Down-regulation of cell adhesion via rho-associated protein kinase (ROCK) pathway promotes tumor cell migration on laminin-511

Down-regulation of cell adhesion via rho-associated protein kinase (ROCK) pathway promotes tumor cell migration on laminin-511

Author’s Accepted Manuscript Down-regulation of cell adhesion via rhoassociated protein kinase (ROCK) pathway promotes tumor cell migration on laminin...

1MB Sizes 0 Downloads 28 Views

Author’s Accepted Manuscript Down-regulation of cell adhesion via rhoassociated protein kinase (ROCK) pathway promotes tumor cell migration on laminin-511 Yamato Kikkawa, Nozomi Harashima, Kazuki Ikari, Shogo Fujii, Fumihiko Katagiri, Kentaro Hozumi, Motoyoshi Nomizu www.elsevier.com/locate/yexcr

PII: DOI: Reference:

S0014-4827(16)30071-4 http://dx.doi.org/10.1016/j.yexcr.2016.04.002 YEXCR10216

To appear in: Experimental Cell Research Received date: 7 January 2016 Revised date: 15 March 2016 Accepted date: 7 April 2016 Cite this article as: Yamato Kikkawa, Nozomi Harashima, Kazuki Ikari, Shogo Fujii, Fumihiko Katagiri, Kentaro Hozumi and Motoyoshi Nomizu, Downregulation of cell adhesion via rho-associated protein kinase (ROCK) pathway promotes tumor cell migration on laminin-511, Experimental Cell Research, http://dx.doi.org/10.1016/j.yexcr.2016.04.002 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

1 Down-regulation of cell adhesion via rho-associated protein kinase (ROCK) pathway promotes tumor cell migration on laminin-511

Yamato Kikkawa, Nozomi Harashima, Kazuki Ikari, Shogo Fujii, Fumihiko Katagiri, Kentaro Hozumi, and Motoyoshi Nomizu

Laboratory of Clinical Biochemistry, Tokyo University of Pharmacy and Life Sciences, Hachioji, Tokyo 192-0392, Japan

Running title: PMA promotes tumor cell migration on laminin-511

#Address correspondence to: Yamato Kikkawa, Ph.D Laboratory of Clinical Biochemistry, Tokyo University of Pharmacy and Life Sciences, 1432-1 Horinouchi,

Hachioji,

Tokyo

192-0392,

Japan,

Phone/Fax:+81-42-676-5670;

[email protected]

Keywords: Cell adhesion, Cell migration, Basement membrane, Integrin, Laminin

E-mail:

2 ABSTRACT Epithelial cells, both normal and precancerous, stably anchor to basement membranes, whereas malignant tumors pass through them to achieve metastasis.

Of basement membrane

components, laminin-511 (α5, β1, γ1; LM-511) has been found to be a major isoform in many adult basement membranes. Several studies have shown that LM-511 promotes not only cell adhesion but also tumor cell migration. Thus, LM-511 can be viewed like two distinct molecules in normal vs. tumor cells; tumor cells seem to be able to alter their response (adhesive vs. migratory) to LM-511. In this study we examined the effects of biologically active molecules on A549 lung adenocarcinoma cell adhesion to LM-511. Of them, phorbol 12-myristate 13-acetate (PMA) induced transition to a rounded cell shape and significantly promoted cell migration on LM-511. The attachment of PMA-treated A549 cells to LM-511 was weaker than that of control cells. PMA-stimulated signaling pathway reduced the binding of integrin α3β1 to LM-511. Cell migration assays using inhibitors for signal transduction and cytoskeletal organization showed that suppression of cell adhesion via the rho-associated protein kinase (ROCK) pathway promoted tumor cell migration on LM-511. Our results suggest that the ROCK pathway is involved in the transition from static to migratory cell behaviors on LM-511.

3 INTRODUCTION A multi-step process must occur for a tumor cell to leave its primary site and invade the surrounding tissues [1]. In order for tumor cells to invade, they must adhere to and actively migrate through the basement membrane into adjacent stroma.

There is growing knowledge about the

importance of the tumor microenvironment. Interaction of tumor cells with the surrounding stroma cells and extracellular matrix is a major contributor to tumor progression [2, 3]. During the progression from precancerous cells to an invasive carcinoma, tumor cells encounter epithelium, fibroblasts, inflammatory cells, vascular elements, connective tissues, and basement membrane components. Any of these cell types, or the molecules produced by them, have the potential to influence tumor progression. The enhancement of tumor cell motility by endogenous and exogenous molecules in the microenvironment is required for subsequent invasion and metastasis. Among the components of tumor microenvironments, basement membranes are implicated in direct interactions with tumor cells. Basement membrane is mainly composed of type IV collagen, laminin, nidogen, and perlecan [4]. To investigate tumor invasion in vitro, Matrigel, an extract derived from the mouse EHS sarcoma, has been used as a reconstituted basement membrane [5]. Of these components, laminin has been thought to be a key molecule in cell adhesion and cell migration during tumor invasion. Laminins are a family of non-collagenous glycoproteins composed of α, β, and γ chains. There are five α chains, three β chains and three γ chains [6]. So far 19 different laminin isoforms have been identified in various cultured cells and tissues [7]. The laminin isoform in Matrigel is laminin-111 (composed of α1, β1, and γ1 chains; LM-111), which is mainly expressed in fetal but not adult tissues. Therefore, many tumor cells only rarely interact with LM-111 in the process of tumor invasion. In contrast, laminin-511 (α5, β1, γ1; LM-511) is a major isoform in many adult basement membranes [8, 9]. Although LM-511 stably anchors epithelial cells in vivo, several in vitro studies

