CellAdhesion&Migration6:5,424–432;September/October2012;G2012LandesBioscience Occupy tissue The movement in cancer metastasis Peta Bradbury,1,2 Ben Fabry3 and Geraldine M. O’Neill1,2,* 1Children’sCancerResearchUnit;KidsResearchInstitute;TheChildren’sHospitalatWestmead;Sydney,NSWAustralia;2DisciplineofPaediatricsandChildHealth; TheUniversityofSydney;Sydney,NSWAustralia;3DepartmentofPhysics;UniversityofErlangen-Nuremberg;Erlangen,Germany Keywords: migration, invasion, metastasis, adhesion, 3D, 2D Abbreviations: 2D, 2-dimensional; 3D, 3-dimensional; ECM, extracellular matrix; CDM, cell-derived matrix Interstitial Dissemination Thecriticalroleofmigrationandinvasionincancermetastasis warrantsnewtherapeuticapproachestargetingthemachinery regulating cell migration and invasion. While 2-dimensional The ability of cancer cells to invade tissue surrounding the (2D) models have helped identify a range of adhesion primary tumor site, leading to the development of metastatic molecules, cytoskeletal components and regulators that are tumors, is one of the hallmarks of cancer.1 Disseminating tumor potentiallyimportantforcellmigration,theuseofmodelsthat cellsencounteravarietyoftissuearchitecturesthroughwhichthey bettermimicthe3-dimensional(3D)environmenthasyielded must transmigrate. While the majority of tumor cells are new insights into the physiology of cell movement. For confronted by collagen-rich connective tissue found around most example, studying cells in 3D models has revealed that organs in the body, invasion in the brain represents a specialized invading cancer cells may switch between heterogeneous form of transmigration due to the unique makeup of the brain invasionmodesandthusevadepharmacologicalinhibitionof interstitium. For more detail we direct readers to an excellent, invasion.Herewesummarizepublisheddatainwhichtherole recent review that provides a comprehensive description of tissue of cell adhesion molecules in 2D vs. 3D migration have been structures through which cancer cells transmigrate in vivo.2 The directly compared and discuss mechanisms that regulate migration speed and persistence in 2D and 3D. Finally we structureandcompositionoftheinterstitialtissuesdeterminesthe discusslimitsof3Dculturemodelstorecapitulatetheinvivo mechanisms that must be employed by the cancer cells for situation. successful navigation. Gritsenko and colleagues2 suggest three generic mechanisms employed by invasive cancer cells: contact guidance along stiff matrix elements through integrin mediated mechanisms, directed migration toward chemokines and growth factors and physical pushing through small spaces. Introduction Cell Migration In Vivo Nearly all adherent cells are able to crawl and migrate under standard 2D cell culture conditions. Time-lapse recordings of The process of cell migration and invasion is, of course, not migrating cells never fail to excite a sense of awe about the restricted to metastatic cancer. Cell migration and invasion complexityofthemigrationprocess,aslamellipodiaandfilopodia occurs during embryonic development and continues through- form, extend and make contact with the matrix, as cells change out adulthood as it is critical for the correct execution of a their shape, lunge forward, contract and detach. Cell migration variety of biological programs. During embryonic development depends on a variety of temporally and spatially orchestrated there are precise migration and invasion events that are parameters, with several hundreds of proteins involved. Although necessary for the final correct organization of the adult tissue. changesincellmigrationafterknockoutormutationofaprotein A particularly well-studied example of this is during the may not be the ultimate proof of a causal relationship; watching development of the mammalian brain, where cells undergo cells migrate and measuring parameters of cell speed and precisely timed migration events that are essential for the final directional persistence have proven invaluable in elucidating the tissue organization and wiring of the fully developed brain (for functional components regulating intrinsic cell migration in 2D. review, see ref. 3). The migration speed and persistence of A number of studies have now begun to examine these same distinct cell populations appear to be important parameters for parameters using 3D culture systems. determining their final destination in the maturing brain tissue. Moreover, cell populations change speed as they migrate throughout different zones of the brain.4 Mutations that alter *Correspondenceto:GeraldineM.O’Neill; speed and directional persistence of migrating neuronal cell Email:[email protected] populations can therefore perturb the organization of the brain Submitted:06/08/12;Revised:07/19/12;Accepted:07/20/12 http://dx.doi.org/10.4161/cam.21559 tissue into discrete layers.5,6 424 CellAdhesion&Migration Volume6Issue5 REVIEW The consequences for changes in speed and persistence of Eventhen,comparativeanalysisofcellmigrationandmorphology invasive cancer cells are currently less well known. Approaches has revealed striking inconsistencies between cells grown on 2D that abrogate cell movement in environments that mimic the surfaces vs. those seeded within 3D matrices.13,18 Therefore, in tissueorganizationoftheinvivotumoraregenerallyconsideredto attempting to recreate a cell’s in vivo environment, it is not bestrongcandidatesfornoveltherapiestotreatmetastaticcancer. sufficient to only consider which ECM group the cell type in Similarly, approaches that donotcompletely abrogate movement questionismostlikelytointeractwith;itisequallyimportantthat but instead slow down the rate of cell migration are expected to the 3D structure and mechanical properties of the matrix closely restrict the dissemination of cells from the primary tumor and mirror that of the in vivo environment. Three of the most increase the window of opportunity for localized and targeted commonly used in vitro 3D matrix models—fibrous gels, delivery of therapies. In the particular case of highly invasive basement membrane extracts (or matrigel) and cell-derived primary brain tumors such as the deadly grade IV gliomas (also matrices—are discussed in the following sections. knownasglioblastoma),thereisextensivecollateraldamagetothe Fibrous gels. Fibrous gels, namely reconstituted collagen and brain tissue as a result of the invading tumor cells, leading to fibrin gels, aim to replicate the major architectural and structural significantlossofqualityoflife.7Thus,evenapartialreductionof components of the extracellular matrix. Collagen and fibrin tumorinvasionmayamelioratesomeofthecollateraldamagethat emerged as ideal candidates for single-protein 3D matrix models occurs in the brains of these patients. Moreover, targeting the as both are major structural components of the extracellular cell’s migration machinery