Séminaire MSC
"Matière et Systèmes Complexes"


Lundi 27 avril 2009 à 11h30
Bâtiment Condorcet, 4ème étage, salle 454 A.

Christophe Zimmer
Computational Imaging & Modeling Group, Institut Pasteur, Paris)

Mapping nuclear organization in living cells by computational imaging

The spatial position of chromosomes and genes inside eukaryotic cell nuclei is not random and plays a role in fundamental biological processes, including gene expression, DNA repair and replication. However, despite its functional importance, the spatial organization of the genome in living cells is still largely uncharted. This is due to several technical limitations, including the diffraction limited resolution of optical microscopy and a lack of analysis tools. To better describe and understand nuclear organization, we have developed an imaging and analysis approach that allows to create high-resolution probability maps of chromosomal loci within the living yeast nucleus.
This is achieved by automated processing of microscopy images containing thousands of cells, combined with fluorescent tagging of two nuclear landmarks. In contrast to the images themselves, the resolution of these probability maps is not limited by diffraction. The technique thereby allowed us to visualize a strong spatial organization, with several genes confined to probabilistic territories significantly smaller than the nucleus. We could observe remodeling of galactose gene territories during transcriptional activation, but also spatial segregation of coregulated genes.
Our data on various loci including many telomeres quantitatively describe the Rabl-configuration of yeast chromosomes. Loci positions strongly correlate with genomic location, suggesting that the spatial organization of the yeast genome may be predictable by relatively simple physical models.