<p>Renal dysplasia, a developmental disorder characterized by defective nephrogenesis and branching morphogenesis, ranks as one of the major causes of renal failure among the pediatric population. The molecular mechanisms underlying the pathogenesis of renal dysplasia are not well understood; however, changes in gene expression are a major contributing factor. In this study, we demonstrate that the levels of activated β-catenin, a transcriptional co-regulator, are elevated in the nuclei of ureteric, stromal, and mesenchymal cells within dysplastic human kidney tissue. To determine the mechanisms by which mesenchymal β-catenin over-expression leads to renal dysplasia, we generated a conditional mouse model in which β-catenin was stabilized exclusively in the metanephric mesenchyme. Kidneys from these mutant mice are remarkably similar to dysplastic human kidneys. In addition, these mutant mice also demonstrate the formation of 4 to 6 ectopic kidneys. While nephrogenesis appeared normal, investigation of ureteric branch pattern revealed ectopic ureteric budding off the Wolffian duct, ectopic branching off the initial ureteric bud stalk and a disorganization of branch patterning. In-situ hybridization of mutant kidneys revealed increased expression of Gdnf, Cret, and Wnt11, key factors that regulate ureteric branch patterning. We further demonstrate that β-catenin directly binds to TCF consensus binding sites within the Gdnf promoter region located 4.9kb, 2.25kb and 2.1kb upstream of the Gdnf transcriptional start site. Molecular cloning of the 4.9kb fragment upstream of a luciferase gene revealed that ß-catenin regulates gene transcription from the 4.9kb consensus site. Consistent with these findings, genetic deletion of β-catenin from the metanephric mesenchyme cell lineage lead to decreased Gdnf expression and a reduction in ureteric branching morphogenesis resulting in renal hypoplasia. Taken together, our findings establish that β-catenin is an essential regulator of Gdnf expression within the metanephric mesenchyme. Furthermore, we have identified a novel disrupted signalling pathway that contributes to the pathogenesis of renal dysplasia. In this pathway, an over-expression of β-catenin directly leads to an over-expression of Gdnf, causing ectopic and disorganized branching morphogenesis and, consequently, renal dysplasia.</p> / Master of Health Sciences (MSc)

Three-dimensional analysis of mandibular growth provides the potential for pedodontists, orthodontists and surgeons to prescribe treatment that works in harmony with the individual growth of the patient. Despite efforts by 3D pioneers, the visualization of growth and development remains reminiscent of the landmark-based cephalometric analyses used in two-dimensional studies. The objective of this study was to identify 3D topographical changes of the mandible during growth and development of adolescent orthodontic patients. Nine pairs of pre- and post-orthodontic cone-beam computed tomography (CBCT) scans were used to generate mandibular surfaces. Surfaces were superimposed on trabecular bone in the anterior mandible using a mutual information algorithm, and topographical changes were visualized and quantified. The intra- and inter-rater intraclass correlation coefficients for surface generation (0.94 and 0.93, respectively) and superimposition (0.96 and 0.82, respectively) demonstrate the reliability of the techniques. The findings of this study support the theories of bone remodeling reported in histological, implant-based, and landmark studies of mandibular growth. / Oral Biology

Proper formation of the heart is a critical developmental event which requires strict regulation of coordinated cardial cell adhesion, alignment, and migration. The simple, tube-like heart of the fruit fly, Drosophila melanogaster, has proven to be an attractive system in which to study the regulatory pathways which control cardiogenesis. This is mainly due to its strikingly similarity to the vertebrate heart during early embryogenesis. In addition, many genes identified in association with congenital heart disease in humans have homologues in Drosophila, suggesting that this model organism has great potential to contribute to cardiovascular research. The extracellular matrix protein encoded by slit is a ligand for the receptors Robo, and Robo2 (lea). Recently, a third receptor for Slit has been identified as the heparin sulfate proteoglycan Syndecan. The main objective of this thesis was to use time lapse confocal imaging in order to develop further understanding of the mechanisms which result in heart assembly defects in slit, robo, lea, and syndecan mutants. We also aimed to gain a better understanding of the role of Syndecan within the Slit-Robo pathway and elucidate its relative contribution to development of the mature heart. In mutants homozygous for slit, as well as mutants doubly heterozygous for robo and lea, cardial cell alignment, adhesion, and synchronized migration were disrupted. The heart phenotype of syndecan homozygous mutants was similar that of slit and robo, lea, however the migration speed of cells to the midline did not seem to be affected. Based on our findings, we hypothesize that Slit may have Syndecan-dependent and Syndecan-independent functions in the heart. / Thesis / Master of Science (MSc)

