INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

The transport of proteins between the nucleus and the cytoplasm is an essential activity of eukaryotic cells. Protein transport across the double lipid bilayer of the nuclear membrane occurs through large macromolecular complexes termed nuclear pore complexes (NPCs). Transport is generally dependent upon specific cis-acting signals within the cargo protein and the corresponding trans-acting factors that mediate the translocation of the protein cargo through the NPC (for reviews, see references 8 and 30). The best-characterized transport pathway is utilized by proteins that contain a classical nuclear localization signal (NLS). The prototypic NLS present in the simian virus 40 (SV40) large T protein contains a short stretch of basic residues that are critically required for nuclear localization (15, 25). NLS-bearing proteins are bound by the heterodimeric importin - complex (7, 16-18, 31, 36), which mediates docking of the receptor-cargo complex at the cytoplasmic face of the NPC. Translocation of the receptor-cargo complex through the NPC is poorly understood, but is thought to involve direct interactions between the importin  subunit and specific components of the NPC (37, 38). Nuclear import is terminated in the nucleus by RanGTP-induced dissociation of the receptor-cargo complex, releasing the NLS-bearing cargo into the nucleus (21, 38). The importin  and  proteins are subsequently recycled back to the cytoplasm to participate in a second round of nuclear import (29).

The export of proteins from the nucleus is accomplished by an analogous mechanism. Proteins that are destined to be exported from the nucleus typically contain nuclear export sequences (NESs) comprised of short clusters of leucine or other hydrophobic residues (12, 47). NES-bearing proteins are recognized by the nuclear export receptor, CRM1, in a RanGTP-dependent manner (4, 13, 14, 33, 45). Although the details of nuclear export are poorly understood, the NES-bearing cargo-CRM1 receptor complex is thought to dock at the nuclear side of the NPC, followed by translocation and dissociation at the cytoplasmic face of the NPC.

A critical aspect of both nuclear import and nuclear export is the asymmetric distribution of RanGTP and RanGDP between the nucleus and the cytoplasm. The GTPase-activating protein for Ran, RanGAP, is located at the cytoplasmic face of the NPC, while the guanine nucleotide exchange factor for Ran, RCC1, is tightly associated with chromatin (19, 34). This differential distribution of RanGAP and RCC1 between the cytoplasm and the nucleus has led to the prediction that Ran will be in the GTP-bound form in the nucleus, while the GDP-bound form will predominate in the cytoplasm. Experimental perturbation of the RanGTP-RanGDP gradient inhibits both nuclear import and nuclear export. For example, the GTP-bound form of the dominant-negative RanQ69L protein inhibits NLS-dependent nuclear import in digitonin-permeabilized cells, presumably by inhibiting binding of an NLS-bearing protein to the importin - receptor complex (21, 38). Likewise, depletion of nuclear RanGTP levels inhibits NES-dependent nuclear export, presumably by preventing a NES-bearing protein from binding to CRM1 in a manner that is competent for transport (4, 39).

