Sistemas alostéricos são característicos de proteínas com um ou mais estados de equilíbrio. Nesse sentido, uma enzima passa por modificações de sua atividade quando um substrato cooperativo se liga a um estado ou outro (1). Estes estados são reconhecidos por possuírem uma conformação mais estável e coexistirem num ensemble. Este trabalho sustenta que tais proteínas oscilem naturalmente entre esses estados. Experimentos de difração de raios-X e RMN, que proporcionam parâmetros de deslocamento anisotrópicos e tempos de relaxação de spin nuclear, já demonstram a coexistência de ambos estados em solução e descrevem o movimento como uma mudança de equilíbrio populacional dos confórmeros (2). Também é possível desenvolver métodos numéricos, como o cálculo de modos normais e a simulação de dinâmica molecular, para associar a geometria proteica a um movimento sobre determinado potencial de campos de força. O sistema adotado para o desenvolvimento desses estudos é a enzima alostérica Glucosamina-6-fosfato Desaminase. Características que defendem seu uso são sua reversibilidade catalítica, rápido equilíbrio cinético e muito baixa afinidade do estado T por ligantes. Sua estrutura também já foi resolvida por experimentos de cristalografia, identificando ambos estados alostéricos. E a caracterização das mudanças estruturais entre os estados T e R está bem estabelecida, identificando diferentes subunidades a distintos graus de rotação e prevendo uma oscilação de baixa frequência entre eles (3). Resultados obtidos neste projeto constituem: (a) uma dinâmica de 100ns partindo do estado T de toda a proteína (hexamérica) solvatada explicitamente, formando um ensemble NVT de 92000 átomos através do programa NAMD, usando o campo de forças CHARMM; (b) análise de componentes principais aproveitando esta dinâmica e usando algoritmos do programa Gromacs; (c) e análise de modos normais, em que os cálculos de minimização de energia foram feitos pelo programa Gromacs sob o campo de forças ENCADV, no vácuo. Análises desses resultados envolvem cálculos de RMSDs e flutuações, trajetórias calculadas para os autovetores oriundos de NMA ou de PCA, fatores de Debye-Waller e a confirmação visual (e gráficos de distância entre resíduos) de aproximação a um estado ou outro. Como a prévia caracterização da movimentação alostérica, identificava duas regiões para cada monômero como representativas de corpos rígidos, também é desenvolvida uma análise por tensores de inércia. Espera-se que, ao longo do tempo, essas subunidades se comportem como corpos quase rígidos e os movimentos destas regiões rígidas correspondam a uma maior representatividade da transição alostérica. Assim, a caracterização dos tensores seria capaz de filtrar movimentos de mais alta frequência que constituem ruído em relação a movimentos funcionais da proteína. - Algoritmos para cálculos matriciais dos tensores foram escritos em Fortran e em TCl. / Allosteric systems are characteristic of proteins with one or more equilibrium states. Such an enzyme experiences a modification of its activity when a cooperative substrate binds to a state or another, thus, establishing a change in population equilibrium (1). These states are recognized by having a more stable conformation, and they coexist in an ensemble. X-ray diffraction and NMR experiments already demonstrated this dynamic equilibrium, and simulation methods, as molecular dynamics and normal mode analysis, generally provide a more complete proof (2).The allosteric enzyme Glucosamine-6-Phosphate Deaminase appeared to be a good model to better understand the equilibrium dynamics as essential to the protein function, given its reversibility of the catalysis and rapid-equilibrium kinetic mechanism. It also has the structure elucidated for both its conformers (3). A computational approach would now give better perspective on how the conformational changes occur. A set of results of this latter kind were obtained: (a) a 100ns dynamic starting at the hexameric T conformer, explicitly solvated, building a NVT ensemble using NAMD program and CHARMM force field; (b) a principal components analysis making use of the calculated dynamic and of the Gromacs algorithms; (c) and normal mode analysis of the T conformer structure (pdb code 1fsf) minimized with Gromacs program using ENCADV vacuum force field. Not only the conventional analyses for these results (fluctuations and projections) were taken, but also an inertia tensor analysis was developed. As the allosteric conformational change, for this protein, was described by the displacement of only two rigid body subunits³, its description by inertia tensors should act as a filter for the high frequency and functionally uninteresting motions, which normally constitute only noise.

