.elsevier.com/locate/bba

Biochimica et Biophysica Acta

The Cas family docking protein, HEF1, promotes the formation of

neurite-like membrane extensions

Sharmilla D. Bargon a, Peter W. Gunning a,b, Geraldine M. O’Neill a,b,*

a Oncology Research Unit, The Children’s Hospital at Westmead, 2145, Sydney, Australiab Discipline of Paediatrics and Child Health, University of Sydney, 2006, Australia

Received 17 May 2005; received in revised form 4 October 2005; accepted 21 October 2005

Available online 14 November 2005

Abstract

The Cas family proteins are a family of adhesion docking molecules that mediate protein–protein interactions and contribute to a number of

signal transduction pathways. Recent studies of two family members, p130Cas and Sin, have suggested that they may play a role in neurite

formation. The current study demonstrates that the third family member, HEF1, can also stimulate the formation of neurite-like processes, in the

presence of Rho kinase inhibitors. The HEF1-promoted processes actively extend from the cell body and resemble neurites both in the manner of

process extension and in the distribution of adhesion-associated molecules. The HEF1-promoted processes are dependent on the presence of an

intact microtubule system and can be inhibited by co-expression of either constitutively active Rac or Cdc42 GTPase. Together, our data support a

role for the Cas proteins in regulating cellular morphologies that contribute to tissue specialization.

D 2005 Elsevier B.V. All rights reserved.

Keywords: Focal adhesion; Integrin; Cas protein; Neurite; Migration; GTPase

1. Introduction

The Cas family of docking proteins are grouped together

based on their overall conserved domain structure and high

degree of sequence homology (reviewed in [1]). Included in the

family are p130Cas/BCAR1 [2,3], Sin/Efs [4,5] and HEF1/

Cas-L/Nedd9 [6,7]. Each protein contains multiple interaction

domains that can mediate interaction with partner molecules

containing poly-proline motifs and Src-homology 2 domains

(SH2) and molecules that interact with phosphorylated serine

and threonine consensus motifs (reviewed in [1]). Both

p130Cas and Sin contain additional poly-proline motifs that

can mediate interactions with SH3-domain containing partner

molecules. A number of studies have indicated that p130Cas

and HEF1 can localize to focal adhesions [2,6,8–10] and

together with their overall protein–protein interaction domain

structure this has led to their description as adhesion docking

molecules. Numerous studies have been carried out to identify

0167-4889/$ - see front matter D 2005 Elsevier B.V. All rights reserved.

doi:10.1016/j.bbamcr.2005.10.008

* Corresponding author. Oncology Research Unit, The Children’s Hospital at

Westmead, 2145, Sydney, Australia. Tel.: +61 2 9845 3116; fax: +61 2 9845

3078.

E-mail address: GeraldiO@chw.edu.au (G.M. O’Neill).

the repertoire of binding proteins (some examples are reviewed

in [1] and [11]) and together these studies have implicated Cas

proteins as playing a role in adhesion signalling and adhesion-

mediated functions.

Currently, it is not clear whether the Cas proteins have

distinct or similar functions. It appears unlikely that they have

redundant function, since knockout of the p130Cas gene in

mice is embryonic lethal [12], suggesting that HEF1 and Sin

cannot compensate for the loss of p130Cas. However, for a

number of functions that have been studied, the molecules

appear to have a similar effect. For example, both p130Cas and

HEF1 have been separately implicated in the promotion of

cellular migration and in both cases this is dependent on an

interaction with Focal Adhesion Kinase (FAK) and requires

intact p130Cas and HEF1 substrate binding domains [13–18].

Both p130Cas and Efs have been shown to stimulate Src kinase

activity and downstream signalling [4,19,20] and all three

molecules have been implicated in the regulation of various

GTPases via recruitment of specific downstream partners

[17,18,21–24]. To date, the majority of the studies have

focused on p130Cas.

Given the large number of molecules known to interact

with Cas proteins (reviewed in [1]), it is probably not

1746 (2005) 143 – 154

http://www

S.D. Bargon et al. / Biochimica et Biophysica Acta 1746 (2005) 143–154144

surprising that the Cas proteins have been implicated in a

wide variety of biological processes. Intriguingly, HEF1 has

been implicated in the apparently unrelated biological

processes of cell migration [13–15] and apoptosis [10,37].

However, a common requirement in both processes is the

integration of information from the external cellular environ-

ment with the regulation of adhesion dynamics and morpho-

logical control. In cell migration, signals received through the

adhesion sites are co-ordinated to regulate adhesion turnover

with lamellipodial protrusion and trailing edge retraction,

while cell rounding during apoptosis is accompanied by

break down of cytoskeletal linkage to cell adhesion sites and

a corresponding loss of focal adhesions. Therefore, we

propose that HEF1 (and likely the other Cas family proteins)

may function by co-ordinating adhesion dynamics that

regulate cellular morphologies required for distinct cellular

functions.

The lack of redundancy implied by the embryonic lethality

of the p130Cas gene knockout [12] suggests that the three

molecules may instead have tissue specific function. Indeed,

while the expression pattern of p130Cas is reported to be

ubiquitous [2], early studies of HEF1 suggested that the

expression of this molecule is more restricted to pre-B and T-

cells and epithelially-derived tissue [6,7,25], while in contrast

Efs is reported to be restricted to brain tissue [5]. However, a

recent study examining the expression pattern of HEF1 mRNA

in rat embryos found that HEF1 is expressed very early in the

developing hindbrain [26]. Together with the fact that one of

the first reports of HEF1 cDNA was in a screen for neural

precursor cell expressed and developmentally down-regulated

molecules (NEDD9 [27]), it therefore appears that HEF1 may

also play a role in neuronal physiology.

Both p130Cas and Sin have been shown to undergo tyrosine

phosphorylation in response to neuritic signals in neuronal cell

types [28–31] and subsequent interaction of Sin with the

adaptor molecule Crk was shown to be important for neurite

formation in cultured PC12 cells [28]. However, Cas protein

involvement in the formation of long neuritic-like processes is

not simply restricted to neuronal-type cells. A number of

studies have now suggested that these molecules may play a

part in the cellular machinery that controls the extension of

long thin membrane processes. For example, a pathway

involving a ternary complex between p130Cas, Crk and the

adaptor molecule CHAT/SHEP1/NSP3 can stimulate the

formation of long branched membrane processes from NIH3T3

fibroblasts [31]. Furthermore, in cultured B-cells and HEK293

cells, over-expression of the GDP-exchange factor (GEF) and

p130Cas/HEF1 interacting partner AND-34/BCAR3/NSP2 can

stimulate the formation of long membrane protrusions [22],

most likely due to the activation of Cdc42 by the GEF activity

of AND-34. Therefore, the Cas proteins may function to

integrate signalling information that then determines cellular

morphologies, including the production of neurites from

neuronal cells. The prevailing view is that activation of Rac

and Cdc42 are required for neurite formation (reviewed in

[32]). Given the array of GTPase regulating molecules that

interact with Cas proteins, including molecules that regulate

Rac and Cdc42 activity [18,21,22,24,31], it is perhaps not

surprising that this family of molecules might contribute to the

formation of neurite-like structures.

