I have set for a challenging study that combined experimental and theoretical approaches in an attempt to resolve a role of small aminoacids in intermolecular interactions. First, I have proposed a hypothesis that described the interaction among individual aminoacids forming D helices of D1 and D2 proteins based on molecular dynamic simulations of a simplified model representing the reaction centre of photosystem II. Stability of the putative interhelical hydrogen bond network connecting D1 and D2 proteins was investigated experimentally with dynamic force spectroscopy using atomic force microscope. The results of both methods are in a full agreement with each other and reveal the key role of D1-Gly208 aminoacid in stability and functionality of photosystem II by providing milieu for weak interactions among three contact points at the cross of D helices: D1-Gly208 (O) and D2-Cys211 (O?), D1-Ser209 (O?) and D2-Ile204 (O), D1-Ser212 (O?) and D2-Gly207 (O). Mutation of the D1-Gly208 led to the increase in probability of the binding among the aforementioned aminoacids, undesirably strengthening the overall interactions among the proteins compromising photosynthetic capacity (D1-Ser208) or disabling of autotrophic growth (D1-Val208).

I have set for a challenging study that combined experimental and theoretical approaches in an attempt to resolve a role of small aminoacids in intermolecular interactions. First, I have proposed a hypothesis that described the interaction among individual aminoacids forming D helices of D1 and D2 proteins based on molecular dynamic simulations of a simplified model representing the reaction centre of photosystem II. Stability of the putative interhelical hydrogen bond network connecting D1 and D2 proteins was investigated experimentally with dynamic force spectroscopy using atomic force microscope. The results of both methods are in a full agreement with each other and reveal the key role of D1-Gly208 aminoacid in stability and functionality of photosystem II by providing milieu for weak interactions among three contact points at the cross of D helices: D1-Gly208 (O) and D2-Cys211 (O?), D1-Ser209 (O?) and D2-Ile204 (O), D1-Ser212 (O?) and D2-Gly207 (O). Mutation of the D1-Gly208 led to the increase in probability of the binding among the aforementioned aminoacids, undesirably strengthening the overall interactions among the proteins compromising photosynthetic capacity (D1-Ser208) or disabling of autotrophic growth (D1-Val208).

Orientadores: Maria de Fatima Sonati, Munir Salomão Skaf / Tese (doutorado) - Universidade Estadual de Campinas, Faculdade de Ciências Médicas / Made available in DSpace on 2018-08-23T23:13:39Z (GMT). No. of bitstreams: 1 Jorge_SusanElisabethDominguesCosta_D.pdf: 8714965 bytes, checksum: 3191d67be1e9be2f9782ce3483bcfd3a (MD5) Previous issue date: 2013 / Resumo: O resumo poderá ser visualizado no texto completo da tese digital / Abstract: The complete abstract is available with the full electronic document / Doutorado / Ciencias Biomedicas / Doutora em Ciências Médicas

Ce manuscrit présente une description théorique des propriétés de transport exceptionnelles des liquides dans les nanotubes de carbone (CNT). La perméabilité de ces canaux dépasse largement ce qui est prévu par les équations de l'hydrodynamique et la condition limite de non-glissement. Au cours des dernières années, plusieurs groupes ont effectué des expériences d'écoulement de liquides dans des membranes de CNT. Une perméabilité très supérieure à l'attente classique a été observée. Dans ce contexte, nous avons mené une étude exhaustive du frottement liquide/solide qui apparaît pendant l'écoulement d'un fluide dans un CNT, à l'aide de simulations de dynamique moléculaire. Le coefficient de frottement a été mesuré pour différents systèmes en utilisant plusieurs méthodes indépendantes. Les simulations ont montré que le coefficient de frottement était indépendant du confinement, mais qu'il dépendait considérablement de la courbure de la paroi. Pour l'eau dans un CNT, le coefficient de frottement diminue avec le rayon du tube. Nous avons ensuite établi une expression approchée du coefficient de frottement, qui le relie à des propriétés microscopiques de l'interface entre le liquide et la paroi. Cette expression reproduit la dépendance du coefficient de frottement avec la courbure, et permet de l'expliquer à partir des trois paramètres statiques suivants : la densité surfacique de l'eau, la rugosité de la paroi et la commensurabilité entre les structures de la paroi et de la première couche d'eau à l'interface. En résumé, notre étude a permis une compréhension détaillée du frottement de l'eau dans les CNT, qui explique l'origine de sa valeur extrêmement basse. / Within the scope of this thesis, a theoretical study of liquid flow in graphitic nanopores was performed. More precisely, a combination of numerical simulations and analytic approach was used to establish the special properties of carbon nanotubes for fluid transport: Molecular dynamics flow simulations of different liquids in carbon nanotubes exhibited flow velocities that are 1-3 orders of magnitude higher than predicted from the continuum hydrodynamics framework and the no-slip boundary