Enhancer of zeste acts as a major developmental regulator of Ciona intestinalis embryogenesis.

The paradigm of developmental regulation by Polycomb group (PcG) proteins posits that they maintain silencing outside the spatial expression domains of their target genes, particularly of Hox genes, starting from mid embryogenesis. The Enhancer of zeste [E(z)] PcG protein is the catalytic subunit of the PRC2 complex, which silences its targets via deposition of the H3K27me3 mark. Here, we studied the ascidian Ciona intestinalis counterpart of E(z). Ci-E(z) is detected by immunohistochemistry as soon as the 2- and 4-cell stages as a cytoplasmic form and becomes exclusively nuclear thereafter, whereas the H3K27me3 mark is detected starting from the gastrula stage and later. Morpholino invalidation of Ci-E(z) leads to the total disappearance of both Ci-E(z) protein and its H3K27me3 mark. Ci-E(z) morphants display a severe phenotype. Strikingly, the earliest defects occur at the 4-cell stage with the dysregulation of cell positioning and mitotic impairment. At later stages, Ci-E(z)-deficient embryos are affected by terminal differentiation defects of neural, epidermal and muscle tissues, by the failure to form a notochord and by the absence of caudal nerve. These major phenotypic defects are specifically rescued by injection of a morpholino-resistant Ci-E(z) mRNA, which restores expression of Ci-E(z) protein and re-deposition of the H3K27me3 mark. As observed by qPCR analyses, Ci-E(z) invalidation leads to the early derepression of tissue-specific developmental genes, whereas late-acting developmental genes are generally down-regulated. Altogether, our results suggest that Ci-E(z) plays a major role during embryonic development in Ciona intestinalis by silencing early-acting developmental genes in a Hox-independent manner.

The non-proliferative nature of ascidian folliculogenesis as a model of highly ordered cellular topology distinct from proliferative epithelia.

Previous studies have addressed why and how mono-stratified epithelia adopt a polygonal topology. One major additional, and yet unanswered question is how the frequency of different cell shapes is achieved and whether the same distribution applies between non-proliferative and proliferative epithelia. We compared different proliferative and non-proliferative epithelia from a range of organisms as well as Drosophila melanogaster mutants, deficient for apoptosis or hyperproliferative. We show that the distribution of cell shapes in non-proliferative epithelia (follicular cells of five species of tunicates) is distinctly, and more stringently organized than proliferative ones (cultured epithelial cells and Drosophila melanogaster imaginal discs). The discrepancy is not supported by geometrical constraints (spherical versus flat monolayers), number of cells, or apoptosis events. We have developed a theoretical model of epithelial morphogenesis, based on the physics of divided media, that takes into account biological parameters such as cell-cell contact adhesions and tensions, cell and tissue growth, and which reproduces the effects of proliferation by increasing the topological heterogeneity observed experimentally. We therefore present a model for the morphogenesis of epithelia where, in a proliferative context, an extended distribution of cell shapes (range of 4 to 10 neighbors per cell) contrasts with the narrower range of 5-7 neighbors per cell that characterizes non proliferative epithelia.

[Would D'Arcy Thompson have predicted a topological control of apoptosis?]. / D'Arcy Thompson aurait-il prédit un contrôle topologique de l'apoptose ?

The laws that drive morphogenesis remain a major biological question. Today's views emphasize molecular autonomous processes rather than physical and mechanical constraints proposed by d'Arcy Thompson earlier on. In Ciona intestinalis oocyte, follicular cells formed by two distinct sets of geometrically-ordered epithelial monolayers positioned over the egg control apoptosis, implying that physically-predetermined shapes play a role in the control of cell determinism. In follicular cells ideally positioned over the spherical geometry of the egg, a drastic, optimized and polarized inward apoptosis sequence directly results from this positioning, suggesting the existence of some apoptotic master cells which control the destiny of neighboring cells. This concept could shed a new light on the origin of massive apoptosis phases that take place during embryogenesis in vertebrates (e.g., cavitation, inter-digitation). It could also be applied to specific therapeutic strategies to fight cancer.

