Altered expression of human Scribble is associated with invasive epithelial cancers, however, its role in tumour development remains unclear. Mutations in Drosophila Scribble result in loss of polarity, overproliferation and 3D-tumourous overgrowth of epithelial cells. Using complementation studies in Drosophila we recently demonstrated that expression of human Scribble can also regulate polarity and restrict tissue overgrowth. Here, we have undertaken a detailed study of human Scribble function in the polarized mammary cell line, MCF10A. We show that although Scribble does not seem to be required for apical-basal polarity or proliferation control in MCF10A cells, Scribble is essential for the control of polarity associated with directed epithelial cell migration. Scribble-depleted MCF10A cells show defective in vitro wound closure and chemotactic movement. The cells at the wound edge fail to polarize, show reduced lamellipodia formation and impaired recruitment of Cdc42 and Rac1 to the leading edge. Furthermore, we show that this function is relevant in vivo as Scribble mutant mice show defective epidermal wound healing. This data identifies an essential role for mammalian Scribble in the regulation of the polarity specifically involved in directed epithelial migration.
Scribble is a tumour suppressor that is targeted for degradation by the E6 protein of high-risk human papilloma viruses (HPV) 16 and 18 that are the major causative factors for cervical carcinoma (Nakagawa and Huibregtse, 2000; Thomas et al., 2005). Scribble expression is significantly downregulated in HPV-positive invasive cervical cancers (Massimi et al., 2004; Nakagawa et al., 2004), but it also appears to be lost in a majority of invasive breast cancers (Navarro et al., 2005) and shows altered expression in colon cancer (Gardiol et al., 2006). Thus loss of Scribble is correlated with cancer development, but it is currently not clear whether it represents an important step in the progression of malignant disease (reviewed by Humbert et al. (2003)). In Drosophila loss of scribble (scrib) results in the loss of epithelial polarity, a failure to differentiate and overproliferation (Bilder and Perrimon, 2000), suggesting it may serve to directly control tumourigenic growth. In addition it was recently shown that in combination with activation of oncogenes such as Ras, Raf or Notch, loss of Scribble leads to the development of invasive and metastatic epithelial neoplasias in Drosophila larvae (Brumby and Richardson, 2003; Pagliarini and Xu, 2003). While in this context Scribble serves to negatively regulate cell motility and invasion, during Drosophila development Scribble is required to positively regulate the migration of epithelial sheets during dorsal closure of the embryo (Bilder et al., 2003), implying the role of Scribble in controlling cell movement may be cell type and/or context dependent. The single human homologue of Scribble is highly concentrated at epithelial cell junctions and expression of Scribble can rescue the polarity and overgrowth defects of Drosophila scribble mutants suggesting that Scribble may function as a tumour suppressor and regulate polarity in mammalian cells (Dow et al., 2003; Navarro et al., 2005). The recent identification and analysis of mouse Scribble (scrb1) as the gene mutated in the mouse circletail (Crc) (Murdoch et al., 2003) and rumpelstilzchen (rumz) (Zarbalis et al., 2004) mutants have indicated an important role for Scribble in mammalian development and planar cell polarity, however, the neonatal lethality of these mutants has so far precluded detailed analysis of Scribble function in epithelia.
We have undertaken a detailed analysis of Scribble function in the nontransformed and highly polarized human mammary cell line, MCF10A to elucidate the cellular function of this potentially important tumour suppressor. Here we show that although human Scribble is not required for apical-basal polarity or proliferation control in MCF10A cells, Scribble is essential for the control of polarity associated with directed epithelial cell migration in vitro. In addition, our work describes a novel important function for Scribble in regulating wound healing in vivo.
Scribble depletion is not sufficient to disrupt cell morphology or apical-basal polarity in MCF10A acini
To directly assess the effect of Scribble loss on the nontransformed epithelial cell line MCF10A we used retroviral-mediated RNA interference (Brummelkamp et al., 2002) to reduce the level of Scribble protein. Using two short hairpin RNA (shRNA) vectors targeted to different regions of the human Scribble transcript we could specifically reduce Scribble protein expression by 90% (shScrib6) and 60% (shScrib7) (Figure 1a). Hereafter cell populations stably expressing shScrib6 and shScrib7 vectors are jointly referred to as ScribbleKD.
