Three-dimensional matrix induces sustained activation of ERK1/2 via Src/Ras/Raf signaling pathway

Cell Biology International 32 (2008) 229e234www.elsevier.com/locate/cellbi

Three-dimensional matrix induces sustained activation ofERK1/2 via Src/Ras/Raf signaling pathway

Ralica Damianova a, Nadezhda Stefanova a, Edna Cukierman b,Albena Momchilova c, Roumen Pankov a,*

a Department of Cytology, Histology and Embryology, Biological Faculty, Sofia University ‘‘St. Kliment Ohridski’’, 1164 Sofia, Bulgariab Division of Basic Science, Tumor Cell Biology, Fox Chase Cancer Center, Philadelphia, PA 19111, USA

c Institute of Biophysics, Bulgarian Academy of Sciences, 1113 Sofia, Bulgaria

Received 23 June 2007; revised 13 July 2007; accepted 27 August 2007

Abstract

Research in cell signaling often depends on tissue culture, but the artificial substrates used to grow cells in vitro are likely to distort the con-clusions, particularly when adhesion-mediated signaling events are investigated. Studies of signal transduction pathways operating in cellsgrown in three-dimensional (3D) matrices provide a better system, giving a closer insight of the cell signaling in vivo. We compared thesteady-state levels of ERK1/2 activity in primary human fibroblasts, induced by cell-derived 3D fibronectin matrix or fibronectin, coated onflat surfaces. 3D environment caused ERK1/2 stimulation concomitant with a 2.5-fold increase in Ras GTP loading and Src activation. Underthese conditions FAK autophosphorylation was suppressed. Treatment with Src inhibitor PP2 abolished these effects indicating that 3D fibro-nectin matrix activated ERK1/2 through Src/Ras/Raf pathway, bypassing FAK. These observations suggest that within in vivo-like conditions Srcmay have a leading role in the induction of sustained ERK1/2 activation.� 2007 International Federation for Cell Biology. Published by Elsevier Ltd. All rights reserved.

Keywords: Three-dimensional matrix; Extracellular matrix; Fibronectin; Src; Ras; ERK

1. Introduction

Our current knowledge about the signaling pathways isbased mainly on studies carried out with artificially polarizedcells cultured in vitro. Under these conditions cells are forcedto adjust to unnaturally flat and rigid plastic or glass surfacesthat greatly differ from the natural extracellular matrix (ECM),surrounding most cells in living organisms. As the activity ofthe cellular signaling pathways is closely dependent on environ-mental cues, the use of more physiologically relevant experi-mental conditions, like cell-derived 3D ECM, may providea better insight for the signal transduction pathways in vivo.

The use of 3D pure collagen gels and Matrigel matrices haveproven to be a valuable model system for studying the role of

* Corresponding author: Tel.: þ359 2 816 7274.

E-mail address: [email protected] (R. Pankov).

1065-6995/$ - see front matter � 2007 International Federation for Cell Biology.

doi:10.1016/j.cellbi.2007.08.029

mechanical interactions between cells and ECMs in signaling(Grinnell, 2000), cell motility (Friedl et al., 2004; Pizzo et al.,2005), and cancer progression (Bissell and Radisky, 2001;Kleinman and Martin, 2005). However, the major drawbackof these gel-based experimental models is that they are not com-pletely synthesized and arranged by the cells themselves. Thesedisadvantages can be avoided if the matrices are produced com-pletely by cells grown in vitro. Several research groups havedemonstrated that given a permissive environment, fibroblastscan be induced to produce in vitro a three-dimensional structureof ECM material over time (Cukierman et al., 2001; Ahlfors andBilliar, 2007). These cell-derived matrices appear to approxi-mate more closely native tissue rather than the ‘‘reconstituted’’matrices described above and present a unique possibility for invitro studies of cell behavior under in vivo-like conditions.

We have shown that fibroblasts, grown within in vivo-like3D fibronectin-rich matrices, organize distinctive 3D-matrix

Published by Elsevier Ltd. All rights reserved.

