Nuclear protein import: specificity for transport across the nuclear pore

Transport of proteins into the cell nucleus is thought to require specific localization sequences and may be mediated by nuclear pores. Following microinjection into fused cultured cells, nuclear protein import was directly monitored by fluorescence microscopy using B-phycoerythrin (PE; Mr 240,000) coupled to synthetic peptides corresponding to the simian virus 40 (SV-40) large T antigen nuclear localization signal. Peptides with a single amino acid replacement found in a cytoplasmic mutant of T antigen (cT) failed to promote uptake. Further studies with deletion peptides revealed the minimum sequence requirements for efficient nuclear import of PE conjugates to be similar to those previously defined genetically for large T antigen itself. No competitive inhibition of uptake was observed in cells expressing nuclear or cytoplasmic T antigen. Nuclear import was time- and temperature-dependent. The lectin wheat germ agglutinin (WGA) binds to glycoproteins bearing O-linked GlcNAc on the cytoplasmic face of the nuclear pore in vitro [J.A. Hanover et al. (1987) J. Biol. Chem. 262, 9887-9894] and in vivo. Microinjection of WGA into the cytoplasm of living cells did not alter the diffusion of dextran (Mr 10,000) into the nucleus, but blocked the uptake of PE conjugates. This inhibition was reversed when a competing saccharide was introduced into the cytoplasm.     

Synthetic peptides containing a region of SV40 large T-antigen involved in nuclear localization direct the transport of proteins into the nucleus

We studied the mechanism of transport of proteins into the nucleus using synthetic peptides containing the nuclear location signal sequence of Simian virus 40 (SV 40) large T-antigen. When chick erythrocytes containing a synthetic large T-antigen nuclear translocation signal were fused with SV 40-transformed human fibroblasts, the migration of native large T-antigen into the chick nuclei was suppressed. Migration of proteins detected by human specific antinuclear autoimmune antibody was not blocked. An analog of the nuclear location signal peptide did not inhibit entry of large T-antigen into the chick nuclei. This result suggests that the peptide blocked the migration of only native large T-antigen into the nucleus, and that the signal of the majority of nuclear proteins for nuclear transport is not the same as that of the large T-antigen. The synthetic peptide was conjugated chemically with bovine serum albumin (BSA) and introduced into the cytoplasm of cultured human cells by red blood cell ghost-mediated microinjection. The BSA-synthetic peptide conjugate was found predominantly in the nucleus within 2 h after its introduction into the cells. BSA conjugated with the cross-linking reagent alone was not transported into the nucleus. Acetylated synthetic peptide was not effective in promoting nuclear localization of BSA. Mild trypsin treatment of the BSA-synthetic peptide conjugate suppressed nuclear localization. Conjugates of the synthetic peptide with phycoerythrin (Mr about 150 kD) and with secretory IgA (Mr about 380 kD) were both found in the nucleus very shortly after their introduction into the cytoplasm. These results suggest that the synthetic peptide containing the nuclear location signal sequence provides exogenous proteins with the ability to migrate into the nucleus. But, since a conjugate of the synthetic peptide with IgM (Mr about 940 kD) did not migrate into the nucleus after its microinjection, there may be a size limit in nuclear transport of conjugated proteins.     

Inhibition of Nuclear Protein Import by a Monoclonal Antibody against a Novel Class of Nuclear Pore Proteins

Nuclear import of proteins is mediated by the nuclear pore complexes in the nuclear envelope and requires the presence of a nuclear localization signal (NLS) on the karyophilic protein. In this paper, we describe studies with a monoclonal antibody, Mab E2, which recognizes a class of nuclear pore proteins of 60-76 kDa with a common phosphorylated epitope on rat nuclear envelopes. The Mab E2-reactive proteins fractionated with the relatively insoluble pore complex-containing component of the envelope and gave a finely punctate pattern of nuclear staining in immunofluorescence assays. The antibody did not bind to any cytosolic proteins. Mab E2 inhibited the interaction of a simian virus 40 large T antigen NLS peptide with a specific 60-kDa NLS-binding protein from rat nuclear envelopes in photoaffinity labeling experiments. The antibody blocked the nuclear import of NLS--albumin conjugates in an in vitro nuclear transport assay with digitonin-permeabilized cells, but did not affect passive diffusion of a small non-nuclear protein, lysozyme, across the pore. Mab E2 may inhibit protein transport by directly interacting with the 60-kDa NLS-binding protein, thereby blocking signal-mediated nuclear import across the nuclear pore complex.     