may in turn sensitize the cells to matrix. It is worth noting, however, that interstitial tissue is different therapies. One example of such sensitization is tumor composedofheterogenousmixturesofECMcomponents,which cells lacking expression of the invasion/metastasis promoter these gels do not recapitulate.19 Nevertheless, due to the NEDD9. These cells appear to concomitantly become sensitized abundance of collagen and fibrin in vivo, reconstituted fibrous to the effects of Src kinase inhibition.8 Similarly, decreasing the gels are physiologically compatible with numerous cells types. persistence of the cancer cell movement will necessarily decrease Fibrin gels are particularly suitable to study migratory events the numbers of successful escapers from the primary tumor site. involved in wound healing,20,21 while collagen gels are more But to confirm these speculations and expectations, any anti- suitable to study interstitial cell migration and metastatic invasivetherapymustfirstdemonstrateitseffectivenessinmodels invasion.10,22,23 that mimic the natural in vivo 3D organization of the relevant Notallfibrousgelsarecreatedalike,however.Seeminglyminor tissue surrounding a particular tumor. protocol variations can have major effects on the gel properties. For instance, an increase in fiber concentration enhances fiber 3D Culture Models density,24,25 resulting in an overall increase in the mechanical strength of the gel.26 Additionally, the effective pore size of the The interactions between cells and the external environment, fibrous meshwork decreases at higher fiber concentrations,27 thus either with the ECM or with neighboring cells, are in a literal increasing steric hindrance for migrating cells.28,29 The gelation sense vitally important. Once thought to solely provide physical temperature will also affect the mechanical and structural details and structural support, the ECM is now known to regulate and ofthegel;forinstance,theporesizeandfibrildiameterdecreases influence survival,9 proliferation,9 migration and adhesion,10,11 and the number of fibrils formed increases at higher polymeriza- morphology,12 internal cellular structures13 and signaling path- tion temperatures.27,30 ways.11 During tumorigenesis, cellular interaction with the ECM The production details of harvesting, dissolving and purifying canbecomederegulatednotonlyasaresultofalteredintegrinand collagenorfibrinfromdifferentanimaltissuesarealsoasourceof focaladhesionproteinexpression,butalsobecausetumorcellscan substantial variability. This is especially true in the case of rearrange the local native tissue architecture, alter its chemical collagen, wherefibersharvestedforexamplefrombovine Achilles composition,structureandstiffness,whichinturnoftenpromotes tendon demonstrate a higher denaturation temperature and a cellmigrationandinvasion.14,15Forexample,inbreastcarcinomas thicker fiber diameter when compared with collagen harvested the formation of the distinctive palpable “lump,” known as from rat tail tendon.27,31,32 The method for dissolving the desmoplasia, is attributed to tumor cell secretion of platelet collagen, whether with acid or by proteolytic digestion (i.e., derived growth factor (PDGF). This induces the deposition of pepsin or trypsin) similarly influences the structure and large quantities of collagen and collagen cross-linkers by polymerization kinetics of the reconstituted gels. Acid-dissolved surrounding fibroblasts, thus increasing local tissue stiffness and collagen will begin the fibrillogenesis process more rapidly,31,33,34 rigidity.16,17 showing characteristic fiber lengths and diameters that closely Invivo,mostinvadingcancer cells encountertwomajorECM resemble those observed in vivo.35 In comparison, collagen groups.Thefirstisthebasementmembranethatdirectlyinteracts extracted by proteolytic enzyme digestion shows significantly with both the epithelium and endothelium layers. The second is smaller fiber diameters and reduced fibril formation, thus theinterstitial matrix.Itharbors numerouscollagenisoforms and reducing the gels’ strength and rigidity.30,31,36 fibronectinandisthoughttocontributetotheoverallmechanical Basement membrane extracts. Matrigel is a reconstituted strength of the tissue. In vitro, the extracellular matrix is most basement membrane matrix product extracted from Engelbreth- commonlyplanar,tissueculture-treatedplasticorsometimesglass Holm-Swarm (EHS) mouse sarcoma cells. When polymerized, coatedwithcollagen,fibronectin orotherECM-derived proteins. matrigel mimics the microenvironment of in vivo extracellular www.landesbioscience.com CellAdhesion&Migration 425 basement membrane.37-39 Composed chiefly of laminin, collagen but decreasing speeds in 2D.57 Without exception, the knock- IV isoforms and heparan sulfate proteoglycans, basement down or loss of cytoplasmic members of focal adhesion sites membranes provide a unique signaling platform for cells that (p130Cas, NEDD9, vinculin, talin, FAK, VASP, paxillin and adhere to it.40,41 Matrigel 3D substrates can restore the normal Hic-5) inhibited cell migration speed in 3D cultures, yet again morphological characteristics and specific cell functions that varied results are seen in 2D cultures. Depletion of these would otherwise be lost under standard cell culture conditions, molecules variously increased 2D speed (p130Cas, NEDD9 and especially in epithelial-,42,43 endothelial-44 and Schwan-derived vinculin knockout fibroblasts), decreased 2D speed (talin and cells.45 But this is not generally true for all cells; for instance FAK) or did not alter speed relative to controls (vinculin shRNA fibroblasts that are not normally in contact with a basement in HT1080 fibroblasts and VASP). With the exception of EGF membrane in vivo adopt a noticeably uncharacteristic rounded stimulation and high a5β1 expression, all treatments resulted in morphology and exhibit a non-migratory phenotype when reduced intrinsic persistence of migration in 3D. As was seen for cultured in matrigel.12,39 the speed data, there is no clear correlation between effects on Cell-derivedmatrix.Whenculturedathighdensity,fibroblasts migration persistence in 2D and effects seen in 3D. “bioengineer” their own extracellular matrix.46 This cell-derived matrix(CDM)iscomposed of fibronectin,47 collagentypes Iand Forces in Cell Migration II,48 hyaluronic acid49 and heparin sulfate proteoglycans.49,50 CDMs mirror the flexibility and malleability of an in vivo The disparate results for 2D and 3D speed and persistence in ECM, as cells seeded onto a CDM can reorganize and modulate exactly the same cell background highlight critical differences the matrix, recapitulating the migration and invasion events of between these different model systems. Cell migration in 2D both physiological and metastatic processes observed in differsinonefundamentalaspectfrom3Dmigrationinthatsteric