© 2017. Published by The Company of Biologists Ltd | Journal of Cell Science (2017) 130, 1064-1077 doi:10.1242/jcs.195552

RESEARCH ARTICLE A RhoG-mediated signaling pathway that modulates invadopodia dynamics in breast cancer cells Silvia M. Goicoechea, Ashtyn Zinn, Sahezeel S. Awadia, Kyle Snyder and Rafael Garcia-Mata*

ABSTRACT micropinocytosis, bacterial uptake, and leukocyte One of the hallmarks of cancer is the ability of tumor cells to invade trans-endothelial migration (deBakker et al., 2004; Ellerbroek et al., surrounding tissues and metastasize. During metastasis, cancer cells 2004; Jackson et al., 2015; Katoh et al., 2006, 2000; van Buul et al., degrade the extracellular matrix, which acts as a physical barrier, by 2007). Recent studies have revealed that RhoG plays a role in tumor developing specialized -rich membrane protrusion structures cell invasion and may contribute to the formation of invadopodia called invadopodia. The formation of invadopodia is regulated by Rho (Hiramoto-Yamaki et al., 2010; Kwiatkowska et al., 2012). , a family of that regulates the actin . Invadopodia are actin-rich adhesive structures that form in the Here, we describe a novel role for RhoG in the regulation of ventral surface of cancer cells and allow them to degrade the invadopodia disassembly in human breast cancer cells. Our results extracellular matrix (ECM) (Gimona et al., 2008). Formation of show that RhoG and Rac1 have independent and opposite roles invadopodia involves a series of steps that include the disassembly in the regulation of invadopodia dynamics. We also show that SGEF of focal adhesions and stress fibers, and the relocalization of several (also known as ARHGEF26) is the exchange factor responsible of their components into the newly formed invadopodia (Hoshino for the activation of RhoG during invadopodia disassembly. When et al., 2012; Oikawa et al., 2008). Invadopodia assembly starts with the expression of either RhoG or SGEF is silenced, invadopodia the formation of actin- and cortactin-rich puncta, followed by the are more stable and have a longer lifetime than in control cells. recruitment of adhesion proteins, such as vinculin and paxillin, and Our findings also demonstrate that RhoG and SGEF modulate the proteinases that allow ECM degradation (Hoshino et al., 2013). of paxillin, which plays a key role during invadopodia Even though many of the molecular components required for disassembly. In summary, we have identified a novel signaling invadopodia formation have been identified, the signaling pathways pathway involving SGEF, RhoG and paxillin phosphorylation, which that regulate these events are still poorly understood (Linder et al., functions in the regulation of invadopodia disassembly in breast 2011). Invadopodia formation is controlled by the integrated cancer cells. activity of several GTPases, including Cdc42, RhoA, RhoC and Rac1 (Spuul et al., 2014). Cdc42 promotes invadopodia formation KEY WORDS: RhoG, Invadopodia, SGEF, Guanine-nucleotide in almost every system tested (Ayala et al., 2009; Di Martino et al., exchange factors, Src, Paxillin, Rac1 2014; Moreau et al., 2006, 2003; Nakahara et al., 2003; Tatin et al., 2006). However, generalized conclusions cannot be drawn from INTRODUCTION studies of other Rho GTPases, since their activities can both inhibit Rho GTPases control many aspects of cell behavior ranging from or promote invadopodia formation depending on the experimental the regulation of cytoskeletal organization, cell motility and cell conditions or cell type used (Spuul et al., 2014). polarity, to nuclear expression and control of In this study, we describe a novel function for RhoG in the (Hodge and Ridley, 2016). Rho proteins cycle between an active regulation of invadopodia dynamics in human breast cancer cells. (GTP-bound) and an inactive (GDP-bound) state. The activation of Our findings describe a signaling pathway involving SGEF and Rho proteins involves the exchange of GDP for GTP, which is RhoG that regulates invadopodia disassembly independently of catalyzed by specific guanine-nucleotide-exchange factors (GEFs). Rac1 through regulation of paxillin phosphorylation. Once activated, Rho GTPases interact with a wide variety of downstream effectors to modulate their activity and/or localization. RESULTS The hydrolysis of GTP to GDP, a reaction that is stimulated by RhoG is a negative regulator of invadopodia formation GTPase-activating proteins (GAPs), inactivates the GTPases and We have previously shown that the breast cancer cell line SUM159 terminates the signal. With more than 80 Rho GEFs, 70 Rho GAPs forms invadopodia when treated with phorbol esters such phorbol and over 100 effectors, cells regulate the activity of Rho proteins 12,13-dibutyrate (PDBu) and 12-O-tetradecanoylphorbol-13- through multiple pathways, thus acting as key nodes for signal acetate (PMA), and that the formation of these structures integration and dissemination (Bustelo et al., 2007; Rossman et al., correlates with their metastatic potential (Goicoechea et al., 2009). 2005; Tcherkezian and Lamarche-Vane, 2007). To determine the requirement of RhoG in invadopodia formation, RhoG, a Rho related to Rac, has been associated we generated stable SUM159 cell lines in which RhoG expression with , outgrowth, dynamics, was silenced using lentivirally encoded shRNA. We used two different RhoG-specific shRNAs (shRNA#1 and shRNA#4) to rule Department of Biological Sciences, University of Toledo, Toledo, OH 43606, USA. out off-target effects. A stable cell line expressing a non-targeting shRNA was used as control (CTRL). Fig. 1A shows that RhoG *Author for correspondence ([email protected]) silencing was efficient, especially for shRNA #4. SUM159 cells are S.M.G., 0000-0003-1107-389X; R.G.-M., 0000-0002-7116-4411 able to form invadopodia spontaneously, so we first looked at the effect of RhoG depletion in untreated cells. To identify invadopodia

Received 22 July 2016; Accepted 14 January 2017 structures, we stained the cells with the invadopodia markers Journal of Cell Science

1064 RESEARCH ARTICLE Journal of Cell Science (2017) 130, 1064-1077 doi:10.1242/jcs.195552

Fig. 1. RhoG negatively regulates invadopodia formation in SUM159 cells. (A) Cell lysates from SUM159 cells stably expressing non-targeting (CTRL) or RhoG-specific shRNAs (shRNA#1 and shRNA #4) were analyzed by western blotting and probed for RhoG and Rac1, and for as a loading control. (B) Quantification of cortactin- and actin-containing invadopodia in CTRL and RhoG KD cells. Results are expressed as the percentage of cells with invadopodia. Data are mean±s.e.m. for at least three independent experiments. Cells were either untreated, or treated with PDBu or PMA. (C) CTRL or RhoG KD SUM159 cells were treated with PDBu for 30 min and stained with anti-cortactin antibody (red), Alexa-Fluor-488–phalloidin (green) and Hoechst 33342 (blue). Arrowheads indicate representative invadopodia. (D) Cell lysates from CTRL, RhoG KD and RhoG KD cells expressing shRNA resistant Myc– RhoG (rescue) were immunoblotted with anti-RhoG and -Myc antibodies. Tubulin was used as a loading control. (E) Quantification of cortactin- and actin-containing invadopodia in CTRL, RhoG KD and rescue cells. Data are mean±s.e.m. of three independent experiments. (F) Cell lysates from CTRL, Myc–RhoG wt (RhoG wt) or Myc–RhoG Q61L (RhoG Q61L) cells were immunoblotted with anti-RhoG and -Myc antibodies. Tubulin was used as a loading control. (G) SUM159 cells transiently transfected with Myc- tagged wild-type (RhoG wt) or constitutively active RhoG (RhoG Q61L) were treated with PDBu for 30 min and stained with anti-cortactin antibody (red), anti-Myc antibody (green), Alexa-Fluor-647–phalloidin (magenta) and Hoechst 33342 (blue). Arrowheads indicate transfected cells and arrows indicate representative invadopodia. (H) Invadopodia were quantified in CTRL (non-transfected cells), and RhoG wt and RhoG Q61L transfected cells. Results are representative of three independent experiments in which at least 100 cells per experiment were counted. Data are mean±s.e.m. (error bars). Scale bars: 10 µm. *P<0.05; **P<0.01; ***P<0.001; ns, not significant.

cortactin and actin. We found that silencing RhoG induced a depletion in cells treated with phorbol esters. In PDBu-treated cells, significant increase in the number of spontaneous invadopodia the number of cells with invadopodia increased significantly in when compared to CTRL cells, from 7% in CTRL cells to 14% and RhoG-knockdown (KD) cells, from 35% in CTRL cells to 64% 23% in RhoG shRNA#1 and shRNA#4 cells, respectively (Fig. 1B, and 81% in RhoG shRNA#1 and shRNA#4 cells respectively untreated; Fig. S1A). We next looked at the effect of RhoG (Fig. 1B,C). Similarly, the percentage of cells with invadopodia Journal of Cell Science