Docking is a computer simulation method used to predict the preferred orientation of two interacting chemical species that has been successfully applied to numerous macromolecules over the years. However, non-traditional targets have inherent difficulties associated with their screening. Large interfaces, lack of obvious binding sites, and transient pockets are some examples. Additionally, most natural ligands of challenging targets are inadequate models for identifying or designing new ligands. Therefore, it is not surprising that customary techniques of structure-based virtual screening are incompatible with these non-traditional targets. We hypothesized that an integrative virtual screening campaign comprised of docking followed by refinement of best receptor–ligand complexes would effectively identify small-molecule ligands of challenging receptors. We targeted the single-stranded DNA (ssDNA) binding groove of the human RAD52, and a cryptic allosteric pocket of the Helicobacter pylori Glutamate Racemase (GR). In this project, we first determined which docking method was more appropriate for each studied non-traditional target, and then examined how good our two-step docking workflow was in finding novel active ligand scaffolds. This research developed a powerful layered virtual screening workflow for the discovery of lead compounds against challenging protein targets. Furthermore, we successfully applied a statistical analysis method, which used receiver operating characteristic (ROC) curves, to validate the selected docking protocol that would be used in the screening campaigns. Using the validated workflow, we identified a natural compound that competes with ssDNA to bind to RAD52. The performed screening campaigns also provided new insights into the studied binding pockets, as well as structure-activity relationships (SAR) and binding determinants of the ligands. Our achievements reinforce the power of the ROC curve analysis approach in directing the search for the most appropriate docking protocol and helping to speed up drug discovery in pharmaceutical research.

No description available.

Microbial degradation of aromatic hydrocarbons is imperative for maintaining the global carbon cycle and removing potentially toxic aromatic xenobiotics. This thesis focuses on the characterization of a pyruvate-specific class II aldolase (BphI) and acetaldehyde dehydrogenase (BphJ), the final two enzymes of the bph meta-cleavage pathway in Burkholderia xenovorans LB400. This pathway is responsible for the degradation of the industrial pollutant polychlorinated biphenyls (PCB) and therefore mechanistic characterization of these enzymes can be applied to improve pollutant degradation. BphI catalyzes the aldol cleavage of 4-hydroxy-2-oxoacids to pyruvate and an aldehyde while BphJ transforms aldehydes to acyl-CoA, using NAD+ and CoASH as cofactors. Size-exclusion chromatography was used to determine that the oligomeric unit of the BphI-BphJ complex is a heterotetramer. The aldolase BphI was shown to exhibit a compulsory order mechanism and utilize 4-hydroxy-2-oxoacids with an S configuration at C4. The generation of BphI active site variants allowed for the proposal of a catalytic mechanism and a greater understanding as to how stereospecificity occurs. Using steady-state kinetic assays, Arg-16 was demonstrated to be essential for catalysis. Molecular modeling of the substrate and pH dependency (wild-type pKa of ~7, lost in H20A and H20S variants) were used to identify His-20 as the catalytic base. Tyr-290 was originally proposed to be the catalytic acid. However, this was refuted as a Tyr-290 (Y290F) variant did not affect the catalytic efficiency of the enzyme. Instead, the variant was observed to exhibit a loss of stereochemical control. From the crystal structure of an orthologous aldolase-dehydrogenase complex, solvent isotope effect studies, and a proton inventory, a water molecule was implicated as the catalytic acid. Based on their position within the crystal structure, Leu-87 and Leu-89 were implicated in substrate specificity. Replacement of Leu-89 with alanine effectively increased the length of the active site, allowing for the accommodation of longer aldehyde substrates. In contrast, Leu-87 was responsible for hydrophobic stabilization of the C4-methyl of the substrate. Double variants L87N;Y290F and L87W;Y290F were constructed to enable the binding of 4(R)-hydroxy-2-oxoacids. Polarimetric analysis confirmed that the double variants were able to synthesize 4-hydroxy-2-oxoacids of up to 8 carbons in lengths, which were of the opposite stereoisomer to those produced by the wild-type enzyme. Cys-131 was identified as the catalytic thiol that forms an acyl-enzyme intermediate in the dehydrogenase, BphJ. This