vivo.11,12,51 However, as with all protocols, there are limitations hindrance is absent. In 2D, the cell needs to overcome only the to this method also. Somewhat problematic is the poorly defined frictional (drag) forces from the surrounding liquid and the 2D andhighlyvariablecompositionoffibroblasticCDMs.Moreover, surface.At aspeedof only afew microns per minute,liquid drag althoughCDMsdisplaya3Dmatrixorganization,thesematrices can be neglected, as the force of a single myosin motor would be are typically thin and therefore subject to the rigidity of the sufficienttopropelthecellforward.Thus,thebetteracelladheres underlying2Dsurface.Regardless,theuseofCDMhaspermitted to its substrate, the greater the contractile force needed to insights into the signaling pathways that govern proliferation, overcome the adhesive friction from cell-matrix adhesions; as a morphology, cell-matrix attachments and individual cell migra- general rule, a cell with poorer adhesion can migrate faster on a tion, all of which impact and promote tumorigenesis.11,12,51 2D surface. Poorer adhesion can either be a consequence of a lowermatrixproteindensityonthesurface,58oraconsequenceof Cell Adhesion a reduced expression level of adhesion proteins on the cell (for examples, see Table 1). Of course, a minimum amount of The role of focal adhesions has been thoroughly studied in 2D adhesion is essential for cell movement as otherwise the cell culturesandhasprovidedimportantinsightintothemechanisms cannot polarize or activate its contractile machinery.58 Thus, in a ofcellmovementin2D.However,asinvestigatorshaveturnedto 2D context, changes in focal adhesion dynamics59,60 dominantly the analysis of focal adhesions in 3D, there has been conflicting effect cell migration speed and likely explain the disparate effects reports regarding their presence and detection in 3D.12,52,53 We on 2D cell migration seen following depletion of focal adhesion- direct readers to the excellent review by Harunaga54 for a associated molecules (Table 1). comprehensiveanalysisoftheissueofcell-ECMadhesions in3D In 3D, in addition to frictional forces, the cell also has to culture models. The upshot of which is that focal adhesions do overcomeforcesthatarisefromthesterichindranceofthematrix exist in 3D, but the composition, mechanical properties and network, provided that the pores and crevices through which the structure ofthe3Dmatrixprofoundlyaffects theappearance and cellmigratesaresmallerthanthecellitself.Here,thecellhastwo distribution of focal adhesions.12 options. It can either deform itself until it can fit through the InTable1 wesummarizedatafromstudiesinwhichmigration pores; or it can deform the network until the pores are large parameters of speed and persistence have been directly compared enough for passage. Switching between cell body deforming vs. between 2D and 3D culture conditions in the same cell matrix deforming migration strategies is evidenced by changes in background. Treatments that are expected to promote invasion cell morphology from rounded cell shapes to elongated cell and metastasis in vivo such as stimulation with EGF, complete shapes.61Deforming thecell bodysufficientlytosqueezethrough oncogenic transformation following combined expression of smallporesrequiresforcesforovercomingtheelasticandfrictional ErbB2 and 14-3-3f and high a5β1 expression all promoted forces of the cytoskeleton. The cell can decrease cytoskeletal faster migration speed in 3D cultures, yet had variable effects in elasticity (stiffness) and friction by depolymerizing cytoskeletal 2D (Table 1). For example, breast epithelial cells induced to filaments, but although this would reduce the forces necessary to undergo oncogenic transformation displayed reduced migration deform the cell until it fits through a pore, it also reduces the speed in 2D,55 while EGF treatment of glioblastoma cells force-generatingcapacityoftheacto-myosincontractileapparatus. increased 2D speed.56 Similarly, cells from a metastatic Analogously, deforming the matrix network requires forces for progression series showed increasing speeds in 3D collagen gels, overcoming the elastic and frictional forces of the matrix. By 426 CellAdhesion&Migration Volume6Issue5 Table1.Studiesreportingcellspeedandpersistencein3Dculturescomparedwith2Dmigrationparametersinthesamecellbackground Adhesion Cellline 2D 3D molecule Speed Persistence Speed Persistence p130CasshRNA52 HT1080 ↑ ↑ 3Dcollagen ↓ fibrosarcoma ↓ NEDD9knockout60,† Fibroblast ↑ ND 3Dcollagen ↓ ↓ Vinculinknockout87 Fibroblast ↑ ↑ 3Dcollagen ND ↓ VinculinshRNA52 HT1080 — — 3Dcollagen ↓ fibrosarcoma Nochange b3integrin Fibroblast ↑ ↓ CDM# ↓ knockout59,* ↑ EGFstimulation56 U87MG ↑ ↓ 3Dcollagen|| 3Dcollagen glioblastoma ↑ ↑ b1integrin DU-145 Biphasic ND Lowmatrigel%: ND inhibition52 prostatecancer ↑ Highmatrigel%: ↓ TalinshRNA52 HT1080 ↓ ↑ 3Dcollagen ↓ fibrosarcoma ↓ FAKshRNA52 HT1080 ↓ ↑ 3Dcollagen ↓ fibrosarcoma ↓ VASPshRNA52 HT1080 — — 3Dcollagen ↓ fibrosarcoma ↓ Transformation MCF10anon- (1)↑ (1)↓ “Compliant” “Compliant”3Dcollagen4: progression transformed (2)↓ (2)↓ 3Dcollagen4: (1)— series:55 breastepithelia (3)↓ (3)↓ (1)— (2)— (1)+ErbB2 (2)— (3)↓ (2)+14-3-3f (3)↑ “Stiff” (3)+ErbB2+14-3-3f “Stiff” 3Dcollagen1: 3Dcollagen1: (1)↓ (1)↓ (2)↓ (2)↓ (3)↓ (3)↑ Metastasisprogressionseries:57 NmuMgmouse ↓ ↓ Collagengel: Collagengel: (1)67NR—tumorigenic breastepithelial (relative (relative ↑ 66c14 (2)168FARN—micrometastases line to to (relativeto ↓ (3)4T07—secondarymicrometastases 67NR) NmuMG) 67NR) (4)66c14—secondaryTumors PaxillinsiRNA94," MDA-MB-231 ND ND CDM# ↓ breastcancer ↓ Hic-5siRNA94 MDA-MB-231 ND ND CDM# ↓ breastcancer ↓ Higha5b162 MDA-MB-231 ND ND 3Dcollagen ↑ breastcancer ↑ *Reducedfocaladhesionlengthandincreasedratesoffocaladhesionturnover;†increasedratesoffocaladhesiondisassembly;4elasticmodulus=103Pa; 1elasticmodulus=391Pa;||effectswerecomparedacrossarangeofcollagenconcentrations(2,3and4mg/mL);"measureddynamicsofGFP-taggedtalin adhesionsinCDMandshowedmultipleroles—paxillindepletioncausedincreasednumbersofhighlydynamic,peripheral,short-livedadhesions,increased stability of more centrally located adhesions accompanied by reduced rates of adhesion assembly and disassembly. #cell-derived matrix; ND, not determined;—,nochange. secreting matrix-digesting enzymes, the cell can soften the cell migration in 3D give clues as to whether cells are using the network structure or increase its porosity, thereby reducing the rounded (cell body deforming) vs. elongated (matrix deforming) necessary forces for deforming the matrix network until the cell cellmigrationstrategies?Initialdescriptionssuggestedthatthereis canfitthrough.Butatthesametimethisalsolimitstheabilityof nolossofcellspeedbetweenthetwomodes.10However,absolute the cell to adhere to the matrix, which is a prerequisite for cell speed depends on