Prion diseases are a group of fatal neurodegenerative diseases affecting almost all mammals including humans. The diseases are caused by formation of the misfolded isoform of the cellular prion protein (PrPC) to the disease causing PrPSc. The focus on this work has been to characterize molecular changes in persistently scrapie-infected murine neuronal cells possibly contributing to prion-induced neurodegeneration. PrPC is localized to lipid rafts in the plasma membrane and this is also the place where it is suggested that the conformational change into PrPSc occurs. This work shows an increased expression of active Src kinase in scrapie-infected cells resulting in an increased overall tyrosine phosphorylation of several proteins. Additionally, an increase in the specific tyrosine kinase activity of Fyn is shown. Interestingly, the membrane distribution of Fyn from non-raft to raft-domains followed that of PrPSc in scrapie-infected cells as analyzed by immunoblotting of flotation-fractions after ultracentrifugation of Triton X-100 extracted cell lysates. This indicates a persistent Fyn activation, probably due to clustering of intracellular Fyn kinases due to PrPSc accumulation in lipid rafts. In addition to an increased Src family kinase activity in scrapie-infected cells these cells also express an increased number of insulin receptor (IR)/insulin-like growth factor-1 receptor (IGF-1R) hybrid receptors, and these receptors display an altered protein glycosylation of the IR subunits. Additionally, ScN2a cells do not respond to LPS-stimulation with NO production, putatively due to the lack of CD14 mRNA. Together, these findings may have pathological implications leading to neuronal cell death in prion diseases via several mechanisms which are discussed in this thesis.

Two independent endosymbiotic events gave rise to mitochondria and chloroplasts. Despite the fact that both organelles have their own small genome the majority of organellar proteins are encoded in the nucleus, synthesized in the cytosol and imported into the organelles. The targeting information for most organellar proteins is located in an N-terminal extension called a targeting peptide. Targeting peptides are cleaved off after import by organellar processing peptidases. The cleaved targeting peptides are toxic to organellar functions and are degraded by the PreP peptidasome, the metalloendopeptidase which is the main topic of this thesis. We have overexpressed, purified and determined the first structure of a plant mitochondrial targeting peptide, the F1β presequence from Nicotiana plumbaginifolia, by NMR in a membrane mimetic environment. The structure showed that the targeting peptide formed two helices separated by an unstructured domain. The N-terminal helix being amphipatic. The F1β targeting peptide has been used as a model substrate for the mitochondrial and chloroplast PreP peptidasome. In Arabidopsis thaliana the PreP peptidasome is present as two isoforms, AtPreP1 and AtPreP2. We have shown that both forms are expressed and dually targeted to mitochondria and chloroplasts. Both AtPreP1 and AtPreP2 degrade targeting peptides and other non-related unstructured peptides up to 65 amino acid residues. Substrate specificity studies showed that both PreP isoforms have a preference for positively charged amino acid residues in the P1′ position and small uncharged residues in the P1 position. Mapping of cleavage sites revealed unique cleavage sites for both isoforms. We have generated and characterized both single and double AtPreP1 and AtPreP2 knockouts in A. thaliana. AtPreP1 was shown to be the major isoform. The double knockout exhibited a chlorotic phenotype with altered mitochondrial and chloroplast morphology. Furthermore,mitochondria were partially uncoupled. Throughout the development there was a slower growth rate and 40% lower biomass production. These results show that the PreP peptidasome is important for efficient organellar functions and normal plant development.

Heparan sulfate (HS) and heparin are linear polysaccharide chains covalently O-linked to serine residues within the core proteins, so called HS proteoglycans (PGs) or heparin PG. HSPGs are produced by almost all mammalian cells and known to play important roles in developmental processes, physiological and pathological conditions; whereas heparin PG is produced by mast cells and best known as an anticoagulant in clinic. Biosynthesis of HS/heparin occurs in Golgi compartment and involves many enzymes, one of which is glucuronyl C5-epimerase (Hsepi) that catalyzes the conversion of D-glucuronic acid (GlcA) to L-iduronic acid (IdoA). Heparanase is an enzyme involved in metabolism of HS; it cleaves the linkage between GlcA and glucosamine residues in HS/heparin chains. Heparanase is expressed essentially by all cells and found up-regulated in many metastatic tumors. This thesis focuses on the structure and functions of HS/heparin through studies on the implications of Hsepi and heparanase. My study demonstrated that the modification catalyzed by Hsepi is critical for HS-dependent function of growth factors, especially FGF2; heparanase is involved in regulation of HS biosynthesis and matrix metalloproteinases expression; moreover, my experimental data demonstrated the functions of heparin in mast cells, showing cleavage of heparin by heparanase contributes to modulation of protease storage in mast cells.

