<div>Ebolavirus, a genus of filoviruses, are responsible for outbreaks that cause up to 90% fatality, including the recent outbreak in West Africa that has resulted in over 28,603 reported cases and 11,301 deaths according to the WHO. Inhibitors of Vacuolar-ATPase (V-ATPase), a key protein complex that is responsible for endosomal acidification and represents a unique method to block this common viral pathway. V-ATPase inhibitors have previously been explored as therapies for many diseases but have failed due to high toxicity. Diphyllin is a natural, arylnaphthalene lignan that represents a novel structural class of V-ATPase inhibitors with a greater selectivity index than previous V-ATPase inhibitors. Diphyllin has shown promising anti-tumor and anti-osteoclast activity, as well as strong anti-viral activity against Influenza and Dengue viruses. </div><div>Herein, novel modifications of the lactone and phenol functional groups of diphyllin were explored for the ability to enhance the potency or therapeutic selectivity of the diphyllin core. Four initial sets of derivatives were synthesized and assayed for activity against ebolavirus infection, inhibition of cellular endosomal acidification, cytotoxicity and biochemical inhibition of isolated V-ATPase. Modification of diphyllin’s lactone functional group reduced both activity and selectivity, while alkylation of the phenol groups significantly enhanced activity. The incorporation of basic heterocycles to the alkyl group created an alkylamino series of derivatives that exhibited significantly improved therapeutic selectivity compared to diphyllin. Further investigation of the alkylamino class indicated that they retained activity against Marburgvirus infection, a filovirus related to Ebolavirus. Alkylamino derivatives inhibited ebolavirus infection of human macrophages at low micromolar levels with no apparent cytotoxicity.</div><div>Further investigation of the alkylamino class of diphyllin derivatives was conducted to determine if potency and/or therapeutic selectivity could be optimized. The addition of a 1-methylpiperazine moiety to the end of the alkyl chain improved potency 1260-fold over diphyllin, though therapeutic selectivity was not improved. The modification of the alkylamino linker to an acetamide eliminated cytotoxicity but decreased derivative activity against V-ATPase activity. To evaluate if the cytotoxicity evidenced by the alkylamino derivatives was evidenced in organisms, the derivative toxicity was assessed in zebrafish and mouse models. Derivatives displayed toxicity in a zebrafish developmental model but were all at least 10-fold less toxic than the known V-ATPase inhibitor bafilomycin A1. Three derivatives were well tolerated in CD-1 mice when administered at therapeutically relevant concentrations and caused no abnormal changes in their blood chemistry. Overall, these results demonstrate that the alkylamino and acetamide diphyllin phenol derivatives should be further studied as therapies for ebolavirus infection in addition to other V-ATPase mediated diseases.</div><div><br></div>

<p>Since 1981, HIV/AIDS has affected over 70 million individuals worldwide. Due to the incorporation of Combination Antiretroviral Therapy (cART), this deadly virus has now become a manageable chronic illness with a reduction in mortality and morbidity rates. Combination therapy targets multiple stages of the HIV replication cycle including fusion, entry, reverse transcription, integration, and maturation. The HIV-1 protease enzyme is responsible for cleavage and processing of viral polyproteins into mature enzymes and is a common therapeutic target for inhibition of HIV. To date, there have been many protease inhibitors approved by the FDA and introduced into the market. However, mutations within the protease enzyme has rendered some of these inhibitors ineffective. This has led to an ever-growing need to develop novel protease inhibitors to combat drug resistance through mutations. Described herein is the design, synthesis, and biological evaluation of HIV-1 protease inhibitors featuring a novel hexahydropyrrolofuran (HPF) bicyclic scaffold as a P₂ ligand to target binding interactions with Asp29 and Asp30. The HPF ligand provides a molecular handle that allows for further structure-activity discoveries within the enzyme. The HIV-1 protease inhibitors discussed feature carbamate, carboxamide, and sulfonamide derivatives which displayed good to excellent activity.</p>

Title: Multifunctional biomolecular assemblies for parallelized analysis of biomolecular interactions Author: Markéta Bocková Department / Institute: Institute of Physics, Charles University Supervisor of the doctoral thesis: Prof. Jiří Homola, Ph.D., DSc., Institute of Photonics and Electronics, The Czech Academy of Sciences Abstract: Surface plasmon resonance (SPR) biosensors represent the most advanced optical method for the direct, real-time monitoring of biomolecular interactions without the need for labelling. This doctoral thesis aims to advance the SPR biosensor method and to expand its utility in the investigation of biomolecular interactions. This encompasses activities on two major fronts of SPR biosensor research - immobilization methods and biosensing methodologies. Methods for the immobilization of biomolecules were researched with the aim of enabling the immobilization of a broad range of biomolecules on the SPR biosensor surface in a spatially controlled manner. The development of novel biosensing methodologies was pursued in order to address the current limitations of SPR biosensors associated with non-specific adsorption and limited analyte transport, and thus to improve the accuracy and robustness of SPR biosensor measurements. Finally, advances in the development of immobilization...