In Drosophila, scribble mutant epithelial cells lose their characteristic epithelial morphology and become disorganized and multilayered. When grown on cell-culture plastic MCF10A ScribbleKD cells showed no obvious changes in cell morphology compared to control cells, each population displaying the characteristic epithelial cobblestone morphology (Figure 1b). It was possible that any effects on gross morphology resulting from Scribble knockdown would only be evident in a polarized ‘organotypic’ setting so we made use of a well-established culture system that promotes the formation of differentiated 3D acini structures when MCF10A cells are cultured in Matrigel (Petersen et al., 1992; Debnath et al., 2003). Under these culture conditions ScribbleKD cells formed spherical clonal acini that were morphologically indistinguishable from parental and cells expressing a control hairpin targeting EGFP (shEGFP) control acini (Figure 1c). As reported for Madin–Darby canine kidney (MDCK) epithelial cells (Dow et al., 2003; Navarro et al, 2005), Scribble was localized to points of cell–cell contact in MCF10A acini and overlapped with E-cadherin (Figure 1d, top row; data not shown). Knockdown of Scribble protein after 20 days in Matrigel culture was confirmed by immunostaining for Scribble (Figure 1d, top row). In both control and ScribbleKD acini we saw no evidence of multilayering of cells and all acini showed a polarized (basal) nucleus typical of luminal epithelial cells (Figure 1d). In Drosophila and Caenorhabditis elegans, loss of Scribble (or LET-413) completely disrupts the formation and distribution of the zonula adherens (Bilder and Perrimon, 2000; Legouis et al., 2000). Although some cells within ScribbleKD acini appeared to have reduced E-cadherin protein at the membrane, we did not detect any major overall disruption to the distribution of the adherens junctions, as marked by E-cadherin and β-catenin (Figure 1d, row 2 and 3). In addition, consistent with previous Scribble RNA interference studies in MDCK cells (Qin et al., 2005), we did not observe any difference in E-cadherin or β-catenin protein levels between control and ScribbleKD cells as measured by Western analysis (data not shown). We next examined apical-basal polarity in MCF10A acini following loss of Scribble. In control acini, apical-basal polarity is evident by the restricted distribution of α6-integrin to the basal region of each cell within acini and the apical localization (above the nucleus) of the Golgi apparatus (Figure 1d, row 4 and 5). In ScribbleKD acini both the basal distribution of α6-integrin and the apical localization of the Golgi apparatus were unaffected, suggesting that these cells can form polarized acini. However, in some cases the structure of the Golgi did appear slightly disorganized as compared to control acini (Figure 1d, bottom row). Taken together, the data described shows that in this organotypic 3D assay, depletion of Scribble protein does not induce disruption of cell–cell junctions, loss of apical-basal polarity or cell multilayering. In addition, Scribble depletion did not alter proliferation in either 2D or 3D culture or exit from the cell cycle (Supplementary Figure 1). Thus, Scribble is not essential for the regulation of apical-basal polarity or proliferation control in MCF10A cells.
Scribble is required for directed cell migration of MCF10A epithelial cells
To investigate the possible role of Scribble in migration of epithelial cells we analysed the directed migration of MCF10A cells in an in vitro scratch ‘wound-healing’ assay (Figure 2). Following scratching, the cells at the leading edge of the wound developed broad lamellipodia and migrated forward as a cohesive epithelial sheet. In the presence of EGF, wild-type and shEGFP control cells completely filled the wounded area 24 h following scratch wounding (Figure 2). In contrast, ScribbleKD cells failed to completely fill the denuded area after 24 h (Figure 2). Moreover, the effect of Scribble knockdown on cell migration was dose dependent as the decrease in migration reflected the level of Scribble protein.
To independently assess the requirement for Scribble in cell migration we used Boyden chamber assays to measure the movement of the MCF10A cell lines towards a chemotactic stimulus. Wild-type, EGF-starved cells plated in the upper compartment of the Boyden chamber effectively migrated towards EGF present in the lower chamber as compared to cells plated without a chemoattractant, which did not migrate (Figure 3). Consistent with the wound-healing assay, ScribbleKD cells showed significantly reduced migration towards a chemotactic stimulus and the level of chemotactic cell migration was reflective of the level of Scribble protein within the cells (Figure 3). A similar dose-dependent reduction in wound healing and chemotaxis was observed in ScribbleKD cells when hepatocyte growth factor (HGF) was used as a chemokinetic stimulus (data not shown), indicating that the defect in cell migration resulting from Scribble knockdown was not due to a specific defect in the response to EGF, but a more general deficiency in cell migration.
It was possible that the defect observed in wound healing and chemotactic migration in ScribbleKD cells was due to a general defect in the ability of cells to move, as has been shown in T cells with reduced Scribble expression (Ludford-Menting et al., 2005). To examine this possibility we tracked the movement of individual cells on collagen IV substrate in the presence of EGF but lacking a directional cue (i.e. scratch). All MCF10A cell lines migrated randomly on the collagen IV substrate (Figure 4a), although we observed significant variation in the average velocity of individual migrating cells (Figure 4b). ScribbleKD cells showed a small decrease in average migration velocity compared to control cells, however, this difference was not statistically significant (P>0.05; Student's t-test). We obtained similar data using collagen I as a substrate (data not shown). This suggests that the defect in sheet migration and chemotactic movement is specifically related to the ability of ScribbleKD cells to respond to directional cues.
Scribble is required for cell polarization following scratch-induced directional cues
In astrocytes and fibroblasts, cells at the leading edge of a wound polarize the nucleus, actin cytoskeleton and microtubule organizing centre (MTOC)/Golgi network in the axis of migration (Hall, 2005). This process involves both tethering of the MTOC/Golgi network and rearward movement of the nucleus and is largely regulated by the small GTPase Cdc42 (Etienne-Manneville and Hall, 2001; Gomes et al., 2005). We analysed the polarity of leading edge cells 6 h following scratch wounding of MCF10As. In 70–80% of control cells at the wound edge the Golgi was polarized within a 120° arc behind the nucleus (Figure 5a and b). In contrast, ScribbleKD cells showed no polarization and the Golgi remained essentially randomly distributed around the nucleus (Figure 5a and b). In addition ScribbleKD showed a disorganization of the cell cytoskeleton as the microtubules appeared disordered and ‘curled’ at the leading edge as opposed to migrating control MCF10A cells that had microtubules directed linearly in the plane of migration (Figure 5c). Manneville and co-workers have previously shown that the localized activation of the small GTPase Cdc42 is essential for polarization of the MTOC following scratch wounding (Etienne-Manneville and Hall, 2001). We observed a small but similar induction of Cdc42 activation following wounding of MCF10A cells in both control and ScribbleKD cells (Figure 5d), indicating Scribble is not regulating the global activation of Cdc42 to control polarity.
