Oncogene (2014) 33, 4203–4212 & 2014 Macmillan Publishers Limited All rights reserved 0950-9232/14 www.nature.com/onc

ORIGINAL ARTICLE Macrophage contact induces RhoA GTPase signaling to trigger tumor cell intravasation

M Roh-Johnson1,2, JJ Bravo-Cordero1,2, A Patsialou1,2, VP Sharma1,2, P Guo1,2, H Liu3, L Hodgson1,2 and J Condeelis1,2

Most cancer patients die as a result of metastasis, thus it is important to understand the molecular mechanisms of dissemination, including intra- and extravasation. Although the mechanisms of extravasation have been vastly studied in vitro and in vivo, the process of intravasation is still unclear. Furthermore, how cells in the tumor microenvironment facilitate tumor cell intravasation is still unknown. Using high-resolution imaging, we found that macrophages enhance tumor cell intravasation upon physical contact. Macrophage and tumor cell contact induce RhoA activity in tumor cells, triggering the formation of actin-rich degradative protrusions called invadopodia, enabling tumor cells to degrade and break through matrix barriers during tumor cell transendothelial migration. Interestingly, we show that macrophage-induced invadopodium formation and tumor cell intravasation also occur in patient-derived tumor cells and in vivo models, revealing a conserved mechanism of tumor cell intravasation. Our results illustrate a novel heterotypic cell contact-mediated signaling role for RhoA, as well as yield mechanistic insight into the ability of cells within the tumor microenvironment to facilitate steps of the metastatic cascade.

Oncogene (2014) 33, 4203–4212; doi:10.1038/onc.2013.377; published online 23 September 2013 Keywords: cancer; tumor cell intravasation; macrophages; RhoA biosensor; invadopodia; heterotypic cell contact

INTRODUCTION landmark composed of a perivascular macrophage in contact More than 1 in 3 people will develop cancer in their lifetime, and with a tumor cell at blood vessels had been identified and termed B1500 people die from cancer each day in the United States. as TMEM (tumor microenvironment of metastasis). In a case- Understanding the mechanistic basis of specific steps of metastasis controlled study of metastastic and nonmetastatic breast cancers, is critical for the identification of robust early prognostic markers. TMEM density in breast tumors at initial resection was associated 6 Understanding how tumor cells escape the primary tumor and with risk of metastasis, suggesting that macrophages, tumor cells enter the vasculature (a process termed as intravasation) is a key and endothelial cells cooperate for tumor cell entry into the step in designing treatment strategies of the disease. Multiphoton- vasculature. However, little is known about the cell biological based intravital imaging of rodent mammary adenocarcinoma1 mechanisms that exist between these three cell types during and human tumors2 has revealed that tumor cells and intravasation. macrophages cooperate during several key steps of the To gain a closer look at the molecular mechanisms of tumor cell metastatic cascade. The polarization and subsequent motility of transendothelial migration, in vitro experiments have focused invasive tumor cells toward the blood vessels are dependent on a extensively on tumor cell and endothelial cell behavior. Often, paracrine loop of signaling with macrophages.3,4 Tumor cells models of tumor cell extravasation or exit of tumor cells from the respond to macrophage-secreted epidermal growth factor (EGF), vasculature are used, as the apical surface of the endothelium can and, in turn, macrophages respond to tumor cell-secreted colony- be easily accessed by plating endothelial monolayers and seeding 7–11 stimulating factor-1 (CSF-1). Macrophages have also been shown to tumor cells on these monolayers. From these studies, an have a role in tumor cell entry into the vasculature in vitro and extensive view of how tumor cells affect the endothelial cell in vivo, as treatment with drugs that affect macrophage function architecture has been elucidated, and the specific steps of tumor results in a decreased number of circulating tumor cells.1 However, cell adhesion and intercalation during transmigration have been as both the tumor cells’ ability to migrate toward the blood vessel described.7,8,12,13 A common downstream mechanism of tumor and the actual intravasation step are necessary for the entry of cell transendothelial migration is the opening of the endothelial tumor cells into the circulation, it is still unclear how macrophages monolayer. Endothelial cells lose their adherens junctions and affect tumor cell intravasation specifically. Furthermore, recent tight junctions as gaps open in the monolayer, allowing for in vitro studies examining the mechanism of tumor cell tumor cell passage. Molecular pathways leading to adhesion intravasation have shown that the presence of macrophages dissolution have also been identified.12,14,15 What is less clear, increases tumor cell intravasation, but the mechanisms of this however, is how tumor cells undergo intravasation, and whether enhancement are still unclear.1,5 common mechanisms exist between intravasation and extra- Intravital imaging of the tumor microenvironment has shown vasation. Given that tumor cells withstand shear flow stresses in that tumor cells intravasate into the vasculature at sites near the blood vessels and that different cells are present at the sites of macrophages.3 Because of these visualizations, a microanatomic intravasation and extravasation, we predict that different modes

1Department of Anatomy and Structural Biology, Albert Einstein College of Medicine, Bronx, NY, USA; 2Gruss-Lipper Biophotonics Center, Bronx, NY, USA and 3The Ben May Department for Cancer Research, University of Chicago, Chicago, IL, USA. Correspondence: Dr M Roh-Johnson, Anatomy and Structural Biology, Albert Einstein College of Medicine, 1301 Morris Park Avenue, Price Center 208, Bronx, NY 10461, USA. E-mail: [email protected] Received 2 May 2013; revised 1 July 2013; accepted 16 July 2013; published online 23 September 2013 Macrophages induce RhoA in tumor cell intravasation M Roh-Johnson et al 4204 of cell–cell interactions occur when tumor cells intravasate of the endothelium facing the bottom chamber (Supplementary compared with that when they extravasate. As such, given that Figures S1C–F). macrophages are visualized at the sites of tumor cell intravasation in vivo, we sought to explore the link between the close association of macrophages and tumor cells and the subsequent Macrophages enhance tumor-cell intravasation in vitro tumor cell intravasation. To examine the interactions between tumor cells, macrophages Using dissolution of endothelial cell–cell adhesion as a readout and the endothelium during intravasation, we added a highly of tumor cell transendothelial migration, we dissected the metastatic triple-negative human breast cancer cell line, MDA-MB- relationship between macrophages and tumor cells with an 231, and the BAC1.2F5 murine macrophage cell line to the in vitro model of intravasation. In particular, we investigated transwells and analyzed the interactions with fixed and live whether the association of macrophages and tumor cells imaging (Figure 1). We have previously shown that the inter- reflected a cell biological mechanism that supports intra- actions between these cells are not hindered because of species vasation. Using high-resolution imaging, combined with in vitro difference.16,17 With fixed imaging of en face views of the and in vivo models, we tested whether macrophages enhance transwells, we found that tumor cells and macrophages made tumor cell intravasation and the cell biological and signaling close interactions with each other at the level of the endothelium mechanisms that regulate this process. Our results reveal that (Figure 1b). Live imaging of tumor cell transmigration in the macrophages facilitate tumor-cell intravasation by inducing transwells revealed that tumor cells preferentially underwent the activation of the RhoA signaling pathway, resulting in the transendothelial migration at sites where macrophages were formation of invadopodia and enabling the tumor cells to initiate present (Figures 1c and d; Supplementary Movie 1). Tumor cells intravasation by penetrating through the basement membrane of took B2–3 h to undergo transendothelial migration (Figure 1c), the vasculature. consistent with published in vivo and in vitro studies indi- cating that tumor transmigration occurs on the order of several hours5,18–21 in contrast to neutrophil migration, which occurs on a RESULTS faster time scale.22 Validation of in vitro intravasation assay To determine whether macrophages facilitated tumor cell To specifically examine the interactions between tumor cells, intravasation in our assay, we compared tumor cell transmigration macrophages and endothelial cells during tumor-cell intravasation, in transwells in which macrophages were either present or we developed an in vitro model to mimic the appropriate endo- absent. Tumor cells exhibited a basal level of intravasation thelial monolayer polarity. Using primary human microvascular independently of macrophages; however, tumor cells exhibited endothelial cells (HUVECs), we formed ‘upside-down’ endothelial a threefold increase in intravasation in the presence of macro- monolayers using transwells for support (Figure 1a). The endo- phages (Figures 2a–c). We tested several macrophage-to-tumor thelium was intact, impermeable to large molecules and dis- cell ratios in our assay and determined that a fourfold increase played high transendothelial electrical resistance (Supplementary in macrophage-to-tumor cell ratio elicited a robust tumor cell Figures S1A and B). To image the transwells, transwells were intravasation response (Supplementary Figure S2A). As the placed in glass-bottom dishes and imaged en face at the level BAC1.2F5 macrophage cell line was used for these experiments, of the endothelium (Figure 1a and Supplementary Figure S1C). we also tested for the ability of other macrophage cell lines to We assessed monolayer polarity by immunostaining the endo- enhance tumor cell intravasation. Using an immortalized bone thelium with polarity markers, apical ZO-1 and basal collagen IV marrow-derived macrophage cell line (iBMM)23 as well as the and found that proper polarity was established, with the basal RAW264.7 macrophage cell line, we found that tumor cell trans- side of the endothelium facing the transwell and the apical side endothelial migration was enhanced (Supplementary Figure S2B),

