1373
Journal of Cell Science 110, 1373-1386 (1997)
Printed in Great Britain © The Company of Biologists Limited 1997
JCS8154
Microtubule-entrained kinase activities associated with the cortical
cytoskeleton during cytokinesis
Gary R. Walker*, Charles B. Shuster and David R. Burgess†
234 Langley Hall, Department of Biological Sciences, University of Pittsburgh, Pittsburgh, PA 15260, USA
*Present address: Department of Biological Sciences, Youngstown State University, Youngstown, OH 44555-0001, USA
†Author for correspondence (e-mail: [email protected])
SUMMARY
Research over the past few years has demonstrated the
central role of protein phosphorylation in regulating
mitosis and the cell cycle. However, little is known about
how the mechanisms regulating the entry into mitosis contribute to the positional and temporal regulation of the
actomyosin-based contractile ring formed during cytokinesis. Recent studies implicate p34cdc2 as a negative
regulator of myosin II activity, suggesting a link between
the mitotic cycle and cytokinesis. In an effort to study the
relationship between protein phosphorylation and cytokinesis, we examined the in vivo and in vitro phosphorylation
of actin-associated cortical cytoskeletal (CSK) proteins in
an isolated model of the sea urchin egg cortex. Examination of cortices derived from eggs or zygotes labeled with
32P-orthophosphate reveals a number of cortex-associated
phosphorylated proteins, including polypeptides of 20, 43
and 66 kDa. These three major phosphoproteins are also
detected when isolated cortices are incubated with
[32P]ATP in vitro, suggesting that the kinases that phos-
phorylate these substrates are also specifically associated
with the cortex. The kinase activities in vivo and in vitro
are stimulated by fertilization and display cell cycledependent activities. Gel autophosphorylation assays,
kinase assays and immunoblot analysis reveal the presence
of p34cdc2 as well as members of the mitogen-activated
protein kinase family, whose activities in the CSK peak at
cell division. Nocodazole, which inhibits microtubule
formation and thus blocks cytokinesis, significantly delays
the time of peak cortical protein phosphorylation as well as
the peak in whole-cell histone H1 kinase activity. These
results suggest that a key element regulating cortical contraction during cytokinesis is the timing of protein kinase
activities associated with the cortical cytoskeleton that is in
turn regulated by the mitotic apparatus.
INTRODUCTION
midzone in cytokinesis (Cao and Wang, 1996; Wheatley and
Wang, 1996). However, these findings are in contrast to recent
results in which mechanical disruption or removal of the
spindle midzone have no effect on cytokinesis (Zhang and
Nicklas, 1996). While the exact role of astral microtubules (or
spindle midzone) remains unknown, the recruitment of chromosomal passenger proteins such as the TD-60 antigen and
CENPIN to the equatorial zone underscore how the timing and
positioning of cleavage furrow formation is tightly coordinated
with chromatid separation during mitosis (see reviews by
Earnshaw and Bernat, 1991; Margolis and Andreasson, 1993).
While the identity of the stimulus for contractile ring
formation remains unknown, the positional information
imparted by the mitotic apparatus must be tightly coordinated
with the mechanisms that regulate the cell cycle. The timing
of mitosis in embryonic cells is coordinated by maturation
promoting factor (MPF), a complex comprised of p34cdc2
kinase and cyclin B that drives the G2/M transition (reviewed
by Nurse, 1990). p34cdc2 is thought to facilitate the remodeling of the three filament systems during the onset of mitosis
by phosphorylating a number of cytoskeletal substrates (Lamb
et al., 1990). These include nuclear lamins (Ward and
Cytokinesis in animal cells represents the terminal event of
mitosis in which cytoplasm is partitioned into two daughter
cells through the constriction of an actomyosin contractile ring
(see reviews by Mabuchi, 1986; Satterwhite and Pollard, 1992;
Fishkind and Wang, 1995). Micro-manipulation experiments
using echinoderm eggs have demonstrated that the position and
induction of the cleavage furrow is influenced by the mitotic
apparatus (Rappaport, 1986, 1996). This positional information is imparted to the cortex via the astral microtubules during
early anaphase. The stimulus emanates from the spindle poles
at a rate of 6-8 µm/minute (Rappaport, 1973), suggesting that
the delivery of the signal is mediated by microtubule-based
molecular motors. Supportive evidence for a kinesin family
member in transmitting the signal has come from antibody
injection experiments, where antibodies specific to kinesin
family members result in mitotic errors and a failure to
complete cytokinesis (Wright et al., 1993). While experiments
in spherical echinoderm eggs have implicated the astral array
as being the primary determinant of furrow positioning, recent
experiments in cultured cells propose a role for the spindle
Key words: Cell cycle, Cytokinesis, Protein phosphorylation,
Cytoskeleton
1374 G. R. Walker, C. B. Shuster and D. R. Burgess
Kirschner, 1990), microtubule-associated proteins such as
MAP4 (Ookata et al., 1995), motor proteins such as Eg5 and
CENP E (reviewed by Vernos and Karsenti, 1996), and
vimentin (Chou et al., 1990). Caldesmon and non-erythroid
spectrin have also been shown to be phosphorylated by p34cdc2,
and this phosphorylation is accompanied by a concomitant
reduction of caldesmon/spectrin association with the actin
cytoskeleton (Yamashiro et al., 1991; Fowler and Adam, 1992).
It is postulated that phosphorylation of these proteins results in
the massive reorganization of the cytoskeleton that accompanies cell division.
With regard to cytokinesis, the regulation of myosin II
activity represents one of the most relevant examples of cell
cycle regulation of an actin-associated protein during mitosis.
Studies of myosin II light chain phosphorylation in higher
eukaryotes reveal that p34cdc2 kinase phosphorylates the regulatory light chain (Satterwhite et al., 1992; Yamakita et al.,
1994) on residues previously shown to inhibit myosin ATPase
activity in vitro (Nishikawa et al., 1984; Bengur et al., 1987).
From these findings, it has been proposed that p34cdc2 inhibits
myosin activity until anaphase when cyclin B destruction
would result in a reduction of p34cdc2 activity (Satterwhite and
Pollard, 1992). At this time, MLCK may then stimulate myosin
filament formation and ATPase activity. Thus, the regulation of
p34cdc2 activity is hypothesized to provide the ‘timer’ that
regulates cortical contraction. Recent genetic analysis of the
light chain phosphorylation sites in Dictyostelium (Ostrow et
al., 1994), however, suggests that myosin II regulation in lower
eukaryotes may be more complicated than that proposed in the
Satterwhite and Pollard model (Satterwhite and Pollard, 1992).
There is an increasing body of experimental evidence to
suggest that protein kinases coordinate the transition from
mitosis to cytokinesis, and may also control the position and
organization of the contractile ring. Studies using an antibody
against a phosphoepitope of glial fibrillary acidic protein
(GFAP) reveal the presence of a protein kinase activity that is
not active until the metaphase-anaphase transition (Sekimata et
al., 1996). In Saccharomyces cerevisiae the Dbf2/Dbf20
protein kinases are activated upon dephosphorylation
following the metaphase-anaphase transition (Toyn and
Johnston, 1994). Polo kinase, however, may represent the most
attractive candidate for a late mitotic kinase that acts to
regulate spindle-cortical dynamics. Originally identified as
a gene required for proper spindle assembly in Drosophila
(Llamazares et al., 1991), polo is highly conserved throughout
the phylogenetic tree (Ohkura et al., 1995; Golsteyn et al.,
1995). Polo kinase activity is activated at the G2/M transition,
but extends beyond the metaphase-anaphase transition (Fenton
and Glover, 1993; Golsteyn et al., 1995). Polo mutants in S.
pombe result in elongated spindles, and a failure in actin ring
assembly and septum formation (Ohkura et al., 1995). Localization of human polo-like kinase (PLK1) in tissue culture cells
reveals that upon anaphase onset, PLK1 translocates from the
mitotic spindle to the equatorial plane (Golsteyn et al., 1995).
While the substrates of this kinase remain unknown, studies in
mammalian cells indicate that PLK1 interacts with
CHO1/MKLP-1, a kinesin-like molecule (Lee et al., 1995), and
the Xenopus homolog phosphorylates the p34cdc2-activating
phosphatase, cdc25 (Kumagai and Dunphy, 1996). Thus, while
polo kinases may be essential to spindle regulation, the
extended activity of polo during mitosis (in comparison to
p34cdc2), its redistribution to the equatorial zone, and its association with microtubule motors make it an attractive candidate
for the microtubule-based regulation of cortical remodeling
during the onset of cytokinesis.
Accumulating genetic and biochemical evidence point
towards a role for protein kinases not only in regulating cortical
remodeling at the G2/M transition, but also in the formation
and induction of the contractile ring. However, the identity of
the kinases, their substrates within the cortical cytoskeleton,
and the functional consequences of this regulation remain
elusive. Previous work in this laboratory developed a model of
the sea urchin zygote cortical actin cytoskeleton that retains the
biochemical and functional characteristics of the intact,
cleaving blastomere (Walker et al., 1994). In this study, we
employed this detergent-extracted preparation of the sea urchin
zygote to ask whether there are elements of the isolated cortex
subject to regulation by kinases acting late in mitosis. Results
of these studies indicate there are three major polypeptides
(p20, p43 and p66) that are phosphorylated by kinases associated with the actin cortex, and 2-dimensional electrophoresis
and immunoprecipitation studies reveal that p20 is myosin
regulatory light chain. As for the kinases that phosphorylate
these substrates, the combined approaches of phosphoamino
acid analyses, in-gel kinase assays, western blotting, and
phenyl-Sepharose chromatography reveal the presence of
multiple kinases associated with the actin cortex. Of these
cortical kinases, p34cdc2 and a homolog of mitogen-activated
protein kinase (MAP kinase), demonstrate cyclic fluctuations
in activities (but not levels) through the cell cycle. The activation of cortical protein phosphorylation is sensitive to nocodazole-, but not to staurosporine-treatment of cleaving embryos,
suggesting that the recruitment or activation of cortical kinase
activity is dependent upon the integrity of the mitotic
apparatus, but not protein kinase C (PKC). The cell cycle- and
microtubule-dependent phosphorylation of these cortical substrates suggest that these polypeptides may represent targets of
multiple regulatory mechanisms, where p34cdc2 and other
cortical kinases regulate the timing and formation of the contractile ring.
