Interaction of cortactin and N-WASp with Arp2/3 complex.

Abstract

Dynamic actin assembly is required for diverse cellular processes and often involves activation of Arp2/3 complex. Cortactin and N-WASp activate Arp2/3 complex, alone or in concert. Both cortactin and N-WASp contain an acidic (A) domain that is required for Arp2/3 complex binding.
We investigated how cortactin and the constitutively active VCA domain of N-WASp interact with Arp2/3 complex. Structural studies showed that cortactin is a thin, elongated monomer. Chemical crosslinking studies demonstrated selective interaction of the Arp2/3 binding NTA domain of cortactin (cortactin NTA) with the Arp3 subunit and VCA with Arp3, Arp2, and ARPC1/p40. Cortactin NTA and VCA crosslinking to the Arp3 subunit were mutually exclusive; however, cortactin NTA did not inhibit VCA crosslinking to Arp2 or ARPC1/p40, nor did it inhibit activation of Arp2/3 complex by VCA. We conducted an experiment in which a saturating concentration of cortactin NTA modestly lowered the binding affinity of VCA for Arp2/3; the results of this experiment provided further evidence for ternary complex formation. Consistent with a common binding site on Arp3, a saturating concentration of VCA abolished binding of cortactin to Arp2/3 complex.
Under certain circumstances, cortactin and N-WASp can bind simultaneously to Arp2/3 complex, accounting for their synergy in activation of actin assembly. The interaction of cortactin NTA with Arp2/3 complex does not inhibit Arp2/3 activation by N-WASp, despite competition for a common binding site located on the Arp3 subunit. These results suggest a model in which cortactin may bridge Arp2/3 complex to actin filaments via Arp3 and N-WASp activates Arp2/3 complex by binding Arp2 and/or ARPC1/p40.

Full text

Page 1

Current Biology, Vol. 12, 1270–1278, August 6, 2002, 2002 Elsevier Science Ltd. All rights reserved.PII S0960-9822(02)01035-7
Interaction of Cortactin and N-WASp
with Arp2/3 Complex
(ARPC1–5)[2].Alone,Arp2/3complexisinactive;activa-
tion requires binding of another cellular protein. Upon
activation, Arp2/3 complex binds to the side of a preex-
isting actin filament and nucleates a new daughter fila-
ment, forming a branch at a 70? angle from the mother
filament [3].
The best-characterized activators of Arp2/3 complex
are members of the Wiskott-Aldrich Syndrome protein
(WASp) family [2]. This family includes WASp, which is
expressed in hematopoietic cells, and SCAR/WAVE1–3
and N-WASp, which are expressed ubiquitously. WASp
family members bind Arp2/3 complex via a C-terminal
acidic (A) domain containing a conserved tryptophan
residue.TheCterminusalsocontainsaverprolinhomol-
ogy and connecting (VC) domain, which binds mono-
meric actin and is important for nucleation activity [2].
Other recently identified Arp2/3 activators include cor-
tactin in mammals, Listeria monocytogenes ActA, CAR-
MIL in Dictyostelium, Abp1p and Pan1p in budding
yeast, and type I myosin in fission yeast, all of which
bind Arp2/3 complex via an acidic domain similar to that
found in WASp [2].
The multiplicity of Arp2/3activators raises two related
questions. First, how do various Arp2/3 activators bind
tothe complexto causeits activation,and, second,how
are the signals from different Arp2/3 activators inte-
grated?Arp2/3 complexaloneisinactive, andactivation
apparently requires a large conformational change [4,
5]. SCAR1/WAVE1 and ActA have been chemically
crosslinked to the Arp2, Arp3, and ARPC1/p40 subunits
of Arp2/3 complex, which suggests that these subunits
may transduce the switch to the active conformation
[6]. Activation of Arp2/3 complex is also promoted by
its binding to the side of an actin filament. The addition
of actin filaments increases the rate of Arp2/3 complex
activation by WASp/SCAR proteins [7, 8]. An antibody
to the ARPC1/p34 subunit of Arp2/3 complex inhibits
binding of the complex to the side of actin filaments
and reduces the efficiency of nucleation [9]. Activation
of Arp2/3 complex by cortactin and Abp1p appears to
involve bridging Arp2/3 complex to actin filaments
[10–12].
Cortactin is a major src kinase substrate and F-actin
binding protein [13] that is overexpressed in a number
of epithelial carcinomas, including breast cancer and
head and neck cancer [14]. Overexpression of cortactin
increases the number of breast cancer metastases in
nude mice [15] and increases the motility of cultured
fibroblasts [16]. Cortactin binds to Arp2/3 complex via
its N terminus, to F-actin via its central repeats domain,
and to a number of signaling molecules, including src
kinases, Shank, CortBP, ZO-1, and dynamin, via its C-ter-
minalproline-richandSH3domains[17].Incells,cortac-
tin is targeted to sites of active actin assembly by Rac1
[18] andcolocalizes with Arp2/3 complexin lamellipodia
[18, 19] and in the actin comet tails of vesicles [19, 20].
Cortactin also localizes to many of the same sites as
N-WASp,includingvesicles[19–21]andpodosomes[22,
23]. Finally, N-WASp and cortactin apparently form a
Alissa M. Weaver,1John E. Heuser,1
Andrei V. Karginov,2Wei-lih Lee,1
J. Thomas Parsons,2and John A. Cooper1,3
1Department of Cell Biology and Physiology
Washington University School of Medicine
St. Louis, Missouri 63110
2Department of Microbiology and Cancer Center
University of Virginia Health Sciences Center
Charlottesville, Virginia 22908
Summary
Background: Dynamic actin assembly is required for
diverse cellular processes and often involves activation
of Arp2/3 complex. Cortactin and N-WASp activate
Arp2/3 complex, alone or in concert. Both cortactin and
N-WASp contain an acidic (A) domain that is required
for Arp2/3 complex binding.
Results: We investigated how cortactin and the consti-
tutively active VCA domain of N-WASp interact with
Arp2/3 complex. Structural studies showed that cortac-
tin is a thin, elongated monomer. Chemical crosslinking
studiesdemonstratedselectiveinteractionoftheArp2/3
binding NTA domain of cortactin (cortactin NTA) with
the Arp3 subunit and VCA with Arp3, Arp2, and ARPC1/
p40. Cortactin NTA and VCA crosslinking to the Arp3
subunit were mutually exclusive; however, cortactin
NTA did not inhibit VCA crosslinking to Arp2 or ARPC1/
p40, nor did it inhibit activation of Arp2/3 complex by
VCA.We conductedanexperiment inwhich asaturating
concentration of cortactin NTA modestly lowered the
binding affinity of VCA for Arp2/3; the results of this
experiment provided further evidence for ternary com-
plex formation. Consistent with a common binding site
on Arp3, a saturating concentration of VCA abolished
binding of cortactin to Arp2/3 complex.
Conclusions: Under certain circumstances, cortactin
and N-WASp can bind simultaneously to Arp2/3 com-
plex, accounting for their synergy in activation of actin
assembly. The interaction of cortactin NTA with Arp2/3
complex does not inhibit Arp2/3 activation by N-WASp,
despite competition for a common binding site located
on the Arp3 subunit. These results suggest a model in
which cortactin may bridge Arp2/3 complex to actin
filaments via Arp3 and N-WASp activates Arp2/3 com-
plex by binding Arp2 and/or ARPC1/p40.
Introduction
Cell migration, vesicle movement,and many other cellu-
lar processes require rapid induction of actin polymer-
ization [1]. Nucleation of actin to form new filaments
often occurs by activation of Arp2/3 complex, a seven-
protein complex containing two actin-related proteins
(Arp2 and Arp3) and five unrelated polypeptides
3Correspondence: jcooper@cellbio.wustl.edu