230 R. Damianova et al. / Cell Biology International 32 (2008) 229e234

adhesion structures (Cukierman et al., 2001, 2002; Yamadaet al., 2003) that are more biologically relevant to living organ-isms than classical focal and fibrillar adhesions (Geiger et al.,2001), formed by cells plated on glass or plastic surfaces.Activated integrin receptors in these 3D-matrix adhesionsmediate increased proliferation, migration and attachment(Cukierman et al., 2001), suggesting a different type of signaltransduction within the in vivo-like matrices. Although thedata are still sparse, comparisons between signaling eventswithin 3D matrices (3D signaling) and signal transductiononto two-dimensional coated surfaces (2D signaling) showsome unexpected differences. For example, in 3D cultures,the steady-state level of activity of extracellular signal-regu-lated kinases (ERK1/2) is enhanced when compared withERK1/2 activation level, sustained by 2D signaling. Interest-ingly, the activity of focal adhesion kinase (FAK) which medi-ates integrin activation of ERK1/2 (Zhao et al., 1998) issuppressed by 3D environment (Cukierman et al., 2001; Ishiiet al., 2001).

Since the duration and the level of ERK1/2 signaling con-trols events such as cell proliferation, differentiation or death,it is important to identify its key upstream regulators operatingin 3D signaling. To gain further insight on the pathways lead-ing to activation of ERK1/2 in in vivo-like conditions wetested the steady-state activity of several upstream ERK1/2activators in a model system employing primary human fibro-blasts (HFFs), grown within cell-derived 3D fibronectin matri-ces. Cells grown onto fibronectin-coated dishes were used as2D signaling controls. We found that a sustained activationof Src/Ras pathway is responsible for the observed enhancedERK1/2 activity in in vivo-like 3D environment.

2. Materials and methods

2.1. Cell culture and preparation of three-dimensionalextracellular matrix

Primary human foreskin fibroblasts (HFF) were a gift from S. Yamada

(NIDCR, NIH) and were used at passages 9e18. NIH 3T3 cells were pur-

chased from ATCC. All cell lines were cultured in Dulbecco’s modified Eagle

medium (DMEM) containing 10% fetal bovine serum, 100 U/ml penicillin,

and 100 mg/ml streptomycin.

Cell-derived 3D matrices were prepared as described by Cukierman

(2002). Briefly, 3T3 cells were cultured on coverslips or culture dishes in

DMEM supplemented with 10% fetal bovine serum. After w8 days, cells

were removed from the deposited matrix by lysis with PBS containing 0.5%

Triton X-100, 20 mM NH4OH at 37 �C. The matrices were washed with

PBS and kept at 4 �C in PBS/antibiotics until use.

2.2. Reagents and antibodies

Antibodies to phospho-FAK (Tyr397) and anti-phospho-Paxillin (Tyr31)

were purchased from BioSource; anti-phospho-ERK1/2 (Thr202/Tyr204)

from Cell Signaling; anti-Ras, anti-phospho-Raf-1 (Ser338) and anti-phospho-

Src (Tyr416) from Upstate; and anti-actin antibody from Sigma. Phalloidin con-

jugated with Alexa 594 were obtained from Molecular Probes and anti-human

FN antibody, directly labeled with FITC was prepared as described by Akiyama

et al., 1989. Ras activation kit was acquired from Upstate. Src inhibitor PP2 was

purchased from Calbiochem.

2.3. Western immunoblotting and pull-down assay

Cells were lysed using RIPA buffer (50 mM HEPES, pH 7.5, 150 mM

NaCl, 10% glycerol, 1.5 mM MgCl2, 1 mM EGTA, 1 mM sodium vanadate,

10 mM sodium pyrophosphate, 100 mM NaF, 1% Triton X-100, 1% sodium

deoxycholate, 0.1% SDS, protease inhibitor mixture (Boehringer), and

1 mM phenylmethylsulfonyl fluoride). Protein samples were mixed with equal

volume of 2� reducing SDSePAGE sample buffer and resolved on 4e12%

gradient gels (Novex). After electrotransfer to nitrocellulose membranes the

filters were blocked with 5% non-fat dry milk in TBST (150 mM NaCl,

50 mM Tris HCl, 0.1% Tween 20, pH 7.4) and probed with primary antibodies

followed by the appropriate secondary horseradish peroxidase-conjugated an-

tibodies. Immunoblots were visualized using the ECL system and Hyperfilm

X-ray film (Amersham Biosciences).