MATERIALS AND METHODS
Culture of sea urchin zygotes
The gametes of the California urchin Strongylocentrotus purpuratus
were used for all experiments (Marinus, Long Beach, CA). Shedding
of gametes was induced by the injection of 0.5 M KCl and eggs were
de-jellied by brief exposure to pH 5.0 artificial sea water (ASW)
followed by washing 1× in ASW. The vitelline membrane was then
removed by treatment with DTT in ASW (pH 9.1). Eggs were allowed
to settle and then washed 5× in ASW by settling.
Eggs were fertilized with 10 µl dry sperm for every 1 ml packed
eggs in ASW. The zygotes (1 ml) were cultured in 50 mls of ASW at
15-16°C in a spinner flask after briefly washing out excess sperm. In
some experiments nocodazole at 10 µM (Sigma Co., St Louis, MO)
was added to experimental flasks at 10 minutes after fertilization to
promote microtubule disassembly.
Preparation of sea urchin egg/zygote cytoskeletons
Cortical CSKs were isolated as detailed by Walker et al. (1994).
Briefly, eggs or embryos (1 ml packed cells) were washed once in 10
ml of isolation buffer containing 1 M glycerol, 5 mM EGTA, 5 mM
MgCl2, 10 µM Na3VO4 and 20 mM Pipes (pH 7.3 unless specified as
Cortical cytoskeleton-associated kinases 1375
6.8). Detergent-extracted cytoskeletons (CSK) were then prepared by
lysing the cells with isolation buffer containing 0.5% Nonidet P-40 for
10 minutes at 0°C. Cortical cytoskeletons were then subjected to gentle
homogenization in a loose-fitting Dounce homogenizer after detergent
extraction. The cytoskeletal material was washed three times in
isolation buffer and pelleted by centrifugation at 3,000 g for 10 minutes
4°C. The pellets were resuspended to 1 ml for phosphorylation studies.
In vivo protein phosphorylation
De-jellied and de-membranated eggs were washed with phosphatefree synthetic sea water (PFSSW). The eggs were then pre-incubated
for 1 hour on ice with 2.5 mCi/ml 32P-orthophosphate in an egg suspension of 1 ml packed eggs in 2 ml PFSSW. Eggs were then diluted
into 50 ml of PFSSW (unfertilized) or 50 ml of PFSSW/sperm suspension (zygote or embryo) and cultured in spinner flasks.
In vitro protein phosphorylation
Cytoskeletal proteins were labeled endogenously by addition of [γ32P]ATP. The reaction was carried out by addition of 200 µl cortex
cytoskeleton suspension (in isolation buffer) to a mixture of 20 µl (0.25
mM) cold ATP, 4 µl [32P]ATP (22.5 µCi) and 1 µl 1 M CaCl2 (free
Ca2+= 23.23 µM). For most experiments the reaction was allowed to
continue for 10 minutes at room temperature and then stopped by
addition of 80 µl hot 4× SDS-sample buffer and boiled for 2 minutes.
Phosphoamino acid analysis
SDS-PAGE gels of 32P-labeled CSK were electrophoretically transferred to Immobilon P membranes (Towbin et al., 1979). The bands
of interest were excised from the blot, cut into pieces and placed into
1.5 ml Eppendorf microfuge tubes. The proteins were then hydrolyzed
by addition of 200 µl 5.7 N HCl and incubated at 110°C for 1 hour.
The hydrolysate was removed and lyophilized. The amino acid
samples were then dissolved in pH 1.9 buffer containing 2.2% formic
acid and 1.3611 M acetic acid supplemented with 0.2 mg/ml each of
phosphoserine, phosphothreonine and phosphotyrosine. Electrophoretic separation of individual amino acids was achieved by thin
layer electrophoresis on cellulose thin layer chromatography plates
(Boyle et al., 1991).
In situ kinase assay
To identify kinases associated with the cortical cytoskeleton, cortices
were resolved by SDS-PAGE, the proteins renatured and subjected to
in situ phosphorylation (Hutchcroft et al., 1991). Briefly, SDS was
removed by washing gels 2× in 20% propanol, 50 mM Tris-HCl, pH
8.3, for 30 minutes each, followed by washing in 50 mM Tris-HCl,
pH 8.3, 5 mM β-ME for 1 hour. Proteins were then denatured by incubating gels in 7 M guanidine, 50 mM Tris-HCl, pH 8.3, for 1 hour at
room temperature. The proteins were slowly renatured by washing 4×
in 300 ml of 50 mM Tris, 5 mM β-ME, 0.04% Tween-40 over 18
hours at 4°C. The gel was then incubated in 100 ml of pre-incubation
buffer (40 mM Hepes, pH 8.0, 10 mM MgCl2, 0.5 mM EGTA, 2 mM
DTT) for 1 hour at 20°C. The phosphorylation of endogenous
renatured kinases was carried out in phosphorylation buffer containing 40 mM Hepes, pH 8.0, 0.5 mM EGTA, 10 mM MgCl2, and 100
µCi [32P]ATP for 2 hours at 20°C. The free 32P was removed by
washing gels 4× 500 ml in 5% TCA, 1% sodium pyrophosphate. The
gels were then stained with Coomassie Blue R-250, dried and the
labeled proteins visualized by autoradiography.
Histone H1 kinase assay
To detect histone H1 kinase activity, resuspended CSK (200 µl in
isolation buffer) were incubated with 20 µl of [32P]ATP/ATP mix
(final=1.5625×10−5 M, 50 µCi), 6.4 µl protein kinase inhibitor (0.08
µM) to inhibit PKA activity, and 20 µl (20 µg) histone H1
(Boehringer-Mannheim, Indianapolis, IN) for 10 minutes at 25°C. In
some assays myelin basic protein (MBP) was added (40 µg) to simultaneously assay for MAP kinase activity. The reaction was stopped by
addition of 1/4 volume 4× SDS-PAGE sample buffer. The samples
were run on SDS-PAGE and autoradiographed. Samples for whole
egg activity were prepared by lysing eggs with 10 volumes of distilled
water followed by freeze thawing.
Immunoblot analysis
SDS-PAGE gels were run and proteins transferred to Immobilon-P
membranes at 24 volts for 90 minutes in a Genie blotting apparatus
(Idea Scientific; Minneapolis, MN). The blots were blocked with 5%
non-fat dry milk, 0.2% Tween-20, TBS for 1 hour at 20°C. The blots
were then incubated with either a 1/2,000 dilution of rabbit anti-MAP
kinase R1(erk1-domain III; Upstate Biotechnology, Lake Placid, NY)
or a 1/1,000 dilution of rabbit anti-human cdc2 kinase (PSTAIR;
Upstate Biotechnology, Lake Placid, NY) in 1% non-fat dry milk,
0.04% Tween-20, TBS for 1 hour at 20°C or overnight at 4°C. Tubulin
was detected by using a 1/2 dilution of a monoclonal mouse anti-alpha
tubulin tissue culture supernatant (generous gift from Dr Charles
Walsh, University of Pittsburgh). Blots were washed 3 times with TBS
containing 0.2% Tween-20. Bound antibodies were dectected by incubation with HRP-conjugated goat anti-rabbit IgGs for 1 hour at room
temperature and blots were visualized using Enhanced Chemiluminescence (Amersham; Arlington Heights, IL).
To quantify the relative fraction of kinases associated with the actin
cortex, cortices were prepared from unfertilized eggs, interphase or
dividing zygotes or, alternatively, whole eggs were placed directly into
hot sample buffer. Equal amounts of cortical and whole egg protein
were resolved by SDS-PAGE, transferred to nitrocellulose, and
probed with antibodies against p34cdc2, MAP kinase, fyn (Upstate
Biochemical, Lake Placid, NY), β subunit CAM kinase II (Zymed,
San Francisco, CA), kinesin heavy chain (provided by John Scholey,
University of California at Davis), or lamin B (provided by Gary
Wessel, Brown University).
MBP kinase preparation
MBP kinase was purified from early sea urchin embryos by the
method of Sanghera et al. (1992). Embryos at first cleavage were
homogenized in 3 ml homogenization buffer containing 40 mM
MOPS, pH 7.2, 120 mM sodium β-D-glycerol phosphate, 10 mM
EGTA, 2 mM EDTA, 2 mM sodium orthovanadate with a PotterElvehjem homogenizer on ice. The sample was clarified by centrifugation at 115,000 g for 60 minutes. The supernatant was recovered
and passed through a 0.45 µm filter. Samples (1 ml) were applied to
a phenyl-Sepharose column (Protein Pak Glass HIC phenyl-5PW,
Waters/Millipore; Bedford, MA). After washing unbound protein
from the column, bound proteins were eluted with a 7 ml linear
gradient of 0-60% ethylene glycol, 250-25 mM NaCl at 0.2
ml/minute. Fractions were then assayed for MBP and H1 kinase activities, and probed for the presence of MAP kinases and p34cdc2 by
western blotting.
Substrate-blot assay
To identify candidate cortical substrates for MBP kinase and p34cdc2, a
substrate blot assay was devised. Briefly, SDS-PAGE gels were run and
proteins transferred to Immobilon-P. The blots were dried and then
blocked with 5% BSA, 30 mM Tris-HCl, pH 7.5, for 1 hour. The blots
were rinsed in reaction buffer (30 mM Tris-HCl, pH 7.35, 10 mM
MgCl2, 2 mM MnCl2). The blots were labeled by incubation with 25
ml reaction buffer containing 250 µCi (2.2 nM) [γ-32P]ATP and either
sea urchin MAP kinase or p34cdc2/cyclin B complex (Uptsate Biomedical, Lake Placid, NY). Parallel control blots were prepared and
incubated in reaction buffer containing 250 µCi (2.2 nM) [γ-32P]ATP
with no exogenously added kinase. The blots were incubated at room
temperature for 60 minutes with agitation. The blots are then washed
twice with TBS/0.05%NP-40 and then twice with TBS. The blots were
then incubated with 1 N KOH for 10 minutes followed by a brief 1 N
KOH rinse. The KOH was removed by washing twice with TBS. Phosphorylated proteins were detected by autoradiography of the dried blots.
1376 G. R. Walker, C. B. Shuster and D. R. Burgess
Other methods
Incorporation of 32P into peptides was quantified by radioanalytical
imaging analysis of Coomassie-stained gels using an Ambis radioanalytical imaging system (AMBIS Systems Inc., San Diego, CA),
directly by scintillation counting, or by scanning densitometry. Actin
was quantified by scanning densitometry of Coomassie-stained gels.
