Degradation of Saccharomyces cerevisiae Transcription Factor Gcn4 Requires a C-Terminal Nuclear Localization Signal in the Cyclin Pcl5

S. cerevisiae strains and growth conditions.All yeast strains used in the present study are either congenic to the S. cerevisiae S288c (RH3239, RH1168, and RH1408) or W303 (RH3237, RH3238, RH3241, EY0140, RH3242, RH2701, RH2702, RH2703, RH2704, RH2706, RH2707, RH2708, RH2709, RH2710, and RH3058) genetic background. Details of the strains are given in Table 1. RH3242 was obtained by replacing the mutant his3-11 allele of yeast strain EY0140 by a wild-type HIS3 allele using BamHI linearized plasmid B1683 (Table 2). Strain RH3239 expressing GFP epitope-tagged version of PCL5 at endogenous levels was obtained by PCR-mediated gene tagging (24). Briefly, primers were designed specific for amplification of the yEGFP-kanMX4 module with homologous sequences to the PCL5 3′ end using plasmid pYM12⁶ as a template. The PCR product was transformed directly into yeast strain RH1168 and plated onto rich medium with 200 μg of G418 (Geneticin; Gibco)/ml. Transformants were replica plated onto the same medium. The GFP-tagged PCL5 version was confirmed by Southern hybridization. Standard methods for genetic crosses and transformation were used (19).

The strains were grown in standard yeast extract-peptone-dextrose (1% yeast extract, 2% peptone, 2% dextrose) and minimal yeast nitrogen base (YNB) medium (1.5 g of yeast nitrogen base lacking amino acids and ammonium sulfate/liter, with 5 g of ammonium sulfate/liter and 2% dextrose, galactose, or raffinose and supplemented with the appropriate amino acids).

Plasmid constructions.The plasmids used in the present study are listed in Table 2, and details of important primers are given in Table S1 in the supplemental material. Plasmid pME2844 expressing PCL5-GFP was obtained by amplifying the PCL5-open reading frame (ORF) with Pfu polymerase and introducing it via SmaI/ClaI into the low-copy-number GFP-C-Fus vector (pME2843) (38). pME2846 expressing PCL5-GFP was constructed by amplifying the PCL5-ORF with Pfu polymerase and cloning it as a SmaI/HindIII fragment into p426MET25 (37). Afterward, a 750-bp BglII-fragment encoding the GFPuv variant of GFP that was amplified from plasmid pBAD-GFP (Clontech, Heidelberg, Germany) was inserted behind the PCL5-ORF.

The construction of plasmids pME2849, pME2850, pME2851, pME2853, pME2854, pME2855, pME2856, pME2857, pME2858, pME2859, pME3370, pME3371, and pME3574 expressing GFP fused to the 3′ end of different PCL5 fragments driven from the MET25 promoter was started by amplifying the GFP-ORF as described before and introducing it via SmaI/ClaI into p426MET25. PCL5 fragments were fused via SpeI/SmaI to the 5′ end of GFP. pME2950 was constructed by amplifying PCL5_bp610-654-GFP (encoding Pcl5_aa204-218-GFP) from plasmid pBAD-GFP by using primers KB57 and KB16 (see Table S1 in the supplemental material) and cloning them via SmaI/ClaI into p426MET25 (37).

Plasmid pME3573 encoding Pcl5_aa1-180-GFP was constructed via amplification of the PCL5-GFP fragment from pME2853 and introducing it as a SmaI/ClaI fragment into p416MET25 (37).

The plasmid pME2951 was obtained by amplifying GFP-ARO7-PCL5_bp619-645 (encoding GFP-Aro7-Pcl5_aa207-215) with Pfu polymerase using the primers RP14 and KB55 (see Table S1 in the supplemental material) and introducing it via SmaI/EcoRI into p426MET25 (37).

The plasmid pME2860 was constructed by amplifying the hybrid ORF PHO80_bp1-219-PCL5_bp235-534-PHO80_bp508-882 (encoding Pho80_aa1-73-Pcl5_aa79-178-Pho80_aa170-294) with Pfu polymerase from plasmid KB1360 and fused it via SpeI/SmaI to the 5′ end of GFP on plasmid pME2849. To obtain pME2948, the hybrid PHO80_bp1-219-PCL5_bp235-534 (encoding Pho80_aa1-73-Pcl5_aa79-178) was amplified via Pfu polymerase from plasmid pME2860 and ligated as SpeI/SmaI fragments in front of the GFP-ORF of pME2849. The plasmid pME3576 was constructed in two steps. First, the PCL5 promoter (coordinates −890 to −3) was amplified with Phusion polymerase as SacI/XbaI fragment using the primers oFLS95 and oFLS96 (see Table S1 in the supplemental material) and ligated in SacI/XbaI-digested pRS416MET25, leading to pME3575. Next, the hybrid ORF PHO80_bp1-219-PCL5_bp235-534-GFP was amplified with the primers oFLS97 and oFLS98 (see Table S1 in the supplemental material) from plasmid pME2948 and inserted as SpeI/ClaI fragment into XbaI/ClaI-digested pME3575.

Construction of plasmids pME3370, pME3577, and pME3578 was started by site-directed mutagenesis of the PCL5-ORF via KOD polymerase using the primers oFLS35 and oFLS36 (see Table S1 in the supplemental material) resulting in the amino acid substitutions K209A, R210A, and R212A. The mutated PCL5***-ORF and the mutated fragments Pcl5_aa61-229*** and Pcl5_aa181-229*** were introduced as a SpeI/SmaI fragment in front of the GFP-ORF of plasmid pME2849.

To obtain pME3371, the Pcl5_aa61-180-encoding fragment was amplified with Pfu polymerase using the primers KB35 and KB100 (see Table S1 in the supplemental material) and ligated as an SpeI/SmaI fragment in front of the GFP-ORF of pME2849.

