Molecular dynamics simulations of proteins in lipid bilayers

With recent advances in X-ray crystallography of membrane proteins promising many new high-resolution structures, molecular dynamics simulations will become increasingly valuable for understanding membrane protein function, as they can reveal the dynamic behavior concealed in the static structures. Dramatic increases in computational power, in synergy with more efficient computational methodologies, now allow us to carry out molecular dynamics simulations of any structurally known membrane protein in its native environment, covering timescales of up to 0.1 micros. At the frontiers of membrane protein simulations are ion channels, aquaporins, passive and active transporters, and bioenergetic proteins.     

Molecular dynamics simulations of hydrophilic pores in lipid bilayers

Hydrophilic pores are formed in peptide free lipid bilayers under mechanical stress. It has been proposed that the transport of ionic species across such membranes is largely determined by the existence of such meta-stable hydrophilic pores. To study the properties of these structures and understand the mechanism by which pore expansion leads to membrane rupture, a series of molecular dynamics simulations of a dipalmitoylphosphatidylcholine (DPPC) bilayer have been conducted. The system was simulated in two different states; first, as a bilayer containing a meta-stable pore and second, as an equilibrated bilayer without a pore. Surface tension in both cases was applied to study the formation and stability of hydrophilic pores inside the bilayers. It is observed that below a critical threshold tension of approximately 38 mN/m the pores are stabilized. The minimum radius at which a pore can be stabilized is 0.7 nm. Based on the critical threshold tension the line tension of the bilayer was estimated to be approximately 3 x 10(-11) N, in good agreement with experimental measurements. The flux of water molecules through these stabilized pores was analyzed, and the structure and size of the pores characterized. When the lateral pressure exceeds the threshold tension, the pores become unstable and start to expand causing the rupture of the membrane. In the simulations the mechanical threshold tension necessary to cause rupture of the membrane on a nanosecond timescale is much higher in the case of the equilibrated bilayers, as compared with membranes containing preexisting pores.     

Molecular dynamics simulations of supported phospholipid/alkanethiol bilayers on a gold(111) surface.

Molecular dynamics simulations have been used to investigate the structure of hybrid bilayers (HB) formed by dipalmitoylphosphatidylcholine (DPPC) lipid monolayers adsorbed on a hydrophobic alkanethiol self-assembled monolayer (SAM). The HB system was studied at 20 degrees C and 60 degrees C, and the results were compared with recent neutron reflectivity measurements (Meuse, C. W., S. Krueger, C. F. Majkrzak, J. A. Dura, J. Fu, J. T. Connor, and A. L. Plant. 1998. Biophys. J. 74:1388) and previous simulations of hydrated multilamellar bilayers (MLB) of DPPC (Tu, K., D. J. Tobias, and M. L. Klein. 1995. Biophys. J. 69:2558; and 1996. 70:595). The overall structures of the HBs are in very good agreement with experiment. The structure of the SAM monolayer is hardly perturbed by the presence of the DPPC overlayer. The DPPC layer presents characteristics very similar to the MLB gel phase at low temperature and to the liquid crystal phase at high temperature. Subtle changes have been found for the lipid/water interface of the HBs compared to the MLBs. The average phosphatidylcholine headgroup orientation is less disordered, and this produces changes in the electric properties of the HB lipid/water interface. These changes are attributed to the fact that the aqueous environment of the lipids in these unilamellar films is different from that of MLB stacks. Finally, examination of the intramolecular and whole-molecule dynamics of the DPPC molecules in the fluid phase HB and MLB membranes revealed that the reorientations of the upper part of the acyl chains (near the acyl ester linkage) are slower, the single molecule protrusions are slightly damped, and the lateral rattling motions are significantly reduced in the HB compared with the MLB.     