the make-up of the surrounding matrix. contractileforcegeneration.Therefore,doesthespeedofintrinsic While partial transformation of breast epithelial cells (Erb2 or www.landesbioscience.com CellAdhesion&Migration 427 14-3-3f overexpression) did not affect cell speed in “compliant” initialtumor formation.A limitationisthatthis doesnotaddress collagen gels (103 Pa elastic modulus), cells had reduced initial migration and invasion away from primary tumor sites. A migrationspeedin“stiff”collagengels(391Paelasticmodulus).55 further important caveat to all of these models is that none Similarly, increasing steric hindrance at high collagen concentra- measure migration and invasion alone. The ability of the tumor tions can inhibit cell invasion.62 Thus the concentration of the cellstodisseminateandformsecondarytumorsisamultifactorial matrix, extent of cross-linking and matrix pore size all critically process, which additionally requires survival in the circulation, determine cell speed and moreover, whether a matrix deforming adhesion to vessel endothelial layers and transmigration through or cell body deforming mode of invasion is possible. Given the theendothelium.Belowwecomparedatafor3Dmigrationspeed ultimate goal of determining the molecular mechanisms that and persistence with in vivo metastasis results in response to promotecancercellinvasionandmetastasis,itwillbeincreasingly altered adhesion molecule expression and activities. important toadopt arange of3Dculture modelsthatcan mimic FAK,p130Cas,talinandvinculin.Todate,alargenumberof the various different extracellular environments encountered by studies have focused on the role of FAK for in vivo metastasis. invading cancer cells in vivo. Two examples from recent studies FAK has long been known to be upregulated in awide variety of withopposingfindingshighlighttheproblemininterpretingdata invasive cancer types (reviewed in ref. 86). Cancer cells in which that are measured under only a single condition: Several studies FAK is either homozygously deleted or downregulated via have reported that invasive, metastatic cancer cells are softer than expression of a dominant negative form of FAK universally theirnonmetastaticcounterpartsderivedfromthesametissue63,64 display reduced metastatic tumor formation (Table 2), matching when measured under non-adherent conditions. By contrast, in the findings that FAK depletion reduces 3D cell migration speed adherent cancer cells, the opposite behavior is seen, with more (Table1).FAKdepletioninhibitedmetastatictumorformationin invasive cells tending to be more contractile and consequently all three categories of in vivo invasion models (see above). stiffer.65,66 Congruent results were also seen between 3D culture migration speedandinvivometastasesforp130Casandtalin(Tables1 and Predictive Power of 3D Invasion Models 2).However,thedataarenotsoclearforvinculin.Thedepletion for Cancer Metastasis orabsenceofvinculinhadnoeffect52orledtoreducedmigration speed87 in 3D models. In contrast, a high level of exogenous The goal of the 3D cell migration models is to better mimic the vinculin expression reduced lung tumor formation in vivo.80 physiological environment. An important question in cancer NEDD9. NEDD9 overexpression promotes invasion and research therefore is how successfully data from the 3D models metastasis in a range of cancer cell types.88 In apparent predicts in vivo metastatic behavior. To address this question we concordance with this, NEDD9 depletion reduces fibroblast have searched the literature for examples where adhesion migration in 3D collagen gels (Table 1). However, while moleculesthathavebeendirectlycomparedin2Dvs.3Dmodels NEDD9−/− tumor cells in a model of oncogenic transformation (Table 1) have been analyzed using in vivo models of metastasis of mammary epithelia had a tendency to form fewer metastatic andinvasion(Table2).Thereisanextensiveliteratureonmouse tumors,67 tail vein injection of these isolated cells revealed models of cancer cell invasion, but for our purposes we have aggressive lung tumor formation.8 Potentially, this may represent limited our analysis to studies where adhesion molecules from a switch to the amoeboid/cell deforming invasion mode that Table1 havebeenmanipulatedbyoverexpressionorknockdown. occurs following NEDD9 depletion,22 which may promote lung It is noticeable that surprisingly few studies have been performed colonization. By contrast, the dense, rigid matrix that is reported to directly test the effects of these molecules on in vivo tumor to characteristically surround breast tumors,89 which is presum- invasion and metastasis—the bulk of studies have focused on ably present in the transformed breast epithelia model, may tumorinitiation.OursearchyieldedexamplesforFAK,p130Cas, obviate the rounded/cell deforming mode of invasion. NEDD9,vinculin,talin,β1integrinanda5β1andavβ3integrin Alternatively, increased metastatic lesions following tail vein receptors (Table 2). injection may reflect changes in survival in circulation, adhesion The in vivo models used in the summarized studies can be and transmigration of the endothelial cell layer of the vessel. divided into three categories: (1) tissue-specific neo-oncogenesis β1 integrin. The effects of β1 integrin depletion on migration that recapitulates carcinoma progression seen in human dis- in matrigel are dependent on the matrigel concentration; at high ease67-72 and orthotopic injection: both have the advantage of concentrations when the gel is less deformable, the loss of β1 mimicking the correct tissue context of the human disease,73-77 impaired the ability of the cells to invade.90 These results are in (2) sub-cutaneous injection at non-orthotopic sites78-80 and agreement with in vivo data where β1 depletion also caused (3) intravenous injection.8,81-85 In the case of the first two reduced metastatic lesions (Table 2). In one of the studies, categories successful metastasis requires the tumor cells to invade increased numbers of tumor emboli in the lymphatic vasculature locally, intravasate into blood or lymph vessels, extravasate at were noted.68 This suggests that the cells in this model were secondary sites and form new tumors. Conversely, intravenous competent to locally invade and intravasate, but were not able to injection does not depend on the formation of a primary tumor, execute the final stages required for secondary tumor formation. thusallowingseparationoftheeffectsofprimarytumorformation a5β1and avβ3integrinreceptors.Thefindingthathighlevel from dissemination into secondary tissues. This approach is a5β1isassociatedwithenhancedmigrationspeedandpersistence frequentlyusedwhenthemoleculeorpathwayofinterestinhibits of breast cancer cells in 3D collagen gels62 appears to agree with 428 CellAdhesion&Migration Volume6Issue5 Table2.Summaryofinvivometastasisdata,targetingtheadhesionmoleculesdescribedinTable1 