Current Biology, Vol. 15, 1078–1089, June 21, 2005, ©2005 Elsevier Ltd All rights reserved.DOI 10.1016/j.cub.2005.05.026
Polo-like Kinase 1 Creates the Tension-Sensing
3F3/2 Phosphoepitope and Modulates the Association
of Spindle-Checkpoint Proteins at Kinetochores
Leena J. Ahonen,1,2,6Marko J. Kallio,1,3,6
John R. Daum,1,6Margaret Bolton,4Isaac A. Manke,5
Michael B. Yaffe,5P. Todd Stukenberg,4
and Gary J. Gorbsky1,*
1Molecular, Cell and Developmental Biology
Oklahoma Medical Research Foundation
Oklahoma City, Oklahoma 73104
2Turku Graduate School of Biomedical Sciences
3VTT Medical Biotechnology and University of Turku
4Department of Biochemistry and Molecular Genetics
University of Virginia Medical School
Charlottesville, Virginia 22908
5Department of Biology
Center for Cancer Research
Massachusetts Institute of Technology
Cambridge, Massachusetts 02139
Conclusions: Plk1 generates the 3F3/2 phosphoepi-
tope at kinetochores that are not under tension and
contributes to the normal kinetochore association of
several key proteins important in checkpoint signaling.
Mechanical tension regulates Plk1 accumulation at ki-
netochores and possibly its kinase activity.
The simultaneous separation of sister chromatids at the
beginning of anaphase ensures the equal distribution of
DNA between daughter cells. Because sister-chromatid
cohesion is dissolved, the start of anaphase is a point
of commitment that cannot be reversed. If a cell enters
anaphase without each chromosome bipolarly attached
to the spindle (amphitelic orientation) through the kinet-
ochore complex, the daughter cells are likely to lose
or gain chromosomes. The fidelity of sister chromatid
separation is controlled by a signaling system termed
the spindle checkpoint [1, 2]. This checkpoint monitors
microtubule-kinetochore interactions and delays the
transition from metaphase to anaphase until all chro-
mosomes establish amphitelic orientation and align at
the metaphase plate. When the kinetochores of a sis-
ter-chromatid pair achieve stable bipolar attachment,
tension develops across the sister kinetochores. Clas-
sic micromanipulation experiments performed on the
meiotic bivalent chromosomes of insect spermatocytes
showed that mechanical tension regulates both micro-
tubule stability at kinetochores and signaling of the
spindle checkpoint [3, 4]. Microtubule binding and me-
chanical tension at kinetochores may regulate the dy-
namic interaction of checkpoint proteins at kinet-
ochores [4–8]. Active checkpoint proteins and protein
complexes inhibit the anaphase-promoting complex/
cyclosome (APC/C), an E3 ubiquitin ligase that plays a
central role in controlling anaphase onset and exit from
mitosis [9–12]. Although the basic principle of the spin-
dle checkpoint is understood, it is not yet clear how
changes in tension and microtubule occupancy at ki-
netochores are translated into the biochemical signals
that regulate checkpoint activity.
The 3F3/2 monoclonal antibody binds to a phos-
phoepitope expressed at the kinetochores of unaligned
or unattached chromosomes that lack tension and sig-
nal the spindle checkpoint [4–6, 13]. When sister kinet-
ochores become fully saturated with microtubules and
equal pulling forces are exerted upon them, the expres-
sion of the 3F3/2 phosphoepitope wanes . This typi-
cally coincides with chromosome alignment at the
metaphase plate. Treatment of metaphase cells with
microtubule drugs, such as nocodazole or taxol, that
abrogate kinetochore tension causes the immediate re-
expression of the 3F3/2 phosphoepitope on all kinet-
ochores . These findings suggest that the kinase
that creates the 3F3/2 phosphoepitope may be regu-
lated by mechanical tension and possibly plays an im-
portant role in signaling the spindle checkpoint.
Background: In mitosis, a mechanochemical system rec-
ognizes tension that is generated by bipolar microtubule
attachment to sister kinetochores. This is translated into
multiple outputs including the stabilization of microtu-
bule attachments, changes in kinetochore protein dy-
namics, and the silencing of the spindle checkpoint.
How kinetochores sense tension and translate this into
various signals represent critical unanswered questions.
The kinetochores of chromosomes not under tension are
specifically phosphorylated at an epitope recognized by
the 3F3/2 monoclonal antibody. Determining the kinase
thatgeneratesthe3F3/2phosphoepitope at kinetochores
should reveal an important component of this system
that regulates mitotic progression.
Results: We demonstrate that Polo-like kinase 1 (Plk1)
creates the 3F3/2 phosphoepitope on mitotic kinet-
ochores. In a permeabilized in vitro cell system, the de-
pletion of Xenopus Plk1 from M phase extract leads to
the loss of 3F3/2 kinase activity. Purified recombinant
Plk1 is sufficient to generate the 3F3/2 phosphoepitope
in this system. Using siRNA, we show that the reduction
of Plk1 protein levels significantly diminishes 3F3/2
phosphoepitope expression at kinetochores. The con-
sensus phosphorylation sites of Plk1 show strong sim-
ilarity to the 3F3/2 phosphoepitope sequence deter-
mined by phosphopeptide mapping. The inhibition of
Plk1 by siRNA alters the normal kinetochore associa-
tion of Mad2, Cenp-E, Hec1/Ndc80, Spc24, and Cdc20
and induces a spindle-checkpoint-mediated mitotic
6These authors contributed equally to this work.
Plk1 Creates the Tension-Sensing 3F3/2 Epitope
Aurora B/Ipl1 is thus far the only mitotic kinase that
has been associated with response to mechanical
tension and has been implicated in the correction of
improper microtubule-kinetochore attachments and in
checkpoint signaling [14–18]. Polo-like kinase 1 (Plk1),
appears to be important in many aspects of mitosis in-
cludingcentrosome maturationandthe activationofCdk1-
cyclinB,as well as the phosphorylation and subsequent
removal of cohesin from chromosome arms [19–21].
Plk1 phosphorylates several APC/C subunits in vitro,
although its role in controlling the APC/C remains con-
troversial because silencing Plk1 expression does not
prevent APC/C activation [9, 22, 23]. Here, we show by
in vitro and in vivo studies that Plk1 creates the tension-
sensing 3F3/2 phosphoepitope on the kinetochores of
mitotic chromosomes and that the activity of Plk1 is
required for the normal kinetochore association of sev-
eral checkpoint proteins, including Mad2, Cenp-E, and
with NEM (Figure 1E). In contrast, purified Aurora A and
Cdk1/cyclin B failed to regenerate the 3F3/2 phospho-
epitope (Figure 1E). Aurora A, a centrosome-associated
member of the Aurora kinase family, has in vitro sub-
strate specificity similar to that of Aurora B. Because
we were unable to synthesize and purify active recom-
binant Aurora B kinase, we employed Aurora A as one
control in this exogenous kinase assay.