Scribble is required to recruit Rac1 and Cdc42 to the leading edge of migrating MCF10A cells
In addition to the failure to polarize in response to a directional cue, ScribbleKD cells failed to form stable lamellipodial protrusions observed in control cells (Supplementary movies 1 and 2). The formation of lamellipodia is thought to be dependent on the localized activation of the small GTPase Rac1 (Nobes and Hall, 1999). Consistent with previously published observations (Shin et al., 2005), retroviral transduction of a dominant-negative Rac1 (or dominant-negative Cdc42) construct into wild-type MCF10A cells lead to a 50% or less decrease in migration in wound closure, an effect comparable to that seen with the shScrib7 hairpin (data not shown). Recently Scribble was shown to associate with Rac1 via a direct interaction with the RacGEF βPix (Audebert et al., 2004, and our unpublished data). We saw that Scribble was colocalized with both βPix and Rac1 at EGF-stimulated membrane ruffles (Figure 6a), suggesting it may be important for the activation/localization of Rac1. To address this we measured Rac1 activation following stimulation with EGF and wounding. We observed no difference between control and ScribbleKD cells in the initial activation of Rac1 following EGF stimulation (Figure 6b) or at later time points (2 and 6 h) following scratch wounding (data not shown), suggesting that Scribble is not required for global activation of this GTPase. Similar to EGF-stimulated ruffles, Scribble colocalized with Rac1 at the leading edge of migrating cells. However, depletion of Scribble resulted in loss of Rac1 and actin-rich puncta from this site (Figure 6c). Given the failure to localize Rac1, we examined the localization of Cdc42 following scratching. Although activation of Cdc42 was not affected (Figure 5d), Scribble-depleted cells had lost the localization of Cdc42 at the leading edge of migrating cells (Figure 6d). In addition, we observed that like Rac1, Cdc42 was co-localized with Scribble at the leading edge (Figure 6d, inset). Together, these data demonstrate that Scribble is required for the recruitment of Rac1 and Cdc42 to the leading edge and for the establishment of polarity, sustained lamellipodia formation and subsequent cell migration.
Scribble is essential for wound healing in vivo
To test whether the requirement for Scribble in promoting cell migration was physiologically relevant in the context of a whole tissue, we made use of an existing mouse scribble mutant allele – rumpelschtilzchen (rumz) – published as ‘Line 90’ (Zarbalis et al., 2004). The rumz mutation is a single base pair change resulting in an isoleucine to lysine substitution at residue 285 (I285K) within the leucine-rich repeat (LRR) region of Scribble (Figure 7a). Homozygous rumz embryos (scribblerumz/rumz) die perinatally as a result of severe morphological defects including craniorachischisis and exencephaly (Zarbalis et al., 2004). Western blot analysis showed that the rumz mutation causes a five-fold reduction of Scribble protein level (Figure 7b), which is most likely due to effects of the mutation on protein stability as we detected no difference in the level of mRNA transcript between scribble+/+ and scribblerumz/rumz keratinocytes (data not shown). In addition, the mutation caused marked mislocalization of Scribble protein away from cell–cell contacts in embryonic keratinocytes (data not shown). Similar mutations in scribble homologues in Drosophila and C. elegans also result in mislocalization of the protein and are suggested to render the protein nonfunctional (Legouis et al., 2000, 2003; Bilder, 2004). Given these observations we asked whether disruption to Scribble function would affect directed cell migration in vivo. We used a previously described method to measure wound healing in embryos at 12.5dpc (McCluskey et al., 1993; Ting et al., 2005). As expected, the majority of control scribble+/+ and scribble+/rumz animals (81%, n=16) showed normal wound-repair 20 h following wounding (Figure 7c). In contrast, only 1/9 scribblerumz/rumz animals (11%) showed normal wound closure and the lesion remained open after 20 h (P<0.001; Fisher exact test). The movement of the epithelial sheet in control embryos was evident from the elongated morphology of the keratinocytes surrounding the wound observed by high magnification scanning electron microscopy (s.e.m.) (Figure 7c, lower row). This cell shape change was not evident in scribblerumz/rumz embryos. This data identifies Scribble as a crucial regulator of epithelial cell migration in the mouse in vivo and provides the first demonstration to date that the polarity regulator Scribble is required for mammalian wound healing.
Here, we show that Scribble is essential for proper epithelial cell movement in response to extracellular directional migration cues such as a chemotactic gradient or during wound healing. Following scratch-induced migration, Scribble is required to polarize the Golgi in the plane of migration and ScribbleKD cells show a disruption to microtubule organization and a significant reduction in directional cell migration. Scribble colocalizes with Rac and Cdc42 at the leading edge of migrating cells and is required for the recruitment of these GTPases to this site. Furthermore, the function of Scribble in regulating cell migration is conserved between different epithelial cell types and species as mutational disruption of Scribble function in the mouse results in defective wound healing of the epidermis in vivo. Taken together the above data identify an essential role for mammalian homologues of Scribble in directed epithelial migration by regulating the establishment of polarity and the formation of lamellipodia at the leading edge of migrating cells. This regulation is critical for wound healing in vivo.