Pull-down experiments were performed with Ras activation kit as sug-

gested by the manufacturer. Briefly, cells were scraped into ice-cold lysis

buffer (25 mM HEPES, pH 7.5, 150 mM NaCl, 10 mM MgCl2, 1 mM

EDTA, 1% Igepal, 10% glycerol, and complete protease cocktail without

EDTA) and centrifuged for 5 min at 14,000 � g. Cleared lysates were incu-

bated with 10 mg Raf1 RBD agarose for 45 min at 4 �C with rotation. The

beads were washed with lysis buffer, heated for 3e5 min at 95 �C in reducing

SDSePAGE sample buffer, and the released proteins were subjected to SDSe

PAGE and immunoblotting.

2.4. Microscopy

Cells for immune-fluorescence analysis were plated on fibronectin (5 mg/

ml) or 3D matrix pre-coated glass coverslips (12 mm, Carolina Biological

Supply Co.) and cultured overnight. Samples were fixed with 4% paraformal-

dehyde in PBS containing 5% sucrose for 20 min and permeabilized with 0.5%

Triton X-100 in PBS for 3 min. Samples were stained with phalloidin conju-

gated with Alexa 594 and anti-human FN antibodies directly labeled with

FITC. Stained samples were mounted in GEL/MOUNT� (Biomeda Corp.)

containing 1 mg/ml 1,4-phenylendiamine (Fluka) to reduce photobleaching.

Immunofluorescent images were obtained with a Zeiss Axiophot microscope

equipped with a CCD camera.

3. Results

3.1. Three-dimensional matrix promoted changes in bothcell morphology and the steady-state levels of FAK andERK1/2 phosphorylation

To avoid the artifacts caused by flat and rigid surfaces ofthe conventional culture dishes, primary human fibroblastswere plated on cell-derived 3D fibronectin-rich matrices. Afterculturing overnight cells successfully entered the matrix andwere completely surrounded by matrix fibrils, losing the dor-soventral polarization typical for fibroblasts grown on glassor plastic surfaces (Fig. 1A). The cell density was kept lowto avoid interactions between neighboring cells. As noted pre-viously (Cukierman et al., 2001; Pankov et al., 2005), cellswithin this in vivo-like 3D environment acquired elongatedshape and had reduced lamella, confined predominantly toone of the cell ends. In contrast, control fibroblasts grownon fibronectin-coated surfaces had a flattened and triangularshape with increased number of peripheral lamellae (Fig 1A,compare FN and matrix). Staining of the activated b1 integrinreceptors revealed formation of typical for the matrix environ-ment 3D matrix adhesions, while fibronectin-coated surfacesinduced arrangement of these receptors in the characteristicfocal adhesions for 2D surfaces (Fig. 1B).

Fig. 1. Three-dimensional matrix induces changes in morphology and signaling in human fibroblasts. (A) Cells onto 2D versus within 3D environments have

different morphologies as shown after staining for F-actin (left panels and red color in the right panels) and fibronectin (green). Primary HFFs were plated for

overnight onto fibronectin or within 3D matrix pre-coated glass coverslips. Cells were fixed and stained with phalloidin conjugated with Alexa 594 and anti-human

FN antibodies directly labeled with FITC. Bar: 20 mm. (B) Cells were prepared as described in ‘‘A’’, fixed and stained with anti-b1 integrin (mAb 9EG7) for

visualization of adhesive contacts. While cells plated on fibronectin-coated coverslips (FN) organized typical focal adhesions (arrowhead), fibroblasts within

3D matrix formed thin and elongated 3D matrix adhesions (arrow). Bar: 20 mm. (C) Total lysates from HFFs, grown as described in ‘‘A’’ were analyzed using

the indicated phosphospecific (anti-pFAK [Tyr397], anti-pPaxillin [Tyr31] and anti-pERK1/2 [Thr202/Tyr204]) or general (anti-actin) antibodies.

