341

Cell polarity: the ups and downs of the Par6/aPKC complex Domingos Henrique
y and Franc¸ois Schweisguthz A signaling complex in which atypical protein kinase C associates with a regulatory protein, Par6, plays an essential role in establishing cell polarity. Recent studies in organisms ranging from worms to mammals have highlighted some of the conserved mechanisms by which the assembly, localization and activity of this complex are regulated. Recent work is also beginning to unravel how this complex acts in concert with additional molecular complexes to establish and maintain polarity. Addresses
Instituto Medicina Molecular, Faculdade de Medicina de Lisboa, Av. Prof. Egas Moniz, 1649-028 Lisboa, Portugal e-mail: [email protected] y Instituto Gulbenkian Cieˆncia, Ap. 14, 2795 Oeiras, Portugal z CNRS UMR 8542, Ecole Normale Supe´rieure, 46 rue d’Ulm,75230 Paris, France e-mail: [email protected]

Current Opinion in Genetics & Development 2003, 13:341–350 This review comes from a themed issue on Pattern formation and developmental mechanisms Edited by Anne Ephrussi and Olivier Pourquie´ 0959-437X/$ – see front matter ß 2003 Elsevier Science Ltd. All rights reserved. DOI 10.1016/S0959-437X(03)00077-7

Abbreviations A–P anterior–posterior Apc Adenomatous polyposis coli aPKC atypical protein kinase C CRIB Cdc42/Rac-interactive-binding FRET fluorescence resonance energy transfer GBD GTPase-binding domain GFP green fluorescent protein GSK-3b glycogen synthase kinase-3b JAM-1 junctional adhesion molecule 1 Lgl lethal giant larvae MDCK Madin–Darby canine kidney PDZ PSD95/Discs large/ZO1 TJ tight junctions ZA zonula adherens

Introduction Polarity is a fundamental property of all eukaryotic cells. During the development of metazoans, regulation of cell polarity is essential for the spatio-temporal regulation of cell and tissue morphogenesis. Cell polarity is also involved in the regulation of cell identity. During asymmetric cell division, fate asymmetry relies on the polarized distribution and unequal segregation of cell-fate determinants at mitosis. The polarized distribution of key regulatory molecules not only underlies the generawww.current-opinion.com

tion of differentiated cell types in developing tissues but also controls the establishment of the anterior–posterior (A–P) body axis in flies and worms. Finally, cell polarity is a key aspect of cell differentiation, underlying cell migration and morphogenesis. In recent years, a signaling complex of three proteins, Par3, Par6 and atypical protein kinase C (aPKC), has emerged as a central player in the mechanisms that regulate cell polarity in the different cell types of various organisms ranging from worms to mammals (for review, see [1,2]). Recent work indicates that the core of this signaling complex is composed of Par6 and aPKC. Here, we review some recent studies that shed light on how the output of the Par6–aPKC complex is regulated and how this complex directs the establishment of cell polarity. In particular, we discuss findings suggesting that the signaling potential of the complex resides with aPKC and its kinase activity, with Par6 functioning as a key regulator of aPKC kinase activity and with other components, like Par3, functioning as structural scaffolds or as regulators of the kinase activity and/or specificity.

The Par proteins: from Caenorhabditis elegans to mammals The C. elegans par genes (par1 to par6) were first identified in a genetic screen for maternal-effect mutations affecting the unequal partitioning of polar granules to the posterior cell during the asymmetric division of the one-cell embryo [3]. Two of these genes, par-3 and par-6, encode PSD95/Discs-large/ZO1 (PDZ) domain proteins that colocalize at the anterior pole of the embryo [4,5]. A third protein, aPKC, encoded by the pkc-3 gene, was later shown to bind Par3 and to co-localize with Par3 and Par6, leading to the proposal that the three proteins form a complex [6] (Figure 1, Table 1). Work from several laboratories has established the pivotal role of the Par3– Par6–aPKC complex in establishing A–P polarity and in regulating asymmetric division in the C. elegans one-cell embryo (for review, see [7,8]). The initial cue for the anterior localization of the Par3– Par6–aPKC complex involves the microtubule-nucleating activity of the sperm asters, which act to exclude the complex from the posterior cortex of the cell [9–11]. Although little is known about the mechanisms that mediate this activity, a recent study has revealed that the anterior localization of the Par3–Par6–aPKC complex results from a two-step process [12]. In a first initiation step, the Par3–Par6–aPKC complex is excluded from the posterior cortex in response to the activity of the sperm asters. In a second step, another Par protein, Par2 (a RING Current Opinion in Genetics & Development 2003, 13:341–350

342 Pattern formation and developmental mechanisms

Figure 1

Par3/Baz (1465 aa)

PDZ C1 OPR motif Kinase CRIB

aPKC (606 aa) Par6 (351 aa) Cdc42 (191 aa) Cdc42•GDP Cdc42•GTP

Inactive Par6–aPKC

Active, cortical Par3–Cdc42–Par6–aPKC Current Opinion in Genetics & Development

Domain organization of Cdc42, Par3, Par6 and aPKC proteins. Domains involved in direct protein–protein interactions are connected by a red bar. Binding of CDC42 GTP to the inactive Par6–aPKC complex induces a conformational change in Par6 that leads to aPKC activation. Par3 may participate in targeting the active Cdc42–Par6–aPKC complex to specific sites at the cell cortex.

finger protein encoded by the par-2 gene [13]) acts to prevent the posterior spreading of the Par3–Par6–aPKC complex. The Par2 protein has no role in the first step, and its localization to the posterior cortex, where it maintains the exclusion of the Par3–Par6–aPKC complex, only happens when the complex is displaced anteriorly by the sperm asters [12]. Two other proteins, the class-II non-muscle myosin NMY2 and the 14.3.3 protein family member Par5, were also shown to participate in the initial exclusion of the Par3–Par6–aPKC complex from the posterior cortex [12]; NMY-2 is necessary to prevent accumulation of Par3–Par6– aPKC at the posterior cortex and Par5 seems to have a dual function, first promoting the posterior removal of the Par3– Par6–aPKC complex induced by the sperm asters and later mediating the ability of Par3–Par6–aPKC to remove Par2 from the anterior cortex [12,14].

Thus, an important question concerns how the Par3– Par6–aPKC complex associates with the cell cortex, a process in which Par3 seems to play a specific role. Indeed, Par3 is initially present at the cortex of both par-6 and pkc-3 mutant embryos, whereas the converse is not true: in par-3 mutants, Par6 and aPKC never reach the cortex [5,15,16]. This suggests that one of the roles of Par3 in the complex is to recognize binding partners at the cell membrane. Such an interaction may either provide a platform for the assembly of the complex at the membrane or target the pre-formed complex to the cortex. Future work should elucidate these points and reveal the mechanisms involved in localizing the Par3–Par6–aPKC complex to the anterior pole of the one-cell embryo, where it triggers a molecular cascade that directs the establishment of A–P polarity.

