No description available.

In dieser Arbeit wurden ungesättigte Vinyl- und Ethynylphosphonothiolate hergestellt und als Linker für Cystein-selektive Proteinmodifikationen verwendet. Zunächst wurde, ausgehend von ungesättigten Phosphoniten und elektrophilen Disulfiden als Startmaterialien, eine Syntheseroute zur Herstellung von ungesättigten Phosphonothiolaten entwickelt. Die hohe Chemoselektivität dieser Reaktion ermöglichte die Einführung von Vinylphosphonothiolaten auf ungeschützten Modellpeptiden und dem Protein Ubiquitin. Es konnte gezeigt werden, dass ungesättigte Phosphonothiolate unter neutralen bis leicht basischen Bedingungen selektiv mit Thiolen reagieren und sich daher als Linker für Cystein-selektive Proteinmodifikationen eignen. Die Vielseitigkeit der hier entwickelten Biokonjugationsmethode wurde in drei Anwendungen demonstriert. Ausgehend von dem Vinylphoshonothiolat-modifizierten Ubiquitin konnten homogene Ubiquitin-Protein-Konjugate erzeugt werden; zum einen ein nicht hydrolysierbares Diubiquitin-Konjugat und zum anderen ein Ubiquitin-α–Synuclein-Konjugat. Des Weiteren wurden ungesättigte Phosphonothiolate als Linker in Antikörper-Wirkstoff-Konjugaten getestet. Hier zeigte sich, dass insbesondere Vinylphosphonothiolat-Linker Potential zur Herstellung von stabilen Antikörper-Konjugaten aufweisen. Schließlich wurden Vinylphosphonothiolate als Linker verwendet, um sowohl Chaperon-bindende Antikörper als auch Deubiquitinasen (DUBs) mit photoreaktiven Crosslinkern auszustatten um damit dynamische Protein-Interaktionen zu untersuchen. Insgesamt ermöglicht das hier entwickelte Verfahren die chemoselektive Umwandlung von elektrophilen Disulfiden in elektrophile Vinyl- und Ethynylphosphonothiolate, wodurch Reaktivität für eine Thioladdition induziert wird. Dadurch können zwei komplexe, thiolhaltige Moleküle selektiv konjugiert werden, was insbesondere für die Herstellung von homogen modifizierten Peptid- und Proteinkonjugaten von Bedeutung ist. / In this work, unsaturated vinyl- and ethynylphosphonothiolates were synthesised and used as linkers for cysteine-selective protein modifications. First, a synthetic route for the generation of unsaturated phosphonothiolates was developed, using unsaturated phosphonites and electrophilic disulfides as starting materials. The high chemoselectivity of this reaction enabled the introduction of vinylphosphonothiolates on unprotected model peptides and the protein ubiquitin. It could then be shown that unsaturated phosphonothiolates react selectively with thiols under neutral to slightly basic conditions and are therefore suitable as linkers for cysteine-selective protein modifications. The versatility of the herein developed bioconjugation method was demonstrated in three applications. First, starting from the vinylphosphonothiolate-modified ubiquitin, homogeneous ubiquitin-protein conjugates could be generated, in particular a non-hydrolyzable diubiquitin conjugate and a ubiquitin-α-synuclein conjugate. Second, the suitability of unsaturated phosphonothiolates as linkers for the generation of stable antibody-drug conjugates was tested. Vinylphosphonothiolate linkers thereby showed potential to produce stable antibody conjugates. Finally, vinylphosphonothiolates were used as linkers to conjugate both chaperone-binding antibodies and deubiquitinases (DUBs) to photo-reactive crosslinkers in order to investigate dynamic protein interactions. Overall, the herein developed methodology enables the chemoselective conversion of electrophilic disulfides into electrophilic vinyl- and ethynylphosphonothiolates, which in turn react selectively with thiols. As a result, two complex, thiol-containing molecules can be selectively conjugated, which is particularly important for the production of homogeneously modified peptide and protein conjugates.

