The Japanese Society of Developmental Biologists

Develop. Growth Differ. (2012) 54, 266276

doi: 10.1111/j.1440-169X.2012.01346.x

Review Article

Molecular mechanisms of cell shape changes that contribute to vertebrate neural tube closure
Makoto Suzuki, 1,2 * Hitoshi Morita 1 and Naoto Ueno 1,2 *
Division of Morphogenesis, Department of Developmental Biology, National Institute for Basic Biology, Nishigonaka 38, Myodaiji, Okazaki, 444-8585, Aichi, Japan; and 2Department of Basic Biology, School of Life Science, The Graduate University of Advanced Studies, Nishigonaka 38, Myodaiji, Okazaki, 444-8585, Aichi, Japan
1

During early development of the central nervous system, the neuroepithelial cells undergo dynamic changes in shape, cumulative action of which cause the neural plate to bend mediolaterally to form the neural tube. The apicobasal elongation changes the cuboidal cells into columnar ones, whereas apical constriction minimizes the cell apices, causing them to adopt wedge-like shapes. To achieve the morphological changes required for the formation of a hollow structure, these cellular changes must be controlled in time and space. To date, it is widely accepted that spatial and temporal changes of the cytoskeletal organization are fundamental to epithelial cell shape changes, and that noncetrosomal microtubules assembled along apicobasal axis and actin laments and non-muscle myosin II at the apical side are central machineries of cell elongation and apical constriction, respectively. Hence, especially in the last decade, intracellular mechanisms regulating these cytoskeletons have been extensively investigated at the molecular level. As a result, several actin-binding proteins, Rho/ROCK pathway, and cellcell adhesion molecules have been proven to be the central regulators of apical constriction, while the regulatory mechanisms of cell elongation remain obscure. In this review, we rst describe the distribution and role of cytoskeleton in cell shape changes during neural tube closure, and then summarize the current knowledge about the intracellular proteins that directly modulate the cytoskeletal organization and thus the neural tube closure. Key words: actomyosin, apical constriction, cell elongation, epithelial remodeling, microtubule.

Introduction
Three-dimensional change of epithelial sheet is the basic strategy for multicellular organisms to form elaborate structures out of simple cell layers. In vertebrates central nervous system (CNS) development, a at epithelial sheet, called the neural plate, is formed by the end of gastrulation (Stern 2005), as a thickening of the ectoderm on the dorsal surface of the embryo (Fig. 1A). The neural plate is composed of undifferentiated neural progenitors, which express unique molecular markers and have clear cellcell contacts on their apical side. Then the neural plate continuously narrows and lengthens along the anterior-posterior axis, which

*Author to whom all correspondence should be addressed. Email: msuzuki@nibb.ac.jp; nueno@nibb.ac.jp Received 29 December 2011; revised 1 March 2012; accepted 5 March 2012. 2012 The Authors Development, Growth & Differentiation 2012 Japanese Society of Developmental Biologists

is called convergent extension (Davidson & Keller 1999; Ybot-Gonzalez et al. 2007b). Simultaneously, the lateral borders of the neural plate elevate to form ridges, called the neural folds, parallel to the anteriorposterior axis. As development proceeds, the neural folds further elevate and bend the neural plate to form the neural groove (Fig. 1B). Finally, the edges of the neural folds meet at the dorsal midline, where they fuse to form a hollow epithelial structure, the neural tube, which is the anlage of whole CNS, and now positioned beneath the overlying surface ectoderm (Karfunkel 1974; Copp et al. 2003) (Fig. 1C). Alternatively, teleost sh and the posterior part of other vertebrates form neural tube through a distinct mechanism from the epithelial bending described above, where mesenchymal progenitor cells condense to form a solid rod that undergoes transformation to an epithelial tube (Lowery & Sive 2004). As the embryo develops, neuroepithelial cells proliferate and differentiate to build a multi-layered, well-patterned complex brain and spinal cord. Moreover, as the neural tube is one of the rst organs

Cell shape change in neural tube closure

267

(A)

(B)

(C)

Fig. 1. Process of vertebrate neural tube closure. Dorsal views (Top) and transverse sections (Bottom) of generalized amphibian embryos in early (A), middle (B), and late (C) neurulae. (A) By the end of gastrulation, a at neural plate (blue) is formed as a thickening of the ectoderm on the dorsal surface of the embryo. (B) The lateral borders of the neural plate elevate to form the neural folds. As development proceeds, the neural folds further elevate and bend the neural plate to form the neural groove. (C) The edges of the neural folds eventually meet at the dorsal midline, where they fuse to form a hollow structure, the neural tube, which is positioned beneath the overlying surface ectoderm (light blue).

developed during embryogenesis and is representative of epithelial morphogenesis, it is an excellent model to study the principle of epithelial cell behaviors and their role in developmental processes. Failure of neural tube closure causes congenital malformations, collectively called neural tube defects (NTDs), including anencephaly and spina bida, which occurs in approximately 1 in 1000 pregnancies in humans (Copp et al. 2003). Therefore, understanding the cellular and molecular mechanisms of neural tube closure is also of clinical interest to reveal the pathogenesis of NTDs in detail.

Cell shape changes in neural tube closure

During neural tube closure, the neuroepithelial cells dramatically change their shapes (Karfunkel 1974). Prior to neurulation, the shape of the neuroepithelial cell is cuboidal (Fig. 2Aa). Then the cells start to elongate along an apicobasal direction and become columnar (Schroeder 1970; Burnside 1971; Schoenwolf & Franks 1984) (Fig. 2Ab). The volume of neuroepithelial cells remains relatively constant, and hence apical and basal surface areas are decreased to a similar extent in this phase (Burnside 1973). In addition, cell apices are specically constricted and minimized, causing cells to adopt wedge- or bottle-like shape from columnar ones (Fig. 2Ac). These movements are called cell elongation and apical constriction, respectively, the coordinated actions of which generate the main physical force to cause the neural plate to bend (Glaser 1914; Karfunkel 1974) (Fig. 2B). This cellular morphogenesis-based intrinsic force leads to complete neural tube closure cooperatively with the

extrinsic forces that are generated by the medially directed cell movements in non-neural ectoderm (Karfunkel 1974; Hackett et al. 1997; Morita et al. 2012). Cell elongation precedes apical constriction, whereas to what extent the time courses of two movements are overlapped varies among organisms: these are almost entirely overlapped in Xenopus (Lee et al. 2007a), whereas in chick and mouse, neuroepithelial cells fully elongate prior to apical constriction (Schoenwolf & Franks 1984). Another different characteristic between Xenopus and other organisms is the numbers of cell layers in the neural plate: the neural plates in amniotes and urodele amphibians are pseudostratied, whereas the neural plate in Xenopus is stratied, and only the supercial cells undergo apical constriction to become wedge-shaped (Schroeder 1970). To achieve the complex morphological changes required for tube formation, cell elongation and apical constriction must be controlled in time and space, but how these apparently independent events are spatiotemporally regulated at the molecular level is unveiled. In this review, we rst describe the distributions and roles of two major cytoskeletons, actin laments and microtubules in apical constriction and cell elongation, respectively. Then we will summarize the current understanding of the molecular mechanisms of cell shape changes that contribute to vertebrate neural tube closure.