Gastrulation involves a complex series of morphogenetic processes that unfold in precise coordination alongside tissue genetic patterning. During the gastrulation of Drosophila melanogaster, a monolayer of cells known as the blastoderm undergoes a sequence of tissue rearrangements that reveal axis determination and cell fates. The Drosophila model provides a comprehensive toolkit for studying embryogenesis, offering insights into its streamlined developmental program, which efficiently produces the final organism while maintaining robustness. This prompts an exploration of the mechanisms that coordinate individual morphogenetic events. Active forces, generated by cytoskeletal-driven cell shape changes, exert themselves on the viscoelastic blastoderm tissue, resulting in passive transmission. Traditional analysis involved studying these events at molecular, cellular, and tissue scales. However, the emergence of single-plane illumination microscopy, enabling whole volume (in toto) imaging of developing embryos with high temporal and spatial resolution, has reshaped this approach. This advancement in microscopy allows us to investigate how morphogenetic events synergise or hinder one another and how the embryo integrates this information to coordinate gastrulation. A recent development in gastrulation research involves identifying regions of heightened friction between the cells and the eggshell within the embryo. Integrins expressed near the midgut primordium have been found to mediate this interaction, challenging the notion that cell movement is the sole driving force of morphogenesis. Instead, static regions emerge as vital contributors to mechanical stability and the preservation of left-right symmetry during germ band extension. This role of integrins is not confined to Drosophila alone, as similar findings have been observed in the beetle Tribolium castaneum, emphasizing their role in orchestrating tissue flows and raising questions about their conservation. Consequently, my focus centered in understanding the physical forces underlying gastrulation, encompassing their interplay, balance, and conflicts in the developing embryo. I considered not only the active forces generated by cell shape changes, but also the passive forces transmitted in the surrounding tissue and the friction generated by the interaction with the physical constraint of the eggshell. In the initial chapter of this thesis, I delved into a lesser-explored morphogenetic event: the formation of the cephalic furrow. This transient tissue fold delineates the head from the trunk but differs from other invaginations in that it does not give rise to further differentiated structures. Instead, it unfolds later in development without leaving a trace of its origin. By employing in toto imaging alongside genetic and photo-manipulation techniques, I investigated the influence of the active forces driving the furrow formation on the head-trunk boundary tissue, and the consequences upon their absence. I found that the invagination is driven by active forces that propagate in the surrounding tissue. By genetic inhibition of the furrow, the presence of ectopic buckling was confirmed in the head-trunk boundary, appearing in between mitotic domains in a rather stochastic fashion. Next, I assessed the impact of further morphogenetic events in the head-trunk boundary tissue using whole embryo imaging and quantitative strain analysis. From my observations, I was able to conclude that the cephalic furrow primes tissue folding to dissipate forces coming from two sources, the local mitotic domains' expansion and the remote germ band compression. These discoveries suggest that the cephalic furrow and its genetic patterning may have evolved in response to the coalition of compressive forces in the head-trunk boundary tissue. The second chapter focused on understanding the ensemble of tissue flows while considering the newfound importance of static regions, mediated by the integrin subunit scab. The possibility to visualise the whole embryo upon mechanical photo-manipulation and integrin mutation, allowed me to uncover how enhanced friction's localised forces contribute to the directionality of tissue movements during the posterior midgut invagination. I was able to verify that cell-to-shell attachment helps maintaining the speed and direction of germ band extension, a highly mechanically unstable process. At the same time, I explored the effects of two milder expression sites of scab not addressed in the past on