RNA in situ hybridization shows the ssh transcripts to be maternally supplied in the embryo and to be expressed broadly in embryonic and imaginal tissues (Niwa, 2002).
Animal development requires that positional information act on the genome to control cell fate and cell shape. The primary determinant of animal cell shape is the cytoskeleton and thus the mechanisms by which extracellular signals influence the cytoskeleton are crucial for morphogenesis. In the developing Drosophila compound eye, localized polymerization of actin functions to constrict the apical surface of epithelial cells, both at the morphogenetic furrow and later to maintain the coherence of the nascent ommatidia. As elsewhere, actin polymerization in the developing eye is regulated by ADF/cofilin (Twinstar in Drosophila), which is activated by Slingshot (Ssh), a cofilin phosphatase. Ssh acts in the developing eye to limit actin polymerization in the assembling ommatidia, but not in the morphogenetic furrow. While Ssh controls cell shape, surprisingly there are no direct or immediate consequences for cell type. Ssh protein becomes apically concentrated in cells that express elevated levels of the Sevenless (Sev) receptor-tyrosine kinase (RTK), even those that receive no ligand. This is interpreted as a non-signal driven, RTK-dependent localization of Ssh to allow for locally increased actin filament turnover. It is suggested that there are two modes of actin remodeling in the developing eye: a non-RTK, non-Ssh mediated mechanism in the morphogenetic furrow, and an RTK and Ssh-dependent mode during ommatidial assembly (Rogers, 2005).
Despite some sequence similarities between Drosophila Ssh and MAP Kinase phosphatases, no such activity, could be detected in vivo or in vitro. Consistent with the results of others, it was found that Ssh normally limits actin polymerization in the developing eye. However, this activity is found to be regionally specific: it is limited to the assembling ommatidia, and does not appear to function in the morphogenetic furrow, despite the intense regulation of F-actin there. Recently, a second, structurally unrelated cofilin phosphatase, chronophin (CIN) has been found in vertebrates. A CIN ortholog or some other cofilin phosphatase may control F-actin dynamics in the morphogenetic furrow (Rogers, 2005).
It is suggested that the colocalization of Sev and Ssh in later eye development may be functionally significant. It is interesting to note that the elevated apical deposits of Ssh antigen are seen at a time and place where elevated Sev expression is known to be occurring and also where there is elevated F-actin. In other systems, Ssh has been found to be associated with multi-protein complexes that include signaling receptors or scaffolds. Stimulation of human cells with neuregulin-1beta (an Egfr ligand), triggers lamellipodium formation, results in the dephosphorylation of Ssh, its release from the scaffolding protein 14-3-3 and its translocation to the F-actin rich lamellipodium. LIM Kinase is also activated by neuregulin-1beta and SSH1L. LIMK, actin and 14-3-3 can form a multiprotein complex in vivo. Therefore, it has been suggested that the local activation of SSH1L and LIM Kinase may be coordinately regulated as part of a complex, increasing cofilin-dependent actin filament turnover in areas of intense actin remodeling such as lamellipodia. Also, in cultured rat mammary adenocarcinoma cells engineered to overexpress the RTK EGFR, the application of liganded beads elicits locally elevated actin polymerization, with locally elevated cofilin (Rogers, 2005).
Thus, in the developing eye, local and transiently elevated RTK levels may lead, over a period of hours, to the stabilization of apical actin through the recruitment of cofilin and the molecules that regulate it, such as Ssh. Thus, the coincident Sev and Ssh apical pattern may represent a transient state of elevated RTK expression, which then serves to elicit the apical constrictions that regulate the Armadillo-positive junctional structures, which then stabilize the ommatidial cell clusters. Later in life, this apical constriction is remodeled to form the light sensing rhabdomere: the site of opsin function. It is further suggested that this concentration of Ssh must remain localized to the apical tip of the cell, otherwise (as in mutant clones), the polymerization of actin over-runs and the differentiation and ultimately the viability of the cell is affected. Thus the local, limited, but long-lasting concentration of Ssh at the tips of developing photoreceptor cells in response to RTK levels may serve to stabilize the ommatidial cell cluster. Thus, while no direct biochemical evidence is available, the tight apical co-localization of Sev and Ssh is consistent with a direct interaction between Ssh and a complex that includes RTKs (Rogers, 2005).