enzyme was shown to exhibit similar specificity constants for acetaldehyde and propionaldehyde and utilize aliphatic aldehydes from two to five carbons in length as substrates. The enzyme was able to use either NAD+ or NADP+ as the cofactor. Finally, we demonstrated that aldehydes produced in the aldolase reaction are not released into the bulk solvent but are channeled directly to the dehydrogenase, providing the first biochemical determination of substrate channeling in any aldolase-dehydrogenase complex. / Chapter 3 - Reprinted (adapted) with permission from Baker, P., Carere, J., and Seah, S. Y. (2011) Probing the Molecular Basis of Substrate Specificity, Stereospecificity, and Catalysis in the Class II Pyruvate Aldolase, BphI, Biochemistry 50: 3559-3569. Copyright (2011) American Chemical Society. Chapter 4 - Reprinted (adapted) with permission from Baker, P., and Seah, S. Y. (2012) Rational design of stereoselectivity in the class II pyruvate aldolase BphI, J Am Chem Soc 134: 507-513. Copyright (2012) American Chemical Society. Chapter 6 - Reprinted (adapted) with permission from Baker, P., Hillis, C., Carere, J., and Seah, S. Y. (2012) Protein-protein interactions and substrate channeling in orthologous and chimeric aldolase-dehydrogenase complexes, Biochemistry 51: 1942-1952. Copyright (2012) American Chemical Society. / National Science and Engineering Research Council of Canada (NSERC), Ontario Graduate Scholarship in Science and Technology

Ce travail de thèse PhD concerne l'application d'une approche intégrée pour obtenir une meilleure compréhension des mécanismes d'action de Akt et CXCR4, surexprimées dans différents cancers humains. Efforts récents dans le développement et l'évaluation biologique (activité antiproliférative) de petites entités moléculaires inhibitrices d’Akt, une sérine/thréonine protéine-kinase, ont conduit à l'identification de nouveaux inhibiteurs pyrrolo[1,2-a]quinoxaline, conçus et préparés via une stratégie multi-étapes. Certains des composés synthétisés ont montré une activité contre les lignées cellulaires leucémiques testées (Jurkat, U266, K562, U937 et HL60) supérieure à celle de composé de référence A6730. En outre, des résultats préliminaires menés sur Akt puis sur le domaine de PH isolé d’Akt, ont montré que ils peuvent être considérés comme des inhibiteurs allostériques potentiels. La seconde partie des travaux concerne la conception et la synthèse de deux nouvelles séquences peptidiques contenant quelques acides aminés "disorder promoting" et une unité CPC. Les études CD, RMN et MD ont fait ressortir leur flexibilité et ont démontré leurs capacités à assumer des ensembles de conformations stabilisées par un réseau de liaisons hydrogène. Ensuite, nous avons étudié l'effet de la chaîne alkyle reliée sur la conformation des peptides. Des études de fluorescence et DLS ont été réalisées pour évaluer les CMC et la dimension des agrégats supramoléculaires. Les tests biologiques ont souligné que ces édifices moléculaires (peptides amphiphiles nommés, PAs) montrent ainsi des activités prometteuses, voire plus que la molécule de référence (AMD3100). / This PhD thesis work has been covered in the application of an integrated approach to get a better understanding about the mechanism of action of two systems: Akt and CXCR4, proteins overexpressed in different human cancers. On the basis of previous results obtained on the antiproliferative activities of small molecule inhibitors of Akt, a serine/threonine protein kinase, a novel series of pyrrolo[1,2-a]quinoxaline derivatives have been designed and synthesized via multistep heterocyclization process. Some compounds showed promising activities against all leukemia cell lines tested (Jurkat, U266, K562, U937 and HL60), even better than the reference compound (A6730) one. In addition, docking results, conducted on the isolated PH domain, showed that these new compounds could be considered as allosteric inhibitors. The second workpackage reports on the design and the synthesis of two new peptidic sequences containing a few amino acids “disorder promoting” and a CPC unit, the CXCL12 binding motif towards CXCR4. The peptide structural preferences were analysed by CD, NMR and MD techniques that highlighted their flexibility and demonstrated the ability of these peptides to assume conformational ensembles stabilized by a network of transient and dynamic H-bonds. Afterwards we studied the alkyl chain effect on the conformation of the peptide portion. Solution fluorescence and DLS studies have been performed to evaluate CMC and the dimension of supramolecular aggregates (named peptide amphiphiles, PAs). Biological tests pointed out that these molecular buildings show promising activities, even higher than the reference molecule (AMD3100) one.

