A Screen for Genes Involved in the Anaphase Proteolytic Pathway Identifies tsm1⁺, a Novel Schizosaccharomyces pombe Gene Important for Microtubule Integrity

Corresponding author: J. Richard McIntosh, MCD Biology, Campus Box 347, University of Colorado, Boulder, CO 80309-0347, richard.mcintosh{at}colorado.edu (E-mail).

Communicating editor: M. D. ROSE

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The growth of several mitotic mutants of Schizosaccharomyces pombe, including nuc2-663, is inhibited by the protease inhibitor N-Tosyl-L-Phenylalanine Chloromethyl Ketone (TPCK). Because nuc2⁺ encodes a presumptive component of the Anaphase Promoting Complex, which is required for the ubiquitin-dependent proteolysis of certain proteins during exit from mitosis, we have used sensitivity to TPCK as a criterion by which to search for novel S. pombe mutants defective in the anaphase-promoting pathway. In a genetic screen for temperature-sensitive mitotic mutants that were also sensitive to TPCK at a permissive temperature, we isolated three tsm (TPCK-sensitive mitotic) strains. Two of these are alleles of cut1⁺, but tsm1-512 maps to a novel genetic location. The tsm1-512 mutation leads to delayed nuclear division at restrictive temperatures, apparently as a result of an impaired ability to form a metaphase spindle. After shift of early G₂ cells to 36°, tsm1-512 arrests transiently in the second mitotic division and then exits mitosis, as judged by spindle elongation and septation. The chromosomes, however, often fail to segregate properly. Genetic interactions between tsm1-512 and components of the anaphase proteolytic pathway suggest a functional involvement of the Tsm1 protein in this pathway.

THE onset of anaphase is a key cell cycle transition whose success and precision are vital for cell survival. A primary role in the start of anaphase is thought to be played by a large, multisubunit complex known as the Anaphase Promoting Complex (APC) or cyclosome (reviewed in KING et al. 1996 ). Studies involving multiple organisms suggest that the APC functions as a cell-cycle-regulated ubiquitin-protein ligase, mediating the poly-ubiquitination of specific proteins that must be destroyed for anaphase to begin (KING et al. 1995 ; SUDAKIN et al. 1995 ; ZACHARIAE and NASMYTH 1996 ; ZACHARIAE et al. 1996 ). Such proteins are degraded by the 26S proteasome, which displays multiple proteolytic activities (PETERS 1994 ; DESHAIES 1995 ; HILT and WOLF 1996 ; KING et al. 1996 ).

Despite recently acquired information about the structure and function of the APC and proteasome, little is known about the substrates of their mitotic activities (HOCHSTRASSER 1995 ). The first identified and best-studied target of the APC is cyclin B, whose degradation leads to inactivation of M-phase-promoting-factor (MPF) (GLOTZER et al. 1991 ; MURRAY 1995 ). It is clear, however, that the degradation of cyclin B alone cannot account for all anaphase events (HOLLOWAY et al. 1993 ; SURANA et al. 1993 ; IRNIGER et al. 1995 ; YAMANO et al. 1996 ). Only a few noncyclin targets of anaphase-specific proteolysis have thus far been identified. One of them is the fission yeast Cut2p, which localizes to mitotic spindles (UZAWA et al. 1990 ; FUNABIKI et al. 1996B ). The failure to degrade Cut2p blocks sister chromatid separation by a mechanism that has not yet been identified, but it is thought to involve an interacting protein, Cut1p (FUNABIKI et al. 1996A , FUNABIKI et al. 1996B ). The Saccharomyces cerevisiae proteins Pds1p (COHEN-FIX et al. 1996 ; YAMAMOTO et al. 1996 ) and Ase1p (PELLMAN et al. 1995 ; JUANG et al. 1997 ) are also thought to be important targets for APC-dependent proteolysis during anaphase. Several other proteins have been reported to be degraded at anaphase, but it remains to be determined whether their degradation occurs via a ubiquitin-dependent mechanism (e.g., BROWN et al. 1994 ; STRATMANN and LEHNER 1996 ). Clearly, more components of the anaphase proteolytic pathway remain to be identified before these detached examples can be connected to one or more coherent control mechanisms. The goal of the this study was to identify additional genes whose functions are important at anaphase onset.

To seek such genes we have tested temperature-sensitive mitotic mutants in fission yeast for their sensitivity to the protease inhibitor N-tosyl-L-phenylalanine chloromethyl ketone (TPCK). It has long been known that mutants that fail to grow and divide at a restrictive temperature can also be hypersensitive at permissive temperatures to drugs that perturb the same cell cycle process. For example, Schizosaccharomyces pombe cells that carry temperature-sensitive tubulin mutations are often sensitive at their permissive temperature to sublethal concentrations of microtubule depolymerizing drugs (UMESONO et al. 1983 ). This phenomenon is usually attributed to the synergy of two factors, neither of which is sufficient in itself to cause lethality: an impaired function of the gene product, resulting from a mutation, and a perturbation of the same or a closely related process by the drug. We reasoned that an increased sensitivity of a temperature-sensitive mitotic mutant to a protease inhibitor might indicate the presence of a mutation in a gene that is involved in the proteolytic pathway that acts at anaphase onset.

TPCK was chosen for this study because it is likely to inhibit an important step in the anaphase initiation pathway. TPCK or the similar compound TLCK (N-p-tosyl-L-lysine chloromethyl ketone) are potent inhibitors of cyclin B degradation in clam embryo extracts in vitro (LUCA and RUDERMAN 1989 ). TLCK can transiently delay fertilized sea urchin eggs at their metaphase-anaphase transition (PENN et al. 1976 ) and perturb mitotic progression of either clam embryos (F. LUCA, personal communication) or mammalian cells (E. GRISHCHUK and J. R. MCINTOSH, unpublished results). Initially, both compounds were used for drug sensitivity tests on different strains of S. pombe (see below), but TPCK proved to be more suitable, due largely to its greater chemical stability. Here we show that several mitotic mutants of S. pombe, including nuc2-663, are hypersensitive to TPCK at their permissive temperatures. A genetic screen for nuc2^ts-like strains that fail in mitosis at elevated temperatures and are hypersensitive to TPCK at 25° has identified three TPCK-Sensitive-Mitotic (TSM) mutants, one of which (tsm1-512) contains defective microtubules and fails in accurate chromosome segregation under restrictive conditions.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

S. pombe methods:
The yeast strains used in this study are listed in Table 1 and Table 2. All media and growth conditions were as described by MORENO et al. 1991 , unless otherwise stated. Edinburgh minimal medium (EMM) and malt extract were purchased from Bio 101, Inc. (Vista, CA); yeast extract and agar were purchased from Difco (Detroit). Yeast extract medium with supplements (YES) was used, unless otherwise stated. Solid medium was made by adding 2% agar. Phloxine B (Sigma, St. Louis) was added up to 5 mg/liter to assess cell health: because sick or dead cells cannot pump out this red dye, they consequently become stained.

