requirement for the budding yeast polo kinase cdc5 in ... · cdc5-1 mutation (24), suggesting that...

EUKARYOTIC CELL, Mar. 2008, p. 444–453 Vol. 7, No. 31535-9778/08/$08.00�0 doi:10.1128/EC.00283-07Copyright © 2008, American Society for Microbiology. All Rights Reserved.

Requirement for the Budding Yeast Polo Kinase Cdc5 in ProperMicrotubule Growth and Dynamics�†

Chong J. Park,1‡ Jung-Eun Park,1‡ Tatiana S. Karpova,2 Nak-Kyun Soung,1 Li-Rong Yu,3Sukgil Song,1 Kyung H. Lee,1 Xue Xia,4 Eugene Kang,1 Ilknur Dabanoglu,1 Doo-Yi Oh,1

James Y. Zhang,1 Young Hwi Kang,1 Stephen Wincovitch,5 Tim C. Huffaker,4Timothy D. Veenstra,3 James G. McNally,2 and Kyung S. Lee1*

Laboratory of Metabolism,1 Laboratory of Receptor Biology and Gene Expression,2 and Laboratory of Experimental Carcinogenesis,5

Center for Cancer Research, National Cancer Institute, National Institutes of Health, Bethesda, Maryland 20892; Laboratory ofProteomics and Analytical Technologies, Advanced Technology Program, SAIC-Frederick, Inc., National Cancer Institute—Frederick,

Frederick, Maryland 217023; and Department of Molecular Biology and Genetics, Cornell University,Ithaca, New York 148534

Received 4 August 2007/Accepted 16 December 2007

In many organisms, polo kinases appear to play multiple roles during M-phase progression. To provide newinsights into the function of the budding yeast polo kinase Cdc5, we generated novel temperature-sensitive cdc5mutants by mutagenizing the C-terminal noncatalytic polo box domain, a region that is critical for propersubcellular localization. One of these mutants, cdc5-11, exhibited a temperature-sensitive growth defect with anabnormal spindle morphology. Strikingly, provision of a moderate level of benomyl, a microtubule-depolymer-izing drug, permitted cdc5-11 cells to grow significantly better than the isogenic CDC5 wild type in a FEAR (cdcFourteen Early Anaphase Release)-independent manner. In addition, cdc5-11 required MAD2 for both cellgrowth and the benomyl-remedial phenotype. These results suggest that cdc5-11 is defective in proper spindlefunction. Consistent with this view, cdc5-11 exhibited abnormal spindle morphology, shorter spindle length,and delayed microtubule regrowth at the nonpermissive temperature. Overexpression of CDC5 moderatelyrescued the spc98-2 growth defect. Interestingly, both Cdc28 and Cdc5 were required for the proper modifi-cation of the spindle pole body components Nud1, Slk19, and Stu2 in vivo. They also phosphorylated thesethree proteins in vitro. Taken together, these observations suggest that concerted action of Cdc28 and Cdc5 onNud1, Slk19, and Stu2 is important for proper spindle functions.

Found from budding yeast to mammalian cells, the polokinases are a conserved subfamily of Ser/Thr protein kinasesthat play pivotal roles during the cell cycle and proliferation (2,46). In addition to the N-terminal kinase domain, they arecharacterized by the presence of a highly conserved polo boxdomain (PBD) in the C-terminal noncatalytic region (16). Inmammalian cells, multiple Plks (Plk1 to -4) with distinct reg-ulation and functions appear to exist. However, the genomes ofDrosophila melanogaster, Schizosaccharomyces pombe, and Sac-charomyces cerevisiae each contain only one apparent Plk1homolog (Polo [44], Plo1 [30], and Cdc5 [20], respectively).Overexpression of the mammalian polo-like kinase Plk1 com-plements the defect associated with the temperature-sensitivecdc5-1 mutation (24), suggesting that the critical functions ofPlk1 and Cdc5 are fundamentally conserved.

It is widely appreciated that Plk1 and its homologs playmultiple roles during M-phase progression, including mitoticentry, metaphase/anaphase transition, and cytokinesis. Severalobservations suggest that polo kinases play critical roles in

regulating spindle functions. The initial findings showed thatmutations in the Drosophila polo result in defects in bipolarspindle formation (26, 44). Subsequent studies in fission yeastalso disclosed that loss of Plo1 function leads to a mitotic arrestas a result of a monopolar spindle (30). In vertebrates, micro-injection of anti-Plk1 antibody into cultured cells or anti-Plx1(the Xenopus Plk1 homolog) into Xenopus embryos leads to adefect in centrosome maturation and bipolar spindle forma-tion (23, 34). Thus, the roles of the polo kinases in regulatingmicrotubule function have been largely conserved throughoutevolution.

However, the molecular mechanism through which Plk1 andits functional homologs regulate the spindle function is stillelusive. It has been shown that addition of active recombinantDrosophila Polo rescues impaired microtubule nucleation ac-tivity of salt-stripped centrosomes in vitro (9), suggesting thatPolo contributes to microtubule nucleation and growththrough a yet-unidentified Polo substrate(s) at the centro-somes. In cultured mammalian cells, Plk1 phosphorylates anddisplaces a centrosomal protein, Nlp, and this event is thoughtto permit the establishment of a centrosomal scaffold impor-tant for microtubule nucleation (5). Plk1 also appears to reg-ulate microtubule dynamics by either positively or negativelyregulating various components associated with microtubules. Ithas been reported that Plk1 phosphorylates and diminishes themicrotubule-stabilizing activity of TCTP (48). On the otherhand, Xenopus Plx1 has been suggested to stabilize microtu-

* Corresponding author. Mailing address: Laboratory of Metabo-lism, Center for Cancer Research, National Cancer Institute, NIH,Bethesda, MD 20892. Phone: (301) 496-9635. Fax: (301) 496-8419.E-mail: [email protected].

‡ C.J.P. and J.-E.P. contributed equally to this work.† Supplemental material for this article may be found at http://ec

.asm.org/.� Published ahead of print on 4 January 2008.

444

on April 11, 2020 by guest

http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

bules by negatively regulating a microtubule-destabilizing pro-tein, Stathmin/Op18 (4). However, how Plk1 and its homologsregulate these seemingly dissimilar events and how theseevents are coordinated with other cell cycle processes have yetto be further investigated. The identification of additional Plk1substrates and revealing of previously uncharacterized path-ways are likely important to shed light on the mechanismunderlying Plk1-dependent spindle regulation.