Mere binding of Plk1 protein, not its kinase activity,
might be responsible for generating immunoreactivity
with the 3F3/2 antibody. To test this possibility, we first
incubated the cells with Plk1 (or with Plk1-depleted ex-
tract supplemented with recombinant Plk1) in the ab-
sence of ATP, washed away unbound kinase, and then
incubated the coverslips with or without ATP and mea-
sured the generation of the 3F3/2 phosphoepitope by
immunofluorescence. Plk1 was able to bind in the ab-
sence of ATP, but kinetochores became labeled with
the 3F3/2 antibody only in the samples incubated with
ATP (Figure S2). Therefore, the binding of Plk1 without
kinase activity is insufficient to generate 3F3/2 antibody
reactivity at kinetochores. The simplest explanation for
these results is that Plk1 directly catalyzes the 3F3/2
phosphoepitope at kinetochores, although it remains
formally possible that Plk1 is required for an NEM-
insensitive downstream event that actually catalyzes
the phosphoepitope. Plk1 is clearly required for the re-
generation of the phosphoepitope in the permeabilized
Results and Discussion
Plk1 Generates the 3F3/2 Kinetochore
Phosphoepitope In Vitro
To identify the kinase that creates the 3F3/2 phospho-
epitope at kinetochores, we used a previously de-
scribed permeabilized cell assay [24, 25]. Ptk1 cells on
coverslips were lysed, dephosphorylated, and treated
with N-ethylmaleimide (NEM), which inactivates endoge-
nous kinetochore bound kinase(s) that create the
3F3/2 phosphoepitope . The kinetochores of these
cytoskeletons were then treated with cell extracts or
purified kinases in the presence of a serine/threonine
phosphatase inhibitor, Microcystin-LR, and the regen-
eration of the 3F3/2 phosphoepitope was monitored by
indirect immunofluorescence (Figure 1; see Figure S1
in the Supplemental Data available with this article
Previously, we had shown that the 3F3/2 phospho-
epitope can be generated on NEM-treated kineto-
chores by mammalian cell extract . Here, we dem-
onstrate that the 3F3/2 phosphoepitope can also be
created by an extract prepared from Xenopus eggs,
which are naturally arrested in second meiotic meta-
phase (Figures 1C and 1D; Xenopus M phase ext). This
Xenopus M phase extract was immunodepleted with an
antibody against Plk1 or Aurora B to identify the 3F3/2
phosphoepitope kinase. Immunoblots confirm that
these extracts were specifically depleted of the corre-
sponding kinases (Figure 1A), while they remained in M
phase, as measured by H1 kinase activity (Figure 1B).
The 3F3/2 kinetochore phosphoepitope was no longer
generated after Plk1 depletion, whereas mock-depleted
or Aurora-B-depleted extracts retained full capacity to
catalyze the 3F3/2 phosphoepitope (Figures 1C and
1D). The supplementation of Plk1-depleted extract with
purified recombinant Plk1 rescued this activity, demon-
strating that Plk1 was the only immunodepleted protein
required for the generation of the 3F3/2 phosphoepi-
tope (Figures 1C and 1D). Finally, the 3F3/2 phospho-
epitope was regenerated at kinetochores when purified
recombinant Plk1 was applied to permeabilized cells in
the absence of cytoplasm or extract after treatment
The 3F3/2 Phosphoepitope Overlaps the Plk1
Consensus Phosphorylation Motif
Oriented-peptide library screening was performed to
determine the consensus phosphorylation motif recog-
nized by the 3F3/2 monoclonal antibody . Bead-
immobilized 3F3/2 antibody was incubated with a
phosphothreonine-oriented peptide library correspond-
ing to the sequence X-X-X-X-pT-X-X-X-X, where pT is
a fixed phosphothreonine residue, and X denotes a
roughly equimolar mixture of all amino acids except
Cys, Ser, and Thr. After being washed extensively, the
3F3/2 bound phosphopeptides were eluted from the
antibody column and sequenced by Edman degrada-
tion. The relative amount of every amino acid in each
position in the bound peptides reveals how strongly
that amino acid is selected during the 3F3/2 binding.
As shown in Figures 2A and 2B, the 3F3/2 binding motif
shows strong selection for Phe and Tyr in the pThr−4
position, Leu and Met in the pThr−3 position, Glu and
Asp in the pThr−2 position, and Asn in the pThr+1 posi-
tion. Additional selection for Asp and Glu in the pThr−3
position, Ile, Leu, Met, His, Gln, and Trp in the pThr−1
position, and Asp, Glu, His, Met, Gln, and Tyr in the
pThr+1, +2, +3, and +4 positions was also observed.
On the basis of these data, the optimal 3F3/2 binding
motif appears to be (F/Y)-(L/M)-(D/E)-(I/L/M)-(pThr/
pSer)-N, although additional selection is present be-
yond that listed in this simplified motif. As with all anti-
bodies, kinases, and modular binding domains, 3F3/2
binding proteins do not necessarily have to match the
optimal motif at each position; rather, the motif reveals
the rank order of optimal amino acids that the anti-
Figure 1. Plk1 Creates the 3F3/2 Phosphoepitope on Kinetochores of Permeabilized Ptk1 Cells
(A) Western-blot analysis showing Xenopus M phase extracts immunodepleted with preimmune serum (lane 1), anti-Plk1 serum (lane 2), and
anti-Aurora B antibody (lane 3).
(B) Extracts immunodepleted of Plk1 or Aurora B maintain high levels of H1 kinase activity as determined with anti-phospho-H1 antibody
(C) Micrographs of Ptk1 cells from the permeabilized in vitro cell assay. Extracts were used to phosphorylate the kinetochores of lysed Ptk1
cells after dephosphorylation and NEM (N-ethylmaleimide)-inactivation of the endogenous kinases. Xenopus M phase extract generates the
3F3/2 phosphoepitope on the kinetochores of Ptk1 cells in prometaphase. The phosphoepitope is no longer generated on kinetochores after
Plk1 depletion, whereas Aurora B depletion has no effect. Recombinant Plk1 restores the 3F3/2 kinase activity of Plk1-depleted extract.
(D) Normalized fluorescence intensities of 3F3/2 signals at the kinetochores (n = 100) of cycling prometaphase Ptk1 cells (n = 10). The bars
indicate mean ± SEM.
(E) Recombinant Plk1 generates the 3F3/2 phosphoepitope on NEM-treated kinetochores, whereas recombinant Aurora A and Cdk1 do not.
Kinetochores were identified with Crest human autoimmune serum. In the merge panel, kinetochores are in green, 3F3/2 antibody in red, and
DNA (DAPI) in blue. The scale bars represent 10 ?m.
Intriguingly, there is partial overlap between the opti-
mal phosphoepitope binding motif of 3F3/2 and a motif
that appears to be phosphorylated by Plk1 on the basis
of mapped phosphorylation sites from known in vivo
substrates (Figures 2B and 2C), namely f-(D/E)-X-
(S/T)-(Aliphatic)-(D/E), in which f denotes hydrophobic
amino acids such as Leu and Met . This partial over-
lap between the Plk1 and 3F3/2 motifs strongly sug-
gests that Plk1 generates 3F3/2 binding sites on a sub-
set of its substrates. In this context, it is important to
note that the limited number of bona fide mapped
in vivo substrates used to derive the Plk1 phosphoryla-
tion motif do not necessarily correspond to those with
3F3/2 epitopes. Furthermore, the optimal Plk1 phos-
Plk1 Creates the Tension-Sensing 3F3/2 Epitope
phorylation motif in the context of totally degenerate
peptides is unknown, although a Plk1 consensus phos-
phorylation motif determined by alanine scanning of the
Cdc25C phosphorylation site is in reasonably good
agreement with the in vivo substrate-derived motif
shown above .
Plk1 Accumulates at Higher Levels on Unaligned
Kinetochores and Colocalizes with the
We immunostained vertebrate tissue culture cells (PANC,
CFPAC, HeLa, and LLCPK) to determine the subcellular
localization of Plk1 with a polyclonal anti-Plk1 antibody.
