Linearized esculentin-2EM shows pH dependent antibacterial activity with an alkaline optimum

Abstract

Here the hypothesis that linearized esculentin 2EM (E2EM-lin) from Glandirana emeljanovi possesses pH dependent activity is investigated. The peptide showed weak activity against Gram-negative bacteria (MLCs ≥ 75.0 μM) but potent efficacy towards Gram-positive bacteria (MLCs ≤ 6.25 μM). E2EM-lin adopted an α-helical structure in the presence of bacterial membranes that increased as pH was increased from 6 to 8 (↑ 15.5-26.9%), whilst similar increases in pH enhanced the ability of the peptide to penetrate (↑ 2.3-5.1 mN m^-1) and lyse (↑ 15.1-32.5%) these membranes. Theoretical analysis predicted that this membranolytic mechanism involved a tilted segment, that increased along the α-helical long axis of E2EM-lin (1-23) in the N → C direction, with - < µH > increasing overall from circa - 0.8 to - 0.3. In combination, these data showed that E2EM-lin killed bacteria via novel mechanisms that were enhanced by alkaline conditions and involved the formation of tilted and membranolytic, α-helical structure. The preference of E2EM-lin for Gram-positive bacteria over Gram-negative organisms was primarily driven by the superior ability of phosphatidylglycerol to induce α-helical structure in the peptide as compared to phosphatidylethanolamine. These data were used to generate a novel pore-forming model for the membranolytic activity of E2EM-lin, which would appear to be the first, major reported instance of pH dependent AMPs with alkaline optima using tilted structure to drive a pore-forming process. It is proposed that E2EM-lin has the potential for development to serve purposes ranging from therapeutic usage, such as chronic wound disinfection, to food preservation by killing food spoilage organisms.

Keywords: Linearized esculentin 2EM (E2EM-lin); Preference for gram-positive bacteria; Tilted peptide; pH dependent with alkaline optimum; α-Helical structure.

Fig. 1

The potential of AMPs for tilted peptide formation. In A, extended hydrophobic moment plot methodology showed that the data points representing E2EM-lin (1–23) [red square (A), < µH > = 0.67 and < H > = 0.09] and E2EM-lin (25–37) [red square (B), < µH > = 0.43 and < H   = 0.06] lay in the shaded area of the plot diagram, indicating the potential to form tilted peptide structure (A). In B, amphiphilic profiling revealed that E2EM-lin (1–23) [red line (A)] possessed a hydrophobicity gradient along the α-helical long axis, which is characteristic of tilted peptides [40, 54]. The hydrophobicity gradient formed by E2EM-lin (1–23) increased along the α-helical long axis in the N → C direction, with − < µH > increasing overall from circa − 0.8 to − 0.3 (B). In contrast, the amphiphilic profiling of E2EM-lin (25–37) showed that this segment formed no discernable hydrophobicity gradient over residues 25–37, which suggested that this region does not form tilted structure (B). In C, E2EM-lin (1–23) was modelled as a two-dimensional axial projection, which revealed that this α-helix possessed a hydrophobic face of circa 180° and a hydrophilic face, which is rich in lysine, aspartic acid and glycine residues (C). In D, E2EM-lin (25–37) was also modelled as a two-dimensional axial projection and was found to form a α-helix, possessing a hydrophobic face of 120° and a hydrophilic face rich in polar residues and lysine residues (D). (Color figure online)

Fig. 2

The effect of pH on the conformation of E2EM-lin in the presence of bacterial membranes. This figure shows the effect of changing pH on the conformational behaviour of E2EM-lin in the presence of SUVs mimetic of bacterial membranes, which are those representing E. coli (A), P. aeruginosa (B), B. subtilis (C) and S. aureus (D) at pH 6 (Blue), pH 7 (Orange) and pH 8 (Green). In all cases, these curves possesses minima in the range 210–224 nm and a maxim around 193 nm, which is typical of α-helical structure [42]. Analysis of these spectra showed that the levels of α-helicity possessed by E2EM-lin was enhanced as pH increased from pH 6 to pH 8 in the case of E. coli (26.9–44.3%), P. aeruginosa (37.8–46.3%), B. subtilis (44.6–71.5%) and S. aureus (53.8–74.9%) (Table 3). (Color figure online)

Fig. 3

The effect of pH on the ability of E2EM-lin to penetrate and lyse bacterial membranes. This figure shows the effect of changing pH on the interactions of E2EM-lin with lipid monolayers mimetic of bacterial membranes. These interactions are represented by the maximal surface pressure changes induced by the peptide in the case of monolayers mimetic of membranes from E. coli (A), P. aeruginosa (B), B. subtilis (C) and S. aureus (D) at pH 6 (Blue), 7 (Orange) and 8 (Green). Data derived from these charts showed that the interaction of E2EM-lin with these monolayers were enhanced as pH increased from pH 6 to pH 8 in the case of E. coli (2.9–5.2 mN m⁻¹), P. aeruginosa (1.3–4.8 mN m⁻¹), B. subtilis (4.5–9.2 mN m⁻¹) and S. aureus (5.9–9.9 mN m⁻¹) (Table 4A). (Color figure online)

Fig. 4

A model for the pH dependent, membranolytic antibacterial action of E2EM-lin. This figure is revised from previous work [37, 53] and shows a schematic representation of the pH dependent antimicrobial action proposed for E2EM-lin. Initially, the peptide interacts with the bacterial membrane surface and forms α-helical structure (represented as cylinders) with a hydrophobic surface (red) and a hydrophilic surface (blue) (A). The α-helical structure formed by E2EM-lin (1–23) possesses a hydrophobicity gradient and the levels of this structure are enhanced by increasing pH (A). This tilted segment then promotes pore formation by E2EM-lin via membrane insertion and the adoption of a transmembrane orientation, which is stabilized by the surface interactions of E2EM-lin (25–37) (B, C). Potentially, E2EM-lin can form a toroidal pore (B) or a barrel stave pore (C) and the major difference between these pore types is that in the former pore, the membrane leaflets deform to allow the lipid head-group region to remain in contact with the hydrophilic face of the E2EM membrane spanning region, which is not observed in the latter pore [54]. In both cases, increasing pH promotes higher levels of membranolysis, which are maximal under alkaline conditions (B, C). For clarity, two monomers of E2EM-lin are shown in this pore forming process but it has been predicted that the involvement of higher order oligomers of the peptide are probable [57]. (Color figure online)