In the PCL5-GFP-NES fusion (pME2861), the nuclear export sequence (NES) from PKI (inhibitor of the cAMP-dependent protein kinase) (54) was engineered at the 3′ end of PCL5-GFP in two steps. The PCL5-ORF was amplified with Pfu polymerase and inserted into p426MET25 as a SmaI/HindIII fragment as the first step. Second, the GFP-NES cassette was amplified using the primers KB5 and KB40 (encoding GFP-NESaa GMDELYKNELALKLAGLDINKTKLTLA) (see Table S1 in the supplemental material) and introduced as a BglII fragment at the 3′ end of PCL5.

To construct a fusion of glutathione S-transferase (GST) with full-length Kap95, the KAP95 coding region was amplified by PCR using Pfu polymerase and the primers Kap95-BamHI-START and Kap95-STOP-XbaI (see Table S1 in the supplemental material) and inserted as a BamHI/XbaI fragment in-frame downstream of the GST-ORF into pYGEX-2T to yield pME3372. To obtain plasmid pME3447, the KAP104-ORF was amplified by PCR using the primers oFLS74BamHI-KAP104 and oFLS75KAP104-SpeI (see Table S1 in the supplemental material) and inserted as a BamHI/SpeI fragment in-frame downstream of the GST-ORF into pYGEX-2T (46). Plasmid pME3448 expressing GST-PSE1 was similarly constructed using the oligonucleotides oFLS76BamHI-PSE1 and oFLS77PSE1-SpeI (see Table S1 in the supplemental material). To receive plasmid pME3579 expressing a ninefold myc epitope-tagged fusion of the NLS-mutated PCL5***-ORF (containing the amino acid substitutions K209A, R210A, and R212A), the mutated PCL5***-ORF was amplified with Pfu polymerase and inserted as a SmaI/HindIII fragment into p425GAL1. A 360-bp BglII fragment carrying myc⁹ was introduced into a BglII site in front of the third amino acid of Pcl5***. Similarly, plasmids pME3580 encoding myc⁹-Pcl5_aa181-229 and pME3581 encoding myc⁹-Pcl5_aa181-229*** with mutated NLS motif were constructed. After amplification of the corresponding fragment with Pfu polymerase and insertion as a SmaI/HindIII fragment into p425GAL1 a 360-bp BglII fragment encoding myc⁹ was introduced after the first amino acid of the respective Pcl5 fragment.

To express KAP95, the ORF was amplified as a BamHI/SalI fragment and introduced into BamHI/SalI-digested p425MET25 (37) to yield pME3583.

Plasmid pME2848, expressing a triple myc epitope-tagged version of GCN4 under the control of the GAL1 promoter, was obtained by amplifying GCN4 with Pfu polymerase and subsequent insertion in p415GAL1 (37) as blunt/HindIII fragment. A 120-bp BamHI fragment carrying the triple myc epitope was inserted into a BglII restriction site after the fifth amino acid of Gcn4. To construct plasmid pME3572, the triple myc epitope-tagged version of GCN4 was amplified as a SpeI/ClaI fragment from pME2848 using Pfu polymerase and inserted into p414GAL1 (37).

Protein analysis. (i) Shutoff-Western procedure.Yeast cells were pregrown in selective minimal medium with glucose as carbon source. Cells were collected by centrifugation and incubated in minimal medium containing 2% galactose to express myc³-GCN4 from the GAL1 promoter. After 3 h of induction, 2% glucose was added to shut off the GAL1 promoter. Samples were analyzed at the indicated time points after promoter shutoff (0-min time point).

(ii) Purification of GST fusions.Yeast strains expressing GST, GST-PHO85, GST-KAP95, GST-KAP104, or GST-PSE1, together with a myc-tagged version of PCL5, were pregrown in selective minimal medium containing raffinose as the carbon source. Two percent galactose was added to induce the expression of the GAL1-driven fusions. After 3 h of induction, cells were collected by centrifugation, and protein extracts were prepared and incubated with glutathione-Sepharose overnight at 4°C. The beads were repeatedly washed and collected to purify GST fusions and any associated proteins (3). Samples were denatured in sodium dodecyl sulfate (SDS) loading dye (65°C, 15 min) and analyzed by Western hybridization.

(iii) Whole-cell extracts of S. cerevisiae.Extracts were prepared from yeast cultures grown to exponential phase. Cells were washed in ice-cold buffer B (50 mM Tris-HCl [pH 7.5], 1 mM EDTA, 50 mM dithiothreitol), lysed with glass beads in 200 μl of buffer B plus PIM (phenylmethylsulfonyl fluoride, tosyl-l-lysine-chloromethylketone, tosyl-l-phenylalanine-chloromethylketone, p-aminobenzamidine-HCl, and o-phenanthroline [1 mM each]) plus 3% Triton X-100 plus 0.8% SDS at 4°C, and spun at 3,500 rpm for 15 min to remove glass beads and large cell debris. Extracts (10 μl) were removed to determine the total protein concentration using a protein assay kit from Bio-Rad (Munich, Germany). Proteins were denatured in SDS loading dye by heating at 65°C for 15 min. Proteins were subjected to SDS-polyacrylamide gel electrophoresis and transferred to nitrocellulose membranes. GFP and myc fusion proteins, Cdc28, and eIF-2 were detected using enhanced chemiluminescence technology (Amersham, United Kingdom). For the first incubation, monoclonal mouse anti-GFP (Clontech), monoclonal mouse anti-myc (9E10), polyclonal rabbit anti-GST (Santa Cruz Biotechnologies, Santa Cruz, CA), polyclonal rabbit anti-Cdc28, or polyclonal rabbit anti-eIF-2 antibodies were used. Peroxidase-coupled goat anti-rabbit or goat anti-mouse immunoglobulin G was used as a secondary antibody (Dianova, Hamburg, Germany).

