INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Although eukaryotic ribosomes function in the cytoplasm, the synthesis, processing, and assembly of the ribosomal subunits in Saccharomyces cerevisiae and higher eukaryotes occur in the nucleolus. The entire ribosome is composed of four rRNA species and ~75 ribosomal proteins (r-proteins) distributed between two subunits. The 18S, 5.8S, and 25S rRNAs are derived from a single 35S rRNA precursor that is synthesized by RNA polymerase I and then processed by a series of endonucleolytic and exonucleolytic cleavages (reviewed by Kressler et al., 1999; Venema and Tollervey, 1999). The 5S rRNA is synthesized separately by RNA polymerase III and associates with the 60S preribosomal subunit early in assembly. The mature 40S ribosomal subunit contains the 18S rRNA and ~32 r-proteins, whereas the 60S subunit is composed of the 5S, 5.8S, and 25S rRNAs and ~45 r-proteins. Proper assembly of each ribosomal subunit requires the coordination of several events, including the synthesis and import of r-proteins, the synthesis and processing of rRNA, and the concomitant assembly of r-proteins into the preribosomal subunits. Although a pathway for 35S rRNA maturation has been well defined through both genetic and biochemical approaches (reviewed by Kressler et al., 1999; Venema and Tollervey, 1999), less is known about the association of r-proteins with the rRNA and the export of the assembled subunits out of the nucleus.

All nucleocytoplasmic transport occurs through the nuclear pore complex (NPC). The yeast NPC is composed of multiple copies of ~30 different nuclear pore proteins referred to as nucleoporins (Rout et al., 2000). Together, these nucleoporins form the overall structure of the NPC, which consists of a membrane-embedded central core with fibrils protruding from its cytoplasmic face and a basket-like structure that extends out on the nuclear side (Yang et al., 1998; Stoffler et al., 1999; Allen et al., 2000). The active nuclear pore can accommodate transport of large macromolecules, including export of the ribosomal subunits.

Transport of cargo into and out of the nucleus requires not only interactions with the NPC but also soluble transport factors. The best defined transport path is the import of proteins containing a classic nuclear localization signal (NLS) by the importin / receptor (reviewed by Corbett and Silver, 1997; Gorlich and Kutay, 1999; Wente, 2000). Importin /Srp1 recognizes and binds to proteins containing a NLS and together with importin /Rsl1 travels through the NPC. Thirteen importin  homologues termed karyopherins were identified in S. cerevisiae by sequence comparison to importin  (Gorlich et al., 1997). These karyopherins act as receptors for specific import and export cargoes. Transport substrates for several of the karyopherins have already been identified (reviewed by Gorlich and Kutay, 1999). The nucleotide-bound state of the small GTPase Ran/Gsp1 imparts directionality to transport (Gorlich et al., 1996; Izaurralde et al., 1997). Dissociation of the karyopherin-cargo complex in the nucleus is mediated by RanGTP association (Rexach and Blobel, 1995; Chi et al., 1996; Gorlich et al., 1996), whereas GTP hydrolysis in the cytoplasm results in the release of export cargo (Bischoff and Gorlich, 1997; Floer et al., 1997; Lounsbury and Macara, 1997). In yeast, the high concentration of RanGTP in the nucleus is maintained by the guanine nucleotide exchange factor Prp20 (Amberg et al., 1993; Kadowaki et al., 1993), and the cytoplasmic pool of RanGDP is generated by the GTPase-activating protein Rna1 (Becker et al., 1995). The mechanism by which the transport complexes travel through the NPC is not as clear.

Microinjection experiments in Xenopus oocytes demonstrated that ribosomal subunit export is an energy-dependent, factor-mediated, and unidirectional process that requires components of the NPC (Bataille et al., 1990). In addition, export of microinjected 40S ribosomal subunits does not compete with export of tRNA out of the nucleus, indicating separate pathways for export (Pokrywka and Goldfarb, 1995). A role for the GTPase Ran in ribosome assembly or export was proposed based on the observation that rRNA processing is delayed in yeast strains bearing conditional alleles of either of the Ran regulators PRP20 or RNA1 (Traglia et al., 1989; Kadowaki et al., 1993). These early observations were recently confirmed for both the 40S and 60S ribosomal subunits with the use of two independent assays (Hurt et al., 1999; Moy and Silver, 1999). Mutation of the Ran regulators Rna1, Prp20, and Yrb1 caused the 40S ribosomal subunit to accumulate in the nucleus of yeast, as determined by in situ hybridization to 20S rRNA (Moy and Silver, 1999). In addition to the Ran regulators, a subset of nucleoporin mutations and a temperature-sensitive mutation of the nuclear export sequence receptor Xpo1/Crm1 also blocked export of the 40S ribosomal subunit (Moy and Silver, 1999). The 60S ribosomal subunit also accumulated in the nucleus of yeast strains bearing conditional alleles of RNA1 and PRP20, as determined by the localization of a fusion between the ribosomal protein L25 (Rpl25) and the green fluorescent protein (GFP) (Hurt et al., 1999). In addition, temperature-sensitive mutations of the nucleoporins Nsp1, Nup49, and Nic96 also caused mislocalization of the 60S ribosomal subunit (Hurt et al., 1999). It is likely that the factors identified to date represent only a subset of the transport factors required for ribosome export.