4 have shown that LM-511 promotes not only cell adhesion but also tumor cell migration [10-13]. LM-511 can behave like two distinct molecules in normal and tumor tissues. This is because tumor cells that are under the influence of biologically active molecules could alter their adhesion to LM-511. However it is unclear what sorts of molecules influence cell adhesion and migration on LM-511. Laminins activate multiple signal transduction pathways, including various components such as G-proteins, mitogen activated protein kinases, phosphatases, small GTPases of the Rho family, and cytoskeleton components [14]. In vitro studies have shown that the binding of integrin α3β1 to LM-511 activates Rac via the p130(Cas)-CrkII-DOCK180 pathway [11]. Cell adhesion to LM-511 also prevents apoptosis via Csk, phosphatidylinositol 3-kinase/Akt- and MEK1/ERK-dependent pathways [15, 16]. The studies of developing organs in laminin α5 deficient mice have shown that laminin α5 modulates the Sonic hedgehog pathway, the Wnt pathway, and the PI3Kinase/Akt pathway [17, 18]. However, despite the accumulated studies of outside-in signaling pathways triggered by binding of laminins to cell surface receptors, there are little data that biologically active molecules can activate inside-out signaling pathways to modulate cell adhesion and migration on LM-511. In this study we found that a potent tumor-promoting agent (PMA) promoted cell migration of A549 lung adenocarcinoma cells adhering to LM-511. Cell migration was associated with weakened cell adhesion to LM-511 via integrin α3β1. We also examined inside-out signaling pathways that modulate cell migration on LM-511. This study is the first report demonstrating that suppression of cell adhesion via inside-out signaling can promote tumor cell migration on LM-511.

5 MATERIALS AND METHODS Antibodies and reagents Monoclonal antibodies against Lu/B-CAM (87207, IgG2a and BRIC221) were purchased from R&D systems (Minneapolis, MN) and Serotec (Oxford, UK), respectively.

Monoclonal

antibodies against human integrin α3 (P1B5, IgG1) and β1 (6S6, IgG1) were purchased from Millipore (Temecula, CA). Rat monoclonal antibody against human integrin α6 (GoH3, IgG2a) was from BD Biosciences (Bedford, MA). Anti-human vinculin (hVIN-1) monoclonal antibody was purchased from Sigma-Aldrich (St. Louis, MO). Anti-human integrin β1 (TS2/16, IgG1) monoclonal antibody was prepared from the conditioned medium of hybridoma cells purchased from the American Type Culture Collection (Manassas, VA). Recombinant LM-511 was produced in HEK293 cells triple-transfected with mouse laminin α5, β1 and γ1 chains and purified as previously described [19]. Human fibronectin and epidermal growth factor (EGF) were purchased from BD Biosciences (Bedford, MA). Hepatocyte growth factor (HGF), Y27632, and Blebbistatin were purchased from Wako (Osaka, Japan). Lysophosphatidic acid and GSK429286 were purchased from Sigma-Aldrich. PMA were purchased from Roche (Mannheim, Germany), respectively. Transforming growth factor-β1 and ML141 were purchased from R&D systems (Minneapolis, MN). Cytochalasin B, LIMKi3, and NSC23766 were purchased from Calbiochem (Darmstadt, Germany).

Cell culture A549 (lung adenocarcinoma), A431 (epidermoid carcinoma), and MCF-7 (breast adenocarcinoma) were purchased from Health Science Research Resources Bank (Osaka, Japan). The cells were maintained in DMEM containing 10% FBS (Life Technologies, Carlsbad, CA).

K562

human erythroleukemia cells transfected with cDNA encoding human integrin α3A subunit were kindly

6 provided by Dr. Arnoud Sonnenberg (The Netherlands Cancer Institute) and maintained in RPMI1640 containing 10% fetal calf serum and 1 mg/ml G418 (Sigma)

Cell migration assay Cells were removed with cell dissociation buffer (Life Technologies) and suspended in serum-free DMEM. The cells were plated on cover glasses coated with either LM-511 (0.8 nM) or fibronectin (40 nM) and blocked with 1% BSA. After culturing for 1 hour, cells adhering to the substrata were treated with biologically active molecules, antibodies, or inhibitors. The cells were incubated in the presence of control DMSO or PMA for another hour. Biologically active molecules, antibodies, and inhibitors were solved with PBS(-) or DMSO. DMEM containing PBS(-) and DMSO was used as vehicle-treated control. The used monoclonal antibodies were classified into three kinds of immunoglobulin classes, mouse IgG1, IgG2a, and rat IgG2a. To concise the control experiments, control IgG subclasses were combined and used as control. Three hours post-plating, cell migration was monitored using a Biozero (Keyence, Osaka, Japan). Ten cells were randomly selected in the field. Video images were collected with a CCD camera at 10-min intervals using the BZ-Viewer and BZ-Analyzer (Keyence). The positions of nuclei were tracked to quantify cell motility. Velocities were calculated in micrometers per 8 hours using Image-J.

Quantification of cell morphology and immunocytochemistry As described above, A549 cells were removed with cell dissociation buffer and suspended in serum-free DMEM. LM-511 (0.8 nM) or fibronectin (40 nM)-coated cover glasses were blocked with 1%BSA in PBS(-), and A549 cells were plated on them. The cells adhering to substrata were incubated in the absence or presence of PMA for 3 hours. For quantification of cell morphology, the image of 20

7 cells captured using a Biozero were imported into Image-J to measure area of the cell bodies. For immunocytochemistry, the cells were fixed with 4% paraformaldehyde in PBS(-). The fixation was quenched with 0.1 M glycine buffer (pH 7.5). The fixed cells were permeabilized with 1% Triton X-100 in PBS(-) and blocked in 10% normal goat serum and then incubated with anti-vinculin monoclonal antibody to label focal adhesions. The bound antibody was detected with secondary antibody conjugated to Alexa488 (Life Technologies). Actin filaments and nuclei were stained with Alexa594 Phalloidin (Life Technologies) and Hoechst 33258 (Life Technologies). After several washes, sections were mounted in 90% glycerol containing 0.1x PBS and 1 mg/ml p-phenylenediamine. Images were captured using a FluoView FV1000D IX81 confocal laser scanning microscope system (Olympus, Tokyo, Japan).