Adhesion Invivomodel Metastasisphenotype molecule FAK Intravenousinjection Failedtoformlungtumors;cellsretainedinthelungcapillarybedwere ofFAK−/−tumorcells82 roundedandlackedmembraneextensionsintothevessel ActivatedErbB2mammarytumormodelwith Metastaticlungtumorsallnegativeforcellswithhomozygousdeletionof epithelial-specificFAKdeletion69 FAK Mammaryepithelium-specificdeletionofFAKin Metastaticlungtumorsallnegativeforcellswithhomozygousdeletionof MMTV-PyVmTMouseTumorModel70 FAK Mammaryepithelium-specificdeletionofFAKin Reducedlungtumormetastases MMTV-PyVmTMouseTumorModel71 Orthotopicinjectionofpancreaticcancercellstreatedwith Preventedformationoflivermetastases FAKsiRNA73 p130Cas Mammarytumormodel—injectionofcellsexpressing Inhibitslungcolonization induciblep130CasshRNA77 Athymicnudemiceinjectedsub-cutaneouslywith Exogenousp130Casexpressionincreasedformationofmetastaticlung p130Cas−/−fibroblaststransformedwithoncogenic tumorsaftersurgicalremovalofprimarytumors;authorscommentthat“the Srcandexpressingp130Cas78 capacityofthecellstoinvadethroughmatrigelwasstronglycorrelatedwith theircapacitytoinvadeandmetastasizeinvivo” Vinculin Exogenousvinculinexpressioninhighlymetastaticrat Highestlevelsofvinculinexpressionsuppressedformationoflung adenocarcinomainjectedintofootpad80 metastases,lowtomoderateexpressorsformedtumorsinlymphnodes closetoinjectionsitebutfailedtoformlungmetastases NEDD9 Mammaryepithelium-specificdeletionofNEDD9in Trendtofewerlungmetastases MMTV-PyVmTMouseTumorModel67 TailveininjectionofNEDD9-nullprimarytumors8 Tumorsformedofnull-celllinesexhibitedincreasedaggressiveness,withall injectedmicegeneratingsecondarytumors Talin Tailveininjectionsofprostatecancercelllinestreated Reducednumbersofmetastaticlunglesions withtalinshRNA84 avb3 Orthotopicinjectionintomammaryfatpadwithmammary Droveuniqueformationofbonemetastases;authorsshowincreased carcinomacelllineexpressingexogenousb376 haptotacticandchemotacticresponsetobone-matrixproteinsand solublefactors MDA-MB-435breastcancercellsexpressingconstitutively Enhancedlungcolonization activeavb3injectedintomousetailvein81 Intravenousinjectionofmetastaticavb3negative Re-expressionofb3inmetastatic,b3negativelinesreducedlung melanomacellsexpressingexogenousb391 colonization b1integrin Conditionaldeletionofb1integrinfrommammaryepithelia, Significantlyreducedformationoflungmetastases crossedwithMMTV/activatederbB272 Orthotopicinjectionofpancreaticcancercellstreated Absenceofanymetastatictumors;controlstreatedwitha2ora3integrin withb1integrinsiRNA74 subunitsiRNAdisplayedmetastatictumors Conditionaldeletionofb1integrinfrompancreatic Lossofb1expressioninducedincreasedtumorcellemboliinthelymphatic bcellscrossedwithRip1Tag2mice68 vasculaturebutnometastasisformation;similarly,tailveininjectionsofb tumorcellslackingb1integrinexpressiondidnotformmetastases Ras-myctransformedb1nullEScellsinjected Reducednumbersandsizeofmetastaticfociinthelung sub-cutaneously79 a5b1integrin HT-29coloncancercellsexpressingexogenousa5 Significantlyreducedlungandextrapulmonarymetastases integrininjectedintravenously85 LewisLungCarcinomacellsexpressinga5shRNAinjected Fewerlungtumors intotailvein83 reports that a5 depletion reduced lung colonization in a tail vein model.91 Fibroblasts lacking β3 integrin expression have signific- injection model.83 However, when a5 and β1 are analyzed antlyfastermigrationspeedsandreducedmigrationpersistencein separately, the findings are conflicting. Exogenous a5 expression a 3D CDM assay.59 It is difficult to estimate to what degree the was shown to significantly reduce lung and extrapulmonary usage of different cell types has contributed to the disparate metastasesarisingfrominjectedcoloncancercells.85Upregulation findings.Forinstance,thedifferentialexpressionofotherintegrin of β3 expression increased bone and lung metastases,76,81 yet receptors in different cell types may have influenced the invasion decreased lung tumor formation in an invasive melanoma and metastasis behavior. In particular, it has been suggested that www.landesbioscience.com CellAdhesion&Migration 429 cross-talkbetweenrecyclinga5β1andaVβ3receptorsmaybean and at the metastatic niches that characterize different tumors. importantmechanismofregulatingmigrationpersistence.92These Such tailored 3D models to map the journey taken by datahighlighttheneedfordirectcomparisonbetween3Dculture metastasizing cancer cells are still lacking for interactions with models and in vivo metastasis models for validation. the vasculature, host immune responses and other cell-cell interactions, and can generally mimic only short-term (, 1 Conclusion week) conditions.93 But even with these limitations, 3D models are essential for gaining a basic understanding of cell migration Our comparison of the effects of a handful of studies that have and invasion through tissue and will ultimately lead to novel investigated adhesion proteins in 2D vs. 3D culture models with cancer therapies. Invasion away from the primary tumor site in vivo measurements of metastasis has revealed a complex fundamentally underpins metastatic progression, and treatments picture. For instance, it is notable that without exception, the that target invasion therefore hold the promise of significantly depletion of the adhesion molecules summarized in Table 1 improving survival rates for most cancers. resultedindecreasedmigrationpersistencein3Dmodelsbutnot always in reduced metastasis formation in vivo (Table 2). To Acknowledgments accurately predict in vivo invasion behavior, 3D model systems Work in the authors’ labs is supported by National Health and are required that faithfully replicate the physical characteristics of MedicalResearchCouncilofAustraliagrant632515(G.M.O.), the environments encountered by transmigrating cancer cells. In NSWCancerCouncilgrantRG12-06(G.M.O.andB.F.),and particular it will be critical to develop model systems mimicking grants from the German Science Foundation (D.F.G. and the interstitial tissue organization both at the primary tumor site B.F.). References 10. WolfK,MüllerR,BorgmannS,BröckerEB,FriedlP. 19. Sabeh F, Shimizu-Hirota R, Weiss SJ. Protease- Amoeboid shape change and contact guidance: T- dependent versus -independent cancer cell invasion 1. HanahanD,WeinbergRA.Thehallmarksofcancer. lymphocyte crawling through fibrillar collagen is programs: three-dimensional amoeboid movement Cell2000;100:57-70;PMID:10647931;http://dx.doi. independentofmatrixremodelingbyMMPsandother revisited. J Cell Biol 2009; 185:11-9; PMID: org/10.1016/S0092-8674(00)81683-9 proteases.Blood2003;102:3262-9;PMID:12855577; 19332889;http://dx.doi.org/10.1083/jcb.200807195 2. GritsenkoPG,IlinaO,FriedlP.Interstitialguidanceof http://dx.doi.org/10.1182/blood-2002-12-3791 20. GeerDJ,SwartzDD,AndreadisST.Fibrinpromotes cancer invasion. J Pathol 2012; 226:185-99; PMID: 11. Cukierman E, Pankov R, Stevens DR, Yamada KM. migration in a three-dimensional in vitro model of 22006671;http://dx.doi.org/10.1002/path.3031 Taking cell-matrix adhesions to the third dimension. wound regeneration. Tissue Eng 2002; 8:787-98; 3. Kulesa PM, Bailey CM, Kasemeier-Kulesa JC, Science2001;294:1708-12;PMID:11721053;http:// PMID:12459057; http://dx.doi.org/10.1089/ McLennanR.Cranialneuralcrestmigration:newrules dx.doi.org/10.1126/science.1064829 10763270260424141 foranoldroad.DevBiol2010;344:543-54;PMID: 12. HakkinenKM,HarunagaJS,DoyleAD,YamadaKM. 