京都大学 / 0048 / 新制・課程博士 / 博士(生命科学) / 甲第18431号 / 生博第311号 / 新制||生||41(附属図書館) / 31289 / 京都大学大学院生命科学研究科高次生命科学専攻 / (主査)教授 豊島 文子, 教授 藤田 尚志, 教授 松田 道行 / 学位規則第4条第1項該当 / Doctor of Philosophy in Life Sciences / Kyoto University / DFAM

Cellular communication requires the transport of chemical messengers between intracellular compartments and from cell to cell. The regulated exocytosis of a secretory vesicle at the plasma membrane involves the merger of two bilayers, with markedly different lipid composition, within a millisecond time scale. The spatial and temporal control of the protein and lipid complement at these fusion sites is essential. A highly conserved family of proteins are known to drive this fusion event; SNAP-25 and syntaxin-1 (t-SNAREs) associate at the plasma membrane in a 1:1 stoichiometry to provide a binding site for the vesicle-membrane protein synaptobrevin (v-SNARE). The formation of this complex and subsequent fusion requires accessory proteins for efficient calcium-triggered exocytosis; which of these proteins facilitate the initial attachment of vesicle to the plasma membrane prior to fusion is still under debate. Specific sites for vesicle fusion have been proposed and the organisation of lipids and proteins at these fusion sites has been extensively investigated with limited spatial and temporal resolution; however the presence of raft-forming lipids at these sites as well as the arrangement of SNARE proteins at the molecular level is still under contention. The data presented within this thesis aims to elucidate the protein and lipid environment at the fusion site using super-resolution microscopy and advanced vesicle tracking. Under diffraction-limited microscopy the t-SNAREs are visualised as 200 nm homogenous clusters; however I have used single molecule localisation microscopy to reveal a more complex heterogeneous molecular arrangement. Quantification of lipid order exclusively at the plasma membrane provided insight into the influence of cholesterol-induced lipid arrangement on SNAP-25 localisation. In addition the t-SNARE interaction was investigated using TCSPC-FLIM identifying two lipid-order-dependent conformations in distinct clusters at the plasma membrane. Extensive vesicle tracking at optimum sampling rates demonstrated the ‘sampling’ behaviour of LDCVs and allowed characterisation of vesicle fusion sites. In summary I find that vesicles exhibit preference for residence and probably fusion at regions of plasma membrane with a low t-SNARE density; these proteins appear to exert control over exocytosis by adopting alternative conformations that are under cholesterol-induced regulation.

The nuclear envelope is a highly organised double membrane system that separates the activities of the nuclear and cytoplasmic compartments in eukaryotic systems. The wide range of functions recently associated with the NE and the identification of hundreds of proteins associated with this cellular structure indicates that it is a major signalling node for the cell. Recent work indicates NE functions in signalling innate immune responses to herpesviruses. The viruses, on the other hand, often target or usurp NE functions in different ways. The NE is also a physical barrier that must be overcome for viruses like the herpesviridae that assemble capsids in the nucleus. This thesis addresses two important questions: 1) How do herpesviruses cross the NE after new viral particles are produced in the nucleus? and 2) What is the nuclear envelope role of NET23/STING in the activation of immune factors upon herpesvirus infection? To address the first question, I followed two different approaches. The first used the isolation of microsomes from HSV-1 infected cells to identify possible host factors involved during herpesvirus exit through the NE on the prediction that such proteins would disperse into the ER during infection. I identified a group of vesicle fusion proteins that play a role in this herpesvirus exit through the NE. Depletion of three identified vesicle fusion proteins decreased the growth of HSV-1 in host cells, yielding accumulation of viral particles in the nucleus. The second approach was to follow the fate of nuclear envelope transmembrane proteins (NETs) during HSV-1 infection. To address the question of how NET23/STING is involved in innate immunity I tested the hypothesis that this NET acts as a transport receptor to carry signals through the peripheral channels of the NPC when central channel transport is blocked by pathogens. FRAP was used to quantify the mobility of NET23/STING upon the induction of the innate immune response, finding an increase of the mobility for this protein in the NE. To further elucidate its role within the NE I tested whether some NE-NET23/STING binding partners were being redistributed between the nucleus and cytoplasm during innate immune responses. This revealed two of these binding partners normally redistribute upon innate immune response activation and this is blocked in cells knocked down for NET23/STING. Finally, I confirmed that NET23/STING contributes to chromatin remodelling during infection involving an increase in the H3K9Me3 epigenetic mark. Collectively, these data argue the identification of novel host proteins involved in herpesvirus nuclear egress and the finding of a new role for NET23/STING within the NE.

