Simulation studies on bacteriorhodopsin bundle of transmembrane alpha segments

Bacteriorhodopsin (BR) is a membrane protein which pumps protons through the plasma membrane. Seven transmembrane BR helical segments are subjected to simulation studies in order to investigate the packing process of transmembrane helices. A Monte Carlo simulated annealing protocol is employed to optimize the helix bundle system. Helix packing is optimized according to a semi-empirical potential mainly composed of six components: a bilayer potential, a crossing angle potential, a helix dipole potential, a helix-helix distance potential, a helix orientation potential and a helix-helix distance restraint potential (a loop potential). Necessary parameters are derived from theoretical studies and statistical analysis of experimentally determined protein structures. The structures from the simulations are compared with the experimentally determined structures in terms of geometry. The structures generated show similar shapes to the experimentally suggested structure even without the helix-helix distance restraint potential. However, the relative locations of individual helices were reproduced only when the helix-helix distance restraint potential was used with restraint conditions. Our results suggest that transmembrane helix bundles resembling those observed experimentally may be generated by simulations using simple potentials.

Simulation studies on bacteriorhodopsin alpha-helices

Bacteriorhodopsin (BR) is a membrane protein which pumps protons through the plasma membrane. Transmembrane BR helical segments are subjected to simulation studies in order to investigate the effect of bilayer environment in various simulation conditions. A bilayer potential is introduced to the system to mimic the lipid membrane. The structures from the simulations are compared with the experimentally determined structures in terms of geometrical properties. Electrostatic contribution to the helix packing is also investigated. The simulation results show that the packing geometry of the transmembrane helices is highly affected by the bilayer potential. The results obtained from the simulations may be used for further simulation studies and analysis in investigating transmembrane helix packing.

Structure and dynamics of the pore-lining helix of the nicotinic receptor: MD simulations in water, lipid bilayers, and transbilayer bundles

Multiple nanosecond duration molecular dynamics simulations on the pore-lining M2 helix of the nicotinic acetylcholine receptor reveal how its structure and dynamics change as a function of environment. In water, the M2 helix partially unfolds to form a molecular hinge in the vicinity of a central Leu residue that has been implicated in the mechanism of ion channel gating. In a phospholipid bilayer, either as a single transmembrane helix, or as part of a pentameric helix bundle, the M2 helix shows less flexibility, but still exhibits a kink in the vicinity of the central Leu. The single M2 helix tilts relative to the bilayer normal by 12 degrees, in agreement with recent solid state NMR data (Opella et al., Nat Struct Biol 6:374-379, 1999). The pentameric helix bundle, a model for the pore domain of the nicotinic receptor and for channels formed by M2 peptides in a bilayer, is remarkably stable over a 2-ns MD simulation in a bilayer, provided one adjusts the pK(A)s of ionizable residues to their calculated values (when taking their environment into account) before starting the simulation. The resultant transbilayer pore shows fluctuations at either mouth which transiently close the channel. Proteins 2000;39:47-55.

Membrane simulations: bigger and better?

Molecular dynamics simulations of biological membranes have come of age. Simulations of pure lipid bilayers are extending our understanding of both optimal simulation procedures and the detailed structural dynamics of lipids in these systems. Simulation methods established using simple bilayer-embedded peptides are being extended to a wide range of membrane proteins and membrane protein models, and are beginning to reveal some of the complexities of membrane protein structural dynamics and their relationship to biological function.

Potassium channels: watching a voltage-sensor tilt and twist

The fourth transmembrane helix (S4) is the primary voltage-sensor of voltage-gated ion channels. Recent studies have used fluorescence resonance energy transfer as a spectroscopic ruler to determine the nature and magnitude of the voltage-induced movement of S4 that leads to channel opening.