Immunofluorescence microscopy was performed as described previously (Walker et al., 1994). Microtubules were localized with antibodies against α-tubulin and visualized with Cy3-conjugated goat
anti-mouse IgG (Chemicon; Temicula, Ca).
Electrophoretic analysis of isolated cortex proteins was performed
according to the procedure of Laemmli (1970) using 10 or 12% acrylamide gels. SDS polyacrylamide minislab gels were run according to
the procedure of Matsudaira and Burgess (1978). Approximate
molecular masses were determined using unstained or pre-stained
molecular mass standards (Sigma Co, St Louis, MO).
RESULTS
Cytoskeletal protein phosphorylation in the isolated
sea urchin cortex
The cortical cytoskeleton (CSK) prepared in this study is an
actin-rich, detergent-insoluble cytoskeleton that retains the
basic morphology of the original blastomeres (Fig. 1.), but
lacks the overlaying plasma membrane. This preparation,
whose structural, biochemical, and functional characteristics
are described more fully elsewhere (Walker et al., 1994),
possesses the known egg cortical cytoskeletal proteins actin,
myosin II, spectrin, and fascin. As shown by Coomassie
staining, actin is the most predominant protein in the CSK
preparation (Fig. 1C, lane 3). Electron microscopy of isolated
cortices reveals that the cortex consists of an anastomosing
meshwork of filaments that retains microvillar actin cores as
well as the hyaline layer (Walker et al., 1994).
To explore the role of kinases in modulating the actin
cytoskeleton during cytokinesis, whole eggs or zygotes were
labeled with 32P-orthophosphate and the phosphoproteins associated with the isolated cortices compared with those found in
the whole egg (Fig. 1C). While whole cell samples from unfertilized eggs contain numerous phosphorylated polypeptides
(Fig. 1C, lane 2), the vast majority of these polypeptides are
not associated with the actin cortex (Fig. 1C, lane 4). In
contrast, cortices isolated from interphase (Fig. 1C, lane 5) or
cleavage-stage zygotes (Fig. 1C, lane 6) following in vivo
labeling reveal a distinct subset of whole egg or embryo phosphoproteins present in the isolated actin cortex. While some
high molecular mass phosphoproteins are shared by dividing
embryos and unfertilized eggs, cortices derived from dividing
blastomeres contain phosphoproteins not evident in the unfertilized egg CSK, including phosphoproteins co-migrating with
actin (~43 kDa) and serum albumin (66 kDa). Other phosphoproteins associated with the cortex in unfertilized and cleaving
eggs include a 20 and 36 kDa species. For all four phosphoproteins, there is a dramatic increase in phosphate incorporation when cortices are isolated from cleavage-stage embryos.
Protein kinase activities are associated with the
cortical cytoskeleton
To ask whether kinase activities are themselves associated with
the cortical cytoskeleton, cortices were derived from zygotes,
incubated in vitro with [γ-32P]ATP, and the phosphoproteins
Fig. 1. Isolation of cortical cytoskeletons and in vivo
phosphorylation of cortical proteins. (A-B) Phase contrast
micrographs of intact (A), and detergent-extracted (B) cleaving
blastomeres. Bar, 10 µm. (C) In vivo labeling of unfertilized eggs
and fertilized zygotes. Unfertilized eggs were incubated in the
presence of 32P-orthophosphate prior to fertilization and preparation
of detergent-resistant cortices (CSK). Lane 1, total egg protein;
Coomassie stain. Lane 2, total egg protein; autoradiogram. Lane 3,
CSK, unfertilized egg; Coomassie stain. Lane 4, CSK, unfertilized
egg; autoradiogram. Lane 5, CSK, interphase zygote; autoradiogram.
Lane 6, CSK, cleavage-stage zygote; autoradiogram.
analyzed by SDS-PAGE. Control experiments indicate that
peak in vitro 32P incorporation into cortical protein occurs
approximately 10 minutes after initiation of phosphorylation in
both unfertilized and cleavage stage cortices (data not shown).
Incubations longer that 10 minutes result in reduced 32P incorporation, possibly due to the actions of phosphatases associated with the cortex. Having determined the optimum conditions for phosphate incorporation, all subsequent in vitro
phosphorylation reactions were carried out for 10 minutes.
Examination of in vitro-labeled cortices reveals that p20,
p43, and p66 are phosphorylated in cortices derived from
cleavage-stage embryos (Fig. 2, lanes 3 and 6). In contrast, the
36 kDa polypeptide observed in in vivo-labeled eggs and
zygotes is not observed in the in vitro-labeled cortices. The
highest level of protein phosphorylation is observed in cortices
isolated from cleavage-stage blastomeres (Fig. 2, lanes 3 and
6) in comparison to those from unfertilized eggs (Fig. 2, lanes
1 and 4) or from interphase zygotes (Fig. 2, lanes 2 and 5).
Because pH-dependent protein tyrosine kinase activities have
been reported in sea urchin eggs, in vitro phosphorylation
Cortical cytoskeleton-associated kinases 1377
kDa
Fig. 2. Identification of in vitro phosphorylated cortical proteins.
Detergent-resistant cortices were derived from unfertilized eggs
(lanes 1 and 4), interphase zygotes (lanes 2 and 5), or cleavage-stage
blastomeres (lanes 3 and 6). Cortices were isolated and incubated in
the presence of [γ-32P]ATP, at pH 6.8 (lanes 1-3) or pH 7.3 (lanes 46). Cortical protein was then analyzed by SDS-PAGE and
autoradiography. The major polypeptide species phosphorylated in
vitro include those of 66, 43, and the 20 kDa (arrows).
reactions were carried out at pH values corresponding to the
cytoplasmic pH of unfertilized (pH 6.8; Fig. 2, lanes 1-3) or
fertilized (Fig. 2; lanes 4-6) eggs (Jiang et al., 1990). Slight
quantitative, but no qualitative differences in protein phosphorylation are observed in cortices phosphorylated in vitro at pH
6.8 versus, pH 7.3, suggesting that the pH-dependent protein
tyrosine kinase may not be the kinase phosphorylating
cleavage-stage cortical polypeptides. The similar patterns of
phosphorylation observed between in vivo and in vitro-labeled
cortices suggest that the kinases phosphorylating these substrates during cytokinesis are, in fact, specifically associated
with the cortical cytoskeleton.
Changes in cortical protein phosphorylation during
the cell cycle
To follow the phosphorylation of p20, p43, and p66 through
the cell cycle, cortices were isolated from embryos at different
stages through the first two cell cycles, and labeled in vitro. As
shown in Fig. 3, the level of p43 and p20 phosphorylation is
highest in cortices isolated from cleavage-stage blastomeres
(Fig. 3, lane 4), decreases in interphase isolated cortices (Fig.
3, lanes 5 and 6) and is again elevated at the time of the second
cleavage (Fig. 3, lane 7). A more variable pattern of in vitro
phosphorylation of p66 occurs following 90 minutes post-fertilization (lane 4), with no consistent cycling observed
following the first cleavage.
Identification and partial characterization of in vitrolabeled phosphoproteins
Biochemical as well as genetic lines of experimentation have
underscored the requirement of myosin II function in cytokinesis (Mabuchi and Okuno, 1977; Kiehart et al., 1982;
DeLozanne and Spudich, 1987; Knecht and Loomis, 1987;
Karess et al., 1991), and we have shown that myosin II is
present in isolated cortices that undergo ATP-dependent contraction in vitro (Walker et al., 1994). Additionally, several
Fig. 3. Cell cycle-dependent in vitro phosphorylation of cortical CSK
proteins. Cortical cytoskeletons prepared at different times following
fertilization were labeled in vitro and the resulting phosphoproteins
analyzed by SDS-PAGE and autoradiography. Lane 1, unfertilized
egg. Lane 2, 50 minutes post-fertilization (interphase). Lane 3, 70
minutes post-fertilization (prophase, 1st division). Lane 4, 90
minutes post-fertilization (anaphase-telophase, 1st division). Lane 5,
120 minutes post-fertilization (interphase, 2 cell stage). Lane 6, 130
minutes post-fertilization (interphase-prophase, 2-cell stage). Lane 7,
150 minutes post-fertilization (anaphase-telophase, 2nd division).
reports indicate that myosin II regulatory light chain (LC 20)
is phosphorylated in dividing cells, and can be phosphorylated
in vivo and in vitro by p34cdc2 kinase (Yamakita et al., 1994;
Satterwhite et al., 1992). To determine whether the 20 kDa
polypeptide phosphorylated in vivo and in vitro is myosin LC
20, cortices were labeled with [32P]ATP in vitro, solubilised,
and anti-sea urchin egg myosin antibodies used to immunoprecipitate myosin II and its associated light chains (Fig. 4A).
Autoradiography of these immunoprecipitations revealed that
myosin LC 20, and to a lesser extent myosin heavy chain, is
phosphorylated in vitro. Confirmation that the phosphorylated
20 kDa protein is LC20 came from two-dimensional
IEF/PAGE, where the 20 kDa phosphoprotein migrates with a
similar mobility (Fig. 4B, lower arrow) as purified egg myosin
II 20 kDa light chain (data not shown).
To partially characterize the other major cortical phosphoproteins, two-dimensional PAGE and non-equilibrium pH gel
electrophoresis were used to determine the isoelectric points of
p43 and p66. A series of phosphorylated species corresponding to the 66 kDa phosphoprotein can be detected focusing at
a slightly more acidic pI (pI ≈5) than actin (Fig. 4B). Conventional two-dimensional IEF/PAGE did not reveal any polypeptides corresponding to the 45 kDa species. However, phosphorylated species corresponding to the 43 kDa polypeptide
can be detected by non-equilibrium pH gel electrophoresis
(NEpHGE) (Fig. 4C), whose pI values were in the region of
≈9.