the mid-dorsal and ventral-anterior region of the embryo. I determined that the first contributes to the shift of the cephalic furrow posteriorly, which is required to allow mitotic domains divisions in the dorsal side of the head. Upon integrin depletion, ectopic buckling in between mitotic domains was observed, resembling our results in the first chapter. Lastly, thanks to whole volumetric imaging, the newly found expression site in the ventral-anterior region was determined to stabilise the head in face of torque generation of the deviating germ band. These results indicate a potential early safety mechanisms against symmetry breaking in the unstable stages of germ band extension, before posterior midgut invagination. In the concluding chapter, I explored the origins of left-right organismal symmetry instability during germ band extension. Analysing cartographic projections of the whole blastoderm surface, I discovered an inherent chirality identified in symmetric embryos at both the cellular and local tissue levels. By quantification of geometric features, along with tissue strain and curl rates, I found significant and consistent differences in all specimens in a specific region of the embryo in the lateral posterior side. These finding raises questions about the potential role of early chiral determinants in embryogenesis, that could translate local asymmetries into organismal twisting. In summary, this thesis underscores the significance of a multi-scale and interdisciplinary approach to embryonic development. The view of genetic patterning setting up the canvas for morphogenesis is taken to the next level, considering this canvas as an active material that needs fine coordination of single cellular events and the forces generated by them. These forces, in turn, shape molecular components and gene expression, culminating in a dynamic picture of a somewhat unstable, though robust, equilibrium of embryonic development.:I Introduction 1 Tissue morphogenesis 2 Embryonic development of Drosophila melanogaster 3 Modern methods for volumetric imaging 4 Project motivation and questions II Results 5 The role of the cephalic furrow. 6 Integration of tissue flows and static regions 7 Left-Right (a)symmetry 8 Discussion 9 Appendix

La morphogenèse de la glande mammaire résulte de la coordination de différentes voies, incluant l'apoptose, la prolifération, la différenciation et la dynamique des cellules souches/progénitrices. La transition épithéliale-mésenchymateuse (EMT) semble être impliquée dans ces voies de signalisation. Ainsi, nous nous sommes concentrés sur le facteur de transcription Slug, un gène clé régulant l'EMT, et son implication dans la morphogenèse de la glande mammaire. Dans un premier temps, en utilisant un modèle de souris transgéniques Slug-Lacz, nous avons localisé Slug dans une sous-population couvrant 10 à 20% des cellules basales du tubule et des cap cells du bourgeons terminal, coexprimant les marqueurs P-cadhérine, CK5, CD49f. Ensuite, nous avons montré par des expériences in vitro de perte et de gain de fonction, que Slug régulait la différenciation et la prolifération des cellules épithéliales mammaires. De plus, nous avons trouvé que Slug inhibait l'apoptose, promouvait la motilité cellulaire, et permettait l'émergence et la croissance de mammosphères clonales. Ce dernier point montre l'implication de Slug dans les cellules souches, qui est renforcé par le fait que des cellules primaires déficientes pour Slug étaient incapables de donner des mammosphères secondaires. Par ailleurs, nous avons pu observer in vivo que les souris déficientes pour Slug présentaient un retard de développement de la glande mammaire, possédant moins de cellules en prolifération, et une surexpression des marqueurs des cellules luminale CK8/18, GATA3 et ER. D'autres gènes régulant l'EMT sont retrouvés surexprimés, suggérant un mécanisme de compensation, qui peut expliquer le fait que le retard de développement de la glande mammaire est rattrapé à l'âge adulte. Les glandes mammaires Slug-knockout présentaient également des branchements excessifs, évoquant une différenciation précoce, similaire aux glandes mammaires de souris déficientes pour la P-cadhérine, exprimée dans les cellules basales. Sachant cela, nous avons constaté que la P-cadhérine était diminuée dans les glandes mammaires Slug-knockout, et dans les cellules CommaDβ traitées par siRNA ciblant Slug. Nous avons alors trouvé que Slug se liait directement au promoteur de la P-cadhérine et l'activait, et que cette dernière intervenait dans certains effets fonctionnels de Slug, tels que la croissance de mammosphères, la différenciation et la migration cellulaire. Ainsi, nous avons montré l'importance d'une nouvelle voie de signalisation Slug/P-cadhérine dans les capacités souches/progénitrices des cellules épithéliales mammaires, intégrant la différenciation et la motilité cellulaire, et nous avons maintenant