Here we describe an assay to follow the localization of the 60S ribosomal subunit in S. cerevisiae with the use of a fusion between the ribosomal protein L11b (Rpl11b) and GFP. This assay differs from the Rpl25-GFP export assay described previously (Hurt et al., 1999) in that nuclear accumulation of Rpl11b-GFP can be detected under conditions of logarithmic growth. A large panel of mutant yeast strains were screened for defects in 60S ribosomal subunit assembly and transport with the use of this assay. Surprisingly, deletion of KAP120, one of the nonessential karyopherins, caused a strong accumulation of the 60S ribosomal subunit in the nucleus. Further characterization revealed that kap120 cells have a slight delay in rRNA processing and a significant reduction in free 60S ribosomal subunits.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Yeast Strains

Table 1 lists the yeast strains used in this study. The strains PSY2083 and PSY2084 were constructed by replacing the ORF of RPL11A and RPL11B with HIS3 by means of a PCR-based method (Baudin et al., 1993). The haploid strain PSY685 was transformed with a HIS3 PCR product targeted to the RPL11A locus to generate PSY2083, and PSY581 was transformed with a HIS3 PCR product targeted to the RPL11B locus to create PSY2084. Disruptions of RPL11A and RPL11B were verified by Southern blot analysis (Southern, 1975).

Yeast strains lacking wild-type copies of both RPL11A and RPL11B were generated by crossing PSY2083 carrying a wild-type copy of RPL11A on a URA3 CEN plasmid to PSY2084. The diploid was sporulated, and tetrads were analyzed. HIS⁺ and 5-FOA-sensitive (FOA^S) spores in tetrads that segregated 2:2 HIS⁺:his and 2:2 FOA^S:FOA^R (5-FOA-resistant) were identified as the double mutant strain rpl11a::HIS3 rpl11b::HIS3.

Plasmids

The LEU2 CEN (pPS2167) and TRP1 CEN (pPS2168) RPL11B-GFP plasmids were constructed as follows. Yeast genomic DNA was amplified by PCR with a 5' primer that annealed 891 base pairs 5' of the ORF for RPL11B (5'GCATCTACTAGTGCAGGATTACGAAGACTTC3') and a 3' primer that annealed to the 3' end of RPL11B excluding the stop codon (5'CAGTCACTCGAGCTTTATCGAGCACATCAGCG3'). The PCR product was digested with SpeI and XhoI and cloned into the SpeI-XhoI site of either a LEU2 or a TRP1 CEN vector that contained the ORF of GFP followed by the 3' untranslated region of NUF2 (Kahana and Silver, 1998). To generate the URA3 CEN RPL11B-GFP plasmid (pPS2169), a NotI-XhoI fragment of pPS2167 was cloned into the NotI-XhoI site of a URA3 CEN plasmid that contained GFP and the 3' untranslated region of NUF2 (Kahana and Silver, 1998). The wild-type RPL11A clone YCp50L16A (RPL11A URA3 CEN) was provided by J. Woolford (Carnegie Mellon University, Pittsburgh, PA) (Rotenberg et al., 1988).

Separation of Ribosomal Subunits

Yeast cells expressing Rpl11b-GFP from pPS2167 were grown in 100 ml of synthetic complete medium lacking leucine (leu) to a density of 1 × 10⁷ cells/ml at 25°C. Cells were harvested by centrifugation at 2000 rpm at room temperature and washed with buffer B (50 mM Tris-Cl, pH 7.5, 50 mM NaCl, 1 mM DTT [Foiani et al., 1991]). Lysate was prepared by glass-bead disruption of cell pellets resuspended in buffer B containing protease inhibitors (1 mM PMSF and 2.5 µg/ml each pepstatin A, leupeptin, chymostatin, and aprotinin). Lysate (200 µg of total RNA) was loaded onto a 10 ml 7-30% sucrose gradient prepared in buffer B and centrifuged at 39,000 rpm for 2.5 h at 4°C. Sucrose gradient profiles were recorded by absorption at 254 nm, and 0.5-ml fractions were collected.

Total protein was precipitated from each 0.5 ml sucrose gradient fraction by the addition of ice-cold trichloroacetic acid to a final concentration of 12% and incubation overnight at 20°C. The fractions were pelleted by centrifugation at 14,000 rpm for 15 min at 4°C, and pellets were washed two times with 200 µl of ice-cold 100% acetone. Dried pellets were resuspended in 90 µl of 1× gel sample buffer (10% glycerol, 2% SDS, 0.1 M DTT, 0.1% bromphenol blue), and 10 µl of each fraction was separated on a 10% SDS-polyacrylamide gel followed by transfer to a nitrocellulose membrane by standard methods (Sambrook et al., 1989). For GFP detection, anti-GFP was used at a 1:2500 dilution in PBST (PBS, 0.25% Tween 20) plus 2.5% milk for 1 h at 25°C. HRP-conjugated anti-rabbit secondary antibody (Jackson Immunoresearch, West Grove, PA) was used at a dilution of 1:5000 in PBST plus milk, and ECL (Amersham Pharmacia Biotech, Uppsala, Sweden) was used to detect immunoreactive bands.

Rpl11b-GFP Localization Assay

Yeast strains expressing Rpl11b-GFP were grown in selective medium (12 ml) to 1-2 × 10⁷ cells/ml at 25°C. Strains that were ade2 were grown in selective medium supplemented with 20 µg/ml adenine sulfate. Before shifting the culture to the nonpermissive growth temperature, a 3-ml aliquot was removed and cells were fixed by the addition of formaldehyde to 3.7% and incubation with agitation at 25°C. After 30 min, the fixed cells were collected, washed two times with 1 ml of 0.1 M potassium phosphate, pH 6.5 (KPi), and stored in 1 ml of solution P (0.1 M KPi, pH 6.5, 1.2 M sorbitol) at 4°C. The remaining culture was shifted to either 37°C for temperature-sensitive strains or 15°C for cold-sensitive strains. The length of the shift depended on the yeast strain being tested and the onset of its growth defect as determined from the literature. At the end of the temperature shift, a 3-ml aliquot was fixed as described above with incubation at the shift temperature (37 or 15°C). The culture was then shifted back to 25°C, and aliquots were removed at 30 and 60 min and fixed at 25°C as described. After permeabilization with 0.5% Triton X-100 for 10 min at room temperature, cell nuclei were stained with DAPI at a final concentration of 1 µg/ml for 3 min at room temperature. Cells were washed two times with 1 ml of 1× PBS and stored at 4°C in 100-200 µl of 1× PBS. Rpl11b-GFP was visualized by fluorescence microscopy as described previously with the use of a Nikon (Garden City, NY) microscope (Ferrigno et al., 1998). Images of cells were captured with a Princeton Instruments (Trenton, NJ) Micromax camera and Metamorph Imaging software (Universal Imaging, Westchester, PA).

