published_or_final_version / Pathology / Doctoral / Doctor of Philosophy

<p>Individual proteins, and small collections of proteins, have been extensively studied for at least two hundred years. Today, more than 350 genomes have been completely sequenced and the proteomes of these genomes have been at least partially mapped. The inventory of protein coding genes is the first step toward understanding the cellular machinery. Recent studies have generated a comprehensive data set for the physical interactions between the proteins of <i>Saccharomyces cerevisiae</i>, in addition to some less extensive proteome interaction maps of higher eukaryotes. Hence, it is now becoming feasible to investigate important questions regarding the evolution of protein-protein networks. For instance, what is the evolutionary relationship between proteins that interact, directly or indirectly? Do interacting proteins co-evolve? Are they often derived from each other? In order to perform such proteome-wide investigations, a top-down view is necessary. This is provided by network (or graph) theory.</p><p>The proteins of the cell may be viewed as a community of individual molecules which together form a society of proteins (nodes), a network, where the proteins have various kinds of relationships (edges) to each other. There are several different types of protein networks, for instance the two networks studied here, namely metabolic networks and protein-protein interaction networks. The metabolic network is a representation of metabolism, which is defined as the sum of the reactions that take place inside the cell. These reactions often occur through the catalytic activity of enzymes, representing the nodes, connected to each other through substrate/product edges. The indirect interactions of metabolic enzymes are clearly different in nature from the direct physical interactions, which are fundamental to most biological processes, which constitute the edges in protein-protein interaction networks.</p><p>This thesis describes three investigations into the evolution of metabolic and protein-protein interaction networks. We present a comparative study of the importance of retrograde evolution, the scenario that pathways assemble backward compared to the direction of the pathway, and patchwork evolution, where enzymes evolve from a broad to narrow substrate specificity. Shifting focus toward network topology, a suggested mechanism for the evolution of biological networks, preferential attachment, is investigated in the context of metabolism. Early in the investigation of biological networks it seemed clear that the networks often display a particular, 'scale-free', topology. This topology is characterized by many nodes with few interaction partners and a few nodes (hubs) with a large number of interaction partners. While the second paper describes the evidence for preferential attachment in metabolic networks, the final paper describes the characteristics of the hubs in the physical interaction network of <i>S. cerevisiae</i>.</p>

G protein-coupled receptors (GPCRs) form a link between the cell and their environment when signaling pathways are activated upon ligand binding. However, the ligands and functions for many GPCRs remain to be determined. G protein-coupled receptor 50 (GPR50) is one such orphan, and its exact role is yet unknown. There is however emerging functional and genetic evidence suggesting a function for GPR50 in psychiatric illness and lipid metabolism. It was hypothesised that investigating GPR50’s protein-protein interactions would lead to a greater understanding of the role of GPR50 in normal brain functioning and in mental illness. Putative protein interactors were initially isolated by a yeast two-hybid study and were further tested here. To address GPR50’s links to mental illness, the GPR50∆502-505 deletion variant associated with mood disorders was also investigated. To test this hypothesis I sought to confirm some of the key yeast two-hybrid interactions. Using co-immunoprecipitation and immunocytochemistry the interaction of GPR50 with reticulon family members Nogo-A, Nogo-C and RTN3, and with cell-cell adhesion molecule CDH8 and lipid-associated protein ABCA2 were validated. In order to identify the location of interactions, subcellular fractionation of mouse brain and rt-PCR and immunohistochemistry in developing and adult mouse brain were performed. GPR50 and several interactors were found to be enriched at the synapse by subcellular fractionation of whole adult brain, and at embryonic day 18 (E18) and 5 weeks by rt-PCR. Colocalisation of GPR50 and interactors was found in the amygdala, hypothalamus, cortex and specific brain stem nuclei by immunohistochemistry. The discovery of GPR50 expression in noradrenergic, serotonergic and dopaminergic nuclei in the adult brain stem suggests a further role for GPR50 in neurotransmitter signaling and stress. To investigate the function of GPR50 two assays were performed that measure processes which are known to be affected by Nogo and RTN3: The first assay was a neurite outgrowth assay in Neuroscreen-1 cells, a PC12 cell clone. A significant increase in neurite length was detected after transient overexpression of GPR50 and this effect was increased in the GPR50∆502-505/T532A variant. Additionally GPR50-overexpression resulted in an increase in filopodia formation suggesting a role in actin dynamics. As a second functional assay in vitro BACE1 activity assays were performed in HEK293 cells. GPR50 but not GPR50∆502-505/T532A overexpression resulted in a significant increase in BACE1 activity. Lastly a final series of pilot experiments were performed to gain insight into the secondary structure of the C-terminal domain and the effects of the polymorphisms on structure. The 35kDa GPR50 C-terminal domain was purified and Circular Dichroism studies indicated a predominantly unstructured protein with increased a- helical content in the GPR50∆502-505 variant. The results in this thesis indicate a role for GPR50 in neuronal development and synaptic functioning. The results also strengthen an association with major mental illness, with links to several disease mechanisms.

