pubs.acs.org/Langmuir © 2010 American Chemical Society
Molecular Structure of the Dioctadecyldimethylammonium Bromide (DODAB) Bilayer Dorota Jamroz,* Mariusz Kepczynski,* and Maria Nowakowska Faculty of Chemistry, Jagiellonian University, Ingardena 3, 30-060 Krak ow, Poland Received June 8, 2010. Revised Manuscript Received July 14, 2010 Dioctadecyldimethylammonium bromide (DODAB) is a double-chained quaternary ammonium surfactant that assembles in water into bilayer structures. This letter reports the molecular dynamics (MD) computer simulations of the DODAB bilayer at 25 °C. The simulations show that the surfactant membrane arranges spontaneously into the rippled phase (Pβ’) at that temperature. The ordering within the chain fragment closest to the hydrophilic head (carbon atoms 1-5) is relatively low. It grows significantly for the carbon atoms located in the center of the membrane (atoms 6-17). The C6-C17 chain fragments are well aligned and tilted by ca. 15° with respect to the bilayer normal.
Introduction Dioctadecyldimethylammonium bromide (DODAB) is a synthetic double-chained cationic surfactant that tends to aggregate spontaneously in aqueous solution with the formation of bilayer structures.1 The morphology of structures, which are formed in the DODAB dispersions at room temperature, strongly depends on the method of preparation. As was shown previously, upon sonication of the dispersion mostly bilayer fragments are formed.2 However, the extrusion process results in the formation of nonspherical faceted vesicle structures.3 The DODAB vesicles, called cationic liposomes in the literature, have found widespread use both in fundamental studies on interfacial phenomena4 and in practical applications; they may be used as DNA carrier systems for gene transfection5,6 and as vehicles for drug delivery.7,8 Therefore, DODAB dispersions have attracted substantial experimental interest over the last three decades. Several experimental techniques have been used in the study of DODAB dispersions, including differential scanning calorimetry (DSC),9-11 solid-state NMR,9 small X-ray scattering (SAXS),9 fluorescence spectroscopy,12 and electron spin resonance (ESR).13 Although DODAB is one of *Corresponding authors. Phone: þ48 12 6632263 or þ48 12 6632020. Fax: þ48 12 6340515. E-mail:
[email protected] or kepczyns@ chemia.uj.edu.pl. (1) Kunitake, T.; Okahata, Y. J. Am. Chem. Soc. 1977, 99, 3860–3861. (2) Kepczynski, M.; Bednar, J.; Kuzmicz, D.; Wydro, P.; Nowakowska, M. Langmuir 2010, 26, 1551–1556. (3) Lopes, A.; Edwards, K.; Feitosa, E. J. Colloid Interface Sci. 2008, 322, 582–588. (4) Gonc- alves da Silva, A. M.; Rom~ao, R. S.; Lucero Caro, A.; Rodrı´ guez Patino, J. M. J. Colloid Interface Sci. 2004, 270, 417–425. (5) Li, P.; Li, D.; Zhang, L.; Li, G.; Wang, E. Biomaterials 2008, 29, 3617–3624. (6) Barreleiro, P. C.; May, R. P.; Lindman, B. Faraday Discuss. 2003, 122, 191–201. (7) Shi, G.; Guo, W.; Stephenson, S. M.; Lee, R. J. J. Controlled Release 2002, 80, 309–319. (8) Pacheco, L. F.; Carmona-Ribeiro, A. M. J. Colloid Interface Sci. 2003, 258, 146–154. (9) Saveyn, P.; Van der Meeren, P.; Zackrisson, M.; Narayanan, T.; Olsson, U. Soft Matter 2009, 5, 1735–1742. (10) Feitosa, E.; Barreleiro, P. C. A.; Olofsson, G. Chem. Phys. Lipids 2000, 105, 201–213. (11) Coppola, L.; Youssry, M.; Nicotera, I.; Gentile, L. J. Colloid Interface Sci. 2009, 338, 550–557. (12) Feitosa, E.; Rosa Alves, F.; Niemiec, A.; Real Oliveira, M. E. C. D.; Castanheira, E. M. S.; Baptista, A. L. F. Langmuir 2006, 22, 3579–3585. (13) Benatti, C. R.; Feitosa, E.; Fernandez, R. M.; Lamy-Freund, M. T. Chem. Phys. Lipids 2001, 111, 93–104.