Molecular dynamics simulations of the interaction of phospholipid bilayers with polycaprolactone

The molecular interaction between common polymer chains and the cell membrane is unknown. Molecular dynamics simulations offer an emerging tool to characterise the nature of the interaction between common degradable polymer chains used in biomedical applications, such as polycaprolactone, and model cell membranes. Herein we characterise with all-atomistic and coarse-grained molecular dynamics simulations the interaction between single polycaprolactone chains of varying chain lengths with a phospholipid membrane. We find that the length of the polymer chain greatly affects the nature of interaction with the membrane, as well as the membrane properties. Furthermore, we next utilise advanced sampling techniques in molecular dynamics to characterise the two-dimensional free energy surface for the interaction of varying polymer chain lengths (short, intermediate, and long) with model cell membranes. We find that the free energy minimum shifts from the membrane-water interface to the hydrophobic core of the phospholipid membrane as a function of chain length. Finally, we perform coarse-grained molecular dynamics simulations of slightly larger membranes with polymers of the same length and characterise the results as compared with all-atomistic molecular dynamics simulations. These results can be used to design polymer chain lengths and chemistries to optimise their interaction with cell membranes at the molecular level.     

Nociceptin and dynorphin A(1-17) produce antianalgesia through independent systems in mice

The administration of dynorphin A(1-17), Dyn, intrathecally (i.t.) or of nociceptin, intracerebroventricularly (i.c.v.) produces antianalgesic actions against i.t. morphine in the tail flick test in mice. The antianalgesic action of nociceptin is mediated by spinal PGE2 and attenuated by i.t. PGD2 or indomethacin. The Dyn response is mediated by release of IL1beta in the spinal cord to activate an ascending pathway to the brain and in turn releases IL1beta in the brain which activates a descending pathway to the spinal cord. The present work investigated the possibility that the action of IL1beta in the Dyn system might release prostaglandins so that the Dyn and nociceptin antianalgesic systems would overlap at these points. The results indicated that in the Dyn system neither the IL1beta in the spinal cord or brain implicated prostaglandin release because i.t. and i.c.v. PGD2 and indomethacin did not affect Dyn-induced antianalgesia. In addition, nociceptin-induced antianalgesia did not involve components in the Dyn system. Thus, the Dyn and nociceptin antianalgesic systems did not overlap and each were independent systems.     

Molecular dynamics simulations of C-terminal decapeptide of gastrin-releasing peptide in DMPC bilayers: structure, stability and orientation of the peptide hormone within the bilayers

Gastrin-releasing peptide (GRP) is a member of bombesin-like peptides and bombesin and neuromedin B are other members of this family. They act on receptors that belong to the GPCR superfamily and exert important physiological functions upon binding to their receptors. The biologically active C-terminal decapeptide of GRP (GRP10) was studied in explicit DMPC bilayers using molecular dynamics simulations. In the initial conformation, the peptide was placed perpendicular to the membrane plane and the peptide-membrane complex with approximately 20,000 atoms was simulated for a period of 8 ns. After a 5 ns simulation, GRP10 adopted a tilted orientation and the tilt angle with respect to the bilayer normal was approximately 60 masculine. Analysis of the interactions of individual residues indicated the role of histidine residues in maintaining a tilted orientation.     

Molecular dynamics simulation of the structure of dimyristoylphosphatidylcholine bilayers with cholesterol, ergosterol, and lanosterol

Five molecular dynamics computer simulations were performed on different phospholipid:sterol membrane systems in order to study the influence of sterol structure on membrane properties. Three of these simulated bilayer systems were composed of a 1:8 sterol:phospholipid ratio, each of which employed one of the sterol molecules: cholesterol, ergosterol, and lanosterol. The two other simulations were of a bilayer with a 1:1 sterol:phospholipid ratio. These simulations employed cholesterol and lanosterol, respectively, as their sterol components. The observed differences in simulations with cholesterol and lanosterol may have their implication on the form of the phospholipid/sterol phase diagram.     