21. BrowerJ,BlumbergS,CarrollE,PastarI,BremH,Chen 20399765;http://dx.doi.org/10.1016/j.ydbio.2010.04. Directcomparisonsofthemorphology,migration,cell W.Mesenchymalstemcelltherapyanddeliverysystems 010 adhesions,andactincytoskeletonoffibroblastsinfour innonhealingwounds.AdvSkinWoundCare2011;24: 4. Cameron DB, Kasai K, Jiang Y, Hu T, Saeki Y, different three-dimensional extracellular matrices. 524-32, quiz 533-4; PMID:22015751; http://dx.doi. KomuroH.Fourdistinctphasesofbasket/stellatecell Tissue Eng Part A 2011; 17:713-24; PMID: org/10.1097/01.ASW.0000407648.89961.a6 migration after entering their final destination (the 20929283; http://dx.doi.org/10.1089/ten.tea.2010. 22. Sanz-MorenoV,GadeaG,AhnJ,PatersonH,MarraP, molecularlayer)inthedevelopingcerebellum.DevBiol 0273 PinnerS,etal.Racactivationandinactivationcontrol 2009; 332:309-24; PMID:19500566; http://dx.doi. 13. ElsdaleT,BardJ.Collagensubstrataforstudiesoncell plasticityoftumorcellmovement.Cell2008;135:510- org/10.1016/j.ydbio.2009.05.575 behavior. J Cell Biol 1972; 54:626-37; PMID: 23;PMID:18984162;http://dx.doi.org/10.1016/j.cell. 5. Britto JM, Tait KJ, Johnston LA, Hammond VE, 4339818;http://dx.doi.org/10.1083/jcb.54.3.626 2008.09.043 KalloniatisM,TanSS.Alteredspeedsandtrajectories 14. Wetzels RH, Holland R, van Haelst UJ, Lane EB, 23. LeesJG,BachCT,O’NeillGM.Interiordecoration: of neurons migrating in the ventricular and subven- Leigh IM, Ramaekers FC. Detection of basement tropomyosininactindynamicsandcellmigration.Cell tricular zones of the reeler neocortex. Cereb Cortex membrane components and basal cell keratin 14 in AdhMigr2011;5:181-6;PMID:21173575;http://dx. 2011; 21:1018-27; PMID:20847150; http://dx.doi. noninvasiveandinvasivecarcinomasofthebreast.AmJ doi.org/10.4161/cam.5.2.14438 org/10.1093/cercor/bhq168 Pathol1989;134:571-9;PMID:2466404 24. Kotlarchyk MA, Shreim SG, Alvarez-Elizondo MB, 6. SimóS,JossinY,CooperJA.Cullin5regulatescortical 15. Levental KR,YuH,Kass L,LakinsJN, EgebladM, EstradaLC,SinghR,ValdevitL,etal.Concentration layering by modulating the speed and duration of Erler JT, et al. Matrix crosslinking forces tumor independentmodulation oflocal micromechanicsina Dab1-dependentneuronalmigration.JNeurosci2010; progression by enhancing integrin signaling. Cell fibringel.PLoSOne2011;6:e20201;PMID:21629793; 30:5668-76; PMID:20410119; http://dx.doi.org/10. 2009; 139:891-906; PMID:19931152; http://dx.doi. http://dx.doi.org/10.1371/journal.pone.0020201 1523/JNEUROSCI.0035-10.2010 org/10.1016/j.cell.2009.10.027 25. ZhuYK,UminoT,LiuXD,WangHJ,RombergerDJ, 7. Furnari FB, Fenton T, Bachoo RM, Mukasa A, 16. Shao ZM, Nguyen M, Barsky SH. Human breast SpurzemJR,etal.Effectofinitialcollagenconcentra- Stommel JM, Stegh A, et al. Malignant astrocytic carcinoma desmoplasia is PDGF initiated. Oncogene tiononfibroblastmediatedcontractionofcollagengels. glioma: genetics, biology, and paths to treatment. 2000; 19:4337-45; PMID:10980609; http://dx.doi. Chest2000;117(Suppl1):234S-5S;PMID:10843925; Genes Dev 2007; 21:2683-710; PMID:17974913; org/10.1038/sj.onc.1203785 http://dx.doi.org/10.1378/chest.117.5_suppl_1.234S http://dx.doi.org/10.1101/gad.1596707 17. WalkerRA.Thecomplexitiesofbreastcancerdesmo- 26. Miron-MendozaM,SeemannJ,GrinnellF.Thediffer- 8. SinghMK,IzumchenkoE,Klein-SzantoAJ,Egleston plasia. Breast Cancer Res 2001; 3:143-5; PMID: entialregulationofcellmotileactivitythroughmatrix BL, Wolfson M, Golemis EA. Enhanced genetic 11305947;http://dx.doi.org/10.1186/bcr287 stiffness and porosity in three dimensional collagen instabilityanddasatinibsensitivityinmammarytumor cellslackingNEDD9.CancerRes2010;70:8907-16; 18. Yamada KM, Pankov R, Cukierman E. Dimensions matrices. Biomaterials 2010; 31:6425-35; PMID: PMID:20940402; http://dx.doi.org/10.1158/0008- anddynamicsinintegrinfunction.BrazJMedBiolRes 20537378; http://dx.doi.org/10.1016/j.biomaterials. 5472.CAN-10-0353 2003;36:959-66;PMID:12886449;http://dx.doi.org/ 2010.04.064 10.1590/S0100-879X2003000800001 9. Chen CS, Mrksich M, Huang S, Whitesides GM, Ingber DE. Geometric control of cell life and death. Science1997;276:1425-8;PMID:9162012;http://dx. doi.org/10.1126/science.276.5317.1425 430 CellAdhesion&Migration Volume6Issue5 27. YangYL,LeoneLM,KaufmanLJ.Elasticmoduliof 41. McGuirePG,SeedsNW.Theinteractionofplasmino- 55. BakerEL,SrivastavaJ,YuD,BonnecazeRT,Zaman collagen gels can be predicted from two-dimensional genactivatorwithareconstitutedbasementmembrane MH.Cancercellmigration:integratedrolesofmatrix confocal microscopy. Biophys J 2009; 97:2051-60; matrix and extracellular macromolecules produced by mechanics and transforming potential. PLoS One PMID:19804737; http://dx.doi.org/10.1016/j.bpj. culturedepithelialcells.JCellBiochem1989;40:215- 2011; 6:e20355; PMID:21647371; http://dx.doi.org/ 2009.07.035 27; PMID:2504731; http://dx.doi.org/10.1002/jcb. 10.1371/journal.pone.0020355 28. GobeauxF,MosserG,AngloA,PanineP,DavidsonP, 240400210 56. Kim HD, Guo TW, Wu AP, Wells A, Gertler FB, Giraud-Guille MM, et al. Fibrillogenesis in dense 42. Li ML, Aggeler J, Farson DA, Hatier C, Hassell J, Lauffenburger DA. Epidermal growth factor-induced collagensolutions:aphysicochemicalstudy.JMolBiol Bissell MJ. Influence of a reconstituted basement enhancement of glioblastoma cell migration in 3D 2008; 376:1509-22; PMID:18234220; http://dx.doi. membrane and its components on casein gene arisesfromanintrinsicincreaseinspeedbutanextrinsic org/10.1016/j.jmb.2007.12.047 expressionandsecretioninmousemammaryepithelial matrix- and proteolysis-dependent increase in persist- 29. MickelW,MünsterS,JawerthLM,VaderDA,Weitz cells. Proc Natl Acad Sci U S A 1987; 84:136-40; ence. Mol Biol Cell 2008; 19:4249-59; PMID: DA,SheppardAP,etal.Robustporesizeanalysisof PMID:3467345; http://dx.doi.org/10.1073/pnas.84.1. 