COR15A is a cold regulated disordered protein from Arabidopsis thaliana that contributes to freezing tolerance in plants by protecting membranes. It belongs to the (LEA) Late Embryogenesis Abundant group of proteins that accumulate during the later stage of seed development and are expressed in various parts of the plant. During freezing-induced cellular dehydration, COR15A transitions from a disordered structure to a mostly α-helical structure that binds and stabilizes chloroplast membranes when cells dehydrate due to freezing. We hypothesize that increasing the transient α-helicity of COR15A under normal conditions will increase its ability to bind and protect chloroplast membranes when cells are frozen. To test this hypothesis, conserved glycine residues were mutated to alanine to increase α-helicity. NMR spectroscopy was used to examine structural changes of these mutants compared to wildtype in 0% and 20% TFE. The impact of these mutations on the stability of model membranes during a freeze-thaw cycle was investigated by fluorescence spectroscopy. The results of these experiments showed the mutants had a higher content of α-helical secondary structure than wildtype in 0% and 20% TFE. Increased α-helicity of the COR15A mutants improved membrane stabilization during freezing. Altogether, our results suggest the conserved glycine residues are important for maintaining the disordered structure of the protein.

The proper packaging of the genome in eukaryotic nuclei is essential for proper gene expression and cell function. Chromatin at the large scale is divided into two major compartments heterochromatin and euchromatin. Heterochromatin compromises the transcriptionally inactive tightly packaged regions of chromatin, while euchromatin is the transcriptionally active region of chromatin. The Heterochromatin Protein family (HP1) proteins are epigenetic hallmarks of constitutive heterochromatin. Recent evidence suggests human HP1α undergoes liquid-liquid phase separation suggesting a role for HP1 phase separation in the formation of compacted heterochromatin within HP1 droplets. Phase separation is a biophysical property of proteins with intrinsically disordered domains which are protein domains that lack a defined secondary structure and have the ability to undertake multiple conformations. In this thesis, I investigated the ability of C. elegans HP1 homologs HPL-2A and HPL-1 to phase separate utilizing directed mutations to elucidate the intermolecular interactions that govern this phenomenon and different assays to assess their phase separation. I concluded that HPL-2A is a bona fide phase separating protein that selectively condenses chromatin. HPL-2A’s phase separation depends on specific interactions, mainly dimerization and the presence of lysine and arginine residues in the hinge region. HPL-2A has a specific IDR that drives its phase separation which is the hinge region as the CTE and NTE are not essential for its phase separation.

No description available.

Serum amyloid A (SAA) is a small, evolutionarily well-conserved, acute-phase protein best known as the protein precursor for amyloid A amyloidosis. During acute injury, infection, or inflammation, SAA plasma concentration rapidly rises 1000-fold, but the benefit of this dramatic increase is unclear. SAA functions in the innate immune response, cell signaling, and lipid homeostasis. Most SAA circulates on plasma high-density lipoproteins (HDL), where it reroutes HDL for lipid recycling. The aim of this dissertation is to provide a structural basis for understanding SAA-lipid interactions and to elucidate the structure-function relationship in this ancient protein. SAA is an intrinsically disordered protein that acquires ~50% helical structure when bound to lipids, and is ~80% helical in three available atomic-resolution x-ray crystal structures. We took advantage of these crystal structures of lipid-free SAA to propose the binding site for various lipids, including lipids in HDL. We postulated that SAA, as a monomer, binds lipids via two amphipathic helices, h1 and h3, that form a concave hydrophobic surface, and that the curvature of this surface defines the binding preference of SAA for HDL versus larger lipoproteins. Next, we used murine SAA1.1 and a membrane-mimicking model phospholipid, palmitoyl-oleoyl phosphocholine (POPC), to reconstitute SAA-lipid complexes and characterize their overall structure, stability and stoichiometry using an array of spectroscopic, electron microscopic, and biochemical methods. We observed preferential formation of ~10 nm particles that mimic HDL size, accompanied by the α-helical folding. To probe the local protein conformation and dynamics in these SAA-POPC particles, we used hydrogen-deuterium exchange mass spectrometry. Analysis of the amount and the kinetics of deuterium uptake clearly established h1 and h3 as the lipid-binding site. Moreover, we determined that SAA binding to lipid follows a mixed model that combines induced fit, promoting α-folding in h3, with conformational selection, stabilizing pre-existing conformations in h1 and around the h2-h3 linker. Taken together, our results provided the structural basis necessary for understanding SAA-lipid interactions, which are central to beneficial functions of SAA as a housekeeping molecule, and to its misfolding in amyloid. This research sets the stage for understanding SAA interactions with its numerous other functional ligands.