Pulse-Chase Labeling of rRNA

Yeast cultures (100 ml) were grown in synthetic complete medium lacking methionine (met) to 1 × 10⁷ cells/ml at 25°C. Cells were harvested by centrifugation at 2000 rpm and resuspended in 3 ml of met medium. To label rRNA specifically, 250 µCi of [³H-methyl]-methionine (specific activity, 70-85 Ci/mmol; Amersham Pharmacia Biotech) was added, and cells were incubated for 3 min at 25°C. At 3 min, unlabeled methionine in met medium was added to a final concentration of 5.1 mM, and at chase times of 15 s, 2 min 15 s, and 9 min 15 s, 1 ml aliquots of cells were removed from the reaction. Cells were collected by brief centrifugation, the supernatant was removed, and the pellets were frozen on dry ice. Processing of each sample took ~45 s, so this time was included in the time of chase, giving 1-, 3-, and 10-min chase time points.

Total RNA was extracted from labeled cells by hot acid phenol treatment as described previously (Lundblad, 1997), and 10,000 cpm of each sample was separated on a 1.2% formaldehyde agarose gel. RNA was transferred to a Hybond-N⁺ membrane (Amersham Pharmacia Biotech) by vacuum transfer (VacuGene XL, Amersham Pharmacia Biotech), and the membrane was sprayed with EN³HANCE (New England Nuclear, Boston, MA) before exposure to film for 3 d at 80°C.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Rpl11b-GFP Serves as a Marker for the 60S Ribosomal Subunit in S. cerevisiae

Rpl11 is one of the ~45 ribosomal proteins that together with 5S, 5.8S, and 25S rRNA forms the 60S ribosomal subunit in S. cerevisiae. Like many of the r-proteins in yeast, Rpl11 is expressed from two gene copies, RPL11A and RPL11B (Woolford et al., 1979; Leer et al., 1984). These two copies code for 99% identical proteins; however, the level of expression from each copy differs significantly (Rotenberg et al., 1988). Two-thirds of Rpl11 expressed in yeast is from the B copy and one-third is from the A copy. Deletion of both RPL11A and RPL11B is lethal, indicating that Rpl11 is an essential component of the 60S subunit in yeast (Rotenberg et al., 1988). Rpl11 associates with the preribosomal subunit late in the assembly pathway, presumably after the 35S rRNA precursor is cleaved into its 40S and 60S components (Kruiswijk et al., 1978; Kressler et al., 1999). We chose to use Rpl11b as a marker for the 60S subunit because it is an essential component of the 60S subunit, it binds specifically to the 60S preribosomal subunit, and it is expressed at a higher level than Rpl11a. Because a fusion between Rpl11a and -galactosidase was functional in yeast (Tsay et al., 1994), we predicted that the addition of the much smaller GFP to the carboxyl terminus of Rpl11b would also produce a functional fusion protein. In addition, the GFP tag would allow visualization of the 60S ribosomal subunit in living cells.

For Rpl11b-GFP to serve as a reporter for the 60S ribosomal subunit, it must meet three criteria. First, Rpl11b-GFP must be able to functionally replace endogenous Rpl11; second, the localization of Rpl11b-GFP in the cell must reflect that of the 60S ribosomal subunit; and third, the majority of Rpl11b-GFP expressed in the cell must be incorporated into the 60S ribosomal subunit.

A low-copy LEU2 plasmid bearing GFP fused in frame to the 3' end of RPL11B was constructed. The 5' promoter of RPL11B was used to drive expression of the fusion protein. A genetic background in which RPL11B is required for viability was generated to verify that RPL11B-GFP was functional in yeast. A diploid strain disrupted for one copy of each RPL11A and RPL11B by replacement with HIS3 and carrying a wild-type copy of RPL11A on a URA3 CEN plasmid was constructed. The diploid was sporulated, and the resulting tetrads were analyzed to identify tetrads that segregated 2:2 his:HIS⁺ and 2:2 FOA^R:FOA^S, indicating the presence of two wild-type (RPL11A RPL11B) and two double mutant spores (rpl11a::HIS3 rpl11b::HIS3), respectively. An example tetrad is shown in Figure 1A (left), where spores a and d do not grow on synthetic complete medium containing FOA and thus require the RPL11A URA3 CEN plasmid for cell viability, and spores b and c, the wild-type spores, are FOA-resistant. Spores a and d were no longer FOA-sensitive when transformed with the RPL11B-GFP LEU2 CEN plasmid (Figure 1A, right), indicating that RPL11B-GFP is a functional gene fusion.

View larger version (39K): [in this window] [in a new window]   Figure 1.   The RPL11B-GFP gene fusion is functional in yeast. (A) Four spores of a representative tetrad from the cross between PSY2083 (rpl11a::HIS3) covered by YCp50L11A URA3 CEN and PSY2084 (rpl11b::HIS3) streaked on synthetic complete medium containing FOA (left panel). Spores a and d are HIS+ and FOAS and spores b and c are his and FOAR. Spores a-d were transformed with pPS2167 (RPL11B-GFP LEU2 CEN) and streaked on synthetic complete medium containing FOA (right panel). (B) Rpl11b-GFP expressed from a CEN plasmid in wild-type yeast was visualized by fluorescence microscopy of living cells (left panel). Arrows point to a vacuole (v) and the nucleus (n). Nomarski, phase-contrast image of cells; Rpl11b-GFP, fluorescence signal. Yeast lysate prepared from wild-type yeast cells expressing Rpl11b-GFP was probed with anti-GFP to detect Rpl11b-GFP (right panel). Five micrograms of total protein of lysate was loaded. Migration of a 10-kDa ladder of molecular mass markers is shown.