F-actin/Myosin II
Actin lament (F-actin) is a major component of the cytoskeleton, composed of monomeric actin (G-actin).

2012 The Authors Development, Growth & Differentiation 2012 Japanese Society of Developmental Biologists

268

M. Suzuki et al.

(A) (a)

(B)

(a)

(b)

(b) (c)

Fig. 2. Cell shape changes that contribute to neural tube closure. (A) The process of cell shape change. (a) Prior to neurulation, the shape of the neuroepithelial cell is cuboidal. (b) Cell elongation. The neuroepithelial cells rst elongate along apicobasal axis to become columnar. (c) Apical constriction. After cell elongation starts, cell apices are constricted and minimized, causing cells to adopt wedge or bottle-like shapes from columnar ones. Apical constriction generates the main physical force to cause the neural plate to bend to form the neural tube. (B) Scanning electron micrographs of a transverse slice through the anterior spinal level of a stage 17 (middle neurula) Xenopus laevis embryo. Rectangular area of (a) is enlarged in (b). Dashed lines in (b) indicate the outlines of neuroepithelial cells undergoing cell elongation and apical constriction. nt, notochord; so, somite.

In the neuroepithelial cells, as well as other epithelial cells, circular bands of actin laments exist at the apical junction (Fig. 3A). During neural tube closure, these actin lament bands become thickened as the cell apices become increasingly constricted (Baker & Schroeder 1967; Schroeder 1970; Burnside 1971; Karfunkel 1971) (Fig. 3C). These observations propose the possibility that actin lament bands act as contractile purse strings, causing the constriction of the apex of each neuroepithelial cell. If actin laments are disrupted by exposure to cytochalasin, which inhibits actin polymerization, the neuroepithelial cells fail to constrict their apices and become wedge-shaped despite the treated cells retaining their elongated shape (Karfunkel 1972; Linville & Shepard 1972). The neuroepithelial cells have non-muscle myosin II that binds to actin laments, comprising two heavy chains, two essential light chains and two regulatory light chains (Quintin et al. 2008). Myosin II motor activities are regulated by phosphorylation of the regulatory myosin light chain (MLC). The Rho-associated kinase (ROCK), an effector of the small GTPase Rho, is the major kinase that phosphorylates MLC. This phosphorylation increases the ATPase activity of MLC, which moves myosin II along actin laments, and in turn generates contractile force between actin laments. In the neural plate, myosin II is apically localized, and its

localization overlapped with that of F-actin (Lee & Nagele 1985; Rolo et al. 2009). In chick embryos, exposure to a myosin II inhibitor, blebbistatin, caused defects in apical constriction, resulting in severe NTDs (Kinoshita et al. 2008). Similarly, in Xenopus, its knockdown by antisense morpholino oligonucleotide (MO) causes delayed and defective neural tube closure, due to impairment of both convergent extension and bending of the neural plate. In such knockdown embryos, apical F-actin accumulation and apical constriction do not occur (Rolo et al. 2009). These indicate that the actin cytoskeleton and contractile force generated by non-muscle myosin II are the main actuators of apical constriction. Hence, molecular aspects of apical constriction have been extensively investigated as discussed later.

Microtubule
Microtubule is another essential component of the cytoskeleton as well as actin lament, composed of a- and b-tubulin. Microtubules nucleate from a centrosome, the most important type of microtubule-organizing centers (MTOCs) containing c-tubulin, a minus-end anchoring protein, or some noncentrosomal sites (Job et al. 2003; Musch 2004; Bartolini & Gundersen 2006). In the at neural plate cells and non-neural ectodermal

2012 The Authors Development, Growth & Differentiation 2012 Japanese Society of Developmental Biologists

Cell shape change in neural tube closure

269

(A)

(B)

(C)

Fig. 3. Distributions of actomyosin and microtubule cytoskeletons during cell shape change. (A) In cuboidal neuroepithelial cell prior to neurulation, thin circular bands of actin laments (red) exist at apical junction, while microtubules (green) are diffusively distributed in cytoplasm. (B) During cell elongation, noncetrosomal c-tubulin particles (yellow) are distributed apically. Noncetrosomal microtubules polymerize and assemble parallel to the apicobasal axis. (C) During apical constriction, actin lament bands become thickened. Nonmuscle myosin II (blue) actively slides and generates contractile force along apical actin laments as the cell apices become increasingly constricted. Intracellular and transmembrane proteins that regulate each cytoskeletal components are shown in the colored boxes.

cells, microtubules are mainly distributed in cell cortex and randomly oriented throughout the cell. However, in the neural cells undergoing apicobasal elongation, microtubules polymerize and assemble parallel to the axis of elongation of cells (Waddington & Perry 1966; Messier 1969; Schroeder 1970; Burnside 1971; Karfunkel 1971, 1972) (Fig. 3B). In addition, noncetrosomal c-tubulin is distributed apically as a thick cloud (Lee et al. 2007a). This accumulation of c-tubulin is correlated with the assembly of the robust apicobasally aligned microtubule arrays, suggesting that microtubules originated from noncetrosomal c-tubulin are involved in the elongation of the neural cells (Fig. 3B). Consistent with this, if neuroepithelial cells are treated with microtubule polymerization inhibitors, vinblastine and colchicine prior to the time when those cells normally elongate, they fail to do so. The disruption of microtubules also affects the cells that have already elongated prior to the treatment, causing their rounding up (Karfunkel 1971, 1972). Moreover, inhibitor treated-embryos neither initiate nor maintain the elevation of the neural folds, resulting in the failure of neural tube closure (Karfunkel 1971, 1972) Thus, microtubule organization is important for both the initiation of cell elongation and its maintenance, and such cellular behaviors are required for correct neural tube closure. The possible mechanisms by which microtubules affects cell elongation have been proposed by Burnside (1971), who suggested, from detailed electron microscopic observations, that microtubules contribute to cell elongation by transport of cytoplasmic materials toward the extending ends of the cell. A recent study suggests other functions of microtubules as a positive regulator of cellcell and cellextracellular matrix (ECM) adhesions during cell elongation (Suzuki et al. 2010,

discussed later), implying that multi-functional effects of microtubules are required for cell elongation of neuroepithelial cells.

Shroom3
Shroom3 was rst identied as a responsible gene of a recessive, lethal gene trap mutation in mice that causes exencephaly and spina bida, of which the neural folds mushroom away from the dorsal midline (Hildebrand & Soriano 1999). Similar defects can be seen in Xenopus and chick embryos with knockdown of orthologous gene by its specic MO, and RNAi, respectively (Haigo et al. 2003; Nishimura & Takeichi 2008). Now the function of Shroom3 has also been discovered in other models of epithelial bending, including vertebrate lens, gut, and Drosophila embryo (Hagens et al. 2006; Lee et al. 2009; Bolinger et al. 2010; Chung et al. 2010; Plageman et al. 2010, 2011b), suggesting a conserved role of Shroom3 as a key regulator of epithelial cell shape change in metazoan. In the neural plate, shroom3 mRNA is dynamically expressed reecting the spatial and temporal pattern of apical constriction (Hildebrand & Soriano 1999; Haigo et al. 2003). However, how shroom3 expression is regulated transcriptionally is not well understood, although Pax6 and Pitx1 have been shown to activate shroom3 expression in the lens placode and the developing gut, respectively (Chung et al. 2010; Plageman et al. 2010). Shroom3 includes a PDZ (PSD-95/Dgl/ZO-1) domain in the N-terminus, and two Apx/Shrm domains (ASD1 and ASD2) centrally and C-terminally, which are conserved among the gene family consisting of shroom14, and putative EVH1 and PDZ binding sites

