Most of bacteria have a single circular chromosome. During replication of DNA, this circularity can lead to two sister chromatids topologically linked (catenanes and dimers). These topological links have to be solved in order to allow good segregation of genetic information between the two daughter cells during cell division. Bacteria possess a highly conserved machinery: the tyrosine recombinases XerC XerD that are capable to resolve dimers and some catenanes, by catalyzing a crossover at the specific site dif located in the Ter region of the chromosome. During this process they realize two sequentialstrand exchanges.The Xer reaction is spatiotemporally controlled by a protein of the divisome: FtsK. FtsK is a pump that translocates DNA through the septum of division. When FtsK meets a synapse that consists of two dif loaded by XerC and XerD, it activates XerD catalysis that initiates first strand exchange. Secondly XerC catalyzes a second strand exchange independently of FtsK. To date the activation mechanism of XerD is not well understood. Some mobile elements solve their multimeric states (like plasmids) or integrate their genome into the chromosome of their host by using XerCD recombinases. Such integrative elements are named IMEXs (Integrative Mobile Element using Xer). The mobile elements studied before my thesis all used recombination pathways initiated by catalysis of XerC and not requiring activation of XerD .During my PhD I studied at first the integration mechanism / excision of a new class IMEXs using as a model the TLC phage Vibrio cholerae, the bacterium responsible for cholera. By genetic approaches I demonstrated that TLCphi uses a recombination pathway initiated by XerD catalysis and independently of FtsK. My work has also shown that the phage excision participates in the evolution of pandemic strains of V. cholerae. In the second part, I identified a phage factor that allows TLC to bypass the activation of XerD by FtsK. This factor was a protein of unknown function with a HTH domain and a DUF3653 domain. DUF3653 are found in many IMEXs. Using molecular biology approaches, I studied the mechanism of action of this protein. I reproduced the recombination reaction in vitro and demonstrated that this factor activates XerD by directly interacting with it. Finally, we were interested to study disparities between Xer recombination in E.coli and V.cholerae. In particular, the Xer recombination seems to act only on dimers in E.coli while it is also active on monomers in V.cholerae. We have demonstrated that these differences in behaviors do not come from Xer themselves or their activation by FtsK. They result from different choreographies of chromosome segregation between these two bacteria and are also dependent on growth rates.