The localization of Rpl11b-GFP in wild-type yeast cells grown to early log phase was determined by fluorescence microscopy. Rpl11b-GFP is predominantly cytoplasmic and for the most part excluded from both the vacuoles (v) and the nucleus (n) (Figure 1B, arrows). This localization is consistent with the steady-state cytoplasmic location of 60S ribosomal subunits active in translation. To verify that the GFP signal seen in the cytoplasm is due to intact Rpl11b-GFP, lysate prepared from wild-type yeast cells expressing Rpl11b-GFP was run on a 10% SDS-polyacrylamide gel and immunoblotted with anti-GFP (Figure 1B, right). A single polypeptide migrating just under 50 kDa was present, consistent with the predicted molecular mass of ~47 kDa for a fusion between ~20-kDa Rpl11b and 27-kDa GFP.

The incorporation of Rpl11b-GFP into the 60S ribosomal subunit was verified by preparing lysate from wild-type yeast cells expressing Rpl11b-GFP and separating the individual ribosomal subunits by high-speed sucrose gradient centrifugation. Fractions from the sucrose gradient were assayed for the presence of Rpl11b-GFP by separation on a 10% SDS gel and immunoblotting with anti-GFP. As seen in Figure 2, Rpl11b-GFP sedimented with the 60S ribosomal subunit and only trace amounts of Rpl11b-GFP were detectable in the 40S subunit and soluble fractions of the sucrose gradient. The gradient profile and distribution of Rpl11b-GFP were similar for a yeast strain that has GFP integrated at the 3' end of the genomic copy of RPL11B (our unpublished results).

View larger version (18K): [in this window] [in a new window]   Figure 2.   Rpl11b-GFP is incorporated in the 60S ribosomal subunit. Yeast lysate from wild-type cells expressing Rpl11b-GFP from a CEN plasmid was separated on a 7-30% sucrose gradient. The sucrose gradient profile was recorded by absorbance at 254 nm (upper panel). Fractions 8-19 collected from the sucrose gradient and lysate (5 µg of total protein) were probed with anti-GFP to detect Rpl11b-GFP (bottom panel).

60S Ribosomal Subunit Localization Assay

An assay was developed to screen for factors involved in ribosome subunit synthesis or export with the use of Rpl11b-GFP as a marker for the 60S ribosomal subunit. Mutant yeast cells expressing Rpl11b-GFP were grown to early log phase in liquid culture at 25°C, shifted to the nonpermissive growth temperature to induce the mutant phenotype, and then shifted back to the permissive growth temperature to induce new ribosomal protein synthesis. Yeast cells were formaldehyde fixed, their nuclei were stained with DAPI at each step of the assay, and the localization of Rpl11b-GFP was determined by fluorescence microscopy. The location and distribution of Rpl11b-GFP in cells was not affected by the fixing procedure. The shift back to the permissive temperature was required because synthesis of ribosomal proteins, and thus of the reporter Rpl11b-GFP, is reduced at 37°C in yeast (Gorenstein and Warner, 1976; Kim and Warner, 1983; Hurt et al., 1999). A yeast strain that is defective in 60S ribosomal subunit assembly or transport should accumulate Rpl11b-GFP in the nucleus upon shifting back to the permissive temperature if the rate of ribosome synthesis is faster than the rate at which the mutant phenotype reverses. The 60S ribosomal subunit should remain cytoplasmic in wild-type yeast cells throughout this assay. As seen in Figure 3, Rpl11b-GFP is cytoplasmic in a wild-type strain after a 1-h shift to 37°C (Figure 3a) and after shifting the cells back to 25°C for 30 min (Figure 3c). The GFP signal was not as bright at 37°C, consistent with a reduction in r-protein synthesis at this temperature. Because this assay depends on the synthesis, import, and assembly of Rpl11b-GFP into the 60S subunit, mutants that slow the import of r-proteins, inhibit rRNA synthesis or processing, or reduce import of any factor required for ribosome transport may also affect Rpl11b-GFP localization.

View larger version (50K): [in this window] [in a new window]   Figure 3.   Rpl11b-GFP localization assay. Wild-type (PSY580) or mtr4-1 cells expressing Rpl11b-GFP from a CEN plasmid were grown to early log phase, shifted to 37°C for 1 h, and then shifted back to 25°C for up to 1 h. Yeast cells were fixed and incubated with DAPI to stain the nuclei. Rpl11b-GFP was visualized by fluorescence microscopy. Panels a, b, e, and f show cells shifted to 37°C for 1 h, and panels c, d, g, and h show cells shifted back to 25°C for 30 min.

Yeast Strains Defective in Ribosome Assembly Accumulate Rpl11b-GFP in the Nucleus

Proper assembly of 60S ribosomal subunits is required for their export from the nucleus (Gorlich and Kutay, 1999). Therefore, yeast mutants that are defective in 60S ribosomal subunit assembly may alter the steady-state localization of Rpl11b-GFP. To test this prediction, several ribosomal assembly mutants, including mtr4-1/dob1-1, nop1-3, fal1-1, nsr1, and rat1-1, were screened for effects on Rpl11b-GFP localization. Mtr4p/Dob1p is involved in the processing of 7S rRNA to mature 5.8S rRNA, and mutations cause defects in 60S ribosomal subunit assembly (de la Cruz et al., 1998). The location of Rpl11b-GFP in mtr4-1 cells grown to early log phase, shifted to 37°C for 1 h, and then shifted back to 25°C was determined by fluorescence microscopy. As predicted for an assembly mutant, Rpl11b-GFP accumulated in the nucleus of >70% of mtr4-1 cells as early as 30 min after shifting back to the permissive temperature (Figure 3g). The fluorescent signal overlaps with the DAPI signal, indicating that Rpl11b-GFP is in the nucleus of mtr4-1 cells (Figure 3h). Similar results were obtained with the dob1-1 mutation, although the defect was even more pronounced in this strain. Rpl11b-GFP accumulated in the nucleus of >90% of dob1-1 cells after the shift back to 25°C.

