Current single molecule localisation microscopy methods allow for multicolour imaging of macromolecules in cells, and for a degree quantification on molecule numbers in one colour. However, that has not yet been an attempt to develop tools capable of quantitative imaging with multiple colours in cells. This work addressed this challenge by designing linker peptides with chemospecific groups to allow attachment of activator and emitter dyes for STORM imaging, and a targeting module. The design ensured a stoichiometric ratio of targeting module to activator and emitter dyes. Peptides with HaloTag ligands attached were labelled with various activator and emitter pairs and used to label HaloTag fusions of S. pombe and mouse embryonic stem cells. These peptides were found to bind non-specifically to various areas of both cell types, and did not localise to HaloTag protein, whereas controls did. Another peptide was also labelled with activator-emitter pairs and attached to expressed anti-GFP and ant-mCherry nanobodies via native chemical ligation. The labelled anti-GFP nanobody was to demonstrate ensemble and single molecule imaging in S. pombe, as well as characterisation on single molecule surfaces in comparison to a conventional randomly labelled antibody. The stoichiometrically labelled nanobody had a more consistent number of photons detected per localisation, number of localisation per molecule and number of blinks per molecule, which implied that it could be more useful than randomly labelled nanobodies for counting experiments. It was also shown to be capable of specific laser activation for STORM imaging with both an Alexa405Cy5 and Cy3Cy5 pairs. These anti-GFP and anti-mCherry nanobodies and peptide linker are new tools for both counting and multicolour imaging in super-resolution, which could be widely applied to constructs that are already tagged with GFP or mCherry.

[pt] Defensinas de planta, componentes inatos do sistema imune das plantas, são peptídeos antifúngicos, catiônicos, com estrutura primária rica em cisteína. Evidência dada pela literatura demonstrou que trechos de esfingolipídios complexos na membrana dos fungos, contendo manosildiinositolfosforilceramida e glicosilceramida, são sítios de ligação seletivos para as defensinas de planta isoladas de Dahlia merckii e Raphanus sativus, respectivamente. Entretanto, desconhece-se se as defensinas de planta interagem direta ou indiretamente com alvos intracelulares dos fungos. A fim de identificar interações físicas e diretas do tipo proteína- proteína, um sistema de duplo-híbrido, em levedura, baseado no fator de transcrição GAL4, foi construído utilizando-se como isca, a defensina da planta Pisum sativum, Psd1 (Pisum sativum defensin 1). Proteínas alvos, capazes de interagirem com o peptídeo Psd1, foram detectadas através do rastreamento de uma biblioteca de cDNA do fungo Neurospora crassa. Do resultado deste rastreamento, nove dentre quinze candidatos, selecionados pelo método do duplo-híbrido, foram identificados como proteínas nucleares da N. crassa. Um clone, detectado com alta freqüência neste rastreamento, apresentou homologia de seqüência com a proteína ciclina F, relacionada com o controle do ciclo celular. O ensaio de co-purificação utilizando a proteína conjugada a glutationa S-transferase (GST) validou in vitro o resultado obtido pelo sistema duplohíbrido. Análise por microscopia de fluorescência da Psd1, conjugada a FITC, e, dos núcleos do fungo Fusarium solani, marcados com DAPI, demonstrou in vivo a co-localização da defensina de planta Psd1 com os núcleos do fungo. Para pesquisar o modo de ação da Psd1 ao nível do ciclo celular, utilizou-se o modelo multicelular da retina de ratos neonatais, em desenvolvimento. Neste modelo, a migração nuclear intercinética, correlacionada com as transições de fase de S para M do ciclo celular, foi observada na presença da Psd1. Verificouse que Psd1 impediu a migração nuclear em neuroblastos, parando o ciclo celular na transição de S para G2. Estes resultados revelaram modos de ação da defensina de planta Psd1 sobre a fisiologia nuclear. / [en] Plant defensins, innate components of the plant immune