Wing hairs made by ssh null cells are much thicker than normal, twisted, and sometimes bifurcated at their tips. In hypomorphs, hairs are much better shaped; nevertheless, they exhibit the terminal splitting phenotype. The splitting phenotype is also seen in bristles and lateral branches of the arista, which is the terminal segment of the antenna. Closer observation of surfaces of wild-type bristles reveals parallel striation patterns, which are known to reflect alignments of actin bundles underneath pupal cuticles. This pattern is distorted in mutant bristles. Compound eyes that are derived from ssh clones show disorganization of the regular ommatidia pattern and have lost many interommatidial bristles (Niwa, 2002).
To assess the importance of the phosphatase activity of Ssh during development, whether expression of a catalytically inactive form could rescue the lethality of ssh strong mutants and the hair/bristle phenotype in hypomorphs was examined. To design such an inactive form, previous studies on protein tyrosine phosphatases and the MKPs were followed, in which the conserved Cys residue in the catalytic pocket is replaced by Ser. Ssh(wt) expression rescues ssh mutants from both the lethality and the hair/bristle phenotype, whereas Ssh(CS) does not (Niwa, 2002).
In wings of the examined hypomorph, 89 hairs out of 392 showed the tip splitting; this phenotype was recovered by Ssh(wt) expression, but not by Ssh(CS) expression. Subcellular localization of Ssh(wt) and that of Ssh(CS) are indistinguishable, and both proteins have the ability to bind F actin. These results support the view that the catalytic activity of Ssh is required for its in vivo roles. In the wild-type background, expression of either Ssh(wt) or Ssh(CS) by using da-GAL4, has no effect on hair/bristle morphogenesis (Niwa, 2002).
The p21 activated kinase (Pak) family of protein kinases are involved in many cellular functions like re-organisation of the cytoskeleton, transcriptional control, cell division, and survival. These pleiotropic actions are reflected in a plethora of known interacting proteins and phosphorylation substrates. Yet, the integration of a single Pak protein into signalling pathways controlling a particular developmental process are less well studied. For two of the three known Pak proteins in Drosophila melanogaster, D-Pak and Mushroom bodies tiny (Mbt), distinct functions during eye development have been established. This study undertook a genetic approach to identify proteins acting in the Mbt signalling pathway during photoreceptor cell morphogenesis. The genetic screen identified the actin depolymerisation factor Twinstar/Cofilin as one target of Mbt signalling. Twinstar/Cofilin becomes phosphorylated upon activation of Mbt. However, biochemical and genetic experiments question the role of the LIM domain protein kinase (Limk) as a major link between Mbt and Twinstar/Cofilin as it has been suggested for other PAK proteins. Constitutive activation of Mbt not only disturbs the actin cytoskeleton but also affects adherins junctions organisation indicating a requirement of the protein in cell adhesion dependent processes during photoreceptor cell differentiation (Menzel, 2007).
Morphogenesis is a fundamental process during the development of a multi-cellular organism, which not only requires the specification of the various cell types, but also their assembly into complex tissues by means of cell shape changes, cell movement, sorting processes, and elimination of surplus cells. Many of these events depend on precisely controlled modulation of cell-cell or cell-extracellular matrix contacts (Menzel, 2007).
Adherens junctions are specialized membrane structures that mediate adhesion between epithelial cells and provide a link to the actin cytoskeleton. The central component of adherens junctions is the cadherin-catenin complex. Cadherins constitute a family of transmembrane proteins that mediate Ca2+-dependent cell-cell adhesion. The intracellular domain of cadherin binds to β-Catenin (Drosophila: Armadillo), which in turn can associate with the actin binding protein α-Catenin. However, the prevailing dogma that cadherins at adherens junctions are linked to the actin cytoskeleton through β-Catenin and α-Catenin in a stable quaternary complex has recently been questioned by the finding that α-Catenin is associated in a mutually exclusive manner with either cadherin-β-Catenin or with actin (Menzel, 2007).