A growing body of evidence suggests that the conservedC-terminal domain of polo kinases, termed the PBD, plays acrucial role in targeting the catalytic activities of these enzymesto specific subcellular locations, such as spindle poles, kineto-chores, and the midbody (25, 40). These findings suggest thatthe PBD is a multifunctional domain capable of interactingwith diverse cellular proteins at specific subcellular structures.To aid our understanding of the function of PBD and to fur-ther investigate the role of mammalian Plk1 in regulating spin-dle function, we employed a genetically amenable buddingyeast organism and studied the function of the Plk1 homologCdc5 by randomly mutagenizing its C-terminal PBD. Charac-terization of one of the obtained temperature-sensitive cdc5mutants, cdc5-11, revealed that Cdc5 is required for properspindle microtubule dynamics and growth. Intriguingly, Cdc5phosphorylated several spindle pole body (SPB) or spindle-associated proteins, such as Nud1, Slk19, and Stu2, in vitro,raising the possibility that Cdc5 regulates spindle functions bydirectly phosphorylating these proteins.

MATERIALS AND METHODS

Strains, growth conditions, and cell counts. The yeast strains used in this studyare shown in Table 1. Cells were cultured in YEP (1% yeast extract, 2% Bactopeptone) supplemented with 2% glucose. Synthetic minimal medium (37) sup-plemented with the appropriate nutrients was employed to select for plasmidmaintenance. Yeast transformation was carried out by the lithium acetatemethod (17). All the cells were counted after cell aggregates were separated bysonication with a Sonicator Model W-225R (Heat Systems-Ultrasonics, Inc.,Plainview, NY) at 40% duty with no. 4 output for 4 seconds.

Strain and plasmid construction. The cdc5-11 mutants (KLY2466 andKLY2962) were generated as described previously (31). Briefly, the temperature-sensitive cdc5-11 allele, which does not support cell viability at 37°C, was firstintegrated at the TRP1 locus of a W303-1A-derived cdc5� strain (KLY2372) thatwas kept viable by the presence of a URA3-based YCplac33-CDC5 plasmid. TheYCplac33-CDC5 plasmid was shuffled out by plating onto 5-fluoro-orotic acid.

To alleviate the mitotic exit defect of the cdc5 mutant, a dominant allele ofCDC14 (CDC14TAB6-1) (38) was integrated at the HIS3 locus. Strain KLY5426(cdc28-as1 cdc5� plus pGAL1-cdc5-1) has been described previously (1). Com-plete deletions of CDC5 (cdc5�::kanMX6) and MAD2 (mad2�::URA3) weregenerated by the one-step gene disruption method (27). To generate strainsexpressing NUD1-HA, SLK19-HA, or STU2-HA under the respective endoge-nous promoter control, the corresponding loci were C-terminally tagged with aPCR fragment containing three hemagglutinin (HA) epitopes (HA3), essentiallyby using the method described by Longtine et al. (27). Strain KLY3928, whichexpresses a heat-inducible degron mutant for Cdc5 (cdc5-dg), was generated bybackcrossing strain KLY1546 (W303-1A background) with strain YYW37 (a giftof S. Elledge, Harvard Medical School, Boston, MA) three times. In order toinhibit the cdc28-as1 activity, the cells were treated with 0.5 �M 1NM-PP1 (a giftof K. Shokat, University of California, San Francisco, CA) at least 20 min afterrelease from the �-factor block to allow passage through G1 phase.

To generate plasmid pKL3838 (Table 2), a MAD2 genomic clone (a gift of DanBurke, University of Virginia Medical Center, Charlottesville, VA) was digestedwith PvuII and BamHI and then inserted into the pRS313 vector digested withEcoRV and BamHI. Plasmid pCJ187 was generated by inserting a SacI-XhoIfragment containing the TUB4 open reading frame into the pRS426 vectordigested with the corresponding enzymes.

Flow cytometry analyses. To examine cell cycle progression, the strains werearrested with �-factor, released into fresh medium, and then harvested at theindicated time points. The cells were washed twice with H2O, fixed with 70%ethanol, and then treated with RNase A (1 mg/ml) in phosphate-buffered salinefor 30 min at 37°C. After the cells were disrupted by sonication for 1 min, theywere stained with propidium iodide (50 �g/ml) in phosphate-buffered saline.Flow cytometry analyses were performed with the Cellquest program (BectonDickinson, San Jose, CA).

� phosphatase treatment and immunoblotting analyses. To dephosphorylatecellular proteins, approximately 100 �g of total cellular lysates was treated with400 units of � phosphatase (New England Biolabs, Ipswich, MA) at room tem-

TABLE 1. Strains used in this study

Strain Genotype Source or reference

KLY1546a MATa his3-11,15 leu2-3,112 trp1-1 ura-3-1 Laboratory stockKLY2470 KLY1546 LEU2::TUB1-GFP cdc5�::KanMX6 TRP1::CDC5-HA3 31KLY2466 KLY1546 LEU2::TUB1-GFP cdc5�::KanMX6 TRP1::cdc5-11-HA3 This studyKLY4733 KLY2470 mad2�::URA3 This studyKLY4731 KLY2466 mad2�::URA3 This studyKLY2970 KLY2470 HIS3::CDC14TAB6-1 31KLY2962 KLY2466 HIS3::CDC14TAB6-1 This studyYYW37 MATa cdc5-dg::URA3 15KLY3928 KLY1546 cdc5-dg::URA3 This studyKLY4208 KLY3928 LEU2::TUB1-GFP This studyKLY4440 MATa ura3-52 lys2-801am ade2-101och trp1�63 his3�200 leu2�1 spc98-2 M. WineyKLY5426 MATa cdc28-as1 cdc5�::HphMX4 � YCplac33-GAL1-HA-EGFP-cdc5-1 1KLY5851 KLY5426 HphMX4�::KanMX6 NUD1-HA::HphMX4 This studyKLY5824 KLY5426 SLK19-HA::KanMX6 This studyKLY5839 KLY5426 STU2-HA::KanMX6 This study

a KLY1546 is in a W303-1A genetic background.

TABLE 2. Plasmids used in this study

Name Description Source or reference

pRS313 CEN HIS3 39pRS316 CEN URA3 39pRS426 2� URA3 39YEp351 2� LEU2 14YCplac111 2� LEU2 11pKL3838 pRS313; MAD2 This studypKL2690 pRS316; CDC5 32pCJ187 pRS426; TUB4 This studypHS89 pRS316; SPC98 43pKL1171 YEp351; CDC5 32pKL743 YCplac111; CDC5 32

VOL. 7, 2008 ROLE OF Cdc5 IN PROPER MICROTUBULE FUNCTION 445


http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

perature for 30 min. Immunoblotting analyses were carried out with eitheranti-HA antibody or anti-Cdc28 antibody as described previously (40). Proteinsthat interacted with the antibodies were detected by the enhanced-chemilumi-nescence Western detection system (Amersham, Arlington Heights, IL).