system, are cationic, antifungal peptides, with a cysteine- rich primary structure. Evidence from the literature demonstrated that fungus membrane patches containing complex sphingolipids, mannosyldiinositolphosphorylceramide and glucosylceramides, are selective binding sites for the plant defensins isolated from Dahlia merckii and Raphanus sativus, respectively. However, whether the plant defensins interact directly or indirectly with fungus intracellular targets is unknown. To identify direct physical protein-protein interactions, a GAL4-based yeast two-hybrid system was constructed, using the plant peptide, Pisum sativum defensin 1 (Psd1), as the bait protein. Target proteins, capable of interacting with the bait Psd1, were detected by screening a Neurospora crassa cDNA library. In this screening, nine out of fifteen two-hybrid candidates were identified as N. crassa nuclear proteins. One clone, detected with high frequency in the screening, presented sequence similarity to a N. crassa cyclin F, related to the cell cycle control. The GST pull- down co purification assay corroborated this two-hybrid result in vitro. Fluorescence microscopy analysis of FITC- conjugated Psd1 and DAPI-stained Fusarium solani nuclei demonstrated in vivo the co-localization of the plant peptide Psd1 and the fungus nuclei. We used the developing retina of neonatal rats as a multicellular model to study Psd1 mode of action at the cell cycle level. In this model, we observed in vivo the interkinetic nuclear migration, correlated to the transitions from S to M-phase of the cell cycle, in the presence of the Psd1 peptide. It was shown that Psd1 impaired nuclear migration of neuroblasts by arresting the cell cycle at the S to G2- phase transition. These results revealed modes of action of the plant defensin Psd1 upon the nuclear physiology.

Coating colors are applied to the base paper in order to maximize the performance of the end product. Coating colors are complex colloidal systems, mainly consisting of water, binders, and pigments. To understand the behavior of colloidal suspensions, an understanding of the interactions between its components is essential.

Actin is one of the most abundant and highly conserved proteins in eukaryotic cells. The globular protein assembles into long filaments, which form a variety of different networks within the cytoskeleton. The dynamic reorganization of these networks - which is pivotal for cell motility, cell adhesion, and cell division - is based on cycles of polymerization (assembly) and depolymerization (disassembly) of actin filaments. Actin binds ATP and within the filament, actin-bound ATP is hydrolyzed into ADP on a time scale of a few minutes. As ADP-actin dissociates faster from the filament ends than ATP-actin, the filament becomes less stable as it grows older. Recent single filament experiments, where abrupt dynamical changes during filament depolymerization have been observed, suggest the opposite behavior, however, namely that the actin filaments become increasingly stable with time. Several mechanisms for this stabilization have been proposed, ranging from structural transitions of the whole filament to surface attachment of the filament ends. The key issue of this thesis is to elucidate the unexpected interruptions of depolymerization by a combination of experimental and theoretical studies. In new depolymerization experiments on single filaments, we confirm that filaments cease to shrink in an abrupt manner and determine the time from the initiation of depolymerization until the occurrence of the first interruption. This duration differs from filament to filament and represents a stochastic variable. We consider various hypothetical mechanisms that may cause the observed interruptions. These mechanisms cannot be distinguished directly, but they give rise to distinct distributions of the time until the first interruption, which we compute by modeling the underlying stochastic processes. A comparison with the measured distribution reveals that the sudden truncation of the shrinkage process neither arises from blocking of the ends nor from a collective transition of the whole filament. Instead, we