The present investigation is concerned with determining the role of calcium, phospholipids, and phospholipid metabolites on transepithelial sodium and water transport in response to antidiuretic hormone (ADH). These studies utilize the frog skin for determining sodium transport and amphibian urinary bladder for water flow measurements and scanning electron microscopy of cell surface morphology. The results demonstrate that phospholipids and phospholipid metabolites containing arachidonic acid stimulate transepithelial sodium transport through amiloride sensitive channels and the action of these lipids involves the synthesis of prostaglandins. These lipids also inhibited the increase in water flow induced by ADH, and this effect was prevented with prostaglandin synthesis inhibitors. Prostaglandins alter intracellular calcium concentrations and agents effecting calcium metabolism alter cell surface morphology and the changes in surface substructure induced by ADH. These observations support the hypothesis that alterations in membrane permeability to water and ions may involve metabolism of membrane phospholipids and prostaglandin biosynthesis.

Dehydrins are intrinsically disordered plant stress-proteins. Repetitively in their sequence are some highly conserved stretches of 7-17 residues, the so called K-, S-, Y- and lysine rich segments. This thesis aims to give insight into the possible role dehydrins have in the stressed plant cell with main focus on membrane interaction and protection. The work includes four recombinant dehydrins from the plant Arabidopsis thaliana: Cor47 (SK3), Lti29 (SK3), Lti30 (K6) and Rab18 (Y2SK2). Initially, we mimicked crowded cellular environment in vitro to verify that dehydrins are truly disordered proteins. Thereafter, the proposal that the compulsory K-segment determines membrane binding was tested. Experiments show that only Lti30 and Rab18 bind, whereas Cor47 and Lti29 does not. As Lti30 and Rab18 binds they assembles vesicles into clusters in vitro, a feature used to characterize the interaction. From this it was shown that membrane binding of Lti30 is electrostatic and determined by global as well as local charges. Protonation of histidine pairs flanking the K-segments works as an on/off-binding switch. By NMR studies it was shown that the K-segments form a dynamic α-helix upon binding, so called disorder-to-order behaviour. Also, dehydrins electrostatic interaction with lipids can be further tuned by posttranslational phosphorylation or coordination of calcium and zinc ions. Finally, specific binding of Rab18 to inositol lipids, mainly PI(4,5)P2, is reported. The interaction is mainly coordinated by two arginines neighboring one of the K-segments. In conclusion, the K-segments are indeed involved in the binding of dehydrins to membrane but only in combination with extensions (Lti30) or modified (Rab18). / <p>At the time of the doctoral defense, the following papers were unpublished and had a status as follows: Paper 4: Manuscript. Paper 5: Manuscript.</p>

Thèse numérisée par la Division de la gestion de documents et des archives de l'Université de Montréal.

