Molecular Dynamics Simulation Study of the Interaction of Cationic

Oct 4, 2012 - Recent studies have shown that OPEs cause catastrophic damage to large unilamellar vesicles. This study uses classical molecular dynamic...
0 downloads 0 Views 1MB Size
Download PDF
Letter pubs.acs.org/Langmuir

Molecular Dynamics Simulation Study of the Interaction of Cationic Biocides with Lipid Bilayers: Aggregation Effects and Bilayer Damage Eric H. Hill,†,‡ Kelly Stratton,† David G. Whitten,‡ and Deborah G. Evans*,†,§ †

The Nanoscience and Microsystems Program, ‡The Center for Biomedical Engineering, and §The Department of Chemistry and Chemical Biology, University of New Mexico, Albuquerque, New Mexico 87131, United States S Supporting Information *

ABSTRACT: A novel class of phenylene ethynylene polyelectrolyte oligomers (OPEs) has been found to be effective biocidal agents against a variety of pathogens. The mechanism of attack is not yet fully understood. Recent studies have shown that OPEs cause catastrophic damage to large unilamellar vesicles. This study uses classical molecular dynamics (MD) simulations to understand how OPEs interact with model lipid bilayers. All-atom molecular dynamics simulations show that aggregates of OPEs inserted into the membrane cause significant structural damage and create a channel, or pore, that allows significant leakage of water through the membrane on the 0.1 μs time scale.

■

■

INTRODUCTION

Controlling pathogenic bacteria is a global healthcare issue. Gram-negative pathogen Pseudomonas aeruginosa is a leading cause of infection in hospitals and is associated with up to 1.4 million deaths per year. The rapid emergence of multidrugresistant bacteria has dramatically reduced effective treatment options, prompting the search for new classes of antimicrobial agents with alternative mechanisms of action that do not induce resistance. A number of studies have demonstrated that a novel class of phenylene ethynylene-based conjugated polyelectrolyte oligomers (OPEs) exhibit significant light-activated biocidal activity and killing efficacy in the dark against a broad spectrum of pathogens.1−4 Recent fluorescence experiments show that some OPEs can cause irreversible and catastrophic damage to large unilamellar vesicles, and it has also been observed that exposure to a specific OPE (OPE-3) results in a very quick disruption of the vesicle.3 The mechanism of how the lipid bilayer is damaged by this class of compounds is not yet well understood. The purpose of this study is to use classical molecular dynamics (MD) simulations to determine how OPE-3 interacts with and damages the structural integrity of lipid bilayers. Molecular dynamics simulations have been effectively used to help determine the mechanism of disruption of lipid membranes by antimicrobial peptides, including recent studies on magainin and melittin.5−18 In this study, we use full atomistic MD simulations to elucidate the nature of OPE−lipid bilayer interactions, to follow the subsequent changes to the lipid bilayer after OPE insertion, and to determine the mechanism of OPE-induced damage on bacterial cell membrane mimics on the molecular level. © 2012 American Chemical Society

METHODS

Simulation Setup and Details. The chemical structure of the specific series of OPEs chosen for this study is shown in Figure 1a. The

Figure 1. (Left) Chemical structure the cationic OPEs studied. OPE-3 has n (number of repeat units) equal to 3. (Right) Starting configuration of three OPEs inserted fully into the lipid bilayer. The phosphorus atoms of DOPE/DOPG are yellow, and the three OPE-3s are green. The lipid tails are gray, and solvated NaCl ions are shown. The inset shows a top-down view of the upper leaflet of the bilayer. simulations discussed in this article focus on a three-repeat-unit oligomer (OPE-3) that has been shown to have potent antimicrobial activity in the dark.1−4 All simulations use a bacterial cell membrane mimic with DOPE and DOPG phospholipids in a 4:1 ratio that is solvated and ionized with 0.19 mM NaCl and OPEs starting from various positions inserted into the membrane. All molecular dynamics simulations were performed with the CHARMM36 lipid force field Received: August 3, 2012 Revised: October 3, 2012 Published: October 4, 2012 14849

