Volume 3
†
Number 1
†
March 2010
10.1093/biohorizons/hzq010
Advance Access publication 13 February 2010
.........................................................................................................................................................................................................................................
Research article
Protein phosphatase 2A contributes to separase
regulation and the co-ordination of anaphase
Christopher P. Wardlaw*
University of Manchester, Manchester, UK.
* Corresponding author: Genome Damage and Stability Centre, Science Park Road, University of Sussex, Falmer, Brighton, East Sussex BN1 9RQ, UK.
Tel: þ44 01273873118. Email: [email protected]
Supervisor: Dr Stephen Taylor, Faculty of Life Science, University of Manchester, Michael Smith Buliding, Oxford Road, Manchester, M13 9PT, UK.
........................................................................................................................................................................................................................................
This study explores the role of the phosphatase Pp2A in regulation of anaphase onset in human cells. During the mitotic cell cycle, cells
replicate their DNA in S-phase giving sister chromatids. These chromatids remain tethered together by the cohesin ring until anaphase.
The onset of anaphase is triggered by the activation of separase, a protease which cleaves the cohesin ring structure, thereby allowing
the sister chromatids to be pulled to opposite ends of the spindle. Prior to anaphase, separase is held in check by one of two inhibitors,
namely securin or cyclin B1. Recently, it has been shown that securin-bound separase also binds the protein phosphatase, Pp2A.
Importantly, the binding of Pp2A is regulated by separase autocleavage; upon activation, separase autocleaves and releases Pp2A.
Strikingly, expression of a non-cleavable separase induces premature sister chromatid separation. Here, we show that the ability of
non-cleavable separase to prematurely induce chromatid disjunction requires its catalytic activity. These data lend weight to a handover
model whereby separase is initially inhibited by securin; then as securin is degraded, separase autocleaves, Pp2A is released thereby
allowing cyclin B1 binding; this in turn maintains separase inhibition until cyclin B1 is degraded. One exciting extension of this
model is that the release of Pp2A provides a burst of phosphatase activity just prior to chromatid separation, perhaps to ‘forewarn’
the cell that anaphase onset is imminent. For example, Pp2A activation may ensure that kinetochore – microtubule interactions are stabilized to ensure that all the chromatids are locked onto their K-fibres at the point when sister chromatid cohesion is lost. This study has
important implications in understanding how defects in separase regulation can lead to aneuploidy and diseases such as cancer.
Key words: prophase, anaphase, chromatid, cohesin, separase, Pp2A.
Submitted on 19 October 2009; accepted on 29 January 2010
........................................................................................................................................................................................................................................
Introduction
Somatic cells divide by a process known as mitosis. The
mitotic cell cycle consists of a gap phase, G1; DNA synthesis,
S-phase; a second gap phase, G2; and nuclear and cytoplasmic division, M-phase. For cells to accurately divide
and segregate their DNA, one copy of each replicated
DNA molecule has to ultimately reside in each daughter
cell. These duplicate DNA molecules, known as sister chromatids, therefore have to be kept together from S-phase
until the metaphase –anaphase transition in M-phase,
where they segregate and one sister moves to each spindle
pole prior to cytokinesis. Failure to keep sister chromatids
tethered together can lead to premature segregation and
aneuploidy in the resulting cells. Aneuploidy in the embryo
leads to birth defects or inviability, in the adult, it leads to
malignant cancers.1
The protein complex which is responsible for tethering
duplicate DNA molecules together and preventing premature
segregation is cohesin.2 Cohesin is loaded onto the DNA
during S-phase along the entire length of the chromosomes.
It most likely encircles the sister chromatids tethering them
together.3 – 6 In higher eukaryotes, the majority of the
cohesin molecules along the chromatid arms are removed
via the prophase pathway at the beginning of M-phase.7,8
During this process, the Ssc3 subunit of the cohesin
complex is phosphorylated by the Polo (Plk1) and Aurora
B kinases which then leads to the dissociation of the
cohesin from the DNA.8 – 10 The centromeric cohesin is protected from the prophase pathway by Shugoshin (Sgo1).
.........................................................................................................................................................................................................................................
#
The Author 2010. Published by Oxford University Press. This is an Open Access article distributed under the terms of the Creative Commons
Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/2.5), which permits unrestricted non-commercial use, distri66
bution, and reproduction in any medium, provided the original work is properly cited.
Research article
Bioscience Horizons
† Volume 3 † Number 1 † March 2010
.........................................................................................................................................................................................................................................
Sgo1 localizes to the centromeric cohesin via an interaction
with centromere-associated Pp2A phosphatase. The centromeric cohesin keeps the sister chromatids tethered together
until the activation of separase at the metaphase–anaphase
transition.11,12 Separase is a cysteine protease which cleaves
the kleisin subunit of cohesin, thus leading to the complete
loss of cohesion and the separation of the sister chromatids.13 – 19 Separase activity is regulated via inhibition by
securin or cyclin B1. Securin binds to both the N- and
C-termini of separase preventing access to the separase
active site.13,20,21 Cyclin B1 binding and inhibition of separase first needs separase to be phosphorylated at two distinct
sites. First, separase is phosphorylated by cyclin-dependent
kinase 1 (Cdk1) of the cyclin B1 –Cdk1 complex on
S1126. This then allows the recruitment of Plk1 which phosphorylates separase on S1399. Cyclin B1 can then bind the
phosphorylated S1339 and inhibit separase.22 – 24 Both
securin and cyclin B1 are targets of the anaphase promoting
complex/cyclosome (APC/C), a ubiquitin ligase whose
activity is controlled by the spindle checkpoint. The spindle
checkpoint is active and prevents APC/C activation until
all the chromosomes are aligned in a bi-orient fashion in
metaphase. Once this has occurred, APC/C is activated targeting securin and cyclin B1 for degradation, thus causing
relief of separase inhibition and anaphase onset.25,26
Recently, the relative importance of the two separase
inhibitors securin and cyclin B1 was addressed in human
tissue culture.22 Overexpression of separase titrates out its
securin inhibition as securin levels cannot exceed a particular
threshold regardless of the level of separase. In this situation,
cyclin B1 is thought to compensate for the lack of securin
inhibition of separase and cells progress through mitosis normally. If a separase allele that harbours an S:A mutation at
S1126 is overexpressed, thus preventing both mechanisms
of inhibition, the sister chromatids separate 5 min prematurely.22 This suggested that cyclin B1 is only required for
separase inhibition immediately before anaphase onset,
leading to a handover model in which inhibition of separase
is handed over from securin to cyclin B1 just prior to the
metaphase to anaphase transition.22 A third binding
partner of mammalian separase has also been identified.