Veränderungen in phosphorylierungsabhängigen Signalwegen, Akkumulation von Proteinaggregaten im Gehirn und neuronaler Zelltod sind Neurodegenerationskennzeichen und Indikatoren für überlappende molekulare Mechanismen. Um Einblicke in die involvierten Signalwege zu erhalten, wurde mit Hilfe eines modifizierten Hefe-Zwei-Hybrid (Y2H)-Systems für 71 Proteine, die mit neurologischen Erkrankungen assoziiert sind, proteomweit nach Protein-Protein Interaktionen (PPIs) gesucht. Für 21 dieser Proteine wurden PPIs identifiziert. Das Gesamtnetzwerk besteht aus 79 Proteinen und 90 PPIs von denen 5 phosphorylierungsabhängig sind. Ein Teil dieser PPIs wurde in unabhängigen Interaktionsassays mit einer Validierungsrate von 66 % getestet. Der netzwerkbasierte Versuch verbindet erfolgreich neurologische Erkrankungen untereinander aber auch mit zellulären Prozessen. Ser/Thr-Kinase abhängige PPIs verknüpfen zum Beispiel das Parkinson Protein 7 (PARK7, DJ1) mit den E3 Ligase Komponenten ASB3 und RNF31 (HOIP). Die Funktion dieser Proteine bekräftigt den Zusammenhang zwischen dem Ubiquitin-Proteasom-System und der Parkinson Krankheit (PD). Neurofibromin 2 (NF2, merlin) Isoformen und PARK7 interagieren mit der regulatorischen PI3K Untereinheit p55-gamma (PIK3R3). Diese PPIs basieren auf Tyr-Kinase Aktivität im modifizierten Y2H System und funktionellen PIK3R3 pTyr-Erkennungsmodulen (SH2 Domänen) in co-IP und Venus PCA Versuchen. Dies verknüpft den PI3K/AKT Überlebenssignalweg mit zwei unterschiedlichen neurologischen Erkrankungsphenotypen: dem PD assoziierten neuronalen Zelltod und der Neurofibromatose Typ 2-assoziierten Tumorentstehung. Die vergleichende Beobachtung von PIK3R3, AOF2 (KDM1A, LSD1) Interaktionen auf NF2 Isoformlevel offenbart eine Bevorzugung von Isoform 7 bei zytoplasmatischer Lokalisation, wohingegen Isoform 1 PPIs an der Membran lokalisiert sind. Das modifizierungsabhängige und isoformspezifische PPI Netzwerk ermöglichte neue Hypothesen zu molekularen Pathomechanismen. / Alterations in phosphorylation-dependent signalling pathways, accumulation of aggregated proteins in the brain and neuronal apoptosis are common to neurodegeneration and implicate overlapping molecular mechanism. To gain insight into involved pathways, a modified yeast-two hybrid (Y2H) system was applied to screen 71 proteins associated with neurological disorders in a proteome-wide manner. For 21 of these proteins interactions were identified including 5 phosphorylation-dependent ones. In total, the network connected 79 proteins through 90 protein-protein interactions (PPIs). A fraction of these Y2H PPIs was tested in secondary interaction assays with a validation rate of 66 %. The described network-based approach successfully identified proteins associated with more than one disorder and cellular functions connected to specific disorders. In particular, the network revealed Ser/Thr kinase-dependent PPIs between the Parkinson protein 7 (PARK7, DJ1) and the E3 ligase components ASB3 and RNF31 (HOIP). The function of these proteins further substantiates the established connection between Parkinson’s disease (PD) and ubiquitination-mediated proteasome (dis)functions. Neurofibromin 2 (NF2, merlin) isoforms and PARK7 were identified as PI3K regulatory subunit p55-gamma (PIK3R3) interactors. These PPIs required Tyr kinase coexpression in the modified Y2H system and functional PIK3R3 pTyr-recognition modules (SH2 domains) in co-IP and Venus PCA experiments. This finding implicates the PI3K/AKT survival pathway in PD-associated neuronal apoptosis and Neurofibromatosis type 2-associated tumour formation. Investigation of PIK3R3, AOF2 (KDM1A, LSD1) and EMILIN1 PPIs on NF2 isoform level revealed preferential isoform 7 binding and cytoplasmic or membrane localisation of these PPIs for isoform 7 or 1, respectively. The generated modification-dependent and isoform-specific PPI network triggered many hypotheses on the molecular mechanisms implicated in neurological disorders.

