Daptomycin Pore Formation and Stoichiometry Depend on Membrane Potential of Target Membrane

Abstract

Daptomycin is a calcium-dependent lipodepsipeptide antibiotic clinically used to treat serious infections caused by Gram-positive pathogens. Its precise mode of action is somewhat controversial; the biggest issue is daptomycin pore formation, which we directly investigated here. We first performed a screening experiment using propidium iodide (PI) entry to Bacillus subtilis cells and chose the optimum and therapeutically relevant conditions (10 µg/ml daptomycin and 1.25 mM CaCl₂) for the subsequent analyses. Using conductance measurements on planar lipid bilayers, we show that daptomycin forms nonuniform oligomeric pores with conductance ranging from 120 pS to 14 nS. The smallest conductance unit is probably a dimer; however, tetramers and pentamers occur in the membrane most frequently. Moreover, daptomycin pore-forming activity is exponentially dependent on the applied membrane voltage. We further analyzed the membrane-permeabilizing activity in B. subtilis cells using fluorescence methods [PI and DiSC₃(5)]. Daptomycin most rapidly permeabilizes cells with high initial membrane potential and dissipates it within a few minutes. Low initial membrane potential hinders daptomycin pore formation.

Keywords: Bacillus subtilis; Staphylococcus aureus; antimicrobial lipopeptides; daptomycin; membrane; pore formation; single pore conductance.

FIG 1

Permeabilization of B. subtilis cytoplasmic membrane. The permeabilization of B. subtilis cytoplasmic membrane induced by daptomycin was measured as the increase in fluorescence intensity of propidium iodide. B. subtilis cells were resuspended in buffer containing 10 mM HEPES (pH 7.2) and 0.5% glucose. Daptomycin at concentrations of 10, 8, 6, 4, 2, and 0 µg/ml was tested in the presence of 5, 2.5, 1.25, and 0 mM CaCl₂. Representative results from three independent experiments performed in duplicate are shown. cps, counts per second.

FIG 2

Histograms of daptomycin single-pore conductance. Pore-forming activity of daptomycin (10 µg/ml) was tested in 1 M KCl, 1.25 mM CaCl₂ and 10 mM Tris (pH 7.4) with different constant voltages (U_m) applied to the DPhPG membrane, as follows: 10 mV (A), 50 mV (B), 100 mV (C), and 150 mV (D). (C and D) Insets show a detailed view of the respective histograms in the range of 0 to 5,000 pS. (E) Typical ion current recordings of daptomycin channels for each histogram are shown; the time scale of each whole trace is 10 min. Note the different scales for U_m of 10 mV together with the 50 mV trace (50 pA), 100 mV (500 pA), and 150 mV trace (1,000 pA). To overcome the bin-edge effects, the histograms were created from pore openings (n = 163 to 766) using kernel density estimation (rectangular kernel with 50 pS width for panels A, B, and C and D insets or 200 pS width for panels C and D). At all the membrane voltages tested, we observed no pore formation before the addition of daptomycin.

FIG 3

Model of daptomycin pore stoichiometry and shape. (A) Histogram of pore conductances recorded at 100 mV (using the data from Fig. 2C) fitted to the model (red). Blue curves show the model components corresponding to individual oligomers of n subunits (see the text for details). The most probable G_4m value is 31.5 pS, with concavity of 2.8, and the most frequent are tetramers and pentamers. The data points represent the overlap of multiple kernel density estimates with varied bin sizes in the range of 10 to 80 pS. (B) Individual polygons differing by number of constituting molecules n. The S_n values show the relative area of the polygon (or relative pore conductance) normalized to the S₄ value. (C) The concave monomers change the relative cross-sections (gray hatched lines) of individual oligomers and allow the existence of a biangular assembly. (D) The convex monomers (with negative concavity) exhibit more dramatic cross-section differences between oligomers with low n numbers. Note that the value of concavity does not express the real shape but only the tendency of individual monomers either to expand or obstruct the pore lumen. The scheme (presented in panels B to D) was drawn by hand and should be taken as an approximation. The details of the model with other less-probable pore geometries are specified in the supplemental material (Fig. S4 and Table S1).

FIG 4

Voltage-dependent daptomycin channel openings. (A) Histogram of single-daptomycin-channel openings (1 M KCl, 1.25 mM CaCl₂, 10 mM Tris [pH 7.4]) with linearly rising voltage (from −20 to 150 mV) applied to the membrane. The histogram with 10 mV bin size was created from pore openings (n = 737) under 134 voltage ramps. (B) Representative ion current recordings showing individual daptomycin pores in the linearly rising voltage ramp (upper line). The pore marked with an asterisk (*) opened at 130 mV and closed at 7 mV (marked with an arrow). Another pore marked with a number sign (#) opened at 138 mV and remained in an opened state for 15 min (not shown).

FIG 5

Daptomycin permeabilizing activity of B. subtilis cytoplasmic membrane; effect of daptomycin on membrane potential and propidium iodide entry. (A) Effect of different adjusted values of ΔΨ of the target B. subtilis membrane on the ability of daptomycin (Dap; 10 µg/ml) to dissipate membrane potential. (B) Effect of daptomycin (10 µg/ml) on membrane potential in intact B. subtilis cells. In this experiment, the cells were resuspended in the same buffers as in panel A. However, here, we did not trigger partial membrane depolarization by valinomycin. Therefore, the cells retained their physiological membrane potential before daptomycin addition. Using calibration linear fit in Fig. S2, membrane potential was estimated to be −110 mV. (C) Membrane permeabilization induced by daptomycin (10 µg/ml) was measured as the increase in the fluorescence intensity of the membrane-impermeable dye propidium iodide (PI). B. subtilis cells were resuspended in buffers with different K⁺_out concentrations (same buffers as in panels A and B). The membrane potential was adjusted to the desired value of ΔΨ (in millivolts) by the addition of 4 µM valinomycin (for details, see Materials and Methods). Representative baselines for a buffer with a 7 mM K⁺_out concentration with (green dash-dot line) and without (green solid line) valinomycin addition are shown. Representative results from three independent experiments performed in duplicate are shown.

FIG 6

Individual events observed on bacterial cells after daptomycin treatment.