The phosphatase Pp2A has been shown to bind inactive,
non-cleaved, separase within its regulatory region (amino
acids 1278–1556) but not inhibit it.27 Separase undergoes
an autocleavage event after the relief of its inhibition; this
autocleavage causes the binding of Pp2A to be disrupted
and Pp2A to dissociate from separase. Overexpression of
separase with mutated autocleavage sites (separase noncleavable, NC) prevents autocleavage and Pp2A dissociation,
causing sister chromatids to separate prematurely in a similar
way to separase S:A. It was then shown to be the increased
binding of Pp2A to separase NC causing the premature
sister chromatid separation, not the inhibition of separase
autocleavage per se.27
There are a number of possible explanations for the phenotype of premature sister chromatid separation associated
with increased binding of Pp2A to separase NC. The first
is that the binding of Pp2A to separase disrupts the separase
cyclin B1 –Cdk1 interaction by preventing cyclin B1 –Cdk1
from accessing its binding site S1339. The binding sites on
separase for Pp2A and cyclin B1 are close, with the cyclin
B1 binding site being within the regulatory region of separase
also bound by Pp2A. Prevention of separase autocleavage
leads to Pp2A being constitutively bound to separase, and
this may prevent cyclin B1 binding, resulting in an uninhibited separase as with the S:A mutant.27,28 Pp2A bound to
separase has been shown to be catalytically active and a
reduction in S1126 phosphorylation has been seen in separase NC compared with wild-type (WT).27 This means that
Pp2A maybe preventing cyclin B1 binding by removing the
phosphorylation at S1126.23,27 Another possible explanation
of how the increased binding of Pp2A to separase may cause
premature sister separation involves the activity of the prophase pathway. The overexpression of separase NC may be
titrating the Pp2A away from the centromere and preventing
Sgo1 localization, thus causing loss of cohesion via the prophase pathway.27 It has recently been shown that Sgo1 is
present and active at the mitotic centromere until the separation of chromatids at anaphase onset.29 Also levels of Plk1
and Aurora B are still high in metaphase, so the prophase
pathway of cohesion loss is still potentially active.30
This study aimed to distinguish between these two possibilities. We reasoned that if the separase N:C mutant
induced premature disjunction by titrating Sgo1 away from
centromeres, separase NC need not be catalytically active.
In contrast, if separase autocleavage is required to facilitate
cyclin-B1-mediated inhibition, then the premature
disjunction should require separase NC to be catalytically
active. Therefore, we generated a catalytically inactive, noncleavable mutant by substituting cysteine 2029 with an
alanine in the separase NC transgene. Using flow cytometry,
time-lapse microscopy and metaphase spreads, we show
that the separase NC C:A double mutant does not induce
premature chromatid disjunction.
Materials and Methods
DNA manipulations
The separase NC plasmid used was a pcDNA/FRT/TO/
Myc/Separase-NC plasmid.27 To render separase NC, catalytically inactive C2029 was mutated to alanine (described
later) using Quick change (Stratagene) site-directed mutagenesis as previously described.22 Plasmid DNA was subjected
to a sequencing PCR reaction using the BigDye Terminator
Sequencing Kit (Applied Biosystems) according to the manufacturer’s guidelines. The amplified DNA was then ethanol
precipitated using a standard ethanol NaAC protocol
.........................................................................................................................................................................................................................................
67
Research article
Bioscience Horizons † Volume 3 † Number 1 † March 2010
.........................................................................................................................................................................................................................................
(0.3 M NaAC pH 4.8, 1 ml glycogen blue 1 mg/ml) and
stored at 2208C until being sequenced by the Stopford
sequencing service.
PBS) for 30 min before being analysed using a CyAn flow
cytometer (DakoCytomation) driven by Summit software
(DakoCytomation).
Cell culture and generation of cell lines
Time-lapse microscopy
HEK Flp-In T-REx-293 (Lac Zeo TO) cells (Invitrogen) were
used for this study.31 These cells were cultured in Dulbecco’s
modified Eagle’s media (DMEM) with 10% foetal calf serum
(FSC), 2 mM L-glutamate, 100 mg/ml streptomycin and
100 U/ml penicillin (all Invitrogen). They were cultured in
a 378C, 5%, CO2 humidity-controlled incubator. This parental cell line was selected for using 200 mg/ml Zeocin
(Invitrogen) and 15 mg/ml Blasticidin (Roche). Stable cell
lines were created using a standard Lipofectamine-Plus transfection protocol with the separase transgenes being integrated via FRT/Flp-mediated recombination (Invitrogen).31
Transfected cells were selected for by addition of 15 mg/ml
blasticidin and 150 ml/ml hygromycin (both Roche). The
expression of separase transgenes was induced by 1 mg/ml
tetracycline (Sigma).
Mitotic progression was analysed via time-lapse microscopy
on cells seeded into an 8-well cover glass chamber slide
(Nunc) and transiently transfected with GFP-histone H2B.
Transient transfection of GFP-histone H2B was carried out
using a pcDNA-5 based expression vector (Invitrogen) containing the GFP fusion protein.33 This was transfected
using Lipofectamine. Six hundred nanograms of GFP-H2B
and 600 ng of Myc DNA giving a total of 1.2 mg of DNA
were transfected. The Myc DNA was used merely to increase
the concentration of transfected DNA to the required 1.2 mg.
