Cell, Vol. 105, 645–655, June 1, 2001, Copyright 2001 by Cell Press
Emi1 Is a Mitotic Regulator that Interacts
with Cdc20 and Inhibits
the Anaphase Promoting Complex
appears to be mediated by F box proteins. Substrate
recognition by the APC is less well understood.
Although present throughout the cell cycle, the APC
is inactive while cyclin B accumulatesin S, G2, and early
M phase. The APC is activated by binding the WD-
repeat-containing proteins Cdc20 or Cdh1 (reviewed in
Page and Hieter, 1999). In somatic cells, Cdc20 and
Cdh1 binding to the APC is differentially regulated, re-
sulting in a peak of APCCdc20activity in mitosis and
APCCdh1activity in G1(Kramer et al., 2000; Shirayama et
however, Cdh1 is not expressed, and only APCCdc20is
active (Kramer et al., 2000; Lorca et al., 1998). Cdc20
and Cdh1 may serve as substrate adaptors for the APC,
much like the proposed function for F box proteins in
the SCF, although how they activate the APC is not
or a KEN box motif for ubiquitin-mediated proteolysis
(Pfleger and Kirschner, 2000). Nonetheless, the timing
of APC substrate destruction is varied, suggesting that
additional factors control the timing of APC activity. For
example, cyclin A destruction begins in prometaphase,
inhibitors of sister chromatid separation (securins) are
destroyed at the metaphase-anaphase transition, and
proteolysis of the mitotic kinase Plk1 and the spindle-
associated protein Ase1 occurs as cells exit mitosis
(Geley et al., 2001; den Elzen and Pines, 2001; Charles
et al., 1998; Juang et al., 1997; Shirayama et al., 1998).
Mitotic kinases, including cyclin B/Cdc2, Plk1, and
Hieter, 1999; Zachariae and Nasmyth, 1999). Notably,
mitotic APC phosphorylation promotes its activation by
Cdc20 (Kramer et al., 2000; Rudner and Murray, 2000;
Shteinberg et al., 1999). However, mitotic APC phos-
phorylation is not sufficient to explain the timing of APC
activity, because APC from interphase extract can be
activated in vitro by Cdc20 or Cdh1 (Fang et al., 1998a;
or Cdh1 at any stage in the cell cycle (Schwab et al.,
1997; Visintin et al., 1997).
APCCdc20activity is also restrained by the spindle
checkpoint (SC), a regulatory pathway conserved in
yeast and vertebrates (reviewed in Straight and Murray,
1997). The SC protein Mad2 functions on unattached
kinetochores in prometaphase to inhibit the APC until
Cdc20 to inhibit APC activity (Alexandru et al., 1999;
Fang et al., 1998b; Hwang et al., 1998; Kallio et al., 1998;
Kim et al., 1998; Li et al., 1997; Wassmann and Benezra,
1998), although how Mad2 inhibits APCCdc20is not clear.
The APC also localizes to centrosomes and the mitotic
spindle (Tugendreich et al., 1995), where it may direct
the local degradation of critical substrates (Clute and
Pines, 1999; Huang and Raff, 1999).
shares homology with the Drosophila protein regulator
of cyclin A (Rca1), a positive regulator of cyclin A (Dong
et al., 1997). Loss of Rca1 blocks embryos in G2of cell
cycle 16. Rca1 overexpression in G1causes precocious
cyclin A/Cdk2 activation by increasing cyclin A protein
levels without affecting its transcription. How Rca1 me-
diates this effect is not known.
Julie D.R. Reimann,1Ellen Freed,1Jerry Y. Hsu,1
Edgar R. Kramer,2Jan-Michael Peters,2
and Peter K. Jackson1,3
1Departments of Pathology, Microbiology and
Immunology, Cancer Biology, and Biophysics
Stanford University School of Medicine
300 Pasteur Drive
Stanford, California 94305
2Research Institute of Molecular Pathology
Dr. Bohr-Gasse 7
We have discovered an early mitotic inhibitor, Emi1,
which regulates mitosis by inhibiting the anaphase
promoting complex/cyclosome (APC). Emi1 is a con-
served F box protein containing a zinc binding region
essential for APC inhibition. Emi1 accumulates before
mitosis and is ubiquitylated and destroyed in mitosis,
independent of the APC. Emi1 immunodepletion from
cycling Xenopus extracts strongly delays cyclin B accu-
mulation and mitotic entry, whereas nondestructible
Emi1 stabilizes APC substrates and causes a mitotic
tion. Our results suggest that Emi1 regulates progres-
sion through early mitosis by preventing premature
APC activation, and may help explain the well-known
delay between cyclin B/Cdc2 activation and cyclin B
(MPF), a kinase complex composed of cyclin B and
Cdc2 (Meijer et al., 1989; Minshull et al., 1989; Murray
and Kirschner, 1989). Mitotic exit requires MPF inactiva-
tion, which is achieved by cyclin B destruction. Cyclin
B is degraded by ubiquitin-dependent proteolysis, trig-
gered by the anaphase promoting complex/cyclosome
(APC) ubiquitin ligase (reviewed in Zachariae and Na-
smyth, 1999). However, activated MPF must mediate
chromatin condensation, nuclear envelope breakdown,
and spindle formation before cyclin B is degraded. How
this critical delay between MPF activation and APC acti-
vation is achieved remains unclear.
The vertebrate APC is composed of at least eleven
subunits, including APC2, a cullin family member, and
APC11, a RING-H2 finger protein (Gmachl et al., 2000;
Yu et al., 1998). The APC shares homology with other
ubiquitin ligases, including the SCF (Skp1, Cullin, and F
box protein) ubiquitin ligase, which contains a cullin and
a RING-H2 finger protein as its catalytic core (reviewed
Figure 1. Emi1 Is an F Box Protein Related to Drosophila Rca1
(A) Clustal W alignment of Emi1 and homologs. Xl, Xenopus laevis; Dm, Drosophila melanogaster; Hs, Homo sapiens; and Mm, Mus musculus.
Black ? identity, dark gray ? highly conserved, and light gray ? less highly conserved. The F box, zinc binding region (ZBR), and possible
NLS sequences are boxed.
(B) Emi1 protein (accession # AF319594) schematic, key features, and variant proteins. Emi1 ? wild type; EL198AA ? mutated in 2 conserved
F box residues; Emi1-N-terminus (NT) ? amino acids 1–193; Emi1-C-terminus (CT) ? amino acids 248–392; Emi1-?ZBR ? amino acids 1–338;
Emi1-5P ? substitution of alanine for serine or threonine in all five SP/TP sites; and C346S ? substitution of cysteine 346 with serine. GST-
Skp1 or GST was incubated with
and autoradiography (right).
(C) Characterization of Emi1 antibodies. Rabbit reticulocyte lysate (RRL) programmed with Emi1 (lane 1), unprogrammed RRL (lane 2), Xenopus
XTC cell lysate (lane 3), and interphase Xenopus egg extract (lane 4) were resolved by SDS-PAGE and immunoblotted with affinity-purified
anti-Emi1 or MBP-Emi1 blocked antibodies.