dx.doi.org/10.1021/la303158c | Langmuir 2012, 28, 14849−14854

Langmuir

Letter

and the NAMD software package.20 The 4:1 DOPE/DOPG lipid bilayer was built and pre-equilibrated using CHARMM-GUI19 and then equilibrated for 775 ps. After the insertion of one or more OPEs, each combined simulation system was subjected to a series of energyminimization steps and then propagated using a small time step of 1 fs for 775 ps to ensure that the lipid bilayer was not destabilized by the insertion process. A number of different simulations have been performed using one or more OPEs interacting with the lipid bilayer. The initial configuration of three OPEs used as a starting configuration is shown in Figure 1b. A discussion of the evolution of this system along a molecular dynamics trajectory is the focus of this article. Other configurations of OPEs for different molecular dynamics simulations are depicted in Figures S1−S6 of the Supporting Information. In Figure 1b, the starting configuration has three OPEs, shown in green, placed with their backbones 21 Å apart in a triangular pattern and fully inserted into the lipid bilayer. Waters were modeled using the TIP3 water model, and all simulations were carried out in the NPT ensemble using periodic boundary conditions. Full system electrostatics was used. Details of all simulation configurations are shown in Table S1 in the Supporting Information. The Langevin temperature thermostat was used without coupling to hydrogens and with a damping coefficient of 1/ps and was set at physiological conditions, 310.15 K. The Langevin barostat with an oscillation period of 50 fs and a decay of 25 fs was employed to maintain the pressure at 1 atm. The OPE force fields were obtained from SwissParam21 and optimized by electronic structure calculations using Gaussian 03.22 Simulation Analysis. Lipid bilayer properties before and after OPE insertion were characterized using the bilayer thickness, lateral area per lipid, and atomic density profiles across the lipid bilayer. The bilayer thickness was measured using the phosphate groups on the upper and lower leaflets with the program GridMAT-MD.23 The permeation rate of water through the lipid bilayer was estimated by counting the number of water molecules that crossed the interfacial planes during a simulation trajectory. The identification of particular water molecules and all visualizations and graphical images were obtained using VMD.24

in Figures S2 and S3 in the Supporting Information). From these simulation results, it was observed that one or more OPEs returned to associate strongly with the upper leaflet or were pushed further out of the bilayer. Other OPEs in these simulations were observed to associate with the phosphate headgroups on the lower leaflet and insert more fully into the lipid bilayer along the trajectory. Interestingly, for the OPEs that inserted further into the lipid layer, two cationic quaternary ammonium groups on the same phenyl ring (with a distance of less than 15 Å between them) bridged the upper and lower bilayer leaflets, leading to a significant distortion of the lipid bilayer. Given the limitations of running full atomistic molecular dynamics simulations beyond 100 ns and anticipating that the insertion process into the lipid bilayer from solution is expected to take much longer with numerous modes of insertion likely (based on our shorter time scale results in Figures S1−S3), it is reasonable to consider initial OPE configurations where one or more OPEs are initially placed fully inserted into the bilayer (e.g., the initial configurations depicted by the green OPEs in Figure 1b and those shown in Figures S4−S6). Using the initial configurations of OPE-3s fully inserted into the lipid bilayer, molecular dynamics simulation trajectories were monitored for up to 100 ns. Starting from these configurations, significant thinning of the lipid bilayer was observed around 20 ns, even when only one OPE is inserted (Figure S4). The lipid bilayer thickness was also monitored as the system evolved from the initial configuration of three OPEs depicted in Figure 1b. The average bilayer thickness was calculated over the course of the simulation trajectory. Snapshots of the average lipid bilayer thickness, as determined from Gridmat-MD,23 are shown in Figure 2. The dramatic changes observed in bilayer thickness are due to the attraction between the cationic quaternary ammonium groups on the OPE and the phosphate groups on the phospholipids. The initial cause of bilayer thinning is similar to the half-insertion simulations, where the electrostatic attraction of the cationic groups on the OPE backbone pull the phosphate headgroups of the phospholipids toward the center of the lipid bilayer. Beyond 30 ns along this simulation trajectory with the three OPEs fully inserted, a much more drastic reduction in the thickness of the bilayer is observed, starting from an initial average thickness of 37 Å to around 10−20 Å in the immediate vicinity of the OPEs in the later stages of the simulation. Changes in the morphology of the lipid bilayer can also be seen clearly from the area occupied per lipid, which increases noticeably for lipids in the vicinity of the OPES (Figure S7 in the Supporting Information). During the course of this simulation, two OPEs were observed to associate closely with each other in the lipid bilayer. This can be seen clearly in Figure 2, where two OPEs are aggregated with one another at 40 ns, for example. The third OPE was somewhat removed from the other two laterally. The two OPEs that were closer to one another had their cationic groups facing toward one another and aligned at similar depths in the lipid bilayer. The close proximity of the charged groups in the lipids further contributed to the disruption of the lipid bilayer and most likely accounts for the dramatic decrease in membrane thickness observed at later times in the simulation trajectory. The cationic groups of the aligned and aggregated OPEs in the lipids ultimately result in extensive damage to the integrity of the bilayer. The presence of the cationic groups on the

