INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

All living cells integrate signals from the environment to modify the activity of genes required for mitotic division, differentiation, or stationary phase. For the fission yeast Schizosaccharomyces pombe, nutritional signals direct life cycle choices. As key nutrients become limited, cells exit the mitotic cycle and enter either G₀ stationary phase or a program of sexual differentiation (6, 10). The choice between G₀ and differentiation is governed by the presence of mating-specific pheromones. Starved cells respond to pheromones produced by cells of the opposite mating type by undergoing transient G₁ arrest, followed by conjugation and meiosis (8, 18, 33, 49). As expected from the need for both nutrient limitation and pheromone signaling, many signal transduction pathways converge to regulate differentiation. These include pathways regulated by cyclic AMP (cAMP), the ras1-regulated mitogen-activated protein (MAP) kinase pathway and the stress-activated MAP kinase pathway (see reference 51 for a review). However, a variety of studies indicate that each phase of the fission yeast life cycle can be governed by the activity of Ran1 (Pat1) kinase (referred to as Ran1 hereafter) and its substrates.

Inactivation of Ran1 is necessary and sufficient to divert cells from the mitotic cell cycle into the meiotic developmental program (16, 17, 32). Experiments examining the phenotype of cells carrying a ran1 temperature-sensitive allele suggest that Ran1 is regulated by stepwise inactivation of the kinase (3, 23). Limiting nutritional conditions trigger partial inactivation of Ran1. This allows cells to accumulate in G₁ (3, 8), the only cell cycle stage permissive for conjugation (10). Following conjugation, continued starvation of the diploid zygote and activation of the mating pheromone pathway lead to full inactivation of Ran1 (24, 49). This promotes meiosis. The complex phenotypes attributed to loss of Ran1 indicate that its activity is likely regulated by a variety of mechanisms.

Attenuation of Ran1 activity provokes expression of genes that function during sexual differentiation as elements of a cascading circuit (31). The most upstream is the product of the ste11 gene, which encodes a transcription factor required for expression of most meiosis-specific genes (43). Ste11 is an in vitro substrate for Ran1 and a likely physiological target for the kinase. Ran1 phosphorylation negatively regulates the transcription factor, perhaps by hindering its nuclear localization (20). Ste11 binds a specific DNA sequence, the TR box, found upstream of genes it regulates. Ste11 is required for expression of the mating type genes (43), and these regulate meiotic commitment. The mating type locus is not a single genetic entity but includes four genes. matP_c and matP_m are functional in plus cells, and matM_c and matM_m are functional in minus cells. matP_c and matM_c control production of pheromones and pheromone receptors essential for conjugation (see reference 7 for a review). Pheromone signaling is also required for expression of matM_m and matP_m, both of which are necessary for meiosis (19, 49). matP_m and matM_m directly provoke transcription of mei3 (24, 45, 49). As described below, the product of the mei3 gene is a critical meiotic activator.

mei3 is expressed only in diploid cells competent to undergo meiosis (24). All available data are consistent with the hypothesis that Mei3 activates meiosis by inhibition of Ran1 kinase. The inhibitor contains a region, RKD, which resembles two regions in the Ste11 substrate for Ran1 (20). Structure-function studies indicate that the Mei3 RKD region is critical for association with Ran1. Cells containing mei3 alleles with Mei3 RKD mutations conjugate well but sporulate inefficiently (47). It is hypothesized that during meiosis, inactivation of Ran1 by Mei3 leads to accumulation of hypophosphorylated, and hence active, forms of Ste11. This leads to induction of meiosis-specific genes. One of these, Mei2, is an RNA-binding protein required for premeiotic DNA synthesis (3, 48). Like Ste11, Mei2 is negatively regulated by Ran1 phosphorylation. It has been suggested that the phosphorylation state of Ste11 readies cells for meiosis, while that of Mei2 determines meiotic commitment (48).

