Indiana University-Purdue University Indianapolis (IUPUI) / Coxiella burnetii, the zoonotic agent of human Q fever and chronic endocarditis, is an obligate intracellular bacterial pathogen. The Coxiella intracellular niche, a large, lysosome-like parasitophorous vacuole (PV), is essential for bacterial survival and replication. There is growing evidence that host cell cholesterol trafficking plays a critical role in PV development and maintenance, prompting an examination of the role of cholesterol-binding host protein ORP1L (Oxysterol binding protein-Related Protein 1, Long) during infection. ORP1L is a multi-functional cholesterol-binding protein involved in late endosome/lysosome (LEL) trafficking, formation of membrane contact sites between LEL and the endoplasmic reticulum (ER), and cholesterol transfer from LEL to the ER. ORP1L localizes to the PV at novel membrane contact sites between the ER and the PV membrane. Ectopically expressed ORP1L in Coxiella-infected cells localizes to the PV membrane early during infection, before significant PV expansion and independent of other PV-localized proteins. Further, the N-terminal ORP1L Ankyrin repeats are both necessary and sufficient for PV localization, suggesting that protein-protein interactions, and not protein-lipid interactions, are primarily involved in PV association. Coxiella employs a Type IVB Secretion System (T4BSS) to translocate effector proteins into the host cytoplasm and manipulate various cellular functions. ORP1L is not found on the PV of a Coxiella mutant lacking a functional T4BSS, indicating a secreted bacterial protein is likely responsible for ORP1L recruitment. We identified a Coxiella mutant with a transposon insertion in CBU_0352 that exhibits a 50% decrease in ORP1L recruitment, suggesting that Coxiella CBU_0352 interacts directly or indirectly with ORP1L. Finally, we found that ORP1L depletion using siRNA alters PV dynamics, resulting in smaller yet more fusogenic Coxiella PVs. Together, these data suggest that ORP1L is specifically recruited to the PV, where it plays a novel role in Coxiella PV development and interactions between the PV and the host cell.

STARD3 est une protéine endosomale de la famille START (Steroidogenic Acute Regulatory (StAR) Related lipid Transfer), qui lie le cholestérol. STARD3 module l’organisation de la cellule en formant des sites de contact membranaire entre les endosomes et le réticulum endoplasmique (RE). Le lien entre les sites de contact membranaire et le transport du cholestérol n’était pas compris. Dans ce travail, nous montrons que STARD3 en interagissant avec les protéines VAPs (VAMP–Associated Proteins) bâtit une machine moléculaire autonome qui transporte le cholestérol au niveau des contacts RE–endosomes. Ce transport permet la formation de membranes internes dans les endosomes et est potentiellement impliqué dans le fonctionnement de ces organites. De plus, nous avons étudié la fonction de STARD3 dans la maladie Niemann Pick type C, qui est caractérisée par une anomalie du transport de cholestérol dans les endosomes. / STARD3 is an endosomal sterol-binding protein which belongs to the START protein family. Remarkably, STARD3 modulates the cellular organization by creating membrane contact sites between the endoplasmic reticulum (ER) and endosomes. The link between ER-endosome contact sites and cholesterol transport was not understood. In this work, we showed that STARD3 and its ER–resident partner, VAMP–associated protein (VAP), assemble into a machine that allows a highly efficient transport of cholesterol within ER–endosome contacts. This cholesterol transport provides building blocks for endosome inner membranes formation, and is probably involved in endosome dynamics. Furthermore, we studied STARD3 function in Niemann Pick type C disease, a condition characterized by an impairment of endosomal cholesterol export.