Seit Jahrhunderten wird menschliches Wissen in Form von natürlicher Sprache ausgetauscht und in Dokumenten schriftlich aufgezeichnet. In den letzten Jahren konnte man auf dem Gebiet der Lebenswissenschaften eine exponentielle Zunahme wissenschaftlicher Publikationen beobachten. Diese Dissertation untersucht die automatische Extraktion von Beziehungen zwischen Eigennamen. Innerhalb dieses Gebietes beschäftigt sich die Arbeit mit der Steigerung der Robustheit für die Relationsextraktion. Zunächst wird der Einsatz von Ensemble-Methoden anhand von Daten aus der "Drug-drug-interaction challenge 2013" evaluiert. Ensemble-Methoden erhöhen die Robustheit durch Aggregation unterschiedlicher Klassifikationssysteme zu einem Modell. Weiterhin wird in dieser Arbeit das Problem der Relationsextraktion auf Dokumenten mit unbekannten Texteigenschaften beschrieben. Es wird gezeigt, dass die Verwendung des halb-überwachten Lernverfahrens self training in solchen Fällen eine höhere Robustheit erzielt als die Nutzung eines Klassifikators, der lediglich auf einem manuell annotierten Korpus trainiert wurde. Zur Ermittlung der Robustheit wird das Verfahren des cross-learnings verwendet. Zuletzt wird die Verwendung von distant-supervision untersucht. Korpora, welche mit der distant-supervision-Methode erzeugt wurden, weisen ein inhärentes Rauschen auf und profitieren daher von robusten Relationsextraktionsverfahren. Es werden zwei verschiedene Methoden untersucht, die auf solchen Korpora trainiert werden. Beide Ansätze zeigen eine vergleichbare Leistung wie vollständig überwachte Klassifikatoren, welche mit dem cross-learning-Verfahren evaluiert wurden. Um die Nutzung von Ergebnissen der Informationsextraktion zu erleichtern, wurde die semantische Suchmaschine GeneView entwickelt. Anforderungen an die Rechenkapazität beim Erstellen von GeneView werden diskutiert und Anwendungen auf den von verschiedenen Text-Mining-Komponenten extrahierten Daten präsentiert. / For several centuries, a great wealth of human knowledge has been communicated by natural language, often recorded in written documents. In the life sciences, an exponential increase of scientific articles has been observed, hindering the effective and fast reconciliation of previous finding into current research projects. This thesis studies the automatic extraction of relationships between named entities. Within this topic, it focuses on increasing robustness for relationship extraction. First, we evaluate the use of ensemble methods to improve performance using data provided by the drug-drug-interaction challenge 2013. Ensemble methods aggregate several classifiers into one model, increasing robustness by reducing the risk of choosing an inappropriate single classifier. Second, this work discusses the problem of applying relationship extraction to documents with unknown text characteristics. Robustness of a text mining component is assessed by cross-learning, where a model is evaluated on a corpus different from the training corpus. We apply self-training, a semi-supervised learning technique, in order to increase cross-learning performance and show that it is more robust in comparison to a classifier trained on manually annotated text only. Third, we investigate the use of distant supervision to overcome the need of manually annotated training instances. Corpora derived by distant supervision are inherently noisy, thus benefiting from robust relationship extraction methods. We compare two different methods and show that both approaches achieve similar performance as fully supervised classifiers, evaluated in the cross-learning scenario. To facilitate the usage of information extraction results, including those developed within this thesis, we develop the semantic search engine GeneView. We discuss computational requirements to build this resource and present some applications utilizing the data extracted by different text-mining components.