1065 RESEARCH ARTICLE Journal of Cell Science (2017) 130, 1064-1077 doi:10.1242/jcs.195552 increased from 31% in CTRL cells to 79% of cells in PMA-treated SUM159 cells, using specific shRNAs. We tested two independent RhoG KD cells (Fig. 1B; Fig. S1B). Based on these results, and shRNAs for each of the GEFs and compared them to cells unless otherwise indicated, we used shRNA#4 for the rest of these expressing a non-targeting shRNA (CTRL). The KD efficiency was studies (referred to as RhoG KD). We next performed a rescue verified either by quantitative real-time PCR (qRT-PCR; Fig. 3A; experiment by re-expressing shRNA-resistant Myc-RhoG in RhoG Fig. S3A) or western blotting (Fig. S3B). Cells were treated with KD cells (Rescue). Myc–RhoG expresses at levels that are PDBu, and invadopodia structures were identified in cells stained comparable to endogenous RhoG levels in RhoG KD cells, and for cortactin and actin (Fig. 3B; Fig. S3C). Our results show that the restores the cells ability to form invadopodia to control levels number of cells with invadopodia increased significantly when (Fig. 1D,E). We also characterized the effects of silencing RhoG on SGEF was silenced (from 42% in CTRL cells, to 70% and 85% invadopodia formation in two additional human breast cancer cell upon treatment with SGEF shRNA#4 and shRNA#5, respectively, lines, invasive MDA-MB-231, commonly used as a model to study Fig. 3C), whereas silencing PLEKHG6, ephexin 4 or Trio had no invadopodia (Fig. S2A,B), and non-invasive MCF7 cells (Fig. S2C,D). significant effect on invadopodia numbers (Fig. S3D). The In both cell lines, silencing RhoG significantly increased the observed in SGEF KD cells was similar to that number of cells with invadopodia, which suggest the effects observed in RhoG KD cells, both in terms of the number of cells observed are not cell line specific. To investigate the role of RhoG in that formed invadopodia and their appearance (Figs 3B and 1C). more detail, we overexpressed either Myc-tagged wild-type (RhoG Based on these results, we used shRNA#5 for the rest of these wt) or constitutively active RhoG (RhoG Q61L) in SUM159 cells studies (referred to as SGEF KD). To determine whether SGEF (Fig. 1F) and tested the ability of the cells to form invadopodia function was mediated through RhoG we attempted to rescue the (Fig. 1G). Overexpression of either RhoG wt or RhoG Q61L effects of RhoG KD on invadopodia by expressing exogenous significantly inhibited the formation of invadopodia (Fig. 1H). As Myc–SGEF in RhoG KD cells. Our results showed that Myc–SEGF expected, the Q61L mutant, which is locked in an active expression could not rescue the RhoG KD phenotype (Fig. 3D,E), conformation had a greater inhibitory effect than the wild type. suggesting that the role of SGEF in invadopodia is mostly mediated Taken together, our results suggest a novel role for RhoG as a by RhoG. Consistent with the results showed in Fig. 1H, negative regulator of invadopodia formation in breast cancer cells. overexpression of Myc–SGEF wt, significantly inhibited the formation of invadopodia (Fig. 3F,G). In contrast, overexpression Enhanced invadopodia formation after RhoG depletion of a catalytically inactive mutant of SGEF (Myc–SGEF 446/621, requires Src activity which has E446A and N621A ) (Ellerbroek et al., 2004) Phorbol esters mimic diacylglycerol (DAG), which activates protein had no effect (Fig. 3F,G), which demonstrates that SGEF exchange C (PKC) family members. The role and positioning of PKC activity, and thus RhoG activation, is required. Only 5% of the cells in these pathways remains largely unknown, but it involves the transfected with Myc–SGEF wt formed invadopodia, whereas cells function of Src family (Hai et al., 2002; Tatin et al., 2006). transfected with the Myc–SGEF 446/621 show invadopodia levels Because both FAK and Src activities have been shown to be comparable to those of non-transfected cells (∼30%) (Fig. 3F). upregulated to promote invadopodia formation, and both play Interestingly, Myc–SGEF 446/621 localized to invadopodia critical roles in cell invasion (Murphy and Courtneidge, 2011), we (Fig. 3G). We also observed Myc–SGEF wt at invadopodia, wanted to determine whether RhoG modulates invadopodia although at a much lower frequency (∼1.25% of Myc–SGEF wt formation through a FAK- and/or Src-dependent pathway (FAK is transfected cells) (Fig. S3E). This suggests that SGEF could localize also known as PTK2). Western blot analysis showed that there was transiently to invadopodia and be released to the cytoplasm after the no difference in the expression level of endogenous FAK and Src exchange reaction is completed. In the absence of catalytic activity, between CTRL and RhoG KD cells (Fig. 2A,B). However, the SGEF may not be efficiently released and would thus be found at activity of both increased by ∼2-fold in the absence of RhoG invadopodia. (Fig. 2A,B). We next assayed the ability of CTRL and RhoG KD Our results suggest that RhoG activity needs tight regulation cells to form invadopodia when either Src or FAK were inhibited. during invadopodia formation, so we next investigated the kinetics Our results show that inhibition of Src with PP2 significantly of RhoG activation at different time points following PDBu inhibited invadopodia formation in both CTRL and RhoG KD cells exposure. We measured RhoG activity by using GST–ELMO to (Fig. 2C,D), whereas inhibition of FAK with PF-573228 had no pulldown GTP-RhoG (van Buul et al., 2007). Our results show a effect (Fig. 2E,F). These findings suggest that the increase in Src rapid and transient decrease of RhoG activity in the first 5 min activity in RhoG KD cells may contribute to the increase observed of PDBu treatment followed by a peak of activation at 15 min in the number of cells with invadopodia, and that RhoG may (Fig. 3H,J). We obtained similar kinetics of RhoG activation after function downstream of Src to regulate invadopodia formation in treatment of cells with PMA (Fig. S4A,B). To test whether SGEF SUM159 cells. plays a role in the regulation of RhoG activation during PDBu stimulation, we measured RhoG activity in SGEF KD cells at SGEF functions upstream of RhoG to regulate invadopodia different times following PDBu addition. Interestingly, the peak of formation RhoG activation observed at 15 min was completely abrogated by Several GEFs have been described to stimulate nucleotide exchange SGEF KD (Fig. 3I,J). In contrast, the activity of RhoG in the on RhoG (Bellanger et al., 2000; Blangy et al., 2000; Bustelo et al., absence of PDBu was not significantly affected in the absence of 2007; D’Angelo et al., 2007; Damoulakis et al., 2014; Ellerbroek SGEF. Therefore, our results show that SGEF specifically mediates et al., 2004; Krishna Subbaiah et al., 2012; May et al., 2002; the PDBu-induced activation of RhoG that occurs at 15 min. Wennerberg et al., 2002). To identify the GEF that regulates RhoG during invadopodia formation, we performed a candidate-based Enhanced invadopodia formation in RhoG- and SGEF- shRNA screen. We stably silenced the best characterized RhoG- depleted cells is not sufficient for invasion specific GEFs, including SGEF (also known as ARHGEF26), In order to determine whether the invadopodia that formed ephexin 4 (also known as ARHGEF16), PLEKHG6 and Trio in following RhoG depletion retained their capacity to degrade the Journal of Cell Science

1066 RESEARCH ARTICLE Journal of Cell Science (2017) 130, 1064-1077 doi:10.1242/jcs.195552

Fig. 2. Src activity is necessary for increased invadopodia formation after RhoG KD. Western blot analysis shows increased levels of phospho-Src (Src p418) (A) and phospho-FAK (FAK p397) (B) in RhoG-depleted cells. Data are mean±s.e.m. of three independent experiments. (C–F) CTRL and RhoG KD cells were cultured in the presence of vehicle control, 5 μM PP2 (C,D) or 5 μM PF-573228 (E,F) for 30 min followed by a 30 min incubation with PDBu. The efficiency of the inhibitors was tested by immunoblotting for phospho-Src (Src p418) (C) or phospho-FAK (FAK p397) (E). Quantifications in D and F show the percentage of cells with invadopodia, as determined by cortactin and actin staining. Data are mean±s.e.m. of three independent experiments. *P<0.05; **P<0.01; ***P<0.001; ns, not significant.

ECM, we plated CTRL and RhoG KD cells on Oregon Green 488- invadopodia and increased degradation capacity in RhoG- and conjugated gelatin-coated coverslips and assessed invadopodia SGEF-deficient cells are not sufficient to enhance invasiveness. It is activity by quantifying the area of degradation (visible as dark on possible that the invasion defect observed results from the inability the green background) (Artym et al., 2009; Chen et al., 1985; Martin of cells to migrate in the absence of RhoG, since RhoG has been et al., 2012). Our results show that the number of cells that form previously shown to play a role during cell migration (Hiramoto- invadopodia on gelatin matrix is also increased when RhoG is Yamaki et al., 2010; Katoh et al., 2006). Supporting this, our results knocked down (CTRL=35%; RhoG KD=71%, Fig. 4A). RhoG- show that RhoG KD also impairs cell migration in SUM159 cells depleted cells also exhibited larger areas of matrix degradation (Fig. S2G). In summary, our results suggest that proper coordination compared to CTRL cells, which showed discrete areas of of matrix degradation and migration is required for efficient degradative puncta (Fig. 4B). Quantification showed a ∼9-fold invasion. increase in degradation area in RhoG KD cells relative to CTRL cells (Fig. 4C). These results demonstrate that invadopodia are still Rac1 stimulates invadopodia formation functional in the absence of RhoG. We obtained similar results, RhoG is a key upstream regulator of Rac1 in migrating cells when we silenced RhoG in MDA-MB-231 cells (Fig. S2E,F). We (Elfenbein et al., 2009; Hiramoto et al., 2006; Katoh et al., 2006; then assayed the ability of serum-starved CTRL, RhoG KD and Katoh and Negishi, 2003). However, RhoG has also been shown to RhoG KD cells expressing a Myc-tagged shRNA-resistant version function independently of Rac1 (Samson et al., 2010; Wennerberg of RhoG (Rescue) to invade through Matrigel-coated membranes et al., 2002). To determine whether the role of RhoG in invadopodia (Fig. 4D). Our results show that, even though there are more cells depends on Rac1, we first generated stable cell lines in which Rac1 with invadopodia in RhoG KD cells than in CTRL cells, their ability expression was silenced using lentivirally encoded shRNA. Since to invade through Matrigel was significantly impaired. The ability knockdown of Rac1 was not efficient for any of the four shRNAs to invade was partially restored by re-expressing Myc–RhoG tested (not shown), we generated Rac1-knockout cells (Rac1 KO) (Fig. 4D). As observed in RhoG KD cells, SGEF KD cells also using the double nicking RNA-guided Cas9 nucleases from the displayed a lower invasive capacity compared to CTRL cells microbial CRISPR/Cas system. After isolation of single cell