Immunofluorescence confocal microscopy. Indirect immunofluorescence wasperformed as described previously (25). Microtubules were stained usingYOL1/34 rat anti-tubulin antibody (Accurate Chemical and Scientific Corp.,New York) and goat anti-rat CY3 antibody (Jackson Immunoresearch Labora-tories, West Grove, PA). DNA was visualized with 4�,6-diamidino-2-phenylin-dole (DAPI). The stained cells were viewed under a Zeiss LSM 510 confocalmicroscope equipped with HeNe, argon visible-light, and argon UV lasers.

Microtubule regrowth assay and measurement of microtubule length. Toperform microtubule regrowth assays, spindles were first disrupted by incubatingthe cells with 15 �g/ml of nocodazole for 2 h. The cells were then released intofresh medium and fixed at the indicated time points. Confocal images wereacquired at room temperature using a Zeiss LSM 510 system mounted on a ZeissAxiovert 100 M microscope with an oil immersion Plan-Neofluar 100�/1.3 ob-jective lens. To measure the spindle length in S phase, cells were arrested withhydroxyurea for 2 h and then fixed for confocal microscopic analyses. The spindlelength (the length of the Tub1-green fluorescent protein [GFP] fluorescencesignal) was measured by collecting 50 z slices with an interval of 0.1 �m and atotal stack size of 4.90 �m. All confocal datasets had a frame size of 512 pixelsby 512 pixels and a scan zoom of 2 and were line averaged two times. Allz-stacked images were analyzed using Zeiss AIM software version 3.2 sp1 (CarlZeiss GmbH, Heidelberg, Germany). Three-dimensional measurements wereacquired by using the orthogonal tool in the Zeiss AIM software. Starting at thez slice that corresponded to the beginning point of the microtubule staining,the beginning point was marked with the crosshair tool as the zero coordinate.The microtubule was measured in three dimensions by marking the microtubulepoint on each z slice and recording the measurement (in �m) from one point tothe next throughout the entire z stack.

Time-lapse imaging. Cells were mounted on agarose pads as described previ-ously (47). Time-lapse imaging was carried out at 37°C on a wide-field micro-

scope imaging system, which consisted of an inverted Nikon TE300 microscopewith a 60� 1.4-numeric-aperture objective (Nikon), a Lambda 10-2 filterchanger, and an I-Pentamax camera (Princeton Instruments/Roper Scientific,Trenton, NJ). Images were acquired with a GFP filter set (Chroma TechnologyCorp., Rockingham, VT) with excitation light attenuated to 10% of transmissionwith a neutral-density filter. The image system was controlled by Metamorphsoftware (Molecular Devices, Downington, PA).

RESULTS

The cdc5-11 mutant exhibits a temperature-sensitive growthdefect with abnormal spindle morphology. Cdc5 plays criticalroles at multiple stages during M-phase progression. To gen-erate cdc5 mutants defective in distinct stages of M phase, wepreviously generated various temperature-sensitive cdc5 mu-tants by mutagenizing the C-terminal PBD of Cdc5 (31). Here,we describe one of these mutants, cdc5-11, which is largelydefective in both microtubule dynamics and regrowth (see be-low). Sequence analyses revealed that the cdc5-11 allele pos-sessed a single point mutation (W565R) within the PB1 motifof the PBD (Fig. 1A). The cdc5-11 mutant grew well at 23°Cbut exhibited a severe temperature-sensitive growth defect at37°C (Fig. 1B). Provision of a centromeric CDC5 plasmid fullycomplemented this defect (data not shown). Flow cytometryanalyses of the cells being released from the �-factor block at37°C revealed that cdc5-11 arrested at a point after DNAreplication (2N DNA content), whereas wild-type CDC5 cellswent through the cell cycle normally under the same conditions

FIG. 1. The cdc5-11 mutant exhibits a temperature-sensitive growth defect and arrests at a late stage of the cell cycle. (A) Sequence alignmentbetween wild-type CDC5 and the cdc5-11 mutant. cdc5-11 possesses the W565R mutation in the C-terminal PBD. Gray box, the kinase domainof Cdc5; PB1 and PB2, the PB1 and PB2 motifs of the PBD. (B) Strains KLY2470 (CDC5) and KLY2466 (cdc5-11) were grown on YEP-glucosefor 3 days at the indicated temperatures. (C) For flow cytometry analyses, strains KLY2470 (CDC5) and KLY2466 (cdc5-11) were culturedovernight, arrested in G1 by �-factor treatment at 23°C, washed, and transferred into YEP-glucose medium at 37°C. Samples were taken at theindicated time points for analyses. (D) Strains KLY2470 and KLY2466 were cultured overnight and then shifted to 37°C for 3.5 h. Total cellularproteins were prepared and then analyzed to determine the levels of Cdc5-HA3 or cdc5-11-HA3 expression (top) or the level of Cdc28 as aninternal loading control (bottom).

446 PARK ET AL. EUKARYOT. CELL


http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

(Fig. 1C). The steady-state level of the cdc5-11 protein wassimilar to that of the wild-type Cdc5 at 23°C but was signifi-cantly diminished, yet detectable, at 37°C (Fig. 1D).

We then closely examined the phenotype associated with thecdc5-11 mutation by shifting the cultures to 37°C for 3.5 h. Incontrast to wild-type CDC5, a large fraction of the cdc5-11mutant exhibited a large-budded morphology with divided nu-clei (data not shown), a phenotype similar to that of the pre-viously characterized cdc5-1 mutant (20). Unlike the cdc5-1mutant, however, the spindle morphology of the cdc5-11 mu-tant was aberrant in a large fraction of the population. Whencultured at the restrictive temperature for 3.5 h, approximately25% of the cdc5-11 mutants exhibited a spindle morphologythat was noticeably bent and often discontinuous (Fig. 2A).Under the same conditions, wild-type CDC5 cells did not ex-hibit any discernible spindle defect (Fig. 2A). These observa-

tions suggest that Cdc5 activity is required for proper spindlefunction and that an intact PBD is required for this event.

Benomyl suppresses the cdc5-11 growth defect. If the spindledefect in the cdc5-11 mutant were in part due to the lack ofmicrotubule dynamics, then this defect could be remedied bythe provision of benomyl, a microtubule-depolymerizing agent.To examine this possibility, we tested the viability of cdc5-11and its isogenic wild-type CDC5 by culturing the cells in thepresence of various concentrations of benomyl at 30°C, a semi-permissive temperature that does not seriously induce thecdc5-11 growth defect. Strikingly, provision of 15 �g/ml ofbenomyl caused the cdc5-11 mutant to outgrow the isogenicwild-type CDC5 cells (Fig. 2B). A similar but less dramaticeffect was also observed in the presence of either 7.5 �g/ml or30 �g/ml of benomyl (data not shown). Loss of a spindlecheckpoint component, MAD2, abolished the benomyl-depen-