Die dynamische Regulation von Protein-Protein Interaktionen (PPIs) ist wichtig für den Ablauf von biologischen Prozessen. Die circadiane Uhr, die einen ~24 Stunden Rhythmus generiert und eine Vielzahl von physiologischen Parametern steuert kann auch die Dynamik von PPIs regulieren. Um neue Erkenntnisse über regulatorische Mechanismen innerhalb des molekularen Oszillators zu gewinnen, habe ich zunächst alle möglichen PPIs zwischen 46 circadianen Komponenten mittels eines systematischen yeast-two-hybid (Y2H) Screens bestimmt. Dabei habe ich 109 bis dahin noch unbekannte PPIs identifiziert und einen repräsentativen Anteil mittels Co-Immunopräzipitationsexperimenten in humanen Zellen validiert. Unter den neuen PPIs habe ich bis dahin unbekannte Modulatoren der CLOCK/BMAL1 Transaktivierung identifiziert und dabei die Rolle der Proteinphosphatase 1 (PP1) als dynamischen Regulator der BMAL1 Stabilität funktionell charakterisiert. Das experimentelle PPI Netzwerk wurde mit bereits aus der Literatur bekannten PPIs und Interaktionspartnern ergänzt. Eine systematische RNAi Studie belegte außerdem die Relevanz der aus der Literatur stammenden Interaktoren für die ~24 Stunden Periodizität. Um eine Aussage über die Dynamik der PPIs im Netzwerk treffen zu können, wurden circadiane mRNA Expressionsdaten in das PPI Netzwerk integriert. Systematische Perturbationsstudien, in denen alle Komponenten des experimentellen Netzwerkes mittels RNAi herunterreguliert oder überexprimiert wurden, zeigten eine essentielle Bedeutung für die dynamischen PPIs innerhalb des circadianen Oszillators auf. Desweiteren wurden im circadianen PPI Netzwerk funktionelle Module identifiziert, welche dynamisch organsiert sind. Durch eine systemweite Analyse des humanen Proteoms wurden viele dynamische PPIs identifiziert, die biologische Prozesse wie z.B. Signaltransduktion und Zellzyklus miteinander verbinden. Rhythmische PPIs sind daher von Bedeutung für die zeitliche Organisation zellulärer Physiologie. / Essentially all biological processes depend on protein-protein interactions (PPIs). Timing of such interactions is crucial for regulatory function. Although circadian (~24 hrs) clocks constitute fundamental cellular timing mechanisms regulating important physiological processes PPI dynamics on this timescale are largely unknown. To elucidate so far unknown regulatory mechanisms within the circadian clockwork, I have systematically mapped PPIs among 46 circadian components using high-throughput yeast-two-hybrid (Y2H) interaction experiments. I have identified 109 so far uncharacterized interactions and successfully validated a sub-fraction via co-immunoprecipitation experiments in human cells. Among the novel PPIs, I have identified modulators of CLOCK/BMAL1 function and further characterized the role of protein phosphatase 1 (PP1) in the dynamic regulation of BMAL1 abundance. Furthermore, to generate a more comprehensive circadian PPI network, the experimental network was enriched and extended with additional interactions and interaction partners from literature, some of which turned out to be essential for normal circadian dynamics. The integration of circadian mRNA expression profiles allowed us to determine the interaction dynamics within our network. Systematic genetic perturbation studies (RNAi and overexpression in oscillating human cells) revealed a crucial role of dynamic regulation (via rhythmic PPIs) for the molecular clockwork. Furthermore, dynamic modular organization as a pervasive circadian network feature likely contributes to time-of-day dependent control of many cellular processes. Global analysis of the proteome regarding circadian regulation of biological processes via rhythmic PPIs revealed time-of-day dependent organization of the human interactome. Circadian PPIs dynamically connect many important cellular processes like signal transduction and cell cycle, which contribute to temporal organization of cellular physiology.

Membrane protein dynamics is of great importance for living organisms. The precise localization of proteins composing a synapse on the membrane facing a nerve terminus is essential for proper functioning of the nervous system. In muscle fibers, the nicotinic acetylcholine is densely packed under the motor nerve termini. A receptor associated protein, rapsyn, acts as a linker between the receptor and the other components of the synaptic suramolecular assembly. Advances in fluorescence microscopy have allowed to measure the behavior of a single receptor in the cell membrane. In this work single-molecule microscopy was used to track the motion of ionotropic acetylcholine (nAChR) and serotonin (5HT3R) receptors in the plasma membrane of cells. We present methods for measuring single-molecule diffusion and their analysis. Single molecule tracking has shown a high dependence of acetylcholine receptors diffusion on its associated protein rapsyn. Comparing muscle cells that either express rapsyn or are devoid of it, we found that rapsyn plays an important role on receptor immobilization. A three-fold increase of receptor mobility was observed in muscle cells devoid of rapsyn. However, in these cells, a certain fraction of immobilized receptors was also found immobile. Furthermore, nAChR were strongly confined in membrane domains of few tens of nanometers. This showed that membrane composition and membrane associated proteins influence on receptor localization. During muscle