Transport of proteins between the nucleus and the cytoplasm provides an effective mechanism for regulation of gene expression. A striking example of how gene expression can be regulated at the level of protein transport between the nucleus and the cytoplasm is provided by the NF-B/Rel family of transcription factors and their inhibitory IB proteins (reviewed in reference 5). For example, the IB protein is able to both inhibit nuclear import of NF-B/Rel proteins and direct the export of NF-B/Rel proteins from the nucleus (2, 3, 20, 22, 43). The ability of IB to act both in the cytoplasm and in the nucleus requires that IB itself travel through the NPC. The second ankyrin repeat of IB is critically required for nuclear localization of IB and is able to functionally substitute for a classical NLS (42). In this report, we have utilized a combination of in vitro and in vivo approaches to provide mechanistic insight into nuclear shuttling of IB. Our results indicate that nuclear import of IB is accomplished by a Ran-independent mechanism, while nuclear export of IB requires the Ran-dependent CRM1 nuclear export receptor.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Construction of recombinant DNA molecules. The IB clones used in this study were derived from the avian IB cDNA clone isolated by Davis et al. (9) and constructed by standard techniques (44). To construct the GST (glutathione S-transferase)-IB expression vector, an EcoRI fragment containing 69 bp of 5' nontranslated sequence, the entire 954 bp of the IB open reading frame, and 762 bp of 3' nontranslated sequence was cloned into the EcoRI site of pGEX-2T in the proper orientation for expression. The GST-IB-N69 expression vector contains an IB insert that lacks the first 69 codons of the IB open reading frame. The GST-IB-C51 expression vector contains an IB insert that contains a termination codon at codon 268 (41). The GST-IB-Ank2 expression vector contains a deletion which removes codons 98 to 142 (42). The GST-IB-ARD (ankyrin repeat domain) expression vector contains codons 70 to 254 of IB cloned into the SmaI site of pGEX3X. The GFP (green fluorescent protein)-IB expression vectors were constructed by inserting the various IB fragments from the GST-based vectors into the GFP-C3 eukaryote expression vector (Clontech). The GST-NLS expression vector contains an oligonucleotide which encodes an NLS derived from the SV40 large T protein inserted into the SmaI site of pGEX1. The GST-M9 expression vector contains an oligonucleotide which encodes the M9 nuclear import sequence from the hnRNP A1 protein cloned into the SmaI site of pGEX1 (35). The pQE32-derived expression vectors for wild-type Ran and the RanQ69L protein and the pQE60-derived expression vector for the importin (45-462) protein were obtained from Dirk Gorlich (University of Heidelberg). A pET-based expression vector for the importin -binding domain of importin  (IBB) was obtained from Steve Adam (Northwestern University). A pQE32-derived expression vector for RanBP1 was obtained from Iain Mattaj's laboratory (EMBL, Heidelberg, Germany).

Expression and purification of recombinant proteins. The recombinant GST fusion proteins were expressed in Escherichia coli strain BL21(DE3). Cultures were grown to an optical density at 600 nm of 0.4 and induced with 0.2 mM isopropyl thiogalactoside (IPTG) (Sigma) for 4 h at 20°C. The bacterial cell pellets were resuspended in ice-cold phosphate-buffered saline (PBS [pH 7.4]) containing 0.1% Triton X-100; 0.2 mM dithiothreitol (DTT); 1 mM phenylmethylsulfonyl fluoride; and 1 µg (each) of antipain, aprotinin, leupeptin, pepstatin, and soybean trypsin-chymotrypsin inhibitor per ml. Lysis of the cell pellets was conducted by brief sonication on ice. The lysates were cleared by centrifugation at 14,000 × g for 15 min at 4°C and the soluble GST fusion proteins were bound to glutathione-agarose beads (Sigma) for 30 min at room temperature. The glutathione-agarose beads were extensively washed with ice-cold PBS (pH 7.4), and the recombinant GST fusion proteins were eluted with 10 mM reduced glutathione in 50 mM Tris-Cl (pH 8.0) and dialyzed against transport buffer (20 mM HEPES (pH 7.3), 110 mM potassium acetate, 5 mM sodium acetate, 2 mM magnesium acetate, 1 mM EGTA).

In vitro nuclear import assay. HeLa cells were grown in Dulbecco's modified Eagle's medium (low glucose) supplemented with 10% fetal bovine serum (FBS) in a 37°C, 5% CO₂ incubator. Approximately 16 h prior to the transport assays, 6.5 × 10⁵ HeLa cells were plated onto 35 mm-diameter plates containing glass coverslips. The in vitro nuclear transport assays were conducted essentially as described in reference 1. In brief, cells on coverslips were permeabilized with 50 µg of digitonin per ml (Calbiochem) in transport buffer for 5 min on ice. The transport reactions were typically conducted for 30 min at 30°C, except where noted. A standard 50-µl transport reaction contained an energy-regenerating system (1 mM ATP, 0.1 mM GTP, 5 mM creatine phosphate, 20 U of creatine phosphokinase per ml), protease inhibitor mix, 2 mM DTT, and 15 µl of rabbit reticulocyte lysate (50 mg of total protein per ml). The unlabeled import substrates were added to a final concentration of 100 µg/ml, while the fluorescein-labeled import substrates were added to a final concentration of 50 µg/ml. For some samples, HeLa cells growing on coverslips were pretreated with 10 nM leptomycin B for 30 min prior to digitonin permeabilization, and 10 nM leptomycin B was included throughout the 30-min time course of the transport reactions. For the energy-dependence experiments, the reticulocyte lysates were pretreated with 25 U of apyrase per ml for 30 min prior to the import reactions, which were performed in the absence of ATP, creatine phosphate, and creatine phosphokinase.

