Predicting the three-dimensional (3D) structure of proteins is a central problem in biology. These computationally predicted 3D protein structures have been successfully applied in many fields of biomedicine, e.g. family assignments and drug discovery. The accurate detection of remotely homologous templates is critical for the successful prediction of the 3D structure of proteins. Also, the prediction of one-dimensional (1D) protein structures such as secondary structures and shape strings are useful for predicting the 3D structure of proteins and important for understanding the sequence-structure relationship. In addition, the prediction of the functional sites of proteins, such as metal-binding sites, can not only reveal the important function of proteins (even in the absence of the 3D structure) but also facilitate the prediction of the 3D structure. Here, three novel methods in the field of protein structure prediction are presented: PREDZINC, a method for predicting zinc-binding sites in proteins; Frag1D, a method for predicting the 1D structure of proteins; and FragMatch, a method for detecting remotely homologous proteins. These methods compete satisfactorily with the best methods previously published and contribute to the task of protein structure prediction. / At the time of the doctoral defense, the following paper was unpublished and had a status as follows: Paper 3: Manuscript. / Protein structure prediction

Histone deacetylases (HDACs) are a class of enzymes that play a crucial role in DNA expression by removing an acetyl group from the ɛ-N-acetyl lysine residue on histone proteins. Out of 18 isoforms of HDAC enzymes which are classified into 4 classes, only 11 of them are metalloenzymes that require zinc for its catalytic activity. HDACs are considered promising target for drug development in cancer and other parasitic diseases due to their role in gene expression. Histone deacetylase inhibitors (HDACi) can cause cell cycle arrest, and induce differentiation or apotosis. While HDACi shows promising antitumor effects, their mechanism of action and selectivity against cancer cells have not been adequately defined yet. In addition, low oral bioavailability, short half-life time, bone marrow toxicity, and cardiotoxicity limit their use in clinic. Therefore, there is considerable interest in developing compounds with selectivity and specificity towards individual family members of HDACs. The prototypical pharmacophore for HDAC inhibitors consist of a metal-binding moiety that coordinates to the catalytic metal ion within the HDAC active site, a capping group that interacts with the residues at the entrance of the active site and a linker that appropriately positions the metal-binding moiety and capping group for interactions in the active site. It has been shown that modification of cap, cap linking moiety, linker or zinc binding group (ZBG) shows promises of superior potency and isoform selectivity. My thesis research involves manipulating different aspects of the pharmacophoric model to yield not only more potent, selective, and effective drugs but also to help understand the biology of HDAC isoforms. In addition, I was successful in extending studies on HDAC isoforms to other zinc metalloenzymes such as leishmanolysin (gp63) and spliceosome associated zinc-metalloenzymes to understand biology of these zinc metalloenzymes by developing potent and selective small molecule inhibitors. This will aid in improvement of existing therapeutics for treatment of cancer, leishmania, malaria and other genetic disorders.

