Oncogene (1999) 18, 2783 ± 2788
ã 1999 Stockton Press All rights reserved 0950 ± 9232/99 $12.00
http://www.stockton-press.co.uk/onc
SHORT REPORT
Speci®c accumulation of Rho-associated kinase at the cleavage furrow
during cytokinesis: cleavage furrow-speci®c phosphorylation of
intermediate ®laments
Hidetaka Kosako1, Hidemasa Goto1,2, Maki Yanagida1, Kaori Matsuzawa1,3, Masatoshi Fujita4,
Yasuko Tomono5, Tohru Okigaki5, Hideharu Odai6, Kozo Kaibuchi7 and Masaki Inagaki*,1
1
Laboratory of Biochemistry, Aichi Cancer Center Research Institute, Nagoya, Aichi 464-8681, Japan; 2Department of Pediatrics,
Mie University School of Medicine, Tsu 514-8507, Japan; 3Institute of Immunological Science, Hokkaido University, Sapporo
060-0815, Japan; 4Laboratory of Viral Oncology, Aichi Cancer Center Research Institute, Nagoya, Aichi 464-8681, Japan;
5
Division of Molecular and Cell Biology, Shigei Medical Research Institute, Okayama 701-0202, Japan; 6Central Laboratories
for Key Technology, Kirin Brewery Company Limited, Yokohama 236-0004, Japan; 7Division of Signal Transduction,
Nara Institute of Science and Technology, Ikoma 630-0101, Japan
The small GTPase Rho and one of its targets, Rhoassociated kinase (Rho-kinase), are implicated in a wide
spectrum of cellular functions, including cytoskeletal
rearrangements, transcriptional activation and smooth
muscle contraction. Since Rho also plays an essential
role in cytokinesis, Rho-kinase may possibly mediate
some biological aspects of cytokinesis. Here, using a
series of monoclonal antibodies that can speci®cally
recognize distinct phosphorylated sites on glial ®brillary
acidic protein (GFAP) and vimentin, phosphorylation
sites by Rho-kinase in vitro were revealed to be identical
to in vivo phosphorylation sites on these intermediate
®lament (IF) proteins at the cleavage furrow in dividing
cells. We then found, by preparing two types of antiRho-kinase antibodies, that Rho-kinase accumulated
highly and circumferentially at the cleavage furrow in
various cell lines. This subcellular distribution during
cytokinesis was very similar to that of ezrin/radixin/
moesin (ERM) proteins and Ser19-phosphorylated myosin
light chain. These results raise the possibility that Rhokinase might be involved in the formation of the
contractile ring by modulating these F-actin-binding
proteins during cytokinesis and in the phosphorylation
and regulation of IF proteins at the cleavage furrow.
Keywords: cell cycle; cytokinesis; cleavage furrow; Rhoassociated kinase (Rho-kinase); intermediate ®lament
Cell-division cycle is the fundamental means by which
all living cells are propagated, and comprises a
complicated series of cytoplasmic and nuclear events
which are elaborately coordinated under the control.
The cell cycle control system is thought to be based on
a series of protein phosphorylation/dephosphorylation
(Hunt, 1991; Norbury and Nurse, 1992; Nigg, 1993).
Recent studies demonstrated that the complexes of
cyclins and cyclin-dependent protein kinases (cdks)
play pivotal roles in controlling the cell cycle at G1-S
and G2-M transitions (Reed, 1992; Pines, 1993; Sherr,
1993). On the other hand, much less is known about
*Correspondence: M Inagaki
Received 8 June 1998; revised 17 November 1998; accepted 15
December 1998
mechanisms at metaphase-anaphase transition and
during the cytokinetic phase of eukaryotic cells.
Using site- and phosphorylation state-speci®c antibodies (Inagaki et al., 1997a), we previously detected
cleavage furrow kinase (CF kinase) activity that
phosphorylates Thr7, Ser13 and Ser38 on human glial
®brillary acidic protein (GFAP) at the cleavage furrow
during cytokinesis (Matsuoka et al., 1992). This protein
kinase activity was observed not only in astroglial cells
but also in other cultured cells in which GFAP was
ectopically expressed (Sekimata et al., 1996). These
®ndings imply a possible existence of a protein kinase
which is speci®cally activated at the cleavage furrow
and may play some important roles in cytokinesis.