Regulation of directional migration by mammalian Scribble
Following scratch-induced migration, the direction of cell movement is specified by a free (leading) edge where activation of the small GTPase Cdc42 stimulates the reorientation of the Golgi and MTOC to the front of the nucleus and subsequent directional extension of microtubules (Nobes and Hall, 1999; Etienne-Manneville and Hall, 2001). While Scribble is required for the establishment of polarity following scratch-induced migration, our data indicate that this defect is not caused by a failure in the global activation of Cdc42, but rather the impaired localization of Cdc42 to the leading edge where it can define the direction of migration. It has been suggested that Scribble might act as a scaffold protein to regulate the localization of membrane associated signalling components and in this way it could mediate the recruitment and localized activation of proteins like Cdc42. It is also possible that Scribble may act downstream of Cdc42 to regulate Golgi/MTOC reorientation through Par6, GSK3β and adenomatous polyposis coli (APC), the latter of which forms ternary complexes with small GTPases at the leading edge (Etienne-Manneville and Hall, 2003; Watanabe et al., 2004). In the astrocyte migration model, Cdc42 activation leads to recruitment of Par6/aPKCζ to the leading edge which in turn activates aPKCζ and results in phosphorylation of GSK3β (Etienne-Manneville and Hall, 2001, 2003). Inhibition of GSK3β through phosphorylation of Ser9 results in relocalization of APC to microtubule plus ends and subsequent binding to Dlg1 thus allowing attachment of microtubules to the plasma membrane (Etienne-Manneville et al., 2005). It is unclear how closely this model applies to epithelial migration. Scratch wound healing of MCF10A cells results in GSK3β phosphorylation at Ser9, however, we did not detect any differences in the induction of GSK3β phosphorylation between Scribble-knockdown and control cells (Supplementary Figure 2). In addition, in contrast to the situation in astrocyte migration, initial experiments indicate no significant colocalization of APC and Dlg1 at the leading edge of migrating control MCF10A cells (our unpublished data). Further experiments will be needed to determine the functional relationship between Dlg1 and Scribble, and Scribble and the Par/aPKC complex in epithelial migration.
In addition to the failure to polarize following a scratch, ScribbleKD cells also showed a delay in lamellipodium formation at the leading edge. Lamellipodia formation is known to be largely dependent on activation of Rac1 at the membrane, which promotes actin polymerization and membrane extension (Nobes and Hall, 1999). While Scribble knockdown did not affect Rac activation per se, it was required for Rac1 localization at the leading edge following scratch-induced migration, indicating that Scribble is essential for the appropriate polarized membrane localization of activated Rac1. In support of a direct role for Scribble in mediating Rac function, others have shown that Scribble can physically associate in a complex with Rac1 following stimulation of PC12 cells (Audebert et al., 2004). In addition, Scribble constitutively associates with the RacGEF βPix and the Arf-GAP protein GIT1 that was recently shown to be required for the restricted activation of Rac1 at the leading edge of migrating epithelial cells (Audebert et al., 2004; Nishiya et al., 2005). This presents the intriguing possibility that Scribble and GIT1 associate in a complex with βPix to coordinately regulate the spatially restricted activation of Rac1. The use of FRET-based live-cell imaging will be crucial to fully define the role of Scribble in the spatio-temporal activation of Rac1. Nevertheless, taken together, the recruitment and colocalization of Scribble with Rac1 and Cdc42 at the leading edge of normal cells, the impaired recruitment of Rac1 (and Cdc42) at the leading edge observed following depletion of Scribble, and the reported association of Scribble and Rac1, point to an important role for these Rho-GTPases in the regulation of Scribble-dependent migration.
To determine whether Scribble was essential in the regulation of directed cell migration in vivo in a physiologically relevant context, we made use of the existing Scribble mutant rumpelstilzchen. The rumpelstilzchen mutation is a single amino acid substitution in the LRR domain of Scribble that results in decreased protein expression and mislocalization of Scribble into the cytoplasm (shown here) and causes severe morphological abnormalities in the embryo (Zarbalis et al., 2004). Using an in vivo assay we have shown that Scribble is essential for proper wound healing in the mouse embryo, likely through the regulation of directed epithelial cell migration. Consistent with this idea, the circletail mouse (a second Scribble mutant mouse) exhibits an eye open at birth phenotype, which is commonly associated with disruptions to known cell migration regulators such as epidermal growth factor receptor (EGFR) and c-Jun (Berkowitz et al., 1996; Zenz et al., 2003; Mine et al., 2005). Thus, both in vitro and in vivo data presented here supports a critical role for Scribble in regulating directed migration in mammalian epithelium.