FIG. 2. The cdc5-11 mutant exhibits a benomyl-remedial growth defect with aberrant spindle structures. (A) Strains KLY2470 (CDC5) andKLY2466 (cdc5-11) cultured at 37°C for 3.5 h were fixed and subjected to immunostaining with anti-tubulin antibody. Cells with aberrant spindlestructures were quantified (right). The error bars represent standard deviations. (B) Strains KLY2470 (CDC5), KLY4733 (CDC5 mad2�),KLY2466 (cdc5-11), and KLY4731 (cdc5-11 mad2�) were cultured overnight, serially diluted, spotted onto either YEP-glucose or YEP-glucosecontaining 15 �g/ml of benomyl, and then incubated at 30°C. (C) Strains KLY4731 (cdc5-11 mad2�) and KLY4733 (CDC5 mad2�) transformedwith either control vector or a centromeric MAD2 plasmid (pMAD2) were spotted on a minimal plate to select for pMAD2 and then incubated ateither 23°C or the semipermissive 34°C. Two independent transformants of KLY4731 were tested. (D) Strains KLY2470 (CDC5), KLY2466(cdc5-11), and KLY2962 (cdc5-11 CDC14TAB6-1) were streaked onto YEPD and incubated at the indicated temperatures. (E) Strains KLY2470,KLY2466, KLY2970 (CDC5 CDC14TAB6-1), and KLY2962 were cultured overnight and spotted on YEP-glucose or YEP-glucose plus 15 �g/ml ofbenomyl.



http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

dent suppression of cdc5-11 viability (Fig. 2B). As expected ifthe cdc5-11 mutant possesses an intrinsic spindle defect, pro-vision of centromeric MAD2 significantly enhanced the viabil-ity of the cdc5-11 mad2� mutant (Fig. 2C). Notably, loss of acomponent of the anaphase-promoting complex, CDC23,which induces a mitotic exit delay, failed to suppress thecdc5-11 growth defect (see Fig. S1 in the supplemental mate-rial). These observations suggest that the rescue of the cdc5-11defect by benomyl is not due to simple provision of a longertime in mitosis. Thus, we concluded that the cdc5-11 mutant isimpaired in proper microtubule dynamics and that provision ofbenomyl helps promote the microtubule function, and there-fore cell growth, in a manner that requires a functional spindlecheckpoint.

It has been shown that a signaling network known as theFEAR (cdc Fourteen Early Anaphase Release) network andits effector, Cdc14 phosphatase, play critical roles in spindlestability during anaphase (33, 42) and also in meiotic spindledisassembly (29). Since Cdc5 is also a component of the FEARnetwork (41), we tested whether the spindle defect in cdc5-11is FEAR independent by providing the CDC14TAB6-1 allele,which is constitutively liberated from the Net1 tether atthe nucleolus (38). Our results showed that provision ofCDC14TAB6-1 moderately alleviated but did not completelyremedy the cdc5-11 growth defect (Fig. 2D). Furthermore, thecdc5-11 CDC14TAB6-1 double mutant also grew significantlybetter than the corresponding CDC5 CDC14TAB6-1 cells in thepresence of 15 �g/ml of benomyl (Fig. 2E). Taken together,

these results suggest that cdc5-11 possesses a microtubuledefect(s) that is independent of the FEAR network.

cdc5-11 is delayed in spindle growth, with a shorter spindlelength. The benomyl-remedial phenotype associated with thecdc5-11 mutation suggests that cdc5-11 is defective in propermicrotubule function. Thus, we carried out time-lapse studiesto closely monitor the microtubule growth as cells proceededthrough the cell cycle. The results showed that wild-type CDC5rapidly elongated spindles approximately 40 min (Fig. 3B) af-ter release from the �-factor block (Fig. 3A and B). In contrast,cdc5-11 displayed significantly shorter spindle length fromearly in the cell cycle and grew at a significantly lower rate thanthe respective wild-type CDC5 (Fig. 3A and B). Direct mea-surement of the spindle length in hydroxyurea-arrested (S-phase) cells showed that the cdc5-11 mutant possessed spindleswith an average length of 0.94 �m, whereas wild-type CDC5displayed an average size of 1.14 �m (Fig. 3C). The heat-inducible degron mutant for Cdc5, cdc5-dg, also exhibited asignificantly shorter spindle length (an average length of 0.88�m) than the isogenic wild type (Fig. 3C). These observationssuggest that Cdc5 is required for proper spindle elongationfrom early in the cell cycle.

cdc5-11 is impaired in microtubule regrowth. Next, we ex-amined the capacity of cdc5-11 to regrow the microtubulesafter the spindles were depolymerized with nocodazole. Ascells were released into fresh medium, samples were fixed formicroscopic observation. Nine minutes after release from no-codazole, wild-type CDC5 regenerated the detectable size of

FIG. 3. cdc5-11 is severely retarded in spindle elongation. (A and B) Strains KLY2470 (CDC5) and KLY2466 (cdc5-11) were arrested in G1with �-factor at 23°C and then released into fresh medium at 37°C. Under the same conditions, both CDC5 (n � 10) and cdc5-11 (n � 12) cellswere monitored by time-lapse video microscopy as described in Materials and Methods. Representative cells for each strain are shown in panelA. The time is given in minutes after �-factor release (time � 0). Images acquired for wild-type CDC5 and the cdc5-11 mutant were then analyzedto determine the average spindle length at each time point (B). An arrow at the 40-min time point corresponds to a stage in which the wild-typeCDC5 cells possess medium-size buds with a moderately elongated intranuclear spindle. The error bars indicate standard deviations. (C) StrainsKLY2470, KLY2466, and KLY4208 (cdc5-dg) were arrested in S phase with hydroxyurea for 2 h, fixed, and then analyzed by confocal microscopyand Zeiss AIM software. The error bars indicate standard deviations.



http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

spindles in 40% of the population (Fig. 4A and B). Under thesame conditions, the cdc5-11 mutant regenerated discerniblespindles in only 12% of the population, while the cdc5-dgmutant produced spindles at a much lower rate (Fig. 4A andB). Among the cells with detectable spindles, measurement ofspindle length after nocodazole release revealed that bothcdc5-11 and cdc5-dg possessed significantly shorter spindlesthan the isogenic wild type. At 21 min after release, wild-typeCDC5 exhibited an average spindle length of 3.31 �m, whereas

cdc5-11 and cdc5-dg displayed mean spindle lengths of 2.21�m and 1.85 �m, respectively, at the same time point (Fig. 4C).These results suggest that Cdc5 is required for proper spindlenucleation or elongation, or both, and that cdc5-11 is defectivein these events.

