Portal de Periódicos da CAPES

Apc10 and Ste9/Srw1, two regulators of the APC–cyclosome, as well as the CDK inhibitor Rum1 are required for G1 cell-cycle arrest in fission yeast

1998; Springer Nature; Volume: 17; Issue: 18 Linguagem: Inglês

10.1093/emboj/17.18.5388

1460-2075

Kin‐ichiro Kominami, Helena M. B. Seth-Smith, Takashi Toda,

Microtubule and mitosis dynamics

Article15 September 1998free access Apc10 and Ste9/Srw1, two regulators of the APC–cyclosome, as well as the CDK inhibitor Rum1 are required for G1 cell-cycle arrest in fission yeast Kin-ichiro Kominami Kin-ichiro Kominami Laboratory of Cell Regulation, Imperial Cancer Research Fund, PO Box 123, 44 Lincoln's Inn Fields, London, WC2A 3PX UK Search for more papers by this author Helena Seth-Smith Helena Seth-Smith Present address: Institute of Biotechnology, University of Cambridge, Tennis Court Road, Cambridge, CB2 1QT UK Search for more papers by this author Takashi Toda Corresponding Author Takashi Toda Laboratory of Cell Regulation, Imperial Cancer Research Fund, PO Box 123, 44 Lincoln's Inn Fields, London, WC2A 3PX UK Search for more papers by this author Kin-ichiro Kominami Kin-ichiro Kominami Laboratory of Cell Regulation, Imperial Cancer Research Fund, PO Box 123, 44 Lincoln's Inn Fields, London, WC2A 3PX UK Search for more papers by this author Helena Seth-Smith Helena Seth-Smith Present address: Institute of Biotechnology, University of Cambridge, Tennis Court Road, Cambridge, CB2 1QT UK Search for more papers by this author Takashi Toda Corresponding Author Takashi Toda Laboratory of Cell Regulation, Imperial Cancer Research Fund, PO Box 123, 44 Lincoln's Inn Fields, London, WC2A 3PX UK Search for more papers by this author Author Information Kin-ichiro Kominami1, Helena Seth-Smith2 and Takashi Toda 1 1Laboratory of Cell Regulation, Imperial Cancer Research Fund, PO Box 123, 44 Lincoln's Inn Fields, London, WC2A 3PX UK 2Present address: Institute of Biotechnology, University of Cambridge, Tennis Court Road, Cambridge, CB2 1QT UK *Corresponding author. E-mail: [email protected] The EMBO Journal (1998)17:5388-5399https://doi.org/10.1093/emboj/17.18.5388 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Many eukaryotic cells arrest the cell cycle at G1 phase upon nutrient deprivation. In fission yeast, during nitrogen starvation, cells divide twice and arrest at G1. We have isolated a novel type of sterile mutant, which undergoes one additional S phase upon starvation and, as a result, arrests at G2. Three loci (apc10, ste9/srw1 and rum1) were identified. The apc10 mutants, previously unidentified, show, in addition to sterility, temperature-sensitive growth with defects in chromosome segregation. apc10+ is essential for viability, encodes a conserved protein (a homologue of budding yeast Apc10/Doc1) and is required for ubiquitination and degradation of mitotic B-type cyclins. Apc10 does not co-sediment with the 20S APC–cyclosome, a ubiquitin ligase for B-type cyclins, and in the apc10 mutant the 20S complex is intact, suggesting that it is a novel regulator for this complex. A subpopulation of Apc10 does co-immunoprecipitate with the anaphase-promoting complex (APC). A second gene, ste9+. srw1+, encodes a member of the fizzy-related family, also regulators of the APC. Finally, Rum1 is a cyclin-dependent kinase (CDK) inhibitor which exists only in G1. The results suggest that dual downregulation of CDK, one via the APC and the other via the CDK inhibitor, is a universal mechanism that is used to arrest cell cycle progression at G1. Introduction The molecular bases of commitment to the proliferative cycle or controlled cell-cycle arrest are of fundamental importance in eukaryotic growth control. Depending on the external cues, dividing cells cease division and arrest at a specific cell-cycle stage. Negative growth factors, tumour suppressor genes and deleterious stress all lead to such stage-specific cell-cycle arrest. For example, during development in multicellular organisms, developing cells arrest at G1 before S phase in response to nutritional deprivation or differentiation stimuli following a limited number of cell divisions. Mutations that are unable to do this result in