Improved Side Chain Dynamics in MARTINI Simulations of Protein–Lipid Interfaces

GōMartini 3: From large conformational changes in proteins to environmental bias corrections

Coarse-grained modeling has become an important tool to supplement experimental measurements, allowing access to spatio-temporal scales beyond all-atom based approaches. The GōMartini model combines structure- and physics-based coarse-grained approaches, balancing computational efficiency and accurate representation of protein dynamics with the capabilities of studying proteins in different biological environments. This paper introduces an enhanced GōMartini model, which combines a virtual-site implementation of Gō models with Martini 3. The implementation has been extensively tested by the community since the release of the reparametrized version of Martini. This work demonstrates the capabilities of the model in diverse case studies, ranging from protein-membrane binding to protein-ligand interactions and AFM force profile calculations. The model is also versatile, as it can address recent inaccuracies reported in the Martini protein model. Lastly, the paper discusses the advantages, limitations, and future perspectives of the Martini 3 protein model and its combination with Gō models.

Optimized Protein-Excipient Interactions in the Martini 3 Force Field

The high doses of drugs required for biotherapeutics, such as monoclonal antibodies (mAbs), and the small volumes that can be administered to patients by subcutaneous injections pose challenges due to high-concentration formulations. The addition of excipients, such as arginine and glutamate, to high-concentration protein formulations can increase solubility and reduce the tendency of protein particle formation. Molecular dynamics (MD) simulations can provide microscopic insights into the mode of action of excipients in mAb formulations but require large system sizes and long time scales that are currently beyond reach at the fully atomistic level. Computationally efficient coarse-grained models such as the Martini 3 force field can tackle this challenge but require careful parametrization, testing, and validation. This study extends the popular Martini 3 force field toward realistic protein-excipient interactions of arginine and glutamate excipients, using the Fab domains of the therapeutic mAbs trastuzumab and omalizumab as model systems. A novel all-atom to coarse-grained mapping of the amino acid excipients is introduced, which explicitly captures the zwitterionic character of the backbone. The Fab-excipient interactions of arginine and glutamate are characterized concerning molecular contacts with the Fabs at the single-residue level. The Martini 3 simulations are compared with results from all-atom simulations as a reference. Our findings reveal an overestimation of Fab-excipient contacts with the default interaction parameters of Martini 3, suggesting a too strong attraction between protein residues and excipients. Therefore, we reparametrized the protein-excipient interaction parameters in Martini 3 against all-atom simulations. The excipient interactions obtained with the new Martini 3 mapping and Lennard-Jones (LJ) interaction parameters, coined Martini 3-exc, agree closely with the all-atom reference data. This work presents an improved parameter set for mAb-arginine and mAb-glutamate interactions in the Martini 3 coarse-grained force field, a key step toward large-scale coarse-grained MD simulations of high-concentration mAb formulations and the stabilizing effects of excipients.

Martini3-IDP: improved Martini 3 force field for disordered proteins

Coarse-grained (CG) molecular dynamics (MD) is widely used for the efficient simulation of intrinsically disordered proteins (IDPs). The Martini model, one of the most popular CG force fields in biomolecular simulation, was reported to yield too compact IDP conformations, limiting its applications. Addressing this, we optimized the bonded parameters based on fitting to reference simulations of a diverse set of IDPs at atomistic resolution, resulting in a Martini3-based disordered protein model coined Martini3-IDP. This model leads to expanded IDP conformations, greatly improving the reproduction of the experimentally measured radii of gyration. Moreover, contrary to ad-hoc fixes based on scaling of protein-protein or protein-water interactions, Martini3-IDP keeps the overall interaction balance underlying Martini 3. To validate that, we perform a comprehensive testing including full-length multidomain proteins, IDP-lipid membrane binding and IDP-small molecule binding, confirming its ability to successfully capture the complex interplay between disordered proteins and diverse biomolecular components. Finally, the recently emerging concept of biomolecular condensate, through liquid-liquid phase separation, was also reproduced by Martini3-IDP for a number of both homotypic and heterotypic systems. With the improved Martini3-IDP model, we expand the ability to simulate processes involving IDPs in complex environments, at spatio-temporal scales inaccessible with all-atom models.

Dynamics and activation of membrane-bound B cell receptor assembly

B-cell receptor (BCR) complexes are expressed on the surface of a B-cell and are critical in antigen recognition and modulating the adaptive immune response. Even though the relevance of antibodies has been known for almost a hundred years, the antigen-dependent activation mechanism of B-cells has remained elusive. Several models have been proposed for BCR activation, including cross-linking, conformation-induced oligomerization, and dissociation activation models. Recently, the first cryo-EM structures of the human B-cell antigen receptor of the IgM and IgG isotypes have been published that validates the asymmetric organization of the BCR complex. Here, we carry out extensive molecular dynamics simulations to probe the conformational changes upon antigen binding and the influence of the membrane lipids. We identify two critical dynamical events that could be associated with antigen-dependent activation of BCR. First, antigen binding causes increased flexibility in regions distal to the antigen binding site. Second, antigen binding alters the rearrangement of IgM transmembrane helices, including the relative interaction of Igα/Igβ that mediates intracellular signaling. Furthermore, these transmembrane rearrangements lead to changes in localized lipid composition. Our work indirectly supports the conformational-change induced models of BCR activation and contributes to the understanding of the antigen-dependent activation mechanism of BCRs.

Dynamics and Activation of Membrane-Bound B Cell Receptor Assembly

B-cell receptor complexes (BCR) are expressed on the surface of a B-cell and are critical in antigen recognition and modulating the adaptive immune response. Even though the relevance of antibodies has been known for almost a hundred years, the antigen-dependent activation mechanism of B-cells has remained elusive. Several models have been proposed for BCR activation, including cross-linking, conformation-induced oligomerization, and dissociation activation models. Recently, the first cryo-EM structures of the human B-cell antigen receptor of the IgM and IgG isotypes were published that validated the asymmetric organization of the BCR complex. Here, we have carried out extensive molecular dynamics simulations to probe the conformational changes upon antigen binding and the influence of the membrane lipids. We identified two critical dynamical events that could be associated with antigen-dependent activation of BCR. First, antigen binding caused increased flexibility in regions distal to the antigen binding site. Second, antigen binding altered the rearrangement of IgM transmembrane helices, including the relative interaction of Igα/Igβ that mediates intracellular signaling. Furthermore, these transmembrane rearrangements led to changes in localized lipid composition.