Centrosomal targeting of Syk kinase is controlled by its catalytic activity and depends on microtubules and the dynein motor.

The nonreceptor Syk kinase is detected in epithelial cells, where it acts as a tumor suppressor, in addition to its well-established role in immunoreceptor-based signal transduction in hematopoietic cells. Thus, several carcinomas and melanomas have subnormal concentrations of Syk. Although Syk is mainly localized at the plasma membrane, it is also present in centrosomes, where it is involved in the control of cell division. The mechanisms responsible for its centrosomal localization and action are unknown. We used wild-type and mutant fluorescent Syk fusion proteins in live-cell imaging (fluorescence recovery after photobleaching, total internal reflection fluorescence, and photoactivation) combined with mathematical modeling to demonstrate that Syk is actively transported to the centrosomes via the microtubules and that this transport depends on the dynein/dynactin molecular motor. Syk can only target the centrosomes if its kinase activity is intact and it is catalytically active at the centrosomes. We showed that the autophosphorylated Y130 Syk residue helps to uncouple Syk from the plasma membrane and to promote its translocation to the centrosome, suggesting that the subcellular location of Syk depends on its autophosphorylation on specific tyrosine residues. We have thus established the details of how Syk is trafficked intracellularly and found evidence that its targeting to the centrosomes is controlled by autophosphorylation.

Cell death and renewal during prey capture and digestion in the carnivorous sponge Asbestopluma hypogea (Porifera: Poecilosclerida).

The sponge Asbestopluma hypogea is unusual among sponges due to its peculiar carnivorous feeding habit. During various stages of its nutrition cycle, the sponge is subjected to spectacular morphological modifications. Starved animals are characterized by many elongated filaments, which are crucial for the capture of prey. After capture, and during the digestion process, these filaments actively regress before being regenerated during a subsequent period of starvation. Here, we demonstrate that these morphological events rely on a highly dynamic cellular turnover, implying a coordinated sequence of programmed cell death (apoptosis and autophagy), cell proliferation and cell migration. A candidate niche for cell renewal by stem cell proliferation and differentiation was identified at the base of the sponge peduncle, characterized by higher levels of BrdU/EdU incorporation. Therefore, BrdU/EdU-positive cells of the peduncle base are candidate motile cells responsible for the regeneration of the prey-capturing main sponge body, i.e. the dynamic filaments. Altogether, our results demonstrate that dynamics of cell renewal in sponge appear to be regulated by cellular mechanisms as multiple and complex as those already identified in bilaterian metazoans.

Ezrin promotes actin assembly at the phagosome membrane and regulates phago-lysosomal fusion.

Phagosome maturation is defined as the process by which phagosomes fuse sequentially with endosomes and lysosomes to acquire an acidic pH and hydrolases that degrade ingested particles. While the essential role of actin cytoskeleton remodeling during particle internalization is well established, its role during the later stages of phagosome maturation remains largely unknown. We have previously shown that purified mature phagosomes assemble F-actin at their membrane, and that the ezrin-radixin-moesin (ERM) proteins ezrin and moesin participate in this process. Moreover, we provided evidence that actin assembly on purified phagosomes stimulates their fusion with late endocytic compartments in vitro. In this study, we further investigated the role of ezrin in phagosome maturation. We engineered a structurally open form of ezrin and demonstrated that ezrin binds directly to the actin assembly promoting factor N-WASP (Neural Wiskott-Aldrich Syndrome Protein) by its FERM domain. Using a cell-free system, we found that ezrin stimulates F-actin assembly on purified phagosomes by recruiting the N-WASP-Arp2/3 machinery. Accordingly, we showed that the down-regulation of ezrin activity in macrophages by a dominant-negative approach caused reduced F-actin accumulation on maturing phagosomes. Furthermore, using fluorescence and electron microscopy, we found that ezrin is required for the efficient fusion between phagosomes and lysosomes. Live-cell imaging analysis supported the notion that ezrin is necessary for the fusogenic process itself, promoting the transfer of the lysosome content into the phagosomal lumen.