Figure 1. Tumor cells physically interact with macrophages during transendothelial migration. (a) Diagram of transwell experiment. (b) Fixed image of apical section of transwell with tumor cell (MDA-MB-231-GFP, green) and macrophage (CMPTX-labeled, red) physically interacting in the endothelium (blue, DAPI). Asterisk marks tumor cells in contact with macrophages at endothelium. (c) Still images from en face view of movie of tumor cell (red) transmigrating through endothelium (green) next to a macrophage (purple). Images are taken every 40 min. The dark region of tumor cell is nucleus as dtomato space fill was used. White dashed lines mark edges of endothelial cells (bottom panel). Trans- endothelial migration of tumor cell is depicted by the broadening of outlined area within the endothelium. (d) IMARIS 3D reconstruction of cells shown in (c) revealing tumor cell (red) penetrating through endothelium (green) in the z axis next to a macrophage (blue).

Oncogene (2014) 4203 – 4212 & 2014 Macmillan Publishers Limited Macrophages induce RhoA in tumor cell intravasation M Roh-Johnson et al 4205

Figure 2. Macrophages facilitate 231 transendothelial migration. (a) Merge of apical 12 microns of HUVEC transwells. MDA-MB-231 cells alone (green) and endothelium marked by phalloidin–rhodamine (red). (b) Merge of apical 12 microns of HUVEC transwells with MDA-MB-231 cells (green) and macrophages (unlabeled) added. Endothelium marked by phalloidin–rhodamine (red). (c) Quantification of number of MDA-MB- 231 cells exposed to the apical surface of HUVEC transwells per field of view (512 Â 512 mm), as a relative number to MDA-MB-231 GFP control. Mean±s.e.m, t test Po0.05. n ¼ 25, six independent experiments. (d) Tight junctions (ZO-1, white) immunostaining of transwells when the tumor cell (MDA-MB-231 tomato, red) is undergoing transendothelial migration when in contact with a macrophage (green). White arrows mark absence of ZO-1 immunostaining. (e) Quantification of tumor cells undergoing transmigration on transwells when in contact or not in contact with macrophages. n ¼ 22, five independent experiments. suggesting that this phenomenon was due to a general must degrade the basal lamina of the endothelium during mechanism by macrophages and not specific to a particular transmigration. As breast cancer cell lines were known to form macrophage cell line. As BAC1.2F5 macrophages generated the invadopodia, actin-rich structures that degrade matrix, we tested most consistent and robust macrophage-induced response, this whether macrophages were capable of inducing the formation cell line was used for the remainder of the study. of invadopodia in breast cancer cells during transendothelial Although tumor cells exhibited a basal level of intravasation migration, as the formation of invadopodia during this process independently of macrophages (Figures 2a and c), we specifically has been speculated to occur in vivo and in vitro but never were interested in how macrophages affected tumor cell unequivocally shown.24–27 To address this question, we first transmigration as there was a strong correlation between TMEM assayed for the formation of macrophage-induced invadopodia in formation and distant metastases in patients.6 Therefore, we two-dimensional (2D) assays and then assayed for the formation focused our studies on macrophage-induced transendothelial of macrophage-induced invadopodia during in vitro intravasation. migration. To determine the effects of tumor cell transendothelial In our 2D assay, the formation of invadopodia was detected migration on endothelial cell–cell adhesion, we examined cell by subcellular punctate structures formed at the ventral surface adhesion in transwells during tumor cell crossing. Using multi- of cells that are positive for two invadopodium markers, Tks5 photon-based intravital imaging, tumor cell protrusions aligned and Cortactin.28,29 Tks5 is an adaptor protein necessary for along endothelial cell–cell contacts in vivo, suggesting that invadopodium formation30 and Cortactin is an actin-binding these protrusive structures were probing the endothelial cell protein shown to localize and function at invadopodia.28,29,31,32 adhesions (Supplementary Figure S3). Indeed tumor cells have MDA-MB-231 cell lines have been shown to form invadopodia been shown to insert into endothelial monolayers between readily when plated in full serum and/or upon EGF stimulation, endothelial cells in in vitro models.7 Consistent with these even in the absence of macrophages.28,29,33,34 To study the role of in vivo results, in our in vitro intravasation assay, endothelial macrophages in the formation and function of invadopodia in cells lost cell–cell adhesion in regions of tumor cell intravasation tumor cells, we first serum-starved the tumor cells and then (Figure 2d), suggesting that tumor cells were undergoing introduced macrophages. When MDA-MB-231 cells were in serum- paracellular migration (between endothelial cells) versus transcel- starvation conditions, they formed few invadopodia, with lular migration (through endothelial cells) (Figure 2d). This a majority of the cells forming none at all (control panels; paracellular migration was also reflected in live movies in which Figures 3a and b). However, when serum-starved MDA-MB-231 tumor cells were visualized to migrate between endothelial cells cells were cocultured in the presence of macrophages, we found (Figure 1c). To determine whether the observed paracellular that tumor cells in direct contact with macrophages formed an migration was induced by macrophages, we quantified the average of five functional invadopodia, which contained both number of paracellular migration events that occurred when Cortactin and Tks5 invadopodium markers and were capable tumor cells were in contact with macrophages compared with of degrading matrix (Figures 3a and b). These findings were in those that were independent of macrophages. Tumor cells were contrast to the average number of invadopodia formed by seen to be in contact with macrophages 86% of the time during tumor cells not in physical contact with macrophages (Figure 3a), paracellular transmigration (Figure 2e), suggesting that the which was similar to that of tumor cells that were plated alone intravasation is a macrophage-induced phenomenon. (Figure 3b). These experiments suggest that direct physical contact was necessary between the tumor cells and macrophages for macrophage-induced invadopodium formation. To confirm Macrophages induce invadopodium formation in tumor cells this hypothesis, we performed macrophage-conditioned media upon physical contact experiments to stimulate tumor cells. Macrophage-conditioned From the above results, we determined that the presence of media did not increase the number of invadopodia compared macrophages correlated with tumor cell intravasation; therefore, with control media (Figure 3c), further suggesting that physical we next sought to determine the cell biological mechanism by contact between macrophages and tumor cells was required for which this induction occurred. We hypothesized that tumor cells enhanced invadopodium formation.

& 2014 Macmillan Publishers Limited Oncogene (2014) 4203 – 4212 Macrophages induce RhoA in tumor cell intravasation M Roh-Johnson et al 4206

Figure 3. Macrophages induce invadopodium formation in MDA-MB-231 cells. (a,b) Immunostaining of MDA-MB-231 cells for Tks5 and Cortactin in the presence and absence of macrophages on 405-labeled gelatin matrix. Top panels shows MDA-MB-231 cell in contact with macrophage (red). Inset is zoomed region of white box in panels. Arrows mark invadopodia, defined as Cortactin- and Tks5-positive invadopodia that degrade matrix. Bottom panels show a MDA-MB-231 cell that is not in contact with a macrophage. (b) Quantification of the number of invadopodia per cell when MDA-MB-231 cells were plated with macrophages or when MDA-MB-231 cells were plated alone. Mean±s.e.m, ANOVA P ¼ 0.0003. n ¼ 60 (alone), n ¼ 18 (no contact), n ¼ 41 (contact) cells; six independent experiments. (c) Quantification of the number of invadopodia per cell when MDA-MB-231 cells were plated in BAC1.2F5-conditioned media. Mean±s.e.m as a ratio of control; n ¼ 34 (ctrl), 30 (conditioned) cells; two independent experiments.