In addition to the observation that HEF1 is expressed in the

developing hindbrain, Merrill et al. [26] have reported that

HEF1 mRNA is induced by retinoic acid treatment of

neuroblastoma cells, and this occurs prior to neurite extension.

Therefore, this has led us to ask whether, similar to p130Cas

and Sin, HEF1 can also promote neurite-like membrane

extensions and whether this occurs using similar mechanisms

to that described for p130Cas and Sin. To determine whether

HEF1 can promote membrane process formation we have

examined cells treated with Rho kinase inhibitors, a treatment

previously demonstrated to stimulate neurite-like process

formation in both neuronal cell types [33] and in non-neuronal

cell types [34]. Time-lapse imaging of epithelial cells expres-

sing inducible exogenous HEF1 reveals that HEF1 promotes

active process extension in the presence of Rho-kinase

inhibitors. Comparison of HEF1-promoted processes in epi-

thelial cells with neurites from neuronal cells reveals a similar

sub-cellular distribution of adhesion-associated molecules,

including HEF1, between the two structures, suggesting that

the HEF1 promoted processes are neurite-like. Further,

formation of the HEF1-promoted processes is demonstrated

to require intact microtubules and is inhibited by constitutively

active Rac and Cdc42.

2. Materials and methods

2.1. Cell lines, reagents and antibodies

MCF7 cell lines engineered to express tetracycline-regulatable HEF1 have

been previously described [13]. Cell lines used include HEF1.M1, HEF1.M2

and the vector control cell line CM1. Cells were maintained in Dulbecco’s

Modified Eagles Medium (DMEM) plus 10% Foetal Bovine Serum (FBS)

supplemented with 1 Ag ml�1 tetracycline to repress expression of the

transgene and 2 Ag ml�1 puromycin. For induction of HEF1 expression, cells

were trypsinized and replated into media without antibiotics and grown for 24

to 48 h. B35 rat neuroblastoma cells were maintained in DMEM plus 10%

FBS and 5 mM l-glutamine.

Constitutively active Rac (pEGFP.RacL61) and dominant negative Rac

(pEGP.RacN17) constructs were obtained from Beric Henderson (Westmead

Millenium Institute) with the kind permission of Mark Phillips (New York

University School of Medicine). Constitutively active Cdc42L61 and

dominant negative Cdc42N17 were purchased from Upstate Biotechnology

(Lake Placid, NY, USA). The Cdc42 cDNA inserts were sub-cloned into

EcoR1–XhoI digested pEGFP, to create in-frame fusions with GFP.

Constructs were verified by Western blot analysis of transfected cell lysates.

Monoclonal antibody to paxillin was from BD Transduction Laboratories

(Franklin Lakes, NJ, USA) and to polymerized-tubulin was from Sternberger

monoclonals (Maryland, USA). The monoclonal anti-HEF1 antibody (clone

2G9) is from ImmuQuest (Cleveland, UK) and anti-h1integrin (clone P4C10)

from Invitrogen Life Technologies (Carlsbad, CA, USA). TRITC-phalloidin

was purchased from Sigma Aldrich (St. Lois, MO, USA), vectashield

mounting medium from Vector Laboratories (Berlingame, CA, USA), alexa

488-conjugated donkey anti-mouse and alexa 488-conjugated donkey anti-

rabbit from Jackson Immunological Labs (West Grove, PA, USA). Y-27632

and H-1152 were from Calbiochem (La Jolla, CA, USA). Dibutyryl-cyclic

AMP and Nocodazole were from Sigma Aldrich (St. Lois, MO, USA).

Transfection reagent Lipofectamine 2000 was obtained from Invitrogen Life

Technologies (Carlsbad, CA, USA) and used according to the manufacturer’s

protocol.

S.D. Bargon et al. / Biochimica et Biophysica Acta 1746 (2005) 143–154 145

2.2. Rho kinase inhibition

Cells were either treated for 1, 4 and 24 h as indicated. In initial

experiments, HEF1.M1 cells were grown on coverslips under inducing

conditions for 48 h and then Y-27632 (10 AM) added for 4 h, cells were

then fixed and stained with phalloidin. Rho kinase inhibition with H-1152 was

achieved as described above, with a concentration of 10 AM H-1152.

2.3. Neurite production

B35 cells were plated onto glass coverslips 24 h prior to the addition of

drugs. To stimulate neurite production, either 1 mM dbCAMP or 10 AM Y-

27632 was added to cells in media with 0.5% FBS and cells were incubated for

a further 24 h before fixing and immunofluorescence analysis.

2.4. Immunofluorescence

Cells for immunofluorescence analysis were grown on glass coverslips,

fixed with 4% paraformaldehyde (PFA) in PBS and permeabilized with 0.2%

Triton X-100 in PBS with 0.5% bovine serum albumin (BSA), with the

exception of cells stained with anti-polymerized tubulin. Cell permeabilization

for microtubule staining was performed with �20 -C methanol for 5 min.

Antibody dilutions and immunostaining were performed in 0.5% BSA in

PBS.

2.5. Image capture and analysis

Epifluorescence and phase contrast images were captured using a Spot II-

cooled CCD digital camera (Diagnostics Instruments) mounted on an

Olympus Bx50 microscope with 40� and 60� objectives. DIC time-lapse

imaging was performed using a SPOT RT SE CCD camera mounted on a

Leica DMIRBE inverted microscope with a 40� oil objective. Images were

captured at 15-min intervals. Image analysis was performed using Image-Pro

plus Version 4.0 (Media Cybernetics) and final images prepared in Adobe

Photoshop.

2.6. Monitoring process formation

To analyse the effects of Rac and Cdc42 GTPases on process formation,

HEF1.M1 cells were induced overnight and transfected with GFP fusion

constructs the following day. Cells were incubated for 6 h in the presence of the

transfection complexes, then trypsinized and replated onto coverslips in 6-well

dishes. The following day cells were treated with Y-27632 and incubated for 4

h, then fixed and immunostained as described. Cells were co-stained with DAPI

to monitor nuclear morphology and only cells with healthy (non-apoptotic)

nuclei, were included in the analysis. HEF.M1 cells grown under non-inducing

conditions (that is in the presence of tetracycline) were transfected for 6 h in the

absence of tetracycline and replated in the presence of tetracycline to inhibit

HEF1 production.