The data for all of the ribosomal assembly mutants tested are summarized in the fourth column of Table 2. The temperature-sensitive nop1-3 strain that has defects in rRNA methylation and 60S subunit assembly mislocalized Rpl11b-GFP to the nucleus of 20-25% of cells after shifting back to the permissive temperature. Unexpectedly, fal1-1 and to a greater extent nsr1 cells also accumulated Rpl11b-GFP in the nucleus. Both mutations inhibit cleavage of the 35S rRNA at sites that divide the precursor into the 40S (20S RNA) and 60S precursor RNAs (27SA₂ RNA). These strains have a reduced level of mature 18S rRNA and therefore are considered to be primarily defective in 40S ribosomal subunit assembly (Kondo and Inouye, 1992; Kondo et al., 1992; Lee et al., 1992; Kressler et al., 1997). The localization of Rpl11b-GFP was not altered in yeast cells bearing a mutation of the 5' to 3' exonuclease Rat1 that is involved in the conversion of the 7S_S RNA precursor to 5.8S_S rRNA (Amberg et al., 1992; Henry et al., 1994).

Several Nucleoporin Mutations Cause Rpl11b-GFP to Accumulate in the Nucleus

Eighteen nucleoporin mutants were screened for their effect on Rpl11b-GFP localization, and the results for two of these mutants are shown in Figure 4A. Nup157 is a nonessential nucleoporin that is one of the core nuclear pore proteins (Aitchison et al., 1995b). Rpl11b-GFP remained cytoplasmic in nup157-2 cells throughout the localization assay (Figure 4A, a and c). In contrast, mutation of NUP120 (rat2-2), a nucleoporin implicated in mRNA export, accumulated Rpl11b-GFP in the nucleus of >50% of cells 30 min after shifting back to the permissive temperature (Figure 4A, g). The change in Rpl11b-GFP localization in nup120 cells was not due to accumulation of free Rpl11b-GFP because Rpl11b-GFP remained in the 60S ribosomal subunit throughout the assay, as determined by sucrose gradient analysis (our unpublished results).

View larger version (63K): [in this window] [in a new window]   Figure 4.   Rpl11b-GFP localization in nucleoporin and karyopherin mutants. Yeast strains nup157-2, rat2-2, msn5, and kap120 expressing Rpl11b-GFP from a CEN plasmid were grown to early log phase, shifted to 37°C for 1 h, and then shifted back to 25°C for up to 1 h. Cells were fixed, and their nuclei were stained with DAPI. (A) Fluorescence microscopy images of nup157-2 cells shifted to 37°C (a and b), nup157-2 cells shifted back to 25°C for 30 min (c and d), rat2-2 cells shifted to 37°C (e and f), and rat2-2 cells 30 min after shifting back to 25°C (g and h). (B) Fluorescence microscopy images of msn5 cells shifted to 37°C (a and b), msn5 cells shifted back to 25°C for 30 min (c and d), kap120 cells shifted to 37°C (e and f), and kap120 cells 30 min after shifting back to 25°C (g and h).

The results for all of the nucleoporin mutants screened for mislocalization of the 60S ribosomal subunit are summarized in Table 2. Temperature-sensitive alleles of the essential nucleoporins NSP1, NUP1, NUP49, NUP82, NIC96, and NUP159 all caused a strong accumulation of Rpl11b-GFP in the nucleus after shift back to the permissive temperature (Table 2). In addition, several of these strains accumulated Rpl11b-GFP in the nucleus at the permissive temperature (Table 2). Interestingly, nup85 cells accumulated Rpl11b-GFP in the nucleus at the permissive temperature of 25°C before shifting to 37°C and appeared to concentrate Rpl11b-GFP in the nucleolus after 1 h at 37°C. Not all of the nucleoporin mutants tested accumulated Rpl11b-GFP in the nucleus (Table 2), indicating that there is some degree of specificity for the 60S ribosomal subunit.

Transport Factors

Several known transport factors, including mutations of the RAN regulators PRP20 and RNA1, were screened for accumulation of Rpl11b-GFP in the nucleus, and the results are shown in Table 2. Both prp20-1 and rna1-1 cells accumulated Rpl11b-GFP in the nucleus after shift back to the permissive temperature; however, the defect was present in only ~25% of the cells. A temperature-sensitive mutation of the Ran-binding protein Yrb1 caused a strong accumulation of Rpl11b-GFP in the nucleus. Mutations in the hnRNP-like protein Npl3 and its importer Mtr10 also caused accumulation of Rpl11b-GFP in the nucleus. Rpl11b-GFP also accumulated in the nucleus of cells bearing mutant alleles of RAT8, a gene that codes for a helicase that is implicated in mRNA export (Snay-Hodge et al., 1998; Tseng et al., 1998). Not all mutations that cause defects in mRNA export mislocalized Rpl11b-GFP, because mutation of the RNA export factor Mex67 did not block export of Rpl11b-GFP from the nucleus.

