Influenza-virus-liposome lipid mixing is leaky and largely insensitive to the material properties of the target membrane.

Monolayer intrinsic curvature, void stabilization, and membrane rupture tension have been suggested as important factors determining the rate of membrane fusion. Here, we have studied the kinetics of fusion between influenza virus and target liposomes as a function of various target membrane material properties. In order to examine the fusion process directly, a simple prebinding step is used and proven to be adequate to achieve fusion-rate-limiting kinetics. To test the hypothesis about membrane curvature and void stabilization, we studied the lipid mixing kinetics with dioleoylphosphatidylcholine (DOPC)/ganglioside GD1a (GD1a) liposomes containing lysooleoylphosphatidylcholine (LPC, positive curvature), dioleoyglycerol (DOG, negative curvature), arachidonic acid (AA, negative curvature), and hexadecane (HD, void stabilization). DOG, AA, and HD (at 4 mol%) showed no significant effect on the fusion kinetics, while LPC reversibly inhibited influenza HA mediated fusion only at very high concentrations. Using target liposomes with different membrane rupture tension values, no obvious correlation between membrane rupture tension and the rate of lipid mixing was observed. Moreover, a reported potential antiviral compound, tert-butylhydroquinone (t-b-HQ) (Bodian et al., 1993), showed no significant effect on the kinetics of influenza fusion. Finally leakage of liposome contents was detected during lipid mixing. For encapsulated molecules smaller than 450 MW, the kinetics of leakage is very similar to the kinetics of lipid mixing. In fact, leakage was also detected for encapsulated molecules up to 10 000 MW, suggesting that HA mediated lipid mixing is a very leaky process. Since "nonleaky fusion" has been the foundation of influenza fusion models, our work suggests the need for a major revision in the modeling of this process.