Rac1-PAK1 regulation of Rab11 cycling promotes junction destabilization

ARTICLE Rac1-PAK1 regulation of Rab11 cycling promotes junction destabilization Jennifer C. Erasmus 1 *, Kasia Smolarczyk 1 *, Helena Brezovjakova 1 , Noor F. Mohd-Naim 1 , Encarnación Lozano 1 , Karl Matter 2 ,and Vania M.M. Braga 1  Rac1 GTPase is hyperactivated in tumors and contributes to malignancy. Rac1 disruption of junctions requires its effector PAK1, but the precise mechanisms are unknown. Here, we show that E-cadherin is internalized via micropinocytosis in a PAK1–dependent manner without catenin dissociation and degradation. In addition to internalization, PAK1 regulates E-cadherin transport by fine-tuning Rab small GTPase function. PAK1 phosphorylates a core Rab regulator, RabGDIβ, but not RabGDIα. Phosphorylated RabGDIβpreferentially associates with Rab5 and Rab11, which is predicted to promote Rab retrieval from membranes. Consistent with this hypothesis, Rab11 is activated by Rac1, and inhibition of Rab11 function partially rescues E-cadherin destabilization. Thus, Rac1 activation reduces surface cadherin levels as a net result of higher bulk flow of membrane uptake that counteracts Rab11-dependent E-cadherin delivery to junctions (recycling and/or exocytosis). This unique small GTPase crosstalk has an impact on Rac1 and PAK1 regulation of membrane remodeling during epithelial dedifferentiation, adhesion, and motility. Introduction The small GTPase Rac1 plays a key role in the regulation of cell–cell adhesion and epithelial function in health and disease. Rac1 is essential for the formation and maintenance of cadherin contacts and differentiated epithelial tissues (McCormack et al., 2013). Yet, in a cancer context, uncontrolled Rac1 activation often correlates with metastatic behavior and poor prognosis, with cell–cell contact disruption, cell detachment, and enhanced migration (Porter et al., 2016). In addition to upregulation of Rac1 protein and mRNA levels, dysfunctional Rac1 signaling in tumors is also achieved by point mutations that increase Rac1 activation and hyperactivation of endogenous Rac1 by upstream regulators (exchange factors, oncogenes, or growth factor receptors; Maldonado et al., 2020;Olson, 2018;Porter et al., 2016). The impact and relevance of Rac1 in tumor progression is consistent with the breadth of its various activating mechanisms and the variety of tumor types affected (Maldonado et al., 2020). Here, we investigate the mechanisms by which inappropriate Rac1 activation perturbs cell–cell contacts as part of a malignancy program. In SCCf12 cells, activated Rac1 promotes E-cadherin internalization in a clathrin-independent manner (Akhtar and Hotchin, 2001). In normal keratinocytes, overexpression of active Rac1 requires signaling from its effector, PAK1, to remove E-cadherin from junctions (Lozano et al., 2008). PAK1 belongs to a family of serine/threonine kinases that has fundamental roles in different cellular processes (Kumar et al., 2017), including epithelial differentiation and morphogenesis in numerous organisms (Bahri et al., 2010; Pirraglia et al., 2010;Tay et al., 2010;Vlachos et al., 2015). Destabilization of cadherin-dependent junctions by PAK1 activation is consistent with the role of other PAK family members in the adhesion of tumor cell lines (Fram et al., 2014;Ismail et al., 2017;Morse et al., 2016;Selamat et al., 2015) and the wellestablished PAK1 function in promoting tumor migration and metastasis (Kumar and Li, 2016). The cellular processes by which PAK1 activity could mediate junction disassembly are not known. Rac1/PAK1 signaling can activate ROCK1 and thus cell contraction, which could contribute to junction perturbation; however, our previous work shows that cells flatten out upon Rac1 expression, and inhibition of ROCK does not rescue Rac1-dependent defects (Lozano et al., 2008). We hypothesize two alternative mechanisms. First, PAK1 could phosphorylate proteins found at cadherin complexes and modulate their binding affinity and/ or internalization, thereby weakening cell–cell adhesion. E-cadherin cytoplasmic tail has distinct motifs required for its internalization that are masked by the interaction with p120 CTN or β-catenin ............................................................................................................................................................................. 1 National Heart and Lung Institute, Faculty of Medicine, Imperial College London, London, UK; 2 Institute of Ophthalmology, University College London, London, UK. *J.C. Erasmus and K. Smolarczyk contributed equally to this paper; Correspondence to Vania M.M. Braga: [email protected];N.F.Mohd-Naim’s present address is Pengiran Anak Puteri Rashidah Sa’adatul Bolkiah Institute of Health Sciences, Universiti Brunei Darussalam, Bandar Seri Begawan, Brunei; E. Lozano’s present address is Department of Cell Biology, Faculty of Medicine, Complutense University, Madrid, Spain. © 2021 Erasmus et al. This article is available under a Creative Commons License (Attribution 4.0 International, as described at https://creativecommons.org/licenses/by/ 4.0/). Rockefeller University Press https://doi.org/10.1083/jcb.202002114 1of19 J. Cell Biol. 2021 Vol. 220 No. 6 e202002114 Downloaded from http://rupress.org/jcb/article-pdf/220/6/e202002114/1821346/jcb_202002114.pdf by Universitaria Sevilla Biblio user on 22 November 2023 (Kowalczyk and Nanes, 2012). It is feasible that PAK1 phosphorylation of cadherin or catenins could destabilize the complex and facilitate E-cadherin internalization. Indeed, unique Ser/Thr phosphorylation sites on the E-cadherin cytoplasmic tail have been shown to enhance (Lickert et al., 2000;McEwen et al., 2014)orweaken(Dupre-Crochet et al., 2007) its interaction with β-catenin. Furthermore, binding between α-catenin and β-catenin is strongly reduced by casein kinase II phosphorylation of α-catenin (Escobar et al., 2015;Ji et al., 2009)oratdifferent residues in β-catenin (Bek and Kemler, 2002). Second, Rac1 and subsequent PAK1 activation could modulate the trafficking of E-cadherin complexes, per se. Different oncogenes and destabilizing stimuli are known to modify the turnover rate of E-cadherin complexes by accelerating their internalization or preventing recycling back to the cell surface (Goldenring, 2013;Kowalczyk and Nanes, 2012). The various routes by which E-cadherin can traffic to and from cell–cell contacts are controlled by Rabs, a family of small GTPases that coordinate the formation of intracellular vesicles and vesicular docking, fusion, and motility (Wandinger-Ness and Zerial, 2014). Rac1 engagement with trafficking machinery and Rab GTPase signaling could play a role in the destabilization of cadherin adhesion. Rac1 signaling is known to crosstalk with Rab GTPases via modulation of the localization and activity levels of each other (Bouchet et al., 2016;Chen et al., 2014;Diaz et al., 2014;Margiotta et al., 2017;Mori et al., 2014;Shim et al., 2010) or via shared activators or effectors (Bouchet et al., 2018; Carroll et al., 2013;Kunita et al., 2007;Topp et al., 2004). We favor the possibility that Rac1 coordination with Rab upstream regulators may control Rab activation/inactivation cycling, which is strictly coupled to Rab localization at different vesicular compartments. Similar to Rho GTPases, Rabs are activated by guanine nucleotide exchange factors (GEFs), inactivated by GTPase-activating proteins(GAPs), and sequestered by Rab GTP-dissociation inhibitor (RabGDI; Stenmark, 2009). The brain-specific RabGDIαand the ubiquitously expressed RabGDIβ(Nishimura et al., 1994) retrieve an inactive Rab from the donor vesicular compartment and keep it in a cytosolic pool until its delivery to an acceptor vesicle or organelle, enabling localized Rab activation by GEFs (Shinde and Maddika, 2018). We identify novel mechanisms by which Rac1 and PAK1 signaling disrupt E-cadherin adhesion in normal keratinocytes. Rac1 activation induces E-cadherin internalization via micropinocytosis in a PAK1-dependent manner: There is no dissociation of catenins, and cadherin complexes are not targeted for degradation within the timeframe analyzed. Our data indicate that regulation of E-cadherin trafficking occurs by two mechanisms. First, activation of Rab11 by Rac1. Rab11 operates at the crossroads between endocytic and exocytic transport: (i) slow recycling of internalized cargo to polarized regions of epithelial membrane and (ii) delivery of transmembrane proteins from the trans-Golgi network via exocytosis, which may or may not occur via the recycling compartment (McDermott and Kim, 2015;Welz et al., 2014). Second, a specific phosphorylation of RabGDIβ, but not RabGDIα, by PAK1. Such post-translation modification increases RabGDIβaffinity to selected Rabs, thereby interfering with specific trafficking routes. Thus, our data reveal novel PAK1 functions in intracellular trafficking with impact on the modulation of cell–cell contact in pathological conditions. In addition, the direct interplay between PAK1, RabGDIβ,andRab11hassignificant importance for other PAK1 functions that require membrane remodeling during motility, ruffling, and