2012 The Authors Development, Growth & Differentiation 2012 Japanese Society of Developmental Biologists

270

M. Suzuki et al.

(Hagens et al. 2006). Shroom3 protein is localized to the adherens junctions, and co-localizes with b-catenin and F-actin. In shroom3 mutants and knockdown embryos, apical constriction is signicantly inhibited, associated with reductions in apical accumulation of F-actin and phosphorylated MLC (pMLC) along the apical junctions (Hildebrand & Soriano 1999; Haigo et al. 2003; Lee et al. 2007a; Nishimura & Takeichi 2008). Conversely, ectopic expression of Shroom3 induces apical constriction, apical accumulation of F-actin and pMLC in embryonic epithelial cells of Xenopus and MDCK cells (Haigo et al. 2003; Hildebrand 2005). At the molecular level, the central region of the Shroom3 containing ASD1 directly binds F-actin and determines apical localization, and that ASD2 triggers apical constriction, whereas the PDZ domain is dispensable for the Shroorm3 activity. Treatment with blebbistatin or a ROCK inhibitor Y-27632 reverts the constriction phenotype caused by gain of Shroom3 function in MDCK cells, indicating that ROCK-induced myosin II activity is required for Shroom3-induced apical constriction (Hildebrand 2005). However, in Xenopus, dominant-negative Rap1a and Rap1GAP inhibit ectopic Shroom3-induced apical constriction, suggesting that Rap1 GTPase also might act downstream of Shroom3 (Haigo et al. 2003). We will further discuss the functional relationship between Shroom3 and Rho/ROCK pathway later. In addition to apical constriction, Shroom3 has an important function in cell elongation as well (Lee et al. 2007a, 2009). Disruption of Shroom3 function in neural plate cells by MO or a dominant-negative form of Shroom3 blocks apicobasal elongation. Defects in cell elongation are associated with a failure to assemble thick, apicobasally aligned microtubules, whereas overall microtubule is not disrupted. Strikingly, knockdown of Shroom3 completely eliminates the apical accumulation of c-tubulin in the neural plate cells. Ectopic Shroom3 in epithelial cells of the Xenopus blastula is sufcient to induce apical accumulation of c-tubulin, apicobasally aligned microtubules, and apicobasal cell elongation. Interestingly, the total level of c-tubulin is not affected by knockdown or overexpression of Shroom3, and there is no evidence of a physical interaction between Shroom3 and c-tubulin (Lee et al. 2007a). These observations suggest that Shroom3 drives apicobasal elongation by indirectly recruiting noncentrosomal c-tubulin to the apical side and inducing the assembly of a discrete population of aligned microtubules along the apicobasal axis. The other shroom family proteins Shroom1 and Shroom2 are expressed in the deep layer cells of the Xenopus neural plate that undergoes apicobasal elongation, and have the activities of recruiting c-tubulin to the apical

side in Xenopus epithelial cells (Fairbank et al. 2006; Lee et al. 2007a). Moreover, knockdown of Shroom2 function blocks apicobasal elongation in deep neural cells, causing NTDs (Lee et al. 2009). These observations demonstrate that the Shroom family has broader roles in controlling cell elongation during neural tube closure by coordinating assembly of microtubule organization.

Rho and ROCK

The family of small Rho GTPases, including Rho, Rac and Cdc42, are key regulators of actin cytoskeleton organization, and are required for diverse cellular functions, including cell shape change (Jaffe & Hall 2005). The activities of Rho proteins are controlled by its binding to GTP: they are active in the GTP-bound state and inactive in the GDP-bound state. In chick and Xenopus embryos, Rho is accumulated in the apical regions of the neural plate cells where it overlaps with F-actin, b-catenin, and pMLC (Kinoshita et al. 2008). Inhibition of Rho by Tat-C3 disrupts its apical accumulation and MLC phosphorylation, and leads to defects in neural tube closure (Kinoshita et al. 2008). Conversely, in MDCK cells, apically targeted, constitutively active RhoA induces apical constriction (Plageman et al. 2011a). This RhoA-mediated apical constriction is dependent on ROCK and myosin II function as Y-27632 and blebbistatin prevented the cells from apical constriction, respectively. These results indicate that apically localized, active RhoA is required and sufcient for myosin II-mediated apical constriction. Rho-associated kinases (ROCK1 and ROCK2), serine/threonine protein kinases, regulate the myosin II activity directly by phosphorylating MLC and indirectly by inactivating myosin phosphatase (Quintin et al. 2008). Inhibition of ROCKs in chick embryo with Y-27632 causes a reduction of pMLC and failure of neural tube closure (Wei et al. 2001; Kinoshita et al. 2008). Interestingly, ROCKs physically interact with the ASD2 domain of Shroom3 via their central regions (C1 domain), and their interaction is required for the Shroom3-mediated phosphorylation of MLC in MDCK cells (Nishimura & Takeichi 2008). Similarly, a dominant-negative RhoA inhibits Shroom3-induced apical constriction (Plageman et al. 2011b). In the neural plate, ROCK1 is colocalized with Shroom3 in the apical junctions and disruption of their interaction by Shroom3 RNAi or the ROCK C1 domain mutant reduces apical ROCK1 and pMLC and induces failure of neural tube closure (Nishimura & Takeichi 2008). These results suggest that ROCK1 is recruited to the apical junctions by Shroom3 and then leads to the

2012 The Authors Development, Growth & Differentiation 2012 Japanese Society of Developmental Biologists

Cell shape change in neural tube closure

271

phosphorylation of MLC dependently of active RhoA, and that this process is required for the Shroom3 function in apical constriction. Similarly, Shroom3-induced apical constriction is dependent on apical localization of RhoA (Hildebrand 2005). Interestingly, basolaterally targeted, constitutive active RhoA recruits Shroom3 to the basal end of the cell and induces basal constriction (Plageman et al. 2011a). In these cells, a tight-junction adaptor protein ZO-1 is also mislocalized, indicating the reversal of apicobasal polarity. These indicate that although the apical recruitment of ROCK1 is dependent on Shroom3, the apical localization of Shroom3 is further dependent on apically localized active RhoA. Additional studies will be necessary to resolve the mutual interaction between the RhoA/ROCK pathway and Shroom3 and its signicance in apical constriction. During embryogenesis, the Rho activity is regulated by noncanonical Wnt/PCP pathway (Wallingford & Habas 2005). In the chick and Xenopus neural plate, the expression of dominant-negative mutants of Wnt/ PCP components, Xdd1, XFrz7-dC, and XWnt11-dC cause a reduced apical accumulation of Rho, the expansion of cell apices, and neurulation defects (Kinoshita et al. 2008). Similarly, suppression of the activity of Daam1, a mediator of Rho activation in this pathway, causes failure of apical constriction in Xenopus embryo (Habas et al. 2001; Liu et al. 2011). These results implicate that the Rho activity in the neural plate is positively regulated by Wnt/PCP pathway. Rho in the neural plate is also negatively regulated by p190 RhoGAP, which stimulates GTP hydrolyzing activity of Rho to promote the inactive state (Jaffe & Hall 2005). In the mice lacking p190 RhoGAP, approximately 30% of homozygous embryos exhibit cranial neural tube closure defects, resulting in exencephaly. Interestingly, mutant neuroepithelial cells have an increased accumulation of F-actin in the basal side and a reduced apical constriction, suggesting that the aberrant Rho activity is associated with abnormal cell shape changes (Brouns et al. 2000).

