Multiscale Molecular Dynamics Simulations Research Articles

Molecular Simulations Reveal the Dynamics of the T-Cell Receptor in a T-Cell Model Membrane

The T-cell antigen receptor (TCR) initiates immune responses by recognising a wide variety of foreign peptides presented by Major Histocompatibility Complex (MHC) proteins. The TCR consists of non-covalently associated TCRαβ, CD3εγ, CD3εδ and ζζ dimers. The CD3 and ζζ subunits have long intracellular tails that regulate TCR signalling. Despite significant knowledge of the TCR structure and topology, lack of information on the arrangement of its cytoplasmic tails limits our understanding of the molecular mechanism of intracellular signal propagation in a T-cell. In this study, we used molecular modelling to model the entire TCR structure and performed multi-scale molecular dynamics simulations of the TCR embedded in a complex lipid bilayer which mimics the T-cell membrane. Our simulations revealed conformational changes within the transmembrane region of the resting TCR, and also showed that the intracellular signalling tails of CD3 and ζζ subunits fold into an intertwined conformation forming contacts with each other. Additionally, the tails strongly interacted with the inner leaflet of the membrane via cationic stretches, whilst intracellular tyrosine side-chains transiently penetrated into the hydrophobic region of the membrane. Phosphatidyl inositol phosphate (PIP) lipids and cholesterol formed preferential protein-lipid interactions creating a unique local lipid environment around the TCR. Our study provides insights into the TCR's complete structure and dynamics in a membrane in atomic resolution. Further, this potentially forms the basis to study the molecular mechanism of initiation of TCR signalling.

Kinase inhibitors allosterically disrupt a regulatory interaction to enhance PKCα membrane translocation

The eukaryotic kinase domain has multiple intrinsically disordered regions whose conformation dictates kinase activity. Small molecule kinase inhibitors (SMKIs) rely on disrupting the active conformations of these disordered regions to inactivate the kinase. While SMKIs are selected for their ability to cause this disruption, the allosteric effects of conformational changes in disordered regions is limited by a lack of dynamic information provided by traditional structural techniques. In this study, we integrated multiscale molecular dynamics simulations with FRET sensors to characterize a novel allosteric mechanism that is selectively triggered by SMKI binding to the protein kinase Cα domain. The indole maleimide inhibitors BimI and sotrastaurin were found to displace the Gly-rich loop (G-loop) that normally shields the ATP-binding site. Displacement of the G-loop interferes with a newly identified, structurally conserved binding pocket for the C1a domain on the N lobe of the kinase domain. This binding pocket, in conjunction with the N-terminal regulatory sequence, masks a diacylglycerol (DAG) binding site on the C1a domain. SMKI-mediated displacement of the G-loop released C1a and exposed the DAG binding site, enhancing protein kinase Cα translocation both to synthetic lipid bilayers and to live cell membranes in the presence of DAG. Inhibitor chemotype determined the extent of the observed allosteric effects on the kinase domain and correlated with the extent of membrane recruitment. Our findings demonstrate the allosteric effects of SMKIs beyond the confines of kinase catalytic conformation and provide an integrated computational-experimental paradigm to investigate parallel mechanisms in other kinases.

Using multiscale molecular dynamics simulations to obtain insights into pore forming toxin mechanisms.

Pore forming toxins (PFTs) are virulent proteins released by several species, including many strains of bacteria, to attack and kill host cells. In this article, we focus on the utility of molecular dynamics (MD) simulations and the molecular insights gleaned from these techniques on the pore forming pathways of PFTs. In addition to all-atom simulations which are widely used, coarse-grained MARTINI models and structure-based models have also been used to study PFTs. Here, the emphasis is on methods and techniques involved while setting up, monitoring, and evaluating properties from MD simulations of PFTs in a membrane environment. We draw from several case studies to illustrate how MD simulations have provided molecular insights into protein-protein and protein-lipid interactions, lipid dynamics, conformational transitions and structures of both the oligomeric intermediates and assembled pore structures.

