A Karyophilic Protein Forms a Stable Complex with Cytoplasmic Components Prior to Nuclear Pore Binding

Targeting of karyophilic proteins to nuclear pores is known to require several cytoplasmic factors, including the nuclear location signal-binding protein. Using a digitonin-permeabilized cell-free transport assay, we have obtained a cytoplasmic fraction containing factors that specifically bind to karyophilic protein and support the nuclear binding step of the transport. Components in this fraction form a stable complex with the karyophile through interaction with nuclear location signal. Since this complex shows nuclear pore binding activity prior to nuclear entry in the absence of other cytosolic factors, we call it nuclear pore-targeting complex. It consists of karyophilic protein and four proteins of 54, 56, 66, and 90 kDa. In our reconstitution experiments, a complex with 54 and 90 kDa proteins is capable of targeting karyophiles to the nuclear pores. Targeting of karyophilic proteins to nuclear pores is known to require several cytoplasmic factors, including the nuclear location signal-binding protein. Using a digitonin-permeabilized cell-free transport assay, we have obtained a cytoplasmic fraction containing factors that specifically bind to karyophilic protein and support the nuclear binding step of the transport. Components in this fraction form a stable complex with the karyophile through interaction with nuclear location signal. Since this complex shows nuclear pore binding activity prior to nuclear entry in the absence of other cytosolic factors, we call it nuclear pore-targeting complex. It consists of karyophilic protein and four proteins of 54, 56, 66, and 90 kDa. In our reconstitution experiments, a complex with 54 and 90 kDa proteins is capable of targeting karyophiles to the nuclear pores. The selective nuclear import of karyophilic proteins is directed by short amino acid sequences termed nuclear location signals (NLSs) 1The abbreviations used are:NLSnuclear location signalWGAwheat germ agglutininAPCallophycocyaninbBSAbiotinylated bovine serum albuminhsc7070-kDa heat-shock cognate proteinRan/TC4Ras-related nuclear protein. (1Silver P.A. Cell. 1991; 64: 489-497Abstract Full Text PDF PubMed Scopus (394) Google Scholar). The process of mediated nuclear import is known to involve at least two steps: ATP-independent binding to the cytoplasmic face of nuclear pores, followed by translocation through the nuclear pore complex dependent on ATP hydrolysis (2Newmeyer D.D. Forbes D.J. Cell. 1988; 52: 641-653Abstract Full Text PDF PubMed Scopus (371) Google Scholar, 3Richardson W.D. Mills A.D. Dilworth S.M. Laskey R.A. Dingwall C. Cell. 1988; 52: 655-664Abstract Full Text PDF PubMed Scopus (374) Google Scholar). Wheat germ agglutinin (WGA), which binds to a family of nuclear pore complex proteins, is known to inhibit the latter step of transport (4Finlay D.R. Newmeyer D.D. Price T.M. Forbes D.J. J. Cell Biol. 1987; 104: 189-200Crossref PubMed Scopus (377) Google Scholar, 5Yoneda Y. Imamoto-Sonobe N. Yamaizumi T. Uchida T. Exp. Cell Res. 1987; 173: 586-595Crossref PubMed Scopus (177) Google Scholar). Physiological evidence has indicated that karyophilic proteins are complexed in the cytoplasm upon active nuclear import (6Breeuwer M. Goldfarb D.S. Cell. 1990; 60: 999-1008Abstract Full Text PDF PubMed Scopus (202) Google Scholar); however, no such complex has yet been identified on a molecular basis. Here we show in vitro evidence that a karyophilic protein actually forms a complex with cytoplasmic factors for nuclear pore targeting in the active nuclear transport process. In view of these results, we propose that the first step of transport can be divided into two steps: formation of the pore-targeting complex in the cytoplasm and subsequent binding of the complex to the nuclear pores. nuclear location signal wheat germ agglutinin allophycocyanin biotinylated bovine serum albumin 70-kDa heat-shock cognate protein Ras-related nuclear protein. PtK2 cells were grown and plated for in vitro assay, as described previously (7Okuno Y. Imamoto N. Yoneda Y. Exp. Cell Res. 1993; 206: 134-142Crossref PubMed Scopus (66) Google Scholar). Synthetic