Page 2

Cortactin, N-WASp, and Arp2/3 Complex
1271
Table 1. Physical Parameters of Cortactin
Ellipsoid Models R Stokes
s*
Massf/f0
ProlateOblateVel sedGel Filtration
kDa
63.5
S4,w
2.09
S20,w
3.315 1.8293 A˚a
29 A˚b
144 A˚a
12 A˚b
48 A˚ 59 A˚
aMajor lengths of ellipsoid models calculated from the experimental data sets.
bMinor lengths of ellipsoid models calculated from the experimental data sets.
molecularcomplexinsrc-transformedcellsandareboth
important for the formation of certain actin-rich struc-
tures, such as podosomes [23] and enteropathogenic
E. coli-induced pedestals [24, 25].
Cortactin binds and activates Arp2/3 complex [10, 11,
18]. Cortactin also stabilizes Arp2/3-induced branched
actin networks and inhibits their disassembly [10]. In
previous studies, activation of Arp2/3 complex by cor-
tactin wasadditive or evensynergistic withthe constitu-
tivelyactiveVCAdomainofN-WASp(VCA)[10,11].Both
cortactin and N-WASp have a single acidic (A) domain,
which is involved in Arp2/3 binding. Thus, one might
expect that cortactin and N-WASp would compete for
binding to Arp2/3 complex; however, this simple view
appears to be inconsistent with the apparent synergy
between cortactin and N-WASp VCA with respect to
Arp2/3 activation. To address this paradox, we have
now investigated how cortactin and N-WASp VCA inter-
act with Arp2/3 complex, using physical and functional
assays.TheresultsindicatethatcortactinandVCAcom-
peteforbindingtotheArp3subunitbutmaystillsimulta-
neously bind Arp2/3 complex because VCA also binds
to the Arp2 and ARPC1/p40 subunits. The N-WASp VCA
in this ternary complex is fully functional in its ability to
activate Arp2/3 complex, and this provides a model for
howcortactinandN-WASpinteractwithArp2/3complex
to promote actin assembly.
an asymmetric, elongated molecule [27]. A prolate (ci-
gar-shaped) ellipsoid model of cortactin gives a length
to width ratio of 10:1 (Table 1). Gel filtration chromatog-
raphy of cortactin yielded a single peak, with a Stokes
radius of 59 A˚ (Table 1; also see the Supplementary
Material), slightly larger than the Stokes radius of 48 A˚
determined by sedimentation velocity. Both measure-
ments of the Stokes radius are substantially larger than
would be expected for a globular protein of molecular
mass 63.5 kDa. For example, bovine serum albumin
(BSA), a relatively spherical protein with a similar molec-
ular mass (66kDa) has a Stokes radius of36 A˚. The large
Stokesradiusandhighfrictionalcoefficientindicatethat
the cortactin monomer has an asymmetric shape.
“Deep etch” electron microscopy [28, 29] of cortactin
(Figure 1) showed it to be a slender threadlike molecule,
Results
Cortactin Is an Elongated Flexible Monomer
To understand how cortactin interacts with Arp2/3 com-
plex and actin filaments, we first determined the shape
and size of cortactin. Previous studies suggested that
cortactin may be a multimer [11, 26]. We determined
the native molecular mass of cortactin by equilibrium
sedimentation(Table1;alsoseetheSupplementaryMa-
terialavailable withthisarticleonline). Amolecularmass
of 63.5 kDa was calculated from the experimentally de-
termined effective reduced molecular weight (?) and
from a partial specific volume of 0.7074. The molecular
weight predicted from the amino acid sequence is 61.5
kDa. Thus, under these conditions, recombinant cortac-
tin,whichbindsandactivatesArp2/3complex,isprimar-
ily a monomer. Consistent with this result, we have not
observedactinfilamentbundlingactivitywithourprepa-
rations of cortactin [10].
Sedimentation velocity experiments were performed
to determine the hydrodynamic properties of cortactin
(Table 1; also see the Supplementary Material). The fric-
tional coefficient for cortactin (f/f0)is 1.8, consistent with
Figure 1. Deep Etch Electron Microscopy
These images are ideally viewed with red-green (left-right) three-
dimensional glasses. Rows 1 and 2: individual cortactin molecules
from a very low-angle (6?), relatively “grainy” platinum replica, em-
ployed to highlight such very thin molecules. The first image in the
top left corner shows an image before digital post-processing into
a three-dimensional masked image (compare to the adjacent image
on the right). Row 3: possible associations of Arp2/3 complexes
with the ends of cortactin molecules, from a standard 12? replica.
Row 4: (Left two panels): F-actin “branchpoints” provided for scale,
generated in a VCA ? Arp2/3 ? G-actin-stimulated polymerization
experiment; (Right two panels): Arp2/3 molecules alone, without
cortactin.

