Cholesterol, sphingolipids, and glycolipids: what do we know about their role in raft-like membranes?

Chemistry and Physics of Lipids 184 (2014) 82–104

Contents lists available at ScienceDirect

Chemistry and Physics of Lipids journal homepage: www.elsevier.com/locate/chemphyslip

Review

Cholesterol, sphingolipids, and glycolipids: What do we know about their role in raft-like membranes? Tomasz Róg a , Ilpo Vattulainen a,b, * a b

Department of Physics, Tampere University of Technology, Tampere, Finland MEMPHYS-Center for Biomembrane Physics, University of Southern Denmark, Odense, Denmark

A R T I C L E I N F O

A B S T R A C T

Article history: Received 17 August 2014 Received in revised form 24 October 2014 Accepted 25 October 2014 Available online 1 November 2014

Lipids rafts are considered to be functional nanoscale membrane domains enriched in cholesterol and sphingolipids, characteristic in particular of the external leaflet of cell membranes. Lipids, together with membrane-associated proteins, are therefore considered to form nanoscale units with potential specific functions. Although the understanding of the structure of rafts in living cells is quite limited, the possible functions of rafts are widely discussed in the literature, highlighting their importance in cellular functions. In this review, we discuss the understanding of rafts that has emerged based on recent atomistic and coarse-grained molecular dynamics simulation studies on the key lipid raft components, which include cholesterol, sphingolipids, glycolipids, and the proteins interacting with these classes of lipids. The simulation results are compared to experiments when possible. ã 2014 Elsevier Ireland Ltd. All rights reserved.

Keywords: Lipid raft Membrane protein Peptide Receptor Membrane domain Molecular dynamics simulations Coarse-grained simulations

Contents 1. 2.

3.

4.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cholesterol effect on phosphatidylcholines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ordering and condensing effects of cholesterol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Ordering and condensing effects of cholesterol-like small molecules . . . . . . . . . . . . 2.2. Cholesterol and the eye lens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3. Interactions of small molecules with membranes in cholesterol-rich environments 2.4. Cholesterol effect on membrane permeability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.5. Flip-flops and lipid desorption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.6. Effects of sterol structure alterations on bilayer properties . . . . . . . . . . . . . . . . . . . . 2.7. Use of cholesterol in drug carriers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.8. Role of cholesterol in lipid oxidative stress . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.9. Sphingomyelin and sphingomyelin-cholesterol interactions . . . . . . . . . . . . . . . . . . . . . . . . . Pure sphingomyelin bilayers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Sphingomyelin in binary mixtures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. Ternary mixtures of sphingomyelin, cholesterol, and PC . . . . . . . . . . . . . . . . . . . . . . 3.3. Sphingomyelin diversity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4. Glycolipids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Lipopolysaccharide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1.

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

83 84 84 85 86 87 87 88 88 89 89 90 90 90 90 91 92 92

Abbreviations: SM, sphingomyelin; OSM, oleoyl-SM; PC, phosphatidylcholine; POPC, 1-palmitoyl-2-oleoyl-PC; DOPC, dioleoyl-PC; DMPC, dimyristoyl-PC; DPPC, dipalmitoyl-PC; PGPC, 1-palmitoyl-2-glutaryl-sn-glycero-3-phosphocholine; POPE, 1-palmitoyl-2-oleoyl-phosphatidylethanolamine; GalCer, galactosyl-cerebroside; GM1, Gal5-b1,3-GalNAc4-b1,4-(NeuAc3-a2,3)-Gal2-b1,4-Glc1-b1,1-Cer; GM3, NeuAc3-a2,3-Gal2-b1,4-Glc1-b1,1-Cer; LPS, Lipopolysaccharide; MD, molecular dynamics; CG, coarse-grained; QM, quantum mechanical; Lo, liquid-ordered; Ld, liquid-disordered; NBD, nitrobenzoxadiazole; PEG, poly(ethylene) glycol; NH, amide group; OC, carbonyl group; OH, hydroxyl group; NMR, nuclear magnetic resonance. * Corresponding author at: Department of Physics, Tampere University of Technology, Tampere, Finland. E-mail address: Ilpo.Vattulainen@tut.fi (I. Vattulainen). http://dx.doi.org/10.1016/j.chemphyslip.2014.10.004 0009-3084/ ã 2014 Elsevier Ireland Ltd. All rights reserved.

T. Róg, I. Vattulainen / Chemistry and Physics of Lipids 184 (2014) 82–104

4.2. 4.3.

5.

6. 7. 8.

Glycerol-based glycolipids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cerebrosides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Galactosylcerebroside . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3.1. GM1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3.2. GM3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3.3. Sugar adsorption at the water-membrane interface . . . . . . . . . . . . . . . . . . . . . . . 4.4. Effects of cholesterol on proteins and peptides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Indirect cholesterol effects: hydrophobic mismatch and lateral pressure profiles 5.1. 5.2. Cholesterol effect on peptide adsorption and insertion into membranes . . . . . . Specific interactions between proteins and cholesterol . . . . . . . . . . . . . . . . . . . . 5.3. Coarse-grained models of cholesterol and their applications . . . . . . . . . . . . . . . . . . . . . Force field issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conflict of interest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Transparency document . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. Introduction Biological membranes contain a complex mixture of numerous lipid species (Niemela et al., 2009). For instance, in a very recent study, thousands of lipid species were found in a single sample of adipose tissue (Pietilainen et al., 2011). This huge diversity of lipids arises from, for example, variations in polar headgroups (neutral, charged, or carbohydrate), variations in length and (un) saturation of the hydrocarbon chains, variations in the main lipid backbone (glycerol vs. sphingosine), and all possible combinations of these variants. Lipids are not randomly distributed in cells (van Meer et al., 2008) but they are highly organized according to their function. For instance, on a cellular level charged cardiolipins are typically found only in mitochondria, while sphingolipids are mainly found in the external leaflet of cell membranes. The well-defined organization of lipids also relates to how the lipids are distributed in the membranes themselves. For example, many cellular membranes are asymmetric. The internal (intracellular, cytosolic) leaflet of plasma membranes is typically composed of charged phosphatidylserines, large amounts of phosphatidylethanolamines, and a smaller number of phosphatidylcholines (PCs). In contrast, the outer (extracellular) leaflet is largely composed of sphingolipids, which include a great fraction of glycolipids, and PCs. Cholesterol, being present in both leaflets, is also an important component of the cell membrane, though the details of its transmembrane distribution remain debated (Maxfield and Mondal, 2006). Because sphingolipids have predominantly saturated chains, and PCs are predominately unsaturated, the composition of the extracellular leaflet can be approximated as a mixture of cholesterol, saturated lipids, and unsaturated lipids. In model membranes, the behavior of this lipid mixture is well characterized (Almeida, 2009). It is known that cholesterol separates with the saturated lipids into a liquid-ordered (Lo) phase while unsaturated lipids segregate into a liquid-disordered (Ld) phase (Rheinstädter and Mouritsen, 2013). In model membranes, the domains in question are relatively large and can hence be visualized quite easily. In membranes of living cells, the situation is more complicated as the discussion below brings out. The existence of domains enriched in cholesterol and sphingolipids was postulated already decades ago. Eventually, they became known as lipid rafts, and nowadays they are known or suggested to be associated with numerous cellular functions (Lingwood et al., 2009; Lingwood and Simons, 2010). Yet as biological membranes are much more complex than model membranes, the characterization and visualization of lipid rafts in cells have been largely incomplete. This lack of identifiable lipid rafts in cells may be

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

83

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . .

92 92 93 93 94 94 94 94 95 96 96 97 98 98 98 98 98

explained by the presence of membrane proteins, which can interfere with lipid-lipid interactions and thus reduce the tendency of lipids to phase separate (Yethiraj and Weisshaar, 2007). It is also known that proteins alter the dynamics of lipids around them, and consequently there are protein-lipid complexes containing of the order of hundreds of lipids migrating together with individual proteins (Niemelä et al., 2009). It is possible that these are the smallest functional units able to carry out the functions in cell membranes. Given that the size of such complexes is about 10 nm or less, the challenge to identify them with even super-resolution microscopy techniques has been too demanding until now. For the same reason, the structure of nanoscale rafts in cell membranes remains unclear. Molecular dynamics (MD) simulations have become an important tool in structural biology. The MD method can provide information about atomic-scale mechanisms that are often inaccessible with current experimental techniques, thus atomistic simulations are often used to complement experimental studies (Lyubartsev and Rabinovich, 2011; Vattulainen and Róg, 2011; Hug, 2012). MD simulations can also yield important insight into largescale behavior such as phase separation and diffusion, when simplified (coarse-grained) molecular models are used instead of atomistic ones. The information emerging from atomistic and molecular simulations can be highly useful in interpreting experimental results. For example, MD simulations of fluorescent probes or spin labels can show how the presence of the probe changes the properties of a biological system, and how the behavior of the probe might differ from that of the surrounding lipids (Loura and Ramalho, 2011; Jurkiewicz et al., 2012; Kemmerer et al., 2013; Timr et al., 2014). Also, atomistic simulations can be carried out under conditions that match the ones used in experiments, and hence simulation data can reveal nanoscale mechanisms that are out of reach through experiments. Recently, this strategy was used to identify the mechanism how cholesterol is able to inhibit the function of a glycoreceptor in a lipid membrane (Lingwood et al., 2011b). The primary limitations of MD simulations have been the small time and length scales for modeling objects of interest. However, significant advances over the last decade have improved the scales that can be reached through simulations. For example, atomic MD simulations of b2-adrenergic receptors embedded in a lipid bilayer recently achieved a timeframe of 30 ms (Rosenbaum et al., 2011). Also, the size capacity of simulated systems has increased to allow simulations of objects as large as lipoproteins (Murtola et al., 2011) or polymer-coated bilayers that are relevant to drug delivery (Lehtinen et al., 2012). Further, the models developed for MD simulations are constantly under intensive development. For

84


instance, new lipid models have been developed within the framework of the CHARMM and Amber force fields, which can reproduce a variety of experimental data (Klauda et al., 2010; Jämbeck and Lyubartsev, 2012a). Moving on, due to increasing efficiency of the simulation programs and the continuously increasing computing capacity, there is a great demand for new analysis tools, and a lot of work is being devoted to reach this goal. For instance, as the sizes of simulated lipid membrane structures are increasing steadily, analysis of local curvature effects due to membrane undulations becomes important. Standard analysis tools for considering these and many other complex problems are currently lacking (Gapsys et al., 2013). However, due to the commitment of the people in the field, the situation will get better in this regard, too. In this review, we concentrate on recent atomistic MD simulation studies relevant to lipid rafts. These include interactions of cholesterol with PCs, sphingomyelins, and glycolipids (see Fig. 1). We also discuss studies on membrane proteins and peptides by focusing on the influence of cholesterol on their properties. Although we mainly discuss the recent views that have emerged from atomistic simulations, we also briefly discuss the insight gained from coarse-grained simulations. Comparison to experimental studies is made whenever possible and appropriate, thereby considering the validity of molecular simulations and the added value emerging from the coupling of experiments with simulations. 2. Cholesterol effect on phosphatidylcholines 2.1. Ordering and condensing effects of cholesterol Cholesterol affects a variety of membrane properties such as the ordering of lipid hydrocarbon chains (the so-called ordering effect) and membrane packing (the so-called condensing effect). A large set of atomistic MD studies that have discussed these effects has been reviewed in previous articles (Róg et al., 2009a; Berkowitz, 2009; Pandit and Scott, 2009). Therefore, here, we will only discuss recent accomplishments in MD studies. The tilt of cholesterol steroid ring was previously shown to be an important parameter that correlates with the degree that cholesterol and other sterols could condense and order membrane lipids (Aittoniemi et al., 2006; Olsen et al., 2009). The idea was first presented by Aittoniemi et al., who showed for a number of sterols that the (decreasing) tilt of a sterol correlates with its (increasing)

capability to order the acyl chain conformations of nearby lipids. More recently, Khelashvili et al. (2010a) showed that, with increasing cholesterol concentrations, the cholesterol hydroxyl group was induced to migrate more towards the water phase, and this location correlated with a smaller tilt angle. They hence suggested that cholesterol concentration (ranging from 0.01 to 40 mol%) affects the tilt and its correlation to changes in bilayer properties (Khelashvili et al., 2010a). The changes in cholesterol tilt observed as a function of cholesterol concentration have been explained by the interplay between entropy, which leads to random cholesterol orientations, and the unfavorability of direct cholesterol–cholesterol interactions. The vertical location of cholesterol in the membrane has also been suggested to be correlated with the sterol ordering capability (Róg and Pasenkiewicz-Gierula, 2003). Furthermore, Khelashvili and Harries (2013a) have given support for the idea that the cholesterol tilt and splay moduli are quantitatively connected to the mechanical properties induced by cholesterol in a lipid bilayer. Importance of cholesterol tilt was also shown in ternary mixtures of saturated DPPC with unsaturated DOPC and cholesterol (de Joannis et al., 2011), where cholesterol tilt was shown to correlate with the relative affinity of DPPC and DOPC for cholesterol-rich membrane patches. A review that focuses on the importance of cholesterol tilt was recently published by Khelashvili and Harries (2013b). The condensing effect of cholesterol in a DOPC bilayer was carefully examined by Alwarawrah et al., (2010). They performed simulations with 14 different cholesterol concentrations, ranging from 0 to 66 mol%. A detailed analysis showed that bilayer condensation was mostly the result of packing at the depth where the steroid ring was located. A detailed analysis of the surface area per DOPC and per cholesterol molecule showed that large differences were obtained with different methods; for example, the area per cholesterol ranged from 0.27 nm2 to 0.43 nm2. For comparison, recent experimental results suggested areas of 0.23– 0.26 nm2 (Gallova et al., 2010). The difference may be explained by difficulties to decompose the surface area to the hard crosssectional area of a molecule and the free area (volume) around it. This complex issue and the influence of cholesterol on free volume distribution inside lipid bilayers have been discussed in detail by Falck et al. (2004) and Edholm and Nagle (2005). One of the most discussed mechanisms associated with cholesterol condensing and ordering effects is the so-called “umbrella model” (Huang and Feigenson, 1999). In that model, cholesterol, which has a very small headgroup, avoids exposing its

Fig. 1. Typical membrane lipid structures. (Left) Chemical and three-dimensional structures of cholesterol; (right) chemical structures of (top) dipalmitoylphosphatidylcholine (DPPC), and (bottom) stearoyl-sphingomyelin (SSM).


hydrophilic ring to water by avoiding close cholesterol–cholesterol contacts and by forming complexes with phospholipids (Fig. 2). To test that hypothesis, Dai et al. (2010) performed simulations of preformed cholesterol clusters in two PC bilayers (DPPC and DOPC). They found that cholesterol clusters were unstable and quickly disintegrated. The presence of a large cluster was associated with numerous bilayer instabilities and deformations, including flip-flop motions of cholesterol molecules rarely observed in MD studies. A lack of direct cholesterol–cholesterol contacts has also been demonstrated on two-dimensional radial distribution functions. These functions showed that cholesterol preferred to be located in the second coordination shell to avoid direct contact with other cholesterol molecules (Martinez-Seara et al., 2010). Interestingly, these two-dimensional radial distribution functions showed that the distribution of lipid molecules around cholesterol molecules had a three-fold symmetry (Fig. 3a), which arose from the presence of methyl groups on one of the sides of the sterol ring (b-side). When two methyl groups were removed from the cholesterol ring, the three-fold symmetry was no longer observed (Fig. 3b). The observed differences in the symmetry of the radial distribution functions for these two different cholesterol derivatives might be related to differences in their effects on the phase behavior of the binary bilayers. Indeed, some sterols that do not have three-fold symmetry also do not induce liquid-ordered phases (e.g., lanosterol; Beattie et al., 2005). Cholesterol also affects other properties of lipid bilayers, which are most likely related to the above-discussed effects. For example, studies on membrane electroporation showed that, when membrane cholesterol concentrations were above 10 mol%, a higher minimum electric field was required to form pores through the bilayer (Fernandez et al., 2010). This is in line with earlier findings that have suggested cholesterol to render lipid bilayers more stable in the sense that cholesterol-containing membranes are more resistant to formation of water pores in tensionless membranes, as well as to formation of pores due to mechanical stress, compared to bilayers without cholesterol (Gurtovenko et al., 2010; Shigematsu et al., 2014). In studies of laterally heterogeneous lipid bilayers composed of patches in the liquid-ordered and liquid-disordered phases, it was shown that electroporation occurs in the liquiddisordered phase and that the strength of the effect induced by an electric field depends on the size of the disordered domains (Reigada, 2014). Interestingly, electroporation was not observed at the domain boundaries. The potential issue of the influence of the simulation protocol (effect of intensity and time of the electric pulses) on the mechanism of electroporation and the role of

85

cholesterol has also been discussed in the literature (Casciola et al., 2014). Cholesterol interactions with unsaturated PCs have been investigated in numerous theoretical and experimental studies (e.g., Martinez-Seara et al., 2008; Wydro et al., 2011). A recent study (Zhao et al., 2011) that investigated cholesterol interactions with two isomers of a lipid with a conjugated double bond in the hydrocarbon chains (cis 9 trans 11 vs. trans 10 cis 12) found that cholesterol had a similar influence on both isoforms. However, the effect of cholesterol on lipids with a conjugated double bond was found to be lower than on lipids with a single double bond such as POPC (Khelashvili et al., 2014). Another study (Rosetti and Pastorino, 2012) used coarse-grained (CG) simulations to investigate the behavior of different mixtures of saturated and unsaturated PCs when combined with cholesterol. They observed that symmetrical (two similarly unsaturated tails) and asymmetrical lipids (one saturated and one unsaturated tail) behaved differently, and the importance of the degree of saturation was documented. In combined neutron diffraction and CG simulations, Kucerka et al. (2010) found that cholesterol showed a strong preference for saturated DMPC molecules over polyunsaturated lipids in mixed bilayers. Moreover, the orientation of cholesterol in saturated DMPC (vertical) was different than its orientation in polyunsaturated lipids (horizontal). Pan and co-workers investigated the effects of cholesterol on ether lipids in combined MD simulation and X-ray scattering studies (Pan et al., 2012). In those studies, cholesterol formed hydrogen bonds mainly with the phosphate groups of ether lipids. However, in ester-type lipids, the cholesterol hydroxyl group mainly formed a hydrogen bond with carbonyl oxygens, located deeper in the bilayer. This altered hydrogen bond pattern caused a shift in cholesterol’s position, more towards the water phase, and it reduced the interactions between the cholesterol ring and the hydrocarbon tails. 2.2. Ordering and condensing effects of cholesterol-like small molecules Cholesterol is not the only molecule in nature that can modify lipid bilayer properties via ordering and condensing effects. Comparative MD simulation studies have shown that diacylglycerol could affect lipid bilayers in a manner similar to cholesterol, but its effect was found to be weaker (Alwarawrah et al., 2012). They postulated that the mechanism of diacylglycerol action was the umbrella effect. Diacylglycerol has a small headgroup compared to the hydrophobic portion, similar to cholesterol. Another group of membrane modifiers, xanthophylls, were

Fig. 2. Schematic representation of the umbrella model. Panel (a) shows the packing of membrane lipids (mushroom-shaped headgroup, grey acyl tails) and cholesterol (small circular headgroup, black square body) in the membrane at low cholesterol concentrations. Wavy lines indicate the aqueous extracellular solution. (b) At intermediate cholesterol concentrations, membrane lipid tails condense, and the headgroups can offer cholesterol (its hydrophobic body) more protection from the aqueous solution. Panel (c) shows the behavior of preformed cholesterol clusters at the beginning of a simulation (reprinted with permission from Dai et al., 2010).