predict a local transition process occurring at random sites within the filament. The combination of additional experimental findings and our theoretical approach confirms the notion of a local transition mechanism and identifies the transition as the photo-induced formation of an actin dimer within the filaments. Unlabeled actin filaments do not exhibit pauses, which implies that, in vivo, older filaments become destabilized by ATP hydrolysis. This destabilization can be identified with an acceleration of the depolymerization prior to the interruption. In the final part of this thesis, we theoretically analyze this acceleration to infer the mechanism of ATP hydrolysis. We show that the rate of ATP hydrolysis is constant within the filament, corresponding to a random as opposed to a vectorial hydrolysis mechanism. / Aktin ist eines der am häufigsten vorkommenden und am stärksten konservierten Proteine in eukaryotischen Zellen. Dieses globuläre Protein bildet lange Filamente, die zu einer großen Vielfalt von Netzwerken innerhalb des Zellskeletts führen. Die dynamische Reorganisation dieser Netzwerke, die entscheidend für Zellbewegung, Zelladhäsion, und Zellteilung ist, basiert auf der Polymerisation (dem Aufbau) und der Depolymerisation (dem Abbau) von Aktinfilamenten. Aktin bindet ATP, welches innerhalb des Filaments auf einer Zeitskala von einigen Minuten in ADP hydrolysiert wird. Da ADP-Aktin schneller vom Filamentende dissoziiert als ATP-Aktin, sollte ein Filament mit der Zeit instabiler werden. Neuere Experimente, in denen abrupte dynamische Änderungen während der Filamentdepolymerisation beobachtet wurden, deuten jedoch auf ein gegenteiliges Verhalten hin: Die Aktinfilamente werden mit der Zeit zunehmend stabiler. Mehrere Mechanismen für diese Stabilisierung wurden bereits vorgeschlagen, von strukturellen Übergängen des gesamten Filaments bis zu Wechselwirkungen der Filamentenden mit dem experimentellen Aufbau. Das zentrale Thema der vorliegenden Dissertation ist die Aufklärung der unerwarteten Unterbrechungen der Depolymerisation. Dies geschieht durch eine Kombination von experimentellen und theoretischen Untersuchungen. Mit Hilfe neuer Depolymerisationexperimente mit einzelnen Filamenten bestätigen wir zunächst, dass die Filamente plötzlich aufhören zu schrumpfen und bestimmen die Zeit, die von der Einleitung der Depolymerisation bis zum Auftreten der ersten Unterbrechung vergeht. Diese Zeit unterscheidet sich von Filament zu Filament und stellt eine stochastische Größe dar. Wir untersuchen daraufhin verschiedene hypothetische Mechanismen, welche die beobachteten Unterbrechungen verursachen könnten. Die Mechanismen können experimentell nicht direkt unterschieden werden, haben jedoch verschiedene Verteilungen für die Zeit bis zur ersten Unterbrechung zur Folge. Wir berechnen die jeweiligen Verteilungen, indem wir die zugrundeliegenden stochastischen Prozesse modellieren. Ein Vergleich mit der gemessenen Verteilung zeigt, dass der plötzliche Abbruch des Depolymerisationsprozesses weder auf eine Blockade der Enden, noch auf einen kollektiven strukturellen Übergang des gesamten Filaments zurückzuführen ist. An Stelle dessen postulieren wir einen lokalen Übergangsprozess, der an zufälligen Stellen innerhalb des Filaments auftritt. Die Kombination von weiteren experimentellen Ergebnissen und unserem theoretischen Ansatz bestätigt die Vorstellung eines lokalen Übergangsmechanismus und identifiziert den Übergang als die photo-induzierte Bildung eines Aktindimers innerhalb des Filaments. Nicht fluoreszenzmarkierte Aktinfilamente zeigen keine Unterbrechungen, woraus folgt, dass ältere Filamente in vivo durch die ATP-Hydrolyse destabilisiert werden. Die Destabilisierung zeigt sich durch die Beschleunigung der Depolymerisation vor der Unterbrechung. Im letzten Teil der vorliegenden Arbeit untersuchen wir diese Beschleunigung mit theoretischen Methoden, um auf den Mechanismus der ATP-Hydrolyse zu schließen. Wir zeigen, dass die Hydrolyserate von ATP innerhalb des Filaments konstant ist, was dem sogenannten zufälligen Hydrolysemechanismus entspricht und im Gegensatz zum sogenannten vektoriellen Mechanismus steht.

