A Novel Role for the Mitotic Spindle during DNA Segregation in Yeast: Promoting 2{micro}m Plasmid-Cohesin Association

The 2µm circle plasmid in Saccharomyces cerevisiae isa model for a stable, high-copy-number, extrachromosomal "selfish"DNA element. By combining a partitioning system and an amplificationsystem, the plasmid ensures its stable propagation and copynumber maintenance, even though it does not provide any selectiveadvantage to its host. Recent evidence suggests that the partitioningsystem couples plasmid segregation to chromosome segregation.We now demonstrate an unexpected and unconventional role forthe mitotic spindle in the plasmid-partitioning pathway. Thespindle specifies the nuclear address of the 2µm circleand promotes recruitment of the cohesin complex to the plasmid-partitioninglocus STB. Only the nuclear microtubules, and not the cytoplasmicones, are required for loading cohesin at STB. In cells recoveringfrom nocodazole-induced spindle depolymerization and G₂/M arrest,cohesin-STB association can be established coincident with spindlerestoration. This postreplication recruitment of cohesin isnot functional in equipartitioning. However, normally acquiredcohesin can be inactivated after replication without causingplasmid missegregation. In the mtw1-1 mutant yeast strain, theplasmid cosegregates with the spindle and the spindle-associatedchromosomes; by contrast, a substantial number of the chromosomesare not associated with the spindle. These results are consistentwith a model in which the spindle promotes plasmid segregationin a chromosome-linked fashion.

INTRODUCTION

Top

Introduction

The 2µm plasmid is an extrachromosomal selfish DNA elementin the Saccharomyces cerevisiae nucleus that harbors a partitioningsystem and a copy number control system for its stable high-copy-numberpropagation (2, 11, 21). Despite a copy number of approximately60 per cell, the plasmid molecules exist as a close-knit clusterand segregate as one unit (18, 22). This effective reductionin copy number provides the rationale for the existence of anactive segregation mechanism. The partitioning system has abeguiling simplicity. It consists of just two plasmid-encodedproteins, Rep1p and Rep2p, and a cis-acting locus, STB, whichincludes roughly six copies of a tandem array of a 65-bp consensussequence. Yet the plasmid manifests a degree of stability comparableto that of the yeast chromosomes, the loss rate being as lowas 10^–4 to 10^–5 per generation.

The remarkable efficiency of the partitioning system appearsto stem from its ability to feed into the segregation mechanismestablished for the host chromosomes (14, 20, 22). The kineticsof chromosome segregation and 2µm circle segregation closelyparallel each other during the yeast cell cycle (22). In severalmutant yeast strains that missegregate chromosomes at the nonpermissivetemperature (ipl1-2 and ndc10-2, for example), the plasmid showsa strong tendency to missegregate in tandem with the bulk ofthe chromosomes (14). The yeast cohesin complex, critical forthe segregation of each pair of sister chromatids to oppositecell compartments, is recruited to the STB locus (8, 14) ina Rep1p- and Rep2p-dependent manner (14). Mutations in Rep1pthat interrupt its interaction with Rep2p or STB or both preventcohesin-STB association and lead to high plasmid loss rates(26). Similarly, altered chromatin structure at STB due to thersc2 ${Delta}$ mutation blocks Rep1p (but not Rep2p) binding to STB, disruptscohesin assembly at this locus, and causes plasmid missegregation(25, 26). The timing of cohesin association and the lifetimeof the cohesin-associated state during the cell cycle are virtuallythe same for the plasmid and the chromosomes (1, 14, 19). Whenthe dissociation of the cohesin complex during anaphase is blocked,the duplicated plasmid clusters fail to separate, as is thecase for sister chromatids (14). Taken together, these observationsare consistent with a general model in which cohesin associationserves to hold together duplicated plasmid clusters, and cohesindissociation triggers their movement to daughter cells.

In this report, we demonstrate a role for the nuclear microtubulesin 2µm plasmid segregation that is distinct from theirrole in chromosome segregation. Upon disassembly of the mitoticspindle by nocodazole treatment or by inactivation of nuclearmicrotubules by a conditional mutation in TUB2, cohesin assemblyat the STB locus is selectively abolished with no ill effecton cohesin acquisition by the chromosomes. The block in cohesin-STBassociation caused by spindle depolymerization is not due tocell cycle delay or arrest in G₂/M per se and is not dependenton the activation of the spindle checkpoint. Consistent withprevious observations, disruption of cohesin-STB association,caused in this case by the lack of spindle integrity, resultsin plasmid missegregation. These findings reveal a hitherto-unsuspectedaspect of the molecular selfishness of the 2µm plasmid:not only does it poach an essential molecular component of chromosomesegregation (namely, cohesin), but it carries out the act byenlisting the assistance of the principal infrastructure ofthe host mitotic apparatus. Evidence from a "spindle recovery"experiment and a programmed cohesin inactivation assay suggeststhat the timing of cohesin assembly at STB, likely concomitantwith DNA replication, is critical for equal plasmid segregation.When a subset of the chromosomes are detached from the spindleby the mtw1-1 mutation, the plasmid still reveals a strikingproclivity to localize to the same cell compartment as doesthe spindle, the bud being strongly preferred over the mother.Collectively, these observations reveal a hitherto-unsuspectedrole for the mitotic spindle in DNA segregation in yeast. Inconjunction with earlier findings, they also suggest that thespindle-supported plasmid segregation is coupled to chromosomesegregation.

MATERIALS AND METHODS

Top

Materials and Methods

Strains and plasmids. The yeast strains and plasmids employed in this study are listedin Table 1. The strains used for chromatin immunoprecipitation(ChIP) assays contained an integrated copy of the MCD1 genetagged with the hemagglutinin (HA) epitope (14). Their relevantgenetic configurations are described in the text and indicatedin the figures. The reporter plasmids containing Lac operatorarrays and bound by the green fluorescent protein-Lac repressoror yellow fluorescent protein (YFP)-Lac repressor were visualizedby fluorescence microscopy (14, 22). For some of the assays,the endogenous 2µm circles served as the reporter.