2.7. Transwell assay of cellular migration

Prior to assay of cell migration, cells were grown under inducing or non-

inducing conditions for 48 h. 6�104 cells in serum-free media were then

plated into the top chamber of an 8-Am pore transwell filter (Falcon) mounted

in 24-well dishes. The media in the bottom wells contained foetal bovine

serum. Cells under non-induced conditions were supplemented with

tetracycline and other inhibitors were added coincident with plating into the

top well of the trans-well filter. Following 6 h incubation at 37 -C and 5%

CO2 cells from the top of the filter were removed with a cotton bud and cells

on the underside were stained with Diff Quik (Lab Aids, Australia). Filters

were then examined with a 40� objective of a light microscope (Olympus)

and the total number of nuclei positive cells determined for 8 randomly

selected fields. Assays were carried out in triplicate (duplicate samples in

each assay) and the average number of cells migrating under each

experimental condition determined.

3. Results

3.1. HEF1 promotes membrane process formation

In light of recent data suggesting that the Cas proteins

p130Cas and Efs contribute to the formation of membrane

processes [28,29,31,22], we tested whether the third Cas family

protein, HEF1, has similar effects on cell morphology.

HEF1.M1 and HEF1.M2, two cell lines engineered to express

HEF1 under the control of a tetracycline-regulatable promoter

and a vector control cell line CM1 [13], were used as the model

to examine HEF1 effects on cellular morphologies. Cells were

treated with the Rho kinase inhibitor, Y-27632, a treatment

previously shown to promote neurite-like process formation

[33,34]. HEF1.M1 cells were induced for HEF1 expression and

then treated for 4 h with 10 AM Y-27632 [35]. Under these

conditions, and as early as 1 h after drug addition, HEF1.M1

cells exhibited a striking arborized phenotype with numerous

branched processes emanating from the cell body (Fig. 1A).

Confirming that this phenotype is due to increased HEF1

expression, there is little evidence of the branched phenotype in

either CM1 vector control cells or in uninduced HEF1.M1

clone treated with Y-27632 for the same length of time (Fig.

1B). Quantitation of process formation further confirmed that

HEF1 promotes process formation and this occurs for up to 24

h after addition of Y-27632 (Fig. 1C). Interestingly, clone

HEF1.M1, which expresses a higher level of HEF1 protein

than clone HEF1.M2 (Fig. 1D) also shows the greatest number

of processes, therefore suggesting that there may be a dose-

dependent effect of HEF1 on process formation. Finally, HEF1

is normally detected as two bands on western blots and there is

an interesting loss of the upper band in the cells treated with the

Rho kinase inhibitor. This suggests that there may be an

alteration in the phosphorylation of HEF1, similar to results we

have previously described [10].

To test whether the observed phenotype is dependent on

inhibition of Rho kinase activity and not due to non-specific

action of Y-27632, cells were treated with a second Rho kinase

inhibitor, H-1152 [36]. Treatment with H-1152 recapitulated

the results seen with Y-27632 with the cells induced to express

HEF1 again displaying branched processes (Fig. 1E). There-

fore, the inhibition of Rho kinase activity appears to be

sufficient for HEF1-mediated increases in process formation.

Since we have earlier shown that HEF1 expression can

promote apoptosis [10,37], it is possible that the processes

reflect membrane retraction fibres in dying cells. However,

three lines of evidence suggest this is not the case. Firstly, the

length of processes exceeds the size of the original cells.

Secondly, the processes contain bundles of microtubules (see

Fig. 4A) which are absent from retraction fibres [38]. Finally,

time-lapse imaging demonstrates that the majority of the

processes are formed by active membrane protrusion. Analysis

of processes in cells treated with Rho kinase inhibitors

demonstrated that 59% of processes exhibited active membrane

extension (Fig. 2B). There was a mixture of phenotypes, with

some processes representing trailing edges as has been

previously reported by others [39,40], others resulting from

S.D. Bargon et al. / Biochimica et Biophysica Acta 1746 (2005) 143–154146

active protrusion of membrane and the third phenotype was a

combination of the two, that is, active protrusion from the end

of a trailing edge. Examples of active membrane extension are

shown in Fig. 2A (arrows). In the first panel, ruffling

extensions are indicated by the arrows. In progressive frames

the membranes are clearly extending away from the cell body

Fig. 2. Membrane processes are actively extended in HEF1-expressing cells treated with Rho kinase inhibitors. (A) DIC time lapse images of induced HEF1.M1 cells

treated with Y-27632. Frames were taken at 15 min intervals, 40� objective. Indicated are examples of actively extending processes (arrows) and a trailing edge

(arrow head). (B) Membrane process formation was observed in time lapse series and the percentage of distinct process phenotypes calculated. A total of 71

individual processes were measured. (C) Migration of HEF1.M1 cells was assessed using a transwell migration assay. Cells were grown under induced (black bars)

or non-induced (white bars) conditions and the number of cells migrating to the under-side of the filter determined. (D) Transwell migration assays were performed in

the presence of the Rho kinase inhibitors, Y-27632 and H-1152 and the number of cells migrating to the under-side of the filter calculated and expressed as a percent

of the total number of cells migrating in the absence of the inhibitors. HEF1.M1 cells were grown under inducing (black bars) or non-inducing conditions (white

bars). *P <0.05, Student’s t-test.

S.D. Bargon et al. / Biochimica et Biophysica Acta 1746 (2005) 143–154 147

and are reminiscent of neurite-like membrane extensions. An

example of a trailing edge is also indicated in Fig. 2B (arrow

heads). Together, the data suggest that a combination of

behaviours contribute to overall process formation: anchoring

of the trailing edge, active process extension and a combination

of both. Importantly, the majority of the processes (59%)

represent active membrane extension.