86


Fig. 3. (A) Two-dimensional density distributions for cholesterol molecules around a tagged cholesterol in a DSPC bilayer that contains 20 mol% cholesterol. The results are based on atomistic simulations. Schematic representations (centers) show the structure of the tagged cholesterol in the membrane. (B) Distribution of the sterol around DCHOL (that is a cholesterol-like molecule with the methyl groups C18, C19, and C21 removed) in an DSPC bilayer with 20 mol% D-cholesterol. The different faces of cholesterol are distinguished as follows: the smooth a-face corresponds to the region x < 0, and the rough b-face corresponds to x > 0. The b-face is divided into two subfaces: b1 for y > 0 and b2 for y < 0 (reprinted from Martinez-Seara et al., 2010). The figure illustrates the lack of 3-fold symmetry after the removal of the methyl groups.

postulated to play a role similar to that of cholesterol in prokaryotic membranes; for example, xanthophylls could induce a liquidordered phase in the bilayer membrane (Subczynski et al., 2012a). MD simulation studies of lutein in lipid bilayers showed that this xanthophyll molecule could orient vertically, perpendicular to the membrane surface (Pasenkiewicz-Gierula et al., 2012). Meanwhile, hopanoids, another group of chemical compounds found in bacteria, have a structure similar to cholesterol ring (with an additional 6-member ring and no hydroxyl group located in the first ring). Not surprisingly, hopanoids induce both ordering and condensing effects, though to a smaller extent than cholesterol (Poger and Mark, 2013). The strength of these effects and the similarity to cholesterol behavior depend strongly on the details of the structure of the given hopanoid. For instance, diplopterol has experimentally been shown to have ordering properties as well as the ability to promote the formation of the Lo phase, thus it has been suggested as a bacterial surrogate of cholesterol (Sáenz et al., 2012). Interestingly, also rigid linear fluorescent probes have been observed to have ordering effects similar to those of cholesterol: e.g., pyrene increases local order around it in the liquid-disordered (fluid) phase, but it decreases the order in the solid-ordered (gel) phase, similarly to cholesterol (Curdova et al., 2007). This supports the view that any rigid linear-like molecule that stands upright

along the membrane normal direction would have ordering properties similar to cholesterol. The extent of ordering would depend on the molecule in question, but the behavior in general is expected to be quite alike. The connection to the discussion above about the importance of sterol tilt in determining the level of ordering is obvious and may concern the tilt of other rigid and linear sterol-like molecules, too. 2.3. Cholesterol and the eye lens In the human eye lens, fiber cells have membranes with one of the highest cholesterol concentrations known (Subczynski et al., 2012b). The fiber cell membrane cholesterol-to-phospholipid ratio is 1 in the newborn, and it increases to 1.5 over a typical lifetime. Not surprisingly, the functions of cholesterol in these membranes have been intensively studied, as reviewed by Subczynski et al. (2012b). Three MD simulation studies have addressed the problem of a specific cholesterol effect at high concentrations related to the lens function. In simulations of a bilayer with 50 mol% cholesterol, it was shown that cholesterol decreased membrane roughness compared to pure cholesterol-free bilayers (Plesnar et al., 2012). Apparently, this effect might be important for the light scattering property of the lens, which confers its transparency. Cholesterol was shown to be mobile and well hydrated when organized into a


cholesterol bilayer; in contrast when organized into a crystal, it was less hydrated and completely rigid (Plesnar et al., 2013). In simulations of a DMPC bilayer with 66.7 mol% cholesterol, cholesterol formed pure micro domains; however, when the cholesterol concentration was 50 mol%, micro domains did not occur (O’Connor and Klauda, 2011). This observation agreed with those reported in previous experimental studies (Raguz et al., 2011). Finally, the role of cholesterol and its esters has been studied in the context of CG models for tear fluid films. These will be discussed in Section 6. 2.4. Interactions of small molecules with membranes in cholesterolrich environments Numerous small molecules are preferentially located at the interface between an aqueous solution and the membrane (water– membrane interface). For example, many amino acid side chains are preferentially localized to the membrane–water interface (MacCallum et al., 2008); also, numerous drug molecules localize to this interface (Cramariuc et al., 2012; Boggara et al., 2012; Forst et al., 2014; Kopec and Khandelia, 2014). Cholesterol can change the properties of both the lipid bilayer hydrocarbon core and the water–membrane interface. For example, cholesterol can affect headgroup hydration, headgroup mobility, water ordering, the pattern of hydrogen bonding (Pasenkiewicz-Gierula et al., 2000), the dipole potential (Haldar et al., 2012), and membrane smoothness (discussed above). Therefore, one could expect that cholesterol might affect the behavior of water molecules, and that it can directly interact with water molecules. In our recent studies, we showed that neurotransmitters such as dopamine and its precursor L-dopa exhibit strong interactions with bilayers that contain cholesterol (Orlowski et al., 2012). However, we did not observe direct interactions between these molecules and cholesterol. Dopamine and L-dopa localized to the water-membrane interface, similarly to another neurotransmitter, melatonin (Drolle et al., 2013). Melatonin was also shown to reduce membrane order and membrane thickness. In lipid bilayers with cholesterol, melatonin was shown to decrease cholesterol ordering and condensing effects (Choi et al., 2014). Direct interactions between drug molecules and cholesterol were observed with ketoprofen (a non-steroidal anti-inflammatory drug), aspirin, and piroxicam. However, cholesterol only interacted with the ionized forms, not the neutral forms, of these drug molecules (Markiewicz and Pasenkiewicz-Gierula, 2011). Another study conducted a systematic comparison of bilayers composed of DPPC and sphingomyelin, mixed in bilayers with various levels of cholesterol. They showed that decreasing cholesterol concentrations allowed deeper ethanol penetration into the bilayer (Polley and Vemparala, 2013). Interestingly, ethanol mostly disturbed the hydrogen bonds between cholesterol and sphingomyelin. Pyrene, a fluorescent probe often used in lipid bilayer studies, was shown to localize in the glycerol region of a pure POPC bilayer, and it slightly increased the ordering of acyl tails in fluid membranes (Loura et al., 2013). In gel membranes, however, pyrene increases the disorder around the probe, thereby rendering the membrane fluid in the vicinity of pyrene (Curdova et al., 2007). In bilayers with cholesterol, the position of pyrene did not change; however the presence of label slightly decreased the ordering of the acyl chains. Importantly, the presence of cholesterol reduced hydration in the region that the probe resided. Another commonly used fluorescent probe is diphenylhexatriene, which is used to gauge membrane fluidity through fluorescence anisotropy. Franova et al. studied how accurately this view holds true in cases where the order of a membrane is varied systematically by increasing cholesterol concentration

87

(Franova et al., 2010). It turned out that while the qualitative trends of fluorescence anisotropy measurements were correct, the quantitative results should be taken cautiously since the analysis of fluorescence anisotropy measurement data is often based on a limited number of terms (the lowest terms in Legendre polynomials). The disagreement between the correct data and the results arising from the commonly used fluorescence anisotropy data analysis increases, as the membrane becomes more and more ordered. Sodium ions are known to stably bind with phospholipids’ phosphate and carbonyl groups (Jungwirth 2014; Berkowitz and Vacha, 2012; Vacha et al., 2009). MD simulations together with j-potential measurements showed that cholesterol decreases the degree of sodium binding at a water–membrane interface (Magarkar et al., 2014b). The mechanism of this effect seems to be associated to cholesterol’s hydroxyl group, and in particular its competition for the same binding partners in PC molecules (carbonyl and phosphate groups), as well as the fact that cholesterol introduces a relatively large portion of its hydrophobic surface into the membrane–water interface, thus excluding ions. This particular effect might be important in cholesterol’s interaction with peripheral proteins, as, e.g., it was shown that cholera toxin binds weaker with a membrane and its glycolipid receptor when cholesterol is present (Lingwood et al., 2011b). Contrary to Na+, hydrated protons (H3O+) were shown to have the same affinity to lipid bilayers with and without cholesterol, locating similarly to Na+ between the phosphate and carbonyl groups (Yamashita, 2014). This result, however, has to be verified experimentally or through quantum-mechanical (QM) calculations. 2.5. Cholesterol effect on membrane permeability Cholesterol is known to decrease membrane permeability to water, gases, and small organic molecules (Ohvo-Rekila et al., 2002). The effect of cholesterol on the water permeability of a saturated DPPC bilayer was examined by means of free energy calculations (Saito and Shinoda, 2011). Those studies showed that the free energy barrier increased with increasing cholesterol concentration. Interestingly, cholesterol modified the shape of the free energy profile. With no cholesterol in the bilayer, the highest value of free energy was observed at the bilayer center. When the membrane contained cholesterol, the free energy increased at the level of the cholesterol ring system, but remained unchanged at the bilayer center. These changes resulted from a reduced number of cavities in the bilayer region, due to a reduced number of gauche conformations in the hydrocarbon chains next to the cholesterol ring. Similar results were obtained for a bilayer composed of sphingomyelins; however, even without cholesterol, a pure sphingomyelin bilayer has low permeability (Saito and Shinoda, 2011). Substantial reduction of water permeation by cholesterol was shown in unbiased 2 ms MD simulations (Hong et al., 2014). In a pure POPC bilayer, a single event was observed every 17 ns, while in a POPC–CHOL 2:1 mixture the observations were made every 114 ns. Interestingly, the events where water translocation through a bilayer took place were very rapid, as they lasted for about 1 ns (order of magnitude). Cholesterol was also shown to affect the free energy profile of hypericin, a large aromatic compound (Eriksson and Eriksson, 2011), across the lipid bilayer. The optimum free energy for hypericin was achieved when its location in the bilayer was the same as that of a cholesterol ring. The presence of cholesterol increased the free energy barrier for translocation between the membrane leaflets at the membrane center. This effect depended on the degree of hypericin bromination. Hypericin derivatives with numerous bromine substituents were more hydrophilic, and thus

88


they were less affected by the presence of cholesterol. Cholesterol was also shown to decrease the permeability coefficient of fullerene (C60) as well as to decrease the free energy of transfer of fullerene from water to the membrane (Sun et al., 2014). MD simulations and free energy calculations predict a decrease of permeability for ibuprofen (that is the most commonly used nonsteroidal anti-inflammatory drug) (Khajeh and Modarress, 2014a) and 5-fluorouracil, a drug used in cancer treatment (Khajeh and Modarress, 2014b). Extensive studies analyzed the partitioning of ammonia, ethanol, propane, benzene, and neopentane in four different types of bilayers, including POPE, POPC, DPPC, and DMPC. With cholesterol concentrations of 0–50 mol%, the solute partitioning into the lipid tail decreased; however, the observed effect was stronger than expected, based on the results of experimental studies (Wennberg et al., 2012). This discrepancy was explained by a lack of lateral inhomogeneity in the cholesterol distribution in the studied systems. Nevertheless, some general observations were possible. For example, cholesterol had a greater effect on the partitioning of large molecules, particularly in saturated lipids. The effect was also greater in bilayers composed of phosphatidylethanolamine than in those composed of PC. 2.6. Flip-flops and lipid desorption Lipids’ flip-flop motions and desorption from a membrane are challenging research topics for MD simulations, since the time scales of these processes can be prohibitively long. Therefore, when MD simulations of membranes are being carried out, it is rare to observe spontaneous flip-flops or lipid desorption events even in very long simulations. What is usually done instead is to employ free energy calculations (umbrella sampling) for considerations of the related free energy barriers, which quite certainly are high. Studies of cholesterol effect on flip-flops and desorption of DPPC showed that cholesterol increases the free energy barrier for flipflop motions, but it decreases the free energy of desorption (Bennett et al., 2009a). These studies showed that the translocation mechanism with and without cholesterol is different – in bilayers without cholesterol, a water pore was formed when the headgroup was located in the bilayer center, but in membranes with 20 and 40 mol% of cholesterol pores were not observed to form. Cholesterol flip-flops and desorption in bilayers with various compositions (Bennett et al., 2009b) showed that, in agreement with experimental data, cholesterol flip-flops occur on time scales shorter than 1 s. Flip-flop rate was shown to be dependent on bilayer composition: higher rates were observed in unsaturated bilayers, while the rate was slower in saturated bilayers with a high cholesterol concentration. Cholesterol was further shown to have a high affinity for more ordered and rigid bilayers. These studies were extended to compare flip-flops in a simple POPC bilayer with similar events in rigid raft models composed of SM–POPC–CHOL 1:1:1 mixtures (Bennett and Tieleman, 2012). The estimated lifetime of flip-flop in a POPC bilayer was found to be 20 ms, while in a raft-like bilayer it increased to 30 min. A similar increase of flip-flop lifetime was observed for ceramide and diacylglycerol. Again, cholesterol showed strong preferences for a rigid raft-like bilayer over a more fluid POPC system, while ceramide and diacylglycerol showed almost no preference. A similar conclusion was drawn from free energy studies by Jo et al. (2010), who observed a lower barrier for flip-flop motion in polyunsaturated bilayers compared to monounsaturated membranes. Two-dimensional free energy plots showing how the free energy depends on cholesterol position along a bilayer normal and cholesterol tilt indicated that in the bilayer center cholesterol adopts a perpendicular orientation with respect to bilayer normal.

Here, it should be stressed that although various studies have shown similar qualitative trends, the free energy values found for cholesterol translocation in simulation studies with different force fields do not match. A systematic comparison between the studies based on different force fields cannot be performed, since the temperatures considered in simulation have not been fully comparable. Nevertheless, in the case of a POPC bilayer studied at 303 K with the CHARMM force field, the observed barrier was 29 kJ/mol (Jo et al., 2010). A study at 310 K with the OPLS-AA force field, the barrier was found to be 18 kJ/mol (Neuvonen et al., 2014), and at 323 K with Berger lipids the barrier was 23 kJ/mol (Bennett and Tieleman, 2012). The observed differences seem to be inconsistent and may result from force field inaccuracies or the protocols used in free energy calculations. In this context, one of the major problems in free energy calculations is sampling. Recent studies have shown that especially for lipids with long tails, consideration of desorption from a lipid bilayer requires very long sampling times due to slow relaxation processes, related to local bilayer deformations (Neale et al., 2011). Experimental atomic force microscopy and MD simulation studies for extraction of cholesterol from the Lo and Ld phases (Stetter et al., 2014) showed, not surprisingly, that the force necessary to extract cholesterol from the Lo phase was higher than in the Ld phase. This resulted from stronger nonpolar interactions between cholesterol and the membrane in the Lo phase (condensation effect). In these studies, two energy barriers were observed: one correlated with the length of the steroid ring, and the second with the length of the whole cholesterol molecule. 2.7. Effects of sterol structure alterations on bilayer properties Apart from cholesterol, other sterols are also of high interest, mostly due to their connection with metabolic illnesses, like Smith–Lemli–Opits syndrome and desmosterolosis (Haas et al., 2001). From a structural point of view, although other sterols are closely similar to cholesterol, they cannot substitute for cholesterol in membrane functions. Thus, comparative studies between cholesterol and its analogues are highly interesting. Cholesterol is, by far, the most commonly occurring sterol; the only sterol that can compete in abundance is ergosterol, which is found mostly in fungus. The importance of cholesterol’s structure has thus been investigated in numerous studies, where its membrane properties have been compared to other structurally similar sterols, in both MD and experimental studies (reviewed by Róg et al., 2009a; Mannock et al., 2010) or with the bioinformatics Apache software (Wenz, 2012). Recently, several cholesterol analogues were studied, including 7-dehydrocholesterol, dehydroergosterol, oxysterols, and sterols with a modified headgroup. Liu et al. studied the effect of 7dehydrocholesterol, a precursor in the cholesterol biosynthetic pathway, on lipid bilayer properties (Liu et al., 2011). They found that 7-dehydrocholesterol was almost identical to cholesterol in its effects on membrane order and condensation. This result was slightly different from previous studies, where a very small difference was observed (Róg et al., 2008). This might result from the differences in force field parameters used, but it is most likely a question of data interpretation and error bar estimations, because most of the results were very similar between studies. The effects of 7-dehydrocholesterol on membrane properties were also investigated experimentally (Shrivastava et al., 2008). That study showed that, at high concentrations, 7-dehydrocholesterol affected membrane properties differently than cholesterol. Nevertheless, the MD simulations (Róg et al., 2008) were performed at a 20 mol% concentration, which was within the error range for experimental fluorescence anisotropy data for both studied sterols. However, infrared spectroscopic studies showed that sterol


interactions with DPPC and EggPC lipids were stronger with cholesterol than with 7-dehydrocholesterol (Chen and Tripp, 2012). Dehydroergosterol is a fluorescent derivative of ergosterol, with additional double bonds in the steroid rings, and cholestatrienol is an analogous fluorescent derivative of cholesterol. MD simulation studies showed that both sterol derivatives had properties similar to cholesterol (Robalo et al., 2013). Both increased the ordering of hydrocarbon chains, but slightly less than cholesterol. Not surprisingly, cholestatrienol behavior was closer to that of cholesterol than dehydroergosterol. Dehydroergosterol was found to rotate slightly faster than cholesterol; thus, it was not fully representative of cholesterol for investigating dynamic behavior (Pourmousa et al., 2014). Fluorescent nitrobenzoxadiazole (NBD)labeled cholesterol, with the NBD-label attached to the tail of cholesterol, has also been shown not to reproduce cholesterol behavior very well (Robalo et al., 2013b). It was observed that the NBD-label preferentially locates closer to the water–membrane interface than the bilayer center. Consequently, the labeled molecules tend to adopt a bent conformation that is not typical for cholesterol, and as a result they affect membrane order less than cholesterol. BODIPY-cholesterol in which a BODIPY moiety is attached to the short chain of cholesterol has been shown to have partly similar behavior, that is, the BODIPY moiety alters membrane packing around it and has a minor tendency to locate itself in the membrane–water region (Hölttävuori et al., 2008). Yet the simulations gave support to an idea that BODIPY-cholesterol partitions more favorably to the raft-like Lo phase compared to the Ld phase. b-sitosterol is a major phytosterol. It differs from cholesterol in the tail structure. MD simulations of model bilayers that included various concentrations of b-sitosterol showed that, with increasing concentrations, bilayers became resistant to the disordering effects of cryosolvents, like dimethyl sulfoxide, propylene glycol, ethylene glycol, glycerol, and methanol (Hughes et al., 2013). One of the quite intriguing questions is whether cholesterol is the final product of the evolutionary process to optimize several sterol properties in plasma membranes of eukaryotic cells, where the concentration of cholesterol is particularly large. This topic was explored in studies, where the ordering capability of cholesterol was compared with synthetic sterols whose structure was highly similar to that of cholesterol. In practice, the roles of individual side (methyl) groups of cholesterol were tested one by one in a systematic manner (Róg et al., 2007a; Pöyry et al., 2008). It turned out that it is very difficult to modify cholesterol further by changing the individual details of its structure, and at the same time expect the ordering capability of cholesterol to improve substantially. Though sterol synthesis is rather difficult, 18,19-di-nor-cholesterol (Dchol), one of the sterols considered in our previous simulation studies (Pöyry et al., 2008) was recently synthetized (MydockMcGrane et al., 2014). Very recent experimental studies of Dchol (Krause et al., 2014) showed that the ordering capability of Dchol is lower than that of cholesterol, in agreement with the predictions of earlier simulations (Róg et al., 2007a; Pöyry et al., 2008). Nearest neighbors’ interaction energies were shown to be lower in the case of Dchol compared to cholesterol. These results indicate that cholesterol is better in inducing the Lo phase and that the structure of cholesterol (with both rough and flat faces) is quite optimal, meaning that it is indeed difficult to design better (synthetic) sterols with an even higher ordering ability compared to cholesterol. As to oxysterols, 25-hydroxycholesterol has been shown to affect lipid bilayers that contained cholesterol (Olsen et al., 2011). The presence of this oxysterol affected cholesterol’s position in the bilayer, by shifting it up into the water phase; that shift resulted in higher cholesterol hydration, and from a physiological point of

89

view it became more prone to migrating between different bilayers. Also, 27-hydroxycholesterol has been shown to have lipid disordering properties (Bielska et al., 2014). CG simulations with the MARTINI force field showed that changing the cholesterol hydroxyl group to a group with less polarity leads to an inhibition of domain formation (Perlmutter and Sachs, 2009). These sterols also adopted a perpendicular orientation, and the headgroup deeply penetrated the hydrocarbon phase, similar to the behavior of ketosterone (Róg et al., 2008a) and cholestenone (Neuvonen et al., 2014) studied in atomic simulations. Meanwhile, cholesteryl hemisuccinate is an ester of cholesterol with a larger charge on the headgroup; it is often used to crystallize membrane proteins (see Section 5.3) or for drug delivery (Straubinger, 1993). MD simulations and fluorescence studies showed that cholesteryl hemisuccinate increased membrane ordering, but to a lesser degree than cholesterol (Kulig et al., 2014a,b; Massey, 1998). Similar effects were observed in the case of cholesterol sulfate (Smondyrev and Berkowitz, 2000), which is also a cholesterol ester but with a large headgroup (larger than hydroxyl). Effects of cholesterol structure modifications on flipflop motion were also studied using an implicit membrane model (Parisio and Ferrarini, 2010; Parisio et al., 2012). These studies showed that the effect of changes in the hydrophobic part of a sterol do not affect flip-flop motions, while a decreasing polarity of the head group increases the rate of these motions. However, in the case of ketosterone, the effect was found to be rather small in disagreement with other CG and atomistic simulations.