Cell adhesion and inhibition assays Cell adhesion assays using LM-511 and fibronectin were performed as described previously [19]. Briefly, 96-well plates (Nunc, Roskilde, Denmark) were incubated with proteins at 4 °C overnight, and then blocked with PBS(-) containing 1% BSA for 1 hour at 37°C. The cells were cultured in serum-free DMEM with control DMSO or PMA for 1 hour and removed with cell dissociation buffer (Life Technologies), and then plated on the coated wells and incubated at 37°C for 1 hour. The alpha3A/K562 cells were only treated with control DMSO or Cytochalasin B to inhibit cell spreading completely. Adherent cells were fixed with 4% formaldehyde and stained with Diff-Quik (International Reagents Corp.). The stained cells were counted under the microscope. For inhibition assays, 96-well microtiter plates were coated with 0.8 nM of LM-511, and then blocked with PBS(-) containing 1% BSA for another hour. The cells were suspended in serum-free DMEM at a density of 4x105 cells/ml and preincubated with antibodies or/and control IgGs at room temperature for 10 min.

8 50 µl of cell suspension were added to wells coated with LM-511. After incubation at 37 °C for 0.5 hour, the attached cells were stained as described above.

Flow cytometric analysis A549 cells were cultured in serum free DMEM with control DMSO or PMA for 1 hour. The cells were removed with cell dissociation buffer and suspended in PBS(-) containing 0.1% BSA and 1 mM EDTA. The cells were incubated with antibodies for 1 hour at 4°C. After washing with PBS(-) containing 0.1% BSA and 1 mM EDTA, the cells were incubated with Alexa488-labeled secondary antibody for 1 hour at 4°C. The cells were then analyzed on a FACScalibur flow cytometer (Becton Dickinson, San Jose, CA).

Statistical analyses Velocities of cell migration and area of cell body are presented as box- and -whisker plot. Boxes indicate the range between upper and lower quartiles. Horizontal line represents median. Whiskers reach from maximum to minimum.

The results of cell adhesion and inhibition assays are

presented as mean ± Standard deviation. Statistical significance was determined using Dunnett’s multiple comparison test or t test, as indicated in the figure legends.

9 RESULTS Effects of biologically active molecules on cell migration on LM-511 Epithelial cells adhering to basement membrane are often exposed to biologically active molecules such as tumor-promoting agents. To examine effects of biologically active molecules on cells adhering to LM-511, cell migration assays were performed using A549 lung adenocarcinoma cells. Adhesion and migration of A549 cells on LM-511 have been well-characterized in previous studies [11, 13, 15, 16]. We chose biologically active molecules that possibly promote cell migration. Lysophosphatidic acid (LPA) and phorbol 12-myristate 13-acetate (PMA), Epidermal growth factor (EGF), Hepatocyte growth factor (HGF), or Transforming growth factor-β1 (TGF-β1) were added to the culture media and incubated for one hour to stimulate the cells. After the incubation, cell movements were traced for 8 hours using time-lapse video microscopy. Of the biologically active molecules, PMA was the only one to significantly promote the migration of A549 cells on LM-511 within the timeframe of the experiment (Fig. 1A). Although PMA also promoted cell migration on fibronectin, it more effectively enhanced cell migration on LM-511 (Fig. 1B). We also observed that MCF-7 breast carcinoma cells stimulated with PMA exhibited increased cell migration on LM-511, but the migration of A431 cells was not increased in the presence of PMA (Fig. 1C).

The morphology of PMA-stimulated A549 cells on LM-511 Control A549 cells exhibited a well-spread morphology on LM-511 or fibronectin-coated surfaces (Fig. 2A). Although PMA-stimulated A549 cells on fibronectin maintained a well-spread morphology with a greater cell-substratum contact area, the cells on LM-511 exhibited a rounded cell shape with a few short pseudopods. To quantify cell morphology, we measured the area of the cell bodies. As shown in Fig. 2B, PMA significantly disturbed cell spreading on LM-511, indicating that

10 the cells decreased their contact area with the substratum. Cell adhesion that is associated with cell motility requires actin rearrangements. We therefore examined the organization of actin filaments and focal adhesions in PMA-stimulated A549 cells on LM-511 and on fibronectin (Fig. 3). The cells were stained with phalloidin to label F-actin (filamentous actin). As described in previously, fibronectin induced typical actin stress fibers in A549 cells, and the actin was detected as peripheral spike-like protrusions when the cells were on LM-511 [11]. On the other hand, actin filaments were decreased in PMA-stimulated A549 cells plated on either substrate. The cells were also stained with anti-vinculin antibody to label focal adhesions. Fibronectin induced focal adhesions at the terminus of actin stress fibers. Although actin filaments in A549 cells on fibronectin were decreased by stimulation with PMA, the number of focal adhesions was increased. In the cells on LM-511, both actin stress fibers and focal adhesions were greatly reduced by PMA-stimulation. Thus, actin stress fibers and focal adhesions seem to be associated with the static state of cells rather than with cell migration.

The adhesion of PMA-stimulated A549 cells to LM-511 To investigate the rounded cell shape on LM-511, cell adhesion assays were performed using PMA-stimulated A549 cells (Fig. 4A). A549 cells were grown to confluency in growth media and then stimulated in serum-free DMEM containing PMA for 2 hours.

The control and

PMA-stimulated cells were plated on wells coated with LM-511 and incubated for 1 hour. The attachment of PMA-treated A549 cells to LM-511 was weaker than that of control cells (Fig. 4A). The rounded cell shape seemed to be due to the reduced adhesion of PMA-stimulated cells to LM-511. We also examined if PMA altered the expression of laminin receptors. Flow cytometric analysis showed

11 that the treatment of PMA did not influence the expression of laminin receptors such as integrins and Lu/B-CAM (Fig. 5). LM-511 is a potent cell-adhesive protein capable of being recognized by both α3β1 and α6β1 integrins, both major cell surface receptors [20]. The adhesion of A549 cells to LM-511 depends predominantly on integrin α3β1 [10]. Inhibition assays using function-blocking antibodies showed that the adhesion of PMA-stimulated A549 cells to LM-511 still depended on integrin α3β1 (Fig. 4B). Although mAb 87207 to Lu/B-CAM alone had no effect on adhesion of control cells to LM-511 [13], the adhesion of PMA-stimulated A549 cells to LM-511 was slightly reduced in the presence of the antibody. Our previous study also showed that mAb 87207 to Lu/B-CAM attenuates the inhibitory effects of anti-integrin α3 and β1 mAbs. However the attenuation activity of this antibody was not observed in the adhesion of PMA-stimulated A549 cells to LM-511. We further examined if PMA-stimulation influenced receptor binding using K562 transfectants expressing α3β1 (alpha3A/K562). As shown in Figure 4C, PMA-stimulation reduced the adhesion of alpha3A/K562 cells to LM-511. The rounded cell shape seems to be due to the reduced adhesion of PMA-stimulated cells to LM-511 through integrin α3β1.