As described above, Mei3 does not regulate conjugation but functions only during meiosis to inactivate Ran. The mechanisms used to partially inactivate Ran1 so that G₁ arrest and conjugation can proceed have not been described. To identify other components of the Ran1 signal transduction pathway, we searched for protein partners of Ran1. A fission yeast cDNA library was screened using the two-hybrid interaction trap. We identified RACK1 (Cpc2) as a Ran1-interacting protein. RACK1 functions in mammalian cells to regulate at least two unrelated kinases, protein kinase C (PKC) (27) and Src (5). The fission yeast version of RACK1, named Cpc2 to conform to the Neurospora crassa designation (28), is 77% homologous to the human protein. In spite of its strong conservation in many organisms, cpc2 is not an essential gene in fission yeast. Cells containing a cpc2 null allele are highly elongated due to a G₂ cell cycle delay. The cells accumulate inefficiently in G₁ when deprived of nitrogen, and conjugation and meiosis are impaired. These phenotypes argue that cpc2 acts in opposition to ran1, perhaps to regulate the activity of the kinase. In support of this hypothesis, cpc2 functions upstream of ran1 and a physical association between the two proteins is observed. Fluorescence microscopy using a Ran1-green fluorescent protein (GFP) fusion indicates that Cpc2 may function to alter the cellular localization of Ran1. Consistent with the high level of homology observed between RACK1 and Cpc2, rat RACK1 is capable of functionally complementing a cpc2 null allele.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Strains and media. The genotypes of the S. pombe strains used in this study are as follows: SP66, h⁹⁰ leu1-32 ade6-M216; SP712, h^+N leu1-32 ura4-D18 ade6-M210 ran1⁺ OP (OP refers to genes expressed at high levels under the control of the constitutive adh promoter); SP713, h^+N leu1-32 ura4-D18 ade6-M210 ran1^K47R+ OP; SP870, h⁹⁰ leu1-32 ura4-D18 ade6-M210 cpc2::pCPC2-24; SPB5, h⁹⁰ leu1-32 ade6-M210; SPB67, h^+N leu1-32 ura4-D18 ade6-M210/h leu1-32 ura4-D18 ade6-M216; SPB68, h⁹⁰ mei2::LacZ; SPB76, h^+N leu1-32 ura4-D18 ade6-M216 cgs1::ura4; SPB90, h^S leu1-32 ura4::fbp-lacZ fbp1::ura4 ade6-M216; SPB93, h^S leu1-32 cgs1.1 ura4::fbp-lacZ fbp1::ura4 ade6-M210; SPB173, h^+N; SPB182, h^+N ura4-D18 cpc2::ura4; SPB188, h⁹⁰ leu1-32 ura4-D18 ade6-M210 cpc2::ura4; SPB190, h⁹⁰ leu1-32 ura4-D18 cpc2::pCPC2-24; SPB191, h⁹⁰ leu1-32 ura4-D18 ade6-M210 mei2t^s ran1::pRAN1-90 (ura⁺); SPB214, h⁹⁰ leu1-32 ura4-D18 ade6-M216 ran1-114 cpc2::ura4; SPB217, h⁹⁰ leu1-32 ade6-M216 ran1-114; SPB273, h⁹⁰; SPB307, h⁹⁰ ura4-D18 cpc2::ura4 mei2-lacZ; SPB317, h⁹⁰ ura4-D18 cpc2::ura4; SPB318, h⁹⁰ leu1-32 git2-1; SPB319, h⁹⁰ ura4-D18 leu1-32 cpc2::ura4 git2::LEU2; SPB320, h^S leu1-32 ura4::fbp-lacZ fbp1::ura4 cpc2::ura4 ade6-M216.

leu2-3 ura3-52 trp1-1901 his3-200 ade2-101 gal4 gal80 URA3::GAL1-LacZ LYS::GAL1-HIS3 cyh

Yeast two-hybrid screen. Plasmids and details of the library screen were essentially as previously described (12). The ran1 gene was fused in frame to the GAL4 DNA-binding domain (DBD) in pAS2. Following transformation of this plasmid into Y190, a Western blot using Ran1 antibodies (23) was used to verify that the fusion protein was produced intact. A fission yeast cDNA library fused to the activation domain of GAL4 in the vector pACT2 (provided by Steve Elledge) was transformed into the Y190/ran1-DBD strain. The transformants were selected on plates containing 3-amino triazole (3-AT) at 30°C for 3 to 4 days. Approximately 10⁶ transformants were obtained. To eliminate one class of false positives, each library plasmid was tested to identify those capable of activating the LacZ reporter gene in the absence of ran1-DBD. ran1-DBD was reintroduced into those candidate strains and tested for activation of LacZ. Measurement of -galactosidase activity was accomplished using a permeabilized-cell assay. A total of 53 colonies were obtained, and DNA was isolated from each. Limited DNA sequence analysis of each identified 30 clones representing nine unique open reading frames.

Oligonucleotides. The sequences of the oligonucleotides used in this study are as follows: GB-RKDI, 5'CTTCTTTCCGGTTCTGCAGACGCGTCCATCATTTTGTG3'; GB-RKDII, 5'GGTTGTTTCTGGTTCCGCGGACGCGACCATTAAGATTTG3'; CPC2-Xma, 5'TCTTGTCCCGGGAACCAGAAA3'; CPC2-HIII, 5'CGGTACCTTGAAGCTTTGGGA3'; CPC2-35, 5'CTCGAAGGTCACTCTGGATG3'; CPC2-927C, 5'TACTTGGTAACTTGCCAGACAC3'; CPC2-Nde, 5'TACCATATGACCGATGGTGGTCACTC3'; CPC2-Bam, 5'AGAGGATCCACCATCGGTGATAGTGT3'.