condition. These results support previous experiments performed by several groups reporting exceptionally high flow rates for water in carbon nanotube membranes. The reason for this important flow enhancement with respect to the expectation was so far unclear. In this work, a careful investigation of the water/graphite friction coefficient which we identified as the crucial parameter for fast liquid transport in the considered systems was carried out. In simulations, the friction coefficient was found to be very sensitive to wall curvature: friction is independent of confinement for water between at graphene walls with zero curvature, while it increases with increasing negative curvature (water at the outside of the tube), and it decreases with increasing positive curvature (water inside the tube), eventually leading to quasi frictionless flow for water in a single file configuration in the smallest tubes. A similar behaviour was moreover found with several other liquids, such as alcohol, alcane and OMCTS. urthermore, a theoretical approximate expression for the friction coefficient is presented which predicts qualitatively and semi-quantitatively its curvature dependent behavior. Moreover, a deeper analysis of the simulations according to the proposed theoretical description shed light on the physical mechanisms at the origin of the ultra low liquid/solid friction in carbon nanotubes. In fine, it is due to their perfectly ordered molecular structure and their atomically smooth surface that carbon nanotubes are quasi-perfect liquid conductors compared to other membrane pores like, for example, nanochannels in amorphous silica. The newly gained understanding constitutes an important validation that carbon nanotubes operate as fast transporters of various liquids which makes them a promising option for different applications like energy conversion or filtration on the molecular level.

The binding of different proteins involved in processes such as transcription, replication and chromatin compaction to regions of the genome is regulated by the structure of DNA. Thus, DNA structure acts as the crucial link modulating evolutionary selection of the DNA sequence based on its own function, and the function of the proteins it encodes. The aim of this work is to examine the role of intrinsic, sequence-dependent structural variations vis-a -vis the protein-induced variations, in allowing DNA to assume geometries necessary for binding by proteins. For this purpose, we carried out analyses of datasets of X-ray crystal structures of free and protein-bound DNA, and molecular dynamics simulation studies of few free DNA structures and a protein-DNA complex. Each of the projects described below will appear as a separate chapter in the thesis. Analysis of X-ray crystal structure datasets Dataset of high-resolution X-ray crystal structures of free and protein-bound DNA This project was initiated with the aim of investigating the variation in A-and B-forms of DNA and the role they play in the binding of proteins. However, a survey of the existing literature indicated that the terms ‘A-DNA’ and ‘B-DNA’ were being used rather loosely and several different parameters at the local structural level were being used by various investigators to characterise these structures. Hence a systematic study was taken up to analyse all high-resolution free DNA structures comprising of sufficient number of contiguous Watson-Crick basepairs, irrespective of how they were classified by the existing databases. We also carried out a study of double-helical, Watson-Crick basepaired, free RNA structures for comparison. The structures in the RNA dataset were observed to rigidly assume the A-form and hence the average values of different parameters for that dataset were used to characterise the A-form. The analysis of free DNA and RNA structures was accompanied by an analysis of protein-bound DNA crystal structures. DNA structures bound to the helix-turn-helix motif in proteins were also analysed separately. The analysis of free DNA and RNA structures allowed us to pinpoint the parameters suitable for discriminating A-and B-forms of DNA at the local structural level. The results illustrated that the free DNA molecule, even in the crystalline state, samples a large amount of conformational space, encompassing both the A-and the B-forms. Most protein-bound DNA structures, including those with large, smooth curvature, were observed to assume the B-form. The A-form was observed to be limited to a small number of dinucleotide steps in DNA structures bound to the proteins belonging to a few specific families. Thus our study highlighted the structural versatility of B-form DNA, which allowed it to take up a range of global geometries to accommodate most DNA-binding protein motifs. Dataset of X-ray crystal structures of the nucleosome The study of high-resolution structures of free and protein-bound DNA was followed by an analysis of a dataset of X-ray crystal structures of the nucleosome, which is the fundamental repeating unit of the eukaryotic chromosome, and has been shown to play an important role in transcription regulation. Our results indicated that there is an ensemble of dinucleotide and trinucleotide level