231R. Damianova et al. / Cell Biology International 32 (2008) 229e234

The observed morphological changes occurred concomi-tantly with alterations in the activity of integrin-associatedfocal adhesion kinase. Western blot analysis with phosphospe-cific antibodies demonstrated down-regulation in the steady-state phosphorylation level of FAK on Tyr397 when cellswere cultured within 3D matrix overnight (Fig. 1C, p-FAK,compare FN with Matr.) or 24 h (not shown), confirming ourprevious results (Cukierman et al., 2001). Since integrin-medi-ated autophosphorylation at Tyr397 correlates with the catalyticactivity of FAK (Calalb et al., 1995) and leads to recruitment ofSrc and subsequent downstream activation of extracellularsignal-regulated kinases (Parsons, 2003), we tested the effectof 3D matrix on the phosphorylation level of ERK1/2. Thesteady-state level of ERK1/2 phosphorylation at Thr202/Tyr204 was considerably enhanced in cells cultured within3D matrix when compared to the control fibroblasts, grownon fibronectin-coated dishes (Fig. 1B, p-ERK1/2). This wassurprising in view of the fact that these kinases are a downstreamtarget of fibronectin and integrin-stimulated FAK signalingpathway (Schlaepfer and Hunter, 1997) which was inhibited.At the same time, the phosphorylation of another FAK sub-strate, the adaptor protein paxillin (Brown and Turner, 2004),was not significantly influenced (Fig. 1B, p-Pax) suggestingthat 3D signaling differs from the well studied signaling cas-cades operating in 2D signaling.

3.2. Three-dimensional matrix activated small GTPaseRas

These results prompted us to test the activation of the up-stream positive regulator of ERKdthe small GTP-bindingprotein Ras (Marshall 1995). Ras is activated downstream ofa variety of transmembrane receptors alternating between

GTP- and GDP-bound conformations, where the GTP-boundconformation represents the active state. Pull-down assay, per-formed with immobilized Raf-1 Ras-binding domain (Zhanget al., 1993) demonstrated a more than 2.5-fold increase(2.69 � 0.21, n ¼ 3) in the steady-state level of GTP-boundRas in cells grown in 3D matrices when compared with cellscultured on fibronectin-coated dishes (Fig. 2). This change re-flected only differences in the activation, since the total levelsof Ras protein in the total lysates, prepared from both samples,were similar (Fig. 2, total Ras).

Activation of Ras was further confirmed by increased phos-phorylation of the serine/threonine kinase Raf, which func-tions downstream of Ras and upstream of ERK1/2 (Avruchet al., 1994). In unstimulated cells, Raf is primarily localizedin the cytosol, which upon Ras activation is recruited to theplasma membrane followed by conformational changes, phos-phorylation and hetero-oligomerization leading to its full acti-vation (Morrison and Cutler, 1997). Phosphorylation of Raf atSer338, which is considered a key adhesion-dependent regula-tory step for efficient activation of ERK (Edin and Juliano,2005), was enhanced in cells grown within 3D matrix(Fig. 2, p-Raf).

3.3. Sustained activation of ERK1/2 and Ras wasmediated by matrix-induced Src activation

Studies have shown that there are multiple and potentiallyparallel integrin-stimulated signaling pathways upstream ofRas that depend upon the activity of tyrosine kinases Srcand FAK (Schlaepfer et al., 1998). Given that FAK was sup-pressed, we examined the possibility for Ras activationthrough Src. We probed the activation of Src by testing thephosphorylation level of Tyr 416 since the addition of

Fig. 2. Three-dimensional matrix activates Ras. Primary fibroblasts were cul-

tured onto fibronectin (FN), or within 3D matrices (Matr) for overnight and

used in pull-down experiments for determination of Ras-GTP. The densitom-

etry signals from the GTP bound forms were normalized to the corresponding

densitometry signals from the lysates. The differences are presented as fold

change compared to cells plated on fibronectin-coated dishes. Error bar indi-

cates SD. The lysates were also probed with anti-pRaf 1[Ser338] antibody

to verify the effect of the activated Ras on this downstream target. Actin

was used as an internal control for protein loading.Fig. 3. Activated Src stimulates Ras and ERK1/2 in cells grown in 3D matri-

