Transbilayer lipid redistribution accompanies poly(ethylene glycol) treatment of model membranes but is not induced by fusion

Abstract

Small, unilamellar vesicles (SUV) or large, unilamellar vesicles (LUV) containing a small amount of N-(7-nitro-2,1,3-benzoxadiazol-4-yl)phosphatidylethanolamine (NBD-PE) or the corresponding phosphatidylserine (NBD-PS) were made asymmetric in labeled lipid by reduction of outer leaflet probe with externally added sodium dithionite. Following removal of dithionite, transbilayer lipid redistribution (presumably due to lipid flip-flop) was indicated by a loss of fluorescence intensity upon readdition of dithionite. Vesicle rupture and fusion in the presence of PEG were measured by changes in the fluorescence of trapped Tb3+ complexed with dipicolinic acid (DPA) or by the increase of fluorescence from 8-aminonaphthalene-1,3,6-trisulfonic acid (ANTS) coencapsulated with a quenching agent. NBD-PE redistributed slowly (approximately 2%/h) in all symmetrically labeled vesicles examined, while NBD-PS did not. NBD-PE redistribution was not accelerated by treatment of vesicles with PEG below concentrations that induced vesicle rupture or fusion, but was enhanced at or above these PEG concentrations. SUV prepared from hen egg yolk phosphatidylcholine (egg PC) or from dioleoylphosphatidylcholine (DOPC)/dilinolenoylphosphatidylcholine (diC(18:3)PC) (85/15) mixtures were shown to fuse without rupturing in the presence of appropriate concentrations of PEG. Matching the osmolalities inside and outside the vesicle mitigated against rupture but did not prevent fusion. Under these conditions, NBD-PE flip-flop was proportional to the amount of fusion, but with different proportionality constants for the two lipid systems, while NBD-PS flip-flop did not occur. Redistribution of total mass from the outer to the inner leaflet during the fusion process could be detected in terms of a change in the ratio of dithionite-reducible probe to total probe. Both probes detected inwardly directed redistribution of lipid mass under conditions that induced fusion of SUV. We conclude that inward mass redistribution must accompany PEG-mediated SUV fusion, but that lipid flip-flop is not mechanistically related to the fusion process.