Abstract

Horizontal gene transfer (HGT) allows rapid exchanges of large genetic elements and is known to play an important role in bacterial evolution and adaptation. Conjugative transfer of genomic islands (GIs) has recently been reported in the opportunistic pathogen Pseudomonas aeruginosa. PAPI-1, one of the largest pathogenicity islands of P. aeruginosa, encodes several putative virulence genes and a major regulator of biofilm formation and antibiotic-resistant traits and was found to be horizontally transferable into strains lacking it. The conjugation of PAPI-1 island transfer is mediated by type IV pilus, which is encoded by ten genes located in PAPI-1. Nevertheless, the acquisition mechanism of PAPI-1 is currently not well understood. The first part of this thesis was aimed at identifying the receptor for conjugative transfer on the bacterial cell surface. Based on previous knowledge on bacterial conjugation, we designed and performed a series of mating experiments and analyzed transfer efficiency between PAPI-1 donor and recipient strains. Our data showed that A-band lipopolysaccharide (LPS) is required to initiate PAPI-1 transfer, supporting the idea that this structure acts as a receptor for conjugative type IV pilus in recipient strains. These results were verified by PAPI-1 transfer inhibition experiments with outer membrane (OM) or LPS preparations. The addition of a low amount of OM or LPS derived from strains producing A-band decreased PAPI-1 transfer efficiency by 80% compared to controls. In the second part, we demonstrated that P. aeruginosa strains which already acquired a copy of PAPI-1 almost completely lost the ability to receive additional copies of the island. Combination of strains with or without PAPI-1 were mated in-pair to investigate the redundancy in PAPI-1 transfer. The surface exclusion of PAPI-1 was characterized by investigating the effects of the addition of OM and LPS derived from strains with or without PAPI-1. In addition, LPS of the different strains were analyzed by western blot using antibodies directed against different parts of the molecule and by testing the in-vitro binding capacity of LPS to pilin protein. All experiments indicated that the strains carrying PAPI-1 produced much less A-band LPS compared to those lacking the island and lost the ability to bind to conjugative pilin. Finally, the screening of a series of mutants highlighted a role for two PAPI-1 genes in an entry exclusion activity, possibly through PAPI-1 island destabilization. This study contributes with a step forward in the understanding of the acquisition of genomic islands in P. aeruginosa, which may be generalized to other gram-negative bacteria and may lead to the future development of new strategies to limit the spread of virulence or resistance functions in populations of pathogenic bacteria.