GFP fluorescence microscopy.Yeast strains harboring plasmids encoding Pcl5-GFP fusion proteins were grown to early log phase and analyzed under sated and starved conditions. Amino acid starvation was induced by transferring these leu2-deficient yeast cells from minimal medium containing leucine to minimal medium lacking leucine for 1 h. Cells from 1-ml portions of the cultures were harvested by centrifugation and immediately viewed in vivo on a Zeiss Axiovert microscope by either differential interference contrast microscopy (DIC) or fluorescence microscopy using a GFP filter set (AHF Analysentechnik AG, Tübingen, Germany) or, in the case of DAPI (4′,6′-diamidino-2-phenylindole) staining, a standard DAPI filter set. DAPI staining was used for visualization of nuclei. Cells were photographed using a Hamamatsu-Orca-ER digital camera and Improvision Openlab software (Improvision, Coventry, United Kingdom).

Growth tests.For spot dilution assays, yeast strains were precultured to the same optical densities (optical density at 600 nm [OD₆₀₀] = 1) and spotted onto appropriate YNB media, as indicated. Tenfold dilutions, starting with 3 × 10⁴ cells per 20 μl, were spotted onto the plates, incubated for 3 days at 30°C, and photographed under white light. For assaying growth in liquid culture, overnight cultures of the yeast strains were pregrown in YNB medium with raffinose as the carbon source, gently harvested, washed with sterile water, and diluted in 100 ml of the appropriate selective medium with either glucose or galactose as the carbon source to a final OD₆₀₀ of 0.1. Cells were incubated with shaking at 30°C. The OD₆₀₀ was measured every hour. A graph was plotted by using OD₆₀₀ values, and the doubling times during logarithmic growth were calculated.

Pcl5 is a nuclear protein.The CDK Pho85 and the Pcl5 target Gcn4 are localized in the nucleus (45). Therefore, we reasoned that the cyclin Pcl5 has to be transported in the nucleus at least in sated cells, when Gcn4 is rapidly degraded. Since Pcl5 is itself an unstable protein with a half-life of <3 min (52), monitoring of the subcellular localization of the protein is a challenge. Overexpression of Pcl5 carrying an epitope tag at the C terminus stabilizes the protein in contrast to a more physiological unstable Pcl5 with an N-terminal tag (2). When PCL5 is expressed from its endogenous promoter, no stabilization was observed using a C-terminal epitope tag (2). We analyzed Pcl5 fusions to GFP at low and high levels of expression by cloning C- or N-terminally GFP-tagged versions of Pcl5 on both high- and low-copy vectors under the repressible MET25 promoter to increase expression levels and to study Pcl5 localization independently of its expression. The effect of native levels of expression and regulation of PCL5 was tested after chromosomal integration of the ORF for GFP at the 3′ terminus of the PCL5-ORF.

Expression of the different Pcl5-GFP fusion proteins was verified by Western hybridization of S. cerevisiae cell extracts using monoclonal anti-GFP antibodies (data not shown). The functionality of all Pcl5-GFP hybrids was tested by their ability to suppress Gcn4 overexpression toxicity. High overexpression of GCN4 inhibits cellular growth, possibly by the interference of Gcn4 with other transcriptional activation pathways (55). A pcl5-deficient yeast strain is hypersensitive to even moderately overexpressed GCN4 fused to the GAL1 promoter (52). Whereas GFP by itself is unable to suppress overexpression toxicity of Gcn4 on solid medium, as well as in liquid culture, all Pcl5 fusions tested carrying either C-terminal (see Fig. 2B) or N-terminal GFP (data not shown) complemented the pcl5 mutant phenotype, indicating that the addition of GFP does not influence the localization of Pcl5 or interfere with kinase activation or substrate recognition. A GFP fusion of PCL5, integrated in a single copy at the original locus, is already sufficient to suppress Gcn4 toxicity (data not shown).

Localization of Pcl5 was monitored by fluorescence microscopy. Pcl5-GFP expressed from the endogenous PCL5 locus, from the MET25 promoter on a centromeric (CEN) plasmid or from the MET25 promoter on a high-copy (2μm) plasmid were all predominantly localized within the nucleus, which was confirmed by DAPI staining (Fig. 1A). Similar results were obtained with N-terminal GFP-fusions of Pcl5 (data not shown). A quantitative analysis of transformed cells (n = 200) revealed that 60 to 80% of the cells expressed Pcl5-GFP were visible in the nucleus depending on promoter strength and copy number. The occurrence of cells where Pcl5-GFP is not detectable decreases when PCL5-GFP is expressed from the MET25 promoter on a high-copy (2μm) plasmid, but overexpression does not alter the localization.

• Open in new tab
• Download powerpoint

FIG. 1.

Pcl5-GFP is transported into the S. cerevisiae nucleus. (A) Nuclear localization of high and low amounts of the functional cyclin Pcl5-GFP. Localization was analyzed in S. cerevisiae cells (RH3239) expressing Pcl5-GFP fusion protein derived from its endogenous PCL5 promoter at the chromosomal locus or pcl5 mutant cells (RH3238) expressing Pcl5-GFP fusion protein from the MET25 promoter on a low-copy (CEN) plasmid (pME2844) or on a high-copy (2μm) plasmid (pME2846). Cells were grown to early log phase at 30°C and analyzed by DIC microscopy, DAPI staining, and fluorescence microscopy (GFP). “X/200” in parentheses means “X” number of 200 cells expressing PCL5-GFP detectable in the nucleus. (B) Nuclear localization of Pcl5-GFP is independent of the availability of amino acids or the proteins Gcn4, Pho85, or Pho81. Nuclear localization of Pcl5-GFP expressed from the MET25 promoter on a high-copy plasmid (pME2846) in pcl5 (RH3238) under amino acid starvation conditions (−Leu), pho85 (RH3242), gcn4 (RH1408), and pho81 (RH3241) mutant strains. Yeast cells were grown and analyzed as described above. Leucine starvation (−Leu) was induced in leucine-auxotrophic yeast cells by shifting to synthetic medium lacking leucine.