Les mécanismes de reconnaissance moléculaire sont à la base de nombreuses fonction biologiques essentielles (transduction du signal, régulation de l’expression génique, stimulation du système immunitaire,...). La compréhension des phénomènes physiques et chimiques à la base de ces phénomènes est fondamentale pour de nombreuses applications telles que la conception de médicaments, le développement d’outils diagnostiques ou tout autre procédé biotechnologique. Dans cette étude, nous avons étudié de manière extensive l’interaction entre la métalloélastase du macrophage (MMP-12) et le RXP470.1, un inhibiteur puissant et sélectif. En combinant des approches de cristallographie, de microcalorimétrie (ITC) et des tests enzymatiques, nous avons pu quantifier l’importance énergétique du transfert d’un seul proton suite à la liaison du RXP470.1, et mettre en lumière l’importance des contributions entropiques. Ainsi, la protonation du Glu219, un résidu catalytique, permet de compenser une enthalpie de liaison intrinsèque défavorable. Cette protonation est rendue possible par le large shift de pKa que subit le Glu219 en réponse à la liaison du RXP470.1 (pKalibre = 5.7 ± 0.1 / pKalié = 10 ± 0.04). Enfin, cette étude est la première, à notre connaissance, ayant combiné données d’affinité, thermodynamiques et structurales pour aborder le rôle du groupe chélatant. Nous avons ainsi étudié deux analogues du RXP470.1 variant seulement par la nature de leur pince à zinc. Ces modifications se traduisent par un effet marqué sur les profils d’affinité et de sélectivité ainsi que sur la signature énergétique des composés étudiés. L’étude des facteurs B, associés à l’analyse des structures cristallographiques, de ces inhibiteurs en complexe avec la MMP-12, suggère que des différences mineures de structures peuvent engendrer des variations de mobilité importantes au niveau des résidus impliqués dans l’interaction inhibiteur - enzyme. Ces différences trouvent leur origine dans un positionnement très légèrement différent du groupe chélatant par rapport au zinc.Pris dans leur ensemble, ces résultats pointent la nécessité de combiner un ensemble d’approches expérimentales pour décrire la complexité des interactions protéine/ligand. Ces associations doivent permettre d’évaluer le potentiel de méthodes théoriques capables de décrire des systèmes complexes. / Protein-ligand recognition mechanisms are essential to many fundamental biological functions such as signal transduction, gene regulation or stimulation of the immune system. Understanding the physical and chemical phenomenon upon protein-ligand binding is essential for many practical applications such as drug design, ligand based diagnostic tools and any other study based on biotechnology. In this study, we extensively explored the interaction between human macrophage metallo elastase MMP-12 and RXP470.1, a potent and selective inhibitor. By combining high resolution X-ray crystallography, FRET based enzyme assays and Isothermal Titration Calorimetry, we were able to highlight the importance of entropic contributions and to quantify the importance of a single proton transfer upon RXP470.1 binding. We show, here, how the protonation of Glu219 upon RXP470.1 binding rescues an otherwise unfavourable binding enthalpy. This protonation is made possible by the large pKa shift Glu219 undergoes as RXP470 enters MMP12h To our knowledge, this study is also the first to address the zinc binding group effect from affinity, thermodynamlc and structural data. We tested two RXP470 analogues, which only differ by their zinc-binding group. We show that this, apparently minor, change has great consequences regarding their affinity profiles and thermodynamic signatures. In addition, the analysis of the b factors, associated to the X-ray structures of these compounds in complex with MMP-12, suggests that small modifications of the zinc binding group might imply important mobility variations of the residues involved in the protein-ligand interaction. These modifications are initiated by a small shift of the zinc binding groups positioning in the active site.Taken together, these results point toward the necessity to combine several experimental approaches to describe the complexity of protein-ligand interactions. These associations should allow the evaluation of new theoretical methods able to describe complex systems.

No description available.

In this thesis the properties of proteins and membranes are studied by molecular dynamics simulations. The subject is decomposed into parts addressing free energy calculations in proteins, mechanical inclusion models for lipid bilayers, phase transitions and structural correlations in lipid bilayers and atomistic lipid bilayer models. The work is based on results from large scale computer simulations, quantum mechanical and continuum models. Efficient statistical sampling and the coarseness of the models needed to describe the ordered and disordered states are of central concern. Classical free energy calculations of zinc binding, in metalloproteins, require a quantum mechanical correction in order to obtain realistic binding energies. Classical electrostatic polarisation will influence the binding energy in a large region surrounding the ion and produce reasonable equilibrium structures in the bound state, when compared to experimental evidence. The free energy for inserting a protein into a membrane is calculated with continuum theory. The free energy is assumed quadratic in the mismatch and depend on two elastic constants of the membrane. Under these circumstances, the free energy can then be written as a line tension multiplied by the circumference of the membrane inclusion. The inclusion model and coarse grained particle simulations of the membranes show that the thickness profile around the protein will be an exponentially damped oscillation. Coarse-grained particle simulations of model membranes containing mixtures of phospholipid and cholesterol molecules at different conditions were performed. The gel-to-liquid crystalline phase transition is successively weakened with increasing amounts of cholesterol without disappearing even at a concentration of cholesterol as high as 60%. A united atom parameterization of diacyl lipids was constructed. The aim was to construct a new force field that retains and improves the good agreement for the fluid phase and at the same time produces a gel phase at low temperatures, with properties coherent with experimental findings. The global bilayer tilt obtains an azimuthal value of 31◦ and is aligned between lattice vectors in the bilayer plane. It is also shown that the model yield a correct heat of melting as well as heat capacities in the fluid and gel phase of DPPC. / <p>QC 20140919</p>

No description available.