Les cellules eucaryotes sont caractérisées par le cloisonnement des organelles par des membranes. La communication entre les différents compartiments cellulaires est assurée par deux voies de transport : le transport vésiculaire et transport non-vésiculaire. Le transport vésiculaire permet à la fois le trafic des protéines et des lipides d'un compartiment à un autre, alors que le transport non-vésiculaire permet uniquement le trafic des lipides. En effet, les lipides jouent un rôle essentiel dans l'organisation cellulaire. Au cours de ma thèse, je me suis intéressé au rôle des lipides dans le trafic intracellulaire, en utilisant la levure comme organisme modèle. Dans une première partie de ma thèse, j'ai étudié les hélices amphipathiques qui permettent le ciblage des protéines vers des compartiments cellulaires spécifiques. Dans une étude précédente, réalisé au laboratoire a montré que ces hélices amphipathiques interagissaient directement avec les lipides membranaires, ce qui permet un adressage spécifique des protéines en fonction des environnements lipidiques dans la cellule. Deux hélices amphipatiques ont fait l’objet de cette étude : le motif ALPS qui cible les vésicules de la voie sécrétoire précoce, et alpha-synucléine qui reconnaît et fixe les vésicules du compartiment trans-Golgi-membrane plasmique. Dans cette première partie de la thèse j’ai cherché à identifier des motifs similaires à celui d’alpha-synucléine dans les protéines de levure, et de déterminer leurs rôles dans la cellule. Dans une seconde partie de ma thèse, en collaboration avec le laboratoire du Dr Thierry Galli, j'ai étudié de nouveaux composants impliqués dans le métabolisme lipidique aux sites de contact entre le réticulum endoplasmique et la membrane plasmique. Les sites de contact membranaires sont des régions de proches appositions (de l'ordre de 10 à 30 nm) entre deux membranes, généralement entre la membrane du réticulum endoplasmique (RE) et une autre organelle. Ce sont principalement des sites de transfert des lipides et d'ions. Maja Petkovic dans le laboratoire de Thierry Galli a fait la découverte que la protéine SNARE du RE, Sec22, interagit avec une syntaxine (Stx1) de la membrane plasmique dans les neurones, ce qui permet un nouveau mécanisme de contact entre ces deux membranes. J’ai donc essayé de voir si ce mécanisme est conservé chez la levure. Les résultats que j'ai obtenus ont confirmé que la levure Sec22 est capable d'interagir avec une protéine SNARE SSO1 localisée à la membrane plasmatique et homologue de Stx1. J'ai trouvé par co-immunoprecipitation que Sec22 et SSO1 deux interagissent avec les protéines de transfert des lipides localisées aux sites de contact. L'utilisation d'une sonde spécifique au Phosphatidylinositol-4 phosphate (PI4P), nous a permis de montrer que Sec22 est impliquée dans la régulation du niveau de PI4P à la membrane plasmique. Pour disséquer les deux fonctions de Sec22, dans la voie sécrétoire et aux sites de contact, nous avons utilisé l'approche des suppresseurs multicopies dans la levure. Parmi les suppresseurs identifiés, nous avons trouvé le Sfh1, une protéine qui a un rôle potentiel dans le transfert des lipides. Ces résultats confirment bien ceux obtenus par l’équipe de Thierry Galli, montrant que Sec22 a un nouveau rôle aux sites de contact entre le RE et la membrane plasmique et suggèrent que ce complexe SNARE pourrait être impliqué dans transfert de lipides chez la levure. / Eukaryotic cells are characterized by their internal membrane compartmentalization, with the various specialized organelles of the cell bounded by lipid membranes. Communication between different cellular compartments occurs via two transport pathways: vesicular transport and non-vesicular transport. Vesicular transport carries both proteins and lipids from one compartment to another in cells, whereas non-vesicular transport carries only lipids. An emerging idea is the important role that lipids play in cellular organization. Lipid binding amphipathic helices such as the ALPS (amphipathic lipid packing sensor) motif are targeted to membranes of a specific lipid composition, and hence act to transfer information encoded in membrane lipids to the vesicle trafficking machinery. The lipid composition of the membranes of different organelles is therefore of great importance. One mechanism that cells use to maintain the distinct lipid compositions of organelles is lipid transport, which occurs preferentially at membrane contact sites (MCS). MCS