fluid uptake. Results Keratinocytes expressing activated Rac1—constitutively active Q61L mutation, similar to the activating Q61L and G12V mutations found in oncogenic Ras—had junctions disrupted in a characteristic pattern: E-cadherin receptors were removed from the cell corners first (Fig. 1 A, arrows; Braga et al., 2000). Following activation of Rac1 in SCCf12 keratinocytes, enlarged vesicles containing E-cadherin complexes were observed (Akhtar and Hotchin, 2001). In our hands, normal keratinocytes also had numerous smaller intracellular vesicles in which E-cadherin and Rac1 colocalized (Fig. 1 A, yellow arrowheads). To understand the mechanism by which Rac1 promotes E-cadherin internalization, we initially assessed whether catenins were selectively released from internalized cadherin complexes. Normal keratinocytes were injected with constitutively active Rac1 (myc-Rac1 Q61L ) and costained for E-cadherin and catenins (Fig. 1 B). Images were segmented to eliminate junctional staining, and fluorescence signals in the resulting cytoplasmic area were quantified. Upon Rac1 activation, there was no significant difference in the percentage of internalized E-cadherin (vesicular pool) that colocalized with α-catenin, β-catenin, or p120 CTN , e.g., 70–80% of internalized cadherins (vesicular pool) colocalized with catenins (Fig. 1 C). Similarly, the Pearson coefficient of the cytoplasmic catenins that colocalized with E-cadherin did not differ, albeit it has reduced values at around 0.5 (Fig. 1 D). Reduced Pearson coefficient values may reflect the contribution of the cytosolic pool of catenins associated with distinct partners (i.e., not on vesicles). Unfortunately, the striking Rac1-specific phenotype at junctions (with augmented fluorescence signal in the middle of contacts; Lozano et al., 2008) makes it challenging to compare the relative cadherin-catenin association at junctions with the vesicular pool (Brezovjakova et al., 2019). Similar analyses with E-cadherin immunoprecipitation from transduced (TAT-Rac1 Q61L ) or transfected cells (myc-Rac1 Q61L ) lysates did not show the release of catenins from the complex (Fig. S1 C). Together with previous reports (Akhtar and Hotchin, 2001), our results suggest that activation of Rac1 signaling does not selectively remove catenins from E-cadherin during internalization. To validate the above results biochemically, cell surface levels and internalized levels were measured following treatment with cell-permeable fusion proteins encoding activated Rac1 (TAT-Rac Q61L )and/ortheautoinhibitorymotifof PAK1 (TAT–autoinhibitory domain of PAK1 [PAK AID ]; Fig. 2). Controls showed that the TAT-fusion proteins reproduced our results with transfection of active Rac1 and inhibition of PAK1 on junctions (Fig. S1;Lozano et al., 2008). There was no significant degradation of E-cadherin complexes during the timeframe evaluated, as total levels of E-cadherin and catenins were unaltered by adding TAT-Rac1 Q61L or control peptide TAT by itself Erasmus et al. Journal of Cell Biology 2of19 Rac1 and PAK1 control Rab GTPase cycling https://doi.org/10.1083/jcb.202002114 Downloaded from http://rupress.org/jcb/article-pdf/220/6/e202002114/1821346/jcb_202002114.pdf by Universitaria Sevilla Biblio user on 22 November 2023 (Fig. 2, A and B). However, in the presence of activated Rac1, E-cadherin surface levels were substantially reduced within 4 h with a corresponding increase in the internalized E-cadherin pool (Fig. 2, C–F). The same profile was observed with α-and β-catenin levels (Fig. 2, C and E), consistent with their cointernalization with E-cadherin shown in immunofluorescence experiments (Fig. 1, B and C). Thus, Rac1 activation does not seem to promote extensive dissociation of catenins before cadherin internalization (Figs. 1 and S1 C). When endogenous PAK1 activation by Rac1 was inhibited (TAT-Rac Q61L +TAT-PAK AID ), a decrease in surface levels of cadherin complexes (Fig. 2 C) and co-internalization of E-cadherin and catenins was prevented (Fig. 2, E and F;seealsoFig. S2 A). TheseeffectswerespecificforE-cadherin,assurfaceandinternalized levels of β1 integrins were not perturbed to similar extent as E-cadherin (Fig. 2, G and H). We concluded that, upon constitutive activation of Rac1 and PAK1, E-cadherin complexes are internalized, but not degraded. Figure 1. Active Rac1 is internalized with E-cadherin and catenins. Active Rac1 (pRK5-myc-Rac Q61L ) was microinjected and expressed for 3 h. (A and B) Keratinocytes were fixed and stained with myc-tag and E-cadherin antibodies (A) or with antibodies against catenins (B). (A) E-cadherin and Rac Q61L are cointernalized. (B–D) Colocalization of internalized E-cadherin and catenins. (B) Cells were labeled for active Rac1, cadherin and α-catenin, β-catenin, or p120 CTN . (C and D) Quantification of internalized pools (see Materials and methods). (C) The percentage of E-cadherin pixels on intracellular vesicles that colocalize with catenins was quantified. (D) Pearson coefficient shows the colocalization of the internal pool of catenins (cytosolic and vesicular) with internalized E-cadherin (vesicular pool). Merged images are shown on the right columns (A and B) and zoom images are shown in the bottom row (A). Arrows show loss of E-cadherin at junctions, arrowheads point to vesicles containing E-cadherin and catenins. Scale bars = 2 µm. Images are representative of three independent biological experiments (thereafter n= 3), and error bars represent SD. Erasmus et al. Journal of Cell Biology 3of19 Rac1 and PAK1 control Rab GTPase cycling https://doi.org/10.1083/jcb.202002114 Downloaded from http://rupress.org/jcb/article-pdf/220/6/e202002114/1821346/jcb_202002114.pdf by Universitaria Sevilla Biblio user on 22 November 2023 PAK1 phosphorylates cadherin tail and β-catenin to strengthen their interaction We initially addressed the potential destabilization of cadherin complexes by phosphorylation. In in vitro kinase assays, purified PAK1 kinase phosphorylated E-cadherin tail and β-catenin fusion proteins and the positive control maltose-binding protein (MBP; Fig. 3, A and B). A weaker phosphorylation of α-catenin was also observed but not investigated further here. It is feasible that the identified phosphorylation of cadherin tail or β-catenin by PAK1 may contribute to the release of cadherin complexes from junctions. In vitro reconstitution assays were set up to evaluate the modulation of cadherin-catenin interaction by PAK1 phosphorylation. Stronger interaction of β-catenin with phosphorylated GST–E-cadherin tail was observed (Fig. 3 C). In addition, phosphorylation of preassembled E-cadherin complexes by PAK1 also enhanced the presence of catenins in the precipitated samples (Fig. 3, D and E). Contrary to expectations, PAK1 phosphorylation of E-cadherin tail does not reduce β-catenin association. Instead, phosphorylation promotes a more stable complex in vitro (Fig. 3, C and D). Consistent with our findings, no Figure 2. PAK1 is necessary for internalization of E-cadherin upon Rac Q61L overexpression. Keratinocytes were treated with cell-permeable TAT or TATRac Q61L in the presence or absence of TAT-PAK AID to inhibit endogenous PAK activation. (A and B) Following a time course, cells were surface biotinylated and processed to show total protein levels. (C–F) Alternatively, proteins were precipitated with streptavidin to monitor surface levels (C and D) or internalized (E and F) levels of E-cadherin and associated catenins. (B, D, and F) Quantification of E-cadherin levels. Representative blots from one independent biological replicate are shown on the left and quantification is shown in graphs on the right (additional replicates are shown in Fig. S2 A). (G and H) Quantification of surface (G) and internalized levels (H) of β1-integrin (n=3). Erasmus et al. Journal of Cell Biology 4of19 Rac1 and PAK1 control Rab GTPase cycling https://doi.org/10.1083/jcb.202002114 Downloaded from http://rupress.org/jcb/article-pdf/220/6/e202002114/1821346/jcb_202002114.pdf by Universitaria Sevilla Biblio user on 22 November 2023 changes in endogenous cadherin complex stoichiometry were observed by various precipitation approaches (Fig. S1 C), despite strong disruption of cadherin-mediated adhesion. Putative PAK1 phosphorylation sites on β-catenin were predicted by mass spectrometry at Thr551, Ser552, or Ser675 (Fig. S2 B), in line with previous studies (Rennefahrt et al., 2007; Taurin et al., 2006). The mutant S675A was unable to be phosphorylated by PAK1 in vitro, suggesting that β-catenin is phosphorylated at a single site by PAK1 (Fig. S2 C) and confirming previous findings (Zhu et al., 2012). PKA, PAK1, and PAK4 phosphorylate β-catenin at Ser675, promoting stabilization and increased transcription of β-catenin responsive genes in cell lines (Hino et al., 2005;Li et al., 2012;Selamat et al., 2015;Zhu et al., 2012). However, we were unable to demonstrate higher transcription levels of β-catenin responsive genes in normal keratinocytes (data not shown). The impact of β-catenin Ser675 phosphorylation on cadherin adhesion function or stability has not been determined. In vitro complex reconstitution showed