Constitutive nuclear localization of Pcl5 is independent of amino acid starvation and the known interacting proteins Gcn4, Pho85, and Pho81.Amino acid starvation results in Gcn4 stabilization. We analyzed whether Gcn4 stability regulation may be caused by an altered localization of Pcl5 dependent on the availability of amino acids. Pcl5-GFP localization was analyzed in leucine-auxotrophic yeast cells briefly shifted to synthetic medium lacking leucine, conditions that cause the stabilization of Gcn4 in PCL5 wild-type cells (26). Promoter shutoff experiments with a pcl5 strain expressing GAL1-driven myc³-GCN4 from the low-copy plasmid pME3572, together with PCL5-GFP under the control of the MET25 promoter from the high-copy plasmid pME2846, revealed that under leucine starvation conditions Gcn4, if at all, is only slightly stabilized (data not shown). Nevertheless, Fig. 1B demonstrates that Pcl5 is constitutively located in the nucleus in leucine-starved cells.

The amino acid sequence of Pcl5 contains neither a monopartite basic NLS as in the simian virus 40 (SV40) large tumor antigen (23) nor the bipartite motif of nucleoplasmin (46). Therefore, we examined whether Pcl5 is indirectly imported into the nucleus by interaction with another nuclear protein. The CKI inhibitor Pho81 is involved in amino acid-dependent stabilization of Gcn4, and both Pho85 and Pho81 are able to interact with the cyclin Pcl5 (3). Gcn4, Pho85, and Pho81 are exclusively nuclear proteins independently of the availability of amino acids (45). We tested whether the substrate Gcn4, the kinase Pho85, or the inhibitor Pho81 are required for Pcl5 nuclear import by analyzing the localization of Pcl5-GFP in the corresponding S. cerevisiae mutant strains. The cyclin was detected predominantly in the nucleus in all tested mutant strains (Fig. 1B) without any significant differences in frequency compared to the GCN4 PHO85 PHO81 wild-type strain shown in Fig. 1A. This indicates that nuclear import of Pcl5 does not require these Pcl5 interacting partners and suggests that Pcl5 possesses an autonomous NLS for direct interaction with karyopherins.

Nuclear localization of yeast Pcl5 is required for Gcn4 degradation.Previous experiments have suggested that phosphorylation and ubiquitination, which are the first two steps of the Gcn4 degradation pathway, are restricted to the nucleus (45). Since Pcl5 is a predominantly nuclear protein (see Fig. 1, above), we wanted to verify that nuclear localization of the Gcn4-specific cyclin Pcl5 is a prerequisite for efficient Gcn4 degradation in sated cells.

The localization of Pcl5-GFP was manipulated by fusing an NES to the C terminus of the Pcl5-GFP hybrid. The used NES sequence derived from the polypeptide inhibitor (PKI) of the cyclic AMP-dependent protein kinase. It represents a short and hydrophobic motif with high leucine content (LALKLAGLDI) (39). Western analysis confirmed that this hybrid, expressed under the MET25 promoter, migrated at the expected size of 54 kDa (Fig. 2A). Quantitative fluorescence microscopy revealed that, similar to PCL5-GFP 80% of the transformed cells expressed PCL5-GFP-NES but in contrast to the nuclear localization of the Pcl5-GFP fusion, the Pcl5-GFP-NES was predominantly localized in the cytoplasm (Fig. 2B). Expression of PCL5-GFP-NES from the stronger GAL1 promoter also resulted in a clear cytoplasmic localization (data not shown).

• Open in new tab
• Download powerpoint

FIG. 2.

Nuclear localization of yeast Pcl5 is required for Gcn4 degradation. (A) PCL5-GFP-NES is correctly expressed as 54-kDa equivalent to PCL5-GFP. Yeast pcl5 cells (RH3238) were transformed to express either PCL5-GFP (pME2846) or PCL5-GFP-NES (pME2861) from the high-copy plasmids under the control of the MET25 promoter. Cells were grown to early log phase, and the expression of the GFP fusion proteins was analyzed by Western blotting with monoclonal anti-GFP antibodies. (B) Pcl5-GFP-NES is transported out of the nucleus into the cytoplasm. Localization of the fusion proteins Pcl5-GFP (pME2846) and Pcl5-GFP-NES (pME2861) expressed under the control of the MET25 promoter on high-copy (2μm) plasmid was analyzed in a pcl5 mutant strain (RH3238) by fluorescence microscopy (GFP), DAPI staining, and DIC microscopy. The numbers in parentheses represent the total amount of cells investigated (n = 200) and the respective amount showing the displayed localization. (C) Pcl5-GFP-NES is incapable of suppressing the toxicity of overexpressed GCN4 in the absence of a functional PCL5 gene. Wild-type cells (RH3237) and pcl5 mutant cells (RH3238) expressing GAL1-driven myc³-GCN4 from the low-copy plasmid pME2848 alone or together with GFP (pME2849), PCL5-GFP (pME2846), or PCL5-GFP-NES (pME2861) under the control of the MET25 promoter on a high-copy plasmid were spotted in 10-fold dilutions on glucose (repressing conditions) and galactose (inducing conditions) to induce expression of GCN4 driven by the GAL1 promoter. As controls, wild-type and pcl5 mutant cells transformed with the empty vector were used (control). The given doubling times are the average of at least three independent growth tests in liquid culture with glucose (repressing conditions) or galactose (inducing condition) as a carbon source. (D) Pcl5-GFP-NES is unable to induce Gcn4 degradation. The same wild-type and pcl5 yeast strains as in panel C were transformed to express myc³-GCN4 from plasmid pME2848 alone or together with PCL5-GFP (pME2846) or PCL5-GFP-NES (pME2861). Protein levels of myc³-Gcn4 or Cdc28 were determined at the indicated time points after the shift to glucose medium. The numbers given below each lane indicate the remaining Gcn4-percentage compared to Cdc28 as an internal standard.