35S-labeled, in-vitro-translated (IVT) proteins, bound to glutathione agarose, and analyzed by SDS-PAGE
lization by inhibiting APC activity. Emi1 accumulates in
S phase and is destroyed in mitosis independent of
the APC, but dependent on Cdk phosphorylation. Emi1
overexpression blocks cells in mitosis with high cyclin
B levels and inhibits cyclin B ubiquitylation in vitro. Emi1
immunodepletion from egg extracts prevents cyclin B
accumulation and mitotic entry. Emi1 binds Cdc20 and
Cdc20 can rescue Emi1-induced cyclin B stabilization
in vitro. The zinc binding region of Emi1 interacts with
Cdc20 and is required for Emi1 to inhibit APCCdc20in
vitro. Collectively, our data indicate that Emi1 controls
the timing of APC activity through Cdc20 regulation.
discussed elsewhere (Regan-Reimann et al., 1999). We
cloned a full-length Xenopus Emi1 oocyte cDNA (see
Experimental Procedures). The predicted Emi1 protein
is 392 residues long with an F box, a zinc binding region
(ZBR), and five possible Cdk phosphorylation sites (Fig-
tion sequences. BLAST search revealed that Emi1 has
homology to the Drosophila protein Rca1 (Figure 1A).
Emi1 and Rca1 are similar in size, placement of func-
tional domains, and share 25% similarity (16% identity).
Emi1 is 43% similar (35% identical) to human Fbx5, a
recently identified F box protein of unknown function
(Cenciarelli etal., 1999). Mutationor deletion ofthe Emi1
F box abrogates binding to Skp1 in vitro (Figure 1B).
Xenopus Emi1 and its homologs contain 8 cysteines
and a histidine in the C terminus that are highly con-
served and may comprise two zinc binding domains
(Figure 1A). The spacing of the cysteines and histidine
in Emi1/Rca1, C-x(2)-C-x(14–30)-C-x(4)-C-x(4)-C-x(2)-
Xenopus Emi1 Is a Cell Cycle Regulated Protein
Related to Drosophila Regulator of Cyclin A (Rca1)
We initially isolated Emi1 in a yeast two-hybrid screen
for Skp1 binding proteins. Details of this screen are
Cell Cycle Regulation by Emi1
described DRIL (TRIAD) cysteine-rich motif (van der
Reijden et al., 1999).
Affinity purified antibodies against Xenopus Emi1 rec-
ognize a protein of the expected molecular mass (44
kDa) in egg extracts and Xenopus XTC lysates, which
is blocked by preincubation of the antibodies with Emi1
protein (Figure 1C). Antibodies also recognize in vitro
grammed reticulocyte lysate.
Emi1 Protein Levels Oscillate in a Cell
We examined the Emi1 protein in the cell cycle of the
early embryo. In fertilized eggs, Emi1 levels increase in
S phase and decrease in M phase (Figure 2A). Emi1 is
presentin CSF-arrestedeggs andpersists afterfertiliza-
tion through the longer first interphase, during pronu-
Extracts made from activated eggs reproduce cell
cycle events in vitro (reviewed in Murray, 1991). Both
endogenous Emi1 and exogenous IVT Emi1 added to
these extracts are ubiquitylated in mitosis (Figure 2B).
Emi1 destruction requires the proteasome because IVT
Emi1 is stabilized when the proteasome inhibitor MG-
132 is added to mitotic egg extracts (data not shown).
Because Emi1 is mitotically destroyed, we tested
cyclin B fragment was incubated in Xenopus egg ex-
cyclin B (?90). In these ?90 extracts, the APC is active
and cyclin B is degraded (King et al., 1995). IVT Emi1
protein is destroyed in ?90 extracts, but not interphase-
arrested extracts. APC immunodepletion or addition of
to inhibit APC-mediated proteolysis (Holloway, 1993;
King et al., 1995), prevented cyclin B destruction,
whereas a control peptide did not (Figure 2C). However,
Emi1 was destroyed with similar kinetics whether the
APC was depleted or blocked by destruction box pep-
tides or a control peptide (Figure 2C). Thus, Emi1 does
not appear to be an APC substrate in the egg.
To investigate the sequence requirements for Emi1
destruction, we constructed N- or C-terminal Emi1 frag-
ments (Figure 1B). IVT Emi1 N terminus (Emi1-NT) was
destroyed with kinetics similar to full-length Emi1 in ?90
extracts (t1/2≈ 10 min), whereas the C terminus (Emi1-
CT) was stable (t1/2? 100 min; Figure 2D). Because the
N terminus contains four of five possible Cdk phosphor-
ylation sites in Emi1, we mutated serine or threonine to
alanine in all five sites and found that this Emi1-5P mu-
tant was stable in ?90 extracts compared to wild type
(Figure 2E). Interestingly, the N terminus of Emi1 identi-
fied Xenopus cyclins B1 and B2 as interacting proteins
several times in a yeast two-hybrid screen (data not
shown). We do not yet know whether Emi1 is a Cdk
substrate in vivo, but we found that full-length Emi1 and
phosphorylated (Figure 2E). Further, Emi1 binds the mi-
totic cyclins A and B in vitro and Emi1 is a phosphopro-
tein in egg extracts (data not shown). Thus, a plausible
model is that phosphorylation of Emi1 by mitotically
active kinases triggers the APC-independent destruc-
tion of Emi1.
Figure 2. Emi1 Is Destroyed in Mitosis
(A) Emi1 levels fluctuate in the embryonic cell cycle. Fertilized eggs
were incubated at 23?C, equal numbers of embryos removed at the
indicated times, and processed for immunoblotting with anti-Emi1
antibodies (upper panel) and for histone H1 kinase activity of immu-
noprecipitated cyclin B1 (lower panel).
(B) Emi1 is ubiquitylated in cycling extracts. Left: Activated Xenopus
at 23?C. Aliquots were removed at the indicated times and analyzed
as determined by cyclin B ubiquitylation and histone H1 kinase
activity. Right: Interphase and mitotic extract with added FLAG-
with anti-Emi1 sera, and analyzed by immunoblotting for FLAG-
ubiquitin. * ? IgG band.
(C) Emi1 destruction does not require the APC.35S-labeled IVT Emi1
or N-terminal cyclin B fragment was added to ?90 extracts treated
with either destruction box (D-box) peptide, scrambled peptide
(control), or depleted of the APC with anti-Cdc27 antibodies. Ali-
quots were removed at the indicated times and analyzed by SDS-
PAGE and autoradiography.
(D) In mitotic extracts, Emi1 and its N terminus are unstable; the C
terminus is stable.35S-labeled IVT full-length, N-terminal, or C-terminal
Emi1 was added to ?90 extracts and assayed for stability as in (C).
(E)Mutation ofthefive possibleCdkphosphorylation sitesstabilizes
Emi1.35S-labeled IVT wild-type Emi1 or a mutant in all five SP/TP
sites (Emi1-5P) was added to ?90 extracts and assayed for stability
as in (C), left. Equimolar amounts of purified MBP-Emi1, MBP-Emi1-
NT, MBP-Emi1-CT, or MBP-Emi1-5P were incubated with purified
cyclin B/Cdc2 in the presence of [32P]-?ATP. Proteins were analyzed
by SDS-PAGE and autoradiography (right).
Emi1 Inhibits APC Activity in Xenopus Egg Extracts
The oscillation of Emi1 in Xenopus embryos and the
G2 arrest seen in Rca1-deficient Drosophila embryos
suggested that like cyclin B, Emi1 accumulation may be
important for mitotic entry and that Emi1 destruction
may be necessary for mitotic exit. To test whether Emi1
destruction is required for mitotic exit, we analyzed the
effect of Emi1 addition to Xenopus extracts. Addition of
Figure 3. Emi1 Prevents the Ubiquitin-Mediated Destruction of APC Substrates and Inhibits Mitotic Exit
(A) Emi1 prevents cyclin A and B destruction and mitotic exit in cycling egg extracts. Activated Xenopus cycling egg extracts were incubated
with either buffer (?) or 1 ?M purified MBP-Emi1 (?). Aliquots were removed at the indicated times and assayed for DNA morphology (graph)
or Xenopus cyclins A and B by immunoblotting (lower panels).
