Planar Polarity and Tissue Morphogenesis

Leading Edge

Review Planar Polarity and Tissue Morphogenesis Jennifer A. Zallen1,*

Developmental Biology Program, Sloan-Kettering Institute, New York, NY 10021, USA *Correspondence: [email protected] DOI 10.1016/j.cell.2007.05.050 1

Planar polarity is a global, tissue-level phenomenon that coordinates cell behavior in a two-dimensional plane. The Frizzled/planar cell polarity (PCP) and anterior-posterior (AP) patterning systems for planar polarity operate in a variety of cell types and provide direction to cells with different morphologies and behaviors. These two systems involve different sets of proteins but both use directional cues provided locally by communication between neighboring cells. This review describes our current understanding of the mechanisms that transmit directional signals from cell to cell and compares the strategies for generating global systems of spatial information in stationary and dynamic cell populations. Introduction Planar polarity is a property of multicellular tissues that demonstrate a coordination of cell behavior across a two-dimensional sheet, orthogonal to the axis of apical-basal polarity. This global property of tissue structure requires the establishment of asymmetry within cells and the alignment of these asymmetries in cells located dozens or hundreds of cells apart. Planar polarity applies to cells with diverse morphologies and behaviors and is readily apparent in the alignment of fish scales, bird feathers, and animal fur. In Drosophila, planar polarity is revealed by the precise arrangement of hairs and bristles on the surface of the adult animal: for example, each of the ?30,000 wing cells generates a single hair that forms at the distal surface of the cell and extends in a distal direction (Figure 1A). Each cell in a tissue could theoretically align with the body axes by navigating independently in response to a secreted cue from a distant source. However, two decades of research on planar cell polarity have demonstrated that the spatial information that organizes planar polarity is transmitted locally from one cell to the next. Input from neighboring cells can influence the behavior of individual cells as well as the orientation of groups of cells that respond as a unit to directional cues. Several different manifestations of planar polarity have been shown to occur through a common mechanism in which signaling by the Frizzled transmembrane receptor activates the conserved planar cell polarity (PCP) pathway (reviewed in Adler [2002]; Axelrod and McNeill [2002]; Mlodzik [2002]). Perhaps best known as one of several receptors for the Wnt family of secreted proteins (Logan and Nusse, 2004), Frizzled is believed to function in a Wnt-independent fashion in PCP (Lawrence et al., 2002).

Planar polarity is not limited to cells that are fixed in space. A striking example is elongation of the body axis during development, a process that involves the coordination of multiple cell behaviors including cell rearrangement, cell division, and cell-shape changes (Keller et al., 2000; Myers et al., 2002a; Wallingford et al., 2002). During tissue elongation, cells establish polarity in response to input from the cells in their immediate environment (Irvine and Wieschaus, 1994; Ninomiya et al., 2004). However, this environment is continually changing, indicating that planar polarity is dynamically remodeled on an ongoing basis. Polarized cell behavior during axis elongation requires the PCP pathway in vertebrates as well as an alternative, conserved system that derives spatial information from the organization of cell fates along the anterior-posterior (AP) axis (Irvine and Wieschaus, 1994; Ninomiya et al., 2004). The mechanisms that impose spatial organization on moving cells use distinct strategies to maintain the integrity of polarizing information in a rearranging population. The Frizzled/PCP and AP patterning systems for generating planar polarity operate in a variety of cell types and provide direction to cells with different morphologies and behaviors. This review describes the mechanisms that transmit directional signals locally between cells and compares the strategies for generating global systems of spatial information in stationary and dynamic cell populations. The PCP Pathway Provides Direction to Single Cells and Cell Groups Genetic and molecular studies in Drosophila have been instrumental in uncovering the Frizzled-dependent PCP pathway, a conserved mechanism for generating planar cell polarity in a range of cell and tissue types (Figure 1; reviewed in Adler [2002]; Axelrod and McNeill [2002]; Cell 129, June 15, 2007 ©2007 Elsevier Inc.  1051

Figure 1. Planar Polarity in Epithelia (A and B) Drosophila wing cells produce single cytoskeletal extensions that become the hairs of the adult wing. Wing hairs point distally (to the right) in wild-type animals (A) but are misoriented in frizzled mutants (B). (C and D) Hairs on the mouse paw point away from the body in Frizzled6 heterozygotes (C) but display a characteristic swirling pattern in Frizzled6 homozygotes (D). (E and F) Sensory receptor cells of the mouse cochlea form a crescent-shaped array of actin-rich stereocilia that face the cochlear lumen (E). This alignment is disrupted in mutants for the mammalian Strabismus homolog Vangl2 (top and bottom rows in [F]). Images courtesy of Paul Adler (A and B), Jeremy Nathans (C and D), and Matt Kelley (E and F). Images in (C) and (D) reproduced with permission from Guo et al. (2004), Proc. Natl. Acad. Sci. USA 101, 9277– 9281. Copyright 2004, National Academy of Sciences, USA.

Mlodzik [2002]). The core PCP pathway consists of the cell-surface proteins Frizzled, Strabismus (or van Gogh), and Flamingo (or Starry night) and the cytoplasmic proteins Dishevelled, Prickle, and Diego (Figure 2). In Drosophila, these proteins are most noticeably required for planar polarity in epithelial tissues that generate external structures of the adult, such as the hairs on the surface of the animal that point distally on the wing and posteriorly on the abdomen. This striking manifestation of planar polarity has facilitated the isolation of mutants that are defective for the global alignment of hairs and instead display characteristic swirling patterns (Figure 1B; Gubb and Garcia-Bellido, 1982; Wong and Adler, 1993). In mutants defective for the core PCP proteins, wing hairs initiate from a central location on the apical cell surface and subsequently point in inappropriate directions (Wong and Adler, 1993). In addition to directing the polarity of individual cells, the core PCP proteins can also influence the orientation of groups of cells. Mechanosensory bristles of the Drosophila adult are multicellular structures that arise from a single sensory organ precursor. The first division of the sensory organ precursor cell is reproducibly oriented parallel to the AP axis in a process that requires Frizzled, Dishevelled, and Flamingo (Gho and Schweis1052  Cell 129, June 15, 2007 ©2007 Elsevier Inc.

guth. 1998; Lu et al., 1999), suggesting a mechanism by which the polarity of a single cell can influence the organization of its descendants. In a variation on this theme, PCP proteins can also orient groups of postmitotic cells in the Drosophila eye. Ommatidia are composed of multiple cells in which the R3 and R4 photoreceptor cells differentiate in an asymmetric fashion and the entire structure rotates as a unit in opposite directions in the dorsal and ventral halves of the eye (Zheng et al., 1995; Wolff and Rubin, 1998). Oriented cell division and group-cell rotation therefore represent two mechanisms by which the PCP pathway regulates the organization of multicell assemblies. The role of Frizzled signaling in planar epithelial organization is conserved in vertebrates. The hair on the skin of the adult mouse normally grows in a posterior direction along the body surface (Figure 1C) but adopts aberrant patterns in mutants defective for the mouse Frizzled6 homolog (Figure 1D; Guo et al., 2004). Similarly, mechanosensory hair cells of the mouse cochlea extend a series of polarized actin-rich stereociliary bundles that point toward the fluid-filled lumen of the cochlea and respond to mechanosensory input (Figure 1E; reviewed in Dabdoub and Kelley [2005]). Cochlear cell polarity is disrupted in mice defective for mammalian homologs of Flamingo, Strabismus, Frizzled, and Dishevelled (Figure 1F; reviewed in Dabdoub and Kelley [2005]; Jones and Chen [2007]; Wang and Nathans [2007]). Several core PCP proteins are also required for elongation of the mammalian cochlear duct and elongation and closure of the mammalian neural tube, two processes of epithelial morphogenesis that involve planar polarity but whose cellular basis has not been determined (Jones and Chen, 2007; Wang and Nathans, 2007).

Figure 2. The Core Planar Cell Polarity Pathway in Drosophila A schematic wing epithelial cell is shown at early (A and B) and late (C) pupal stages. Apical is up, distal is to the right. (A) PCP proteins are initially present in the cytoplasm. (B) Frizzled and Strabismus are recruited to the apical surface by Flamingo and recruit Dishevelled and Prickle, respectively. (C) The core PCP proteins sort out into proximal and distal domains in a process that requires the activity of all six proteins. Flamingo, Strabismus, and Prickle localize to the proximal cell surface, and Flamingo, Frizzled, Dishevelled, and Diego localize to the distal surface. A proximodistally oriented microtubule network (gray lines in [C]) is required to deliver Frizzled to the distal surface. Flamingo (also known as Starry night) is an atypical cadherin with seven transmembrane domains. Frizzled is a seven-transmembrane protein, and Strabismus (also known as van Gogh) is a fourtransmembrane protein. Dishevelled is a cytoplasmic protein with DIX (Dishevelled-Axin), PDZ (PSD-95, Discs large, ZO-1), and DEP (Dishevelled, EGL-10, Pleckstrin) domains. Prickle (also known as Spiny legs) is a cytoplasmic protein with one PET domain and three LIM domains. Diego is a cytoplasmic protein with six ankyrin repeats.