The activity of the Pcl5-GFP-NES construct was first tested by assaying the ability of this protein to suppress Gcn4 overexpression toxicity as described previously. The Gcn4 toxicity assay revealed that, in contrast to nuclear Pcl5-GFP, the activity of cytoplasmic Pcl5-GFP-NES is decreased in the suppression of Gcn4-mediated growth inhibition (Fig. 2C). The activity of Pcl5-GFP-NES was further tested by assaying its ability to promote Gcn4 degradation. A pcl5 strain was transformed with a GAL-myc³-GCN4 plasmid and either Pcl5-GFP or Pcl5-GFP-NES. A promoter shutoff experiment of the GAL-myc³-GCN4 construct revealed that Pcl5-GFP efficiently complemented the Gcn4 degradation defect of the pcl5 mutant. In contrast, expression of the cytoplasmic Pcl5-GFP-NES was unable to complement the pcl5 mutant, indicating that Pcl5-GFP-NES is unable to degrade Gcn4 (Fig. 2D). This suggests that nuclear localization of Pcl5 is required for efficient Gcn4 degradation.

The C-terminal part of Pcl5 directs nuclear localization.To determine which part of Pcl5 is responsible for its nuclear localization, we constructed a set of Pcl5-GFP fusions. The first 60 N-terminal amino acids of Pcl5 are followed by the central part of the protein (amino acids [aa] 61 to 180) containing the predicted cyclin box domain (see Fig. 6). This central domain, which is conserved between the different Pho85 cyclins (32), is followed by an additional 50 aa at the C terminus. The subcellular localization of a set of truncated Pcl5-GFP hybrids was analyzed to identify the Pcl5 cis-acting sequences responsible for nuclear import (Fig. 3A and 6). First, all fusion constructs were driven from the repressible MET25 promoter on a 2μm plasmid to facilitate visualization of truncated Pcl5-GFP hybrids. Expression of Pcl5-GFP derivatives was verified by Western analysis of pcl5 mutant cell extracts using monoclonal anti-GFP antibodies (Fig. 3B).

• Open in new tab
• Download powerpoint

FIG. 3.

The C-terminal part of Pcl5 directs nuclear localization. (A) Yeast pcl5 mutant strain RH3238 was transformed to express either GFP alone (pME2849) or in N-terminal fusion with Pcl5_aa1-229 (pME2846), Pcl5_aa1-95 (pME2850), Pcl5_aa1-127 (pME2851), Pcl5_aa1-180 (pME2853), Pcl5_aa61-229 (pME2854), Pcl5_aa111-180 (pME2855), Pcl5_aa111-229 (pME2856), Pcl5_aa153-229 (pME2857), Pcl5_aa181-229 (pME2859), Pcl5_aa204-218 (pME2950), Aro7-Pcl5_aa207-215 (pME2951), Pcl5_aa181-229*** (with the mutated NLS motif marked by asterisks) (pME3578), Pcl5_aa1-229*** (pME3577), Pcl5_aa61-229*** (pME3370), Pcl5_aa61-180 (pME2858), Pcl5_aa79-178 (pME3574), Pho80_aa1-73-Pcl5_aa79-178-Pho80_aa170-294 (pME2860), or Pho80_aa1-73-Pcl5_aa79-178 (pME2948) from the MET25 promoter on a high-copy plasmid. In addition, Pho80_aa1-73-Pcl5_aa79-178 was expressed from a low-copy plasmid under the control of the PCL5 promoter (pME3576). GFP signals were analyzed by fluorescence microscopy (GFP) or DIC microscopy. Nuclear localization is indicated by white colored arrows. An arrowhead only marks plasma membrane localization. Accumulation in the cytoplasm is highlighted by a rendered blank arrow. A dashed arrow is used in the case of cytoplasmic aggregates. The differences in sizes of the cells shown are due to different enlargement. “X/200” in parentheses represents the fraction “X” of 200 cells displaying the marked staining pattern. (B) Yeast pcl5 cells transformed with the same high-copy plasmids as in panel A beside pME2849 encoding GFP alone and the low-copy plasmid pME3576 expressing Pho80_aa1-73-Pcl5_aa79-178 were grown to early log phase, and the expression of the GFP fusion proteins was analyzed by immunoblotting with monoclonal anti-GFP antibodies. Protein extracts of the cells are blotted in the same order as in panel A (left panel, top to bottom; right panel, top to bottom).

Localization of GFP and the different Pcl5-GFP fusion proteins was examined by fluorescence microscopy in sated pcl5 cells. The native GFP protein by itself was found throughout the cell in both the nucleus and the cytoplasm, whereas Pcl5-GFP displayed a distinctly nuclear localization, as shown above (Fig. 1 and 3A). In contrast, N-terminal sequences of Pcl5 extending between positions 1 and 95 or between positions 1 and 127 showed a very different staining pattern, displaying a pericellular staining distinctive of plasma membrane localization (Fig. 3A). This indicates that the N-terminal region of Pcl5, extending from residues 1 to 95, is associated with the plasma membrane. Pcl5-GFP constructs lacking the N- or C-terminal domain (Pcl5_aa1-180 and Pcl5_aa61-229) showed predominantly nuclear staining, but cytoplasmic and membrane dot-like staining was also observed (Fig. 3). Expression of the Pcl5 aa 1 to 180 from a low-copy (CEN) plasmid did not change its localization (data not shown). The conserved central part of Pcl5 (aa 61 to 180) and the contained cyclin box (aa 79 to 178) are visualized via GFP only in an aggregate-like form spread over the whole cell (Fig. 3A) even when a classical SV40 large tumor antigen NLS is fused to the C terminus (data not shown). The C-terminal domain of the protein, either by itself (Pcl5_aa181-229) or with part of the cyclin box (Pcl5_aa111-229 and Pcl5_aa153-229) efficiently targeted the fusion protein to the nucleus (Fig. 3A). These data suggest that the carboxy-terminal 49 aa residues of Pcl5 include an NLS.