parameters that can give rise to similar global nucleosome structures. We therefore raise doubts about the use of the best resolved nucleosome structure as the template to calculate the energy required by putative nucleosome-forming sequences for adopting the nucleosome structure. Based on our results, we have proposed that the local and global level structural variability of DNA may act as a significant factor influencing the formation of nucleosomes in the vicinity of high-plasticity genes, and in determining the probability of binding by regulatory proteins. Molecular dynamics simulation studies of free and protein-bound DNA structures The analysis of crystal structure databases was complemented by molecular dynamics (MD) studies to investigate the dynamic evolution of the DNA structure in its free and protein-bound states. The following three simulation studies were carried out: Study to examine the biological relevance of the presence of 5-methyl group in thymine nucleotides An investigation of the biological relevance of the 5-methyl group in thymine nucleotides was carried out. For this purpose, comparison of molecular dynamics studies on structures with sequences d(CGCAAAUUUGCG)2and d(CGCAAATTTGCG)2was carried out. Our results showed that the presence of the thymine 5-methyl group was necessary for the A-tract to assume characteristic properties such as a narrow minor groove. It was also shown to modulate local level structural parameters and consequently, the curvature of the longer DNA fragment in which the A-tract was embedded. The analysis also provided possible explanation for the experimentally observed interaction of A-tracts with drugs and DNase-I in the presence and the absence of the thymine 5-methyl group. This project was the first of a series of MD studies, and hence several protocols were tested before finalising the correct protocol. Simulations were carried out using the Berendsen temperature equilibration scheme as well as the Langevin temperature equilibration scheme on both the structures. The Langevin temperature equilibration scheme was found to be unsuitable for nucleic acid simulations, as it caused long-term and possibly permanent disruption of the double-helical structure at the terminal and the neighbouring two positions in the sequence. The Berendsen temperature equilibration scheme was not observed to cause such disruptions. Simulations were also carried out on both structures, with or without initialising the initial ion positions. The position of minimum electrostatic potential, where AMBER8 placed the first counterion, was observed to act as a minimum energy trap from which the counterion could not escape even during the course of several nanoseconds of simulation. Hence, the actual simulations were carried out using the Berendsen temperature equilibration scheme, and after randomisation of initial ion positions. The results of protocol testing have been reported in an appendix. Study of DNA bending and curvature An analysis of DNA bending and curvature was carried out, by MD simulation on structures of three, ∼thirty basepair long sequences, namely, d(G-3(CA4T4G)-C)2, d(G-3(CT4A4G)-C)2and d(T-GACTA5T-GACTA6T-GACTA5T-G). For each molecule, snapshots belonging to a particular global geometry (linear, curved, bent in a particular direction etc.) were grouped together, and the average values of the dinucleotide step parameters for different groups were compared. It was observed that for all the three molecules, the average values for groups corresponding to different global geometries were within 1of each other, indicating that ensemble average values of dinucleotide level parameters are incapable of predicting the global geometry of a DNA molecule. Study of the TraR-Trabox complex The study on DNA bending and curvature was followed by simulations of a protein-DNA complex comprising of the bacterial quorum sensing transcription factor TraR with its promoter region known as Trabox. Simulations of a protein-free wild-type Trabox and a Trabox with two mutations in the spacer region were also carried out. Grouping of DNA snapshots in all the three simulations based on average values of dinucleotide parameters in the spacer region shows how selection of the ‘right’ DNA geometry by proteins works at several levels. The number of snapshots of free mutated Trabox assuming a geometry favourable for protein-binding in terms of average twist alone are less than one-fourth of the corresponding number for free wild-type Trabox. When one applies further selection criteria in terms of other parameters such as roll and slide, the number of mutated Trabox snapshots with a geometry favourable for protein-binding drops to less than 0.5%ofthe total number of MD snapshots. Thus our results highlight how sequence-dependent changes in the structrure of DNA regions, adjacent to those that directly hydrogen-bond to proteins, can also critically influence processes such as transcription. General Conclusion Overall, our results indicate that intrinsic, sequence-dependent structural variations in free B-DNA allow it to sample a large volume of the double-helical conformational space, and assume global geometries that can accomodate most DNA-binding proteins.