Differential roles of PtdIns(4,5)P2 and phosphorylation in moesin activation during Drosophila development.

The ezrin, radixin and moesin (ERM) proteins regulate cell membrane architecture in several cellular contexts. Current models propose that ERM activation requires a PtdIns(4,5)P(2)-induced conformational change, followed by phosphorylation of a conserved threonine. However, how these inputs contribute in vivo to orchestrate ERM activation is poorly understood. We addressed this issue by evaluating the contribution of PtdIns(4,5)P(2) and phosphorylation to the regulation of moesin during Drosophila development. Unexpectedly, we found that a form of moesin that cannot be phosphorylated displayed significant activity and could substitute for the endogenous product during wing morphogenesis. By contrast, we also show that PtdIns(4,5)P(2) binding is essential for moesin recruitment to the membrane and for its subsequent phosphorylation. Our data indicate that PtdIns(4,5)P(2) acts as a dosing mechanism that locally regulates ERM membrane recruitment and activation, whereas cycles of phosphorylation and dephosphorylation further control their activity once they have reached the cell cortex.

Topological control of life and death in non-proliferative epithelia.

Programmed cell death is one of the most fascinating demonstrations of the plasticity of biological systems. It is classically described to act upstream of and govern major developmental patterning processes (e.g. inter-digitations in vertebrates, ommatidia in Drosophila). We show here the first evidence that massive apoptosis can also be controlled and coordinated by a pre-established pattern of a specific 'master cell' population. This new concept is supported by the development and validation of an original model of cell patterning. Ciona intestinalis eggs are surrounded by a three-layered follicular organization composed of 60 elongated floating extensions made of as many outer and inner cells, and indirectly spread through an extracellular matrix over 1200 test cells. Experimental and selective ablation of outer and inner cells results in the abrogation of apoptosis in respective remaining neighbouring test cells. In addition incubation of outer/inner follicular cell-depleted eggs with a soluble extract of apoptotic outer/inner cells partially restores apoptosis to apoptotic-defective test cells. The 60 inner follicular cells were thus identified as 'apoptotic master' cells which collectively are induction sites for programmed cell death of the underlying test cells. The position of apoptotic master cells is controlled by topological constraints exhibiting a tetrahedral symmetry, and each cell spreads over and can control the destiny of 20 smaller test cells, which leads to optimized apoptosis signalling.

Alteration of sarcoplasmic reticulum ca release in skeletal muscle from calpain 3-deficient mice.

Mutations of Ca(2+)-activated proteases (calpains) cause muscular dystrophies. Nevertheless, the specific role of calpains in Ca(2+) signalling during the onset of dystrophies remains unclear. We investigated Ca(2+) handling in skeletal cells from calpain 3-deficient mice. [Ca(2+)](i) responses to caffeine, a ryanodine receptor (RyR) agonist, were decreased in -/- myotubes and absent in -/- myoblasts. The -/- myotubes displayed smaller amplitudes of the Ca(2+) transients induced by cyclopiazonic acid in comparison to wild type cells. Inhibition of L-type Ca(2+) channels (LCC) suppressed the caffeine-induced [Ca(2+)](i) responses in -/- myotubes. Hence, the absence of calpain 3 modifies the sarcoplasmic reticulum (SR) Ca(2+) release, by a decrease of the SR content, an impairment of RyR signalling, and an increase of LCC activity. We propose that calpain 3-dependent proteolysis plays a role in activating support proteins of intracellular Ca(2+) signalling at a stage of cellular differentiation which is crucial for skeletal muscle regeneration.

Using Ciona to study developmental programmed cell death.

Ciona intestinalis, a member of Tunicates, the closest group to vertebrates, has emerged as an appropriate organism for the study of developmentally regulated programmed cell death. First, because massive phases of apoptosis occur all along embryogenesis. Second, because the lecithotrophic mode of development is associated with autophagic process occurring during juvenile formation. Third, because the biochemical cell death machinery is close to that found in mammals. Altogether, the Ciona system contributes to identify new specific regulatory pathways and to explain how molecular mechanisms of programmed cell death evolved from invertebrates to vertebrates.