In light of recent evidence indicating that, in addition to
serine-threonine kinases such as p34cdc2, there are tyrosine
kinases active during mitosis (Roche et al., 1995), phosphoamino acid analysis was performed to determine the relative
distribution of phosphoamino acids in LC 20, p43, and p66. As
shown in Fig. 5, phosphoamino acid analysis of in vitro-labeled
p20, p43, and p66 from cleavage-stage blastomeres reveals that
p66 is evenly phosphorylated on all three phosphoamino acids
(Fig. 5A). Phosphoserine is the major phosphoamino acid in
the 43 kDa polypeptide, which is also phosphorylated to a
lesser extent on threonine and tyrosine (Fig. 5B). Phosphoserine is the major phosphoamino acid found in LC20 (Fig. 5C)
with phosphothreonine present at a much lower level. Thus,
from phosphoamino analysis of in vitro-labeled p20, p43 and
1378 G. R. Walker, C. B. Shuster and D. R. Burgess
Fig. 4. Characterization of in vitro phosphorylated kinase substrates.
(A) In vitro-labeled cortices derived from cleaving blastomeres were
solubilized and immune precipitations were performed using preimmune (lane 1) or anti-egg myosin polyclonal antibodies (lane 2).
Samples were then resolved by SDS-PAGE and visualized by
autoradiography. The arrow indicates the position of the 20 kDa
phosphoprotein. (B) Two-dimensional IEF/SDS-PAGE analysis of in
vitro-labeled cortices. The position of the 20 kDa phosphoprotein
(which has the same relative mobility as myosin LC20, as
determined by IEF/SDS-PAGE of purified egg myosin), is indicated
by the lower arrow. The upper arrow indicates the position of a series
of species corresponding to the 66 kDa polypeptide, with a pI
slightly more acidic than actin (a). (C) Non-equilibrium pH gel
electrophoresis (NEphGE) of cortical CSK phosphopeptides. The
arrow indicates the basic position of the 43 kDa polypeptide.
p66, there appears to be both serine-threonine and tyrosine
kinase activities associated (and active) with the actin cortex
of dividing blastomeres.
Identification of cortical kinase activities
As a first measure toward characterizing candidate kinases
present in the isolated sea urchin cortex, cortices were resolved
by SDS-PAGE, renatured, and assayed for kinase autophosphorylation (Fig. 6A). Three major autophosphorylating
kinases, with approximate molecular masses of 84, 45 and 42
kDa, can be detected by this method in unfertilized and
cleavage-stage cortices (Fig. 6A). These kinases appear
distinct from the major kinase substrates detected by in vivo
and in vitro phosphorylation experiments based on mobility in
SDS-PAGE. Phosphoamino acid analysis of the 45 kDa and 84
kDa kinases indicate that both are autophosphorylated on
threonine and serine residues (data not shown). No evidence of
tyrosine autophosphorylation could be detected by this assay.
When cortical kinases were renatured in PAGE gels containing immobilized myelin basic protein (MBP) as a substrate,
kinase activity is readily detected in CSKs from both unfertilized eggs and cleaving zygotes (Fig. 6B). However, no additional kinase activities are detected by this method. Using this
Fig. 5. Phosphoamino acid analysis of in vitro-labeled LC 20, p43,
and p66. The 66 kDa (A) 43 kDa (B) and myosin LC20 (C)
polypeptides from in vitro-labeled cortices were resolved by SDSPAGE, excised, and analyzed by two-dimensional phosphoamino
acid analysis. (D) The relative positions of phosphoamino acid
standards: pS, phosphoserine; pT, phosphothreonine; pY,
phosphotyrosine.
Fig. 6. Identification of cortical kinase
activities. Sea urchin egg cortices were
resolved by SDS-PAGE, renatured,
and assayed for in situ kinase activity
(A) or myelin basic protein (MBP)
kinase activity (B). Lane 1,
unfertilized egg. Lanes 2-7, 50, 70, 90,
120, 130 and 150 minutes post-fertilization, respectively. The
prominent kinases identified by both autophosphorylation and MBP
kinase activity include polypeptides of 84, 45 and 42 kDa.
(C) Inclusion of exogenously-added histone H1 into in vitro
phosphorylation reactions. Lane 1, unfertilized egg. Lane 2,
interphase zygote. Lane 3, cleaving blastomere. A dramatic increase
in H1 kinase activity is seen in CSKs isolated from cleaving zygotes
(lane 3) relative to equivalent amounts of CSKs isolated from
unfertilized eggs (lane 1) or interphase cells (lane 2).
assay system, the activity of these kinases also fluctuate
through the cell cycle, with the highest activities found in CSKs
isolated from cleavage-stage zygotes (Fig. 6A and B, lanes 4
and 7). The presence of MBP kinase activity was confirmed by
inclusion of MBP in CSK in vitro labeling reactions, where
peak MBP kinase activity can be detected in cortices isolated
immediately prior to cleavage (data not shown).
Cortical cytoskeleton-associated kinases 1379
Cortical CSK phosphorylation is independent of
protein kinase C
Protein kinase C (PKC) is thought to play an important role in
egg activation and mitosis (Mabuchi and Takano-Ohmuro,
1990; Bement and Capco, 1991). To determine if PKC plays a
role in cytokinesis or the phosphorylation of cortex-associated
proteins, zygotes were continuously incubated in staurosporine
beginning at different times following fertilization (Fig. 7).
Examination of staurosporine-treated zygotes reveals, as
expected, that cleavage is sensitive to staurosporine treatment
in a time-dependent manner, with incubations in the drug
starting any time up to 40 minutes after fertilization blocking
cleavage in over 75% of zygotes (Fig. 7A). Progressively later
initiation of staurosporine treatment resulted in a progressively
lesser impact on cleavage. If drug treatments were begun after
about 70 minutes post-fertilization (corresponding to the onset
of mitosis), zygotes divide normally. Hoechst dye was used to
monitor the state of the mitotic cycle in samples of staurosporine treated zygotes (data not shown). Staurosporine
treatment beginning 10 minutes post fertilization resulted in
80% of zygotes remaining in interphase with only 20% having
a telophase stage mitotic apparatus or having cleaved. In
contrast, 47% of zygotes continuously cultured in staurosporine beginning at 70 minutes post fertilization had
cleaved or were in telophase and another 45.5% possessed a
mitotic apparatus in some stage of mitosis.
Cortices were prepared from staurosporine-treated and
control cultures at the time of first cleavage in control cultures
and the in vitro phosphorylation of LC20 and p43 monitored
(Fig. 7B). Continuous treatment of zygotes beginning 10
minutes post fertilization with staurosporine inhibits by over
60% the level of in vitro phosphorylation of p43 and LC20 over
that in controls. In contrast, initiation of staurosporine
treatment at 70 minutes, which delayed entry into mitosis,
results in less than a 40% inhibition of p43 or LC20 phosphorylation relative to that in cortices from control cultures. Consistent with these results was the finding that incubating
isolated cleavage stage cortices in H7, another potent PKC
inhibitor (Hidaka et al., 1984), did not affect the in vitro phosphorylation of p43 or LC20 (data not shown).
Characterization of cell cycle kinases associated
with the cortical cytoskeleton
Results of in vitro labeling as well as in-gel phosphorylation
assays suggest that kinases associated with the cortex may
mediate the phosphorylation of p20, p43 and p66. And since
the phosphorylation of cortical proteins by endogenous kinases
A
80
Percent Cleavage
70
60
50
40
30
20
10
0
20
40
60
80
Time Post-fertilization of Drug Administration (minutes)
B
120
Relative Protein Phosphorylation
Histone H1 serves as a substrate for many kinases, including
the catalytic component of maturating promoting factor,
p34cdc2 (Langan et al., 1989). To ask whether an analogous
activity is associated with cortices derived from dividing sea
urchin blastomeres, histone H1 was mixed with CSKs isolated
from zygotes at different stages in the presence of [γ-32P]ATP,
and the phosphorylated histone H1 resolved by SDS-PAGE and
autoradiography (Fig. 6C). As shown in Fig. 6C, histone H1 is
phosphorylated with a peak in H1 kinase activity in cortices
isolated from cleavage stage embryos (Fig. 6, lane 3). These
results indicate that both H1 and MBP kinase activities
exhibited by the isolated cortices appear to be cyclic in nature,
with both activities peaking at the time of cell division.
43 kD
100
LC 20
80
60
40
20
0
Unfertilized
Stauro
10' PF
Stauro
70' PF
Cleavage
Fig. 7. Effects of staurosporine on the inhibition of cleavage and
phosphorylation of cortical CSK proteins. (A) Sea urchin eggs were
fertilized, and staurosporine was administered to batches at different
times post-fertilization. The zygotes cultured in staurosporine were
fixed after the time of first cleavage (110 minutes, PF), and the
percentage of embryos that underwent cytokinesis determined. (B) In
vitro phosphorylation of p43 (
) and LC20 (
) was determined
in cortices isolated from staurosporine-treated zygotes where the
drug was applied at different time points after fertilization.
appears to be cell cycle-regulated, we sought to identify
specific kinases known to play roles in cell cycle and growth
regulation by western blotting of isolated cortical CSK
proteins. Western blotting with anti-MAP kinase antibodies
reveal the presence of a 49-50 kDa polypeptide present in
cortices throughout the cell cycle (Fig. 8, MAPK). In addition,
the MAP kinase R2 (erk1-CT) antibody recognizes two other
peptides of about 46 kDa and 44 kDa (data not shown). Antibodies to p34cdc2 kinase recognize several species of approximately 34 kDa (Fig. 8, cdc2). Additionally, antibodies specific
for the tyrosine kinase c-fyn, recognize a single 60 kDa species
constitutively associated with the cortical cytoskeleton (data
not shown). In addition, we have recently demonstrated the
presence of e-abl, another tyrosine kinase family member
within the cortical cytoskeleton (Walker et al., 1996). Western
blotting of cortical and whole cell fractions reveals that while
the levels of the 49 kDa MAP kinase, p34cdc2, and fyn in the
isolated cortex increase following fertilization (Fig. 8), statistical analyses indicate that the small changes in the fraction of
kinases associated with the cortex from interphase to cell
1380 G. R. Walker, C. B. Shuster and D. R. Burgess
prophase (Ward and Kirschner, 1990), and therefore represent
another control for nonspecific trapping of soluble molecules.
Western blotting of cortical and whole cell fractions reveals
that while there exists a minor amount of contaminating lamin
B present in the cortices (probably due to the occasional
nucleus contaminating the preparation), the percentage of
lamin present in dividing and interphase cortices does not
appreciably change although lamins are solubilised during
mitosis, and could be trapped within the cortex (Table 1).
Therefore, based on these experiments and controls, it is likely
that the kinase activities detected in the cortex are specifically
associated and are not artifactually trapped.
