ã
Oncogene (2001) 20, 6418 ± 6434
2001 Nature Publishing Group All rights reserved 0950 ± 9232/01 $15.00
www.nature.com/onc
Cortactin: coupling membrane dynamics to cortical actin assembly
Scott A Weed*,1 and J Thomas Parsons2
1
2
Department of Craniofacial Biology, University of Colorado Health Sciences Center, Denver, Colorado, CO 80262, USA;
Department of Microbiology, Health Sciences Center, University of Virginia, Charlottesville, Virginia, VA 22903, USA
Exposure of cells to a variety of external signals causes
rapid changes in plasma membrane morphology. Plasma
membrane dynamics, including membrane ru‚e and
microspike formation, fusion or ®ssion of intracellular
vesicles, and the spatial organization of transmembrane
proteins, is directly controlled by the dynamic reorganization of the underlying actin cytoskeleton. Two
members of the Rho family of small GTPases, Cdc42
and Rac, have been well established as mediators of
extracellular signaling events that impact cortical actin
organization. Actin-based signaling through Cdc42 and
Rac ultimately results in activation of the actin-related
protein (Arp) 2/3 complex, which promotes the formation of branched actin networks. In addition, the activity
of both receptor and non-receptor protein tyrosine
kinases along with numerous actin binding proteins
works in concert with Arp2/3-mediated actin polymerization in regulating the formation of dynamic cortical
actin-associated structures. In this review we discuss the
structure and role of the cortical actin binding protein
cortactin in Rho GTPase and tyrosine kinase signaling
events, with the emphasis on the roles cortactin plays in
tyrosine phosphorylation-based signal transduction, regulating cortical actin assembly, transmembrane receptor
organization and membrane dynamics. We also consider
how aberrant regulation of cortactin levels contributes to
tumor cell invasion and metastasis. Oncogene (2001) 20,
6418 ± 6434.
Keywords: Rho GTPase; actin cytoskeleton; tyrosine
kinase substrate; Arp2/3 complex; vesicle tracking;
invasion
Introduction
Dynamic changes in the cortical actin cytoskeleton are
central to a wide variety of cellular events, including
cell motility, adhesion, endo- and phagocytosis,
cytokinesis, movement of intracellular particles
through the cytoplasm and organization of transmembrane proteins (Mitchison and Cramer, 1996; Ridley,
2000; Qualmann et al., 2000; Chimini and Chavrier,
2000; Mandato et al., 2000; Beckerle, 1998; Wheal et
al., 1998). The actin cytoskeleton at the cell periphery
*Correspondence: SA Weed; E-mail: [email protected]
consists of a highly organized meshwork of ®lamentous
(F-) actin closely associated with the overlying plasma
membrane (Small et al., 1995). In motile cells, F-actin
underlies two main types of protrusive membranes;
lamellipodia, which contain short ®laments of F-actin
linked into orthogonal arrays and ®lopodia, comprised
of bundled, crosslinked actin ®laments (Rinnerthaler et
al., 1988). Work within the last decade has demonstrated that Rac and Cdc42, two members of the Rho
family of small GTPases, govern lamellipodia and
®lopodia formation (Hall, 1998). Cdc42 and Rac play a
pivotal role in the transmission of extracellular signals
that induce cortical cytoskeletal rearrangement (Zigmond, 1996), eliciting their e€ects through a large
number of binding (or `e€ector') proteins (Aspenstrom,
1999; Bishop and Hall, 2000). It is now well established
that members of the Wiskott ± Aldrich syndrome
protein (WASp) family serve as the e€ector proteins
for Cdc42 and Rac-mediated cortical actin polymerization by directly binding to and activating the actin
nucleation activity of the actin related protein (Arp) 2/
3 complex (Higgs and Pollard, 1999). Arp2/3 complex
activity is essential for the protrusion of the leading
edge during cell migration (Borisy and Svitkina, 2000;
Pollard et al., 2000).
Activation of Cdc42 and Rac are also closely
coordinated with tyrosine kinase signaling pathways.
In the case of receptor tyrosine kinases such as the
EGFR or PDGFR, ligand binding resulting in cortical
actin rearrangements responsible for membrane ru‚ing
and motility are mediated by downstream activation
Cdc42 and Rac (Liu and Burridge, 2000). EGFR and
PDGFR activation also result in activation of cellular
Src (c-Src, or Src) family nonreceptor tyrosine kinases
(Parsons and Parsons, 1997). The activity of Src and
related kinases are responsible for the tyrosine
phosphorylation of numerous actin-binding proteins
that impact cortical actin cytoskeleton organization
and cell shape (Brown and Cooper, 1996; Kellie et al.,
1991). This is highlighted in cells expressing oncogenic
forms of Src, which causes dramatic alterations in the
actin cytoskeleton resulting in aberrant cell adhesion
and morphology (Jove and Hanafusa, 1987). While
Cdc42, Rac and Src are all activated during integrinmediated cell adhesion, divergent pathways are utilized
in regulating GTPase and Src activity, suggesting that
tyrosine phosphorylation of cytoskeletal Src substrates
can occur independently of Cdc42 and Rac activation
(Clark et al., 1998). Src-mediated tyrosine phosphor-
Cortactin and membrane dynamics
SA Weed and JT Parsons
ylation creates binding sites for proteins with Src
homology (SH) 2 domains, allowing for phosphotyrosine-based assembly of macromolecular signaling
complexes as well as directly in¯uencing substrate
activity (Parsons and Parsons, 1997; Brown and
Cooper, 1996). It is now clear that proteins involved
in both Rho family GTPase and tyrosine kinase
signaling pathways play important cooperative and
integrative roles in cortical actin cytoskeletal signaling
(Zigmond, 1996; Parsons, 1996). In this review we will
consider the structure and role of the Src substrate and
cortical actin-associated protein cortactin, with emphasis on signaling pathways involved in cortactin tyrosine
phosphorylation as well as the role of cortactin in
dynamic membrane events involving Arp2/3-mediated
cortical actin reorganization.
Structural organization of cortactin and related proteins
Cortactin was initially identi®ed as a tyrosine phosphorylated protein in v-Src infected chicken embryo
®broblasts (Kanner et al., 1990). Subsequent cloning of
the cDNA encoding cortactin revealed a novel protein
with a unique domain structure (Wu et al., 1991).
Structural predictions based on the amino acid
sequence indicate that cortactin is comprised of several
domains (Figure 1). The ®rst 84 ± 94 amino-terminal
residues are largely unstructured, containing a large
number of acidic residues between amino acids 15 and
35 and is referred to as the amino terminal acidic
domain (NTA; Weed et al., 2000). The NTA region is
Figure 1 Structure of cortactin and comparison with related
proteins. Domain organization of murine cortactin (top), HS1
(middle) and Abp1 (bottom). Percent amino acid sequence
similarities for homologous domains of cortactins from other
species are shown compared to the murine protein. The percent
similarity of related domains in HS1 and Abp1 are indicated.
Boxed numbers denote complete cortactin repeats. D. melanogaster contains four, S. purpuratus ®ve and HS1 three complete
tandem repeats, and were therefore compared to the corresponding repeats in cortactin. NTA, amino-terminal acidic region; helix,
predicted a-helical region; P-rich, proline-rich region; SH3, Src
homology 3 domain; ADF-H, actin-depolymerizing factor
homology domain. Genebank accession numbers for the analysed
sequences are U03184 (murine), M98343 (human), M73705
(chicken), AB009998 (D. melanogaster) and AF064260 (S.
purpuratus) cortactin, D42120 (HS1) and NM013810 (mAbp)
followed by a series of six complete 37 amino acid
tandemly repeating segments and one incomplete
segment 20 residues in length. The complete repeats
are predicted to form a helix-turn-helix structure (Wu
et al., 1991) and have been termed cortactin repeats
(Sparks et al., 1996a) since they show no sequence
similarity to other known repeats except in HS1 (see
below). The repeats region is followed by a a-helical
domain 48 ± 52 amino acids in length, a proline-rich
domain of variable length abundant in tyrosine, serine
and threonine residues, and a Src homology (SH) 3
domain at the distal carboxyl terminus. The cortactin
SH3 domain shares signi®cant homology to SH3
domains from Src-family kinases, various adapter and
cytoskeletal proteins (Wu et al., 1991; Sparks et al.,
1996a).
Cortactin genes have now been cloned from a
number of diverse organisms. In addition to chicken,
complete cDNA sequences have been reported for
human (Schuuring et al., 1993), rat (Ohoka and Takai,
1998), mouse (Zhan et al., 1993; Miglarese et al., 1994),
Drosophila (Katsube et al., 1998) and the sea urchin
Strongylocentrouts purpuratus (Genebank accession
#AF064260). All cortactins are structurally similar to
each other, with the SH3 domain showing the greatest
amount of evolutionary conservation while the prolinerich domain the least (Figure 1). With the exception of
the proline-rich domain, vertebrate cortactins show a
particularly strong similarity, demonstrating greater
than 87% identity in all other domains. Invertebrate
cortactins show a similar pattern, with the SH3 domain
at least 71% identical while the proline-rich regions are
less than 20% homologous to the mouse form. The
large degree of conservation in the SH3 domain
suggests that this domain binds ligands conserved
across a broad evolutionary spectrum, performing a
similar function(s) in diverse species. The proline-rich
segments in invertebrate cortactins are considerably
longer than the similar regions in the vertebrate
proteins, accounting in part for the low degree of
homology (Katsube et al., 1998).
Only one cortactin gene has been identi®ed to date,
and in most cell types is responsible for a single
cortactin protein product that often migrates as two
separate bands in SDS ± PAGE with relative molecular
weights of 80 and 85 kDa (Wu et al., 1991). This is
considerably larger than the predicted 61 ± 63 kDa
molecular mass based on the primary sequence (Wu
et al., 1991; Miglarese et al., 1994) and is presumably
due to the proline-rich domain in the carboxyl
terminus. The two observed cortactin bands likely
represent di€erent conformational states, since both
forms are observed following expression of recombinant cortactin constructs in eukaryotic and prokaroytic
systems (Wu et al., 1991; Huang et al., 1997). This is
further supported by the observation of a single
cortactin band following denaturing SDS ± PAGE in
5 M urea (Huang et al., 1997). However, cortactin
isoforms have been recently identi®ed in rat brain that
are products of alternative mRNA splicing (Ohoka and
Takai, 1998). These shorter forms result from the
6419
Oncogene
Cortactin and membrane dynamics
SA Weed and JT Parsons
6420
removal of exons from the more prevalent original
form (termed cortactin-A) resulting in the translation
of proteins with either ®ve (cortactin-B) or four
(cortactin-C) complete tandem repeats. Currently it is
not clear if alternative spliced forms of cortactin are
present in other tissues.
Northern blot analysis indicates that cortactin is
expressed in nearly all mammalian tissues (Miglarese et
al., 1994; Du et al., 1998). One notable exception is in
hematopoietic cells, where the related protein HS1 is
expressed in place of cortactin (Kitamura et al., 1989).
HS1 is structurally related to cortactin, with the NTA,
cortactin repeats and SH3 domain showing the greatest
homology. HS1 di€ers from cortactin in that it
contains only three complete cortactin repeats (Figure
1), and that the helical and proline-rich domains are
inverted with respect to cortactin (Kitamura et al.,
1995). In spite of these di€erences, HS1 shares
functional similarities with cortactin, serving as a Srcfamily tyrosine kinase substrate (Takemoto et al., 1995;
Brunati et al., 1999) as well as playing a unique and
essential role in antigen-receptor induced B cell clonal
expansion (Taniuchi et al., 1995).
Another protein that structurally resembles cortactin
is murine actin-binding protein 1 (mAbp; Kessels et al.,
2000). The carboxyl terminus of mAbp1 contains a
helical, proline-rich and SH3 domain that shares
moderate sequence similarity with the cognate domains
in cortactin (Figure 1). While mAbp1 does not contain
a NTA domain or cortactin repeats, the amino
terminus contains an actin depolymerizing factor
homology (ADF-H) domain, an actin-binding domain
found in a number of cytoskeletal proteins (Lappalainen et al., 1998). mAbp1 and cortactin appear to be
functionally similar in many respects (see below).