Translocation of beta-catenin into the nucleus independent of interactions with FG-rich nucleoporins

beta-Catenin nuclear import has been found to be independent of classical nuclear localization signal (NLS) nuclear import factors. Here, we test the hypothesis that beta-catenin interacts directly with nuclear pore proteins to mediate its own transport. We show that beta-catenin, unlike importin-beta, does not interact detectably with Phe/Gly(FG)-repeat-rich nuclear pore proteins or nucleoporins (Nups). Moreover, unlike NLS-containing proteins, beta-catenin nuclear import is not inhibited by wheat germ agglutinin (WGA) or excess importin-beta. These results suggest beta-catenin nuclear translocation does not involve direct interactions with FG-Nups. However, beta-catenin has two regions that can target it to the nucleus, and its import is cold sensitive, indicating that beta-catenin nuclear import is still an active process. Transport is blocked by a soluble form of the C-cadherin cytoplasmic domain, suggesting that masking of the nuclear targeting signal may be a mechanism of regulating beta-catenin subcellular localization.     

Reconstitution of nuclear protein transport with semi-intact yeast cells

We have developed an in vitro nuclear protein import reaction from semi- intact yeast cells. The reaction uses cells that have been permeabilized by freeze-thaw after spheroplast formation. Electron microscopic analysis and antibody-binding experiments show that the nuclear envelope remains intact but the plasma membrane is perforated. In the presence of ATP and cytosol derived from yeast or mammalian cells, a protein containing the nuclear localization sequence (NLS) of SV40 large T-antigen is transported into the nucleus. Proteins with mutant NLSs are not imported. In the absence of cytosol, binding of NLS- containing proteins occurs at the nuclear envelope. N-ethylmaleimide treatment of the cytosol as well as antibodies to the nuclear pore protein Nsp1 inhibit import but not binding to the nuclear envelope. Yeast mutants defective in nuclear protein transport were tested in the in vitro import reaction. Semi-intact cells from temperature-sensitive nsp1 mutants failed to import but some binding to the nuclear envelope was observed. On the other hand, no binding and thus no import into nuclei was observed in semi-intact nsp49 cells which are mutated in another nuclear pore protein. Np13 mutants, which are defective for nuclear protein import in vivo, were also deficient in the binding step under the in vitro conditions. Thus, the transport defect in these mutants is at the level of the nucleus and the point at which nuclear transport is blocked can be defined.     

Spatial changes in calcium signaling during the establishment of neuronal polarity and synaptogenesis

Calmodulin (CaM) potentiates Ca(2+)-dependent signaling pathways in both the cytoplasm and nucleus. We have investigated the mechanism of CaM nuclear transport using tissue culture cell microinjection and a permeabilized cell import assay. The inhibition of CaM import by the translocation inhibitor wheat germ agglutinin (WGA) and by chilling, indicates that CaM import is facilitated, but because ATP depletion does not affect CaM import, the mechanism does not appear to be active. Chilling and WGA arrest persist in ATP-depleted cells, indicating that CaM is not retained in the cytoplasm by an ATP-dependent mechanism. In permeabilized cells, both Ca(2+)-CaM and Ca(2+)-free CaM are sensitive to extract-dependent WGA and chilling import inhibition. Titration experiments in microinjected and permeabilized cells indicate that a saturable cytosolic factor(s) mediates chilling and WGA arrest.     