(Fig. 4E). These results suggest that enhanced formation of colonies, gene knockout efficiency was assayed by western Journal of Cell Science

1067 RESEARCH ARTICLE Journal of Cell Science (2017) 130, 1064-1077 doi:10.1242/jcs.195552

Fig. 3. See next page for legend. Journal of Cell Science

1068 RESEARCH ARTICLE Journal of Cell Science (2017) 130, 1064-1077 doi:10.1242/jcs.195552

Fig. 3. SGEF regulates invadopodia formation. (A) Cell lysates from contrast, in the absence of RhoG, the number of cells with SUM159 cells stably expressing non-targeting (CTRL) or SGEF-specific invadopodia did not decrease as much, and remained stable over shRNAs (shRNA#4 and shRNA #5) were analyzed by qRT-PCR for time at ∼70% (Fig. 6A). These results suggest that, in CTRL cells, expression of SGEF. (B) CTRL and SGEF KD SUM159 cells were treated with PDBu for 30 min and stained with anti-cortactin antibody (red), Alexa-Fluor- following a rapid burst of initial invadopodia assembly in response 488–phalloidin (green) and Hoechst 33342 (blue). Arrowheads indicate to PDBu, invadopodia start to disassemble and the percentage of representative invadopodia. (C) Quantification of cortactin- and actin- cells showing invadopodia stabilizes, probably reflecting an containing invadopodia in CTRL and SGEF KD cells expressed as the equilibrium between assembly and disassembly. In contrast, the percentage of cells with invadopodia. Data are mean±s.e.m. of four percentage of cells with invadopodia remains significantly higher in independent experiments. (D) Cell lysates from CTRL, RhoG KD and RhoG RhoG KD cells, which may suggest that invadopodia are more KD cells expressing Myc–SGEF (rescue Myc–SGEF) were immunoblotted with anti-RhoG and -Myc antibodies. Tubulin was used as a loading control. stable or longer lived in the absence of RhoG. (E) Quantification of cortactin- and actin-containing invadopodia in CTRL, To investigate the impact of RhoG and SGEF on invadopodia RhoG KD and rescue Myc–SGEF cells. Data are mean±s.e.m. of three lifetime, we analyzed the dynamics of invadopodia by time-lapse independent experiments in which at least 200 cells per experiment were confocal microscopy in SUM159 cells expressing mCherry–cortactin counted. (F) Quantification of cortactin- and actin-containing invadopodia in after PDBu treatment (Fig. 6B). In CTRL cells, invadopodia were SUM159 cells transiently transfected with either Myc-tagged wild-type SGEF short-lived and motile with an average lifetime of ∼10 min (Fig. 6B, (SGEF wt) or catalytically inactive SGEF (SGEF 446/621). Data are mean± C; Movie 1). However, when either RhoG or SGEF were silenced, s.e.m. of at least three independent experiments in which at least 200 cells per experiment were counted. (G) Representative images of cells transfected with the lifetime of invadopodia increased significantly to over 30 min, either Myc-tagged wild-type SGEF (SGEF wt) or catalytically inactive SGEF and they often persisted in a single location for time periods >1 h (SGEF 446/621). Cells were stained with anti-cortactin antibody (red), anti-Myc (Fig. 6B,C; Movies 4 and 5). When RhoG or SGEF were re- antibody (green), Alexa-Fluor-647–phalloidin (magenta) and Hoechst 33342 expressed in RhoG KD or SGEF KD cells, respectively, invadopodia (blue). Arrows indicate transfected cells and arrowheads indicate lifetime returned to control levels (Fig. 6C; Movies 6 and 7). These – representative invadopodia. (H J) CTRL (H) and SGEF KD (I) cells were results demonstrate the importance of RhoG and SGEF in the treated with PDBu for the indicated times. Active RhoG was precipitated from total lysates using GST–ELMO and immunoblotted with RhoG antibody. regulation of invadopodia dynamics, most likely during disassembly. (J) For quantification, active RhoG levels were normalized to total RhoG levels. Paxillin has previously been found to be a component of Data are mean±s.e.m. of at least three independent experiments. Scale bars: invadopodia in other systems, and its tyrosine phosphorylation has 10 µm. *P<0.05; **P<0.01; ***P<0.001. been shown to play a role in invadopodia disassembly (Badowski et al., 2008). To test whether RhoG and SGEF regulate paxillin blotting (Fig. 5A). Remarkably, invadopodia structures were tyrosine phosphorylation, we first examined the levels of completely absent in Rac1 KO cells (Fig. 5B). PDBu-treated Rac1 phosphorylated (phospho)-paxillin (PXN p118) and total paxillin KO cells showed a very consistent morphology, with actin and (PXN) in CTRL, RhoG KD and SGEF KD cells by western blotting cortactin concentrated at defined regions of the leading edge in using both total and phospho-specific paxillin antibodies. small lamellipodia-like structures (Fig. 5B). Quantitative analysis in Quantification of immunoblots showed that paxillin tyrosine- the three single cell colonies analyzed showed that Rac1 KO phosphorylation decreased ∼2.5- and ∼1.8-fold in RhoG KD and completely blocked the ability of cells to form invadopodia as SGEF KD cells respectively compared to CTRL cells (Fig. 7A). compared to CTRL cells (Fig. 5C; Movies 1 and 2). Re-expression Similar effects were also observed in MDA-MB-231 and MCF7 of Myc-tagged Rac1 in KO cells (rescue) restored the ability of the cells (Fig. S4C,D) with a ∼1.5- and ∼1.7-fold decrease, cells to form invadopodia to CTRL levels (Fig. 5D, Movie 3). To respectively. We also investigated the distribution of phospho- explore the role of Rac1 in more detail, we overexpressed Myc– paxillin (PXN p118, red) and total paxillin (PXN, green) at Rac1 in SUM159 cells (Rac1 OE) and tested the cells for their invadopodia by immunofluorescence in CTRL and RhoG KD ability to form invadopodia. As expected from the Rac1 KO results, SUM159 cells. Our results show that, even though total paxillin was cells overexpressing Rac1 formed more invadopodia than CTRL present at invadopodia in both CTRL and RhoG KD cells, phospho- cells (Fig. 5E). We also attempted to rescue RhoG KD cells with paxillin accumulated in invadopodia structures only in CTRL cells, Myc-tagged Rac1 (RhoG KD+Myc–Rac1). The rationale for this and was virtually absent in RhoG KD invadopodia (Fig. 7B). Line- experiment was that if Rac1 is being activated downstream of RhoG, scan analysis through invadopodial structures confirmed these an excess of Rac1 would be able to rescue RhoG deficiency. Fig. 5F results, revealing a significant decrease of phospho-paxillin shows that ectopic expression of Rac1 could not rescue the RhoG fluorescence intensity on RhoG KD invadopodia (PXN p118, red KD invadopodia phenotype. In contrast, RhoG KD Myc–Rac1 line) (Fig. 7C). In addition, the average ratio of phosphorylated to rescue cells formed even more invadopodia than RhoG KD cells total paxillin fluorescence intensity in invadopodia was two times suggesting an additive effect. Taken together, these results higher in CTRL cells than in RhoG KD cells (Fig. 7D). These demonstrated that Rac1 and RhoG function independently and results show that SGEF and RhoG regulate the levels of paxillin play opposing roles in the regulation of invadopodia formation in phosphorylation, which is critical for invadopodia disassembly. breast cancer cells. Both Src and FAK have been associated with paxillin phosphorylation (Bellis et al., 1995; Schaller and Parsons, 1995). RhoG is involved in invadopodia dynamics However, our results show that even though the phosphorylation of To further characterize the differences in invadopodia formation both kinases increased when RhoG is silenced (Fig. 2A,B), paxillin between CTRL and RhoG KD cells, we analyzed invadopodia phosphorylation levels are significantly decreased (Fig. 7A), which formation over time by fixing and staining SUM159 cells at suggests that the effect of RhoG on paxillin phosphorylation may be different times after PDBu addition. Surprisingly, at early time independent of Src and FAK. To determine whether Src and/or points, the number of cells with invadopodia was almost identical FAK are involved in RhoG-mediated regulation of paxillin between CTRL and RhoG KD cells, peaking at ∼70–80% (Fig. 6A). phosphorylation, we analyzed the levels of phospho-paxillin in In CTRL cells, the number of cells with invadopodia decreased CTRL or RhoG KD cells in the presence or absence of Src or FAK rapidly to ∼40% at 30 min, and remained stable even after 3 h. In inhibitors. Our results show that treating CTRL or RhoG KD with a Journal of Cell Science

1069 RESEARCH ARTICLE Journal of Cell Science (2017) 130, 1064-1077 doi:10.1242/jcs.195552

Fig. 4. Silencing RhoG increases matrix degradation but not invasion. (A) CTRL and RhoG KD cells were cultured on Oregon Green 488-conjugated gelatin and stained with anti-cortactin antibody (red). Invadopodia quantification in CTRL and RhoG KD cells is expressed as the percentage of cells with invadopodia. Data are mean±s.e.m. of three independent experiments. (B) Representative images showing matrix degradation in CTRL and RhoG KD cells. Scale bars: 20µm. (C) Area of matrix degraded per cell area. Data are mean±s.e.m. of three independent experiments. (D) Invasion assay in CTRL, RhoG KD and RhoG KD/Myc- RhoG rescue cells. Representative images showing cells that have invaded across a Matrigel-coated transwell membrane are presented. Results are representative of at least four individual experiments. (E) Invasion assay of CTRL and SGEF KD cells. Representative images are showing on the left. Results are representative of at least four individual experiments. *P<0.05; **P<0.01.