Table 1 Domain composition and partners of polarity regulators. Fly homologue

Mouse homologue

Domains

Partners

aPKC Par6 Bazooka (Par3) Cdc42 Crumbs (Crb) Stardust (Std) Discs-lost (Dlt) Scribble (Scrib) Discs-large (Dlg) Lethal giant larvae (Lgl)

PKCd/PKCz Par6A–D ASIP (Par3) Cdc42 Crb PALS1 PATJ mScrib mDlg mLgl

Kinase, C1, OPR CRIB, PDZ, OPR PDZ – EGF, LamG, TM L27, PDZ, SH3, GUK MRE, PDZ LRR, PDZ PDZ, SH3, GUK WD40

Par6, Par3, GSK-3b* aPKC,Par3,Lgl, Std/PALS1 aPKC, Par6, JAM-1 Par6 Std/PALS1 Crumbs, Dlt/PATJ, Par6 – GUKHolder GUKHoldery MyosinII, syntaxin-4, Par6

Summary of our current knowledge on structural domains and interacting partners of the proteins that form the three molecular complexes considered in this review. *The interaction may be indirect. yMany interactors of Dlg have been characterized in flies and mammals; see for instance [65] and references therein.

Current Opinion in Genetics & Development 2003, 13:341–350

www.current-opinion.com

Cell polarity: the ups and downs of the Par6/aPKC complex Schweisguth and Henrique 343

Which molecular mechanisms are involved downstream of the Par3–Par6–aPKC complex is an important question that has become even more crucial with the realization that similar complexes, involving homologues of the C. elegans proteins, are core regulators of cell polarity in other organisms, from flies to vertebrates (Table 1). Indeed, the proteins encoded by the Drosophila par-6, aPKC and par-3 (bazooka) genes also form a complex [17–19]. This complex is known to regulate several developmental processes (for review, see [1,8]): A–P polarization of the presumptive oocyte in the female germ-line, apical–basal polarization of epithelial cells in the embryo, establishment of asymmetry in mitotic neuroblasts and sensory organ precursor cells, and polarized migration of border cells in the egg chamber. In all these processes, the three proteins appear to co-localize extensively, suggesting that, like their C. elegans homologues, the Drosophila proteins Par-6 and aPKC, together with Par3, act as a signaling complex. In mammals, homologues of Par6 (Par6A–D [20]), aPKC (aPKCl and aPKCz [21]) and of Par3 (also known as ASIP [6]) have also been identified (Table 1). In fact, an interaction between Par6 and aPKC proteins was first shown in Madin–Darby canine kidney (MDCK) cells, where they form a complex with Par3 [20,22]. The three proteins were shown to co-localize at tight junctions (TJs) [6,20,23,24] and studies in cultured epithelial cells indicate that the Par3–Par6–aPKC complex promotes the formation of epithelial TJs and thereby contributes to the establishment and maintenance of apical–basal polarity [23–25]. Finally, Par6 and aPKC have also been described as regulators of cell polarity in migrating primary rat astrocytes [26]. In these cells, Par3 does not appear to co-localize with Par6 and aPKC and is therefore not part of the complex [26].

motif [29]. However, the Par6 CRIB motif is not complete, lacking two conserved histidines at the C-terminal half, and alone is not sufficient to bind Cdc42 [20]. The PAR6 GTPase-binding domain (GBD) is actually formed by the semi-CRIB motif and the adjacent PDZ domain, which has been shown to be absolutely required to form a stable Cdc42 GTP–Par6 complex [20]. Recently, the crystal structure of a complex consisting of Par6 bound to activated Cdc42 with the mutation Q61L has been solved [30], confirming that the PDZ domain indeed contributes to the normal binding of activated Cdc42, by extending and stabilizing the b-sheet structure formed by the partial CRIB motif. In addition, the peptide-binding region of the PDZ domain maps to a region opposite to the intra-molecular b-sheet formed by the PDZ domain and the partial CRIB motif [30], suggesting that the PDZ domain can promote the simultaneous binding of Cdc42 GTP and Par3 (or other PDZ interactors), further modulating the aPKC activity in the complex.

Activation of the Par6–aPKC complex by Cdc42

Cdc42 is an important regulator of cell polarity [31,32]. In yeast, Cdc42 is not only required to establish polarity in response to extrinsic or intrinsic spatial cues but is also sufficient to establish polarity in the absence of any such cues via a self-organizing mechanism [33]. Cdc42 is also important for the polarization of various cell types in metazoans, and accumulating evidence suggests that this involves its participation in the Par6–aPKC complex. For instance, it has been reported that Cdc42, Par6 and aPKC regulate the scratch-induced repolarization of migrating astrocytes [26]. In this experimental model, scratching a confluent monolayer of primary rat astrocytes leads to a reorganization of the microtubule network and a re-positioning of the Golgi apparatus along an axis defined by the wound (Figure 2a). Scratching induces localized activation of integrins at the leading edge of astrocytes, which leads to a fourfold increase in activation of Cdc42. This is accompanied by the recruitment of Par6–aPKC to the leading edge and by an increase in aPKC activity, which is dependent on Cdc42 activation. These findings imply that the correct polarization of astrocytes results from the localized activation of the Par6–aPKC complex at the leading edge by Cdc42 GTP. Consistent with this, overexpression of a kinase-dead aPKC or of dominant-negative versions of Cdc42 and Par6 was shown to inhibit the repolarization of the astrocytes [26].

An important advance towards an understanding of how aPKC activity can be regulated within the complex with Par6 has come from the identification of Cdc42 as a Par6 partner (Figure 1). In a search for novel effectors of the GTPase Cdc42 using a yeast two-hybrid approach, three groups have identified human Par6 as a direct interactor with an activated form of Cdc42, Cdc42(V12) [20,27,28]. In a parallel study, Par6 was also shown to co-immunoprecipitate with Cdc42(V12) [22]. Cdc42, in the GTP-bound state, interacts with proteins that contain a short conserved sequence called a Cdc42/Rac-interactive-binding (CRIB)

This raises the question of how the binding of Cdc42 GTP to Par6 stimulates the kinase activity of the associated aPKC. The N-terminal region of Par6 binds to the regulatory domain of aPKC [20] and overexpression of this region by itself in COS cells significantly increases the kinase activity of the linked aPKC [24]. This N-terminal region of Par6 must therefore have an intrinsic potential to bind and stimulate the kinase activity of aPKC. However, this potential seems to be suppressed by the adjacent GBD, as overexpression of the full Par6 protein is not able

These studies have highlighted the fundamental importance of the Par6–aPKC signaling complex and introduced two basic questions; the first concerns what is upstream (i.e. how the complex is localized and activated) and the second concerns what is downstream (i.e. how the complex regulates polarity). We address these two questions below in the context of epithelial cells.