cell differentiation, the fraction of immobile nAChR diminished along with the decreasing nAChR and stable rapsyn expression levels. The importance of rapsyn in nAChR immobilization has been further confirmed by measurements in HEK 293 cells, where co-expression of rapsyn increased immobilization of the receptor. nAChR is a ligand-gated ion-channel of the Cys-loop family. In mammals, members of this receptor family share general structural and functional features. They are homo- or hetero-pentamers and form a membrane-spanning ion channel. Subunits have three major regions, an extracellular ligand binding domain, a transmembrane channel and a large intracellular loop. 5HT3R was used as a model to study the effect of this loop on receptor mobility. Single-molecule tracking experiments on receptors with progressively larger deletions in the intracellular loop did not show a dependence of the size of the loop on the diffusion coefficient of mobile receptors. However, two regions were identified to play a role in receptor mobility by changing the fractions of immobile and directed receptors. Interestingly, a prokaryotic homologue of cys-loop receptors, ELIC, devoid of a large cytoplasmic loop was found to be immobile or to show directed diffusion similar as the wild-type 5HT3R. The scaffolding protein rapsyn stabilizes nAChR clusters in a concentration dependent manner. We have measured the density and self-interactions of rapsyn using FRET microscopy. Point-mutations of rapsyn, known to provoke myopathies, destabilized rapsyn self-interactions. Rapsyn-N88K, and R91L were found at high concentration in the cytoplasm suggesting that this modification disturbs membrane association of rapsyn. A25V was found to accumulate in the endoplasmic reticulum. Fluorescent tools to measure intracellular concentration of calcium ions are of great value to study the function of neurons. Rapsyn is highly abundant at the neuromuscular junction and thus is a genuine synaptic marker. A fusion protein of rapsyn with a genetically encoded ratiometric calcium sensor has been made to probe synapse activity. This thesis has shown that the combined use of biologically relevant system and modern fluorescence microscopy techniques deliver important information on pLGIC behaviour in the cell membrane. / <p>QC 20151217</p>

No description available.

L'étude structurale de l'interactome cellulaire peut conduire à des découvertes intéressantes sur les bases moléculaires de certaines pathologies. La modélisation par homologie et l'amarrage de protéines ("protein docking") sont deux approches informatiques pour modéliser la structure tri-dimensionnelle (3D) d'une interaction protéine-protéine (PPI). Des études précédentes ont montré que ces deux approches donnent de meilleurs résultats quand des données expérimentales sur les PPIs sont prises en compte. Cependant, les données PPI ne sont souvent pas disponibles sous une forme facilement accessible, et donc ne peuvent pas être re-utilisées par les algorithmes de prédiction. Cette thèse présente une approche systématique fondée sur l'extraction de connaissances pour représenter et manipuler les données PPI disponibles afin de faciliter l'analyse structurale de l'interactome et d'améliorer les algorithmes de prédiction par la prise en compte des données PPI. Les contributions majeures de cette thèse sont de : (1) décrire la conception et la mise en oeuvre d'une base de données intégrée KBDOCK qui regroupe toutes les interactions structurales domaine-domaine (DDI); (2) présenter une nouvelle méthode de classification des DDIs par rapport à leur site de liaison dans l'espace 3D et introduit la notion de site de liaison de famille de domaines protéiques ("domain family binding sites" ou DFBS); (3) proposer une classification structurale (inspirée du système CATH) des DFBSs et présenter une étude étendue sur les régularités d'appariement entre DFBSs en terme de structure secondaire; (4) introduire une approche systématique basée sur le raisonnement à partir de cas pour modéliser les structures 3D des complexes protéiques à partir des DDIs connus. Une interface web (http://kbdock.loria.fr) a été développée pour rendre accessible le système KBDOCK / Understanding how the protein interactome works at a structural level could provide useful insights into the mechanisms of diseases. Comparative homology modelling and ab initio protein docking are two computational methods for modelling the three-dimensional (3D) structures of protein-protein interactions (PPIs). Previous studies have shown that both methods give significantly better predictions when they incorporate experimental PPI information. However, in general, PPI information is often not available in an easily accessible way, and cannot be re-used by 3D PPI modelling algorithms. Hence, there is currently a need to develop a reliable framework to facilitate the reuse of PPI data. This thesis presents a systematic knowledge-based approach for representing, describing and manipulating 3D interactions to study PPIs on a large scale and to facilitate knowledge-based modelling of protein-protein complexes. The main contributions of this thesis are: (1) it describes an integrated database of non-redundant 3D hetero domain interactions; (2) it presents a novel method of describing and clustering DDIs according to the spatial orientations of the binding partners, thus introducing the notion of "domain family-level binding sites" (DFBS); (3) it proposes a structural classification of DFBSs similar to the CATH classification of protein folds, and it presents a study of secondary structure propensities of DFBSs and interaction preferences; (4) it introduces a systematic case-base reasoning approach to model on a large scale the 3D structures of protein complexes from existing structural DDIs. All these contributions have been made publicly available through a web server (http://kbdock.loria.fr)