Our live cell imaging analysis also suggested that Rho

kinase inhibition blocked HEF1 promoted cell migration. To

quantitate the effects of Rho kinase inhibition on HEF1-

promoted cell migration, we initially confirmed that the

induction of HEF1 expression following tetracycline with-

drawal resulted in approximately 5-fold enhanced cell migra-

tion of the HEF1.M1 cell line under induced conditions when

Fig. 1. HEF1 promotes membrane process formation. (A) Phase contrast images of H

AM Y-27632 for 1 h (right panel). Arrows indicate processes. (B) HEF1.M1 cells a

inducing conditions as indicated. Cells in the bottom panels were all treated with 10 AQuantitation of process formation in two clones, HEF1.M1 and HEF1.M2 and CM

conditions as indicated and then incubated in the presence of Y-27632 for 1, 4 or

counted and expressed per 100 nuclei. (D) Western blot analysis of HEF1 expressio

27632. (E) Phase contrast image of induced HEF1.M1 cells treated with 10 AM H

compared with uninduced control cells (Fig. 2C). Next, we

tested the effect of Rho kinase inhibition on HEF1-promoted

cell migration by using both Y-27632 and H-1152. In transwell

migration assays, the use of both inhibitors significantly

reduced HEF1-promoted cell migration to less than 60% of

that seen in control cells (Fig. 2D). In contrast, Rho kinase

inhibition of cells grown under non-inducing conditions had no

effect on the extent of cell migration (Fig. 2D). These data

correlate with earlier findings that Rho kinase inhibition can

inhibit cell migration by preventing tail retraction [39,40].

However, we note that in our cells we observe loss of the

motile morphology (that is loss of obvious leading and trailing

edges) and a switch to active membrane processes emanating

from around the cell periphery. Therefore, HEF1-promoted

EF1.M1 cells grown under induced conditions (left panel) and treated with 10

nd CM1 vector control cells grown under inducing conditions or control non-

M Y-27632. Cells are stained with TRITC-phalloidin to show cell structure. (C)

1 vector control cell line. Cells were grown under inducing or non-inducing

24 h. Cells were then fixed and immunostained and the number of processes

n in induced HEF1.M1 and HEF1.M2 cells in the presence and absence of Y-

-1152. Arrows indicate processes.

S.D. Bargon et al. / Biochimica et Biophysica Acta 1746 (2005) 143–154148

process formation occurs in the absence of both apoptosis and

cell migration, suggesting that process formation represents a

previously undescribed function of the HEF1 protein.

3.2. HEF1-promoted processes have a similar distribution of

adhesion molecules to that observed in neurites

Since the HEF1-promoted processes are reminiscent of

neurites, the pattern of HEF1 sub-cellular localization in the

HEF1-promoted processes was next compared with the

distribution of HEF1 in neuronal cells induced to produce

neurites. To carry out these experiments, we employed the rat

neuronal cell line B35, which has previously been shown to

produce neurites in response to 1 mM dbCAMP [41]. Western

blot analysis of B35 protein lysates confirms that the HEF1

Fig. 3. HEF1 promoted processes have a similar distribution of adhesion molecules

and induced HEF1.M1 cells probed with anti-HEF1 antibody. (B) Western blot of ex

HEF1.M1 cells grown under induced conditions (left hand panels) and B35 cells (

paxillin and anti-h-1 integrin antibodies as indicated. Arrows indicate focal adhes

membrane accumulation. Scale bar 25 Am. (D) Induced HEF1.M1 cells treated with

and anti-h1 integrin antibodies as indicated. Shown in the middle and bottom panels

fixed and immunostained as above. Images show detail of antibody stained membr

antibody cross-reacts with endogenous rat HEF1 and further

demonstrates that these cells express endogenous HEF1 protein

(Fig. 3A). Interestingly, examination of HEF1 protein expres-

sion in B35 cells treated with dbCAMP reveals a similar loss of

the upper form of HEF1 (Fig. 3B) as was seen in the Y-27632

treated HEF1.M1 and HEF1.M2 lysates (Fig. 1D). However,

no obvious difference is observed in the Y-27632 treated B35

cells (Fig. 3B).

To our knowledge, this is the first examination of HEF1

protein in a neuronal cell type, we therefore initially established

the sub-cellular distribution of HEF1 in control B35 cells.

HEF1 localizes to the perinuclear region of the B35 cells and

can also be detected as discrete punctate staining in the

cytoplasm (Fig. 3C, arrow heads). However, no localization to

classical focal adhesion structures is detected, although the

to that observed in neurites. (A) Western blot of total protein extracts from B35

tracts from control B35 cells (con.) and treated with Y-27632 and dbCAMP. (C)

right hand panels). Cells were fixed and immunostained with anti-HEF1, anti-

ions, arrow heads indicate punctate staining and asterisks indicate regions of

10 AM Y-27632 (top panels) and immunostained with anti-HEF1, anti-paxillin

are B35 neuronal cells treated with 1 mM dbCAMP or 10 AMY-27632 for 24 h,

ane processes enlarged from the inset images. Scale bar, 5 Am.

S.D. Bargon et al. / Biochimica et Biophysica Acta 1746 (2005) 143–154 149

HEF1 antibody can detect adhesion-associated HEF1 as

demonstrated by positive adhesion staining in the HEF1.M1

cells (Fig. 3C arrows). Paxillin staining further confirms that

classical focal adhesion structures do exist in the B35 cells

(Fig. 3C arrows). Hence, in the B35 cells, HEF1 is localized to

punctate regions within the cytoplasm. We next compared the

pattern of HEF1 staining to h1-integrin staining and found that

in contrast, h1-integrin staining appears to accumulate at the

membrane edge in both HEF1.M1 and B35 cells (Fig. 3C,

asterisks).

To determine whether the HEF1-promoted processes have a

similar molecular structure to neurites, the staining patterns of

HEF1 and paxillin were compared between HEF1.M1 induced

cells treated with Y-27632 and B35 cells treated with either Y-

27632 or dbCAMP (Fig. 3D). This analysis revealed firstly that

the HEF1 and paxillin staining patterns appear identical in both

HEF1.M1 and B35 processes (Fig. 3D). Secondly, we can

detect two populations of paxillin staining: punctate staining

throughout the cytoplasm of the process (Fig. 3D arrows) and

more dash-like adhesion staining at regions projecting from the

edges of the processes (Fig. 3D, arrow heads). This is a

prominent feature of the paxillin staining in all three conditions

of process formation. In contrast, HEF1 is restricted to punctate

staining throughout the cytoplasm of the processes in all three

treatments.

The HEF1 staining pattern appears similar to h1-integrinpositive ‘‘point contacts’’ that have been previously reported

[45,46]. Therefore we compared the distribution of HEF1 with

that of h1-integrin. The sub-cellular localization of h1-integrinis indistinguishable between the three conditions of membrane

process formation and further, appears similar to the distribu-

Fig. 4. HEF1-promoted processes require intact microtubules. (A) HEF1.M1 cells gr

an antibody to polymerized tubulin. (B) HEF1.M1 cells grown under inducing co

tubulin antibodies. Note the absence of organized microtubule structure. (C) Phas

presence of nocodazole and following the addition of Y-27632 or H-1152 as indica

tion of HEF1. However, we note that since both antibodies are

mouse monoclonals it is not possible to directly measure

whether HEF1 and h1-integrin are co-localized (Fig. 3C).