View this table:
[in this window]
[in a new window]
TABLE 1. Yeast strains and plasmids used in this study

Assigning plasmid locations with respect to the spindle pole or the nuclear periphery. The method for localizing plasmids within the nucleus was adaptedfrom a previously published study (7). Assays were carried outwith a yeast strain housing a reporter plasmid tagged by theLac operator-YFP-Lac repressor. The nuclear membrane was labeledusing a fusion between the cyan fluorescent protein (CFP) andthe nuclear pore protein Nic96p. To localize a reporter plasmidwith respect to the spindle pole, the latter was labeled witha fusion between CFP and the Spc29 protein. The plasmid residencezone was determined by optical sectioning of the yeast nucleusalong the vertical axis (Z-series sectioning). For each sample,21 consecutive sections of 0.25-µm thickness each wereexamined, spanning 5 µm of the total thickness. Thosesections simultaneously displaying fluorescence from the nuclearmembrane (or the spindle pole) and that from the reporter plasmidwere stacked, deconvolved, and projected as two-dimensionalimages. The plasmid signals were localized relative to the nuclearperiphery by the use of concentric circles that define fourzones of equal area. For zonal demarcations, the diameter ofeach nucleus was estimated as follows. The circumference ofthe nuclear membrane was traced along the midpoints betweenthe outer and inner edges of the cyan fluorescence from thetagged Nic96p. A set of farthest end-to-end distances on thismembrane contour were measured and averaged. The thickness ofthe membrane was exaggerated in the two-dimensional projectionsbecause multiple nuclear cross sections were layered over eachother. Similar protocols of Z-series sectioning and deconvolutionwere employed in the mapping of plasmids with respect to theneighboring spindle pole. For a group of plasmid foci withina nucleus, the distance of each focus from the proximal spindlepole or the nearest point on the membrane was measured, summed,and averaged. Metamorph software (Universal Imaging Corporation,Downingtown, PA) was used to obtain the spherical reconstructionof the nucleus and to shrink the sphere to one-fourth, one-half,and three-fourths its volume.

Determination of relative plasmid copy numbers. DNA samples were digested with HinDIII, fractionated by electrophoresisin agarose, and blotted on to nylon membranes. They were hybridizedto a radioactive probe prepared by randomly primed replicationof a DNA fragment (obtained by PCR) containing the 2µmcircle REP1 region and the yeast HIS3 marker. The ratio of theintensities of the REP1 and the HIS3 bands in the autoradiogramwas estimated using Bio-Rad Image Analysis software (MolecularImager).

Other protocols. The methodologies for chromosome spread preparation, ChIP assays,and fluorescence microscopy of reporter plasmids or cytoskeletalstructures have been described previously (14, 22).

RESULTS

Top

Results

Integrity of the partitioning system and the mitotic spindle specifies the plasmid address inside the nucleus. Previous results have shown that the Rep1 and Rep2 proteinswithin the yeast nucleus tend to localize more strongly at orclose to the spindle pole (22). Furthermore, depolymerizationof microtubules using nocodazole causes the Rep proteins tobecome rather dispersed in the nucleus. At the same time, thereis an increase in the width of the plasmid residence zone measuredby confocal Z-series sectioning, indicating a loss of compactnessof the plasmid clusters.

To precisely map reporter plasmids tagged with YFP fluorescencein the nucleus, we used a procedure adapted from Heun et al.(7). The nuclear envelope or the spindle pole, visualized byCFP fusions to Nic96p or Spc29p, respectively, provided thereference landmarks in this assay. Two-dimensional projectionsof deconvolved Z-series fluorescence images from 100 individualcells were used to allocate plasmids within the nucleus. Thehost strain was [cir⁺] and hence supplied an STB-containingreporter plasmid with the trans-acting components, Rep1p andRep2p, of the partitioning system. As shown in Fig. 1A, a plasmid(shown in green) harboring STB was present as a tight clusterin close proximity to the spindle pole (shown in red). Whenthe reporter plasmid lacked STB, the cluster was loosely organized,with a nearly twofold increase in the spacing between the spindlepole and plasmid foci. A similar increase was observed for theSTB reporter in a [cir⁰] host, presumably due to lack of theRep proteins (data not shown). Nocodazole treatment resultedin a roughly 2.5-fold increase in the distance between the spindlepole and the STB reporter, comparable to that caused by thelack of the Rep-STB system. By contrast, the effect of nocodazoleon the reporter lacking STB was modest. The results were essentiallyidentical for cells arrested in G₁ with ${alpha}$ -factor (shown here)and those in late G₂/M in which replicated plasmid clustershad segregated into opposite cell compartments (data not shown).

View larger version (29K):
[in this window]
[in a new window]
FIG. 1. Delocalization of the 2µm plasmid and blockage of cohesin recruitment to the partitioning locus by nocodazole. (A) Reporter plasmids containing LacO arrays were visualized by tagging them with YFP-Lac repressor (yellow fluorescence). The spindle pole body was tagged by using Spc29p fused to CFP (cyan fluorescence). Following Z-series fluorescence microscopy, the sections containing the plasmid and the spindle pole were superimposed and projected in two dimensions. The cyan and yellow were converted to red (spindle pole) and green (plasmid), respectively,using Adobe Photoshop software. One hundred cells were assayed for each sample. (B to D) Chromatin immunoprecipitations were carried out in [cir⁺] yeast strains, the wild type, and the indicated mutants, containing an HA-tagged version of the MCD1 gene at its native chromosomal locus. The antibodies used for immunoprecipitations are indicated above the respective panels. WCE, whole-cell extracts, employed as positive controls for the PCRs; Chr V, one of the arm cohesin binding sites on chromosome V identified by Tanaka et al. (19).

In a corollary assay, we mapped plasmid positions to four arbitrarilydefined zones of equal area comprising the nuclear cross section.According to this assignment, plasmid occupancy was confinedprimarily to zones 2 and 3, was independent of the partitioningsystem (presence or absence of STB on the plasmid), and wasnot affected by nocodazole treatment (see Fig. S1A in the supplementalmaterial). Similar nonrandom distributions of plasmids wereverified by reconstructing the nucleus in three dimensions anddividing it into four spherical shells of equal volume (seeFig. S1B in the supplemental material).

The above observations indicate that the global nuclear localizationof plasmids to defined nuclear zones, with the center of thenucleus as the point of reference, is not dependent on the integrityof the mitotic spindle or that of the Rep-STB partitioning system.However, the picture changes when the spindle pole is the pointof reference. Both the spindle and the partitioning system areessential for the tight clustering of plasmid foci, as wellas their specific address proximal to the spindle pole. Perhapsnocodazole breaks the microtubule tether that holds the 2µmplasmid at this locale.

Effects of nocodazole treatment on the association of the Mcd1 protein or the Rep1 and Rep2 proteins with the STB locus. The remarkably similar effects of nocodazole treatment or thelack of the partitioning system on the specific nuclear addressof the 2µm plasmid suggest a potential role for the mitoticspindle in plasmid segregation. Therefore, we tested whethernocodazole would interfere with plasmid-cohesin association,a key step in the partitioning process, as indicated by severallines of evidence (14, 26). The results are shown in Fig. 1B.

