www.bba-direct.com

Biochimica et Biophysica Acta 1693 (2004) 73–80

Pro-apoptotic signaling pathway activated by echistatin in GD25 cells

Elena Alimentia, Simona Tafurib, Antonio Scibellia, Danila d’Angeloa, Laura Mannac,Luigi Michele Pavonec, M. Antonietta Belisariob, Norma Staianoa,*

aDipartimento di Strutture, Funzioni e Tecnologie Biologiche, Universita di Napoli Federico II, Naples, ItalybDipartimento di Biochimica e Biotecnologie Mediche, Universita di Napoli Federico II, Naples, Italy

cDipartimento di Scienze Cliniche Veterinarie, Universita di Napoli Federico II, Naples, Italy

Received 30 January 2004; received in revised form 18 May 2004; accepted 19 May 2004

Available online 15 June 2004

Abstract

Disintegrins, low molecular weight RGD-containing polypeptides isolated from snake venoms, have seen use as integrin antagonists in

the field of tumor biology and angiogenesis. In this study, we investigated the molecular mechanism by which the disintegrin echistatin

affects cell adhesion and signaling resulting in an apoptotic response in the GD25 cell system. Wild-type GD25 cells, which lack expression

of the h1 family of integrin, and stable transfectants expressing the A isoform of h1 integrin subunit were used.

Nanomolar concentrations of echistatin detached fibronectin- and vitronectin-adherent GD25 cells from immobilized substratum.

However, prior to inducing detachment of adherent cells, echistatin caused apoptosis as measured by caspase-3 activation. Either cell

detachment or apoptotic response induced by echistatin were more pronounced on fibronectin-adherent GD25 cells than on vitronectin-

adherent ones.

GD25 cell exposure to echistatin caused a reduction of tyrosine phosphorylation levels of pp125FAK, whereas it didn’t affect either Shc

tyrosine phosphorylation levels or Shc–Grb2 functional association. The down-regulation of pp125FAK preceded apoptosis and cell

detachment induced by echistatin. Our results indicate that pp125FAK and not Shc pathway is involved in echistatin-induced apoptotic

response in the GD25 cell system.

D 2004 Elsevier B.V. All rights reserved.

Keywords: Echistatin; GD25 cell; Apoptosis; FAK; Shc

1. Introduction

Disintegrins are a family of low molecular weight

proteins isolated from Crotalidae and Vipiridae snake

venoms [1]. They contain an Arg–Gly–Asp (RGD) motif

which represents a common adhesion recognition site [2].

However, disintegrins with other recognition motifs such as

KGD, VGD or MLD have been identified [3–5]. Disinte-

grins bind with high affinity integrin receptors on platelet

and cell surface, thus inhibiting both platelet aggregation

and cell adhesion [6]. Additional structural features other

than the RGD motif, including the amino acids flanking the

RGD sequence, the pattern of intramolecular S–S bridges,

and the presence in their molecule of other recognition sites

affect disintegrin activity [7]. Thus, structurally distinct

0167-4889/$ - see front matter D 2004 Elsevier B.V. All rights reserved.

doi:10.1016/j.bbamcr.2004.05.007

* Corresponding author. Tel.: +39-81-2536108; fax: +39-81-2536097.

E-mail address: staianor@unina.it (N. Staiano).

disintegrins may differently act on cell adhesion and sig-

naling [8,9].

Disintegrins have seen use as integrin antagonists in the

fields of tumor biology and angiogenesis [10,11]. Various

disintegrins have been shown to inhibit angiogenesis either

in vitro and in vivo by inducing apoptosis [12]. Echistatin

was shown to induce an apoptotic response occurring prior

to cell detachment in avh3-transfected human embryonic

kidney-293 epithelial cells [13]. However, the molecular

mechanism of action of disintegrins in causing apoptosis has

not been fully clarified yet.

We previously showed that echistatin inhibits experi-

mental Lewis lung carcinoma-induced metastasis in vivo

[14] and we demonstrated that this disintegrin detaches

fibronectin-adherent melanoma cells by down-regulating

pp125FAK phosphorylation [15]. The echistatin-induced

dephosphorylation of pp125FAK Tyr 118 residue resulted

in a decrease of paxillin tyrosine phosphorylation and in the

disruption of pp125FAK-Src and pp125FAK-paxillin func-

E. Alimenti et al. / Biochimica et Biophysica Acta 1693 (2004) 73–8074

tional complexes [16]. Herein, we also demonstrated that a

reduction of tyrosine phosphorylation of pp125FAK, paxillin

and Shc is involved in apoptosis of COS cells induced by

ochratoxin A, a mycotoxin with nephrotoxic, teratogenic

and carcinogenic activity which seems to be involved in the

pathogenesis of Balkan Endemic Nephropathy and urinary

tract tumors [17].

In this study, we evaluated the ability of echistatin to

affect cell adhesion and signaling in the mouse cell line

GD25. These cells, derived from the embryonic stem cell

clone G201, lack expression of the h1 family of integrin

heterodimers due to the disruption of the h1 subunit gene

[18]. By using wild-type GD25 cells and stable transfectants

of these cells expressing h1A isoform of h1 integrin subunit,

the molecular mechanism by which echistatin induces an

apoptotic response was investigated. In particular, the FAK

and Shc signaling pathways activated by the disintegrin

interaction with integrin receptor(s) were characterized.

