to PG-containing bilayers, which has been described by the electrostatic attrac- tion and surface partitioning model. [2] FRET experiments supported the clus- tering of GM1, but not PG, by MG2. Quenching data suggested that MG2 is bound to the sugar region of GM1. The bound peptide assumed a helical struc- ture and induced the leakage of calcein and the coupled flip-flop of lipids, indicating the peptide also forms a toroidal pore in GM1-containing vesicles. However, the membrane permeabilization activity against GM1-containing membranes was weaker than that against PG-doped liposomes in accordance with the trapping of the peptide in the sugar region. These data shed light on AMP-human cell interaction. References. [1] Matsuzaki K (2009) Control of cell selectivity of antimicrobial peptides. Biochim Biophys Acta 1788: 1687-1692. [2] Seelig J (2004) Thermodynamics of lipid-peptide interactions. Biochim Biophys Acta 1666: 40-50. 3091-Pos Board B246 Characterization of Antimicrobial Peptide Insertion in Tethered Bilayer Lipid Membranes by Pulse Amperometry and Linear Sweep Voltammetry Methods Charles G. Cranfield 1,2 , Bruce Cornell 3 , Stephan L. Grage 4 , Paul Duckworth 5 , Sonia Carne 3 , Anne S. Ulrich 4 , Boris Martinac 1,2 . 1 Victor Chang Cardiac Research Institute, Darlinghurst, NSW, Australia, 2 St Vincent’s Clinical School, University of NSW, Sydney, 2052, Australia, 3 Surgical Diagnostics Pty Ltd., Roseville, NSW, Australia, 4 Karlsruhe Institute of Technology, Karlsruhe, Germany, 5 eDAQ Pty Ltd, Denistone East, NSW, Australia. We describe new techniques to study the insertion of pore forming antimicro- bial peptides (AMPs) into tethered bilayer lipid membranes (tBLMs). A conse- quence of tethering a membrane to a gold surface is that electrical contact to the PBS bathing solution is intrinsically capacitive, preventing the direct applica- tion of a steady-state voltage across the bilayer. However, by using pulsed waveforms, defined potentials may be expressed across the membrane for tens to hundreds of milliseconds, and the resulting I-V plots provide valuable data about AMP insertion rates and voltage dependence (Fig 1). Using this technique in the presence of PGLa, we demonstrate how AMP inser- tion into zwitterionic and negatively charged lipid membranes can be rapidly measured and compared. To better understand the voltage dependence of AMP insertion into tBLMs, ramped potentials can also be applied which can determine the potential thresholds of peptide insertion and pore formation. Advantages of using tBLMs:  tBLMs are more robust and longer last- ing than black lipid membranes, or mi- cropipette patches;  Easier, quicker sample preparation than conventional voltage clamp experiments;  Physiologically relevant AMP concen- trations can be used (cf. NMR). 3092-Pos Board B247 All-Atom Molecular Dynamic Simulations of Piscidin 1 and Piscidin 3 in Lipid Bilayers Bradley S. Perrin 1 , Richard M. Venable 1 , Mukesh Sharma 2 , Richard W. Pastor 1 , Myriam W. Cotten 3 . 1 NHLBI/NIH, Rockville, MD, USA, 2 Harvard Medical School, Boston, MA, USA, 3 Hamilton College, Clinton, NY, USA. Piscidin 1 (p1) and the less active piscidin 3 (p3) are a-helical, amphipathic, and cationic antimicrobial peptides (AMPs). They adsorb onto the anionic outer membrane of pathogenic species and induce leakage beyond a threshold pep- tide concentration. While different mechanisms of membrane disruption have been proposed, an atomic-level description of the events leading to cell death is lacking and elucidating these mechanisms aid in the antibiotic design. Here, p1 and p3 in three different lipid bilayers (3:1 DMPC:DMPG, 1:1 POPE:POPG, and 4:1 POPC:Cholesterol) are studied by solid-state NMR spec- troscopy and all-atom molecular dynamics (MD) simulations to identify factors that differentiate the structure, orientation, and depth of insertion of piscidin. 15N-H dipolar coupling calculated for the membrane-bound peptides are in good agreement with values measured by NMR. The tilts of the peptides in the bilayer (t) determined by fitting dipolar waves to the coupling data agree with those of the NMR-derived and MD-average structures when deviation from an ideal a-helix is included; when ideal values are assumed for the dipolar wave fitting, deviations range from 4  to 6  . While the instantaneous tilt fluc- tuates approximately 5 10  in the simulation, averaging over a time series of structures yields the same tilt as one for an average structure (as might be ob- tained from NMR-based structure determination). This is because t z 90. For transmembrane helices (t z 0), the average over a MD time series may be sig- nificantly different from a single averaged structure. Correlation between t and the depth of insertion for p1 shows that as the peptide becomes more buried, the peptide tilts to bury the C-terminus. Moreover, the depth of insertion is 0.5 to 1.5A ˚ greater for p3 than p1, which likely reduces the activity of p3. 