Molecular and Cellular Mechanisms of PCP Signaling The presence of Frizzled protein, even above a threshold level, is not sufficient for a cell to achieve the correct orientation. Instead, it is local differences in Frizzled activity between neighboring cells that provide the directional information required for planar polarity. These observations have led to a longstanding model in which a gradient of Frizzled activity, perhaps in response to a graded ligand, determines the direction of planar polarity. An alternative model proposes that cells do not respond to a global gradient but instead rely on the local head-totail alignment of neighboring cells through an asymmetric distribution of core PCP proteins. These two models make different predictions about the role of the global spatial cues that act either at the tissue level to establish a gradient of Frizzled activity or at the cellular level to determine the site of Frizzled localization. Relative Frizzled Activity Is Important for Planar Polarity It is widely accepted that differences in the activity of the Frizzled transmembrane protein in adjacent cells provide instructive information that is capable of generating planar polarity. In a classic experiment, an ectopic gradient of Frizzled expression—created by dripping hot wax on Drosophila pupae expressing a heat-shock-inducible frizzled transgene—was shown to redirect wing hairs to point away from the site of highest Frizzled activity (Adler et al., 1997). This result, together with observations in other tissues, led to the hypothesis that a gradient of Frizzled activity provides the global spatial information that establishes the direction of planar cell polarity (Zheng et al., 1995; Adler et al., 1997; Lawrence et al., 2004). For example, wild-type wing hairs could orient distally by setting the site of hair formation to point

down the Frizzled gradient. However, direct evidence for a gradient of PCP protein expression or activity has not been reported. Other lines of evidence demonstrate that planar polarity is influenced by local interactions between cells. When clones of frizzled mutant cells are generated in otherwise wild-type tissue, the hairs made by wild-type cells near the clone point toward the frizzled mutant cells (Figure 3B; Vinson and Adler, 1987; Lawrence et al., 2004). Conversely, wild-type hairs point away from mutant cells that lack the Strabismus transmembrane protein (Figure 3C; Taylor et al., 1998). This unusual behavior of wild-type cells around frizzled or strabismus mutant clones can appear several rows of cells away from the original site where differences occur. The reorientation is blocked if the cells either inside or outside of the clone are defective for Flamingo (Chae et al., 1999; Lawrence et al., 2004; Casal et al., 2006), demonstrating that the Flamingo cadherin is required for cells to both send and receive polarizing information. By contrast, mutant clones defective for the Dishevelled or Prickle cytoplasmic proteins primarily affect cells within the clone (Figure 3A; Klingensmith et al., 1994; Theisen et al., 1994; Gubb et al., 1999), indicating that these proteins are required to respond to directional signals, but not to send them. PCP Proteins Are Often Asymmetrically Localized Planar polarized hair formation in the Drosophila wing is accompanied by an underlying molecular polarity that includes an asymmetric distribution of core PCP proteins. The core PCP proteins are first recruited to the apical cell surface and subsequently segregate into complementary apical subdomains before the onset of hair formation (Figure 2). Flamingo localizes to proximal and distal surfaces (Usui et al., 1999; Shimada et al., Cell 129, June 15, 2007 ©2007 Elsevier Inc.  1053

Figure 3. Local Cell Interactions Determine Wing Hair Direction Proximal is to the left, distal is to the right. (A) The distal orientation of wild-type Drosophila wing hairs is unaffected by a nearby clone of prickle mutant cells. (B) Wild-type cells distal to a frizzled mutant clone reorient their hairs to point backward toward the clone. (C) Wild-type cells proximal to a strabismus mutant clone reorient their hairs to point away from the clone. (D) Dsh:GFP is distributed in a characteristic zigzag pattern, running vertically in the figure. (E) Mosaic GFP expression reveals that Dsh:GFP localizes to distal surfaces (blue arrows) and not to proximal surfaces (red arrows). Wild-type cells bordering a clone lacking Dsh:GFP are indicated by yellow dots. (F) Endogenous Prickle localizes to proximal and not distal surfaces in wild-type cells next to a prickle mutant clone. Wild-type cells bordering a clone lacking Prickle are indicated by yellow dots. (G–J) Two models to explain how directional information is propagated from cell to cell. These schematic diagrams do not indicate the true localization of the core PCP proteins, which are concentrated at the apical surface. In the feedback model (G and H), wild-type cells concentrate Frizzled (blue) at distal cell surfaces and Strabismus (red) at proximal surfaces (G). These distributions are maintained by cooperative interactions between Frizzled and Strabismus in neighboring cells and antagonistic interactions within each cell. In a frizzled mutant clone (two central cells in [H]), Strabismus (red) is uniformly distributed at the cell surface and recruits Frizzled (blue) to the surface of wild-type cells facing the clone. As a result, the Frizzled-Strabismus feedback loop is reversed in cells distal to the clone, and this altered polarity propagates to cells further from the clone border. In the averaging model (I and J), a high proximal→low distal gradient of Frizzled activity (blue) is maintained by a mechanism in which Frizzled activity in each cell is set as an average of its neighbors (I). In the presence of a clone of cells lacking frizzled (two central cells in [J]), the averaging process instructs wild-type cells to reduce their Frizzled activity, causing cells distal to the clone to reorient their hairs to point down the local Frizzled gradient. Images courtesy of Paul Adler (A–C) and Jeff Axelrod (D–F).

2001), whereas Frizzled, Dishevelled, and Diego localize specifically to the distal surface (Axelrod, 2001; Strutt, 2001; Das et al., 2004) and Prickle and Strabismus localize to the proximal surface (Tree et al., 2002; Bastock et al., 2003). Frizzled can interact directly with Dishevelled (Wong et al., 2003), and Strabismus can associate with Prickle (Bastock et al., 2003; Jenny et al., 2003), indicating that proximal and distal cell domains consist of at least two protein complexes (Figure 2). The Flamingo cadherin, which is capable of mediating homophilic adhesion (Usui et al., 1999), recruits the other core PCP proteins to the region of the adherens junctions (Strutt, 2002). Once at the surface, the activity of all six core PCP proteins is required for any of them to achieve a planar polarized distribution (Strutt, 2002), indicating that these proteins participate in a regulatory loop rather than a strict linear pathway. Asymmetric PCP protein localization is observed in several epithelial tissues in which core PCP proteins have been shown to function, including the R3/R4 boundary in the Drosophila eye, the dividing bristle precursor cell, and the mouse cochlear epithelium (reviewed in Jones and Chen [2007]; Seifert and Mlodzik [2007]; Wang and Nathans [2007]). These observations suggest that polarized protein distribution is a key property of PCP function in epithelia. However, some cells can retain planar polarity in the absence of detectable molecular asymmetries (Strutt and Strutt, 2007). In a link between cell and tissue organization, the asymmetric localization of PCP proteins is sensitive not 1054  Cell 129, June 15, 2007 ©2007 Elsevier Inc.

only to the activity of PCP proteins in the same cell but also to PCP activity in adjacent cells. Frizzled activity in one cell is required for Prickle localization in its distal neighbor, whereas Prickle is required for the localization of Dishevelled in the adjacent proximal cell, suggesting that Frizzled and Prickle interact indirectly across cell boundaries (Tree et al., 2002). Within a cell, there is evidence that Prickle can block the association between Frizzled and Dishevelled (Tree et al., 2002), although other studies find that Prickle overexpression does not disrupt Dishevelled membrane localization (Bastock et al., 2003). Conversely, the Diego protein can associate with Prickle and Strabismus (Das et al., 2004) and may counteract Prickle activity to allow Frizzled-Dishevelled complexes to form at the distal surface (Jenny et al., 2005). These interactions suggest a mechanism by which proximal Strabismus-Prickle complexes and distal Frizzled-Dishevelled complexes form in mutually exclusive cellular domains. Several PCP proteins that interact in vitro are found in different parts of the cell (Tree et al., 2002; Bastock et al., 2003; Das et al., 2004; Jenny et al., 2005), raising the question of whether these interactions occur in vivo. These proteins could come into contact when they are first recruited to the apical surface of wing cells during the establishment of polarity or in mesenchymal cells in which PCP protein localization is not obviously exclusive (see below). Alternatively, there may be a dynamic and ongoing antagonism between PCP proteins at opposing cell surfaces that serves to maintain polarized protein localization.