La polarité cellulaire, organisation asymétrique du matériel cellulaire dans l'espace et le temps, est fréquemment observée en biologie. Elle est nécessaire pour de nombreux mécanismes cellulaires essentiels allant de la division cellulaire et la migration au développement et la croissance polarisée. Comprendre comment la cellule génère et maintient cette polarité est crucial, les défauts de polarité étant liés à des maladies graves comme le cancer ou les maladies neurodégénératives. Chez la levure Saccharomyces cerevisiae, la polarité cellulaire est établie lorsque le module de la Rho GTPase Cdc42, qui comprend le facteur d'échange de nucléotide guanine (GEF) Cdc24 et la protéine scaffold Bem1, localise à un unique site à la membrane plasmique pour activer Cdc42 et ainsi, établir un axe de polarité utilisé pour la croissance et la division cellulaire. Les mécanismes responsables de l'activation de Cdc42 à un site unique au cortex pendant l'établissement de la polarité sont essentiels mais largement inconnus. En utilisant des expériences complémentaires d'imagerie in vivo et des expériences in vitro, je mis en évidence que le ciblage avide du module de Cdc42 à la membrane plasmique implique des interactions multivalentes entre des lipides anioniques et le module de Cdc42. En détail, j'ai démontré que la combinaison de plusieurs phospholipides anioniques, comprenant PS, PI4P et PI(4,5)P2, est nécessaire à la localisation de Bem1 et Cdc24 in vivo. J'ai identifié des groupements cationiques interagissant avec des lipides (CLICs) dans l'extrémité N-terminale de Bem1 qui étaient nécessaires et suffisants pour interagir avec des phospholipides anioniques. Réduire l’interaction de Bem1 avec les lipides en mutant la séquence CLICs a fortement diminué la localisation de Bem1 au niveau du cortex ainsi que la signalisation de Cdc42. En plus des CLICs de Bem1, le domaine PX de Bem1 et le domaine PH de Cdc24 augmentent davantage l'avidité du module GTPase pour les lipides anioniques et la combinaison des trois domaines est essentielle pour l'établissement de la polarité cellulaire. Ces résultats définissent pour la première fois le mécanisme de ciblage avide des activateurs de Cdc42 à la membrane plasmique pendant l'établissement de l'axe de polarité. / Cell polarity, the asymmetric organization of cell material in space and time, is frequently observed in biology. It is required for numerous essential cellular processes ranging from cell division and migration to development and polarized growth. Addressing how cells generate and maintain polarity is crucial, since defects in polarity are linked to severe diseases including cancer and neurodegeneration. In the budding yeast Saccharomyces cerevisiae, cell polarity is established when the Cdc42 Rho GTPase module, which includes the Guanine nucleotide Exchange Factor (GEF) Cdc24 and the scaffold protein Bem1, accumulate at a unique site on the plasma membrane to activate Cdc42 and establish the polarity axis used for cell growth and division. The mechanisms responsible for the site-specific activation of Cdc42 at the cortex during polarity establishment are essential but are largely unknown. Using complementary in vivo imaging and in vitro experiments, I found that the avid targeting of the Cdc42 GTPase module to the plasma membrane involves multivalent anionic lipid-Cdc42 module interactions. I found that a combination of anionic phospholipids, including PS, PI4P and PI(4,5)P2, are necessary for Bem1 and Cdc24 localization in vivo. I identified Cationic-enriched Lipid Interacting Clusters (CLICs) in the N-terminus of Bem1 that were necessary and sufficient for anionic phospholipid interactions. Reducing Bem1 lipid binding by mutating the CLICs strongly diminished the localization of Bem1 at the cortex and Cdc42 signaling. In addition to the Bem1 CLICs, the Bem1 PX domain and the Cdc24 PH domain increased the avidity of the GTPase module for anionic lipids, and a combination of all three domains was essential for the establishment of cell polarity. The results of my thesis define a mechanism of avid targeting of Cdc42 activators to the cortex during polarity axis establishment.

