Short Report 2141 VANGL2 regulates membrane trafficking of MMP14 to control cell polarity and migration

B. Blairanne Williams1, V. Ashley Cantrell1, Nathan A. Mundell1, Andrea C. Bennett1, Rachel E. Quick1 and Jason R. Jessen1,2,* 1Division of Genetic Medicine/Department of Medicine, and 2Department of Cancer Biology, Vanderbilt University Medical Center, Nashville, TN 37232, USA *Author for correspondence ([email protected])

Accepted 11 January 2012 Journal of Cell Science 125, 2141–2147 ß 2012. Published by The Company of Biologists Ltd doi: 10.1242/jcs.097964

Summary Planar cell polarity (PCP) describes the polarized orientation of cells within the plane of a tissue. Unlike epithelial PCP, the mechanisms underlying PCP signaling in migrating cells remain undefined. Here, the establishment of PCP must be coordinated with dynamic changes in cell adhesion and (ECM) organization. During gastrulation, the membrane type-1 matrix (MT1-MMP or MMP14) is required for PCP and convergence and extension cell movements. We report that the PCP protein Vang-like 2 (VANGL2) regulates the endocytosis and cell-surface availability of MMP14 in manner that is dependent on focal adhesion kinase. We demonstrate that zebrafish trilobite/vangl2 mutant embryos exhibit increased Mmp14 activity and decreased ECM. Furthermore, in vivo knockdown of Mmp14 partially rescues the Vangl2 loss-of-function convergence and extension phenotype. This study identifies a mechanism linking VANGL2 with MMP14 trafficking and suggests that establishment of PCP in migrating gastrula cells requires regulated proteolytic degradation or remodeling of the ECM. Our findings implicate matrix as downstream effectors of PCP and suggest a broadly applicable mechanism whereby VANGL2 affects diverse morphogenetic processes.

Key words: VANGL2, MMP14 (MT1-MMP), Planar cell polarity, Gastrulation, Fibronectin, Migration

Introduction it is unknown whether PCP signaling proteins themselves It has been a decade since homologues of fly planar cell polarity regulate degradation and remodeling of the ECM. Our previous

Journal of Cell Science (PCP) were shown to be required for convergence and work used HT-1080 cells to probe the relationship between extension movements during frog and fish gastrulation (Djiane human VANGL2, MMP14 and cell–ECM interactions. et al., 2000; Wallingford et al., 2000; Jessen et al., 2002). Knockdown of VANGL2 using siRNA increased secreted Underlying these movements are polarized cell behaviors that levels of active MMP2 and promoted invasion through an ECM contribute to mediolateral narrowing and anterior–posterior substrate (Cantrell and Jessen, 2010). Because MMP2 is activated elongation of the developing embryo. Loss of PCP by MMP14 at the cell surface (Sato et al., 1994), we hypothesized function in vivo disrupts polarity, as indicated by changes in that Vangl2 might influence PCP in migrating cells by directly cell elongation, orientation and protrusive activity (Wallingford regulating Mmp14 activity. In this report, we identify VANGL2 et al., 2000; Topczewski et al., 2001; Jessen et al., 2002). As a as a regulator of MMP14 endocytosis and demonstrate that result of altered cell polarity, zebrafish mutants including this membrane-tethered metalloproteinase acts downstream of trilobite/vang-like 2 (vangl2) exhibit a distinct convergence and Vangl2 to control zebrafish convergence and extension cell extension phenotype characterized by shortened and broadened movements. body axes (Sepich et al., 2000; Jessen et al., 2002). Although it is thought that Wnt-mediated PCP signaling impacts the actin Results and Discussion cytoskeleton to regulate polarity and protrusion dynamics in VANGL2 regulates MMP14 endocytosis migrating cells, few downstream effector proteins have been Both endocytosis and recycling of MMP14 provide possible identified. Previously, we showed that zebrafish membrane type- mechanisms to control the level of proteolytic activity in 1 (Mmp14) is also required for cell polarized cells (Jiang et al., 2001; Remacle et al., 2003; polarity and convergence and extension movements exhibiting a Poincloux et al., 2009). To determine whether VANGL2 genetic interaction with the Wnt co-receptor knypek/glypican4 regulates MMP14 trafficking, we performed an in vitro (Coyle et al., 2008). The temporal requirement for Mmp14 biotinylation assay using a cleavable form of biotin to quantify function during late gastrulation coincides with both that of cell-surface, internalized and recycled MMP14 levels in cells Vangl2 (Sepich et al., 2000) and the appearance of a fibrillar transfected with siRNA against VANGL2. As shown in Fig. 1A, extracellular matrix (ECM) meshwork (Latimer and Jessen, VANGL2-knockdown cells had increased levels of cell-surface 2010), suggesting a functional relationship between cell polarity MMP14 but not total MMP14 protein (supplementary material and proteolysis. However, despite previous connections between Fig. S1). After we stimulated endocytosis, VANGL2-knockdown PCP and ECM assembly (Goto et al., 2005; Dzamba et al., 2009), cells had reduced levels of internalized MMP14 (Fig. 1B,C). To 2142 Journal of Cell Science 125 (9)