Molecular Dynamics Simulation to Uncover the Mechanisms of Protein Instability During Freezing.

Freezing is a common process applied in the pharmaceutical industry to store and transport biotherapeutics. Herewith, multi-scale molecular dynamics simulations of Lactate dehydrogenase (LDH) protein in phosphate buffer with/without ice formation performed to uncover the still poorly understood mechanisms and molecular details of protein destabilization upon freezing. Both fast and slow ice growing conditions were simulated at 243K from one or two-side of the simulation box, respectively. The rate of ice formation at all-atom simulations was crucial to LDH stability, as faster freezing rates resulted in enhanced structural stability maintained by a higher number of intramolecular hydrogen bonds, less flexible protein's residues, lower solvent accessibility and greater structural compactness. Further, protein aggregation investigated by coarse-grained simulations was verified to be initiated by extended protein structures and retained by electrostatic interactions of the salt bridges between charged residues and hydrogen bonds between polar residues of the protein. Lastly, the study of free energy of dissociation through steered molecular dynamics simulation revealed LDH was destabilized by the solvation of the hydrophobic core and the loss of hydrophobic interactions. For the first time, experimentally validated molecular simulations revealed the detailed mechanisms of LDH destabilization upon ice formation and cryoconcentration of solutes.

Multiscale simulations of the hydration shells surrounding spherical Fe3O4 nanoparticles and effect on magnetic properties.

Iron oxide magnetic nanoparticles (NPs) are excellent systems in catalysis and in nanomedicine, where they are mostly immersed in aqueous media. Even though the NP solvation by water is expected to play an active role, the detailed structural insight at the nanostructure oxide/water interface is still missing. Here, based on our previous efforts to obtain accurate models of dehydrated Fe3O4 NPs and of their magnetic properties and through multiscale molecular dynamics simulations combining the density functional tight binding method and force field, we unravel the atomistic details of the short range (chemical) and long range (physical) interfacial effects when magnetite nanoparticles are immersed in water. The influence of the first hydration shell on the structural, electronic and magnetic properties of Fe3O4 NPs is revealed by high-level hybrid density functional calculations. Hydrated Fe3O4 NPs possess larger magnetic moment than dehydrated ones. This work bridges the large gap between experimental studies on solvated Fe3O4 NPs and theoretical investigations on flat Fe3O4 surfaces covered with water and paves the way for further study of Fe3O4 NPs in biological environments.

Computational studies of membrane pore formation induced by Pin2

Understanding, at the molecular level, the effect of AMPs on biological membranes is of crucial importance given the increasing number of multidrug-resistant bacteria. Being part of an ancient type of innate immunity system, AMPs have emerged as a potential solution for which bacteria have not developed resistance. Traditional antibiotics specifically act on biosynthetic pathways, while AMPs may directly destabilize the lipid membrane, but it is unclear how AMPs affect the membrane’s stability. We performed multiscale molecular dynamics simulations to investigate the structural features leading to membrane pores formation on zwitterionic and anionic membranes by the antimicrobial peptide (AMP) Pandinin 2 (Pin2). Some experimental reports propose that Pin2 could form barrel-stave pores, while others suggest that it could form toroidal pores. Since there is no conclusive evidence of which type of pore is formed by Pin2 on bilayers, performing molecular dynamics simulations on these systems could shed some light on whether or not or what type of pore Pin2 forms on model membranes. Our results are focused on a detailed description of the pore formation by Pin2 in POPC and POPE:POPG membranes., which strongly suggest that Pin2 forms a toroidal pore and not a barrel-shaped pore; this type of pore also affects the membrane properties. In the process, a phospholipid remodeling in the POPE:POPG membrane takes place. Moreover, the pores formed by Pin2 indicate that they are selective for the chlorine ion. There are no previous ion selectivity reports for other AMPs with similar physicochemical properties, such as melittin and magainin. Communicated by Ramaswamy H. Sarma