Evidence for existence of a nuclear pore complex-mediated, cytosol-independent pathway of nuclear translocation of ERK MAP kinase in permeabilized cells

The classical mitogen-activated protein kinase (MAPK, also known as ERK) pathway is widely involved in eukaryotic signal transductions. In response to extracellular stimuli, MAPK becomes activated and translocates from the cytoplasm to the nucleus. At least two pathways for the nuclear import of MAPK are shown to exist; passive diffusion of a monomer and Ran-dependent active transport of a dimer, the detailed molecular mechanism of which is unknown. In this study, we have reconstituted nuclear import of MAPK in vitro by using digitonin-permeabilized cells with GFP-fused MAPK protein (GFP-MAPK), which is too large to pass through the nuclear pore by passive diffusion. GFP-MAPK was able to accumulate in the nucleus irrespective of its phosphorylation state. This import of GFP-MAPK occurred even in the absence of any soluble cytosolic factors or ATP but was inhibited by wheat germ agglutinin or an excess amount of importin-beta or at low temperatures. Moreover, MAPK directly bound to an FG repeat region of nucleoporin CAN/Nup214 in vitro. Taken together, these results suggest the third pathway for nuclear import of MAPK, in which MAPK passes through the nuclear pore by directly interacting with the nuclear pore complex.     

Depletion of calcium from the lumen of endoplasmic reticulum reversibly inhibits passive diffusion and signal-mediated transport into the nucleus

Nuclear pore complexes provide channels for molecular transport across the nuclear envelope. Translocation of most proteins and RNAs through the pore complex is mediated by signal- and ATP-dependent mechanisms, while transport of small molecules is accomplished by passive diffusion. We report here that depletion of calcium from the lumen of the endoplasmic reticulum and nuclear envelope with ionophores or the calcium pump inhibitor thapsigargin rapidly and potently inhibits signal mediated transport of proteins into the nucleus. Lumenal calcium depletion also inhibits passive diffusion through the pore complex. Signal-mediated protein import and passive diffusion are rapidly restored when the drugs depleting lumenal calcium are removed and cells are incubated at 37 degrees C in calcium-containing medium. These results indicate that loss of calcium from the lumen of the endoplasmic reticulum and nuclear envelope reversibly affects properties of pore complex components located on the nuclear/cytoplasmic membrane surfaces, and they provide direct functional evidence for conformational flexibility of the pore complex. These methods will be useful for achieving reversible inhibition of nucleocytoplasmic trafficking for in vivo functional studies, and for studying the structure of the passive diffusion channel(s) of the pore complex.     

Antibodies against 70-kD heat shock cognate protein inhibit mediated nuclear import of karyophilic proteins

Previously, we found that anti-DDDED antibodies strongly inhibited in vivo nuclear transport of nuclear proteins and that these antibodies recognized a protein of 69 kD (p69) from rat liver nuclear envelopes that showed specific binding activities to the nuclear location sequences (NLSs) of nucleoplasmin and SV-40 large T-antigen. Here we identified this protein as the 70-kD heat shock cognate protein (hsc70) based on its mass, isoelectric point, cellular localization, and partial amino acid sequences. Competition studies indicated that the recombinant hsc70 expressed in Escherichia coli binds to transport competent SV-40 T-antigen NLS more strongly than to the point mutated transport incompetent mutant NLS. To investigate the possible involvement of hsc70 in nuclear transport, we examined the effect of anti-hsc70 rabbit antibodies on the nuclear accumulation of karyophilic proteins. When injected into the cytoplasm of tissue culture cells, anti-hsc70 strongly inhibited the nuclear import of nucleoplasmin, SV-40 T-antigen NLS bearing BSA and histone H1. In contrast, anti-hsc70 IgG did not prevent the diffusion of lysozyme or 17.4-kD FITC-dextran into the nuclei. After injection of these antibodies, cells continued RNA synthesis and were viable. These results indicate that hsc70 interacts with NLS-containing proteins in the cytoplasm before their nuclear import.     

Formation of the Nuclear Envelope Permeability Barrier Studied by Sequential Photoswitching and Flux Analysis

In higher eukaryotes, the nuclear envelope breaks down during mitosis. It reforms during telophase, and nuclear import is reestablished within <10 min after anaphase onset. It is widely assumed that import functionality simultaneously leads to the exclusion of bulk cytoplasmic proteins. However, nuclear pore complex assembly is not fully completed when import capacity is regained, which raises the question of whether the transport and permeability barrier functions of the nuclear envelope are indeed coupled. In this study, we therefore analyzed the reestablishment of the permeability barrier of the nuclear envelope after mitosis in living cells by monitoring the flux of the reversibly photoswitchable fluorescent protein Dronpa from the cytoplasm into the nucleus after photoactivation. We performed many consecutive flux measurements in the same cell to directly monitor changes in nuclear envelope permeability. Our measurements at different time points after mitosis in individual cells show that contrary to the general view and despite the rapid reestablishment of facilitated nuclear import, the nuclear envelope remains relatively permeable for passive diffusion for the first 2 h after mitosis. Our data demonstrate that reformation of the permeability barrier of nuclear pore complexes occurs only gradually and is uncoupled from regaining active import functionality.     