Fig. 8. Cell cycle-dependent and growth-related kinases are
associated with the cortical CSK. Equivalent amounts of cortical
CSKs derived from: lane 1, unfertilized eggs; lane 2, 50 minutes
post-fertilization (PF); lane 3, 70′ PF; lane 4, 90′ PF (first cleavage);
lane 5, 120′ PF; lane 6, 130′ PF; lane 7, 150′ PF (second cleavage)
were resolved by SDS-PAGE, transferred to nitrocellulose, and
probed for the presence of p49 MAP kinase (top panel), tubulin
(middle panel) and p34cdc2 (bottom panel).
cytokinesis (Table 1) are not significantly different (P≥0.05 for
all).
Although we found that the cortex represents less than 2%
of total egg protein, experiments were performed to determine
whether the cortex-associated kinase activities were due to
non-specific trapping (Fig. 8; Table 1). Tubulin, which is
largely soluble in unfertilized eggs cannot be detected in the
CSK prior to fertilization, suggesting that there is minimal
trapping of soluble proteins in these preparations (Fig. 8, Tub).
Tubulin only becomes detectable after fertilization when it
becomes specifically associated with the CSK in the form of
short microtubules as detected by immunofluorescence (data
not shown). Levels of kinesin heavy chain, whose association
with the mitotic apparatus has previously been shown to be
detergent-sensitive (Wright et al., 1991), do not appreciably
change between interphase and dividing cells. Additionally,
while the β subunit of calcium-calmodulin dependent kinase II
can be easily detected in whole cell lysates, it cannot be
detected in any cortical preparations (data not shown). Because
nuclei are normally extruded during the detergent extraction
and Dounce homogenization of the cortices, lamin B was used
as a probe for nuclear contamination. Additionally, lamins are
hyperphosphorylated and subsequently solubilised during
Table 1. Percentage of proteins associated with the cortical
cytoskeleton
p34cdc2
MAP kinase
fyn
CAM kinase
Kinesin HC
Lamin B
Unfertilized
Interphase
Mitotic
1.48
1.01
2.81
n.d.
4.7
1.8
2.4
3.3
3.4
n.d.
3.6
1.4
1.0
1.0
1.3
n.d.
3.54
1.2
Equal loads of cortical and whole cell protein derived from unfertilized
eggs, interphase or dividing zygotes were resolved by SDS-PAGE, transferred
to nitrocellulose, and probed for the presence of several kinases, kinesin
heavy chain, and lamin B. The relative fractions of each protein were
determined by normalizing the signal detected in the cortex (which represents
1.5% of total cell protein) to the corresponding whole cell signal. n=6.
n.d., not detected.
Identification of cortical substrates for the MBP- and
p34cdc2 kinases
Western blotting as well as histone H1 and MBP kinase assays
indicate that cell cycle- and signaling-associated kinases are
active within the actin cortex of dividing cells. To identify
potential substrates of p34cdc2 and MAP kinase, an enriched
MBP/MAP kinase fraction was prepared from cleavage-stage
zygotes fractionated by phenyl-Sepharose chromatography
(Sanghera et al., 1992). This highly reproducible method has
been employed in a number of different systems to purify
MAP kinases (Ray and Sturgill, 1988; Sanghera et al., 1992;
Mamajiwalla and Burgess, 1995). Analysis of fractions eluted
from the phenyl-Sepharose column by histone H1- and MBP
kinase assays indicate that MBP kinase activity elutes as 2
peaks, a minor peak co-eluting with the major H1 kinase
activity, and a major peak eluting at the end of the gradient
(Fig. 9A). Western blotting of these fractions with MAP kinase
antibodies reveal that the 49 kDa cross reactive species (Fig.
9B, lane 1) is present in the major peak of MBP kinase activity,
whereas other cross-reactive species are not detected in this
fraction. Additionally, this peak fraction showed low H1 kinase
activity and no p34cdc2 by immunoblotting (data not shown).
This partially purified MBP kinase fraction, along with commercially available p34cdc2 was used in a substrate blot assay,
where cortical proteins were resolved by SDS-PAGE, transferred to nitrocellulose, then renatured and incubated in the
presence of [32P]ATP and purified kinases. The embryo MBP
kinase fraction consistently phosphorylates polypeptides of 43
and a 66 kDa on renatured blots (Fig. 9B, lane 2). The
denatured cortical cytoskeletal proteins on these immobilized
blots demonstrate no endogenous autophosphorylation activity
(Fig. 9B, lane 3) if blots are incubated in the presence of
[32P]ATP but without added kinase. In contrast, use of purified
p34cdc2 kinase predominantly phosphorylates a 66 kDa cortical
cytoskeletal protein in this assay (Fig. 9B, lane 4).
Microtubules of the mitotic apparatus affect the
timing of cortical kinase activity
Previous work has shown that astral microtubules of the mitotic
apparatus are crucial for the positioning and induction of the
contractile ring in cleaving echinoderm zygotes (reviewed by
Rappaport, 1986, 1996). To determine whether microtubule
integrity plays a role in regulating cortical kinase activities or
cortical CSK protein phosphorylation, nocodazole was added
to fertilized zygotes to disassemble microtubules in vivo (Fig.
10A). The isolated control cortex CSK contains detectable
amounts of tubulin, based on immunofluorescence microscopy
(Fig. 10A) and western blotting (Fig. 10B, lanes 1 and 2).
Cortical cytoskeleton-associated kinases 1381
Fig. 9. Isolation of sea urchin egg MBP kinase and the identification
of specific cortical CSK substrates for the MBP and p34cdc2 kinases.
(A) Whole-cell extracts from cleavage-stage embryos were applied to
phenyl-Sepharose, and eluted with a linear gradient of 0-60%
ethylene glycol, 250-25 mM NaCl. The eluted fractions were assayed
for MBP (solid line) and H1 (dotted line) kinase activities. MBP
kinase activity is found in both flow-through fractions and in a
retained peak fraction (fraction 77). The retained peak fraction
containing polypeptides of 49 and 52 kDa immunoreactive to antiMAP kinase (B, lane 1) but possessed minimal H1 kinase activity
(panel A) and no p34cdc2 by western blotting (data not shown).
Significant histone H1 kinase activity is only found in the flow
through. (B) Phenyl-Sepharose-enriched MBP kinase (fractions 7480), and purified p34cdc2 were used in a substrate blot assay to detect
kinase substrates within isolated cortices resolved by SDS-PAGE,
transferred to nitrocellulose. Autoradiography of nitrocellulose strips
incubated with egg MBP kinase and [γ-32P]ATP reveals that species
of 43 and 66 kDa are the major species phosphorylated (lane 2). No
endogenous autophosphorylation activity is detected when blots are
incubated with [32P]ATP in the absence of exogenous protein kinase
(lane 3). Commercially obtained p34cdc2/cyclin B complex (UBI,
Lake Placid, NY) predominantly phosphorylates a 66 kDa protein
(lane 4).
Nocodazole treatment of eggs prior to cortex isolation drastically reduces the amount of tubulin present in the cortex (Fig.
10B), disrupts all zygote and cortical microtubules (Fig. 9A),
blocking mitosis and cytokinesis.
Zygotes were treated with nocodazole, processed for wholecell H1 kinase activity (Fig. 11) and for determination of in
vitro/in vivo phosphorylation of cortical CSK proteins (Fig.
Fig. 10. Microtubule disassembly by nocodazole. (A) Detergentextracted cortices were derived from either normal cleaving
blastomeres (upper panels) or from zygotes treated with nocodazole
(lower panels). Cortices were fixed and processed for anti-tubulin
localization. Left panels, Nomarski images. Right panels, antitubulin immunofluorescence. Bar, 10 µm. (B) Microtubule disruption
confirmation by immunoblot analysis. Cortical CSK proteins, shown
by Coomassie staining (Left panel) were transferred to nitrocellulose
and probed with an antibody to α-tubulin (Right panel). Arrows
indicate the position of tubulin in either control (lanes 1-2) or
nocodazole treated-samples (lanes 3-4).
12). Histone H1 kinase activity in whole cell homogenates
(Fig. 11) made from nocodazole-treated zygotes reveals that
nocodazole significantly delays the peak of H1 kinase activity
through the time course corresponding to the first two cell
cycles (Fig. 11). Examination of in vivo and in vitro labeling
of p43 and LC20 reveals that both p43 and LC20 phosphorylation are affected by nocodazole disruption of microtubules
(Fig. 12A-D), with nocodazole delaying the peak in p43 and
LC20 phosphorylation until after the time of cleavage of
control zygotes. Quantitation of p43 phosphorylation from
cortices labeled in vivo or in vitro reveals that phosphorylation
levels in nocodazole-treated cultures is on average 25% of
8
fractions of p34cdc2 and MAP kinase associated with the
cortex does not change during the cycle, their respective
activities peak concomitant with the onset of cleavage. The
activation of cortical kinase activities may be dependent on
microtubules, since nocodazole-treatment of zygotes results
in a delay in cortical LC 20/p43 phosphorylation. These
results point toward a mechanism whereby the mitotic
apparatus directs the position of the cleavage furrow by the
local delivery or activation of cortical kinases.
+nocodazole
untreated
6
4
150
100
2
50
Relative Histone H1 Phosphorylation
1382 G. R. Walker, C. B. Shuster and D. R. Burgess
Time Post-fertilization
Fig. 11. Effect of nocodazole on whole cell histone H1 kinase
activity. Whole cell histone H1kinase activity was assayed in
untreated control cultures (h) or embryos continuously incubated
with nocodazole (d).
controls prepared at the time of first cell division (Fig. 12E).
When a single batch of zygotes was used for both whole cell
H1 kinase and in vitro phosphorylation assays, the delayed
peak of histone H1 phosphorylation activity determined in
whole cell homogenates from nocodazole-treated embryos correlated precisely with the delayed peak of p43 and LC20 phosphorylation seen in vivo and in vitro.
