• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of jbcAbout JBCASBMBSubmissionsSubscriptionsContactJBCThis Article
J Biol Chem. Apr 2, 2010; 285(14): 10519–10526.
Published online Feb 5, 2010. doi:  10.1074/jbc.M109.068890
PMCID: PMC2856259

Identification of Specific Lipid-binding Sites in Integral Membrane Proteins*An external file that holds a picture, illustration, etc.
Object name is sbox.jpg

Abstract

Protein-lipid interactions are increasingly recognized as central to the structure and function of membrane proteins. However, with the exception of simplified models, specific protein-lipid interactions are particularly difficult to highlight experimentally. Here, we used molecular dynamics simulations to identify a specific protein-lipid interaction in lactose permease, a prototypical model for transmembrane proteins. The interactions can be correlated with the functional dependence of the protein to specific lipid species. The technique is simple and widely applicable to other membrane proteins, and a variety of lipid matrices can be used.

Keywords: Lipid, Membrane Bilayer, Membrane Proteins, Molecular Dynamics, Phosphatidylethanolamine, Lactose Permease, Major Facilitator Superfamily, Protein-Lipid Interaction, Secondary Transporter, Specific Interaction

Introduction

The molecular mechanisms underlying the influence of the lipid environment on the function of membrane proteins remain unclear. Lipid composition is known to have a modulatory effect on membrane protein activity, and for a number of membrane proteins a clear correlation was found between protein activity and bulk properties of the membrane bilayer such as fluidity (1). Membrane proteins are anchored in their lipid environment through nonspecific protein-lipid interactions (2, 3), and different fluidic properties are therefore expected to play a modulatory role on membrane protein activity. However, there is also increasing evidence for specifically bound lipids that are necessary to achieve biological function (4). For example, the presence of phosphatidylethanolamine (PE)3 in the bilayer has been found to be essential for the structure and function of a number of transmembrane transporter proteins (5,10). Unfortunately, crystallographic evidence of such a dependence is exceptional. First, membrane protein structure remains rare, with the structures of less than 200 unique proteins deposited in the Protein Data Bank. Second, membrane protein purification and crystallization require the use of detergent, which tends to delipidate the target protein. Thus only a few examples of crystal structures show lipids, known to be important for function, specifically associated to the protein. The yeast cytochrome bc1 complex contains a number of bound phospholipids, which suggest specific roles both for the structural and functional integrity of the protein (11). The crystal structure of the Thermochromatium tepidum photosynthetic reaction center shows one PE and seven detergent molecules on its molecular surface (12). In the crystal structure of the Rhodobacter sphaeroides photoreaction center, a specific interaction with cardiolipin has been found (13). This binding site has subsequently been suggested to be a conserved feature of these reaction centers (14). More recently, a tightly bound cholesterol molecule was observed in the crystal structure of a protein G-coupled receptor (15), triggering the question of the general presence of nonannular cholesterol-binding sites for these proteins (16).

Molecular modeling and other computational studies are therefore playing an increasingly relevant role in the study of membrane proteins (17); these include the use of molecular dynamics simulation to study dynamic events and conformational pathways (18). The installment of membrane protein-targeted structural genomics and the rapid development of high throughput structural biology techniques are expected to result in a significant increase in the number of available membrane protein structures in the near future (19). But it is doubtful that these structures will show physiologically associated lipid molecules; the crystallization procedure typically delipidates the protein, and only the strongest bound lipid molecules may withstand the purification procedure and show sufficient electron density to be resolved at the atomic level. Molecular modeling may play an important role here. With the increase in computing power, molecular dynamics simulations are now routinely applied, and physics-based force fields are capable of treating a wide range of molecules and molecular interaction (20, 21).

In this study, we use molecular modeling and dynamics to characterize a specific protein-lipid interaction between lactose permease (LacY) and PE. LacY, which has been crystallized at a resolution of 2.95 Å (22), is a paradigm for the major facilitator superfamily (MFS) (23) that represents as much as 25% of all membrane transport proteins, with over 15,000 sequence members identified to date (24). Members of the MFS show a common architecture (25), and for a number of these transporters the presence of PE in the bilayer is required for proper structure and activity (26). By carefully positioning the protein in a selection of lipid matrices and performing molecular dynamics simulations, we identify a limited number of strong interactions between specific amino acids and individual lipid molecules. The strength of this technique lies in the fact that it leads to a molecular picture of the lipid-protein interactions involved. The only limitation is the number of membrane protein structures available with a sufficient resolution; a large number of lipid matrices in which proteins can be inserted are available, and the technology required to create new ones mimicking all kinds of lipid membrane compositions is available and operational (20).