Ebola Virus Matrix Protein VP40 Single Mutations G198R and G201R Significantly Enhance Plasma Membrane Localization

Viral proteins frequently undergo single or multiple amino acid mutations during replication, which can significantly alter their functionality. The Ebola virus matrix protein VP40 is multifunctional but primarily responsible for creating the viral envelope by binding to the inner leaflet of the host cell plasma membrane (PM). Changes to the VP40 surface cationic charge via mutations can influence PM interactions, resulting in altered viral assembly and budding. A recent mutagenesis study evaluated the effects of several mutations and found that mutations G198R and G201R enhanced VP40 assembly at the PM and virus-like particle budding. These two mutations lie in the loop region of the C-terminal domain (CTD), which directly interacts with the PM. To understand the role of these mutations in PM localization at the molecular level, we performed both all-atom and coarse-grained molecular dynamics simulations using a dimer-dimer configuration of VP40, which contains the CTD-CTD interface. Our studies indicate that the location of mutations on the outer surface of the CTD regions can lead to changes in membrane binding orientation and degree of membrane penetration. Direct PI(4,5)P2 interactions with the mutated residues seem to further stabilize and pull VP40 into the PM, thereby enhancing interactions with numerous amino acids that were otherwise infrequently or completely inaccessible. These multiscale computational studies provide new insights at the atomic and molecular level as to how VP40-PM interactions are altered through single amino acid mutations. Given the high case fatality rates associated with Ebola virus disease in humans, it is essential to explore the mechanisms of viral assembly in the presence of mutations to mitigate the severity of the disease and understand the potential of future outbreaks.

Disordered regions in the IRE1α ER lumenal domain mediate its stress-induced clustering

Conserved signaling cascades monitor protein-folding homeostasis to ensure proper cellular function. One of the evolutionary conserved key players is IRE1, which maintains endoplasmic reticulum (ER) homeostasis through the unfolded protein response (UPR). Upon accumulation of misfolded proteins in the ER, IRE1 forms clusters on the ER membrane to initiate UPR signaling. What regulates IRE1 cluster formation is not fully understood. Here, we show that the ER lumenal domain (LD) of human IRE1α forms biomolecular condensates in vitro. IRE1α LD condensates were stabilized both by binding to unfolded polypeptides as well as by tethering to model membranes, suggesting their role in assembling IRE1α into signaling-competent stable clusters. Molecular dynamics simulations indicated that weak multivalent interactions drive IRE1α LD clustering. Mutagenesis experiments identified disordered regions in IRE1α LD to control its clustering in vitro and in cells. Importantly, dysregulated clustering of IRE1α mutants led to defects in IRE1α signaling. Our results revealed that disordered regions in IRE1α LD control its clustering and suggest their role as a common strategy in regulating protein assembly on membranes.

OLIVES: A Go̅-like Model for Stabilizing Protein Structure via Hydrogen Bonding Native Contacts in the Martini 3 Coarse-Grained Force Field

Coarse-grained molecular dynamics simulations enable the modeling of increasingly complex systems at millisecond timescales. The transferable coarse-grained force field Martini 3 has shown great promise in modeling a wide range of biochemical processes, yet folded proteins in Martini 3 are not stable without the application of external bias potentials, such as elastic networks or Go̅-like models. We herein develop an algorithm, called OLIVES, which identifies native contacts with hydrogen bond capabilities in coarse-grained proteins and use it to implement a novel Go̅-like model for Martini 3. We show that the protein structure instability originates in part from the lack of hydrogen bond energy in the coarse-grained force field representation. By using realistic hydrogen bond energies obtained from literature ab initio calculations, it is demonstrated that protein stability can be recovered by the reintroduction of a coarse-grained hydrogen bond network and that OLIVES removes the need for secondary structure restraints. OLIVES is validated against known protein complexes and at the same time addresses the open question of whether there is a need for protein quaternary structure bias in Martini 3 simulations. It is shown that OLIVES can reduce the number of bias terms, hereby speeding up Martini 3 simulations of proteins by up to ≈30% on a GPU architecture compared to the established Go̅MARTINI Go̅-like model.

Rescaling protein-protein interactions improves Martini 3 for flexible proteins in solution

Multidomain proteins with flexible linkers and disordered regions play important roles in many cellular processes, but characterizing their conformational ensembles is difficult. We have previously shown that the coarse-grained model, Martini 3, produces too compact ensembles in solution, that may in part be remedied by strengthening protein-water interactions. Here, we show that decreasing the strength of protein-protein interactions leads to improved agreement with experimental data on a wide set of systems. We show that the 'symmetry' between rescaling protein-water and protein-protein interactions breaks down when studying interactions with or within membranes; rescaling protein-protein interactions better preserves the binding specificity of proteins with lipid membranes, whereas rescaling protein-water interactions preserves oligomerization of transmembrane helices. We conclude that decreasing the strength of protein-protein interactions improves the accuracy of Martini 3 for IDPs and multidomain proteins, both in solution and in the presence of a lipid membrane.

Assessing the Martini 3 protein model: A review of its path and potential

Coarse-grained (CG) protein models have become indispensable tools for studying many biological protein details, from conformational dynamics to the organization of protein macro-complexes, and even the interaction of proteins with other molecules. The Martini force field is one of the most widely used CG models for bio-molecular simulations, partly because of the enormous success of its protein model. With the recent release of a new and improved version of the Martini force field - Martini 3 - a new iteration of its protein model was also made available. The Martini 3 protein force field is an evolution of its Martini 2 counterpart, aimed at improving many of the shortcomings that had been previously identified. In this mini-review, we first provide a general overview of the model and then focus on the successful advances made in the short time since its release, many of which would not have been possible before. Furthermore, we discuss reported limitations, potential directions for model improvement and comment on what the likely future development and application avenues are.