Fig. 1. VANGL2 regulates MMP14 endocytosis in vitro. (A) Quantification of cell-surface MMP14 in cells transfected with non-targeting control (NT) siRNA or siRNA to knock down VANGL2. MESNA treatment removes cell-surface biotin. Graph of representative experimental data normalized to GAPDH expression from total protein extracts. (B) Endocytosis was quantified after shifting the temperature to 37˚C and MESNA treatment to remove non-internalized proteins. MMP14 recycling was quantified after incubating the cells an additional hour followed by a second MESNA treatment. Graph shows representative data from one experiment. (C) Quantification of all internalization and recycling data from three independent experiments. Graph depicts the mean percentage of internalized MMP14 (total intracellular MMP14 divided by total cell-surface MMP14; NT siRNA594.6%, VANGL2 siRNA548.6%) and recycled MMP14 (total intracellular after endocytosis divided by total intracellular after 1 hour of recycling; NT siRNA, 25%; VANGL2 siRNA, 33%) with s.d. (D) Endocytosis assay performed

Journal of Cell Science using an antibody to label cell-surface MMP14 at 4˚C followed by a temperature shift to 37˚C. Although more MMP14 is observed at the plasma membrane in VANGL2-knockdown cells (arrows), less is internalized compared with controls. (E) Transferrin endocytosis following VANGL2 knockdown. HT-1080 cells transfected with NT or VANGL2 siRNA were labeled at 4˚C with Alexa-Fluor-594-labeled transferrin followed and shifted to 37˚C. Similar numbers of transferrin-positive intracellular vesicles were observed between control and VANGL2-knockdown cells (mean number of vesicles counted per cell with s.d. from three independent experiments; NT538614, n5186 cells; VANGL2543611, n5221 cells).

test whether MMP14 recycling is impaired, after endocytosis Fig. S2). These data indicate that VANGL2 specifically regulates cells were treated with MESNA and incubated at 37˚C for an endocytosis of MMP14; however, it is possible that VANGL2 additional hour followed by a second MESNA treatment. Here, influences the trafficking of additional membrane proteins in western blot of biotin-labeled proteins gives a quantitative polarized cells. For example, studies from fly and fish suggest measurement of the remaining non-recycled pool of MMP14. that PCP proteins regulate endocytosis of cadherins to modulate Our data show that a similar ratio of intracellular MMP14 was cell adhesion (Classen et al., 2005; Ulrich et al., 2005). recycled to the cell surface in both control and VANGL2- knockdown cells (Fig. 1B,C). To confirm these findings, we VANGL2 and MMP14 colocalization in HT-1080 cells performed an antibody-uptake assay to visualize MMP14- MMP14 expression in vitro reflects its regulation by endocytosis positive endocytic vesicles. Cells were labeled with antibody and recycling in that little protein is present on the plasma against MMP14 and either fixed or shifted to 37˚C to permit membrane with the majority localized to intracellular vesicles endocytosis. VANGL2-knockdown cells were confirmed to have (Steffen et al., 2008). Recruitment of fly Van Gogh to the plasma a decreased ability to internalize cell-surface MMP14 (Fig. 1D). membrane is thought to be a crucial event for PCP signaling in To determine whether loss of VANGL2 globally disrupts epithelial cells (Strutt and Strutt, 2009). However, recent in vitro endocytosis, we performed endocytosis assays using and in vivo studies have identified VANGL2 expression within fluorescently labeled transferrin (to ascertain clathrin-mediated vesicular compartments (Merte et al., 2010; Cha et al., 2011). endocytosis) or EEA1 antibody (to detect early endosomes). In VANGL2 and MMP14 colocalization in HT-1080 cells was neither case did we observe a difference between control and detected at the plasma membrane (Fig. 2A and supplementary VANGL2-knockdown cells (Fig. 1E and supplementary material material Fig. S3A). VANGL2 was also expressed in vesicle VANGL2 regulates MMP14 trafficking 2143