that phosphomimetic β-catenin (S675D) interacted with recombinant E-cadherin tail more efficiently than WT or nonphosphorylatable mutant β-catenin (S675A; Fig. S2 D). However, a phophomimetic β-catenin mutant did not show stronger binding for α-catenin (Fig. 3 F). These results indicate that PAK1 phosphorylation of β-catenin enhances the affinity for cadherin tail, but not α-catenin. Using an antibody against phosphorylated β-catenin at Ser675, a pool of endogenous phosphorylated β-catenin was detected at junctions during homeostasis and at stable cell-cell contacts (Fig. 3 G). In Rac1-expressing cells, phosphorylated β-catenin was also present at disrupted junctions and in intracellular tubules and vesicles (Fig. 3 G, zoom). Taken together, these data strongly indicate that E-cadherin is in complex with catenins inside cells and that PAK1 phosphorylation unexpectedly enhances the association between cadherin and β-catenin. Following Rac1 activation, E-cadherin is internalized via micropinocytosis The unexpected finding that E-cadherin is internalized by Rac1 activation without significant dissociation of catenins in primary keratinocytes (Figs. 1,2,and3) is consistent with data from tumor cell lines that the disruption of cell–cell adhesion may not use the classical internalization routes (Akhtar and Hotchin, 2001). To identify the E-cadherin vesicular pool in the cytoplasm, we costained samples expressing Rac1 with E-cadherin antibodies and a panel of intracellular markers (Fig. 4). In spite of considerable perturbation of junctions, there was no substantial overlap between E-cadherin–containing vesicles and markers of clathrinor caveolin-dependent internalization (transferrin or caveolin, respectively), early endosomes, or late endosome/lysosome compartment (CD63; Fig. 4 A). By feeding cells with fluorescently labeled BSA, we could also exclude macropinocytosis, a typical membrane turnover process induced by Rac1 and PAK1 (Fig. 4 B; Dharmawardhane et al., 2000). In contrast, treating keratinocytes with a smaller fluorescent compound (dextran; 10 kD), substantial colocalization with E-cadherin on small vesicles was observed (Fig. 4 B). The micropinosome compartment and its intracellular trafficking routes are poorly defined. Nevertheless, our data Figure 3. PAK1 phosphorylation strengthens the interaction between E-cadherin and catenins. (A and B) In vitro kinase assay using different GSTtagged proteins as substrates and purified PAK1 kinase domain in the presence or absence of 32 P-ATP. Images show radioactively phosphorylated proteins. PAK1 autophosphorylation (A and B) and MBP (B) are used as internal controls. (C–F) In vitro reconstitution of cadherin complex with or without PAK1 phosphorylation. Diagrams on the left represent proteins used and numbers show the order of phosphorylation and binding. Blots show the amount of cleaved catenins that interact with GST–E-cadherin cytoplasmic tail (C–E) or GST–β-catenin (F) under each condition. Graphs on the right show quantification of the protein interaction, and values are expressed relative to controls (nonphosphorylated proteins). (G) Rac Q61L -transfected keratinocytes were stained with β-catenin pS675 antibody. Inset is a zoom of cell highlighted by the white square. Arrow shows enlargement of junctional staining, and arrowhead points to labeled intracellular tubular structures. Scale bar = 10 µm or 4 µm (zoom; n= 3). Samples were analyzed with ttest, and error bars represent SD. *, P < 0.05; **, P = 0.0002. Erasmus et al. Journal of Cell Biology 5of19 Rac1 and PAK1 control Rab GTPase cycling https://doi.org/10.1083/jcb.202002114 Downloaded from http://rupress.org/jcb/article-pdf/220/6/e202002114/1821346/jcb_202002114.pdf by Universitaria Sevilla Biblio user on 22 November 2023 indicate that, rather than macropinocytosisasshowninkeratinocyte tumor cell lines (Akhtar et al., 2000), Rac1 activation promotes E-cadherin internalization via fluid uptake that is likely to be micropinocytosis. PAK1 phosphorylates and modulates RabGDIβfunction Our results suggest that the junction defects caused by active Rac1 in normal keratinocytes cannot be explained by a PAK1driven, looser association between cadherin and catenins or Figure 4. Rac1 activation promotes E-cadherin internalization via fluid uptake. (A) Keratinocytes expressing activated Rac1 (Rac1 Q61L ) were fixed and stained with antibodies against the tag, E-cadherin, and various intracellular markers to visualize caveolae (caveolin), early endosomes (EEA1), or late endosomes/lysosomes (CD63). (A and B) Alternatively, keratinocytes expressing activated Rac1 were incubated with Alexa Fluor 568-Transferrin (A; A568transferrin) to detect clathrin-dependent internalized vesicles, Texas Red-BSA to label macropinocytosis, or Texas Red-Dextran to show micropinocytosis (B) for 30 min. After incubation, cells were stained and imaged using confocal microscopy. Merged files are shown in the last column. White arrows show intracellular marker staining, white arrowheads show cadherin vesicular staining, and yellow arrowheads show colocalization of E-cadherin with intracellular marker. Images are representative of three independent experiments. Scale bars = 2 µm. Erasmus et al. Journal of Cell Biology 6of19 Rac1 and PAK1 control Rab GTPase cycling https://doi.org/10.1083/jcb.202002114 Downloaded from http://rupress.org/jcb/article-pdf/220/6/e202002114/1821346/jcb_202002114.pdf by Universitaria Sevilla Biblio user on 22 November 2023 promotion of β-catenin nuclear function (data not shown). We reasoned that Rac1 signaling could modulate E-cadherin complex internalization by targeting the endocytic machinery. In mammalian cells, in addition to macropinocytosis (Dharmawardhane et al., 2000), PAK1 participation in intracellular trafficking has been reported with internalization (Karjalainen et al., 2008)and glucose uptake (Tunduguru et al., 2017); however, its endocytic roles are generally thought to occur via PAK1-dependent cytoskeletal reorganization rather than a direct modulation of the trafficking machinery. PAK1 may dissociate cell-cell junctions by regulating E-cadherin intracellular transport rather than complex stability. We speculated that PAK1 may phosphorylate RabGDI, the regulator of Rab small GTPase retrieval and delivery to different intracellular compartments (Shinde and Maddika, 2018). The RabGDI counterpart, RhoGDI, is a key regulator of Rho GTPases. PAK1 phosphorylates RhoGDI at Ser101 and Ser174, which promotes its dissociation from Rac1, but not RhoA (Ard et al., 2012; DerMardirossian et al., 2004). Similar phosphorylation of RabGDI could modulate their affinity for different Rabs and thus target specific trafficking pathways. RhoGDI was aligned with RabGDI sequences to determine whether the PAK1-phosphorylated amino acids are conserved in RabGDI (Fig. 5 A, highlighted in green). The main RhoGDI phosphorylated site by PAK1—Ser174—was replaced by a lysine (position 357) in RabGDIα(brain specific) and RabGDIβ(ubiquitously expressed). RhoGDI Ser101 corresponded to Thr248 found in both RabGDIαand RabGDIβand could be a site for PAK1 phosphorylation. However, in vitro kinase assay with purified proteins showed that PAK1 phosphorylated RabGDIβ, but not RabGDIα(Fig. 5 B). Such exclusive phosphorylation of RabGDIβ(Fig. 5 B) suggests that it is unlikely that T248 is the PAK1 site phosphorylated on RabGDIβand that alternative sites must exist. To provide insights into additional residues that PAK1 could phosphorylate on RabGDIβ, we aligned RabGDIβsequence with known PAK1 substrates. PAK1 substrates were grouped according to their consensus phosphorylated motifs and aligned with RabGDIβ(Fig. S3). From this analysis, putative PAK1 phosphorylation sites could be Ser285, Ser330, and Ser382 (Fig. S3). These amino acids are conserved among different species (Fig. S4, highlighted in gray). Two of the predicted RabGDIβphosphosites by PAK1 (Ser285 and Ser330) are also conserved in RabGDIαand were thus excluded from our consideration. The predicted phospho-site (Ser382) is found only in RabGDIβ and thus the likely site to be phosphorylated by PAK1 (Fig. S4, highlighted in yellow; Fig. 5 A, highlighted in cyan). Following in vitro kinase assays using PAK1 kinase, WT RabGDIβ,oramutant unable to be phosphorylated (S382A; Fig. 5 C) were run on Phostag gels, which is a qualitative assessment using retardation of phosphorylated proteins as a readout. A mobility shift was observed with WT RabGDIβ, consistent with addition of a negative charge by phosphorylation. Mutation to alanine at residue 382 (S382A) abolished the shift, suggesting that this is the site where PAK1 phosphorylated RabGDIβ(Fig. 5 C). Serine 382 localizes at the C terminus of RabGDIβ, outside the RabGDI sequence conserved regions (Fig. 5 D;Luan et al., 2000; Schalk et al., 1996;Wu et al., 1998). This residue is also away from known amino acids that modulate Rab binding on RabGDIβ (i.e., Tyr39 and Tyr249; Shisheva et al., 1999)orRabGDIα(Ser121 and Ser45; Cavalli et al., 2001). Yet, in the 3D structure, all of these residues are in close proximity within the predicted Rab binding platform (Fig. 5, E and F;Luan