Ena/VASP
Ena/VASP are a family of closely related, conserved actin-binding proteins. Vertebrates have three paralogous proteins, Mena (mammalian enabled), VASP (vasodilator stimulated phosphoprotein), and EVL (Ena/ VASP like) (Krause et al. 2003). Ena/VASP proteins have a conserved structural organization composed of N- and C-terminal Ena/VASP homology domains (EVH1 and EVH2) and a proline-rich central region. The EVH1 domain mediates subcellular localization by binding to the motif D/EFPPPP in binding partners, and the

EVH2 domain mediates binding to G- and F-actin. In the series of development of compound mutant mice of Ena/VASP family, it have been shown that triple null mutants (Mena/;VASP/;EVL/) and double null mutants (Mena/;VASP/;EVL+/+) exhibit cranial neural tube closure defects, resulting in exencephaly at high frequency (Menzies et al. 2004; Kwiatkowski et al. 2007). Since no single null mutant shows NTDs, Ena/ VASP proteins may redundantly function in mice. In Xenopus, an orthologue of Ena (Xena) is highly expressed in the neural plate, whereas expression of VASP and Xevl is relatively weak (Roffers-Agarwal et al. 2008). Xena protein is localized to the cell cortex in the neuroepithelial cells where it overlaps with b-catenin and vinculin, an Ena binding partner. Inhibition of Xena by MO or a dominant negative form (dnMena) causes failure of neural tube closure associated with the defects in apical constriction and apical accumulation of F-actin (Roffers-Agarwal et al. 2008). These suggest that Ena regulates apical constriction by facilitating the apical accumulation or remodeling of F-actin. Molecular mechanisms of Ena/VASP functions in apical constriction have been shown in mice and MDCK cells. Mena-decient mice that are heterozygous for another actin-binding protein Prolin I (Mena/;prolin I/+) also exhibit cranial neural tube closure defects, resulting in exencephaly, suggesting that Prolin I is involved in Mena-mediated apical constriction (Lanier et al. 1999). Furthermore, a recent study suggests the role of Ena/VASP in Shroom3-meditad apical constriction (Plageman et al. 2010). Shroom3 has a proline-rich sequence (FPn) that matches the consensus binding site of EVH1 in Ena/VASP, which is unique among the four Shroom proteins. In MDCK cells, deletion mutant of Shroom3 lacking EVH1 binding site shows the suppressed activity in apical constriction. Conversely, dnMena consisting of only the EVH1 domain suppresses Shroom3-induced apical constriction. Consistent with this, dnMena inhibits apical constriction during epithelial invagination in the prospective lens pits (Plageman et al. 2010). Apical localization of VASP is reduced or absent in Shroom3 mutants, and apical recruitment of VASP is dependent on EVH1 binding site in Shroom3 in MDCK cells. Together, these suggest that Ena/VASP physically and functionally interact with Shroom3 and such interaction is essential for Shroom3-mediated apical constriction and apical recruitments of Ena/VASP themselves (Plageman et al. 2010).

Myristoylated alanine-rich C kinase substrate

Myristoylated Alanine-Rich C Kinase Substrate (MARCKS) and MacMARCKS (also called F52 and

2012 The Authors Development, Growth & Differentiation 2012 Japanese Society of Developmental Biologists

272

M. Suzuki et al.

MRP) are closely related actin-binding, membraneassociated proteins (Arbuzova et al. 2002). MARCKS proteins have a basic effector domain that is phosphorylated by PKC and interacts with other molecules, such as calmodulin, F-actin and acidic phospholipids. MARCKS proteins have been implicated in several cellular events related to actin cytoskeleton, such as cell motility and adhesion (Arbuzova et al. 2002). The roles of MARCKS proteins in neural tube closure were identied from the analyses of knockout mice (Stumpo et al. 1995; Wu et al. 1996; Chen et al. 1996). 25% of MARCKS homozygote exhibits exencephaly and 63 100% (varies between independent studies) of MacMARCKS homozygote exhibits failure of neural tube closure and severe NTDs consisting of exencephaly and spina bida. In the chick neural plate, MARCKS is transiently accumulated at the apical side during processes of bending (Zolessi & Arruti 2001). On the other hand, in MDCK cells, MacMARCKS is localized at the basolateral membrane and colocalized with E-cadherin (Myat et al. 1998). The detailed mechanisms of how MARCKS and MacMARCKS proteins regulate cell shape change in the neuroepithelial cells remains to be investigated.

Epb4.1l5/Lulu
Erythrocyte protein band 4.1-like 5 (Epb4.1l5, also known as YMO1, Limulus, Lulu, Lulu1, Mosaic Eyes) is a member of the FERM-domain family. These family members link membrane-associated proteins to the actin cytoskeleton. The mammalian genome encodes at least 50 FERM proteins, including the actin-binding proteins ezrin, moesin, merlin (Tepass 2009). Epb4.1l5 has one paralogue in mammals, Epb4.1l4b (also known as Ehm2, Lulu2). Both genes are orthologues of Drosophila FERM proteins Yurt and produce two alternatively spliced isoforms, which share FERM and FERM-adjacent (FA) domains in their N-terminus. In an ENU mutagenesis screen, a mouse mutation, named limulus (lulu), has been isolated in which the neural plate shows an irregularly wide shape and forms ectopic folds, resulting in failure of neural tube closure (Lee et al. 2007b). The lulu mutation lies within the FERM domain of Epb4.1l5 and thus disrupts both isoforms. In the lulu neuroepithelial cells, apicobasal polarity is not affected, but apical accumulations of F-actin and Myosin II extend to the basal side, causing ectopic actomyosin activities at basal positions. Together with the observation that Lulu protein is enriched in the apical side, it is suggested that Epb4.1l5/Lulu function to anchor the actomyosin contractile machinery to the apical membrane that is required for the neural plate bending (Lee et al. 2007b).