No description available.

p53 is a transcription factor and an important tumor suppressor protein that becomes activated due to DNA damage. Because of its role as a tumor suppressor, mutations in the gene that encodes it are found in over 50% of human cancers. The N-terminal transactivation domain (NTAD) of p53 is intrinsically disordered and modulates the function and interactions of p53 in the cell. Its disordered structure allows it to be controlled closely by post-translation modifications that regulate p53’s ability to bind DNA and interact with regulatory binding partners. p53 is an attractive target for developing cancer therapeutics, but its intrinsically disordered region makes it difficult for traditional experimental techniques to resolve its heterogeneous conformational ensemble. This challenge necessitates the use of techniques that can capture the transient structural features and interactions of p53 to aid in designing effective drugs that can modulate and stabilize its activity. Hybrid-resolution (HyRes) II is a coarse-grained molecular dynamics force field that was parameterized specifically to capture the dynamics of IDPs and can give insight into secondary structure propensity and how post-translational modifications affect the structural ensemble of the protein. Nanopore experiments allow for real-time, single-molecule studies of protein dynamics and interactions with binding partners through characteristic changes in the current that passes through the nanopore. Pairing nanopore experiments with simulations can give insight into the molecular detail of IDP ensembles and interactions, revealing a fuller picture of how p53 is controlled in stressed cell conditions and how its structure is affected due to various modifications and small molecules with therapeutic implications. Herein, we show the HyRes II force field can capture the complex, long-range dynamics of the p53 tetramer and provide molecular-level detail of the p53 autoinhibition mechanism, which is enhanced by the phosphorylation of the NTAD. Secondly, we use the MspA nanopore to capture the differences in events of the wild-type NTAD and a cancer-associated NTAD mutant. Lastly, we detect a small molecule binding to the WT NTAD using nanopore sensing. This approach of integrating MD simulations and nanopore experiments can be applied to the study of other IDPs which are prevalent in biology and integral to human health and disease.

Intrinsically disordered proteins (IDPs) are an abundant class of proteins and protein regions which rapidly change between multiple structures without an equilibrium position. IDPs exist as a series of conformational ensembles of semi-stable conformations that can be adopted based on a hilly landscape of shallow free energy minima. Disordered sequences share characteristic features differentiating them from globular proteins, including low sequence complexity, low occurrence of hydrophobic residues, high polar and charged residue content, and high flexibility. IDPs are commonly involved in regulation in the cell, and frequently function as, or interact with, hub proteins in protein-protein interaction networks, making them an important class of macromolecules for understanding regulatory and other processes. Given their functional importance, these proteins are widely studied. Many analytical techniques are used, though rapid conformational sampling by IDPs makes it difficult to detect many states with NMR or other techniques. Computational approaches such as molecular dynamics are increasingly used to probe the binding and conformational sampling of these proteins, allowing for observation of factors that cannot be observed with traditional analytical methods such as NMR, such as differing conformational ensembles and the dipoles of individual residues. Here, we studied the role of electrostatic interactions in IDP protein-protein interaction using molecular dynamics simulations performed with the Drude-2019 force field (FF), a polarizable model that allows for more accurate representation of electrostatics, an important factor for highly charged systems like IDPs. For this project, a prototypical protein with disordered regions, p53, was simulated with two protein partners, the nuclear coactivator domain of the CREB binding protein (CBP), and the E3 ubiquitin-protein ligase mouse double minute 2 (MDM2). p53 is widely studied, and the p53 transactivation domain (TAD) is disordered and binds to many structurally diverse partners, making this protein domain a useful model for probing the role of electrostatic interactions formed by IDPs at protein-protein binding interfaces. We found that the Drude-2019 FF allows for simulation of the p53 TAD with Cα chemical shifts comparable to those observed with NMR, supporting that the Drude-2019 FF performs well in simulating IDPs. We observed large relative change in sidechain dipole moments when comparing the p53 TAD alone and when bound to either CBP or MDM2. We observed that aliphatic and aromatic amino acids experienced the largest relative change in sidechain dipole moments, and that there is sensitivity to binding shown in this dipole response. The largest percent changes in sidechain dipole moment were found to localize at and around the binding interface. Understanding the binding interactions of IDPs at a fundamental level, including the role of electrostatic interactions, may help with targeting IDPs or their partners for drug design. / Master of Science in Life Sciences / Many proteins adopt one main structure, and these proteins are called ordered proteins. Intrinsically disordered proteins (IDPs) are an abundant category of proteins which adopt multiple structures, and transition between these different structures is based on factors such as the environment around them, modifications, or interactions with other macromolecules. The flexible structures of IDPs allow them to bind to multiple different partners and to regulate processes in the cell. Since IDPs often regulate processes important to cell function, when these proteins are mutated, misfolded, or otherwise mis-regulated the resulting issues are associated with disease states. IDPs are widely studied with analytical techniques, but because IDPs frequently change shape it can be difficult to observe certain behaviors or certain factors with these techniques. Computational approaches, such as molecular dynamics (MD). MD is the study of molecular motion and interaction, and can allow observation of factors that would be difficult or impossible to observe otherwise, such as the varying structures of IDPs or the dipole moments of specific amino acids within the proteins. For this project we wanted to probe the role of dipole moments, which are charge-based interactions, in the binding of IDPs to protein partners, to better understand how IDPs bind to different partners. We used the p53 protein as an example of IDP binding and simulated it alone and bound to two other proteins, the CREB binding protein (CBP), and the E3 ubiquitin-protein ligase mouse double minute 2 (MDM2). We observed that our simulations were comparable to experiments done with nuclear magnetic resonance spectroscopy, which served to validate that our simulations were realistic. We observed that the dipole moments of the proteins change when simulating the proteins alone and in complex, and that the largest relative changes in dipole are observed for regions of the proteins involved in binding. Probing the role of charge-based interactions in protein-protein binding interactions for IDPs can help us to greater understand these interactions at a more fundamental level and could help with targeting IDPs or their partners for drug design or other problems.