Genetic techniques:
For complementation tests involving mutants sensitive to TPCK or 36°, heterozygous diploid strains were obtained by using two complementing alleles of the same nutritional marker, ade6-M210 and ade6-M216. For tsm1-512 and nuc2-663 the tests were also performed with stable diploid strains obtained by using mat2-102 (McI#189, Table 2) and mat2-102 leu1-32 (McI#192, Table 1), respectively. For crosses between nuc2-663 and other strains, pnuc2⁺ LEU2 plasmid was used to complement the sterility of the nuc2^- strain. After the heterozygous diploids were obtained by selection on EMM with appropriate supplements, the pnuc2⁺ plasmid was lost by allowing growth without selection, and the leu^- diploids were sporulated.

All genetic crosses were performed at 25°, except those that involved cold-sensitive alleles of nda2⁺ and nda3⁺, which were done at 29°. More than 49 tetrads were examined for crosses between tsm1-512 (strains McI#190 and McI#199, Table 1) and each of the following strains: cut4-533 (FY201, Table 2), mts2-1 (McI#142, Table 2), mts3-1 (McI#143, Table 2), nda2-52 (PN779, Table 2), nda3-311 (PN780, Table 2), and nuc2-663 (McI#131, Table 1). Tetrad analysis was also used to identify double mutants between tsm1-512 and cdc2-33 (PN54, Table 2), cdc13-117 (strain PN117, Table 2), cut1-tsm2 (strain 637BC1-d, Table 1), cut2-364 (FY249, Table 2), and cut9-665 (FY205, Table 2).

Test for TPCK-sensitivity:
TPCK was purchased from Sigma. A stock solution of TPCK (500 mg/ml) was prepared in dimethyl sulfoxide (DMSO) and stored at -20°. Sensitivity to this drug was assessed by replica plating freshly streaked or patched cells on YES dishes containing 1% DMSO, 5 µg/ml phloxine B, and 120 µM (or as indicated) TPCK. Control dishes contained 1% DMSO and phloxine B. After 48 hr of incubation at 25° (or 24 hr at 32° for cold-sensitive strains) the colors of strains grown on control and TPCK-containing dishes were compared. TPCK-sensitive strains were identified by their pinker color on TPCK-containing dishes, due to the accumulation of phloxine in dead or dying cells. Table 2 presents average scores awarded by four different individuals in at least three experiments.

Dilution assays (Figure 1 and Figure 3) were performed using log-phase cells that were spun and resuspended to achieve 10⁸ cells/ml (leftmost columns). Serial dilutions in YES were made in a 96-well plate; they were 10-fold for the first two columns and 5-fold for the remaining three columns. Cells were transferred to YES dishes with or without TPCK and incubated at 25° for 3 days to allow colony growth.

View larger version (45K): In this window In a new window Download PPT slide   Figure 1. TPCK sensitivity assessed by serial dilution. Strains were analyzed for growth at 25° on YES dishes, with and without TPCK, by serial dilution of the equivalent number of cells. The nuc2 strain used is McI#129 without the pnuc2+ plasmid (Table 2); genotypes for other strains are as described in Table 2. Control plate (left panel) contained 1% DMSO and no TPCK. Plate on the right panel contained 1% DMSO and 120 µM TPCK. Photographs were taken on the third day of growth. Variations in the colony size between different strains on the control dish are mostly due to the background nutritional markers. Note that the degree of sensitivity for different TPCK-sensitive strains corresponds well to the results presented in Table 2.

View larger version (36K): In this window In a new window Download PPT slide   Figure 2. TPCK sensitivity of nuc2-663. (A) Plating assay at 25° on YES dishes supplemented with 1% DMSO and TPCK at different concentrations. Four days after plating of the equivalent number of cells, the surviving colonies were counted and their number was expressed as a percent of the number of colonies that grew without TPCK. (B) wt B and nuc2-663 cells incubated at 25° in YES, with or without 80 µM TPCK. Cells were fixed and stained with DAPI and Calcofluor. a, wt B cells incubated with TPCK and 1% DMSO for 8 hr; b, nuc2-663 cells incubated with 1% DMSO and no TPCK for 8 hr; c and d, nuc2-663 cells incubated with TPCK and 1% DMSO for 8 hr; e and f, nuc2-663 cells incubated with TPCK and 1% DMSO for 21 hr. Arrow in f points to an elongated cell that has at least five nuclei, suggesting that DNA synthesis can continue in the presence of TPCK. Bar, 5 µm.

View larger version (49K): In this window In a new window Download PPT slide   Figure 3. TPCK sensitivity of tsm mutant strains. Strains were analyzed for growth at 25° with and without TPCK by serial dilution of the equivalent number of cells. All tsm strains are leu1-32 ura4-D18 h-. Genotypes for the other strains are given in Table 2. Photographs were taken on the fourth day of incubation of plates containing 1% DMSO (left) or 1% DMSO and 112 µM TPCK (right).

TPCK-sensitivity playing assays at 25° (Figure 2A) were performed by plating log-phase wtB and nuc2-663 cells (strain McI#129 without plasmid pnuc2⁺, Table 2) onto YES agar that contained 1% DMSO and different concentrations of TPCK; fresh plates were poured for each experiment. Figure 2B presents results from a single experiment, but similar results were obtained in two additional experiments. The exact percent of surviving colonies was variable from one experiment to the next, probably because of the variability in TPCK concentration that resulted from its instability in the hot medium used to pour plates.

To examine whether the TPCK sensitivity of nuc2-663 cosegregated with its temperature sensitivity, the nuc2-663 strain McI#120 (Table 1) was initially crossed to a wild-type strain (wt A, which is PN513, Table 2). The results suggested that there was a second site modifier of nuc2-663 in one of these two strains. For example, at 25° on control plates containing phloxine alone, ~50% of the temperature-sensitive colonies grown from random spores appeared smaller and more darkly stained than the remaining 50% of the temperature-sensitive progeny. Moreover, some TPCK-sensitive colonies were not temperature sensitive. Subsequent crosses between the progeny were performed to isolate nuc2-663 strains (e.g., McI#129, Table 1) and wild-type strains that were more representative of the nuc2-663 background (e.g., wt B, which is McI#126, Table 2). When these strains were backcrossed to each other, the temperature-sensitive progeny were no longer heterogeneous; in these strains, TPCK sensitivity and temperature sensitivity always cosegregated (total of 530 progeny colonies examined). However, temperature-sensitive progeny from the crosses between the isolated nuc2-663 (e.g., McI#129, Table 1) and original wt A strain remained heterogeneous, even though temperature and TPCK sensitivities always cosegregated (total of 306 progeny colonies examined). The TPCK sensitivities of different nuc2-663 strains appear to be identical (e.g., Table 2). Note that examples of phenotypic differences between distinct genetic backgrounds are rare in S. pombe.

Screen for TPCK-sensitive mutants:
Ethylmethane-sulfonate (EMS) mutagenesis was performed as described by ALFA et al. 1993 . Nine different experiments were carried out to screen ~162,000 colonies; lethality averaged 50%. Half of the mutageneses were done with wt A and half with wt B. After 3 hr of incubation in EMS, cells were washed in 10% sodium thiosulfate, followed by two washes in YES medium, and plated on YES dishes at 25°. Three days later, the colonies were tested for TPCK sensitivity at 25°. Colonies that accumulated more phloxine on TPCK-containing plates than on control plates were patched or streaked on YES dishes and tested for growth at 36°. Mutant phenotypes of the temperature-sensitive strains were examined using a fluorescence microscope to visualize DNA and the septum, as described below. The three tsm mutants studied here were all isolated in the wt A background. They were backcrossed first to strain PN556 and then to wt A.