The involvement of laminin receptors in cell migration promoted by PMA We also examined whether integrin α3β1 and Lu/B-CAM are involved in cell migration promoted by PMA. Cell migration assays were performed in the presence of monoclonal antibodies against integrin subunits.

The antibodies against integrin α3 and β1, but not against 6, inhibited the

migration of PMA-stimulated A549 cells on LM-511 (Fig. 6A).

We next examined whether

Lu/B-CAM was involved in cell migration on LM-511 using mAb 87207 (Fig. 6B). The antibody decreased the velocity of PMA-stimulated A549 cells to the control level observed with no

12 PMA-stimulation and no inhibitory antibody. Our previous study showed that the binding of a high affinity form of integrin α3β1 to LM-511 impairs cell migration. To activate integrin α3β1, we used an antibody against integrin β1 (TS2/16) that stimulates cell adhesion to extracellular matrix proteins, including LM-511 [19]. Similar to the effects of mAb 87207, TS2/16 antibody decreased the velocity of PMA-stimulated A549 cells to the control level. However, neither of these antibodies could reduce the velocity of cell migration to the basal level completely when PMA was used. Our results show that competitive binding of Lu/B-CAM and low affinity integrin α3β1 promotes PMA-induced cell migration on LM-511.

Signal transduction in PMA-stimulated cell migration on LM-511 Small GTPases of the Rho family (Rho, Rac, and Cdc42) are involved in integrin signal transduction and influence the cytoskeleton’s arrangement [21, 22]. We examined if these regulators were involved in PMA-stimulated cell migration on LM-511 using specific inhibitors (Fig. 7A). Y27632 and GSK429286 were used as inhibitors of Rho signaling, because both reagents inhibit ROCK, which is a downstream target of Rho GTPase. Although the Cdc42 inhibitor did not influence PMA-stimulated cell migration on LM-511, cell migration was significantly reduced by Rac and ROCK inhibitors. Y27632 and GSK429286 also decreased control cell migration on LM-511. The results show that cell migration on LM-511 is mainly modulated via Rho/ROCK pathway. ROCK is a key regulator of actin organization; it phosphorylates LIM kinase, myosin light chain (MLC) and MLC phosphatase [23]. Cytochalasin B, which inhibits actin polymerization, drastically decreased the migration of PMA-stimulated A549 cells on LM-511 (Fig. 7B). Although blebbistatin, an inhibitor of nonmuscle myosin II, did not influence cell migration on LM-511, LIMKi3 slightly inhibited the migration of PMA-stimulated A549 cells (Fig. 7B).

13 The effect of the inhibitors on the morphology of PMA-stimulated A549 cells was investigated (Fig. 8). Y27632 induced a flattened cell shape that is similar to control A549 cells on LM-511. The cell morphology with stable adhesion seems to impair cell migration on LM-511. Even if the inhibitors of myosin II and LIM kinase were combined, the morphology of PMA-stimulated A549 cells still exhibited a rounded or spindle cell shape. Although the TS2/16 antibody alone did not influence the morphology of PMA-stimulated A549 cells, the antibody induced a flattened cell shape in the presence of myosin II and LIMK inhibitors.

These results show that the ROCK pathway

down-regulates the binding of integrin α3β1 to LM-511.

14 DISCUSSION In this study we show that a tumor-inducing agent, PMA, promotes the migration of A549 lung adenocarcinoma cells on LM-511. PMA is a phorbol ester that stimulates protein kinase C to modulate various downstream signaling pathways.

We also observed that MCF-7 breast

adenocaricinoma cells stimulated with PMA show increased migration on LM-511. However, A431 epidermoid carcinoma cells did not show increased motility in the presence of PMA. These cells may require more genetic alterations to be tumor-initiated. Several previous studies have reported that LPA, EGF, HGF, and TGF-β1 promote migration of A549 cells [24-27]. However, the migration of A549 cells on LM-511 was not promoted by treatment with these factors. LM-511 may thus serve as an anti-tumor molecule that can promote resistance to carcinogenesis caused by growth factor stimulation. We also show that PMA immediately induces a rounded cell shape on LM-511. Although it looks like epithelial-mesenchymal transition (EMT), it is unclear if PMA-stimulation is involved in EMT events on A549 cells adhering to LM-511. Previous study showed that EMT events are induced on A549 cells with cell-cell contacts [28]. In the future it will be necessary to examine the effect of PMA in suitable culture condition for inducing EMT and the expression of mesenchymal markers. Although TGF-β is well known to induce EMT events in A549 cells [29], this phenotypic transition requires a few days. The migration of mesenchymal-like cells induced by TGF-β1 may be facilitated by culture on LM-511. Consistent with the previous report [11], actin filaments of intermediate length were observed in A549 cells adhering to LM-511, as opposed to the longer filaments when adhering to fibronectin. The actin filaments were further shortened and mostly disappeared in PMA-stimulated A549 cells on LM-511; focal adhesions were also mostly absent from these cells. These results show that cell migration on LM-511 is enhanced, with a concomitant suppression of actin stress fibers and focal adhesion formation. On the other hand, the formation of actin stress fibers and focal adhesions in

