Article Scrib Controls Cdc42 Localization and ... - Semantic Scholar

Background: Mammalian Scribble (Scrib) plays a con- served role in polarization of epithelial and neuronal cells. Polarization is essential for migrat...
33 downloads 443 Views 1MB Size
Download PDF
Love PNG Images
Current Biology 16, 2395–2405, December 19, 2006 ª2006 Elsevier Ltd All rights reserved

DOI 10.1016/j.cub.2006.10.026

Article Scrib Controls Cdc42 Localization and Activity to Promote Cell Polarization during Astrocyte Migration Nae¨l Osmani,1,2 Nicolas Vitale,3 Jean-Paul Borg,4,5,6 and Sandrine Etienne-Manneville1,2,* 1 Cell Polarity and Migration Group Institut Pasteur 25 rue du Dr Roux 75724 Paris cedex 15 France 2 Institut Curie Unite´ mixte de recherche 144 Centre National de la Recherche Scientifique 26 rue d’Ulm 75248 Paris cedex 05 France 3 Institut des Neurosciences Cellulaires et Inte´gratives Unite mixte de recherche 7168/LC2 Centre National de la Recherche Scientifique Universite´ Louis Pasteur 5 rue Blaise Pascal 67084 Strasbourg cedex France 4 Inserm, U599 Centre de Recherche en Cance´rologie de Marseille Pharmacologie Mole´culaire F-13009 Marseille France 5 Institut Paoli-Calmettes Marseille F-13009 France 6 Universite´ Me´diterrane´e F-13007 Marseille France

Summary Background: Mammalian Scribble (Scrib) plays a conserved role in polarization of epithelial and neuronal cells. Polarization is essential for migration of a variety of cell types; however, the function of Scrib in this context remains unclear. Scrib has been shown to interact with bPIX, a guanine nucleotide exchange factor for the small GTPases Rac and Cdc42. Cdc42 controls cell polarity from yeast to mammals during asymmetric cell division and epithelial cell polarization, as well as during cell migration. Cdc42 is, in particular, required for polarization and orientation of astrocytes in a scratch-induced polarized migration assay. Using this assay, we characterized Scrib function during polarized cell migration. Results: Depletion of Scrib by siRNA or expression of dominant-negative constructs inhibits astrocyte polarization. Like Cdc42, Scrib controls protrusion formation, cytoskeleton polarization, and centrosome and Golgi reorientation. Scrib interacts and colocalizes with bPIX at the front edge of polarizing astrocytes. Perturbation of

*Correspondence: [email protected]

Scrib localization or of Scrib-bPIX interaction inhibits bPIX polarized recruitment. We further show that bPIX is required for astrocyte polarization and that both the Scrib-binding motif and the GEF activity of bPIX are essential for its function. Scrib and bPIX control Cdc42 activation and localization during astrocyte polarization. Thereby, Scrib regulates Cdc42-dependent APC and Dlg1 recruitment to the leading edge to promote cell orientation. Conclusion: We conclude that Scrib plays a key role in the establishment of cell polarity during migration. By interacting with bPIX, Scrib controls localization and activation of the small GTPase Cdc42 and regulates Cdc42-dependent polarization pathways. Introduction Cell polarity is an essential feature of all eukaryotic cells, both during development and in the adult organism. A polarized cellular organization is critical for cell division, differentiation, and morphogenesis, as well as for a variety of cell functions. Genetic analyses have revealed that several proteins that control cell polarity are conserved in many different contexts. The small Rho GTPase Cdc42, for instance, is involved in the control of polarity throughout evolution from yeast to mammals and regulates asymmetric cell division, epithelial cell differentiation, and directed cell migration [1]. Cdc42 function in cell polarity is, at least in part, mediated by the regulation of the Par6-Par3-atypicalPKC (aPKC) polarity complex. The Par proteins were first identified for their role during the C. elegans zygote first asymmetric division [2]. The Par6/Par3/aPKC complex was later shown to regulate polarization of epithelial and neuronal cells both in Drosophila and in mammals [3]. Other evolutionarily conserved protein complexes, including the Crumbs and Scribble complexes, take an essential part in the polarization process [4, 5]. In Drosophila, Scribble controls apico-basal polarity in epithelial cells, synaptic function, and neuroblast asymmetric divisions [6, 7]. scrib (scribble) has been shown to interact genetically with dlg (discs large) and lgl (lethal giant larvae) in Drosophila melanogaster [8]. Although there is no evidence for a physical interaction between Scrib, Dlg, and Lgl, the three proteins localize at the lateral domain of epithelial cells and form the socalled Scrib complex. The Scrib complex contributes to the formation of the basolateral surface by repressing both the Par and Crumbs complexes and restraining the extension of the apical domain. Scrib encodes a multidomain scaffold protein belonging to the LAP (LRR and PDZ) family, which includes Erbin and Densin-180. Scrib contains 16 leucine-rich repeats (LRR domain), four PDZ (PSD-95, Dlg, ZO-1) domains, and a carboxy-terminal region. The functions of Scrib in epithelial-cell polarization are mediated in part by the LRR domain, which tethers Scrib to the plasma membrane [9]. On the other hand, Scrib PDZ domains have been shown to interact