ces. (A) HFFs were plated onto fibronectin-coated dishes (FN) or within 3D

fibronectin matrices (Matr) overnight. Lysates were probed with anti-pSrc

[Tyr416] and anti-actin for protein loading control. The densitometry signals

from the active p-Src [Tyr416] were normalized to the above-mentioned load-

ing control. Differences are presented as fold change compared to cells plated

on fibronectin-coated dishes. Error bar indicates S.D. (B) HFFs were plated on

three-dimensional matrices, cultured for 12 h and than incubated in medium

containing 20 mM Src inhibitor PP2 (þ) or medium with equal volume

DMSO (�) for an additional hour. After the treatment cells were used for

pull-down assay for activated Ras, performed as described in Fig. 2. The

same lysates were used to probe the phosphorylation level of phospho-Src

[Tyr416] and phospho-ERK1/2[Thr202/Tyr204].

232 R. Damianova et al. / Cell Biology International 32 (2008) 229e234

phosphate to this residue stimulates complete activation of Srcand provides a binding site for SH2 domains of other cellularproteins (Roskoski, 2005). Western blotting with phosphospe-cific antibodies demonstrated a 2.5-fold increase (2.48 � 0.15,n ¼ 3) of the phosphorylated forms of Src within 3D fibronec-tin matrix when compared to Src, present in lysates from thesame cells cultured on fibronectin-coated plates (Fig. 3A).This result supported our initial assumption that Src may par-ticipate in the activation of Ras and ERK1/2 in cells grownwithin 3D environment.

To test this possibility further we used PP2, a potent and se-lective small molecule inhibitor of the Src family of proteintyrosine kinases (Hanke et al., 1996) that has been extensivelyused to examine the role of these kinases in signaling. Treat-ment of HFFs with PP2, plated within 3D matrices, led tosuppression of Src and a strong reduction of the amount ofGTP-bound Ras and the phosphorylated forms of ERK1/2(Fig. 3B). Based on these data, it appears that Src activationwithin the 3D environment is necessary for the sustained con-stitutive activation of ERK1/2 and that Ras is an intermediatestep in this signaling pathway.

4. Discussion

In this study we explored the steady-state level of activationof several key signaling molecules in primary human fibro-blasts grown under two different conditions: (1) within invivo-like three-dimensional matrix and (2) conventional mono-layer culture. The obtained results demonstrated that the 3Dmatrix induced more than a twofold increase in the steady-state level of activation of the tyrosine kinase Src, whilst theother integrin-signaling partner (the tyrosine kinase FAK)was suppressed. At the same time, the sustained levels of ac-tivation of the members of the Ras/Raf/ERK pathway werealso elevated, suggesting that at least some of the 3D matrix

stimulatory signals were routed through Src kinase. The Src-mediated stimulation of this pathway was confirmed aftertreatment of cells with the specific Src inhibitor PP2, whichalso lowered the steady-state levels of Ras and ERK1/2.

The observed differences in 3D signaling were most prob-ably due to the three-dimensionality of the matrix, rather thanto its composition. Our previous studies have demonstratedthat physically flattening the matrix or solubilizing the matrixand spreading its components on a flat substrate are sufficientto abolish the formation of 3D matrix adhesions and ERK-stimulated cell proliferation (Cukierman et al., 2001, 2002).Although cell-derived 3D matrices contain laminin, collagenI and heparan sulfate proteoglycan, in addition to their keyfibronectin component (our unpublished results) the above ob-servations made possible to simplify the experimental systemand compare the signaling events occurring in 3D matrix withthose taking place in cells, plated on fibronectin-coated surfaces.

The attenuation of FAK autophosphorylation at Tyr397 incells, grown in 3D matrices, although noted before (Cukier-man et al., 2001; Ishii et al., 2001), was somewhat unexpectedin view of the fact that integrin ligation and clustering at focaladhesions in monolayer cell cultures promotes recruitment andstrong phosphorylation of FAK (see Parsons, 2003; Mitra andSchlaepfer, 2006). Ligand occupied and clustered a5b1