peptides containing SV40 large T antigen wild-type NLS (CYGGPKKKRKVEDP: T-peptide) or transport incompetent point-mutated NLS (CYGGPKTKRKVEDP: mutant T-peptide) were purchased from Peptide Institute (Osaka, Japan), and WGA from E˙Y Laboratories, Inc. Histone H1 purified from calf thymus (8Imamoto N. Matsuoka Y. Kurihara T. Kohno K. Miyagi M. Sakiyama F. Okada Y. Tsunasawa S. Yoneda Y. J. Cell Biol. 1992; 119: 1047-1061Crossref PubMed Scopus (149) Google Scholar) and lysozyme (Sigma) were conjugated to CNBr-activated Sepharose 4B (Pharmacia) by a standard method at a protein concentration of 2 mg/ml gel. For preparation of biotinylated bovine serum albumin (bBSA), 100 μl of 2 mg/ml sulfosuccinimidyl 6-(biotinamido) hexanoate (NHS-LC-biotin) (Pierce) dissolved in distilled water was mixed with 800 μl of 2 mg/ml bovine serum albumin (BSA) (Sigma) dissolved in 0.1 M NaHCO3, pH 8.5, incubated for 1 h at room temperature, and dialyzed against 0.1 M potassium phosphate buffer, pH 7.2. Allophycocyanin (APC) (Calbiochem) and bBSA were conjugated with T-peptide or mutant T-peptide to produce T-APC, T-bBSA, and mutant T-APC as described previously (10Yoneda Y. Arioka T. Imamoto-Sonobe N. Sugawa H. Shimonishi Y. Uchida T. Exp. Cell Res. 1987; 170: 439-452Crossref PubMed Scopus (43) Google Scholar). All the conjugates contained 10-15 peptides per carrier molecule (analyzed by SDS-PAGE). Ehrlich ascites tumor cells were freshly harvested from the abdominal cavity of mice, washed twice in phosphate-buffered saline (137 m M NaCl, 2.7 m M KCl, 8.1 m M Na2HPO4, 1.5 m M KH2PO4, pH 7.2), once in washing buffer (10 m M HEPES, pH 7.3, 110 m M CH3COOK, 2 m M (CH3COO)2Mg, 2 m M dithiothreitol) and then lysed in lysis buffer (5 m M HEPES, pH 7.3, 10 m M CH3COOK, 2 m M (CH3COO)2Mg, 2 m M dithiothreitol, 20 μM cytochalasin B, 1 m M phenylmethylsulfonyl fluoride, 1 μg/ml each aprotinin, leupeptin, pepstatin). Extracts were clarified by sequential centrifugation (1,500 × g for 15 min; 15,000 × g for 20 min; 100,000 × g for 30 min) as described by Adam et al. (11Adam S.A. Sterne Marr R. Gerace L. J. Cell Biol. 1990; 111: 807-816Crossref PubMed Scopus (771) Google Scholar) For total lysate preparation, the clarified extract was concentrated to 20-40 mg/ml protein concentration by vacuum dialysis against transport buffer (20 m M HEPES, pH 7.3, 110 m M CH3COOK, 2 m M (CH3COO)2Mg, 5 m M CH3COONa, 0.5 m M EGTA, 2 m M dithiothreitol, 1 μg/ml each aprotinin, leupeptin, pepstatin) in a collodion apparatus (Schleicher 206: 134-142Crossref PubMed Scopus (66) Google Scholar, 11Adam S.A. Sterne Marr R. Gerace L. J. Cell Biol. 1990; 111: 807-816Crossref PubMed Scopus (771) Google Scholar). 10 μl of testing cytosols containing 100 ng of transport substrates or isolated, and reconstituted complexes were incubated with permeabilized cells for 30 min, with or without 1 m M ATP, 5 m M creatine phosphate, and 20 units/ml creatine phosphokinase at 30°C, or on ice, as described in the respective figure legends. After incubation, cells were fixed with 3.7% formaldehyde in transport buffer. To visualize T-bBSA localization, the fixed cells were permeabilized with 0.5% Triton X-100 in phosphate-buffered saline and incubated with 2 μg/ml fluorescein isothiocyanate-avidin (Pierce) in phosphate-buffered saline containing 10 mg/ml BSA for 1 h at room temperature. The samples were then examined using an Axiophot microscope (Carl Zeiss, Inc.). Staining of nuclear binding was faint and could be observed clearly only with an oil immersion lens. For quantitative analysis, the image was focused on a plane corresponding to a section through the center of the nucleus. Most of the nuclei in each field of view were usually in the same focal plane. Fluorescence intensity was quantified by scanning photographic negatives with a dual wavelength TLC scanner (Shimadzu CS-930). hsc70 was purified from the cytoplasm of Ehrlich ascites tumor cells as described previously (8Imamoto N. Matsuoka Y. Kurihara T. Kohno K. Miyagi M. Sakiyama F. Okada Y. Tsunasawa S. Yoneda Y. J. Cell Biol. 1992; 119: 1047-1061Crossref PubMed Scopus (149) Google Scholar). For immunoblotting, proteins were separated by 10% SDS-PAGE