Every organism must contend with invaders and thus have an active system of defence and offence. A broad spectrum of molecular mechanisms has evolved in nature to perform these functions. Important contributors to a fast response are small peptides consisting of 20-40 amino acids. The activity of these peptides does not seem to be correlated to specific receptors, rather they act by perturbing the barrier function of cell membranes. However, the molecular mechanism of their action is still a matter of debate. An understanding of how these peptides distinguish between bacterial and mammalian cell membranes would allow the design of novel peptide antibiotics, which could selectively kill bacteria. Interest in this research area has grown, since the number of bacterial strains being resistant to conventional antibiotics has been increasing dramatically in recent years.

Using X-ray reflectivity (XRR) and grazing-incidence diffraction (GID) techniques on ID10B, we studied the effects of the antimicrobial peptide, peptidyl-glycylleucine-carboxyamide (PGLa), on lipid monolayers formed at the air/water water interface. This peptide, consisting of 21 amino acid residues, is found in the skin secretion of the South African clawed frog, Xenopus laevis. Monolayers composed of the negatively-charged phospholipid distearoyl-phosphatidylglycerol (DSPG) mimicked a bacterial cell membrane, whereas a mammalian cell membrane was mimicked by the zwitterionic distearoyl-phosphatidylcholine (DSPC).

Fig. 23: Grazing-incidence diffraction of a DSPG monolayer at different surface pressures: () 20 mN/m, () 25 mN/m, () 30 mN/m, () 35 mN/m, () 40 mN/m.

The Bragg peaks of the GID spectra (Figure 23), recorded from the monolayer of DSPG molecules, are characteristic for a two-dimensional (2D) ordering of the lipids in the monolayer plane. These peaks disappear completely from the spectra at any surface pressure upon addition of peptides to the monolayer (molar ratio of lipid:peptide studied: 2.5:1, 5:1, 25:1), which indicates destruction of the initial order of the 2D lattice. The XRR measurements, which give information about the electron-density profile normal to the monolayer surface, demonstrate that, in addition to the loss of the lateral arrangement, the film also looses its structure across the layer. In contrast, XRR and GID measurements performed on monolayers of DSPC and DSPC/PGLa showed that there is no interaction between the lipid and peptide molecules, i.e. the structure of the DSPC monolayer stays intact upon addition of the peptide. Therefore, these two types of molecules do not mix at a molecular level at the air/water interface, but form two independent phases. A sketch of the molecular organisation for a monolayer of the pure lipids and in the presence of PGLa is presented in Figure 24.

Fig. 24: Models of the organisation of the lipid monolayer in the absence and presence of the frog skin antimicrobial peptide, PGLa. Diagrams on the right illustrate variation of the electron density () versus coordinate (z).

Both diffraction techniques clearly show that the antimicrobial peptide PGLa does not interact with DSPC. In contrast, PLGa was shown to interact strongly with DSPG monolayers leading to destruction of the 2D packing of these lipids, subsequently forming a disordered homogeneous molecular mixture of lipids and peptides. This would naturally result in a perturbation of the membrane functionality. Hence, the electrostatic interaction between the positively-charged peptides and these negatively-charged lipid molecules, which are the predominant components in bacterial membranes, is an important step in the discrimination between bacterial and mammalian cell membranes by antimicrobial peptides.

Principal Publication and Authors
O. Konovalov (a), I. Myagkov (b), B. Struth (a), K. Lohner (c), European Biophysics Journal 31, 428, (2002).
(a) ESRF
(b) Research Institute for Physical Problems, Zelenograd (Russia)
(c) Institut für Biophysik und Röntgenstrukturforschung, Austrian Academy of Sciences, Graz (Austria)