15 the cells adhering to fibronectin seems to contribute to the static state. Because laminins potently promote cell migration, several groups have tried to identify a cellular apparatus mediating cell adhesion and migration, but it has not been discovered yet. A specific cellular apparatus may be unnecessary for cell migration on LM-511. Our results showed that the stimulation of PMA down-regulates the adhesion of A549 cells to LM-511. As consist with common observations of multiple papers, the migration of A549 cells was dependent on the strength of cell adhesion to LM-511. Cell adhesion assays using the transfectants showed that this is due to the down-regulation of integrin α3β1 binding. Previously we showed that Lu/B-CAM acts to disturb integrin-mediated cell attachment to LM-511 rather than to promote cell anchoring [13]. In this study, however, Lu/B-CAM partially contributed to the cells anchoring to LM-511. Because the binding of integrin α3β1 is reduced with the stimulation of PMA, the role of Lu/B-CAM seems to increase in adhesion to LM-511. In accord with the results of inhibition assays using antibodies, PMA-stimulated cell migration was significantly inhibited in the presence of integrin α3β1 or Lu/B-CAM antibody. However the antibodies did not reduce the velocity of cell migration to the basal (non-PMA-stimulated) level. The active form of integrin α3β1 contributes to the static state rather than to a migratory behavior [13]. These results suggest that integrin α3β1 is in an intermediately active state on PMA-stimulated A549 cells. The activation potency of the TS2/16 antibody may be insufficient to suppress PMA-stimulated cell migration on LM-511 completely. In addition to binding LM-511, integrin α3β1 also binds to LM-332 [30]. A549 cells anchoring to LM-332 showed similarly increased motility in the presence of PMA (data not shown). Although LM-332 is restrictively distributed in adult tissues, it would also be a scaffold for tumor cells that become stimulated with biologically active molecules.

16 There are two types of integrin-related signal transduction [31]. The first is direct signaling in which stimulation of integrins by extracellular proteins triggers intracellular signaling events. The second is integrin modulation of inside-out signaling, for instance, in which a signaling pathway activated by growth factors influences integrin-mediated cell anchorage. In general, integrin direct signaling leads to accumulation of phosphorylated proteins and cytoskeletal molecules at adhesion sites. The binding of integrin α3β1 to LM-511 activates Rac via the p130(Cas)-CrkII-DOCK180 pathway [11]. Cell adhesion to LM-511 prevents apoptosis via Csk, phosphatidylinositol 3-kinase/Akt- and MEK1/ERK-dependent pathways [15, 16]. However, despite potent cell adhesion activity, studies of outside-in signaling activated by laminins (including LM-511) have not progressed. Rather than the occurrence of outside-in signaling, cell adhesion to LM-511 seems to act with inside-out signaling (Fig. 9). In previous integrin works PMA-stimulation has been shown to activate integrins and increase cell adhesion [32]. Two groups reported that PMA-stimulation leads to phosphorylation of the cytoplasmic domain of integrin α3 [33, 34]. Therefore, our results suggest that the phosphorylated integrin α3β1 is a low affinity form for LM-511. On the other hand, Zhang et al. showed that cell spreading on LM-332 is mediated through the phosphorylation of integrin α3 [34]. Together with, the phosphorylation level of integrin α3 seems to modulate the binding of ligand. Our results also suggest that the ROCK pathway leads to the phosphorylation of integrin α3β1 directly or indirectly. Zhang et al. reported that activation of PKC by PMA occurs in parallel with PKC translocation to cellular membranes, and activated PKC further associates with transmembrane-4 superfamily (TM4SF) [35]. Of TM4SF proteins, CD151 is constitutively associated with integrin α3β1 and involved in tumorigenesis [36, 37].

Because PKC

dose not phosphorylate integrin α3β1 directly [34], PKC and CD151 complex may modulate the ROCK pathway leading to phosphorylation of integrin α3β1. As described above, the opposite effects of PMA

17 are observed in A549 cells on fibronectin and LM-511 substrates. It seems to be due to the activation of fibronectin-binding integrins and suppression of integrin α3β1 through inside-out signaling. Recent stem cell studies have reported that LM-511 and its derivatives are substrata permitting sustained self-renewal of human embryonic stem cells (hESCs) and induced pluripotent stem cells (hiPSCs) in xeno-free culture conditions [38, 39]. The adhesion of hESCs and hiPSCs to LM-511 or its derivatives is mediated through integrin α6β1. Because Y27632 improves the adhesion of hESCs on feeder cells and permits cell survival [40], the inhibitor is often added to stem cell culture media. However, Y27632 does not improve the adhesion of hESCs and hiPSCs to the LM-511 E8 domain that is recognized by integrins [39], suggesting that the ROCK pathway dose not link to cell adhesion mediated through integrin α6β1. Based on our results, the ROCK pathway specifically modulates the binding of integrin α3β1 and seems to be involved in a transition from static to migratory cell behaviors on LM-511.

18 ACKNOWLEDGMENTS This work was supported in part by grants to Y.K. (26112718, 26430123) from the Ministry of Education, Sciences Sports and Culture, Japan. We thank Dr. Jeffrey H. Miner for comments on the manuscript. We also thank Ms. Shiori Uda and Ms. Momoko Nakao for technical assistance.

DISCLOSURE STATEMENT The authors have no conflict of interest.

19 REFERENCES [1]

D. Hanahan, R.A. Weinberg, Hallmarks of cancer: the next generation, Cell 144 (2011)

646-674. [2]

R. Kalluri, M. Zeisberg, Fibroblasts in cancer, Nat Rev Cancer 6 (2006) 392-401.

[3]

M.W. Pickup, J.K. Mouw, V.M. Weaver, The extracellular matrix modulates the hallmarks of cancer, EMBO Rep 15 (2014) 1243-1253.

[4]

J.H. Miner, P.D. Yurchenco, Laminin functions in tissue morphogenesis, Annu Rev Cell Dev Biol 20 (2004) 255-284.

[5]

H.K. Kleinman, G.R. Martin, Matrigel: basement membrane matrix with biological activity, Semin Cancer Biol 15 (2005) 378-386.