une meilleure compréhension de son rôle dans l'agressivité de certains cancers du sein. / Mammary gland morphogenesis results from the coordination of different pathways, including apoptosis, proliferation, differentiation, and stem/progenitor cell dynamics. Epithelial-mesenchymal transition (EMT) appears to be involved in these signalling pathways. Thus, we focused on transcription factor Slug, a key gene regulating EMT, and its involvement in mammary gland morphogenesis. First, using a Slug–LacZ transgenic mice model, we located Slug in a subpopulation covering about 10–20% basal duct cells and cap cells of terminal end bud, coexpressed with basal markers P-cadherin, CK5 and CD49f. Then, we have shown by in vitro experiments of loss and gain of function that Slug regulated the differentiation and proliferation of mammary epithelial cells. Moreover, we found that Slug inhibited apoptosis, promoted cell motility, and allowed the emergence and growth of clonal mammospheres. This last point shows the involvement of Slug in stem cells, which is reinforced by the fact that primary cells deficient for Slug were unable to give secondary mammospheres. Furthermore, we observed in vivo that mice deficient for Slug showed delayed development of the mammary gland, with less proliferating cells, and overexpression of markers of luminal cells CK8/18, GATA3 and ER. Other genes regulating EMT are found overexpressed, suggesting a compensatory mechanism, which can explain the fact that the delayed development of the mammary gland is caught up in adulthood. The Slug-knockout mammary glands also showed overbranching, evoking an early differentiation, similar to the mammary glands of mice deficient in P-cadherin, expressed in the basal cells. Knowing this, we found that P-cadherin was decreased in Slug-knockout mammary glands, and in CommaDβ cells treated with siRNA targeting Slug. We then found that Slug binds directly to the promoter of the P-cadherin and activated it, and that P-cadherin was involved in some functional effects of Slug, such as mammospheres growth, differentiation and cell migration. Thus, we have shown the importance of a new signalling pathway Slug/P-cadherin in the capacity of mammary epithelial stem/progenitor cells, integrating differentiation and cell motility, and we now have a better understanding of its role in the aggressiveness of some breast cancers.

Le bassin supérieur de l’oued Djemaa qui fait partie de la montagne du Djurdjura est affecté par une importante morphogenèse. Les mouvements de terrain présentent une grande variété dans le secteur d’étude et jouent un rôle prépondérant dans l’évolution des versants. L’étude géomophologique détaillée, une enquête auprès des populations locales, des analyses géotechniques et l’analyse sédimentologique des formations déplacées nous ont permis de faire une typologie des mouvements de terrain et de dégager les facteurs de leur développement et de leur localisation. Les fortes précipitations d’hiver et de printemps sont à l’origine de tous les mouvements de terrain et des ravinements, mais l’exposition et la lithologie permettent de comprendre les formes différentes rencontrées sur les adrets et sur les ubacs. Les activités humaines jouent un rôle croissant dans l’accélération de la morphogenèse. Dans cette étude, un système d’Information Géographique a été utilisé pour produire les cartes des risques de dégradation du milieu qui sont des outils pour la planification et la gestion de l’utilisation du sol dans le bassin supérieur de l’oued Djemaa. Il permet de croiser la vulnérabilité avec l’aléa. Le résultat de ces croisements que nous avons codifiés débouche sur une carte globale de niveaux de risque / The upper basin of the oued Djemaa which includes to the mountain of Djurdjura is affected by an important morphogenesis. The ground movements introduces a large variety in the study area and play a preponderant role in the evolution of mountainsides. The geomorphology detailed study an investigation into the local populations, geotechtonic and analyses it sedimentology of displaced trainings allows us to do ground typology movement and to chow theire development factors and their location. The big rainfalls of winter and sprinter are the origin of these ground movements and gully erosions, but exposition and lithology allow to understand different forms on northern and southern slope. Human activities play more and more an important role in morphogenesis acceleration. In this study, a geographical information system has been used to provide degradation maps of risks of which are tools for planning and management of soil using in the upper basin of the oued Djemaa. He allows to cross vulnerability witch vagary. The result of these crossroads which we codified results in a total map of risk levels