Transfected cells were induced with tetracycline for 6 h and
viewed using a Zeiss Axiovert 200 microscope equipped with
an environmental chamber which maintained 378C and 5%
CO2.33 Images were taken every 2 min at five positions per
well using a Cool Snap HQ CCD camera (Photometrics)
driven by Metamorph software (Universal imaging) over a
24 h period.
Immunoblotting
Soluble cell proteins were collected and suspended in 6
SDS buffer (7% stacking buffer (0.5 M Tris pH 6.8),
1 mM SDS, 0.93 mM DTT, 3% glycerol, 0.05 mM
Bromophenol blue) for western blotting. In summary, 10 ml
of the cell lysates was loaded into a 3–8% gradient SDS
Tris –acetate gel (Invitrogen). They were run in Tris –
acetate SDS running buffer (Invitrogen) for 2 h at 200 V.
The proteins were then transferred to a nitrocellulose membrane (Whatman) via wet transfer in high molecular weight
transfer buffer (50 mM Tris, 380 mM glycine, 0.2% (w/v)
SDS, 5% methanol) at 25 V for 16 h at 48C. The membrane
was then stained with ponceau red, washed in TBS þ Tween
(TBST) and blocked in 5% milk (Marvel) TBST for 1 h. The
membrane was then probed using the following primary
antibodies in 5% milk TBST: 4A6 (mouse a-myc, Upstate,
1:1000), XJ11-1B12 (mouse a-separase, AbCam, 1:1000)
and SBR1.1 (sheep a-BubR132 1:1000). The secondary antibodies used were Hrp goat a-mouse and Hrp rabbit a-sheep
(both from Zymed at 1:2000 in 5% milk TBST). Following
washes in TBST, the membrane was incubated with equal
volumes of supersignal west Pico reagents (Perbio) and
exposed to Biomax MR autoradiography film (Kodak) for
between 1 and 5 min.32
Flow cytometry
DNA content was measured using flow cytometry. Cells were
induced with tetracycline (Sigma) for 24 h, washed in PBS
and collected in ice-cold PBS. To a final concentration of
70%, 100% ice-cold ethanol was added in a drop-wise
manner, and the cells were stored at 2208C for 24 h. They
were then washed in PBS and incubated in propidium
iodide (PI) solution (40 mg/ml PI, 50 mg/ml RNase A in
Metaphase spreads
Metaphase spreads were created 24 h post-induction with
tetracycline (Sigma) and 2 h treatment with 0.2 mg/ml nocodazole (Sigma). Cells were collected, resuspended and incubated in 0.45% hypotonic buffer (32 mM KCl, 0.5 mM
EDTA, 16 mM HEPES, pH 7.4). The cells were then pelleted, resuspended in ice-cold methanol:acetic acid (3:1)
and stored at 2208C for 24 h. These fixed cells were then
dropped onto acetic acid covered glass slides and allowed
to dry. They were then stained with 1 mg/ml Hoechst
(Sigma) in PBS for 2 min, washed in PBS, mounted and visualized on a Zeiss Axioskop 2 microscope with a CoolSnap
HQ CCD camera (Photometrics) driven by Metamorph
software (Universal imaging).
Results
A separase NC catalytically inactive mutant was created to
further understand the premature chromatid separation phenotype caused by the overexpression of separase NC, and the
role of Pp2A in the metaphase to anaphase transition.
Site-directed mutagenesis was carried out on the plasmid
pcDNA5/FRT/TO/Myc/SeparaseNC.27 The mutagenesis
aimed to mutate a cysteine residue (C2029) within the catalytic domain of separase, which is essential for catalytic
activity (Fig. 1A). To mutate C2029, two primers used in a
previous study to generate a catalytically inactive separase
were utilized.27 These primers mutate base pairs 6082 and
6083 from TG to GC, therefore changing cysteine 2029 to
an alanine (Fig. 1). To ensure that the mutagenesis had
worked successfully, the plasmid DNA was subjected to
.........................................................................................................................................................................................................................................
68
Bioscience Horizons
Research article
† Volume 3 † Number 1 † March 2010
.........................................................................................................................................................................................................................................
Figure 1. Creation of separase NC C:A. (A) Diagram showing the domains and important residues within separase. The blue represents the series of ARM
repeats; the red shows the regulatory region which contains the cyclin B1 and Pp2A binding sites, and the autocleavage sites (A/C). The orange represents
the inactive fold and the yellow the active fold. C2029 is within this active fold, at the C-terminus, within the catalytic site. Base pair and amino acid
sequence of the catalytic site are shown with C2029 highlighted in red.56 (B) The sequencing data in the form of four-peak graphs. On the left separase
NC without the C:A mutation, on the right separase NC C:A. Mutation from TGT to GCT can be seen, thus changing the amino acid sequence from C to A
(highlighted in red).
diagnostic restriction digest (data not shown) and sequencing
reactions (Fig. 1). It can be seen in Fig. 1B that the mutagenesis was successful and cysteine 2029 had been mutated to an
alanine, thus likely creating separase NC C:A.
Generation of separase NC and NC C:A stable cell lines
The separase NC and NC C:A mutant genes were integrated
at a pre-existing FRT site within the 293 cells genome under
the control of a tetracycline-inducible CMV promoter. A
western blot was carried out to ensure that the separase
mutants were expressed strongly in response to tetracycline
(Fig. 2). Separase is a 230 kDa protein, so a strong band of
this size representing the mutant separase alleles should be
apparent in the tetracycline induced but not the uninduced
lanes.
It can be seen from Fig. 2 that both the Myc-tagged separase NC and NC:CA are expressed upon tetracycline induction. It can also be seen that there is no expression of the
mutant alleles in the absence of tetracycline, but endogenous
separase is still expressed as expected at low, WT, levels.