and transferred electrophoretically on to nitrocellulose sheets. Western blots were probed with anti-hsc70 rabbit serum (8Imamoto N. Matsuoka Y. Kurihara T. Kohno K. Miyagi M. Sakiyama F. Okada Y. Tsunasawa S. Yoneda Y. J. Cell Biol. 1992; 119: 1047-1061Crossref PubMed Scopus (149) Google Scholar) or anti-Ran/TC4 rabbit serum (9Tachibana T. Imamoto N. Seino H. Nishimoto T. Yoneda Y. J. Biol. Chem. 1994; 269: 24542-24545Abstract Full Text PDF PubMed Google Scholar) in TBS buffer (20 m M Tris-HCl, pH 7.5, 0.3 M NaCl) containing 1% skim milk for 2 h at room temperature after blocking with TBS buffer containing 3% skim milk for 4 h. Rabbit antibodies were detected with alkaline phosphatase-conjugated goat antibodies to rabbit IgG (Bio-Rad) by the standard method. Previously we showed that cytoplasmic injection of antibodies against 70-kDa heat-shock cognate protein (hsc70) inhibited nuclear import in living cells (8Imamoto N. Matsuoka Y. Kurihara T. Kohno K. Miyagi M. Sakiyama F. Okada Y. Tsunasawa S. Yoneda Y. J. Cell Biol. 1992; 119: 1047-1061Crossref PubMed Scopus (149) Google Scholar). Using a digitonin-permeabilized cell-free transport assay, we confirmed that hsc70 is a necessary but not sufficient cytoplasmic factor to support active nuclear import of SV40 T-antigen NLS-containing karyophile (7Okuno Y. Imamoto N. Yoneda Y. Exp. Cell Res. 1993; 206: 134-142Crossref PubMed Scopus (66) Google Scholar). Therefore, in this study, we attempted to identify other cytoplasmic factors required for nuclear import, initially by fractionating cytosolic extracts in two distinct manners, and examining their ability to support the nuclear import of SV40 T-antigen NLS (T-peptide) conjugates (T-APC and T-BSA) in a digitonin-permeabilized cell-free transport assay. As one of two fractionation procedures, total cytosol prepared from Ehrlich ascites tumor cells was fractionated by Q-Sepharose ion exchange chromatography as described under “Materials and Methods.” Each fraction separated by Q-Sepharose, as well as hsc70 alone, supported little or no nuclear import by itself (Fig. 1 A, b-f). However, combination of the two fractions eluted with 200 m M KCl and 550 m M KCl, termed Q200 and Q550, respectively, reconstituted the nuclear import of T-peptide conjugates (Fig. 1 A, g). Addition of hsc70 alone to Q200 or Q550, or other combinations of two separated fractions, did not reconstitute the import (data not shown). The other procedure was based on previous evidence which indicated that a small basic karyophilic protein, histone H1, is complexed in the cytoplasm, and therefore accumulates in the nucleus via mediated import pathway as do SV40 T-antigen NLS-containing karyophiles, whereas a small basic non-karyophilic protein, lysozyme, enters the nucleus by passive diffusion when injected into the cytoplasm of living cells (6Breeuwer M. Goldfarb D.S. Cell. 1990; 60: 999-1008Abstract Full Text PDF PubMed Scopus (202) Google Scholar). Our previous evidence indicated that hsc70 is involved in nuclear import of both histone H1 and T-peptide conjugates (8Imamoto N. Matsuoka Y. Kurihara T. Kohno K. Miyagi M. Sakiyama F. Okada Y. Tsunasawa S. Yoneda Y. J. Cell Biol. 1992; 119: 1047-1061Crossref PubMed Scopus (149) Google Scholar). Here, we found that incubation of total cytosol with histone H1-conjugated Sepharose resulted in depletion of transport activity for T-peptide conjugates, whereas incubation with lysozyme-conjugated Sepharose did not (Fig. 1 B, a-c). Among hsc70 alone and the Q-Sepharose fractions, the addition of fraction Q550 to the H1-depleted cytosol restored the transport of T-peptide conjugates (Fig. 1 B, h). Pretreatment of permeabilized cells with WGA or treatment of Q550 with N-ethylmaleimide (NEM) inhibited the transport of T-peptide conjugates in all these reconstitution experiments (data not shown). With incubation on ice in the absence of ATP, the transport substrate accumulated at the nuclear rim in the presence of unfractionated cytosol. Q550, alone or supplemented with H1-depleted cytosol, yielded the same rim staining (Fig. 1 C, a, c, and d). No such rim staining was observed in the presence of H1-depleted cytosol alone, other