Conserved role for Scribble in cell migration
Data presented in this study not only describes a role for Scribble in regulating mammalian cell migration, it identifies that this function of Scribble is conserved in multiple epithelial cell types and between different species. The role of Scribble as a highly conserved regulator of all types of cell motility is further highlighted by two recent publications showing that Scribble controls the migration of nonepithelial cell types, namely motor neurons in zebrafish and murine T lymphocytes (Ludford-Menting et al., 2005; Wada et al., 2005). In zebrafish, zygotic mutation of scribble in the landlocked (llk) mutant or Scribble knockdown results in failure of nVII motor neuron migration and loss of maternal Scribble expression causes defects in convergent extension during embryogenesis (Wada et al., 2005). Knockdown of Scribble in murine MD45T lymphocytes results in a profound defect in cell polarity and a loss of uropod formation. In addition, Scribble depletion in these cells results in a striking loss of cell motility. While it is clear that Scribble is essential for normal regulation of cell migration, one observation that remains to be reconciled is that in some contexts, loss of Scribble enhances motility of epithelial cells. Recently, Qin et al. (2005) have shown that RNAi knockdown of Scribble in MDCK cells causes an increase in the migration of these epithelial cells. This change is reported to be mediated through disruption of E-cadherin-mediated cell adhesion, however, this function of Scribble may be cell type specific as we did not observe any change in junction formation (Figure 2) or cell adhesion (our unpublished observation, AS Yap, personal communication). Indeed, it is likely that the effect of Scribble loss on cell behaviour in all species is dependent on cellular context. It is clear at least in Drosophila that the developmental state, cellular environment and genetic background has a major influence on the effect of Scribble loss on cell migration. In the Drosophila embryo Scribble is required for cell migration during dorsal closure (Bilder and Perrimon, 2000). However, in other settings such as in cooperation with an activated allele of ras, loss of Scribble can induce spontaneous invasion and metastasis (Pagliarini and Xu, 2003). Interestingly, this context-specific function for Scribble may represent a more common theme for polarity proteins in migration (Humbert et al., 2006). In the Drosophila ovary, loss of Bazooka/Par3 or Par6 function in border cell epithelial migration results in disorganization of the border cell cluster and impaired migration (Pinheiro and Montell, 2004). In contrast, loss of Bazooka/Par3 in the context of activated Ras in the eye imaginal disc epithelium can give rise to invasive tumours in Drosophila (Pagliarini and Xu, 2003). In mammalian cells, the dissolution of cell–cell junctions following loss of Par3 (Chen and Macara, 2005) or Par6 (Ozdamar et al., 2005) has been proposed to promote migration and invasion of tumour cells, however, Par6/aPKC complex function is required for migration of a number of cell types such as astrocytes and fibroblasts (Etienne-Manneville and Hall, 2001; Cau and Hall, 2005).
The context-dependent regulation of migration by Scribble may have important consequences for tumourigenesis. Studies in a number of different human cancers have now shown that loss of Scribble and Dlg is associated with more invasive and aggressive malignancies (Watson et al., 2002; Boussioutas et al., 2003; Cavatorta et al., 2004; Lin et al., 2004; Nakagawa et al., 2004; Navarro et al., 2005; Gardiol et al., 2006). It is now clear that Scribble plays a critical role in the regulation of directed cell migration during development and tissue homeostasis (e.g. wound healing). To reconcile these observations, we would predict that similar to Drosophila, genetic factors such as activating mutations in Ras may switch the activity of Scribble from a positive regulator of migration in normal development and tissue homeostasis, to a negative regulator of cell motility in the tumour setting. The development of conditional mouse alleles for Scribble together with the use of existing mouse models of cancer will facilitate examining the requirement for Scribble in regulating migration in pathological situations such as cancer invasion and metastasis.
Materials and methods
Cell culture and antibodies
MCF10A cells were cultured in DMEM:F12 (Dulbecco's Modified Eagle Medium: F12) supplemented with 5% donor horse serum (Gibco, NY, USA), 10 μg/ml insulin (Actrapid, Novo Nordisk Pharmaceuticals Ltd., UK), 0. 5 μg/ml hydrocortisone (Sigma Aldrich, St Louis, MO, USA), 20 ng/ml EGF (PeproTech Cytolab Ltd., Rehovot, Israel), 100 ng/ml cholera toxin (Sigma), penicillin (100U/ml) and streptomycin (100 μg/ml) and maintained at 37°C in 5% CO2. Primary mouse keratinocytes were isolated from E17.5 embryos and cultured as described (Redvers and Kaur, 2005). Amphotrophic retrovirus was produced by cotransfection of 293T cells with RD114 amphotrophic packaging plasmid and pRetroSUPER-based retroviral vectors. Cells infected with pRetroSUPER-based virus were selected in 2 μg/ml puromycin for 1 week before use in experiments. Details of shRNA cloning and sequences are available as Supplementary methods. Phase contrast images of cells were captured on a Leica DMIRB inverted microscope using a × 10 objective (Leica, NA 0.22) and a Spot RT Slider cooled CCD camera (Diagnostics Instruments, Sterling Heights, MI, USA). Antibodies used for Western blotting: α-Scribble monoclonal Ab (7C6.D10; Dow et al., 2003) 1:100, α-tubulin (Sigma), 1:10 000. Antibodies used for immunofluorescence: α-Scribble goat polyclonal 1:200 (Santa Cruz Biotechnology Inc., Santa Cruz, CA, USA, C20), α-βPix rabbit polyclonal, 1:400 (Chemicon, San Jose, CA, USA), α-β-catenin mouse monoclonal, 1:200 (Transduction Labs BD, Franklin Lakes, NJ, USA), α-GM130 mouse monoclonal, 1:400 (BD Biosciences, San Jose, NJ, USA), α-α6-integrin mouse monoclonal, 1:400 (GoH3; Chemicon), α-Rac1 mouse monoclonal, 1:400 (BD), α-tubulin mouse monoclonal, 1:2000 (Sigma). Secondary fluorophores were sourced from Molecular Probes (Eugene, OR, USA): Alexa 568 phalloidin, donkey α-goat Alexa 488, donkey α-rabbit Alexa 594, donkey α-mouse Alexa 647.