Page 3

Current Biology
1272
with a width close to the resolution limit of the platinum
that was used to replicate the molecules after freeze
drying. This platinum forms ?20 A˚beads in such a rep-
lica.Wealsooccasionallysawlargerthreadsthatproba-
bly represented self-associated or aggregated cortac-
tin. When mixed with Arp2/3 complex, cortactin was
seen as a slender thread attached at one end to Arp2/3
complex,aformationthatisconsistentwiththeprevious
identification of the Arp2/3 binding site near the N termi-
nus of cortactin. We were not able to visualize cortactin
bound to actin filaments or at filament branch points,
probably because cortactin is much thinner than the
actin filaments. The average length of optimally repli-
cated cortactin molecules was 220 ? 30 A˚(mean ? SD,
n ? 91), which is slightly shorter than the 290 A˚length
predicted by elliptical modeling of the results of the
velocity sedimentation studies (Table 1, prolate model).
Collectively, these hydrodynamic and imaging studies
indicatethatcortactinisamonomeric,elongatedprotein
with a length of 220–290 A˚.
Chemical Crosslinking Studies: How Cortactin
and N-WASp Interact with the Subunits
of Arp2/3 Complex
To identify the subunits within Arp2/3 complex that may
provide binding interfaces with cortactin and VCA,
chemical crosslinking studies were performed. Purified
Arp2/3 complex was incubated with the zero-length
crosslinker EDC/NHS in the presence of GST-VCA or
theN-terminalArp2/3bindingdomainofcortactin(NTA).
Crosslinked products were identified by Western blot-
ting, by using antibodies specific for individual subunits
of Arp2/3 complex. We used a range of protein concen-
trations to define the relative importance and specificity
of the crosslinked products. At protein concentrations
close to the Kdfor binding to Arp2/3 complex [11], VCA
crosslinked to the Arp3, Arp2, and ARPC1/p40 subunits
(Figure 2). These subunits have been reported to bind
to the Arp2/3 activators Scar and ActA [6]. At higher
concentrations,a faintcrosslinking
ARPC2/p34 was also observed.
Cortactin NTA was crosslinked only to the Arp3 sub-
unit at low cortactin concentrations (Figure 2). Faint
crosslinkingproductswerealsoobservedwiththeArp2,
ARPC1/p40, ARPC2/p34, and ARPC3/p21 subunits;
however, these bands were evident only when the cor-
tactin NTA concentration was at least ten-fold higher
than that required for crosslinking to Arp3 and above
the reported Kdfor cortactin binding to Arp2/3 complex
[11, 18]. No crosslinking of cortactin or VCA to ARPC5/
p16 or ARPC4/p20 was observed at any concentration
(data not shown). Full-length cortactin and GST-NTA
had the same subunit crosslinking pattern as untagged
NTA (data not shown).
Figure 2. Chemical Crosslinking Analysis of Cortactin NTA and
GST-VCA to Arp2/3 Complex
Purified cortactin NTA (0–5 ?M) or GST-VCA (0–2.5 ?M) were incu-
bated with Arp2/3 complex and crosslinked for 30 min. Western
blots were probed by using antibodies against individual Arp2/3
complex subunits as indicated. The uncrosslinked subunits are the
major, lower band in each case. Crosslinked products, if present,
are indicated as “xN” (for NTA) and “xV” (for VCA). These blots were
purposely overexposed in order to see faint crosslinking bands.
n ? 3.
product with
Cortactin NTA inhibited crosslinking of GST-VCA to the
Arp3 subunit but not to Arp2 or ARPC1/p40 (Figure 3A).
In the converse experiment, increasing concentrations
of GST-VCA inhibited crosslinking of cortactin NTA to
Arp3 (Figure 3B). The simplest interpretation of these
results is that cortactin NTA and N-WASp VCA compete
for binding to the Arp3 subunit. An alternative explana-
tion is that NTA and VCA bind simultaneously and alter
the conformation of Arp2/3 complex in such a way as
to abrogate the chemical crosslinking reaction. This al-
ternative is less likely since crosslinking of VCA or NTA
to Arp3 is lost in both converse experiments; however,
we tested the model further with a physical binding
experiment that did not involve chemical crosslinking.
In this experiment, we used a quantitative GST pull-
down assay [30] to determine the apparent Kdfor VCA
binding to Arp2/3 complex, in the presence or absence
of saturating cortactin NTA. Arp2/3 complex (50 nM)
was incubated with increasing concentrations of GST-
VCA coupled to glutathione-agarose (0–8 ?M), in the
Cortactin Competes with N-WASp VCA for Binding
to Arp3, but Not to Arp2 or ARPC1/p40
Since both cortactin NTA and N-WASp VCA crosslinked
totheArp3subunit,weaskedwhethertheywouldcompete
with each other. Arp2/3 complex (0.5 ?M) and GST-VCA
(0.2 ?M) were incubated with increasing concentrations
of cortactin NTA (0–80 ?M) and were then crosslinked.

Page 4

Cortactin, N-WASp, and Arp2/3 Complex
1273
Figure 4. The Arp2/3 Binding Domain of Cortactin Does Not Inhibit
theAbilityofN-WASpVCAtoStimulateArp2/3-MediatedNucleation
(A) Cortactin domain structure. The N-terminal acidic domain (NTA)
(amino acids 1–84) is the Arp2/3 binding domain used for competi-
tion studies in (B). The central repeats domain is required for F-actin
bindingandactivationofArp2/3complex,butdoesnotcontributeto
Arp2/3 binding.
(B) Competition between VCA and cortactin: pyrene actin polymer-
ization assay. A GST fusion of the VCA fragment of N-WASp was
incubated with Arp2/3 complex in the presence or absence of the
Arp2/3 binding domain of cortactin (Cort-NTA) in polymerization
buffer for 10 min at 25?C. Actin polymerization was initiated by
the addition of monomeric actin (7.5% pyrene labeled) and was
monitored by continuous measurement of fluorescence at 386 nm.
Curves had 12.5 nM Arp2/3 complex ? 0.5 nM GST-VCA with the
following additions: (b) no additions, (c) 2 ?M GST-NTA, (d) 11 ?M
GST-NTA, and (e) 22 ?M GST-NTA. Other controls (“Controls”) in-
clude actin alone, Arp2/3 alone, and Arp2/3 ? 22 ?M GST-NTA.
Activation of Arp2/3 complex with higher concentrations of VCA
(“Arp2/3 ? 12.5 nM VCA”) shows that our competition studies were
done under conditions in which Arp2/3 complex was not saturated
by the GST-VCA. n ? 4.
Figure 3. Cortactin and VCA Compete for Binding to Arp3, but Not
to Arp2 or ARPC1/p40
(A) Effects of NTA on GST-VCA crosslinking to Arp2/3 complex
subunits.PurifiedArp2/3complex(0.5?M)wasincubatedwithGST-
VCA (0.2 ?M) in the presence or absence of NTA (5–80 ?M), as
indicated at the top. Antibody probes (top to bottom): anti-Arp3,
-Arp2, -ArpC1/p40. Uncrosslinked Arp2/3 complex subunits are la-
beled “Arp3”, “Arp2”, and “p40”. Crosslinked products are labeled
as “NTAx” or “GST-VCAx” plus the appropriate Arp2/3 subunit.
n ? 3.
(B) Effect of GST-VCA on NTA crosslinking to Arp3. Purified Arp2/3
complex (0.5 ?M) was incubated with NTA (1 ?M) in the presence or
absence of GST-VCA (0–5 ?M), as indicated at the top. Crosslinked
products are labeled “NTAxArp3” or “GST-VCAxArp3”. n ? 3.
(C) Effect of NTA on the Kdfor VCA binding to Arp2/3 complex.
Binding of Arp2/3 complex (50 nM) to GST-VCA beads (0–8 ?M) in
the absence (solid circles) or presence (open squares) of 30 ?M
presence or absence of 30 ?M cortactin NTA. The cor-
tactin NTA concentration was approximately 25-fold the
reported Kdfor binding to Arp2/3 complex [11, 18]. The
GST-VCA beads were rapidly pelleted, and the superna-
cortactin NTA was determined by supernatant depletion [30]. GST-
VCA-Arp2/3 complex (“Bound VCA”) is shown plotted against the
freeGST-VCAconcentration(“FreeVCA”)(averageofthreeindepen-
dent experiments, error bars represent standard error). Apparent Kd
values for VCA binding to Arp2/3 complex were 0.76 ? 0.01 ?M and
1.21?0.09 ?M(mean?SE), intheabsenceor presenceofcortactin
NTA, respectively (p ? 0.05).