Neuerliche Fortschritte im Verständnis biomolekularer Motoren rücken ihre Anwendung als Nanomaschinen in den Bereich des Möglichen. So könnten sie zum Beispiel als Nanoroboter arbeiten, um in molekularen Fabriken kleine – aber dennoch komplizierte – Strukturen auf winzigen Förderbändern herzustellen, um Netzwerke molekularer Nanodrähte und Transistoren für elektronische Anwendungen zu assemblieren oder sie könnten in adaptiven Materialien patrouillieren und diese, wenn nötig, reparieren. In diesem Sinne besitzen biomolekulare Motoren das Potenzial, die Basis für die Konstruktion, Strukturierung und Wartung nanoskaliger Materialien zu bilden. / Recent advances in understanding how biomolecular motors work have raised the possibility that they might find applications as nanomachines. For example, they could be used as molecule- sized robots that work in molecular factories where small, but intricate structures are made on tiny assembly lines, that construct networks of molecular conductors and transistors for use as electrical circuits, or that continually patrol inside “adaptive” materials and repair them when necessary. Thus biomolecular motors could form the basis of bottom-up approaches for constructing, active structuring and maintenance at the nanometer scale.

<p>Approximately 6.3 million bone fractures occur annually in the USA, resulting in considerable morbidity, deterioration in quality of life, loss of productivity and wages, and sometimes death (e.g. hip fractures). Although anabolic and antiresorptive agents have been introduced for treatment of osteoporosis, no systemically-administered drug has been developed to accelerate the fracture healing process. To address this need, we have undertaken to target a bone anabolic agent selectively to fracture surfaces in order to concentrate the drug’s healing power directly on the fracture site. We report here that conjugation of dasatinib to a bone fracture-homing oligopeptide via a releasable linker reduces fractured femur healing times in mice by ~60% without causing overt off-target toxicity or remodeling of nontraumatized bones. Thus, achievement of healthy bone density, normal bone volume, and healthy bone mechanical properties at the fracture site is realized after only 3-4 weeks in dasatinib-targeted mice, but requires ~8 weeks in PBS-treated controls. Moreover, optimizations have been implemented to the dosing regimen and releasing mechanisms of this targeted-dasatinib therapy, which has enabled us to cut the total doses by half, reduce the risk of premature release in circulation, and still improve upon the therapeutic efficacy. These efforts might reduce the burden associated with frequent doses on patients with broken bones and lower potential toxicity brought by drug degradation in the blood stream. In addition to dasatinib, a few other small molecules have also been targeted to fracture surfaces and identified as prospective therapeutic agents for the acceleration of fracture repair. In conclusion, in this dissertation, we have successfully targeted dasatinib to bone fracture surfaces, which can significantly accelerate the healing process at dasatinib concentrations that are known to be safe in oncological applications. A modular synthetic method has also been developed to allow for easy conversion of a bone-anabolic warhead into a fracture-targeted version for improved fracture repair.</p><p></p>

