Highly reproducible ion channels of the lipopeptide antibiotic syringomycin E demonstrate unprecedented involvement of the host bilayer lipids. We find that in addition to a pronounced influence of lipid species on the open-channel ionic conductance, the membrane lipids play a crucial role in channel gating. The effective gating charge, which characterizes sensitivity of the conformational equilibrium of the syringomycin E channels to the transmembrane voltage, is modified by the lipid charge and lipid dipolar moment. We show that the type of host lipid determines not only the absolute value but also the sign of the gating charge. With negatively charged bilayers, the gating charge sign inverts with increased salt concentration or decreased pH. We also demonstrate that the replacement of lamellar lipid by nonlamellar with the negative spontaneous curvature inhibits channel formation. These observations suggest that the asymmetric channel directly incorporates lipids. The charges and dipoles resulting from the structural inclusion of lipids are important determinants of the overall energetics that underlies channel gating. We conclude that the syringomycin E channel may serve as a biophysical model to link studies of ion channels with those of lipidic pores in membrane fusion.

The cyclic lipodepsipeptide, syringomycin E, when incorporated into planar lipid bilayer membranes, forms two types of channels (small and large) that are different in conductance by a factor of sixfold. To discriminate between a cluster organization-type channel structure and other possible different structures for the two channel types, their ionic selectivity and pore size were determined. Pore size was assessed using water-soluble polymers. Ion selectivity was found to be essentially the same for both the small and large channels. Their reversal (zero current) potentials with the sign corresponding to anionic selectivity did not differ by more than 3 mV at a twofold electrolyte gradient across the bilayer. Reduction in the single-channel conductance induced by poly(ethylene glycol)s of different molecular weights demonstrated that the aqueous pore sizes of the small and large channels did not differ by more than 2% and were close to 1 nm. Based on their virtually identical selectivity and size, we conclude that large syringomycin E channels are clusters of small ones exhibiting synchronous opening and closing.

No abstract

Cecropin A (CecA) and cecropin B (CecB) added to one side of a bilayer formed from equimolar mixtures of DOPS and DOPE, DPhPS and DPhPE, or DOPS, DOPE, and Chol leads to the formation of well-defined and well-reproducible ion channels of different conductance levels while cecropin P1 (CecP1) does not induce pore formation at micromolar concentrations. We found three populations of CecA channels: pores with weak cationic selectivity, pores with weak anionic selectivity, and pores that were nonselective. The dipole modifiers, flavonoids and styryl dyes, were used to modulate the channel-forming activity of CecA and CecB. The mean conductance of single CecA channels is affected by the influence of dipole modifiers on the lipid packing in the membrane. A decrease in the membrane dipole potential is accompanied by a decrease in the steady-state transmembrane current induced by CecA and CecB in cholesterol-free and cholesterol-containing bilayers. The observed changes in the channel-forming activity might be caused by an increase in the energy barrier for the interfacial accumulation of cecropin monomers. This finding indicates that the negative pole of the cecropin dipole is inserted into the membrane.

We evaluated the effect of agents modifying the membrane dipole potential: phloretin, 6-ketocholestanol and RH 421 on the properties of single channels formed by lipodepsipeptide syringomycin E (SRE) in planar lipid bilayers. SRE forms two conductive states in lipid bilayers: "small" and "large." Large SRE channels are clusters of several small ones, demonstrating synchronous openings and closures. The increase in the membrane dipole potential led to (i) an increase in SRE channel conductance, (ii) an increase in the channel's lifetime, and (iii) a decrease in a number of synchronously operating small channels in the clusters. Overall, the results support the model of the small SRE channel synchronization in the cluster as voltage-dependent orientation of the lipid dipoles associated with the channel pores.