How do PCP proteins become asymmetrically localized? In the early Drosophila wing, microtubules have been shown to align preferentially along the proximaldistal axis, parallel to the apical cell surface (Eaton et al., 1996; Turner and Adler, 1998; Shimada et al., 2006). Live-imaging studies reveal that Frizzled:GFP is transported in a distally biased fashion in vesicles that contain Flamingo (Shimada et al., 2006), reminiscent of the microtubule-dependent motility of Dishevelled vesicles in the Xenopus egg (Miller et al., 1999). Flamingo vesicles are found in association with microtubules, and microtubule disorganization results in the loss of Frizzled and Flamingo from the surface and a disruption of Frizzled: GFP transport (Shimada et al., 2006). These results suggest that planar polarity may be established by the directional transport of Frizzled to the distal cell surface along a polarized microtubule network. Frizzled Disparities between or within Cells? New Models for Planar Cell Polarity The polarized distribution of PCP proteins in epithelial tissues has given rise to a model in which heterophilic protein interactions participate in a feedback loop that propagates molecular polarities from cell to cell (Tree et al., 2002; Amonlirdviman et al., 2005). In this “feedback” mechanism (Figures 3G and 3H), planar polarity could occur through an asymmetric distribution of Frizzled in cells with comparable levels of Frizzled activity. Frizzled accumulation at the distal surface of a cell would recruit or retain Strabismus on the proximal surface of the adjacent cell, stabilizing their complementary distributions. Strabismus-Prickle and Frizzled-Dishevelled complexes could then segregate into reciprocal domains in both cells through an antagonistic effect of the proximal Prickle protein on Frizzled-Dishevelled association. As a result, Frizzled is predicted to accumulate on the distal surface of the neighboring cell, and the cycle would continue. Importantly, a mathematical model of this feedback loop based on interactions among four core PCP proteins—Frizzled, Dishevelled, Prickle, and Strabismus—can reproduce both wild-type planar polarity and complex patterns that arise in mosaic situations (Amonlirdviman et al., 2005). These simulations demonstrate that, given a distal bias in Frizzled localization, local interactions among PCP proteins could provide a robust mechanism for generating planar cell polarity in the absence of a Frizzled gradient. Although the feedback model dispenses with the requirement for a gradient of Frizzled activity, what this model leaves unexplained is the nature of the initial input that biases Frizzled localization to determine the direction of planar polarity. An alternative model set out to address this issue by proposing that a single mechanism—a gradient of Frizzled activity across the population—accounts for both local interactions between cells and global tissue organization. This model is based on the hypothesis that Frizzled acts as a receptor for a graded extracellular ligand (Vinson and Adler, 1987; Adler et al., 1997; Lawrence et al., 2004). A global gradient of a Frizzled

activator, for example, would provide one way to create differences in Frizzled activity between cells. In its original form, however, this model did not explain how cells are able to reorient at a distance from a local disparity in Frizzled activity, such as at the boundary of a mosaic clone. A potential solution to this problem is provided by a variation of the gradient model in which cells respond to a Frizzled gradient by adjusting their Frizzled activity to equal an average of their neighbors (Lawrence et al., 2004). In wild-type animals (Figure 3I), this “averaging” mechanism would reinforce the global gradient in response to local variations in cell number and position. In a mosaic situation (Figure 3J), wild-type cells near a frizzled mutant clone would reduce their levels of Frizzled signaling in response to the clone, introducing a new difference between these cells and cells further from the clone border. An averaging mechanism could generate differences in Frizzled activity that propagate for several rows of cells away from a frizzled mutant clone in a novel mechanism for translating tissue-level information into single-cell polarity. The dust has not settled on the debate between averaging and feedback models for propagating directional information across a field of cells. It is an open question whether a global gradient of PCP activity or differences in the composition of PCP complexes at adjacent cell surfaces enable a cell to align with its neighbors. Both models are plausible but currently incomplete and require components that have not been identified. Although an intercellular interaction between Frizzled and Strabismus provides the basis for cell communication in the feedback model, direct interactions between these proteins have not been reported. Flamingo, a homophilic adhesion protein (Usui et al., 1999), has been proposed to bridge interactions between PCP proteins across cells (Lawrence et al., 2004; Klein and Mlodzik, 2005; Le Garrec et al., 2006), but it has not been shown to bind directly to any of the PCP proteins. In addition, Prickle is not required for the transmission of planar polarity signals in some cases (Adler et al., 2000; Lawrence et al., 2004; Strutt and Strutt, 2007), indicating that other proteins contribute to the feedback loop (Amonlirdviman et al., 2005). Similarly, the mechanisms proposed to detect and respond to local differences in Frizzled activity in the averaging model have not been defined, and no proof of the existence of a gradient of Frizzled activity, long sought in the field, has been obtained. Ultimately, a comprehensive description of the molecular mechanisms that communicate directional information from cell to cell will be essential to understanding the processes that organize cell and tissue polarity. Gradients of Surface Proteins Set the Direction of Polarity Defining the nature of the elusive spatial input that sets the global direction of planar polarity would constitute an important step toward understanding the downstream pathways that propagate directional information between cells. Components of such global spatial Cell 129, June 15, 2007 ©2007 Elsevier Inc.  1055

Figure 4. Polarized Cell Intercalation and Tissue Elongation (A–D) Schematic diagram of intercalating mesenchymal cells in the Xenopus dorsal mesoderm. Anterior is to the left, medial is down. (A) Mesenchymal cells initially extend protrusions in random directions. (B) At gastrulation, these cells elongate along the mediolateral axis and extend lamellipodial protrusions from their medial and lateral surfaces. (C) Cell intercalation leads to a narrowing of the tissue along the mediolateral axis (convergence) and a lengthening of the tissue along the anterior-posterior (AP) axis (extension). (D) Intercalary behavior is induced by the juxtaposition of different cell types along the AP axis, perhaps through global gradients of spatial information such as the chordin BMP antagonist (gray) and the Brachyury transcription factor (green). (E–H) Intercalation of epithelial cells in the Drosophila germband ectoderm. Anterior is to the left, ventral is down. (E–G) Cell outlines visualized with a GFP marker in a time-lapse movie shown at 10 min intervals. Four linked interfaces perpendicular to the AP axis (black lines circumscribed by red cells in [E]) contract together to produce a seven-cell rosette (F) that resolves parallel to the AP axis (red cells in [G], black lines indicate new interfaces), promoting elongation. Nearby rosettes of six to seven cells are indicated. (H) Overlapping stripes of pair-rule gene expression provide spatial information necessary for rosette formation and axis elongation (each stripe is about three to four cells wide), shown for the Eve (gray) and Runt (green) transcription factors. Schematic diagrams in (A)–(C) adapted from Keller et al. (2000). Images in (E)–(G) courtesy of Ori Weitz.

systems are predicted to be expressed or activated in a precise spatial pattern. Four-jointed, a Golgi protein, and Dachsous, a large atypical cadherin with 27 extracellular cadherin repeats, are required for planar polarity in Drosophila and are expressed in complementary gradients in the wing, eye, and abdomen—exactly what would be expected for a global spatial cue (Strutt and Strutt, 2005). These proteins act genetically upstream of the uniformly expressed Fat protein, another atypical cadherin with 34 extracellular cadherin repeats, and are proposed to generate a gradient of Fat activity (Yang et al., 2002). Reversing the direction of the Dachsous and Four-jointed gradients is sufficient to reverse the direction of ommatidial polarity, and flattening both gradients randomizes polarity, demonstrating that these components provide critical spatial information in the eye (Simon, 2004). Genetic interactions indicate that Fat acts upstream of Frizzled signaling (Strutt and Strutt, 2005). In one model, Fat could regulate Frizzled signaling to create a gradient of Frizzled activity. Alternatively, Fat could be required for the formation of a polarized microtubule network or for Frizzled localization to the distal cell surface. The Fat/Dachsous/Four-jointed pathway is an ideal candidate for providing spatial input into the core PCP pathway, but several observations are not consistent with this view. Although reversing the Dachsous and Fourjointed gradients reverses the direction of wing hair polarity, flattening these gradients has no effect (Zeidler et al., 2000; Matakatsu and Blair, 2004; Simon, 2004), indicating that Dachsous and Four-jointed do not provide critical spatial information in the wing, although they may function redundantly with other cues. Moreover, a molecular link between Fat activity and Frizzled signaling has not been identified. Despite the temptation to propose a role for the conserved Fat extracellular domain in cell-cell commu1056  Cell 129, June 15, 2007 ©2007 Elsevier Inc.