Separase NC C:A lacks the G2/M phenotype seen
in separase NC
It was previously shown that cells expressing separase NC
accumulate DNA contents equal to or greater than 4N.27
Figure 2. Creation of stable inducible separase NC and NC C:A cell lines.
Cells were induced to express the mutant separase alleles for 24 h, 10 ml
of cell lysates was run on a SDS gel and immune-blotted. In the top
panel, anti-Myc antibody which only binds the exogenous Myc-tagged
mutant separase alleles was used. The middle panel shows anti-separase
antibody, which binds both endogenous and exogenous separase. The
bottom panel is a loading control using anti-BubR1 anitbody.
This therefore suggested that overexpression of separase
NC caused a G2/M defect which is most likely due to premature sister chromatid separation. To test if overexpression
of separase NC C:A caused a phenotype similar to that of
separase NC, flow cytometry was carried out. It was performed on tetracycline induced and uninduced separase
NC and NC C:A. As a control, a tetracycline-inducible
separase WT 293 cell line was also used (Fig. 3).22
.........................................................................................................................................................................................................................................
69
Research article
Bioscience Horizons † Volume 3 † Number 1 † March 2010
.........................................................................................................................................................................................................................................
Figure 3. Overexpression of separase NC C:A does not cause a G2/M defect. (A) Facs profiles of asynchronous uninduced separase WT, NC and NC C:A
cells. Cells were induced with tetracycline for 24 h and processed by flow cytometry. (B) Graph showing approximate percentage of cells with a 4Nþ
content of DNA. Data acquired using summit software (Dakocytomation). Note the increase in the percentage of cells with a 4Nþ content seen for separase NC but not NC C:A.
As expected, an increase in cells with a 4Nþ content can
be seen for induced separase NC in Fig. 3, which is in line
with previous reports.27 Induction of WT separase shows a
slight increase in the percentage of cells with a 4Nþ
content, with 31% of cells accumulating a 4Nþ content
compared with 24% in the uninduced cells. However,
there is no increase in the percentage of cells with a 4Nþ
content with induction of separase NC C:A, 27% of the
induced and uninduced populations have a 4Nþ DNA
content (Fig. 3). This would suggest that overexpression of
separase NC C:A may not be causing the same G2/M
defect as overexpression of separase NC. To confirm that
there is in fact no G2/M defect in cells overexpressing separase NC C:A, further analysis was carried out in the form of
time-lapse microscopy.
Separase NC C:A exhibits WT mitotic progression
It was reported that separase NC has an increased length of
mitosis and undergoes repeated rounds of chromosome
search and capture/alignment due to the premature loss of
cohesion.27 To see if this can also be observed for separase
NC C:A, mitotic progression time-lapse microscopy was performed on separase WT, NC and NC C:A. The time spent in
mitosis was then analysed by measuring the time from
nuclear envelope breakdown to decondensation of the
chromosomes (Fig. 4).
From Fig. 4A, it can be seen that cells overexpressing WT
separase took 36 min to progress from nuclear envelope
breakdown to chromosome decondensation. In contrast,
the separase NC induced cell takes 202 min before decondensation of its chromosomes. The cell had already broken
down its nuclear envelope, when the imaging had started;
so, this is an underestimate of the time taken in mitosis.
The cell aligns its chromosomes at the 20 min point, then
the chromosomes separate, they start to realign again by
86 min, separate again, then realign at 190 min, the chromosomes then decondense without separation and mitotic exit.
These repeated rounds of search and capture seen for separase NC are in line with that already reported.27 It is worth
noting that not all the separase NC cells decondensed their
chromosomes without proper division, many of them
simply died or underwent an aberrant division without first
equally separating their chromosomes (data not shown).
The induced separase NC C:A cell breaks down its nuclear
envelope, aligns its chromosomes on the metaphase plate
by 26 min and separates its chromosomes in an orderly
fashion, before decondensation occurs at 38 min. The separase NC C:A cell does not go through the repeated rounds of
search and capture seen in the separase NC cell and
.........................................................................................................................................................................................................................................
70
Bioscience Horizons
Research article
† Volume 3 † Number 1 † March 2010
.........................................................................................................................................................................................................................................
Figure 4. Separase NC C:A has no defect in mitotic progression. (A) Cells were transiently transfected with GFP Histone H2B, induced with tetracycline for
6 h and visualized by time-lapse microscopy. Numbers represent time in minutes. For WT and NC C:A, 0 represents nuclear envelope breakdown and end
of mitosis is classed as chromosome decondensation. For NC, 0 min is just after nuclear envelope breakdown. (B) Box and whisker chart showing time
spent in mitosis. Start of mitosis is defined as nuclear envelope breakdown and the end by decondensation of chromosomes. Between 8 and 16 cells
were counted for each cell line. The box represents the inter-quartile range and the whiskers the range. The median is shown within the box. The
P-values show the significance using the Kruskal –Wallis test (non-parametric ANOVA).
progresses through mitosis in a WT-like fashion. The average
time spent in mitosis shows that the tetracycline-induced
separase NC cells spend a significantly longer time in
mitosis than both the induced WT, induced NC C:A, and
uninduced cells (Fig. 4B). As expected, there was some variation in the time spent in mitosis within each cell line, but the
median time spent in mitosis by separase NC was over three
times that of WT and NC C:A (Fig. 4B).34 Overall, 86% of
induced separase NC cells spent over 100 min in mitosis
compared with only 13% of induced WT, 6% of induced
NC C:A and 0% of uninduced WT cells. This suggests that
induction of separase NC C:A has no significant effect on
the length of, or progression through, mitosis. However,
this may not necessarily mean that the separase NC C:A
does not separate its chromatids in a premature or aberrant
manner. It may be causing a chromatid separation defect
that does not activate the spindle checkpoint, therefore not
causing the repeated rounds of search and capture seen in
the induced separase NC cells. Even though this was
unlikely, further analysis of chromatid separation was
carried out.