Construction of a Cpc2 null allele. Oligonucleotides CPC2-Nde and CPC2-Bam were used in a PCR to obtain the entire cpc2 cDNA. The linear fragment was cloned into the commercially obtained TA vector (InVitrogen). Inverse PCR of this plasmid using primers CPC2-Xma and CPC2-HIII was used to insert HindIII and XmaI restriction sites into cpc2. A HindIII-to-XmaI fragment of ura4 was cloned into the inverse PCR product. A linear DNA fragment containing cpc2 disrupted with ura4 was obtained using primers CPC2-35 and CPC2-927C in a PCR. The linear fragment was purified using a Qiagen column and used to transform SPB67. Stable URA⁺ transformants were identified, and a Southern blot verified the correct replacement. Sporulation of the diploid produced the haploid strain SPB188, which contains cpc2::ura4 (also referred to as cpc2).

Oligonucleotide mutagenesis. The Cpc2 RKD motifs were mutagenized using oligonucleotides GB-RKDI and GBRKDII. All oligonucleotides were designed with the assistance of the OLIGO software package (National Biosciences Inc., Plymouth, Minn.). The identities of the mutations were confirmed using restriction enzyme analysis.

Expression and purification of recombinant proteins. The plasmids used for expression of recombinant proteins were pRAN1.83 (His-hemagglutinin [HA]-Ran1), pRAN1.87 (His-HA-Ran1^K47R), and pSTE11.27 (His-p39^ste11). The construction of each and expression in bacteria have been described previously (20). pCPC2.2 was used for expression of His-HA-Cpc2. The soluble portion of bacterial whole-cell extract was incubated with 1.0 ml of Ni²⁺-nitrilotriacetic acid agarose. The mixture was packed into a column and washed sequentially with 10 column volumes each of wash buffer I (300 mM NaCl, 20 mM HEPES [pH 7.5], 5 mM MgCl₂, 10% glycerol), wash buffer II (wash buffer I with 100 mM NaCl), and wash buffer III (wash buffer II plus 20 mM imidazole). Proteins were eluted from the column in wash buffer III containing 100 mM imidazole. Imidazole was removed by dialysis of the purified protein against wash buffer I.

Copurification of GST-tagged proteins from yeast. Plasmids pALT4-GST and pCPC2-24 were constructed to produce HA epitope-tagged glutathione S-transferase (GST) and HA-GST-tagged Cpc2, respectively. The plasmids were transformed into SP66 (wild type) or SPB191. SPB191 contains an integrated plasmid which produces HA-tagged ran1 under control of the nmt1 promoter from plasmid pREP41 (2). The four strains were grown in thiamine-free medium to allow induction of ran1. Pelleted cells were washed in HE buffer (50 mM Tris HCl [pH 8.0], 5 mM EDTA, 1 mM dithiothreitol) and resuspended in 0.5 ml of HE buffer containing 1 mM phenylmethylsulfonyl fluoride. Following addition of sterile glass beads, the cells were broken by vortexing for 15 min at 4°C. Lysates were removed to a fresh tube, and the glass beads were washed in 2.0 ml of RIPA buffer (HE plus 0.1% sodium dodecyl sulfate SDS, 1% Triton X-100, 0.5% deoxycholate). Clarified lysate was obtained by centrifugation at 12,000 × g for 10 min. Protein concentrations were determined using bovine serum albumin as the standard in the Bio-Rad system. Lysate (5 mg) was purified using glutathione-agarose (Sigma). The beads were resuspended in Laemmli sample buffer prior to separation by SDS-7.5 to 15% polyacrylamide gel electrophoresis (PAGE). Western blotting was performed as previously described (20). The primary antibody used for detection of Ran1 was a mixture of R30, R48, and R99 monoclonal antibodies (23) diluted 1:1,000 in 5% dry milk. For detection of HA fusion proteins, a commercially available antibody was used (12CA5; Boehringer Mannheim Biochemicals, Indianapolis, Ind.). Immunoreactive proteins were visualized using the Dupont NEN Research Products chemiluminescence assay kit. Alternatively, the purified proteins were used for a kinase assay. In that case, proteins bound to the agarose beads were washed in RIPA buffer as described above. The beads were resuspended in 50 µl of 1×KAB (50 mM Tris-HCl [pH 7.5], 5 mM MgCl₂, 1 mM EDTA, 0.2 mM dithiothreitol) in the final step. Kinase assays utilized 25 µl of bead-bound protein and p39^Ste11 as the substrate (20).

Plasmids. Steve Elledge provided hybrid plasmids pAS2 and pACT2 and the fission yeast cDNA library. The entire coding region of ran1 was fused in frame to the DBD of GAL4 as an NdeI-to-BamHI fragment. cpc2 or cpc2 RKD was fused to pACT2 as an NheI-to-BamHI fragment obtained from a pALT4 vector.

Flow cytometry. Either wild-type (SPB173) or cpc2 (SPB182) cells were grown to 10⁷/ml and then shifted to medium lacking nitrogen for the designated times. Cells were fixed and stained with propidium iodide as previously described (1). DNA fluorescence was measured using a FACScan (Becton Dickinson).