Lesions in the human Lyst gene are associated with the lysosomal disorder Chediak Higashi Syndrome. The absence of Lyst causes the formation of enlarged lysosome related compartments in all cells. This defect results in severe immunodeficiency, neurological dysfunction, and ultimately in death. Despite decades of research, the mechanism for how these enlarged compartments arise is not well established. Two opposing models have been proposed for Lyst function. The fission model describes Lyst as a positive regulator of fission from lysosomal compartments, while the fusion model identifies Lyst as a negative regulator of fusion between lysosomes. To date, a consensus on which model is correct has not been reached. This thesis details my investigation of Lyst function using Dictyostelium discoideum. To establish a definitive model for the function of the Dictyostelium Lyst ortholog, LvsB, we used assays that distinguish between defects in vesicle fusion versus fission. We compared the phenotype of cells defective in LvsB with that of two known fission defect mutants ([mu]3 and WASH null mutants). The temporal localization characteristics of the post-lysosomal marker vacuolin, as well as vesicular acidity and fusion dynamics of LvsB null cells are distinct from those of both fission defect mutants. These distinctions are predicted by the fusion defect model and implicate LvsB as a negative regulator of vesicle fusion. This work also presents evidence that LvsB antagonizes the function of two fusion regulatory proteins, Rab14 and dLIP5. The Dictyostelium Rab14 GTPase is known to stimulate lysosome fusion, and here we implicate dLIP5 as a promoter of Rab14 activity. Constitutive activation of Rab14 increases vesicle fusion in wild type cells but not in dLIP5 mutant cells. Thus, Rab14 activity is dependent on dLIP5. Additionally, the aberrant vesicle morphology and fusion phenotypes of LvsB mutant cells are suppressed by expression of dominant inactive Rab14 or disruption of dLIP5. This suppression suggests that LvsB antagonizes Rab14 activity to negatively regulate vesicle fusion. These studies validate the fusion model for LvsB function and provide new insights into the relationships that dictate vesicle fusion regulation. By extension, we propose that Lyst negatively regulates vesicle fusion by antagonizing the activity of a RabGTPase. / text

No description available.

No description available.

Proper neuronal function requires the preservation of appropriate neural excitability. An adaptive increase in neural excitability after exposure to agents that depress neuronal signaling blunts the sedative drug effects upon subsequent drug exposure. This adaptive response to drug exposure leads to changes in drug induced behaviors such as tolerance, withdrawal and addiction. Here I use Drosophila melanogaster to study the cellular and neuronal components which mediate behavioral tolerance to the anesthetic benzyl alcohol. I demonstrate that rapid tolerance to benzyl alcohol is a pharmacodynamic mechanism independent of drug metabolism. Furthermore, tolerance is a cell autonomous response which occurs in the absence of neural signaling. Using genetic and pharmacological manipulations I find the synapse to play an important role in the development of tolerance. In addition, the neural circuits that regulate arousal and sleep also alter benzyl alcohol sensitivity. Beyond previously described transcriptional mechanisms I find a post-translational role of the Ca2+-activated K+-channel, slowpoke in the development of tolerance. Finally, I explore a form of juvenile onset tolerance, which may have origins that differ from rapid tolerance. The implications of this study go beyond tolerance in Drosophila melanogaster to benzyl alcohol and can shed light on human drug tolerance, withdrawal and addiction. / text