DCC regulates cell adhesion in human colon cancer derived HT-29 cells and associates with ezrin.

The deleted in colorectal cancer (DCC) gene encodes a 170- to 190-kDa protein of the Immunoglobulin superfamily. Firstly identified as a tumor suppressor gene in human colorectal carcinomas, the main function for DCC has been described in the nervous system as part of a receptor complex for netrin-1. Moreover, roles in mucosecretory cell differentiation and as inducer of apoptosis have also been reported. DCC knockout mice supported a crucial role for this gene in axonal migration, yet questioned its implication in tumor suppression and mucosecretory differentiation. The work presented here demonstrates that a DCC-transfected HT-29 colonic human cell line (HT-29/DCC) displays an increase in cell-cell adhesion to the detriment of cell-matrix interactions: HT-29/DCC cells exhibit more and better-structured desmosomes while focal adhesions and hemidesmosomes are disrupted. HT-29/DCC cells show no changes in adherent junctions but upon treatment with TPA, HT-29/DCC cells show resistance to scattering, and maintain E-cadherin in the membrane. In addition, the actin cytoskeleton is affected in HT-29/DCC cells: stress fibers are disrupted while cortical actin remains intact. We identified a putative ERM-M (ezrin/radixin/moesin and merlin) binding domain in the juxtamembrane region of the DCC protein. In vitro pull-down assays demonstrate the interaction of the DCC cytoplasmic domain with the N-terminal region of ezrin and merlin, and co-immunoprecipitation assays in transiently DCC-transfected COS-1 cells showed that the interaction between DCC and ezrin also takes place in vivo. Altogether, our results suggest that DCC could regulate cell adhesion and migration through its association with ERM-M proteins.

Fertilization regulates apoptosis of Ciona intestinalis extra-embryonic cells through thyroxine (T4)-dependent NF-kappaB pathway activation during early embryonic development.

In Ciona intestinalis, the elimination of extra-embryonic test cells during early stage of development is delayed by a fertilization signal. Test cells undergo a caspase-dependent apoptosis event repressed by thyroxine (T4)-activated NF-kappaB. When apoptosis was experimentally blocked, the hatching stage was delayed. The incubation of unfertilized eggs with a 1-h-fertilized egg extract or purified T4 restored apoptosis in test cells at a similar timing than found in fertilized eggs. Ciona expresses specific genes forming a functional IkappaB/NF-kappaB pathway. One, Ci-p65, was transiently induced upon fertilization via T4 and found to exert its anti-apoptotic role in test cells nuclei as well as in a reconstituted cell system. Blocking NF-kappaB activity by dexamethasone-induced overexpression of Ci-IkappaB abrogated the repression of apoptosis in test cells. Overall, the data are consistent for defining a central coupling role of both T4 and NF-kappaB during early embryo development.

The Syk tyrosine kinase localizes to the centrosomes and negatively affects mitotic progression.

We showed previously that the spleen tyrosine kinase Syk is expressed by mammary epithelial cells and that it suppresses malignant growth of breast cancer cells. The exact molecular mechanism of its tumor-suppressive activity remains, however, to be identified. Here, we show that Syk colocalizes and copurifies with the centrosomal component gamma-tubulin and exhibits a catalytic activity within the centrosomes. Moreover, its centrosomal localization depends on its intact kinase activity. Centrosomal Syk expression is persistent in interphase but promptly drops during mitosis, obviously resulting from its ubiquitinylation and proteasomal degradation. Conversely, unrestrained exogenous expression of a fluorescently tagged Discosoma sp. red fluorescent protein (DsRed)-Syk chimera engenders abnormal cell division and cell death. Transient DsRed-Syk overexpression triggers an abrupt cell death lacking hallmarks of classic apoptosis but reminiscent of mitotic catastrophe. Surviving stable DsRed-Syk-transfected cells exhibit multipolar mitotic spindles and contain multiple abnormally sized nuclei and supernumerary centrosomes, revealing anomalous cell division. Taken together, these results show that Syk is a novel centrosomal kinase that negatively affects cell division. Its expression is strictly controlled in a spatiotemporal manner, and centrosomal Syk levels need to decline to allow customary progression of mitosis.