A role of adherens junctions as a focal point for intracellular signalling has emerged. Notably, the formation of cadherin-mediated cell contacts influences the activity of the RhoGTPases Cdc42, Rac and Rho. These proteins function as molecular switches, cycling between an active GTP-bound conformation and an inactive, GDP-bound state. In its active form, RhoGTPases bind to a vast number of downstream effector proteins and one of the major effects is the reorganisation of the actin cytoskeleton (Menzel, 2007).
A family of proteins, which are influenced by RhoGTPases in its activity and localisation, are the p21-activated kinases (Pak). Pak proteins are characterised by a C-terminal serine/threonine kinase domain and a N-terminal p21-binding domain (PBD) required for binding of Rac- or Cdc42-like RhoGTPases. Based on additional structural features, which determine the regulation of the kinase activity and the binding to other proteins, Pak proteins can be classified into two subgroups. From the six Pak proteins known in Homo sapiens (Hs), three (HsPak1-3) have been assigned to group 1, whereas HsPak4-6 belong to the group 2. Similar, the three Pak proteins found in Drosophila melanogaster can be classified as group 1 (D-Pak, D-Pak3) and group 2 (Mbt/D-Pak2) members, respectively. The role of Pak proteins as regulators of the cytoskeleton, cell morphology and cell motility is well established. Different Pak proteins can induce the formation of lamellipodia, filopodia, and membrane ruffles as well as the dissolution of stress fibers and the disassembly of focal adhesions. A function of Pak proteins at the centrosomes of dividing cells has been described. Deregulation of Pak protein activity can result in oncogenic transformation. The pleiotropic functions of Pak proteins are reflected in the plethora of known interacting proteins and phosphorylation substrates. This obviously raises the question of differential activation of effector pathways by different Pak proteins in a cell-type specific manner (Menzel, 2007).
The developing Drosophila eye is used as a model system to study the role of the group 2 Pak protein Mbt in tissue morphogenesis. The adult eye with its 800 single eye units (ommatidia), each containing eight photoreceptor cells as well as non-neuronal cone, pigment, and bristle cells, arises from a monolayer epithelium, the eye imaginal disc. Differentiation of the different cell types of the eye is initiated in third instar larvae, when an indentation known as the morphogenetic furrow starts to move from posterior to anterior across the eye disc. Anterior to the furrow, cells continue to proliferate, posterior to it, the photoreceptor cells and lens-secreting cone cells become specified in a sequential manner through a series of inductive cell-cell interactions. At pupal stage, pigment and bristle cells are added to complete the ommatidium structure and surplus cells are eliminated by programmed cell death. After recruitment and determination, photoreceptor cells undergo massive morphological changes to form the rhabdomeres, the light-sensitive structures. Rhabdomeres are apical membrane specializations. Alignment of rhabdomeres to the ommatidial optical axis is achieved by involution of the apical domains and adherens junctions into the epithelial layer followed by their massive expansion in distal-proximal direction (Menzel, 2007).
Loss of function studies of the Pak proteins Mbt and D-Pak provided evidence for distinct localisation and function of these proteins during photoreceptor cell development. D-Pak is required in growth cones to control guidance of photoreceptor cell axons, whereas recruitment of Mbt to adherens junctions is dependent on a functional RhoGTPase binding site (Schneeberger, 2003). In mbt mutant animals, recruitment of the photoreceptor cells appear largely normal, but their final differentiation is disturbed. The adherens junctions become highly disorganized and rhabdomeres fail to elongate properly leading to the suggestion that Mbt acts as a downstream effector of activated RhoGTPases at adherens junctions to regulate photoreceptor cell morphogenesis (Schneeberger, 2003). This study shows that activation of Mbt disturbs the co-ordinated assembly of the ommatidium structure by interfering with the actin cytoskeleton and adherens junctions. Given the multitude of potential candidate targets of Mbt, a genetic screen was establised to identify components that act downstream of Mbt to control photoreceptor cell morphogenesis. This screen identified the actin depolymerisation factor Twinstar as one target of Mbt signalling. Yet, biochemical and genetic experiments question the role of LIM domain protein kinase (Limk) as a major link between Mbt and Twinstar (Menzel, 2007).