Protonation of lysine residues inverts cation/anion selectivity in a model channel

A dimeric alamethicin analog with lysine at position 18 in the sequence (alm-K18) was previously shown to form stable anion-selective channels in membranes at pH 7.0 [Starostin, A. V., R. Butan, V. Borisenko, D. A. James, H. Wenschuh, M. S. Sansom, and G. A. Woolley. 1999. Biochemistry. 38:6144-6150]. To probe the charge state of the conducting channel and how this might influence cation versus anion selectivity, we performed a series of single-channel selectivity measurements at different pH values. At pH 7.0 and below, only anion-selective channels were found with P(K(+))/P(Cl(-)) = 0. 25. From pH 8-10, a mixture of anion-selective, non-selective, and cation-selective channels was found. At pH > 11 only cation-selective channels were found with P(K(+))/P(Cl(-)) = 4. In contrast, native alamethicin-Q18 channels (with Gln in place of Lys at position 18) were cation-selective (P(K(+))/P(Cl(-)) = 4) at all pH values. Continuum electrostatics calculations were then carried out using an octameric model of the alm-K18 channel embedded in a low dielectric slab to simulate a membrane. Although the calculations can account for the apparent pK(a) of the channel, they fail to correctly predict the degree of selectivity. Although a switch from cation- to anion-selectivity as the channel becomes protonated is indicated, the degree of anion-selectivity is severely overestimated, suggesting that the continuum approach does not adequately represent some aspect of the electrostatics of permeation in these channels. Side-chain conformational changes upon protonation, conformational changes, and deprotonation caused by permeating cations and counterion binding by lysine residues upon protonation are considered as possible sources of the overestimation.

Simulations of ion permeation through a potassium channel: molecular dynamics of KcsA in a phospholipid bilayer

Potassium channels enable K(+) ions to move passively across biological membranes. Multiple nanosecond-duration molecular dynamics simulations (total simulation time 5 ns) of a bacterial potassium channel (KcsA) embedded in a phospholipid bilayer reveal motions of ions, water, and protein. Comparison of simulations with and without K(+) ions indicate that the absence of ions destabilizes the structure of the selectivity filter. Within the selectivity filter, K(+) ions interact with the backbone (carbonyl) oxygens, and with the side-chain oxygen of T75. Concerted single-file motions of water molecules and K(+) ions within the selectivity filter of the channel occur on a 100-ps time scale. In a simulation with three K(+) ions (initially two in the filter and one in the cavity), the ion within the central cavity leaves the channel via its intracellular mouth after approximately 900 ps; within the cavity this ion interacts with the Ogamma atoms of two T107 side chains, revealing a favorable site within the otherwise hydrophobically lined cavity. Exit of this ion from the channel is enabled by a transient increase in the diameter of the intracellular mouth. Such "breathing" motions may form the molecular basis of channel gating.

Exploring models of the influenza A M2 channel: MD simulations in a phospholipid bilayer

The M2 protein of influenza A virus forms homotetrameric helix bundles, which function as proton-selective channels. The native form of the protein is 97 residues long, although peptides representing the transmembrane section display ion channel activity, which (like the native channel) is blocked by the antiviral drug amantadine. As a small ion channel, M2 may provide useful insights into more complex channel systems. Models of tetrameric bundles of helices containing either 18 or 22 residues have been simulated while embedded in a fully hydrated 1-palmitoyl-2-oleoyl-sn-glycerol-3-phosphatidylcholine bilayer. Several different starting models have been used. These suggest that the simulation results, at least on a nanosecond time scale, are sensitive to the exact starting structure. Electrostatics calculations carried out on a ring of four ionizable aspartate residues at the N-terminal mouth of the channel suggest that at any one time, only one will be in a charged state. Helix bundle models were mostly stable over the duration of the simulation, and their helices remained tilted relative to the bilayer normal. The M2 helix bundles form closed channels that undergo breathing motions, alternating between a tetramer and a dimer-of-dimers structure. Under these conditions either the channel forms a pocket of trapped waters or it contains a column of waters broken predominantly at the C-terminal mouth of the pore. These waters exhibit restricted motion in the pore and are effectively "frozen" in a way similar to those seen in previous simulations of a proton channel formed by a four-helix bundle of a synthetic leucine-serine peptide (, Biophys. J. 77:2400-2410).