No description available.

Diblock-Copolymers (DBCs), created by covalently joining two chemically distinct polymer blocks, spontaneously form various nanoscale morphologies such as lamellae, cylinders, spheres, etc. due to the chemical incompatibility of its constituent blocks. This effect is called microphase separation in the literature. Because of this self-organizing property DBCs find applications in many areas e.g. in creating selective membranes, and in polymer based modern electronic devices like organic photovoltaics where the internal morphology plays an important role in determining the performance of the device. Many such modern devices are based on thin film technologies and uses copolymer nanocomposites as it exhibits advantageous electrical, optical, and mechanical properties. Also, DBC can direct the spatial distribution of nanoparticles (NPs) in the polymer matrix via microphase separation. Generally, two types of NPs are distinguished with respect to their monomer affinity: selective NPs which prefer one component of DBC, and non-selective NPs which interact equally with both components of DBC. In this work, using molecular dynamics simulations and analytical calculations, we explore the effect of adding both types of NP in the copolymer matrix considering a thin film (or confined) geometry. We consider a cylinder forming DBC melt confined by purely repulsive walls in slit geometry and study the behavior of the system upon adding non-selective NPs. Two models of non-selective interactions between the monomers and NPs are applied, i.e repulsive and weakly attractive interactions (athermal and thermal cases respectively). Spatial distribution of NPs in the copolymer matrix is sensitive to the NP-monomer interaction behavior. We focus on the thermal case and discuss, in particular, the following points: (1) role of diblock and polymer-wall interfaces, (2) spatial distribution of NPs, and (3) NP segregation and uptake behavior by the copolymer film. The uptake of NPs by the copolymer film in the thermal case displays a non-monotonic dependence on temperature which can be explained qualitatively using a mean-field model. In general, addition of non-selective NPs do not affect the copolymer morphology and the NPs are preferentially localized at the interface between microphase domains. Morphological transitions are observed when adding selective NPs to the copolymer matrix. By varying the amount of selective NPs and diblock composition we systematically explore the various structures formed by the nanocomposites under confinement and constructed the corresponding phase diagram in diblock composition and NP concentration. We also discuss the NP induced orientation transition of lamellar structure and study the stability of lamellar phases formed by the nanocomposites. To study the commensurability and wetting transition of horizontally oriented lamellar phase formed by the nanocomposites we have developed a mean field model based on the strong segregation theory. Our model predicts that it is possible to reduce the frustration in a film of fixed thickness by properly tuning the NP-monomer interaction strength. Furthermore, the model predicts a discontinuous transition between the non-wetted phase (where a dense NP layer is present in the polymer-substrate interface) and wetted phase (where the substrate is covered by polymers). Finally, we extend our study to non-equilibrium where we apply a shear flow field to copolymer thin films. Here, we study the flow behavior, lamellae deformation and change of pair-wise interaction energy, and macroscopic response like kinetic friction coefficient and viscosity of the copolymer thin film with and without NPs. / Lösungen von Diblock-Copolymeren (DBC), welche durch die kovalente Bindung zweier chemisch unterschiedlicher linearer Polymerblöcke entstehen, können spontan mikroskopische Strukturen ausbilden, welche je nach dem Grad der chemischen Kompatibiliät der