Whole-cell protein lysate was resolved on standard Tris-glycine sodium dodecyl sulphate–polyacrylamide gels and transferred to polyvinylidene difluoride membrane (Millipore, Billerica, MA, USA). Membranes were blocked in 2% Skim milk/0.2% Tween-20 in phosphate-buffered saline for 30 min, incubated with appropriate primary and secondary antibodies in blocking solution and developed with LumiLight (Roche Diagnostics, Mannheim, Germany) or visualized using an Odyssey infrared imager (LiCor Biosciences, Lincoln, NE, USA), as per the manufacturers recommendations. Signal intensity was determined by densitometry using Image J freeware (http://rsb.info.nih.gov/ij/) or Odyssey imaging software, respectively.
MCF10A organotypic cultures
Three-dimensional organotypic cultures were performed essentially as previously described (Debnath et al., 2003). Briefly, subconfluent MCF10A cells (80% confluent) were trypsinized, resuspended in 3D culture media and plated at 4000 cells/well using the overlay method in growth factor reduced Matrigel (BD Biosciences). Phase contrast images were captured as described for 2D cultures using a × 5 objective (Leica, NA 0.11). Cultures were immunolabelled for confocal microscopy as previously described (Debnath et al., 2003). Images were acquired on an MRC-1000 scanning confocal microscope (BioRad, Hercules, CA, USA) using a × 20 objective (Leica, NA 0.5) and processed using Image J freeware and Adobe Photoshop CS software (Adobe Systems Inc., San Jose, CA, USA).
In vitro wound healing assay
Confluent MCF10A monolayers were EGF-starved overnight and ‘wounded’ by scratching the surface with a P200 micropipette tip. Media containing 20 ng/ml EGF (PeproTech) and 1 ng/ml Mitomycin C (Kyowa, Bristol-Myers Squibb, NY, USA) (to inhibit cell division) was applied and wound closure was followed over 24 h on an Olympus IX81 inverted live-cell microscope equipped with a motorized stage and a 37°C heated chamber (Olympus Corp., Shinjuku-ku, Tokyo, Japan). Images were captured using a × 4 objective (Olympus, NA 0.13) on an ORCA/ER CCD camera (Hamamatsu Photonic, Hamamatsu, Japan). Closure of the wound was quantitated using MetaMorph 6.3 software (Molecular Devices Corp., Downington, PA, USA) as the average distance moved by the leading edge over time. For immunofluorescent staining cells were fixed in 3.7% paraformaldehyde 6 h following scratching, permeabilized with 0.3% Triton X-100 for 5 min, blocked with 2% bovine serum albumin for 1 h and incubated with primary antibodies in block overnight at 4°C. Cells were incubated with appropriate secondary antibodies and mounted in ProLong Gold (Molecular Probes).
Transwell chemotactic migration assay
Subconfluent MCF10A cells (70–80% confluent) were EGF-starved overnight and plated in assay media (200 μl) at 5 × 104 cells/well in the upper chamber of 24-well format transwell plates (Falcon BD Labwork, Franklin Lakes, NJ, USA). EGF (20 ng/ml) containing media (600 μl) was placed in the lower chamber. After 16 h, cells were fixed in 10% buffered formalin and stained with 4,6-diamidino-2-phenylindole (DAPI) (0.5 μg/ml). Nonmigrated cells on the upper side of the filter were removed by swabbing with cotton wool. Cells were imaged on a Zeiss Axioskop 2 fluorescence microscope using a × 10 objective (Zeiss, NA 0.3) and a Spot RT/SE CCD camera (Diagnostics). Values were calculated as the average of two fields counted on each filter.
Cell movement on collagen substrate
Subconfluent MCF10A cells (70–80% confluent) were EGF-starved overnight plated in assay media at 5 × 103 cells/cm2 on collagen-coated cell culture plastic. Cells were allowed to adhere for 6 h without EGF then complete media was applied. The cells were imaged over 12 h on an Olympus IX81 inverted live-cell microscope as described for scratch assays. Cell movement was quantitated using MetaMorph 6.3 software (Molecular Devices Corp.) by tracking the position of individual cells and calculating the average distance moved as a function of time.
Rac1/Cdc42 activation assay
Confluent cells were wounded as described for in vitro wound healing assay subconfluent MCF10A cells (70–80% confluent) were EGF-starved overnight and stimulated with prewarmed complete media (containing 20 ng/ml EGF) for the indicated times. Rac1-GTP and Cdc42-GTP levels were assessed using the Upstate Biotechnology (Lake Placid, NY, USA) Rac1 activation assay kit according to manufacturers instructions.
In vivo embryo wound healing
E12.5 embryos were dissected from the uterus in sterile Tyrode's saline and the left hind limb was amputated with the aid of fine dissecting forceps. Embryos attached to their yolk sac were cultured in 2 ml of a 1:3 mix of rat serum: Tyrode's saline for 20 h, fixed and processed for s.e.m. as previously described (Ting et al., 2005). Gold-coated samples were imaged using a Phillips 515 scanning electron microscope at 20 kV. Statistical significance for the wound closure assays was determined using a Fisher's exact test.
microtubule organizing centre
Audebert S, Navarro C, Nourry C, Chasserot-Golaz S, Lecine P, Bellaiche Y et al. (2004). Mammalian Scribble forms a tight complex with the betaPIX exchange factor. Curr Biol 14: 987–995.
Berkowitz EA, Seroogy KB, Schroeder JA, Russell WE, Evans EP, Riedel RF et al. (1996). Characterization of the mouse transforming growth factor alpha gene: its expression during eyelid development and in waved 1 tissues. Cell Growth Differ 7: 1271–1282.
Bilder D . (2004). Epithelial polarity and proliferation control: links from the Drosophila neoplastic tumor suppressors. Genes Dev 18: 1909–1925.
