Trans-kingdom conservation of mechanism between bacterial actifensin and eukaryotic defensins

Abstract

Antimicrobial peptides are defense molecules found across all domains of life holding promise for developing therapies against drug-resistant pathogens. Actifensin, from Actinomyces ruminicola DPC7226, exhibits potent activity against gram-positive bacteria and shares structural similarities with eukaryotic defensins. This study characterized actifensin's mechanism of action and therapeutic potential. The findings revealed that actifensin inhibits peptidoglycan synthesis by binding lipid II (Kd = 30 ± 20 nM). Unlike defensins, it also binds lipid I (Kd = 24 ± 27 nM) without significant difference, suggesting the N-acetyl glucosamine moiety of lipid II is not required for complexation. Membrane disruption was not observed with DiSC₃(5) fluorescence, or synthetic unilamellar liposomes, indicating indirect cell death via cell wall weakening, visualised by phase contrast microscopy. Actifensin showed no haemolytic activity or toxicity up to 128 µg/ml in human erythrocytes and Hep G2 cells. The peptide was not immunogenic, demonstrating no induction of LDH release in PBMCs or any effect on TLR-mediated signalling. Structural motif analysis identified actifensin as part of a conserved trans-kingdom defensin subfamily, GXGCP, distinct from XTCD peptides in more recently evolved arthropods. These findings emphasise the conserved structure-function relationship of antimicrobials across kingdoms, suggesting a shared evolutionary history of defensins and highlight the therapeutic potential for them or their variants.

Fig. 1. Actifensin complexes with the peptidoglycan precursor molecule, lipid II.

a Spot-on-lawn assays using stress-response bioreporter strains of Bacillus subtilis 168: pAC6-P_yorB (DNA), pAC6-P_yvgS (RNA), pAC6-P_yheI (protein), and pAC6-P_ypuA (cell wall). Actifensin (AfnA) and plectasin (Plec) induce β-galactosidase expression from the ypuA promoter, indicating interference with cell wall biosynthesis. Results are compared with control antibiotics: Rif (rifampicin), Clind (clindamycin), Cipro (ciprofloxacin), and Vanc (vancomycin). b Basic schematic of cell wall precursor compound structures and the cycle of their incorporation into peptidoglycan. PGN peptidoglycan. c Results of lipid complexation assays using purified cell wall precursors at varying molar ratios of actifensin to lipid. Complexation was assessed based on the lipid band intensity on the membrane: absence (−), decreased intensity (+/−), or unaffected intensity (+). ND not determined. d Antagonization assays of the actifensin-induced lia cell wall stress response in B. subtilis P_lia-lux, performed at 1:1 molar ratios of cell wall precursor lipids to actifensin.

Fig. 2. Actifensin’s action interferes with cell wall biosynthesis without directly impacting the cell membrane.

a Phase contrast microscopy images of B. subtilis 168 treated with mechanistically distinct compounds. Lipid II-binding agents induce a weakened cell wall resulting in membrane blebs. b Phase contrast images of B. subtilis P_xyl-gfp-minD over 60 min following treatment with actifensin, the known membrane-interacting bacteriocin, nisin, and no treatment. c DiSC₃(5) fluorescence of B. subtilis 168 cells treated with increasing concentrations of actifensin and the membrane-depolarising compound valinomycin. d Carboxyfluorescein (CF) efflux assays from unilamellar liposomes lack of CF release with actifensin in the presence and absence of lipid II, where nisin forms pores in the presence of lipid II.

Fig. 3. Actifensin’s effects on human cells.

a Relative viability of human epithelial cells (Hep G2) following 24 h of actifensin treatment, calculated as a percentage of untreated cells. b Relative viability of human erythrocytes following actifensin treatment for 24 h. Actifensin does not modulate TNF release in PBMCs stimulated with c LPS, d Pam3CSK4, or e heat-inactivated S. aureus (6.5 × 10⁸ cfu/ml in stationary phase or 6.9 × 10⁸ cfu/ml in mid-logarithmic phase). Bar graphs show mean + SEM (n = 3–5). n.d. not detected.

Fig. 4. Conserved structures and spread of the GXGCP and XTCD subfamilies of the CSαβ peptides.

a Schematic and exemplary peptide structures displaying conserved features and mechanism-related residues separating the peptide subfamilies. b Taxonomic phylogram of species that encode CSαβ peptides coloured by the type of peptides produced (yellow—GXGCP, blue—XTCD) and images displaying spread of peptide presence across phyla.