Les rétrovirus sont des virus à ARN, enveloppés présents dans de nombreuses espèces animales de rente, chez les animaux de compagnie et chez l’homme. Une des particularités des rétrovirus concerne l’intégration du génome viral au sein du génome de la cellule infectée; cette intégration est réalisée par une enzyme virale, l’intégrase. Le projet de cette thèse vise à mieux comprendre le fonctionnement de cette enzyme notamment en identifiant des facteurs cellulaires interagissant avec celle-ci, facteurs qui pourraient être des agents favorisant le processus intégratif ou, au contraire, des agents restrictifs. Les intégrases de deux modèles de rétrovirus ont été utilisées dans cette étude : L’intégrase de RAV1, un rétrovirus exogène aviaire du genre des alpharétrovirus appartenant au sous-groupe A de la famille des ASLV. Cette enzyme virale est largement étudiée soit au niveau structural ou fonctionnel, mais les données concernant ses partenaires cellulaires sont rares et insuffisantes. La seconde intégrase est celle du PERV A/C, un rétrovirus endogène porcin du genre gammarétrovirus. Aucune information sur cette enzyme n’a été décrite jusqu’à présent. Ces deux enzymes, en fusion avec une étiquette 6xHistidine, ont été donc produites en bactérie, et en cellules d’insecte puis purifiées sur colonne d’affinité en FPLC. Leurs activités catalytiques ont été testées in vitro. Ces tests permettent de valoriser la capacité de l’intégrase à exercer principalement les 2 fonctions dont elle est responsable in vivo, le clivage en 3’ et le transfert de brins, et une activité qu’elle exerce exclusivement in vitro, la désintégration. Les protéines pures et actives ont ensuite servies à la vérification de leur interaction avec une protéine cellulaire, Brd2. La technique ‘Far western blot’ a ainsi permis de valider l’interaction entre l’intégrase de PERV et la protéine cellulaire, puis d’identifier les domaines de l’intégrase et de Brd2 impliqués dans cette interaction. A terme, l’identification de ce facteur cellulaire et la validation de son rôle dans le processus intégratif permettront de mieux comprendre ce processus particulier développé par les rétrovirus et pourront conduire au développement d’inhibiteurs dirigés contre cette interaction / A critical step for retroviral replication is the stable integration of the provirus genome into the genome of its host; this integration is realized by a viral enzyme, the integrase. The aim of this work was to better understand the functioning of the integrase, particularly, by identifying host factors that might interact with it, and which could be factors favoring the integration process or, restrictive factors. Therefore, we used two models of retroviral integrases: The integrase of RAV1, an alpharetrovirus belonging to the subgroup A of the family of ALSV. Although this viral enzyme is widely studied, still not enough data are available about its cellular cofactors. The second enzyme studied here is the integrase of PERV, a gammaretrovirus. No studies of either PERV integrase activities in vitro or of proteins interacting with this viral enzyme have been available until now. In the present study, we have expressed the PERV and ALSV integrases as fusion proteins with a 6xHistidine Tag in both Escherichia coli and insect SF9 cells. After that, we analysed their ability to mediate catalytic activities (3’-end processing, strand transfer and disintegration) in vitro. We also investigated the interaction of these two viral enzymes with the cellular protein Brd2, using the Far western blot method. Our results validate Brd2 as a cofactor of PERV integrase and point to the important role of particular domains of the PERV integrase and Brd2 in mediating the interaction. Finally, this study contibute to a better understanding of the precise interaction between cellular proteins and integrase, and may lead in the future to the development of protein-protein interaction inhibitors

Proteins have an indispensable role in the cell. They carry out a wide variety of structural, catalytic and signaling functions in all known biological systems. To perform their biological functions, proteins establish interactions with other bioorganic molecules including other proteins. Therefore, protein-protein interactions is one of the central topics in molecular biology. My thesis is devoted to three different topics in the field of protein-protein interactions. The first one focuses on solvent contribution to protein interfaces as it is an important component of protein complexes. The second topic discloses the structural and functional potential of fluorine's unique properties, which are attractive for protein design and engineering not feasible within the scope of canonical amino acids. The last part of this thesis is a study of the impact of charged amino acid residues within the hydrophobic interface of a coiled-coil system, which is one of the well-established model systems for