Hypothalamo-posterior pituitary system in Brattleboro rats: immunoreactive levels of leucine-enkephalin, dynorphin (1-17), dynorphin (1-8) and alpha-neo-endorphin

The levels of immunoreactive leucine-enkephalin, alpha-neo-endorphin, dynorphin (1-17) and dynorphin (1-8) have been determined in the hypothalamus and posterior pituitary from male and female Brattleboro rats homozygous (unable to produce vasopressin) and heterozygous (producing vasopressin) for diabetes insipidus, and from male and female Long Evans rats. In the hypothalamus we found no significant differences in the levels of these peptides while there were great differences in extracts from the posterior pituitary: female homozygous animals have greatly reduced levels in all four peptides compared to the heterozygous controls. In male homozygous animals the differences in the dynorphin (1-17) and leucine-enkephalin levels were small whereas the concentrations of alpha-neo-endorphin and dynorphin (1-8) showed a significant decrease compared to the male heterozygous controls. The results indicate a reduction in opioid peptides linked to the vasopressin deficiency in a partially sex dependent manner.     

Nitric oxide synthase inhibitors influence dynorphin A (1-17) immunoreactivity in the rat brain following hyperthermia

Summary.  The possibility that nitric oxide synthase (NOS) inhibitors influence dynorphin immunoreactivity following hyperthermia was examined in a rat model using a pharmacological approach. Previous reports from our laboratory show that hyperthermia induces an upregulation of NOS in several brain regions that seems to be instrumental in causing cell injury. Recent reports suggest that nitric oxide (NO) can influence dynorphin neurotransmission in the normal brain as well as in several pathological states. Since dynorphin is neurotoxic in different animal models of brain or spinal cord injury, it may be that the peptide will contribute to the cell injury in hyperthermia. The present investigation was carried out to determine whether hyperthermia can influence dynorphin immunoreactivity in the brain, and if so, whether inhibition of NOS will influence the peptide distribution in the brain following heat stress. Rats subjected to hyperthermia at 38°C for 4 h in a biological oxygen demand incubator (BOD) resulted in a marked upregulation of dynorphin immunoreactivity in several brain regions e.g., cerebral cortex, hippocampus, cerebellum and brain stem. Pretreatment of rats with two potent NOS inhibitors, L-NAME (30 mg/kg/day, i.p. for 7 days) or L-NMMA (35 mg/kg/day, i.p. for 7 days) significantly attenuated the dynorphin immunoreactivity in the brain. These drugs were also able to reduce hyperthermia induced blood-brain barrier (BBB) permeability, brain edema formation and cell injury. Taken together, our results suggest that (i) hyperthermia has the capacity to upregulate dynorphin immunoreactivity in the brain, (ii) inhibition of NOS considerably attenuates the dynorphin immunoreaction following heat stress and (iii) upregulation of dynorphin is somehow contributing to hyperthermia induced brain damage, not reported earlier. Received July 3, 2001 Accepted August 6, 2001 Published online July 31, 2002     

Molecular dynamics simulations of adrenocorticotropin (1-24) peptide in a solvated dodecylphosphocholine (DPC) micelle and in a dimyistoylphosphatidylcholine (DMPC) bilayer

The structure and interactions of the 1-24 fragment of the adrenocorticotropin hormone, ACTH (1-24), with membrane have been studied by molecular dynamics (MD) simulation in an NPT ensembles in two explicit membrane mimics, a dodecylphosphocholine (DPC) micelle and a dimyristoylphosphatidylcholine (DMPC) bilayer. The starting configuration of the peptide/lipid systems had the 1-10 segment of the peptide lying on the surface of the model membrane, the same as the equilibrated structure (by MD) of ACTH (1-10) in a DPC micelle. The simulations showed that the peptide adopts the surface-binding mode and essentially the same structure in both systems. Thus the results of this work lend support to the assumption that micelles are reasonable mimics for biological membranes for the study of peptide binding. The 1-10 segment is slightly tilted from the parallel orientation to the interface and interacts strongly with the membrane surface while the more polar 11-24 segment shows little tendency to interact with the membrane surface, preferring to reside primarily in the aqueous phase. Furthermore, the 1-10 segment of the peptide binds to the DPC micelle in essentially the same way as ACTH (1-10). Thus the MD results are in excellent agreement with the model of interaction of ACTH (1-24) with membrane derived from NMR experiments. The secondary structure and the hydration of the peptide and the interactions of specific residues with the lipid head groups have also been analyzed.     