233R. Damianova et al. / Cell Biology International 32 (2008) 229e234

integrin receptors are also characteristic for fibroblasts grownin 3D matrices, even though they are organized in a new typeof adhesiond3D matrix adhesions (Cukierman et al., 2001).The discrepancy in FAK phosphorylation can be explainedwith differences in the composition of the integrin cytoplasmiccomplexes, formed in 3D matrix adhesions and the composi-tion of complexes, found in classical focal adhesions (Geigeret al., 2001). Another possible explanation can be related tothe stiffness of the substrate. It has been shown that FAKTyr397 phosphorylation strongly depends on the rigidity ofthe substrate (Shi and Boettiger, 2003; Wozniak et al.,2003). Since cell-derived 3D matrices are pliable (Cukiermanet al., 2001, 2002) they may not be able to induce such strongFAK phosphorylation like the fibronectin-coated plastic orglass surfaces. Interestingly, similar pliability is a feature ofthe natural tissue-derived 3D matrices (Cukierman et al.,2001), which raises the possibility that the high FAK activa-tion typical for conventional monolayer cultures is to some ex-tent artificial.

Despite FAK suppression, 3D matrix stimulated the steady-state activity of the extracellular signal regulated kinases 1 and2. A similar increase of ERK1/2 phosphorylation at Thr202/Tyr204 was found 12 h (Fig. 1B) and 24 h after plating (notshown). Since sustained activation of these kinases is requiredfor cell proliferation (see Sharrocks, 2006; Meloche andPouysse, 2007) the latter result is in agreement with our pre-vious findings, demonstrating an increased proliferation rateof HFFs within the 3D fibronectin matrix (Cukierman et al.,2001). The stimulatory effect of the matrix is most probablydue to its three-dimensionality, rather than to its composition,as flattening of the matrix by mechanical force abolishes theeffect on cell proliferation (Cukierman et al., 2001).

Correspondingly, the canonical upstream members of theERK signaling cascade Ras and Raf were also activated.The strong 2.5-fold increase in the steady-state Ras GTP load-ing appears to be a specific result of the interactions of fibro-blasts with three-dimensional environment, since we haverecently documented that other members of the Ras superfa-milydRho and Cdc42dare not affected and Rac1 is evensuppressed under the same conditions (Pankov et al., 2005).

The observed elevation of the steady-state activity of Rasand ERK1/2 was found to be Src dependent, as treatmentwith PP2 abolished the stimulatory effect of the 3D matrix.Numerous examples of activation of Ras signaling throughstimulation of Src exist, utilizing as an intermediate step thephosphorylation of Shc and subsequent binding to Grb2(Rozakis-Adcock et al., 1992; Dilworth et al., 1994) orPlcg1 activation (Zhang et al., 2004). Several members ofthe Src family have also been shown to associate with andphosphorylate Ras GTPase-activating protein (Ellis et al.,1990; Brott et al., 1991) thus influencing Ras activity.

While Src mediated activation of Ras/Raf/ERK signalingpathways is documented by many studies, the upstream Srcactivators, operating in 3D cell-derived matrix, are still notclarified. A well-studied integrin signaling route involvesFAK-Src complex (see Mitra and Schlaepfer, 2006), but itdepends on FAK phosphorylation at Tyr397 which was

suppressed. Activation of Src, independently of FAK, havebeen described for a4b1 (Hisa et al., 2005) and b3 (see Shattil2005) integrin cytoplasmic tail; however in 3D matrix, theenhanced proliferation rate of the fibroblasts is mediated bya5b1integrin (Cukierman et al., 2001). As Src is also activatedby a number of receptor tyrosine kinases it is possible thattheir interaction with integrin receptors in 3D matrix adhesionsis sufficient to induce the observed sustained Src stimulation.In accordance with this possibility it has been demonstratedthat integrin engagement is sufficient to activate several growthfactor receptors (see Cabodi et al., 2004; Lee and Juliano,2004).

These possibilities still need experimental evaluation, butour study supports the notion that both FAK and c-Src can pro-mote ERK1/2 activation through more than one integrin-medi-ated signaling pathway (Schlaepfer et al., 1998; Lee andJuliano, 2004) and demonstrates that Src, not FAK, influencessteady-state activity and is up-regulated within the 3D matrixenvironment. It also indicates the existence of different bal-ances between signal transduction pathways in 3D versus 2Dsignaling, and emphasizing the need for further clarificationof cell signaling in vivo.

Acknowledgments

This work was supported by the Bulgarian National Fundfor Scientific Research (Grant BU-B-1/05 and partially grant1404/04).