Rearrangement of the actin cytoskeleton occurs in a variety of cellular processes and structures and involves a wide spectrum of proteins. Among these, the gelsolin superfamily proteins (GSPs) control actin organization by severing filaments, capping filament ends and bundling filaments. Structural changes within the GSPs are key in controling their functions. This thesis is aimed in understanding the activation mechanisms of the C-terminal halves of GSPs through investigating the atomic structures of gelsolin, adseverin and villin. X-ray crystallography was used to determine the structures of C-terminal fragments of these 3 proteins. The results demonstrate that: 1) The structure of the activated form of the C-terminal half of gelsolin displays an open conformation, with the actin-binding site on gelsolin domain 4 (G4) fully exposed and all three type-II calcium binding sites (CBS) occupied. Neither actin nor the type-I calcium, which is normally sandwiched between actin and G4, is required to achieve this conformation. 2) Calcium ions at both type-I and type-II CBSs of gelsolin were exchangable within the crystals. Extraction of calcium ions from the CBSs triggered local conformation changes which we speculate are the initial steps toward restoration of the arrangement of domains found in the calcium-free inactive form of gelsolin in solution. 3) The long helix of G6 in the calcium-bound structure is similar to the helix of calcium-free isolated villin domain 6 (V6). 4) The conformation of the C-terminal half of adseverin in the active state is similar to that of gelsolin. These results suggest that the C-terminal halves of GSPs are activated before forming a complex with actin. The activation involves straightening the helix of domain 6 which is a key component in the global conformation changes of C-terminal halves of these proteins. The results also suggest that a calcium ion may bind to the type-I CBS on domain 4 of the active conformation of GSPs concurrently with forming the complex with actin, hence, stabilizing the GSP:actin complex.

In both pro- and eukaryotes 20-30% of all genes encode alpha-helical transmembrane domain proteins, which act in various and often essential capacities. Notably, membrane proteins play key roles in disease and they constitute more than half of all known drug targets. The natural abundance of membrane proteins is in general too low to conveniently isolate sufficient material for functional and structural studies. Therefore, most membrane proteins have to be obtained through overexpression. Escherichia coli is one of the most successful hosts for overexpression of recombinant proteins. While the production of soluble proteins is comparably straightforward, overexpression of membrane proteins remains a challenging task. The yield of membrane localized recombinant membrane protein is usually low and inclusion body formation is a serious problem. Furthermore, membrane protein overexpression is often toxic to the host cell. Although several reasons can be postulated, the basis of these difficulties is not completely understood, preventing the design of rational strategies to improve membrane protein overexpression yields. The objective of my Ph.D. studies has been to improve membrane protein overexpression in E. coli by a) understanding membrane protein overexpression from the perspective of membrane protein biogenesis, b) systematically investigating the physiological response to overexpression of membrane proteins and c) engineering strains that are optimized for membrane protein overexpression based on insights resulting from these studies. By working toward these objectives, I was able to identify and alleviate one of the major bottlenecks of membrane protein overexpression in E. coli: saturation of the Sec-translocon could be overcome by harmonizing translation and membrane insertion of the recombinant membrane protein. This minimized the toxic effects of overexpression and thus resulted in increased membrane protein-producing biomass.