To further investigate if Cdc5 plays an important role inmicrotubule function, we examined whether CDC5 geneticallyinteracts with components important for microtubule nucle-ation. Overexpression of CDC5 partially remedied the growth

FIG. 4. Cdc5 is required for proper spindle microtubule regrowth. (A to C) Strains KLY2470 (CDC5), KLY2466 (cdc5-11), and KLY4208(cdc5-dg) were cultured at 23°C overnight. The cultures were then shifted to 37°C and treated with 15 �g/ml of nocodazole for 2 h before theywere released into fresh medium. At the indicated time points, samples were harvested for confocal microscopy (A). Quantification of the cellswith tubulin-GFP signals (B) and measurement of spindle length (C) were carried out as described in Materials and Methods. The error barsindicate standard deviations. (D) Strain KLY4440 (spc98-2) was transformed with the indicated constructs. The resulting transformants werecultured overnight, serially diluted, spotted onto YEP-glucose, and then incubated at the indicated temperatures. (E) Strain KLY2466 (cdc5-11)was transformed with the indicated plasmids. The transformants were selected and used for spindle regrowth assays as in panel A. The error barsindicate standard deviations.



http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

defect associated with the spc98-2 mutation (Fig. 4D). Further-more, as expected if Tub4 promotes Cdc5 function in micro-tubule nucleation, overexpression of TUB4 significantly en-hanced the ability of the cdc5-11 mutant to regrow the spindlesafter release from nocodazole block (Fig. 4E; see Fig. S2 in thesupplemental material).

Both Cdc5 and Cdc28 activities are required for properphosphorylation of Nud1, Slk19, and Stu2. To provide newinsights into the mechanism through which Cdc5 contributes toproper spindle function, we then examined whether Cdc5 candirectly regulate any of the known components important forspindle function. To this end, cdc5� cells kept viable by theexpression of a weakly functional cdc5-1 allele under the con-trol of the GAL1 promoter were transformed with either con-trol vector or a centromeric plasmid expressing wild-typeCDC5 from its native promoter (pCDC5). Modification of var-ious SPB or spindle-associated proteins (Nud1, Slk19, Stu2,Ase1, Spc97, Spc98, and Tub4) was then examined in thepresence or absence of pCDC5. Among the components ex-amined, we observed that modification of Nud1, Slk19, andStu2 was greatly diminished upon depletion of Cdc5 activity(see Fig. S3 in the supplemental material). Similar results werealso observed when the cdc5-11 or cdc5-11 CDC14TAB6-1 mu-tant was shifted to the nonpermissive temperature (see Fig. S4in the supplemental material). These findings suggest thatCdc5 is required for proper modification of Nud1, Slk19, andStu2. It has been reported that the PBD of Plk1 binds to aphosphorylated epitope that is frequently primed by Cdk1 (10).In line with this view, modification of Nud1, Slk19, and Stu2proteins also required proper Cdc28 activity (see Fig. S5 in thesupplemental material).

We then examined whether Cdc28 and Cdc5 activitiescooperate to achieve maximal phosphorylation of Nud1, Slk19,and Stu2 in vivo. To this end, a cdc28-as1 cdc5� double mutantkept viable by the expression of the cdc5-1 allele under GAL1promoter control was transformed with either a control vectoror a centromeric pCDC5 to regulate Cdc5 activity. The kinaseactivity of cdc28-as1 was acutely inhibited by the cell-perme-able inhibitor 4-amino-1-tert-butyl-3-(1-napthylmethyl) pyra-zolo[3,4-d]pyrimidine) (1NM-PP1) (3). In the presence of bothcdc28-as1 and Cdc5 activities, Nud1, Slk19, and Stu2 wereefficiently modified to generate the slowly migrating forms ascells proceeded from G1 to M (Fig. 5A to C). Treatment oftotal cellular lysates with � phosphatase eliminated or greatlyreduced the slowly migrating forms (Fig. 5D), suggesting thatthese forms were generated through multiple phosphorylationevents. In the absence of Cdc5, cdc28-as1 significantly inducedthe phosphorylated forms of Nud1, Slk19, and Stu2 (Fig. 5A toC). Inhibition of cdc28-as1 greatly diminished the levels ofphosphorylation on these proteins (Fig. 5A to C; compare thedimethyl sulfoxide [DMSO]-treated samples with the 1NM-PP1-treated samples in the cells containing vector), suggestingthat Cdc28 phosphorylates Nud1, Slk19, and Stu2 in vivo. Sim-ilarly, the presence of Cdc5 appeared to significantly increasethe levels of phosphorylation on all of these proteins, whichwas further pronounced in the presence of cdc28-as1 activity(Fig. 5A to C). Taken together, these results suggest thatCdc28 and Cdc5 phosphorylate Nud1, Slk19, and Stu2, eithercooperatively or independently, to achieve the maximal level ofphosphorylation during mitosis.

DISCUSSION

cdc5-11 is defective in proper spindle function. A growingbody of evidence suggests that the function of polo kinase inbipolar spindle assembly is conserved throughout evolution.However, the mechanism through which Plk1 contributes tomicrotubule function has been elusive. More direct evidencefor the involvement of polo kinase in regulating microtubulefunction came from the work of de Carcer et al., which dem-onstrated that provision of active recombinant Drosophila Polorescues impaired microtubule nucleation activity of the salt-stripped centrosomes in vitro (9). Other studies showed thatPlk1 phosphorylates and decreases the microtubule-stabilizingactivity of TCTP, an event thought to promote microtubuledynamics during anaphase (48). The Xenopus polo kinase ho-molog Plx1 has also been shown to phosphorylate Stathmin/Op18 and to stabilize microtubules by negatively regulating themicrotubule-destabilizing activity of the latter (4). Althoughhow these seemingly disparate events are coordinated duringcell cycle progression is not clearly understood, these observa-tions suggest that Plk1 and its homologs in various organismsregulate the microtubule dynamics by either positively or neg-atively regulating various components associated with micro-tubules.

In an effort to better understand the role of polo kinase inmicrotubule function, we generated and characterized a bud-ding yeast polo kinase, cdc5-11 mutant. Our results showedthat the cdc5-11 mutant exhibited a temperature-sensitivegrowth defect, with a substantial fraction (25%) of the mu-tants displaying improper spindle structures. Several lines ofevidence suggest that the observed spindle defect is a conse-quence of the cdc5-11 mutation. First, provision of benomyl,which diminished the viability of the isogenic wild type, dra-matically enhanced cdc5-11 viability at a semipermissive tem-perature. Second, as expected if the intrinsic spindle defectexists in cdc5-11, loss of MAD2 caused significant deteriorationin the growth of cdc5-11, but not in that of the correspondingCDC5 wild type. Third, the cdc5-11 mutant appeared to bedefective in microtubule dynamics and exhibited significantlyslower microtubule nucleation and spindle growth than theisogenic wild type. Tests of other available cdc5 mutants re-vealed that, except for cdc5-3, which is defective in the Swe1regulatory pathway (31), many cdc5 mutants also possess var-ious degrees of the benomyl-remedial growth defect (see Fig.S6A in the supplemental material). Furthermore, a cdc5 mu-tant that displays tethered Cdc5 populations at the SPB andthe bud neck (cdc5� bearing pCDC5�C-CNM67 andpCDC5�C-CDC12) (32) exhibited benomyl-dependent growthenhancement (see Fig. S6B in the supplemental material),suggesting that normal localization of Cdc5 to the nucleus andspindles is important for proper microtubule function. Exam-ination of the temperature-sensitive cdc5-dg mutant exhibiteda much more drastic defect in spindle nucleation and growth,although how much of this defect is directly attributable to theother Cdc5 depletion-induced mitotic defects is difficult toassess.