18632979; http://dx.doi.org/10.1091/mbc.E08-05- filamentousnetworksfromthree-dimensionalconfocal 136 0501 microscopy. Biophys J 2008; 95:6072-80; PMID: 43. LiW,MachuleD,GaoC,DenBestenPK.Growthof 57. Indra I, Undyala V, Kandow C, Thirumurthi U, 18835899; http://dx.doi.org/10.1529/biophysj.108. ameloblast-lineagecellsinathree-dimensionalMatrigel Dembo M, Beningo KA. An in vitro correlation of 135939 environment.EurJOralSci2006;114(Suppl1):159- mechanical forces and metastatic capacity. Phys Biol 30. McPhersonJM,WallaceDG,SawamuraSJ,ContiA, 63,discussion164-5,380-1;PMID:16674679;http:// 2011; 8:015015; PMID:21301068; http://dx.doi.org/ CondellRA,WadeS,etal.Collagenfibrillogenesisin dx.doi.org/10.1111/j.1600-0722.2006.00308.x 10.1088/1478-3975/8/1/015015 vitro:acharacterizationoffibrilqualityasafunctionof 44. KubotaY,KleinmanHK,MartinGR,LawleyTJ.Role 58. DiMilla PA, Stone JA, Quinn JA, Albelda SM, assembly conditions. Coll Relat Res 1985; 5:119-35; of laminin and basement membrane in the morpho- Lauffenburger DA. Maximal migration of human PMID:3924470; http://dx.doi.org/10.1016/S0174- logical differentiation of human endothelial cells into smooth muscle cells on fibronectin and type IV 173X(85)80034-0 capillary-likestructures.JCellBiol1988;107:1589-98; collagenoccursatanintermediateattachmentstrength. 31. Zeugolis DI, Paul RG, Attenburrow G. Factors PMID:3049626; http://dx.doi.org/10.1083/jcb.107.4. JCellBiol1993;122:729-37;PMID:8335696;http:// influencing the properties of reconstituted collagen 1589 dx.doi.org/10.1083/jcb.122.3.729 fiberspriortoself-assembly:animalspeciesandcollagen 45. Carey DJ, Todd MS, Rafferty CM. Schwann cell 59. WorthDC,Hodivala-DilkeK,RobinsonSD,KingSJ, extractionmethod.JBiomedMaterResA2008;86: myelination:inductionbyexogenousbasementmem- MortonPE,GertlerFB,etal.Alphavbeta3integrin 892-904; PMID:18041730; http://dx.doi.org/10. brane-likeextracellularmatrix.JCellBiol1986;102: spatiallyregulatesVASPandRIAMtocontroladhesion 1002/jbm.a.31694 2254-63; PMID:3086325; http://dx.doi.org/10.1083/ dynamicsandmigration.JCellBiol2010;189:369-83; 32. RaubCB,SureshV,KrasievaT,LyubovitskyJ,Mih jcb.102.6.2254 PMID:20404115; http://dx.doi.org/10.1083/jcb. JD, Putnam AJ, et al. Noninvasive assessment of 46. PorterKR,VanameeP.Observationsontheformation 200912014 collagengelmicrostructureandmechanicsusingmulti- of connective tissue fibers. Proc Soc Exp Biol Med 60. ZhongJ,BaquiranJB,BonakdarN,LeesJ,ChingYW, photon microscopy. Biophys J 2007; 92:2212-22; 1949;71:513-6;PMID:18139799 PugachevaE,etal.NEDD9stabilizesfocaladhesions, PMID:17172303; http://dx.doi.org/10.1529/biophysj. 47. Hedman K, Vaheri A, Wartiovaara J. External increases binding to the extra-cellular matrix and 106.097998 fibronectinofculturedhumanfibroblastsispredomi- differentially effects 2D versus 3D cell migration. 33. Bannister DW,BurnsAB.Pepsin treatment ofavian nantlyamatrixprotein.JCellBiol1978;76:748-60; PLoSOne2012;7:e35058;PMID:22509381;http:// skincollagen.Effectsonsolubility,subunitcomposition PMID:344328; http://dx.doi.org/10.1083/jcb.76.3. dx.doi.org/10.1371/journal.pone.0035058 andaggregationproperties.BiochemJ1972;129:677- 748 61. Sanz-Moreno V, Marshall CJ. The plasticity of 81;PMID:4572795 48. VaheriA,KurkinenM,LehtoVP,LinderE,TimplR. cytoskeletaldynamicsunderlyingneoplasticcellmigra- 34. SnowdenJM,SwannDA.Theformationandthermal Codistribution of pericellular matrix proteins in tion. Curr Opin Cell Biol 2010; 22:690-6; PMID: stabilityofinvitroassembledfibrilsfromacid-soluble cultured fibroblasts and loss in transformation: fibro- 20829016; http://dx.doi.org/10.1016/j.ceb.2010.08. and pepsin-treated collagens. Biochim Biophys Acta nectin and procollagen. Proc Natl Acad Sci U S A 020 1979;580:372-81;PMID:42446;http://dx.doi.org/10. 1978;75:4944-8;PMID:217006;http://dx.doi.org/10. 62. MierkeCT,FreyB,FellnerM,HerrmannM,FabryB. 1016/0005-2795(79)90149-1 1073/pnas.75.10.4944 Integrin a5β1 facilitates cancer cell invasion through 35. Forgacs G, Newman SA, Hinner B, Maier CW, 49. Hedman K, Kurkinen M, Alitalo K, Vaheri A, enhancedcontractileforces.JCellSci2011;124:369- SackmannE.Assemblyofcollagenmatricesasaphase Johansson S, Höök M. Isolation of the pericellular 83; PMID:21224397; http://dx.doi.org/10.1242/jcs. transitionrevealedbystructuralandrheologicstudies. matrixofhumanfibroblastcultures.JCellBiol1979; 071985 BiophysJ2003;84:1272-80;PMID:12547807;http:// 81:83-91; PMID:383722; http://dx.doi.org/10.1083/ 63. GuckJ,SchinkingerS,LincolnB,WottawahF,Ebert dx.doi.org/10.1016/S0006-3495(03)74942-X jcb.81.1.83 S, Romeyke M, et al. Optical deformability as an 36. NotbohmH,MoslerS,MüllerPK,BrinckmannJ.In 50. Narayanan K, Leck KJ, Gao S, Wan AC. Three- inherentcellmarkerfortestingmalignanttransforma- vitroformationandaggregationofheterotypiccollagen dimensional reconstituted extracellular matrix scaf- tionandmetastaticcompetence.BiophysJ2005;88: Iand IIIfibrils.Int JBiolMacromol 1993;15:299- folds for tissue engineering. Biomaterials 2009; 30: 3689-98; PMID:15722433; http://dx.doi.org/10. 304; PMID:8251445; http://dx.doi.org/10.1016/ 4309-17; PMID:19477508; http://dx.doi.org/10. 1529/biophysj.104.045476 0141-8130(93)90030-P 1016/j.biomaterials.2009.04.049 64. RemmerbachTW,WottawahF,DietrichJ,LincolnB, 37. Timpl R, Martin GR, Bruckner P, Wick G, 51. PankovR,EndoY,Even-RamS,ArakiM,ClarkK, Wittekind C, Guck J. Oral cancer diagnosis by Wiedemann H.Natureofthecollagenousproteinin Cukierman E, et al. A Rac switch regulates random mechanicalphenotyping.CancerRes2009;69:1728- atumorbasementmembrane.EurJBiochem1978;84: versusdirectionallypersistentcellmigration.JCellBiol 32; PMID:19223529; http://dx.doi.org/10.1158/ 43-52; PMID:648517; http://dx.doi.org/10.1111/j. 2005; 170:793-802; PMID:16129786; http://dx.doi. 0008-5472.CAN-08-4073 1432-1033.1978.tb12139.x org/10.1083/jcb.200503152 65. KochTM,MünsterS,BonakdarN,ButlerJP,FabryB. 38. OrkinRW,GehronP,McGoodwinEB,MartinGR, 52. Fraley SI, Feng Y, Krishnamurthy R, Kim DH, 3DTractionforcesincancercellinvasion.PLoSOne ValentineT,SwarmR.Amurinetumorproducinga Celedon A, Longmore GD, et al. A distinctive role 2012; 7:e33476; PMID:22479403; http://dx.doi.org/ matrixofbasementmembrane.JExpMed1977;145: for focal adhesion proteins in three-dimensional cell 10.1371/journal.pone.0033476 204-20; PMID:830788; http://dx.doi.org/10.1084/ motility. Nat Cell Biol 2010; 12:598-604; PMID: 66. Kraning-Rush CM, Califano JP, Reinhart-King CA. jem.145.1.204 20473295;http://dx.doi.org/10.1038/ncb2062 Cellular traction stresses increase with increasing 39. Kleinman HK. Preparation of basement membrane 53. Kubow KE, Horwitz AR. Reducing background metastatic potential. PLoS One 2012; 7:e32572; components from ehs tumors. Curr Protoc Cell Biol fluorescence reveals adhesions in 3D matrices. Nat PMID:22389710; http://dx.doi.org/10.1371/journal. 