Structure and dynamics of K channel pore-lining helices: a comparative simulation study

Isolated pore-lining helices derived from three types of K-channel have been analyzed in terms of their structural and dynamic features in nanosecond molecular dynamics (MD) simulations while spanning a lipid bilayer. The helices were 1) M1 and M2 from the bacterial channel KcsA (Streptomyces lividans), 2) S5 and S6 from the voltage-gated (Kv) channel Shaker (Drosophila melanogaster), and 3) M1 and M2 from the inward rectifier channel Kir6.2 (human). In the case of the Kv and Kir channels, for which x-ray structures are not known, both short and long models of each helix were considered. Each helix was incorporated into a lipid bilayer containing 127 palmitoyloleoylphosphatidylcholine molecules, which was solvated with approximately 4000 water molecules, yielding approximately 20, 000 atoms in each system. Nanosecond MD simulations were used to aid the definition of optimal lengths for the helix models from Kv and Kir. Thus the study corresponds to a total simulation time of 10 ns. The inner pore-lining helices (M2 in KcsA and Kir, S6 in Shaker) appear to be slightly more flexible than the outer pore-lining helices. In particular, the Pro-Val-Pro motif of S6 results in flexibility about a molecular hinge, as was suggested by previous in vacuo simulations (, Biopolymers. 39:503-515). Such flexibility may be related to gating in the corresponding intact channel protein molecules. Analysis of H-bonds revealed interactions with both water and lipid molecules in the water/bilayer interfacial region. Such H-bonding interactions may lock the helices in place in the bilayer during the folding of the channel protein (as is implicit in the two-stage model of membrane protein folding). Aromatic residues at the extremities of the helices underwent complex motions on both short (<10 ps) and long (>100 ps) time scales.

Simulation studies of the interaction of antimicrobial peptides and lipid bilayers

Experimental studies of a number of antimicrobial peptides are sufficiently detailed to allow computer simulations to make a significant contribution to understanding their mechanisms of action at an atomic level. In this review we focus on simulation studies of alamethicin, melittin, dermaseptin and related antimicrobial, membrane-active peptides. All of these peptides form amphipathic alpha-helices. Simulations allow us to explore the interactions of such peptides with lipid bilayers, and to understand the effects of such interactions on the conformational dynamics of the peptides. Mean field methods employ an empirical energy function, such as a simple hydrophobicity potential, to provide an approximation to the membrane. Mean field approaches allow us to predict the optimal orientation of a peptide helix relative to a bilayer. Molecular dynamics simulations that include an atomistic model of the bilayer and surrounding solvent provide a more detailed insight into peptide-bilayer interactions. In the case of alamethicin, all-atom simulations have allowed us to explore several steps along the route from binding to the membrane surface to formation of transbilayer ion channels. For those antimicrobial peptides such as dermaseptin which prefer to remain at the surface of a bilayer, molecular dynamics simulations allow us to explore the favourable interactions between the peptide helix sidechains and the phospholipid headgroups.

Molecular dynamics of synthetic leucine-serine ion channels in a phospholipid membrane

Molecular dynamics calculations were carried out on models of two synthetic leucine-serine ion channels: a tetrameric bundle with sequence (LSLLLSL)(3)NH(2) and a hexameric bundle with sequence (LSSLLSL)(3)NH(2). Each protein bundle is inserted in a palmitoyloleoylphosphatidylcholine bilayer membrane and solvated by simple point charge water molecules inside the pore and at both mouths. Both systems appear to be stable in the absence of an electric field during the 4 ns of molecular dynamics simulation. The water motion in the narrow pore of the four-helix bundle is highly restricted and may provide suitable conditions for proton transfer via a water wire mechanism. In the wider hexameric pore, the water diffuses much more slowly than in bulk but is still mobile. This, along with the dimensions of the pore, supports the observation that this peptide is selective for monovalent cations. Reasonable agreement of predicted conductances with experimentally determined values lends support to the validity of the simulations.

Putting the parts together

The structures of several different domains and subunits of potassium channels have recently been solved. Reassembling these fragments into a working model of an intact voltage-gated channel will be a major challenge.

The mechanism of channel formation by alamethicin as viewed by molecular dynamics simulations

Alamethicin is a 20-residue channel-forming peptide that forms a stable amphipathic alpha-helix in membrane and membrane-mimetic environments. This helix contains a kink induced by a central Gly-X-X-Pro sequence motif. Alamethicin channels are activated by a cis positive transbilayer voltage. Channel activation is suggested to correspond to voltage-induced insertion of alamethicin helices in the bilayer. Alamethicin forms multi-conductance channels in lipid bilayers. These channels are formed by parallel bundles of transmembrane helices surrounding a central pore. A change in the number of helices per bundle switches the single channel conductance level. Molecular dynamics simulations of alamethicin in a number of different environments have been used to explore its channel-forming properties. These simulations include: (i) alamethicin in solution in water and in methanol; (ii) a single alamethicin helix at the surface of a phosphatidylcholine bilayer; (iii) single alamethicin helices spanning a phosphatidylcholine bilayer; and (iv) channels formed by bundles of 5, 6, 7 or 8 alamethicin helices spanning a phosphatidylcholine bilayer. The total simulation time is c. 30 ns. Thus, these simulations provide a set of dynamic snapshots of a possible mechanism of channel formation by this peptide.

