Dans de nombreuses pathologies digestives dont les maladies inflammatoires chroniques intestinales (MICI), l'intégrité de la barrière épithéliale est rompue. Cette perte d'intégrité est notamment due à la réduction ou la perte d'expression des jonctions adhérentes composées du complexe E-cadhérine/caténines. Il est donc important d'identifier de nouvelles molécules capables de réguler ce complexe dans les MICI. C'est dans ce contexte que nous nous sommes intéréssés à une levure non pathogène, Saccharomyces boulardii (Sb) utilisée dans la prévention et le traitement de désordres gastro-intestinaux et qui présente des bénéfices thérapeutiques chez les patients atteints de MICI, notamment en régulant l'intégrité de la barrière intestinale. L'étude que nous avons réalisée sur des explants tissulaires provenant de patients atteints de MICI nous a permis de mettre en évidence que le surnageant de Sb (Sbs) protège la morphologie tissulaire et maintient l'expression de la E-cadhérine à la membrane. In vitro nous avons également pu montrer que Sbs accélère la ré-expression de la E-cadhérine à la membrane en régulant son recyclage par les endosomes (Rab11A), entrainant ainsi la restauration et le renforcement de la barrière épithéliale intestinale. / Some intestinal pathologies including inflammatory bowel disease (IBD) are associated with an altered barrier function. The reduction or the lost of adherens junctions composed by E-cadherin/catenins complex are linked to changes in the barrier integrity. Characterization of molecules targeting the E-cadherin/catenins complex during IBD is crucial for the development of alternative therapies. From this perspective, we focus ours studies on a non pathogenic yeast, Saccharomyces boulardii, used to prevent and treat gastro-intestinal disorders and may have beneficial effects in IBD treatment, including the regulation of barrier integrity. Ours studies on colonic explants from IBD patients showed that Sb supernatant (Sbs) protects epithelial morphology and maintains E-cadherin expression at the cell surface. In vitro study pointed out that Sbs accelerated the recovery of E-cadherin at the cell membrane. This process involved the modulation of the recycling of E-cadherin by endosomes (Rab11A), leading to restoration and strengthening of intestinal barrier function.

Les cellules de mélanome humain expriment différents antigènes tumoraux qui sont reconnus par les lymphocytes T cytotoxiques CD8 + (CTL) induisant des réponses spécifiques de la tumeur in vivo. Cependant, chez les patients atteints de mélanome l'efficacité de la réponse naturelle des CTL ou stimulée par thérapie est limitée. Les mécanismes sous-jacents de l'échec de la phase effectrice des CTL contre les mélanomes sont encore largement méconnus. Notre hypothèse est que l'efficacité limitée des CTL dans leur combat contre les tumeurs est le résultat d'une balance défavorable entre la capacité des CTL à tuer les tumeurs et une résistance tumorale intrinsèque à l'activité cytolytique des CTL. Au cours de ma thèse je me suis concentrée sur la dynamique moléculaire qui se produit à la synapse lytique afin de pouvoir identifier un mécanisme précoce mis en place par les cellules de mélanome face à l'attaque des CTL. En combinant l'utilisation d'approches de microscopie de pointe et des outils moléculaires, j'ai pu montrer que, lors de l'interaction avec les CTL, les cellules de mélanome humain subissent une activation de leur trafic vésiculaire endosomal et lysosomal, lequel est intensifié à la synapse lytique et corrèle avec la dégradation par la cathepsine de la perforine et un défaut de pénétration d'entrée du granzyme B. De plus, j'ai démontré que le blocage du trafic lysosomal dépendant de SNAP23, la modification du pH (intra-vésiculaire) et l'inhibition de l'activité lysosomale protéotlytique des cellules de mélanome permet de restaurer leur sensibilité à l'attaque des CTL. Nos résultats révèlent une stratégie sans précédent d' " auto-défense " des cellules de mélanome à la synapse immunologique basée sur une sécrétion lysosomale massive et sur la dégradation de la perforine sécrétée par les CTL. Ainsi pouvoir interférer avec cette stratégie synaptique d'auto-défense des cellules de mélanome pourrait contribuer à potentialiser les réponses des CTL et les immunothérapies chez les patients atteints de mélanome. / Human melanoma cells express various tumor antigens that are recognized by CD8+ cytotoxic T lymphocytes (CTL) and elicit tumor-specific responses in vivo. However, natural and therapeutically enhanced CTL responses in melanoma patients are of limited efficacy. The mechanisms underlying the failure of CTL effector phase against melanomas are still largely elusive. Our hypothesis is that the limited efficacy of CTL in their fight against tumors is the result of an unfavorable balance between CTL ability to kill tumors and an intrinsic tumor resistance to CTL cytolytic activity. During my thesis I focused on the molecular dynamics occurring at the lytic synapse in order to identify possible "early response-mechanism" of melanoma cells to CTL attack. Using a combination of cutting edge microscopy approaches and molecular tools, I showed that upon conjugation with CTL, human melanoma cells undergo an exacerbated late endosome/lysosome trafficking, which is intensified at the lytic synapse and is paralleled by cathepsin-mediated perforin degradation and deficient granzyme B penetration. Abortion of SNAP-23-dependent lysosomal trafficking, pH perturbation or impairment of lysosomal proteolytic activity restores susceptibility to CTL attack. Our results reveal an unprecedented strategy of melanoma cell "self-defense" at the immunologic synapse based on a lysosome secretory burst and perforin degradation at the lytic synapse. Interfering with this synaptic self-defense strategy might be instrumental to potentiate CTL-mediated therapies in melanoma patients.