The [cir⁺] host strain, in which ChIP assays were performed,contained an HA-tagged version of the MCD1 gene in its nativechromosomal context. After 3 h of treatment with nocodazole,80% or more of the cells were arrested in G₂/M with 2C DNA content,as indicated by light microscopy and fluorescence-activatedcell sorter (FACS) analysis (data not shown). Association ofMcd1p (the reporter for cohesin) with the STB DNA could notbe detected in the drug treated cells; by contrast, its associationwith an arm binding site on chromosome V (19) was normal (Fig.1B, lane 2). Cohesin binding at centromeres was also unaffectedby nocodazole (data not shown). As expected from previous work(14), the chromosomal and STB loci in the untreated cells wereassociated with Mcd1p (or the cohesin complex, by inference)(Fig. 1B, lane 4).

We showed previously that the Rep1 and Rep2 proteins are essentialfor cohesin recruitment to STB (14). The observed nocodazoleeffect could result from either one or both of the Rep proteinsbeing dislodged from STB. However, ChIP assays using antibodiesto Rep1p or Rep2p showed that nocodazole did not affect Repprotein association with STB (Fig. 1B, lanes 7 and 8). Consistentwith prior results, the chromosomal cohesin binding site wasnot occupied by the Rep proteins (Fig. 1B, lanes 7 and 8).

Is the disruption of cohesin-STB association upon nocodazoletreatment caused by spindle depolymerization or rather by cellcycle delay or arrest in G₂/M? To resolve this issue, ChIP analysiswas repeated with the cdc20-1 yeast strain which, at the nonpermissivetemperature, arrests at the metaphase-anaphase transition priorto cohesin disassembly (10). In the absence of nocodazole treatment,Mcd1p was detected at STB at the permissive (26°C) and atthe nonpermissive (37°C) temperatures (Fig. 1C, lanes 2and 8). Nocodazole treatment eliminated this association atboth temperatures (Fig. 1C, lanes 5 and 11). In addition, ChIPresults from a [cir⁺] mad2 ${Delta}$ strain revealed that triggering ofthe spindle checkpoint is not a prerequisite for the blockadeof STB-cohesin association by nocodazole (Fig. 1D, compare lanes11 and 5).

The data from the nocodazole-ChIP assays suggest a role, eitherdirect or indirect, for microtubules in the association of thecohesin complex with the 2µm plasmid. Strictly, our resultscan also be accommodated by the direct action of nocodazoleon the cohesin-Rep-STB complex. However, the chances of a spindleindependent action of nocodazole have been ruled out by furtherexperiments (see below).

Cohesin recruitment to STB is dependent on nuclear but not cytoplasmic microtubules. To verify that the nocodazole effects observed in the previousset of experiments (Fig. 1B to D) are mediated through the depolymerizationof microtubules, the ChIP analysis was repeated using cold-sensitivetubulin mutants that affect cytoplasmic microtubules (tub2-104)or nuclear microtubules (tub2-402) or both (tub2-401) (9).

In the wild-type strain and the tub2-104 mutant strain (affectedonly in cytoplasmic microtubules), the association of Mcd1pwith STB was normal at 30°C and 11°C, the permissiveand nonpermissive temperatures, respectively, for the mutant(Fig. 2A, lanes 2 and 5 of rows 1 and 2). In the tub2-401 (affectedin nuclear and cytoplasmic microtubules) and tub2-402 (affectedonly in nuclear microtubules) strains, Mcd1p was present onSTB at 30°C but absent at 11°C (Fig. 2A, compare lanes2 to lanes 5 of rows 3 and 4). At the same time, occupancy ofthe chromosomal locus by Mcd1p was not sensitive to any of thethree mutations at either temperature (Fig. 2A, lanes 8 and11 of rows 2 to 4).

View larger version (35K):
[in this window]
[in a new window]
FIG. 2. Plasmid-cohesin association and plasmid segregation in tubulin mutants. (A) Chromatin immunoprecipitations were performed in [cir⁺] yeast strains containing an HA-tagged MCD1. The 30°C caption refers to log-phase cells grown at this temperature. The 11°C caption indicates cells first grown to mid-log phase at 30°C and then shifted to 11°C for 18 h. This treatment caused 90 to 95% of the mutant cells to be arrested in the large budded state. The antibodies used for immunoprecipitations as well as the experimental and control lanes are indicated as in Fig. 1. (B) The patterns of microtubules at 11°C in the wild-type strain and the tubulin mutants (roughly 100 cells per sample) were visualized with the aid of YFP-Tub2p, and pseudocolored in green. (C) Chromosome segregation and plasmid segregation were followed in large budded tub2-402 cells grown at 30°C or those arrested in the large budded state at 11°C. Plasmids were visualized by LacO-bound green fluorescent protein-Lac repressor and chromosomes by DAPI staining. (D) Nuclear elongation was monitored with Nic96p-CFP, and the cyan fluorescence was recolored in red. For the segregation data in panels C and D, the set of numbers in each column was the average of at least 100 cells.

We have verified that the mutant strains arrest at 11°Cas large budded cells with a 2C DNA content and exhibit theexpected spindle phenotypes (Fig. 2B) (9). Consistent with thelack of cytoplasmic microtubules that are required for nuclearelongation and spindle extension, nearly all of the tub2-104cells showed brightly stained short spindles (Fig. 2B, row 2).In roughly 70% of the tub2-402 cells, the tubulin fluorescencewas localized at the periphery of the 4',6'-diamidino-2-phenylindole(DAPI) staining region and towards the cytoplasm, with no indicationof nuclear microtubules (Fig. 2B, row 3). The remaining cellscontained only fluorescent dots, likely specifying spindle poles,and no spindle structure was discernible (data not shown). Inthe tub2-401 mutant, >90% of the cells showed little or nosign of organized fluorescence (Fig. 2B, row 4).

The analyses with the tubulin mutants demonstrate that the associationof cohesin with the STB locus requires the integrity of onlynuclear microtubules and is unaffected by the lack of cytoplasmicmicrotubules.

Patterns of 2µm plasmid segregation in the absence of nuclear microtubules. Our earlier work indicated that certain mutations, ipl1-2 andndc10-2, for example, that affect specific steps of the chromosomesegregation pathway also cause missegregation of the 2µmplasmid in a chromosome-coupled manner (14, 22). That is, theplasmid almost always stays in tandem with the bulk of the chromosomes.Since the plasmid and chromosomes respond differently to thelack of nuclear microtubules in cohesin loading, is it possibleto uncouple the two in their segregation patterns in the tub2-402strain at 11°C? Note, though, that chromosome missegregationis inevitable in the mutant at this temperature because of theabsence of a functional spindle.