2. Materials and methods

2.1. Antibodies

Horseradish peroxidase conjugated goat anti-(mouse

IgG) Ig and horseradish peroxidase conjugated goat anti-

(rabbit IgG) Ig were purchased from Sigma Chemical Co.

(St. Louis, MO, USA); rabbit polyclonal anti-human FAK

IgG from Upstate Biotechnology (Lake Placid, NY, USA);

monoclonal anti-phosphotyrosine (PY20) IgG, monoclonal

mouse anti-Grb2 IgG and rabbit polyclonal anti-Shc Ig from

Transduction Laboratories (Lexington, KY, USA).

2.2. Chemicals

Aprotinin, bovine serum albumin (BSA), Dulbecco

minimum Eagle’s medium (DMEM), human plasma

fibronectin, leupeptin, orthovanadate, pepstatin, N-acetyl-

Asp-Glu-Val-Asp-7-amino-4-methylcoumarin (Ac-DEVD-

AMC), 1,4-piperazinediethane sulfonic acid (PIPES),

ethylenediaminetetraacetic acid (EDTA), 3-[(3-cholamido-

propyl)dimethylammonio]-1-propanesulfonate (CHAPS),

1,1-dichloro-2,2-bis(4-chlorophenyl)ethane (DDT) were

purchased from Sigma Chemical Co.; L-glutamine and tryp-

sin from ICN Biomedicals, Inc. (Aurora, OH, USA); fetal

bovine serum (FBS) fromHyclone Laboratories, Inc. (Logan,

UT, USA); protein A-agarose and protein G-agarose from

Santa Cruz Biotechnology, Inc. (Santa Cruz, CA, USA).

2.3. Cell culture

GD25 cells were kindly provided by Prof. Francesco

Retta (Dipartimento di Genetica, Biologia e Biochimica,

Universita di Torino, Torino, Italy). GD25 cells were

grown on plastic Petri dishes in DMEM supplemented

with 10% FBS and 1% L-glutamine at pH 7.4, and cultured

under standard cell culture conditions (37 jC, 5% CO2). In

the culture medium of GD25 h1A cells hygromycin B was

also added. Cells were harvested for propagation or cell

attachment studies by treatment with 0.25% trypsin/0.02%

EDTA in phosphate-buffered saline (PBS), pH 7.2, and

with 5 mM EDTA in PBS, respectively. Cells were then

washed with DMEM and resuspended in complete DMEM

for propagation or in DMEM with 2% FBS for adhesion

experiments.

2.4. Substrate coating

Tissue culture plastic dishes (35� 10 mm) or 96-multi-

well plastic dishes (Costar, Cambridge, MA, USA) were

coated overnight at 4 jC by incubation with human plasma

fibronectin (10 Ag/ml) diluted in PBS with 1 mM CaCl2 and

1 mM MgCl2, or with vitronectin (5 Ag/ml) diluted in PBS.

After coating, dishes were treated with 1% BSA in PBS for

30 min at 37 jC to block free binding sites on the plastic.

Before plating cells, dishes were rinsed twice with PBS.

2.5. Cell adhesion assay

Freshly suspended cells (100 Al, 75� 104 cells/ml) were

plated onto 96-multiwell plates previously coated with

fibronectin or vitronectin and allowed to adhere for 3 h at

37 jC. Non-adherent cells were removed by gentle washing

with PBS. Adherent cells were exposed to 5 Ag/ml of

echistatin for increasing time intervals (0–6 h) or to

different concentrations of echistatin (0.1–20 Ag/ml) for

3 h and incubated at 37 jC.At the indicated time, detached cells were gently re-

moved by washing with PBS, adherent cells were fixed, and

stained by crystal violet. The colour yields were measured

by a Bio-Rad ELISA reader equipped with a 570 nm filter.

The percentage of cell detachment induced by echistatin

was evaluated as previously described [15].

2.6. Apoptosis assay

Freshly suspended cells (1�106 cells) were plated on

substratum-coated dishes, allowed to adhere at 37 jC for 3 h.

Non-adherent cells were removed by gentle washing with

serum-free medium. Adherent cells were exposed to in-

creasing amounts of echistatin (0.1–10 Ag/ml) for 3 h or to

5 Ag/ml of echistatin for increasing time intervals (0–180

min). At the end of the incubation time, cells were lysed by

ice-cold lysis buffer containing 10 mM Tris–HCl, pH 7.5,

125 mM NaCl, 1% Triton, 10 mM Na4P2O7, 10 mM

Na2HPO4, 22.7 Ag/ml aprotinin (0.1 trypsin inhibitor units/

ml), 10 Ag/ml leupeptin, 0.4 mM Na3VO4. The lysates were

clarified by centrifugation at 12,000� g for 10 min at 4 jC.The amount of proteins in the samples was determined by the

Bio-Rad DC protein assay (Bio-Rad Laboratories). Lysates

containing 150 Ag of proteins were incubated with 50 AM of

Ac-DEVD-AMC in 1 ml of reaction buffer (10 mM PIPES,

E. Alimenti et al. / Biochimica et Biophysica Acta 1693 (2004) 73–80 75

pH 7.4, 2 mM EDTA, 0.1% CHAPS, 5 mM DTT) for 5 h at

37 jC. Fluorescence derived from the release of the 7-amino-

4-methylcoumarin moiety was followed using a Perkin-

Elmer with E excitation at 390 nm and E emission at 460

nm. The data obtained are reported as the variation (D) of the

fluorescence of the treated sample lysate minus the fluores-

cence of the appropriate control sample lysate.