Current Biology 2396

with ZO2, Zyxin, APC [10–12], and more recently with bPIX/Cool-1, a guanine exchange factor (GEF) for Rac and Cdc42 [13]. However, the role of these interactions in Scrib function is not known yet. In addition to their role in cell polarity, scrib, dlg, and lgl also behave as tumor suppressors and cause massive proliferation in Drosophila imaginal discs when they are mutated (for a review, see [5]). They cooperate with oncogenic Ras in the transformation of Drosophila eye-disc cells and favor metastasis in activated Ras cells [14]. Mammalian Scrib, Lgl, and Dlg share conserved sequences and functions with their Drosophila homologs. Scrib, Lgl, and Dlg localize to the baso-lateral domain of epithelial cells [9, 15, 16]. Lgl and Scrib have been shown to regulate E-cadherin-mediated epithelial cell-cell junctions and contribute to apico-basal polarity [17, 18]. Mammalian Dlg and Scrib may also control cell proliferation [19, 20]. Both proteins are downregulated during malignant progression, and their expression pattern is correlated with loss of cell polarity and tissue architecture in the colon [21]. Additionally, loss of Lgl is correlated with the development of metastases [22]. These observations point to the participation of Scrib, Lgl, and Dlg in a signaling pathway controlling cell polarity and migratory behavior and led us to further investigate the role of these proteins during polarization of migrating cells. A polarized migrating cell is characterized by a protruding front and a retracting rear. During oriented cell migration, the cell polarity axis is oriented in a direction defined by external factors such as a chemoattractant concentration gradient or a physical stimulus. In socalled wound-healing assays, scratching the cell monolayer induces cell polarization and migration in a direction perpendicular to the wound. In the case of primary astrocytes, polarization corresponds to an elongation of the cell, which forms a protrusion in the direction of migration. The orientation of this polarity axis can be detected by (i) the direction of cell protrusion and cell migration, (ii) the orientation of the microtubule and actin networks, and (iii) the position of the centrosome and the Golgi apparatus relative to that of the nucleus. Using this assay, we previously demonstrated that, after wounding, the Rho GTPase Cdc42 was activated and recruited to the leading edge of the cells [23]. Localized Cdc42 activity results in a spatially confined signaling, which controls cell polarization and cell orientation. Active Cdc42 recruits to the leading edge the polarity complex formed by Par6 and the atypical PKC PKCz and leads to the clustering of Adenomatous Polyposis Coli (APC) at microtubule plus ends [24]. The Cdc42-dependent Par6-aPKC pathway also controls the polarized recruitment of Dlg to the leading edge. Dlg, together with APC, controls microtubule-network polarization and centrosome positioning [25]. The mechanisms controlling Cdc42 recruitment and activation remain unclear. In this article, we investigate the function of Scrib during polarization of migrating cells by using the astrocyte wound-healing assay and show that Scrib is essential for cell polarization and cell orientation. We find that Scrib controls the recruitment of bPIX to the leading edge of migrating cells and consequently regulates Cdc42 localization and activity to promote cell polarization.

Figure 1. Scrib Plays a Critical Role in Astrocyte Polarization Primary astrocytes were nucleofected with control siRNA (CTL) or siRNAs directed against two different Scrib sequences (Scrib-1, Scrib-2) and incubated for 72 hr. (A) Sixteen hours after wounding of the cell monolayer, cells were fixed and stained with phalloidin (upper panels) and anti-tubulin antibodies (lower panels). (B) Eight hours after wounding, cells were fixed and stained with anti-pericentrin (centrosome, red), anti-GM130 (Golgi apparatus, yellow), and Hoechst (nucleus, blue). White dotted lines indicate the scratch orientation. The scale bar represents 10 mm. (C) Percentage of wound-edge cells having their centrosome in the quadrant that is in front of the nucleus and facing the wound (light blue on the diagram). Random orientation of the centrosome with respect to the wound edge corresponds to a value of 25%. Results shown are means 6 SEM of 3–5 independent experiments, for a total of at least 500 cells being scored.

Results Scrib Controls Astrocyte Polarization and Orientation We used two siRNAs directed against distinct Scrib sequences to reduce Scrib expression in primary astrocytes (Figure S1A). A scratch in a control astrocyte monolayer induces the formation of long protrusions filled with an elongated microtubule network and actin fibers aligned along the protrusion axis (Figure 1A, left panel), as previously reported [23]. Depletion of

Scrib in Polarized Cell Migration 2397

Figure 2. Role of Scrib Domains in Astrocyte Polarization After scratching, cells of the first row were microinjected with the indicated DNA constructs (pEGFP as a control). (A) Scrib constructs used in this study. (B) Sixteen hours after wounding of the cell monolayer, cells were fixed. Expressing cells were scored as protruding when they formed protrusions at least three times longer than wide. Representative images are shown in the right panels. (C) Cells were fixed and stained with antipericentrin (centrosome, red) and Hoechst (nucleus, blue). Centrosome reorientation was assessed in expressing cells of the first row. Microinjected cells are marked with a star (right panels). Expression of the construct is shown in the insert. Results shown are means 6 SEM of 3–5 independent experiments, with a total of at least 200 cells being scored. White dotted lines show scratch positions. The scale bar represents 10 mm.

endogenous Scrib inhibits protrusion formation, and Scrib-depleted cells show a short and disorganized microtubule network and nonpolarized actin structures (Figure 1A, right panels). Furthermore, depletion of Scrib significantly perturbs scratch-induced centrosome and Golgi reorientation (Figures 1B and 1C; also Figure S1C). These results indicate that Scrib is involved in both polarization and orientation of migrating astrocytes. Scrib contains an LRR domain, four PDZ domains, and noncharacterized central and carboxy-terminal domains. The PDZ domains have been shown to mediate Scrib interaction with several proteins, including the

Cdc42 and Rac guanine exchange factor bPIX [13]. Various Scrib constructs were microinjected in astrocytes of the wound edge (Figure 2A). Overexpression of Scrib WT, which localizes to multiple sites on the plasma membrane and in the cytoplasm (Figure 2B, right panel), evokes multiple, randomly oriented protrusions associated with a random centrosome orientation (Figures 2B and 2C). Scrib-induced multiple protrusions are inhibited by coexpression of a dominant-negative Cdc42 (+ N17 Cdc42). Similar results are obtained after expression of Scrib with a point mutation (Scrib PL) that disrupts the LRR structure and delocalizes Scrib from the