A Practical Guide to All-Atom and Coarse-Grained Molecular Dynamics Simulations Using Amber and Gromacs: A Case Study of Disulfide-Bond Impact on the Intrinsically Disordered Amyloid Beta

Intrinsically disordered proteins (IDPs) pose challenges to conventional experimental techniques due to their large-scale conformational fluctuations and transient structural elements. This work presents computational methods for studying IDPs at various resolutions using the Amber and Gromacs packages with both all-atom (Amber ff19SB with the OPC water model) and coarse-grained (Martini 3 and SIRAH) approaches. The effectiveness of these methodologies is demonstrated by examining the monomeric form of amyloid-β (Aβ42), an IDP, with and without disulfide bonds at different resolutions. Our results clearly show that the addition of a disulfide bond decreases the β-content of Aβ42; however, it increases the tendency of the monomeric Aβ42 to form fibril-like conformations, explaining the various aggregation rates observed in experiments. Moreover, analysis of the monomeric Aβ42 compactness, secondary structure content, and comparison between calculated and experimental chemical shifts demonstrates that all three methods provide a reasonable choice to study IDPs; however, coarse-grained approaches may lack some atomistic details, such as secondary structure recognition, due to the simplifications used. In general, this study not only explains the role of disulfide bonds in Aβ42 but also provides a step-by-step protocol for setting up, conducting, and analyzing molecular dynamics (MD) simulations, which is adaptable for studying other biomacromolecules, including folded and disordered proteins and peptides.

Scrutinising the Conformational Ensemble of the Intrinsically Mixed-Folded Protein Galectin-3

Galectin-3 is a protein involved in many intra- and extra-cellular processes. It has been identified as a diagnostic or prognostic biomarker for certain types of heart disease, kidney disease and cancer. Galectin-3 comprises a carbohydrate recognition domain (CRD) and an N-terminal domain (NTD), which is unstructured and contains eight collagen-like Pro-Gly-rich tandem repeats. While the structure of the CRD has been solved using protein crystallography, current knowledge about conformations of full-length galectin-3 is limited. To fill in this knowledge gap, we performed molecular dynamics (MD) simulations of full-length galectin-3. We systematically re-scaled the solute-solvent interactions in the Martini 3 force field to obtain the best possible agreement between available data from SAXS experiments and the ensemble of conformations generated in the MD simulations. The simulation conformations were found to be very diverse, as reflected, e.g., by (i) large fluctuations in the radius of gyration, ranging from about 2 to 5 nm, and (ii) multiple transient contacts made by amino acid residues in the NTD. Consistent with evidence from NMR experiments, contacts between the CRD and NTD were observed to not involve the carbohydrate-binding site on the CRD surface. Contacts within the NTD were found to be made most frequently by aromatic residues. Formation of fuzzy complexes with unspecific stoichiometry was observed to be mediated mostly by the NTD. Taken together, we offer a detailed picture of the conformational ensemble of full-length galectin-3, which will be important for explaining the biological functions of this protein at the molecular level.

Building complex membranes with Martini 3

The Martini model is a popular force field for coarse-grained simulations. Membranes have always been at the center of its development, with the latest version, Martini 3, showing great promise in capturing more and more realistic behavior. In this chapter we provide a step-by-step tutorial on how to construct starting configurations, run initial simulations and perform dedicated analysis for membrane-based systems of increasing complexity, including leaflet asymmetry, curvature gradients and embedding of membrane proteins.

Multiscale simulations reveal TDP-43 molecular-level interactions driving condensation

The RNA-binding protein TDP-43 is associated with mRNA processing and transport from the nucleus to the cytoplasm. TDP-43 localizes in the nucleus as well as accumulating in cytoplasmic condensates such as stress granules. Aggregation and formation of amyloid-like fibrils of cytoplasmic TDP-43 are hallmarks of numerous neurodegenerative diseases, most strikingly present in >90% of amyotrophic lateral sclerosis (ALS) patients. If excessive accumulation of cytoplasmic TDP-43 causes, or is caused by, neurodegeneration is presently not known. In this work, we use molecular dynamics simulations at multiple resolutions to explore TDP-43 self- and cross-interaction dynamics. A full-length molecular model of TDP-43, all 414 amino acids, was constructed from select structures of the protein functional domains (N-terminal domain, and two RNA recognition motifs, RRM1 and RRM2) and modeling of disordered connecting loops and the low complexity glycine-rich C-terminus domain. All-atom CHARMM36m simulations of single TDP-43 proteins served as guides to construct a coarse-grained Martini 3 model of TDP-43. The Martini model and a coarser implicit solvent C⍺ model, optimized for disordered proteins, were subsequently used to probe TDP-43 interactions; self-interactions from single-chain full-length TDP-43 simulations, cross-interactions from simulations with two proteins and simulations with assemblies of dozens to hundreds of proteins. Our findings illustrate the utility of different modeling scales for accessing TDP-43 molecular-level interactions and suggest that TDP-43 has numerous interaction preferences or patterns, exhibiting an overall strong, but dynamic, association and driving the formation of biomolecular condensates.

Martini 3 Coarse-Grained Force Field for Cholesterol

Cholesterol plays a crucial role in biomembranes by regulating various properties, such as fluidity, rigidity, permeability, and organization of lipid bilayers. The latest version of the Martini model, Martini 3, offers significant improvements in interaction balance, molecular packing, and inclusion of new bead types and sizes. However, the release of the new model resulted in the need to reparameterize many core molecules, including cholesterol. Here, we describe the development and validation of a Martini 3 cholesterol model, addressing issues related to its bonded setup, shape, volume, and hydrophobicity. The proposed model mitigates some limitations of its Martini 2 predecessor while maintaining or improving the overall behavior.