FAK inhibitor did not have a significant effect in phospho-paxillin activation during invadopodia disassembly. Silencing SGEF, but levels when compared to non-treated cells. In contrast, inhibition of not other RhoG-specific GEFs, phenocopied the results obtained in Src family kinases promoted a decrease in phospho-paxillin levels RhoG KD cells with a significant increase in the number of cells that in both CTRL and RhoG KD cells (Fig. 7E,F). These results form invadopodia. SGEF has been shown to direct actin suggests there are two pools of phospho-paxillin in the cells, one cytoskeleton remodeling mainly at the dorsal surface of the cells, that is regulated by RhoG independently of Src, and the other is where it promotes the formation of dorsal ruffles (Ellerbroek et al., dependent on Src and independent of RhoG. 2004; Patel and Galán, 2006; van Buul et al., 2007). These actin-rich dorsal structures share several of its molecular components with DISCUSSION invadopodia and podosomes, as well as with focal adhesions, Cells assemble invadopodia during cell invasion, which involves a suggesting that some of the signaling pathways controlling their dramatic rearrangement of the actin cytoskeleton, a process that is assembly may also be conserved (Buccione et al., 2004). In regulated by the integrated activity of several GTPases (Spuul et al., addition, recent studies have shown that SGEF may play a role 2014). While RhoG is known to promote migration and invasion during invasion in human papillomavirus-mediated cervical cancer, and has been recently linked to invadopodia, the regulation and role cancer and glioblastoma (Fortin Ensign et al., 2013; of RhoG during invadopodia formation are still unknown. The data Krishna Subbaiah et al., 2012; Wang et al., 2013). The molecular reported here describe a novel function for RhoG as a regulator of mechanisms by which SGEF contributes to these processes are invadopodia disassembly in human breast cancer cells. We showed unclear. Our studies suggest that SGEF activity is tightly regulated that when RhoG expression is silenced, invadopodia are more stable during the lifetime of invadopodia and that it is recruited to and live longer, whereas its overexpression has the opposite effect. invadopodia where it activates RhoG during invadopodia

We also identified SGEF as the exchange factor that regulates RhoG disassembly. Journal of Cell Science

1070 RESEARCH ARTICLE Journal of Cell Science (2017) 130, 1064-1077 doi:10.1242/jcs.195552

Fig. 5. Rac1 is necessary for invadopodia formation in SUM159 cells. (A) Cells lysates from CTRL and Rac1 KO SUM159 cells were analyzed by western blotting and probed for Rac1 and tubulin, as a loading control. (B) CTRL and Rac1 KO cells were treated with PDBu for 30 min and stained with anti-cortactin antibody (red), Alexa-Fluor-488–phalloidin (green) and Hoechst 33342 (blue). Arrowheads indicate representative invadopodia. Scale bars: 10 µm. (C) Quantification of cortactin- and actin-containing invadopodia in CTRL and Rac1 KO cell lines expressed as percentage of number of cells with invadopodia. Data are mean±s.e.m. of at least three independent experiments in which at least 200 cells per experiment were counted. (D) Cell lysates from CTRL, Rac1KO and Rac1 KO cells expressing Myc–Rac1 (rescue) were immunoblotted with anti-Rac1 and -Myc antibodies. Tubulin was used as a loading control (left panel). Quantification of cortactin- and actin-containing invadopodia in CTRL, Rac1 KO and rescue cells (right panel). Data are mean±s.e.m. of at least three independent experiments in which at least 200 cells per experiment were counted. (E) Cell lysates from CTRL and cells expressing Myc–Rac1 (Rac1 OE) were immunoblotted with anti-Rac1 and -Myc antibodies. Tubulin was used as a loading control (left panel). Quantification of cortactin- and actin-containing invadopodia in CTRL cells and cells expressing Myc-Rac1 (right panel). Data are mean±s.e.m. of at least three independent experiments in which at least 200 cells per experiment were counted. (F) Cell lysates from CTRL, RhoG KD and RhoG KD cells expressing Myc–Rac1 (RhoG KD+Myc-Rac1) were immunoblotted with anti-RhoG and -Myc antibodies. Tubulin was used as a loading control (left panel). Quantification of cortactin- and actin-containing invadopodia in CTRL, RhoG KD and RhoG KD +Myc-Rac1 cells (right panel). Data are mean±s.e.m. of at least three independent experiments in which at least 200 cells per experiment were counted.*P<0.05; **P<0.01; ***P<0.001.

RhoG can directly influence the activity of Rac1 by forming a with Rac1 being essential for invadopodia formation whereas RhoG is ternary complex with its effector protein ELMO and the Rac1 involved in their disassembly. Rac1 has been recently shown, together exchange factor (also known as DOCK1) (Brugnera et al., with Trio and Pak1, to function in invadopodia disassembly in rat 2002; Katoh and Negishi, 2003). However, RhoG can also signal mammary adenocarcinoma cells (Moshfegh et al., 2014). In that independently of Rac1 (Gauthier-Rouviere et al., 1998; Samson et al., study, silencing the expression of Rac1 or Trio was also shown to 2010; Wennerberg et al., 2002). Our results demonstrate that Rac1 and promote an increase in invadopodia lifetime, without affecting

RhoG play opposing roles in the regulation of invadopodia dynamics, invadopodia numbers (Moshfegh et al., 2014). Even though our Journal of Cell Science

1071 RESEARCH ARTICLE Journal of Cell Science (2017) 130, 1064-1077 doi:10.1242/jcs.195552

Fig. 6. Silencing RhoG or SGEF increases invadopodia lifetime. (A) CTRL or RhoG KD SUM159 cells were treated with PDBu for the indicated times and stained for cortactin and actin as invadopodia markers. Invadopodia were quantified and expressed as the percentage of cells with invadopodia. Data are mean± s.e.m. of three independent experiments in which at least 200 cells were analyzed per condition. (B) Representative time series of invadopodia formation in CTRL and RhoG KD cells. SUM159 cells expressing mCherry–cortactin were imaged for 120 min at 15 s intervals following the addition of PDBu. White arrows point to an invadopodia cluster forming in the nuclear region. Scale bars: 20 µm. (C) Invadopodia lifetime increases when RhoG and SGEF expression are silenced. Re-expression of RhoG and SGEF in RhoG KD and SGEF KD cells respectively restores the lifetime to the levels of CTRL cells. Results are representative of three independent experiments in which at least five cells were analyzed per condition. The box represents the 25–75th percentiles, and the median is indicated. The whiskers show the 5th to 95th percentiles. *P<0.05; **P<0.01. results contradict those of Moshfegh and colleagues (2014), they are PMA treatment followed by a peak of activation at 15 min in agreement with several reports which show Rac1 is required for (invadopodia formation peaks at 10 min after PDBu treatment). invadopodia and podosome formation in different cell lines Phorbol esters stimulate the formation of podosomes and (Furmaniak-Kazmierczak et al., 2007; Harper et al., 2010; Lin et al., invadopodia by activating PKCα, which lies upstream of Src 2014; Nascimento et al., 2011; Pignatelli et al., 2012; Wheeler et al., (Gatesman et al., 2004; Hai et al., 2002; Tatin et al., 2006) but the 2006). Moreover, our results show that silencing SGEF or RhoG mechanisms that connect them with SGEF and RhoG regulation are affects both invadopodia numbers and lifetime, whereas silencing not known and are the focus of our future studies. Trio has no significant effect on the number of cells that form Efficient turnover of invadopodia is critical for effective cell invadopodia. Our results also differ with recent studies that analyzed invasion (Chan et al., 2009). It has been reported that RhoG and the role for RhoG in invadopodia. In glioblastoma cells, depletion of SGEF promote cell migration and invasion in several cell lines RhoG inhibits invadopodia formation (Kwiatkowska et al., 2012), including glioblastoma cells, MDA-MB-231 breast cancer cells, PC- whereas in rat breast cancer cells it appears to have no significant effect 3 cells and HeLa cells (Chatterjee et al., 2011; Fortin (Moshfegh et al., 2014). The differences between these studies and the Ensign et al., 2013; Hiramoto-Yamaki et al., 2010; Katoh et al., results reported here may not be necessarily contradictory. In the 2006; Krishna Subbaiah et al., 2012; Kwiatkowska et al., 2012). studies by Kwiatkowska et al., invadopodia formation was determined Here, we also show that RhoG- and SGEF-deficient SUM159 cells by measuring matrix degradation area (Kwiatkowska et al., 2012), show impaired invasion, despite showing enhanced invadopodia whereas in the studies by Moshfegh et al. invadopodia were followed formation and ECM degradation. These results are consistent with directly in live rat breast cancer cells following EGF stimulation the notion that the ability of cancer cells to form invadopodia alone is (Moshfegh et al., 2014). Since matrix degradation does not not sufficient for invasion to occur (Chan et al., 2009). This necessarily correlate with invadopodia number, it is difficult to uncoupling between invadopodia formation and invasion has been compare these studies. Alternatively, these distinct effects might described before, and may originate from the fact that several of the reflect differences specific to the cell types used or the experimental proteins involved in invadopodia formation also play a role during conditions, as already reported for other GTPases (Spuul et al., 2014). adhesion and migration. Silencing the expression of other proteins, Little is known regarding the extracellular signals that regulate including FAK, laminin-332 and ezrin, has also been shown to RhoG activity. In this study, we show that there is a rapid and induce an increase in invadopodia while simultaneously decreasing transient decrease of RhoG activity in the first 5 min after PDBu and their invasive capacity (Chan et al., 2009; Hoskin et al., 2015; Liu Journal of Cell Science