Current Biology 2398

Figure 3. Scrib Controls bPIX Recruitment to the Leading Edge of Migrating Astrocytes (A and B) Cells were fixed at the indicated time after wounding. Cells were stained with anti-Scrib (A) or anti-bPIX (B) antibodies. The percentage of front-row cells showing Scrib or bPIX recruitment to the leadingedge plasma membrane was scored. Results shown are means 6 SEM of 3–5 independent experiments, with a total of at least 500 cells being scored. Representative images are shown on the right. White dotted lines show scratch positions. The scale bar represents 10 mm. (C) Cells were fixed 4 hr after wounding and stained with Scrib and bPIX antibodies. The lower panel shows the merged image (Scrib in red, bPIX in green).The scale bar represents 10 mm. (D) Cells were lysed at the indicated time after wounding. Immunoprecipitations with antiScrib or irrelevant (CTL) antibodies were performed and analyzed by western blotting with Scrib (lower panels) or bPIX antibodies (upper panels). Blots shown are representative of three independent experiments. (E and F) Cells were nucleofected with the indicated siRNAs (E) or microinjected with the indicated constructs (F). Cells were scratched, fixed 4 hr later, and stained with anti-bPIX. The percentage of cells in the front row showing bPIX recruitment at the leadingedge plasma membrane was scored. Representative images are shown on the right. Microinjected cells are marked with a star (right panels). Expression of the construct is shown in the insert. White dotted lines show scratch positions. The scale bar represents 10 mm. (G) Four hours after scratching, cells were fixed and stained with Scrib antibodies. The percentage of front-rpw cells showing Scrib recruitment to the leading-edge plasma membrane was scored. Results shown are means 6 SEM of 3–5 independent experiments, with a total of 200–500 cells being scored.

plasma membrane without affecting its interaction with bPIX [9]. Expression of truncated Scrib forms missing the four PDZ domains (Scrib DPDZ, Scrib PL DPDZ) or of constructs only encoding the four PDZ domains (PDZ) or the carboxy-terminal domain (Cter) strongly inhibits both cell protrusion and centrosome orientation. This phenotype is similar to that observed after expression of a dominant-negative form of Cdc42 (N17-Cdc42) but differs from that induced by a dominant-negative form of Rac (N17-Rac) [23]; the latter inhibits protrusion formation but not centrosome orientation (Figures 2B and 2C). In contrast, expression of a construct lacking both the PDZ and the carboxy-terminal domain (Scrib PL Nter) has no effect. Taken together, these results suggest that mislocalization of Scrib perturbs cell polarization and that PDZ and carboxy-terminal domains of Scrib are required for Scrib functions. They also suggest that Scrib may act in the same signaling pathway as Cdc42 to control astrocyte polarization and orientation. This led us to

investigate the localization of Scrib and its interaction with bPIX during astrocyte polarization. Scrib Is Recruited with bPIX to the Leading Edge of Migrating Cells In confluent nonmigrating astrocytes, Scrib localizes at cell-cell contacts and in the cytoplasm (data not shown). Immediately after scratching the monolayer, Scrib cannot be detected at the wound edge. However, as soon as 30 min after scratching, Scrib is recruited to the leading edge of front row of cells (Figure 3A). bPIX is cytoplasmic in confluent cells and between 30 min and 4 hr after wounding progressively appears at the cell leading edge (Figure 3B), where it colocalizes with Scrib (Figure 3C). Four hours after scratching, bPIX also localizes in focal adhesion-like structures, as previously reported [26]. In contrast, Scrib is not visible in these structures (Figure 3C). Immunoprecipitation of Scrib showed that bPIX constitutively interacts with Scrib in astrocytes (Figure 3D). This interaction increases by a factor

Scrib in Polarized Cell Migration 2399

4 6 0.7 (mean 6 SEM of three independent experiments) between 1 and 4 hr after wounding. Because Scrib and bPIX are recruited together to the leading edge of migrating cells, we investigated the role of Scrib in bPIX localization. Depletion of Scrib, or expression of Scrib PDZ domains, which perturbs the interaction between Scrib and bPIX, strongly inhibits bPIX recruitment to the leading edge (Figures 3E and 3F). Overexpression of Scrib-WT, Scrib PL, or Scrib DPDZ, which cannot bind bPIX, also perturbs bPIX recruitment to the leading edge (Figure 3F). bPIX depletion does not affect Scrib recruitment (Figure 3G). These observations suggest that Scrib is involved in a multimolecular complex required for bPIX localization through various interaction domains. Taken together, these results show that, after the scratching of the cell monolayer, Scrib is recruited to the leading edge and controls the localization of its interacting partner bPIX. bPIX Is Required for Astrocyte Polarization and Orientation Two distinct siRNAs were used to deplete bPIX expression. Although one of the siRNAs (siRNA bPIX-1) almost totally inhibits bPIX expression, the other (siRNA bPIX-2) reduces bPIX expression to a lower extent (Figure S1B). In both cases, bPIX depletion inhibits protrusion formation, elongation of the microtubule network, and polarization of the actin cytoskeleton (Figure 4A). bPIX is also required for centrosome and Golgi apparatus reorientation after wounding (Figure 4B), the level of inhibition correlating with the protein-expression level (Figure 4C; also Figure S1C). These observations show an essential role of bPIX during scratch-induced astrocyte polarization. We used different constructs to further investigate the role of bPIX activity in astrocyte polarization (Figure 4D). Overexpression of bPIX (WT) has no effect on astrocyte polarization (Figures 4E and 4F). In contrast, expression of a truncated bPIX form lacking the three carboxy-terminal amino acids involved in PDZ-mediated Scrib interaction (DTNL) [13] strongly inhibits protrusion formation and centrosome reorientation. A similar phenotype is obtained after expression of the last 15 amino acids including the carboxy-terminal PDZ binding domain of bPIX (bPIX Cter) [13]. Astrocyte polarization was not modified by expression of a control protein (bPIX Cter DTNL) that encompasses the last 15 residues of bPIX without the PDZ binding site and that has no affinity for Scrib. Finally, expression of an inactive form of bPIX lacking the Dbl (diffuse B-cell lymphoma) homology domain (DDH) inhibits both protrusion formation and centrosome orientation in migrating astrocytes to a similar extent as dominant-negative Cdc42 (Figures 4E and 4F). Taken together, these results show that bPIX, like Scrib, is involved in the regulation of cell protrusion and cell orientation during scratch-induced astrocyte polarization and point to an essential role of bPIX guanine-exchange activity. Scrib and bPIX Control Cdc42 Recruitment to the Leading Edge, Cdc42 Activity, and the Cdc42-Dependent Polarity Pathway We have previously shown that scratching an astrocyte monolayer promotes Cdc42 recruitment and activation to the leading edge and that both localization and