Measurement of -galactosidase activity. Total cell extract was prepared from cells containing a mei2-lacZ gene after they attained a density of 10⁷/ml. A 20-µg protein sample was used to assay -galactosidase activity (1). Activity is expressed as A₄₂₀ × 1.7/0.0045 × protein (micrograms) × volume × time. fbp1-lacZ activity was measured in 10⁷ permeabilized cells.

Measurement of viability in stationary phase. Cells were grown to 10⁸/ml at 30°C. At this time (day 0) and each day thereafter (days 1 to 4), a portion of the cells was removed and stained with FUN-1 (Molecular Probes, Eugene, Oreg.) as previously described (25, 53). Cells were immediately examined under a microscope. Viable cells contained red intravacuolar structures, and nonviable cells stained yellow-green.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Isolation of cpc2 using a two-hybrid library screen. A fission yeast cDNA library was screened by two-hybrid analysis using ran1-DBD) as bait. Approximately 10⁶ colonies were obtained following cotransformation of ran1-DBD and the library plasmids into yeast. Plasmid inserts from LacZ-positive transformants were sequenced, and one of these, cpc2, is the subject of this paper. cpc2 was chosen for analysis because the predicted protein has regions of homology with other proteins known to directly interact with Ran1 (see below). The interaction between Ran1 and Cpc2 was quantitatively assessed in the two-hybrid system using S. cerevisiae strain Y190. This strain contains a LacZ reporter gene controlled by the upstream activating sequence of GAL4. We observed that Cpc2 paired with Ran1 produced sixfold higher -galactosidase activity than Ran1 paired with the ACT2 vector. This interaction is comparable to, but not quite as strong as, the interaction between Ran1 and Ste11, a transcription factor that directly associates with Ran1 (Fig. 1A and reference 20).

View larger version (51K): [in this window] [in a new window]   FIG. 1.   Ran1 interacts with the highly conserved RACK1 (Cpc2) protein. (A) Two-hybrid analysis of the interaction between Ran1 and Cpc2. The reporter strain (Y190) permits detection of protein-protein interaction through transcriptional activation of a GAL1-LacZ gene. Y190 cells were transformed with the indicated plasmids. Ran1 was fused to the GAL4 DBD. Ste11, Cpc2, and the Cpc2 RKD were each fused to GAL4 activation domain. Cpc2 RKD contains defined mutations in both Cpc2 RKD domains (see below). A permeabilized-cell assay was used to measure -galactosidase activity. The results represent the mean values obtained using three independent transformants. (B) Amino acid sequence comparison of regions within Ste11, Cpc2, and Mei3. The arrow marks the phosphoacceptor position occupied by serine or threonine residues in the Ste11 substrate for Ran1 (20). The asterisks over Mei3 residues indicate those required for Mei3 function (47). The amino acids R36, K38, R125, and K127 in Cpc2 were each changed to alanine in Cpc2 RKD (indicated with asterisks). The superscript numbers correspond to amino acid positions. (C) Amino acid comparison of Homo sapiens (Hs) and Rattus norvegicus (Rn; reference 36) RACK1, S. pombe (Sp) Cpc2, and S. cerevisiae (Sc) Cpc2 (15). Identical amino acids are indicated using a dark box.

Ran1 and Cpc2 associate in cell lysates. The presence of RKD sequences in Cpc2 indicates that if, in fact, Ran1 and Cpc2 are associated, the interaction may be direct. To determine if the two polypeptides are physically connected, we assayed cell lysates for copurification of Ran1 and Cpc2. To accomplish this, cells were transformed with plasmids expressing GST, Cpc2 fused to GST (GST-Cpc2), or Ran1. To permit immunological detection, all proteins were also tagged with HA1 sequences. Soluble proteins were prepared from cells expressing the fusion proteins either individually or in pairs. An immunoblot developed with anti-HA1 antibodies was used to examine the steady-state level of each fusion protein (Fig. 2A). Since Ran1 and GST-Cpc2 migrate close to one another, the immunoblot was also developed using anti-Ran1 antibodies (Fig. 2B). These experiments allowed unequivocal identification of Ran1 and GST-Cpc2 and verified that the proteins were intact. Partial purification of GST fusion proteins was accomplished on glutathione beads. Bead-bound proteins were separated by SDS-PAGE and analyzed in an immunoblot using anti-HA1 antibodies (Fig. 2C) or anti-Ran1 antibodies (Fig. 2D). This experiment showed that Ran1 does not inherently bind glutathione beads or copurify with GST (Fig. 2C and D, lane 3). In contrast, Ran1 was present in partially purified complexes containing GST-Cpc2 (Fig. 2C and D, lane 4).