Sistemas alostéricos são característicos de proteínas com um ou mais estados de equilíbrio. Nesse sentido, uma enzima passa por modificações de sua atividade quando um substrato cooperativo se liga a um estado ou outro (1). Estes estados são reconhecidos por possuírem uma conformação mais estável e coexistirem num ensemble. Este trabalho sustenta que tais proteínas oscilem naturalmente entre esses estados. Experimentos de difração de raios-X e RMN, que proporcionam parâmetros de deslocamento anisotrópicos e tempos de relaxação de spin nuclear, já demonstram a coexistência de ambos estados em solução e descrevem o movimento como uma mudança de equilíbrio populacional dos confórmeros (2). Também é possível desenvolver métodos numéricos, como o cálculo de modos normais e a simulação de dinâmica molecular, para associar a geometria proteica a um movimento sobre determinado potencial de campos de força. O sistema adotado para o desenvolvimento desses estudos é a enzima alostérica Glucosamina-6-fosfato Desaminase. Características que defendem seu uso são sua reversibilidade catalítica, rápido equilíbrio cinético e muito baixa afinidade do estado T por ligantes. Sua estrutura também já foi resolvida por experimentos de cristalografia, identificando ambos estados alostéricos. E a caracterização das mudanças estruturais entre os estados T e R está bem estabelecida, identificando diferentes subunidades a distintos graus de rotação e prevendo uma oscilação de baixa frequência entre eles (3). Resultados obtidos neste projeto constituem: (a) uma dinâmica de 100ns partindo do estado T de toda a proteína (hexamérica) solvatada explicitamente, formando um ensemble NVT de 92000 átomos através do programa NAMD, usando o campo de forças CHARMM; (b) análise de componentes principais aproveitando esta dinâmica e usando algoritmos do programa Gromacs; (c) e análise de modos normais, em que os cálculos de minimização de energia foram feitos pelo programa Gromacs sob o campo de forças ENCADV, no vácuo. Análises desses resultados envolvem cálculos de RMSDs e flutuações, trajetórias calculadas para os autovetores oriundos de NMA ou de PCA, fatores de Debye-Waller e a confirmação visual (e gráficos de distância entre resíduos) de aproximação a um estado ou outro. Como a prévia caracterização da movimentação alostérica, identificava duas regiões para cada monômero como representativas de corpos rígidos, também é desenvolvida uma análise por tensores de inércia. Espera-se que, ao longo do tempo, essas subunidades se comportem como corpos quase rígidos e os movimentos destas regiões rígidas correspondam a uma maior representatividade da transição alostérica. Assim, a caracterização dos tensores seria capaz de filtrar movimentos de mais alta frequência que constituem ruído em relação a movimentos funcionais da proteína. - Algoritmos para cálculos matriciais dos tensores foram escritos em Fortran e em TCl. / Allosteric systems are characteristic of proteins with one or more equilibrium states. Such an enzyme experiences a modification of its activity when a cooperative substrate binds to a state or another, thus, establishing a change in population equilibrium (1). These states are recognized by having a more stable conformation, and they coexist in an ensemble. X-ray diffraction and NMR experiments already demonstrated this dynamic equilibrium, and simulation methods, as molecular dynamics and normal mode analysis, generally provide a more complete proof (2).The allosteric enzyme Glucosamine-6-Phosphate Deaminase appeared to be a good model to better understand the equilibrium dynamics as essential to the protein function, given its reversibility of the catalysis and rapid-equilibrium kinetic mechanism. It also has the structure elucidated for both its conformers (3). A computational approach would now give better perspective on how the conformational changes occur. A set of results of this latter kind were obtained: (a) a 100ns dynamic starting at the hexameric T conformer, explicitly solvated, building a NVT ensemble using NAMD program and CHARMM force field; (b) a principal components analysis making use of the calculated dynamic and of the Gromacs algorithms; (c) and normal mode analysis of the T conformer structure (pdb code 1fsf) minimized with Gromacs program using ENCADV vacuum force field. Not only the conventional analyses for these results (fluctuations and projections) were taken, but also an inertia tensor analysis was developed. As the allosteric conformational change, for this protein, was described by the displacement of only two rigid body subunits³, its description by inertia tensors should act as a filter for the high frequency and functionally uninteresting motions, which normally constitute only noise.