HeLa cell transfections. HeLa cells were purchased from American Type Culture Collection and cultured in Dulbecco's modified Eagle's medium containing 10% FBS. HeLa cells were transfected with the GFP-IB expression plasmids by using Fugene 6 according to the manufacturor's instructions (Boehringer Mannheim). Expression of the GFP-IB fusion proteins was confirmed by immunoblot analysis. Localization of the GFP-IB fusion proteins was determined by indirect immunofluorescence with an anti-GFP antibody (Chemicon).

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Nuclear shuttling of IB in digitonin-permeabilized HeLa cells. A digitonin-permeabilized cell assay was used to characterize nuclear import of IB. IB was expressed as a GST fusion protein in E. coli, and the ability of the GST-IB fusion protein to accumulate in the nuclei of digitonin-permeabilized HeLa cells was monitored by indirect immunofluorescence with a monoclonal antibody directed against GST. The NLS from the SV40 large T protein was fused to GST (GST-NLS) to serve as a positive control for nuclear import. Nuclear staining was not observed in the absence of an import substrate (Fig. 1A). In the presence of reticulocyte lysate and an energy-regenerating system, GST-NLS was efficiently imported into the nucleus (Fig. 1C). Under these conditions, neither GST nor a GST fusion protein containing a mutant NLS was imported into the nucleus (data not shown).

View larger version (83K): [in this window] [in a new window]   FIG. 1.   Nuclear accumulation of IB in digitonin-permeabilized cells is leptomycin B sensitive. HeLa cells were either left untreated (A to C) or were treated with 10 nM leptomycin B (LMB [D to F]) for 30 min prior to permeabilization with digitonin. Digitonin-permeabilized HeLa cells were incubated for 30 min at 30°C in the presence of reticulocyte lysate and an energy-regenerating system. Leptomycin B (10 nM) was included in the import reaction for the samples shown in panels D to F. The cells were incubated in the absence of import substrate (A and D) and in the presence of either GST-IB (B and E) or GST-NLS (C and F). The cellular localization of the import substrates was visualized by indirect immunofluorescence with a monoclonal antibody directed against GST.

IB contains multiple functional domains that contribute to rapid shuttling between the cytoplasm and the nucleus. To define regions within IB that specify nuclear import and nuclear export, we constructed a series of mutant GST-IB fusion proteins (Fig. 2). These mutant proteins were expressed in E. coli, purified with glutathione-agarose, and assayed for their ability to accumulate in the nucleus of digitonin-permeabilized HeLa cells in the absence and presence of leptomycin B (Fig. 3). To ensure that nuclear export of IB was not saturated by excess input protein, the ability of leptomycin B to increase nuclear accumulation of the mutant IB proteins was measured at 2.5 µg of input protein.

View larger version (53K): [in this window] [in a new window]   FIG. 2.   Domain organization of IB. IB contains an N-terminal signal response domain, a central ARD comprised of six ankyrin repeats, and a C-terminal PEST domain. The location of two canonical leucine-rich NESs (amino acids 45 to 55 and 273 to 283) in IB is indicated (LLLL). The structure of the mutant IB proteins used in these experiments is diagrammed.

View larger version (69K): [in this window] [in a new window]   FIG. 3.   Definition of nuclear import and NESs in IB. Nuclear import reactions were performed with the indicated protein substrates in either the absence (A, C, E, G, I, and K) or presence of 10 nM leptomycin B (B, D, F, H, J, and L). The cellular localization of the import substrates was determined by indirect immunofluorescence with either a monoclonal antibody directed against GST or by direct fluorescence with FITC-labeled proteins. Parallel experiments with both detection methods were done for all protein samples, with the exception of the GST-IB-C251 protein, which was only detected by direct fluorescence.