Cells were simultaneously labeled with Crest human
autoimmune serum, which detects centromeres/kinet-
ochores at all stages of the cell cycle. In agreement
with previous studies [30, 31], Plk1 accumulated at ki-
netochores and spindle poles in early mitosis, translo-
cated to the spindle midzone at the onset of anaphase,
and concentrated at the midbody during telophase
(Figure 3A; Figure S3). However, unlike the previous de-
scriptions of Plk1 localization, we observed significant
differences in Plk1 staining among the kinetochores of
individual prometaphase cells. The concentration of Plk1
was high at the kinetochores of unaligned chromo-
somes and lower at the kinetochores of chromosomes
that had congressed to the spindle equator (Figure 3A).
This localization pattern was verified by transfecting
HeLa and LLCPK cells with a plasmid expressing GFP-
Plk1. GFP-Plk1 localized to kinetochores, spindle poles,
and the spindle midzone in mitotic cells (Figure S4).
Similarly, GFP-Plk1 preferentially accumulated at the ki-
netochores of unaligned chromosomes in prometa-
phase cells (Figure 3B).
Drugs that disassemble or hyperstabilize the micro-
tubules activate the spindle checkpoint. However, in
some cases they have different effects on the accumu-
lation of spindle-checkpoint markers at kinetochores.
For example, the 3F3/2 phosphoepitope and BubR1
accumulate to high levels on the kinetochores of cells
treated with both kinds of microtubule drugs, whereas
Mad2 accumulation at kinetochores increases to a
much greater extent in cells treated with microtubule
depolymerizers [13, 32, 33]. These differences have
contributed to speculation that the spindle checkpoint
may respond via distinct pathways to the loss of micro-
tubule occupancy and to the loss of mechanical ten-
sion at kinetochores [34–36]. We measured GFP-Plk1
accumulation at the kinetochores of untreated meta-
phase cells (which exhibit both microtubule attachment
plus tension), in cells treated with nocodazole (which
exhibit neither microtubule attachment nor tension),
and in cells treated with taxol (which exhibit microtu-
bule attachment but little or no tension). Both nocoda-
zole and taxol increased the concentration of Plk1 at
Figure 2. The 3F3/2 Phosphoepitope Motif and Plk1 Phosphoryla-
tion Motif Show Similar Amino Acid Sequence Specificity
(A) Determination of the 3F3/2 phosphoepitope motif by oriented
peptide library screening. The panels shown correspond to dif-
ferent positions in the motif in relation to the phosphothreonine and
demonstrate the enrichment of each amino acid in that position in
the 3F3/2 bound peptides normalized to the starting library mixture.
Residues showing particularly strong selection in several of the se-
quencing cycles are denoted by asterisks. Cys, Ser, and Thr were
omitted from the library to circumvent oxidation during synthesis.
(B) The overlap of the optimal 3F3/2 motif and consensus Plk1
phosphorylation motif from Nakajima et al. . f denotes hy-
drophobic amino acids and (Al) denotes aliphatic amino acids.
(C) Plk1 phosphorylation sites in known in vivo substrates .
kinetochores (Figure 3C), indicating that this accumula-
tion is stimulated by loss of mechanical tension. The
increased accumulation of Plk1 at kinetochores after
treatment with nocodazole or taxol is consistent with
the observation that both drugs also increase the 3F3/2
phosphoepitope levels at the kinetochores [13, 25, 37].
It also explains why in permeabilized cells, the kinet-
ochores of unaligned chromosomes show greater abil-
ity to regenerate the 3F3/2 phosphoepitope with their
endogeneous kinetochore bound kinase . However,
results from micromanipulation experiments performed
with permeabilized insect spermatocytes are consis-
tent with the idea that mechanical tension may directly
inhibit 3F3/2 kinase activity or access to substrate .
Thus, mechanical tension may have multiple roles in
regulating Plk1 concentration and activity at kinet-
ochores. Reduced association and activity of Plk1 may
participate in stabilization of microtubule-kinetochore
attachments and in silencing the spindle checkpoint.
If Plk1 is responsible for generating the 3F3/2 phos-
phoepitope at kinetochores in vivo, Plk1 should localize
to the same region of the kinetochore as the 3F3/2
phosphoepitope. Using high-resolution immunofluores-
cence in LLCPK cells, we found that Plk1 and 3F3/2
were closely colocalized (Figures 4A and 4B). Plk1 was
concentrated exterior to the inner centromere marker
Aurora B and colocalized with the structural kinetochore
protein Hec1/Ndc80 (Figure 4C). Immunoelectron micro-
scopic analyses indicate that the 3F3/2 phosphoepitope
is concentrated in the central electron lucent layer,
whereas Hec1/Ndc80 appears to reside primarily in the
outer dense plaque of the trilaminar kinetochore struc-
ture [37, 38]. Thus, within the limits of light microcopic
resolution, our analysis suggests that Plk1 and the
3F3/2 phosphoepitope colocalize at the mid-to-outer
layers of the kinetochore trilaminar plate.
Inhibition of Plk1 Expression by siRNA Causes a
Checkpoint-Mediated M Phase Arrest and Reduces
the 3F3/2 Phosphoepitope at Kinetochores In Vivo
HeLa cells were transfected with siRNA against Plk1,
and the efficiency of knockdown was quantified by
Western blotting. Plk1 protein level was reduced by
R85% in extract from siRNA-treated cultures in com-
parison to extract from control cells (Figure 5A). As pre-
becomes undetectable by the time cells reach metaphase, and
only poles are labeled (lower row). The cells were labeled with Crest
automimmune serum to detect kinetochores (green in the merge
panel) and anti-Plk1 antibody (red), and they were counterstained
with DAPI (DNA, blue).
(B) HeLa cells transfected with GFP-Plk1 plasmid showing the lo-
calization of the fluorescent fusion protein. GFP-Plk1 (green in the
merge panel) localizes to all kinetochores in early prometaphase
(top row), but later in prometaphase (lower row), the kinetochores
of unaligned chromosomes show stronger signals (circle). Spindle
poles are marked with arrows. The cells were stained with DAPI
(DNA, red in the merge panel).
(C) GFP-Plk1 accumulation at kinetochores increases upon treat-
ment with the microtubule-depolymerizing drug nocodazole or the
microtubule-hyperstabilizing drug taxol. The graph indicates mean
intensity ± SEM normalized to 100% for the nocodazole-treated
population. The micrographs show a representative LLCPK cell
from each category. The scale bars represent 10 ?m.
Figure 3. Plk1 Accumulates at the Kinetochores of Unaligned Chro-
(A) Mitotic PANC cells labeled with polyclonal anti-Plk1 antibody
showing the accumulation of the protein at kinetochores. In the top
row, two prometaphase cells show Plk1 accumulation at kinet-
ochores; the mid-prometaphase cell on the right shows Plk1 accu-
mulation at the kinetochores of unaligned chromosomes (circles),
whereas the kinetochores of chromosomes at the metaphase plate
(between the circles) show very weak signals. The early-prometa-
phase cell on the left side shows Plk1 accumulation on all kinet-
ochores. Spindle poles are marked with arrows. The Plk1 signal
Plk1 Creates the Tension-Sensing 3F3/2 Epitope
Figure 5. Inhibition of Plk1 Expression with siRNA Induces Spindle-
Checkpoint-Mediated Prometaphase Arrest
(A) A Western blot of M phase HeLa cell extracts probed with poly-
clonal antibody to Plk1 shows an w85% reduction in Plk1 protein
level after siRNA treatment.
(B) A micrograph depicting three mitotic HeLa cells from a popula-
tion treated with Plk1 siRNA. Two cells show the typical monopolar
Plk1 knockdown phenotype, and the third mitotic cell has formed
a normal bipolar spindle. The cells were labeled with antibody
against tubulin (red), and DNA was counterstained with DAPI
(green). The scale bar represents 10 ?m.
