Florian Turbant

Under specific conditions, some proteins can self-assemble into fibrillar structures called amyloids. Initially, these proteins were associated with neurodegenerative diseases in eucaryotes. Nevertheless, they have now been identified in the three domains of life. In bacteria, they are involved in diverse biological processes and are usually useful for the cell. For this reason, they are classified as “functional amyloids”. In this work, we focus our analysis on a bacterial functional amyloid called Hfq. Hfq is a pleiotropic regulator that mediates several aspects of genetic expression, mainly via the use of small noncoding RNAs. Our previous work showed that Hfq amyloid-fibrils interact with membranes. This interaction influences Hfq amyloid structure formation and stability, but the specifics of the lipid on the dynamics of this process is unknown. Here, we show, using spectroscopic methods, how lipids specifically drive and modulate Hfq amyloid assembly or, conversely, its disassembly. The reported effects are discussed in light of the consequences for bacterial cell life.

Due to their two-cell membranes, Gram-negative bacteria are particularly resistant to antibiotics. Recent investigations aimed at exploring new target proteins involved in Gram-negative bacteria adaptation helped to identify environmental changes encountered during infection. One of the most promising approaches in finding novel targets for antibacterial drugs consists of blocking noncoding RNA-based regulation using the protein cofactor, Hfq. Although Hfq is important in many bacterial pathogens, its involvement in antibiotics response is still unclear. Indeed, Hfq may mediate drug resistance by regulating the major efflux system in Escherichia coli, but it could also play a role in the influx of antibiotics. Here, using an imaging approach, we addressed this problem quantitatively at the single-cell level. More precisely, we analyzed how Hfq affects the dynamic influx and efflux of ciprofloxacin, an antibiotic from the group of fluoroquinolones that is used to treat bacterial infections. Our results indicated that the absence of either whole Hfq or its C-terminal domain resulted in a more effective accumulation of ciprofloxacin, irrespective of the presence of the functional AcrAB-TolC efflux pump. However, overproduction of the MicF small regulatory RNA, which reduces the efficiency of expression of the ompF gene (coding for a porin involved in antibiotics influx) in a Hfq-dependent manner, resulted in impaired accumulation of ciprofloxacin. These results led us to propose potential mechanisms of action of Hfq in the regulation of fluoroquinolone fluxes across the E. coli envelope.

The possible carrier role of Outer Membrane Vesicles (OMVs) for small regulatory noncoding RNAs (sRNAs) has recently been demonstrated. Nevertheless, to perform their function, these sRNAs usually need a protein cofactor called Hfq. In this work we show, by using a combination of infrared and circular dichroism spectroscopies, that Hfq, after interacting with the inner membrane, can be translocated into the periplasm, and then be exported in OMVs, with the possibility to be bound to sRNAs. Moreover, we provide evidence that Hfq interacts with and is inserted into OMV membranes, suggesting a role for this protein in the release of sRNA outside the vesicle. These findings provide clues to the mechanism of host–bacteria interactions which may not be defined solely by protein–protein and protein–outer membrane contacts, but also by the exchange of RNAs, and in particular sRNAs.