fractions eluted from Q-Sepharose, or transport buffer. In addition, mutant T-peptide conjugates, with or without Q550, failed to yield the same rim staining. Thus, it appears that rim staining obtained with Q550 shows the ATP-independent and NLS-dependent nuclear binding step of the transport. These results suggest that fraction Q550 contains factors that bind to karyophilic protein and support the nuclear pore binding step of transport. As shown in Fig. 2 A, upon incubation, T-BSA formed a complex in Q550 that was detectable by sucrose gradient analysis. No other Q-Sepharose fractions caused a shift in the T-BSA, indicating that complex formation occurred specifically in Q550. The complex formation in Q550 was markedly reduced in the presence of free but to a by mutant that complex formation was inhibited by active 5 × to 1 × of free T-peptide over T-peptide in T-BSA was required for the The for concentration of free T-peptide for be to the interaction of the T-peptide with of free T-peptide by free proteins in Q550, which be in over To the of this complex, we the complex by Superdex TM200 HR 10/30 and examined nuclear binding activity. As shown in Fig. 2 B, a protein complex of kDa containing a transport substrate in fraction showed nuclear rim binding activity in the absence of other cytoplasmic fractions. To such rim staining actually shows the nuclear binding step of transport prior to nuclear we examined transport in the presence of ATP at As shown in Fig. 2 C, the transport substrate in the complex in fraction showed only nuclear rim in the presence of In the addition of histone H1-depleted cytosol to fraction supported nuclear entry of the transport substrate by nuclear only in the presence of rim nuclear staining obtained in the presence of the H1-depleted fraction was observed upon of permeabilized cells with whereas rim staining obtained in the absence of H1-depleted cytosol was only inhibited Fig. a, c, and The of fraction Q200 of histone H1-depleted cytosol the same results (data not shown). These results showed that in Q550, a T-peptide forms a stable complex which binds to the nuclear pore with no other cytoplasmic fractions and enters the nucleus dependent on in both H1-depleted cytosol and Q200 and on ATP we this complex nuclear pore-targeting complex. To and identify of the complex, BSA in immobilized avidin was incubated with Q550. proteins of molecular 54, 56, 66, and 90 kDa were eluted with after extensive washing with transport buffer. All four proteins specifically from T-bBSA in the presence of free T-peptide (data not shown). showed that eluted materials did not both hsc70 and Q550 contained these two (Fig. 3 To these four proteins actually the nuclear pore-targeting complex, T-bBSA was mixed with eluted materials and the was dialyzed and examined for targeting activity. As shown in Fig. and C, T-bBSA reconstituted a complex with the four proteins and showed nuclear binding activity in the absence of other the of this study, we found that Q550 the histone H1-depleted cytosol. Therefore, the four proteins which reconstituted the protein complex with T-peptide conjugates were to be by histone eluted with from were found to reconstitute the targeting activity with T-bBSA as shown in Fig. we confirmed that four proteins of molecular 54, 56, 66, and 90 kDa in the eluted materials were to T-bBSA after reconstitution of the targeting activity (Fig. 4 c, These results show that histone all of the targeting complex from total cytosol. To the targeting activity of the complex on the presence of all four proteins, the proteins were separated by gradient from T-bBSA in immobilized with molecular of 54 and 90 kDa were eluted the and of the targeting activity of protein complexes reconstituted with fractions containing either or proteins that the targeting activity on the presence of and proteins (Fig. indicating that these proteins can reconstitute the targeting activity. In this study, we a nuclear pore-targeting complex from a fraction Q550. Q550 was and supported only the binding step of transport. The addition of histone H1-depleted cytosol or Q200 was required for transport activity. of fraction Q550 of fraction A separated