The in vivo phosphorylation sites on GFAP by CF
kinase were recently shown to be phosphorylated by
Rho-kinase (Matsui et al., 1996) (also called ROK
(Leung et al., 1995) or ROCK (Ishizaki et al., 1996)) in
a GTP×Rho-dependent manner in vitro (Kosako et al.,
1997). More recently, vimentin was shown to be
phosphorylated at Ser38 and Ser71 by Rho-kinase in
vitro (Goto et al., 1998). A rabbit polyclonal antibody
against phosphorylated vimentin-Ser71 revealed that
this site was also phosphorylated at the cleavage
furrow during cytokinesis (Goto et al., 1998). In the
present work, we prepared rat monoclonal antibodies
TM38 and TM71, which can speci®cally recognize the
phosphorylation of vimentin at Ser38 and Ser71,
respectively (Figure 1a and b). Immunocytochemical
studies using TM38 and TM71 revealed that these two
sites were phosphorylated at the cleavage furrow
during cytokinesis in U251 human glioma cells
(Figure 1c and d). Together with previous immunocytochemical studies (Matsuoka et al., 1992; Tsujimura et
al., 1994; Ogawara et al., 1995; Takai et al., 1996;
Sekimata et al., 1996; Inagaki et al., 1997b), these
results indicate that both CF kinase and Rho-kinase
phosphorylate Thr7, Ser13 and Ser38 on GFAP and Ser38
and Ser71 on vimentin but do not phosphorylate Ser8 on
GFAP and Ser6, Ser33, Ser50, Ser55 and Ser82 on vimentin
(Table 1).
In contrast, in vitro sites phosphorylated by Cdc2
kinase, cAMP-dependent protein kinase (A kinase),
protein kinase C (C kinase) and Ca2+/calmodulindependent protein kinase II (CaM kinase II) were
clearly di€erent from in vivo sites phosphorylated by
Accumulation of Rho-kinase at the cleavage furrow
H Kosako et al
2784
CF kinase (Table 1). All these results are consistent
with the idea that Rho-kinase is responsible for CF
kinase activity during cytokinesis. However, there
remains the possibility that other protein kinase(s)
with similar substrate speci®city to Rho-kinase may be
CF kinase.
Figure 1 Reactivity of rat monoclonal antibodies TM38 and TM71. Vimentin peptides PV38 (CSTRTYphosphoS38LGSAL), V38
(CSTRTYS38LGSAL), PV71 (CAVRLRphosphoS71SVPGV) and V71 (CAVRLRS71SVPGV) were synthesized as described
(Tsujimura et al., 1994). Rat monoclonal antibodies against PV38 (termed TM38) and against PV71 (termed TM71) were
produced as described by Takai et al. (1996). (a) Vimentin phosphorylated by Rho-kinase, the catalytic subunit of cAMP-dependent
protein kinase (A kinase), Ca2+/calmodulin-dependent protein kinase II (CaM kinase II), Cdc2 kinase or protein kinase C (C
kinase) was prepared as described previously (Goto et al., 1998). After SDS ± PAGE, samples were transferred onto a polyvinylidene
di¯uoride (PVDF) membrane. The membrane was immunoblotted with TM38 or TM71, and then stained with Coomassie Brilliant
Blue (CBB). Immunoreactive bands were detected by horseradish peroxidase-conjugated secondary antibodies and the ECL Western
blotting detection system (Amersham Pharmacia Biotech). (b) Speci®city of TM38 and TM71 determined by an inhibition assay.
Vimentin phosphorylated by Rho-kinase was immunoblotted with TM38 or TM71 preincubated with bu€er alone or with 50 mg/ml
V38, PV38, V71 or PV71. (c and d) Late mitotic U251 cells were doubly stained with 1B8 (mouse monoclonal anti-vimentin;
Tsujimura et al., 1994) and TM38 or TM71. Indirect immuno¯uorescence microscopy was performed as described previously (Goto
et al., 1998). FITC-conjugated goat anti-rat immunoglobulins (BioSource International) and Texas Red-conjugated sheep antimouse immunoglobulins (Amersham Pharmacia Biotech) were used for secondary antibodies. DNAs were stained with 0.5 mg/ml
DAPI (4',6-diamidine-2-phenylindole-dihydrochloride; Boehringer Mannheim). Scale bars, 10 mm
Accumulation of Rho-kinase at the cleavage furrow
H Kosako et al
To examine the subcellular distribution of Rhokinase during cytokinesis, we prepared two kinds of
anti-Rho-kinase antibodies (anti-CAT and anti-COIL)
by immunizing two rabbits with two recombinant
Table 1
fragments of bovine Rho-kinase. Immunoblot analysis
revealed that both of these antibodies strongly and
speci®cally reacted with 160 kDa Rho-kinase in total
extracts from MDBK bovine epithelial cells (Figure 2a).
