Phosphate-dependent aggregation of KLn peptides affects their membranolytic activity

In this study, we investigate how the length of amphiphilic β-sheet forming peptides affects their interaction with membranes. Four polycationic model peptides with lengths from 6 to 18 amino acids were constructed from simple Lys-Leu repeats, giving [KL]n=3,5,7,9. We found that (1) they exhibit a pronounced antimicrobial activity with an intriguing length dependent maximum for [KL]5 with 10 amino acids; (2) their hemolytic effect, on the other hand, increases steadily with peptide length. CD analysis (3) and TEM (4) show that all peptides-except for the short [KL]3-aggregate into amyloid-like fibrils in the presence of phosphate ions, which in turn has a critical effect on the results in (1) and (2). In fact, (5) vesicle leakage reveals an intrinsic membrane-perturbing activity (at constant peptide mass) of [KL]5 > [KL]9 > [KL]7 in phosphate buffer, which changes to [KL]5 ≈ [KL]7 ≈ [KL]9 in PIPES. A specific interaction with phosphate ions thus explains the subtle balance between two counteracting effects: phosphate-induced unproductive pre-aggregation in solution versus monomeric membrane binding and vigorous lipid perturbation due to self-assembly of the bound peptides within the bilayer. This knowledge can now be used to control and optimize the peptides in further applications.In this study, we investigate how the length of amphiphilic β-sheet forming peptides affects their interaction with membranes. Four polycationic model peptides with lengths from 6 to 18 amino acids were constructed from simple Lys-Leu repeats, giving [KL]n=3,5,7,9. We found that (1) they exhibit a pronounced antimicrobial activity with an intriguing length dependent maximum for [KL]5 with 10 amino acids; (2) their hemolytic effect, on the other hand, increases steadily with peptide length. CD analysis (3) and TEM (4) show that all peptides-except for the short [KL]3-aggregate into amyloid-like fibrils in the presence of phosphate ions, which in turn has a critical effect on the results in (1) and (2). In fact, (5) vesicle leakage reveals an intrinsic membrane-perturbing activity (at constant peptide mass) of [KL]5 > [KL]9 > [KL]7 in phosphate buffer, which changes to [KL]5 ≈ [KL]7 ≈ [KL]9 in PIPES. A specific interaction with phosphate ions thus explains the subtle balance between two counteracting effects: phosphate-induced unproductive pre-aggregation in solution versus monomeric membrane binding and vigorous lipid perturbation due to self-assembly of the bound peptides within