[6]

M. Aumailley, L. Bruckner-Tuderman, W.G. Carter, R. Deutzmann, D. Edgar, P. Ekblom, J. Engel, E. Engvall, E. Hohenester, J.C. Jones, H.K. Kleinman, M.P. Marinkovich, G.R. Martin, U. Mayer, G. Meneguzzi, J.H. Miner, K. Miyazaki, M. Patarroyo, M. Paulsson, V. Quaranta, J.R. Sanes, T. Sasaki, K. Sekiguchi, L.M. Sorokin, J.F. Talts, K. Tryggvason, J. Uitto, I. Virtanen, K. von der Mark, U.M. Wewer, Y. Yamada, P.D. Yurchenco, A simplified laminin nomenclature, Matrix Biol. 24 (2005) 326-332.

[7]

M. Durbeej, Laminins, Cell Tissue Res 339 (2010) 259-268.

[8]

J.H. Miner, B.L. Patton, S.I. Lentz, D.J. Gilbert, W.D. Snider, N.A. Jenkins, N.G. Copeland, J.R. Sanes, The laminin alpha chains: expression, developmental transitions, and chromosomal locations of alpha1-5, identification of heterotrimeric laminins 8-11, and cloning of a novel alpha3 isoform, J. Cell Biol. 137 (1997) 685-701.

[9]

L.M. Sorokin, F. Pausch, M. Frieser, S. Kroger, E. Ohage, R. Deutzmann, Developmental regulation of the laminin alpha5 chain suggests a role in epithelial and endothelial cell

20 maturation, Dev. Biol. 189 (1997) 285-300. [10]

Y. Kikkawa, N. Sanzen, K. Sekiguchi, Isolation and characterization of laminin-10/11 secreted by human lung carcinoma cells. laminin-10/11 mediates cell adhesion through integrin alpha3 beta1, J. Biol. Chem. 273 (1998) 15854-15859.

[11]

J. Gu, Y. Sumida, N. Sanzen, K. Sekiguchi, Laminin-10/11 and fibronectin differentially regulate integrin-dependent Rho and Rac activation via p130(Cas)-CrkII-DOCK180 pathway, J Biol Chem 276 (2001) 27090-27097.

[12]

N. Pouliot, N. Kusuma, Laminin-511: A multi-functional adhesion protein regulating cell migration, tumor invasion and metastasis, Cell Adh Migr 7 (2013) 142-149.

[13]

Y. Kikkawa, T. Ogawa, R. Sudo, Y. Yamada, F. Katagiri, K. Hozumi, M. Nomizu, J.H. Miner, The lutheran/basal cell adhesion molecule promotes tumor cell migration by modulating integrin-mediated cell attachment to laminin-511 protein, J Biol Chem 288 (2013) 30990-31001.

[14]

V. Givant-Horwitz, B. Davidson, R. Reich, Laminin-induced signaling in tumor cells, Cancer Lett 223 (2005) 1-10.

[15]

J. Gu, A. Fujibayashi, K.M. Yamada, K. Sekiguchi, Laminin-10/11 and fibronectin differentially prevent apoptosis induced by serum removal via phosphatidylinositol 3-kinase/Akt- and MEK1/ERK-dependent pathways, J Biol Chem 277 (2002) 19922-19928.

[16]

J. Gu, S. Nada, M. Okada, K. Sekiguchi, Csk regulates integrin-mediated signals: involvement of differential activation of ERK and Akt, Biochem Biophys Res Commun 303 (2003) 973-977.

[17]

J. Gao, M.C. DeRouen, C.H. Chen, M. Nguyen, N.T. Nguyen, H. Ido, K. Harada, K. Sekiguchi, B.A. Morgan, J.H. Miner, A.E. Oro, M.P. Marinkovich, Laminin-511 is an epithelial message

21 promoting dermal papilla development and function during early hair morphogenesis, Genes Dev 22 (2008) 2111-2124. [18]

L. Ritie, C. Spenle, J. Lacroute, A.L. Bolcato-Bellemin, O. Lefebvre, C. Bole-Feysot, B. Jost, A. Klein, C. Arnold, M. Kedinger, D. Bagnard, G. Orend, P. Simon-Assmann, Abnormal Wnt and PI3Kinase signaling in the malformed intestine of lama5 deficient mice, PLoS One 7 (2012) e37710.

[19]

Y. Kikkawa, T. Sasaki, M.T. Nguyen, M. Nomizu, T. Mitaka, J.H. Miner, The LG1-3 tandem of laminin alpha5 harbors the binding sites of Lutheran/basal cell adhesion molecule and alpha3beta1/alpha6beta1 integrins, J Biol Chem 282 (2007) 14853-14860.

[20]

Y. Kikkawa, N. Sanzen, H. Fujiwara, A. Sonnenberg, K. Sekiguchi, Integrin binding specificity of laminin-10/11: laminin-10/11 are recognized by alpha 3 beta 1, alpha 6 beta 1 and alpha 6 beta 4 integrins, J Cell Sci 113 ( Pt 5) (2000) 869-876.

[21]

P. Friedl, K. Wolf, M.M. Zegers, Rho-directed forces in collective migration, Nat Cell Biol 16 (2014) 208-210.

[22]

M.M. Zegers, P. Friedl, Rho GTPases in collective cell migration, Small GTPases 5 (2014) e28997.

[23]

N. Rath, M.F. Olson, Rho-associated kinases in tumorigenesis: re-considering ROCK inhibition for cancer therapy, EMBO Rep 13 (2012) 900-908.

[24]

E.K. Kim, S.J. Yun, J.M. Ha, Y.W. Kim, I.H. Jin, D.H. Woo, H.S. Lee, H.K. Ha, S.S. Bae, Synergistic

induction

of

cancer

cell

migration

regulated

by

Gbetagamma

and

phosphatidylinositol 3-kinase, Exp Mol Med 44 (2012) 483-491. [25]

C.R. Hauck, D.J. Sieg, D.A. Hsia, J.C. Loftus, W.A. Gaarde, B.P. Monia, D.D. Schlaepfer, Inhibition of focal adhesion kinase expression or activity disrupts epidermal growth

22 factor-stimulated signaling promoting the migration of invasive human carcinoma cells, Cancer Res 61 (2001) 7079-7090. [26]

S. Radtke, M. Milanovic, C. Rosse, M. De Rycker, S. Lachmann, A. Hibbert, S. Kermorgant, P.J. Parker, ERK2 but not ERK1 mediates HGF-induced motility in non-small cell lung carcinoma cell lines, J Cell Sci 126 (2013) 2381-2391.

