The correct patterning of embryos requires the exchange of information between cells. This is in part achieved by the proper distribution of signaling molecules, many of which exert their function by establishing gradients of concentration. Because of this property they were named "morphogens", or "form giving" substances. Among these, proteins belonging to the Hedgehog (Hh) family have received much attention, owing to their unusual double lipid modification and their involvement in human disease, causing congenital birth defects and cancer. Great efforts have been made in order to elucidate the mechanisms by which Hh molecules are propagated in the embryo. However, no conclusive evidence exists to date to which structures these molecules localize and how they, despite their membrane association, establish a gradient of concentration. Therefore, I decided to study the distribution of the vertebrate Hh homolog, Sonic Hedgehog (Shh) in developing zebrafish embryos. By fluorescently tagging Shh proteins, I found that these localize to discrete punctate structures at the membranes of expressing cells. These were often regions from which filopodial protrusions emanated from the cells. Puctate deposits of Shh were also located outside of expressing cells. In dividing cells, Shh accumulated at the cleavage plane. Furthermore, by making use of confocal microscopy and time lapse analysis, I visualized Shh proteins moving in filopodial extensions present between cells. This suggests a novel mechanism of Shh distribution, which relies on the direct contact of cells by filopodia for the exchange of signaling proteins. In a second part of my thesis, I characterized genes implicated in regulating Shh protein distribution and signaling function. I cloned three zebrafish genes belonging to the Ext1 (exostosin) family of glycosyltransferases required for the synthesis of Heparan Sulfate Proteoglycans and established a tentative link of these genes to somitic Hh signaling. In addition, I characterized the developmental expression and function of zebrafish Rab23, a small GTPase, which acts as a negative regulator of the Shh signaling pathway. Performing knock-down experiments of zebrafish Rab23, I found that Rab23 functions in left-right axis specification, a process previously shown to depend on proper Shh signaling.

Morphogens are secreted signalling molecules that are expressed in restricted groups of cells within the developing tissue. From there, they are secreted and travel throughout the target field and form concentration gradients. These concentration profiles endow receiving cells with positional information. A number of experiments in Drosophila demonstrated that the morphogen Decapentaplegic (Dpp) forms activity gradients by inducing the expression of several target genes above distinct concentration thresholds at different distances from the source. This way, Dpp contributes to developmental fates in the target field such as the Drosophila wing disc. Although the tissue distribution as well as the actual shape and size of the Dpp morphogen concentration gradient has been visualized, the cell biological mechanisms through which the morphogen forms and maintains a gradient are still a subject of debate. Two hypotheses as to the dominant mechanism of movement have been proposed that can account for Dpp spreading throughout the Drosophila wing imaginal target tissue: extracellular diffusion and planar transcytosis, i. e. endocytosis and resecretion of the ligand that is thereby transported through the cells. Here, I present data indicating that implications of a theoreticalanalysis of Dpp spreading, where Dpp transport through the target tissue is solely based on extracellular diffusion taking into account receptor binding and subsequent internalization, are inconsistent with experimental results. By performing Fluorescence Recovery After Photobleaching (FRAP) experiments, I demonstrate a key role of Dynamin-mediated endocytosis for Dpp gradient formation. In addition, I show that most of GFP-Dpp traffics through endocytic compartments at the receiving epithelial cells, probably recycled through apical recycling endosomes (ARE). Finally, a Dpp recycling assay based on subcellular photouncage of ligand is presented to address specifically the Dpp recycling event at the receiving cells.

During cell communication, cells produce secreted signals termed morphogens, which traffic through the tissue until they are received by target, responding cells. Using the fruit fly Drosophila melanogaster as a model organism, I have studied transforming growth factor-beta (TGF-beta) signal from the secreting to the receiving cells in the developing wing epithelial cells and at the neuromuscular junctions. Cell culture studies have suggested that cells modulate morphogenetic signaling by expressing the receptors and secreting the ligand in spatially defined areas of the cell. Indeed, I have found that TGF-beta ligands, receptors and R-Smads show a polarized distribution both in the epithelial cells and at the synapses. My results indicate that the cellular junctions define a signaling domain within the plasma membrane, to which TGF-beta signaling machinery is targeted. In the context of epithelial cells, the junctions play a role in TGF-beta signaling regulation through their component beta-cat. A complex forms between beta-cat and the R-Smad Mad, but the mechanism by which beta-cat modulates signaling is not yet understood. At the synapse, the sub-cellular localization of TGF-beta pathway components indicates the occurrence of an anterograde signal. Moreover, my results suggest a scenario in which TGF-beta signaling is coupled with synaptic activity: quanta of growth factor, released upon neurostimulation together with neurotransmitter quanta, could modulate therefore the development and the function of the synapse.