protein-protein interactions studies. I. The majority of proteins interact in vivo in solution, thus studies of solvent impact on protein-protein interactions could be crucial for understanding many processes in the cell. However, though solvent is known to be very important for protein-protein interactions in terms of structure, dynamics and energetics, its effects are often disregarded in computational studies because a detailed solvent description requires complex and computationally demanding approaches. As a consequence, many protein residues, which establish water-mediated interactions, are neither considered in an interface definition. In the previous work carried out in our group the protein interfaces database (SCOWLP) has been developed. This database takes into account interfacial solvent and based on this classifies all interfacial protein residues of the PDB into three classes based on their interacting properties: dry (direct interaction), dual (direct and water-mediated interactions), and wet spots (residues interacting only through one water molecule). To define an interaction SCOWLP considers a donor–acceptor distance for hydrogen bonds of 3.2 Å, for salt bridges of 4 Å, and for van der Waals contacts the sum of the van der Waals radii of the interacting atoms. In previous studies of the group, statistical analysis of a non-redundant protein structure dataset showed that 40.1% of the interfacial residues participate in water-mediated interactions, and that 14.5% of the total residues in interfaces are wet spots. Moreover, wet spots have been shown to display similar characteristics to residues contacting water molecules in cores or cavities of proteins. The goals of this part of the thesis were: 1. to characterize the impact of solvent in protein-protein interactions 2. to elucidate possible effects of solvent inclusion into the correlated mutations approach for protein contacts prediction To study solvent impact on protein interfaces a molecular dynamics (MD) approach has been used. This part of the work is elaborated in section 2.1 of this thesis. We have characterized properties of water-mediated protein interactions at residue and solvent level. For this purpose, an MD analysis of 17 representative complexes from SH3 and immunoglobulin protein families has been performed. We have shown that the interfacial residues interacting through a single water molecule (wet spots) are energetically and dynamically very similar to other interfacial residues. At the same time, water molecules mediating protein interactions have been found to be significantly less mobile than surface solvent in terms of residence time. Calculated free energies indicate that these water molecules should significantly affect formation and stability of a protein-protein complex. The results obtained in this part of the work also suggest that water molecules in protein interfaces contribute to the conservation of protein interactions by allowing more sequence variability in the interacting partners, which has important implications for the use of the correlated mutations concept in protein interactions studies. This concept is based on the assumption that interacting protein residues co-evolve, so that a mutation in one of the interacting counterparts is compensated by a mutation in the other. The study presented in section 2.2 has been carried out to prove that an explicit introduction of solvent into the correlated mutations concept indeed yields qualitative improvement of existing approaches. For this, we have used the data on interfacial solvent obtained from the SCOWLP database (the whole PDB) to construct a “wet” similarity matrix. This matrix has been used for prediction of protein contacts together with a well-established “dry” matrix. We have analyzed two datasets containing 50 domains and 10 domain pairs, and have compared the results obtained by using several combinations of both “dry” and “wet” matrices. We have found that for predictions for both intra- and interdomain contacts the introduction of a combination of a “dry” and a “wet” similarity matrix improves the predictions in comparison to the “dry” one alone. Our analysis opens up the idea that the consideration of water may have an impact on the improvement of the contact predictions obtained by correlated mutations approaches. There are two principally novel aspects in this study in the context of the used correlated mutations methodology : i) the first introduction of solvent explicitly into the correlated mutations approach; ii) the use of the definition of protein-protein interfaces, which is essentially different from many other works in the field because of taking into account physico-chemical properties of amino acids and not being exclusively based on distance cut-offs. II. The second part of the thesis is focused on properties of fluorinated amino acids in protein environments. In general, non-canonical amino acids with newly designed side-chain functionalities are powerful tools that can be used to improve structural, catalytic, kinetic and thermodynamic properties of peptides and proteins, which otherwise are not feasible within the use of canonical amino acids. In this context fluorinated amino acids have increasingly gained in importance in protein chemistry because of fluorine's unique properties: high electronegativity and a small atomic size. Despite the wide use of fluorine in drug design, properties