2.8. Use of cholesterol in drug carriers Liposomes (vesicles) formed from lipid bilayers are among the most promising drug carriers in drug delivery, highlighting their high pharmaceutical interest. It is not surprising that cholesterol is a commonly used component in liposome formulations (Gabizon et al., 1994), as due to the above-mentioned properties, cholesterol renders liposomes mechanically more stable and less permeable. To increase blood stream circulation time, liposomes are often covered with neutral polymer-like poly(ethylene) glycol (PEG) molecules, which shield those “stealth” or “PEGylated” liposomes from interactions with other macromolecules (Bunker, 2012). MD simulations of cholesterol in PEGylated lipid bilayers composed of saturated PCs have shown a substantial difference in both cholesterol behavior as well as the behavior of the PEGylated bilayer with and without cholesterol (Magarkar et al., 2014a). PEG has been observed to specifically enter lipid bilayers next to the rough b-face of the cholesterol molecule; yet in studies of lipid bilayers composed of saturated lipids, in the gel phase PEG did not penetrate the bilayer core (Magarkar et al., 2012) but in the case of fluid bilayers such penetration was observed. Overall, cholesterol in PEGylated bilayers tends to be located higher at the water– membrane interface. Cholesterol effects have also been examined in lipid bilayers composed of catanionic surfactants, which potentially can be used as DNA carriers (Kuo and Chang, 2014). Cholesterol affected these bilayers similarly to typical lipids by decreasing the surface area and by increasing the degree of order in the hydrocarbon chain region. 2.9. Role of cholesterol in lipid oxidative stress Cholesterol behavior in the presence of oxidized lipids was recently investigated both via MD simulations and experiments (Štefl et al., 2014). In these studies, one focused on 1-palmitoyl-2glutaryl-sn-glycero-3-phosphocholine (PGPC) oxidized lipids with a truncated tail ending with a carboxylic group. As the carboxylic

90


group locates at the water–membrane interface, a void under the carboxylic group is formed. Cholesterol was observed to preferentially locate inside the void under the carboxylic group, thereby decreasing PGPC-induced effects on the whole bilayer structure. Another study that is of potentially great interest dealt with the interactions of cholesterol with oxidized lipid species (Khandelia et al., 2014). The simulation results and dynamic light scattering experimental data was interpreted to imply that cholesterol protects lipid bilayers from disruption caused by lipid oxidation, and that this is due to complementary shapes of cholesterol and oxidized lipids which allows cholesterol to sequester conical shaped oxidized lipid species away from phospholipids. There was also support for an idea that mixtures of cholesterol and oxidized lipids self-assemble into bilayers like lysolipid–cholesterol mixtures. 3. Sphingomyelin and sphingomyelin-cholesterol interactions Sphingomyelin is the most common sphingolipid in nature. It is also the most commonly studied; its structure is shown in Fig. 1. Sphingomyelin has three functional groups in the backbone region, an amide (NH), a carbonyl (OC), and a hydroxyl group (OH). Numerous variants of this basic structure exist. These include molecules with additional hydroxyl groups, modifications in the double bond regions, variations in the tail linkage, etc. Detailed reviews of sphingomyelin types and their properties have been published previously (Ramstedt and Slotte, 2002; Slotte, 2013). 3.1. Pure sphingomyelin bilayers Four older studies have presented simulations of bilayers composed solely of sphingomyelins that focused on its hydrogen bonding properties and compared its properties with those of DPPC bilayers (Mombelli et al., 2003; Hyvönen and Kovanen, 2003; Chiu et al., 2003; Niemela et al., 2004). Those studies showed that sphingomyelin bilayers were much more ordered and condensed than bilayers composed of PC molecules with hydrocarbon tails of the same length. In those studies, the area per sphingomyelin molecule was about 1–1.2 nm2 smaller than the area per PC molecule in a DPPC bilayer. The studies showed that a large number of intramolecular hydrogen bonds formed between the hydroxyl groups and phosphate ester oxygens. In contrast, intermolecular hydrogen bonds were formed mainly with an amide group as the hydrogen donor. A slightly different hydrogen bonding pattern was observed by Róg and Pasenkiewicz-Gierula (2006). In this study, it was found that the hydroxyl groups formed intramolecular hydrogen bonds with carbonyl oxygens, but the bilayer remained more ordered and condensed than the equivalent PC bilayer. This subtle difference between the different studies most likely resulted from the different parameterizations used; it pointed out the need for a specific parameterization study of the sphingosine moiety as well as for a careful validation of force fields. This was also stressed in studies by Mombelli et al. (2003), where three different parameter choices for simulations were compared. They showed that the different parameters resulted in substantially different bilayer properties. In most recent studies of sphingomyelin, it was shown that the hydroxyl group was predominately (99% of time) involved in hydrogen bonds with the phosphate oxygen, and the amide group was involved in intermolecular hydrogen bonds with the carbonyl and hydroxyl groups (Venable et al., 2014). The model used in this study is currently only specifically developed for sphingomyelin and will be discussed more extensively in Section 7. Niemela et al. (2006) conducted a systematic comparison of the properties of sphingomyelin and acyl moieties with tails of 16– 24 residues. Not surprisingly, the membrane thickness increased

with the length of the acyl tails. However, there were also moderate differences in other properties, like the surface area per lipid (a small decrease) and the ordering of the acyl tails (a small increase). Also, lateral diffusion slightly slowed down with increases in tail length, but the hydrogen bond pattern was nearly unaffected. Those investigators also studied the effects of changing the position of a single double bond to 4 different locations. The presence of double bonds, as expected, increased the area per lipid and decreased tail ordering. 3.2. Sphingomyelin in binary mixtures Interactions between cholesterol and sphingomyelin were studied in binary mixtures of cholesterol and sphingomyelin via MD simulation, with a focus on hydrogen bonding (Róg and Pasenkiewicz-Gierula, 2006; Khelashvili et al., 2005). Those studies showed formation of a high number of hydrogen bonds, including the interactions of cholesterol with all the functional groups of sphingomyelin. The number of hydrogen bonds formed depended on cholesterol concentration, temperature, and the sphingomyelin type (sphingomyelin-16 vs. sphingomyelin-18). Importantly, the number of hydrogen bonds formed was higher than that observed in a corresponding mixture of cholesterol and PC molecules. For example, in our studies, cholesterol formed twice as many hydrogen bonds with sphingomyelin than it formed with PC (Róg and Pasenkiewicz-Gierula, 2006). We also observed that a substantial number of charge-pairs formed between cholesterol oxygens and the choline groups of sphingomyelin. Zidar et al. (2009) studied five different cholesterol concentrations in a sphingomyelin bilayer, and they found that the packing of lipids was different at low concentrations. These different packing modes were attributed to liquid-ordered and liquid-disordered phases. Zhang et al. (2007) compared the energies of interaction between cholesterol and sphingomyelin, sphingomyelin with an oleoyl acyl tail (OSM), and POPC. Surprisingly, the energies of interaction were very similar in each case. In addition, they found that the flat a-face of cholesterol preferred to interact with the saturated sphingosine tail of the OSM molecule; in contrast, the rough b-face of cholesterol showed no preference for either of the OSM tails. Metcalf and Pandit (2012) studied sphingomyelin in a binary mixture with ceramide, a sphingolipid with a hydroxyl headgroup. They tested these mixtures over the whole range of concentrations (from 0–100 mol%) at four different temperatures. In their systematic study, a range of properties was calculated, which provided solid characteristics of the studied systems. They showed that increasing the concentration of ceramide increased the ordering of the hydrocarbon tails, and at higher concentrations, the authors observed an umbrella-like effect of the sphingomyelin headgroup shielding the ceramide from water. 3.3. Ternary mixtures of sphingomyelin, cholesterol, and PC From the point of view of raft formation, the most interesting studies considered ternary (or higher) lipid mixtures. Unfortunately, those studies are not the most abundant in the literature. The main reason for their scarcity is that, with an increasing number of components, the system size has to be increased to obtain proper statistics; in addition, the simulation time should be sufficiently lengthened to allow proper mixing of the components. Recent studies of lipid mixing showed that in two-component systems, the convergence of lipid mixing requires time scales of the order of 400 ns (Hong et al., 2014). In more complex systems characterized by domain formation with a multitude of different lipid types, and with protein crowding included, the time scales needed for true mixing will be significantly larger, of the order of


seconds or minutes. To avoid time-scale limitations related to the mixing processes, in the first papers to consider raft-like membranes, Pandit et al. (2004a) performed simulations of sphingomyelin–cholesterol clusters in DOPC bilayers, or they observed the early stages of domain formation in a threecomponent bilayer (Pandit et al., 2004b). Those early studies covered only a 5-ns timeframe. A recent review concentrating on simulations of lipid domains discusses this topic quite broadly (Bennett and Tieleman, 2013). Niemela et al. (2007) studied the characteristics of mixtures of sphingomyelin, cholesterol, and POPC, as a model for raft domains. The most interesting observation made in those studies that covered a substantially longer time scale was related to the large differences in the lateral pressure profiles across the membrane compared to non-raft bilayers. The shape of the profile most likely affected the functionality of membrane proteins (discussed more in the next sections). Another key feature that emerged from the study by Niemela et al. dealt with lateral dynamics. It turned out that the lateral diffusion rates of lipids are highly sensitive to the physical properties of different membrane domains, and the diffusion rates in raft-like regions are distinctly different from those in more fluid membranes. In a more recent study, an atomistic 10 ms simulation of a lipid bilayer in the Lo/Ld coexistence revealed the existence of a substructure inside the Lo phase, formed by saturated hydrocarbon chains organized locally into hexagonal structures (Sodt et al., 2014). Currently, it is still rare to achieve a ms time scale in atomic simulations; thus, phase separation and raft formation are not well studied with this methodology. However, these phenomena can be studied with less computationally demanding methods; for example, with mean-fields models (Tumaneng et al., 2011) and CG molecular models discussed below. Interesting conclusions could be drawn about the interactions between cholesterol, sphingomyelin, and POPC, when the first two components were at very low concentrations (Aittoniemi et al., 2007). Those atomistic studies indicated that hydrogen bonding alone could not explain the higher affinity of cholesterol for sphingomyelin. Instead, one must also consider the contributions of charge-pairs between cholesterol and the choline groups, and van Der Waals interactions. Apajalahti et al. (2010) studied the problem of lipid diffusion by focusing on concerted lipid motions in many-component raft-like membranes. In agreement with earlier atomistic simulation studies by Falck et al. (2008) and quasi-elastic neutron scattering experiments by Busch et al. (2010), they found that lipids diffuse in the membrane plane as dynamical complexes of tens to hundreds of lipids. The role of cholesterol, explored by Apajalahti et al. (2010), was observed to be prominent in slowing down the lateral motion in raft-like membrane domains, and in the vicinity of boundaries between physically different membrane regions. Sphingolipids are typically found on the extracellular leaflet of the cell membrane, together with cholesterol and PCs; in contrast, the intracellular leaflet is composed of charged phosphatidylserine, phosphatidylethanolamine, and PC (van Meer et al., 2008). Bhide et al. (2007) studied the role of this asymmetry in MD simulation studies. They constructed symmetric and asymmetric bilayers that included sphingomyelin and phosphatidylserine. The symmetric and asymmetric models had similar properties. Polley et al. (2012) studied asymmetric lipid bilayers, with one leaflet composed of POPC, sphingomyelin, and cholesterol, and the other leaflet composed of POPC and cholesterol. They observed a coupling between the two layers that, at long range, affected the tilt of sphingomyelin molecules in the ordered domain. Given that the time scale issue is still quite difficult to handle in atomistic simulations, one often uses CG models to consider the long-time behavior. With the MARTINI force field, CG simulations

91

of ternary mixtures of saturated lipids, cholesterol, and unsaturated lipids were studied on a time scale of 1–20 ms (Risselada and Marrink, 2008). A mixture of 63 lipid species, mimicking the extracellular and intracellular leaflets of a protein-free cell membrane, was studied for 40 ms (Ingolfsson et al., 2014). These studies showed a small enrichment of cholesterol in the extracellular leaflet, as well as heterogeneous lipid distributions in both leaflets. The locations of cholesterol-rich regions in the two leaflets were shown to be correlated, but at the same time these regions were also shown to fluctuate rapidly, and thus to be transient. Nanodomains of gangliosides and smaller clusters of phosphatidylinositols were observed, too. The MARTINI force field was also used by Hakobyan and Heuer (2014) to study mixing properties of a ternary mixture of cholesterol with saturated and unsaturated lipids. The authors systematically modified force field parameters that affect the rigidity of lipids and cholesterol to observe how rigidity affects domain formation.

3.4. Sphingomyelin diversity The natural diversity of sphingolipids and advances in modern organic chemistry have provided an opportunity to understand the basic sphingomyelin structure by comparing its properties with those of specific natural or synthetic analogues. In extensive experimental studies, Slotte et al. investigated a large set of sphingomyelins to understand the importance of the basic structure (Sergelius et al., 2012; Maula et al., 2011; Sergelius and Slotte, 2011; Jaikishan and Slotte, 2011; Ekholm et al., 2011; Jaikishan et al., 2010). A summary of these works was published by Ramstedt and Slotte (2006). To date, there are two studies (Björkbom et al., 2010, 2011) where MD simulations support experimental data to investigate sphingomyelins with altered structures; in the first study, the choline group was replaced with ethanolamine, and in the second study, the amide and hydroxyl groups were methylated. Natural sphingomyelins typically have a PC headgroup. In contrast, glycerol-based phospholipids often have other types of headgroups, but phosphatidylethanolamine is the most common. In combined experimental and MD studies, Björkbom et al. (2010) systematically studied the effect of substituting choline (trimethyl) with ethanolamine (no methyl) or an intermediate with one or two methyl groups (Fig. 4). Those studies showed that the methyl groups in choline (and intermediates between choline and ethanolamine) formed charge-pairs with cholesterol oxygen. These charge pairs were more frequently formed than the hydrogen bonds formed between cholesterol and lipids, via a hydrogen from the amine or intermediate groups. Thus, sphingomyelin species which have more methyl groups formed stronger interactions with cholesterol, and in turn, cholesterol caused stronger condensation and ordering in bilayers composed of these lipids. This result can also be interpreted in light of the umbrella model, because the large choline group would be more effective than a small amine or intermediate in shielding cholesterol from unfavorable interactions with water (Fig. 4). MD simulations were also used to study three analogues of sphingomyelin with methylated functional groups. The first analogue had a methylated amide group, the second had a methylated hydroxyl group, and the third had methylations on both the hydroxyl and the amide group (Björkbom et al., 2011). MD simulations showed that fewer hydrogen bonds formed between cholesterol and the modified sphingomyelin molecules. That reduction was partly due to the lack of a partner for hydrogen bonding, but it was also due to the fact that the cholesterol position had shifted towards the water phase. At the same time, a larger number of interactions with water and choline group were

92


Fig. 4. Representation of typical interactions of headgroups on sphingomyelin analogues. Abbreviations: N-palmitoyl ceramide phosphoethanolamine (CPE), N-palmitoyl ceramide phosphoethanolamine-N-methyl (CPE-Me), and N-palmitoyl ceramide phosphoethanolamine-N,N-dimethyl (CPE-Me2), N-palmitoyl-sphingomyelin (PSM), cholesterol (CHOL), hydrogen bond (HB), charge pair (CP) (reprinted from Björkbom et al., 2010).

observed. This observation showed that cholesterol–sphingomyelin interactions were sensitive to any structural changes in sphingomyelin and that the sphingomyelin moiety should be considered a functional unit, because all its components are important. Methylation also affected the sphingomyelin–sphingomyelin interactions. In this case, methylation of the hydroxyl group had a drastic effect, but methylation of the amide group only moderately reduced the interaction. These changes in interaction strength were most likely responsible for the lower stability of the gel phase (the main transition temperature Tm was reduced up to 7 C) observed experimentally and the higher disorder in the liquid phase observed both experimentally and in MD simulations. 4. Glycolipids Glycolipids comprise a large, diverse group of lipids that serve numerous cellular functions (Degroote et al., 2004; Kasahara and Sanai, 1999; Curatolo, 1987). Glycolipids are major structural lipids in plants, particularly in photosynthetic membranes, and in the external membrane of gram negative bacteria. However, in animals, they occur in smaller quantities, and their functions are mainly related to cell signaling. These differences in function are associated with the different structures of bacterial, plant, and animal glycolipids. Bacterial glycolipids are based on lipid A, with a long oligosaccharide chain attached; plant glycolipids are typical, glycerol-based lipids, and animal glycolipids belong to the class of sphingolipids. Studies on glycolipids are scarce; thus, in this chapter, we review all the studies, rather than only those related to raft cerebrosides. Computational and structural studies on glycolipids were reviewed previously by Patel and Balaji (2011), DeMarco (2012), Marsh (2012), and Manna et al. (2014). Here we focus on atomistic simulations. We discuss CG models and their results only in a limited manner, since due to the importance of chemical details in carbohydrates, atomistic models are often the model of choice to describe their behavior realistically enough. 4.1. Lipopolysaccharide Five pioneering studies studied bacterial lipopolysaccharide (LPS). Interactions between LPS and Ca2+ ions (Soares et al., 2008; Shroll and Straatsma, 2002) or goethite (Shroll and Straatsma, 2003) were considered. They also described the behavior of a long oligosaccharide known as O-Antigen (Lins and Straatsma, 2001). The membrane models used in those studies were large; thus, only limited timeframes (1–12 ns) were simulated. From the point of view of current knowledge on ion dynamics (e.g., Stepniewski

et al., 2010), particularly Ca2+ ions (Magarkar et al., 2012), this time scale appears to be too short to equilibrate the system; therefore, these studies will most likely have to be extended in future. A more recent study described a 200-ns MD simulation of electroporation in the context of an Escherichia coli external membrane model (Piggot et al., 2011). That paper described the characteristics of the bilayer and discussed the interactions between Mg2+ ions and LPS. 4.2. Glycerol-based glycolipids Glycerol-based glycolipids are typically found in plant membranes, where they may comprise 50 mol% of cell lipids and 75 mol % of lipids in photosynthetic membranes (Dörmann and Benning 2002). The most common glycerol-based glycolipids are digalactosides. In our initial studies of glycolipids, we chose to model monogalactoside and mono-glucoside (Róg et al., 2005; Róg et al., 2007b), because these lipids form lipid bilayers with a main phase transition in the temperature range accessible to MD simulations. Additionally, experimental data on the surface area per lipid molecule were available (Köberl et al., 1998), and that data facilitated the validation of MD simulation results. In contrast, digalactosides form hexagonal phases, and in physiological conditions they coexist in the bilayer with different lipids and proteins (Shipley et al., 1973). In our initial studies, the MD simulation results were consistent with experimental data and provided the structural characteristics of glycolipid bilayers. More recently, Kapla et al. (2012) studied mixtures of DMPC and monogalactosyl at three different concentrations. Interestingly, they compared the NMR dipolar couplings of various segments of lipid molecules obtained in MD and experimental studies. This comparison indicated specific parts of the molecule that were not adequately parameterized, which will greatly facilitate future methodology development. In an early study, Iida-Tanaka et al. performed mixed NMR and MD simulations of the bis-sulfated glycolipid, (HSO3) 2–2,6Mana-2Glca-1-sn-2,3-O-alkylglycerol with a focus on the sugar headgroup conformation (Iida-Tanaka et al., 2000). Later studies examined the anhydrous bilayer forms of dodecyl b-maltoside, dodecyl b-cellobioside, dodecyl b-isomaltoside, and a C12C10 branched b-maltoside; they showed that the glycolipid headgroup structure had substantial effects on bilayer properties (Manickam Achari et al., 2012). 4.3. Cerebrosides Animal glycolipids belong to a heterogeneous class of lipids called cerebrosides. Currently, cerebroside characterizations have included about 300 different oligosaccharide head groups


connected to 60 different hydrophobic components (Degroote et al., 2004; Hakomori, 2007). However, there are reasons to suspect that these numbers are severe underestimations (Lingwood, 2011a). Such glycolipids play a major role in a variety of cellular functions, such as cell–cell recognition (Hakomori, 1981), cells adhesion (Hakomori et al., 1998), signal transduction (Kasahara and Sanai, 2000), and protein trafficking (Patterson et al., 2008). Glycolipids play a vital role in development and differentiation (Yamashita, 1999), viral infections (Harouse, 1991), and several metabolic syndromes (Puri et al., 1999). Although glycolipids are typically found in low concentrations in the outer leaflets of cell membranes (Degroote et al., 2004), in specialized membranes they may be found in concentrations on the order of 30 mol%. Particularly high concentrations of glycolipids of this type have been reported in neurons and myelin (Stoffel and Bosio, 1997). Thus, it is not surprising that glycolipids play an important role in nerve growth, regeneration (Vyas, 2002), and function (Coetzee et al., 1996). Another cell type with elevated glycolipid concentrations is the epithelial cell (Degroote et al., 2004). The most recent review of cerebroside functions was published by Lingwood (2011a). In cell membranes, glycolipids are mainly found in lipid rafts; however, their behavior may be complex (Westerlund and Slotte, 2009; Gupta and Surolia, 2010). Although a large number of animal glycolipids have been identified, only four cerebroside species have been studied in MD simulations. These included galactosylcerebroside (Hall et al., 2010, 2011) and the gangliosides GD1 (Roy and Mukhopadhyay, 2001), GM1 (Patel and Balaji, 2007, 2008; Sega et al., 2006; Shang et al., 2010; Jedlovszky et al., 2009; Vasudevan and Balaji, 2001; DeMarco et al., 2010; Roy and Mukhopadhyay, 2002; Lingwood et al., 2011b; Mondal and Mukhopadhyay, 2008), and GM3 (Sega et al., 2004, 2007; DeMarco et al., 2009). 4.3.1. Galactosylcerebroside The simplest cerebroside is galactosylcerebroside (GalCer), with a headgroup that comprises a single galactose. In 200-ns simulations of GalCer in a mixture of lipids that mimicked a raft

93

environment, 5 mol% concentrations of GalCer did not affect the structural properties of the membrane, including the order and the surface area per lipid (Hall et al., 2010). In contrast, GalCer reduced the lateral diffusion rate of membrane lipids. This effect was attributed to the interdigitation into the opposite leaflet of a GalCer bilayer. Raising the GalCer concentration to 10 mol% reduced its translational diffusion further; however, this also weakened interdigitation (Hall et al., 2011). Most likely the reduced translational diffusion rate resulted from a strong network of hydrogen bonds. GalCer was also studied at higher concentrations of 10 and 25 mol% in charged DPPG bilayers (Zaraiskaya and Jeffrey, 2005). 4.3.2. GM1 GM1 (Gal5-b1,3-GalNAc4-b1,4-(NeuAc3-a2,3)-Gal2-b1,4Glc1-b1,1-Cer) is the most studied ganglioside in MD simulations. This high interest in GM1 is due to its function as a receptor for numerous proteins, particularly toxins, like cholera toxin and vero toxin but also Alzheimer’s b amyloid peptide (Matsuzaki et al., 2010). Two different studies described the behavior of systems where GM1 was embedded in a lipid bilayer composed of POPC and cholesterol. MD simulations displayed the first 40 ns. The results suggested that cholesterol accumulated around the GM1 molecule (Mondal and Mukhopadhyay, 2008). In more extensive, 300-ns MD simulations, the conformation of the GM1 headgroup was compared in two bilayers; one composed of POPC and cholesterol and the other of cholesterol only. Those results indicated that the presence of cholesterol affected the GM1 conformation, which resulted in a tighter interaction between the headgroup and the membrane surface (Fig. 5; Lingwood, 2011b). Consequently, the GM1 became cryptic, and its receptor function for bacterial toxin was restricted, as was shown experimentally. A similar mechanism of GM1 headgroup response for the presence of cholesterol in a membrane was proposed based on 1 ps simulations of lipids in vacuum (Yahi et al., 2010). In this case, parallel experimental studies indicated that cholesterol might optimize the recognition of Alzheimer’s b amyloid peptide by glycolipids.