2.7. Protein immunoprecipitation and Western blotting

Freshly suspended cells (1�106 cells) were plated on

substrate-coated dishes and allowed to adhere at 37 jC for

3 h. Non-adherent cells were removed by two gentle wash-

ings with serum-free medium, adherent cells were exposed to

20 Ag/ml of echistatin for 1 or 3 h. At the end of the

incubation time, the proteins were detergent-extracted by

ice-cold lysis buffer, as described in Apoptosis assay. The

lysates were clarified by centrifugation at 12,000� g for 10

min at 4 jC and the amount of proteins in the samples was

determined by the Bio-Rad DC protein assay.

Lysates containing equal amount of proteins were incu-

bated with anti-pp125FAK (1 Ag 100 Ag proteins) or anti-Shc

Ig (0.25 Ag 100 Ag proteins) for 6 h at 4jC. Protein A- or

protein G-agarose was then added to the samples and

incubated overnight at 4jC. Beads were sedimented by

brief centrifugation and immunoprecipitates were washed

extensively with cold lysis buffer. Proteins were resus-

pended in SDS sample buffer, boiled for 5 min in Laemmli

buffer and run on 10% SDS/polyacrilamide gel.

After electrophoresis, the proteins were transferred to

nitrocellulose using a Mini trans-blot apparatus (Bio-Rad

Laboratories) according to manufacturer’s instructions.

Membranes were blocked for 1 h at 42jC with blocking

buffer containing 5% BSA in Tris/NaCl/Pi and incubated

overnight with a 1:2000 dilution of antibody PY20 to probe

for phosphotyrosine-containing proteins. The blots were

washed three times with Tris/NaCl/Pi/Tween (150 mM

NaCl, 20 mM Tris–HCl, pH 7.4, 0.3% Tween 20) and

incubated for 1 h with peroxidase conjugated anti-(mouse

IgG) Ig diluted 1:3000 in Tris/NaCl/Pi, 1% BSA.

To evaluate the amount of Grb2 co-precipitating with

Shc, cellular extracts were immunoprecipitated with anti-

Shc Ig as previously described, and separated by SDS gel

electrophoresis. After electrophoresis, the proteins were

transferred to nitrocellulose using a Mini trans-blot appara-

tus (Bio-Rad Laboratories) according to manufacturer’s

instructions. The membranes were blocked for 1 h with

blocking buffer containing 5% milk in Tris/NaCl/Pi and

incubated overnight with a 1:500 dilution of antibody Grb2.

The blots were washed three times with Tris/NaCl/Pi/Tween

(150 mM NaCl, 20 mM Tris–HCl, pH 7.4, 0.3% Tween 20)

and incubated for 1 h with peroxidase conjugated anti-

(rabbit IgG) Ig diluted 1:3000 in Tris/NaCl/Pi, 2.5% milk.

The proteins were visualized by an ECL chemilumines-

cence kit (Amersham Corp., Little Chalfont, UK). The same

blots were stripped and reprobed using anti-pp125FAK or

anti-Shc Ig to confirm equal loading of pp125FAK or Shc in

the lysates.

3. Results

3.1. Echistatin causes detachment of GD25 cells from

immobilized fibronectin or vitronectin

The ability of echistatin to promote cell detachment from

immobilized substrates was evaluated. Echistatin induced

GD25 cell detachment from immobilized fibronectin, but

this effect was more pronounced on fibronectin-adherent

GD25 h1A cells than on wild-type GD25 cells. Cell

treatment with 20 Ag/ml of echistatin for 3 h induced 87%

of GD25 h1A cell detachment from immobilized fibronec-

tin, while the same dose caused 76% of wild-type GD25 cell

detachment from immobilized fibronectin (Fig. 1, panel A).

Echistatin promoted cell detachment from immobilized

vitronectin, but this effect was more pronounced on wild-

type GD25 cells than on GD25 h1A: 3 h of cell treatment

with 20 Ag/ml echistatin caused 45% and 54% of cell

detachment for GD25 h1A cells and wild-type GD25 cells,

respectively (Fig. 1, panel A).

The time-course of echistatin-induced cell detachment

confirmed the observation that the disintegrin is more active

to promote cell detachment from fibronectin than from

vitronectin and that wild-type cells are easier detached from

vitronectin than GD25 h1A cells, whereas GD25 h1A are

more easily detached from immobilized fibronectin than

wild-type cells (Fig. 1, panel B).