The C terminus of Pcl5 contains a PVKRPRESD motif between positions 207 and 215. The structure of this motif resembles the c-myc NLS (PAAKRVKLD), where the proline and aspartic acid residues flanking the basic cluster play an important role in nuclear import (31). We tested the function of this putative NLS motif by mutating the basic cluster to alanine (PVAAPAESD). Pcl5_aa1-229***-GFP containing the mutated motif shows nuclear but also cytoplasmic staining, with dot-like aggregates similar to Pcl5_aa1-180-GFP and Pcl5_aa61-229-GFP. Pcl5 aa 61 to 229 lacking the N terminus are sufficient for nuclear localization. Therefore, we analyzed the effect of the mutated putative NLS motif of this Pcl5 fragment on localization. Figure 3A shows that this Pcl5_aa61-229***-GFP fusion displayed a predominant cytoplasmic staining with additional membrane dot-like structures, suggesting that the putative NLS motif plays an active role for nuclear localization. Furthermore, we analyzed the localization of the C-terminal domain containing the mutated cluster fused to GFP. This Pcl5_aa181-229***-GFP is, in contrast to the nonmutated C-terminal domain, found in the cytoplasm, as well as in the nucleus, similar to GFP alone (Fig. 3A), corroborating the importance of the putative NLS motif. To test whether this putative NLS motif is sufficient to direct nuclear import, we constructed fusions encompassing this motif N and C terminally to two different reporters. The sequence including aa 204 to 218 of Pcl5 was fused to the N terminus of GFP (pME2950), and the Pcl5_aa207-215 sequence was fused to C terminus of GFP-Aro7 (pME2951), an exclusively cytoplasmic protein (45). Neither of the constructs was found in the nucleus (Fig. 3A). Together, these data indicate that aa 207 to 215 of Pcl5 are necessary but not sufficient for nuclear Pcl5 localization.

Pcl5 nuclear transport requires the importin Kap95.We analyzed the subcellular localization of Pcl5-GFP in a set of S. cerevisiae mutant strains that are defective in particular β-importins. Yeast strains deleted for the nonessential importin genes kap114, kap123, nmd5, pdr6, and sxm1 were cultured and analyzed at a temperature of 30°C. Strains carrying the temperature-sensitive mutations kap95, mtr10, kap104, pse1, or pse1/kap123 were analyzed at the permissive temperature of 20°C and the restrictive temperature of 30°C. Localization studies showed that Pcl5-GFP was nuclear in the kap114, kap123, nmd5, pdr6, sxm1, mtr10, kap104, pse1, or pse1/kap123 mutant strains. A cytoplasmic accumulation of Pcl5-GFP was observed in the kap95 mutant cells at their restrictive temperature of 30°C, suggesting that Kap95 is necessary for nuclear localization of Pcl5 (Fig. 4A). Transformation of kap95 mutant cells with KAP95 under the control of the MET25 promoter on a 2μm plasmid restored nuclear localization of Pcl5-GFP at the restrictive temperatures of 30 and 37°C (Fig. 4A). The high expression of KAP95 results also in a 30% subpopulation of yeast cells, which showed Pcl5-GFP aggregate formation similar to kap95 cells.

• Open in new tab
• Download powerpoint

FIG. 4.

The mutation of the β-importin encoding KAP95 prevents the import of Pcl5 into the nucleus. (A) Nuclear import of a functional Pcl5-GFP fusion protein expressed from the high-copy plasmid pME2846 under the control of the MET25 promoter was analyzed in five temperature-sensitive importin mutant strains by fluorescence microscopy (GFP) and DIC microscopy. Pcl5 translocation is not affected in the mutant strains mtr10 (RH2701), kap104 (RH2702), pse1 (RH2703), or pse1/kap123 (RH2706), whereas the kap95 (RH2704) mutation impairs the uptake of Pcl5-GFP at the restrictive temperature of 30°C. Expression of KAP95 under the control of the MET25 promoter on a high-copy plasmid (pME3583) in kap95 mutant cells (RH2704) restores the uptake of Pcl5-GFP to the nucleus (∼ 40%) at 30 and 37°C. Furthermore, five mutant strains with the nonconditional importin mutations kap114 (RH3058), kap123 (RH2707), nmd5 (RH2708), pdr6 (RH2709), and sxm1 (RH2710) were examined for subcellular localization of Pcl5-GFP by fluorescence microscopy. Nuclear import of Pcl5 was unaffected in all five mutant strains and was indistinguishable from that in the wild-type (wt) control (RH3237). (B) Protein-protein interaction of Pcl5 with Kap95. Yeast pcl5 mutant strain RH3238 was transformed to express either myc⁹-PCL5 (pME2865) with GST (GST on pYGEX-2T), GST-KAP104 (pME3447), or GST-PSE1 (pME3448) as negative controls or myc⁹-PCL5 (pME2865), together with GST-PHO85 (pME2866), as a positive control. In addition, yeast strain RH3238 was transformed to express myc⁹-PCL5 (pME2865), together with GST-KAP95 (pME3372). Protein levels of the fusion proteins were determined by Western blotting with rabbit anti-GST and mouse anti-myc antibodies before and after incubation with glutathione-Sepharose. The left part (Input) represents the GST, GST-Kap95, GST-Kap104, GST-Pse1, and myc⁹-Pcl5 before glutathione-Sepharose incubation to ensure that the initial protein extracts contain similar amounts of the fusion proteins. On the right (Beads) the elutions of the glutathione beads are shown. The GST fusion proteins are marked with an arrow in the respective lane of the elutions. (C) Mutation of the C-terminal NLS motif in Pcl5 does not impede the interaction to Kap95. This is the same experiment as shown in panel B. Yeast pcl5 mutant strain RH3238 was transformed to express either myc⁹-PCL5 (pME2865) together with GST (GST on pYGEX-2T), GST-PHO85 (pME2866), or GST-KAP95 (pME3372). In addition, pcl5 mutant cells (RH3238) were transformed to express myc⁹-PCL5*** with a mutated NLS motif (pME3577), together with GST (GST on pYGEX-2T) or GST-KAP95 (pME3372). Input (left part) represents the initial protein extracts before glutathione-Sepharose incubation. On the right side (Beads) the elutions of the beads are shown.