Fig. 2. VANGL2 and MMP14 colocalization in HT-1080 cells. (A) GFP–VANGL2 and MMP14 expression at membrane protrusions (arrows). (B–D) GFP–VANGL2 and MMP14 expression shown in relation to different vesicular markers (EEA1, RAB11 and VAMP1). Green arrows, VANGL2 single-positive vesicles; yellow arrows, VANGL2 and MMP14 double-positive vesicles; white arrows, triple-positive vesicles. Scale bars: 20 mm in top panels; lower panels are magnified views of the boxed regions. (E) Quantification of vesicles double positive for VANGL2 and MMP14 or vesicular markers. (F) Quantification of triple positive vesicles. For each graph, the regions used for vesicle counting are indicated by boxes in B–D and in supplementary material Fig. S3B (for LAMP1). Error bars show s.e.m. Journal of Cell Science

populations that labeled positive for MMP14, EEA1, RAB11 adhesion was shown to influence cell polarity and migration (recycling endosomes), LAMP1 (late endosomes) and VAMP1 (Takino et al., 2007), suggesting that regulation of cell-surface (post-Golgi vesicles) (Fig. 2B–F and supplementary material Fig. MMP14 activity is coordinated with integrin–ECM adhesion and S3A,B). Here, co-labeling of VANGL2 with MMP14 and EEA1 signaling. This notion is supported by data showing that or RAB11 represented the largest fraction of triple positive increased focal adhesion kinase (FAK) activity, including Y397 vesicles compared with LAMP1 or VAMP1. Neither VANGL2 autophosphorylation suppresses MMP14 endocytosis through nor MMP14 expression overlapped with the cis-Golgi marker effects on endophilin A2 phosphorylation and inhibition of GM130 (supplementary material Fig. S3C). Taken together, our dynamin (Wu et al., 2005). Notably, injection of dominant- data demonstrate that the cellular distribution of VANGL2 is negative dynamin into frog embryos was shown to inhibit dynamic and HT-1080 cells do not constitutively express this convergent extension gastrulation cell movements (Jarrett protein at the plasma membrane, but rather recruit VANGL2 (and et al., 2002). Based on these data, we examined the effects of MMP14) to specific membrane domains. These results are VANGL2 knockdown on focal adhesion formation and FAK consistent with the notion that VANGL2 could influence MMP14 phosphorylation. Using paxillin–GFP to label focal adhesions endocytosis by regulating protein interactions and/or actin (Turner, 1998) in HT-1080 cells, we found that VANGL2- cytoskeletal dynamics at either the plasma membrane or within knockdown cells had fewer focal adhesions than controls, endosomal vesicles. and the remaining adhesions were largely restricted to the cell periphery (Fig. 3A). Phosphorylation of FAK was analyzed by VANGL2 regulates MMP14 endocytosis downstream of immunolabeling polarized cells in a wounded monolayer and by focal adhesion kinase western blot. Both assays demonstrated increases in Y397- Although regulation of MMP14 exocytosis facilitates targeting of phosphorylated FAK in VANGL2-knockdown cells (Fig. 3B), activity to sites of focal degradation (Bravo-Cordero suggesting that VANGL2 function antagonizes FAK activation. et al., 2007; Steffen et al., 2008), the concept of endocytic In addition to affecting MMP14 endocytosis, phosphorylation of removal of plasma membrane MMP14 as a means of restricting FAK at Y397 induces disassembly of focal adhesions (Hamadi proteolytic degradation is less well characterized. In vitro, et al., 2005); thus our data also suggest that VANGL2 regulates MMP14-mediated cleavage of ECM at sites of cell–matrix focal adhesion turnover. To test whether the ability of VANGL2 2144 Journal of Cell Science 125 (9) Journal of Cell Science Fig. 3. VANGL2 regulates focal adhesion formation and FAK phosphorylation. (A) VANGL2-knockdwon paxillin–GFP cells have a reduced number of focal adhesions (arrows). Graph shows number of focal adhesions per cell with s.d. and medians (black boxes). For each siRNA, focal adhesions were counted (three independent experiments, n530 total cells analyzed, total number of focal adhesions counted, NT51921, VANGL251304) and the Student’s t-test was used to determine significance (control NT versus VANGL2, P50.0000178). (B) Loss of VANGL2 function increases the amount of Y397-phosphorylated FAK. Immunolabeling of Y397-phosphorylated FAK in NT and VANGL2-knockdown cells 4 hours after wounding (images taken at identical settings). Arrows indicate leading edge of the migrating cells. Western blot shows Y397-phosphorylated FAK and total FAK expression in NT and VANGL2-knockdown cells. Two samples from four independent experiments are shown. (C) Y397-phosphorylated FAK expression in HT-1080 cells treated with DMSO or FAK inhibitor (20– 55 mm) for 24 hours. (D) Cell-surface biotinylated MMP14 in control NT and VANGL2 (V2) siRNA transfected cells treated with DMSO or FAK inhibitor (35 mm). (E) Endocytosis assay performed on siRNA transfected cells treated with FAK inhibitor. Note reduced cell-surface MMP14 on VANGL2-knockdown cells at 4˚C (compare with 4˚C DMSO-treated cells) and normal endocytosis at 37˚C (compare with Fig. 1D).

