Universiẗat, Diss., 2001--Mainz. / Auch als gedr. Diss.

La publication du génome de la paramécie (Aury, 2006) a révélé une séquence atypique particulièrement intéressante pour les études de génomique évolutive. Au cours de cette thèse, j’ai mené une analyse bioinformatique détaillée de ce génome en me concentrant particulièrement sur les trois points suivants : 1) Le rôle de deux classes distinctes de petits ARN fonctionnels non codants, l’une intervenant dans les processus de régulation de l’expression des gènes tandis que l’autre participe aux réarrangements génomiques (élimination de fragments d’ADN) associés au cycle sexuel de la paramécie. 2) L’évolution des paires de gènes après une duplication globale de génome (WGD). En effet, avec une WGD relativement récente précédée de deux autres WGDs plus anciennes encore bien visibles, la paramécie est un modèle de choix pour cette étude. Nous avons pu montrer que la rétention des deux copies d’un gène après une WGD est fortement corrélée au niveau d’expression des gènes. Nous proposons un modèle basé sur les coûts et bénéfices de l’expression des gènes pour expliquer cette observation. 3) L’analyse de contraintes sélectives sur les introns pour produire des messagers détectables par le Nonsense-Mediated mRNA Decay (NMD). Ces contraintes sélectives, mises en évidence initialement chez la paramécie, se sont avérées être présentes chez tous les eucaryotes que nous avons pu analyser, ce qui nous amène à questionner l’efficacité des mécanismes d’épissage chez les eucaryotes et le rôle du NMD dans la prévention des erreurs d’épissage. L’ensemble de ces analyses a permis de mieux comprendre un certain nombre de mécanismes évolutifs universels / This work presents a detailed analysis of the paramecium genome, with focusing more precisely on the 3 following topics : 1) The role of two distinct classes of small non-coding RNAs. The first one (siRNAs) being involved in post-transcriptional gene silencing while the other (scanRNAs) plays a crucial role during the massive genomic rearrangements that occur in ciliates after sexual reproduction (Lepère et al. 2009). 2) The evolution of duplicated genes following Whole-Genome Duplications (WGDs). Indeed, the paramecium genome contains evidences for 3 successive WGDs (Aury et al. 2006), which explains why this organisms is perfectly well suited for such an analysis. We show that retention of duplicated genes is strongly correlated to their expression level and we propose a model based on cost and benefit of gene expression to explain this pattern. 3) The analysis of the extremely tiny introns in paramecium (99% of introns are less than 20-33nt in length) revealed the presence of a translational control of splicing in eukaryotes. This work suggests that splicing errors are frequent and that eukaryotic cells rely on the Nonsense-mediated mRNA Decay to detect aberrant transcripts produced by splicing errors (Jaillon et al. 2008). These analyses provide new insights on several evolutionary mechanisms that shape the genomes of eukaryotes

Cilia and flagella are hair-like appendages of eukaryotic cells. They are actively bending structures that exhibit regular beat patterns and thereby play an important role in many different circumstances where motion on a cellular level is required. Most dramatic is the effect of nodal cilia whose vortical motion leads to a fluid flow that is directly responsible for establishing the left-right axis during embryological development in many vertebrate species, but examples range from the propulsion of single cells, such as the swimming of sperm, to the transport of mucus along epithelial cells, e.g. in the ciliated trachea. Cilia and flagella contain an evolutionary highly conserved structure called the axoneme, whose characteristic architecture is based on a cylindrical arrangement of elastic filaments (microtubules). In the presence of a chemical fuel (ATP), molecular motors (dynein) exert shear forces between neighbouring microtubules, leading to a bending of the axoneme through structural constraints. We address the following two questions: How can these organelles generate regular oscillatory beat patterns in the absence of a biochemical signal regulating the activity of the force generating elements? And how can the beat patterns be so different for apparently very similar structures? We present a theoretical description of the axonemal structure as an actively bending elastic cylinder, and show that in such a system bending waves emerge from a non-oscillatory state via a dynamic instability. The corresponding beat patterns are solutions to a set of coupled partial differential equations presented herein.

Cilia and flagella are hair-like appendages of eukaryotic cells. They are actively bending structures that exhibit regular beat patterns and thereby play an important role in many different circumstances where motion on a cellular level is required. Most dramatic is the effect of nodal cilia whose vortical motion leads to a fluid flow that is directly responsible for establishing the left-right axis during embryological development in many vertebrate species, but examples range from the propulsion of single cells, such as the swimming of sperm, to the transport of mucus along epithelial cells, e.g. in the ciliated trachea. Cilia and flagella contain an evolutionary highly conserved structure called the axoneme, whose characteristic architecture is based on a cylindrical arrangement of elastic filaments (microtubules). In the presence of a chemical fuel (ATP), molecular motors (dynein) exert shear forces between neighbouring microtubules, leading to a bending of the axoneme through structural constraints. We address the following two questions: How can these organelles generate regular oscillatory beat patterns in the absence of a biochemical signal regulating the activity of the force generating elements? And how can the beat patterns be so different for apparently very similar structures? We present a theoretical description of the axonemal structure as an actively bending elastic cylinder, and show that in such a system bending waves emerge from a non-oscillatory state via a dynamic instability. The corresponding beat patterns are solutions to a set of coupled partial differential equations presented herein.