Page 5

Current Biology
1274
tants were analyzed for depletion of Arp2/3 complex by
Western blotting. From these measurements, we calcu-
lated the amount of Arp2/3 complex associated with the
GST-VCAbeads.The concentrationofGST-VCA-Arp2/3
complex was plotted against the free GST-VCA concen-
tration,andapparentKdvaluesweredeterminedbyleast
squares analysis (Figure 3C). The apparent Kdfor VCA
binding to Arp2/3 complex was 0.76 ? 0.01 ?M and
1.21 ? 0.09 ?M (mean ? SE, n ? 3), in the absence and
presence of cortactin NTA, respectively. These values
were significantly different (p ? 0.05). The fact that VCA
does bind Arp2/3 complex in the presence of a saturat-
ing concentration of cortactin provides evidence that a
ternary VCA-Arp2/3-cortactin complex can form and is
consistent with a model in which cortactin blocks only
oneofthemultipleVCAbindingsiteswithinArp2/3com-
plex. Competition at this site is apparently sufficient to
decrease the apparent binding constant for the VCA-
Arp2/3 interaction.
When the converse pull-down experiment was per-
formedwithGST-NTAbeads,asaturatingconcentration
of VCA (14 ?M) completely blocked Arp2/3 binding to
the beads (see the Supplementary Material). Therefore,
physical binding studies also support a model in which
cortactin NTA and N-WASp VCA compete for a single
binding site, i.e., Arp3, within Arp2/3 complex, and
N-WASp VCA has additional binding sites.
Cortactin Does Not Inhibit Activation
of Arp2/3 Complex by N-WASp VCA
Totestforfunctionalcompetitionbetweencortactinand
N-WASp VCA, we asked whether cortactin NTA inhibits
the ability of VCA to activate Arp2/3 complex. Cortactin
NTA alone, without the F-actin binding domain, binds
but does not activate Arp2/3 complex [10, 18]. We de-
signedpyreneactinpolymerizationassayswithasubsa-
turating concentration of VCA (0.5 nM) and a low con-
centration of Arp2/3 complex (12.5 nM), to which we
added excess NTA. Concentrations of NTA up to 22 ?M
did not affect activation of Arp2/3 complex by VCA (Fig-
ure 4). Experiments with bothtagged and untagged pro-
teins(VCAwithGST-NTA,GST-VCAwithNTA,GST-VCA
with GST-NTA) yielded similar results (data not shown).
We calculated that virtually all of the Arp2/3 complex
should have been bound to NTA (calculated for 22 ?M
NTA), assuming equilibrium conditions and conserva-
tion of mass and using previously determined Kds of
1.0–1.3 ?M and 0.2–0.75 ?M for the binding of cortactin
and VCA, respectively, to Arp2/3 complex [11, 18], (Fig-
ure 3C).
Figure 5. Effect of VCA Mutations on Arp2/3 Activation and Subunit
Binding
(A) Acidic domain sequences of N-WASp and cortactin with indica-
tions of the mutants used in this study. The asterisks show the point
mutation sites.
(B) Pyrene actin polymerization assay: wild-type or mutant GST-
VCA, as indicated, was added to 12.5 nM Arp2/3 and 2.5 ?M actin
inpolymerizationbuffer.ControlsincludedArp2/3alone,actinalone,
and each GST-VCA protein alone. n ? 3.
(C)Crosslinkingstudies:wild-typeormutantGST-VCA,asindicated,
(0.4 ?M) was incubated with 0.5 ?M Arp2/3 complex.
(D) Densitometry analysis of crosslinking results: the degree of
crosslinking of wild-type or mutant VCAs was evaluated by densi-
tometry of films from the subunit-specific Western blots, such as
the ones shown in (C). The data are plotted as the percentage of
Binding of the VCA DDW Motif to Arp3 Is Not
Necessary for Arp2/3 Complex Activation
TheapparentcompetitionofcortactinNTAforVCAbind-
ing to the Arp3 subunit combined with the lack of func-
tionalinhibitionsuggeststhatbindingofVCAtotheArp3
wild-typecrosslinking(mean ?standarderrorfor threeindependent
experiments). The black solid bars represent wild-type GST-VCA,
the diagonally hatched bars represent GST-W503A, and the gray
solid bars represent GST-?498–505.

