Tyrocidine A interactions with saccharides investigated by CD and NMR spectroscopies.

Tyrocidines are a family of cyclic decapeptides produced by the soil bacterium, Brevibacillus parabrevis. These antibiotic peptides can be used to prevent infections in agriculture and food industry but also to prepare antimicrobial lozenges, creams, and dressings for medical applications. It has been observed that the tyrocidines interact with saccharides such as cellulose from their soil environment, as well as sugars in culture media and glycans in fungal cell walls. Here, we investigated the interactions of tyrocidines with glucose, sucrose, and cellotetraose (as cellulose model) in a quantitative fashion utilising CD and NMR spectroscopy. The CD and NMR spectra of tyrocidine A (TrcA) were analysed as a function of solvent composition, and the spectral properties agree with the formation of oligomeric structures that are governed by ß-sheet secondary structures once the acetonitrile content of the solvent is increased. Saccharides seem to also induce TrcA spectral changes reverting those induced by organic solvents. The CD spectral changes of TrcA in the presence of glucose agree with new ordered H-bonding, possibly ß-sheet structures. The amides involved in intramolecular H-bonding remained largely unaffected by the environmental changes. In contrast, amides exposed to the exterior and/or involved in TrcA intermolecular association show the largest 1 H chemical shift changes. CD and NMR spectroscopic investigations correlated well with TrcA-glucose interactions characterized by a dissociation constant around 200 µM. Interestingly, the association of cellotetraose corresponds closely to the additive effect from four glucose moieties, while a much higher dissociation constant was observed for sucrose. Similar trends to TrcA for binding to the three saccharides were observed for the analogous tyrocidines, tyrocidine B, and tyrocidine C. These results therefore indicate that the tyrocidine interactions with the glucose monosaccharide unit are fairly specific and reversible.

Oligomerisation of tryptocidine C, a Trp-rich cyclodecapeptide from the antimicrobial tyrothricin complex.

Tryptocidine C (TpcC, cyclo[D-Phe1-Pro2-Trp3-D-Trp4-Asn5-Gln6-Trp7-Val8-Orn9-Leu10]) is a broad-spectrum antimicrobial peptide in the tyrothricin complex produced by a soil bacterium, Brevibacillus parabrevis. Electrospray mass spectrometric studies reveal the oligomerisation of TpcC into dimers and higher oligomers, analogous to tyrocidine C (TrcC, Trp7 replaced by Tyr7). Ion mobility mass spectrometry (IMMS) further confirms the formation of stable peptide dimers and tetramers with diameters of 2.7 nm and 3.3 nm, respectively, calculated from collisional cross section (CCS). Molecular dynamic simulations and docking studies support the formation of amphipathic dimers, with a diameter of 2.5 ± 0.07 nm calculated from low energy model CCS. Circular dichroism and IMMS studies point towards dynamic hydrogen-bonded conformational changes up to 28-33 µM after which the structures become more static (or in equilibrium). Fluorescence studies indicate aromatic stacking of Trp residues with a CMC of 18 µM in aqueous solutions. The concentration and time dependent interaction of Trp in oligomers indicate cooperativity in the TpcC oligomerisation that leads to the formation of higher order microscopic structures. Scanning electron microscopy studies unequivocally shows that TpcC forms nanospheres with a mean diameter of 25 nm. Repeated smaller oligomeric units, possibly dimers and tetramers, self-assemble to form these nanospheres.

Following tyrothricin peptide production by Brevibacillus parabrevis with electrospray mass spectrometry.

