The nucleus of mammalian cells has been proven to be highly organised. A recent study on interphase chromosome positioning has identified low serum induced rapid chromosome repositioning. Chromosome 10 initially localised at an intermediate position in normal proliferating human dermal fibroblasts (HDF) was found to relocate to the nuclear periphery 15 minutes after the cells have been incubated in low serum. Whereas chromosome X has remained in a peripheral position. The relocation of chromosome 10 has been shown to be dependant on both actin and myosin functions. In this project we have further investigated the possible role of nuclear myosin I in chromosome 10 repositioning. Using siRNA to block the expression of the nuclear myosin I (NMI) we were able to identify this nuclear myosin as necessary for the rapid repositioning of chromosome 10. Furthermore, using image analysis software we investigated the effect of the NMI knock down on the overall nuclear size and shape. The analysis has revealed that while the nuclear size of normal proliferating cells remained unchanged after the low serum incubation both in cells expressing the NMI and NMI depleted cells, the knock down of the NMI seems to have affected the nuclear shape when the cells were subjected to the serum incubation. On the other hand, the analysis of the chromosome territories area has revealed significant differences in the chromosome territories sizes before and after the low serum incubation, in normal proliferating HDF cells .

L'analyse de l'organisation à grande échelle des chromosomes, par des approches d'imagerie et de biologie moléculaire, constitue un enjeu important de la biologie. Il est maintenant établi que l'organisation structurelle du génome est un facteur déterminant dans tous les aspects des " transactions " génomiques: transcription, recombinaison, réplication et réparation de l'ADN. Bien que plusieurs modèles aient été proposés pour décrire l'arrangement spatial des chromosomes, les principes physiques qui sous-tendent l'organisation et la dynamique de la chromatine dans le noyau sont encore largement débattus. Le noyau est le compartiment de la cellule dans lequel l'ADN chromosomique est confiné. Cependant, la mesure quantitative de l'influence de la structure nucléaire sur l'organisation du génome est délicate, principalement du fait d'un manque d'outils pour déterminer précisément la taille et la forme du noyau. Cette thèse est organisée en deux parties: le premier axe de mon projet était d'étudier la dynamique et les propriétés physiques de la chromatine dans le noyau de la levure S. cerevisiae. Le deuxième axe visait à développer des techniques pour détecter et quantifier la forme et la taille du noyau avec une grande précision. Dans les cellules de levure en croissance exponentielle, j'ai étudié la dynamique et les propriétés physiques de la chromatine de deux régions génomiques distinctes: les régions codant les ARN ribosomiques regroupés au sein d'un domaine nucléaire, le nucléole, et la chromatine du nucléoplasme. Le mouvement de la chromatine nucléoplasmique peut être modélisé par une dynamique dite de " Rouse ". La dynamique de la chromatine nucléolaire est très différente et son déplacement caractérisé par une loi de puissance d'exposant ~ 0,7. En outre, nous avons comparé le changement de la dynamique de la chromatine nucléoplasmique dans une souche sauvage et une souche porteuse d'un allèle sensible à la température (ts) permettant une inactivation conditionnelle de la transcription par l'ARN polymérase II. Les mouvements chromatiniens sont beaucoup plus importants après inactivation transcriptionnelle que dans la souche témoin. Cependant, les mouvements de la chromatine restent caractérisés par une dynamique dite de " Rouse ". Nous proposons donc un modèle biophysique prenant en compte ces résultats : le modèle de polymère dit "branched-Rouse". Dans la deuxième partie, j'ai développé "NucQuant", une méthode d'analyse d'image permettant la localisation automatique de la position de l'enveloppe nucléaire du noyau de levures. Cet algorithme comprend une correction post-acquisition de l'erreur de mesure due à l'aberration sphérique le long de l'axe Z. "NucQuant" peut être utilisée pour déterminer la géométrie nucléaire dans de grandes populations cellulaires. En combinant " NucQuant " à la technologie microfluidique, nous avons pu estimer avec précision la forme et la taille des noyaux en trois dimensions (3D) au cours du cycle cellulaire. "NucQuant" a également été utilisé pour détecter la distribution des regroupements locaux de complexes de pore nucléaire (NPCs) dans des conditions différentes, et a révélé leur répartition non homogène le long de l'enveloppe nucléaire. En particulier, nous avons pu montrer une distribution particulière sur la région de l'enveloppe en contact avec le nucléole. En conclusion, nous avons étudié les propriétés biophysiques de la chromatine, et proposé un modèle dit "branched Rouse-polymer" pour