Macrophage-induced invadopodia form and are necessary for derived from a patient with triple-negative breast cancer, TN1, tumor-cell intravasation which are capable of invasion and spontaneous metastasis.36 As the formation of invadopodia was induced by macrophages As these primary cells do not grow in culture as a cell line, on 2D matrices, we next tested whether invadopodia could be TN1 tumors were grown orthotopically in mice, and the tumor induced by macrophages during tumor cell intravasation. Tumor cells were dissociated and assayed for their ability to undergo cells were first transfected with Cortactin-tagRFP to assess the macrophage-dependent transendothelial migration and invado- formation of invadopodia. Regions of tumor-cell transendo- podium formation in vitro. thelial migration were detected by a loss of endothelial cell–cell We found that TN1 cells exhibited macrophage-induced invado- adhesion, as visualized by tight junction immunostaining. Tumor podium formation (Figures 5a–d; Supplementary Figure S5). We cells indeed accumulated the invadopodium marker, Cortactin, in further found that TN1 cells preferentially underwent intravasation areas of the tumor cell that penetrated through endothelial when in contact with macrophages, consistent with our findings barriers (Figure 4a, red arrow) forming an invasive front similar to with MDA-MB-231 cells (Figures 5e–i). When we quantified the that previously shown in three-dimensional collagen matrices,33,35 amount of transmigration that occurred, we found that 86% of suggesting that invadopodia formed during intravasation. We also TN1 cells physically contacted macrophages when undergoing found that an independent invadopodium marker, Tks5, is also paracellular migration, which was similar to our findings with enriched in areas of the tumor cell that penetrate the endothelium MDA-MB-231 cells (Figure 5i, compare to Figure 2e). Thus, our (Supplementary Figure S4A, green arrow). To determine whether experiments with patient-derived breast tumor cells recapitulate invadopodia are indeed necessary for efficient intravasation, our in vitro intravasation results with cell lines. Tks5 was knocked down in tumor cells, and were assessed for their ability to undergo intravasation. Consistent with published Physical contact between macrophages and tumor cells correlates results, cells with either transient or stable knockdown of Tks5 with invadopodium formation in vivo 30 did not form invadopodia on 2D substrates; Figures 4b–e; Our in vitro results with both cell lines and patient-derived cells Supplementary Figures S4B and C). Transient or stable Tks5 KD suggest that the contact between macrophages and tumor cells cells also did not undergo intravasation even in the presence of induces invadopodium formation in tumor cells to facilitate tumor macrophages, and their level of crossing was similar to back- cell intravasation. To validate these findings in vivo, we generated ground levels (Figure 4f; Supplementary Figure S4D). Further- orthotopic MDA-MB-231 tumors in immunocompromised mice. more, treating tumor cells with an MMP inhibitor, GM6001, Once tumors attained the size of 1–1.2 cm, the tumors were which has been shown to inhibit the ability of invadopodia to 32 removed, sectioned and immunostained for blood vessels, macro- degrade matrix, resulted in the inability of tumor cells to cross phages and invadopodia. Specifically, we tested whether tumor the endothelium (Figures 4f and g), further suggesting invado- cells upregulate invadopodium markers when in physical contact podia are required for intravasation. Given that Tks5 KD cells do 30 with macrophages near blood vessels in vivo. We found that this not exhibit motility defects, these experiments strongly suggest was indeed the case, and tumor cells that were in physical contact that the matrix-degrading ability of macrophage-induced invado- with macrophages formed cortactin-rich structures next to blood podium formation is necessary for tumor-cell intravasation. vessels, compared with cells at the blood vessel that were not in contact with macrophages (Figures 6a and b). These experiments Patient-derived breast tumor cells form invadopodia and suggest that physical contact between macrophages and tumor physically contact macrophages during tumor-cell intravasation cells induces invadopodium formation at blood vessels in vivo. As the MDA-MB-231 breast cancer cell line is triple-negative for the estrogen receptor, progesterone receptor and human Macrophages activate RhoA signaling in tumor cells to induce epidermal growth factor receptor 2, we sought to validate our invadopodium formation upon physical contact findings with another triple-negative receptor cell type. To add We have shown that macrophages induce tumor cell invadopo- clinical relevance to our findings, we used breast tumor cells dium formation upon physical contact for intravasation. We next

Oncogene (2014) 4203 – 4212 & 2014 Macmillan Publishers Limited Macrophages induce RhoA in tumor cell intravasation M Roh-Johnson et al 4207

Figure 4. Invadopodia are required for transendothelial migration. (a) Invadopodia form as MDA-MB-231 cells undergo transendothelial migration. Immunostaining for ZO-1 (white in left panel, purple in merge) of transwells with MDA-MB-231 cells transfected with Cortactin- tagRFP (red). White arrowheads mark disassembled tight junctions. Red arrow marks Cortactin enrichment in tumor cell as the cell penetrates through the endothelium. (b) Tks5 smartpool siRNA knockdown western blot with actin as loading control. Quantification of Tks5 fluorescence below. (c) qRT-PCR of control and Tks5 siRNA cells to show tks5 mRNA reduction in Tks5 siRNA cells. n ¼ 3 replicates. (d) Representative images of control and Tks5 siRNA MDA-MB-231 cells immunostained for Ttks5 (green) and Cortactin (purple) and matrix degradation (black clearings). Arrows mark invadopodia. (e) Quantification of invadopodium number per cell in control and Tks5 siRNA MDA- MB-231 cells plated in complete media. Mean±s.e.m, t-test Po0.05. n ¼ 15 (ctrl), n ¼ 27 (tks5) cells; three independent experiments. (f) Quantification of macrophage-induced in vitro intravasation when MDA-MB-231 cells were treated with control siRNA, Tks5 siRNA and GM6001 with and without macropahges. Mean±s.e.m. n ¼ 10 (ctrl), n ¼ 15 (ctrl þ MF), n ¼ 14 (Tks5), n ¼ 14 (Tks5 þ MF), n ¼ 11 (MMP inhib); two independent experiments for each. ANOVA P ¼ 1.4E-11. (g) Representative images of MDA-MB-231 treated with GM6001 for transendothelial cell crossing. x views of transwells showing tumor cells (green) and endothelium (red), with cartoon depiction to the right. Dark clearing is filter (arrow). sought to determine the molecular mechanism of this heterotypic macrophages. RhoA biosensor expressing cells that did not cell contact. RhoA, a member of the Rho family of GTPases, was contact macrophages maintained a constant level of RhoA activity shown to be involved in invadopodium formation,37,38 and RhoC (Figures 7a and c; Supplementary Figure S7A; Supplementary GTPase has been shown to affect invadopodium morphology and Movies 2 and 3). However, when macrophages contacted tumor function without affecting invadopodium number.38 As we had cells (Figure 7b, top DIC panels and inset), there was a global visualized changes in invadopodium number upon macrophage increase in RhoA activity that was sustained for 30 min (Figure 7c; contact, we hypothesized that RhoA activity may be induced Supplementary Figure S7B; Supplementary Movies 4 and 5). The upon contact with macrophages. Furthermore, RhoA acts as a increase in RhoA activity was concomitant with multiple cell central node of signaling when cells are mechanically stimulated protrusions globally around the tumor cell (Figure 7b; during processes such as cell À cell interactions.39 Consistent Supplementary Figure S7B; white arrows), a previously described with previous data revealing that knocking down RhoA expression indicator of increased RhoA activity.40 To determine whether this led to reduced matrix degradation,37 we found that knocking global increase in RhoA activity led to the formation of down RhoA expression in tumor cells reduced invadopodium invadopodia, we first imaged RhoA biosensor-expressing tumor formation (Supplementary Figures S6A and B) and inhibited the cells for RhoA activity and then subsequently assessed the same ability for macrophages to induce invadopodium formation cells for their ability to form invadopodia upon RhoA activation. (Supplementary Figure S6C). We further found that knocking We found that tumor cells that were in contact with macrophages down RhoA expression resulted in decreased tumor cell intravasa- and exhibited a global increase in RhoA activity formed tion in vitro (Supplementary Figure S6D). We next tested whether significantly more invadopodia than control (Figures 7d and e). constitutive active mutations of RhoA was sufficient to induce Furthermore, these experiments suggest that macrophages invadopodia in the absence of macrophage contact and found induce RhoA GTPase signaling in tumor cells to trigger that this was indeed the case, as tumor cells expressing either invadopodium formation for intravasation. G14V-RhoA or F30L-RhoA constitutive active mutations formed more invadopodia than WT RhoA in the absence of macrophage contact (Supplementary Figure S6E). DISCUSSION We next used a RhoA biosensor in tumor cells to monitor the Although there is a breadth of research exploring how tumor RhoA activation in real time upon physical contact with cells affect endothelial cell À cell adhesion, and cytoskeletal