Regulation of Nud1, Slk19, and Stu2 by Cdc28 and Cdc5.Although it is widely appreciated that polo kinase is requiredfor proper bipolar spindle assembly in various organisms, theunderlying mechanism through which polo kinase contributes



http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

to the spindle function is still largely elusive. In an effort toidentify potential Cdc5 targets that are important for Cdc5-mediated spindle regulation, we examined whether Cdc5 ac-tivity is required for proper modification of some of the pre-viously characterized components critical for spindle function.Among the components that we examined, we observed thatproper modification of Nud1, Slk19, and Stu2 requires Cdc5

function in vivo (see Fig. S3 and S4 in the supplemental ma-terial). Cdc5 also phosphorylated these proteins in vitro (seeFig. S7 in the supplemental material), raising the possibilitythat Cdc5 directly phosphorylates and regulates the proteins.

It has been shown that Nud1 localizes at the SPB and playsa critical role in coordinating cytoplasmic microtubule organi-zation with mitotic exit (13). Slk19 is a component of the

FIG. 5. Requirement for Cdc28 and Cdc5 for the modification of Nud1, Slk19, and Stu2. (A to C) Strains KLY5851 (NUD1-HA3) (A),KLY5824 (SLK19-HA3) (B), and KLY5839 (STU2-HA3) (C) were individually transformed with either pRS315 (control vector) or pKL743(YCplac111-CDC5). The resulting transformants were cultured in YEP-galactose overnight and arrested in G1 by �-factor treatment. To depletethe weakly functional cdc5-1 under the control of the GAL1 promoter, the cells were then released into YEP-glucose medium containingnocodazole (noc.). Either control DMSO or 0.5 �M of 1NM-PP1 was added to the cultures 20 min after G1 release. Total cellular proteins wereseparated by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis for immunoblotting analysis with anti-HA antibody. The arrowsindicate the Cdc28-dependent, but Cdc5-independent, slowly migrating form. (D) Strains KLY5851 (left), KLY5824 (middle), and KLY5839(right) bearing pCDC5 (pKL743) were cultured as for panels A to C and then harvested after being released into nocodazole-containing mediumfor 140 min. Total cellular lysates were prepared, either treated with �-phosphatase or left untreated, and then subjected to immunoblottinganalyses. Subsequently, the same membranes were stained with Coomassie brilliant blue (CBB) for loading controls.



http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

FEAR network (41, 49) and translocates from the kineto-chores to the spindle midzone during anaphase, an event thatis critical for the stability of the anaphase spindle (42). Stu2localizes to the spindles, SPBs, and cytoplasmic microtubules(21) and appears to be required for mitotic spindle elongationand microtubule dynamics by destabilizing the plus ends in vivo(36, 45). Thus, although further studies are required to betterunderstand the mechanism through which Cdc5 contributes tothe regulation of Nud1, Slk19, and Stu2, the distinct functionsof these proteins at multiple subcellular locations help explainthe complexity of the spindle defect associated with thecdc5-11 mutation.

A growing body of evidence suggests that Cdk1-dependentphosphorylation onto a protein functions as a docking site forthe subsequent PBD-dependent Plk1 function (28). Consistentwith this model, acute inhibition of the Cdc28 activity drasti-cally diminished the levels of Nud1, Slk19, and Stu2 modifica-tions (see Fig. S5 in the supplemental material) and appearedto lessen the Cdc5-dependent modification of these proteins invivo (Fig. 5). Cdc28 also phosphorylated Nud1, Slk19, and Stu2in vitro (see Fig. S7 in the supplemental material), suggestingthat Cdc28 directly regulates these proteins. It should benoted, however, that Cdc5 is a substrate of the anaphase-promoting complex–Cdh1 complex (7), which is negatively reg-ulated by Cdc28 (18). In agreement with this observation,inhibition of cdc28-as1 with 1NM-PP1 significantly downregu-lated the Cdc5 level (see Fig. S8 in the supplemental material),thus raising the possibility that the diminished levels of Nud1,Slk19, and Stu2 phosphorylation in the 1NM-PP1-treated cellswere in part due to the decreased amount of Cdc5. Neverthe-less, Cdc28 appeared to induce the phosphorylated forms ofNud1, Slk19, and Stu2 even in the absence of Cdc5 (Fig. 5A toC), suggesting a direct role of Cdc28 in regulating these pro-teins. In support of this view, both Slk19 and Stu2 possess apotential PBD-binding S-pT/pS-P motif (the S189 residue forSlk19 and the S603 residue for Stu2) and GST-fused PBD, butnot the corresponding PBD H538A K540M (PBD/AM) phos-pho-Ser/Thr pincer mutant (10), bound to phosphorylatedStu2 (see Fig. S9A in the supplemental material). (Our at-tempt to detect the interaction between the PBD and thephosphorylated form of Slk19 was hampered by the unstablenature of the latter protein.) Furthermore, Cdc5 efficientlyinduced Stu2 modification only in the presence of Cdc28 ac-tivity (see Fig. S9B in the supplemental material). Thecdc5(W517F V518A L530A) mutant, which exhibits a crippledPBD function (40), or the cdc5(N209A) mutant, which lacksthe kinase activity (8), failed to induce this modification (seeFig. S9C in the supplemental material). These observationssuggest that the Cdc28-dependent phosphorylation onto Stu2promotes the Stu2-Cdc5 interaction through the PBD and thatthis step is critical for subsequent Cdc5-dependent Stu2 phos-phorylation. However, it should be noted that vertebrateCdc25C, whose PBD docking site is normally generated byCdc2 prior to the G2/M transition (10), can be directly phos-phorylated and activated by the Xenopus polo-like kinase ho-molog Plx1 in vitro (22), suggesting the existence of phosphory-lation-independent interaction between the polo-like kinaseand its substrates. Taken in aggregate, our results suggest thatCdc28 and Cdc5 phosphorylate and regulate the functions ofNud1, Slk19, and Stu2 either cooperatively or independently.

Determination of the Cdc28- and Cdc5-dependent phosphory-lation sites on these proteins and further investigation of thesignificance of each phosphorylation event are likely critical tobetter understand the underlying mechanism through whichCdc28 and Cdc5 regulate the microtubule function.