MATERIALS AND METHODS

Modeling Setup

We take the LacY crystal structure with the highest available resolution (2.95 Å), representing wild-type LacY in neutral pH, that is open on the cytoplasmic side and therefore susceptible to gradient sensing and sugar transport (22). With lipid chain lengths near C18 often being the most favorable for function (27, 28), we employ POPC, POPE, and POPG bilayers that have been extensively used in MD simulations and for which parameter sets are available (29,31). The bilayer thickness of these bilayers (32) shows good agreement with the calculated hydrophobic thickness of LacY (33), minimizing the effect of hydrophobic mismatch. Previous simulations of LacY, investigating the mechanism of sugar transport, have also been performed in a POPE bilayer (34).

Lactose permease (Protein Data Bank code 2cfq) was reoriented with its principal axis aligned with the z axis, rotated 10° over the x axis, and placed in the center of an equilibrated POPE bilayer (30) with all waters removed and the x and y coordinates expanded by a factor of 4. This orientation was found to yield the best final alignment between phosphates and interfacial arginines. The entire system was subjected to 100 steps of steepest descent energy minimization, applying 105 kJ/nm2 position restraints on the protein non-hydrogen atoms. In subsequent iterations, the system was restored to the reference area per lipid by shrinking the lipid x and y coordinates by 2% and deleting all lipids that had their phosphorus atom at a distance closer than 6 Å to any Cα atom of the protein (35). Every iteration was accompanied by 100 steps of steepest descent, and after eight iterations the deflation was increased to 5% per iteration.

The original box size of the resulting system of LacY and 298 POPE lipids was expanded by 1 nm perpendicular to the bilayer surface (now ~10 × 10 × 7 nm3) and solvated with roughly 9000 water molecules. The atoms in the palmitoyl and oleoyl chains were given a van der Waals radius of 6 Å to avoid solvation of the hydrophobic bilayer core. The system was neutralized by adding chloride ions (36) and was subjected to 1000 steps of steepest descent energy minimization and 100 ps of MD using the weak position restraint (103 kJ/nm2 on the non-hydrogen protein atoms), followed by 10 ns of free MD.

Previous simulations of PE systems have reduced the ethanolamine partial charges to correspond to the lysine parameterization (37). To eliminate artificial association between the ethanolamine moiety and acidic side chains of the protein because of an overpolarization of the lipid partial charges, a new charge set was also developed by fitting atomic point charges to reproduce the electrostatic potential obtained from ab initio HF/6–31G* calculations (38), keeping symmetry considerations (39) and summing hydrogen charges onto the nonpolar carbons (40). For all purposes of this work, the results obtained with both charge sets were found to be similar, and all analyses presented henceforth were carried out using the latter charge set. Both charge sets are listed in Table 1.

TABLE 1
Lipid head group partial charges

Molecular dynamics simulations were performed with the Gromacs 3.3 package (41), using the Gromos96 43a2 force field (40), with Berger parameters for the lipid tails (42). A time step of 2 fs was employed. All systems were coupled to a temperature bath of T = 310 K with a coupling constant τT = 0.1 ps. Protein, lipids, and solvent (ions plus water) were coupled independently. Single point charge water was employed (43). Pressure was maintained using semi-isotropic pressure coupling (44) at p = 1 bar with a coupling constant of τp = 1.0 ps. van der Waals interactions were cut off at a distance of 1.4 nm, and electrostatic interactions were calculated with the particle mesh Ewald method (45), using fourth-order splining and a grid spacing of 0.12 nm. Equations of motion for the water molecules were solved analytically (46), and all covalent bonds in the system were constrained in the MD simulations (47).

The entire procedure of positioning, equilibration, and simulation was repeated using POPE lipids with monomethylated, dimethylated, and trimethylated ethanolamine (the trimethylation resulting in POPC) and in a bilayer consisting of POPG lipids. The POPG bilayer was obtained by mutating every POPE lipid into POPG as before (48). Methylation of the PE lipids was achieved by mutating a lipid amine hydrogen into methyl while at the same time modifying all other related parameters (bond lengths, angles, nonbonded parameters, etc.). Additional ab initio calculations had shown it not to be necessary to modify the partial charges for the hydrogen to methyl mutation, which corresponds to the way the original POPE parameters were derived from POPC (30).