In MDCK cells, overexpression of Epb4.1l5 induces apical constriction (Nakajima & Tanoue 2009). Similar reduction of apical area by Epb4.1l5 is also observed in other epithelial cells, including DLD1, NRK52E and Caco-2 cells. The activity of Epb4.1l5 for apical constriction requires both of the FERM and FA domains, and either the FERM or FA domain alone does not have such ability. Epb4.1l5-mediated apical constriction recruits F-actin, myosin II and its phosphorylated forms to apical junctions, and blebbistatin or Y-27632 suppressed cell shape change, indicating that ROCKmediated myosin II activity is required for Epb4.1l5induced apical constriction. Interestingly, although this situation is quite similar to Shroom3, knockdown of Shroom3 inhibits neither Epb4.1l5-induced apical constriction nor apical accumulation of pMLC, suggesting that Epb4.1l5 functions independently of Shroom3 (Nakajima & Tanoue 2009). The detailed mechanism of Epb4.1l5-induced apical constriction is not known so far. During epithelial-mesenchymal transition in transforming growth factor (TGF) b-stimulated NMuMG cells, Epb4.1l5 binds to p120-catenin through its FERM domain and inhibits its function to stabilize E-cadherin at cellcell contacts (Hirano et al. 2008). On the other hand, paralogous molecule Epb4.1l4b physically interacts with p114RhoGEF, a Rho-specic guanine nucleotide exchanging factor via FERM domain, and activates the p114RhoGEF catalytic activity (Terry et al. 2011). In addition, interaction between Epb4.1l4b and p114RhoGEF is negatively regulated by an apical polarity protein aPKC via phosphorylation of FA domain of Epb4.1l4b (Nakajima & Tanoue 2011). Whether Epb4.1l5 functions with p120-catenin, p114RhoGEF, and aPKC in the neural plate needs to be validated.

N-cadherin and nectin-2

In addition to intracellular molecules, cell adhesion molecules also have crucial roles in cellular morphogenesis in neural tube closure. Cadherins are calciumdependent cellcell adhesion proteins composed of a large family of more than 100 proteins (Suzuki & Takeichi 2008). Intracellularly, cadherins link the cortical actin through a-, b-, c-, p120-catenins, and EPLIN, indicating that cadherins function in assembling actin dependently of cellcell contacts (Abe & Takeichi 2008). N-cadherin is one of the type I sub-family members, characterized by ve extracellular cadherin domains. In vertebrates, N-cadherin expression is evident in the neural plate, and concentrated in the apical cytoplasm (Detrick et al. 1990; Fujimori et al. 1990; Radice et al. 1997). Depletion of N-cadherin in Xenopus by MO causes failure of neural tube closure

2012 The Authors Development, Growth & Differentiation 2012 Japanese Society of Developmental Biologists

Cell shape change in neural tube closure

273

(Nandadasa et al. 2009). The MO-receiving cells show expansion of apical surface associated with reductions of apical F-actin and pMLC, suggesting that N-cadherin is required for the maintenance of cell surface tension and contractility by linking the F-actin to the apical region. In mouse, N-cadherin/ embryos show the almost normal-appearing neuroepithelium, and only some embryos show undulation of the neural tube, suggesting that the defects in the neural tube are less severe than those in Xenopus (Radice et al. 1997). However, recent genetic analyses show that doubly heterozygous embryos with Shroom3 (Shroom3+/Gt; N-cadherin+/) displayed exencephaly at moderate frequency, suggesting that N-cadherin functions with Shroom3 to modulate the cytoskeletal machinery driving apical constriction (Plageman et al. 2011b). Another cell adhesive molecule participating in apical actin assembly with N-cadherin is nectin-2 (Morita et al. 2010). Nectin is a member of the immunoglobulin (Ig) superfamily and contains three extracellular Iglike domains, a single-pass transmembrane domain and a binding motif for afadin, an F-actin-binding protein (Takai et al. 2008). Nectin is preferentially localized to adherens junctions, and mediates cell adhesion in various cell types (Takai et al. 2008). In Xenopus embryo, nectin-2 is the predominant form expressed among four nectin family members. From the beginning of neurulation, its expression becomes more prominent in the neural plate where apical constriction takes place. Depletion of nectin-2 by MO causes defects in neural tube closure. The nectin-2 knockdown cells neither show apical constriction nor the apical F-actin accumulation. Conversely, ectopic expression of nectin-2 induces apical constriction accompanied by the accumulation of F-actin in the non-neural ectodermal cells. Nectin-2 physically interacts with N-cadherin via extracellular domain in the cis mode and directly recruits N-cadherin to the apical junctions (Morita et al. 2010). Furthermore, co-expression of nectin-2 and N-cadherin induce apical constriction and F-actin accumulation more efciently, and double knockdown synergistically exaggerates the failure of apical constriction in the neural plate. These results suggest that nectin-2 cooperatively functions with N-cadherin in apical constriction by recruiting N-cadherin and F-actin to the apical region.

signature) box, a motif recently identied in RBCC proteins (Short & Cox 2006). In addition, both proteins form homo- and hetero-dimers (Buchner et al. 1999; Cainarca et al. 1999) and physically interact with Mig12 (Mid1 interacting G12-like protein) (Berti et al. 2004). In humans, MID1 is responsible for X-linked Opitz G/BBB syndrome, which is characterized by hypertelorism, hypospadias, and additional midline defects (Quaderi et al. 1997). MID proteins are expressed in the neural plate in human, mouse, chick, and Xenopus (Buchner et al. 1999; Pinson et al. 2004; Granata et al. 2005; Suzuki et al. 2010), and knockdown of both of MID1 and MID2, or Mig12 by MOs cause NTDs in Xenopus (Hayes et al. 2007; Suzuki et al. 2010). In knockdown embryos, the neural plate cells fail to undergo cell elongation and apical constriction, which are associated with the defects in the apical F-actin accumulation and apicobasal arrays of microtubules. Furthermore, knockdown embryos have defects in apical localizations of a tight junction protein ZO-1, an adherens junction protein C-cadherin, and basal distribution of an extracellular matrix protein laminin (Suzuki et al. 2010). These results suggest that the regulation of microtubule organization by the MID1/ MID2/Mig12 complex is required for cell shape change and epithelial integrity in the neural plate as well as in other epithelial tissues (Suzuki et al. 2010).

Concluding remarks
In this review, we summarized the role of cytoskeleton and the intracellular and cell adhesion proteins directly modulating the cytoskeletal organization in cell shape changes during neural tube closure. The signicance of cytoskeleton in early CNS development was proven even at early 70s, but it is only since the twenty-rst century when the molecular mechanisms have been investigated in detail. Particularly, in the last decade, Shroom3 and Rho/ROCK pathway have been proven as the central regulators of apical constriction. Moreover, several actin-binding proteins and cellcell adhesion molecules have been shown to participate in apical F-actin assembly and the activation of non-muscle myosin II. On the other hand, although the regulatory mechanism of cell elongation remains relatively obscure, the recent studies have revealed the importance of Shroom3 and MID proteins in the regulation of microtubules and its cellular role in this process. What is the next question to be solved? The suggested function of Shroom3 seems to be a key scaffold to recruit Rho/ROCK pathway to the apical junction (Nishimura & Takeichi 2008; Plageman et al. 2011b). However, it is not yet known whether such activity of Shroom3 is regulated post-transcriptionally

MID1/MID2 and Mig12

MID1 and its paralogue MID2 encode conserved proteins belonging to the RBCC/TRIM (N-terminal RING nger-B box-coiled coil/tripartite motif) superfamily (Short & Cox 2006). MID1 and MID2 associate with microtubules through COS (C-terminal subgroup one