Les épithéliums jouent le rôle fondamental de barrière physique et chimique chez les Métazoaires. Les jonctions adhérentes, par le biais de la protéine transmembranaire E-cadhérine (E-cad), assurent une grande partie de l’adhésion intercellulaire. Malgré cette robustesse, les épithéliums peuvent subir des remodelages considérables pendant l’embryogenèse ou la cicatrisation. Lors de la gastrulation de l’embryon de Drosophile, l’épithélium ventro-latéral (la bandelette germinale) subit une élongation le long de l’axe antéropostérieur induite par l’intercalation cellulaire. Le remodelage polarisé des jonctions cellulaires est à la base de ce phénomène: les jonctions parallèles à l’axe dorsoventral (DV) rétrécissent et forment de manière irréversible de nouvelles jonctions parallèles à l’axe antéropostérieur (AP). Ce remodelage dépend de l’enrichissement du moteur moléculaire Myosine II (MyoII) dans les jonctions DV, qui induit une anisotropie de tension. Les protéines des jonctions adhérentes (E-cad, &#946;-catenin) sont, elles aussi, polarisées : elles sont enrichies dans les jonctions AP. Néanmoins, nous ne savions pas si cette polarité de l’adhésion avait un rôle dans le remodelage des jonctions, et nous ne connaissions pas les mécanismes contrôlant cette localisation asymétrique. L’un des mécanismes les mieux connus de la modulation de l’adhésion cellulaire est l’endocytose des protéines d’adhésion. A ce titre, je me suis intéressé au rôle de l’endocytose Clathrine dépendante (ECD) pendant l’intercalation cellulaire. J’ai ainsi pu montrer que l’ECD de E-cad est régulée à la hausse dans la bandelette germinale au niveau jonctionnelle, plus particulièrement au niveau des jonctions DV (qui rétrécissent). L’ECD d’E-cad est nécessaire à l’intercalation et à la distribution polarisée d’E-cad. Elle est régulée par l’organisation de l’actine: la formine Diaphanous ainsi que le moteur moléculaire Myosine II accélèrent le recrutement de la machinerie d’endocytose (AP2 et Clathrine) et régulent la polarité de l’ECD dans l’embryon. Elles sont contrôlées par RhoGEF2, qui est enrichie dans les jonctions DV, et induisent l’endocytose par un mécanisme de clustering latéral d’E-cad. Dans la seconde partie de ma thèse, je me suis intéressé au couplage entre E-cad et la dynamique de MyoII. En effet, l’intercalation dépend aussi de flux contractiles de MyoII qui ont lieu préférentiellement en direction des jonctions DV. J’ai ainsi pu montrer que la direction des flux est induite par les anisotropies de forces d’ancrage de MyoII. Les faibles niveaux d’E-cad et le fort taux d’endocytose dans les jonctions DV augmentent la probabilité de générer une anisotropie d’ancrage et induisent davantage de flux de MyoII vers les jonctions DV. Ce projet met en lumière le rôle fondamental du couplage entre E-cad et MyoII dans la régulation de la morphogenèse. / Epithelia build up strong mechanical and chemichal barriers in Metazoans. Adherens junctions, through the adhesion provided by the transmembrane protein E-cadherin (E-cad), are essential for the mechanical integrity of the tissue. Yet, epithelia can be dramatically remodeled during embryogenesis or wound healing. During gastrulation of Drosophila embryo, the ventrolateral epithelium (the germ band) undergoes a massive elongation along the anteroposterior (AP) axis, driven by cell-cell intercalation. This is based on the polarized remodeling of intercellular junctions whereby junctions parallel to the dorsoventral axis (DV) shrink and form new junctions along AP axis. This remodeling is mediated by the planar polarized enrichment of Myosin II (MyoII) in DV junctions, which generates high tension. Adhesion proteins are also planar polarized, E-cad is enriched in AP junctions, but we did not know if this polarity contributed to cell-cell intercalation and the mechanism driving this polarity. As such, I have studied the role of Clathrin mediated endocytosis (CME) during germ band extension. I have shown that E-cad CME is specifically upregulated at the junction plane in the germ band, and planar polarized (enriched in DV shrinking junctions). It is required for cell-cell intercalation and the planar polarized distribution of E-cad. E-cad CME is regulated by the concerted action of the Formin Diaphanous and Myosin-II, which accelerates CME through the lateral clustering of E-cad. They are controlled by RhoGEF2, which is also enriched in DV junctions. In the second part of my PhD, I have studied the coupling between E-cad and MyoII dynamics. Indeed, planar polarized contractile flows of MyoII are required for DV junction shrinkage, but we did not know the mechanism driving the polarity of these flows. I have shown that the transient anisotropy of anchoring forces between two facing junctions triggers flow. As such, the low steady state amount of E-cad and the high rate of CME in DV junctions trigger more anisotropy and polarize the flow. These results outline the strong crossregulation between E-cad and MyoII and their concerted action in morphogenesis.