a failure to differentiate normally (Thomas et al., 1994; Kipreos et al., 1996). Exposure to DNA-damaging agents or spindle drugs results in a transient G1 or G2 arrest or mitotic blockage, respectively, to ensure the repair of DNA lesions or the fidelity of chromosome segregation (collectively called checkpoint; Hartwell and Weinert, 1989; Murray, 1995). At the heart of cell-cycle regulation lie a number of master regulating cyclin-dependent kinases (CDKs) and their regulatory partner cyclins (Nurse, 1990). Cells have developed a number of molecular strategies to regulate CDK activities and the endpoint of checkpoint pathways often lies in regulation of CDKs (Morgan, 1995; Nurse, 1997; Hwang et al., 1998; Kim et al., 1998). Recent molecular analysis has shed light upon the importance of the ubiquitin-dependent proteolysis in the regulation of cell-cycle progression, which is catalysed by a large mechanochemical complex called a proteasome. Ubiquitin, a 76-residue ubiquitous protein, acts as a recognition tag for proteasome-dependent proteolysis (Hochstrasser, 1996; Hershko, 1997). Ubiquitin is transferred to substrate proteins in a multiubiquitinated form via a series of enzymatic reactions comprising E1 (a ubiquitin-activating enzyme), E2 (a ubiquitin-conjugating enzyme) and sometimes E3 (a ubiquitin ligase). E3 is a crucial determinant of substrate selectivity and the timing of degradation. Classical E3s include Ubr1, which targets proteins for degradation by the N-end rule pathway (Varshavsky, 1996), and HECT-domain-containing proteins such as E6AP, which is responsible for the ubiquitination of p53 in the presence of human papilloma virus-encoded protein E6 (Huibregtse et al., 1995). Recent analysis indicates that there are more E3s than was originally thought, both HECT-related members and other novel E3s. The anaphase promoting complex (APC)–cyclosome is a large E3 complex which was discovered as a factor required for the ubiquitin-dependent degradation of B-type cyclins at anaphase (Hershko et al., 1994; King et al., 1995; Sudakin et al., 1995). In sharp contrast to classical E3s, which are composed of a single component, the APC–cyclosome is a 20–22S multiprotein complex. Recent molecular analyses, both genetic and biochemical, have begun to identify components of this complex. Many of the components isolated to date are well conserved in evolution from yeasts to vertebrates (Irniger et al., 1995; King et al., 1995; Tugendreich et al., 1995; Peters et al., 1996; Yamashita et al., 1996; Zachariae et al., 1996, 1998; Yamada et al., 1997; Yu et al., 1998). These include members of a TPR-repeat family, Cdc27/Nuc2/Apc3, Cdc16/Cut9/Apc6, Apc7 and Cdc23/Apc8, and non-TPR members such as Cut4/Apc1, Rsi1/Apc2, Apc4 and Apc5. Apart from the E3 activities of the APC–cyclosome as a whole complex, however, the specific or distinct functions, if any, of each subunit in the ubiquitination reaction remain to be addressed. Moreover, mechanisms of how the APC–cyclosome is activated and inactivated during cell-cycle progression are unknown. Mitotic cyclins were the first substrates to be identified for the APC–cyclosome, the destruction mechanisms of which have been most extensively characterized (Glotzer et al., 1991; King et al., 1996). A nine-residue sequence in B-type cyclins is recognized for ubiquitination and degradation, and is called the 'destruction box'. The destruction box exists not only in B-type cyclins but also in several other proteins including budding yeast Pds1 and Ase1 and fission yeast Cut2, which have also been shown to be degraded via the APC–cyclosome (Cohen-Fix et al., 1996; Funabiki et al., 1996; Juang et al., 1997). Accordingly, the APC–cyclosome is required for the timely destruction of a group of proteins whose role is to induce anaphase, by different means, including inactivation of CDK/cyclins, sister chromatid separation and spindle morphogenesis. Fission yeast cells arrest cell-cycle progression at G1 phase upon nutritional starvation. This arrest is a prerequisite for sexual differentiation (Egel and Egel-Mitani, 1974; Kumada et al., 1995). We have previously proposed the existence of mechanisms that actively restrain DNA