View larger version (52K): [in this window] [in a new window]   FIG. 2.   Ran1 and Cpc2 physically interact. Wild-type cells (SP66, lanes 1 and 2) or cells producing high levels of Ran1 (SPB191, lanes 3 and 4) were transformed with a plasmid expressing either GST (lanes 1 and 3) or GST-Cpc2 (lanes 2 and 4). All expressed proteins were epitope tagged with HA1 sequences. Cells were grown to 107/ml in complete medium and shifted to thiamine-free medium for protein induction. Soluble extract was prepared after 18 h of growth. Proteins were separated by SDS-7.5 to 15.0% PAGE and immunoblotted with anti-HA (A) or anti-Ran1 (B) antibody. A portion (2 mg) of each soluble extract was applied to glutathione-agarose beads. Bead-bound material was examined in an immunoblot assay using anti-HA (C) or anti-Ran1 (D) antibody. The positions of HA-Ran1, HA-GST-Cpc2, and HA-GST are marked.

Loss of cpc2 causes pleiotropic cell cycle defects. Because two independent assays indicate that Ran1 and Cpc2 physically associate, we examined cells for a functional connection between the two proteins. To accomplish this, a cpc2 null allele was constructed. A portion of the predicted cpc2 open reading frame was deleted and replaced with the ura4 gene to form cpc2::ura4. A linear DNA fragment containing cpc2::ura4 was transformed into diploid cells (SPB67). Stable URA4⁺ transformants were analyzed using a Southern blot to identify a simple replacement of cpc2 (data not shown). One representative diploid was chosen for further analysis. The cells were allowed to undergo meiosis, and spores were collected. Random spore analysis showed that URA⁺ and URA cells were equally represented. This result indicates that cpc2 is not an essential gene. During this analysis, it was noted that cpc2 cells formed small colonies compared to wild-type cells (Fig. 3A). Microscopic examination of the cells showed that they were elongated, a phenotype observed in cells producing activated Ran1 and a hallmark of mitotic delay (Fig. 3B). To examine the role of cpc2 in division, cell growth was monitored. This experiment revealed that cpc2 doubled in 3.0 h, as did wild-type cells (Fig. 3C). Thus, loss of cpc2 allows cells to grow normally but delays cell division. In this and other experiments, it was noted that cpc2 cells consistently ceased dividing at a two- to fourfold lower density than wild-type cells. These phenotypes are associated with loss of cpc2 because expression of cpc2 cDNA completely restores the wild-type phenotype (Fig. 3A and data not shown). Next, we examined the fate of cells exiting the cell cycle.

View larger version (54K): [in this window] [in a new window]   FIG. 3.   Loss of cpc2 causes pleiotropic defects. (A) cpc2+ (SPB173), cpc2 (SPB182) (left side), or cpc2 (SPB188) (right side) cells transformed with the indicated plasmids, spread on minimal medium plates, and incubated for 60 h at 30°C. (B) Photomicrographs of cpc2+ (SPB173), cpc2 (SPB182), and ran1OP (SP712) cells. (C) Growth curve constructed by growing cells (cpc2+ [SPB173] or cpc2 [SPB182]) to stationary phase and inoculating them at a density of 105/ml into fresh YEA medium. A portion of each culture was removed at the indicated times, and cells were counted using a hemocytometer. (D) cpc2+ (SPB173) or cpc2 (SPB182) cells patched onto minimal-medium plates and incubated at 30°C. At the indicated times, a portion of the cells was resuspended in water and the cells were counted and scored as either a vegetative cell, a zygote, or an ascus-containing spores. Differentiated cells include diploid zygotes plus asci. (E) Flow cytometry of cpc2+ (SPB173) or cpc2 (SPB182) cells shifted to nitrogen-free medium for the indicated times prior to fluorescence-activated cell sorter analysis. (F) cpc2+ (SPB173), cpc2 (SPB182), ran1OP (SP712), and ranK47ROP (SP713) cells were streaked on minimal-medium plates and incubated at either 30 or 37°C.

cpc2 functions independently of the PKA pathway. Cell cycle delay, failure to differentiate, and stationary-phase defects are phenotypes associated with activation of the cAMP-dependent protein kinase (PKA) pathway (3, 8, 9, 26). This, and the observation that RACK1 interacts with cAMP phosphodiesterase in mammalian cells (52), led us to test the hypothesis that cpc2 downregulates cAMP levels. In fission yeast, alteration of cAMP levels or of PKA activity causes pleiotropic defects. Transcription of genes regulated by nitrogen or glucose availability is impaired when cAMP levels are altered. The fructose-1,6-bisphosphatase gene (fbp1) is repressed in glucose-grown cells. Shifting the cells to glycerol-containing medium leads to derepression of fbp1 (14, 17). Addition of cAMP to cells (13) or mutations that increase PKA activity repress fbp1 transcription, even following a shift to glycerol medium. We used cells containing an fbp1-lacZ reporter gene to examine glucose repression in cpc2 cells. Either wild-type, cgs1 (cgs1 encodes the regulatory subunit of PKA), or cpc2 cells were shifted from glucose- to glycerol-containing medium. cgs1 cells fail to derepress the fusion gene immediately following the shift to glycerol (Fig. 4A and reference 50). In contrast, fbp1-lacZ expression in both wild-type and cpc2 cells increased to the same extent within 3 h of the nutritional shift and reached maximal expression at 6 h (Fig. 4A). Thus, cpc2 does not perform a major function in cAMP-regulated transcription of glucose-sensitive genes.