Karyopherins

Mutations of importin  (SRP1), importin  (RSL1), and 12 additional karyopherins were screened for their effect on 60S ribosomal subunit localization (Table 2). Temperature-sensitive mutations of the NLS receptor SRP1 (srp1-31) and of the importin  receptor RSL1 (rsl1-4) both caused Rpl11b-GFP to accumulate in the nucleus. A cold-sensitive mutation of CSE1 (cse1-1), the export receptor for Srp1, also caused Rpl11b-GFP to accumulate in cells shifted to 15°C. The results for two nonessential karyopherins, MSN5 and KAP120, are shown in Figure 4B. Deletion of MSN5 did not alter the cytoplasmic localization of Rpl11b-GFP at either 37 or 25°C (Figure 4B, a-d), whereas deletion of KAP120 caused a surprisingly strong mislocalization of Rpl11b-GFP in the nucleus at 25°C (Figure 4B, g). Rpl11b-GFP was trapped in the nucleus of >90% of kap120 cells even before shifting to 37°C, whereas at 37°C, when ribosomal protein synthesis was reduced, Rpl11b-GFP was distributed throughout the cell (Figure 4B, e).

Deletion of NMD5, the yeast homologue of the mammalian ribosomal protein importer RanBP7 (Gorlich et al., 1997; Jakel and Gorlich, 1998), also caused a strong nuclear accumulation of Rpl11b-GFP at 25°C. Deletion of KAP123, one of the ribosomal protein importers in yeast (Rout et al., 1997), and a temperature-sensitive mutant of KAP104, the yeast homologue of the mammalian ribosomal protein importer transportin (Jakel and Gorlich, 1998), also caused the 60S subunit to accumulate in the nucleus to a lesser extent. Although Pse1 has been implicated in the import of ribosomal proteins (Rout et al., 1997), the 60S reporter Rpl11b-GFP remained in the cytoplasm of pse1-1 cells throughout the localization assay. Interestingly, Rpl11b-GFP also accumulated in ~30% of xpo1-1 cells that were first shifted to 37°C for 1 h and then shifted back to 25°C for 1 h. Xpo1 is the receptor for nuclear export signal-containing proteins (Stade et al., 1997). Deletion of the karyopherins KAP114, SXM1, and LOS1 had no effect on the localization of the 60S ribosomal subunit (Table 2).

Deletion of KAP120 Causes a Delay in rRNA Processing and a Reduction in the Level of 60S Ribosomal Subunits

Because Kap120 is one of the few karyopherins whose transport substrate(s) has not yet been identified and whose function is not known, we chose to characterize the 60S export defect in kap120 cells in more detail. Processing of the 35S rRNA precursor was analyzed in kap120 cells to investigate whether nuclear accumulation of Rpl11b-GFP was due to a defect in ribosomal rRNA maturation. The 35S rRNA precursor can be specifically labeled with [³H]methionine because rRNA is highly methylated in vivo. To follow the processing of the 35S rRNA precursor, KAP120 and kap120 strains were grown to 1 × 10⁷ cells/ml at 25°C, pulse labeled with [³H]methionine for 3 min, and then chased with an excess of unlabeled methionine for up to 10 min. Conversion of the 35S rRNA precursor to the mature 25S and 18S rRNAs was nearly complete after 10 min of chase with unlabeled methionine for wild-type KAP120 cells (Figure 5). In contrast, deletion of KAP120 led to an accumulation of the 35S rRNA precursor that was still detectable after 10 min of chase (Figure 5). A 23S aberrant RNA species was also detected in kap120 cells at early chase times (Figure 5). Although rRNA processing was delayed in kap120 cells, the levels of mature 25S and 18S rRNA were not depleted significantly (Figure 5).

View larger version (82K): [in this window] [in a new window]   Figure 5.   Processing of rRNA in KAP120 and kap120 cells. KAP120 and kap120 cells were grown at 25°C to a density of 1 × 107 cells/ml, pulse labeled for 3 min with [3H-methyl]methionine, and chased with an excess of unlabeled methionine. Chase time points of 1, 3, and 10 min for each yeast strain are shown. The 35S rRNA precursor and processing intermediates are labeled.

Because a delay in rRNA processing is often indicative of a ribosome assembly defect, the ratio of 60S to 40S ribosomal subunits in kap120 and KAP120 cells was determined. Lysates prepared from kap120 and KAP120 cells grown at 25°C were subjected to sucrose gradient sedimentation to separate the individual subunits. The subunit peaks were detected by their absorption at 254 nm. As seen in the sucrose gradient profiles in Figure 6, deletion of KAP120 reduced the 60S ribosomal subunit peak significantly compared with the wild-type KAP120 strain. The level of 40S ribosomal subunits, however, was reduced only slightly in kap120 cells (Figure 6).

View larger version (12K): [in this window] [in a new window]   Figure 6.   Deletion of KAP120 causes a large reduction in 60S ribosomal subunit levels. Yeast lysates prepared from KAP120 and kap120 cells expressing Rpl11b-GFP from a CEN plasmid were centrifuged through 7-30% sucrose gradients, and gradient profiles were collected by absorbance at 254 nm. An overlay of the gradient profiles for KAP120 and kap120 cells is shown to emphasize the difference in peak heights for the 60S and 40S ribosomal subunits. Equal amounts of total RNA (200 µg) were loaded for each gradient.

The accumulation of Rpl11b-GFP in the nucleus of kap120 cells was not an indirect effect of trapping mRNA in the nucleus. Polyadenylated RNA was localized throughout the cytoplasm of kap120 cells, as determined by in situ hybridization with a digoxigenin-labeled oligo-dT probe (our unpublished results). In addition, nuclear accumulation of Rpl11b-GFP in kap120 cells was not due to free Rpl11b-GFP because Rpl11b-GFP was incorporated in the 60S ribosomal subunit, as determined by sucrose gradient analysis (our unpublished results). Although it is possible that Kap120 is another ribosomal protein importer, it does not import either Rpl11 or Rpl25. Rpl11b-GFP and a Rpl25-NLS-GFP-GFP reporter are both localized in the nucleus of kap120 cells at 25°C, indicating that their import is not blocked. In contrast, the Rpl25-NLS-GFP-GFP reporter is distributed in both the cytoplasm and the nucleus of cells deleted for KAP123, an Rpl25 importer (our unpublished results).