Fig. 5. Cholesterol (Chol) affects the GM1 conformation in lipid bilayers. (Left) Snapshots of the GM1 molecule from atomistic MD simulations in (upper panels) POPC bilayers and (lower panels) POPC-cholesterol bilayers. Note that three cholesterol molecules are associated with the acyl tails of one GM1 molecule. (Right) Snapshots of the whole bilayer system. Cholesterol is shown in gold, and other lipids are shown in standard colors. POPC is shown as transparent stick and GM1 as van der Waals spheres (reprinted from Lingwood et al., 2011b).(For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

94


In contrast to the studies in multicomponent bilayers, GM1 has also been frequently simulated in binary bilayers. Patel and Balaji performed 20-ns MD simulations of DPPC and GM1 mixtures at concentrations that ranged from 4 to 25 mol% in symmetric and asymmetric bilayers (GM1 was inserted only into one leaflet of the bilayer) (Patel and Balaji, 2008). Those studies were performed at 325 K; thus, in the liquid-disordered phase. They showed that increasing concentrations of GM1 caused decreases in the area per lipid and increases in the ordering of acyl chains. This result contrasted with the observed lack of effect that GalCer showed on membrane order and area per lipid (Hall et al., 2010); this discrepancy was most likely due to the different phase states of the studied bilayers (liquid-ordered vs. liquid-disordered). Those studies also showed that the hydration layer of GM1 molecules consisted of about 30 water molecules, and 23 were hydrogenbonded. In 11 independent 10-ns simulations of a single GM1 molecule in a DPPC bilayer, Patel and Balaji studied the conformation of the GM1 headgroup (Patel and Balaji, 2007). In a similar study of one GM1 molecule inserted into a DOPC bilayer, Jedlovszky et al. (2009) identified two distinct GM1 headgroup conformations. The GM1 headgroup conformation was also described in a micelle environment (Vasudevan and Balaji, 2001). However, these three studies did not address the question of whether the time scale covered and the numbers of molecules in the system were sufficient to provide an equilibrium structure of a complex carbohydrate group. Sega et al. (2006) observed the formation of charge-pairs between the carboxyl group of GM1 and the choline group of DOPC. Shang et al. (2010) considered the effect of a mica surface on GM1 clustering in MD studies. By combining NMR and MD simulations, DeMarco et al. (2010) developed methods for characterizing the structure of membrane-bound, flexible molecules. When they compared GM1/sphingomyelin/ cholesterol and GM1/POPC bilayers, they found that GM1 molecules were more condensed in sphingomyelin/cholesterol bilayers due to GM1 clustering, which was not observed in POPC bilayers (Mori et al., 2012). 4.3.3. GM3 GM3 (NeuAc3-a2,3-Gal2-b1,4-Glc1-b1,1-Cer) ganglioside has a simpler headgroup than GM1, composed of three sugar residues, including a single sialic acid. GM3 has generated high interest due to its role in regulating membrane protein functions; for example, the epidermal growth factor receptor (Coskun et al., 2011). DeMarco and Woods studied the effect of the GM3 attachment to the lipid molecule and the presence of a membrane surface on the carbohydrate structure. The equilibrated GM3 structures that they obtained in a DMPC bilayer were then used to dock two proteins, sialoadhesin and Wheat germ agglutinin, to the membrane surface (DeMarco and Woods, 2009). They also studied bilayers composed solely of GM3 molecules and found that these bilayers exhibited slow relaxation properties. The primary explanation for that finding was related to the slow dynamics in the torsion angles of the headgroup (Sega et al., 2004). In continuing studies, this model of the GM3 lipid bilayer provided a good reproduction of the experimental wide-angle X-ray scattering spectrum of the bilayer (Sega et al., 2007). Short picosecond simulations (in vacuum) of the GM3 dimer with a-Synuclein showed possible interactions between the peptide and lipids. One found the importance of tyrosine 39 and the role of the specific basic-aromatic-basic motive in these interactions (Fantini and Yahi, 2011).

4.4. Sugar adsorption at the water-membrane interface Interactions between free sugars and PCs were investigated in two recent computational studies. One study used MD simulations to investigate the interactions between a DPPC bilayer and two disaccharides, gentiobiose and trechalose. They applied elevated temperatures to compare the stabilizing properties of these disaccharides on the lipid bilayer (Horta et al., 2010). That study showed that the more flexible gentiobiose was weaker than the more rigid trechalose for stabilizing the bilayer. The second study used QM calculations to investigate the interaction between mannose and a PC headgroup. They showed that hydrogen bonding and an unconventional CH . . . O bond formed by the choline group were highly important for the interaction (Parthasarathi et al., 2011). 5. Effects of cholesterol on proteins and peptides Proteins are important components of membranes, and their interactions with lipids are highly important for their function (Soubias and Gawrisch, 2012; Smith, 2012). The importance of cholesterol in these interactions has been recognized (Dopico et al., 2012; Song et al., 2014). Cholesterol can have indirect effects, by changing the membrane properties, or direct effects, by interacting in ways that affect protein structure. The most important membrane parameters affected by cholesterol, which are of key importance for membrane proteins, are the membrane thickness, which affects hydrophobic matching, the lateral pressure profile associated with membrane elasticity, and membrane fluidity/ viscosity, which is related to the lifetime of functional complexes formed by proteins and lipids. Another cholesterol effect, which might be considered indirect, is the modification of the water– membrane interface properties, which can change the interactions between peripheral proteins and the membrane. Direct interactions between cholesterol and proteins might also affect protein function. In many cases, it may be difficult to discriminate between direct and indirect cholesterol effects that contribute to changes in protein conformation and hence function. 5.1. Indirect cholesterol effects: hydrophobic mismatch and lateral pressure profiles The importance of membrane thickness and hydrophobic matching between lipids and transmembrane proteins is well recognized (Nyholm et al., 2007; Holt and Killian, 2010; Li et al., 2012). Among the possible effects of hydrophobic mismatch are membrane deformations, changes in protein tilt, protein aggregation, and changes in protein conformation. In addition, recent CG studies showed that hydrophobic mismatch could affect the lateral diffusion of transmembrane peptides (Ramadurai et al., 2010). Cholesterol is known to increase the membrane thickness, depending on its hydrocarbon chain length. Accordingly, a change in membrane thickness can affect hydrophobic matching between proteins and lipids. Our previous studies showed that cholesterolinduced increases in membrane thickness affected peptide tilting and conformation (Róg et al., 2008b). More recent studies showed similar results; moreover, they found that the membrane in the vicinity of a peptide could adjust to details in the peptide structure (such as length), but only in cholesterol-free bilayers (Kaiser et al., 2011). This lack of membrane adaptation might be a driving force for the sorting of peptides and lipids according to their hydrophobic chain lengths. Hydrophobic mismatch indicated by cholesterol was shown experimentally to affect dimerization of transmembrane helices (Sparr et al., 2005).


CG simulations that could run for relatively long times have provided interesting observations concerning protein segregation and aggregation. Schmidt and Weiss observed that, in heterogeneous bilayers, proteins segregated towards environments with minimal mismatching, but in homogenous bilayers, protein clustering occurred (Schmidt and Weiss, 2010). In another CG study, where cholesterol was explicitly present, de Meyer et al. (2010a) considered cholesterol effects on protein aggregation under hydrophobic mismatching conditions. They proposed a lipid shell mechanism which affected protein–protein interactions. That study showed that cholesterol could decrease repulsive interactions between proteins, but only under positive mismatch conditions. The notion that lateral pressure can affect protein structure is not new (Cantor, 1997, 1999a,b); however, in MD simulations, it was only recently demonstrated that ion channels could be induced to open with changes in lateral pressure (Ollila et al., 2011). MD simulations also showed that changes in ion channels induced by the anesthetic drug, ketamine, were correlated with changes in the lateral pressure profile (Jerabek et al., 2010). On the other hand, it is well documented that cholesterol and other sterols have significant and specific effects on lateral pressure profiles (Ollila et al., 2007). 5.2. Cholesterol effect on peptide adsorption and insertion into membranes Cholesterol modifications of the water–membrane interface can also affect peripheral proteins or peptide associations with membranes. For instance, a recent review summarizes the importance of lipid composition and also cholesterol in protein toxin entry into cells (Sandvig et al., 2014). Extensive free energy

95

studies have shown that the lipid bilayer composition affects amino acid insertions into the membrane (Johansson and Lindahl, 2009). Antimicrobial peptides are a well-known example of peptides, whose membrane interactions depend on lipid compositions (Teixeira et al., 2012); these antimicrobials do not destroy cholesterol-rich bilayers. In the MD literature, only one previous study investigated antimicrobial peptides in the context of cholesterol (Murzyn et al., 2004). However, cholesterol effects on interactions between other peptides and the membrane surface have been studied recently. Two peptides (T-20 and T1249) that inhibit HIV envelope fusion with the cell membrane were investigated in the context of the cholesterol effects on peptidemembrane interactions (do Canto et al., 2011, 2012). Both peptides localized to the membrane-water interface and interacted with lipid headgroups. These interactions were substantially weaker in membranes that contained cholesterol, and both peptides were more deeply embedded into cholesterol-free lipid bilayers. Apparently, T1249 showed stronger interactions with bilayers than T-20, and this correlated with its stronger inhibitor property. Phospholamban is a 52-amino acid peptide with a single transmembrane helix and a short extracellular segment, which also forms a helix. MD simulation studies of phospholamban in a bilayer with and without cholesterol showed that cholesterol stabilized the bent peptide conformation. Thus, the extracellular helix localized to the membrane-water interface in cholesterolcontaining bilayers, and it projected out towards the water phase in cholesterol-free bilayers (Manna and Mukhopadhyay, 2011). Similarly, studies of the b-amyloid (Ab) peptide insertion into lipid bilayers showed that these peptides preferred an inserted state in cholesterol-enriched membranes and a surface location in cholesterol-depleted membranes (Qiu et al., 2011). Interestingly, in the surface location, Ab showed a tendency to form b-sheets.

Table 1 Proteins co-crystallize with cholesterol or cholesteryl hemisuccinate. Ligand

Protein

PDB ID

Reference

Cholesterol

b-2-adrenergic receptor

2RH1 3D4S 3PDS 3NY9 3NYA 3NY8 3AM6 4DKL 4EIY 4IB4 4NC3 4NTJ 4PXZ 4OR2 4M48 3GKI 2ZXE 3A3Y 3KDP 4HYT 4HQJ 3WGV 3WGU 3ZPQ 3ZPR 2Y01 2Y00 2Y02 2Y03 2Y04 4AYW

Cherezov et al., 2007 Hanson et al., 2008 Rosenbaum et al., 2011 Wacker et al., 2010 Wacker et al., 2010 Wacker et al., 2010 Wada et al., 2011 Manglik et al., 2012 Liu et al., 2012 Wacker et al., 2013 Liu et al., 2013 Zhang et al., 2014a Zhang et al., 2014b Wu et al., 2014 Penmatsa et al., 2013 Kwon et al., 2009 Shinoda et al., 2009 Ogawa et al., 2009 Morth et al., 2007 Laursen et al., 2013 Nyblom et al., 2013 Kanai et al., 2013 Kanai et al., 2013 Christopher et al., 2013 Christopher et al., 2013 Warne et al., 2011 Warne et al., 2011 Warne et al., 2011 Warne et al., 2011 Warne et al., 2011 Shintre et al., 2013

Rhodopsin m-Opioid receptor Adenosine receptor Serotonin receptor P2Y12 receptor Metabotropic glutamate receptor Dopamine transporter N-terminal domain of NPC1 Sodium–potassium pump

Cholesteryl hemisuccinate

b-2-adrenergic receptor

ABCB10, a human ATP-binding cassette transporter

96


These observations might be important in understanding the molecular mechanisms of Alzheimer disease. Here, it is worth to stress that the role of cholesterol and sphingolipids in Alzheimer disease is recognized (Fantini and Yahi, 2010). Dynorphin, a natural ligand for opioid receptors, was found to localize close to the water–membrane interface in cholesterol-rich bilayers (Khelashvili et al., 2010b). This location is close to the postulated ligand entry pathway. Caveolin-1 was shown to preferentially interact with lipid bilayers that contained cholesterol; this was highlighted with partitioning free energy calculations (Sengupta, 2012). 5.3. Specific interactions between proteins and cholesterol Lipids are often found in crystallographic structures of membrane proteins, indicating strong and specific lipid–protein interactions. For instance, charged lipid species such as cardiolipin have been found in crystal structures of mitochondrial proteins (Paradies et al., 2014; Hunte, 2005). Recent atomistic simulations (Pöyry et al., 2013) have given consistent results, showing how cardiolipin diffuses to the protein complex and binds to a position that crystallographic studies suggested to be its binding site. Related support has been given by CG studies, too (Arnarez et al., 2013a,b). Cholesterol has been co-crystallized with several proteins, and currently the PDB database includes 30 entries of protein structures where cholesterol is present, and 23 out of these 30 structures represent membrane proteins (Table 1). A large fraction of these proteins belongs to G-protein coupled receptors such as b-2-adrenergic receptor, rhodopsin, m-opioid receptor, adenosine receptor, serotonin receptor, P2Y12 receptor, and metabotropic glutamate receptor. From other types of proteins, cholesterol has been co-crystallized with sodium-potassium pump, N-terminal domain of NPC1, and dopamine transporter. Additionally eight structures of membrane proteins have been cocrystallized with cholesteryl hemisuccinate (CHS), and four watersoluble proteins with cholesterol sulfate. Table 1 summarizes the PDB codes of all membrane proteins co-crystallized with cholesterol and cholesteryl succinate. It is worth noting that the majority of the structures listed in Table 1 have been published during the last few years. Only a few recent simulation studies have investigated the specific interactions between cholesterol and proteins. This is due to the microsecond time scale needed to properly sample the lipid environment around a protein, and such a large time scale is rarely achieved in current MD simulations (reviewed by Sadiq et al., 2013). The role of cholesterol in the structure of the human A2A adenosine receptor was studied in simulations on the microsecond time scale (Lyman et al., 2009; Lee et al., 2013). Those studies showed that cholesterol stabilized helix II through direct binding. In extensive MD simulations, Khelashvili et al. (2009) investigated the spatial distribution of cholesterol around rhodopsin. They found three regions of high cholesterol density that affected the protein structure. Similarly, Cang et al. found that cholesterol preferentially bound at 7 places on the b-2-adrenergic receptor (Cang et al., 2013). Three of those places were previously identified in experimental studies. Cheng et al. (2009) investigated cholesterol and anionic lipid interactions with nicotinic acetylcholine receptors. Other experimental studies showed that cholesterol could stabilize another G-protein coupled receptor, the serotonin1A receptor (Saxena and Chattopadhyay, 2012). That effect was sensitive to the particular sterol structure (Jafurulla et al., 2013). Subsequent MD simulations with the MARTINI CG model showed that cholesterol could bind specifically to a particular site on the receptor that exhibited a specific amino acid motif (Sengupta and Chattopadhyay, 2012).

Aquaporin simulations in cholesterol-free and cholesterol-rich bilayers showed that cholesterol stabilization of the protein–lipid interactions depended on the hydrogen bonding lifetime (O’Connor and Klauda, 2011). Cholesterol was found to form specific stacking interactions with protein side chains. CG simulations of the connexin 26 hemichannel in various lipid environments showed that the frequency of cholesterol flops increased in the vicinity of the protein (Hung and Yarovsky, 2011). The effect of cholesterol on transmembrane helix dimerization through GxxxG motifs was studied in CG simulations by Prakash et al. (2011). They showed that different dimerization patterns occurred at low and high cholesterol concentrations. It is often asked whether the effects of cholesterol on membrane protein stability and function are specifically due to cholesterol, or would other sterols have largely the same effects. While there is no general answer to this question, one can highlight the importance and uniqueness of cholesterol with an example. Desmosterol, an immediate precursor of cholesterol in the Bloch pathway of sterol synthesis, was used to replace cholesterol in raft membranes. The consequence of this replacement was that the signaling of the insulin receptor was impaired (Vainio et al., 2006). This impairment in protein function took place despite the fact that the only structural difference between cholesterol and desmosterol is an additional double bond in the short hydrocarbon chain of desmosterol. Yet this seemingly negligible difference weakens the ordering capability of desmosterol compared to that of cholesterol, and hence membranes rich in desmosterol and more fluid than those rich in cholesterol (Vainio et al., 2006). It remains an open question, though, whether the impairment of the insulin receptor function is due to changes in membrane properties, or due to changes in specific sterol–protein binding. 6. Coarse-grained models of cholesterol and their applications In recent literature, CG models have become particularly interesting, because they have enabled analyses of time scales that are inaccessible in atomic simulations (Stansfeld and Sansom, 2011; Marrink and Tieleman, 2013). This feature makes it possible to observe rare events and long-time scale phenomena like phase separation (Risselada and Marrink, 2008), phase transitions (Marrink et al., 2005), protein diffusion (Niemela et al., 2010), desorption of cholesterol from cholesterol monolayers by cyclodextrin (Lopez et al., 2011), curvature effects (Risselada et al., 2011), effects of electric potential (Loubet et al., 2013), and flip-flop motions (Ogushi et al., 2012). Here we discuss some of the recent key articles dealing with CG models and related to the long-time behavior, thereby complementing the other CG model results discussed earlier in this review. Probably the most popular basis for CG models is the MARTINI force field, which includes lipids, proteins, and carbohydrates (Marrink et al., 2007; Monticelli et al., 2008; Lopez et al., 2009). Cholesterol is one of the many molecules parameterized in the MARTINI description. However, alternative CG models exist especially for lipids (Orsi and Essex, 2011; Lu and Voth, 2009; Hadley and McCabe, 2012; de Meyer and Smit, 2009). The de Meyer and Smit model uses a simple CG description, where a cholesterol molecule is described as a five-bead, flat planar ring with a twobead tail; this description is sufficient to partly reproduce the phase behavior of DMPC-cholesterol bilayers (de Meyer and Smit, 2009; de Meyer et al., 2010b). In all of these models, one has to keep in mind that the coarse-grained “cholesterol” is not cholesterol in a true sense, but rather a sterol molecule having some of the generic properties of sterols. It is a plain fact that there is a price to pay in dealing with coarse-grained models, and with cholesterol it means that CG models of cholesterol are likely not the most appropriate approach to examine detailed processes such


as specific binding of cholesterol with its receptors. Instead, for large-scale phenomena where details are not important, such as diffusion or phase separation and phase behavior, the CG models are often the method of choice. An improved version of the MARTINI-compatible model for cholesterol, better capturing the cholesterol ring asymmetry due to C18 and C19 methyl groups, and the cholesterol tail complexity was recently shown to perform better in a wide range of cholesterol concentrations than the original MARTINI model (Daily et al., 2014). A slightly different model for cholesterol was also developed by Hadley and McCabe, (2010). In that model, the cholesterol ring system was represented with 6 beads, and two of these beads project out of the ring plane, mimicking the asymmetry of the cholesterol ring caused by the two out-of-plane methyl groups. The model was successfully verified in simulations of a cholesterol crystal, which is rather difficult to simulate with the CG methodology. The Hadley and McCabe model was also used in studies on the self-assembly of stratum corneum lipids (Hadley and McCabe, 2012). CG models have also been used in studies on lung surfactant monolayers, where cholesterol is an important component, and in studies on tear fluid, which forms a lipid layer and where cholesterol oleates are present (Duncan et al., 2011; Kulovesi et al., 2012; Rantamaki et al., 2011; Wizert et al., 2014). CG models have also been applied to simulations of lipids droplets (Chaban and Khandelia, 2014a,b), where the studies suggested that a large fraction of cholesterol is actually located in the lipid droplet core, and not necessarily in the surface layer region that covers the core of a droplet. Comparative studies that examined phase separation in systems composed of DPPC, cholesterol, and di-18:2-PC showed that phase separation occurred in both CG and atomic models, but lipid demixing was about 40 times slower in the atomic simulations (Hakobyan and Heuer, 2013). Flip-flop motions of cholesterol were studied in asymmetric bilayers, where one leaflet was composed of PC and the other of phosphatidylserine. Those studies showed that the cholesterol distribution was not symmetric; cholesterol showed a preference for phosphatidylserine and saturated lipids (Yesylevskyy and Demchenko, 2012). Addition of cholesterol to a curved asymmetric DOPC/DOPS lipid bilayer was shown to promote the formation of highly curved regions with either saddle-like or sphere-like topology (Yesylevskyy and Ramseyer, 2014). This makes membranes topologically more heterogeneous. Domains of DOPS were in this case more sensitive to the presence of cholesterol. 7. Force field issues Force field development is an important topic for MD simulation studies (Monticelli and Tieleman et al., 2012; Antila and Salonen, 2012). As to lipids, the progress in the field has been particularly dynamic during the recent years. This is highlighted by new lipid models that are compatible with, e.g., CHARMM (Klauda et al., 2010, 2012; Lim et al., 2012; Kang and Klauda, 2014), AMBER (Dickson et al., 2014; Jambeck and Lyubartsev, 2012a,b, 2013), TraPPE (Bhatnagar et al., 2013), OPLS-AA (Maciejewski et al., 2014), and polarizable (Davis and Patel 2009; Bauer et al., 2011; Chowdhary et al., 2013) force fields. A more detailed review of lipids’ force fields can be found in Manna et al. (2014). Here we briefly discuss force field issues related to raft components: sphingolipids, cholesterol, and glycolipids. The only specific parameterization of the sphingosine moiety has been performed in the framework of the CHARMM force field (Venable et al., 2014). In this parameter set, seven dihedral angles were specifically derived for molecular fragments best