to control MMP14 endocytosis is directly mediated by FAK Matrix metalloproteinases are downstream effectors of activation, we used an inhibitor of FAK Y397 phosphorylation Vangl2 function (Fig. 3C) in conjunction with siRNA knockdown. Our data show Our data suggest that matrix metalloproteinases act downstream of that treatment of VANGL2-knockdown HT-1080 cells with FAK VANGL2 in vitro. To corroborate these findings in zebrafish, we inhibitor rescues both the VANGL2-dependent increase in cell- tested whether trilobitem209/vangl2 mutant embryos (Jessen et al., surface MMP14 levels and defective MMP14 endocytosis 2002) exhibit increased metalloproteinase activity and have altered (Fig. 3D,E). FAK is a principle kinase implicated in integrin ECM organization. First, proteins were collected from wild-type signalling, and phosphorylation of Y397 is a key determinant of and trilobitem209/vangl2 homozygous mutants at various focal adhesion stability and turnover (Hamadi et al., 2005). Based embryonic stages and incubated with fluorogenic protease on our data, we propose a model whereby VANGL2 regulates substrates. Quantification of fluorescence showed a striking MMP14 endocytosis at sites of cell–matrix adhesion downstream increase in activity in trilobitem209/vangl2 mutants that of effects on FAK activation. However, we do not rule out was reduced to near background levels by treatment with the additional points of intersection between VANGL2 and MMP14, metalloproteinase inhibitor GM6001 (Fig. 4A). Interestingly, such as in EEA1-positive early endosomes. embryonic protease activity increases in wild-type embryos VANGL2 regulates MMP14 trafficking 2145

between mid-gastrulation and the end of gastrulation (80% with cell adhesion, polarity and fibronectin fibrillogenesis epiboly-1 somite) correlating with the appearance of ECM (Dzamba et al., 2009), it is possible that disrupted polarity in (Latimer and Jessen, 2010) and the onset of the trilobitem209/ trilobitem209/vangl2 mutants also contributes to loss of ECM. vangl2 mutant phenotype (Sepich et al., 2000). To determine However, despite having a severe cell polarity defect (Topczewski whether increased protease activity in trilobitem209/vangl2 mutant et al., 2001), fibronectin is not reduced in knypekm119/glypican4 embryos is mediated by Mmp14, we performed a gelatin protease mutants (N.A.M., R.E.Q. and J.R.J., unpublished data). Although assay using embryos injected with mmp14 morpholino. Notably, we do not exclude the contributions of cell polarity and adhesion to knockdown of Mmp14 in trilobitem209/vangl2 homozygous ECM assembly, our data suggest a significant role for matrix mutants reduced metalloproteinase activity to near wild-type metalloproteinases in regulating ECM degradation and remodeling levels (Fig. 4A), indicating that Mmp14 is aberrantly active when during zebrafish gastrulation. Vangl2 function is reduced. To test the effect of Vangl2 loss The above data suggest that loss of Vangl2 function in of function on ECM assembly, we performed fibronectin zebrafish trilobite mutants disrupts Mmp14 trafficking, resulting immunolabeling on whole-mount and sectioned embryos. in increased cell-surface expression and degradation of ECM. Consistent with their increased metalloproteinase activity, To confirm this hypothesis, we tested whether the Vangl2 loss- trilobitem209/vangl2 mutants had decreased levels of fibronectin of-function convergence and extension phenotype is suppressed (Fig. 4B). Fibronectin assembly into a fibrillar matrix is a cell- by inhibition of Mmp14 activity. Injection of a vangl2 antisense mediated process requiring engagement of integrin receptors and morpholino produced a moderate or severe convergence generation of cell tension (Schwarzbauer and Sechler, 1999). and extension phenotype in ,95% of wild-type embryos Because inhibition of PCP signaling in frog embryos interferes (Fig. 4C,D). By contrast, co-injection of mmp14 and vangl2 Journal of Cell Science