Motile microorganisms navigate through their environment using special molecular machinery in order to sense gradients of various signals: chemotaxis (reactions to chemical compounds) and phototaxis (to light) sensory cascades. Transmembrane receptors play a central role in these cascades as they receive input signals and transmit them inside the cell, where they modulate activity of the kinases CheA, which are tightly bound to their cytoplasmic domains. CheA further phosphorylates the response regulator protein CheY, which regulates the flagella. At the same time, CheA phosphorylates and, by means of this, activates another response regulator, CheB, which, along with the constantly active CheR protein, catalyzes two opposite reactions: methylation and demethylation of the specific glutamic acid residues located at the cytoplasmic domain of the receptors. The latter reactions establish the adaptation mechanism, which allows microbes to sense in a very broad range of the input signal intensities. Many functional, structural and dynamical aspects of the signal propagation through the prokaryotic receptors as well as a mechanism of the signal amplification remain still unclear. In the present thesis we have used various techniques of computational biophysics, chiefly molecular dynamics (MD) simulations, in order to approach these problems. In Chapter 3, we have carried out MD simulations of the isolated linker domain (HAMP) from the E. coli Tsr chemoreceptor. The MD simulations revealed highly dynamical nature of this domain, which allows for interconversion between several metastable states. These metastable states feature a number of structural and dynamical properties, which were previously reported for HAMP domains of various receptors obtained from different organisms. It allowed us to reconcile numerous experimental data and to hypothesize that different HAMP domains share similar mechanism of their action. In Chapter 4, we have performed MD simulations of the whole cytoplasmic domain of the Tsr chemoreceptor. The simulations revealed a mechanism for the inter-domain coupling between the HAMP domain and a part of the cytoplasmic domain adjacent to the HAMP, the adaptation subdomain, by means of the regulated unfolding of a short linker region termed the stutter. Also, we have found that the reversible methylation/demethylation of the cytoplasmic domain affects its flexibility and symmetry. Altogether, these findings suggest a mechanism of signal propagation at the level of an individual chemoreceptor dimer. In Chapter 5, we have built a model of the trimer-of-dimers of the archaeal phototaxis receptor complex (NpSRII:NpHtrII). Subsequent MD simulations revealed an important role of dynamics in signal transduction and, potentially, in the kinase activation. In Chapter 6, we have reconstructed a whole transmembrane lattice formed by the NpSRII:NpHtrII complexes. The concave shape of the obtained lattice naturally explains polar localization of the receptor arrays in prokaryotic cells. At the same time, additional MD simulations of an individual unit of this lattice (a dimer of the photosensor) revealed global motional modes in its transmembrane region, which presumably co-occur with its activation and can spread across the tightly packed transmembrane arrays allowing for the signal amplification.