Normal segregation of the plasmid and of chromosomes (roughequality between the two cell compartments in the distributionof plasmid-associated fluorescence, as well as DAPI staining)seen in nearly all of the cells at 30°C was virtually absentat 11°C (Fig. 2C, row 1). Instead, three classes of aberrantplasmid segregation (Fig. 2C, rows 2 to 4) were observed. Inthe majority of cases, the entire set of plasmid foci was coincidentwith the chromosome mass indicated by the DAPI stain (Fig. 2C,row 2) or clearly separated from it (Fig. 2C, row 3). In theremainder, plasmid foci were split into two unequal groups,at least one of which was localized outside the DAPI stainingarea (Fig. 2C, row 4). Caution is warranted, however, in interpretingthe plasmid missegregation patterns because of the nuclear phenotypesof the tub2-402 mutant at 11°C (Fig. 2D). As revealed byCFP-tagged Nic96p, the nucleus was contained entirely withinone cell compartment in approximately 60% of the cells (Fig.2D, row 2). Interestingly, the primary missegregation classcomprised approximately the same fraction of cells, with plasmidand chromosomes confined to the same cell compartment (62%)(Fig. 2C, row 2). Hence, this phenotype could be accounted forsimply by nonelongated nuclei at 11°C and is not particularlyinformative. Cells in which the whole set of plasmid foci haddistinctly separated from the chromosomes accounted for $~$ 24%of the population (Fig. 2C, row 3); those in which a subsetof plasmid foci had moved away from the chromosomes constituted $~$ 12% of the population (Fig. 2C, row 4). The sum of these twoclasses (36%) correlated well with the fraction of cells inwhich nuclei had elongated enough to straddle the two cell compartments(38%) (Fig. 2D, row 3).

Thus, a mutation that specifically disrupts nuclear spindleassembly and abolishes cohesin association with the plasmidbut not with chromosomes is able to uncouple, at least to alarge extent, plasmid from chromosomes in their nuclear distributionpatterns.

Relocalization of the 2µm plasmid in chromosome spreads and restoration of plasmid-cohesin association, following recovery from nocodazole treatment. The Rep1 and Rep2 proteins and the associated 2µm circlemolecules can be localized to a subregion of the DAPI stainingzone in preparations of yeast chromosome spreads (14). A reporterplasmid that lacks the STB sequence was not detected in thespreads; nor is an STB-containing plasmid if Rep1p or Rep2por both are missing in the host cell. Since nocodazole is knownto cause dispersal of the Rep proteins in the yeast nucleus,expand the plasmid residence zone, and affect the plasmid addresswith respect to the spindle pole (Fig. 1A) (22), the followingquestions become relevant. Does nocodazole affect the localizationof the plasmid in chromosome spreads? If so, can the localizationbe reestablished when cells are allowed to recover in drug-freemedium? Does the recovery lead to reassociation of cohesin withSTB? Finally, and most significantly, how does the plasmid segregatefollowing recovery? The answers are provided in the resultsshown in Fig. 3 to 5. In these assays, cells blocked in G₁ with ${alpha}$ -factor were released into nocodazole-containing medium withoutthe pheromone for 3 h to impose G₂/M arrest and subsequentlyallowed to recover in nocodazole-free medium.

View larger version (25K):
[in this window]
[in a new window]
FIG. 3. Plasmid localization in chromosome spreads, cohesin STB association, and spindle restoration during recovery of nocodazole-treated cells. Cells arrested in G₁ were released in the presence of nocodazole for 3 h before the drug was washed off, and recovery in drug-free medium was permitted. Time zero refers to the start of recovery. At each time point, 100 to 200 cells were scored, and the percentages with the indicated profiles are listed. (A) The chromosome spreads were probed for the presence of the reporter plasmid and Rep1p using antibodies to the plasmid-bound Lac repressor and to native Rep1 protein, respectively. They were visualized using fluorescein-conjugated (plasmid) or Texas red-conjugated (Rep1p) secondary antibodies. (B) ChIP assays were carried out with antibodies to HA-tagged Mcd1 protein being expressed by the host strain. (C) The spindle structure was visualized by tagging native Tub1p with YFP. NS, no spindle fluorescence detected; PS, point spindle (presumably spindle pole); SP, short spindle; ES, elongated spindle; PS*, point spindle in telophase.

View larger version (29K):
[in this window]
[in a new window]
FIG. 5. Plasmid and chromosome segregations, following recovery of cells subjected to nocodazole treatment. The experimental protocol for nocodazole treatment and recovery was the same as for the data shown in Fig. 3. (A) Chromosome and reporter plasmid distributions were followed as described for the data shown in Fig. 2C in normal large budded cells from a log-phase culture, in cells arrested in the large budded state with nocodazole, and in cells at 60 min of recovery that had not yet undergone cytokinesis. The data set for each sample was obtained from the analysis of approximately 150 to 250 cells. (B) DNA samples from G₁-arrested cells (lanes 1 and 2) and those treated with nocodazole for 3 h immediately following release (lanes 3 and 4) were subjected to Southern blot analysis as described in Materials and Methods. The left and right panels show shorter and longer exposures, respectively, of the same autoradiogram. For reliable quantitation, the band intensities for the plasmid (Pl) were obtained from the shorter exposure, and those for chromosome (Ch) from the longer exposure. Lanes 1 and 3 contained twice the DNA inputs in lanes 2 and 4, respectively. The radioactively labeled hybridization probe harbored the 2µm circle REP1 and the chromosomal HIS3 sequences.

In contrast to chromosome spreads prepared from G₁ cells (priorto nocodazole treatment), those from nocodazole-arrested cells(0 min) failed to display the Rep1 protein or the STB-containingreporter plasmid in roughly 80% of the cases (Fig. 3A, lefttwo panels). However, at 45 min into the recovery period, thespreads showed the presence of Rep1p as well as the plasmidin >85% of the cells (Fig. 3A, right). At this time point,nearly all of the cells contained elongated spindles (Fig. 3C).The pattern for Rep2p was the same as that for Rep1p (data notshown).

ChIP assays revealed that the Mcd1 protein, and hence cohesinby implication, was absent at STB after nocodazole treatment(0 min); it reappeared at STB at approximately 30 min from thestart of recovery, was most prominent at 45 min, and was nearlygone by 60 min (Fig. 3B, middle). By contrast, Mcd1p was presentat the chromosome V binding site from 0 to 45 min, before disappearingalmost completely at 60 min (Fig. 3B, bottom). The marked reductionin Mcd1p at STB and the chromosomal site over the 45- to 60-mininterval is consistent with its cleavage and consequent cohesindisassembly during anaphase. As shown in Fig. 3C, there wasa strong correlation between the kinetics of spindle restorationand the reestablishment of STB-cohesin association during recoveryfrom nocodazole treatment.