Together, these data suggest that the HEF1 promoted processes

have a similar distribution of adhesion molecules to that

observed in neurites. In summary, HEF1 and h1 integrin

display punctate staining along the length of the processes,

while paxillin is additionally localized to peripheral adhesion

structures.

3.3. HEF1-promoted processes require intact microtubules

Characteristic of neurites (reviewed in [38]) and long

membrane processes formed by other, non-neuronal cells

[33,34,42] is the appearance of parallel bundles of micro-

tubules within the processes. Immunostaining of induced

HEF1.M1 cells treated with Y-27632 using an antibody to

polymerized tubulin shows that HEF1-mediated processes also

contain bundles of microtubules (Fig. 4A). Currently, it is not

clear whether microtubules are required for the initiation of

neurite outgrowth, however they do appear to be necessary for

maintaining outgrowth [38] and furthermore were necessary for

process extension in non-neuronal cells [34,42]. Therefore,

induced HEF1.M1 cells treated were treated with nocodazole to

assess the role of polymerized microtubules in HEF1-promoted

process formation. The efficacy of the nocodazole in depoly-

merizing the microtubules is indicated by loss of microtubule

staining (Fig. 4B). Induced HEF1.M1 cells co-treated with both

nocodazole and either Y-27632 or H-1152 display morpholo-

gies that are indistinguishable from cells treated with nocoda-

zole alone and there is an absence of the membrane extensions

own under inducing conditions and treated with Y-27632. Cells are stained with

nditions and treated with 1 AM nocodazole and stained with anti-polymerized

e contrast images of HEF1.M1 cells grown under inducing conditions in the

ted.

S.D. Bargon et al. / Biochimica et Biophysica Acta 1746 (2005) 143–154150

observed following Rho kinase inhibition alone (Fig. 4C). We

note that the cells treated with nocodazole adopt an altered

morphology, with control cells expressing HEF1 losing their

characteristic motile phenotype (Fig. 4C). Similar to other

studies, it appears that HEF1 promoted process formation

requires intact microtubules.

3.4. Active Rac and Cdc42 inhibit HEF1-promoted processes

Previous data have implied a role for Cas proteins in

regulation of Rac and Cdc42 activity ([22] and reviewed in [1])

and given the role of these molecules in control of cell shape

(reviewed in [43]) and neurite outgrowth [32] we questioned

whether HEF1 might mediate different phenotypes via activa-

tion of these molecules. To test this, induced HEF1.M1 cells

were transfected with plasmids expressing either constitutively

active -Rac (Rac1L61) or -Cdc42 (Cdc42L61), and treated with

Y-27632. Surprisingly, expression of both constitutively active

constructs inhibited HEF1-promoted process formation in

the presence of Y-27632 (Fig. 5A and B and 6A and B).

Consequently, HEF1 does not appear to promote process

formation via the activation of either Rac or Cdc42. In contrast,

Fig. 5. Constitutively active Rac inhibits HEF1-mediated process formation. (A) HE

expressing either GFP control, GFP-tagged RacL61 or RacN17, as indicated, and tr

transfected with constructs expressing GFP, GFP-tagged RacL61 or RacN17, we

membrane processes calculated.

induced HEF1.M1 cells transfected with plasmids expressing

dominant negative Rac (RacN17) and Cdc42 (Cdc42N17) and

treated with Y-27632 still display process formation (Figs. 5A

and B and 6A and B). Together, these data suggest that

inhibition of Rac and Cdc42 activities are necessary for process

formation. However, suppression of the activity of Rac and

Cdc42 alone is not sufficient to promote process formation as

HEF1.M1 cells grown under non-inducing conditions and

transfected with either RacN17 or Cdc42N17 then treated with

Y-27632 did not display processes formation (Figs. 5B and

6B). Together, these data suggest that HEF1 does not regulate

process formation via activation of Rac or Cdc42 and indeed

suppression of the activity of these molecules is necessary,

although not sufficient, for HEF1-promoted process formation.

4. Discussion

Following earlier studies suggesting that the Cas proteins

p130Cas and Sin can promote neurite formation, we now show

that the third Cas family member, HEF1, can also promote

neurite-like process formation, in epithelially-derived cancer

cells. These data represent a previously undescribed function

F1.M1 cells grown under inducing conditions were transfected with constructs

eated with Y-27632 (see arrows). (B) Induced and non-induced HEF1.M1 cells

re treated with Y-27632 and the percentage of GFP-positive cells displaying

Fig. 6. Constitutively active Cdc42 inhibits HEF1-mediated process formation. (A) HEF1.M1 cells grown under inducing conditions were transfected with constructs

expressing either GFP control, GFP-tagged Cdc42L61 or Cdc42N17, as indicated, and treated with Y-27632 (see arrows). (B) Induced and non-induced HEF1.M1

cells transfected with constructs expressing GFP, GFP-tagged Cdc42L61 or Cdc42N17, were treated with Y-27632 and the percentage of GFP-positive cells

displaying processes calculated.

S.D. Bargon et al. / Biochimica et Biophysica Acta 1746 (2005) 143–154 151

for HEF1. In earlier data, we and others have shown that HEF1

can variously promote apoptosis [10,37] and migration [13–

15]. In the current study we demonstrate that the formation of

neurite structures occurs in the absence of either apoptosis or

migration. The HEF1-promoted processes are molecularly

similar to neuronal cell-derived neurites, both in their

requirement for microtubules and the distribution of adhe-

sion-related molecules. The formation of such membrane

extensions requires the co-ordinated interaction of multiple

cytoskeletal and cellular signalling molecules. Given the

protein docking function of the Cas proteins, it is possible that

they may play a role in helping to localize molecules required

for this process.

There are many parallels in the processes of cell migration

of non-neuronal cells and neurite extension (reviewed in [44]).

Both processes require the co-ordination of adhesion dynamics

with morphological controls [44,47], and so it is perhaps not

surprising that a molecule that has been shown to stimulate

migration in a number of different cell types [13–15] can also

stimulate the production of neurite-like membrane processes. It

is interesting that HEF1 promoted process formation requires

the inhibition of Rho kinase. In earlier studies induction of

HEF1 expression in the HEF1.M1 cells was shown to stimulate

Rho kinase mRNA [13]. An intriguing question is whether the

increased Rho kinase expression in response to elevated levels

of HEF1 might be required to balance membrane dynamics at

the leading edge. This could serve to promote cell migration

[13] rather than the membrane process extension we have

observed following Rho kinase inhibition. Indeed, the activa-

tion of Rho kinase has been previously suggested to restrict

membrane protrusion to the leading edge of migrating cells

[39].