Cells use energy to maintain order, as living systems are inherently non-equilibrium. Or- der in the cytoplasm is achieved by compartmentalization. One type of compartment that gained interest in recent years is membraneless organelles (MLOs). Observations of the liquid-like properties of MLOs led to their interpretation in analogy to Liquid-Liquid Phase Separation (LLPS). However, LLPS alone implies a passive closed system that tends towards equilibrium, which is incompatible with the physical nature of the cell. It is unclear then what non-equilibrium interactions give rise to the dynamics of MLOs in the cell. We sought to decipher the regulatory interactions that give rise to active condensation in the actomyosin cortex of C. elegans. The components of the actomyosin cortex, F- actin and its branching nucleation module Arp2/3 and N-WASP (WSP-1 in C. elegans) have been described as a phase separated system in previous reports. In vitro, phase separated N-WASP compartments do not have the non-equilibrium growth and disas- sembly dynamics observed in the multicomponent clusters in vivo. Therefore, our goal is to examine WSP-1, Arp2/3 and F-actin interactions in the endogenous context. We chose the stage in which the quiescent oocyte cortex becomes actively contractile. During the transition out of quiescence, we observed transient WSP-1 Arp2/3 F-actin puncta that assemble and disassemble. To capture growth dynamics for all puncta, we devel- oped a novel phase portrait analysis tool. The phase portrait allows us to simultaneously study puncta growth and disassembly rates as a function of internal composition. The growth rate dependence on internal composition reflects the non-trivial changes to nu- cleation profiles that accompany condensation in active, open, multi-component systems. We observed superlinear WSP-1 growth rates consistent with condensation. Further, we identified the in vivo equivalent of a nucleation barrier for WSP-1 condensation. The in vivo nucleation barrier increases with branching F-actin reaction, which tunes con- densation. Correspondingly, the reactive components WSP-1 and Arp2/3 are important for condensate dynamics. Combining condensation and the branching reaction, we for- mulated a coarse-grained model which captures non-equilibrium condensate dynamics. Altogether, our results showed that WSP-1 grows like condensation, and its growth is steered away from equilibrium by Arp2/3 mediated branching reaction. In summary, combining high-resolution imaging, quantitative analysis and theory, we identified the interactions that could explain non-equilibrium condensation in the acto- myosin cortex. The living dynamics that arise from the interplay between condensation and reaction. The interplay between physical processes (like condensation) and biological regulation (such as reactions) may be a common organizing principle behind MLO for- mation, as well as other non-equilibrium processes in the cell. The methods and concepts developed in this work hold the promise to deepen our understanding of how living cells regulate their dynamic organization, in order to maintain themselves in a non-equilibrium ordered state.:1 Introduction 1 1.1 Evolving concepts of cellular organization 1 1.2 Condensation of biomolecules 3 1.2.1 Terminology for biomolecular condensates 5 1.2.2 Technical considerations for identifying liquid-like properties and LLPS 7 1.2.3 Thermodynamics of condensation 10 1.2.4 The problem of an equilibrium description of living systems 13 1.2.5 Towards active condensation 14 1.3 Actomyosin cortex self-organization 16 1.3.1 F-actin treadmilling and nucleation 17 1.3.2 N-WASP and Arp2/3 regulation 18 1.3.3 Multivalent interactions in condensation of transmembrane receptors and actin regulators 22 1.3.4 Cortex activation in C. elegans 23 2 Aims 25 3 Results 26 3.1 C. elegans cortical activation begins at fertilization 26 3.1.1 C. elegans oocytes as an ex utero model for cortex self-organization 27 3.2 WSP-1, Arp2/3 and F-actin form dynamic multicomponent phases 32 3.2.1 Capping proteins outcompete Formin in WSP-1 Arp2/3 puncta preventing F-actin elongation 32 3.2.2 WSP-1 and Arp2/3 are required for punctate F-actin formation and dynamics 34 3.2.3 Summary of the characterization of cortical activation 34 3.3 Establishment of systematic phase portrait analysis for multicomponent clusters 36 3.3.1 Non-equilibrium features of the multicomponent puncta 36 3.3.2 Recording intensity traces of multicomponent cluster over time 37 3.3.3 