Most urinary tract infections in humans and dogs are caused by uropathogenic strains of , and increasing antimicrobial resistance among these pathogens has created a need for a novel approach to therapy. Bacterial iron uptake and metabolism are potential targets for novel antimicrobial therapy, as iron is a limiting factor in . growth during infection. As a trivalent metal of similar atomic size to iron (III), gallium can interact with a wide variety of biomolecules that normally contain or interact with iron. Gallium compounds disrupt bacterial iron metabolism, are known to accumulate at sites of infection and inflammation in mammals, exert antimicrobial activity against multiple bacterial pathogens in vitro, and may be good candidates as novel antimicrobial drugs. We assessed the suitability of orally administered gallium maltolate as a potential new antimicrobial therapy for urinary tract infections by evaluating its distribution into the bladder, its activity against uropathogenic . in vitro, and its pharmacokinetics and efficacy in a mouse cystitis model. Using a novel application of synchrotron-based analytical methods, we confirmed the distribution of gallium to the bladder mucosa and characterized the relationship between iron and gallium distribution in the bladder. In vitro experiments with human and canine uropathogenic . isolates demonstrated that gallium maltolate exerts antimicrobial effects in a time-dependent, bacteriostatic manner. Minimum inhibitory concentrations ranged from 0.144 µmol/mL to >9.20 µmol/mL with a median of 1.15 µmol/mL. Isolates resistant to ampicillin, ciprofloxacin, or with decreased susceptibility to cephalothin were susceptible to the antimicrobial activity of gallium maltolate, suggesting that resistance to conventional antimicrobials does not predict resistance to gallium maltolate. Pharmacokinetic studies in healthy mice and in a mouse model of urinary tract infection confirmed that gallium is absorbed into systemic circulation after oral administration of gallium maltolate. Gallium is slowly eliminated from the body, with a trend toward longer terminal half-lives in blood and bladder for infected mice relative to healthy mice. This study did not reveal any statistically significant effect of infection status on maximum blood gallium concentrations (4.46 nmol/mL, 95% confidence interval 2.75 nmol/mL – 6.18 nmol/mL and 4.80 nmol/mL, 95% confidence interval 2.53 nmol/mL – 7.06 nmol/mL in healthy and infected mice, respectively) or total gallium exposure in blood and kidney as represented by area under the concentration vs. time curves. Gallium exposure in the bladder was significantly greater for mice with urinary tract infections than for healthy mice. The investigation of gallium distribution within tissues represented a novel application of synchrotron-based analytical techniques to antimicrobial pharmacokinetics. Prior to analysing tissue samples, a library of x-ray absorption spectra was assembled for gallium compounds in both the hard and soft x-ray ranges. The suitability of hard x-ray fluorescence imaging and scanning and transmission x-ray microscopy for localizing and speciating trace elements in tissues was subsequently assessed. Of these methods, only hard x-ray microprobe analysis was well-suited to the analysis and was successfully used for this application. This approach confirmed that gallium arrives at the bladder mucosa after oral administration of gallium maltolate. Furthermore, comparison of iron and gallium distribution within the bladder mucosa indicated that these elements are similarly but not identically distributed and that they do not significantly inhibit one another’s distribution. This finding suggests that gallium may be distributed in part via pathways that do not involve iron. Despite the favorable distribution characteristics of gallium and the persistence of gallium in target tissues following the oral administration of gallium maltolate, its efficacy in a mouse model of urinary tract infection was disappointing. In this study, no statistically significant difference was detected between gallium maltolate, ciprofloxacin and sham treatments in their ability to eliminate bacteria in the urinary tracts. The failure of ciprofloxacin treatment to render bladder tissue culture-negative for an organism that is classified as ciprofloxacin-susceptible in vitro is consistent with observations from other research groups. The similar lack of efficacy observed for gallium maltolate may be related to the large gap between minimum inhibitory concentrations observed in vitro and gallium concentrations observed in tissues from treated mice, but may also be related to the small study size if the effect size of gallium maltolate treatment is small. Given the magnitude of the difference between tissue concentrations and minimum inhibitory concentrations, it may not be possible to increase the dose sufficiently to achieve therapeutic concentrations without causing toxicity. While the results of these experiments suggest that orally administered gallium maltolate may not be a reasonable antimicrobial drug candidate for treating urinary tract infections caused by uropathogenic . , it may be useful for other applications. Other bacterial pathogens may be more susceptible to the antimicrobial effects of gallium maltolate, and local or topical administration could produce much higher concentrations than we observed following oral administration. Continued development of the synchrotron-based analytical techniques used in these experiments could provide new and important opportunities to investigate antimicrobial distribution and metabolism within cells and tissues, particularly for metal-based drugs.