We had previously observed that the attachment to

immobilized fibronectin and vitronectin which bind a5h1

and avh3 integrin receptors on cell surface, respectively, is

efficient for both wild-type GD25 cells and the GD25 h1A

cells. Furthermore, h1A transfected GD25 cells showed a

higher percentage of adhesion to immobilized fibronectin

than wild-type GD25 cells, whereas the percentage of wild-

type GD25 cells adhering to immobilized vitronectin was

higher than that of transfected-ones [19]. This may be due to

the observed overexpression of the avh3 integrin receptor,

specific for vitronectin, on wild-type cells which lack the h1

integrin subunit [20]. Thus, the results obtained indicate that

echistatin is able to promote cell detachment by competing

with both immobilized fibronectin and vitronectin for their

respective integrin receptor. However, echistatin activity

was more pronounced on fibronectin-adherent GD25 cells

than on vitronectin-adherent GD25 cells, thus suggesting

that the disintegrin has higher affinity for fibronectin recep-

tor than for vitronectin receptor.

3.2. Echistatin promotes apoptosis of GD25 cells adhering

to immobilized fibronectin or vitronectin

The ability of echistatin to promote apoptosis in adherent

GD25 cells was evaluated by measuring the activation of

Fig. 2. Apoptosis of adherent cells induced by echistatin. Fibronectin-

(5,n) or vitronectin- (D,E) adherent cells were treated with different

doses of echistatin for 3 h (panel A) or with 5 Ag/ml of echistatin for

different time intervals (panel B). Control cells were treated with fresh

culture medium for the same time intervals than echistatin-treated cells.

Each value represents the average of six determinations performed in

triplicateF standard deviation indicated by vertical bars.

Fig. 1. Cell detachment induced by echistatin. Fibronectin- (5,n) or

vitronectin- (D,E) adherent cells were treated with different doses of

echistatin for 3 h (panel A) or with 5 Ag/ml of echistatin for different time

intervals (panel B). Control cells were treated with fresh culture medium for

the same time intervals than echistatin-treated cells. Each value represents

the average of six determinations performed in triplicateF standard

deviation indicated by vertical bars.

E. Alimenti et al. / Biochimica et Biophysica Acta 1693 (2004) 73–8076

caspase-3, which represents an early hallmark during apo-

ptosis [21]. Echistatin caused apoptosis in both wild-type

GD25 and GD25 h1A cells cultured on either immobilized

vitronectin or fibronectin. Fig. 2 reports the increase in the

fluorescence as measured in echistatin-treated cells with

respect to control (untreated) cells.

Echistatin-induced apoptotic response was more pro-

nounced on fibronectin-adherent GD25 h1A cells than

on wild-type GD25 cells, whereas vitronectin-adherent

wild-type GD25 cells were more sensitive to apoptotic

activity of echistatin with respect to GD25 h1A cells

(Fig. 2, panel A).

The time-course of echistatin-induced apoptotic response

confirmed the observation that the disintegrin is more active

on fibronectin-adherent GD25 cells than on vitronectin-

adherent ones (Fig. 2, panel B).

Interestingly, a significant increase in the fluorescence

which is strictly correlated to the activation of caspase-3 was

observed at echistatin doses lower than those causing cell

detachment and at an earlier time than that required to

induce cell detachment.

3.3. Echistatin activity on tyrosine phosphorylation and

functional association of focal adhesion associated-proteins

The ability of echistatin to modulate tyrosine phosphory-

lation levels of the focal adhesion-associated proteins FAK

and Shc was evaluated. A reduction of tyrosine phosphory-

Fig. 3. Echistatin activity on tyrosine phosphorylation of pp125FAK in adherent cells. Wild-type GD25 (lanes 1 and 2) and GD25 h1A cells (lanes 3 and 4) were

allowed to adhere to immobilized fibronectin (panel A) or vitronectin (panel B) for 3 h and then exposed to 20 Ag/ml of echistatin for 1 h (upper blots, lanes 2

and 4). Control cells were treated with fresh culture medium for 1 h (upper blots, lanes 1 and 3). Cell lysates were immunoprecipitated with an anti-FAK Ig and

subsequently separated by electrophoresis. After Western blotting, tyrosine phosphorylated proteins were visualized by an anti-phosphotyrosine Ig. The same

membranes used in the upper blots of panels A and B were stripped and reprobed with an anti-FAK Ig to ensure that the same amount of pp125FAK was

immunoprecipitated from all samples (lower blots). Molecular mass markers are indicated on the left. The position of pp125FAK is indicated by the arrows on

the left. Similar results were obtained from three separate experiments of identical design.