In vivo (CF kinase) and in vitro (Cdc2 kinase, A kinase, C kinase, CaM kinase II and Rho-kinase) phosphorylation sites on GFAP and
vimentin, and monoclonal antibodies that recognize site-speci®c phosphorylation
Antibody
site
TMG7
Thr7
GFAP
YC10
KT13
8
Ser
Ser13
Cdc2 kinase
A kinase
C kinase
CaM kinase II
Rho-kinase
CF kinase
7
++
7
7
+++
+++
++
+++
+++
+
7
7
7
+++
+++
+++
+++
+++
KT34
Ser38
MO6
Ser6
YT33
Ser33
TM38
Ser38
Vimentin
TM50
Ser50
4A4
Ser55
TM71
Ser71
MO82
Ser82
7
+++
+++
+++
+++
+++
7
+++
++
7
7
7
7
7
+++
7
7
7
7
+++
+++
+++
+++
+++
7
+++
+++
7
7
7
+++
7
7
7
7
7
7
7
7
7
+++
+++
7
+
7
+++
7
7
Phosphorylation of individual sites by Cdc2 kinase, A kinase, C kinase, CaM kinase II and Rho-kinase was examined by immunoblotting with
the indicated site- and phosphorylation-speci®c monoclonal antibodies as shown in Figure 1a. Phosphorylation of these sites by CF kinase was
examined by immunostaining U251 cells with the indicated antibodies. Rat monoclonal antibody TMG7 against GFAP peptide PG7
(CERRRVphosphoT7SAARR) was produced as described previously (Takai et al., 1996)
Figure 2 Reactivity of rabbit polyclonal antibodies anti-CAT and anti-COIL. GST-CAT (amino acids 6 ± 553; catalytic domain of
Rho-kinase) and GST-COIL (amino acids 421 ± 701; coiled-coil domain of Rho-kinase) were prepared as described (Amano et al.,
1997). Rabbit polyclonal antibodies against CAT and COIL were prepared as follows: after removing the GST portion from GST ±
CAT or GST ± COIL by thrombin, CAT or COIL protein was injected into a rabbit at 4 week intervals. The antiserum was
obtained 2 weeks after the third injection and anity-puri®ed by A-Gel 10 (Bio-Rad) coupled with CAT or COIL protein. (a)
Extracts from MDBK cells (10 mg of protein; lanes 2, 4 and 6) and puri®ed bovine Rho-kinase (5 ng of protein; lanes 3, 5 and 7)
were stained with Coomassie Brilliant Blue (CBB) or immunoblotted with anti-CAT (aCAT) or anti-COIL (aCOIL) antibody (each
0.4 mg/ml). Molecular weight standards were electrophoresed on lane 1. (b) U251 cells arrested in early mitosis were obtained by
using TN-16 (3-(1-anillinoethylidene)-5-benzylpyrrolidine-2,4-dione; Wako) as described previously (Goto et al., 1998) with slight
modi®cations. Cell lysates prepared at the indicated times after the removal of TN-16 were immunoblotted with anti-COIL (upper)
or 1B8 (lower). Microscopic observation and FACS analysis revealed the synchronized progression of cell cycle: 0 h, early mitosisrich; 0.5 h, late mitosis-rich; 6 h, G1 phase-rich; 14 h, S phase-rich (data not shown). (c and d) Confocal microscopic images of
interphase, metaphase and telophase MDBK cells stained with anti-CAT or anti-COIL antibody (each 4 mg/ml). FITC-conjugated
goat anti-rabbit immunoglobulins (BioSource International) were used for secondary antibodies. DNAs were stained with 0.5 mg/ml
propidium iodide (Sigma). Images represent horizontal optical sections. Scale bars, 10 mm
2785
Accumulation of Rho-kinase at the cleavage furrow
H Kosako et al
2786
Rho-kinase protein levels did not change during cell
cycle progression (Figure 2b). Immunocytochemical
studies with anti-CAT and anti-COIL using confocal
laser scanning microscope revealed that Rho-kinase
speci®cally accumulated at the cleavage furrow of
telophase MDBK cells (Figure 2c and d). Confocal
Figure 3 Distribution of Rho-kinase in late mitotic MDBK cells.