1072 RESEARCH ARTICLE Journal of Cell Science (2017) 130, 1064-1077 doi:10.1242/jcs.195552

Fig. 7. RhoG knockdown decreases paxillin phosphorylation at invadopodia. (A) Cell lysates from CTRL, RhoG KD and SGEF KD SUM159 cells were analyzed by western blotting and probed for phospho-paxillin (PXN p118), total paxillin (PXN) and tubulin as a loading control (left panel). The quantification represents the average of at least three independent experiments. Data are mean±s.e.m. (right panel). (B) CTRL and RhoG KD SUM159 cells were treated with PDBu for 30 min, fixed and stained with Alexa-Fluor-647–phalloidin (magenta), and for PXN (green) and PXN p118 (red). Scale bars: 10 µm. (C) Graphs indicate fluorescent intensity in arbitrary units (A.U.) of PXN p118 (red) with respect to PXN (green) and F-actin (blue) over the indicated line scan in (B). (D) Total fluorescence intensity for PXN and PXN p118 in individual invadopodia from CTRL and RhoG KD cells were measurements of at least 50 invadopodia per condition. Data are mean±s.e.m. (E) CTRL and RhoG KD cells were cultured in the presence of vehicle control, 5 μM PP2 or 5 μM PF-573228 for 30 min. Cell lysates were analyzed by western blotting and probed for total paxillin (PXN) and phosho-paxillin (PXN p118). (F) Quantification of PXN p118 to PXN ratio in cells treated with vehicle, PP2 and PF-573228. Results represent the mean±s.e.m. of at least three independent experiments. *P<0.05, **P<0.01. et al., 2010). It is possible that the invasion defect observed is related and invadopodia disassembly independently of Rac1 (Hiramoto to impaired migration. Migration is also inhibited in RhoG KD cells, et al., 2006; Katoh et al., 2006). as has been shown previously by others (Hiramoto-Yamaki et al., Despite the significant advances in the characterization of the very 2010; Katoh et al., 2006). Our results and those of others suggest that early stages of invadopodia assembly and the role of adhesion

RhoG may be working in parallel to regulate migration through Rac1 proteins in promoting invadopodia maturation (Beaty and Condeelis, Journal of Cell Science

1073 RESEARCH ARTICLE Journal of Cell Science (2017) 130, 1064-1077 doi:10.1242/jcs.195552

whereas Rac1 activity is absolutely essential. We propose that a yet to-be-identified RhoG-specific GAP downregulates RhoG during invadopodia assembly. During invadopodia disassembly, SGEF is targeted to invadopodia, where it activates RhoG. Active RhoG then promotes the phosphorylation of the adaptor protein paxillin, which stimulates the disassembly of invadopodia. Taken together, our observations suggest that RhoG promotes invadopodia turnover as the cell protrudes and prepares for tissue invasion. Our findings provide novel insights into the mechanisms of RhoG signaling in breast cancer invasion.

MATERIALS AND METHODS Cell lines Three breast cancer cell lines were used: MCF7, MDA-MB-231 and SUM159. The cell lines were a gift from Carol Otey (UNC-Chapel Hill, NC). MCF7 and MDA-MB-231 cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) (GIBCO, Grand Island, NY) containing 10% fetal bovine serum (FBS) and antibiotics (penicillin-streptomycin). SUM159 cells were cultured in Ham’s F12 with 5% calf serum, 5 μg/ml , 1 μg/ml hydrocortisone and antibiotics. All cell lines were grown at 37°C and 5% CO2. All experiments were conducted with early passage cells that were passaged no more than 15 times. was tested regularly by staining with Hoechst 33342 (83218, AnaSpec Inc., San Jose, CA).

Reagents Fig. 8. Model of RhoG function in invadopodia. Phorbol esters, including PDBu and PMA, stimulate PKC, which induces the formation of invadopodia in Antibodies against the following proteins were used: RhoG (sc-26484), Trio a Src-dependent fashion. Assembly of invadopodia requires Rac1 activity. In (sc-6060), cortactin (sc-11408) and Myc (9E10, sc-131) (Santa Cruz, Santa contrast, RhoG activity needs to be downregulated for invadopodia to form. Cruz, CA); tubulin (T9028, Sigma, St. Louis, MO); pY118-paxillin (2541), During disassembly, SGEF is recruited to invadopodia, where it activates total FAK (13009), pY397 FAK (8556), total Src (2108) and pY418 Src RhoG, which promotes the phosphorylation of paxillin. (2101) (, Danvers, MA); total paxillin (610051) and Rac1 (610650; BD Biosciences, San Jose, CA). The antibody dilutions used are 2014), our understanding of the mechanisms involved in listed in Table S1. Secondary antibodies were: Alexa Fluor 488 and Alexa Fluor 594-conjugated anti-mouse-IgG and anti-rabbit-IgG secondary invadopodia disassembly is limited. A report by Badowski and antibodies (A11008, A11001, A11005 and R37117) and Alexa Fluor-488 colleagues describes a role for paxillin in invadopodia disassembly (A12379) and Alexa Fluor-647 (A22287) conjugated to phalloidin (Life (Badowski et al., 2008). Paxillin tyrosine phosphorylation (Y31 and Technologies, Carlsbad, CA), and horseradish peroxidase (HRP)- Y118) promotes ERK protein activation, which activates calpain and conjugated anti-mouse-IgG, anti-rabbit-IgG and anti-goat-IgG secondary stimulates disassembly at the inner rim of the invadopodia ring in antibodies (715-035-151, 711-035-152 and 705-035-147; Jackson Rous sarcoma virus (RSV)-transformed cells (Badowski et al., Immunoresearch, West Grove, PA). PP2 (529573, Calbiochem, San 2008). These results are in agreement with a large body of literature Diego, CA). Phorbol-12,13-dibutyrate (PDBu) (P1269), Phorbol 12- describing the role of paxillin phosphorylation in the regulation of myristate 13-acetate (PMA) (P8139) and PF-573228 (PZ0117) were from focal adhesion disassembly and cell migration (Brown and Turner, Sigma, St Louis, MO. 2004; Nakamura et al., 2000; Petit et al., 2000; Vindis et al., 2004; Webb et al., 2004; Zaidel-Bar et al., 2007). Supporting these Transfections, immunofluorescence and treatments observations, we found a decrease in paxillin phosphorylation and Transfection of SUM159 cells was performed using Lipofectamine 2000 an increase in invadopodia lifetime in both RhoG and SGEF KD cells, (Life Technologies, Carlsbad, CA). For immunofluorescence, MCF7, MDA-MB-231 and SUM159 cells grown on coverslips were fixed in suggesting a role in invadopodia disassembly. Surprisingly, even 3.7% paraformaldehyde, permeabilized with 0.1% Triton X-100 in PBS and though paxillin phosphorylation at Y31 and Y118 is mostly associated then incubated with primary antibody for 1 h at room temperature. Primary with FAK and Src kinases (Brown and Turner, 2004), we found that antibodies were detected with Alexa Fluor 488- and Alexa Fluor 568- Src and FAK activities are not correlated with RhoG-mediated conjugated anti-mouse-IgG or anti-rabbit-IgG antibodies. Images were regulation of paxillin phosphorylation. Instead, we detected two pools acquired on an Olympus IX81 inverted microscope using a PlanApo N of phospho-paxillin, one that is regulated by RhoG independently of 60×1.42 NA oil objective lens and a XM10 camera (Olympus, Tokyo, Src, and the other is dependent on Src and independent of RhoG. FAK, Japan). Image processing and quantitative analysis was performed using on the other hand, had no effect on RhoG-mediated phosphorylation ImageJ. Invadopodia formation was induced by the addition of 1 μM PDBu μ of paxillin. Taken together, our results suggest that RhoG promotes or 7.5 M PMA. To inhibit Src, cells were treated with PP2 (5 µM) or paxillin phosphorylation through a yet to be identified kinase. DMSO diluents for 30 min before treatment with PDBu. For FAK inhibition, we incubated cells with 5 μM PF-573228 for 30 min before Alternatively, RhoG may be mediating the effect on paxillin by PDBU treatment. Live imaging was performed with a Leica SP8 confocal inhibiting a phosphatase, such as SHP-2 or PTP-PEST (also known as microscope using a PlApo CS2 N 63×1.4 NA objective (Leica, Wetzlar, PTPN11 and PTPN12, respectively), which have been shown to Germany), and equipped with an environmental chamber that controls modulate paxillin phosphorylation (Cote et al., 1999; Jamieson et al., temperature, CO2 and humidity (Tokai Hit, Fujinomiya, Japan). 2005; Mañes et al., 1999; Shen et al., 2000, 1998). In our model, RhoG and Rac1 have independent and opposite Cell lysis and immunoblotting roles in the regulation of invadopodia dynamics (Fig. 8). RhoG Cells cultured on 100 mm tissue culture dishes were rinsed with PBS and activity needs to be downregulated for invadopodia to assemble, then scraped into a lysis buffer containing 50 mM Tris-HCl pH 7.4, 10 mM Journal of Cell Science