Blöcke beispielsweise lamellen-, zylinder- oder kugelartige Formen zeigen. Dieses Phänomen wird meist als Mikrophasenseparation bezeichnet. Aufgrund dieser selbstorganisierenden Eigenschaft finden DBCs Anwendungen in vielen Bereichen der Forschung und der Industrie. Beispielsweise zur Erzeugung selektiver Membranen oder in moderner polymerbasierter Elektronik, wie organischen Solarzellen, wo die innere Struktur eine wichtige Rolle spielt um die Leistungsfähigkeit zu erhöhen. Viele moderne Geräte basieren auf der Technologie dünner Schichten und nutzen Copolymer-Nanokomposite um elektrische, optische oder mechanische Eigenschaften zu verbessern. In Folge der Mikrophasenseparation kann man mit Hilfe von DBC die räumliche Verteilung von Nanopartikeln (NP) in der Polymermatrix kontrollieren. Man unterscheidet im Allgemeinen zwischen zwei Arten von NP: selektive NP, welche eine der beiden Komponenten der DBC bevorzugen und nicht-selektive NP, welche mit beiden Komponenten gleichartig wechselwirken. In der vorliegenden Arbeit nutzen wir molekulardynamische Simulationen und analytische Rechnungen um den Eigenschaften zu studieren, welche eine Zugabe von selektiven und nicht-selektiven NP auf eine dünnschichtige Copolymermatrix hat. Wir betrachten eine zylinderformende Schmelze aus DBC, welche in einem dünnen Film, zwischen zwei harten Wänden eingeschränkt ist, und untersuchen das Verhalten des Systems unter Zugabe nicht-selektiver NP. Zwei Modelle nicht-selektiver Wechselwirkungen werden angenommen: ausschließlich repulsive (athermische) Wechselwirkungen und schwach anziehende (thermische) Wechselwirkungen. Die räumliche Verteilung der NP ist abhängig von dem jeweiligen Wechselwirkungsverhalten. Wir konzentrieren uns hierbei auf den thermischen Fall und diskutieren speziell folgende Schwerpunkte: (1.) die Rolle der sich ausbildenden Grenzschichten, (2.) die räumliche Verteilung der NP und (3.) die Abscheidung der NP, sowie die Aufnahmefähigkeit derselben durch die Polymermatrix. Im thermische Fall zeigt die Aufnahme der NP durch die Copolymerschicht eine nicht-monotone Abhängigkeit von der Temperatur, was mit Hilfe eines Mean-Field Modells erklärt werden kann. Die Zugabe nicht-selektiver NP hat keinen Einfluss auf die Struktur der Copolymermatrix und die NP werden vorzugsweise an der Grenzschicht der jeweiligen Mikrophasen gefunden. Im Gegensatz dazu kann man durch die Zugabe selektiver NP eine Strukturveränderung in der Copolymermatrix feststellen. Durch Veränderung der Menge der NP und der Zusammensetzung der DBC können wir systematisch unterschiedliche Strukturen des räumlich eingeschränkten Nanokomposits erzeugen und ein entsprechendes Phasendiagram bezüglich der NP Konzentration und der DBC Zusammensetzung erstellen. Wir untersuchen auch die durch NP induzierte Orientierung der Lamellenstruktur und analysieren ihre Stabilität. Um den sogenannten Kommensurabilitäts- und Benetzungsübergang in horizontal orientierten Lamellenstrukturen zu untersuchen haben wir ein Mean-Field Modell entwickelt, welches auf der Annahme der 'starken Segregation' basiert. Unser Modell macht die Vorhersage, dass es möglich ist die Frustration in einem Kompositfilm zu reduzieren, indem man die NP-Monomer-Wechselwirkung entsprechend anpasst. Zusätzlich sagt das Modell einen diskontinuierlichen Übergang zwischen der unbenetzten Phase (Ausbildung einer dichten NP Konzentration an der Polymer-Substrat Grenzschicht) und der benetzten Phase (das Substrat ist ausschließlich vom Polymerkomposit bedeckt) voraus. Abschließend weiten wir unsere Untersuchungen auf Nicht-Gleichgewichtszustände aus und induzieren durch Scherung der Substratwände einen Strömungprofil im Kompositfilm. Dabei analysieren wir das Strömungsverhalten, die Lamellendeformation und die Änderung der paarweisen Wechselwirkungsenergie. Wir untersuchen auch makroskopische Größen, wie den kinetischen Reibungskoeffizienten und die Viskosität, je in An- und Abwesenheit von Nanopartikeln.