2012 The Authors Development, Growth & Differentiation 2012 Japanese Society of Developmental Biologists

274

M. Suzuki et al.

such as by phosphorylation and/or ubiquitination. Similarly, although Rho/ROCK pathway is active in the neural plate (Kinoshita et al. 2008; Nishimura & Takeichi 2008), which RhoGEF activates Rho, and whether the additional pathway exists upstream of Rho as well as Wnt/PCP pathway are still unclear. Identication of possible upstream factor(s) of the Rho and Shroom3 would be important for full understanding of the molecular bases of the initiation and maintenance of the apical constriction. In addition, to understand spatial and temporal dynamics of cell shape change, addressing how actin-binding and adhesion proteins regulate F-actin assembly in apical side at single molecular level is intriguing. Furthermore, the possible role of secreted factors is of particular interest. The neural plate expresses a variety of growth factors and morphogen, including bone morphogenetic protein (BMP) and sonic hedgehog and their antagonists. From genetic and in vivo manipulation analyses using mouse and chick embryos, it has been shown that antagonism of BMP signaling and its regulation by sonic hedgehog is necessary and sufcient for correct bending of the neural plate (YbotGonzalez et al. 2002, 2007a). BMP blockade by noggin overexpression can induce endocytosis of apical protein Par3 and cell adhesive molecule N-cadherin as a possible cause of apical constriction (Eom et al. 2011). In an independent study (Lee & Harland 2010), disrupting endocytosis with dominant-negative dynamin causes inefcient apical constriction and neural tube closure defects in Xenopus, suggesting that BMPdependent regulation of intracellular trafcking affects apical constriction. As roles of secreted factors have been also proposed in other apically constricting cells, including bottle cells (Choi & Sokol 2009) and Drosophila mesodermal invagination and eye imaginal disc (Costa et al. 1994; Corrigall et al. 2007; Escudero et al. 2007), it is intriguing to examine other signaling molecules as possible players in cell shape changes, for understanding how neural tube closure is spatially and temporally regulated at the whole tissue level. To answer questions discussed above, comparative approaches with other model systems and quantitative analysis combined with the live imaging analyses in high resolution would be promising approaches.

tory was supported by KAKENHI (23111529, 23770261 to M.S.; 21370102, 22127007 to N.U.) from the Japan Society for the Promotion of Science.

References
Abe, K. & Takeichi, M. 2008. EPLIN mediates linkage of the cadherin catenin complex to F-actin and stabilizes the circumferential actin belt. Proc. Natl Acad. Sci. USA 105, 1319. Arbuzova, A., Schmitz, A. A. & Vergeres, G. 2002. Cross-talk unfolded: MARCKS proteins. Biochem. J. 362, 112. Baker, P. C. & Schroeder, T. E. 1967. Cytoplasmic laments and morphogenetic movement in the amphibian neural tube. Dev. Biol. 15, 432450. Bartolini, F. & Gundersen, G. G. 2006. Generation of noncentrosomal microtubule arrays. J. Cell Sci. 119, 41554163. Berti, C., Fontanella, B., Ferrentino, R. & Meroni, G. 2004. Mig12, a novel Opitz syndrome gene product partner, is expressed in the embryonic ventral midline and co-operates with Mid1 to bundle and stabilize microtubules. BMC Cell Biol. 5, 9. Bolinger, C., Zasadil, L., Rizaldy, R. & Hildebrand, J. D. 2010. Specic isoforms of drosophila shroom dene spatial requirements for the induction of apical constriction. Dev. Dyn. 239, 20782093. Brouns, M. R., Matheson, S. F., Hu, K. Q., Delalle, I., Caviness, V. S., Silver, J., Bronson, R. T. & Settleman, J. 2000. The adhesion signaling molecule p190 RhoGAP is required for morphogenetic processes in neural development. Development 127, 48914903. Buchner, G., Montini, E., Andol, G., Quaderi, N., Cainarca, S., Messali, S., Bassi, M. T., Ballabio, A., Meroni, G. & Franco, B. 1999. MID2, a homologue of the Opitz syndrome gene MID1: similarities in subcellular localization and differences in expression during development. Hum. Mol. Genet. 8, 1397 1407. Burnside, B. 1971. Microtubules and microlaments in newt neuralation. Dev. Biol. 26, 416441. Burnside, B. 1973. Microtubules and microlaments in amphibian neurulation. Am. Zool. 13, 9891006. Cainarca, S., Messali, S., Ballabio, A. & Meroni, G. 1999. Functional characterization of the Opitz syndrome gene product (midin): evidence for homodimerization and association with microtubules throughout the cell cycle. Hum. Mol. Genet. 8, 13871396. Chen, J., Chang, S., Duncan, S. A., Okano, H. J., Fishell, G. & Aderem, A. 1996. Disruption of the MacMARCKS gene prevents cranial neural tube closure and results in anencephaly. Proc. Natl Acad. Sci. USA 93, 62756279. Choi, S. C. & Sokol, S. Y. 2009. The involvement of lethal giant larvae and Wnt signaling in bottle cell formation in Xenopus embryos. Dev. Biol. 336, 6875. Chung, M. I., Nascone-Yoder, N. M., Grover, S. A., Drysdale, T. A. & Wallingford, J. B. 2010. Direct activation of Shroom3 transcription by Pitx proteins drives epithelial morphogenesis in the developing gut. Development 137, 13391349. Copp, A. J., Greene, N. D. & Murdoch, J. N. 2003. The genetic basis of mammalian neurulation. Nat. Rev. Genet. 4, 784 793. Corrigall, D., Walther, R. F., Rodriguez, L., Fichelson, P. & Pichaud, F. 2007. Hedgehog signaling is a principal inducer of Myosin-II-driven cell ingression in Drosophila epithelia. Dev. Cell 13, 730742.

Acknowledgments
We apologize to the colleagues whose work has not been cited or cited only indirectly because of space limitations. We thank members of the N.U. laboratory for valuable discussions and comments, and Hanaichi UltraStructure Research Institute for technical support for scanning electron microscopy. Work in our labora-