rendre compte de ces propriétés. De plus, nous avons développé "NucQuant", un algorithme d'analyse d'image permettant de faciliter l'étude de la forme et la taille nucléaire. Ces deux travaux combinés vont permettre l'étude des liens entre la géométrie du noyau et la dynamique de la chromatine. / Chromosome high-order architecture has been increasingly studied over the last decade thanks to technological breakthroughs in imaging and in molecular biology. It is now established that structural organization of the genome is a key determinant in all aspects of genomic transactions. Although several models have been proposed to describe the folding of chromosomes, the physical principles governing their organization are still largely debated. Nucleus is the cell’s compartment in which chromosomal DNA is confined. Geometrical constrains imposed by nuclear confinement are expected to affect high-order chromatin structure. However, the quantitative measurement of the influence of the nuclear structure on the genome organization is unknown, mostly because accurate nuclear shape and size determination is technically challenging. This thesis was organized along two axes: the first aim of my project was to study the dynamics and physical properties of chromatin in the S. cerevisiae yeast nucleus. The second objective I had was to develop techniques to detect and analyze the nuclear 3D geomtry with high accuracy. Ribosomal DNA (rDNA) is the repetitive sequences which clustered in the nucleolus in budding yeast cells. First, I studied the dynamics of non-rDNA and rDNA in exponentially growing yeast cells. The motion of the non-rDNA could be modeled as a two-regime Rouse model. The dynamics of rDNA was very different and could be fitted well with a power law of scaling exponent ~0.7. Furthermore, we compared the dynamics change of non-rDNA in WT strains and temperature sensitive (TS) strains before and after global transcription was actived. The fluctuations of non-rDNA genes after transcriptional inactivation were much higher than in the control strain. The motion of the chromatin was still consistent with the Rouse model. We propose that the chromatin in living cells is best modeled using an alternative Rouse model: the “branched Rouse polymer”. Second, we developed “NucQuant”, an automated fluorescent localization method which accurately interpolates the nuclear envelope (NE) position in a large cell population. This algorithm includes a post-acquisition correction of the measurement bias due to spherical aberration along Z-axis. “NucQuant” can be used to determine the nuclear geometry under different conditions. Combined with microfluidic technology, I could accurately estimate the shape and size of the nuclei in 3D along entire cell cycle. “NucQuant” was also utilized to detect the distribution of nuclear pore complexes (NPCs) clusters under different conditions, and revealed their non-homogeneous distribution. Upon reduction of the nucleolar volume, NPCs are concentrated in the NE flanking the nucleolus, suggesting a physical link between NPCs and the nucleolar content. In conclusion, we have further explored the biophysical properties of the chromatin, and proposed that chromatin in the nucleoplasm can be modeled as "branched Rouse polymers". Moreover, we have developed “NucQuant”, a set of computational tools to facilitate the study of the nuclear shape and size. Further analysis will be required to reveal the links between the nucleus geometry and the chromatin dynamics.

The nucleus is a mesoscopic system that retains features from both the quantum and macroscopic worlds. A basic property of a macroscopic body is its shape. Nuclear shapes can be deduced from experimental data as they influence the excitation mode of the nucleus and hence the energies and lifetimes of its excited levels. Various short-lived nuclei were created in fusion-evaporation experiments performed at international heavy-ion accelerator facilities. The emitted γ rays and, in some experiments, also the charged particles and neutrons emitted in the reactions were detected. The studied neutron-deficient isotopes were either selected by the type and number of particles emitted in the reactions, or by using their characteristic decays. The excited states of the different isotopes were extracted from the γ-ray analyses. Spectroscopic properties, such as the lifetimes of the excited states or the angular distribution of the emitted γ rays were measured when possible. The experimentally obtained level schemes together with the other spectroscopic information were used to deduce the excitation modes and the shapes of the studied nuclei. The detector systems are described in the first chapter and in the second chapter some techniques used to extract information from the experimental data are explained. Finally, a brief theoretical overview on the nuclear models which were used to interpret the experimental results is given. / QC 20100621

No description available.