nication, Fat protein lacking its extracellular domain has been reported to rescue planar polarity when expressed in fat mutant cells (Matakatsu and Blair, 2006). However, other studies observe no effect of ectopic Fat expression in the absence of its extracellular domain (Casal et al., 2006). Interestingly, genetic experiments in the Drosophila abdomen indicate that Fat and Dachsous can act independently of the Frizzled pathway (Casal et al., 2006). Clones of cells that overexpress Fat or the Dachsous extracellular domain can trigger a wave of repolarization in cells that lack Frizzled (Casal et al., 2006), providing evidence for a polarity mechanism that does not require Frizzled activity. Moreover, disruption of Flamingo activity enhances the planar polarity defects in animals lacking Dachsous (Casal et al., 2006), indicating that Flamingo has functions that do not require Dachsous. These studies indicate that the Fat/Dachsous/Four-jointed system does not merely act as an input into the Frizzled pathway but can provide a distinct source of spatial information in the establishment of planar polarity. Planar Polarity on the Move: Cell Rearrangement during Tissue Morphogenesis In addition to organizing polarity in nonmotile tissues where cells maintain contact with a stable cohort of neighbors, planar polarity systems can also operate in motile populations experiencing dynamic morphogenetic changes. Elongation of the body axis is a conserved developmental process that arises when a systematic orientation is imposed on processes of cell motility, cell division, cell-shape changes, or cell rearrangement. These polarized behaviors can take directional cues from the classical PCP system as well as a novel mechanism that uses spatial information provided by the pattern of cell fate specification along the AP axis. Like PCP, planar polarity in rearranging tissues is determined by

local cell interactions, but unlike PCP, the neighborhood of each cell is constantly in flux. These features present a challenge to traditional planar polarity systems, which use a range of strategies to allow moving cells to reorient in response to a changing cellular environment. The PCP Pathway Is Required for Polarized Cell Motility A principal mechanism for axis elongation in vertebrates and invertebrates is cell intercalation, in which cells insert between their neighbors along one axis, causing the tissue to elongate in a perpendicular direction (Figure 4; Keller et al., 2000; Myers et al., 2002a; Wallingford et al., 2002). Intercalation occurs in mesenchymal cells during elongation of the dorsal mesoderm and posterior neural ectoderm in Xenopus (Shih and Keller, 1992; Elul and Keller, 2000) and the dorsal mesoderm and ectoderm in zebrafish (Concha and Adams, 1998; Myers et al., 2002b; Glickman et al., 2003). In contrast to epithelial cells, these mesenchymal cells lack formal adherens junctions and apicalbasal polarity but are still able to orient with respect to the plane of the tissue. The cellular basis of intercalation is best understood in Xenopus, in which cells extend polarized lamellipodial protrusions in the direction of cell movement (Figure 4B; Shih and Keller, 1992; Elul and Keller, 2000). These dynamic protrusions may generate tractile forces that enable cells to actively crawl over neighboring cells or an extracellular matrix (ECM) substrate (Keller et al., 2000). Axis elongation in Xenopus and zebrafish requires the transmembrane PCP proteins Frizzled, Strabismus, and Flamingo and the cytoplasmic PCP proteins Dishevelled and Prickle (Keller, 2002; Myers et al., 2002a; Wallingford et al., 2002). Several PCP proteins are also required for elongation of the mammalian cochlea and elongation and closure of the neural tube (reviewed in Jones and Chen [2007]; Wang and Nathans [2007]), but intercalation is independent of some components of the core PCP mechanism in Drosophila (Zallen and Wieschaus, 2004). The dynamic reorganization of mesenchymal cells in elongating vertebrate tissues requires some of the same proteins that generate planar polarity in Drosophila epithelial cells, but several lines of evidence suggest key differences in the way the PCP pathway functions in these two cell types. First, polarized motility in mesenchymal cells requires the Wnt11 and Wnt5 ligands for the Frizzled receptor (Heisenberg et al., 2000; Tada and Smith, 2000; Kilian et al., 2003), whereas epithelial planar polarity appears to be independent of Wnt-family proteins in Drosophila (Lawrence et al., 2002). These extracellular ligands, when they do act, may not provide critical spatial information: uniform expression of Wnt11 or Wnt5 can restore axis elongation in zebrafish silberblick/Wnt11 mutants, and overexpression of a truncated Dishevelled protein can bypass the requirement for Wnt11 activity (Heisenberg et al., 2000; Tada and Smith, 2000; Kilian et al., 2003). Second, although PCP proteins are asymmetrically distributed in nonmotile epithelial tissues such as the Drosophila wing, eye, and notum and the mammalian cochlea (Jones and Chen, 2007; Seifert and Mlodzik, 2007; Wang and Nathans, 2007), the evidence for

polarized PCP protein localization in intercalating cells is less clear. Dishevelled localizes to the surface of intercalating mesenchymal cells in Xenopus in a punctate pattern that does not display obvious planar asymmetry (Wallingford et al., 2000), and mammalian Dishevelled2 localizes uniformly to the surface of epithelial cells in the elongating neural plate (Wang et al., 2006). Prickle and Dishevelled are excluded from the notochord-muscle boundary in ascidians but do not display a polarized localization at interfaces between intercalating cells (Jiang et al., 2005). At later stages, Prickle becomes anteriorly enriched in cells of the ascidian notochord (Jiang et al., 2005) and the zebrafish neural keel (Ciruna et al., 2006). Another feature that distinguishes PCP activity in motile and nonmotile cell types is an unexpected role for PCP proteins in organizing the fibronectin-rich ECM that surrounds the Xenopus notochord (Goto et al., 2005). This ECM substrate is required for the intercalation of notochord cells (Marsden and DeSimone, 2003; Davidson et al., 2006), where it may restrict protrusive activity to a twodimensional plane as an alternative to epithelial apicalbasal polarity. Disruption of Frizzled, Strabismus, or Prickle activity interferes with the spatial organization of the ECM, causing it to interdigitate inappropriately between intercalating cells rather than specifically accumulating at the tissue boundary (Goto et al., 2005). Surprisingly, an artificial fibronectin substrate can restore polarized cell motility to Prickle-overexpressing cells (Goto et al., 2005), indicating that ECM organization is a primary output of Prickle activity. However, an ectopic substrate cannot rescue cells that overexpress Frizzled or Strabismus (Goto et al., 2005), indicating that these PCP proteins also affect other processes important for intercalation. Polarized Cell Behavior in Epithelial Morphogenesis During axis elongation in the Drosophila germband epithelium, cells change position while remaining interconnected by a network of adherens junctions that maintains the continuity of the epithelial sheet. These cell rearrangements are accompanied by a polarized localization of proteins that accumulate in the vicinity of the adherens junctions and are required for effective elongation (Bertet et al., 2004; Zallen and Wieschaus, 2004), reminiscent of the asymmetric distribution of core PCP proteins. Myosin II and filamentous actin are concentrated at interfaces between anterior and posterior germband cells, and the adherens junction proteins DE-cadherin, Armadillo/β-catenin, and Bazooka/ PAR-3 are enriched in the reciprocal dorsal-ventral interfaces (Bertet et al., 2004; Zallen and Wieschaus, 2004; Blankenship et al., 2006). Unlike core PCP components, polarized proteins in the Drosophila germband have established cell biological activities that link them directly to cell behavior. Actin-myosin contractility can generate dramatic changes in cell shape, and myosin II is capable of destabilizing intercellular adhesion, whereas Bazooka/PAR3 promotes adherens junction stability and localization (Sahai and Marshall, 2002; Harris and Peifer, 2004; Chen and Macara, 2005). Although myosin II activation by Rho kinase acts as an effector of the PCP system to determine Cell 129, June 15, 2007 ©2007 Elsevier Inc.  1057