97

representing the sphingomyelin backbone. The used fragments were relatively large and the parameterization was based on computationally demanding high-level quantum-mechanical calculations (for details, see (Klauda et al., 2004)). The obtained force field was validated against experimental NMR data, showing good agreement between simulations and experiments, with only few exceptions. Most of the atomistic MD simulations of lipid bilayers with cholesterol have used default force field parameters for given atom types present in a cholesterol molecule. This approach has given satisfactory results with the Slipids force field (Jambeck and Lyubartsev, 2013), as the order parameter data of acyl chains matched experimental findings. However, in the case of the socalled Berger lipids, the cholesterol model based on the GROMOS force field was shown to reproduce the order parameters of hydrocarbon chains and cholesterol segments at small cholesterol concentrations, but to overestimate the ordering effect of cholesterol for larger cholesterol concentrations (above 30 mol%) (Mendes Ferreira et al. 2013). Tieleman and co-workers (Tieleman et al., 2006) showed the interaction between Berger lipids’ hydrocarbon tails and peptides (parameterized with the GROMOS force field) to be too strong, leading to membrane condensation. These two studies indicate lack of transferrability between the GROMOS force field and Berger lipids, or their parameters; however, the extent of this problem can be reduced using straight combination rules (Tieleman et al., 2006). A more careful parameterization of cholesterol was performed for the CHARMM force field C36c (Lim et al., 2012). In this model, the torsion angle at the branch of the cholesterol tail (around the bond C20—C22) was reparameterized. To validate the model, a mixture of heptane representing hydrocarbons and decalin representing the first two rings of cholesterol was first simulated over a wide range of concentrations, and the obtained molar volumes were compared with experimental data. Deviation of MD data from experimental findings was small, indicating sufficient accuracy of the non-bonded parameters in the model. At a second stage, simulations of lipid bilayers with cholesterol were performed, showing quite good agreement of MD simulation results with experiments, though the observed deviations also suggested more parameterization work to be done. Relatively few MD simulation studies are available on lipid bilayers composed of glycolipids. One explanation is the lack of a force field parameter necessary to parameterize glycolipids. Although all major force fields have an extension for simulating carbohydrates, the number of parameterized residues might be limited to only basic hexose units (Damm et al., 1997). The largest set of parameters for simulating carbohydrates is known as GLYCAM (Kirschner et al., 2008), a force field developed to be compatible with the AMBER force field. However, there is no adequate model of lipids that is compatible with this force field. Some attempts were made to develop a model of lipids compatible with the GLYCAM force field (Tessier et al., 2008), but the model does not correctly reproduce the membrane phase state with tensionless simulations. Some of the problems related to carbohydrates and membrane parameterization were discussed in more detail in previous reviews (Woods and Tessier, 2010; Fadda and Woods, 2010; Pastor and MacKerell, 2011). As to CG models, one can consider the MARTINI model as an example. In the framework of MARTINI, basic sugar units and several glycolipids were recently parameterized (Lopez et al., 2013). However, one has to be aware of the limitations of CG models in studies of highly chiral molecules such as sugars. For example, there are 8 stereoisomers of the most common sugars, D-aldohexoses (e.g. glucose and galactose), which in the MARTINI framework would be described by the same residue. This may be problematic in considerations of molecular properties where details do matter,

98


since in practice the differences between these residues are substantial. For instance, lipid bilayers comprised of glucose- and galactose-based lipids having the same structure except for the glyco-headgroup were shown to have quite different membrane properties, such as membrane fluidity (Róg et al., 2007b). Differences in chirality are particularly important when specific interactions between glycolipids and proteins are considered. 8. Summary In this review, we have discussed recent results based on atomistic and coarse-grained MD simulations of lipid bilayers related to lipid rafts. Particular attention has been paid to the roles of cholesterol, sphingomyelin, and glycolipids in dictating membrane properties, and to the consequences of their interactions with membrane proteins and receptors. Based on the examples discussed in this paper, it is easy to conclude that the insight emerging from MD simulations is substantial and provides a great deal of added value for experimental research, and it is likely that the future work in the field will unite the two approaches more and more often. Given that the accuracy of simulation models improves all the time, the efficiency of programs used in simulations improves similarly, too, and the computing resources available for simulations also get better, there is no doubt that the role of atomistic and molecular simulations in considerations of membrane-associated processes and phenomena will increase. Perhaps some day the reasons for the quite unique properties of cholesterol will be understood in detail, too. Conflict of interest The authors declare that there are no conflicts of interest. Transparency document The Transparency document associated with this article can be found in the online version. Acknowledgments This work was supported, in part, by the Academy of Finland (project funding (T.R., I.V.); Center of Excellence grant (T.R., I.V.)), the Sigrid Juselius Foundation (I.V.), and the European Research Council Advanced Grant CROWDED-PRO-LIPIDS (T.R., I.V.). We also thank CSC–IT Center for Science (CSC, Espoo, Finland) and the HorseShoe (DCSC) supercluster computing facility at the University of Southern Denmark for computer resources. References Aittoniemi, J., Niemela, P.S., Hyvonen, M.T., Karttunen, M., Vattulainen, I., 2007. Insight into the putative specific interactions between cholesterol, sphingomyelin, and palmitoyl-oleoyl phosphatidylcholine. Biophys. J. 92, 1125–1137. Aittoniemi, J., Róg, T., Niemelä, P.S., Pasenkiewicz-Gierula, M., Karttunen, M., Vattulainen, I., 2006. Tilt: major factor in sterols’ ordering capability in membranes. J. Phys. Chem. B 110, 25562–25564. Almeida, P.F.F., 2009. Thermodynamics of lipid interactions in complex bilayers. Biochim. Biophys. Acta 1788, 72–85. Alwarawrah, M., Dai, J., Huang, J., 2010. A molecular view of the cholesterol condensing effect in DOPC lipid bilayers. J. Phys. Chem. B 114, 7516–7523. Alwarawrah, M., Dai, J., Huang, J., 2012. Modification of lipid bilayer structure by diacylglycerol: a comparative study of diacylglycerol and cholesterol. J. Chem. Theory Comput. 8, 749–758. Antila, H.S., Salonen, E., 2012. Polarizable force fields. In: Monticelli, L., Salonen, E. (Eds.), Biomolecular Simulations Methods and Protocols Series: Methods in Molecular Biology, vol. 924. Springer, New York, pp. 215–242. Apajalahti, T., Niemela, P.S., Govindan, P.N., Miettinen, M.S., Salonen, E., Marrink, S.J., Vattulainen, I., 2010. Concerted diffusion of lipids in raft-like membranes. Faraday Discuss. 144, 411–430.

Arnarez, C., Marrink, S.-J., Periole, X., 2013a. Identification of cardiolipin binding sites oncytochrome c oxidase at the entrance of proton channels. Sci. Rep. 3, 1263. Arnarez, C., Mazat, J.-P., Elezgaray, J., Marrink, S.-J., Periole, X., 2013b. Evidence for cardiolipin binding sites on the membrane-exposed surface of the cytochrome bc1. J. Am. Chem. Soc. 135, 3112–3120. Bauer, B.A., Lucas, T.R., Meninger, D.J., Patel, S., 2011. Water permeation through DMPC lipid bilayers using polarizable charge equilibration force fields. Chem. Phys. Lett. 508, 289–294. Beattie, M.E., Veatch, S.L., Stottrup, B.L., Keller, S.L., 2005. Sterol structure determines miscibility versus melting transitions in lipid vesicles. Biophys. J. 89, 1760–1768. Bennett, W.F.D., Tieleman, D.P., 2012. Molecular simulation of rapid translocation of cholesterol, diacylglycerol, and ceramide in model raft and nonraft membranes. J. Lipid Res. 53, 421–429. Bennett, W.F.D., MacCallum, J.L., Tieleman, D.P., 2009a. Thermodynamic Analysis of the effect of cholesterol on dipalmitoylphosphatidylcholine lipid membranes. J. Am. Chem. Soc. 131, 1972–1978. Bennett, W.F.D., MacCallum, J.L., Hinner, M.J., Marrink, S.J., Tieleman, D.P., 2009b. Molecular view of cholesterol flip-flop and chemical potential in different membrane environments. J. Am. Chem. Soc. 131, 12714–12720. Bennett, W.F.D., Tieleman, D.P., 2013. Computer simulations of lipid membrane domains. Biochim. Biophys. Acta 1828, 1765–1776. Berkowitz, M.L., 2009. Detailed molecular dynamics simulations of model biological membranes containing cholesterol. Biochim. Biophys. Acta 1788, 86–96. Berkowitz, M.L., Vacha, R., 2012. Aqueous solutions at the interface with phospholipid bilayers. Acc. Chem. Res. 45, 74–82. Bhatnagar, N., Kamath, G., Potoff, J.J., 2013. Biomolecular simulations with the transferable potentials for phase equilibria: extension to phospholipids. J. Phys. Chem. B 117, 9910–9921. Bhide, S.Y., Zhang, Z., Berkowitz, M.L., 2007. Molecular dynamics simulations of SOPS and sphingomyelin bilayers containing cholesterol. Biophys. J. 92, 1284– 1295. Bielska, A.A., Olsen, B.N., Gale, S.E., Mydock-McGrane, L., Krishnan, K., Baker, N.A., Schlesinger, P.H., Covey, D.F., Ory, D.S., 2014. Side-chain oxysterols modulate cholesterol accessibility through membrane remodeling. Biochemistry 53, 3042–3051. Björkbom, A., Róg, T., Kankaanpää, P., Lindroos, D., Kaszuba, K., Kurita, M., Yamaguchi, S., Yamamoto, T., Jaikishan, S., Paavolainen, L., Päivärinne, J., Nyholm, T.K.M., Katsumura, S., Vattulainen, I., Slotte, J.P., 2011. N- and Omethylation of sphingomyelin markedly affects its membrane properties and interactions with cholesterol. Biochim. Biophys. Acta 1808, 1179–1186. Björkbom, A., Róg, T., Kaszuba, K., Kurita, M., Yamaguchi, S., Lonnfors, M., Nyholm, T. K.M., Vattulainen, I., Katsumura, S., Slotte, J.P., 2010. Effect of sphingomyelin headgroup size on molecular properties and interactions with cholesterol. Biophys. J. 99, 3300–3308. Boggara, M.B., Mihailescu, M., Krishnamoorti, R., 2012. Structural association of nonsteroidal anti-inflammatory drugs with lipid membranes. J. Am. Chem. Soc. 134, 19669–19676. Bunker, A., 2012. Poly(ethylene glycol) in drug delivery, why does it work, and can we do better? All atom molecular dynamics simulation provides some answers. Phys. Procedia 34, 24–33. Busch, S., Smuda, C., Pardo, L.C., Unruh, T., 2010. Molecular mechanism of long-range diffusion in phospholipid membranes studied by quasi-elastic neutron scattering. J. Am. Chem. Soc. 132, 3232–3233. Cang, X., Du, Y., Mao, Y., Wang, Y., Yang, H., Jiang, H., 2013. Mapping the functional binding sites of cholesterol in b2-adrenergic receptor by long-time molecular dynamics simulations. J. Phys. Chem. B 117, 1085–1094. Cantor, R.S., 1997. The lateral pressure profile in membranes: a physical mechanism of general anesthesia. Biochemistry 36, 2339–2344. Cantor, R.S., 1999a. The influence of membrane lateral pressures on simple geometric models of protein conformational equilibria. Chem. Phys. Lipids 101, 45–56. Cantor, R.S., 1999b. Lipid composition and the lateral pressure profile in bilayers. Biophys. J. 76, 2625–2639. Casciola, M., Bonhenry, D., Liberti, M., Apollonio, F., Tarek, M., 2014. A molecular dynamic study of cholesterol rich lipid membranes: comparison of electroporation protocols. Bioelectrochemistry 100, 11–17. Chaban, V.V., Khandelia, H., 2014a. Lipid structure in triolein lipid droplets. J. Phys. Chem. B 118, 10335–10340. Chaban, V.V., Khandelia, H., 2014b. Distribution of neutral lipids in the lipid droplet core. J. Phys. Chem. B 118, 11145–11151. Chen, C., Tripp, C.P., 2012. A comparison of the behavior of cholesterol, 7-dehydrocholesterol and ergosterol in phospholipid membranes. Biochim. Biophys. Acta 1818, 1673–1681. Cheng, M.H., Xu, Y., Tang, P., 2009. Anionic lipid and cholesterol interactions with a4b2 nAChR: insights from MD simulations. J. Phys. Chem. B 113, 6964–6970. Cherezov, V., Rosenbaum, D.M., Hanson, M.A., Rasmussen, S.G.F., Thian, F.S., Sun, T., Choi, H.-J., Kuhn, P., Weis, W.I., Kobilka, B.K., Stevens, R.C., 2007. High resolution crystal structure of an engineered human b2-adrenergic G-protein coupled receptor. Science 318, 1258–1265. Chiu, S.W., Vasudevan, S., Jakobsson, E., Mashl, R.J., Scott, H.L., 2003. Structure of sphingomyelin bilayers: a simulation study. Biophys. J. 85, 3624–3635. Choi, Y., Attwood, S.J., Hoopes, M.I., Drolle, E., Karttunen, M., Leonenko, Z., 2014. Melatonin directly interacts with cholesterol and alleviates cholesterol effects in dipalmitoylphosphatidylcholine monolayers. Soft Matter 10, 206–213.

T. Róg, I. Vattulainen / Chemistry and Physics of Lipids 184 (2014) 82–104 Chowdhary, J., Harder, E., Lopes, P.E.M., Huang, L., MacKerell Jr., A.D., Roux, 2013. A polarizable force field of dipalmitoylphosphatidylcholine based on the classical Drude model for molecular dynamics simulations of lipids. J. Phys. Chem. B 117, 9142–9160. Christopher, J., Brown, J., Dore, A., Errey, J., Koglin, M., Marshall, F.H., Myszka, D., Rich, R.L., Tate, C.G., Tehan, B., Warne, T., Congreve, M., 2013. Biophysical fragment screening of the b1-adrenergic receptor: identification of high affinity arylpiperazine leads using structure-based drug design. J. Med. Chem. 56, 3446–3455. Coetzee, T., Fujita, N., Dupree, J., Shi, R., Blight, A., Suzuki, K., Suzuki, K., Popko, B., 1996. Myelination in the absence of galactocerebroside and sulfatide: normal structure with abnormal function and regional instability. Cell 86, 209– 219. Coskun, U., Grzybek, M., Drechsel, D., Simons, K., 2011. Regulation of human EGF receptor by lipids. Proc. Natl. Acad. Sci. USA 108, 9044–9048. Cramariuc, O., Róg, T., Vattulainen, I., 2012. Drug-lipid membrane interaction mechanisms revealed through molecular simulations. Curr. Phys. Chem. 2, 379–400. Curatolo, W., 1987. Glycolipid function. Biochim. Biophys. Acta 906, 137–160. Curdova, J., Capkova, P., Plasek, J., Repakova, J., Vattulainen, I., 2007. Free pyrene probes in gel and fluid membranes: perspective through atomistic simulations. J. Phys. Chem. B 111, 3640–3650. Dai, J., Alwarawrah, M., Huang, J., 2010. Instability of cholesterol clusters in lipid bilayers and the cholesterol’s umbrella effect. J. Phys. Chem. B 114, 840–848. Daily, M.D., Olsen, B.N., Schlesinger, P.H., Ory, D.S., Baker, N.A., 2014. Improved coarse-grained modeling of cholesterol-containing lipid bilayers. J. Chem. Theory Comput. 10, 2137–2150. Damm, W., Frontera, A., Tirado-Rives, J., Jorgensen, W.L., 1997. OPLS all-atom force field for carbohydrates. J. Comput. Chem. 18, 1955–1970. Davis, J.E., Patel, S., 2009. Charge equilibration force fields for lipid environments: applications to fully hydrated DPPC bilayers and DMPC-embedded gramicidin A. J. Phys. Chem. B. 113, 9183–9196. de Joannis, J., Coppock, P.S., Yin, F., Mori, M., Zamorano, A., Kindt, J.T., 2011. Atomistic simulation of cholesterol effects on miscibility of saturated and unsaturated phospholipids: implications for liquid-ordered/liquid-disordered phase coexistence. J. Am. Chem. Soc. 133, 3625–3634. de Meyer, F., Smit, B., 2009. Effect of cholesterol on the structure of a phospholipid bilayer. Proc. Nat. Acad. Sci. USA 106, 3654–3658. de Meyer, F.J.M., Benjamini, A., Rodgers, J.M., Misteli, Y., Smit, B., 2010a. Molecular simulation of the DMPC-cholesterol phase diagram. J. Phys. Chem. B 114, 10451–10461. de Meyer, F.J.-M., Rodgers, J.M., Willems, T.F., Smit, B., 2010b. Molecular simulation of the effect of cholesterol on lipid-mediated protein–protein interactions. Biophys. J 99, 3629–3638. Degroote, S., Wolthoorn, J., van der Meer, G., 2004. The cell biology of glycosphingolipids. Cell Develop. Biol. 15, 375–387. DeMarco, M.L., 2012. Three-dimensional structure of glycolipids in biological membranes. Biochemistry 51, 5725–5732. DeMarco, M.L., Woods, R.J., 2009. Atomic-resolution conformational analysis of the GM3 ganglioside in a lipid bilayer and its implications for ganglioside-protein recognition at membrane surfaces. Glycobiology 19, 344–355. DeMarco, M.L., Woods, R.J., Prestegard, J.H., Tian, F., 2010. Presentation of membrane-anchored glycosphingolipids determined from molecular dynamics simulations and NMR paramagnetic relaxation rate enhancement. J. Am. Chem. Soc. 132, 1334–1338. Dickson, C.J., Madej, B.D., Skjevik Å, A., Betz, R.M., Teigen, K., Gould, I.R., Walker, R.C., 2014. Lipid14: the Amber lipid force field. J. Chem. Theory Comput. 10, 865–879. do Canto, A.M.T.M., Carvalho, A.J.P., Ramalho, J.P.P., Loura, L.M.S., 2011. Molecular dynamics simulations of T-20HIV fusion inhibitor interacting with model membranes. Biophys. Chem. 159, 275–286. do Canto, A.M.T.M., Carvalho, A.J.P., Ramalho, J.P.P., Loura, L.M.S., 2012. Molecular dynamics simulation of HIV fusion inhibitor T-1249: insights on peptide–lipid interaction. Comput. Math. Methods Med Article ID 1151854. Dopico, A.M., Bukiya, A.N., Singh, A.K., 2012. Large conductance, calcium- and voltage-gated potassium (BK) channels: regulation by cholesterol. Pharmaco. Therap. 135, 133–150. Drolle, E., Ku9 cerka, N., Hoopes, M.I., Choi, Y., Katsaras, J., Karttunen, M., Leonenko, Z., 2013. Effect of melatonin and cholesterol on the structure of DOPC and DPPC membranes. Biochim. Biophys. Acta 1828, 2247–2254. Duncan, S.L., Dalal, I.S., Larson, R.G., 2011. Molecular dynamics simulation of phase transitions in model lung surfactant monolayers. Biochim. Biophys. Acta 1808, 2450–2465. Dörmann, P., Benning, C., 2002. Galactolipids rule in seed plants. Trends Plant Sci. 7, 112–118. Edholm, O., Nagle, J.F., 2005. Areas of molecules in membranes consisting of mixtures. Biophys. J. 89, 1827–1832. Ekholm, O., Jaikishan, S., Lonnfors, M., Nyholm, T.K.M., Slotte, J.P., 2011. Membrane bilayer properties of sphingomyelins with amide-linked 2- or 3-hydroxylated fatty acids. Biochim. Biophys. Acta 1808, 727–732. Eriksson, E.S.E., Eriksson, L.A., 2011. The influence of cholesterol on the properties and permeability of hypericin derivatives in lipid membranes. J. Chem. Theory Comput. 7, 560–574. Fadda, E., Woods, R.J., 2010. Carbohydrates and protein–carbohydrate interactions: motivation, issues and prospects. Drug Discov. Today 15, 596–609.