Engineering charge selectivity in alamethicin channels

The peptide alamethicin provides a system for engineering ion channel charge selectivity. To define alamethicin charge selectivity experimentally, we measured single-channel current-voltage relationships in KCl gradients using covalently linked peptide dimers. Two factors were found to contribute to the charge selectivity of these channels: (i) the ionic strength of the surrounding solutions; and (ii) the distribution of fixed charge on the peptide. Native alamethicin channels exhibited either cation selectivity or anion selectivity depending on which end of the channel was at the low salt side of the membrane. When the glutamine residue at position 18 in the sequence was replaced with a lysine residue, an anion-selective channel was obtained regardless of which end of the channel was at the low salt side of the membrane.

Simulation of the packing of idealized transmembrane alpha-helix bundles

The aim of this study is to investigate if the packing motifs of native transmembrane helices can be produced by simulations with simple potentials and to develop a method for the rapid generation of initial candidate models for integral membrane proteins composed of bundles of transmembrane helices. Constituent residues are mapped along the helix axis in order to maintain the amino acid sequence-dependent properties of the helix. Helix packing is optimized according to a semi-empirical potential mainly composed of four components: a bilayer potential, a crossing angle potential, a helix dipole potential and a helix-helix distance potential. A Monte Carlo simulated annealing protocol is employed to optimize the helix bundle system. Necessary parameters are derived from theoretical studies and statistical analysis of experimentally determined protein structures. Preliminary testing of the method has been conducted with idealized seven Ala(20) helix bundles. The structures generated show a high degree of compactness. It was observed that both bacteriorhodopsin-like and delta-endotoxin-like structures are generated in seven-helix bundle simulations, within which the composition varies dependent upon the cooling rate. The simulation method has also been employed to explore the packing of N = 4 and N = 12 transmembrane helix bundles. The results suggest that seven and 12 transmembrane helix bundles resembling those observed experimentally (e.g., bacteriorhodopsin, rhodopsin and cytochrome c oxidase subunit I) may be generated by simulations using simple potentials.

Effective diffusion coefficients of K+ and Cl- ions in ion channel models

We have used molecular dynamics simulations, corresponding to a total simulation time of 11 ns, to investigate the effective short-time local diffusion coefficient of potassium and chloride ions in a series of model ion channels. These models, which include channels formed by the fungal peptide alamethicin, by a synthetic leucine-serine peptide, and by the pore-lining M2 helix bundle of the nicotinic acetylcholine receptor, have a range of different secondary structures, diameters and hydrophobicities. We find that the diffusion coefficients of both ions are appreciably reduced in the narrower channels, the extent of the reduction being similar for both the anionic and cationic species. This suggests that a difference in mobility cannot be the source of the ion selectivity exhibited by some of the channels (for example, the leucine-serine peptide). We find no evidence for a reduction in mobility of either ion in the nAChR model. These results are broadly in line with a previous similar study of Na+ ions, and may be useful in Poisson-Nernst-Planck, Eyring rate theory or Brownian dynamics calculations of channel conductance.

Surface binding of alamethicin stabilizes its helical structure: molecular dynamics simulations

Alamethicin is an amphipathic alpha-helical peptide that forms ion channels. An early event in channel formation is believed to be the binding of alamethicin to the surface of a lipid bilayer. Molecular dynamics simulations are used to compare the structural and dynamic properties of alamethicin in water and alamethicin bound to the surface of a phosphatidylcholine bilayer. The bilayer surface simulation corresponded to a loosely bound alamethicin molecule that interacted with lipid headgroups but did not penetrate the hydrophobic core of the bilayer. Both simulations started with the peptide molecule in an alpha-helical conformation and lasted 2 ns. In water, the helix started to unfold after approximately 300 ps and by the end of the simulation only the N-terminal region of the peptide remained alpha-helical and the molecule had collapsed into a more compact form. At the surface of the bilayer, loss of helicity was restricted to the C-terminal third of the molecule and the rod-shaped structure of the peptide was retained. In the surface simulation about 10% of the peptide/water H-bonds were replaced by peptide/lipid H-bonds. These simulations suggest that some degree of stabilization of an amphipathic alpha-helix occurs at a bilayer surface even without interactions between hydrophobic side chains and the acyl chain core of the bilayer.