Page 6

Cortactin, N-WASp, and Arp2/3 Complex
1275
subunit may not be necessary for VCA activation of
Arp2/3 complex. To test this hypothesis, we mutated
the DDW binding motif that is shared with cortactin
(Figure 5A). We truncated the C terminus by eight amino
acids(?498–505),thusdeletingtheDDWmotif.Inpyrene
actin polymerization assays, ?498–505 VCA activated
Arp2/3 complex as well as did wild-type VCA (Figure
5B). This experiment was performed with GST fusions
of VCA; untagged versions also showed no difference
between wild-type and mutant VCAs (data not shown).
To test for interaction of the mutant VCA with the Arp3
subunit, chemical crosslinking was performed. Cross-
linking of the ?498–505 VCA to Arp3 was decreased
by 81 ? 7% relative to wild-type VCA. Crosslinking to
ARPC1/p40 was decreased by 37 ? 7%, and crosslink-
ing to Arp2 was not changed (representative blot shown
in Figure 5C, quantitation by densitometry of films from
three separate experiments graphed in Figure 5D). A
VCA mutant in which the conserved tryptophan in the
DDW motif was changed to an alanine (W503A) had
unchanged crosslinking to Arp3, Arp2, or ARPC1/p40
and unchanged activity (Figure 5). It is possible that the
removal of acidic residues from ?498–505 VCA affected
the crosslinking of VCA to Arp3 without fully abolishing
thebindingofVCAtoArp3.However,thesedatasupport
the prediction that the shared DDW motif in VCA and
cortactinisinvolvedinArp3bindingandthatthisinterac-
tion is not necessary for activation of Arp2/3 complex
by N-WASp VCA.
Figure 6. Effect of Cortactin Mutation on Binding and Activation of
Arp2/3 Complex
(A) Arp2/3 pull-down: Arp2/3 complex binding by cortactin proteins
was assayed by bead pull-down assays. Cortactin proteins (un-
tagged) or GST (as a negative control) were covalently bound to
Sepharose. Equal volumes of beads were incubated with mouse
brain lysate. Protein bound to the beads was analyzed with an anti-
Arp3 Western blot. The amount of protein on the beads was as
follows:cortactin,80?g;cortactinW22A,160?g;cortactinfragment
1–269, 180 ?g; cortactin fragment 350–546, 180 ?g; GST, 40 ?g;
and BeadsOnly, 0 ?g.This blot waspurposely overexposedin order
to visualize any faint bands. Lanes: cortactin: full-length cortactin;
cortactin W22A: full-length cortactin with a point mutation in the
Arp2/3 binding domain; cortactin 1–269: amino acids 1–269 of cor-
tactin; cortactin 350–546: amino acids 350–546 of cortactin; GST:
glutathione S-transferase; Brain Lysate: 40 ?g protein loaded as a
control. n ? 3.
(B)Pyreneactinpolymerizationassay:500nMfull-lengthHis-tagged
cortactin or His-tagged W22A mutant cortactin was coincubated
with 100 nM Arp2/3 complex and 2.5 ?M actin in polymerization
buffer. n ? 3.
Mutation of Tryptophan 22 Abolishes Binding
and Activation of Arp2/3 Complex by Cortactin
Cortactin binds Arp3 primarily and also has a conserved
tryptophan as part of a DDW motif in its acidic domain.
We tested the importance of this tryptophan in the inter-
action of cortactin with Arp2/3 complex. The W22A cor-
tactin mutant protein did not bind Arp2/3 complex in a
pull-down assay (Figure 6A). Cortactin W22A also did
not activate Arp2/3 complex in a pyrene actin polymer-
ization assay (Figure 6B). Thus, the conserved trypto-
phan in the DDW motif is essential for the interaction of
cortactin with Arp2/3 complex. The DDW motif may be
most important for proteins that bind primarily to Arp3.
Discussion
Model for Binding of Cortactin and N-WASp VCA
to Arp2/3 Complex
Cortactin and N-WASp share a common acidic domain
that binds Arp2/3 complex, but they are able to cooper-
ate and even synergize in activating Arp2/3 complex
[10, 11]. Our results here provide an explanation for
this apparent paradox by demonstrating that a ternary
complex between cortactin, Arp2/3 complex, and
N-WASpVCAcanform.CortactinandN-WASpcompete
for binding to one subunit, the Arp3 subunit, of Arp2/3
complex. N-WASp has additional interactions with the
Arp2andARPC1/p40subunits,whicharenotfoundwith
cortactin and are not competed by cortactin (see model
in Figure 7).
These conclusions are based on several results. In a
chemicalcrosslinkingexperiment,cortactinblockedthe
interaction of N-WASp VCA with the Arp3 subunit, but
not with the Arp2 or ARPC1/p40 subunits. In a physical
binding (pull-down) experiment, cortactin caused a
modest decrease in the apparent affinity of VCA for
Arp2/3 complex, but did not prevent the interaction. By
contrast, N-WASp VCA completely blocked both the
binding of cortactin to Arp2/3 complex in a pull-down
assay and the chemical crosslinking of cortactin to the
Arp3 subunit.
TheinteractionofN-WASpVCA withtheArp3subunit,
which is competed by cortactin, appears to play little
or no role in the activation of Arp2/3 complex. An
N-WASp VCAtruncation mutant,lacking theDDW motif,
no longer can be crosslinked to Arp3 but activates
Arp2/3 complex normally. Also, high concentrations of

Page 7

Current Biology
1276
Figure 7. Model of Coactivation by N-WASp