The development of a multicellular organism, such as a human or an animal, begins with the fertilization of an egg cell. Thereupon the organism grows by repeated cell divisions until the adult size is reached and growth stops. Although it is known that intrinsic mechanisms determine the final size of developing organs and organisms, the basic principles of growth control are still poorly understood. However, there is strong evidence that certain morphogens, which are a special class of signaling molecules, act as growth factors and play a key role in growth control. In this work, growth control is studied from a mainly theoretical viewpoint. A discrete vertex model describing the organization of cells by a network of polygons is used, including a description of the cell cycle and a description of dynamical morphogen distributions. Self-organized growth is studied by introducing growth rules that govern cell divisions based on the local morphogen level. This discrete description is complemented by a continuum theory to gain further insight into the dynamics of self-organized growth processes. The theoretical description is applied to the developing wing of the fruit fly Drosophila melanogaster. In the developing wing, which is an epithelium consisting of single-layered cell sheets, the morphogen Decapentaplegic (Dpp) acts as a key growth factor. Experimental data shows that the Dpp distribution is dynamic and adapts to the size of the developing wing. Two mechanisms that rely on a regulatory molecule species and lead to such a dynamic behaviour of the Dpp distribution are studied. Several growth rules are tested and the resulting growth behaviour is quantitatively compared to experimental data of the developing wing. A particular growth rule, that triggers a cell division when the local morphogen level has increased by a certain relative amount, is found to be consistent with experimental observations under normal and several perturbed conditions. It is shown that mechanical stresses that arise due to spatial growth inhomogeneities can have a stabilizing effect on the growth process.

Morphogens are secreted signalling molecules that are expressed in restricted groups of cells within the developing tissue. From there, they are secreted and travel throughout the target field and form concentration gradients. These concentration profiles endow receiving cells with positional information. A number of experiments in Drosophila demonstrated that the morphogen Decapentaplegic (Dpp) forms activity gradients by inducing the expression of several target genes above distinct concentration thresholds at different distances from the source. This way, Dpp contributes to developmental fates in the target field such as the Drosophila wing disc. Although the tissue distribution as well as the actual shape and size of the Dpp morphogen concentration gradient has been visualized, the cell biological mechanisms through which the morphogen forms and maintains a gradient are still a subject of debate. Two hypotheses as to the dominant mechanism of movement have been proposed that can account for Dpp spreading throughout the Drosophila wing imaginal target tissue: extracellular diffusion and planar transcytosis, i. e. endocytosis and resecretion of the ligand that is thereby transported through the cells. Here, I present data indicating that implications of a theoreticalanalysis of Dpp spreading, where Dpp transport through the target tissue is solely based on extracellular diffusion taking into account receptor binding and subsequent internalization, are inconsistent with experimental results. By performing Fluorescence Recovery After Photobleaching (FRAP) experiments, I demonstrate a key role of Dynamin-mediated endocytosis for Dpp gradient formation. In addition, I show that most of GFP-Dpp traffics through endocytic compartments at the receiving epithelial cells, probably recycled through apical recycling endosomes (ARE). Finally, a Dpp recycling assay based on subcellular photouncage of ligand is presented to address specifically the Dpp recycling event at the receiving cells.

During cell communication, cells produce secreted signals termed morphogens, which traffic through the tissue until they are received by target, responding cells. Using the fruit fly Drosophila melanogaster as a model organism, I have studied transforming growth factor-beta (TGF-beta) signal from the secreting to the receiving cells in the developing wing epithelial cells and at the neuromuscular junctions. Cell culture studies have suggested that cells modulate morphogenetic signaling by expressing the receptors and secreting the ligand in spatially defined areas of the cell. Indeed, I have found that TGF-beta ligands, receptors and R-Smads show a polarized distribution both in the epithelial cells and at the synapses. My results indicate that the cellular junctions define a signaling domain within the plasma membrane, to which TGF-beta signaling machinery is targeted. In the context of epithelial cells, the junctions play a role in TGF-beta signaling regulation through their component beta-cat. A complex forms between beta-cat and the R-Smad Mad, but the mechanism by which beta-cat modulates signaling is not yet understood. At the synapse, the sub-cellular localization of TGF-beta pathway components indicates the occurrence of an anterograde signal. Moreover, my results suggest a scenario in which TGF-beta signaling is coupled with synaptic activity: quanta of growth factor, released upon neurostimulation together with neurotransmitter quanta, could modulate therefore the development and the function of the synapse.