Un tissu épithélial est composé de cellules polarisées, étroitement liées les unes aux autres par des jonctions adhérentes. La perte de ces jonctions adhérentes est la première étape dans le développement des cancers au niveau des tissus épithéliaux. Il est donc important de comprendre les mécanismes d’attachement inter-cellulaire. Pour étudier ces interactions, nous utilisons comme modèle l’embryon de drosophile, où une fine régulation des jonctions adhérentes est requise pour l’une des étapes précoces de développement. Durant cette étape du développement, les cellules épithéliales changent de voisines le long de l’axe antéro-postérieur sans perdre leur adhérence cellulaire. Ce processus d’intercalation cellulaire est dû au recrutement polarisé du moteur moléculaire Myosine II au niveau des jonctions qui se désassemblent. Il a été mis en évidence qu’au cours de ce processus la perte de fonction de la voie JAK/STAT perturbe la localisation de la Myosine II. Au cours de ma thèse, j’ai réalisé un crible génétique dans un contexte mutant pour le ligand de la voie JAK/STAT pour me permettre d’identifier des interacteurs potentiellement impliqués dans le contrôle spatial de Myosine II. J’ai pu mettre en évidence plusieurs gènes pouvant être impliqués dans cette intercalation. Parmi ces candidats, je me suis focalisée sur celui montrant le plus fort phénotype : le gène CG13992. La caractérisation de ce gène a fut la seconde étape de mon travail de thèse (car seules les séquences nucléotidiques et protéiques étaient connues). Les résultats obtenus ont permis de mettre en évidence l’implication de ce gène dans la localisation de la Myosine II mais ils restent à confirmer. / Epithelial tissue is composed of polarized cells, which are closely attached to each other by adherens junctions. The loss of adherens junctions is often a key step in the development of cancer in epithelial tissues. It is therefore important to understand the mechanisms of attachment between the cells. To study such epithelial plasticity, we use the Drosophila embryo as a model system, where a fine regulation of adherens junctions is required for one of the early processes of development: germ band elongation. During this process, epithelial cells change their neighbors along the anterior-posterior axis (cell cell intercalation) without loss of cell adhesion. Polarized recruitment of the molecular motor Myosin II at the junctions, that disassemble and reassemble, underlies the intercalation process. In part, intercalation relies on the normal activity of the the JAK / STAT pathway that is crucial for the spatial control of Myosin II. During my PhD, I conducted a genetic screen, in a mutant for the ligand of the JAK / STAT pathway, designed to identify second site interactors for Myosin II control. I identified several genes that appear to be involved in the intercalation process. Among these candidates, I focused on one with the strongest phenotype: the gene CG13992. The functional characterization of this gene was the second stage of my thesis (because only the nucleotide and the protein sequences were known). Preliminary results highlight the involvement of this gene in the localization of Myosin II that remain to be confirmed.