et al., 2000). These analyses indicate that PAK1 phosphorylation at Ser382 may regulate the interaction between RabGDIβand Rabs. The localization of RabGDI in cells is unknown. To investigate whether PAK1 phosphorylation alters RabGDIβlocalization, pEGFP-RabGDIβWT and mutants (S382A or S382D) were expressed in keratinocytes by themselves and showed cytoplasmic distribution and an unexpected junctional localization (Fig. S5 A, arrows). In addition to the predicted cytoplasmic localization, in the presence of activated Rac1 (pRFP-Rac1 Q61L ;Fig. 6 A), all RabGDIβconstructs colocalized with Rac1 at junctions and at tubular structures originating from cell–cell contacts (Fig. 6 A, merge zoom, arrows). When the intensity at junctions was quantified, there were no differences in RabGDIβlevels at steady state (Fig. 6 B). Following Rac1 activation, a small but significant increase in nonphosphorylatable RabGDIβ(S382A) intensity at cell–cell contacts was observed (Fig. 6 B). We next evaluated the potential influence of Rac1 on the localization of pEGFP-RabGDIβ fluorescence at the contacting interface between neighboring cells (Fig. 6, C–E). Rac1 activation reduced the number of pixels of both RabGDIβmutants at the contacting interface area (Fig. 6 D), while there was no significant change in the levels of WT RabGDIβ. When coverage index was considered (i.e., RabGDIβ pixel length that covered the contacting interface length), there was a consistent reduction of all constructs at junctions (Fig. 6 E). We concluded that RabGDIβphosphorylation does not alter its junctional localization at steady state or in response to Rac1 activation (Fig. 6, B and D). Rac1 signaling activates Rab11 The functional significance of RabGDIβphosphorylation was tested on its ability to interact with different Rabs (Fig. 7). GST or different GST-RabGDIβfusion proteins were incubated with keratinocyte lysates at steady state and precipitated Rabs detected by Western blots (Fig. 7 A). Phosphomimetic RabGDI mutant (S382D) showed a significant increase in interaction with endogenous Rab5 and Rab11 compared with WT or nonphosphorylated forms (Fig. 7 B). No differences were detected in the association with Rab7 or Rab22, indicating that distinct Rabs can be discriminated by phosphomimetic RabGDIβ. Thus, the differential binding of RabGDIβto Rabs may suggest that the retrieval of selected Rabs from membranes is modified by phosphorylation at Ser382. We focused on the potential role of Rab11 in the disassembly of junctions caused by Rac1. PAK1 was not able to phosphorylate Rab11 in vitro (data not shown), but it is feasible that Rac1 signaling could modulate Rab11 activity. For these experiments, we tested Rac1 mutants containing the constitutively active mutation Q61L or the tumor-derived, fast-cycling mutation P29S, highly prevalent in melanomas (Porter et al., 2016). Expression of these mutants is predicted to mimic WT Rac1 gene amplification or mRNA overexpression in tumors (Porter et al., 2016). Erasmus et al. Journal of Cell Biology 7of19 Rac1 and PAK1 control Rab GTPase cycling https://doi.org/10.1083/jcb.202002114 Downloaded from http://rupress.org/jcb/article-pdf/220/6/e202002114/1821346/jcb_202002114.pdf by Universitaria Sevilla Biblio user on 22 November 2023 Figure 5. PAK1 phosphorylates RabGDIβ.(A)Sequence alignment of RabGDIα, RabGDIβ, and RhoGDI. PAK1 phosphorylation sites on RhoGDI are highlighted in green. Based on PAK1 phosphorylation motifs of known substrates (Fig. S3), putative phosphorylated site by PAK1 on RabGDIβis highlighted (light blue). (B) In vitro kinase assay using purified proteins incubated with PAK1 kinase with and without radioactive ATP:RabGDIα,RabGDIβ, and MBP (positive control) or GST (negative control). PAK1 autophosphorylation is shown. (C) RabGDIβWT or nonphosphorylatable mutant (S382A) were phosphorylated in vitro as described in B and separated using a Phostag gel to show mobility retardation of phosphorylated proteins. (D) Schematic diagram of RabGDIα/βdomain structure showing shared conserved domains (sequence conserved region [SCR]; Schalk et al., 1996) and phosphorylated amino acids on RabGDIαor RabGDIβknown to modulate Rab binding. (E and F) Crystal structure of RabGDI-α(Protein Data Bank accession no. 1GND) mapping the different phosphorylation sites. Phosphorylated residue identified in RabGDIβis shown in purple font (S382). (F) Zoom of the Rab binding platform on RabGDIαshows the proximity of residues identified in this work and in the literature. Erasmus et al. Journal of Cell Biology 8of19 Rac1 and PAK1 control Rab GTPase cycling https://doi.org/10.1083/jcb.202002114 Downloaded from http://rupress.org/jcb/article-pdf/220/6/e202002114/1821346/jcb_202002114.pdf by Universitaria Sevilla Biblio user on 22 November 2023 Using GST-FIP3 pull-down assays (Franco et al., 2014), Rab11 was activated following expression of either Rac1 mutant (Fig. 7, C and D). Furthermore, the oncogenic Rac1 P29S was also able to disrupt cell-cell contacts, although less efficiently than with Rac Q61L (Fig. 7 E). We concluded that, rather than inactivating Rab11, Rac1 mutants promote Rab11 activation. We surmise that it is likely that Rab11-dependent transport participates in cadherin adhesion disruption. To test whether Rab11 function is necessary downstream of Rac1 activation, we used two approaches: Expression of Rab11 mutants (Fig. 8) or depletion of endogenous Rab11 (Fig. 9). Following expression of constitutively active Rab11 (S20V) or dominant-negative Rab11 (S25N), samples were stained with anti–E-cadherin antibodies (Fig. 8 A) and the percentage of contacting interface length covered by cadherin staining was quantified (Fig. 8 B). In the presence of activated Rac1, preventing endogenous Rab11 activation with the dominant-negative Rab11 S25N partially rescued the perturbation of cell–cell contacts (Fig. 8 B). In contrast, coexpression with GTP-locked Rab11 (Rab11 S20V ) had no effect. These results are consistent with the interpretation that activation of endogenous Rab11 by Rac1 (Fig. 7, C and D) is required for junction disruption. We confirmed the above data by depletion of endogenous Rab11 in the presence of Rac1 activation (Fig. 9). Rab11 depletion Figure 6. RabGDIβlocalizes at cell–cell contacts and cytoplasm of keratinocytes. (A) pEGFP-RabGDIβWT and mutants nonphosphorylatable (S382A) or phosphomimetic (S382D) were expressed in keratinocytes in the presence of activated Rac1 (mRFP-Rac Q61L ; see also Fig. S5 A). Cells were fixed and representative confocal images are shown. (B) Staining levels of RabGDIβat junctions between coexpressing cells were quantified by measuring the total intensity of exogenous proteins at cell–cell contacts. (C) Diagram showing the quantified junctional areas: Contacting membrane between neighboring cells and the staining fragments of the junction marker. (D and E) Graphs show the area of the contacting interface that contains RabGDIβpixels (D; interface occupancy) and the length of RabGDIβfluorescence that covers the contacting interface length (E; coverage index). Scale bar = 20 µm or 7 µm (zoom). Arrows point to tubular structures at cell–cell contacts where Rac1 and RabGDIβcolocalize. Statistical analyses performed with Kruskal-Wallis with Dunn’smultiple comparison test; error bars represent SEM. *, P < 0.05; **, P < 0.01; ***, P < 0.001. Erasmus et al. Journal of Cell Biology 9of19 Rac1 and PAK1 control Rab GTPase cycling https://doi.org/10.1083/jcb.202002114 Downloaded from http://rupress.org/jcb/article-pdf/220/6/e202002114/1821346/jcb_202002114.pdf by Universitaria Sevilla Biblio user on 22 November 2023 15 min (60 mM glutathione, 75 mM NaCl, 10 mM EDTA, 75 mM NaOH, and 1% BSA). Cells were lysed in radioimmunoprecipitation assay buffer (150 mM NaCl, 20 mM Tris 7.4, 0.1% SDS, 1% Triton, 0.5% deoxycholate, 5 mM EDTA) with protease inhibitors, and an aliquot was separated to measure the total amount of E-cadherin. Internalized biotinylated proteins were recovered from lysates by coprecipitation with streptavidin beads. The amount of internalized and total E-cadherin was quantified by Western blot. To determine surface protein levels, keratinocytes were first treated with TAT-proteins for various time points, and then surface biotinylation was performed, followed by washing with quenching reagent and PBS. Cells were lysed in radioimmunoprecipitation assay buffer, and surface biotinylated proteins were recovered using streptavidin beads and identified by Western blot. Quantification of the total level of E-cadherin was normalized to actin expression. Values of total E-cadherin and surface proteins at time x (t X −level observed in time x) were quantified as a ratio of the amount at time 0 (t 0 ). Internalized proteins were quantified as a ratio of the amount at time x (t X )to the amount of internalized protein after 2 h (t 2 ). Production and purification of recombinant fusion proteins For GST-fusion proteins, bacteria were cultured overnight and induced with 0.3 mM IPTG (Calbiochem) for 3–5 h at 30°C or overnight at 16°C. The bacterial pellet was resuspended in lysis buffer (50 mM Tris HCl [pH 7.5], 100 mM NaCl, 5 mM DTT, 1 mM