Wnt signalling belongs to conserved pathways and mediates cell fate decision, development, regeneration and adult tissue homeostasis. Disruption or misregulation of Wnt signalling pathway often leads to disease. Wnt proteins are hydrophobic glycoproteins which need a special receptor for transport from Golgi Apparatus to cell surface, which is called MIG-14 in Caenorhabditis elegans and Wntless (Wls) in mammals. In this study, I focus on understanding mechanisms that regulate MTM-6 protein activity. MTM-6, a lipid phosphatase associated with endosomal membrane, has been recently identified as a regulator of MIG-14/Wls trafficking in Caenorhabditis elegans. Silencing of mtm-6 leads to misregulation of some Wnt-directed processes, such as migration of Q neuroblasts progeny. This study reports identification of novel mtm-6 genetic interactors that have been found to influence migration of Q neuroblasts progeny through Wnt signalling. New knowledge about mtm-6 genetic interactions bring us near to understanding of Wnt signalling regulation. Keywords: Caenorhabditis elegans, MTM-6, SEL-5, Wntless, Wnt, endosomes, phosphoinositides, retromer

Actin polymerization facilitated by the Arp2/3 complex plays a critical role in a wide range of cellular processes such as motility, endocytosis and cargo recycling. Activation and appropriate localization of the Arp2/3 complex is mediated by an interaction with the nucleation-promoting factor (NPF). WASH complex is the major endosomal NPF which plays a crucial role in the cargo recycling back to the trans-Golgi network (TGN) or plasma membrane. It is composed of five subunits: WASH1, SWIP, FAM21, CCDC53 and strumpellin. While WASH1 and FAM21 have been extensively studied, much less is known about strumpellin, a protein causally implicated in the onset of hereditary spastic paraplegia (HSP). This work focuses on strumpellin function in the cells, showing that only full-length protein incorporates into the WASH complex. In a strumpellin knock out cell line, we demonstrated that loss of strumpellin resulted in destabilization of the other WASH complex subunits. Still, an incomplete WASH complex without strumpellin was assembled. Cells also displayed enlarged endosomal subdomains and WASH complex nucleation activity on endosomes was largely diminished as assessed by loss of the actin patches. Finally, the absence of strumpellin was also accompanied by the accumulation of glucose transporter 1 (GLUT1)...