Validation of Equilibration Procedure

The orientational evolution of the α-helices in the simulation is used to validate the equilibration procedure and overall structural stability of the system. LacY contains 12 transmembrane helices, two of which with a kink in the middle, and one additional short helix on the cytosolic side. Table 2 shows the residues defining the helices and their angle with the normal bilayer averaged over the four simulations with neutral lipids. Helical axes were calculated using a rotational least squares fitting procedure (49). It is known that TM helices may tilt or flex to match the hydrophobic thickness of a membrane (3, 50). We indeed observed this for the longer helices, e.g. tilt for TM1 and TM4 and flex for TM3 and TM8. The shorter TM helices may span the bilayer without a tilt, e.g. TM5 and TM9. The observed behavior corresponds to what has been found by NMR experiments and MD simulations of the individual LacY helices TM1 and TM5 (50). A secondary structure analysis shows all helices retain their helical structure throughout the simulations (data not shown). Also, no change in helix orientation (less than 10° range of change in angle) was found (with the exception of TM6 and TM3.2 that varied within a 20° range). In addition, the relatively short helix 7.c showed some degree of motion, with a deviation up to 20° around its average. TM6 is imperfect in the crystal structure, and we find it to fully adopt the α-helical structure in all simulations, most likely due to the stabilizing influence of the lipid environment. The same happens with TM3.2, which is located next to TM6. The average helical angles and standard deviations were not found to differ significantly when calculated over the full trajectory or just the second half of it. Taken together with the correspondence to initial values, we conclude that the positioning and equilibration procedure has had no destabilizing effect on the overall protein structure.

TABLE 2
Transmembrane helix angles

Detection of Lipid-mediated Salt Bridges

The presence of salt bridges and hydrogen bonds between LacY and lipids was determined using simple distance criteria. Residues involved in salt bridges and hydrogen bonds to individual lipids were determined using the criterion that at any point during the combined simulations the donor-acceptor distance fell below the threshold of 4 Å. The upper limit of 4 Å constitutes a weak interaction. By using the obtained residues, and the requirement that two of them must simultaneously be bound to the same lipid, lipid-mediated salt bridges were determined using a distance criterion of 4 Å of the residue phosphate (donor-phosphorus), 3 Å of residue amine or residue hydroxy (acceptor-donor), and 4 Å of residue choline (acceptor-nitrogen) distance. Cumulative presence (Δtcumul) of individual lipid-mediated salt bridges was calculated as combined fractional presence over the entire simulation time. The persistence factor F was calculated as shown in Equation 1,

equation image

where atom is either the lipid nitrogen, hydroxy oxygen, or phosphorus atom; MSF is the mean square fluctuation, and MSFmin and MSFmax equal 0.156 and 0.653 nm2, respectively. Lipid-mediated salt bridges were retained if they showed Δtcumul > 10.0%.

Residue Conservation

LacY sequences were obtained by querying the NCBI nonredundant sequence data base of January 22, 2009, with BLAST (51), using the BLOSUM62 matrix and default parameters, and the Escherichia coli LacY sequence as query. All sequences with an E-value better than 1.0e-10 were retained and aligned using MUSCLE (52), resulting in 92 unique sequences with the following (selected) keyword occurrences: permease, 90; oligosaccharide/H+, 54; galactoside, 39; hypothetical protein, 38; lactose, 31; transport, 21; symporter,: 20; and lactose-proton, 20.

RESULTS

Interactions between protein and lipids can manifest themselves in a number of ways as follows: lipid acyl chains can settle on a hydrophobic surface patch created by one or two TM helices; the lipid carbonyl or phosphate groups can act as acceptor for hydrogen bonds emanating from the protein; and salt bridges may be formed between opposing charges, e.g. phosphate and arginine or lysine. Hydrophobic interactions are the weakest of these, and the strongest interactions are made by salt bridges, which are in fact a particularly strong form of a hydrogen bond.

In the following sections we analyze the occurrence of individual and double hydrogen bonds and salt bridges. We then identify specific protein-lipid interactions from the presence of lipid-mediated salt bridges. The strongest of these is validated by additional simulations in lipid matrices with increasing degrees of methylation and correlated with experimental data and sequence conservation in the family and superfamily.

Hydrophobic Interactions and Single Hydrogen Bonds

Annular lipids were found to show limited diffusion (as calculated from the mean square deviation of lipid tails), in accordance with spectroscopic data (53), but no clear transition in distribution between annular and bulk lipids was observed. We could therefore not identify any single lipid as being associated with the protein on the basis of spatial fluctuation and hydrophobic interaction.