from extracts by and Cell. 1992; Full Text PDF PubMed Scopus Google Scholar). To NLS purified from bovine S.A. Gerace L. Cell. 1991; Full Text PDF PubMed Scopus Google hsc70 (7Okuno Y. Imamoto N. Yoneda Y. Exp. Cell Res. 1993; 206: 134-142Crossref PubMed Scopus (66) Google Scholar, N. Matsuoka Y. Kurihara T. Kohno K. Miyagi M. Sakiyama F. Okada Y. Tsunasawa S. Yoneda Y. J. Cell Biol. 1992; 119: 1047-1061Crossref PubMed Scopus (149) Google Scholar, Y. Cell. Biol. 1992; PubMed Scopus Google and (9Tachibana T. Imamoto N. Seino H. Nishimoto T. Yoneda Y. J. Biol. Chem. 1994; 269: 24542-24545Abstract Full Text PDF PubMed Google Scholar, 1993; PubMed Scopus Google Scholar, F. J. Gerace L. J. Cell Biol. 1993; PubMed Scopus Google Scholar) have been identified as factors involved in nuclear protein Adam and Adam Adam S.A. J. Cell Biol. 1994; PubMed Scopus Google Scholar) the and of bovine cytosolic kDa protein required for the binding step of transport and that both NLS and protein are necessary for the binding from and molecular and proteins identified in this be of bovine and the other Q550 contained hsc70 and proteins, these were not of the nuclear pore-targeting complex (Fig. 3 The protein identified in this did not with rabbit antibodies against the heat-shock protein (data not shown). experiments be to hsc70 and a in the step of transport. of hsc70 with (8Imamoto N. Matsuoka Y. Kurihara T. Kohno K. Miyagi M. Sakiyama F. Okada Y. Tsunasawa S. Yoneda Y. J. Cell Biol. 1992; 119: 1047-1061Crossref PubMed Scopus (149) Google Scholar, N. Matsuoka Y. T. Okada Y. Uchida T. Yoneda Y. J. Biol. Chem. 1990; Full Text PDF PubMed Google Scholar) or targeting complex the formation of the targeting complex in the cytoplasm or the pore-targeting of the complex. these proteins be required for an other step in the transport process. Histone specifically cytosolic factors that the complex with T-peptide conjugates for the These results support the that both histone H1 and T-peptide conjugates transport to into the the NLS of histone H1 is Since proteins were to histone (Fig. 4 it is yet the four proteins identified in this form a targeting complex with histone H1 or with other karyophilic associated with histone It other of karyophilic proteins such as karyophiles containing NLS and form a nuclear pore-targeting complex with the same as of the complex identified in this WGA is well known to inhibit the translocation step of but not the binding However, in the study, nuclear rim binding of a transport substrate was inhibited by WGA in the presence of H1-depleted cytosol (Fig. 2 C, and only inhibited in the absence of H1-depleted cytosol (Fig. 2 C, The of WGA on the nuclear pore binding step in vitro was previously by Adam and Adam Adam S.A. J. Cell Biol. 1994; PubMed Scopus Google Scholar). that H1-depleted cytosol contains which transport substrate at nuclear pores when the subsequent translocation step is The of rim binding by observed in the absence of H1-depleted cytosol, be to a of activity in the permeabilized proteins of molecular 54, 56, 66, and 90 kDa were found in the pore-targeting complex formed in Q550. found that when Q550 was fractionated by gel adding transport no fractions supported pore targeting of T-peptide conjugates (data not shown). These results that at least two factors are required for the targeting activity, indicating that both the and proteins identified in this are for the targeting activity. our of the targeting activity of the complex was not when the complex was formed by adding proteins to the The of and proteins in nuclear which are associated with T-peptide conjugates in an NLS-dependent Since we used karyophiles containing NLS peptides per it is that these two proteins with NLS from and proteins and as NLS in with other cytoplasmic factors in a fraction other Q550. and proteins with and proteins to complex formation or the complex once experiments are to these two proteins actually have a in nuclear of each of the targeting complex identified in this study, with of formation, nuclear pore and to of the molecular of nuclear protein are to and for the of rabbit antibodies and to H. Seino and T. Nishimoto for anti-Ran/TC4