PMSF; a cocktail of protease inhibitors leupeptin, pepstatin, and pefabloc at 5 µg/ml each). Proteins were dialyzed against dialysis buffer (15 mM Tris HCl [pH 7.5], 150 mM NaCl, and 0.1 mM DTT). His-tagged TAT-fusion proteins were purified from bacteria pellet lysed in Lysis Buffer (standard buffer[SB]3mMMgCl 2 , 5 µg/ml lysozyme, 10 µg/ml DNase, 1 mM PMSF; cocktail of protease inhibitors, as above) on Nicharged chelating-sepharose beads (GE Healthcare) followed by washes with SB (30 mM Tris HCl [pH 7.5], 100 mM NaCl, 1mMβ-mercaptoethanol, 5 mM MgCl 2 )plus20mMimidazole. His-tagged fusion proteins were eluted from beads with elution buffer (SB + 1 M imidazole) and dialyzed against SB. GST-FIP3 (Franco et al., 2014) and GST-p120 CTN were prepared as described. In vitro kinase assay Recombinant PAK1 kinase domain (4 µg) was incubated with 10 µCi [γ32 P]-ATP (PerkinElmer) and GST-fusion proteins (0.4 nmol) trapped on glutathione-sepharose beads in phosphorylation buffer (50 mM Hepes [pH 7.3], 10 mM MgCl 2 ,10mMsodium fluoride, and 2 mM MnCl 2 ) containing 40 µM cold ATP. Reaction was incubated for 5 min at 30°C and terminated by washes with phosphorylation buffer and addition of SDS sample buffer. Proteins were separated in SDS-PAGE gel and phosphorylation was visualized by autoradiography. Alternatively, kinase assay was also used for RabGDIβ-phosphorylation detection with nonradioactive ATP and samples run on Phos-tag gels (Wako Pure Chemical Industries) to determine whether phosphorylation had occurred. For identification of phosphorylation sites, in vitro kinase reactions were analyzed by mass spectrometry (commercially, FingerPrints Proteomics, Dundee University). In vitro binding assay For GST pull-down assay, recombinant GST-fusion protein (catenins or cadherin cytoplasmic tail) was coupled for 1 h at 4°C to 20 µl glutathione-Sepharose beads (GE Healthcare), and then beads were washed with phosphorylation buffer (50 mM Hepes [pH 7.3], 10 mM MgCl 2 , 10 mM sodium fluoride, and 2 mM MnCl 2 ) and incubated with 4 µg recombinant PAK1 kinase domain and 10 mM cold ATP for 5 min at 30°C. Reaction was terminated by washes with ice-cold phosphorylation buffer. After kinase reaction, beads were incubated with different concentrations of cleaved proteins (0.05 µg and 0.1 µg in phosphorylation buffer containing 0.3 M NaCl and 0.5% Triton X-100) for 30 min at 4°C, followed by washing three times with phosphorylation buffer plus 0.3 M NaCl and 0.5% Triton X-100. Proteins bound to beads were eluted with SDS sample buffer containing freshly added DTT (0.1 M) and analyzed by SDSPAGE and immunoblotting. Alternatively, protein complexes were formed in vitro before PAK1 phosphorylation. GST-RabGDIβ WT ,GST-RabGDIβ S302A , and GST-RabGDIβ S302D were used to pull down endogenous Rabs from keratinocyte lysates. Briefly, cells were lysed in lysis buffer (50 mM Tris HCL, 0.5% NP-40, 150 mM NaCl, 1 mM MgCl 2 ,1mMEDTA, 1 mM PMSF, and protease inhibitor cocktail, as above). Lysates were incubated with GST-RabGDIβbeads for 1 h at 4°C and washed, and samples were used for Western blots to detect associated Rabs. To determine levels of Rab11 activation, cells were transfected with constitutively active Rac1 constructs overnight, washed in cold PBS, and lysed in 50 mM Tris HCL (pH 7.4), 100 mM NaCl, 1% NP-40, 10% glycerol, 10 mM MgCl 2 , 10 mM sodium fluoride, 1 mM sodium orthovanadate, 1 mM PMSF, and protease inhibitor cocktail, as above. Samples were processed for pull downs using GST-FIP3 essentially as described in Franco et al. (2014), except for shorter spins of 2 min and three washes after the pulldown. Quantification and statistics Quantification of junction phenotypes used a semiautomated custom-made software, Junction Mapper (Brezovjakova et al., 2019). Quantification of the levels of E-cadherin at junctions was performed using the parameter Coverage Index that measures the proportion of the length of contacting interface that is covered by cadherin staining (Lozano et al., 2008), the percentage of the area of the contacting interface that contains pixels of the junction marker, measurement of the pixel length of contacting interface, and length of the staining of the junction marker. Images were processed using Adobe Photoshop and Adobe Illustrator. Colocalization parameters were measured using SimplePCI 6 software. Western blot films in the linear range exposure were scanned and identified bands quantified using WCIF ImageJ software. Quantification of the levels of active Rab11 were performed as described (Nola et al., 2011). Levels of endogenous Rab11 associated with GST-FIP3 (active Rab11) were then normalized to the Erasmus et al. Journal of Cell Biology 16 of 19 Rac1 and PAK1 control Rab GTPase cycling https://doi.org/10.1083/jcb.202002114 Downloaded from http://rupress.org/jcb/article-pdf/220/6/e202002114/1821346/jcb_202002114.pdf by Universitaria Sevilla Biblio user on 22 November 2023 total levels of Rab11 found in the respective lysates. Samples expressing GFP were used to calculate the basal line of Rab11 activation and arbitrarily set as one. Stimulated samples were expressed as fold change to levels found in GFP samples (basal levels). RabGDIβinteractionwith distinct Rabs was quantified in a similar way by normalizing the levels of associated Rabs with mutants S382A or S382D to the amount detected associated with WT RabGDIβand expressed as fold change. Error bars represent error of the means, unless stated otherwise in figure legends. Statistical analysis was performed with ttests or Kruskal-Wallis with Dunn’s multiple comparison test using GraphPad Prism. Data distribution was assumed to be nonparametric, but this was not formally tested. Online supplemental material Fig. S1 shows cellular effects of TAT proteins used in this study. Fig. S2 shows that PAK1 phosphorylation of β-catenin does not participate in junction perturbation in keratinocytes. Fig. S3 shows the search for putative phosphorylation sites in RabGDIβ.Fig. S4 showsthealignmentofRabGDIαand RabGDIβ from human, bovine, and mouse species. Fig. S5 shows that localization of RabGDIβor Rab11 mutants at junctions does not impair cell–cell contacts. Acknowledgments We would like to acknowledge the contribution of Dr. M. Carstens for cloning and initial optimization of in vitro experiments with β-catenin mutants, and Ms. A.C. Garai for cloning the RabGDIβWT in mammalian expression vectors. Funding was provided by the Medical Research Council (K. Smolarczyk), Cancer Research UK (E. Lozano), Brunei Government PhD studentship (N.F. Mohd-Naim), and Biotechnology and Biological Sciences Research Council (J.C. Erasmus and K. Matter). The authors declare no competing financial interests. Author contributions: K. Smolarczyk and J.C. Erasmus contributed to the investigation, methodology, formal analyses, validation, visualization, and writing the original draft. E. Lozano contributed to the investigation. N.F. Mohd-Naim contributed to validation. H. Brezovjakova helped with formal analyses, visualization, and validation. K. Matter participated with resources and writing (review and editing). V.M.M. Braga was involved in the conceptualization, project administration, writing, and funding acquisition. Submitted: 20 February 2020 Revised: 21 September 2020 Accepted: 4 February 2021 References Akhtar, N., and N.A. Hotchin. 2001. RAC1 regulates adherens junctions through endocytosis of E-cadherin. Mol. Biol. Cell. 12:847–862. https:// doi.org/10.1091/mbc.12.4.847 Akhtar, N., K.R. Hudson, and N.A. Hotchin. 2000. Co-localization of Rac1 and E-cadherin in human epidermal keratinocytes. Cell Adhes. Commun. 7: 465–476. https://doi.org/10.3109/15419060009040304 Alory, C., and W.E. Balch. 2001. Organization of the Rab-GDI/CHM superfamily: the functional basis for choroideremia disease. Traffic. 2: 532–543. https://doi.org/10.1034/j.1600-0854.2001.20803.x Ard, R., K. Mulatz, H. Abramovici, J.C. Maillet, A. Fottinger, T. Foley, M.R. Byham, T.A. Iqbal, A. Yoneda, J.R. Couchman, et al. 2012. Diacylglycerol kinase ζregulates RhoA activation via a kinase-independent scaffolding mechanism. Mol. Biol. Cell. 23:4008–4019. https://doi.org/10.1091/mbc .e12-01-0026 Bahri, S., S. Wang, R. Conder, J. Choy, S. Vlachos, K. Dong, C. Merino, S. Sigrist, C. Molnar, X. Yang, et al. 2010. The leading edge during dorsal closure as a model for epithelial plasticity: Pak is required for recruitment of the Scribble complex and septate junction formation. Development. 137:2023–2032. https://doi.org/10.1242/dev.045088 Banworth, M.J., and G. Li. 2018. Consequences of Rab GTPase dysfunction in genetic or acquired human diseases. Small GTPases. 9:158–181. Bek, S., and R. Kemler. 2002. Protein kinase CKII regulates the interaction of beta-catenin with alpha-catenin and its protein stability. J. Cell Sci. 115: 4743–4753. https://doi.org/10.1242/jcs.00154 Bouchet, J., I. Del R´ ıo-Iñiguez, R. Lasserre, S. Agüera-Gonzalez, C. Cuche, A. Danckaert, M.W. McCaffrey, V. Di Bartolo, and A. Alcover. 2016. Rac1Rab11-FIP3 regulatory hub coordinates vesicle traffic with actin remodeling and T-cell activation. EMBO J. 35:1160–1174. https://doi.org/10 .15252/embj.201593274 Bouchet, J., M.W. McCaffrey, A. Graziani, and A. Alcover. 