2001;Chapter10:Unit102. Cell Biol 2011; 13:3-5, author reply 5-7; PMID: pone.0032572 40. KleinmanHK,McGarveyML,LiottaLA,RobeyPG, 21173800;http://dx.doi.org/10.1038/ncb0111-3 TryggvasonK,MartinGR.Isolationandcharacteriza- 54. Harunaga JS, Yamada KM. Cell-matrix adhesions in tion of type IV procollagen, laminin, and heparan 3D. Matrix Biol 2011; 30:363-8; PMID:21723391; sulfate proteoglycan from the EHS sarcoma. http://dx.doi.org/10.1016/j.matbio.2011.06.001 Biochemistry 1982; 21:6188-93; PMID:6217835; http://dx.doi.org/10.1021/bi00267a025 www.landesbioscience.com CellAdhesion&Migration 431 67. Izumchenko E, Singh MK, Plotnikova OV, 76. SloanEK,PouliotN,StanleyKL,ChiaJ,MoseleyJM, 86. McLean GW,CarragherNO,AvizienyteE,EvansJ, Tikhmyanova N, Little JL, Serebriiskii IG, et al. Hards DK, et al. Tumor-specific expression of BruntonVG,FrameMC.Theroleoffocal-adhesion NEDD9 promotes oncogenic signaling in mammary alphavbeta3integrinpromotesspontaneousmetastasis kinaseincancer—anewtherapeuticopportunity.Nat tumor development. Cancer Res2009; 69:7198-206; ofbreastcancertobone.BreastCancerRes2006;8: RevCancer2005;5:505-15;PMID:16069815;http:// PMID:19738060; http://dx.doi.org/10.1158/0008- R20; PMID:16608535; http://dx.doi.org/10.1186/ dx.doi.org/10.1038/nrc1647 5472.CAN-09-0795 bcr1398 87. Mierke CT, Kollmannsberger P,Zitterbart DP, Diez 68. Kren A, Baeriswyl V, Lehembre F, Wunderlin C, 77. CabodiS,TinnirelloA,BisaroB,TornilloG,delPilar G, Koch TM, Marg S, et al. Vinculin facilitates cell StrittmatterK,AntoniadisH,etal.Increasedtumorcell Camacho-Leal M, Forni G, et al. p130Cas is an invasion into three-dimensional collagen matrices. J disseminationandcellularsenescenceintheabsenceof essentialtransducerelementinErbB2transformation. Biol Chem 2010; 285:13121-30; PMID:20181946; beta1-integrin function. EMBO J2007; 26:2832-42; FASEB J 2010; 24:3796-808; PMID:20505116; http://dx.doi.org/10.1074/jbc.M109.087171 PMID:17541405; http://dx.doi.org/10.1038/sj.emboj. http://dx.doi.org/10.1096/fj.10-157347 88. O’NeillGM,SeoS,SerebriiskiiIG,LessinSR,Golemis 7601738 78. BrábekJ,ConstancioSS,SiesserPF,ShinNY,PozziA, EA. A new central scaffold for metastasis: parsing 69. Lahlou H, Sanguin-Gendreau V, Frame MC, Muller HanksSK.Crk-associatedsubstratetyrosinephosphor- HEF1/Cas-L/NEDD9. Cancer Res 2007; 67:8975-9; WJ.Focaladhesionkinasecontributestoproliferative ylationsitesarecriticalforinvasionandmetastasisof PMID:17908996; http://dx.doi.org/10.1158/0008- potential of ErbB2 mammary tumour cells but is SRC-transformedcells.MolCancerRes2005;3:307- 5472.CAN-07-1328 dispensableforErbB2mammarytumourinductionin 15; PMID:15972849; http://dx.doi.org/10.1158/ 89. Paszek MJ, Zahir N, Johnson KR, Lakins JN, vivo. Breast Cancer Res 2012; 14:R36; PMID: 1541-7786.MCR-05-0015 RozenbergGI,GefenA,etal.Tensionalhomeostasis 22373082;http://dx.doi.org/10.1186/bcr3131 79. BrakebuschC,WennerbergK,KrellHW,WeidleUH, and the malignant phenotype. Cancer Cell 2005; 8: 70. LahlouH,Sanguin-GendreauV,ZuoD,CardiffRD, SallmyrA,JohanssonS,etal.Beta1integrinpromotes 241-54; PMID:16169468; http://dx.doi.org/10.1016/ McLeanGW,FrameMC,etal.Mammaryepithelial- butisnotessentialformetastasisofras-myctransformed j.ccr.2005.08.010 specificdisruptionofthefocaladhesionkinaseblocks fibroblasts. Oncogene 1999; 18:3852-61; PMID: 90. ZamanMH,TrapaniLM,SieminskiAL,MackellarD, mammarytumorprogression.ProcNatlAcadSciUSA 10445848;http://dx.doi.org/10.1038/sj.onc.1202770 GongH,KammRD,etal.Migrationoftumorcellsin 2007; 104:20302-7; PMID:18056629; http://dx.doi. 80. Rodríguez Fernández JL, Geiger B, Salomon D, 3Dmatricesisgovernedbymatrixstiffnessalongwith org/10.1073/pnas.0710091104 Sabanay I, Zöller M, Ben-Ze’ev A. Suppression of cell-matrixadhesionandproteolysis.ProcNatlAcadSci 71. Provenzano PP, InmanDR,EliceiriKW,Beggs HE, tumorigenicity in transformed cells after transfection USA2006;103:10889-94;PMID:16832052;http:// Keely PJ. Mammary epithelial-specific disruption of with vinculin cDNA. J Cell Biol 1992; 119:427-38; dx.doi.org/10.1073/pnas.0604460103 focal adhesion kinase retards tumor formation and PMID:1400584; http://dx.doi.org/10.1083/jcb.119.2. 91. DanenEH,vanKraatsAA,CornelissenIM,RuiterDJ, metastasisinatransgenicmousemodelofhumanbreast 427 van Muijen GN. Integrin beta 3 cDNA transfection cancer. Am J Pathol 2008; 173:1551-65; PMID: 81. Felding-Habermann B, O’Toole TE, Smith JW, intoahighlymetastaticalphavbeta3-negativehuman 18845837; http://dx.doi.org/10.2353/ajpath.2008. Fransvea E, Ruggeri ZM, Ginsberg MH, et al. melanomacelllineinhibitsinvasionandexperimental 080308 Integrinactivationcontrolsmetastasisinhumanbreast metastasis.BiochemBiophysResCommun1996;226: 72. Huck L, Pontier SM, Zuo DM, Muller WJ. beta1- cancer.ProcNatlAcadSciUSA2001;98:1853-8; 75-81; PMID:8806594; http://dx.doi.org/10.1006/ integrin is dispensable for the induction of ErbB2 PMID:11172040; http://dx.doi.org/10.1073/pnas.98. bbrc.1996.1313 mammary tumors but plays a critical role in the 4.1853 92. WhiteDP,CaswellPT,NormanJC.alphavbeta3and metastaticphaseoftumorprogression.ProcNatlAcad 82. Pylayeva Y, Gillen KM, Gerald W, Beggs HE, alpha5beta1integrinrecyclingpathwaysdictatedown- Sci U S A 2010; 107:15559-64; PMID:20713705; ReichardtLF,GiancottiFG.Ras-andPI3K-dependent streamRhokinasesignalingtoregulatepersistentcell http://dx.doi.org/10.1073/pnas.1003034107 breasttumorigenesisinmiceandhumansrequiresfocal migration. J Cell Biol 2007; 177:515-25; PMID: 73. DuxburyMS,ItoH,ZinnerMJ,AshleySW,Whang adhesionkinasesignaling.JClinInvest2009;119:252- 17485491;http://dx.doi.org/10.1083/jcb.200609004 EE. Focal adhesion kinase gene silencing promotes 66;PMID:19147981 93. YamadaKM,CukiermanE.Modelingtissuemorpho- anoikisandsuppressesmetastasisofhumanpancreatic 83. RomanJ,RitzenthalerJD,Roser-PageS,SunX,HanS. genesis and cancer in 3D. Cell 2007; 130:601-10; adenocarcinoma cells. Surgery 2004; 135:555-62; alpha5beta1-integrin expression is essential for tumor PMID:17719539; http://dx.doi.org/10.1016/j.cell. PMID:15118593; http://dx.doi.org/10.1016/j.surg. progressioninexperimentallungcancer.AmJRespir 2007.08.006 2003.10.017 Cell Mol Biol 2010; 43:684-91; PMID:20081050; 94. DeakinNO,TurnerCE.Distinctrolesforpaxillinand 74. Grzesiak JJ, Tran Cao HS, Burton DW, Kaushal S, http://dx.doi.org/10.1165/rcmb.2009-0375OC Hic-5 in regulating breast cancer cell morphology, Vargas F, Clopton P, et al. Knockdown of the 84. Sakamoto S, McCann RO, Dhir R, Kyprianou N. invasion,andmetastasis.MolBiolCell2011;22:327- β(1) integrin subunit reduces primary tumor growth Talin1 promotes tumor invasion and metastasis via 41;PMID:21148292;http://dx.doi.org/10.1091/mbc. andinhibitspancreaticcancermetastasis.IntJCancer focaladhesionsignalingandanoikisresistance.Cancer E10-09-0790 2011; 129:2905-15; PMID:21491421; http://dx.doi. Res 2010; 70:1885-95; PMID:20160039; http://dx. org/10.1002/ijc.25942 doi.org/10.1158/0008-5472.CAN-09-2833 75. HsuMY,ShihDT,MeierFE,VanBelleP,HsuJY, 85. Schirner M, Herzberg F, Schmidt R, Streit M, Elder DE, et al. Adenoviral gene transfer of beta3 SchöningM,HummelM,etal.Integrinalpha5beta1: integrin subunit induces conversion from radial to apotentinhibitorofexperimentallungmetastasis.Clin verticalgrowthphaseinprimaryhumanmelanoma.Am Exp Metastasis 1998; 16:427-35; PMID:10091938; JPathol1998;153:1435-42;PMID:9811334;http:// http://dx.doi.org/10.1023/A:1006581424490 dx.doi.org/10.1016/S0002-9440(10)65730-6 432 CellAdhesion&Migration Volume6Issue5