(B) Excess Emi1 does not affect the kinetics of cyclin B/Cdc2 activation in egg extracts. Activated Xenopus cycling egg extracts were incubated
with buffer or 1 ?M purified MBP-Emi1. Aliquots were removed at the indicated times and processed for the histone H1 kinase activity of
immunoprecipitated cyclin B1.
(C) Emi1 stabilizes securin and geminin.35S-labeled IVT Xenopus securin or geminin protein was incubated in ?90 extracts treated with buffer
(control) or 1 ?M purified MBP-Emi1. Aliquots were removed at the indicated times and analyzed by SDS-PAGE and autoradiography.
(D) Emi1 inhibits cyclin B ubiquitylation in mitotic extracts. An125I-labeled cyclin B N-terminal fragment was incubated in ?90 extracts treated
with 2.5 ?M purified MBP (left) or MBP-Emi1 (right). Aliquots were removed at the indicated times and analyzed as in (C).
(E–G)35S-labeled IVT cyclin B N-terminal fragment was added to ?90 extracts treated with Emi1 variants. Aliquots were removed at indicated
times, resolved by SDS-PAGE and quantitated by Phosphorimager.
(E) The Emi1 C terminus is sufficient to block cyclin B destruction. Additions: buffer (?) or 1 ?M purified MBP-Emi1 (?), MBP-Emi1-NT (?),
or MBP-Emi1-CT (?). (F) The Emi1 ZBR but not the F box domain is required to block cyclin B destruction. Additions: buffer (?), 1 ?M purified
MBP-Emi1 (?), MBP-EL198AA (?), MBP-Emi1?ZBR (?), or MBP-Emi1–5P (?). (G) ZBR mutations fail to inhibit cyclin B destruction. Additions:
buffer (?), 1 ?M purified MBP-Emi1 (?), MBP-C341S (?), MBP-C346S (?), MBP-C351S (?), MBP-C354S/C356S (?), or MBP-C364S (?).
(H) Injection of Emi1 blocks Xenopus embryos in mitosis with high Cdk kinase activity. One pmol purified MBP-Emi1, MBP-Emi1-NT, MBP-
Emi1-CT, or MBP was injected into one blastomere (right side) of two-cell stage Xenopus embryos. Embryos were photographed 2.5 hr after
injection (left panel). For kinase assays, both blastomeres of two-cell stage embryos were injected and extracts from injected embryos assayed
for histone H1 kinase activity 2.5 hr post-injection (right panel). Unfertilized eggs and equivalent aliquots of interphase and ?90 extracts were
assayed as controls.
the destruction of endogenous cyclins A and B and
mitotic exit (Figure 3A). Addition of equimolar amounts
of MBP or another Xenopus F box protein had no effect
on cyclin stability or mitosis (data not shown). Excess
Emi1 did not affect the timing of mitotic entry or MPF
activation in egg extracts, as analyzed by DNA morphol-
ogy (Figure 3A) or cyclin B/Cdc2 kinase activity (Figure
3B). By quantitative immunoblotting, we estimate Emi1
to be ?300 nM in interphase egg extracts. As little as
100 nM additional Emi1 protein stabilizes cyclins A and
B. However, we see a stronger delay between cyclin B/
Cdc2 activation and cyclin B destruction with 300 nM
to 1 ?M Emi1 protein concentrations, likely because
Emi1 is itself destroyed in mitosis.
We found that Emi1 also inhibits the destruction of
two other known APC substrates, securin and geminin,
affects APC substrate ubiquitylation, we measured
cyclin B ubiquitylation in ?90 extracts treated with puri-
fied MBP or MBP-Emi1 protein. Addition of MBP-Emi1
strongly reduced the ubiquitylation of an iodinated
amino-terminal fragment of cyclin B containing the de-
struction box, whereas MBP did not (Figure 3D).
To determine which domains of Emi1 are required
to block cyclin B destruction, we tested several Emi1
mutants (schematic, Figure 1B). Cyclin B was destroyed
in ?90 extracts treated with buffer (control) or an MBP-
Emi1-NT fusion protein, but was stabilized in the pres-
ence of MBP fusions to wild-type Emi1, Emi1-5P, the F
box mutant (EL198AA), or Emi1-CT (Figures 3E and 3F).
Therefore, the Cdk sites, the F box, and the region N-ter-
minal to the F box are not required for Emi1 to stabilize
cyclin B; however, the C terminus is both necessary and
sufficient. An Emi1 truncation mutant missing the C-ter-
minal ZBR (Emi1-?ZBR) was incapable of stabilizing cy-
clin B (Figure 3F). Further, mutations of conserved ZBR
residues cysteine 341 (C341S) or cysteine 346 (C346S)
to serine greatly reduced the ability of Emi1 to inhibit
sary for Emi1 to inhibit APC activity.
To test whether Emi1 affects the cell cycle in vivo, we
injected the protein into one blastomere of a two-cell
stage Xenopus embryo. Emi1 caused a stable cell cycle
blastomere continued to divide normally (Figure 3H).
Embryos injected in both blastomeres with Emi1 had a
Cell Cycle Regulation by Emi1
Figure 4. Transfection ofEmi1 intoXTC Cells
Causes a Mitotic Block
(A) Emi1 localization. XTC cells were labeled
with affinity-purified antibodies to Emi1, anti-
?-tubulin, and Hoechst 33258 dye. Anti-Emi1
(“Block”). The Emi1 staining (red) and ?-tubu-
lin (green) images were merged (Merge) to
show Emi1 spindle localization.
(B) Deconvolution image of Emi1 spindle lo-
calization. XTC cells were labeled as in (A),
and the Emi1 staining (red) and ?-tubulin
(green) images were merged (Merge) to show
the Emi1 spindle localization.
(C) Emi1 overexpression causes a mitotic in-
dex increase. XTC cells were cotransfected
with GFP and myc-tagged constructs ex-
pressing Emi1 variants. Cells were fixed and
stained with anti-?-tubulin antibody and
Hoechst33258 dye.Thenumberof GFPposi-
tive mitotic cells was quantitated based on
DNA and spindle morphology.
(D) Flow cytometric analysis of Emi1-trans-
fected XTC cells. Cells were fixed, labeled
with propidium iodide, and analyzed by flow
cytometry. The table lists the % GFP positive
cells in each cell cycle stage for each trans-
fection. “*” ? The percentage mitotic for the
Emi1-5P mutant is likely an underestimate
because many cells expressing this mutant
(E and F) Emi1 overexpression blocks cells
in prometaphase. XTC cells were transfected
with either myc vector or myc-tagged Emi1
sualize ?-tubulin and DNA. Promet. ? normal
prometaphase cell, met. ? normal meta-
phase cell (E). The number of GFP-positive
cells in each mitotic phase was quantitated
as in (C) (shown in [F]).
high level of histone H1 kinase activity similar to that
detected in ?90 extracts, whereas uninjected and con-
trol-injected embryos had H1 kinase levels similar to
interphase extracts (Figure 3H). As in cycling extracts,
the Emi1 C terminus with an intact ZBR was also neces-
sary and sufficient to mediate the mitotic block in vivo
and wild-type and N-terminal Emi1 are unstable in vivo
(Figure 3H and data not shown). In summary, Emi1
blocks the cell cycle at mitosis both in vitro and in vivo
and prevents the ubiquitin-mediated destruction of
known APC substrates in vitro.
EL198AA, Emi1-5P, or Emi1-CT caused a mitotic index
increase compared to vector, whereas neither Emi1-
NT nor the C346S point mutant had a significant effect
(Figure 4C). The mitotic block was confirmed by flow
cytometric analysis of DNA content (Figure 4D). The
stable Emi1 mutants (Emi1-CT and Emi1-5P) caused a
stronger mitotic delay than the unstable wild-type Emi1.