It should be pointed out that nocodazole treatment also resultedin the exclusion of the 2µm plasmid from chromosome spreadsin G₁-arrested cells (Fig. 4A). When the drug was washed offwithout relieving G₁ arrest, the plasmid regained its associationwith the spreads and its spindle pole proximal localization(Fig. 4B). Upon subsequent release of the cells from pheromonearrest, the 2µm plasmid showed a normal segregation patternin the ensuing cell cycle (Fig. 4C). Furthermore, during nocodazolerecovery of a strain harboring smc3-42(Ts), the plasmid clusterwas detected at 45 min in chromosome spreads from cells maintainedat the permissive temperature (26°C) or shifted to 37°Cat the start of recovery (Fig. 4D).

View larger version (24K):
[in this window]
[in a new window]
FIG. 4. Cohesin has no role in the association of the 2µm plasmid with the spindle pole and chromosome spreads. (A) Chromosome spreads prepared from ${alpha}$ -factor-arrested cells were probed for the 2µm plasmid with antibodies to the Lac repressor (see Fig. 3A). (B) Plasmid distances from the spindle pole body were estimated as described in the legend to Fig. 1A. (C) The segregation patterns of the reporter plasmid and chromosomes were scored as in Fig. 2C. (D) The protocols for G₁ arrest, nocodazole treatment, and subsequent cell recovery were the same as those for the data shown in Fig. 3A. For each of the assays represented in panels A to D, the results are displayed as the mean values obtained after scoring at least 100 cells.

The fact that plasmid localization in chromosome spreads andthat the appearance of cohesin at STB in the recovering cellsis coincident with or closely follows spindle reformation suggestspossible spatial and/or functional regulation of plasmid-cohesinassociation by the spindle. Furthermore, the cohesin complexitself is not required for the specific nuclear location ofthe plasmid or its association with chromosomes, as indicatedby the spreads.

Association of cohesin with STB following spindle reassembly is not functional in effecting equal segregation of the plasmid. If the chromosomal paradigm for the role of cohesin in segregationapplies to the 2µm plasmid, cohesin loaded on the plasmidin a replication-independent manner is likely to be nonfunctional.We tested whether cohesin acquired subsequent to replicationduring recovery from nocodazole can mediate plasmid partitioning(Fig. 5A). The plasmid segregation patterns observed in thisexperiment could be divided into four types: equal segregation(Fig. 5A, row 1); unequal segregation (row 2), and completemissegregation (rows 3 and 4). Row 4, though, comprises a specialclass in which the chromosomes (stained with DAPI) failed tosegregate (see below).

Equal segregation of the plasmid dropped from approximately95% in large budded cells of the control group (not treatedwith nocodazole) to only about 30% in cells that had recoveredfrom nocodazole but had not yet completed cell division (Fig.5A, row 1). Correspondingly, there was an increase in the classof cells missegregating the plasmid, with unequal segregation(approximately 45%) (Fig. 5A, row 2) much more prevalent thannonsegregation (approximately 9%) (Fig. 5A, row 3). In earlierexperiments, it was noted that plasmids lacking a functionalRep-STB system showed an equal segregation frequency between20 and 40% when cells were scored at 75 min after release fromG₁ arrest (22). The values are comparable to the 30% seen hereafter recovery from nocodazole, with the distinction that thepartitioning machinery (Rep1p-Rep2p-STB) was intact in the presentinstance (Fig. 5A, row 1). In other words, the absence of thespindle during the cell cycle window between G₁ and G₂/M orthe lack of the Rep-STB system had the same functional consequencewith respect to plasmid segregation. These results would beconsistent with the concerted or sequential action of the mitoticspindle and the Rep/STB system in facilitating one or more steps,including cohesin recruitment, of the plasmid segregation pathway.

In three of the four cell types displayed in Fig. 5A (rows 1to 3), chromosomes segregated normally after recovery from nocodazole,as deduced from the equivalence of DAPI staining in the twocell compartments. Since nocodazole has no effect on the replicationcoupled recruitment of cohesin at the chromosomal locales, pairedsister chromatids are expected to complete the subsequent stepsin partitioning, once the spindle has been put in place. Thefraction of cells in which both plasmid and chromosomes wererestricted to the same cell compartment even after transferto nocodazole-free medium (approximately 16%) (Fig. 5A, row4) likely indicates the preanaphase state of these cells ora failure on their part to resume the cell cycle.

There is no reason, a priori, to suspect that nocodazole hasany adverse effect on 2µm circle replication. However,there is no experimental evidence that rules out this possibility.Missegregation caused by nocodazole can indeed be explainedif the plasmid fails to replicate or grossly underreplicatesin the absence of the spindle. The relative copy numbers ofthe 2µm circle in cells arrested in G₁ (1C chromosomecontent) and those arrested in G₂/M with nocodazole (2C chromosomecontent) were therefore estimated by Southern hybridization(Fig. 5B). The intensity of the plasmid band was divided bythat of the corresponding chromosome band for each of the lanes1 to 4. The ratio of these values for lanes 3 and 1 was 0.93;that for lanes 2 and 4 was 1.32 (or close to 1). Thus, the molarratio of plasmid to chromosome remained unchanged between theG₁ and G₂/M cells. Since chromosome duplication proceeds normallyin nocodazole treated cells, so must plasmid replication.

In summary, despite the reemergence of the 2µm plasmidin chromosome spreads and the reassociation of cohesin withSTB in response to spindle restoration in cells rescued fromnocodazole arrest, the plasmid partitioning process is irreversiblydamaged for the remainder of the cell cycle.

Cohesin functions in 2µm plasmid and chromosome segregation during the same window of the cell cycle. Our previous work indicated that the timely cleavage of Mcd1pis as essential for the separation of the duplicated plasmidclusters as it is for the separation of sister chromatids (22).The present work suggests that the late acquisition of cohesinby the plasmid in a postreplication fashion, as in the nocodazolerecovery experiment, is ineffective in mediating equal partitioning(Fig. 3 and 4). In a separate experiment, it was noted thatinactivating cohesin prior to DNA replication with the helpof a temperature-sensitive (Ts) mutation in Smc1p leads to themissegregation of chromosomes and the plasmid (26). All of theabove observations are consistent with a critical timing inthe cell cycle when cohesin acquisition by the plasmid is functional.

We have now followed plasmid segregation in a yeast strain carryinga single Ts allele of the MCD1 gene by releasing G₁ cells from ${alpha}$ -factor arrest and inactivating cohesin at different times duringprogression of the cell cycle (Fig. 5A). The 0-min or 60-mintemperature-shifted samples revealed high levels of plasmidand chromosome missegregation (unequal numbers of plasmid focior unequal amounts of DAPI in the two cell compartments) (Fig.6A, bottom). The frequency of equal segregation improved markedlywhen the temperature shift was delayed until 120 min, tendingtowards that observed in the subpopulation of large budded cellsmaintained at 23°C (Fig. 6A, top). The FACS data showedthat chromosome doubling was completed in these cells at 120min. Thus, blocking cohesin recruitment to STB entirely (0-minshift to 37°C) or inactivating cohesin in the midst of theS phase (60-min shift to 37°C) led to plasmid (and chromosome)missegregation. By contrast, the temperature shift imposed atthe end of S phase (120 min) did not.