No description available.

Doctor of Philosophy / Department of Chemistry / Daniel A. Higgins / This dissertation describes the application of wide-field single-molecule fluorescence microscopy techniques to investigations of DNA-surface interactions on chemically graded organosilane surfaces. The adsorption and desorption behaviors of double-stranded (ds) plasmid DNA along the amino-trimethoxysilane and octyl-trichlorosilane gradients were explored as a function of solution pH, solution ionic strength and surface properties. The results provide an improved fundamental understanding of DNA interactions with different surfaces and are certain to aid in the development and advancement of DNA-based biological and biomedical devices. Three distinct experiments were performed in completion of the work for this dissertation. In the first study, total internal reflection fluorescence (TIRF) microscopy was employed to study DNA interactions with aminosilane gradient surfaces under relatively acidic and basic environments. Electrical potentials were applied to assist DNA adsorption and desorption. The single-molecule data clearly showed that DNA capture and release was achieved on the monolayer and submonolayer coated regions of the aminosilane gradient surface under relatively basic pH conditions, with the help of an electrical potential. Meanwhile, DNA adsorption was found to be quasi-reversible on the multilayers at the high aminosilane end of the gradient in the relatively acidic solution. The results of these studies demonstrate the importance of manipulating the electrostatic interactions of DNA with charged surfaces in order to achieve DNA capture and release. The fundamental knowledge of the DNA-surface interactions gained in these studies will be helpful in diverse fields ranging from the layer-by-layer assembly of polyelectrolyte-based thin films to the selective electronic sensing of charged biomolecules. In the second study, the local dielectric properties of the least polar environments in dsDNA were assessed by using the solvatochromic dye, nile red, as a polarity-sensitive probe. TIRF spectroscopic imaging methods were employed in these studies. Although the dielectric constant within the least polar regions of dsDNA was previously predicted by theoretical and computational methods, no experimental measurements of its value had been reported to date. The results provide important knowledge of the factors governing the polarity of DNA microenvironments to which intercalators bind, and provide vital experimental support for the values determined in computational studies. In the third study, TIRF microscopy and single molecule tracking methods were employed to study DNA interactions with an opposed two-component C8-silane and aminosilane gradient surface as a function of solution pH. The mobility of surface-adsorbed DNA molecules was explored and quantified in these studies. The preliminary results further demonstrated the importance of electrostatic interactions over hydrophobic interactions in governing the adsorption of DNA to surfaces. The mobility of surface-adsorbed DNA was found to be largely independent of position along the two-component gradient. These studies were originally undertaken as a route to observation of cooperative effects that are believed to govern DNA-surface binding. Unfortunately, no clear evidence of cooperative effects was observed at the mixed regions of the two-component gradient surface.

No description available.