activity of Cdc42 are essential for astrocyte polarization [23]. Cdc42 recruitment to the leading edge was visualized with a GFP-tagged construct previously shown to localize similarly to the endogenous protein [27] (Figure 5A). Using confocal imaging and membrane staining with a lipophilic dye, we have verified that increased GFP signal at the leading edge is not due to membrane ruffling (Figure S2). Cdc42 recruitment is dramatically reduced in Scrib-depleted cells and in siRNA bPIX-1 nucleofected cells. The less efficient siRNA bPIX-2 only partially reduces Cdc42 recruitment (Figure 5B). As shown by a PAK-CRIB pulldown assay, depletion of Scrib and bPIX also strongly reduces wound-induced Cdc42 activation, whereas control siRNA has no effect (Figures 5C and 5D). Finally, to confirm the inhibition of Cdc42 activity, we analyzed the effects of Scrib and bPIX depletion on Cdc42-dependent polarity pathways. We have previously reported that Cdc42 activity is required for adenomatous polyposis coli (APC) clustering at microtubule plus ends and for Dlg1 recruitment to the leading edge, both events being required for centrosome reorientation [25]. Depletion of Scrib or bPIX with siRNA strongly impaired APC clustering (Figure 6A) and Dlg1 recruitment (Figure 6B). Interestingly, bPIX depletion inhibits Dlg1 localization without perturbing Scrib recruitment to the leading edge (Figures 6B and 3G). Taken together, these results demonstrate that Scrib and bPIX are required for Cdc42 localization, for Cdc42 activation, and for downstream events in the Cdc42-dependent polarity pathway. Discussion Using dominant-negative constructs and specific siRNAs, we demonstrate here that Scrib is essential for polarization and correct orientation of migrating cells. scrib has been characterized as a tumor-suppressor gene, and its expression decreases with malignant progression [21]. As reported in epithelial cells [17], the orientation more than the speed of migration was dramatically perturbed (Figure 1 and data not shown). Similar effects were observed after depletion of APC and Dlg in astrocytes [25], suggesting that these tumor suppressors contribute to cell migration essentially by controlling orientation rather than speed. We might speculate that decreased expression of these genes will result in random and uncontrolled cell migration. Scrib plays a dual role in cell polarity by regulating Cdc42 activation and Cdc42 recruitment to the leading edge. Scrib PDZ and carboxy-terminal domains play a key role in Scrib polarity function. Although we do not know the nature of the carboxy-terminal binding partner, it is tempting to speculate that the role of the PDZ domain is to bind the GEF bPIX. bPIX forms a functional link between two major polarity proteins, Scrib and Cdc42. Accordingly, the TNL carboxy-terminal residues and the exchange activity of bPIX are essential (Figure 4). bPIX has been described as a potential GEF for Cdc42 and Rac [28]. In astrocytes, bPIX depletion inhibits Cdc42 activation upon wounding (Figure 5C), and expression of bPIX DDH impairs centrosome and Golgi reorientation (Figure 5C and data not shown), which are

Current Biology 2400

Figure 4. bPIX Is Involved in Astrocyte Polarization Primary astrocytes were nucleofected with the indicated siRNAs or microinjected with the indicated DNA constructs. (A) Cytoskeleton polarization. Sixteen hours after wounding of the cell monolayer, cells were fixed and stained with tubulin antibodies (lower panels) and phalloidin (upper panel). (B) Eight hours after wounding, cells were fixed and stained with anti-pericentrin (centrosome in red), anti-GM130 (Golgi in yellow), and Hoechst (nucleus in blue). White dotted lines indicate the scratch orientation. The scale bar represents 10 mm. (C) Centrosome orientation. Results shown are the mean 6 SEM of 3–5 independent experiments, with a total of at least 500 cells being scored. (D) bPIX constructs used in this study. (E) Protrusion formation. Sixteen hours after wounding of the cell monolayer, cells were fixed. Expressing cells were scored as protruding when they formed protrusions at least three times longer than wide (labeled cells in the right panels). (F) Centrosome reorientation was assessed in expressing cells of the first row. Cells were fixed and stained with anti-pericentrin (centrosome, red) and Hoechst (nucleus, blue). Microinjected cells are marked with a star (right panels). Expression of the construct is shown in the insert. Results shown are means 6 SEM of 3-5 independent experiments, with a total of at least 200 cells being scored. Representative images are shown in the right panels. White dotted lines show scratch positions. Tje scale bar represents 10 mm.