DISCUSSION
While genetic analyses as well as morphological approaches
continue to identify components of the contractile ring, the
molecular mechanisms that determine the spatial and
temporal regulation of cleavage furrow formation remain
elusive. In an effort to define the potential targets of cell cycle
regulation within the actin cortex, we used a preparation of
the sea urchin embryo cortex previously described as
retaining the biochemical and contractile characteristics of
the intact zygote cortex (Walker et al., 1994). Using in vivo
and in vitro labeling techniques, three polypeptides were
identified that are the major phosphoproteins associated with
the cortex at the time of cleavage. Immunoprecipitation
studies reveal that the 20 kDa phosphoprotein is the myosin
regulatory light chain (LC 20), which has been previously
shown to be a substrate of p34cdc2 and MLCK (Satterwhite et
al., 1992; Yamakita et al., 1994). The identity of the other two
polypeptides, p43 and p66, remain unknown at this time,
although the presence of phosphoserine/threonine as well as
phosphotyrosine residues suggests that these molecules are
the substrates of at least two different protein kinases. In-gel
kinase assays, western blotting, as well as biochemical fractionation of sea urchin lysates reveal that two cell cycle-associated protein kinases, p34cdc2 and a homolog of MAP kinase,
the tyrosine kinases fyn and abl (Walker et al., 1996), and
several other unidentified kinases are present and active
within the cortex. Control experiments indicate that the association of these kinases with the cortex is not due to nonspecific trapping of soluble molecules or of differential association of these kinases during the cell cycle. While the actual
Cell cycle-regulated kinase activities are associated
with the cortical cytoskeleton
Our results indicate that p34cdc2, a sea urchin homolog of the
erk family of kinases, tyrosine kinases, and possibly other
kinases are active and specifically associated with the actinrich cortical cytoskeletons of dividing blastomeres. These
results are not surprising since there are many examples of
specific cytoskeletal associated kinases, such as FAK (Dash et
al., 1995), pp60src (Weernink and Rijksen, 1995), and MLCK
(De Lanerolle et al., 1981), which retain activity as part of an
isolated CSK complex. While we have identified at least two
cell cycle regulated serine/threonine kinases (a MAP kinase
and p34cdc2) that are associated with the cortex of the dividing
cell it is likely, on the basis of in-gel kinase assays, that additional serine/threonine kinases are also associated with the
CSK. By implication, myosin light chain kinase is also active
within these cortices, since we have previously reported
cortical contractility (Walker et al., 1994) and now demonstrate
phosphorylation of myosin light chain in cortices derived from
dividing cells. We have not, however, mapped the phosphorylation sites on LC 20, and therefore cannot specifically attribute
the phosphorylation of cortical LC20 to MLCK. Additionally,
our findings that multiple kinases can be identified in an in-gel
kinase assay (including kinases of 84, 44 and 42 kDa) are consistent with the presence of several as yet unidentified kinases
in the cortical CSK. The fact that neither staurosporine nor H7, when applied to whole eggs immediately prior to cell
division or to cortices isolated from dividing cells (Fig. 7), have
any effect on the phosphorylation of the cortical phosphoproteins or cytokinesis suggests that protein kinase C is not active
or essential in the cortical CSK of dividing cells. It will be of
great interest to determine whether the kinases active within
the cortical cytoskeleton act to mediate the anaphase-telophase
transition.
How the kinases are associated with the cortex is not clear.
MAP kinases have been reported to associate with the mitotic
apparatus (Verlhac et al., 1993), and as well as with microtubules within hippocampal neurons (Morishima-kawashima
and Kosik, 1996). Cyclin B specifically associates with MAP
4, thus providing a docking site for p34cdc2 with the mitotic
apparatus (Ookata et al., 1995), and myosin light chain kinase
has been shown to redistribute from actin-containing stress
fibers in interphase cells (De Lanerolle et al., 1981) to kinetochore spindle fibers during mitosis (Guerriero et al., 1981). Our
results suggest that it is unlikely that intact microtubules are
involved in the quantitative association of kinases with the
cortical CSK since we found disruption of microtubules with
nocodazole delayed peak cortex-associated kinase activity but
did not inhibit the activity of cortical kinases. It is entirely
possible that the cortex-associated kinases are part of large
complexes that includes their specific substrates since the p66,
B
In vivo p43 phosphorylation
0
50
100
Time (min)
0
D
40
60
80
Time (min)
100
120
Post-fertilization
50
100
Time (min)
In vitro p43 phosphorylation
C
150
Post-fertilization
140
150
Post-fertilization
In vitro LC20 phosphorylation
A
In vivo LC20 phosphorylation
Cortical cytoskeleton-associated kinases 1383
40
60
80
Time (min)
100
120
140
Post-fertilization
100
E
Fig. 12. In vitro and in vivo
phosphorylation of cortical CSK proteins
in nocodazole-treated embryos. (A-D) In
vivo (A,B) and in vitro (C,D)
phosphorylation reactions were performed
on control (solid lines) or nocodazoletreated (hatched lines) embryos, and the
levels of p43 (A,C) and LC20 (B,D)
phosphorylation were assayed. (E) p43
phosphorylated in vivo or in vitro was
assayed at the time of normal cleavage (90
minutes post-fertilization) in control- or
nocodazole-treated embryos. n=3.
Peak Protein Phosphorylation
Control
80
60
Nocodazole
40
20
0
In Vivo 43 kd
Cleavage stage
p43 and LC20 phosphoproteins are phosphorylated in vitro in
these detergent models of the cell cortex.
Control experiments indicate that the presence of these
cortical kinases is not due to non-specific trapping (Fig. 8,
Table 1). We found that molecules such as tubulin and nuclear
lamin are not retained within the cortex when both molecules
are in a soluble, non-filamentous form. Additionally, the
abundant cytoplasmic protein kinase calcium-calmodulin
dependent kinase II, cannot be detected associating with the
cortex. Additional evidence against non-specific trapping
comes from the in vivo phosphorylation of a 36 kDa cortical
protein that is not phosphorylated under in vitro labeling conditions (Figs 1 and 2). These results indicate that the kinase(s)
responsible for phosphorylating p36 is soluble and not retained
during CSK preparation. Thus, those kinases identified by ingel kinase assays, western blotting and biochemical fractiona-
In Vitro 43 kd
Cleavage stage
tion are, in all likelihood, specifically associated with the
cortical cytoskeleton.
We have also identified tyrosine kinase activity associated
with the cortex, as detected by phosphoamino acid analysis of
p66 (Fig. 5). We have recently shown that an abl-related
tyrosine kinase, termed e-abl (Moore and Kinsey, 1994), is
active and associated with the cortical CSK from fertilized eggs
(Walker et al., 1996). Immunoblot analysis of the isolated
cortical cytoskeletons reveals that in addition to e-abl, the src
homolog fyn is also present. However, it remains to be seen
whether these are, in fact, the kinases that phosphorylate p66
during cytokinesis. Src is known to be associated with the
membrane cytoskeleton and is active at the G2/M transition
(Chackalaparamil and Shalloway, 1988). Additionally,
antibody injection experiments suggest that src family
members (in particular, fyn) are required for proper cell
1384 G. R. Walker, C. B. Shuster and D. R. Burgess
division (Roche et al., 1995). Indeed, cytokinesis defects have
been observed in mice deficient for the fyn gene (Yasunaga et
al., 1996). Alternatively, tyrosine phosphorylation of p66 and
p43 may be mediated by dual specificity kinases such as the
MAP kinase kinase (MEK), which play a role in regulating cell
cycle-related events (reviewed by Ahn et al., 1992). Understanding the putative relationship between tyrosine- or dual
specificity kinases and p66 may lend considerable insights into
the role of tyrosine kinases in cortical remodeling during
mitosis, as well as their role in regulating the actin cytoskeleton in general.
The identities of the other CSK phosphoproteins (p43 and
p66) are not known at this time. Recently, Bachman and
McClay (1995) have shown that the sea urchin egg possesses
a 75 kDa member of the ERM family of actin-binding and
tyrosine-phosphorylated proteins. Moesin redistributes to the
cell cortex upon fertilization, where its localization is
dependent on actin filaments (Bachman and McClay, 1995).
Preliminary immunoblot analysis indicates that moesin is
present in the isolated cortical CSK; and while immunoprecipitation analysis indicates that sea urchin moesin is phosphorylated following fertilization, we find that moesin and p66
are immunologically distinct proteins (data not shown). The
identity of the 36 and 43 kDa cortical CSK phosphoproteins
are also unknown at this time. It is unlikely that the p43 is a
MAP kinase, since phosphoamino acid analysis reveals only a
limited amount of phosphotyrosine in p43, and two-dimensional gel analysis indicates that p43 has a very basic isoelectric point.
Regulation of cortical cytoskeletal kinase activity by
the cell cycle and the mitotic apparatus
Our finding that the kinase activities associated with the
cortical cytoskeleton peak at cell division suggests that these
kinases may play a key role in the regulation of cytokinesis.
Both p34cdc2 and MAPK are well documented as regulators of
cell growth and division. There is an extensive literature concerning p34cdc2 and its role in regulating the G2/M transition
(Nurse, 1990). In terms of its role in regulating cytokinesis, it
has been proposed that p34cdc2 activity acts as a timer for
cytokinesis by negatively regulating cortical contractility until
anaphase, when pp34cdc2 activity falls off (Satterwhite and
Pollard, 1992). This regulation is thought to be accomplished
through the phosphorylation of myosin regulatory light chain
(Satterwhite et al., 1992; Yamakita et al., 1994). However,
aside from the regulatory light chain, caldesmon (Yamashiro et
al., 1991), and spectrin (Fowler and Adam, 1992), few other
substrates of p34cdc2 have been identified that associate with
the actin cytoskeleton. p66 represents a potential novel
substrate for p34cdc2, and further characterization will reveal
how phosphorylation modulates the activities of this polypeptide and its association with the cortical cytoskeleton.
Our results from in vitro labeling studies have identified a
histone H1 kinase activity that is highest in cortices isolated
from cleaving blastomeres (Fig. 6C). However, H1 kinase
activity has been widely shown to undergo an abrupt decline
during anaphase, and it has been proposed that this loss of
p34cdc2 activity signals the onset of contractile ring formation
and contraction (Satterwhite and Pollard, 1992). Our results
suggest that the H1 kinase activity associated with the actin
cytoskeleton may be either mediated by a second, possibly
novel mitotic kinase or, alternatively, the cortex-associated H1
kinase activity may represent a sequestered pool of active
p34cdc2, whose activity lingers on after the cytoplasmic activity
has declined due to cyclin degradation If this is indeed the case,
determining the spatial distribution of p34cdc2 and cyclin within
the cortex relative to the position of the furrow may yield
important insights into the mechanisms by which the position
of the furrow is determined.