1074 RESEARCH ARTICLE Journal of Cell Science (2017) 130, 1064-1077 doi:10.1242/jcs.195552

MgCl2, 150 mM NaCl, 1% Triton X-100 and EZBlock protease inhibitor CRISPR/Cas9-mediated KO cocktail (BioVision, Mipitas, CA). The supernatant was collected after The Rac1 gene was knocked out using CRIPS/Cas9 double nickase plasmids centrifugation at 16,800 g for 10 min. For immunoblotting, lysates were boiled (Santa Cruz Biotechnology, Santa Cruz, CA). Briefly, cells were transfected in 2× Laemmli buffer, and 20 μg of protein were resolved by SDS-PAGE. The with the plasmid mixture (2 gRNA plasmids; strand A, 5′-AGACACGAT- proteins were transferred onto PVDF and immunoblotted with the indicated CGAGAAACTGA-3′; strand B, 5′-TTTAGTTCCCACTAGGATGA-3′), antibodies. Immunocomplexes were visualized using the Immobilon Western and selected with puromycin 24 h after transfection. After selection, single Millipore Chemiluminescence HRP substrate (Millipore, Billerica, MA). cell colonies were isolated by serial dilution. The efficiency of the knockout was confirmed by western blotting. RhoG activity assay Active RhoG pulldown experiments were performed as described Lentiviral constructs and transduction previously (van Buul et al., 2007). Briefly, SUM159 cells were lysed in pLKO lentiviral non-targeting shRNA control was from Sigma (SHC016- 1EA). pLKO.1 shRNAs for human RhoG (#1 TRCN0000048018, #4 50 mM Tris-HCl pH 7.4, 10 mM MgCl2, 150 mM NaCl, 1% Triton X-100, and EZBlock protease inhibitor cocktail. After clearing the lysates by TRCN0000048021), SGEF (#4 TRCN0000048291, #5 TRCN0000048292), centrifugation at 14,000 g for 5 min, the protein concentrations of the ephexin 4 (#1 TRCN0000047503, #3 TRCN0000047507), PLEKHG6 (#1 supernatants were determined, and equal amounts of total protein were TRCN0000128030, #4 TRCN0000148892) and Trio (#1 TRCN0000000871, incubated with 50 μg of glutathione transferase (GST)–ELMO (GST fusion #5 TRCN0000010561) were from Open Biosystems (Huntsville, AL). protein containing the full-length RhoG effector ELMO) bound to Lentiviruses were prepared at the Lenti-shRNA Core Facility (UNC-Chapel glutathione–Sepharose beads (GE Healthcare, Pittsburgh, PA), and rotated Hill, NC). Cells were infected with lentivirus particles overnight. The for 30 min at 4°C. Subsequently the beads were washed four times in lysis following day, the infection medium was removed and replaced with complete buffer. Pull-downs and lysates were then immunoblotted for RhoG. medium containing puromycin (2.5 µg/ml) to select for shRNA-expressing cells. Total cell lysates were subjected to western blot analysis for protein Gelatin degradation assay expression as described above. For some shRNAs, single cell colonies were Oregon Green 488-conjugated gelatin-coated coverslips were prepared as isolated by serial dilution. described previously (Martin et al., 2012). Coverslips were coated with 50 μg/ml poly-D-lysine for 15 min, washed with PBS and cross-linked with qRT-PCR 0.5% glutaraldehyde for 15 min. Coverslips were then inverted on a 60 μl Total RNA was purified from SGEF KD, PLEKHG6 KD and Ephexin4 KD drop of 1 mg/ml Oregon Green 488-conjugated gelatin (Molecular Probes, SUM159 cells using Trizol (Life Technologies, CA) and was treated with ThermoFisher) for 20 min. After washing with PBS, coverslips were DNAse I (NEB, Ipswich, MA). Reverse was carried out using quenched with 5 mg/ml sodium borohydride for 5 min followed by washes the iScript cDNA Synthesis kit (BioRad, Hercules, CA) on 1 µg of total with PBS. Finally, they were transferred into complete growth medium for RNA. qRT-PCR was performed with equal amounts of cDNA using the Taq 1 h before use. Cells were seeded and cultured on cross-linked gelatin for PCR Master Mix kit (Qiagen, Valencia, CA); primer sequences will be 16 h and then fixed for immunofluorescence studies. For each experimental made available upon request. condition, 25 images were taken in a random fashion. To quantify the gelatin degradation activity of invadopodia, we calculated the degradation area Statistical analysis observed in images using ImageJ software and normalized the Values calculated from at least three independent experiments were ’ t measurements to the total cell area in each image. Results are expressed compared by a Student s -test using GraphPad Prism (La Jolla, CA), and P as the percentage of the cell area that was degraded. At least 100 cells per <0.05 was considered statistically significant. Error bars represent the s.e.m. experiment were analyzed. Acknowledgements We would like to thank Carol Otey (UNC-Chapel Hill, NC) for sharing the cell lines Migration and invasion assays utilized in these studies and James Bear (UNC-Chapel Hill, NC) for the mCherry– Migration assays were carried out using 24-well non-coated Transwell plates cortactin cDNA. (Corning, Lowel, MA) and invasion was analyzed using BD BioCoat growth-factor-reduced Matrigel Invasion Chambers (BD Biosciences, Competing interests Bedford, MA). After 2 h of serum starvation, cells (1.5×104) were added The authors declare no competing or financial interests. to the upper chamber. The bottom chamber was filled with medium Author contributions containing 10% FBS. Cells were allowed to migrate or invade for 16 h. Cells Conceptualization and Methdology, S.M.G. and R.G.-M.; Formal Analysis, R.G.-M, at the upper side of the membrane were removed using a Q-tip. Cells on the S.M.G., A.Z. and K.S.; Investigation, R.G.-M, S.M.G., A.Z., S.A. and K.S. Writing – bottom surface were fixed and stained using Diff-Quick (IHC World, LLC, Original Draft, Writing – Review and Editing and Visualization, S.M.G. and R.G.-M. Woodstock, MD). Cells were counted from at least four individual Supervision, S.M.G. and R.G.-M. Project Administration, R.G-M. Funding experiments performed in triplicates. Acquisition, S.M.G. and R.G.-M.

DNA constructs Funding R.G.-M. was supported by the National Institutes of Health (1R21CA194776-01A1, Generation of eukaryotic expression vectors pCMV-myc-RhoG-wt, wild 1R15CA199101-01A1 and 1R03CA197227-01A1), Ohio Cancer Research, and type (WT) and mutant (E446A-N621A) Myc-tagged SGEF has been the deArce-Koch Memorial Endowment fund. S.M.G. was supported by the National previously described (Ellerbroek et al., 2004). The RhoG Q61L was Institutes of Health (1R03CA161136-01). Deposited in PMC for release after generated by site-directed mutagenesis using the Quickchange Site-Directed 12 months. Mutagenesis kit (Stratagene, Santa Clara, CA). mCherry–Cortactin cDNA was a gift from James Bear, UNC-Chapel Hill, NC. Data availability Supplementary movies are also available from the Dryad Digital Repository (http://dx.doi.org/10.5061/dryad.q1j50). Overexpression using adenoviral system shRNA-resistant versions of human Myc–RhoG, Myc–Rac1, Myc–SGEF – Supplementary information and mCherry Cortactin were subcloned into pAd/CMV/V5-DEST using Supplementary information available online at Gateway recombination technology (Life Technologies Carlsbad, CA). http://jcs.biologists.org/lookup/doi/10.1242/jcs.195552.supplemental shRNA resistance in RhoG and SGEF was achieved by introducing silence mutations in at least three bases within the corresponding shRNA targeting References region. Virus particles were produced using the Virapower Adenoviral Artym, V. V., Yamada, K. M. and Mueller, S. C. (2009). ECM degradation assays

Expression System (Life Technologies, Carlsbad, CA). for analyzing local cell invasion. Methods Mol. Biol. 522, 211-219. Journal of Cell Science