Amyloid plaques, a neuropathological hallmark of Alzheimer’s disease (AD), are largely composed of amyloid beta (Aβ) peptide, derived from cleavage of amyloid precursor protein (APP) by β- and γ-secretase. The endosome is increasingly recognized as an important crossroads for APP and the secretases, with major implications for APP processing and amyloidogenesis. Amongst various posttranslational modifications affecting APP, ubiquitination of cytodomain lysines may represent a key signal controlling endosomal sorting. Here, we show that substitution of APP COOH-terminal lysines with arginines disrupts APP ubiquitination, though the pool of ubiquitinated APP is small or transient. Nonetheless, this small deficiency in ubiquitination can have a significant impact on APP, such that the number of lysines mutated trends toward an increase in APP metabolism. An APP mutant lacking all COOH-terminal lysines undergoes the most pronounced increase in processing, leading to accumulation of both secreted and intracellular Aβ40, without change in Aβ42. This phenotype is abolished by artificial ubiquitination of APP using rapalog-mediated proximity inducers. Lack of APP COOH-terminal lysines does not affect APP endocytosis, but leads to a redistribution of APP from endosomal intraluminal vesicles (ILVs) to the endosomal limiting membrane, with subsequent decrease in APP COOH-terminal fragment (CTF) content of secreted exosomes, but minimal effects on APP lysosomal degradation. Both the secreted and intracellular increase in Aβ40 is abolished by depletion of presenilin 2 (PSEN2), recently shown to be enriched on the endosomal limiting membrane compared to presenilin 1 (PSEN1). In a separate set of studies, we found that a familial AD mutant, L723P, which occurs immediately next to a string of three lysines in the juxtamembrane region, behaves more similarly to other FAD-causing mutations. APP L723P exhibits a selective increase in Aβ42, and a delay in degradation, but no change in exosomal content, despite some missorting to the endosomal limiting membrane. Our findings demonstrate that ubiquitin can act as a signal for endosomal sorting at five lysines in the APP cytodomain, disruption of which prevents sequestration of APP in ILVs and results in the processing of a larger pool of APP-CTF by PSEN2 on the endosomal membrane.

At the neuronal synapse, neurotransmitter-filled synaptic vesicles (SVs) fuse with the presynaptic plasma membrane during activity. Following exocytosis, SVs must be retrieved for neurotransmission to be maintained. Several modes of SV recycling have been identified. During mild neuronal activity, clathrin-mediated endocytosis has been regarded as the dominant SV retrieval mode, however the recently identified ultrafast endocytosis mode may also be important in this condition. During elevated activity, activity-dependent bulk endocytosis (ADBE) is the dominant SV retrieval pathway. In ADBE, large invaginations are formed from the plasma membrane, which then undergo scission to create bulk endosomes. In a second distinct step, SVs bud from these endosomes and specifically repopulate the reserve SV pool. However, since its first identification, only few molecules have been shown to participate in ADBE. The aim of this PhD was to identify novel molecules and elucidate the molecular mechanism of ADBE. To achieve this, two independent biochemical approaches were designed to purify and enrich bulk endosomes from primary neuronal cultures. In the first approach, bulk endosomes and SVs were labelled with a dye, FM1-43, using a strong stimulus. Cells were broken mechanically and the post nuclear supernatant, that contains all intracellular organelles, was collected. The supernatant was then subjected to subcellular fractionation using discontinuous Nycodenz gradients. This stimulated sample was always processed in parallel with a basal sample, where no neuronal stimulus was applied, in order to visualise activity dependent FM loading. After different fractionation protocols were applied, bulk endosomes were efficiently separated from SVs, as revealed by tracking fluorescence in different fractions. The fractionation results were further validated by electron microscopy, where bulk endosomes and SVs were labelled with horseradish peroxidase and purified using the established protocol. Immunoblotting against selected SV cargo proteins from stimulated bulk endosome and SV samples, indicated the specific and preferential localisation of VAMP4 on bulk endosomes, in contrast to other SV cargo. The molecular identity of bulk endosomes was also approached by submitting the bulk endosome fractions to semi-quantitative mass spectrometry. This analysis revealed many different proteins that were identified in bulk endosome samples and quantification approaches further indicated proteins that can be localised on bulk endosomes and have a potential role in ADBE. A second magnetic isolation approach was designed, to purify bulk endosomes using a completely different methodology. In this case, bulk endosomes were specifically labelled with iron nanoparticles, which are preferentially taken up by bulk endosomes since they are larger than SVs. The cells were broken as before and post nuclear supernatant was acquired. In this case, the supernatant was submitted to magnetic isolation that separated iron beads labelled structures from all other intracellular organelles. An extensive immunoblotting analysis of magnetic bulk endosomes validated that VAMP4 and syndapin I, two essential ADBE proteins, were enriched in these purified samples. These magnetic bulk endosomes were also analysed using semi-quantitative MS and revealed many proteins with a potential role in ADBE. Significant overlap between the two independent methods was observed, further validating these approaches. Combining these two methods with bioinformatics tools allowed the identification of the molecular signature of ADBE as well as novel key candidates for this process. Specific molecules were investigated for their role in ADBE and SV recycling using a variety of different real-time fluorescent imaging assays. A major focus was on rab small GTPases. High molecular weight dextran uptake was used to specifically study the role of these proteins in ADBE, as it preferentially reports uptake via larger bulk endosomes. A pH sensitive chimeric protein, synaptophysin-pHluorin, was used to investigate the role of these proteins in CME. Additional imaging assays were used to answer emerging questions regarding the function and localisation of these targets in the presynapse. Using these approaches, rab11A and rab35 were found to promote ADBE and accelerate clathrin-mediated endocytosis. This effect was specific to high intensity stimulation, while SV exocytosis was not affected. Further research on the role of both novel and established ADBE molecules will provide key future insights into the mechanism of both bulk endosome generation/scission and subsequent SV reformation. A very promising group is rab proteins and now evidence for their implication in SV recycling is presented here. Identification and characterisation of new targets will allow to investigate the role of ADBE in neurotransmission in both physiology and pathophysiology.