99

Falck, E., Patra, M., Karttunen, M., Hyvonen, M.T., Vattulainen, I., 2004. Lessons of slicing membranes Interplay of packing, free area, and lateral diffusion in phospholipid/cholesterol bilayers. Biophys. J. 87, 1076–1091. Falck, E., Róg, T., Karttunen, M., Vattulainen, I., 2008. Lateral diffusion in lipid membranes through collective flows. J. Am. Chem. Soc. 130, 44–45. Fantini, J., Yahi, N., 2010. Molecular insights into amyloid regulation by membrane cholesterol and sphingolipids: common mechanisms in neurodegenerative diseases. Expert Rev. Mol. Med. 12, e27. Fantini, J., Yahi, N., 2011. Molecular basis for the glycosphingolipid-binding specificity of a-synuclein: key role of tyrosine 39 in membrane insertion. J. Mol. Biol. 408, 654–669. Fernandez, M.L., Marshall, G., Sagues, F., Reigada, R., 2010. Structural and kinetic molecular dynamics study of electroporation in cholesterol-containing bilayers. J. Phys. Chem. B 114, 6855–6865. Forst, G., Cwiklik, L., Jurkiewicz, P., Schubert, R., Hof, M., 2014. Interactions of betablockers with model lipid membranes: molecular view of the interaction of acebutolol, oxprenolol, and propranolol with phosphatidylcholine vesicles by time-dependent fluorescence shift and molecular dynamics simulations. Eur. J. Pharma. Biopharma. 87, 559–569. Franova, M., Repakova, J., Capkova, P., Holopainen, J.M., Vattulainen, I., 2010. Effects of DPH on DPPC-cholesterol membrane with varying concentration of cholesterol: from local perturbations to limitations in fluorescence anisotropy experiments. J. Phys. Chem. B 114, 2704–2711. Gabizon, A., Catane, R., Uziely, B., Kaufman, B., Safra, T., Cohen, R., Martin, F., Huang, A., Barenholz, Y., 1994. Prolonged circulation time and enhanced accumulation in malignant exudates of doxorubicin encapsulated in polyethylene-glycol coated liposomes. Cancer Res. 54, 987–992. Gallova, J., Uhrikova, D., Kucerka, N., Teixeirac, J., Balgavy, P., 2010. Partial area of cholesterol in monounsaturated diacylphosphatidylcholine bilayers. Chem. Phys. Lipids 163, 765–770. Gapsys, V., de Groot, B.L., Briones, R., 2013. Computational analysis of local membrane properties. J. Comput. Aided Mol. Des. 27, 845–858. Gupta, G., Surolia, A., 2010. Glycosphingolipids in microdomain formation and their spatial organization. FEBS Lett. 584, 1634–1641. Gurtovenko, A.A., Anwar, J., Vattulainen, I., 2010. Defect-mediated trafficking across cell membranes: insights from in silico modeling. Chem. Rev. 110, 6077–6103. Haas, D., Kelley, R.J., Hoffmann, G.F., 2001. Inherited disorders of cholesterol biosynthesis. Neuropediatrics 32, 113–122. Hadley, K.R., McCabe, C., 2010. A structurally relevant coarse-grained model for cholesterol. Biophys. J 99, 2896–2905. Hadley, K.R., McCabe, C., 2012. A simulation study of the self-assembly of coarsegrained skin lipids. Soft Matter 8, 4802–4814. Hakobyan, D., Heuer, A., 2013. Phase separation in a lipid/cholesterol system: comparison of coarse-grained and united-atom simulations. J. Phys. Chem. B 117, 3841–3851. Hakobyan, D., Heuer, A., 2014. Key molecular requirements for raft formation in lipid/cholesterol membranes. PLoS One 9, e87369. Hakomori, S., 1981. Glycosphingolipids in cellular interaction differentiation, and oncogenesis. Annu. Rev. Biochem. 50, 733–764. Hakomori, S., Handa, K., Iwabuchi, K., Yamamura, S., Prinetti, A., 1998. New insights in glycosphingolipid function: glycosignaling domain, a cell surface assembly of glycosphingolipids with signal transducer molecules, involved in cell adhesion coupled with signaling. Glycobiology 8, XI–XIX. Hakomori, S.I., 2007. Structure and function of glycosphingolipids and sphingolipids: recollections and future trends. Biochim. Biophys. Acta 1780, 325–346. Haldar, S., Kanaparthi, R.K., Samanta, A., Chattopadhyay, A., 2012. Differential effect of cholesterol and its biosynthetic precursors on membrane dipole potential. Biophys J. 102, 1561–1569. Hall, A., Róg, T., Karttunen, M., Vattulainen, I., 2010. Role of glycolipids in lipid rafts: a view through atomistic molecular dynamics simulations with galactosylceramide. J. Chem. Phys. B 114, 7797–7807. Hall, A., Róg, T., Vattulainen, I., 2011. Effect of galactosylceramide on dynamics of cholesterol-rich lipid membranes. J. Phys. Chem. B 115, 14424–14434. Hanson, M.A., Cherezov, V., Roth, C.B., Griffith, M.T., Jaakola, V.-P., Chien, E.Y.T., Velasquez, J., Kuhn, P., Stevens, R.C., 2008. A specific cholesterol binding site is established by the 2.8 Å structure of the human b2-adrenergic receptor in an alternate crystal form. Structure 16, 897–905. Harouse, J.M., 1991. Inhibition of entry of HIV-1 in neural cell-lines by antibodies against galactosyl ceramide. Science 253, 320–323. Holt, A., Killian, J.A., 2010. Orientation and dynamics of transmembrane peptides: the power of simple models. Eur. Biophys. J. 39, 609–621. Hong, C., Tieleman, D.P., Wang, Y., 2014. Microsecond molecular dynamics simulations of lipid mixing. Langmuir 30, 11993–12001. Horta, B.A.C., Peric-Hassler, L., Hünenberger, P.H., 2010. Interaction of the disaccharides trehalose and gentiobiose with lipid bilayers: a comparative molecular dynamics study. J. Mol. Graphics Model. 29, 331–346. Huang, J.Y., Feigenson, G.W., 1999. A microscopic interaction model of maximum solubility of cholesterol in lipid bilayers. Biophys. J. 76, 2142–2157. Hug, S., 2012. Classical molecular dynamics in a nutshell. In: Monticelli, L., Salonen, E. (Eds.), Biomolecular Simulations Methods and Protocols Series: Methods in Molecular Biology, vol. 924. Springer, New York, pp. 127–252. Hughes, Z.E., Malajczuk, C.J., Mancera, R.L., 2013. The effects of cryosolvents on DOPC-b-Sitosterol bilayers determined from molecular dynamics simulations. J. Phys. Chem. B 117, 3362–3375.

100


Hung, A., Yarovsky, I., 2011. Gap junction hemichannel interactions with zwitterionic lipid, anionic lipid, and cholesterol: molecular simulation studies. Biochemistry 50, 1492–1504. Hunte, C., 2005. Specific protein-lipid interactions in membrane proteins. Biochem. Soc. Trans. 33, 938–942. Hyvönen, M.T., Kovanen, P.T., 2003. Molecular dynamics simulation of sphingomyelin bilayer. J. Phys. Chem. B 107, 9102–9108. Hölttävuori, M., Uronen, R.L., Repakova, J., Salonen, E., Vattulainen, I., Panula, P., Li, Z., Bittman, R., Ikonen, E., 2008. BODIPY-cholesterol: a new tool to visualize sterol trafficking in living cells and organisms. Traffic 9, 1839–1849. Iida-Tanaka, N., Fukase Koichi Utsumi, H., Ishizuka, I., 2000. Conformational studies on a unique bis-sulfated glycolipid using NMR spectroscopy and molecular dynamics simulations. Eur. J. Biochem. 267, 6790–6797. Ingolfsson, H.I., Melo, M.N., van Eerden, F.J., Arnarez, C., Lopez, C., Wassenaar, T.A., Periole, X., de Vries, A.H., Tieleman, D.P., Marrink, S.J., 2014. Lipid organization of the plasma membrane. J. Am. Chem. Soc. 136, 14554–14559. Jafurulla, Md., Rao, B.D., Sreedevi, S., Ruysschaert, J.-M., Covey, D.F., Chattopadhyay, A., 2013. Stereospecific requirement of cholesterol in the function of the 2 serotonin -1A receptor. Biochim. Biophys. Acta 1838, 158–163. Jaikishan, S., Björkbom, A., Slotte, J.P., 2010. Sphingomyelin analogs with branched N-acyl chains: the position of branching dramatically affects acyl chain order and sterol interactions in bilayer membranes. Biochim. Biophys. Acta 1798, 1987–1994. Jaikishan, S., Slotte, J.P., 2011. Effect of hydrophobic mismatch and interdigitation on sterol/sphingomyelin interaction in ternary bilayer membranes. Biochim. Biophys. Acta 1808, 1940–1945. Jedlovszky, P., Sega, M., Vallauri, R., 2009. GM1 ganglioside embedded in a hydrated DOPC membrane: a molecular dynamics simulation study. J. Phys. Chem. B 113, 4876–4886. Jerabek, H., Pabst, G., Rappolt, M., Stockner, T., 2010. Membrane mediated effect on ion channels induced by the anesthetic drug ketamine. J. Am. Chem. Soc. 132, 7990–7997. Jo, S., Rui, H., Lim, J.B., Klauda, J.B., Im, W., 2010. Cholesterol flip-flop: insights from free energy simulation studies. J. Phys. Chem. B 114, 13342–13348. Johansson, A.C.V., Lindahl, E., 2009. The role of lipid composition for insertion and stabilization of amino acids in membranes. J. Chem. Phys. 130, 185101. Jungwirth, P., 2014. Ions at biological interfaces. Encyclopedia of Applied Electrochemistry. Springer, Berlin, pp. 1131. Jurkiewicz, P., Cwiklik, L., Jungwirth, P., Hof, M., 2012. Lipid hydration and mobility: an interplay between fluorescence solvent relaxation experiments and molecular dynamics simulations. Biochimie 94, 26–32. Jämbeck, J.P.M., Lyubartsev, A.P., 2012a. Derivation and systematic validation of a refined all-atom force field for phosphatidylcholine lipids. J. Phys. Chem. B 116, 3164–3179. Jambeck, J.P.M., Lyubartsev, A.P., 2012b. An extension and further validation of an all-atomistic force field for biological membranes. J. Chem. Theory Comput. 8, 2938–2948. Jambeck, J.P.M., Lyubartsev, A.P., 2013. Another piece of the membrane puzzle: extending Slipids further. J. Chem. Theory Comput. 9, 774–784. Kaiser, H.-J., Orlowski, A., Róg, T., Nyholm, T.K.M., Chai, W., Feizi, T., Lingwood, D., Vattulainen, I., Simons, K., 2011. Lateral sorting in model membranes by cholesterol mediated hydrophobic matching. Proc. Natl. Acad. Sci. USA 108, 16628–16633. Kanai, R., Ogawa, H., Vilsen, B., Cornelius, F., Toyoshima, C., 2013. Crystal structure of a Na1-bound Na1,K1-ATPase preceding the E1P state. Nature 502, 201–206. Kang, H., Klauda, J.B., 2014. Molecular dynamics simulations of palmitoyloleoylphosphatidylglycerol bilayers. Mol. Sim doi:http://dx.doi.org/10.1080/ 08927022.2014.926548 in press. Kapla, J., Stevensson, B., Dahlberg, M., Maliniak, A., 2012. Molecular dynamics simulations of membranes composed of glycolipids and phospholipids. J. Phys. Chem. B 116, 244–252. Kasahara, K., Sanai, Y., 1999. Possible roles of glycosphingolipids in lipid rafts. Biophys. Chem. 82, 121–127. Kasahara, K., Sanai, Y., 2000. Functional roles of glycosphingolipids in signal transduction via lipid rafts. Glycoconjug. J. 17, 153–162. Kemmerer, S., Voss, J.C., Faller, R., 2013. Molecular dynamics simulation of dipalmitoylphosphatidylcholine modified with a MTSL nitroxide spin label in a lipid membrane. Biochim. Biophys. Acta 1828, 2770–2777. Khajeh, A., Modarress, H., 2014a. The influence of cholesterol on interactions and dynamics of ibuprofen in a lipid bilayer. Biochim. Biophys. Acta 1838, 2431–2438. Khajeh, A., Modarress, H., 2014b. Effect of cholesterol on behavior of 5-fluorouracil (5-FU) in a DMPC lipidbilayer a molecular dynamics study. Biophys. Chem. 187, 43–50. Khandelia, H., Loubet, B., Olzynska, A., Jurkiewicz, P., Hof, M., 2014. Pairing of cholesterol with oxidized phospholipid species in lipid bilayers. Soft Matter 10, 639–647. Khelashvili, G., Grossfield, A., Feller, S.E., Pitman, M.C., Weinstein, H., 2009. Structural and dynamic effects of cholesterol at preferred sites of interaction with rhodopsin identified from microsecond length molecular dynamics simulations. Proteins 76, 403–417. Khelashvili, G., Harries, D., 2013a. How cholesterol tilt modulates the mechanical properties of saturated and unsaturated lipid membranes. J. Phys. Chem. B 117, 2411–2421.

Khelashvili, G., Harries, D., 2013b. How sterol tilt regulates properties and organization of lipid membranes and membrane insertions. Chem. Phys. Lipids 169, 113–123. Khelashvili, G., Johner, N., Zhao, G., Harries, D., Scott, H.L., 2014. Molecular origins of bending rigidity in lipids with isolated and conjugated double bonds: the effect of cholesterol. Chem. Phys. Lipids 178, 18–26. Khelashvili, G., Mondal, S., Andersen, O.S., Weinstein, H., 2010b. Cholesterol modulates the membrane effects and spatial organization of membrane-penetrating ligands for G-protein coupled receptors. J. Phys. Chem. B 114, 12046–12057. Khelashvili, G., Pabst, G., Harries, D., 2010a. Cholesterol orientation and tilt modulus in DMPC bilayers. J. Phys. Chem. B 114, 7524–7534. Khelashvili, G.A., Pandit, S.A., Scott, H.L., 2005. Self-consistent mean-field model based on molecular dynamics: application to lipid-cholesterol bilayers. J. Chem. Phys. 123, 034910. Kirschner, K.N., Yongye, A.B., Tschampel, S.M., Daniels, C.R., Foley, B.L., Woods, R.J., 2008. GLYCAM06: a generalizable biomolecular force field. Carbohydrates. J. Comput. Chem. 29, 622–655. Klauda, J.B., Garrison, S.L., Arora, G., Jiang, J., Sandler, S.I., 2004. HM-IE: a quantum chemical hybrid method for accurate interaction energies. J. Phys. Chem. A 108, 107–112. Klauda, J.B., Monje, V., Kim, T., Im, W., 2012. Improving the CHARMM force field for polyunsaturated fatty acid chains. J. Phys. Chem. B 116, 9424–9431. Klauda, J.B., Venable, R.M., Freites, J.A., O’Connor, J.W., Tobias, D.J., MondragonRamirez, C., Vorobyov, I., MacKerell Jr., A.D., Pastor, 2010. Update of the CHARMM all-atom additive force field for lipids: validation on six lipid types. J. Phys. Chem. B 114, 7830–7843. Kopec, W., Khandelia, H., 2014. Reinforcing the membrane-mediated mechanism of action of the anti-tuberculosis candidate drug thioridazine with molecular simulations. J. Comput. Aided. Mol. Des. 28, 123–134. Krause, M.R., Wang, M., Mydock-McGrane, L., Covey, D.F., Tejada, E., Almeida, P.F., Regen, S.L., 2014. Eliminating the roughness in cholesterol’s b-face: does it matter? Langmuir 30, 12114–12118. Kucerka, N., Marquard, D., Harroun, T.A., Wassall, Nieh M.-P., de Jong, S.R., Schafer, D. H., Marrink, L.V., Katsaras, S.J., 2010. Cholesterol in bilayers with PUFA chains: doping with DMPC or POPC results in sterol reorientation and membranedomain formation. Biochemistry 49, 7485–7493. ska, A., Tynkkynen, J., Javanainen, M., Manna, M., Róg, _ n Kulig, W., Jurkiewicz, P., Olzy T., Hof, M., Vattulainen, I., Jungwirth, P., 2014a. Experimental determination and computational interpretation of biophysical properties of lipid bilayers enriched by cholesteryl hemisuccinate. Biochim Biophys Acta doi:http://dx. doi.org/10.1016/j.bbamem.2014.10.032 in press. Kulig, W., Tynkkynen, J., Javanainen, M., Manna, M., Róg, T., Vattulainen, I., Jungwirth, P., 2014b. How well does cholesteryl hemisuccinate mimic cholesterol in saturated phospholipid bilayers? J. Mol. Model. 20, 2121. Kulovesi, P., Telenius, J., Koivuniemi, A., Brezesinski, G., Vattulainen, I., Holopainen, J. M., 2012. The impact of lipid composition on the stability of the tear fluid lipid layer. Soft Matter 8, 5826–5834. Kuo, A.-T., Chang, C.-H., 2014. Cholesterol-induced condensing and disordering effects on a rigid catanionic bilayer: a molecular dynamics study. Langmuir 30, 55–62. Kwon, H.J., Abi-Mosleh, L., Wang, M.L., Deisenhofer, J., Goldstein, J.L., Brown, M.S., Infante, R.E., 2009. Structure of N-terminal domain of NPC1 reveals distinct subdomains for binding and transfer of cholesterol. Cell 137, 1213–1224. Köberl, M., Hinz, H.J., Rapp, G., 1998. Temperature scanning simultaneous small- and wide-angle X-ray scattering studies on glycolipid vesicles: areas, expansion coefficients and hydration. Chem. Phys. Lipids 91, 13–37. Laursen, M., Yatime, L., Nissen, P., Fedosova, N.U., 2013. Crystal structure of the high-affinity Na+K+-ATPase-ouabain complex with Mg2+ bound in the cation binding site. Proc. Natl. Acad. Sci. USA 110, 10958–10963. Lee, J.Y., Patel, R., Lyman, E., 2013. Ligand-dependent cholesterol interactions with the human A2A adenosine receptor. Chem. Phys. Lipids 169, 39–45. Lehtinen, J., Magarkar, A., Stepniewski, M., Hakola, S., Bergman, M., Róg, T., Yliperttula, M., Urtti, A., Bunker, A., 2012. Analysis of cause of failure of new targeting peptide in PEGylated liposome: molecular modeling as rational design tool for nanomedicine. Eur. J. Pharma. Sci. 46, 121–130. Li, L.B., Vorobyov, I., Allen, T.W., 2012. The role of membrane thickness in charged protein–lipid interactions. Biochim. Biophys. Acta 1818, 135–145. Lim, J.B., Rogaski, B., Klauda, J.B., 2012. Update of the cholesterol force field parameters in CHARMM. J. Phys. Chem. B 116, 203–210. Lingwood, C.A., 2011a. Glycosphingolipid functions. Cold Spring Harbor Persp. Biol. 3, 149–174. Lingwood, D., Binnington, B., Róg, T., Vattulainen, I., Chai, W., Feizi, T., Grzybek, M., Coskun, U., Lingwood, C., Simons, K., 2011b. Cholesterol regulates glycolipid conformation and receptor activity. Nature Chem. Biol. 7, 260–262. Lingwood, D., Kaiser, H.-J., Levental, I., Simons, K., 2009. Lipid rafts as functional heterogeneity in cell membranes. Biochem. Soc. Trans. 37, 955–960. Lingwood, D., Simons, K., 2010. Lipid rafts as a membrane-organizing principle. Science 327, 46–50. Lins, R.D., Straatsma, T.P., 2001. Computer simulation of the rough lipopolysaccharide membrane of Pseudomonas aeruginosa. Biophys. J. 81, 1037–1046. Liu, W., Chun, E., Thompson, A.A., Chubukov, P., Xu, F., Katritch, V., Han, G.W., Roth, C. B., Heitman, L.H., Jzerman, I., Cherezov, A.P., Stevens, V., 2012. Structural basis for allosteric regulation of GPCRs by sodium ions. Science 337, 232–236. Liu, W., Wacker, D., Gati, C., Han, G.W., James, D., Wang, D., Nelson, G., Weierstall, U., Katritch, V., Barty, A., Zatsepin, N.A., Li, D., Messerschmidt, M., Boutet, S.,

