RESUMO
Mesentericin Y105 (Mes-Y105) is a bacteriocin secreted by Leuconostoc mesenteroides which is particularly active on Listeria. It is constituted by 37 residues and reticulated by one disulfide bridge. It has two W residues, W18 and W37, which can be studied by fluorescence. Two single substituted W/F analogues were synthesized (Mes-Y105/W18 and Mes-Y105/W37) to differentiate the local environment around each W and to study their changes in the presence of lipid vesicles. Fluorescence experiments show that, for the pure Trp-analogues, W18 and W37 are fully exposed to solvent whatever pH and buffer conditions. In the presence of lipid vesicles, both became buried. Lipid affinities were estimated: they are weak for zwitterionic phospholipids but an order of magnitude higher for negatively charged phosphatidylserine (PS) and phosphatidylglycerol (PG) lipids. On negatively charged PG lipids, Mes-Y105 and Mes-Y105/W37 display comparable lipid affinities. A decrease in lipid affinity is observed for Mes-Y105/W18 compared to Mes-Y105, which means that W37 would seem to be required for increased lipid selectivity. In the lipid-bound state W18 is strongly dehydrated, probably embedded into the acyl chains, while W37 stands more at the interface. Mes-Y105 was also studied by polarization modulation infrared reflection absorption spectroscopy (PMIRRAS), alone and in various phospholipid environments, to obtain structural information and to assess lipid perturbations. At nanomolar concentrations close to those required for anti-Listeria activity, Mes-Y105 forms films at the air/water interface and inserts into negatively charged lipid monolayers. In situ infrared data show that Mes-Y105 binding only affects the polar head group vibrations while the lipid order of the acyl chains remains unaffected. The PMIRRAS show that Mes-Y105 folds into an N-terminal antiparallel beta-sheet followed by an alpha-helix, both structures being tilted (40 degrees) compared to the normal at the interface, which is in agreement with the thickness estimated by Brewster angle microscopy (BAM). All these data support the proposal of a new model for Mes-Y105 at the membrane interface.