Potential polo kinase substrates in vertebrates. Studies ofpotential polo kinase substrates in a genetically amenable bud-ding yeast organism may allow us to better understand themechanism through which vertebrate polo kinases regulatevarious spindle functions. In this regard, it is noteworthy thatNud1 exhibits a limited homology with human Centriolin,which is important for cytoplasmic microtubule organizationand the late cytokinetic events (12). This observation hints thatsome of the functions of Nud1 and Centriolin are conservedthroughout evolution. Slk19 is the budding yeast member ofthe human TACC (transforming acidic coiled-coil) family ofproteins, whose deregulation has been implicated in the devel-opment of certain types of human cancers (35). Like Slk19,TACC proteins participate in controlling mitotic spindle dy-namics. In addition, Stu2 belongs to the XMAP215/Dis1 MAPfamily (19), which regulates microtubule plus-end assembly,microtubule nucleation, and anchorage to the centrosomes.The human homolog of this family, TOGp, has also beenisolated and appears to stimulate bipolar spindle assembly (6),although how it is regulated is not known. Providing thatCdc28 and Cdc5 directly phosphorylate Nud1, Slk19, and Stu2,it will be interesting to further investigate whether Cdk1and polo kinase cooperatively regulate the functions of thesehigher-eukaryotic homologs in their respective organisms.

ACKNOWLEDGMENTS

We are grateful to Susan Garfield for technical support and criticalreading of the manuscript.

This work was supported in part by NCI intramural grants (K.S.L.and J.G.M.), an NCI fund under contract N01-CO-12400 (T.D.V.),and NIH grant R01 GM040479 (T.C.H.).

REFERENCES

1. Asano, S., J. E. Park, K. Sakchaisri, L. R. Yu, S. Song, P. Supavilai, T. D.Veenstra, and K. S. Lee. 2005. Concerted mechanism of Swe1/Wee1 regu-lation by multiple kinases in budding yeast. EMBO J. 24:2194–2204.

2. Barr, F. A., H. H. Sillje, and E. A. Nigg. 2004. Polo-like kinases and theorchestration of cell division. Nat. Rev. Mol. Cell Biol. 5:429–440.

3. Bishop, A. C., K. Shah, Y. Liu, L. Witucki, C. Kung, and K. M. Shokat. 1998.Design of allele-specific inhibitors to probe protein kinase signaling. Curr.Biol. 8:257–266.

4. Budde, P. P., A. Kumagai, W. G. Dunphy, and R. Heald. 2001. Regulation ofOp18 during spindle assembly in Xenopus egg extracts. J. Cell Biol. 153:149–158.

5. Casenghi, M., P. Meraldi, U. Weinhart, P. I. Duncan, R. Korner, and E. A.Nigg. 2003. Polo-like kinase 1 regulates Nlp, a centrosome protein involvedin microtubule nucleation. Dev. Cell 5:113–125.

6. Cassimeris, L., and J. Morabito. 2004. TOGp, the human homolog ofXMAP215/Dis1, is required for centrosome integrity, spindle pole organiza-tion, and bipolar spindle assembly. Mol. Biol. Cell 15:1580–1590.

7. Charles, J. F., S. L. Jaspersen, R. L. Tinker-Kulberg, L. Hwang, A. Szidon,and D. O. Morgan. 1998. The Polo-related kinase Cdc5 activates and isdestroyed by the mitotic cyclin destruction machinery in S. cerevisiae. Curr.Biol. 8:497–507.

8. Cheng, L., L. Hunke, and C. F. J. Hardy. 1998. Cell cycle regulation of theSaccharomyces cerevisiae polo-like kinase Cdc5p. Mol. Cell. Biol. 18:7360–7370.

9. de Carcer, G., M. do Carmo-Avides, M. J. Lallena, D. M. Glover, and C.Gonzalez. 2001. Requirement of Hsp90 for centrosomal function reflects itsregulation of polo kinase stability. EMBO J. 20:2878–2884.

10. Elia, A. E., P. Rellos, L. F. Haire, J. W. Chao, F. J. Ivins, K. Hoepker, D.Mohammad, L. C. Cantley, S. J. Smerdon, and M. B. Yaffe. 2003. Themolecular basis for phospho-dependent substrate targeting and regulation ofPlks by the polo-box domain. Cell 115:83–95.



http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

11. Gietz, R. D., and A. Sugnino. 1988. New yeast-Escherichia coli shuttle vectorsconstructed with in vitro mutagenized yeast genes lacking six-base pair re-striction sites. Gene 74:527–534.

12. Gromley, A., A. Jurczyk, J. Sillibourne, E. Halilovic, M. Mogensen, I. Groisman,M. Blomberg, and S. Doxsey. 2003. A novel human protein of the maternalcentriole is required for the final stages of cytokinesis and entry into S phase.J. Cell Biol. 161:535–545.

13. Gruneberg, U., K. Campbell, C. Simpson, J. Grindlay, and E. Schiebel. 2000.Nud1p links astral microtubule organization and the control of exit frommitosis. EMBO J. 19:6475–6488.

14. Hill, J. E., A. M. Myers, T. J. Koerner, and A. Tzagoloff. 1993. Yeast/E. colishuttle vectors with multiple unique restriction sites. Yeast 2:163–167.

15. Hu, F., Y. Wang, D. Liu, Y. Li, J. Qin, and S. J. Elledge. 2001. Regulation ofthe Bub2/Bfa1 GAP complex by Cdc5 and cell cycle checkpoints. Cell 107:655–665.

16. Hudson, J. W., A. Kozarova, P. Cheung, J. C. Macmillan, C. J. Swallow, J. C.Cross, and J. W. Dennis. 2001. Late mitotic failure in mice lacking Sak, apolo-like kinase. Curr. Biol. 11:441–446.

17. Ito, H., Y. Fukuda, K. Murata, and A. Kimura. 1983. Transformation ofintact yeast cells treated with alkali cations. J. Bacteriol. 153:163–168.

18. Jaspersen, S. L., J. F. Charles, and D. O. Morgan. 1999. Inhibitory phos-phorylation of the APC regulator Hct1 is controlled by the kinase Cdc28 andthe phosphatase Cdc14. Curr. Biol. 9:227–236.

19. Kinoshita, K., B. Habermann, and A. A. Hyman. 2002. XMAP215: a keycomponent of the dynamic microtubule cytoskeleton. Trends Cell Biol. 12:267–273.

20. Kitada, K., A. L. Johnson, L. H. Johnston, and A. Sugino. 1993. A multicopysuppressor gene of the Saccharomyces cerevisiae G1 cell cycle mutant genedbf4 encodes a protein kinase and is identified as CDC5. Mol. Cell. Biol.13:4445–4457.

21. Kosco, K. A., C. G. Pearson, P. S. Maddox, P. J. Wang, I. R. Adams, E. D.Salmon, K. Bloom, and T. C. Huffaker. 2001. Control of microtubule dy-namics by Stu2p is essential for spindle orientation and metaphase chromo-some alignment in yeast. Mol. Biol. Cell 12:2870–2880.