Entre as estratégias que os protozoários do gênero Leishmania apresentam para o escape da resposta imune do hospedeiro vertebrado, há a ocorrência de um tipo de morte celular programada, conhecida como apoptose. Quando em contato com o macrófago, a Leishmania é fagocitada de forma silenciosa, evitando a resposta inflamatória do hospedeiro vertebrado. Alguns autores defendem que a Leishmania mimetiza a apoptose, expondo entre outras moléculas um fosfolipídio que sinalizaria para o macrófago que está em apoptose, e esse mecanismo é denominado na literatura como mimetismo apoptótico. O objetivo desta tese foi elucidar como ocorre esse escape com o enfoque nos fosfolipídios presentes e expostos em parasitas mutantes de L. (L.) amazonensis com características fenotípicas distintas, utilizando diferentes estratégias: transfecção com cosmídeos contendo frações do genoma de L. (L.) amazonensis; identificação e clonagem do gene pi4k contido no cosmídeo em vetor de expressão em Leishmania; seleção de parasitas resistentes a miltefosina, mantidos ou não sob pressão do antibiótico, seleção de parasitas na 28a passagem em cultura; seleção de parasitas purificados de macrófagos de linhagem RAW. A caracterização desses mutantes foi realizada em relação à ligação de anexina V-FITC, infectividade em macrófagos da linhagem RAW, tomada de fosfolipídios fluorescentes (NBD), IC50 de células tratadas com os antibióticos duramicina, miltefosina e anfotericina B. De acordo com os ensaios de ligação à anexina V-FITC, identificamos que os mutantes pi4k-pSNBR e os mutantes resistentes à miltefosina apresentaram maior ligação à anexina V-FITC. O gene que codifica a fosfatidilinositol (PI) 4-kinase, fez com que os parasitas que continham tanto o cosmídeo como o superexpressor pi4k, apresentassem menor infectividade em relação ao controle selvagem. O mesmo ocorreu para os parasitas resistentes à miltefosina. Em contrapartida, os parasitas derivados desses resistentes, mas mantidos sem pressão do antibiótico, recuperaram os valores de infectividade comparáveis ao grupo controle. Interessante é que os parasitas resistentes à miltefosina, mantidos ou não sob pressão, assim como o parasita superexpressor do gene pi4k apresentaram maior ligação à anexina V-FITC em relação ao controle selvagem, indicando que a ligação à anexina V-FITC não se correlaciona com infectividade. Parasitas resistentes à miltefosina, mantidos ou não sob pressão apresentaram maior sensibilidade à duramicina, e quando tratados com anfotericina B, esses parasitas apresentaram maior resistência. Uma outra abordagem analisada nessa tese foi elucidar qual o fosfolipídio é reconhecido pelo macrófago durante a infecção por L. (L.) amazonensis. Como resultado, observamos que os lipossomas contendo PC levam à diminuição dose-dependente da infecção, o que não foi visto em PS ou PC:PE. Esse resultado sugere a importância de PC para o estabelecimento da infectividade. / Among the strategies that Leishmania protozoans present to escape the immune response of the vertebrate host, there is a type of programmed cell death, known as apoptosis. When in contact with the macrophage, Leishmania is phagocytized in a silent manner, avoiding the inflammatory response of the vertebrate host. Some authors argue that Leishmania mimics apoptosis, exposing among other molecules a phospholipid that would signal to the macrophage that is in apoptosis, and this mechanism is denominated in the literature as apoptotic mimicry. The objective of this thesis was to elucidate how this escape occurs with the focus on the phospholipids present and exposed in mutant parasites of L. (L.) amazonensis, with distinct phenotypic characteristics, using different strategies, such as selection of parasites showing higher attachment to annexin V-FITC. After transfection with cosmids containing the genome of L. (L.) amazonensis; identification and cloning of the pi4k gene contained in the cosmid in Leishmania expression vector; selection of parasites resistant to miltefosine, whether or not under antibiotic pressure, selection of parasites at the 28th passage in culture; selection of purified strains of RAW lineage macrophages. The characterization of these mutants was performed in relation to the annexin V-FITC binding, infectivity in RAW lineage macrophages, fluorescent phospholipid (NBD) uptake, IC50 of cells treated with the antibiotics duramycin, miltefosine and amphotericin B. According to the binding assays to Annexin V-FITC, we have identified that pi4k-pSNBR mutants and miltefosin-resistant mutants showed higher attachment to annexin V-FITC. The gene coding for phosphatidylinositol (PI) 4-kinase caused the parasites containing both the cosmid and the pi4k superexpressor to have less infectivity than the wild-type control. The same occurred for parasites resistant to miltefosine. In contrast, the parasites derived from these resistant, but kept without antibiotic pressure, recovered infectivity values, comparable to the control group. Interestingly, miltefosine-resistant parasites, whether or not under pressure, as well as the overexpressing parasite of the pi4k gene showed greater attachment to annexin V-FITC over wild-type control, indicating that attachment to annexin V-FITC does not correlate with infectivity. Parasites resistant to miltefosine, whether or not under pressure, showed greater sensitivity to duramycin, and when treated with amphotericin B, these parasites showed greater resistance. Another approach analyzed in this thesis was to elucidate the phospholipid recognized by the macrophage during infection by L. (L.) amazonensis. As a result, we observed that PC-containing liposomes lead to dose-dependent decrease of infection, which has not been seen in PS or PC: PE. This result suggests the importance of PC for the establishment of infectivity.