■

RESULTS AND DISCUSSION Our simulations focus on the same DOPE/DOPG composition used in experimental studies of dye leakage from large unilamellar vesicles, where OPE-3 is shown to be very effective at causing leakage.3 To validate the force field parameters for the OPEs, we also performed simulations of OPEs in water solvent. In these simulations, an interesting phenomenon was observed: the OPEs readily aggregate to form complexes of two or three in solution. The major interaction that leads to the formation of these aggregates is the cofacial association of the backbone phenyl rings. Experimental results show that the OPEs readily form complexes with scaffolds such as carboxymethylcellulose, DNA, or anionic quencher 1,2dianthroquinone sulfonate.25 The average length of OPE-3 in water is 45 Å, and the average distance between the two cationic quaternary ammonium groups on a single benzene ring is 13.5 Å. For comparison, the average thickness of the solvated and equilibrated DOPE/DOPG lipid bilayer prior to insertion was found to be 37 Å. Simulations to study the interaction of OPE-3s with the lipid bilayer were started with the OPEs outside of the lipid bilayer, as depicted in Figure S1. The OPEs were found to associate with the lipid bilayer within 10 ns. In this simulation and in several similar simulations, some of the cationic OPE side groups become embedded in the upper leaflet in less than 50 ns, as the cationic quaternary ammonium groups associate with the phosphate headgroups of DOPG and DOPE. Molecular dynamics simulations were also performed where the OPEs were initially positioned as half-inserted into the membrane (as 14850

dx.doi.org/10.1021/la303158c | Langmuir 2012, 28, 14849−14854

Langmuir

Letter

Figure 2. Average lateral bilayer thickness along the simulation trajectory for the initial OPE configuration in Figure 1b. The OPE positions are shown in black, and the total area shown in each box is 75 Å × 75 Å. The cationic side groups are shown in magenta.

Figure 3. Snapshots along the simulation trajectory show the timeline of the formation of a water channel and pore in the lipid bilayer. (Top) A transverse view of the lipid bilayer, showing only the OPEs in green, the P in the lipid headgroups in yellow, and the water molecules in blue. Other atoms are omitted for clarity. (Bottom) A cross-sectional view of the membrane showing a view of the upper leaflet. The DOPEs are shown in orange, the DOPGs are shown in red, the OPEs are shown in green, and the water molecules are shown in blue.

OPEs. The water molecules, shown in blue in Figure 3b, are shown to be closely associated with this pair of OPES-3s. This view also shows clearly how the water associates along the channel created by the OPEs and shows that the lipid bilayer is significantly damaged after about 60 ns. At 70 ns, one of the two OPEs involved in the formation of the pore slipped mostly out of the pore and associated with the upper leaflet of the lipid bilayer. At 60−80 ns, significant gaps or holes are created when the bilayer is viewed from the upper leaflet. The lipid morphology in the bilayer is observed to change drastically along the simulation trajectory (Figure S8). In addition, lipids are also observed to move from one leaflet to the other leaflet in the area that surrounds the OPEs. In Figure 4, snapshots taken along this simulation trajectory show a lipid transitioning between leaflets.

aggregated OPEs provides a polar channel across the lipid bilayer, in a region where the thickness of the bilayer has significantly decreased. As a result of the cationic groups in this damaged region, an infiltration of water into the lipid bilayer is observed after a fairly short time (20−40 ns) along the simulation trajectory, followed by the creation of a water channel or “pore” as the simulation evolves. In Figure 3, a timeline of these changes to the lipid bilayer and the infiltration of water, starting from the configuration in Figure 1b, are clearly illustrated. Thinning of the bilayer around the OPEs is observed within 20 ns, and water infiltration into the lipid bilayer is observed at around 40 ns. By 60 ns, the presence of a channel of water between the upper and lower leaflets is clearly seen. In Figure 3b, a cross-sectional view of the lipid bilayer from the upper leaflet clearly shows how these changes are brought about by the aggregation of two of the 14851

dx.doi.org/10.1021/la303158c | Langmuir 2012, 28, 14849−14854

Langmuir

Letter

Figure 4. DOPG in the region of the pore (shown in red) between 40 and 60 ns on the lower leaflet flipping and becoming stable on the upper leaflet after 100 ns.