Fig. 4. Zebrafish trilobitem209/vangl2 mutant embryos have increased matrix metalloproteinase activity and decreased ECM. (A) Graph of zebrafish protease assay data. Protein samples were incubated with DQgelatin or DQcollagen IV substrates in assay buffer for 2 hours at 30˚C before quantifying fluorescence using a microplate reader. Injection of mmp14 morpholino reduced protease activity in trilobitem209/vangl2 mutants to wild-type levels, whereas treatment with GM6001 reduced protease levels to near background. Data shown are the average of three independent experiments performed in triplicate with s.d. Background fluorescence from negative controls was subtracted. hpf, hours post fertilization. (B) Representative confocal images of fibronectin immunolabeling shown in whole-mount and cross-section (wild type, n514; trilobitem209/vangl2 mutants, n510). For whole-mount images, z-series images were flattened into a single montage to show all fibronectin. E, ectoderm; PSM, pre-somitic mesoderm; Nc, notochord. (C,D) Knockdown of Mmp14 partially rescues the vangl2 loss- of-function convergence and extension defect. Wild-type embryos were injected with vangl2 and/or mmp14 morpholinos and fixed for whole-mount in situ hybridization at ,12-somite stage. Embryos were sorted into three phenotypic classes and data from three independent experiments pooled (NIC, n5134; mmp14, n543; vangl2, n583; vangl2+mmp14, n5103). Images depict representative embryos with each phenotype (lateral and dorsal views). Graph shows percentage of embryos with each phenotype. NIC, non-injected control. 2146 Journal of Cell Science 125 (9)