Cytological techniques:
The cellular localization of DNA was visualized by staining with 4',6'-diamidino-2-phenylindole dihydrochloride (DAPI; Sigma). Septum material was stained with Calcofluor white M2R (Sigma). Cells were fixed either in 3.7% formaldehyde (Electron Microscopy Sciences, Fort Washington, PA), as described by MORENO et al. 1991 , or, if the same samples were to be used for flow cytometry, in ice-cold 70% ethanol. A typical staining solution contained 50% glycerol, DAPI (1 µg/ml), Calcofluor (50 µg/ml), and p-phenylenediamine (1 mg/ml) to slow the rate of fluorescence photobleaching.

Immunofluorescent staining of microtubules was carried out as described in ALFA et al. 1993 with slight modifications. Freshly opened formaldehyde was used for each experiment (Electron Microscopy Sciences). Cell wall digestion was carried out at 37° for about 25 min in 0.1 mg/ml Novozyme 234 (Sigma) and 0.3 mg/ml Zymolase 20T (ICN Biomedicals Inc., Costa Mesa, CA). By the end of the incubation period, 80–90% of cells had lost the integrity of their cell walls, as judged by a loss of birefringence in 1% sodium dodecyl sulfate (SDS). Microtubules were stained using mouse monoclonal antibodies against Drosophila -tubulin (kindly provided by Dr. M. FULLER, Stanford University) and Texas Red goat anti-mouse antibodies (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA). Stained cells were viewed with a Zeiss (Thornwood, NY) fluorescence microscope, using a x100 achromatic objective and either photographed on 35-mm film (Kodak TMAX p3200) or recorded in digital form, using an Empix CCD camera and the Metamorph imaging system (Universal Imaging, Inc., West Chester, PA). Images were processed using Adobe Photoshop (Adobe Systems, Inc.). For flow cytometry a protocol from ALFA et al. 1993 was used with slight modifications. Following fixation in 70% ethanol and using a rehydration in 50 mM sodium citrate, cells were incubated for 2 hr at 37° in 1 mg/ml RNase and then stained with 5 µg/ml propidium iodide (Sigma). Samples were analyzed using a FACScan (Becton Dickinson, San Jose, CA) and the software Lysis II.

Synchronized cultures:
Populations of wild-type and tsm-512 cells were synchronized using log-phase cells harvested by centrifugation. After resuspension in about 0.5 ml of YES, cells were loaded onto the top of a 26-ml linear gradient of lactose (10 to 26% lactose in YES). The gradient tubes were then spun in an HB-4 swinging bucket rotor, using a Sorvall (Newtown, CT) centrifuge at ~500 rpm for 9–10 min. The top 1–1.5 ml of the cell suspension was collected and the cells were washed, resuspended in YES at about 10⁶ cells/ml, and incubated at 36°.

Isolation of the cut1⁺ gene:
Standard molecular techniques were used as described in SAMBROOK et al. 1989 . For PCR we followed protocols from INNIS et al. 1990 . Sequencing was carried out using dideoxy chain termination (United States Biochemical, Cleveland). All S. pombe transformations were done using a lithium acetate procedure, described in MORENO et al. 1991 or as modified in ELBLE 1992 . An S. pombe genomic library in the shuttle vector pUR19 was the gift of Dr. A. M. CARR (BARBET et al. 1992 ). In this library we identified a genomic fragment that could rescue the temperature sensitivity of tms1-637. This fragment contained the cut1⁺ gene, as determined by restriction mapping, by partial sequencing with vector-specific primary, and by PCR with cut1⁺ specific primers. The cut1 gene used to transform different strains, including nuc2-633, was a 7.2-kb genomic fragment, identical at its 3' end but ~330 bp shorter at its 5' end than the pCUT1-1 fragment isolated by UZAWA et al. 1990 (Figure 4).

View larger version (52K): In this window In a new window Download PPT slide   Figure 4. Asynchronous cultures of tsm1-512 and isogenic wt A cells grown in YES and shifted at 0 hr to 36°. Graphs represent average results of three independent experiments. (A) Cell number (broken lines) was determined by counting with a hemocytometer. Colony forming units (CFU, solid lines) were determined by plating fixed volumes of a diluted cell suspension on YES dishes at 25° and counting the number of colonies several days later. Open symbols, control wt A cells; closed symbols, tsm1-512 cells (normalized to the same starting cell number). (B) Frequencies of cells with different abnormal morphologies among tsm1-512 cells grown in YES and shifted at 0 hr to 36°. Cell morphology at indicated times was determined after fixation in ethanol and staining with DAPI and Calcofluor. At least 300 cells were examined for each data point. , cells with condensed chromosomes (D, panel b); , cells with a septum but showing abnormal nuclear division (D, panels c–g); , overseptated cells (D, panel h) and cells undergoing cytokines following abnormal nuclear division, e.g., D, panel i. (C) Frequency of cells with septum and different configurations of nuclear DNA among tsm1-512 cells grown in YES and shifted at 0 hr to 36°. The sum of these phenotypes corresponds to the line marked with open squares on B; , abnormally separated chromosome masses (D, panels e and f); , condensed chromosomes (D, panel d); , "cut" phenotype (D, panel c); , interphase nucleus (D, panel g). (D) Examples of wild-type control cells (panel a) and tsm1-512 cells (panels b–i) grown in YES at 36°. Cells were fixed in ethanol, rehydrated, and stained with DAPI and Calcofluor. Panel b, cell with condensed chromosomes; c, "cut" phenotype; d, cell with septum and condensed chromosomes; e and f, cells with septum and abnormally separated chromosome masses; g, cell with septum and single interphase nucleus; h, overseptated cell; i, anucleated and binucleated daughter cells.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Several mitotic mutants of S. pombe, including nuc2-663, show increased sensitivity to the protease inhibitor TPCK:
To test our supposition that mutants defective in the pathway for anaphase initiation might be hypersensitive to TPCK, we examined 38 strains of S. pombe for growth at their permissive temperatures in the presence of 120 µM TPCK, as described in MATERIALS AND METHODS. Strains containing the mutations cut1-206, cut6-81, cdc15-140, and nuc2-663 were the most sensitive of all those tested (Table 2). This result was confirmed by an assay in which serial dilutions of these and wild-type strains were transferred to YES plates containing 120 µM TPCK (Figure 1). Growth of nuc2-663 at 25° is completely inhibited by this concentration of TPCK.