Role of nuclear pore complex in simian virus 40 nuclear targeting.

Cytoplasmically injected simian virus 40 (SV40) virions enter the nucleus through nuclear pore complexes (NPCs) and can express large T antigen shortly thereafter (J. Clever, M. Yamada, and H. Kasamatsu, Proc. Natl. Acad. Sci. USA 88:7333-7337, 1991). The nuclear import of the protein components of introduced SV40 was reversibly arrested by chilling and energy depletion, corroborating our previous observation that the nuclear entry of injected SV40 is blocked in the presence of wheat germ agglutinin and an antinucleoporin monoclonal antibody (mAb414), general inhibitors of NPC-mediated import. The nuclear accumulation of virion protein components and large T antigen in nonpermissive NIH 3T3 cells was similar to that in the permissive host, indicating that the ability to use NPCs as a route of nuclear entry appears to be a general property of the injected virus. Injected virions were capable of completing their lytic cycle and forming plaques in permissive cells. During the early phase of SV40 infection, the cytoplasmic injection of mAb414 effectively blocked nuclear T-antigen accumulation for up to 8 h of infection but had very little effect after 12 h of infection. The time-dependent interference with nuclear T-antigen accumulation by the antinucleoporin antibody is consistent with the hypothesis that the infecting virions enter the nucleus through NPCs. The interference study also suggests that the early phase of infection consists of at least two steps: a step for virion cell entry and intracytoplasmic trafficking and a step for virion nuclear entry followed by large-T-antigen gene expression and subsequent nuclear localization of the gene product. Virions were visualized as electron-dense particles in ultrathin sections of samples in which transport was permitted or arrested. In the former cells, electron-dense particles were predominantly observed in the nucleus. The virions were distributed randomly and nonuniformly in the nucleoplasm but were not observed in heterochromatin or in nucleoli. In the latter cells, the electron-dense particles were seen intersecting the nuclear envelope, near the inner nuclear membrane, and in NPCs. In tangential cross sections of NPCs, which appeared as donut-shaped structures, a spherical electron-dense particle was observed in the center of the structure. Immunoelectron microscopy revealed that NPCs were selectively decorated with 5-nm colloidal gold particles-anti-Vp1 immunoglobulin G at the cytoplasmic entrance to and in NPCs, confirming that the morphologically observed electron-dense particles in NPCs contain the viral structural protein. These results support the hypothesis that the nuclear import of SV40 is catalyzed through NPCs by an active transport mechanism that is similar to that of other karyophiles.     

Functional role of newly formed pore complexes in postmitotic nuclear reorganization

Many nuclear proteins are released into the cytoplasm at prometaphase and are transported back into the daughter nuclei at the end of mitosis. To determine the role of this reentry in nuclear remodelling during early interphase, we experimentally manipulated nuclear protein uptake in dividing cells. Recently we and others have shown that signal-dependent, pore complex-mediated uptake of nuclear protein is blocked in living cells on microinjection of the lectin wheat germ agglutinin (WGA), or of antibodies such as PI1 that are directed against WGA-binding pore complex glycoproteins. In the present study, we microinjected mitotic PtK₂ cells with WGA or antibody PI1 and followed nuclear reorganization of the daughter cells by immunofluorescence and electron microscopy. The inhibitory effect on nuclear protein uptake was monitored by co-injection of the karyophilic protein nucleoplasmin. When injected by itself early in mitosis, nucleoplasmin became sequestered into the daughter nuclei as they entered telophase. In contrast, nucleoplasmin was excluded from the daughter nuclei in the presence of WGA or antibody PI1. Although PtK₂ cells with blocked nuclear protein uptake completed cytokinesis, their nuclei showed a telophaselike organization characterized by highly condensed chromatin surrounded by a nuclear envelope containing a few pore complexes. These findings suggest that pore complexes become functional as early as telophase, in close coincidence with nuclear envelope reformation. They further indicate that the extensive structural rearrangement of the nucleus during the telophase-G1 transition is dependent on the influx of karyophilic proteins from the cytoplasm through the pore complexes, and is not due solely to chromosome-associated components.Abbreviations WGA wheat germ agglutinin - GlcNAc N-acetylglucosamine     