and Cortactin
(A) N-WASp alone binding to the Arp2, p40,
and Arp3 subunits of Arp2/3 complex.
N-WASp(red)isshownboundtoArp2/3com-
plex (pink) at an actin (aqua) end-to-side
branch point. Also shown are the N-WASp
activators cdc42 and phosphatidylinositol
4,5-bisphosphate(PIP2).Thearrangementsof
Arp3, Arp2, and ARPC1/p40 are based on
previous crosslinking and structural studies
[4, 5].
(B) N-WASp and cortactin bound to Arp2/3
complex at an actin filament branch point.
Cortactin (dark blue) binds to Arp3 and dis-
places the extreme C-terminal region of
N-WASp. Cortactin is also shown binding ac-
tin filaments via its central repeats domain and a third molecule (“X”), such as dynamin or SHANK, via its SH3 domain. Cortactin is drawn
here as binding the mother filament, but it may bind the daughter filament instead.
cortactin NTA compete away the VCA-Arp3 interaction
but have no effect on the activation of Arp2/3 complex
byVCA.Therefore,theinteractionsofN-WASpVCAwith
other subunits, namely, Arp2 and ARPC1/p40, may be
more important for activation.
The combination of our chemical crosslinking, bind-
ing, and functional studies indicates that a ternary func-
tional complex of cortactin, N-WASp, and Arp2/3 com-
plex can form. In this ternary complex, the F-actin
binding site of cortactin probably promotes and stabi-
lizes the interaction of Arp2/3 complex with the mother
actin filament, and N-WASp induces strong activation
of Arp2/3 complex to nucleate the formation of the
daughter filament. The interaction of cortactin with
Arp2/3 complex can be abolished by excess N-WASp
VCA, so whether a ternary complex forms in vivo will
depend on the microenvironment, including the local
concentrations of cortactin and activated N-WASp.
N-WASp and cortactin were coimmunoprecipitated
from lysates of src-transformed fibroblasts [23], consis-
tent with the formation of a ternary complex. Immunolo-
calization studies reveal that cortactin and N-WASp co-
localize at certain sites of actin polymerization, such as
podosomes [23]; however, some actin-rich structures,
such as Vaccinia-induced actin tails, contain cortactin
but not N-WASp [31].
F-actin, maycause Arp2/3complex toassume thesame
or different conformational states. In other words, the
conformational pathway followed by Arp2/3 complex
may differ depending on its interactions with N-WASp,
cortactin, actin monomers, and actin filaments.
Integration of Signaling Pathways Directed
at Arp2/3 Complex
We found that cortactin is long, thin, and flexible and
that it exists as a monomer, based on a combination of
structural approaches.Electron micrographsreveal that
one end of cortactin binds Arp2/3 complex, and this is
consistent with previous studies localizing the Arp2/3
binding domain to the N-terminal portion of cortactin
[18]. Together, all our findings are consistent with the
idea that cortactin can bindsimultaneously to three pro-
teins — Arp2/3 complex at its N terminus, an actin fila-
ment in its central region, and a third molecule, such
as dynamin, Shank, CortBP, or ZO-1 [17], via the SH3
domain near its C terminus. One recent biochemical
study confirms that all three interactions can exist at
the same time (Hou et al., personal communication).
In addition to binding a number of signaling proteins
at its SH3 domain, cortactin is also a physiologically
significant substrate for tyrosine phosphorylation by src
kinases [17, 32, 33]. The presence of these multiple
interactions and modifications may allow cortactin to
integrate diverse signals directed at Arp2/3 complex.
Signals can also arrive at Arp2/3 complex via N-WASp,
by its activation by PIP2, Cdc42, and Grb2 and/or Nck,
and by its interaction with WIP [2]. The finding that cor-
tactin and N-WASp can simultaneously bind and coop-
erate in activation of Arp2/3 complex suggests that di-
verse signals may be integrated at the level of Arp2/3
complex as well.
Mechanisms for Arp2/3 Complex Activation
by Cortactin and N-WASp
The mechanisms by which cortactin and N-WASp acti-
vate Arp2/3 complex and enhance actin assembly are
different. Both bind Arp2/3 complex, but in different
ways. Cortactin binds Arp2/3 complex via Arp3, and
N-WASpbindsviaArp3,Arp2,andArpC1/p40.TheDDW
motif in cortactin is essential for Arp2/3 activation,
whereas the same motif inN-WASp VCA is not. N-WASp
binds monomeric actin, enhancing nucleation [2], and
cortactin binds actin filaments, stabilizing actin branch-
points [10].
Structural studies suggest that Arp2/3 complex un-
dergoes a majorconformational change upon activation
[4, 5]. The number and nature of the conformational
states available to Arp2/3 complex is not known.
N-WASp and cortactin, in combination with G- and
Conclusions
We investigated the interactions of cortactin and
N-WASp with Arp2/3 complex. Cortactin and N-WASp
compete for a common site on the Arp3 subunit;
N-WASp has additional interactions with the ARPC1/
p40 and Arp2 subunits, which are not competed by
cortactin. In cells, diverse signaling pathways that im-