Nuclear shuttling of IB in transfected HeLa cells. The failure of the C-terminal NES of IB to contribute to nuclear export of IB in digitonin-permeabilized HeLa cells was surprising, because several reports have indicated that this canonical leucine-rich NES is required for IB-mediated nuclear export of NF-B/Rel proteins (3, 42). To further characterize nuclear shuttling of IB in vivo, we constructed fusion proteins between the GFP and the various IB proteins. These fusion proteins were expressed in HeLa cells, and the ability of leptomycin B to alter the nuclear-cytoplasmic distribution of fusion proteins was measured (Fig. 4 and Table 1).

View larger version (55K): [in this window] [in a new window]   FIG. 4.   Nuclear shuttling of GFP-IB chimeric proteins in transfected HeLa cells. HeLa cells transfected with expression vectors for the indicated GFP-IB fusion proteins were left untreated or were treated with leptomycin B for 1 h. The cellular localization of the fusion proteins was detected by indirect immunofluorescence with a chicken antibody raised against GFP. More than 200 cells were examined for each fusion protein (see Table 1 for quantitation), and representative cells are shown.

View this table: [in this window] [in a new window]   TABLE 1.   Localization of GFP-IB fusion proteinsa

Nuclear import of IB is energy and temperature dependent and is blocked by a dominant-negative importin  protein. To determine whether nuclear accumulation of IB reflects active nuclear transport or passive diffusion through the NPC, cytosolic extracts were pretreated with apyrase to deplete the extracts of high-energy phosphate compounds. The ability of the energy-depleted extracts to support nuclear import of GST-IB or GST-NLS in the absence of an energy-regenerating system was determined. Nuclear import of both GST-IB (Fig. 5D) and GST-NLS (Fig. 5H) was inhibited when the cytosolic extracts were pretreated with apyrase. Similar results were obtained when hexokinase and glucose were used to deplete the energy pools present in the cytosolic extracts (data not shown).

View larger version (80K): [in this window] [in a new window]   FIG. 5.   Nuclear import of IB is energy and temperature dependent. Nuclear import reactions into digitonin-permeabilized HeLa cells were performed with either GST-IB (A to D) or GST-NLS (E to H). Leptomycin B (10 nM) was included in the nuclear import reactions with GST-IB (A to C). The import reactions were carried out under standard conditions (A and E), in the presence of WGA (B and F), on ice (C and G), or in the presence of 25 U of apyrase per ml (D and H). The cellular localization of the import substrates was determined by indirect immunofluorescence with a monoclonal antibody directed against GST. Panels A and E were from equivalent exposure settings. Panels B to D and F to H were exposed for 10 to 20 times longer than panels A and E.

View larger version (47K): [in this window] [in a new window]   FIG. 6.   Binding of IB to the nuclear membrane in the presence of a dominant-negative importin  protein. Nuclear import reactions using the indicated proteins were performed in the absence (A, C, and E) or presence (B, D, and F) of the dominant-negative importin (45-462) protein. Leptomycin B (10 nM) was included in the import reactions shown in panels A and B. The images shown are representative of more than 50 cells that were examined for each import substrate. The import substrates were visualized by indirect immunofluorescence by confocal laser scanning microscopy.

Rate-limiting factors for nuclear import of IB are not lost during the digitonin-permeabilization procedure. To investigate the dependence of IB import on soluble protein factors, the ability of GST-IB to accumulate in the nuclei of HeLa cells in the absence of exogenously added cytosol was determined. Nuclear import of GST-NLS was strictly dependent on the presence of exogenously added cytosol (Fig. 7E and F). In contrast, GST-IB readily accumulated in the nuclei of HeLa cells, even when cytosol was not added to the import reactions (Fig. 7A and B). Leptomycin B was included in the nuclear import reactions for GST-IB. However, the ability of GST-IB to accumulate in the nucleus in the absence of exogenously added soluble factors was not an artifact resulting from the inclusion of leptomycin B in the import reaction, because the GST-IB-N69 protein efficiently accumulated in HeLa cell nuclei in a cytosol-independent manner in the absence of leptomycin B (Fig. 7C and D).

View larger version (74K): [in this window] [in a new window]   FIG. 7.   Nuclear import of IB is independent of exogenous cytosol. Nuclear import reactions using the indicated proteins were carried out either in the absence of reticulocyte lysate (A, C, and E) or in the presence of 100 µg of reticulocyte lysate. Leptomycin B (LMB) was included in the nuclear import reactions using GST-IB (A and B). The cellular localization of the import substrates was determined by indirect immunofluorescence with a monoclonal antibody directed against GST.