MAP kinases have been shown to play roles in the spindle
assembly checkpoint pathway during mitosis (Minshull et al.,
1994), microtubule dynamics (Gotoh et al., 1991b; Karsenti et
al., 1984), and are thought to associate with microtubule organizing centers at mitosis (Verlhac et al., 1993). Xenopus MAP
kinase has been shown to be phosphorylated during meiotic
maturation and the embryonic cell cycle (Ferrell et al., 1991),
implying that its activity may be modulated directly or indirectly by p34cdc2 (Gotoh et al., 1991a). Similarly, Pelech et al.
(1988) reported that sea urchin zygote whole cell MAP kinase
activity cycles with histone H1 activity during the cell cycle.
Our findings that cortex-associated MAP kinase activity cycles
through the embryonic cell cycle, and peaks at cell division
confirm and extend the results of Pelech et al. (1988, 1990).
One of the more intriguing findings of our studies is that
nocodazole disruption of microtubules, which interferes with
the stimulation of cleavage furrow formation by disrupting the
mitotic asters, delays the peak of p34cdc2 and other cortex-associated kinase activities (Figs 11 and 12). Both in vivo phosphorylation of specific cortical phosphoproteins and in vitro
kinase activity of the cortex-associated kinases are significantly
altered by the disruption of microtubules (Fig. 12). The study
of surface contraction waves (SCW) in Xenopus eggs, which
occur in a microtubule-independent manner (Hara et al., 1980),
have called into question whether microtubules influence
cyclic MPF activity. More recent evidence suggests that the
apparent microtubule independence of MPF in early embryos
is due to a low ratio of chromatin:cytoplasm, which inhibits the
normal MAP kinase-mediated spindle checkpoint pathway
(Minshull et al., 1994). Our results resemble the delay in peak
H1 kinase activity seen in nocodazole-treated Xenopus cycling
extracts (Minshull et al., 1994), more than the observations of
SCWs that have been shown to be regulated by pathways
distinct than those mediating cleavage furrow formation
(Asada-Kubota and Kubota, 1991).
Our results suggest that p34cdc2 regulates the activity of the
cortex-associated kinases that phosphorylate LC20 and p43, or
that the regulatory pathways that control p34cdc2 activity also
modulate these cortical kinases. However, since disruption of
microtubules did not prevent but only delayed the activation of
the cortical kinases coincident with the peak in p34cdc2 activity,
it is possible that the kinases are resident in the cortical CSK
rather than being transported along microtubules to the cortex.
Rappaport and Rappaport (1993) have provided evidence that
the cortex is responsive to the signal from the mitotic apparatus
over an extended, but finite, part of the cell cycle. If p34cdc2
acts to entrain cortical kinase activities (required as a part of
the signaling cascade for cytokinesis) during this period of
responsiveness, then a delay in p34cdc2 activity beyond this
window would not signal cleavage furrow formation, in spite
of a subsequent elevation of activity. Our results are therefore
consistent with a model for cytokinesis involving the astral
microtubule-dependent activation of protein kinases, including
Cortical cytoskeleton-associated kinases 1385
some associated with the cortical actin cytoskeleton, which
play a role in the signaling cascade for cytokinesis.
The authors thank Dr David McClay (Duke University) for
providing antibodies against sea urchin moesin, Dr Charles Walsh for
anti-tubulin antibodies, Dr John Scholey (University of California at
Davis) for anti-sea urchin kinesin antibodies, and Dr Gary Wessel
(Brown University) for antibodies against sea urchin lamin B. The
authors also thank Drs Karl Fath and Susan Gilbert for critical evaluation of the manuscript. This work was supported in part by NIH
GM 40086 (awarded to D.R.B.) and a Pittsburgh Cancer Institute Fellowship (C.B.S.).
REFERENCES
Ahn, N. G., Seger, R. and Krebs, E. G. (1992). The mitogen-activated protein
kinase activator. Curr. Opin. Cell Biol. 4, 992-999.
Asada-Kubota, M. and Kubota, H. Y. (1991). Furrow-related contractions are
inhibited but furrow-unrelated contractions are not affected in af mutant eggs
of Xenopus laevis. Dev. Biol. 147, 354-362.
Bachman, E. S. and McClay, D. R. (1995). Characterization of moesin in the
sea urchin Lytechinus variegatus: redistribution to the plasma membrane
following fertilization is inhibited by cytochalasin B. J. Cell Sci. 108, 161171.
Bement, W. M. and Capco, D. G. (1991). Analysis of inducible contractile
rings suggests a role for protein kinase C in embryonic cytokinesis and
wound healing. Cell Motil. Cytoskel. 20, 145-157.
Bengur, A. R., Robinson, E. E., Appella, E. and Sellers, J. R. (1987).
Sequence of the sites phosphorylated by protein kinase C in the smooth
muscle light chain. J. Biol. Chem. 262, 7613-7617.
Boyle, W., van der Geer, J. P. and Hunter, T. (1991). Phosphopeptide
mapping and phosphoamino acid analysis by two dimensional separation on
thin-layer cellulose Plates. Meth. Enzymol. 201, 110-149.
Chackalaparamil, I. and Shalloway, D. (1988). Altered phosphorylation and
activation of pp60c-src during fibroblast mitosis. Cell 52, 801-810.
Chou, Y. H., Bischoff, J. R., Beach, D. and Goldman, R. D. (1990).
Intermediate filament reorganization during mitosis is mediated by p34cdc2
phosphorylation of vimentin. Cell 62,1063-1071.
Cao, L. and Wang, Y. L. (1996). Signals from the midzone are required for the
stimulation of cytokinesis in cultured epithelial cells. Mol. Biol. Cell 7, 225232.
Dash, D., Aepfelbacher, M. and Siess, W. (1995). The association of
pp125FAK, pp60Src, CDC42Hs and Rap1B with the cytoskeleton of
aggregated platelets is a reversible process regulated by calcium. FEBS Lett.
363, 231-234.
De Lanerolle, P., Adelstein, R. S., Feramisco, J. R. and Burridge, K. (1981).
Characterization of antibodies to smooth muscle myosin kinase and their use
in localizing myosin kinase in non-muscle cells. Proc. Nat. Acad. Sci. USA
78, 4738-4742.
DeLozanne, A. and Spudich, J. A. (1987). Disruption of the Dictyostelium
myosin heavy chain gene by homologous recombination. Science 236, 10861091.
Earnshaw, W. C. and Bernat, R. L. (1991). Chromosomal passengers: toward
an integrated view of mitosis. Chromosoma 100, 139-146.
Fenton, B. and Glover, D. M. (1993). A conserved mitotic kinase active at late
anaphase-telophase in syncytial Drosophila embryos. Nature 363, 637-640.
Ferrell, J. J., Wu, M., Gerhart, J. C. and Martin, G. S. (1991). Cell cycle
tyrosine phosphorylation of p34cdc2 and a microtubule-associated protein
kinase homolog in Xenopus oocytes and eggs. Mol. Cell. Biol. 11, 19651971.
Fishkind, D. J. and Wang, Y. I. (1995). New horizons for cytokinesis. Curr.
Opin. Cell Biol. 7, 23-31.
Fowler, V. M. and Adam, E. J. H. (1992). Spectrin redistributes to the cytosol
and is phosphorylated during mitosis in cultured cells. J. Cell Biol. 119,
1559-1572.
Golsteyn, R. M., Mundt, K. E., Fry, A. M. and Nigg, E. A. (1995). Cell cycle
regulation of the activity and subcellular localization of PLK1, a human
protein kinase implicated in mitotic spindle function. J. Cell Biol. 129, 16171628.
Gotoh, Y., Moriyama, K., Matsuda, S., Okumura, E., Kishimoto, T.,
Kawasaki, H., Suzuki, K., Yahara, I., Sakai, H. and Nishida, E. (1991a).
Xenopus M phase MAP kinase: isolation of its cDNA and activation by MPF.
EMBO J. 10, 2661-2668.
Gotoh, Y., Nishida, E., Matsuda, S., Shiina, N., Kosako, H., Shiokawa, K.,
Akiyama, T., Ohta, K. and Sakai, H. (1991b). In vitro effects on
microtubule dynamics of purified Xenopus M phase-activated MAP kinase.
Nature 349, 251-254.
Guerriero, V., Rowley, D. R. and Means, A. R. (1981). Production and
characterization of an antibody to myosin light chain kinase and intracellular
localization of enzyme. Cell 27, 449-458.
Hidaka, H., Inagaki, M., Kawamoto, S. and Sasaki, Y. (1984).
Isoquinolinesulfonamides, novel and potent inhibitors of cyclic nucleotide
dependent protein kinase and protein kinase C. Biochemistry 23, 5036-5041.
Hara, K., Tydeman, P. and Kirschner, M. (1980). A cytoplasmic clock with
the same period as the Division cycle in Xenopus eggs. Proc. Nat. Acad. Sci.
USA 77, 462-466.
Hutchcroft, J., Anostario, M., Harrison, M. L. and Geahlen, R. L. (1991).
Renaturation and assay of protein kinases after electrophoresis in sodium
dodecyl sulfate-polyacrylamide gels. Meth. Enzymol. 200, 417-423.
Jiang, W., Veno, P. A., Wood, R. W., Peaucellier, G. and Kinsey, W. H.
(1990). pH regulation of an egg cortex tyrosine kinase. Dev. Biol. 146, 81-88.
Karess, R. E., Chang, X.-J., Edward, K. A., Kulkarni, S., Aguilera, I. and
Kiehart, D. P. (1991). The regulatory light chain of nonmuscle myosin is
encoded by spaghetii-squash, a gene required for cytokinesis. Cell 65, 11771189.
Karsenti, E., Newport, J., Hubble, R. and Kirschner, M. (1984). The
Interconversion of metaphase and interphase microtubule srrays, as studied
by the injection of centrosomes and nuclei into Xenopus eggs. J. Cell Biol. 98,
1730.
Kiehart, D. P., Mabuchi, I. and Inoue, S. (1982). Evidence that myosin does
not contribute to force production in chromosome movement. J. Cell Biol. 94,
165-178.
Knecht, D. A. and Loomis, W. F. (1987). Antisense RNA inactivation of
myosin heavy-chain gene expression in Dictyostelium discoideum. Science
236, 1081-1086.