& 2014 Macmillan Publishers Limited Oncogene (2014) 4203 – 4212 Macrophages induce RhoA in tumor cell intravasation M Roh-Johnson et al 4208

Figure 6. Physical contact between macrophages and tumor cells at blood vessels correlates with increased invadopodium formation in vivo.(a) Immunostained sections from MDA-MB-231 orthotopic tumor. Cortactin (green) is enriched in tumor cell (cells indicated by DAPI, blue) in contact with macrophage (CD68, red) at blood vessel (CD31, white). White arrow marks tumor cell with increased Cortactin intensity when in contact with macrophage. Red arrow marks tumor cell with absence of cortactin immunostaining when not in contact with macrophage. (b) Quantification of the percentage of cortactin-positive cells at blood vessel when in contact and not in contact with macrophages. n ¼ 6 sections from three tumors. found in patients developing distant metastasis may reflect the cell biological mechanism of macrophage-induced invadopodia described in this study.

Invadopodium formation is the first step in tumor cell intravasation Figure 5. Patient-derived breast tumor cells form invadopodia and Our results reveal that invadopodia form and are necessary for contact macrophages during transendothelial migration. (a) TN1-GFP intravasation. Although this phenomenon has been alluded to in cells plated with and without macrophages (red) on 405-labeled the literature, it has never been clearly shown. In vivo studies show gelatin-coated dishes and immunostained with Tks5 (red) and that knocking down genes required for invadopodium formation Cortactin (white). Macrophage is labeled with CMPTX (red) and is such as N-Wasp, Arg and Tks5, or overexpressing proteins marked by asterisk. Arrow marks the invadopodium in TN1 cell in resulting in excess invadopodia, results in decreased distant contact with macrophage. (b) As both the macrophage and Tks5 metastasis and/or number of circulating tumor cells.24,41–43 Several immunostaining are in the same channel, GFP is used to specifically steps are required from primary tumor growth to distant detect tumor cells (TN1-GFP). A dashed line outlines the macrophage metastasis, so it is difficult to ascertain which specific step of (MF) and a solid line outlines the tumor cell (TC). (c) Zoomed image of tumor cell in (a) to reveal invadopodium (Tks5 and Cttn-positive) the metastatic cascade is perturbed when tumor cells are unable in TN1-GFP tumor cell (arrow). (d) Quantification of macrophage- to form invadopodia. By taking an in vitro approach, we are able to induced invadodopdia in TN1-GFP cells. n ¼ 24 (Tn1 alone), n ¼ 10 show that invadopodia do indeed form during transendothelial (TN1 þ MF) cells; two independent experiments. t-test Po0.05. migration, and that invadopodia are also necessary for tumor cell (e) TN1-GFP cells plated on transendothelial migration transwells. transendothelial migration to occur. Furthermore, we confirm our Arrows mark areas of ZO-1 disassembly (white in left panel, purple in findings in vivo and show using a blood vessel marker to clearly merge). (f) IMARIS 3D reconstruction of TN1 cell in (e)toshowaTN1- demarcate where vessels lie in the tumor that tumor cells form GFP (green) cell in z axis penetrating through endothelium, as tight invadopodia at sites of intravasation. Interestingly, invadopodia junctions (pink) are disassembling. (g) z projection of same TN-1-GFP have been suggested to be dispensable for extravasation in vivo,43 cell in (e) undergoing transendothelial migration (endothelium marked by white ZO-1 immunostaining in left panel, purple ZO-1 revealing a potentially marked difference between mechanisms of in merge in right panel) when contacting macrophages (red). intravasation versus extravasation. (h) Cartoon depiction of macrophage-induced TN1-GFP trans- Previous experiments suggest that upon tumor cell attachment migration in (g). (i) Quantification of TN1-GFP cells undergoing onto the endothelium, endothelial cells activate Src and other in vitro intravasation when in contact and not in contact with signaling cascades such as p38 and ERK1/2 leading to a decreased macrophages. n ¼ 7; two independent experiments. endothelial cell À cell adhesion and changes in the endothelial cell actin architecture to facilitate tumor cell crossing.44–50 Whether invadopodia are required to directly disassemble mechanisms of extravasation have been vastly studied, little is cell À cell adhesion is unknown. However, we hypothesize that known about the initial steps of tumor cell intravasation and how the interaction between perivascular macrophages and tumor other cells within the tumor microenvironment affect this process. cells to induce invadopodium formation comprises only the first Our experiments yield an unexpected finding that macrophages step in tumor transmigration, and that invadopodium formation is induce RhoA activity and the subsequent enhanced invadopo- necessary for the tumor cells to penetrate through the basement dium formation in breast cancer cells and that this induction membrane of the blood vessel to then make contact with the requires physical contact between the two cell types. We have underlying endothelial cells. Subsequent downstream steps further shown that this macrophage-induced invadopodium requiring loss of endothelial cell À cell adhesion is necessary for formation facilitates tumor cell intravasation using breast cancer completion of tumor cell crossing. Tumor cells also adopt changes cell lines and patient-derived tumor cells, consistent with the in their cytoskeleton to facilitate transmigration.12,14,51 Consistent hypothesis that invadopodium formation is necessary for tumor with previous endothelial cell À cell adhesion studies, we have cells to degrade the basal lamina overlying blood vessels. These shown that endothelial cells clearly detach from one another as experiments yield mechanistic insight into the formation of TMEM tumor cells cross through the monolayer, and that tumor cells sites in patients who develop distant metastasis. The tripartite primarily undergo paracellular transendothelial migration in our structure involving tumor cells, macrophages and endothelial cells in vitro assays.

Oncogene (2014) 4203 – 4212 & 2014 Macmillan Publishers Limited Macrophages induce RhoA in tumor cell intravasation M Roh-Johnson et al 4209

Figure 7. Macrophages induce RhoA signaling in tumor cells to form invadopodia. (a) RhoA biosensor expressing MDA-MB-231 cells not in contact with macrophages. DIC panels (top) and FRET panels (bottom) reveal no global increase in RhoA activity nor protrusive structures. Time interval is indicated in minutes relative to macrophage addition. DIC panel corresponding to time 0 contains inset of TRITC to show the absence of cell tracker red-labeled macrophage. (b) RhoA biosensor expressing cell in contact with macrophage. DIC panels (top) reveals close association of a macrophage and a tumor cell. DIC panel corresponding to time 0 contains inset of TRITC to reveal cell tracker red-labeled macrophage. FRET panels (bottom) show increased RhoA activity upon macrophage contact as well as multiple protrusions (white arrows). Time interval is indicated in minutes relative to macrophage addition. The pseudocolored scale in (a) and (b) represents ratio limits of black (1.0) to red (1.5) for RhoA activity. Scale bar is 10 mm. (c) Quantification of RhoA biosensor activity in cells contacting macrophages (red) compared with cells not contacting macrophages (blue). Plotted is mean±s.e.m. Red arrow marks time in minutes when macrophages are added. n ¼ 12 cells (contact), n ¼ 15 cells (no contact); three independent experiments. (d) RhoA biosensor-expressing cells were imaged for 1 h, and subsequently fixed and immunostained for Cortactin and Tks5. Asterisk in Tks5 panel marks macrophage. Boxed area in images is shown on right to reveal punctae that are both Tks5 and Cortactin-positive. (e) Quantification of invadopodium formation (Tks5- and Cortactin-positive structures) in RhoA biosensor-expressing cells that contact macrophages versus cells that do not contact macrophages. Plotted are all values, as well as mean±s.e.m, t-test Po0.05. n ¼ 20 (no contact), n ¼ 11 (contact); two independent experiments.