Janus Kinase (JAK) activity specifies the cell fates of the follicular epithelium during Drosophila oogenesis by establishing a gradient of JAK activity with highest levels at the A/P poles. Unpaired (Upd), a ligand for the pathway, is expressed and secreted exclusively from the polar cells potentially establishing the JAK activity gradient. This project proposed that Upd acts as a morphogen to directly establish the JAK activity gradient, specifying the fates of the follicular epithelium. The aims of this work were to investigate the extracellular distribution of Upd and, in addition, factors that may be involved. Furthermore, upd3, a gene encoding a protein with sequence similarity to Upd, is also co-expressed with upd in the polar cells. An additional aim of this project was to determine what role, if any, Upd3 plays in follicular development. Immunostaining was used to reveal Upd distribution during oogenesis. The data revealed an Upd gradient on the apical membrane of the follicular epithelium. By virtue of the extracellular gradient, Upd fulfills the requirements necessary to be classified as a morphogen. Some morphogens are dependent on heparan sulphate proteoglycans (HSPGs) for distribution. Using mitotic recombination to make mosaics, this work reveals that Dally, a glypican, is essential for the distribution of Upd and establishment of the JAK gradient during oogenesis. The data suggests Dally is involved with stability of extracellular Upd. Mosaic analysis of an additional HSPGs revealed that they are not essential for the Upd gradient or JAK activity during oogenesis. upd3 mutant flies have small eyes and outstretched wings, a phenotype consistent reduced JAK activity. In upd3 mutant ovaries it is shown that there is a higher frequency of deteriorating egg chambers, a higher frequency of egg chamber fusions, and a decrease in border cells per egg chamber compared to wildtype controls; all of which support a reduction of JAK activity. Furthermore, ovarian phenotypes of upd3 get worse as the fly ages suggesting that upd3 is required over time. The data presented suggests that Upd3 does act to maintain JAK activity in the ovary as the fly ages.