are regions of close appositions, on the order of 10 to 30 nm, between two membranes, generally between the membrane of the endoplasmic reticulum (ER) and another organelle. In my thesis, I addressed two aspects of how lipids and their transport function in intracellular trafficking, using yeast as a model system. First, I studied amphipathic motifs that mediate targeting of proteins to specific compartments in cells. Lipid binding amphipathic helices were shown in a previous study in the laboratory to mediate specific targeting to distinct lipid environments via direct protein-lipid interactions, both in vitro and in cells. One of these, the ALPS motif, targets vesicles of the early secretory pathway. The other, alpha-synuclein, targets vesicles travelling between the late Golgi, the plasma membrane and endosomes. I studied new potential alpha-synuclein-like motifs in yeast proteins, and their roles in cells. In a second project, in collaboration with the laboratory of Dr. Thierry Galli, I studied new compenents involved in lipid metabolism at contact sites between the endoplasmic reticulum and the plasma membrane. Maja Petkovic in the laboratory of Thierry Galli made the important discovery that the ER-localized SNARE protein Sec22 interacts with a plasma membrane syntaxin in neurons, thus providing a novel mechanism for mediating close contact between these two membranes. I addressed the question of whether this mechanism is conserved in yeast. The results I obtained confirmed that yeast Sec22 is able to interact with a SNARE protein localized to the plasma membrane, Sso1. I found by co-immunoprecitation that Sec22 and Sso1 both interact with lipid transfer proteins localized to ER-plasma membrane contact sites. Using a specific probe for phosphatidylinositol-4 phosphate (PI4P), we showed that Sec22 was involved in regulating the level of PI4P at the plasma membrane. These results extend to yeast those obtained by Maja Petkovic, Thierry Galli and colleauges showing that Sec22 has a novel role at ER-plasma membrane contact sites, and suggest that this SNARE complex might be implicated in lipid transfer at these sites in yeast.

Ca2+ is a universal and versatile intracellular messenger, regulating a vast array of biological processes due to variations in the frequency, amplitude, spatial and temporal dynamics of Ca2+ signals. Increases in cytosolic free Ca2+ concentration ([Ca2+]c) are due to influx from either an infinite extracellular Ca2+ pool or from the more limited intracellular Ca2+ stores. Stimulation of the endogenous muscarinic (M3) receptors of human embryonic kidney (HEK) cells with carbachol results in the activation of phospholipase C (PLC) and formation of inositol 1,4,5-trisphosphate (IP3), activation of IP3 receptors (IP3Rs), release of Ca2+ from the endoplasmic reticulum (ER), and activation of store-operated Ca2+ entry (SOCE). Lysosomes are the core digestive compartments of the cell, but their importance as signalling organelles is also now widely appreciated. Accumulating evidence indicates that lysosomal Ca2+ is important for their physiological functions. Lysosomal Ca2+ release triggers fusion during membrane trafficking and, through calmodulin, it regulates lysosome size. Luminal Ca2+ is critical for regulation of lysosomal biogenesis and autophagy during starvation through the transcription factor, TFEB. Furthermore, aberrant lysosomal Ca2+ is associated with some lysosomal storage diseases. Lysosomes in mammalian cells have long been suggested to accumulate Ca2+ via a low-affinity Ca2+-H+ exchanger (CAX). This is consistent with evidence that dissipating the lysosomal H+ gradient increased [Ca2+]c and decreased lysosomal free [Ca2+], and with the observation that lysosomal Ca2+ uptake was followed by an increase in pHly. Furthermore, heterologous expression of Xenopus CAX in mammalian cells attenuated carbachol-evoked Ca2+ signals. However, there is no known CAX in mammalian cells, and so the identity of the lysosomal Ca2+ uptake pathway in mammalian cells is unresolved. Using mammalian cells loaded with a fluorescent Ca2+ indicator, I show that dissipating the pHly gradient pharmacologically or by siRNA-mediated knockdown of an essential subunit of the H+ pump, increases the amplitude of IP3-evoked cytosolic Ca2+ signals without affecting those evoked by SOCE. A genetically encoded low-affinity Ca2+ sensor expressed on the lysosome surface reports larger increases in [Ca2+]c than the cytosolic sensor, but