Strains containing the nuc2-663 mutation were examined in more detail. The TPCK and temperature sensitivities of nuc2-663 cells appear to result from the same mutation, as indicated by the meiotic cosegregation of these phenotypes (see MATERIALS AND METHODS). Moreover, both phenotypes were recessive and could be rescued by the nuc2⁺ gene on a plasmid. The TPCK sensitivity of this strain was further assessed by counting the number of colonies that grew from nuc2-663 or wild-type cells plated at 25° on YES dishes containing different concentrations of the drug. nuc2-663 cells have decreased viability in the presence of 20–100 µM TPCK when compared with isogenic wild-type cells (Figure 2A). To examine the phenotype of this drug sensitivity, log-phase cultures of nuc2-663 and wild-type control cells were incubated at 25° in liquid YES medium containing 80 µM TPCK. After 8 hr of incubation, wild-type cells continued to grow normally (Figure 2B, panel a). Many nuc2-663 cells, however, appeared to be blocked in mitosis with highly condensed or missegregated chromosomes and multiple septa (Figure 2B, panels c and d). This phenotype is similar to the phenotype of nuc2-663 cells grown at a restrictive temperature (HIRANO et al. 1988 ). In 80 µM TPCK many of the nuc2-663 cells remained alive, by the criterion that they excluded phloxine for more than 24 hr. Elongated cells with multiple cytokinetic furrows and with abnormal DNA distribution were often seen (Figure 2B, panel f), suggesting that this concentration of TPCK does not inhibit cell growth and elongation but inhibits some cell division process(es) in nuc2-663. Elongated and multiseptated cells were not found in wild-type cultures, even at concentrations of TPCK that were lethal for these cells.

A genetic screen for S. pombe mitotic mutants that are sensitive to TPCK:
To identify TPCK-sensitive mitotic mutants that were also temperature sensitive, we plated EMS-mutagenized cells at 25° and, after 3 days growth, tested them from TPCK sensitivity. The drug-sensitive isolates were then examined for an inability to grow at 36°. From strains that fulfilled these criteria, the mutants with mitotic abnormalities were identified by their phenotype at a restrictive temperature, for example, an abnormal distribution of DNA or the appearance of an aberrant septum. Out of 162,000 strains examined, only three met all our screening criteria and were therefore called tsm for TPCK-sensitive mitotic (Figure 3). Both the temperature and drug sensitivities of each mutant are recessive. They cosegregate meiotically, as revealed by tetrad and random spore analysis, and are therefore due to single gene mutations.

Linkage analysis suggested that the mutations in tsm2 and 3 mapped to the same locus (map distance 0.1 cM). This inference was confirmed by their failure to complement each other's temperature sensitivity. At 36° both mutant strains displayed the classical "cut" phenotype, like that seen in the TPCK-sensitive strain cut1-206 (UZAWA et al. 1990 ). In crosses between cut1-206 and tsm2-637, wild-type progeny were obtained only rarely (the recombination frequency was ~3.7 x 10^-5). Moreover, neither tsm2-637 nor tsm3-338 could complement the temperature sensitivity of cut1-206, and the temperature sensitivities of tsm2-637 and tsm3-338 were rescued by a plasmid-borne cut1⁺ gene. We concluded that tsm2 and tsm3 are alleles of the cut1⁺ gene.

Because nuc2-663 and at least three cut1^ts alleles result in the phenotype of sensitivity to the protease inhibitor TPCK, we surmised that these genes might act in the same pathway at anaphase onset. To test this hypothesis genetically, we made a heterozygous diploid by crossing cut1-tsm2 (strain 637BC1-d, Table 1) and nuc2-663 (strain McI#131, Table 1), as described in MATERIALS AND METHODS. Seventy-one azygotic asci were examined, 54 of which produced four spores that germinated on YES at 25° to yield TT-10, NPD-11, and PD-5. In all cases the cut1-tsm2 nuc2-663 double mutants were dead (the spores germinated, but after two to three divisions the cells elongated abnormally, branched, and died). The synthetic lethality of nuc2-663 and cut1-tsm2 suggests that Nuc2p and Cut1p may function in the same pathway at anaphase onset. This inference is strengthened by the finding that nuc2-663 cells could not be transformed with a multicopy plasmid carrying the cut1⁺ gene regulated by its endogenous promoter; the same plasmid transformed wt or tsm1-512 cells with high efficiency. Apparently, overexpression of cut1⁺ in the nuc2-663 background is toxic.

Tsm1-512 cells have defective microtubules, and at restrictive temperatures they fail in accurate chromosome segregation:
Strain tsm1-512 is the only representative of a second locus identified in our genetic screen. Genetic mapping of the tsm1 locus was performed by mitotic haploidization of the stable heterozygous tsm1-512/+ mat2-102/+ diploid, as described in ALFA et al. 1993 . Random spore analysis revealed linkage between tsm1-512 and two loci on Chromosome 2: ade1 (strain PN750, Table 1) and puc1 (strain FY89, Table 2).

Strains containing the tsm1-512 mutation failed to grow at 36° on YES medium but could grow, albeit slowly, on minimal medium with supplements (data not shown). This phenotype is also exhibited by two other mitotic mutants of S. pombe: cut2-364 and cut4-533 (YAMASHITA et al. 1996 ). Tsm alleles of cut1⁺ and cut1-206 can also be partially rescued by growth on minimal medium at 32° (data not shown); a similar phenotype has been described for temperature-sensitive alleles of several S. pombe genes important for G₂/M transition (HAGAN et al. 1988 ; FANTES 1989 ; MOLZ et al. 1989 ), although nuc2-663 does not grow on EMM at any restrictive temperature. It seems likely that the temperature sensitivity of these mitotic mutants is medium dependent not because the mutants are affected by some specific ingredient(s) in the culture medium, but because they fail to adjust to the changes in the timing of cell cycle progression that are caused by growth on different media.

To examine the morphology of tsm1-512 cells at 36°, we used the more stringent condition of growth: YES medium. An asynchronous culture of tms1-512 continued to increase in cell number for about 3 hr after the shift to 36° but then slowed down relative to wild-type controls (Figure 4A). After 4 hr, cell viability gradually decreased. Diverse abnormal phenotypes were seen, as is common for an asynchronous culture of a temperature-sensitive (ts) mutant at its restrictive temperature, where different cells have grown past their restriction points for different periods of time. To examine the phenotypes of these dying cells we fixed them, stained with DAPI and Calcofluor, and determined the fractions of cells that displayed the most prominent categories of abnormal phenotypes (Figure 4B). Slightly elongated cells with condensed chromosomes (Figure 4D, panel b) peaked at about 4 hr, followed by a peak in cells that had a septum but in which nuclear division had been abnormal (Figure 4D, panels c–g).

The majority of cells in these cultures separated their nuclear DNA, but such separation was defective (Figure 4C, closed triangles). This category included cells that had apparently equal masses of separating chromatin, but whose division was abnormal with respect to timing (delayed relative to septation) and/or positioning (the nucleus was shifted toward one end of the cell) (Figure 4D, panel e). In many cells chromosome separation was clearly unequal (e.g., Figure 4D, panel f). Often the chromosomes remained unseparated, despite the presence of a septum (Figure 4C, open triangles; Figure 4D, panel d), a phenotype also seen in nuc2-663^ts. Infrequently, we found cells with a "cut" phenotype (Figure 4D, panel c) or cells with a single interphase nucleus in the presence of a septum (Figure 4D, panel g). At ~5 hr after the shift there was a sharp increase in abnormal, postmitotic cells (Figure 4B). These included cells undergoing massive septation (Figure 4D, panel h) and cells that completed cytokinesis following an abnormal nuclear division, for example, anucleate and binucleate cells (Figure 4D, panel i). Consistent with a mixture of terminal phenotypes, the DNA content of the cells in the population varied considerably, from very low values (presumably anucleate or aneuploid cells) to very high (presumably binucleated cells undergoing a new round of DNA synthesis) (Figure 5).