Overexpression of the APC inhibitor Mad2 in XTC cells
caused a mitotic index increase similar to Emi1 (?9%;
data not shown).
DNA and spindle morphology examination revealed
that cells transfected with Emi1, EL198AA, Emi1-5P, or
Emi1-CT accumulated predominantly in prometaphase
(Figures 4E and 4F). Cyclin A destruction (which is
blocked by Emi1), occurs in prometaphase and cyclin A
overexpression causes a prometaphase delay in human
cells (den Elzen and Pines, 2001) and in XTC cells (our
unpublished data). In contrast, Mad2 does not stabilize
cyclin A and blocks predominantly in metaphase when
transfected into XTC cells (J.D.R.R. and P.J., unpub-
Emi1 Overexpression in Somatic Cells
Causes a Mitotic Block
To examine Emi1 subcellular localization, we stained
Xenopus XTC cells with affinity-purified antibodies to
Emi1. In interphase, the protein localizes in a punctate
pattern in the nucleus and the cytoplasm, with some
perinuclear concentration (Figure 4A). In mitotic cells,
Emi1 localized throughout the cell and particularly at
the spindle (Figures 4A and B).
Because Emi1 can stabilize several APC substrates,
which are each destroyed at specific times in mitosis,
we tested more precisely when in mitosis Emi1 blocks
by overexpressing epitope-tagged Emi1 variants in so-
matic cells. Because Emi1 is unstable in mitotic XTC
cells (data not shown), the myc-tagged Emi1 variants
were cotransfected with a GFP expression construct to
mark transfected cells. Transfection of wild-type Emi1,
Emi1 Depletion Prevents Cyclin B Accumulation
and Mitotic Entry
If Emi1 normally inhibits APC activity in interphase, then
Emi1 depletion from cycling egg extracts might block
ing Emi1 immunodepletion (Figure 5D), we examined
cyclin B accumulation and DNA morphology as markers
Figure 5. Emi1 Depletion Prevents Mitotic Entry in Egg Extracts
(A) Emi1 depletion prevents cyclin B accumulation in Xenopus cycling extracts. Equal aliquots were removed at the indicated times from
preimmunesera-depleted, Emi1-depleted,or Emi1-depletedcycling extractspreincubated witheither 300nM MBP-Emi1,0.13 volumesextract,
or beads from the Emi1 depletion. Samples were processed for immunoblotting with anti-cyclin B2 and anti-Orc1 antibodies (as a loading
control). Exposure time is the same for all blots.
(B and C) Emi1-depleted cycling extracts fail to enter mitosis. Sperm DNA was added to preimmune sera-depleted, Emi1-depleted, or Emi1-
depleted cycling extracts preincubated with either 300 nM MBP-Emi1, 0.2 volumes extract, 6 ?M GST-Mad2, or 60 ?g/ml GST-?90 cyclin B.
Aliquots were removed at the indicated times, fixed onto slides, and DNA visualized by Hoechst 33258 staining (B). The number of interphase
and mitotic figures was quantitated (C).
(D) Equal amounts of undepleted, preimmune sera-depleted and Emi1-depleted extracts were resolved by SDS-PAGE and processed for
immunoblotting with anti-Emi1 antibodies. Emi1 depletion removes ?80% of the protein.
of mitotic entry. In cycling extracts, cyclin B normally
peaks by 80 min and is destroyed by 120 min. In Emi1-
ure 5A). Addition of beads from the Emi1 immunode-
pletion rescued the accumulation and subsequent
destruction of cyclinB. Addition of 300nM purified Emi1
protein rescued cyclin B accumulation but blocked its
destruction (Figure 5A). This is likely because excess
Emi1 is not completely destroyed, thus inhibiting the
APC and stabilizing cyclin B.
The effect of Emi1 depletion on mitosis was verified
by examining DNA morphology in cycling extracts. In
control extracts, demembranated sperm DNA was
highly condensed by 60 min, indicating onset of mitosis,
ogy by 90 min (Figures 5B and 5C). In Emi1-depleted
extracts, nuclei remained intact with DNA decondensed
(Figures 5B and 5C). Addition of undepleted extract or
purified Emi1 to Emi1-depleted extracts rescued mitotic
entry. Although Emi1-depleted extracts rescued with
undepleted extract progressed past metaphase, ex-
tracts rescued with Emi1 protein did not, presumably
because Emi1 blocks APC-dependent securin destruc-
tion and thus sister chromatid separation. It is not clear
whether our assay in egg extracts can easily distinguish
prometaphase and metaphase or whether, like the spin-
dle checkpoint, a mechanism necessary to block in pro-
metaphase fails to function in egg extracts.
If Emi1 depletion prematurely activates the APC, then
addition of the APC inhibitor Mad2 should also rescue
mitotic entry. Mad2 addition to Emi1-depleted extracts
did rescue mitotic entry (Figures 5B and 5C), although,
much like rescue with Emi1 protein, the extracts did
not progress beyond metaphase. To test whether the
inability of Emi1-depleted extracts to enter mitosis was
primarily due to their failure to accumulate cyclin B,
we also tested whether nondestructible ?90 cyclin B
addition rescued mitotic entry. ?90 addition to depleted
extracts rescued nuclear envelope breakdown and mi-
totic DNA condensation, indicating that nondestructible
cyclin B can overcome the requirement for Emi1 in mi-
totic entry (Figures 5B and 5C).
Emi1 Interacts with the APC Activator Cdc20
To better understand how Emi1 controls APC activity,
we looked for interacting proteins by yeast two-hybrid
screens of a Xenopus oocyte library using Emi1 as the
bait. Screening with full-length Emi1 identified only Skp1,
therefore we tested Emi1-NT and Emi1-CT for interacting
proteins as well (see Experimental Procedures). As pre-
viously mentioned, the Emi1-NT bait identified cyclin B.
Cdc20. To validate this interaction, we took several ap-
proaches. First, Emi1 and Cdc20 coimmunoprecipitate
from egg extracts (Figure 6A). This interaction appears
to be APC independent, since we were unable to detect
the APC subunit APC2 in the precipitate. Second, in-
Cell Cycle Regulation by Emi1
Figure 6. Emi1p Interacts with Cdc20 and Inhibits APC Activation by Cdc20
(A) Emi1 coimmunoprecipitates with Cdc20 but not APC2. Preimmune (PI) or anti-Emi1 immunoprecipitates from interphase egg extract were
assayed by immunoblotting for APC2 (upper panel) and Cdc20 (lower panel).
(B) Sucrose gradient cosedimentation of Emi1 and Cdc20. Interphase egg extract was fractionated on a 10%–40% sucrose gradient, and
fractions analyzed by immunoblotting with antibodies to the indicated proteins.
(C) Emi1 and Cdc20 cofractionate during gel filtration chromatography. Interphase egg extract was resolved on a Resource Q anion exchange
column and fractions containing Emi1 chromatographed on an S-300 gel filtration column and immunoblotted for Emi1 and Cdc20 (left panel).
Preimmune (PI) or anti-Emi1 immunoprecipitates from a 100 kDa–200 kDa fraction were assayed by immunoblotting for Cdc20 (right).
(D) Emi1 and Cdc20 associate in baculovirus coinfection. SF9 cells were coinfected with baculovirus-expressed Emi1 and Cdc20, precipitated
with preimmune (PI) or anti-Emi1 antisera, and analyzed for Cdc20 by immunoblotting.
