An Electrospray Ionization Mass Spectrometry Study on the "In Vacuo" Hetero-Oligomers Formed by the Antimicrobial Peptides, Surfactin and Gramicidin S.

It was previously observed that the lipopeptide surfactants in surfactin (Srf) have an antagonistic action towards the highly potent antimicrobial cyclodecapeptide, gramicidin S (GS). This study reports on some of the molecular aspects of the antagonism as investigated through complementary electrospray ionization mass spectrometry techniques. We were able to detect stable 1:1 and 2:1 hetero-oligomers in a mixture of surfactin and gramicidin S. The noncovalent interaction between GS and Srf, with the proposed equilibrium: GS~Srf↔GS+Srf correlated to apparent K d values of 6-9 µM in gas-phase and 1 µM in aqueous solution. The apparent K d values decreased with a longer incubation time and indicated a slow oligomerization equilibrium. Furthermore, the low µM K dapp values of GS~Srf↔GS+Srf fell within the biological concentration range and related to the 2- to 3-fold increase in [GS] needed for bacterial growth inhibition in the presence of Srf. Competition studies indicated that neither Na+ nor Ca2+ had a major effect on the stability of preformed heterodimers and that GS in fact out-competed Ca2+ and Na+ from Srf. Traveling wave ion mobility mass spectrometry revealed near symmetrical peaks of the heterodimers correlating to a compact dimer conformation that depend on specific interactions. Collision-induced dissociation studies indicated that the peptide interaction is most probably between one Orn residue in GS and the Asp residue, but not the Glu residue in Srf. We propose that flanking hydrophobic residues in both peptides stabilize the antagonistic and inactive peptide hetero-oligomers and shield the specific polar interactions in an aqueous environment. Graphical Abstract ᅟ.

Inhibition of malaria parasite blood stages by tyrocidines, membrane-active cyclic peptide antibiotics from Bacillus brevis.

Tyrothricin, a complex mixture of antibiotic peptides from Bacillus brevis, was reported in 1944 to have antimalarial activity rivalling that of quinine in chickens infected with Plasmodium gallinaceum. We have isolated the major components of tyrothricin, cyclic decapeptides collectively known as the tyrocidines, and tested them against the human malaria parasite Plasmodium falciparum using standard in vitro assays. Although the tyrocidines differ from each other by conservative amino acid substitutions in only three positions, their observed 50% parasite inhibitory concentrations (IC(50)) spanned three orders of magnitude (0.58 to 360 nM). Activity correlated strictly with increased apparent hydrophobicity and reduced total side-chain surface area and the presence of ornithine and phenylalanine in key positions. In contrast, mammalian cell toxicity and haemolytic activities of the respective peptides were considerably less variable (2.6 to 28 microM). Gramicidin S, a structurally analogous antimicrobial peptide, was less active (IC(50)=1.3 microM) and selective than the tyrocidines. It exerted its parasite inhibition by rapid and selective lysis of infected erythrocytes as judged by fluorescence and light microscopy. The tyrocidines, however, did not cause an overt lysis of infected erythrocytes, but an inhibition of parasite development and life-cycle progression.

Analyses of dose-response curves to compare the antimicrobial activity of model cationic alpha-helical peptides highlights the necessity for a minimum of two activity parameters.

To assess and compare different model Leu-Lys-containing cationic alpha-helical peptides, their antimicrobial activities were tested against Escherichia coli as target organism over a broad peptide concentration range. The natural cationic alpha-helical peptides magainin 2 and PGLa and the cyclic cationic peptide gramicidin S were also tested between comparison. The dose-response curves differed widely for these peptides, making it difficult to rank them into an activity order over the whole concentration range. We therefore compared five different inhibition parameters from dose-response curves: IC(min) (lowest concentration leading to growth inhibition), IC(50) (concentration that gives 50% growth inhibition), IC(max) (related to minimum inhibition concentration and minimum bactericidal concentration), inhibition concentration factor (IC(F); describing the increase in concentration of the peptide between minimum and maximum inhibition), and activity slope (A(S); related to the Hill coefficient). We found that these parameters were covariant: two of them sufficed to characterize the dose dependence and hence the activity of the peptides. This was corroborated by showing that the dose dependences followed the Hill equation, with a small, constant aberration. We propose that the activity of antimicrobial peptides can readily be characterized by both IC(50) and IC(F) (or A(S)) rather than by a single parameter and discuss how this may relate to investigations into their mechanisms of action.