(C) The prometaphase arrest of Plk1 siRNA-treated cells is due to
activation of the spindle checkpoint. The aurora kinase inhibitor
ZM447439 overrides the spindle checkpoint and causes mitotic
exit in cells arrested by Plk1 siRNA, by taxol, or by nocodazole at
low concentration. It does not cause mitotic exit in cells arrested
with a high concentration of nocodazole or with the proteasome
Figure 4. Plk1 and the 3F3/2 Phosphoepitope Colocalize at the Ki-
netochores of Unaligned Chromosomes
(A) The panels show micrographs from a LLCPK cell at late pro-
metaphase. A partially congressed chromosome is highlighted
(white box) and shown at higher magnification. The fluorescent in-
tensities of the 3F3/2 (red in merge image and plot) and Plk1 (green)
antibody labeling were line scanned along the white dotted arrow,
and the intensites were plotted. The phase-constrast image shows
the location of the scanned chromosome (white arrow) and the
nearest spindle pole (yellow arrow).
(B) Similar analysis of a cell treated with nocodazole.
(C) Partially congressed chromosomes in LLCPK cells colabeled
with Plk1 (green) and Aurora B (red) antibodies (upper row), and
with Plk1 (green) and Hec1/Ndc80 (red) antibodies (lower row).
Line-scan plots show that Plk1 localizes exterior to Aurora B and
colocalizes with 3F3/2 epitope and Hec1/Ndc80.
viously described [39, 40], siRNA against Plk1 resulted
in mitotic arrest or delay that was followed by cell
death. To characterize this phenotype in more detail,
we transfected cells with Plk1 siRNA and processed
the cells for immunofluorescence or followed them by
time-lapse microscopy. Our results showed that 68% of
the M-phase-arrested cells after Plk1 siRNA treatment
exhibited a monopolar phenotype, in which chromo-
somes were arrayed around the unseparated spindle
poles (Figure 5B). In agreement with previous studies
[39, 40], we found that the inhibition of Plk1 expression
by siRNA induced arrest in a prometaphase-like state
for several hours and eventually resulted in cell death
with characteristics of apoptosis (data not shown).
To test whether the prometaphase-like arrest result-
ing from silencing Plk1 expression was due to the acti-
vation of the spindle checkpoint, we treated cells with
Plk1 siRNA to induce mitotic arrest and subsequently
with ZM447439, a chemical inhibitor of Aurora B kinase.
In vertebrate cells, the inhibition of Aurora B abrogates
the spindle checkpoint and, as a consequence, leads
to mitotic exit [15–17]. ZM447439 induced mitotic exit
in cells arrested in M phase by Plk1 siRNA and in cells
that were first arrested in M phase with nocodazole
(330 nM) and then transferred into medium containing
5 ?M taxol or a low (66 nM) concentration of nocoda-
zole (Figure 5C). Consistent with previous results with
ZM447439, override of the checkpoint did not occur
over the time course of the experiment in cells arrested
with a high concentration (330 nM) of nocodazole (Fig-
ure 5C). As expected, cotreatment with the proteasome
inhibitor MG132 blocked ZM447439-induced mitotic
exit because proteasome activity is required for mitotic
exit downstream of the spindle checkpoint (Figure 5C).
These results are entirely consistent with other recent
reports showing that reduction of Plk1 levels by treat-
ment with siRNA leads to mitotic arrest that is depen-
dent on active signaling by the spindle checkpoint
To assess the effect of Plk1 inhibition on the expres-
sion of the 3F3/2 kinetochore phosphoepitope in vivo,
we analyzed mitotic chromosomes that were isolated
from cells arrested in M phase with Plk1 siRNA, with
nocodazole, or with Plk1 siRNA and then incubated
with nocodazole. Chromosomes were isolated both in
the presence and absence of the phosphatase inhibitor
Microcystin-LR, and the intensity of labeling with the
3F3/2 antibody was quantified. M phase chromosomes
prepared from nocodazole-arrested cells in the pres-
ence of the phosphatase inhibitor exhibited robust ex-
pression of the 3F3/2 kinetochore phosphoepitope.
Chromosomes isolated from Plk1 siRNA-treated cells
in presence of phosphatase inhibitor showed markedly
reduced 3F3/2 phosphoepitope expression (Figures
6A and 6B). This reduction in 3F3/2 phosphoepitope
was also observed on kinetochores of chromosomes
isolated from Plk1 siRNA-treated cells that had subse-
quently been treated with nocodazole to remove micro-
tubule-kinetochore interactions (Figure 7A). Chromo-
somes isolated in the absence of the phosphatase
inhibitor from these sources lost the 3F3/2 phosphoepi-
tope owing to the action of endogenous kinetochore
phosphatases. If chromosomes were first isolated in
the absence of phosphatase inhibitor and then subse-
Figure 6. Plk1 Suppression by siRNA Significantly Reduces Expres-
sion of the 3F3/2 Phosphoepitope at Kinetochores In Vivo
(A) Micrographs of isolated chromosomes from control nocoda-
zole-arrested cells and Plk1 siRNA-treated cultures show that re-
ducing Plk1 expression causes significant diminution of 3F3/2
phosphoepitope expression. The insets show a higher-magnifica-
tion view of individual chromosomes. 3F3/2 labeling is shown in
red, Crest autoimmune serum in green, and DNA staining (DAPI)
(B) The graph shows quantification of the 3F3/2 phosphoepitope
at the kinetochores of chromosomes isolated in the presence or
absence of the phosphatase inhibitor Microcystin-LR, and after en-
dogenous rephosphorylation. In endogenous rephosphorylation,
kinetochore bound kinase recreates the 3F3/2 phosphoepitope
in vitro on dephosphorylated kinetochores in the presence of ATP
and Microcystin-LR. The bars indicate mean ± SEM. The scale bar
represents 5 ?m.
quently supplied with ATP and the phosphatase inhibi-
tor, Microcystin-LR, endogenous kinase at the kinet-
ochores could regenerate the 3F3/2 phosphoepitope
(Figure 6B). The chromosomes from the nocodazole-
treated samples regenerated the 3F3/2 phosphoepi-
tope to high levels, whereas the chromosomes from the
Plk1 siRNA-treated population regenerated it only to
low levels. It might have been expected that even a
residual amount of kinase could regenerate high levels
Plk1 Creates the Tension-Sensing 3F3/2 Epitope
of 3F3/2 phosphoepitope. This did not occur, suggest-
ing that in Plk1 siRNA-treated cells, the remaining ki-
nase may have access to a restricted portion of the
substate or that the amount of the substrate is also
reduced. The fact that rephosphorylation was limited
when the amount of endogenous Plk1 was reduced by
siRNA is also consistent with our previous finding that
the endogenous kinase on kinetochores of aligned meta-
phase chromosomes showed limited ability to regener-
ate the 3F3/2 phosphoepitope in the permeabilized cell
model . In summary, these results demonstrate that
the knockdown of Plk1 limits 3F3/2 phosphoepitope
generation in vivo and the ability of isolated chromo-
somes to regenerate the epitope in vitro.