The aim of the thesis project was to examine what form(s) of Amyloid beta (Aβ) (25-­‐35) peptide interact with phospholipids in vitro and the implications of this for the mechanism of Alzheimer’s Diseases (AD). The mechanism of AD is thought to involve protein folding and misfolding. An increasing amount of evidence has shown that protein misfolding plays an important role in the biological and pathological processes of AD. Although seen as the biomedical markers of those diseases, the roles of amyloid aggregates themselves are still not fully understood. Whether the aggregates, or the monomer, or some other intermediates of Aβ cause AD is still unknown. In order to investigate the membrane-­‐interaction of Aβ and its implications for AD, two forms of Aβ, namely levorotary and dextrorotary (L-­‐ and D-­‐) Aβ isomers were used. Evidence has shown that L-­‐ and D-­‐ peptide can each form aggregates in a humid environment. However, when mixed together, L-­‐ and D-­‐ peptides tend not to form any aggregates. Using the mixtures of L-­‐ and D-­‐ peptides at different proportions and as well as using L-­‐ and D-­‐ alone can help us to determine the toxic form of Aβ. Phospholipids have been used to mimic membrane bilayers. Biological membranes in vivo are a complicated system. They contain three types of lipids, namely phospholipids, glycolipids, and steroids. Different types of cells and different membranes have different proportions of those lipids. Studying the interaction between Aβ and membranes in vivo can be extremely difficult. Artificial membranes, which only contains one kind of lipids, on the other hand, are a useful tool for the study of molecular interactions. Phospholipids are the most abundant type of membrane lipid and thus that can be seen as representative of cell membranes. The interactions of Aβ and different kinds of phospholipids have been investigated in this project. This thesis discusses the secondary structure of Aβ in different environment, the interaction between Aβ and phospholipids at the air-­‐water surface, and the location of Aβ in membranes during the interaction. The study provides useful information of the mechanisms and the origin of AD. At the end of the thesis, a discussion chapter analyses the difficulties of studying Aβ and AD and the potentials and inadequacies of this research.

Phospholipids (PLs) are a component of cell membranes, biological fluids and tissues. These compounds are problematic for the bioanalytical chemist, especially when PLs are not the analytes of interest. PL interference with bioanalysis highly impacts reverse-phase chromatographic methods coupled with mass spectrometric detection. Phospholipids are strongly retained on hydrophobic columns, and can cause significant ionization suppression in the mass spectrometer, as they out-compete analyte molecules for ionization. Strategies for improving analyte detection in the presence of PLs are reviewed, including in-analysis modifications and sample preparation strategies. Removal of interfering PLs prior to analysis seems to be most effective at moderating the matrix effects from these endogenous cellular components, and has the potential to simplify chromatography and improve column lifetime. Products targeted at PL removal for sample pre-treatment, as well as products that combine multiple modes of sample preparation (i.e. Hybrid SPE), show significant promise in mediating the effect on PL interference in bioanalysis.