E. Alimenti et al. / Biochimica et Biophysica Acta 1693 (2004) 73–80 77

lation levels of pp125FAK was observed in protein lysates

from wild-type GD25 and GD25 h1A cells cultured either

on immobilized fibronectin or vitronectin and exposed to

echistatin (Fig. 3, panels A and B, upper blots). To confirm

equal loading of pp125FAK in all lanes, the blots were

stripped and reprobed with an anti-FAK antibody (Fig. 3,

Fig. 4. Echistatin activity on tyrosine phosphorylation of Shc in adherent cells. W

allowed to adhere to immobilized fibronectin (panel A) or vitronectin (panel B) for

and 4). Control cells were treated with fresh culture medium for 1 h (upper blots, la

subsequently separated by electrophoresis. After Western blotting, tyrosine phosph

membranes used in upper blots of panels A and B were stripped and repro

immunoprecipitated from all samples (lower blots). Molecular mass markers are in

on the left. Similar results were obtained from three separate experiments of iden

panels A and B, lower blots). The cells were exposed to the

same dose of echistatin for 3 h (data not shown); however,

the results obtained are consistent with the ones obtained

after 1 h of cell treatment.

Next, the effect of echistatin treatment of adherent wild-

type GD25 and GD25 h1A cells on Shc tyrosine phosphory-

ild-type GD25 (lanes 1 and 2) and GD25 h1A cell (lanes 3 and 4) were

3 h and then exposed to 20 Ag/ml of echistatin for 1 h (upper blots, lanes 2

nes 1 and 3). Cell lysates were immunoprecipitated with an anti-Shc Ig, and

orylated proteins were visualized by an anti-phosphotyrosine Ig. The same

bed with an anti-Shc Ig to ensure that the same amount of Shc was

dicated on the left. The position of Shc isoforms are indicated by the arrows

tical design.

E. Alimenti et al. / Biochimica et Biophysica Acta 1693 (2004) 73–8078

lation levels was investigated. Only the p52 isoform of Shc

appears to be phosphorylated in wild-type GD25 and GD25

h1A cells adherent to immobilized substrata. A decrease of

tyrosine phosphorylation of p52 Shc isoform was only

observed in vitronectin-adherent GD25 h1A cells exposed

to echistatin (Fig. 4, panel B, upper blot), whereas the

disintegrin did not modify the tyrosine phosphorylation

levels of Shc protein of both fibronectin- and in vitronec-

tin-adherent wild-type GD25 and GD25 h1A cells (Fig. 4,

panels A and B, upper blots). In order to evaluate whether

the Shc loading was equal in all lanes, the blots were

stripped and reprobed with an anti-Shc antibody (Fig. 4,

panels A and B, lower blots).

Finally, the action of echistatin on the functional associ-

ation of Shc with Grb2 was evaluated. Shc immune com-

plexes prepared from lysates of substrata-adherent wild-type

GD25 and GD25 h1A cells were probed with anti-Grb2 Ig

in order to determine the presence of co-precipitating Grb2.

The association of Grb2 with Shc was observed in wild-type

GD25 and GD25 h1A cells held in suspension (data not

shown) or adherent to both substrata. The addition of

echistatin to vitronectin- and fibronectin-adherent GD25

cells did not cause a reduction of the amount of Grb2 co-

precipitating with Shc in both wild-type GD25 and GD25

h1A cells (Fig. 5, panel A and B, upper blots). The same

blots were stripped and reprobed with an anti-Shc antibody

to ensure that the loading of the protein was equal in all

lanes (Fig. 5, panels A and B, lower blots).

Fig. 5. Echistatin activity on functional association of Shc with Grb2 in adherent ce

allowed to adhere to immobilized fibronectin (panel A) or vitronectin (panel B) for

and 4). Control cells were treated with fresh culture medium for 1 h (upper blots, la

subsequently separated by electrophoresis. After Western blotting, the association o

in upper blots of panels A and B were stripped and reprobed with an anti-Shc Ig

samples (lower blots). Molecular mass markers are indicated on the left. The positi

from three separate experiments of identical design.

4. Discussion

The h1 integrin subunit is expressed in most mammalian

cells except mature erythrocytes [22]. Three splice variants

of the protein have been described and some cells express

more than one variant at the same time [23]. The different

forms of the protein have specific functions, as indicated by

the inability of h1B isoform to promote cell migration, to

activate pp125FAK or to localize to focal contacts [20].

Echistatin, a 49 amino acid disintegrin found in the

venom of Echis carinatus, is a potent inhibitor of ligand

binding to aIIbh3, avh3 and a5h1 [24]. The disintegrin uses

different structural motifs, including the RGD loop and C

terminus, to selectively interact with a5h1 and avh3 recep-

tors [25]. Here, we show that echistatin is able to detach

GD25 cells from either immobilized fibronectin or vitro-

nectin. The activity of disintegrin was specific, not due to

toxicity, and was dependent on the dose and the time of cell

exposure. In particular, echistatin detached the GD25 h1A

cells from fibronectin at a greater extent than wild-type

GD25 cells. Conversely, the disintegrin detached the wild-

type GD25 cells from vitronectin at a greater extent than

GD25 h1A cells (Fig. 1). Our results suggest that echistatin

binds with higher affinity the integrin a5h1 than integrin

avh3; this finding is in agreement with the observation of a

lower binding affinity of echistatin for avh3 and of the

dissociable nature of the disintegrin interaction with avh3

receptor [26].