Multi-scale coarse-grained molecular dynamics simulation to investigate the thermo-mechanical behavior of shape-memory polyurethane copolymers

Shape-memory polyurethanes (SMPUs) are smart semi-crystalline polymers with heat-induced shape memory. The segmented SMPU copolymer simulated in this study consisted of an isocyanate (hard segment) that acts as a netpoint to stabilize the network, and a polyol (soft segment) that, owing to polymer crystallization, acts as a molecular switch. The ratio of hard to soft segments is a determining factor in the thermo-mechanical behavior and shape-memory performance of SMPU copolymers. However, the limitations of the scale of the conventional all-atom molecular dynamics simulation makes it difficult to observe mesoscale phenomena such as the phase-segregated morphology and polymer crystallization. To address this problem, we have developed a coarse-grained (CG) molecular dynamics (MD) model with fewer degrees of freedom by treating multiple atoms as a single bead. We derived the CG intra- and inter-bead potentials of the CG MD model such that the structural and thermodynamic properties would be identical to those of the all-atom reference model. Consequently, polymer crystallization was verified to occur at room temperature with the developed CG potentials. We subsequently investigated the effects of hard-segment content (HSC) on the thermal transition at SMPU melting temperature. We also investigated the influence of HSC on the thermo-elastic behavior and shape-memory properties of SMPU copolymers by simulating a 4-step thermo-mechanical cycle. The results revealed that the microstructure significantly affects macroscopic shape-memory behavior. We expect that this study will be used to improve the prediction and design of the thermo-mechanical behavior of segmented polyurethane or other semi-crystalline polymers.

Multiscale molecular dynamics simulation study of polyoxyethylated alcohols self-assembly in emulsion systems

Although nonionic surfactants have been widely employed as emulsifiers, their self-assembly behaviors in emulsion systems are not well understood. Herein, we report a multiscale molecular dynamics simulation study on the nonionic surfactant aggregations in alcohol ethoxylate (AEO)/water/octane/hexanol quaternary emulsion systems. By systematically examining the effects of chain length, degree of branching in surfactants and the addition of co-surfactants, we find that the stability of self-assembled emulsions is majorly affected by the hydrophilic chain length and chain distribution at the oil-water interfaces. The presence of branch chains in surfactants can promote the self-assembly. Moreover, the addition of co-surfactants benefits the formation of interface films and facilitates the formation of micro-emulsions. By unraveling the self-assembly mechanism of nonionic surfactants in emulsion systems from a multiscale perspective, this study provides fundamental insights towards the preparation and applications of nonionic-surfactant-based emulsions.

Activation Microswitches in Adenosine Receptor A2A Function as Rheostats in the Cell Membrane.

Although multiple components of the cell membrane modulate the stability and activation of G protein-coupled receptors (GPCRs), insights into the dynamics of GPCR structures come from biophysical studies conducted in detergents. This is because of the challenges of studying activation in a multicomponent lipid bilayer. To understand the role of cellular membrane lipids and cations in GPCR activation, we performed multiscale molecular dynamics simulations (56 μs) on three different conformational states of adenosine receptor A2AR, in both the cell membrane-like lipid bilayer and in detergent micelles. Molecular dynamics (MD) simulations show that the phosphatidylinositol bisphosphate (PIP2) interacts with the basic residues in the intracellular regions of A2AR, thereby reducing the flexibility of the receptor in the inactive state and limiting the transition to the active-intermediate state. In the G protein-coupled fully active state, PIP2 stabilizes the GPCR:G protein complex. Such stiffening effects are absent in non-ionic detergent micelles, and therefore, more transitions have been observed in detergents. The inter-residue distances that change significantly upon GPCR activation are known as activation microswitches. The activation microswitches show different levels of activation in the cell membrane, in the pure POPC bilayer, and in detergents. Thus, the temporal heat map of different activation microswitches calculated from the MD simulations suggests a rheostat model of GPCR activation microswitches rather than the binary switch model. These simulation results connect the chemistry of cell membrane lipids to receptor activity, which is useful for the design of detergents mimicking the cell membrane.

