The lexicon of antimicrobial peptides: a complete set of arginine and tryptophan sequences

Abstract

Our understanding of the activity of cationic antimicrobial peptides (AMPs) has focused on well-characterized natural sequences, or limited sets of synthetic peptides designed de novo. We have undertaken a comprehensive investigation of the underlying primary structural features that give rise to the development of activity in AMPs. We consider a complete set of all possible peptides, up to 7 residues long, composed of positively charged arginine (R) and / or hydrophobic tryptophan (W), two features most commonly associated with activity. We found the shortest active peptides were 4 or 5 residues in length, and the overall landscapes of activity against gram-positive and gram-negative bacteria and a yeast were positively correlated. For all three organisms we found a single activity peak corresponding to sequences with around 40% R; the presence of adjacent W duplets and triplets also conferred greater activity. The mechanistic basis of these activities comprises a combination of lipid binding, particularly to negatively charged membranes, and additionally peptide aggregation, a mode of action previously uninvestigated for such peptides. The maximum specific antimicrobial activity appeared to occur in peptides of around 10 residues, suggesting ‘diminishing returns’ for developing larger peptides, when activity is considered per residue of peptide.

Antimicrobial Peptide Mimics for Clinical Use: Does Size Matter?

“…By focussing on only using arginine and tryptophan and synthetic amino acids they showed across a series of key papers that a high antimicrobial activity could be obtained once a "sufficient" number of cationic residues and hydrophobic elements were assembled (32,54,55). The order of assembly appeared to only play a minor role (32) and this was recently verified in an elegant and extensive study by Dodson and co-workers who assembled and evaluated all (252 out of 254) possible RW peptides up to seven amino acids in length (56). As visually presented in Figure 3, weak activity against Gram-positive bacteria is seen for some tetrapeptides while almost all hexa-and heptapeptides display a high activity, seemingly independent of sequence, provided sufficient charge and hydrophobicity is provided (Figure 3).…”

“…Dissection of the hydrophobic contribution into a minimal hydrophobic volume (65) and its placement (39) instead of a number of hydrophobic residues also generates a more accurate understanding of the essential structural requirements for high antimicrobial activity (65). While the initial "2+2" pharmacophore remains quite accurate for linear peptides, parallel and independent developments have (56). The diagram highlights how the antimicrobial activity of the peptides increases with peptide length and that how most peptides are active as heptapeptides if sufficient R and W residues are combined in a non-sequence dependent manner.…”

AMPing Up the Search: A Structural and Functional Repository of Antimicrobial Peptides for Biofilm Studies, and a Case Study of Its Application to Corynebacterium striatum, an Emerging Pathogen

“…AMPs in B-AMP are diverse in terms of length, structure, sources, and activity, which is important for several reasons. The length of AMPs is known to influence a range of activities, including antimicrobial effects, hemolytic activity and membrane binding and accumulation ( Bahar and Ren, 2013 ; Li et al., 2017 ; Clark et al., 2021 ). Further, AMPs possess a vast diversity in secondary structures, such as α-helices, β-sheets, coils, loops, and combinations of these structures; AMPs from all structural groups have been observed to exhibit antimicrobial activity, and certain AMPs are also known to change conformation on interaction with target structures ( Epand and Vogel, 1999 ; Bahar and Ren, 2013 ).…”

Multitalented Synthetic Antimicrobial Peptides and Their Antibacterial, Antifungal and Antiviral Mechanisms

“…In the barrel-stave , interaction among peptides is a prerequisite as they mimic a transmembrane pore, whereas, in the case of the toroidal model, peptides are loosely arranged [ 64 , 65 ]. Despite the perturbation of the membrane seems to vary depending on the peptides, actually, the mechanisms of action are not completely well-defined and they are partially overlapping [ 66 ]. Moreover, all these models are based on the membrane perturbation but, then, the killing effect is not always enough to provide antimicrobial activity [ 67 ].…”