Allosteric inhibition of kinases provides high selectivity and potency due to lower evolutionary pressure in conserving allosteric vs. orthosteric sites. The former are regions distinct from the kinase active site, yet, when perturbed through allosteric effectors, induce conformational and/or dynamical changes that in turn modulate kinase function. Protein kinases involved in cyclic nucleotide signalling are important targets for allosteric inhibition due to their association with diseases, from infections to Cushing’s syndrome. This dissertation specifically focuses on elucidating the molecular mechanism of allosteric inhibition in the cAMP-dependent protein kinase (PKA) and the Plasmodium falciparum cGMP-dependent protein kinase (PfPKG), which are targets for a generalized tumor predisposition commonly referred to as Carney Complex and for malaria, respectively. In chapters 2 and 3, we focus on the agonism-antagonism switch (i.e. allosteric pluripotency) observed as the phosphorothioate analog of cAMP, Rp-cAMPS (Rp), binds to PKA. Utilizing Nuclear Magnetic Resonance (NMR), Molecular Dynamics (MD) simulations and Ensemble Allosteric Model (EAM), we determined that two highly homologous cAMP-binding domains respond differently to Rp, giving rise to a conformational ensemble that includes excited inhibition-competent states. The free energy difference between this state and the ground inhibition-incompetent state is tuned to be similar to the effective free energy of association of the regulatory (R) and catalytic (C) subunits, leading to allosteric pluripotency depending on conditions that perturb the R:C affinity. The general significance of these results is a re-definition of the concept of allosteric target to include not only the isolated allosteric receptor, but also its metabolic and proteomic sub-cellular environment. In chapter 4, we utilize a mutant that silences allosteric pluripotency to reveal that the agonism-antagonism switch of PKA not only arises from the mixed response of tandem domains, but also from the mixed response of allosteric regions within a single domain that mediates interactions with Rp. In chapter 5, the allosteric inhibition of PfPKG associated with malaria is induced through base-modified cGMP-analogs and the underlying inhibitory mechanism is determined. We show that, when bound to a PfPKG antagonist, the regulatory domain of PfPKG samples a mixed intermediate state distinct from the native inhibitory and active conformations. This mixed state stabilizes key cGMP-binding regions, while perturbing the regions critical for activation, and therefore it provides an avenue to preserve high affinity, while promoting significant inhibition. Overall, in this thesis, previously elusive mechanisms of allosteric inhibition were elucidated through the combination of NMR, MD, and EAM methods. Through this integrated approach, we have unveiled an emerging theme of inhibitory ‘mixed’ states, either within a single domain or between domains, which offer a simple but effective explanation for functional allostery in kinases. / Thesis / Candidate in Philosophy

No description available.

Au niveau de la synapse, la liaison des neurotransmetteurs aux récepteurs membranaires induit l’ouverture de canaux ioniques. Le Récepteur de la Glycine (RGly) est un récepteur ionotrope impliqué dans des troubles neuronaux tels que l’addiction, la douleur chronique, ou l’hyperekplexie ; pour cette raison il est important de développer des nouveaux traitements ciblant ce récepteur. Nous avons utilisé des simulations de Dynamiques Moléculaire (DM) et d’électrophysiologie numérique afin d’évaluer la fonction des structures du RGly disponibles et montré qu’aucune d’entre elles ne satisfait les propriétés fonctionnelles de l’état ouvert. Grâce aux simulations de DM, nous avons caractérisé une nouvelle conformation du RGly, qui est compatible avec cet état. Nous avons souligné le rôle majeur des portails latéraux pour la perméation des ions. Nous avons proposé un protocole, nommé pharmacologie dépendante de l’état, pour identifier des molécules modulatrices de protéines allostériques. / Signals within neurons are mostly transmitted through chemical synapses. Signal transduction arises from the binding of neurotransmitters to membrane receptors in order to open ion channels. The Glycine Receptor (GlyR) is an ionotropic receptor which is involved in several neurological disorders such as addiction, chronic pain, or hyperekplexia. Because of its implication in human diseases, it is interesting to design novel drugs targeting this receptor. We used Molecular Dynamics (MD) simulations and computational electrophysiology to probe the function of available GlyR structures. We showed that none of the experimental structures display the physiological behavior of the conductive state. Using MD simulations, we captured a novel conformation of the GlyR compatible with a conductive state and demonstrated the importance of lateral portals for ionic permeation. Lastly, we proposed an original protocol, named state-based pharmacology, to discover modulators of allosteric proteins.