the number of wing hairs downstream of Frizzled signaling (Winter et al., 2001), myosin II polarization in Drosophila is an early step in the establishment of planar polarity (Blankenship et al., 2006). The planar polarized distribution of contractile actinmyosin structures is likely to generate the forces that produce changes in cell shape and interactions during intercalation. Consistent with this possibility, interfaces that accumulate myosin II are preferentially eliminated during axis elongation (Bertet et al., 2004). However, the extrapolation of single-interface behaviors to the complex dynamics of cell populations does not turn out to be straightforward. Germband cells continually vary the number and nature of neighbors in their immediate environment and participate in a wide range of cell interactions (Zallen and Zallen, 2004; Blankenship et al., 2006). Live-imaging studies were instrumental in revealing the logic behind this cellular disorder, in which germband cells organize locally to generate multicellular rosette structures that form and resolve in a directional fashion (Figures 4E–4G; Blankenship et al., 2006). During rosette formation, two columns of cells constrict their shared interfaces in tandem (Figure 4E), bringing as many as 11 cells into contact at a single point (Figure 4F). These multicellular rosette structures rapidly resolve in a perpendicular direction to produce a pronounced elongation of the multicellular array (Figure 4G). Rosette behaviors appear to provide an efficient mechanism for intercalation: groups of cells rearrange locally to elongate along the AP axis, and the sum of these behaviors across the germband results in net extension of the entire tissue. A majority of germband cells participate in multiple rounds of rosette formation that, together with local neighbor exchange (Bertet et al., 2004), can account for most of the total change in germband morphology (Blankenship et al., 2006). Although neighbor exchange may be sufficient for elongation in some cases (Glickman et al., 2003), higher-order rosette behaviors could ensure that the necessary amount of elongation is achieved in a limited time window. Rosette structures may reflect the outcome of a coordinated contractile network that spans multiple cells, reminiscent of the actin-myosin cables that contribute to epithelial advance in Drosophila and zebrafish (Franke et al., 2005; Koppen et al., 2006). It will be informative to see whether live-imaging approaches reveal higher-order rosette behaviors in other examples of epithelial reorganization, such as morphogenesis of the mammalian neural tube. Anterior-Posterior Patterning Systems Direct Polarized Cell Behavior Cell intercalation in Xenopus and Drosophila is organized in a two-dimensional plane by spatial cues provided by the arrangement of different cell fates along the AP axis, a system that is distinct from the core PCP pathway (Irvine and Wieschaus, 1994; Lengyel and Iwaki, 2002; Ninomiya et al., 2004). This alternative mechanism for establishing planar polarity was originally suggested by the observation that striped expression of pair-rule 1058  Cell 129, June 15, 2007 ©2007 Elsevier Inc.

AP patterning genes is necessary for axis elongation in Drosophila (Figure 4H; Irvine and Wieschaus, 1994). Embryos that lack AP pattern display severe defects in axis elongation and rosette formation (Irvine and Wieschaus, 1994; Blankenship et al., 2006), and striped pair-rule gene expression is required for the planar polarized distribution of cytoskeletal and junctional proteins (Zallen and Wieschaus, 2004; Blankenship et al., 2006). An ectopic source of Eve or Runt is capable of recruiting an AP marker to interfaces between cells with different levels of Eve or Runt expression, regardless of their orientation (Zallen and Wieschaus, 2004). These results suggest that differences in cell fate along the AP axis determine the length and orientation of actin-myosin cables during axis elongation. AP patterning systems also generate planar polarity during axis elongation in vertebrates. The juxtaposition of explants from different AP locations along the Xenopus notochord is sufficient to generate intercalary behavior in culture, whereas combining explants from similar AP positions has no effect (Ninomiya et al., 2004). AP pattern in Xenopus acts upstream or independently of PCP signaling, as Frizzled-defective cells can instruct neighboring cells to intercalate but fail to respond to AP signals (Ninomiya et al., 2004). Although AP pattern is required for the polarized localization of adherens junction proteins in Drosophila (Zallen and Wieschaus, 2004; Blankenship et al., 2006), intercalating mesenchymal cells in the Xenopus embryo lack defined adherens junctions and display a distinct set of cell behaviors (Keller et al., 2000). Despite these differences, the Bazooka/PAR-3 binding proteins PAR-6 and atypical protein kinase C are concentrated at sites of lamellipodial formation in the Xenopus notochord along with the XGAP Arf GTPase regulator (Hyodo-Miura et al., 2006), and knockdown experiments demonstrate that XGAP is required for polarized protrusive activity (Hyodo-Miura et al., 2006). Identification of the target genes that mediate the effects of AP pattern on cell polarity and behavior will reveal the extent to which these spatial systems are conserved between vertebrates and invertebrates. Generating and Orienting Asymmetry along Planar Axes Based on current knowledge, the AP patterning system appears to be dedicated to the organization of moving cells, whereas the PCP pathway often (but not always) functions in static epithelial tissues in which neighbor relationships remain constant. Several factors may influence the strategy for generating planar polarity in different tissue contexts. First, cell movements can occur on a faster timescale than the polarization of stationary cells: planar polarity in the Drosophila wing is established over several hours, whereas intercalation in the Drosophila germband is completed in 30–40 min. Second, planar polarity in nonmotile tissues displays a remarkable precision that is often relevant to tissue function, but dynamic behaviors during morphogenesis may tolerate a greater degree of variability. Third, whereas the core PCP mechanism acts as an intermediary to mediate a

Figure 5. Oriented Cell Divisions Shape Tissue Morphology (A) Schematic diagram of a cubic cell dividing to give rise to two equal daughter cells. The epithelial layer is drawn parallel to the plane of the page. For clarity, daughter cells are shown slightly separated. Top right: When the daughter cells are cubes, the elongation factor is 1.6 (twice the cube root of one-half). This factor is independent of cell shape as long as the total cell volume is conserved and daughter cells are equal in size and have the same shape and orientation as the original cell. Bottom right: A dividing cell constrained to a fixed height (edges in red are the same length). The square face (apical surface) of each daughter cell has half the area of the square face of the original cell, producing an elongation factor of 1.4. (B and C) Oriented cell divisions account for ?25% of the total elongation during epiboly in the wild-type zebrafish dorsal ectoderm (B), and misoriented divisions correlate with reduced elongation in embryos that overexpress a mutant Dishevelled protein (C). Schematic diagrams in (A) courtesy of Richard Zallen. Images in (B) and (C) reprinted by permission from Macmillan Publishers Ltd.: Nature, Gong et al. (2004), 430, 689–693, copyright 2004.

range of tissue-specific cellular responses, AP patterning cues may communicate directly with the effector proteins that generate cell behavior in a mechanism that favors speed and adaptability over precision. A further distinction between planar polarity systems in static and dynamic tissues is suggested by evidence that intercalating cells use planar polarity cues to generate asymmetry, in contrast to planar polarity systems in nonmotile tissues that in some cases provide direction to cells that already possess inherent asymmetry. Xenopus mesodermal cells that overexpress Strabismus, Dishevelled, or dominant-negative Dishevelled fail to elongate or form polarized lamellipodia (Wallingford et al., 2000; Goto and Keller, 2002), and zebrafish ectodermal cells mutant for Pipetail/Wnt5 do not acquire an elongated morphology (Kilian et al., 2003), consistent with an absence of planar polarity in these cells. Similarly, aspects of polarized protein localization are eliminated in Drosophila AP patterning mutants (Zallen and Wieschaus, 2004; Blankenship et al., 2006). By contrast, planar polarity systems in some epithelial tissues act on cells that are already asymmetric but require the PCP pathway for their orientation. For example, sensory organ precursor cells divide along random axes in PCP mutants but segregate the Numb protein to a single daughter cell (Gho and Schweisguth, 1998; Lu et al., 1999), ommatidial clusters in frizzled mutants specify distinct R3 and R4 photoreceptor cell fates in a random fashion 89% of the time (Zheng et al., 1995), and misoriented cochlear cells in mouse PCP mutants elaborate a polarized stereociliary array (Curtin et al., 2003; Montcouquiol et al., 2003). These results are consistent with the interpretation that moving cells use polarity cues to orient and generate polarity simultaneously, whereas stationary cells in some cases use the PCP pathway solely for their orientation. An exception is the Drosophila wing, in which wing hairs grow from a central location on the apical surface in core PCP mutants, indicating a loss of planar polarity (Wong and Adler, 1993). A dual require-