An anion-selective analogue of the channel-forming peptide alamethicin

The peptide alamethicin self-assembles to form helix bundle ion channels in membranes. Previous macroscopic measurements have shown that these channels are mildly cation-selective. Models indicate that a source of cation selectivity is a zone of partial negative charge toward the C-terminal end of the peptide. We synthesized an alamethicin derivative with a lysine in this zone (replacing the glutamine at position 18 in the sequence). Microscopic (single-channel) measurements demonstrate that dimeric alamethicin-lysine18 (alm-K18) forms mildly anion-selective channels under conditions where channels formed by the parent peptide are cation-selective. Long-range electrostatic interactions can explain the inversion of ion selectivity and the conductance properties of alamethicin channels.

Membrane proteins: A tale of barrels and corks

Crystal structures have been solved for two bacterial outer membrane proteins, FhuA and FepA, which mediate active transport of chelated iron. Analysis of ligand-induced changes in the structure of FhuA has provided our first structural insights into an active transport mechanism for a complex solute.

An extensive and diverse gene family of nicotinic acetylcholine receptor alpha subunits in Caenorhabditis elegans

Using reverse transcription-polymerase chain reactions the transcription of eight novel candidate nicotinic acetylcholine receptor (nAChR) alpha subunit genes has been demonstrated in the nematode Caenorhabditis elegans. Together with five other alpha subunit genes described elsewhere by ourselves (unc-38) and other workers (deg-3, acr-4, Ce21 and acr-6), this is now the largest known family of nAChR alpha subunit genes in a single species. By homology we have identified four groups of alpha subunits: DEG-3-like; ACR-16[Ce21]-like; UNC-38-like and ACR-8-like. Five C. elegans nAChR alpha subunits contain a modification in loop C of the ACh binding site in which the normally conserved Tyr-x-Cys-Cys, is replaced by a distinct motif (Tyr-x-x-Cys-Cys). Variation is also found in the channel lining M2 regions, including the replacement in four subunits of the highly conserved leucine at the 9' position by valine and most notably, the replacement in all ACR-8-like subunits of the highly conserved glutamic acid at the -1' position by histidine. Restrained molecular dynamics simulations have been used to generate homo-pentameric M2 helix bundle models for alpha subunits and possible functional implications examined. The calculated electrostatic potential energy profile for the M2 region of ACR-8 differs strikingly from that of ACR-16[Ce21] largely due to the presence of histidine at the -1' position, suggesting a possible perturbation of nAChR channel action permeability in the presence of this subunit type.

Defining the transmembrane helix of M2 protein from influenza A by molecular dynamics simulations in a lipid bilayer

Integral membrane proteins containing at least one transmembrane (TM) alpha-helix are believed to account for between 20% and 30% of most genomes. There are several algorithms that accurately predict the number and position of TM helices within a membrane protein sequence. However, these methods tend to disagree over the beginning and end residues of TM helices, posing problems for subsequent modeling and simulation studies. Molecular dynamics (MD) simulations in an explicit lipid and water environment are used to help define the TM helix of the M2 protein from influenza A virus. Based on a comparison of the results of five different secondary structure prediction algorithms, three different helix lengths (an 18mer, a 26mer, and a 34mer) were simulated. Each simulation system contained 127 POPC molecules plus approximately 3500-4700 waters, giving a total of approximately 18,000-21,000 atoms. Two simulations, each of 2 ns duration, were run for the 18mer and 26mer, and five separate simulations were run for the 34mer, using different starting models generated by restrained in vacuo MD simulations. The total simulation time amounted to 11 ns. Analysis of the time-dependent secondary structure of the TM segments was used to define the regions that adopted a stable alpha-helical conformation throughout the simulation. This analysis indicates a core TM region of approximately 20 residues (from residue 22 to residue 43) that remained in an alpha-helical conformation. Analysis of atomic density profiles suggested that the 18mer helix revealed a local perturbation of the lipid bilayer. Polar side chains on either side of this region form relatively long-lived H-bonds to lipid headgroups and water molecules.