3093-Pos Board B248 Interactions of Two Amphipathic Cell-Penetrating Peptides with Complex Model Membranes: Insights from Molecular Dynamics Simulations Jean Helie 1 , Francesca Milletti 2 , Mickael Lelimousin 1 , Charlotte M. Deane 1 , Mark S.P. Sansom 1 . 1 Oxford University, Oxford, United Kingdom, 2 Hoffmann-La Roche, Nutley, NJ, USA. Cell-penetrating peptides (CPP) can permeate cellular membranes and are therefore attractive vectors for gene therapy and drug delivery. However their uptake mechanisms are still poorly understood [1]. Peptides classified as CPP are usually enriched in basic residues and thus positively charged but except from this shared characteristic display a wide range of physico-chemical prop- erties. A unifying interpretation of experimental results is rendered all the more difficult by the great diversity in experimental setups; nevertheless there is evidence that some CPP can be internalized by both endocytic and direct trans- location pathways. In particular, peptide concentration and amphipathicity have been shown to be important for membrane disruption and passive perme- ation. Membrane lipid composition and electrostatic properties also appear to play a crucial role in the peptides activity. Here we apply multi-scale molecular dynamics simulations to gain molecular level insights into the interactions of both a primary amphipathic CPP (Trans- portan) and a secondary amphipathic CPP (Penetratin) with model membranes. Coarse-grained (CG) simulations were performed to investigate the behaviour of the peptides over several microseconds in large, asymmetric bilayers with complex, biologically relevant lipid compositions. The influence of peptide secondary structure was also explored. The CG simulations highlighted differ- ent lipid bilayer perturbations by the two peptides. Membrane permeation was investigated further using more detailed simulation methods (both CG simula- tions with polarisable water model and atomistic representations). [1] A. Ziegler, Thermodynamic studies and binding mechanisms of cell- penetrating peptides with lipids and glycosaminoglycans, Adv. Drug Del. Rev. 60 (2008) 580-597. 3094-Pos Board B249 High Resolution Structures and Structure-Function Relationships in Histidine-Rich Antimicrobial Peptides from Cod Mark McDonald, Michael Mannion, Damien Pike, Krystina Lewis, Mitchell Browne, Matthew L. Rise, Valerie Booth. Memorial University of Newfoundland, St. John’s, NL, Canada. Two paralogous antimicrobial peptide (AMP) sequences, Gad-1 and Gad-2, were previously identified from an Atlantic cod (Gadus morhua) expressed se- quence tag database. Both peptides are rich in histidine, suggesting that their activity might be pH dependent. Indeed, minimal inhibitory concentration (MIC) assays with Gram-negative bacteria demonstrate that the activity of Gad-2 is substantially higher at pH 5 than it is at pH 7, whereas the activity of Gad-1 at pH 5 is similar to its activity at pH 7. The paralogues also appear to differ from each other in their level of activity against Gram-negative bacte- ria, with Gad-1 exhibiting more activity than Gad-2, even at pH 5. Clues to the origin of the different pH dependencies of the activity of the two peptides, as well as the difference in activity of Gad-1 and Gad-2 at pH 5, were provided by circular dichroism (CD) studies and high resolution NMR structures in so- dium dodecyl sulfate (SDS) micelles at pH 5. Gad-1 takes on a helical config- uration from residue 4 to 21. Gad-2 also appears to be predominantly helical but has a kink at the junction between H10 and H11. This deformation in the helix is likely due to the electrostatic repulsion between histidine sidechains at pH 5. 3095-Pos Board B250 Membrane Perturbing Effects of Antimicrobial Peptides: A Systematic Spectroscopic Analysis Daniela Roversi 1 , Lorenzo Giordano 1 , Marta De Zotti 2 , Gianfranco Bocchinfuso 1 , Andrea Farrotti 1 , Sara Bobone 1 , Antonio Palleschi 1 , Kyung-Soo Hahm 3 , Yoonkyung Park 4 , Ferdinando Formaggio 2 , Claudio Toniolo 2 , Lorenzo Stella 1 . 1 University of Rome Tor Vergata, Rome, Italy, 2 University of Padova, Padova, Italy, 3 BioLeaders Corporation, Daejeon, Korea, Republic of, 4 Chosun University, Gwangju, Korea, Republic of. Antimicrobial peptides (AMPs) exhibit a strong activity against a wide range of microorganisms, mainly by perturbing the permeability of bacterial membranes through the formation of pores. However, AMPs effects on membrane proper- ties probably extend beyond pore-formation. We performed a systematic 600a Wednesday, February 6, 2013