www.current-opinion.com

Current Opinion in Genetics & Development 2003, 13:341–350

344 Pattern formation and developmental mechanisms

Figure 2

(a) Migrating astrocytes

Leading edge: Cdc42–Par6–aPKC Active Apc Microtubules

(b) Mitotic neuroblast

Apical Par3–Par6–aPKC Inactive Lgl Active Lgl Basal determinants

(c) Epithelial cell

Crb complex

Par3–Par6–aPKC

Scrib complex Apical Crb complex Apical Par3–Par6–aPKC ZA Active Lgl Basolateral Scrib complex Current Opinion in Genetics & Development

Functional analysis of the Par6–aPKC complex. (a) Migrating astrocyte: the active Par6–aPKC complex localizes at the leading edge of migrating astrocytes and reorganizes microtubule polarity via the regulation of Apc. (b) Asymmetrically dividing neuroblast: the apical Par3–Par6–aPKC complex inactivates Lgl at the apical cortex, via direct aPKC-mediated phosphorylation. Lgl activity is thereby restricted to the basal cortex, where it directs the localization of cell-fate determinants. (c) Epithelial cell: the crb and Par3–Par6–aPKC complexes localize apical to the ZA whereas the Lgl and the Scrib complex localize along basolateral membranes. Direct interaction between the Crb and Par3–Par6–aPKC complex is mediated by a Std–Par6 interaction. Genetic evidence suggests that the main function of the Crb complex is to antagonize the activity of the Scrib complex, whereas the main function of the Scrib complex is to antagonize the activity of the Par3–Par6–aPKC complex. Also, the apical localization of the Crb complex depends on the activity of the Par3–Par6–aPKC complex. See text for details.

to stimulate aPKC activity [24]. This is not the case if activated Cdc42 is present, which results in a significant increase of the kinase activity. Together, these results led to a model (Figure 1) in which the kinase activity of the Par6–aPKC complex is turned on by the binding of Cdc42 GTP to Par6’s GBD, relieving the GBD-mediated inhibition [24]. In support of this model, FRET analysis [30] indicates that the binding of Cdc42 GTP to Par6 induces a conformational change in Par6 that might increase the affinity of its N-terminal domain for aPKC and allow a switch in the kinase activity of the complex. Current Opinion in Genetics & Development 2003, 13:341–350

Thus, the emerging picture is that aPKC activation within the Par6 complex is mediated by Cdc42 GTP through double-negative regulation, whereby activated Cdc42 inhibits the Par6-dependent inhibition of aPKC. This suggests a simple mechanism for producing specific and highly localized aPKC activity within a cell: targeting the inactive Par6–aPKC complex to a defined cortical domain and regulating its timely activation by Cdc42. In such a scenario, additional components of the Par6–aPKC complex might regulate its targeting to specific cortical domains where it can be activated. www.current-opinion.com

Cell polarity: the ups and downs of the Par6/aPKC complex Schweisguth and Henrique 345

Regulatory role(s) of Par3 Par3 is a good candidate to be such a regulatory partner of the Par6–aPKC core. Par3 directly binds aPKC [6] and is phosphorylated in vitro by aPKC [22,34], and colocalization of Par6–aPKC with Par3 is seen in the C. elegans one-cell embryo [5,15], in Drosophila epithelial cells [18] and neuroblasts [17,19,35], and in mammalian cells [6,20,23,24]. Par3 may therefore be a conserved scaffold for the whole complex. As discussed above, C. elegans Par3 plays a role in targeting the Par6–aPKC complex to the cortex. However, recent results indicate that Par3 is not always found associated with Par6–aPKC (see below), raising the possibility that it may act as a sitespecific anchor for the cytosolic Par6–aPKC complex. In mammalian cells, Par3 directly binds, via its PDZ domains, the C terminus of the junctional adhesion molecule 1 (JAM-1) [36,37], a single-pass membrane protein belonging to the immunoglobulin superfamily. JAM-1 directly associates with the PDZ3 domain of ZO1, which itself directly binds the C terminus of claudin, a major structural constituent of the TJ [36]. Thus, JAM-1 may recruit Par3 to the claudin-based structures that will eventually form TJs, thereby providing a scaffold to assemble an active Par6–aPKC complex, which is necessary for correct TJ formation. In another recent study [38], Par3 was shown to directly bind, via its PDZ1 domain, to the C terminus of the immunoglobulin-like cell–cell adhesion molecules Nectin-1 and -3 in neuroepithelial cells. Whether Nectin-1 and -3, which localize at adherens junctions, recruit the Par6–aPKC complex via Par3 remains to be investigated. Whereas genetic data from worms and flies indicate that Par3 plays a primarily positive role in aPKC signaling, a negative regulatory function has also been proposed on the basis of the observation that addition of the aPKCbinding domain of Par3 to purified aPKC leads to a twofold reduction in the kinase activity [22]. Additional experiments are, however, required to validate this hypothesis, which is for the moment solely based on an in vitro result obtained in the absence of Par6. We note, however, that inhibition of aPKC by Par3 might help to prevent ‘ectopic’ aPKC activation until the complex arrives at the right place to be activated by Cdc42 GTP in a timely fashion. We further speculate that aPKC activation might then lead to Par3 phosphorylation and thereby avert Par3-mediated inhibition.

Phosphorylation targets of aPKC As mentioned above, the identification of direct phosphorylation targets of aPKC is expected to provide key insights into how asymmetrically localized Par6–aPKC regulates polarity. Recently, two targets of aPKC that might act as polarity regulators have been identified. The first, Lethal giant larvae (Lgl), is a membrane-associated protein with five WD40 repeats that was originally identified in Drosophila as the product of a tumor-suppressor www.current-opinion.com

