Back

Dynamic exchange of antimicrobial peptides stabilize persistent lipid pores

Sengel, J. T.; Zan, B.; Szabo, M.; Ulmschneider, J.; Wallace, M.

2025-09-18 biophysics
10.1101/2025.09.18.677094 bioRxiv
Show abstract

Antimicrobial peptides (AMPs) are a promising broad-spectrum complement to conventional antibiotics. But despite their relatively simple structures, the reported modes of action of AMPs are diverse, ranging from formation of stable ion-channel-like structures, or disruption of biochemical pathways, to comprehensive membrane solubilization. This apparent complexity of function is a significant hurdle in their future application and rational design. Here, we combine single-molecule tracking with optical single-channel recording in synthetic mimics of bacterial membranes to image the pores formed by AMPs in real time, mapping the diffusion of individual AMPs relative to sites of membrane permeablization. Using alamethicin, magainin-II and melittin as archetypes of three principal membrane-disrupting AMP classes, we observe AMP molecules freely diffusing on the membrane, but with an enhancement of local peptide surface density within a nanoscopic region surrounding each pore locus. We do not observe this enhancement for indolicidin, an AMP which is not believed to use pore formation as its main mode of toxicity. Corroborated by molecular dynamics simulations that replicate our experiments, these results suggest defects formed by membrane-active AMPs are persistent but dynamic structures, stabilised by individual peptides free to diffuse into and out of the pore. Whilst alamethicin, magainin-II and melittin are often placed in different mechanistic pore-forming groups, the broad similarities between observations in these different AMPs encourages a continuum view of pore-forming ability, rather than discrete categories of mechanism.

Matching journals

The top 11 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.