(E) Cdc20 rescues cyclin B destruction.35S-labeled IVT N-terminal cyclin B was added to mitotic Xenopus egg extracts treated with buffer
(?), 1 ?M purified MBP-Emi1 (?), 1 ?M MBP-Emi1 plus 1 ?M His-Cdc20 (?), or 1 ?M MBP-Emi1 plus 3 ?M His-Cdc20 (?). Aliquots were
removed at the indicated times, resolved by SDS-PAGE and quantitated on a Phosphorimager.
(F) Cdc20 interacts with both the N terminus and the ZBR of Emi1 in vitro. Purified MBP-Emi1 fusion protein variants and purified baculovirus-
panel). Purified GST-Emi1, GST-Emi1-NT, GST-Emi1-CT?ZBR (residues 248–334), or GST-Emi1-CTZBR (residues 335–364) was incubated with
35S-labeled in vitro translated (IVT) Cdc20(1–158), bound to glutathione agarose, and analyzed by SDS-PAGE and autoradiography (lower
(G) Inhibition of Cdc20-mediated activation of the APC by Emi1. IVT Cdc20 (2–8) or rabbit reticulocyte lysate (1) was incubated for 30 min
with buffer (1 and 2) or purified bacterially expressed 1 ?M MBP-Emi1 (3), 3 ?M MBP-Emi1 (4), 6 ?M MBP-Emi1 (5), 3 ?M MBP-Emi1-CT (6),
6 ?M MBP (7), or 20 ?M GST-Emi1-CT?ZBR (8). APC was immunopurified from mitotic egg extracts with anti-Cdc27 beads, then incubated
with the Cdc20/protein mixtures for 1 hr. APC beads were washed, and assayed for cyclin ubiquitylation activity using a
N-terminal Xenopus cyclin B substrate.
terphase extracts separated on a sucrose gradient or
gel filtration column showed that Emi1 and Cdc20 co-
fractionate (Figures 6B and 6C). Emi1 is found in two
higher molecular weight pools, ?100–200 kDa and
?300–500 kDa. Cdc20 cofractionates in the ?100–200
kDa complex and Cdc20 coimmunoprecipitates with
that does not cofractionate with Emi1 cofractionates
with the ?1.5 MDa APC complex (Figure 6B); however,
Cdc20 binds weakly to the inactive interphase APC
(Fang et al., 1998a; Kramer et al., 2000). Interestingly, a
slower migrating form of Cdc20 is consistently seen in
some of the fractions containing Emi1. The nature and
significance of this modification is not known. We can
also reconstitute the interaction between Emi1 and
Cdc20 with baculovirus or using purified proteins (Fig-
ures 6D and 6F), further supporting a direct interaction.
Emi1 Can Block Cdc20-Dependent APC
Activation In Vitro
If Emi1 inhibits Cdc20 activation of the APC, then Cdc20
protein should rescue the Emi1 block to cyclin B de-
struction. Baculovirus-expressed Cdc20 addition to mi-
totic extracts rescued the Emi1-induced block to cyclin
B destruction in a dose-dependent manner (Figure 6E),
supporting the hypothesis that Emi1 prevents Cdc20
from activating the APC. This result also indicates that
Emi1 is present, reinforcing our other observations that
Emi1 does not directly inhibit the APC enzymatic com-
plex (see discussion).
We knew that the Emi1 N terminus interacts with
Cdc20 from our two-hybrid screen, but the C terminus
of Emi1 (Emi1-CT) also bound Cdc20 in vitro (Figure
6F). We confirmed the Cdc20-Emi1-CT interaction in the
Figure 7. A Model for Emi1 Regulation of the
Anaphase Promoting Complex
yeast two-hybrid system. Interestingly, we also ob-
served in both yeast two-hybrid and in vitro binding
assays, that the Cdc20 N terminus from residues 1–158,
but not the WD repeat domain, is sufficient for binding
to Emi1 (data not shown). Because the C-terminal ZBR
is required for Emi1 to inhibit APC activity, we tested the
conserved region of the ZBR, GST-Emi1-CTZBR (resi-
dues 335–364), interacts with the Cdc20-NT, whereas
the C-terminal fragment without the ZBR, GST-Emi1-
CT?ZBR (residues 248–334), does not (Figure 6F).
We utilized this binding information to test whether
the interaction between Emi1 and Cdc20 is required
for Emi1 to inhibit Cdc20 from activating the APC in a
reconstituted system. Addition of full-length Emi1 pro-
tein prevented Cdc20 from activating APC immunopuri-
fied from mitotic egg extract in a dose-dependent fash-
ion (Figure 6G). The Emi1 C terminus, which contains
the ZBR, is sufficient to inhibit cyclin B ubiquitylation in
this purified system whereas the Emi1-CT?ZBR, which
fails to bind Cdc20, does not inhibit.
son et al., 2000), mitotically phosphorylated Emi1 may
be an SCF substrate.
Emi1 destructionmay be influenced byits association
with Cdc20. An interesting possibility is that phosphory-
lation by cyclin B/Cdc2 triggers the dissociation of Emi1
and Cdc20, thereby promoting Emi1 destruction. In-
deed, we found that Cdc20 addition not only rescued
the Emi1 block of APC activity, but also stabilized Emi1
in mitotic extracts (J.D.R.R. and P.J., unpublished data)
suggesting that Emi1 is more stable when complexed
Emi1 Inhibits the APCCdc20Complex
Cdc20 exists in high molecular weight complexes both
with and independent of the APC (Kramer et al., 1998;
Lorca et al., 1998). Emi1 and Cdc20 coimmunoprecipi-
tate from interphase extracts in a complex independent
of the APC, suggesting a model where Emi1 sequesters
Cdc20 from the APC (Figure 7). We do not yet know if
Emi1 is associated with E3 activity, but one possibility
is that Cdc20 is a substrate of an Emi1-containing SCF
complex. However, addition of Emi1 to egg extracts
does not destabilize Cdc20 (J.D.R.R. and P.J., unpub-
lished data), and the F box is not required for Emi1 to
block cyclin B destruction, making it unlikely that Emi1
directs Cdc20 destruction. Cdc20 is also stable in the
early embryo (Kramer et al., 2000; Lorca et al., 1998) so
the ability of Emi1 to regulate the cell cycle is unlikely
to require Cdc20 destruction.
Our in vitro APC inhibition assays and rescue experi-
ments indicate that Emi1 is a direct Cdc20 inhibitor. The
Emi1 ZBR is required to inhibit the APC and binds to
Cdc20 in vitro. The ZBR appears to cooperate with the
Emi1 N terminus to bind Cdc20 and may prevent the
interaction of Cdc20 with APC substrates. Importantly,
Emi1 does not inhibit the substrate and Cdc20-indepen-
dent ubiquitylation activity of the APC2/APC11 core
complex (Gmachl et al., 2000), further indicating that
Emi1 inhibits APC activity through Cdc20 and not at
the level of the APC enzymatic machinery (J.D.R.R., B.
Gardner, and P.J., unpublished data). Further indicating
activity in vitro, or SCF-dependent events (DNA replica-
tion and mitotic entry) in egg extracts.
We have identified an APC inhibitor called Emi1, which
is required for mitotic entry. Emi1 is unstable in mitosis
and expression of nondestructible versions of the pro-
tein or overexpression of the wild-type protein causes
a mitotic block in embryos and somatic cells. Emi1 de-
struction is APC-independent in the egg and appears
to require phosphorylation by mitotically active Cdks.