Kumagai, A. and Dunphy, W. G. (1996). Purification and molecular cloning
of Plx1, a Cdc25-regulatory kinase from Xenopus egg extracts. Science 273,
1377-1380.
Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of
the head of bacteriophage T4. Nature 227, 680-685.
Lamb, N. J. C., Fernandez, A., Watrin, A., Labbé, J.-C. and Cavadore, J.C. (1990). Microinjection of p34cdc2 kinase induces marked changes in cell
shape, cytoskeletal organization, and chromatin structure in mammalian
fibroblasts. Cell 60, 151-165.
Langan, T. A., Gautier, J., Lohka, M., Hollingsworth, R., Moreno, S.,
Nurse, P., Maller, J. and Sclafani, R. A. (1989). Mammalian growthassociated H1 histone kinase: a homolog of cdc2/CDC28 protein kinase
controlling mitotic entry in yeast and frog cells. Mol. Cell Biol. 9, 3860-3868.
Lee, K. S., Yuan, Y. L., Kuriyama, R. and Erickson, R. L. (1995). PLK is an
M-phase specific protein kinase and interacts with a kinesin-like protein
CHO1/MKLP-1. Mol. Cell Biol. 15, 7143-7151.
Llamazares, S., Moreira, A., Tavares, A., Girdham, C., Spruce, B.,
Gonzalez, C., Karess, R., Glover, D. M. and Sunkel, C. E. (1991). Polo
encodes a protein kinase homolog required for mitosis. Genes Dev. 5, 21532165.
Mabuchi, I. and Okuno, M. (1977). The effect of myosin antibody on the
division of starfish blastomeres. J. Cell Biol. 74, 251-263.
Mabuchi, I. and Takano-Ohmuro, H. (1990). Effects of inhibitors of myosin
light chain kinase and other protein kinases on the first cell division of sea
urchin eggs. Dev. Growth Differ. 32, 549-556.
Mabuchi, I. (1986). Biochemical aspects of cytokinesis. Int. Rev. Cytol. 101,
175-213.
Mamajiwalla, S. N. and Burgess, D. R. (1995). Differential regulation of the
activity of the 42 kd mitogen activate protein kinase (p42mapk) during
enterocyte differentiation in vivo. Oncogene 11, 377-386.
Margolis, R. L. and Andreassen, P. R. (1993). The telophase disc: its possible
role in mammalian cell cleavage. BioEssays 15, 201-207.
Matsudaira, P. T. and Burgess, D. R. (1978). SDS microslab linear gradient
polyacrylamide gel electrophoresis. Anal. Biochem. 87, 386-396.
Minshull, J., Sun, H., Tonks, N. K. and Murray, A. W. (1994). A MAP
kinase-dependent spindle assembly checkpoint in Xenopus egg extracts. Cell
79, 475-486.
Moore, K. L. and Kinsey, W. H. (1994). Identification of an abl-related protein
tyrosine kinase in the cortex of the sea urchin egg: Possible role at
fertilization. Dev. Biol. 164, 444-455.
1386 G. R. Walker, C. B. Shuster and D. R. Burgess
Morishima-Kawashima, M. and Kosik, K. S. (1996). The pool of MAP
kinase associated with microtubules is small but constitutively active. Mol.
Biol. Cell 7, 893-905.
Nishikawa, M., Sellers, J. R., Adelstein, R. S. and Hidaka, H. (1984). Protein
kinase C modulates in vitro phosphorylation of the smooth muscle heavy
meromyosin by myosin light chain kinase. J. Biol. Chem. 259, 8808-8814.
Nurse, P. (1990). Universal control mechanism regulating onset of M-phase.
Nature 344, 503-508.
Ohkura, H., Hagan, I. M. and Glover, D. M. (1995). The conserved
Schizosaccharomyces pombe kinase plo1, required to form a bipolar spindle,
the actin ring, and septum, can drive septum formation in G1 and G2 cells.
Genes Dev. 9, 1059-1073.
Ookata, K., Hisanaga, S., Bulinski, J. C., Murofuschi, H., Aizawa, H., Itoh,
T., Hotani, H., Okumura, E., Tachibana, K. and T. Kishimoto. (1995).
Cyclin B interaction with microtubule-associated protein 4 (MAP4) targets
p34cdc2 kinase to microtubules and is a potential regulator of M-phase
microtubule dynamics. J. Cell Biol. 128, 849-862.
Ostrow, B. D., Chen, P. and Chisholm, R. L. (1994). Expression of a myosin
regulatory light chain phosphorylation site mutant complements the
cytokinesis and developmental defects of Dictyostelium RMLC null cells. J.
Cell Biol. 127, 1945-1955.
Pelech, S. L., Tombes, R. M., Miejer, L. and Krebs, E. G. (1988). Activation
of myelin basic protein kinases during echinoderm oocyte maturation and
egg fertilization. Dev. Biol. 130, 28-36.
Pelech, S. L., Sangera, J. S. and Daya-Makin, M. (1990). Protein kinase
cascades in meiotic and mitotic cell cycle control. Biochem. Cell Biol. 68,
1297-1330.
Rappaport, R. (1973). On the rate of movement of the cleavage stimulus in
sand dollar eggs. J. Exp. Zool. 183, 115-119.
Rappaport, R. (1986). Establishment of the mechanism of cytokinesis. Int.
Rev. Cytol. 101, 245-281.
Rappaport, R. and Rappaport, B. N. (1993). Duration of division-related
events in cleaving sand dollar eggs. Dev. Biol. 158, 265-273.
Rappaport, R. (1996). Cytokinesis in Animal Cells. Cambridge University
Press, Cambridge.
Ray, L. B. and Sturgill, T. W. (1988). Insulin-stimulated microtubuleassociated protein kinase is phosphorylated on tyrosine and threonine in vivo.
Pro. Nat. Acad. Sci. USA 85, 3753-3757.
Roche, S., Fumagalli, S. and Courtneidge, S. A. (1995). Requirement for Src
family protein tyrosine kinases in G2 for fibroblast cell division. Science 15,
1567-1569.
Sanghera, J. S., Peter, M., Nigg, E. A. and Pelech, S. L. (1992).
Immunological characterization of avian MAP kinases: Evidence for nuclear
localization. Mol. Biol. Cell 3, 775-787.
Satterwhite, L. L., Lohka, M. H., Wilson, K. L., Scherson, T. Y., Cisek, L. J.,
Cordon, J. L. and Pollard, T. D. (1992). Phosphorylation of myosin-II
regulatory light chain by cyclin-p34cdc2: a mechanism for the timing of
cytokinesis. J. Cell Biol. 118, 595-605.
Satterwhite, L. L. and Pollard, T. D. (1992). Cytokinesis. Curr. Opin. Cell
Biol. 4, 43-52.
Sekimata, M., Tsujimura, K., Tanaka, J., Takeuchi, Y., Inagaki, N. and
Inagaki, M. (1996). Detection of protein kinase activity specifically
activated at metaphase-anaphase transition. J. Cell Biol. 132, 635-641.
Towbin, H., Staehelin, T. and Gordon, G. (1979). Electrophoretic transfer of
proteins from polyacrylamide gels to nitrocellulose sheets: procedure and
some applications. Proc. Nat. Acad. Sci. USA 76, 4350-4354.
Toyn, J. H. and Johnston, L. H. (1994). Dbf2 and Dbf20 protein kinases of
budding yeast activated after the metaphase to anaphase cell cycle transition.
EMBO J. 13, 1103-1113.
Vernos, I. and Karsenti, E. (1996). Motor involved in spindle assembly and
chromosome segregation. Curr. Opin. Cell Biol. 8, 4-9.
Verlhac, M. H., De, P. H., Maro, B., Cobb, M. H. and Clarke, H. J. (1993).
MAP kinase becomes stably activated at metaphase and is associated with
microtubule-organizing centers during meiotic maturation of mouse oocytes.
Dev. Biol. 158, 330-340.
Walker, G., Kane, R. E. and Burgess, D. R. (1994). Isolation and
characterization of a sea urchin zygote cortex that supports in vitro
contraction and reactivation of furrowing. J. Cell Sci. 107, 2239-2248.
Walker, G. R., Burgess, D. R. and Kinsey, W. H. (1996). Fertilization
promotes selective association of the Abl-kinase with the egg cytoskeleton.
Eur. J. Cell Biol. 70, 165-171.
Ward, G. E. and Kirschner, M. W. (1990). Identification of cell cycleregulated phosphorylation sites on nuclear lamin C. Cell 61, 561-577.
Weernink, P. and Rijksen, G. (1995). Activation and translocation of c-Src to
the cytoskeleton by both platelet-derived growth factor and epidermal growth
factor. J. Biol. Chem. 270, 2264-2267.
Wheatley, S. P. and Wang, Y. L. (1996). Midzone microtubule bundles are
continuously required for cytokinesis in cultured epithelial cells. J. Cell Biol.
135, 981-989.
Wright, B. D., Henson, J. H., Wedamean, K. P., Willy, P. J., Morand, J. N.
and Scholey, J. M. (1991). Subcellular localization and sequence of sea
urchin kinesin heavy chain: evidence for its association with membranes in
the mitotic apparatus and interphase cytoplasm. J. Cell Biol. 113, 817-833.
Wright, B. D., Terasaki, M. and Scholey, J. M. (1993). Roles of kinesin and
kinesin-like proteins in sea urchin embryonic cell division: evaluation using
microinjection. J. Cell Biol. 123, 681-689.
Yamakita, Y., Yamashiro, S. and Matsumura, F. (1994). In vivo
phosphorylation of regulatory light chain of myosin II during mitosis of
cultured cells. J. Cell Biol. 124, 129-137.
Yamashiro, S., Yamakita, Y., Hosoya, H. and Matsumura, F. (1991).
Phosphorylation of non-muscle caldesmon by p34cdc2 kinase during mitosis.
Nature 349, 169-172.
Yasunaga, M., Yagi, T., Hanzawa, N., Yasuda, M., Yamanashi, Y.,
Yamamoto, T., Aizawa, S., Miyauchi, Y. and Nishikawa, S. I. (1996).
Involvement of Fyn tyrosine kinase in progression of cytokinesis of B
lymphocyte progenitor. J. Cell Biol. 132, 91-99.
Zhang, D. and Nicklas, B. R. (1996). ‘Anaphase’ and cytokinesis in the
absence of chromosomes. Nature 382, 446-468.
(Received 20 January 1997 - Accepted 4 April 1997)