Inhibition of nuclear accumulation of karyophilic proteins in living cells by microinjection of the lectin wheat germ agglutinin

The lectin wheat germ agglutinin (WGA), which has been reported to inhibit nuclear protein uptake in vitro by isolated nuclei (Finlay et al. (1987) J. Cell Biol. 104, 189), also blocks, on microinjection into living cells, the migration of proteins into the cell nucleus. Radioactively labeled nuclear proteins were injected into the cytoplasm of Xenopus oocytes and their reentry into the nucleus was analyzed in the presence or absence of WGA by two-dimensional gel electrophoresis. In another set of experiments, fluorescently labeled nucleoplasmin was injected, alone or together with WGA, into the cytoplasm of rat hepatoma cells, and its nucleocytoplasmic distribution was studied by quantitative laser fluorescence microscopy. The results indicate that WGA inhibits the uptake of karyophilic proteins in general, independent of their sizes. Since the nucleocytoplasmic flux of a dextran with Mr 10,000 was not affected it can be excluded that WGA acts by a general blockade or constriction of the functional pore channel. At reduced WGA concentrations, the rate but not the final extent of nuclear protein accumulation was decreased. These findings support the concept that the O-glycosidically bound carbohydrates of certain nuclear pore complex proteins are exposed to the pore interior and that these regions are probably involved in nucleocytoplasmic translocation processes.     

Shigella protein IpaH(9.8) is secreted from bacteria within mammalian cells and transported to the nucleus.

Various pathogenic bacteria such as Shigella deliver effector proteins into mammalian cells via the type III secretion system. The delivered Shigella effectors have been shown to variously affect host functions required for efficient bacterial internalization into the cells. In the present study, we investigated the IpaH proteins for their ability to be secreted via the type III secretion system and their fate in mammalian cells. Upon incubation in a medium containing Congo red, the bacteria secrete IpaH into the medium, but secretion of IpaH occurs later than that of IpaBCD. Immunofluorescence microscopy indicated that IpaH(9.8) is secreted from intracellular bacteria and transported into the nucleus. On microinjection of the protein, intracellular IpaH(9.8) is accumulated at one place around the nucleus and transported into the nucleus. This movement seems to be dependent on the microtubule network, since nuclear accumulation of IpaH(9.8) is inhibited in cells treated with microtubule-destabilizing agents. In nuclear import assay, IpaH(9.8) was efficiently transported into the nucleus, which was completely blocked by treatment with wheat germ agglutinin. The nuclear transport of IpaH(9.8) does not depend on host cytosolic factors but is partially dependent on ATP/GTP, suggesting that, like beta-catenin, IpaH(9.8) secreted from intracellular Shigella can be transported into the nucleus.     

Inhibition of in vitro nuclear transport by a lectin that binds to nuclear pores

Selective transport of proteins is a major mechanism by which biochemical differences are maintained between the cytoplasm and nucleus. To begin to investigate the molecular mechanism of nuclear transport, we used an in vitro transport system composed of a Xenopus egg extract, rat liver nuclei, and a fluorescently labeled nuclear protein, nucleoplasmin. With this system, we screened for inhibitors of transport. We found that the lectin, wheat germ agglutinin (WGA), completely inhibits the nuclear transport of fluorescently labeled nucleoplasmin. No other lectin tested affected nuclear transport. The inhibition by WGA was not seen when N-acetylglucosamine was present and was reversible by subsequent addition of sugar. When rat liver nuclei that had been incubated with ferritin-labeled WGA were examined by electron microscopy, multiple molecules of WGA were found bound to the cytoplasmic face of each nuclear pore. Gel electrophoresis and nitrocellulose transfer identified one major and several minor nuclear protein bands as binding 125I-labeled WGA. The most abundant protein of these, a 63-65-kD glycoprotein, is a candidate for the inhibitory site of action of WGA on nuclear protein transport. WGA is the first identified inhibitor of nuclear protein transport and interacts directly with the nuclear pore.     