The view of adherens junctions as static structures that maintain epithelial integrity appears to be at odds with the requirement of dynamic cell shape changes during development. The assembly of the ommaditia of the Drosophila eye from an initially unpatterned epithelium provides an example. Photoreceptor cells become specified during third larval instar, but their final morphology is established at pupal stage. Simultaneously, the non-neuronal cone, pigment, and bristle cells must form a precise cell lattice that requires cell shape changes, movements and elimination of surplus cells. In spite of these dynamic processes, the integrity of adherens junctions as a basic feature of epithelia cells has to be maintained throughout eye development. The prevailing view that E-Cadherin is stably linked via β-Catenin and α-Catenin to the actin cytoskeleton has recently been questioned. It was shown that monomeric α-Catenin binds to E-Cadherin/β-Catenin, whereas dimeric α-Catenin does not and instead binds with high affinity to actin to influence Arp2/3-mediated actin polymerisation. If it is not α-Catenin that provides a physical link between adherens junctions and the actin cytoskeleton, what other proteins could mediate such a link (Menzel, 2007)?
Genetic and biochemical evidence is presented that the Pak protein Mbt, which localizes in a RhoGTPase dependent manner to adherens junctions, could provide a link to the actin cytoskeleton during photoreceptor cell morphogenesis. Constitutive activation of Mbt in photoreceptor cells causes clustering of F-actin, which finally leads to complete disruption of the adult eye morphology. In addition, the positioning of nuclei at apical levels is disturbed. The finding that mutations in twinstar and the phosphatase-encoding slingshot act as modifiers of the eye phenotype of a hypomorphic mbt allele suggest that these proteins work together to control photoreceptor cell morphogenesis. Additional evidence comes from the phenotypic analysis of the corresponding mutants. In twinstar mutants, elevated levels of F-actin, disruption of the ommatidial architecture and misshapen rhabdomeres are observed. Loss of Slingshot function results in increased F-actin levels and defects in the apical movement of photoreceptor cell nuclei. So far, no evidence is available for a direct interaction between Slingshot and Mbt. In vertebrates, Pak4 inactivates Slingshot and activates Limk by phosphorylation. Slingshot plays a dual role. It dephosphorylates and thereby down-regulates the activity of Limk to phosphorylate Cofilin at serine 3 and it directly dephosphorylates Cofilin at serine 3. Thus Pak4 activation results in a decrease of Cofilin activity and actin filament turnover. A major role of D-Limk in Mbt signalling is questioned by biochemical and genetic experiments. Activated Mbt is able to induce Twinstar phosphorylation. However, despite binding and phosphorylation of D-Limk by activated Mbt, no strong increase of D-Limk activity was seen. Furthermore, genetic experiments with a recently identified null mutation in D-Limk show that Mbt signalling is not completely blocked in the absence of D-Limk function. Yet, the strong synergistic effect upon complete removal of Mbt and D-Limk function indicates the requirement of both proteins in eye development. If it is not Mbt, what other proteins could be involved in regulation of D-Limk function during eye morphogenesis? Recent experiments implicate a role of the Par3 protein as a regulator of vertebrate Limk-2 activity towards Cofilin. Interestingly, the Drosophila Par3 homologue Bazooka localizes to and is required for maintenance of adherens junctions during photoreceptor cell morphogenesis (Menzel, 2007).
It is also evident that regulation of Twinstar activity by Pak proteins is not restricted to eye development. Mutations in twinstar lead to proliferation defects of central brain neuroblasts and the neurons derived from these neuroblasts show axon growth defects. In this study, genetic evidence was presented that Twinstar function in axon growth is activated by Slingshot and inhibited by D-Limk, which in turn is regulated by D-Pak and Rock-kinases. In vertebrates, Limk and Slingshot control growth cone motility and morphology via Cofilin phosphorylation. Knock-out of Limk in mice leads to changes in Cofilin phosphorylation and the actin cytoskeleton resulting in abnormalities in spine morphology and in synaptic function. A dominant-negative version of mouse Pak1 causes changes in spine number and synaptic morphology. Thus, different Pak proteins might not only act as general regulators of Cofilin activity, but they could also provide spatial and temporal control even in a single cell (Menzel, 2007).