Photoinduced DNA Lesions in Dormant Bacteria: The Peculiar Route Leading to Spore Photoproducts Characterized by Multiscale Molecular Dynamics*.

Some bacterial species enter a dormant state in the form of spores to resist to unfavorable external conditions. Spores are resistant to a wide series of stress agents, including UV radiation, and can last for tens to hundreds of years. Due to the suspension of biological functions, such as DNA repair, they accumulate DNA damage upon exposure to UV radiation. Differently from active organisms, the most common DNA photoproducts in spores are not cyclobutane pyrimidine dimers, but rather the so-called spore photoproducts. This noncanonical photochemistry results from the dry state of DNA and its binding to small, acid-soluble proteins that drastically modify the structure and photoreactivity of the nucleic acid. Herein, multiscale molecular dynamics simulations, including extended classical molecular dynamics and quantum mechanics/molecular mechanics based dynamics, are used to elucidate the coupling of electronic and structural factors that lead to this photochemical outcome. In particular, the well-described impact of the peculiar DNA environment found in spores on the favored formation of the spore photoproduct, given the small free energy barrier found for this path, is rationalized. Meanwhile, the specific organization of spore DNA precludes the photochemical path that leads to cyclobutane pyrimidine dimer formation.

Investigation of the relationship between nanochannel width and mass transfer characteristics for dense methane nanofluidics

In this paper, we analytically investigate the influence of the nanochannel width on mass transfer characteristics through the rough nanochannel by using non–equilibrium multiscale molecular dynamics simulation. The mass transfer characteristics are analyzed by fitting the curve correlation between the diffusion coefficient and nanochannel width. The diffusion coefficient is calculated using the revised Green–Kubo relation in different nanochannels. The numerical results demonstrate that the relationship between diffusion coefficient and the inverse of nanochannel width is parabolic. Moreover, the diffusion coefficients show significant variations in the small rough nanochannel, and gradually approach the values of bulk system with increasing the width of nanochannel. In addition, to explore its mechanism, the local atom number density, velocity profile, velocity autocorrelation function, the escape rate of molecule, local diffusion coefficient and projection radial distribution function are given in different widths of nanochannel under given boundary condition. Simulation results show that the local structure and number density of fluid influence the transfer mass characteristics. These results of this work may be conducive to further understand the mass transfer behaviors of nanofluidics, and guide the optimal design of nano–devices.

Concurrent Multiscale Simulations of Rough Lubricated Contact of Aluminum Single Crystal

In this paper, a concurrent multiscale simulation strategy coupling atomistic and continuum models was proposed to investigate the three-dimensional contact responses of aluminum single crystal under both dry and lubricated conditions. The Hertz contact is performed by using both the multiscale and full molecular dynamics (MD) simulations for validation. From the contact area, kinetic energy and stress continuity aspects, the multiscale model shows good accuracy. It can also save at least five times the computational time compared with the full MD simulations for the same domain size. Furthermore, the results of lubricated contact show that the lubricant molecules could effectively cover the contact surfaces; thereby separating the aluminum surfaces and bearing the support loads. Moreover, the surface topography could be protected by the thin film formed by the lubricant molecules. It has been found that the contact area decreases obviously with increasing the magnitude of load under both dry and lubricated contacts. Besides, a decrease in contact area is also seen when the number of lubricant molecules increases. The present study has confirmed that the dimension of lubricated contacts could be greatly expanded during the simulation using the proposed multiscale method without sacrificing too much computational time and accuracy.