lls. Wild-type GD25 (lanes 1 and 2) and GD25 h1A cell (lanes 3 and 4) were

3 h and then exposed to 20 Ag/ml of echistatin for 1 h (upper blots, lanes 2

nes 1 and 3). Cell lysates were immunoprecipitated with an anti-Shc Ig, and

f the proteins was visualized by an anti-Grb2 Ig. The same membranes used

to ensure that the same amount of Shc was immunoprecipitated from all

on Grb2 is indicated by the arrows on the left. Similar results were obtained

E. Alimenti et al. / Biochimica et Biophysica Acta 1693 (2004) 73–80 79

Echistatin was able to promote apoptosis in both types of

GD25 cells cultured on immobilized substrata. The effect of

echistatin was more pronounced on GD25 h1A cells than on

wild-type GD25 cells adhering to fibronectin and more

pronounced on wild-type GD25 cells than on GD25 h1A

cells adhering to vitronectin (Fig. 2). Furthermore, the

apoptotic response induced by echistatin in the GD25 cells

occurred prior to cell detachment. Cell substratum detach-

ment usually induces an apoptotic response (anoikis); how-

ever, echistatin binding to avh3 integrin has been shown to

induce cell death resulting in an apoptotic signal with

different kinetics than the apoptotic signal induced by

matrix detachment [13]. This apoptotic inducing effect of

echistatin may be due to the ability of disintegrin to cause

integrin-mediated death (IMD) [23]. This form of apoptosis

is biologically and biochemically distinct from anoikis,

which is due to cell detachment from substratum. Herein,

IMD is defined as the apoptosis of adherent cells induced by

the recruitment of caspase-8 to the h subunit tails of

unligated integrins. This is a cellular process whereby

adherent cells that find themselves within an inappropriate

extracellular matrix (into the tissues), or those exposed to

integrin antagonists would be actively cued to undergo

apoptosis. Interestingly, while h1 and h3 integrins can

promote this process, h5 does not [27]. Echistatin may

cause IMD on GD25 cells adhering to vitronectin and

fibronectin by binding avh3 and a5h1 integrins.

Because of the complexity of signaling pathways in-

volved in apoptotic response by chemicals and because of

the cell-type specificity of signaling activation that leads to

apoptosis, whether FAK and Shc pathways are involved in

the apoptotic signals induced by echistatin was investigated.

Thus, the ability of echistatin to modulate tyrosine phos-

phorylation levels of the two proteins was evaluated.

Echistatin treatment caused a reduction of tyrosine phos-

phorylation of pp125FAK in protein lysates from GD25 cells

immobilized either on fibronectin or vitronectin (Fig. 3). No

differences were observed in the signaling pathway activat-

ed by echistatin treatment of either wild-type GD25 cells or

transfected-ones. However, integrin signaling independent

of ligand recognition has been described [27].

FAK is a crucial molecule in integrating signals from

integrins and receptor tyrosine kinases in processes such as

cell survival, proliferation and motility [28]. Herein, FAK

phosphorylation levels are closely associated with the sta-

bility of focal adhesions and cell–matrix interactions. Dur-

ing cell adhesion process, turnover of focal adhesions

requires activation of FAK and its functional interactions

with several proteins involved in integrin signaling [29]. For

example, FAK-deficient cells spread more slowly on extra-

cellular matrix proteins exhibiting an increased number of

prominent focal adhesions; these cells migrate poorly in

response to chemotactic signals [30]. Furthermore, cells

lacking paxillin expression, Cas expression or Src family

kinase expression all exhibit defects in focal adhesion

turnover [31]. The disassembly of focal adhesions and the

subsequent GD25 cell detachment from the substratum

induced by echistatin seem to be later events than focal

adhesion protein dephosphorylation. Although disruption of

focal adhesion signaling through FAK clearly precedes cell

detachment after echistatin induced apoptosis, our data also

suggest that caspase-3 activation is not responsible for the

former. Our results suggest that the dephosphorylation of

FAK may contribute to the loss of focal adhesions and cell

detachment.

Using human breast cancer cell lines, Xu et al. [32]

showed that FAK signaling pathway is involved in inhibit-

ing death receptor or death receptor-related apoptosis. The

overexpression in the cells of the carboxyl-terminal domain

of FAK, that is without kinase activity, induced apoptosis

independently of cell adhesion, but required Fas-associating

protein with death domain (FADD), caspase-8 and caspase-

3 [32].

The adapter protein Shc has also been linked to cyto-

skeletal organization of cells under different contexts [33].

Some reports suggest distinct roles for Shc and FAK in focal

adhesions and cell motility [34,35]. However, how Shc

plays a role in the cytoskeletal reorganization in different

cell types is not well understood [36]. GD25 cell exposure

to echistatin does not appear to affect either tyrosine

phosphorylation levels of Shc isoforms (Fig. 4) or the

functional association Shc–Grb2 (Fig. 5).

Our results indicate that FAK pathway, but not Shc

pathway is involved in the pro-apoptotic signals induced

by echistatin in GD25 cell system, during integrin-mediated

death process. However, further studies are needed in order

to identify FAK downstream molecules involved in the

echistatin-induced apoptotic response. These studies aimed

at a better understanding of the molecular mechanism of

action of disintegrins may open the way to therapeutic

targets for human diseases which require a modulation of

integrin function.