La voie de signalisation activée par la protéine Sonic Hedgehog (Shh) est connue pour son rôle majeur au cours de l’embryogenèse et en particulier dans la prolifération et la spécification cellulaire ou encore le guidage axonal au cours de l’établissement des structures du système nerveux. Depuis quelques années, ce morphogène a aussi été identifié comme un régulateur important de plusieurs processus physiologiques du cerveau adulte comme le maintien de la neurogenèse ou la régulation de l’activité électrique de certains neurones (Traiffort et al., 2010). La suractivation de la voie Shh dans un cerveau sain entraine une augmentation significative de la prolifération des cellules progénitrices des oligodendrocytes (OPCs), la source des oligodendrocytes matures, les cellules responsables de la formation des gaines de myéline (Loulier et al., 2006). Au cours de ma thèse, j’ai étudié le potentiel que représente l’activation de la voie Shh dans la régulation de ces progéniteurs dans un contexte de démyélinisation. Pour cela, j’ai utilisé une souris transgénique plp-GFP, chez laquelle la protéine fluorescente verte est exprimée par les cellules du lignage oligodendrocytaire. Après avoir caractérisé le profil d’expression de la GFP dans le cerveau mature de ces souris, j’ai mis au point un modèle de démyélinisation focale par injection stéréotaxique d’un détergent spécifique de la myéline, la lysolécithine (LPC). J’ai identifié les cellules du lignage oligodendrocytaire comme source directe de protéines Shh au sein de la lésion à un temps très précoce après l’injection de LPC. Les gènes cibles de la voie Shh sont aussi fortement induits dans cette population cellulaire à une période plus tardive, correspondant à la différenciation des OPCs en cellules matures. L’utilisation d’adénovirus codant soit pour Shh lui-même soit pour son antagoniste physiologique Hip, m’a permis de réaliser des expériences de gain et de perte de fonction et ainsi d’analyser comment la modulation de la voie Shh peut influencer sur le processus de régénération des oligodendrocytes suite à une lésion. La surexpression de Shh permet d’augmenter la prolifération des OPCs mais aussi d’accélérer leur différenciation, aboutissant à un nombre plus élevé d’oligodendrocytes matures plus précocement au cours du processus de remyélinisation. De plus, il est intéressant de constater que la densité des cellules astrocytaires et microgliales, notamment associées au processus inflammatoire, diminue dans la lésion chez les animaux ayant reçu l’adénovirus Shh comparés au animaux contrôles. A l’inverse, le blocage de la voie induit l’arrêt complet de la production de nouveaux oligodendrocytes. Au-delà de l’amélioration de notre compréhension de la physiologie et de la régulation du lignage oligodendrocytaire dans le cerveau adulte, l’ensemble de ce travail montre de quelle manière la voie Shh peut représenter une nouvelle piste dans la recherche de cibles thérapeutiques dans les affections de la myéline telles que la sclérose en plaques. / The Sonic Hedgehog (Shh) signaling pathway is known for its role during embryogenesis and in particular for controlling cell proliferation and specification, as well as axon guidance. In recent years, this morphogen has also been identified as an important regulator of several physiological processes in the adult brain such as the maintenance of neurogenesis or the regulation of the electrophysiological propreties of mature neurons (Traiffort et al., 2010). Overactivation of the Shh pathway in a healthy brain causes a significant increase in the proliferation of oligodendrocyte progenitor cells (OPCs), the source of mature oligodendrocytes, the cells responsible for the formation of myelin sheaths (Loulier et al., 2006).In my thesis, I studied the effects of the Shh pathway activation on OPC regulation in the context of demyelination. To that purpose, I used a plp-GFP transgenic mouse, in which the green fluorescent protein (GFP) is expressed by cells belonging to the oligodendrocyte lineage. After characterization of the expression pattern of GFP in the mature brain of these mice, I developed a model of focal demyelination by stereotaxic injection of lysolecithin (LPC). I identified the oligodendrocyte lineage cells as a source of Shh protein within the lesion, soon after the LPC injection. Target genes of the Shh pathway are also strongly induced in this cell population, at a time corresponding to the differentiation of OPCs into mature cells. The use of adenoviral vectors encoding either Shh itself or its physiological antagonist Hip allowed me to conduct gain- and loss-of-function experiments. This way I could analyze how the modulation of Shh pathway may influence the regeneration ofoligodendrocytes after injury. Shh overexpression increases the survival and proliferation of OPCs but also accelerates their differentiation, resulting in a higher number of mature oligodendrocytes earlier during the remyelination process. In addition, the density of astrocytes and microglia, associated with the inflammatory process, is decreased in animalsreceiving the Shh adenoviral vector compared to control animals. Altogether these effects are associated with a reduction of the lesion. Conversely, blocking the pathway induced a complete arrest of new oligodendrocyte production. Besides the fundamental knowledge gained about the molecular mechanism involved in the oligodendroglial precursor cells survival, proliferation, differentiation and myelin repair in vivo, this project should also give valuable insights concerning the potential use of pharmacological modulators of Shh signaling as a novel therapeutic approach for the treatment of multiple sclerosis and other myelin diseases.

Les gradients morphogénétiques contrôlent l'émergence de polarités axiales au cours du développement. Bien que la dynamique d'établissement de ces gradients soit bien comprise, la précision des mécanismes d'activation agissant en aval restent à élucider. Nous abordons cette question avec le gradient de Bicoid qui fournit rapidement une réponse transcriptionnelle robuste dans l'embryon de drosophile. Cette robustesse survient malgré le challenge imposé par de fréquentes mitoses dépourvues de transcription. Un calcul théorique intégrant les paramètres physiques de Bicoid (concentration, diffusion) indique que la mesure précise de concentration de Bicoid ne peut être effectuée à chaque interphase en 5-6mn. Il a donc été proposé que l'acquisition rapide de cette robustesse repose sur une mémorisation de l'état transcriptionnel au cours des divisions. Pour tester cette hypothèse, j'ai adapté à l'embryon de drosophile le système MS2 d'étiquetage des ARN dans les cellules vivantes et démontré qu'il permettait de suivre la dynamique transcriptionnelle dans un organisme pluricellulaire vivant. De manière inattendue, le rapporteur MS2 s'exprime aussi postérieurement ce qui m'a empêché de tester l'hypothèse de mémorisation. J'ai montré que cette expression postérieure est due à la présence de sites de fixation pour le facteur de transcription Zelda dans la cassette MS2. Un nouveau rapporteur MS2, dépourvu de ces sites récapitule l'expression endogène et fournit un outil de choix pour tester l'hypothèse de mémorisation. Ce travail ouvre de nouvelles perspectives pour comprendre la dynamique transcriptionnelle sur laquelle repose l'émergence des patrons d'expression développementaux. / Morphogen gradients provide concentration-dependent positional information along polarity axes. Although the dynamics of these gradients is well described, precision and noise in the activation processes acting downstream remain unclear. To address this question, we study the response to the Bicoid gradient that elicits very rapidly a robust transcriptional response in young fly embryos. This robustness occurs despite the challenge imposed by frequent mitoses during which transcription is interrupted suggesting that nuclei measure the Bicoid concentration during the 5-6 mn interphases. Modeling using statistical mechanics and Bicoid physical parameters do not account for accurate measurement of Bicoid concentration in such a short period. It was proposed that rapid robustness of the Bicoid response relies on a memorization process allowing nuclei to recall Bicoid concentration from previous cycles. To understand how the Bicoid system resists to the challenge imposed by mitosis, I have adapted the MS2 RNA-tagging approach to fly embryos and shown that it can be used to quantify transcription dynamics in a living multicellular organism. Unexpectedly, the MS2 reporter was also expressed in the posterior of the embryo making it impossible to directly test the memorization hypothesis. I have shown that this posterior expression is due to binding sites for the transcription factor Zelda unexpectedly localized in the MS2 cassette. A newly engineered MS2 reporter removing those sites faithfully reproduces the endogenous expression providing a powerful tool to test the memory hypothesis. This work opens new avenue to decipher the transcription dynamics underlying pattern formation.