Cdc42- but not Rac-dependent events. Moreover, we have previously shown that Cdc42 activation is dependent on Src-like tyrosine kinase activity and integrin engagement [23]. bPIX activity toward Cdc42 has been recently shown to be regulated by Src phosphorylation [29]. It is therefore tempting to speculate that integrins and Src kinases are involved in Scrib and bPIX activation

or recruitment to the leading edge, or both. Taken together, our results strongly support the idea that bPIX is directly responsible for Cdc42 activation during astrocyte polarization. Our observations differ from the previous finding that bPIX is not involved in centrosome reorientation in migrating fibroblasts [30]. Brain-specific isoforms of bPIX, bPIX-b(L) and b2PIX, might account

Scrib in Polarized Cell Migration 2401

Figure 5. Scrib and bPIX Regulate Scratch-Induced Cdc42 Recruitment and Activation (A) pEGFP or pEGFP-Cdc42 constructs were microinjected in confluent nonmigrating astrocytes or in cells just after wounding. Eight hours later, cells were fixed, and the percentage of front-row cells showing an increased GFP fluorescence at the leading-edge plasma membrane was scored (left panel). (B) Primary astrocytes were nucleofected with the indicated siRNAs. Seventy-two hours later, the cell monolayer was scratched, and woundedge cells were microinjected with pEGFP-Cdc42. Cells were fixed 8 hr later, and the percentage of front-row cells showing an increased GFP fluorescence at the leading-edge plasma membrane was scored (left panel). Results shown are means 6 SEM of 3–5 independent experiments, with a total of at least 200 cells being scored. Representative images are shown on the right. White dotted lines show scratch positions. The scale bar represents 10 mm. (C) Astrocytes were nucleofected with control (CTL), Scrib (Scrib-1; Scrib-2), or bPIX (bPIX-1; bPIX-2) siRNAs. Cells were lysed just after (0) or 30 min (30’) after scratching. Lysates were incubated with agarose beads coupled to GST-PAK-CRIB (GST-PAK). Affinity-purified GTP-bound Cdc42 (top panel) and total Cdc42 (lower panel) were analyzed by western blotting. Blots shown are representative of 3–5 experiments. (D) Quantification of wound-induced Cdc42 activation in control and siRNA-treated cells. Results are expressed as Cdc42 activity 30 min after wounding over basal Cdc42 activity. In each condition, basal activity was defined as the GTP-Cdc42 level present in just-wounded astrocytes. Results shown are means 6 SEM of 3–5 independent experiments.

for these differences [31, 32]. However, we could not detect these proteins by western blotting. Alternatively, the exact nature of integrins engaged during cell polarization, the level of Scrib expression, or both may account for cell specificity. bPIX, but not Scrib, can also be observed at focal adhesions, where it may contribute to their turnover (Figure 3C and [26]). Moreover, only a minor fraction of bPIX was found associated with Scrib (data not shown), suggesting that bPIX is associated with different complexes that play distinct roles during cell migration. For instance, bPIX may be involved in Rac regulation during astrocyte migration. Rac and PAK are both involved in astrocyte-protrusion formation [23]. Moreover, they both directly interact with bPIX via its SH3 domain [33, 34]. In agreement with the model recently proposed by ten Klooster and collegues, bPIX may activate Cdc42

to induce PAK phosphorylation, as well as recruitment and activation of Rac [34]. Activation and recruitment of Cdc42 appear to be two separate events. Scrib overexpression, which inhibits bPIX and Cdc42 recruitment to the leading edge, does not seem to inhibit scratch-induced activation of Cdc42. This is supported by the fact that Scrib overexpression does not inhibit protrusion formation but rather induces random Cdc42-dependent cell protrusions (Figure 2B). In these conditions, APC forms clusters at microtubule plus ends at the tip of these multiple protrusions (data not shown), suggesting that Cdc42-dependent polarity pathway is activated but not properly localized [24]. Overexpression of Scrib WT may dissociate a multi-molecular complex by titrating a front-edge membrane-associated protein required for the localization of bPIX and Cdc42 but not for Cdc42 activation. The

Current Biology 2402

Figure 6. Scrib and bPIX Regulate APC and Dlg1 Recruitment to the Leading Edge Astrocytes were nucleofected with control (CTL), Scrib (Scrib-1; Scrib-2), or bPIX (bPIX1; bPIX-2) siRNAs. (A) Cells were fixed 8 hr after wounding and stained with tubulin (green) and APC (red) antibodies (right panels). The percentage of front-row cells showing APC clustering at the leading edge was scored. Representative images are shown in the right panels. The scale bar represents 10 mm. Enlargements of boxed regions showing microtubule plus ends at the leading edge are shown on the right. The scale bar represents 5 mm. (B) Cells were fixed 6 hr after wounding and stained with Dlg1 antibodies (right panels). The percentage of front-row cells showing Dlg1 enrichment at the leading edge was scored. Results shown are means 6 SEM of 3–5 independent experiments, with a total of at least 500 cells being scored. Representative images are shown in the right panels. The scale bar represents 5 mm. White dotted lines show scratch positions.