Bilder D, Perrimon N . (2000). Localization of apical epithelial determinants by the basolateral PDZ protein Scribble. Nature 403: 676–680.
Bilder D, Schober M, Perrimon N . (2003). Integrated activity of PDZ protein complexes regulates epithelial polarity. Nat Cell Biol 5: 53–58.
Boussioutas A, Li H, Liu J, Waring P, Lade S, Holloway AJ et al. (2003). Distinctive patterns of gene expression in premalignant gastric mucosa and gastric cancer. Cancer Res 63: 2569–2577.
Brumby AM, Richardson HE . (2003). scribble mutants cooperate with oncogenic Ras or Notch to cause neoplastic overgrowth in Drosophila. EMBO J 22: 5769–5779.
Brummelkamp TR, Bernards R, Agami R . (2002). Stable suppression of tumorigenicity by virus-mediated RNA interference. Cancer Cell 2: 243–247.
Cau J, 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.
Cavatorta AL, Fumero G, Chouhy D, Aguirre R, Nocito AL, Giri AA et al. (2004). Differential expression of the human homologue of Drosophila discs large oncosuppressor in histologic samples from human papillomavirus-associated lesions as a marker for progression to malignancy. Int J Cancer 111: 373–380.
Chen X, Macara IG . (2005). Par-3 controls tight junction assembly through the Rac exchange factor Tiam1. Nat Cell Biol 7: 262–269.
Debnath J, Muthuswamy SK, Brugge JS . (2003). Morphogenesis and oncogenesis of MCF-10A mammary epithelial acini grown in three-dimensional basement membrane cultures. Methods 30: 256–268.
Dow LE, Brumby AM, Muratore R, Coombe ML, Sedelies KA, Trapani JA et al. (2003). hScrib is a functional homologue of the Drosophila tumour suppressor Scribble. Oncogene 22: 9225–9230.
Etienne-Manneville S, Hall A . (2001). Integrin-mediated activation of Cdc42 controls cell polarity in migrating astrocytes through PKCzeta. Cell 106: 489–498.
Etienne-Manneville S, Hall A . (2003). Cdc42 regulates GSK-3beta and adenomatous polyposis coli to control cell polarity. Nature 421: 753–756.
Etienne-Manneville S, Manneville JB, Nicholls S, Ferenczi MA, Hall A . (2005). Cdc42 and Par6-PKCzeta regulate the spatially localized association of Dlg1 and APC to control cell polarization. J Cell Biol 170: 895–901.
Gardiol D, Zacchi A, Petrera F, Stanta G, 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.
Gomes ER, Jani S, Gundersen GG . (2005). Nuclear movement regulated by Cdc42, MRCK, myosin, and actin flow establishes MTOC polarization in migrating cells. Cell 121: 451–463.
Hall A . (2005). Rho GTPases and the control of cell behaviour. Biochem Soc Trans 33: 891–895.
Humbert P, Russell S, Richardson H . (2003). Dlg, Scribble and Lgl in cell polarity, cell proliferation and cancer. Bioessays 25: 542–553.
Humbert PO, Dow LE, Russell SM . (2006). The Scribble and Par complexes in polarity and migration: Friends or Foes? Trends Cell Biol, in press.
Legouis R, Gansmuller A, Sookhareea S, Bosher JM, Baillie DL, Labouesse M . (2000). LET-413 is a basolateral protein required for the assembly of adherens junctions in Caenorhabditis elegans. Nat Cell Biol 2: 415–422.
Legouis R, Jaulin-Bastard F, Schott S, Navarro C, Borg JP, Labouesse M . (2003). Basolateral targeting by leucine-rich repeat domains in epithelial cells. EMBO Rep 4: 1096–1102.
Lin HT, Steller MA, Aish L, Hanada T, Chishti AH . (2004). Differential expression of human Dlg in cervical intraepithelial neoplasias. Gynecol Oncol 93: 422–428.
Ludford-Menting MJ, Oliaro J, Sacirbegovic F, Cheah ET, Pedersen N, Thomas SJ et al. (2005). A network of PDZ-containing proteins regulates T cell polarity and morphology during migration and immunological synapse formation. Immunity 22: 737–748.
Massimi P, Gammoh N, Thomas M, Banks L . (2004). HPV E6 specifically targets different cellular pools of its PDZ domain-containing tumour suppressor substrates for proteasome-mediated degradation. Oncogene 23: 8033–8039.
McCluskey J, Hopkinson-Woolley J, Luke B, Martin P . (1993). A study of wound healing in the E11.5 mouse embryo by light and electron microscopy. Tissue Cell 25: 173–181.
Mine N, Iwamoto R, Mekada E . (2005). HB-EGF promotes epithelial cell migration in eyelid development. Development 132: 4317–4326.
Murdoch JN, Henderson DJ, Doudney K, Gaston-Massuet C, Phillips HM, Paternotte C et al. (2003). Disruption of scribble (Scrb1) causes severe neural tube defects in the circletail mouse. Hum Mol Genet 12: 87–98.
Nakagawa S, Huibregtse JM . (2000). Human scribble (Vartul) is targeted for ubiquitin-mediated degradation by the high-risk papillomavirus E6 proteins and the E6AP ubiquitin-protein ligase. Mol Cell Biol 20: 8244–8253.
Nakagawa S, Yano T, Nakagawa K, Takizawa S, Suzuki Y, Yasugi T et al. (2004). Analysis of the expression and localisation of a LAP protein, human scribble, in the normal and neoplastic epithelium of uterine cervix. Br J Cancer 90: 194–199.