The tyrocidines and analogues are cationic cyclodecapeptides [cyclo (D-Phe1-L-Pro2-L-(Phe3/Trp3)-D-(Phe4/Trp4)-L-Asn5-L-Gln6-L-(Tyr7/Phe7/Trp7)-L-Val8-L-(Orn9/Lys9)-L-Leu10], produced together with the neutral linear pentadecapeptide gramicidins, in the antibiotic tyrothricin complex by Brevibacillus parabrevis. Despite discovery 80 years ago, it was still uncertain whether these peptides are secreted or sequestered intracellularly. We resolved this by utilising high resolution electrospray mass spectrometry to confirm the predominantly intracellular sequestration of the peptides in the tyrothricin complex. A "peptidomics" approach allowed us to map the intracellular production of 16 cyclodecapeptides and 6 gramicidins over 16 days of culturing. Gramicidin production remained relatively constant, with Val-gramicidin A the predominant analogue produced throughout the 16 day fermentation period. The tyrothricin cyclodecapeptides have four variable positions and there was a culturing time related shift from the Phe-rich A analogues, containing a L-Phe3-D-Phe4 aromatic dipeptide unit, to the Trp-rich C analogues with L-Trp3-D-Trp4. For the other variable aromatic residue position, Tyr7 was preferentially incorporated above Trp7, with a minor incorporation of Phe7 over the whole culturing period. For the variable basic amino acid residue, there was time-sensitive shift from Orn9 to Lys9 incorporation. Modulation of the cyclodecapeptide profile over time does not correlate with the reported non-ribosomal peptide synthetase affinity, specifically for Trp in the variable aromatic residue positions, indicating additional supply-demand control in the cyclodecapeptides production by B. parabrevis. These novel observations are not only of importance for production and purification of selected peptide analogues from the tyrothricin complex, but also for insight into microbial control of non-ribosomal peptide production that extends beyond the peptide synthetase machinery.

The Multifaceted Antibacterial Mechanisms of the Pioneering Peptide Antibiotics Tyrocidine and Gramicidin S.

Cyclic ß-sheet decapeptides from the tyrocidine group and the homologous gramicidin S were the first commercially used antibiotics, yet it remains unclear exactly how they kill bacteria. We investigated their mode of action using a bacterial cytological profiling approach. Tyrocidines form defined ion-conducting pores, induce lipid phase separation, and strongly reduce membrane fluidity, resulting in delocalization of a broad range of peripheral and integral membrane proteins. Interestingly, they also cause DNA damage and interfere with DNA-binding proteins. Despite sharing 50% sequence identity with tyrocidines, gramicidin S causes only mild lipid demixing with minor effects on membrane fluidity and permeability. Gramicidin S delocalizes peripheral membrane proteins involved in cell division and cell envelope synthesis but does not affect integral membrane proteins or DNA. Our results shed a new light on the multifaceted antibacterial mechanisms of these antibiotics and explain why resistance to them is virtually nonexistent.IMPORTANCE Cyclic ß-sheet decapeptides, such as tyrocidines and gramicidin S, were among the first antibiotics in clinical application. Although they have been used for such a long time, there is virtually no resistance to them, which has led to a renewed interest in this peptide class. Both tyrocidines and gramicidin S are thought to disrupt the bacterial membrane. However, this knowledge is mainly derived from in vitro studies, and there is surprisingly little knowledge about how these long-established antibiotics kill bacteria. Our results shed new light on the antibacterial mechanism of ß-sheet peptide antibiotics and explain why they are still so effective and why there is so little resistance to them.

Antifungal peptides: To be or not to be membrane active.

Most antifungal peptides (AFPs), if not all, have membrane activity, while some also have alternative targets. Fungal membranes share many characteristics with mammalian membranes with only a few differences, such as differences in sphingolipids, phosphatidylinositol (PI) content and the main sterol is ergosterol. Fungal membranes are also more negative and a better target for cationic AFPs. Targeting just the fungal membrane lipids such as phosphatidylinositol and/or ergosterol by AFPs often translates into mammalian cell toxicity. Conversely, a specific AFP target in the fungal pathogen, such as glucosylceramide, mannosyldiinositol phosphorylceramide or a fungal protein target translates into high pathogen selectivity. However, a lower target concentration, absence or change in the specific fungal target can naturally lead to resistance, although such resistance in turn could result in reduced pathogen virulence. The question is then to be or not to be membrane active - what is the best choice for a successful AFP? In this review we deliberate on this question by focusing on the recent advances in our knowledge on how natural AFPs target fungi.