LacY contains 17 negatively and 24 positively charged residues plus a number of otherwise hydrogen bond-capable residues; all these are potential candidates for protein-lipid-specific interactions. We determined for all hydrogen bond-capable residues of the protein whether they were at one time or another favorably bonded to a hydrogen bond-complementary (but not necessarily opposingly charged) lipid head group. For many of the candidate residues, a large population of distances indicative of binding was found, but it is impossible to distinguish between specific and nonspecific protein-lipid interactions. As expected, the strongest interactions observed were those involving negatively (Asp and Glu) or positively charged residues (Arg and Lys), the latter of which are responsible for the anchoring of a membrane protein in its environment (3). Interactions through the phosphate group are not lipid head group-specific, because they do not discriminate between different types of phospholipids. Also, single salt bridges and hydrogen bonds have a limited lifetime, with fluctuations on a 1–5-ns time scale (2). Lipids showing interactions through the amine, hydroxy, or choline group would therefore show too fast an exchange rate with bulk lipids to call them specific.

Lipid-mediated Salt Bridges

Lipids that are simultaneously bound to two different residues, e.g. through both their positively and negatively charged moieties, will show significantly longer residence times, and their diffusion away requires a two-step process. We therefore determined for every salt bridge and hydrogen bond, whether the same lipid was simultaneously bound to another salt bridge or hydrogen bond. Although cation-π interactions can be described using the current force field (31), and the PE amine group would be prone to such interactions, we have found no significant lipid-mediated salt bridges involving the π-clouds of aromatic residues. The list of lipid-mediated salt bridges is supplied in Table 3. As additional requirement, we impose the presence of at least 10% of the simulation time (amounting to 1 ns), resulting in a total of nine lipid-mediated salt bridges (listed in Table 4).

TABLE 3
Lipid-mediated salt bridges
TABLE 4
Lipid-bridged residue-residue contacts

Of the nine lipid-mediated salt bridges, five, including the three strongest, showing a Δtcumul >15%, involve both an acidic and basic residue, and four of these occur in the PE bilayer (the fifth one with PC). Two bridges are found involving a PC lipid, and only one with PG. Most lipid-mediated salt bridges show similarly strong interactions, with persistence factor values lying between 5 and 10. An exception is the bridge involving Glu-215 and Lys-211, which shows a weaker interaction on the amine side, with Fdonor = 2.9, related to an increased local fluctuation (Table 3). Unmistakably, however, the interactions with Asp-68 and Lys-69, two neighboring residues, stand out, with significantly larger persistence factors. In addition, this lipid-mediated salt bridge is weaker in the case of PC and may therefore be responsible for PE preference. We validate the results by a closer investigation of this bridge, both in terms of the molecular details of the simulation and sequence conservation in the lactose permease family and major facilitator superfamily.

Binding of PE to Asp-68 and Lys-68 in Lipid Matrices with Increasing Degrees of Methylation

In the independent studies of two MFS transporters, including LacY (54), transporter activity was investigated using PE lipids with various degrees of methylation. In both cases, a triple methylation (i.e. PC lipids) abolished activity completely, indicating a resident hydrogen on the ethanolamine moiety of PE to be essential (9, 54). We performed two additional simulations of LacY in lipid bilayer systems composed of mono- and dimethylated POPE lipids and analyzed the four simulations together to provide a molecular basis for PE specificity.

Fig. 1 (E and F) shows binding of Lys-69 to PE to occur already in the equilibration phase. Weak position restraints keep the protein heavy atoms in place, but the lipid is free to move, and the phosphate group of PE is gradually moving closer to Lys-69 to form a salt bridge at the final stages of energy minimization (POPE) or the first step of position restraint MD (Me-POPE, dimethyl-POPE, and POPC). The bridge is maintained throughout the remainder of the simulations, showing no significant difference in average distance.

FIGURE 1.
Distances of Asp-68 and Lys-69 to the lipid. Distance of Asp-68 Oδ atoms to lipid N (A–D) and Lys-69 Nζ to lipid P (E–H) for the last steps of the equilibration (A, B, E, and F) and for the MD simulation (C, D, G, and ...

The distance evolution between the lipid and Asp-68 is shown in Fig. 1 (A–D). Although a few lipids were found to bind to Asp-68 during the course of the simulations, in all four simulations a single lipid is continuously bound to Lys-69, and in every instance this is the same lipid that ultimately binds Asp-68. In the crystal and initial structure, Asp-68 is directed toward the inside of the protein and is hydrogen-bonded to Lys-131. It is remarkable that POPC shows a movement of the choline group away from Asp-68, and all three PE simulations an approach of the amine moiety to Asp-68 (Fig. 1B). After the side chain is released, the salt bridge is formed in the initial steps of the free MD simulation for POPE and monomethylated POPE. Data for the full simulation (Fig. 1D) shows binding to occur progressively later upon increased methylation. Fig. 1D shows initial binding of one of the two amine methyl groups at t = 1.5 ns, after which a rotation of the dimethylated amine group at t = 2 ns rotates the single hydrogen toward Asp-68. Binding of PC to Asp-68 (Fig. 1D) occurs only at t = 2.4 ns, and at an equilibrium distance more than 1 Å larger (0.37 versus 0.25 nm), making the rather significant difference between strong and weak hydrogen bonding.