gene [39]. New work in Drosophila [40] and mammals [41,42] has now shown that Lgl interacts directly with Par6, via its partial CRIB and PDZ domains, and with aPKC. Furthermore, sequential co-immunoprecipitation experiments reveal that Lgl, Par6 and aPKC are present in the same complex. However, Par3 does not appear to be part of this complex either in flies or in mammals [40,41], and further studies suggest that Lgl and Par3 compete to bind to the Par6–aPKC complex [42]. The presence of Lgl in the Par6–aPKC complex leads to its direct phosphorylation by aPKC at highly conserved serine residues [40,41,42], and this phosphorylation has been shown, in both Drosophila neuroblasts and mammalian cells, to control the intracellular localization of Lgl and its function in regulating cell polarity. In Drosophila neuroblasts, Lgl activity is known to be required for the basal localization of the cell-fate determinants Numb and Pros during asymmetric division [43,44]. It is not, however, required for the apical localization of the Par3–Par6–aPKC complex, suggesting that Lgl acts downstream of the apical complex to promote the localization of cell-fate determinants to the basal pole. Lgl itself is not asymmetrically distributed, and is found uniformly distributed at the cortex and in the cytoplasm [43,44]. However, as aPKC localizes apically, this might mean that apical Lgl is phosphorylated whereas the remaining pool of Lgl across the cell is unphosphorylated. In the absence of specific antibodies to confirm this, Betschinger et al. [40] used functional assays to investigate which pool of Lgl is functional. It was found that overexpression of a mutant version of Lgl that lacks the aPKC phosphorylation sites leads to the mislocalization of Numb, Pros and Miranda, an adaptor protein of Pros, around the whole cortex. This effect is not seen with wild-type Lgl, suggesting that the active form of the protein is the unphosphorylated version. In these experiments, the apical localization of the Par6–aPKC complex is unaffected. This led the authors to suggest a model (Figure 2b) in which aPKC phosphorylates and thereby inhibits Lgl at the apical cortex [40]. As a consequence, active Lgl is only present at the basal cortex, where it promotes the proper localization of Miranda, Numb and Pros (Figure 2b). How Lgl acts in this process is still unclear. Lgl directly interacts with myosin II [45], and the basal localization of cell-fate determinants has recently been shown to involve myosin-based transport [43,44,46,47], suggesting that Lgl might regulate this type of transport. Alternatively, by analogy to the role of yeast Lgl-like proteins in vesicle docking during exocytosis [48,49], active Lgl might direct vesicle trafficking within the cell. In either case, the localized apical activity of the Par6–aPKC complex is translated into a clear asymmetry of Lgl activity within the neuroblast, which eventually leads to the proper localization of fate determinants to the opposite, basal pole. Current Opinion in Genetics & Development 2003, 13:341–350

346 Pattern formation and developmental mechanisms

In mammals, a Lgl homologue (mLgl) is present in the baso-lateral domain of fully polarized MDCK cells, and this localization was shown to be dependent on its phosphorylation [42,50]. These findings are consistent with the notion that apically-located aPKC phosphorylates Lgl and thereby prevents localization of Lgl to the apical membrane of MDCK cells. The non-phosphorylated ‘active’ mLgl will thereby accumulate at the baso-lateral domain, where it might regulate baso-lateral exocytosis [50] via an interaction with syntaxin-4, a component of the baso-lateral exocytic machinery. However, in partially polarized MDCK cells, mLgl was also found to colocalize with aPKC and Par6 at points of cell–cell contact [42]. Interestingly, phosphorylation of mLgl is increased during polarization of MDCK cells and this is coincident with the progressive segregation of mLgl from Par6–aPKC along the apicobasal axis. This reinforces the suggestion that aPKC-mediated phosphorylation promotes the segregation of mLgl from these points of cell–cell contact to the baso-lateral membrane during the polarization of MDCK cells. On the basis of these results, a model was proposed by Yamanaka et al. [42] in which mLgl is bound to the inactive Par6–aPKC complex during polarization of MDCK cells and antagonizes the Par3-dependent anchoring of this complex to points of cell–cell contacts. This inhibition is counteracted by Cdc42, which would activate aPKC at points of cell–cell contact and thereby lead to mLgl being phosphorylated, released from the Par6–aPKC complex and restricted to the baso-lateral domain [42]. In another recent study [41], the role of mLgl phosphorylation was assayed during the polarization of migrating astrocytes. In these cells, mLgl colocalizes with aPKC at the leading edge, and a mutant version of mLgl that lacks the aPKC phosphorylation sites inhibits, albeit weakly, the cells’ polarization. Together, these studies show that the aPKC-dependent phosphorylation of Lgl is an important step in the mechanisms downstream of the Par6– aPKC complex that regulate the establishment of polarity in different cell types. Another target of the Par6–aPKC complex has been recently identified in rat primary astrocytes: the glycogen synthase kinase-3b (GSK-3b) [51]. It was shown that GSK-3b physically associates in a complex with aPKC, Par6 and possibly Cdc42. However, only unphosphorylated GSK-3b kinase could be immunoprecipitated with aPKC, indicating that phosphorylation leads to its dissociation from the complex. To address the role of this phosphorylation event (which occurs at serine 9, and which is thought to inhibit the kinase activity of GSK3b [52]), the authors used a scratch assay in primary astrocytes; as discussed above, this assay has previously shown that the complex Cdc42–Par6–aPKC controls cell polarization following scratching [26]. GSK-3b was shown Current Opinion in Genetics & Development 2003, 13:341–350

to become phosphorylated at the leading edge of polarizing astrocytes, where the activity of the Cdc42–Par6– aPKC complex is localized, and further studies indicate that GSK-3b phosphorylation is indeed dependent on the activity of Cdc42, Par6 and aPKC. Thus, polarizing astrocytes have low GSK-3b activity at the leading edge, due to its phosphorylation by the Cdc42–Par6–aPKC complex. This localized inactivation of GSK-3b may promote the accumulation of non-phosphorylated Adenomatous polyposis coli (Apc), which is known to be a direct target of GSK-3b and which regulates the interactions of microtubule ends with the cortex [53]. In fact, Apc was found to accumulate at the leading edge, in a Cdc42- and aPKC-dependent manner, where it associates with the plus ends of microtubules [26]. This seems to be important for the correct polarization of astrocytes, as overexpression of Apc molecules that are unable to bind microtubules leads to disruption of the process. Therefore, active, unphosphorylated Apc is needed at leading edges to regulate microtubule dynamics and the polarization of migrating astrocytes. Together, these findings illustrate another mechanism by which localized activity of the Par6–aPKC complex might lead to polarization of a cell — the modulation of microtubule dynamics via inhibition of the GSK-3b-dependent phosphorylation of Apc (Figure 2a).

Three distinct complexes act in a hierarchical manner to initiate epithelial polarity The establishment of apico-basal polarity in epithelial cells may be the process in which the function of the Par3–Par6– aPKC complex is best understood. Genetic and biochemical studies in both Drosophila and mammalian cells have uncovered the organizing role of the Par3–Par6–aPKC complex during the biogenesis of epithelia (for review, see [1]). In this process, the Par3–Par6–aPKC complex acts together with two other conserved protein complexes (Figure 2c and 3, Table 1). One complex is apical and contains the transmembrane protein Crumbs [54], the protein Stardust [55,56] (PALS1 in mammalian cells [57]) and the protein Discs-lost [58] (PATJ in mammalian cells [59,60]). This complex will be referred to here as the Crb complex. The other localizes basal to the zonula adherens (ZA), a specialized region of the cell that separates the apical and baso-lateral membrane domains and plays a central role in establishing the correct apico-basal polarity of epithelia [61]. It includes the proteins Scribble [62,63] and Dlg [64], which are physically linked via adaptor molecules [65]. This complex will be referred to here as the Scrib complex. As mentioned above, this complex exhibits a strong functional interaction with Lgl but there is to date no evidence for a direct physical interaction between Lgl and components of the Scrib complex. Two recent papers [66,67] have analyzed in great detail how these three complexes — Par3–Par6–aPKC, www.current-opinion.com

Cell polarity: the ups and downs of the Par6/aPKC complex Schweisguth and Henrique 347

Figure 3

Crumbs (2139 aa)

EGF-like LamG

Crb complex

Stardust (1083 aa) Discs-lost (871 aa)

TM L27 MRE PDZ

Scribble (1465 aa) Scrib complex Discs-large (960 aa)

SH3 GUK LRR

Lgl (1161 aa)

WD40 Current Opinion in Genetics & Development

Domain organization of proteins regulating epithelial polarity. Domains involved in direct protein–protein interaction are connected by a red bar. The indirect Dlg–Scrib interaction is shown by a yellow bar.