Figure 5. (a) Nitrogen atoms in four cationic quaternary ammonium groups are illustrated in gray, red, yellow, and purple. The backbones of the OPEs creating the pore are shown in green. Waters are shown as diffuse blue spheres, and a specific water molecule is shown in dark blue. (b) Two specific water molecules that pass through the pore are monitored as they pass through the channel: the distance between each water molecule and the colored nitrogen atoms is shown in that color (red, yellow, purple, or gray) as the simulation proceeds.

pass through the pore, 153 in the +z direction and 195 in the −z direction. After the formation of the water channel at 30 ns, the initial progress of water molecules through the pore was observed to occur via a simple mechanism: the water molecules individually move down a “ladder” created by the multiple cationic side chains of the OPEs to end up on the opposite leaflet. The progress of individual water molecules crossing through this pore was followed to illustrate this mechanism clearly. In Figure 5, the proximity of specific water molecules to the nitrogen atom in the cationic quaternary ammonium groups is tracked in time along the simulation trajectory. For the water molecule shown in the top panel of Figure 5b, it is clear that the water molecule “jumps” from close proximity to the nitrogen atom shown in purple to the yellow and red nitrogen atoms at 39.29 ns. Similar jumps are observed for other water molecules seen traversing the pore at other snapshots early in the simulation. Essentially, the water molecules associate strongly with the cationic groups in the lipid layer and jump from one cationic group to the next along a polar ladder created by the OPE side chains. Once the pore has grown in size, the water molecules can freely flow from leaflet to leaflet.

This mechanism is aided by the significant OPE-induced distortions of the phosphate headgroup positions in the bilayer. This lipid movement begins to occur after 45 ns in the simulation trajectory and is fully completed by 100 ns. Similar lipid flip-flop events have also been observed in simulations of pore-forming antimicrobial peptides such as magainin with lipid bilayers.26−30 The creation of a polar channel across the lipid bilayer along this simulation trajectory is also shown to result in a significant leakage of water through the lipid bilayer in the vicinity of the OPEs. The passage of water molecules through the channel occurs as a result of the presence of the cationic groups along the channel and the proximity of the upper and lower leaflets in this damaged zone. The formation of the water channel is observed at around 30 ns in Figure 3. The passage of specific water molecules was followed through this channel. Water molecules that crossed the bilayer were detected by defining planes parallel to the bilayer leaflets. Water molecules that cross these planes from either direction (+z or −z) were identified and tracked individually. Between 30 and 120 ns along the simulation trajectory, 346 water molecules were detected to 14852

dx.doi.org/10.1021/la303158c | Langmuir 2012, 28, 14849−14854

Langmuir

■

The formation of a pore following the insertion of three OPEs is observed from other starting configurations, where the initial positions of the OPEs in the bilayer are altered. For example, for another simulation trajectory shown in Figure S6, similar aggregation of the OPEs and lipid bilayer damage has been observed. In the latter simulation, a pore suddenly forms at 100 ns, and once formed, it seems to undergo more rapid growth and is larger than in the simulation from the initial configuration shown in Figure 1b. In Figure S6, the three OPEs all span the membrane during the entire process of pore formation and growth. It is interesting that the insertion of one or two OPEs into the bilayer (Figures S4 and S5) results in thinning of the bilayer and even water moving along the cationic ladder. However, no formation of a large pore for any sustained amount of time is observed.

CONCLUSIONS Molecular dynamics simulations of OPE-3 interacting with a model bacterial membrane mimic show that these OPEs strongly associate with and disrupt the lipid bilayer. Although a single OPE can insert into and partially distort the membrane, the aggregation of three OPEs in the lipid bilayer results in extensive damage and the formation of a water channel and a pore through which significant water leakage occurs. Multiple cationic side chains provide a favorable polar environment inside the lipid bilayer for the incursion and passage of water molecules. These simulations provide insight into the mechanism associated with the antimicrobial activity of this class of compounds in the dark. The cationic side groups of OPE-3 initially create a ladder for water molecules to leak through a pore in the lipid bilayer on the nanosecond time scale. This corroborates recent experimental observations of rapid vesicle leakage caused by OPE-3.3 This new knowledge of the mechanism will be used in future synergistic experimental and computational studies to design antimicrobial agents based on the phenylene-ethynylene motif with higher efficacy and specificity. ASSOCIATED CONTENT

S Supporting Information *

Molecular dynamics timelines of all other simulations discussed, the analysis of the area per lipid on upper and lower leaflets, and the lipid morphology tracked along a simulation trajectory. This material is available free of charge via the Internet at http://pubs.acs.org.