View larger version (23K): In this window In a new window Download PPT slide   Figure 5. DNA content of tsm1-512 cells grown in YES and shifted to 36° at 0 hr. Cells were fixed in ethanol at indicated times and processed for analysis by flow cytometry. 5000 cells were counted for each data set. The "control" was a starved wild-type diploid culture, which shows 2n and 4n peaks of DNA.

Further insight into the cytological defects of tsm1-512 cells was obtained by examining their interphase and mitotic microtubules. Wild-type interphase S. pombe cells contained numerous cytoplasmic microtubules, most of which run between the cell tips, roughly parallel to the cell's axis (Figure 6, panel a). In mutant cells grown at 25°, however, there were fewer microtubules than normal; most of these were short, and they tended to localize near the middle of the cell (Figure 6, panel b). Upon shift to restrictive temperature, the fraction of tsm1-512 cells that stained with antitubulin lowered progressively with time (Figure 7). Moreover, the interphase cells that did stain contained only one to two cytoplasmic microtubules, and no cells were found with a wild-type microtubule network (Figure 6, panel c). As cells accumulated in mitosis, their microtubules largely disappeared; ~83% of cells with condensed chromosomes (n = 840) had no microtubule staining, and ~13% contained one or two tubulin-positive dots (Figure 6, panels d and e). In anaphase cells (n = 192) short or long spindles were seen in ~64% of cells (Figure 6, panels f and g). All spindles seen looked essentially normal, and there were no broken, V-, or X-shaped spindles. Consistent with the results described in previous paragraphs, however, two major defects were observed. First, almost all elongating spindles were in cells with well-developed septa (Figure 6, panels g and h), implying a significant delay in nuclear division relative to septation. Second, the separating masses of DNA, which were always located at the ends of the spindles, were often unequal, indicating a failure in proper chromosome segregation (Figure 6, panels f and g). Finally, almost none of the microtubules that are characteristically associated with the forming septum were seen, although some cells contained what appeared to be the remnant of a spindle or a post-anaphase array (Figure 6, panel i).

View larger version (124K): In this window In a new window Download PPT slide   Figure 6. Morphology of cytoplasmic and spindle microtubules in wild-type and tsm1-512 mutant cells grown in YES. Cells were fixed and processed for immunostaining as described in MATERIALS AND METHODS. Panels a and a', wild-type control cells; all other images are of the tsm1-512 cells grown either at 25° (b and b') or at 36° for 3–5 hr (the remaining panels). Panels a–i show tubulin staining; panels a'–i' show DNA stained with DAPI; and panels g''–i'' are bright field images. To allow antibody penetration, the yeast cell wall was removed, which precludes staining with Calcofluor to visualize the septum. The bright field images are included to allow visualization of well-developed septa (indicated with black arrowheads) in cells that contain a spindle. White arrowhead on panel h' points to one of the daughter nuclei caught in the developing septum. Bar in panel i', 5 µm.

View larger version (18K): In this window In a new window Download PPT slide   Figure 7. Frequency of cells with tubulin staining in wild-type (open bars) and tsm1-512 (closed bars) cultures as a function of time after a shift at 36°. Average results of three independent experiments are shown. Error bars represent standard errors of the mean. More than 300 cells were examined for each time point in each experiment. Cells were considered to show staining even if only a single dot of fluorescence was observed.

The finding that tsm1-512 cells have abnormal microtubules prompted us to examine the possibility of genetic interactions between this mutant and the cold-sensitive alleles of - and ß-tubulin, nda2-52 and nda3-311 (TODA et al. 1983 ; UMESONO et al. 1983 ). In a study of 93 tetrads dissected from a cross between tsm1-512 and nda3-311 and 115 tetrads from a cross between tsm1-512 and nda2-52, no double mutant colonies were found to be viable on YES at 25–29°, even though these conditions permit growth of strains carrying each mutation independently. The double mutant spores could germinate, but they failed to divide even once. This synthetic lethality between tsm1-512 and tubulin mutants is consistent with a notion that the tsm1-512 mutation compromises the integrity of microtubules. The mutation does not, however, lead to an increased sensitivity to the microtubule-depolymerizing drug thiabendazole (TBZ), because growth of tms1-512 at 25° on TBZ-containing medium was indistinguishable from that of wt A.

The analysis of asynchronous cultures demonstrated that tsm1-512 failed to divide properly at the restrictive temperature. It was, however, puzzling that the abnormal phenotypes accumulated only after 3 hr (one division of a wild-type cell takes ~2.3 hr under these conditions) and that viability decreased so gradually (Figure 4). We also wanted to learn whether the uncoordinated division of tsm1-512 resulted from a delay in nuclear division relative to septation or from a premature onset of cytokinesis. To address these questions we analyzed the phenotypes of tsm1-512 cells synchronized prior to a shift to restrictive temperature.

Wild-type and mutant cells in early G₂ phase were obtained as described in MATERIALS AND METHODS, resuspended in YES medium, and incubated at 36°. These cultures were fixed at the times indicated and stained with DAPI and Calcofluor. Three categories of cells were scored: those with condensed but unseparated chromosomes, those in anaphase, and those that stained with Calcofluor (examples of such phenotypes can be seen in Figure 4D, panel a). Frequency of the latter category measured the cell plate index (the fraction of cells with forming septa), except for the two last time points in tsm1-512 cultures, where it also included overseptated cells. Judged by these three criteria, tsm1-512 cells appear to undergo normal division for one cell cycle after a shift to a restrictive temperature in early G₂ (Figure 8). Quantitative analysis of the DNA content revealed a normal timing during the first round of DNA synthesis in this culture (data not shown). The timing of the first division and the entry into a second division were also almost indistinguishable in wild-type and mutant cultures. The second division of tsm1-512 cells, however, was marked by a pronounced increase in the frequency of cells with condensed chromosomes. About 20 min later, the cells septated, although almost no anaphase cells were found. This suggests that septation in mutant cells does not occur prematurely, nor is it significantly delayed. Chromosome condensation also begins at a normal time but persists while chromosome segregation is delayed. This combination of normal cytokinesis with delayed and abnormal karyokinesis leads to an increase in the number of abnormal and dying cells.

View larger version (25K): In this window In a new window Download PPT slide   Figure 8. Synchronized cultures of wild-type (wt A, top) and tsm1-512 (McI#190, bottom) cells growing in YES at 36°. At the times indicated, cells were fixed in ethanol, rehydrated, and stained with DAPI and Calclfluor. At least 300 cells were examined for each time point. Graphs represent results of a single experiment; however, similar results were obtained in two additional experiments. , cells with condensed, unseparated chromosomes; , anaphase cells; , cells with Calcofluor staining (cell plate index); , cells undergoing cytokinesis following abnormal nuclear division, e.g., Figure 4D, panel i.