Cortactin as a target for tyrosine and serine/threonine
protein kinases
Following the observation that cortactin is tyrosine
phosphorylated in cells transformed by v-Src, a wealth
of data have indicated that cortactin tyrosine phosphorylation is closely correlated with Src activity in a
number of signaling pathways (Table 1). Src kinase
activity is downregulated following phosphorylation of
a carboxyl-terminal tyrosine (codon 527 in chicken) by
C-terminal Src Kinase (Csk), which creates an
intramolecular binding site for the Src SH2 domain
that in turn holds the kinase in a conformationally
inactive state (Nada et al., 1991). While cells from
Csk7/7 mouse embryos display elevated kinase activity
of the Src family members Src, Fyn and Lyn (Imamoto
and Soriano, 1993; Nada et al., 1993), studies in Src7/7
Csk7/7 and Fyn7/7 Csk7/7 ®broblasts indicate that Src
is primarily responsible for cortactin tyrosine phosphorylation resulting from Csk knockout (Thomas et
al., 1995). In accordance with these studies, it is now
clear that activation of Src in response to diverse cell
stimuli points to a direct physiological role for Srcmediated cortactin tyrosine phosphorylation. Contin-
Oncogene
uous stimulation of quiescent ®broblasts with FGF-1
(bFGF) or EGF leads to a rapid induction of cortactin
tyrosine phosphorylation (within 5 min) that decreases
within 1 h, increases again within 4 ± 7 h and is
sustained for 12 ± 24 h (Maa et al., 1992; Zhan et al.,
1993). While the signi®cance of the biphasic nature of
cortactin tyrosine phosphorylation under these conditions is not clear, overexpression of Src enhances EGFinduced cortactin tyrosine phosphorylation (Maa et al.,
1992) while activation of Src and coimmunoprecipitation with cortactin is achieved following FGF-1
treatment (Zhan et al., 1994). These studies support a
role for Src in growth factor-induced cortactin tyrosine
phosphorylation. In addition, Src has been implicated
as the kinase responsible for tyrosine phosphorylation
of cortactin in cell-matrix and cell-cell adhesion events.
In NIH3T3 ®broblasts, a5b1 integrin-mediated cell
adhesion and spreading on ®bronectin results in
elevated cortactin tyrosine phosphorylation (Vuori
and Ruoslahti, 1995), a condition where Src is
activated following binding to activated focal adhesion
kinase (FAK) (Schaller et al., 1994). In N18 neuroblastoma cells, cortactin is tyrosine phosphorylated
following binding of N-syndecan (syndecan 3) with
heparin-binding growth-associated molecule (HBGAM), and both cortactin and Src copurify following
anity chromotography using the N-syndecan cytoplasmic domain (Kinnunen et al., 1998). Similarly,
binding of CD44 to hyaluronic acid (HA) in ovarian
tumor cells leads to activation and association of Src
with the CD44 cytoplasmic domain and subsequent
cortactin tyrosine phosphorylation (Bourguignon et al.,
2001). Engagement of avb3 integrin with soluble
vitronectin, binding of ICAM-1 to aLb2 or a4b1
integrin, as well as peroxide-mediated injury in
endothelial cells also induces Src activation and
subsequent cortactin tyrosine phosphorylation (Durieu-Trautmann et al., 1994; Bhattacharya et al., 1995;
Li et al., 2000). Finally, invasion of epithelial cells by
either Shigella ¯exneri (Dehio et al., 1995) or
Chlamydia trachomatis (Fawaz et al., 1997) leads to
increased cortactin tyrosine phosphorylation, and in
the case of Shigella, Src overexpression enhances
cortactin tyrosine phosphorylation upon bacteria entry.
Collectively, the strong correlation between Src activation and cortactin tyrosine phosphorylation in these
studies suggest that cortactin is phosphorylated directly
by Src (Table 1).
Coimmunoprecipitation of Src with cortactin suggests that Src directly binds cortactin (Zhan et al.,
1994; Okamura and Resh, 1995; van Damme et al.,
1997). In support of this, cortactin directly interacts
with the recombinant SH2 domain of Src and other
SH2 domain-containing proteins, an interaction that is
prevented in the presence of a phosphopeptide containing the optimal Src SH2 binding sequence (-pYEEI-;
Songyang et al., 1993; Okamura and Resh, 1995).
These results indicate that tyrosine phosphorylation of
cortactin creates phospho-dependent binding site(s) for
Src and perhaps other SH2 domain-containing proteins. Src directly phosphorylates cortactin in vitro
Cortactin and membrane dynamics
SA Weed and JT Parsons
Table 1
Type of stimulis
Growth factors
FGF-1
EGF
PDGF
CSF-1
Thromboxane A2
Adhesion receptors
a5b1 integrin
aIIBb3 integrin
avb3 integrin
syndecan 3
CD44
ICAM-1
Bacterial invasion
S. flexneri
C. trachomatis
Oncogenic transformation
V-Src infection
Csk7/7
Others
Ca2+
m-opioids
H2O2
Cell shrinkage
6421
Cellular conditions that promote cortactin tyrosine phosphorylation
Cell type
Implicated kinase
Comments
fibroblast
fibroblast
fibroblast
fibroblast
platelet
Src
Src
Fer
Fer
Syk
fibroblast
platelet
Src
Syk
endothelia
neuroblastoma
ovarian carcinoma
endothelia
Src
Src
Src
Src
epithelia
epithelia
Src
?
Dehio et al. (1995)
Fawaz et al. (1997)
fibroblast
fibroblast
v-Src
Src
Wu et al. (1991)
Nada et al. (1994)
keratinocyte
neuron
endothelia
epithelia
Fyn
?
Src
Fer
Enhanced with Src expression
Ectopically expressed
Binding to FN
Binding to thrombin or fibrinogen;
downstream of Src
Binding to VN
Binding to HB-GAM
Binding to HA
Binding to aLb2 or a4b1 integrin
Increased during differentiation
Downstream of Fyn
Reference
Zhan et al. (1993)
Maa et al. (1992)
Kim and Wong et al. (1998)
Kim and Wong (1998)
Gallet et al. (1999)
Vuori and Ruoslahti (1995)
Rosa et al. (1997);
Gao et al. (1997)
Bhattachayra et al. (1995)
Kinnunen et al. (1998)
Bourguignon et al. (2001)
Durieu-Trautmann et al. (1994)
Calautti et al. (1995)
Mangoura (1997)
Li et al. (2000)
Kapus et al. (1999)
Table abbreviations: FGF-1, ®broblast growth factor-1; EGF, epidermal growth factor; PDGF, platelet-derived growth factor; TxA2,
thromboxane A2; FN, ®bronecton; VN, vitronectin; HB-GAM, heparin-binding growth-associated molecule; HA, hyaluronic acid
(Huang et al., 1997) on three tyrosine residues in
murine cortactin (tyrosine-421, tyrosine-466 and tyrosine-482) located within the proline-rich domain
(Huang et al., 1998) (Figure 2a). Expression of a
cortactin construct where tyrosines 421, 466 and 482
were mutated to phenylalanine dramatically reduced
cortactin tyrosine phosphorylation following expression in v-Src transformed 3T3 ®broblasts, indicating
that these residues are targeted by Src in vivo (Huang et
al., 1998). Although the sequential order by which Src
phosphorylates cortactin is unknown, the presence of a
consensus Src SH2 binding sequence next to tyrosine
421 (Figure 2c) raises the possibility that this residue
may be initially phosphorylated by Src, thereby
allowing for the stabilization of Src with cortactin
through an interaction between the Src SH2 domain
and phosphorylated tyrosine 421. Stabilization of Src
at tyrosine 421 may act to further facilitate the
phosphorylation of cortactin at tyrosines 466 and
482. Evidence for the importance of Src SH2 domain
binding in cortactin tyrosine phosphorylation is highlighted by the marked reduction of tyrosine phosphorylation of cortactin in Csk7/7 cells treated with Src
SH2 inhibitory compounds (Violette et al., 2001).
While phosphorylation of cortactin by Src appears to
be functionally linked to some aspects of actin
organization and cell motility, the full signi®cance of
Src mediated cortactin tyrosine phosphorylation is not
yet understood.
A second nonreceptor tyrosine kinase that is
involved in growth factor-mediated cortactin tyrosine
phosphorylation is Fer. Fer, along with Fps/Fes,
comprise a small and distinct tyrosine kinase subfamily
that is has unique structural characteristics, including
the presence of an amino terminal coiled-coil sequence
proximal to a central SH2 domain and a carboxyl
terminal catalytic domain (Hao et al., 1989). NIH3T3
cells treated with PDGF leads to Fer activation and
cortactin tyrosine phosphorylation (Downing and
Reynolds, 1991; Kim and Wong, 1998) which is
markedly reduced in ®broblasts derived from Fer7/7
mice (Craig et al., 2001). Fer immunoprecipitates with
cortactin, and the Fer SH2 domain interacts with
cortactin in cell extracts, suggesting that Fer directly
interacts with and phosphorylates cortactin. This is
further supported by the suppression of CSF-1
receptor-mediated cortactin tyrosine phosphorylation
by expression of kinase de®cient Fer (Kim and Wong,
1998). Other evidence supporting a role for Fer in
cortactin tyrosine phosphorylation has come from the
analysis of signaling events involved in cell shrinkage.
Hyperosmolarity elevates tyrosine phosphorylation of
cortactin as well as the activity of the Fyn, while
suppressing Src activity (Kapus et al., 1999). The
requirement for Fyn in this process is further underscored
by the reduction of cortactin tyrosine phosphorylation in
hyperosmotic Fyn7/7, but not Src7/7 cells (Kapus et
al., 1999). A rapid reduction in cell volume also
activates Fer, which is suppressed by pretreatment with
the Src family inhibitor PP1 as well as in Fyn7/7 cells
(Kapus et al., 2000). Expression of kinase de®cient Fer
also inhibits osmotic-induced cortactin tyrosine phosphorylation, and hyperosmotic conditions fail to
signi®cantly increase the tyrosine phosphorylation level
Oncogene
Cortactin and membrane dynamics
SA Weed and JT Parsons
6422
Figure 2 Diagrammatic representation of cortactin-interacting proteins and sequence alignments of known and putative
interaction sites. (a) Mapping of cortactin binding proteins and phosphorylation sites. Binding of Arp2/3 complex occurs through
the three amino acid ± DDW -motif within the NTA domain. The actin-binding domain is located within the repeats region,
requiring the fourth repeat and possibly adjacent sequences. Tyrosine phosphorylation by Src and Fer occurs within the proline-rich
domain at Y421, Y466 and Y482, creating putative binding sites for SH2 domains. Putative serine phosphorylation by ERK1/2
occurs at S405 and S418. The SH3 domain interacts with numerous PDZ-containing sca€olding proteins as well with dynamin and
CBP90. See text for details. (b) The cortactin Arp2/3 binding site is conserved among other Arp2/3 binding proteins. Sequence
alignment of various Arp2/3 activating proteins with the cortactin Arp2/3 binding site reveals a conserved DDW- or -DEWsequence (except for Scar-1) (boxed) required for Arp2/3 binding and activation. Displayed sequences are from Miglarese et al.
(1994) (cortactin), Kitamura et al. (1995) (HSl), Drubin et al. (1990) (Abp1p), Dujon et al. (1994) (myo3p), Derry et al. (1995)
(WASP), Fukuoka et al. (1997) (N-WASP) and Bear et al. (1998) (Scar-1). (c) Src tyrosine phosphorylation sites on cortactin
conform to a consensus SH2-binding motif. Comparison of cortactin phosphorylation sites with known SH2 domain-binding motifs
that mediate stable complex formation with activated Src. P0, position of Src-phosphorylated tyrosine; P1 ± P3, position of ¯anking
residues; c, aliphatic residues. Sequence data is from Songyang et al. (1993) (consensus), Schaller et al. (1994) (FAK), Nakamoto et
al. (1996) (CAS) and Guappone et al. (1998) (AFAP) and Huang et al. (1998) (cortactin). (d) Alignment of proline-rich sequences
contained within cortactin SH3 domain-binding proteins. Proteins demonstrated or proposed to bind to the cortactin SH3 domain
and their putative or con®rmed SH3 binding sites(s) are aligned with the consensus binding sequence (top) obtained by peptide
phage display library screening (Sparks et al., 1996). Proteins with a consensus-conforming -KPPVPPKP- sequence are displayed in
the upper half of the alignment (binding of mXin has yet to be demonstrated) while non-conforming proteins and their interaction
sequences are shown underneath. Boxes indicate the positions of conserved proline residues. Sequences are from Du et al. (1998)
(CortBP1), Naisbitt et al. (1999) (Shank 3), Wang et al. (1999) (mXin), Takahisa et al. (1996) (ZO-1), Ohoka and Takai (1998)
(CBP90), and Cook et al. (1994) (Dynamin)
of a cortactin construct containing the Src-targeted
point mutations at tyrosines 421, 466 and 482 (Kapus
et al., 2000). Collectively these data indicate that
osmotic-induced tyrosine phosphorylation of cortactin
by Fer requires upstream activity of Fyn, and that Fer
directly phosphorylates one or more of the same
tyrosine residues targeted by Src. The functional
signi®cance of Fer-mediated cortactin tyrosine phosphorylation is unclear. It is also not known if Fer is
responsible for the Fyn-mediated increase in cortactin
tyrosine phosphorylation observed during calciumOncogene
induced keratinocyte di€erentiation (Calautti et al.,
1995).
