc finger domain of PLAG1, which is responsible for DNA binding. This domain can also localize the protein to the nucleus, although no candidate NLS consensus signals could be identified using Psort. The mechanism by which this domain targets PLAG1 to the nucleus, either by direct interaction with one of the several importin  proteins or indirectly via proteins that bind to the zinc finger domain, is still unknown. Such redundancy in nuclear targeting is not uncommon for transcription factors, enzymes, or structural proteins whose nuclear import is essential. A seemingly superfluous NLS was found in the 42-kDa B-cell-specific activator protein. The N-terminal DNA-binding domain termed the paired box and a defined NLS in the central domain of B-cell-specific activator protein are redundant in nuclear targeting of B-cell-specific activator protein (20). Multiple functional NLSs were identified in DNA topoisomerases I, II, and II (21, 22). The histone H1⁰ contains multiple sequence elements for nuclear targeting (23).

The contribution of passive diffusion in nuclear localization of PLAG1 has not yet been established. With a molecular mass of 55 kDa, PLAG1 might be able to pass through nuclear pores by diffusion. Nevertheless, small proteins are often equipped with NLSs to ensure fast nuclear accumulation by active transport. The results presented in this work focus on active transport, because fusion to -galactosidase and GFP adds an additional 102 kDa (75 and 27 kDa, respectively) to the molecular mass, thus severely impairing if not completely preventing passive diffusion.

The PLAG1 family members PLAGL1 and PLAGL2 did not interact with karyopherin 2 in vitro. In addition, an analysis with Psort did not reveal candidate NLS consensus signals in PLAGL1 or PLAGL2. Nevertheless, both are shown to be nuclear proteins (9, 24), in agreement with their function as transcription factors. It is possible that their zinc finger domain is responsible for their nuclear import.

As a result of alternative splicing of PLAG1, different isoforms of the protein are predicted starting either at methionine 83 or methionine 100. These truncated PLAG1 products lack the identified NLS and do not bind to karyopherin 2 in a GST pull-down assay. In these isoforms, the zinc finger domain probably ensures nuclear targeting.

The fact that PLAG1 is equipped with an extra NLS absent in the other PLAG family members and in the PLAG1 isoforms suggests that its nuclear import might be of particular physiological relevance. Indeed, the nuclear targeting directed by NLS1 is more efficient in comparison to the nuclear targeting by the zinc finger domain. A cytoplasmic protein is driven to the nucleus in 100% of the cells after fusion to NLS1. Introduction of the DNA-binding domain of PLAG1, in contrast, confers nuclear localization in only 47% of the cells (see table in Fig. 5). However, it is not yet clear why PLAG1 needs a highly efficient mechanism for nuclear import mediated by NLS1, which is in contrast to the other family members and the isoforms that lack the identified NLS. It would not be surprising to learn that the extra NLS of PLAG1 has some yet uncharacterized physiological role. A nuclear localization signal in chicken heat shock transcription factor 3 (cHSF), for example, targets the protein to the nucleus and in addition is essential for its stress-induced dimer-to-trimer transition (25).

In the context of the whole cell, the interaction between PLAG1 and karyopherin 2 is not exclusive. PLAG1 has to compete with other NLS-containing cargo proteins for karyopherin 2 in order to be transported into the nucleus. Some import substrates such as small nuclear ribonuclear proteins, histones, and ribosomal subunits are imported into the nucleus in high but relatively constant amounts, whereas other import substrates such as transcription factors are only required in the nucleus for short periods in small amounts in response to specific cellular signals. The mechanisms inducing rapid changes in signal binding affinity such as phosphorylation or intermolecular masking are essential to ensure dynamic exchanges between nucleus and cytoplasm. In the case of PLAG1, it seems that the protein uses the nuclear import machinery to transport all of the highly overexpressed PLAG1 protein into the nucleus despite the competition with other cargo proteins. Therefore, it seems that either PLAG1 is not regulated at the level of nuclear import and is always rapidly transported into the nucleus to affect its target genes or the mechanism that decreases the binding affinity and allows competition of other NLS-containing proteins is not functional in 293T cells.

	FOOTNOTES

^* This work was supported in part by the Geconcerteerde Onderzoeksactie (GOA, 1997-2001), the Fonds voor Wetenschappelijk Onderzoek Vlaanderen (FWO), and the Algemene Spaar en Llrente Kas-programma voor Kankeronderzoek.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

An "Aspirant" of the FWO.

^§ Postdoctoral Fellow of the FWO.

^¶ To whom correspondence should be addressed. Tel.: 32-016-34-59-87; Fax: 32-016-34-60-73; E-mail: wim.vandeven@med.kuleuven.ac.be.

Published, JBC Papers in Press, March 6, 2002, DOI 10.1074/jbc.M112112200

² Hensen, K., Van Valckenborgh, I. C., Kas, K., Van de Ven, W. J., and Voz, M. L. (2002) Cancer Res. 62, 1510-1517.

	ABBREVIATIONS

The abbreviations used are: PLAG1, pleomorphic adenoma gene-like 1 protein; PLAGL2, pleomorphic adenoma gene-like 2 protein; GST, glutathione S-transferase; GFP, green fluorescent protein; SD, synthetic dropout; DAPI, 4'-6-diamine-2-phenylindole-dihydrochloride; IGF-II, insulin-like growth factor II gene.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				S. Heckl and U. Vogel Nuclear Trafficking of a Gadolinium Conjugate in Nude Mice Xenografted with Human LN-229 Glioma J. Pharmacol. Exp. Ther., November 1, 2006; 319(2): 657 - 662. [Abstract] [Full Text] [PDF]

				L. Heath, A.-L. Williamson, and E. P. Rybicki The Capsid Protein of Beak and Feather Disease Virus Binds to the Viral DNA and Is Responsible for Transporting the Replication-Associated Protein into the Nucleus J. Virol., July 15, 2006; 80(14): 7219 - 7225. [Abstract] [Full Text] [PDF]

				J. Declercq, F. Van Dyck, C. V. Braem, I. C. Van Valckenborgh, M. Voz, M. Wassef, L. Schoonjans, B. Van Damme, L. Fiette, and W. J.M. Van de Ven Salivary Gland Tumors in Transgenic Mice with Targeted PLAG1 Proto-Oncogene Overexpression Cancer Res., June 1, 2005; 65(11): 4544 - 4553. [Abstract] [Full Text] [PDF]

				N. Shomron, M. Reznik, and G. Ast Splicing Factor hSlu7 Contains a Unique Functional Domain Required to Retain the Protein within the Nucleus Mol. Biol. Cell, August 1, 2004; 15(8): 3782 - 3795. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 277/22/19673    most recent M112112200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Braem, C. V. Articles by Voz, M. L. Search for Related Content PubMed PubMed Citation Articles by Braem, C. V. Articles by Voz, M. L. Social Bookmarking                      What's this?