References

Ahlfors J-EW, Billiar KL. Biomechanical and biochemical characteristics of

a human fibroblast- produced and remodeled matrix. Biomaterials 2007;

28:21e83.

Akiyama SK, Yamada SS, Chen WT, Yamada KM. Analysis of fibronectin

receptor function with monoclonal antibodies: roles in cell adhesion,

migration, matrix assembly, and cytoskeletal organisation. J Cell Biol

1989;109:863e75.

Avruch J, Zhang XF, Kyriakis JM. Raf meets Ras: completing the framework

of a signal transduction pathway. Trends Biochem Sci 1994;19:279e83.

Bissell MJ, Radisky D. Putting tumours in context. Nat Rev Cancer 2001;

1:46e54.

Brott BK, Decker S, O’Brien MC, Jove R. Molecular features of the viral and

cellular Src kinases involved in interactions with the GTPase-activating

protein. Mol Cell Biol 1991;11:5059e67.

Brown MC, Turner CE. Paxillin: adapting to change. Physiol Rev 2004;84:

1315e39.

Cabodi S, Moro L, Bergatto E, Boeri Erba E, Di Stefano P, Turco E, et al.

Integrin regulation of epidermal growth factor (EGF) receptor and of

EGF-dependent responses. Biochem Soc Trans. 2004;32:438e42.

Calalb MB, Polte TR, Hanks SK. Tyrosine phosphorylation of focal adhesion

kinase at sites in the catalytic domain regulates kinase activity: a role for

Src family kinases. Mol Cell Biol 1995;15:954e63.

Cukierman E. Preparation of extracellular matrices produced by cultured fibro-

blasts. Curr Protocols Cell Biol 2002;10.9.1e10.9.14.

Cukierman E, Pankov R, Stevens DR, Yamada KM. Taking cell-matrix adhe-

sions to the third dimension. Science 2001;294:1708e12.

Cukierman E, Pankov R, Yamada KM. Cell interactions with three-dimen-

sional matrices. Curr Opin Cell Biol 2002;14:633e9.

Dilworth SM, Brewster CEP, Jones MD, Lanfrancone L, Pelicci G, Pelicci PG.

Transformation by polyoma virus middle T-antigen involves the binding

and tyrosine phosphorylation of Shc. Nature 1994;367:87e90.

234 R. Damianova et al. / Cell Biology International 32 (2008) 229e234

Edin ML, Juliano RL. Raf-1 Serine 338 Phosphorylation plays a key role in

adhesion-dependent activation of extracellular signal-regulated kinase by

epidermal growth factor. Mol Cell Biol 2005;25:4466e75.

Ellis C, Moran M, McCormick F, Pawson T. Phosphorylation of GAP and

GAP-associated proteins by transforming and mitogenic tyrosine kinases.

Nature 1990;343:377e81.

Friedl P, Hegerfeldt Y, Tusch M. Collective cell migration in morphogenesis

and cancer. Int J Dev Biol 2004;48:441e9.

Geiger B, Bershadsky A, Pankov R, Yamada KM. Transmembrane crosstalk

between the extracellular matrixecytoskeleton crosstalk. Nat Rev Mol

Cell Biol 2001;2:793e805.

Grinnell F. Fibroblast-collagen-matrix contraction: growth-factor signalling

and mechanical loading. Trends Cell Biol 2000;9:362e5.

Hanke JH, Gardner JP, Dow RL, Changelian PS, Brissette WH, Weringer EJ,

et al. Discovery of a novel, potent, and Src family-selective tyrosine kinase

inhibitor. Study of Lck- and FynT-dependent T cell activation. J Biol Chem

1996;271:695e701.

Hisa DA, Lim ST, Bernard-Trifilo JA, Mitra SK, Tanaka S, den Hertog J, et al.

Integrin a4b1 promotes focal adhesion kinase-independent cell motility

via a4 cytoplasmic domain-specific activation of c-Src. Mol Cell Biol

2005;25:9700e12.

Ishii I, Tomizawa A, Kawachi H, Suzuki T, Kotani A, Koshushi I, et al. His-

tological and functional analysis of vascular smooth muscle cells in a novel

culture system with honeycomb-like structure. Atherosclerosis 2001;158:

377e84.