ment for extracellular cues in generating and orienting planar polarity would provide an elegant mechanism to allow moving cells to dynamically remodel their polarity in response to changes in their local environment. Beyond PCP: New Outputs of Planar Polarity Systems Planar polarity is a property of many tissues that reflects the organization of different cellular processes and behaviors. Recent studies indicate that familiar polarized proteins can contribute to new features of tissue organization. Some aspects of two-dimensional tissue structure are independent of the core PCP proteins, indicating that additional systems for generating planar polarity remain to be discovered. Polarized Cell Division Influences Tissue Structure One way to regulate the organization of cell populations is through the division of one cell into two: without a change in volume, this process can theoretically produce a 60% elongation parallel to the spindle long axis (Figure 5A; Keller, 2006). The systematic alignment of mitotic spindles in dividing populations could therefore lead to significant tissue elongation. Cell division in the zebrafish dorsal ectoderm is oriented parallel to the animal-vegetal axis at gastrulation (Figure 5B; Concha and Adams, 1998; Gong et al., 2004), and misoriented divisions are associated with a failure to elongate the body axis in embryos with disrupted Dishevelled, Wnt11/Silberblick, or Strabismus activity (Figure 5C; Gong et al., 2004). Oriented cell divisions also contribute to elongation of the Drosophila wing (Baena-Lopez et al., 2005) and the mammalian kidney (Fischer et al., 2006). In zebrafish, the Strabismus homolog Trilobite is required for the allocation of daughter cells to opposing sides of the neural tube following division (Ciruna et al., 2006). This process involves an unusual cell division that produces daughter cells with mirror-image symmetry through the polarized localization of a zebrafish PAR-3 homolog (Tawk et al., 2007). Inhibiting cell division resCell 129, June 15, 2007 ©2007 Elsevier Inc.  1059

cues the neural tube defects of trilobite mutants (Ciruna et al., 2006), demonstrating that the sole function of PCP signaling in this context is to maintain tissue structure during division. Tissue Organization and the Reversal of Disorder Epithelial populations display a trend toward increased disorder in tissues undergoing cell intercalation or division (Zallen and Zallen, 2004; Gibson et al., 2006), but this tendency is not inevitable. In the Drosophila wing, cells transition from a topologically disordered state to an ordered pattern of predominantly hexagonal cells aligned with two edges oriented parallel to the proximal-distal axis (Classen et al., 2005). These observations reveal a novel mechanism of tissue organization in which cells can regulate their neighbor relationships and align as a population with the tissue axes to achieve a highly ordered pattern unusual for epithelia. Establishment of an ordered cellular array, sometimes presumed to be a low-energy default state, instead seems to reflect an active process that in the Drosophila wing requires PCP proteins, adherens junction components, and the vesicular trafficking machinery (Classen et al., 2005). Mutations in PCP proteins do not completely block the ordering process, demonstrating that additional pathways contribute to planar organization in the developing wing. A distinct property of tissue organization is observed in the epithelial cells that generate planar polarized denticle structures, the actin- and microtubule-rich protrusions that line the Drosophila larval cuticle. Cells that produce denticles first elongate along the dorsal-ventral axis and then initiate actin extensions from their posterior surface, producing the characteristic organization of denticles into rows (Colosimo and Tolwinski, 2006; Price et al., 2006; Walters et al., 2006). A strong loss of Frizzled or Strabismus function only disrupts planar polarity in a subset of denticles (Price et al., 2006; Walters et al., 2006), indicating that additional mechanisms regulate denticle organization in the embryo. One possibility is the Fat and Dachsous protocadherins that are required for denticle polarity at later larval stages (Casal et al., 2006). Denticle polarity involves some of the same proteins involved in axis elongation, including a planar polarized network of actin filaments (Price et al., 2006; Walters et al., 2006), an AP enrichment of myosin II whose activity is required for cell and denticle morphology (Walters et al., 2006), and a complementary dorsal-ventral enrichment of adherens junction proteins (Colosimo and Tolwinski, 2006; Price et al., 2006). In contrast to intercalating cells, denticle-producing cells do not change neighbors, indicating that polarized cytoskeletal and junctional structures participate in distinct cell biological processes at different stages of development. Conclusion The mechanisms that organize multicellular populations operate through local cell interactions that are biased by global patterns of spatial information. Genetic and 1060  Cell 129, June 15, 2007 ©2007 Elsevier Inc.

molecular studies of planar polarity have identified a conserved role for the Frizzled-dependent PCP pathway in epithelial organization. Distinct strategies are used for the organization of rearranging cell populations that respond to information from the Frizzled/PCP and AP patterning systems. The challenge of maintaining planar polarity in a dynamic cellular environment is met by a variety of strategies, including using PCP signaling to organize the ECM and linking AP patterning cues directly to the cytoskeletal and junctional proteins that generate cell behavior. A detailed description of the molecular and cellular basis of planar polarity will provide insight into mechanisms that mediate communication between cells to establish planar polarity, and live-imaging studies will be essential to understanding how these polarities are dynamically remodeled during tissue morphogenesis. The current challenge is to link widely used systems for providing spatial information to unique properties of cell behavior and to understand how the polarized behaviors of individual cells are integrated at the population level to generate three-dimensional tissue structure. Acknowledgments I thank the members of my lab for many valuable discussions and R. Zallen, P. Adler, J. Axelrod, K. Anderson, E. Wieschaus, C. Bargmann, N. Denef, L. Davidson, Z. Gitai, N. Tolwinski, and E. Lai for comments on the review. Work in my laboratory is supported by a Burroughs Wellcome Fund Career Award in the Biomedical Sciences, a March of Dimes Basil O’Connor Starter Scholar Research Award, and a Searle Scholar Award. References Adler, P.N. (2002). Planar signaling and morphogenesis in Drosophila. Dev. Cell 2, 525–535. Adler, P.N., Krasnow, R.E., and Liu, J. (1997). Tissue polarity points from cells that have higher Frizzled levels towards cells that have lower Frizzled levels. Curr. Biol. 7, 940–949. Adler, P.N., Taylor, J., and Charlton, J. (2000). The domineering non-autonomy of frizzled and van Gogh clones in the Drosophila wing is a consequence of a disruption in local signaling. Mech. Dev. 96, 197–207. Amonlirdviman, K., Khare, N.A., Tree, D.R., Chen, W.S., Axelrod, J.D., and Tomlin, C.J. (2005). Mathematical modeling of planar cell polarity to understand domineering nonautonomy. Science 307, 423–426. Axelrod, J.D. (2001). Unipolar membrane association of Dishevelled mediates Frizzled planar cell polarity signaling. Genes Dev. 15, 1182–1187. Axelrod, J.D., and McNeill, H. (2002). Coupling planar cell polarity signaling to morphogenesis. ScientificWorldJournal 15, 434–454. Baena-Lopez, L., Baonza, A., and Garcia-Bellido, A. (2005). The orientation of cell divisions determines the shape of Drosophila organs. Curr. Biol. 15, 1640–1644. Bastock, R., Strutt, H., and Strutt, D. (2003). Strabismus is asymmetrically localised and binds to Prickle and Dishevelled during Drosophila planar polarity patterning. Development 130, 3007– 3014. Bertet, C., Sulak, L., and Lecuit, T. (2004). Myosin-dependent junction remodelling controls planar cell intercalation and axis elonga-

