Secretory processes underpin the emergence of cellular diversity in complex multicellular organisms. However, our understanding of the basic mechanisms controlling the different secretory and endosomal compartments involved remains surprisingly incomplete. During my DPhil I have studied a specialised epithelial cell type in the male Drosophila accessory glands, the secondary cell, which contains unusually large intracellular compartments that are accessible to detailed morphological study. I characterise the organisation, ultrastructure and molecular composition of this cell's secretory and endosomal compartments, and I employ specific Rab GTPases, conserved coordinators of membrane trafficking and identity, to define multiple compartmental subtypes. By developing super-resolution and time-lapse microscopy approaches in these cells, I show that numerous intraluminal vesicles (ILVs) are formed within Rab11-labelled secretory compartments and released into the accessory gland lumen as exosomes, the first clear demonstration in eukaryotic cells of exosome biogenesis within a non-late endosomal compartment. Biogenesis of these ILVs is dependent on evolutionarily conserved Endosomal Sorting Complexes Required for Transport (ESCRT) 0-III genes and involves loading of compartment-specific cargoes. Work by others, some in collaboration with me, has shown that these novel mechanisms are conserved in human cells. I show that dense-core granules, the structures employed to package proteins and other molecules destined for regulated secretion, form within large non-cored Rab6- positive compartments, in a process that seems to involve inputs from both the Golgi and recycling endosomal pathways. Further analysis has revealed roles for specific Rabs, for ILVs, and for the conserved fibrillar protein Mfas/TGFBI in different aspects of DCG formation. I also show that DCGs are not only secreted, but can also be degraded by fusion to acidic endosomal compartments. Remarkably, there is evidence that mammalian cells may employ all of these mechanisms and defects in these processes may be linked to diseases like cancer, diabetes and neurodegenerative disorders. Hence my work has established a new system to study complex secretory mechanisms, which can now be developed to model specific disease processes in the future. In summary, I have discovered several novel cell biological mechanisms controlling exosome biology, dense-core granule biogenesis, regulated secretion, and endolysosomal trafficking. Some of these already appear relevant to human health and disease, suggesting that the secondary cell system has considerable further potential for unravelling the fundamental processes underlying eukaryotic secretion in the future.

Candida albicans est un organisme commensal présent dans le microbiote, qui peut cependant provoquer des infections superficielles mais aussi systémiques, engageant alors le pronostic vital chez les patients immunodéprimés. La transition entre forme bourgeonnante et forme filamenteuse hyphale hautement polarisée, ce qui nécessite une réorganisation du cytosquelette et un trafic membranaire soutenu, est associée à la virulence. Chez les eucaryotes, les GTPases de la famille Rab (Ras related protein in the brain) et leurs régulateurs jouent un rôle central dans le trafic membranaire. L'objectif de ce travail est de comprendre le rôle de ces protéines, en particulier de Ypt6, l'homologue de Rab6 humain, dans la transition morphologique et la virulence de C. albicans. Dans ce but, j'ai construit des mutants « perte de fonction » et déterminé que YPT6 n'est pas essentiel à la viabilité, mais est critique pour l'intégrité de la paroi cellulaire et la croissance hyphale invasive ; les hyphes du mutant ypt6 sont plus courtes que celles de la souche sauvage. En outre, YPT6 est critique pour la virulence dans deux modèles murins de candidose. Lors de la croissance hyphale, Ypt6 est co-localisé avec Arl1, une GTPase de la famille Arf (ADP Ribosylation Factor), également nécessaire pour la croissance hyphale et la virulence de C. albicans. De plus, la surexpression de YPT6 compense spécifiquement le défaut de croissance hyphale du mutant de délétion arl1, mais pas l'inverse. La délétion de YPT6 résulte également en une augmentation du nombre de citernes Golgiennes, suggérant que l'intégrité du Golgi est altérée dans ce mutant. Utilisant de l'imagerie sur cellules vivantes, j'ai montré que la distribution d’Abp1 (Actin binding protein 1), qui est un rapporteur des sites d’endocytose, est aussi altérée dans le mutant ypt6, en ceci qu’elle n’est plus restreinte à l’apex de l’hyphe, comme observé dans les cellules sauvages. Ces données suggèrent que le défaut de maintien de la croissance hyphale du mutant ypt6 est au moins en partie associé à une altération de la distribution des sites d’endocytose. En résumé, j’ai identifié le rôle de Ypt6 dans la croissance hyphale invasive et la virulence du pathogène fongique opportuniste de l’homme C. albicans, et mis en évidence une interaction entre deux GTPases, Ypt6 et Arl1, lors du processus de croissance hyphale. / Candida albicans is a harmless constituent of the human microbiota that causes superficial infections as well as life threatening infections in immune compromised individuals. The transition from a budding form to the highly polarized hyphal form is associated with virulence and requires cytoskeleton reorganization and sustained membrane trafficking. In a range of eukaryotes, Ras related protein in the brain (Rab) G proteins and their regulators have been shown to play a central role in membrane traffic. The objective of this work is to understand the role of Rab proteins, in particular Ypt6, the homolog of Human Rab6, in the morphological transition and virulence of C. albicans. To this aim, I generated loss of function mutants and found that YPT6 is not essential for viability, yet was critical for cell wall integrity and invasive hyphal growth, with ypt6 hyphal filaments shorter compared to that of the wild type (WT). Furthermore, YPT6 was important for virulence in two murine candidiasis models. I determined that Ypt6 was localized at the late Golgi compartment during hyphal growth, where it co-localized with Arl1, a small GTPase of the Arf (ADP Ribosylation Factor) family, also required for hyphal growth and virulence. Interestingly, overexpression of YPT6 specifically rescued the hyphal growth defect of the arl1 mutant, but not the converse. Further characterization of the ypt6 deletion mutant showed that the number of Golgi cisternae is increased in this mutant compared to that of WT strain, suggesting an alteration of Golgi integrity. In addition, using live cell imaging I showed that the distribution of Actin binding protein 1 (Abp1), which is a reporter for actin patches, was altered in the ypt6 mutant, in that it was no longer restricted to the tip of the filament, as is observed in WT cells. These data suggest that the defect in hyphal growth maintenance of the ypt6 deletion mutant is at least partly associated with an alteration of the distribution of endocytic sites. Thus, I identified a critical role of Ypt6 during invasive hyphal growth and virulence in the human fungal opportunistic pathogen C. albicans and revealed an interaction between Ypt6 and Arl1 in the hyphal growth process.

No description available.