only when the Ca2+ signals are evoked by IP3R rather than SOCE. Using cells expressing single IP3R subtypes, I demonstrate that each of the three IP3R subtypes can deliver Ca2+ to lysosomes. I conclude that IP3Rs release Ca2+ within near-lysosome microdomains that fuel a low-affinity lysosomal Ca2+ uptake system. The temporal relationship between the increase in pHly and reduced Ca2+ sequestration suggests that pHly affects the organization of the microdomain rather than the Ca2+ uptake mechanism. I show that abrogation of the lysosome H+ gradient does not acutely prevent uptake of Ca2+ into lysosomes, but disrupts junctions with the ER where the exchange of Ca2+ occurs. The dipeptide, glycyl-L-phenylalanine 2-naphthylamide (L-GPN), is much used to disrupt lysosomes and release Ca2+ from them. The mechanism is widely assumed to require cleavage of GPN by cathepsin C, causing accumulation of amino acid residues, and osmotic lysis of lysosomal membranes. I show, using LysoTracker Red and Oregon Green-dextran to report pHly, that L-GPN is effective in HEK cells lacking functional cathepsin C, following CRISPR-Cas9-mediated gene disruption. Furthermore, D-GPN, which is resistant to cleavage by cathepsin C, is as effective as L-GPN at increasing pHly, and it is similarly effective in cells with and without cathepsin C. L-GPN and D-GPN increase cytosolic pH, and the effect is similar when the lysosomal V-ATPase is inhibited with bafilomycin A1. This is not consistent with GPN releasing the acidic contents of lysosomes. I conclude that the effects of GPN on lysosomes are not mediated by cathepsin C. Both L-GPN and D-GPN evoke Ca2+ release, the response is unaffected by inhibition or knock-out of cathepsin C, but it requires Ca2+ within the ER. GPN-evoked increases in [Ca2+]c require Ca2+ within the ER, but they are not mediated by ER Ca2+ channels amplifying Ca2+ release from lysosomes. GPN increases [Ca2+]c by increasing pHcyt, which then directly stimulates Ca2+ release from the ER. I conclude that physiologically relevant increases in pHcyt stimulate Ca2+ release from the ER independent of IP3 and ryanodine receptors, and that GPN does not selectively target lysosomes. I conclude that all three IP3R subtypes selectively deliver Ca2+ to lysosomes, and that the low pH within lysosomes is required to maintain the junctions between ER and lysosomes, but not for lysosomal Ca2+ uptake. I suggest that GPN lacks the specificity required to allow selective release of Ca2+ from lysosomes.

La communication intercellulaire est essentielle pour le développement et la survie d'organismes multicellulaires. Dans le règne végétale, une des voies privilégiée pour la communication intercellulaire est la voie symplastique qui implique des canaux aux dimensions nanométriques connectant les cellules entre elles, leur permettant d'échanger directement photo-assimilats, miARN, protéines, oligoéléments etc. Observés pour la première fois en 1880 par le botaniste autrichien Eduard Tangl (Tangl 1880; Kohler & Carr 2006), ils ont longtemps été considérés comme de simples trous perméables permettant la diffusion de matériel cellulaire (Lee & Lu 2011; Oparka & Roberts 2001). Etant donné leurs taille nanoscopique, ce n'est que dans les années 1960, avec la démocratisation de la Microscopie Électronique en Transmission (MET) qui permet d'atteindre , que les premiers modèles ultrastructuraux sont établis (Lopez-Saez 1965; Robards 1970). Ils font état d'un canal d'environ 30 à 40 nm de diamètre avec un élément central cylindrique traversant le pore, appelé le desmotubule, connecté au Réticulum Endoplasmique des deux cellules (Figure 1 of our review Tilsner et al. 2016). Dans les années 1980 notre compréhension des plasmodesmes a quelque peu évolué et nous savons maintenant que ces structures ne sont pas de simples trous mais des structures membranaires très spécialisées et régulées (Lucas & Lee 2004; Faulkner & Maule 2011; Furuta et al. 2012). Le modèle ultrastructural actuel découle de la congrégation d'études ultrastructurale, physiologiques et pharmacologiques plus ou moins anciennes dépeignant une structure morphologiquement très souple et changeant de conformation au cours du développement. Les plasmodesmes peuvent réguler leur ouverture/fermeture par la constriction de leurs extrémités grâce à l'accumulation entre la membrane plasmique et la paroi végétale d'un polymère de sucre, la callose qui va pousser la membrane plasmique contre le desmotubule et en obstruer