Page 8

Cortactin, N-WASp, and Arp2/3 Complex
1277
Chemical Crosslinking
Purified proteins were coincubated in 60 mM KCl, 2 mM MgCl2, 20
mM HEPES, 0.1 mM ATP, 0.025% Thesit (pH 7.4) for 10 min at 25?C
before the addition of 1.5 mM EDC/NHS, added from a fresh 1:1
mixture of 100 mM stocks in DMSO (Pierce Chemical Company).
Crosslinking reactions were terminated after 30 min by the addition
of SDS-PAGE sample buffer, such that the final concentration of
Tris was 100 mM. The reactions were analyzed by SDS-PAGE on
vertical slab gels [41]. Using the Mini-Transblot system (BioRad),
proteins were transferred to nitrocellulose membranes. Membranes
were blocked overnight in 1% BSA (w/v), 0.1% (v/v) Tween-20 in
150 mM NaCl, 20 mM NaH2PO4 (pH 7.4), then probed with
1:20,000–1:500,000 dilutions of affinity-purified Arp2/3 subunit-spe-
cific rabbit antibodies. After washing, blots were then probed with
1:60,000 dilutions of anti-rabbit-HRP secondary antibodies (Bio-
Source), washed, and developed by Enhanced Chemiluminescence
(Amersham Pharmacia Biotech), with exposure to Hyperfilm (Amer-
sham Pharmacia Biotech). Densitometry of the films was performed
with an Epson scanner and NIH Image software.
pinge on N-WASp and cortactin may be integrated at
the level of Arp2/3 complex and actin assembly.
Experimental Procedures
Chemicals
Unless otherwise noted, chemicals and reagents were purchased
from Sigma Chemical or Fisher Scientific.
Plasmid Construction
The GST-VCA construct was kindly provided by Marie-France Car-
lier, CNRS, Gif-sur-Yvette, France. Mutation of the GST-VCA [34]
and GST-cortactin [18] plasmids was performed with the Quik-
Change Site-Directed Mutagenesis Kit (Stratagene).
Antibodies
Polyclonal antibodies directed against Arp3, p41, p16 [35], p34, p21
[36], and Arp2 and p20 [37] were previously described.
Protein Purification
Arp2/3 complex was purified from bovine brain by GST-VCA affinity
chromatography [34] or bovine thymus by conventional chromatog-
raphy[8].Inallcases,whencompared(inbothchemicalcrosslinking
and pyrene actin polymerization assays), the method of purification
did not affect the experimental results. GST-VCA was purified by
glutathione affinity chromatography, as previously described [34].
ForthepreparationofuntaggedVCAs,GST-VCAbeadswererotated
at room temperature for 90 min with 15 U/ml bovine thrombin (Cal-
biochem) in 150 mM NaCl, 2.5 mM CaCl2, 50 mM Tris, 1 mM DTT
(pH 7.5). The supernatant was collected, and untagged VCAs were
further purified by anion exchange chromatography on a Mono-Q
column (Pharmacia) with a gradient of 30–500 mM NaCl in 10 mM
Tris, 1 mM EDTA (pH 8.0). Actin was purified from chicken pectoralis
skeletal muscle [38] and was gel filtered [39]. Full-length and smaller
domains of cortactin protein were purified from BL21 E. coli by
glutathione affinity chromatography followed by GST-tag cleavage
and removal as previously described [18], followed by anion ex-
change chromatography on a Mono-Q column (Pharmacia) with a
gradient of 40–500 mM NaCl in 10 mM Tris, 1 mM EDTA (pH 8.0).
Cortactin proteins were concentrated by rebinding to a Mono-Q
column and then step eluting. Protein concentrations were calcu-
lated from the UV absorption at 280 nm, based on the tryptophan
and tyrosine content of each protein, except for the untagged VCAs,
which were quantitated by UV absorption at 205 nM due to a lack
of tryptophan or tyrosine residues in the mutant VCAs.
KdDeterminations by GST-VCA Pull-Down
of Purified Arp2/3 Complex
GST-VCA bound to glutathione-agarose was quantitated by analyz-
ing a range of bead volumes along with a standard curve of bovine
serum albumin by SDS-PAGE, followed by Coomassie staining and
densitometry of the gel (NIH Image). GST-VCA beads (0–8 ?M) were
incubated with 50 nM Arp2/3 complex in the presence or absence
of 30 ?M NTA in 60 mM KCl, 2 mM MgCl2, 20 mM HEPES (pH
7.4) ? 4 mg/ml bovine serum albumin. Glutathione-agarose was
also added to various samples so that all of the bead volumes
were equivalent. The samples were agitated for 10 min at room
temperature, before centrifugation. The supernatants were col-
lected and analyzed by SDS-PAGE, followed by immunoblotting
for Arp3, as described above. Depletion of the supernatant was
determinedbydensitometryofthefilms(NIHImage),andconcentra-
tions of bound and free VCA were calculated from the amount of
Arp2/3 complex depleted from the supernatant. The binding results
were plotted as bound VCA versus free VCA with Kaleidograph
software and were fit by the following equation:
[VCAbound] ? {(Kd? [Arp2/3] ? [VCA]) ? ?((Kd?
[Arp2/3] ? [VCA])2? (4•[Arp2/3]•[VCA]))}/2.
Statistical analysis was performed with Kaleidograph software.
Arp2/3 Pull-Downs from Mouse Brain Lysate
Recombinant untagged cortactin proteins or GST were covalently
linked to CNBr-Sepharose. A 50% suspension (40 ?l) of beads was
incubated with 800 ?l (4.5 mg/ml) of mouse brain lysate for 30 min
before washing in 50 mM KCl, 1 mM EDTA, 0.1 mM ATP, 1% NP40,
20 mM HEPES (pH 7.4), as previously described [18].
Electron Microscopy
Purified proteins were diluted to 5 ?g/ml in 70 mM KCl, 3 mM MgCl2,
30 mM HEPES (pH 7.4). Samples were adsorbed to mica flakes at
25?C, quick frozen, and freeze dried as described [28, 29]. Replicas
were observed at 70,000? magnification in a standard transmission
electronmicroscopeoperatingat100kV.Suitableareaswerephoto-
graphed on film as stereo pairs, by using ? 10? tilt of the micro-
scope’s goniometer stage, and were converted to digital images at
5? higher magnification with a Kodak 560 digital copy camera.
Stereo pairs were then converted to anaglyphs, and optimally sepa-
rated molecules were selected by hand as 300 ? 300 pixel crops
with Adobe PhotoShop. Images were masked by superimposing
identical three-dimension dark and light images, viewing them with
stereo glasses, and burning through the top layer. These masks
were used to measure the length of cortactin molecules and to
highlight the molecules. The three-dimensionality of the images is
preserved by using this process.
Supplementary Material
Supplementary Material including the Experimental Procedures and
two figures is available at http://images.cellpress.com/supmat/
supmatin.htm.
Acknowledgments
We are grateful to Dr. Matt Welch for a generous gift of Arp2/3
complex subunit-specific antibodies and to Robyn Roth for help
with the electron microscopy studies. We would also like to thank
Yan Li and Michelle Kaiser for technical assistance with protein
preparationandDr.MartinWearandDr.NaomiMorissetteforcritical
reading of the manuscript. This work was supported by National
Institutes of Health (NIH) grant GM 38542 to J.A.C. and by NIH-NCI
grant CA-29243 to J.T.P. A.M.W. is supported by a Howard Hughes
Medical Institute Postdoctoral Research Fellowship for Physicians.
Actin Polymerization Assays
Pyrene actin polymerization assays were performed as described
[40] on a PTI QuantaMaster spectrofluorometer. Cortactins and/or
the GST-VCA fragment of human N-WASp were incubated with
Arp2/3 complex in 50 mM KCl, 2 mM MgCl2, 20 mM HEPES, 0.1 mM
ATP, 0.025% Thesit (pH 7.4) (“polymerization buffer”) at 25?C. Actin
polymerization was initiated by the addition of monomeric actin
(7.5% pyrene labeled) and was monitored by continuous measure-
ment of fluorescence at 386 nm.
Received: February 4, 2002
Revised: May 31, 2002
Accepted: June 13, 2002
Published: August 6, 2002