tion. Nature 429, 667–671. Blankenship, J.T., Backovic, S., Sanny, J.S.P., Weitz, O., and Zallen, J.A. (2006). Multicellular rosette formation links planar cell polarity to tissue morphogenesis. Dev. Cell 11, 459–470. Casal, J., Struhl, G., and Lawrence, P.A. (2006). Two separate molecular systems, Dachsous/Fat and Starry night/Frizzled, act independently to confer planar cell polarity. Development 133, 4561–4572. Chae, J., Kim, M.J., Goo, J.H., Collier, S., Gubb, D., Charlton, J., Adler, P.N., and Park, W.J. (1999). The Drosophila tissue polarity gene starry night encodes a member of the protocadherin family. Development 126, 5421–5429. Chen, X., and Macara, I.G. (2005). Par-3 controls tight junction assembly through the Rac exchange factor Tiam1. Nat. Cell Biol. 7, 262–269. Ciruna, B., Jenny, A., Lee, D., Mlodzik, M., and Schier, A.F. (2006). Planar cell polarity signalling couples cell division and morphogenesis during neurulation. Nature 439, 220–224. Classen, A.K., Anderson, K.I., Marois, E., and Eaton, S. (2005). Hexagonal packing of Drosophila wing epithelial cells by the planar cell polarity pathway. Dev. Cell 9, 805–817. Colosimo, P.F., and Tolwinski, N.S. (2006). Wnt, Hedgehog, and junctional Armadillo/β-catenin establish planar polarity in the Drosophila embryo. PloS ONE 1, e9. Concha, M.L., and Adams, R.J. (1998). Oriented cell divisions and cellular morphogenesis in the zebrafish gastrula and neurula: a time-lapse analysis. Development 125, 983–994. Curtin, J.A., Quint, E., Tsipouri, V., Arkell, R.M., Cattanach, B., Copp, A.J., Henderson, D.J., Spurr, N., Stanier, P., Fisher, E.M., et al. (2003). Mutation of Celsr1 disrupts planar polarity of inner ear hair cells and causes severe neural tube defects in the mouse. Curr. Biol. 13, 1129–1133. Dabdoub, A., and Kelley, M.W. (2005). Planar cell polarity and a potential role for a Wnt morphogen gradient in stereociliary bundle orientation in the mammalian inner ear. J. Neurobiol. 64, 446–457. Das, G., Jenny, A., Klein, T.J., Eaton, S., and Mlodzik, M. (2004). Diego interacts with Prickle and Strabismus/Van Gogh to localize planar cell polarity complexes. Development 131, 4467–4476. Davidson, L.A., Marsden, M., Keller, R., and DeSimone, D.W. (2006). Integrin α5β1 and fibronectin regulate polarized cell protrusions required for Xenopus convergence and extension. Curr. Biol. 16, 833–844. Eaton, S., Wepf, R., and Simons, K. (1996). Roles for Rac1 and Cdc42 in planar polarization and hair outgrowth in the wing of Drosophila. J. Cell Biol. 135, 1277–1289. Elul, T., and Keller, R. (2000). Monopolar protrusive activity: a new morphogenic cell behavior in the neural plate dependent on vertical interactions with the mesoderm in Xenopus. Dev. Biol. 224, 3–19. Fischer, E., Legue, E., Doyen, A., Nato, F., Nicolas, J.F., Torres, V., Yaniv, M., and Pontoglio, M. (2006). Defective planar cell polarity in polycystic kidney disease. Nat. Genet. 38, 21–23.

Glickman, N.S., Kimmel, C.B., Jones, M.A., and Adams, R.J. (2003). Shaping the zebrafish notochord. Development 130, 873–887. Gong, Y., Mo, C., and Fraser, S.E. (2004). Planar cell polarity signalling controls cell division orientation during zebrafish gastrulation. Nature 430, 689–693. Goto, T., and Keller, R. (2002). The planar cell polarity gene strabismus regulates convergence and extension and neural fold closure in Xenopus. Dev. Biol. 247, 165–181. Goto, T., Davidson, L., Asashima, M., and Keller, R. (2005). Planar cell polarity genes regulate polarized extracellular matrix deposition during frog gastrulation. Curr. Biol. 15, 787–793. Gubb, D., and Garcia-Bellido, A. (1982). A genetic analysis of the determination of cuticular polarity during development in Drosophila melanogaster. J. Embryol. Exp. Morphol. 68, 37–57. Gubb, D., Green, C., Huen, D., Coulson, D., Johnson, G., Tree, D., Collier, S., and Roote, J. (1999). The balance between isoforms of the prickle LIM domain protein is critical for planar polarity in Drosophila imaginal discs. Genes Dev. 13, 2315–2327. Guo, N., Hawkins, C., and Nathans, J. (2004). Frizzled6 controls hair patterning in mice. Proc. Natl. Acad. Sci. USA 101, 9277–9281. Harris, T.J., and Peifer, M. (2004). Adherens junction-dependent and -independent steps in the establishment of epithelial cell polarity in Drosophila. J. Cell Biol. 167, 135–147. Heisenberg, C.P., Tada, M., Rauch, G.J., Saude, L., Concha, M.L., Geisler, R., Stemple, D.L., Smith, J.C., and Wilson, S.W. (2000). Silberblick/Wnt11 mediates convergent extension movements during zebrafish gastrulation. Nature 405, 76–81. Hyodo-Miura, J., Yamamoto, T.S., Hyodo, A.C., Iemura, S., Kusakabe, M., Nishida, E., Natsume, T., and Ueno, N. (2006). XGAP, an ArfGAP, is required for polarized localization of PAR proteins and cell polarity in Xenopus gastrulation. Dev. Cell 11, 69–79. Irvine, K.D., and Wieschaus, E. (1994). Cell intercalation during Drosophila germband extension and its regulation by pair-rule segmentation genes. Development 120, 827–841. Jenny, A., Darken, R.S., Wilson, P.A., and Mlodzik, M. (2003). Prickle and Strabismus form a functional complex to generate a correct axis during planar cell polarity signaling. EMBO J. 22, 4409–4420. Jenny, A., Reynolds-Kenneally, J., Das, G., Burnett, M., and Mlodzik, M. (2005). Diego and Prickle regulate Frizzled planar cell polarity signalling by competing for Dishevelled binding. Nat. Cell Biol. 7, 691–697. Jiang, D., Munro, E.M., and Smith, W.C. (2005). Ascidian prickle regulates both mediolateral and anterior-posterior cell polarity of notochord cells. Curr. Biol. 15, 79–85. Jones, C., and Chen, P. (2007). Planar cell polarity signaling in vertebrates. Bioessays 29, 120–132. Keller, R. (2002). Shaping the vertebrate body plan by polarized embryonic cell movements. Science 298, 1950–1954. Keller, R. (2006). Mechanisms of elongation in embryogenesis. Development 133, 2291–2302.

Franke, J.D., Montague, R.A., and Kiehart, D.P. (2005). Nonmuscle myosin II generates forces that transmit tension and drive contraction in multiple tissue during dorsal closure. Curr. Biol. 15, 2208– 2221.

Keller, R., Davidson, L., Edlund, A., Elul, T., Ezin, M., Shook, D., and Skoglund, P. (2000). Mechanisms of convergence and extension by cell intercalation. Philos. Trans. R. Soc. Lond. B Biol. Sci. 355, 897–922.

Gho, M., and Schweisguth, F. (1998). Frizzled signalling controls orientation of asymmetric sense organ precursor cell divisions in Drosophila. Nature 393, 178–181.

Kilian, B., Mansukoski, H., Barbosa, F.C., Ulrich, F., Tada, M., and Heisenberg, C.P. (2003). The role of Ppt/Wnt5 in regulating cell shape and movement during zebrafish gastrulation. Mech. Dev. 120, 467–476.

Gibson, M.C., Patel, A.B., Nagpal, R., and Perrimon, N. (2006). The emergence of geometric order in proliferating metazoan epithelia. Nature 442, 1038–1041.

Klein, T.J., and Mlodzik, M. (2005). Planar cell polarization: an emerging model points in the right direction. Annu. Rev. Cell Dev.

Cell 129, June 15, 2007 ©2007 Elsevier Inc.  1061

Biol. 21, 155–176. Klingensmith, J., Nusse, R., and Perrimon, N. (1994). The Drosophila segment polarity gene dishevelled encodes a novel protein required for response to the wingless signal. Genes Dev. 8, 118–130. Koppen, M., Fernandez, B.G., Carvalho, L., Jacinto, A., and Heisenberg, C.-P. (2006). Coordinated cell-shape changes control epithelial movement in zebrafish and Drosophila. Development 133, 2671–2681. Lawrence, P.A., Casal, J., and Struhl, G. (2002). Towards a model of the organisation of planar polarity and pattern in the Drosophila abdomen. Development 129, 2749–2760. Lawrence, P.A., Casal, J., and Struhl, G. (2004). Cell interactions and planar polarity in the abdominal epidermis of Drosophila. Development 131, 4651–4664. Le Garrec, J.F., Lopez, P., and Kerszberg, M. (2006). Establishment and maintenance of planar epithelial cell polarity by asymmetric cadherin bridges: a computer model. Dev. Dyn. 235, 235–246. Lengyel, J.A., and Iwaki, D.D. (2002). It takes guts: the Drosophila hindgut as a model system for organogenesis. Dev. Biol. 243, 1–19. Logan, C.Y., and Nusse, R. (2004). The Wnt signaling pathway in development and disease. Annu. Rev. Cell Dev. Biol. 20, 781–810. Lu, B., Usui, T., Uemura, T., Jan, L., and Jan, Y.N. (1999). Flamingo controls the planar polarity of sensory bristles and asymmetric division of sensory organ precursors in Drosophila. Curr. Biol. 9, 1247–1250.