LRR domain of Scrib has been shown to be essential for Scrib localization at the plasma membrane in epithelial cells [9]. Consistent with this result, expression of Scrib PL in migrating astrocytes also promotes the formation of multiple disoriented Cdc42-dependent protrusions (Figure 2B). Determining which molecule recruits Scrib to the plasma membrane, probably via the LRR domain, will be the goal of further investigations. Our results suggest that Cdc42 activation first occurs at various cellular locations and that active Cdc42 is then recruited to the leading edge. The kinetics of Scrib, bPIX, and Cdc42 recruitment to the leading edge do not exactly parallel those of Cdc42 activation, which is initiated within 5 min after wounding and reaches a maximum at 1 hr (Figure 3 and [23]). GFP-Cdc42 localizes not only at the plasma membrane but also on the Golgi apparatus and on numerous vesicular structures

dispersed throughout the cytosol (Figure 5A and data not shown). In the absence of leading-edge recruitment, active Cdc42 may localize to any of these structures. bPIX has been involved in the recycling of integrins, Rac, and PAK via the ARF GTPase activating protein GIT1 [26, 35, 36]; it may also be involved in Cdc42 trafficking and delivery to plasma-membrane sites. Together with previous observations showing the role of Scrib in epithelial and neuronal cell polarization as well as in planar polarity [17, 37, 38], our findings strengthen the idea that Scrib is a major and conserved regulator of cell polarity. Although cell polarization requires conserved proteins, how these proteins interact seems to depend on the cellular context [1]. As in epithelial and neuronal cells [6, 37], Scrib and Dlg1 both localize in the same region of the plasma membrane. However, in contrast with Drosophila, where scrib, dlg, and

Scrib in Polarized Cell Migration 2403

Proligo. SiRNAs directed against the GFP sequence or corresponding to a random nucleotide sequence were used as control siRNAs. SiRNAs were introduced into cells by nucleofection according to the vendor’s instructions (Amaxa GmbH). Via this technique, astrocyte transfection efficiency reached approximately 80%. Cell Culture and Scratch-Induced Migration Primary rat astrocytes were prepared as described previously [23, 39]. For scratch-induced assays, cells were seeded on poly-L-ornithine-coated coverslips or 90 mm diameter dishes and were grown in serum to confluence. The medium was changed 16 hr before scratching. Individual wounds (approximately 300 mm wide) were made with a microinjection needle. For biochemical analysis, multiple wounds were made with an eight-channel pipette (0.1–2 ml tips). In all cases, wounds were closed in about 24 hr. Nuclear microinjections were performed in wound-edge cells in the first 30 min after scratching. Expression vectors were used at 100–200 mg/ml. Protein expression was detectable about 1 hr after microinjection.

Figure 7. Diagram Showing Scrib and bPIX Functions during Astrocyte Polarization After wounding, Scrib is recruited to the leading edge. Scrib interacts with bPIX and controls bPIX localization. In turn, bPIX promotes Cdc42 activation and localization at the leading edge. Cdc42 controls two distinct signaling pathways, promoting (1) Rac- and PAKdependent protrusion formation and (2) centrosome and Golgi reorientation through APC clustering and Dlg1 localization at the leading edge.

lgl produce very similar phenotypes [6], Scrib is involved in both generation of a polarity axis (elongated cell morphology) and orientation of this polarity axis toward the wound, whereas Dlg1 only regulates orientation [25]. Moreover, Scrib depletion inhibits Dlg1 recruitment to the leading edge, whereas Dlg1 depletion has no effect on Scrib localization (Figure 6B, data not shown), suggesting that Scrib lies upstream of Dlg1 in the signaling cascade (Figure 7). Conclusions We have shown here that Scrib is a critical player in the polarization of migrating cells, confirming the conserved role of Scrib in the control of cell polarity. Scrib, through its association with the exchange factor bPIX, controls Cdc42, another essential polarity protein. By regulating Cdc42 activity, Scrib acts upstream of Dlg1 and is involved in the same molecular pathway controlling cell orientation. Experimental Procedures Materials Scrib antibody was from Santa Cruz Biotechnology, anti-bPIX from Chemicon, anti-atubulin from Sigma-Aldrich, anti-Dlg1 from Transduction Laboratories, phalloidin-rhodamine from Molecular Probes, and anti-pericentrin from BabCO. Anti-APC was a gift from I. Na¨thke (University of Dundee, Dundee, Scotland, UK). Secondary antibodies were all obtained from Jackson ImmunoResearch Laboratories. The DiI lipophilic dye was from Molecular Probes. DNA Constructs and siRNAs Cdc42, Rac, Scrib, and bPIX constructs have been previously described [13, 23]. Two siRNA duplexes corresponding to rat Scrib (GenBank/EMBL/DDBJ accession no. XM_343266) starting at nucleotides 949 (siRNA Scrib-1) and 2525 (siRNA Scrib-2) and to rat bPIX (GenBank/EMBL/DDBJ accession no. AF044673) starting at nucleotides 247 (siRNA bPIX-1) and 741 (siRNA bPIX-2) were obtained from

Immunofluorescence and Image Quantification Cells were stained as described previously [23]. Epifluorescence images of fixed cells mounted in mowiol were obtained on a microscope (model DM6000 Leica) equipped with a 403 NA 1.25 and a 633 NA 1.4 objective lens and were recorded on a CCD camera (CoolSNAP HQ; Roper Scientific) with MetaMorph software (Universal Imaging Corp.). Cell-Polarization Assays Protrusion Formation Protrusion formation was determined 16 hr after wounding following fixation and microtubule staining. Images of cells transfected with siRNAs were taken randomly along the wound. For quantification of protrusion formation, cells of the wound edge were microinjected with the indicated constructs just after wounding. Expressing cells, revealed with the appropriate staining, were scored as protruding if they displayed large membrane protrusion(s) at least three times longer than wide. Centrosome Positioning Centrosome reorientation was determined as described previously [23, 39]. Centrosomes located in front of the nucleus, within the quadrant facing the wound, were scored as correctly oriented. In this assay, a score of 25% represents the absolute minimum corresponding to random centrosome positioning. Protein Localization. Cdc42 localization. Microinjection of GFP-tagged Cdc42 WT was used for visualization of Cdc42 because of the lack of anti-Cdc42 antibodies that work in immunofluorescence. The percentage of cells showing an increased GFP fluorescence at the leading edge was determined 8 hr after wounding. Similar results were obtained 4 hr after wounding (data not shown). Scrib, bPIX, or Dlg1 localization. Cells were fixed and stained with the appropriate antibody. For the scoring of polarized protein recruitment, cells in which the protein was specifically enriched at the leading edge were counted as positive. APC localization. Cells were stained with APC and tubulin antibodies. The percentage of cells showing APC clusters at microtubule tips of the leading edge was measured. In general, only very few migrating astrocytes show membrane ruffling at the leading edge. However, in order to avoid artifact from increased fluorescence due to membrane ruffling, these cells were not included in the analysis. Immunoprecipitation and Cdc42-Activity Assay Experiments were performed as previously described [23]. Image quantification was performed with the ImageJ software. Intensity of the Cdc42-GTP band was normalized by the intensity of the corresponding amount of total Cdc42. Results were expressed as activation over basal Cdc42 activity observed in just-wounded, nonmigrating astrocytes. Supplemental Data Supplemental Data include additional Experimental Procedures and two figures and are available online at http://www.current-biology. com/cgi/content/full/16/24/2395/DC1/.