The dissociation of fast hydrogen and nitrogen molecular ions with kinetic energies ranging from 200 to 2000 eV/atom is studied for grazing collisions with various fcc metal surfaces. Within this energy range, the dissociation is either caused by electron capture into antibonding molecular states or by vibrational and rotational excitation. In contrast to hydrogen, nitrogen is chemically inert and interacts mainly elastically with the surfaces; thus there is no dissociation via electron capture. The processes of vibrational and rotational excitation are simulated using a molecular dynamics simulation with interaction potentials based on density functional theory. The comparison with the data obtained from Time-Of-Flight experiments reveals that an additional electronic effect has to be taken into account: The intramolecular bond of the molecules is softened due to electronic screening during the interaction with the surface. Hence, the softened molecules are more likely to dissociate through elastic collisions with surface atoms. The dissociation of hydrogen at low energies on metallic surfaces is dominated by electronic processes. An analysis of the kinetic energy distributions of the scattered dissociation products reveals information about the energy which is released during the dissociation process. The model of electronically induced dissociation is clearly confirmed by this method. However, an increasing contribution of additional mechanical processes becomes apparent at higher energies.

Les interactions protéine-protéine sont souvent contrôlées par de petits domaines protéiques qui régulent les chemins de signalisation au sein des cellules eucaryotes. Les domaines PDZ sont parmi les domaines les plus répandus et les plus étudiés. Ils reconnaissent spécifiquement les 4 à 10 acides aminés C-terminaux de leurs partenaires. Tiam1 est un facteur d'échange de GTP de la protéine Rac1 qui contrôle la migration et la prolifération cellulaire et dont le domaine PDZ lie les protéines Syndecan-1 (Sdc1), Caspr4 et Neurexine. Des petits peptides ou des molécules peptidomimétiques peuvent potentiellement inhiber ou moduler son activité et être utilisés à des fins thérapeutiques. Nous avons appliqué des approches de dessin computationnel de protéine (CPD) et de calcul d'énergie libre par simulations dynamique moléculaire (DM) pour comprendre et modifier sa spécificité. Le CPD utilise un modèle structural et une fonction d'énergie pour explorer l'espace des séquences et des structures et identifier des variants protéiques ou peptidiques stables et fonctionnels. Nous avons utilisé le programme de CPD Proteus, développé au laboratoire, pour redessiner entièrement le domaine PDZ de Tiam1. Les séquences générées sont similaires à celles des domaines PDZ naturels, avec des scores de similarité et de reconnaissance de pli comparables au programme Rosetta, un outil de CPD très utilisé. Des séquences contenant environ 60 positions mutées sur 90, ont été testées par simulations de DM et des mesures biophysiques. Quatre des cinq séquences testées expérimentalement (par nos collaborateurs) montrent un dépliement réversible autour de 50°C. Proteus a également déterminer correctement la spécificité de la liaison de quelques variants protéiques et peptidiques. Pour étudier plus finement la spécificité, nous avons paramétré un modèle d'énergie libre semi-empirique de Poisson-Boltzmann ayant la forme d'une énergie linéaire d'interaction, ou PB/LIE, appliqué à des conformations issues de simulations de DM en solvant explicite de complexes PDZ:peptide. Avec trois paramètres ajustables, le modèle reproduit correctement les affinités expérimentales de 41 variants, avec une erreur moyenne absolue de 0,4~kcal/mol, et donne des prédictions pour 10 nouveaux variants. Le modèle PB/LIE a ensuite comparé à la méthode non-empirique de calcul d'énergie libre par simulations alchimiques, qui n'a pas de paramètre ajustable et qui prédit correctement l'affinité de 12 complexes Tiam1:peptide. Ces outils et les résultats obtenus devraient nous permettre d'identifier des peptides inhibiteurs et auront d'importantes retombées pour l'ingénierie des interactions PDZ:peptide. / Small protein domains often direct protein-protein interactions and regulate eukaryotic signalling pathways. PDZ domains are among the most widespread and best-studied. They specifically recognize the 4-10 C-terminal amino acids of target proteins. Tiam1 is a Rac GTP exchange factor that helps control cellmigration and proliferation and whose PDZ domain binds the proteins syndecan-1 (Sdc1), Caspr4, and Neurexin. Short peptides and peptidomimetics can potentially inhibit or modulate its action and act as bioreagents or therapeutics. We used computational protein design (CPD) and molecular dynamics (MD) free energy simulations to understand and engineer its peptide specificity. CPD