tularensis in macrophages. While H3 and H4 were located in regions of little importance for VipB binding, H1

and H2 overlapped with regions crucial for the interaction. Although the F. tularensis T6SS is phylogenetically only distantly related to other T6SSs, domains structurally very similar to the four Temsirolimus research buy helices with the same specific locations were predicted in an extensive number of homologues of other Gram-negative bacteria. These structural similarities also correlated to a functional relationship, as evidenced by our demonstration of both native and heterologous interactions between the A-B homologues of 6 Gram-negative bacteria, including Vibrio, despite rather low levels of amino acid identities. Thus, the evidence indicates that the H2, and possibly also the H1, helices are essential for the formation of the A/B complex due to the strong preservation of these structures despite different evolutionary DAPT molecular weight origins. In view of this background, we wanted to further

characterize the previously identified interaction of the H2 helix of VipA using a targeted mutagenesis approach. Residues within the conserved α-helix of VipA were exchanged to alanine and the resulting mutants tested in a B2H system. By this approach, several residues important for the VipB interaction were identified, i.e. D104, V106, V110, P111 and L113. Interestingly, out of these, V106, V110 and L113 were homologous to the residues V105, V109 and I112 respectively of the F. tularensis

homologue IglA, which when mutated resulted in diminished IglB binding [6]. This confirms that the mechanism behind A/B complex formation is conserved in distantly related pathogens. The small but consistent defect in VipB-binding, however, had no visible effect on VipB expression/stability or Hcp secretion in vitro, although many mutants D104A, V110A and L113A were all less efficient at competing with E. coli when tested in a bacterial competition assay. These results resemble those obtained with IglA, for which mutants V109A and L115A showed a defect in IglB binding, but not on IglB stability, yet both mutants were completely unable to grow within host cells and were also avirulent in mice [6]. Thus, even subtle defects in the A-B interaction have drastic impact on the competitive ability of T6S-containing pathogens, as well as on their ability to successfully infect host cells. By combining two or more of the single substitutions that resulted in a defect in VipB-binding, an additive effect was apparent; the ability to interact with VipB binding was poor or abolished in both B2H and Y2H systems, and similarly to a vipA null mutant, these multiple substitution mutants were unable to support stable VipB, Hcp secretion, and to compete with E. coli in a bacterial competition assay. This is the first time that this type of systematic mapping has been carried out in Vibrio.