The role of tumor-associated macrophages during tumor-cell tumor cells require physical cell À cell contact to induce RhoA intravasation activation and subsequent invadopodium formation. GTPases are Macrophages are critical factors in metastasis, and several macro- involved in the formation of homotypic cell À cell contacts in vivo phage subtypes reside within the tumor microenvironment. We and in vitro, mainly through regulating the integrity of junctions hypothesize that the perivascular macrophages comprise a specific within a sheet of cells.54–58 Although RhoA activity has been shown macrophage subtype that facilitates tumor cell intravasation by in nascent homotypic cell À cell contacts,59 it remained unclear how inducing invadopodium formation in the tumor cell. This does not heterotypic cell contacts regulate RhoA activity in real time. In preclude any other role for macrophages in inducing perturbations addition, although the EGF/CSF-1 paracrine loop of signaling was inthebloodvesselsorsignalingthatmayoccurbetween identified between tumor cells and macrophages, the intracellular macrophages and endothelial cells during intravasation. Multi- signaling pathways induced by macrophages in the tumor photon-based intravital imaging reveals that, although tumor- microenvironment were elusive. Indeed, the EGF/CSF-1 paracrine associated macrophages are highly motile, the perivascular macro- loop of signaling is also required for both macrophage-induced phages remain stationary, closely docked at blood vessels, and that, invadopodium formation and transendothelial migration in some instances, perivascular macrophages can be seen in close (Supplementary Figure S9). Thus, it is possible that either the association with intravasating tumor cells.4 In this work, we show unprocessed forms of CSF-1 and EGF are involved in macrophage- that macrophages induce RhoA activation and the subsequent induced invadopodium formation, or it remains to be determined formation of invadopodia during transmigration. Certainly, other cell which upstream contact-mediated signaling between the cells in types in the tumor microenvironment also affect tumor cell the tumor microenvironment is important for invadopodium invadopodium formation and intravasation (Supplementary Figure formation during transmigration. We hypothesize that the yet S8), revealing the complexity of the tumor microenvironment. unidentified contact-mediated ligand À receptor pair will activate Interestingly, only those cell types tested that are capable of the RhoA pathway, resulting in increased invadopodium formation inducing invadopodium formation in tumor cells are also capable of in tumor cells at blood vessels. Our results illustrate a novel role for enhancing tumor cell intravasation (Supplementary Figure S8). We RhoA in real time in heterotypic cell À cell contact signaling. The predict that invadopodium formation is necessary for several steps global RhoA increase in the tumor cell, not just at the site of cell of the metastastic cascade, and that there will be different levels of contact, suggests that RhoA signaling stimulates invadopodium regulation by macrophages and other factors in the tumor formation, not merely the location where invadopodia will form. microenvironment at each of these steps. Work exploring upstream signaling pathways regulating RhoA activity during intravasation is currently underway.

A novel heterotypic cell contact role for RhoA signaling Recent advances in imaging RhoA activity in vivo reveal sub- Clinical significance of macrophage-induced intravasation cellular spatial regulation of RhoA activity in cancer models.52,53 Using patient-derived breast tumor cells, we confirmed our Consistent with these findings, we reveal that macrophages and findings that macrophages induce both invadopodium formation