Crb and Scrib — functionally interact to establish apical– basal polarity in early Drosophila embryos. Although losing par3, crb, std, scrib or lgl functions affects the generation of the ZA at gastrulation similarly, both genetic interactions and analysis of late embryonic phenotypes indicate that these three complexes are functionally distinct and that the apically located Par3–Par6–aPKC complex is at the top of a genetic hierarchy regulating epithelial polarity. First, loss of Crb complex activity delays but does not prevent the acquisition of normal polarity, with crb and std mutant cells exhibiting an enlarged baso-lateral domain at late embryogenesis. Conversely, the absence of the Scrib complex or Lgl does not block polarity establishment, with lgl and scrib mutant cells eventually acquiring a normal polarity and showing an enlarged apical domain. Thus, the Crb and Scrib complexes act antagonistically to confer an apical and basal character, respectively, to the membrane. Second, double mutant analysis revealed that the primary function of the Crb complex is to antagonize the activity of the Scrib complex and of Lgl, which in turn act together to inhibit the function of the Par3–Par6–aPKC complex (Figure 2c). The results of these genetic studies are entirely consistent with the inhibitory role proposed for mLgl in the regulation of the Par3–Par6–aPKC complex in polarizing MDCK cells [42]. Third, the Par3–Par6–aPKC complex appears to act before and upstream of the Crb and Scrib complexes. Indeed, the Par3–Par6–aPKC complex localizes to the forming ZA before Crb, and the early ZA defects seen in www.current-opinion.com

par3 mutant embryos precede those seen in crb mutants. Moreover, apical localization of the Crb complex is dependent on the Par3–Par6–aPKC complex, but not vice-versa. Thus, the Par3–Par6–aPKC complex functions before the Crb complex and is required for the apical localization of the latter, whereas the Crb complex mainly acts to maintain the apical localization of the Par3– Par6–aPKC complex. However, little is known about the molecular details underlying these genetic interactions. This gap is being filled by biochemical work on mammalian MDCK cells that has uncovered not only a direct interaction between Par6–aPKC and mLgl [41,42] but also a direct interaction between the Par3–Par6–aPKC and Crb complexes, with Par6 directly binding to PALS1 [68]. This interaction involves the PDZ domain of Par6 and a new domain of PALS1, named U1. Furthermore, this interaction seems to be facilitated by activated Cdc42. As PALS1 binds to the cytoplasmic end of the transmembrane protein Crb, the Par6–PALS1 interaction might explain how the Crb complex maintains the apical localization of the Par3–Par6–aPKC complex (Figure 2c). Alternatively, this interaction might explain how the Par3–Par6–aPKC complex controls the correct apical localization of the Crb complex; this may involve the phosphorylation of Crb by aPKC (S Sotillos and S Campuzano, personal communication). Future studies should reveal the details of the molecular intimacy displayed by these three molecular complexes. Nevertheless, a question remains. How does the forming ZA functionally interact with the Par3–Par6– aPKC complex? Establishing the molecular and functional links between these two complexes is an important goal for future research. Current Opinion in Genetics & Development 2003, 13:341–350

348 Pattern formation and developmental mechanisms

Conclusions An emerging conclusion from the studies described here is the central role of the Par6–aPKC complex in regulating the establishment of polarity in a variety of distinct cell types, from the one-cell embryo in C. elegans to migrating astrocytes and epithelial cells in mammals. In all these cells, the function of the Par6–aPKC complex involves fine-tuned spatio-temporally tight regulation of the signaling activity of aPKC, which, in turn, regulates the activity of proteins organizing cytoskeletal architecture (Apc) and vesicular trafficking (Lgl). The protein Par6 is an unique and versatile molecule in the way it regulates aPKC activity within the cell. It constitutively inhibits the kinase activity of aPKC via a unique structure that shields the regulatory domain of aPKC, thereby preventing ectopic signaling; via its partial CRIB–PDZ domain, it receives temporal and spatial inputs from various extracellular stimuli converging onto Cdc42, thus releasing inhibition of the aPKC’s kinase activity in a highly accurate manner; via its PDZ domain, it interacts with Par3 and PALS1 and thereby regulates the targeting of aPKC to specific microenvironments within the cell; and finally, via PALS1, it also helps to recruit substrates such as Lgl or Crb to aPKC. These studies have also illustrated how the activity of the Par6–aPKC complex integrates with other molecular machinery to implement and stabilize the initial asymmetry created by the localized activation of aPKC. In fact, the integration of Par6–aPKC signaling activity with other signaling pathways may be a general feature of cell polarization. This notion is nicely illustrated by a recent and elegant study showing that, in mitotic neuroblasts of Drosophila, the apical Par6–aPKC complex acts redundantly with a second signaling complex that includes the Gai subunit of the heterotrimeric G protein to regulate mitotic spindle asymmetry during unequal division [69]. A major challenge in the future will be to understand how localized activation of Par6– aPKC acts globally through distinct pathways to orchestrate, in time and space, physiological changes in cell polarity.

Acknowledgements We thank Fundac¸a˜ o para a Cieˆncia e Tecnologia and the Centre National de la Recherche Scientifique for funding. We also thank Y Bellaı¨che and members of our laboratories for critical reading.

References and recommended reading Papers of particular interest, published within the annual period of review, have been highlighted as:  of special interest  of outstanding interest 1.

Wodarz A: Establishing cell polarity in development. Nat Cell Biol 2002, 4:E39-E44.

2.

Kemphues K: PARsing embryonic polarity. Cell 2000, 101:345-348.

Current Opinion in Genetics & Development 2003, 13:341–350

3.

Kemphues KJ, Priess JR, Morton DG, Cheng NS: Identification of genes required for cytoplasmic localization in early C. elegans embryos. Cell 1988, 52:311-320.

4.

Etemad-Moghadam B, Guo S, Kemphues KJ: Asymmetrically distributed PAR-3 protein contributes to cell polarity and spindle alignment in early C. elegans embryos. Cell 1995, 83:743-752.

5.

Hung TJ, Kemphues KJ: PAR-6 is a conserved PDZ-domaincontaining protein that colocalizes with PAR-3 in Caenorhabditis elegans embryos. Development 1999, 126:127-135.