replication under starvation. This proposal was based on the analysis of sterile mutant nuc2-663 (Kumada et al., 1995). In nuc2-663 mutant cells, instead of arresting at G1, one additional S phase occurs under starvation conditions, and results in a G2 arrest with defects in sexual differentiation. Interestingly, Nuc2 is also required for the mitotic cycle, since nuc2-663 shows, in addition to sterility, temperature-sensitive growth with metaphase arrest at the restrictive temperature (Hirano et al., 1988). It is of particular interest that a subsequent study showed that Nuc2, a homologue of budding yeast Cdc27 and vertebrate Apc3, is an ubiquitous subunit of the APC–cyclosome (King et al., 1995; Yamada et al., 1997). This has prompted us to perform systematically a novel genetic screen, in the hope that we could identify more genes that are involved in the degradation of B-type cyclins in order to maintain the G1 arrest required for sexual differentiation. In the current study, we have undertaken a large-scale screen of mutants which show phenotypes similar to nuc2, and are sterile due to an inability to arrest in G1 phase. We have identified apc10+, which encodes a novel regulator of the APC–cyclosome. In addition, genetic analysis of isolated mutants adds support for the simple scheme, in which G1 cell-cycle arrest requires dual approaches to the downregulation of CDK activity, one acting via the degradation of B-type cyclins and the other via a direct physical inhibition by the CDK inhibitor. Results Isolation of novel types of sterile mutants which fail to arrest at G1, with defects in the inhibition of S phase In order to identify systematically genes that are required for proper cell-cycle arrest at G1 phase following nitrogen starvation, we undertook a large-scale screen for sterile mutants. We focused on mechanisms that restrain DNA replication under starvation conditions. To isolate mutants which are defective in this block to DNA replication, we chose the following strategy. Sterile mutants were first isolated by iodine staining and then the DNA content of each mutant under starvation conditions was measured with flow cytometry (FACS). To distinguish mutants which are defective in restraining DNA replication from those defective in entering mitosis, which also arrest at G2 (Shiozaki and Russell, 1996), the morphology was examined. We sought strains that arrested at a size as small as that of the wild type (<8 μm), which usually divides twice and arrests at G1 before S phase; mutants that are defective in the G2/M transition typically arrest at a larger cell size (Kumada et al., 1995). A total of 1008 independent sterile mutants were identified following iodine staining (see Materials and methods). FACS analysis showed that, as expected, most of these mutants (95%) were normal in terms of cell-cycle arrest at G1; however, 56 mutants arrested with the 2C DNA content typical of G2 rather than G1 (1C) cells. Cell morphology showed that 28 of these mutants appeared to arrest at a small cell size (<8 μm) and so were selected for further analysis. Figure 1 shows a typical example of DNA content (left panels), cell size distribution (middle) and cell morphology (right) in wild type (A) and three sterile mutants that typify the class, we isolated (B, apc10; C, rum1; and D, ste9/srw1, see below). For comparison, an example of a G2-arrest mutant with a larger cell size is also shown (E). Figure 1.DNA content and cell size of novel sterile mutants under nitrogen starvation. (A) Wild type (TP108–3A, Table III), (B) apc10-27, (C) rum1-7, (D) ste9-12 and (E) 6-42 cells were grown in a low nitrogen medium supplemented with leucine and uracil (25 μg/ml) for 48 h at 27°C. DNA content (left), forward scattering (middle) and cell morphology (right panels) are shown. Bar, 10 μm. Download figure Download PowerPoint Classification of mutants defective in G1-arrest and S-phase inhibition In addition to the sterile phenotype, 14 out of 28 strains showed temperature-sensitive (ts) growth. In particular, nine of these mutants showed the 'cut' phenotype at the restrictive temperature (Figure 2), as septation and cytokinesis occurred in the absence of chromosome segregation. This