(a) Confocal images on ®ve serial focal planes of a single
telophase cell stained with anti-CAT and propidium iodide. These
serial optical sections were obtained at 0.8 mm intervals. Similar
results were obtained with anti-COIL antibody. (b) Confocal
images of early anaphase, late anaphase, early telophase, mid
telophase and late telophase cells stained with anti-COIL and
propidium iodide. Images represent individual optical sections.
Similar results were obtained with anti-CAT antibody. Scale bars,
10 mm
images on ®ve serial focal planes of a single
telophase cell stained with anti-CAT showed that
Rho-kinase accumulated circumferentially in the
equatorial region (Figure 3a). This cleavage furrowspeci®c accumulation of Rho-kinase appeared at late
anaphase and was maintained until late telophase,
then gradually decreased at the exit of mitosis
(Figure 3b).
To investigate whether the accumulation of Rhokinase at the cleavage furrow is a general occurrence
in various types of cells during cytokinesis, Swiss 3T3
cells and HeLa cells were stained with the anti-CAT
antibody. As shown in Figure 4a and b, Rho-kinase
accumulated at the cleavage furrow in both cell lines.
Furthermore, U251 cells were doubly labeled with
anti-CAT and KT13 (anti-phosphorylated GFAPSer13) or TM71. As shown in Figure 4c and d,
GFAP-Ser13 and vimentin-Ser71 were phosphorylated
near the area at which Rho-kinase accumulated.
However, Rho-kinase does not colocalize with these
phosphorylated intermediate ®lament proteins (Figure
4c and d), and this needs a possible explanation.
Vimentin has been shown to be attached to the
plasma membrane (Georgatos and Blobel, 1987). In
addition, the recent study using GFP/vimentin
chimeras clearly demonstrated that the vimentin
network constantly moves in a wavy manner in
living cells (Ho et al., 1998). This dynamic behavior
of intermediate ®laments and continued ingression of
the cleavage furrow may cause their transient
attachment to the plasma membrane, where the
phosphorylation by Rho-kinase might occur.
The phosphorylation of GFAP and vimentin by
Rho-kinase has been shown to lead to disassembly
of their ®lament structures in vitro (Kosako et al.,
1997; Goto et al., 1998). Most recently, we found
that the expression of mutant GFAP, where the
Rho-kinase phosphorylation sites were substituted to
alanine residues, impaired cytokinetic segregation of
GFAP ®laments, resulting in the formation of an
unusually long bridge-like structure between the
unseparated daughter cells (Yasui et al., 1998).
These observations suggest that the phosphorylation-dependent disassembly of intermediate ®laments
is required for proper execution and completion of
cytokinesis.
Rho-kinase is activated by forming a complex with
the GTP-bound active form of Rho in vitro (Matsui et
al., 1996; Ishizaki et al., 1996). The small GTPase Rho
(Van Aelst and D'Souza-Schorey, 1997; Hall, 1998) is
known to play a critical role in cytokinesis (Kishi et al.,
1993; Mabuchi et al., 1993; Drechsel et al., 1997) and
localize to the cleavage furrow (Takaishi et al., 1995).
In the present study, we found that Rho-kinase also
localized to the cleavage furrow, suggesting that Rhokinase binds to Rho in vivo and phosphorylates several
proteins during cytokinesis.
So far, myosin binding subunit of myosin phosphatase (Kimura et al., 1996), myosin light chain (MLC;
Amano et al., 1996), ezrin/radixin/moesin (ERM)
proteins (Matsui et al., 1998; Fukata et al., 1998),
adducin (Kimura et al., 1998), GFAP (Kosako et al.,
1997) and vimentin (Goto et al., 1998) have been
identi®ed as the putative physiological substrates for
Rho-kinase. Especially, ERM proteins (Sato et al.,
1991) and Ser19-phosphorylated MLC (Matsumura et
Accumulation of Rho-kinase at the cleavage furrow
H Kosako et al
al., 1998; Rho-kinase phosphorylates MLC at Ser19 in
vitro; Amano et al., 1996) have been shown to be
highly concentrated at the cleavage furrow. So we
compared the subcellular distribution of ERM proteins
and Ser19-phosphorylated MLC with that of Rhokinase. As shown in Figure 4e, we observed almost
colocalization of Rho-kinase and ERM proteins at the
cleavage furrow in rat 3Y1 cells. In addition, the
staining pattern of anti-Ser19-phosphorylated MLC
antibody at the cleavage furrow (Figure 4f) was very
similar to that of anti-CAT (Figure 2c) and anti-COIL
(Figure 2d) in telophase MDBK cells. These results
suggest that Rho-kinase phosphorylates and regulates
these F-actin-binding proteins during cytokinesis.