of fluorine in protein environments have not been yet extensively studied. The aims of this part of the dissertation were: 1. to analyze the basic properties of fluorinated amino acids such as electrostatic and geometric characteristics, hydrogen bonding abilities, hydration properties and conformational preferences (section 3.1) 2. to describe the behavior of fluorinated amino acids in systems emulating protein environments (section 3.2, section 3.3) First, to characterize fluorinated amino acids side chains we have used fluorinated ethane derivatives as their simplified models and applied a quantum mechanics approach. Properties such as charge distribution, dipole moments, volumes and size of the fluoromethylated groups within the model have been characterized. Hydrogen bonding properties of these groups have been compared with the groups typically presented in natural protein environments. We have shown that hydrogen and fluorine atoms within these fluoromethylated groups are weak hydrogen bond donors and acceptors. Nevertheless they should not be disregarded for applications in protein engineering. Then, we have implemented four fluorinated L-amino acids for the AMBER force field and characterized their conformational and hydration properties at the MD level. We have found that hydrophobicity of fluorinated side chains grows with the number of fluorine atoms and could be explained in terms of high electronegativity of fluorine atoms and spacial demand of fluorinated side-chains. These data on hydration agrees with the results obtained in the experimental work performed by our collaborators. We have rationally engineered systems that allow us to study fluorine properties and extract results that could be extrapolated to proteins. For this, we have emulated protein environments by introducing fluorinated amino acids into a parallel coiled-coil and enzyme-ligand chymotrypsin systems. The results on fluorination effect on coiled-coil dimerization and substrate affinities in the chymotrypsin active site obtained by MD, molecular docking and free energy calculations are in strong agreement with experimental data obtained by our collaborators. In particular, we have shown that fluorine content and position of fluorination can considerably change the polarity and steric properties of an amino acid side chain and, thus, can influence the properties that a fluorinated amino acid reveals within a native protein environment. III. Coiled-coils typically consist of two to five right-handed α-helices that wrap around each other to form a left-handed superhelix. The interface of two α-helices is usually represented by hydrophobic residues. However, the analysis of protein databases revealed that in natural occurring proteins up to 20% of these positions are populated by polar and charged residues. The impact of these residues on stability of coiled-coil system is not clear. MD simulations together with free energy calculations have been utilized to estimate favourable interaction partners for uncommon amino acids within the hydrophobic core of coiled-coils (Chapter 4). Based on these data, the best hits among binding partners for one strand of a coiled-coil bearing a charged amino acid in a central hydrophobic core position have been selected. Computational data have been in agreement with the results obtained by our collaborators, who applied phage display technology and CD spectroscopy. This combination of theoretical and experimental approaches allowed to get a deeper insight into the stability of the coiled-coil system. To conclude, this thesis widens existing concepts of protein structural biology in three areas of its current importance. We expand on the role of solvent in protein interfaces, which contributes to the knowledge of physico-chemical properties underlying protein-protein interactions. We develop a deeper insight into the understanding of the fluorine's impact upon its introduction into protein environments, which may assist in exploiting the full potential of fluorine's unique properties for applications in the field of protein engineering and drug design. Finally we investigate the mechanisms underlying coiled-coil system folding. The results presented in the thesis are of definite importance for possible applications (e.g. introduction of solvent explicitly into the scoring function) into protein folding, docking and rational design methods. The dissertation consists of four chapters: ● Chapter 1 contains an introduction to the topic of protein-protein interactions including basic concepts and an overview of the present state of research in the field. ● Chapter 2 focuses on the studies of the role of solvent in protein interfaces. ● Chapter 3 is devoted to the work on fluorinated amino acids in protein environments. ● Chapter 4 describes the study of coiled-coils folding properties. The experimental parts presented in Chapters 3 and 4 of this thesis have been performed by our collaborators at FU Berlin. Sections 2.1, 2.2, 3.1, 3.2 and Chapter 4 have been submitted/published in peer-reviewed international journals. Their organization follows a standard research article structure: Abstract, Introduction, Methodology, Results and discussion, and Conclusions. Section 3.3, though not published yet, is also organized in the same way. The literature references are summed up together at the end of the thesis to avoid redundancy within different chapters.