View larger version (27K): [in this window] [in a new window]   FIG. 4.   Interactions between cpc2 and the cAMP pathway. (A) Induction of the fbp1-lacZ fusion gene. Cells with the indicated relevant genotype and the ura4::fbp-lacZ reporter gene were shifted from glucose (filled symbols) to glycerol (empty symbols) medium. -Galactosidase activity was measured at the indicated times using a permeabilized-cell assay. Each point represents three samples (standard error of the mean, <2.0%). Wild-type (wt; SPB90), cgs1 mutant (SPB93), and cpc2 mutant (SPB320) cells were used. (B) Induction of the mei2-lacZ fusion gene. Cells were grown to 5 × 106/ml and washed in medium devoid of a nitrogen source. The culture was divided between two flasks. NH4Cl was added to one sample (closed symbols); the other represents nitrogen-depleted medium (open symbols). -Galactosidase activity was measured at the indicated times using 20 µg of soluble cell protein. Wild-type (SPB68) and cpc2 mutant (SPB307) cells were used. (C) Survival in stationary phase. Cells were grown to saturation (2 × 108/ml; day 0) and incubated for an additional 4 days (days 1 to 4). A portion of the culture was removed each day and stained with FUN-1. Wild-type (SPB173) cgs1 mutant (SPB76), and cpc2 mutant (SBP182) cells were used. (D) Sexual differentiation of cells grown on phorbol myristate acetate for 3 days at 30°C. Asci and zygotes in a portion of each sample were measured using a hemocytometer. Wild-type (SPB273), git2 mutant (SPB318), cpc2 mutant (SPB317), and git2 cpc2 mutant (SPB319) cells were used.

Genetic interactions between Ran1 and Cpc2. If cpc2 and ran1 function on the same genetic pathway, cpc2 could potentially function as an upstream regulator or a downstream effector for the kinase. An epistasis test was used to discriminate between these possibilities. Cells containing the ran1-114 allele undergo haploid sporulation when the cells are incubated under restrictive conditions. Loss of cpc2 does not abolish conjugation and meiosis but retards the process (Fig. 3D). Conditions that partially inactivate ran1-114 were used to determine if the sporulation rate of ran1 cells is dependent on cpc2. We observed that neither wild-type cells nor cpc2 cells sporulated to a significant level after 20 h on malt extract medium at 30°C. At that time, 80% of the ran1-114 cells contained spores. Importantly, ran1-114 sporulation was only partially independent of cpc2. Inactivation of ran1 reversed the differentiation defect caused by loss of cpc2, but sporulation was accelerated if cells contained a functional cpc2 allele (Fig. 5A). We infer from this result that loss of cpc2 enhances ran1 residual activity. Since inactivation of ran1 allows cpc2 cells to regain a function, then cpc2 is most likely upstream of ran1 if, in fact, they are on the same pathway.

View larger version (34K): [in this window] [in a new window]   FIG. 5.   Genetic interactions between ran1 and cpc2. (A) Wild-type (SP66), ran1 mutant (SPB217) cpc2 mutant (SPB188), or ran1 and cpc2 mutant (SPB214) cells were patched onto malt extract plates and incubated at 32°C for the indicated periods of time. A portion of the patches was resuspended in water, and the cells were counted and scored as a vegetative cell, a zygote, or an ascus containing spores. A differentiated cell includes zygotes and cells with spores. (B) Either wild-type (SP66) or cpc2 mutant (SPB182) cells were transformed with a control plasmid (pALT2) or a plasmid containing adh-ran1 (pRAN1-81).

Cpc2 allows nuclear accumulation of Ran1. Mammalian RACK1 is believed to regulate PKC function by serving as an anchoring protein for the activated kinase (37). Thus, the cellular localization of Ran1 in wild-type and cpc2 cells was examined. For these experiments, Ran1 was fused to GFP and expressed under control of the nmt1 promoter. Induction of the fusion protein was accomplished by removal of thiamine from the medium. A detectable GFP signal was first noted 7 h following induction, and the signal was concentrated in the nucleus (Fig. 6A and reference 47). Early nuclear concentration of Ran1 was independent of cpc2. At 12 h following induction, a striking difference in the cellular distribution of Ran1-GFP became evident. Ran1-GFP displayed a prominent, punctate cytoplasmic staining in cpc2 cells that was not observed in wild-type cells. In addition, the fusion protein did not accumulate to high levels in the nucleus in the absence of cpc2 (Fig. 6A and data not shown). The pattern is specific for Ran1-GFP because it was not observed for the other nuclear proteins tested, such as Ste11-GFP and Mei3-GFP, or for GFP alone (data not shown). The level of Ran1-GFP in the cells was examined in an immunoblot. This experiment showed that the fusion protein had the predicted molecular weight and that the steady-state level of Ran1-GFP in cpc2 cells was comparable to that in wild-type cells (Fig. 6B).