A long synthetic peptide containing a nuclear localization signal and its flanking sequences of SV40 T-antigen directs the transport of IgM into the nucleus efficiently

Synthetic short peptides containing only the nuclear localization signal (NLS) direct the transport of nonnuclear proteins into the nucleus. As a conjugate of the synthetic peptide with immunoglobulin M (IgM) did not enter the nucleus, there was believed to be a size limit for nuclear transport of NLS-conjugated proteins. However, we found that IgM conjugated with purified nucleoplasmin, a nuclear protein of Xenopus oocytes, rapidly accumulated in the nucleus. For direct comparison with the short peptide, we prepared a long peptide containing the NLS and its flanking sequences of SV40 large T-antigen and its mutated long peptide, in which possible phosphorylation sites located at the amino terminal of the NLS were changed to alanine. Kinetic experiments showed that wild-type long peptide-IgM conjugates were almost entirely taken up into the nucleus within 30 min after their injection, whereas almost 60 min was required for the mutated long peptide-IgM conjugates to enter the nucleus of all the cells examined, and there was no apparent accumulation of short peptide-IgM conjugates in the nucleus within 60 min. These results indicate that even when the kinetics of transport are affected by amino acid substitutions, the long peptide directs the transport of large molecules such as IgM into the nucleus.     

Novel properties of the nucleolar targeting signal of human angiogenin

The polypeptide ligand angiogenin, a potent inducer of angiogenesis, localizes in the nucleus/nucleolus subsequent to endocytosis by relevant cell types. This study examines the kinetic properties of the nucleolar targeting signal (NTS) of angiogenin (IMRRRGL(35)) at the single cell level. We show that the NTS is sufficient to target green fluorescent protein (GFP), but not beta-galactosidase, to the nucleolus of rat hepatoma cells. Mutation of Arg(33) to Ala within the NTS abolishes targeting activity. Nuclear/nucleolar import conferred by the NTS of angiogenin is reduced by cytosolic factors as well as ATP and is independent of importins and Ran. The NTS also confers the ability to bind to nuclear/nucleolar components which is inhibited by ATP hydrolysis; nonhydrolysable GTP analogs prevent nuclear accumulation in the absence of an intact nuclear envelope through an apparent cytoplasmic retention mechanism. Since the lectin wheat germ agglutinin does not inhibit transport, we postulate a mechanism for angiogenin nuclear/nucleolar import involving passive diffusion of angiogenin through the nuclear pore and NTS-mediated nuclear/nucleolar retention, and with cytoplasmic retention modulating the process. This pathway is clearly distinct from that of conventional signal-mediated nuclear protein import.     

Two co-existing mechanisms for nuclear import of MAP kinase: passive diffusion of a monomer and active transport of a dimer.

In response to extracellular stimuli, mitogen-activated protein kinase (MAPK, also known as ERK) translocates from the cytoplasm to the nucleus. MAP kinase kinase (MAPKK, also know as MEK), which possesses a nuclear export signal (NES), acts as a cytoplasmic anchor of MAPK. Here we show evidence that tyrosine (Tyr190 in Xenopus MPK1/ERK2) phosphorylation of MAPK by MAPKK is necessary and sufficient for the dissociation of the MAPKK-MAPK complex, and that the dissociation of the complex is required for the nuclear translocation of MAPK. We then show that nuclear entry of MAPK through a nuclear pore occurs via two distinct mechanisms. Nuclear import of wild-type MAPK (mol. wt 42 kDa) was induced by activation of the MAPK pathway even in the presence of wheat germ agglutinin or dominant-negative Ran, whereas nuclear import of beta-galactosidase (beta-gal)-fused MAPK (mol. wt 160 kDa), which occurred in response to stimuli, was completely blocked by these inhibitors. Moreover, while a dimerization-deficient mutant of MAPK was able to translocate to the nucleus upon stimulation, this mutant MAPK, when fused to beta-gal, became unable to enter the nucleus. These results suggest that monomeric and dimeric forms of MAPK enter the nucleus by passive diffusion and active transport mechanisms, respectively.