Acknowledgements

This work was supported by a grant from Polo delle

Scienze e Tecnologie per la Vita, Universita degli Studi di

Napoli Federico II, Napoli, Italy.

References

[1] R.J. Gould, M.A. Polokoff, P.A. Friedma, T.F. Huang, J.C. Halt, J.J.

Cook, S. Niewiarowski, Disintegrins: a family of integrin inhibitory

proteins from viper venoms, Proc. Soc. Exp. Biol. Med. 195 (1990)

168–171.

[2] M.A. McLane, C. MarcinKiewicz, S. Vijay-Kumar, I. Wierzbieka-

Patynowsky, S. Niewiarowski, Viper venom disintegrins and related

molecules, Proc. Soc. Exp Biol. Med. 219 (1998) 109–119.

[3] C. Brando, C. Marcinkiewicz, B. Goldman, M.A. McLane, S.

Niewiarowski, EC3, a heterodimeric disintegrin from Echis carinatus,

inhibits human and murine alpha4 integrin and attenuates lymphocyte

infiltration of Langherhans islets in pancreas and salivary glands in

E. Alimenti et al. / Biochimica et Biophysica Acta 1693 (2004) 73–8080

nonobase diabetic mice, Biochem. Biophys. Res. Commun. 267

(2000) 413–417.

[4] J.J. Calvete, M. Jurgens, C. Marcinkiewicz, A. Romero, M. Schrader,

S. Niewiarowski, Disulphide-bond pattern and molecular modelling

of the dimeric disintegrin EMF-10, a potent and selective integrin

alpha5beta1 antagonist from Eristocophis macmahoni venom, Bio-

chem J. 345 (2000) 573–581.

[5] Y. Fujii, D. Okuda, Z. Fujimoto, T. Morita, H. Mizuno, Crystallization

and preliminary crystallographic studies of dimeric disintegrins from

the venom of two Agkistrodon snakes, Acta Crystallogr., D Biol.

Crystallogr. 58 (2002) 145–147.

[6] C.P. Blobel, J.M. White, Structure, function and evolutionary rela-

tionship of proteins containing a disintegrin domain, Curr. Opin. Cell

Biol. 4 (1992) 760–765.

[7] E.A. Clark, M. Trikha, F.S. Markland, J.S. Brugge, Structurally dis-

tinct disintegrins contortrostatin and multisquamatin differentially reg-

ulate platelet tyrosine phosphorylation, J. Biol. Chem. 269 (1994)

21940–21943.

[8] A. Scaloni, E. Di Martino, N. Miraglia, A. Pelagalli, R. Della Morte,

N. Staiano, P. Pucci, Amino acid sequence and molecular modelling

of glycoprotein IIb– IIIa and fibronectin receptor iso-antagonists from

Trimeresurus elegans venom, Biochem. J. 319 (1996) 775–782.

[9] T.F. Huang, What have snakes taught us about integrins? Cell. Mol.

Life Sci. 54 (1998) 527–540.

[10] F.S. Markland, Snake venoms and the hemostatic system, Toxicon 36

(1998) 1749–1800.

[11] T.F. Huang, C.H. Yeh, W.B. Wu, Viper venom components affecting

angiogenesis, Haemostasis 31 (2001) 192–206.

[12] C.H. Yeh, H.C. Peng, T.F. Huang, Accutin, a new disintegrin, inhibits

angiogenesis in vitro and in vivo by acting as integrin avh3 antagonist

and inducing apoptosis, Blood 92 (1998) 3268–3276.

[13] D.L. Brassard, E. Maxwell, M. Malkowski, T.L. Nagabhushan, C.C.

Kumar, L. Armstrong, Integrin avh3-mediated activation of apoptosis,

Exp. Cell Res. 251 (1999) 33–45.

[14] D. Spalletti-Cernia, C. Squillacioti, R. Della Morte, P. Laccetti, N.

Staiano, Echistatin inhibits Lewis lung carcinoma cell–matrix adhe-

sion in vitro and experimental metastasis in vivo, Int. J. Oncol. 11

(1997) 757–763.

[15] N. Staiano, C. Garbi, C. Squillacioti, S. Esposito, E. Di Martino, M.A.

Belisario, L. Nitsch, P. Di Natale, Echistatin induces decrease of

pp125FAK phosphorylation, disassembly of actin cytoskeleton and

focal adhesion, and detachment of fibronectin-adherent melanoma

cells, Eur. J. Cell Biol. 73 (1997) 298–305.

[16] R. Della Morte, C. Squillacioti, C. Garbi, P. Derkinderen, M.A.

Belisario, J. Girault, P. Di Natale, L. Nitsch, N. Staiano, Echistatin

inhibits pp125FAK autophosphorylation, paxillin phosphorylation and

pp125FAK-paxillin interaction in fibronectin-adherent melanoma

cells, Eur. J. Biochem. 267 (2000) 5047–5054.

[17] A. Scibelli, S. Tafuri, M.C. Ferrante, E. Alimenti, B. Naso, A.