resulted in two incompletely separated daughter cells associated in a side-by-side manner (Hirano et al., 1986). These phenotypes are extremely similar to the phenotype of nuc2-663 cells grown at the lower range of its restrictive temperature (33°C; Kumada et al., 1995). However, plasmids containing nuc2+ did not suppress either the temperature sensitivity or the sterility of these nine mutants, indicating that they are unlikely to be allelic to nuc2. Pair-wise complementation tests using protoplast fusion showed that nine mutants were all allelic and represent a novel locus, designated apc10 (Table I). Figure 2.Cut phenotypes of the ts apc10 mutant. The apc10-27 mutant (KK1427, Table III) was grown in rich medium at 26°C and shifted to 35.5°C. Cells were collected hourly, fixed with 3.7% formaldehyde, stained with DAPI [(A) shows 4 h samples, with cut cells indicated by arrows] and (B) the percentage of cut cells was also measured. Bar, 10 μm. Download figure Download PowerPoint Table 1. Complementation groups and characteristics of G1-arrest-defective mutants Loci Alleles Suppression by multicopy tsa rum1+ ste9+ apc10 3, 5, 27, 71, 202 yesb no − 104, 127, 193, 207 no no − ste9 12, 39, 44, 53, 120, 150, 165, 223 no yes + rum1 7, 173, 204 yes no + sat1 11 yesb no − sat2 32 no yes + sat3 48 no no + sat4 86 no no − sat5 100 no no + sat6 103 no no + sat7 110 no no − sat8 146 no no + a Temperature sensitive; + indicates growth at 36°C. b Only sterile phenotypes were suppressed. Our preliminary analysis of known sterile mutants by FACS analysis showed that, in addition to the nuc2 mutant, ste9 and rum1 are defective in arresting at G1 under nutritional starvation. Complementation tests of the remaining 19 mutants with the ste9 and rum1 mutants showed that eight were allelic to ste9 and three were allelic to rum1, and that none of them was allelic to nuc2 (Table I). The other eight mutants all belonged to distinct complementation groups, which were designated sat1-8 (starvation-induced arrest). Next we asked whether multicopy plasmids containing either rum1+, nuc2+ or ste9+/srw1+ suppressed the sterile mutants that were isolated in this study. The sterile, but not ts phenotypes of five out of nine apc10 alleles were suppressed by elevating the dosage of the rum1+ gene (Table II). This result suggests that since Rum1 is the CDK-inhibitor of Cdc2/Cdc13 and Cdc2/Cig2 (Correa-Bordes and Nurse, 1995; Jallepalli and Kelly, 1996; Martin-Castellanos et al., 1996), inhibition of CDK activity could partially bypass the normal requirement of Apc10 for the mating pathway under starvation conditions. Table 2. Suppression of sterile phenotypes by high gene dosage of rum1+ Mutantsa % of conjugationb Vector rum1+ Δrum1 <1 100c rum1-7 <1 63 apc10-3 2.5 30 apc10-5 3.9 39 apc10-27 5.5 24 apc10-71 5.8 45 apc10-104 <1 <1 a The following strains were used: Δrum1 (PN1012, Table III), rum1-7 (KK137), apc10-3 (KK143), apc10-5 (KK135), apc10-27 (KK1427), apc10-71 (KK1371) and apc10-104 (KK13104). b Mutant cells containing pREP3-rum1 were streaked on minimal plates supplemented with uracil and thiamine and incubated for 3 days at 28°C. The percentage of cells which had conjugated or sporulated was scored. At least 500 cells were observed in each case. c A conjugation frequency of ∼10% (65 out of 698 cells) was obtained. In the table, this value is shown as 100%, and the efficiency of conjugation of each mutant is shown relative to this value. Cloning of the apc10+ gene The apc10+ gene was cloned by complementation of the temperature sensitivity using a fission yeast genomic library. Five different plasmids which contained overlapping inserts were obtained. These plasmids suppressed both the ts and the sterile phenotype. Restriction mapping and subsequent subcloning indicated that the internal XhoI site was essential for suppressing activity (Figure 3A). The nucleotide sequence of this region revealed one 189-amino-acid open reading frame (ORF), which was interrupted by two putative introns. The identity of this ORF as the apc10+ gene was confirmed by subcloning the ORF into a plasmid under the direct control of the thiamine-repressible nmt1 promoter (pREP1-apc10+, Figure 3A; see