15076 DOI: 10.1021/la102324p
the most frequently studied vesicle-forming cationic synthetic amphiphiles, the structure and properties of its bilayer are still not fully understood. The molecular dynamics simulations (MD) have become a great tool in the theoretical analysis of complex systems at the molecular level. Computer simulations provide a complementary view to experiments, revealing a level of detail that is often very difficult or even impossible to achieve experimentally. MD studies of lipid bilayers such as dipalmitoylphosphatidylcholine (DPPC)14 and dimyristoylphosphatidylcholine (DMPC)15,16 have been reported. In this letter, we report the results of MD simulations of the DODAB system, which were performed to gain insight into the molecular organization of the surfactant in its bilayer. The hydrated DODAB bilayer was simulated for 90 ns, and detailed information regarding the organization, orientation, and conformation of the surfactant molecules was obtained from the resulting trajectory. Finally, the structural parameters have been compared with several experimental findings.
Methods Initial Structure. Four arbitrarily chosen DODAB conformers with elongated chains were used to build the bilayer. Their geometries were optimized on the semiempirical PM3 level using Gaussian 03 program.17 The initial structure was generated by arranging the conformers, picked up in a random fashion, into an (14) Berger, O.; Edholm, O.; Jahnig, F. Biophys. J. 1997, 72, 2002–2013. (15) Chiu, S.-W.; Clark, M.; Balaji, V.; Subramaniam, S.; Scott, H. L.; Jakobsson, E. Biophys. J. 1995, 69, 1230–1245. (16) Smondyrev, A. M.; Berkowitz, M. L. J. Comput. Chem. 1999, 20, 531–545. (17) 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.
Published on Web 08/30/2010
Langmuir 2010, 26(19), 15076–15079
Jamr oz et al.
Letter
8 8 2 rectangular lattice with dimensions of 6.4 6.4 5.0 nm3. Each molecule was rotated afterwards around the z axis (perpendicular to the bilayer) by a random angle χ (| χ | e 180°) and tilted by a random angle θ (|θ| e 30°). The resulting bilayer of 128 DODAB molecules was energy minimized with a steep descent method to relax any possible short contact, after which it was hydrated with 3464 water molecules (27 water molecules/ DODAB molecule). The hydrated bilayer was submitted once more to the energy-minimization procedure. The DODAB molecule was parametrized using the Berger14 lipid force field. The parameters for the Br- anion were those from Joung and Cheatham,18 together with the SPC water model. Simulation Parameters. Simulations were performed with the Gromacs version 4.0 package.19 Periodic boundary conditions were applied in all three directions. The simulated system was maintained at a temperature of 298 K and under a pressure of 1 bar according to the NPT ensemble regime. The temperature was controlled by the velocity rescaling thermostat,20 and the pressure was controlled by the isotropic Parrinello-Rahman barostat (except for the first 100 ps where the Berendsen barostat was used). The simulation was carried out for 90 ns total, with the first 10 ns considered to be the equilibration process and the remaining 80 ns treated as the productive run. A time step of 2 fs was used (1 fs for the first 100 ps). The long-range electrostatic potential was calculated using the particle-mesh Ewald (PME) method with a Coulomb cutoff radius of 1.0 nm. The Lennard-Jones potential was calculated with a twin range cutoff with the radii set at 1.0 and 1.4 nm and the pair list updated every 10 steps. The bond lengths were constrained with the LINCS algorithm.
Results and Discussion To study the structural parameter of the DODAB membrane, a planar bilayer of 64 DODAB molecules on each leaflet and 27 water molecules per surfactant was simulated at a constant temperature of 25 °C, which is below the experimentally found gel-liquid transition temperature of 44-45 °C.10 The time profile of the potential energy (a), the system density (b), and the area per DODAB molecule (c) for the first 30 ns of the simulation were plotted (Supporting Information, Figure s1). The system was observed to equilibrate within the first 20 ns, during which all of the parameters converged to constant values. To ensure maximum reliability, the trajectory analysis was performed on data from the final 50 ns of the simulation. The average density of the hydrated DODAB bilayer is very close to that of pure water and equals 0.99 g cm-3. This value agrees well with that determined experimentally (ca. 0.98 g cm-3).9 The area per DODAB molecule was calculated as the ratio of the xy face area of the simulation box to the number of DODAB molecules per monolayer (64). The obtained value of 58 A˚2 is in very good agreement with the reported experimental result of 56 A˚ 2, determined at 20 °C using Langmuir monolayer measurements.2 Figure 1 shows snapshots taken from MD simulations at t = 0 (A) and 90 ns (B). The initial structure was regular with the DODAB molecules arranged into two parallel monolayers. During the simulation, a wavelike arrangement of the surfactant molecules appeared. Such a structure developed spontaneously within the first 20 ns of the simulation and remained unchanged for the whole simulation period. To make sure that this molecular arrangement is not an artifact due to specific simulation conditions, we performed additional simulations of three other systems of the same size. They differed in the initial conformation and (18) Joung, I. S.; Cheatham, T. E. J. Phys. Chem. B 2008, 112, 9020–9041. (19) Hess, B.; Kutzner, C.; van der Spoel, D.; Lindahl, E. J. Chem. Theory Comput. 2008, 4, 435–447. (20) Bussi, G.; Donadio, D.; Parrinello, M. J. Chem. Phys. 2007, 126, 14101.