22. Kumagai, A., and W. G. Dunphy. 1996. Purification and molecular cloning ofPlx1, a Cdc25-regulatory kinase from Xenopus egg extracts. Science 273:1377–1380.

23. Lane, H. A., and E. A. Nigg. 1996. Antibody microinjection reveals anessential role for human polo-like kinase 1 (Plk1) in the functional matura-tion of mitotic centrosomes. J. Cell Biol. 135:1701–1713.

24. Lee, K. S., and R. L. Erikson. 1997. Plk is a functional homolog of Saccha-romyces cerevisiae Cdc5, and elevated Plk activity induces multiple septationstructures. Mol. Cell. Biol. 17:3408–3417.

25. Lee, K. S., T. Z. Grenfell, F. R. Yarm, and R. L. Erikson. 1998. Mutation ofthe polo-box disrupts localization and mitotic functions of the mammalianpolo kinase Plk. Proc. Natl. Acad. Sci. USA 95:9301–9306.

26. Llamazares, S., A. Moreira, A. Tavares, C. Girdham, B. A. Spruce, C.Gonzalez, R. E. Karess, D. M. Glover, and C. E. Sunkel. 1991. polo encodesa protein kinase homolog required for mitosis in Drosophila. Genes Dev.5:2153–2165.

27. Longtine, M. S., A. McKenzie, D. J. Demarini, N. G. Shah, A. Wach, A.Brachat, P. Philippsen, and J. R. Pringle. 1998. Additional modules forversatile and economical PCR-based gene deletion and modification in Sac-charomyces cerevisiae. Yeast 14:953–961.

28. Lowery, D. M., D. Lim, and M. B. Yaffe. 2005. Structure and function ofpolo-like kinases. Oncogene 24:248–259.

29. Marston, A. L., B. H. Lee, and A. Amon. 2003. The Cdc14 phosphatase andthe FEAR network control meiotic spindle disassembly and chromosomesegregation. Dev. Cell 4:711–726.

30. Ohkura, H., I. M. Hagan, and D. M. Glover. 1995. The conserved Schizo-saccharomyces pombe kinase plo1, required to form a bipolar spindle, theactin ring, and septum, can drive septum formation in G1 and G2 cells. GenesDev. 9:1059–1073.

31. Park, C. J., S. Song, P. R. Lee, W. Shou, R. J. Deshaies, and K. S. Lee. 2003.Loss of CDC5 function in Saccharomyces cerevisiae leads to defects in Swe1pregulation and Bfa1p/Bub2p-independent cytokinesis. Genetics 163:21–33.

32. Park, J.-E., C. J. Park, K. Sakchaisri, T. Karpova, S. Asano, J. McNally, Y.Sunwoo, S.-H. Leem, and K. S. Lee. 2004. Novel functional dissection of thelocalization-specific roles of budding yeast polo kinase Cdc5p. Mol. Cell.Biol. 24:9873–9886.

33. Pereira, G., and E. Schiebel. 2003. Separase regulates INCENP-Aurora Banaphase spindle function through Cdc14. Science 302:2120–2124.

34. Qian, Y. W., E. Erikson, C. Li, and J. L. Maller. 1998. Activated polo-likekinase Plx1 is required at multiple points during mitosis in Xenopus laevis.Mol. Cell. Biol. 18:4262–4271.

35. Raff, J. W. 2002. Centrosomes and cancer: lessons from a TACC. Trends CellBiol. 12:222–225.

36. Severin, F., B. Habermann, T. Huffaker, and T. Hyman. 2001. Stu2 promotesmitotic spindle elongation in anaphase. J. Cell Biol. 153:435–442.

37. Sherman, F., G. R. Fink, and J. B. Hicks. 1986. Laboratory course manualfor methods in yeast genetics. Cold Spring Harbor Laboratory, Cold SpringHarbor, NY.

38. Shou, W., J. H. Seol, A. Shevchenko, C. Baskerville, D. Moazed, Z. W. Chen,J. Jang, A. Shevchenko, H. Charbonneau, and R. J. Deshaies. 1999. Exitfrom mitosis is triggered by Tem1-dependent release of the protein phos-phatase Cdc14 from nucleolar RENT complex. Cell 97:233–244.

39. Sikorski, R. S., and P. Hieter. 1989. A system of shuttle vectors and yeasthost strains designed for efficient manipulation of DNA in Saccharomycescerevisiae. Genetics 122:19–27.

40. Song, S., T. Z. Grenfell, S. Garfield, R. L. Erikson, and K. S. Lee. 2000.Essential function of the polo box of Cdc5 in subcellular localization andinduction of cytokinetic structures. Mol. Cell. Biol. 20:286–298.

41. Stegmeier, F., R. Visintin, and A. Amon. 2002. Separase, polo kinase, thekinetochore protein Slk19, and Spo12 function in a network that controlsCdc14 localization during early anaphase. Cell 108:207–220.

42. Sullivan, M., C. Lehane, and F. Uhlmann. 2001. Orchestrating anaphase andmitotic exit: separase cleavage and localization of Slk19. Nat. Cell Biol.3:771–777.

43. Sundberg, H. A., and T. N. Davis. 1997. A mutational analysis identifies threefunctional regions of the spindle pole component Spc110p in Saccharomycescerevisiae. Mol. Biol. Cell 8:2575–2590.

44. Sunkel, C. L., and D. M. Glover. 1988. polo, a mitotic mutant of Drosophiladisplaying abnormal spindle poles. J. Cell Sci. 89:25–38.

45. van Breugel, M., D. Drechsel, and A. Hyman. 2003. Stu2p, the budding yeastmember of the conserved Dis1/XMAP215 family of microtubule-associatedproteins, is a plus end-binding microtubule destabilizer. J. Cell Biol. 161:359–369.

46. van de Weerdt, B. C., and R. H. Medema. 2006. Polo-like kinases: a team incontrol of the division. Cell Cycle 5:853–864.

47. Waddle, J. A., T. S. Karpova, R. H. Waterston, and J. A. Cooper. 1996.Movement of cortical actin patches in yeast. J. Cell Biol. 132:861–870.

48. Yarm, F. R. 2002. Plk phosphorylation regulates the microtubule-stabilizingprotein TCTP. Mol. Cell. Biol. 22:6209–6221.

49. Zeng, X., J. A. Kahana, P. A. Silver, M. K. Morphew, J. R. McIntosh, I. T.Fitch, J. Carbon, and W. S. Saunders. 1999. Slk19p is a centromere proteinthat functions to stabilize mitotic spindles. J. Cell Biol. 146:415–425.



http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

requirement for the budding yeast polo kinase cdc5 in ... · cdc5-1 mutation (24), suggesting that...

Documents