View larger version (32K):
[in this window]
[in a new window]
FIG. 6. Effect of programmed inactivation of cohesin on plasmid segregation. (A) G₁-arrested MCD1 (Ts) cells at 23°C were released into growth medium, and aliquots were shifted to 37°C immediately (0 min) and at the indicated times. The extent of DNA replication at each shift time was monitored by FACS analysis. Cells were scored for plasmid and chromosome segregation at 180 min. (B) The experimental scheme was similar to that outlined in panel A. Following release from G₁, portions of the smc3-42(Ts) MCD1-nc cells were shifted to the nonpermissive temperature. Cells shifted at 150 min showed twice the haploid DNA content. These cells and control cells maintained at 26°C were examined by microscopy at 210 min. In the data shown in panels A and B, for the cells maintained at the permissive temperature, plasmid segregation was analyzed only in the subpopulation of large budded cells. Each column in the segregation data sets was obtained by examining over 100 cells.

In the above experiment, there is a caveat that the Mcd1p subunitof cohesin could have potentially been cleaved at 120 min, sothat the temperature shift was superfluous for cohesin disassembly.A variation of this assay was performed on a yeast strain containinga Ts allele of SMC3 (smc3-42), the native MCD1 under its ownpromoter, plus the MCD1-nc allele (encoding noncleavable Mcd1p)under the inducible GAL promoter (Fig. 6B). The strain alsoharbored a fluorescence-tagged 2µm plasmid in one caseand a fluorescence-tagged chromosome in the other. The cohesincomplex assembled during growth in galactose, while being thermallyunstable, would be immune to anaphase cleavage of Mcd1p. G₁-arrestedcells at 26°C were induced with galactose for 60 min, releasedinto pheromone-free galactose medium, and subjected to temperatureshift as in the experiment shown in Fig. 6A. FACS analysis ofculture aliquots indicated that >90% of the cells completedS phase at 150 min. Cells shifted at this time point, and thecontrol cells maintained at 26°C were examined for plasmid-chromosomesegregation at 210 min.

The predominant fraction of cells maintained at 26°C revealedthe fluorescence tagged reporter chromosome as a single dot(paired sister chromatids; 87%) and the reporter plasmid asa single cluster (unseparated sister clusters, presumably; 75%)(Fig. 6B). The remaining cells contained two well-segregatedchromosomal dots (13%) or plasmid clusters (25%). This resultagrees with the previous observation that plasmids cannot bepartitioned evenly into daughter cells in the absence of Mcd1pcleavage (22). In cells shifted to 37°C, normal segregationincreased to 82% for the chromosome and 83% for the plasmid.Thus, once DNA replication has been completed (and presumablythe duplicated plasmid clusters have been paired by cohesin),thermal inactivation of cohesin can yield equal segregation.

The 2µm plasmid cosegregates with the spindle and/or spindle-associated chromosomes in the mtw1-1 mutant. The Mtw1 complex, a component of the kinetochore, is requiredto establish the biorientation of sister chromatids that ismonitored by the Ipl1/Aurora protein kinase (6, 17). In themtw1-1 mutant yeast strain, the action of the Ipl1 kinase causeskinetochores to dissociate from the spindle at the nonpermissivetemperature. Since the spindle checkpoint is activated in thismutant, cells are arrested in G₂/M, and inhibition of the anaphase-promotingcomplex causes sister chromatids to remain paired by cohesin.Despite checkpoint activation, the chromosomes do segregateinto two separate masses in a significant fraction of the arrestedcells (6, 17). Strikingly, the short spindle, typical of thevast majority of the mtw1-1 cells, shows a strong tendency topreferentially migrate into the bud (17). These characteristicsegregation phenotypes of the chromosomes and spindle in themtw1-1 mutant prompted us to ask how the 2µm plasmid wouldsegregate in this background. Would the plasmid follow the spindle-freechromosomes or the spindle or neither?

The chromosome segregation patterns in the mtw1-1 cell populationarrested at 37°C, as revealed by DAPI, are depicted in Fig.7A (classes I through V). Roughly half the population consistedof cells with two well-separated chromosome masses, as indicatedby DAPI staining (panel A, II to IV). The DAPI zones were roughlyequal in class II, but unequal in classes III and IV. Note,however, that the DAPI equivalence in class II does not indicatenormal chromosome segregation, since sister chromatids mostoften tend to cosegregate in the mtw1-1 mutant due to lack ofcohesin cleavage. In the other two classes, the chromosomesexisted as a single unresolved entity, either confined to onecell compartment (class I) or stretched along the bud neck (classV). In almost all the class I and class V cells, the chromosomes,the reporter plasmid, and the spindle were not separated fromeach other. Since we wished to know whether the 2µm plasmidpreferentially associated with the spindle or with the chromosomesin the mtw1-1 background, these cells were not included in theanalysis presented in Fig. 7B and C.

View larger version (40K):
[in this window]
[in a new window]
FIG. 7. Segregation patterns of chromosomes, plasmid, and spindle in the mtw1-1 mutant yeast strain at the nonpermissive temperature. (A) Chromosome distributions in the mutant at 37°C, followed by DAPI staining, can be divided into five classes (I to V). (B) Segregation profiles of two fluorescence-tagged reporter plasmids, one harboring STB (+STB) and the other lacking it (–STB), among the class II to IV cells (see panel A) are displayed. The green fluorescence from plasmids is pseudocolored in red. (C) The relative dispositions of the STB-containing plasmid and the spindle in this group of cells (panel A, classes II to IV), as revealed by immunofluorescence, are presented. Approximately 200 cells were assayed for each of the analyses shown in panels A to C.

The class II to IV cells could be divided into five subtypes(Fig. 7B, subtypes a through e) with respect to the segregationof the reporter plasmid they harbored: equal segregation (subtypea), complete missegregation (b and c), and unequal segregation(d and e). Equal segregation of the 2µm plasmid (harboringSTB), which is the norm in wild-type cells, occurred only inapproximately 6% of the mutant cells. Unequal segregation wasobserved in roughly 12% (subtypes d and e). In the largest fractionof cells (82%; subtypes b and c), the plasmid cluster was presentonly in one cell compartment. Of these, the bud residents (67%;subtype b) outnumbered the mother residents (17%; subtype c)nearly four to one. The strongly preferred localization of the2µm plasmid in the bud correlates well with a similarbias noted by Pinsky et al. (17) for the migration of the shortspindle in mtw1-1. By contrast to the 2µm plasmid, equal,unequal, and all-or-none segregation events for the ARS reporterplasmid (lacking STB) were comparable to those observed previouslywith wild-type yeast strains.

