Zero current potential and conductance of ionic channels formed by polyene antibiotic amphotericin B in a lipid bilayer were studied in various electrolyte solutions. Nonpermeant magnesium and sulphate ions were used to independently vary the concentration of monovalent anions and cations as well as to maintain the high ionic strength of the two solutions separated by the membrane. Under certain conditions the channels select very strongly for anions over cations. They are permeable to small inorganic anions. However, in the absence of these anions the channels are practically impermeable to any cation. In the presence of a permeant anion the contribution of monovalent cations to channel conductance grows with an increase in the anion concentration. The ratio of cation-to-anion permeability coefficients is independent of the membrane potential and cation concentration, but it does depend linearly on the sum of concentrations of a permeant anion in the two solutions. These results are accounted for on the assumption that a cation can enter only an anion-occupied channel to form an ionic pair at the center of the channel. The cation is also assumed to slip past the anion and then to leave the channel for the opposite solution. This model with only few parameters can quantitatively describe the concentration dependences of conductance and zero current potential under various conditions.

Summary. The purpose of the work presented in this paper was to determine experimentally the actual mechanism of lipid bilayer conductivity in the presence of tetrachlorotrifluoromethylbenzimidazole (TTFB). The capacitance and conductance of lipid bilayers were measured with a current-clamp method, as a function of equal TTFB concentrations (10 -7 to 5 • 10-5 M) in the two aqueous phases. The voltage across the membrane was measured as a function of time during rectangular current pulses. If the hydrogen buffer capacity of the solution is low, the voltage response to long current pulses has two components. Tile slow component is due to hydrogen ion concentration changes in the unstirred layers. This component disappears if the buffer capacity is made high enough. Membrane capacitance and conductance can be determined from the fast component of the voltage response. The conductance increases with the square of TTFB concentration (pH 2 to 7) and the capacitance is 0.4 laF/cm 2 for the range of concentration used. If solutions of low buffer capacity are used, shifts of hydrogen ion concentrations near the membrane give rise to a complicated dependence of the membrane potential on pH given a unit pH difference between the two aqueous solutions (protonic potential). This dependence can be explained if the membrane permeability for neutral uncoupler molecules (TH) is high enough. The membrane permeability coefficient is determined: PTn= 10 cm/sec, in other experiments the dependence of short-circuit current on pH difference between the two solutions was also measured, with the pH value on one side fixed in a given experiment. These complicated nonmonotonic dependences can be described using a mathematical equation with only two parameters: (1) the dissociation constant of TTFB in water (pK=5.2), and (2) the proportionality factor between short-circuit current and TTFB concentration squared. These data can be formally interpreted to mean that the membrane is permeable only to TzHand TH, where T is TTFB anion and H is hydrogen ion. However, this model does not explain the high current values obtained because of the limited rate constant of T2H-formation in aqueous solution. An alternative model is proposed. It is shown that the T2H-is not formed in aqueous phase but rather within the membrane. The TzH-can be the intramembrane charge carrier if its life-time is long enough. If the average lifetime is short, current might be carried through the membrane by proton exchange between TH and T-, when they collide. This mechanism could also account for the action of uncouplers of oxidative phosphorylation other than TTFB.

No abstract

No abstract

No abstract