The Glycosphingolipid GM3 Modulates Conformational Dynamics of the Glucagon Receptor

The extracellular domain (ECD) of class B1 G-protein-coupled receptors (GPCRs) plays a central role in signal transduction and is uniquely positioned to sense both the extracellular and membrane environments. Although recent studies suggest a role for membrane lipids in the modulation of class A and class F GPCR signaling properties, little is known about the effect of lipids on class B1 receptors. In this study, we employed multiscale molecular dynamics simulations to access the dynamics of the glucagon receptor (GCGR) ECD in the presence of native-like membrane bilayers. Simulations showed that the ECD could move about a hinge region formed by residues Q122–E126 to adopt both closed and open conformations relative to the transmembrane domain. ECD movements were modulated by binding of the glycosphingolipid GM3. These large-scale fluctuations in ECD conformation may affect the ligand binding and receptor activation properties. We also identify a unique phosphatidylinositol (4,5)-bisphosphate (PIP2) interaction profile near intracellular loop (ICL) 2/TM3 at the G-protein-coupling interface, suggesting a mechanism of engaging G-proteins that may have a distinct dependence on PIP2 compared with class A GPCRs. Given the structural conservation of class B1 GPCRs, the modulatory effects of GM3 and PIP2 on GCGR may be conserved across these receptors, offering new insights into potential therapeutic targeting.

Protein Fibrils Formed by Rationally Designed α-Helical Peptides.

Fibrillar structures of proteins play essential roles in normal life events as well as diseases. It is of great importance to understand the principles by which proteins organize into fibrils. Here, we report a rationally designed α-helical peptide that can aggregate into fibrils. Mutation studies indicate that the helicity of the peptide is crucial for fibril formation. Multiscale molecular dynamics simulations demonstrated that the peptide may assemble in a quasiregular pattern, which is different from both the coiled coil and cross-α structures reported before. Our study provides a new helical peptide design that produces a fibrillar structure and contributes to the understanding of fibrillar structures formed by α-helices.

Membrane mediated toppling mechanism of the folate energy coupling factor transporter

Energy coupling factor (ECF) transporters are responsible for the uptake of micronutrients in bacteria and archaea. They consist of an integral membrane unit, the S-component, and a tripartite ECF module. It has been proposed that the S-component mediates the substrate transport by toppling over in the membrane when docking onto an ECF module. Here, we present multi-scale molecular dynamics simulations and in vitro experiments to study the molecular toppling mechanism of the S-component of a folate-specific ECF transporter. Simulations reveal a strong bending of the membrane around the ECF module that provides a driving force for toppling of the S-component. The stability of the toppled state depends on the presence of non-bilayer forming lipids, as confirmed by folate transport activity measurements. Together, our data provide evidence for a lipid-dependent toppling-based mechanism for the folate-specific ECF transporter, a mechanism that might apply to other ECF transporters.

Insight into the interface between Fe3O4 (001) surface and water overlayers through multiscale molecular dynamics simulations.

In this work, we investigate the Fe3O4 (001) surface/water interface by combining several theoretical approaches, ranging from a hybrid functional method (HSE06) to density-functional tight-binding (DFTB) to molecular mechanics (MM). First, we assess the accuracy of the DFTB method to correctly reproduce HSE06 results on structural details and energetics and available experimental data for adsorption of isolated water, dimers, and trimers up to a water monolayer. Second, we build two possible configurations of a second and a third overlayer and perform molecular dynamics simulations with DFTB, monitoring the water orientation, the H-bond network, and the ordered water structure formation. To make our models more realistic, we then build a 12 nm-thick water multilayer on top of the Fe3O4 (001) surface slab model, which we investigate through MM-molecular dynamics (MD). The water layer structuring, revealed by the analysis of the atomic positions from a long MM-MD run for this large MM model, extends up to about 6-7 Å and nicely compares with that observed for a water trilayer model. However, MM and DFTB MD simulations show some discrepancy due to the poor description of the Fe⋯OH2 distance in MM that calls for further work in the parameterization of the model.

Computational Investigation of Voltage-Gated Sodium Channel β3 Subunit Dynamics.