morpholinos consistently rescued this defect in a small at a concentration of 35 mm (chosen because it provided inhibition of FAK Y397 percentage of embryos, as indicated by normal anterior– phosphorylation without significant loss of cell viability). posterior body length and mediolateral width (Fig. 4C,D). In Quantification of VANGL2 expression, focal adhesions, and statistics these experiments, high doses (10 ng/embryo) of mmp14 To quantify colocalization of VANGL2 in vesicles, we generated an HT-1080 morpholino were required to achieve consistent rescue, cell line stably transfected with a GFP–VANGL2 fusion protein. This construct whereas lower doses (5 ng/embryo) had little effect. The exhibits an identical (but brighter) expression pattern to that detected by the VANGL2 antibody. Cells were fixed in 4% paraformaldehyde (PFA) and Vangl2-knockdown phenotype was exacerbated in some subjected to immunolabeling for MMP14, EEA1, RAB11, LAMP1 or VAMP1, as mmp14 morpholino-injected embryos, indicating a requirement indicated. The z-stack from each field containing the highest number of GFP- for tight regulation of Mmp14 activity during embryogenesis. expressing vesicles was digitally analyzed using ImageJ software to determine Indeed, we previously showed that high doses of mmp14 the total number of VANGL2, MMP14 or additional marker-positive vesicles in three independent experiments. Fields of ,400 mm2 (indicated by boxes in morpholinos disrupt cell polarity and induce a convergence and Fig. 2B–D and supplementary material Fig. S3B) were quantified in .20 cells extension defect (Coyle et al., 2008). Taken together, our data labeled with VANGL2 and MMP14 and 6–10 cells labeled with all other markers. show that Mmp14 acts downstream of Vangl2 to mediate the The multi-point tool in ImageJ software was used to count focal adhesions in establishment or maintenance of PCP. images of non-targeting control (NT) and VANGL2-knockdown cells. For comparison of the differences between means, we used a two-tailed Student’s t- The ability of Vangl2 to regulate the proteolytic activity of test. Mmp14 in migrating gastrula cells can partly explain how cell polarization is propagated across a field of cells. We propose that Protease assay a proper balance of matrix metalloproteinase activity is required Zebrafish were maintained under standard conditions and embryos staged during gastrulation to regulate integrin–ECM interactions and the according to morphology. Embryos were dechorionated with , deyolked and placed in ice-cold lysis buffer without protease inhibitors (15 mM NaCl, alignment or stabilization of membrane protrusions along 10 mM HEPES, 0.1% Triton X-100, pH 8). Proteins were quantified and assays fibronectin. Asymmetrical Vangl2 expression in polarized cells performed using either DQgelatin or DQcollagen IV as directed (Invitrogen). In would promote or exclude Mmp14 and proteolytic activity from some experiments, 100 mM GM6001 metalloproteinase inhibitor (Calbiochem) or DMSO were added. Alternatively, embryos were injected at the one cell-stage with certain cell surfaces and thereby influence cell polarity mmp14 morpholino, as described below. autonomously and non-autonomously. One prediction from this model is that excess metalloproteinase activity and loss of ECM Rescue experiments in trilobite/vangl2 mutant embryos would disrupt polarized The vangl2 morpholino (Gene Tools, LLC; 59-AGTTCCACCTTACTCCTGA- membrane protrusive activity. In agreement, migrating cells in GAGAAT-39) was injected into single-cell embryos at 3 ng/embryo. Morpholinos to both zebrafish Mmp14 isoforms (Coyle et al., 2008) were injected at doses of trilobite/vangl2 mutants are able to generate protrusions, but they 5 ng of each isoform per embryo (10 ng total). We compared non-injected controls cannot be polarized and/or stabilized (Jessen et al., 2002). Future with embryos injected with vangl2 morpholino or a combination of vangl2 and studies will be needed to establish how Vangl2 regulates Mmp14 mmp14a/mmp14b morpholinos. Embryos were fixed at the ,12-somite stage and trafficking in vivo and to identify the specific roles of Mmp14 analyzed by in situ hybridization. Probes used were zic1 (forebrain), egr2 (hindbrain rhombomeres) and myod1 (somites). Embryos were placed in different (and Mmp2) for cell polarity and directed migration underlying phenotypic classes: (1) mild or partially rescued, less severe than the moderate convergence and extension. vangl2 phenotype; (2) moderate, similar to the moderate vangl2 phenotype; (3) severe, worse than the moderate vangl2 phenotype.