6.

Izumi Y, Hirose T, Tamai Y, Hirai S, Nagashima Y, Fujimoto T, Tabuse Y, Kemphues KJ, Ohno S: An atypical PKC directly associates and colocalizes at the epithelial tight junction with ASIP, a mammalian homologue of Caenorhabditis elegans polarity protein PAR-3. J Cell Biol 1998, 143:95-106.

7.

Rose LS, Kemphues KJ: Early patterning of the C. elegans embryo. Annu Rev Genet 1998, 32:521-545.

8.

Pellettieri J, Seydoux G: Anterior–posterior polarity in C. elegans and Drosophila — PARallels and differences. Science 2002, 298:1946-1950.

9.

Sadler PL, Shakes DC: Anucleate Caenorhabditis elegans sperm can crawl, fertilize oocytes and direct anterior–posterior polarization of the one-cell embryo. Development 2000, 127:355-366.

10. Wallenfang MR, Seydoux G: Polarization of the anterior– posterior axis of C. elegans is a microtubule-directed process. Nature 2000, 408:89-92. 11. O’Connell KF, Maxwell KN, White JG: The spd-2 gene is required for polarization of the anteroposterior axis and formation of the sperm asters in the Caenorhabditis elegans zygote. Dev Biol 2000, 222:55-70. 12. Cuenca AA, Schetter A, Aceto D, Kemphues K, Seydoux G:  Polarization of the C. elegans zygote proceeds via distinct establishment and maintenance phases. Development 2003, 130:1255-1265. GFP fusions of several proteins involved in A–P polarity were created and their localization followed by time-lapse microscopy during polarization in the living C. elegans one-cell embryo. This analysis provides evidence for an establishment phase and a maintenance phase during AP polarization. The authors were also able to establish a temporal and functional hierarchy between the par genes known to control A–P polarity. For instance, analysis of GFP–Par6 localization in par2 embryos revealed that initial localization of Par6 is independent of Par2. 13. Levitan DJ, Boyd L, Mello CC, Kemphues KJ, Stinchcomb DT: par-2, a gene required for blastomere asymmetry in Caenorhabditis elegans, encodes zinc-finger and ATP-binding motifs. Proc Natl Acad Sci USA 1994, 91:6108-6112. 14. Morton DG, Shakes DC, Nugent S, Dichoso D, Wang W, Golden A, Kemphues KJ: The Caenorhabditis elegans par-5 gene encodes a 14-3-3 protein required for cellular asymmetry in the early embryo. Dev Biol 2002, 241:47-58. 15. Tabuse Y, Izumi Y, Piano F, Kemphues KJ, Miwa J, Ohno S: Atypical protein kinase C cooperates with PAR-3 to establish embryonic polarity in Caenorhabditis elegans. Development 1998, 125:3607-3614. 16. Watts JL, Etemad-Moghadam B, Guo S, Boyd L, Draper BW, Mello CC, Priess JR, Kemphues KJ: par-6, a gene involved in the establishment of asymmetry in early C. elegans embryos, mediates the asymmetric localization of PAR-3. Development 1996, 122:3133-3140. 17. Petronczki M, Knoblich JA: DmPAR-6 directs epithelial polarity and asymmetric cell division of neuroblasts in Drosophila. Nat Cell Biol 2001, 3:43-49. 18. Wodarz A, Ramrath A, Grimm A, Knust E: Drosophila atypical protein kinase C associates with Bazooka and controls polarity of epithelia and neuroblasts. J Cell Biol 2000, 150:1361-1374. 19. Kuchinke U, Grawe F, Knust E: Control of spindle orientation in Drosophila by the Par-3-related PDZ-domain protein Bazooka. Curr Biol 1998, 8:1357-1365. www.current-opinion.com

Cell polarity: the ups and downs of the Par6/aPKC complex Schweisguth and Henrique 349

20. Joberty G, Petersen C, Gao L, Macara IG: The cell-polarity protein Par6 links Par3 and atypical protein kinase C to Cdc42. Nat Cell Biol 2000, 2:531-539. 21. Moscat J, Diaz-Meco MT: The atypical protein kinase Cs. Functional specificity mediated by specific protein adapters. EMBO Rep 2000, 1:399-403. 22. Lin D, Edwards AS, Fawcett JP, Mbamalu G, Scott JD, Pawson T: A mammalian PAR-3–PAR-6 complex implicated in Cdc42/Rac1 and aPKC signalling and cell polarity. Nat Cell Biol 2000, 2:540-547. 23. Suzuki A, Yamanaka T, Hirose T, Manabe N, Mizuno K, Shimizu M, Akimoto K, Izumi Y, Ohnishi T, Ohno S: Atypical protein kinase C is involved in the evolutionarily conserved par protein complex and plays a critical role in establishing epithelia-specific junctional structures. J Cell Biol 2001, 152:1183-1196. 24. Yamanaka T, Horikoshi Y, Suzuki A, Sugiyama Y, Kitamura K, Maniwa R, Nagai Y, Yamashita A, Hirose T, Ishikawa H et al.: PAR-6 regulates aPKC activity in a novel way and mediates cell–cell contact-induced formation of the epithelial junctional complex. Genes Cells 2001, 6:721-731.

stabilize the initial asymmetry. The end-result is full polarization of the yeast cell. The authors suggest that in vivo extrinsic and intrinsic polarity signals might use this feedback loop to establish polarity along a defined axis. 34. Hirose T, Izumi Y, Nagashima Y, Tamai-Nagai Y, Kurihara H, Sakai T, Suzuki Y, Yamanaka T, Suzuki A, Mizuno K et al.: Involvement of ASIP/PAR-3 in the promotion of epithelial tight junction formation. J Cell Sci 2002, 115:2485-2495. 35. Schober M, Schaefer M, Knoblich JA: Bazooka recruits Inscuteable to orient asymmetric cell divisions in Drosophila neuroblasts. Nature 1999, 402:548-551. 36. Itoh M, Sasaki H, Furuse M, Ozaki H, Kita T, Tsukita S: Junctional adhesion molecule (JAM) binds to PAR-3: a possible mechanism for the recruitment of PAR-3 to tight junctions. J Cell Biol 2001, 154:491-497. 37. Ebnet K, Suzuki A, Horikoshi Y, Hirose T, Meyer Zu Brickwedde MK, Ohno S, Vestweber D: The cell polarity protein ASIP/PAR-3 directly associates with junctional adhesion molecule (JAM). EMBO J 2001, 20:3738-3748.

25. Gao L, Joberty G, Macara IG: Assembly of epithelial tight junctions is negatively regulated by Par6. Curr Biol 2002, 12:221-225.

38. Takekuni K, Ikeda W, Fujito T, Morimoto K, Takeuchi M, Monden M, Takai Y: Direct binding of cell polarity protein PAR-3 to cell–cell adhesion molecule nectin at neuroepithelial cells of developing mouse. J Biol Chem 2003, 278:5497-5500.