The use of recombinant protein is a cornerstone in many structural and functional studies. The enteric bacterium Escherichia Coli is the most commonly used organism for producing recombinant proteins. E. coli has several advantages over other expression hosts, but also one major disadvantage - the protein of interest does not always adopt its native conformation. Instead the protein might form large insoluble aggregates, inclusion bodies, within the cell. In particular, the heterologous overexpression of eukaryotic and membrane proteins are troublesome. In this thesis, methods are described that can be used to increase the likelihood of overexpressing eukaryotic proteins as well as membrane proteins. In particular, a novel method is described that can distinguish between bacterial colonies expressing soluble proteins from those expressing inclusion bodies. The method utilizes the fact that inclusion bodies are of a considerable size and can be removed by filtration. Using this screening method in combination with methods that alter the physical properties of proteins, we have shown that the likelihood of overexpression in E. coli can be dramatically increased.

The principles of natures own light-driven water splitting catalyst, Photosystem II (PSII), can in the future inspire us to use water as electron and proton source to generate light-driven H2 production. To mimic this challenging step, it is important to understand how the enzyme system can oxidize water. The mechanism of light-driven water oxidation in PSII is in this thesis addressed by EPR spectroscopy. P680+ is a strong oxidant formed by light-oxidation of the chlorophyll species P680 positioned in the center of PSII. The redox active tyrosine-Z (YZ) can reduce P680+ and the YZ• radical is formed. This transient radical is further reduced by the CaMn4-cluster, which is the binding site of the substrate water molecules. In a cyclic process called the S-cycle, this catalytic cluster accumulates four oxidizing equivalents to evolve one molecule of O2 and to oxidize two molecules of water. We can induce the YZ• radical at cryogenic temperatures in the different oxidation states of the catalytic S-cycle and observe this in metalloradical EPR signals. These metalloradical EPR signals are here characterized and used to deduce mechanistic information from the intact PSII. The "double nature" of these spin-spin interaction signals, so called split EPR signals, makes them excellent probes to both YZ oxidation and, when YZ• is present, also to the S-states of the CaMn4-cluster. The metalloradical EPR signals presented here, form a way to study the transient YZ• radical in active PSII that has not been depleted of the catalytic metal cluster. This depleting method that has often been used in the past to study YZ is not representing studies of a mechanistically relevant material. The previously suggested disorder around YZ and accessibility to the bulk can be artifactual properties induced in the mechanistically defect PSII. On the contrary, our observation that proton coupled electron transfer from YZ to the light induced P680+ can occur in a high yield at cryogenic temperatures, suggests a well ordered catalytic site in the protein positioned for optimal performance. The optimized positioning of the redox components found in PSII might be a feature also important to build in an efficient water oxidizing catalyst.

Quantum chemical studies of enzymatic reactions are able to provide detailed insight into mechanisms and catalytic strategies. The energetic feasibility of proposed mechanisms can be established, and new possible reaction pathways can be put forward. The role of the involved active site residues can be analyzed in detail and the origins for experimentally observed selectivities can be investigated. Density functional theory (DFT), in particular the hybrid functional B3LYP, is the method of choice in this kind of studies. In this thesis, the reaction mechanisms of several enzymes have been explored using the B3LYP functional. The studied enzymes include limonene epoxide hydrolase (LEH), soluble epoxide hydrolase (sEH), haloalcohol dehalogenase (HheC), and nitrile hydratase (NHase). Transition states and intermediates along various reaction pathways were optimized and evaluated. For the three epoxide-transforming enzymes, the role of the proposed catalytic residues could be confirmed. Analysis of in silico mutations helped to quantify the effect of various functional groups on the barriers and regioselectivities of epoxide opening. A detailed analysis of the factors governing the enzymatic regioselectivities is given. For nitrile hydratase, various putative first- and second-shell mechanisms have been studied. Active site models based on both the Co(III)-NHase and the Fe(III)-NHase were employed. The studied mechanisms include general base-catalyzed reaction pathways with water as nucleophile as well as two pathways involving cysteine-sulfenate as nucleophile. Several computed mechanisms exhibit similar barriers, making it difficult to pinpoint the true NHase mechanism. / QC 20100811