Journal of Cell Science Materials and Methods Endocytosis and recycling assays Acknowledgements HT-1080 cell culture, siRNA-mediated knockdown of VANGL2, and We thank Alissa Weaver, Department of Cancer Biology, Vanderbilt internalization and recycling assays were performed as described (Remacle et al., University Medical Center for paxillin–GFP HT-1080 cells and J. 2003; Cantrell and Jessen, 2010). Biotinylated proteins were isolated using streptavidin Dynabeads (Invitrogen). After magnetic separation, samples were Dunlap for technical assistance. boiled in Laemmli buffer and analyzed by western blot. The blots were stripped and re-probed for GAPDH. Densitometry was performed using ImageJ Funding (rsb.info.nih.gov/ij) to quantify MMP14 and GAPDH protein bands, and data This work was supported by grants to J.R.J. from the American were normalized to GAPDH. MMP14 and transferrin internalization were visualized fluorescently as described (Remacle et al., 2003). Cancer Society [grant number RSG 0928101]; the National Science Foundation [grant number IOS 0950849]; and in part by a National Microscopy Cancer Institute Cancer Center Support Grant [grant number P30 Imaging was performed using a Leica DMI6000B epifluorescence microscope CA068485] (for use of the confocal microscope provided by the Cell equipped with an ORCA monochromatic camera (Hamamatsu), a 406 oil Imaging Shared Resource). objective, and Simple PCI software (Hamamatsu). Certain images (control and experimental) were manipulated using the brightness and contrast functions of the above software. Images of zebrafish embryos were collected using a Leica MZ16F Supplementary material available online at stereomicroscope and DFC280 camera. Confocal images were captured using an http://jcs.biologists.org/lookup/suppl/doi:10.1242/jcs.097964/-/DC1 Olympus FV1000 laser-scanning microscope using a 1006 oil objective (26 optical zoom) and an optical thickness (z) of 0.67 mm (provided by the VUMC References Cell Imaging Shared Resource). Bravo-Cordero, J. J., Marrero-Diaz, R., Megı´as, D., Genı´s, L., Garcı´a-Grande, A., Garcı´a, M. A., Arroyo, A. G. and Montoya, M. C. (2007). MT1-MMP proinvasive Immunolabeling, antibodies and FAK inhibition activity is regulated by a novel Rab8-dependent exocytic pathway. EMBO J. 26, Immunolabeling and cell monolayer wounding were performed as described 1499-1510. (Cantrell and Jessen, 2010). Primary antibodies were used against the following Cantrell, V. A. and Jessen, J. R. (2010). The planar cell polarity protein Van Gogh- proteins: MMP14 for localization studies and western blot (MAB3328, Millipore), Like 2 regulates tumor cell migration and matrix metalloproteinase-dependent MMP14 for endocytosis assay (Z-812, Santa Cruz Biotechnology), VANGL2 invasion. Cancer Lett. 287, 54-61. (AF4815, R&D Systems), LAMP1 (Developmental Studies Hybridoma Bank), Cha, S. W., Tadjuidje, E., Wylie, C. and Heasman, J. (2011). The roles of maternal EEA1 and RAB11 (C45B10 and D4F5, Cell Signaling Technologies), VAMP1 Vangl2 and aPKC in Xenopus oocyte and embryo patterning. Development 138, (ab3346, Abcam), GM130 (P-20, Santa Cruz Biotechnology), FAK and fibronectin 3989-4000. (RB-477 and AB-10, Lab Vision Corporation), Y397-P FAK (KAP-TK131, Enzo Classen, A. K., Anderson, K. I., Marois, E. and Eaton, S. (2005). Hexagonal packing Life Sciences) and GAPDH (clone 6C5, Ambion). The FAK inhibitor PF-573,228 of Drosophila wing epithelial cells by the planar cell polarity pathway. Dev. Cell 9, was dissolved in DMSO at 10 mM and used as described (Slack-Davis et al., 2007) 805-817. VANGL2 regulates MMP14 trafficking 2147