Peptide models of dynorphin A(1-17) incorporating minimally homologous substitutes for the potential amphiphilic beta strand in residues 7-15

Two peptide models of dynorphin A(1-17) have been synthesized. These peptides incorporate a minimally homologous substitute sequence for residues 6-17, including alternating lysine and valine residues substituting for the potential amphiphilic beta-strand structure in positions 7-15. Model 1 retains Pro10 from the native sequence, but model 2 does not. Compression isotherms of peptide monolayers at the air-water interface and CD spectra of peptide films adsorbed from aqueous solution onto siliconized quartz slides were evaluated by comparison to those of idealized amphiphilic alpha-helical, beta-sheet, and disordered peptides. Dynorphin A(1-17) was mostly disordered, whereas beta-endorphin was alpha helical. Dynorphin model 1 had properties similar to those of dynorphin A(1-17) at these interfaces, but model 2 formed strongly amphiphilic beta sheets. In binding assays to mu-, delta-, and kappa-opioid receptors in guinea pig brain membranes, model 1 reproduced the high potency and selectivity of dynorphin A(1-17) for kappa receptors, and model 2 was only 3 times less potent and less selective for these receptors. Both peptide models retained the high, kappa-selective agonist activity of dynorphin A(1-17) in guinea pig ileum assays, and like dynorphin A(1-17), model 1 had little activity in the rat vas deferens assay. In view of the minimal homology of the modeled dynorphin structures, these studies support current models of membrane-catalyzed opioid ligand-receptor interactions and suggest a role for the amphiphilic alpha-helical and beta-strand structures in beta-endorphin and dynorphin A(1-17), respectively, in this process.     

MK-801降低鞘内注射强啡肽A(1-17)对福尔马林诱导的大鼠痛反应的增强效应

本实验用福尔马林试验在动物痛模型上观察了鞘内单纯注射生理盐水 (NS)、NMDA受体阻断剂MK 80 1、阿片受体阻断剂纳洛酮 (naloxone)、强啡肽A [DynA (1 17) ]以及先用MK 80 1或纳洛酮再注射DynA (1 17)对动物的行为痛反应的影响。大鼠后肢脚掌皮下注射福尔马林后出现的行为痛反应显示有 2个时相 ,即首先出现持续较短的第一时相和 3～ 6min后出现的持续较长的第二时相。实验结果显示 ,各组的第一时相无明显差异 ;而第二时相则有差异 :鞘内注射DynA (1 17)组第二时相痛反应持续时间 (489 5± 2 2 5s)明显较单纯鞘内注射NS组(3 44 7± 12 9s)、MK 80 1组 (3 3 1 4± 2 0 7s)和纳洛酮组 (3 5 2 5± 18 4s)长 (均为P <0 0 1) ;而先用NMDA受体阻断剂MK 80 1后再注射DynA (1 17) ,则第二时相行为痛反应的持续时间 (2 85 7± 19 4s)较单纯注射DynA (1 17)组明显缩短 (P <0 0 1) ,但与单纯鞘内注射MK 80 1组相比无明显差异 ;先用阿片受体阻断剂纳洛酮后再注射DynA (1 17) ,则动物的第二时相行为痛反应 (473 8± 17 8s)与单纯注射DynA (1 17)组相比无明显差异 ,而与单纯注射NS组或纳洛酮组相比则明显增强 (分别为P <0 0 1)。因此本实验结果提示 :(1)在脊髓水平的DynA(1 17)具有促进福尔马林所诱导的第二     

Molecular dynamics simulation of melittin in a dimyristoylphosphatidylcholine bilayer membrane.