26. Etienne-Manneville S, Hall A: Integrin-mediated activation of Cdc42 controls cell polarity in migrating astrocytes through PKCf. Cell 2001, 106:489-498.

39. Mechler BM, McGinnis W, Gehring WJ: Molecular cloning of lethal(2)giant larvae, a recessive oncogene of Drosophila melanogaster. EMBO J 1985, 4:1551-1557.

27. Johansson A, Driessens M, Aspenstrom P: The mammalian homologue of the Caenorhabditis elegans polarity protein PAR-6 is a binding partner for the Rho GTPases Cdc42 and Rac1. J Cell Sci 2000, 113:3267-3275.

40. Betschinger J, Mechtler K, Knoblich JA: The Par complex directs  asymmetric cell division by phosphorylating the cytoskeletal protein Lgl. Nature 2003, 22:326-330. This paper identifies Lgl as a Par6–aPKC phosphorylation target and proposes a simple model for the accumulation of fate determinants at the basal pole of asymmetrically dividing neuroblasts based on the inhibition of Lgl at the opposite pole. Using immunoprecipitation with anti-Par6 antibody in Drosophila embryos followed by mass spectroscopy sequencing, the authors report the presence of Lgl in the Par6–aPKC complex. The authors show that Lgl is phosphorylated by aPKC and that phosphorylated Lgl is inactive and cytoplasmic. They therefore suggest that phosphorylation of Lgl by the apical Par6–aPKC complex serves to inactivate it at the apical pole of neuroblasts, thus restricting its activity to the basal pole, where it directs the process of localizing Miranda and associated basal determinants.

28. Qiu RG, Abo A, Steven Martin G: A human homolog of the C. elegans polarity determinant Par-6 links Rac and Cdc42 to PKCf signaling and cell transformation. Curr Biol 2000, 10:697-707. 29. Burbelo PD, Drechsel D, Hall A: A conserved binding motif defines numerous candidate target proteins for both Cdc42 and Rac GTPases. J Biol Chem 1995, 270:29071-29074. 30. Garrard SM, Capaldo CT, Gao L, Rosen MK, Macara IG,  Tomchick DR: Structure of Cdc42 in a complex with the GTPasebinding domain of the cell polarity protein, Par6. EMBO J 2003, 22:1125-1133. This paper describes the crystal structure of a complex between Cdc42 with the mutation Q61L and the GBD of Par6 (which consists of semiCrib and PDZ), providing a coherent molecular basis to understand the biological function of this complex. These structural studies reveal that the way in which Par6 interacts with Cdc42 GTP is unique and different from the mode in which Cdc42 GTP binds WASP or PAK. This is due to the contribution of the Par6’s PDZ domain to Cdc42 GTP binding, which stabilizes the interaction with the partial CRIB domain. These studies further indicate that the structural motifs in the PDZ domain that allow its interaction with ‘classical’ PDZ partners are still available for protein– protein interactions in presence of Cdc42 GTP. Thus, Par6 can simultaneously bind Cdc42 and other PDZ partners, such as Par3. Finally, FRET analysis indicated that binding of Cdc42 GTP induces a conformational alteration in Par6 that may in turn regulate aPKC activity. 31. Johnson DI: Cdc42: An essential Rho-type GTPase controlling eukaryotic cell polarity. Microbiol Mol Biol Rev 1999, 63:54-105. 32. Pruyne D, Bretscher A: Polarization of cell growth in yeast. I. Establishment and maintenance of polarity states. J Cell Sci 2000, 113:365-375. 33. Wedlich-Soldner R, Altschuler S, Wu L, Li R: Spontaneous cell  polarization through actomyosin-based delivery of the Cdc42 GTPase. Science 2003, 299:1231-1235. Remarkable study on the ability of yeast cells to polarize in the absence of any spatial cue, and on the central roles of Cdc42 and the actin cytoskeleton in the process. The mechanism should involve a positive feedback loop involving Cdc42 and actin polymerization. The process can be initiated randomly at the plasma membrane at a point where a stochastic rise in the concentration of activated Cdc42 leads to localized formation of actin cables. This results in a higher delivery rate of vesicles to that point, which, by carrying more activated Cdc42, thus reinforce and www.current-opinion.com

41. Plant PJ, Fawcett JP, Lin DC, Holdorf AD, Binns K, Kulkarni S,  Pawson T: A polarity complex of mPar-6 and atypical PKC binds, phosphorylates and regulates mammalian Lgl. Nat Cell Biol 2003, 5:301-308. This paper reports the interaction in mammalian cells of mLgl and the Par6–aPKC complex and characterizes this interaction biochemically. The role of mLgl phosphorylation was assessed in a fibroblast scratch assay and nicely complements the results shown in [40] on Drosophila neuroblasts. 42. Yamanaka T, Horikoshi Y, Sugiyama Y, Ishiyama C, Suzuki A,  Hirose T, Iwamatsu A, Shinohara A, Ohno S: Mammalian Lgl forms a protein complex with PAR-6 and aPKC independently of PAR-3 to regulate epithelial cell polarity. Current Biology 2003, 13:734-743. This study identifies mLgl-1 and mLgl-2 as direct phosphorylation targets of aPKCl. Initially identified as Par6 interactors, mLgl-1 and mLgl-2 were found in complexes containing aPKCl and Par6B, but not Par3. mLgl is shown to transiently colocalize with aPKC and Par6 at points of cell–cell contact in polaryzing MDCK cells. As polarity is established, mLgl is segregated from the complex to the baso-lateral domain and this correlates with an increase in its phosphorylation by aPKC. 43. Peng CY, Manning L, Albertson R, Doe CQ: The tumoursuppressor genes lgl and dlg regulate basal protein targeting in Drosophila neuroblasts. Nature 2000, 408:596-600. 44. Ohshiro T, Yagami T, Zhang C, Matsuzaki F: Role of cortical tumour-suppressor proteins in asymmetric division of Drosophila neuroblast. Nature 2000, 408:593-596. 45. Strand D, Jakobs R, Merdes G, Neumann B, Kalmes A, Heid HW, Husmann I, Mechler BM: The Drosophila lethal(2)giant larvae tumor suppressor protein forms homo-oligomers and is associated with nonmuscle myosin II heavy chain. J Cell Biol 1994, 127:1361-1373. Current Opinion in Genetics & Development 2003, 13:341–350

