The Ectodomain of HA2 of Influenza Virus Promotes Rapid PH Dependent Membrane Fusion

To better understand the roles of different regions of influenza hemagglutinin in membrane fusion, we have studied the fusion properties of large unilamellar vesicles in the presence of constructs comprising the 127 amino acid ectodomain of the HA2 fragment (FHA2) as well as mutated forms of FHA2 containing single amino acid substitutions, the 95 amino acid truncated form of FHA2 lacking the N-terminal fusion peptide (SHA2), the 20 amino acid N-terminal fusion peptide and the ten amino acid peptide corresponding to the kinked loop region of FHA2. The 100 nm liposomes were made from dioleoylphosphatidylethanolamine, dioleoylphosphatidylcholine and cholesterol in equimolar ratio. At pH 5 a high rate of lipid mixing was observed with FHA2 present, even at very low molar concentrations, whereas much lower rates were observed using the shorter constructs: SHA2, the fusion peptide, and the loop peptide. Concentrations of FHA2 which promoted extensive lipid mixing also induced leakage of aqueous contents. Marked effects of FHA2 were also observed with liposomes of egg phosphatidylcholine. All of the changes observed with the liposomes were highly pH-dependent, with only negligible changes occurring at pH 7. The results demonstrate the potent action of FHA2 in promoting lipid mixing and demonstrate the contribution of other regions of the ectodomain of FHA2, in addition to the fusion peptide, to the mechanism of acceleration of membrane fusion. The results also indicate that the pH dependence of fusion is not due solely to changes in the interactions between the HA1 and HA2 subunits. Thus, the "spring loaded energy" is not required to bring about the apposition of the two membranes, considering that FHA2 is already in its thermostable conformation. The acidic amino acid residues in the kinked loop region appear to play a particularly important role in the pH-dependent fusion process as demonstrated by the marked loss of lipid mixing activity of mutant forms of FHA2.