Proteins are the functional units of cells. However, a major portion of the proteome does not have a known functional annotation. This dissertation explores protein -protein interactions, involving these uncharacterized or unknown function proteins. Initially, protein – protein interactions were tested and analyzed for paralogous proteins in Escherichia coli. To expand this concept further and to get an overview, protein – protein interactions were analyzed using ‘comparative interactomics’ for four pathogenic bacterial species including Escherichia coli, Yersinia pestis, Vibrio cholerae and Staphylococcus aureus. This approach was used to study unknown function protein pairs as well as to focus on uncharacterized hub proteins. The dissertation aims at using protein – protein interactions along with other research data about proteins as a possible approach to narrow down on functions of proteins.

Mouse polyomavirus (MPyV) is small non-enveloped DNA virus. Although this virus has been studied for almost 60 years, it still remains unclear, how can virus transport its genetic information to the cell nucleus. Also, the mechanism of virion morphogenesis is not well understood. First part of this work is focused on endocytic pathway which is used by MPyV for trafficking toward the cell nucleus. Using dominant negative mutant of caveolin-1 we showed that caveolin-1dependent endocytic pathway, described for SV40, is not used by MPyV for productive infection. MPyV is transported to early endosomes. Acidic milieu of endosomes is indispensable for productive infection. Preventing virus localisation into early endosomes (dominant negative mutant of Rab 5 GTPase) or endosomes alkalisation (by ammonium chloride or bafilomycin A1) led to dramatic decrease of virus infectivity. Alkalisation of endosomes entailed retention of MPyV in early endosomes. It indicates that virus is further transported to late endosomes. Finally, we confirmed by FRET that MPyV is in perinuclear space localized into recycling endosomes. Another poor characterized process is virion morphogenesis. To characterize the participation of cellular proteins in virion precursor complexes, nuclear as well as whole-cell lysates of infected cells or...

This thesis provides a physics account of the ability of cells to integrate environmental information to make complex decisions, a process commonly known as signaling. It strives to address the following questions: (i) How do cells relate the state of the environment (e.g. presence/absence of specific molecules) to a desired response such as gene expression? (ii) How can cells robustly transfer information? (iii) Is there a biophysical limit to a cells' ability to process information? (iv) Can we use the answers to the above questions to formulate biophysical principles that inform us about the evolution of signaling? Throughout, I borrow techniques from non-equilibrium statistical physics, statistical learning theory, information theory and information geometry to construct biophysical models capable of making quantitative experimental predictions. Finally, I address the connection of energy expenditure and biological efficiency by zeroing in on a process unique to eukaryotic cells-- nuclear transport. The thesis concludes with a discussion of our theory and its implications for synthetic biology.