Le moteur flagellaire bactérien (BFM) est un complexe moléculaire qui permet aux bactéries de nager dans un milieu liquide. La rotation du moteur est générée à l’interface entre deux éléments clés: les protéines formant le stator (MotA and MoB) et l’anneau C “switching complex” à la base du rotor. Les stators sont des modules du moteur structurellement et fonctionnellement différentiables du reste du moteur, et leurs association et dissociation dynamique autour du rotor contrôle la génération du couple. Quand une protéine fluorescente (PF) est fusionnée à MotB, le moteur est en état de marche mais une réduction générale de la mobilité de la cellule a été observée. La raison précise d’une telle réduction de mobilité n’a pas été étudiée.Le but de cette étude est de comprendre comment la fusion PF de la protéine du stator modifie la génération du couple et le sens de rotation du moteur. C’est particulièrement important car le tag FP se trouve à l’interface entre le stator et le rotor, là où le couple et le changement du sens de rotation sont produits. Trois différentes PFs (eGFP, YPet, Dendra2) ont été fusionnées à la protéine MotB. Malgré la haute similarité de leurs structures, notre analyse a montré que les trois stators fusionnés génèrent des couples différents. Les stators marqués avec YPet produisent un couple moyen similaire au WT (stators sans tag PF), alors que les stators marqués avec eGFP et Dendra2 produisent respectivement 70% et 40% du couple moyen du WT. De plus, les moteurs utilisant les stators fusionnés ont montré des capacités de changement de sens de rotation réduites. Lors d’un changement de sens de rotation, la valeur absolue de la vitesse des moteurs WT ne change pas. Cette “symétrie” de vitesse lors du changement n’apparaît pas avec les moteurs à stators fusionnés et le changement peut être accompagné d’une importante diminution (~30%) de la vitesse absolue.En observant par microcopie TIRF avec détection de molécules uniques, des stators marqués dans un moteur en état de marche, les signaux de fluorescence sont détectés à la membrane comme prévu pour ces protéines, montrant une population de stators diffusant dans celle-ci. Les clusters fluorescents étaient visibles au centre des cellules en rotation, attachés au couvre-glace par une seule flagelle, confirmant que le tag de fluorescence peut être visualisé dans des moteurs en état de marche. Dans un second projet développé dans le laboratoire Bertus Beaumont à TU Delft, en prenant le BFM en tant que système modèle d’évolution expérimentale, sa modularité et son « évolubilité » ont été explorés pour apprendre les détails au niveau moléculaire de l’évolution de ce type de machine. Les stators de E.coli ont été échangés par un set de 21 stators étrangers homologues. L’expérience a révélé que les protéines du stator peuvent être échangées entre espèces de bactéries distantes et certains stators non compatibles peuvent être modifiés positivement par un procédé d’évolution pour devenir fonctionnels. Au cours de cette évolution, les bactéries ont accumulé des mutations avantageuses dans leurs gènes MotA et MotB étrangers, tout particulièrement dans leur domaine fonctionnel. Des mutations identiques dans des stators différents ont été observées, indiquant que l’évolution peut se reproduire. L’analyse fonctionnelle au niveau d’un seul moteur a révélé que ces mutations avantageuses amélioraient la génération du couple et/ou la capacité du moteur à changer de sens. Les investigations détaillées du génotype et du phénotype du BFM modifié par évolution apportés par cette étude, pourraient donner une idée sur la façon dont des machines moléculaires comme le BFM ont évolué, et les effets fonctionnels des mutations bénéfiques qui facilitent l'intégration fonctionnelle. / The bacterial flagellar motor (BFM) is the macromolecular complex which allows bacteria to swim in liquid media. Located at the base of the flagellum, anchored in the cell membrane, this remarkably small (~45nm) yet powerful rotary motor rotates each flagellum of the cell switching between counterclockwise (CCW) and clockwise (CW) direction. The motor rotation is generated at the interface between the two key components of the motor: the stator protein complexes (each composed of 4 MotA and 2 MotB proteins) and the C- ring protein complex at the base of the rotor. The stator complexes are structurally and functionally discernible modules of the motor, and their dynamical association and dissociation around the rotor controls the generation of torque.The first project of this study aims to investigate how the FP tag on the stator protein modifies the torque generation and switching of the motor. This is particularly important because the fluorescent protein tag lies at the interface between stator and rotor, where torque and switching are produced. Three different FPs (eGFP, YPet, Dendra2) were fused to MotB. Interestingly, despite the high similarity of their structures, our analysis revealed that the three fusion stators generate different torque. Furthermore, in the presence of fusion stators, the motor showed significantly impaired switching abilities. When