Expulsion mechanism of the substrate-translocating subunit in ECF transporters

Energy-coupling factor (ECF)-type transporters mediate the uptake of micronutrients in many bacteria. They consist of a substrate-translocating subunit (S-component) and an ATP-hydrolysing motor (ECF module) Previous data indicate that the S-component topples within the membrane to alternately expose the binding site to either side of the membrane. In many ECF transporters, the substrate-free S-component can be expelled from the ECF module. Here we study this enigmatic expulsion step by cryogenic electron microscopy and reveal that ATP induces a concave-to-convex shape change of two long helices in the motor, thereby destroying the S-component's docking site and allowing for its dissociation. We show that adaptation of the membrane morphology to the conformational state of the motor may favour expulsion of the substrate-free S-component when ATP is bound and docking of the substrate-loaded S-component after hydrolysis. Our work provides a picture of bilayer-assisted chemo-mechanical coupling in the transport cycle of ECF transporters.

Self-Association of ACE-2 with Different RBD Amounts: A Dynamic Simulation Perspective on SARS-CoV-2 Infection

Transmissibility of SARS-CoV-2 initially relies on its trimeric Spike-RBDs to tether the ACE-2 on host cells, and enhanced self-association of ACE-2 engaged with Spike facilitates the viral infection. Two primary packing modes of Spike-ACE2 heteroproteins exist potentially due to discrepant amounts of RBDs loading on ACE-2, but the resultant self-association difference is inherently unclear. We used extensive coarse-grained dynamic simulations to characterize the self-association efficiency, the conformation relevance, and the molecular mechanism of ACE-2 with different RBD amounts. It was revealed that the ACE-2 hanging two/full RBDs (Mode-A) rapidly dimerized into the heteroprotein complex in a compact "linear" conformation, while the bare ACE-2 showed weakened self-association and a protein complex. The RBD-tethered ectodomains of ACE-2 presented a more upright conformation relative to the membrane, and the intermolecular ectodomains were predominantly packed by the neck domains, which was obligated to the rapid protein self-association in a compact pattern. Noted is the fact that the ACE-2 tethered by a single RBD (Mode-B) retained considerable self-association efficiency and clustering capability, which unravels the interrelation of ACE-2 colocalization and protein cross-linkage. The molecular perspectives in this study expound the self-association potency of ACE-2 with different RBD amounts and the viral activity implications, which can greatly enhance our comprehension of SARS-CoV-2 infection details.

NanoModeler CG: A Tool for Modeling and Engineering Functional Nanoparticles at a Coarse-Grained Resolution

Functionalized metal nanoparticles (NPs) are macromolecular assemblies with a tunable physicochemical profile that makes them interesting for biotechnology, materials science, and energy conversion. In this regard, molecular simulations offer a way to scrutinize the structural and dynamical features of monolayer-protected NPs and their interactions with relevant matrices. Previously, we developed NanoModeler, a webserver that automates the preparation of functionalized gold NPs for atomistic molecular dynamics (MD) simulations. Here, we present NanoModeler CG (www.nanomodeler.it), a new release of NanoModeler that now also allows the building and parametrizing of monolayer-protected metal NPs at a coarse-grained (CG) resolution. This new version extends our original methodology to NPs of eight different core shapes, conformed by up to 800,000 beads and coated by eight different monolayer morphologies. The resulting topologies are compatible with the Martini force field but are easily extendable to any other set of parameters parsed by the user. Finally, we demonstrate NanoModeler CG's capabilities by reproducing experimental structural features of alkylthiolated NPs and rationalizing the brush-to-mushroom phase transition of PEGylated anionic NPs. By automating the construction and parametrization of functionalized NPs, the NanoModeler series offers a standardized way to computationally model monolayer-protected nanosized systems.

Facilitating CG Simulations with MAD: The MArtini Database Server

The MArtini Database (MAD - https://mad.ibcp.fr) is a web server designed for the sharing of structures and topologies of molecules parametrized with the Martini coarse-grained (CG) force field. MAD can also convert atomistic structures into CG structures and prepare complex systems (including proteins, lipids, etc.) for molecular dynamics (MD) simulations at the CG level. It is dedicated to the generation of input files for Martini 3, the most recent version of this popular CG force field. Specifically, the MAD server currently includes tools to submit or retrieve CG models of a wide range of molecules (lipids, carbohydrates, nanoparticles, etc.), transform atomistic protein structures into CG structures and topologies, with fine control on the process and assemble biomolecules into large systems, and deliver all files necessary to start simulations in the GROMACS MD engine.

A scoring function for the prediction of protein complex interfaces based on the neighborhood preferences of amino acids

Proteins often assemble into functional complexes, the structures of which are more difficult to obtain than those of the individual protein molecules. Given the structures of the subunits, it is possible to predict plausible complex models via computational methods such as molecular docking. Assessing the quality of the predicted models is crucial to obtain correct complex structures. Here, an energy-scoring function was developed based on the interfacial residues of structures in the Protein Data Bank. The statistically derived energy function (Nepre) imitates the neighborhood preferences of amino acids, including the types and relative positions of neighboring residues. Based on the preference statistics, a program iNepre was implemented and its performance was evaluated with several benchmarking decoy data sets. The results show that iNepre scores are powerful in model ranking to select the best protein complex structures.

Exploring CRD mobility during RAS/RAF engagement at the membrane

During the activation of mitogen-activated protein kinase (MAPK) signaling, the RAS-binding domain (RBD) and cysteine-rich domain (CRD) of RAF bind to active RAS at the plasma membrane. The orientation of RAS at the membrane may be critical for formation of the RAS-RBDCRD complex and subsequent signaling. To explore how RAS membrane orientation relates to the protein dynamics within the RAS-RBDCRD complex, we perform multiscale coarse-grained and all-atom molecular dynamics (MD) simulations of KRAS4b bound to the RBD and CRD domains of RAF-1, both in solution and anchored to a model plasma membrane. Solution MD simulations describe dynamic KRAS4b-CRD conformations, suggesting that the CRD has sufficient flexibility in this environment to substantially change its binding interface with KRAS4b. In contrast, when the ternary complex is anchored to the membrane, the mobility of the CRD relative to KRAS4b is restricted, resulting in fewer distinct KRAS4b-CRD conformations. These simulations implicate membrane orientations of the ternary complex that are consistent with NMR measurements. While a crystal structure-like conformation is observed in both solution and membrane simulations, a particular intermolecular rearrangement of the ternary complex is observed only when it is anchored to the membrane. This configuration emerges when the CRD hydrophobic loops are inserted into the membrane and helices α3-5 of KRAS4b are solvent exposed. This membrane-specific configuration is stabilized by KRAS4b-CRD contacts that are not observed in the crystal structure. These results suggest modulatory interplay between the CRD and plasma membrane that correlate with RAS/RAF complex structure and dynamics, and potentially influence subsequent steps in the activation of MAPK signaling.