uses a structural model and an energy function to explore the space of sequences and structures and identify stable and functional protein or peptide variants. We used our in-house Proteus CPD package to completely redesign the Tiam1 PDZ domain. The designed sequences were similar to natural PDZ domains, with similarity and fold recognition scores comarable to the widely-used Rosetta CPD package. Selected sequences, containing around 60 mutated positions out of 90, were tested by microsecond MD simulations and biophysical experiments. Four of five sequences tested experimentally (by our collaborators) displayed reversible unfolding around 50°C. Proteus also accurately scored the binding specificity of several protein and peptide variants. As a more refined model for specificity, we parameterized a semi-empirical free energy model of the Poisson-Boltzmann Linear Interaction Energy or PB/LIE form, which scores conformations extracted from explicit solvent MD simulations of PDZ:peptide complexes. With three adjustable parameters, the model accurately reproduced the experimental binding affinities of 41 variants, with a mean unsigned error of just 0.4 kcal/mol, andgave predictions for 10 new variants. The PB/LIE model was tested further by comparing to non-empirical, alchemical, MD free energy simulations, which have no adjustable parameters and were found to give chemical accuracy for 12 Tiam1:peptide complexes. The tools and insights obtained should help discover new tight binding peptides or peptidomimetics and have broad implications for engineering PDZ:peptide interactions.

Les mitochondries sont des organites dynamiques dont la morphologie dépend de l’équilibre fusion/fission de leurs membranes. Ce processus essentiel à la survie cellulaire est nommé dynamique mitochondriale et sa dérégulation est associée à des troubles neurologiques. Cependant les mécanismes précis régissant la dynamique mitochondriale ne sont pas élucidés. Cette thèse porte sur la protéine Fzo1p, une grande GTPase de la superfamille des Dynamin-related-Protein. C’est un élément clé impliqué dans la fusion mitochondriale de la membrane externe de la levure. Sa structure et sa dynamique ont été étudiées par modélisation et simulations de dynamiques moléculaires tout-atome dans une bicouche lipidique solvatée. Le modèle structural obtenu tient compte de données expérimentales, de template structuraux, et de modèles ab initio du domaine transmembranaire de Fzo1p. Ce modèle a été validé expérimentalement par mutagenèse dirigée. Des permutations de charges ont confirmé des ponts salins à longue distance prédits dans le modèle. En outre, des mutations ont montré que les domaines coiled-coil de Fzo1p, contrairement à sa partie N-terminale, sont indispensables à sa fonction. L’ensemble des résultats expérimentaux et in silico met en évidence l’implication des domaines charnières dans le changement conformationnel de Fzo1p, ainsi que des résidus critiques affectant sa stabilité. Les précisions atomiques obtenues sur l’interaction de Fzo1p avec le GDP permet de formuler des hypothèses sur le mécanisme moléculaire de la catalyse du GTP pour la fusion membranaire; voire à la compréhension de la dynamique mitochondriale. / Mitochondria are dynamic organelles whose morphology is determined by fusion and fission of their membranes. This essential process is known as mitochondrial dynamics. Defects in mitochondrial dynamics are associated with neurological disorders making the investigation of physiological relevance. However, the precise sequence of events that lead mitochondrial dynamics are still not well characterised. Fzo1p, a large GTPase of the Dynamin-Related Proteins superfamily, is a key component in mitochondrial outer membrane fusion in yeast. During this PhD project I built a model of the protein Fzo1p. The structure and dynamics of the model was investigated through molecular modelling and all-atom molecular dynamics simulation in a fully hydrated lipid bilayer environment. The Fzo1p structural model integrates information from several template structures, experimental knowledge, as well as ab initio models of the transmembrane segments. The model is validated experimentally through directed mutagenesis, for instance charge-swap mutations confirm predicted long-distance salt bridges. A series of mutants indicate that coiled-coil domains are required for protein function at variance with its N-terminal region. Overall, the experimental and in silico approaches pinpoint the hinge domains involved in the putative conformational change and identifies critical residues affecting protein stability. Finally, key Fzo1p-GDP interactions provide insights about the molecular mechanism of membrane fusion catalysis. The model provides insight on atomic level and proposes a structure that will be instructional to understanding mitochondrial membrane fusion.