Cortactin tyrosine phosphorylation is also observed
during platelet di€erentiation and activation. Although
Src activation and cortactin tyrosine phosphorylation
are coincident with thrombin-mediated platelet activation following binding to aIIBb3 integrin (Fox et al.,
1993), there is also evidence that the hematopoieticspec®c kinase Syk plays an important role in cortactin
tyrosine phosphorylation during platelet maturation
and activation. Syk is homologous to the lymphocyte-
Cortactin and membrane dynamics
SA Weed and JT Parsons
speci®c ZAP-70 kinase, containing two amino-terminal
SH2 domains and a carboxyl-terminal catalytic domain
(Chan et al., 1992; Muller et al., 1994). Syk is activated
following ®brinogen ligation to aIIBb3 integrin, and
requires the intact cytoplasmic domains of both
integrin subunits (Gao et al., 1997). Defects in aIIBb3
integrin are found in individuals with Glanzmann
thrombasthenia, which experience excessive bleeding
due to defective platelet activation and aggregation
(Nurden and Caen, 1974). Platelets from Glanzmann
patients display reduced cortactin tyrosine phosphorylation following ®brinogen stimulation (Rosa et al.,
1997) indicating that aIIBb3 integrin-induced activation
of Syk is required for cortactin tyrosine phosphorylation. Syk autophosphorylation is inhibited in the
presence of catalytic inactive Src, indicating that Src
activity is required for Syk activation and subsequent
cortactin tyrosine phosphorylation in a manner
analogous to that of the Fyn/Fer pathway described
above. In addition, integrin-independent cortactin
tyrosine phosphorylation attributable to Syk activation
occurs following treatment of platelets with thromboxane analogues and in TPA-induced megakaryocyte
di€erentiation of K562 leukemic cells (Maruyama et
al., 1996; Gallet et al., 1999). TPA treatment of K562
cells leads to Src activation in the presence of kinase
inactive Syk, further suggesting that Src acts upstream
of Syk (Maruyama et al., 1996). Furthermore,
thromboxane stimulation of platelets does not elevate
Src kinase activity, indicating that Syk can act
independent of Src in phosphorylating cortactin (Gallet
et al., 1999). In these cases, cortactin forms complexes
with Syk prior to simulus addition, suggesting that the
interaction between Syk and cortactin is phosphotyrosine independent. Although the role of Syk-induced
cortactin phosphorylation is not known, it likely plays
an important role in regulating platelet morphology
(Gallet et al., 1999).
In addition to serving as a tyrosine kinase substrate,
cortactin is also phosphorylated on serine and
threonine residues (Wu et al., 1991). Serine and
threonine phosphorylation of cortactin is enhanced in
human squamous carcinoma cell lines following
stimulation with EGF, causing a shift from the 80 to
the 85 kDa cortactin form (van Damme et al., 1997).
Treatment of 293 cells with EGF leads to a similar
shift that is blocked by the MEK inhibitor PD98059,
indicating that EGF-induced activation of the RasRaf-MEK-ERK pathway is responsible for cortactin
serine and threonine phosphorylation (Campbell et al.,
1999). Based on peptide mapping data, ERK1/2 has
been proposed to phosphorylate cortactin at two ERK
consensus sites, serine-405 (-403PVSP) and serine-418
(-416PSSP) within the proline-rich domain (Campbell
et al., 1999). MEK activates a number of serine/
threonine kinases other than ERKs, which may in part
contribute to additional cortactin phosphorylation
events (Kolch, 2000). Since serine 405 and 418 lie
within putative SH3 binding sites, and although these
sequences do not conform to known cortactin SH3
domain-binding sequences (Figure 2d), it has been
proposed that ERK-mediated serine phosphorylation
may regulate an intramolecular interaction between the
cortactin SH3 domain and sequences within the
proline-rich domain. This would render cortactin in
an `open' conformation, allowing binding of the SH3
domain to other ligands (Campbell et al., 1999). This
model has been proposed for other proteins (Taylor et
al., 1998), and is indirectly supported by the presence
of a single 85 kDa band expressed from a cortactin
construct containing a disruptive point mutation within
the SH3 domain (Du et al., 1998). Alternatively, serine/
threonine phosphorylation of cortactin within the
proline-rich domain may regulate SH3-mediated binding of heterologous proteins, as has been demonstrated
for Sos and FAK (Corbalan-Garcia et al., 1996; Ma et
al., 2001).
6423
The role of cortactin in cortical actin organization
Localization and biochemical studies point to cortactin
playing an important role in regulating cortical actin
assembly and organization. In most cell types cortactin
localizes in cytoplasmic punctate structures of unknown composition concentrated at the perinuclear
region, and also with F-actin at sites of dynamic
peripherial membrane activity (Figure 3a,b). Cortactin
translocates from the cytoplasm to the periphery in
response to many of the same stimuli that induce its
tyrosine phosphorylation, including growth factor
treatment, integrin activation and bacterial entry
(Ozawa et al., 1995; Weed et al., 1998; Cantarelli et
al., 2000). These events also lead to activation of Rac
(Hartwig et al., 1995; Clark et al., 1998; Mounier et al.,
1999), which together with Cdc42 are responsible for
controlling the formation of cortical actin networks
(Ridley et al., 1992; Kozma et al., 1995). Rac activation
is required for cortactin translocation to the cell
periphery (Weed et al., 1998), and cortactin disperses
from cortical actin networks following treatment with
the F-actin disrupting drug cytochalasin D (Wu and
Parsons, 1993). Cortical localization of cortactin is
therefore closely associated with Rac-mediated actin
assembly.
In agreement with its localization to the cortical
actin network, cortactin interacts directly with F-actin
®laments through sequences in the repeats domain (Wu
and Parsons, 1993). Deletion mapping analysis has
determined that the fourth cortactin repeat is responsible for F-actin binding and cortical localization
(Weed et al., 2000). Cortactins -A, -B and -C all bind
F-actin (Ohoka and Takai, 1998) whereas the ®rst
three cortactin repeats do not (He et al., 1998),
con®rming the importance of the fourth cortactin
repeat in F-actin binding. Cortactin binds to the sides
of actin ®laments as determined by negative stained
electron microscopy of recombinant cortactin/F-actin
mixtures (H Wu and JT Parsons, unpublished observations) and by competition binding experiments with
myosin subfragment 1 (Ohoka and Takai, 1998). HS1
also binds actin through the amino-terminal cortactin
Oncogene
Cortactin and membrane dynamics
SA Weed and JT Parsons
6424
Figure 3 Cortactin localizes at sites of dynamic actin assembly.
Colocalization of cortactin (a, c, e) with F-actin-containing
cellular structures (b, d, f). (a) and (b) Localization of cortactin
and F-actin in Swiss 3T3 cells spreading on ®bronectin. Cortactin
selectively localizes with cortical F-actin within lamellipodia and
not with the stress ®ber network. A signi®cant pool of cortactin is
also found in the perinuclear region. (c) and (d) Redistribution of
cortactin and F-actin into podosomes in v-Src transformed
NIH3T3 cells. In addition to its cortical and perinuclear
localization, cortactin is also present in `rosettes' or podosomes
resultant from transformation by oncogenic non-receptor tyrosine
kinases. (e) and (f) Cortactin localization is coincident with Factin tails in Listeria monocytogenes-infected Swiss 3T3 cells.
Arrowheads denote sites of cortactin and F-actin colocalization
repeats domain, and the fourth repeat in cortactin as
well as the three repeats in HS1 bind to phosphatidylinositol 4,5 bisphosphate (PIP2) (He et al., 1998),
which has been shown to regulate the interaction of
several actin-binding proteins with F-actin (Sechi and
Wehland, 2000). Recombinant cortactin has been
reported to crosslink F-actin (Huang et al., 1997), an
activity attributed to the sixth cortactin repeat (Ohoka
and Takai, 1998). This is somewhat controversial, since
recombinant constructs lacking the fourth but containing the sixth repeat fail to bind F-actin (Weed et al.,
2000) and crosslinking has not been detected by other
groups under similar conditions (H Wu and JT
Parsons, unpublished observations; Weaver et al.,
Oncogene
2001). Whereas binding of cortactin to F-actin is not
in¯uenced by Src phosphorylation (Wu and Parsons,
1993), Src phosphorylation has been reported to inhibit
the crosslinking ability of cortactin (Huang et al.,
1997). In addition, PIP2 binding to cortactin also
inhibits F-actin crosslinking (He et al., 1998), although
in this case it is not known if this is due to an inability
of the cortactin/PIP2 complex to bind actin. Since
cortactin contains a single F-actin binding site (Weed
et al., 2000) and multimeric forms of cortactin have not
been detected (Ohoka and Takai, 1998; Wu and
Parsons, unpublished observations) cortactin must
crosslink F-actin by an undetermined mechanism that
will require further elucidation.
Activated (GTP-bound) Cdc42 and Rac interact with
a number of proteins that regulate assembly of the
cortical actin cytoskeleton (Aspenstrom, 1999). It is
now well established that members of the Wiskott ±
Aldrich syndrome protein family, or WASps, play a
pivotal role in mediating Cdc42 and Rac e€ects on the
actin cytoskeleton (Takenawa and Miki, 2001). The
prototypic member, WASp, binds directly to Cdc42
and induces clusters of actin ®laments in cells
dependent on the unmasking of a C-terminal acidic
region (Machesky and Insall, 1998). Expression of
WASp is restricted to lymphoid and hemopoietic cells
while a second family member closely related to
WASp, termed N-WASp, also directly binds Cdc42.
N-WASp is ubiquitously expressed and contributes
directly to Cdc42-induced ®lopodia formation in most
cell types (Miki et al., 1998a). WASp and N-WASp are
also activated by molecules other than Cdc42,
including PIP2 (Higgs and Pollard, 2000; Rohatgi et
al., 2000), the proteins WISH (Fukuoka et al., 2001)
and WIP (Martinez-Quiles et al., 2001), indicating that
multiple signaling pathways are utilized in WASp
activation. Another WASP-like protein, termed Scar1
or WAVE, plays a central role in Rac-induced cortical
actin dynamics (Bear et al., 1998; Miki et al., 1998b).
Scar1/WAVE is structurally similar to WASp, sharing
strong homology to the carboxy-terminal acidic
domain (Miki et al., 1998b). The activation of Scar1/
Wave by Rac is indirect and is mediated by IRSp53
(Miki et al., 2000). A Rac/IRSp53 complex directly
binds and activates Scar1/WAVE, which in turn
promotes lamellipodia and membrane ru‚ing. Scar1/
Wave colocalizes with cortical actin within membrane
ru‚es induced by activated Rac (Miki et al., 1998b).
These data indicate that Scar1/WAVE functions in a
manner similar to N-WASP to promote actin polymerization within lamellipodia.
The Arp2/3 protein complex is now ®rmly established in directly mediating signals from the Cdc42/NWASP (Rohatgi et al., 1999) and Rac/IRSp53/Scar1/
WAVE (Machesky and Insall, 1998) pathways to
induce polymerization of the cortical actin cytoskeleton. The Arp2/3 complex consists of the two
unconventional actins Arp (actin-related protein) 2
and 3 along with ®ve unrelated proteins (Pollard et al.,
2000). Arp2/3 complex binds to the sides and pointed
(or slow-growing) ends of actin ®laments (Mullins et
Cortactin and membrane dynamics
SA Weed and JT Parsons
al., 1998), nucleating de novo actin polymerization,
producing barbed end (fast growing) `daughter'
®laments attached to the sides of preexisting `mother'
®laments (Machesky et al., 1999; Amann and Pollard,
2001). Binding of the carboxy-terminal acidic region of
WASp family proteins enhances Arp2/3-mediated actin
nucleation (Machesky et al., 1999) and in the case of
Scar1/Wave, this region binds directly to the 21 kDa
Arp2/3 subunit (Machesky and Insall, 1998). This
interaction forms a molecular complex indirectly
linking Rac to the actin cytoskeleton through IRSp53.