Two cooperative binding sites sensitize PI(4,5)P2 recognition by the tubby domain

Phosphoinositides (PIs) are lipid signaling molecules that operate by recruiting proteins to cellular membranes via PI recognition domains. The dominant PI of the plasma membrane is phosphatidylinositol 4,5-bisphosphate [PI(4,5)P₂]. One of only two PI(4,5)P₂ recognition domains characterized in detail is the tubby domain. It is essential for targeting proteins into cilia involving reversible membrane association. However, the PI(4,5)P₂ binding properties of tubby domains have remained enigmatic. Here, we used coarse-grained molecular dynamics simulations to explore PI(4,5)P₂ binding by the prototypic tubby domain. The comparatively low PI(4,5)P₂ affinity of the previously described canonical binding site is underpinned in a cooperative manner by a previously unknown, adjacent second binding site. Mutations in the previously unknown site impaired PI(4,5)P₂-dependent plasma membrane localization in living cells and PI(4,5)P₂ interaction in silico, emphasizing its importance for PI(4,5)P₂ affinity. The two-ligand binding mode may serve to sharpen the membrane association-dissociation cycle of tubby-like proteins that underlies delivery of ciliary cargo.

Asynchronous Reciprocal Coupling of Martini 2.2 Coarse-Grained and CHARMM36 All-Atom Simulations in an Automated Multiscale Framework

The appeal of multiscale modeling approaches is predicated on the promise of combinatorial synergy. However, this promise can only be realized when distinct scales are combined with reciprocal consistency. Here, we consider multiscale molecular dynamics (MD) simulations that combine the accuracy and macromolecular flexibility accessible to fixed-charge all-atom (AA) representations with the sampling speed accessible to reductive, coarse-grained (CG) representations. AA-to-CG conversions are relatively straightforward because deterministic routines with unique outcomes are achievable. Conversely, CG-to-AA conversions have many solutions due to a surge in the number of degrees of freedom. While automated tools for biomolecular CG-to-AA transformation exist, we find that one popular option, called Backward, is prone to stochastic failure and the AA models that it does generate frequently have compromised protein structure and incorrect stereochemistry. Although these shortcomings can likely be circumvented by human intervention in isolated instances, automated multiscale coupling requires reliable and robust scale conversion. Here, we detail an extension to Multiscale Machine-learned Modeling Infrastructure (MuMMI), including an improved CG-to-AA conversion tool called sinceCG. This tool is reliable (∼98% weakly correlated repeat success rate), automatable (no unrecoverable hangs), and yields AA models that generally preserve protein secondary structure and maintain correct stereochemistry. We describe how the MuMMI framework identifies CG system configurations of interest, converts them to AA representations, and simulates them at the AA scale while on-the-fly analyses provide feedback to update CG parameters. Application to systems containing the peripheral membrane protein RAS and proximal components of RAF kinase on complex eight-component lipid bilayers with ∼1.5 million atoms is discussed in the context of MuMMI.

Key aspects of the past 30 years of protein design

Proteins are the workhorse of life. They are the building infrastructure of living systems; they are the most efficient molecular machines known, and their enzymatic activity is still unmatched in versatility by any artificial system. Perhaps proteins' most remarkable feature is their modularity. The large amount of information required to specify each protein's function is analogically encoded with an alphabet of just ∼20 letters. The protein folding problem is how to encode all such information in a sequence of 20 letters. In this review, we go through the last 30 years of research to summarize the state of the art and highlight some applications related to fundamental problems of protein evolution.

Improved Parameterization of Phosphatidylinositide Lipid Headgroups for the Martini 3 Coarse-Grain Force Field

Phosphoinositides are a family of membrane phospholipids that play crucial roles in membrane regulatory events. As such, these lipids are often a key part of molecular dynamics simulation studies of biological membranes, in particular of those employing coarse-grain models because of the potential long times and sizes of the involved membrane processes. Version 3 of the widely used Martini coarse-grain force field has been recently published, greatly refining many aspects of biomolecular interactions. In order to properly use it for lipid membrane simulations with phosphoinositides, we put forth the Martini 3-specific parameterization of inositol, phosphatidylinositol, and seven physiologically relevant phosphorylated derivatives of phosphatidylinositol. Compared to parameterizations for earlier Martini versions, focus was put on a more accurate reproduction of the behavior seen in both atomistic simulations and experimental studies, including the signaling-relevant phosphoinositide interaction with divalent cations. The models that we develop improve upon the conformational dynamics of phosphoinositides in the Martini force field and provide stable topologies at typical Martini time steps. They are able to reproduce experimentally known protein-binding poses as well as phosphoinositide aggregation tendencies. The latter was tested both in the presence and absence of calcium and included correct behavior of PI(4,5)P₂ calcium-induced clusters, which can be of relevance for regulation.