Three-dimensional polarization sensitizes hepatocytes to Fas/CD95 apoptotic signalling.

Maintenance of epithelial cell shape and polarity determines many vital cell functions, including the appropriate response to external stimuli. Murine hepatocytes cultured in a three-dimensional Matrigel matrix formed highly polarized organoids characterized by specific localization of an ERM (ezrin/radixin/moesin) protein, radixin, at microvillus-lined membrane domains. These apical domains surrounded a lumen and were bordered by tight junctions. The hepatocyte organoids were functional as judged by the high level of albumin secretion and accumulation of bilirubin. Stimulation of the Fas/CD95 death receptor, which is highly hepatotoxic in vivo, was a strong inducer of apoptosis in the polarized organoids. This was in sharp contrast to the monolayer hepatocyte cultures, which were protected from death by exacerbated NF-kappaB signalling following engagement of the death receptors. Thus, hepatocytes in polarized, functional organoids modulate an intracellular signal transduction pathway, allowing the recapitulation of their physiological response to an apoptotic stimulus.

Phosphoinositide binding and phosphorylation act sequentially in the activation mechanism of ezrin.

Ezrin, a membrane-actin cytoskeleton linker, which participates in epithelial cell morphogenesis, is held inactive in the cytoplasm through an intramolecular interaction. Phosphatidylinositol 4,5-bisphosphate (PIP2) binding and the phosphorylation of threonine 567 (T567) are involved in the activation process that unmasks both membrane and actin binding sites. Here, we demonstrate that ezrin binding to PIP2, through its NH2-terminal domain, is required for T567 phosphorylation and thus for the conformational activation of ezrin in vivo. Furthermore, we found that the T567D mutation mimicking T567 phosphorylation bypasses the need for PIP2 binding for unmasking both membrane and actin binding sites. However, PIP2 binding and T567 phosphorylation are both necessary for the correct apical localization of ezrin and for its role in epithelial cell morphogenesis. These results establish that PIP2 binding and T567 phosphorylation act sequentially to allow ezrin to exert its cellular functions.

Identification of a novel Ezrin-binding site in syndecan-2 cytoplasmic domain.

ERM (Ezrin/Radixin/Moesin) proteins are crosslinkers between plasma membrane proteins and the actin cytoskeleton, thereby involved in the formation of cell adhesion sites. Earlier work showed that Ezrin links syndecan-2 to the actin cytoskeleton. Here we provide evidence that the Ezrin N-terminal domain binds to the syndecan-2 cytoplasmic domain with an estimated K(D) of 0.71 microM and without the requirement of other proteins. We also studied the regions in the syndecan-2 cytoplasmic domain implicated in the binding to Ezrin. By truncating the syndecan-2 cytoplasmic domain and by oligopeptide competition assays we show that the Ezrin-binding sequence is not located in the positively charged juxtamembrane region (RMRKK), but in the neighboring sequence DEGSYD. We therefore conclude that the consensus sequence for Ezrin binding is unique among membrane proteins, suggesting a distinct regulation.

Mechanical constraint imposed on plasma membrane through transverse phospholipid imbalance induces reversible actin polymerization via phosphoinositide 3-kinase activation.

Platelets were used to explore the effect of membrane curvature induced by phospholipid excess on cell shape and on organization of the actin cytoskeleton. We showed that the addition of short chain analogues of phospholipids to the outer leaflet of plasma membrane of resting platelets immediately induced a shape change with long filopodia formation containing newly polymerized actin. Cells recovered rapidly their discoid shape and their initial F-actin content only with the phosphatidylserine analogue, which was transported to the inner leaflet by aminophospholipid translocase. Filopodia formation and actin polymerization were inhibited in platelets pre-incubated with cytochalasin D. Both wortmannin and LY294002, two unrelated inhibitors of phosphoinositide 3-kinase, considerably reduced actin polymerization and filopodia formation. Phospholipid imbalance was accompanied by a reversible translocation of phosphoinositide 3-kinase from cytoplasm to plasma membrane. In agreement with a role for PI 3-kinase, when phospholipids were added to platelets, PtdIns(3,4)P2 increased two-fold and Akt protein was partly phosphorylated. A similar shape change was also observed in nocodazole-treated L929 fibroblasts which were incubated with the similar phospholipid analogues. In those nucleated cells, where the microtubule cytoskeleton was disrupted, a major actin-dependent membrane extension was induced by addition of short chain phospholipids that required the functional integrity of PI 3-kinase. We conclude that any physical constraint acting on plasma membrane and resulting on local changes in membrane curvature is sufficient to initiate transient actin polymerization via phosphoinositide 3-kinase activation.