Shiga toxin and cholera toxin are members of the AB5 family of protein exotoxins. The A subunit is the enzymatic subunit, whereas the pentameric B subunit binds cell surface receptors and carries the A subunit to the endoplasmic reticulum (ER) where it can be released into the cytosol. The B-subunits (STxB or CTxB) mediate toxin traffic along the retrograde pathway from the plasma membrane to the ER via early/recycling endosomes and the Golgi apparatus. It is unknown if STxB requires transport through the Golgi, or if it is just kinetically favorable. It is also unknown if the recycling endosome (RE) plays a role in the retrograde transport of STxB and CTxB. The first goal of this dissertation research was to demonstrate that transport through the Golgi is required for STxB to reach the ER. Using aluminum fluoride treatment, a simple temperature block, and cytoplast studies, I show that Golgi transport is necessary for STxB to reach the ER. The second goal of this dissertation research was to tease apart how STxB and CTxB move through early and recycling endosomes as well as elucidate a mechanism of how STxB exits endosomes en route to the Golgi. The role of the RE in STxB and CTxB transport is unclear. I used transferrin colocalization and temperature block studies to show that STxB and CTxB traffic through the RE. I then used HRP ablation of the RE to show that STxB requires the RE to reach the Golgi. I also examined the role of an RE-specific protein, EHD1, in exit of STxB from the RE. EHD1 has been previously shown to regulate recycling Tfn exit from the RE but its role in STxB transport is unknown. Expression of a dominant negative form of EHD1 arrested STxB at the RE and prevented it from reaching the Golgi. Together, these results suggest that STxB and CTxB transit the RE, STxB requires a functional RE for normal retrograde trafficking, and that STxB exit from the RE is regulated by EHD1.

Endosomes are dynamic vesicular structures which transport cargo molecules internalized into the cell via endocytosis. Endosomal trafficking of cargo involves a large number of individual endosomes that regularly interact with each other via fusion and fission and thus form a dynamic network wherein endocytosed cargo is sorted and transported to various other intracellular compartments. In this study we present a general theoretical framework that takes into account individual endosomes and several key microscopic interaction processes among them. By combining theory with quantitative experiments, we seek to address the fundamental question of how the behaviour of the endosomal network emerges from the interactions among many individual endosomes of different sizes and cargo contents. Our theory is based on distributions of endosomes of various sizes and cargo amount. We compare our theory to experimental time course distributions of LDL, a degradative cargo, in a population of early endosomes. Early endosomes display a broad distribution of cargo with a characteristic power law, which we show is a consequence of stochastic fusion events of cargo carrying early endosomes. A simple model can quantitatively describe time-dependent statistics of LDL distributions in individual early endosomes. From fits of the theory to experimental data we can determine key parameters of endosomal trafficking such as the endosome fusion rate and the fluxes of cargo into and out of the network. Our theory predicts several experimentally confirmed scaling behaviours, which arise as a result of endosome fusion. Our theory provides a link between the dynamics at individual endosome level and average properties of the endosomal network. We show from our theory that some features of the endosomal distributions, which arise from interactions among individual endosomes, are sensitive to alterations in chosen parameters. This provides a direct means to study perturbation experiments wherein the cargo distribution can vary in response to changes of the endocytic system. Our analysis provides a powerful tool for the study of genetic and chemical perturbations that may alter specific systems properties and for extracting various kinetic rates involved in endosomal trafficking from only still images at different points.