Nuclear import of IB is not blocked by saturation of the NLS-dependent or M9-dependent import pathways. Nuclear import of proteins that utilize known nuclear import pathways, such as the NLS-dependent pathway or the M9-dependent pathway, can be inhibited by saturation of the respective import receptor (30). As anticipated, we found that a 100-fold molar excess of GST-NLS (Fig. 8D) or a 7-fold molar excess of IBB (Fig. 8F) markedly reduced nuclear import of FITC-labeled GST-NLS. Likewise, a 100-fold molar excess of GST-M9 significantly inhibited nuclear import of FITC-labeled GST-M9 (Fig. 8J). However, nuclear import of FITC-labeled GST-IB-N69 was not inhibited by the addition of a 100-fold molar excess of GST (Fig. 8A), GST-NLS (Fig. 8C), a 7-fold molar excess of IBB (Fig. 8E), or a 100-fold molar excess of GST-M9 (Fig. 8I). Similar results were obtained with the full-length GST-IB protein when the nuclear import reactions were carried out in the presence of leptomycin B (data not shown). Taken together, these results indicate that nuclear import of GST-IB is independent of either the NLS-dependent or the M9-dependent nuclear import pathway.

View larger version (56K): [in this window] [in a new window]   FIG. 8.   Nuclear import of IB is independent of NLS- and M9-mediated nuclear import pathways. Nuclear import reactions with fluorescein-labeled GST-IB-N69 (A, C, E, G, and I), fluorescein-labeled GST-NLS (B, D, and F), or fluorescein-labeled GST-M9 (H and J) were performed. The import reactions were performed in the presence of a 100-fold molar excess of unlabeled GST (A, B, G, and H), GST-NLS (C and D), or GST-M9 (I and J). For the import reactions shown in panels E and F, the IBB was included at a final concentration of 30 µM. The cellular localization of the import substrates was determined by direct fluorescence.

IB utilizes a high-throughput nuclear import pathway. The saturability of both nuclear import and nuclear export of IB was determined by the ability of FITC-labeled GST-IB to accumulate in the nuclei of digitonin-permeabilized HeLa cells following addition of unlabeled GST-IB. These experiments were performed with limiting amounts of FITC-GST-IB protein to ensure that nuclear export was not titrated out simply by an excess of the FITC-labeled protein. Although FITC-labeled GST-IB does not efficiently accumulate in the nucleus in the absence of leptomycin B (Fig. 9A), addition of a 100-fold molar excess of unlabeled GST-IB resulted in a significant nuclear accumulation of FITC-labeled GST-IB (Fig. 9B), consistent with the notion that nuclear export of IB requires one or more rate-limiting factors that can be titrated out with an excess of IB protein.

View larger version (59K): [in this window] [in a new window]   FIG. 9.   High capacity of the IB nuclear import pathway. Nuclear import reactions with FITC-GST-IB were performed in the absence (A and C) or presence (B and D) of 10 nM leptomycin B (LMB). The import reactions were performed in the absence (A and B) or presence (C and D) of a 100-fold molar excess of unlabeled GST-IB. The cellular localization of the import substrates was determined by direct fluorescence.

Nuclear import of IB is independent of GTP hydrolysis by Ran. The directionality of classical nuclear import and nuclear export pathways is determined by the differential distribution of the GTP-bound and GDP-bound forms of Ran between the nucleus and the cytoplasm (21). To determine if nuclear shuttling of IB is sensitive to perturbation of the asymmetric Ran-nucleotide gradient, nuclear accumulation of GST-IB was determined in the presence of the dominant-negative mutant RanQ69L protein bound to GTP. RanQ69L-GTP blocked nuclear import of GST-NLS (compare Fig. 10G and H) and GST-M9 (data not shown). In contrast, RanQ69L-GTP did not inhibit nuclear import of GST-IB in the presence of leptomycin B (compare Fig. 10C and D) or of GST-IB-N69 in the absence of leptomycin B (compare Fig. 10E and F). Consistent with the results obtained with the RanQ69L protein, addition of the wild-type Ran protein preloaded with GTPS inhibited nuclear import of GST-NLS but did not inhibit nuclear import of either GST-IB or GST-IB-N69 (data not shown). Furthermore, while addition of GTPS or 5'-guanylylimdodiphosphate (GMP-PNP) completely inhibited nuclear import of GST-NLS and GST-M9 (data not shown), neither GTPS nor GMP-PNP blocked nuclear import of GST-IB or GST-IB-N69 (data not shown).