A existência de uma maior variedade de aminoácidos codificados pelo código genético tem estimado estudos sobre os mecanismos de síntese, reconhecimento e incorporação desses resíduos nas cadeias polipeptídicas nascentes. Um exemplo é a via de incorporação de selenocisteína evento cotraducional dirigido pelo códon UGA. Em bactérias, essa via conta com uma complexa maquinaria molecular composta por: Selenocisteína Sintase (SelA), Fator de Elongação Específico de Reconhecimento (SelB), Selenofosfato Sintetase (SelD), tRNA específico (SelC ou tRNAsec), sequência específica no mRNA (Sequência de Inserção de Selenocisteínas - SECIS) e Aminoacil tRNA Sintetase (aaRS). Pelo fato do selênio ter uma toxicidade elevada em ambientes celulares, é fundamental a compreensão do mecanismo catalítico e razão estequiométrica na formação dos complexos da via na etapa de incorporação junto ao tRNAsec, bem como sua caracterização estrutural foram os objetivos deste trabalho. A proteína SelA foi expressa e purificada para utilização em análises envolvendo microscopia de força atômica, microscopia eletrônica de transmissão com contraste negativo e em gelo vítreo foram realizadas nos complexos SelA e SelA-tRNAsec, visando obter um modelo estrutural e a razão estequiométrica dos complexos. A fim de compreender o mecanismo de passagem do selênio, ensaios de anisotropia de fluorescência e de microcalorimetria, corroborados pelas análises de troca de hidrogênio-deutério acoplado a espectrometria de massa e espectroscopia de infravermelho, elucidaram a formação e estequiometria do complexo ternário SelAtRNA sec-SelD. Tentativas de cristalização e análises cristalográficas também foram realizadas, no entanto, sem sucesso. Com os resultados obtidos foi possível propor que o reconhecimento de SelD e, consequentemente, a entrega do selenofosfato, seja uma etapa crucial da via de incorporação de selenocisteínas. / The existence of a greate variety of amino acids encoded by the genetic code has stimulated the study of the mechanisms of synthesis, recognition and incorporation of these residues in the nascent polypeptide chains. An example of genetic code expansion is the selenocysteine incorporation pathway an event cotraducional by the UGA codon. In bacteria, this pathway has a complex molecular machinery comprised: Selenocysteine Synthase (SelA), Specific Elongation Factor (SelB), Selenophosphate Synthetase (SelD), tRNA-specific (SelC or tRNAsec), Specific mRNA Sequence (SElenocysteine Insertion Sequence - SECIS) and Aminoacyl tRNA Synthetase (aaRS). Because selenium has high toxicity in cellular environments; it is essential for cell survival the association of this compound with proteins, in this case, selenoprotens and the associated proteins involved in the selenocysteine synthesis. Therfore the understanding of the catalytic mechanism, stoichiometric ratio, protein complex formation with the tRNAsec, and its structural characterization were the objectives of this work. The SelA protein was expressed and purified to used in analyzes involving atomic force microscopy, transmission electron microscopy with negative stain and in vitreous ice were performed in the complex SelA and SelA-tRNAsec in order to obtain a structural model of the complex and the stoichiometric ratio of its components. To study the selenium association with protein of the synthesis pathway, fluorescence anisotropy assays and isothermal titration calorimetry corroborated by the analysis hydrogen-deuterium exchange coupled to mass spectrometry and infrared spectroscopy were employed.Crystallization attempts were made and preliminary crystallographic analyzes were also performed, however, so far unsuccessfuly. The results obtained were possible to develop the hypothesis about the SelD recognition and, consenquently, the selenophosphate delivery, a crucial stage of the selenocysteine incorporation pathway.