Inhibition of Plk1 Expression Inhibits Normal
Kinetochore Association of Mad2, Cenp-E,
Hec1/Ndc80, and Spc24
We examined whether the inhibition of Plk1 expression
by siRNA affects the accumulation of kinetochore pro-
teins that have been shown to be important in spindle-
checkpoint signaling. We isolated chromosomes from
cells arrested in mitosis with nocodazole, with siRNA
to Plk1, and from siRNA cultures that were also treated
with nocodazole. We quantified the kinetochore-asso-
ciated 3F3/2 phosphoepitope, Mad2, Cenp-E, two com-
ponents of the Ndc80 complex (Hec1/Ndc80 and
Spc24), Bub1 kinase, and BubR1 kinase. We also mea-
sured the labeling of kinetochores with human Crest
autoimmune antibodies that were used to identify ki-
netochores on the isolated chromosomes. Our results
show that chromosomes isolated from cells arrested in
M phase with Plk1 siRNA exhibited significant reduc-
tions in the levels of Mad2, Cenp-E, Hec1/Ndc80, and
Spc24, whereas Bub1, BubR1, and Crest antigens were
not significantly affected (Figure 7A). The treatment of
siRNA-arrested cultures with nocodazole for 45 min re-
sulted in the partial restoration of Mad2, Cenp-E, Hec1/
Ndc80, and Spc24, although their levels remained nota-
bly reduced in comparison to chromosomes from noco-
Figure 7. Inhibition of Plk1 Expression with siRNA Reduces the Ki-
netochore Accumulation of Mad2, Cenp-E, Hec1, and Spc24, but
not Bub1, BubR1, or Crest Autoimmune Antigens
(A) Quantification of various kinetochore-associated proteins on
isolated mitotic HeLa cell chromosomes. 3F3/2 phosphoepitope,
Mad2, Cenp-E, Hec1, and Spc24 levels were reduced after Plk1
siRNA treatment in the presence or absence of nocodazole. Bub1,
BubR1, and Crest antigens were only slightly altered. Bars indicate
mean ± SEM.
(B) The quantification of 3F3/2, Mad2, and BubR1 at the kinet-
ochores of chromosomes isolated from cells treated with nocoda-
zole, Plk1 siRNA, or Plk1 siRNA and then incubated with nocoda-
zole. Chromosomes were isolated in the presence or absence of
the phosphatase inhibitor Microcystin-LR. The kinetochore accu-
mulation of Mad2 in vivo is decreased by siRNA to Plk1, and this
is partially restored by nocodazole. Retention of Mad2 at kinet-
ochores during chromosome isolation is not affected by the pres-
ence or absence of the phosphatase inhibitor. The kinetochore ac-
cumulation of BubR1 is unaffected in cells treated with Plk1 siRNA.
However, retention of BubR1 during chromosome isolation is re-
duced in chromosomes isolated without phosphatase inhibitor.
Bars indicate mean ± SEM.
dazole-treated cells. These results showing the partial
retention of checkpoint proteins at kinetochores of cells
treated with siRNA to Plk1 may account for the pro-
metaphase delay phenotype of Plk1 knockdown cells.
We speculate that although Plk1 is substantially re-
duced in cells treated with siRNA to Plk1, residual
amounts of Plk1 allow checkpoint signaling that is suffi-
cient to induce a delay in M phase. This delay may oc-
cur because even slightly reduced Plk1 levels result in
defective attachments of all chromosomes to the spin-
dle microtubules. Thus, although each individual chro-
mosome in a cell with reduced Plk1 may have a weak-
ened checkpoint signaling potential, the sum of signaling
from all the chromosomes may still induce significant
delays in M phase. Moreover, although a clear mitotic
delay occurs in cells treated with siRNA to Plk1, the
delay may be limited in duration in comparison to con-
trol cells with normal Plk1 levels; in these cells, the
spindle checkpoint is activated with microtubule drugs.
A potentially analogous situation occurs with siRNA
targeting of the Hec1/Ndc80 complex. In most studies,
siRNA inhibition of Hec1/Ndc80 complex proteins
caused a prometaphase arrest [41, 42]. However, Mer-
aldi et al.  showed that whereas a partial repression
of Ndc80 complex proteins results in a mitotic-arrest
phenotype, more complete knockdown abrogated the
Previous studies have implicated the preservation of
kinetochore phosphorylation and the 3F3/2 phospho-
epitope with retention of the checkpoint proteins Mad2
and BubR1 at kinetochores [34, 44]. We prepared chro-
mosomes in the presence or absence of the phospha-
tase inhibitor Microcystin-LR from cultures arrested
with nocodazole, with siRNA to Plk1, or with siRNA to
Plk1 followed by nocodazole treatment. In the absence
of phosphatase inhibitor, the 3F3/2 kinetochore phos-
phoepitope was almost completely lost from chromo-
somes in each of the three conditions (Figure 7B). The
association of Mad2 with kinetochores was reduced on
chromosomes from Plk1 siRNA-treated cells, indicating
that Plk1 is required for kinetochore accumulation
in vivo. However, Mad2 labeling was not signicantly af-
fected by the presence or absence of Microcystin-LR,
suggesting that neither the 3F3/2 kinetochore phospho-
epitope nor other phosphorylations preserved by Micro-
cystin-LR are required to retain Mad2 at kinetochores
when cells are lysed (Figure 7B). In contrast, BubR1
kinetochore accumulation was notably reduced when
Microcystin-LR was omitted but was not affected by
reduced Plk1 expression. This suggests that, unlike
Mad2, full retention of BubR1 at kinetochores after cell
lysis requires preservation of one or more phosphoryla-
tions distinct from the 3F3/2 phosphoepitope and inde-
pendent of Plk1. Aurora B kinase may regulate BubR1
association with kinetochores because the inhibition of
Aurora B kinase activity depletes BubR1 from kinet-
ochores in vivo [16, 17].
Figure 8. Inhibition of Plk1 Expression with siRNA Reduces the Ac-
cumulation of Cdc20 at Kinetochores
(A) Cycling HeLa cells were mock transfected or transfected with
Plk1 siRNA and subsequently treated with DMSO, nocodazole, or
monastrol. Cells were then fixed and processed for immunolabeling
with Crest autoimmune serum (green in merge panel) and anti-
Cdc20 antibody (red) and counterstained with DAPI for DNA (blue).
The scale bars represent 10 ?m.
(B) Quantification of kinetochore bound Cdc20. The bars indicate
mean ± SEM. Cdc20 accumulation at kinetochores of mitotic cells
is markedly reduced by Plk1 siRNA in all conditions.
sessing whether kinetochores are attached to microtu-
bules and under tension and in disseminating this infor-
mation to the rest of the cell [45–47]. Therefore, we
carried out additional analyses of the effects of Plk1
siRNA on the kinetochore association of Cdc20. Be-
cause Cdc20 is released from kinetochores under all
conditions when cells are lysed, the analysis of isolated
chromosomes is not useful. Instead, we quantified ki-
netochore-associated Cdc20 in whole cells simulta-
neously fixed and permeabilized. The inhibition of Plk1
expression with siRNA significantly reduced Cdc20 ac-
cumulation at kinetochores (Figures 8A and 8B). Cdc20
Plk1 Is Required for the Normal Kinetochore
Association of Cdc20 in Mammalian Cells
We and others have postulated that the binding, modifi-
cation, and release of the APC/C activator Cdc20 at
kinetochores may function in the mechanism for as-
Plk1 Creates the Tension-Sensing 3F3/2 Epitope
3F3/2 Phosphopeptide Library Screening
A phosphothreonine-oriented degenerate peptide library con-
sisting of the sequence Met-Ala-X-X-X-X-pThr-X-X-X-X-Ala-Lys-
Lys-Lys, in which X denotes all amino acids except Cys, Ser, and
Thr, was synthesized with N-α-FMOC-protected amino acids and
standard BOP/HOBt coupling chemistry. Peptide library screening
was performed with 125 ?l of Protein G beads complexed to a
saturating amount of the 3F3/2 monoclonal antibody (w1 mg). 3F3/2
beads were loaded into a 1 ml column and incubated with 0.45 mg
of the peptide library mixture for 10 min at RT in Tris-buffered saline
(TBS; 50 mM Tris-HCl [pH 8.0] and 150 mM NaCl). Unbound pep-
tides were removed from the column by three washes with TBS
containing 1.0% NP-40 and then three washes with Phosphate-
buffered saline (150 mM NaCl, 3 mM KCl, 10 mM Na2HPO4, and
2 mm KH2PO4[pH 7.6]). Bound peptides were eluted with 30%
acetic acid for 10 min at RT, lyophilized, resuspended in H2O, and
sequenced by automated Edman degradation on a Procise protein
microsequencer. Selectivity values for each amino acid were deter-
mined by comparing the relative abundance (mole percentage) of
each amino acid at a particular sequencing cycle in the recovered
peptides to that of each amino acid in the original peptide library
mixture at the same position.
