Bacterial factors drive the differential targeting of Guanylate Binding Proteins to Francisella and Shigella

Guanylate-Binding Proteins (GBPs) are interferon-inducible GTPases that play a key role in cell autonomous responses against intracellular pathogens. Seven GBPs are present in humans. Despite sharing high sequence similarity, subtle differences among GBPs translate into functional divergences that are still largely not understood. A key step for the antimicrobial activity of GBPs towards cytosolic bacteria is the formation of supramolecular GBP complexes on the bacterial surface. Such complexes are formed when GBP1 binds lipopolysaccharide (LPS) from Shigella and Salmonella and further recruits GBP2, 3, and 4. Here, we investigated GBPs recruitment on Francisella novicida, a professional cytosol-dwelling pathogen with an atypical tetra-acylated LPS. Co-infection experiments demonstrated that GBPs target preferentially S. flexneri compared to F. novicida. F. novicida was coated by GBP1 and GBP2 in human macrophages but escaped targeting by GBP3 and GBP4. GBP1 and GBP2 features that drive recruitment to F. novicida were investigated revealing that GBP1 GDPase activity is required to initiate GBP recruitment to F. novicida but facultative to target S. flexneri. Furthermore, analysis of chimeric GBP2/5 proteins identified a central domain in GBP2 necessary and sufficient to target F. novicida. Finally, a F. novicida {Delta}lpxF mutant with a penta-acylated lipid A was targeted by GBP3 suggesting that lipid A tetra-acylation contributes to escape from GBP3. Altogether our results indicate that GBPs have different affinity for different bacteria and that the repertoire of GBPs recruited onto cytosolic bacteria is dictated by GBP-intrinsic features and specific bacterial factors, including the structure of the lipid A. IMPORTANCEFew bacteria have adapted to thrive in the hostile environment of the cell cytosol. As a professional cytosol-dwelling pathogen, S. flexneri secretes several effectors to block cytosolic immune effectors, including GBPs. This study illustrates a different approach of adapting to the host cytosol: the stealth strategy developed by F. novicida. F. novicida bears an atypical hypoacylated LPS, which does not elicit neither TLR4 nor caspase-11 activation. Here, this atypical LPS was shown to promote escape from GBP3 targeting. Furthermore, the lower affinity of GBPs for F. novicida allowed to decipher the different domains that govern GBP recruitment to the bacterial surface. This study illustrates the importance of investigating different bacterial models to broaden our understanding of the intricacies of host-pathogen interactions.