Importantly, our data extend earlier observations of HEF1

mRNA expression [26] and now demonstrates HEF1 protein

expression in differentiated neuronal cell neurites. Therefore,

HEF1 is expressed at the right time [26] and in the right place

(as demonstrated by our staining of neuritic processes) to play a

role in neuronal cell architecture. Both the HEF1 promoted

processes in the HEF1.M1 cultures and the B35 cell neuritic

processes showed very similar pattern of adhesion molecule

staining: punctate HEF1, paxillin and h1-integrin staining

throughout the extent of the processes and additional discrete

dash-like paxillin staining at small projections from the edges

of the processes. Similar staining patterns to the HEF1 and h1-

S.D. Bargon et al. / Biochimica et Biophysica Acta 1746 (2005) 143–154152

integrin staining have previously been described as point

contacts [45,46], while vinculin has been previously shown to

give a similar distribution in neurites [45] to the results shown

here for paxillin. In the same way that adhesion dynamics are

important for migration of non-neuronal cells [47], they are

also crucial for neuronal cell architecture [44]. A key event

following Rho kinase inhibition in non-neuronal cells, is the

loss of stable focal adhesions and the formation instead of

rapidly turning over, pre-cursor focal complexes [48]. This is

also true in the switch from a non-migratory to a migratory

phenotype in fibroblast cell types where rapidly turning over

focal complexes become predominant [49,50]. Indeed, our data

demonstrate loss of large HEF1 and paxillin positive adhesions

in the HEF1.M1 cells following Rho kinase inhibition,

suggesting altered adhesion dynamics in these cells. Similarly,

dash-like paxillin positive focal adhesions are reduced in size

in B35 cells treated with Y-27632. Crucially, the same pattern

of altered paxillin adhesion staining is also seen in the B35

cells treated with dbCAMP. Therefore it is possible that the

morphological similarities between the HEF1 promoted pro-

cesses and the neuritic processes may be due to altered

adhesion dynamics.

In contrast to current data on the mechanism of neurite

extension suggesting a positive role for Rac and Cdc42

activation (reviewed in [32]), we find that constitutively active

Rac and Cdc42 inhibit HEF1 promoted membrane process

formation, while dominant negative Rac and Cdc42 have no

effect. Currently it is not possible to determine whether this is a

similar mechanism to that for Sin-promoted neurite outgrowth.

Sin-promoted neurite outgrowth requires an interaction with

Crk and while the authors speculated that Rac activation is a

likely target downstream of this interaction [28], it will be

necessary for this to be tested before a direct comparison can be

made with the HEF1 results presented here. However, similar

to our data, neurite-like extensions formed in NIH-3T3 cells in

response to the inhibition of ROCK and the ubiquitin ligase

Cbl were also shown to be independent of Rac and Cdc42

activity [34]. Of note, there is some contradictory data

regarding the absolute requirement for activated Rac and

Cdc42 in neurite extension. Studies have shown that constitu-

tively active Rac can inhibit total neurite length [51] and inhibit

neurite extension in Drosophila [52]. While activated Rac, and

dominant negative and activated Cdc42 have no effect on DRG

neurite outgrowth [53]. These apparently contradictory results

may be due to a requirement for localized activity of GTPases,

at restricted regions within the cell [54]. In our experiments, the

exogenous mutant GTPases are expressed throughout the cell

and presumably normal controls on the cycling of GTP and

GDP throughout the cell are therefore disrupted. However, if

this is the case, it is still difficult to understand why the mutant

GTPase constructs would have a positive effect on neurite

formation in one set of studies and a negative effect in the other

set of studies. An alternative explanation for the apparently

discrepant results is that there may be multiple pathways that

can determine process formation. The activation of each

pathway may be dependent on tissue and developmental

context.

The control of cell morphology is crucial to normal cellular

function and contributes to the specialization of different

tissues. Integral to morphological control is the ability to

regulate the spatial organization of the cytoskeleton and

associated signalling molecules. In the case of neuritic

extension the correct organization of these components is

required for successful neuronal cell function. Together with

this study, it now appears that the Cas family proteins may

contribute to the regulation of neurite extension, although

currently it is not clear to what extent each member of the

family might contribute to neuronal function. In a broader

context, we propose that the Cas proteins may be involved in

the co-ordination of adhesion site dynamics and signalling that

regulates specialized cellular morphologies. Further study of

the interactions between Cas proteins and their cognate

partners should illuminate the role of these molecules in

neurite extension.

Acknowledgments

P. Gunning is a Principal Research Fellow of the NHMRC

(#163626) and G. O’Neill is the NSW Cancer Council

Research Fellow. This study was further supported by a

Howard Florey Centenary Research Fellowship from the

NHMRC of Australia (#147117) to G. O’Neill and an NHMRC

project grant (#117409) to P. Gunning.

References

[1] G.M. O’Neill, S.J. Fashena, E.A. Golemis, Integrin signalling: a new

Cas(t) of characters enters the stage, Trends Cell Biol. (2000) 111–119.

[2] R. Sakai, A. Iwamatsu, N. Hirano, S. Ogawa, T. Tanaka, H. Mano, Y.

Yazaki, H. Hirai, A novel signaling molecule, p130, forms stable

complexes in vivo with v-Crk and v-Src in a tyrosine phosphorylation-

dependent manner, EMBO J. (1994) 3748–3756.

[3] A. Brinkman, F.S. van der, E.M. Kok, L.C. Dorssers, BCAR1, a human

homologue of the adapter protein p130Cas, and antiestrogen resistance in

breast cancer cells, J. Natl. Cancer Inst. (2000) 112–120.

[4] K. Alexandropoulos, D. Baltimore, Coordinate activation of c-Src by

SH3- and SH2-binding sites on a novel p130Cas-related protein, Sin,

Genes Dev. (1996) 1341–1355.

[5] M. Ishino, T. Ohba, H. Sasaki, T. Sasaki, Molecular cloning of a cDNA

encoding a phosphoprotein, Efs, which contains a Src homology 3 domain

and associates with Fyn, Oncogene (1995) 2331–2338.

[6] S.F. Law, J. Estojak, B. Wang, T. Mysliwiec, G. Kruh, E.A. Golemis,

Human enhancer of filamentation 1, a novel p130cas-like docking protein,

associates with focal adhesion kinase and induces pseudohyphal growth in

Saccharomyces cerevisiae, Mol. Cell Biol. (1996) 3327–3337.