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

To further our understanding of 60S ribosomal subunit assembly and export, we have designed an assay that uses Rpl11b-GFP to follow 60S ribosomal subunit localization in yeast. Mutant yeast strains that are defective for 60S ribosomal subunit assembly or export accumulate Rpl11b-GFP in the nucleus upon shifting cells from the nonpermissive growth temperature back to the permissive growth temperature. The shift back to the permissive growth temperature is required to increase the nuclear pool of Rpl11b-GFP available for ribosome biogenesis because r-protein synthesis in yeast is reduced under mild stress conditions (Gorenstein and Warner, 1976; Kim and Warner, 1983). Using this assay, we identified mutations in several nucleoporins and transport factors that trap the 60S ribosomal subunit in the nucleus, including mutation of the karyopherin Kap120.

The Rpl11b-GFP localization assay described here differs from the 60S export assay that uses Rpl25-GFP as the tag for the 60S ribosomal subunit (Hurt et al., 1999). Nuclear accumulation of Rpl11b-GFP can be detected in cells during logarithmic growth when they are actively producing ribosomes, whereas Rpl25-GFP accumulation is observed when cells are shifted from conditions of starvation (saturation) to fresh medium (Hurt et al., 1999). The two r-proteins also bind at different points on the ribosome biogenesis pathway (Kruiswijk et al., 1978; Kressler et al., 1999). Rpl25 binds to the 35S rRNA precursor before it is divided into its 40S (20S rRNA) and 60S (27SA₂ rRNA) specific components, whereas Rpl11 associates later in the pathway, most likely with the 27SA₂ preribosomal particle. Although the differences between these two 60S localization assays may seem minor, they have the potential to identify a different set of factors required for ribosomal subunit assembly and export. Because Rpl25 binds directly to the rRNA, it is likely to be buried within the 60S ribosomal subunit (El-Baradi et al., 1984, 1987; Yeh and Lee, 1998). Rpl11 binds later in ribosome assembly and appears to be located at the surface of the 60S ribosomal subunit (Tsay et al., 1994). Thus, Rpl11 may be available for recognition by the ribosome assembly or export machinery.

The distribution of Rpl11b-GFP in yeast cells is sensitive to defects in ribosome synthesis, as expected. Rpl11b-GFP is predominantly cytoplasmic in wild-type yeast cells at all temperatures assayed, consistent with the cytoplasmic localization of active 80S ribosomes. In contrast, Rpl11b-GFP accumulates in the nucleus of yeast strains bearing temperature-sensitive mutations in the 60S ribosome assembly factors Mtr4/Dob1 and Nop1. Although rat1-1 cells have a defect in 5.8S rRNA processing (Amberg et al., 1992; Henry et al., 1994), Rpl11b-GFP did not accumulate in this strain. It is possible that a defect in rat1-1 cells was not detected because in its absence the exonuclease Xrn1 completes 5.8S rRNA processing (Henry et al., 1994; Petfalski et al., 1998). Rpl11b-GFP did accumulate in fal1-1 and nsr1 cells, which are defective in cleavage of the 35S rRNA precursor into the 20S and 27SA₂ rRNAs (Kondo and Inouye, 1992; Kondo et al., 1992; Lee et al., 1992; Kressler et al., 1997). Rpl11b-GFP may accumulate in the nucleus of these strains because it binds to the 27SA₂ preribosomal particle after this major cleavage event (Kressler et al., 1999). Thus, we were able to detect defects in ribosome assembly mutants with the use of Rpl11b-GFP as a marker for the 60S ribosomal subunit. In contrast, Rpl25-GFP did not mislocalize in cells bearing a mutation in the 60S subunit assembly factor Nop1 (Hurt et al., 1999), and we were unable to detect a defect in mtr4-1 cells with the use of the Rpl25-GFP assay (our unpublished results).

The 60S ribosomal subunit accumulates in the nucleus of several yeast strains bearing mutations in nucleoporins. The nucleoporin mutants nup49-313, nic96-1, nsp1, nup82108, nup116-5, and nup120 cause Rpl11b-GFP to accumulate in the nucleus (Table 2). The 40S ribosomal subunit also accumulates in the nucleus of these strains (Moy and Silver, 1999; T. Moy and P. Silver, unpublished results), suggesting that these nucleoporins may be involved in the export of both ribosomal subunits. In addition, the yeast strains nup85, nup133, nup159, nup1, nup40, and gle1-4 all accumulated Rpl11b-GFP in the nucleus to various degrees (Table 2). These strains have also been shown to accumulate poly(A)⁺ RNA in the nucleus at nonpermissive growth temperatures (Bogerd et al., 1994; Doye et al., 1994; Gorsch et al., 1995; Li et al., 1995; Goldstein et al., 1996; Murphy et al., 1996; Murphy and Wente, 1996). Some of these nucleoporins may play a role in the export of both poly(A)⁺ RNA and 60S ribosomal subunits. Rpl11b-GFP accumulation in nup1-8, nup133, gle1-4, and nup159 cells may be due to a delay in ribosome assembly and not an export defect, because the 40S ribosomal subunit accumulated in the nucleolus of nup1-8, nup133, and gle1-4 cells (Moy and Silver, 1999) and rRNA processing was inhibited in nup159 cells (Del Priore et al., 1996; Goldstein et al., 1996).