T. Róg, I. Vattulainen / Chemistry and Physics of Lipids 184 (2014) 82–104 Williams, G.J., Koglin, J.E., Seibert, M.M., Wang, C., Shah, S.T., Basu, S., Fromme, R., Kupitz, C., Rendek, K.N., Grotjohann, I., Fromme, P., Kirian, R.A., Beyerlein, K.R., White, T.A., Chapman, H.N., Caffrey, M., Spence, J.C., Stevens, R.C., Cherezov, V., 2013. Serial femtosecond crystallography of G protein-coupled receptors. Science 342, 1521–1524. Liu, Y.Z., Chipot, C., Shao, X.G., Cai, W.S., 2011. The effects of 7-dehydrocholesterol on the structural properties of membranes. Phys. Biol. 8, 056005. Lopez, C.A., de Vries, A.H., Marrink, S.J., 2011. Molecular mechanism of cyclodextrin mediated cholesterol extraction. PLoS Comput. Biol. 7, e1002020. Lopez, C.A., Rzepiela, A.J., de Vries, A.H., Dijkhuizen, L., Hunenberger, P.H., Marrink, S. J., 2009. Martini coarse-grained force field: extension to carbohydrates. J. Chem. Theory Comput. 5, 3195–3210. Lopez, C.A., Sovova, Z., van Eerden, F.J., de Vries, A.H., Marrink, S.J., 2013. Martini force field parameters for glycolipids. J. Chem. Theory Comput. 9, 1694–1708. Loubet, B., Lomholt, M.A., Khandelia, H., 2013. Tension moderation and fluctuation spectrum in simulated lipid membranes under an applied electric potential. J. Chem. Phys. 139, 164902. Loura, L.M.S., Martins do Canto, A.M.T., Martins, J., 2013. Sensing hydration and behavior of pyrene in POPC and POPC/cholesterol bilayers: a molecular dynamics study. Biochim. Biophys. Acta 1828, 1094–1101. Loura, L.M.S., Ramalho, J.P.P., 2011. Recent developments in molecular dynamics simulations of fluorescent membrane probes. Molecules 16, 5437–5452. Lu, L., Voth, G.A., 2009. Systematic coarse-graining of a multicomponent lipid bilayer. J. Phys. Chem. B 113, 1501–1510. Lyman, E., Higgs, C., Kim, B., Lupyan, D., Shelley, J.C., Farid, R., Voth, G.A., 2009. A role for a specific cholesterol interaction in stabilizing the apo configuration of the human A2A adenosine receptor. Structure 17, 1660–1668. Lyubartsev, A.P., Rabinovich, A.L., 2011. Recent development in computer simulations of lipid bilayers. Soft Matter 7, 25–39. MacCallum, J.L., Bennett, W.F.D., Tieleman, D.P., 2008. Distribution of amino acids in a lipid bilayer from computer simulations. Biophys. J. 94, 3393–3404. Maciejewski, A., Pasenkiewicz-Gierula, M., Cramariuc, O., Vattulainen, I., Róg, T., 2014. Refined OPLS-AA force field for saturated phosphatidylcholine bilayers at full hydration. J. Phys. Chem. B 118, 4571–4581. Magarkar, A., Róg, T., Bunker, A., 2014a. Molecular dynamics simulation of PEGylated membranes with cholesterol: building towards the DOXILR formulation. J. Phys. Chem. C 118, 15541–15549. Magarkar, A., Dhawan, V., Kallinteri, P., Viitala, T., Elmowafy, M., Róg, T., Bunker, A., 2014b. Cholesterol level affects surface charge of lipid membranes in physiological environment. Sci. Rep. 4, 5005. Magarkar, A., Karakas, E., Stepniewski, M., Rog, T., Bunker, A., 2012. Molecular dynamics simulation of PEGylated bilayer interacting with salt ions: a model of the liposome surface in the bloodstream. J. Phys. Chem. B 116, 4212–4219. Manglik, A., Kruse, A.C., Kobilka, T.S., Thian, F.S., Mathiesen, J.M., Sunahara, R.K., Pardo, L., Weis, W.I., Kobilka, B.K., Granier, S., 2012. Crystal structure of the m-opioid receptor bound to a morphinan antagonist. Nature 485, 321–326. Manickam Achari, V., Seng Nguan, H., Heidelberg, T., Bryce, R.A., Hashim, R., 2012. Molecular dynamics study of anhydrous lamellar structures of synthetic glycolipids: effects of chain branching and disaccharide headgroup. J. Phys. Chem. B 116, 11626–11634. Manna, M., Mukhopadhyay, C., 2011. Cholesterol driven alteration of the conformation and dynamics of phospholamban in model membrane. Phys Chem. Chem. Phys. 13, 20188–20198. Manna, M., Róg, T., Vattulainen, I., 2014. The challenges of understanding glycolipid functions: an open outlook based on molecular simulations. Biochim. Biophys. Acta 1841, 1130–1145. Mannock, D.A., Lewis, R.N.A.H., McMullen, T.P.W., McElhaney, R.N., 2010. The effect of variations in phospholipid and sterol structure on the nature of lipid-sterol interactions in lipid bilayer model membranes. Chem. Phys. Lipids 163, 403– 448. Markiewicz, M., Pasenkiewicz-Gierula, M., 2011. Comparative model studies of gastric toxicity of nonsteroidal anti-inflammatory drugs. Langmuir 27, 6950– 6961. Marrink, S.J., Risselada, H.J., Yefimov, S., Tieleman, D.P., de Vries, A.H., 2007. The MARTINI force field: coarse grained model for biomolecular simulations. J. Phys. Chem. B 111, 7812–7824. Marrink, S.J., Risselada, J., Mark, A.E., 2005. Simulation of gel phase formation and melting in lipid bilayers using a coarse grained model. Chem. Phys. Lipids 135, 223–244. Marrink, S.J., Tieleman, D.P., 2013. Perspective on the Martini model. Chem. Soc. Rev. 42, 6801–6822. Marsh, D., 2012. Bilayer dimensions and hydration of glycolipids. Chem. Phys. Lipids 165, 23–31. Martinez-Seara, H., Róg, T., Karttunen, M., Vattulainen, I., Reigada, R., 2010. Cholesterol induces specific spatial and orientational order in cholesterol/ phospholipid membranes. PLoS One 5, e11162. Martinez-Seara, H., Róg, T., Pasenkiewicz-Gierula, M., Vattulainen, I., Karttunen, M., Reigada, R., 2008. Interplay of unsaturated phospholipids and cholesterol in membranes: Effect of double bond position. Biophys. J. 95, 3295–3305. Massey, J.B., 1998. Effect of cholesteryl hemisuccinate on the interfacial properties of phosphatidylcholine bilayers. Biochim. Biophys. Acta 1415, 193–204. Matsuzaki, K., Kato, K., Yanagisawa, K., 2010. Ab polymerization through interaction with membrane gangliosides. Biochim. Biophys. Acta 1801, 868–877. Maula, T., Kurita, M., Yamaguchi, S., Yamamoto, T., Katsumura, S., Slotte, J.P., 2011. Effects of sphingosine 2N- and 3O-methylation on palmitoyl ceramide properties in bilayer membranes. Biophys. J 101, 2948–2956.

101

Maxfield, F.R., Mondal, M., 2006. Sterol and lipid trafficking in mammalian cells Biochem. Soc. Trans. 34, 335–339. Mendes Ferreira, T., Coreta-Gomes, F., Ollila, O.H.S., Joao Moreno, M., Vaz, W.L.C., Topgaard, D., 2013. Cholesterol and POPC segmental order parameters in lipid membranes: solid state 1H–13C NMR and MD simulation studies. Phys. Chem. Chem. Phys. 15, 1976–1989. Metcalf, R., Pandit, S.A., 2012. Mixing properties of sphingomyelin ceramide bilayers: a simulation study. J. Phys. Chem. B 116, 4500–4509. Mombelli, E., Morris, R., Taylor, W., Fraternali, F., 2003. Hydrogen bonding propensities of sphingomyelin in solution and in a bilayer assembly: a molecular dynamics study. Biophys. J. 84, 1507–1517. Mondal, S., Mukhopadhyay, C., 2008. Molecular level investigation of organization in ternary lipid bilayer: a computational approach. Langmuir 24, 10298–10305. Monticelli, L., Kandasamy, S.K., Periole, X., Larson, R.G., Tieleman, D.P., Marrink, S.J., 2008. The MARTINI coarse-grained force field: extension to proteins. J. Chem. Theory Comput. 4, 819–834. Monticelli, L., Tieleman, D.P., 2012. Force fields for classical molecular dynamics. In: Monticelli, L., Salonen, E. (Eds.), Biomolecular Simulations Methods and Protocols Series Methods in Molecular Biology, vol. 924. Springer, New York, pp. 197–214. Mori, K., Mahmood, M.I., Neya, S., Matsuzaki, K., Hoshino, T., 2012. Formation of GM1 ganglioside clusters on the lipid membrane containing sphingomyeline and cholesterol. J. Phys. Chem. B 116, 5111–5121. Morth, J.P., Pedersen, B.P., Toustrup-Jensen, M.S., Sorensen, T.L., Petersen, J., Andersen, J.P., Vilsen, B., Nissen, P., 2007. Crystal structure of the sodium– potassium pump. Nature 450, 1043–1049. Murtola, T., Vuorela, T., Hyvonen, M.T., Marrink, S.J., Karttunen, M., Vattulainen, I., 2011. Low-density lipoprotein: structure, dynamics and lipid–apoB-100 interactions. Soft Matter 7, 8135–8141. Murzyn, K., Róg, T., Pasenkiewicz-Gierula, M., 2004. Interactions of magainin2 amide with membrane lipids. Lect. Notes Comput. Sci. 3037, 325–331. Mydock-McGrane, L.N., Rath, P., Covey, D.F., 2014. Synthesis of a smoothened cholesterol: 18,19-Di-nor-cholesterol. J. Org. Chem. 79, 5636–5643. Neale, C., Bennett, W.F.D., Tieleman, D.P., Pomes, R., 2011. Statistical convergence of equilibrium properties in simulations of molecular solutes embedded in lipid bilayers. J. Chem. Theory Comput. 7, 4175–4188. Neuvonen, M., Manna, M., Mokkila, S., Javanainen, M., Róg, T., Liu, Z., Bittman, R., Vattulainen, I., Ikonen, E., 2014. Enzymatic oxidation of cholesterol: properties and functional effects of cholestenone in cell membranes. PLoS ONE 9, e103743. Niemela, P.S., Castillo, S., Sysi-Aho, M., Oresic, M., 2009. Bioinformatics and computational methods for lipidomics. J. Chromatogr. B 877, 2855–2862. Niemela, P.S., Hyvonen, M.T., Vattulainen, I., 2004. Structure and dynamics of sphingomyelin bilayer Insight gained through systematic comparison to phosphatidylcholine. Biophys. J. 87, 2976–2989. Niemela, P.S., Hyvonen, M.T., Vattulainen, I., 2006. Influence of chain length and unsaturation on sphingomyelin bilayers. Biophys. J. 90, 851–863. Niemela, P.S., Miettinen, M.S., Monticelli, L., Hammaren, H., Bjelkmar, P., Murtola, T., Lindahl, E., Vattulainen, I., 2010. Membrane proteins diffuse as dynamic complexes with lipids. J. Am. Chem. Soc. 132, 7574. Niemela, P.S., Ollila, S., Hyvonen, M.T., Karttunen, M., Vattulainen, I., 2007. Assessing the nature of lipid raft membranes. PLoS Comput. Biol. 3, e34. Nyblom, M., Poulsen, H., Gourdon, P., Reinhard, L., Andersson, M., Lindahl, E., Fedosova, N., Nissen, P., 2013. Crystal structure of Na+, K(+)-ATPase in the Na (+)-bound state. Science 342, 123–127. Nyholm, T.K.M., Ozdirekcan, S., Killian, J.A., 2007. How protein transmembrane segments sense the lipid environment. Biochemistry 46, 1457–1465. O’Connor, J.W., Klauda, J.B., 2011. Lipid membranes with a majority of cholesterol: applications to the ocular lens and aquaporin 0. J. Phys. Chem. B 115, 6455–6464. Ogawa, H., Shinoda, T., Cornelius, F., Toyoshima, C., 2009. Crystal structure of the sodium–potassium pump (Na+,K+-ATPase) with bound potassium and ouabain. Proc. Natl. Acad. Sci. USA 106, 13742–13747. Ogushi, F., Ishitsuka, R., Kobayashi, T., Sugita, Y., 2012. Rapid flip-flop motions of diacylglycerol and ceramide in phospholipid bilayers. Chem. Phys. Lett. 522, 96– 102. Ohvo-Rekila, H., Ramstedt, B., Leppimaki, P., Slotte, J.P., 2002. Cholesterol interactions with phospholipids in membranes. Prog. Lip. Res. 41, 66–97. Ollila, O.H.S., Louhivuori, M., Marrink, S.J., Vattulainen, I., 2011. Protein shape change has a major effect on the gating energy of a mechanosensitive channel. Biophys. J. 100, 1651–1659. Ollila, O.H.S., Róg, T., Karttunen, M., Vattulainen, I., 2007. Role of sterol type on lateral pressure profiles of lipid membranes affecting membrane protein functionality: comparison between cholesterol, desmosterol, 7-dehydrocholesterol and ketosterol. J. Struc. Biol. 159, 311–323. Olsen, B.N., Schlesinger, P.H., Baker, N.A., 2009. Perturbations of membrane structure by cholesterol and cholesterol derivatives are determined by sterol orientation. J. Am. Chem. Soc. 131, 4854–4865. Olsen, B.N., Schlesinger, P.H., Ory, D.S., Baker, N.A., 2011. 25-hydroxycholesterol increases the availability of cholesterol in phospholipid membranes. Biophys. J. 100, 948–956. Orlowski, A., Bunker, A., Pasenkiewicz-Gierula, M., Vattulainen, I., Männistö, P.T., Róg, T., 2012. Strong preferences of dopamine and L-dopa towards lipid head group: importance of phosphatidylserine and its implication for neurotransmitters metabolism. J. Neurochem. 122, 681–690. Orsi, M., Essex, J.W., 2011. The ELBA force field for coarse-grain modeling of lipid membranes. PLoS One 6, e28637.

102


Pan, J., Cheng, X., Heberle, F.A., Mostofian, B., Kucerka, N., Drazba, P., Katsaras, J., 2012. Interactions between ether phospholipids and cholesterol as determined by scattering and molecular dynamics simulations. J. Phys. Chem. B 116, 14829– 14838. Pandit, S.A., Jakobsson, E., Scott, H.L., 2004b. Simulation of the early stages of nanodomain formation in mixed bilayers of sphingomyelin cholesterol, and dioleylphosphatidylcholine. Biophys. J. 87, 3312–3322. Pandit, S.A., Scott, H.L., 2009. Multiscale simulations of heterogeneous model membranes. Biochim. Biophys. Acta 1788, 136–148. Pandit, S.A., Vasudevan, S., Chiu, S.W., Mashl, R.J., Jakobsson, E., Scott, H.L., 2004a. Sphingomyelin-cholesterol domains in phospholipid membranes: atomistic simulation. Biophys. J 87, 1092–1100. Paradies, G., Paradies, V., De Benedictis, V., Ruggiero, F.M., Petrosillo, G., 2014. Functional role of cardiolipin in mitochondrial bioenergetics. Biochi. Biophys. Acta 1837, 408–417. Parisio, G., Ferrarini, A., 2010. Solute partitioning into lipid bilayers: an implicit model for nonuniform and ordered environment. J. Chem. Theory Comput. 6, 2267–2280. Parisio, G., Sperotto, M.M., Ferrarini, A., 2012. Flip-Flop of steroids in phospholipid bilayers: effects of the chemical structure on transbilayer diffusion. J. Am. Chem. Soc. 134, 12198–12208. Parthasarathi, R., Tian, J., Redondo, A., Gnanakaran, S., 2011. Quantum chemical study of carbohydrate-phospholipid interactions. J. Phys. Chem. A 115, 12826– 12840. ski, K., Murzyn, K., Markiewicz, M., 2012. Pasenkiewicz-Gierula, M., Baczyn Orientation of lutein in a lipid bilayer–revisited. Acta Biochim. Polon. 59, 115–118. Pasenkiewicz-Gierula, M., Róg, T., Kitamura, K., Kusumi, A., 2000. Cholesterol effects on the phosphatidylcholine bilayer polar region: a molecular simulation study. Biophys. J. 78, 1376–1389. Pastor, R.W., MacKerell Jr., R.Y., 2011. Development of the CHARMM force field for lipids. J. Phys. Chem. Lett. 2, 1526–1532. Patel, Balaji, P.V., 2007. Characterization of the conformational and orientational dynamics of ganglioside GM1 in a dipalmitoylphosphatidylcholine bilayer by molecular dynamics simulations. Biochim. Biophys. Acta 1768, 1628–1640. Patel, R.Y., Balaji, P.V., 2008. Characterization of symmetric and asymmetric lipid bilayers composed of varying concentrations of ganglioside GM1 and DPPC. J. Phys. Chem. B 112, 3346–3356. Patel, R.Y., Balaji, P.V., 2011. Structure and dynamics of glycosphingolipids in lipid bilayers: insights from molecular dynamics simulations. Int. J. Carbohydrate Chem Article ID 950256. Patterson, G.H., Hirschberg, K., Polishchuk, R.S., Gerlich, D., Phair, R.D., LippincottSchwartz, J., 2008. Transport through the golgi apparatus by rapid partitioning within a two-phase membrane system. Cell 133, 1055–1067. Penmatsa, A., Wang, K.H., Gouaux, E., 2013. X-ray structure of dopamine transporter elucidates antidepressant mechanism. Nature 503, 85–90. Perlmutter, J.D., Sachs, J.N., 2009. Inhibiting lateral domain formation in lipid bilayers: simulations of alternative steroid headgroup chemistries. J. Am. Chem. Soc. 131, 16362–16363. Pietilainen, K.H., Róg, T., Seppanen-Laakso, T., Virtue, S., Gopalacharyulu, P., Tang, J., Rodriguez-Cuenca, S., Maciejewski, A., Naukkarinen, J., Ruskeepaa, A.-L., Niemela, P.S., Yetukuri, L., Tan, C.Y., Velagapudi, V., Castillo, S., Nygren, H., Hyotylainen, T., Rissanen, A., Kaprio, J., Yki-Jarvinen, H., Vattulainen, I., VidalPuig, A., Oresic, M., 2011. Association of lipidome remodeling in the adipocyte membrane with acquired obesity in humans. PLoS Biol 9, e1000623. Piggot, T.J., Holdbrook, D.A., Khalid, S., 2011. Electroporation of the E. coli and S. aureus membranes: molecular dynamics simulations of complex bacterial membranes. J. Phys. Chem. B 115, 13381–13388. Plesnar, E., Subczynski, W.K., Pasenkiewicz-Gierula, M., 2013. Comparative computer simulation study of cholesterol in hydrated unary and binary lipid bilayers and in an anhydrous crystal. J. Phys. Chem. B 117, 8758–8769. Plesnar, E., Subczynski, W.K., Pasenkiewicz-Gierula, M., 2012. Saturation with cholesterol increases vertical order and smoothes the surface of the phosphatidylcholine bilayer: a molecular simulation study. Biochim. Biophys. Acta 1818, 520–529. Poger, D., Mark, A.E., 2013. The relative effect of sterols and hopanoids on lipid bilayers: when comparable is not identical. J. Phys. Chem. B 117, 16129–16140. Polley, A., Vemparala, S., 2013. Partitioning of ethanol in multi-component membranes: effects on membrane structure. Chem. Phys. Lipids 166, 1–11. Polley, A., Vemparala, S., Rao, M., 2012. Atomistic simulations of a multicomponent asymmetric lipid bilayer. J. Phys. Chem. B 116, 13403–13410. Pourmousa, A.M., Róg, T., Mikkeli, R., Vattulainen, I., Solanko, L.M., Wustner, D., Holmgaard List, N., Kongsted, J., Karttunen, M., 2014. Cholesterol and its analogues: the effect of dehydroergosterol on cell membranes. J. Phys. Chem. B 118, 7345–7357. Prakash, A., Janosi, L., Doxastakis, M., 2011. GxxxG motifs phenylalanine, and cholesterol guide the self-association of transmembrane domains of ErbB2 receptors. Biophys. J. 101, 1949–1958. Puri, V., Watanabe, R., Dominguez, M., Sun, X.F., Wheatley, C.L., Marks, D.L., Pagano, R.E., 1999. Cholesterol modulates membrane traffic along the endocytic pathway in sphingolipid storage diseases. Nature Cell Biol. 1, 386–388. Pöyry, S., Cramariuc, O., Postila, P.A., Kaszuba, K., Sarewicz, M., Osyczka, A., Vattulainen, I., Róg, T., 2013. Atomistic simulations indicate cardiolipin to have an integral role in the structure of the cytochrome bc1 complex. Biochim. Biophys. Acta 1827, 769–778.