Die integralen Vesikelmembranproteine Synaptophysin und Synaptobrevin interagieren in adulten Neuronen. Zusätzlich bildet Synaptobrevin mit den Plasmamembranproteinen Syntaxin und synaptosome-associated protein 25kDa (SNAP25) den SNAP-Rezeptor (SNARE)-Proteinkomplex, der Voraussetzung für die Fusion zwischen synaptischen Vesikeln und präsynaptischer Membran ist. Mit Synaptophysin interagierendes Synaptobrevin bindet jedoch nicht an den SNARE-Proteinen. Es wird daher vermutet, dass der Synaptophysin/Synaptobrevin-Komplex eine Art Reservepool für Synaptobrevin bei erhöhter neuronaler Aktivität darstellt und die Verfügbarkeit von Synaptobrevin während der Exozytose reguliert. Mit verschiedenen Ansätzen wurde versucht, den auf dem Vesikel befindlichen Komplex genauer zu charakterisieren und in seiner Funktion näher zu beschreiben. Nach Stimulation mit exozytosevermittelnden Substanzen dissoziierte der Synaptophysin/ Synaptobrevin-Komplex, sowohl unter nativen Bedingungen als auch bei Blockierung des finalen Fusionsereignisses. Dieser Prozess war calciumabhängig, konnte jedoch nicht durch die direkte Wirkung von Calcium ausgelöst werden. Die Untersuchung des Komplexes mit Hilfe von clostridialen Neurotoxinen zeigte, dass Synaptobrevin bevorzugt in Bindung an Synaptophysin und als Dimer gespalten wurde. Die Spaltung des SNARE-Proteins SNAP25 hatte keinen Einfluss auf die Komplexbildung. Die Verringerung des Cholesterolgehaltes der Membran führte zur Abnahme der Interaktion von Synaptophysin und Synaptobrevin, umgekehrt zeigte sich ein Anstieg bei zusätzlicher Cholesterolapplikation. In weiteren Experimenten konnte der C-terminale Teil des Synaptobrevins als für die Bindung zu Synaptophysin entscheidende Abschnitt identifiziert werden. Weiterhin konnte die erfolgreiche Translokation von rekombinanten Konstrukten aus Botulinumtoxin D und einem angekoppelten funktionstüchtigen Protein ins Zytosol gezeigt werden. / The vesicle associated membrane proteins synaptophysin and synaptobrevin interact in ma-ture neurones. Additionally synaptobrevin forms a complex with the plasma membrane pro-teins syntaxin and synaptosome-associated protein 25kDa (SNAP25), better known as the SNAP-Receptor (SNARE) complex, which is a prerequisite for fusion of the presynaptic and vesicle membranes. These two protein complexes however are mutually exclusive. It is as-sumed that the synaptophysin/synaptobrevin complex resembles a reserve pool for synapto-brevin and regulates the availability of synaptobrevin for the fusion process in case of in-creased synaptic activity. Different approaches where chosen to characterize this protein complex and to examine its function in more detail. After excessive stimulation the synaptophysin/synaptobrevin complex dissociates, even when the final fusion process is blocked. This step was dependent on the presence of cal-cium, though it could not be triggered directly by calcium administration. When using clos-tridial neurotoxins, synaptobrevin was preferentially cleaved in its homodimeric form and in the complex with synaptophysin. Cleavage of SNAP25 had no effect on the complex forma-tion. Depletion of cholesterol content decreases the interaction of synaptophysin with synap-tobrevin, while cholesterol treatment increases interaction. Further experiments indicated that synaptophysin binds to the the carboxy-terminal transmembrane part of synaptobrevin. Fur-thermore it could be shown that proteins attached to botulinum toxin can be delivered to the cytosol of neuronal cells, being fully active.

No description available.