Langmuir 2010, 26(19), 15076–15079
Figure 1. Snapshots of the DODAB bilayer at t = 0 (A) and 90 ns (B). (A stretch of three simulation boxes along the x direction is shown.) Surfactant molecules in the upper layer are yellow, and those in the lower layer are violet. The hydrophilic heads (ammonium groups) are represented as spheres, similarly to the bromine ions (orange). The water molecules are displayed as small spheres in cyan.
arrangement of DODAB molecules and slightly altered the number of water molecules. In all cases, the system rapidly evolved into the rippled structure of a very similar ripple geometry and molecular organization. Its characteristic feature is the presence of two regions with very different molecular ordering patterns. In the first region, stretched approximately for 3 nm, the hydrocarbon chains of the DODAB molecules in the upper and lower monolayers are interdigitated, similarly to a zipper in the zipped position, which results in a reduced bilayer thickness. The hydrocarbon chains are stretched out and tightly packed in a manner characteristic of the gel phase. The other domain, of some 4 nm in length, may be characterized by noticeable chain disorder that is peculiar to the liquid phase of a membrane. The upper and lower monolayers are separated by a distance of approximately double the chain length (the zipper in the unzipped position), and in this domain, the membrane thickness is significantly larger. Both domains alternate in a periodic manner, giving a characteristic, asymmetric sawtooth profile. Such a membrane structure is known as the ripple phase (Pβ’) and was observed in both the X-ray scattering experiments21 and the simulation on lipid bilayers.22-24 The ripple phase appears at the temperature below the main transition temperature Tm and it is considered to be an intermediate phase between a gel and a liquid. Figure 2 shows the (symmetrical) mass density distribution along the normal to the bilayer. The presence of the two domains of distinctly different molecular arrangements is supported by double maxima appearing on the density profile curves for the ammonium headgroup and for the Br- anion. The first, narrower maxima correspond to the domain of “zipped” DODAB layers, and the other, much broader maxima, to the “unzipped” layers region. The distance between the corresponding maxima gives an estimate of the bilayer thickness, defined as the average distance between the centers of mass of the hydrophilic groups in both (21) Sengupta, K.; Raghunathan, V. A.; Katsaras, J. Phys. Rev. E 2003, 68, 31710. (22) de Vries, A. H.; Yefimov, S.; Mark, A. E.; Marrink, S. J. Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 5392–5396. (23) Lenz, O.; Schmid, F. Phys. Rev. Lett. 2007, 98, 58104. (24) Sun, X.; Gezelter, J. D. J. Phys. Chem. B 2008, 112, 1968–1975.
DOI: 10.1021/la102324p
15077
Letter
Figure 2. Snapshot of a 2 nm slice of the DODAB bilayer (upper panel). Mass density profiles of the various groups along the bilayer normal (lower panel). The density profiles were averaged over the last 50 ns of the trajectory. The lines are colored as follows: (blue) water, (violet) ammonium headgroups, (red) hydrocarbon chains, (orange) terminal CH3 groups, and (green) bromine ions.
layers. It equals ca. 2.4 nm in the “zipped” domain and ca. 4.0 nm in the “unzipped” one. The experimentally determined thickness of the DODAB membrane in the gel state is equal to 3.42 nm.9 This value reflects the average thickness of the surfactant membrane. Fitting four Gaussian curves to the density profile of the ammonium headgroups allowed us to estimate the percentage of DODAB molecules in the zipped domain as 34%. The average membrane thickness, calculated as a weighted sum of both values, gives 3.46 nm. The density profile curve for the terminal carbon atoms of the hydrocarbon chains reflects the wide distribution of these atoms far beyond the central region of the bilayer. The distribution of Br- anions shows that these ions do not diffuse into the thick water layer but rather remain close to the positively charged hydrophilic heads. The deuterium order parameter, SCD, is an experimentally obtainable indicator of the hydrocarbon chain order state (Supporting Information). The SCD order parameter for both DODAB chains (marked as a and b) is shown in Figure 3A. The ordering patterns for both chains are very similar. The ordering within the chain fragment closest to the hydrophilic head (carbon atoms 1-5) is relatively low; it grows significantly for the middle-chain carbon atoms (atoms 6-17), reaching a maximum value of SCD = 0.37, and then it decreases slightly for the last three atoms of the chain. Thus, the hydrocarbon chain order is higher in the center of the DODAB membrane than in the region close to the ammonium headgroups. These findings show that the structure of the hydrophobic region of the DODAB bilayer is quite different from that of the phospholipid membrane, where the highly ordered region of the acyl chains is close to the carbonyl groups and the chain ordering drops precipitously toward the center of the membrane.14-16 The orientation of the chains has been analyzed by calculating the distribution of the hydrocarbon chain z-tilt angle (Figure 3B). 15078 DOI: 10.1021/la102324p
Jamr oz et al.