[7] M. Minegishi, K. Tachibana, T. Sato, S. Iwata, Y. Nojima, C. Morimoto,

Structure and function of Cas-L, a 105-kD Crk-associated substrate-

related protein that is involved in beta 1 integrin-mediated signaling in

lymphocytes, J. Exp. Med. (1996) 1365–1375.

[8] M.T. Harte, M. Macklem, C.L. Weidow, J.T. Parsons, A.H. Bouton,

Identification of two focal adhesion targeting sequences in the adapter

molecule p130(Cas), Biochim. Biophys. Acta (2000) 34–48.

[9] T. Nakamoto, R. Sakai, H. Honda, S. Ogawa, H. Ueno, T. Suzuki, S.

Aizawa, Y. Yazaki, H. Hirai, Requirements for localization of p130cas to

focal adhesions, Mol. Cell. Biol. (1997) 3884–3897.

[10] G.M. O’Neill, E.A. Golemis, Proteolysis of the docking protein HEF1

and implications for focal adhesion dynamics, Mol. Cell. Biol. (2001)

5094–5108.

[11] A.H. Bouton, R.B. Riggins, P.J. Bruce-Staskal, Functions of the adapter

S.D. Bargon et al. / Biochimica et Biophysica Acta 1746 (2005) 143–154 153

protein Cas: signal convergence and the determination of cellular

responses, Oncogene (2001) 6448–6458.

[12] H. Honda, H. Oda, T. Nakamoto, Z. Honda, R. Sakai, T. Suzuki, T. Saito,

K. Nakamura, K. Nakao, T. Ishikawa, M. Katsuki, Y. Yazaki, H. Hirai,

Cardiovascular anomaly, impaired actin bundling and resistance to Src-

induced transformation in mice lacking p130Cas, Nat. Genet. (1998)

361–365.

[13] S.J. Fashena, M.B. Einarson, G.M. O’Neill, C. Patriotis, E.A. Golemis,

Dissection of HEF1-dependent functions in motility and transcriptional

regulation, J Cell Sci. (2002) 99–111.

[14] Y. Ohashi, S. Iwata, K. Kamiguchi, C. Morimoto, Tyrosine phosphory-

lation of Crk-associated substrate lymphocyte-type is a critical element in

TCR- and beta 1 integrin-induced T lymphocyte migration, J. Immunol.

(1999) 3727–3734.

[15] G.A. van Seventer, H.J. Salmen, S.F. Law, G.M. O’Neill, M.M.

Mullen, A.M. Franz, S.B. Kanner, E.A. Golemis, J.M. van Seventer,

Focal adhesion kinase regulates beta1 integrin-dependent T cell

migration through an HEF1 effector pathway, Eur. J. Immunol. (2001)

1417–1427.

[16] L.A. Cary, D.C. Han, T.R. Polte, S.K. Hanks, J.L. Guan, Identification of

p130Cas as a mediator of focal adhesion kinase-promoted cell migration,

J. Cell Biol. (1998) 211–221.

[17] R.L. Klemke, J. Leng, R. Molander, P.C. Brooks, K. Vuori, D.A. Cheresh,

CAS/Crk coupling serves as a ‘‘molecular switch’’ for induction of cell

migration, J. Cell Biol. (1998) 961–972.

[18] D.A. Cheresh, J. Leng, R.L. Klemke, Regulation of cell contraction and

membrane ruffling by distinct signals in migratory cells, J. Cell Biol.

(1999) 1107–1116.

[19] M.R. Burnham, P.J. Bruce-Staskal, M.T. Harte, C.L. Weidow, A. Ma, S.A.

Weed, A.H. Bouton, Regulation of c-SRC activity and function by the

adapter protein CAS, Mol. Cell. Biol. (2000) 5865–5878.

[20] L. Xing, C. Ge, R. Zeltser, G. Maskevitch, B.J. Mayer, K. Alexandro-

poulos, c-Src signaling induced by the adapters Sin and Cas is mediated

by Rap1 GTPase, Mol. Cell. Biol. (2000) 7363–7377.

[21] K. Vuori, H. Hirai, S. Aizawa, E. Ruoslahti, Introduction of p130cas

signaling complex formation upon integrin-mediated cell adhesion: a role

for Src family kinases, Mol. Cell. Biol. (1996) 2606–2613.

[22] D. Cai, K.N. Felekkis, R.I. Near, G.M. O’Neill, J.M. van Seventer, E.A.

Golemis, A. Lerner, The GDP exchange factor AND-34 is expressed in B

cells, associates with HEF1, and activates Cdc42, J. Immunol. (2003)

969–978.

[23] T. Gotoh, D. Cai, X. Tian, L.A. Feig, A. Lerner, p130Cas regulates the

activity of AND-34, a novel Ral, Rap1, and R-Ras guanine nucleotide

exchange factor, J. Biol. Chem. (2000) 30118–30123.

[24] V.C. Dodelet, C. Pazzagli, A.H. Zisch, C.A. Hauser, E.B. Pasquale, A

novel signaling intermediate, SHEP1, directly couples Eph receptors to R-

Ras and Rap1A, J. Biol. Chem. (1999) 31941–31946.

[25] S.N. Manie, A.R. Beck, A. Astier, S.F. Law, T. Canty, H. Hirai, B.J.

Druker, H. Avraham, N. Haghayeghi, M. Sattler, R. Salgia, J.D. Griffin,

E.A. Golemis, A.S. Freedman, Involvement of p130(Cas) and

p105(HEF1), a novel Cas-like docking protein, in a cytoskeleton-

dependent signaling pathway initiated by ligation of integrin or antigen

receptor on human B cells, J. Biol. Chem. (1997) 4230–4236.

[26] R.A. Merrill, A.W. See, M.L. Wertheim, M. Clagett-Dame, Crk-associated

substrate (Cas) family member, NEDD9, is regulated in human neuro-

blastoma cells and in the embryonic hindbrain by all-trans retinoic acid,

Dev. Dyn. (2004) 564–575.

[27] S. Kumar, Y. Tomooka, M. Noda, Identification of a set of genes with

developmentally down-regulated expression in the mouse brain, Biochem.

Biophys. Res. Commun. (1992) 1155–1161.

[28] L.T. Yang, K. Alexandropoulos, J. Sap, c-SRC mediates neurite

outgrowth through recruitment of Crk to the scaffolding protein Sin/Efs

without altering the kinetics of ERK activation, J. Biol. Chem. (2002)

17406–17414.

[29] S. Fagerstrom, S. Pahlman, E. Nanberg, Protein kinase C-dependent

tyrosine phosphorylation of p130cas in differentiating neuroblastoma

cells, J. Biol. Chem. (1998) 2336–2343.