For N-WASp, the interaction is direct since Cdc42
directly binds N-WASp to stimulate ®lopodia (Miki et
al., 1998a). The observation that carboxy-terminal
fragments of WASP-family proteins are more active
than the full-length protein at promoting actin
nucleation coupled with the requirement for activated
Cdc42 or Rac to elicit actin polymerization has led to
the proposal that activation of N-WASp or Scar1/
WAVE is achieved by GTPase-mediated targeting to
the cortical membrane. This is followed by a
conformational change exposing the carboxy-terminal
acidic domain to allow binding of the Arp2/3 complex
and subsequent actin polymerization (Machesky et al.,
1999; Rohatgi et al., 1999). The sidebinding, pointed
end capping and nucleation ability of Arp2/3 complex
results in the formation of Y-shaped branched networks observed in vitro (Mullins et al., 1998; Blanchoin
et al., 2000a) and in vivo (Svitkina and Borisy, 1999).
Phosphate released from aged ADP-Pi actin ®laments
leads to disassembly of Arp2/3-induced actin networks
at branch points (Blanchoin et al., 2000a,b). The
debranching process is enhanced by the activity of
actin depolymerizing factor (ADF)/co®lin, which also
depolymerizies and severs F-actin ®laments (Maciver,
1998). ADF/co®lin localizes at sites distal to that of
Arp2/3 complex, placing it in the proper position to
breakdown Arp2/3-induced actin networks within
lamellipodia (Svitkina and Borisy, 1999). Co®lin
severing activity is abolished following phosphorylation
by the Rac e€ector LIM kinase (Yang et al., 1998),
which in turn is activated by a second Cdc42/Rac
e€ector, PAK (Edwards et al., 1999). The coordinated
activities of LIM kinase and PAK therefore serve to
attenuate actin networks resultant from Arp2/3 activation.
In addition to WASp family proteins, cortactin also
binds to Arp2/3 complex and stimulates actin nucleation activity (Weed et al., 2000; Weaver et al., 2001;
Uruno et al., 2001). Binding of cortactin to Arp2/3 is
mediated through a three amino acid motif in the NTA
domain (-20DDW) conserved in many Arp2/3 activating proteins (Figure 2b). Cortactin's ability to stimulate
Arp2/3 activity requires both the DDW motif and the
actin binding domain (Weaver et al., 2001; Uruno et
al., 2001), domains that are also required for Racmediated localization of cortactin to cortical actin
networks (Weed et al., 2000). While the ability of
cortactin to stimulate Arp2/3 activity is weaker than
that of N-WASp, cortactin enhances N-WASp-induced
Arp2/3 activation (Weaver et al., 2001). This synergy
suggests that cortactin can function cooperatively with
WASp proteins in addition to independently activating
Arp2/3 complex. Evidence for an alternative mechanism of Arp2/3 activation by cortactin is supported by
the lower anity of cortactin for Arp2/3 complex
(Weed et al., 2000) and by the inability of cortactin to
interact with individual Arp2/3 subunits in a yeast twohybrid assay (LM Machesky, personal communication). In addition, N-WASP and cortactin are di€erentially localized in Vaccinia-induced actin tails (Zettl
and Way, 2001). Cortactin also inhibits debranching of
aged Arp2/3 networks (Weaver et al., 2001) suggesting
that cortactin also serves to stabilize the interaction
between Arp2/3 complex and mother/daughter ®lament
networks. The association of cortactin with mobile
cortical actin structures determined by live cell imaging
points to a functional role for cortactin in regulating
Arp2/3 activity (Dai et al., 2000; Kaksonen et al.,
2000).
Activation of Cdc42 and Rac are required for cell
migration (Nobes and Hall, 1999), with the resultant
Arp2/3 activity widely believed to be responsible for
protrusive-based cell motility mediated by these
GTPases (Borisy and Svitkina, 2000). The ability of
cortactin to bind and activate Arp2/3 complex
indicates a role for cortactin in actin-based motility
events. Enhanced cell migration in NIH3T3 and
endothelial cells overexpressing cortactin supports this
notion (Patel et al., 1998; Huang et al., 1998). While
cortactin stimulated cell motility can be attributed to
enhanced Arp2/3 activity, Src phosphorylation of
cortactin also appears to play an important role in
cell migration. Src activation and cortactin tyrosine
phosphorylation are also closely coordinated in ®broblast cell motility induced by FGF-1 (LaVallee et al.,
1998; Liu et al., 1999) or in carcinoma cells by CD44
binding to HA (Bourguignon et al., 2001). Rac is also
activated under these conditions (Johnston et al., 1995;
Oliferenko et al., 2000), and is responsible for
recruiting Src to the cell periphery (Fincham et al.,
1996). Collectively these data point to a role for
cortactin in integrating cell motility signaling pathways
involving activation of Arp2/3 complex and Src.
Furthermore, several intracellular pathogens utilize
Arp2/3 complex-based actin polymerization for intracellular movement (Frischknecht and Way, 2001) and
cortactin is found in the `rocketing' actin tails of many
(if not all) of these microbes, including Listeria (Figure
3e,f), Shigella and Vaccinia (Zettl and Way, 2001).
Pathogen-induced Arp2/3 actin networks are structurally similar to those found in lamellipodia, suggesting
a similar mechanism of assembly (Cameron et al.,
2001). While in the case of Vaccinia, cellular infection
results in Src activation and cortactin tyrosine
phosphorylation (Frischknecht et al., 1999a), it is not
likely that cortactin tyrosine phosphorylation contributes to Vaccinia motility since tyrosine phosphorylation is restricted to the virus body and is not present in
actin tails (Frischknecht et al., 1999b). The lack of
tyrosine phosphorylation in Vaccinia (and endosome)
actin tails suggests that Src phosphorylation of
6425
Oncogene
Cortactin and membrane dynamics
SA Weed and JT Parsons
6426
cortactin does not in¯uence cortactin's ability to
activate and stabilize Arp2/3-mediated networks.
Function of cortactin SH3-domain binding proteins
Cortactin interacts with several proteins though its
carboxyl terminal SH3 domain. The binding of SH3
domains to proline-rich sequence motifs has been well
established (Kay et al., 2000) and based on screening of
phage display peptide libraries, the cortactin SH3
domain binds to a consensus sequence of +PPCPXKP
(Sparks et al., 1996b). Yeast two hybrid screening of
brain libraries with the cortactin SH3 domain led to
the identi®cation of ®rst cortactin SH3 domain-binding
protein, termed cortactin binding protein 1 (CortBP1)
(Du et al., 1998). Subsequent studies determined that
CortBP1 is identical ProSAP1 (Boeckers et al., 1999a)
and to a splice variant of Shank2 (Naisbitt et al.,
1999), which is a member of the Shank family of
sca€olding proteins (Sheng and Kim, 2000). Abundantly expressed in neurons, CortBP1 and the related
protein Shank3 contain a canonical cortactin SH3
domain-binding sequence (KPPVPPKP) (Figure 2d)
which mediates binding to cortactin (Du et al., 1998).
Shank proteins serve as molecular sca€olds at postsynaptic sites (PSD) of excitatory synapses, and are
characterized by the presence of an amino-terminal
PSD95/dlg/ZO-1 (PDZ) domain. The CortBP1 PDZ
domain directly binds to the carboxyl terminus of the
transmembrane type 2 somatostatin receptor (Zitzer et
al., 1999) and the a-latrotoxin receptor CL1 (Kreienkamp et al., 2000; Tobaben et al., 2000). In addition,
Shank1 and Shank3 contain a set of seven ankyrin
repeats and an SH3 domain proximal to the PDZ
domain (Sheng and Kim, 2000). While a ligand for the
Shank SH3 domain has not been reported, the ankyrin
repeats interact with a novel protein of unknown
function termed sharpin, which can multmerize
through an amino terminal coiled-coil region (Lim et
al., 2001). The PDZ domain of CortBP1 and Shank3
also interacts with the carboxyl terminus of GKAP
(Naisbitt et al., 1999; Boeckers et al., 1999b), a
cytosolic PSD protein that binds through its amino
terminal region to the C-terminal guanylate kinase-like
domain of PSD-95 (Kim et al., 1997). The amino
terminal region of PSD-95 contains three tandem PDZ
motifs, the ®rst two which are capable of directly
interacting with the carboxyl terminus of both the
NMDA receptor (Kornau et al., 1995) and the Shaker
K+ channel (Kim et al., 1995). Additionally, the PDZ
domain of Shank3 directly interacts with the carboxyl
terminus of the group 1 metabotropic receptors
mGluR1a and mGluR5 as well as indirectly binding
to these receptors through Homer (Tu et al., 1999).
Homer is a protein localized in PSDs that posses an
Enabled/VASP homology domain (EVH) that interacts
with a proline-rich segment on Shank3 proximal to the
cortactin binding site (Tu et al., 1999). The Homer
EVH domain is capable of simultaneously interacting
with Shank3 and either the carboxyl terminus of
Oncogene
mGluR receptors (Xiao et al., 1998) or with the
cytoplasmic amino terminus of inositol trisphosphate
receptors (IP3R) (Tu et al., 1998). Homer therefore
serves as a sca€old for coupling mGluR and IP3
receptors to Shank3, and links these receptors to
NMDA and Shaker K+ signaling pathways. CortBP1/
Shank proteins also contain a sterile alpha motif
(SAM) domain at the extreme carboxyl terminus that
serves as a homo- and/or heterodimerization domain
(Naisbitt et al., 1999). The carboxyl terminus of Homer
proteins contains a coiled-coil domain that functions in
an analogous manner to SAM domains (Xiao et al.,
1998). The ability of Shank and Homer proteins to
dimerize in various combinations contributes to the
overall complexity of receptor organization at the PSD.
The precise level of bridging between various receptors
within the PSD is likely to be further complicated given
the abundant alternative splicing that occurs in both
Shank and Homer gene products (Lim et al., 1999;
Soloviev et al., 2000). Shanks, cortactin and Arp2/3
complex are present in neuronal growth cones (Du et
al., 1998; Goldberg et al., 2000) and given the ability to
bind to both CortBP1 and Shank 3 molecular `bridges',
cortactin can serve to link multiple postsynaptic
receptors to sites of Arp2/3 mediated actin assembly
(Figure 4). Therefore, clustering and organization of
receptor components at the synapse may in part be
attributable to Arp2/3 mediated actin polymerization
regulated by cortactin in a manner similar to that
described for the movement and formation of acetylcholine receptor clusters (Dai et al., 2000). Shank
proteins have been recently observed in non-neuronal
tissues (Redecker et al., 2001), indicating that they may
also participate in receptor organization at sites other
than the PSD. It of interest to note that the cortactin
SH3 domain-binding sequence found in Shanks is also
present in the murine form of Xin, a muscle-speci®c
protein implicated in cardiac morphogenesis (Wang et
al., 1999). The cortactin SH3 domain also interacts
with a novel brain-speci®c protein of unknown
function termed CBP90 though a proline-rich sequence
that di€ers from the binding sequence in Shank
proteins (Figure 2d; Ohoka and Takai, 1998).
The epithelial tight junction is another region of
subcellular specialization where cortactin may participate in organizing transmembrane receptor networks
(Figure 4). Tight junctions form a barrier between
adjacent polarized epithelia that restricts the movement
of immune cells and solutes between cells. In addition,
they establish a di€usion barrier for integral membrane
proteins within individual cells by dividing the plasma
membrane into apical and basolateral domains
(Tsukita et al., 1999). Tight junctions are associated
with an apical cytoskeletal structure known as the
terminal web (Fath et al., 1993). Cortactin is localized
at the terminal web in human small bowel epithelium
(Wu and Montone, 1998) and the SH3 domain
associates with a proline-rich motif in ZO-1 (Katsube
et al., 1998), a protein selectively localized to the
epithelial tight junction (Stevenson et al., 1986). The
ZO-1 proline-rich domain contains a homologous
Cortactin and membrane dynamics
SA Weed and JT Parsons
6427
Figure 4 Cortactin serves to link diverse sca€olding protein complexes to Arp2/3-driven actin polymerization. Schematic
representation of the proposed molecular interactions within the neuronal post-synaptic density (top), epithelial tight junction
(bottom left) and in receptor mediated endocytosis (bottom right). In all cases, cortactin (red) is proposed to interact through its
SH3 domain with proline-rich segments within numerous sca€olding proteins, including Shank proteins, ZO-1 and dynamin 2.
