In the selection or design of antimicrobial peptides, the key role played by cationic amino acids and chain length on the inhibitory potency and specificity is not clear. A fundamental study was conducted using chemically synthesized homopeptides of L-Lys and L-Arg ranging from 7 to 14 residues. Their effect on growth inhibition was evaluated over a wide range of Gram-positive bacteria at different levels of concentration. Interestingly, at lower concentrations (10 μM), Lys homopeptides with odd number of residues, especially with 11 residues, showed a broader inhibitory activity than those with even number of residues. At higher peptide concentrations (>20 μM), the inhibitory activity of Lys homopeptides was directly related to the number of residues in the chain. In contrast, Arg homopeptides, at lower concentrations, did not exhibit a defined pattern of bacterial inhibition related to the number of residues; however, at higher concentrations (>20 μM), the inhibitory effects were more pronounced. Lys homopeptides at concentrations up to 300 μM showed a remarkably lower toxicity against CHSE-214 cells. Arg homopeptides exhibited negligible cytotoxicity up to chain length of 11 residues at concentrations lower than 100 μM, but an abrupt increase in toxicity resulted when the peptide chain length reached 12 amino acid residues and higher concentrations. All synthesized homopeptides displayed characteristic polyproline II helix conformation in both buffer and liposomes, as shown by CD spectroscopy. This result suggests that short Lys homopeptides with an odd number of residues (9 and 11) have a broad spectrum of activity against Gram-positive bacterial cells compared with Arg homopeptides, which in turn showed a considerably higher selectivity toward those cells. By investigating the differences between Lys and Arg homopeptides, this study contributes to the understanding of their mechanism of growth inhibition and selectivity. Thus, it provides further guidelines for a rational design of short antimicrobial peptides.

Previous work demonstrated that Lys homopeptides with an odd number of residues (9, 11 and 13) were capable of inhibiting the growth of Gram-positive bacteria in a broader spectrum and more efficiently than those with an even number of Lys residues or Arg homopeptides of the same size. Indeed, all Gram-positive bacteria tested were totally inhibited by 11-residue Lys homopeptides. In the present work, a wide variety of Gram-negative bacteria were used to evaluate the inhibitory activity of chemically synthesized homopeptides of L-Lys and L-Arg ranging from 7 to 14 residues. Gram-negative bacteria were comparatively more resistant than Gram-positive bacteria to Lys homopeptides with an odd number of residues, but exhibited a similar inhibition pattern than on Gram-positive bacteria. CD spectra for the odd-numbered Lys homopeptides in anionic lipid dimyristoylphosphatidylglycerol, and Escherichia coli membrane extract increased polyproline II content, as compared to those measured in phosphate buffer solution. Lys and Arg homopeptides were covalently linked to rhodamine to visualize the peptide interactions with E. coli cells using confocal laser scanning microscopy. Analysis of Z-stack images showed that Arg homopeptides indeed appear to be localized intracellularly, while the Lys homopeptide is localized exclusively on the plasma membrane. Moreover, these Lys homopeptides induced membrane disruption since the Sytox fluorophore was able to bind to the DNA in E. coli cultures.

Previous work demonstrated that lysine homopeptides adopt a polyproline II (PPII) structure. Lysine homopeptides with odd number of residues, especially with 11 residues (K11), were capable of inhibiting the growth of a broader spectrum of bacteria than those with an even number. Confocal studies also determined that K11 was able to localize exclusively in the bacterial membrane, leading to cell death. In this work, the mechanism of action of this peptide was further analyzed focused on examining the structural changes in bacterial membrane induced by K11, and in K11 itself when interacting with bacterial membrane lipids. Moreover, alanine and proline scans were performed for K11 to identify relevant positions in structure conformation and antibacterial activity. To do so, circular dichroism spectroscopy (CD) was conducted in saline phosphate buffer (PBS) and in lipidic vesicles, using large unilamellar vesicles (LUV), composed of 2-dimyristoyl-sn-glycero-3-phosphoglycerol (DMPG) or bacterial membrane lipid. Antimicrobial activity of K11 and their analogs was evaluated in Gram-positive and Gram-negative bacterial strains. The scanning electron microscopy (SEM) micrographs of Staphylococcus aureus ATCC 25923 exposed to the Lys homopeptide at MIC concentration showed blisters and bubbles formed on the bacterial surface, suggesting that K11 exerts its action by destabilizing the bacterial membrane. CD analysis revealed a remarkably enhanced PPII structure of K11 when replacing some of its central residues by proline in PBS. However, when such peptide analogs were confronted with either DMPG-LUV or membrane lipid extract-LUV, the tendency to form PPII structure was severely weakened. On the contrary, K11 peptide showed a remarkably enhanced PPII structure in the presence of DMPG-LUV. Antibacterial tests revealed that K11 was able to inhibit all tested bacteria with an MIC value of 5 µM, while proline and alanine analogs have a reduced activity on Listeria monocytogenes. Besides, the activity against Vibrio parahaemolyticus was affected in most of the alanine-substituted analogs. However, lysine substitutions by alanine or proline at position 7 did not alter the activity against all tested bacterial strains, suggesting that this position can be screened to find a substitute amino acid yielding a peptide with increased antibacterial activity. These results also indicate that the PPII secondary structure of K11 is stabilized by the interaction of the peptide with negatively charged phospholipids in the bacterial membrane, though not being the sole determinant for its antimicrobial activity.

Antifreeze glycoproteins (AFGPs) are considered to be the most efficient means to reduce ice damage to cell tissues since they are able to inhibit growth and crystallization of ice. The key element of antifreeze proteins is to act in a non-colligative manner which allows them to function at concentrations 300-500 times lowers than other dissolved solutes. During the past decade, AFGPs have demonstrated tremendous potential for many pharmaceutical and food applications. Presently, the only route to obtain AFGPs involves the time consuming and expensive process of isolation and purification from deep-sea polar fishes. Unfortunately, it is not amenable to mass production and commercial applications. The lack of understanding of the mechanism through which the AFGPs inhibit ice growth has also hampered the realization of industrial and biotechnological applications. Here we report the structural motifs that are essential for antifreeze activity of AFGPs, and propose a unified mechanism based on both recent studies of short alanine peptides and structure activity relationship of synthesized AFGPs.