■

REFERENCES

(1) Tang, Y.; Corbitt, T. S.; Parthasarathy, A.; Zhou, Z.; Schanze, K. S.; Whitten, D. G. Light-Induced Antibacterial Activity of Symmetrical and Asymmetrical Oligophenylene Ethynylenes. Langmuir 2011, 27, 4956−62. (2) Wang, Y.; Corbitt, T. S.; Jett, S. D.; Tang, Y.; Schanze, K. S.; Chi, E. Y.; Whitten, D. G. Direct Visualization of Bactericidal Action of Cationic Conjugated Polyelectrolytes and Oligomers. Langmuir 2011, 28, 65−70. (3) Wang, Y.; Tang, Y.; Zhou, Z.; Ji, E.; Lopez, G. P.; Chi, E. Y.; Schanze, K. S.; Whitten, D. G. Membrane Perturbation Activity of Cationic Phenylene Ethynylene Oligomers and Polymers: Selectivity against Model Bacterial and Mammalian Membranes. Langmuir 2010, 26, 12509−12514. (4) Wang, Y.; Chi, E. Y.; Schanze, K. S.; Whitten, D. G. Membrane Activity of Antimicrobial Phenylene Ethynylene Based Polymers and Oligomers. Soft Matter 2012, 8, 8547−8558. (5) Shepherd, C. M.; Vogel, H. J.; Tieleman, D. P. Dermaseptin S3 with Lipid Bilayers: Molecular-Dynamics Simulations. Biochem. J. 2003, 370, 233−243. (6) Ivanov, I.; Vemparala, S.; Pophristic, V.; Kuroda, K.; DeGrado, W. F.; McCammon, J. A.; Klein, M. L. Characterization of Nonbiological Antimicrobial Polymers in Aqueous Solution and at Water−Lipid Interfaces from All-Atom Molecular Dynamics. J. Am. Chem. Soc. 2006, 128, 1778−1779. (7) Leontiadou, H.; Mark, A. E.; Marrink, S. J. Antimicrobial Peptides in Action. J. Am. Chem. Soc. 2006, 128, 12156−12161. (8) Sengupta, D.; Leontiadou, H.; Mark, A. E.; Marrink, S. Toroidal Pores Formed by Antimicrobial Peptides Show Significant Disorder. Biochim. Biophys. Acta 2008, 1778, 2308−2317. (9) Irudayam, S. J.; Berkowitz, M. L. Binding and Reorientation of Melittin in a POPC Bilayer: Computer Simulations. Biochim. Biophys. Acta 2012, 1818, 2975−2981. (10) Manna, M.; Mukhopadhyay, C. Cause and Effect of MelittinInduced Pore Formation: A Computational Approach. Langmuir 2009, 25, 12235−12242. (11) Tsai, C.-W.; Hsu, N.-Y.; Wang, C.-H.; Lu, C.-Y.; Chang, Y.; Tsai, H.-H. G.; Ruaan, R.-C. Coupling Molecular Dynamics Simulations with Experiments for the Rational Design of Indolicidin-Analogous Antimicrobial Peptides. J. Mol. Biol. 2009, 392, 837−854. (12) Dittmer, J.; Thøgersen, L.; Underhaug, J.; Bertelsen, K.; Vosegaard, T.; Pedersen, J. M.; Schiøtt, B.; Tajkhorshid, E.; Skrydstrup, T.; Nielsen, N. C. Incorporation of Antimicrobial Peptides into Membranes: A Combined Liquid-State NMR and Molecular Dynamics Study of Alamethicin in DMPC/DHPC Bicelles. J. Phys. Chem. B 2009, 113, 6928−6937. (13) Irudayam, S. J.; Berkowitz, M. L. Influence of the Arrangement and Secondary Structure of Melittin Peptides on the Formation and Stability of Toroidal Pores. Biochim. Biophys. Acta 2011, 1808, 2258− 2266. (14) Soliman, W.; Bhattacharjee, S.; Kaur, K. Interaction of an Antimicrobial Peptide with a Model Lipid Bilayer Using Molecular Dynamics Simulation. Langmuir 2009, 25, 6591−6595. (15) Santo, K. P.; Berkowitz, M. L. Difference between Magainin-2 and Melittin Assemblies in Phosphatidylcholine Bilayers: Results from Coarse-Grained Simulations. J. Phys. Chem. B 2012, 116, 3021−3030. (16) Lee, H.; Larson, R. G. Lipid Bilayer Curvature and Pore Formation Induced by Charged Linear Polymers and Dendrimers: The Effect of Molecular Shape. J. Phys. Chem. B 2008, 112, 12279− 12285. (17) Lin, J.-H.; Baumgaertner, A. Molecular Dynamics Simulations of Hydrophobic and Amphiphatic Proteins Interacting with a Lipid Bilayer Membrane. Comp. Theor. Polym. Sci. 2000, 10, 97−102. (18) Woo, H.-J.; Wallqvist, A. Spontaneous Buckling of Lipid Bilayer and Vesicle Budding Induced by Antimicrobial Peptide Magainin 2: A Coarse-Grained Simulation Study. J. Phys. Chem. B 2011, 115, 8122− 8129.