Protein interactions occurring through biomolecular interfaces play an important role in the circle of life. These interactions are responsible for cellular function, including RNA transcription, protein translation, cell division and cell death among many others. There are different types of interactions based on the strength and the duration of the association. Transient interactions govern most steps of the cellular metabolism, where the associations between two or more molecules are responsive to environmental cues. Among the participants of transient interactions, intrinsically disordered proteins are employed in signaling and other regulatory events within the cell. These proteins exhibit allosteric regulation and gain secondary structure when they bind other proteins or small molecules. In this doctoral thesis work, the biochemical and biophysical principals governing protein associations are investigated and using protein engineering tools, novel biomolecular interfaces are engineered, with potential applications in different areas of biotechnology. The first part of the thesis (Chapter 2) focuses on the investigation of supramolecular enzyme association among tricarboxylic acid cycle enzymes, specifically between citrate synthase and mitochondrial malate dehydrogenase. In this study, the interactions between these enzymes are examined, both among their natural and synthetically produced recombinant versions. In addition, mutational analysis of the amino acid residues at the complex interface was performed to explore the importance of the positively charged patch connecting the active sites of the enzymes. It was discovered that the channeling of the negatively charged intermediate is severely impaired upon mutation of surface residues contributing to the electrostatic channeling. This work provides an important insight into understanding the coupled reaction-transport systems and metabolon formation in general. In addition, it constitutes a great example for substrate channeling in leaky systems, which are relevant to most biological processes. The next section of the thesis (Chapter 3) focuses on an intrinsically disordered peptide, the β-roll. This peptide is isolated from the Block V repeats-in-toxin (RTX) domain of adenylate cyclase from Bordetella pertussis. It is disordered in the absence of calcium and it folds into a β-roll secondary structure composed of two parallel β-sheet faces upon binding to calcium ions. This way, the peptide can transition between its unfolded state and the β-roll structure in a reversible way. We have utilized the allosteric regulation of this domain as a tool to engineer new protein interfaces. In its folded state, the peptide has two faces, serving as binding surfaces available for interaction with other proteins. Our work involved the alteration of the residues, which form these faces upon calcium binding, via combinatorial protein design techniques. The potential of this peptide is evaluated as a cross-linking domain for hydrogel formation. By rationally engineering the two faces of the folded β-roll to contain leucine residues, we have created hydrophobic interfaces, serving as environmentally-responsive cross-linking domains. When there is no calcium, the β-roll domains remain unstructured, delocalizing the leucine rich patches. After calcium binding, the β-rolls fold and the leucine rich faces are exposed creating a hydrophobic driving force for self-assembly. This way, we showed that the β-roll peptide can function as a biomaterials building block capable of proteinaceous hydrogel formation, only in the presence of calcium. The next study (Chapter 4) demonstrates the utilization of this peptide as an alternative scaffold for biomolecular recognition applications. A library of mutant β-rolls was constructed by randomizing the amino acid residues on one of the β-sheet forming faces. Mutant peptides demonstrating an affinity for hen egg white lysozyme were selected, which was chosen as a model target molecule. The thermodynamic parameters of the interactions between the β-roll mutants and the lysozyme were quantified. Upon performing further protein engineering (e.g. concatenation of the single mutants on the DNA level), a mutant with mid-nanomolar affinity was identified. Affinity chromatography experiments showed that this mutant was capable of capturing the target, in the presence of calcium. The captured target was easily released upon removal of the calcium ions. The reversibility of the calcium binding allowed the engineered molecular interface to be controllable. Throughout this study, the β-roll peptide was explored as an allosterically-regulated protein switch for on/off biomolecular recognition, which can be mediated by simply changing the calcium concentration, allowing control over the binding behavior between molecules. The last part of the thesis (Chapter 5) expands on the calcium dependent network formation study. A hydrogel construct was genetically built by fusing the cross-linking β-roll domain and the lysozyme binding β-roll mutant, resulting in a smart biomaterial with dual-functionality. The network-assembly and target capture functions of this construct were tested by various assays including hydrogel erosion experiments. This allosterically-regulated biomaterial exhibited promising results, where calcium-dependent lysozyme entrapment within the assembled network and lysozyme capture on the hydrogel surface were demonstrated. The work presented in this thesis demonstrates different approaches to understand and engineer molecular interfaces in both natural and recombinant systems. In the future, these approaches and the knowledge gained from these studies can be further built upon for different biotechnological applications and can also be applied to other synthetic systems.