Separase NC C:A cells do not show premature
segregation of sister chromatids
Metaphase spreads were carried out to identify the number
of separated chromatids in metaphase for separase NC C:A
compared with separase NC and WT. This is a good indicator of premature segregation and was performed to help
confirm the data shown in Figs 3 and 4. Cells were
induced with tetracycline for 24 h, then treated with nocodazole for 2 h to break down the mitotic spindle, therefore preventing transition into anaphase. The cells were then fixed
and the DNA was stained. Example pictures of metaphase
.........................................................................................................................................................................................................................................
71
Research article
Bioscience Horizons † Volume 3 † Number 1 † March 2010
.........................................................................................................................................................................................................................................
Figure 5. Separase NC C:A does not separate its chromatids prematurely. Cells were induced for 24 h with tetracycline and treated with nocodazole for 2 h.
They were then prepared for metaphase spreads. Magnification of chromatids from the spread is shown below the main images (A) WT separase, (B) NC
separase, and (C) NC C:A separase. (D) Between 150 and 300 chromatids were counted from a number of images within the same experiment and the
percentage of separated chromatids was calculated. Note the increased number of separated chromatids upon induction of separase NC but not NC C:A.
spreads can be seen in Fig. 5A –C for separase WT, NC and
NC C:A, respectively.
It can be seen in Fig. 5A that the majority of chromatids
remain paired in WT separase cells in both the presence
and absence of tetracycline. This suggests, at these levels of
expression, WT separase does not cause premature segregation of sister chromatids. The magnification of chromatids
shown below the spread of Fig. 5A shows tight pairing of
chromatids, especially at the centromeric regions which is
consistent with intact cohesin.2 Figure 5B shows that when
induced separase NC cells have a dramatic increase in the
number of unpaired chromatids, shown by the scattered
appearance of the chromatids in the spread. In Fig. 5C, it
can be seen that the separase NC phenotype is not present
for induced separase NC C:A. Both the tetracycline
induced and uninduced NC C:A cells’ chromatids remain
tightly paired in metaphase, as seen for WT. This is confirmed in Fig. 5D which shows the percentage of separated
chromatids. It shows that 66% of the chromatids from
induced separase NC cells prematurely segregate in metaphase compared with ,10% for induced WT and NC
C:A. The data in Fig. 5 suggest that overexpression of
separase NC C:A does not cause premature separation of
sister chromatids.
Discussion
The aim of this study was to further characterize the phenotype of premature sister chromatid segregation caused by the
overexpression of separase NC.27 This was done in order to
shed light on the role of Pp2A binding to mammalian separase and the regulation of this binding by separase
autocleavage. To test which of the two hypothesis surrounding this premature segregation phenotype was most likely,
disruption of separase inhibition by cyclin B1 or activation
of the prophase pathway, C2029 of separase NC was
mutated to alanine, rendering it catalytically inactive.
Catalytic activity of separase is required for premature
sister chromatid separation in separase NC
The results presented in Figs 3– 5 confirm previous reports
that prevention of separase autocleavage leads to an increase
in cells with a 4Nþ DNA content due to premature segregation of sister chromatids, leading to repeated rounds of
.........................................................................................................................................................................................................................................
72
Bioscience Horizons
Research article
† Volume 3 † Number 1 † March 2010
.........................................................................................................................................................................................................................................
search and capture and thus mitotic delay.27 Eventual breakdown of cyclin B1 most likely occurs, thus causing the chromatids to decondense and exit mitosis without proper
division. Some of these cells may then progress to the next
G1 as tetraploids, therefore contributing, along with the
mitotic delay, to the increased 4Nþ DNA content seen in
Fig. 3. Some other separase NC cells simply died while
others underwent an aberrant division, attempting to separate unaligned and decondensing chromatids, also likely
causing aneuploidy in the daughter cells.
A separase NC phenotype cannot be seen for separase NC
C:A. As described above, the percentage of cells with a 4Nþ
DNA content is almost identical for induced and uninduced
cells (Fig. 3). The progression through mitosis for separase
NC C:A in Fig. 4 is almost identical to WT. They also do
not show an increased level of sister chromatid separation
in metaphase (Fig. 5C and D). The overexpression of separase NC C:A showed the same phenotype as separase C:A.22
For example, separase C:A in the 2006 study spent approximately the same time in mitosis as the separase NC C:A in
this study.22 This further confirms that mutation of the cleavage sites does not have an effect when separase is catalytically inactive. The fact overexpression of WT separase
caused a stronger phenotype than separase NC C:A adds
weight to the notion that the catalytic activity of separase
is required for premature sister chromatid separation
(Figs 3 and 5).35
Pp2A may disrupt cyclin B1 inhibition of separase
The lack of premature chromatid separation in separase NC
C:A along with the phenotypes described for the other separase mutants suggests that separase needs to be catalytically
active and uninhibited for premature separation to
occur.22,27 Therefore, it is likely that the premature segregation in separase NC cells is caused by Pp2A affecting
cyclin B1 inhibition of separase. Thus meaning separase
NC is unlikely to be titrating Pp2A away from the centromere and causing loss of cohesion by the activity of a
prophase-like pathway. It is most probable that Pp2A is preventing cyclin B1 binding to separase rather than displacing
already bound separase. This is because previous studies
showed that Pp2A binds separase, along with securin, in
interphase. Whereas, cyclin B1 cannot bind to separase
until mitosis, as it requires Cdk1 to be active for S1126 phosphorylation.22 – 24,36 This suggests Pp2A binds to separase
first and thus prevents cyclin B1 binding and inhibiting
separase. From this study, it is impossible to tell if Pp2A prevents cyclin B1 binding by removing the phosphate from
S1126 or by excluding cyclin B1 from its binding site on
separase. It is most probably a combination of both
methods, with Pp2A removing the phosphate group from
S1126, while also making the kinetics of cyclin B1 binding
to its site much less likely even in the presence of
phosphorylation.