The dimorphic fungus Ustilago maydis is a pathogen of maize and it was used for decades to understand the molecular basis of plant pathogenicity aspects. Recently, much effort went into understanding the cell biology that underlies the virulence of U. maydis. It was shown previously that early endosomes (EEs) move bidirectionally within fungal hyphal cells. Although it was shown that the motility of EEs facilitates growth of the infectious hypha and mutants defective for kinesin-3 (Kin3), the major EE transporter, exhibit impaired polarized growth, the importance of EEs and their motility in plant colonization is not known. The first part of this thesis is focused on the role of EE motility during plant infection. In collaboration with Natalie Steinberg, who performed the plant infection assays, I used a synthetic molecular anchor, K1rPX, to block the motility of EEs at early and late stages during the host plant infection and I found that EE motility is essential during the first two days of pathogenic development, when infectious hyphae exhibit most prominent elongation, whereas blockage of EE motility after 3 days post infection does not inhibit plant colonization. Moreover, I documented that the blockage of EE motility during early stages of the infection causes high plant defence response, which means that the pathogen becomes recognized by the host plant defence system. These results indicate that EE motility is crucial during initial stages of the plant host infection and enables colonization by U. maydis and additionally suggests involvement of EEs in some defence response machinery. The second part of the thesis addresses the relationship between Kin3, the major motor for EE motility, and the microtubule (MT) array. I demonstrate here that Kin3 uses all MT tracks available in the cell, which is in contrast to published results in other systems. In the third part I focused on the interaction between Kin3 and the EEs. I found that the pleckstrin homology (PH) domain localized at the distal part of the Kin3 tail is of minor importance for EE association. This conclusion is supported by in vivo experiments, showing that truncated Kin3PH, which lacks the PH domain, was still able to bind to the organelles. By systematic truncation of parts of the Kin3 tail I found two adjacent regions, a DUF3694 domain and a "linker" region, that are important for binding of Kin3 to EEs. By using a synthetic anchor composed of Kin1 rigor domain and selected Kin3 domains I proved that both domains anchor the EEs to MTs and inhibit EE motility. I also showed that the PH domain is not able to block EE motility. In collaboration with Dr. Nicholas Harmer, who performed structural modelling of selected PH domains, I demonstrated that the PH domain is likely to interact with the motor domain of Kin3. This result was confirmed by using a yeast-two hybrid approach and a protein affinity assay. This indicates a globular organization of the Kin3 motor, which was confirmed by a split-YFP assay in living cells. Deletion of the PH domain and most probably lack of intramolecular interaction between the tail and motor domain reduces Kin3 motility parameters like velocity, frequency and run length indicating that the interaction of the PH domain with the motor domain has a role in the control of Kin3 motility.

Utilization of iron by hemoglobin-producing cells is highly efficient. The acquisition of iron from plasma requires the binding of diferric transferrin (Tf) to its cognate receptor (Tf-R) on the erythroid cell membrane, followed by internalization of the Tf - Tf-R complexes via receptor-mediated endocytosis. Through a poorly understood mechanism, iron is targeted to mitochondria, the site of heme biosynthesis. We believe that a direct interaction between iron-containing endosomes and mitochondria is essential for iron transfer to mitochondria and its efficient incorporation into heme. / In order to illustrate the interaction between endosomes and mitochondria, we have employed flow cytometry. Flow cytometry analysis of reticulocytes (erythrocyte precursors which still synthesize hemoglobin) stained with fluorescent dyes specific to mitochondria and endosomes revealed three distinct populations: mitochondria, endosomes and a population labeled with both dyes. This double-labeled population suggests a population composed of endosomes associated with mitochondria. Using non-fluorescent diferric-Tf, we were able to remove the double population, leaving only the endosomal and the mitochondrial population. This finding has confirmed that the double population is the result of the interaction between the two organelles. / Additionally, we established a cell-free assay consisting of fluorescent mitochondria and endosomes isolated from erythroid cells. Using confocal microscopy, we demonstrated a colocalization between the two organelles. We repeated the assay using fluorescent mitochondria and endosomes isolated from HeLa spinner cells. Using the mitochondrial uncoupler CCCP, we were able to significantly reduce the colocalization between the two organelles, indicating that the interaction between the organelles is specific and that the mitochondrial potential is a requirement for organellar interaction. / Based on our results from flow cytometry and confocal microscopy, we conclude that a specific and direct interaction exists between the two organelles.