Probability flux of composition in the phase portrait show a closed cycle 38 3.3.4 WSP-1 F-actin puncta have a preferred joint concentration 38 3.3.5 The phase portrait is robust to cell-to-cell noise 41 3.3.6 Choosing the appropriate bin size 41 3.4 Existence of a tuned critical size and signatures of active condensation 45 3.4.1 Growth rate dependence on internal composition 45 3.4.2 Stoichiometric growth laws of WSP-1 F-actin clusters 47 3.4.3 Estimation of WSP-1 cluster critical size in vivo 47 3.4.4 Theoretical description of WSP-1 and F-actin interactions in regulating puncta dynamics 48 3.4.5 Summary of 2D phase portrait findings 52 3.5 Towards three dimensional phase portrait analysis of the reaction network 54 3.6 Initial assessment of the compartment’s external environment 54 3.7 Identification of modulators of puncta dynamics 56 3.7.1 CDC-42 controls cortical levels of WSP-1 56 3.7.2 RHO-1 and Formin CYK-1 are not involved in WSP-1 F-actin condensate dynamics 58 3.7.3 WSP-1 and Arp2/3 dynamics are independent of NCK-1 and VAB-1 58 3.7.4 Arp2/3 regulates condensate dynamics 60 3.8 Summary of perturbations 63 4 Conclusions and outlook 64 4.1 Concluding remarks 64 4.2 Discussion 66 4.3 Future directions 67 4.3.1 Realizing the full potential of the phase portraits in identifying biochemical interactions 67 4.3.2 Resolving the ultrastructure of condensates . 70 4.3.3 Further investigation of the biological function 71 4.3.4 Applying full-dynamic data acquisition to other membraneless organelles 71 5 Materials and Methods 72 5.1 C.elegans maintenance and strains 72 5.2 Sample preparation 72 5.2.1 In utero imaging 72 5.2.2 Oocyte imaging 73 5.2.3 C.elegans HaloTag staining 73 5.2.4 Oocyte chemical inhibitor treatments 73 5.3 RNAi Feeding 73 5.4 Microscopy 73 5.4.1 Spinning disk microscopy 73 5.4.2 SIM-TIRF microscopy 74 5.5 TIRF microscopy 74 5.6 Phase portrait analysis pipeline 74 5.7 Kymographs 76 / Zellen verbrauchen Energie, um Ordnung aufrechtzuerhalten, da lebende Systeme von Natur aus ungleichgewichtig sind. Ordnung im Zytoplasma wird durch Kompartimen- tierung erreicht. Eine Art von Kompartiment, das in den letzten Jahren an Interesse gewonnen hat, sind membranlose Organellen (engl.: membraneless organelles, MLOs). Beobachtungen der flu ̈ssigkeits ̈ahnlichen Eigenschaften dieser MLOs fu ̈hrten zu ihrer In- terpretation in Analogie zur Flu ̈ssig-Flu ̈ssig-Phasentrennung (engl.: liquid-liquid phase separation, LLPS). Die LLPS allein impliziert jedoch ein passives geschlossenes System, das zum Gleichgewicht neigt und mit der physikalischen Natur der Zelle nicht kompatibel ist. Es war bisher nicht bekannt, welche Ungleichgewichtswechselwirkungen die Dynamik von MLOs in der Zelle hervorrufen. Wir wollten die regulatorischen Wechselwirkungen entschlu ̈sseln, die zu aktiver Konden- sation im Aktomyosin-Kortex von C. elegans fu ̈hren. Die Komponenten des Aktomyosin- Kortex, F-Aktin und seines verzweigten Nukleationsmoduls Arp2/3 und N-WASP (WSP- 1 in C. elegans) wurden in fru ̈heren Studien als phasengetrenntes System beschrieben. In vitro weisen phasengetrennte N-WASP-Kompartimente allerdings nicht dieselben un- gleichgewichtigen Wachstums- und Zerlegungsdynamiken auf, die in kultivierten Zellen beobachtet werden. Daher wollten wir die Wechselwirkungen zwischen WSP-1, Arp2/3 und F-Aktin im Kontext des Fadenwurms C. elegans untersuchen. Wir haben das C.elegans Lebenstadium gew ̈ahlt, in dem die ruhende Eizellenrinde aktiv kontraktil wird. Wa ̈hrend des U ̈bergangs aus der ruhigen in die aktive Periode konnten wir voru ̈bergehende WSP- 1 Arp2/3 F-Aktin-Puncta beobachten, die sich zusammensetzen und zerlegen. Um die Wachstumsdynamik fu ̈r alle Puncta zu erfassen, haben wir ein neuartiges Tool zur Anal- yse von Phasenportr ̈ats entwickelt. Das Phasenportr ̈at ermo ̈glicht es uns, gleichzeitig die Wachstums- und die Zerlegungsraten von Puncta in Abha ̈ngigkeit der inneren Zusam- mensetzung zu messen. Die Abha ̈ngigkeit der Wachstumsrate von der inneren Zusam- mensetzung spiegelt die nicht trivialen A ̈nderungen der Nukleationsprofile wider, die mit der Kondensation in aktiven, offenen Mehrkomponentensystemen einhergehen. Wir kon- nten superlineare WSP-1-Wachstumsraten beobachten, die mit der Kondensation u ̈bere- instimmen. Ferner konnten wir das In-vivo-A ̈quivalent einer Nukleationsbarriere fu ̈r die WSP-1-Kondensation identifizieren. Die In-vivo-Nukleationsbarriere nimmt mit der verzweigten F-Actin-Reaktion zu, die die Kondensation reguliert. Dementsprechend sind die reaktiven Komponenten WSP-1 und Arp2/3 wichtig fu ̈r die Dynamik des Konden- sats. Wir haben die Kondensations- und Verzweigungsreaktionen kombiniert, um damit ein grobko ̈rniges Modell zu formulieren, das die Ungleichgewichtskondensationsdynamik erfasst. Insgesamt haben unsere Ergebnisse gezeigt, dass WSP-1 kondensiert und diese Kondensation durch Arp2/3-vermittelte Verzweigungsreaktionen aus dem Gleichgewicht gebracht wird. Zusammenfassend konnten wir durch Kombination von hochauflo ̈sender Bildgebung, quan- titativer Analyse und Theorie die Wechselwirkungen identifizieren, die die Ungleichgewicht- skondensation im Aktomyosin-Kortex erkla ̈ren ko ̈nnten. Die Dynamik im lebendem Sys- tem ergibt sich aus dem Zusammenspiel von Kondensation und Reaktion. Die Interaktion zwischen physikalischen Prozessen (wie Kondensation) und biologischen Regulationen (wie Reaktionen) kann ein gemeinsames Organisationsprinzip hinter der MLO-Bildung sowie anderen Ungleichgewichtsprozessen in der Zelle sein. Die in dieser Arbeit entwickel- ten Methoden und Konzepte k ̈onnen daher helfen, unser Versta ̈ndnis daru ̈ber zu vertiefen, wie lebende Zellen ihre r ̈aumlich-zeitlichen Proteinverteilungen dynamisch regulieren, um sich in einem ungleichgewichtigen, geordneten Zustand zu halten.:1 Introduction 1 1.1 Evolving concepts of cellular organization 1 1.2 Condensation of biomolecules 3 1.2.1 Terminology for biomolecular condensates 5 1.2.2 Technical considerations for identifying liquid-like properties and LLPS 7 1.2.3 Thermodynamics of condensation 10 1.2.4 The problem of an equilibrium description of living systems 13 1.2.5 Towards active condensation 14 1.3 Actomyosin cortex self-organization 16 1.3.1 F-actin treadmilling and nucleation 17 1.3.2 N-WASP and Arp2/3 regulation 18 1.3.3 Multivalent interactions in condensation of transmembrane receptors and actin regulators 22 1.3.4 Cortex activation in C. elegans 23 2 Aims 25 3 Results 26 3.1 C. elegans cortical activation begins at fertilization 26 3.1.1 C. elegans oocytes as an ex utero model for cortex self-organization 27 3.2 WSP-1, Arp2/3 and F-actin form dynamic multicomponent phases 32 3.2.1 Capping proteins outcompete Formin in WSP-1 Arp2/3 puncta preventing F-actin elongation 32 3.2.2 WSP-1 and Arp2/3 are required for punctate F-actin formation and dynamics 34 3.2.3 Summary of the characterization of cortical activation 34 3.3 Establishment of systematic phase portrait analysis for multicomponent clusters 36 3.3.1 Non-equilibrium features of the multicomponent puncta 36 3.3.2 Recording intensity traces of multicomponent cluster over time 37 3.3.3 Probability flux of composition in the phase portrait show a closed cycle 38 3.3.4 WSP-1 F-actin puncta have a preferred joint concentration 38 3.3.5 The phase portrait is robust to cell-to-cell noise 41 3.3.6 Choosing the appropriate bin size 41 3.4 Existence of a tuned critical size and signatures of active condensation 45 3.4.1 Growth rate dependence on internal composition 45 3.4.2 Stoichiometric growth laws of WSP-1 F-actin clusters 47 3.4.3 Estimation of WSP-1 cluster critical size in vivo 47 3.4.4 Theoretical description of WSP-1 and F-actin interactions in regulating puncta dynamics 48 3.4.5 Summary of 2D phase portrait findings 52 3.5 Towards three dimensional phase portrait analysis of the reaction network 54 3.6 Initial assessment of the compartment’s external