1075 RESEARCH ARTICLE Journal of Cell Science (2017) 130, 1064-1077 doi:10.1242/jcs.195552

Ayala, I., Giacchetti, G., Caldieri, G., Attanasio, F., Mariggio, S., Tete, S., Gimona, M., Buccione, R., Courtneidge, S. A. and Linder, S. (2008). Assembly Polishchuk, R., Castronovo, V. and Buccione, R. (2009). Faciogenital dysplasia and biological role of podosomes and invadopodia. Curr. Opin. Cell Biol. 20, protein Fgd1 regulates invadopodia biogenesis and extracellular matrix degradation 235-241. and is up-regulated in prostate and breast cancer. Cancer Res. 69, 747-752. Goicoechea, S. M., Bednarski, B., Garcıa-Mata,́ R., Prentice-Dunn, H., Kim, H. J. Badowski, C., Pawlak, G., Grichine, A., Chabadel, A., Oddou, C., Jurdic, P., and Otey, C. A. (2009). Palladin contributes to invasive motility in human breast Pfaff, M., Albiges-Rizo, C. and Block, M. R. (2008). Paxillin phosphorylation cancer cells. Oncogene 28, 587-598. controls invadopodia/podosomes spatiotemporal organization. Mol. Biol. Cell 19, Hai, C.-M., Hahne, P., Harrington, E. O. and Gimona, M. (2002). Conventional 633-645. protein kinase C mediates phorbol-dibutyrate-induced cytoskeletal remodeling in Beaty, B. T. and Condeelis, J. (2014). Digging a little deeper: the stages of a7r5 smooth muscle cells. Exp. Cell Res. 280, 64-74. invadopodium formation and maturation. Eur. J. Cell Biol. 93, 438-444. Harper, K., Arsenault, D., Boulay-Jean, S., Lauzier, A., Lucien, F. and Dubois, Bellanger, J.-M., Astier, C., Sardet, C., Ohta, Y., Stossel, T. P. and Debant, A. C. M. (2010). Autotaxin promotes cancer invasion via the lysophosphatidic acid (2000). The Rac1- and RhoG-specific GEF domain of Trio targets filamin to receptor 4: participation of the cyclic AMP/EPAC/Rac1 signaling pathway in remodel cytoskeletal actin. Nat. Cell Biol. 2, 888-892. invadopodia formation. Cancer Res. 70, 4634-4643. Bellis, S. L., Miller, J. T. and Turner, C. E. (1995). Characterization of tyrosine Hiramoto, K., Negishi, M. and Katoh, H. (2006). is regulated by RhoG and phosphorylation of paxillin in vitro by focal adhesion kinase. J. Biol. Chem. 270, promotes Rac-dependent cell migration. Exp. Cell Res. 312, 4205-4216. 17437-17441. Hiramoto-Yamaki, N., Takeuchi, S., Ueda, S., Harada, K., Fujimoto, S., Negishi, Blangy, A., Vignal, E., Schmidt, S., Debant, A., Gauthier-Rouviere, C. and Fort, M. and Katoh, H. (2010). Ephexin4 and EphA2 mediate cell migration through a P. (2000). TrioGEF1 controls Rac- and Cdc42-dependent cell structures through RhoG-dependent mechanism. J. Cell Biol. 190, 461-477. the direct activation of rhoG. J. Cell Sci. 113, 729-739. Hodge, R. G. and Ridley, A. J. (2016). Regulating Rho GTPases and their Brown, M. C. and Turner, C. E. (2004). Paxillin: adapting to change. Physiol. Rev. regulators. Nat. Rev. Mol. Cell Biol. 17, 496-510. 84, 1315-1339. Hoshino, D., Jourquin, J., Emmons, S. W., Miller, T., Goldgof, M., Costello, K., Brugnera, E., Haney, L., Grimsley, C., Lu, M., Walk, S. F., Tosello-Trampont, Tyson, D. R., Brown, B., Lu, Y., Prasad, N. K. et al. (2012). Network analysis of A. C., Macara, I. G., Madhani, H., Fink, G. R. and Ravichandran, K. S. (2002). the focal adhesion to invadopodia transition identifies a PI3K-PKCalpha invasive Unconventional Rac-GEF activity is mediated through the Dock180-ELMO signaling axis. Sci. Signal. 5, ra66. complex. Nat. Cell Biol. 4, 574-582. Hoshino, D., Branch, K. M. and Weaver, A. M. (2013). Signaling inputs to Buccione, R., Orth, J. D. and McNiven, M. A. (2004). Foot and mouth: podosomes, invadopodia and podosomes. J. Cell Sci. 126, 2979-2989. invadopodia and circular dorsal ruffles. Nat. Rev. Mol. Cell Biol. 5, 647-657. Hoskin, V., Szeto, A., Ghaffari, A., Greer, P. A., Cote, G. P. and Elliott, B. E. Bustelo, X. R., Sauzeau, V. and Berenjeno, I. M. (2007). GTP-binding proteins of the (2015). Ezrin regulates focal adhesion and invadopodia dynamics by altering Rho/Rac family: regulation, effectors and functions in vivo. Bioessays 29, 356-370. calpain activity to promote breast cancer cell invasion. Mol. Biol. Cell 26, Chan, K. T., Cortesio, C. L. and Huttenlocher, A. (2009). FAK alters invadopodia 3464-3479. and focal adhesion composition and dynamics to regulate breast cancer invasion. Jackson, B. C., Ivanova, I. A. and Dagnino, L. (2015). An ELMO2-RhoG-ILK J. Cell Biol. 185, 357-370. network modulates microtubule dynamics. Mol. Biol. Cell 26, 2712-2725. Chatterjee, M., Sequeira, L., Jenkins-Kabaila, M., Dubyk, C. W., Pathak, S. and Jamieson, J. S., Tumbarello, D. A., Halle, M., Brown, M. C., Tremblay, M. L. and van Golen, K. L. (2011). Individual GTPases mediate aspects of prostate Turner, C. E. (2005). Paxillin is essential for PTP-PEST-dependent regulation of cancer cell and bone marrow endothelial cell interactions. J. Signal. Transduct. cell spreading and motility: a role for paxillin kinase linker. J. Cell Sci. 118, 2011, 541851. 5835-5847. Chen, W.-T., Chen, J.-M., Parsons, S. J. and Parsons, J. T. (1985). Local Katoh, H. and Negishi, M. (2003). RhoG activates Rac1 by direct interaction with degradation of fibronectin at sites of expression of the transforming gene product the Dock180-binding protein Elmo. Nature 424, 461-464. pp60src. Nature 316, 156-158. Katoh, H., Yasui, H., Yamaguchi, Y., Aoki, J., Fujita, H., Mori, K. and Negishi, M. Cote, J.-F., Turner, C. E. and Tremblay, M. L. (1999). Intact LIM 3 and LIM 4 (2000). Small GTPase RhoG is a key regulator for neurite outgrowth in PC12 cells. domains of paxillin are required for the association to a novel polyproline region Mol. Cell. Biol. 20, 7378-7387. (Pro 2) of protein-tyrosine phosphatase-PEST. J. Biol. Chem. 274, 20550-20560. Katoh, H., Hiramoto, K. and Negishi, M. (2006). Activation of Rac1 by RhoG Damoulakis, G., Gambardella, L., Rossman, K. L., Lawson, C. D., Anderson, regulates cell migration. J. Cell Sci. 119, 56-65. K. E., Fukui, Y., Welch, H. C., Der, C. J., Stephens, L. R. and Hawkins, P. T. Krishna Subbaiah, V., Massimi, P., Boon, S. S., Myers, M. P., Sharek, L., Garcia- (2014). P-Rex1 directly activates RhoG to regulate GPCR-driven Rac signalling Mata, R. and Banks, L. (2012). The invasive capacity of HPV transformed cells and actin polarity in . J. Cell Sci. 127, 2589-2600. requires the hDlg-dependent enhancement of SGEF/RhoG activity. PLoS Pathog. D’Angelo, R., Aresta, S., Blangy, A., Del Maestro, L., Louvard, D. and Arpin, M. 8, e1002543. (2007). Interaction of Ezrin with the novel guanine nucleotide exchange factor Kwiatkowska, A., Didier, S., Fortin, S., Chuang, Y., White, T., Berens, M. E., PLEKHG6 promotes RhoG-dependent apical cytoskeleton rearrangements in Rushing, E., Eschbacher, J., Tran, N. L., Chan, A. et al. (2012). The small epithelial cells. Mol. Biol. Cell 18, 4780-4793. GTPase RhoG mediates glioblastoma cell invasion. Mol. Cancer 11, 65. deBakker, C. D., Haney, L. B., Kinchen, J. M., Grimsley, C., Lu, M., Klingele, D., Lin, C.-W., Sun, M.-S., Liao, M.-Y., Chung, C.-H., Chi, Y.-H., Chiou, L.-T., Yu, J., Hsu, P.-K., Chou, B.-K., Cheng, L.-C., Blangy, A. et al. (2004). Phagocytosis of Lou, K.-L. and Wu, H.-C. (2014). Podocalyxin-like 1 promotes invadopodia apoptotic cells is regulated by a UNC-73/TRIO-MIG-2/RhoG signaling module formation and metastasis through activation of Rac1/Cdc42/cortactin signaling in and armadillo repeats of CED-12/ELMO. Curr. Biol. 14, 2208-2216. breast cancer cells. Carcinogenesis 35, 2425-2435. Di Martino, J., Paysan, L., Gest, C., Lagree, V., Juin, A., Saltel, F. and Moreau, V. Linder, S., Wiesner, C. and Himmel, M. (2011). Degrading devices: invadosomes (2014). Cdc42 and Tks5: a minimal and universal molecular signature for in proteolytic cell invasion. Annu. Rev. Cell Dev. Biol. 27, 185-211. functional invadosomes. Cell Adh. Migr. 8, 280-292. Liu, S., Yamashita, H., Weidow, B., Weaver, A. M. and Quaranta, V. (2010). Elfenbein, A., Rhodes, J. M., Meller, J., Schwartz, M. A., Matsuda, M. and Laminin-332-beta1 integrin interactions negatively regulate invadopodia. J. Cell Simons, M. (2009). Suppression of RhoG activity is mediated by a syndecan 4- Physiol. 223, 134-142. synectin-RhoGDI1 complex and is reversed by PKCalpha in a Rac1 activation Mañes, S., Mira, E., Gómez-Mouton, C., Zhao, Z. J., Lacalle, R. A. and Martınez-́ pathway. J. Cell Biol. 186, 75-83. A, C. (1999). Concerted activity of tyrosine phosphatase SHP-2 and focal Ellerbroek, S. M., Wennerberg, K., Arthur, W. T., Dunty, J. M., Bowman, D. R., adhesion kinase in regulation of cell motility. Mol. Cell. Biol. 19, 3125-3135. DeMali, K. A., Der, C. and Burridge, K. (2004). SGEF, a RhoG guanine Martin, K. H., Hayes, K. E., Walk, E. L., Ammer, A. G., Markwell, S. M. and Weed, nucleotide exchange factor that stimulates macropinocytosis. Mol. Biol. Cell 15, S. A. (2012). Quantitative measurement of invadopodia-mediated extracellular 3309-3319. matrix proteolysis in single and multicellular contexts. J. Vis. Exp., e4119. Fortin Ensign, S. P., Mathews, I. T., Eschbacher, J. M., Loftus, J. C., Symons, May, V., Schiller, M. R., Eipper, B. A. and Mains, R. E. (2002). Kalirin Dbl- M. H. and Tran, N. L. (2013). The Src homology 3 domain-containing guanine homology guanine nucleotide exchange factor 1 domain initiates new axon nucleotide exchange factor is overexpressed in high-grade gliomas and promotes outgrowths via RhoG-mediated mechanisms. J. Neurosci. 22, 6980-6990. tumor necrosis factor-like weak inducer of -fibroblast growth factor- Moreau, V., Tatin, F., Varon, C. and Genot, E. (2003). Actin can reorganize into inducible 14-induced cell migration and invasion via tumor necrosis factor podosomes in aortic endothelial cells, a process controlled by Cdc42 and RhoA. receptor-associated factor 2. J. Biol. Chem. 288, 21887-21897. Mol. Cell. Biol. 23, 6809-6822. Furmaniak-Kazmierczak, E., Crawley, S. W., Carter, R. L., Maurice, D. H. and Moreau, V., Tatin, F., Varon, C., Anies, G., Savona-Baron, C. and Genot, E. Cote, G. P. (2007). Formation of extracellular matrix-digesting invadopodia by (2006). Cdc42-driven podosome formation in endothelial cells. Eur. J. Cell Biol. primary aortic smooth muscle cells. Circ. Res. 100, 1328-1336. 85, 319-325. Gatesman, A., Walker, V. G., Baisden, J. M., Weed, S. A. and Flynn, D. C. (2004). Moshfegh, Y., Bravo-Cordero, J. J., Miskolci, V., Condeelis, J. and Hodgson, L. Protein kinase C{alpha} activates c-Src and induces podosome formation via (2014). A Trio-Rac1-Pak1 signalling axis drives invadopodia disassembly. Nat. AFAP-110. Mol. Cell. Biol. 24, 7578-7597. Cell Biol. 16, 574-586. Gauthier-Rouviere, C., Vignal, E., Meriane, M., Roux, P., Montcourier, P. and Murphy, D. A. and Courtneidge, S. A. (2011). The ‘ins’ and ‘outs’ of podosomes Fort, P. (1998). RhoG GTPase controls a pathway that independently activates and invadopodia: characteristics, formation and function. Nat. Rev. Mol. Cell Biol.