Mutation of either Nup120 or Nup85, two members of the Nup84 nuclear pore subcomplex, causes nuclear accumulation of Rpl11b-GFP. The Nup84 subcomplex also contains Nup145-C, Seh1, and Sec13 (Siniossoglou et al., 1996, 2000). It has been proposed that this subcomplex plays a role in mRNA export because mutations in Nup120, Nup85, Nup84, and Nup145 trap poly(A)⁺ RNA in the nucleus (Aitchison et al., 1995a; Heath et al., 1995; Goldstein et al., 1996; Siniossoglou et al., 1996; Dockendorff et al., 1997). The 60S subunit accumulated in nup120 cells after a 1-h shift to 37°C, and a slight defect was also detected at the permissive temperature. It does not appear that nuclear accumulation of Rpl11b-GFP in nup120 cells is due to a defect in ribosome assembly, because the ratio of 60S to 40S subunits is similar to that of wild-type cells in all conditions assayed (our unpublished results). In nup85 cells, Rpl11b-GFP accumulates in the entire nucleus at 25°C and is concentrated in the nucleolus at 37°C, suggesting that both ribosome assembly and export are affected by this mutation. It is possible that the Nup120 and Nup85 portion of the Nup84 subcomplex plays a role in 60S ribosomal subunit export in addition to mRNA export.

Consistent with a role for Ran in ribosomal subunit export, mutation of the Ran regulators Rna1, Prp20, and Yrb1 resulted in nuclear accumulation of Rpl11b-GFP. The 40S ribosomal subunit also accumulates in the entire nucleus of rna1-1, prp20-1, and yrb1-1 cells (Moy and Silver, 1999). It is not clear whether Ran and its regulators play a direct role in ribosomal subunit export or if the defect in these mutants is due to a block in other transport processes. The importin  and importin  mutants srp1-31 and rsl1-4 also caused 60S subunit accumulation, presumably due to a delay in the import of factors required for 60S ribosomal subunit assembly or export.

The mutants rat8-1 and rat8-2 accumulated Rpl11b-GFP in the nucleus. Rat8 is a member of the DEAD box family of helicases that is localized in the cytoplasm and associates with the NPC in yeast cells (Snay-Hodge et al., 1998; Tseng et al., 1998; Hodge et al., 1999). A role for Rat8 in mRNA export has been proposed (Snay-Hodge et al., 1998). The 60S export defect in rat8 mutants was not due to a ribosome assembly defect, because processing of the 35S rRNA in rat8-1 cells was delayed only slightly and the ratio of 60S to 40S subunits as well as the polyribosome profiles of rat8-1 and rat8-2 cells were similar to those of wild-type cells (our unpublished results). We were also able to detect the 60S export defect in rat8-1 cells with the use of the Rpl25-GFP export assay described previously (our unpublished results). Rpl25-GFP accumulated in both the nucleolus and the nucleus of rat8-1 cells within 15 min of shifting cells from the nonpermissive growth temperature back to the permissive temperature (our unpublished results). It is possible that Rat8 is involved in both mRNA and 60S ribosomal subunit export. A role for a helicase in the release of mRNA or 60S ribosomal subunits from the cytoplasmic face of the NPC is an intriguing possibility.

A subset of karyopherin mutants mislocalize Rpl11b-GFP to the nucleus. Mutation of KAP104 and deletions of NMD5 and KAP123 all lead to nuclear accumulation of the 60S ribosomal subunit. Because Kap123 is a putative yeast r-protein importer (Rout et al., 1997) and both Nmd5 and Kap104 are homologous to mammalian r-protein importers (Jakel and Gorlich, 1998), it is possible that mutation of these factors slows r-protein import, which in turn inhibits ribosome subunit assembly. In contrast, mutation of Pse1 and Sxm1, two karyopherins implicated in r-protein import (Rosenblum et al., 1997; Rout et al., 1997), did not affect Rpl11b-GFP localization. Because ribosome biogenesis is essential for cell viability, it is likely that there are multiple r-protein importers in yeast. Mutation of the nuclear export sequence receptor Xpo1/Crm1 also mislocalizes Rpl11b-GFP to the nucleus, although the defect occurs in only 30% of cells. Because Rna1 accumulates in the nucleus of xpo1-1 cells at the nonpermissive temperature (Feng et al., 1999), it is possible that disruption of the RanGTP gradient causes the 60S mislocalization defect in these cells.

We have identified a potential role for the karyopherin Kap120 in the assembly or export of 60S ribosomal subunits in yeast. Deletion of KAP120 causes a strong accumulation of Rpl11b-GFP in the nucleus at 25°C (Figure 4B) and a large reduction in the level of mature 60S ribosomal subunits (Figure 6). The decrease in 60S ribosomal subunits is not simply due to a deficiency in the level of mature 25S rRNA because deletion of KAP120 only delays (does not block) processing of the 35S rRNA precursor (Figure 5). The 60S ribosomal subunit defect is also not caused by a block in mRNA export because poly(A)⁺ RNA is cytoplasmic in kap120 cells (our unpublished results). Deletion of KAP120 appears to affect only the 60S ribosomal subunit. The 40S subunit is not mislocalized in kap120 cells (Moy and Silver, 1999), and nearly wild-type levels of 40S ribosomal subunits are present (Figure 6). At both 37°C (Figure 4B) and at saturation (our unpublished results), conditions in which ribosome synthesis in the cell is reduced, Rpl11b-GFP is distributed throughout the cytoplasm of kap120 cells. This redistribution of Rpl11b-GFP indicates that deletion of KAP120 does not cause a complete block in assembly or export. There are several potential roles for Kap120 in ribosome biogenesis and export. Kap120 may be an importer for factor(s) involved in ribosomal subunit assembly, a r-protein importer, or a factor that recognizes fully assembled ribosomal subunits for export. We are currently investigating the role of Kap120 in ribosome assembly and export in more detail.