To learn the frequency of different abnormal phenotypes, we scored each kind as a percent of the cells with Calcofluor staining. In the first wave of divisions, >85% of tsm1-512 cells that had Calcofluor staining looked normal, a value comparable to the >90% of cells that appear normal in a wild-type culture under identical conditions. In the second division, however, there was a large increase in the abnormal phenotypes described for asynchronous cultures (Figure 4). Abnormal separation of chromatin masses predominated (Figure 4D, panels e and f): 5 hr after the temperature shift >59% of cells with Calcofluor staining had abnormally segregated their chromosomes, whereas 18% still had condensed chromosomes (Figure 4D, panel d). The fraction of septated cells with a "cut" phenotype or a single interphase nucleus was <8% at all times examined.

These data suggest that after a shift to 36° during early G₂, most of the tsm1-512 cells can divide normally once, but their second division is abnormal, leading to cell death. This finding accounts for the slow changes in morphology and viability observed when an asynchronous culture of the tsm1-512 cells is shifted to restrictive temperature.

Genetic interactions between tsm1-512 and components of the anaphase proteolytic pathway suggest a functional involvement of the Tsm1 protein in this pathway:
To examine the possibility that tsm1⁺ is involved in the anaphase proteolytic pathway, we analyzed the progeny of genetic crosses between tsm1-512 and temperature-sensitive alleles of the genes that are known to act in this pathway. We found no synthetic lethal interactions between tsm1-512 and any of the following temperature-sensitive mutations: cdc2-33, cdc13-117 (HAGAN et al. 1988 ), cut1-tsm2, cut2-364, and cut9-665 (SAMEJIMA and YANAGIDA 1994 ). Each double mutant grows well at 25°, even though some of them show a reduction in their minimal restrictive temperature. On the other hand, double mutant spores of tsm1-512 and either cut4-533 or each of the proteasome mutants mts2 (GORDON et al. 1993 ) and mts3 (GORDON et al. 1996 ; SEEGAR et al. 1996) germinated efficiently on YES at 25°, but only pinpoint-sized colonies would form. Two representative colonies of each of these double mutants were streaked at 25° on YES, and the plates were examined several days later. The growth of these strains was severely retarded, although colonies of various sizes arose during prolonged incubation, presumably due to the genetic instability of these double mutant strains (data not shown). Double mutants containing tsm1-512 and nuc2-663 were identified in tetrad progeny. Although all spores were able to germinate at 25°, double mutant cells were not viable, indicating a synthetic lethal interaction between tsm1-512 and nuc2-663. These genetic interactions are diagrammed in Figure 9; they suggest that the protein product of the tsm1 gene is involved in the proteolytic pathway that acts at anaphase onset.

View larger version (10K): In this window In a new window Download PPT slide   Figure 9. Schematic representation of genetic interactions between selected S. pombe genes. Synthetic lethal interactions have been reported for some pairs of temperature-sensitive alleles of these genes: cut1ts, cut2-364 (UZAWA et al. 1990 ) and cut9-665, nuc2-663 (SAMEJIMA and YANAGIDA 1994 ). Arrows indicate synthetic lethal interactions, and solid lines indicate almost-lethal interactions.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

TPCK sensitivity of a mitotic mutant can identify genes involved in the anaphase proteolytic pathway:
Because proteolytic degradation of certain proteins is required for proper exit from mitosis, we asked whether sensitivity to the protease inhibitor TPCK can be used to seek mutants defective in anaphase onset. By examining 38 different ts mitotic mutants of S. pombe for their sensitivity to TPCK, we have identified four sensitive strains; nuc2-663, cut1-206, cut6-81, and cdc15-136. Each of these genes has previously been cloned and characterized.

The protein products of nuc2⁺, cut1⁺, and cdc15⁺ are all likely to be involved in the proteolytic pathway that acts at anaphase onset. Cdc15p is a key element in mediating the cytoskeletal rearrangements required for cytokinesis (FANKHAUSER et al. 1995 ). It may be a substrate for APC-dependent degradation, because it contains a PEST sequence, which is present in some short-lived proteins that are degraded by the ubiquitin-dependent pathway (RECHSTEINER and ROGERS 1996 ). Nuc2p is a component of the APC that is required for ubiquitination of certain target proteins at anaphase (HIRANO et al. 1988 ; YAMADA et al. 1997 ). We considered the possibility that the TPCK sensitivity of nuc2-663 might arise from some factor other than its relation to a ubiquitination pathway at anaphase, but this does not appear to be the case. The temperature and TPCK sensitivities of this strain cosegregate, and both phenotypes are rescued by the same plasmid that carries the nuc2⁺ gene, suggesting that the two phenotypes are due to the same mutation. Moreover, the phenotypes under the two restrictive conditions are similar.

The exact function of the cut1⁺ gene product at anaphase is not known, but it may be indirectly related to the proteolytic pathway (HIRANO et al. 1986 ; UZAWA et al. 1990 ). The intracellular level of Cut1p does not change during the cell cycle, but it interacts genetically and physically with Cut2p, which is localized to the metaphase spindle and is degraded at anaphase via the ubiquitin-dependent pathway (FUNABIKI et al. 1996A , FUNABIKI et al. 1996B ). These authors have proposed that the essential function of the Cut1p-Cut2p complex is regulated by proteolysis of Cut2p. The genetic interactions we have described between alleles of nuc2⁺ and cut1⁺, plus the lethality of Cut1p overexpression in nuc2-663 cells, support the notion of intimate, albeit perhaps indirect, involvement of Cut1p in the anaphase proteolytic pathway.

The relationship between Cut6p and the pathway for anaphase initiation is at present obscure. Mutations in cut6⁺, which encodes acetyl CoA carboxylase, lead to apparently normal chromosome segregation and cytokinesis, but the daughter nuclei are of unequal size (SAITOH et al. 1996 ). It is unclear how impaired fatty acid metabolism produces this postmitotic defect, nor can one say whether it is related to anaphase-specific proteolysis.

Because three of the four TPCK-sensitive strains of S. pombe carry mutations in genes whose products are likely to be involved in the proteolytic events at anaphase, we concluded that this collateral phenotype could be useful for identifying genes important in anaphase onset. However, mutations in some other genes that are involved in anaphase proteolysis (e.g., cut9⁺, cut2⁺, mts2⁺, and mts3⁺) do not show increased sensitivity to TPCK, so the absence of TPCK sensitivity cannot be taken as an indication that a given gene is not connected to this pathway. There are at least two possible explanations for these observations. First, not every mutation that renders a protein temperature sensitive will also lead to TPCK sensitivity. Indeed, different degrees of TPCK sensitivity are seen for different alleles of the cut1⁺ gene, all of which are equally temperature sensitive (Figure 3). Second, the exact mechanism of TPCK sensitivity is not known, so its effects may arise from some activity other than the inhibition of proteases.