2018. The functional interplay of Rab11, FIP3 and Rho proteins on the endosomal recycling pathway controls cell shape and symmetry. Small GTPases. 9: 310–315. https://doi.org/10.1080/21541248.2016.1224288 Braga, V.M.M., L.M. Machesky, A. Hall, and N.A. Hotchin. 1997. The small GTPases Rho and Rac are required for the establishment of cadherindependent cell-cell contacts. J. Cell Biol. 137:1421–1431. https://doi.org/10 .1083/jcb.137.6.1421 Braga, V.M.M., M. Betson, X. Li, and N. Lamarche-Vane. 2000. Activation of the small GTPase Rac is sufficient to disrupt cadherin-dependent cellcell adhesion in normal human keratinocytes. Mol. Biol. Cell. 11: 3703–3721. https://doi.org/10.1091/mbc.11.11.3703 Brezovjakova, H., C. Tomlinson, N. Mohd Naim, P. Swiatlowska, J.C. Erasmus, S. Huveneers, J. Gorelik, S. Bruche, and V.M. Braga. 2019. Junction Mapper is a novel computer vision tool to decipher cell-cell contact phenotypes. eLife. 8:e45413. https://doi.org/10.7554/eLife.45413 Bryant, D.M., M.C. Kerr, L.A. Hammond, S.R. Joseph, K.E. Mostov, R.D. Teasdale, and J.L. Stow. 2007. EGF induces macropinocytosis and SNX1modulated recycling of E-cadherin. J. Cell Sci. 120:1818–1828. https://doi .org/10.1242/jcs.000653 Carroll, B., N. Mohd-Naim, F. Maximiano, M.A. Frasa, J. McCormack, M. Finelli, S.B.Thoresen,L.Perdios,R.Daigaku,R.E.Francis,etal.2013.TheTBC/ RabGAP Armus Coordinates Rac1 and Rab7 Functions during Autophagy. Dev. Cell. 25:15–28. https://doi.org/10.1016/j.devcel.2013.03.005 Cavalli, V., F. Vilbois, M. Corti, M.J. Marcote, K. Tamura, M. Karin, S. Arkinstall, and J. Gruenberg. 2001. The stress-induced MAP kinase p38 regulates endocytic trafficking via the GDI:Rab5 complex. Mol. Cell. 7: 421–432. https://doi.org/10.1016/S1097-2765(01)00189-7 Chen, W., Y. Feng, D. Chen, and A. Wandinger-Ness. 1998. Rab11 is required for trans-golgi network-to-plasma membrane transport and a preferential target for GDP dissociation inhibitor. Mol. Biol. Cell. 9:3241–3257. https://doi.org/10.1091/mbc.9.11.3241 Chen, P.I., K. Schauer, C. Kong, A.R. Harding, B. Goud, and P.D. Stahl. 2014. Rab5 isoforms orchestrate a “division of labor”in the endocytic network; Rab5C modulates Rac-mediated cell motility. PLoS One. 9:e90384. https://doi.org/10.1371/journal.pone.0090384 de Beco, S., C. Gueudry, F. Amblard, and S. Coscoy. 2009. Endocytosis is required for E-cadherin redistribution at mature adherens junctions. Proc. Natl. Acad. Sci. USA. 106:7010–7015. https://doi.org/10.1073/pnas .0811253106 Chen, Y., Y. Deng, J. Zhang, L. Yang, X. Xie, and T. Xu. 2009. GDI-1 preferably interacts with Rab10 in insulin-stimulated GLUT4 translocation. Biochem J. 422(2):229–235. https://doi.org/10.1042/BJ20090624 DerMardirossian, C., A. Schnelzer, and G.M. Bokoch. 2004. Phosphorylation of RhoGDI by Pak1 mediates dissociation of Rac GTPase. Mol. Cell. 15: 117–127. https://doi.org/10.1016/j.molcel.2004.05.019 DerMardirossian, C., G. Rocklin, J.-Y. Seo, and G.M. Bokoch. 2006. Phosphorylation of RhoGDI by Src regulates Rho GTPase binding and cytosol-membrane cycling. Mol. Biol. Cell. 17:4760–4768. https://doi.org/ 10.1091/mbc.e06-06-0533 Desclozeaux, M., J. Venturato, F.G. Wylie, J.G. Kay, S.R. Joseph, H.T. Le, and J.L. Stow. 2008. Active Rab11 and functional recycling endosome are Erasmus et al. Journal of Cell Biology 17 of 19 Rac1 and PAK1 control Rab GTPase cycling https://doi.org/10.1083/jcb.202002114 Downloaded from http://rupress.org/jcb/article-pdf/220/6/e202002114/1821346/jcb_202002114.pdf by Universitaria Sevilla Biblio user on 22 November 2023 required for E-cadherin trafficking and lumen formation during epithelial morphogenesis. Am. J. Physiol. Cell Physiol. 295:C545–C556. https://doi.org/10.1152/ajpcell.00097.2008 Dharmawardhane, S., A. Schürmann, M.A. Sells, J. Chernoff, S.L. Schmid, and G.M. Bokoch. 2000. Regulation of macropinocytosis by p21-activated kinase-1. Mol. Biol. Cell. 11:3341–3352. https://doi.org/10.1091/mbc.11.10 .3341 Diaz, J., P. Mendoza, R. Ortiz, N. Diaz, L. Leyton, D. Stupack, A.F. Quest, and V.A. Torres. 2014. Rab5 is required in metastatic cancer cells for Caveolin-1-enhanced Rac1 activation, migration and invasion. J. Cell Sci. 127: 2401–2406. https://doi.org/10.1242/jcs.141689 Dovas, A., Y. Choi, A. Yoneda, H.A. Multhaupt, S.H. Kwon, D. Kang, E.S. Oh, and J.R. Couchman. 2010. Serine 34 phosphorylation of rho guanine dissociation inhibitor (RhoGDIalpha) links signaling from conventional protein kinase C to RhoGTPase in cell adhesion. J. Biol. Chem. 285: 23296–23308. https://doi.org/10.1074/jbc.M109.098129 Dupre-Crochet, S., A. Figueroa, C. Hogan, E.C. Ferber, C.U. Bialucha, J. Adams, E.C. Richardson, and Y. Fujita. 2007. Casein kinase 1 is a novel negative regulator of E-cadherin-based cell-cell contacts. Mol. Cell. Biol. 27: 3804–3816. https://doi.org/10.1128/MCB.01590-06 Egami, Y., T. Taguchi, M. Maekawa, H. Arai, and N. Araki. 2014. Small GTPases and phosphoinositides in the regulatory mechanisms of macropinosome formation and maturation. Front. Physiol. 5:374. https://doi .org/10.3389/fphys.2014.00374 Escobar, D.J., R. Desai, N. Ishiyama, S.S. Folmsbee, M.N. Novak, A.S. Flozak, R.L. Daugherty, R. Mo, D. Nanavati, R. Sarpal, et al. 2015. α-Catenin phosphorylation promotes intercellular adhesion through a dual-kinase mechanism. J. Cell Sci. 128:1150–1165. https://doi.org/10.1242/jcs.163824 Fram, S., H. King, D.B. Sacks, and C.M. Wells. 2014. A PAK6-IQGAP1 complex promotes disassembly of cell-cell adhesions. Cell. Mol. Life Sci. 71: 2759–2773. https://doi.org/10.1007/s00018-013-1528-5 Franco, I., F. Gulluni, C.C. Campa, C. Costa, J.P. Margaria, E. Ciraolo, M. Martini, D. Monteyne, E. De Luca, G. Germena, et al. 2014. PI3K class II α controls spatially restricted endosomal PtdIns3P and Rab11 activation to promote primary cilium function. Dev. Cell. 28:647–658. https://doi.org/ 10.1016/j.devcel.2014.01.022 Frasa, M.A., F.C. Maximiano, K. Smolarczyk, R.E. Francis, M.E. Betson, E. Lozano, J. Goldenring, M.C. Seabra, A. Rak, M.R. Ahmadian, and V.M.M. Braga. 2010. Armus is a Rac1 effector that inactivates Rab7 and regulates E-cadherin degradation. Curr. Biol. 20:198–208. https://doi.org/10.1016/ j.cub.2009.12.053 Frasa,M.A.,K.T.Koessmeier,M.R.Ahmadian,andV.M.Braga.2012.Illuminating the functional and structural repertoire of human TBC/RABGAPs. Nat. Rev. Mol. Cell Biol. 13:67–73. https://doi.org/10.1038/nrm3267 Goldenring, J.R. 2013. A central role for vesicle trafficking in epithelial neoplasia: intracellular highways to carcinogenesis. Nat. Rev. Cancer. 13: 813–820. https://doi.org/10.1038/nrc3601 Goody, R.S., A. Rak, and K. Alexandrov. 2005. The structural and mechanistic basis for recycling of Rab proteins between membrane compartments. Cell. Mol. Life Sci. 62:1657–1670. https://doi.org/10.1007/s00018-005 -4486-8 Ha, K.D., S.M. Bidlingmaier, and B. Liu. 2016. Macropinocytosis Exploitation by Cancers and Cancer Therapeutics. Front. Physiol. 7:381. https://doi .org/10.3389/fphys.2016.00381 Hino, S., C. Tanji, K.I. Nakayama, and A. Kikuchi. 2005. Phosphorylation of β-catenin by cyclic AMP-dependent protein kinase stabilizes β-catenin through inhibition of its ubiquitination. Mol. Cell. Biol. 25:9063–9072. https://doi.org/10.1128/MCB.25.20.9063-9072.2005 Ismail, A.F., S. Oskay Halacli, N. Babteen, M. De Piano, T.A. Martin, W.G. Jiang, M.S. Khan, P. Dasgupta, and C.M. Wells. 2017. PAK5 mediates cell: cell adhesion integrity via interaction with E-cadherin in bladder cancer cells. Biochem. J. 474:1333–1346. https://doi.org/10.1042/BCJ20160875 Ji, H., J. Wang, H. Nika, D. Hawke, S. Keezer, Q. Ge, B. Fang, X. Fang, D. Fang, D.W. Litchfield, et al. 2009. EGF-induced ERK activation promotes CK2mediated disassociation of alpha-Catenin from beta-Catenin and transactivation of beta-Catenin. Mol. Cell. 36:547–559. https://doi.org/10 .1016/j.molcel.2009.09.034 Jou, T.-S., S.-M. Leung, L.M. Fung, W.G. Ruiz, W.J. Nelson, and G. Apodaca. 2000. Selective alterations in biosynthetic and endocytic protein traffic in Madin-Darby canine kidney epithelial cells expressing mutants of the small GTPase Rac1. Mol. Biol. Cell. 11:287–304. https://doi.org/10 .1091/mbc.11.1.287 Karjalainen, M., E. Kakkonen, P. Upla, H. Paloranta, P. Kankaanp¨ a¨ a, P. Liberali, G.H. Renkema, T. Hyypi¨ a, J. Heino, and V. Marjom¨ aki. 2008. A Raft-derived, Pak1-regulated entry participates in alpha2beta1 integrindependent sorting to caveosomes. Mol. Biol. Cell. 19:2857–2869. https:// doi.org/10.1091/mbc.e07-10-1094 Kowalczyk, A.P., and B.A. Nanes. 2012. Adherens junction turnover: regulating adhesion through cadherin endocytosis, degradation, and recycling. Subcell. Biochem. 60:197–222. https://doi.org/10.1007/978-94 -007-4186-7_9 Krauthammer, M., Y. Kong, B.H. Ha, P. Evans, A. Bacchiocchi, J.P. McCusker, E. Cheng, M.J. Davis, G. Goh, M. Choi, et al. 2012. Exome sequencing identifies recurrent somatic RAC1 mutations in melanoma. Nat. Genet. 