These multiprotein sca€olds are in turn linked to the Arp2/3 complex (dark green) and associated actin cytoskeleton through
sequences within the cortactin NTA and repeat domains. Protein domains are: a, cortactin NTA; rep, cortactin repeats; 3, SH3; PZ,
PDZ; PP, proline-rich; SM, SAM; GK, guanylate kinase-like domain; ANK, ankyrin-like repeats; EVH, Enabled/VASP homology;
CC, coiled-coil, RH, RBCK1 homology; PH, pleckstrin homolgy; GED, GTPase e€ector domain; ADF, actin depolymerizing factor
homology; VC, verprolin/co®lin and acidic region in N-WASP
sequence to the consensus cortactin SH3 domainbinding motif (Figure 2d). The ZO-1 domain structure
is similar to that of PSD-95, containing three amino
terminal PDZ domains, an SH3 domain, a guanylate
kinase-like domain and a carboxyl-terminal proline
rich motif (Willott et al., 1993; Itoh et al., 1993). Based
on its localization and domain structure, ZO-1 is
thought to participate in the biogenesis and regulation
of tight junction assembly (Sheth et al., 1997).
Consistent with this, the guanylate kinase-like domain
of ZO-1 interacts with cytoplasmic motifs of occludin
(Furuse et al., 1994; Fanning et al., 1998) while the
®rst PDZ domain binds the carboxyl terminus of
claudin (Itoh et al., 1999a), two transmembrane
proteins that are implicated in forming the paracellular
seal between adjacent cells (Furuse et al., 1993,
1998a,b). ZO-1 contains a centrally located SH3
domain that interacts with the transcription factor
ZONAB (Balda and Matter, 2000). ZO-1 is homologous to two other tight junction associated proteins,
ZO-2 and ZO-3 (Jesaitis and Goodenough, 1994;
Haskins et al., 1998) which share the same domain
structure as ZO-1. ZO-1 forms binary complexes with
ZO-2 and ZO-3 (Wittchen et al., 1999), with ZO-1/ZO-
2 binding mediated through the second PDZ domains
of both proteins (Fanning et al., 1998; Itoh et al.,
1999b). The ®rst PDZ domain of ZO-2 and ZO-3 also
bind to the carboxyl terminus of claudin (Itoh et al.,
1999a) and the amino-terminal region of both proteins
also associate with occludin (Itoh et al., 1999b;
Haskins et al., 1998). All three ZO proteins also bind
F-actin (Wittchen et al., 1999), which in ZO-1 is
mediated through sequences in the proline-rich domain
(Fanning et al., 1998), thus linking and organizing
occludin and claudin clusters within the tight junction
membrane. The ZO-1 proline-rich domain also
mediates interaction with the Ras e€ector AF-6
(Yamamoto et al., 1997). While the interactions of
the various ZO proteins are more complex than PSD
sca€olding proteins in terms of their association with
the actin cytoskeleton, the binding of cortactin to ZO1 suggests that the tight junction complex in part is
linked to Arp2/3 complex, and further suggests that
indirect regulation of transmembrane complexes by
attachment to sites of cortical actin assembly is a
hallmark of cortactin function. How Arp2/3 driven
actin polymerization contributes to tight junction
organization remains unexplored.
Oncogene
Cortactin and membrane dynamics
SA Weed and JT Parsons
6428
Oncogene
Besides transmembrane receptor complex organization, cortactin also plays a role in coupling Arp2/3
complex-mediated actin dynamics to endocytosis and
vesicle movement by binding to dynamin 2 (Figure 4).
Dynamins comprise a family of a large GTPases that
play a fundamental role in the ®ssion and recycling of
membrane vesicles during receptor mediated endocytosis (Schmid, 1997). Dynamins contain a GTP
hydrolysis domain within the amino terminus, followed
by a linker region, a phospholipid binding pleckstrin
homology (PH) domain, a GTPase e€ector domain and
a proline-rich domain at the carboxyl terminus (Schmid
et al., 1998). The SH3 domain of cortactin can interact
with two di€erent proline motifs within the dynamin 2
proline-rich domain as determined by peptide competition studies (McNiven et al., 2000). Like cortactin,
dynamin 2 translocates into membrane ru‚es following
PDGF stimulation of NIH3T3 cells. The growth factorinduced tranlocation of dynamin 2 to lamellipodia is
inhibited in cells expressing cortactin constructs lacking
the SH3 domain, pointing to a functional requirement
for cortactin in targeting dynamin 2 to sites of
membrane ru‚ing (McNiven et al., 2000). A second
protein that may participate in localizing dynamin to
lamellipodia in a similar manner is mAbp1. The SH3
domain of mAbp1 also binds to the dynamin prolinerich region and is localized with dynamin at sites of
endocytosis (Kessels et al., 2001). mAbp1 is a Src
substrate (Lock et al., 1998), binds F-actin at two sites
through its ADF-H domain, and is translocated to the
cortical cytoskeleton following Rac activation (Kessels
et al., 2000). While the yeast homolog of mAbp1,
Abp1p, interacts with and activates Arp2/3 complex
through two DDW motifs (Goode et al., 2001), mAbp1
does not contain a DDW motif and is therefore not
likely to participate in Arp2/3 complex activation. A
third SH3 domain containing protein that couples the
actin cytoskeleton to dyamin is syndapin, which also
localizes with dynamin at sites of endocytosis (Qualmann et al., 1999). Syndapin directly binds N-WASp,
and overepxression of syndapin leads to ®lopodia
formation dependent on Arp2/3 complex activity
(Qualmann and Kelly, 2000) providing a second
molecular link between dynamin and Arp2/3 complex
in mammalian cells. Overexpression of the mAbp1 or
syndapin SH3 domain inhibits endocytosis (Kessels et
al., 2001; Qualmann and Kelly, 2000), further highlighting the functional importance of linking dynamin
to the cortical actin skeleton in endocytic events. The
association of dynamin with clathrin-coated pits is
actin-independent, and is mediated through amphiphysins I and II (Wigge and McMahon, 1998). The
amphiphysin SH3 domain binds to sequences in the
dynamin proline-rich region (David et al., 1996) while
the amino terminal region of amphiphysins interact with
clathrin (Ramjaun et al., 1997) and the clathrin-binding
adaptor protein AP2 (David et al., 1996). How the
activities of cortactin, mAbp1, syndapin and amphipysin regulate the cortical actin cytoskeleton and
coordinate with dynamin GTPase activity during
endocytic membrane ®ssion is highly complex. It has
been proposed that Arp2/3 activity is involved in
regulating membrane morphology and movement in
pre- and post-endocytic events (Qualmann et al., 2000).
Support for the involvement of cortactin in post
endocytic vesicle dynamics comes from the localization
of cortactin to actin tails of cytoplasmic rocketing
endosomes (Kaksonen et al., 2000; Schafer et al., 2000),
an activity that is dependent on Arp2/3 activation
(Taunton et al., 2000). Actin tails on endosomes are
strikingly similar to tails produced by intracellular
pathogens, suggesting that cortactin plays a similar role
by the simultaneous stimulation and stabilization of
Arp2/3-induced actin networks. Taken together, these
results suggest that Arp2/3 activation and actin network
stabilization is a conserved cellular function of
cortactin, and this activity is utilized directly for actinbased motility as well as indirectly through the
interaction of SH3 binding proteins to dynamically
regulate transmembrane protein organization.
Overexpression and role of cortactin in human tumors
A large body of evidence indicates that cortactin plays
an important role in human neoplasia. The human
cDNA encoding cortactin (termed EMS1) was isolated
by the cloning of genes found in human carcinoma cell
lines containing ampli®ed copies of the chromosome
11q13 region (Schuuring et al., 1991). The ampli®ed
region typically ranges from 3 ± 5 Mb in size and
contains several DNA markers, including the oncogenes cyclin D1 (PRAD1), INT2/FGF3, HSTF1/
FGF4 and BCL1 that are often found in a variety
of human tumors (Lammie and Peters, 1991).
Ampli®cation of 11q13 occurs most frequently in
carcinomas of the head and neck (29%), ovary
(20%), bladder (15%), breast (13%) and lung (13%)
(Schuuring, 1995). Individuals with carcinomas containing chromosome 11q13 ampli®cations have a poor
prognosis (Schuuring et al., 1992; Meredith et al.,
1995). While overexpression of FGF3, FGF4 and
BCL1 is rarely observed in the above listed cancers
(Liscia et al., 1989; Fantl et al., 1990), in the majority
of cases expression levels of cyclin D1 and cortactin
are dramatically increased (Bringuier et al., 1996; Hui
et al., 1998). Cyclin D1 is a cell cycle regulatory
protein that is essential for passage through G1
(Baldin et al., 1993), and overexpression in human
tumors is thought to lead to a loss of cell cycle
regulation (Schuuring, 1995). Ampli®cation of the
cortactin locus can occur with or without cyclin D1
ampli®cation, and elevated cortactin gene levels
correlate with poor patient prognosis (Rodrigo et al.,
2000), suggesting that ampli®cation of the cortactin
gene directly contributes to tumor severity. Cell lines
isolated from patients with 11q13 ampli®cation exhibit
increased cortactin mRNA and protein levels (Patel et
al., 1996; Campbell et al., 1996), providing strong
circumstantial evidence that cortactin overexpression
plays an important role in tumors with 11q13
ampli®cation.
Cortactin and membrane dynamics
SA Weed and JT Parsons
Based on observations of cortactin localization and
function in non-transformed cells, overexpression of
cortactin in human tumors has been proposed to result
in increased cell migration and metastatic potential
(Schuuring, 1995). Support for this comes from
increased cell migration but not growth rates in
NIH3T3 ®broblasts overexpressing cortactin (Patel et
al., 1998). Although not directly addressed, the increase
in cell motility is likely due to increased Arp2/3
activation and lamellipodia extension as described
above. In addition, cortactin redistributes into aberrant
cell-matrix contact sites in v-Src transformed cells and
in head and neck squamous cell carcinoma lines
overexpressing cortactin (Figure 3c,d; Wu et al., 1991;
Schuuring et al., 1993). These contact sites, known as
podosomes or rosettes, consist of invaginations of
ventral plasma membrane ringed by circular actin-rich
structures abundant in tyrosine phosphorylated cytoskeletal adhesion proteins (Tarone et al., 1985).
Podosomes are also found in osteoclasts (Marchisio
et al., 1984) where they contribute to bone resorption
(Tanaka et al., 1996) and in macrophages (Davies and
Stossel, 1977) where they are involved in cell motility
(Linder et al., 1999). In v-Src transformed cells,
podosomes are sites of increased extracellular matrix
degradation (Chen et al., 1985) caused by elevated
membrane metalloproteinase activity (Chen and Wang,
1999). The resulting decrease in cell-matrix adhesion
likely contributes to the rounded transformed phenotype observed in v-Src transformed cells (Kellie et al.,
1991) and therefore may lead to increased metastatic
potential. The actin cytoskeleton surrounding podosomes is highly labile (Ochoa et al., 2000), sensitive to
changes in Rac activity (Ory et al., 2000), and is
regulated in macrophages by WASp-mediated Arp2/3
activation (Linder et al., 1999, 2000). The presence of
cortactin in podosomes of Src transformed cells
coupled with the appearance of podosomes in tumor
cells that overexpress cortactin strongly suggests that
cortactin plays a role in regulating podosome actin and
membrane dynamics. This is supported by the
coincident localization of dynamin 2 with cortactin in
the podosomes of multiple cell types (Ochoa et al.,
2000).
A role for cortactin in tumor cell invasion is
further borne out by the localization of cortactin
within invadopodia of invasive breast cancer cells
(Bowden et al., 1999). Invadopodia are unique to
transformed cells, occurring in the center of the cell
and extending well into the neighboring substratum
(Chen and Wang, 1999). Invadopodia exhibit robust
membrane bound matrix metaloproteinase activity
(Kelly et al., 1994) which facilitates the invasion of
human breast cancer cells (Kelly et al., 1998). Tumor
cell motility and invasion also requires the activity of
Cdc42 and Rac (Keely et al., 1997; Banyard et al.,
2000), which in part contributes to endocytosis and
proteolysis of extracellular matrix components (Kheradmand et al., 1998; Banyard et al., 2000). Racmediated recruitment of cortactin into podosomes
and invadopodia would therefore serve to enhance
the ability of cortactin to regulate the Arp2/3mediated assembly of the cortical actin meshwork, a
process that would be further attenuated in tumor
cells with elevated cortactin levels. The observation
that overexpression of the cortactin NTA domain in
invasive MDA-MB-231 breast cancer cells disrupts
large cytoplasmic actin aggregates (which may
provide the structural precurser elements for invadopodia formation) (Uruno et al., 2001) also supports a
role for cortactin in tumor cell adhesion and
invasion.
