Investigation of cationicity and structure of pseudin-2 analogues for enhanced bacterial selectivity and anti-inflammatory activity

Abstract

Pseudin-2 (Ps), isolated from the frog Pseudis paradoxa, exhibits potent antibacterial activity and cytotoxicity. To develop antimicrobial peptides with anti-inflammatory activity and low cytotoxicity, we designed Ps analogues with Lys substitutions, resulting in elevated amphipathic α-helical structure and cationicity. We further substituted Gly¹¹ with Pro (Ps-P analogues) to increase bacterial cell selectivity. Ps analogues retained antimicrobial activity and exhibited reduced cytotoxicity, whereas Ps-P analogues exhibited lower cytotoxicity and antimicrobial activity. Tertiary structures revealed that Ps has a linear α-helix from Leu² to Glu²⁴, whereas Ps-P has a bend at Pro¹¹ between two short α-helixes. Using various biophysical experiments, we found that Ps analogues produced much higher membrane depolarization than Ps-P analogues, whereas Ps-P analogues may penetrate bacterial cell membranes. Ps and its analogue Ps-K18 exhibited potent anti-inflammatory activity in LPS-stimulated RAW264.7 and mouse dendritic cells via a mechanism involving the Toll-like receptor 4 (TLR4) pathway. These activities may arise from their direct inhibition of the formation of TLR4-MD-2_LPS complex, implying that amphipathic α-helical structure with an optimum balance between enhanced cationicity and hydrophobicity may be essential for their anti-inflammatory activity. The bent structure provided by Pro substitution plays an important role in enhancing bacterial cell selectivity and cell penetration.

Figure 1

(A) Head-on view of the amphipathic N-terminal helix from Leu² to Gln²⁴ of Ps. The superpositions of the 20 lowest energy structures were calculated from the NMR data and ribbon diagram of the lowest energy structure for (B) Ps and (C) Ps-P in 200 mM DPC micelles. The hydrophobic residues are indicated in red, and the hydrophilic residues are shown in blue.

Figure 2

Concentration-dependent toxic activities of Ps and its analogues toward (A) human RBCs, (B) mouse macrophage-derived RAW264.7 cells and (C) mouse bone marrow-derived dendritic cells.

Figure 3

Concentration-response curves of calcein leakage induced by the peptides from (A) egg yolk phosphatidylethanolamine (EYPE)/egg yolk phosphatidylglycerol (EYPG) (7:3, w/w), (B) EYPG/cardiolipin (CL) (6:4, w/w) and (C) EYPC/cholesterol (CH) large unilamellar vesicles (LUVs; 10:1, w/w).

Figure 4

(A) Concentration-dependent membrane depolarization of intact E. coli KCTC 1682, monitored as an increase in diSC₃-5 fluorescence. (B) Permeabilization of the outer membrane of E. coli KCTC 1682 monitored as an increase in NPN fluorescence intensity. (C) Permeabilization of the inner membrane of E. coli ML35p monitored as hydrolysis of ONPG by β-galactosidase. (D) Membrane depolarization of S. aureus KCTC 1621, monitored as an increase in diSC₃-5 fluorescence.

Figure 5

(A) Inhibition of nitrite production by peptides (10 μM and 20 μM) in LPS-stimulated RAW264.7 cells. (B) Disaggregation of LPS by the peptides. Enhancement of the intensity of FITC-labelled LPS as a function of the concentration of peptides. Effects of Ps and Ps-K18 on LPS-induced expression of inflammatory cytokines in (C) RAW264.7 cells and (D) dendritic cells as determined by enzyme-linked immunosorbent assay.

Figure 6

Effects of Ps and Ps-K18 on LPS-stimulated expression of pro-inflammatory cytokines in (A) RAW264.7 cells and (B) dendritic cells determined by reverse transcriptase polymerase chain reaction. Inhibition of inflammatory-related protein expression in LPS-stimulated (C) RAW264.7 cells and (D) dendritic cells.

Figure 7

Binding between TLR4 and (A) Ps, (B) Ps-K18 and (C) Ps-K14-K18, as measured by bio-layer interferometry (BLI).