View larger version (36K): [in this window] [in a new window]   FIG. 6.   Localization of the Ran1-GFP fusion protein. (A) cpc2+ (SPB5) or cpc2 (SPB188) cells were transformed with a plasmid (pRAN1-GFP) expressing Ran1-GFP under control of the nmt promoter. Transformants were grown to 5 × 106 cells/ml in minimal medium containing thiamine (Thia). The cells were washed thoroughly in thiamine-free medium, and incubation was continued in the absence of thiamine for 7 or 12 h. The cells were then examined using a microscope. (B) The above cells were used to prepare whole-cell extract. A portion of the extract was separated by SDS-PAGE and used in a Western blot assay. The proteins were detected using anti-Ran1 antibody.

Cpc2 does not alter Ran1 activity in vitro. One hypothesis consistent with the results obtained to this point is that cpc2 functions upstream of ran1 and negatively regulates its activity. The mechanism employed by Cpc2 to inactivate Ran1 was further investigated. An in vitro kinase assay was used to measure Ran1 activity in partially purified yeast lysates to determine if Cpc2 inhibits Ran1 substrate phosphorylation. Ran1 was expressed as a GST fusion protein in wild-type cells, cpc2 cells, or cells expressing high levels of cpc2. Affinity-purified Ran1 was incubated with p39^ste11 as the substrate and radioactive ATP (Fig. 7A). This experiment revealed that Ran1 substrate phosphorylation was unaltered by Cpc2. To more accurately quantitate the effect of Cpc2 on Ran1, both proteins were expressed in and purified from bacteria. (HIS)₆-tagged Cpc2 was purified by Ni²⁺ affinity chromatography, and increasing amounts of Cpc2 were added to (HIS)₆-Ran1 in an in vitro kinase assay (Fig. 7B). This analysis showed that addition of Cpc2 had no effect on Ran1 substrate phosphorylation. Thus, Cpc2 does not inhibit Ran1 activity under these conditions. As shown in Fig. 1B, Cpc2 contains serine and threonine residues (Ser-39 and Thr-128) at positions corresponding to the phosphoacceptor serine and threonine residues in Ste11 (20). However, the in vitro kinase assays also demonstrate that Cpc2 is not likely to function as a substrate for Ran1 (data not shown).

View larger version (27K): [in this window] [in a new window]   FIG. 7.   Cpc2 does not alter Ran1 substrate phosphorylation in vitro. (A) Western blot (top) and in vitro kinase (bottom) assays of extracts obtained from cpc2+ (SP66), cpc2 (SPB188), or cpc2OP (SPB870::CPC2-24) cells transformed with a plasmid (pRAN1-95) producing Ran1 under control of the nmt promoter. Ran1 is tagged at the amino terminus with GST-HA1 sequences. Twenty hours following induction of Ran1, cell lysate was prepared. Ran1 was affinity purified from the lysates and incubated in the in vitro kinase reaction. p39ste11 was used as the substrate (20). (B) Recombinant p52ran1 and recombinant p39ste11 were incubated with [32P]ATP in the absence (lane 1) or the presence (lanes 2 to 6) of increasing concentrations of recombinant Cpc2. Phosphorylation of p39ste11 was monitored using SDS-PAGE and autoradiography.

Mammalian RACK1 is a functional homologue of the fission yeast gene. We wished to determine if Cpc2 and mammalian RACK1 are functional, as well as structural, homologues. Recently, it was shown that N. crassa cpc-2 is a functional homologue of the S. cerevisiae CPC2 gene (15). To investigate this, rat RACK1 (which is identical to the human protein) was expressed in yeast under control of the adh promoter. This plasmid was introduced into a cpc2 strain. As controls, the cells were transformed with the vector alone or with a plasmid expressing fission yeast cpc2. Individual transformants were examined in several assays (Fig. 8). As previously observed, loss of cpc2 caused temperature-sensitive growth. Additionally, cpc2 cells formed small colonies compared to wild-type cells. Expression of mammalian RACK1 reversed both defects as efficiently as yeast cpc2. The transformants were also examined to determine if the mammalian gene was able to reverse the cpc2 sexual differentiation defect. For this assay, cells were plated on minimal medium for 2 days and stained with iodine vapor to observe spore-containing cells. Both yeast cpc2 and mammalian RACK1 restored the ability of cpc2 cells to sporulate. This experiment provides evidence that mammalian RACK1 is a structural and functional homologue of fission yeast cpc2.