accumulation at kinetochores remained dependent on
Plk1 in cells treated with nocodazole or with the Eg5
inhibitor monastrol, which generates monopolar spin-
dles similarly to the effect of Plk1 knockdown. Our re-
sults suggest that Plk1 markedly affects the kinet-
ochore association of Cdc20 in vivo. We speculate that
Plk1 may be involved in regulation of the rapid turnover
of Cdc20 at kinetochores and may thereby contribute
to the activation and dissemination of the spindle
checkpoint [46, 48].
Here, we demonstrate that Plk1 accumulates at una-
ligned kinetochores and generates the 3F3/2 phos-
phoepitope, whose expression was previously found to
be regulated by mechanical tension [4–6]. Because
Plk1 levels at kinetochores decrease with bipolar at-
tachment, at least a portion of Plk1 regulation may in-
volve alteration of its affinity for kinetochores. This con-
trol could be due to mechanical tension imparted by
the attachment of microtubules. On the basis of previ-
ous micromanipulation studies , mechanical tension
may also directly inhibit the kinase activity of Plk1 or its
access to substrate. We hypothesize that the reduced
association of Plk1 or the reduction of its kinase activity
at kinetochores in response to bipolar attachment and
inter-kinetochore tension may stabilize microtubule at-
tachment to kinetochores and contribute to the deacti-
vation of the spindle checkpoint.
Transfections and Collection of Cells
HeLa or LLCPK cells on coverslips were transfected with murine
pShuttle-CMV-HA-EGFP-Plk1 plasmid (kind gift from Dr. Kyung
Lee) with Effectene transfection reagent according to the manufac-
turer’s (Qiagen) protocol. Some cells were further treated with vari-
ous drugs for immunofluorescence analysis: 20 ?M MG132 for 2 hr,
3 ?M nocodazole for 1 hr, and 0.6 ?M taxol for 1 hr. For siRNA-
mediated Plk1 knockdown, we utilized a published target se-
quence: 5#-AAG ATC ACC CTC CTT AAA TAT-3# (Plk1 Duplex IV,
Dharmacon) . HeLa cells were transfected at w70% confluency
with Oligofectamine (Invitrogen) according to the manufacturer’s
recommendations. Plk1 siRNA was used at 50150 nM. Plk1 siRNA-
arrested cells were harvested from the culture plate by wash-off
at 2430 hr after transfection and used for immunoblot analysis to
examine the specificity of rabbit anti-Plk1 antibody (Plk1 NT, Up-
state) and the efficiency of Plk1 knockdown.
Phosphorylation of Detergent-Lysed Ptk1 Cells
Ptk1 cells growing on coverslips were rinsed briefly in 50 mM Tris
(pH 7.5) and 5 mM MgSO4(TM buffer) and extracted for 5 min in
TM buffer containing 1% CHAPS, 1 mM DTT, and 5 ?g/ml protease
inhibitor coctail (pepstatin A, leupeptin, and Pefabloc SC). This was
followed by dephosphorylation for an additional 5 min in TM buffer
containing DTT and protease inhibitors. Next, the cells on cover-
slips were incubated for 10 min in 5 mM N-ethylmaleimide (NEM)
in TM buffer to inactivate endogenous kinases. After being rinsed,
cells were phosphorylated with either mitotic Xenopus egg extracts
or recombinant kinases  in TM buffer containing 1 mM ATP,
1 mM DTT, 5 ?g/ml protease inhibitors, and 400 nM Microcystin-
LR for 20−40 min.
Mitotic HeLa cells, newly arrested either by incubation with noco-
dazole for 4 hr or by Plk1 siRNA transfection as described above,
were collected by wash-off. These cells were subsequently incu-
bated in 24-well tissue-culture plates with either media alone, 5 ?M
taxol, 330 nM nocodazole, or 66 nM nocodazole in the presence or
absence of 25 ?M MG132 for 45 min. All cultures were incubated
for 4 hr after ZM447439 addition to allow cells that had exited mito-
sis to adhere to the plate surface. The wells were rinsed, and the
adherent cells were fixed in 2% formaldehyde/1% Triton X-100/
PHEM containing a 1:10,000 dilution of SyberGold (Molecular
Probes). The relative number of interphase cells (the measurement
of mitotic exit) was determined with a plate reader (Tecan GENios)
that measured the fluorescence of SyberGold bound to nucleic
acid in these fixed, adherent cells.
Immunodepletion of Mitotic Xenopus Extracts
Extracts were prepared as described previously . Interphase
extracts were driven into mitosis by the addition of 66 nM nonde-
gradable glutathione S-transferase (GST)-cyclin B for 20 min at RT.
Extracts were immunodepleted with preimmune serum, anti-Plk1
(Xenopus homolog Plx1, hereafter named Plk1) serum, or poly-
clonal anti-Aurora B antibody. Anti-Plk1 serum was a kind gift from
Jan-Michael Peters (Research Institute of Molecular Pathology, Vi-
enna, Austria). The anti-Aurora B polyclonal antibody was pre-
viously characterized . Plk1 and Aurora B were depleted from
150 ?l of mitotic extract by three consecutive incubations with 25
?l of protein A Sepharose beads (Amersham Biosciences) coupled
to either 20 ?g of sera or 2 ?g of affinity-purified antibody. The
antibody-conjugated beads were subsequently washed five times
in XB (10 mM HEPES [pH 7.7], 1 mM MgCl2, 100 mM KCl, and 50
mM sucrose) containing an additional 200 mM NaCl and 0.1% Brij
35. One microliter equivalent of extract was analyzed by SDS-PAGE
and immunoblot to determine the efficiency of depletion. The his-
tone H1 activity of preimmune, Plk1, and Aurora-B-depleted ex-
tracts was determined as previously described .
HeLa cells were transfected with Plk1 siRNA as described above.