PIP2-induced membrane binding of the vinculin tail competes with its other binding partners

Vinculin plays a key role during the first phase of focal adhesion formation and interacts with the plasma membrane through specific binding of its tail domain to the lipid phosphatidylinositol 4,5-bisphosphate (PIP₂). Our understanding of the PIP₂-vinculin interaction has been hampered by contradictory biochemical and structural data. Here, we used a multiscale molecular dynamics simulation approach, in which unbiased coarse-grained molecular dynamics were used to generate starting structures for subsequent microsecond-long all-atom simulations. This allowed us to map the interaction of the vinculin tail with PIP₂-enriched membranes in atomistic detail. In agreement with experimental data, we have shown that membrane binding is sterically incompatible with the intramolecular interaction between vinculin's head and tail domain. Our simulations further confirmed biochemical and structural results, which identified two positively charged surfaces, the basic collar and the basic ladder, as the main PIP₂ interaction sites. By introducing a valency-disaggregated binding network analysis, we were able to map the protein-lipid interactions in unprecedented detail. In contrast to the basic collar, in which PIP₂ is specifically recognized by an up to hexavalent binding pocket, the basic ladder forms a series of low-valency binding sites. Importantly, many of these PIP₂ binding residues are also involved in maintaining vinculin in a closed, autoinhibited conformation. These findings led us to propose a molecular mechanism for the coupling between vinculin activation and membrane binding. Finally, our refined binding site suggests an allosteric relationship between PIP₂ and F-actin binding that disfavors simultaneous interaction with both ligands, despite nonoverlapping binding sites.

Cholesterol plays a decisive role in tetraspanin assemblies during bilayer deformations

The tetraspanin family plays key roles in many physiological processes, such as, tumour invasion, cell motility, virus infection, cell attachment and entry. Tetraspanins function as molecular scaffolds organized in microdomains with interesting downstream cellular consequences. However, despite their relevance in human physiology, the precise mechanisms of their various functions remain elusive. In particular, the full-length CD81 tetraspanin has interesting cholesterol-related properties that modulate its activity in cells. In this work, we study the opening transition of CD81 under different conditions. We propose that such conformational change is a collaborative process enhanced by simultaneous interactions between multiple identical CD81 tetraspanins. With molecular dynamics simulations we describe the crucial role of a ternary lipid bilayer with cholesterol in CD81 conformational dynamics, observing two emergent properties: first, clusters of CD81 collectively segregate one tetraspanin while favouring one opening transition, second, cumulative cholesterol sequestering by CD81 tetraspanins inhibits large membrane deformations due to local density variations.

Structure of the human signal peptidase complex reveals the determinants for signal peptide cleavage

The signal peptidase complex (SPC) is an essential membrane complex in the endoplasmic reticulum (ER), where it removes signal peptides (SPs) from a large variety of secretory pre-proteins with exquisite specificity. Although the determinants of this process have been established empirically, the molecular details of SP recognition and removal remain elusive. Here, we show that the human SPC exists in two functional paralogs with distinct proteolytic subunits. We determined the atomic structures of both paralogs using electron cryo-microscopy and structural proteomics. The active site is formed by a catalytic triad and abuts the ER membrane, where a transmembrane window collectively formed by all subunits locally thins the bilayer. Molecular dynamics simulations indicate that this unique architecture generates specificity for SPs based on the length of their hydrophobic segments.

Binding of Ca2+-independent C2 domains to lipid membranes: A multi-scale molecular dynamics study

C2 domains facilitate protein interactions with lipid bilayers in either a Ca²⁺-dependent or -independent manner. We used molecular dynamics (MD) simulations to explore six Ca²⁺-independent C2 domains, from KIBRA, PI3KC2α, RIM2, PTEN, SHIP2, and Smurf2. In coarse-grained MD simulations these C2 domains formed transient interactions with zwitterionic bilayers, compared with longer-lived interactions with anionic bilayers containing phosphatidylinositol bisphosphate (PIP₂). Type I C2 domains bound non-canonically via the front, back, or side of the β sandwich, whereas type II C2 domains bound canonically, via the top loops. C2 domains interacted strongly with membranes containing PIP₂, causing bound anionic lipids to cluster around the protein. Binding modes were refined via atomistic simulations. For PTEN and SHIP2, CG simulations of their phosphatase plus C2 domains with PIP₂-containing bilayers were also performed, and the roles of the two domains in membrane localization compared. These studies establish a simulation protocol for membrane-recognition proteins.

•

Binding of Ca²⁺-independent C2 domains to membranes was explored by MD simulation

•

C2 domains from KIBRA, PI3KC2α, RIM2, PTEN, SHIP2, and Smurf2 were compared

•

C2 domains formed longer-lived interactions with lipid bilayers containing PIP₂

•

For PTEN and SHIP2, simulations of their phosphatase plus C2 domains were performed

Martini 3: a general purpose force field for coarse-grained molecular dynamics

The coarse-grained Martini force field is widely used in biomolecular simulations. Here we present the refined model, Martini 3 ( http://cgmartini.nl ), with an improved interaction balance, new bead types and expanded ability to include specific interactions representing, for example, hydrogen bonding and electronic polarizability. The updated model allows more accurate predictions of molecular packing and interactions in general, which is exemplified with a vast and diverse set of applications, ranging from oil/water partitioning and miscibility data to complex molecular systems, involving protein-protein and protein-lipid interactions and material science applications as ionic liquids and aedamers.

Protocol for Simulations of PEGylated Proteins with Martini 3

Enhancement of proteins by PEGylation is an active area of research. However, the interactions between polymer and protein are far from fully understood. To gain a better insight into these interactions or even make predictions, molecular dynamics (MD) simulations can be applied to study specific protein-polymer systems at molecular level detail. Here we present instructions on how to simulate PEGylated proteins using the latest iteration of the Martini coarse-grained (CG) force-field. CG MD simulations offer near atomistic information and at the same time allow to study complex biological systems over longer time and length scales than fully atomistic-level simulations.

Protein-ligand binding with the coarse-grained Martini model

The detailed understanding of the binding of small molecules to proteins is the key for the development of novel drugs or to increase the acceptance of substrates by enzymes. Nowadays, computer-aided design of protein–ligand binding is an important tool to accomplish this task. Current approaches typically rely on high-throughput docking essays or computationally expensive atomistic molecular dynamics simulations. Here, we present an approach to use the recently re-parametrized coarse-grained Martini model to perform unbiased millisecond sampling of protein–ligand interactions of small drug-like molecules. Remarkably, we achieve high accuracy without the need of any a priori knowledge of binding pockets or pathways. Our approach is applied to a range of systems from the well-characterized T4 lysozyme over members of the GPCR family and nuclear receptors to a variety of enzymes. The presented results open the way to high-throughput screening of ligand libraries or protein mutations using the coarse-grained Martini model.