& 2014 Macmillan Publishers Limited Oncogene (2014) 4203 – 4212 Macrophages induce RhoA in tumor cell intravasation M Roh-Johnson et al 4210 and intravasation in vitro. Broadly, the close association of macro- primer 50-ACCCAAGGACAACAACCTGT-30; Tks5 reverse primer 50-AGCGAGC phages and tumor cells at the level of the endothelium lends AGTGCTAAAGGAG-30) was performed as described.43 credence to the finding of TMEM sites in resected tumor tissue of breast cancer patients. Thus, our results support the value of using Invadopodium assay the number of TMEM sites as a prognostic marker of the risk of The 405 gelatin-labeled Mattek dishes were prepared as previously distant metastasis. described.61 Tumor cells were plated in complete media for 4 h on the Alexa 405-labeled gelatin dishes. Dishes were fixed and immunostained for cortactin and tks5 as previously described.28 Cells were imaged with a MATERIALS AND METHODS Deltavision Core Microscope (Applied Precision, Issaquah, WA, USA) using a Cell lines CoolSnap HQ2 camera,  60 1.42 N.A. oil immersion objective and MDA-MB-231 and Jurkat T cells were cultured in 10% FBS/DMEM. MDA-MB- softWorx software. Fixed cells were imaged in PBS at room temperature. 231 cells were serum-starved in 0.5% FBS/0.8% BSA in DMEM for 16 h Invadopodia were detected as punctate structures that were positive for before macrophage induction studies. BAC1.2F5 cells were cultured in 10% both cortactin and tks5 and capable of degrading Alexa 405-gelatin. FBS/MEM supplemented with 2 mML-glutamine, 22 mg/ml L-asparagine and 3000 U/ml of purified human recombinant CSF-1 (generously provided by Macrophage-induced invadopodia assay Richard Stanley, Albert Einstein College of Medicine). Human umbilical vein endothelial cells (HUVECs, Lonza, Allendale, NJ, USA) were cultured in EGM- MDA-MB-231 and TN1-GFP cells were serum-starved for 16 h. BAC1.2F5 2 (Lonza) and only used between passage 2–4. Immortalized bone cells were cell tracker-labeled (CMPTX, Invitrogen). A total of 25K MDA-MB- marrow-derived macrophages23 were cultured in 10% FBS/MEM 231 cells were incubated with 125K BAC1.2F5 cells in serum-starvation media on 405-labeled gelatin-coated dishes for 4 h, fixed and immunos- supplemented with 2 mML-glutamine, 22 mg/ml L-asparagine and 10 000 U/ml of purified human recombinant CSF-1. RAW cells were cultured tained for invadopodium markers as described above. For conditioned in 10% FBS/RPMI. HL-60 cells were cultured and differentiated as media experiments, BAC1.2F5 cells were incubated with complete media, described.60 TN1 cells were isolated and stably labeled to express GFP as with the exception that 300 U/ml CSF-1 was used rather than 3000 U/ml described36 and maintained by passage through orthotopic injections of CSF-1, for 24 h, and subsequently used for assays. Unconditioned complete mice (Supplementary Materials and methods). media with 300 U/ml CSF-1 was used as a control. To assess other cell types’ ability to induce tumor cell invadopodia, 125K of HL-60 cells or Jurkat Tcells were incubated with 25K MDA-MB-231 cells DNA siRNA and transfection and cell labeling in serum-starvation media as described above. A total of 1  106 MDA-MB-231 cells were transfected by 2 mgeachof Cortactin-tagRFP28 and GFP-tks5 (kindly provided by Sara Courtneidge) or by 1.5 mg each of RhoA-WT, RhoA-F30L and RhoA-G14V using the Lonza In vitro intravasation assay Nucleofection Kit V protocol 24 h before the experiment using manufacturer Eight-micron-pore-sized transwell inserts (Millipore) were inverted, and the conditions. Control nonsilencing siRNA was from Qiagen (Valencia, CA, filter was plated with 50 ul of 2.5 mg/ml of growth factor-reduced matrigel USA). Human-specific tks5 and RhoA siGenome Smart Pool were from (BD Biosciences, San Jose, CA, USA) in DMEM media (Invitrogen) for 1 h at Dharmacon (Pittsburgh, PA, USA). A total of 1  106 MDA-MB-231 cells were room temperature. Excess matrigel was removed, and 100K HUVECs were transfected with 2 mM siRNA using the Lonza Nucleofection Kit V 72 h (for plated on the matrigel-coated filters on the inverted transwells in 50 ml tks5) and 96 h (for RhoA) before each experiment. Immunoblot analysis was EGM-2 for 4 h in a 37 1C CO2 incubator. Transwells were then flipped right performed to confirm knockdown for each experiment. BAC1.2F5 and side up into a 24-well plate (Corning, Big Flats, NY, USA), and tumor cells HUVECs were labeled with cell tracker dyes (CMFDA, CMPTX from and macrophages or other cell types were added as described Invitrogen, Carlsbad, CA, USA) before the experiments. Stable cell lines or (Supplementary Materials and methods). To assess the extent of tumor MDA-MB-231-EGFP and MDA-MB-231-dTomato cells were prepared as cell transendothelial migration, transwells were washed twice with PBS and described,16 with the exception that dTomato was inserted into the EGFP then fixed in 4% paraformaldehyde/PBS for 20 min, permeabilized with site in the EGFP-C1 vector (Clontech, Mountainview, CA, USA). 0.1% Triton X-100 for 10 min and stained with rhodamine–phalloidin (Invitrogen) and DAPI for 1 h. Transwells were placed in PBS on a Mattek dish (MatTek Corp, Ashland, MA, USA) and imaged on an inverted Cloning RhoA constitutive active mutants Multiphoton Olympus FV1000-MPE microscope (Olympus, Center Valley, Expression constructs for the RhoA F30L and G14V mutants were produced PA, USA) with a  25 NA 1.05 water immersion objective and a Spectra and cloned into the pTRIEX-4 backbone (Novagen, Billerica, MA, USA) as Physics (Irvine, CA, USA) Mai Tai-DeepSee laser set to 880 nm. Two- described (Supplementary Materials and methods). micrometer step Z-stacks ofen face views were acquired throughout the depth of the transwell, and only those tumor cells that breached the Inhibitors and blocking antibodies endothelium were scored as a positive transendothelial migration event. Approximately six fields of view were acquired for each transwell. For in vitro transendothelial migration and invadopodia formation assays, Transwells were immunostained with ZO-1 (Invitrogen) and Collagen IV the mouse CSF-1 receptor was blocked with antimouse CSF-1R blocking (Abcam, Cambridge, MA, USA) antibodies following a previously described antibody (AFS98, Novus Biologicals, Littleton, CO, USA) at a concentration protocol62 and imaged as described (Supplementary Materials and of 5 mM, and a Rat IgG1 k isotype control blocking antibody was also used methods). at 5 mM. To block the EGF receptor, Iressa (AstraZeneca, Wilmington, DE, To live image the transwells, transwells were set up as described above, USA) was used at 5 mM. To block matrix metalloproteases, GM6001 (BML- and, before imaging on an inverted Multiphoton Olympus FV1000-MPE E1300-0001, Enzo Life Sciences, Farmingdale, NY, USA) was used at 5 mM. microscope with a  25 NA 1.05 water immersion objective and a Spectra Physics Mai Tai-DeepSee laser set to 880 nm, the transwells were placed in Western blotting and quantification Mattek dishes with L-15/10% FBS. Two-micromete step Z-stacks, encom- Cells were transfected with tks5, RhoA or control siRNA for 72 or 96 h, lysed passing the most apical 10 microns of the transwells were taken every with SDS-PAGE sample buffer, sonicated and boiled at 95 1C. To maintain 10 min for 3 h. TN1 cells, tumors of B1 cm in diameter were excised and trimmed, mechanically dissociated with scalpels and then enzymatically digested for Transwell permeability and TEER measurements 1 hour at 48 1C (Liberase TH, Roche, Indianapolis, IN, USA). Samples were filtered twice into a single-cell suspension in PBS/2% FBS, red blood cells Endothelial monolayers were grown on transwells as described above. lysed with ACK buffer (Invitrogen, A10492-01) and cells washed twice with A concentration of 1 mg/ml 60K texas red dextran was added to the top PBS/2% FBS on ice. well. A volume of 50 ul of media was taken from the bottom well after 30 and 60 min, and fluorescence measurements were determined with the described settings for costar 96-well plates, 555/640 using a RNA extraction and PCR amplification Fluostar Optima (BMG LabTech, Cary, NC, USA). Transendothelial electrical RNA was extracted from triplicate plates of control and tks5 siRNA 72 h resistance measurements were recorded on the same transwells using an knockdown plates with the RNeasy Mini kit (Qiagen). The RNA was EVOM2 epithelial Voltohmmeter (World Precision Instruments, Sarasota, evaluated, reverse-transcribed and quantitative PCR analysis (Tks5 forward FL, USA).