Ce manuscrit de thèse est composé de trois parties dédiées aux développements technologiques nécessaires à l'étude de la polarité et des contraintes mécaniques dans les cellules épithéliales. La première partie décrit les développements technologiques et méthodologiques qui ont été réalisés en micro-fabrication et traitement de surface, acquisition et analyse d'image, et mesure des forces de traction. La deuxième partie décrit l'étude de l'organisation spatiale du système d'adhérence des cellules épithéliales. De la régulation de leur polarité à celle de leur fonction, l'architecture des cellules épithéliales est profondément liée à leur système d'adhérence. Nous avons utilisé les micropatrons adhésifs pour contrôler la géométrie de la matrice extra-cellulaire pour examiner l'effet de l'adhérence des cellules avec la matrice sur la position des zones d'adhérence intercellulaire. Nos résultats montrent que l'organisation spatiale de l'adhérence cellule-matrice joue un rôle déterminant sur celle de l'adhérence intercellulaire. Ils montrent également que cette organisation dirige ensuite la position du centrosome et l'orientation de l'ensemble de la polarité interne. Lors d'une réorganisation spatiale de l'épithélium, comme c'est le cas au cours de la transition épithélium-mésenchyme, les systèmes d'adhérence et la polarité interne subissent tous les deux de profondes modifications. Néanmoins, les cellules semblent capables de les réguler de façon indépendante selon le type de stimulus qui induit la réorganisation. La dernière partie est une analyse des paramètres physiques impliqués dans l'architecture épithéliale. En parallèle des régulations biochimiques, les contraintes mécaniques jouent également un rôle fondamental dans la régulation des processus morphogenétiques. L'association de l'ensemble de nos développements technologiques (patterning de substrat déformable, logiciel de détection et de mesure de force, contrôle du positionnement des cellules) nous a permis d'analyser précisément les propriétés mécaniques des architectures multicellulaires. Nous avons découvert que l'organisation spatiale du système adhérence était un régulateur majeur de l'intensité et de la répartition des forces intra-cellulaires. Cette observation nous a permis de proposer une modification du modèle actuel de distribution des contraintes dans un épithélium qui prend en compte l'anisotropie des forces inter-cellulaires en réponse à l'hétérogénéité de la matrice extra-cellulaire. Ce nouveau modèle physique permet de rendre compte des positions adoptées par les cellules en réponse aux différentes géométries de la matrice extra-cellulaire. / This thesis dissertation is comprised of three major parts. The first part devotes to all the technological developments that have been realized in my thesis study. These developments in microfabrication, in image acquisition and analysis, and in the traction force analysis had solved various problems we have encountered during our study of epithelial architecture. The second part describes the study of the spatial organization of the adhesion systems in epithelia. From their polarity, their functioning, to their remodeling, the epithelial architecture is deeply linked with the adhesion systems. With the capability to well define the location of cell-matrix interaction, we examined how the intercellular adhesion was organized according to the cell-matrix adhesion. Our results highlighted the instructive role of cell-matrix adhesion in organizing the intercellular adhesion. This organization subsequently governed the internal polarity which was indicated by the centrosome positioning. During epithelial remodeling, both the adhesion system and internal polarity were subjected to modification. Nevertheless they could be regulated differently depending on the context of remodeling. The last part is focused on the physical aspect of the epithelial architecture. Apart from the biochemical signaling network, mechanical force is also a substantial ingredient in morphogenesis. Together with our techniques in micropatterning the soft gel, the development of software for traction force microscope, and our knowledge of cell-cell positioning, we were able to analyze precisely the mechanical property of the multicellular architecture. We found that the cellular contractility was modulated by the spatial organization of the adhesion system. It permitted us to complete the current physical model of epithelial geometry with an anisotropic term for contractility. This new physical model could effectively account for the cell positioning on various matrix geometries.