Our observation is that a phospholipid (PE or PC) is recruited by Lys-69, followed by a rotation of the Asp-68 side chain to bind its amine or choline group, thus forming a double salt bridge or lipid-mediated salt bridge. Fig. 2 shows an image of this lipid-mediated salt bridge.

FIGURE 2.
Final structure of the POPE simulation showing the lipid-mediated salt bridge between Asp-68 and Lys-69. Distances indicated are between the closest hydrogen and oxygen atoms. Only amine hydrogens are shown.

Sequence Conservation in LacY and the MFS

The wide variety of substrates of MFS transporters is reflected in the large sequence variation going from one subfamily to the other, but all members of the MFS share a highly conserved cytoplasmic motif (55) that includes the acidic Asp-68 and the basic Lys-69, Arg-73, and Lys-74. Individual mutations have shown Asp-68, but also a resident positive charge, to be required for activity (55, 56). In addition, Asp-68 was found to be related to the proton gradient-sensing mechanism but not to substrate transport (9, 56, 57). Our strategy for detecting lipid-mediated salt bridges directly and unambiguously identifies Asp-68 as the most relevant PE-interacting residue. This is the first study where the PE dependence of LacY (5) and the functional importance of Asp-68 (55) can be directly linked together.

For the residues involved in the three strongest lipid-mediated salt bridges of Table 4, residue conservation in lactose permease is listed in boldface in Table 5. As opposed to the MFS, the multiple sequence alignment of (putative) lactose permease sequences is less restrictive and still shows Asp-68 and Lys-69 to remain the only such bridge between two conserved residues. The other lipid-mediated salt bridges are likely to be only phosphate-specific, due to the limited conservation of the acceptor residues involved, e.g. Asp-190 and Glu-215. Arg-73 and Lys-74 are located toward the periphery of the protein, at a small distance from Asp-68/Lys-69, and could therefore be involved in the initial recruitment of PE, shuttling a suitable phospholipid toward Asp-68 by the use of its flexible lysine side chain. In most LacY family members, a lysine is in fact found at position 73. Snorkeling actions of the lysine side chain are not an uncommon phenomenon (2, 3) and could in this case be related to lipid recruitment.

TABLE 5
Residue conservation

DISCUSSION

Specific Interaction between Asp-68 and PE That Contributes to the Gradient-sensing Mechanism

Using molecular modeling and dynamics, we identify a specific protein-lipid interaction in LacY between Asp-68 and PE by investigating the existence of lipid-mediated salt bridges of LacY embedded in five different lipid matrices. Our simulations show a consistent and strong hydrogen bond between (non-, mono-, and dimethylated) POPE amine groups and Asp-68 that is significantly weaker in the case of PC. In every instance, the bond is formed to a free hydrogen of the amine group with the speed of formation being inversely related to the degree of methylation. Asp-68-bound PE is initially bound to Lys-69 through its phosphate entity before being recruited by Asp-68. The nearby residue Lys-74 may have a facilitating function in the recruitment of PE. This lipid-mediated salt bridge is the only such bridge existing between conserved residues in LacY.

Asp-68 and Lys-69 are also part of a highly conserved motif in the MFS. Together with the PE dependence of MFS transporters (26) and the involvement of Asp-68 in the energy-coupling mechanism (9, 56, 57), the interaction between Asp-68 and PE may constitute a specific protein-lipid interaction involved in ΔpH sensing. This suggests that not only secondary and tertiary structure elements (25) but also functional elements such as the proton gradient-sensing mechanism have been conserved in secondary transporters during evolution. The fact that Lys-69 has not been found to be crucial for transport is in accord with our findings; only a single resident positive charge is required in the conserved motif (55), and only a single positively charged residue in principle suffices to recruit a phospholipid that can subsequently be shuttled toward Asp-68.