In addition to the role in re-organisation of the actin cytoskeleton, overexpression experiments indicate that Mbt is also able to influence the adhesive properties of cells. Differential adhesion is crucial to establish the final retinal pattern. It has been demonstrated that an apical protein complex consisting of Crumbs, Stardust, and DPATJ controls assembly of adherens junctions and is essential for rhabdomere maintenance. Differential adhesion, by expression of DE-Cadherin alone or together with DN-Cadherin has been suggested as a mechanism that controls cell shape changes. Expression of DN-Cadherin in cone cells (in addition to DE-Cadherin, which is expressed in all cells in an ommatidium) confers on them specific adhesion properties in order to adopt a cell shape that minimizes surface contacts with surrounding cells. Integration of other proteins into the adhesive complex could be another mechanism to modify adhesive contacts. Hibris and IrreC/rst, transmembrane proteins of the immunoglobulin superfamily, become integrated into the adhesion complex and form an intercellular link only at the interface between the presumptive primary pigment cells and secondary/tertiary pigment cells to control proper cell sorting during pupal eye development. Interference with DE-Cadherin function blocks proper localisation of IrreC/rst and results in cell sorting and survival defects. It is therefore conceivable that neuronal expression of activated Mbt influences DE-Cadherin-mediated cell adhesion of photoreceptor and bristle cells with the surrounding pigment and cone cells. In this way, motility, sorting or survival of all retinal cells can be affected. Evidence for a function of other group 2 PAK proteins in regulation of cell adhesion comes from studies with human PAK4 and Xenopus X-Pak5. Expression of an activated version of human Pak4 leads to anchorage-independent growth of fibroblasts. X-Pak5 is expressed in regions of the embryo that undergo extensive cell movements during gastrulation. It was shown that X-Pak5 localises to cell-cell junctions and regulates in a calcium-dependent manner cell-cell adhesion. An activated version of X-Pak5 decreased cell adhesiveness, while kinase-dead X-Pak5 increased adhesion. However, in all cases the molecular targets of the PAK proteins to exert these effects remain to be identified (Menzel, 2007).
Abe, H., et al. (1996). Xenopus laevis actin-depolymerizing factor/cofilin: a phosphorylation-regulated protein essential for development. J. Cell Biol. 132: 871-885. 8603919
Aizawa H., et al. (2001). Phosphorylation of cofilin by LIM-kinase is necessary for semaphorin 3A-induced growth cone collapse. Nat. Neurosci. 4: 367-373. 11276226
Amano, T., Tanabe, K., Eto, T., Narumiya, S. and Mizuno, K. (2001) LIM-kinase 2 induces formation of stress fibres, focal adhesions and membrane blebs, dependent on its activation by Rho-associated kinase-catalysed phosphorylation at threonine-505. Biochem. J. 354: 149-159. 11171090
Arber S., et al. (1998). Regulation of actin dynamics through phosphorylation of cofilin by LIM-kinase. Nature 393: 805-809. 9655397
Baum, B., Li, W. and Perrimon, N. (2000). A cyclase-associated protein regulates actin and cell polarity during Drosophila oogenesis and in yeast. Curr. Biol. 10: 964-973. 10985383
Baum, B. and Perrimon, N. (2001). Spatial control of the actin cytoskeleton in Drosophila epithelial cells. Nat. Cell Biol. 3: 883-890. 11584269
Endo, M., et al. (2003). Control of growth cone motility and morphology by LIM kinase and Slingshot via phosphorylation and dephosphorylation of cofilin. J. Neurosci. 23(7): 2527-37. 12684437
Gunsalus, K. C., et al. (1995). Mutations in twinstar, a Drosophila gene encoding a cofilin/ADF homologue, result in defects in centrosome migration and cytokinesis. J. Cell Biol. 131: 1243-1259. 8522587
Hsieh, S. H., Ferraro, G. B. and Fournier, A. E. (2006). Myelin-associated inhibitors regulate cofilin phosphorylation and neuronal inhibition through LIM kinase and Slingshot phosphatase. J. Neurosci. 26(3): 1006-15. 16421320