Computer-aided design of protein-ligand binding is important for the development of novel drugs. Here authors present an approach to use the recently re-parametrized coarse-grained Martini model to perform unbiased millisecond sampling of protein-ligand binding interactions of small drug-like molecules.

Biophysical characterization of p53 core domain aggregates

Aggregation is the cause of numerous protein conformation diseases. A common facet of these maladies is the transition of a protein from its functional native state into higher order forms, such as oligomers and amyloid fibrils. p53 is an essential tumor suppressor that is prone to such conformational transitions, resulting in its compromised ability to avert cancer. This work explores the biophysical properties of early-, mid-, and late-stage p53 core domain (p53C) aggregates. Atomistic and coarse-grained molecular dynamics (MD) simulations suggest that early- and mid-stage p53C aggregates have a polymorphic topology of antiparallel and parallel β-sheets that localize to the core amyloidogenic sequence. Both topologies involve similar extents of interstrand mainchain hydrogen bonding, while sidechain interactions could play a role in regulating strand orientation. The free energy difference between the antiparallel and parallel states was within statistical uncertainty. Negative stain electron microscopy of mature fibrils shows a wide distribution of fiber widths, indicating that polymorphism may extend to the quaternary structure level. Circular dichroism of the fibrils was indicative of β-sheet rich structures in atypical conformations. The Raman spectrum of aggregated p53C was consistent with a mixture of arranged β-sheets and heterogeneous structural elements, which is compatible with the MD findings of an ordered β-sheet nucleus flanked by disordered structure. Structural polymorphism is a common property of amyloids; however, because certain polymorphs of the same protein can be more harmful than others, going forward it will be pertinent to establish correlations between p53C aggregate structure and pathology.

Molecular mechanism for bidirectional regulation of CD44 for lipid raft affiliation by palmitoylations and PIP2

The co-localization of Cluster-of-Differentiation-44 protein (CD44) and cytoplasmic adaptors in specific membrane environments is crucial for cell adhesion and migration. The process is controlled by two different pathways: On the one hand palmitoylation keeps CD44 in lipid raft domains and disables the linking to the cytoplasmic adaptor, whereas on the other hand, the presence of phosphatidylinositol-4,5-biphosphate (PIP2) lipids accelerates the formation of the CD44-adaptor complex. The molecular mechanism explaining how CD44 is migrating into and out of the lipid raft domains and its dependence on both palmitoylations and the presence of PIP2 remains, however, elusive. In this study, we performed extensive molecular dynamics simulations to study the raft affinity and translocation of CD44 in phase separated model membranes as well as more realistic plasma membrane environments. We observe a delicate balance between the influence of the palmitoylations and the presence of PIP2 lipids: whereas the palmitoylations of CD44 increases the affinity for raft domains, PIP2 lipids have the opposite effect. Additionally, we studied the association between CD44 and the membrane adaptor FERM in dependence of these factors. We find that the presence of PIP2 lipids allows CD44 and FERM to associate in an experimentally observed binding mode whereas the highly palmitoylated species shows no binding affinity. Together, our results shed light on the sophisticated mechanism on how membrane translocation and peripheral protein association can be controlled by both protein modifications and membrane composition.

Cytoskeleton protein complex involving with association of CD44 and ERMs is critical for cancer-related cellular adhesion and migration. The protein interactions are found to be modulated by chemical modification and membrane microenvironments, but the inherent mechanism is unclear. We obtained molecular dynamic details of CD44 localization switching between raft/non-raft subdomains regulated by palmitoylations and PIP2 molecules. Binding of PIP2 on the palmitoylated CD44 enables it to release from lipid raft, revealing an exceptional role of PIP2 in mediating protein translocation. PIP2 is beneficial for CD44 to associate with the active domain of ERM, in a nearly crystal structure mode. The molecular information will enhance our understanding for PIP2 regulation to protein translocation and membrane association.

Capturing Choline-Aromatics Cation-π Interactions in the MARTINI Force Field

Cation−π interactions play an important role in biomolecular recognition, including interactions between membrane phosphatidylcholine lipids and aromatic amino acids of peripheral proteins. While molecular mechanics coarse grain (CG) force fields are particularly well suited to simulate membrane proteins in general, they are not parameterized to explicitly reproduce cation−π interactions. We here propose a modification of the polarizable MARTINI coarse grain (CG) model enabling it to model membrane binding events of peripheral proteins whose aromatic amino acid interactions with choline headgroups are crucial for their membrane binding. For this purpose, we first collected and curated a dataset of eight peripheral proteins from different families. We find that the MARTINI CG model expectedly underestimates aromatics–choline interactions and is unable to reproduce membrane binding of the peripheral proteins in our dataset. Adjustments of the relevant interactions in the polarizable MARTINI force field yield significant improvements in the observed binding events. The orientation of each membrane-bound protein is comparable to reference data from all-atom simulations and experimental binding data. We also use negative controls to ensure that choline–aromatics interactions are not overestimated. We finally check that membrane properties, transmembrane proteins, and membrane translocation potential of mean force (PMF) of aromatic amino acid side-chain analogues are not affected by the new parameter set. This new version “MARTINI 2.3P” is a significant improvement over its predecessors and is suitable for modeling membrane proteins including peripheral membrane binding of peptides and proteins.