Materials and methods). pREP1-apc10+ was slightly toxic when transformed into wild-type cells and modest cell elongation was observed (on average cells were 30% longer than the wild type; Figure 3B). FACS analysis indicated that no increase in ploidy was evident, suggesting G2 delay in the cell cycle (data not shown). Figure 3.Cloning, overexpression and disruption of the apc10+ gene. (A) The restriction map and subcloning strategy for the apc10+ gene are shown. Two introns are depicted with open boxes. A single XhoI site was essential for complementation as, when it was disrupted (shown with closed triangle), the resulting subclone lost the complementing activity. (B) Cells which had been transformed with a vector (pREP1, left, or pREP1–apc10+, right) were streaked on minimal medium (derepressed) and incubated at 30°C for 2 days. (C) Tetrad analysis of a heterozygous apc10+. Δapc10 diploid (KK300, Table III) is shown. (D) Germinating cells from a nonviable apc10-deleted spore are shown. After germination, cells divided several times and arrested with either 'cut' (arrow heads) or elongated cell (arrow) phenotypes. Bar, 10 μm. Download figure Download PowerPoint The apc10+ gene was disrupted with the PCR-based gene targeting method (Bähler et al., 1998). Two viable and two inviable spores were obtained from 20 dissected tetrads and all the viable colonies were Ura−, indicating that the apc10+ gene is essential for cell viability (Figure 3C). Inviable spores germinated, divided several times and then ceased division. Two kinds of terminal morphology was observed; elongated cells which occasionally had multiple septa (Figure 3D, arrow) or with a cut phenotype (arrow head), which was similar to the defective phenotypes of the ts apc10 mutants. The apc10+ gene encodes an evolutionarily conserved protein A database search revealed that Apc10 is a highly conserved protein: homologues exist in vertebrates, invertebrates and budding yeast, with 49% identity between the fission yeast and human proteins in the central 80 amino acid residues and 38% between fission and budding yeasts in the central 150 amino acid residues (the reported nucleotide sequence of the human homologue of Apc10 is incomplete and its C-terminus is unsequenced; Figure 4A and B). During the preparation of this manuscript, we learned that the budding yeast homologue of Apc10, called Doc1, had been identified and proposed to be a component of the APC–cyclosome (Hwang and Murray, 1997). Figure 4.Amino acid sequence comparison of fission yeast Apc10 with putative homologues from other organisms. (A) An overall homology shared between Apc10-related proteins is shown. Species are indicated by the following abbreviations: Sp, fission yeast; Hs, human (DDBJ/EMBL/GenBank accession No. AA234328); Mm, mouse (AA472445); Dm, fly (AA141768); Ce, worm (Z73972); and Sc, budding yeast (Z72762). A human protein (KIAA0076, D38548) which shows a homology to both Apc10 (closed box) and the C-terminal region of cullins and Apc2 (shadowed box) is also shown. (B) The amino acid sequence of the central 150 amino acid residues of Apc10 is compared with that of Apc10 homologues in other species and to the human ORF KIAA0076. Closed boxes with white letters show identical amino acids, while shadowed boxes with black letters represent conservative amino acids. Download figure Download PowerPoint In addition to homologues from different organisms, there is a human gene (KIAA0076) which encodes a protein distantly related to Apc10 (Figure 4B). Although the function of this human protein is not known, it is interesting to note that it also contains a region of 200 amino acid residues near the C-terminus which is related to sequences in cullins, a protein family involved in ubiquitin-dependent protein degradation (Figure 4A, 30% identify to cullin-1; Kipreos et al., 1996; Mathias et al., 1996; Willems et al., 1996). In the apc10 mutant the level of mitotic B-type cyclin Cdc13 is not downregulated at G1 phase Given the phenotypic similarity between apc10 and the nuc2 mutant that is defective in one of the APC–cyclosome subunits (Kumada et al., 1995; Yamada et al., 1997), the possibility that Apc10 is also involved in regulating