Molecular dynamics trajectories of melittin in an explicit dimyristoyl phosphatidylcholine (DMPC) bilayer are generated to study the details of lipid-protein interactions at the microscopic level. Melittin, a small amphipathic peptide found in bee venom, is known to have a pronounced effect on the lysis of membranes. The peptide is initially set parallel to the membrane-solution interfacial region in an alpha-helical conformation with unprotonated N-terminus. Solid-state nuclear magnetic resonance (NMR) and polarized attenuated total internal reflectance Fourier transform infrared (PATIR-FTIR) properties of melittin are calculated from the trajectory to characterize the orientation of the peptide relative to the bilayer. The residue Lys7 located in the hydrophobic moiety of the helix and residues Lys23, Arg24, Gln25, and Gln26 at the C-terminus hydrophilic form hydrogen bonds with water molecules and with the ester carbonyl groups of the lipids, suggesting their important contribution to the stability of the helix in the bilayer. Lipid acyl chains are closely packed around melittin, contributing to the stable association with the membrane. Calculated density profiles and order parameters of the lipid acyl chains averaged over the molecular dynamics trajectory indicate that melittin has effects on both layers of the membrane. The presence of melittin in the upper layer causes a local thinning of the bilayer that favors the penetration of water through the lower layer. The energetic factors involved in the association of melittin at the membrane surface are characterized using an implicit mean-field model in which the membrane and the surrounding solvent are represented as structureless continuum dielectric material. The results obtained by solving the Poisson-Bolztmann equation numerically are in qualitative agreement with the detailed dynamics. The influence of the protonation state of the N-terminus of melittin is examined. After 600 ps, the N-terminus of melittin is protonated and the trajectory is continued for 400 ps, which leads to an important penetration of water molecules into the bilayer. These observations provide insights into how melittin interacts with membranes and the mechanism by which it enhances their lysis.     

Molecular dynamics simulations of individual alpha-helices of bacteriorhodopsin in dimyristoylphosphatidylcholine. II. Interaction energy analysis.

The concepts of hydrophobicity and hydrophobic moments have been applied in attempts to predict membrane protein secondary and tertiary structure. The current paper uses molecular dynamics computer calculations of individual bacteriorhodopsin helices in explicit dimyristoylphosphatidylcholine bilayers to examine the atomic basis of these approaches. The results suggest that the types of interactions between a particular amino acid and the surrounding bilayer depend on the position and type of the amino acid. In particular, aromatic residues are seen to interact favorably at the interface region. Analysis of the trajectories in terms of hydrophobic moments suggests the presence of a particular face that prefers lipid. The results of these simulations may be used to improve secondary structure prediction methods and to provide further insights into the two-stage model of protein folding.     

Molecular dynamics simulations of model trans-membrane peptides in lipid bilayers: a systematic investigation of hydrophobic mismatch

Hydrophobic mismatch, which is the difference between the hydrophobic length of trans-membrane segments of a protein and the hydrophobic width of the surrounding lipid bilayer, is known to play a role in membrane protein function. We have performed molecular dynamics simulations of trans-membrane KALP peptides (sequence: GKK(LA)nLKKA) in phospholipid bilayers to investigate hydrophobic mismatch alleviation mechanisms. By varying systematically the length of the peptide (KALP15, KALP19, KALP23, KALP27, and KALP31) and the lipid hydrophobic length (DLPC, DMPC, and DPPC), a wide range of mismatch conditions were studied. Simulations of durations of 50-200 ns show that under positive mismatch, the system alleviates the mismatch predominantly by tilting the peptide and to a smaller extent by increased lipid ordering in the immediate vicinity of the peptide. Under negative mismatch, alleviation takes place by a combination of local bilayer bending and the snorkeling of the lysine residues of the peptide. Simulations performed at a higher peptide/lipid molar ratio (1:25) reveal slower dynamics of both the peptide and lipid relative to those at a lower peptide/lipid ratio (1:128). The lysine residues have favorable interactions with specific oxygen atoms of the phospholipid headgroups, indicating the preferred localization of these residues at the lipid/water interface.     