TPCK, and a related compound TLCK, are the amino acid derivatives of chloromethyl ketone (reviewed in POWERS 1977 ). They were originally designed as specific inhibitors of the serine proteases, chymotrypsin and trypsin, respectively, although inhibition of other serine proteases has been observed (SCHOELLMANN and SHAW 1963 ; SHAW et al. 1965 ; SHAW 1975 ). Irreversible inhibition takes place by covalent bond formation between the active-site histidine and a methylene group of the inhibitor (ONG et al. 1965 ; ANTONOV 1993 ). Chloromethyl ketones can also react with a thiol group, so they can inactivate some of the cysteine proteases as well. However, TPCK and TLCK are relatively ineffective in inhibition of the purified 20S and 26S multicatalytic proteases (HOUGH et al. 1987 ). Both of these inhibitors have been widely used to study the role of proteases in such diverse biological processes as fertilization, signal transduction, cell growth, protein synthesis, and virus maturation (e.g., O'NEILL 1974 ; PONG et al. 1975 ; ETLINGER and GOLDBERG 1977 ; RICHERT et al. 1979 ; HILLER et al. 1984 ). These experiments are difficult to interpret, however, because TPCK and TLCK can also modify nonproteolytic enzymes and other proteins (e.g., KUPFER et al. 1979 ; KINZEL and KONIG 1980 ; MCCRAY and WEIL 1982 ; HUBBARD et al. 1984 ; SOLOMON et al. 1985 ; CROMLISH and ROEDER 1989 ; MURAKAMI and PIOMELLI 1991 ). Based on the comparison of effective inhibitor concentrations used in these studies, it seems likely that proteolytic enzymes are the primary targets of TPCK in our work, where the inhibitor was used at 80–120 µM. Other effects, however, such as the inhibition of esterase(s) and/or protein synthesis, cannot be rigorously excluded.

Possible role for Tsm1p in mitosis:
Tsm1-512 is a novel locus identified in our genetic screen. The phenotype of the tsm1-512 mutation differs from that of other S. pombe mitotic mutants previously characterized (HIRANO et al. 1986 ; SAMEJIMA et al. 1993 ). For example, when tsm1-512 cells are incubated at restrictive temperature, they transiently arrest with condensed chromosomes but without metaphase spindles. As with most "cut" mutants of S. pombe, cytokinesis is not restrained and proceeds normally, leading to cell death. The defects in spindle formation of tsm1-512 are not permanent, however, because most of the cells eventually segregate their DNA: in synchronized cultures, ~60% of cells with septa showed two separated masses of DNA. Even though the microtubules in the anaphase spindles looked normal by tubulin immunofluorescence, the resulting chromosome distribution was often unequal, implying that spindle microtubules and/or kinetochores did not function properly. The observed changes in the microtubule morphology were time- and cell-cycle stage dependent (Figure 6 and Figure 7), suggesting that the immunostaining procedure reliably reproduces the in vivo changes. Moreover, the same conclusions were drawn when microtubules were visualized using a plasmid expressing an -tubulin gene fused with that encoding a green fluorescent protein, kindly provided by D.-Q. DING and Y. HIRAOKO (E. GRISHCHUK and J. R. MCINTOSH, unpublished results). We also found that the temperature sensitivity of tsm1-512 could be suppressed by overexpression of an S. pombe protein that is related to S. cerevisiae Pac2p (HOYT et al. 1997 ) and human Cofactor E (TIAN et al. 1996 ), a microtubule-associated protein involved in the folding of ß-tubulin (E. GRISHCHUK and J. R. MCINTOSH, unpublished results). These results, as well as the synthetic lethality observed between tsm1-512 and the alleles of - and ß-tubulin, nda2 and nda3, strongly suggest that Tsm1p is important for microtubule integrity.

We analyzed genetically the possibility that tsm1⁺, like nuc2⁺, is involved in the anaphase proteolytic pathway by creating double mutants between tsm1-512 and temperature-sensitive alleles of the genes that are known to act in this pathway. Strong synthetic interactions were found between tsm1-512 and mutations in two components of the APC—nuc2⁺ and cut4⁺. Mutants in genes encoding two subunits of the proteasome, mts2 and mts3, were almost lethal when combined with tsm1-512, whereas temperature-sensitive alleles of cut1⁺, cut2⁺, cdc2⁺, and cdc13⁺ did not show strong interactions. These results suggest that tsm1-512 is involved in some aspect of the anaphase proteolytic pathway(s) that is different from the cut2/cut1 and cdc13/cdc2 branches. Based on the results presented above, we think that tsm1⁺ may act in a pathway important for anaphase spindle function. The most straightforward model is that the Tsm1p, or an interacting protein, is a substrate for ubiquitin-dependent proteolysis at anaphase. A protein like this has been found in S. cerevisiae: Ase1p, a microtubule-associated protein, whose APC-mediated degradation is required for both the formation of a mitotic spindle and the completion of anaphase B (PELLMAN et al. 1995 ; JUANG et al. 1997 ).

Unlike nuc2-663 and cut1-206, tsm1-512 cells synchronized in early G₂ phase and shifted to the restrictive temperature undergo a normal first division. The execution point for the tsm1 function may therefore lie earlier in the cell cycle or even in the previous cell cycle. Mutation in the sad1⁺ gene, whose product localizes to the spindle pole body, leads to a failure in spindle formation in the second nuclear division after a comparable synchrony and temperature shift (HAGAN and YANAGIDA 1995 ). This may indicate that certain events required for normal mitotic division occur early in the S. pombe cell cycle. Finally, Tsm1p might act in the proteolytic pathway late in mitosis, but failure to execute its function might lead to an abnormal subsequent mitosis, because of some kind of poisoning effect. This has been observed for Ase1p, as inappropriate expression of stable Ase1p during G₁ interfered with normal spindle assembly (JUANG et al. 1997 ). Cloning of the tsm1⁺ gene should help to address these possibilities and elucidate the function of its product.

	FOOTNOTES

¹ Present address: Institute of Gene Biology, Moscow, Russia 117334.

	ACKNOWLEDGMENTS

We are grateful to Drs. R. BARTLETT, P. FANTES, S. FORSBURG, C. GORDON, B. GRALLERT, P. NURSE, S. SAZER, K. SONG, and M. YANAGIDA for providing strains and to Dr. A. M. CARR for providing an S. pombe genomic library. We thank Drs. M. WINEY, S. SAZER, S. FORSBURG, and F. LUCA for helpful discussions and reading of the manuscript and members of the McIntosh laboratory for providing reagents and experimental advice. This work was supported by GM-33787 from the National Institutes of Health and PRP-84 from the American Cancer Society to J.R.M.

Manuscript received October 6, 1997; Accepted for publication April 9, 1998.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ALFA, C., P. FANTES, J. HYAMS, M. MCLEOD and E. WARBRICK, 1993 Experiments with Fission Yeast. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

ANTONOV, V. K., 1993 Irreversible group-specific inhibitors, pp. 257–259 in Chemistry of Proteolysis. Springer-Verlag, Berlin.

BARBET, N., W. J. MURIEL, and A. M. CARR, 1992  Versatile shuttle vectors and genomic libraries for use with Schizosaccharomyces pombe.. Gene 114:59-66[Medline].

