We have two ears and one mouth, so we should listen more than we say." (Zeno of Citium) i Aim of the study The work presented in this study deals with the application of a solid-state 2 H-NMR method to study the action of a natural antimicrobial peptide (AMP), PGLa, on biological membranes. Such AMPs are promising compounds for developing new drugs and are interesting parts of the immune system. Though they are known to interact strongly with membranes, their mechanism of action against

more »

... is not yet fully understood and more investigations are needed. To this aim, one can use solid-state NMR spectroscopy. Having a suitable model for the structure and the motion of the peptide, its orientation in biological membranes can be calculated from several local orientational constraints of the peptide backbone measured by NMR spectroscopy. Hence, its biophysical mechanism of action can be understood. A 2 H-labeling method has been developed for model transmembrane peptides and is used here on AMPs for the first time. A similar method using 19 F-labels in the potentially structure disturbing residue CF 3 -phenylglycine (CF 3 -Phg) has been previously used to study the orientation of the PGLa peptide in DMPC lipid membranes. For this study, PGLa peptides carrying 2 H-labels at selected sites first need to be synthesized and characterized. Through several replacements (Ala, Ile or Gly replaced by Ala-d 3 ) and using different motional models, a rigorous analysis of the 2 H-NMR method is carried out, and its reliability for such amphipathic peptide is assessed. Using an appropriate motional model and the minimal number of data points to calculate an accurate orientation of the peptide, a reliable and simple method for further detailed experiments is validated. This 2 H-labeling strategy is compared to the CF 3 -Phg strategy, and their respective advantages are discussed. From a biological point of view, AMPs like PGLa are assumed to permeabilize cytoplasmic bacterial membranes by forming pores. In such pores, the peptides are supposed to be inserted in a transmembrane state, and the lipid head groups are tilted to participate in the pore structure. This study has to confirm this mechanism of action for PGLa, and it investigates the influence of parameters like the peptide concentration, the presence of anionic lipids, and the sample hydration. Another relevant parameter for the membrane permeabilization can be the presence of other peptides. PGLa and magainin 2 (Mag2) peptides have indeed been shown to act synergistically in this process. Hence, the orientational and dynamic behavior of PGLa in the presence of Mag2 is monitored using the same NMR methodology to shed light on how this synergism can be explained and to assess its critical structural features. ii