Heparan sulphate proteoglycans (HSPGs) are cell surface/secreted molecules expressed by all cells. HSPGs consist of carbohydrate side-chains attached to a core protein and are involved in regulating key signalling pathways in the developing mammalian brain via sugar-protein interactions. It has been hypothesized, in the ‘heparan sulphate (HS) code hypothesis’, that the specificity for the interaction between the HSPGs and particular signalling pathways is encoded by its HS side-chain. HS has an enormous variety of structures due to postsynthetic modification. Hs2st and Hs6st1 are enzymes involved in generating different HS structures by sulphating the 2-carbon or 6-carbon molecule of the sugar backbone respectively. Fibroblast growth factors (Fgfs) are a family of signalling molecules crucial for forebrain development. Some of its members such as Fgf8 are morphogens which pattern the forebrain via regulated gradient formation while others such as Fgf2 drive neurogenesis and cell proliferation. One of the main molecular consequences of Fgf signalling is activation of extracellular signal-regulated kinase (Erk) where the activation of Erk then drives developmental events such as neurogenesis or cell migration. Based on previous studies on the HS code hypothesis, we hypothesized that differential sulphation regulates Fgf signalling in a specific manner depending on the HS sulphation pattern. We performed binding assays on Hs2st-/- mice to ascertain the molecular mechanism behind the role of differential sulphation in Erk signalling through Fgf2 in the forebrain. We found that differential sulphation also has an important role to play in regionally targeting Fgf2/Erk signalling through regulating the formation of active signalling complexes. Studying the Fgf8/Erk signalling axis at E14.5 developing mouse corticoseptal boundary (CSB) revealed increased Fgf8 levels and Erk hyperactivation in both Hs2st and Hs6st1 null mutants. The dysregulation of Fgf8/Erk signalling at the CSB also highly correlates with the high expression of Hs2st and Hs6st1 at the CSB. A closer look into the molecular phenotypes of Hs2st-/- and Hs6st1-/- CSB revealed differences between them in which Hs6st1-/- CSB has higher Fgf8 levels compared to Hs2st-/- CSB. To elucidate the mechanisms underlying Hs2st and Hs6st1 role at the CSB, we investigated the formation and interpretation of Fgf8/Erk signalling gradient using Fgf8 bead assays in mice with Hs2st and Hs6st1 loss of function throughout development. We found that differential sulphation has a complex effect on Fgf8 gradient formation and interpretation in the forebrain in which Hs2st acts to stabilise the Fgf8 distribution through regulating Fgf8 levels through time while Hs6st1 acts to stabilise the Fgf8 distribution by maintaining the shape of the Fgf8 gradient through restricting Fgf8 levels during the formation of the Fgf8 distribution. In addition, we found Hs2st and Hs6st1 both function to increase the sensitivity of the CSB to Fgf8 for an Erk response although through different modes of action. Therefore, we conclude that differential HS sulphation plays a specific role in Fgf/Erk signalling depending on the HS sulphation pattern.