Pp2A may mediate the mutually exclusive
binding of securin and cyclin B1 to separase
By acting to prevent cyclin B1 binding to separase, Pp2A
maybe functioning under normal cellular conditions as an
integral regulator of cyclin B1 –Cdk1 localization in mitosis
and the handover of separase inhibition from securin to
cyclin B1.24,28,37 By binding securin-inhibited separase in
interphase, Pp2A prevents cyclin B1 binding to separase in
early mitosis, thus leaving cyclin B1 –Cdk1 free to act upon
its many other targets (Fig. 6).22,36 When the chromatids
are aligned on the metaphase plate, APC/C is activated by
relief of the spindle checkpoint causing the degradation of
securin and cyclin B1.25,26 There may be a higher level of
cyclin B1 than securin in the cell; in fact, securin has been
shown to be oncogenic at high levels so is tightly regulated,
thus meaning the majority of securin maybe degraded
before cyclin B1.21,38 Furthermore, in Saccharomyces
cerevisiae, the cyclin B1 homologue Clb2 is degraded in
two phases: first at anaphase onset and then again later in
mitosis.39,40 This suggests that there may still be cyclin B1
present after securin degradation at the metaphase–anaphase
transition. It has also been suggested that APC/C may form a
complex with some securin molecules in prometaphase.
Ubiquitination of securin can then occur rapidly when
APC/C is activated targeting these securin molecules for
Figure 6. Handover/prepare for anaphase model. Separase is normally inhibited by securin. Securin-bound separase also has Pp2A bound. Securin is
degraded, separase autocleaves and Pp2A is released. The released Pp2A goes to the kinetochore and dephosphorylates substrates increasing the stability
of the microtubule – kinetochore interaction in preparation for anaphase. Meanwhile, the loss of Pp2A from separase and the larger pool of cyclin B1 compared with securin leads to the phosphorylation of and the cyclin B1 inhibition of separase. This is handover of separase inhibition. Five minutes later cyclin
B1 is degraded and separase is active to cleave cohesin.28,37
.........................................................................................................................................................................................................................................
73
Research article
Bioscience Horizons † Volume 3 † Number 1 † March 2010
.........................................................................................................................................................................................................................................
degradation more quickly than cyclin B1.41 This securin
degradation leaves some separase molecules uninhibited
and free to autocleave which releases Pp2A. Cyclin B1 can
then bind and inhibit separase for the final 5 min before anaphase onset (Fig. 6). If separase does not autocleave, as in
separase NC, then Pp2A is not released and cyclin B1
cannot inhibit separase, causing cohesin cleavage to occur
5 min prematurely.27
Pp2A release may signal anaphase is about to occur
Figure 6 describes that the autocleavage and subsequent
release of Pp2A occurs 5 min before chromatid separation
to mediate the handover of separase inhibition from
securin to cyclin B1. This begs the question, is the release
of Pp2A from separase significant in any other aspects of
chromatid separation? It is possible that this sudden increase
in free Pp2A and the subsequent burst in phosphatase
activity maybe playing a role. This burst of phosphatase
activity maybe signalling to other cellular components that
anaphase is about to occur, as the release of Pp2A should
always occur 5 min prior to anaphase onset. There are a
number of different targets that Pp2A may act on. The first
possible target maybe the centrosomes, where Pp2A has
been shown to localize both in this study (data not shown)
and in previous studies.42 – 45 The dephosphorylation of centrosome components may lead to the elongation of the
spindle needed for anaphase to progress. The second possible
targets, and the ones that we favour, are kinetochore components and regulators (Fig. 6). Kinetochores only remain
stably bound to microtubules when under tension, this is
the process that underlines the search and capture model
for chromosome alignment and control of the spindle checkpoint.46 However, kinetochores are also stochastic and the
two kinetochores of a pair of chromatids act independently
of each other. This means that the kinetochores are constantly under varying amounts of tension due to the
dynamic polymerization and depolymerization of kinetochore microtubules.47 If a decrease in tension were to
occur just before anaphase, once the spindle checkpoint
had been satisfied, leading to the kinetochore detaching
from its microtubule, it would have disastrous consequences
for the cell, as the chromatid would not move to its pole.
Furthermore, a sudden decrease in tension occurs when
cohesion is lost, again a mechanism must be in place to
prevent kinetochore microtubule detachment at anaphase
onset.48 We therefore hypothesize that it is this release of
Pp2A from separase that acts as a signal telling the kinetochores to remain bound, even in the absence of tension, as
anaphase is about to occur. This would therefore change
the behaviour of the kinetochore –microtubule interaction
from a dynamic tension stabilized form of attachment, as
seen in prometaphase, to a fixed attachment. Pp2A could
do this by dephosphorylating, and therefore regulating, proteins involved in tension sensing and destabilization of
kinetochore binding, such as aurora B, Incenp or Cenp-E
(Fig. 6).49,50
Aurora B has previously been shown to be a target of
Pp2A, with Pp2A negatively regulating aurora B activity in
human cells.51 Also, aurora B has been shown to delocalize
from the kinetochore at anaphase onset, Pp2A could play a
role in this process.52 Furthermore, in Drosophila
embryos, Pp2A has been shown to play a role in
microtubule– chromosome attachment. A reduction in
Pp2A leads to a reduction in kinetochore –microtubule
attachment, thus suggesting Pp2A can stabilize the
kinetochore –microtubule binding.42 This all suggests that
Pp2A, once released from separase, may initiate a prewarning signal for anaphase that securin has been degraded.
This then leads to all kinetochores becoming firmly bound to
the microtubules in a tension-independent manner ready for
cohesion loss, thus giving a defined role for the separase
handover model, separase autocleavage and Pp2A release.