[27]

J. Liu, G. Hu, D. Chen, A.Y. Gong, G.S. Soori, T.J. Dobleman, X.M. Chen, Suppression of SCARA5 by Snail1 is essential for EMT-associated cell migration of A549 cells, Oncogenesis 2 (2013) e73.

[28]

M. Kawata, D. Koinuma, T. Ogami, K. Umezawa, C. Iwata, T. Watabe, K. Miyazono, TGF-beta-induced epithelial-mesenchymal transition of A549 lung adenocarcinoma cells is enhanced by pro-inflammatory cytokines derived from RAW 264.7 macrophage cells, J Biochem 151 (2012) 205-216.

[29]

K. Miyazono, Transforming growth factor-beta signaling in epithelial-mesenchymal transition and progression of cancer, Proc Jpn Acad Ser B Phys Biol Sci 85 (2009) 314-323.

[30]

K. Miyazaki, Laminin-5 (laminin-332): Unique biological activity and role in tumor growth and invasion, Cancer Sci 97 (2006) 91-98.

[31]

P. Keely, L. Parise, R. Juliano, Integrins and GTPases in tumour cell growth, motility and invasion, Trends Cell Biol 8 (1998) 101-106.

[32]

C. Gimond, A. Sonnenberg, Activation States of Integrins, Integrin-Ligand Interaction, Springer US, Boston, MA, 1997, pp. 219-240.

[33]

G.O. Delwel, F. Hogervorst, I. Kuikman, M. Paulsson, R. Timpl, A. Sonnenberg, Expression and function of the cytoplasmic variants of the integrin alpha 6 subunit in transfected K562 cells. Activation-dependent adhesion and interaction with isoforms of laminin, J Biol Chem

23 268 (1993) 25865-25875. [34]

X.A. Zhang, A.L. Bontrager, C.S. Stipp, S.K. Kraeft, G. Bazzoni, L.B. Chen, M.E. Hemler, Phosphorylation of a conserved integrin alpha 3 QPSXXE motif regulates signaling, motility, and cytoskeletal engagement, Mol Biol Cell 12 (2001) 351-365.

[35]

X.A. Zhang, A.L. Bontrager, M.E. Hemler, Transmembrane-4 superfamily proteins associate with activated protein kinase C (PKC) and link PKC to specific beta(1) integrins, J Biol Chem 276 (2001) 25005-25013.

[36]

L.M. Sterk, C.A. Geuijen, J.G. van den Berg, N. Claessen, J.J. Weening, A. Sonnenberg, Association of the tetraspanin CD151 with the laminin-binding integrins alpha3beta1, alpha6beta1, alpha6beta4 and alpha7beta1 in cells in culture and in vivo, J Cell Sci 115 (2002) 1161-1173.

[37]

H.X. Wang, Q. Li, C. Sharma, K. Knoblich, M.E. Hemler, Tetraspanin protein contributions to cancer, Biochem Soc Trans 39 (2011) 547-552.

[38]

S. Rodin, L. Antonsson, O. Hovatta, K. Tryggvason, Monolayer culturing and cloning of human pluripotent stem cells on laminin-521-based matrices under xeno-free and chemically defined conditions, Nat Protoc 9 (2014) 2354-2368.

[39]

T. Miyazaki, S. Futaki, H. Suemori, Y. Taniguchi, M. Yamada, M. Kawasaki, M. Hayashi, H. Kumagai, N. Nakatsuji, K. Sekiguchi, E. Kawase, Laminin E8 fragments support efficient adhesion and expansion of dissociated human pluripotent stem cells, Nat Commun 3 (2012) 1236.

[40]

K. Watanabe, M. Ueno, D. Kamiya, A. Nishiyama, M. Matsumura, T. Wataya, J.B. Takahashi, S. Nishikawa, S. Nishikawa, K. Muguruma, Y. Sasai, A ROCK inhibitor permits survival of dissociated human embryonic stem cells, Nat Biotechnol 25 (2007) 681-686.

24 [41]

L. Xiao, M. Eto, M.G. Kazanietz, ROCK mediates phorbol ester-induced apoptosis in prostate cancer cells via p21Cip1 up-regulation and JNK, J Biol Chem 284 (2009) 29365-29375.

25 FIGURE LEGENDS Fig. 1. Effects of PMA on cell migration on LM-511. (A) Effects of biologically active molecules. Substrata were coated with LM-511 (0.8 nM), and A549 cells were incubated in the presence of Lysophosphatidic acid (LPA, 3 μM), Phorbol 12-myristate 13-acetate (PMA, 100 nM), Epidermal growth factor (EGF, 60 ng/ml), Hepatocyte growth factor (HGF, 60 ng/ml), and Transforming growth factor β1 (TGF-β, 60 ng/ml). Biologically active molecules were solved in PBS(-) or DMSO and diluted in DMEM. The media containing PBS(-) or DMSO were used as vehicle-treated control. The diluted vehicle did not influence cell behaviors. Cell movements were monitored by time-lapse video microscopy. Quantification of cell motility was evaluated by velocity as described under “Materials and Methods.” *, P<0.01 by Dunnett’s multiple comparison test; n=10. (B) Dose dependency of PMA. Substrata were coated with LM-511 (0.8 nM) or fibronectin (40 nM), and A549 cells were incubated in the presence of PMA (0, 1, 10, 100 nM). (C) Effect of PMA on the other cells plated on LM-511. A431 and MCF-7 cells were attached to LM-511 (0.8 nM) and incubated without or with PMA (100 nM).

Fig. 2. Effect of PMA on cell morphology. (A) Morphology of PMA-stimulated A549 cells on LM-511. The cells were attached to substrata coated with LM-511 (0.8 nM, upper panel) or fibronectin (40 nM, lower panel) and incubated for 3 hours in the absence (left panel) or presence (right panel) of PMA (100 nM). (B) Quantification of cell morphology. Images of cells were captured and used to measure cell area. *, P<0.01 by Dunnett’s multiple comparison test; n=20.