Eighteen hours after transfection, mitotic cells were removed and
discarded from the culture by shake-off and pipetting. Cells ar-
rested in mitosis by Plk1 siRNA were then permitted to accumulate
for 3–4 hr. These freshly arrested mitotic cells were harvested, and
their chromosomes were isolated as described below. Plk1 siRNA-
arrested cells were exposed to 330 nM nocodazole for the final 45
min of the 3–4 hr culture period. Nontransfected HeLa cells were
arrested in mitosis by incubation with 330 nM nocodazole and har-
vested for chromosome isolation in the same manner as the mito-
tics collected from the Plk1 siRNA-transfected cell culture. Har-
vested mitotic cells were washed in 10 mM HEPES (pH 7.4), 40 mM
KCl, 5 mM EGTA, 4 mM MgSO4, and 400 nM Microcystin-LR by
centrifugation. Mitotics were lysed in 60 mM Pipes, 25 mM HEPES
(pH 6.9), 10 mM EGTA, 4 mM MgSO4(PHEM), 0.5% Triton X-100,
1 mM DTT, 400 nM Microcystin-LR, and 5 ?g/ml protease inhibitor
cocktail. The extracts were centrifuged through a cushion of lysis
buffer containing 10% glycerol over poly-L-lysine-treated glass
coverslips at 1500 × g for 10 min at 4°C in order to collect chromo-
somes for immunofluorescence labeling. The chromosome-coated
coverslips were then fixed in PHEM and 1.5% formaldehyde for 15
min and processed for immunofluorescence analysis as described
in the Supplemental Experimental Procedures.
of a phosphoepitope at the kinetochores of moving chromo-
somes. J. Cell Biol. 122, 1311–1321.
14. Tanaka, T.U., Rachidi, N., Janke, C., Pereira, G., Galova, M.,
Schiebel, E., Stark, M.J., and Nasmyth, K. (2002). Evidence that
the Ipl1-Sli15 (Aurora kinase-INCENP) complex promotes chro-
mosome bi-orientation by altering kinetochore-spindle pole
connections. Cell 108, 317–329.
15. Kallio, M.J., McCleland, M.L., Stukenberg, P.T., and Gorbsky,
G.J. (2002). Inhibition of aurora B kinase blocks chromosome
segregation, overrides the spindle checkpoint, and perturbs
microtubule dynamics in mitosis. Curr. Biol. 12, 900–905.
16. Ditchfield, C., Johnson, V.L., Tighe, A., Ellston, R., Haworth, C.,
Johnson, T., Mortlock, A., Keen, N., and Taylor, S.S. (2003). Au-
rora B couples chromosome alignment with anaphase by
targeting BubR1, Mad2, and Cenp-E to kinetochores. J. Cell
Biol. 161, 267–280.
17. Hauf, S., Cole, R.W., LaTerra, S., Zimmer, C., Schnapp, G., Wal-
ter, R., Heckel, A., van Meel, J., Rieder, C.L., and Peters, J.M.
(2003). The small molecule Hesperadin reveals a role for Aurora
B in correcting kinetochore-microtubule attachment and in
maintaining the spindle assembly checkpoint. J. Cell Biol. 161,
18. Biggins, S., and Murray, A.W. (2001). The budding yeast protein
kinase Ipl1/Aurora allows the absence of tension to activate
the spindle checkpoint. Genes Dev. 15, 3118–3129.
19. Lane, H.A., and Nigg, E.A. (1996). Antibody microinjection re-
veals an essential role for human polo-like kinase 1 (Plk1) in
the functional maturation of mitotic centrosomes. J. Cell Biol.
20. Liu, X., and Erikson, R.L. (2002). Activation of Cdc2/cyclin B
and inhibition of centrosome amplification in cells depleted of
Plk1 by siRNA. Proc. Natl. Acad. Sci. USA 99, 8672–8676.
21. Gimenez-Abian, J.F., Sumara, I., Hirota, T., Hauf, S., Gerlich, D.,
de la Torre, C., Ellenberg, J., and Peters, J.M. (2004). Regula-
tion of sister chromatid cohesion between chromosome arms.
Curr. Biol. 14, 1187–1193.
22. van Vugt, M.A., van de Weerdt, B.C., Vader, G., Janssen, H.,
Calafat, J., Klompmaker, R., Wolthuis, R.M., and Medema, R.H.
(2004). Polo-like kinase-1 is required for bipolar spindle forma-
tion but is dispensable for anaphase promoting complex/
Cdc20 activation and initiation of cytokinesis. J. Biol. Chem.
23. Kraft, C., Herzog, F., Gieffers, C., Mechtler, K., Hagting, A.,
Pines, J., and Peters, J.M. (2003). Mitotic regulation of the hu-
man anaphase-promoting complex by phosphorylation. EMBO
J. 22, 6598–6609.
24. Renzi, L., Gersch, M.S., Campbell, M.S., Wu, L., Osmani, S.A.,
and Gorbsky, G.J. (1997). MPM-2 antibody-reactive phosphor-
ylations can be created in detergent-extracted cells by kinet-
ochore-bound and soluble kinases. J. Cell Sci. 110, 2013–2025.
25. Waters, J.C., Chen, R.H., Murray, A.W., and Salmon, E.D.
(1998). Localization of Mad2 to kinetochores depends on
microtubule attachment, not tension. J. Cell Biol. 141, 1181–
26. Campbell, M.S., Daum, J.R., Gersch, M.S., Nicklas, R.B., and
Gorbsky, G.J. (2000). Kinetochore “memory” of spindle check-
point signaling in lysed mitotic cells. Cell Motil. Cytoskeleton
27. Yaffe, M.B., and Cantley, L.C. (2000). Mapping specificity deter-
minants for protein-protein association using protein fusions
and random peptide libraries. Methods Enzymol. 328, 157–170.
28. Barr, F.A., Sillje, H.H., and Nigg, E.A. (2004). Polo-like kinases
and the orchestration of cell division. Nat. Rev. Mol. Cell Biol.
29. Nakajima, H., Toyoshima-Morimoto, F., Taniguchi, E., and Ni-
shida, E. (2003). Identification of a consensus motif for Plk
(Polo-like kinase) phosphorylation reveals Myt1 as a Plk1 sub-
strate. J. Biol. Chem. 278, 25277–25280.
30. Golsteyn, R.M., Mundt, K.E., Fry, A.M., and Nigg, E.A. (1995).
Cell cycle regulation of the activity and subcellular localization
of Plk1, a human protein kinase implicated in mitotic spindle
function. J. Cell Biol. 129, 1617–1628.
31. Arnaud, L., Pines, J., and Nigg, E.A. (1998). GFP tagging re-
veals human Polo-like kinase 1 at the kinetochore/centromere
region of mitotic chromosomes. Chromosoma 107, 424–429.
Supplemental Experimental Procedures and several supplemental
figures are available at http://www.current-biology.com/cgi/
We thank Jasminder Weinstein, Tim Yen, Rey Chen, and Steve Tay-
lor for generously providing antibodies. We are grateful to Elaine
Brown of Astra Zeneca for providing ZM447439. We thank Dr. Ky-
ung Lee for providing the plasmid encoding EGFP-Plk1. L.J.A. and
M.J.K. are supported by the Academy of Finland (8206930), the
European Commission (MC002697), and the Finnish Cultural Foun-
dation. J.R.D. and G.J.G. are supported by the National Institute of
General Medical Sciences (NIGMS; GM50412). M.B. and P.T.S. are
supported by the NIGMS (GM63045).
Received: November 16, 2004
Revised: May 4, 2005
Accepted: May 5, 2005
Published online: May 19, 2005
1. Musacchio, A., and Hardwick, K.G. (2002). The spindle check-
point: Structural insights into dynamic signalling. Nat. Rev. Mol.
Cell Biol. 3, 731–741.
2. Cleveland, D.W., Mao, Y., and Sullivan, K.F. (2003). Centro-
meres and kinetochores: From epigenetics to mitotic check-
point signaling. Cell 112, 407–421.
3. Nicklas, R.B., and Koch, C.A. (1969). Chromosome microma-
nipulation. 3. Spindle fiber tension and the reorientation of mal-
oriented chromosomes. J. Cell Biol. 43, 40–50.
4. Li, X., and Nicklas, R.B. (1995). Mitotic forces control a cell-
cycle checkpoint. Nature 373, 630–632.