An alamethicin channel in a lipid bilayer: molecular dynamics simulations

We present the results of 2-ns molecular dynamics (MD) simulations of a hexameric bundle of Alm helices in a 1-palmitoyl-2-oleoylphosphatidylcholine bilayer. These simulations explore the dynamic properties of a model of a helix bundle channel in a complete phospholipid bilayer in an aqueous environment. We explore the stability and conformational dynamics of the bundle in a phospholipid bilayer. We also investigate the effect on bundle stability of the ionization state of the ring of Glu18 side chains. If all of the Glu18 side chains are ionised, the bundle is unstable; if none of the Glu18 side chains are ionized, the bundle is stable. pKA calculations suggest that either zero or one ionized Glu18 is present at neutral pH, correlating with the stable form of the helix bundle. The structural and dynamic properties of water in this model channel were examined. As in earlier in vacuo simulations (Breed et al., 1996 .Biophys. J. 70:1643-1661), the dipole moments of water molecules within the pore were aligned antiparallel to the helix dipoles. This contributes to the stability of the helix bundle.

Ion channels: structure of a molecular brake

A combination of crystallographic and mutagenesis studies on the HERG K+ channel, a key determinant of cardiac excitability, has suggested how the protein's extramembraneous amino-terminal domain might act as a 'molecular brake' that slows down channel deactivation.

Interactions of alpha-helices with lipid bilayers: a review of simulation studies

Membrane proteins, of which the majority seem to contain one or more alpha-helix, constitute approx. 30% of most genomes. A complete understanding of the nature of helix/bilayer interactions is necessary for an understanding of the structural principles underlying membrane proteins. This review describes computer simulation studies of helix/bilayer interactions. Key experimental studies of the interactions of alpha-helices and lipid bilayers are briefly reviewed. Surface associated helices are found in some membrane-bound enzymes (e.g. prostaglandin synthase), and as stages in the mechanisms of antimicrobial peptides and of pore-forming bacterial toxins. Transmembrane alpha-helices are found in most integral membrane proteins, and also in channels formed by amphipathic peptides or by bacterial toxins. Mean field simulations, in which the lipid bilayer is approximated as a hydrophobic continuum, have been used in studies of membrane-active peptides (e.g. alamethicin, melittin, magainin and dermaseptin) and of simple membrane proteins (e.g. phage Pf1 coat protein). All atom molecular dynamics simulations of fully solvated bilayers with transmembrane helices have been applied to: the constituent helices of bacteriorhodopsin; peptide-16 (a simple model TM helix); and a number of pore-lining helices from ion channels. Surface associated helices (e.g. melittin and dermaseptin) have been simulated, as have alpha-helical bundles such as bacteriorhodopsin and alamethicin. From comparison of the results from the two classes of simulation, it emerges that a major theoretical challenge is to exploit the results of all atom simulations in order to improve the mean field approach.

Peptide-bilayer interactions: simulations of dermaseptin B, an antimicrobial peptide

Dermaseptins, a family of antimicrobial peptides, are believed to act by forming amphipathic alpha-helices which associate with the cell membrane, leading to its permeabilisation and disruption. A simple mean field method is described for simulation of the interactions of peptides with lipid bilayers which includes an approximate representation of the electrostatic effects of the head-group region of the bilayer. Starting from an atomistic model of a PC phospholipid bilayer we calculate an average electrostatic potential along the bilayer normal. By combining the interaction of the peptide with this electrostatic potential and with the hydrophobic core of the membrane we arrive at a more complete description of peptide-bilayer energetics than would be obtained using sidechain hydrophobicities alone. Using this interaction potential in MD simulations of the frog skin peptide dermaseptin B reveals that the lipid bilayer stabilises the alpha-helical conformation of the peptide. This is in agreement with FTIR data. A surface associated orientation thus appears to be the most stable arrangement of the peptide, at least at zero ionic strength and without taking account of possible peptide-peptide interactions.