Translocation of cationic amphipathic peptides across the membranes of pure phospholipid giant vesicles.

The ability of amphipathic polypeptides with substantial net positive charges to translocate across lipid membranes is a fundamental problem in physical biochemistry. These peptides should not passively cross the bilayer nonpolar region, but they do. Here we present a method to measure peptide translocation and test it on three representative membrane-active peptides. In samples of giant unilamellar vesicles (GUVs) prepared by electroformation, some GUVs enclose inner vesicles. When these GUVs are added to a peptide solution containing a membrane-impermeant fluorescent dye (carboxyfluorescein), the peptide permeabilizes the outer membrane, and dye enters the outer GUV, which then exhibits green fluorescence. The inner vesicles remain dark if the peptide does not cross the outer membrane. However, if the peptide translocates, it permeabilizes the inner vesicles as well, which then show fluorescence. We also measure translocation, simultaneously on the same GUV, by the appearance of fluorescently labeled peptides on the inner vesicle membranes. All three peptides examined are able to translocate, but to different extents. Peptides with smaller Gibbs energies of insertion into the membrane translocate more easily. Further, translocation and influx occur broadly over the same period, but with very different kinetics. Translocation across the outer membrane follows approximately an exponential rise, with a characteristic time of 10 min. Influx occurs more abruptly. In the outer vesicle, influx happens before most of the translocation. However, some peptides cross the membrane before any influx is observed. In the inner vesicles, influx occurs abruptly sometime during peptide translocation across the membrane of the outer vesicle.

Hemolytic activity of membrane-active peptides correlates with the thermodynamics of binding to 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine bilayers.

Understanding the mechanisms of antimicrobial, cytolytic and cell-penetrating peptides is important for the design of new peptides to be used as cargo-delivery systems or antimicrobials. But these peptides should not be hemolytic. Recently, we designed a series of such membrane-active peptides and tested several hypotheses about their mechanisms on model membranes. To that end, the Gibbs free energy of binding to 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC) vesicles was determined experimentally. Because the main lipid components of the outermost monolayer of erythrocyte membranes are zwitterionic, like POPC, we hypothesized that the Gibbs free energy of binding of these peptides to POPC would also be a good indicator of their hemolytic activity. Now, the hemolytic activity of those synthetic peptides was examined by measuring the lysis of sheep erythrocyte suspensions after peptide addition. Indeed, the Gibbs free energy of binding was in good correlation with the hemolytic activity, which was represented by the concentration of peptide in solution that produced 50 % hemolysis. Furthermore, with two exceptions, those peptides that caused graded dye release from POPC vesicles were also hemolytic, while most of those that caused all-or-none release were not.