Page 9

Current Biology
1278
References
21. Taunton, J., Rowning, B.A., Coughlin, M.L., Wu, M., Moon, R.T.,
Mitchison, T.J., and Larabell, C.A. (2000). Actin-dependent pro-
pulsion of endosomes and lysosomes by recruitment of
N-WASP. J. Cell Biol. 148, 519–530.
22. Wu, H., Reynolds, A.B., Kanner, S.B., Vines, R.R., and Parsons,
J.T. (1991). Identification and characterization of a novel cy-
toskeleton-associated pp60src substrate. Mol. Cell. Biol. 11,
5113–5124.
23. Mizutani, K., Miki, H., He, H., Maruta, H., and Takenawa, T.
(2002).EssentialroleofneuralWiskott-Aldrichsyndromeprotein
in podosome formation and degradation of extracellular matrix
in src-transformed fibroblasts. Cancer Res. 62, 669–674.
24. Kalman, D., Weiner, O.D., Goosney, D.L., Sedat, J.W., Finlay,
B.B., Abo, A., Bishop, J.M. (1999). Enteropathogenic E. coli acts
through WASP and Arp2/3 complex to form actin pedestals.
Nat. Cell Biol. 1, 389–391.
25. Cantarelli, V.V., Takahashi, A., Yanagihara, I., Akeda, Y., Imura,
K., Kodama, T., Kono, G., Sato, Y., Iida, T., and Honda, T. (2002).
Cortactin is necessary for F-actin accumulation in pedestal
structures induced by enteropathogenic Escherichia coli infec-
tion. Infect. Immun. 70, 2206–2209.
26. Huang, C., Ni, Y., Wang, T., Gao, Y., Haudenschild, C.C., and
Zhan, X. (1997). Down-regulation of the filamentous actin cross-
linking activity of cortactin by Src-mediated tyrosine phosphor-
ylation. J. Biol. Chem. 272, 13911–13915.
27. Cantor, C.R., and Schimmel, P.R. (1980). Biophysical Chemistry
Part II: Techniques for the Study of Biological Structure and
Function. (San Francisco: W. H. Freeman and Company).
28. Heuser, J.E. (1983). Procedure for freeze-drying molecules ad-
sorbed to mica flakes. J. Mol. Biol. 169, 155–195.
29. Heuser, J. (1989). Protocol for 3-D visualization of molecules
on mica via the quick-freeze, deepetch technique. J. Electron
Microsc. Tech. 13, 244–263.
30. Lee, W.L., Bezanilla, M., and Pollard, T.D. (2000). Fission yeast
myosin-I, Myo1p, stimulates actin assembly by Arp2/3 complex
and shares functions with WASp. J. Cell Biol. 151, 789–799.
31. Zettl, M., and Way, M. (2001). New tricks for an old dog? Nat.
Cell Biol. 3, E74–E75.
32. Okamura, H., and Resh, M.D. (1995). p80/85 cortactin associ-
ates with the Src SH2 domain and colocalizes with v-Src in
transformed cells. J. Biol. Chem. 270, 26613–26618.
33. Shah, K., and Shokat, K.M. (2002). A chemical genetic screen
for direct v-Src substrates reveals ordered assembly of a retro-
grade signaling pathway. Chem. Biol. 9, 35–47.
34. Egile, C., Loisel, T.P., Laurent, V., Li, R., Pantaloni, D., Sanso-
netti, P.J., and Carlier, M.F. (1999). Activation of the CDC42
effector N-WASP by the Shigella flexneri IcsA protein promotes
actin nucleation by Arp2/3 complex and bacterial actin-based
motility. J. Cell Biol. 146, 1319–1332.
35. Yarar, D., To, W., Abo, A., and Welch, M.D. (1999). The Wiskott-
Aldrich syndrome protein directs actin-based motility by stimu-
lating actin nucleation with the Arp2/3 complex. Curr. Biol. 9,
555–558.
36. Welch,M.D., DePace,A.H., Verma,S., Iwamatsu,A., andMitchi-
son, T.J. (1997). The human Arp2/3 complex is composed of
evolutionarily conserved subunits and is localized to cellular
regions of dynamic actin filament assembly. J. Cell Biol. 138,
375–384.
37. Gournier, H., Goley, E.D., Niederstrasser, H., Trinh, T., and
Welch, M.D. (2001). Reconstitution of human Arp2/3 complex
reveals critical roles of individual subunits in complex structure
and activity. Mol. Cell 8, 1041–1052.
38. Spudich,J.A.,andWatt,S.(1971).Theregulationofrabbitskele-
tal muscle contraction. J. Biol. Chem. 246, 4866–4871.
39. Pollard, T.D. (1986). Rate constants for the reactions of ATP-
and ADP-actin with the ends of actin filaments. J. Cell Biol. 103,
2747–2754.
40. Cooper,J.A.,Walker,S.B.,andPollard,T.D.(1983).Pyreneactin:
documentation of the validity of a sensitive assay for actin poly-
merization. J. Muscle Res. Cell Motil. 4, 253–262.
41. Laemmli, U.K. (1970). Cleavage of structural proteins during the
assemblyoftheheadofbacteriophageT4.Nature227,680–685.
1. Borisy,G.G.,andSvitkina,T.M.(2000).Actinmachinery:pushing
the envelope. Curr. Opin. Cell Biol. 12, 104–112.
2. Higgs,H.N.,andPollard,T.D.(2001).Regulationofactinfilament
network formation through Arp2/3 complex: activation by a di-
verse array of proteins. Annu. Rev. Biochem. 70, 649–676.
3. Mullins, R.D., Heuser, J.A., and Pollard, T.D. (1998). The interac-
tion of Arp2/3 complex with actin: nucleation, high affinity
pointed end capping, and formation of branching networks of
filaments. Proc. Natl. Acad. Sci. USA 95, 6181–6186.
4. Volkmann, N., Amann, K.J., Stoilova-McPhie, S., Egile, C., Win-
ter, D.C., Hazelwood, L., Heuser, J.E., Li, R., Pollard, T.D., and
Hanein, D. (2001). Structure of Arp2/3 complex in its activated
state and in actin filament branch junctions. Science 293, 2456–
2459.
5. Robinson, R.C., Turbedsky, K., Kaiser, D.A., Marchand, J.-B.,
Higgs, H.N., Choe, S., and Pollard, T.D. (2001). Crystal structure
of Arp2/3 complex. Science 294, 1679–1684.
6. Zalevsky, J., Grigorova, I., and Mullins, R.D. (2001). Activation
of the Arp2/3 complex by the Listeria ActA protein: ActA binds
two actin monomers and three subunits of the Arp2/3 complex.
J. Biol. Chem. 276, 3468–3475.
7. Machesky, L.M., Mullins, R.D., Higgs, H.N., Kaiser, D.A., Blan-
choin, L., May, R.C., Hall, M.E., and Pollard, T.D. (1999). Scar,
a WASp-related protein, activates nucleation of actin filaments
by the Arp2/3 complex. Proc. Natl. Acad. Sci. USA 96, 3739–
3744.
8. Higgs, H.N., Blanchoin, L., and Pollard, T.D. (1999). Influence
of the C terminus of Wiskott-Aldrich syndrome protein (WASp)
and the Arp2/3 complex on actin polymerization. Biochemistry
38, 15212–15222.
9. Bailly, M., Ichetovkin, I., Grant, W., Zebda, N., Machesky, L.M.,
Segall, J.E., and Condeelis, J. (2001). The F-actin side binding
activity of the Arp2/3 complex is essential for actin nucleation
and lamellipod extension. Curr. Biol. 11, 620–625.
10. Weaver, A.M., Karginov, A.V., Kinley, A.W., Weed, S.A., Li, Y.,
Parsons, J.T., and Cooper, J.A. (2001). Cortactin promotes and
stabilizes Arp2/3-induced actin filament network formation.
Curr. Biol. 11, 370–374.
11. Uruno, T., Liu, J., Zhang, P., Fan Yx, Y., Egile, C., Li, R., Mueller,
S.C., and Zhan, X. (2001). Activation of Arp2/3 complex-medi-
atedactinpolymerizationbycortactin.Nat.CellBiol.3,259–266.
12. Goode, B.L., Rodal, A.A., Barnes, G., and Drubin, D.G. (2001).
Activation of the Arp2/3 complex by the actin filament binding
protein Abp1p. J. Cell Biol. 153, 627–634.
13. Wu, H., and Parsons, J.T. (1993). Cortactin, an 80/85-kilodalton
pp60src substrate, is a filamentous actin-binding protein en-
riched in the cell cortex. J. Cell Biol. 120, 1417–1426.
14. Schuuring, E., Verhoeven, E., Litvinov, S., and Michalides, R.J.
(1993). The product of the EMS1 gene, amplified and overex-
pressed in human carcinomas, is homologous to a v-src sub-
strate and is located in cell-substratum contact sites. Mol. Cell.
Biol. 13, 2891–2898.
15. Li, Y., Tondravi, M., Liu, J., Smith, E., Haudenschild, C.C., Kacz-
marek, M., and Zhan, X. (2001). Cortactin potentiates bone me-
tastasis of breast cancer cells. Cancer Res. 61, 6906–6911.
16. Patel, A.S., Schechter, G.L., Wasilenko, W.J., and Somers, K.D.
(1998). Overexpression of EMS1/cortactin in NIH3T3 fibroblasts
causes increased cell motility and invasion in vitro. Oncogene
16, 3227–3232.
17. Weed, S.A., and Parsons, J.T. (2001). Cortactin: coupling mem-
brane dynamicsto corticalactin assembly.Oncogene 20,6418–
6434.
18. Weed, S.A., Karginov, A.V., Schafer, D.A., Weaver, A.M., Kinley,
A.W., Cooper, J.A., and Parsons, J.T. (2000). Cortactin localiza-
tion to sites of actin assembly in lamellipodia requires interac-
tions with F-actin and the Arp2/3 complex. J. Cell Biol. 151,
29–40.
19. Kaksonen,M.,Peng,H.B.,andRauvala,H.(2000).Associationof
cortactin with dynamic actin in lamellipodia and on endosomal
vesicles. J. Cell Sci. 113, 4421–4426.
20. Schafer, D.A., D’Souza-Schorey, C., and Cooper, J.A. (2000).
Actin assembly at membranes controlled by ARF6. Traffic 1,
892–903.