Phosphorylation of ERM proteins by Rho-kinase
interferes with their head-to-tail association in vitro
(Matsui et al., 1998), leading to the activation as Factin/plasma membrane cross-linkers (Tsukita et al.,
1997). MLC phosphorylation at Ser19 is believed to
promote the contractility of actomyosin in cells
Figure 4 (a and b) Distribution of Rho-kinase in various cell lines. HeLa and Swiss 3T3 cells were stained with anti-CAT antibody.
(c and d) A telophase U251 cell was doubly stained with anti-CAT (green) and KT13 (mouse monoclonal anti-phosphoSer13 of
GFAP; Sekimata et al., 1996) or TM71 (red). (e) A telophase 3Y1 cell was doubly stained with anti-CAT (green) and CR22 (red;
mouse monoclonal anti-ERM; Sato et al., 1991). (f) A telophase MDBK cell was stained with pp2b (rabbit polyclonal antiphosphoSer19 of MLC; Matsumura et al., 1998) and propidium iodide. Scale bars, 10 mm
2787
Accumulation of Rho-kinase at the cleavage furrow
H Kosako et al
2788
(Huttenlocher et al., 1995). Therefore, Rho-kinase may
be implicated in the formation of the contractile ring,
an actomyosin-based cytoskeletal structure just beneath
the plasma membrane.
While this paper was under the review, Narumiya
and co-workers reported that citron kinase, another
Rho target with the structural similarity to Rho-kinase,
localizes to the cleavage furrow and may play an
important role in the contractile process of cytokinesis
(Madaule et al., 1998). Further analyses of di€erent
and redundant functions of both Rho-binding kinases
will help to elucidate the molecular mechanism of
cytokinesis downstream of Rho.
Acknowledgements
We thank Drs S Yonemura and S Tsukita for providing
anti-ERM mAb (CR22); Dr F Matsumura for providing
anti-phosphorylated MLC-Ser 19 (pp2b); N Takahashi for
preparing GST-CAT; F Shigei for ®nancial support to YT
and TO; K Ando for technical assistance; K Kuromiya for
the secretarial services and M Ohara for critique of the
manuscript. This work was supported in part by Grants-inAid for Scienti®c Research and Cancer Research from the
Ministry of Education, Science, Sports and Culture of
Japan; Japan Society of the Promotion of Science Research
for the Future; special coordination funds from the Science
and Technology Agency of the Government of Japan; and
a grant from Bristol-Myers-Squibb.
References
Amano M, Ito M, Kimura K, Fukata Y, Chihara K, Nakano
T, Matsuura Y and Kaibuchi K. (1996). J. Biol. Chem.,
271, 20246 ± 20249.
Amano M, Chihara K, Kimura K, Fukata Y, Nakamura N,
Matsuura Y and Kaibuchi K. (1997). Science, 275, 1308 ±
1311.
Drechsel DN, Hyman AA, Hall A and Glotzer M. (1997).
Curr. Biol., 7, 12 ± 23.
Fukata Y, Kimura K, Oshiro N, Saya H, Matsuura Y and
Kaibuchi K. (1998). J. Cell Biol., 141, 409 ± 418.
Georgatos SD and Blobel G. (1987). J. Cell Biol., 105, 105115.
Goto H, Kosako H, Tanabe K, Yanagida M, Sakurai M,
Amano M, Kaibuchi K and Inagaki M. (1998). J. Biol.
Chem., 273, 11728 ± 11736.
Hall A. (1998). Science, 279, 509 ± 514.
Ho CL, Martys JL, Mikhailov A, Gundersen GG and Liem
RKH. (1998). J. Cell Sci., 111, 1767 ± 1778.
Hunt T. (1991). Semin. Cell Biol., 2, 213 ± 222.
Huttenlocher A, Sandborg RR and Horwitz A. (1995). Curr.
Opin. Cell Biol., 7, 697 ± 706.
Inagaki M, Inagaki N, Takahashi T and Takai Y. (1997a). J.
Biochem., 121, 407 ± 414.
Inagaki N, Goto H, Ogawara M, Nishi Y, Ando S and
Inagaki M. (1997b). J. Biol. Chem., 272, 25195 ± 25199.