Les interactions protéine–protéine (IPP) jouent un rôle primordial dans les processus physiologiques. L’inhibition de ces interactions ouvre la voie à la conception de nouvelles molécules à visée thérapeutique. Les structures secondaires en hélice α sont fréquemment impliquées dans les interactions entre protéines auxquelles elles peuvent contribuer de manière significative. La conception de molécules, mimant ce motif de reconnaissance et pouvant interagir avec la protéine cible tout en inhibant la reconnaissance avec le partenaire naturel, représente une voie innovante pour trouver de nouveaux candidats médicaments. Les oligomères d’urée aliphatique, une classe de foldamères qui adoptent une structure secondaire en hélice bien définie et proche de l’hélice α, ont été proposés comme mimes d’hélice α pour inhiber les IPP. Au cours de cette thèse, nous nous sommes d’abord intéressés à la conception de foldamères d’oligourée et de chimères oligoamide/oligourée pour cibler des surfaces de protéine. Nous avons sélectionné le récepteur nucléaire de la vitamine D (VDR) comme modèle d’étude en raison de son intérêt thérapeutique, et des connaissances structurales disponibles. Les protéines partenaires de VDR (coactivateurs) interagissant via une courte région structurée en hélice α, nos recherches ont portés sur des mimes d’hélices inspirés des séquences de coactivateurs. Dans une seconde partie, nous nous sommes intéressés à la génération et à l’étude de dimères covalents de foldamères, qui pourraient être utilisés pour couvrir des surfaces d’interaction plus larges. En effet, les interactions protéine-protéine montrent souvent un mode d’interaction plus complexe qu’une simple hélice, faisant intervenir des structures tertiaires et quaternaires de type coiled coils, qui peuvent aussi servir de point de départ pour la conception de nouvelles classes d’inhibiteurs. / Protein-protein interactions (PPI) have a key role in physiological processes. The inhibition of these PPI may lead to new therapeutic strategies. Secondary structures in α-helix are frequently involved in protein interactions where they may contribute significantly to binding. Designing molecules which mimic the helical motif for protein surface recognition and inhibition of the natural partner represents an innovative path to discover new drug candidates. Aliphatic urea oligomers, a class of foldamers that adopt a well-defined H-bonded helical secondary structure with good similarity to the α-helix have been proposed as possible α-helix mimics to inhibit protein-protein interactions. The first part of this PhD project was dedicated to the design and synthesis of oligoureas and oligourea/α-peptide chimeras for specific protein surface recognition. We have selected the vitamin D receptor as a potential target, mainly because (i) it is therapeutically relevant; (ii) its protein partner (coactivators) interact through a short region which adopts an α-helical structure upon binding and (iii) structures at atomic resolution were available to enable the design of effective mimetics. In the second part, we investigated methods to generate foldamer covalent dimers that could potentially be used to cover larger interaction surfaces. The rationale is that the binding interface is often more complex than a single helix and may involve tertiary and quaternary structures such as coiled coils which in turns may also serve as a basis for the design of new classes of inhibitors.

Antibodies and antibody fragments are the largest class of biotherapeutics in development with many products already available in the clinic. Antibodies are promising due to their naturally high affinity and specificity for biological targets. A key stumbling block to biopharmaceutical development compared to small molecule drugs is the general requirement for a stable liquid formulation, which is often difficult to obtain due to issues with aggregation, phase separation, particle formation, and chemical instabilities. Aberrant solution behaviour limits the production, storage and delivery of the monoclonal antibody. Biopharmaceutical solution behaviour is determined by weak, transient protein-protein and protein-solvent interactions. An attractive interaction potential between proteins in solution can lead to association. Irreversible association occurs when proteins undergo large scale structural changes and aggregate. Reversible association is less severe, but can lead to undesirable solution properties such as high viscosity, phase separation and opalescence, which can lead to difficulties throughout the downstream processing and formulation steps. These problems can become exacerbated during formulation of antibodies when trying to achieve high protein concentrations often required for effective antibody dosage. Firstly, we studied the domains of the Fab fragment using statistical models and continuum electrostatic calculations and found that the CH1 domain is more soluble than the other domains and has properties of intrinsically disordered like proteins which is supported by observations in the literature. We then investigated the immunoglobulin superfamily and found 11 proteins which may have a similarly disordered nature. We present a new web server for predicting protein solubility from primary sequence using an in-house algorithm that weighs the contribution of various sequence properties for predicting solubility. Lastly, we conducted physical characterisation of an antibody and human serum albumin in pharmaceutically relevant buffers and found that the interaction potential can be modelled using spherical models from low to high protein concentration. We hope that the work outlined in this thesis will contribute to the theoretical understanding and modelling of protein solution behaviour.