La plupart du temps, les bactéries vivent au sein de biofilms : un tissu biologique accroché sur des surfaces (molles ou solides), qui est composé de bactéries et de matrice extracellulaire. Lors de cette thèse nous avons étudié les mécanismes qui contrôlent la formation d’un biofilm à l’interface eau-air par Bacillus subtilis (BS). D’abord, nous avons observé l’évolution phénotypique de BS au cours du développement de sa population en milieu liquide, dans des conditions de culture standard (pas de biofilm in fine). Nous avons constaté que la population exhibe différents types cellulaires (phénotypes) au cours du temps. Puis, afin d’observer les étapes de la formation d’un biofilm, nous avons créé une expérience qui permet de suivre l’évolution macroscopique de la concentration des bactéries et sa répartition spatiale au sein du milieu de culture. Nous constatons qu’une accumulation de bactéries se forme en dessous de l’interface eau-air avant même l’apparition d’un biofilms. Cette accumulation est concomitante avec de la convection dans le fluide (bioconvection). Le biofilm apparait lors de la phase de croissance de la population en bactérie pour une concentration moyenne dans le milieu de culture de l’ordre de 10¹³ bactéries/m³. Ensuite, nous avons formulé un modèle continu qui renseigne sur l’évolution de l’environnement des bactéries. Ce modèle reproduit la bioconvection observée dans les expériences et révèle son effet sur la concentration en dioxygène dissous dans la culture. Enfin, nous avons construit un modèle hybride continu-discret qui décrit la transition de bactéries déconnectées (nomades) vers des bactéries connectées formant un biofilm solide (sédentaires). Chaque bactérie est considérée comme une particule individuelle. Le modèle tient compte des forces de contact, ainsi que les forces élastiques qui peuvent s’établir entre les bactéries lorsqu’elles sont liées par de la matrice. Un nombre minimal d’aptitudes biologiques a été utilisé pour modéliser les bactéries : la division cellulaire qui leur permet de coloniser le milieu de culture, la motilité et l’aérotactisme qui explique leur migration vers la surface liquide, le quorum sensing (QS) et la différenciation cellulaire qui leur permet de passer du phénotype nomade (motile) au phénotype sédentaire (producteur de matrice). L’environnement en dioxygène des bactéries et les propriétés hydrodynamiques du milieu sont décrits par des champs continus. Le modèle reproduit toutes les étapes de la formation d'un biofilm observées dans nos expériences et confirme la nécessité de certaines aptitudes biologiques. Il montre que le seuil de QS joue un rôle majeur dans la morphologie du biofilm et sa cinétique de formation. En revanche, le taux de consommation de dioxygène par les bactéries ne semble pas avoir de rôle important. Enfin, nous avons établi que la bioconvection agit comme un retardateur de la formation du biofilm. / Most of the time, the bacteria live inside the biofilm: the biological tissue that is attached to the surface (soft or solid), is made of the bacteria and of extracellular matrix. According to this thesis we study the mechanics that control the formation of a biofilm at the interface water-air by Bacillus subtilis (BS). First, we absorbed the phenotype evolution of the BS during the development of its population in liquid medium, under the conditions of a standard culture (no of biofilm in fine). We noticed that the population exhibits different cellular types (phenotypes) during this time. Then, after absorbing the different stages of the development in the formation of a biofilm, we created an experience that allows us to follow the macroscopic evolution of the concentration of the bacteria and its special distribution in the middle of the culture. We found that an accumulation of the bacteria forms under the surface water-air, even before the appearance of a biofilm. This accumulation is concomitant with the conservation in the liquid (bioconvection). The biofilm appears during the growth phase of the bacterial population in an average concentration in the culture medium of about 10¹³ bacteria / m³. Then, we formed a continuous model that shows us the evolution of the environment of the bacteria. This model reproduced the bioconvection that was observed in the experiments and reveals its effect on the concentration of oxygen in the biological culture. Finally we built a hybrid continuous-discreet model that described the transition of the disconnected bacteria (nomads) through the connected bacteria that form a solid biofilm (sedentary). Each bacteria is considered as an individual particle. The model takes the contact forces under consideration, as well as the elastic forces that can settle between the bacteria when linked by the matrix. A minimum number of biological skills were used to form a model from the bacteria; the cellular division that allowed it to colonize the medium biological culture, the motility and the aerotaxi that explains its migration towards the liquid surface, the quorum sensing (QS) and the cellular differentiation that allows them to spend nomadic phenotype (motile) sedentary phenotype (producer of matrix). The dioxygen environment of the bacteria and its middle hydrodynamic properties are described by continuous fields. The model reproduces all the formation steps of an observed biofilm in our experiment and confirms the need of certain biological skills. It shows that the threshold of QS plays a major role in the morphology and biofilm formation kinetics. On the other hand, the rate of diocygen consumption by the bacteria does not seem to have any significant role. Finally, we established that the bioconvection reacts as a retardant in the biofilm formation.