Role of Alternative Lipids

Although PE is reportedly required for proper LacY topogenesis and functioning (5, 58), it has been proposed that it can be replaced by a nonendogenous lipid, monoglucosyl-diacylglycerol (GlcDAG), a lipid found in Acholeplasma laidlawii (59). This may seem incompatible with the hypothesis proposed here considering the structural differences between PE and GlcDAG. We performed docking studies of the glucosyl moiety to identify binding modes with the LacY pocket. Surprisingly, the glucosyl can interact with both Asp-68 and Lys-69 in a PE-like fashion (supplemental Fig. S1). In the absence of the phosphate group, Lys-69 now interacts with the O4 of the sugar ring, whereas Asp-68 interacts with O3, leading to a motif conformation very similar to the one observed in the POPE simulation. In addition, O2 is found to be hydrogen-bonded to the backbone NH of Lys-69. Note that in diglucosyl-diacylglycerol (DGlcDAG), the O2 atom is involved in the connection between the two sugar rings and would therefore not be available for binding, although binding of the second sugar ring to the motif would be prohibited by the bulkiness of DGlcDAG as a whole. Although full-fledged MD simulations of LacY in GlcDAG and DGlcDAG bilayers should be done (a technically challenging process that goes beyond the scope of this paper), these results provide an explanation on why GlcDAG appears to allow LacY function and not DGlcDAG (64). In contrast, and in accord with experimental data (54), the simulations of LacY in a POPG bilayer show no affinity of the glycerol side chain to the acidic Asp-68 (Table 1), and it is found to bind the phosphate groups of itself and neighboring lipids.

Implications of Results on the Functioning of Membrane Proteins

Our findings show that specific protein-lipid interactions may be identified through the identification of simultaneous occurrence of multiple (strong) interactions involving the same lipid. The occurrence of such interactions provides further evidence to the fact that specific protein-lipid interactions indeed play a role in the functioning of membrane proteins (4, 10). The significance of membrane proteins that constitute ~25% of genomic sequences (19) is not reflected in the number of membrane structures that are currently available in the Protein Data Bank. Yet membrane proteins represent 70% of current drug targets (60). Specific lipid-binding residues may therefore also present a yet unexplored area for drug targeting.

This is the first computational study ever to identify a specific protein-lipid interaction and provide the molecular basis for lipid species specificity at the same time. The technique applied here can be used on any membrane protein structure and embedded in a variety of lipid matrices, including mixtures with lipid species that do not form bilayer structures on themselves like cholesterol and cardiolipin. The approach is a thorough but simple application of molecular modeling and dynamics, a tried and tested technique.

Supplementary Material

Supplemental Data:

*The Genome and Network Bioinformatics Laboratory is a partner of the BioSapiens Network of Excellence funded under the Sixth Framework Program of the European Commission (LSHG-CT-2003-503265).

An external file that holds a picture, illustration, etc.
Object name is sbox.jpgThe on-line version of this article (available at http://www.jbc.org) contains supplemental Fig. S1.

3The abbreviations used are:

PE
phosphatidylethanolamine
MFS
major facilitator superfamily
POPE
1-palmitoyl-2-oleoyl-l-phosphatidylethanolamine
POPC
1-palmitoyl-2-oleoyl-l-phosphatidylcholine
POPG
1-palmitoyl-2-oleoyl-l-phosphatidylglycerol
PC
phosphatidylcholine
MD
molecular dynamics
GlcDAG
monoglucosyl-diacylglycerol
DGlcDAG
diglucosyl-diacylglycerol.