View larger version (87K): [in this window] [in a new window]   FIG. 10.   Nuclear import of IB is not inhibited by a dominant-negative Ran protein. Nuclear import reactions with the indicated substrate proteins were performed in the absence (A, B, and E to H) or presence (C and D) of 10 nM leptomycin B (LMB). Parallel import reactions were performed in the absence (A, C, E, and G) or presence (B, D, F, and H) of 5 µg of the RanQ69L protein preloaded with GTP. Localization of the import substrates was determined by indirect immunofluorescence with a monoclonal antibody directed against GST.

View larger version (41K): [in this window] [in a new window]   FIG. 11.   RanQ69L-GTP inhibits nuclear export of IB through sequestration of RanBP1. Nuclear import reactions were performed with 2.5 µg of FITC-GST-IB in the absence of leptomycin B (LMB). RanQ69L-GTP (5 µg) was added to the import reactions shown in panels B to D, and RanBP1 (15 µg) was added to the import reaction shown in panel D. Localization of FITC-GST-IB was determined by direct fluorescence with confocal laser scanning microscopy. A single z-section of representative cells is shown in each panel.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

In the present study, we have utilized both in vitro and in vivo approaches to understand how IB travels between the cytoplasm and the nucleus. The results of our experiments lead to three important conclusions: first, that IB rapidly shuttles between the cytoplasm and the nucleus; second, that IB contains multiple domains that specify nuclear import and nuclear export; and third, that nuclear import of IB is mediated by an import pathway that is mechanistically distinct from classical nuclear import pathways.

Nuclear shuttling of IB. We find that leptomycin B markedly increased nuclear accumulation of IB in both digitonin-permeabilized cells and in transiently transfected HeLa cells. Our finding that nuclear accumulation of IB was increased by leptomycin B in digitonin-permeabilized cells is in apparent contrast to a recent report in which nuclear accumulation of IB in digitonin-permeabilized cells was observed in the absence of leptomycin B treatment (46). However, a careful titration of input FITC-GST-IB protein revealed that leptomycin B-independent nuclear accumulation of IB was observed at high levels of input protein, suggesting that one or more limiting export factors were titrated out by an excess of IB. It has been established that rate-limiting nuclear export factors can be lost during permeabilization with digitonin (10, 26). It is likely that subtle differences in the conditions used for permeabilization resulted in more complete loss of CRM1 in the experiments reported by Turpin et al. (46), such that their digitonin-permeabilized cells were no longer competent for nuclear export of IB.

IB contains multiple cis-acting sequences for nuclear import and export. The distribution of IB between the nucleus and the cytoplasm can be altered by specific mutations with IB (24, 42). For example, we have previously demonstrated that mutations within the second ankyrin repeat of IB resulted in a marked relocalization of ectopically expressed IB from the nucleus to the cytoplasm (42). Consistent with this result, we find that the integrity of the second ankyrin repeat is required for nuclear import of IB in digitonin-permeabilized cells. However, mutation or deletion of the second ankyrin repeat does not completely abolish nuclear import of IB, because leptomycin B increased nuclear accumulation of the GFP-IB-Ank2 protein in transfected HeLa cells. Likewise, Hope and coworkers found that GFP-IB constructs lacking the second ankyrin repeat were able to accumulate in the nucleus in a leptomycin B-dependent manner (24). In our previous work, we found that the second ankyrin repeat was able to direct nuclear localization of a heterologous cytoplasmic protein (42). The other ankyrin repeats of IB also possess this nuclear import function (42). Taken together, the data support a model in which IB contains multiple cis-acting nuclear import sequences within its ARD. Because the steady-state distribution of IB between the nucleus and the cytoplasm is the consequence of rapid shuttling between the nucleus and the cytoplasm, mutation of any individual nuclear import sequence will decrease the rate of nuclear import of IB, resulting in a net increase in cytoplasmic IB.