An infinitesimal fraction of the practically infinite sequence space has achieved enormous functional diversity of proteins during evolution. Intrinsically disordered proteins (IDPs) which lack a fully defined three-dimensional structure are the most likely precursors to today's proteins because of their flexible conformation and functional diversity. But how have these proteins evolved into often rigid and highly specialized protein structures? This evolutionary trajectory has the greatest support in the theory of induced fold whereby the development of the structure was mediated by the interaction and coevolution of primordial unstructured proteins with different cofactors or RNA molecules. Although some random sequences from the sequence space which is not used by nature are also able to form folded proteins the more suitable candidates for evolution of structure and function appear to be random sequences with a high content of disordered which have low aggregation propensity. The selected random protein sequences with high disorder content have been structurally characterized in this work for their further use in evolutionary studies. Three artificial proteins were selected from a random-sequence library based on previous study in our laboratory. In the present work they were purified and...

Les domaines thymosineβ/WH2 sont une famille de protéines intrinsèquement désordonnées impliqués dans le remodelage du cytosquelette d’actine. Ces domaines de 20 à 50 acides aminés existent seuls ou au sein de protéines modulaires, isolés ou répétés. Ils exercent de nombreuses fonctions : ils séquestrent des monomères d’actine, promeuvent l’assemblage du filament, nucléent, fragmentent et coiffent les filaments. Tous les domaines WH2 interagissent de manière similaire avec l’actine via une hélice amphipathique N-terminale suivie d’un brin central et d’une région C-terminal plus ou moins longue et dynamique. Une étude antérieure a montré que la fonction des domaines βT/WH2 isolés était liée à la dynamique du complexe avec l’actine déterminée par une combinaison d’interactions intermoleculaires le long de l’ensemble de la séquence. Les mécanismes expliquant la multifonctionnalité des domaines WH2 répétés restent vagues. Ce travail de thèse présente tout d’abord la production d’actine recombinante, sauvage et mutée dans le système baculovirus/Sf9 pour la biologie structurale ainsi que le développement de stratégies de marquage isotopique en cellules d’insectes. La deuxième partie s’intéresse à la caractérisation structurale et dynamique de domaines WH2 seules en solution : deux domaines isolés et deux protéines contenant deux domaines WH2. Les hélices amphipathiques N-terminales sont partiellement repliées avec des populations variant selon les protéines. La préstructuration des régions C-terminales est plus variable, complètement désordonnée ou partiellement hélicoïdale selon les protéines. La dernière partie présente l’étude de l’interaction de ces protéines avec l’actine. / WH2 domains are a family of intrinsically disordered proteins involved in actin cytoskeleton remodeling. These short domains, isolated or repeated in various actin binding proteins display a low sequence identity and a large panel of functions such as sequestration of actin monomers, promotion of unidirectional assembly, nucleation, fragmentation, filament capping. All WH2 domains fold similarly upon actin binding. They form an extended interface along actin, with an amphipathic N-terminal helix followed by an extended central strand and a more dynamic C-terminal region. Previous work on single βT/WH2 domains showed that function was linked to the dynamics of the complex with actin which is determined by a combination of intermolecular interactions throughout the sequence. The multifunctionality of WH2 tandem repeats is still elusive. The present work first describes production of recombinant wild-type and mutant actin in insect cells and isotopic 15N-labeling for NMR spectroscopy. As a first step to gain insight into the folding upon binding mechanism of functionally different WH2 repeats, we investigated the conformational behavior of two single domains and two tandem repeats free in solution by NMR. The N-terminal amphipatic helix is partially formed but with various propensities depending on the proteins while the C-terminal region that may form an helix in the complex may be either completely disordered or partially formed in absence of actin. Investigation of WH2:actin interaction for the same four proteins is described in the last chapter.