44:1006–1014. https://doi.org/10.1038/ng.2359 Kumar, R., and D.Q. Li. 2016. PAKs in Human Cancer Progression: From Inception to Cancer Therapeutic to Future Oncobiology. Adv. Cancer Res. 130:137–209. https://doi.org/10.1016/bs.acr.2016.01.002 Kumar, R., R. Sanawar, X. Li, and F. Li. 2017. Structure, biochemistry, and biology of PAK kinases. Gene. 605:20–31. https://doi.org/10.1016/j.gene .2016.12.014 Kunita, R., A. Otomo, H. Mizumura, K. Suzuki-Utsunomiya, S. Hadano, and J.E. Ikeda. 2007. The Rab5 activator ALS2/alsin acts as a novel Rac1 effector through Rac1-activated endocytosis. J. Biol. Chem. 282: 16599–16611. https://doi.org/10.1074/jbc.M610682200 Li, Y., Y. Shao, Y. Tong, T. Shen, J. Zhang, Y. Li, H. Gu, and F. Li. 2012. Nucleocytoplasmic shuttling of PAK4 modulates β-catenin intracellular translocation and signaling. Biochim. Biophys. Acta. 1823:465–475. https://doi.org/10.1016/j.bbamcr.2011.11.013 Lickert, H., A. Bauer, R. Kemler, and J. Stappert. 2000. Casein kinase II phosphorylation of E-cadherin increases E-cadherin/β-catenin interaction and strengthens cell-cell adhesion. J. Biol. Chem. 275:5090–5095. https://doi.org/10.1074/jbc.275.7.5090 Liu, W., E.Y. Yuen, and Z. Yan. 2010. The stress hormone corticosterone increases synaptic alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptors via serumand glucocorticoid-inducible kinase (SGK) regulation of the GDI-Rab4 complex. J. Biol. Chem. 285:6101–6108. https://doi.org/10.1074/jbc.M109.050229 Lock, J.G., and J.L. Stow. 2005. Rab11 in recycling endosomes regulates the sorting and basolateral transport of E-cadherin. Mol. Biol. Cell. 16: 1744–1755. https://doi.org/10.1091/mbc.e04-10-0867 Lozano, E., M.A.M. Frasa, K. Smolarczyk, U.G. Knaus, and V.M.M. Braga. 2008. PAK is required for the disruption of E-cadherin adhesion by the small GTPase Rac. J. Cell Sci. 121:933–938. https://doi.org/10.1242/jcs .016121 Luan, P., A. Heine, K. Zeng, B. Moyer, S.E. Greasely, P. Kuhn, W.E. Balch, and I.A. Wilson. 2000. A new functional domain of guanine nucleotide dissociation inhibitor (α-GDI) involved in Rab recycling. Traffic. 1: 270–281. https://doi.org/10.1034/j.1600-0854.2000.010309.x Maldonado, M.D.M., J.I. Medina, L. Velazquez, and S. Dharmawardhane. 2020. Targeting Rac and Cdc42 GEFs in Metastatic Cancer. Front. Cell Dev. Biol. 8:201. https://doi.org/10.3389/fcell.2020.00201 Margiotta, A., C. Progida, O. Bakke, and C. Bucci. 2017. Rab7a regulates cell migration through Rac1 and vimentin. Biochim. Biophys. Acta Mol. Cell Res. 1864:367–381. https://doi.org/10.1016/j.bbamcr.2016.11.020 McCormack, J., N.J. Welsh, and V.M. Braga. 2013. Cycling around cell-cell adhesion with Rho GTPase regulators. J. Cell Sci. 126:379–391. https:// doi.org/10.1242/jcs.097923 McDermott, H., and K. Kim. 2015. Molecular dynamics at the endocytic portal and regulations of endocytic and recycling traffics. Eur. J. Cell Biol. 94: 235–248. https://doi.org/10.1016/j.ejcb.2015.04.003 McEwen, A.E., M.T. Maher, R. Mo, and C.J. Gottardi. 2014. E-cadherin phosphorylation occurs during its biosynthesis to promote its cell surface stability and adhesion. Mol. Biol. Cell. 25:2365–2374. https://doi .org/10.1091/mbc.e14-01-0690 Ming, Z., C. Guo, M. Jiang, W. Li, Y. Zhang, N. Fan, Y. Zhong, X. Meng, and S. Yang. 2014. Bioinformatics analysis of Rab GDP dissociation inhibitor beta and its expression in non-small cell lung cancer. Diagn. Pathol. 9: 201. https://doi.org/10.1186/s13000-014-0201-0 Mori, M., K. Saito, and Y. Ohta. 2014. ARHGAP22 localizes at endosomes and regulates actin cytoskeleton. PLoS One. 9:e100271. https://doi.org/10 .1371/journal.pone.0100271 Morse, E.M., X. Sun, J.R. Olberding, B.H. Ha, T.J. Boggon, and D.A. Calderwood. 2016. PAK6 targets to cell-cell adhesions through its N-terminus in a Cdc42-dependent manner to drive epithelial colony escape. J. Cell Sci. 129:380–393. https://doi.org/10.1242/jcs.177493 Nishimura, N., H. Nakamura, Y. Takai, and K. Sano. 1994. Molecular cloning and characterization of two rab GDI species from rat brain: brainspecific and ubiquitous types. J. Biol. Chem. 269:14191–14198. https:// doi.org/10.1016/S0021-9258(17)36773-X Erasmus et al. Journal of Cell Biology 18 of 19 Rac1 and PAK1 control Rab GTPase cycling https://doi.org/10.1083/jcb.202002114 Downloaded from http://rupress.org/jcb/article-pdf/220/6/e202002114/1821346/jcb_202002114.pdf by Universitaria Sevilla Biblio user on 22 November 2023 Nola, S., R. Daigaku, K. Smolarczyk, M. Carstens, B. Martin-Martin, G. Longmore, M. Bailly, and V.M. Braga. 2011. Ajuba is required for Rac activation and maintenance of E-cadherin adhesion. J. Cell Biol. 195: 855–871. https://doi.org/10.1083/jcb.201107162 Olson, M.F. 2018. Rho GTPases, their post-translational modifications, disease-associated mutations and pharmacological inhibitors. Small GTPases. 9:203–215. https://doi.org/10.1080/21541248.2016.1218407 Paterson, A.D., R.G. Parton, C. Ferguson, J.L. Stow, and A.S. Yap. 2003. Characterization of E-cadherin endocytosis in isolated MCF-7 and chinese hamster ovary cells: the initial fate of unbound E-cadherin. J. Biol. Chem. 278:21050–21057. https://doi.org/10.1074/jbc.M300082200 Pirraglia, C., J. Walters, and M.M. Myat. 2010. Pak1 control of E-cadherin endocytosis regulates salivary gland lumen size and shape. Development. 137:4177–4189. https://doi.org/10.1242/dev.048827 Polgar, N., and B. Fogelgren. 2018. Regulation of Cell Polarity by ExocystMediated Trafficking. Cold Spring Harb. Perspect. Biol. 10:a031401. https://doi.org/10.1101/cshperspect.a031401 Porter, A.P., A. Papaioannou, and A. Malliri. 2016. Deregulation of Rho GTPases in cancer. Small GTPases. 7:123–138. https://doi.org/10.1080/ 21541248.2016.1173767 Qin, X., J. Wang, X. Wang, F. Liu, B. Jiang, and Y. Zhang. 2017. Targeting Rabs as a novel therapeutic strategy for cancer therapy. Drug Discov. Today. 22:1139–1147. https://doi.org/10.1016/j.drudis.2017.03.012 Rennefahrt, U.E., S.W. Deacon, S.A. Parker, K. Devarajan, A. Beeser, J. Chernoff, S. Knapp, B.E. Turk, and J.R. Peterson. 2007. Specificity profiling of Pak kinases allows identification of novel phosphorylation sites. J. Biol. Chem. 282:15667–15678. https://doi.org/10.1074/jbc .M700253200 Sabatini, P.J., M. Zhang, R.V. Silverman-Gavrila, and M.P. Bendeck. 2011. Cadherins at cell-autonomous membrane contacts control macropinocytosis. J. Cell Sci. 124:2013–2020. https://doi.org/10.1242/jcs .076901 Schalk, I., K. Zeng, S.K. Wu, E.A. Stura, J. Matteson, M. Huang, A. Tandon, I.A. Wilson, and W.E. Balch. 1996. Structure and mutational analysis of Rab GDP-dissociation inhibitor. Nature. 381:42–48. https://doi.org/10.1038/ 381042a0 Schill, N.J., and R.A. Anderson. 2009. Out, in and back again: PtdIns(4,5)P(2) regulates cadherin trafficking in epithelial morphogenesis. Biochem. J. 418:247–260. https://doi.org/10.1042/BJ20081844 Selamat, W., P.L. Tay, Y. Baskaran, and E. Manser. 2015. The Cdc42 effector kinase PAK4 localizes to cell-cell junctions and contributes to establishing cell polarity. PLoS One. 10:e0129634. https://doi.org/10.1371/ journal.pone.0129634 Shim, J., S.M. Lee, M.S. Lee, J. Yoon, H.S. Kweon, and Y.J. Kim. 2010. Rab35 mediates transport of Cdc42 and Rac1 to the plasma membrane during phagocytosis. Mol. Cell. Biol. 30:1421–1433. https://doi.org/10.1128/MCB .01463-09 Shinde, S.R., and S. Maddika. 2018. Post translational modifications of Rab GTPases. Small GTPases. 9:49–56. https://doi.org/10.1080/21541248.2017 .1299270 Shisheva, A., S.R. Chinni, and C. DeMarco. 1999. General role of GDP dissociation inhibitor 2 in membrane release of Rab proteins: modulations of its functional interactions by in vitro and in vivo structural modifications. Biochemistry. 38:11711–11721. https://doi.org/10.1021/bi990200r Steger, M., F. Tonelli, G. Ito, P. Davies, M. Trost, M. Vetter, S. Wachter, E. Lorentzen, G. Duddy, S. Wilson, et al. 2016. Phosphoproteomics reveals that Parkinson’s disease kinase LRRK2 regulates a subset of Rab GTPases. eLife. 5:e12813. https://doi.org/10.7554/eLife.12813 Stenmark, H. 2009. Rab GTPases as coordinators of vesicle traffic. Nat. Rev. Mol. Cell Biol. 10:513–525. https://doi.org/10.1038/nrm2728 Takahashi, S., K. Kubo, S. Waguri, A. Yabashi, H.W. Shin, Y. Katoh, and K. Nakayama. 2012. Rab11 regulates exocytosis of recycling vesicles at the plasma membrane. J. Cell Sci. 125:4049–4057. https://doi.org/10.1242/jcs .102913 Taurin, S., N. Sandbo, Y. Qin, D. Browning, and N.O. Dulin. 2006. Phosphorylation of β-catenin by cyclic AMP-dependent protein kinase. J. Biol. Chem. 281:9971–9976. https://doi.org/10.1074/jbc.M508778200 Tay, H.G., Y.W. Ng, and E. Manser. 2010. A vertebrate-specific Chp-PAKPIX pathway maintains E-cadherin at adherens junctions during zebrafish epiboly. PLoS One. 5:e10125. https://doi.org/10.1371/journal .pone.0010125 Topp, J.D., N.W. Gray, R.D. Gerard, and B.F. Horazdovsky. 2004. Alsin is a Rab5 and Rac1 guanine nucleotide exchange factor. J. Biol. Chem. 279: 24612–24623. https://doi.org/10.1074/jbc.M313504200 Tunduguru, R., J. Zhang, A. Aslamy, V.A. Salunkhe, J.T. Brozinick, J.S. Elmendorf, and D.C. Thurmond. 2017. The actin-related p41ARC subunit contributes to p21-activated kinase-1 (PAK1)-mediated glucose uptake into skeletal muscle cells. J. Biol. Chem. 292:19034–19043. https://doi .org/10.1074/jbc.M117.801340 Vlachos, S., S. Jangam, R. Conder, M. Chou, T. Nystul, and N. Harden. 