Kleinman HK, Martin GR. Matrigel: basement membrane matrix with biolog-

ical activity. Semin Cancer Biol. 2005;15:378e86.

Lee JW, Juliano R. Mitogenic signal transduction by integrin- and growth

factor receptor-mediated pathways. Mol Cells 2004;17:188e202.

Marshall MS. Ras target proteins in eukaryotic cells. FASEB J 1995;9:

1311e8.

Mitra SK, Schlaepfer DD. Integrin-regulated FAK-Src signaling in normal and

cancer cells. Curr Opin Cell Biol. 2006;18:516e23.

Meloche S, Pouysse J. The ERK1/2 mitogen-activated protein kinase pathway

as a master regulator of the G1- to S-phase transition. Oncogene 2007;

26:3227e39.

Morrison DK, Cutler RE. The complexity of Raf-1 regulation. Curr Opin Cell

Biol 1997;9:174e9.

Pankov R, Endo Y, Even-Ram S, Araki M, Clark K, Cukierman E, et al. A Rac

switch regulates random versus directionally persistent cell migration.

J Cell Biol 2005;170:793e802.

Parsons JT. Focal adhesion kinase: the first ten years. J Cell Sci 2003;116:

1409e16.

Pizzo AM, Kokini K, Vaughn LC, Waisner BZ, Voytik-Harbin SL. Extracellu-

lar matrix (ECM) microstructural composition regulates local cell-ECM

biomechanics and fundamental fibroblast behavior: a multidimensional

perspective. J Appl Physiol 2005;98:1909e21.

Rozakis-Adcock M, McGlade J, Mbamalu G, Pelicci G, Daly R, Li W, et al.

Association of the Shc and Grb2/Sem5 SH2-containing proteins is impli-

cated in activation of the Ras pathway by tyrosine kinases. Nature 1992;

360:689e92.

Roskoski Jr R. Src kinase regulation by phosphorylation and dephosphoryla-

tion. Biochem Biophys Res Commun 2005;331:1e14.

Schlaepfer DD, Hunter T. Focal adhesion kinase overexpression enhances

ras-dependent integrin signaling to ERK2/mitogen-activated protein kinase

through interactions with and activation of c-Src. J Biol Chem 1997;272:

13189e95.

Schlaepfer DD, Jones KC, Hunter T. Multiple Grb2-mediated integrin-

stimulated signaling pathways to ERK2/mitogen-activated protein kinase:

summation of both c-Src- and focal adhesion kinase-initiated tyrosine

phosphorylation events. Mol Cell Biol 1998;18:2571e85.

Sharrocks AD. Cell cycle: Sustained ERK signalling represses the inhibitors.

Curr Biol 2006;16:R540e2.

Shattil SJ. Integrins and Src: dynamic duo of adhesion signaling. Trends Cell

Biol 2005;15:399e403.

Shi Q, Boettiger D. A novel mode for integrin-mediated signaling: tethering is

required for phosphorylation of FAK Y397. Mol Biol Cell 2003;14:

4306e15.

Wozniak MA, Desai R, Solski PA, Der CJ, Keely PJ. ROCK-generated

contractility regulates breast epithelial cell differentiation in response to

the physical properties of a three-dimensional collagen matrix. J Cell

Biol 2003;163:583e95.

Yamada KM, Pankov R, Cukierman E. Dimensions and dynamics in integrin

function. Braz J Med Biol Res 2003;36:959e66.

Zhang SQ, Yang W, Kontaridis MI, Bivona TG, Wen G, Araki T, et al. Shp2

regulates SRC family kinase activity and Ras/Erk activation by controlling

Csk recruitment. Mol Cell 2004;13:341e55.

Zhang XF, Settleman J, Kyriakis JM, Takeuchi-Suzuki E, Elledge SJ,

Marshall MS, et al. Normal and oncogenic p21ras proteins bind to the

amino-terminal regulatory domain of c-Raf-1. Nature 1993;364:308e13.

Zhao JH, Reiske H, Guan JL. Regulation of the cell cycle by focal adhesion

kinase. J Cell Biol 1998;143:1997e2008.