Oncogene (2014) 4203 – 4212 & 2014 Macmillan Publishers Limited Macrophages induce RhoA in tumor cell intravasation M Roh-Johnson et al 4211 In vivo imaging and immunostaining transendothelial migration in a mouse model of breast cancer. J Cell Sci 2011; All experiments conducted in mice were in accordance with the National 124(Pt 13): 2120–2131. Institutes of Health regulation on the use and care of experimental animals 2 Patsialou A, Bravo-Cordero JJ, Wang Y, Entenberg D, Liu H, Clarke MF et al. and approved by the Albert Einstein College of Medicine Animal Use Intravital multiphoton imaging reveals multicellular streaming as a crucial com- Committee. MDA-MB-231 orthotopic tumors were generated by injecting ponent of in vivo cell migration in human breast tumors. IntraVital 2013 (in press). MDA-MB-231 cells expressing GFP in sterile PBS/20% collagen I into the 3 Wyckoff J, Wang W, Lin EY, Wang Y, Pixley F, Stanley ER et al. A paracrine loop mammary fat pad of severe combined immunodeficiency mice (SCID) mice between tumor cells and macrophages is required for tumor cell migration in (NCI, Frederick, MD, USA). Tumors grew for B3 months until they attained mammary tumors. Cancer Res 2004; 64: 7022–7029. a size of 1 cm. To image tumor cell interactions with the blood vessel 4 Wyckoff JB, Wang Y, Lin EY, Li JF, Goswami S, Stanley ER et al. Direct visualization in vivo, mice were injected through tail vein with 50 ml of rhodamine- of macrophage-assisted tumor cell intravasation in mammary tumors. Cancer Res labeled Ricinus Communis Agglutinin I (Vector Labs). Animals were killed 2007; 67: 2649–2656. immediately after injection, the tumors were removed and imaged in PBS 5 Zervantonakis IK, Hughes-Alford SK, Charest JL, Condeelis JS, Gertler FB, on an Olympus FV1000-MPE microscope with a  25 NA 1.05 water Kamm RD. Three-dimensional microfluidic model for tumor cell intravasation immersion objective and a Spectra Physics Mai Tai-DeepSee laser set to and endothelial barrier function. Proc Natl Acad Sci USA 2012; 109: 880 nm. Z-stacks of 2 um step size were taken for each field of view. For 13515–13520. immunostaining of tumor sections, tumors were removed and frozen on 6 Robinson BD, Sica GL, Liu YF, Rohan TE, Gertler FB, Condeelis JS et al. Tumor dry ice in OCT Compound (TissueTek, VWR, Radnor, PA, USA), sectioned, microenvironment of metastasis in human breast carcinoma: a potential prog- fixed, immunostained and imaged as described (Supplementary Materials nostic marker linked to hematogenous dissemination. Clin Cancer Res 2009; 15: and methods). 2433–2441. 7 Reymond N, Im JH, Garg R, Vega FM, Borda d’Agua B, Riou P et al. Cdc42 promotes RhoA biosensor imaging and quantification transendothelial migration of cancer cells through beta1 integrin. J Cell Biol 2012; MDA-MB-231 cells were transiently transfected with the RhoA biosensor40 199: 653–668. using the Lonza Nucleofection protocol described above and plated directly 8 Reymond N, Riou P, Ridley AJ. Rho GTPases and cancer cell transendothelial onto MatTek dishes, serum starved overnight and imaged on a single migration. Methods Mol Biol 2012; 827: 123–142. plane for three-color imaging every 2 min using a previously described 9 Jin F, Brockmeier U, Otterbach F, Metzen E. New insight into the SDF-1/CXCR4 axis protocol.38,63 Ratiometric calculations on the CFP and YFP FRET emissions of in a breast carcinoma model: hypoxia-induced endothelial SDF-1 and tumor cell RhoA were performed as described.38,40 Cells were subsequently fixed and CXCR4 are required for tumor cell intravasation. Molecular Cancer Res 2012; 10: immunostained using methods described above and quantified with the 1021–1031. DeltaVision Core microscope using methods described above. For more 10 Haidari M, Zhang W, Caivano A, Chen Z, Ganjehei L, Mortazavi A et al. information, see Supplementary Materials and methods. Integrin alpha2beta1 mediates tyrosine phosphorylation of vascular endothelial cadherin induced by invasive breast cancer cells. J Biol Chem 2012; 287: 32981– 32992. Statistical Analysis 11 Haidari M, Zhang W, Chen Z, Ganjehei L, Warier N, Vanderslice P et al. Myosin Statistical analysis was conducted using an unpaired, two-tailed Student’s light chain phosphorylation facilitates monocyte transendothelial migration by t-test when comparing two data sets. Statistical significance was defined as dissociating endothelial adherens junctions. Cardiovascular Res 2011; 92: a Po0.05. For multiple comparisons, ANOVA was used. All graphs are 456–465. displayed as mean±s.e.m, unless otherwise indicated. The ‘n’ in s.e.m. was 12 Voura EB, Sandig M, Kalnins VI, Siu C. Cell shape changes and cytoskeleton derived from the number of fields of view (for tumor cell intravasation) or reorganization during transendothelial migration of human melanoma cells. number of cells (for invadopodium formation). Cell Tissue Res 1998; 293: 375–387. 13 Voura EB, Sandig M, Siu CH. Cell-cell interactions during transendothelial migra- tion of tumor cells. Microscopy Res Tech 1998; 43: 265–275. CONFLICT OF INTEREST 14 Stoletov K, Montel V, Lester RD, Gonias SL, Klemke R. High-resolution imaging of JC holds equity in and is a member of the Scientific Advisory Board of MetaStat, Inc. the dynamic tumor cell vascular interface in transparent zebrafish. Proc Nat Acad Sci USA 2007; 104: 17406–17411. 15 Tremblay PL, Auger FA, Huot J. Regulation of transendothelial migration of colon ACKNOWLEDGEMENTS cancer cells by E-selectin-mediated activation of p38 and ERK MAP kinase- sOncogene 2006; 25: 6563–6573. We thank Brian Beaty for advice regarding invadopodium experiments; David 16 Patsialou A, Wyckoff J, Wang Y, Goswami S, Stanley ER, Condeelis JS. Invasion of Entenberg, Vera Desmarais, Jeffrey Wyckoff and the Analytical Imaging Facility for help human breast cancer cells in vivo requires both paracrine and autocrine loops with microscopy; Richard Stanley, Fernando Macian, Esther Arwert, Allison Harney and involving the colony-stimulating factor-1 receptor. Cancer Res 2009; 69: Veronika Micolski for guidance and reagents; Yarong Wang for help with animal 9498–9506. injections; and the Albert Einstein shRNA Core, the Flow Cytometry Core facility and 17 Das SK, Stanley ER, Guilbert LJ, Forman LW. Human colony-stimulating factor the Histology and Comparative Pathology Core for help with experiments. We also (CSF-1) radioimmunoassay: resolution of three subclasses of human colony- thank Jeffrey Segall, Dianne Cox, Anne Bresnick and Aviv Bergmann for helpful stimulating factors. Blood 1981; 58: 630–641. discussions, and the Condeelis, Segall, Cox and Hodgson labs for advice and reagents. 18 Koop S, MacDonald IC, Luzzi K, Schmidt EE, Morris VL, Grattan M et al. Fate of This work was funded by NIH CA150344 (JC), NIH CA100324 (AP), NIH GM093121 (LH melanoma cells entering the microcirculation: over 80% survive and extravasate. and JJB-C), Post-doctoral fellowship from Susan G. Komen for the Curer KG111405 Cancer Res 1995; 55: 2520–2523. (VPS) and NIH CA159663 (MRJ). 19 Kramer RH, Nicolson GL. Interactions of tumor cells with vascular endothelial cell monolayers: a model for metastatic invasion. Proc Natl Acad Sci USA 1979; 76: 5704–5708. AUTHOR CONTRIBUTIONS 20 Nicolson GL. Metastatic tumor cell attachment and invasion assay utilizing MR-J and JC conceived the project idea. MR-J, JJB-C and LH performed RhoA vascular endothelial cell monolayers. J Histochem Cytochem 1982; 30: 214–220. biosensor experiments; LH cloned RhoA mutant constructs; AP maintained TN1- 21 Ohigashi H, Shinkai K, Mukai M, Ishikawa O, Imaoka S, Iwanaga T et al. In vitro GFP cells, created the MDA-MB-231 dtomato cell line and performed qRT-PCR; VPS invasion of endothelial cell monolayer by rat ascites hepatoma cells. Jap J Cancer optimized protocols regarding invadopodium assays and helped with microscopy; Res 1989; 80: 818–821. 22 Proebstl D, Voisin MB, Woodfin A, Whiteford J, D’Acquisto F, Jones GE et al. PG and MR-J performed IMARIS 3D reconstructions; and HL created the TN1-GFP Pericytes support neutrophil subendothelial cell crawling and breaching of cells. MR-J performed the rest of the experiments. All authors contributed to venular walls in vivo. J Exp Med 2012; 209: 1219–1234. discussions related to the manuscript. MR-J and JC wrote the manuscript. 23 Yu W, Chen J, Xiong Y, Pixley FJ, Dai XM, Yeung YG et al. CSF-1 receptor structure/ function in MacCsf1r À / À macrophages: regulation of proliferation, differentia- tion, and morphology. J Leukoc Biol 2008; 84: 852–863. REFERENCES 24 Gligorijevic B, Wyckoff J, Yamaguchi H, Wang Y, Roussos ET, Condeelis J. N-WASP- 1 Roussos ET, Balsamo M, Alford SK, Wyckoff JB, Gligorijevic B, Wang Y et al. mediated invadopodium formation is involved in intravasation and lung Mena invasive (MenaINV) promotes multicellular streaming motility and metastasis of mammary tumors. J Cell Sci 2012; 125(Pt 3): 724–734.