REFERENCES

1. Vigh L., Escribá P. V., Sonnleitner A., Sonnleitner M., Piotto S., Maresca B., Horváth I., Harwood J. L. (2005) Prog. Lipid Res. 44, 303–344 [PubMed]
2. Deol S. S., Bond P. J., Domene C., Sansom M. S. (2004) Biophys. J. 87, 3737–3749 [PMC free article] [PubMed]
3. Nyholm T. K., Ozdirekcan S., Killian J. A. (2007) Biochemistry 46, 1457–1465 [PubMed]
4. Lee A. G. (2005) Mol. Biosyst. 1, 203–212 [PubMed]
5. Bogdanov M., Dowhan W. (1995) J. Biol. Chem. 270, 732–739 [PubMed]
6. Wang X., Bogdanov M., Dowhan W. (2002) EMBO J. 21, 5673–5681 [PMC free article] [PubMed]
7. Zhang W., Bogdanov M., Pi J., Pittard A. J., Dowhan W. (2003) J. Biol. Chem. 278, 50128–50135 [PubMed]
8. Zhang W., Campbell H. A., King S. C., Dowhan W. (2005) J. Biol. Chem. 280, 26032–26038 [PubMed]
9. Hakizimana P., Masureel M., Gbaguidi B., Ruysschaert J. M., Govaerts C. (2008) J. Biol. Chem. 283, 9369–9376 [PubMed]
10. Powl A. M., East J. M., Lee A. G. (2008) Biochemistry 47, 12175–12184 [PubMed]
11. Lange C., Nett J. H., Trumpower B. L., Hunte C. (2001) EMBO J. 20, 6591–6600 [PMC free article] [PubMed]
12. Nogi T., Fathir I., Kobayashi M., Nozawa T., Miki K. (2000) Proc. Natl. Acad. Sci. U.S.A. 97, 13561–13566 [PMC free article] [PubMed]
13. McAuley K. E., Fyfe P. K., Ridge J. P., Isaacs N. W., Cogdell R. J., Jones M. R. (1999) Proc. Natl. Acad. Sci. U.S.A. 96, 14706–14711 [PMC free article] [PubMed]
14. Wakeham M. C., Sessions R. B., Jones M. R., Fyfe P. K. (2001) Biophys. J. 80, 1395–1405 [PMC free article] [PubMed]
15. Cherezov V., Rosenbaum D. M., Hanson M. A., Rasmussen S. G., Thian F. S., Kobilka T. S., Choi H. J., Kuhn P., Weis W. I., Kobilka B. K., Stevens R. C. (2007) Science 318, 1258–1265 [PMC free article] [PubMed]
16. Paila Y. D., Tiwari S., Chattopadhyay A. (2009) Biochim. Biophys. Acta 1788, 295–302 [PubMed]
17. Elofsson A., von Heijne G. (2007) Annu. Rev. Biochem. 76, 125–140 [PubMed]
18. Aksimentiev A., Brunner R., Cohen J., Comer J., Cruz-Chu E., Hardy D., Rajan A., Shih A., Sigalov G., Yin Y., Schulten K. (2008) Methods Mol. Biol. 474, 181–234 [PubMed]
19. Carpenter E. P., Beis K., Cameron A. D., Iwata S. (2008) Curr. Opin. Struct. Biol. 18, 581–586 [PMC free article] [PubMed]
20. Guvench O., MacKerell A. D., Jr. (2008) Methods Mol. Biol. 443, 63–88 [PubMed]
21. Klepeis J. L., Lindorff-Larsen K., Dror R. O., Shaw D. E. (2009) Curr. Opin. Struct. Biol. 19, 120–127 [PubMed]
22. Guan L., Mirza O., Verner G., Iwata S., Kaback H. R. (2007) Proc. Natl. Acad. Sci. U.S.A. 104, 15294–15298 [PMC free article] [PubMed]
23. Kaback H. R. (2005) C. R. Biol. 328, 557–567 [PubMed]
24. Saier M. H., Jr. (2000) Microbiology 146, 1775–1795 [PubMed]
25. Law C. J., Maloney P. C., Wang D. N. (2008) Annu. Rev. Microbiol. 62, 289–305 [PMC free article] [PubMed]
26. Dowhan W., Bogdanov M. (2009) Annu. Rev. Biochem. 78, 515–540 [PMC free article] [PubMed]
27. Starling A. P., East J. M., Lee A. G. (1995) Biochem. J. 310, 875–879 [PMC free article] [PubMed]
28. Yuan C., O'Connell R. J., Feinberg-Zadek P. L., Johnston L. J., Treistman S. N. (2004) Biophys. J. 86, 3620–3633 [PMC free article] [PubMed]
29. Tieleman D. P., Berendsen H. J. (1996) J. Chem. Phys. 105, 4871–4880
30. Tieleman D. P., Berendsen H. J. (1998) Biophys. J. 74, 2786–2801 [PMC free article] [PubMed]
31. Lensink M. F., Christiaens B., Vandekerckhove J., Prochiantz A., Rosseneu M. (2005) Biophys. J. 88, 939–952 [PMC free article] [PubMed]
32. Sprong H., van der Sluijs P., van Meer G. (2001) Nat. Rev. Mol. Cell Biol. 2, 504–513 [PubMed]
33. Lomize M. A., Lomize A. L., Pogozheva I. D., Mosberg H. I. (2006) Bioinformatics 22, 623–625 [PubMed]