Emi1 binds the APC activator Cdc20 and Cdc20 can
rescue the Emi1-induced block of cyclin B degradation,
indicating that Cdc20 is the target of Emi1-APC regu-
Identification of an Independent Cell Cycle
Oscillator that Controls APC Activity
Like cyclin B, Emi1 must accumulate for mitotic entry
and be destroyed for mitotic exit. Emi1 is destroyed in
mitosis by ubiquitin-mediated proteolysis and its de-
struction likely requires phosphorylation by mitotic ki-
nases, including cyclin B/Cdc2. We are currently testing
the mechanism of Emi1 destruction in mitosis, but it
does not require the APC. Several F box proteins are
unstable in mitosis, and the SCF has been implicated in
theirdestruction. BecausemanySCF substratesrequire
Emi1 as a Mitotic Timer and Potential Checkpoint
Protein for APC Activation
Cyclin B ubiquitylation activity of APC immunoprecipi-
tated from synchronized HeLa cells increases signifi-
cantly before cyclin B levels decrease, and the APC
Cell Cycle Regulation by Emi1
library (Stratagene). A 1.9 kb clone was sequenced on both strands
and contains stop codons upstream of the 5? start codon and a
3? poly-A tail. In vitro translation produces a 44 kDa species in
Emi1-NT (1–193), Emi1-CT (233–392), or Emi1 full-length (fl) were
cloned into pAS2 and used to screen (2.5 million clones each) a
were verified with fl Emi1 by filter lift ?-galactosidase assay.
subunit Cdc27 is phosphorylated well before cyclin B
levels decrease (Kramer et al., 2000). This delay in APC
suggests the presence of an inhibitor that restrains full
APC activation until nuclear envelope breakdown, spin-
dle assembly, and chromatin condensation have oc-
curred. The delay might be explained in part by Mad2,
which is required for APC inhibition in prometaphase
until chromosomes have been properly aligned at the
metaphase plate (Gorbsky et al., 1998; Taylor and
McKeon, 1997). However, although anti-Mad2 antibody
not affect progression through prophase, when MPF is
also active and Cdc20 is present (Gorbsky et al., 1998).
We considered whether Emi1 cooperates with Mad2
to inhibit Cdc20 as part of the SC pathway. However,
Emi1 does not bind to Mad2 in vitro and Emi1 competes
with Mad2 for binding to Cdc20 in vitro (J.D.R.R. and
P.J., unpublished data). Further, cyclin A destruction is
not inhibited when Mad2 is activated (Geley et al., 2001;
den Elzen and Pines, 2001), whereas Emi1 stabilizes
cyclin A in cycling extracts. In human cells, cyclin A
normally is destroyed in prometaphase (Geley et al.,
2001; den Elzen and Pines, 2001), at a time when Emi1
delay like Mad2 because Emi1 stabilizes cyclin A, which
is not seen with Mad2 overexpression. Consistently, co-
expression of the cyclin A/Cdk2 inhibitor p21Cip1with
Emi1 canreverse the Emi1-induced mitoticblock in XTC
cells (our unpublished data). However, it is possible that
the prometaphase delay induced by Emi1 results from
the stabilization of other as yet unidentified APC sub-
strate(s) that must normally be destroyed in early mitosis.
The observation that Emi1 immunodepletion delays
that Emi1 inhibits the APC in interphase and in early
mitosis, before Mad2 begins to function. Similarly, loss
of the likely Emi1 homolog Rca1 prevents mitotic entry
in Drosophila embryos (Dong et al., 1997). Because APC
inhibition by Mad2 or proteasome inhibition by addition
mitosis in Emi1-depleted extracts, Emi1 most likely af-
for example, its translation. Nonetheless, we cannot ex-
clude that Emi1 may have additional roles in promoting
Recent studies indicate APC activation is spatially as
well as temporally restricted. Notably, cyclin B proteoly-
sis begins first at the spindle poles (Clute and Pines,
1999; Huang and Raff, 1999) and Mad2 activation at
kinetochores restrains securin destruction to prevent
chromosome segregation. Are there sensing mecha-
nisms other than the SC that regulate the APC? Mitotic
events other than kinetochore capture by microtubules,
namely chromatin condensation, centrosome separa-
tion, nuclear envelope breakdown, and spindle forma-
tion must occur sequentially before APC activation. It
is therefore possible that these critical prophase and
prometaphase events are controlled by sensing mecha-
nisms that involve Emi1.
Preparation of Emi1 Full-Length and Mutant Constructs
Emi1 and variants were cloned into pCS2-5mt (myc-tagged), pMAL
or pGEX vectors and Cdc20(1–158) into pCS2 vector. Emi1-pCS2-
5mt site-directed mutants (E198A and L199A [EL198AA], S10A,
S29A, S105A, T123A, S328A [Emi1-5P], C341S, C346S, C354S, and
C356S [C354S/C356S], and C364S), were verified by sequencing.
Emi1 baculovirus was generated using the BAC-TO-BAC system
Emi1 variants were produced as MBP fusion proteins and purified
by standard protocols. Human Cdc20 baculovirus protein was as
described (Kramer et al., 2000).
Bacterially produced MBP-Emi1 was used to raise polyclonal anti-
were affinity purified on a GST-Emi1 column.
Binding Assays and Chromatography
In vitro GST-fusion protein binding reactions were as described (Bai
et al., 1996). In vitro MBP protein binding assays: 100 nM purified
MBP-Emi1, MBP-Emi1-NT, MBP-Emi1-CT, or MBP was incubated
with 100 nM His-Cdc20 in Buffer 1 (50 mM Tris [pH 7.5], 100 mM
NaCl, and 0.1% NP-40), supernatants bound to amylose beads,
washed 4 times with buffer 2 (50 mM Tris [pH 7.5], 300 mM NaCl,
and 1% NP-40), and bound proteins resolved by SDS-PAGE and
SF9 cells coinfected with Emi1 and Cdc20 baculoviruses were lysed
in RIPB (100 mM NaCl, 50 mM ?-glycerophosphate, 5 mM EDTA,
0.1% Triton X-100, and 1 mM DTT), and lysates precleared with
4 times in RIPB, and analyzed by anti-Cdc20 immunoblots.
Interphase egg extract was diluted 1:5 in buffer (100 mM KOAc [pH
7.2], 2.5 mM Mg(OAc)2, 5 mM EGTA, 2 mM DTT, 10 mM Tris [pH
7.2],80 mM?-glycerophosphate,and 100mMsucrose), andcleared
at 40,000 rpm (SW50.1 rotor, 1 hr, 4?C). Lysate was resolved on a
10%–40% w/v sucrose gradient, centrifuged in an SW40.1 rotor
(30,000 rpm, 18 hr, 4?C), and fractions analyzed by SDS-PAGE and
High speed interphase Xenopus egg extract supernatants were
fractionated on a Resource Q column, and eluted with a 0–0.5 M
NaCl gradient. Pooled Emi1-containing fractions were separated on
an S-300 gel filtration column. Egg extract or the 100–200 kDa frac-
tion immunoprecipitated with anti-Emi1 or PI sera (as above) were
analyzed by anti-Cdc20 or anti-APC2 immunoblot.
HistoneH1 kinaseactivity(Murray, 1991)andcyclinB kinaseactivity
(Jackson et al., 1995) were analyzed as described. In vitro cyclin B
phosphorylation experiments: 1 ?M purified MBP-Emi1 or MBP-
Emi1 variants were incubated with 2 units cyclin B/Cdc2 (NEB) in ki-
nase buffer plus 66 ?M ATP and 0.25 ?Ci/?l [32p]-?ATP) (15min, RT),
Xenopus Extracts and Embryos
activated with calcium ionophore A23187. To assay DNA morphol-
ogy, sperm nuclei were added, fixed at various times, and DNA
labeled (Hoechst 33258). Endogenous cyclin A and B levels were
assayed by immunoblots with anti-Xl cyclin B2 or anti-Xl cyclin A1
mouse monoclonal antibodies (S. Geley and T. Hunt). Mitotic ex-
tracts were made by addition of nondegradable ?90 sea urchin
Emi1 Cloning and Yeast Two-Hybrid Screen
A partial cDNA isolated in a Skp1 yeast two-hybrid screen (Regan-
Reimann et al., 1999) was used to screen a Xenopus ovary cDNA
Embryo Injection Experiments
9.2 nl of 100 ?M protein was injected into one blastomere at the
two-cell stage. Injected embryos were transferred to 0.1? MMR
with 3% Ficoll. H1 kinase activity in injected embryos was assayed
as described (Fang et al., 1998b).