& 2014 Macmillan Publishers Limited Oncogene (2014) 4203 – 4212 Macrophages induce RhoA in tumor cell intravasation M Roh-Johnson et al 4212 25 Yamaguchi H. Pathological roles of invadopodia in cancer invasion and 45 Hu Y, Szente B, Kiely JM, Gimbrone Jr. MA. Molecular events in transmembrane metastasisEur J Cell Biol 2012; 91(11-12): 902–907. signaling via E-selectin. SHP2 association, adaptor protein complex formation and 26 Saltel F, Daubon T, Juin A, Ganuza IE, Veillat V, Genot E. Invadosomes: intriguing ERK1/2 activation. J Biol Chem 2001; 276: 48549–48553. structures with promise. Eur J Cell Biol 2011; 90(2-3): 100–107. 46 Laferriere J, Houle F, Taher MM, Valerie K, Huot J. Transendothelial migration of 27 Linder S, Wiesner C, Himmel M. Degrading devices: invadosomes in proteolytic colon carcinoma cells requires expression of E-selectin by endothelial cells and cell invasion. Ann Rev Cell Dev Biol 2011; 27: 185–211. activation of stress-activated protein kinase-2 (SAPK2/p38) in the tumor cells. 28 Mader CC, Oser M, Magalhaes MA, Bravo-Cordero JJ, Condeelis J, Koleske AJ et al. J Biol Chem 2001; 276: 33762–33772. An EGFR-Src-Arg-cortactin pathway mediates functional maturation of invado- 47 Rousseau S, Houle F, Landry J, Huot J. p38 MAP kinase activation by vascular podia and breast cancer cell invasion. Cancer Res 2011; 71: 1730–1741. endothelial growth factor mediates actin reorganization and cell migration in 29 Oser M, Mader CC, Gil-Henn H, Magalhaes M, Bravo-Cordero JJ, Koleske AJ et al. human endothelial cells. Oncogene 1997; 15: 2169–2177. Specific tyrosine phosphorylation sites on cortactin regulate Nck1-dependent 48 Walz G, Aruffo A, Kolanus W, Bevilacqua M, Seed B. Recognition by ELAM-1 of the actin polymerization in invadopodia. J Cell Sci 2010; 123(Pt 21): 3662–3673. sialyl-Lex determinant on myeloid and tumor cells. Science 1990; 250: 1132–1135. 30 Seals DF, Azucena Jr. EF, Pass I, Tesfay L, Gordon R, Woodrow M et al. The adaptor 49 Weiss TW, Mehrabi MR, Kaun C, Zorn G, Kastl SP, Speidl WS et al. Prostaglandin E1 protein Tks5/Fish is required for podosome formation and function, and for the induces vascular endothelial growth factor-1 in human adult cardiac myocytes protease-driven invasion of cancer cells. Cancer Cell 2005; 7: 155–165. but not in human adult cardiac fibroblasts via a cAMP-dependent mechanism. 31 Bowden ET, Barth M, Thomas D, Glazer RI, Mueller SC. An invasion-related com- J Mol Cell Cardiol 2004; 36: 539–546. plex of cortactin, paxillin and PKCmu associates with invadopodia at sites of 50 Yoshida M, Takano Y, Sasaoka T, Izumi T, Kimura A. E-selectin polymorphism extracellular matrix degradation. Oncogene 1999; 18: 4440–4449. associated with myocardial infarction causes enhanced leukocyte-endothelial 32 Clark ES, Whigham AS, Yarbrough WG, Weaver AM. Cortactin is an essential interactions under flow conditions. Arterioscler Thromb Vasc Biol 2003; 23: regulator of matrix metalloproteinase secretion and extracellular matrix degra- 783–788. dation in invadopodia. Cancer Res 2007; 67: 4227–4235. 51 Li B, Zhao WD, Tan ZM, Fang WG, Zhu L, Chen YH. Involvement of Rho/ROCK 33 Magalhaes MA, Larson DR, Mader CC, Bravo-Cordero JJ, Gil-Henn H, Oser M et al. signalling in small cell lung cancer migration through human brain microvascular Cortactin phosphorylation regulates cell invasion through a pH-dependent endothelial cells. FEBS Letts 2006; 580: 4252–4260. 52 McGhee EJ, Morton JP, Von Kriegsheim A, Schwarz JP, Karim SA, Carragher NO pathway. J Cell Biol 2011; 195: 903–920. et al. FLIM-FRET imaging in vivo reveals 3D-environment spatially regulates 34 Beaty BT, Sharma VP, Bravo-Cordero JJ, Simpson MA, Eddy RJ, Koleske AJ et al. RhoGTPase activity during cancer cell invasion. Small GTPases 2011; 2: 239–244. beta1 integrin regulates Arg to promote invadopodial maturation and matrix 53 Timpson P, McGhee EJ, Morton JP, von Kriegsheim A, Schwarz JP, Karim SA et al. degradation. Mol Biol Cell 2013; 24: 1661–1675. Spatial regulation of RhoA activity during pancreatic cancer cell invasion driven 35 Bravo-Cordero JJ, Marrero-Diaz R, Megias D, Genis L, Garcia-Grande A, Garcia MA by mutant p53. Cancer Res 2011; 71: 747–757. et al. MT1-MMP proinvasive activity is regulated by a novel Rab8-dependent 54 Carmona-Fontaine C, Matthews HK, Kuriyama S, Moreno M, Dunn GA, Parsons M exocytic pathway. EMBO J 2007; 26: 1499–1510. et al. Contact inhibition of locomotion in vivo controls neural crest directional 36 Liu H, Patel MR, Prescher JA, Patsialou A, Qian D, Lin J et al. Cancer stem cells from migration. Nature 2008; 456: 957–961. human breast tumors are involved in spontaneous metastases in orthotopic 55 Ponik SM, Trier SM, Wozniak MA, Eliceiri KW, Keely PJ. RhoA is down-regulated at mouse models. Proc Natl Acad Sci USA 2010; 107: 18115–18120. cell-cell contacts via p190RhoGAP-B in response to tensional homeostasis. Mol 37 Sakurai-Yageta M, Recchi C, Le Dez G, Sibarita JB, Daviet L, Camonis J et al. The Biol Cell 2013; 24: 1688–1699. interaction of IQGAP1 with the exocyst complex is required for tumor cell inva- 56 Witzel S, Zimyanin V, Carreira-Barbosa F, Tada M, Heisenberg CP. Wnt11 controls sion downstream of Cdc42 and RhoA. J Cell Biol 2008; 181: 985–998. cell contact persistence by local accumulation of Frizzled 7 at the plasma mem- 38 Bravo-Cordero JJ, Oser M, Chen X, Eddy R, Hodgson L, Condeelis J. A novel brane. J Cell Biol 2006; 175: 791–802. spatiotemporal RhoC activation pathway locally regulates cofilin activity at 57 Arthur WT, Noren NK, Burridge K. Regulation of Rho family GTPases by cell-cell invadopodia. Curr Biol 2011; 21: 635–644. and cell-matrix adhesion. Biol Res 2002; 35: 239–246. 39 Lessey EC, Guilluy C, Burridge K. From mechanical force to RhoA activation. Bio- 58 Nethe M, de Kreuk BJ, Tauriello DV, Anthony EC, Snoek B, Stumpel T et al. Rac1 chemistry 2012; 51: 7420–7432. acts in conjunction with Nedd4 and dishevelled-1 to promote maturation of cell- 40 Pertz O, Hodgson L, Klemke RL, Hahn KM. Spatiotemporal dynamics of RhoA cell contacts. J Cell Sci 2012; 125(Pt 14): 3430–3442. activity in migrating cells. Nature 2006; 440: 1069–1072. 59 Omelchenko T, Hall A. Myosin-IXA regulates collective epithelial cell migration by 41 Blouw B, Seals DF, Pass I, Diaz B, Courtneidge SA. A role for the podosome/ targeting RhoGAP activity to cell-cell junctions. Curr Biol 2012; 22: 278–288. invadopodia scaffold protein Tks5 in tumor growth in vivo. Eur Cell Biol 2008; 87(8- 60 Sharma VP, DesMarais V, Sumners C, Shaw G, Narang A. Immunostaining evidence 9): 555–567. for PI(4,5)P2 localization at the leading edge of chemoattractant-stimulated HL-60 42 Lovett DH, Cheng S, Cape L, Pollock AS, Mertens PR. YB-1 alters MT1-MMP traf- cells. J Leukoc Biol 2008; 84: 440–447. ficking and stimulates MCF-7 breast tumor invasion and metastasis. Biochem 61 Sharma VP, Entenberg D, Condeelis J. High-resolution live-cell imaging and time- Biophys Res Commun 2010; 398: 482–488. lapse microscopy of invadopodium dynamics and tracking analysis. Methods Mol 43 Gil-Henn H, Patsialou A, Wang Y, Warren MS, Condeelis JS, Koleske AJ. Arg/Abl2 Biol 2013; 1046: 343–357. promotes invasion and attenuates proliferation of breast cancer in vivo. Oncogene 62 Lipschutz JH, O’Brien LE, Altschuler Y, Avrahami D, Nguyen Y, Tang K et al. 2013; 32: 2622–2630. Analysis of membrane traffic in polarized epithelial cells. Curr Protoc Cell BiolS 44 Hu Y, Kiely JM, Szente BE, Rosenzweig A, Gimbrone Jr. MA. E-selectin-dependent 2001; 15.5.1–15.5.18. signaling via the mitogen-activated protein kinase pathway in vascular endo- 63 Spiering D, Hodgson L. Multiplex imaging of Rho family GTPase activities in living thelial cells. J Immunol 2000; 165: 2142–2148. cells. Methods Mol Biol 2012; 827: 215–234.

Supplementary Information accompanies this paper on the Oncogene website (http://www.nature.com/onc)

Oncogene (2014) 4203 – 4212 & 2014 Macmillan Publishers Limited