switching direction of the rotation, the absolute value of the speed of WT motors does not change, whereas this symmetry of speed upon switching is not observed in the fusion stator motors, and switching can be accompanied with a significant (~30%) decrease in absolute speed. Both the impaired torque generation and the switching ability were improved by introducing a rigid linker between the stator and the FP tag. Taken together, this study provides a further insight into the dynamics of the stator and rotor interaction at its interface.When the cells carrying the fluorescently labeled stators were observed in a custom made TIRF-fluorescence microscope with single molecule capability, the fluorescence signals were detected as concentrated clusters in the membrane as expected for these membrane proteins around the motors, together with a population of stators diffusing in the membrane. Fluorescent clusters were visible at the center of rotating cells tethered to the glass slide by a single flagellum, confirming that the fluorescent tags can be visualized in functioning motors.In a second project developed in Bertus Beaumont lab at TU Delft, taking BFM as an experimental evolutionary model system, its modularity and evolvability have been explored to learn the molecular details of the evolution of molecular machines. The stators of E.coli have been exchanged by a set of 21 homologue foreign stators. The experiments revealed that the stator proteins can be exchanged between distant bacteria species, and some of the non-compatible stators can be positively modified by evolution to become functional. Those evolved strains accumulated beneficial mutations in their foreign motA and motB genes, especially on their functional domains. Identical mutations in different stators were common, indicating that evolution is repeatable. The functional investigation at the single motor level revealed that those beneficial mutations improved the torque generation and/or the switching ability of the motor. The detailed genotype and phenotype investigations of the evolutionary modified BFM may bring an insight into how molecular machines such as BFM have evolved as well as the functional effects of the beneficial mutations that facilitate functional integration.

Dans la première partie de cette thèse, nous étudions la cinétique de la complexation d'un double brin d'ADN par la protéine NCp7. Pour ce faire, nous étudions l'évolution des propriétés mécaniques de l'ADN au fur et à mesure de sa complexation, en étirant la complexe ADN/NCp7 à l'aide d'un montage de piégeage optique. Nous avons observé que la longueur de persistance du complexe diminue au fur et à mesure de la complexation. En utilisant un modèle statistique décrivant l'évolution de la flexibilité de l'ADN complexé par NCp7. Notre principal résultat est que la fraction//phi de paires de bases ayant réagi n'est pas une fonction linéaire du temps aux faibles //phi. Nous interprétons nos résultats en supposant que l'adsorption de NCp7 sur l'ADN est fortement coopérative. Dans deuxième chapitre, nous décrivons la dynamique de particules sondes dans une suspension vitreuse colloïdale de Laponite. La Laponite est une particule colloïdale discoïdale de 25nm de diamètre et de 0.92 nm d'épaisseur. Nous utilisons une expérience de microscopie en onde évanescente, et suivons le mouvement de particules fluorescentes de latex. Nous imageons ensuite ces particules. Nous montrons que, pour un mouvement possédant une seule échelle de temps caractéristique, elle est simplement une fonction linéaire du temps. Nous obtenons que, quelle que soit leur taille, le mouvement des particules sondes peut être décrit par une succession de deux modes dynamiques, où le mode le plus rapide correspond à la diffusion des particules dans un fluide viscoélastique. / In the first part of this thesis, we study the kinetics of the complexation of a double-stranded DNA byNCp7 protein. To do this, we study the evolution of mechanical properties of DNA and its complexation by stretching the DNA/NCp7 complex with a optical trap. We observed that the persistence length of the complex decreases progressively during the complexation. Using astatistical model we describe the evolution of the flexibility of DNA complexed with NCp7. Our main result is that the fraction phi of base pairs that have reacted is not a linear function of time at low phi.We interpret our results assuming that the adsorption of NCp7 on DNA is highly cooperative. In the second chapter, we describe the dynamics of probe particles in a colloidal glassy suspension of Laponite. Laponite is a colloidal discoidal particle of 25 nm in diameter and 0.92 nm thick. We take advantage of evanescent wave microscopy, and follow the movement of fluorescent latex particles.Then we image these particles. We show that for a movement that has a single characteristic time scale, it is simply a linear function of time. We find that, what ever their size, the motion of probe particles can be described by a succession of two dynamic modes, where the fastest mode corresponds to the diffusion of particles in a viscoelastic fluid.