environment 54 3.7 Identification of modulators of puncta dynamics 56 3.7.1 CDC-42 controls cortical levels of WSP-1 56 3.7.2 RHO-1 and Formin CYK-1 are not involved in WSP-1 F-actin condensate dynamics 58 3.7.3 WSP-1 and Arp2/3 dynamics are independent of NCK-1 and VAB-1 58 3.7.4 Arp2/3 regulates condensate dynamics 60 3.8 Summary of perturbations 63 4 Conclusions and outlook 64 4.1 Concluding remarks 64 4.2 Discussion 66 4.3 Future directions 67 4.3.1 Realizing the full potential of the phase portraits in identifying biochemical interactions 67 4.3.2 Resolving the ultrastructure of condensates . 70 4.3.3 Further investigation of the biological function 71 4.3.4 Applying full-dynamic data acquisition to other membraneless organelles 71 5 Materials and Methods 72 5.1 C.elegans maintenance and strains 72 5.2 Sample preparation 72 5.2.1 In utero imaging 72 5.2.2 Oocyte imaging 73 5.2.3 C.elegans HaloTag staining 73 5.2.4 Oocyte chemical inhibitor treatments 73 5.3 RNAi Feeding 73 5.4 Microscopy 73 5.4.1 Spinning disk microscopy 73 5.4.2 SIM-TIRF microscopy 74 5.5 TIRF microscopy 74 5.6 Phase portrait analysis pipeline 74 5.7 Kymographs 76

<p>The opioid epidemic has reached new highs in the past decade, with opioid overdose becoming the leading cause of death for Americans age 18-45 in 2019. While there is ongoing research into novel opioids that may avoid the development of tolerance and addiction, there is also research into alternative targets downstream of the opioid receptor. Adenylyl cyclase type 1 (AC1) is one such downstream effector of the opioid receptors that is crucial for opioid-based analgesia. By selectively targeting AC1, bypassing the opioid receptor, we may avoid the development of tolerance and addiction seen with opioid therapy while still treating chronic pain. In this report we have optimized a novel scaffold of selective AC1 inhibitors, the pyrimidinones.Through two generations of pyrimidinones analogs we have yielded the lead compound AC10084, an AC1 inhibitor as efficacious as morphine in vivo with improved aqueous solubility relative to previous AC1 inhibitor scaffolds. Furthermore, utilizing our pyrimidinone scaffold in a pharmacophore-based virtual screen has yielded yet another novel scaffold of AC1 inhibitors, the dithiophenes. In total, our highly collaborative project provides strong evidence that AC1 can be selectively targeted for chronic pain management and that AC1 inhibitors may represent a future alternative for traditional opioids.</p> <p><br></p>

Le dessin computationnel de protéine, ou CPD, est une technique qui permet de modifier les protéines pour leur conférer de nouvelles propriétés, en exploitant leurs structures 3D et une modélisation moléculaire. Pour rendre la méthode de plus en plus prédictive, les modèles employés doivent constamment progresser. Dans cette thèse, nous avons abordé le problème de la représentation explicite de la flexibilité du squelette protéique. Nous avons développé une méthode de dessin "multi-états", qui se base sur une bibliothèque discrète de conformations du squelette, établie à l'avance. Dans un contexte de simulation Monte Carlo, le paysage énergétique d'une protéine étant rugueux, les changements de squelettes ne peuvent etre acceptés que moyennant certaines précautions. Aussi, pour explorer ces conformations, en même temps que des mutations et des mouvements de chaînes latérales, nous avons introduit un nouveau type de déplacement dans une méthode Monte Carlo existante. Il s'agit d'un déplacement "hybride", où un changement de squelette est suivi d'une courte relaxation Monte Carlo des chaînes latérales seules, après laquelle un test d'acceptation est effectué. Pour respecter une distribution de Boltzmann des états, la probabilité doit avoir une forme précise, qui contient une intégrale de chemin, difficile à calculer en pratique. Deux approximations sont explorées en détail: une basée sur un seul chemin de relaxation, ou chemin "générateur" (Single Path Approximation, ou SPA), et une plus complexe basée sur un ensemble de chemins, obtenus en permutant les étapes élémentaires du chemin générateur (Permuted Path Approximation, ou PPA). Ces deux approximations sont étudiées et comparées sur deux protéines. En particulier, nous calculons les énergies relatives des conformations du squelette en utilisant trois méthodes différentes, qui passent réversiblement d'une conformation à l'autre en empruntent des chemins très différents. Le bon accord entre les méthodes, obtenu avec de nombreuses paramétrisations différentes, montre que l'énergie libre se comporte bien comme une fonction d'état, suggérant que les états sont bien échantillonnés selon la distribution de Boltzmann. La méthode d'échantillonnage est ensuite appliquée à une boucle dans le site actif de la tyrosyl-ARNt synthétase, permettant d'identifier des séquences qui favorisent une conformation, soit ouverte, soit fermée de la boucle, permettant en principe de contrôler ou redessiner sa conformation. Nous décrivons enfin un travail préliminaire visant à augmenter encore la flexibilité du squelette, en explorant un espace de conformations continu et non plus discret. Ce changement d'espace oblige à restructurer complètement le calcul des énergies et le déroulement des simulations, augmente considérable le coût des calculs, et nécessite une parallélisation beaucoup plus agressive du logiciel de simulation. / Computational protein design is a method to modify proteins and obtain new properties, using their 3D structure and molecular modelling. To make the method more predictive, the models need continued improvement. In this thesis, we addressed the problem of explicitly representing the flexibility of the protein backbone. We developed a "multi-state" design approach, based on a small library of backbone conformations, defined ahead of time. In a Monte Carlo framework, given the rugged protein energy landscape, large backbone motions can only be accepted if precautions are taken. Thus, to explore these conformations, along with sidechain mutations and motions, we have introduced a new type of Monte Carlo move. The move is a "hybrid" one, where the backbone changes its conformation, then a short Monte Carlo relaxation of the sidechains is done, followed by an acceptation test. To obtain a Boltzmann sampling of states, the acceptation probability should have a specific form, which involves a path integral that is difficult to calculate. Two approximate forms are explored: the first is based on a single relaxation path, or "generating path" (Single Path Approximation or SPA). The second is more complex and relies on a collection of paths, obtained by shuffling the elementary steps of the generating path (Permuted Path Approximation or PPA). These approximations are tested in depth and compared on two proteins. Free energy differences between the backbone conformations are computed using three different approaches, which move the system reversibly from one conformation to another, but follow very different routes. Good agreement is obtained between the methods and a wide range of parameterizations, indicating that the free energy behaves as a state function, as it should, and strongly suggesting that Boltzmann sampling is verified. The sampling method is applied to the tyrosyl-tRNA synthetase enzyme, allowing us to identify sequences that prefer either an open or a closed conformation of an active site loop, so that in principle we can control, or design the loop conformation. Finally, we describe preliminary work to make the protein backbone fully flexible, moving within a continuous and not a discrete space. This new conformational space requires a complete reorganization of the energy calculation and Monte Carlo simulation scheme, increases simulation cost substantially, and requires a much more aggressive parallelization of our software.

No description available.

No description available.