Leaflet by Leaflet Synergistic Effects of Antimicrobial Peptides on Bacterial and Mammalian Membrane Models

Abstract

Antimicrobial peptides (AMPs) offer significant hope in the fight against antibiotic resistance. Operating via a mechanism different from that of antibiotics, they target the microbial membrane and ideally should damage it without impacting mammalian cells. Here, the interactions of two AMPs, magainin 2 and PGLa, and their synergistic effects on bacterial and mammalian membrane models were studied using electrochemical impedance spectroscopy, atomic force microscopy (AFM), and fluorescence correlation spectroscopy. Toroidal pore formation was observed by AFM when the two AMPs were combined, while individually AMP effects were confined to the exterior leaflet of the bacterial membrane analogue. Using microcavity-supported lipid bilayers, the diffusivity of each bilayer leaflet could be studied independently, and we observed that combined, the AMPs penetrate both leaflets of the bacterial model but individually each peptide had a limited impact on the proximal leaflet of the bacterial model. The impact of AMPs on a ternary, mammalian mimetic membrane was much weaker.

Figure 1

Relative variation of (a) resistivity (ΔR) and (b) capacitance (ΔQ) of the E. coli and mammalian model membrane bilayer upon treatment with an equimolar mixture of magainin 2 and PGLa. All EIS measurements were performed within the frequency range of 0.05–10⁵ Hz at a dc bias voltage of 0 V with an ac amplitude of 0.01 V in 0.01 M PBS at 22 ± 1 °C. The data are means ± SD. The two peptides were premixed at an equimolar ratio in these experiments and then introduced into the electrochemical cell containing the MSLB systems. After incubation at each concentration for 30 min, data points were recorded.

Figure 2

Representative AFM images of the E. coli bilayer (a) without AMPs or in the presence of (b) 2 μM magainin 2 (MG2), (c) 2 μM PGLa, and (d) an equimolar mixture of magainin 2 and PGLa (1 μM maganin 2 and 1 μM PGLa) with a total concentration of 2 μM in the contact buffer. The line scans across images a–d are shown in panels e–h, respectively. Images b–d were captured following peptide incubation for 30 min.

Figure 3

AFM images of a ternary DOPC/SM/Chol (2:2:1) bilayer (a) in the absence and (b) in the presence of an equimolar mixture of magainin 2 and PGLa with total concentration of 2 μM. (c) Line scan across image b.

Figure 4

Representative autocorrelation functions (ACFs) obtained for labeled (a) DOPE-ATTO655 (in the upper leaflet) and (b) DOPE-ATTO532 (in the lower leaflet) E. coli MSLBs in the absence of AMPs (black circles) or in the presence of PGLa (red circles), magainin 2 (pink circles), and an equimolar mixture of magainin 2 and PGLa (brown circles) with a concentration of 2 μM in each case.

Figure 5

Schematic representation of a microcavity-supported lipid bilayer (MSLB) array and plausible membrane–peptide association in model bilayers of (A) a bacterial membrane (E. coli extract) and (B) a mammalian membrane (DOPC/SM/cholesterol).