6429
Perspectives
Regulation of the cortical actin cytoskeleton is
fundamental to a wide range of dynamic cellular
events. Signaling pathways utilized in the control of
plasma membrane dynamics and organization require
reorganization of the cortical actin cytoskeleton.
Dynamic changes in the actin skeleton require the
coordinated activity of tyrosine kinase and Rho
GTPase signaling pathways. Molecules at the `crossroads' of these two pathways, such as cortactin, play
an important role in integrating and mediating
cellular responses. As discussed above, cortactin plays
important roles in tyrosine kinase signaling pathways
that involve alterations in cell morphology. Regulation of the cortical cytoskeleton is central to changes
in cell morphology, and the direct role of cortactin in
Arp2/3-mediated actin dynamics indicates that an
important function of cortactin is to regulate actinbased motility. The interaction of cortactin with PDZ
sca€olding proteins and with dynamin indicates that
cortactin serves to link transmembrane receptor
complexes and membrane vesicle dynamics to areas
of active actin polymerization. This linkage is likely
to be critical for the organization of receptor
complexes and for vesicle motility. Increased cortactin
levels in human cancers augments tumor cell
adhesion, motility and invasion in a manner that
likely involves changes in Arp2/3 dynamics. In spite
of recent advances in understanding cortactin function, much remains to be discovered. Although
tyrosine phosphorylation of cortactin is associated
with cell motility and morphological changes, the
molecular consequences of tyrosine and/or serine/
threonine phosphorylation remain unclear. Does
tyrosine phosphorylation of cortactin by kinases such
as Src, Fer or Syk create binding sites for SH2containing proteins and if so, what is their contribution to cortactin function? Do other tyrosine kinases
phosphorylate cortactin? How does MEK-meditated
serine/threonine phosphorylation impact cortactin
function, and are other kinases involved? While the
identi®cation of cortactin SH3 domain-binding proteins such as CortBP1, ZO-1 and dynamin point to a
indirect role for cortactin in membrane organization,
how cortactin binding ultimately in¯uences synaptic
organization, tight junction assembly and vesicle
biogenesis is unknown. The existence of Shank
Oncogene
Cortactin and membrane dynamics
SA Weed and JT Parsons
6430
proteins in non-neuronal tissues also raises the
likelihood that additional transmembrane protein
complexes are regulated by cortactin. What are the
events leading to chromosome 11q13 ampli®cation
and cortactin overexpression? Finally, how does
cortactin in¯uence podosome and invadopodia activity? These questions and others will likely be the
subject of much investigation in the years to come.
Acknowledgments
We wish to thank Drs Helene Marquis and Steve Anderson
(University of Colorado Health Sciences Center) for
providing Listeria infected and v-Src transformed cells,
Dr Alan Fanning (Yale University) for helpful discussions
and Julie Head for technical support. JT Parsons acknowledges the support of NIH grants CA40042 and CA29243.
SA Weed is a member of the University of Colorado
Cancer Center.
References
Amann KJ and Pollard TD. (2001). Nat. Cell. Biol., 3, 306 ±
310.
Aspenstrom P. (1999). Curr. Opin. Cell Biol., 11, 95 ± 102.
Balda MS and Matter K. (2000). EMBO J., 19, 2024 ± 2033.
Baldin V, Likas MJ, Marcote M, Pagano J, Bartek J and
Draetta G. (1993). Genes Dev., 7, 812 ± 821.
Banyard J, Anand-Apte B, Symons M and Zetter B. (2000).
Oncogene, 19, 580 ± 591.
Bear JE, Rawls JF and Saxe CL. (1998). J. Cell Biol., 142,
1325 ± 1335.
Beckerle MC. (1998). Cell, 95, 741 ± 748.
Bhattacharya S, Fu C, Bhattacharya J and Greenberg S.
(1995). J. Biol. Chem., 270, 16781 ± 16787.
Bishop AL and Hall A. (2000). Biochem. J., 348, 241 ± 255.
Blanchoin L, Amann KJ, Higgs HN, Marchand J-B, Kaiser
DA and Pollard TD. (2000a). Nature, 404, 1007 ± 1011.
Blanchoin L, Pollard TD and Mullins RD. (2000b). Curr.
Biol., 10, 1273 ± 1282.
Boeckers TM, Kreutz MR, Winter C, Zuschratter W, Smalla
K-H, Sanmarti-Vila L, Wex H, Langnaese K, Bockmann
J, Garner CC and Gundel®nger ED. (1999a). J. Neurosci.,
19, 6506 ± 6518.
Boeckers TM, Winter C, Smalla K-H, Kreutz MR,
Bockmann J, Seidenbecher C, Garner CC and Gundel®nger ED. (1999b). Biochem. Biophys. Res. Comm., 264,
247 ± 252.
Borisy GG and Svitkina TM. (2000). Curr. Opin. Cell Biol.,
12, 104 ± 112.
Bourguignon LYW, Zhu H, Shao L and Chen Y-W. (2001).
J. Biol. Chem., 276, 7327 ± 7336.
Bowden ET, Barth M, Thomas D, Glazer RI and Mueller
SC. (1999). Oncogene, 18, 4440 ± 4449.
Bringuier PP, Tamimi Y, Schuuring E and Schalken J.
(1996). Oncogene, 12, 1747 ± 1753.
Brown MT and Cooper JA. (1996). Biochem. Biophys. Acta,
1287, 121 ± 149.
Brunati AM, Donella-Deana A, James P, Quadroni M,
Contri A, Marin O and Pinna LA. (1999). J. Biol. Chem.,
274, 7557 ± 7564.
Calautti E, Missero C, Stein PL, Ezzell RM and Paolo Dotto
G. (1995). Genes Dev., 9, 2279 ± 2291.
Cameron LA, Svitkina TM, Cignjevic D, Theriot JA and
Borisy GG. (2001). Curr. Biol., 11, 130 ± 135.
Campbell DH, de Fazio A, Sutherland RL and Daly RJ.
(1996). Int. J. Cancer, 68, 485 ± 492.
Campbell DH, Sutherland RL and Daly RJ. (1999). Cancer
Res., 59, 5376 ± 5385.
Cantarelli V, Takahashi A, Akeda Y, Nagayama K and
Honda T. (2000). Infect. Immun., 68, 382 ± 386.
Chan AC, Iwashima M, Turck CW and Weiss A. (1992).
Cell, 71, 649 ± 662.
Oncogene
Chen WT and Wang JY. (1999). Ann. NY Acad. Sci., 878,
361 ± 371.
Chen W-T, Chen J-M, Parsons SJ and Parsons JT. (1985).
Nature, 316, 156 ± 158.
Chimini G and Chavrier P. (2000). Nat. Cell. Biol., 2, E191 ±
E196.
Clark EA, King WG, Brugge JS, Symons M and Hynes RO.
(1998). J. Cell Biol., 142, 573 ± 586.
Cook TA, Urrutia R and McNiven MA. (1994). Proc. Natl.
Acad. Sci. USA, 91, 644 ± 648.
Corbalan-Garcia S, Yang SS, Degenhardt KR and Bar-Sagi
D. (1996). Mol. Cell. Biol., 16, 5674 ± 5682.
Craig AWB, Zirngibl R, Williams K, Cole L-A and Greer
PA. (2001). Mol. Cell. Biol., 21, 603 ± 613.
Dai Z, Luo X, Xie H and Peng HB. (2000). J. Cell. Biol., 150,
1321 ± 1334.
David C, McPherson PS, Mundigl O and De Camilli P.
(1996). Proc. Natl. Acad. Sci. USA, 93, 331 ± 335.
Davies WA and Stossel TP. (1977). J. Cell. Biol., 75, 941 ±
955.
Dehio C, Prevost M-C and Sansonetti PJ. (1995). EMBO J.,
14, 2471 ± 2482.
Derry JM, Wiedemann P, Blair P, Wang Y, Kerns JA,
Lemahieu V, Godfrey VL, Wilkinson JE and Francke U.
(1995). Genomics, 29, 471 ± 477.
Downing JR and Reynolds AB. (1991). Oncogene, 6, 607 ±
613.
Drubin DG, Mulholland J, Zhu Z and Botstein D. (1990).
Nature, 343, 288 ± 290.
Du Y, Weed SA, Xiong W-C, Marshall TD and Parsons JT.
(1998). Mol. Cell. Biol., 18, 5838 ± 5851.
Dujon B, Alexandraki D, Andre B, Ansorge W, Baladron V,
Ballesta JPG, Banrevl A, Bolle PA, Bolotin-Fukuhara M,
Bossier P, Bou G, Boyer J, Bultrago MJ, CheÂret G,
Colleaux L, Daignan-Fornier B, del Rey F, Dion C,
Domdey H, DuÈsterhoÈft A, DuÈsterhus S, Entian K-D, Er¯e
H, Esteban PF, Feldmann H, Fernandes L, Fobo GM,
Fritz C, Fukuhara H, Gabel C, Gallion L, CarciaCantelejo M, Garcia-Ramirez JJ, Gent ME, Ghazvini
M, Go€eau A, GonzaleÂz A, Grothues D, Guerreiro P,
Hegemann J, Hewitt N, Hilger F, Hollenberg CP, Horaitis
O, Indge KJ, Jacquier A, James CM, Jauniaux JC,
Jimenez A, Keuchel H, Kirchrath L, Kleine K, KoÈtter P,
Legrain P, Liebi S, Louis EJ, Maia e Silva A, Marck C,
Monnier A-L, MoÈsti D, MuÈller S, Obermaler B, Olover
SG, Pallier C, Pascolo S, Pfei€er F, Phillippsen P, Planta
RJ, Pohl FM, Pohl TM, PoÈhlmann R, Portetelle D,
Purnelle B, Puzos V, Ramezani Rad M, Rasmussen SW,
Remacha M, Revuelta JL, Richard G-F, Rieger M,
Rodrigues-Pousada C, Rose M, Rupp T, Santos MA,
Schwager C, Sensen C, Skala J, Soares H, Sor F,
Cortactin and membrane dynamics
SA Weed and JT Parsons
Stegemann J, Tettelin H, Thierry A, Tzermia M,
Urrestarazu LA, van Dyck L, van Vilet-Reedijk JC,
Valens M, Vandenbol M, Vilela C, Vissers S, von
Wettstein D, Voss H, Wiemann S, Xu G, Zimmermann
J, Haasemann M, Becker I and Mewes HW. (1994).
Nature, 369, 371 ± 378.
Durieu-Trautmann O, Chaverot N, Cazaubon S, Strosberg
AD and Couraud P-O. (1994). J. Biol. Chem., 269, 12536 ±
12540.
Edwards DC, Sanders LC, Bokoch GM and Gill GN. (1999).
Nat. Cell. Biol., 1, 253 ± 259.
Fanning AS, Jameson BJ, Jesaitis LA and Anderson JM.
(1998). J. Biol. Chem., 273, 29745 ± 29753.
Fantl V, Richards MA, Smith R, Lammie GA, Johnstone G,
Allen D, Gregory W, Peters G, Dickson C and Barnes
DM. (1990). Eur. J. Cancer, 26, 423 ± 429.
Fath KR, Mamajiwalla SN and Burgess DR. (1993). J. Cell.
Sci., 17, 65 ± 73.
Fawaz F, van Ooij C, Homola E, Mutka SC and Engel JN.
(1997). Infect. Immun., 65, 5301 ± 5308.
Fincham VJ, Unlu M, Brunton VG, Pitts JD, Wyke JA and
Frame MC. (1996). J. Cell. Biol., 135, 1551 ± 1564.
Fox JEB, Lipfert L, Clark EA, Reynolds CC, Austin CD and
Brugge JS. (1993). J. Biol. Chem., 268, 25973 ± 25977.
Frischknecht F and Way M. (2001). Trends Cell. Biol., 11,
30 ± 38.
Frischknecht F, Cudmore S, Moreau V, Reckmann I,
Rottger S and Way M. (1999b). Curr. Biol., 9, 89 ± 92.
Frischknecht F, Moreau V, Rottger S, Gon¯oni S, Reckmann I, Superti-Furga G and Way M. (1999a). Nature,
401, 926 ± 929.
Fukuoka M, Miki H and Takenawa T. (1997). Gene, 196,
43 ± 48.