To verify the suspected cosegregation of the 2µm plasmidand the spindle in the mtw1-1 mutant, the two were simultaneouslylocalized by immunofluorescence (Fig. 7C): the plasmid usingantibodies to the LacO bound repressor and the spindle usingantibodies to tubulin (Tub1p). In cells containing two equallysegregated plasmid clusters (Fig. 7C-1), the spindle was almostalways present in both cell compartments. When there was a singleplasmid cluster confined to the bud compartment (C-2) or themother compartment (C-3), the spindle almost invariably colocalizedto the same compartment. In the case of unequal plasmid segregation(C-3 and C-4), the spindle most often stayed in tandem withthe larger of the two clusters.

As a control, we followed fluorescence-tagged chromosome IVin a similar assay (see Fig. S2 in the supplemental material).Unlike the 2µm plasmid, chromosome IV showed a bias, albeitsmall, in the opposite direction. It was more often locatedaway from the spindle ( $~$ 60%) than associated with it ( $~$ 40%) (seeFig. S2 in the supplemental material) (17). Unlike chromosomeIV, chromosomes III and VII showed a strong tendency to remainassociated with the short spindle in the mtw1-1 mutant (17).We also repeated this assay with the mtw1-1 mad2 ${Delta}$ double mutant,which is not blocked in metaphase and is competent for cohesincleavage (17). The spindle phenotype in the large budded cellsof the double mutant was the same as that in the metaphase-arrestedsingle mutant. In addition, the 2µm plasmid remained tightlyassociated with the short spindle.

The plasmid, chromosome, and spindle patterns in the mtw1-1mutant demonstrate that, during segregation, the 2µm plasmidmaintains its coupling with the spindle and/or the subset ofspindle-associated chromosomes.

Spindle-mediated localization of the 2µm plasmid in chromosome spreads and cohesin recruitment to STB are independent of chromosome spindle attachment. The behavior of the 2µm plasmid in the mtw1-1 mutant isconsistent with two models for plasmid segregation. In one,plasmid segregation is spindle dependent but chromosome independent.In the other, segregation is also dependent on a spindle-associatedchromosome. If the plasmid is tethered to a chromosome, it cangain access to the spindle indirectly via kinetochore attachmentto the spindle. This step may reinforce plasmid-chromosome associationthrough additional spindle-mediated interactions. Furthermore,it can set the stage for critical events in plasmid segregationfacilitated by the spindle, for example, cohesin recruitment.We therefore tested whether plasmid localization in chromosomespreads and cohesin association with STB is affected when chromosome-spindleassociation is blocked by the kinetochore mutation ndc10-2.

As shown in Fig. 8A, the plasmid cluster was detected in chromosomespreads at the permissive and nonpermissive temperatures inthe ndc10-2 mutant. It disappeared from spreads upon nocodazoletreatment and reappeared following nocodazole removal at bothtemperatures. Similarly, in ChIP analysis, cohesin associationwith STB was not affected by the mutation, and nocodazole blockedthis association at 26°C and 37°C (Fig. 8B). Furthermore,in a nocodazole recovery assay using a synchronized cell population(analogous to that shown in Fig. 3), the mutation did not alterthe temporal patterns of plasmid localization, spindle restorationand cohesin-STB association (data not shown).

View larger version (20K):
[in this window]
[in a new window]
FIG. 8. Plasmid localization in chromosome spreads and cohesin association with STB in the absence of chromosome spindle attachment. (A) In chromosome spreads prepared from the ndc10-2 mutant strain, the 2µm reporter plasmid (harboring LacO arrays) was probed using antibodies to LacI or to the Rep1 protein. (B) A derivative of the ndc10-2 strain expressing HA-tagged Mcd1p was used for the ChIP assays.(C) The results of this study are in agreement with a spindle-dependent and chromosome-associated mode of 2µm plasmid segregation. The results shown in panels A and B indicate that chromosome (kinetochore) attachment to the spindle does not affect stages I and II of the pathway, namely, plasmid clustering, localization, and cohesin recruitment to STB. The absolute requirement of a functional spindle during stage II, namely plasmid-cohesin association, is a surprising outcome from this study. As demonstrated by previous work, stage II is also dependent on the integrity of the Rep-STB system and the activity of the RSC2 chromosome-remodeling complex (8, 14, 22, 25, 26). Cohesin disassembly is a critical step for the final stage (stage III) for plasmid segregation (14; this study). Observations of 2µm plasmid behavior in mutant yeast strains that missegregate chromosomes provide cumulative circumstantial evidence that spindle-mediated plasmid segregation also requires plasmid-chromosome association (22; this study).

Thus, plasmid localization as well as cohesin recruitment tothe STB locus can occur normally in the presence of an intactspindle even when the chromosomes are blocked from spindle attachment.

DISCUSSION

Top

Discussion

The most significant and least-suspected outcome from the presentstudy concerns the role of the mitotic spindle in the segregationof the 2µm plasmid. The integrity of the spindle is essentialfor the specific nuclear localization of the plasmid, its associationwith chromosome spreads, and the recruitment of the cohesincomplex to its partitioning locus. Our current results, alongwith previous work (14, 26), suggest that the assembled cohesincomplex plays similar functional roles in the partitioning ofthe plasmid and the yeast chromosomes. Yet, the plasmid utilizesa completely different mechanism from that of the chromosomes,involving the plasmid partitioning system and the host mitoticspindle to acquire the cohesin complex.

The nuclear spindle adds a spatial and/or structural dimension to plasmid-cohesin association. The displacement of the plasmid cluster from its normal nuclearlocale or from chromsome spreads by nocodazole treatment, alongwith the accompanying loss of STB-cohesin association, suggestsa spatial-structural contribution of the spindle to plasmidpartitioning. At this time, the nature of the interaction betweenthe plasmid and spindle is not understood. The interaction maybe indirect, perhaps mediated through one or more spindle-associatedproteins. A relevant question is whether, in addition to itsrole in channeling cohesin to STB, the spindle may also mediatethe trafficking of plasmid clusters during segregation. In time-lapseassays, the 2µm plasmid and kinetochore proteins showalmost identical localization and dynamics throughout the cellcycle (S. Velmurugan, unpublished data). The plasmid appearsto be situated at the right place to potentially interact witha plus-end motor protein.

The timing of cohesin action during plasmid segregation. According to the prevalent models for cohesin assembly at chromosomalloci, an advancing replication fork pauses at a precohesin sitethat is already occupied by the Smc1 and Smc3 protein subunitsof the complex (3, 23). The recruitment of the Mcd1 proteinthen requires the exchange of the resident DNA polymerase ${alpha}$ withinthe replication complex for a novel polymerase ${sigma}$ (23). We donot know whether these steps hold true in the organization ofcohesin at STB during a normal cell cycle. However, consistentwith this mechanism, the Smc subunits of the complex are foundassociated with STB and not with other regions of the 2µmplasmid before the initiation of replication (26).