Fig. 3. Localization of actin and vinculin in PMA-stimulated A549 cells on LM-511. Cells were attached to substrata coated with LM-511 (0.8 nM) or fibronectin (40 nM) and incubated for 3 hours in the presence of PMA (100 nM). The cells were stained with phalloidin, anti-vinculin antibody, and

26 Hoechst 33258 for actin filaments (red), focal adhesions (green), and nuclei (blue), respectively. PMA-stimulation disturbed the formation of actin stress fibers in A549 cells.

Fig. 4. Analysis of cell adhesion to LM-511. (A) Adhesion of PMA-stimulated A549 cells to LM-511. A549 cells were cultured in serum-free DMEM with control DMSO or PMA for 1 hour. The cells were non-enzymatically removed from culture dishes.

96-well plates were coated with increasing

concentrations of the proteins and incubated with control cells (closed circles), or PMA-stimulated cells (closed squares) at 37 °C for 1 hour. Adherent cells were stained with Diff-Quik and counted. Each point represents the mean of triplicate assays. Bars, standard deviation. *, P<0.01 by t test. (B) Inhibitory effects of integrins and Lu/B-CAM mAbs on the adhesion of PMA-stimulated A549 cells to LM-511. Cells pre-incubated with function-blocking antibodies against the indicated integrin subunits or/and Lu-B-CAM were added to LM-511-coated wells. Control IgG subclasses were combined and used as control IgGs. High concentration of control IgGs did not influence the cell behaviors. After incubation for 30 min, the attached cells were stained and counted. Values are expressed as percentages of the number of cells adhering in the absence of antibody. Each column represents the mean of triplicate assays. Bars, standard deviation. *, P<0.05; **, P<0.10 by t test. (C) Adhesion of PMA-stimulated α3A/K562 cells to LM-511. The cells were cultured in serum free DMEM with control DMSO or PMA for 1 hour. Control cells (closed circles) or PMA-stimulated transfectants (closed squares) were plated on LM-511-coated wells and incubated in the presence of Cytochalasin B (1 μg/ml) at 37 °C for 1 hour. Adherent cells were counted as described above. Bars, standard deviation. *, P<0.01 by t test.

Fig. 5. Flow cytometric analyses of integrin α3/α6/β1 and Lu/B-CAM expression. The expression of

27 integrin α3/α6/β1 and Lu/B-CAM is shown as a solid line (control) and a dotted line (PMA). The gray area indicates the negative control.

Stimulation with PMA did not reduce the expression of

LM-511-binding integrins α3β1 and Lu/B-CAM.

Fig. 6. Inhibitory effects of anti-laminin receptor mAbs on cell migration. (A) Migration of A549 cells on LM-511 in the presence of function blocking antibodies to integrins. The cells were incubated with the indicated control IgG subclasses (20 μg/ml) or antibodies (10 μg/ml) and plated on wells coated with LM-511 (0.8 nM). The adherent cells were incubated in the presence of control DMSO or PMA (100 nM). Cell movements were tracked at 10-min intervals over a span of 8 hours. Cell motility was evaluated by velocity. *, P<0.01 by Dunnett’s multiple comparison test; n=10. (B) Migration of A549 cells on LM-511 in the presence of anti- Lu/B-CAM (87207) or integrin β1 (TS2/16 [activating]) mAb. After cells adhered to LM-511, indicated control IgG subclasses (20 μg/ml) or the antibodies (10 μg/ml) were added to the media and incubated in the absence or presence of PMA. Cell motility was measured as described above.

Fig. 7. Effects of inhibitors of Rho family signaling pathway and stress fiber organization on cell migration and morphology. Migration of A549 cells on LM-511 was assayed in the presence of inhibitors of Rho family signaling pathway (A) and actin stress fiber organization (B). The cells were attached to substrata coated with LM-511 (0.8 nM) and treated with ML141 (cdc42 inhibitor, 10 μM), NSC237661 (Rac1 inhibitor, 10 μM), Y27632 (ROCK inhibitor, 10 μM), GSK429286 (ROCK inhibitor, 10 μM), (Cytochalasin B (CytoB, actin polymerization inhibitor, 1μg/ml), Blebbistatin (Bleb, Myosin II inhibitor, 15 μM) or LIMKi3 (LIM-kinase inhibitor, 20 μM). After the treatment with inhibitors, the cells were incubated in the absence or presence of PMA (100 nM). Reagents were solved

28 in PBS(-) or DMSO and diluted in DMEM. Assay media containing PBS(-) and DMSO were used as vehicle-treated control. Cell movements were monitored by time-lapse video microscopy as described above.

Cell motility was evaluated by velocity. *, P<0.01; **, P<0.05 by Dunnett’s multiple

comparison test; n=10.

Fig. 8. Effects of inhibitors acting downstream of ROCK signaling and anti-integrin antibody on cell morphology on LM-511. Morphology of PMA-stimulated A549 cells on LM-511 in the presence of the indicated inhibitors, Y27632 (10 μM), Blebbistatin (15 μM), LIMKi3 (20 μM) and/or anti-integrin β1 (TS2/16 [activating]) mAb (10 μg/ml). After cells adhered to LM-511, the inhibitor and/or antibody were added to the media and then incubated in the absence or presence of PMA for 3 hours.

Fig. 9. Schematic of signaling pathways modulating cell migration on LM-511.

PMA stimulates

protein kinase C (PKC) and leads to activation of the Rho/ROCK pathway [41].

Overdriven

Rho/ROCK pathway disturbs actin filament stabilization mediated through LIMK and MLC. The activation of the Rho/ROCK pathway phosphorylates integrin α3β1, directly or indirectly (indicated by the gray arrow), and results in weak cell adhesion to LM-511. The shortened actin filaments and weakened receptor binding promote tumor cell migration on LM-511.

29

30

31

32

33

34

35

36

37

38

Highlights PMA-stimulation significantly promotes the migration of lung adenocarcinoma A549 cells on laminin-511 (LM-511) The cell migration is associated with weakened cell adhesion to LM-511 via integrin α3β1 ROCK signaling pathway modulates tumor cell migration on LM-511