Understanding these pathways of cohesion loss is important for understanding cancer formation. Heterozygous
mutations in separase have shown to increase epithelial
cancer 8-fold.53 Also, overexpression of separase is seen in
over 60% of human breast tumours and in numerous other
cancer types.54,55
Acknowledgements
I would like to thank Dr Stephen Taylor and all the members
of the Taylor lab for their help and support during this
project, with particular thanks to Dr Anthony Tighe for
his supervision.
Funding
This project was funded by the Faculty of Life Sciences
University of Manchester and Cancer Research UK.
Author biography
Christopher Wardlaw studied molecular biology with industrial experience at the University of Manchester. He spent his
placement year at the National Institute for Medical
Research in London looking at novel targets of the DNA
damage checkpoint protein Mec 1. He is now undertaking
a PhD degree at the Genome Damage and Stability Center
in Sussex where he is researching into his main interest of
DNA damage checkpoints. He aspires to continue and
excel in this field.
References
1. Weaver BA, Cleveland DW (2005) Decoding the links between mitosis,
cancer, and chemotherapy: the mitotic checkpoint, adaptation, and cell
death. Cancer Cell 8: 7–12.
.........................................................................................................................................................................................................................................
74
Bioscience Horizons
Research article
† Volume 3 † Number 1 † March 2010
.........................................................................................................................................................................................................................................
2. Michaelis C, Ciosk R, Nasmyth K (1997) Cohesins: chromosomal proteins that
prevent premature separation of sister chromatids. Cell 91: 35 –45.
3. Haering CH, Lowe J, Hochwagen A et al. (2002) Molecular architecture of
SMC proteins and the yeast cohesin complex. Mol Cell 9: 773 –788.
4. Gruber S, Haering CH, Nasmyth K (2003) Chromosomal cohesin forms a ring.
Cell 112: 765 –777.
25. Taylor SS, Scott MI, Holland AJ (2004) The spindle checkpoint: a quality
control mechanism which ensures accurate chromosome segregation.
Chromosome Res 12: 599 –616.
26. Musacchio A, Salmon ED (2007) The spindle-assembly checkpoint in space
and time. Nat Rev Mol Cell Biol 8: 379 –393.
5. Ivanov D, Nasmyth K (2007) A physical assay for sister chromatid cohesion in
vitro. Mol Cell 27: 300 –310.
27. Holland AJ, Bottger F, Stemmann O et al. (2007) Protein phosphatase 2A and
separase form a complex regulated by separase autocleavage. J Biol Chem
282: 24623 –24632.
6. Haering CH, Farcas AM, Arumugam P et al. (2008) The cohesin ring concatenates sister DNA molecules. Nature 454: 297– 301.
28. Holland AJ, Taylor SS (2008) Many faces of separase regulation. SEB Exp Biol
Ser 59: 99 –112.
7. Losada A, Hirano M, Hirano T (2002) Cohesin release is required for sister
chromatid resolution, but not for condensin-mediated compaction, at the
onset of mitosis. Genes Dev 16: 3004 –3016.
29. Lee J, Kitajima TS, Tanno Y et al. (2008) Unified mode of centromeric protection
by shugoshin in mammalian oocytes and somatic cells. Nat Cell Biol 10: 42 –52.
8. Hauf S, Roitinger E, Koch B et al. (2005) Dissociation of cohesin from chromosome arms and loss of arm cohesion during early mitosis depends on phosphorylation of SA2. PLoS Biol 3: e69.
9. Sumara I, Vorlaufer E, Stukenberg PT et al. (2002) The dissociation of cohesin
from chromosomes in prophase is regulated by Polo-like kinase. Mol Cell 9:
515– 525.
30. Morgan DO (1999) Regulation of the APC and the exit from mitosis. Nat Cell
Biol 1: E47 – E53.
31. Tighe A, Johnson VL, Taylor SS (2004) Truncating APC mutations have dominant effects on proliferation, spindle checkpoint control, survival and
chromosome stability. J Cell Sci 117: 6339 –6353.
10. Gimenez-Abian JF, Sumara I, Hirota T et al. (2004) Regulation of sister chromatid cohesion between chromosome arms. Curr Biol 14: 1187 –1193.
32. Taylor SS, Hussein D, Wang Y et al. (2001) Kinetochore localisation and phosphorylation of the mitotic checkpoint components Bub1 and BubR1 are differentially regulated by spindle events in human cells. J Cell Sci 114:
4385 –4395.
11. Salic A, Waters JC, Mitchison TJ (2004) Vertebrate shugoshin links sister
centromere cohesion and kinetochore microtubule stability in mitosis. Cell
118: 567 –578.
33. Morrow CJ, Tighe A, Johnson VL et al. (2005) Bub1 and aurora B cooperate to
maintain BubR1-mediated inhibition of APC/CCdc20. J Cell Sci 118:
3639 –3652.
12. McGuinness BE, Hirota T, Kudo NR et al. (2005) Shugoshin prevents dissociation of cohesin from centromeres during mitosis in vertebrate cells.
PLoS Biol 3: e86.
34. Gascoigne KE, Taylor SS (2008) Cancer cells display profound intra- and
interline variation following prolonged exposure to antimitotic drugs.
Cancer Cell 14: 111 –122.
13. Ciosk R, Zachariae W, Michaelis C et al. (1998) An ESP1/PDS1 complex regulates loss of sister chromatid cohesion at the metaphase to anaphase transition in yeast. Cell 93: 1067 – 1076.
35. Zhang N, Ge G, Meyer R et al. (2008) Overexpression of separase induces
aneuploidy and mammary tumorigenesis. Proc Natl Acad Sci USA 105:
13033 –13038.
14. Uhlmann F, Lottspeich F, Nasmyth K (1999) Sister-chromatid separation at
anaphase onset is promoted by cleavage of the cohesin subunit Scc1.
Nature 400: 37– 42.
36. Stemmann O, Zou H, Gerber SA et al. (2001) Dual inhibition of sister chromatid separation at metaphase. Cell 107: 715 –726.