[30] C. Anneren, K.A. Reedquist, J.L. Bos, M. Welsh, GTK, a Src-related

tyrosine kinase, induces nerve growth factor-independent neurite out-

growth in PC12 cells through activation of the Rap1 pathway. Relation-

ship to Shb tyrosine phosphorylation and elevated levels of focal adhesion

kinase, J. Biol. Chem. (2000) 29153–29161.

[31] A. Sakakibara, Y. Ohba, K. Kurokawa, M. Matsuda, S. Hattori, Novel

function of Chat in controlling cell adhesion via Cas–Crk–C3G-pathway-

mediated Rap1 activation, J. Cell Sci. (2002) 4915–4924.

[32] E.E. Govek, S.E. Newey, L. Van Aelst, The role of the Rho GTPases in

neuronal development, Genes Dev. (2005) 1–49.

[33] M. Hirose, T. Ishizaki, N. Watanabe, M. Uehata, O. Kranenburg, W.H.

Moolenaar, F. Matsumura, M. Maekawa, H. Bito, S. Narumiya, Molecular

dissection of the Rho-associated protein kinase (p160ROCK)-regulated

neurite remodeling in neuroblastoma N1E-115 cells, J. Cell Biol. (1998)

1625–1636.

[34] R.M. Scaife, D. Job, W.Y. Langdon, Rapid microtubule-dependent

induction of neurite-like extensions in NIH 3T3 fibroblasts by inhibition

of ROCK and Cbl, Mol. Biol. Cell (2003) 4605–4617.

[35] T. Ishizaki, M. Uehata, I. Tamechika, J. Keel, K. Nonomura, M. Maekawa,

S. Narumiya, Pharmacological properties of Y-27632, a specific inhibitor

of rho-associated kinases, Mol. Pharmacol. (2000) 976–983.

[36] M. Ikenoya, H. Hidaka, T. Hosoya, M. Suzuki, N. Yamamoto, Y.

Sasaki, Inhibition of rho-kinase-induced myristoylated alanine-rich C

kinase substrate (MARCKS) phosphorylation in human neuronal cells

by H-1152, a novel and specific Rho-kinase inhibitor, J. Neurochem.

(2002) 9–16.

[37] S.F. Law, G.M. O’Neill, S.J. Fashena, M.B. Einarson, E.A. Golemis, The

docking protein HEF1 is an apoptotic mediator at focal adhesion sites,

Mol. Cell. Biol. (2000) 5184–5195.

[38] L. Dehmelt, S. Halpain, Actin and microtubules in neurite initiation: are

MAPs the missing link? J. Neurobiol. (2004) 18–33.

[39] R.A. Worthylake, K. Burridge, RhoA and ROCK promote migration by

limiting membrane protrusions, J. Biol. Chem. (2003) 13578–13584.

[40] R.A. Worthylake, S. Lemoine, J.M. Watson, K. Burridge, RhoA is

required for monocyte tail retraction during transendothelial migration,

J. Cell Biol. (2001) 147–160.

[41] M. Boukhelifa, M.M. Parast, J.G. Valtschanoff, A.S. LaMantia, R.B.

Meeker, C.A. Otey, A role for the cytoskeleton-associated protein palladin

in neurite outgrowth, Mol. Biol. Cell (2001) 2721–2729.

[42] T. Omelchenko, J.M. Vasiliev, I.M. Gelfand, H.H. Feder, E.M.

Bonder, Mechanisms of polarization of the shape of fibroblasts and

epitheliocytes: separation of the roles of microtubules and Rho-

dependent actin-myosin contractility, Proc. Natl. Acad. Sci. U. S. A.

(2002) 10452–10457.

[43] C.D. Nobes, A. Hall, Rho, rac, and cdc42 GTPases regulate the assembly

of multimolecular focal complexes associated with actin stress fibers,

lamellipodia, and filopodia, Cell (1995) 53–62.

[44] J.S. da Silva, C.G. Dotti, Breaking the neuronal sphere: regulation of

the actin cytoskeleton in neuritogenesis, Nat. Rev., Neurosci. (2002)

694–704.

[45] C.O. Arregui, S. Carbonetto, L. McKerracher, Characterization of

neural cell adhesion sites: point contacts are the sites of interaction

between integrins and the cytoskeleton in PC12 cells, J. Neurosci.

(1994) 6967–6977.

[46] A. Renaudin, M. Lehmann, J. Girault, L. McKerracher, Organization of

point contacts in neuronal growth cones, J. Neurosci. Res. (1999)

458–471.

[47] A.J. Ridley, M.A. Schwartz, K. Burridge, R.A. Firtel, M.H. Ginsberg, G.

Borisy, J.T. Parsons, A.R. Horwitz, Cell migration: integrating signals

from front to back, Science (2003) 1704–1709.

[48] T. Tsuji, T. Ishizaki, M. Okamoto, C. Higashida, K. Kimura, T.

Furuyashiki, Y. Arakawa, R.B. Birge, T. Nakamoto, H. Hirai, S.

Narumiya, ROCK and mDia1 antagonize in Rho-dependent Rac activa-

tion in Swiss 3T3 fibroblasts, J. Cell Biol. (2002) 819–830.

[49] S.K. Sastry, K. Burridge, Focal adhesions: a nexus for intracellular

signaling and cytoskeletal dynamics, Exp. Cell Res. (2000) 25–36.

[50] K. Rottner, A. Hall, J.V. Small, Interplay between Rac and Rho in the

control of substrate contact dynamics, Curr. Biol. (1999) 640–648.

[51] T. Kubo, T. Yamashita, A. Yamaguchi, H. Sumimoto, K. Hosokawa, M.

S.D. Bargon et al. / Biochimica et Biophysica Acta 1746 (2005) 143–154154

Tohyama, A novel FERM domain including guanine nucleotide

exchange factor is involved in Rac signaling and regulates neurite

remodeling, J. Neurosci. (2002) 8504–8513.

[52] M.J. Allen, X. Shan, R.K. Murphey, A role for Drosophila Drac1 in

neurite outgrowth and synaptogenesis in the giant fiber system, Mol. Cell.

Neurosci. (2000) 754–765.

[53] A.E. Fournier, B.T. Takizawa, S.M. Strittmatter, Rho kinase inhibition

enhances axonal regeneration in the injured CNS, J. Neurosci. (2003)

1416–1423.

[54] K. Aoki, T. Nakamura, M. Matsuda, Spatio-temporal regulation of Rac1

and Cdc42 activity during nerve growth factor-induced neurite outgrowth

in PC12 cells, J Biol. Chem. (2004) 713–719.