2012 The Authors Development, Growth & Differentiation 2012 Japanese Society of Developmental Biologists

Cell shape change in neural tube closure

275

Costa, M., Wilson, E. T. & Wieschaus, E. 1994. A putative cell signal encoded by the folded gastrulation gene coordinates cell shape changes during Drosophila gastrulation. Cell 76, 10751089. Davidson, L. A. & Keller, R. E. 1999. Neural tube closure in Xenopus laevis involves medial migration, directed protrusive activity, cell intercalation and convergent extension. Development 126, 45474556. Detrick, R. J., Dickey, D. & Kintner, C. R. 1990. The effects of Ncadherin misexpression on morphogenesis in Xenopus embryos. Neuron 4, 493506. Eom, D. S., Amarnath, S., Fogel, J. L. & Agarwala, S. 2011. Bone morphogenetic proteins regulate neural tube closure by interacting with the apicobasal polarity pathway. Development 138, 31793188. Escudero, L. M., Bischoff, M. & Freeman, M. 2007. Myosin II regulates complex cellular arrangement and epithelial architecture in Drosophila. Dev. Cell 13, 717729. Fairbank, P. D., Lee, C., Ellis, A., Hildebrand, J. D., Gross, J. M. & Wallingford, J. B. 2006. Shroom2 (APXL) regulates melanosome biogenesis and localization in the retinal pigment epithelium. Development 133, 41094118. Fujimori, T., Miyatani, S. & Takeichi, M. 1990. Ectopic expression of N-cadherin perturbs histogenesis in Xenopus embryos. Development 110, 97104. Glaser, O. C. 1914. On the mechanism of morphological differentiation in the nervous system. Anat. Rec. 8, 525551. Granata, A., Savery, D., Hazan, J., Cheung, B. M., Lumsden, A. & Quaderi, N. A. 2005. Evidence of functional redundancy between MID proteins: implications for the presentation of Opitz syndrome. Dev. Biol. 277, 417424. Habas, R., Kato, Y. & He, X. 2001. Wnt/Frizzled activation of Rho regulates vertebrate gastrulation and requires a novel Formin homology protein Daam1. Cell 107, 843854. Hagens, O., Ballabio, A., Kalscheuer, V., Kraehenbuhl, J. P., Schiafno, M. V., Smith, P., Staub, O., Hildebrand, J. & Wallingford, J. B. 2006. A new standard nomenclature for proteins related to Apx and Shroom. BMC Cell Biol. 7, 18. Haigo, S. L., Hildebrand, J. D., Harland, R. M. & Wallingford, J. B. 2003. Shroom induces apical constriction and is required for hingepoint formation during neural tube closure. Curr. Biol. 13, 21252137. Hackett, D. A., Smith, J. L. & Schoenwolf, G. C. 1997. Epidermal ectoderm is required for full elevation and for convergence during bending of the avian neural plate. Dev. Dyn. 210, 397406. Hayes, J. M., Kim, S. K., Abitua, P. B., Park, T. J., Herrington, E. R., Kitayama, A., Grow, M. W., Ueno, N. & Wallingford, J. B. 2007. Identication of novel ciliogenesis factors using a new in vivo model for mucociliary epithelial development. Dev. Biol. 312, 115130. Hildebrand, J. D. & Soriano, P. 1999. Shroom, a PDZ domaincontaining actin-binding protein, is required for neural tube morphogenesis in mice. Cell 99, 485497. Hildebrand, J. D. 2005. Shroom regulates epithelial cell shape via the apical positioning of an actomyosin network. J. Cell Sci. 118, 51915203. Hirano, M., Hashimoto, S., Yonemura, S., Sabe, H. & Aizawa, S. 2008. EPB41L5 functions to post-transcriptionally regulate cadherin and integrin during epithelial-mesenchymal transition. J. Cell Biol. 182, 12171230. Jaffe, A. B. & Hall, A. 2005. Rho GTPases: biochemistry and biology. Annu. Rev. Cell Dev. Biol. 21, 247269.

Job, D., Valiron, O. & Oakley, B. 2003. Microtubule nucleation. Curr. Opin. Cell Biol. 15, 111117. Karfunkel, P. 1971. The role of microtubules and microlaments in neurulation in Xenopus. Dev. Biol. 25, 3056. Karfunkel, P. 1972. The activity of microtubules and microlaments in neurulation in the chick. J. Exp. Zool. 181, 289 301. Karfunkel, P. 1974. The mechanisms of neural tube formation. Int. Rev. Cytol. 38, 245271. Kinoshita, N., Sasai, N., Misaki, K. & Yonemura, S. 2008. Apical accumulation of Rho in the neural plate is important for neural plate cell shape change and neural tube formation. Mol. Biol. Cell 19, 22892299. Krause, M., Dent, E. W., Bear, J. E., Loureiro, J. J. & Gertler, F. B. 2003. Ena/VASP proteins: regulators of the actin cytoskeleton and cell migration. Annu. Rev. Cell Dev. Biol. 19, 541564. Kwiatkowski, A. V., Rubinson, D. A., Dent, E. W., Edward van Veen, J., Leslie, J. D., Zhang, J., Mebane, L. M., Philippar, U., Pinheiro, E. M., Burds, A. A., Bronson, R. T., Mori, S., Fassler, R. & Gertler, F. B. 2007. Ena/VASP Is Required for neuritogenesis in the developing cortex. Neuron 56, 441 455. Lanier, L. M., Gates, M. A., Witke, W., Menzies, A. S., Wehman, A. M., Macklis, J. D., Kwiatkowski, D., Soriano, P. & Gertler, F. B. 1999. Mena is required for neurulation and commissure formation. Neuron 22, 313325. Lee, C., Scherr, H. M. & Wallingford, J. B. 2007a. Shroom family proteins regulate gamma-tubulin distribution and microtubule architecture during epithelial cell shape change. Development 134, 14311441. Lee, C., Le, M. P. & Wallingford, J. B. 2009. The shroom family proteins play broad roles in the morphogenesis of thickened epithelial sheets. Dev. Dyn. 238, 14801491. Lee, H. Y. & Nagele, R. G. 1985. Studies on the mechanisms of neurulation in the chick: interrelationship of contractile proteins, microlaments, and the shape of neuroepithelial cells. J. Exp. Zool. 235, 205215. Lee, J. D., Silva-Gagliardi, N. F., Tepass, U., McGlade, C. J. & Anderson, K. V. 2007b. The FERM protein Epb4.1l5 is required for organization of the neural plate and for the epithelial-mesenchymal transition at the primitive streak of the mouse embryo. Development 134, 20072016. Lee, J. Y. & Harland, R. M. 2010. Endocytosis is required for efcient apical constriction during Xenopus gastrulation. Curr. Biol. 20, 253258. Linville, G. P. & Shepard, T. H. 1972. Neural tube closure defects caused by cytochalasin B. Nat. New. Biol. 236, 246247. Liu, W., Komiya, Y., Mezzacappa, C., Khadka, D. K., Runnels, L. & Habas, R. 2011. MIM regulates vertebrate neural tube closure. Development 138, 20352047. Lowery, L. A. & Sive, H. 2004. Strategies of vertebrate neurulation and a re-evaluation of teleost neural tube formation. Mech. Dev. 121, 11891197. Menzies, A. S., Aszodi, A., Williams, S. E., Pfeifer, A., Wehman, A. M., Goh, K. L., Mason, C. A., Fassler, R. & Gertler, F. B. 2004. Mena and vasodilator-stimulated phosphoprotein are required for multiple actin-dependent processes that shape the vertebrate nervous system. J. Neurosci. 24, 80298038. Messier, P. E. 1969. Effects of beta-mercaptoethanol on the ne structure of the neural plate cells of the chick embryo. J. Embryol. Exp. Morphol. 21, 309329. Morita, H., Nandadasa, S., Yamamoto, T. S., Terasaka-Iioka, C., Wylie, C. & Ueno, N. 2010. Nectin-2 and N-cadherin interact