Fukuoka M, Suetsugu S, Miki H, Fukami K, Endo T and
Takenawa T. (2001). J. Cell. Biol., 152, 471 ± 482.
Furuse M, Fujita T, Hiiragi T, Fujimoto K and Tsukita Sh.
(1998a). J. Cell. Biol., 141, 1539 ± 1550.
Furuse M, Hirase T, Itoh M, Nagafuchi A, Yonemura S,
Tsukita Sa and Tsukita Sh. (1993). J. Cell. Biol., 123,
1777 ± 1778.
Furuse M, Itoh M, Hirase T, Nagafuchi A, Yonemura S,
Tsukita Sa and Tsukita Sh. (1994). J. Cell. Biol., 127,
1617 ± 1626.
Furuse M, Sasaki H, Fujimoto K and Tsukita Sh. (1998b). J.
Cell. Biol., 143, 391 ± 401.
Gallet C, Rosa J-P, Habib A, Lebret M, Levy-Toledano S
and Maclouf J. (1999). J. Biol. Chem., 274, 23610 ± 23616.
Gao J, Zoller KE, Ginsberg MH, Brugge JS and Shattil SJ.
(1997). EMBO J., 16, 6414 ± 6425.
Goldberg DJ, Foley MS, Tang D and Grabham PW. (2000).
J. Neurosci. Res., 60, 458 ± 467.
Goode BL, Rodal AA, Barnes G and Drubin DG. (2001). J.
Cell. Biol., 153, 627 ± 634.
Guappone AC, Weimer T and Flynn DC. (1998). Mol.
Carcin., 22, 110 ± 119.
Hall A. (1998). Science, 279, 509 ± 574.
Hao QL, Heisterkamp N and Gro€en J. (1989). Mol. Cell.
Biol., 9, 1587 ± 1593.
Hartwig JH, Bokoch GM, Carpenter CL, Janmey PA,
Taylor LA, Toker A and Stossel TP. (1995). Cell, 82,
643 ± 653.
Haskins J, Gu L, Wittchen ES, Hibbard J and Stevenson BR.
(1998). J. Cell. Biol., 141, 199 ± 208.
He H, Watanabe T, Zhan X, Huang C, Schuuring E, Fukami
K, Takenawa T, Kumar CC, Simpson RJ and Murata H.
(1998). Mol. Cell. Biol., 18, 3829 ± 3837.
Higgs HN and Pollard TD. (1999). J. Biol. Chem., 274,
32531 ± 32534.
Higgs HN and Pollard TD. (2000). J. Cell. Biol., 150, 1311 ±
1320.
Huang C, Liu J, Haudenschild CC and Zhan Xi. (1998). J.
Biol. Chem., 273, 25770 ± 25776.
Huang C, Ni Y, Wang T, Gao Y, Haudenschild CC and
Zhan X. (1997). J. Biol. Chem., 272, 13911 ± 13915.
Hui R, Ball JR, Macmillian RD, Kenny FS, Prall OWJ,
Campbell DH, Cornish AL, McClelland RA, Daly RJ,
Forbes JF, Blamey RW, Musgrove EA, Robertson JFR,
Nicholson RI and Sutherland RL. (1998). Oncogene, 17,
1053 ± 1059.
Imamoto A and Soriano P. (1993). Cell, 73, 1117 ± 1124.
Itoh M, Furuse M, Morita K, Kubota K, Saitou M and
Tsukita Sh. (1999a). J. Cell. Biol., 147, 1351 ± 1363.
Itoh M, Morita K and Tsukita S. (1999b). J. Biol. Chem.,
274, 5981 ± 5986.
Itoh M, Nagafuchi A, Yonemura S, Kitani-Yasuda T,
Tsukita Sa and Tsukita Sh. (1993). J. Cell. Biol., 121,
491 ± 502.
Jesaitis LA and Goodenough DA. (1994). J. Cell. Biol., 124,
949 ± 961.
Johnston CL, Cox HC, Gomm JJ and Coombes RC. (1995).
Biochem. J., 306, 609 ± 616.
Jove R and Hanafusa H. (1987). Annu. Rev. Cell. Biol., 3,
31 ± 56.
Kaksonen M, Peng HB and Ravuala H. (2000). J. Cell. Sci.,
113, 4421 ± 4426.
Kanner SB, Reynolds AB, Vines RR and Parsons JT. (1990).
Proc. Natl. Acad. Sci. USA, 87, 3328 ± 3332.
Kapus A, Di Ciano C, Sun J, Zhan X, Kim L, Wong TW and
Rostein OD. (2000). J. Biol. Chem., 275, 32289 ± 32298.
Kapus A, Szaszi K, Sun J, Rizoli S and Rotstein OD. (1999).
J. Biol. Chem., 274, 8093 ± 8102.
Katsube T, Takahisa M, Ueda R, Hashimoto N, Kobayashi
M and Togashi S. (1998). J. Biol. Chem., 273, 29672 ±
29677.
Kay BK, Williamson MP and Sudol M. (2000). FASEB J.,
14, 231 ± 241.
Keely PJ, Westwick JK, Whitehead IP, Der CJ and Parise
LV. (1997). Nature, 390, 632 ± 636.
Kellie S, Horvath AR and Elmore MA. (1991). J. Cell. Sci.,
99, 207 ± 211.
Kelly T, Mueller SC, Yeh Y and Chen W-T. (1994). J. Cell.
Physiol., 158, 299 ± 308.
Kelly T, Yan Y, Osborne RL, Athota AB, Rozypal TL,
Colclasure JC and Chu WS. (1998). Clin. Exp. Metastasis,
16, 501 ± 512.
Kessels MM, Engqvust-Goldstein AEY and Drubin DG.
(2000). Mol. Biol. Cell, 11, 393 ± 421.
Kessels MM, Engqvist-Goldstein AEY, Drubin DG and
Qualmann B. (2001). J. Cell. Biol., 153, 351 ± 366.
Kheradmand F, Werner E, Tremble P, Symons M and Werb
Z. (1998). Science, 280, 898 ± 902.
Kim E, Naisbitt S, Hsueh Y-P, Rao A, Rothschild A, Craig
AM and Sheng M. (1997). J. Cell. Biol., 136, 669 ± 678.
Kim E, Niethammer M, Rothschild A, Jan YN and Sheng M.
(1995). Nature, 378, 85 ± 88.
Kim L and Wong TW. (1998). J. Biol. Chem., 273, 23542 ±
23548.
Kinnunen T, Kaksonen M, Saarinen J, Kalkkinen N, Peng
HB and Rauvala H. (1998). J. Biol. Chem., 273, 10702 ±
10708.
Kitamura D, Kaneko H, Miyagoe Y, Ariyase T and
Watanabe T. (1989). Nucl. Acids. Res., 17, 9367 ± 9379.
6431
Oncogene
Cortactin and membrane dynamics
SA Weed and JT Parsons
6432
Oncogene
Kitamura D, Kaneko H, Taniuchi I, Akagi K, Yamamura K
and Watanabe T. (1995). Biochem. Biophys. Res. Comm.,
208, 1137 ± 1146.
Kolch W. (2000). Biochem. J., 351, 289 ± 305.
Kornau HC, Schenker LT, Kennedy MB and Seeburg PH.
(1995). Science, 269, 1737 ± 1740.
Kozma R, Ahmed S, Best A and Lim L. (1995). Mol. Cell.
Biol., 15, 1942 ± 1952.
Kreienkamp H-J, Zitzer H, Gundel®nger ED, Richter D and
Bockers TM. (2000). J. Biol. Chem., 275, 32387 ± 32390.
Lammie G and Peters G. (1991). Cancer Cells, 3, 413 ± 417.
Lappalainen P, Kessels MM, Cope MJTV and Drubin DG.
(1998). Mol. Biol. Cell, 9, 1951 ± 1959.
LaVallee TM, Prudovsky IA, McMahon GA, Hu X and
Maciag T. (1998). J. Cell. Biol., 141, 1647 ± 1658.
Li Y, Liu J and Zhan X. (2000). J. Biol. Chem., 275, 37187 ±
37193.
Lim S, Sala C, Yoon J, Park S, Kuroda S, Sheng M and Kim
E. (2001). Mol. Cell. Neurosci., 17, 385 ± 397.
Lim S, Naisbitt S, Yoon J, Hwang J-I, Suh P-G, Sheng M
and Kim E. (1999). J. Biol. Chem., 274, 29510 ± 29518.
Linder S, Higgs H, Hufner K, Schwarz K, Pannicke U and
Aepfelbacher M. (2000). J. Immunol., 165, 221 ± 225.
Linder S, Nelson D, Weiss M and Aepfelbacher M. (1999).
Proc. Natl. Acad. Sci. USA, 96, 9648 ± 9653.
Liscia DS, Merlo GR, Garrett C, French D, MarianiCostantini R and Callahan R. (1989). Oncogene, 4, 1219 ±
1224.
Liu BP and Burridge K. (2000). Mol. Biol. Cell, 20, 7160 ±
7169.
Liu J, Huang C and Zhan X. (1999). Oncogene, 18, 6700 ±
6706.
Lock P, Abram CL, Gibson T and Courtneidge SA. (1998).
EMBO J., 17, 4346 ± 4357.
Ma A, Richardson A, Schaefer EM and Parsons JT. (2001).
Mol. Biol. Cell, 12, 1 ± 12.
Maa M-C, Wilson LK, Moyers JS, Vines RR, Parsons JT
and Parsons SJ. (1992). Oncogene, 7, 2429 ± 2438.
Machesky LM and Insall RH. (1998). Curr. Biol., 8, 1347 ±
1356.
Machesky LM, Mullins RD, Higgs HN, Kaiser DA,
Blanchoin L, May RC, Hal ME and Pollard TD. (1999).
Proc. Natl. Acad. Sci. USA, 96, 3739 ± 3744.
Maciver SK. (1998). Curr. Opin. Cell Biol., 10, 140 ± 144.
Mandato CA, Benink HA and Bement WM. (2000). Cell
Motil. Cytoskeleton, 45, 87 ± 92.
Mangoura D. (1997). J. Neurosci. Res., 50, 391 ± 401.
Marchisio PC, Cirillo D, Naldini L, Primavera MV, Teti A
and Zambonin-Zallone A. (1984). J. Cell Biol., 99, 1696 ±
1705.
Martinez-Quiles N, Rohatgi R, Anton IM, Medina M,
Saville SP, Miki M, Yamaguchi H, Takenawa T, Hartwig
JH, Geha RS and Ramesh N. (2001). Nat. Cell Biol., 3,
484 ± 491.
Maruyama S, Kurosaki T, Sada K, Yamanashi Y,
Yamamoto T and Yamamura H. (1996). J. Biol. Chem.,
271, 6631 ± 6635.
McNiven MA, Kim L, Krueger EW, Orth JD, Cao H and
Wong TW. (2000). J. Cell Biol., 151, 187 ± 198.
Meredith SD, Levine PA, Burns JA, Ga€ey MJ, Boyd JC,
Weiss LM, Erickson NL and Williams ME. (1995). Arch.
Otolaryngol. Head Neck Surg., 121, 790 ± 794.
Miglarese MR, Hannion-Henderson J, Wu H, Parsons JT
and Bender TP. (1994). Oncogene, 9, 1989 ± 1997.
Miki H, Yamaguchi H, Suetsugu S and Takenawa T. (2000).
Nature, 408, 732 ± 735.
Miki H, Suetsugtu K and Takenawa T. (1998b). EMBO J.,
17, 6932 ± 6941.
Miki H, Sasaki T, Taki Y and Takenawa T. (1998a). Nature,
391, 93 ± 96.
Mitchison TJ and Cramer LP. (1996). Cell, 84, 371 ± 397.
Mounier J, Laurent V, Hall A, Fort P, Carlier MF,
Sansonetti PJ and Egile C. (1999). J. Cell Sci., 112,
2069 ± 2080.
Muller B, Cooper L and Terhorst C. (1994). Immunogenetics,
39, 359 ± 362.
Mullins RD, Heuser JA and Pollard TD. (1998). Proc. Natl.
Acad. Sci. USA, 95, 6181 ± 6186.
Nada S, Okada M, Aizawa S and Nakagawa H. (1994).
Oncogene, 9, 3571 ± 3578.
Nada S, Okada M, MacAuley A, Cooper JA and Nakagawa
H. (1991). Nature, 351, 69 ± 72.
Nada S, Yagi T, Takeda H, Tokunaga T, Nakagawa H,
Ikawa Y, Okada M and Aizawa S. (1993). Cell, 73, 1125 ±
1135.