Voltage-gated sodium (Nav) channels form the basis for the initiation of the action potential in excitable cells by allowing sodium ions to pass through the cell membrane. The Nav channel α subunit is known to function both with and without associated β subunits. There is increasing evidence that these β subunits have multiple roles that include not only influencing the voltage-dependent gating but also the ability to alter the spatial distribution of the pore-forming α subunit. Recent structural data has shown possible ways in which β1 subunits may interact with the α subunit. However, the position of the β1 subunit would not be compatible with a previous trimer structure of the β3 subunit. Furthermore, little is currently known about the dynamic behavior of the β subunits both as individual monomers and as higher order oligomers. Here, we use multiscale molecular dynamics simulations to assess the dynamics of the β3, and the closely related, β1 subunit. These findings reveal the spatio-temporal dynamics of β subunits and should provide a useful framework for interpreting future low-resolution experiments such as atomic force microscopy.

Emergence of chromatin hierarchical loops from protein disorder and nucleosome asymmetry

Protein flexibility and disorder is emerging as a crucial modulator of chromatin structure. Histone tail disorder enables transient binding of different molecules to the nucleosomes, thereby promoting heterogeneous and dynamic internucleosome interactions and making possible recruitment of a wide-range of regulatory and remodeling proteins. On the basis of extensive multiscale modeling we reveal the importance of linker histone H1 protein disorder for chromatin hierarchical looping. Our multiscale approach bridges microsecond-long bias-exchange metadynamics molecular dynamics simulations of atomistic 211-bp nucleosomes with coarse-grained Monte Carlo simulations of 100-nucleosome systems. We show that the long C-terminal domain (CTD) of H1-a ubiquitous nucleosome-binding protein-remains disordered when bound to the nucleosome. Notably, such CTD disorder leads to an asymmetric and dynamical nucleosome conformation that promotes chromatin structural flexibility and establishes long-range hierarchical loops. Furthermore, the degree of condensation and flexibility of H1 can be fine-tuned, explaining chromosomal differences of interphase versus metaphase states that correspond to partial and hyperphosphorylated H1, respectively. This important role of H1 protein disorder in large-scale chromatin organization has a wide range of biological implications.

Flow-Induced Self-Assembly of Spider Silk from Multi-Scale Simulations

Dragline spider silk proteins or spidroins self-assemble into an outstandingly tough fiber. The combination of ductility and strength relies in its microscopic structure within the fiber: small and strong beta-sheet crystals formed by poly-alanine repeats embedded into a flexible amorphous matrix of glycine-rich repeats. Flow is a critical factor for the assembly of the highly disordered silk proteins into this elaborated structure within the fibril. However, the mechanism of flow-induced silk self-assembly remains elusive. We studied oligomer formation of tethered repetitive dragline peptides under uniform flow using non-equilibrium multi-scale molecular dynamics simulations at different flow rates. In atomistic Molecular Dynamics simulations, we found poly-alanine repeats to primarily drive the self-assembly, confirming crystal formation to promote fibrillation. We used this finding in more coarse-grained hydrodynamic simulations with aminoacid resolution, treating the silk proteins as semi-flexible block copolymers. We observed that medium to high flow velocities (>20 cm/s) increase alignment and crystallization of the silk peptides. High velocities (> 50 cm/s) result in extensions close to the contour length of the protein (>90%), thereby slowing down the assembly process as protein fluctuations are largely abolished. Our results yield a microscopic understanding of flow-induced silk assembly, which is likely relevant also for other flow-dependent proteins.

Cholesterol Controls Dynamics of the Metabotropic Glutamate Receptor 2 via an Ionic-Lock

The Metabotropic glutamate receptor (mGluR) 2 is a key G-protein coupled receptor for the function of the central nervous system. The lipid environment, in particular cholesterol, has been shown to strongly regulate mGluR2. Nevertheless, the molecular mechanism of such regulation, remained unknown. By using a combination of multi-scale molecular dynamics simulations and biochemical assays, we determined the cholesterol binding fingerprint around the transmembrane domain of mGluR2, composed of at least two, but potentially three more, specific binding sites.