the stability of B-type cyclin Cdc13 was addressed. The APC–cyclosome is active from the onset of anaphase throughout the G1 phase and continuously degrades Cdc13 (Yamano et al., 1996; Correa-Bordes et al., 1997). The level of Cdc13 in G1-arrested cells was measured in ts mutants defective in the Cdc10 transcription factor which arrest in G1 at Start (cdc10-129 and apc10-27cdc10-129). As shown in Figure 5A, Cdc13 levels decreased in the cdc10 mutant at the restrictive temperature (lanes 1–3), whereas in the double mutant no significant decrease was observed (lanes 4–6). This was confirmed by densitometric measurement of the level of Cdc13 at each time point, using tubulin as an internal loading control (Figure 5B). The stability of Cdc13 in the apc10cdc10 mutant is not ascribable to a failure of the cells to arrest at G1 in the double mutant, as FACS analysis confirmed the accumulation of G1-arrested cells at 2 and 4 h at the restrictive temperature (Figure 5C). It should be noted that in the apc10cdc10 mutant, the kinetics of G1 arrest were somewhat delayed compared with a single cdc10 mutant. These data indicate that Apc10 is required for the degradation of Cdc13 at G1 phase. Figure 5.Apc10 is required for the degradation and ubiquitination of Cdc13. (A) cdc10 single (lanes 1–3, cdc10, Table III) or apc10cdc10 double mutants (lanes 4–6, TP426–9C) were grown in rich medium at 26°C and shifted to 35.5°C. Cell extracts were prepared at 0 (lanes 1 and 4), 2 (lanes 2 and 5) and 4 h (lanes 3 and 6) at 35.5°C, run on SDS–PAGE, immunoblotted with anti-Cdc13 (upper) and anti-tubulin antibodies (lower panel). (B) Densitometric calibration of the data in (A) is shown. At each time point, the relative level of the Cdc13 was measured by using the level of tubulin as an internal control. The level at time 0 was assigned 100%. (C) Samples used in (A) were fixed and processed for flow cytometry. (D) mts2 single (mts2, Table III) or apc10mts2 double mutants (TP429–2A) containing pREP1–cdc13+ were grown in minimal medium at 25°C in the presence of 2 μM thiamine. Cells were harvested by filtration, and expression of the cdc13+ gene was induced by replacing the medium with minimal medium which lacked thiamine (at a density of 2×105 cells/ml). After further growth for 16 h at 25°C, the culture was divided into two, and one half was incubated at 25°C (lanes 1, 3, 5 and 7), whilst the other half was shifted to 35.5°C (lanes 2, 4, 6 and 8). Both cultures were then incubated for a further 4 h. Cell extracts were prepared from these cultures and immunoblotted with either anti-Cdc13 (lanes 1–4) or anti-ubiquitin antibody (lanes 5–8). Download figure Download PowerPoint Apc10 is required for the ubiquitination of Cdc13 The results presented above suggest that Apc10 is involved in some step in the pathway resulting in the proteolysis of Cdc13. If Apc10 is a subunit or a regulator of the APC–cyclosome, Cdc13 should not be ubiquitinated in the apc10 mutant. To address this question, a double mutant containing mutations in apc10 and a proteasome mutant were constructed. mts2-1 is a ts mutant which is defective in subunit 4 of the 26S proteasome (Gordon et al., 1993) and polyubiquitinated intermediate forms of Cdc13 accumulate in this mutant at the restrictive temperature (Yamashita et al., 1996). Cdc13 was expressed under the control of the thiamine-repressible nmt1 promoter in mts2-1 and apc10-27mts2-1 mutants, and the temperature of the culture was shifted to the restrictive temperature. Immunoblot analysis of protein extracts prepared from samples at different time points at the permissive (25°C) and at the restrictive temperature (35.5°C) is shown in Figure 5D. Consistent with a previous report (Yamashita et al., 1996), the higher molecular forms of Cdc13, which were detected with anti-ubiquitin antibody, appeared after temperature shift-up of the mts2-1 mutant culture (Figure 5D, compare lanes 1, 2, 5 and 6). In sharp contrast, however, no higher forms of ubiquitinated Cdc13 were detectable in the apc10mts2 double mutant (Figure 5D, lanes 3, 4, 7 and 8). These results demonstrate that Apc10 is required for ubiquitination and degradation of Cdc13. Apc10 genetically and