Next, we wanted to know whether there is a direct interaction between Kap95 and Pcl5. Kap95/Pcl5 interaction was investigated by an in vivo coprecipitation assay under conditions when cellular Gcn4 is unstable. GST-KAP95 and myc⁹-PCL5 were expressed from the GAL1 promoter. The protein fusions were induced and purified with glutathione beads to isolate the GST fusion and its associated proteins. Figure 4B shows that myc⁹-Pcl5 copurifies with GST-Kap95 under the tested condition. To exclude that the physical interaction of Pcl5 and Kap95 is an overexpression effect, we tested the two other importin fusions, GST-PSE1 and GST-KAP104, concerning their interaction with the cyclin Pcl5. Neither GST-Pse1 nor GST-Kap104 copurified myc⁹-Pcl5 in comparison to GST-Kap95 (Fig. 4B). We also analyzed the interaction of GST-Kap95 and Pcl5*** fragments with an altered C-terminal NLS motif. myc⁹-Pcl5_aa1-229*** copurifies with GST-Kap95 like the nonmutated myc⁹-Pcl5 (Fig. 4C), whereas neither myc⁹-Pcl5_aa181-229 nor myc⁹-Pcl5_aa181-229*** interacted with GST-Kap95 under the tested conditions (data not shown). This suggests that a tight interaction between Kap95 and Pcl5 requires more than the C-terminal domain of Pcl5. The N-terminal myc⁹-tag might impair binding of the C-terminal domain alone.

In summary, we have shown that Kap95 is required for accurate nuclear import of the yeast cyclin Pcl5 by a physical interaction.

Analysis of the activity of the Pcl5-GFP hybrids.We used a set of different truncated Pcl5-GFP versions described above to define the specificity domain for the Pho85/Pcl5 complex, which promotes Gcn4 phosphorylation and subsequent degradation. The function of all Pcl5-GFP hybrids was first tested in the Gcn4 overexpression toxicity assay on solid medium as well as in liquid culture. We found that in the pcl5 background the toxicity of GCN4 moderately overexpressed from the GAL1 promoter can only be suppressed by Pcl5_aa61-229 or Pcl5_aa1-180 (Fig. 5A), but not by any of the other truncated Pcl5-GFP hybrids. For a more detailed analysis, i.e., to determine whether the truncated Pcl5 versions are able to promote Gcn4 degradation, we tested the effect of the truncated PCL5-GFP hybrids on Gcn4 degradation. Wild-type PCL5 and pcl5 mutant cells were transformed to express GAL-myc³-GCN4, as well as the truncated Pcl5-GFP fusions. Promoter shutoff experiments of myc³-Gcn4 revealed a rapid degradation of this protein when Pcl5 aa 61 to 229 or aa 1 to 180 are expressed (Fig. 5B). The N-terminal domain with the central part (aa 1 to 180) expressed from a low-copy (CEN) plasmid is still able to suppress GCN4 overexpression toxicity and degrades efficiently Gcn4 (Fig. 5B). The central domain of Pcl5 (aa 61 to 180) containing the cyclin box domain (aa 79 to 178) (shown schematically in Fig. 6) is unable to direct nuclear import on its own and showed no Pcl5 function in the Gcn4 toxicity assay (Fig. 3A and 5A). In agreement, Gcn4 is highly stable when PCL5 encoding aa 61 to 180 is present by simultaneous expression (Fig. 5B). Similar results were observed for Pcl5_aa79-178-GFP (Fig. 3A and data not shown). A fusion of nuclear localized Pcl5 aa 61 to 180 to SV40 large tumor antigen NLS motif (pME3371) was not sufficient to restore nuclear localization and subsequently Gcn4 degradation (data not shown).

• Open in new tab
• Download powerpoint

FIG. 5.