2012 The Authors Development, Growth & Differentiation 2012 Japanese Society of Developmental Biologists

276

M. Suzuki et al.

through extracellular domains and induce apical accumulation of F-actin in apical constriction of Xenopus neural tube morphogenesis. Development 137, 13151325. Morita, H., Kajiura-Kobayashi, H., Takagi, C., Yamamoto, T. S., Nonaka, S. & Ueno, N. 2012. Cell movements of the deep layer of non-neural ectoderm underlie complete neural tube closure in Xenopus. Development 139, 14171426. Musch, A. 2004. Microtubule organization and function in epithelial cells. Trafc 5, 19. Myat, M. M., Chang, S., Rodriguez-Boulan, E. & Aderem, A. 1998. Identication of the basolateral targeting determinant of a peripheral membrane protein, MacMARCKS, in polarized cells. Curr. Biol. 8, 677683. Nakajima, H. & Tanoue, T. 2009. Epithelial cell shape is regulated by Lulu proteins via myosin-II. J. Cell Sci. 123, 555566. Nakajima, H. & Tanoue, T. 2011. Lulu2 regulates the circumferential actomyosin tensile system in epithelial cells through p114RhoGEF. J. Cell Biol. 195, 245261. Nandadasa, S., Tao, Q., Menon, N. R., Heasman, J. & Wylie, C. 2009. N- and E-cadherins in Xenopus are specically required in the neural and non-neural ectoderm, respectively, for F-actin assembly and morphogenetic movements. Development 136, 13271338. Nishimura, T. & Takeichi, M. 2008. Shroom3-mediated recruitment of Rho kinases to the apical cell junctions regulates epithelial and neuroepithelial planar remodeling. Development 135, 14931502. Pinson, L., Auge, J., Audollent, S., Mattei, G., Etchevers, H., Gigarel, N., Razavi, F., Lacombe, D., Odent, S., Le Merrer, M., Amiel, J., Munnich, A., Meroni, G., Lyonnet, S., Vekemans, M. & Attie-Bitach, T. 2004. Embryonic expression of the human MID1 gene and its mutations in Opitz syndrome. J. Med. Genet. 41, 381386. Plageman Jr, T. F., Chung, M. I., Lou, M., Smith, A. N., Hildebrand, J. D., Wallingford, J. B. & Lang, R. A. 2010. Pax6dependent Shroom3 expression regulates apical constriction during lens placode invagination. Development 137, 405 415. Plageman Jr, T. F., Chauhan, B. K., Yang, C., Jaudon, F., Shang, X., Zheng, Y., Lou, M., Debant, A., Hildebrand, J. D. & Lang, R. A. 2011a. A Trio-RhoA-Shroom3 pathway is required for apical constriction and epithelial invagination. Development 138, 51775188. Plageman Jr, T. F., Zacharias, A. L., Gage, P. J. & Lang, R. A. 2011b. Shroom3 and a Pitx2-N-cadherin pathway function cooperatively to generate asymmetric cell shape changes during gut morphogenesis. Dev. Biol. 357, 227234. Quaderi, N. A., Schweiger, S., Gaudenz, K., Franco, B., Rugarli, E. I., Berger, W., Feldman, G. J., Volta, M., Andol, G., Gilgenkrantz, S., Marion, R. W., Hennekam, R. C., Opitz, J. M., Muenke, M., Ropers, H. H. & Ballabio, A. 1997. Opitz G/BBB syndrome, a defect of midline development, is due to mutations in a new RING nger gene on Xp22. Nat. Genet. 17, 285291. Quintin, S., Gally, C. & Labouesse, M. 2008. Epithelial morphogenesis in embryos: asymmetries, motors and brakes. Trends Genet. 24, 221230. Radice, G. L., Rayburn, H., Matsunami, H., Knudsen, K. A., Takeichi, M. & Hynes, R. O. 1997. Developmental defects in mouse embryos lacking N-cadherin. Dev. Biol. 181, 6478. Roffers-Agarwal, J., Xanthos, J. B., Kragtorp, K. A. & Miller, J. R. 2008. Enabled (Xena) regulates neural plate morphogenesis, apical constriction, and cellular adhesion required for neural tube closure in Xenopus. Dev. Biol. 314, 393403.

Rolo, A., Skoglund, P. & Keller, R. 2009. Morphogenetic movements driving neural tube closure in Xenopus require myosin IIB. Dev. Biol. 327, 327338. Schoenwolf, G. C. & Franks, M. V. 1984. Quantitative analyses of changes in cell shapes during bending of the avian neural plate. Dev. Biol. 105, 257272. Schroeder, T. E. 1970. Neurulation in Xenopus laevis. An analysis and model based upon light and electron microscopy. J. Embryol. Exp. Morphol. 23, 427462. Short, K. M. & Cox, T. C. 2006. Subclassication of the RBCC/ TRIM superfamily reveals a novel motif necessary for microtubule binding. J. Biol. Chem. 281, 89708980. Stern, C. D. 2005. Neural induction: old problem, new ndings, yet more questions. Development 132, 20072021. Stumpo, D. J., Bock, C. B., Tuttle, J. S. & Blackshear, P. J. 1995. MARCKS deciency in mice leads to abnormal brain development and perinatal death. Proc. Natl Acad. Sci. USA 92, 944948. Suzuki, M., Hara, Y., Takagi, C., Yamamoto, T. S. & Ueno, N. 2010. MID1 and MID2 are required for Xenopus neural tube closure through the regulation of microtubule organization. Development 137, 23292339. Suzuki, S. C. & Takeichi, M. 2008. Cadherins in neuronal morphogenesis and function. Dev. Growth Differ. 50(Suppl. 1), S119S130. Takai, Y., Ikeda, W., Ogita, H. & Rikitake, Y. 2008. The immunoglobulin-like cell adhesion molecule nectin and its associated protein afadin. Annu. Rev. Cell Dev. Biol. 24, 309342. Tepass, U. 2009. FERM proteins in animal morphogenesis. Curr. Opin. Genet. Dev. 19, 357367. Terry, S. J., Zihni, C., Elbediwy, A., Vitiello, E., Leefa Chong San, I. V., Balda, M. S. & Matter, K. 2011. Spatially restricted activation of RhoA signalling at epithelial junctions by p114RhoGEF drives junction formation and morphogenesis. Nat. Cell Biol. 13, 159166. Waddington, C. H. & Perry, M. M. 1966. A note on the mechanisms of cell deformation in the neural folds of the Amphibia. Exp. Cell Res. 41, 691693. Wallingford, J. B. & Habas, R. 2005. The developmental biology of Dishevelled: an enigmatic protein governing cell fate and cell polarity. Development 132, 44214436. Wei, L., Roberts, W., Wang, L., Yamada, M., Zhang, S., Zhao, Z., Rivkees, S. A., Schwartz, R. J. & Imanaka-Yoshida, K. 2001. Rho kinases play an obligatory role in vertebrate embryonic organogenesis. Development 128, 29532962. Wu, M., Chen, D. F., Sasaoka, T. & Tonegawa, S. 1996. Neural tube defects and abnormal brain development in F52-decient mice. Proc. Natl Acad. Sci. USA 93, 21102115. Ybot-Gonzalez, P., Cogram, P., Gerrelli, D. & Copp, A. J. 2002. Sonic hedgehog and the molecular regulation of mouse neural tube closure. Development 129, 25072517. Ybot-Gonzalez, P., Gaston-Massuet, C., Girdler, G., Klingensmith, J., Arkell, R., Greene, N. D. & Copp, A. J. 2007a. Neural plate morphogenesis during mouse neurulation is regulated by antagonism of Bmp signalling. Development 134, 32033211. Ybot-Gonzalez, P., Savery, D., Gerrelli, D., Signore, M., Mitchell, C. E., Faux, C. H., Greene, N. D. & Copp, A. J. 2007b. Convergent extension, planar-cell-polarity signalling and initiation of mouse neural tube closure. Development 134, 789 799. Zolessi, F. R. & Arruti, C. 2001. Apical accumulation of MARCKS in neural plate cells during neurulation in the chick embryo. BMC Dev. Biol. 1, 7.

2012 The Authors Development, Growth & Differentiation 2012 Japanese Society of Developmental Biologists