Naisbitt S, Kim E, Tu JC, Xiao B, Sala C, Valtschano€ J,
Weinberg RJ, Worley PF and Sheng M. (1999). Neuron,
23, 569 ± 582.
Nakamoto T, Ssakai R, Ozawa Y, Yazaki Y and Hirai H.
(1996). J. Biol. Chem., 271, 8959 ± 8965.
Nobes CD and Hall A. (1999). J. Cell Biol., 144, 1235 ± 1244.
Nurden AT and Caen JP. (1974). Br. J. Haematol., 28, 253 ±
260.
Ochoa G-C, Slepnev VI, Ne€ L, Ringstad N, Takei K,
Daniell L, Kim W, Cao H, McNiven M, Baron R and De
Camilli P. (2000). J. Cell Biol., 150, 377 ± 389.
Ohoka Y and Takai Y. (1998). Genes Cells, 3, 603 ± 612.
Okamura H and Resh MD. (1995). J. Biol. Chem., 270,
26613 ± 26618.
Oliferenko S, Kaverina I, Small JV and Huber LA. (2000). J.
Cell Biol., 148, 1159 ± 1164.
Ory S, Munari-Silem Y, Fort P and Jurdic P. (2000). J. Cell
Sci., 113, 1177 ± 1188.
Ozawa K, Kashiwada K, Takahashi M and Sobue K. (1995).
Exp. Cell Res., 221, 197 ± 204.
Parsons JT. (1996). Curr. Opin. Cell Biol., 8, 146 ± 152.
Parsons JT and Parsons SJ. (1997). Curr. Opin. Cell Biol., 9,
187 ± 792.
Patel AM, Incognito LS, Schechter GL, Wasilenko WJ and
Somers KD. (1996). Oncogene, 12, 31 ± 35.
Patel AM, Schechter GL, Wasilenko WJ and Somers KD.
(1998). Oncogene, 16, 3227 ± 3232.
Pollard TD, Blanchoin L and Mullins RD. (2000). Ann. Rev.
Biophys. Biomol. Struct., 29, 545 ± 576.
Qualmann B and Kelly RB. (2000). J. Cell Biol., 148, 1047 ±
1061.
Qualmann B, Kessels MM and Kelly RB. (2000). J. Cell
Biol., 150, F111 ± F116.
Qualmann B, Roos J, DiGregorio PJ and Kelly RB. (1999).
Mol. Biol. Cell, 10, 501 ± 513.
Ramjaun AR, Micheva KD, Bouchelet I and McPherson PS.
(1997). J. Biol. Chem., 272, 16700 ± 16706.
Redecker P, Gundel®nger ED and Boeckers TM. (2001). J.
Histochem. Cytochem., 49, 639 ± 648.
Ridley A. (2000). J. Cell Biol., 150, F107 ± F109.
Ridley AJ, Patterson HF, Johnston CL, Diekmann D and
Hall A. (1992). Cell, 70, 401 ± 410.
Rinnerthaler G, Geiger B and Small JV. (1988). J. Cell Biol.,
106, 747 ± 760.
Rodrigo JP, Garcia LA, Ramos S, Lazo PS and Suarez C.
(2000). Clin. Cancer Res., 6, 3177 ± 3182.
Cortactin and membrane dynamics
SA Weed and JT Parsons
Rohatgi R, Ho HH and Kirschner MW. (2000). J. Cell Biol.,
150, 1299 ± 1309.
Rohatgi R, Ma L, Miki H, Lopez M, Kirchhausen T,
Takenawa T and Krischner MW. (1999). Cell, 97, 221 ±
231.
Rosa J-P, Artcanuthurry V, Grelac F, Maclouf J, Caen JP
and Levy-Toledano S. (1997). Blood, 12, 4385 ± 4392.
Schafer DA, S'Souza-Schorey C and Cooper JA. (2000).
Trac, 1, 892 ± 903.
Schaller MD, Hildebrand JD, Shannon JD, Fox JW, Vines
RR and Parsons JT. (1994). Mol. Cell. Biol., 14, 1680 ±
1688.
Schmid SL. (1997). Ann. Rev. Biochem., 66, 511 ± 548.
Schmid SL, McNiven MA and De Camilli P. (1998). Curr.
Opin. Cell Biol., 10, 504 ± 512.
Schuuring E. (1995). Gene, 159, 83 ± 96.
Schuuring E, Verhoeven E, Litvinov S and Michalides
RJAM. (1993). Mol. Cell. Biol., 13, 2891 ± 2898.
Schuuring E, Verhoeven E, Mooi WJ and Michalides RJAM.
(1991). Oncogene, 7, 355 ± 361.
Schuuring E, Verhoeven E, van Tinteren H, Peterse JL,
Nunnik B, Thunnissen FBJM, Devilee P, Cornelisse CJ,
van de Vijver MJ, Mooi WJ and Michalides RJAM.
(1992). Cancer Res., 52, 5229 ± 5234.
Sechi AS and Wehland J. (2000). J. Cell Sci., 113, 3685 ±
3695.
Sheng M and Kim E. (2000). J. Cell Sci., 113, 1851 ± 1856.
Sheth B, Fesenko I, Collins JE, Moran B, Wild AE,
Anderson JM and Fleming TP. (1997). Development,
124, 2027 ± 2037.
Small JV, Herzog M and Anderson K. (1995). J. Cell Biol.,
129, 1275 ± 1286.
Soloviev M, Ciruela F, Chan WY and McIlhinney RA.
(2000). J. Mol. Biol., 295, 1185 ± 1200.
Songyang Z, Shoelson SE, Chaudhuri M, Gish G, Pawson T,
Haser WG, King F, Roberts T, Ratnofsky S, Lechleider
RJ, Neel BG, Birge RB, Fajardo JE, Chou MM, Hanafusa
H, Scha€hausen B and Cantley LC. (1993). Cell, 72, 767 ±
778.
Sparks AB, Ho€man NG, McConnell SJ, Fowlkes DM and
Kay BK. (1996a). Nat. Biotechnol., 14, 741 ± 744.
Sparks AB, Rider JE, Ho€man NH, Fowlkes DM, Quilliam
LA and Kay BK. (1996b). Proc. Natl. Acad. Sci. USA, 93,
1540 ± 1544.
Stevenson BR, Siliciano JD, Mooseker MS and Goodenough
DA. (1986). J. Cell Biol., 103, 755 ± 766.
Svitkina TM and Borisy GG. (1999). J. Cell. Biol., 145,
1009 ± 1026.
Takahisa M, Togashi S, Suzuki T, Kobayashi M, Murayama
A, Kondo K, Miyake T and Udea R. (1996). Genes Dev.,
10, 1783 ± 1795.
Takemoto Y, Furuta M, Li X-K, Strong-Sparks WJ and
Hashimoto Y. (1995). EMBO J., 14, 3403 ± 3414.
Takenawa T and Miki H. (2001). J. Cell Sci., 114, 1801 ±
1809.
Tanaka S, Amling L, Ne€ A, Peyman E, Uhlmann E, Levy
JB and Baron R. (1996). Nature, 383, 528 ± 531.
Taniuchi I, Kitamura D, Maekawa Y, Fukuda T, Kishi H
and Watanabe T. (1995). EMBO J., 14, 3664 ± 3678.
Tarone G, Cirillo D, Giancotti FG, Comoglio PM and
Marchisio PM. (1985). Exp. Cell Res., 159, 141 ± 157.
Taunton J, Rowning BA, Coughlin ML, Wu M, Moon RT,
Mitchison TJ and Larabell CA. (2000). J. Cell Biol., 148,
519 ± 530.
Taylor JM, Hildebrand JD, Mack CP, Cox ME and Parsons
JT. (1998). J. Biol. Chem., 273, 8063 ± 8070.
Thomas SM, Soriano P and Imamoto A. (1995). Nature, 376,
267 ± 271.
Tobaben S, Sudhof TC and Stahl B. (2000). J. Biol. Chem.,
275, 36204 ± 36210.
Tsukita S, Furuse M and Itoh M. (1999). Curr. Opin. Cell
Biol., 11, 628 ± 633.
Tu JC, Xiao B, Naisbitt S, Yuan JP, Petralia RS, Brakeman
P, Doan A, Aakalu VK, Lanahan AA, Shang M and
Worley PF. (1999). Neuron, 23, 583 ± 592.
Tu JC, Xiao B, Yuan J, Lanahan A, Leo€ert K, Li M, Linden
D and Worley PF. (1998). Neuron, 21, 717 ± 726.
Uruno T, Liu J, Zhang P, Fan Y-X, Egile C, Li R, Mueller
SC and Zhan X. (2001). Nat. Cell Biol., 3, 259 ± 266.
van Damme H, Brok H, Schuuring-Scholtes E and
Schuuring E. (1997). J. Biol. Chem., 272, 7374 ± 7380.
Violette SM, Guan W, Bartlett C, Smith JA, Bardelay C,
Antoine E, Rickles RJ, Mandine E, Van Schravendijk
MR, Adams SE, Lynch BA, Shakespeare WC, Yang M,
Jacobsen VA, Takeuchi CS, Macek KJ, Bohacek RS,
Dalgarno DC, Weigele M, Lesuisse D, Sawyer TK and
Baron R. (2001). Bone, 28, 54 ± 64.
Vuori K and Ruoslahti E. (1995). J. Biol. Chem., 270,
22259 ± 22262.
Wang D-Z, Reiter RS, Lin JL-C, Wang Q, Williams HS,
Krob SL, Schultheiss TM, Evans S and Lin JJ-C. (1999).
Development, 126, 1281 ± 1294.
Weaver AM, Karginov AV, Kinley AW, Weed SA, Li Y,
Parsons JT and Cooper JA. (2001). Curr. Biol., 11, 370 ±
374.
Weed SA, Du Y and Parsons JT. (1998). J. Cell Sci., 111,
2433 ± 2443.
Weed SA, Karginov AV, Schafer DA, Weaver AM, Kinley
AW, Cooper JA and Parsons JT. (2000). J. Cell Biol., 151,
29 ± 40.
Wheal HV, Chen Y, Mitchell J, Schachner M, Maerz W,
Wieland H, Van Rossum D and Kirsch J. (1998). Prog.
Neurobiol., 55, 611 ± 640.
Wigge P and McMahon HT. (1998). Trends Neurosci., 21,
339 ± 344.
Willott E, Balda MS, Fanning AS, Jameson B, Van Itallie C
and Anderson JM. (1993). Proc. Natl. Acad. Sci. USA, 90,
7834 ± 7838.
Wittchen ES, Haskins J and Stevenson BR. (1999). J. Biol.
Chem., 274, 35179 ± 35185.
Wu H and Montone KT. (1998). J. Histochem. Cytochem.,
46, 1189 ± 1191.
Wu H and Parsons JT. (1993). J. Cell Biol., 120, 1417 ± 1426.
Wu H, Reynolds AB, Kanner SB, Vines RR and Parsons JT.
(1991). Mol. Cell. Biol., 11, 5113 ± 5124.
Xaio B, Tu JC, Petralia RS, Yuan J, Doan A, Breder C,
Ruggiero A, Lanahan AA, Wenthold RJ and Worley PF.
(1998). Neuron, 21, 707 ± 716.
Xiao B, Tu JC, Petralia RS, Yuan J, Doan A, Breder C,
Ruggiero A, Lanahan AA, Wenthold RJ and Worley PF.
(1998). Neuron, 21, 707 ± 716.
Yamamoto T, Harada N, Kano K, Taya S, Canaani E,
Matsuura Y, Mioguchi A, Ide C and Kaibuchi K. (1997).
J. Cell Biol., 139, 785 ± 795.
Yang N, Higuchi O, Ohashi K, Nagata K, Wada A,
Kangawa K, Nishida E and Mizuno K. (1998). Nature,
393, 809 ± 812.
Zettl M and Way M. (2001). Nat. Cell Biol., 3, E74 ± E75.
6433
Oncogene
Cortactin and membrane dynamics
SA Weed and JT Parsons
6434
Oncogene
Zhan X, Hu X, Hamton B, Burgess WH, Friesel R and
Maciag T. (1993). J. Biol. Chem., 268, 24427 ± 24431.
Zhan X, Plourde C, Hu X, Friesel R and Maciag T. (1994). J.
Biol. Chem., 269, 20221 ± 20224.
Zigmond SH. (1996). Curr. Opin. Cell Biol., 8, 66 ± 73.
Zitzer H, Richter D and Kreienkamp H-J. (1999). J. Biol.
Chem., 274, 18153 ± 18156.