Functional analysis of Pcl5-Pho80 hybrids fused to GFP. (A) The Pho80-Pcl5_aa79-178-Pho80 hybrid is able to suppress the overexpression toxicity of Gcn4 like full-length-Pcl5 in pcl5 yeast cells. Wild-type cells (RH3237) and pcl5 mutant cells (RH3238) expressing myc³-GCN4 from the GAL1 promoter (pME2848), together with MET25-PCL5_aa61-229-GFP (pME2854, 2μm), MET25-PCL5_aa61-180-GFP (pME2858, 2μm), MET25-PCL5_aa1-180-GFP (pME2853, 2μm), MET25-PCL5_aa1-180-GFP (pME3573, CEN), MET25-PHO80_aa1-73-PCL5_aa79-178-PHO80_aa170-294-GFP (pME2860, 2μm), or MET25-PHO80_aa1-73-PCL5_aa79-178-GFP (pME2948, 2μm), and PCL5prom-PHO80_aa1-73-PCL5_aa79-178-GFP (pME3576, CEN) were spotted in 10-fold dilutions on glucose and galactose to induce expression of GCN4 driven by the GAL1 promoter. The plates were incubated for 3 days at 30°C. Furthermore, these cells were used for growth tests in liquid culture with glucose or galactose as a carbon source, and the doubling times of three or more independent cultures were determined. (B) The Pho80-Pcl5_aa79-178-Pho80 hybrid is able to promote Gcn4 degradation like full-length Pcl5 in pcl5 yeast cells. Protein levels of myc³-Gcn4, Cdc28, or eIF-2 were determined by Western blotting at the indicated time points after the GAL1 promoter shutoff in the same transformed yeast cells as described in panel A. The numbers given below each lane indicate the remaining Gcn4 percentage compared to Cdc28 or eIF-2 as internal standard quantified with a Kodak imaging station of the gel shown.

• Open in new tab
• Download powerpoint

FIG. 6.

Domain analysis of yeast cyclin Pcl5. (A) On the left side of the panel are listed Pcl5 fragments fused to the N terminus of GFP. pME2846 (Pcl5_aa1-229), pME3577 (Pcl5_aa1-229*** containing the three amino acid substitutions K209A, R210A, and R212A marked by asterisks), pME2850 (Pcl5_aa1-95), pME2851 (Pcl5_aa1-127), pME2853 (Pcl5_aa1-180), pME2858 (Pcl5_aa61-180), pME2854 (Pcl5_aa61-229), pME3370 (Pcl5_aa61-229*** containing the three amino acid substitutions K209A, R210A, and R212A), pME2855 (Pcl5_aa111-180), pME2856 (Pcl5_aa111-229), pME2857 (Pcl5_aa153-229), pME2859 (Pcl5_aa181-229), pME3578 (Pcl5_aa181-229*** containing the three amino acid substitutions K209A, R210A, and R212A), pME3574 (Pcl5_aa79-178), pME2860 (Pho80_aa1-73-Pcl5_aa79-178-Pho80_aa170-294), pME2948 (Pho80_aa1-73-Pcl5_aa79-178), and pME2849 (GFP alone). A summary of the results is shown on the right. The columns indicate the subcellular localization of the different PCL5 or PCL5-PHO80 deletions (Fig. 3) and their ability to complement the pcl5 phenotype of Gcn4 toxicity or to promote Gcn4 degradation (Fig. 5). CB, cyclin box; N, nuclear; PM, plasma membrane; C, cytoplasm; +, yeast cells are able to complement the pcl5 phenotype or to degrade Gcn4, respectively; −, yeast cells are not able to complement the pcl5 phenotype or rather to degrade Gcn4. (B) Scheme of identified Pcl5 domains. The relative positions of the different domains within the full-length Pcl5 protein are shown. The region from aa 1 to 95 represents a putative plasma membrane binding site motif, and the middle part consisting of aa 79 to 178 is required for the right substrate specificity. The Pcl5 carboxyl terminus of aa 207 to 215 is required for nuclear localization. CB, cyclin box.

The C-terminal NLS of Pcl5 can be replaced by the N-terminal NLS of Pho80.We analyzed whether the Pcl5 NLS can be replaced by the NLS of another Pho85 cyclin without loss of function. The rationale for these experiments was the question whether nuclear localization represents a distinct domain or whether it overlaps with other functions of the cyclin. Recently, it was shown that a Pho80-Pcl5-Pho80 hybrid is able to mediate Gcn4 degradation (2). The cyclin Pho80 normally acts as the specificity factor of Pho85 for the transcription factor Pho4 and is unable to promote Gcn4 degradation (52). Since the 49 carboxy-terminal amino acids of Pcl5 are required for nuclear localization, we first analyzed whether a Pho80-Pcl5 hybrid carrying the N and C termini of Pho80 is able to mediate nuclear localization of the central Pcl5 domain. The functional Pho80-Pcl5-Pho80-GFP fusion protein (Fig. 5) is localized in the yeast nucleus in contrast to the Pcl5_aa61-180-GFP and Pcl5_aa79-178-GFP hybrid (Fig. 3A). Furthermore, we analyzed which part of this Pho80-Pcl5-Pho80 hybrid, which carries aa 79 to 178 of Pcl5 to provide Gcn4 specificity to the CDK, is responsible for nuclear localization. Fusions of either the Pho80 N or C terminus to this Pcl5 fragment were analyzed for their subcellular localization (N-Pho80-Pcl5-GFP and Pcl5-Pho80-C-GFP). Strikingly, only the Pho80-Pcl5-GFP-fusion carrying the Pho80 N terminus is localized in the nucleus (Fig. 3A) and is able to suppress the overexpression toxicity of Gcn4 (Fig. 5A). Accordingly, Gcn4 is degraded in the presence of PHO80-PCL5 encoding Pho80_aa1-73-Pcl5_aa79-178-GFP, even when expressed from a low-copy (CEN) plasmid under the control of the PCL5 promoter (Fig. 5B). In contrast, a Pcl5-Pho80-GFP hybrid containing only the Pho80 C terminus is found as aggregates over the whole cell, similar to the localization of the Pcl5 central part alone (data not shown and Fig. 3A). Thus, in contrast to Pcl5, which primarily depends on a C-terminal NLS, Pho80 is transported into the nucleus via an N-terminal NLS. Our data demonstrate that the C-terminal Pcl5 NLS can be replaced by the N-terminal Pho80 NLS without loss of substrate specificity, and therefore the nuclear localization regions of cyclins Pcl5 or Pho80 represent distinct domains, which can be combined in a modular fashion and exchanged.