Current Biology 2404

Acknowledgments This work was supported by the Centre National de la Recherche Scientifique, the Pasteur Institute, the Agence Nationale pour la Recherche, the Fondation de France, the Institut pour la Recherche sur la Moelle e´pinie`re et l’Ence´phale and the Association pour la Recherche contre le Cancer. We would like to thank Prof. Daniel Louvard and Dr. Monique Arpin for their support and the PlateForme d’Imagerie Dynamique imaging facilitiy for technical help. N.O. is funded by the Ministe`re de la Recherche. Received: July 28, 2006 Revised: September 26, 2006 Accepted: October 12, 2006 Published online: November 2, 2006 References 1. Etienne-Manneville, S. (2004). Cdc42—The centre of polarity. J. Cell Sci. 117, 1291–1300. 2. Nance, J. (2005). PAR proteins and the establishment of cell polarity during C. elegans development. Bioessays 27, 126–135. 3. Macara, I.G. (2004). Par proteins: Partners in polarization. Curr. Biol. 14, R160–R162. 4. Medina, E., Lemmers, C., Lane-Guermonprez, L., and Le Bivic, A. (2002). Role of the Crumbs complex in the regulation of junction formation in Drosophila and mammalian epithelial cells. Biol. Cell. 94, 305–313. 5. Humbert, P., Russell, S., and Richardson, H. (2003). Dlg, Scribble and Lgl in cell polarity, cell proliferation and cancer. Bioessays 25, 542–553. 6. Bilder, D., Li, M., and Perrimon, N. (2000). Cooperative regulation of cell polarity and growth by Drosophila tumor suppressors. Science 289, 113–116. 7. Albertson, R., and Doe, C.Q. (2003). Dlg, Scrib and Lgl regulate neuroblast cell size and mitotic spindle asymmetry. Nat. Cell Biol. 5, 166–170. 8. Tanentzapf, G., and Tepass, U. (2003). Interactions between the crumbs, lethal giant larvae and bazooka pathways in epithelial polarization. Nat. Cell Biol. 5, 46–52. 9. Navarro, C., Nola, S., Audebert, S., Santoni, M.J., Arsanto, J.P., Ginestier, C., Marchetto, S., Jacquemier, J., Isnardon, D., Le Bivic, A., et al. (2005). Junctional recruitment of mammalian Scribble relies on E-cadherin engagement. Oncogene 24, 4330–4339. 10. Petit, M.M., Crombez, K.R., Vervenne, H.B., Weyns, N., and Van de Ven, W.J. (2005). The tumor suppressor Scrib selectively interacts with specific members of the zyxin family of proteins. FEBS Lett. 579, 5061–5068. 11. Metais, J.Y., Navarro, C., Santoni, M.J., Audebert, S., and Borg, J.P. (2005). hScrib interacts with ZO-2 at the cell-cell junctions of epithelial cells. FEBS Lett. 579, 3725–3730. 12. Takizawa, S., Nagasaka, K., Nakagawa, S., Yano, T., Nakagawa, K., Yasugi, T., Takeuchi, T., Kanda, T., Huibregtse, J.M., Akiyama, T., et al. (2006). Human scribble, a novel tumor suppressor identified as a target of high-risk HPV E6 for ubiquitinmediated degradation, interacts with adenomatous polyposis coli. Genes Cells 11, 453–464. 13. Audebert, S., Navarro, C., Nourry, C., Chasserot-Golaz, S., Lecine, P., Bellaiche, Y., Dupont, J.L., Premont, R.T., Sempere, C., Strub, J.M., et al. (2004). Mammalian Scribble forms a tight complex with the betaPIX exchange factor. Curr. Biol. 14, 987–995. 14. Tapon, N. (2003). Modeling transformation and metastasis in Drosophila. Cancer Cell 4, 333–335. 15. Musch, A., Cohen, D., Yeaman, C., Nelson, W.J., RodriguezBoulan, E., and Brennwald, P.J. (2002). Mammalian homolog of Drosophila tumor suppressor lethal (2) giant larvae interacts with basolateral exocytic machinery in Madin-Darby canine kidney cells. Mol. Biol. Cell 13, 158–168. 16. Muller, B.M., Kistner, U., Veh, R.W., Cases-Langhoff, C., Becker, B., Gundelfinger, E.D., and Garner, C.C. (1995). Molecular characterization and spatial distribution of SAP97, a novel presynaptic protein homologous to SAP90 and the Drosophila discs-large tumor suppressor protein. J. Neurosci. 15, 2354–2366.