physically interacts with a universal subunit of the APC–cyclosome in vivo As a first step towards addressing the question as to whether Apc10 is a subunit or a regulator of the APC–cyclosome, the possibility of a physical interaction between Nuc2 and Apc10 was examined. The N-terminus of Apc10 was tagged with a c-myc peptide (pREP41–myc–apc10+). This plasmid rescued the apc10 deletion, indicating that tagging did not interfere with the Apc10 function. Immunoprecipitation was performed with the deletion strain containing pREP41–myc–apc10+ using anti-myc antibody. A non-related anti-HA antibody (or mock) was used as a control. As shown in Figure 6A, Apc10 formed a complex with Nuc2 (lanes 3). Co-precipitation was not observed when anti-HA (or mock) was used as a primary antibody (lanes 2 and 4). Unfortunately, myc–Apc10 migrated as a 24 kDa band, which overlapped with a light chain of IgG in these extracts (25 kDa). However, the precipitation of myc–Apc10 was evident because the intensity of the band around 24 kDa was much stronger with anti-myc antibody than that with anti-HA (Figure 6A, compare lanes 3 and 4). This result supports the notion that Apc10 is a component of the APC–cyclosome, although it does not mean that it is a stoichiometric subunit of this complex (see below). Figure 6.Apc10 interacts with a subunit of the APC–cyclosome but is not a core component. (A) A Δapc10 strain containing a plasmid carrying myc-tagged apc10+ (pREP41–myc–apc10+, KK301, Table III) was grown in minimal medium in the absence of thiamine (derepressed conditions) at 30°C. A cell extract (1 mg) was prepared and immunoprecipitation was performed using either anti-myc (lane 3), anti-HA antibody (lane 4) or mock treatment of no antibody (lane 2). Total cell extracts (30 μg) were also run (lane 1). Immunoblotting was performed with anti-Nuc2 (upper panel) or anti-myc (lower) antibody. Due to a similar size between myc–Apc10 (24 kDa) and the IgG light chain (25 kDa, IgG-L), these two proteins overlapped on the gel (lanes 3 and 4). However, note that in contrast to a clear difference in the intensity of bands around 24 kDa which were precipitated with anti-myc and with anti-HA (lower panel in lanes 3 and 4), the amount of the IgG heavy chain (upper panel) was indistinguishable with these two antibodies (IgG-H, 55 kDa, cross-reacted with anti-rabbit secondary antibody, upper panel). (B) Cell extracts prepared from the same strain used in (A) were separated through a 15–40% sucrose gradient centrifugation and fractions were analysed by immunoblotting with anti-Nuc2 (top), or anti-myc antibody (bottom panel). The same filter used in immunoblotting with anti-Nuc2 was stripped off and reprobed with anti-Cut9 antibody (middle). Positions of sedimentation markers (19.2S, thyroglobulin; 4.5S, BSA) and fraction numbers (frac. no.) are also shown. Total cell extracts (15 μg) used in centrifugation were run in the leftmost panel (crude). The second lane was blank (no proteins were run). (C) Cell extracts prepared from an apc10-27 strain (TPR27-1, Table III) incubated at 35.5°C for 4 h were separated through a sucrose gradient as in (B), and immunoblotting was performed. Download figure Download PowerPoint Next, the possibility of a genetic interaction between nuc2 and apc10 was addressed. It was found that ts nuc2-663 and apc10-27 were synthetically lethal: no double mutants were viable at 26°C, the permissive temperature for both single mutants. It should be noted that a similar synthetic lethality was also reported between nuc2-663 and cut9-665, which is defective in another subunit of the APC–cyclosome (Samejima and Yanagida, 1994; Yamada et al., 1997). Apc10 is not a core component of the APC–cyclosome To examine whether Apc10 exists in the 20S APC–cyclosome complex, a sucrose gradient centrifugation was performed with cell extracts prepared from exponentially growing cells which contain the myc-tagged apc10+ gene (in the chromosomal apc10 gene deletion background, KK301, Table III). As reported previously (Yamashita et al., 1996), the majority of the Nuc2 protein sedimented around 20S (Figure 6B, top panel). Also, Cut9 co-sedimented with Nuc2, although compared with Nuc2, Cut9 also s