Earlier results (14, 26) and the present data demonstrate thatlack of functional cohesin prior to or during DNA replication,as well as postreplication loading of cohesin on STB, leadsto plasmid missegregation. In sum, the data suggest that functionalcohesion is possible only in a replication-associated manner.The results agree with a cohesin-mediated binary mode of countingsister plasmid clusters, as is the established mechanism forsister chromatids.

The association of the 2µm plasmid with the mitotic spindle during segregation. The strong association of the 2µm plasmid with the spindlein the mtw1-1 mutant, while a substantial number of chromosomesare detached from it, places constraints on one of the currentlyentertained models for plasmid segregation in which replicatedplasmid clusters hitchhike on sister chromatids (22). Not allchromosomes are uniformly affected by the mtw1-1 mutation. Whereaschromosome III is detached from the spindle quite infrequently,chromosome IV is detached in more than half the cell population(see Fig. S2 in the supplemental material) (17). In a revisedhitchhiking model, the plasmid cluster can be tethered onlyto a member of the chromosome subset that remains stably attachedto the spindle in the mtw-1 background.

As revealed by the ndc10-2 mutation, plasmid localization tochromosome spreads or cohesin association with STB can occurnormally in the presence of a functional spindle even when chromosomesare not attached to the spindle. Yet, this mutation causes the2µm plasmid to stay put with the bulk of the chromosomesin the mother, even though spindle elongation is normal (12,14). This phenotype is not alleviated by combining the ndc10-2mutation with mad2 ${Delta}$ to inactivate the spindle checkpoint andpromote cohesin disassembly (Velmurugan, unpublished). Thus,spindle elongation and cohesin cleavage are not sufficient topromote plasmid segregation when chromosomes fail to do so.We have no reason to suspect that the ndc10-2 mutation itselfmay interfere with plasmid-spindle association. We failed todetect by ChIP assays any interaction between STB and representativemember proteins of the inner, middle, and outer kinetochorecomplexes (S. Mehta, unpublished observations). Therefore, analternative to the hitchhiking model that proposes spindle-dependentbut chromosome-independent plasmid segregation (14) is alsoinadequate.

2µm plasmid segregation: spindle and chromosome dependence? A tentative scheme for 2µm plasmid segregation that accommodatescurrent as well as previous observations is summarized in Fig.8C. In this model, the plasmid cluster interacts with the mitoticspindle as well as the chromosomes, although the nature of theseinteractions is as yet undefined. The three stages of plasmidsegregation are (i) clustering, specific nuclear localization,and chromosome association of the plasmid; (ii) cohesin recruitmentduring replication and pairing of duplicated plasmid clusters;and (iii) cohesin dissociation and movement of the clustersaway from each other.

The outcomes from this study underscore the importance of thespindle at all three stages of the plasmid segregation pathway.By contrast, sister chromatids that have been replicated andcohesed in the absence of the spindle can segregate normallywhen the spindle function is provided. Cohesin itself has norole during stage I, and stages I and II can be executed inthe absence of chromosome-spindle attachment. Thus, plasmidinteractions with a chromosome do not follow passively fromthe independent association of each entity with the spindle.Rather, the spindle appears to actively promote plasmid alliancewith a chromosome, and the two then become coentities in thesegregation process.

New insights into potential functional links between the mitotic spindle and sister chromatid cohesin. Two recent reports, based on a synthetic genetic array analysisusing a ctf8 deletion strain and a microarray hybridizationassay using a ctf4 deletion strain, have brought to light apreviously unrecognized role for mitotic spindle integrity insister chromatid cohesion (13, 24). Ctf8p is a component ofthe alternative replication factor C-like complex Ctf18-RFC,and Ctf4p is a constituent of the replication fork that bindsDNA polymerase ${alpha}$ . These factors contribute to sister chromatidcohesion in S. cerevisiae; however, the mechanisms are not understood.Among the synthetically lethal deletions uncovered using ctf8 ${Delta}$ were bim1 ${Delta}$ , kar3 ${Delta}$ , and vik1 ${Delta}$ (13), and each of the latter threedeletions caused impaired cohesion at centromeric and arm sites.The kar3 deletion, but not bim1 ${Delta}$ or vik1 ${Delta}$ , was also identifiedin the screen with ctf4 ${Delta}$ (24).

KAR3 codes for a minus-end-directed microtubule motor protein(5) that is required for spindle integrity, karyogamy duringmating (15), and proper mitotic spindle positioning (4). Thedata from Mayer et al. (13) suggest that the interaction ofKar3p with one of its accessory factors Vik1p is important insister chromatid cohesion. An interaction between Kar3p andthe cohesin subunit Smc1p had been reported earlier (16). Bim1pis a plus-end microtubule binding protein that is required forthe correct orientation of the mitotic spindle and, as a result,for polarized cell growth. Its newly discovered role in cohesionis consistent with earlier results that link a lack of Bim1pto chromosome missegregation.

The above findings suggest that the spindle-cohesin connectionis not exclusive to the 2µm plasmid, as we had originallybelieved. Nevertheless, the spindle effect on the plasmid ismuch more dramatic than that on the chromosomes. In a preliminaryscreen using the synthetic genetic array deletion strains, wenoticed an elevated loss of a 2µm reporter plasmid inctf8 ${Delta}$ ; furthermore, ChIP assays have revealed the associationof Kar3p with the STB locus (Mehta, unpublished). It is possiblethat the yeast plasmid has adapted and optimized one of theless prominent cellular mechanisms for establishing sister chromatidcohesion to serve a principal function in its own partitioning.

ACKNOWLEDGMENTS

This work was supported by funds from the National Institutesof Health (GM64363). S.M. received partial support through theWilliam Livingston Fellowship from the University of Texas atAustin.

We are grateful to David Botstein, Douglas Koshland, Sue Biggins,and Vincent Guacci for providing strains and reagents.

FOOTNOTES

* Corresponding author. Mailing address for S. Velmurugan: Section of Molecular Genetics & Microbiology, University of Texas at Austin, Austin, TX 78712. Phone: (512) 471-0966. Fax: (512) 471-5546. E-mail: jayaram{at}mail.utexas.edu. Mailing address for M. Jayaram: Section of Molecular Genetics & Microbiology, University of Texas at Austin, Austin, TX 78712. Phone: (512) 471-0966. Fax: (512) 471-5546. E-mail: jayaram{at}mail.utexas.edu.

Supplemental material for this article may be found at http://mcb.asm.org/.

REFERENCES

Top