Navarro C, Nola S, Audebert S, Santoni MJ, Arsanto JP, Ginestier C et al. (2005). Junctional recruitment of mammalian Scribble relies on E-cadherin engagement. Oncogene 24: 4330–4339.
Nishiya N, Kiosses WB, Han J, Ginsberg MH . (2005). An alpha4 integrin-paxillin-Arf-GAP complex restricts Rac activation to the leading edge of migrating cells. Nat Cell Biol 7: 343–352.
Nobes CD, Hall A . (1999). Rho GTPases control polarity, protrusion, and adhesion during cell movement. J Cell Biol 144: 1235–1244.
Ozdamar B, Bose R, Barrios-Rodiles M, Wang HR, Zhang Y, Wrana JL . (2005). Regulation of the polarity protein Par6 by TGFbeta receptors controls epithelial cell plasticity. Science 307: 1603–1609.
Pagliarini RA, Xu T . (2003). A genetic screen in Drosophila for metastatic behavior. Science 302: 1227–1231.
Petersen OW, Ronnov-Jessen L, Howlett AR, Bissell MJ . (1992). Interaction with basement membrane serves to rapidly distinguish growth and differentiation pattern of normal and malignant human breast epithelial cells. Proc Natl Acad Sci USA 89: 9064–9068.
Pinheiro EM, Montell DJ . (2004). Requirement for Par-6 and Bazooka in Drosophila border cell migration. Development 131: 5243–5251.
Qin Y, Capaldo C, Gumbiner BM, Macara IG . (2005). The mammalian Scribble polarity protein regulates epithelial cell adhesion and migration through E-cadherin. J Cell Biol 171: 1061–1071.
Redvers RP, Kaur P . (2005). Serial cultivation of primary adult murine keratinocytes. Methods Mol Biol 289: 15–22.
Shin I, Kim S, Song H, Kim HR, Moon A . (2005). H-Ras-specific activation of Rac-MKK3/6-p38 pathway: its critical role in invasion and migration of breast epithelial cells. J Biol Chem 280: 14675–14683.
Thomas M, Massimi P, Navarro C, Borg JP, Banks L . (2005). The hScrib/Dlg apico-basal control complex is differentially targeted by HPV-16 and HPV-18 E6 proteins. Oncogene 24: 6222–6230.
Ting SB, Caddy J, Hislop N, Wilanowski T, Auden A, Zhao LL et al. (2005). A homolog of Drosophila grainy head is essential for epidermal integrity in mice. Science 308: 411–413.
Wada H, Iwasaki M, Sato T, Masai I, Nishiwaki Y, Tanaka H et al. (2005). Dual roles of zygotic and maternal Scribble1 in neural migration and convergent extension movements in zebrafish embryos. Development 132: 2273–2285.
Watanabe T, Wang S, Noritake J, Sato K, Fukata M, Takefuji M et al. (2004). Interaction with IQGAP1 links APC to Rac1, Cdc42, and actin filaments during cell polarization and migration. Dev Cell 7: 871–883.
Watson RA, Rollason TP, Reynolds GM, Murray PG, Banks L, Roberts S . (2002). Changes in expression of the human homologue of the Drosophila discs large tumour suppressor protein in high-grade premalignant cervical neoplasias. Carcinogenesis 23: 1791–1796.
Zarbalis K, May SR, Shen Y, Ekker M, Rubenstein JL, Peterson AS . (2004). A focused and efficient genetic screening strategy in the mouse: identification of mutations that disrupt cortical development. PLoS Biol 2: E219.
Zenz R, Scheuch H, Martin P, Frank C, Eferl R, Kenner L et al. (2003). c-Jun regulates eyelid closure and skin tumor development through EGFR signaling. Dev Cell 4: 879–889.
We would like to thank members of the Bissell lab, and J Debnath and S Muthuswamy for advice with the culture of MCF10A cells. We thank S Ellis and N Waterhouse for assistance with confocal and live-cell microscopy, R Redvers for advice on culture of primary keratinocytes and J Hayes for maintenance of the scribblerumz mouse colony. We thank H Richardson and A Brumby for critical reading of the manuscript. This work was supported by project grants from the National Institutes of Health (USA) (PO1 HL53749⁁03) and Australian National Health and Medical Research Council (282400) (SMJ), the Australian Research Council (SMR), the Association for International Cancer Research UK and The Cancer Council Victoria (POH, SMR). LED was supported by a Cancer Council Victoria Postgraduate Cancer Research Scholarship, JSK was supported by The LAVFW Postdoctoral Research Fellowship and POH was supported by a Career Development Award from the Australian National Health and Medical Research Council.
About this article
Cite this article
Dow, L., Kauffman, J., Caddy, J. et al. The tumour-suppressor Scribble dictates cell polarity during directed epithelial migration: regulation of Rho GTPase recruitment to the leading edge. Oncogene 26, 2272–2282 (2007). https://doi.org/10.1038/sj.onc.1210016
- wound healing
The EMBO Journal (2021)
Scribble mutation disrupts convergent extension and apical constriction during mammalian neural tube closure
Developmental Biology (2021)
Biomimetic sulfated glycosaminoglycans maintain differentiation markers of breast epithelial cells and preferentially inhibit proliferation of cancer cells
Acta Biomaterialia (2021)
PLOS ONE (2021)
Proceedings of the National Academy of Sciences (2020)