les entrées. Cette modulation permettrait majoritairement de réguler les flux intercellulaires qui impliquent les plasmodesmes. Cependant nos connaissances sur les remaniements membranaires prenant place durant le développement des plasmodesmes et sur la régulation de leur perméabilité sont encore imparfaites.La microscopie électronique en transmission, malgré l'ancienneté de la technique, est l'une des plus résolutive, largement utilisée en biologie. Avec l'amélioration des techniques de préservation d'échantillons, notamment les cryo- méthodes, elle permet d'atteindre à l'heure actuelle des résolutions inférieures à 5 nm en condition contrastée et inclus en résine et peut descendre en dessous du nanomètre pour la cryo-microscopie. Ce potentiel permet aisément l'étude des sous-compartiments cellulaires de l'ordre du µm tel que mitochondries, chloroplastes, noyaux etc. (Frey et al. 2002) mais permet également l'étude ultrastructurale précise de structures de l'ordre de la dizaine de nm (Beck et al. 2007; Al-Amoudi et al. 2007).En revanche, dans son utilisation classique, la microscopie électronique ne permet pas d'accéder à la troisième dimension de l'espace, rendant difficile l'interprétation de structure à l'architecture quelque peu compliquée. En effet, les images produites ne sont que des projections en deux dimensions d'objets en trois dimensions. Cela a mené au développement de la tomographie électronique en transmission (Crowther et al. 1970), méthode basée sur un concept mathématiques formulé par Johann Radon au XIXe siècle. Ce n'est que dans les années 2000 que la tomographie électronique a pris un essor significatif grâce au couplage avec des méthodes d'automatisation informatiques. / Plasmodesmata were first observed by Austrian botanist Eduard Tangl in 1880. He devoted himself to studying the anatomy and cytology of plants and his greatest discovery, of course, was the observation and first characterization of plasmodesmata (Tangl 1880, 1884 and 1885). Despite not having access to their ultrastructure, he observed thin striations (see front page engraving) between cotyledon cells of Strychnos nuxvomica and in the endosperm of seeds and described them as being conductive ducts. Already at the time, he was evoking the idea that these strands "unite them [the cells] to an entity of higher order", in other words formulating the first definition of a symplastic domain. lt is only in 1901 that Strasburger finally names these canals "plasmodesmata". His discovery led to a radical change in our conception of the plant entity and brought in new concepts such as the symplasm (Munch 1930) and transmembrane fluxes between cells, which are now being tackled with great interest by numerous research teams around the globe.Because of their size, plasmodesmata ultrastructure was not accessible until the advent of electron microscopy and they were long thought to be simple holes connecting plant cells one-another with no specific regulation. lt is only with the advent of electron microscopy and chemical fixation that botanists started to gain interest in this structure again. And even with these methods allowing the observation of structures down to several nanometers in size, there are still debates on the nature of the canal, its constituents and physiology (Lopez-Saez J. 1965, Robards A. 1970, Ding et al. 1992, Tilney et al. 1991, Overall and Gunning 1982, Schulz et al. 1995).Nowadays, with the advent of modern cryopreservation and three-dimensional electron tomography methods, great improvements are to be done in the understanding of the ultrastructure and physiology of these mysterious canals. More particularly by understanding the link between the membranous rearrangements taking place in these pores and the molecular transit regulation.My work has led us to view plasmodesmata as specialised Membrane Contact Sites (MCS). Hence, by analogy with MCS found in mammals, yeast and plants, this work embraces an original angle on the speculation of the composition and role of the desmotubule-plasma-membrane tethering complex. The work produced during my thesis allowed me to contribute to the publication of one review and two articles, which will constitute the introduction and two main sub-sections of the results chapter, respectively. The introductory review has been published in 2016 in Annual Review of Plant Biology. The first one is still under reviewing at Nature Plant and the other has been published in The Plant Cell journal in April 2015.