Figure 3. Deuterium order parameters for both DODAB hydrocarbon chains (A) and the distribution of tilt for two molecular axes (B). The first one is defined by the C1 and C5 atoms of each chain, and the second one, by the C6 and C17 atoms.
Figure 4. Distribution of dihedral angles within the hydrophilic headgroups (A) [(blue line) dihedral CN2-C1, (red line) dihedral N-C2, and (green line) dihedral CN4-18] and along the hydrocarbon chains (B) [(red line) dihedral C1-C4, (blue line) dihedral C2-C5, (green line) dihedral C6-C9, and (green open circle) dihedral C12-C15].
Tilt angles for two chain fragments have been considered, with the first fragment covering the five carbon atoms closest to the Langmuir 2010, 26(19), 15076–15079
Jamr oz et al.
nitrogen atom (the C1-C5 fragment) and the other one stretching over the remaining part of the chain (the C6-C17 fragment). It may be noticed that the C6-C17 chain fragments are well-aligned and tilted by ca. 15° with respect to the bilayer normal. No such order may be observed for the C1-C5 chain fragments. The distribution of the tilt angles is very wide, showing that the alignment of these fragments has great orientational freedom. Only a part of them is collinear with the main molecular axis. This is clearly a result of the observed tendency to open both chains at the N atom in order to avoid strong interactions between the initial carbon atoms. Further information concerning the chains conformations may be obtained by analyzing the distribution of the dihedrals along the chains (Figure 4A,B). The dihedral distribution profiles are exactly the same for both chains, so only the results for the “a” chain have been presented. The CN2-C1 dihedral defines the position of the first atom of the “a” chain (C1) with respect to both methyl groups. As may be concluded by comparing the areas under the maxima on the distribution curve, both values of the dihedral angle (i.e., 180 and 60°) are represented with nearly equal probabilities (52 and 48%, respectively). This means that the C1 atom (and C18 for the “b” chain) assumes the trans conformation with respect to one of the methyl groups (180° dihedral angle) and the gauche conformation with respect to the other CH3 group (60° dihedral angle). This finding is contrary to an intuitive expectation that the carbon atom would be preferably found at a position equidistant from both methyl groups (120° dihedral angle). However, the observed conformation seems to be energetically favorable because it secures a greater distance of the initial fragments of both hydrocarbon chains. Indeed, the preferable value of the dihedral defining the relative positions of both chains at the head (the CN4-C18 dihedral) turns out to be close to 120°, which signifies that in this region both chains assume a sterically favorable “open scissors” conformation. The next carbon atom of the chain (C2) assumes with high probability the trans conformation
Langmuir 2010, 26(19), 15076–15079
Letter
with respect to the nitrogen atom (dihedreal N-C2). In regions more remote from the hydrophilic head, the trans conformation of the hydrocarbon chain is by far the dominating one (Figure 4B). The relative contribution of the gauche conformation at the beginning of the chain (dihedral C1-C4) is ca. 32%, and it decreases rapidly to only 6% at greater distances.
Conclusions Our MD calculations show that at 25 °C the DODAB bilayer arranges spontaneously into the rippled phase. Such a phase was observed in numerous membrane systems upon rapid cooling of the liquid phase to below the main transition temperature. Our initial, artificially generated structure with rather considerable intermolecular distances may be considered to be similar to the liquid phase, and imposing the simulation temperature of 25 °C is therefore adequate for its rapid cooling. The ordering of the surfactant molecules within the membrane and their preferential conformation are mainly governed by the repulsion interaction of the two initial fragments of the carbohydrate chains, which, being bound to the same nitrogen atom, have to assume the openscissors conformation in order to minimize the steric hindrance. Acknowledgment. This project was operated within the Foundation for Polish Science Team Programme cofinanced by the EU European Regional Development Fund, PolyMed, TEAM/20082/6. The calculations were performed at The Academic Computer Centre CYFRONET in Krak ow, Poland (grant no. MEiN/SG/ 3700/UJ/089/2006). Supporting Information Available: The time profile of the potential energy, system density, and area per surfactant molecule for the first 30 ns of the simulation. A description of the molecular order parameter. This material is available free of charge via the Internet at http://pubs.acs.org.
DOI: 10.1021/la102324p
15079