Lucisano, N. Staiano, R. Della Morte, Ochratoxin A affects COS

cell adhesion and signaling, Toxicol. Appl. Pharmacol. 192 (2003)

222–230.

[18] K. Wennerberg, L. Lohikangas, D. Gullberg, M. Pfaff, S. Johansson,

R. Fassler, h1 Integrin-dependent and -independent polymerization of

fibronectin, J. Cell Biol. 132 (1996) 227–238.

[19] A. Scibelli, S. Tafuri, E. Alimenti, M.A. Belisario, R. Della Morte, N.

Staiano, Biochemical analysis of the molecular mechanism of action

of natural integrin inhibitors, Vet. Res. Commun. 27 (2003) 627–630.

[20] S.F. Retta, F. Balzac, P. Ferrarsi, A.M. Belkin, R. Fassler, M.J.

Humphries, G. De Leo, L. Silengo, G. Tarone, Beta1-integrin cyto-

plasmic subdomains involved in dominant negative function, Mol.

Biol. Cell 9 (1998) 715–731.

[21] T. Patel, G.J. Gores, S.H. Kaufmann, The role of proteases during

apoptosis, FASEB J. 10 (1996) 587–597.

[22] R.O. Hynes, Integrins: bidirectional, allosteric signaling machines,

Cell 110 (2002) 673–687.

[23] S.F. Retta, G. Cassara, M. D’Amato, R. Alessandro, M. Pellegrino, S.

Degani, G. De Leo, L. Silengo, G. Tarone, Cross talk between beta(1)

and alpha(v) integrins: beta(1) affects beta(3) mRNA stability, Mol.

Biol. Cell 12 (2001) 3126–3138.

[24] C. Marcinkiewicz, S. Vijay-Kumar, M.A. McLane, S. Niewiarowski,

Significance of RGD loop and C-terminal domain of echistatin for

recognition of aIIbh3 and avh3 integrins and expression of ligand-

induced binding site, Blood 90 (1997) 1565–1575.

[25] I. Wierzbicka-Patynowski, S. Niewiarowski, C. Marcinkiewicz, J.J.

Calvete, M.M. Marcinkiewicz, M.A. McLane, Structural require-

ments of echistatin for the recognition of avh3 and a5h11 integrins,

J. Biol. Chem. 274 (1999) 37809–37814.

[26] C.C. Kumar, H. Nie, L. Armstrong, R. Zhang, S. Vijay-Kumar, A.

Tsarbopoulos, Chloramine T-induced structural and biochemical

changes in echistatin, FEBS Lett. 429 (1998) 239–248.

[27] D.G. Stupack, X.S. Puente, S. Boutsaboualoy, C.M. Storgard, D.A.

Cheresh, Apoptosis of adherent cells by recruitment of caspase-8 to

unligated integrins, J. Cell Biol. 155 (2001) 459–470.

[28] M.D. Schaller, Biochemical signals and biological responses elicited by

the focal adhesion kinase, Biochim. Biophys. Acta 1540 (2001) 1–21.

[29] D.J. Webb, J.T. Parsons, A.F. Horwitz, Adhesion assembly, disassem-

bly and turnover in migrating cells—over and over and over again,

Nat. Cell Biol. 4 (2002) E97–E100.

[30] D.J. Sieg, C.R. Hauck, D. Ilic, C.K. Klingbeil, E. Schaefer, C.H.

Damsky, D.D. Schlaepfer, FAK integrates growth-factor and integrin

signals to promote cell migration, Nat. Cell Biol. 2 (2000) 249–256.

[31] R.A. Kinghoffer, C. Sachsenmaier, J.A. Cooper, P. Soriano, Src family

kinases are required for integrin but not PDGFR signal transduction,

EMBO J. 18 (1999) 2459–2471.

[32] L.H. Xu, X. Yang, C.A. Bradham, D.A. Brenner, A.S. Baldwin, R.J.

Craven, W.G. Cance, The focal adhesion kinase suppresses transfor-

mation-associated, anchorage-independent apoptosis in human breast

cancer cells. Involvement of death receptor-related signaling path-

ways, J. Biol. Chem. 275 (2000) 30597–30604.

[33] K.S. Ravichandran, Signaling via Shc family adapter proteins, Onco-

gene 20 (2001) 6322–6330.

[34] J. Gu, M. Tamura, R. Pankov, E. Danen, T. Takino, K. Matsumoto, K.

Yamada, Shc and FAK differentially regulate cell motility and direc-

tionality modulated by PTEN, J. Cell Biol. 146 (1999) 389–403.

[35] L. Barberis, K.K. Wary, G. Fiucci, F. Liu, E. Hirsch, M. Brancaccio,

F. Altruda, G. Tarone, F.G. Giancotti, Distinct roles of the adaptor

protein Shc and focal adhesion kinase in integrin signaling to ERK,

J. Biol. Chem. 275 (2000) 36532–36540.

[36] K.M. Lai, T. Pawson, The Shc A phosphotyrosine docking protein

sensitizes cardiovascular signaling in the mouse embryo, Genes Dev.

14 (2000) 1132–1145.