La phosphoribulokinase (PRK) et la glycéraldéhyde 3-phosphate déshydrogénase (GAPDH) sont deux enzymes-clés du cycle de Calvin-Benson. Leurs activités sont régulées par l’intermédiaire de la CP12, une protéine intrinsèquement désordonnée. Au cours de la transition lumière-obscurité, la GAPDH, la CP12 et la PRK forment un complexe supramoléculaire au sein duquel l’activité des enzymes est inhibée. Dans les travaux présentés ici, nous nous sommes intéressés à la formation de ce complexe et à la dynamique de ses composants. Nous avons montré pour la première fois que les résidus cystéine Cys243 et Cys249 de la PRK sont essentiels à la formation du complexe GAPDH-CP12-PRK et qu’ils peuvent former un pont disulfure en présence de CP12. Nous avons également étudié la dynamique de la CP12 en présence de ses partenaires, et observé que la CP12 adopte une conformation beaucoup plus compacte en présence de GAPDH et de PRK. La glutathionylation (formation d’un pont disulfure mixte entre une molécule de glutathion et un résidu cystéine appartenant à une protéine) est une modification post-traductionnelle associée au stress oxydant qui affecte dix enzymes du cycle de Calvin-Benson, y compris la GAPDH et la PRK. Nous avons étudié l’impact de la glutathionylation sur ces enzymes, et montré que l’inactivation de la PRK naît de l’encombrement du site de fixation de l’ATP.Enfin, la dernière partie de ces travaux est centrée sur l’adénylate kinase 3 de C. reinhardtii, une enzyme impliquée dans le métabolisme de l’ATP et qui possède une extension similaire à la CP12. Cette première étude montre que cette extension augmente la stabilité de l’ADK 3 et intervient dans sa glutathionylation. / Phosphoribulokinase (PRK) and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) are two key enzymes of the Calvin-Benson and their activities are redox-regulated through the intervention of CP12, a intrinsically disordered protein. During the light-to-dark transitions, GAPDH, CP12 and PRK form a supramolecular complex in which the enzymes are strongly inhibited; this complex is dissociated during the dark-to-light transition and the active enzymes are released.In the work presented here, we studied the formation of the complex and the dynamics of its components. For the first time, we showed that two cysteine residues of PRK, Cys243 and Cys249, are essential to the assembly of the GAPDH-CP12-PRK complex, and can form a disulfide bridge in presence of CP12.Glutathionylation (the formation of a mixed disulfide bridge linking one glutathione molecule and a cysteine residue from a protein) is a post-translational modification associated with oxidative stress that affects ten of the Calvin-Benson enzymes, including GAPDH and PRK, and we show that the inactivation of PRK by glutathionylation is caused by the blockage of the ATP binding site by glutathione.The last part of this work is centered around adenylate kinase 3 from C. reinhardtii, an enzyme tied to the energetic metabolism of the cells that presents a CP12-like C-terminal extension. Our results suggest that this CP12-like “tail” improve the stability of ADK 3 and participates in tis glutathionylation.