34. Yin Y., Jensen M. Ø., Tajkhorshid E., Schulten K. (2006) Biophys. J. 91, 3972–3985 [PMC free article] [PubMed]
35. Kandt C., Ash W. L., Tieleman D. P. (2007) Methods 41, 475–488 [PubMed]
36. Chandrasekhar J., Spellmeyer D. C., Jorgensen W. L. (1984) J. Am. Chem. Soc. 106, 903–910
37. Marrink S. J., Mark A. E. (2002) Biochemistry 41, 5375–5382 [PubMed]
38. Guest M. F., Bush I. J., van Dam H. J., Sherwood P., Thomas J. M., van Lenthe J. H., Havenith R. W., Kendrick J. (2005) Mol. Phys. 103, 719–747
39. Dupradeau F. Y., Cézard C., Lelong R., Stanislawiak E., Pêcher J., Delepine J. C., Cieplak P. (2008) Nucleic Acids Res. 36, D360–D367 [PMC free article] [PubMed]
40. Van Gunsteren W. F., Billeter S. R., Eising A. A., Hünenberger P. H., Krüger P., Mark A. E., Scott W. R., Tironi I. G. (1996) Biomolecular Simulation: The GROMOS96 Manual and User Guide, Hochschulverlag AG an der ETH Zürich, Zürich, Switzerland
41. Van Der Spoel D., Lindahl E., Hess B., Groenhof G., Mark A. E., Berendsen H. J. C. (2005) J. Comput. Chem. 26, 1701–1718 [PubMed]
42. Berger O., Edholm O., Jähnig F. (1997) Biophys. J. 72, 2002–2013 [PMC free article] [PubMed]
43. Berendsen H. J., Postma J. P., Van Gunsteren W. F., Hermans J. (1981) in Intermolecular Forces ( Pullman B., editor. ed) pp. 331–342, Reidel, Dordrecht, The Netherlands
44. Berendsen H. J., Postma J. P., DiNola A., Haak J. R. (1984) J. Chem. Phys. 81, 3684–3690
45. Essman U., Perela L., Berkowitz M. L., Darden T., Lee H., Pedersen L. G. (1995) J. Chem. Phys. 103, 8577–8592
46. Miyamoto S., Kollman P. A. (1992) J. Comput. Chem. 13, 952–962
47. Hess B., Bekker H., Berendsen H. J., Fraaije J. G. (1997) J. Comput. Chem. 18, 1463–1473
48. Lensink M. F. (2008) Methods Mol. Biol. 443, 161–179 [PubMed]
49. Christopher J. A., Swanson R., Baldwin T. O. (1996) Comput. Chem. 20, 339–345 [PubMed]
50. Yeagle P. L., Bennett M., Lemaître V., Watts A. (2007) Biochim. Biophys. Acta 1768, 530–537 [PubMed]
51. Altschul S. F., Madden T. L., Schäffer A. A., Zhang J., Zhang Z., Miller W., Lipman D. J. (1997) Nucleic Acids Res. 25, 3389–3402 [PMC free article] [PubMed]
52. Edgar R. C. (2004) Nucleic Acids Res. 32, 1792–1797 [PMC free article] [PubMed]
53. Devaux P. F., Seigneuret M. (1985) Biochim. Biophys. Acta 822, 63–125 [PubMed]
54. Chen C. C., Wilson T. H. (1984) J. Biol. Chem. 259, 10150–10158 [PubMed]
55. Jessen-Marshall A. E., Paul N. J., Brooker R. J. (1995) J. Biol. Chem. 270, 16251–16257 [PubMed]
56. Yamaguchi A., Ono N., Akasaka T., Noumi T., Sawai T. (1990) J. Biol. Chem. 265, 15525–15530 [PubMed]
57. Mazurkiewicz P., Poelarends G. J., Driessen A. J., Konings W. N. (2004) J. Biol. Chem. 279, 103–108 [PubMed]
58. Bogdanov M., Dowhan W. (1998) EMBO J. 17, 5255–5264 [PMC free article] [PubMed]
59. Xie J., Bogdanov M., Heacock P., Dowhan W. (2006) J. Biol. Chem. 281, 19172–19178 [PMC free article] [PubMed]
60. Lundstrom K. (2007) J. Cell. Mol. Med. 11, 224–238 [PMC free article] [PubMed]
61. Chiu S. W., Clark M., Balaji V., Subramaniam S., Scott H. L., Jakobsson E. (1995) Biophys. J. 69, 1230–1245 [PMC free article] [PubMed]
62. Mulliken R. S. (1955) J. Chem. Phys. 23, 1833–1840
63. Bayly C. I., Cieplak P., Cornell W. D., Kollman P. A. (1993) J. Phys. Chem. 97, 10269–10280
64. Wikström M., Kelly A. A., Georgiev A., Eriksson H. M., Klement M. R., Bogdanov M., Dowhan W., Wieslander A. (2009) J. Biol. Chem. 284, 954–965 [PMC free article] [PubMed]

Articles from The Journal of Biological Chemistry are provided here courtesy of American Society for Biochemistry and Molecular Biology
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...