Pharmacological characterization of beta-endorphin- and dynorphin A(1-17)-induced feeding using G-protein alpha-subunit antisense probes in rats

Antisense (AS) oligodeoxynucleotides targeting G-protein alpha-subunits distinguish feeding responses of morphine and its metabolite, as well as nocturnal and deprivation-induced feeding. The present study examined whether feeding elicited by beta-endorphin (betaEND) or dynorphin A(1-17) was altered by ventricularly-applied G(i)alpha(1), G(i)alpha(2), G(i)alpha(3), G(s)alpha, G(o)alpha, G(q)alpha or G(x/z)alpha AS probes, or a nonsense (NS) control. The betaEND-induced feeding was reduced by the G(i)alpha(1) and G(x/z)alpha AS probes, and increased by G(i)alpha(2) or G(i)alpha(3) AS treatment. Dynorphin-induced feeding was attenuated by G(i)alpha(1) and G(o)alpha AS treatment. Yet, G(s)alpha or G(q)alpha AS and NS treatments failed to alter opioid agonist-induced feeding. These data provide initial characterization of potential effector signaling pathways mediating betaEND and dynorphin-induced feeding.     

Proton NMR and CD solution conformation determination and opioid receptor binding studies of a dynorphin A(1-17) model peptide

The opioid peptide dynorphin A(1-17) contains a peptide segment in residues 7-15 with the potential to form an amphiphilic beta-strand. This amphiphilic structure may, like the amphiphilic alpha-helices found in many other peptide hormones, be an important determinant of its interactions with membranes and receptors. In order to investigate and characterize these interactions, we have synthesized a 17-residue dynorphin analogue (YGGFLKKVKPKVKVKSS) that incorporates a peptide model of this amphiphilic secondary structure with minimized homology (25%) relative to the native sequence. This peptide exhibits the full biological potency of dynorphin in assays of kappa-opioid receptor binding, and is more selective for this type of opioid receptor than the natural peptide. The conformation of the model peptide in aqueous solution has been investigated in detail by NMR spectroscopy. The values of the NH-CH alpha coupling constants together with rotating frame NOEs indicate the presence of an amphiphilic structure together with some beta-strand structure in residues 7-15, and demonstrate that a peptide model that stabilizes this structure in aqueous solution and enhances kappa-opioid receptor selectivity can be successfully designed using using alternating lysine and valine residues.     

Dynorphin(1-10)amide: a potent and selective analog of dynorphin(1-13)

Dynorphin(1-10)amide was more potent than Dynorphin(1-13) in inhibiting the twitch of the mouse vas deferens (IC50 of Dynorphin(1-10)amide = 0.3 nM and IC50 of Dynorphin (1-13) = 4.0 nM). Binding assays indicated that two opioid peptides had similar profiles in that they enhanced dihydromorphine (DHM) binding in picomolar concentrations but displaced DHM binding in nanomolar concentrations (IC50 for Dynorphin(1-10)amide = 5 nM). In the mouse tail-flick assay, however, Dynorphin(1-10)amide showed a more selective action on morphine-induced analgesia. Although Dynorphin(1-10)amide had no significant analgesic activity by itself, it differed from the (1-13) analog by neither potentiating nor antagonizing morphine in naive animals. In tolerant animals, on the other hand, 50 microgram of this analog administered icv shifted the ED50 of morphine from 43.0(33.0-55.9) to 17.0 (12.4-23.3). Thus, Dynorphin(1-10)amide appears to be a more potent and selective analog of Dynorphin(1-13).