■

■

Letter

AUTHOR INFORMATION

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank the Defense Threat Reduction Agency for supporting this research through grants W911NF07-1-0079 and HDTRA108-1-0053. K.S. was supported by the NSMS program through the National Science Foundation Research Experience for Undergraduates (REU). E.H.H. was partially supported by the NSMS program through a U.S. Department of Education GAANN award. We thank the Center for Advanced Research Computing at the University of New Mexico for providing access to computing resources. 14853

dx.doi.org/10.1021/la303158c | Langmuir 2012, 28, 14849−14854

Langmuir

Letter

(19) Jo, S.; Kim, T.; Iyer, V. G.; Im, W. CHARMM-GUI: A WebBased Graphical User Interface for CHARMM. J. Comput. Chem. 2008, 29, 1859−1865. (20) Phillips, J. C.; Braun, R.; Wang, W.; Gumbart, J.; Tajkhorshid, E.; Villa, E.; Chipot, C.; Skeel, R. D.; Kale, L.; Schulten, K. Scalable Molecular Dynamics with NAMD. J. Comput. Chem. 2005, 26, 1781− 1802; http://www.ks.uiuc.edu/Research/namd/. (21) Zoete, V.; Cuendet, M. A.; Grosdidier, A.; Michielin, O. SwissParam, a Fast Force Field Generation Tool for Small Organic Molecules. J. Comput. Chem. 2011, 32, 2359−2368. (22) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03, revision C.02; Gaussian, Inc.: Wallingford, CT, 2004. (23) Allen, W. J.; Lemkul, J. A.; Bevan, D. R. GridMAT-MD: A Gridbased Membrane Analysis Tool for Use With Molecular Dynamics. J. Comput. Chem. 2009, 30, 1952−1958. (24) Humphrey, W.; Dalke, A.; Schulten, K. VMD - Visual Molecular Dynamics. J. Mol. Graphics 1996, 14, 33−38. (25) Tang, Y.; Hill, E. H.; Zhou, Z.; Evans, D. G.; Schanze, K. S.; Whitten, D. G. Synthesis, Self-Assembly, and Photophysical Properties of Cationic Oligo(p-phenyleneethynylene)s. Langmuir 2011, 27, 4945−4955. (26) Fattal, E; Nir, S.; Parente, R. A.; Szoka, F. C., Jr. Pore-Forming Peptides Induce Rapid Phospholipid Flip-Flop in Membranes. Biochemistry 1994, 33, 6721−6731. (27) Matsuzaki, K.; Murase, O.; Fujii, N.; Miyajima, K. An Antimicrobial Peptide, Magainin 2, Induced Rapid Flip-Flop of Phospholipids Coupled with Pore Formation and Peptide Translocation. Biochemistry 1996, 35, 11361−11368. (28) Tieleman, D. P.; Marrink, S. Lipids Out of Equilibrium: Energetics of Desorption and Pore Mediated Flip-Flop. J. Am. Chem. Soc. 2006, 128, 12462−12467. (29) Gurtovenko, A. A.; Vattulainen, I. Molecular Mechanism for Lipid Flip-Flops. J. Phys. Chem. B 2007, 111, 13554−13559. (30) Hwang, H.; Schatz, G. C.; Ratner, M. A. Coarse-Grained Molecular Dynamics Study of Cyclic Peptide Nanotube Insertion into a Lipid Bilayer. J. Phys. Chem. A 2009, 113, 4780−4787.

14854

dx.doi.org/10.1021/la303158c | Langmuir 2012, 28, 14849−14854