350 Pattern formation and developmental mechanisms

46. Lu B, Ackerman L, Jan LY, Jan YN: Modes of protein movement that lead to the asymmetric localization of partner of Numb during Drosophila neuroblast division. Mol Cell 1999, 4:883-891. 47. Petritsch C, Tavosanis G, Turck CW, Jan LY, Jan YN: The Drosophila myosin VI Jaguar is required for basal protein targeting and correct spindle orientation in mitotic neuroblasts. Dev Cell 2003, 4:273-281. 48. Kagami M, Toh-e A, Matsui Y: Sro7p, a Saccharomyces cerevisiae counterpart of the tumor suppressor l(2)gl protein, is related to myosins in function. Genetics 1998, 149:1717-1727. 49. Lehman K, Rossi G, Adamo JE, Brennwald P: Yeast homologues of tomosyn and lethal giant larvae function in exocytosis and are associated with the plasma membrane SNARE, Sec9. J Cell Biol 1999, 146:125-140. 50. Musch A, Cohen D, Yeaman C, Nelson WJ, Rodriguez-Boulan E,  Brennwald PJ: Mammalian homolog of Drosophila tumor suppressor lethal (2) giant larvae interacts with basolateral exocytic machinery in Madin–Darby canine kidney cells. Mol Biol Cell 2002, 13:158-168. This paper describes the localization of mLgl in polarized MDCK cells and correlates this to the phosphorylation status of the protein. Consistent with studies on the yeast Lgl homologue, the authors report that mLgl binds to syntaxin4, raising the possibility that mLgl may be involved in targeting specific proteins to the basolateral membrane during polarization of MDCK cells. 51. Etienne-Manneville S, Hall A: Cdc42 regulates GSK-3b and  adenomatous polyposis coli to control cell polarity. Nature 2003, 421:753-756. The authors describe the GSK-3b kinase as a downstream target of aPKC that can be specifically inactivated by aPKC phosphorylation. Inactivation of GSK-3b occurs specifically at the leading edge of polarizing astrocytes and might lead to accumulation of Apc at microtubule ends. Although the contribution of Apc to the polarization of these cells is not clear, the findings in this paper link two pathways that together might coordinate several aspects of cell polarity in various cell types. 52. Harwood JA: Regulation of GSK-3: a cellular multiprocessor. Cell 2001, 105:821-824. 53. Mogensen MM, Tucker JB, MacKie JB, Prescott AR, Nathke IS: The adenomatous polyposis coli protein unambiguously localizes to microtubule plus ends and is involved in establishing parallel arrays of microtubule bundles in highly polarized epithelial cells. J Cell Biol 2002, 157:1041-1048. 54. Tepass U, Theres C, Knust E: Crumbs encodes an EGF-like protein expressed on apical membranes of Drosophila epithelial cells and required for organization of epithelia. Cell 1990, 61:787-799. 55. Bachmann A, Schneider M, Theilenberg E, Grawe F, Knust E: Drosophila Stardust is a partner of Crumbs in the control of epithelial cell polarity. Nature 2001, 414:638-643. 56. Hong Y, Stronach B, Perrimon N, Jan LY, Jan YN: Drosophila Stardust interacts with Crumbs to control polarity of epithelia but not neuroblasts. Nature 2001, 414:634-638. 57. Roh MH, Makarova O, Liu CJ, Shin K, Lee S, Laurinec S, Goyal M, Wiggins R, Margolis B: The Maguk protein, Pals1, functions as an adapter, linking mammalian homologues of Crumbs and Discs Lost. J Cell Biol 2002, 157:161-172. 58. Bhat MA, Izaddoost S, Lu Y, Cho KO, Choi KW, Bellen HJ: Discs Lost, a novel multi-PDZ domain protein, establishes and maintains epithelial polarity. Cell 1999, 96:833-845.

Current Opinion in Genetics & Development 2003, 13:341–350

59. Roh MH, Liu CJ, Laurinec S, Margolis B: The carboxyl terminus of zona occludens 3 binds and recruits a mammalian homologue of discs lost to tight junctions. J Biol Chem 2002, 277:27501-27509. 60. Lemmers C, Medina E, Delgrossi MH, Michel D, Arsanto JP, Le Bivic A: hINADl/PATJ, a homolog of discs lost, interacts with crumbs and localizes to tight junctions in human epithelial cells. J Biol Chem 2002, 277:25408-25415. 61. Knust E, Bossinger O: Composition and formation of intercellular junctions in epithelial cells. Science 2002, 298:1955-1959. 62. Bilder D, Li M, Perrimon N: Cooperative regulation of cell polarity and growth by Drosophila tumor suppressors. Science 2000, 289:113-116. 63. Bilder D, Perrimon N: Localization of apical epithelial determinants by the basolateral PDZ protein Scribble. Nature 2000, 403:676-680. 64. Woods DF, Hough C, Peel D, Callaini G, Bryant PJ: Dlg protein is required for junction structure, cell polarity, and proliferation control in Drosophila epithelia. J Cell Biol 1996, 134:1469-1482. 65. Mathew D, Gramates LS, Packard M, Thomas U, Bilder D, Perrimon N, Gorczyca M, Budnik V: Recruitment of scribble to the synaptic scaffolding complex requires GUK-holder, a novel DLG binding protein. Curr Biol 2002, 12:531-539. 66. Tanentzapf G, Tepass U: Interactions between the crumbs,  lethal giant larvae and bazooka pathways in epithelial polarization. Nat Cell Biol 2003, 5:46-52. See annotation [67]. 67. Bilder D, Schober M, Perrimon N: Integrated activity of PDZ  protein complexes regulates epithelial polarity. Nat Cell Biol 2003, 5:53-58. These two papers present complementary results that identify a temporal and genetical hierarchy in the way three interacting molecular complexes regulate epithelial apico-basal polarity in Drosophila embryos. A precise temporal observation of mutant phenotypes, alone and in combination, together with a detailed analysis of alterations in protein localization, allowed the authors to show that a major function of the apical Crumbs complex is to restrict the activity of the Scribble complex to a more basal position. In turn, the Scribble complex mainly acts to antagonize the activity of the more apical Baz/Par3–Par6–aPKC complex, which appears to play a more central role in positioning the ZA. 68. Hurd TW, Gao L, Roh MH, Macara IG, Margolis B: Direct  interaction of two polarity complexes implicated in epithelial tight junction assembly. Nat Cell Biol 2003, 5:137-142. This paper describes the interaction of Par6 with PALS1 in polarized MDCK cells, thus revealing a molecular link between the two protein complexes that localize to the apical region of these cells. This interaction involves the PDZ domain of Par6 and the N terminus of PALS1 and is functionally important for the correct apico-basal polarization of MDCK cells. 69. Cai Y, Yu F, Lin S, Chia W, Yang X: Apical complex genes control  mitotic spindle geometry and relative size of daughter cells in Drosophila neuroblast and pI asymmetric divisions. Cell 2003, 112:51-62. Through the analysis of double mutant combinations, the authors elegantly show that the apical Baz/Par3–Par6–aPKC complex acts redundantly with a second signaling complex, the Pins–Gai complex, to regulate mitotic spindle positioning in asymmetrically dividing neuroblasts in Drosophila.

www.current-opinion.com