Rac1 and Cdc42Hs. Mol. Biol. Cell 9, 1379-1394. 12, 413-426. Journal of Cell Science

1076 RESEARCH ARTICLE Journal of Cell Science (2017) 130, 1064-1077 doi:10.1242/jcs.195552

Nakahara, H., Otani, T., Sasaki, T., Miura, Y., Takai, Y. and Kogo, M. (2003). Shen, Y., Lyons, P., Cooley, M., Davidson, D., Veillette, A., Salgia, R., Griffin, Involvement of Cdc42 and Rac small G proteins in invadopodia formation of J. D. and Schaller, M. D. (2000). The noncatalytic domain of protein-tyrosine RPMI7951 cells. Cells 8, 1019-1027. phosphatase-PEST targets paxillin for dephosphorylation in vivo. J. Biol. Chem. Nakamura, K., Yano, H., Uchida, H., Hashimoto, S., Schaefer, E. and Sabe, H. 275, 1405-1413. (2000). Tyrosine phosphorylation of paxillin alpha is involved in temporospatial Spuul, P., Ciufici, P., Veillat, V., Leclercq, A., Daubon, T., Kramer, I. J. and regulation of paxillin-containing focal adhesion formation and F-actin organization Génot, E. (2014). Importance of RhoGTPases in formation, characteristics, and in motile cells. J. Biol. Chem. 275, 27155-27164. functions of invadosomes. Small GTPases 5, e28195. Nascimento, C. F., de Siqueira, A. S., Pinheiro, J. J. V., Freitas, V. M. and Jaeger, Tatin, F., Varon, C., Génot, E. and Moreau, V. (2006). A signalling cascade R. G. (2011). Laminin-111 derived peptides AG73 and C16 regulate invadopodia involving PKC, Src and Cdc42 regulates podosome assembly in cultured activity of a human adenoid cystic carcinoma cell line. Exp. Cell Res. 317, endothelial cells in response to phorbol ester. J. Cell Sci. 119, 769-781. 2562-2572. Tcherkezian, J. and Lamarche-Vane, N. (2007). Current knowledge of the large Oikawa, T., Itoh, T. and Takenawa, T. (2008). Sequential signals toward podosome RhoGAP family of proteins. Biol. Cell 99, 67-86. formation in NIH-src cells. J. Cell Biol. 182, 157-169. van Buul, J. D., Allingham, M. J., Samson, T., Meller, J., Boulter, E., Garcıa-́ ́ Patel, J. C. and Galan, J. E. (2006). Differential activation and function of Rho Mata, R. and Burridge, K. (2007). RhoG regulates endothelial apical cup GTPases during Salmonella-host cell interactions. J. Cell Biol. 175, 453-463. assembly downstream from ICAM1 engagement and is involved in leukocyte Petit, V., Boyer, B., Lentz, D., Turner, C. E., Thiery, J. P. and Vallés, A. M. (2000). trans-endothelial migration. J. Cell Biol. 178, 1279-1293. Phosphorylation of tyrosine residues 31 and 118 on paxillin regulates cell Vindis, C., Teli, T., Cerretti, D. P., Turner, C. E. and Huynh-Do, U. (2004). EphB1- migration through an association with CRK in NBT-II cells. J. Cell Biol. 148, mediated cell migration requires the phosphorylation of paxillin at Tyr-31/Tyr-118. 957-970. J. Biol. Chem. 279, 27965-27970. Pignatelli, J., Tumbarello, D. A., Schmidt, R. P. and Turner, C. E. (2012). Hic-5 Wang, H., Li, S.-H., Li, H., Li, C., Guan, K., Luo, G., Yu, L., Wu, R.-Q., Zhang, X., promotes invadopodia formation and invasion during TGF-beta-induced epithelial-mesenchymal transition. J. Cell Biol. 197, 421-437. Wang, J. et al. (2013). SGEF enhances EGFR stability through delayed EGFR Rossman, K. L., Der, C. J. and Sondek, J. (2005). GEF means go: turning on RHO trafficking from early to late endosomes. Carcinogenesis. 34, 1976-1983. GTPases with guanine nucleotide-exchange factors. Nat. Rev. Mol. Cell Biol. 6, Webb, D. J., Donais, K., Whitmore, L. A., Thomas, S. M., Turner, C. E., Parsons, 167-180. J. T. and Horwitz, A. F. (2004). FAK-Src signalling through paxillin, ERK and Samson, T., Welch, C., Monaghan-Benson, E., Hahn, K. M. and Burridge, K. MLCK regulates adhesion disassembly. Nat. Cell Biol. 6, 154-161. (2010). Endogenous RhoG is rapidly activated after epidermal growth factor Wennerberg, K., Ellerbroek, S. M., Liu, R.-Y., Karnoub, A. E., Burridge, K. and stimulation through multiple guanine-nucleotide exchange factors. Mol. Biol. Cell Der, C. J. (2002). RhoG signals in parallel with Rac1 and Cdc42. J. Biol. Chem. 21, 1629-1642. 277, 47810-47817. Schaller, M. D. and Parsons, J. T. (1995). pp125FAK-dependent tyrosine Wheeler, A. P., Wells, C. M., Smith, S. D., Vega, F. M., Henderson, R. B., phosphorylation of paxillin creates a high-affinity binding site for Crk. Mol. Cell. Tybulewicz, V. L. and Ridley, A. J. (2006). Rac1 and Rac2 regulate macrophage Biol. 15, 2635-2645. morphology but are not essential for migration. J. Cell Sci. 119, 2749-2757. Shen, Y., Schneider, G., Cloutier, J.-F., Veillette, A. and Schaller, M. D. (1998). Zaidel-Bar, R., Milo, R., Kam, Z. and Geiger, B. (2007). A paxillin tyrosine Direct association of protein-tyrosine phosphatase PTP-PEST with paxillin. phosphorylation switch regulates the assembly and form of cell-matrix adhesions. J. Biol. Chem. 273, 6474-6481. J. Cell Sci. 120, 137-148. Journal of Cell Science

1077