5. Nicklas, R.B., Ward, S.C., and Gorbsky, G.J. (1995). Kinet-
ochore chemistry is sensitive to tension and may link mitotic
forces to a cell cycle checkpoint. J. Cell Biol. 130, 929–939.
6. Nicklas, R.B., Campbell, M.S., Ward, S.C., and Gorbsky, G.J.
(1998). Tension-sensitive kinetochore phosphorylation in vitro.
J. Cell Sci. 111, 3189–3196.
7. Hoffman, D.B., Pearson, C.G., Yen, T.J., Howell, B.J., and
Salmon, E.D. (2001). Microtubule-dependent changes in as-
sembly of microtubule motor proteins and mitotic spindle
checkpoint proteins at PtK1 kinetochores. Mol. Biol. Cell 12,
8. Nicklas, R.B., Waters, J.C., Salmon, E.D., and Ward, S.C.
(2001). Checkpoint signals in grasshopper meiosis are sensi-
tive to microtubule attachment, but tension is still essential. J.
Cell Sci. 114, 4173–4183.
9. Kotani, S., Tanaka, H., Yasuda, H., and Todokoro, K. (1999).
Regulation of APC activity by phosphorylation and regulatory
factors. J. Cell Biol. 146, 791–800.
10. Wassmann, K., and Benezra, R. (1998). Mad2 transiently asso-
ciates with an APC/p55Cdc complex during mitosis. Proc. Natl.
Acad. Sci. USA 95, 11193–11198.
11. Sudakin, V., Chan, G.K., and Yen, T.J. (2001). Checkpoint inhibi-
tion of the APC/C in HeLa cells is mediated by a complex of
BUBR1, BUB3, CDC20, and MAD2. J. Cell Biol. 154, 925–936.
12. Sudakin, V., and Yen, T.J. (2004). Purification of the mitotic
checkpoint complex, an inhibitor of the APC/C from HeLa cells.
Methods Mol. Biol. 281, 199–212.
13. Gorbsky, G.J., and Ricketts, W.A. (1993). Differential expression
Plk1 Creates the Tension-Sensing 3F3/2 Epitope
32. Chen, R.H., Waters, J.C., Salmon, E.D., and Murray, A.W.
(1996). Association of spindle assembly checkpoint compo-
nent XMAD2 with unattached kinetochores. Science 274, 242–
33. Jablonski, S.A., Chan, G.K., Cooke, C.A., Earnshaw, W.C., and
Yen, T.J. (1998). The hBUB1 and hBUBR1 kinases sequentially
assemble onto kinetochores during prophase with hBUBR1
concentrating at the kinetochore plates in mitosis. Chro-
mosoma 107, 386–396.
34. Logarinho, E., Bousbaa, H., Dias, J.M., Lopes, C., Amorim, I.,
Antunes-Martins, A., and Sunkel, C.E. (2004). Different spindle
checkpoint proteins monitor microtubule attachment and ten-
sion at kinetochores in Drosophila cells. J. Cell Sci. 117,
35. Skoufias, D.A., Andreassen, P.R., Lacroix, F.B., Wilson, L., and
Margolis, R.L. (2001). Mammalian mad2 and bub1/bubR1 re-
cognize distinct spindle-attachment and kinetochore-tension
checkpoints. Proc. Natl. Acad. Sci. USA 98, 4492–4497.
36. Carvalho, A., Carmena, M., Sambade, C., Earnshaw, W.C., and
Wheatley, S.P. (2003). Survivin is required for stable checkpoint
activation in taxol-treated HeLa cells. J. Cell Sci. 116, 2987–
37. Campbell, M.S., and Gorbsky, G.J. (1995). Microinjection of mi-
totic cells with the 3F3/2 anti-phosphoepitope antibody delays
the onset of anaphase. J. Cell Biol. 129, 1195–1204.
38. DeLuca, J.G., Dong, Y., Hergert, P., Strauss, J., Hickey, J.M.,
Salmon, E.D., and McEwen, B.F. (2005). Hec1 and nuf2 are core
components of the kinetochore outer plate essential for organ-
izing microtubule attachment sites. Mol. Biol. Cell 16, 519–531.
39. Spankuch-Schmitt, B., Bereiter-Hahn, J., Kaufmann, M., and
Strebhardt, K. (2002). Effect of RNA silencing of polo-like ki-
nase-1 (PLK1) on apoptosis and spindle formation in human
cancer cells. J. Natl. Cancer Inst. 94, 1863–1877.
40. Sumara, I., Gimenez-Abian, J.F., Gerlich, D., Hirota, T., Kraft,
C., De La Torre, C., Ellenberg, J., and Peters, J.M. (2004). Roles
of polo-like kinase 1 in the assembly of functional mitotic spin-
dles. Curr. Biol. 14, 1712–1722.
41. DeLuca, J.G., Howell, B.J., Canman, J.C., Hickey, J.M., Fang,
G., and Salmon, E.D. (2003). Nuf2 and Hec1 are required for
retention of the checkpoint proteins Mad1 and Mad2 to kinet-
ochores. Curr. Biol. 13, 2103–2109.
42. Martin-Lluesma, S., Stucke, V.M., and Nigg, E.A. (2002). Role
of Hec1 in spindle checkpoint signaling and kinetochore re-
cruitment of Mad1/Mad2. Science 297, 2267–2270.
43. Meraldi, P., Draviam, V.M., and Sorger, P.K. (2004). Timing and
checkpoints in the regulation of mitotic progression. Dev. Cell
44. Waters, J.C., Chen, R.H., Murray, A.W., Gorbsky, G.J., Salmon,
E.D., and Nicklas, R.B. (1999). Mad2 binding by phosphory-
lated kinetochores links error detection and checkpoint action
in mitosis. Curr. Biol. 9, 649–652.
45. Kallio, M., Weinstein, J., Daum, J.R., Burke, D.J., and Gorbsky,
G.J. (1998). Mammalian p55CDC mediates association of the
spindle checkpoint protein Mad2 with the cyclosome/ana-
phase-promoting complex, and is involved in regulating ana-
phase onset and late mitotic events. J. Cell Biol. 141, 1393–
46. Howell, B.J., Moree, B., Farrar, E.M., Stewart, S., Fang, G., and
Salmon, E.D. (2004). Spindle checkpoint protein dynamics at
kinetochores in living cells. Curr. Biol. 14, 953–964.
47. Tang, Z., Shu, H., Oncel, D., Chen, S., and Yu, H. (2004). Phos-
phorylation of Cdc20 by Bub1 provides a catalytic mechanism
for APC/C inhibition by the spindle checkpoint. Mol. Cell 16,
48. Kallio, M.J., Beardmore, V.A., Weinstein, J., and Gorbsky, G.J.
(2002). Rapid microtubule-independent dynamics of Cdc20 at
kinetochores and centrosomes in mammalian cells. J. Cell Biol.
49. Stukenberg, P.T., Lustig, K.D., McGarry, T.J., King, R.W., Kuang,
J., and Kirschner, M.W. (1997). Systematic identification of mi-
totic phosphoproteins. Curr. Biol. 7, 338–348.
50. Bolton, M.A., Lan, W., Powers, S.E., McCleland, M.L., Kuang,
J., and Stukenberg, P.T. (2002). Aurora B kinase exists in a
complex with survivin and INCENP and its kinase activity is
stimulated by survivin binding and phosphorylation. Mol. Biol.
Cell 13, 3064–3077.
51. Murray, A.W. (1991). Cell cycle extracts. Methods Cell Biol. 36,