17. Qin, Y., Capaldo, C., Gumbiner, B.M., and Macara, I.G. (2005). The mammalian Scribble polarity protein regulates epithelial cell adhesion and migration through E-cadherin. J. Cell Biol. 171, 1061–1071. 18. Yamanaka, T., Horikoshi, Y., Izumi, N., Suzuki, A., Mizuno, K., and Ohno, S. (2006). Lgl mediates apical domain disassembly by suppressing the PAR-3-aPKC-PAR-6 complex to orient apical membrane polarity. J. Cell Sci. 119, 2107–2118. 19. Bilder, D. (2004). Epithelial polarity and proliferation control: Links from the Drosophila neoplastic tumor suppressors. Genes Dev. 18, 1909–1925. 20. Dow, L.E., Brumby, A.M., Muratore, R., Coombe, M.L., Sedelies, K.A., Trapani, J.A., Russell, S.M., Richardson, H.E., and Humbert, P.O. (2003). hScrib is a functional homologue of the Drosophila tumour suppressor Scribble. Oncogene 22, 9225–9230. 21. Gardiol, D., Zacchi, A., Petrera, F., Stanta, G., and Banks, L. (2006). Human discs large and scrib are localized at the same regions in colon mucosa and changes in their expression patterns are correlated with loss of tissue architecture during malignant progression. Int. J. Cancer 119, 1285–1290. 22. Schimanski, C.C., Schmitz, G., Kashyap, A., Bosserhoff, A.K., Bataille, F., Schafer, S.C., Lehr, H.A., Berger, M.R., Galle, P.R., Strand, S., et al. (2005). Reduced expression of Hugl-1, the human homologue of Drosophila tumour suppressor gene lgl, contributes to progression of colorectal cancer. Oncogene 24, 3100–3109. 23. Etienne-Manneville, S., and Hall, A. (2001). Integrin-mediated Cdc42 activation controls cell polarity in migrating astrocytes through PKCz. Cell 106, 489–498. 24. Etienne-Manneville, S., and Hall, A. (2003). Cdc42 regulates GSK3 and adenomatous polyposis coli (APC) to control cell polarity. Nature 421, 753–756. 25. Etienne-Manneville, S., Manneville, J.B., Nicholls, S., Ferenczi, M., and Hall, A. (2005). Cdc42 and Par6-PKCz regulate the spatially localized association of Dlg1 and APC to control cell polarization. J. Cell Biol. 170, 895–901. 26. Nayal, A., Webb, D.J., Brown, C.M., Schaefer, E.M., VicenteManzanares, M., and Horwitz, A.R. (2006). Paxillin phosphorylation at Ser273 localizes a GIT1-PIX-PAK complex and regulates adhesion and protrusion dynamics. J. Cell Biol. 173, 587–589. 27. Michaelson, D., Silletti, J., Murphy, G., D’Eustachio, P., Rush, M., and Philips, M.R. (2001). Differential localization of Rho GTPases in live cells: Regulation by hypervariable regions and RhoGDI binding. J. Cell Biol. 152, 111–126. 28. Manser, E., Loo, T.H., Koh, C.G., Zhao, Z.S., Chen, X.Q., Tan, L., Tan, I., Leung, T., and Lim, L. (1998). PAK kinases are directly coupled to the PIX family of nucleotide exchange factors. Mol. Cell 1, 183–192. 29. Feng, Q., Baird, D., Peng, X., Wang, J., Ly, T., Guan, J.L., and Cerione, R.A. (2006). Cool-1 functions as an essential regulatory node for EGF receptor- and Src-mediated cell growth. Nat. Cell Biol. 8, 945–956. 30. Cau, J., and Hall, A. (2005). Cdc42 controls the polarity of the actin and microtubule cytoskeletons through two distinct signal transduction pathways. J. Cell Sci. 118, 2579–2587. 31. Rhee, S., Yang, S.J., Lee, S.J., and Park, D. (2004). betaPix-b(L), a novel isoform of betaPix, is generated by alternative translation. Biochem. Biophys. Res. Commun. 318, 415–421. 32. Koh, C.G., Manser, E., Zhao, Z.S., Ng, C.P., and Lim, L. (2001). Beta1PIX, the PAK-interacting exchange factor, requires localization via a coiled-coil region to promote microvillus-like structures and membrane ruffles. J. Cell Sci. 114, 4239–4251. 33. Mott, H.R., Nietlispach, D., Evetts, K.A., and Owen, D. (2005). Structural analysis of the SH3 domain of beta-PIX and its interaction with alpha-p21 activated kinase (PAK). Biochemistry 44, 10977–10983. 34. ten Klooster, J.P., Jaffer, Z.M., Chernoff, J., and Hordijk, P.L. (2006). Targeting and activation of Rac1 are mediated by the exchange factor beta-Pix. J. Cell Biol. 172, 759–769. 35. Di Cesare, A., Paris, S., Albertinazzi, C., Dariozzi, S., Andersen, J., Mann, M., Longhi, R., and de Curtis, I. (2000). p95-APP1 links membrane transport to Rac-mediated reorganization of actin. Nat. Cell Biol. 2, 521–530.

Scrib in Polarized Cell Migration 2405

36. Matafora, V., Paris, S., Dariozzi, S., and de Curtis, I. (2001). Molecular mechanisms regulating the subcellular localization of p95-APP1 between the endosomal recycling compartment and sites of actin organization at the cell surface. J. Cell Sci. 114, 4509–4520. 37. Roche, J.P., Packard, M.C., Moeckel-Cole, S., and Budnik, V. (2002). Regulation of synaptic plasticity and synaptic vesicle dynamics by the PDZ protein Scribble. J. Neurosci. 22, 6471– 6479. 38. Montcouquiol, M., Rachel, R.A., Lanford, P.J., Copeland, N.G., Jenkins, N.A., and Kelley, M.W. (2003). Identification of Vangl2 and Scrb1 as planar polarity genes in mammals. Nature 423, 173–177. 39. Etienne-Manneville, S. (2006). Wound healing in primary astrocytes. Methods Enzymol. 406, 565–578.