View larger version (85K): [in this window] [in a new window]   FIG. 8.   Functional complementation of cpc2 by expression of the mammalian gene. Either cpc2+ (SP66) or cpc2 (SPB190) cells were transformed with a control plasmid (pIRT2), the fission yeast cpc2 gene (pCPC2.10), or the gene for rat RACK1 (pRACK1.1). (A) The transformants were spread on minimal-medium plates and incubated at 30°C to observe colony morphology (left side of panel). In addition, a representative colony was streaked for single colonies and incubated at 37°C to examine viability (right side of panel). (B) The morphology of the cells was examined using Nomarski optics. (C) Patches of cells were incubated at 30°C for 2 days and stained using iodine vapor to observe spore-containing cells. The patched cells were examined under a microscope to count vegetative cells, zygotes, and spore-containing cells. The bar graph represents the percentages of cells that have conjugated or formed spores. Three independent samples were measured.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

Ran1 is a key regulator of the developmental switch between mitotic cell growth and meiosis. Activation of Ran1 inhibits sexual differentiation (23), while inactivation of the kinase initiates G₁ arrest, conjugation, and meiosis (3, 16, 17, 32). We screened a fission yeast library using a two-hybrid interaction assay to identify protein partners for Ran1. Here, we describe one of these, the product of the cpc2 gene. Fission yeast Cpc2 is a structural and functional homologue of mammalian RACK1 proteins. RACK1 was initially isolated as a PKC-binding protein. It is hypothesized to function as an anchoring protein for activated PKC to enable access of the kinase to physiologically important substrates. Evidence is presented that Cpc2 interacts with Ran1. Analysis of cells containing the cpc2 null allele indicates that it is important for cell cycle progression, as well as efficient conjugation and meiosis. Cpc2 may regulate Ran1 activity by controlling cellular localization of the kinase.

The interaction between Ran1 and Cpc2 is specific and may be direct. Three lines of evidence obtained using different experimental approaches indicate that Ran1 and Cpc2 interact. Firstly, Cpc2 was identified as a Ran1-interacting protein in a two-hybrid library screen. In support of this finding, we found Ran1 and Cpc2 physically associated in cell extracts. Cpc2 contains two regions that resemble a motif (known as RKD) previously identified in both the Ste11 substrate and the Mei3 inhibitor of Ran1. Thus, the interaction between Ran1 and Cpc2 may be mediated by the RKD-like sequences of Cpc2. Mutagenesis of specific residues in both Cpc2 RKD regions diminishes the interaction observed in the two-hybrid assay. One interpretation of this observation is that the RKD regions mediate Cpc2 interaction with Ran1. However, it can be argued that the mutations, which are in conserved regions of the protein, are required for functional folding of the protein. In that case, failure of the polypeptide to interact with Ran1 could be a general consequence of misfolding. We favor the former interpretation. Expression of Cpc2 RKD on a plasmid in cpc2 cells causes partial suppression of cpc2-associated defects (data not shown). This suggests that cpc2 RKD is a weak but functional, allele, perhaps producing a product with reduced affinity for Ran1. High-level expression of the altered protein may compensate for an affinity defect but is less likely to restore a function lost because of incorrect folding.

Structural and functional conservation of Cpc2. One of the most significant findings presented here is the identification of a structurally and functionally conserved RACK1 homologue as a partner for a kinase. It is somewhat surprising to identify the highly conserved RACK1 protein interacting with a protein that does not appear to have a structural counterpart in other organisms. The closest relative to Ran1 in budding yeast is Sks1p. The two proteins share 52% homology, and this is limited to their kinase domains. Conversely, the PKA pathway is conserved from yeast to mammals. As reported here, the phenotypes of cells containing cpc resemble those caused by high cAMP levels. These observations, and the finding that RACK1 interacts with a mammalian phosphodiesterase, provide a rationale for the hypothesis that Cpc2 regulates cAMP levels in yeast. However, our studies failed to identify a connection between Cpc2 and the PKA pathway. We also investigated phenotypes associated with loss of the two identified fission yeast PKC genes to search for PKC interaction with Cpc2 (Shor and McLeod, unpublished data). To date, no interaction has been demonstrated in yeast. The interaction between RACK1 proteins and specific signal transduction pathways may vary depending on the particular cell type examined or the signal to which the cell is responding. However, one common theme emerging from studies of Cpc2 (RACK1) in diverse systems is that Cpc2 (RACK1) alters cell cycle progression and, in particular, may augment exit from the mitotic cell cycle into either G₀ or a differentiation pathway. S. pombe cpc2 cells are cell cycle delayed and slow to respond to signals that induce G₁ arrest. In NIH 3T3 cells, expression of RACK1 causes a G₁/G₀ delay, as well as a reduced growth rate (5). Budding yeast cells devoid of CPC2 enter G₀ prematurely (15). Both S. pombe and N. crassa devoid of cpc-2 exhibit fertility defects (28). On the other hand, RACK1 specifically associates with the integrin  subunit in lymphoblastoid cells (21), suggesting that some integrin signaling pathways function through a RACK1-PKC pathway. Notably, yeast cells appear to lack integrins. Recent evidence indicates that RACK1 binds pleckstrin homology domains from some proteins. The authors speculate that RACK1 scaffolding proteins are multivalent organizers that coordinate complex signaling systems (35). With this intriguing observation in mind, further genetic analysis is under way to identify other potential partners for Cpc2.