Les membranes cellulaires naturelles sont des structures complexes et posent de nombreux problèmes lorsqu'elles sont étudiées dans leur forme native. Par conséquent, des systèmes modèles lipidiques plus simples sont souhaitables pour étudier les composants des membranes cellulaires et leur interaction avec les molécules biologiques. Immobiliser ces modèles lipidiques sur des surfaces solides métalliques, pour former des bicouches biomimétiques supportées (SLB pour Supported Lipid Bilayer en anglais), est encore plus avantageux grâce leur adaptabilité à de nombreuses techniques de caractérisation de surface, telles que la microscopie de force atomique (AFM), la spectroscopie de résonance des plasmons de surface (SPR), l’électrochimie et les spectroscopies vibrationnelles (IR, Raman). Former ces bicouches lipidiques supportées par fusion des vésicules a toujours été la technique la plus adaptée vue sa simplicité et son efficacité. Cependant, cette technique exige des conditions expérimentales critiques comme la nécessité de surfaces planes lisses et hydrophiles (mica, verre…), des vésicules à base de phospholipides zwitterioniques en phase fluide, une concentration élevée en lipides, et une longue durée d’incubation (>1h). Dans cette thèse, nous visons à développer une nouvelle méthode simple, rapide et polyvalente permettant de former une large gamme de bicouches biomimétiques supportées, de type zwitterionique et anionique, en phase gel et fluide sur un substrat d’or. Cette nouvelle approche consiste en l’utilisation des réactions d’appariement ionique générées par un système rédox confiné en surface pour induire l’assemblage de phospholipides et former la bicouche lipidique. Le premier objectif de cette thèse est d’étudier le comportement électrochimique d’une monocouche auto-assemblée de ferrocényldodécanethiolates (FcC12SAu) en présence de molécules amphiphiles avec des groupes anioniques de types carboxyle (sel d’acide gras) et phosphate (groupes qu’on trouve dans les phospholipides) et une simple chaîne hydrocarbonée. Dans le même contexte, nous viserons également l’utilisation des réactions d’appariement ionique pour induire l’assemblage des surfactants n-alkyl carboxylate et n-alkyl phosphate à l’interface SAM/électrolyte. Le second objectif de ce travail de thèse consiste en l’utilisation du système rédox confiné en surface pour déclencher par appariement ionique l’assemblage des phospholipides (molécules amphiphiles à double chaînes hydrocarbonées) pour former des bicouches biomimétiques supportées sur une surface d’or, à partir de vésicules unilamellaires, à température ambiante et en quelques minutes. La couverture de surface en ferrocènes et l’hydrophobicité/hydrophilicité de la surface seront altérées par la suite pour investiguer l’effet sur la formation des bicouches lipidiques supportées. / Natural cell membranes are complex structures and may present many problems when studied in their native form. It is therefore desirable to have simpler lipid bilayer systems to study the components of cell membranes and their interaction with biological molecules. Immobilizing these lipid membranes on metallic solid surfaces, to form Supported Lipid Bilayers (SLB), is more advantageous due to the integrity with a wide range of surface-sensitive characterization techniques, such as atomic force microscopy (AFM), surface plasmon resonance spectroscopy (SPR), electrochemistry and vibrational spectroscopies (IR, Raman). The preparation of SLBs by vesicle fusion has always been the most suitable technique due to its simplicity and efficiency, but it requires critical experimental conditions such as the need for smooth and hydrophilic flat surfaces (mica, glass...), vesicles based on zwitterionic phospholipids in fluid phase, high lipid concentration, and lengthy SLB preparation times (>1h). In this thesis, we aim to develop a new simple, fast, and versatile method to form a wide range of supported biomimetic bilayers using zwitterionic and anionic phospholipid vesicles in gel and fluid phase on a gold substrate. This new approach consists in the use of ionic pairing reactions generated by a surface-confined redox system to induce the assembly of phospholipids and form the lipid bilayer. The first part of this thesis focuses on studying the electrochemical behavior of a self-assembled monolayer of ferrocenyldodecanethiolates (FcC12SAu) in the presence of amphiphilic molecules containing a carboxyl (fatty acid salt) and phosphate anionic group and a single hydrocarbon chain. This part will also focus on the use of ion-pairing reactions to induce the assembly of n-alkyl carboxylate and n-alkyl phosphate surfactants at the SAM/electrolyte interface. The second and main objective of this thesis work was subsequently devoted to the use of the surface-confined redox system to trigger by ion-pairing the assembly of phospholipids (amphiphilic molecules with double hydrocarbon chains) to form biomimetic bilayers supported on a gold surface from unilamellar vesicles at room temperature and within minutes. The surface coverage of ferrocenes and the hydrophobicity/hydrophilicity of the surface will be altered later to investigate the effect on the formation of supported lipid bilayers.