37. Queralt E, Uhlmann F (2005) More than a separase. Nat Cell Biol 7: 930 –932.
15. Uhlmann F, Wernic D, Poupart MA et al. (2000) Cleavage of cohesin by the
CD clan protease separin triggers anaphase in yeast. Cell 103: 375 –386.
38. Uhlmann F (2007) What is your assay for sister-chromatid cohesion? EMBO J
26: 4609– 4618.
16. Jager H, Herzig A, Lehner CF et al. (2001) Drosophila separase is required for
sister chromatid separation and binds to PIM and THR. Genes Dev 15:
2572 –2584.
39. Yeong FM, Lim HH, Padmashree CG et al. (2000) Exit from mitosis in budding
yeast: biphasic inactivation of the Cdc28-Clb2 mitotic kinase and the role of
Cdc20. Mol Cell 5: 501 –511.
17. Siomos MF, Badrinath A, Pasierbek P et al. (2001) Separase is required for
chromosome segregation during meiosis I in Caenorhabditis elegans. Curr
Biol 11: 1825– 1835.
40. Wasch R, Cross FR (2002) APC-dependent proteolysis of the mitotic cyclin
Clb2 is essential for mitotic exit. Nature 418: 556 –562.
18. Kumada K, Yao R, Kawaguchi T et al. (2006) The selective continued linkage
of centromeres from mitosis to interphase in the absence of mammalian
separase. J Cell Biol 172: 835 –846.
19. Wirth KG, Wutz G, Kudo NR et al. (2006) Separase: a universal trigger for
sister chromatid disjunction but not chromosome cycle progression. J Cell
Biol 172: 847– 860.
20. Funabiki H, Yamano H, Kumada K et al. (1996) Cut2 proteolysis required for
sister-chromatid seperation in fission yeast. Nature 381: 438– 441.
21. Zou H, McGarry TJ, Bernal T et al. (1999) Identification of a vertebrate sisterchromatid separation inhibitor involved in transformation and tumorigenesis. Science 285: 418 –422.
22. Holland AJ, Taylor SS (2006) Cyclin-B1-mediated inhibition of excess separase is required for timely chromosome disjunction. J Cell Sci 119:
3325 –3336.
23. Boos D, Kuffer C, Lenobel R et al. (2008) Phosphorylation-dependent binding
of cyclin B1 to a Cdc6-like domain of human separase. J Biol Chem 283:
816 –823.
24. Gorr IH, Boos D, Stemmann O (2005) Mutual inhibition of separase and Cdk1
by two-step complex formation. Mol Cell 19: 135– 141.
41. Jeganathan KB, Baker DJ, van Deursen JM (2006) Securin associates
with APCCdh1 in prometaphase but its destruction is delayed by
Rae1 and Nup98 until the metaphase/anaphase transition. Cell Cycle 5:
366– 370.
42. Snaith HA, Armstrong CG, Guo Y et al. (1996) Deficiency of protein phosphatase 2A uncouples the nuclear and centrosome cycles and prevents attachment of microtubules to the kinetochore in Drosophila microtubule star
(mts) embryos. J Cell Sci 109: 3001– 3012.
43. Murphy MB, Levi SK, Egelhoff TT (1999) Molecular characterization and
immunolocalization of Dictyostelium discoideum protein phosphatase 2A.
FEBS Lett 456: 7 –12.
44. Horn V, Thelu J, Garcia A et al. (2007) Functional interaction of Aurora-A and
PP2A during mitosis. Mol Biol Cell 18: 1233 –1241.
45. Schlaitz AL, Srayko M, Dammermann A et al. (2007) The C. elegans RSA
complex localizes protein phosphatase 2A to centrosomes and regulates
mitotic spindle assembly. Cell 128: 115 –127.
46. Zhou J, Yao J, Joshi HC (2002) Attachment and tension in the spindle assembly checkpoint. J Cell Sci 115: 3547 –3555.
47. Maiato H, DeLuca J, Salmon ED et al. (2004) The dynamic kinetochore–
microtubule interface. J Cell Sci 117: 5461 – 5477.
.........................................................................................................................................................................................................................................
75
Research article
Bioscience Horizons † Volume 3 † Number 1 † March 2010
.........................................................................................................................................................................................................................................
48. Palframan WJ, Meehl JB, Jaspersen SL et al. (2006) Anaphase inactivation of
the spindle checkpoint. Science 313: 680 –684.
49. Lens SM, Medema RH (2003) The survivin/Aurora B complex: its role in coordinating tension and attachment. Cell Cycle 2: 507 –510.
50. Biggins S, Walczak CE (2003) Captivating capture: how microtubules attach
to kinetochores. Curr Biol 13: R449 –R460.
51. Sugiyama K, Sugiura K, Hara T et al. (2002) Aurora-B associated protein phosphatases as negative regulators of kinase activation. Oncogene 21: 3103–3111.
52. Murata-Hori M, Wang YL (2002) The kinase activity of aurora B is required for
kinetochore –microtubule interactions during mitosis. Curr Biol 12: 894 –899.
53. Shepard JL, Amatruda JF, Finkelstein D et al. (2007) A mutation in separase
causes genome instability and increased susceptibility to epithelial cancer.
Genes Dev 21: 55 –59.
54. Pati D (2008) Oncogenic activity of separase. Cell Cycle 7: 3481– 3482.
55. Meyer R, Fofanov V, Panigrahi A et al. (2009) Overexpression and mislocalization of the chromosomal segregation protein separase in multiple human
cancers. Clin Cancer Res 15: 2703 –2710.
56. Viadiu H, Stemmann O, Kirschner MW et al. (2005) Domain structure of
separase and its binding to securin as determined by EM. Nat Struct Mol
Biol 12: 552 –553.
........................................................................................................................................................................................................................................
.........................................................................................................................................................................................................................................
76