The role of the Type IV pili system in the virulence of Francisella tularensis

Abstract: Francisella tularensis is a Gram-negative intracellular pathogen causing the zoonotic disease tularemia. F. tularensis can be found almost all over the world and has been recovered from several animal species, even though the natural reservoir of the bacterium and parts of its life cycle are still unknown. Humans usually get infected after handling infected animals or from bites of blood-feeding arthropod vectors. There are four subspecies of F. tularensis: the highly virulent tularensis (Type A) that causes a very aggressive form of the disease, with mortality as high as 60% if untreated, the moderately virulent holarctica (Type B) and mediasiatica, and the essentially avirulent subspecies F. novicida. So far, our knowledge of the molecular mechanisms that would explain these differences in virulence among the subspecies is poor. However, recent developments of genetic tools and access to genomic sequences have laid the ground for progress in this research field. Analysis of genome sequences have identified several regions that differ between F. tularensis subspecies. One of these regions, RD19, encodes proteins postulated to be involved in assembly of type IV pili (Tfp), organelles that have been implicated in processes like twitching motility, biofilm formation and cell-to-cell communication in pathogenic bacteria. While there have been reports of pili-like structures on the surface of F. tularensis, these have not been linked to the Tfp encoding gene clusters until now. Herein, I present evidence that the Francisella pilin, PilA, can complement pilin-like characteristics and promote assembly of fibers in a heterologous system in Neisseria gonorrhoeae. pilA was demonstrated to be required for full virulence of both type A and type B strains in mice when infected via peripheral routes. A second region, RD18, encoding a protein unique to F. tularensis and without any known function, was verified to be essential for virulence in a type A strain. Interestingly, the non-licensed live vaccine strain, LVS (Type B), lacks both RD18 and RD19 (pilA) due to deletion events mediated by flanking direct repeats. The loss of RD18 and RD19 is responsible for the attenuation of LVS, since re-introducing them in cis could restore the virulence to a level similar to a virulent type B strain. Significantly, these deletion events are irreversible, preventing LVS to revert to a more virulent form. Therefore, this important finding could facilitate the licensing of LVS as a vaccine against tularemia.