The work presented in this dissertation is a composite of experiments in the growth of gold nanoparticles with specific optical properties of interest. The goal is to synthesize these gold nanoparticles using soybean extract for not only shape control, but for propensity as a biocompatible delivery system. The optical properties of these nanoparticles has found great application in coloring glass during the Roman empire and, over the centuries, has grown into its own research field in applications of nanoparticulate materials. Many of the current functions include use in biological systems as biosensors and therapeutic applications, thus making biocompatibility a necessity. Current use of cetyltrimethylammonium bromide leads to rod-shaped gold nanoparticles, however, the stability of these gold nanoparticles does not endure for extended periods of time in aqueous media. In my research, two important components were found to be necessary for stable, anisotropic growth of gold nanoparticles. In the first experiments, it was found that bromide played a key role in shape control. Bromide exchange on the gold atoms led to specific packing of the growing crystals, allowing for two-dimensional growth of gold nanoparticles. It was also discerned that soybean lecithin contained ligands that blocked specific gold facets leading to prismatic gold nanoparticle growth. These gold nanoprisms give a near infrared plasmon absorption similar to that of rod-shaped gold nanoparticles. These gold nanoprisms are discovered to be extremely stable in aqueous media and remain soluble for extended periods of time, far longer than that of gold nanoparticles grown using cetyltrimethylammonium bromide. Since soy lecithin has a plethora of compounds present, it became necessary to discover which compound was responsible for the shape control of the gold nanoprisms in order to optimize the synthesis and allow for a maximum yield of the gold nanoprisms. Many of these components were identified by high performance liquid chromatography and liquid chromatography-mass spectrometry. However, re-spike of these components into growth solutions did not enhance the growth of gold nanoprisms. Upon separating the shapes of the gold nanoparticles using gel electrophoresis, addition of KCN to the separated gold nanoparticles allowed us to extract the culpable ligands for shape control. Analysis of these ligands by mass spectrometry elucidated the identity of PA and upon re-spike of the PA into a growth solution of PC95, the growth of a near-infrared plasmon absorption was seen. The stability of these gold nanoparticles was tested with and without the addition of decane thiol and it was concluded that addition of the thiol allowed for improved stability of the gold nanoparticles towards cyanide. It was determined that at a concentration of 2 μM decanethiol, spherical gold nanoparticles remained stable to cyanide at the expense of the prismatic gold nanoparticles. However, at 5 μM decanethiol, both spherical and prismatic gold nanoparticles retained stability to cyanide in aqueous conditions.

Biological membranes mainly constituent lipid molecules along with some proteins and steroles. The properties of the pure lipid bilayers as well as in the presence of other constituents (in case of two or three component systems) are very important to be studied carefully to model these systems and compare them with the realistic systems. Molecular dynamic simulations provide a good opportunity to model such systems and to study them at microscopic level where experiments fail to do. In this thesis we study the structural and dynamic properties of the pure phospholipid bilayers and the phase behavior of phospholipid bilayers when other constituents are present in them. Material and structural properties like area per lipid and area compressibility of the phospholipids show a big scatter in experiments. These properties are studied for different system sizes and it was found that the increasing undulations in large systems effect these properties. A correction was applied to area per lipid and area compressibility using the Helfrich theory in Fourier space. Other structural properties like order of the lipid chains, electron density and radial distribution functions are calculated which give the structure of the lipid bilayer along the normal and in the lateral direction. These properties are compared to the X-ray and neutron scattering experiments after Fourier transform. Thermodynamic properties like heat capacity and heat of melting are also calculated from derivatives of energies available in molecular dynamics. Heat capacity on the other hand include quantum effect and are corrected for that by applying quantum correction using normal mode analysis for a simple as well as ambiguous system like water. Here it is done for SPC/E water model. The purpose of this study is to further apply the quantum corrections on macromolecules like lipids by using this technique. Furthermore the phase behavior of two component systems (phospholipids/cholesterol) is also studied. Phase transition in these systems is observed at different cholesterol concentrations as a function of temperature by looking at different quantities (as an order parameter) like the order of chains, area per molecule and partial specific area. Radial distribution functions are used to look at the in plane structure for different phases having a different lateral or positional order. Adding more cholesterol orders the lipid chains changing a liquid disordered system into a liquid ordered one and turning a solid ordered system into a liquid ordered one. Further more the free energy of domain formation is calculated to investigate the two phasecoexistence in binary systems. Free energy contains two terms. One is bulk freeenergy which was calculated by the chemical potential of cholesterol moleculein a homogeneous system which is favorable for segregation. Second is thefree energy of having an interface which is calculated from the line tension of the interface of two systems with different cholesterol concentration which in unfavorable for domain formation. The size of the domains calculated from these two contributions to the free energy gives the domains of a few nm in size. Though we could not find any such domains by directly looking at our simulations. / <p>QC 20120913</p>

Thesis (M.Sc.), Memorial University of Newfoundland, 2000. / Bibliography: leaves 115-118.