Pöyry, S., Róg, T., Karttunen, M., Vattulainen, I., 2008. Significance of cholesterol methyl groups. J. Phys. Chem. B 112, 2922–2929. Qiu, L., Buie, C., Reay, A., Vaughn, M.W., Cheng, K.H., 2011. Molecular dynamics simulations reveal the protective role of cholesterol in b-amyloid proteininduced membrane disruptions in neuronal membrane mimics. J. Phys. Chem. B 115, 9795–9812. Raguz, M., Mainali, L., Widomska, J., Subczynski, W.K., 2011. The immiscible cholesterol bilayer domain exists as an integral part of phospholipid bilayer membranes. Biochim. Biophys. Acta 1808, 1072–1080. Ramadurai, S., Holt, A., Schafer, L.V., Krasnikov, V.V., Rijkers, D.T.S., Marrink, S.J., Killian, J.A., Poolman, B., 2010. Influence of hydrophobic mismatch and amino acid composition on the lateral diffusion of AT peptides. Biophys. J. 99, 1447– 1454. Ramstedt, B., Slotte, J.P., 2002. Membrane properties of sphingomyelins. FEBS Lett. 531, 33–37. Ramstedt, B., Slotte, J.P., 2006. Sphingolipids and the formation of sterol-enriched ordered membrane domains. Biochim. Biophys. Acta 1758, 1945–1956. Rantamaki, A., Telenius, J., Koivuniemi, A., Vattulainen, I., Holopainen, J.M., 2011. Lessons from the biophysics of interfaces lung surfactant and tear fluid. Prog. Retinal Eye Res. 30, 204–215. Reigada, R., 2014. Electroporation of heterogeneous lipid membranes. Biochim. Biophys. Acta 1838, 814–821. Rheinstädter, M.C., Mouritsen, O.G., 2013. Small-scale structure in fluid cholesterol– lipid bilayers. Curr. Opin. Colloid Interface Sci. 18, 440–447. Risselada, H.J., Marrink, S.J., 2008. The molecular face of lipid rafts in model membranes. Proc. Natl. Acad. Sci. USA 105, 17367–17372. Risselada, H.J., Marrink, S.J., Muller, M., 2011. Curvature-dependent elastic properties of liquid-ordered domains result in inverted domain sorting on uniaxially compressed vesicles. Phys. Rev. Lett. 106, 148102. Robalo, J.R., Martins do Canto, A.M.T., Carvalho, A.J.P., Ramalho, J.P.P., Loura, L.M.S., 2013. Behavior of fluorescent cholesterol analogues dehydroergosterol and cholestatrienol in lipid bilayers: a molecular dynamics study. J. Phys. Chem. B 117, 5806–5819. Robalo, J.R., Ramalho, J.P.P., Loura, L.M.S., 2013b. NBD-labeled cholesterol analogues in phospholipid bilayers: insights from molecular dynamics. J. Phys. Chem. B 117, 13731–13742. Róg, T., Bunker, A., Vattulainen, I., Karttunen, M., 2007b. Effect of glucose and galactose headgroup on lipid bilayer properties. J. Phys. Chem. B 111, 10146– 10154. Róg, T., Murzyn, K., Karttunen, M., Pasenkiewicz-Gierula, M., 2008b. Non-polar interactions between trans-membrane helical peptide and phosphatidylcholines, sphingomyelins and cholesterol molecular dynamics simulation studies. J. Pept. Sci. 14, 374–382. Róg, T., Pasenkiewicz-Gierula, M., 2003. Comparison of effects of epicholesterol and cholesterol on the phosphatidylcholine bilayer: a molecular dynamics simulation study. Biophys. J. 84, 1818–1826. Róg, T., Pasenkiewicz-Gierula, M., 2006. Cholesterol–sphingomyelin interactions: a molecular dynamics simulation study. Biophys. J. 91, 3756–3767. Róg, T., Pasenkiewicz-Gierula, M., Vattulainen, I., Karttunen, M., 2009a. Ordering effects of cholesterol and its analogues. Biochim. Biophys. Acta 1788, 97–121. Róg, T., Stimson, L.M., Pasenkiewicz-Gierula, M., Vattulainen, I., Karttunen, M., 2008a. Replacing the cholesterol hydroxyl group with the ketone group facilitates sterol flip-flop and promotes membrane fluidity. J. Phys. Chem. B 112, 1946–1952. Róg, T., Vattulainen, I., Jansen, M., Ikonen, E., Karttunen, M., 2008. Comparison of cholesterol and its direct precursors along the biosynthetic pathway: effects of cholesterol, desmosterol and 7-dehydrocholesterol on saturated and unsaturated lipid bilayers. J. Chem. Phys 129, 154508. Róg, T., Vattulainen, I., Karttunen, M., 2005. Modeling glycolipids: take one. Cell. Mol. Biol. Lett. 10, 625–630. Róg, T., Vattulainen, I., Pasenkiewicz-Gierula, M., Karttunen, M., 2007a. What happens if cholesterol is made smoother: importance of methyl substituents in cholesterol ring structure on phosphatidylcholine-sterol interaction. Biophys. J 92, 3346–3357. Rosenbaum, D.M., Zhang, C., Lyons, J.A., Holl, R., Aragao, D., Arlow, D.H., Rasmussen, S.G.F., Choi, H.J., Devree, B.T., Sunahara, R.K., Chae, P.S., Gellman, S.H., Dror, R.O., Shaw, D.E., Weis, W.I., Caffrey, M., Gmeiner, P., Kobilka, B.K., 2011. Structure and function of an irreversible agonist-b(2) adrenoceptor complex. Nature 469, 236–240. Rosetti, C., Pastorino, C., 2012. Comparison of ternary bilayer mixtures with asymmetric or symmetric unsaturated phosphatidylcholine lipids by coarse grained molecular dynamics simulations. J. Phys. Chem. B 116, 3525–3537. Roy, D., Mukhopadhyay, C., 2001. GD1a in phospholipid bilayer: a molecular dynamics simulation. J. Biomolec. Struc. Dyn. 18, 639–646. Roy, D., Mukhopadhyay, C., 2002. Molecular dynamics simulation of GM1 in phospholipid bilayer. J. Biomolec. Struct. Dyn. 19, 1121–1132. Sadiq, S.K., Guixà-González, R., Dainese, E., Pastor, M., De Fabritiis, G., Selent, J., 2013. Molecular modeling and simulation of membrane lipid-mediated effects on GPCRs. Curr. Med. Chem 20, 22–38. Sáenz, J.P., Sezgin, E., Schwille, P., Simons, K., 2012. Functional convergence of hopanoids and sterols in membrane ordering. Proc. Natl. Acad. Sci. USA 109, 14236–14240. Saito, H., Shinoda, W., 2011. Cholesterol effect on water permeability through DPPC and PSM lipid bilayers: a molecular dynamics study. J. Phys. Chem. B 115, 15241– 15250.

T. Róg, I. Vattulainen / Chemistry and Physics of Lipids 184 (2014) 82–104 Sandvig, K., Bergan, J., Kavaliauskiene, S., Skotland, T., 2014. Lipid requirements for entry of protein toxins into cells. Prog. Lipid Res. 54, 1–13. Saxena, R., Chattopadhyay, A., 2012. Membrane cholesterol stabilizes the human serotonin1A receptor. Biochim. Biophys. Acta 1818, 2936–2942. Schmidt, U., Weiss, M., 2010. Hydrophobic mismatch-induced clustering as a primer for protein sorting in the secretory pathway. Biophys. Chem. 151, 34–38. Sega, M., Jedlovszky, P., Vallauri, R., 2006. Molecular dynamics simulation of GM1 gangliosides embedded in a phospholipid membrane. J. Mol. Liq. 129, 86– 91. Sega, M., Vallauri, R., Brocca, P., Cantu, L., Melchionna, S., 2007. Short-range structure of a GM3 ganglioside membrane: comparison between experimental WAXS and computer simulation results. J. Phys. Chem. B 111, 10965–10969. Sega, M., Vallauri, R., Brocca, P., Melchionna, S., 2004. Molecular dynamics simulation of a GM3 ganglioside bilayer. J. Phys. Chem. B 108, 20322–20330. Sengupta, D., 2012. Cholesterol modulates the structure, binding modes, and energetics of caveolin-membrane interactions. J. Phys. Chem. B 116, 6–14564. Sengupta, D., Chattopadhyay, A., 2012. Identification of cholesterol binding sites in the serotonin1A receptor. J. Phys. Chem. B 116, 12991–12996. Sergelius, C., Niinivehmas, S., Maula, T., Kurita, M., Yamaguchi, S., Yamamoto, T., Katsumura, S., Pentikainen, O.T., Slotte, J.P., 2012. Structure-activity relationship of sphingomyelin analogs with sphingomyelinase from Bacillus cereus. Biochim. Biophys. Acta 1818, 474–480. Sergelius, C., Slotte, J.P., 2011. Membrane properties of and cholesterol’s interactions with a biologically relevant three-chain sphingomyelin: 3O-palmitoyl-Npalmitoyl-D-erythro-sphingomyelin. Biochim. Biophys. Acta 1808, 2841–2848. Shang, Z., Mao, Y., Tero, R., Liu, X., Hoshino, T., Tanaka, M., Urisu, T., 2010. Clustering effects of GM1 and formation mechanisms of interdigitated liquid disordered domains in GM1/SM/cholesterol-supported planar bilayers on mica surfaces. Chem. Phys. Lett. 497, 108–114. Shigematsu, T., Koshiyama, K., Wada, S., 2014. Molecular dynamics simulations of pore formation in stretched phospholipid/cholesterol bilayers. Chem. Phys. Lipids 183, 43–49. Shinoda, T., Ogawa, H., Cornelius, F., Toyoshima, C., 2009. Crystal structure of the sodium-potassium pump at 2.4 A resolution. Nature 459, 446–450. Shintre, C.A., Pike, A.C.W., Li, Q., Kim, J., Barr, A.J., Von Delft, F., Shrestha, S., Gou, L., Yang, J., Berridge, G., Ross, J., Stansfeld, P.J., Edwards, M.S.P., Sans, A.M., Bountra, C., Marsden, B.D., Von Delft, F., Bullock, A.N., Gileadi, O., Burgess-Brown, N., Carpenter, E.P., 2013. Structures of ABCB10: a human ATP-binding cassette transporter in apo- and nucleotide-bound states. Proc. Natl. Acad. Sci. USA 110, 9710–9715. Shipley, G.G., Green, J.P., Nichols, B.W., 1973. Phase behavior of monogalactosyl digalactosyl, and sulfoquinovosyl diglycerides. Biochim. Biophys. Acta 311, 531–544. Shrivastava, S., Devi Paila, Y., Dutta, A., Chattopadhyay, A., 2008. Differential effects of cholesterol and its immediate biosynthetic precursors on membrane organization. Biochemistry 47, 5668–5677. Shroll, R.M., Straatsma, T.P., 2002. Molecular structure of the outer bacterial membrane of Pseudomonas aeruginosa via classical simulation. Biopolymers 65, 395–407. Shroll, R.M., Straatsma, T.P., 2003. Molecular basis for microbial adhesion to geochemical surfaces: computer simulation of Pseudomonas aeruginosa adhesion to goethite. Biophys. J. 84, 1765–1772. Slotte, J.P., 2013. Molecular properties of various structurally defined sphingomyelins–correlation of structure with function. Prog. Lip. Res 52, 206–219. Smith, A.W., 2012. Lipid–protein interactions in biological membranes: a dynamic perspective. Biochim. Biophys. Acta 1818, 172–177. Smondyrev, A.M., Berkowitz, M.L., 2000. Molecular dynamics simulation of dipalmitoylphosphatidylcholine membrane with cholesterol sulfate. Biophys. J. 78, 1672–1680. Soares, T.A., Straatsma, T.P., Lins, R.D., 2008. Influence of the B-band O-antigen chain in the structure and electrostatics of the lipopolysaccharide membrane of Pseudomonas aeruginosa. J. Braz. Chem. Soc. 19, 312–320. Sodt, A.J., Logan Sandar, M., Gawrisch, K., Pastor, R.W., Lyman, E., 2014. The molecular structure of the liquid-ordered phase of lipid bilayers. J. Am. Chem. Soc. 136, 725–732. Song, Y., Kenworthy, A.K., Sanders, C.R., 2014. Cholesterol as a co-solvent and a ligand for membrane proteins. Protein Sci. 23, 1–22. Soubias, O., Gawrisch, K., 2012. The role of the lipid matrix for structure and function of the GPCR rhodopsin. Biochim. Biophys. Acta 1818, 234–240. Sparr, E., Ash, W.L., Nazarov, P.V., Rijkers, D.T.S., Hemminga, M.A., Tieleman, D.P., Killian, J.A., 2005. Self-association of transmembrane-helices in model membranes. Importance of helix orientation and role of hydrophobic mismatch. J. Biol. Chem. 280, 39324–39331. Stansfeld, P.J., Sansom, M.S.P., 2011. From coarse grained to atomistic: a serial multiscale approach to membrane protein simulations. J. Chem. Theory Comput. 7, 1157–1166. ska, A., Amaro, M., Savchenko, D., Deyneka, A., Hermetter, A., _ n Štefl, M., Šachl, R., Olzy Cwiklik, L., Humpolí9 cková, J., Hof, M., 2014. Comprehensive portrait of cholesterol containing oxidized membrane. Biochim. Biophys. Acta 1838, 1769–1776. Stepniewski, M., Bunker, A., Pasenkiewicz-Gierula, M., Karttunen, M., Róg, T., 2010. Effects of the lipid bilayer phase state on the water membrane interface. J. Chem. Phys. B. 114, 11784–11792. Stetter, F.W.S., Cwiklik, L., Jungwirth, P., Hugel, T., 2014. Single lipid extraction: The anchoring strength of cholesterol in liquid-ordered and liquid-disordered phases. Biophys. J. 107, 1167–1175.

103

Stoffel, W., Bosio, A., 1997. Myelin glycolipids and their functions. Curr. Opin. Neurobio. 7, 654–661. Straubinger, R.M., 1993. pH-Sensitive liposomes for delivery of macromolecules into cytoplasm of cultured-cells. Methods Enzymol. 221, 361–376. Subczynski, W.K., Raguz, M., Widomska, J., Mainali, L., Konovalov, A., 2012b. Functions of cholesterol and the cholesterol bilayer domain specific to the fiber-cell plasma membrane of the eye lens. J. Membrane Biol. 245, 51–68. Subczynski, W.K., Wisniewska-Becker, A., Widomska, J., 2012a. Can macular xanthophylls replace cholesterol in formation of the liquid-ordered phase in lipid-bilayer membranes. Acta Biochim. Polon 59, 109–114. Sun, D., Lin, X., Gu, N., 2014. Cholesterol affects C translocation across lipid bilayers. Soft Matter 10, 2160–2168. Teixeira, V., Feio, M.J., Bastos, M., 2012. Role of lipids in the interaction of antimicrobial peptides with membranes. Progr. Lip. Res. 51, 149–177. Tessier, M.B., DeMarco, M.L., Yongye, A.B., Woods, R.J., 2008. Extension of the Glycam06 biomolecular force field to lipids lipid bilayers and glycolipids. Mol. Simul. 34, 349–364. Tieleman, D.P., MacCallum, J.L., Ash, W.L., Kandt, C., Xu, Z.T., Monticelli, L., 2006. Membrane protein simulations with a united-atom lipid and all-atom protein model: lipid–protein interactions, side chain transfer free energies and model proteins. J. Phys. Cond. Matt. 18, 1221–1234. Timr, S., Bondar, A., Cwiklik, L., Stefl, M., Hof, M., Vazdar, M., Lazar, J., Jungwirth, P., 2014. Accurate determination of the orientational distribution of a fluorescent molecule in a phospholipid membrane. J. Phys. Chem. B 118, 855–863. Tumaneng, P.W., Pandit, S.A., Zhao, G., Scott, H.L., 2011. Self-consistent mean-field model for palmitoyloleoylphosphatidylcholine–palmitoyl sphingomyelin–cholesterol lipid bilayers. Phys. Rev. E 83, 031925. Vacha, R., Siu, S.W.I., Petrov, M., Bockmann, R.A., Barucha-Kraszewska, J., Jurkiewicz, P., Hof, M., Berkowitz, M.L., Jungwirth, P., 2009. Effects of alkali cations and halide anions on the DOPC lipid membrane. J. Phys. Chem. A 113, 7235–7243. Wacker, D., Fenalti, G., Brown, M.A., Katritch, V., Abagyan, R., Cherezov, V., Stevens, R. C., 2010. Conserved binding mode of human beta2 adrenergic receptor inverse agonists and antagonist revealed by X-ray crystallography. J. Am. Chem. Soc. 132, 11443–11445. Wacker, D., Wang, C., Katritch, V., Han, G.W., Huang, X.P., Vardy, E., McCorvy, J.D., Jiang, Y., Chu, M., Siu, F.Y., Liu, W., Xu, H.E., Cherezov, V., Roth, B.L., Stevens, R.C., 2013. Structural features for functional selectivity at serotonin receptors. Science 340, 615–619. Wada, T., Shimono, K., Kikukawa, T., Hato, M., Shinya, N., Kim, S.Y., Kimura-Someya, T., Shirouzu, M., Tamogami, J., Miyauchi, S., Jung, K.-H., Kamo, N., Yokoyama, S., 2011. Crystal structure of the eukaryotic light-driven proton-pumping rhodopsin: acetabularia rhodopsin II from marine alga. J. Mol. Biol. 411, 986–998. Vainio, S., Jansen, M., Koivusalo, M., Róg, T., Karttunen, M., Vattulainen, I., Ikonen, E., 2006. J. Biol. Chem. 281, 348–355. van Meer, G., Voelker, D.R., Feigenson, G.W., 2008. Membrane lipids: where they are and how they behave. Nature Rev. Mol. Cell Biol. 9, 112–124. Warne, A., Moukhametzianov, R., Baker, J.G., Nehme, R., Edwards, P.C., Leslie, A.G.W., Schertler, G.F.X., Tate, C.G., 2011. The structural basis for agonist and partial agonist action on a b(1)-adrenergic receptor. Nature 469, 241–244. Vasudevan, S.V., Balaji, P.V., 2001. Dynamics of ganglioside headgroup in lipid environment: molecular dynamics simulations of GM1 embedded in dodecylphosphocholine micelle. J. Phys. Chem. B 105, 7033–7041. Vattulainen, I., Róg, T., 2011. Lipid simulations: a perspective on lipids in action. Cold Spring Harbor Perspect. Biol. 3, a004655. Venable, R.M., Sodt, A.J., Rogaski, B., Rui, H., Hatcher, E., MacKerell Jr., A.D., Pastor, R. W., Klauda, J.B., 2014. CHARMM all-atom additive force field for sphingomyelin: elucidation of hydrogen bonding and of positive curvature. Biophys. J. 107, 134–145. Wennberg, C.L., van der Spoel, D., Hub, J.S., 2012. Large influence of cholesterol on solute partitioning into lipid membranes. J. Am. Chem. Soc. 134, 5351–5361. Wenz, J.J., 2012. Predicting the effect of steroids on membrane biophysical properties based on the molecular structure. Biochim. Biophys. Acta 1818, 896–906. Westerlund, B., Slotte, J.P., 2009. How the molecular features of glycosphingolipids affect domain formation in fluid membranes. Biochim. Biophys. Acta 1788, 194–201. Wizert, D., Iskander, R., Cwiklik, L., 2014. Organization of lipids in the tear film: a molecular-level view. PLoS One 9, e92461. Woods, R.J., Tessier, M.B., 2010. Computational glycoscience: characterizing the spatial and temporal properties of glycans and glycan–protein complexes. Curr. Struct. Biol. 20, 575–583. Wu, H., Wang, C., Gregory, K.J., Han, G.W., Cho, H.P., Xia, Y., Niswender, C.M., Katritch, V., Meiler, J., Cherezov, V., Conn, P.J., Stevens, R.C., 2014. Structure of a class C GPCR metabotropic glutamate receptor 1 bound to an allosteric modulator. Science 344, 58–64. Vyas, A.A., 2002. Gangliosides are functional nerve cell ligands for myelinassociated glycoprotein (MAG): an inhibitor of nerve regeneration. Proc. Natl. Acad. Sci. USA 99, 8412–8417. Wydro, P., Knapczyk, S., Lapczynska, M., 2011. Variations in the condensing effect of cholesterol on saturated versus unsaturated phosphatidylcholines at low and high sterol concentration. Langmuir 27, 5433–5444. Yahi, N., Aulas, A., Fantini, J., 2010. How cholesterol constrains glycolipid conformation for optimal recognition of Alzheimer’s b amyloid peptide (Ab1-40). PLoS ONE 5, e9079.

104


Yamashita, T., 1999. A vital role for glycosphingolipid synthesis during development and differentiation. Proc. Natl. Acad. Sci. USA 96, 9142–9147. Yamashita, T., 2014. Properties of a hydrated excess proton near the cholesterolcontaining phospholipid bilayer. JPS Conf. Proc. 1, 013086. Yesylevskyy, S.O., Demchenko, A.P., 2012. How cholesterol is distributed between monolayers in asymmetric lipid membranes. Eur. Biophys. J. 41, 1043–1054. Yesylevskyy, S.O., Ramseyer, C., 2014. Determination of mean and Gaussian curvatures of highly curved asymmetric lipid bilayers: the case study of the influence of cholesterol on the membrane shape. Phys. Chem. Chem. Phys. 16, 17052–17061. Yethiraj, T., Weisshaar, 2007. Why are lipid rafts not observed in vivo? Biophys. J. 93, 3113–3119. Zaraiskaya, Jeffrey, K.R., 2005. Molecular dynamics simulations and 2H NMR study of the GalCer/DPPG lipid bilayer. Biophys. J. 88, 4017–4031. Zhang, J., Zhang, K., Gao, Z.G., Paoletta, S., Zhang, D., Han, G.W., Li, T., Ma, L., Zhang, W., Muller, C.E., Yang, H., Jiang, H., Cherezov, V., Katritch, V., Jacobson, K.A.,

Stevens, R.C., Wu, B., Zhao, Q., 2014b. Agonist-bound structure of the human P2Y12 receptor. Nature 509, 119–122. Zhang, K., Zhang, J., Gao, Z.-G., Zhang, D., Zhu, L., Han, G.W., Moss, S.M., Paoletta, S., Kiselev, E., Lu, W., Fenalti, G., Zhang, W., Muller, C.E., Yang, H., Jiang, H., Cherezov, V., Katritch, V., Jacobson, K.A., Stevens, R.C., Wu, B., Zhao, Q., 2014a. Structure of the human P2Y12 receptor in complex with an antithrombotic drug. Nature 509, 115–118. Zhang, Z., Bhide, S.Y., Berkowitz, M.L., 2007. Molecular dynamics simulations of bilayers containing mixtures of sphingomyelin with cholesterol and phosphatidylcholine with cholesterol. J. Phys. Chem. B 111, 12888–12897. Zhao, G., Subbaiah, P.V., Mintzer, E., Chiu, S.-W., Jakobsson, E., Scott, H.L., 2011. Molecular dynamic simulation study of cholesterol and conjugated double bonds in lipid bilayers. Chem. Phys. Lipids 164, 811–818. Zidar, J., Merzel, F., Hodoscek, M., Rebolj, K., Sepcic, K., Macek, P., Janezic, D., 2009. Liquid-ordered phase formation in cholesterol/sphingomyelin bilayers: allatom molecular dynamics simulations. J. Phys. Chem. B 113, 15795–15802.

What do We Know about Neonatal Cognition?

What do mind readers know and what do we know about mind readers?

Vaping and health: what do we know about e-cigarettes?

What Do We Know about Opioids and the Kidney?

What we (don't) know about what we know.

ACT and Recovery: What We Know About Their Compatibility.

What we do not know about ADHD… yet.

What do we need to know about your NHP?

What do we know by now about the genus Naegleria?

What do we know about implicit false-belief tracking?

What we know but do not understand about nidovirus helicases.

Chemotherapy-induced peripheral neuropathy: What do we know about mechanisms?

What do we know about canine osteosarcoma treatment? Review.

What do we know about the safe surgery checklist now?

What do we really know about amyotrophic lateral sclerosis?

Fungal Melanin: What do We Know About Structure?

What do we know about cancer? A bicentennial perspective.

What do we know about phytotherapy of benign prostatic hyperplasia?

What do we know about gliotransmitter release from astrocytes?

What do we know about pediatric renal microlithiasis?

What do we know about Late Onset Huntington's Disease?

Mental health interventions for traumatized asylum seekers and refugees: What do we know about their efficacy?

The AIDS virus. What we know and what we can do about it.

What we do not know about the role of colchicine in pericarditis in 2014.