Coyle, R. C., Latimer, A. and Jessen, J. R. (2008). Membrane-type 1 matrix Sato, H., Takino, T., Okada, Y., Cao, J., Shinagawa, A., Yamamoto, E. and Seiki, M. metalloproteinase regulates cell migration during zebrafish gastrulation: evidence for (1994). A matrix metalloproteinase expressed on the surface of invasive tumour cells. an interaction with non-canonical Wnt signaling. Exp. Cell Res. 314, 2150-2162. Nature 370, 61-65. Djiane, A., Riou, J., Umbhauer, M., Boucaut, J. and Shi, D. (2000). Role of frizzled 7 Schwarzbauer, J. E. and Sechler, J. L. (1999). Fibronectin fibrillogenesis: a paradigm in the regulation of convergent extension movements during gastrulation in Xenopus for extracellular matrix assembly. Curr. Opin. Cell Biol. 11, 622-627. laevis. Development 127, 3091-3100. Sepich, D. S., Myers, D. C., Short, R., Topczewski, J., Marlow, F. and Solnica- Dzamba, B. J., Jakab, K. R., Marsden, M., Schwartz, M. A. and DeSimone, D. W. Krezel, L. (2000). Role of the zebrafish trilobite in gastrulation movements of (2009). Cadherin adhesion, tissue tension, and noncanonical Wnt signaling regulate convergence and extension. Genesis 27, 159-173. fibronectin matrix organization. Dev. Cell 16, 421-432. Slack-Davis, J. K., Martin, K. H., Tilghman, R. W., Iwanicki, M., Ung, E. J., Autry, Goto, T., Davidson, L., Asashima, M. and Keller, R. (2005). Planar cell polarity genes C., Luzzio, M. J., Cooper, B., Kath, J. C., Roberts, W. G. et al. (2007). Cellular regulate polarized extracellular matrix deposition during frog gastrulation. Curr. Biol. characterization of a novel focal adhesion kinase inhibitor. J. Biol. Chem. 282, 14845- 15, 787-793. 14852. Hamadi, A., Bouali, M., Dontenwill, M., Stoeckel, H., Takeda, K. and Ronde´, P. Steffen, A., Le Dez, G., Poincloux, R., Recchi, C., Nassoy, P., Rottner, K., Galli, T. (2005). Regulation of focal adhesion dynamics and disassembly by phosphorylation and Chavrier, P. (2008). MT1-MMP-dependent invasion is regulated by TI-VAMP/ of FAK at tyrosine 397. J. Cell Sci. 118, 4415-4425. VAMP7. Curr. Biol. 18, 926-931. Strutt, H. and Strutt, D. (2009). Asymmetric localisation of planar polarity proteins: Jarrett, O., Stow, J. L., Yap, A. S. and Key, B. (2002). Dynamin-dependent Mechanisms and consequences. Semin. Cell Dev. Biol. 20, 957-963. endocytosis is necessary for convergent-extension movements in Xenopus animal cap Takino, T., Saeki, H., Miyamori, H., Kudo, T. and Sato, H. (2007). Inhibition of explants. Int. J. Dev. Biol. 46, 467-473. membrane-type 1 matrix metalloproteinase at cell-matrix adhesions. Cancer Res. 67, Jessen, J. R., Topczewski, J., Bingham, S., Sepich, D. S., Marlow, F., 11621-11629. Chandrasekhar, A. and Solnica-Krezel, L. (2002). Zebrafish trilobite identifies Topczewski, J., Sepich, D. S., Myers, D. C., Walker, C., Amores, A., Lele, Z., new roles for Strabismus in gastrulation and neuronal movements. Nat. Cell Biol. 4, Hammerschmidt, M., Postlethwait, J. and Solnica-Krezel, L. (2001). The 610-615. zebrafish glypican knypek controls cell polarity during gastrulation movements of Jiang, A., Lehti, K., Wang, X., Weiss, S. J., Keski-Oja, J. and Pei, D. (2001). convergent extension. Dev. Cell 1, 251-264. Regulation of membrane-type matrix metalloproteinase 1 activity by dynamin- Turner, C. E. (1998). Molecules in focus Paxillin. Int. J. Biochem. Cell Biol. 30, 955- mediated endocytosis. Proc. Natl. Acad. Sci. USA 98, 13693-13698. 959. Latimer, A. and Jessen, J. R. (2010). Extracellular matrix assembly and organization Ulrich, F., Krieg, M., Scho¨tz, E. M., Link, V., Castanon, I., Schnabel, V., during zebrafish gastrulation. Matrix Biol. 29, 89-96. Taubenberger, A., Mueller, D., Puech, P. H. and Heisenberg, C. P. (2005). Wnt11 Merte, J., Jensen, D., Wright, K., Sarsfield, S., Wang, Y., Schekman, R. and Ginty, functions in gastrulation by controlling cell cohesion through Rab5c and E-cadherin. D. D. (2010). Sec24b selectively sorts Vangl2 to regulate planar cell polarity during Dev. Cell 9, 555-564. neural tube closure. Nat. Cell Biol. 12, 41-46. Wallingford, J. B., Rowning, B. A., Vogeli, K. M., Rothba¨cher, U., Fraser, S. E. and Poincloux, R., Liza´rraga, F. and Chavrier, P. (2009). Matrix invasion by tumour cells: Harland, R. M. (2000). Dishevelled controls cell polarity during Xenopus a focus on MT1-MMP trafficking to invadopodia. J. Cell Sci. 122, 3015-3024. gastrulation. Nature 405, 81-85. Remacle, A., Murphy, G. and Roghi, C. (2003). Membrane type I-matrix Wu, X., Gan, B., Yoo, Y. and Guan, J. L. (2005). FAK-mediated src phosphorylation metalloproteinase (MT1-MMP) is internalised by two different pathways and is of endophilin A2 inhibits endocytosis of MT1-MMP and promotes ECM degradation. recycled to the cell surface. J. Cell Sci. 116, 3905-3916. Dev. Cell 9, 185-196. Journal of Cell Science