Molecular analysis of microscopic ezrin dynamics by two-photon FRAP.

Ezrin plays a key role in coupling signal transduction to cortical cell organization. This actin-membrane linker undergoes a series of conformational changes that modulate its interactions with various partners and its localization in membrane or cytosolic pools. Its mobility and exchange rates within and between these two pools were assessed by two-photon fluorescence recovery after photobleaching in epithelial cell microvilli. Analysis of ezrin mutants with an altered actin-binding site revealed three ezrin membrane states of different mobilities and exchange properties, reflecting sequential association with membrane components and F-actin in the context of a fast overall turnover.

Tail regression in Ciona intestinalis (Prochordate) involves a Caspase-dependent apoptosis event associated with ERK activation.

Two apoptotic events take place during embryonic development of Ciona intestinalis. The first concerns extra-embryonic cells and precedes hatching. The second controls tail regression at metamorphosis, occurs through a polarized wave originating from tail extremity, and is caspase dependent. This was shown by: (1) in vivo incorporation of a fluorescent marker of caspase activation in different cell types of the tail; (2) detection of an activated form of caspase 3-like protein by western blotting; and (3) failure of 30% of larvae to undergo metamorphosis after treatment of fertilized eggs with a pan-caspase inhibitor. In addition, Ciona embryos express a single ERK protein, specifically phosphorylated at metamorphosis. ERK activation was shown to be located in cells of the tail. Addition of MEK inhibitor in the culture medium prevented ERK activation and metamorphosis. In silico analysis of Ciona genome pointed to 15 caspases with high homology with humans, and a single ERK gene with high homology to both mammalian ERK1 and ERK2. It is concluded that the sequence of events leading to metamorphosis includes ERK phosphorylation followed by caspase-dependent apoptosis and tail regression.

Phosphoinositides regulate membrane-dependent actin assembly by latex bead phagosomes.

Actin assembly on membrane surfaces is an elusive process in which several phosphoinositides (PIPs) have been implicated. We have reconstituted actin assembly using a defined membrane surface, the latex bead phagosome (LBP), and shown that the PI(4,5)P(2)-binding proteins ezrin and/or moesin were essential for this process (). Here, we provide several lines of evidence that both preexisting and newly synthesized PI(4,5)P(2), and probably PI(4)P, are essential for phagosomal actin assembly; only these PIPs were routinely synthesized from ATP during in vitro actin assembly. Treatment of LBP with phospholipase C or with adenosine, an inhibitor of type II PI 4-kinase, as well as preincubation with anti-PI(4)P or anti-PI(4,5)P(2) antibodies all inhibited this process. Incorporation of extra PI(4)P or PI(4,5)P(2) into the LBP membrane led to a fivefold increase in the number of phagosomes that assemble actin. An ezrin mutant mutated in the PI(4,5)P(2)-binding sites was less efficient in binding to LBPs and in reconstituting actin assembly than wild-type ezrin. Our data show that PI 4- and PI 5-kinase, and under some conditions also PI 3-kinase, activities are present on LBPs and can be activated by ATP, even in the absence of GTP or cytosolic components. However, PI 3-kinase activity is not required for actin assembly, because the process was not affected by PI 3-kinase inhibitors. We suggest that the ezrin-dependent actin assembly on the LBP membrane may require active turnover of D4 and D5 PIPs on the organelle membrane.