Ishizaki T, Maekawa M, Fujisawa K, Okawa K, Iwamatsu
A, Fujita A, Watanabe N, Saito Y, Kakizuka A, Morii N
and Narumiya S. (1996). EMBO J., 15, 1885 ± 1893.
Kimura K, Ito M, Amano M, Chihara K, Fukata Y,
Nakafuku M, Yamamori B, Feng J, Nakano T, Okawa
K, Iwamatsu A and Kaibuchi K. (1996). Science, 273,
245 ± 248.
Kimura K, Fukata Y, Matsuoka Y, Bennett V, Matsuura Y,
Okawa K, Iwamatsu A and Kaibuchi K. (1998). J. Biol.
Chem., 273, 5542 ± 5548.
Kishi K, Sasaki T, Kuroda S, Itoh T and Takai Y. (1993). J.
Cell Biol., 120, 1187 ± 1195.
Kosako H, Amano M, Yanagida M, Tanabe K, Nishi Y,
Kaibuchi K and Inagaki M. (1997). J. Biol. Chem., 272,
10333 ± 10336.
Leung T, Manser E, Tan L and Lim L. (1995). J. Biol. Chem.,
270, 29051 ± 29054.
Mabuchi I, Hamaguchi Y, Fujimoto H, Morii N, Mishima M
and Narumiya S. (1993). Zygote, 1, 325 ± 331.
Madaule P, Eda M, Watanabe N, Fujisawa K, Matsuoka T,
Bito H, Ishizaki T and Narumiya S. (1998). Nature, 394,
491 ± 494.
Matsui T, Amano M, Yamamoto T, Chihara K, Nakafuku
M, Ito M, Nakano T, Okawa K, Iwamatsu A and
Kaibuchi K. (1996). EMBO J., 15, 2208 ± 2216.
Matsui T, Maeda M, Doi Y, Yonemura S, Amano M,
Kaibuchi K, Tsukita Sa and Tsukita Sh. (1998). J. Cell
Biol., 140, 647 ± 657.
Matsumura F, Ono S, Yamakita Y, Totsukawa G and
Yamashiro S. (1998). J. Cell Biol., 140, 119 ± 129.
Matsuoka Y, Nishizawa K, Yano T, Shibata M, Ando S,
Takahashi T and Inagaki M. (1992). EMBO J., 11, 2895 ±
2902.
Nigg EA. (1993). Curr. Opin. Cell Biol., 5, 187 ± 193.
Norbury C and Nurse P. (1992). Annu. Rev. Biochem., 61,
441 ± 470.
Ogawara M, Inagaki N, Tsujimura K, Takai Y, Sekimata M,
Ha MH, Imajoh-Ohmi S, Hirai S, Ohno S, Sugiura H,
Yamauchi T and Inagaki M. (1995). J. Cell Biol., 131,
1055 ± 1066.
Pines J. (1993). Trends Biochem. Sci., 18, 195 ± 197.
Reed SI. (1992). Annu. Rev. Cell Biol., 8, 529 ± 561.
Sato N, Yonemura S, Obinata T, Tsukita Sa and Tsukita Sh.
(1991). J. Cell Biol., 113, 321 ± 330.
Sekimata M, Tsujimura K, Tanaka J, Takeuchi Y, Inagaki N
and Inagaki M. (1996). J. Cell Biol., 132, 635 ± 641.
Sherr CJ. (1993). Cell, 73, 1059 ± 1065.
Takai Y, Ogawara M, Tomono Y, Moritoh C, Imajoh-Ohmi
S, Tsutsumi O, Taketani Y and Inagaki M. (1996). J. Cell
Biol., 133, 141 ± 149.
Takaishi K, Sasaki T, Kameyama T, Tsukita Sa, Tsukita Sh
and Takai Y. (1995). Oncogene, 11, 39 ± 48.
Tsujimura K, Ogawara M, Takeuchi Y, Imajoh-Ohmi S, Ha
MH and Inagaki M. (1994). J. Biol. Chem., 269, 31097 ±
31106.
Tsukita Sa, Yonemura S and Tsukita Sh. (1997). Trends
Biochem. Sci., 22, 53 ± 58.
Van Aelst L and D'Souza-Schorey C. (1997). Genes Dev., 11,
2295 ± 2322.
Yasui Y, Amano M, Nagata K, Inagaki N, Nakamura H,
Saya H, Kaibuchi K and Inagaki M. (1998). J. Cell Biol.,
143, 1249 ± 1258.