Sahai, E., and Marshall, C.J. (2002). ROCK and Dia have opposing effects on adherens junctions downstream of Rho. Nat. Cell Biol. 4, 408–415. Seifert, J.R., and Mlodzik, M. (2007). Frizzled/PCP signalling: a conserved mechanism regulating cell polarity and directed motility. Nat. Rev. Genet. 2, 126–138. Shih, J., and Keller, R. (1992). Cell motility driving mediolateral intercalation in explants of Xenopus laevis. Development 116, 901–914. Shimada, Y., Usui, T., Yanagawa, S., Takeichi, M., and Uemura, T. (2001). Asymmetric colocalization of Flamingo, a seven-pass transmembrane cadherin, and Dishevelled in planar cell polarization. Curr. Biol. 11, 859–863. Shimada, Y., Yonemura, S., Ohkura, H., Strutt, D., and Uemura, T. (2006). Polarized transport of Frizzled along the planar microtubule arrays in Drosophila wing epithelium. Dev. Cell 10, 209–222. Simon, M.A. (2004). Planar cell polarity in the Drosophila eye is directed by graded Four-jointed and Dachsous expression. Development 131, 6175–6184. Strutt, D., and Strutt, H. (2007). Differential activities of the core planar polarity proteins during Drosophila wing patterning. Dev. Biol. 302, 181–194. Strutt, D.I. (2001). Asymmetric localization of frizzled and the establishment of cell polarity in the Drosophila wing. Mol. Cell 7, 367–375. Strutt, D.I. (2002). The asymmetric subcellular localisation of components of the planar polarity pathway. Semin. Cell Dev. Biol. 13, 225–231.

Marsden, M., and DeSimone, D.W. (2003). Integrin-ECM interactions regulate cadherin-dependent cell adhesion and are required for convergent extension in Xenopus. Curr. Biol. 13, 1182–1191.

Strutt, H., and Strutt, D. (2005). Long-range coordination of planar polarity in Drosophila. Bioessays 27, 1218–1227.

Matakatsu, H., and Blair, S.S. (2004). Interactions between Fat and Dachsous and the regulation of planar cell polarity in the Drosophila wing. Development 131, 3785–3794.

Tada, M., and Smith, J.C. (2000). Xwnt11 is a target of Xenopus Brachyury: regulation of gastrulation movements via Dishevelled, but not through the canonical Wnt pathway. Development 127, 2227–2238.

Matakatsu, H., and Blair, S.S. (2006). Separating the adhesive and signaling functions of the Fat and Dachsous protocadherins. Development 133, 2315–2324. Miller, J.R., Rowning, B.A., Larabell, C.A., Yang-Snyder, J.A., Bates, R.L., and Moon, R.T. (1999). Establishment of the dorsalventral axis in Xenopus embryos coincides with the dorsal enrichment of dishevelled that is dependent on cortical rotation. J. Cell Biol. 146, 427–437. Mlodzik, M. (2002). Planar cell polarization: do the same mechanisms regulate Drosophila tissue polarity and vertebrate gastrulation? Trends Genet. 18, 564–571. Montcouquiol, M., Rachel, R.A., Lanford, P.J., Copeland, N.G., Jenkins, N.A., and Kelley, M.W. (2003). Identification of Vangl2 and Scrb1 as planar polarity genes in mammals. Nature 423, 173–177. Myers, D.C., Sepich, D.S., and Solnica-Krezel, L. (2002a). Convergence and extension in vertebrate gastrulae: cell movements according to or in search of identity? Trends Genet. 18, 447–455. Myers, D.C., Sepich, D.S., and Solnica-Krezel, L. (2002b). Bmp activity gradient regulates convergent extension during zebrafish gastrulation. Dev. Biol. 243, 81–98. Ninomiya, H., Elinson, R.P., and Winklbauer, R. (2004). Antero-posterior tissue polarity links mesoderm convergent extension to axial patterning. Nature 430, 364–367. Price, M.H., Roberts, D.M., McCartney, B.M., Jezuit, E., and Peifer, M. (2006). Cytoskeletal dynamics and cell signaling during planar polarity establishment in the Drosophila embryonic denticle. J. Cell Sci. 119, 403–415.

1062  Cell 129, June 15, 2007 ©2007 Elsevier Inc.

Tawk, M., Araya, C., Lyons, D.A., Reugels, A.M., Girdler, G.C., Bayley, P.R., Hyde, D.R., Tada, M., and Clarke, J.D. (2007). A mirrorsymmetric cell division that orchestrates neuroepithelial morphogenesis. Nature 446, 797–800. Taylor, J., Abramova, N., Charlton, J., and Adler, P.N. (1998). Van Gogh: a new Drosophila tissue polarity gene. Genetics 150, 199– 210. Theisen, H., Purcell, J., Bennett, M., Kansagara, D., Syed, A., and Marsh, J.L. (1994). dishevelled is required during wingless signaling to establish both cell polarity and cell identity. Development 120, 347–360. Tree, D.R., Shulman, J.M., Rousset, R., Scott, M.P., Gubb, D., and Axelrod, J.D. (2002). Prickle mediates feedback amplification to generate asymmetric planar cell polarity signaling. Cell 109, 371– 381. Turner, C.M., and Adler, P.N. (1998). Distinct roles for the actin and microtubule cytoskeletons in the morphogenesis of epidermal hairs during wing development in Drosophila. Mech. Dev. 70, 181–192. Usui, T., Shima, Y., Shimada, Y., Hirano, S., Burgess, R.W., Schwarz, T.L., Takeichi, M., and Uemura, T. (1999). Flamingo, a seven-pass transmembrane cadherin, regulates planar cell polarity under the control of Frizzled. Cell 98, 585–595. Vinson, C.R., and Adler, P.N. (1987). Directional non-cell autonomy and the transmission of polarity information by the frizzled gene of Drosophila. Nature 329, 549–551. Wallingford, J.B., Rowning, B.A., Vogeli, K.M., Rothbacher, U., Fraser, S.E., and Harland, R.M. (2000). Dishevelled controls cell polar-

ity during Xenopus gastrulation. Nature 405, 81–85. Wallingford, J.B., Fraser, S.E., and Harland, R.M. (2002). Convergent extension: the molecular control of polarized cell movement during embryonic development. Dev. Cell 2, 695–706. Walters, J.W., Dilks, S.A., and Dinardo, S. (2006). Planar polarization of the denticle field in the Drosophila embryo: roles for Myosin II (zipper) and fringe. Dev. Biol. 297, 323–339. Wang, J., Hamblet, N.S., Mark, S., Dickinson, M.E., Brinkman, B.C., Segil, N., Fraser, S.E., Chen, P., Wallingford, J.B., and Wynshaw-Boris, A. (2006). Dishevelled genes mediate a conserved mammalian PCP pathway to regulate convergent extension during neurulation. Development 133, 1767–1778. Wang, Y., and Nathans, J. (2007). Tissue/planar cell polarity in vertebrates: new insights and new questions. Development 134, 647–658.

Wong, H.C., Bourdelas, A., Krauss, A., Lee, H.J., Shao, Y., Wu, D., Mlodzik, M., Shi, D.L., and Zheng, J. (2003). Direct binding of the PDZ domain of Dishevelled to a conserved internal sequence in the C-terminal region of Frizzled. Mol. Cell 12, 1251–1260. Wong, L.L., and Adler, P.N. (1993). Tissue polarity genes of Drosophila regulate the subcellular location for prehair initiation in pupal wing cells. J. Cell Biol. 123, 209–221. Yang, C.H., Axelrod, J.D., and Simon, M.A. (2002). Regulation of Frizzled by fat-like cadherins during planar polarity signaling in the Drosophila compound eye. Cell 108, 675–688. Zallen, J.A., and Wieschaus, E. (2004). Patterned gene expression directs bipolar planar polarity in Drosophila. Dev. Cell 6, 343–355. Zallen, J.A., and Zallen, R. (2004). Cell-pattern disordering during convergent extension in Drosophila. J. Phys. Condens. Matter 16, S5073–S5080.

Winter, C.G., Wang, B., Ballew, A., Royou, A., Karess, R., Axelrod, J.D., and Luo, L. (2001). Drosophila Rho-associated kinase (Drok) links Frizzled-mediated planar cell polarity signaling to the actin cytoskeleton. Cell 105, 81–91.

Zeidler, M.P., Perrimon, N., and Strutt, D.I. (2000). Multiple roles for four-jointed in planar polarity and limb patterning. Dev. Biol. 228, 181–196.

Wolff, T., and Rubin, G.M. (1998). Strabismus, a novel gene that regulates tissue polarity and cell fate decisions in Drosophila. Development 125, 1149–1159.

Zheng, L., Zhang, J., and Carthew, R.W. (1995). frizzled regulates mirror-symmetric pattern formation in the Drosophila eye. Development 121, 3045–3055.

Cell 129, June 15, 2007 ©2007 Elsevier Inc.  1063