B. Gardner for baculovirus protein, and G. Fang, G. Gorbsky, P.
Tavormina, L. Furstenthal, and A. Eldridge for comments on the
manuscript. This research was supported by the NIGMS Medical
Scientist Training Grant GM07365 (J.D.R.R.), Boehringer Ingelheim
and by the Austrian Industrial Research Promotion and Austrian
Science Promotion Funds (J.P. and E.R.K.), Cancer Biology Training
Grant CA09302 and HHMI (J.Y.H.), and NIH grants GM54811 and
Received November 10, 2000; revised April 23, 2001.
Degradation and Ubiquitylation Assays
Interphase or mitotic extracts were incubated at 23?C for 60 min
with 4.6 ng/?l FLAG-ubiquitin, 1 ?M ubiquitin aldehyde, and 2 mM
MG-132. Time pointswere diluted in RIPB,immunoprecipitated with
mouse anti-Emi1 sera or PI sera and analyzed by anti-FLAG (Sigma)
immunoblots. Substrate degradation in ?90 or cycling extracts:35S-
labeled IVT protein was added and extracts incubated (23?C). Ali-
quots were removed, resolved by SDS-PAGE, and quantitated on
a Phosphorimager. The cyclin B substrate was an N-terminal sea
urchin cyclin B fragment (aa 13–91)-protein A fusion (Glotzer et al.,
1991). Extracts were treated with 1 mM Hs cyclin B destruction box
peptide or a scrambled version, or depleted of the APC with anti-
To assay Emi1’s effect on APC activity, 1 ?M MBP fusion protein,
1 ?M control protein, or buffer was added. To assay Emi1’s effect
on cyclin B ubiquitylation, 2.5 ?M MBP-Emi1 or MBP was incubated
in?90extracts (20min),withiodinated sea-urchincyclin-Bfragment
as described (Kramer et al., 2000).
In Vitro APC Assay
Mitotic extract anti-Cdc27 immunoprecipates were incubated (1hr,
4?C) with IVT hCdc20 preincubated with Emi1, control protein, or
buffer, washed in XB?, and assayed for cyclin ubiquitylation as
described (Fang et al., 1998a), using a35S-labeled IVT Xl cyclin B1
(aa 2–97) fragment as substrate.
Alexandru, G., Zachariae, W., Schleiffer, A., and Nasmyth, K. (1999).
Sister chromatid separation and chromosome re-duplication are
regulated by different mechanisms in response to spindle damage.
EMBO J. 18, 2707–2721.
Bai, C., Sen, P., Hofmann, K., Ma, L., Goebl, M., Harper, J.W., and
Elledge, S.J. (1996). SKP1 connects cell cycle regulators to the
ubiquitin proteolysis machinery through a novel motif, the F box.
Cell 86, 263–274.
Cenciarelli, C., Chiaur, D.S., Guardavaccaro, D., Parks, W., Vidal,
M., and Pagano, M. (1999). Identification of a family of human F box
proteins. Curr. Biol. 9, 1177–1179.
Charles, J.F., Jaspersen, S.L., Tinker-Kulberg, R.L., Hwang, L., Szi-
don, A., and Morgan, D.O. (1998). The Polo-related kinase Cdc5
activates and is destroyed by the mitotic cyclin destruction machin-
ery in S. cerevisiae. Curr. Biol. 8, 497–507.
Clute, P., and Pines, J. (1999). Temporal and spatial control of cyclin
B1 destruction in metaphase. Nat. Cell Biol. 1, 82–87.
Dong, X., Zavitz, K.H., Thomas, B.J., Lin, M., Campbell, S., and
Zipursky, S.L. (1997). Control of G1 in the developing Drosophila
eye: rca1 regulates Cyclin A. Genes Dev. 11, 94–105.
den Elzen, N., and Pines, J. (2001). Cyclin A is destroyed in prometa-
phase and can delay chromosome alignment and anaphase. J. Cell
Biol. 153, 121–135.
Fang, G., Yu, H., and Kirschner, M.W. (1998a). Direct binding of
CDC20 protein family members activates the anaphase-promoting
complex in mitosis and G1. Mol. Cell 2, 163–171.
Fang, G., Yu, H., and Kirschner, M.W. (1998b). The checkpoint pro-
tein MAD2 and the mitotic regulator CDC20 form a ternary complex
with the anaphase-promoting complex to control anaphase initia-
tion. Genes Dev. 12, 1871–1883.
Freed, E., Lacey, K.R., Huie, P., Lyapina, S.A., Deshaies, R.J.,
tin ligase localize to the centrosome and regulate the centrosome
duplication cycle. Genes Dev. 13, 2242–2257.
Geley, S., Kramer, E., Giefferes, C., Gannon, J., Peters, J.-M., and
Hunt, T. (2001). Anaphase-promoting Complex/Cyclosome-depen-
dent proteolysis of human cyclin A starts at the beginning of mitosis
and is not subject to the spindle assembly checkpoint. J. Cell Biol.
Glotzer, M., Murray, A.W., and Kirschner, M.W. (1991). Cyclin is
degraded by the ubiquitin pathway. Nature 349, 132–138.
Gmachl, M., Gieffers, C., Podtelejnikov, A.V., Mann, M., and Peters,
J.M. (2000). The RING-H2 finger protein APC11 and the E2 enzyme
UBC4 are sufficient to ubiquitinate substrates of the anaphase-
promoting complex. Proc. Natl. Acad. Sci. USA 97, 8973–8978.
Gorbsky, G.J., Chen, R.H., and Murray, A.W. (1998). Microinjection
of antibody to Mad2 protein into mammalian cells in mitosis induces
premature anaphase. J. Cell Biol. 141, 1193–1205.
Holloway, S.L., Glotzer, M., King, R.W., and Murray, A.W. (1993).
Anaphase is initiated by proteolysis rather than by the inactivation
of maturation-promoting factor. Cell 73, 1393–1402.
Huang, J., and Raff, J.W. (1999). The disappearance of cyclin B at
the end of mitosis is regulated spatially in Drosophila cells. EMBO
J. 18, 2184–2195.
Hwang, L.H., Lau, L.F., Smith, D.L., Mistrot, C.A., Hardwick, K.G.,
Hwang, E.S., Amon, A., and Murray, A.W. (1998). Budding yeast
Cdc20: a target of the spindle checkpoint. Science 279, 1041–1044.
Anti-Emi1 rabbit polyclonal or PI sera were covalently coupled to
protein A Affiprep beads (Bio-Rad). Beads were washed (20 mM
HEPES [pH 7.7] and 100 mM KCl), incubated with cycling extracts
(0.3 ?l beads/?l extracts, 45 min, 4?C), and samples cleared (3 min,
3000 rpm, 4?C). The process was repeated 2 more times (30 min,
depleted extracts were preincubated with 0.2 volumes undepleted
extract, 300 nM MBP-Emi1, 0.3 ?l depletion beads/?l extract, 60
ng/?l ?90, or 0.3 mg/ml GST-Mad2 (10 min, 4?C) prior to cycling.