2015. A Pak-regulated cell intercalation event leading to a novel radial cell polarity is involved in positioning of the follicle stem cell niche in the Drosophila ovary. Development. 142:82–91. https://doi.org/10.1242/dev .111039 Wandinger-Ness, A., and M. Zerial. 2014. Rab proteins and the compartmentalization of the endosomal system. Cold Spring Harb. Perspect. Biol. 6:a022616. https://doi.org/10.1101/cshperspect.a022616 Welz, T., J. Wellbourne-Wood, and E. Kerkhoff. 2014. Orchestration of cell surface proteins by Rab11. Trends Cell Biol. 24:407–415. https://doi.org/ 10.1016/j.tcb.2014.02.004 West, J.J., and T.J. Harris. 2016. Cadherin trafficking for tissue morphogenesis: control and consequences. Traffic. 17:1233–1243. https://doi.org/10 .1111/tra.12407 Winter, J.F., S. H¨ opfner, K. Korn, B.O. Farnung, C.R. Bradshaw, G. Marsico, M. Volkmer, B. Habermann, and M. Zerial. 2012. Caenorhabditis elegans screen reveals role of PAR-5 in RAB-11-recycling endosome positioning and apicobasal cell polarity. Nat. Cell Biol. 14:666–676. https://doi.org/10 .1038/ncb2508 Wirtz-Peitz, F., and J.A. Zallen. 2009. Junctional trafficking and epithelial morphogenesis. Curr. Opin. Genet. Dev. 19:350–356. https://doi.org/10 .1016/j.gde.2009.04.011 Woichansky, I., C.A. Beretta, N. Berns, and V. Riechmann. 2016. Three mechanisms control E-cadherin localization to the zonula adherens. Nat. Commun. 7:10834. https://doi.org/10.1038/ncomms10834 Wu, S.-K., P. Luan, J. Matteson, K. Zeng, N. Nishimura, and W.E. Balch. 1998. Molecular role for the Rab binding platform of guanine nucleotide dissociation inhibitor in endoplasmic reticulum to Golgi transport. J. Biol. Chem. 273:26931–26938. https://doi.org/10.1074/jbc.273.41.26931 Wu, Y., K. Moissoglu, H. Wang, X. Wang, H.F. Frierson, M.A. Schwartz, and D. Theodorescu. 2009. Src phosphorylation of RhoGDI2 regulates its metastasis suppressor function. Proc. Natl. Acad. Sci. USA. 106: 5807–5812. https://doi.org/10.1073/pnas.0810094106 Yan, Z., Z.G. Wang, N. Segev, S. Hu, R.D. Minshall, R.O. Dull, M. Zhang, A.B. Malik, and G. Hu. 2016. Rab11a Mediates Vascular Endothelial-Cadherin Recycling and Controls Endothelial Barrier Function. Arterioscler. Thromb. Vasc. Biol. 36:339–349. https://doi.org/10.1161/ATVBAHA.115 .306549 Zhu, G., Y. Wang, B. Huang, J. Liang, Y. Ding, A. Xu, and W. Wu. 2012. A Rac1/ PAK1 cascade controls β-catenin activation in colon cancer cells. Oncogene. 31:1001–1012. https://doi.org/10.1038/onc.2011.294 Erasmus et al. Journal of Cell Biology 19 of 19 Rac1 and PAK1 control Rab GTPase cycling https://doi.org/10.1083/jcb.202002114 Downloaded from http://rupress.org/jcb/article-pdf/220/6/e202002114/1821346/jcb_202002114.pdf by Universitaria Sevilla Biblio user on 22 November 2023 Supplemental material Erasmus et al. Journal of Cell Biology S1 Rac1 and PAK1 control Rab GTPase cycling https://doi.org/10.1083/jcb.202002114 Downloaded from http://rupress.org/jcb/article-pdf/220/6/e202002114/1821346/jcb_202002114.pdf by Universitaria Sevilla Biblio user on 22 November 2023 Figure S1. Cellular effects of TAT proteins used in this study. (A) Keratinocytes were left untreated or incubated with constitutively active Rac1 (TATRac1 Q61L ) in the presence or absence of PAK1 autoinhibitory domain (TAT-PAK AID ). Additionally, cells were treated with dominant-negative Rac1 (TATRac1 T17N ), TAT-PAK AID , or TAT peptide by themselves. Cells were fixed and stained for E-cadherin. (B) Quantification of the phenotypes shown in A. Intensity levels of E-cadherin at cell–cell contacts were measured and normalized to the length of each junctions. Nontreated values were arbitrarily set as 100. (C) Keratinocytes were transduced with TAT-Rac1 Q61L or TAT and lysates immunoprecipitated with anti–β-catenin or anti–E-cadherin antibodies. Alternatively, cells were transfected with pRK5myc-Rac1 Q61L (+) or empty vector (−) and immunoprecipitated with anti–E-cadherin antibodies. Precipitated complexes (IP) were probed for the presence of endogenous α-catenin, β-catenin, or E-cadherin. Levels of catenins and cadherin proteins in lysates are shown (input). Statistical significance was analyzed using Student’sttest. Scale bar = 20 µm (n= 3). *, P < 0.0002. Erasmus et al. Journal of Cell Biology S2 Rac1 and PAK1 control Rab GTPase cycling https://doi.org/10.1083/jcb.202002114 Downloaded from http://rupress.org/jcb/article-pdf/220/6/e202002114/1821346/jcb_202002114.pdf by Universitaria Sevilla Biblio user on 22 November 2023 Figure S2. Internalization assays and PAK1 phosphorylation of β-catenin increases its interaction with cadherin tail. (A) Additional replicates of internalization assays shown in Fig. 2. Keratinocytes were treated with cell-permeable TAT or TAT-Rac Q61L in the presence or absence of TAT-PAK AID to inhibit endogenous PAK activation. Following a time course, cells were surface biotinylated and proteins were precipitated with streptavidin to monitor surface levels (left) or internalized (right) levels of E-cadherin and associated catenins. As a control, surface and internalized β1-integrin levels were measured. (B) Peptides identified by mass spectrometry as potential serine phosphorylation sites for PAK1 on β-catenin following in vitro phosphorylation assays. Putative threonine (T) is highlighted in red font and serines (S) are shown in blue font. (C) Identification of PAK1 phosphorylation site. In vitro PAK phosphorylation assay using GST–β-catenin WT and alanine mutants (T551A, S552A, or S675A). (D) Association of E-cadherin cytoplasmic tail with GST-tagged β-catenin WT and mutants (nonphosphorylatable β-cat S675A or phosphomimetic β-cat S675D ) using pull-down assays. (E) Specificity of antibody against phosphorylated β-catenin (pS675). Keratinocyte lysates were incubated in the presence or absence of calf intestinal phosphatase, and Western blots were probed with anti–β-catenin antibodies raised against the phosphorylated form (pS675) or the C-terminal region (C-term; n=3). Erasmus et al. Journal of Cell Biology S3 Rac1 and PAK1 control Rab GTPase cycling https://doi.org/10.1083/jcb.202002114 Downloaded from http://rupress.org/jcb/article-pdf/220/6/e202002114/1821346/jcb_202002114.pdf by Universitaria Sevilla Biblio user on 22 November 2023 Figure S3. Search for putative phosphorylation sites in RabGDIβ.PAK1 is a promiscuous kinase that phosphorylates a variety of cellular substrates. As a PAK1 phosphorylation motif has not been defined on RabGDI proteins, known PAK1 substrates were grouped according to the phospho-site (nine separate groups). Substrate names are shown on the left of each row. RabGDIβsequence was blasted to identify potential homologies. Two putative phosphorylation sites on RabGDIβwere identified at positions S285 and S330 compared with Group 2 (RS) and Group 3 (R-S) substrates. Another potential site was identified at position 382 compared with Group 8 (KY-S). Serine 382 is not found in RabGDIαand thus it is likely the RabGDIβamino acid phosphorylated by PAK1. No other putative phosphorylation sites were identified compared with other groups. Erasmus et al. Journal of Cell Biology S4 Rac1 and PAK1 control Rab GTPase cycling https://doi.org/10.1083/jcb.202002114 Downloaded from http://rupress.org/jcb/article-pdf/220/6/e202002114/1821346/jcb_202002114.pdf by Universitaria Sevilla Biblio user on 22 November 2023 Figure S4. Alignment of RabGDIαand RabGDIβfrom human, bovine, and mouse species. Alignment of full-length amino acid sequences of RabGDIα (GDIA) and RabGDIβ(GDIB) was done using UniProt Align. Sequences from different species are shown with the access number on the left of each row. Conservation of putative phosphorylation sites across different genes is highlighted as follows. Highlighted amino acid residues: Tyrosine (magenta), serine (cyan), or predicted phosphorylation sites by alignment with PAK substrate motifs (gray). Serine 382 is found exclusively in RabGDIβ. See text for more details. Erasmus et al. Journal of Cell Biology S5 Rac1 and PAK1 control Rab GTPase cycling https://doi.org/10.1083/jcb.202002114 Downloaded from http://rupress.org/jcb/article-pdf/220/6/e202002114/1821346/jcb_202002114.pdf by Universitaria Sevilla Biblio user on 22 November 2023 Figure S5. Localization of RabGDIβor Rab11 mutants at junctions does not impair cell–cell contacts. (A) Keratinocytes were transfected EGFP-RabGDIβ WT or mutants nonphosphorylatable (S382A) or phosphomimetic (S382D). After overnight incubation, cells were fixed and confocal images collected. Arrows point to RabGDIβlocalization at junctions. (B) Keratinocytes were transfected with Rab11 siRNA or nontargeting control oligos, fixed, and stained for E-cadherin and nuclei. (C) Keratinocytes were microinjected with constitutively active Rab11 (S20V) or dominant-negative Rab11 (S25N). Cells were fixed, stained for E-cadherin, and imaged in a confocal microscope to detect E-cadherin and the GFP tag. Scale bar = 20 µm (A and B) or 40 µm (C). Erasmus et al. Journal of Cell Biology S6 Rac1 and PAK1 control Rab GTPase cycling https://doi.org/10.1083/jcb.202002114 Downloaded from http://rupress.org/jcb/article-pdf/220/6/e202002114/1821346/jcb_202002114.pdf by Universitaria Sevilla Biblio user on 22 November 2023