Flotillins are the principal scaffolding proteins of the bacterial Functional Membrane Microdomains (FMM), equivalent to eukaryotic lipid rafts. The SuperFlotillin project aims to find Flo* gain-of-function mutants of Bacillus subtilis flotillins (FloA and/or FloT) that would organize lipid rafts in a more efficient manner. They would be then used to confine multiple enzymes to perform biological reactions and eventually produce biofuels among others compounds of interest.
To isolate Flo* mutants we first attempted to find a clear phenotype of B. subtilis deletion mutants lacking one or both flotillin genes (FloA and FloT) that would differentiate them from the wild type strain. This would allow screening of the Flo* mutants. We tested the differences (growth curve, viability on plates, colony aspect, cell morphology, spore formation etc.) between ∆floA, ∆floT single deletion and ∆floA∆floT double deletion mutants with the wild type strains of B. subtilis under multiple stress conditions (antibiotics, osmotic stress, heat/cold shock, detergents, extreme pH, limited nutrient availability, etc.). No significant differences were detected between flotillin deletion mutants with respect to their parental strains. Moreover, the FloA and FloT overproduction did not have any detectable effect on the resistance to these stress conditions.
FloA-scaffolded lipid rafts are crucial in other bacteria such as antibiotic resistant strains of pathogenic Staphylococcus aureus. Therefore, we also investigated the role of their structural domains in their properties. This will help to target specific residues and obtain Flo* mutants that will enhance their scaffolding capacity. We designed multiple mutations in flotillins from B. subtilis (FloA and FloT), S. aureus (FloA), Salmonella typhimurium (YqiK) and Escherichia coli (YqiK). From those, we already studied in details sequential truncations at their N-terminal ends to determine the residues involved in membrane anchoring. These mutants fused to mCherry flourorescent protein were overproduced in E. coli. The full-length protein was detected bound to the bacterial membrane using confocal microscope. The deletion mutants abolished the membrane attachment, proportionally to the extent of the deleted region. Our results point to a key role of a conserved prolin residue at the N-terminal end in the membrane attachment. We predict that this prolin may form a hairpin loop to anchor the protein to lipid bilayer.
Strikingly both S. typhimurium and E. coli YqiK proteins lacking this predicted hairpin loop (YqiK∆Nt20 in both cases) had a dramatic effect on the host E. coli strain. When overproduced they inhibited the cell growth, viability, division and morphology. The remaining cells were significantly deformed and presented abnormal bulges. We postulate that these mutants, unable to bind to membrane, titer their interacting partners, likely cell wall synthesis proteins, sequestering them in the cytoplasm and thus these partners are unable to function correctly at membranes.
The vast majority of prokaryotic flotillins, contrary to their eukaryotic counterparts, are encoded by an operon containing at least another gene, encoding a NfeD-like protein. It is likely involved in the activation of flotillins, being the C-terminal OB-fold domain crusial in this process. Therefore,we included NfeD in our studies on flotillins and their gain-of-function mutants. Our bacterial two-hybrid assay results show that staphylococcal NfeD strongly self-interacts and also interacts with FloA. The OB-fold alone does not interact with flotillin nor self-interact, suggesting that it has to be membrane-anchored to perform its biological function. It was confirmed by using a truncated NfeD protein lacking OB-fold.
We also focused on the structural details of the flotillin molecular assemblies. We cloned and optimized the purification of flotillins and their NfeD proteins from different bacterial species. The purified FloA from S. aureus did not form any particular structures in solution. Given that flotillins are membrane-attached proteins, we decided to study their behavior in a membrane-context, considering lipids as their essential cofactor. Accordingly, we performed lipid monolayer assay with the purified flotillins. Strikingly, we detected a presence of long, straight, tubular structures on lipid monolayers formed by staphylococcal FloA.