Kaji, N., et al. (2003). Cell cycle-associated changes in Slingshot phosphatase activity and roles in cytokinesis in animal cells. J. Biol. Chem. 278(35): 33450-5. 12807904
Meberg, P. J., et al. (1996). Actin depolymerizing factor and cofilin phosphorylation dynamics: response to signals that regulate neurite extension. Cell Motil. Cytoskeleton 39: 172-190. 9484959
Menzel, N., Schneeberger, D. and Raabe, T. (2007). The Drosophila p21 activated kinase Mbt regulates the actin cytoskeleton and adherens junctions to control photoreceptor cell morphogenesis. Mech. Dev. 124: 78-90. Medline abstract: 17097274
Moon, A. and Drubin, D. G. (1995). The ADF/cofilin proteins: stimulus-responsive modulators of actin dynamics. Mol. Biol. Cell 6: 1423-1431. 8589446
Nagaoka, R., Abe, H., Kusano, K. and Obinata, T. (1995). Concentration of cofilin, a small actin-binding protein, at the cleavage furrow during cytokinesis. Cell Motil. Cytoskeleton 30: 1-7. 7728864
Nagata-Ohashi, K., et al. (2004). A pathway of neuregulin-induced activation of cofilin-phosphatase Slingshot and cofilin in lamellipodia. J. Cell Biol. 165(4): 465-71. 15159416
Nishita, M., et al. (2004). Phosphoinositide 3-kinase-mediated activation of cofilin phosphatase Slingshot and its role for insulin-induced membrane protrusion. J. Biol. Chem. 279(8): 7193-8. 14645219
Nishita, M., et al. (2005). Spatial and temporal regulation of cofilin activity by LIM kinase and Slingshot is critical for directional cell migration. J. Cell Biol. 171(2): 349-59. 16230460
Niwa, R., et al. (2002). Control of actin reorganization by Slingshot, a family of phosphatases that dephosphorylate ADF/Cofilin. Cell 108: 233-246. 11832213
Ohashi, K., Nagata, K., Maekawa, M., Ishizaki, T., Narumiya, S. and Mizuno, K. (2000). Rho-associated kinase ROCK activates LIM-kinase 1 by phosphorylation at threonine 508 within the activation loop. J. Biol. Chem. 275: 3577-3582. 10652353
Ohta, Y., et al. (2003). Differential activities, subcellular distribution and tissue expression patterns of three members of Slingshot family phosphatases that dephosphorylate cofilin. Genes Cells 8(10): 811-24. 14531860
Schneeberger, D. and Raabe, T. (2003). Mbt, a Drosophila PAK protein, combines with Cdc42 to regulate photoreceptor cell morphogenesis. Development 130: 427-437. Medline abstract: 12490550
Soosairajah, J., et al. (2005). Interplay between components of a novel LIM kinase-slingshot phosphatase complex regulates cofilin. EMBO J. 24(3): 473-86. 15660133
Tanaka, K., Nishio, R., Haneda, K. and Abe, H. (2005a). Functional involvement of Xenopus homologue of ADF/cofilin phosphatase, slingshot (XSSH), in the gastrulation movement. Zoolog Sci. 22(9): 955-69. 16219976
Tanaka, K., Okubo, Y. and Abe, H. (2005b). Involvement of slingshot in the Rho-mediated dephosphorylation of ADF/cofilin during Xenopus cleavage. Zoolog. Sci. 22(9): 971-84. 16219977
Toshima, J., Toshima, J. Y., Amano, T., Yang, N., Narumiya, S. and Mizuno, K. (2001a). Cofilin phosphorylation by protein kinase testicular protein kinase 1 and its role in integrin-mediated actin reorganization and focal adhesion formation. Mol. Biol. Cell 12: 1131-1145. 11294912
Toshima, J., Toshima, J. Y., Takeuchi, K., Mori, R. and Mizuno, K. (2001b). Cofilin phosphorylation and actin reorganization activities of testicular protein kinase 2 and its predominant expression in testicular Sertoli cells. J. Biol. Chem. 276: 31449-31458. 11418599
Wang, Y., Shibasaki, F. and Mizuno, K. (2005). Calcium signal-induced cofilin dephosphorylation is mediated by Slingshot via calcineurin. J. Biol. Chem. 280(13): 12683-9. 15671020
Yang, N., et al. (1998). Cofilin phosphorylation by LIM-kinase 1 and its role in Rac-mediated actin reorganization. Nature 393: 809-812. 9655398
date revised: 15 March 2008
Home page: The Interactive Fly © 1997 Thomas B. Brody, Ph.D.
The Interactive Fly resides on the
Society for Developmental Biology's Web server.