The energetics of protein-lipid interactions as viewed by molecular simulations

Membranes are formed from a bilayer containing diverse lipid species with which membrane proteins interact. Integral, membrane proteins are embedded in this bilayer, where they interact with lipids from their surroundings, whilst peripheral membrane proteins bind to lipids at the surface of membranes. Lipid interactions can influence the function of membrane proteins, either directly or allosterically. Both experimental (structural) and computational approaches can reveal lipid binding sites on membrane proteins. It is, therefore, important to understand the free energies of these interactions. This affords a more complete view of the engagement of a particular protein with the biological membrane surrounding it. Here, we describe many computational approaches currently in use for this purpose, including recent advances using both free energy and unbiased simulation methods. In particular, we focus on interactions of integral membrane proteins with cholesterol, and with anionic lipids such as phosphatidylinositol 4,5-bis-phosphate and cardiolipin. Peripheral membrane proteins are exemplified via interactions of PH domains with phosphoinositide-containing membranes. We summarise the current state of the field and provide an outlook on likely future directions of investigation.

Order–disorder skewness in alpha-synuclein: a key mechanism to recognize membrane curvature

Currently, membrane curvature is understood as an active mechanism to control cells spatial organization and activity.

MARTINI-based simulation method for step-growth polymerization and its analysis by size exclusion characterization: a case study of cross-linked polyurethane

Simulation studies of step-growth polymerization, e.g., polymerization of polyurethane systems, hold great promise due to having complete control over the reaction conditions and being able to perform an in-depth analysis of network structures. In this work, we developed a (completely automated) simulation method based on a coarse-grained (CG) methodology, i.e., the MARTINI model, to study the cross-linking reaction of a diol, a tri-isocyanate molecule and one-hydroxyl functional molecule to form a polyurethane network without and with dangling chains. This method is capable of simulating the cross-linking reactions not only up to very high conversions, but also under rather complicated reaction conditions, i.e., a non-stoichiometric ratio of the reactants, solvent evaporation and multi-step addition of the reactants. We introduced a novel network analysis, similar to size-exclusion chromatography based on graph theory, to study the growth of the network during the polymerization process. By combining the reaction simulations with these analysis methods, a set of correlations between the reaction conditions, reaction mechanisms and final network structure and properties is revealed. For instance, a two-step addition of materials for the reaction, i.e., first the dangling chain to the tri-isocyanate and then the diol, leads to the highest integrated network structure. We observed that different reaction conditions lead to different glass transition temperatures (T_g) of the network due to the distinct differences in the final network structures obtained. For example, by addition of dangling chains to the network, the T_g decreases as compared to the network without dangling chains, as also is commonly observed experimentally.

Characterization of Lipid-Protein Interactions and Lipid-Mediated Modulation of Membrane Protein Function through Molecular Simulation

The cellular membrane constitutes one of the most fundamental compartments of a living cell, where key processes such as selective transport of material and exchange of information between the cell and its environment are mediated by proteins that are closely associated with the membrane. The heterogeneity of lipid composition of biological membranes and the effect of lipid molecules on the structure, dynamics, and function of membrane proteins are now widely recognized. Characterization of these functionally important lipid-protein interactions with experimental techniques is however still prohibitively challenging. Molecular dynamics (MD) simulations offer a powerful complementary approach with sufficient temporal and spatial resolutions to gain atomic-level structural information and energetics on lipid-protein interactions. In this review, we aim to provide a broad survey of MD simulations focusing on exploring lipid-protein interactions and characterizing lipid-modulated protein structure and dynamics that have been successful in providing novel insight into the mechanism of membrane protein function.

Transmembrane domain dimerization induces cholesterol rafts in curved lipid bilayers

Are the dimerization of transmembrane (TM) domains and the reorganization of the lipid bilayer two independent events?

A Practical View of the Martini Force Field

Martini is a coarse-grained (CG) force field suitable for molecular dynamics (MD) simulations of (bio)molecular systems. It is based on mapping of two to four heavy atoms to one CG particle. The effective interactions between the CG particles are parametrized to reproduce partitioning free energies of small chemical compounds between polar and apolar phases. In this chapter, a summary of the key elements of this CG force field is presented, followed by an example of practical application: a lipoplex-membrane fusion experiment. Formulated as hands-on practice, this chapter contains guidelines to build CG models of important biological systems, such as asymmetric bilayers and double-stranded DNA. Finally, a series of notes containing useful information, limitations, and tips are described in the last section.

Interactions of the EphA2 Kinase Domain with PIPs in Membranes: Implications for Receptor Function

EphA2 is a member of the receptor tyrosine kinase family. Interactions of the cytoplasmic region of EphA2 with the cell membrane are functionally important and yet remain incompletely characterized. Molecular dynamics simulations combined with biochemical studies reveal the interactions of the transmembrane, juxtamembrane (JM), and kinase domains with the membrane. We describe how the kinase domain is oriented relative to the membrane and how the JM region can modulate this interaction. We highlight the role of phosphatidylinositol phosphates (PIPs) in mediating the interaction of the kinase domain with the membrane and, conversely, how positively charged patches at the kinase surface and in the JM region induce the formation of nanoclusters of PIP molecules in the membrane. Integration of these results with those from previous studies enable computational reconstitution of a near complete EphA2 receptor within a membrane, suggesting a role for receptor-lipid interactions in modulation of EphA2.

Bending Lipid Bilayers: A Closed-Form Collective Variable for Effective Free-Energy Landscapes in Quantitative Biology

Curvature-related processes are of major importance during protein-membrane interactions. The illusive simplicity of membrane reshaping masks a complex molecular process crucial for a wide range of biological functions like fusion, endo- and exocytosis, cell division, cytokinesis, and autophagy. To date, no functional expression of a reaction coordinate capable of biasing molecular dynamics simulations to produce membrane curvature has been reported. This represents a major drawback given that the adequate identification of proper collective variables to enhance sampling is fundamental for restrained dynamics techniques. In this work, we present a closed-form equation of a collective variable that induces bending in lipid bilayers in a controlled manner, allowing for straightforward calculation of free energy landscapes of important curvature-related events, using standard methods such as umbrella sampling and metadynamics. As a direct application of the collective variable, we calculate the bending free energies of a ternary lipid bilayer in the presence and the absence of a Bin/Amphiphysin/Rvs domain with an N-terminal amphipathic helix (N-BAR), a well-known peripheral membrane protein that induces curvature.