This thesis is devoted to theoretical studies of the role of nuclear vibrations on nonlinear and linear absorption, pulse propagation, and resonant scattering of light. The molecular parameters needed for the simulations are obtained through suitable quantum chemical calculations, which are compared with available experimental data. The first part of the thesis addresses to modeling of ampli ed spontaneous emission (ASE) in organic chromophores recently studied in a series of experiments. To explain the threshold behavior of the ASE spectra we invoke the idea of competition between di erent ASE channels and non-radiative quenching of the lasing levels. We show that the ASE spectrum changes drastically when the pump intensity approaches the threshold level, namely, when the ASE rate approaches the rate of vibrational relaxation or the rate of solute-solvent relaxation in the rst excited state. According to our simulations the ASE intensity experiences oscillations. Temporal self-pulsations of forward and backward propagating ASE pulses occur due to two reasons: i) the interaction of co- and counter-propagating ASE, and ii) the competition between the ampli ed spontaneous emission and o -resonant absorption. In the second part of the thesis we explore two-photon absorption taking into account nuclear vibrational degrees of freedom. The theory, applied to the N101 molecule [p-nitro-p'- diphenylamine stilbene], shows that two-step absorption is red shifted relative to one-photon absorption spectrum in agreement with the measurements. The reason for this e ect is the one-photon absorption from the first excited state. Simulations show that two mechanisms are responsible for the population of this state, two-photon absorption and offresonant one-photon absorption by the wing of the spectral line. In the third part of the thesis we study multi-photon dynamics of photobleaching by a periodical sequence of short laser pulses. It is found that the photobleaching as well as the uorescence follow double-exponential dynamics. The fourth part of the thesis is devoted to the role of the nuclear dynamics in x-ray spectroscopy. Our studies show that the vibronic coupling of close lying core excited states strongly a ects the resonant x-ray Raman scattering from ethylene and benzene molecules. We demonstrate that the manifestation of the non-adiabatic e ects depends strongly on the detuning of photon energy from the top of photoabsorption. The electronic selection rules are shown to break down when the excitation energy is tuned in resonance with the symmetry breaking vibrational modes. Selection rules are then restored for large detuning. We obtained good agreement with experiment. Finally, our multi-mode theory is applied to simulations of the resonant Auger and x-ray absorption spectra of the ethyne molecule. / QC 20100813

The molecular structure of a protein decides its function, its way to interact with other molecules. Using X-ray crystallography methods, a 3-dimensional, atomic model of a macromolecule can be determined. In this thesis work, the X-ray structures of two different proteins involved in human diseases were studied: FocB, which is associated with urinary tract infections, and transthyretin, which is the causative of hereditary systemic transthyretin amyloidosis. FocB is a 12 kDa protein which binds DNA in an oligomeric fashion. It is involved in the regulation of the expression of bacterial surface organelles (fimbriae), responsible for the adhesion to specific receptors in host tissue. Specifically, FocB regulates the expression of one fimbrial type found in uropathogenic E. coli (UPEC): F1C. Our FocB structure revealed it to be an all-alpha helical protein with an atypical helix-turn-helix (HTH) motif. Residues previously found important for DNA-binding in the FocB homologue PapB, were not located in the putative “recognition helix” of the HTH-motif. FocB was also found to bind to the minor groove of the DNA. Together with homology searches showing that the DNA-interactions possible for FocB are greatly diversified, these findings indicated a DNA-interaction different from the typical DNA-interaction of a HTH-protein. Transthyretin (TTR) is a plasma protein involved in transport of thyroxin (T4) and retinol. Mutated TTR is also the cause of the neurodegenerative disease hereditary systemic transthyretin amyloidosis, which is characterized by systemic deposition of TTR amyloid fibrils. The amyloid occurs through a process of TTR tetramer destabilization and partial unfolding. A common way to inhibit amyloid formation is to design small molecules that bind unoccupied thyroxin binding sites and stabilize the tetrameric form of the protein. The structural characterization of the binding of chloride and iodide ions to TTR revealed that two of three previously identified halogen binding pockets in the T4-binding site were just as optimal for halide binding. In addition, a third halide-binding site, bridging two TTR subunits, was found. In biochemical experiments, chloride and iodide ions were shown to stabilize the TTR structure and inhibit the TTR aggregation and/or amyloid formation, with iodide ions doing so more efficiently than the chloride ions. In the search for new TTR amyloid-inhibiting drugs, the identified halide-binding sites in the T4-binding pocket are possible starting points for structure-based drug design.