BROWN, K. D., R. M. COULSON, T. J. YEN, and D. W. CLEVELAND, 1994  Cyclin-like accumulation and loss of the putative kinetochore motor CENP-E results from coupling continuous synthesis with specific degradation at the end of mitosis. J. Cell Biol. 125:1303-1312[Abstract/Free Full Text].

COHEN-FIX, O., J. M. PETERS, M. W. KIRSCHNER, and D. KOSHLAND, 1996  Anaphase initiation in Saccharomyces cerevisiae is controlled by the APC-dependent degradation of the anaphase inhibitor Pds1p. Genes Dev. 10:3081-3093[Abstract/Free Full Text].

CROMLISH, J. A. and R. G. ROEDER, 1989  Human transcription factor IIIC (TFIIIC). J. Biol. Chem. 264:18100-18109[Abstract/Free Full Text].

DESHAIES, R. J., 1995  Make it or brake it: the role of ubiquitin-dependent proteolysis in cellular regulation. Trends Cell Biol. 5:428-434.

ELBLE, R., 1992  A simple and efficient procedure for transformation of yeasts. Biotechniques 13:18-20[Medline].

ETLINGER, J. D. and A. L. GOLDBERG, 1977  A soluble ATP-dependent proteolytic system responsible for the degradation of abnormal proteins in reticulocytes. Proc. Natl. Acad. Sci. USA 74:55-58.

FANKHAUSER, C., A. REYMOND, L. CERUTTI, S. UTZIG, and K. HOFMANN et al., 1995  The S. pombe cdc15 gene is a key element in the reorganization of F-actin at mitosis. Cell 82:435-444[Medline].

FANTES, P. A., 1989 Cell cycle controls, pp. 128–204 in Molecular Biology of the Fission Yeast, edited by A. NASIM, P. YOUNG and B. F. JOHNSON. Academic Press, Inc., New York.

FUNABIKI, H., H. YAMANO, K. KUMADA, K. NAGAO, and T. HUNT et al., 1996a  Cut2 proteolysis required for sister-chromatid separation in fission yeast. Nature 381:438-441[Medline].

FUNABIKI, H., K. KUMADA, and M. YANAGIDA, 1996b  Fission yeast Cut1 and Cut2 are essential for sister chromatid separation, concentrate along the metaphase spindle and form large complexes. EMBO J. 15:6617-6628[Medline].

GLOTZER, M., A. W. MURRAY, and M. W. KIRSCHNER, 1991  Cyclin is degraded by the ubiquitin pathway. Nature 349:132-138[Medline].

GORDON, C., G. MCGURK, P. DILLON, C. ROSEN, and N. D. HASTLE, 1993  Defective mitosis due to a mutation in the gene for a fission yeast 26S protease subunit. Nature 366:355-357[Medline].

GORDON, C., G. MCGURK, M. WALLACE, and N. D. HASTIE, 1996  A conditional lethal mutant in the fission yeast 26 S protease subunit mts3⁺ is defective in metaphase to anaphase transition. J. Biol. Chem. 271:5704-5711[Abstract/Free Full Text].

HAGAN, I. M. and J. S. HYAMS, 1988  The use of cell division cycle mutants to investigate the control of microtubule distribution in the fission yeast Schizosaccharomyces pombe.. J. Cell Sci. 89:343-357[Abstract/Free Full Text].

HAGAN, I. and M. YANAGIDA, 1995  The product of the spindle formation gene sad1⁺ associates with the fission yeast spindle pole body and is essential for viability. J. Cell Biol. 129:1033-1047[Abstract/Free Full Text].

HAGAN, I., J. HAYLES, and P. NURSE, 1988  Cloning and sequencing of the cyclin-related cdc13⁺ gene and a cytological study of its role in fission yeast mitosis. J. Cell Sci. 91:587-595[Abstract/Free Full Text].

HILLER, R., A. SCHAEFER, R. ZIBIRRE, H. R. KABACK, and G. KOCH, 1984  Factors influencing the accumulation of tetraphenylphosphonium cation in HeLa cells. Mol. Cell. Biol. 4:199-202[Abstract/Free Full Text].

HILT, W. and D. H. WOLF, 1996  Proteasomes: destruction as a program. Trends Biochem. Sci. 21:96-102[Medline].

HIRANO, T., S. FUNAHASHI, T. UEMURA, and M. YANAGIDA, 1986  Isolation and characterization of Schizosaccharomyces pombe cut mutants that block nuclear division but not cytokinesis. EMBO J. 5:2973-2979[Medline].

HIRANO, T., Y. HIRAOKA, and M. YANAGIDA, 1988  A temperature-sensitive mutation of the Schizosaccharomyces pombe gene nuc2⁺ that encodes a nuclear scaffold-like protein blocks spindle elongation in mitotic anaphase. J. Cell Biol. 106:1171-1183[Abstract/Free Full Text].

HOCHSTRASSER, M., 1995  Ubiquitin, proteasomes, and the regulation of intracellular protein degradation. Curr. Opin. Cell Biol. 7:215-223[Medline].

HOLLOWAY, S. L., M. GLOTZER, R. W. KING, and A. W. MURRAY, 1993  Anaphase is initiated by proteolysis rather than by the inactivation of maturation-promoting factor. Cell 73:1393-1402[Medline].

HOUGH, R., G. PRATT, and M. RECHSTEINER, 1987  Purification of two high molecular weight proteases from rabbit reticulocyte lysate. J. Biol. Chem. 262:8303-8313[Abstract/Free Full Text].

HOYT, M. A., J. P. MACKE, B. T. ROBERTS, and J. R. GEISER, 1997  Saccharomyces cerevisiae PAC2 functions with CIN1, 2 and 4 in a pathway leading to normal microtubule stability. Genetics 146:849-857[Abstract].

HUBBARD, J. R., A. J. BARRETT, and M. KALIMI, 1984  Tosyl-lysyl chloromethane alters glucocorticoid-receptor complex nuclear binding and physical properties. Endocrinology 115:65-72[Abstract].

INNIS, M. A., D. H. GELFAND, J. J. SNINSKY and T. J. WHITE, 1990 PCR Protocols: A Guide to Methods and Applications. Academic Press, Inc., New York.

IRNIGER, S., S. PIATTI, C. MICHAELIS, and K. NASMYTH, 1995  Genes involved in sister chromatid separation are needed for B-type cyclin proteolysis. Cell 81:269-277[Medline].

JUANG, Y. L., J. HUANG, J. M. PETERS, M. E. MCLAUGHLIN, and C. Y. TAI et al., 1997  APC-mediated proteolysis of Asel and the morphogenesis of the mitotic spindle. Science 275:1311-1314[Abstract/Free Full Text].

KING, R. W., J. PETERS, S. TUGENDREICH, M. ROLFE, and P. HIETER et al., 1995  A 20S complex containing CDC27 and CDC16 catalyzes the mitotic-specific conjugation of ubiquitin to cyclin B. Cell 81:279-288[Medline].

KING, R. W., R. J. DESHAIES, J. M. PETERS, and M. W. KIRSCHNER, 1996  How proteolysis drives the cell cycle. Science 274:1