Tissue Culture, Immunofluorescence, and Flow Cytometry
Fugene 6 reagent was used for transfections (Roche Molecular Bio-
chemicals). pEGFP-C1 (Clontech), and myc-Emi1 constructs were
cotransfected (1:10). Ninety-eight percent of GFP-positive in-
terphase cells were also myc labeled. Cells were processed for
immunofluorescence or flow cytometry 72 hr post-transfection. Im-
munofluorescence: cells were grown on cover slips and fixed in
methanol (?20?C) or 2% paraformaldehyde with similar results.
(Freed et al., 1999), and blocked in IFWB with 5% normal donkey
serum. 1? Antibodies were: affinity-purified anti-Emi1 (1.5 ?g/ml);
tant; 1:1); anti-myc mAb 9E10 (1 ?g/ml). Texas Red or fluorescein-
conjugated donkey 2? antibodies (Jackson Immunoresearch) were
used at 1:150, and Hoechst dye at 5 ?g/ml. Fluorescent cells were
visualized and digitally imaged or examined by deconvolution mi-
croscopy (Freed et al., 1999). Flow cytometric analysis of PI-stained
using MultiCycle AV software (Phoenix Flow Systems, Inc.)
We are very grateful to S. Geley and T. Hunt for cyclin A and B
antibodies and unpublished results, H.-T. Yu and G. Fang for APC
reagents, M. Kirschner for geminin and securin cDNAs, M. Dobles
and P. Sorger for GST-Mad2 cDNA, C. Crumpton for flow cytometry,
Cell Cycle Regulation by Emi1
Jackson, P.K., Chevalier, S., Philippe, M., and Kirschner, M.W.
(1995). Early events in DNA replication require cyclin E and are
blocked by p21CIP1. J. Cell Biol. 130, 755–769.
Jackson, P.K., Eldridge, A.G., Freed, E., Furstenthal, L., Hsu, J.Y.,
Kaiser, B.K., and Reimann, J. D. (2000). The lore of the RINGs:
substrate recognition and catalysis by ubiquitin ligases. Trends Cell
Biol. 10, 429–439.
Juang, Y.L., Huang, J., Peters, J.M., McLaughlin, M.E., Tai, C.Y.,
and Pellman, D. (1997). APC-mediated proteolysis of Ase1 and the
morphogenesis of the mitotic spindle. Science 275, 1311–1314.
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/anaphase-promoting
complex, and is involved in regulating anaphase onset and late
mitotic events. J. Cell Biol. 141, 1393–1406.
Kim, S.H., Lin, D.P., Matsumoto, S., Kitazono, A., and Matsumoto,
T. (1998). Fission yeast Slp1: an effector of the Mad2-dependent
spindle checkpoint. Science 279, 1045–1047.
King, R.W., Peters, J.M., Tugendreich, S., Rolfe, M., Hieter, P., and
Kirschner, M.W. (1995). A 20S complex containing CDC27 and
CDC16 catalyzes the mitosis-specific conjugation of ubiquitin to
cyclin B. Cell 81, 279–288.
Kotani, S., Tanaka, H., Yasuda, H., and Todokoro, K. (1999). Regula-
tion of APC activity by phosphorylation and regulatory factors. J.
Cell Biol. 146, 791–800.
Kramer, E.R., Gieffers, C., Ho ¨lzl, G., Hengstschla ¨ger, M., and Peters,
J.M. (1998). Activation of the human anaphase-promoting complex
by proteins of the CDC20/Fizzy family. Curr. Biol. 8, 1207–1210.
Kramer, E.R., Scheuringer, N., Podtelejnikov, A.V., Mann, M., and
Peters, J.M. (2000). Mitotic regulation of the APC activator proteins
CDC20 and CDH1. Mol. Biol. Cell 11, 1555–1569.
Li, Y., Gorbea, C., Mahaffey, D., Rechsteiner, M., and Benezra, R.
(1997). MAD2 associates with the cyclosome/anaphase-promoting
complex and inhibits its activity. Proc. Natl. Acad. Sci. USA 94,
Lorca, T., Castro, A., Martinez, A.M., Vigneron, S., Morin, N., Sigrist,
S., Lehner, C., Dore ´e, M., and Labbe ´, J.C. (1998). Fizzy is required
J. 17, 3565–3575.
Meijer, L., Arion, D., Golsteyn, R., Pines, J., Brizuela, L., Hunt, T.,
and Beach, D. (1989). Cyclin is a component of the sea urchin egg
M-phase specific histone H1 kinase. EMBO J. 8, 2275–2282.
Minshull, J., Blow, J.J., and Hunt, T. (1989). Translation of cyclin
mRNA is necessary for extracts of activated xenopus eggs to enter
mitosis. Cell 56, 947–956.
Murray, A.W. (1991). Cell cycle extracts. Methods Cell Biol. 36,
Murray, A.W., and Kirschner, M.W. (1989). Cyclin synthesis drives
the early embryonic cell cycle. Nature 339, 275–280.
new subunits and regulators. Annu. Rev. Biochem. 68, 583–609.
Pfleger, C.M., and Kirschner, M.W. (2000). The KEN box: an APC
recognition signal distinct from the D box targeted by Cdh1. Genes
Dev. 14, 655–665.
Regan-Reimann, J.D.,Duong, Q.V.,and Jackson, P.K.(1999). Identi-
fication of novel F box proteins in Xenopus laevis. Curr. Biol. 9,
Rudner, A.D., and Murray, A.W. (2000). Phosphorylation by Cdc28
activates the Cdc20-dependent activity of the anaphase-promoting
complex. J. Cell Biol. 149, 1377–1390.
Schwab, M., Lutum, A.S., and Seufert, W. (1997). Yeast Hct1 is a
regulator of Clb2 cyclin proteolysis. Cell 90, 683–693.
Shirayama, M., Zachariae, W., Ciosk, R., and Nasmyth, K. (1998).
Saccharomyces cerevisiae. EMBO J. 17, 1336–1349.
Shteinberg, M., Protopopov, Y., Listovsky, T., Brandeis, M., and
Hershko, A. (1999). Phosphorylation of the cyclosome is required
point in budding yeast. Methods Enzymol. 283, 425–440.
Taylor, S.S., and McKeon, F. (1997). Kinetochore localization of mu-
rine Bub1 is required for normal mitotic timing and checkpoint re-
sponse to spindle damage. Cell 89, 727–735.
Tugendreich, S., Tomkiel, J., Earnshaw, W., and Hieter, P. (1995).
CDC27Hs colocalizes with CDC16Hs to the centrosome and mitotic
spindle and is essential for the metaphase to anaphase transition.
Cell 81, 261–268.
van der Reijden, B.A., Erpelinck-Verschueren, C.A., Lo ¨wenberg, B.,
and Jansen, J.H. (1999). TRIADs: a new class of proteins with a
novel cysteine-rich signature. Protein Sci. 8, 1557–1561.
Visintin, R., Prinz, S., and Amon, A. (1997). CDC20 and CDH1: a
family of substrate-specific activators of APC-dependent proteoly-
sis. Science 278, 460–463.
Wassmann, K., and Benezra, R. (1998). Mad2 transiently associates
with an APC/p55Cdc complex during mitosis. Proc. Natl. Acad. Sci.
USA 95, 11193–11198.
M.W. (1998). Identification of a cullin homology region in a subunit
of the anaphase-promoting complex. Science 279, 1219–1222.
Zachariae, W., and Nasmyth, K. (1999). Whose end is destruction:
cell division and the anaphase-promoting complex. Genes Dev. 13,
Zachariae, W., Schwab, M., Nasmyth, K., and Seufert, W. (1998).
Control of cyclin ubiquitination by CDK-regulated binding of Hct1
to the anaphase promoting complex. Science 282, 1721–1724.
The sequence of the Emi1 protein reported in this paper has
been deposited in the Protein Data Bank with accession number